Ecological Research Series
                                METABOLISM  OF
MERCURY  COMPOUNDS IN MICROORGANISMS
                            Environmental Research Laborato
                              Office of Research and Development
                              U.S. Environmental Protection Agency
                             Narragansett, Rhode Island 02882

-------
                     RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped into
five series.   These five broad categories were established to
facilitate further development and application of environmental
technology.  Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in
related fields.  The five series are:

          1.   Environmental Health Effects Research
          2.   Environmental Protection Technology
          3.   Ecological Research
          k.   Environmental Monitoring
          5.   Socioeconomic Environmental Studies

This report has been assigned to the Ecological Research series.
This series describes research on the effects of pollution on
humans, plant and animal species, and materials.  Problems are
assessed for their long- and short-term influences.  Investi-
gations include formation, transport, and pathway studies to
determine the fate of pollutants and their effects.  This work
provides the technical basis for setting standards to minimize
undesirable changes in living organisms in the aquatic,
terrestrial and atmospheric environments.
This document is available to the public through the National
Technical Information Service, Springfield, Virginia  22l6l.

-------
                                    EPA-600/3-75-007
                                    October 1975
    METABOLISM OF MERCURY COMPOUNDS

           IN MICROORGANISMS
                  by

Rita R, Colvell and John D. Nelson, Jr.
        University of Maryland
     College Park, Maryland  207^2
           Grant No.  802529
            Project Officer

              C.  S. Hegre
   Environmental  Research Laboratory
   Narragansett,  Rhode Island  02882
 U.S.  ENVIRONMENTAL PROTECTION AGENCY
  OFFICE OF RESEARCH AND DEVELOPMENT
   ENVIRONMENTAL RESEARCH LABORATORY
   NARRAGANSETT, RHODE ISLAND  02882

-------
                            DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory,
U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor
does mention of trade names or commercial products constitute
endorsement or recommendation for use.
                                  11

-------
                                ABSTRACT
This report describes the physiology and ecology of mercury-resistant,
mercury-metabolizing bacteria from Chesapeake Bay.  Evidence is present-
ed which establishes a role for bacteria in the cycling of mercury in
the estuarine environment.

Mercury-resistant aerobic, heterotrophic bacteria were isolated from
water and sediment at six stations, representing various levels of en-
vironmental quality, in upper Chesapeake Bay.  These organisms were
found to be resistant to an array of inorganic and organic mercury com-
pounds and heavy metal ions.  It was also observed that in individual
cultures demonstrating resistance to EgCl2 and phenylmercurie acetate,
the resistance was adaptive and was also related to metabolic capability
for degrading the compounds to elemental mercury (Hg°).  However, cul-
tures in the process of adaptation evidenced delayed growth and cell di-
vision, which were dependent upon mercury concentration.  The cultures
also developed a variety of morphological irregularities associated with
cell wall and cytoplasmic membrane synthesis and function.

From the results of a survey of Hg° production among a group of randomly
selected, BgCl2-resistant bacteria and mixed natural microbial popula-
tions, it was established that the enumeration of mercury-resistant bac-
teria by plate counting is a valid index of potential Hg** metabolism J£
situ.

The population of mercury-resistant bacteria, primarily Pseudomonas spp.,
varied quantitatively in time and from station to station.  The distri-
bution of mercury-resistant bacteria was significantly different in water
and sediment, from station to station, and seasonally.  let, the propor-
tion of Hg2^ resistant bacteria among the total, viable, heterotrophio
bacterial population reached a reproducible maximum in spring and was
positively correlated with water turbidity, dissolved oxygen concentra-
tion, and mercury concentration in the sediment.

These findings and the observation of the evolution of Hg° from freshly
collected water and sediment suggest that bacteria may contribute sub-
stantially to the mobilization and transformation of mercury from exist-
ing deposits in Chesapeake Bay, specifically, and in the aquatic environ-
ment, in general.

This report was submitted in fulfillment of Grant #802529 by the Univer-
sity of Maryland under the partial sponsorship of the Environmental Pro-
tection Agency.  Work was completed as of November 1, 1974.
                                    iii

-------
                                CONTENTS


Sections

    I        CONCLUSIONS                                              1

   II        RECOMMENDATIONS                                          2

  III        INTRODUCTION                                             3

   IV        PHYSIOLOGICAL AND CULTURAL CHARACTERISTICS OF
             MERCURY-RESISTANT BACTERIA                               4

    V        MICROBIAL ECOLOGY OF MERCURY-RESISTANT BACTERIA
             IN CHESAPEAKE BAY                                       24

   VI        EFFECTS OF MERCURIC CHLORIDE UPON GROWTH AND MORPHOL-
             OGY OF SELECTED STRAINS OF MERCURY-RESISTANT BACTERIA   48

  VII        THE ROLE OF BACTERIA IN THE MOVEMENT OF MERCURY
             THROUGH A SIMPLIFIED FOOD CHAIN                         64

 VIII        REFERENCES                                              74

   IX        LIST OF PUBLICATIONS                                    81

    X        GLOSSARY OF TERMS                                       83

-------
                                FIGURES
No..
1
2
3
4
5
6
7
8
9
10
11

Average distribution of genera in total and HgCl2 resistant
populations
Comparative percentages of mercury-resistant isolates from
sediment and water
Induction of resistance to PMA by Paeudomonas 2/t/t
Uptake and metabolism of FMA by Pseudomonas 244
Change in mercury contained in bioreactor atmosphere as a
function of time
Uptake and metabolism of FMA by induced and non-induced
resting cells of Pseudomonas 244
TSrpe-rl mental sampling p1tr*»p -fi^ npp*»-p ntieaapAalre Ray
Elemental mercury evolution by bacteria grovn on agar
plates containing HgCXg
Seasonal variation in percent of HgCl2 resistant bacteria
in -water
Seasonal variation in percent of HgCl2 resistant bacteria
in sediment
Seasonal variation in salinity, temperature, and dissolved
Page
9
12
15
16
18
19
28
30
36
36

      oxygen concentration in surface waters from three locations
      in Chesapeake Bay                                              37

12    Distribution of HgCl2 resistant, aerobic, heterotrophic
      bacteria isolated from water and sediment of Chesapeake
                                                                     41
13    Seasonal population distribution of HgCl2 resistant,
      aerobic, heterotrophic bacteria                                42

14    Mercury resistant bacterial population distributions in
      water and sediment in Chesapeake Bay                           42
                                   vi

-------
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Growth of Bacillus, species 394, in the presence and
absence of 5 ppm of HgCl2
Thin section of cells of culture #394 grown in the pres-
ence of 5 ppm of HgCLj
Thin section of cells of culture #394 grown in the pres-
ence of 15 ppm of HgCLg
Growth and uptake and metabolism of mercuric ion by
Enterobacter . strain 85
Colonial variation in Enterobacter. strain 85, grown in
the presence of 2 ppm of HgCL^
Growth and uptake and metabolism of mercuric ion by
Enterobaoter. strain 85
Thin section of culture #85 cells immediately after
addition to broth containing 2 ppm of HgClg
Thin section of culture #85 cells after 2 h incubation
in broth containing 2 ppm of HgCl2
Thin section of culture #85 cells after 6 h incubation
in broth containing 2 ppm of HgCLj
Thin section of culture #85 cells after 24 h incubation
in broth containing 2 ppm of HgCLj
Thin section of culture #85 cells after 26 h incubation
in broth containing 2 ppm of HgClj
Thin section of culture #85 cells after 28 h incubation
in broth containing 2 ppm of HgGl^
Sterilizable oyster aquarium
Oyster aquarium in operation
Oyster dissection
52
53
53
54
55
56
58
58
59
60
60
61
66
67
68
VI1

-------
TABLES
Ha-
1

2
3
4

5
6
7
8
9

10

11

12
13

U
15

COMPARISON OF RESISTANCE OP SEDIMENT AND WATER POPULATIONS
TO DIFFERENT LEVELS OF HgCl2 AND PMA
EFFECT OF INCUBATION TEMPERATURE ON BACTERIAL VIABLE COUNT
AND RESISTANCE TO HgClo
COMPARATIVE EFFECTS OF AEROBIC AND ANAEROBIC INCUBATION ON
BACTERIAL RESISTANCE TO HgClg
EFFECT OF SALT CONCENTRATION ON BACTERIAL RESISTANCE TO
HgCl2
DISTRIBUTION OF BACTERIAL BgCl2 RESISTANCE AND TOTAL MER-
CURY CONCENTRATIONS AMONG WATER AND SEDIMENT FRACTIONS
CROSS RESISTANCE OF BACTERIAL CULTURES TO MERCURY COMPOUNDS
HEAVY METAL RESISTANCE OF MERCURY-RESISTANT BACTERIA
ADAPTATION TO MERCURY RESISTANCE
METABOLISM OF PHENYIMERCURIC ACETATE BY MERCURY-RESISTANT
BACTERIA
EFFECTS OF SALTS AND INHIBITORS ON PMA METABOLISM BY
PSEUDOMONAS. STRAIN 244
EFFECT OF MAGNESIUM ION CONCENTRATION ON PMA RESISTANCE
IN PSEUDOMONAS. STRAIN 244
DISTRIBUTION OF RADIOACTIVITY IN ^^g-PMA-LABELED CELLS
OF PSEUDOMONAS. STRAIN 244
TOTAL MERCURY CONCENTRATIONS AND PERCENTAGE OF THE TOTAL
VIABLE BACTERIAL COUNT RESISTANT TO 6 PPM OF HgCl2
HgClg METABOLIZING BACTERIA
MERCURY RESISTANCE AND METABOLISM OF BACTERIAL STRAIN
Page

6

7
7

S
10
11
13
14

17

20
/\«*
21
22

29
31
*±f*
 viii

-------
Ha-

16     CHARACTERIZATION OF THE VOLATILE Hg METABOLITE PRODUCED
       BY PSEUDQMONAS. 8TRAIH 5                                     33

17     Hg° EVOLUTION BT NATURAL POPULATIONS OF BACTERIA ISOLATED
       FROM WATER AND SEDIMENT                                      35

18     MULTIPLE REGRESSION ANAUSIS OF SEASONAL DATA                38
19     PHYSIOLOGICAL CHARACTERISTICS OF HgCl2 A® ™* RESISTANT
       BACTERIA                                                     40

20     SURVEY OF MORPHOLOGICAL EFFECTS IN SELECTED uuLruwsS OF
       BACTERIA GROWN IN THE PRESENCE OF HgCl2                      51

21     MERCURY RESISTANCE AND MERCURY METABOLISM OF STRAIN 85
       BEFORE AND AFTER GROWTH IN THE PRESENCE OF HgC^             62

22     EXPERIMENTAL OUTLINE                                         69

23     AVERAGE TOTAL MERCURY CONCENTRATION IN UNTREATED OYSTERS     70

24     RELATIVE DISTRIBUTION OF MERCURY FOLIDWING EXPOSURE OF
       OYSTERS                                                      71

25     TISSUE DISTRIBUTION OF MERCURY AMONG TREATED OYSTERS         72
                                  IX

-------
                             ACKNOWLEDGMENTS
Shiptime aboard the R/V RIDGELY WARFIELD was made possible by Grant
#GD-31707 from the National Science Foundation to the Chesapeake Bay
Institute, Johns Hopkins University.  We are indebted to Dr. E. Taylor
and H. Whaley of the Chesapeake Bay Institute and the crew of the R/V
RIDGELY WARFIELD for their excellent cooperation and assistance during
the field studies.  Shiptime was also provided by the State of Maryland
Department of Natural Resources.

We are also grateful for heavy metal analyses performed by 0. Villa and
J. Marks, Annapolis Field Laboratory, Environmental Protection Agency;
J. F. Kopp, Environmental Monitoring and Support Laboratory, Cincinnati,
Ohio; and W. Blair, National Bureau of Standards, Washington, D.C.
Analytical support was provided via the National Bureau of Standards
(Environmental Protection Agency Grant #R 801002).

We also acknowledge electron microscopists, Z. Vaituzis and L. W. Wan,
for their considerable efforts.

-------
                               SECTION I

                              CONCLUSIONS
The results of this investigation prove that aerobic, heterotrophic bac-
teria can be influential in the mobilization of mercury from existing
resources in Chesapeake Bay.

Mercury-resistant bacteria are numerous throughout Chesapeake Bay in
vater and sediment, but are most prevalent in those areas where high
total mercury levels are found in sediment.  Bacteria "adapt" to mer-
cury, in terms of total numbers and generic distribution in a given lo-
cation and in the ability to resist and reduce mercury, presented in the
form of mercuric and phenylmercuric ions, to the elemental state.  The
former adaptation probably represents a long-term, selective effect, and
the latter a more rapid, or transient, microbial response to mercury
contamination.  Although bacteria adapt successfully, in vitro studies
suggest that bacteria can undergo cell damage in the presence of exces-
sive levels of mercury contamination.  Bacterial growth occurs at a
threshold inhibitory concentration of mercury, with the appearance of
numerous morphological effects, which indicate cell wall and cytoplasmic
membrane damage.

The proportion of mercury-resistant bacteria in the total viable, aero-
bic, heterotrophic bacterial population serves as a valid index of po-
tential mercury mobilization in Chesapeake Bay water and sediment.
Using this criterion, it can be concluded that bacterial mobilization of
mercury, through the formation of Hg°, varies seasonally and is related
to water turbidity, water dissolved oxygen, and total mercury concentra-
tions in sediment.  A reproducible peak in activity occurs in Spring.
This phenomenon can have serious consequences, in terms of effects on
life cycles of higher marine organisms, and the rates of accumulation of
mercury through marine and estuarine food webs.

-------
                              SECTION II

                            RECOMMENDATIONS
We have identified and characterized a group of bacteria which are po-
tentially significant in the mobilization and transformation of mercury
in Chesapeake Bay.  Our results and the reports of others indicate that
Hg is continually mobilized from existing deposits of mercury in the in-
shore marine and estuarine environments.  As mercury levels did not de-
crease significantly during.the 17 months in vhich ve monitored the
several stations in Chesapeake Bay, vhich vere included in our study, Me
assume that processes of addition and mobilization maintain a steady
state level of mercury.  A mercury budget should be calculated to deter-
mine whether production of Hg° correlates with theoretical seasonal
fluctuations predicted from our results.  Clearly a "microbial mobiliza-
tion index" (MMI) will then prove to be a useful parameter for assessing
the environmental impact of presently existing or new reservoirs of mer-
cury.

The fate of Hg°, following release into the aqueous medium, should also
be investigated to determine:  1) the proportion of Hg lost from the en-
vironment through volatilization; and 2) the chemical species and ulti-
mate destination of mercury in solution under in situ conditions.  This
information must eventually be obtained in order to establish the
sources and routes of mercury in marine and estuarine food webs.

Finally, our findings underline  the importance of an understanding of
the microbial population structure in the natural environment.  An un-
equivocal conclusion from our studies is that the bacteria play a sig-
nificant role in the cycling and introduction of mercury and other
pollutants to the marine and estuarine food webs.  More attention must
be placed on understanding  and managing the bacteria, fungi, and other
microorganisms  in the environment.

-------
                              SECTION III

                              INTRODUCTION
Mercury and its compounds (organic and inorganic) are known to undergo a
number of chemical and biological transformations.  Microorganisms, in
particular bacteria, have been documented as agents in the formation of
methyl mercury and elemental mercury from inorganic mercury.  Similarly,
bacteria have been shovn to be capable of degrading PMA, MeHg and other
organomercurials to Hg° and simple carbon compounds.

Inorganic mercury in complexation with organic material is felt to be
the predominant form of mercury in the estuarine environment.  We and
other investigators have observed the evolution of Hg° from samples of
water and sediment.  Based upon these observations and the known meta-
bolic capabilities of bacteria, we hypothesized a role for bacteria in
the transformations and mobilization of mercury in Chesapeake Bay.  This
report describes our investigation of the physiological and ecological
properties of mercury-resistant bacteria.  It also presents evidence for
the relationship of bacterial mercury resistance and metabolism.

-------
                               SECTION IV

               PHYSIOLOGICAL AND CULTURAL CHARACTERISTICS

                     OF MERCURY-RESISTANT BACTERIA
INTRODUCTION
The major objective of this research project vas to determine the mecha-
nism and extent of influence of bacteria in the transformations of mer-
cury which take place in the estuarine environment.  Isolated reports by
other investigators (cited in later sections) of bacterial synthesis and
metabolism of mercury compounds stimulated the initiation of the work
summarized in this report.  The first major premise was that bacterial
resistance to mercury is a function of ability to detoxify mercury by
metabolic transformation.  A corollary was that measurement of numbers
of mercury-resistant bacteria £3 situ should provide an index of real,
or potential, microbiological mercury transformation in Chesapeake Bay.
The following section is a description of the experimental plan followed
in the characterization and isolation of mercury-resistant bacteria from
Chesapeake Bay.

MATERIALS AND METHODS

Isolation. Culture and Identification of Bacteria

Media and methods are described elsewhere (Materials and Methods, Sec-
tion V).

Mercury Analysis

Samples of sediment were allowed to settle, supernatant liquids, if any,
were removed, and the sediments were air dried.  Unfiltered samples of
water were acidified to 0.5 N HN03, and plankton samples were freeze
dried.  All  samples were wet ashed and analyzed for total mercury by
flameless atomic absorption according to  "EPA Provisional Method for
Mercury in Sediment and Water," January 1972.  The vapor phase analyses
of PMA-metabolizing bacteria are described elsewhere  (1).  Six plates of
basal medium containing 0.3 ppm of PMA were inoculated with cultures
grown on slants of the same medium.  The cultures were placed in a her-
metically sealed glass container provided with an on-line dual beam
atomic absorption spectrophotometer.  Six uninoculated plates served as

-------
controls.  Saaples of the vapor phase were analyzed for benzene using a
flow splitter for simultaneous thermocouple and flame ionization detec-
tion.  Saaples were injected under the following conditions t  isothermal
at 50 C, N2 carrier flow at 30 ml/min, 105t Apiezon L on 80/100 Supelco-
port in 0.225 in. by 6 ft. stainless steel columns.  Samples were cali-
brated against 2.0 ml benzene-air mixtures from which cross-integration
of thermocouple and flame ionization detector peak areas yielded a limit
of detection of 0.01 ppm for benzene in air.

Radio Jsotope Experiments

20%g-labeled HgCl2 or PMA were added to suspensions of cells in basal
broth or 0.01 M phosphate or TRIS-buffered (pH 7) "three salts" solutions
and the suspensions were incubated with shaking or agitation with a mag-
netic spin bar or aerated through capillary tubes.  Samples were removed
and added directly to liquid scintillation cocktail or filtered through
0.4-5 micron Millipore filters.  The filters were washed with three 1 ml
volumes of salts and placed in scintillation cocktail to assess cell-
bound activity.  Corrections for chemical quenching and decay of the
isotope were applied.

RESULTS AND DISCUSSION

     1""^ *M»**ts of Bacterial Mercury Metabolism
Resistance of natural populations of aerobic, heterotrophic bacteria was
measured by ability to form colonies on a simple solid growth medium
supplemented with selected inorganic and organic mercury compounds.  The
preparation and incubation of the medium was varied.

Total mercury analysis of natural materials provides only limited infor-
mation with respect to chemical form and biologically available concen-
trations of mercury in a micro environment.  Consequently, mercury
concentrations used in our selective media were arbitrarily chosen
initially.  Samples of water, sediment, and homogenized plankton were
routinely spread on agar containing 6 ppm HgCla (22.1 microgram atoms
of Hg/liter) or 3 ppm phenylaerouric acetate (PMA) (3.9 microgram atoms
of Hg/liter).  PMA was found to be significantly more toxic on a per-
mole-of-Hg basis than HgCl2 and was subsequently reduced in concentra-
tion to 0.3 ppm.  The media were prepared by addition of either aqueous
solutions of HgCl2 in sterile "three salts" solution or alcoholic solu-
tions of PMA to the sterile, molten agar medium.  It was found that
sterilization of the solutions by membrane filtration was unnecessary
and caused a reduction in final mercury concentration of the solutions.

The concentration of resistant bacteria in each sample was expressed as
a percent of the "total viable, aerobic, heterotrophic bacterial count*1
(TVC) obtained by counting colonies appearing after 7 days incubation at
25 C.  The comparative effects of selected concentrations of mercury and
concentrations approximating environmental levels are shown in Table 1 .
Levels as low as -1.2 ppb of PMA or 1 .2 ppb of HgCl2 showed measurable

-------
effects upon bacterial populations.  It is logical to assume that con-
centrations of mercury actually encountered in the environment should
induce detectable effects in the corresponding bacterial flora.

Table 1.  COMPARISON OF RESISTANCE OF SEDIMENT AND WATER POPULATIONS TO
          DIFFERENT LEVELS OF HgCl2 AND PMA
Station
            Test meditm
                                             Percent of

Patuxent
R. mouth
5/1 5 hi





6 ppn HgCl2

0.12 ppm HgCl2
1.2 ppb HgCl2
3 ppm PMA
0.12 ppm PMA
1.2 ppb PMA
Water
0.2

-
72.0
<0.09
-
48.3
Sediment
2.6

75.3
-
0.05
3.7
-
Potomac
R. mouth

 5/16/72
6 ppm HgCl2

0.12 ppm HgCl2

3 ppm PMA

0.12 ppm PMA

1.2 ppb PMA
                                         9.4
  0.76

 93.4

<0.005

  2.5
                                        19.2
 aTVC = total viable, aerobic, heterotrophic bacterial population.   (Re-
  produced with the permission of the Marine Technology Society.)

 The effect of incubation temperature on numbers of mercury-resistant bac-
 teria is shown in Table 2.  As Incubation temperature was increased to
 37 C, apparent mercury resistance increased greater than ten-fold.   This
 result may have been a consequence of increased volatilization of the
 mercury from the medium or a selection for mercury-resistant populations.
 The known non-biological reduction of mercuric ion which occurs in this
 medium (see Section VI) tends to favor the former explanation.  At the
 temperature (15 C) most closely approximating the ia aitu temperature,
 the largest TVC was obtained.

 Spangler e£ al. (2) have reported that mercury-resistant, methyl mercury-
 degrading bacteria metabolize methyl mercury to methane and elemental
 mercury (Hg°) relatively more extensively under anaerobic rather than
 aerobic conditions.  Consequently, plates of HgCl2 medium were inoculated
                                     6

-------
with dilutions of vater and sediment and incubated aerobically and an-
aerobioally, for comparison.  The percent of mercury-resistant bacteria
was always lower for samples taken from three different locations when
plates were incubated anaerobioally (Table 3).  These findings are not
consistent with properties of methyl mercury-metabolizing bacteria as
published by others, but they do agree with those of mercuric ion-
reducing bacteria reported later in this section.

Table 2.  EFFECT OF INCUBATION TEMPERATURE ON BACTERIAL VIABLE COUNT AND
          RESISTANCE TO
Incubation
temperature
(0
37
25
15
2
Incubation
time
(days)
7
7
14
14
Total viable
count (x lOvgram)
1.45
2.60
18.80
11.70
Percent
resistantb
16.7
7.8
1.4
1.1
 A sample of sediment (B2-5/24/73) was diluted and plated on basal
 medium, with and without 6 ppm of HgCLj.  Surface water temperature was
 16.9 C when the sample was taken.

 Percentage of the TVC resistant to 6 ppm of HgCl2.


     Table 3.  COMPARATIVE EFFECTS OF AEROBIC AND ANAEROBIC INCUBA-
               TION ON BACTERIAL RESISTANCE TO
Sample
B2 -
A2 -
EB1-

5/24/73
5/24/73
5/25/73
Percent
Aerobic incubation
7.0
1.3
1.8
resistant8
Anaerobic incubation0
1.0
0.3
0.3
     aSamples of  sediment were diluted and spread on basal medium
       agar,  with  and without 6 ppm HgCl2, and incubated for 7 days
       at  25  C.

       Incubated in BioQuest  (Cookeysville, Maryland) anaerobic jars
       containing  COa-enriched anaerobic atmosphere produced by the
       Gas Pak (BioQuest).

-------
Since an estuary is subjected to extensive changes in salinity, among
other parameters, the effects of the salt content of the growth medium
used for routine isolation of mercury-resistant bacteria was evaluated
for samples taken from sites encompassing a vide range of salinities
(Table 4).  Media of three salinities, including the routine isolation
medium were compared.  Media of the highest (26.59 °/oo) and lowest
(2.66 °/oo) salinities consistently yielded higher proportions of mer-
cury-resistant bacteria than the medium in general use in our studies
(11.3S °/oo).  The effects of aalt concentration may be related to a
salt requirement for mercury metabolism (this Section), or to selective
effects on the bacterial populations in the samples tested.  The extreme
effects of salt observed in the case of samples of lowest salinity (B2-
5/15/72) suggested that the latter argument should also be considered.

Table 4.  EFFECT OF SALT CONCENTRATION ON BACTERIAL RESISTANCE TO HgOLj
Sample
B1 -


EB1 -


B1 -



1/18/72


1/31/72


5/15/72


York R. mouth


B2 -




5/24/73


Salinity of
surface
water (%o)
5.94


10.56


1.56


16.40


5.60


Salinity of
growth
medium* (%o)
2.66
11.38°
26.59
2.66
11.38
26.59
2.66
11.38
26.59
2.66
11.38
26.59
2.66
11.38
26.59
Percent
Water
7.60
7.90
17.00
«,_
—
—
12.90
4.20
50.00
5.30
0.50
0.16
„_
_
—
V
resistant
Sediment
22.90
16.40
19.50
0.79
0.08
0.74
11.50
8.20
29.20
6.30
3.50
5.90
9.25
7.80
11.10
 &Samples of water and sediment were spread on basal medium containing one
  of three concentrations of artificial sea water (HaCl :
  KC1 = 100 : 23 » 3) with and without 6 ppm of HgClg added.

 ^Percent of total viable, heterotrophio bacterial population capable of
  growth in medium containing 6 ppm of HgClg after 7 days  at  25 C.
 °Concentration of salts routinely used in the medium for assay of mer-
  cury resistance.

-------
To define the natural habitats of mercury-resistant bacteria,  water and
sediment samples were separated into filterable, planktonic, and inter-
stitial fractions, respectively (Table 5).  Results shoved that  a non-
uniform distribution of both mercury-resistant bacteria and total mer-
cury concentrations existed among water, plankton, and sediment  samples.
The relative enrichment, i.e., greater concentrations, of mercury In
sediments and living organisms, in comparison to water, is consistent
with other published data  (3, 4) and agrees with the hypothesis  that
planktonic forms of life may be influential in the transport of  mercury,
as well as introduction of mercury, into food chains.  A definite trend
in the data was that mercury-resistant bacterial populations found asso-
ciated with plankton were  relatively larger than those for water or sed-
iment.  Observations of water and sediments indicated that in  most
cases, mercury-resistant bacteria were also distributed non-unifornly
between different particle size fractions.  This suggests that an exam-
ination of mercury levels  and bacterial populations of individual micro-
environments would be the  most logical approach to the problem of de-
fining the relationship of mercury-resistant bacterial population size
to environmental mercury levels.

A variety of bacterial cultures were isolated and characterized  during
the initial 2 years of the investigation.  Fig. 1 shows the average
comparative population distributions of HgClg resistant and total popu-
lations of bacteria.  These distributions were obtained by sampling and
testing cultures randomly  selected from count plates with and  without
added HgCla.  The relatively greater diversity of the total population,
with respect to the generic categories used, and the relative  enrichment
for pgeiyjoiqftT^fl species among the resistant population is evident.
                                      PERCENT
                                     40      60
      PSEUDOMONAS
 ACHROMOBACTER/ALCALIGENES/
     ACINETOBACTER

     CORYNEBACTERIUM
 CYTOPHAGA/FLAVOBACTERIUM
Figure 1.  Average distribution of genera in total and HgCla resistant
           populations.   (Reproduced with the permission of the Marine
           Technology Society.)

-------
Table 5.  DISTRIBUTION OF BACTERIAL HgClg RESISTANCE* AND TOTAL MERCURY CONCENTRATIONS AMONG WATER AND SEDIMENT FRACTIONS
Sample6
A2 - 3/29/72
B2 - 4/03/72
Rhode River
4/13/72
B1 - 5/15/72
A2 - 6/01/72
B2 -10/05/72
EB1 -10/06/72
A2 -12/05/72
B2 - 1/04/73
B2 - 5/24/73
EB1- 5/25/73

Surface water
% Resistant $%J (ppb)
HgC^ PMA
6.5 -- <0.2 ± .02
—
4.8 — <0.2 ± .02
4.2 — <0.2 ± .02
0.3 — —
13.2 — 0.09 ± .01
— _
6.0 — 0.04 ± .01
8.9 22.4 —
8.3 19.6 —
0.08 0.14 —


Source







Filtered Interstitial
water6 Surface sediment water0 Planktona
% Resistant % Resistant $%J (ppm) % Resistant % Resistant $%] (ppm)
HgCl2 HgCl2 PMA HgCl2 HgCl2 PMA
3.1
7.5
7.3
—
<0.4
35.0
—
8.1
—
—
—
7.0
29.8
23.8
8.2
1.5
6.3
4.8
1.5
3.3
6.0
0.31
— 0.
— 0.
~ 0.
— 0.
— 0.
— 0.
— 0.
— 0.
27
82
10
17
37
97
04
20
5.5 0.67
8.2

•
«•
±
±
±
±
±
±
±
±
±

mm*
.03
.02
.01
.01
.02
.04
.00
.08
.01


33
22
5

1
35
31
1



.6
.5
.9
—
.7
.7
.0
.8

MB
• •
11.4 —
31.0 —
—
10.9
2.9 —
—
—
—
18.8
6.1 31.8
0.23 53.5
0.77 ± .04
0.68 + .03
—
—
3.0 + .90
—
—
—
0.09 i .01
0.05 i .00
0.06 ± .05
 Dilutions of material were apread on basal medium,  with and without 6 ppm of HgCl2 or 0.3 ppm of PMA added.   The percent
 of the total, viable, aerobic,  heterotrophic bacterial population resistant  to mercury was calculated.

 Water was filtered through sterile 8 urn pore size Millipore filters.

°Supernatant solution resulting from centrifugation  of sediment at 1,610 x G  for  20 nin.

Collected with #20 nylon mesh plankton net.

eSee Fig. 7, Section V.

-------
The following experiments describe the resistance characteristics of
pure cultures of bacteria grovn in the presence of selected mercury con-
pounds.  Inoculum size, physiological age, and aeration of broth cul-
tures were all found to be factors influencing mercury resistance.
Hence, these parameters were carefully controlled in each experiment.
Two hundred and forty-nine pure, freshly isolated, HgCl^resistant cul-
tures were tested for resistance to a series of types and concentrations
of mercury compounds by replicating from a master agar plate to a set of
mercury-containing plates using sterile Velvetine cloth (Table 6).
Groups of cultures which were strongly resistant to a selected compound
showed a corresponding high resistance to other compounds of mercury.
This suggests that the acquisition of resistance to a single compound of
mercury confers a generalized resistance to mercury compounds.  The com-
parative resistance of bacteria from water and sediments were estab-
lished in an experiment with 131 cultures isolated from samples collect-
ed at a single station in Baltimore Harbor, Maryland (Fig. 2).  In this
particular sample, sediment populations were more resistant to different
types and concentrations of mercury compounds.  Similarly, the percent
of total mercury-resistant bacteria was greater in sediment samples than
in water samples.  Differences in resistance can be explained by the
fact that sediment and water bacterial population distributions are sig-
nificantly different, but not always by intrinsic differences in mercury
resistance, because the relationship between water and sediment in terms
of percent resistance is not consistent (Section V).

  Table 6.  CROSS RESISTANCE OF BACTERIAL CULTURES TO MERCURY COMPOUNDS

                                Percent of cultures resistant
Test compound8
24 ppm PMA
15 ppm PMA
3 ppm Me HgCl
100 ppm HgCl2

24
PMA
—
—
85.8
28.6

15
PMA
~
—
89.6
2^.1
Groupb
3 3
PMA Me HgCl
27.9
— 90.8
6.1
0.0 25.6

100
HgCl2
36.4
63.6
100.0
—

50
HgCl2
9.4
20.8
29.2
—
   jSach compound was incorporated into a solid growth medium.


   Each culture was claa
   cury tolerated (ppm).
Each culture was classified according to the maximum amount of mer-
                                   11

-------
                            WATER
TEST MEDIUM
              PERCENT RESISTANT
20    40   60   80        .    20    4p   6.0
                       n
SEDIMENT
  24 ppM PMA

  15 ppm PMA

   6 ppm PMA

   3 ppm PMA

 3ppm MeH«CI

 lOOppm HgCI£

  SOppmHflCI
    Figure 2.  Comparative percentages of mercury-resistant isolates from
               sediment and water.  (Reproduced with the permission of the
               Marine Technology Society.)


    Mercury resistance in bacteria is often accompanied by resistance to
    other heavy metals and to drugs (5, 6, 7,  8).  Resistance is believed to
    be conferred by factors borne by extra-chromosomal fragments ("plasndds")
    which are transferable.  A group of representative mercury-resistant
    bacteria were screened for resistance to a set of  heavy metal ions using
    a broth medium.  Variable.patterns of resistance,  particularly to Ag ,
    Co2*, Zn2*, Cd2*, and Or46, and Hg2* were  observed (Table 7), suggesting
    that resistance was not generalized, i.e., that determinants of resis-
    tance are independent.  The possibility of plaamid-mediated resistance
    among these bacteria is currently being investigated in our laboratory.

    Bacterial Adtation       c
     It was observed that, after serial-passages of bacteria through growth
     medium in the absence of mercury, resistance to mercury decreased sig-
     nificantly.  When 30 cultures were grown in the absence and presence of
     inorganic and/or organic mercury, followed by replicate plating onto
     mercury-containing media of various types, it was found that growth of
     bacteria in the presence of mercury clearly decreased  sensitivity to
     this metal (Table 8).   "Adaptation" for growth in the  presence of mer-
     cury compounds had been observed earlier by others (5, 9, 10).  This
     phenomenon has been found to be a consequence of the "induction" of
     enzymes  for the detoxification of mercury compounds via reductive de-
     composition of mercury  compounds to volatile Hg° (8, 11, 12).  This
                                     12

-------
activity may be particularly relevant to the short  term response  of
natural bacterial populations to mercury pollution, for the  induction
time frame may be in terms of only minutes or hours,  under the proper
conditions.

      Table 7.  HEAVY METAL RESISTANCE OF MERCURY-RESISTANT  BACTERIA

Culture              Maximum concentration (ppm)* —  Metal ion*3 _ _
  No-     11 >   Pb2+  Ag+1    As*5   Co2+   ^2+°   ^   M2+    Cr+6 ^2+

   72   167. (3d  167.0  16.7  501.0  334.0  334.0   55.1  100.2  668.0  4.0

  119   167.0   167.0  16.7  501.0   55.1   334.0 167.0 167.0    1.7  20.0

  187    16.7   167.0   1.7  501.0   55.1   334.0 110.2   50.1  668.0 16.0

  132    16.7   167.0   1.7  501.0  167.0  334.0 167.0 167.0    16.7  40.0

   85   167.0   167.0   1.7  501.0  167.0  334.0   55.1    50.1  100.2  12.0

  639    16.7   167.0   1.7  501.0  110.2  334.0   55.1  100.2    50.1  12.0

   94    16.7   167.0   1.7  501.0   55.1   334.0   55.1    16.7  334.0 24.0

  244    16.7   167.0   1.7  501.0   55.1   334.0   55.1    50.1  100.2  50.0

  127    16.7   167.0   1.7  501.0   18.4  334.0   18.4   16.7    16.7  24.0

a
 One drop of culture was added to 3 ml of basal  broth containing  dilutions
 of filter-sterilized solution of heavy metal, and  the  tubes were incubated,
 without agitation, at 25 C for 7 or more days.  Tubes  showing turbidity
 were scored positive.  Uninoculated and inoculated controls with and with-
 out metals were included in the assay.
       ion salts added were:  AlClq^HgO,  Pb  (C^Ori^^O, AgNOo,
             , CoCl2'6H20, CuSO^-SttjO,  ZnSO^-7H20,  (3
           , and HgCl2.
CBroth with salt added was adjusted to  pH 7 with 1 N NaOH and filter-
 sterilized.


 Maximum concentrations tested were: Al - 167.0, Pb - 167.0, Ag - 167.0,
 As - 501.0, Co - 668.0, Cu - 334.0, Zn - 334.0, Cd - 334.0, Or - 668.0,
 Eg - 100.0.
                                    13

-------
            Table  8.   ADAPTATION TO MERCUOT RESISTANCE*
Before growth in Hg
Teat medium
3 ppm Me HgCl
24 ppm IMA
24 ppm PMA
24 ppm PMA
^otal number
Medium
grown in
Control
Control
Control
Control
of cultures
% Cultures
resistant
20
3
7
7
tested = 30.
After growth in Kg
Medium
grown in
100 ppm HgCl
100 ppn HgCl
50 ppm HgCl
3 ppm PMA

% Cultures
resistant
63
77
77
73

Paeudomonaa culture #5, when grown in the absence of HgCl2> showed de-
creased resistance to HgCla.  The same culture also evidenced a relative-
ly decreased ability to volatilizeradiplabeled Hg from a buffered
suspension of cells containing ^^Hg-labeled HgCla (Table 15* Section V).
These results indicated the capability to metabolize inorganic Hg in
this bacterial strain is inducible and related to mercury resistance.
The volatile radioactive product of the reaction was isolated and char-
acterized as Hg° (Section II), in agreement with the findings of others
(9, 12).  A simplified system, consisting of non-proliferating cells in
a phosphate-buffered (pH 7.0), artificial, estuarine salts solution
(PES), was used to establish kinetic parameters.  When the suspension was
bubbled with nitrogen, instead of air, Hg° formation was diminished by as
much as 45$.  This observation is possibly related to effect of anaero-
bic sis upon mercury resistance of mixed bacterial populations described
above.  The reaction was dependent upon cell concentration and exhibited
saturation kinetics, characteristics which are attributable to enzyme-
catalyzed reactions.  At 25 C, the half saturation constant  (fin) was 20
ppm of HgCl2 and the mmcimum velocity of reduction was 35.4 ug/mg dry
weight of cella/min.

Similarly, bacterial adaptation to an organic mercury compound, PMA, was
investigated using Paeudomonaa strain 244, which was originally isolated
from PMA-containing medium.  Figure 3 shows the change in resistance of
the culture to a test concentration of 24 ppm of PMA, when grown with and
without 6 ppm of PMA  "inducer."   La contrast  to the non-induced culture,
the induced culture, when plated  on 24 ppm PMA medium, was able to grow
and maintain resistance for a  long period of  time.  The number of PMA
resistant colony forming units  (CPU; in the induced culture  rose 200-fold,
whereas CFU of the non-induced culture decreased 30-fold.  In either case,
these  changes were preceded by a  transient, but reproducible, increase in
resistance prior to the onset  of  growth.  This rapid response is consis-
tent with the induction of  an  enzyme(s).  The experiment was repeated
using  6 ppm of HgCl  as inducer.  In contrast to the previous experiment,
                                     lU

-------
resistance to 24 ppm of PMA rose to only a transient 19-fold increase.
This difference may be the result of loss of Hg   inducer, for it has
been observed that Hg   is chemically reduced and volatilized in the
growth medium employed (Section VI).
          2.2
           i.o
                               3     4
                               TIME (HOURS)
Figure 3.  Induction of resistance to PMA by Pseudomonas 244.  Broth
           with (triangles) and without (circles) 6 ppm of PMA was
           inoculated with cells.  Growth was measured turbidimetrically
           (open symbols), and numbers of cells resistant to 24 ppm of
           PMA/ml were determined (closed symbols).


To ascertain the fate of PMA, the above experiment was repeated, using
Pseudomonas 244 growing in a ^^Hg-labeled PMA medium (Fig. 4).  Dupli-
cate flasks, with and without inoculation, were assayed for cell-bound
and total radioactivity at hourly intervals.  During the period, in
which it was shown in the previous experiment (Fig. 3), that resistance
to PMA increased rapidly, a net loss of 39$ and a cell accumulation of
4$ of the label in the inoculated flask occurred.  The uninoculated
flask showed no significant decrease during the same period of time.

A group of representative bacteria was surveyed for the ability to
volatilize Hg from labeled PMA-containing suspensions (Table 9).  Analy-
sis of the vapor phase of cultures on PMA agar by atomic absorption
spectrophotometry and gas liquid chromatography (1) indicated that the
cultures degraded PMA to Hg° and benzene vapors (Fig. 5).  The apparent
periodicity of Hg  evolution was an artifact of the sampling procedure.
No organomercurials were detectable among the gaseous products.  The
                                   15

-------
reductive decomposition of PMA, methyl mercury  (Me Hg) and ethyl mercury
has also been reported by others  (2, 10, 13).
                            234
                                TIME (HOURS)
 Figure 4.  Uptake and metabolism of FMA by Pseudomonas 244.  Cells vere
            added to broth containing 6.08 ppm of radio labeled FMA after
            5 min. preincubation.  Radioactivity in 0.1 ml of cell sus-
            pension (triangles) and in 1.0 ml of cell suspension collect-
            ed on a membrane filter (open symbols) vere determined.  0.1
            ml samples of uninoculated brotn (squares) vere also analyzed.
            Growth was measured turbidlmetrically (closed symbols).
Washed, non-proliferating cells of
                                                244 retained the ability
 to metabolize FMA for extended periods of time and vere used to study
 the process in detail.  Aerated suspensions of the cells containing
 2°3Hg-labeled PMA in FES vere used to assay decomposition through the
 loss of radioactivity.  Activity was dependent upon cell density and
 aeration driving off the 203ggO formed.  Cultures previously grown and
 harvested from media with and without 6 ppm FMA supplementation vere
 assayed (Fig. 6).  As in the case of the production of Hg° and Eg2*, the
 induced culture was much more active.  Cells of both cultures accumulated
 label rapidly, but the non-induced, non-PMA-metabolizlng culture accumu-
 lated label significantly more than the induced culture.  Thus, for both
 Hg2  and FMA, resistance and metabolism vere related and inducible
 phenomena.
                                     16

-------
Table 9.  METABOLISM OF FHENHMERCURIC ACETATE BY MERCURY-RESISTAHT
          BACTERIA
Isolate
number
244e
187e
94e
127e
72
132
85
21
119
Generic Percent radioactivity reuiiiing*
identification
Paeudofflonaa sp.
Sasad^oaaa «P.
Paeudomonaa an.
Pseudpaonaa ap.
Arthrobacter sp.
Cfrtrobactey ap.
Enterpbacter ap.
Vibrio ap.
Fl *wobact erin?" ap.
1 hourb
52.0
94.8
88.3
82.6
94.8
94.6
92.3
98.2
42.7
4 days0
42. 5d
87.1
73.7
59.9
63.2
65.7
85.1
87.5
43.3
 a0.1 ml PMA in 9556 ethanol, final concentration = 0.4 ug PMA
  (5.75 x 105 cpm/pg), vas added to 0.9 ml of pH 7.0 PES buffer con-
  taining approximately equal quantities of cells.  The suspensions
  were incubated at 25 C, and 100 ul samples were withdrawn.


  With aeration.

 o
  Stationary.


  Isolate 244 was incubated 2 days.

 elsolates 244, 187, 94» and 127 were Pseudomonas sp. types I, II, III,
  and IV, respectively.

  (Reproduced with the permission of the American Society for Micro-
   biology. )
                                  IT

-------
Figure 5.  The change in mercury contained in the bioreactor atmosphere
           as a function of time is plotted for formation of Eg° from
           nine bacterial isolates from the Chesapeake Bay, each growing
           on six agar plates of basal media containing 0.3 ppm PMA.  The
           small open circles represent + std. dev. from the mercury cal-
           ibration curves used.  Hg values plotted here are not addi-
           tively corrected for portions of reactor atmosphere removed at
           each BampUng period.  Note that in order to avoid overlapping,
           the individual curves are not referred to a common zero on the
           ordinate, but the 1,000 ng intervals shown on that scale pro-
           vide an indication of the relative change in the amount of Hg
           present above each isolate referred to a control blank (un-
           inoculated plates).
           (Reproduced with the permission of the American Society for
           Microbiology.)

                                   18

-------
              10
            a.
            o

            0>

           10
           O
           CM
                AA,
                     \
                          V
                                   NON INDUCED



                                O• CELLS
                                A A TOTAL
                     INDUCED
            u


             \
   CELLS
                             NON INDUCED



                             _     INDUCED
-5   0       10       20      30
                 TIME (MIN)
                                                   40
Figure 6.  Uptake and metabolism of FMA by Induced and non-induced rest-
           ing cells of Pseudomonas 244*  The reaction was begun when
           cells were added to the buffered FMA-salts mixture  (arrow).
           0.1/ml samples of cell suspension vere taken at the indicated
           times.
The effect of salinity upon FMA metabolism by Pseudpmonas strain 244 was
tested by varying the total concentration of salts in the three salts
solution, vhich is comprised of NaCl, MgCl2, and KC1.  Cells placed in
three salts  (33% sea water strength) buffered with 0.01 M tris (hydroxy-
methyl) aminomethane (TRIS)-hydrochloride (pH 7.0) were as active as
those in a salts solution at 32£ of the strength of sea water (Table 10).
However, when the salt content was reduced to 8.2$, the cells were in-
active.  Inhibition was partially due to decreased salt concentration and
to TRIS buffer, since the same concentration of salts in the absence of
TRIS gave partial activity.  When 0.01 M phosphate-buffered solutions of
NaCl, MgCl2, and KC1 of equal ionic strength were tested, it was found
that Mg** ion, alone, satisfied the ionic requirement for activity.  In
phosphate-buffered suspensions, a concentration of Mg2"1" Of from 5 - 10 mM
was optimum for FMA degradation.  A similar Mg2* requirement for the
cell-free reduction of Kg2"*" to Hg° by a mercury-resistant strain of
Escherichia coll has also been reported (14).  Magnesium ion concentra-
tion also effects resistance of the organism to FMA, as shown by the
data given in Table 11.  The optimum concentration for resistance was
found to be in the, range 1.2 - 5.6 mM Mg2*.   This observation further
established the certainty of the relationship of FMA resistance and FMA
                                    19

-------
metabolism.  The magnesium effect is particularly interesting in view
of the estuarine locale from which the organism was isolated.  The opti-
mum magnesium concentration approximated the average ia situ concentra-
tions of magnesium.  We have shown that metabolism of PMA by Paeudo-
mpnas strain 244 is dependent upon cell concentration and exhibits
saturation kinetics.  At 25 C, the pH optimum was between pH 6 and 7,
the half saturation constant (fin) was 16.7 ppm PMA, and the Tnnximum
velocity of degradation was 10.6 ug/mg dry weight of cells/min.

Table 10.  EFFECTS OF SALTS AND INHIBITORS ON PMA METABOLISM BY PSEUDO-
           MONAS STRAIN 244

                                  PMA metabolized  (percent of control0)
Experimental condition
82.0$ sea water, Tris buffer.
8.2$ sea water, Tris buffer
8.2$ sea water, no buffer
Tris buffer, alone
NaCl, phosphate bufferb
KC1, phosphate bufferb
MgCLg, phosphate buffer1*
0.01 M KCN°
0.01 M NaN3c
0.01 M NagH AsO^-7H2Oc
Initial rate Net/30 min
99.3 —
15.0
56.1
5.6 -
26.1
0.0 —
99.3 —
56.6 74.2
100.0 82.4
72.0 73.5
 a0.01 M buffered-"three salts" (pH 7.0) containing 6 ppm PMA and ap-
  proximately 0.1 gm dry weight of cells/ml,  aerated at  25 C.


  Concentration equivalent to total ionic strength of "three  salts"  so-
  lution.


 clnhibitors incubated with cells 5 min before reaction  started  in phos-
  phate-buffered three salts.
                                    20

-------
       Table 11.  EFFECT OF MAGNESIUM ION CONCENTRATION ON PMA
                  RESISTANCE IN PSEUDOMONAS STRAIN
Magnesium
concentration
(mM)
0.1
1.2
5.6
11.2
22.3
Growth (davs)a
PMA concentration (ppm)
0
+1
+1
+1
+1
+3
10
+3
+1
+1
+1
+1
15
—
+3
+1
+1
+1
20
—
+3
+3
+3
+3
25
—
+3
+3
+3
+3
30
—
—
+3
+3
+3
       a0.01 ml of culture was transferred to tubes of basal
        broth containing varying concentrations of MgCl2 and
        PMA.  The tubes were incubated at 25 C, with shaking,
        and observed up to 7 days for growth.

         Site of Mercury Binding and Metabolism

The effects of several inhibitors of oxidative phosphorylation were
tested in Pseudomonas strain 244- PMA-metabolizing system to see if
active transport of PMA into the cells was essential for metabolism
(Table 10).  Potassium cyanide (10 mM) caused considerable inhibition of
the initial rate of reaction, but none of the inhibitors caused more
than a 30$ net decrease in the PMA metabolized after 30 min.  The partial
inhibitions may have resulted from the formation of complexes  of phenyl-
mercuric ions with the inhibitors.  Cells treated with sodium arsenate
or sodium azide bound considerably more label than did the control, but
cyanide-treated cells bound less label.  These results suggested that
PMA metabolism takes place on, or near, the cell surface.  A similar con-
clusion was reached by Tonomura et al. (10), who used a PMA-resistant
Pseudonpnafi.  The cellular site for binding of PMA was examined in the
following-ijay.  Cells of Pseudomonas strain 244 were grown in the pres-
ence of *^Hg-labeled PMA for 2/3 generation time, after which the cells
were fractured by mechanical shear.  PMA degradation was stopped iircnedi-
ately by the breakage, and the cells were separated into crude subcellu-
lar fractions by differential centrifugation (Table 12).  Supernatant
solutions were separated from pellets, after sequential centrifugation
at 3,000 x G for 15 min, 35,000 x G for 15 min and 126,000 x G for 1 hr.
Approximately 75$ of the cell-bound radioactivity was present  in partic-
ulate fractions sedimented by 3,000 and 35,000 x G forces.  On the basis
of these findings, it was concluded that the majority of the PMA binds
to components of the cell envelope.
                                   21

-------
Table 12.  DISTRIBUTION OF RADIOACTIVITY IN ^^g-PMA-LABEIED CELTS OF
           PSEUDOMONAS STRAIN
Cell
Cell
3,
35,
126,
126,
fractionb
free
000 x
000 x
000 x
000 x
extract
G
G
G
G
pellet
pellet
pellet
supernatant
Total
counts/minc
1.
2.
5.
8.
1.
59 x
28 x
85 x
06 x
96x
10*
10*
10*
103
10^
Percent radio-
Percent activity recovered
of total in fractions
100.
U.
37.
5.
12.
0
3
4
1
3
••
21.
53.
7.
18.
•^m,
0
7
3
0

  added to 100 ml of fresh broth containing 6 ppm of PMA.   The culture
      incubated for 2 h,  with shaking,  at 20 C,  and 20 microcuries of
      '-labeled PMA were  added.   The culture was incubated an additional
  hour, and labeled cells vere harvested by centrifugation.


 bThe labeled cells, suspended in chilled 0.01 M TRIS-buffered (pH 7.0)
  "three salts", were fragmented by two passages through a French Pres-
  sure Cell.  The resulting cell-free extract was fractionated by dif-
  ferential centrifugation.

 cAliquots of each fraction were collected, and radioactivity was mea-
  sured by liquid scintillation counting.


Cells of induced and non-induced cultures of Pseudomonas strain 244- vere
compared to identify a cell component responsible for mercury resistance.
Preliminary investigations have shown no differences in gross lipid com-
position or in the morphology, by inspection of thin sections of cells
using electron microscopy.  However, striking cytological effects were
observed in cultures vhich were less resistant to mercury and which were
grown in the presence of HgCl2 (Section VI).  Cells of several cultures
grown in the presence of HgCl2 shoved morphological similarities, sug-
gesting that mercury impairs normal cell wall and membrane synthesis and
function.  This observation most likely is related to the propensity of
mercury to bind to the cell envelope.

SUMMARY

This section described the sources of mercury-resistant bacteria and
techniques used in this investigation.  The cultural characteristics of
mercury-resistant bacteria and the physiology of aerobic mercury metabo-
lism vere discussed.
                                   22

-------
Aerobic, heterotrophic bacteria resistant to mercuric ion were isolated
from water, sediment, and plankton.  They are predominately of the
genus, Pseudomonas.  Many of these organisms are also resistant to
organomercurials, FMA (phenylinercuric acetate) and Me HgCl (methylmer-
curic chloride), deriving resistance from their ability to degrade mer-
cury compounds with the formation of Hg°.  The capability of the organ-
isms to produce Hg° from Kg2* and FMA is correlated with their ability
to grow in the presence of compounds of mercury.  Mercury resistance is
also related to conditions of incubation, such as aerobicsis and tempera-
ture of incubation, as well as the salt content of the growth medium and
the physiological age of the cultures.
                                   23

-------
                               SECTION V

            MICRDBIAL ECOLOGY OF MERCURI-RESISTANT BACTERIA

                           IN CHESAPEAKE BAT
INTRODUCTION

Comparatively little is knovn about chemical and biological processes
involved in the availability and mobility of inorganic mercury, the pre-
dominant pollutant form of mercury in the environment and a precursor of
methyl mercury.

Microbial transformations of mercury, other than those involved in the
formation of methyl mercury, have been described by several investiga-
tors.  The reductive decomposition of organic and inorganic mercurials,
with the formation of elemental mercury (Hg°), have been reported for
several genera of mercury resistant bacteria (1, 2, 8, 9, 13, 15, 16).
This phenomenon may account for the low ambient levels of methyl mercury
(17, 18) and, possibly, for the observed loss of Hg from contaminated
sediments and soil (19, 20, 21, 22).  However, the process of microbial
Hg° generation should be emphasized as a potential mechanism for the mo-
bilization of mercury and for the generation of substrate for methyla-
tion.  The latter consideration is supported by evidence that Hg° is
readily oxidized to Hg*^ jn sjltu under the influence of dissolved oxygen
and organic matter (23, 24).

Although direct pollution of the aquatic environment with mercury has
been drastically curtailed during the early 70's, extensive sedimentary
deposits exist which constitute a continuing methyl mercury hazard (25).
In view of this prospect, an understanding of microbial processes in the
generation of Hg° and Bg2  is essential, particularly for evaluating en-
vironmental impact.

This report describes a study of the geographical and seasonal distribu-
tions of aerobic, heterotrophic, mercury-resistant bacteria involved in
the formation of Hg° in Chesapeake Bay.

-------
MATERIALS AND METHODS
Water samples for bacterial plate counts vere collected In sterile dilu-
tion bottles or sterile Niskin bag samplers (General Oceanics, Miami,
Florida) 10 cm below the surface.  Sediment samples for bacterial plate
counts and mercury analysis were taken from the upper 5 » 10 cm layer
using an Ekman dredge or Petite Ponar grab (Wildlife Supply Co., Saginav,
Michigan).  Water samples for mercury analysis were collected in 0.5 N
HNOj-washed bottles and acidified promptly with 5 ml of concn. HN03 per
liter.  Horizontal plankton tows for 10-30 min, using plankton nets
equipped with 20 mesh nylon cloth, provided ample material for analysis.
The collected plankton were centrifuged, resuspended in sterile salts
solution (see below) and either homogenized for bacterial plate counts
or recentrifuged and lyophilized for mercury analysis.  Appropriate con-
trols and experiments eliminating the possibility that lyophilization
affected the mercury analyses were made.

Mercury Analysis

Sediment samples were allowed to settle, the water layer, if any, was
removed, and the sediments were air dried.  Each sample of water or sed-
iment was wet ashed according to Environmental Protection Agency proce-
dures and analyzed by flameless atomic absorption spectrophotometry
(26).  Analyses were performed at the National Environmental Research
Center, Cincinnati, Ohio and at the National Bureau of Standards, Wash-
inton, D.C.

Isolation. Culture, and Identification of Bacteria

The medium employed for determination of the total viable count (TVC)
and enumeration of mercury-resistant bacteria consisted of glucose, 2.0
gmj Casamino acids (Difco Laboratories,  Detroit,  Michigan), 5.0 gm;
least extract (Difco), 1.0 gm; and Bacto agar (Difco), 20 gm;  per liter
of artificial estuarine salts solution,  i.e.,  10.0 gm of NaCl, 2.3 gm of
MgCl2'6H20, and 0.3 gm of KC1 per liter.  The salts mixture was adjusted
to pH 7.2 and autoc laved at 121  C for 15 min.   Mercury-containing media
were prepared by adding freshly prepared solutions of HgClg in sterile
salts solution or PMA in 95/6 ethanol to sterile,  molten agar medium.
Suitable dilutions of material were prepared in sterile salts solution
and spread on agar medium with and without added mercury within 5 hours
of sampling.  The inoculated plates were incubated for 7 days at 25 C,
after which the colonies were counted.   Colonies appearing on mercury-
containing media were selected on a random basis and purified.  The or-
ganisms were identified according to a taxonomic scheme originated by
Shewan (27) which was modified by incorporation of the methods and proce-
dures described by Colwell and Wiebe (28).
                                   25

-------
Assay for Mercury Evolutiop

Cultures were incubated with shaking in Erlenmeyer flasks of broth con-
taining 6 ppm HgCla for 16 h at 25 C and subcultured (20% inoculum size)
in freah medium, with subsequent incubation for 1 to 6 h.  After incuba-
tion, the cells were centrifuged from the medium and washed twice in
0.01 M potassium phosphate-buffered (pH 7.0) estuarine salts (PES).
After resuspension, the cells were divided into three portions:  one was
assayed directly, the second was autoclaved at 121 G for 15 min to serve
as a control, and the third vas centrifuged, resuspended in distilled
water, and dried at 80 G for dry weight determination.  Suspensions of
cells in FES were added to FES in a test tube to 4.9 ml final volume.
The suspensions were aerated with moist air to keep the cells evenly
dispersed and to enhance the evolution of Hg° from solution.  After 5
min aeration, 100 ul of 2"^Hg-labeled HgCla (Amersham Searle Corp.,
Arlington Heights, Illinois) was added.  Samples  (100 ul) were removed
at intervals and mixed with 10.0 ml of 3a70 preblended liquid scintilla-
tion vials.  Duplicate samples were filtered through 0.45 micron Milli-
pore filters  (Millipore Corp., Bedford, Massachusetts) and washed thrice
with 1.0 ml volumes of FES at 25 C.  The filters were immersed in 10.0
ml scintillation cocktail.  Samples were counted in an Intel-technique
model SL-40 liquid scintillation spectrophotometer (Teledyne Isotopes,
Westwood, New Jersey).  Radioactivity, expressed in disintegrations per
min, was calculated automatically with a channels ratio program.  Cor-
rections for  decay  (47 day half life) were made when necessary.

Statistical Analyses

Data expressed  as percent were converted by arc  sine transformation  (29)
to approximate  a normal distribution.  Multiple  stepwise  linear regres-
sions were performed  by an  IBM. 1108 computer using biomedical program
BMD03R from the University  of Maryland Computer  Center  Library.

Isolation of  Microbial Populations Directly from Water and  Sediment

Bacteria and  seston were  recovered from vater  samples by  centrifugation
at  13,200 x g for 15  min.   The greenish-brown  material was  resuspended
 in sterile  FES before use.   Bacteria  were  extracted from sediments by a
modification of the methods of  Balkwill and  Casida  (30) developed by
those authors for soils.  One hundred grams  (wet weight)  of sediment was
 homogenized with 200 ml of  sterile  0.1$  sodium pyrophosphate in  salts
 solution (PES)  in a Waring  blender for two 30  sec. blending intervals.
 The mixture was centrifuged at  365 x  g for 5 min and the  supernatant so-
 lution was decanted and recentrifuged at  365 x g.  The  supernatant  solu-
 tion was again decanted and centrifuged at 13,200 x g for 10 min.   The
 straw-colored supernatant solution was  discarded, and the pellet was re-
 suspended in sterile FES and centrifuged at  475  x g for 10  min.  The
 dark brown pellet was discarded,  and the greenish brown supernatant so-
 lution was retained for use.
                                    26

-------
RESULTS

Experimental s^ml     Sa."
Chesapeake Bay was traversed longitudinally during two cruises, one in
May 1972 and the other in February 1973, to survey total mercury concen-
trations and mercury resistant populations of bacteria in sediments (see
Fig. 7).  The results (Table 13) shoved that a range of mercury concen-
trations exists in Chesapeake Bay sediments, with highest levels present
in the two most industrialized areas, Baltimore Harbor (stations 2 and 3,
Fig. 7) and the Elizabeth River near Norfolk, Virginia (station 20, Fig.
7).  Bacterial resistance to HgClj, expressed as the proportion of the
total viable, heterotrophic, aerobic population capable of growth in a
medium containing 6 ppm of HgCla, was also greatest at these two sites.
The data strongly suggest a possible causal relationship between ambient
mercury levels and numbers of mercury resistant bacteria.

A total of 6 stations at three locations in the upper Bay, shown in Fig.
7 (B-1, 2; A-1, 2; and EB-1, 2), were monitored at approximately 1.5
month intervals over a 17 month period, Fall 1971 through Spring 1973.
Samples of surface water (10 cm depth) and the upper sediment layer were
routinely analyzed for numbers of bacteria resistant to 6 ppm of HgCl2-
and 3 ppm of PMA and for total mercury concentration.  The three stations
encompassed a spectrum of environmental quality and ambient mercury con-
centrations from heavily industrialized Baltimore Harbor to relatively
clean Eastern Bay, the site of important fin and shellfish fisheries.
Differences in mercury concentrations in sediments among the three loca-
tions were significant at the 0.05 level and did not change appreciably
during the 17 month monitoring period (15).   Techniques of sufficient
accuracy and precision for detection of mercury in water samples were
developed somewhat later in the study.  Levels of from 0.00 to 0.68 ppb
were detected in the water from the three sites.   However, there were
insufficient data to draw conclusions as to the relationship of water  to
sediment mercury concentration or to location.   Colgate Creek (station
B-2), a subestuary in Baltimore Harbor, was found to contain the highest
concentrations of mercury found in Chesapeake Bay during the course of
this study.  This site was subsequently found to be heavily contaminated
with high levels of other heavy metals and oil pollutants as well (0.
Villa and J. Marks, personal communication).   The sources of mercury at
this station have not been identified directly, but the indications are
that point sources, such as a paint factory or a hospital sewage outfall,
in the area may be implicated.   The two sites of study which are located
near Annapolis Harbor, A-1  and A-2 shown in Fig.  7, are subject to heavy
commercial marine traffic.

Relationship of Mercury Resistance and Metabolism j.n Bacterial Isolates

From the onset of this investigation, the working hypothesis was that  the
percent of bacteria in the autochthonous microbial population which
demonstrated resistance to mercury could be used as an index of potential
microbial activi'ty in the generation of Hg° from inorganic or organic
                                   27

-------
sources of mercury.  The  hypothesized relationship between mercury re-
sistance and metabolism vas suggested by the published observations
cited previously, and vas substantiated by a series of experiments.
                   CHESAPEAKE BAY
                                                      39*
                                                      30
                                                       39-
                                                       00
                                                       38-
                                                       OO'
                                                       37»
                                                       30'
                                                       3T
                                                       OO'
                                                     7S"30'
 Figure 7.  Experimental sampling sites in upper Chesapeake Bay.   Station
            designations are:  2 = B-2 (Baltimore Harbor);  3 = B-1 (Balti-
            more Harbor); 4 = A-1 (Annapolis);  5 = A-2 (Annapolis);  7 =
            EB-1 (Eastern Bay); 8 = EB-2  (Eastern Bay);  and 20 = Elizabeth
            River.
                                     28

-------
Table 13.  TOTAL MERCURY CONCENTRATIONS AND PERCENTAGE OF THE TOTAL VIABLE
           BACTERIAL COUNT RESISTANT TO 6 ppm OF HgCl2
location*
1
2-Baltimore Harbor (B-2)
3— Ba^i^jor*^ PflT'?®^' (B— 1 )
4-Ar)m»polis (A-1)
5-AxmaDolis (A-2)
6
7-Eaatern Bav (EB-1)
8-Eastern Bav (EB-2)
9
10
11
12
13
14
15
16
17
18
19
20 (May 1972)
20 (February 1973)
&Numbers refer to samplim
Mercury concentration^
Sediment (ppm) Water (ug/l)
0.220 ± 0.004
0.800 ± 0.070d 0.090 ± 0.010
0.590 + 0.6206 0.370 + 0.010
0.060 + 0.040e 0.080 + 0.010
0.200 ± 0.060* 0.490 ± 0.010
0.170 ±0.010
0.080 + 0.030d 0.100 ± 0.020
0.040 ± o.oso8
0.094 ± 0.002
0.015 ± 0.007
0.104 ± 0.013
0.120 + 0.010
0.073 ± 0.002
0.100 ± 0.010
0.052 ±0.011
0.060 ± 0.020
0.023 ±0.009
0.070 ± 0.010
0.007 ± 0.002
0.860 ± 0.030
0.280 ± 0.013 0.070 ± 0.010
* stations. Regularly sampled
Percent
Sediment

22.8
8.2

2.9

1.3




2.6

0.8

1.8

3.5

6.3
2.2
resistant0.
Water

28.5
4.2

2.8

0.5




0.2

9.4

0.7

0.5

5.7
2.3
stations are under-
 lined.  Other stations were sampled in May 1972 and February  1973.

Tfean mercury concentration (dry veight) + average deviation of one  sample,
 except where designated otherwise.   Water values based upon unfiltered
 samples taken October 1972.

C0ne standard deviation is approximately 10$ of "Percent  resistant."  Sta-
 tions 3, 12, 14, 16, 18, and 20 assayed on May 1972 cruise  and stations  2,
 5, 7, and 20 assayed on February 1973 cruise.

 Ninety-five percent confidence interval of the mean mercury concentration
 of 8 samples.

6Ninety-five percent confidence interval of the mean mercury concentration
 of 6 samples.
f>
 Ninety-five percent confidence interval of the mean mercury concentration
 of 9 samples.

^Ninety-five percent confidence interval of the mean mercury concentration
 of 4 samples.

                                    29

-------
                       Table U.  HgCl2 METABOLIZING BACTERIA
Source1*
B-2
B-2
B-2
B-2
B-2
B-2
B-2
B-2
B-2
B-2
B-2
B-2
B-2
B-2
B-2
B-2
EB-1
EB-1
EB-1
EB-1
Culture
1
3
U
5
7
S
10
11
12
13
U
15
16
17
19
20
21
24
26
28
mg Cells
(dry weight)
0.99
2.36
1.17
1.08
0.84
1.17
0.42
0.84
1.02
0.17
1.26
0.72
1.41
0.72
0.81
0.51
0.84
0.84
1.48
1.04

Percent
change in (203Hg)a
Decrease
live
12.6
38.1
1.1
83.4
5.6
15.0
26.1
33.4
9.0
1.7
21.9
14.7
81.6
11.8
31.4
30.4
14.2
33.3
21.0
37.8
Killed
5.4
8.5
5.5
3.2
4.1
5.9
1.1
7.6
2.5
1.2
7.3
2.7
6.5
~
3.0
7.5
2.0
3.2
5.9
4.8
Uninoculated
4.6
4.5
4.6
_
8.0
4.6
4.5
__
4.6
4.5
4.6
— —
-_
8.0
4.5
4.5
4.5
4.5
—
4.5
Cellular0
uptake
22.7
51.8
29.6
4.8
26.6
21.0
31.8
20.0
20.6
8.8
36.5
23.8
16.0
—
30.2
20.2
13.5
11.8
38.2
30.0
Incubation
tine
(h)
48
72
48
1
24
48
72
1
48
72
48
1
1
24
72
72
72
72
1
72
 A suspension of washed cells, maintained at  25 C  in 4.9 ml of FES and aerated via a
 capillary tube, was incubated for 5 min prior to  the addition of 0.1 ml of 203ng-
 labeled HgCl2 (final concentration of  6 ppm  HgCla; appro*. 100,000 CPM/ml).  0.1 ml
 samples were removed at zero time and  at regular  intervals to 10.0 ml scintillation
 cocktail for counting.  Percentage decrease  or uptake was  based upon counts corrected
 for quenching and decay relative to radioactivity in the killed control at t = 0.

 Selected from 6 ppm HgClg count  plates of water and sediment from Colgate Creek and
 sediment from Eastern Bay.   Of the 30  original colonies suboultured from the count
 plates to broth containing 6 ppm HgClg, 22 grew,  with 17 actively metabolizing Hg2*.

30.1  ml aliquots were filtered through  .45 ji  Millipore filters and washed three times
 with 1.0 ml volumes of FES (25 C).

-------
An examination of one of the most active isolates  (culture #5) further
suggested a correlation of resistance and metabolism.  The organism was
grovn in the absence of HgCla, with concomitant diminution of its toler-
ance for, and ability to metabolize, HgCl2  (Table  15).  Experiments uti-
lizing a closed system with a HgBr2-KBr mercury trap  (18), in which the
radioactive volatile product of the reaction was quantitatively collected,
suggested that Hg° evolution accounted for  the loss of Hg from the sus-
pension  (Table 16).  Radioactivity in the acidified trapping solution
was not benzene-extractable, nor were methyl or other alkyl mercury com-
pounds detected in thin layer chromatograms of the dithizone derivative.

Table 15.  MERCURY RESISTANCE AND METABOLISM OF BACTERIAL STRAIN NO. 5
           GROWN IN THE PRESENCE AND ABSENCE OF HgCL,
Cultureb
Grown with HgCL>
Growtha
HgCl2 concentration (ppm)
4 8 12 16 20
+ + -f + +


24
_
 Grown without HgClo
  Tubes were inoculated with 1  drop of culture,  incubated,  with shaking,
  at 25 C,  and observed daily for up to 21  days.

  Cultures  were serially transferred twice  with  or without  6 ppm HgCl2
  in the growth medium.  A 24. hour old culture was used to  inoculate the
  test medium.

 B.  Metabolism
                me                Percent change in (20^Hg)/30 mina
 Culture*3   Dry weight	Decrease in suspension     Uptake by cells

               °-39                     38.8                   13.7

 Grown with—   r\ i ?                      c-i                   -ICA
   out         °'43
 aA suspension of washed cells, maintained at 25 C in 4.9 ml of PES and
  aerated via a papillary tube, was incubated for 5 mln prior to the ad-
  dition of 0.1 ml of HgCl2 (final concentration of 6 ppm HgCl2; approx.
  300,000 CPM/ml).  0.1 ml samples were removed at aero time and at
  regular intervals to 10,0 of scintillation cocktail.  Percentage de-
  crease in the suspension or increase in the cellular (non-filterable)
  radioactivity was based upon counts corrected for quenching.

  The cultures were grown for 2 h at 25 C, with shaking, in broth in the
  presence and absence of 6 ppm HgCl2.  Inocula were cells grown for 24
  h with and without 6 ppm of HgCl2 in the growth medium.  Equal turbidi-
  ties were attained in each culture at the end of the incubation period.
                                     32

-------
Table 16.  CHARACTERIZATION OF THE VOLATILE Hg METABOLITE PRODUCED BY
           PSEUDOMONAS STRAIN 5
Experiment
f
11°
IIId

Sample
Suspension after 60 min
Cells after 60 min
HgBr2-KBr trap after 60 min
Benzene extract of trap
a. Suspension after 90 min
b. Cells after 90 min
c. HgBr2-KBr trap after 90 min
d. First dithizone extract of trap
e. Second dithizone extract of trap
a. Dithizone extract of II c.
b. Thin layer chromatography of III a.
Hg2^ spot plus origin
Percentage of
added f°3Hg)a
0.67
0.98
85.00
0.00
0.80
0.60
96.80
22.80
76.20
-U.OO
u.oo
aA flask containing 9.0 ml of PES containing 203Hg-labeled HgClg was
 connected via plastic tubing to a second flask containing 10.0 ml of
 HgBr2~KBr trapping solution.  The experiment set-up was such that
 the atmosphere was continually recirculated by means of a peristaltic
 pump as the flasks were shaken.  After 15 min incubation, 1.0 ml of
 washed cells in FES was added to the first flask and incubation was
 continued.  The trapping solution was examined for radioactivity
 before and after addition of cells to the system.   Transfer of radio-
 active Hg to the trapping flask was insignificant in the absence of
 the bacterial suspension.  At the end of the incubation, cells were
 filtered through a 0.45 micron Millipore filter and washed thrice
 with 1 .0 ml volumes of PES.

b4.1 ml dry weight of cells with 1.8 ppm 203HgCl2 (131,000 dpm/ml).
 Five volumes of trapping solution were acidified with concentrated
 HC1 to a concentration of 1 N, and were extracted with 1 volume of
 benzene.

C1.9 mg dry weight of cells with 1.5 ppm 203HgCl2 (202,000 dpm/ml).
 One volume of trapping solution was extracted with 2 volumes of a "\%
 solution of dithizone in benzene.  The extraction was repeated.
     volume of trapping solution (Expt.  II) was extracted with 2
 volumes of 0.856 dithizone in benzene.   The dithizonate derivative was
 spotted on a silica gel G plate (250 )i thickness) and developed with
 petroleum ethertdiethyl ether (70:30)  solvent.  Spots were scraped
 off and collected in scintillation vials and suspended in an Aquasol
 (New England Nuclear) gel for counting.  Dithizone extraction and
 chromatography were according to the methods of Westoo (31 ) .
                                  33

-------
The microbial populations of water and sediment, collected by differen-
tial centrifugation, also carried out the reductive process.  Recoveries
of bacteria isolated by this method were essentially quantitative for
water and approximately 1# for sediments, based upon total, viable,
aerobic, heterotrophic bacterial counts.  Assays for total and non-
filterable radioactive mercury in suspensions containing live and sterile
inocula and uninoculated controls were performed as described for the
previous pure culture experiments (Table 17).  Under the conditions of
the assay, the microorganisms slowly released Hg°.  The rate of loss
measured for the sterilized controls indicated some chemical reduction,
i.e., non-biological reduction also occurred.  The major difference be-
tween the Colgate Creek and Eastern Bay samples (B-2 and EB-1), respec-
tively) was, in each case, the amount of mercury bound to non-filterable
material.  Lowering the initial Kg2"*" concentration by an order of magni-
tude decreased the loss of Hg°, but increased binding.

Analysis of Seasonal Data

Figures 9 and 10 illustrate changes in proportions of HgCljj-resistant
bacteria in water and sediment during the period from January 1972
through May 1973.  A distinct periodicity was evident for both water and
sediment, with a major peak in mercury-resistant bacteria observed in the
Spring months of both 1972 and 1973 and a possible secondary peak in the
Fall of 1972.  Figures 9 and 10 also show the relative differences in
resistance among three of the stations.  Several physical parameters
were found to demonstrate a seasonal periodicity  (Fig. 11) characteris-
tic of the Chesapeake Bay estuary  (32).

The data were subjected to multiple stepwise regression analysis to de-
tect possible relationships between the selected physical parameters and
number of resistant organisms and to establish a basis for comparison of
the three sites.  The percentages of bacteria resistant to HgCl2 in
water and sediment were treated as dependent variables, respectively.
Independent variables, whose partial regression coefficients were sig-
nificantly greater than zero  (t test of significance), are presented in
Table 18.  Standard partial regression coefficients were also calculated
to show the relative contribution of each variable to the regression
equations  (29).   When  each site was considered separately, it was evi-
dent that there were differences in factors affecting mercury resistance
among the sites (sample 1, 2 and 3).  When  data for all six stations
were added to the regression  equation,  the  individual measured parame-
ters did  not  adequately account for changes in resistance  (sample 4-).
Three  stations  were  selected on the basis of  similarities in sediment
type and  differences in  sediment concentrations  (sample 5).  Percentage
of resistance in water and sediment for each  site were observed to be
related,  but  differences between water  and  sediment were not significant
 (0.05  significance level).  However, average  differences in percentage
of resistance  in water and sediment, respectively, among the three loca-
tions  were  significant (0.05  significance  level).  Resistance in sedi-
ments  was found to be  correlated positively with  water transparency
 (Secchi disc),  dissolved oxygen, and total mercury concentration in the
 sediment.   Resistance  was not  observed  to be  related to total viable
count.

                                    3U

-------
Table 17.
Hg° EVOLUTION BT NATURAL POPULATIONS OF BACTERIA ISOLATED FROM WATER
AND SEDIMENT
                                                Percent change in (203Hg)a
Sample
  /~HgClp_7  TVC per  Time
    (ppn>7      ml      (h)
                                                    Decrease
Uninocu-  Live sus-
 lated     pension
  Sterile   Cellular
suspension   uptake
B-2 sediment
B-2 sediment
B-2 sediment
B-2 sediment
B-2 sediment
EB-1 sediment
EB-1 sediment
EB-1 sediment
EB-1 sediment
EB-1 sediment
B-2 vater
B-2 vater
B-2 vater
B-2 vater
B-2 vater
B-2 vater
B-2 vater
B-2 vater
B-2 vater
B-2 sediment
B-2 sediment
B-2 sediment
B-2 sediment
EB sediment
EB sediment
EB sediment
EB sediment
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
1.2 x 10$
1.2 x 105
1.2 x 105
1.2 x 105
1.2 x 105
1.0 x 105
1.0 x 105
1.0 x 105
1.0 x 105
1.0 x 105
7.0 x 10-!
7.0 x 10^
7.0 x 10^
7.0 x 105.
7.0 x 105
1.8 x 10^
1.8 x 10°
1.8 x 10°
1.8 x 106
4.6 x 105
4.6 x 1CP
4.6 x 105
4.6 x 105
2.5 x 10^
2.5 x 10°
2.5 x 10°
2.5 x 106
1
2
3
24
48
1
2
3
24
48
1
2
3
24
48
2 5.3
18 8.0
42 4.6
66 4.5
1
2
3
24
1
2
3
24
6.6
10.3
11.0
11.6
13.7
7.4
10.5
8.8
9.1
11.8
7.7
9.3
11.1
15.6
20.6
5.8
10.5
10.2
12.7
0.3
1.8
1.8
4.1
4.0
4.2
5.5
7.7
1.7b
0.0
3.0
6.3
9.9
1.7b
3.9
2.2
9.1
10.7
1.3b
1.0
0.1
0.7
2.4
4.6C
7.8
7.5
7.2
1.4b
4.1
5.5
7.1
4.7
5.7
7.7
13.6
38.3
38.7
40.2
46.4
50.5
9.4
10.9
11.7
15.6
16.8
6.8
7.2
9.0
9.7
10.7
5.8
6.0
7.7
9.6
95.0
101.0
96.5
99.9
40.0
44.3
44.7
57.2
a0.1 ml of 20%gCl2 in PES vas added to a 4.9 ml PES suspension of seston or sedi-
 ment extract (prepared as described in Materials and Methods).   The tubes of sus-
 pension vere aerated, via a capillary tube, vith moistened air for 15 min prior
 to each sampling.  Radioactivity in 0.1 ml samples of suspension and non-filter-
 able material vas measured and corrections for quenching and decay vere made.

^Sterilized by heating at 121 C for 15 min.  No attempt vas made to maintain
 sterility after the start of the incubation.

cSteriliaed vith chloroform.  No attempt vas made to maintain sterility after the
 start of the experiment.
                                        35

-------
Figure 9*  Seasonal variation in percent of HgClg resistant bacteria in
           vater.  The percent of total viable, aerobic, heterotrophic
           bacterial count  (TVC) capable of growth on a solid medium con-
           taining 6 ppm of HgCl2 was determined.  Water samples from Station
           B-1, 1972 (A);  Station B-2, 1972 and 1973 (A,©)j Station A-1,
           1972 (©); Station A-2, 1972 and 1973 («,D); and Station EB-1,
           1972 and 1973 (0,0)* were plated and incubated at 25 C for 1 week.
         30-
       E
       a
       o.
       
-------
       15
      I "0
      >-
      §

      uj 5
      05
      o
       15
                                                                     30 LU
                                                                       o
20 g
  UJ
  
-------
                 Table 18.   MULTIPLE REGRESSION ANALYSIS OF SEASONAL DATA
I
% water
% water
% sediment
% water
% water
% water
% water
% sediment
% sediment
% sediment
% sediment
% sediment
% sediment
% sediment
% water
% water
% sediment
% sediment
% sediment
% sediment
a
*n
3
5
6
7
9
2
4
1
3
4
5
6
3
9
3
2
1
7
8
9
Partial standard
Correlation coefficients regression
Simple Partial Multiple coefficient
40.08557
-0.76012
+0.50568
+0.79797
-0.42607
+0.79752
+0.26347
+0.79752
+0.68272
+0.45798
+0.01878
-0.14795
-0.00870
-0.15035
+0.67893
+0.66167
+0.66167
-0.47979
+0.14737
+0.66394
-0.90433
-0.84457
+0.82607
+0.81540
-0.72052
+0.94433
-0.89642
+0.94434
+0.94348
+0.95714
-0.96038
-0.92919
-0.96131
+0.94953
+0.60978
+0.5777
+0.5777
+0.50305
+0.53308
+0.59526
0.9871b
0.9871
0.9232
0.9640
0.9640
0.9731
0.9731
0.9950
0.9950
0.9950
0.9950
0.9950
0.9950
0.9950
0.7358°
0.8171
0.8834d
0.8834
0.8834
0.8834
-9.14
-1.29
+0.42
+0.85
-0.57
+1.96
-0.95
+0.45
+0.64
+0.49
-8.30
-1.09
-7.30
0.67
0.65
0.71
0.47
0.59
1.55
0.83
Significance
level df
0.05
0.05
0.05
0.05
0.10
0.05
0.10
0.05
0.05
0.01
0.01
0.05
0.01
0.05
0.001
0.05
0.05
0.10
0.10
0.05
10
10
10
11
11
9
9
9
9
9
9
9
9
9
32
17
17
17
17
17
Sample
1-A1, A2
n
n
2-B1, B2
n
3-EB1, EB2
n
n
ii
n
it
n
ti
»
4-A1, A2
B1, B2
EB1, EB2
5-A2, B2
EB1
n
n
n
M
 11 = % waters X2 - % sediment; Xj = TVC water; *4 = TVC sediment; £5 = surface water tem-
 perature; l£ = salinity; Jrj = turbidity; Xg = dissolved oxygen; Xo, = total mercury concen-
 tration in sediment.  Only those variables whose partial regression coefficients were
 significantly >0 are included  (see "Significance level" column).
b, c, d
       F value for analysis of variance for multiple linear regression is significant (0.10,
       0.01, and 0.05 levels of significance, respectively).
                                            38

-------
Distribution of Types  of Mercurv-Reaigfonfl. Bacteria

Representative colonies  from platings of water and sediment on HgCl2~
and FMA-containing media were selected for further study.  Cultures were
purified and identified  to genus for each of the individual samples
examined during 1972 (Table  19).  The majority of HgCla-resistant bac-
teria were oytoohrome  oxidase positive, non-pigment ed, Gram-negative,
asporogenous rods  which  produced H2S, but did not utilize glucose or
require added  sodium ion for growth.  FMA-resistant bacteria differed
primarily in the utilization of glucose and production of diffusible
pigments.  Upon comparison of paired HgCljj-resistant and randomly
selected isolates  from medium without Hg, it was found that differences
in salt requirements and E%S production were not significant (0.05 level
of significance).

Preliminary  results, however, indicated that there were detectable dif-
ferences in  the population distributions of the mercury-resistant and
the total viable,  aerobic, heterotrophic bacterial populations (34-).
The total population is  now being more fully characterized to verify
this observation.  A total of 539 pure cultures isolated on 6 ppm
HgCl2-containing medium, collected in the period from October 1971
through December 1972, were characterized.  On the basis of our modifi-
cation of the  scheme proposed by Shewan et al.  (27),  the cultures were
placed into  one of seven generic groups or an eighth category (less than
4/6 in the majority of  cases) into which were lumped yeasts, molds, and
unidentified bacteria.   Chi-square analysis of the data (0.01  level of
significance)  indicated  that there were significant differences in dis-
tribution spatially (Fig. 12), seasonally (Fig.  13),  and between water
and sediment (Fig. 14.).  The apparent diversity of genera was greatest
at the Eastern Bay stations.  The most conspicuous seasonal changes in
population structure were in the Pseudomonas and Bacillus spp.  (Fig.
12).  Minima in percentage of the former occurred in the spring and
fall, while  a maximum of the latter occurred during the fall destratifi-
cation.  Water and sediment differed primarily in the proportions of
Pgeudopppflp  and Gram positive rods (Bacillus spp.  plus coryneform bao-
teria) .  The genus Pseudpmpnas accounted for 66# of all BgCl2-resistant
bacteria in  water  and sediment.   A much higher proportion,  93$,  of the
70 PMA-resistant isolates were Pflqudomonas spp.,  the  majority of which
were type II (27).

DISCUSSION

                      Stations
The range of mercury concentrations found in sediments of the regularly
sampled stations appears to be related to the nature of the sediments
and their locations (Table 13).  The highest concentrations were found
in the fine eilty-clay sediments located near industrial activity or
industrial plant outfalls.  There was no significant difference in the.
mercury level of the sandy sediments of Station A-1  and the silty-clay
sediments of Eastern Bay.  The difference in mercury concentrations
                                   39

-------
                       Table 19.  PHYSIOLOGICAL CHARACTERISTICS OF HgCl2- AND FMA-RESISTANT BACTERIA

A. HgClg

(6 ppm)
No. of cultures
Percent
B. FMA (3

Cytochromea
oxidase
resistant6
238
68.0
Gram-
negative

353
83.5
Green
Soluble
Pigment
Produced

55
13.0
H Sb
produced

233
55.0
Na+c
required

32
7.6
Bacto OF medium plus ^% glucose*
(aerobic/anaerobic )
Acid/ —

- 61
14.4
Acid/Acid

35
8.3
Alkaline/ —

217
51.3
	 / —

103
24.3
ppm) resistant^
No. of cultures
Percent

64
92.8
69
100.0
15
21.8
40
58.0
6
8.7
38
55.0
2
2.9
28
40.6
0
0
Method of Gaby and Hadley (33).
bF.
rom sodium thiosulfate or L-cysteine.
 Requirement for ]% NaCl for growth.

a
 Anaerobic tubes were layered with sterile mineral oil.   Dasb indicates no change in pH.


3Total of 423 isolates tested.
 Total of 69 isolates tested.

-------
         80
O  60
tr.
ui
a.

   40
         20
            I
                                                    I  2345678
                                                      EASTERN BAY
                                   I  -  -
                                  2345678
                                    ANNAPOLIS
            12345678
               BALTIMORE
                                     SITE
Figure 12.  Distribution of HgCl2 resistant, aerobic, heterotrophic bac-
            teria isolated from water and sediment of Chesapeake  Bay.
            Four hundred and eighteen pure cultures found to be resis-
            tant to 6 ppm of HgCl2 vere isolated betveen January  1972
            and December 1972.  These cultures vere classified into 8
            genera or generic groups:  1 = Pseudomonas. 2 = Vibrio  and
            Aeromonas. 3 - Bacillus. 4 = coryneform, 5 = Ovtophaga  and
            Flavobacterium. 6 = Achromobacter. Alcaljgenes. and Acineto-
            bacter. 7 = Enterobacteriaceae. and 8 - unidentified.   The
            cultures vere isolated from B-1 and B-2 ("Baltimore"),  A-1
            and A-2 ("Annapolis") and EB-1 and EB-2 ("Eastern Bay").

-------
              80
            Z
            UJ
            O 60
            K
            UJ
            CL
               40
               20
                                             12345678
                                               FALL 1972
                                            I 2345678
                                             SUMMER 1972
                    iiili,
                               12345678
                                SPRING 1972
                  12345678
                   WINTER 1972
                                       SEASON
Figure
Seasonal population distribution of HgCl2 resistant aerobic,
heterotrophic bacteria.  Five hundred and thirty-nine pure
cultures found to be resistant to 6 ppm of HgCl  were collected
from Stations B-1, B-2, A-1, A-2, and EB-1, and EB-2 during
January 1972 through December 1972.  "Winter" = 1/13/72 - 1/31/72,
"Spring" = 3/29/72 - 6/1/72, "Summer" = 7/25/72 - 8/1/72, and
•Tall" = 10/6/72 - 12/5/72.  For generic key, see Fig. 12.
                       80
                        70
                        60
                        50
                        40
                      o
                      
-------
observed for the two Baltimore Harbor stations may be related to differ-
ences in hydrocarbon content  (35), for it has been shown that mercury is
associated with the hydrocarbon fraction of sediments in Baltimore Har-
bor.  Although no attempt was made to characterize the chemical form of
the mercury found in the sediments in Chesapeake Bay, apparent contrasts
in bioavailability of mercury, as in the case of Baltimore Harbor, might
well prove to be attributable to the concentration of oil in the sedi-
ment of a given site.

Relationship of Mercury Resistance and Metabolism

We have established previously that the effect of acquisition of resis-
tance to a single mercury compound is pleiotropic, in terms of resis-
tance to other mercury compounds (15).  It was also shown that metabolic
conversion of FMA to Hg° and benzene (1), an inducible mechanism, is
associated with resistance (15).  Similar inducible systems have been
hypothesized for other organisms (11, 14).  These findings concur with
those relating to inorganic mercury and reported in this paper.  Along
this line of reasoning, the synthesis of methyl mercury, which also is
a conversion of mercury to a more mobile state, is thought to be a re-
sistance mechanism in Neurospora (35).  The apparent adaptability of
microorganisms to growth in the presence of mercury or mercury compounds
suggests a possible causal relationship between mercury in the environ-
ment and the degree of biotransformation and mobilization of mercury.
Our findings support the concept of a plate count index of potential in
sfttu mercury metabolism.  We have used concentrations of HgCl2 and FMA
(6 ppm and 3 ppm, respectively) routinely in preparing the media used
in our investigation.  These concentrations are considerably higher than
the mercury concentrations encountered in Chesapeake Bay.  However, our
observations show that concentrations as low as 1.2 ppb of HgCl2 or PMA
will inhibit a measurable portion of the total population (34).

The phenomenon of bacterial reduction of Hg2* is probably related to the
evolution of Hg° from water and sediment which can be directly measured
using fresh, raw sediment from Baltimore Harbor (F.  Brinckman and W.
Iverson, National Bureau of Standards, personal communication) and the
Patapsco River (34) in a closed system fitted with a flameless atomic
absorption spectrophotometer.  Bothner and Carpenter (19) reported that
total mercury concentrations in sediments of Bellingham Bay of Puget
Sound, Washington, decreased according to first order kinetics, follow-
ing the reduction of mercury discharges from a chlor-alkali plant.  They
found that the apparent half life of mercury in these sediments was 1.3
years.  From data published by others, they also calculated half lives
for the conversion of Hg2+ to methyl mercury to be 2.9 to 18 years.  Our
experiments, using similar concentrations of Hg and natural microbial
populations isolated without selective enrichment for mercury resistance,
also indicated a first order process (Table 17).  An average half life
of 12.5 days for Hg   in aerated suspensions was calculated.  It should
also be noted that, in most cases,  heat- or chloroform-sterilized in-
ocula also induced a significant loss of Hg.  The rate of loss observed
in our experiments should be treated as an estimate of the upper limit
                                   1*3

-------
to be expected in the natural environment, since temperatures and cell
concentrations used |n. vitro vere much higher than measured £g situ.
However, the Jin, vitro results clearly indicate a mechanism for the rapid
mobilization of mercury in an estuary.  The generation of Hg° may effect
a net loss from the environment through volatilization, since Holm and
Cox  (38) have shown Hg° added to aqueous systems to be stable for
periods up to 2 weeks.

Analysis of Seasonal Data

We have observed a seasonal fluctuation in numbers of mercury-resistant
bacteria, and presumably mercury metabolism, and have attempted to re-
late it to various physical parameters.  There are precedents for sea-
sonal variations of marine bacteria, suggesting successions of (39, 40),
but  none more apropos than Jernelov's  (4-1) observation of the seasonal
changes in numbers of mercury-methylating bacteria in fish slime.  Mul-
tiple regression analysis  (Table 18) indicated that numbers of mercury-
resistant bacteria are determined by different factors at each sampling
area.  A similar approach vas employed by Brasfield  (42) in her investi-
gation of the incidence of pollution indicator bacteria.

Percentage resistance and mercury concentrations in  sediments were posi-
tively correlated at the Annapolis  and Eastern Bay stations, although
not  significantly at Annapolis.  In contrast, a negative correlation  vas
observed for  samples collected  in Baltimore Harbor, vhere Colgate Creek
sediment contained higher  levels of both Hg and petroleum.  If mercury
concentration is sufficient to  specify bacterial population structure,
then differences in the mercury-resistant populations of similar locales
might reflect the proportion of total mercury which  is biologically
available.

Stations for  comparative analysis were selected on the basis of similar-
ities in sediment type and differences in mercury content.  On this
basis, data for Stations A-1 and EB-2  (sandy sediments) and B-1 were  de-
leted from the  regression  equation. Percentage of mercury-resistant
bacteria in the water column was found to be related to percentage of
mercury-resistant bacteria in the sediment, suggesting a common deter-
minant, most  likely the mercury occurring in the sediments.  The per-
centage of mercury-resistant bacteria  in sediments was found also to  be
correlated with water transparency  and dissolved oxygen.

A peak in sesten levels  in the  upper Chesapeake Bay  coincides with the
 spring run-off from the  Susquehanna River  (32, 4-3).  Smith has reported
that most  of  the mercury in estuarine  water is associated with the par-
ticulates  (44).  We have observed a negative correlation between water
turbidity and presumptive J.R situ bacterial mercury metabolism, which
may correspond to a decreased availability of mercury to the bacterial
population.   This process may be related to the adsorption of Hg to par-
ticulates.   In fact,  the binding of mercury and other heavy metals to
 organic suspended particulates, is  the basis for their removal from
water during sewage treatment (45,  46).   Thus, the  occurrence of a peak

-------
in bacterial mercury metabolism in the spring is inconsistent vith the
elevated suspended sediment load normally present during that season.
However, the lack of coincidence of the two peaks may explain the ap-
parent paradox.  In contrast, the elevated levels of dissolved oxygen
during the spring season, in conjunction with organic complexing agents,
may increase the solubility of Hg (23, 24).

Although analysis of variance for the multiple linear regressions for 3
stations and 6 stations gave significant F values (0.01 and 0.05 levels
of significance, respectively), the multiple correlation coefficients
(0.74 and 0.88, respectively) indicated that the parameters which were
measured could not account for all of the variation in numbers of resis-
tant bacteria.  There are other seasonal influences in Chesapeake Bay
which deserve mention but could not be considered in this study for lack
of time, personnel and funds.  For example, in Chesapeake Bay there is
no spring phytoplankton bloom (47), but total nitrogen and zooplankton
populations do peak during the spring months (32, 43).

The observed relationship between mercury-resistant bacterial popula-
tions and mercury concentration in sediment is compatible vith the con-
cept of bacterial "adaptation."  However, because of differences in
either the chemical form or the availability of mercury and the seasonal
fluctuations in mercury-resistant bacterial population levels, a clear
and unequivocal relationship cannot be established.   That there is, in-
deed, a spring peak in the mercury-resistant bacterial population was
confirmed by the monitoring data obtained during the first half of the
second year.  Unfortunately, because of the plan of work which had been
laid out, the field work had to be terminated before the secondary fall
peak could be monitored.  The secondary peak in the mercury-resistant
bacterial population which was observed in the Fall of 1972 may have
been a consequence of the effects of tropical storm, Agnes, which occurred
in June 1972.  Correspondingly, data for the months preceding and follow-
ing the storm suggest that there were significant, temporary changes in
the bacterial flora following the storm (15).

Population Distribution

Using a taxonomic scheme similar to that employed in this study,
Murchelano and Brown (40) obtained a seasonal distribution of hetero-
trophic bacteria in Long Island Sound which was mucn like that reported
here for mercury-resistant bacteria in Chesapeake Bay.  Total viable
counts reached ni-in-! ma in the summer in both cases, and Pseudomonas spp.
were predominant in Long Island Sound samples (40.6$), as in the case
for mercury-resistant bacteria in Chesapeake Bay (66.0$).  Our results
and those of an earlier study of bacteria found in Chesapeake Bay water
and sediment (49) suggest that the mercury-resistant bacterial popula-
tion is richer in Fseudomonas spp. relative to the total population.
Yet, the total number of Pgeudomonas spp. did not correlate with seasonal
mercury resistance (see Figs. 9, 10, and 13).  However, when the Pseudo-
monas spp. were separated into physiological sub-groups, according to the
method of Shewan et al. (27), it was evident that the proportion of
                                     1*5

-------
non-glucose-utilizing type III organisms correlated vith average season-
al resistance (0.76 correlation).  Interestingly, the tvo most metabol-
lically active, Hg2* reducing strains assayed in a survey of cultures
(Table 14) vere both type III Pseudomonaa spp.  In contrast, glucose
oxidizing Pseudoiaonaa spp. types I and II declined in abundance from
winter through the spring, summer and fall months.

Mercury mercaptide formation, via metabollically-generated sulfhydryl
compounds has been shown to be a mechanism for mercury resistance in
some microorganisms (50, 52).  However, our data show that HgS produc-
tion is not an exclusive property of mercury-resistant bacteria, hence
could not account for the level of mercury resistance in the bacterial
populations studied.  This particular point was carefully examined in
order to establish the validity of a plate count index for potential
mercury metabolism.  Our conclusion was further supported by the fact
that the seasonal distribution of HgS-producing bacteria did not coin-
cide with that of average mercury resistance in the viable, aerobic,
heterotrophic populations in Chesapeake Bay water and sediment.

PMA was found to be considerably more toxic for the bacteria in Chesa-
peake Bay samples than HgCl2.  At 3 ppm, it was also more selective for
pqeiyfenfqiyfrfl spp.  When the FMA concentration was reduced to 0.3 ppm,
inhibition of the total population was found to be comparable to that
induced by the presence of 6 ppm of HgCl2 in the plating medium.  The
relatively greater toxicity of organomercurials for microorganisms is a
general phenomenon  (34, 51, 52) and is very likely related to the lipo-
philic character of these compounds.

In summary, an ecological analysis of the distribution and function of
mercury-resistant bacteria in Chesapeake Bay has been accomplished.  The
seasonal fluctuations in the numbers of these bacteria has been demon-
strated and some of the influencing environmental factors elucidated.
The cyclic oscillation observed in that physiological group of micro-
organisms comprising the mercury-resistant bacteria may be only a single
component of a species succession.  Work underway on the species compo-
sition of the total aerobic, heterotrophic bacterial populations in
Chesapeake Bay should provide some further insight into this fascinating
and heretofore undescribed aspect or bacterial population dynamics in
nature.

SUMMARY

Total ambient mercury concentrations and numbers of mercury-resistant,
aerobic, heterotrophic bacteria at six locations in Chesapeake Bay were
monitored over a 17 month period.  Mercury resistance expressed as the
proportion of the total, viable, aerobic, heterotrophic bacterial popu-
lation, reached a reproducible maximum in spring and was positively
correlated with dissolved oxygen concentration and sediment mercury con-
centration and negatively correlated with water turbidity.

A relationship between mercury resistance and metabolic capability for

-------
reduction of mercuric ion to the metallic state vas established by sur-
veying a number of HgClg-resistant cultures.  The reaction was also ob-
served in microorganisms isolated by differential centrifugation 'of
water and sediment samples.  Mercuric ion exhibited an average half life
of 12.5 days in the presence of approximately 1Cr organismsAl.

Cultures resistant to 6 ppm of mercuric chloride and 3 ppm of phenylmer-
curic acetate  (PMA) were classified into eight generic categories.
Paeudomonas spp. vere the most numerous of those bacteria capable of
metabolizing both compounds; however, PMA was more toxic and was more
selective for Pseudomonas.  The mercury resistant generic distribution
was distinct from that of the total bacterial generic distribution and
differed significantly between water and sediment, positionally, and
seasonally.  The proportion of non-glucose-utilizing mercury resistant
Pseudomonas spp. was found to be positively correlated with total bac-
terial mercury resistance.

It is concluded from this study that numbers of mercury-resistant bac-
teria as established by plate count can serve as a valid index of in
situ Kg** metabolism.

-------
                              SECTION VI

        EFFECTS OF MERCURIC CHLDRIDE UPON GROWTH AND MORPHOLOGY

           OF SEIECTED STRAINS OF MERCURY-RESISTANT BACTERIA
INTRODUCTION

Preceding sections of this report have included experimental evidence
linking bacterial mercury resistance vith %& situ generation of Hg°.
The positive correlation observed between numbers of aerobic, hetero-
trophic bacteria resistant to given levels of mercury under laboratory
conditions and ambient mercury concentration found in Chesapeake Bay
sediments suggests that environmental mercury contamination can exert
an effect upon the population structure of the estuarine microflora.
To elucidate the mechanisms by which mercury manifests selective pres-
sures under natural environmental conditions, ve examined the effects of
inorganic mercury upon growth and morpholoty of representative strains
of bacteria isolated from Chesapeake Bay.

Investigations of mercury resistance in microorganisms have revealed
several possible mechanisms for this resistance.  Phytoplankton (53) and
various bacterial strains (9, 12, 54-) are capable of adapting to growth
in the presence of mercury.  In some bacteria, resistance is attained
through the acquisition of plasmid, an extra chromosomal element which
can be transferred intra- and inter-generically (5, 6, 7, 8, 9).  Micro-
organisms detoxify mercury metabolically by formation of volatile Hg°
(9, 12, 53, 54-j 55), or of mercury mercaptides  (50, 51).

Comparatively little information is available concerning the morphologi-
cal effects of mercury upon microorganisms.  Tingle et al. reported
mitochondrial damage and swelling, pellicular membrane disruption, and
loss of motility in Tetrahymena pyriformis incubated in a solution of
0.5 ppm of HgCl2 (56).Exposure of cells of Pseudpmonas aeruginosa to
HgCl2 caused swelling which was reversible by addition of sulfhvdryl
compounds  (57).  Also, extremely low levels of PMA (0.9 - 3 ppb) causing
gross changes in the cellular form of growing cultures of Phaeodactylum
tricornutum and ChoreIjUi have been described by Nuzzi (52).

The research work described in this report was carried out using cul-
tures of bacteria isolated from Chesapeake Bay.  In addition, the levels
                                   1*8

-------
of mercury employed in the study were selected so that correlation of
results obtained with is situ conditions in Chesapeake Bay could be
achieved.
MATERIALS AND METHODS

Organisms and
Organisms were cultured in an artificial estuarine salts nutrient broth
and agar medium described previously.  (See Section V. )  Cultures used
in the study were isolated by spreading suitable dilutions of water and
sediment on media supplemented with varying concentrations of HgClp or
PMA.

Electron Microscopy

Cells in late log phase cultures were transferred to fresh broth (0.1
ml/ml) in shaker flasks with and without sublet ha 1 concentrations of
HgCl2 added and were incubated at 25 C for 3 to 4- generation times, or
until equal turbidities were obtained during early logarithmic phase of
growth.  The cultures were fixed according to the procedure described by
Kellenberger and Ryter (53).  Ten ml of culture was washed once in bar-
bituric acid (K-R) buffer (pH 6.2), following which the cells were fixed
in 1.0$ osmium tetroxide in K-R buffer and prestained in 0.5% uranyl
acetate, in K-R buffer, prior to stepwise dehydration through a 50$,
75/6, 85%, 95% , and 100$ ethanol and propylene oxide series.  The speci-
mens were embedded in Epon resin (Miller-Stephenson Chemical Co., Inc.,
Danbury, Conn.) (59), and ultra-thin sections were prepared using an 1KB
800 Ultramicrotome III (LKB-Producter AB, Bromma 1, Sweden), fitted with
a diamond knife (E. I. DuPont de Nemours and Co., Wilmington, Del.).
The sections were placed on collodion-coated 200 and 300 micron mesh
size copper grids and stained with a saturated solution of uranyl ace-
tate (60) and lead citrate (61).  An RCA EMU-3E or Hitachi HU-11A micro-
scope (accelerating voltage of 50 kv) was used to view the sections.
Culture 85 , grown without mercury, was added to shaker flasks containing
broth (0.1 ml inoculum/5 ml of broth).  The first, i.e.. control flask
contained no mercury, the second flask, 2 or 4 ppm of 20^Hg-labeled HgCl
(approx. 200,000 dpm/ml).  The flasks were incubated at 34. C (first ex-
periment) or 25 C (second experiment) and optical densities relative to
an uninoculated ^^Hg-labeled blank control were measured using a KLett-
Sommerson Photoelectric Colorimeter (Klett Manufacturing Co., New York,
N.Y.).  Aliquots of cells were removed aseptically from the unlabeled
medium and fixed for electron microscopy or diluted and plated on
nutrient agar for viable counts.  Duplicate 0.1 ml volumes of culture
were removed at appropraite intervals for determination of total and
non-filterable radioactivity by liquid scintillation.

-------
Assay for Hg° Production

The method of assay using 20^Hg-labeled HgClg was as previously described.
The assays vere carried out at 25 C, using 1 ppm of HgCl2.

RESULTS AND DISCUSSION

Survey of Morphological Effects

Representative cultures of bacteria isolated from colonies on agar media
containing HgCl2 or FMA vere grown in broth containing sublethal concen-
trations of HgCl2 in shaker flasks incubated at 25 C.  The experimental
cultures and controls without HgCl2 vere grovn to early log phase and
harvested.  Thin sections vere prepared from fixed and stained specimens,
and the preparations were examined by transmission electron microscopy.
Grovth in the presence of mercury commenced after varying lengths of
time, with an array of morphological effects which did not appear in con-
trol cultures of the same physiological age (Table 20).  The most fre-
quently observed defect involved the cell wall and cell wall synthesis.
This was suggested by the formation of projections from the cell wall,
elongated pleomorphic cells of the Gram-negative bacteria examined, and
irregular cross-wall formation in the Gram-positive bacterial strains.
A number of cultures also showed plasmolysis, indicating interference
by the mercury with cytoplasmic membrane transport phenomena.  These
effects may be related to the finding that most of the mercury bound to
bacterial cells has been found either on the cell wall or cytoplasmic
membrane  (this report, 10, 62).

Growth and Morphology of Representative Gram-positive and Gram-negative
Species

Figure 15 shows growth of Bacillus strain 394 in the presence and absence
of 5 ppm HgCl  .  After a one day delay  ("lag") in the onset of growth,
the cells were able to undergo logarithmic growth at a normal rate.  A
similar lag phase has been observed in the pattern of growth of other
microorganisms in the presence of mercury  (5, 9, 12, 53, 54, 63).

-------
Table 20.  SURVEY OF MORPHOLOGICAL EFFECTS IN SELECTED CULTURES OF BAC-
           TERIA GROWN IN THE PRESENCE OF HgCl2a
       Culture
Genus
               Strain
concentration
    (ppm)
 Lag
phase
 (h)
Morphological effects*
Pseudomonas       94

Citrobacter      132


Flavobacterium   119
                              10
                              20
                              10
Vibrio
                 639
  24     Cell wall outgrovth

  70     Plasmolysis, damaged
         membrane

  40     Cell vail outgrovth,
         plasmolysis, elongated
         cells, irregular meso-
         somes

  72     Giant, pleomorphic and
         rod-shaped cells,
         elongated cells
Bacillus
Arthrobacter
Enterpbac^er
Pseudomonas
Pseudomonas
394
72
85
B-l6d
244
15
5
1
15
25
41
24
0
6
days
0
Irregular septum forma-
tion
Irregular septum forma-
tion
Pleomorphic cells, out-
grovth of cell vail,
plasmolysis
Small spherical cells
bounded by common
outer membrane
None
a
 Selected cultures vere inoculated into shaker flasks of broth vith and
 vithout HgCl2 added and vere incubated at 25 C until early log phase.

b
 Cultures vere isolated and purified from colonies picked from mercury-
 containing nutrient medium inoculated vith water and sediment samples
 collected in Chesapeake Bay and the Potomac River.

<;
 Fixed and stained thin sections of cells grovn, vith and vithout HgCl2,
 to the same cell density vere compared.

d
 Supplied by R. A. MacLeod.
                                   51

-------
                   0.9
                   0.8
                   0.7
                   0.6
                   0.5
                  o
                  h-
                   0.4
                  ,0.3
                   0.2
                   0.
                                10          20

                              INCUBATION TIME (hr.)
                                                      30
Figure 15.  Growth of Bacillus, species 394-, in the presence and absence
            of 5 ppm of HgClg.  Broth in  shaker flasks with (A) and
            without (•) HgCl2 was inoculated with overnight culture
            (0.1 ml/5 ml fresh broth).  Growth was measured turbidimetri-
            cally during growth at 25 C.  Samples were removed from each
            flask (arrows) for the morphological analysis of stained
            thin sections  (see Figs. 15 and 16).
These cells showed irregular cross-wall formation and grossly deformed
outer cell walls  (Fig. 16).  The  effects were noticeably increased at a
concentration of  15 ppm HgClg  (Fig.  17).

A Gram-negative species was selected for a more  detailed investigation
since Gram-negative organisms  constitute 90$ of  the mercury-resistant
bacteria isolated from Chesapeake Bay in the course of the  study.  The
objective, in this instance, was  to  clarify  events occurring while bac-
teria were apparently quiescent in mercury-containing broth and to relate
these events to cytological changes  associated with adaptation.  Entero-
bacter. strain 85, was serially transferred  several times in the absence
of mercury, prior to inoculation  into shaker flasks with and without 2
ppm of 20^Bg-labeled HgCl2.  After a 24 h lag phase, the mercury-contain-
ing culture grew  at a rate comparable to that of the control culture
(Fig. 18).  Although the turbidity of the culture remained  constant, a
nearly 3-log drop in viable count occurred.  Viable count and turbidity
increased at the  end of the lag phase.  An isotope effect was suggested

-------
 Figure 16.  Thin section of cells of culture #394 grown in the presence
             of 5 ppm of HgCl2.  In comparison to cells grown without
             mercury, irregular septa were formed (arrow).  Scale in
             this and subsequent micrographs is one micrometer.
Figure 17.  Thin section of cells of culture #394 grown in the presence
            of 15 ppm of HgCl2.  Cells show irregular cell division and
            pleomorphism.

                                  53

-------
by the slightly shorter lag phase in the unlabeled mercury-containing
culture.  During the initial phase of incubation, radioactivity was
rapidly lost from both the inoculated broth and from the sterile control.
                       100
                                      6 ' ' 24   26   28   SO
                   1.0
 Figure 18.   Growth and uptake and metabolism of mercuric ion by Entero-
             bacter. strain 85.  Shaker flasks containing broth with and
             without 203Hg_ and un-labeled HgCl2 (2 ppm) were inoculated
             with an 18 h culture and incubated at  34 C.  Turbidity of
             the culture without mercury (•) and the culture with labeled
             mercury (A) were measured.  Total (A)  and non-filterable
             (•) activity of the latter were also  measured.  An uninocu-
             lated,    Hg- labeled control flask (O)  was also assayed for
             total radioactivity.  Aliquots were removed from flasks
             without and with unlabeled mercury at  intervals for examina-
             tion by electron microscopy (arrows).  Total viable counts
             of the unlabeled mercury culture (Q)  were obtained by
             spreading suitable dilutions upon basal  nutrient agar.
 Other investigators have shown that reducing agents such as  glucose  (64)
 and yeast extract (63) in growth media promote the reduction of Hgz  .
 The loss of radioactivity from the control broth represented the vola-
 tilization of Bg as Hg° and did not arise from adsorption to the glass

-------
vails of the flask since only 8.9/6 of the radioactivity was recovered in
sequential washes of the vessel using water, 0.5 N HN03, and 10$ Radiac
wash (Atomic Products Corp., Center Moriches, N. T.).  An initial, small
loss of cell-bound mercury coincided with the drop in viable count, in-
dicating lysis of the cells may have occurred.  When the mercury concen-
tration dropped approximately 4-5.056, to 1.1 ppm, growth was initiated.
At this point, the rate of loss of Hg from the broth accelerated in the
inoculated flask, and the rate of cellular uptake, per unit volume of
broth, increased.  The possibility that the former effect resulted from
biological reduction of Hg^"*" was subsequently investigated.

After several hours of incubation, small colonial variants (85-S) of the
HgCl2-containing culture began to appear on the spread plates prepared
for determination of total viable counts (Fig. 19).  These colonies con-
tained non-motile, pleomorphic cells, in contrast to the large colonies
(85-L) comprising the remainder of the population.
Figure 19.  Colonial variation in Enterobacter. strain 85, grown in the
            presence of 1 ppm of HgCl2-  Samples of culture grown in 2
            ppm unlabeled HgCl2 containing broth were diluted and plated
            on basal nutrient medium  (Fig. 17) and incubated at 25 C
            for 7 days.  Small (85-S) and large (85-L) colonial forms
            were observed.
                                    >

-------
The experiment was repeated using  an elevated concentration of mercury
(4 ppm) to test the hypothesis that  the  length of the lag phase demon-
strated by this organism was dependent upon reduction of mercury con-
centration to a specific threshold concentration.   The patterns of
grovth and loss of mercury vere  as observed in the first experiment,
except that the lag phase vas found  to be almost  three days (Fig.  20).
                100?
                                                         10.0
                                                         4.0  O
                                                        ' 2.0
                            20   30   40    50    60
                              INCUBATION TIME (hr.)
                                                    70
Figure 20.  Growth and uptake  and metabolism of mercuric  ion by Entero-
               ter.  strain 85.   Shaker flasks containing  broth with
                  -  and  un-labeled HgClg (4- ppm)  were inoculated with an
            18 h culture and  incubated  at  25  C.   Turbidity (A) and total
            (A)  and non-filterable  (•) radioactivity in the  labeled
            culture were measured.   An  uninoculated,  labeled control
            flask (D)  was also  assayed for total radioactivity.  Total
            viable  counts  (Q) axi^ numbers of gmaii  colonial variants
            (85-S,  Fig. 19) (•)  were obtained by spreading suitable di-
            lutions of  unlabeled mercury containing  medium upon basal
            nutrient agar.
However, during the lag phase, the mercury concentration dropped 70$, to
a level of 1.2 ppm, a concentration closely approximating the concentra-
tion of 1.1 ppm HgCLj concentration at which growth commenced in the
                                    56

-------
first experiment.  An increased isotope effect vas observed in the second
experiment, vhere the difference betveen the lag phase for labeled and
unlabeled cultures was nearly 10 hours.  Again, a drop in viable count
vas accompanied by a drop in cell-bound Hg.  As in the first experiment,
the rate of loss of Hg from the inoculated flask accelerated at the end
of lag phase.  Numbers of small colonial variants comprised approximately
1% of the total viable count, and these small colony variants increased
in parallel vith the total population (Fig. 20).

Fixed and stained thin sections prepared from cells harvested, in the
first experiment, from unlabeled broth vith, and vithout, mercury vere
examined.  Control cells examined at the early, mid, and late logarithmic
phase of grovth vere found to be normal in all respects, viz., array and
morphology of the DNA fibrils, ribosomes and cell envelope, as veil as
mode of division.  At 0 h, the mercury-containing culture vas found to
be identical to the 0 h control (Fig. 21).  After 2 h (Fig. 22), the
cells appeared normal in overall morphology.  The DNA fibrils in some
cells appeared condensed, rather than dispersed throughout the cytoplasm,
and electron dense areas among the ribpsomes vere seen.  Intact 6 h
cells (Fig. 23), resembling 2 h cells, vere accompanied by svollen and
plasmolyzed cells.  Approximately one-third of all the cells at this
stage shoved signs of lysis.  At 2k h (Fig. 24), the culture vas char-
acterized by debris arising from lysed cells, as veil as viable but
pleomorphic cells demonstrating abnormal cross-vail formation.  Electron
dense clusters of ribosomes vere more frequently seen in these cells.
As grovth progressed through the 26th and 28th hours, the cells became
relatively more rod-shaped, although irregular cell vails and electron
dense ribosome clusters vere still present (Figs. 25, 26).  The latter
clusters are consistent vith the preferential intracellular binding of
Hg, but the technique of electron probe microanalysis, vith sufficient
resolution, would be required to clarify this point.

Properties of Mercury—selected C|i1.tiiT**>g

After repeated serial transfers in mercury-free medium, the small colo-
nial variants lost their pleomorphism and regained motility, but retain-
ed their colonial morphology.  II' the mutants resulted as an "adaptation"
to mercury, they should differ from the parent stock, vith respect to
mercury resistance and/or the ability to reduce Hg2+.  Comparative data
on mercury resistance and ability to produce Hg  from HgCl2 of strains
85, 85-L and 85-S are presented in Table 21.  It vas evident that the
isolates previously grovn in mercury, specifically strain 85-S, vere
more resistant to mercury and demonstrated increased capability to me-
tabolize Hg2+.  Thus, it appears that, during the lag phase of growth,
mercury-resistant mutants are selected vhich are capable of releasing
Hg° from the grovth medium.  It is probable that these cells are iden-
tical to the pleomorphic, but viable, cells observed in thin sections
of those cells examined after 24 h.  When the mercury concentration vas
reduced to ca. 1 ppm via chemical and biological reduction of Hg^"1", the
entire population initiated grovth at a normal rate.
                                   57

-------
 Figure 21.   Thin section of culture #85  cells  immediately after addition
             to broth containing 2  ppm of HgC.l2«   The  cells show dispersed
             chromatin,  normal ribosomes  and  cell envelope constituents,
             as -well as  typical gram negative cell division by constric-
             tion.

Figure 22.  Thin section of culture #85 cells after 2 h incubation in
            broth containing 2 ppm of HgCl .  Similar in appearance to
            zero time, except for condensed chromatin, and the appear-
            ance of electron dense groups or ribosomes (arrow).


-------
Figure 23.  Thin section of culture #85 cells after 6 h incubation in
            broth containing 2 ppm of HgCl2-  The most prominent cyto-
            logical features are plasmolysis and lysis in a large por-
            tion of the cells.  Arrows indicate electron dense groups
            of ribosomes.
                                   59

-------
Figure 24.  Thin section of culture #85 cells after 24 h incubation in
            broth containing 2 ppra of HgClg.   Viable, pleomorphic cells
            and fragments of lysed cells are seen.   Cellular division
            occurs by cross -wall formation (arrows), rather than by nor-
            mal constriction.  Numerous electron dense groups of ribo-
            somes are also seen.
Figure 25.  Thin section of culture #85 cells after 26 h incubation in
            broth containing 2 ppm of HgCl2.  Predominately viable, pleo-
            morphic cells are seen.  Electron dense clusters of ribosomes
            are present in all cells.
                                   60

-------


Figure 26.  Thin section of culture #85 cells after 28 h incubation in
            broth containing 2 ppm of HgCl2.  The cells show reversion
            to rod-shaped morphology, although cell wall contours are
            still somewhat irregular.  Electron dense clusters of ribo-
            somes are present in fewer numbers.
                                   6

-------
Table 21.  MERCURY. RESISTANCE AND MERCURY. METABOLISM OF STRAIN 85 BEFORE
           AND AFTER GROWTH IN THE PRESENCE OF
             A.  Mercury resistance"
                                  HgClg concentration (ppm)
             Culture0	24      6      8      10
               85              +      +
               85-L            +      +      +      _

               85-S     .       +      +      +      +       +

             aSee the first experiment (Fig. 18).

              The cultures were serially subcultured in the
              absence of mercury.  Each tube of mercury-
              containing broth was inoculated with 3 drops of
              a log phase culture and incubated, with shaking,
              at 25 C for 4 days.
             c
              Culture 85 was the original inoculum, strain
              numbers 85-L and 85-S refer to small and large
              colonial types isolated during incubation of the
              HgCl2 containing broth.  (See Fig. 19.)
                    a
B.  Formation of Hg°
Percent decrease in 20%g radioactivity/60 m1n
Culture
85
85-L
85-S
Dry weight
(ing)
2.05
1.50
1.90
Live
suspension
9.8
7.1
11.3
Heat killedb
suspension
16.1
3.6
3.0
Net biological
reduction0
0
3.5
8.3
&The assay was carried out at 25 C, with aeration, as described in Materi-
 als and Methods.  The assay contained 1 ppm of 20%gCl2 (approx. 120,000
 dpm/ml).

 Sterile inocula were prepared by autoclaving at 121 C for 15 min.

°The difference in percent decrease between live and sterile inocula.
                                    62

-------
The 4a vitro experiments carried out in this study may be related to con-
ditions in the natural environment.  The positive correlation observed
between mercury concentration and number of mercury-resistant bacteria
in sediment suggests that selection does occur J£ situ.  It is logical
to propose that despite adverse effects of mercury or other heavy metals,
microorganisms will prevail, in some form, because of their remarkable
versatility.  However, it is not yet possible to predict, quantitatively,
the effects such a perturbation would have upon population structure in
a natural ndcrobial community or at the higher ecological level.
                                 63

-------
                              SECTION VII

            THE ROLE OF BACTERIA IN THE MOVEMENT OF MERCURI

                    THROUGH A SIMPLIFIED FOOD CHAIN
INTRODUCTION

Whether mercury levels in the oceans have risen significantly is a high-
ly controversial issue (65, 66, 67).  Undisputable, however, is the fact
that marine animals are capable of concentrating mercury and other heavy
metals to levels several orders of magnitude above that found in sea
water GO, particularly in the proximity of known sources of pollution
(3, 68).  Regardless of the source, the consequences of the sequestra-
tion of mercury are serious because the highly toxic and persistent
alkylated forms of mercury comprise most of the mercury in fish tissue
(22, 69, 70).  The mechanism(s) by which mercury is accumulated in higher
organisms are not completely known.  Barber et al. found no apparent re-
lationship between mercury concentrations in marine fish and ambient sea
water levels of mercury (66).  Similarly, Windom et al. (71) found no
significant differences in mercury concentrations in inshore and offshore
catches of 35 species of North Atlantic fin fish.  However, the amount
of mercury accumulated in fish appears to be related both to the size
(66, 70) and the species of fish involved (66).  It is not known, as yet,
whether given species of fish, such as tuna or swordfish, acquire rela-
tively large amounts of mercury by unique mechanisms, which are species
specific, or from participation in, or association with, a particular
mercury-accumulating food web  (72).  Evidence both for (73, 74) and
against (4, 69, 75) the magnification of mercury concentrations through
trophic levels has been presented.  The question has not been satisfac-
torily answered since none of the investigations were specifically
undertaken to trace mercury flow through a specific food chain, under
controlled conditions.  Mercury uptake has been studied in individual
marine animals such as the oyster (69, 76) and the fiddler crab (77),
but in neither case were the microbiological parameters of the experiment
measured or controlled.

We have investigated the ecology of mercury-resistant bacteria in Chesa-
peake Bay and have concluded that bacteria, by virtue of their ability
to accumulate mercury and reduce Hg2+ to Hg°, are influential in the
mobilization and transformation of mercury in the Chesapeake Bay habitat.
Consequently, an investigation of the role of bacteria in the

-------
introduction of mercury into a simple estuarine food chain was initiated.
Based upon a comparative survey of benthic and pelagic animals in Hawai-
ian waters, Klemmer and luoma (73) concluded that mercury is probably
most readily transported through short food chains directly linked to
benthic organisms.  After considering the available information and the
fact that mo Husks are known to accumulate mercury and other heavy
metals (78), we selected for our study a food chain incorporating bac-
teria and a benthic filter feeder, the American oyster, Crassostrea
virginioa.  A series of experiments were conducted to elucidate the
routes and mechanisms of mercury introduction into aquatic food chains
at the lowest trophic level.

MATERIALS AND METHODS

Aquarium System

The apparatus used in each of the three experiments which were carried
out consisted of a flanged, 14 liter pyrex fermenter jar equipped with a
stainless steel lid which was sealed with a rubber 0-ring and fastened
with bolts (Pig. 27).  The lid had threaded nylon hose fittings wrapped
with teflon tape to prevent leaks.  A rubber serum bottle cap was placed
on the central fitting to permit sampling of the closed system with a
sterile 12 inch canula attached to a sterile syringe.  A rack constructed
from a glass rod was used to suspend the oysters off the bottom of the
jar, above the teflon-coated magnetic stirring bar.  A gas dispersion
tube was connected to an outside air pump via Tygon tubing and a nylon
hose fitting in the lid.  The entire aquarium, complete with fiber glass
air filter, was sterilized by autoclaving at 121  C for 15 min.  Ten
liters of 0.45 micron filter-sterilized, dilute,  artificial sea water
(Seven Seas Marine Mix, Utility Chemical Co.,  Paterson,  N.  J., 33% of
sea water strength) were pumped in,  and the aquarium was immersed in a
15 C constant temperature water bath.  Circulation was provided by an
external, submersible magnetic stirrer powered by water, and the water
in the jar was oxygenated by air pumped through the fiber glass filter
to the submerged gas dispersion tube.  Effluent air from the aquarium
was transported through Tygon tubing to two scrubbers in series,  each
containing 100 ml of HgBr2 - KBr solution (see Section V) to trap vola-
tilized mercury (Fig. 28).

Oysters and Bacteria

Oysters were dredged from Tolly Bar outside of Annapolis Harbor,  Maryland,
in March and April 1974, and maintained in raw bay water at  in situ tem-
perature and salinity until used in the experiments.   The animals were
scrubbed and placed in artificial sea water (approximately 12.0 °/oo
salinity) at 6 C for approximately 1  week.   The temperature  of the water
in which the oysters were held was brought up  to 15 C gradually and was
held for 3-4 days at 15 C prior to placing them in the sterile aquarium
system.  Bacterial cultures were grown in liquid basal medium (see pre-
vious sections) Containing 6 ppm of HgCl2«   Fresh transfers  of overnight
cultures were grown for 3-5 h, centrifuged and resuspended in sterile
salts solution, and added to the aquarium.   Total viable bacterial counts


                                    65

-------
Figure 27.  Sterilizable oyster aquarium.  Sampling port with rubber
            septum (1); air inlet with sterile fiber glass filter (2);
            air outlet (3)j water effluent (4.); magnetic stirring bar
            (5); gas dispersion tube (6).
of aquarium water were determined periodically by spreading suitable di-
lutions of the water on solid basal medium and incubating the plates at
room temperature for 2 weeks.

Experimental Procedures

Six oysters of uniform size were selected for each experiment.  Two were
shucked and freeze dried for total mercury analysis*  The remaining four
were placed in the sterilized aquarium.  Bacterial suspensions were
added through a port in the lid, followed by 1.0 ml of sea water containing

-------
Figure 28.  Oyster aquarium in operation.  HgBr£ - KBr traps  (1); imraer-
            sible, water propelled magnetic stirrer  (2).


-------
10-20 micro C of 2HgCl2 to give a final concentration of 10.2 ppb.
After an equilibration period of 10-15 min, duplicate 1 ml samples of
aquarium water and scrubber solution vere removed for scintillation
counting.  A second 1 ml sample of aquarium water was filtered through a
0.45 micron membrane filter, which was rinsed with three 1 ml volumes of
artificial sea water.  The filters were saved for scintillation counting.
The above sampling procedure was repeated on four successive days.  At
the conclusion of the experiment, the oysters were removed from the
aquarium, shucked, and dissected (Fig. 29).  Tissues were rinsed in arti-
ficial sea water and placed in tubes for scintillation counting.  Fecal
material which had settled to the bottom was aspirated from the aquarium
and collected on 8 micron membrane filters and rinsed in artificial sea
water prior to scintillation counting.  Radioactivity was measured in a
Packard TRICARB Spectrometer equipped with an Auto-Gamma Spectrometer.
The standard deviation was 1$ of total counts for all oyster tissues.
Figure 29.  Oyster dissection.  Oysters were shucked, and the mantle fluid
            was drained and retained.  The oysters were dissected, and each
            of the tissues selected for analysis was rinsed in artificial
            sea water prior to measurement of radioactivity.  Mantle,
            gills + labial palps, visceral mass, and adductor muscle (left
            to right).
                                    ..;

-------
RESULTS AND DISCUSSION

Three experiments, each using four oyster replicates per experiment, were
completed (Table 22).  The oysters were harvested at the same time of
year from the same location.  There was some variation in size, but it
was determined that there was no correlation between size and the amount
of mercury accumulated.  Total mercury concentrations in the oysters
prior to each experiment were nearly equal to the initial concentration
of mercury in the aquarium water (Table 23).  The significantly greater
concentrations of mercury in treated oysters indicated that tracer uptake
was truly a measure of accumulation and not exchange.

The purpose of the first experiment was to establish a base line for mea-
suring the effect of Hg-resistant bacteria on Hg uptake by the oysters.
In the second experiment, culture #14 (see Section V), which accumulates
Hg (12 u gm of HgClg accumulated/mg of cells/15 min from a solution of
6 ppm HgCl2) was added.  In experiment three, culture #5 (see Section V),
which actively reduces Hg2+ to Hg° (25.4- ug of HgCl2 reduced to Hg°/mg of
cells/15 min from a solution of 6 ppm HgCl2) was added.   In experiments
II and III,  a quantity of cells approximately equal to that of the mixed
flora in the aquarium was added.
                      Table 22.   EXPERIMENTAL OUTLINE

                	Oyster tissue wet weight (gm)a	
                Mantle                    Visceral   Adductor
   Experiment   fluid    Mantle    Gills     mass      muscle    Total
Ib
11°
IIId
12.3
12.3
15.3
5.7
3.2
2.7
5.0
2.3
1.9
7.6
5.0
4.1
3.6
1.6
1.2
34.1
24.4
25.2
   Q
    Average of four oysters.


    No bacteria were added.   Total viable bacterial count  increased  from
    3.4 x 1oVml to 2.5 x 105/ml after 4 days.

   £
    Culture #14 added (1.2 x  10? cells).   Total viable bacterial count
    increased from 7.2 x 10^  to 9.6 x 10* after 4 days.


   dCulture #5 added (1.2 x 109 cells).   Total  viable bacterial count
    increased from 1.0 x 1oVml t° 1.4 x  10^/ml after 4  days.
                                   69

-------
  Table 23.   AVERAGE TOTAL MERCURY. CONCENTRATION IN UNTREATED OYSTERS6
Experiment
I
I
II
II
III
III
Total mercury
Dry weight (ppb)b
324. ± 23
97 ±14
201 ± 6
103 ± 25
239 ±99
111 ± 17
concentration
Wet weight (ppb)c
48.6
14.5
30.2
15.5
35.8
16.7
Methyl mercury
spike recovery (56)
100
109
76
83
100
90
aTwo oysters vere selected from each experimental group of six before
 treatment with mercury.

      ± average deviation of three determinations.

GExtrapolated from dry weight concentration.  Dry weight = 15$ of wet
 weight (average).


The results of the three experiments are summarized in Table 24.  Approxi-
mately twice as much mercury was accumulated by the oysters when mercury
resistant bacteria were added to the aquarium (experiments II and III).
The less-than-quantitative recovery obtained may have been a result of
adsorption of mercury onto the aquarium and shell surfaces.  Trapping of
volatilized mercury by the first trap was essentially quantitative, with
less than 0.10$ of the total added radioactivity accumulating in the second
trap.  Approximately 10$ of the added Hg was volatilized, even in the ab-
sence of added mercury metabolizing bacteria.  The volatilization of Hg
and the apparent excretion of Hg in fecal material suggest that the
oysters, themselves, may actively metabolize Hg.  During the course of
each experiment, it was observed that the total concentration of mercury
in the water approached the concentration of mercury in the microparticu-
late fraction of the water.  Particulate mercury increased during the
first 24 h and then decreased.  During this period, the rate of decrease
in dissolved Hg was most rapid.

The relative and absolute accumulations of mercury by individual oyster
tissues are presented in Table 25.  Generally, the largest proportion
of the accumulated Hg was found in the gills  (Table 25A).  In experiments
II and III, the amount found in the gills was statistically significantly
greater than all other tissues examined.  In the latter experiments, the
mantle and visceral mass accumulated significantly greater proportions
of Hg than either the adductor muscle or mantle fluid.  Table 25B shows
the tissue concentrations of mercury.  Comparison of experiment I with
                                     70

-------
experiments II and III indicates that significant increases in mercury
accumulation in gill tissue occurred when mercury-resistant bacteria
vere added to the system.  The distribution of mercury in the tissues
found in this study is consistent with the data published by others.
Pentreath (79) shoved that the stomach and gills vere the primary sites
of both concentration and exchange of heavy metal nuclides in the mussel,
Mvtilus edulia.  Kbpfler (80) reported that in oysters continuously ex-
posed to 50 ppb of Eg as HgCl2, the accumulation of Eg in gills was more
than ten-fold higher than any other tissue.  Mercury accumulated in the
gills is probably entrapped in the microparticulate or dissolved form-via
the mucous secretions of the gills.  Similarly, Penreath found that '^Fe
was associated with the mucous covering of the gills (79).

The results of experiment III vere unexpected, in that relatively less
accumulation would have been predicted due to increased volatilization
of Kg  through bacterial metabolism.  Although less Hg was accumulated
than in experiment II, significantly more was accumulated than in the
control experiment.  The relatively high half saturation constant (20
ppm of HgCl ) (see Section IV) for the evolution of Hg  by the bacterial
culture is a likely explanation.  At the low concentration (10.2 ppb) of
HgCl  in the water, the bacterial cells will evolve Hg  at a relatively
low rate.  However, culture #5 does accumulate Hg, albeit at 1/3 the
rate of culture #14» and may also enhance Hg uptake by the oysters.

CONCIIJSION

In short term experiments,  mercury resistant, mercury accumulating bac-
teria caused relative increases in Hg accumulation in all of the oyster
tissues examined.  Increases in mercury concentration of the gills were
statistically significant,  with concentration factors in excess of three
thousand.  It can be concluded that bacteria may have a demonstrable
effect upon mercury accumulation in those food chains which include  fil-
ter feeding components.
                                  71

-------
                    Table 24.  RELATIVE DISTRIBUTION OF MERCURY FOLLOWING EXPOSURE OF OYSTERS
Percentage of total added mercury* after 4 days

Experiment
I
II
III

Uptake
22.00
39.30
36.90
Oyster
Fecal
material"3
—
2.07
2.10

Adherent
organisms0
—
0.80
0.55

Unfiltered
water
4.20
7.30
5.60

Suspended
particulates
2.S
5.9
4.1

Volatile6
9.25
8.70
8.00

Recovery*
35.5
58.3
53.1
w       a
         The initial mercury concentration vas 10.2 ppb of HgCl2.


        b
         Collected on 8 micron membrane filter after settling of aquarium contents.


        c
         Worms and crustaceans adhering to oyster shells.



         Mercury retained by filtration through 0.45 micron membrane filters.


        8Mercury accumulated in first HgBr2 - KBr trap. Mercury accumulated in the second traps ranged
         from 0.03-0.095S of the total added mercury.

        £
         Experiment I did not include fecal material or adhering organisms.   Mercury adsorbed to aquarium
         and shell surfaces vas not measured in any of the experiments.

-------
                                  Table 25.  TISSUE DISTRIBUTION OF MERCURY AMONG TREATED OYSTERS8
-q
U)
          A.   Relative uptake
Percent*1
Experiment
I
*
II
III
B. Mercury
Experiment
Mantle
fluid
1.05 ± 1.43
1.17 + 0.52
1.25 ±0.54
concentration8-

Mantle
fluid
Mantle
14.51 ± 3.81
12.14 ± 5.86
20.67 ± 5.63

Mean
Mantle
Gills0
51.79 ± 11.63
67.50 + 9.36
57.84 ± 5.44

mercury concentration
Gills0
Visceral Adductor
mass muscle
29.57 + 12.71 3.06+1.11
16.87 + 7.89 2.34 + 0.51
17.70+ 6.45 2.46 + 1.20

(ppb HgCl2 - wet weight )d
Visceral Adductor
mass muscle
Whole
oyster
4.19 + 2.49
9.84 ± 4.77
9.20 + 5.91


Whole
oyster
               I       4.13 + 5.69   162.10 + 128.90    619.50+  474.40   231.90 + 182.90    52.80+  44.10   173.30 + 120.00
              II      10.35 + 8.59   397.05 + 290.00   2974.10 ± 1703.70   338.30 + 262.00   153.25 + 105.76   424.60 + 267.00
             III       8.49 + 6.23   714.78 + 4U.67   2778.65 + 1060.78   408.77 + 238.93   200.02 + 197.69   389.77 + 277.59
           After four days in the presence of HgCL,  (initial concentration of 10.2 ppb).
           Tissues expressed relative to  total uptake of HgCl2 per oyster.  For whole oysters, expressed relative to total
           initial amount of HgCl2 added  to  the aquarium.  Ninety-five percent confidence interval for mean of 4 determinations.
          n
          "Including labial palps.
           Ninety-five percent confidence interval for mean of 4 determinations.  Underlined values are significantly different
           from control values (experiment I).

-------
                             SECTION VIII

                              REFERENCES
 1.  Nelson,  J.D.,  W.  Blair, F.E. Brinckman, R.R. Colwell, and W.P. Iver-
    son.   Biodegradation of phenylmercuric acetate by mercury-resistant
    bacteria.   Appl.  Microbiol. g6:321-326.  (1973)

 2.  Spangler,  W.J., J.L. Spigarelli, J.M. Rose, R.S. Flippin, and H.H.
    Miller.  Degradation of methyl mercury by bacteria isolated from en-
    vironmental samples.  Appl. Microbiol. 25_:488-493.  (1973)

 3.  Klein, D.  and E.  Goldberg.  Mercury in the marine environment.
    Environ. Sci.  Technol. 4:765-768.  (1970)

 4.  Williams,  P.M. and H.V. Weiss.  Mercury in the marine environment:
    Concentration in seavater and in a pelagic food chain.  J. Fish.
    Res.  Board Can. 20:293-295.  (1973)

 5.  Kondo, I., T.  Ishikawa, and H. Nakahara.  Mercury and cadmium resis-
    tances mediated by the penicillinase plasmid in Staphylococcus
    aureus.   J. Bacteriol. H2:1-7.  (1974.)

 6.  Novick, R. and C. Roth.  Plasmid-linked resistance to inorganic
    salts in Stanhvlococcus aureus.  J. Bacteriol. 22:1335-1342.  (1968)

 7-  Smith, D.   R factors mediate resistance to mercury, nickel, cobalt.
    Science 15.6:1114-1115.  (1967)

 8.  Summers, A.O. and E. Lewis.  Volatilization of mercuric chloride by
    mercury-resistant plasmid-bearing strains of Escherichia coli.
    Staphylococcus aureus. and Pseudomonas aeruginosa.  J. Bacteriol.
    H3.: 1070-1072.   (1973)

 9.  Komura, I. and K. Izaki.  Mechanism of mercuric chloride resistance
    in microorganisms.  I. Vaporization of a mercury compound from mer-
    curic chloride by multiple drug resistant strains of Escherichia
    eoli.  J.  Biochem.  (Japan) 20:885-893.  (1971)

10.  Tonomura,  K., K. Maeda, F. Futai, T. Nakagami, and M. Tamada.  Stim-
    ulative vaporization of phenyl mercuric acetate by mercury-resistant
    bacteria.   Nature (London) 212:644-646.  (1968)

-------
11. Furukawa, K. and K. Tonomura.  Induction of metallic mercury-releas-
    ing enzyme in mercury-resistant Pseudomonaa.  Agr. Biol. Chem.
    (Japan) 3j>: 2441-2448.  (1972)

12. Summers, A.O. and S. Silver.  Mercury-resistance in a plasmid bear-
    ing strain of Escherichia coli.  J. Bacteriol. 112:1228-1236.   (1972)

13. Magos, L.A., A. Tuffrey, and T.W. Clarkson.  Volatilization of mer-
    cury by bacteria.  Brit. J. Ind. Med. 21:29-4-298.  (1964)

14. Kbmura, I., T. Funaba, and K. Izaki.  Mechanism of mercuric chloride
    resistance in microorganisms.  II. NADPH-dependent reduction of mer-
    curic chloride and vaporization of mercury from mercuric chloride by
    a multiple drug resistant strain of Escherichia eoli.  J. Biochem.
    (Japan) 22:895-901.  (1971)

15. Nelson, J.D. and R.R. Golvell.  Effects of tropical stora Agnes upon
    the bacterial flora of Chesapeake Bay.  Chesapeake Research Consor-
    tium Pub. #27, U. S. Army Corps of Engineers, Baltimore, Maryland.
    (1974)

16. Tonomura, K., K. Maeda, and F. Futai.  Studies on the action of
    mercury-resistant microorganisms on mercurials.  II. The vaporiza-
    tion of mercurials stimulated by mercury-resistant bacterium.  J.
    Ferment. Technol. 46:685-692.  (1968)

17. Olson, B.H. and R.C. Cooper.  Methylation of mercury by estuarine
    sediments.  In: Abstracts.  Ann. Meeting Amer. Soc. for Microbiol.,
    Miami, Florida,  p. 48.  (1973)

18. Spangler, V.J., J.L. Spigarelli, J.M. Rose, and H.H. Miller.  Methyl-
    mercury:  Bacterial degradation in lake sediments.  Science 180:
    192-193.  (1973)

19. Bothner, M.H. and R. Carpenter.  The rate of mercury loss from con-
    taminated estuarine sediments in Bellingham Bay, Washington.  In:
    Proc. First Ann. National Sci. Foundation Trace Contaminants Confer-
    ence, Oak Ridge National Laboratory, Oak Ridge, Tennessee.   (1973)

20. Gilmour, J.T. and M.S.  Miller.  Fate of a mercuric-mercurous chloride
    fungicide added to turf grass.  J. Environ. Qual. g:145-1^8.  (1973)

21. Klmura, Y. and V. L. Miller.  The degradation of organomercury fungi-
    cides in soil.  J. Agric. Food Chem. 12:253-257.  (1964)

22. Voartal, J.  Transport and transformation of mercury in nature and
    possible routes of exposure.  In.: Mercury in the Environment (L.T.
    Friberg and J. Vostal,  eds.).  CRC Press, Cleveland, Ohio.   pp.  15-
    27.  (1972)
                                   75

-------
23. Jernelov, A.  Factors in the transformation of mercury to methylmer-
    cury.  In: Environmental Mercury Contamination  (R. Hartung and B.D.
    Dinan, eds.).  Ann Arbor Science Publishers,  Inc., Ann Arbor,
    Michigan,  pp. 167-172.  (1972)

24. Lockwood, R.A. and K.Y. Chen.  Chemical transformation of mercury in
    the aquatic environment.  Presented at the 35th  Annual Meeting of the
    American Society of  Limnology  and Oceanography,  March 19-22,
    Tallahassee, Florida.   (1972)

25. Jernelov, A.  Studies in Sweden on feasibility of some methods for
    restoration of mercury-contaminated bodies of water.  Environ. Sci.
    and Technol. 7_:712-718.  (1973)

26. Hatch, W.R. and W. L. Ott.   Determination  of sub-microgram quantities
    of mercury by atomic absorption sDectrophotometry.   Anal. Chem.  £0:
    2085-2087.   (1968)

27. Shewan,  J.M., G. Hobbs, and W. Hodgkiss.  A determinative scheme for
    the identification of certain  genera  of gram-negative bacteria,  with
    special  reference to the Pseudomonadaceae.  J. Appl.  Bacteriol.  2J.:
    379-390.   (1960)

28. Colwell,  R.R. and W.J.  Wiebe.   "Core" characteristics for use in
    classifying  aerobic, heterotrophic bacteria by  numerical taxonomy.
    Bull.  Georgia Acad.  Sci. 2£: 165-185.   (1970)

29. Steel, R.G.D. and J.H.  Torrie. Principles and  Procedures of Statis-
    tics.  McGraw-Hill Book Company,  Inc., New York, N.Y.  (1960)

30. Balkwill,  D.L.  and L.E. Casida, Jr.   Microflora of soil as  viewed by
    freeze-etching.   J.  Bacteriol. HA: 1319-1327.  (1973)

31. Westoo,  G.   Determination  of  methyl  mercury  compounds in food stuffs.
    I.  Methyl mercury compounds in fish,  identification and determination.
    Acta Chem.  Scand. 20:2131-2137.   (1966)

32. Schubel, J.R.   The physical and chemical conditions of the  Chesapeake
    Bay.   J. Wash.  Acad. Sci.  6g: 56-87.   (1972)

33. Gaby,  W. and C.  Hadley.  Analytical laboratory test for the identi-
    fication of Pseudomonas aeruginosa.   J.  Bacteriol. 24:356-358.   (1957)

34. Nelson,  J.D.,  H.L. McClam, and R.R.  Colwell.   The ecology of mercury-
    resistant bacteria  in Chesapeake  Bay.  Ja!  Preprints.  Proceedings
    of the 8th Annual Conference of the Marine Technology Society,
    Sept.  11-13, Washington,  D.C.   (1972)

35. Walker,  J.D. and R.R.  Colwell. Mercury-resistant bacteria and petro-
     leum degradation.   Appl.  Microbiol.  22:285-287.   (1973)
                                     76

-------
36. Landner, L.  Biochemical model for the biological methylation of mer-
    cury suggested from methylation studies J£ vivo vith Neurospora
    crasaa.  Nature  (London) £2P.:452^54.  (1971)

37. Bongers, L.H. and M.N. Khattak.  Sand and gravel overlay for control
    of mercury in sediments.  EPA-16080HVA01/72, U.S. Environmental Pro-
    tection Agency.  U.S. Government Printing Office, Washington,  D.C.
    (1972)

38. Holm, H.W. and M.F. Cox.  Mercury transformations in aquatic  sedi-
    ments.  In: Abstracts.  Ann. Meeting Amer. Soc. Microbiol., Chicago,
    Illinois,  p. 25.  (1974)

39. Kaneko, T. and R.R. Colvell.  Ecology of Vibrio parahaemolTticus in
    Chesapeake Bay.  J. Bacteriol. 112:24-32.(1973)

40. Murchelano, R.A. and C. Broun.  Heterotrophic bacteria in Long Island
    Sound.  Marine Biology 7_:1-6.  (1970)

41. Jernelov, A.  Mercury and food chains.  In; Environmental Mercury
    Contamination (R. Hartung and B.D. Dinan, eds.).  Ann Arbor Science
    Publishers, Inc., Ann Arbor, Michigan,  pp. 174-177.  (1972)

42. Brasfield, H.  Environmental factors correlated vith size of bacterial
    populations in a polluted stream.  Appl.  Microbiol. 24:349-352.   (1972)

43. Biggs, R.B.  Geology and hydrography.  In.: Gross Physical and  Bio-
    logical Effects of Overboard Spoil Disposal in Upper Chesapeake  Bay.
    Contribution No. 397, Natural Resources Institute of the University
    of Maryland, Solomons, Maryland, pp. 7-15.  (1970)

44. Smith, J.D., R.A. Nicholson, and P.J. Moore.   Mercury in water of
    the tidal Thames.  Nature (London) 2J£: 393-394.  (1971)

45. Brown, H.G., C.P. Hensley, G.L. McKinney, and J. L. Robinson.   Effi-
    ciency of heavy metals removal in municipal sewage treatment plants.
    Environ. Letters £:103-114.  (1973)

46. logsdon, G.S. and J.M. Symons.  Mercury removal by conventional  water
    treatment methods.  In;  Proceedings 92nd Annual Conference American
    Water Works Association, Chicago, Illinois.  (1972)

47. Flemer, D.A.  Phytoplankton.  In;  Gross Physical and Biological Ef-
    fects of Overboard Spoil Disposal in Upper Chesapeake Bay.  Contribu-
    tion No. 397, Natural Resources Institute of the University of
    Maryland, Solomons, Maryland,  pp. 16-25.  (1970)

48. Goodwyn, F.  Zooplankton.  In;  Gross Physical and Biological  Effects
    of Overboard*Spoil Disposal in Upper Chesapeake Bay.  Contribution
    No. 397, Natural Resources Institute of the University of Maryland,
    Solomona, Maryland, pp. 39-41.  (1970)
                                    77

-------
49. Lovelace, T.E., H. Tubiash, and R.R.  Colwell.   Quantitative and
    qualitative commensal bacterial flora of  Crassostreq virginica in
    Chesapeake Bay.  Proc. Natl.  Shellfish  Assoc. 58:82-87.   (1968)

50. Ross, I.S. and  K.M.  Old.   Thiol compounds and resistance of
    Pyrenophora avenae to mercury.  Trans.  Br. Mycol.  Soc.  60:301-310.
    (1973)
51. Stutzenberger, F.J.  and E.O.  Bennett.   Sensitivity of nixed popula-
    tions of Staphvloc occ'|ij? au^eua and I
    Appl. Microbiol. 13.:570-574.   (1965j
tions of Staphv lococcua aureua and Eachericbia coli to mercurials.
                                  5)
52. Nuzzi, R.   Toxicity of mercury to phytoplankton.   Mature (London)
    222:33-40.   (1972)

53. Ben-Bassat, D.,  G. • Shelef,  N.  Gruner,  and H.  Shuval.   Growth of
    Chlamydomonas in a medium containing mercury.   Nature (London) 240;
    43-44.   (1972)

54. Nelson,  J.D. and R.R.  Colwell.  Metabolism of mercury compounds by
    bacteria in Chesapeake Bay.  In: Third International Congress on
    Marine Corrosion and Fouling.   Northwestern University Press,
    Evanston,  Illinois, pp.  767-777-  (1973)

55. Furukawa,  K., T. Suzuki,  and K. Tonomura.  Decomposition of organic
    mercurial compounds by mercury-resistant bacteria.  Agr. Biol. Chem.
     (Japan)  3^:128-130.  (1969)

56. Tingle,  L.E., V.A.  Pavlat,  and I.L. Cameron.   Sublethal cytotoxic
    effects  of mercuric chloride on the ciliate Tetrahymena pyriformj,3.
    J.  Protozool. 2p.:301-304.  (1973)

57. Bernheim,  F.  The effect of cyanogen iodide and mercuric chloride
    on the permeability of cells of Pseudomonas aeruginosa and the
    antagonistic action of sulfhydryl compounds.   Proc. Soc. Expt. Biol.
    Med.  12£:444-447.  (1971)

58. Kellenberger, E. and R. Ryter.  Cell wall and cytoplasmic membrane
    of Escherichia coli.  J.  Biophys. Biochem. Cytol. 4*323-326.  (1958)

59.  luft, J.H.  Improvement in epoxy resin embedding material.  J. Bio-
    phys. Biochem. Cytol.  2:409-414*  (1961)

60. Watson,  M.L.  Staining of tissue sections for electron microscopy
    with heavy metals.  J. Biophys. Biochem. Cytol. 4:475-478.  (1958)

61. Venalle, J.H. and R. Coggeshall.  A simplified lead citrate stain
    for use  in electron microscopy.  J. Cell Biol. 25_:407-408.  (1965)
                                    78

-------
62. Troger, R.  Uber den Metallnachweis in quecksilber-oder kupfer-
    behandelten Bakterien.   (Detection of metals in bacteria tested with
    mercury or copper.)  Arch. Mibrobiol.  22.: 186-190.   (1959)

63. Murray, A. and D.  RLdby.  Mercury distribution and transformations
    in growing yeast.  In; Abstracts.   Ann. Meeting Amer.  Soc.  Microbiol.,
    Philadelphia, Pennsylvania,  p. 17.   (1972)

64-. Gillespie, D.C.  Mobilization of  mercury from sediments into guppies
    (Poecilia Kyflj^ij^i.  j. Fish.  Res.  Bd. Canada 22:1035-1041.
    (1772)

65. Anderson, A.A., J.M. Anderson, and L.E. Mayer.   System simulation to
    identify environmental research needs:  Mercury contamination.   OIKDS
    24:231-238.   (1973)

66. Barber, R.T., A. Vijayakumar, and F.A.  Cross.   Mercury concentrations
    in recent and ninety-year-old benthopelagic  fish.   Science  178:
    636-639.  (1972)

67. Weiss, H.V., M. Kbide, and E.D. Goldberg.  Mercury in  a Greenland ice
    sheet:  Evidence of recent input  by man.  Science  174:692-694.   (1971)

68. Parsons, T.R., C.A. Bawden, and V.A. Heath.  Preliminary survey  of
    mercury and other metals contained in  animals from the Fraser River
    mud flats.  J. Fish. Res. Bd. Canada 3J>: 1014-1016.   (1973)

69. Leather land, T.M., J.D. Burton, F.  Culkin, M.J.  McCartney,  and R.J.
    Morris.  Concentrations of some trace metals in pelagic organisms
    and of mercury in Northeast Atlantic Ocean water.   Deep Sea Res.  20:
    #79-685.  (1973)

70. Vestoo, G.  Methylmercury as percentage of total mercury in i'lesh and
    viscera of salmon and sea trout of various ages.   Science 181:567-568.
    (1973)

71. Windom, H., R. Stickney, R. Smith, D. White, and F. Taylor.   Arsenic,
    cadmium, copper, mercury, zinc in some species of  North Atlantic  fin
    fish.  J. Fish. Res. Bd. Canada 20:275-279.   (1973)

72. Holden, A.V.  Mercury in fish and shellfish.  A review.   J.  Food
    Technol. g:1-25.   (1973)

73. KLemmer, H. and S.N. Luoma.  Mercury levels  in marine biota.  Project
    Bull. No. 6, Water Resources Research Center, University of Hawaii,
    Honolulu.  (1973)

74. Peakall, D.B. and R.J. Lovett.   Mercury:  Its occurrence and effects
    in the ecosystem.  BioScience 2^:20-25.  (1972)
                                    79

-------
75. Jernelov, A. and H. Lann.  Mercury accumulation  in food chains.
    OIKDS 22:403-406.   (1971)

76. Cunningham, P. A. and M.R. Tripp.  Accumulation and depuration of mer-
    cury in the American oyster, Grassostrea vireinica.  Marine Biol.
    20:14-19.  (1973)

77. Vernberg, W.B. and  J. O'Hara.   Temperature-salinity  stress and mer-
    cury uptake in the  fiddler crab, Uca pugjlator.   J.  Fish. Res. Bd.
    Canada 22:1491-1494.   (1972)

78. Glatsoff, P.  The American oyster, Crassostrea v^Lrginica  (Qmelin).
    U.S. Fish. Wildlife Service Fishery Bulletin 64:1-480.   (1964)
79.  Pentreath,  R.J.   The accumulation from water of ^Zn,
     and 59pe  by the  mussel,  Mytilus edulis.  J.  Mar.  Biol.  Ass.  U.K.
     5^:138-153.  (1973)

80.  Kbpfler,  F.C.  The accumulation of organic and inorganic mercury com-
     pounds by the  Eastern oyster (Crassostrea virginica).   Bull. Environ.
     Contain. Toxicol. H: 275-280.  (1974)
                                     80

-------
                              SECTION IX

                         LIST OF PUBLICATIONS
Articles

Nelson, J.D., H.L. McClam, and R.R. Colwell.  The ecology of mercury-
resistant bacteria in Chesapeake Bay.  In: Preprints.  Proceedings of the
Eighth Annual Conference of the Marine Technology Society, Washington,
B.C.  pp. 303-312.  (1972)

Nelson, J.D. and R.R. Colwell.  Metabolism of mercury compounds by bac-
teria in Chesapeake Bay.  In: Third International Congress on Marine Cor-
rosion and Fouling.  Northwestern University Press, Evanston, Illinois.
pp. 767-777.  (1973)

Nelson, J.D., F.E. Brinckman, R.R. Colwell, and W.P. Iverson.  Biodegra-
dation of phenylmercuric acetate by mercury-resistant bacteria.  Appl.
Microbiol. 26:321-326.  (1973)

Walker, J.D. and R.R. Colwell.  Mercury-resistant bacteria and petroleum
degradation.  Appl. Microbiol. 22:285-287.  (1974)

Nelson, J.D. and R.R. Colwell.  Ecology of mercury-resistant bacteria in
Chesapeake Bay.  J. Microbial Ecol.  In Press.

Papers Presented

Nelson, J.D., T.B. Elliot, and R.R. Colwell.  Metabolism of mercury com-
pounds by bacteria in Chesapeake Bay.  Presented at the 35th Annual Meet-
ing of the American Society of Limnology and Oceanography, March 19-22,
Tallahassee, Florida.   (1972)

Nelson, J.D., H.L. McClam, and R.R. Colwell.  The ecology of mercury-
resistant bacteria in Chesapeake Bay.  Presented at the Eighth Annual Con-
ference of the Marine Technology Society, Sept. 11-13, Washington, D.C.
(1972)

Nelson, J.D. and R.R. Colwell.  Metabolism of mercury compounds by bac-
teria in Chesapeake Bay.  Presented at the Third International Congress
on Marine Corrosion and Fouling, Oct. 2-6, Gaithersburg, Maryland.  (1972)
                                    81

-------
Nelson, J.D., L.W. Wan, Z. Vaituzis, and R.R. Colvell.  Effects of mer-
curic chloride on the morphology of selected bacterial strains.  Present-
ed at the Annual Meeting of the American Society of Microbiologists,
May 6-11, Miami Beach, Florida.  (1973)

Nelson, J.D. and R.R. Colvell.  Effects of tropical storm Agnes upon the
bacterial flora of Chesapeake Bay.  Chesapeake Research Consortium,
Publication  #27.   (1974)
                                    82

-------
                               SECTION X

                               GLOSSARY


Term                 Definition

CFU                  Colony-forming units

Hg°                  Elemental mercury

MeHg Cl              Methylmercuric chloride

PES                  A 0.01  M phosphate-buffered "three salts solution"
                     (pH 7.0)

PMA                  Phenylmercuric acetate

Three salts          An artificial estuarine salt vater mix of 1$ Nad,
                     0.23% MgCl2-6H20,  and 0.03$ KC1

TVC                  Total viable, aerobic, heterotrophic count of bac-
                     teria by spread plate enumeration
                                 83

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 . REPORT NO.
 EPA-600/3-75-007
             3. RECIPIENT'S ACCESSION-NO.
 . TITLE AND SUBTITLE
 METABOLISM OF MERCURY COMPOUNDS IN MICROORGANISMS
                                                           S. REPORT DATE
                                                           October 1973 (Issuing Date)
                                                           6. PERFORMING ORGANIZATION CODE
 . AUTHOR(S)
                                                           8. PERFORMING ORGANIZATION REPORT NO.
 Rita R. Colvell and John D. Nelson,  Jr.
9. PERFORMING ORGANIZATION NAME AND ADDRESS

 Department  of Microbiology
 University  of Maryland
 College Park, Maryland  207^2
             10. PROGRAM ELEMENT NO.

             1BA022
             R 802529-01
12. SPONSORING AGENCY NAME AND ADDRESS
 Environmental Research Laboratory
 Office  of "Research and Development
 U.S. Environmental Protection.Agency
 Narragansett, Rhode Island   02882
             13. TYPE OF REPORT AND PERIOD COVERED
             14. SPONSORING AGENCY CODE

             EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
 This report describes the physiology and ecology of mercury-resistant and mercuryT-
 metabolizing bacteria from Chesapeake Bay.  Evidence is. presented which establishes
 a role  for bacteria in the cycling of mercury in the estuarine environment.  From
 the results of a. survey of Hg°  production among a group of  randomly selected,
 HgCl2-resistant bacteria and mixed natural microbial populations, it was established
 that the enumeration of mercury-resistant bacteria by plate counting is a valid index
 of potential Hg^+ metabolism in situ.

 The distribution of mercury-resistant bacteria was significantly different in water
 and sediment, from station to  station, and seasonally; the  proportion of Hg2+-
 resistant bacteria among the total, viable, heterotrophic bacterial population
 reached a reproducible maximum in Spring and was positively correlated with water
 turbidity, dissolved oxygen  concentration, and mercury concentration in the sediment.

 These  findings and the observation of the evolution of Hg°  from freshly collected
 water  and sediment suggest that bacteria may contribute substantially to the mobili-
 zation and transformation of mercury from existing deposits in Chesapeake Bay,
 specifically, and in the aquatic environment, in general.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                           c. COSATI Field/Group
 Marine microorganisms
Mercury degrading bacteria
Mercury mobilization
6M
6P
 18. DISTRIBUTION STATEMENT
 RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
     UNCLASSIFIED
                                                                         21. NO. OF PAGES
20. SECURITY CLASS (TMspage)
     UNCLASSIFIED
                                                                         22. PRICE
EPA Form 2220-1 (9-73)
                                                  • v U. S. GOVERNMENT PRINTING OFFICE: 1975-657-695/5322 Region No. 5-11

-------