WATER POLLUTION CONTROL RESEARCH SERIES • 14010 DAY OS/71
       Inorganic Sulfur Oxidation
        by Iron-Oxidizing  Bacteria

ENVIRONMENTAL PROTECTION AGENCY • WATER QUALITY OFFIC1

-------
        HATER POLLUTION CONTROL RESEARCH SERIFS
The yater Pollution Control Research Series describes
the results and prepress in the control and abatement
of pollution in our. Nation's, waters.  They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Hater Duality
Office, Environmental Protection Agency, through inhouse
research and grants and contracts with Federal^ State.
and local aoencies, research institutions, and industrial
oroanizations.

Inquiries pertaining to Hater Pollution Control Research
Reports should he directed to the Head, Project Reports
Systera, Office of Research and Development, V.'ater Quality
Office, Environtnental Protection Agency, Room 1108,
Washington,1 0. C.  20242.

-------
             Inorganic Sulfur Oxidation

             by Iron-Oxidizing Bacteria
                             by
                   Department of  Biology
                    Syracuse University
                    Syracuse, New York
                            for the

               ENVIRONMENTAL  PROTECTION AGENCY
                     Project  #14010 DAY

                         June,  1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, B.C. 20402 - Price $1.25
                         Stock Number 5501-0117

-------
               EPA Review Notice
This report has been reviewed by the Water
Quality Office, EPA, and approved for publication.
Approval does not signify that the contents
necessarily reflect the views and policies of
the Environmental Protection Agency, nor does
mention of trade names or commercial products
constitute endorsement or recommendation for
use.

-------
                          Abstract
The utilization of sulfur and reduced sulfur compounds  by the
iron-oxidizing chemolithotroph Thiobacillus ferrooxidans  was
studied at the biochemical level.  The identification,
characterization and partial purification of the rhodanese
and sulfite oxidase enzymes completed the scheme of sulfur
metabolism in J_. ferrooxidans which leads to energy generation.

The cell envelope lipopolysaccharide (IPS) purified from iron-
grown cells was studied in the electron microscope.  The
partial chemical composition of the LPS revealed unusually
high quantities of Fe^+.  A new colorimetric whole cell assay
to study iron-oxidation kinetics was developed which will be
of benefit to future studies at the molecular level.  The
inorganic pyrophosphatase enzyme, an essential enzyme in
maintaining the energy balance in the cell, was partially
purified and its properties studies.  This is the first
account of the presence of this enzyme in chemolithotrophic
microorganisms.

The effects of organic carbon and energy sources on chemo-
lithotrophic microorganisms were studied.  T_. ferrooxidans
can convert from chemolithotrophic to heterotrophic metabolism
after a long lag in the presence of the organic substrate,
and after some energy is stored from iron oxidation.  Growth
on glucose proceeds much like other heterotrophic gram
negative organisms.  The metabolism of glucose is via the
Entner-Doudoroff pathway.

A new species of Thiobacillus has been isolated from alkaline
mine drainage  (pH > 4.5) and its taxonomic status determined.
The organisms most interesting feature is its ability to
produce basic substances which can increase the pH to as
much as 8.3.

This report was submitted in fulfillment of grant number
14010 DAY under the sponsorship of the Federal Water Quality
Administration.
                             111

-------
                             CONTENTS


Section                                                          n
-                                                          Page

     1.  Conclusions ....................... 1

    11.  Recommendations ..................... 3

   111.  Introduction ...................... 5

    IV.  Studies concerning the metabolism of Thiobacillus
              f e rro o xi dans grown on elemental sulfur.

            Introduction  ....................

            Isolation and properties of the rhodanese enzyme. .   . ?

            Isolation and properties of the sulfite oxidase
                 enzyme  .................... 1 £
            Discussion ..................... 22

     V.  Studies concerning the isolation and identification
              of new Thiobacillus spp.
            Introduction
27
            Isolation of an alkaline producing Thiobacillus
               sp

            Isolation of an acid producing Thiobacillus
                                           -
                                                                   '
            Discussion ...................... 3'0

    VI.  Studies  concerning the metabolism of Thiobacillus
               ferrooxidans grown on iron.

            Introduction  .................... ^7

            The structure and chemical composition of the cell
                  envelope ..................... 
-------
                     CONTENTS CContinued)
                                                          Page
 VII.  Studies concerning the metabolism of Thio-
            bacillus  ferrooxidans  grown on hetero-
            trophic substrates	31

VIII.  Acknowledgments	131

  IX.  References	133

   X.  List of publications resulting from the grant.  ...  143

  XI.  Glossary	145

 XII.  Appendix	147

            List of persons trained under this grant.  ...  147

            Published abstracts of papers delivered
                 before professional meetings	140

            Title of papers delivered to local
                 professional meetings	14C
            Titles of papers delivered at  symposia
                 at professional meetings
                                 Vi
-------
                     USE OF FIGURES

                                                        Page
 1.   Electron micrograph of a  carbon -platinum shadowed
          cell of Thiobacillus strain J isolated
          from alkaline mine drainage .........   32

 2.   Electron micrograph of carbon-platinum shadowed
          cells  of  an acid producing Thiobacillus
          grown  heterotrophically ...........   36

 3.   Electron micrograph of negatively stained
          cells  of  an acid producing Thiob aci 1 lus
          grown  heterotrophically ...........   37

 4.   Electron micrograph of carbon-platinum
          shadowed  cells of an acid producing
          Thiobacillus grown autotrophically .....   33

 5.   Growth parameters of an acid producing
          Thiobacillus grown on thiosulfate ......   33

 6.   Oxygen uptake, measured manometrically, by the
          acid producing Thiobacillus oxidizing
          reduced sulfur compounds  ..........   41
 7.   Oxygen  uptake, measured manometrically by  the
          acid producing Thiobacillus  oxidizing
          sulfite  in  the presence  and  absence of
          organic  compounds   .............   42

 8.   Oxygen  uptake, measured manometrically, by the
          acid producing Thiobacillus  oxidizing
          0.5% yeast  extract  .............   43

 9.   Thin section  of  a T_.  f errooxi dans cell  grown
          autotrophically  on iron  showing a
          multilayer  cell  envelope ..........   ^2-
10.   Thin section of T_.  f errooxi dans  after chloroform-
          methanol extraction .............    53

11.   Thin section of T.  f errooxi dans  after chloroform-
                     _
          methanol  extraction
                           VI1
-------
                                                        Page

12.  Thin section of purified LPS from T_. ferrooxidans
          showing the membrane like tripartite
          structure ...................  5C

13.  Negatively stained purified LPS from T_. ferrooxi-
          dans showing the ribbons and sheets of LPS  .  .  57

14.  Sedimentation velocity pattern of phenol extracted
          LPS from T. f errooxians . ........... 5P»
15.  Sugar components of LPS from T_. f erroo xi dans
16.  Bubble tube assay for iron oxidation by intact
          cells of T_. ferrooxidans	65

17.  Effect of inorganic pyrophosphate concentration
          on inorganic pyrophosphatase activity
          with varying Mg   concentrations	7?

18.  Utilization of glucose by iron-grown T_.
          ferrooxidans	92

19.  Relationship of the rate of iron oxidation
          by resting cells to RuDP carboxylase
          activity	Of

20.  Radiorespirometric pattern for the utilization
          of specific   C-labeled glucose by iron-
          glucose-grown cells of T_.  ferrooxidans	105

21.  Proposed pathway for glucose dissimilation in
          T. ferrooxidans	   125
                           Viii
-------
                         LIST  OF TABLES

                                                           Page
 1.   Purification procedure for rhodanese from sulfur
          grown T_. ferrooxidans .............    1?

 2.   Requirements and specificity for rhodanese enzyme
          from sulfur grown T_.  ferrooxidans  .......    13

 3.   Effect of various inhibitors on rhodanese activity •    Ir

 4.   Purification of sulfite oxidase from T. ferrooxi-
          dans grown on sulfur ..............   19

 5.   Requirements for sulfite  oxidase from T_.  ferrooxi-
          dans. .....................   21

 6.   pH change after growth of Thiobacillus  strain J
          on selected energy sources ...........   2?

 7.   Generation times of Thiobacillus strain J on
          various substrates ...............  31

 8.   Comparative sensitivity to drugs of Thiobacillus
          thioparus and Thiobacillus strain  J grown
          on 1% thiosulfate agar plates ..........  33

 9.   Sugar composition of lipopolysaccharide (LPS)
          from T_. ferrooxidans ...............  5?

10.   Elemental analysis of lipopolysaccharide (LPS)
          from T_. ferrooxidans ............... f f

11.   Buffers tested for use in the iron oxidation assay-  •   £7

12.   Effect of various SO.  containing salts on the rate of
          iron oxidation .................. CO

13.   Effect of various anions on iron oxidation in a system
          containing only Cl" and in the same system
          containing Cl" plus SO 2~ ............ 7n

14.   Inorganic pyrophosphatase activity of  different
           cell preparations ...............   7^1
15.   Purification of inorganic pyrophosphatase ......   7f
                             IX
-------
                                                             Page

16.  Effect of cations on inorganic pyrophosphatase ..... 75

17.  The effect of metabolic inhibitors on inorganic
          pyrophosphatase .................. 73

18.  Effect of organic supplements) on iron oxidation by
          resting cells of T_. ferrooxidans previously
          grown in the iron-supplemented medium ....... 94
19.   Ribulose 1,5-diphosphate carboxylase levels in
          extracts of T_. ferrooxidans grown in the
          iron-supplemented media
20.   Growth of T_. ferrooxidans on organic substrates.  ....  -2

21.   Effect of growth substrate on enzymes involved in
          glucose metabolism by T_. ferrooxidans  ......  ?t?
22.   Requirements for the Entner-Doudoroff enzyme assay .  .  .101

23.   Effect of organic substrate on the activity of
          glucose-6-phosphate dehydrogenase and the
          Entner-Doudoroff enzymes ..............  1 03

24.   Activity of TCA cycle enzymes and NADH oxidase in
          cell extracts of T\ ferrooxidans grown on
          various media ...................  134

                    14
25.   Utilization of   C-labeled glucose by iron-glucose-
          grown T_. ferrooxidans ...............  1 07

26.   Utilization of   C-labeled glucose by glucose-grown
          T . ferrooxidans ..................  1 *%
27.  Purification of glucose-6-phosphate dehydrogenase
          from T_. ferrooxidans grown in an iron- glucose
          medium ...................... •

28.  Purification of glucose-6-phosphate dehydrogenase
          from glucose-grown T_. ferrooxidans ........ 1
 29.  Comparison of reaction velocity on glucose-6-phosphate
           dehydrogenase with NAD+ and NADP+ during
           purification of the enzyme from T_. ferrooxidans. . . .  113

 30.  Electrophoretic migration of purified glucose-6-
           phosphate dehydrogenase from T_. ferrooxidans. ......  114
-------
                                                            Page

31.   Comparison of apparent Michaelis
          constants for partially purified
          T_.  ferrooxidans glucose-6-phosphate
          dehydrogenase 	 117

32.   Effect of adenine nucleotides on T_.
          ferrooxidans glucose-6-phosphate
          dehydrogenase 	 11^
                               XI
-------
                         SECTION I

                        CONCLUSIONS
The studies presented herein describe basic laboratory research
on the structure and function of T_. ferrooxidans grown under
different conditions relating to acid mine drainage.  The
purpose of these basic studies is to try to find a specific
aspect of the structure or metabolism of T_. ferrooxidans which
can be used as a means of abating acid pollution at its source.
Specific conclusions in this regard are as follows:

     (1)  The inhibition of the rhodanese enzyme by iodo-
          acetamide and N-ethyl-maleimide indicate that
          thiol groups are required for activity.

     (2)  The inhibition of the sulfite oxidase enzyme by
          low (40 uM) concentrations of phosphate would be
          of practical use.

     (3)  The absolute requirement of Mg   for inorganic
          pyrophosphatase activity.

     (4)  The inhibition of inorganic pyrophosphatase activity
          by ethylenediaminetetraacetate.

     (5)  The cell envelope lipopolysaccharide is important
          in maintaining a balance between the environmental
          pH and the intracellular pH.

     (6)  When a heterotrophic substrate such as glucose is
          included in the medium, T_. ferrooxidans will convert
          from chemolithotrophic to heterotrophic growth after
          a period of energy storage.

      (7)  Growth of T_. ferrooxidans on glucose as the sole
          carbon and energy source does not result in the
          generation of acid, sulfate or ferric precipitates.

      (8)  The isolation and identification of a Thiobacillus
          species which produces alkaline products.

      (9)  More basic research needs to be done on T_. ferrooxi-
          dans.
-------
                        SECTION II
                     RECOMMENDATIONS

The following recommendations are made based on the afore-
mentioned conclusions of this research:

     (1)  More basic research into the function of the cell
          envelope as it relates to the oxidation of a
          soluble inorganic substrate, like iron;  the
          oxidation of an insoluble inorganic substrate,
          like sulfur;  the oxidation of an organic
          substrate like glucose.

     (2)  Basic research into how the cell envelope
          maintains a balance in pH between the acidic
          environment and the less acid cytoplasm.  It
          appears that the outer cell envelope will be
          the site of the cell which will be ameanible
          to the action of inhibitor compounds in the
          acid mine drainage environment.

     (3)  More research must be done on the effects of
          chemical inhibitors on the key enzymes involved
          in iron and sulfur utilization.  Preferably, the
          inhibitors would be specific enough to not cause
          other pollution problems.

     (4)  More research in the field, relating to trying
          to adapt the iron and sulfur utilization to
          heterotrophic growth.  This is a logical step
          since heterotrophic growth does not lead to the
          generation of acid, sulfate or ferric precipitates.

     (5)  More studies into the isolation and identification
          of microbes which will tolerate and grow in acid
          mine drainage, but will generate alkaline by-
          products.  This would help to raise the pH of
          mine drainage as well as causing less ferric
          precipitate to be formed.  Such an organism
          could successfully compete with T. ferrooxidans
          and render it ineffective since it is obligately
          acidophilic.
-------
                        SECTION III
                        INTRODUCTION
The iron oxidizing bacteria of the genus Thiobacillus
are widely distributed in acid waters associated with
deposits of metal sulfides and sulfide bearing coals.
They are indigenous to environments containing metal
sulfides, oxygen (either atmospheric or in aqueous
solution) and acid conditions.  The iron sulfides,
mascarite and pyrite, are found associated with coal
strata and upon removal of the coal, the sulfur bearing
iron compounds are exposed to air and water.   This
exposure establishes the first step in developing an
ecosystem which will maintain sulfur and iron oxidizing
bacteria.  Both chemical and biological oxidation of the
iron containing minerals lead to sulfuric acid and the
red to red-brown sludge deposits of oxidized iron pre-
cipitates which are associated with acid mine drainage.
The generation of the sulfuric acid and ferric precipi-
tates, called "yellcw boy", are a major water pollution
problem in this country and elsewhere.  It has been
estimated that 10,000 miles of streams and 29,000
surface acres of impoundments and reservoirs are seriously
affected by acid mine drainage.  The drainage from active
and abandoned mines amount to over 4 million tons of
acidity per year in this country.

The magnitude of this pollution problem ana its concomi-
tant economic and environmental problems have • rompted
the study of the causes of this type of pollution, with
hopes that some aspect of the generation of acid mine
drainage may be found that can be exploited to stop the
causative reaction.

The basic reactions involved in the generation of insoluble
ferric precipitates and acid can be expressed chemically as
follows:

         2FeS2+ 2H20 +  702	> 2FeS04 + 2H2S04        (1)
       (pyrite)                    (ferrous sulfate)
         4FeS0  +  20  + 2HS0 - > 2Fe(S0)  + 2H0    (2)
                                >2Fe(OH)3 + 3H2S04      (3)
                                 (ferric hydroxide)
-------
Pyrite oxidation can also occur in the presence of ferric iron


        +  14Fe3+ +  8H20 - > 15Fe2+ + 2So|" + 16H+        (4)
This reaction  (4) also generates acid thus keeping the pH low
so that the reaction may occur in acid streams.  However,
reaction  (2) proceeds slowly below pH 4 in the absence of
a catalyst, but proceeds rapidly in the presence of iron
oxidizing bacteria.  Thus one might consider that the
regulation of  the production of acid mine drainage conditions
depends upon the rate of reaction (2) which has been found
to be  controlled by the presence of iron oxidizing bacteria
(Thiobacillus  ferrooxidans) .  The bacteria have been shown
to accelerate  reaction (2) by a factor of more than 10^ over
the chemical reaction rate  (1).  Thus the generation of Fe3+
in reaction  (2) can cause the chemical oxidation of pyrite
(reaction  (4)) producing more acid.

Thiobacillus ferrooxidans can also oxidize sulfur and sulfur
containing compounds to sulfuric acid.  Understanding the
metabolism of  these sulfur  compounds is also important to
the study of methods for control of acid mine drainage.
In the biogeochemical sulfur cycle in nature, the sulfur
oxidizing bacteria play an important oxidative role by
converting the reduced sulfur compounds (t^S and S°) to
oxidized sulfur compounds (SO^", s20,  , SO.2", etc).
The anaerobic sulfate reducing bacteria complete the cycle by
reducing the oxidized sulfur compounds back to the reduced
state.

When one refers to iron oxidizing bacteria or sulfur oxidizing
bacteria, one is describing the mode of energy generation for
the bacterium, i.e. by oxidation of reduced iron and sulfur
compounds.  These types of bacteria obtain their carbon for
cell growth and reproduction from CC^, thus the organisms are
referred to as chemolithotrophs, or simply, autotrophs.  In
acid mine drainages, there are small but significant quantities
of organic carbon which could support heterotrophic growth,
that is, growth and energy production are derived from the
metabolism of organic compounds.  Thus, it would be important
to study the metabolism of organic compounds by iron and
sulfur oxidizing bacteria to try to determine if there is
some way in which to "turn off" the oxidations of the iron
and sulfur compounds which produce acid and ferric precipi-
tates, and "turn on" oxidation of organic compounds which
would not produce acid or ferric precipitates.
-------
The research performed under this grant involved four general
areas related to the basic study of the microbe which causes
acid mine pollution.  Because of this, the body of the report
will be divided into four separate sections, and each section
will have its own description of materials and methods used,
results obtained and the significance of the results found.
The four general areas involve studies of sulfur metabolism,
the isolation and characterization of new thiobacillus species
from alkaline mine drainage, iron metabolism, and heterotrophic
growth of T. ferrooxidans under acid conditions.
-------
                         SECTION IV
  STUDIES OF THE METABOLISM OF THIOBACILLUS FERROOXIDANS

                 GROWN ON ELEMENTAL SULFUR

Introduction

Thiobacillus ferrooxidans, formerly called Ferrobacillus
ferrooxidans can use elemental sulfur as the sole source of
energy, oxidizing the sulfur to sulfate.  The oxidation is
via sulfite involving the sulfur oxidizing enzyme (2).
Thiosulfate can be formed by a chemical condensation of
sulfite and sulfur.  This thiosulfate has also been found
to be converted to tetrathionate by the thiosulfate
oxidizing enzyme (3).  Thiosulfate can be converted to
sulfur and sulfite by the enzyme rhodanese, first discovered
by Sorbo (4) in 1953 in ox liver.  Sulfite can be oxidized
to sulfate by sulfite oxidase, with the concomitant
production of ATP energy for cell growth.  Both of these
latter enzymes have been studied in other Thiobacillus
species  (5, 6, 7, 8, 9, 10, 11, 12) but their presence in
T_. ferrooxidans was not determined until the present work.
The methods used and results obtained for each enzyme
isolation will be presented separately, followed by a dis-
cussion of both enzymes.

Isolation and Properties of the Rhodanese Enzyme.

                   Materials and Methods

     Organism

     Thiobacillus  ferrooxidans  (Ferrobaci 1 lus ferrooxidans) was
     grown  on the  9K salts-elemental sulfur medium, harvested
     after  7 days  growth,  and washed as previously reported
      (13).

     Preparation of  Cell-Free Extracts and Enzyir.3 Purification

     Cells were treated with 0.5 M Tris-HCl buffer (pH 8.5)
     overnight at  4°C, and then  centrifuged and resuspended
     in  0.01 M Tris-HCl containing 10~3M Na S 03 at pH 7.8
      (Buffer A) to form a  20%  (w/v) cell suspension;  thio-
     sulfate was added to help stabilize the enzyme (5).  The
     cells were disrupted  ia a water-cooled (6-8°C),
-------
10 kcycles/s Raytheon sonic oscillator for 20 min.
Whole cells and debris were removed by centrifugation
at 13,000 x g for 20 min.  The supernatant, containing
the crude extract, was then centrifuged at 105,000  x g
for 1 hr at 4°C in the Beckman model L-2 ultracentrifuge
equipped with a type 65 angular rotor.  The clear,
yellow supernatant from this step was removed, and  the
pellet resuspended in buffer A and recentrifuged at
105,000 x g for 1 h.  The two supernatants (containing
the enzyme) were combined.  About 99% of the enzymatic
activity of the crude cell-free extract was found in the
supernatant fraction.

Supernatant fraction  (27 ml) was treated with cold  1 N
acetic acid to a pH of 5.0, and a volume, equal to  the
supernatant fraction, of 2% streptomycin sulfate was
added;  the resulting turbid solution was stirred for
30 min in an ice bath at 0-4°C.  Precipitated protein
and nucleic acids were removed by centrifugation and
discarded. To the resulting supernatant, solid ammonium
sulfate was slowly added to 25% saturation (while
maintaining the pH at 5.0 with 1 N acetic acid) and
stirred for 30 min in an ice bath.  After centrifuging
and discarding the pellet, the supernatant was treated
with ammonium sulfate to 90% saturation, stirred for
30 min, and allowed to stand for at least 1 h in an
ice bath.  The precipitate was recovered by centrifugation,
resuspended to the original volume with buffer A, and
dialyzed overnight against 21 liters of the same buffer.
The dialysis buffer was then changed and the preparation
dialyzed for an additional 3 h.  To the dialyzed extract,
and equal volume of a 20% slurry of DEAE-cellulose
in buffer A was added and stirred for 1 h at 4°C.  The
enzyme, adsorbed to the cellulose, was recovered after
centrifugation and elution with 30 ml each of buffer A
of 0.05, 0.1, and 0.25 M.  Enzyme of highest specific
activity was eluted at 0.25 M buffer.  Further washing
of  the cellulose at higher concentrations of buffer
removed little additional enzyme.

Protein was estimated according to the procedure of
Lowry et al. (16) with crystalline bovine serum albumin
used as the standard.

Enzyme Assays

Rhodanese was assayed using a modification of the
procedure of Bowen et al. (5) with thiocyanate measured
colorimetrically.  The reaction mixture, unless otherwise
-------
indicated, contained 500 pinoles of Tris-HCl buffer
CpH 8.5), 50 ymoles of Na2S 0   50 ymoles of NaCN
(prepared in 0.1 M Tris-HCl, pH 8.5), enzyme (usually
0.1 ml), and water to a total volume of 5 ml.  The
reaction was initiated with the addition of NaCN to
individual screw-cap reaction tubes, and the reaction
proceeded at room temperature.  At the desired time,
0.1 ml of 38% formaldehyde was added to stop the
reaction and 0.5 ml of the ferric nitrate reagent of
Sorbo (20% Fe(N03)  in 3 N NHOj) was added.  The
reaction tubes were then centrifuged to remove denatured
protein and the intensity of the orange-yellow color
was measured colorimetrically in a Klett-Summerson
photoelectric colorimeter, using a blue (No. 42)
filter.  Thiocyanate levels were determined by comparison
to known standards;  reagent blanks were prepared by
reversing the order of the addition of cyanide and
formaldehyde.  Either the 90% ammonium sulfate fraction
or the DEAE-cellulose eluate was used in all enzyme
assays.

Chemicals
All reagents were obtained from commercial sources;
enzyme grade ammonium sulfate from Nutritional Bio-
chemicals, sodium thiosulfate from J. T. Baker Chemical
Company, sodium cyanide and ferric nitrate from Fisher
Scientific Company, Tris from Sigma Chemical Company,
and DEAE-cellulose from Schleicher and Schull Inc.
All reagents were prepared in glass-distilled water.
                     Results

Purification of the Enzyme

Table  1  lists the enzyme purification steps and shows a
38-fold  purification.  All fractions were stable at room
temperature for approximately 4 h.  After 24 h at 4°C
enzyme activity was gradually lost, which probably
explains the different specific activities obtained from
experiment to experiment (Tables 1 and 2).  The purified
enzyme (DEAE-cellulose eluate) following freezing in a
dry  ice  - 2-methoxyethanol bath, can be stored indefinitely
at -20°C with little loss in activity.  However, 12 h
after  thawing, activity was completely lost at 4°C.
                         11
-------
                                                  Table 1

                 Purification procedure for rhodanese from sulfur-grown T_.  ferrooxidans
         Fraction
Volume   NaSCN
(ml)     formed*
        (ymoles)
                                                  Total
                                        Protein   NaSCN     Specific
                                        ,   , n  formed     activity     Purity
                                        Ung/nuj  (ymoles)  (ymoles/mg)  (fold)
                                                                                           Recovery
r\j
Crude extract        27

105,000 x g          27
Supernate

Supernate, pH 5.0    52
and 2% strepto-
mycin sulfate

0-25% (NH4)2S04      52

supernate
0.500

0.372


0.158



0.150
19.0    135.0

12.5    100.0


 6.3     82.2




 4.1     78.0
                                                                       0.263       1.00

                                                                       0.298       1.13
                                                                       0.251
                                                                       0.366
                                                      0.95
                                                      1.39
100

 74.4


 60.9




 57.8
25-90% (NHJ2S04
precipitate
DEAE -cellulose 31
eluate
0.196
0.08
3.0 60.8
0.08 24.8
0.653
10.00
2.48
38.02
45.0
18.4
         *NaSCN formed (micromoles) per minute per 0.1 ml of enzyme fraction.  Enzyme activity was deter-
          mined for 1 min as described in Materials and Methods.  The amounts of protein used were: crude
          extract, 1,90 mg;  105,000 x g supernatant, 1.25 mg; pH 5.0 and streptomycin supernatant, 0.63 mg;
          0.25% (NH ) SO. supernatant, 0.41 mg; 25-90% (NH ) SO. precipitate, 0.30 mg; DEAE-cellulose eluate
          —        H- £  H*                                 T- ^  ^r
          8 yg.
-------
                                     Table 2
Requirements and specificity for rhodanese enzyme from sulfur grown T_. f e rrpoxid ans
Micromoles
Deletions Additions NaSCN
formed
1.06
NaCN --- 0.08
SO 2" --- 0.11
Enzyme — 0.07
Enzyme Boiled enzyme 0.38
S20 " 50 ymoles cysteine 0.12
SO " 0.1% Mercaptoethanol 0.14
S 02~ 50 ymoles GSH 0.11
L. 
-------
The enzyme reaction was linear with respect to time up
to 10 min, and the reaction was linear with respect to
protein concentration up to 16 yg of protein.

Requirements and Specificity of the Reaction

Cyanide, thiosulfate, and enzyme are required for the
catalytic formation of thiocyanate (Table 2).  Neither
cysteine, mercaptoethanol nor reduced glutathione could
replace thiosulfate in the reaction.   In these and sub-
sequent assays, results represent the average of at
least two separate experiments, and reproducibility was
demonstrated with different batches of enzyme.

Effect of pH on Enzyme Activity

The activity of rhodanese plotted as a function of pH
has a broad pH optimum ranging from pH 7.5-9.0.

Effect of Temperature

Room temperature (25°C) was optimum for the enzyme assay.
Above 25°C, enzyme activity rapidly decreased with tempera-
ture.  However, in the crude 90% ammonium sulfate fraction,
boiling for 5 min did not result in complete loss of
activity (Table 2), due possibly to some protection
afforded by extraneous protein.

Effect of Substrate Concentration

The apparent ^ for thiosulfate and cyanide as determined
by the method of Lineweaver and Burk (14) was 5.8 x 10  M
and 1.1 x 10~^M, respectively.

Effect of Inhibitors on Enzyme Activity

Table 3 shows the effects of inhibitors on rhodanese.
The thiolalkylating reagents iodoacetamide and N-ethylmale-
imide inhibited the enzyme by about 85 and 20%, respectively.
The chelating agent EDTA and o-phenanthroline had no effect
nor did azide, arsenate, arsenite, fluoride, or n-hvdroxy-
mercuribenzoate.  Sulfite, one of the end products of the
reaction, inhibited the enzyme by about 40%.  Except for
mercury which caused 30% inhibition, metal ions were not
tested because they caused the nonenzymatic formation of
thiocyanate (15).
                        14
-------
                         Table  3

    Effect  of various  inhibitors on rhodanese activity*
Addition Concentration (M) e c^ 8
control
None

p-Hydroxymercuribonzoate

EDTA

Arsenate

Arsenite
o-Phenanthroline

NaF
HgCl2

GSH

N-ethylmaleimide
Na2S03

lodoacetamide

NaN3
Mercaptoethanol
—
_3
10 •*
_2
10
_2
10
_2
10 *
ID'3
_2
10 ^
in"3
_2
10 Z
_•*
10 A
io-3
_3
10 *
_•?
10 A
0.1%
100

98.9

101.1

98.9

101.1
100

97.7
71.3

95.4

79.3
63.2

16.1

101.1
98.9
*Enzyme activity was determined after 10 min as described in
 Materials and Methods with additions as indicated.  Protein
 (0.45 mg) of the 90% ammonium sulfate fraction was used.
                             15
-------
Isolation and Properties of the Sulfite Oxida.se Enzyme

                  Materials and Methods

     Organism and Growth Conditions

     Thiobacillus ferrooxidans strain TM was  grown  on the  9K
     salts medium (16)  supplemented with 0.8  ml of  a 0.5 M
     solution of FeS04- 7H 0 per liter of medium.   Colloidal
     sulfur (0.5%) was  added to the salts solution.  The
     culture was grown in three 10-liter quantities of
     medium contained in New Brunswick Model  F-14 fermentors
     and the medium was steamed at 110°C for  30 min prior
     to inoculation.  The initial pH of the medium was  3.2.
     The cells were incubated in the fermentors at  28°C with
     a stirring rate of 150 rpm and an aeration rate of
     750 ml/min.  The fermentors were inoculated with 5 ml
     of a suspension containing previously sulfur-grown cells
     suspended to a final concentration of 10^  cells/ml;
     growth was monitored by following a drop in pH to
     1.5-1.8.  The 30 liters of medium were run through a
     water cooled Sharpies air driven centrifuge and the
     cells collected, scraped from the bowl and suspended
     in distilled water adjusted to pH 3 with H2SO  .  The
     cells were washed once with the acid water by  centrifuga-
     tion at 4°C in a Sorvall RC2-B centrifuge.

     Preparation of Cell-free Extracts and Enzyme Purification.

     All procedures were done at 0-5°C.  The  washed cells  were
     suspended to 10% (w/v) in 0.1 M Tricine-NaOH buffer
     (pH 8.0) to a total volume of 60 ml and  disrupted  in  a
     water cooled, 10 Kcycles/sec Raytheon sonic oscillator
     operating at maximum output.  Three separate 5 min
     treatments were required to break 90% of the cells.
     The oscillator bowl was cooled for 2 min between each
     disruption.  Phase contrast microscopy was used to monitor
     breakage of the cells. Cellular debris was removed by
     centrifugation at  27,000 x g for 20 min  in a Sorvall
     RC2-B centrifuge equipped with a SS-34 rotor.  The super-
     natant was then centrifuged at 105,000 x g for 1.5 h  in
     an International Preparative Ultracentrifuge Model B-60,
     with a type A-211  rotor.  The 105,000 x  g  supernatant
     was collected and  labeled crude cell-free  extract.
                             16
-------
The crude extract was applied to a 2.5 x 40 cm DEAE-
cellulose column equilibrated with 0.02M Tricine-NaOH
buffer at pH 8.0.  The sulfite oxidase enzyme ran with
the solvent front (Tricine-NaOH buffer) and was collected
in about the same volume (60 ml) as had been applied to
the column.  The eluate was then concentrated to a 10 ml
volume by ultrafiltration through an Amicon Model 52
Ultrafiltration Cell with a Type XM-50 filter.  The
residue retained by the ultrafilter was mixed with
calcium phosphate gel (4.0 rog/mg protein) and stirred
for 30 min.  The gel was collected by centrifugatiori
and the pellet eluted with 0.15 M Tricine-NaOH buffer
(pH 8.0).  The purified enzyme was stored at -20°C.
Protein was estimated according to the Lowry procedure
(17) with crystalline bovine albumin as the standard.

Polyacrylamide Disc Gel Electrophoresis - The standard
disc gel electrophoresis procedure of Davis (18) was
used to check the purity of the enzyme.  Protein from
each purification step was electrophoresed in triplicate
in 6.0% polyacrylamide gel.  The protein preparations
were layered on the lower gel in 5% sucrose and electro-
phoresed in a Poly-Analyt disc electrophoresis apparatus
(Buchler Instruments, Inc.,Fort Lee, N. J.).  The upper
buffer was 0.042 M Tris-Glycine (pH 8.9 at 25°C) and
the lower buffer was 0.12 M Tris-HCl  (pH 8.1 at 25°C).
The current was maintained at 2.5 mamp/gel until the
bromophenol blue tracking dye reached about 1-2 mm
from the bottom of the gel.  The gels were removed,
fixed in 20% trichloroacetic acid  (TCA) for 20 min,
stained for 45 min with Coomassie  Blue  (1% aqueous
solution diluted 1:20 with 20% TCA) and then destained
with 10% TCA.

The stained gels were scanned at 500 nm in a Linear
Gel Scanner  (Model  2410, Gilford  Instrument Laboratories,
Inc., Oberlin, Ohio).

Enzyme  Assay  - The  sulfite oxidase was assayed using a
modification  of  the procedure of Charles and Suzuki  (10).
The standard  reaction mixture contained:  0.1 mmole of
Tricine-NaOH buffer at pH  8.0,  1.5 ymole of K Fe(CN)6,
35 ymoles  of Na  SO   (low in PO   ) in  5 mM EDTA, enzyme
and double distilled water to   3.0 ml.  The enzyme
reaction was  followed as the decrease  in absorbance at
420 nm  in  a Perkin-Elmer Model  124 dual beam spectrophoto-
meter,  using the absorbance scale  expander.
                           17
-------
Molecular Weight Determinations

The molecular weight of the sulfite oxidase was determined
losing poly aery 1 amide gel concentrations of 5.5, 6.0, 6.5
and 7.0% according to the method of Hedrick and Smith
(19).  The gels were fixed, stained, scanned and the
R 's measured.  The R  values were plotted against gel
concentration and the slope of the line was determined
using a computer program written by Mr. Amnon Liphshitz
(Department of Biology, Syracuse University).  The
molecular weight was determined from a standard curve
of proteins of known molecular weight (Liphshitz, A.
and Lebowitz, J.:  Biophysical Society Abstracts,
p. 70a, 1971).

All procedures described were conducted at least 3 times
on 3 separate batches of cells grown on sulfur.  The
reproducilibity of the procedures was good and the Rm
of the purified enzyme on the polyaery1amide gel electro-
phoresis was the same in each case.

Determination of Labile Iron

     Labile ferrous and ferric ion were determined on
1 mg of the partially purified enzyme using the procedure
of Suzuki and Silver (20).

Reagents

N-tris (hydroxymethyl)-methyl glycine (Tricine), Horse
heart cytochrome £, type III, EDTA (Na.) were obtained
from Sigma Chemical Co.;  K,Fe(CN)fi from J. T. Baker
Chemical Co.;  NagSO. (low in phosphate), TCA, 2,2'-
dipyridyl and colloidal sulfur from Fisher Scientific
Co.;  Polyaery1amide gel reagents from Canal Industrial
Corp.;  Ca,(PO^)2 gel from Nutritional Biochemicals Corp.;
DEAE cellulose from Schleicher and Schull, Inc.  All
reagents were prepared in double glass distilled water.

                    Results

Purification of Sulfite Oxidase

The purification results for sulfite oxidase are given
in Table 4.  To check for purity, after each step in the
purification scheme, polyacrylamide disc gel electro-
phoresis was done on 160 yg of protein from the extract
-------
                              Table  4



Purification of sulfite oxidase from T.  ferrooxidans  grown on  sulfur



105,000 x g supernate
cone. DEAEC eluate
Ca (PO ) eluate
Total Protein
mg
317
75
7.2
Total -Units
(ymoles Fe(CN)6 reduced/hr)
317
104
53
Units /mg
protein
1.0
1.4
7.3
Fold
pure
1.0
1.4
7.3
-------
for each step.  Comparative scans of the Coomassie blue
stained gels showed a single major peak from the Ca-CPO.)^
eluate extract.  This peak was determined, by triangulation,
to be 75-85% of the protein present.  This major band was
cut from unstained gels, eluted in 0.05M Tricine-NaOH
buffer and enzyme activity determined;  the remaining
protein was re-electrophoresed and only one band was
noted.  An ultraviolet absorption spectrum of the enzyme
revealed only one peak at 270 nm indicating that the
protein must have been contaminated with some nucleic
acid.  Attempts at trying to stain the gels for sulfite
oxidase activity by the method of Fine and Costello
(21) were unsuccessful.  This was probably due to the
electronegativity of the sulfite which reduced the
phenazine methosulfate almost instantaneously thus
precipitating the nitroblue tetrazoleum.

Analysis for labile iron associated with the partially
purified enzyme using the 2,2'-dipyridyl method (20)
revealed no ferrous of ferric ion associated with the
enzyme.

Enzyme Stability

The 7.3 fold pure sulfite oxidase was stored routinely
at -20°C with no appreciable loss of activity upon
thawing.  The enzyme lost all activity by boiling for
15 min.  The enzyme was stable for 8 h at 3°C.  Enzyme
activity was linear with protein concentration to about
250 yg.

Molecular Weight Determination

The purified enzyme was electrophoresed in triplicate
at four different gel concentrations as described.  The
molecular weight was determined to be approximately
41,500.

Electron Acceptors
                                             •7
Sulfite oxidase could couple to either Fe(CN)°~ or oxidized
horse heart cytochrome £.  The K  for sulfite using
Fe(CN)3- as the electron acceptor1 was 0.538 mM.  The Km
for suifite was cytochrome c as electron acceptor was
0.578 mM.  The K  for Fe(CNT63~ was determined to be 250 yM.
The requirements for enzyme  activity are seen in Table 5.
                             20
-------
                         Table 5

  Requirements for sulfite oxidase from T. ferrooxidans
Additions or                     AA   /min x 10
  Deletions
Complete
2-
minus SO
minus enzyme
+boiled enzyme (15 min)
+TCA precipitated enzyme
19
0
10
10
10
The assay mixture was as described in Materials and Methods

The protein concentration was 170 yg.
-------
The Effect of pH and Buffer Concentration

When enzyme activity was determined as a function of pH
in 0.1 M Tricine-NaOH buffer, it was found that there was
no increase in activity from pH 5.4 to pH 9.0.  At pH
10.0 and above, enzyme activity increased and no specific
pH optimum could be determined.  It was assumed that
this phenomenon had no physiological significance, thus
all assays were conducted at pH 8.0.

The effect of increasing the Tricine buffer concentra-
tion on enzyme activity showed a gradual decrease in
activity to 60% at 130 yM.   When phosphate buffer was
added to the reaction mixture in the place of water,
an inhibitory effect was noted which completely inhibited
enzyme activity at a level  of 40 yM.  These data agree
with the sulfite oxidase isolated from T_. thioparus
(12J but not with the T_. no veil us enzyme (10).

Effect of 5'-AMP

A major reaction involving  sulfite oxidation is depen-
dent upon the condensation  of S'-adenosine monophosphate
and sulfite to form adenosine-5'-phosphosulfate (APS).
The reaction is catalyzed by APS reductase (22).

The effect of S'-AMP upon partially purified sulfite
oxidase from T_. ferrooxidans was tested.  When 1.5
ymole of 5'-AMP was added to the reaction mixture, no
effect on enzyme activity was noted.

                   Discussion

Explaining the mechanism of sulfur utilization in T_.
ferrooxidans was the objective of the present studies.
The presence of the rhodanese enzyme in sulfur grown
cells indicates that thiosulfate can be split to form
sulfite and a sulfur compound.  The presence of the
sulfite oxidase indicates that the sulfite can be
oxidized to sulfate with the production of ATP energy
for cell growth.

T_. ferrooxidans is similar  to other thiobacilli (5, 6,
7, 8) in possessing a rhodanese with a fairly broad pH
optimum (pH 7.5-9.0).  The  inhibition of rhodanese
activity by iodoacetamide and N-ethylmaleimide indicates
the involvement of thiol groups at the enzymes active
site.  The K  for thiosulfate and cyanide compares
-------
favorably to that reported for rhodanese from Chromatium
(23).  However, the unusually high value for cyanide
(K  of 10~2M) indicates that this compound is not the
physiological acceptor of the sulfur atom split from
thiosulfate.  Other experiments with mammalian rhodanese
(24) indicate that the S acceptor molecule is dihydro-
lipoate.  Dihydrolipoate was not used in the present
study.

The sulfite oxidase from T_. ferrooxid ans grown on
colloidal sulfur was partially purified and comprised
about 75-85% of the purified protein.  The enzyme could
couple to both ferricyanide and cytochrome £ and
exhibited the same apparent K  for sulfite when either
electron acceptor was used.  The apparent K  for SO,
of  .58 mM is 6.5 times higher than the 1C for sulfite
of the Thiobacillus thioparus enzyme (12) and 17 times
higher than the apparent K  for sulfite of the Thio-
bacillus novellus enzyme (11). This could be due to
the fact that T_. no veil us and T_. thioparus sulfite
oxidases were isolated from cells grown on thiosulfate,
whereas the J_. ferrooxidans enzyme came from cells grown
on  sulfur.  The latter organisms grows slowly on the
thiosulfate and the enzyme from the organism grown on
thiosulfate was not tested.

No  ferrous or ferric ion was found associated with the
partially purified enzyme.  The T_. thioparus (12) enzyme
did contain 1 mole of ferrous ion per mole of enzyme
and chelating agents inhibited the enzyme activity
indicating that the ferrous ion was essential for
activity.  EDTA and 2,2'-dipyridyl did not inhibit
enzyme activity in T_. ferrooxidans.

The pH optimum of the enzyme was high (pH 10.0)
relative to the T_. novellus  (11) and T_. thioparus
enzymes  (12), and is probably not the physiological pH
optimum.  This high pH optimum is interesting since,
T_.  ferrooxidans is strongly acidophilic, where the pH
of  the culture medium prior to harvest was 1.5 to 1.8.
It  is assumed that this enzyme in_ vivo is not in direct
contact with the outside medium.  The pH optimum is
probably closer to neutrality since Blaylock and Nason
(25) have shown, by breaking T_. f e rrooxi dan s in water,
that the intracellular pH is about 4.8-5.0.  We can
offer no explanation for this pH abnormality for the
sulfite oxidase.
                         23
-------
The molecular weight of 41,500 was lower than the 54,000
reported for the T_. thioparus enzyme (12) .  Whether or
not this represents a loss of some enzyme subunit is not
known.  Tests for purity  (gel electrophoresis) did
indicate a homogeneous enzyme, but from the ultraviolet
absorption spectra  (max 270 nm) some nucleic acid con-
tamination is most  likely.

Our results indicate that the sulfite oxidase enzyme
is the enzyme responsible for energy production in
sulfur- grown cells.  The enzyme catalyzes the reduction
of oxidized cytochrome _c.  It has been shown that the
electron transport  system in T_. f^rrooxidans most
likely involves only cytochrome £ and cytochrome  polythionates?
 4 6
-------
The sulfur oxidizing enzyme (2) and thiosulfate oxidizing
enzyme (3) have been previously identified in sulfur-grown
T_. ferrooxidans.  The identification and isolation of
rhodanese and sulfite oxidase completes the metabolic
cycle which leads to energy formation in sulfur grown
Thiobacillus ferrooxidans.
-------
                         SECTION V

      STUDIES CONCERNING THE ISOLATION AND IDENTIFICATION OF

                    NEW THIOBACILLUS SPP.
Introduction

Mine drainages which have a pH of 4.5 or above are considered to
be alkaline.  These drainages, class II, class III, and class
IV, can contain appreciable quantities of ferrous iron (0-1000 mg/1)
as well as sulfate (500-10,000 mg/1).  The interesting question
becomes why are these drainages alkaline when they contain
ferrous and sulfate?  Why have not the iron-oxidizing bacteria
developed and oxidized the ferrous to produce ferric precipi-
tates and converted any reduced sulfur compounds to sulfate
and acid?  These questions prompted the present investigators
to look at the microbial content of alkaline mine drainage and
to try to find an organism which might be producing some product
which keeps the pH from decreasing.  From the standpoint of the
ecology of alkaline mine drainage, those bacteria present in
these drainages might lead to new insights into fundamental
situations which exist in a mining environment.

Isolation of  an alkaline producing Thiobacillus sp.

                   Materials and Methods

A sample of alkaline mine drainage  (pH 6.8) from an underground
mine near Graham, Kentucky, was used to isolate microorganisms
by selective  culture techniques.  The selective medium used
was the salts medium of Starkey (27) supplemented with 1%
sodium thiosulfate as the energy source.  Ten ml of mine water
was added to  the 90 ml of sterilized thiosulfate medium, which
was then incubated on a rotary shaker  (150 rpm) for 3-5 days
at 28 C.  After enrichment, thiosulfate agar plates were streaked
with the enriched culture in order  to isolate individual colonies
of bacteria.  In order to select colonies which produce alkaline
by-products,  phenol red indicator was added to the plates and
the presence  of alkaline producing  colonies was noted as a
coloi change  from orange  (pH  7.0) to purple  (pH 8.0).  The
alkaline producing organism was subsequently restreaked 4-5
times to determine purity of the organism.  This isolate was
used for further studies  and will be referred to as Thiobacillus
strain J since its taxonomic status is  currently under study.
                              27
-------
Substrate specificity

Thiobacillus strain J was tested for its ability to grow and
produce alkaline by-products.  Starkey salts (27) agar plates
were prepared with purified, washed agar (Difco) at 2%.
Phenol red was added to the liquid agar before solidification.
The initial pH's of the plates were adjusted with either H SO
or NH OH.  The autotrophic and heterotrophic substrates were
added at a 1% concentration.  When liquid cultures were used,
50 ml of the salts were contained in 250 ml nephalo flasks
with the appropriate energy source added at 1%.   All cultures
were incubated at 28 C, unless otherwise stated, and liquid
cultures were shaken on the rotary shaker at 150 rpm.

Electron microscopy

Metal shadowing of the specimen was done on electron microscope
grids.  The grids were examined in an RCA 2D electron microscope
operating at 50 kv.

Drug sensitivity

Thiosulfate agar plates containing a lawn of organisms were
prepared and appropriate drug sensitivity discs  (Difco) were
aseptically placed on the plates and the plates  incubated at
30 C.  Zones of inhibition were noted after 3 days incubation.

Mole % G + C
One to two grains  (wet wt) of thiosulfate grown Thiobacillus
strain J was frozen and sent to Dr. Manley Mandel  (M.D.
Anderson Hospital, University of Texas Medical Center,  Houston,
Texas) for cesium chloride density gradient centrifugation
to determine the mole % G + C of isolated DNA.

                          Results

Substrate specificity

The alkaline producing isolate, Thiobacillus strain J,  increased
the pH of the medium into the alkaline range after growth on
thiosulfate, acetate, citrate, pyruvate, glutamate, yeast
extract and nutrient broth (Table 6) .   A decrease  in pH was
noted after growth on fructose, lactose, sucrose,  glucose and
galactose.  Formate, thiourea, thiocyanate, dithionite,  sodium
sulfide, and ferrous sulfate would not support growth of strain
J.  Growth was not inhibited by 5% NaCl and 0.01%  phenol.
-------
                         Table  6




    pH  change  after  growth of Thiobacillus  strain J on
selected
Growth Substrate*
Na2S205
Sodium citrate
Sodium succinate
Sodium pyruvate
Sodium glutamate
Fructose
Lactose
Sucrose
Glucose
Galactose
Yeast Extract**
Nutrient Broth**
energy sources
initial pH
5.9
6.4
6.3
6.1
6.2
6.2
6.1
6.1
6.1
6.1
6.1
6.1

final pH
8.3
9.1
8.7
7.8
8.7
4.7
5.1
4.6
4.1
3.9
8.0
8.3
 *Substrates added to a concentration of 0.5  M.



**Substrates added to 0.5% (w/v).
-------
     It was  found that  strain  J would grow on  1% thiosulfate
     agar plates  in a temperature  range  from 2 C to 41 C.
     If the  pH of the thiosulfate  medium were  made acidic
     or basic with H2SO  or NH OH,  growth occurred at a pH
     range from 3.3 to  10.0.  No visible growth occurred above
     or below the pH's  indicated.

     Generation times were determined for strain J on selected
     autotrophic and heterotrophic substances  and are shown
     in Table 7.

     Electron microscopy

     Thiobacillus strain J is  a gram negative, short rod (2 ym
     long by 0.5 ym wide) with one polar flagellum as seen by
     shadow preparation in Figure  1.   The surface topology of
     this organism is similar  to typical gram  negative bacteria.

     Drug sensitivity

     When Thiobacillus  strain  J was tested for drug sensitivity,
     it was found to be sensitive  to Chloromycetin, Kanamycin,
     Neomycin, Novobiocin, Streptomycin, and Tetracycline
     (Table 8).

     Mole % G + C

     The  mole % G + C of the DNA isolated from Thiobacillus
     strain J was determined to be 64.3  with a cesium chloride
     density of 1.723.

Isolation of^ar^ jyid producingThiobacillus sp.

                   Materials and Methods

     Isolation of a new JIM ob aci 1 lus sp.

     An alkaline  mine water sample was obtained from the Depart-
     ment of Mines and  Mineral Industries of the State of
     Pennsylvania.  The outpourings were from  a bituminous
     coal mine in the Ohio River Valley  and had a recorder
     pH of 7.0.

     The  water sample was assayed  for the presence of iron
     oxidizing, sulfur  oxidizing organisms and heterotrophic
     bacteria.  The former were detected in the standard 9K
-------
                         Table 7




Generation times of Thiobacillus strain J on various substrates,
Growth Substrate*
pyruvate
glutamate
fructose
glucose
sucrose
thiosulfate
Generation Time
00
0.75
1.2
3.0
3.7
6.5
7.3
  "Substrates added to 1% (w/v) in Starkeys salts medium.
                              31
-------
Figure 1.  Electron micrograph of a carbon-platinum shadowed
           cell of Thiobacillus strain J isolated from
           alkaline mine drainage.  Bar marker = 1 pm;
           28,000 X.
-------
                        Table 8




Comparative sensitivity to drugs of Thiobacillus thioparus




 and Thiobacillus strain J grown on 1% thiosulfate agar plates
Drug Thiobacillus thioparus Thiobacillus
strain J
Chlororaycetin
Erythromycin
Kanamycin
Neomycin
Novobiocin
Penicillin
Streptomycin
Tetracycline
s*
R
S
S
S
R
S
S
S
R
S
S
S
R
S
S
 S = sensitive, R = resistant
                            33
-------
medium of Silverman and Lundgren (16) and the hetero-
trophs on nutrient agar (Difco).

To isolate reduced sulfur oxidizing bacteria from the
alkaline mine water, the medium of Starkey was used (27).
Thiosulfate was the energy source.

The thiosulfate and phosphates were autoclaved separately
and added to the salts aseptically.  The final pH of the
medium was 7.0.  When a solid medium was needed, 1.5%
agar was added  (Oxoid Agar, Consolidated Laboratories,
Inc., Chicago Heights, Illinois).  Yeast extract (0.1
and 0.5%) was added to the medium as needed.  Also
studied was Starkey's medium supplemented with 0.1%
glucose and 0.1% sodium glutamate.  To detect acid or
alkaline production, 0.02% phenol red was added to the
medium.  All media were autoclaved for 15 min at
121°C and appropriate additives combined as needed.

To obtain an enrichment culture of Na2§ 0- oxidizing
bacteria, 95 ml of Starkey's medium containing phenol
red  (pH 7.0) was incubated with 5 ml of mine water and
the  flask shaken for 7 days on a reciprocal shaker
at 28 C.  Contents of this flask (5 ml) were trans-
ferred to 95 ml of fresh medium contained in 250 ml
Erylenmeyer flasks and incubated for an additional 4
days.  This flask then served for an inoculum to
streak various  agar containing plates, where the
energy yielding substrates were those described above.

As a result of  these studies, two types of thiosulfate
oxidizing clones were isolated after 7 days of incuba-
tion.  These clones were twice re-transferred on different
media and their purity confirmed.  One organism was an
acid producer and the second formed base.  The acid
producer was readily cultivated on thiosulfate agar and
the  organism was propagated in liquid medium on thiosulfate
to obtain mass  quantities of the organism for further
studies.  This  organism appeared to be more predominant
in this mine water sample and it was selected for further
study.

Growth parameters were examined with the organism grown
on thiosulfate;  thiosulfate utilization, medium pH
changes and cell dry wt were assayed using  conventional
procedures.  Thiosulfate was determined by  the colori-
metric method of Sorbo  (15), pH was measured with a pH
                         34
-------
meter and cell dry wt was determined by weighing a ml of
cell suspension after drying overnight and subtracting
the weight of an uninoculated medium control sample
treated in an identical manner.

The capacity of thiosulfate-grown cells to oxidize
different substrates was assessed manometrically using
intact cells.  For thiosulfate oxidation the reaction
vessel contained 60 ymoles of Tris-HCl buffer (pH, 8.0),
20 ymoles of potassium phosphate, 5 ymoles of Na-S 0
and 4 mg (dry wt) of whole cells.  Two tenths of a ml
of 20% NaOH was used in the center well and water was
added to the reaction vessel to make a final volume
of 2.2 ml.   Sulfite, tetrathionate and elemental sulfur
were also examined in the Warburg flasks.  These addi-
tions were sodium sulfite, in 5 mM EDTA (5 ymoles),
sodium tetrathionate (5 ymoles) and elemental flowers
of sulfur (48 mg).

The effects of various organic substrates on sulfite
oxidation were also examined;  0.5% of yeast extract,
0.5% of glucose and 0.5% of sodium glutamate were the
compounds tested.  They were added directly to the
Warburg flasks.  Oxygen uptake by cells in the pre-
sence of yeast extract alone was also tested.

The major cell types isolated were examined under the
electron microscope with negative stains (1% ammonium
molybdate  (pH 5.2)) and shadowed preparations (carbon-
platinum) .

                     Results

The acid producing Thiobacillus species was a small
gram-negative rod which grew either autotrophically or
heterotrophically.  Autotrophic growth gave clones which
were ropy and yellow-orange in appearance.  On an organic
medium, they  appeared white and mucoid with a raised
center.  Heterotrophically grown cells were often found
in pairs and  grew to a comparatively large size.  Cells
appeared to be motile but flagella were not readily
demonstrated  (Figure 2, 3).  Autotrophically grown cells
also appeared to be motile and somewhat smaller than
heterotrophically grown cells  (Figure 4).

Figure 5 shows the growth parameters of the acid producing
isolate propagated on thiosulfate as the energy supply.
                          35
-------





Figure 2.  Electron micrograph of carbon-platinum shadowed
           cells of an acid producing Thiobacillus grown
           heterotrophically,  which was isolated from alkaline
           mine drainage.  32,900 X.
                               3(=
-------

Figure 3.   Electron micrograph of negatively  stained cells  of
           an acid producing Thiobacillus  grown heterotrophically,
           which was isolated from alkaline mine drainage.
           28,200 X.
-------


Figure 4.  Electron micrograph of carbon-platinum shadowed cells
           of an acid producing Thiobacillus grown autotrophically
           on thiosulfate, which was isolated from alkaline mine
           drainage.  32,900 X.
                              38
-------
         7 -
                                 Thiosulfate
                                 Dry weight
                              -x PH
                 20      4O      60       80

                              Time (hrs)
  40   8.5
  35   8.0
                                                   30 £ 75
                                                      Jt
                                                      •o
                                                   25 | 7.0
                                                      "5

                                                      o
                                                   20 1 6.5
                                                   15    6.0
                                                   10   55
                                                             >,
                                                             8
100
Figure  5.  Growth parameters of an  acid producing Thiobacillus
            grown on thiosulfate.
-------
The cell dry wt increased with time and growth leveled off
as the substrate was diminished.  Thiosulfate was oxidized
up to 60 h with time with the concomitant accumulation of
acid.  The mechanism of oxidation agrees with that suggested
for some of the thiobacillus organisms:
         S2°3  + 40 + H2°	^2S04 + 2H+
Acid accumulated until a pH of about 4.0;  a lower pH
appears to be detrimental to the organism.  Intact cells
of the acid producing isolate also oxidized other reduced
sulfur containing compounds (Figure 6).  Thiosulfate
under these conditions was the preferred substrate with
tetrsthionate, elemental sulfur and sulfite oxidized in
that order.  Organic substrates supplemented the oxidation
of the reduced sulfur compounds (Figure 7).  Yeast extract
was by far the better source of nutrient.  This result
was supported by growth studies in liquid and solid media
as well as in manometric studies (Figure 8).   Yeast
extract, in the absence of any other substrate, was
oxidized readily by this organism.  Endogenous oxidation
represents 0  uptake by the organism in the absence of
any exogenous substrate.

Although alkaline mine water does contain organisms that
can oxidize ferrous iron, the aforementioned thiosulfate
oxidizing isolate  did not oxidize iron during the 7 day
incubation period tested using the standard 9K medium.

                    Discussion

A new Thiobacillus species has been isolated and
characterized from alkaline mine drainage.  The new
species, designated at present as strain J, was studied
taxonomically to try to relate it to known Thiobacilli.
From its growth characteristics and substrate specificity
strain J resembles Thiobacillus noveilus and Thiobacillus
intermedius (28).  The sensitivity to the drugs tested
indicates that strain J resembles T_. thioparus.  The
moles % G + C of the isolated DNA (64.3 mole %) is
extremely close to J_. thioparus and T. intermedius
(63.3 and 64.8 mole % respectively, M. Mandel, personal
communication) indicating a relationship to these two
known species.
-------
               30  60 9O  I2O  ISO  ISO 2IO 240 27O  3OO 33O 36O

                           Time (minutes)
Figure 6.
Oxygen uptake, measured  manometrically,  by the acid
producing Thiobacillus oxidizing reduced sulfur
compounds.  Four  mg  (dry weight) of cells were used
in each manometer flask.
-------
             700
             600
             500
          2
          53
          =  400
          c
          05
             300
             200
              100
•— • S03= + yeast extract
x— x S03= + glutamate
    S03= + glucose
                     •— • No S033
                        i No cells
                             90  120" 150 180  210  240
                             Time (minutes)
Figure 7.  Oxygen uptake, measured manometrically,  by  the acid
           producing Thiobacillus oxidizing  sulfite in the
           presence and absence of organic compounds.   Six
           mg (dry weight) of cells were  used  in each  m: r^meter
           flask.
                              42
-------
             0   30  60  90  120  150 180  210  240  270  300

                             Time (minutes )
Figure 8.  Oxygen uptake, measured manometideally, by  the  acid
           producing Thiobacillus oxidizing 0.5% yeast extract.
           Endogenous (no substrate) is also shown.  Two mg  (dry
           weight) of cells were used in each manometer -flask.
                            43
-------
We feel, however, that there are enough significant
differences between strain J and other known Thiobacilli
to call it a new species.  The mole % G + C is signifi-
cantly different from T. novellus (68.4 mole %).  Also,
T. novellus is sensitive to 5% NaCl, 0.01% phenol and
will not. grow on 0.5% citrate.  The mole % G + C of
strain J is similar to T_. thioparus, but T_. thioparus
has not been shown to grow heterotrophically on any of
the organic substrate  tested here.

The final pH attained after growth on thiosulfate is a
rather important and accepted method of differentiating
the Thiobacilli taxonomically.  T_. intermedius is fairly
 acidophilic  (growth range of pH 2-3) and does not increase
the pH of the medium after growth on thiosulfate (28).
Also, T_  intermedius is sensitive to 5% NaCl and will
not grow on 0.5% glucose, as strain J will.  The final
pH attained after growth of T. novellus is pH 5.0 (28)
which is much lower than that produced from strain J
 (pH 8.3 on thiosulfate, Table 6).  Also, strain J had
the ability to grow over a vast range of temperature
 (2 C to 41 C) and pH (3.3-10.0).  Other known Thio-
bacilli are incapable of this.  Strain J will not grow
 anaerobically using nitrate as the final electron
 acceptor, thus eliminating the possibility that it
 might be T. denitrificans,

The acid producing isolate, studied separately from
 strain J, was also a gram negative motile, short rod
 isolated from alkaline mine drainages.

 The Thiobacillus type isolate forming acid has many
 features similar to Thiobacillus intermedius a facultative
 autotroph.  There were some physiological differences
however, notably the failure of the isolate of this study
 to lower the pH of the medium below 4.0.  T_. intermedius
 lowers the pH to almost 2 when grown on a thiosulfate
 medium.  Also, T_. intermedius^ can oxidize sulfide, a
 compound which was not tested in the present study.
 Other than these differences most test parameters employed
were similar.  Thiobacillus novellus, another facultative
 autotroph, is not motile and is not pH tolerant.  It ^hus
 seems that the Thiobacillus isolated and studied here is
 either a new species of Thiobacillus or a variant of
T. intermedius.
-------
It is of interest that the natural environment of coal
mines established a niche whereby the various  thiobacilli
can survive and evolve.  One would probably expect that
since the organisms are closely related and that their
separation is by relatively minor physiological differences,
the complete spectrum of the organisms composing the genus
Thiobacillus inhabits mine waters.  This naturally remains
to be studied further.  However, it does seem to be appa-
rent that both types of mine water, acid and alkaline,
contain the same thiobacilli but that the predominance
of reduced iron and pyrite in acid drainage favors Thio-
bacillus ferrooxidans.  The alkaline waters favor the
reduced sulfur oxidizers, some of which are not acid
tolerant.  Whether or not these latter organisms contribute
to the alkalinity of the mine waters remains to be demon-
strated conclusively in the laboratory.
-------
                         SECTION VI

     STUDIES  CONCERNING THE METABOLISM OF THIOBACILLUS

                FERROOXIDANS GROWN ON IRON


 Introduction

 When considering the growth of T_. ferrooxidans, either in  acid
 mine drainage or under laboratory conditions,  it is of importance
 to  consider  the structure and function of the  outer layer  which
 surrounds the cell.  It  is this outer layer which is in  contact
 with the aqueous acidic  environment, and it would be reasonable
 to  assume that this layer is important in maintaining a balance
 between the  acidic environment (pH 2.5-3.5) and the less acidic
 cytoplasm  (pH 4.8-5.0 to above (25)).  The general anatomy
 and fine structure of J_. ferrooxidans has been previously
 described  (29).  However, the structure of the cell envelope
 layer has not been described in detail until now.  The first
 part of this section describes the fine structure and chemical
 composition  of the lipopolysaccharide outer layer of T_. ferro-
 oxidans grown on iron.                               ~

 The study of the importance of the outer layer of the cell as
 it  relates to maintaining a pH balance can also be coupled to
 the study of the site of iron oxidation.  In order to determine,
 at  the molecular level, where and how iron oxidation occurs,
 it  would be  beneficial to have a rather accurate kinetic assay
 procedure for determining rates of iron oxidation by whole
 cells.  In the second part of this section, a description  of
 a new and improved method of studying the rates of iron oxida-
.tion by whole cells is described.

 Another aspect of iron oxidation by T_.  ferrooxidans is the
 study of energy production.  As previously described, ATP,
 the cells energy molecule, is produced in the cell as the
 electron released in ferrous oxidation moves through the
 electron transport system.  The ATP molecule can undergo many
 reactions which are essential to cell growth.   When the ATP
 is  broken down to release energy,  sometimes the "high energy"
 molecule pyrophosphate is formed.   In order to convert this
 pyrophosphate molecule to two orthophosphates, which are
 necessary for ATP synthesis, the inorganic pyrophosphatase
 enzyme carries out the necessary reaction.   This reaction  is
 necessary to maintain the balance  between ATP, ADP and AMP
 in  the cell, the so-called "ATP charge" concept.  The
                             47
-------
inorganic pyrophosphatase enzyme was studied in T_. ferrooxidans
grown on ferrous iron.  The enzyme had not been previously
reported in chemoautotrophs until the present work was done,
which is reported in the third part of this section.

The structure and chemical composition of the cell envelope

                  Materials and Methods

     Cell culture.

     Thiobacillus ferrooxidans was propagated in 9K medium
     in 5-gal carboys under forced aeration at 28 C for
     54 hr and harvested as described by Silverman and
     Lundgren (16).  Cells at this time are in the late log
     phase of growth;  approximately 6 g (wet weight) of
     cells are obtained from 96 liters of medium.  Cell
     numbers are about 2 x 108 cells/ml and have a generation
     time of about 8 h.

     Ele ctron mi croscopy

     Cells (as suspensions), peptidoglycan (PG) and lipopoly-
     saccharide (LPS) to be examined as thin sections were
     fixed with 1.5% glutaraldehyde in S-collidine-hydro-
     chloride buffer (0.05 M, pH 7.6) for 10 min at room
     temperature, washed once (for 10 min) with the same
     buffer, and then fixed overnight at room temperature
     in osmium tetroxide (1.0%) in distilled water (pH 6.2).
     The aforementioned treatments were selected based upon
     results of different trials.  These appeared to give
     the best results.  The osmolality of the fixative was
     180 milliosmal, as calculated by using the method of
     Maser, et al  (30).  The fixed cells were embedded in
     2.0% agar prepared with distilled water, and the agar
     was cut into  1-mm cubes.  The embedded cells were
     dehydrated with ethyl alcohol (5 to 100%) and embedded
     in Epon 812 by the method of Luft (31) .  Sections were
     cut on an ultramicrotome with glass knives and then
     stained with  1.0% uranyl acetate (pH 4.5) for 30 to
     60 min at 60 C, followed by lead citrate  (32) for 5
     min at room temperature.  Purified LPS suspended in
     distilled water was also examined by negative staining
     by using methods previously described (33).  Thin
     sections and  negative stain preparations were examined
     in either  an  RCA EMU-2D or in a JEM-7 electron microscope.
                              48
-------
Extraction and purification of lipopolysaccharide.

LPS was extracted and purified by the classical method
of Westphal (34).

After centrifugation for 30 min at 3,000 x g, the aqueous
upper phase was withdrawn, an equal volume of distilled
water (preheated to 75 C) was added to the remaining
phases, and the extraction was repeated.  The aqueous
phase was again withdrawn, pooled with the initially
extracted layer and dialyzed against running tap water
for 6 h and against three changes of distilled water
for 20 h.  The dialyzed aqueous phase LPS was then
lyophilized and weighed.

Lyophilized LPS (50 mg) was dissolved in 7.5 ml of
distilled water containing 0.75 ml of 2% hexadecyltri-
methylammonium bromide (Cetavlon, Eastman Organic
Chemicals, Distillation Products Industries, Rochester,
N. Y.);  the solution was stirred at room temperature
for 15 min before centrifuging at 3,000 x g for 30 min
to partially remove ribonucleic acid (RNA; 35).

RNA associated with LPS was estimated to be 3% by
assuming that absorbancy at 260 nm was due to nucleic
acid and that 50 ug of RNA has an optical density (OD)
of 1.0.  The supernatant extraction was collected,
lyophilized, and dissolved in 3.0 ml of 0.5 M NaCl.
The LPS solution was then added to 10 volumes of
absolute ethanol and held at 0 C for 1 h until a
flocculant precipitate formed.  The ethanol mixture
was centrifuged, and the pellet was collected and
suspended in 4 ml of distilled water.  The solution was
dialyzed for 30 to 48 h against several changes of
distilled water in the cold.  The dialyzed preparation
was lyophilized and stored in a dessicator until use.
Different batches of LPS extracted from T_. ferropxidans
as described above were pooled and used for the sub-
sequent experiments.

Relative density and sedimentation velocity of LPS.

The relative density of LPS was determined by moving
zone centrifugation by using sucrose.  Centrifuge tubes
containing 50.9 to 68.0% sucrose were prepared in a
total volume of 3.6 ml and allowed to equilibrate over-
night.  LPS  (1.0 mg/1.2 ml of water) was placed on top
of the gradient and centrifuged at 60,000 x g for 90 min
                         An
                         T.*
-------
at 20 C.  After centrifugation, LPS was visible as a
slight band in the gradient.

The sedimentation velocity pattern of LPS was obtained
by dissolving 4 rag of LPS in 1.0 ml of 0.066 M phos-
phate buffer (pH 7.1).  The solution was placed in a
double sector cell and centrifuged at 20,410 rev/min
for 33 min in a Spinco model E analytical ultracentri-
fuge.  Photographs of schlieren optic patterns were
taken on metallographic plates (2 x 10 inch plates,
high green contrast, Eastman Kodak Co., Rochester,
N. Y.) at various time intervals.  All photographic
measurements were made on a Nikon enlarger (Nippon,
Kogaku, K. K., Japan) with a calibrated stage micro-
meter.

Chemical analyses of LPS.

The phosphorus content of LPS was determined by the
method of Chen, et al (36), and Taussky and Shorr
(37).  Total nitrogen was determined by converting
the organic nitrogen in 1 mg of lyophilized LPS to
NH  by acid digestion and determining NH_
with Nessler's reagent (38).  Iron was determined
by the method as described by Suzuki and Silver (20).
Magnesium and calcium were estimated by atomic absorp-
tion spectrophotometry (Analytical Methods for Absorp-
tion Spectrophotometry, Perkin-Elmer, Norwalk, Conn.).

Heptose was determined by the procedure of Dische
(39) as modified by Osborn  (40).  Hexose was deter-
mined by a modified anthrone method by using a 2%
solution of anthrone in ethyl acetate (41).  The
method of Aminoff, et al  (42) was used to determine
2-keto-3-deoxyoctulosonic acid.  Hexosamine was
determined by the method of Rondle and Morgan (43).
Glucose and galactose were determined by using
glucostate kits, respectively (Worthington Biochemi-
cal Corp., Freehold, N. J.).

Gas-liquid chromatography.  An F § M model 500 gas
chromatograph was used equipped with a hydrogen flame
ionization detector and 6-ft, U-shaped, 3% SE-30 column
(60 to 90 mesh).  The separation was done at 180 C.
Sugars were first converted to methyl esters without
directly determining yield;  2 mg of LPS was suspended
in 20 ml of dry methanolic 0.5 N HC1 and methanolysis
was done at 80 C for 48 h in a screw-cap test tube
                         50
-------
provided with a Teflon liner.   Lipid released during
methanolysis was removed by extraction with an equal
volume of hexane.  The extraction was repeated three
times.  HC1 was removed by repeated evaporation of the
sample to dryness under reduced pressure.   The dried
sample was taken up in 2 ml of methanol and passed
over a column of Amberlite IR-120 resin in the H form
which was prewashed with methanol.  Pooled washings were
evaporated to dryness under reduced pressure and
redissolved in 1 ml of methanol.  The final solution
was dried under a gentle stream of air.  Sugar-0-
trimethylsilyl (TMS) derivatives were prepared by the
addition of 1 ml of dry pyridine, 0.1 ml of trimethyl
chlorosilane, and 0.1 ml of hexamethyldisilazane.  The
mixture was shaken vigorously for 5 min;  1 to 3 yliters
was used for gas chromatographic analysis  (44) .

                    Results

Electron microscopy of the cell envelope.

Figure 9 shows a thin-section profile of normal iron-
grown cells possessing a multilayered cell envelope
typical of gram-negative bacteria and in particular,
similar to the thiobacilli (45, 46).  A detailed des-
cription of the organization of the layers in the
envelope of the iron-oxidizing chemoautotroph has been
given (29).

Chloroform-methanol extracted cells show an almost
complete removal of the outer layers of the envelope
(Fig. 10) except where two cells are firmly attached
and apparently protected from the solvent (Fig. 11).
The chloroform-methanol extraction removes LPS as
determined by chemical analysis.  The densely stained PG
layer of the cell is still intact after chemical extraction
and is identified in these micrographs based upon compari-
son to normal cells and on previous results of chemical
extraction studies  (33).  Also noted in Fig. 10 and 11
are translucent  areas surrounding the polyhedral bodies
seen in untreated cells  (Fig. 9).  The translucency is
believed due to materials lost by solvent extraction.
The solvent extraction seems to have damaged the cyto-
plasmic membrane but this is difficult to prove for even
untreated cells  are difficult to fix properly.  Proper
fixation of thiobacilli is a general problem which is
still not solved  (45, 46).
                         51
-------
                                        CM



Figure 9.   Thin section of a T_. ferrooxidans cell grown
           autotrophically on iron showing a multilayer
           cell envelope.  The labels are outer layer (OL)
           or lipopolysaccharide (LPS);   middle layer (ML)
           or peptidoglycan (PG);   cytoplasmic membrane (CM);
           ribosomal particles (R);   and membrane vesicle
           (M).   Bar marker represents 0.1 ym.
-------
Figure 10.   Thin section of T_. ferrooxidans after chloroform--
                                  Nucleus  (N), ribosomes (R)
methanol extraction.
and peptidoglycan  (PG).
0.2 ym.
                                     The bar marker represents
                               53
-------

Figure 11.  Thin section of T_. ferrooxidans treated as in Fig.  10.
-------
When phenol-extracted LPS was embedded,  sectioned,  and
examined in the electron microscope (Fig.  12),  typical
unit membrane structures were seen with  an approximate
width corresponding to that observed in  thin sections
of intact cells. The triple-layered structures  are
generally arranged in elongated and coiled formations
of various sizes.  The clearly resolved  areas of the
section of LPS have a measured thickness of about 8 nm.
A negatively stained preparation of purified LPS is
shown in Fig. 13.  The large perforated  sheetlike
structures and ribbon networks are typical of phenol-
extracted LPS from Salmonella typhimurium (47).  No
explanation is available for the holes in the sheets
but this appearance was consistent for many preparations.

Physical and chemical properties of LPS.

LPS extracted by phenol-water accounted  for approximately
4 to 6% of the dry weight of the cells.   The amount of
contaminating nucleic acid initially present was estimated
to be 3%;  this was reduced to 1.6% after Cetavlon  pre-
cipitation.  This level of contaminating RNA agrees with
the level of nucleic acid in purified LPS described by
Burton and Carter (48).

The LPS has a relative density of 1.28 based on results
of sucrose gradient centrifugation and was visible  as
a single band in the 59.4% sucrose fraction.  The LPS
had a sedimentation coefficient of 99.9S (uncorrected)
and an S     of 105 when corrections were made for the
viscosity"'and density of the buffer (Fig.  14).   The
latter value was based on a calculated partial specific
volume (V) of 0.79 cm3/g for the LPS (derived from its
relative density of 1.28).  The concentration dependence
of the sedimentation coefficient was not checked.

Results of chemical analyses of LPS for  sugars are  shown
in Table 9.  Fig. 15 shows a profile of a gas-liquid
chromatogram of the sugar components. The assignment
of each sugar to a particular peak is based on the
retention time of an a-methyl-D-glucoside standard
(Pfanstiehl Chemical Co., Waukegan, 111.)  given an
arbitrary value of 1.0.  The brackets depict the a and
8 isomers of the sugars.  The sugar components identified
in the LPS are similar to those found in the LPS isolated
from gram-negative heterotrophs (49). Nitrogen and phos-
phorus were present in LPS as was iron which was mostly
                         55
-------


Figure 12.   Thin section of purified IPS  from T_.  ferrooxidans
            showing the membrane-like tripartite  structure
            (12 nm).
                           56
-------



         •
Figure 13.   Negatively stained purified LPS  from T_.  ferrooxidjiiis^
            showing the ribbons and sheets of LPS.
-------
                                                                       D
Figure 14.  Sedimentation velocity pattern of phenol  extracted IPS
            from T_. ferrooxidans.  The sample was  centrifuged at 20,410
            rev/min at 20 C.  Approximately 4 mg of LPS was dissolved in
            0.066 M phosphate buffer (pH 7.1).  Time  (min)  after attaining
            speed:  (A) 16, (B) 20, (C) 24, (D)  28.   The bar angle was
            65°.  The sedimentation coefficient  of the LPS  was corrected
            for the viscosity__and density of the buffer. The partial
            specific volume (V) of the LPS was calculated from the
            relative Density of the macromolecule  (1.28), where
            1/1.28 = V.
                                   58
-------
                         Table 9

       Sugar composition of lipopolysaccharide (LPS)

                   from T. ferrooxidans.
                 Carbohydrate
 LPS
Hexose (total)

Heptose

Hexosamine

2-Keto-3-deoxyoctulosonate (KDO)

Glucose

Galactose
39.2

12.7

 4.5

 8.7

 3.0

 6.5
p
  Values of analyses here and in Table 10  represent  averages
  determined from at least three separate determinations
  which were done on samples of LPS extracted from different
  batches of cells.

  Assumed that when OD    minus OD    = 1.07, then
  L-glycero-D-mannoheptose = 1.0 mole.
c                                                          -3
  Assumed KDO had a molar extinction coefficient of 72 x 10
  and a molecular weight of 236.
                             59
-------
          •RHAMNOSIDE
             MANNOSIDE
             n
                                          HEPTOSIDE
                         8

                       MINUTES
                         12
16
Figure 15.
Sugar components of LPS from T_. ferrooxidans.
Sugars were determined by gas chromatography
as described in Materials and Methods.  One
yl of sample was injected, the sensitivity
range was 10, and attenuation at 32.  No KDO
was detected by this method since it is
destroyed during methanolysis.
-------
     in the ferric form.   Other cations present were calcium
     and magnesium (Table 10).

A new whole cell, iron oxidation assay

                  Materials and methods

     Organism and culture procedure.

     T_. ferrooxidans strain TM was grown on the 9K medium
     and in the manner previously described (16) in 16-liter
     glass carboys under forced aeration.  Cells were har-
     vested in a Sharpies centrifuge  after 48 h of growth,
     washined three times with  distilled water (pH 3.0),
     and stored in the cold (6  C) in  acidified 9K salts or
     6-alanine-SO 2~ buffer (pH 3.6,  0.01 M).   Stored cells
     were used within 4 days.

     Development of an assay system.

     Initially, a direct  colorimetric assay was used which
     measured the appearance of small amounts  of Fe^+, in
     the presence of variable  anions, as a red complex formed
     with sodium thiocyanate (2).  The reaction was stopped
     with 5 ml of 2 N HC1 prior to analyzing for Fe3+.  The
     addition of HC1 to tubes  containing Fe^+ gave rise to a
     yellow color, presumably  due to  the formation of a ferric
     chloride complex, and it was possible to estimate the
     color spectrophotometrically at  410 nm.   This Fe  -
     chloride complex formation served as a basis for the
     Fe3+ assay used in this report.

     The assay procedure measuring the appearance of Fe   was
     as follows.  Five acid-washed and thoroughly rinsed
     Bellco colorimeter tubes,  each containing 3.8 ml of
     distilled water adjusted to pH 3.0 with H-SO., were
     placed in a water bath at  35 C;   0.2 ml of a cell sus-
     pension (2 mg, dry weight  of cells) was added to each
     tube, and 0  was bubbled through the suspension via
     Pasteur pipettes.  To each tube, 1 ml of FeSO  • TH^O
     solution containing 2.5 mg of Fe2+ was added in rapid
     succession by use of a repeating syringe.  The oxidation
     was stopped at 0, 1, 3, 6, and 10 min by rapidly adding
     5 ml of 2 N HC1.  The HC1  was added before the addition
     of Fe^+ ion for the zero-time tube.  Tubes were removed
     and analyzed for Fe^* at 410 nm in a Bausch § Lomb
                             61
-------
                      Table 10

    Elemental  analysis of lipopolysaccharide (LPS)
           Element                            LPS
	yg/mg
Nitrogen                     •               9.25

Phosphorus                                  3.89

Mg2+                                        0.13

Ca2+                                        1.31

Fe2"*" and Fe3+                              83.00
-------
Spectronic 20 colorimeter.  The intensity of the color
was directly related to the amount of Fe^+ formed and
agreed well with Fe^+ determinations by the thiocyanate
assay method.

Kineti c analysi-S of^ i ron oxi.datipn^bywholeeel Is .

Five test tubes, each containing 3.8 ml of distilled
water adjusted to pH 3.0 with H-SO., were placed in a
water bath at 35 C.  This temperature had been established
as optimal for the system.  To each tube, 0.2 ml of cell
suspension was added and 0_ was bubbled through the
suspension via Pasteur pipettes.  The rate of gas flowing
through each tube was regulated by a needle valve to
ensure a constant bubbling rate.  At 15-sec intervals,
1 ml of a FeSO. solution containing 2.5 mg (500 yg/ml)
of Fe^"1" was added to each tube.  The oxidation of iron
was stopped in individual tubes by the addition of 5 ml
of 2 N HC1 at intervals of 2, 3.25, 6.50, and 10.75
min.  The development of the yellow ferric chloride complex
was estimated as above.  For these assays, a reaction
tube which received HC1 to stop the reaction at 1 min
served as the blank;  the assay was not considered
linear for the first minute.  The optical density (OD)
at 410 nm was plotted versus reaction time, and the rate
of change as OD  ^ per min was obtained from a line
fitted to the OD values of the reaction tubes.

The effect of whole cells and Fe   substrate concentrations
on the rate of iron oxidation was followed by varying the
concentration of cells or of iron in the appropriate
reaction system.

The pH of the reaction mixture of the assay system was
varied by using different amino acid buffers.  Glycine
and 3-alanine were dissolved in excess dilute H2S04 of
known normality and back-titrated to the desired pH with
KOH.  The final concentrations of components in the assay
system were:  Fe^+, 500 yg/ml; glycine or 3-alanine,
0.05M;  SO   , 0.034 M.  A variety of other buffers were
also tested in order to determine whether they could
replace the 3-alanine-SO 2~ buffer.

                 2-
The effect of SO    and Cl  on iron oxidation was
investigated.  The assay system contained Fe^+  (200 yg/ml,
as FeSO • 7H 0);  the S0^~ content was increased from
0.0036 to 0.05 M by the addition of K2S04.  The Cl~
levels were varied by preparing a series of assay systems
-------
in which all cations were the same and the anions  2_
consisted of various mole percentages of Cl  and SO. .
The test systems contained 8.94 x 10"3M Fe  , 10~2M
S-alanine, and a total anion concentration of 3.79 x 10~2M;
the pH was adjusted to 3.2 with KOH.

All assay results were reproducible and all experiments
were done in triplicate and with different batches of
cells.
                         2_
Anionic replacement of SO.  in the iron assay system.

The assay system to test for replacement of other anions
contained:  1.9 x 10"2M Cl", 4.5 x Wyi Fe  , and
5  x 10  M 8-alanine.  The fact that Cl" was not inhibitory
at these levels and could not satisfy the anionic require-
ments for iron oxidation led to the use of a system
containing Cl  as the only anion to determine whether
other anions could replace the SO.  requirements for
iron oxidation.  The various test anions were added as
the Na or K salt and were tested at 2.5 x IQ-^M, both
in the presence and absence of 2.5 x 10"^1 SO. .

                    Results

The bubble tube-type of iron assay measuring the ferric
chloride complex at 410 nm shows a linear increase in
Fe   after a brief initial lag (Fig. l6);3+the lag could
be due to the mixing requirement.  The Fe   levels can be
compared to those measured by the SCN  assay, which also
shows a linear response.  However, the SCN  assay has
limitations which restrict its use for kinetic analysis;
these limitations are not pertinent to this paper.  The
rate of iron oxidation ( AOD ._ . . ) was found to be
directly proportional to the concentration of cells used
in the assay over the AOD range of 0.02 to 0.085.

The effect of Fe + concentration (from 3.5 to 35.8 mM)
on the rate of iron oxidation revealed an apparent K
of this system to be 4.5 mM.

The standard iron assay procedure and various amino acid
buffers were used to obtain more precise information
concerning the effect of pH on iron oxidation by intact
cells.  Glycine-SO .2- buffer was used over the pH range
of 1.8 to 3.6, and B-alanine-SO.  buffer was used over
the pH range of 3.2 to 4.8.  The pH optimum for Fe
oxidation was broad, extending from 2.4 to 3.6 and a rapid
                        04
-------
.600
    .500
.400
  2.300
  d
  d
    .200
    .100
                                                  -i300
                                                   200 ~
                                                       E
                                                      X
                                                    100
                246
                   TIME, IN MINUTES
                                               10
Figure 16.   Bubble tube assay for iron  oxidation  by intact
            cells of T_. ferrooxidans.   Iron oxidation was
            followed by measuring Fe    with the SCN  complex
            and by measuring the Fe'+-chloride  complex at
            410 nm.
-------
decrease in iron oxidation occurred above pH 3.6 and
below 2.0.

The effect of pH on iron oxidation was investigated
further by observing its effect on the two kinetic para-
meters, K  and V   .   When $-alanine-sulfate buffer was
used in tne~assay system over a pH range of 2.4 to 4.4
with the Fe   concentration varied from 3.5 to 17.9 mM
for each pH, the K  remained constant throughout most
of the pH range.  The change in V    was more significant
in terms of the pH optimum, sincemi£ decreased below
pH 2.8 and above pH 326.  The K  determined in the pre-
sence of 8-alanine-S04~ buffer was 2.2 x 10  M.

Table 11 shows the effects of usi$g various buffers
in the assay system measuring Fe   oxidation by whole
cells.  Both formate and acetate buffer systems were
inhibitory.  Maleate, citrate, and malate buffers were
also unsuitable, owing to inhibition of the biological
oxidation or to some auto-oxidation of iron.  The slight
stimulation of iron oxidation observed with 3-alanine
may be due to additional SO.  added to the system, but
this did not occur in the case of glycine-SO. .

The iron assay system was also used to investigate_the
effects  of SO.  on iron oxidation.  Addition of SO.
to the reaction mixture, ranging from 0.0036 to 0.05 M,
approximately doubled the rate of iron oxidation.  Table
12 shows the effect of the addition of SO.  containing
salts.  The four salts tested all caused a significant
increase in iron oxidation.  The low increase with
                                                   lused
   n * j  n    •*•      ~              *•*•
by
(NH ) SO.  is probably due to  a slight inhibition cause
by NH.+ ions.  It was shown that the addition of SO.
     * •  Ji. Wllw •  JU \f « UW •JtlVniL b>ildfclu> WILW ClVilU-1. t>-i-\Sll \J -A. fc_/W JL
increased the velocity of the reaction but did not affect
the K .
     m

The effects of Cl~ on the rate of iron oxidation indicated
a slight stimulation of iron oxidation occurring at low
Cl  levels, but the rate decreased rapidly at high Cl
levels;  no oxidation was observed when Cl  was the only
anion present.  Since these levels of Cl~ do not inhibit
iron oxidation, the data indicate a specific SO.  require-
The fact that Cl  was not inhibitory and that chloride
could not satisfy the anionic requirements for iron
oxidation led to the use of a system containing Cl~ as
the only anion, to determine whether other anions could
                         66
-------
                           Table 11

      Buffers tested for use in the iron oxidation assay
Buffer system
pH tested
          Results
Formate-formic acid     3.2

Acetate-acetic acid     3.6

Maleate-NaOH            3.6
Citrate-NaOH
Malate-NaOH

Glycine-SO^"
 6-Alanine-SO
            2_
   3.6



   3.6

 2.1-3.7



   3.2
Inhibition (70%) at 5xlO~5M

Inhibition (55%) at 5xlO~2M

Slight auto-oxidation;
no biological oxidations
at 5 x 1(T2M

Slight biological oxidation;
large autooxidations at
5 x 10~2M
Same as maleate

No inhibition or stimulation;
rate of oxidation the same
as without buffer.

Slight stimulation of
oxidation over the control
with no buffer.
                               C7
-------
                   Table 12

                     2-
Effects of various SO.   containing salts  on the


            rate of iron oxidation.
SO ~ salt added

None
K2S04
MgS04
Na2S04
(NH4)2S04
SO in assay
M
0.0036
0.041
0.041
0.041
0.041
AOD410/min

0.033
0.066
0.060
0.059
0.046
-------
               2-
     replace SO.  .   When various  anions were  added to the    ,
     reaction system in the presence  and  absence of  2.5 x 10  M
     SO   to test  for both the  inhibitory effect of  anions and
     their ability to replace S0^~, it was noted9that both
     HPO2' and HAsO~~ could partially replace S0^~  (Table 13).
     Both of these anions stimulated  iron oxidation  in the
     presence of SO  } but the  stimulation was no greater than
     that observed when the equivalent amount of addition    -
     was added.   Borate appeared to be without  effect  and it
     did not replace S0^~.   Nitrate did not  replace  SO^  and
     was inhibitory in the  presence of SO^" ;  molybdate  did
     not replace S0^~ and was  strongly inhibitory.   These
     preliminary results demonstrate that  the requirement for
     S0|" in iron oxidation is quite specific.

Isolation and properties of the inorganic  pyrophosphatase enzyme

                   Materials and Methods

     Organism

     Thiobacillus ferrooxidans strain TM,  was grown  on the 9K
     medium as described by Silverman and  Lundgren  (16).  Cells
     were grown in 16- liter carboys under  forced aeration and
     harvested after 48-60  h using a Sharpies centrifuge.  The
     cells were suspended in 0.1 M 3-alanine sulfate buffer
     (pH 3.6) and centrifuged at 160 x g for 10 min  to remove
     iron;  the supernatant containing cells was centrifuged
     at 13,000 x g for 10 min.  The cells  were  washed  twice
     with the same buffer and stored as a  10% (w/v)  suspension
     until used.

     Preparation of Cell-Free Extracts

     Prior to breakage, washed whole cells were treated  for
     8-12 h with 0.5 M Tris-HCl buffer (pH 8.5) and  then
     centrifuged at 13,000  x g for 10 min.  The cell pellet
     was resuspended in Tris-HCl buffer and the cells  were
     disrupted by treatment in a 10 kcycles/s Raytheon sonic
     oscillator for 20 min.  Unbroken cells and debris were
     removed by cent rifugat ion at 13,000 x g for 15  min  and
     the supernatant was saved.

     Enzyme Assays

     The inorganic pyrophosphatase assay was that of Akagi
     and Campbell (50).  The reaction mixture contained  50 ymoles
                             69
-------
                      Table 13
Effect of various  anions  on iron oxidation in a system
     containing only Cl   and in the  same system
                                    2-
              containing  Cl  plus SO-
*a
Anion added


HAsO2'
4
HPO2'
4
B°~3
NO-
Mo°4~
SO2'
so4
AOD410/min
No SO2' 2.5
4
.005
.015
.000
.000
.000
.030

x 10~3M

.033
.035
.031
.016
.000
.035b

so2-
4






 a                                                   2
   The assay system used to test  anions  other than  SO ~
                                   -2
   contained the  following:   5  x  10 M g-alanine,
   1.9 x 10~2M Cl", and 4.5 x 10~2M Fe2+.  The various
   test anions were added  as  the  Na or K salt and were
   tested at 2.5  x 10  M both in  the presence and absence
   of 2.5 x 10~2M SO2".
 b                                      -2     2-
   This assay contained a  total of  5 x 10  M  SO
-------
of Tris-HCl buffer (pH 8.0), 5 ymoles of MgCl ,  5 ymoles
of Na2P.07, enzyme, and H-0 to a final volume of 0.7 ml.
All components except Na-P 0_ were incubated at  the
temperature of the assay (30 C) for 5 min before the addi-
tion of substrate.  After 10 min. the reaction was stopped
by the addition of 1 ml of 5 N H SO  and immediate chilling
of the reaction tubes.  An appropriate sample was removed
for inorganic phosphate determination, using a modified
method (51).  Unless otherwise indicated, results are
expressed as units per milligram protein.  A unit is 1
pinole of inorganic phosphate liberated per minute.

Protein was measured by the method of Lowry et al. (17)
with crystalline bovine serum albumin as the standard.

Alkaline phosphatase was assayed by the method of Garen
and Levinthal (52).

All results reported represent averages from at  least
three separate assays and results were reproducible using
different batches of cells.

Puri fi cation Procedure

Cell-free extracts were centrifuged at 105,000 x g for 1
h in a type 65 rotor Beckman ultracentrifuge, model L2-65B.
All centrifugations reported were run at 0-4° C.  To the
supernatant, an equal volume of 2% streptomycin  solution
was slowly added and the mixture stirred for 30  min in
the cold.  After standing for 15 min, the mixture was
centrifuged at 13,000 x g for 10 min.  For every milli-
liter of supernatant  (usually 5 ml) 0.01 ml of 1 M MgCl2
solution was added and the solution placed in a boiling
water bath for 5 min.  Denatured protein was removed by
two centrifugations.  Solid (NH )-SO, was added slowly,
with stirring, to the supernatant to 60% saturation, and
the suspension stirred for 30 min in the cold and allowed
to stand for an additional 15 min before centrifugation.
The supernatant was treated with (NH ) SO  to 90%
saturation and again stirred for 30 min in the cold,
allowed to stand for  15 min, and centrifuged.  The
precipitate was dissolved in 0.01 M Tris-HCl buffer
(pH 8.0) and dialyzed overnight in the cold against the
same buffer.  An equal volume  (45 ml) of DEAE-cellulose,
equilibrated with the same buffer, was added to the
dialyzed fraction and the suspensions stirred for 1 h
                       71
-------
in the cold.  This volume was arbitrarily selected based
upon purification experience with other enzymes (2,3).
After centrifugation at 13,000 x g for 10 min, 0.1 M
Tris-HCl buffer (pH 8.0) was added to the DEAE-cellulose
pellet.  This procedure was repeated using increasing
buffer concentrations following each centrifugation.
Concentrations of 0.2 M, 0.3 M, and 0,4 M were each used
once, and 0.5 M Tris-HCl buffer was used twice.  The
supernatants from the two 0.5 M eluates were pooled and
filtered to remove any residual DEAE-cellulose.  This
fraction was then dialyzed for 2-4 h against 40% (w/v)
polyethylene glycol (Carbowax 6000) for the purpose of
concentrating the enzyme.

Spheroplast Formation

Five grams of cells were washed with 0.1 M Tris-HCl
buffer (pH 7.9).  The cell pellet was resuspended in 50
ml of a solution containing 2.0% lipase and 0.025 M
CaCl2 dissolved in 0.1 M Tris-HCl buffer (pH 7.9) which
was previously centrifuged at 600 x g for*10 min to
remove solids.  After stirring at room temperature for
2.5 h, the cells were centrifuged and the pellet resus-
pended in 50 ml of 0.1 M Tris-HCl buffer (pH 7.9) con-
taining 1.0% ethylenediamine tetraacetate (sodium salt),
1.0% lysozyme, and 10.0% sucrose.  The solution was
stirred for 5 h at room temperature and then stored at
0-4°C overnight.  After storage the cells were centri-
fuged and the pellet was suspended in 10.0% sucrose in
0.1 M Tris-HCl buffer (pH 7.9).  For every 2 ml of the
above sucrose solution, 1 ml of H_0, 5 ml of 0.05 M
MgSO  in 0.15 M NH Cl-NH OH buffer (pH 9.1), and 2 ml
of 0715 M NH4C1-MTOH buffer (pH 9.1) were added.
Spheroplasts formed in a majority of the cells in several
Osmotic shocking of whole cells was done according to the
method of Neu and Chou (53).

                    Results

Inorganic Pyrophosphatase Activity

Cell-free extracts possessed more inorganic pyrophosphatase
activity than did intact cells which exhibited about 6%
of the activity of the extracts. Activity is expressed
on the basis of milligrams of protein, and it is assumed
-------
that no preferential release of enzyme occurred during
cell breakage.  Sph crop las ting of iron-grown cells
released about 11% of the enzyme activity whereas
osmotic shocking of whole cells released about 7%  of
the activity  (Table 14).

Properties of the Purified Enzyme

The purification as outlined in Table IS shows about a
21- fold purification of the enzyme.

The activity of the purified enzyme, as determined by
the assay conditions, was linear with increasing protein
concentrations (0.5-6.0 yg) .

The enzyme had a broad pH optimum extending from pH 7.5
to 8.5. Sodium acetate buffer  (50 ymoles) was used from
pH 5.5 to 6.0 and Tris-HCl buffer (50 ymoles) from pH
7.2 to 9.0.  No other buffer systems were studied  for
their effect upon the enzyme.

The enzyme was inactive when Mg   was absent, and  other
cations were poor substitutes;  these were Mn^"1", Co  ,
and Zn   (Table 16).  The two  anions (S04~2 and Cl~)
tested were without effect on the enzyme;  MgSO. was as
effective as
Heat inactivation studies were done for the enzyme, which
was heated at the different temperatures for 10 min.
In the presence of Mg2+, the enzyme was more stable to
heat, and under the conditions of the assay approximately
8% of the activity remained after heating for 10 min at
100°C.  In fact, results of other thermal inactivation
experiments indicate that the enzyme retains 8% residual
activity after heating for 1 h at 100°C.  No explanation
is known for this heat resistance.  The possibility of
the presence of more than one enzyme cannot be ruled out.
Also, since the purification procedure involved a heat
treatment it is not known if a heat- labile enzyme was
removed.  No temperature studies were made on crude cell
free extracts.  Blumenthal et al. (54) did report a heat-
stable enzyme in cell- free extracts of gram-negative
bacteria.

The inorganic pyrophosphatase was unable to hydrolyze
adenosine triphosphate  (ATP), adenosine diphosphate (ADP) ,
adenosine monophosphate  (AMP), glucose 1 -phosphate,
                          73
-------
                          Table 14

       Inorganic pyrophosphatase activity of different

                  cell  preparations
       n                                Percent total
       Preparation                      enzyme activity
1.   Cell activity

       Cell extracts                         100
       Intact cells                             6

2.   Spheroplasts
       Supernatant collected during           11
          spheroplasting
       Spheroplast lysate*                    89

3.   Osmotic Shocking

       Cold water wash                         7
       Lysate of cells following shocking     93
* Supernatant after centrifugation of spheroplasts ruptured
  by sonic disintegration.
-------
                                         Table  15



                        Purification of inorganic pyrophosphatase

I.
II.
III.
IV.
V.
VI.
Fraction
Cell- free extract
105,000 x g supernatant
Streptomycin
Heat (100°C, 5 min)
Ammonium sulfate
DEAE-cellulose
Protein
Cmg/ml)
3.80
2.00
1.10
0.31
0.12
0.04
Activity
(units */mg
protein)
3.99
7.25
6.66
20.06
24.08
80.93
Yield
100
88
85
60
18
10
Purification
(fold)
1.0
1.9
1.7
5.2
6.2
21.0
* A unit is  defined at  1  ymole phosphate  liberated per minute,
-------
                        Table 16
    Effect of cations on inorganic pyrophosphatase.


   Cation                              Percentage Activity

None                                           0
Mg2+                                         100
Mn2+                                           7
Co2+                                           8
Zn2+                                          11*
Ca2+                                        *  0*
Ba2+                                           0*
Fe2+                                           0*
Cu2+                                           0

* Incipient precipitation appeared during the assay.
  Note:  The reaction mixture contained 5 ymoles of each
  cation and 2.2 yg of protein.   Mg  , Ca  , Ba  , and
  Fe   were supplied as chlorides;  Mn  , Zn  , Co  , and
  Cu   were supplied as sulfates.
-------
glucose 6-phosphate, and fructose 6-phosphate.   These
substrates were tested at 5 ymoles each.   The enzyme  did
not act upon p-nitrophenyl phosphate (3 ymoles), a
substrate for alkaline phosphatase;  bis-p-nitrophenyl
phosphate, the substrate for phosphodiasterase,  was not
tested.

Both ethylenediamine tetraacetate fEDTA)  and potassium
fluoride at a concentration of ICT^M strongly inhibited
(90%) inorganic pvrophosphatase.  Guanidine-HCl and
sodium azide (10  M) inhibited the enzyme about 25%.
Two sulfhydryl-binding agents, p-chloromercuribenzoate
and iodoacetamide, had no effect on the enzyme nor did
increasing concentrations of phosphate (up to 10  M),
the end product of the reaction  (Table 17).

Effect of Pyrophosphate and Magnesium Concentrations  on
En zyme Ac t i vi ty.

The rather specific requirement of the enzyme for Mg
prompted further study of this reaction.  Figure 17
shows that except at very high Mg2+ concentrations,
maximum enzyme activity occurred when the Mg2+ to pyro-
phosphate (PPj) ratio was one.   In the region of excess
PP., inhibition was observed;  excess Mg2+ was also
inhibitory.

The apparent K  of the enzyme with respect to either Mg
or PP. was calculated as 1.6 x 10~3M from a Lineweaver-
Burk 1[14) plot.  To obtain this plot the initial reaction
velocities for low substrate concentrations of PPi and a
Mg  /PP. ratio of one were used.

                     Discussion

An understanding of structure-function relationships in
the cell envelope and how they relate to chemolithotrophy
was the purpose of the first part of this section.  The
lipopolysaccharide outer layer of iron grown T. ferrooxidans
isolated by hot phenol-water extraction was similar to
other gram negative LPS molecules as observed in the
negatively stained preparations.  The isolated  LPS were
ribbon-like as well as forming 'distinct sheets  of LPS
containing many visible holes.   Preliminary chemical
analysis indicate that the basic sugar content  of the
LPS was similar to E_. coli and Salmonella  (49)  LPS.
The isolated LPS molecule had a  molecular weight in excess
of a million.  The LPS also contained a rather  large amount
                         77
-------
of ferric iron.  The presence of the ferric iron may
substantiate the role of the outer layer of cell in
ferrous iron oxidation.

The whole cell kinetic assay technique described in the
second part of this section allows one to study the rates
of iron oxidation by iron-oxidizing bacteria.   The
advantages of the assay procedure are its simplicity
and rapidity, as compared to the conventional  Warburg
respirometer.
                                          2_
Using the new method, it was found that SO^ was a
requirement for iron oxidation.   This had been previously
reported (55).  This requirement may relate to the fact
that SO?" is produced by the cell during sulfur oxidation.
It appears, from these data, that the S0|~ is  involved
in the binding of ferrous iron to the cell for subsequent
oxidation.  The sulfate anion were partially replaced by
HPO|~ and to a lesser extent by HAsO$~.  Formate and
molybdate inhibited the iron oxidation reaction, under
the new assay conditions.

The inorganic pyrophosphatase studied was found to be
located in the cells cytoplasm.   The partially purified
enzyme required Mg^* f.or activity and could not be replaced
by Mn2+, Zn  , or Co  .   It is generally believed that the
Mg2+ forms a complex with the pyrophosphate molecule
(MgPPi2~) and it is this complex on which the  enzyme acts.
The K  for the enzyme (1.6  mM) is in general agreement
with inorganic pyrophosphatases  isolated from other
mi crobes.

One aspect of this enzyme which appears repeatedly in the
isolation and characterization of other T_. ferrooxidans
enzymes, is the alkaline pH optimum.  All the  enzymes
isolated to date from J_. ferrooxidans (2,3) have alkaline
pH optima, while the pH of the aqueous environment in
which the cells exist is so low (pH 2.5-3.5).   This
observation further emphasizes the role of the outer
cell envelope in maintaining a pH balance between the
acid environment and the cells cytoplasm.
-------
                          SECTION VII

       STUDIES CONCERNING THE METABOLISM OF THIQBACILLUS

        FERROOXIDANS GROWN ON HETEROTROPHIC SUBSTRATES
Introduction

The acidophilic iron-oxidizing bacteria include the group of
organisms previously referred to as Ferrobacillus
                                                             ^
but now recognized as Thiobacillus ferrooxidans (56, 28).
A chemolithotroph, T_. ferrooxidans obtains energy for the
reduction of carbon dioxide from the oxidation of reduced
iron and sulfur compounds.   Recent studies, however, have shown
that not only this organism, but other autotrophs, can use
organic compounds such as sugars, amino acids and various
metabolic intermediates for energy (57).  Lundgren, et al
(58) have shown that when autotrophically-grown T. ferrooxidans
is transferred to a medium containing iron plus glucose, cells
preferentially utilize ferrous iron and when i»ron is exhausted,
the cells oxidize glucose as an energy source upon transfer
into a glucose-salts medium.

Some strains of iron-oxidizing bacteria are unable to adapt
to glucose, and it has been suggested that this trait may be
considered as a basis for species differentiation  (57).
Interestingly, glucose has been shown to inhibit iron and
elemental sulfur oxidation in manometric experiments using
resting cells prepared from cultures grown on either inorganic
substrate  (59) .   Concomitant with the inhibition of substrate
oxidation, the autotrophic fixation of carbon dioxide is
drastically reduced in the presence of high concentrations of
glucose.  It has recently been found that glucose inhibits
CO- fixation from 40-66 percent when sulfide, thiosulfate,
tetrathionate, dithionate or sulfite served as the source of
reduced sulfur (60).  Other studies (J. H. Tuttle and P. R.
Dugan, Bacteriol. Proc., p. 64, 1969) have shown several
organic acids to inhibit both iron and sulfur oxidation.

This investigation was undertaken to study the effects of
organic metabolites, particularly glucose, on the growth and
metabolism of T_. ferrooxidans in an attempt to understand
metabolic changes induced by the oxidation and subsequent
catabolism of this sugar.  Further, glucose-6-phosphate dehydro-
genase, a key enzyme of glucose dissimilation was studied to
ascertain its possible role in the control of heterotrophic
growth by autotrophic bacteria.
                               01
                               U i
-------
                     Materials and Methods

Organism and Growth Conditions

T_. ferrooxidans strain TM was grown in the 9K  medium as previously
described  (16).

Growth on  Iron-Glucose Media

When it was desired to grow T_. ferrooxidans  on the  9K medium,
supplemented with glucose  (9 KG medium),  the 9K salts and
ferrous sulfate solution  (at pH 2.5) were autoclaved separately.
Upon cooling, the two solutions were combined  and filter-
sterilized glucose was added at 0.5% final  concentration. The
final pH was 2.7.  10 ml  of a culture of iron-grown T_.  ferrooxi-
dans  (5 x  10^  cells) was  aseptically added to  the 9 KG medium
(100 ml total  volume) in  a 250-ml Erlenmeyer flask.  The  flask
was shaken on  a reciprocal shaker at 30  C for  60 h, at which
time  all the ferrous iron was oxidized.   The contents of  this
flask served as an inoculum for 500 ml  of 9  KG medium,  contained
in a  Fernbach  flask.  9K-fructose-grown,  9K-sucrose-grown, and
9K-glutamate-grown cells  were cultured  in a similar manner,
with  the exception that the appropriate  sugar  or amino acid
substrate  was  substituted for glucose.   Cells  were  harvested
as described for  iron-grown cells.

Growth on  Glucose
T_.  ferrooxidans was  readily cultured in a glucose 9K salts
media.   This  was  done  by transferring 10 ml of a 9 KG culture
into  90  ml  of a glucose-salts  medium free from substrate  amounts
of  FeSO..   The glucose was  filter-sterilized at high concentra-
tions  (10%) and added  to sterile  9K-salts solution (pH 2.5)
at  a  final  concentration of 0.5%.   Cells have been continuously
transferred,  and  maintained on the  glucose-salts medium for
over  a year,  with bimonthly transfers  to fresh medium. Large
masses of glucose-grown  cells  were  used for preparing crude
enzyme extracts;  cells  were grown  using a 10% inoculum from
a fresh  starter culture  and inoculated into 500 ml of the
glucose-salts medium contained in 2800-ml Fernbach flasks.
Flasks were shaken at  30 C  on  a 3-tiered New Brunswick rotary
shaker set  at 150 cycles/min.

Cells were  grown  on  several  organic media,  merely by replacing
the appropriate organic  substrate for  glucose.
                               32
-------
In all cases where media contained an organic energy source,
contamination was vigorously monitored by streaking cultures
on a battery of complex media, including nutrient agar,
Saboraud's agar, thioglycollate and nutrient broth, adjusted
to different pH's.  Plates (or tubes) were incubated at various
temperatures for at least ten days.  In no case, was contamina-
tion observed.  Microscopic examination of the cultures was
also used to monitor purity.

Growth Rates

Growth rates for cultures grown on sugars or other organic
substrates were determined using 250-ml Nephalo flasks fitted
with side arras  (Bellco Glass Inc., Vineland, New Jersey).
The flasks contained 20 ml of medium and were shaken at
150 cycles/min at 30 C;  turbidity was followed spectrophoto-
metrically at 550 nm in a Bausch and Lomb Spectronic 20 colori-
meter and absorbance plotted as a function of culture time.
The generation time was calculated from the exponential portion
of the growth curve and is defined as the time required for
absorbance at 550 nm to double.  The specific growth rate
constant,  k, was determined from the expression
where t_ - t  equals the time  interval  required for cell doubling
(61).

Colorimetric Determination  of  Iron

Iron was determined by  the  procedure  described by Suzuki and
Silver  (20) using  2,2'-dipyridyl.

Whole Cell Iron Oxidation Assay

The procedure for  the  colorimetric  assay  used for kinetic
studies of iron oxidation by whole  cells  as  described earlier
in this report was used to  measure  iron oxidation.

Glucose Assay

Glucose was assayed using the  Glucostat enzyme reagent  (Worthing-
ton Biochemical Corp.,  Freehold,  N. J.).   All dilutions of
standards and samples were  made in  10 mM  potassium phosphate,
pH 7.0.  9.0 ml of the  Glucostat  reagent  were placed in colori-
meter tubes to which was  added 1  ml of the sample.  After exactly
                              83
-------
10 min at room temperature, one drop of 4 N HC1 was added to
stop the reaction and stabilize the color.  The tubes were
allowed to stand for 5 min and then read at 400 nm.

Radiorespirometric Studies

The radiorespirometric experiments were performed as described
by Wang et al (62), as modified by Perry and Evans  (63).  The
experiments were done in the laboratory of Dr. J. J. Perry,
Department of Microbiology, North Carolina State University,
Raleigh, N. C.

Preparation of Cell-free Extracts by Sonic Oscillation

Harvested  cells of T_. ferrooxidans were immediately suspended
in the  appropriate buffer  at 4 C, washed three times and
suspended  in  the same buffer overnight at 4 C.  This eliminated
the necessity of the treatment of the cells with ion-exchange
resins  (2) to facilitate rupture of T_. ferrooxidans.  The
cells were centrifuged at "35,000 x g_ for  10 min, suspended as
a 20%  (u/v) suspension in  buffer and disrupted in  a water-cooled
10 kc/sec  Raytheon sonic oscillator for 15 min.  Residual whole
cells  and  debris were removed by centrifugalion at  15,000 x £
for 20  min and the resulting supernatant was used  as the crude
cell-free  extract.   Preparation of crude  cell-free extracts for
ribulose diphosphate  (RuDP) carboxylase assay was  carried out
in this manner;  the buffer, at pH 7.9, contained  50 mM
Tris-HCl,  12  mM Tris-HCl,  12 mM 2-mercaptoethanal, 50 mM
NaHCO   and 15 mM EDTA.  Extracts to be used for assay of
glucose catabolic  enzymes  were prepared from cells suspended
in 50  mM Tris-HCl  (pH 7.9).  Tricarboxylic  (TCA) cycle enzymes
were  assayed  using cell-free extracts cells previously harvested
as described  above and immediately suspended in a potassium
phosphate  (50 mM)-glutathione  (1 mM) buffer at pH  7.4.  Cell
pellets were  twice washed  with this buffer.  The presence of
glutathione  (GSH) or dithiothreitol (1 mM) in the buffer was
essential  for maximum enzymatic activity.  The cells were sus-
pended to  a 20% w/v  suspension in the phosphate-GSH buffer,
placed in  a standard Raytheon cup and N, gas bubbled through
the cell suspension  for 10 min.  The cells were disrupted by
sonic  energy  for 15 min.   After the removal of whole cells and
debris, the supernatant was separated and used for  assay of
the TCA cycle enzymes, as  well as for NADH oxidase  assays.

Cell extracts were also prepared from cells harvested at
different  times during growth in the 9 KG medium,  in this
instance 2 ml of a 20% (w/v) cell suspension were  treated
using  a Branson sonifier,  equipped with a mini-probe
-------
(Branson Instruments, Inc., Plainview, New York).   The cell
suspension was chilled in an ice bath and intermittent bursts
of energy were used, amounting to a total of 5 min.

Protein Determination

Protein was determined coloimetrically by the method of Lowry
et^ si^ (17) using crystalline bovine serum albumin as the
standard.  The spectrophotometric method of Warburg and
Christian, as described by Layne (64) was also used for
estimating protein.

Assay for RuDP Carboxylase

Ribulose 1,5-diphosphate carboxylase  (3-phospho-D-glycerate
carboxylase (dimerizing), EC 4.1.1.39 was determined in a
2-step coupled assay essentially as described by Stukus and
DiCicco  (1970). Step 1 results in the formation of 3-phospho-
glyceric acid.  The reaction mixture, in a total volume of
2.0 ml contained:  Tris-HCl (pH 7.9), 50 mM, Mg§0  , 6 mM;
GSH, 6 mM; NaHCO,,  40 mM;  RuDP, 1 mM, extract and water to
2.0 ml.  The reaction was initiated at 37 C with extract and
run for 5 min.  The reaction was stopped by placing reaction
tubes in a boiling water bath for 3 min.  After cooling in
an ice bath, denatured protein was removed by centrifugation
and the supernatant drawn off and used.   Stet> 2:   The 3-
phosphoglyeerie acid formed in step 1 as a result of RuDP
carboxylase activity was determined in a coupled assay con-
taining 3-phosphoglyceraldehyde dehydrogenase  (230 yg) and
3-phosphoglyeerie acid kinase (2.2 yg).  The reaction mixture
also contained:  Tris-HCl  (pH 7.9),   50 mM;  ATP,  0.67 mM;
MgS04, 6.67 mM;  L-cysteine (pH 7.0), 6.67 mM;  NADH  (pH 7.0),
0.13 mM;  one ml of  step 1 supernatant and water to 3.0 ml.
A unit of activity is defined as the  number of ymoles of
3-phosphogly eerie acid formed in 1 min, as measured by
following the oxidation of NADH at 340 nm.  Activity  is
expressed as units/mg of protein calculated from the  linear
portion of a curve plotting enzyme activity vs protein con-
centration.  A Coleman #124 recording spectrophotometer,
equipped with absorbance expansion from 0 to  0.1 absorbance,
was used for the assay.

Assay for Glucose-6-Phosphate Dehydrogenase

Glucose-6-phosphate  dehydrogenase  (D-glucose-6-phosphate;
NADP+ oxidoreductase, EC  1.1.1.49) was assayed spectrophoto-
metrically at 340 nm in  3  ml cuvettes with a  1 cm  light path,
                               03
-------
by following the rate of appearance of either NADPH or NADH.
The reaction mixture contained:  Tris-HCl (pH 7.9), 15 mM;
MgCl_, 6.67 mM;  glucose-6-phosphate, 0.83 mM;  NADP"1", 0.33 mM
or NAD"1" (pH 7.0), 3.33 mM;  cell-free extract, suitably
diluted in 50 mM Tris (pH 7.9) and water to 3.0 ml.  The
reaction, in the case of crude extracts, was initiated with
either NADP* or NAD*.  Enzyme was used to initiate the reaction
during kinetic experiments.  When NADH was included in the
cuvette, the reduction of NADP+ was measured at 366 nm.  A
Coleman 124 recording spectrophotometer was used routinely
in all assays.  For kinetic studies, a Gilford Nodel 240
spectrophotometer coupled to a Leeds and Northrup Speedomax
G recorder and equipped with a temperature controlled sample
holder, was used at 25 C.

Assay for 6-Phosphogluconate Dehydrogenase

6-phosphoglyconate dehydrogenase  (6-phospho-D-glyconate:
NADP+ oxidoreductase  (decarboxylating), EC 1.1.1.44) was
assayed using the same assay as for glucose-6-phosphate
dehydrogenase except that 2.67 mM 6-phosphoglyconate was
used in place of glucose-6-phosphate.  The potassium salt
of 6-phosphoglyconate was prepared from the barium salt
according to the procedure of Horecker and Smyrniotis  (65).

Assay for Fructose-1, 6-diphosphate Aldolase

Fructose diphosphate aldolase  (fructose-1, 6-diphosphate
D-glyceraldehyde-3-phosphate lyase, EC 4. (1.2.1.3) was
measured spectrophotometrically using a coupled assay
containing triosephosphate isomerase and ct-glycerophosphate
dehydrogenase;  the rate of NADH  oxidation was followed at
340 nm, and the activity of the enzyme calculated from the
initial rate of NADH oxidation.   The reaction mixture con-
tained:  Tris-HCl (pH 7.5), 0.1 M;  fructose-1, 6-diphos-
phate, 5 mM;  0.02 ml a-glycerophosphate dehydrogenase/trio-
sephosphate isomerase (10 mg/ml);  NADH (pH 7.0), 0.25 mM;
cell extract and water to 3.0 ml.  The reaction was initiated
by adding cell extract.

Assay for the Entner-Doudoroff Enzymes

The Entner-Doudoroff enzymes, 6-phosphogluconate dehydrase
(6-phosphogluconate dehydratase, EC 4.2.1.12) and 2-keto-3-
deoxy-6-phosphogluconate (KDPG) aldolase (EC 4.1.2.14) were
assayed by determining the amount of pyruvate formed from
6-phosphoglyconate,  using the procedure of Keele, Hamilton
                               85
-------
and Elkan  (66).  i unit of enzyme activity results in the
formation of 1 ymole of pyruvate from 6-phosphogluconate.
Activity is expressed in units per mg of protein.

Assay for  Isocitrate Dehydrogenase

Isocitrate dehydrogenase (threo-D -isocitrate:  NAD  oxido-
reductase  (decarboxylating) EC 1.1.1.41) and  (threo-D -iso-
citrate:  NADP+ oxidoreductase (decarboxylating), EC 5.1.1.42)
was assayed spectrophotometrically at 340 nm following the
rate of appearance of NADH or NADPH.  The reaction mixture
contained Tris-HCl (pH+7.9), 10 mM;  MnCl2, 0.33 mM;  DL-iso-
citrate, 0.17 mM;  NAD+, 3.33 mM or NADP+, 0.33 mM;  extract
and water to 3\0 ml.

Assay for Aconitase

Aconitase  (citrate (isocitrate) hydro-lyase, EC 4.2.1.3)  was
assayed according to the procedure of Anfinsen (67), based
on the spectrophotometric determination at 240 nm of cis-
aconitic acid.

Assay for Succinic Dehydrogenase

Succinate dehydrogenase [succinate: (acceptor) oxidoreductase,
EC 1.3.99.1] was assayed spectrophotometrically at 600 nm
following the reduction of 2,6-dichlorophenol-indophenol
(DCPIP), mediated by phenazine methosulfate (PMS), according
to the procedure of Arrigoni and Singer (68).

Assay for ct-Ketoglutarate dehydrogenase was assayed spectro-
photometrically at 600 nm by following DCPIP reduction as
described by Watson and Dworkin (69).

Assay for Fumerase

Fumerase ( fumeratehydratase, EC 4.2.1.2) was assayed spectro-
photometrically at 240 nm according to the procedure of Racker
(70).

Assay for NADH, Oxidase

NADH oxidase  [NADH: (acceptor) oxidoreductase, EC 1.6.99] was
assayed spectrophotometrically at 340 nm by following the
oxidation of NADH.  The reaction mixture contained NADH,
0.25 mM;  phosphate buffer (pH 7.4), 66.7 mM;   cell-free
extract and water to 3.0 ml.  The reaction was initiated
with extract.
-------
For all assays controls were routinely performed and consisted
of the reaction mixture minus the specific substrate, or
minus the cell-free extract.  All enzymes were inactivated
by boiling for 5 min.  In all cases, except where noted,
enzyme was expressed as units per mg of protein.  One unit
is the amount of enzyme needed to oxidize or reduce one
ymole of substrate per min.  Assays were always repeated
three times and results were reproducible with different
batches of cell-free extract.

In kinetic experiments with glucose-6-phosphate dehydrogenase,
velocities are given as nmoles of NADPH or NADH formed per
min.  In Lineweaver-Burk plots, the reciprocal of velocity
is given as I/A OD   /min.

Purification of Glucose-6-Phosphate Dehydrogenase

All enzyme purification steps reported were conducted at 0-4 C.
All centrifugations were performed at 35,000 £_ in a Sorvall
RC2-B  centrifuge.  The sequential treatments were:

     Step 1.  The crude cell-free extract of T_. ferrooxidans,
     prepared as described, was frozen in a dry ice-ethanol
     bath of -20 C.  The extract was stored at this tempera-
     ture until needed.  At this time, the frozen extract was
     thawed and then centrifuged for 20 min;  the resulting
     precipitate was discarded.  The remaining supernatant
     was then centrifuged  at 105,000 x £ in a Beckman model
     L-2 preparative ultracentrifuge.  The pellet was again
     discarded, while the  clear yellow-orange supernatant was
     used for enzyme purification.

     Step 2.  To the 105,000 x £ supernatant, solid ammonium
     sulfate was added slowly with stirring over a period of
     15 min until the concentration reached 35% saturation.
     The solution was centrifuged for 15 min and the pellet
     discarded.  The resulting supernatant was brought to
     50% saturation with ammonium sulfate with constant
     stirring.  The mixture was centrifuged for 20 min.  The
     resulting precipitate contained nearly all the enzyme
     activity.

     Step 3.  The 50% ammonium sulfate precipitate was
     suspended in a minimum amount of 0.05 M Tris-HCl buffer,
     pH 7.9, and applied to a Sephadex G-200 column (2.5 x
     30 cm) equilibrated with 0.05 M Tris-HCl (pH 7.9) and
     eluted with the same buffer,  collecting 10 ml fractions.
     Enzyme of high specific activity was usually found in a
                               88
-------
single tube, and this material was used in all future puri-
fication steps.

Step 4.  The eluate from the Sephadex column was fractionated
with ammonium sulfate, first to 35% saturation and then
to 50% saturation as before.  The precipitate was collected
by centrifugation and resuspended in a minimum volume of
0.05 M Tris-HCl, pH 7.9.  This material was either chroma-
tographed on a small Sephadex G-200 column (1.5 x 20 cm)
in the case of enzyme from glucose-grown cells, or was
dialyzed against two 1-liter changes of 0.05 M Tris-HCl
buffer, pH 7.9, in the case of enzyme from iron-glucose
grown cells (9 KG cells).

Step 5.  The enzyme from the glucose-grown cells was
eluted off the small G-200 column by 0.05 M Tris-HCl,
pH 7.9 in 4 ml fractions.  Usually one of these fractions
contained the highest specific activity and this fraction
was used for all subsequent experiments.  Further puri-
fication was unsuccessful.

The dialyzed preparation from iron-glucose-grown cells
was further purified by negative Ca.,(P04)2 &el absorption:
Ca_(PO )_ gel, purchased commercially, was added at a
ratio of 3 mg of calcium phosphate to every 2 mg of
protein.  This mixture was stirred for 15 min slowly and
then centrifuged for 20 min.  The supernatant was then
treated again with additional calcium phosphate (3 mg
to every 2 mg of protein).  The resulting supernatant,
following centrifugation, was used in all cases as the
enzyme from iron-glucose-grown cells.

Disc Gel Electrophoresis

The standard disc gel electrophoretic procedure of Davis
(18) was used.  When gels were to be stained for protein,
gels were first fixed for 20 min in 20% trichloroacetic
acid  (TCA), stained for  40 min with Coomassie blue (1%
aqueous solution diluted 1:20 with 20% TCA), and destained
with 10% TCA.

Stains to detect enzyme  activity were performed in 6% gels
by placing gels in small tubes protected from the light.
The tubes contained the  assay components with the addition
of phenazine methosulfate  (PMS) and nitrobluetetrazoleum
(NET) at 0.056 mg/ml and 0.90 mg/ml respectively.  The
gels were incubated for  15 min at room temperature after
-------
which they were stored in 10% TCA.  Gels placed in tubes
lacking substrate were not stained.  The position of the
enzyme in the gel was determined by scanning stained gels
at 600 nm in a Gilford spectrophotometer, with a linear
transport attachment.  The relative mobility of the stained
enzyme (R ) was defined as the ratio of the migration (cm)
of the enzyme from the origin to the migration (cm) of the
dye front from the origin.
In all cases, enzyme was layered on the lower gel in 5% sucrose
and electrophoresed at pH 9.3, using the anionic system des-
cribed in the Buchler Instruments pamphlet "Instructions for
the Poly-analyst," Buchler Instruments, Inc., Fort Lee, N.J.
The current was maintained at 3 ma per gel.

Molecular Weight Determinations

Molecular weights were determined using gels of different
acrylamide  concentration, according to the method of Hedrick
and Smith  (19).  The gels were stained for activity and R^s
were plotted  against gel concentration;  the slope of this
line was determined using a  computer program written by Mr.
Amnon Liphshitz  (Dept. of Biology, Syracuse University) and
fitted to a standard curve of several proteins of known
molecular weight.

Chemicals and Reagents

The chemicals and reagents used in this study were mainly
procured from commercial sources, and were of the highest
purity available.  The sources of some of these are listed
below.

Sigma Chemical Company:  ADP, AMP, 3', 51 - AMP, ATP, DCPIP,
DTT, GSH, NAD+, NADH, NADP+, NADPH, NBT, PMS, Tris, glucose-6-
phosphate, 6-phosphogluconate, ribulose 1,5-diphosphate.

New England Nuclear Corporation:  glucose-1-  C. glucose-2-
14C, glucose-3-14C, glucose-3,4-14C, glucose-6-14C.

Packard Instrument Corporation:  PPO and POPOP.

Fisher Scientific Company:   2,2'-dipyridyl, ferrous sulfate,
D-glucose, D-fructose,  sucrose, potassium gluconate, sodium
glutamate, sodium acetate,  sodium succinate, sodium pyruvate.
-------
     Nutritional Biochemicals Corporation:   enzyme  grade
     ammonium sulfate, calcium phosphate gel.

     Canalco:  aery1amide, glycine,  ammonium persulfate
     N, N, N1, N'-tetramethylethylenediamine,  N,  N'methylene-
     bisacrylamide, bromphenol blue, Coomassie blue.

     All reagents and buffers were prepared in glass  distilled
     water.

                            Results

Effects of Organic Metabolites _on_ AutotrophicMechanisms

     Utilization of Glucose by Iron-grown Cells

     Iron-grown cells required at least two transfers in  a
     medium containing both iron and glucose (9 KG  medium)
     before glucose could be utilized as a sole source of
     energy.  Figure ISA and 18B show the relationship of
     iron and glucose utilization by iron-grown cells. Heavy
     inocula (5 g wet weight) were used in these experiments
     insuring that all the ferrous iron would  be rapidly
     oxidized and leaving glucose as the sole  energy  source.
     Ferrous iron was oxidized to ferric iron  by 2-3  hr,
     whereas the glucose level remained constant for  over
     62 hr.  When an inoculum of these cells was transferred
     to fresh 9 KG medium, cells quickly oxidized the iron
     and then, after a lag, oxidized the glucose (Fig- 18B).
     This lag, prior to glucose oxidation, had been noted
     previously (58), and was reproducible in  the present
     author's hands.  The increase in the level of  glucose-6-
     phosphate dehydrogenase (G6PD), a key enzyme for the
     dissimilation of glucose, is also shown.   The  utilization
     of glucose by the cells resulted in the accumulation of
     poly-B-hydroxybutyric acid (PHB).

     Iron Oxidation by Resting Cells Cultured  on Iron and
     Organi c Supplements.

     J_. ferrooxidans was cultured on ferrous iron plus an
     organic supplement, and the substrate's effect on the
     iron oxidizing capacity of thiobacilli tested  using
     washed resting cells.  Resting cells were prepared  from
     cultures harvested after exhaustion of ferrous iron.
     The rate of iron oxidation by cells previously grown in
     the various supplemented media varied according  to the
     organic substrate present.  Cells grown in the iron
-------
                                            100  no
                                           120  130
                         CULTURE  AGE  (hr)
Figure l<
Utilization of glucose by iron-grown T.  ferrooxidans.
5 g of iron-grown cells (wet  weight), suspended in
the salts of the 9K medium, were added to 500  ml of_
the iron-glucose supplemented medium (pH 2.7)  contained
in a 2-liter Erlenmeyer flask.  The flask was  incubated
at 30 C and shaken at 150 cycles/min.  At zero time and
intervals thereafter, samples were removed for ferric
iron, glucose, and enzyme assays.  After 62.5  hr of
incubation, 50 ml of medium containing about 0,5 g
of cells were placed in a flask containing fresh
medium and the shaking continued.  Samples were
removed  aseptically and analyzed as before  Fig.    ).
Glucose-6-phosphate dehydrogenase  (G6PD) activity
 (units/mg of protein)  is expressed relative to the
resting  cell inoculum.
-------
glucose medium lost 80% of their iron oxidizing ability
(Table 18), whereas iron-fructose-grown cells oxidized
iron at a rate nearly comparable to cells grown solely
on ferrous iron.  These results demonstrate that once
organic carbon is oxidized by the cells, the autotrophic
energy generating system is affected, however, the degree
depends upon what particular organic metabolite is supplied.

Ribulose Diphosphate Carboxylase Levels

The levels of ribulose diphosphate (RuDP) carboxylase in
cell free extracts prepared from cells grown in the
various supplemented media are shown in Table 19.  Extracts
from iron-fructose and iron-glutamate grown cells possess
about 70% of the carboxylase activity of autotrophically
grown cells.  However, both sucrose and glucose almost
completely repress the synthesis of this enzyme.

Relationship Between Iron Oxidation and RuDP Carboxylase
Activity

The relationship between iron oxidation, as expressed in
resting cells previously grown in a 9 KG medium, and
carboxylase activity in extracts of these cells is
illustrated in Figure 19.  The greatest amount of carboxy-
lase activity (3.5 h) is noted just after completion of
iron oxidation, thereafter enzyme activity diminishes to
the level of the cell inoculum (t ).  Washed resting
cells prepared from cells previously grown in the 9 KG
medium for 1-6 h and then assayed for iron oxidation and
enzyme activity exhibited the highest iron oxidation rate
at 1 hr;  at this time Fe3+ ion is actively accumulating
in the growth medium.  Maximum iron oxidation rates pre-
cede peak RuDP carboxylase activity.  These results
indicate that the two processes, iron oxidation and CCL
fixation, so necessary for autotrophic existence, are
related and that maximum iron oxidation must precede the
maximum expression of carboxylase.  The presence of
glucose in the medium does not affect the RuDP carboxylase
activity or iron oxidation rates, for at the early hours
of growth only ferrous iron is oxidized.

Heterotrophic Growth of T. ferrooxidans

T_. ferrooxidans was  cultured  on  glucose  as the sole
energy source.  Glucose disappeared from the medium by
50 h and the cell doubling time was 4.5 h.  Glucose-
grown cells examined under the phase microscope were
-------
                                       Table 18

        Effect of organic supplement(s) on iron oxidation by resting cells of
                                                                          3.
          T_. f errooxidans previously grown in the iron-supplemented medium—
Growth
substrate (s)
9K (iron)
9 K- fructose
9 K- sucrose
9K-glutamate
9K-glucose
AOD41Q/min
U02£
2.80
1.60
0.95
0.80
0.56
Cell dry
weight (mg)
0.40
0.30
0.38
0.38
0.36
AOD/min/mg
(io2)
7.00
5.41
2.53
2.13
1.57
% of control
100.0
77.3
36.1
30.4
22.4
A_
  In each case, cells were harvested after 60 hr of growth, washed in g-alanine-SO.
  buffer, pH 3.6, and suspended in this buffer.

b_
  Iron oxidation was determined using the kinetic assay as described by Schnaitman
  et al (71).
-------
10
en
                                               Table 19



              Ribulose 1,5-diphosphate carboxylase levels in extracts of T_. ferrooxidans



                               grown in the iron-supplemented media—
Growth
Iron
Iron +
Iron +
Iron +
Iron +
Iron +
substrate-

glucose (60 hr)
glucose
sucrose
fructose
glutamate
Units/mg (10~3)
9.88
1.55
0.54
0.51
6.62
7.17
Autotrophic level
(%)
100.0
15.7
5.5
5.2
67.0
72.6
           All cells were harvested after 66 hr of growth except  for one instance  (60  hr),

           labeled above.


         b

           Organic supplements  were added to a final  concentration  of 0.5%.
-------
     0   I    23456789   10  II

                  CULTURE  AGE (hr)


Figure 19.   Relationship of  the rate of iron oxidation by
            resting cells to RuDP carboxylase activity.  Cells
            were grown in 500 ml of the 9KG medium (pH 2.5)
            using 5 g of iron-grown cells  as an inoculum.
            Fe^+ was  determined in  culture supernatants  at
            the times indicated. For iron oxidation and RuDP
            carboxylase assays, 9 KG-grown cells were harvested
            and washed as previously described.  Iron oxidation
            by resting cells was assayed spectrophotometrically
            at 410 nm.   RuDP carboxylase  (units/mg of protein)
            is expressed relative to the level of the enzyme
            found in  the cell inoculum (zero time).
                               96
-------
     found in pairs and contained large amounts of the
     refractile storage product PHB.  Glucose-grown cells
     were readily transferred to defined media containing
     other sugars or organic substrates in place of glucose
     (57).  Growth rates of the organism grown on five
     different types of  substrates are compared in Table 20.
     Fructose appeared to be the preferred substrate with a
     generation time of 3.4 hr;  all other substrates tested
     gave generation times greater than glucose-grown cells.
     Cells grown heterotrophically on any of the aforementioned
     substrates gradually regained their potential for auto-
     trophic growth when transferred back to the 9K medium.
     Glucose-grown cells showed no detectable iron oxidation
     ability;  cells grown on the other substrates were not
     tested.

     No absolute requirement for vitamins or growth factors
     was noted for cells grown in the organic media. Glycerol,
     acetate, pyruvate, and succinate (at a concentration of
     0.5%) did not support growth.  No other organic substrates
     were tested.

Het e rotrophi c Met abolism of T.  ferrooxi dans

As shown previously, iron-grown cells, once adapted to glucose,
possessed glucose-6-phosphate dehydrogenase, the enzyme which
catalyzes the initial reaction in both the pentose-phosphate
shunt and the Entner-Doudoroff pathways.   These results
suggested that in order to metabolize glucose, drastic changes
in intermediary metabolism must take place..  Studies were
therefore initiated to determine the enzymatic basis for the
change from an autotrophic existence to a heterotrophic one.
Radiorespirometric experiments, using ^C-specifically labeled
glucose, were also performed to help determine the pathways)
through which glucose is metabolized.

     Enzymes of Glucose Catabolism

     Enzymes functioning in the Embden-Meyerhof, Entner-Doudo-
     roff and pentose-phosphate pathways were assayed using
     crude cell-free extracts of T_. ferrooxidans grown auto-
     trophically, mixotrophically (iron + glucose), and hetero-
     trophically.  The results (Table 21) show that during
     autotrophic growth  (iron-grown cells), the levels of the
     glucose catabolic enzymes were low, whereas, as previously
     shown, the capacity for iron oxidation and CCL fixation
     was high.  This low level of the glucose enzymes in
     autotrophically-grown cells indicates that these enzymes
-------
                         Table 20




       Growth of  J_. ferrooxidans on organic substrates
Growth substrate    Generation time    Growth rate  constant



   (0.5%)                (hr)
                   k (hr'1)
Glucose




Sucrose




Fructose




Glutamate




Gluconate
4.5




6.6




3.4




7.4




7.0
.154




.105




.204




.094




.099
-------
                                       Table  21
                                                                                       o
Effect of growth substrate on enzymes  involved in  glucose metabolism by T.  ferrooxidans—
Enzyme
Glucose-6-phosphate +
dehydrogenase (NADP )
Glucose-6-phosphate +
dehydrogenase (NAD )
6-phosphogluconate
dehydrogenase (NADP )
6-phosphogluconate
dehydrogenase . (NAD )
6-phosphogluconate dehydrase
and 2-keto-3-deoxy-6-
phosphogluconate aldolase—
Fructose- 1, 6 -di phosphate
aldolase

Iron -grown
1.29
1.55
3.70
1.23
0.40
1.29
3
Activity (units/mg x 10 )
cells Iron-glucose-
grown cells
52.70
62.00
H.D.t
1.02
12.4
0.47

Glucose -grown
cells
76.67
90.20
N.D.
1.59
15.9
0.88
   In each case,  cells  were  taken  from the  late  log phase of growth.

 — N.D. = not detected under the conditions used.

 — 1 unit of activity results in the  formation of  1 nmole per min. of pyruvate from
      6-phosphogluconate.
-------
function in biosynthesis, and do not appear to be directly
involved in energy metabolism (72).  However, when glucose
serves as an energy source, as with iron-supplemented
medium or in the glucose-salts medium, glucose-6-phosphate
dehydrogenase (both NADP -and NAD -linked) was induced
about 50 times above the autotrophic level.  The presence
of an inhibitor to glucose-6-phosphate dehydrogenase in
crude cell extracts was ruled out, for adding this extract
to cell-free extracts of .E_. coll had no effect;  EL coli
K12 is known to have a high level of glucose-6-phosphate
dehydrogenase.  Nor was inhibition observed when extracts
from iron-grown cells were added to partially purified
glucose-6-phosphate dehydrogenase obtained |rom glucose-
grown T_. ferrooxidans.  Either NADP  or NAD  served as
coenzyme for the glucose-6-phosphate dehydrogenase from
T_. ferrooxidans.  The activity of 6-phosphogluconate
dehydrogenase was low or undetectable in all extracts
tested.  However, the enzymes of the Entner-Doudoroff
pathway  (6-phosphcgluconate dehydrase and 2-keto-3-
deoxy-6-phosphogluconate aldolase) in cell extracts from
iron-supplemented and glucose-grown cells showed a 30-40
fold stimulation over extracts prepared from iron-grown
cells.

The assay for the enzymes of the Entner-Doudoroff pathway
was based on the conversion of 6-phosphogluconate to
pyruvate.  The enzymatic formation of pyruvate was con-
firmed by comparing the absorption spectra of the
dinitrophenylhydrozones of standard pyruvate and the
enzymatically generated compound.  Further, the
                                                  °D540
ratio of both products  (1.12 for the pyruvate standard,
1.07 for the sample) are compatible and correspond to
values obtained by Smith and Gunsalus  (73) .  In each case,
there is an optical density difference of  .015 from 490
to 540 nm.  In the assay ferrous ions  are needed for
maximum activity (Table 22) .  It is also apparent that
the reaction will not proceed without  6-phosphogluconate
or extract.  Boiling for 5 rain destroys over 85% of the
activity.  Reduced glutathione, however, does not appear
to be essential.  The reaction is linear with protein
concentration, up to about 3 mg of protein.

As a further control, the formation of pyruvate with
time using 6-phosphogluconate as substrate was compared,
using extracts of T. ferrooxidans grown on iron (9K) ,
                          IOC
-------
                          Table 22




    Requirements for the Entner-Doudoroff enzyme assay
Deletions Pyruyate formed^
(nmoles)
None
Gluconate-6-phosphate
Cell- free Extract
GSH
Fe++
None (enzyme boiled
for 5 min)
340
0
0
340
224
46
% of control
100
0
0
100
65.9
13.5
a_  represents the amount of pyruvate (nmoles) formed in 30 min,
                             101
-------
iron-supplemented medium (9 KG) and Pseudomonas aeruginosa,
an organism known to degrade glucose by way of the Entner-
Doudoroff pathway (74).

The high levels of glucose-6-phosphate dehydrogenase and
Entner-Doudoroff enzymes were not observed when other
sugars or carbon sources were used for energy substrates
(Table 23).  Glucose-grown cells were readily transferred
to several different organic substrates which supported
good growth (Shafia and Wilkinson, 1969).  However, these
cells showed altered activities for both glucose-6-phos-
phate dehydrogenase and Entner-Doudoroff enzymes.  Cells
grown on sucrose or fructose apparently do not use the
Entner-Doudoroff pathway.  Exactly how these various
sugars are metabolized by J_. ferrooxidans remains to be
determined.

Enzymes of the TCA Cycle

Five representative enzymes of the TCA cycle, plus NADH
oxidase, were assayed in cell-extracts prepared from
iron-grown (autotrophic), iron-glucose-grown (mixo-
trophic), and glucose-grown cells.  Enzyme activities
were generally lower in extracts from iron-grown cells
 (Table 24);  succinic dehydrogenase, a-ketoglutarate
dehydrogenase and NADH oxidase were not detected in
these extracts.  NADP+-linked isocitrate dehydrogenase,
fumarase and aconitase were very low, whereas NAD+-linked
isocitrate dehydrogenase levels were high.  The presence
of glucose (iron-glucose grown cells) activates succinic
dehydrogenase, a-ketoglutarate dehydrogenase and NADH
oxidase and greatly stimulates NADP"1"-linked isocitrate
dehydrogenase, fumarase and aconitase.  NAD+-linked
isocitrate dehydrogenase levels remained unchanged in
the presence of glucose.  This same enzyme pattern was
expressed in extracts prepared from cells grown solely
on glucose;  the TCA cycle enzymes were operative and
NADH oxidase activity was readily apparent.

Radiorespirometric studies

To better understand which pathway(s) was involved in glucose
dissimilation, cells grown on the iron-glucose combination
medium and the glucose-salts medium were placed in radio-
respirometer vessels and C0_ formation followed using
-^C-specifically labeled glucose.  The observed labeling
patterns indicate the nature of the pathways involved (62).
Results of these studies with iron-glucose-grown cells
are given in Figure 20.  A very active decarboxylation
-------
                          Table 23

Effect of organic substrate on the activity of glucose-6-phosphate

      dehydrogenase and the Entner-Doudoroff enzymes
Growth
g^
substrate-
Glucose
Sucrose
Fructose
Glutamate
Gluconate
Relative enzyme
Glucose-6-phosphate
dehydrogenase
1.00
1.08
0.41
0.20
0.25
activity!!
Entner-Doudoroff
enzymes—
1.00
0.06
0.08
0.41
0.46
jl
  All organic substrates were added at 0.5% final concentration;
  cells were taken from the late log phase of growth.
b
  Enzyme activity is  compared to the levels found in glucose
  cells.
  6-phosphogluconate dehydrase and 2-keto-3-deoxy-6-phospho-
  gluconate aldolase.
                              103
-------
                                        Table 24
   Activity of TCA cycle enzymes and NADH oxidase in cell extracts of T_. ferrooxidans
grown on various media-
Enzyme
Is o citrate dehydrogenase
(NADP+)
Isocitrate dehydrogenase
(NAD+)
Aconitase
Succinic dehydrogenase
a-ketoglutarate
dehydrogenase
Fumarase
NADH oxidase

Iron -grown
cells
0.65
33.3
2 . 30
N.D.k
N.D.
33.0
N.D.
Activity Cunits/mg
Iron-glucose-
grown eel. Is
55.4
34.7
1050
54.3
30.4
982
17.0
; x 10'3)
Glucose-grown
cells
61.5
27.3
1690
179
9.70
456
30.4
—  In each case, cells were taken at the late log phase of growth,
—  N.D. = not detected under the conditions used.
-------
                                                   Giuco$£-;-c
                                                   GtUCOSf-J-'*C
                                             	o GLUCOSE 3 uc
                                              	x GlUCOSf-3.*-"
                                             	o Giucose-614t
                         MINUTES
Figure 20.   Radiorespirometric pattern  for the utilization of
             specific -^C-labeled  glucose by iron-glucose-grown
             cells of T_. ferrooxijans.   19.2 mg of cells (dry
             weight) were suspended  in the basal salts solution
             making up the iron containing medium (pH 2.5).
-------
occurred at the C-l position, and a peak respiration rate
was reached within 30 min.  The rate of C0_ evolution of
C-4 labeled glucose  (determined by extrapolation) was
higher than from C-3 labeled glucose;  C-3 was respired
less rapidly, with peak evolution of respiratory CC^
occurring after 90 min.  The enzyme 2-keto-3-deoxy-6-
phosphogluconate aldolase splits glucose between carbons
3 and 4;  thus carbons 1 and 4 will appear as the car-
boxyl groups of the resulting pyruvate molecules.  If
the Emden-Meyerhof pathway were operable, CO- derived
from C-3 and C-4 of glucose would appear before C02
derived from C-l.  However, in T_. ferrooxidans grown
in the iron-glucose medium, it appears that much of
the glucose was dissimilated by the Entner-Doudoroff
pathway, with a strong contribution by the pentose shunt
mechanism.  The high CC^ production from C-6 was
unexpected and suggests that the triose phosphate
produced from glucose dissimilation via the Entner-
Doudoroff scheme, and comprising carbon atoms 4, 5 and
6, may be metabolized in an unorthodox manner.  Analysis
of the cumulative   CC^ recovery from glucose (Table 25)
allows one to calculate the relative contribution of
the Entner-Doudoroff and pentose phosphate pathways.
For iron-glucose-grown cells, 37.1% of the glucose is
channeled through the Entner-Doudoroff pathway, while
62.9% of the administered glucose is catabolized by
this organism via the pentose-phosphate shunt.

A similar radioactive inventory was performed on
glucose-grown cells (Table 26).   In this  medium the
Entner-Doudoroff pathway contributes 77.5% to glucose
catabolism, with the pentose-phosphate shunt responsible
for 22.5% of the glucose dissimilated.

If glucose is metabolized through the  Entner-Doudoroff
pathway, the triose phosphate formed must be diverted
into hexose phosphate.   The hexose-phosphate is then
degraded by the pentose cycle (75).  If this explanation
were correct, *4C02 recovered would be rich from carbon
1 and fairly rich from carbon 6.  The  percent   CCU
recovered substantiate  that carbon  1 is indeed evolved
at a high rate, with C02 evolved from  carbon 6 at sub-
stantial, but lower,  levels.  This  form of glucose
metabolism would also necessitate that  carbon 2 be
retained as acetate,  possibly leading  to  the synthesis
of PHB,  known to accumulate in this  organism.
                        106
-------
                                     Table 25


                    14                                                       a
     Utilization of   C-labeled glucose by iron-glucose-grown T.  ferrooxidans-

Substrate
Glucose-l-14C
Glucose-2-14C
Glucose- 3- 14C
Glucose-3,4-14C
Glucose-6-14C
Radioactive distribution-
Respiratory
C02
89.0
15.2
6.1
31.7
37.3
Supernatant Cells
1.5 5.6
8.7 57.5
3.8 81.0
1.2 60.9
4.0 46.4
14
Total C
recovery
96.1
81.4
90.9
93.8
87.7
19.2 mg (dry wt)  cells  were added to  20  ml  of the  salts  of the  9K  medium.
Counts collected at the end of the experiment.
-------
•jo
                                                Table 26



                  Utilization of   C-labeled glucose by glucose-grown T_. ferrooxidans-

Substrate

Glucose- 1-14C
Glucose-2- C
Glucose-3,4-14C
Glucose- 6- 14C
Radioactivity distribution-
Respiratory Supernatant Cells
co2
42.4 54.9 1.1
20.2 51.7 12.1
18.7 75.8 3.3
17.5 50.8 10.1
Total 14C
recovery

98.4
84.0
98.8
78.4
           53.2 mg  (dry wt) cells were added to 20 ml of the ialts of the 9K medium.
           Counts collected at the end of the experiment.
-------
Glucose-6-phosphate Dehydrogenase

     Purification and stability of T.  ferrooxidans  glucose-
     6-phosphate- dehydrogenase

     Glucose-6-phosphate dehydrogenase, a branch point  enzyme
     leading to both the pentose-phosphate shunt and the
     Entner-Doudoroff pathway, was partially purified from
     both iron-glucose and glucose-grown cells.   The purifi-
     cation procedure involved preparative ultracentrifugation
     to remove membrane fragments, ammonium sulfate fractiona-
     tion, Sephadex chromatography and negative  calcium
     phosphate adsorption.  Glucose-6-phosphate  dehydrogenase
     from iron-glucose grown T_. ferrooxidans was eluted from
     the Sephadex G-200 column just after the major protein
     peak.  This elution profile was obtained also  from
     preparations from glucose-grown cells.  Table  27 gives
     a summary of the purification of glucose-6-phosphate
     dehydrogenase from cell-free extracts of iron-glucose
     grown cells.  The enzyme, after treatment with calcium
     phosphate gel, was purified about 90 fold over the crude
     cell-free extract with a recovery of about  15%. Disc
     gel electrophoresis of this enzyme preparation stained
     for protein showed one large band along with several
     minor bands.

     The enzyme from glucose-grown T.  ferrooxidans  was
     purified about 150 fold with about a 4% recovery
     (Table 28).  This preparation also gave 1 large major
     band on polyaery 1 amide gels stained for protein along
     with 6 or 7 minor electrophoretic components.

     For both purified preparations, neither NADH nor NADPH
     oxidase nor 6-phosphogluconate dehydrogenase,  enzymes
     capable of interfering with the assay of glucose-6-
     phosphate dehydrogenase, were detected.  The enzyme
     from iron-glucose or glucose-grown cells was stored
     frozen at -20 C in small samples.  Upon thawing, 10-15%
     of the activity was lost.  Activity was gradually lost
     at room temperature  (50% within 10 h).

     Enzyme activity of the purified preparation was linear
     with protein from 0 to 20 yg of protein using either
     the NADP+ or NAD*-linked reaction.
-------
                                  Table 27



Purification of glucose-6-phosphate dehydrogenase from T_. ferrooxidans

Fraction
Crude extract
1st Ammonium sulfate
G-200 eluate
2nd Ammonium sulfate
Calcium phosphate gel
grown in
Total
Units
51.3
36 . 5
19.5
18.9
7.5
an iron-glucose medium
Total protein
(mg)
1036.0
86.0
18.0
8.1
1.7

Units
mg
0.05
0.42
0.67
2.33
4.34

Recovery (%)
100,0
71.0
38.1
36.8
14.5
-------
                               Table 28



Purification of glucose-6-phosphate dehydrogenase from glucose-grown



                           T,  ferrooxidans

Fraction Total units
Crude extract 407.3
Ammonium sulfate 239/6
1st G-200 eluate 74.0
2nd G-200 eluate 15,4
Total protein (mg) Units/mg Recovery (%)
5087.3 0.08 100.0
315.0 0.76 58.9
44.1 1.68 18.2
1.3 11.9 3.8
-------
Pyridine Nucleotide Specificity of Glucose-6-phosphate
Dehydrogenase.

T_. ferrooxidans glucose-6-phosphate dehydrogenase, at
each stage of the purification, was capable of utilizing
either NADP+ or NAD"1" as coenzyroe (Table 29).  The ratio
of the NADP+-linked activity with respect to the NAD+-
linked reaction at each purification stage was 0.8-0.9,
indicating that there was a single enzyme capable of
both NADP  or NAD"1" reduction.  The confirmation for
the dual nucleotide specificity of T_. ferrooxidans
glucose-6-phosphate dehydrogenase was shown by activity
staining on polyaery1amide gels.  When the enzyme from
iron-glucose grown cells was electrophoresed and gels
stained for enzyme activity, similar scanning profiles
were obtained in the presence of both coenzymes (Table
30).  In each case, a large enzyme peak (peak a), along
with a smaller peak (peak b) was found.  The enzyme
isolated from glucose-grown cells and stained in the
presence of both NADP"1" and NAD+, exhibited similar
profiles.  The RJJ, values of each enzyme preparation,
stained in the presence of both NADP"1" and NAD+ are
compared.  The data indicate that for both the iron-
glucose and glucose-cell preparation, a single enzyme
catalyzes both NADP -and NAD -linked activity.  Moreover,
the RJJJ'S for the enzyme from both iron-glucose and
glucose-grown-eelIs are the same.  Thus, glucose-6-
phosphate dehydrogenase obtained from growth in a mixo-
throphic (iron-glucose) medium shows the same electro-
phoretic properties as the enzyme isolated from a purely
heterotrophic (glucose) environment.  The partially
purified enzyme from each source exhibits identical
NADP+:NAD+ activity ratios.   The same coenzyme activity
ratios were found for the enzyme from iron-grown cells.
However, in the case of extracts from cells grown
autotrophically, electrophoretic activity stains were
not successful due to the extremely low levels of enzyme
activity.

Molecular Weight of Glucose-6-phosphate Dehydrogenase

Straight lines were obtained when per cent gel concentra-
tion was plotted against log R^ for enzyme isolated from
iron-glucose grown cells and from glucose-grown cells.
The slopes of the lines obtained were essentially the
same for each preparation.  When these slopes were fitted
to a standard curve relating the known molecular weight
                         112
-------
                          Table 29

   Comparison of reaction velocity of glucose-6-phosphate

    dehydrogenase with NAD  and NADP  during purification

            of the enzyme from T.  ferrooxidans—
Fraction*!

Crude extract
105,000 x g Supernatant
1st 35% Ammonium sulfate
Velocity—
NADP+
.099
.092
.086
NAD+
.109
.103
.097
Ratio
NADP+/NAD+
0.91
0.89
0.89
   supernatant

2nd 50% Ammonium sulfate
   precipitate

2nd G-200 eluate^-
.029     .032
.058    .068
0.91
0.85
  Cells were grown in the glucose-salts medium.
  0.1 ml of suitably diluted enzyme from each fraction  was  used.
  Velocity is expressed as AOD_.n/min.
  Enzyme from this step was frozen,  diluted and then assayed.
                             113
-------
                          Table 30

 Electrophoretic migration of purified glucose-6-phosphate

             dehydrogenase from T_. ferrooxidans
Enzyme Source            Coenzyme       Peak       R
                                                    m

Glucose cells              NAD+          a        0.416
Glucose cells             NADP+          a        0.422
b        0.287

a        0.422
b        0.297
Iron-glucose cells         NAD           a        0.417
                                         b        0.298

Iron-glucose cells        NADP           a        0.408
                                         b        0.292
                             114
-------
of several proteins to the slopes obtained for log
vs % gel concentration plots, the molecular weight o
T_. ferrooxidans glucose-6-phosphate dehydrogenase was
determined to be 110,000.

Effect of pH on Glucose-6-phosphate Dehydrogenase Activity

When the enzyme was assayed with NADP  at different
pH's of Tris buffer, a relatively sharp optimum was
obtained at pH 7.8-7.9.  Enzyme activity with NAD+ as
coenzyme differed in that no activity was obtained at
pH's below 7.5;  a relatively broad pH optimum was
observed from pH 8.0-8.5.  This pattern emerged regard-
less of the source of enzyme.

Activation by Magnesium Ions

Glucose-6-phosphate dehydrogenase purified from several
microbial sources has been reported to be activated by
Mg"1"1" ions  (61, 76, 77), yet this cation is not an
absolute requirement for activity.  Recently, it has
been reported that the E_. coli enzyme retained 20% of
its activity in the absence of divalent cations (77).
The enzyme from T_. ferrooxidans retains 50% of its
activity in the absence of MgT+.  The effect of Mg
on the velocity of the reaction catalyzed by T_. ferrooxi-
dans glucose-6-phosphate dehydrogenase was tested.   From
Lineweaver-Burk plots, an apparent K  (Mg"1"1") of 4.0  mM
was obtained.  Mn++ cations were effective at low con-
centrations (0.27 mM) in activating the enzyme;  how-
ever, at higher concentrations, this effect was markedly
diminished. At optimum concentrations of both cations,
Mg++ was substantially more effective.  Fe"1"1" was not
tested since it is autooxidized at pH's above 4.0.  No
other cation was tested.

Apparent Michaelis Constants

The Michaelis constants for NADP+, NAD+, and G-6-P
were obtained by conventional procedures for the enzyme
isolated and purified from iron-glucose-grown cells.
For glucose-6-phosphate dehydrogenase isolated from
glucose-grown cells, Michaelis constants for each of
the above substrates were obtained similarly.  In all
cases, hyperbolic curves were obtained from velocity
versus substrate concentration plots.  There was no
apparent sigmoidicity.  The apparent Km's for the
                          115
-------
substrates for the enzyme from both iron-glucose and
glucose-grown cells were comparable (Table 31) and
indicate along with the electrophoretic data that the
enzyme is the same regardless of whether the cell is
grown under mixotrophic or heterotrophic conditions.

Effect of NADH

NADH has been reported to inhibit glucose-6-phosphate
dehydrogenase isolated from E^.  coli in an allosteric
manner (77).  However, glucose-6-phosphate dehydrogenase
isolated from the chemolithotroph T_. ferrooxidans does
not exhibit a sigmoidal inhibition curve with NADH,
indicating that NADH is not an allosteric effector in
this case.  Other workers (78)  have previously shown
that the enzyme from the facultative chemolithotrophy
Hydrogenomonas H-16 (grown on fructose) is noncom-
petitively inhibited by NADH.  Thus, from results
with at least two different autotrophic species, NADH
does not appear to be an allosteric effector for
glucose-6-phosphate dehydrogenase.  In addition,
Olive et al (79) have recently shown that homogeneous
Leuconostoc mesenteroides glucose-6-phosphate dehydro-
genase is not allosterically affected by NADH.

Effect of ATP

ATP has been implicated as an allosteric effector with
respect to glucose-6-phosphate concentration for glucose-
6-phosphate dehydrogenases isolated and purified from
a number of microbial sources (61, 78).  The effect
of ATP concentration on the velocity of glucose-6-phos-
phate dehydrogenase isolated from T_. ferrooxidans was
assayed with NADP+.  The kinetics for this inhibition
by ATP at various concentrations was done.  In no
instance did ATP affect the hyperbolic nature of the
velocity versus substrate concentration curve.  From
double reciprocal plots it appears that ATP inhibits
in a nearly competitive manner when NADP+ is used as
coenzyme.  Dixon plots confirm the inhibition by ATP
and reflect the apparent sigmoidicity found a K^ for ATP
of approximately 3 mM is obtained at 50% inhibition.
Hill plots of these data give slopes of 0.9-1.2.   ATP
does not appreciably change the slope and the values
for n obtained (•~'l), indicate  that G-6-P binds to one
site on the enzyme.
                          ne
-------
                          Table 31

 Comparison of apparent Michaelis constants for partially

purified T_. ferrooxidans glucose-6-phosphate dehydrogenase
 Enzyme source
                            Apparent K   (M)
                   NADP
                NAD
G-6-P
 Iron-glucose-
    grown cells

 Glucose-grown
    cells
1.02 x 10"5     6.58 x 10"4    5.18 x 10"5
2.43 x 10"5     2.08 x 10~3    5.39 x 10"5
                               117
-------
     ATP exhibits a profound effect on the NAD+-linked reaction.
     The kinetics of the ATP inhibition of the NAD*-linked
     reaction with respect to varying G-6-P concentration
     were tested.  Again, no deviation from normal Michaelis-
     Menten kinetics was observed.  Double reciprocal plots
     indicate competitive inhibition when NAD+ is used as
     coenzyme.  Dixon plots confirm the competitive nature
     of this inhibition.  From these plots, a K. of about
     0.3 mM for ATP was obtained, 10 times less than the
     value obtained with NADP"1" used as coenzyme.  Thus, ATP
     is 10 times more inhibitory to the NAD+-linked reaction.
     Hill plots give slopes of 0.9-1.1 in the absence or
     presence, of ATP for the NAD+-reaction.

     Effect of Other Adenine Nucleotides.

     The inhibition by ATP is not shared by other adenine
     nucleotides (ADP, AMP, cyclic-3', 5' AMP) (Table 32).
     Only in the presence of ATP was substantial inhibition
     (80-90%) noted.

                           Discussion

Effect of Organic Metabolites on Autotrophic Mechanisms

The results of this study document the ability of T_. ferrooxi-
dans to grow on glucose as the sole source of energy.  Shafia
and Wilkinson (57) had confirmed an earlier observation of
such growth by this organism (58).  Implicit in each of these
studies, however, was the observation that autotrophically
grown T_. ferrooxidans failed to grow in the glucose medium
unless the cells were first grown in a ferrous sulfate medium
containing 0.5% glucose.  After transfer, once again, in the
iron-glucose media, the progeny of this second culture were
fully adapted to glucose.  In the present study, large
quantities ( 5 g) of autotrophically grown T_. ferrooxidans,
when incubated in the iron-glucose medium, failed to oxidize
the glucose;  the ferrous sulfate, however, was completely
oxidized within 3 h.  When these cells were subsequently
transferred to fresh iron-glucose media, glucose was fully
oxidized within 60 h.  Noteworthy in this experiment, is the
fact that the cells prefer iron to glucose;  glucose is
utilized some time (10 h) after the completion of iron
oxidation.  It may be that the further addition of the inorganic
substrate, iron, is needed to provide energy for the initial
transport of glucose across the cell membrane.  This appears
to be the case with several of the chemolithotrophic bacteria
                               118
-------
                          Table 32

                                                      Q
      Effect of adenine nucleotides on T_. ferrooxidans—

            glucose-6-phosphate dehydrogenase—
Nucleotide—
	
ATP
ADP
AMP
3', 5' -AMP
ATP + AMP
ATP + 3', 5 '-AMP
ATP + ADP
Velocity
6.77
1.45
5.00
5.65
5.65
1.37
0.73
1.05
Per Cent Activity
100
21.4
73.9
83.5
83.5
20.2
10.8
15.5
jl
  Enzyme obtained  from  glucose-grown  cells.

b_
  The reaction mixture  in  a total volume of  1.0 ml  contained:
  Tris-HCl  (pH 7.9),  15 mM;  G-6-P, 1 mM;  NADP+, 0.35  mM;
  enzyme, 2 yg»  and adenine nucleotide.
  All adenine nucleotides were  at  3 mM  concentration.
-------
examined to date;  e.g., the addition of ammonium chloride
roughly doubled the incorporation of several amino acids by
resting cells of Nitrosomonas europaea (80).  The observed
lag period before glucose is oxidized by T_. ferrooxidans
may also represent a period when the cell is synthesizing
"transport specific" enzymes such as Enzyme II of the phospho-
transferase (PTS) system of Kundig,et  al (81).  No experi-
mental evidence is available to support the transferase
suggestion.  Alternatively (or concurrently), the cell may be
synthesizing necessary inducible enzymes for glucose dissimila-
tion during the lag period after iron oxidation.  One enzyme
involved in the catabolism of glucose is glucose-6-phosphate
dehydrogenase, an enzyme whose activity roughly parallels the
utilization of glucose in the iron-glucose medium.  Poly-B-
hydroxybutyrate  (PHB) synthesis also parallels the utilization
of glucose by T_. ferrooxidans.

Once adapted to glucose from the iron-glucose medium, T_.
ferrooxidans was able to grow on glucose as its sole energy
source.  However, glucose assimilation in a thiosulfate-
glucose medium by T_. intermedius ceases with exhaustion of
thiosulfate (82).  Thus, T. ferrooxidans appears to resemble
J_' novellus and various Hydrogenomonas species which grow on
simple organic compounds.  On glucose, T_. ferrooxidans showed
a generation time of 4.5 h, as opposed to about 8-10 h on
iron  (56) and about 22 h on sulfur.  Glucose-grown cells
differed significantly in morphology, as compared to auto-
trophically grown T_. ferrooxidans (83).  Glucose cells usually
appeared in  pairs or short chains and appeared to possess
many PHB storage granules, causing the cells to bulge (83).

T_. ferrooxidans was also capable of growth on a variety of
organic substrates (57).  In this investigation, fructose
appeared to give the highest growth rate.  The growth stimula-
tion by yeast extract is concentration dependent and suggests
a cofactor requirement for maximum growth, possibly p-amino-
benzoic acid  (PABA), as suggested by Shafia and Wilkinson
(57).  However, no studies have been made to establish this.

J_. ferrooxidans failed to grow on the organic acids acetate,
pyruvate or succinate.  If the hypothesis of Rao and Berger
(84) is correct, this would be expected.  Thus, at the low
pH required for growth, these acids would enter the cell in
an undissociated form, then dissociate and permit the further
passive entry of acids.  As acid accumulates in the cell, there
would be a concomitant decrease in internal pH with a sub-
sequent inhibition of growth.
                               120
-------
In contrast to the studies of Shafia and Wilkinson (57)  with
strain BCU-4, the adaptation to glucose has  a profound effect
on the cells' iron oxidation ability.   This  effect is  also
found when other organic compounds were added to the iron
containing media.  The degree of iron oxidation was dependent
on the particular energy source;  cells taken from an  iron-
glucose medium lost almost 80% of the ability to oxidize
iron, while cells taken from an iron-fructose medium retained
almost 80% of the cells  iron oxidation potential.  In each
case cells were removed from the medium after the FeSO^ had
been oxidized.  There is support for the belief that repression
of the autotrophic machinery is likely, for assays were done
with washed resting cells in the absence of any organic
supplement.  Thus, it would appear that a product of hetero-
trophic metabolism is responsible for the repression of the
autotrophic iron oxidation mechanism.   Currently, there is
only one pathway thought to be responsible for iron oxida-
tion, that is, through the cy to chrome chain  (25, 85, 86)
     Fe^+v  .^oxidized cytochrome c .^  _ reduced cytochrome  a

       d i"                      /   \
     Fe°     reduced cytochrome c     x oxidized cytochrome a
                      ADP + P;     ATP
Iron oxidation involves cytochromes c and a, and the enzymes
iron-cytochrome c reductase and cytochrome oxidase, in that
order.  Any represser of iron oxidation would necessarily have
to act at the level of energy generation, presumably at the
cytochrome chain.  An analogous situation has been described
in Thiobacillus noyellus, where fermentable substrates such as
glucose, glycerol, lactate, ribose and pyrivate repressed most
of the enzymes involved in sulfur oxidation (87).  However,
the degradation of glucose cannot give rise to any of the
intermediates of sulfur oxidation which might then cause a
type of feedback inhibition.  Nevertheless, in T_. ferrooxidans
(59) and in T. novellus, glucose exerts a profound influence
on the sulfur oxidizing systems.  Thus, the effector responsible
for the repression would necessarily be a product of hetero-
trophic metabolism.  This effector would also be a product
of the oxidation of iron or sulfur compounds if this system
adheres to the definition of Magasanik (88) which states that
the synthesis of certain enzymes is regulated by some
                               121
-------
 intracellular products of the enzymes concerned.  The only
 common by-products of both inorganic iron or sulfur metabolism
 and heterotrophic metabolism are ATP and NADH.  Presumably the
 level of these energy-rich compounds would vary depending upon
 the ability of the cell to degrade the organic compound in
 question.  According to Lejohn, Van Caseele and Lees (87), the
 cell is probably adjusting its energy imbalance by reducing
 the production of excess energy through some form of inhibition.
 In the  case of T. ferrooxidans, iron oxidation and the resulting
 formation of ATP~would fall into this category.

 In order for autotrophic organisms to exist, there must be
 some mechanism for assimilating carbon dioxide directly into
 cell substance. The means by which this occurs is through
 the reductive pentose-phosphate cycle or Calvin-Benson pathway
 (89).  The key reaction in this scheme is catalyzed by ribulose
 diphosphate (RuDP) carboxylase.  The reaction catalyzed by
 this protein results in the fixation of 1 molecule of carbon
 dioxide to a molecule of ribulose 1,5-diphosphate with a con-
 comitant dismutation to yield two molecules of phosphoglyeerie
 acid.   In J_. ferrooxidans, a repression of RuDP carboxylase
 activity was found in cell-free extracts prepared from iron-
 glucose, iron-sucrose, iron-fructose, and iron-glutamate grown
 cells.  This has also been shown with several facultatively
 autotrophic species, London and Rittenberg (82) reported
 repression of thiosulfate oxidation in J_. intermedius as well
 as a decreased level of RuDP carboxylase prepared from cells
 propagated in a medium containing thiosulfate and an organic
 supplement.  McFadden and Tu (90) also noted a similar repres-
 sion of enzyme activity in Hydrogenomonas facil'us.

 There is a relationship between inorganic iron oxidation and
 RuDP carboxylase activity.  It was found that maximum iron
 oxidation, during growth and by resting cells, precedes the
 expression of maximum RuDP carboxylase activity.   This is to
 be expected since the oxidation of ferrous iron provides energy
 (ATP) which is then utilized in the reduction of CC^ via the
 Calvin cycle.   The reduction of CO^ is a completely endergonic
 process, requiring large amounts of ATP and NADH.  Any reduction
 in the amount of ATP supplied to the system will result in
 diminished CC^ fixation and lower levels of RuDP carboxylase.

 Heterotrophic Metabolism of T. ferrooxidans.

 Growth on glucose affords the cell an alternate mechanism for
 energy generation through a form of heterotrophic metabolism.
 From the present results, based upon enzymatic studies and
 14c-labeling data, the Embden-Meyerhof pathway does not appear
to be the major route of glucose dissimilation in T.  ferrooxidans.
                              122
-------
Instead, glucose is dissimilated through the Entner-Doudoroff
pathway involving glucose-6-phosphate dehydrogenase, 6-phos-
phogluconate dehydrase, and 2-keto-3-deoxy-6-phosphogluconate
aldolase.  The aforementioned activities are enhanced several
fold during glucose catabolism.  The Embden-Meyerhof pathway
probably plays a minor role in providing carbon skeletions
for biosynthesis.  Thus, this chemoautotroph is similar to
Thiobacillus intermedius (72), Hydrogenomonas eutropha (91) and
Hydrogenomonas H-16 (92) in using the Entner-Doudoroff scheme
for glucose dissimilation.

Pyruvate and 3-phosphoglyceraldehyde are the products of
glucose breakdown through the Entner-Doudoroff pathway.  For
energy to be derived from this degradation, these products
must be channeled through the TCA cycle.  In J_. ferrooxidans
grown on glucose, an increase in activity of TCA cycle enzymes
was noted.  It has been suggested that the TCA cycle merely
serves a biosynthetic function in chemoautotrophs (93), since
there is an apparent absence of a-ketoglutarate dehydrogenase
and low levels of succinic dehydrogenase.  A lack of NADH
oxidase was also reported and given as the reason for the
inability of the complete TCA cycle to function (93).  Several
authors, however, have found NADH oxidase to be present in
extracts from several chemolithotrophs  (72, 94, 95).  From results
of these studies, J_. ferrooxidans appears to have an incomplete
TCA cycle when grown autotrophically, as well as an apparent
absence of NADH  oxidase.  However, when grown heterotrophically
(on glucose) the TCA cycle is functional and the cells exhibit
an increase in NADP"1"-linked isocitrate dehydrogenase, aconi-
tase, fumarase,  succinic dehydrogenase, and a-keto-glutarate
dehydrogenase.  NADH oxidase was also readily detected in
glucose-grown cells.  Evidence is thus  accumulating supporting
the idea that the TCA cycle in chemolithotrophic bacteria is
incomplete and functions as a biosynthetic mechanism when
autotrophic conditions exist.  The presence or absence of
NADH oxidase in  autotrophically grown cells probably reflects
a function of the regulation of its activity  (96).

The presence of  two isocitrate dehydrogenases  (NAD+ - and
NADP+-linked) may well have important regulatory functions
for this organism.  It is known that the NAD+- enzyme is
activated by AMP.  Thus, if the NAD+-and NADP+-enzymes
compete for the  same pool of isocitrate, any excessive energy
drain, such as the Calvin cycle, will favor the NAD"1" enzyme.
This would occur since the relative concentrations of ATP
would decrease, while that of AMP would increase.  Thus,
electrons would be channeled through the electron transport
chain rather than toward biosynthesis.  By contrast, under
                               123
-------
conditions where ATP accumulates in high concentrations, i.e.,
as the result of glucose catabolism, the NADP+-enzyme would
be stimulated.  In addition, citrate would accumulate (due to
the equilibrium of the aconitase reaction), stimulating fatty
acid synthesis by activation of acetyl coenzyme A carboxylase.
Under these conditions acetate could be stored in the form of
PHB.

Thiobacillus intermedius is similar to T_. ferrooxidans in that
an NADP"1"-specific isocitrate dehydrogenase is induced by
glucose, while the activity of the NAD+-enzyme remains constant
(72).

                          14
Labeling experiments with   C-specifically labeled glucose show
an early release of C-l and C-4 labeled C0_, indicating that
the Entner-Doudoroff pathway is operating.  Respiratory CC^
from C-3 is evolved at least 1 h later and at substantially
lower levels.  Thus, the Embden-Meyerhof pathway cannot be of
importance in energy generation, for if this pathway were
operating, the preferential rate of evolution of C-3, 4 would
be greater than that of C-l, since this pathway*yields 2
molecules of pyruvate in which the carboxyl groups are derived
from carbons three and four  of glucose.  If the pentose-
phosphate cycle were functional, the rate of   CO^ evolution
would be C-l > C-2 > C-3 > C-4 > C-6.  In T_. ferrooxidans,
this is  dearly not the observed pattern.  In the Entner-
Doudoroff pathway, the glucose is split between carbon atoms
three and four, with the result, however, that carbons one
and four appear as the carboxyl groups of the resulting pyruvate
molecules.  In addition, if this pathway were the sole means
of glucose degradation, it would follow that C-3 should equal
C-6.  If the Entner-Doudoroff pathway were operating in con-
junction with any of the other pathways, C-l would no longer
equal C-4, and C-3 would not equal C-6.  In these experiments
C02 from C-6 labeled glucose was released earlier than respira-
tory CCL from C-3 labeled glucose.  This was unexpected based
upon established dissimilation schemes for bacteria (62).
Figure 21 represents a scheme which could account for the way
that glucose is dissimilated in T_. ferrooxidans.   One explana-
tion for the high C-6 evolution is the fact that all autotrophic
organisms, both photosynthetic and chemosynthetic, form fructose
1,6-diphosphate by joining one molecule of 3-phosphoglycer-
aldehyde and one molecule of dihydroxyacetone phosphate,
essentially a reversal of the aldolase step of the Embden-
Meyerhof pathway (89).  In addition, autotrophs used phospho-
glyceric acid kinase and phosphoglyceralde dehydrogenase
in the reverse direction (89).   Thus, in the case at hand, when
phosphoglyceraldehyde is formed as a result of the Entner-
                             124
-------
                     HCO
                   I
                   2
                   J M04H
                   4  H
                   5  HCOH
                   6  CH]OP
                     «-4-f
              AW lyrNAO'^X'NADP*

                ^>NADHv{v
-------
Doudoroff scheme, it is able to be isomerized to dihydroxy-
acetone phosphate (Fig. 21).  Triose phosphate isomerase has
been detected in J_.  ferrooxidans (97).  Aldolase then joins
these two molecules into a molecule of fructose 1,6-diphos-
phate;  carbon 6 (of the original glucose molecule) from 3-
phosphoglyceraldehyde is now the number one carbon atom of
fructose diphosphate.  Fructose diphosphate can be easily
converted to fructose-6-phosphate and then to glucose-6-
phosphate by means of fructose diphosphatase and phospho-
glucose isomerase.  Phosphoglucose isomerase has been detected
in extracts of T. ferrooxidans (97).  Glucose-6-phosphate then
recycles through the pentose phosphate pathway or through
the Entner-Doudoroff pathway with the subsequent liberation
of C0? from the original C-6 of glucose.

Glucose is dissimilated primarily (80%) through the Entner-
Doudoroff scheme when J_. ferrooxidans is grown on glucose.
The presence of the autotrophic substrate iron (iron-glucose
cells) appears to repress the En tner-Doudoroff pathway,
since this pathway accounts for only 40% of the glucose
dissimilated under these conditions.  T_. ferrooxidans is
thus similar to Hydrogenomonas H-16 (92) and J_. intermedius
(72) in that the Entner-Doudoroff pathway is repressed by
the chemolithotrophic energy source.

One of the interesting observations found was the formation of
large masses of poly-6-hydroxybutyrate granules in cells of
T_. ferrooxidans grown on glucose (83).  The synthesis of this
storage polymer requires reduced NAD"1".  Reduced NAD"1" is readily
available through glucose-6-phosphate dehydrogenase, and
isocitrate dehydrogenase (NAD+-linked) activity.  NADH
availability for possible PHB synthesis is also shown in
Figure 24.  NADP"1"-linked isocitrate and glucose-6-phosphate
dehydrogenases supply needed NADPH for fatty acid synthesis.
The two isocitrate dehydrogenase activities are probably two
distinct enzymes, since the NAD"1"- and NADP+-linked activities
vary to different degrees depending upon how the cells are
grown.  The ratio of NADP+/NAD -linked activity for glucose-6-
phosphate dehydrogenase is the same regardless of how the
cells are grown, thus indicating there is only one protein
catalyzing these activites.  This subject is discussed in the
following section.

Glucose-6-phosphate Dehydrogenase

Glucose-6-phosphate dehydrogenase catalyzes an important step
in the heterotrophic metabolism of J_. ferrooxidans.  The enzyme
uses either NADP"1" or NAD  as coenzyme based upon the constant
ratio of enzyme activities with either coenzyme during each step
                               126
-------
of the enzyme purification procedure;  and from results of
activity staining in polyacrylanri.de gels.  However, the
NADP+- and NAD*-linked activities do respond differently
to changes in pH and ATP concentration;  these differences
may suggest differences in the formal kinetic mechanism for
these reactions (79).  In addition, it was postulated that
both reactions (NAD+ and NADP+) proceed by an ordered sequen-
tial mechanism, with an isomerization of free enzyme for only
the NAD+-linked reaction.  It is thought that NAD+ reacts with
one form of the enzyme (E) whereas NADH, NADP+ and NADPH
react with another form (E1) (79).

Glucose-6-phosphate dehydrogenase is identical when isolated
from cells  grown mixotrophically  (iron-glucose grown cells)
and cells grown heterotrophically (glucose-grown cells).  The
data showing identical electrophoretic migration patterns,
similar Michaelis constants for substrates, identical
NADP+/NAD+ activity ratios and similar molecular weights
support this claim.  In addition, extracts of autotrophically
grown cells  (iron-grown cells possess NADP+/NAD+ ratios
identical to extracts from mixotrophically and heterotrophically
grown cells.  All evidence supports the fact that glucose-6-
phosphate dehydrogenase is identical in properties, no matter
how drastic  a difference in the organism's primary energy
source.  Thus, there does not appear to be a different enzyme
synthesized  as a result of the induction by glucose, as with
isocitrate dehydrogenase  (72).  Conversely, the low level of
glucose-6-phosphate dehydrogenase found in autotrophic extracts
is due to a  regulation of the synthesis of the one enzyme;  the
organism does not have the genetic capability of synthesizing
large amounts of a different glucose-6-phosphate dehydrogenase
to metabolize glucose heterotrophically.  It would then appear
that complex regulatory processes are operating to control
the synthesis of a single glucose-6-phosphate dehydrogenase.
These processes are affected by the substrate milieu in which
the organism is grown.  In addition, the conditions of growth
do not appear to induce a change  in enzyme property, be it
structural or catalytic.

J_. ferrooxidans glucose-6-phosphate dehydrogenase is activated
by Nig"*"*", and to a lesser extent by low concentrations of Mn++.
This property is shared by a number of bacterial glucose-6-
phosphate dehydrogenases;  among  these sources are E_. coli
(77) and Pseudomonas aeruginosa.  In contrast, however,
glucose-6-phosphate dehydrogenase from Leuconostoc mesenter-
oides (79) and Hydrogenomonas H-16 (78) is not activated by
                              127
-------
the enzyme from T_. ferrooxidans, although inhibited by
NADH, is not affected in an allosteric manner, as in E_. coli. (77)
Instead, the enzyme from T_. ferrooxidans is similar to
glucose-6-phosphate dehydrogenase from the related organism,
Hydrogenomonas H-16 (78), where NADH inhibition is noncompeti-
tive with respect to NADP+.  In addition, the inhibition of
T_. ferrooxidans glucose-6-phosphate dehydrogenase by ATP does
not deviate from normal Michaelis-Menten kinetics as illustrated
by the enzyme isolated from Hydrogenomonas H-16 and Ps.
aeruginosa..  ATP inhibits the enzyme from T_. ferrppxidans
nearly competitively with respect to glucose-6-phosphate
concentration when NADP  is used as the coenzyme;  the K^
was calculated to be approximately 3 mM.  However, when
NAD"1" is the coenzyme, ATP inhibits competitively with respect
to glucose-6-phosphate;  in this case the K^ for ATP was
determined to be 0.3 mM.  There is apparently only one binding
site for glucose-6-phosphate, since the slopes from the Hill
plots gave values of 11 approximately equal to one.  Glucose-
6-phosphate dehydrogenase from another autotrophic organism.
Hydrogenomonas H-16 (78) and from the heterotroph Ps^ aeruginosa
(61), values of n_ varied from 2-3 indicating the presence of
at least 2 binding sites for glucose-6-phosphate by these
enzymes.  The enzyme from each of these sources exhibited
sigmoidal velocity versus glucose-6-phosphate concentration
curves;  ATP enhanced the sigmoidicity of these curves, a
property not shared by the enzyme from T_. ferrooxidans.

T_. ferrooxidans glucose-6-phosphate dehydrogenase preferen-
tially utilizes NADP"1";  the affinity of the enzyme for NADP*
is 60-80 times greater than the affinity for NAD+.  The
glucose-6-phosphate dehydrogenase from Hydrogenomonas H-16
(78) and L^. mesenteroides  (79) shows a 6-fold and 16-fold
preference for NADP"1" respectively, based on Michaelis  con-
stants.

The inhibition by ATP may be important physiologically for
T_. ferrooxidans, especially with respect to the NAD+-linked
reaction, where the K^ is 10 times lower than that for the
NADP+ reaction.  It has been suggested that unidentified
control mechanisms linked to the respiratory system affect
the entry  of  glucose into dissimilatory pathways  (98).
T. ferrooxidans possesses a relatively simple cytochrome
chain:

       NADH	^  cyt c	>cyt a	> 0~
                                12C
-------
The product of the oxidation of NADH through this system is
ATP.  Thus, the reduction of NAD"1", catalyzed by glucose-6-
phosphate dehydrogenase, is needed for the subsequent genera-
tion of ATP.  The ATP formed is then capable of inhibiting
the action of glucose-6-phosphate dehydrogenase, by a type
of product inhibition, thus regulating the flow of carbon
through the Entner-Doudoroff pathway.  This may be important
under autotrophic conditions of growth where relatively large
amounts of ATP and NADH are needed for the endergonic reduc-
tion of CC^.  In addition, the amount of glucose-6-phosphate
in the cell may be a major factor in controlling its oxidation
by glucose-6-phosphate dehydrogenase and the subsequent carbon
flow.  However, the relative concentration of glucose-6-phos-
phate with respect to that of ATP is also a determinant in
governing the activity of glucose-6-phosphate dehydrogenase.
This would appear tenable, since at high concentrations of
substrate, the inhibition by ATP would be relieved.

Recently, it has been discovered that several glucose-6-
phosphate dehydrogenases, including the enzyme from T_. ferrooxi-
dans, are regulated by acetyl coenzyme A (A. Ishaque, personal
communication).  It is particularly interesting that the
NAD"1"-reaction is inhibited to a greater extent than the NADP -
linked reaction.  It is conceivable that the regulation of
glucose-6-phosphate dehydrogenase may be a determining factor
in controlling T_. ferrooxi dans' ability for a dual existence,
autotrophic or heterotrophic.
                              129
-------
                         SECTION VIII
                       ACKNOWLEDGEMENTS
An extensive research project such as this can only be
successful when full cooperation is achieved by people
involved.  The senior author stresses the collaborative nature
of the project and thanks the many contributors involved.
Recognition goes to Dr. Robie Vestal for experimental
expertise and professional guidance;  Dr. Gus Wang for
electron microscopy and physiological techniques;  Dr. Robert
Tabita for vision and the pursuit of fundamentals, Dr. Carl
Schnaitman for kinetic thoughts;  Dr. Jerry Perry for
supporting experiments, Mr. Carl Bodo, Miss Adele Howard,
Mr. Robert Tuttle, and Mike Kaplan for student power and the
pursuit of zealous research.
                               131
-------
                            SECTION IX
                            REFERENCES
 1.  Singer, P. C. and W. Stumm.   1970.  Acid mine drainage:
          The rate-determining step.  Science.  167:1121-1125.

 2.  Silver, M. and D. G. Lundgren.  1968.  Sulfur-oxidizing
          enzyme of Ferrobacillus ferrooxidans (Thiobacillus
          ferrooxidans).   Can. J. Biochem. 46_: 456-461.

 3.  Silver, M. and D. G. Lundgren.  1968.  The thiosulfate-
          oxidizing enzyme of Ferrobacillus ferrooxidans
          (Thiobacillus ferrooxidans).   Can. J. Biochem.
          46_: 1215-1220.

 4.  Sorbo, B. H.  1953.   Crystalline rhodanese.   Acta Chem.
          Scand.  7:238-239.

 5.  Bowen, P. J., J. Butler, and F. C. Happold.   1965.  Some
          properties of the rhodanese system of Thiobacillus
          denitrificans.   Biochem. J.  97:651-657.

 6.  Peck, H. D., Jr. and E. Fisher.  1962.  The oxidation of
          thiosulfate and phosphorylation in extracts of
          Thiobacillus thioparus.  J. Biol. Chem. 257:190-197.

 7.  Taylor, B. F.  1968.  Oxidation of elemental sulfur by an
          enzyme  system from Thiobacillus neapolitanus.  Biochim.
          Biophys. Acta.    170:112-122.

 8.  Charles, A.  H.   1969.  Mechanism of thiosulfate oxidation
          by Thiobacillus intermedius.   Arch. Biochem. Biophys.
          129:124-150.

 9.  Charles, A.  M. and I. Suzuki.   1965.  Sulfite oxidase of a
          facultative autotroph, Thiobacillus novellus.
          Biochem. Biophys. Res. Commun. 19_:686-690.

10.  Charles, A.  M. and I. Suzuki.   1966.  Mechanism of thio-
          sulfate oxidation by Thiobacillus novellus.  Biochim.
          Biophys. Acta.  128:510-521.

11.  Charles, A. M. and I. Suzuki.   1966.  Purification and
          properties of sulfite:  cytochrome c^ oxidoreductase
          from Thiobacillus novellus.  Biochim. Biophys. Acta.
          128:522-534.
                                133
-------
12.   Lyric,  R.  M.  and I.  Suzuki.   1970.   Enzymes  involved  in
          the metabolism of thiosulfate  by Thiobacillus  thio-
          parus.   I.   Survey of enzymes  and properties  of
          sulfite:   cytochrome £ oxidoreductase.   Can.  J.
          Biochem.  48:534-345.

13.   Margalith, P., M.  Silver,  and D.  G.  Lundgren.   1966.
          Sulfur oxidation by the iron bacterium  Ferrobacillus
          ferrooxidans.   J.  Bacteriol. 92:1706-1709.

14.   Lineweaver,  H. and D. Burk.   1934.   The determination of
          enzyme dissociation constants.   J.  Am.  Chem.  Soc.
          56_: 658-666.

15.   Sorbo,  B.   1957.  A colorimetric method for  the deter-
          mination of thiosulfate.   Biochim.  Biophys. Acta.
          ^3:412-416.

16.   Silverman, M.  P. and D. G. Lundgren.  1959.   Studies  on
          the chemosynthetic iron bacterium Ferrobacillus
          ferrooxidans.   I.   An improved medium and a harvesting
          procedure for securing high cell yields.   J.  Bacteriol.
          ^77:642-647.

17.   Lowry,  0.  H.,  N. J.  Rosebrough, A.  L. Farr,  and R.  J.
          Randall.   1951.  Protein measurements with the Folin
          phenol reagent.  J. Biol. Chem. 193:265-275.

18.   Davis,  B.  J.  1964.   Disc electrophoresis  -  II.  Method
          and application to human serum proteins.   Ann. N. Y.
          Acad. Sci.  121:404-427.

19.   Hedrick, J.  L. and A. J. Smith.  1968.  Size and charge
          isomer separation and estimation of molecular
          weights of proteins by disc gel electrophoresis.
          Arch. Biochem.  Biophys. 126:155-164.

20.   Suzuki, I. and M. Silver.   1966. The initial product and
          properties of the sulfur oxidizing enzyme of  thio-
          bacilli.   Biochim. Biophys. Acta. 122:22-35.

21.   Fine, I. H. and L.  A. Costello.  1963.  The  use of starch
          electrophoresis in dehydrogenase studies,  p.  958-972.
          D4 S. P.  Colowick and N.  0. Kaplan (eds).  Methods
          in Enzymology, Vol. 6.  Academic Press.  New  York.
                                134
-------
22.   Peck, H.  D., Jr., T. E. Deacon, and J.  T.  Davidson.   1965.
          Studies on adenosine-5'-phosphosulfate reductase
          from Desulf o yibrio desulfuricans and Thiobacillus
          thioparus.  Biochim. Biophys.  Acta.  96:429-446.

23.   Smith, A. J. and J.  Lascelles.  1966.  Thiosulfate
          metabolism and rhodanese in Chromatium sp.  strain
          D. J. gen. Microbiol. 4_2_: 357-370.

24.   Green, J. R. and J.  Westley.   1961.  Mechanism of rhodanese
          action:Polarographic studies.   J.  Biol. Chem. 236:
          3047-3050.

25.   Blaylock, B. A. and A. Nason.  1963.  Electron transport
          systems of the chemoautotroph  Ferrobaci11us ferrooxi-
          dans.  I.  cytochrome c_-containing iron oxidase.
          J. Biol. Chem.  258:3453-3462.

26.   Aleem, M. I. H., H.  Lees, and D. J. D.  Nicholas.  1963.
          Adenosine triphosphate-dependent reduction of
          nicotinamide adenine dinucleotide by ferrocyto-
          chrome £ in ehemoautotrophic bacteria.  Nature.
          200:759-761.

27.   Starkey, R. L.  1935.  Isolation of some bacteria which
          oxidize thiosulfate.  Soil. Sci. 39:197-219.

28.   Hutchinson, M., K. I. Johnstone, and D. White.  1969.
          Taxonomy of the Genus Thiobacillus:   the outcome of
          numerical taxonomy applied to the group as a whole.
          J. gen. Microbiol. 57:397-410.

29.   Remsen, C. C. and D. G. Lundgren.  1966.   Electron microscopy
          of the cell envelope of Ferrobacillus ferrooxidans
          prepared by freeze-etching and chemical fixation.
          J. Bacteriol. 92_: 1756-1771.

30.   Maser, M. P., T. E. Powell, III, and C. W. Philpott.
          1967.  Relationships among pH, osmolality and
          concentration of  fixative solution.  Stain. Technol.
          42_:175-182.

31.   Luft, J.  H.  1961.  Improvements in epoxy resin embedding
          methods.  J. Biophys. Biochem. Cytol. 9_: 409-414.
                               135
-------
32.  Reynolds,  E.  S.   1963.   The use of lead citrate  at  high pH
          as an electron-opaque stain in electron microscopy.
          J. Cell  Biol,  17:208-212.

33.  Wang, W.  S. and D.  G.  Lundgren.  1968.   Peptoglycan of a
          chemolithotrophic bacterium,  Ferrobacillus  ferrooxi-
          dans. J.  Bacteriol.  95:1851-1856.

34.  Westphal,  0.  and K.  Jann.   1965.  Bacterial  lipopoly-
          saccharides.  Extraction with phenol-water  and
          further  application of the procedure,   p. 83-91.
          _m R. L.  Whistler and M. L. Wolfrom.   (eds.),
          Methods  in carbohydrate chemistry, vol. 5.
          Academic Press, Inc., New  York.

35.  Bellamy,  A. R.  and R.  K. Ralph.  1968.   Recovery and
          purification of nucleic acids by means  of cetyl-
          trimethyl ammonium bromide, p. 156.  J!N_ S.  P.
          Colowick and N. 0. Kaplan  (eds.),  Methods in
          enzymology, vol.  12,  Academic Press,  Inc.,
          New York.

36.  Chen, P.  S.,  T.  Y.  Toribara, and H. Warner.   1956.
          Microdetermination of phosphorus.   Anal. Chem.
          2£: 1756-1758.

37.  Taussky,  H. A.  and E.  Shorr.  1953.  A microcolorimetric
          method for the determination of inorganic phosphorus.
          J. Biol.  Chem.  202:675-685.

38.  Vogel, A.  I.   1951.   Quantitative inorganic  analysis.
          p. 643-644.  Longmans, Green and Co., London.

39.  Dische, Z.  1953.  Quantitative and qualitative  deter-
          mination of heptoses.  J.  Biol. Chem. 204:983-997.

40.  Osborn, M. J.   1963.  Studies on the gram-negative  cell
          wall. Evidence for the role of 2-keto-3-deoxyoctuloso-
          nate in  the lipopolysaccharide of Salmonella
          typhimurium.  Proc. Nat. Acad. Sci.  U.S.A.  50:
          499-506.

41.  Eblein, A. D.  and E. C. Heath.   1965.  The biosynthesis  of
          cell wall lipopolysaccharide in Escherichia coli.
          I.  The  biochemical properties of a uridine diphos-
          phate galactose-4-epimerase less mutant. J. Biol.
          Chem. 240:1919-1925.
                               136
-------
42.   Aminoff,  D.,  W.  T.  J.  Morgon,  and W.  M.  Watkins.   1952.
          Studies  in  immunochemistry.   II.   The  action of
          dilute  alkali  on  the N-acetylhexosamines  and the
          specific blood group mucoids.   Biochem. J.   51:
          329-389.

43.   Rondle, C. J. M. and W.  T.  J.  Morgan.   1955.   The deter-
          mination of glucosamine and galactosamine.   Biochem.
          J. 61:586-589.

44.   Davis, C. E., S. D. Freedman,  H.  Douglas and A.  I. Brande.
          1969.   Analysis of sugars in bacterial endotoxins
          by gas-liquid  chromatography.   Anal.  Biochem. 28:
          243-256.

45.   Mahoney, R.  P. and  M.  R. Edwards.  1966.  Fine structure
          of Thiobacillus thiooxidans.  J.  Bacteriol.  92:
          487-495.

46.   Shively, J.  M.,  G.  L.  Decker and J.  W. Greenawalt.  1970.
          Comparative ultrastructure of the thiobacilli.
          J. Bacteriol.  101:618-627.

47.   Shands, J.  M., J. A. Graham and K.  North.   1967.   The
          morphologic structure of isolated bacterial  lipo-
          polysaccharide.  J. Mol.  Biol.  25:15-21.

48.   Burton, A.  and H. E. Carter.  1964.   Purification and
          characterization  of the lipid A component of the
          lipopolysaccharide from E_. coli.  Biochemistry
          .3:411-418.

49.   Osborn, M.  J.   1969.  Structure and biosynthesis  of the
          bacterial  cell wall.  Annu. Rev. Biochem. 38:501-538.

50.   Akagi, J. M.  and L. L. Campbell.  1963.  Inorganic pyro-
          phasphatase of Desulfovibrio desulfuricans .   J.
          Bacteriol.  86:563-568.

51.   Clark, J. M., Jr.  1964.  Inorganic phosphate determina-
          tion:   Modified Fiske-Subbarow method.  JW Experi-
          mental biochemistry.  J. M. Clark  (ed.)  W.H. Freeman
          and Co., San  Francisco,  p. 32.

52.   Garen, A. and C. Levinthal.  1960.  A fine-structure,
          genetic  and chemical study of the enzyme alkaline
          phosphatase of E_.  coli.   I.  Purification and
          characterization of alkaline phosphatase.  Biochim.
          Biophys. Acta. 38:470-483.
                                137
-------
53.  Neu, H.  C.  and H.  Chou.   1967.   Release of surface enzymes
          in Enterobacteriaceae by osmotic shock.   J.  Bacteriol.
          94_: 1934-1945.

54.  Blumenthal, B.  I., M.  D.  Johnson and E. J.  Johnson.   1967.
          Distribution of heat-labile and heat-stable  inorganic
          pyrophosphatases  among some bacteria.   Can.  J.
          Microbiol. 1_3:1695-1999.

55.  Lazaroff, N.   1963. Sulfate requirement for iron
          oxidation by Thiobacillus ferrooxidans.   J.  Bacteriol.
          85_:78r83.

56.  Unz, R.  and D.  G. Lundgren.  1961.  A comparative nutri-
          tional study of three chemoautotrophic bacteria:
          Ferrobacillus ferrooxidans, Thiobacillus ferrooxidans,
          and Thiobacillus  thiooxidans.  Soil Sci. 92:502-515.

57.  Shafia, F. and R. F. Wilkinson, Jr.  1969.   Growth of
          Ferrobacillus ferrooxidans on organic matter.  J.
          Bacteriol. 97:256-260.

58.  Lundgren, D.  G., K. J. Anderson, C. C. Remsen, and R. P.
          Mahoney.  1964.  Culture, structure, and physiology
          of the chemoautotroph Ferrobacillus ferrooxidans.
          Dev. Ind.  Microbiol. 6_: 250-259.

59.  Silver, M., P.  Margalith, and D. G. Lundgren.  1967.
          Effect of glucose on carbon dioxide assimilation
          and substrate oxidation by Ferrobacillus ferrooxidans.
          J.  Bacteriol. 9^:1765-1769.

60.  Silver, M.  1970.  Oxidation of elemental sulfur and
          sulfur compounds  and C02 fixation of Ferrobacillus
          ferrooxidans  (Thiobacillus ferrooxidans).  Can. J.
          Microbiol. 16_: 845-849.

61.  Lessie, T. and F. C. Neidhardt.  1967.  Adenosine triphos-
          phate linked control of Pseudomonas aeruginosa
          glucose-6-phosphate dehydrogenase.  J. Bacteriol.
          93_: 1337-1345.

62.  Wang, C. H., I. Stern, C. M. Gilmour, S. Klingsoyr, D.
          J. Reed, J. J. Baily, B. E. Christensen and V. H.
          Cheldelin.  1958.  Comparative study of glucose
          catabolism by the radiorespirometric method.  J.
          Bacteriol. 76:207-216.
                                133
-------
63.  Perry, J.  J.  and J. B. Evans.  1967.   Glucose catabolism
          in Micrococcus sodonensis.  J.  Bacteriol. 95:1859-1846,

64.  Layne, E.   1957.  Spectrophotometric and turbidometric
          methods  for measuring proteins,  p. 447-454.  JW
          S. P. Colowick and N. 0. Kaplan,  (eds).   Methods in
          enzymology, vol. 3.  Academic Press, Inc., New York.

65.  Horecker,  B.  L.  and P. Z. Smyrmotis.  1955.  6-phospho-
          gluconate dehydrogenase, p. 323-327.  IN_ S. P.
          Colowick and N. 0.  Kaplan, (eds).  Methods in
          enzymology, vol. 1.  Academic Press, Inc., New
          York.

66.  Keele, B.  K., P. B. Hamilton and G.  H. Elkan.  1969.
          Glucose catabolism in Rhizobium japonicum.  J.
          Bacteriol.  £7:1184-1191.

67.  Anfinson, C.  B.   1955.  Aconitase from pig heart, p.
          695-698.  IN^ S. P.  Colowick and N. 0. Kaplan  (eds).
          Methods in enzymology, vol. 1.   Academic Press,
          Inc., New York.'

68.  Arrigoni, 0.  and T. P. Singer.  1962.  Limitations of
          the phenazine methosulfate assay for succinic and
          related dehydrogenases.  Nature.  195:1256-1258.

69.  Watson, B. F. and M. Dworkin.  1968.   Comparative inter-
          mediary metabolism of negative cells and microcysts
          of Myxocoecus xanthus.  J. Bacteriol'. 96:1465-1475.

70.  Racker, E.  1950.  Spectrophotometric measurement of the
          enzymatic formation of fumaric acid and cis-aconitic
          acids.  Biochim. Biophys. Acta. 4_:211-214.

71.  Schnaitman, C. A., M. S. Korczynski and D. G. Lundgren.
          1969.  Kinetic studies of iron oxidation by whole
          cells of Ferrobacillus ferrooxidans.  J. Bacteriol.
          9£:552-557.

72.  Martin, A. and S. C. Rittenberg.  1970.  Regulation of
          glucose metabolism in Thiobacillus intermedius.
          J. Bacteriol. 104:239-246.
                                139
-------
73.  Smith, R. A. and I. C. Gunsalus.  1957.  Isocitritase:
          Enzyme properties and reaction equilibrium.   J.
          Biol. Chem. 229:305-319.

74.  Eisenberg, R. C. and W. J. Dobrogosz.  1967.   Gluconate
          metabolism in Escherichia coli.  J.  Bacteriol.
          93^:941-949.

75.  Kitos, P. A., C. H. Wang, B.  A. Mohler, T.  E.  King and
          V. H. Cheldelin.  1958.   Glucose and gluconate
          dissimilation in Acetobacter suboxydans.   J.  Biol.
          Chem. 235:1295-1298.

76.  Kornberg, A. and B. L. Horecker.  1955.  Glucose-6-
          phosphate dehydrogenase,  p. 323-327.  JW  S.  P.
          Colowick and N. 0. Kaplan (eds).  Methods in
          enzymology, vol. 1.  Academic Press, Inc.,
          New York.

77.  Sanwall, B. D.  1970.  Regulatory mechanisms involving
          nicotinamide adenine nucleotides as  allosteric
          effectors.   III.  Control of glucose-6-phosphate
          dehydrogenase.  J. Biol.  Chem. 245:1626-1631.

78.  Blackkolb, R. and H. G. Schlegel.  1968.   Regulation
          der glucose-6-phosphat-dehydrogenase aus  Hydrogenomonas
          H-16 durch  ATP und NADH     Arch. Mikrobiol.  63^:
          177-196.

79.  Olive, C., M. Geroch, and H.  R. Levy.  1971.   Glucose-6-
          phosphate dehydrogenase  from Leuconostoc  mesen-
          teroides.  Kinetic studies.  J. Biol.  Chem.  246:
          2047-2057.

80.  Clark, C. and E. L. Schmidt.   1967.  Growth response of
          Nitrosomonas europaea to  amino acids.  J. Bacteriol.
          93j 1302-1308.

81.  Kundig, W., S. Ghosh and S. Roseman.  1964. Phosphate
          bound to histidine as an  intermediate  in  a novel
          phosphotransferase system.  Proc. Natl. Acad. Sci.
          U.S.A. 52^:1067-1074.

82.  London, J. and S. L. Rittenberg.   1966.  Effects  of
          organic matter on the growth of Thiobacillus  inter-
          medius.  J. Bacteriol. 91:1062-1069.
                             140
-------
83.  Wang, W.  S. and D.  G. Lundgren.   1969.   Poly-g-hydroxy-
          butyrate in the chemolithotrophic bacterium Ferro-
          bacillus ferrooxidans.   J.  Bacteriol.  97:947-950.

84.  Rao, G.  S.  and L. R. Berger.  1970.   Basis  of pyruvate
          inhibition in Thiobacillus  thiooxidans.   J. Bacteriol.
          102:462-466.

85.  Din, G.  A., I.  Suzuki and H. Lees.  1967.   Ferrous iron
          oxidation by Ferrobacillus  ferrooxidans.   Purifica-
          tion and properties of Fe^-cyto chrome £ reductase.
          Can. J. Biochem. 45:1523-1546.

86.  Yates, M. G. and A. Nason.  1966.   Electron transport
          systems of the chemoautotroph Ferrobacillus ferrooxi-
          dans.   III.  Purification and properties  of a heat-
          labile iron-cytochrome £ reductase. J.  Biol. Chem.
          241:4872-4880.

87.  Lejohn, H.  B., L. Van Caeseele and H. Lees. 1967.  Cata-
          bolite repression in the facultative  chemoautotroph
          Thiobacillus novellus.   J.  Bacteriol.  94:1484-1491.

88.  Magasanik,  B.  1963.  The genetic and molecular basis for
          catabolite repression, p. 271-286.  IN_H. H. Vogel,
          V. Bryson and J. 0. Lampen (eds), Informational
          macromolecules.  Academic Press, Inc., New York.

89.  Bassham, J. A. and M. Calvin.  1957.  The path of carbon
          in photosynthesis, Prentice Hall, Inc.,  New Jersey.

90.  McFadden, B. A. and C. L. Tu.  1967.  Regulation of
          autotrophic and heterotrophic carbon dioxide fixation
          in Hydrogenomonas facilis.   J. Bacteriol. 95:886-893.

91.  Kuehn, G. D. and B. A. McFadden.  1968.  Enzymes of the
          Entner-Doudoroff pathway in fructose-grown Hydro -
          genomonas eutropha.  Can. J. Microbiol.  14:1259-1260.

92.  Gottshalk, G., V. Eberhardt and H. G. Schlegel.  1965.
          Verwertung von fructose durch Hydrogenomonas H-16.
          Arch.  Mikrobiol. £8:95-108.

93.  Smith, A. J., J. London and R. Y. Stanier.   1967.  Biochemi-
          cal basis of obligate autotrophy in blue-green algae
          and thiobacilli.  J. Bacteriol. 94:972-983.
                                141
-------
94.  Smith, A. J. and D. S. Hoare.   1968.   Acetate assimilation
          ky Nitrobacter agilis in relation to its "obligate
          autotrophy".  J.  Bacteriol.  95:844-855.

95.  Trudinger, P. A. and D. P. Kelly.  1968.  Reduced nicotin-
          amide adenine dinucleotide oxidation by Thipbacillus
          neapolitanus and Thiobacillus strain C.  J. Bacteriol,
          95_: 1962-1963.

96.  Peck, H. D., Jr.  1968.  Energy-coupling mechanisms in
          chemolithotrophic bacteria.   Ann. Rev. Microbiol.
          _22_:489-518.

97.  Anderson, K. J. and D. G. Lundgren.  1969.  Enzymatic
          studies of the iron-oxidizing bacterium Ferrobaci11us
          ferrooxidans;  Evidence for a glycolytic pathway and
          Krebs cycle.  Can. J. Microbiol. 15:73-79.

98.  Cohen, L. H. and W. K. Noell.   1960.   Glucose catabolism
          of rabbit retina before and after development of
          visual function.   J. Neurochem.  5:253-276.
                               142
-------
                           SECTION X

         LIST OF PUBLICATIONS RESULTING FROM THE GRANT
Tabita, F. R., M. Silver and D. G. Lundgren.  1969.  The
     rhodanese enzyme of Ferrobacillus ferrooxidans CThio-
     bacillus ferrooxidans) Can. J. Biochem. 47:1141-1145.

Lundgren, D. G. and F. R. Tabita.  1969.  Biochemical
     biology of metal sulfide oxidizing bacteria.  Symposium
     on pollution control in fuel combustion, processing,
     and mining.  Amer. Chem. Soc., Division of Fuel Chemistry,
     157th National Meeting, Minneapolis, Minn.  Vol. 13(2) :
     60-67.

Schnaitman, C. A., M. S. Korczynski, and D. G. Lundgren.
     1969.  Kinetic studies of iron oxidation by whole cells
     of Ferrobacillus ferrooxidans.  J. Bacteriol. 99:552-557

Wang, W. S., M. S. Korczynski, and D. G. Lundgren.  1970.
     Cell envelope of an iron-oxidizing bacterium:  Studies
     of lipopolysaccharide and peptidoglycan.  *J. Bacteriol.
     104:556-565.

Howard, A. and D. G. Lundgren.  1970.  Inorganic pyrophos-
     phatase from Ferrobacillus ferrooxidans  (Thiobacillus
     ferrooxidans)"Can. J. Biochem. 48: 1302-1307.

Tabita, R., M. Kaplan and D. G. Lundgren.  1970.  Microbial
     ecology of mine drainage.  Third Symposium on Coal
     Mine Drainage Research.  Mellon Institute, Pittsburgh,
     Pa. 94-113.

Tabita, R. and D. G. Lundgren.  1971.  The heterotrophic
     metabolism of the  chemolithotroph Thiobacillus ferrooxi-
     dans .  Submitted to J.  Bacteriol.

Tabita, R. and D. G. Lundgren.  1971.  Utilization of glucose
     and the effect of  organic  compounds on the chemolitho-
     troph Thiobaci1lus ferrooxidans.  Submitted  to J. Bacteriol,

Tabita, R. and D. G. Lundgren.  1971.  Glucose-6-phosphate
     dehydrogenase from the  chemolithotroph Thiobacillus
     ferrooxidans.  Submitted to J. Bacteriol.

Vestal, J. R. and D. G. Lundgren.  1971.  The  sulfite oxidase
     of Thiobacillus ferrooxidans  (Ferrobacillus  ferrooxidans].
     Can. J. Biochem.   In the press.
                                143
-------
Lundgren, D.  G.,  J.  R.  Vestal and F.  R.  Tabita.   1971.
     The microbiology of mine drainage pollution.   In
     R.  Mitchell  (Ed.)   Water Pollution  Microbiology,
     John Wiley and Sons, Inc.  New York.   In the press
                           144
-------
                           SECTION XI
SO
   2-
    2-
    2-
S2°3
S4°6
DEAE- cellulose
Tris
Tricine
EDTA
AOD or AA
TCA cycle
NAD (H)
NADP (H)

RuDP
GSH
ATP
ADP
AMP
cAMP
APS
V
 max
Ki
9K medium

9KG medium

mole % G + C
    GLOSSARY

sulfite
thiosulfate
tetrathionate
diethylaminoethyl cellulose
tris-(hydroxymethyl)-aminomethane
N-tris- (hydroxymethyl)-methyl glycine
ethylenediaminetetraacetate
change in optical density or absorbance
tricarboxylic acid cycle
nicotinamide adenine dinucleotide (reduced)
nicotinamide adenine dinucleotide
     phosphate (reduced)
ribulose diphosphate
reduced glutathione
adenosine triphosphate
adenosine diphosphate
adenosine monophosphate
3',5'-cyclic adenosine monophosphate
adenosine phosphosulfate
relative mobility
Michaelis constant
maximum velocity of an enzyme
enzyme inhibitor constant
ferrous sulfate medium containing
     9000 ppm Fe++
ferrous sulfate medium with 0.5% glucose
     added
the amount of guanine and cytosine in DNA
                               145
-------
S  on                   sedimentation in water at  20°C
 w,^u



EC number               International enzyme code  number




cyt c                   cytochrome £




cyt a                   cytochrome a
                              14C
-------
                         SECTION XII
                           APPENDIX
Persons trained under this grant:

Dr. Augustine Wang, Ph.D.        June 1968
Dr. F. Robert Tabita, Ph.D.     Jan 1971
Adele Howard Cooney, M. S.
Carl A. Bodo, Jr., Ph.D.
Robert C. Tuttle, B. S.
Postdoctoral Fellow:

Dr. J. Robie Vestal
  Jan 1970
  expected
June 1972
  June 1971
Present Address

Dept. of Microbiology
Mississippi State Univ.
State College,
Mississippi

Dept. of Chemistry
Washington State Univ.
Pullman, Washington

Dept. of Microbiology
Upstate Medical Center
State Univ. of New York
Syracuse, New York

Dept. of Biology
Syracuse University
Syracuse, New York

Dept. of Biological
Sciences
Harvard University
Cambridge, Mass.
                 Dept. of Biological
                 Sciences
                 University of
                 Cincinnati
                 Cincinnati, Ohio
                               147
-------
Published abstracts of papers delivered before professional
meetings.

Tabita, F. R., M. S. Silver and D.  G.  Lundgren.  1969.
     Partial purification and properties of rhodanese from
     Ferrobacillus ferrooxidans.   Bacteriol. Proc.  69:64.

Tabita, F. R. and D. G. Lundgren.   1970.  Glucose metabolism
     in Ferrobacillus f errooxidans  (Thiobacillus ferrooxidans)
     Bacteriol. Proc. 70: 125.

Bodo, C. A., Jr. and D. G. Lundgren.   1971.  Uptake and
     oxidation of iron by Thiobacillus ferrooxidans.
     Bacteriol. Proc. 71; p. 44.

Vestal, J. R., D. G. Lundgren and K.  C. Milner.  1971.
     Toxic and immunological differences among lipopoly-
     saccharides from Thiobacillus  f errooxidans grown auto-
     tropically and heterotrophically.  Bacteriol.  Proc,
     71: M20.

Titles of papers delivered to local professional meetings:

D. G. Lundgren.  1969.  Structural and functional relation-
     ships of Ferrobacillus.  Connecticut Valley Branch,
     American Society for Microbiology.  (November)

F. R. Tabita and D. G. Lundgren.   1970.  Growth of iron
     oxidizing bacteria on different organic substrates.
     Central New York Branch, American Society for Micro-
     biology (May).

J. R. Vestal and D. G. Lundgren.   1970.  Metabolism of sulfur
     by Thiobacillus ferrooxidans:   Sulfite oxidase.  Central
     New York Branch, American Society for Microbiology
     (October).

C. A. Bodo and D. G. Lundgren.  1971.   Iron binding and
     oxidation by Thiobacillus ferrooxidans.  Central New
     York Branch, American Society for Microbiology (March).

R. C. Tuttle and D. G. Lundgren.   1971.  Isolation and
     characterization of Thiobacillus  from alkaline mine
     drainage.  Central New York Branch, American Society
     for Microbiology  (March).
                             148
-------
Titles of papers delivered at symposia at professional meetings
Lundgren, D. G., M. Korczynski and G.  Wang.   1969.   Structure
     and function relationships of iron-oxidizing bacteria.
     Symposium:  Structure and biochemistry of autotrophic
     microorganisms.  69th Annual meeting, American Society
     for Microbiology.

Tabita, R. and D. Lundgren.  1971.  Autotrophic and hetero-
     trophic metabolism of iron-oxidizing bacteria.  Sympo-
     sium:  Autotrophy.  71st annual meeting.   American
     Society for Microbiology.
-------
1
Accession Number
w
5
fy 1 Subject Field & Group

SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
SvTQr^ncA ITn-i iro-i^e-i -t-\r n^-^n^.4. — *. r° r\ - t «
      Title
           Inorganic Sulfur Oxidation "by Iron-Oxidizing Bacteria
  10
     Authors)
   Donald G. Lundgren
        16
                                     21
Project Designation

   Grant #14010 DAY
                                         Note
  22
     Citation
      Descriptors (Starred First)
  25
     Identifiers (Starred First)
  07 Abstract
 _^_J         The utilization of sulfur and reduced sulfur  compounds by  the  iron oxidizing
 cnemolithotroph Thiobacillus ferrooxidans was studied at the biochemical  level.  The
 identification, characterization and partial purification  of the rhodanese  and sulfite
 oxidase  enzymes completed the scheme of sulfur metabolism  in T. ferrooxidans which  leads
 to  energy generation.                                         ~ ~~	
     The cell envelope lipopolysaccharide (LPS) purified from iron-grown  cells was  studied
 in  the electron microscope.   The partial chemical composition of the LPS  revealed unusually
 high quantities of Fe-5 .   A new colorimetric whole cell assay to study  iron-oxidation
 kinetics was  developed which will be of benefit to future  studies at the  molecular  level
 The inorganic pyrophosphatase enzyme, an essential enzyme  in maintaining  the energy
 balance  in the cell, was  partially purified and its properties studied.   This is the first
 account  of the presence of this enzyme in chemolithotrophic microorganisms.
     The effects  of organic carbon and energy sources on chemolithotrophic microorganisms
were studied.   T.  ferrooxidans  can convert from chemolithotrophic to heterotrophic
metabolism after a long lag in  the presence of the organic substrate, and after some
energy is  stored from  iron oxidation.  Growth on glucose proceeds much  like other hetero-
trophic  gram  negative  organisms.   The metabolism of glucose is via the Entner-Doudoroff
pathway.
^Abstractor
Donald G. Lundgren
  Institution
 Syracuse  University,
           Syracuse. New York 13210
  WR:102  (REV. JULY 1969)
  WRSIC
SEND. WITH COPY OF DOCUMENT. TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
                         U.S. DEPARTMENT OF THE INTERIOR
                         WASHINGTON. D. C. 20240

                                                  * GPO: 1970 — 389-930
-------