OFFICE MEMORANDUM e STANFORD UNIVERSITY @ OFFICE MEMORANDUM e¢ STANFORD UNIVERSITY e OFFICE MEMORANDUM

To

FROM

SuBJEcT:

Date: April 22, 1975

Joshua Lederberg
Department of Genetics

Draft Proposals for Research Bearing on Water Reclamation

I am very grateful for the opportunity to have participated in the
meeting last Friday and wish it had been possible for me to attend the entire
session. I look forward to seeing the minutes of the further proceedings,

I was glad to see that there was no lack of appreciation of the complexity
of the problem of toxicity evaluation of substances that may be found in water
supplies and the additional burden on that list that reclamation processes
will undoubtedly entail.

One way of expressing this point is to recall that under current FDA
regulations new products must undergo on the average from $1 million to $3 million
of safety testing before a new drug entity can be marketed. (This is apart from
other costs connected with proving the efficacy of the new agent). Since it is
clear that there will be many thousands of chemical substances in our water
supplies that can be detected by modern sophisticated analytical techniques
and potentially many hundreds of thousands of others that could be found if we
push these methodologies to their ultimate possibility of sensitivity, it should
be perfectly obvious how futile it would be to attempt an evaluation of the
safety of additives to water supplies along the same lines as are now imposed
for the definition of new drugs and food additives, We are, of course, dealing
with somewhat different kinds of problems in the sense that our problems with
water supplies have to do with the extremely broad dissemination of substances
in trace amounts even by contrast to the analogous problem with food additives
and the more so with respect to drug entities. This suggests that we must
develop alternative strategies of environmental surveillance that are far more
sensitive to questions of quantitative risk assessment and to the issues of
chronic effects at low dose even than pertain to the other areas of safety
regulation. With the exception of a few selected high priority targets our
research will simply be unable to deal with each entity one at a time,

Consider how difficult it is already to find reasonable standards for specific
identifiable substances already known to be in many water supplies like

chloroform and asbestos where it is obvious that no one has the foggiest idea
what possible bounds of hazard may be reflected in existing circumstances,

It seems to me then that our recommendations for research programs must
be focussed at a number of different levels:including even a deeper analysis
of the research strategy itself,so that what will inevitably be an insufficient
number of dollars to do the job properly may nevertheless give us the best
result still achievable,

over

WNONVYOWAW Jd1ddO © ALISYZAINN GYOAINVIS e

WNONVYOWFW FJD1ddO @ ALISHFAINN GYOINVIS

ALISUZAINN GYOANVILS e@

WNGNVYOWIW F51dd0
At level one I have in mind the improvement of our overall methodology
of risk assessment of environmental chemicals. Most laymen do not have the
foggiest idea what a quandary we are in in trying to obtain useful estimates
of human risk in a toxicological framework. We urgently need great improvements
in our general insight into questions like the variability of individual
human response to a variety of chemicals; the mathematical form of dose-risk
curves; the techniques for extrapolation from animal experimentation at high
levels to the assessment of human population risks. Whatever possibility there
may have been with respect to the attitude of demanding absolute safety and
zero risk with respect to intentional chemical additives, we do not have that
luxury with respect to our water supplies;and we must learn to develop
policies based on optimum risk rather than the fantasy of zero risk. This is
no longer a question of imposing economic burdens on one particular category
of producers or innovators but an area of shared risk in which the entire
community must inevitably participate. There is no economically or socially
feasible way in which water supplies can be furnished to the public that
consist of absolutely pure water to the total exclusion of other material -
even fresh rain water or highly purified distilled water will still contain
detectable amounts of other constituents. And as soon as we have human
congregations, the contamination of water supplies with potentially contagious
agents, not to mention chemical additives, is absolutely inevitable. If this
objective fact is not widely understood, the further pursuit of water hygiene
will be founded on a pernicious fallacy.

A second level of investigation is undoubtedly already given considerable
impetus judging from recent news reports;and this should concern the development
of chemical methodology and the survey of existing water supplies for the actual
prevalence of observable chemical species in different regions of the country.
Our primary mission is the assessment of water reclamation, buUwe can hardly adopt
as a standard for reclaimed water criteria that ignore the ambient situation
with respect to existing water supplies. We should keep in mind, however, that
there is likely to be increasing pressure (1) towards the achievement of even
higher standards of safety for such primary water supplies, but (2) increasing
economic pressure in the opposite direction, These may become more and more
important in the future if we are to assess the ultimate projection of the role
of reclaimed water in future policy.

Hopefully information on the abundance of organic and inorganic contaminants
may enable us to seek out a number of specific high priority targets for more
specific toxicity evaluation. For example, the recent findings of the abundance
of chloroform as an ultimate residue from the interaction of chlorine with
a variety of organic inputs obviously points to the need to reassess the chronic
toxicity of chloroform at rather low concentrations but distributed to the entire
population. The fact that large doses are capable of inducing liver damage and
liver nodules in rodents is only a primary alarm and is hardly a secure basis
for undertaking costly and possibly futile measures to eliminate chloroform from
our water supplies. This will not be an easy task but certainly the first steps
that need to be undertaken are chronic feeding studies to animals at various
levels of chloroform intake. Connected with this should be a very careful
analysis of the biochemical and physiological pathways by which chloroform
exerts its toxic effect since it is unlikely that empirical measurements alone
can be extrapolated to give us reliable information on the impact of parts per

million levels of contamination, (My view is that a judicious combination of

over
empirical research on dose—toxicity relationships and of more basic
theoretical studies on the mechanism of toxic action are indispensable
for extrapolations of chronic toxicity measurements down to the level of
expected human exposure!)

If there are more than a dozen or so chemical species that become highly
suspect owing to their prevalence in water supplies, we will face an almost
impossible level of cost in endeavoring to find really credible answers to the
questions posed by their potential toxicity.

Another level of search should therefore be poised at more global
rather than selective methods of toxicity evaluation of water additives.
When we think of chronic long-term effects, a large part of our concern is
directed to potential carcinogenesis, Fortunately through the work of
Bruce Ames and others we have increasing confidence about the useful role
of simplified microbiological test-systems for guessing at the average
carcinogenic potential of a range of chemicals. This is an approach that has
already been applied to a number of single chemical species although we do
not yet have a good calibration that enables us to make reliable quantitative
estimates of human risk from bacterial experiments and this is certainly one of
the lines of investigation that needs to be pursued. However, even failing that,
one could envisage the development of practical test protocols that involve the
extraction of a significant fraction of the organic load from a water supply,
the elution of these materials from the adsorbent, and the testing of these
concentrates for mutagenic activity in bacterial test-systems. Certainly,
water supplies that gave fluctuating responses in such assays would be
candidates for suspicion and for aggressive pursuit of the individual culprits
in question. This approach bids fair to shorten the efforts that would be involved
in some global assessment of organic chemical aspects of water quality since
it would not necessarily involve the isolation of hundreds of different chemical
entities as the first step in the analysis. Rather, this would wait upon the
finding of a suspicious positive result.

These experiments would be directed at the assessment of the toxic load
of a water supply before the final purification process as an index of the
possible leakthrough even after purification by adsorption had been allowed to
run its course.

Plainly, another very important arena for further research is the thorough
practical and theoretical analysis of the adsorption process itself. We can
hardly afford to rely on the hope that a given specimen of charcoal will be
universally effective in removing all organic chemical constituents with equal
efficacy and with unflagging efficiency throughout the cycle of use of the
material. We need far more sophisticated models of the adsorption process that
take account of the role of the competition between different chemical species,
the saturation of adsorbing sites by one compound as it may influence the
efficiency of adsorption of others, the development of indexes and assays to
monitor the efficiency of adsorption and so forth. With such information we will
be in a position to make theoretical estimates of the level of effective
filtering out of undesirable chemical species by a given process and we will
be in the position of conducting our analysis on the influent which is far more
amenable to chemical analytical procedures than the effluent after purification.
In the process we may also hope to understand adsorbency to the point of being
able to design ever more efficacious agents, processes and protocols for use and
for monitoring.

over
Such studies may reveal which compounds may safely be allowed to enter
into water supplies destined for eventual reclamation,and which will have
to be monitored closer to their proximate source. We may then wish to
do investigations at still another level, of social engineering, for example
to determine the feasibility of taxing the producers of specific effluenf=
in some relation to the cost of their removal. This would also imply monitoring
at the principal sources of novel effluents - here I have in mind special
chemical industries which introduce a still new variety of chemical entities
into the water supply - a monitoring that is likely to be more efficacious
at the source than after admixture with the total water supplies. Furthermore,
the knowledge of specific sources of contamination gives us specific and
detailed information about the characteristics of chemical entities that would
be difficult to decipher in the global pool.

In addition I think that special emphasis does have to be given to the
chemistry and biological consequences of a process like chlorination since this
is such a fundamental part of our overall strategy of water protection and
since chlorination has by now been amply demonstrated to be the source of
an abundant variety of new and sometimes unexpected chemical species of potential
toxicity. In a sense the use of chlorine has converted natural products like
the humic acids into an industrial effluent. The study of these reactions with
chlorine may enable us more intelligently to single out individual high priority
candidates for selective toxicity analysis. In addition, chlorinated derivatives
are quite likely to be among those which are the least biodegradable and therefore
may play a significant role in the problem of water buildup in the context of
studies on reclamation.

All of these problems will have their comparable analogues even if
chlorination were to be replaced by other methods of water disinfection.

To the extent that reclaimed water poses new biological as well as chemical
burdens, we may have to anticipate other kinds of interaction that need to be
taken account of in our overall strategy. For example, if chlorination must be
relied upon to a greater degree as a means of coping with the contaminants present
in reclaimed water, then this must be regarded as one of the costs of reclamation
as opposed to primary water usage.

Besides the products of chlorination reactions that have been identified
in potable water as distributed, we have also to consider the further products
of chemical reaction of residual chlorine with dietary and physiological
constituents, And then we have to consider not only natural dietary constituents
and bodily secretions but also additives that are being evermore widely used
in increasing amounts. Part of the safety evaluation of materials like cyclamates
and saccharin will have to include their interaction with chlorine to the extent
that this remains chemically active at the point of admixture and human intake.
This is a problem that can hardly be ignored with respect to primary water
supplies subject to chlorination but is likely to achieve even greater prominence
to the extent that reclaimed waters may require further disinfection.
The previous discussion focussed on the chemical hazards relevant to
water supply and reclamation. For assessing biological hazards, we also need
substantial new information particularly with respect to the survival of
viruses. Fortunately, in the operation of domestic septic tanks we already
have some traditional usage that may be relevant to this problem but which
may require re~examination from a fresh outlook, Certainly we need to know
much more about the survivability of different microbial and viral species
in the actual circumstances of aquifer replenishment and other modes of
recirculation. As I discussed at the meeting, we can hardly rely in any
quantitative way on existing indices of coliform survival as measures of
the effective level of fecal pollution. The most troublesome point is of
course the possible durability of some microbial and many viral species
which might allow for very substantial build-up and broadcasting of infectious
agents through the reclamation route.

Segregated water supplies may be an avenue of solution especially for
chemical contamination, but even this must be validated by more experimental
work on the extent to which agricultural products are in practice contaminated
by additives in irrigation water. Of course, details of agricultural practice
may have a lot to do with the extent to which contaminating materials,
chemical or biological, find themselves in the final edible product. I do not
know whether there had been studies on the uptake of infectious agents like
hepatitis virus through the rootS of growing plants, and of course each chemical
entity must have rules of its own with respect to the possibility of chemical
concentration in the food cycle. One of the hazards of supply segregation
is that water supplies that would be kept at a level of marginal potability
under a non-segregated regime may allow to become heavily contaminated in a
nonchalant fashion with possibly very dangerous results.

Since wastes must eventually end up in the oceans, whether they go through
a reclamation cycle or not, we should not entirely avoid the questions of the
adequacy of the oceanic reservoir ~ the effects on marine life and eventually
on water quality and other environmental values of the same chemical constituents
about which we are concerned in a water reclamation context. The chlorinated
hydrocarbon insecticides already give us a notorious precedent to illustrate this
point. I am not suggesting that I believe that there is a crisis of planetary
habitability looming within the next 25 years, as some of my fellow biologists
have suggested, but it is not too soon to be applying the same kind of concern
that we express about our own fresh water supplies to an oceanic context.