TRANSMISSION OF VIRUSES BY THE WATER ROUTE

Abstract

Few aspects of the transmission of viruses by water escape consideration in this volume which is based on a symposium sponsored by the Division of Water Supply and Pollution Control of the US Public Health Service at Cincinnati, Ohio, in December 1965. The first impression is of the large volume of material presented in that time, but most of the chapters contain more material than it was possible to present, some data were added subsequently, and several chapters were unscheduled additions to the symposium. The edited discussion has also been included. The book is divided into 5 parts, namely epidemiology, the quantitative recovery of viruses from dilute suspensions, the minimal infective dose, the viruses in water, and the survival of viruses in water and waste water. Each part corresponds to one session of the symposium, and each is believed to represent current knowledge and the ideas now being debated in one important area of this problem. J. W. MOSLEY'S consideration of the epidemiology of virus diseases in relation to drinking water (p. 5), and the following discussion clearly illustrate the difficulties in obtaining rigorous epidemiological proof of water-borne transmission. In the discussion M. GOLDFIELD gives cogent reasons for the view that it is unwise to place too much significance on the absence of epidemiological evidence. In the United States in the years from 1946 to 1960 there were 23 recognized or suspected epidemics of water-borne infectious hepatitis, with 930 cases. During the same period there were 142 episodes of gastroenteritis and diarrhoea with 18, 790 resulting illnesses. These accounted for 62% of all epidemics of water-borne disease and 72% of all cases. If virus infections accounted for only a part of this burden of gastrointestinal illness, as may have been the case, such a possibility calls for more expenditure and effort, and less complacency on the part of water authorities. D. M. MCLEAN'S studies (p. 25) dealing with virus hazards of recreational water, including wading pools and swimming baths, completed the first session. Since extracellular virus cannot reproduce, it is thought unlikely that the amount of virus of human and animal origin in water supplies is ever likely to be large. Techniques sufficiently sensitive to detect small amounts of viruses in large volumes of water are therefore in demand. This demand appears to have been met handsomely, and the range and variety of techniques now available are described and discussed. Some of these are readily applicable in any laboratory with facilities for tissue culture, since relatively simple and inexpensive equipment is utilized. The most attractive in their simplicity and effectiveness are the methods of detecting small quantities of virus in large volumes described by C. WALLIS and J. L. MELNICK (p. 129) (adsorption on aluminium hydroxide which adsorbed all 17 viruses tested except reovirus and gave recovery rates for the enteroviruses of over 80%); and by H. GARTNER (p. 121) whose paper showed the great utility of the soluble alginate ultrafilters especially for non-turbid waters. In such cases viruses may be demonstrable at a concentration of 0.001 ID50 per ml. when sufficient amounts of material are investigated. Of the remaining methods treated in this section, only the electrophoretic method of concentration described by M. BIER, G. C. BRUCKNER, F. C. COOPER and H. E. ROY (p. 57) seems practicable for the ordinary water laboratory. The contribution by J. E. SMITH and R. J. COURTNEY (p. 89) described in considerable detail the use of passive haemagglutination as a quantitative assay for the virion. Although technically very interesting, the method is unlikely to be preferred to the simpler and possibly more effective methods described in the same chapter. N. G. ANDERSON, G. B. CLINE, W. W. HARRIS and J. G. GREEN (p. 75) at Oak Ridge described how the possibilities of centrifugal separation and isolation of viruses from natural waters have been realized by letting a continuous stream of raw water flow over a density gradient in a continuous flow centrifuge rotor. The results obtained with experimental rotors are given; one removed 99.9% of a phage at the rate of 3 litres per hour when operating at 40, 000 r.p.m. Large numbers of particles (108-109 per ml.) with characteristic viral morphology were seen when these rotor systems were applied to isolation from river or sea-water. This contribution opened up a fascinating new field, for these techniques are not only a formidable asset in assessing the dimensions of the problem of water transmission, but enable the morphology, biophysics and ecology of naturally occurring particles of viral dimensions to be explored for the first time in any depth. A paper on the concentration of enteric viruses in water by hydro-extraction and two-phase separation is given by H. I. SHUVAL, S. CYMBALISTA, B. FATAL and N. GOLDBLUM (p. 45), and two papers by D. O. CLIVER on enterovirus detection (pp. 109 and 139). In the following section on the Minimal Infective Dose, S. A. PLOTKIN and M. KATZ (p. 151) reviewed the available data to justify their opinion that the human subject is not less susceptible than the tissue culture. None of the data is inconsistent with this view. KOPROWSKI [Bull. Hyg., 1956, v. 31, 500 and 1198] has shown that 2 PFU of attenuated poliovirus 1 produced infection in 2 out of 3 subjects, while Plotkin and Katz infected 2 of 3 infants with 10 TCD50 of an attenuated poliovirus 3, and a note added in proof indicates that even 1 TCD50 of this strain may be infective for infants. If small amounts are capable of infecting, then their presence in water, however diluted, assumes significance. The subject is extended in the next paper by J. W. BEARD (p. 167) to the " Host-virus interaction in the initiation of infection ". D. G. SHARP (p. 193) contributed a fascinating account of electronmicroscopy and viral particle function. Both the number and the state of aggregation of the particles can be measured, after the careful sedimentation procedures described, with the electron microscope. This leads to the determination of the quality of a virus preparation, and facilitates the interpretation of the observed biological effect. S. L. CHANG (p. 219) considered the statistics of the infective units of animal viruses, and stressed the importance of the aggregation of particles into clumps. The establishment of the fact that their behaviour conforms to the Poisson distribution does not mean that these infective units consist of single particles. The requirements are satisfied equally well by indivisible entities, and in fact clumping appears to be a normal phenomenon in suspensions of animal viruses. The serious effect of this on inactivation rate is discussed and the common types of survival curves observed in the inactivation of enteroviruses by chemical agents illustrated, their difference in form being interpreted according to the percentage of infective units in clumps. Aberrations in survival curves are dealt with by G. BERG, R. M. CLARK, D. BERMAN and S. L. CHANG (p. 235). R. M. CLARK and J. F. NIEHAUS (p. 241) present a mathematical model for viral devitalization which assumes aggregation as the major reason for aberrations. This model was programmed for an IBM 1620 computer with 20-k memory and copies of the programme written in Fortran II are available. Surface waters which are the major sources must abound in viruses parasitic on the aquatic fauna and flora, as well as those derived from higher plants and animals. Domestic animals and the wild life of urban areas must also contribute by way of storm waters which are commonly added to streams and rivers without treatment of any kind. The large number of non-human viruses present cannot be ignored since some viruses produce cancers in species exotic to them, and others may infect generically distant hosts. S. S. KALTER (p. 253) described the known picornaviruses and discussed their occurrence in man and animals. They include 65 human enteroviruses with 50 or more human rhinoviruses. The non-human picornaviruses are as yet unnumbered, but there are at least 76 non-human enteroviruses and at least 3 non-human rhinoviruses. E. A. MIRAND (p. 269) described the transmission of some tumour viruses and presented the results of studies on the transmission of Friend virus, a mouse leukaemia virus, water-borne transmission of which was barely successful experimentally. His statement that no tumourigenicity has been reported for the picornaviruses, arboviruses, herpes viruses or reoviruses may already be called in question by the recent demonstration that Burkitt's lymphoma is associated with reovirus type 3 [see Lancet, 1966, v. 1, 965, and STANLEY, Bull. Hyg., 1966, v. 41, 991]. Mirand's contribution is important to water hygienists inasmuch as the field covered is one with which not many are familiar. It remains to be determined whether small amounts of virus in water, air, soil or food, in the presence of a chemical carcinogen, can be an important tumour hazard in a host. The author notes that KOTIN and WISELEY (Prog. Exp. Tumour Res., 1963, v. 3, 186) found that large increases in the incidence of squamous carcinomas of the lungs occurred in mice when these were exposed to the vapours of ozonized gasoline and simultaneously infected with mouse-adapted strains of influenza virus. The importance of further studies on the synergistic action of chemical carcinogens and viruses in tumourigenesis is stressed, especially as some water sources are contaminated by both. J. E. PRIER and R. RILEY (p. 287) considered the " Significance of water in natural animal virus transmission " and presented some unconvincing data on enterovirus survival, with survival periods longer at 37 degrees C. than at 22 degrees C, a finding contrary to that of other workers and one which may indicate a relatively insensitive method. An inadequately studied as