United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S2-83-045 Aug. 1983
Project Summary
Development and Assessment of
Procedures for Collection,
Chemical Characterization and
Mutagenicity Testing of Ambient
Air
A R. Kolber, T. J. Hughes, T. J. Wolff, L W. Little, C. M. Sparacino, and E. D.
Pellizzari
A chemical and biological testing
protocol was developed to initially eval-
uate the mutagenic/carcinogenic po-
tential of ambient air. , The testing
protocol outlined sampling, chemical
fractionation/identification and muta-
genicity testing procedures for ambient
air particles and vapors. Specifically,
this study: (1) evaluated the three-
stage Massive Air Volume Sampler
(MAVS), developed by Battelle® for
collection of air particles; (2) developed
and utilized a solvent fractionation
scheme for extraction of organic com-
pounds from ambient air particles; (3)
utilized a GC/MS/Computer system
to identify signature mutagens/car-
cinogens, and (4) developed, evaluated,
and utilized modifications of the Ames/
Salmonella typhimurium plate incor-
poration assay to assess the mutagenic
activity of ambient air particles and
vapors.
Air particles were collected in three
size fractions with the MAVS: <1.7,
1.7-3.5, and 3.5-20 jitm (mean particle
diameter) at the following locations:
South Charleston, West Virginia; Baton
Rouge, Louisiana; Lake Charles. Louisi-
ana; Beaumont Texas; Houston, Texas;
Upland, California; and Elizabeth, New
Jersey. Collected particles were sol-
vent fractionated into six chemical
classes: organic acids, organic bases,
nonpolar neutrals (NPN), polynuclear
aromatic hydrocarbons (PNA), polar
neutrals (PN), and cyclohexane in-
solubles (CHI). The mass of each frac-
tion was determined and the muta-
genic potential of certain fractions was
evaluated with the Ames/Salmonella
bioassay. When sample size was suf-
ficient the unfractionated extract was
also tested for mutagenicity. Either
the standard plate incorporation assay,
the spot test, or a newly developed agar
well diffusion assay was employed for
mutagenicity testing.
Test results suggested that the PNA,
PN, acidic, and basic fractions were
mutagenic. Both indirect-acting (SB
dependent) and direct-acting (S9 in-
dependent) mutagens were detected.
Mutagenic/carcinogenic compounds
were qualitatively identified in the frac-
tions by GC/MS/Computer analysis.
After solvent fractionation: (1) muta-
genicity not detected in the unfraction-
ated organic extract was detected in
specific fractions; (2) chemical analysis
was possible after fractionation; potent
mutagens were identified in the frac-
tions by GC/MS/Computer analysis,
and (3) certain minor fractions were
detected as mutagenic after concentra-
tion (e.g., organic bases). Solvent
fractionation, therefore, resulted in a
more meaningful chemical and muta-
genic analysis.
A major technical problem in this air
particle study was insufficient sample
amounts, which precluded comprehen-
-------
sive chemical and biological analysis.
The reasons for insufficient sample
amounts were: the MAVS collected an
unexpectedly low amount of sample,
organic compounds accounted for only
10 to 15 percent of the total air particle
weight and the small mass of some
chemical fractions. Availability of limi-
ted sample prompted the development
of a bioassay priority scheme to opti-
mize the amount of information obtain-
able from a limited amount of sample.
Mutagenicity results did qualitatively
identify probable mutagenic compounds
and chemical-classes. However, the
quantitative determination of mutagenic
activity was secondary to the purpose
of this EPA study. Mutagenicity testing
was utilized only to develop the meth-
odology for a more extensive research
effort to quantitatively evaluate the
mutagenicity of ambient air particles
at these specific sites.
In addition to the development of
techniques to measure the microbial
mutagenic activity of ambient air par-
ticles, a preincubation technique was
developed to quantitatively measure
vapor-phase organic mutagens in the
Ames/Salmonella assay. Techniques
to measure vapor-phase organic com-
pounds with the standard plate incor-
poration protocol of the Ames/Sal-
monella mutagenicity assay were not
adequate. The preincubation technique
was evaluated with Salmonella strain
TA100 and the following pure muta-
genic vapor-phase organic compounds:
ethylene oxide, propylene oxide, bu-
tylene oxide, styrene oxide, ethylene
dibromide, and vinylidene chloride.
Results suggested that the preincuba-
tion technique was three to ten times
more sensitive than the standard plate
incorporation technique in the detec-
tion of vapor-phase mutagens. Vapor-
phase samples were collected on Tenax®
cartridges, and GC/MS/Computer anal-
ysis identified signature mutagenic
and carcinogenic compounds. Muta-
genic ambient air compounds identified
in the vapor samples were primarily
short-chain chlorinated hydrocarbons,
such as tetrachloroethylene. The
combined Tenax®-resin/preincubation
technique can now be utilized to identify
and evaluate potentially mutagenic am-
bient air vapors.
In conclusion, the study objective to
develop an initial biological and chemi-
cal testing protocol for evaluation of
the mutagenic/carcinogenic potential
of ambient air particles and vapors was
accomplished. The usefulness and
necessity of an integrated biological-
chemical approach to assess the muta-
genicity of ambient air was demonstra-
ted. Guidelines for future quantitative
mutagenicity studies at industrial and
rural sites were established.
This Project Summary was developed
by EPA's Health Effects Research La-
boratory. Research Triangle Park. NC.
to announce key findings of the re-
search project that is fully documented
in a separate report of the same title
(see Project Report ordering informa-
tion at back).
Introduction
The Health Effects Research Laboratory
(HERL) of the United States Environmental
Protection Agency sponsored this three-
year study of ambient air pollution, which
was initiated in May 1 977 and concluded
in June 1 980. The goal of the research
was to develop and evaluate an initial
chemical and biological test protocol for
mutagenic assessment of ambient air. The
test protocol included: sampling procedures,
chemical fractionation techniques for re-
moval of organic compounds from air
particles, chemical identification tech-
niques, and microbial mutagenicity testing
procedures. Test agents were ambient air
particles, vapor samples, and pure muta-
genic compounds.
Similar integrated biological/chemical
testing approaches have successfully iden-
tified mutagens present in other complex
environmental mixtures, such as coal gasi-
fication effluents, fly ash from coal-fired
power plants, drinking water, shale oil,
synthetic fuels, automobile and diesel
exhaust and cigarette smoke condensate
(Ames, 1979; Waters et al., 1979). The
biological component of the ambient air
testing protocol was the Salmonella/
mammalian microsome mutagenicity assay,
developed by Dr. Bruce Ames and his
colleagues (Ames et al., 1975b). The
Ames mutagenicity assay is rapid and
economical, requires little space, and de-
tects many pure mutagens at nanogram to
microgram levels (McCann et al., 1975;
McCann and Ames, 1976). The Ames
assay can detect direct-acting mutagens,
and promutagens which require the addi-
tion of a mammalian liver S9 extract (Ames
et al., 1973). The Ames assay is based on
the reversion of histidine-requiring mutants
to histidine independence (reverse muta-
tion) as a result of interaction of the
bacterial DNAwith a mutagenic compound.
Bacterial mutagenicity is proportional to
the number of revertant colonies produced
after incubation in histidine-deficient media
at 37°C for two days. Since 1975, the
Ames assay has been utilized to detect a
wide variety of environmental and pure
mutagens (Ames et al., 1975a; Durston
and Ames, 1974; Kieretal., 1974; McCann
et al., 1975; McCann and Ames, 1976;
Waters et al., 1979). The main value of
the Ames assay is that the assay detects
many carcinogens as mutagens; and 80 to
90 percent correlation between mutagenic
and carcinogenic potential has been demon-
strated (Commoner etal., 1976; McCann
et al., 1976; Sugimura et al., 1976;
McCann and Ames, 1977).
This pilot study determined if the Ames
assay could detect mutagenic compounds
in ambient air in a cost-effective and timely
manner. At the start of the program, the
Ames assay was a "novel" system. Since
that time, numerous other investigators
have utilized the Ames/Salmonella bac-
terial mutagenesis assay to detect muta-
gens in ambient air. For example, initial
studies by Pitts etal. (1977) demonstrated
mutagenic activity in organic extracts from
air particles collected at eight urban Cali-
fornia sites, but not for particles collected
at a rural site. A quantitative study per-
formed by Tokiwa et al. (1980) on air
particulate collected from Japanese cities
found higher mutagenicity in organic ex-
tracts from industrial areas when com-
pared to extracts from rural areas. Both
direct- and indirect-acting mutagens were
detected in the Tokiwa study and similar
findings have been reported since 1977
(Tokiwa et al., 1977; Talcott and Wei,
1977; Dehnenetal., 1977; Commoner et
al., 1978; Teranishi et al., 1978; Talcott
and Harger, 1980; Lofroth, 1978; Alfheim
and Moller, 1979; Moller and Alfheim,
1980). A general overview of the collec-
tion, chemical characterization, and muta-
genicity testing of organic compounds
extracted from ambient air particles was
recently published by Hughes et al. (1980).
In addition to the development of the
Ames/Salmonella mutagenicity assay,
comprehensive studies of ambient air were
made possible by other recent technical
advances. The adaptation of chemical
fractionation techniques originally de-
signed to separate cigarette smoke con-
densate into chemical-classes (Swain et
al., 1969) has served as a model for the
study of many other types of complex
environmental organic mixtures (Com-
moner, 1979; Huisingh et al., 1979;
Pelroy and Peterson, 1979; Guerin et al.,
1979) such as air particles, diesel exhaust
shale oil, and liquified coal samples.
Fractionation techniques reduce the com-
plexity of the organic mixture and permit
meaningful chemical analysis and biological
testing to be performed. For example.
-------
chemical identification of potential muta-
genic/carcinogenic signature compounds
in chemical-class fractions of ambient air
particles has been performed (Pellizzari et
al., 1979). Each chemical-class fraction
contained hundreds of compounds that
might have gone undetected with an anal-
ysis of only the crude extract (Pellizzari,
1979).
The sampling of particles from very
large volumes of air was made possible by
the development of a Massive Air Volume
Sampler (MAVS) (Henry and Mitchell,
1978), which samples over 20,000m3 of
air in 24 hr, and fractionates particles by
size. Size-fractionation of particles is im-
portant because the smaller sized particles
(<2/im) present a greater surface area per
unit weightfor adsorption of organic pollu-
tants. The smaller particles are more easily
inhaled and deposited in the lung and are
more difficult to expel (Lippman et al.,
1979; Schlesinger and Lippman, 1978;
Yeh et al., 1976; Natusch and Wallace,
1974).
Multidisciplinary techniques described
above are a very powerful analytical tool.
This report outlines the development and
utilization of such techniques for muta-
genic analysis of ambient air.
Results
Testing procedures and methods for
ambient air analysis were evaluated with
pure compounds and ambient air samples.
Ambient air particles, collected from seven
cites, were utilized to evaluate: the Massive
Air Volume Sampler (MAVS), the chemical
fractionation scheme, the GC/MS/Com-
puter analysis system, and the Ames/
Salmonella assay. Pure mutagenic com-
pounds were utilized to determine the
biological and chemical recovery after sol-
vent fractionation, and to evaluate the agar
well assay and preincubation techniques
in the Ames/Salmonella assay. A modifi-
cation of the Ames plate incorporation
technique, the agar well diffusion assay,
was developed to optimize the information
available from a limited sample size of
particles. The agar well diffusion assay
was evaluated with known mutagenic com-
pounds. Results with positive mutagenic
compounds suggested that the agar well
assay could qualitatively detect direct- and
indirect-acting mutagens and toxicity on a
single plate, which increased the total
amount of information obtainable from
each fractioa In addition, a solvent fractiona-
tion scheme was developed which divided
crude solvent extract from air particulate
samples into six general chemical-class
fractions: organic acids and bases, polar
and nonpolar neutrals, polynuclear aromatic
hydrocarbons, and cyclohexane insoluble
fractions. The solvent fractionation scheme
was evaluated for chemical and biological
separation and recovery ability to determine
if fractionation was quantitatively accurate,
if known chemicals were fractionated into
their proper chemical-classes (qualitatively
accurate), and if fractionation influenced
the mutagenic activity of known mutagenic
compounds. GC/MS/Computer analysis
identified mutagenic and carcinogenic
compounds in specific chemical classes.
The extractable organics from air particles
collected at several sites were utilized as
test agents to evaluate the effectiveness of
the techniques developed under this
program. Both mutagenic and chemical
testing was performed. Air particles were
sampled with the MAVS and solvent ex-
tracted and fractionated. The mass of each
chemical-class fraction was determined,
signature mutagens/carcinogens were
qualitatively identified by GC/MS/Com-
puter analysis (Table 1), and fractions
(when available) were tested for muta-
genicity with the Ames/Salmonella assay
(Table 2).
A second major objective of this study
was to develop and evaluate biological and
chemical techniques which would quantify
and identify organic vapor-phase mutagens
present in ambient air. Volatile compounds
are vapors at ambient temperatures, and
have been shown to represent a large
portion of ambient air pollutants. Duce
(1978) cited data that measured the global
vapor-phase organic pollutant load at 50 x
1011 g/day, and the particulate organic
matter load at4 x 1011 g/day. Consequent-
ly, the concentrations of vapor-phase or-
ganics are generally* ten to fifty times
greater than the levels of particulate organic
matter. Previous mutagenicity testing has
been conducted on pure vapor-phase
mutagens/carcinogens that have been
identified as present in ambient air, but no
research has been directly performed on
the mutagenicity of ambient air vapors.
Vapor-phase compounds were difficult to
quantitatively evaluate for mutagenic po-
tential due to their volatility and insolubility
in aqueous phases (Rosenkranz et al.,
1979). Consequently, a sensitive quantita-
tive technique to collect and measure the
mutagens present as vapors in ambient air
was developed. Tenax® cartridges de-
veloped by Pellizzari (1979), were utilized
for vapor collection and chemical analysis
of field samples. Modification of the
preincubation technique of Yahagi and
coworkers (Yahagi et al., 1975, 1977;
Nagao et al., 1977), was developed and
tested with pure vapor-phase mutagens.
Results suggested that the preincubation
technique was from three to ten times
more sensitive than the standard pour-
plate technique in the Ames/Salmonella
assay for the detection of the following
vapor-phase mutagens: vinylidenechloride;
styrene, propylene, ethylene and butylene
oxides; and ethylene dibromide (Table 3).
The utilization of this preincubation tech-
nique in the future, in combination with
Tenax® collected vapor-phase field samples,
would potentially allow quantitative chemi-
cal and biological analysis of mutagenic
vapors in ambient air. In this study,
mutagenic vapor-phase mutagens were
qualitatively identified in field samples by
GC/MC/Computer analysis (Table 4).
Conclusions
• A minimal biological and chemical
testing protocol was developed to evaluate
the microbial mutagenic activity of vapors
and the extractable organic matter from
particles in ambient air.
• A Massive Air Volume Sampler
(MAVS) collected ambient air particles in
three size fractions: <1.7 ju.m, 1.7-3.5
ju.m, and 3.5-20/x.m. Particle size was
important in the evaluation of the mutagenic
potential of ambient air particles because
the smaller sized particles (i.e., <1.7 p.m)
contained the greatest amount of surface
area per unit weight were the most respir-
able, contained the greatest measurable
amount of signature carcinogens/muta-
gens, and contained the highest mutagenic
activity.
• A solvent fractionation scheme was
developed to extract and separate organic
compounds from air particles. Crude or-
ganic extracts were portioned into six
general chemical-classes: organic acids
and bases, polar and nonpolar neutrals,
polynuclear aromatic hydrocarbons, and
cyclohexane insoluble fractions. The
chemical and biological recovery of the
scheme was evaluated with pure mutagenic
compounds. Solvent fractionation permitted
meaningful biological analysis; concentra-
ted minor, but highly mutagenic com-
ponents; and permitted the identification
of mutagenic chemical-classes and signa-
ture mutagenic compounds within these
chemical-classes by GC/MS/Computer
analysis (Hughes et al., 1980).
• The agar well diffusion assay, a modi-
fication of the standard Ames/'Salmonella
plate incorporation assay, was developed
to evaluate the mutagenic potential of
these organic fractions from ambient air
particles. The agar well diffusion assay
allowed multiple endpoints to be measured
on a single plate: toxicity, mutagenicity,
activation requirements, and appropriate
-------
Table 1. Mutagenic and/or Carcinogenic Compounds3 Identified in Ambient Particulate Organic Matter (POM) by Chemical Class Fraction. Identi-
fication was Performed by the GC/MS/Computer System
PNA
Anthracene"
Anthraquinone11
Benzanthraceneb-c
Benzanthrene"- c
Benzfluoranthene"
Benzophenanthrene"
Benzo(a)pyrene"- c
Benzeperlene"
PN
Acridine"-0
Aniline13
Benzanthracene"- c
Benzene13
Benzofluoroantheneh
Benzophenanthrene*3
Benzopyrene"' c
Benzo(a)pyreneb- c
ACIDS
Beneneisomersb
Cresol"-0
Napthacene"
Napthalene13
Phenol"
Phenol lsomersb
Phenyf3
p-Tetramethyl-
butylphenolb
BASES NPN
Benzoacrioine13 Benzoguinoline"
Benzoquinolineb Cresor-0
Phenol"
Dicyclohexylamine"
Naphthalene13
Nicotine"-0
2-Nitro-4,6-
dichlorophenof3- c
Chrysene"-0
Fluoranthene"
Methylbenzanthracene"'0
Methylbenzophenanthracene13
Methylbenzophenanthrene
Methylchrysene"
Methylnaphthalene13
Methylphenanthrene"
Methylpyreneb
Methylstearate"
Napthaceneb
Naphthalene13
Perylene"
Phenanthraceneb
Phenanthrene"
Phenylnapthylamineb-c
Pyrene"
Benzoquinoline
Chrysene13-0
Methylbenzanthraceneb
Methylbenzophenanthracene13
Naphthalene"
Phenol"
Pyrene"
Phenol"
Quinoline0
a'Designation of mutagens and carcinogens obtained from: McCann et a/., 7975; Sawicki, 1979; Rinkus and Legator, 1979.
"Carcinogen.
°Mutagen.
Table 2.
Summary of the Mutagenic Activity of Chemical Class Fractions and Crude Organic Extract from Ambient Air Particles (< 1.7(im Mean
Diameter?) Collected at Seven Geographical Locations
«
Agar Well Test
Pour-Plate Test
Site
South Charleston,
West Virginia
Baton Rouge,
Louis/ana
Lake Charles,e
Louisiana
Beaumont Texas
Houston, Texas
Upland, California
Elizabeth, New
Jersey
Crude
organic
extract Acids Bases NPN
++++ ++++ _c _
NT ++++ ++ —
NT ++++ NT ++++
NT NT NT NT
NT ++++ ++++ —
NT ++++ — ++
++++ ++++ ++++ —
PN PNA"
++++ NT1
— NT
++ NT
NT NT
++++ NT
++++ NT
++++ NT
Crude
organic
extract Acids Bases NPN PN
+++++_ -H-
NT NT NT NT —
+ — + + +
NT + — — —
— + + + ++
NT NT NT NT NT
+ NT NT — +
PNA
NT
++++
++++
+
-H-
+
+++
The qualitative ranking is based upon the following mutagenic response with any strain, either with or without S9 addition:
Agar Well Assay Pour-Plate Assay
No. Colonies/Plate Mutagenic Ratio
<5
6-10
11-20
>20
2-5
6-10
11-20
>20
aThe < 1.7ftm particles were generally the most mutagenic-sized fraction, when compared to 1.7-3.5 fim and >3.5 urn sized fractions.
"PNA fraction was not tested in the agar well assay due to the inability of PNAs to migrate in the water-based agar growth media.
c— = negative mutagenic response.
dNT= not tested due to insufficient sample size.
eDifferent sampling dates for agar well and pour-plate testing of Lake Charles, Louisiana samples.
4
-------
Table 3. Comparison of Initial Mutagenicity Slopes (Revertants/iig)a for Six Vapor-Phase Compounds with Three Different Bioassay Procedures
Initial Mutagenicity Slope Coefficient of Variance for S Mutagenic Confidence Intervals for S
Procedure Compound (S) in Revertants/ftg (Std. Error of S/Estimate of S) Response Lower Limit Upper Limit
Preincubation:
Log- Phase
Cells
Pour
Log-Phase
Cells
Pour
Stationary-
Phase Cells
Ethylene Oxide
Styrene Oxide
Ethylene Dibromide
Propylene Oxide
Butylene Oxide
Vinylidene Chloride
Ethylene Oxide
Styrene Oxide
Ethylene Dibromide
Propylene Oxide
Butylene Oxide
Vinylidene Chloride
Ethylene Oxide
Styrene Oxide
Ethylene Dibromide
Propylene Oxide
Butylene Oxide
Vinylidene Chloride
0.15
1.54
1.01
0.65
0.48
0.88
0.37*
0.80
0.34
0.33
0.03
0.86
0.09
0.45
0.66
0.09
0.70
0.47
0.47
0.08
0.74
0.08
0.70
0.20
0.27
0.77
0.28
0.72
70.77
0.73
0.55
0.77
0.28
0.32
0.47
0.27
+ 0.02
+ 7.29
+ 0.77
+ 0.54
+ 0.38
+ 0.46
+ 0.27
+ 0.62
+ 0.74
+ 0.25
-0.68
+ 0.63
-0.07
+ 0.29
+ 0.27
+ 0.32
+ 0.00
+ 0.78
0.27
7.79
7.32
0.76
0.59
7.30
0.53
0.98
0.54
0.42
0.75
7.09
0.79
0.62
7.06
0.76
0.79
0.64
Calculated by the nonlinear regression analysis system of Myers et al. (1981).
bO. 18 if outlier at 125 colonies for spontaneous backmutation rate is removed.
Table 4. Compounds Identified by GC/MS
in Vapor-Phase Organic Samples
from Four Sites" that are
Either Mutagens (McCann et al.,
1975) and/or Carcinogens
(Sawicki, 1979)
Benzene ( and benzene isomer)
Carbon tetrachloride
Chloroform
Ethylene dibromide
Napthalene
Phenol
Trichloroethane
Trichloroethylene
Tetrachloroethylene
Vinyl chloride
Vinylidene chloride
aSites sampled and tested were
Lake Charles, Louisiana; Beau-
mont Texas; Houston, Texas; and
Elizabeth, New Jersey.
dose-ranges. The development of the agar
well assay was necessary due to the small
amount of each fraction available. A bio-
assay priority scheme also was developed
to optimize the amount of information
obtained with limited sample amounts.
• Ambient air particles were collected
from seven U.S. geographical sites: South
Charleston, West Virginia; Baton Rouge,
Louisiana; Lake Charles, Louisiana; Beau-
mont, Texas; Houston, Texas; Upland,
California; and Elizabeth, New Jersey. The
collected particles were utilized as test
agents to evaluate the methods developed
during this EPA program. The particles
were size classified, solvent fractionated,
and tested for mutagenicity in the Ames/
Salmonella mammalian microsome muta-
genicity assay (Table 2). Both the standard
plate incorporation assay.and the agar well
diffusion assay (developed underthis EPA
program) were utilized. Signature mutagens
were identified with a GC/MS/Computer
system (Table 1). Results suggested that:
(1) the mutagenic activity was primarily
present in the PNA, PN, acid, and base
fractions; (2) the mutagenicity increased
as the surface area/unit mass of particulate
increased (i.e., as the diameter of the
particle decreased) (see Hughes et al.,
1980); (3) the chemical fractionation
scheme unmasked mutagenicity which
was not detected in the crude organic
extract(seeTable2, Houston, Texas, pour-
plate assay); (4) toxic effects present in
the crude extract were reduced during
fractionation; (5) PNA and PN fractions
generally required metabolic activation
(S9), while the acid and base fractions
generally did not require S9--the NPN
were generally nonmutagenic; (6) both
frameshift and base substitution mutagens
were detected; (7) the agar well diffusion
test permitted initial qualitative mutagenic
analysis when sample size was less than
10 mg (see Hughes et al., 1 980).
• A GC/MS/Computeranalysis system
identified signature mutagenic/carcino-
genic compounds in ambient air particulate
samples. Chemical identification of these
compounds supported the bioassay muta-
genicity results obtained from field samples.
• An adequate quantitative method
was not available to measure the muta-
genicity of ambient air vapors in the Ames/
Salmonella assay. Consequently, a modifi-
cation of the preincubation technique (in
liquid suspension) was developed for field
testing of ambient air vapors. This prein-
cubation technique was developed and
compared to the standard Ames/ Salmonella
assay for sensitivity. Both stationary- and
logarithmically-grown cells of Salmonella
tester strain TA100 were employed. Six
known vapor-phase mutagens were tested:
butylene oxide, propylene oxide, ethylene
oxide, ethylene dibromide, styrene oxide,
and vinylidene chloride. Results from this
vapor testing (summarized in Table 3)
suggested that: (1) the preincubation tech-
nique (with log-phase cells) was generally
superior in the detection of the mutagenicity
of six known vapor-phase mutagens; (2)
the preincubation assay had a reduced
variance at all dosage levels, and an initial
mutagenicity slope (total revertants//ig)
that was equal to or greater than the plate
incorporation assay for the six vapor-phase
compounds tested; and (3) the preincuba-
tion technique is applicable to ambient air
vapor-phase field samples collected on
Tenax® cartridges. The development of
the preincubation/Tenax® procedure will
potentially permit quantitative chemical
and biological analysis of potentially muta-
genic ambient air vapors.
• Ambient air vapor-phase organic
mutagens were qualitatively analyzed with
-------
a GC/MS/Computer system at Elizabeth,
New Jersey; Lake Charles, Louisiana;
Beaumont Texas; and Houston, Texas.
Identified signature mutagenic and car-
'cinogenic compounds included: benzene
and benzene isomers, chlorinated short-
chain hydrocarbons (chloroform, carbon
tetrachloride, trichloroethylene), and vinyl
and vinylidene chloride (Table 4).
Recommendations
The primary goal of this program was to
develop a minimal testing protocol for
future biological and chemical analysis of
ambient air vapors and particles. Testing
did result in qualitative mutagenic and
chemical data, although testing was per-
formed primarily to aid in the development
of methods for future testing. Preliminary
results suggested that further quantitative
evaluations of similar urban sites is feasible.
A rural site should be included as a com-
parative control.
• Size-classification and chemical frac-
tionation of organic extracts obtained from
air partides should be performed. Since
the percentage of organic material ac-
counted for only 10 to 20 percent of the
total weight of air paniculate samples,
initial biological and chemical testing re-
quires a minimum of 30 g of paniculate
sample. A 30 gram amount of paniculate
sample would allow chemical and biological
evaluation of both the crude organic extract
and chemical-class fractions. Adequate
dose responsive mutagenicity testing re-
quires a minimum of 50 mg of sample for
each chemical-class fraction. Testing of
both the crude extract and subsequent
fractions will provide data on synergistic
and antagonistic effects which can occur
in the crude organic extract Paniculate
samples should be fractionated, and recon-
struction experiments should be performed
to evaluate the effect of fractionation on
mutagenic potential.
• Chemical identification of signature
mutagenic/carcinogenic compounds in
mutagenic fractions and quantification of
these compounds per unit of air samples
is desirable. Additional analytical tech-
niques, such as HPLC and LC/MS, should
be employed to verify the results from the
GC/MS/Computer analysis.
• Quantitative toxicity testing in the
Ames assay should be performed prior to
mutagenicity testing to determine the non-
toxic dose ranges for subsequent muta-
genicity testing. Preferred doses for initial
toxicity testing of organic extracts of air
particles are: 1,000, 500,100,10, and 1
jug/plate. Multiple doses (a minimum of
five doses) should be tested, both with and
without Aroclor-induced S9 metabolic acti-
vation. If initial mutagenic activity requires
S9 metabolic activation, additional testing
should be performed to determine optimal
S9 concentrations. When sample size is
less than 50 mg, it is generally desirable to
perform extensive testing with one Sal-
monella tester strain, rather than to perform
minimal testing with all five tester strains.
Replicate dose response curves are recom-
mended for active fractions Strain priorities
for future testing are: TA98, TA100,
TA1535, TA1 537, and TA1 538. Strain
priority may be different when specific
information is required.
• Atmospheric conditions, time of year,
sampling dates, duration of sampling, and
exact location of sampling sites must be
documented and reported, since these
conditions can affect the genotoxic potential
of ambient air.
• The preincubation assay for vapor-
phase mutagen detection requires further
development to increase the sensitivity of
the S9 metabolic activation system for
promutagens. Specifically, variation in the
concentration of S9 protein and time of
liquid preincubation need further study.
The Tenax®/preincubation technique should
be field tested to quantify and identify
mutagenic ambient air vapors.
• Additional research to verify and
standardize the solvent fractionation scheme
would improve the ability to fractionate
mutagenic compounds into more distinct
chemical classes. Optimal solvents for the
extraction of organics from ambient air
particles should be determined. Collection,
fractionation, and bioassay systems could
be further optimized for sensitivity, ac-
curacy, and reproducibility of results.
A. R. Kolber. T. J. Hughes, C. M. Sparacino, andE. D. Pellizzari are with Research
Triangle Institue. Research Triangle Park, NC 27709; T. J. Wolff is located in
Southport. NC 28405; and L W. Little is with L W. Little Associates, Raleigh,
NC 27608.
Larry Claxton. Joel/en Lewtas, and Michael D. Waters are the EPA Project
Officers (see below).
The complete report, entitled "Development and Assessment of Procedures for
Collection, Chemical Characterization and Mutagenicity Testing of Ambient
Air," (Order No. PB 83-220 046; Cost: $20.50, subject to change) will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officers can be contacted at:
Health Effects .Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
6
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
PS 0000329
6GHCV
------- |