A QUANTETATNE STUDY OF THE ENTER“: VERUSES IN SEWAGE 'E'hesis ka- tho Degree of Ph. D. 7 MECHtGAR STATE UNWERSETY James R. Frey 1961 Us! if) This is to certify that the thesis entitled A Quantitative Study of the Enteric Viruses In Sewage presented by James R. Frey has been accepted towards fulfillment of the requirements for _Eh..D.__ degree in Microbiology CU . 9/] ‘43)?‘7/1c/<. Major professor Date /?/3/7¢er}£ // (/é/ u 0-169 LIBRARY Michigan State University ABSTRACT A QUANTITATIVE STUDY OF THE ENTERIC VIRUSES IN SEWAGE by James R. Frey The purpose of this study was to determine step by step (quantitatively) the reduction of the enteric viruses in sewage during the process of sewage treatment. A total of 616 pad, dip and sludge samples were col- lected from the East Lansing sewage treatment plant from June 6, 1959, through November 7, 1959. Fifty-four samples were positive for the presence of virus when the samples were tested in monkey kidney epithelial monolayers. The pad sampling method proved to be the most satisfactory for the recovery of virus from sewage followed by the sludge and dip samples. Plaque counts were made on each of the original positive samples to determine the number of virus particles per unit volume. A new, improved, semi-microplaque technique for counting enteric viruses was developed and used in counting the virus samples. An increase in virus particles as expressed in plaque forming units per ml (pfu/ml) resulted from the treatment process of sewage through the disposal plant up to and including James R. Frey the final settling tank. The final effluent was exposed to an average of 0.5 parts per million of free residual chlorine for 15 minutes. The virus concentration entering the sewage plant was reduced 94.7 per cent as a result of the treatment process. The potential reservoir of virus disease is great in raw sewage, contaminated water supplies, inadequately treated sewage and activated sludge materials. Chlorine residual should be increased if destruction of viral patho- gens is desired. A QUANTITATIVE STUDY OF THE ENTERIC VIRUSES IN SEWAGE BY James Rf Frey A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Departmen€ of Microbiology and Public Health 1961 r . f f l [’l / ‘ , X ii ACKNOWLEDGEMENTS The author wishes to express his gratitude to Dr. Walter N. Mack for his patience and encouragement through- out the completion of this work. My thanks to Mrs. Gail D. Riegle whose technical assistance was invaluable during the course of this study. I wish to dedicate this thesis to my wife who has made this accomplishment possible. Her understanding and encouragement has been and always will be most appreciated. iii TABLE OF CONTENTS PAGE INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1 REVIEW OF LITERATURE . . . . . . . . . . . . . . . . . 3 MATERIALS AND METHODS . . . . . . . . . . . . . . . . . 20 Collection of samples ' 20 Laboratory processing of samples 23 Tissue culture methods 24 Virus isolation and confirmation 26 Preparation of cell cultures for plaque count determinations 28 Plaque studies 30 Plaque counts 38 Ultracentrifuge concentration of samples 40 RESULTS . . . . . . . . . . . . . . . . . . . . . . . . 42 DISCUSSION . . . . . . . . . . . . . . . . . . . . . . 61 SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . 7O BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 72 iv LIST OF TABLES TABLE PAGE I. Total monthly per cent recovery of virus ‘ from sewage . . . . . . . . . . . . . . . . . 43 II. Per cent virus recovery from dip, pad and sludge samples . . . . . . . . . . . . . . . . 44 III. Total monthly average in pfu/ml at each collecting station . . . . . . . . . . . . . . 46 IV. Concentration in pfu/ml by ultracentrifu- gation of known poliomyelitis virus in sterilized sewage . . . . . . . . . . . . . . 52 V. Total average in pfu/ml of viruses in sewage . . . . . . . . . . . . . . . . . . . . 54 VI. Total monthly average pfu/ml of virus from samples collected on the same date from different stations . . . . . . . . . . . . . . 56 VII. Total average number of viruses in pfu/ml compiled from data where viruses were isolated from two or more stations on the same day . . . . . . . . . . . . . . . . . . . 58 VIII. The per cent of viruses isolated by the pad sampling method from each station throughout the sewage treatment plant . . . . . . . . . . 60 IX. Virus isolations from sewage specimens as compared to isolations obtained from stool samples . . . . . . . . . . . . . . . . . . . 62 LIST OF FIGURES FIGURE Page 1. A diagrammatic stepwise collection by stations of sewage samples throughout the processing plant . . . . . . . . . . . . . 22 2. Individual clustering of cells throughout the cell monolayer prior to full plaque development, fixed and stained with crystal violet (11X) . . . . . . . . . . . . . 36 3. Normal control cell monolayer fixed and stained with crystal violet (11X) . . . . . . . 36 4. Full plaque development 48 to 72 hours after the appearance of small areas of degeneration, fixed and stained with crystal violet (11X) . . 37 5. Normal control cell monolayer fixed and stained with crystal violet (11X) . . . . . . . 37 6. A typical French square bottle overlaid with agar and stained with crystal violet . . . . . 39 INTRODUCTI ON The discovery of antibiotics, correlated with the knowledge that many multiplying viruses produce degenerative changes in cultured cells, has given the virologist a new tool with which to study the viruses in greater detail. The use of tissue cell culture in clinical and epidemiological studies has increased over the last ten years. The rapid development of new methods for growing various primary tissue cell lines is further proof of the practicability of this method for use in studying viruses. The purpose of this study was to determine step by‘ step (quantitatively) the reduction of the enteric viruses in sewage during the process of sewage treatment. The enteric group of viruses at present consists of the polioviruses, Coxsackie viruses, the enteric cytopathogenic human orphan (ECHO) viruses and possibly other unknown viral agents. The presence of these viruses in sewage has been established. Studies have shown a marked seasonal incidence of the enteric viruses in sewage with the highest recoveries occurring during the summer months. Little was known concerning the quanti- tative nature of the enteric viruses in sewage and the number that survive modern day treatment procedures. The efficiency of the sewage treatment process in removing the enteric viruses is of practical public health and epidemiological significance. The possible spread of disease by continuous discharge of treated sewage effluents into surface water is aesthetically disagreeable and important to every individual. REVIEW OF LITERATURE Today there is one factor on which virologists and epidemiologists agree ----- the consistency with which poliomyelitis, as well as the other enteric virus diseases seem to avoid falling into patterns normal to other transmissible diseases. Clinical manifestations of these diseases vary from the usually unrecognized carrier to frank paralysis. One of the first possible answers to this problem was suggested by Sawyer (1915-1916) who demonstrated the active virus of poliomyelitis in rectal washings from a patient 14 days after the onset of paralysis. Prior to this report Kling gt 21. (1912) infected monkeys with colonic flushings from patients suffering from infantile paralysis. Since the publication of these initial reports, numerous investigators have isolated many types of enteric‘ viruses from fecal materials. Among these are: Trask __§.§1. (1938), Howe and Bodian (1939), Kramer gt 31. (1939), Lepine (1939), Lepine gt a1. (1939), Vignac gt 2;. (1939). Gard (1940), Kramer (1940), Piszczek gt 31. (1941), Silverman (1941), McClure and Langmuir (1942), Dalldorf and Sickles (1948), Melnidk gt 31. (1949), Robbins gt 31. (1951), Riordan _E.§l. (1952), Melnick and Agran (1952), Steigman gt a1. (1953), Ramos-Alvarez and Sabin (1954), Honig gt a1. (1956), Ramos-Alvarez and Sabin (1956) and Gelfand £3 21, (1957). Trask gt al.(1938) isolated poliomyelitis virus from 4 cases of the disease. Three isolations were made from the nasopharynx in 3 different cases and 3 more isolations were made from stool specimens in one case. The virus remained viable for 10 weeks in one of the stool samples that was stored in the refrigerator at 4 C. Such findings suggested that these hitherto unrecognized sources of poliomyelitis virus might have been responsible for a high degree of pollution of sewage with the virus. As a result investigators turned to the study of sewage as a possible reservoir and means of naturally transmitting the enteric virus diseases. This led to the first isolation of poliomyelitis virus from sewage by Paul gt 31. (1939). These results supported the hypothesis that improper disposal of feces from patients might have serious public health consequences. Trask (1939) stated, "Authorities are now ready to admit other avenues to the central nervous system besides the nose, since lesions are not found in the olfactory bulbs in fatal human cases, and the disease has been pro- duced accidentally in man by the subcutaneous inoculation of vaccine, and monkeys (except the common rhesus) may be infected by placing virus in their food." Following the initial isolation of poliomyelitis virus from sewage, Levaditi (1940) isolated virus from sewer water in Paris. This was quickly followed by an anonymous report in the Journal of American Medical Association (1940) stating Kling had inoculated a number of monkeys with samples of the contents of Paris city waters. After repeated inoculations the animals developed paralySis thus confirming the previous reports concerning the presence of the poliomyelitis virus in sewage. In a report during the same year (Paul gt 31., 1940), poliomyelitis virus was recovered from urban sewage 6 times in 50 attempts. McDaniel (1940) recovered polio- myelitis virus from sewage in South Carolina. Paul gt_§l, (1940) demonstrated that poliomyelitis virus survived in the sewage flow for several miles. Trask and Paul (1942) established the fact that the- poliomyelitis virus was not detectable in urban sewage during non-epidemic times. This conclusion was subsequently refuted by Mundel gt 2;. (1947) who reported the recovery of poliomyelitis virus from sewage in a treatment plant in Johannesburg, South Africa. No case of poliomyelitis had been reported for the previous two months in the area prior to the initial virus recovery. This finding possibly indicated that continued infection of poliomyelitis was still occurring. A method for concentrating and detecting poliomyelitis virus in sewage was devised by Melnick (1947). This method consisted of ammonium sulfate precipitation followed by ultracentrifugation. Using this method virus was detected in sewage from Imhoff settling tanks, in raw sewage and in sewage leaving a poliomyelitis hospital. Virus was also recovered from the sewage of a residential area in which no personsvfith poliomyelitis were living and in which no polio- myelitis subsequently developed. Virus was recovered only in late summer and fall months. Dalldorf and Sickles (1948) isolated an unidentifiable agent from the feces of a child with paralysis. Isolation of the virus was dependent upon the utilization of day old mice rather than the more expensive monkey. This agent was later found not to be poliomyelitis virus but a new virus which ~ was called Coxsackie virus. Melnick 23 31. (1949) reported finding a filterable agent similar to that reported by Dalldorf. This agent was isolated from sewage and produced characteristic paralysis in day old mice. Clark gt al.(l951) studied the Coxsackie virus in urban sewage and reported that the virus was probably excreted by healthy carriers. Francis et a1. (1953) detected poliomyelitis and Coxsackie viruses from privy contents in Texas, thus supporting the presence of these viruses which had previously been isolated from sewage. Melnick et 31. (1954a) studied two sampling methods in collecting sewage specimens for virus isolations. Gauze pad and dip sampling methods were compared. The gauze pad technique yielded a significantly higher percentage of positive viruses than did the dip sampling method. These methods are described in detail under materials and methods. The seasonal distribution of Coxsackie viruses in urban sewage was investigated by Melnick t 21. (1954b) by testing a total of 1,926 samples which yielded 269 viruses. The viruses appeared mostly in samples collected in the summer and fall, but rarely during the winter and spring. Kelly E; El. (1955) demonstrated again the presence of Coxsackie virus in sewage. The gauze pad device was used as the method of sampling and these data supported anew, the concept that "acceptable" sewage treatment facilities did not always destroy pathogenic agents. Madk.§§_§1. (1958) tested sewage for 24 consecutive months. Viruses were recovered 118 times from 1,403 tests resulting in 8 per cent positive recovery. The greatest per cent of recoveries was made during the months of July, August, September, and October. Tests on the non-chlorinated final effluent from the sewage plant showed that some of the viruses were not removed by the sewage treatment. Types I and III poliomyelitis, ECHO, Coxsackie and 17 unknown viruses were isolated during this investigation. Giovanardi and Berganani (1959) found type I polio- myelitis virus to be the most prevalent of all as the cause of paralytic poliomyelitis in Italy. These investigators isolated this agent during inter-epidemic periods from sewage. Kelly and Sanderson (1959) studied 140 samples of raw and treated sewage collected in the summer and fall of 1957. Samples were obtained from the influent, effluent and acti- vated sludge only. Viruses were isolated from these samples 166 times. The frequency of isolation and the concentration of infectious particles in the samples indicated that sewage treatment did not destroy the viruses. Chlorination for disinfection, often but not always, reduced the frequency of isolation and concentration of the infectious particles. Prior to this report, Casey and Aymond (1940) correlated the distribution of poliomyelitis in Louisiana with case rates in communities with and without water supplies and sewage systems. Data indicated that communities without a water supply and sewage system had 39.7 cases per 100,000 inhabitants. Communities with a water supply and a sewage system had 26.6 cases per 100,000 population while communities with a water supply but no sewage system had a total of 83.6 cases per 100,000 population. Ellsworth (1940) also noted that certain bathing beaches in Massachusetts were affected by sewage pollution which may have resulted in a higher than 9 normal incidence of poliomyelitis. It was suggested that the virus might have been present in sewage polluted waters in sufficient concentrations to cause infection as a result of repeated contact of the water with the nasal passages. Hargreaves (1950) discussed a severe outbreak of poliomyelitis in Cornwall, England, which occurred in 1949. There was an incidence of 0.3 per cent per 1,000 population, 20 cases occurring in 18 days. Investigation showed that the cases were evenly distributed over the city. One interesting aspect of this report was the fact that the city water supply had been augmented by tapping a stream heavily polluted with Escherichia coli. After increased chlorination of the stream supplement, only one additional case was noted. Little (1954) demonstrated that a possible reservoir of infection was present in a town upstream from which inadequately treated sewage had access to the water supply. The polio- myelitis virus was demonstrated in sewage at the time the disease was present in the town. Along with the increased interest in the study of sewage as a source of enteric virus disease transmission the period of viral excretion by infected patients has been investigated. Trask g_.§l. (1938) found that the poliomyelitis virus remained viable for 10 weeks in a stool sample that was stored in the refrigerator. Virus was also recovered 10 from the feces of a child with abortive poliomyelitis on the second, fourteenth and twenty-fifth day after the onset of the disease. In 1940 Kramer reported that the virus of polio- myelitis survived for 6 months in the feces of a contact which had been stored in the refrigerator at 3 to 4 C. Piszczekugthl. (1941) isolated poliomyelitis virus from the feces of 3 of 4 patients about one month after the onset of the disease and also isolated the virus from stools of 6 of 38 contacts. Feces from 14 normal residents in the community gave negative results. Horstmann gt a1. (1944) studied the persistence of virus excretion in the stools of poliomyelitis patients and concluded that the period of virus excretion extended into the seventh and eighth weeks. The virus was isolated from the feces of 61 patients. Again in 1946 Horstmann gt a1. studied the excretion of virus in the stools of patients following acute poliomyelitis. The virus was present in 70 per cent of the cases in the first two weeks, 50 per cent during the third and fourth weeks, 27 per cent at the end of the fifth and sixth weeks, and 13 per cent at the end of the seventh and eighth weeks. In a single instance virus was demonstrated at the twelfth week after the onset of the disease. Rhodes gt_§l. (1950) demonstrated that poliomyelitis virus in a stool specimen, after addition of river water, was infective for monkeys for at least 188 days. 11 The stool sample was diluted in river water at a dilution of 1:200 and stored in the refrigerator at 4 C. These data again suggest the possibility that improperly treated sewage could play an important role in the transmission of enteric viral diseases. Due to the ease with which the enteric viruses could be isolated from feces and sewage materials, several investi— gators undertook studies involving the so-called carrier state and viruses present in normal individuals. Melnick and Agren (1952) isolated both poliomyelitis and Coxsackie viruses from normal infants in Egypt. They obtained samples by rectal swabs and used monkey testicular tissue cultures for the isolations. They reported finding an unidentifiable cyto- pathic agent; however, it failed to produce disease in monkeys, mice and hamsters. The characteristics of this unidentifiable agent were similar to others previously reported in the literature. Prior to this work Robbins gt a1. (1951) isolated two unidentifiable agents from two patients with diseases diagnosed as non-paralytic poliomyelitis. Riordan 2£.§l- (1952) reported that three agents out of a total of 24 iso- lations from the feces of both paralytic and non-paralytic poliomyelitis patients could not be typed. Kibrick _§.§l. (1953), Melnick gt El. (1953), Steigman E£.§l- (1953) and Horstmann (1955), all isolated unidentifiable agents in the 12 stools, blood or rectal swabs from healthy people or patients ill with poliomyelitis, non-paralytic poliomyelitis or aseptic meningitis. These viruses were tentatively called "orphan viruses" by Melnick §t_§l, (1955). The recovery of large numbers of new cytopathogenic viruses in different laboratories from the intestinal tract of man prompted the National Foundation for Infantile Paraylsis to establish the Enteric Cytopathogenic Human Orphan (ECHO) Virus Committee to c1asSify these viruses (1955). In 1957 the Committee on ECHO viruses changed its name to the Committee on Enteric viruses. This name change was based on the fact that the poliomyelitis, Coxsackie and ECHO viruses may all inhabit the alimentary tract of man. Finally the enteric viruses included the 3 types of poliomyelitis viruses, the 19 Group A and 5 Group B Coxsackie viruses and the 19 ECHO viruses. Needless to say, the task of continued classification of new enteric viruses is continuing. Ramos-Alvarez and Sabin (1954) recovered 31 cyto- pathogenic agents from 1,566 healthy American children who had no contact with clinically recognized cases of poliomyelitis. Rectal swabs were used as the sampling method. Poliomyelitis, Coxsackie and ECHO viruses were recovered from normal children with no apparent disease manifestations. In view of these findings Honig gt 31. (1956) investigated endemiologically 13 the enteric virus infections among 136 normal children in two socio-economic groups over a 29 month period. A repeatable seasonal incidence of enteric virus excretion was noted with over 90 per cent isolations occurring during the months of June through October. Again in 1956, Ramos-Alvarez and Sabin (1956) studied the viral flora of 3,336 healthy children in this country and in Mexico using rectal swabs as the sampling method. These data indicated that the highest recovery of the viruses occurred during the months of June, July and August. Gelfand gt 31. (1957) studied the enteric viral flora of normal children in southern Louisiana by testing extracts of fecal specimens. A marked seasonal association was evident with a peak in the summer and autumn months. Melnick §t_gl, (1953) concluded that stool samples were superior to rectal swab samples for virus isolations. During this study viruses were isolated that were not identi- fiable as poliomyelitis virus or Coxsackie virus. These agents were found to belong to the ECHO group of viruses. As a result of the ease with which enteric viruses could be recovered from stool samples and sewage, investi- gators searched for improved methods of isolation and concentration of these viruses. Kling £3 a1. (1939) devised a method for concentrating poliomyelitis virus from fecal material by means of ether extraction. The treated stool 14 sample materials could be kept in the refrigerator in ether up to 53 days before the virus was destroyed. Virus was recovered by use of this method from patients with the abortive form of poliomyelitis. Howe and Bodian (1939) investigated the efficiency of the intranasal inoculation in rhesus monkeys as a means of recovering poliomyelitis virus from stools. By using this route of inoculation, virus was demonstrated in 63 per cent of 174 stool specimens. By testing each specimen with two or more monkeys instead of one, the virus recovery could be increased to 90 per cent. Gard (1940) designed a method for detecting poliomyelitis virus in sewage and from stool samples. The method consisted of making aqueous suspensions of human stools or sewage and precipitating the suspensions with 50 per cent saturation of ammonium sulfate. The precipitate was dialyzed against water to remove the ammonium sulfate without loss of viral activity. Melnick (1947) in studying the various techniques used for isolating enteric viruses from sewage found the most sensitive method to be a combination of ammonium sulfate precipitation followed by concentration in the ultracentrifuge. Kelly (1953) reported the recovery of virus from sewage employing a simple, rapid and comparatively safe method using absorption on, and elution from, ion exchange resins. Stevenson gt El- (1956) developed a method for concentrating dilute virus 15 suspensions'byalum flocculation. The method proved to be more sensitive than the ion exchange method described pre- viously and possessed the additional advantage of being economical and simple. Tissue culture methods have greatly facilitated the isolation and identification of the enteric viruses. Enders .gt.al. (1949) first utilized tissue culture methods by cultivating the 3 strains of poliomyelitis virus in human embryonic tissues. As a result of these techniques the viruses could be studied on a larger scale with less expense than was formerly possible. Milzer gt 31. (1950) succeeded in confirming the propagation of poliomyelitis virus by cultivating the Lansing strain in various human embryonic tissue cultures. Following this report, Robbins §£_§l, (1951), using tissue culture methods, isolated poliomyelitis virus directly from the patient's feces. In 1952 Robbins __t.§l. cultivated poliomyelitis virus in roller-tube cultures of embryonic skin, muscle and fascia, mature human uterus, kidney and testes,and kidney and testicular tissues from rhesus monkeys. Riordan gt a1. (1952) reported isolating and serologically typing strains of poliomyelitis virus from human stools and spinal cords in roller-tube cultures of monkey testes. Scherer gt 31. (1953) demonstrated propagation of all 3 types of poliomyelitis viruses in human uterine 16 carcinoma cell cultures (HeLa) established by Gey 2E.El- (1952). The virus multiplied and caused degeneration and destruction of the cancer cells within 12 to 96 hours after inoculation. These experimental results demonstrated the usefulness of the HeLa cells for quantitating the polio- myelitis virus, measuring poliomyelitis antibodies and general laboratory production of the virus. Syverton and Scherer (1954) found that monkey kidney epithelial cells were more susceptible to infection by poliomyelitis virus than was the HeLa cell line. Weller (1953) stated that conditions observed in yit£2_were different from those obtained.ig ‘ziyg and that human tissue culture systems promised an alternative, in certain cases, to the study of the infective entities of man that possess a high degree of host specificity. Melnick et a1. (1954) compared standard suspensions of monkey kidney epithelial monolayers and fibroblast mono- layer cultures of monkey testes with suspended fragments of kidney epithelial tissue for sensitivity in primary isolation of poliomyelitis virus. Results indicated that the monkey kidney epithelial monolayer tissue culture was more sensitive than the other methods and tissues used in this study. With the development of a procedure by Dulbecco and Vogt (1954) for the preparation of cell suspensions of monkey kidney by trypsinization, an important step in tissue culture methods 17 was achieved. Younger (1954) improved this original technique thus enabling investigators to prepare standard suspensions of monkey kidney cells which could be used to prepare large numbers of replicate cultures for use in iSOlation and identification of the enteric viruses. In 1954 Dulbecco and Vogt (1954) devised a plaque method for counting the enteric viruses. Monkey kidney anui testicular monolayer cells were used in this study with pure lines of the 3 types of poliomyelitis viruses. The authors concluded that each plaque originated from a single virus particle, defined as the virus unit that was inseparable by dilution and that the number of plaques was equivalent to the virus concentration. Wenner and Miller (1954) compared monkey testicular cell cultures, HeLa cell cultures and monkey kidney epithelial cell cultures for recovering poliomyelitis viruses from human sources. Under the conditions of the study, monkey epithelial cell cultures were superior to HeLa or monkey testicular cultures in the primary isolation of poliomyelitis virus from oropharyngeal specimens. Kibrick gt a1. (1953) compared the sensitivity of various human cell cultures with monkey kidney epithelial cell cultures in the detection of poliomyelitis virus from human sources. They found the monkey epithelial cell cultures 18 to be more sensitive than the other cell cultures. Hsuing and Melnick (1955) simplified Dulbecco's method of plaque formation by growing the monkey kidney epithelial cells in ordinary prescription bottles with rubber stoppers. These same authors (1957a) were able to group provisionally 70 isolates of the ECHO viruses by plaque morphology. Serological procedures were then used to type the various groups of viruses. In the hands of the authors this plaque procedure eliminated the time consuming serological grouping that would have been necessary before individual antigenic types could be determined. In the same year Hsuing and Melnick (1957b) compared the susceptibility of kidney cells from different species of monkeys to the enteric viruses. They found that the Macaca cynamalogus kidney cells were highly susceptible to infection by the enteric virus group as were the rhesus monkey cells. As a result of these findings the monkey kidney epithelial cell monolayer is the cell line of choice for use in enteric virus investigations. The present program of isolating viruses from sewage is best summarized by Kabler (1959) who stated, "Enteric viruses and pathogenic fungi have also been recovered from sewage, but it is recognized that numerical results are only useful approximations because of the lack of sensitivity and wide variability in the procedures along with the unpredictable 19 effects of interferring organisms and other organic components." In this work to be presented quantitative recoveries of virus have been obtained. 20 MATERIALS AND METHODS Collection of samples: East Lansing's treatment plant is of the activated sludge type and is unique in the fact that there is little or no industrial waste material processed by the plant. Previous studies by Mack §t_§l, (1958) have established the presence of viral agents in the sewage of this treatment plant. The next course of study was to determine the stepwise virus concentration in sewage during the process of sewage treatment. Based on the above criteria this plant is ideally suited for the exacting methods that are required in the quantitative recovery of viruses from sewage. The designed flow of this particular sewage treatment plant is 3.7 million gallons daily (mgd). Maximum daily flow through the plant occurs between 9:00 a.m. and 12:00 noon. The plant normally serves a residential area of 30,148 people when the university is in session. During the summer months the number of persons decreases by approximately 20,000 people. The final effluent was exposed to an average of 0.5 parts per million (ppm) of chlorine residual for 15 minutes. The concentration varied from 0.2 ppm at night to 4.0 ppm during the heavy flow period. The average hydrogen ion concentration (pH) of the final effluent was between 7.2 and 7.5. 21 Sewage samples were collected three times a week during the period of June 6, 1959, through November 7, 1959. A total of 685 samples were collected during this period. Pad, dip and raw sludge samples were collected at the same period. The pads consisted of 4 1/2 x 4 1/2 inch absorbent cotton placed between two layers of cheese cloth and sewed together around the periphery. The pads were placed in the flowing sewage and upon eXtraction yieldéd approximately 150 ml of liquid. Dip samples were collected by permitting 200 ml of sewage to flow into a wide mouth bottle. The raw sludge was accumulated in a holding tank and pumped to the digestor daily. Raw sludge samples represented a 24 hour collection of 200 ml of sludge. All samples were brought to the laboratory immediately after collection and stored at -20 C. In order to determine the amount of enteric viruses in sewage during the process of sewage treatment, samples were collected (figure 1) stepwise throughout the processing procedure. Samples taken from the influent channel directly behind the bar screen were labeled station A. Samples taken from the weir of the primary settling tank were labeled station B. Samples taken from the spillway of the final settling tank were designated as station C and samples obtained from the chlorinated effluent, at the 22 FIGURE I. A o'IAGRAMA'rIc srspwnss COLLECTION av snruous or SEWAGE SAMPLES THROUGHOUT THE PROCESSING PLANT. PIIMI EFFLUENT ill CAOOAOII 23 outfall to the river, or cascades were labeled station D. Laboratory processing of samples: The collecting bottles containing the frozen pad samples were placed in cool tap water to hasten thawing. After thawing, the pad was removed from the bottle and placed in a large funnel. The liquid content of the pad (approximately 50 ml) was removed into a 25 x 150 mm test tube. To remove the remaining fluid from each pad, the pad was cut in half with a sterile instrument and the pieces of pad were placed into the barrel of a 100 cc syringe. Pressure was applied to the plunger of the syringe removing approximately 50 m1 of additional fluid. To prevent accidental loss of a single sample in processing, each sample after mixing, was divided into two equal parts in separate tubes. The duplicate samples were centrifuged at 3,000 rpm (1,800 x g) for 15 minutes in a horiZontal centrifuge to remove gross particles. Ten milli- liters of the supernatant fluids were removed from each of the tubes and placed in two 16 x 125 mm screw-capped test tubes and labeled. One-hundred and twenty ug of streptomycin and 30,000 units of penicillin were added to each duplicate sample to eliminate viable bacteria. The samples remained at room temperature for 4 hours before they were tested for sterility. Three tubes each containing 10 ml of brain heart inquion broth in 16 X 125 mm screw-capped tubes were 24 inoculated with 0.1 m1 of each sewage sample. The Specimen was considered free of viable bacteria if the brain heart infusion broth indicated no bacterial activity at the end of a 72 hour incubation period at 37 C. If the specimen contained viable bacteria it was centrifuged at 9000 rpm (6000 x g) for 15 minutes in an International refrigerated multi-speed centrifuge, model PR-l. The supernatant fluid was removed and an additional 20,000 units of penicillin and 100 ug of streptomycin were added per sample. The samples were again allowed to remain at room temperature for 4 hours and the sterility tests repeated. A sample was discarded after three such cycles if it continued to show the presence of viable bacteria in sterility tests. The dip and sludge samples were treated in a similar manner . Tissue culture methods: Macaca gynamalogus kidney epithelial monolayer tube cultures were obtained from Microbiological Associates, Bethesda, Maryland. These tubes were used to isolate and confirm the presence of virus in the processed sewage specimens. Monkeys of the same species weighing approximately 1 to 2 kilograms were obtained from Trefflick's Animal and Bird Company, New York City. Kidneys from these animals were 25 processed in the laboratory and served as a source of cells for preparing monolayers for plaque determinations. Lactalbumin hydrolysate medium (Melnick, 1952) was used to grow and maintain the kidney cells. The medium was prepared in the laboratory and consisted of Hank's basal salt solution, lactalbumin hydrolysate, sodium bicarbonate, antibiotics and inactivated calf serum. A ten times con- centrated stock solution of Hank's basal salt solution was prepared and consisted of the following components: Hank's basal salt solution 10X CaCl2 -------------------------------------- 1.4 g Glucose ----------------------------------- —l0.0 g NaC1 ---------------- - ---------------------- 80.0 g KCL ---------------------------------------- 4.0 g MgSO~'7HOH --------------------------------- 2.0 g KHZP 4 ------------------------------------- 0.6 g NaZHPO4-2HOH ------------------------------- 0.6 g Phenol red --------------------------------- 0.2 9 These components were dissolved in 1100 ml of triple distilled water, dispensed in screw capped bottles and autoclaved at 10 lbs. pressure (115 C) for 10 minutes. This was one of the components used in preparing the lactalbumin hydrolysate medium. The formulation of the lactalbumin hydrolysate medium is as follows: 26 Lactalbumin hydrolysate medium per cent Hank's BSS 10X ---------------------------- 10.0 Lactalbumin hydrolysate ------------------- 0.5 NaHCO ------------------------------------ 0.875 Penic1llin ----------------------- — -------- 100 units Streptomycin ------------------------------ 100 ug Inactivated calf serum -------------------- 1.0 (maintenance) 8.0 (growth) Blood was obtained aseptically from two-month-old healthy calves. The blood was allowed to clot and was then stored at 3-4 C overnight. The serum was removed and placed in 16 x 125 mm screw-capped tubes and inactivated at 56 C for 30 minutes. Serum sterility tests were made in brain heart infusion broth in triplicate and the sterile serum was stored at -20 C. Virus isolation and confirmation: The standard tissue culture tubes of monkey kidney cell sheets from Microbiological Associates were used in isolating and confirming the presence of viruses in the processed sewage specimens. Upon arrival at the laboratory the kidney cells were observed and if growth was not confluent, the medium was changed and the tubes were held at 37 C until the cell sheets were fully developed. The cultures were inoculated, in triplicate, with 0.5 ml of the processed sewage and observed daily under the microscope for an average of 14 days for signs of cytopathic effect (OPE). 27 Inocula exceeding this amount proved to be, in many instances, toxic for the supporting cell sheets. The first visible sign of viral activity was the degeneration of the epithelial cells shown by small clusters of cells becoming rounded and transparent. In this early stage, a partial release of the cells from the glass surface of the tubes could be observed. Some of the cells became stellate-like and as viral multipli- cation continued a characteristic rounding of the cell population developed. Finally the cells began to detach from the glass surface which interrupted the continuity of the monolayer. The culture fluid was harvested when the majority of the cells was released from the surface. Cytopathic effect was usually observed from the second to the ninth day after inoculation with the sewage specimens. All three tubes of a series did not always show CPE. The nutrient fluids from tubes demonstrating CPE were harvested. Confirmation of virus isolation consisted of inoculating a fresh monolayer of cells with 0.5 m1 of undiluted harvested nutrient fulid. These cells were observed for CPE. A sewage specimen was considered confirmed and ready for plaque count determinations if CPE occurred in this culture within 14 days. In most cases CPE could be observed within 4 days. The nutrient fluids from all positive cultures were stored at -20 C until plaque counts were made. 28 Preparation of cell cultures forgplaque count determinations: Kidneys were aseptically removed from exsanguinated .M. cynamalogus monkeys and placed in a sterile Petri dish. The kidneys were decapsulated and placed in maintenance medium. The cortices of the kidneys were removed and the remaining portions discarded. The tissue was cut into pieces about 1 cm2, transferred to a trypsinization flask and covered with 0.25 per cent trypsin solution prewarmed to 37 C. The trypsin flask was incubated for 1 hour at 37 C. After incubation the trypsin was decanted and fresh, Prewarmed trypsin was added to the flask. The tissue fragments and trypsin were stirred with a magnetic stirrer for 8 minutes at 37 C. The speed of the stirring was sufficient to allow rapid mixing without foaming. After 8 minutes of stirring the flask was removed from the magnetic stirrer and the tissue fragments were allowed to settle. Thetrypsin solution, containing the free cells and tissue debris, was decanted into a sterile 50 ml conical centrifuge tube. Fresh prewarmed trypsin was added to the trypsinizing flask and the stirring procedure repeated. Similar cycles were continued until the tissue was exhausted. The centrifuge tube containing the trypsin, free cells and cellular debris was centrifuged in a horizontal centrifuge at 600 rpm (75 x g) for 4 minutes. The trypsin 29 solution was removed and the packed cellular material was washed twice with maintenance medium. All the cells obtained from the trypsinizing procedure were accumulated in a sterile 50 m1 centrifuge tube. The suspended cells were filtered through two layers of sterile cheese cloth to remove connective tissue and cellular debris. FolloWing filtration the cells were centrifuged and resuspended in 15 m1 of lactalbumin hydrolysate growth medium. A Spencer, bright-line hemacytometer was used to determine the number of kidney cells obtained as a result of trypsinization. One-tenth milliliter of the cell suspension was added to 0.9 m1 of a crystal violet staining solution. The staining solution consisted of 2.1 per cent citric acid and 0.1 per cent crystal violet dissolved in double distilled water. The 1:10 dilution of the kidney cell suspension was thoroughly mixed and a portion of this suspension was used to fill the counting chamber of the hemacytometer. Only those cells showing both nuclei and attached cytoplasm were counted. The cells appearing in the four large corner squares, normally used When counting white blood cells, were counted and an arithmetic average was obtained. This figure was multiplied by the dilution factor and the number of monkey kidney cells per ml of the original suspension was determined. The cell suspension was then diluted in lactalbumin hydrolysate nutrient 30 growth medium to give 300,000 cells per m1. Three ml of this suspension were seeded into eadh French square culture bottle and after 5 days incubation in a stationary, inclined position at 37 C a confluent sheet of growth developed. After the fifth day of incubation the growth medium was replaced‘and the cell sheets were incubated an additional 2 days to allow for maximum cell sheet development. The cell sheets were then considered ready for use. Plaque studies: Preliminary studies were performed in order to deter- mine which plaque method would prove to be most satisfactory for use in this study. The procedure of Dulbecco and Vogt (1954) was first applied using 1 ounce French square bottles in place of unsealed Petri dishes. Dilute samples of the three known groups of enteric viruses were placed on kidney monolayer cells. The cultures were incubated at 37 C for 1 hour to allow the virus particles to attach to the cells. An overlay medium consisting of 1.5 per cent Noble agar in Earle's salt solution with 0.0017 per cent neutral red was added to each bottle to cover the cell sheet. A commercially prepared stock solution, five times concentrated of Earle's salt solution was used and consisted of the following components: 31 Earle's salt solution 5X 9/1 NaC1 --------------------------------- 34 0 KCl ---------------------------------- 2 .0 CaCl -------------------------------- 1.0 MgSO ~7HOH --------------------------- 1.0 NaH 0 -HPH -------------------------- 0.625 Glucose ------------------------------ 5.0 After the agar solidified, the bottles were inverted and incubated at 37 C for plaque development. The agar confined the virus infection to the immediate area of the cells that were exposed to each infective particle introduced in the inoculum. The neutral red was employed to outline the area of infected cells. With this method the cell sheet degenerated within 48 hours. Decreasing agar concentration, varying the amount of neutral red in the overlay and staining the developing plaques with neutral red after 72 hours incubation improved results, but some of the plaques of the various species of virus used had not developed by this time. The neutral red dye precipitated in some instances resulting in the fomration of crystals on the cell sheets. (Trypan blue was substituted for neutral red under the same conditions but gave equally poor results. Klein and Goodgal (1959) found that monkey kidney monolayer cell cultures underwent degeneration within 24 hours after exposure to white light in the presence of neutral red. Prior to this work Darnell 33 31. (1958) studied the effect 32 of neutral red on plaque formation. They used monolayers of HeLa cells in Petri dishes incubated with type I poliomyelitis virus for 75 minutes at 37 C in 5 per cent carbon dioxide to allow the virus particles to attach to the cell sheet. Half the plates were then overlaid with nutrient agar con- taining neutral red and the other half with the same medium lacking neutral red. The plates were incubated in carbon dioxide atmosphere for 72 hours. At this time the plates, without dye, were stained with Earle's saline solution containing neutral red. Although the plaque size was unaffected, the number of plaques was reduced 50 per cent in the plates containing neutral red. These studies suggested that the use of a plaque technique which did not incorporate neutral red in the procedure would be more advantageous. Attempts were made to follow Holland and McLaren's (1959) plaque technique. Known samples of the three groups of enteric viruses were inoculated into monkey kidney mono- layers grown in 1 ounce French square bottles. A low agar concentration (0.6%) without neutral red was employed in the overlay medium. When the French squares were inverted so that the cell monolayer could be observed microscopically, the agar overlay slipped off the cell sheet surface. The procedure was originally designed for use with viruses whose period of plaque development was known. After plaques had attained 33 optimal size the agar overlay could be removed and the firmly adherent non-cytopathically affected cells of the monolayer could be fixed and stained with crystal violet. Originally this procedure did not necessitate daily observation of the cell sheets for determining the presence of developing plaque areas. A second disappointing feature of this plaque method was that the cell sheets started to degenerate on the third day after the addition of the agar overlay medium. This was probably due to the lack of available nutrient materials. It was necessary in this present work to keep the cell monolayers viable for a minimum of 7 to 8 days and preferably for 14 days so that the slow plaque forming viruses such as the ECHO viruses and some of the Coxsackie viruSes could develop and produce plaques. In 1959 Sommerville developed a microplaque method for counting enteric virus particles. His procedure was based on the small isolated foci of rounded degenerating cells preceding the appearance of a fully developed plaque. No dyes were used and the method did not depend upon an agar overlay. In the hands of the author the formation of these microplaques gave an accurate and reproducible method for counting enteric viruses. The cell sheets were washed three times with maintenance medium to remove unattached virus particles prior to incubation and subsequent plaque development. 34 The monolayers were observed with the high powered lens of the light microscope. The microplaques developed rapidly and in some instances secondary microplaques developed as a result of prolonged incubation. This method was promising, but the rapid development of secondary microplaques combined with the requirement of frequent microscopic examination of the cell sheets rendered the method unsuitable for use in this study. For the purpose of this work a plaque technique was needed in which the cell sheet would remain viable for a minimum of 14 days and not be subject to the disadvantages of the previously mentioned methods. Preliminary studies resulted in an improved semi-microplaque teChnique for deter- mining the virus concentrations in the sewage samples. This technique was found to be reproducible and is described as follows: The maintenance medium was removed from the French square bottles containing confluent sheets of epithelial cells. The cells were inoculated with varying volumes of the sewage samples containing the unknown viruses to insure the formation of plaques that were separate and discrete. Immediately following inoculation the bottles were placed in the incubator at 37 C for 1 hour to allow the viruses to attach to the cells. The bottles were gently rotated every 15 minutes to assure 35 even distribution of the inoculum over the entire surface of the monolayer. The excess inoculum was removed from the bottles and the cell sheets were covered with 2.5 ml of 1.5 per cent Noble agar containing lactalbumin hydrolysate medium with 2 per cent inactivated calf serum. After the agar over- lay solidified, 3 ml of fluid maintenance medium was added to each culture bottle to nourish the cells. The maintenance medium was changed, depending on the condition of the cell monolayers and variations observed in the indicator system in the medium. The inoculated cultures were inverted for incu- bation at 37 C and observed daily for 14 days for plaque development. Using this method the cell sheets remained viable. A Bausch and Lomb series B, stereomicroscope with a variable power pod of IX to 30X total magnification was used to observe the cell sheets for the presence of developing plaques. The first visible signs of plaque formation occurred between the first and the ninth day after inoculation. Small areas of degeneration of the cells were the first signs of plaque development. This was followed by small individual clustering of cells throughout the cell sheets as observed in figure 2. The number of developing plaques was dependent upon the size of the inoculum and the concentration of virus in the specimen. The developed plaques (see figure 4) could 36 Figure 2. Individual clustering of cells throughout the cell monolayer prior to full plaque develop- ment, fixed and stained with crystal violet (11X). Figure 3. Normal control cell monolayer fixed and stained with crystal violet (11X). 37 Figure 4. Full plaque development 48 to 72 hours after the appearance of small areas of degeneration, fixed and stained with crystal violet (11X). Figure 5. Normal control cell monolayer fixed and stained with crystal violet (11X). 38 be observed within 48 to 72 hours after the appearance of the small areas of degeneration. These areas of degeneration and plaque development were compared with normal control cell monolayers (figure 3 and 5). To prevent confluent lysis of the cell sheet, 96 hours after the first visible appearance of a plaque, the cultures were fixed with 70 per cent ethanol for five minutes. After the alcohol was removed the cells were stained with 0.1 per cent crystal violet for five minutes and washed three times with tap water. By permitting the fixed and stained cultures to remain at room temperature overnight the stain penetrated the agar and stained the cell sheets. This staining procedure greatly facilitated plaque counting and preserved the plaques until counts could be made. A typical French square overlaid with agar and stained with crystal violet is presented in figure 6. Plaque counts: Four culture bottles each were inoculated with the undiluted sewage samples containing virus. After development, the plaque areas in each culture were counted and an arithmetic mean of the counts was determined for each sample. The volume of the inoculum varied in samples because of the amount of virus present in each sample. The final counts were computed to represent plaque forming units per milliliter of sample. Figure 6. A typical French square bottle overlaid with agar and stained with crystal violet. 0AM“; Pipotbv Fraction Colochntcmaphy 5 Beck: 39 40 Ultracentrifuge concentration of samples: Viruses were not found in samples from the chlorinated effluent when tested by the methods above. Because of the treatment and the presence of chlorine this was to be expected. To determine whether the effluent samples were completely free of viable viruses, all pad and dip samples collected in July were pooled into two separate groups. The total volume of the pad pool was 175 ml and the dip pool contained 125 m1. Both pools were concentrated in the ultracentrifuge at 42,040 rpm (114,610 x g) for 1 hour in the Spinco model E ultracentrifuge. The supernatant fluids were removed and the sediment from each sample was resuspended in 5 ml of sterile distilled water. Reasons for using this volume of diluent were to permit sufficient volume for virus i801ation followed by repeated plaque counts should a virus be isolated. Virus was not recovered from these samples. Additional pools were made of pad and dip specimens collected in August which were concentrated in the ultracentrifuge. Since virus was not recovered in any of the concentrated pooled samples, further pools were made of the concentrates. All effluent samples for the months of July and August that had been concentrated in the ultracentrifuge were combined in a single sample of approximately 40 ml. This pool was ultracentrifuged at 110,660 x g for 1 hour and reconstituted in 5 ml of sterile 41 distilled water. In addition to the effluent samples collected during July and August two additional pools consisting of 355 ml each were made from effluent samples collected on the same day that virus was isolated from other stations in the plant. These two pools consisted of both dip and pad samples and were also concentrated in the ultracentrifuge. As a result of this treatment all the samples collected from the chlorinated effluent for the months of July and August and on those days when virus was isolated from other stations in the plant were contained in 3 concentrated samples. The 3 pooled ultra- centrifuged samples were tested for the presence of virus in the conventional manner. To have a valid basis for determining the amount of concentration obtained by ultracentrifugation, a known suspension of poliomyelitis virus type II diluted in sterilized sewage was studied. The virus suspension was concentrated in the ultracentrifuge as described previously. Plaque counts of the known suspension before and after ultracentrifugation were used to determine the degree of concentration. 42 RESULTS During 5 consecutive months, June 6, 1959, through November 7, 1959, a total of 616 sewage specimens were tested in monkey kidney tissue culture. Virus was isolated a total of 54 times resulting in 8.8 per cent recoveries. The total per cent virus recoveries per month is illustrated in Table I. As can be seen, the largest per cent of virus isolations occurred in July when 13.1 per cent of the samples contained virus. This was followed by 9.5 per cent for August, 9.5 per cent for September, 6.5 per cent for October and 3.4 per cent for June. Only 24 samples were tested during November and viruses were not encountered. A comparison is made in Table II of the per cent of virus recovered by the three sampling methods. Although the largest number of specimens tested were dip samples (293) only once was virus recovered from this type of sample. When the 268 pad samples were tested, virus was recovered 44 times. In 55 sludge samples, virus was recovered 6 times. The per cent virus recovery from the dip, pad and sludge samples was 0.3, 16.4 and 10.8 per cent respectively. The results show that the pad sampling was superior to dip sampling under the experimental conditions. The dip sampling method detected only the virus present at the time of collection. The pads on the other hand, were exposed to the sewage flow for 48 to 43 Table I. Total monthly per cent recovery of virus from sewage. Month of collection No. positive/No. tested Per cent virus 1959 isolations June 4/117 3.4 July 18/137 13.1 August 10/105 9.5 September 12/125 9.5 October 7/107 6.5 November 0/24- 0.0 Total *51/616 8.8 *This total does not include the three positive pooled and concentrated dip and pad effluent samples. 44 Table II. Per cent virus recovery from dip, pad and sludge samples. Sampling Total samples Total positive Per cent method tested samples positive Dip 293 1 0.3 Pad 268 44 t 16.4. Sludge 55 6 10.8 Total 616 *51 8.8 *This total does not include the three positive pooled and concentrated dip and pad effluent samples. 45 72 hours. The dip sampling method proved disappointing. Originally it was planned to compare the dip and pad sampling methods. After virus was isolated from any Sewage sample the original sample was set aside until all samples had been tested for the presence of virus. The original samples that were previously shown to contain virus were then tested to determine the number of virus particles per m1. Varied volumes from 0.3 to 0.75 ml were used to determine the number of plaque forming units (pfu) contained in each sewage sample. The volume of inoculum depended on the concentration of virus contained in each original sewage sample. Four culture bottles each were inoculated with a given volume of the undiluted sewage sample containing virus. The plaque areas in each culture were counted and an arithmetic mean of the counts was determined for each sample. The final counts were computed to represent plaque forming units per milliliter (pfu/ml) of the original sample. Each plaque represented at least one infectious virus particle. Duplicate plaque counts were determined whenever possible to verify results. The average pfu/ml for each sample for each collecting station each month is presented in Table III. Average plaque counts on the four viruses isolated in June were 22.7 in the influent, 28.7 in the primary settling and 28.7 in the final settling tank. Table III. Total monthly average in pfu/ml at each collecting station. STATIONS Month Number of A B C D collected samples influent primary final sludge effluent sett. sett. June 4 22.7 28.7 28.7 0.0 0.0 July 18 17.2 51.4 43.3 28.4 0.0 August 10 28.6 40.6 195.0 76.0 0.0 September 12 62.8 48.5 36.0 26.5 0.0 October 7 48.5 96.6 0.0 34.2 0.0 47 There was a slight increase in virus particles per ml progressively through the disposal system. No virus was recovered from the sludge samples collected in June. In July there was a greater number of virus particles present in the primary settling samples (51.4) as compared to the influent samples (17.2). The virus particle concentration decreased slightly in the final settling samples (43.3) and the average sludge samples for July contained 28.4 pfu/ml. In August the trend seen in the previous two months continued and 28.6, 40.6 and 195.0 pfu/ml were found in the influent, primary and final settling stations respectively. A total of 76.0 pfu/ml was obtained from the sludge samples for the month of August. During September and October the virus particles per ml were 62.8 and 48.5 pfu/ml in the respective influent samples and in September the virus particles per m1 decreased as treatment was continued. There was 48.5 and 36.0 pfu/ml in the primary and final settling samples. In October there were large numbers of virus particles per m1 (96.6) in the primary settling, but none was found in the final settling. The number of pfu/ml in the sludge for these months decreased respectively to 26.5 and 34.2 pfu/ml. In general the data in Table III do not correlate with the crude percentages as given in Table I. This is 48 understandable as two different methods of analysis were used. The trend presented in Table III illustrates that as the sewage was processed there was a release of individual virus particles, and the sampling of the influent did not necessari- ly represent the total virus particles present in the other locations or sampling stations. Since no virus was recovered from the effluent when the samples were tested for the presence of virus in the conventional manner, the pad and dip samples collected from the effluent for the month of July were combined into four separate pools containing 70, 105, 105 and 120 ml respectively, totaling 400 ml. The samples were concentrated by ultra- centrifugation at 114,610 x g for 1 hour. (All but 5 ml of the supernatant fluid was removed from each sample and 5 m1 of water was added to each concentrated pool. The water was added to the concentrated sewage to give a volume with which to make repeated plaque counts. When these samples were tested in tissue culture no viruses were isolated. The August pad and dip samples were made into four pools containing 95, 70, 100 and 60 ml respectively (total 325 ml). After concentration, each pool was tested for virus in tissue culture. Again no virus was recovered. The concentration was 5 volumes for the July pools and 4 for the August pools. Since virus was not encountered by the above concentrations, 49 the 5 m1 remaining in each of the eight pools for July and August were combined, giving 40 ml. The 40 ml of this pool was combined with 690 m1 of the original supernatant fluid and ultracentrifuged at 114,610 x g for 1 hour. The super- natant fluid was removed permitting 5 ml to remain on the pellet. An additional 5 m1 of water was used to wash the walls of the tubes, giving 10 m1 total volume. This concentrated pool, when tested in tissue culture, was positive for the presence of virus. The total concentration, by volume, was 19.7 times for the four original July samples and 16.1 times for the August samples. The combined concentration was 18.3 times for the pool of the two month samples. When virus was recovered from the effluent samples by pooling and concentration, additional pools were made and concentrated. Tests on the unconcentrated samples in the plant were not found to contain viruses each day and the effluent would probably not contain large amounts of virus on those days. Pad and dip samples from the effluent were pooled only on those days when viruses were already known to be present in the plant. Twelve pad and 12 dip effluent samples collected on June 6 through September 6 (excluding July and August) were combined in a 355 ml pool. A second pool of 17 pad and 16 dip effluent samples of September 12 through October 24 contained 355 ml. The pooled samples 50 were ultracentrifuged at 114,610 x g and all but 11 m1 of the supernatant fluid was removed from the sediment. Three m1 of water was then added to give a sufficient volume of material with which to fill the ultracentrifuge tubes for further concentration. The two 14 m1 pooled samples were ultra- centrifuged again at 110,660 x g for 1 hour. The supernatant fluid was removed permitting 1 ml to remain on the pellet. An additional 9 ml of water was used to wash the tube walls giving a total volume of 10 m1. These concentrated pooled samples were positive for the presence of virus when tested in tissue culture. The total concentration for each pool was 35.7 times by volume. In order to determine the virus particle concentration resulting from ultracentrifugation of the pooled effluent samples a known suspension of poliomyelitis virus type II was diluted in sterilized sewage and concentrated in the ultra- centrifuge. Two ml of a mouse brain suspension containing poliomyelitis virus was added to 98 m1 of sterilized processed sewage. The suspension was thoroughly mixed and divided into two portions of 66 and 34 ml respectively. The smaller portion was used to determine the number of virus particles per ml in the original unconcentrated suspension. The 66 m1 portion was concentrated in the ultracentrifuge in the same manner as those pooled effluent samples of June 6-30 51 and September 12 through October 24. The results of this experiment are presented in Table IV. The average pfu/ml in the 10-1 dilution of the unconcentrated original suSpension was 6.7 as compared to 141.7 in the concentrated suspension. The virus was concentrated 21.1 times for each plaque forming unit contained in 1 ml of the unconcentrated original suspension as a result of ultracentrifugation. This represents an increase of 1.3 logarithmic units. Small volumes of inocula (0.3 to 0.5 m1) of the con- centrated effluent fluids were used in determining pfu/ml as compared to the larger volumes (0.75 ml) used in testing the other unconcentrated samples collected from other stations throughout the plant. The cell monolayers did not tolerate the toxic substances concentrated along with the viruses during ultracentrifugation when larger volumes of the con- centrated effluent samples were used for inoculation. In order to determine the actual number of pfu/ml in the original effluent samples, the number of pfu/ml obtained from each of the 3 pooled and concentrated samples was divided by the number of times the viruses in the known suspension were concentrated (21.1) following ultracentrifugation. The total pfu/ml following concentration for the pooled samples for July and August was 50.6, 37.0 for June 6-30 and September 1-5 and 36.0 for September 12 through October 24. By dividing 52 Table IV. Concentration in pfu/ml by ultracentrifugation of known poliomyelitis virus in sterilized sewage. Sample after Dilution Original sample ultracentrifugation *Ave. count _pfu/ml *Ave. count pfu/ml Original 30.0 50.0 I**TNC **TNC 10"1 4.0 6.7 85.0 141.7 10’2 1.3 2.2 10.1 16.8 -3 10 1.0 .1.7 2.5 4.2 *AVerage of four individual plaque counts. **Too numerous to count. 53 these counts by the concentration factor (21.1) the actual number of pfu/ml in the original unconcentrated effluent samples was 2.4, 1.7 and 1.7 pfu/ml for July and August, June 6-30 and September 1-5, and September 12 through October 24 respectively, or a total average of 1.9 pfu/ml. Based on the corrected average pfu/ml recovered from these samples less than 1 virus particle per ml was contained in the largest volume of inoculum used in determining pfu/ml of the effluent prior to concentration. This would probably account for the failure to recover virus from the chlorinated effluent samples prior to ultracentrifugation and the subsequent recovery of virus following such treatment. The completed sewage treat- ment reduced, but did not eliminate all of the virus particles present in the sewage undergoing treatment. In order to show further the effect of sewage treat- ment on the removal of virus particles contained in the sewage undergoing treatment, the total average number of virus pfu/ml recovered from the sewage for the entire sample collecting period is presented in Table V. The influent contained 34.9 and the primary settling samples contained 52.1 pfu/ml. From the final settling tank a greater number of virus particles (75.7) was obtained as compared to the activated sludge (36.7). This was followed by a drop in the chlorinated effluent to a corrected 1.9 pfu/ml. The general trend of an 54 Table V. Total average of pfu/m1 of viruses in sewage during processing. J t It Total number of Total number of pfu/ml Station positive pad pfu samples A-influent 15 524.4 34.9 B-primary settling 25 1303.0 52.1 C-final settling 4 303.0 75.7 Sludge 6 220.0 36.7 *D-effluent 3 5.7 1.9 *Represents pooled dip and pad samples from the effluent after concentration. 55 increase in virus particles as the sewage underwent treatment up to and including the final settling tank (Table V) correlated well with the similar increase in virus particles (seen in Table III. Based on the data tabulated in Table III (total monthly average in pfu/ml at each collecting station) and Table V (total average in pfu/m1 of viruses in sewage during processing) it appears that there was a release of virus particles as the sewage material underwent treatment. The number was greatly reduced as the sewage passed through the chlorine holding tank. In an attempt to determine if this trend of virus particle increase up to and including the final settling tank was affected by the day to day variations in virus con- centrations entering the plant, the following comparisons were made. There was no single instance where virus was recovered from all collecting stations on the same day; however, virus was recovered 18 times from samples collected on 9 dif- ferent days from two or more stations on the same day. The total monthly average in pfu/m1 recovered from two or more samples collected on the same day is presented in Table VI. There was an increase in pfu/ml in each instance when the number of virus particles recovered from samples collected from the influent were compared with those recovered from 56 Table VI. Total monthly average pfu/ml of virus from samples collected on the same date from different stations. STATIONS Number of A B C D Month samples influent primary final effluent sett. sett. July 10 17.6 57.2 43.3 August 2 0.0 10.0 195.0 *1.9 September 4 64.2 91.7 36.0 October 2 20.9 124.2 0.0 *Represents pooled dip and pad samples from effluent after concentration. 57 the primary settling tank. When the number of virus particles recovered from the influent samples was compared with the number recovered from the final settling tank there was an increase in pfu/m1 in the latter for the months of July and August but a decrease for the months of September and October. In general the number of pfu/ml recovered from both the primary and final settling tanks was greater than the number recovered from the influent. By computing the total average number of virus pfu/ml recovered from each station for the entire collecting period for those days in which two or more viruses were isolated from different stations on the same day (Table VII), the virus particle increase was even more apparent. Total averages of 27.2, 65.0 and 91.4 pfu/ml were recovered from the influent, primary settling tank and final settling tank respectively. Based on the corrected average, a decrease to 1.9 pfu/m1 was obtained from samples collected from the chlorinated effluent. These data indicated that the virus particles were released and increased in number as a result of sewage treatment up to and including the final settling tank. Following chlorination this number decreased to a minimum. In order to study further the effects of the treat— ment on the viruses as they passed through the disposal system the per cent of viruses recovered from the pad samples 58 Table VII. Total average number of viruses in pfu/m1 compiled from data where viruses were isolated from two or more stations on the same day. Total number Average Station positive samples Total pfu pfu/ml A-influent 8 217.5 27.2 B-primary settling 7 454.7 65.0 C-final settling 3 274.3 91.4 *D-effluent 3 5.7 1.9 *Represents pooled dip and pad samples from the effluent after concentration. 59 at each collecting station throughout the plant, including the pooled dip and pad samples collected from the chlorinated effluent, were compiled and presented in Table VIII. As can be seen the per cent virus recovery based on the pad sampling method from the influent, primary settling tank, final settling tank and chlorinated effluent was 25.0, 24.3, 8.3 and 2.6 per cent respectively. It was apparent that the per cent recovery of virus decreased stepwise through the plant as the sewage underwent treatment while the virus concentration expressed in pfu/ml increased as indicated in Tables III, V, VI and VII. By comparing the total average virus pfu/ml entering the disposal plant with the total number leaving the plant in the chlorinated effluent the total per cent reduction of virus particles resulting from the sewage treatment process could be determined. A corrected total average of 1.9 pfu/m1 of virus was obtained from the original chlorinated effluent samples before concentration as compared to 36.0 pfu/ml obtained from the influent samples (Table III). Based on these comparative figures the virus concentration was reduced 94.7, or approximately 95 per cent, as a result of the treatment process. Approximately 5 per cent of the virus particles entering the plant survived the treatment process. Table VIII. \. 60 The per cent of viruses isolated by the pad sampling method from each station throughout the sewage treatment plant. Station Numbergpositive Per cent positive Number tested A-influent B-primary settling C—final settling D-effluent 15/60 25.0 25/103 . 24.3 4/48 8.3 *3/117 2.6 *Represents pooled dip and pad samples from the effluent after concentration. 61 DISCUSSION This was a quantitative study of viruses contained in samples collected from each step in the sewage processing plant. Sewage samples represent the fecal material from a population and this justifies a comparison of the results presented here with the results obtained by collecting stool samples directly from the population. In Table IX a t 31, (1957), comparison is made of the results of Gelfand who isolated the enteric viruses directly from the population, to the work of Mack §__§l, (1958) who isolated the same viruses from East Lansing sewage. Further comparison is made to the results found in the present study. All data are presented in the number of specimens tested and the per cent viruses recovered. The results presented in the comparison (Table IX) were obtained by isolating the viruses in monkey kidney cell cultures. The work of Mack g; 31. (1958) was the result of ultracentrifuge concentration of all samples prior to testing for virus. This explains the lower recoveries of virus obtained in the present study. Because large volumes of inocula were required in a quantitative study, concentration of the samples would not have permitted a sufficient volume of sample to make numerous plaque counts on each sample. 62 m.m mam m.¢a maoa ¢.MH boom amuoe 0.0 gm m.ma mm m.m Ham HoQEo>oz m.o boa a.mm vga H.0m mam nonouoo m.m own o.ma mom 6.m~ mom nmnamummm m.m moa o.mm OOH m.mm mom umsmsa a.ma nmn 6.44 mm m.sm mam sass ¢.m baa H.m om h.oa Nam wasp m>HuHm0Q powwow o>wuflmom pounou o>HuHmom pmumou soapooaaou ucoo Mom .02 ucmo Mom .02 name mom .oz coefloomm m0 nusoz mmma sz mmma ..Hm um xomz hmma ..Hm um mammaow .monEmm Hooum Eonm pocflmuno macaumaomfi ou pmnmmaoo mm msofifloomm mmmBom Eoum mcoflumaoma msufl> .XH magma 63 The marked seasonal incidence of the enteric viruses in feces and sewage has been repeatedly shown by Melnick 1947, Melnick gt 21., 1954a, Honig §t_§1,, 1956, Gelfand .gt'al., 1957 and Mack gt 21,, 1958. The months of June to November were chosen for the present study because of the higher incidence of viruses present in sewage during this period. The present study (Table II) showed a high recovery of viruses from the pad samples (16.4%) and compares favorably with the work of Mack 33 31. (1958), where concentration methods were used. The dip samples, however, produced a low recovery of viruses. This again emphasizes the fact that the pad sampling method produced favorable results without additional procedures. The plaque count results obtained were surprising. It was thought that as the sewage was processed through the disposal system, less virus would be recovered at each step in the processing. This was nOt the case. The total average number of virus particles recovered from the sewage for the entire collecting period showed an increase in the number of virus particles as the sewage underwent treatment. The increase (Table V) in virus particles per ml continued from the influent (34.9 pfu/m1) through the final settling tank where 76.7 pfu/m1 were recovered. The results clearly 64 indicate that there were more virus particles in the plant than was indicated by the recovery of virus at the influent. The data used in compiling Table V included all plaque counts from all samples containing viruses. To attempt to clarify the results obtained in Table V the data from which the table was made were arranged differently. The data of Table VII were compiled in pfu/m1 of viruses contained in samples where viruses were isolated from two or more stations on the same day. When the results given in the two tables (Table V and Table VII) were compared, the trend was found to be the same. More virus particles were again found circulating within the plant than could be accounted for by the pfu/m1 present in the influent. An explanation of the results obtained could be made by the activated sludge treatment in the plant. The procedures involved in the treatment process consisted of three basic steps: primary settling, aeration and final settling. Approximately 60 per cent of the suspended solid materials were removed from the sewage in the primary settling tank. The sewage was then treated by the aeration process to convert the organic materials by zooqloeal action into microbial flora. The activated sludge employed in the aeration process was composed of a gelatinous mass containing a stabilized flora of bacteria and protozoa. The sewage undergoing treatment was combined with the activated sludge and subjected to 65 mechanical aeration. Aeration was accamplished by agitating the combined activated sludge and sewage materials for 5 1/2 hours by means of mechanically driven propellers located near the bottom of the aeration tank. Following agitation the zoogloeal masses settled to the bottom of the aeration tank. The activated sludge was continually recirculated. Ninety per cent of the activated sludge was combined with the effluent of the primary settling tank and circulated back into the aeration tank. The remaining 10 per cent was returned to the primary settling tank and pumped to the digestor along with the settled solids of the primary settling tank. The mechanical and biological action resulting from the aeration process would disperse the organic materials releasing individual virus particles. The released virus particles would recirculate resulting in larger numbers of viruses recovered from the final settling tank (Table VII). The gauze pads were suspended in the sewage flow for 48 to 72 hours. Virus particles recovered in tissue cell cultures represented particles expressed from the pad samples and probably represented an accumulation. The pad did not represent a filter for the entire volume of flow in any station but was a sample. The almost complete lack of Virus in the dip samples gave evidence that the pad accumulated viruses, although for how long and under what 66 physical and chemical conditions viruses survive in pads is not known. One reason that the influent did not show greater isolations of viruses may be that the virus particles were contained in organic material of a size that did not permit passage through the collecting pad. Thus, after the organic materials were acted upon in the various processes of the plant, the pad acted as a filter and accumulated the virus particles. There was a relationship between the number of virus particles recovered from the pads at different stations throughout the plant. A comparison of the total average number of viruses recovered from the influent with the corrected total average recovered from the effluent would give the total per cent reduction of viruses resulting from the treatment process. Based on this crude comparison the virus con- centration entering the plant was reduced approximately 95 per cent as a result of the treatment process. This meant that approximately 5 per cent of the virus concentration survived the treatment process and was discharged in the effluent. Therefore, chlorination reduced but did not eliminate the viable virus particles in the discharged effluent. Investigations have been carried out in the past to determine what effect, if any, conventional chlorination procedures have on the virus of poliomyelitis. It is recog- nized that the efficiency of chlorination is dependent on the 67 interrelation of co-existing conditions. Some of the factors that play prominent roles are the amount of organic material and amount of ammonia present, temperature and pH of the system, chlorine dosage and type of test used to determine residuals. Weidenkopf (1958) studied the inactivation of type I poliomyelitis virus with chlorine. Highly purified virus suspensions containing a minimum amount of organic material were used to hold chlorine losses to insignificant levels. The plaque method of assay was used to determine the per cent survival. Results indicated that the rate of inacti- vation of poliomyelitis virus was independent of the virus concentration. These experiments were carried out in phosphate buffered saline solutions. The presence of great quantities of organic matter in sewage plant effluents would render the establishment of free residual chlorine difficult. Kelly and Sanderson (1959) reported that effluents from secondary treatment plants that had been chlorinated for disinfection contained enteric viruses about a third of the time. Poliomyelitis virus type I, Coxsackie and ECHO viruses were all isolated from both the influent and effluent. Residual chlorine values reported for these samples were near the recommended value of 0.5 ppm. The chlorinationypracticed, however, had some destructive action on viruses, as plaque 68 counts of chlorinated effluents were lower than those obtained from the influent samples. Raw sewage which had been chlori- nated to reduce nuisance in plants, involving concentrations of 0.1 and 0.3 ppm residual chlorine, contained viruses as often as non-chlorinated raw sewage. The effluent was chlorinated throughout the course of this study. The average final effluent exposure in the chlorine contact chamber was 15 minutes. Attempts were made to maintain an average chlorine residual of 0.5 ppm; however, concentrations varied from 0.2 ppm to 4.0 ppm. As a result of this chlorination procedure the virus concentration of the sewage was reduced approximately 95 per cent. These results lend additional support to the results obtained by these previous workers. The potential reservoir of virus disease is great in raw sewage, contaminated water supplies, inadequately treated sewage and activated sludge materials. Chlorine residuals should be increased if destruction of viral pathogens is desired. A total average of 31.9 pfu/ml recovered from the sludge samples (Table V and Table VII) suggest that wet sludge and settled solids are unsuitable for use in fertilizing vegetables, lawns and shrubbery. Proper chlorine dosages, retention times, and residual levels must be devised to fit the specific need. Continuous re-evaluation of the 69 standards for disinfection of sewage may be necessary. Many kinds of viruses have been isolated from sewage and each strain has its individual characteristics of resistance, virulence and stability. An adequate standard for disinfection should take into account the expected range of variation of these properties. 70 SUMMARY A total of 616 sewage samples were tested for the presence of enteric viruses in monkey kidney tissue culture. Gauze pad and raw sludge samples provided the best source of material for isolation of enteric viruses. Fifty-four samples were shown to contain virus. Plaque counts were made on each of the original positive samples to determine the number of virus particles per unit volume. A definite increase in virus particles expressed in plaque forming units per m1 (pfu/m1) occurred in the treatment process of sewage through the plant up to and including the final settling tank. The virus concentration in sewage entering the treatment plant was reduced 94.7 per cent as a result of the treatment and exposure of the final effluent for an average of 15 minutes to 0.5 parts per million of residual free chlorine at a pH of 7.2 to 7.5. This reduction occurred under actual field conditions. 6. 71 A new, improved, semi-micro plaque technique for counting enteric viruses was developed and used in counting the viruses contained in each original positive sample. Sewage samples provide a source of material from which the viral flora of a community may be determined. 72 BIBLIOGRAPHY ANONYMOUS 1940 Poliomyelitis virus in sewer water. J. Amer. Med. Assoc., 114, 2232. CASEY, A. E. AND B. J. AYMOND. 1940 The distribution of poliomyelitis in Louisiana. Science, 21, 17. CLARK, E. M., D. S. KNOWLES, F. T. SHIMADA, A. H. RHODEs, R. C. RITCHIE AND W. L. DONOHUE. 1951 Coxsackie virus in urban sewage. Can. J. Public Health, 42, 103-110. COMMITTEE ON THE ECHO VIRUSES. 1955 Enteric cytopathogenic human orphan (ECHO) viruses. Science, 122, 1187-1188. COMMITTEE ON THE ENTEROVIRUSES, NATIONAL FOUNDATION FOR INFANTILE PARALYSIS. 1957 The enteroviruses. Am. J. Public Health, 41, 1556-1566. DALLDORF, G.‘AND G. M. SICKLES. 1948 An unidentified titerable agent, isolated from the feces of children with paralysis. Science, 108, 61-62. DARNELL, J. E. JR., R. Z. LOCKART, JR. AND T; K. SAWYER, 1958 The effect of neutral red on plaque formation in two virus-cell systems. Virology, é, 567-568. DULBECCO, R. AND M. VOGT. 1954 Plaque formation and iso- lation of pure lines with poliomyelitis viruses. J. Expt. Med., 22, 167-182. ELLSWORTH, S. M. 1940 Can infantile paralysis be spread by bathing in sewage-polluted waters? New Engl. J. Med., 55, 222. ENDERS, J. F., T. H. WELLER AND F. C. ROBBINS. 1949 Cultivation of the Lansing strain of poliomyelitis virus in cultures of various human embryonic tissues. Science, 199, 85-87. FRANCIS, T. JR., G. C. BROWN AND J. D. AINSLIE. 1953 Poliomyelitis in Hidalgo County, Texas, 1948: Polio- myelitis and Coxsackie viruses in privy specimens. Am. J. Hyg., 58, 310-318. 73 GARD, S. 1940 Method for detecting poliomyelitis virus in sewage and stools. J. Exptl. Med., 15, 779. GELFAND, H. M., J. P. FOX AND D. R. LEBLANC. 1957 The 1 enteric viral flora of a population of normal children in southern Louisiana. Am. J. Trop. Med. Hyg., é, 521-530. GEY, G. O., W. D. COFFMAN AND M. T. KUBICEK. 1952. Tissue culture studies of the proliferative capacity of cervical carcinoma and normal epithelium. Cancer Res., 5;, 264-265. GIOVANARDI, A. AND F. BERGAMINI. 1959 Richerche Italiane sui virus enterici. Ann. Sclavo., 5, 94-111. HARGREAVES, E. R. 1950 Poliomyelitis in Cornwall in 1949. Brit. Med. J., 4658, 879-883. HOLLAND, J. J. AND L. C. MCLAREN. 1959 Improved method for staining cell monolayers for virus plaque counts. J. Bacteriol., 15, 596-597. HONIG, E. I., J L. MELNICK, P. ISSACSON, R. PARR, I. L. MEYERS AND M. WALTSON. 1956 An endimiological study of enteric virus infections. J. Exptl. Med., 595, 247-262. HORSTMANN, D. M. 1955 Endemic virus infections in Egypt: Isolation of polio viruses and other tissue culture cytopathogenic agents from infants. Fed. Proc., 55, 466. HORSTMANN, D. M., R. WARD AND J. L. MELNICK. 1944 Persistence of virus excretion in the stools of poliomyelitis patients. J. Am. Med. Assoc., 126, 1061-1062. HORSTMANN, D. M., R. WARD AND J. L. MELNICK. 1946 The isolation of poliomyelitis virus from human extra- neural sources. III. Persistence of virus in stools after acute infections. J. Clin. Invest., 25, 278- 283. HOWE, H. A. AND D. BODIAN. 1939 Production of experimental poliomyelitis from untreated stools. Proc. Soc. Exptl. Biol. Med., 55, 538. 74 HSIUNG, G. C. AND J. L. MELNICK. 1955 Plaque formation with poliomyelitis, Coxsackie, and Orphan (ECHO) viruses in bottle cultures of monkey epithelial cells. Virology, ;, 533-535. HSIUNG, G. D. AND J. L. MELNICK. 1957a Orphan viruses of man and animals. Ann. N. Y. Acad. Sci., 19, 342- 361. HSIUNG, G. D. AND J. L. MELNICK. 1957b Comparative sus- ceptibility of kidney cells from different monkey species to enteric viruses (poliomyelitis, Coxsackie and ECHO groups). J Immunol., 15, 137-146. KABLER, P. 1959 Removal of pathogenic microorganisms by sewage treatment processes. Sewage and Ind. Wastes, 5;, 1373-1382. KELLY, S. M. 1953 Detection and occurrence of Coxsackie viruses in sewage. Am. J. Public Health, 35, 1532- 1538. KELLY, S. M., M. E. CLARK AND M. B. COLEMAN. 1955 Demon- stration of infectious agents in sewage. Am. J. KELLY, S. M. AND W. W. SANDERSON. 1959 The effect of sewage treatment on viruses. Sewage and Ind. Wastes, QL, 683-689. KIBRICK, S., J. F. ENDERS AND F. C. ROBBINS. 1953 An evaluation of the roller tube tissue culture for the isolation of poliomyelitis viruses from feces. J. Immunol., 15, 391-400. KLEIN, S. W. AND S. H. GOODGAL. 1959 Photodynamic inacti- vation of monkey kidney cell monolayers. Science, 130, 629. KLING, C., G. OLIN, J. H. MAGNUSSON AND S. GARD. 1939 Nouvelles recherches sur l'elimination du virus poliomyelitique par 1es matieres fecales. Bull. Acad. de med. Paris, 52;, 826. KLING, C., A. PATTERSON AND W. WERNSTEDT. 1912 The presence of the microbe of infantile paralysis in human beings. State Med. Institute, Stockholm, 5, 5. 75 KRAMER, S. D. 1940 Virus of poliomyelitis: Survival for over 6 months in the feces of a contact. J. Michigan State Med. Soc., 55, 766. KRAMER, S. D., A. G. GILLIAM AND J. G. MOLNER. 1939 Re- covery of the virus of poliomyelitis from the stools of healthy contacts in an institutional outbreak. Public Health Reprs., 53, 1914. LEPINE, P. 1939 On the virus of poliomyelitis in intestinal contents of healthy children. Internat. Bull. Econ. Med. Research Pub. Hyg., A40, 57. LEPINE, P., P. SEDALLIAN AND V. SAUTTER. 1939 Sur la presence du virus poliomyelitique dans 1es matieres fecales et sa longue duree d'elimination chez un porteur sain. Bull. Acad. de med. Paris, 522, 57. LEVADITI, C. 1940 Sur la presence du virus poliomyelitique dans 1es eaux d'egout. Bull. Acad. de med. Paris, 123, 23-30. LITTLE, G. M. 1954 Poliomyelitis and water supply. Can. MACK, W. N., W. L. MALLMANN, H. H. BLOOM AND B. J. KRUEGER. 1958 Isolation of enteric viruses and Salmonellae from sewage. Sewage and Ind. Wastes, 59, 957-962. MCCLURE, G. Y. AND A. D. LANGMUIR. 1942 Search for carriers in an outbreak of acute anterior poliomyelitis in a rural community. The incidence of virus in feces. Am. J. Hyg., §§, 285. MCDANIEL, G. E. 1940 A preliminary discussion of polio- myelitis in South Carolina in 1939. J. South Carolina Med. Assoc., 55, 331. MELNICK, J. L. 1947 Poliomyelitis virus in sewage in epidemic and nonepidemic times. Am. J. Hyg., g5, 240-253. MELNICK, J. L. AND K. AGREN. 1952 Poliomyelitis and Coxsackie viruses isolated from normal infants in Egypt. Proc. Soc. Exptl. Biol. Med., 55, 621-624. 76 MELNICK, J. L., J. EMMONS, J. H. COFFEY AND H. F; SCHOOF. .1954a Seasonal distribution of Coxsackie viruses in urban sewage and flies. Am. J. Hyg., 55, 164-184. MELNICK, J. L., J. EMMONS, E. M. OPTON AND J. H. COFFEY. 1954b Coxsackie viruses from sewage. Methodology including an evaluation of the grab sample and gauze pad collection procedures. Am. J. Hyg., 55, 185-195. MELNICK, J. L., C. RAPPAPORT, D. D. BANKER AND P. N. BHOTT. 1955 Stabilized suspensions of monkey kidney cells suitable for intercontenental shipment. Proc. Soc. Exptl. Biol. Med., 14, 136-138. MELNICK, J. L. AND J. T. RIORDAN. 1952 Poliomyelitis viruses in tissue culture. IV. Protein-free nutrient media in stationary and roller tube cultures. Proc. Soc. Exptl. Biol. Med., £1, 208-213. MELNICK, J. L., J. T. RIORDAN, E. L. CURNEN AND A. D. MACRAE. 1953 Tissue culture pathogenic agents other than poliovirus isolated from cases of suspected polio- myelitis. Fed. Proc., 52, 454. MELNICK, J L., E. W. SHAW AND E. C. CURNEO. 1949 A Virus isolated from patients diagnosed as non-paralytic poliomyelitis or aseptic meningitis. Proc. Soc. Exptl. Biol. Med., 11, 244-249. MILZER, A., O. LEVINSON, K. VANDERBOON AND P. ADELMAN. 1950 Cultivation of poliomyelitis virus (Lansing strain) in human embryonic and placental tissues. Proc. Exptl. Biol. Med., 13, 136-138. MUNDEL, B., J. H. S. GEAR AND D. WILSON. 1947 The distri- bution of virus of poliomyelitis in a sewage pruifi- cation works.in Johannesburg, South Africa. S. African Med. J., 29, 336. PAUL, J. R., J. D. TRASK AND C. S. CULOTTA. 1939 Polio- myelitic virus in sewage. Science, 59, 258. PAUL, J. R., J. D. TRASK AND S. GARD. 1940a Poliomyelitic virus in urban sewage. J. Exptl. Med., 1;, 765. 77 PAUL, J. R., A. J. VIGNEC AND J. D. TRASK. 1940 The clinical epidemiology of poliomyelitis. Proc. Third Inter- nat. Cong. for Microbiology, Williams and Wilkins, Baltimore, Maryland. PISZCZEK, E. A., H. J. SHAUGHNESSY, J. ZICHIS AND S. O. LEVINSON. 1941 Acute anterior poliomyelitis: Study of an outbreak in west suburban Cook County, Illinois; preliminary report. J. Am. Med. Assoc., 551, 1962. RAMOS-ALVAREZ, M. AND A. B. SABIN. 1954 Characteristics of poliomyelitis and other enteric viruses recovered in tissue culture from healthy American children. Proc. Soc. Exptl. Biol. Med., 51, 655-661. RAMOS-ALVAREZ, M. AND A. B. SABIN. 1956 Intestinal viral flora of healthy children demonstrable by monkey kidney tissue culture. Am. J. Public Health, 35, 295-299. RHODES} A. J., E. M. CLARK, D. S. KNOWLES, A. M. GOODFELLOW AND W. L. DONOHUE. 1950 Prolonged survival of human poliomyelitis virus in experimentally infected river water. Can. J. Public Health, 3;, 1950. RIORDAN, J. T., N. LEDINKO AND J. L. MELNICK. 1952 Multiplication of poliomyelitis viruses in tissue cultures of monkey testes. II. Direct isolation and typing of strains from‘human stools and spinal cords in roller tubes. Am. J. Hyg., 55, 339-346. ROBBINS, F. C., J. F. ENDERS, T. H. WELLER AND B. FLORENTINO. 1951 Studies on the cultivation of poliomyelitis viruses in tissue culture. V. The direct isolation and serologic identification of virus strains in tissue culture from patients with non-paralytic polio- myelitis. Am. J. Hyg., 53, 286-292. ROBBINS, F. C., T. H. WELLER AND J. F. ENDERS. 1952 Studies on the cultivation of poliomyelitis viruses in tissue culture. J. Immunol., 55, 673-694. SAWYER, W. A. 1915-16. An epidemiological study of polio- myelitis. Am. J. Trop. Disease, 5, 164. 78 SCHERER, W. F., J. F. SYVERTON AND G. O. GEY. 1953 Studies on the propagation lp_vitro of poliomyelitis viruses. IV. Viral multiplication in a stable strain of human malignant epithelial cells (strain HeLa) derived from an epidermoid carcinoma of the cervix. J. Exptl. Med., 51, 695-710. SILVERMAN, A. C. 1941 Institutional outbreak of polio- myelitis. Am. J. Public Health, 5;, 593. SOMMERVILLE, R. G. 1959 A microplaque method for counting enterovirus particles. Virology, 5, 701-702. STEIGMAN, A. J., U. PENTTI KOKKO AND R. J. SILVERBERG. 1953 Mack virus-serum and gamma globulin neutralization of unidentified agents isolated from suspected non- paralytic poliomyelitis. Proc. Soc. Exptl. Biol. Med., 55, 200-204. STEVENSON, R. B., S. L. CHANG, N. A. CLARKE AND P. W. KABLER. 1956 Concentration of dilute virus suspension by alum floculation. Proc. Soc. Exptl. Biol. Med., .52, 764. SYVERTON, J. T. AND W. F. SCHERER. 1954 Studies on the propagation of strain HeLa human epithelial cells for isolation and typing. J. Lab. Clin. Med., 3;, 286-302. TRASK, J D. 1939 Modes of transmission in poliomyelitis. J. Conn. State Med. Soc., 5, 595. TRASK, J. D. AND J. R. PAUL. 1942 Periodic examination of sewage for the virus of poliomyelitis. J. Exptl. med 0 ' 22, 1-120 TRASK, J. D., A. J. VIGNEC AND J. R. PAUL. 1938 Polio- * myelitis virus in human stools. J. Am. Med. Assoc., 111, 6. VIGNAC, A. J., J. R. PAUL AND J. D. TRASK. 1939 Poliomyelitic t/irus from feces in non-paralytic poliomyelitis; infectivity by various routes. Proc. Soc. Exptl. Biol. Med., 4;, 246. 79 l WEIDENKOPF, S. J. 1958 Inactivation of type I poliomyelitis virus with chlorine. Virology, 2, 56-67. WELLER, T. H. 1953 The application of tissue-culture methods to the study of poliomyelitis. New Engl. J. Med., 249, 186-194. WENNER, H. A. AND C. A. MILLER. 1954 Comparison of methods for recovering poliomyelitis viruses from human sources. Proc. Soc. Exptl. Biol. Med., 55, 11-15. YOUNGER, J. S. 1954 Monolayer tissue cultures, I. Pre- paration of standardization of suspensions of trypsin- dispersed monkey kidney cells. Proc. Soc. Exptl. Biol. Med., 55, 203-205. '2'. 3 .0081 USE 0 »- I. 1.. . t , r _. I,IL'L ,.r . i. y .nloilllullvts Vial, . .,,. 1,, 91.1.-) intuit . . .0 , ”'11113111111111filifll’flllfljljifll1T1"s