United States
Environmental Protection
Agency
Research and Development
Health Effects Research
Laboratory
Research Triangle Park NC 27711
. EPA-600/S1 -84-027 Jan. 1985
&EPA Project Summary
Effects of Selected
Waterborne Particulates on
Cellular and Molecular
Parameters
R. W. Hart, R. Fertel, H. A. I. Newman, and J. R. Blakeslee
This project was initiated to determine
the effects of participate samples, some
concentrated directly from drinking
water supplies, on biological endpoints
in cultured mammalian cells and to
compare them with those caused by
known asbestos materials cultured with
the same cell type. A normal human
fibroblast cell strain and primary Syrian
hamster embryonic cell cultures were
used as test systems. The cellular and
biochemical parameters examined
were: (1) cytotoxicity, (2) virally-direct-
ed cellular transformation, (3) cell mem-
brane composition changes, and (4)
cyclic nucleotide concentration dif-
ferences. Three of the particulate sam-
ples were collected by filtration from
drinking waters from Duluth, Seattle,
and San Francisco. The other three
samples included taconite tailings, at-
tapulgite clay, and particles from asbes-
tos-cement pipe tapping. All samples
were less cytotoxic than Canadian
chrysotile. Three of the samples affect-
ed the ratio of cyclic nucleotides in a
manner similar to what would be ex-
pected if the cells had been treated with
a chemical carcinogen. The samples
had varying effects on glycolipids and
glycoproteins. It was apparent that the
samples induced differing effects rel-
ative to the biological endpoints exam-
ined. Some of these endpoints may be
useful in predicting potential health
hazards of waterborne particulates.
This report was submitted in fulfill-
ment of Grant No. R-804201-02bythe
Chemical Biomedical Environmental
Research Group of the Ohio State
University Research Foundation, Colum-
bus, Ohio, under the sponsorship of the
Environmental Protection Agency.
This Project Summary was developed
by EPA's Health Effects Research L abo-
ratory. Research Triangle Park. NC, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Mineral fibers and particulates are
found in air, soil, and water and thus
reach man by a number of routes. Many
of these materials are relatively inert and,
rather than decomposing, recirculate in
the environment. Most important, how-
ever, their mode of action has not been
determined and a method other than
epidemiology has not been developed to
screen for their potential human health
effects.
Cultured mammalian cell-line studies
provide biological endpoints that can be
used to compare the potential health
•effects of pollutants. A multidisciplinary
approach is necessary to evaluate the
effects of biological endpoints such as
cytotoxicity, enhancement of virally di-
rected cellular transformation, alteration
of cell membrane composition, changes
in cyclic nucleotide ratios, modification of
chemical carcinogen metabolism, and
association with genetic material.
-------
Independent studies have suggested
that asbestos induces biochemical alter-
ations in cells in culture and, further, that
different types of asbestos induce these
alterations to differing extents. Since the
extent of the effect induced by various
forms of asbestos appears to correlate
with the carcinogenic potential of these
agents, as deter mined by epidemiological
studies, it was deemed reasonable that a
double-blind study be performed to fur-
ther examine this correlation with un-
known samples, which may be found or
isolated from drinking water.
Methods and Materials
Samples of particulates, some extract-
ed directly from drinking waters, were
referenced only by a code number and
sent to the grantee. The testing on the
various cell parameters was then done
blind.
Samples 1, 2, and 3 were particulates
collected by filtration from drinking waters'
from Duluth, Seattle, and San Francisco,
respectively The Duluth sample was
taken before the installation of the filtra-
tion plant at Duluth. Chrysotile fibers
have been identified in the particulate
samples from San Francisco and Seattle;
amphibole fibers were identified among
the particulates in the sample from
Duluth. Sample 4 contained the less than
2 micrometer size fraction of taconite
tailings which had been prepared by a
sedimentation separation procedure. Am-
phibole fibers were identified among the
particulates in sample 4. Sample 5 con-
tained attapulgite clay, a non-asbestos
mineral, which consisted of short fibers
in the same range as some chrysotile
asbestos. Attapulgite fibers have been
identified in some water supplies m
Georgia and Florida. Sample 6 consisted
of particulates coUected from water flow-
ing through asbestos cement pipe after
the pipe was tapped. Sample 6 contained
chrysotile, some crocidolite and cement
pieces.
The test systems employed were a
normal human fibroblast cell strain estab-
lished at Columbus and primary Syrian
hamster embryonic cell cultures. The
cellular and biochemical parameters ex-
amined were cytotoxicity, virally-directed
cellular transformation, cell membrane
composition, and cyclic nucleotide con-
centrations.
Results
The dose-response curve for each
unknown and for Canadian chrysotile
was determined over a range of 10~8 to
10 2gramsatlog intervals. At the highest
concentration used, all compounds proved
toxic; however, at the LD50 level, a
difference of three orders of magnitude
was noted between samples. The effect
of each sample on both the total number
of colonies formed as well as the size
distribution of the colonies as a function
of each dose was determined by use of an
Artek Size Discriminator. Studies have
shown that while all samples tested
induced differing shapes and sizes of
colonies relative to the control, each class
of colony produced a similar dose-survival
curve. Relative to potency and due to the
large number of plates per point (40
plates counted at least 4 times via an
automatic colony counter and size dis-
criminator) it is estimated that under the
conditions of these experiments and for
this particular normal human fibroblast
culture, Canadian chrysotile (control) is
more potent than 1 >6>2>5>4 based
upon their respective LDio's, LD50's and
LDao's.
Whereas all six known asbestos sam-
ples previously examined increased the
frequency of Snyder-Theilen Feline Sar-
coma Virus transformation of human
cells in vitro to one degree or another, the
unknown samples did not uniformly do
so. Samples 1, 2,4, 5, and 6 significantly
inhibited viral transformation when cells
were treated 2 hours prior to infection,
whereas substance 3 enhanced trans-
formation at this same time point. Sub-
stance 4, at a concentration of 0.1 fjg/m\,
enhanced transformation when applied
24 hours prior to infection. All substances,
with the exception of 6, affected viral-
directed transformation in a dose-depend-
ent manner.
Relative to the effect of unknown
samples on the cyclic nucleotide response
of cells with increasing doses of asbestos,
no predictable dose-response relation-
ship was noted; however, there were
striking differences in the tissue response
of human cells to the different samples
examined. These responses could be
divided into two (2) groups with samples
1, 2 and 3 having relatively minor effects
and samples 4, 5 and 6 causing a
consistent decrease in the cyclic nucleo-
tide ratio at all concentrations tested. The
most striking effects were seen with
sample 4. The only sample studied which
exhibited an effect similar to those ob-
served previously with known samples of
asbestos was sample 3, which exhibited
the biphasic curve observed previously.
Interestingly, the striking decrease in
ratio observed with certain of these
samples was very similar to that obtame
with chemical carcinogens.
Cell membrane monosialoganglioside
(GMi, GM2) are an index of a simpler ce
surface glycolipid pattern which, in prev
ious studies with known asbestos sarr
pies, appeared to change in a directio
that was approximately equivalent to th
carcinogenic potential of the sample
examined.
In the present studies, samples 1 and;
caused little change in the proportions c
surface gangliosides, whereas sample
3,4, and 5 produced losses in surface GT
with the greatest effect being manifesto'
on the surfaces of cells treated will
samples 4 and 5. The chrysotile-incu
bated cells included in the sample 6 stud
reacted anomalously to the asbestos
making the interpretation of sample I
changes difficult. When sample 6 treatei
cells were compared to the control fo
samples 3-5, there was a loss in GDiAani
a corresponding increase in GMi. GL-4 ii
the neutral glycolipids increased will
samples 1 and 2 treatment and decrease*
with samples 3 and 4 cell incubations
Consistent with its actions on ganglio
sides, among the test samples, sample <•
caused the greatest reductions in GL-4
Again, sample 6 effects were anomalou:
due to the fact that distributions o
glycolipids in the sample 6 control wen
so different from those of the sample 3-E
control.
Very high molecular weight protein;
increased with cell treatment of sample;
1 and 2. The control for samples 3, 4, anc
5 predominantly had a high concentrator
of a high molecular mass glycoprotein or
the cell surfaces. Samples 3-6 allowec
the detection of some of this glycoprotein
but it was in a lower concentration thar
that found in the controls. The loss o
such high molecular mass proteins has
been correlated with cells undergoing
transformation.
Conclusions
Of the unknown samples tested, sam-
ples 1, 6, and 2 appeared in that order to
be the most toxic. Relative to the effect of
these substances on the cyclic nucleotide
ratio, samples 1, 2, and 3 had little effect,
whereas samples 4, 5, and 6 produced a
decrease in the ratio similar to that
induced by chemical carcinogens. The
effect of these samples on viral-directed
transformation was inhibitory, unlike the
asbestos fibers previously studied (1)
which were stimulatory with the excep-
tion of substance 3, which was stimu-
latory when cells were treated 2 hours
before infection.
-------
The order of effectiveness in reduction
of surface complex glycolipids was 5 > 6
>3>4>2>1.The order of effectiveness
in reduction of the high molecular mass
glycoproteins was 5>4>3>6>2>1.
Thus, it is apparent that these samples
induced differing effects relative to the
biological endpoints examined. It is still to
be determined which of these endpoints
is the most useful in predicting potential
health hazards of waterborne particu-
lates.
These studies, taken in combination
with previous findings and those of other
laboratories, suggest that mineral fibers
and particulates may have two effects: (a)
cytotoxic and (b) promotional, with their
primary role in carcinogenesis being the
latter.
Reference
1. Hart, R. W., R. Fertel, H. A. I. Newman,
F. B. Daniel, J. R. Blakeslee, 1979.
Effects of Selected Asbestos Fibers
on Cellular and Molecular Parame-
ters. USEPA-600/1-79-021. 35 p.
R. W. Hart, R. Fertel, H. A. I. Newman, and J. R. Blakeslee are with Ohio State
University, Columbus, OH 43210.
James R. Millette is the EPA Project Officer (see below).
The complete report, entitled "Effects of Selected Waterborne Particulates on
Cellular and Molecular Parameters," (Order No. PB 85-127 496; Cost: $10.00,
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 Officer can be contacted at:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park. NC 27711
*US GOVERNMENT PRINTING OFFICE, 1985 — 559-016/7876
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE & FEES PA
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use $300
000052^
fc^vlk PKuTtC F Hl'M
U'^ S LidKARr
S UtARdHKN 5iKtL I
L H t C A (? U I L 606Ua
------- |