Filtration Method for Bacteriophage Detection

Transcription

1 APPLIED MICROBIOLOGY, Nov., American Society for Microbiology Vol. 13, No. 6 Printed in U.S.A. Filtration Method for Bacteriophage Detection RAYMOND C. LOEHR AND DIANN T. SCHWEGLER Environmental Health Engineering Laboratory, University of Kansas, Lawrence, Kansas Received for publication 19 July 1965 ABSTRACT LOEHR, RAYMOND C. (University of Kansas, Lawrence), AND DIANN T. SCHWEGLER. Filtration method for bacteriophage detection. Appl. Microbiol. 13: A filtration method has been developed which can be used to detect and enumerate phage in low concentrations directly from solution without the need for prior concentration. In this method, a known volume of the phage solution is mixed with a suitable host solution. Samples are filtered through membrane filters; the filter is removed and incubated, and after 24 hr the resultant plaques are counted and the titer is calculated. Escherichia coli B and the coliphage T2 were used in these studies. Host cultures less than 12 hr old produced the best results. Approximately 11 host organisms must be present in the sample taken for filtration. To avoid phage reproduction, all steps prior to filtration must be done in less than 45 min. The method was compared with the soft-agar technique and was shown to be less precise but able to measure phage in lower concentrations. An increasing number of viruses have been found in man (Huebner, 1959) and, hence, have the potential of being part of sewage-polluted waters. Coxsackievirus, poliovirus, and other viruses have been isolated from sewage. Several epidemics and a number of lesser viral diseases have been attributed to water-borne virus (Clarke and Chang, 1959; Mosley, 1963; Weibel et al., 1964). The density of enteric virus in sewage has been indicated as between 2 and 4 units per 1 ml, depending upon the season of the year and the temperature of the sewage (Kelly and Sanderson, 196). Clarke et al. (1964) recently summarized the survival and removability of enteric viruses in water and wastewater. Virological techniques have not been developed to a point where virus enumeration can be recommended as a routine procedure in the microbiological examination of waters. The gauze-pad technique, shown by Melnick et al. (1954) to be superior to grab samples, is the technique currently used to detect viruses in waters suspected of pollution (Wiley et al., 1962). The gauze-pad technique has provided invaluable information on the occurrence of relative numbers of viruses in wastewaters. However, very little is known about the seasonal variations in virus concentration and occurrence in a community, and about the virus pollution in wastewater treatment plant effluents and its effect upon downstream users of such waters. As new epidemiological and diagnostic techniques are developed, it is quite probable that even greater occurrences of disease may be traced 15 to water-borne viruses. To maintain the production of high-quality potable water at water treatment plants and to prevent the transmission of water-borne viruses, the virological quality of wastewater and treated waters should be monitored on a regular basis. Because of the equipment and experience necessary, the gauze-pad technique is not a technique that can be used routinely by health agencies or water treatment personnel to monitor potable and wastewaters. Although many pathogenic bacteria are present in polluted waters, no attempt is made to assess the bacteriological quality of such water by determining the concentration of specific pathogens. Instead, indicator organisms, the coliform group, are used to evaluate the bacteriological quality of water. Since the coliform organism can be used to measure the bacteriological quality of water, the viruses of these bacteria, the coliphage, may be able to be used to measure the virological quality of water. In experiments with the coliphage as a pollution indicator (Committee on Public Health Activities, 1961), the phage content of well waters was found to be much higher in summer than in winter, increased after periods of rain, and higher in wells liable to contamination by surface water. Gildermiester and Watanabe (1932) examined waters suspected of pollution for the phage of the typhoid, paratyphoid, and coliform bacteria. The phage content of such waters correlated with the degree of pollution. Initial research to develop a routine virological method applicable to polluted water is described in this paper. Escherichia coli B and the T2 coli-

2 16 LOEHR AND SCHWEGLER APPL. MICROIBIOL. phage were used to evaluate the method because: (i) bacteriophage can be easily maintained and handled in a laboratory, and (ii) the coliform organisms are used as a measure of the bacteriological quality of water, and procedures for handling and detecting these organisms are common in most local, state, and federal agencies. MATERIALS AND METHODS Routine methods. E. coli strain B and the T2 bacteriophage were initially obtained from the Department of Microbiology, University of Kansas, and from S. E. Luria, Massachusetts Institute of Technology. Subsequent quantities were cultured and maintained in the Environmental Health Engineering Laboratory. Coliform counts were determined by the membrane-filter method (American Public Health Association, 196). The soft-agar technique of Adams (1959) was used to determine the titer of the phage. Tryptose Phosphate Broth was used in place of nutrient broth and was used to maintain and dilute the stock phage solutions. Filtration method. Techniques for virus detection and enumeration rely upon the fact that the virus will adsorb onto a suitable host and, during incubation, will produce cytopathic changes or plaques depending upon the virus and host in question. The filtration method developed in the Environmental Health Engineering Laboratory also relies upon adsorption. Dilute T2 phage solutions were mixed with E. coli B solutions. Samples of this mixture were filtered through sterile membrane filters (Millipore). After filtration, the filter was removed, placed in a sterile disposable petri dish on an absorbent pad previously saturated with M-Endo Broth (Difco), inverted, and incubated at 37 C. After 24 hr, the resultant plaques were counted, and the titer of the phage solution was calculated. Comparison. The filtration technique was evaluated by comparing the results obtained with it and with the soft-agar technique of Adams (1959). The comparison of the methods was done by calculating the apparent titer of the stock phage solution used in the tests. This approach was necessary since different phage dilutions were used with each technique. RESULTS Effect of bacterial host. The age of the host bacterial culture affects the ability of the filtration method to detect the phage (Fig. 1). A broth solution of E. coli B was prepared and incubated at 37 C for up to 12 hr prior to use in the filtration method before any resultant decrease in titer was observed. A 1- to 2-hr bacterial growth period was used in all experiments to keep the overall time requirements to a minimum. The optimal quantity of bacteria in a filtered sample was about 19 to 11 bacteria. An exces-. 1..s = INCUBATION TIME (HOURS) FIG. 1. Effect of age of host bacteria (incubation time) on resultant detection of phage. PFUt = plaque-forming units at time t; PFU1 = plaqueforming units at 1 hr of incubation time. Each point is the average of five replicates. h. h. Bl-. 11 el 2 S -t a TIME (MINUTES) S 8 1 FIG. 2. Latent period of TR phage at 22 to 25 C with Tryptose Phosphate Broth as the dilution medium. Each symbol represents a different run. Each point is the average of five replicates. All symbols start at 1 at time-zero. PFUt = plaque-forming units at time t; PFUo = plaque-forming units at time-zero. sive quantity of bacteria on the filter reduces the apparent plaque count by allowing only small plaques to form. The small plaques were difficult to distinguish and count. An insufficient quantity of bacteria prevents formation of the continuous layer of bacteria necessary to avoid interpreting an area where no bacteria have grown as a plaque. It is important to avoid phage overcrowding on a plate. Our experience indicates that the plaque count per plate or dish that is still individually discernible is approximately 3 to 5 plaques. Latent period. The results (Fig. 2) obtained in determining the rapidity with which the phage would multiply indicate that with both methods the latent period was at least 45 min. All mixing and filtering steps in the filtration method can be completed in 3 min or less. The resultant plaque counts therefore represent the original phage in solution. The latent period experiments were done by mixing the bacteria and phage together at room temperature (22 to 25 C). At the specified times, samples were withdrawn from the mixture and the titer was determined by both methods. Mixing. The degree of mixing of the bacteria

3 VOL. 13, 1965 ; B X-- VFTRATION METHOD FOR PHAGE DETECTION x- X MIXING RATE (RPM) FIG. 3. Effect of mechanical mixing rate on phage detection. Open circles = soft-agar method; crosses = filtration method. PFUt = plaque-forming units at indicated mixing rate. PFUo = plaqueforming units at a mixing rate of 6 rev/min. Each point is the average of five replicates. TABLE 1. Accuracy and precision of soft-agar and filtration techniques for phage enumeration* No. of Avg titer of Technique sam- stock phage solu- SD Range pls tion (1" partipisles/mid) Soft-agar Filtration * Samples analyzed from three different experiments run on three nonconsecutive days. and phage is important in the filtration method. Only the phage particles that contact and adsorb onto a bacterial cell will be measured. Mixng thus will enhance the possibility of contact. E coli B and the T2 phage in Tryptose Phosphate Broth were mechanically mixed for 3 min in a laboratory stirring machine (Phipps & Bird, Richmond, Va.) at various paddle speeds (Fig. 3). The phage titer of samples of the mixture was determined by both the filtration and the soft-agar methods. Both methods indicated that the titer decreased at high mixing rates. Because of the increase in titer that occurred with moderate mixing rates, a mixing rate of 45 rev/min for 3 min was incorporated as a part of the filtration method. Mixing at 45 rev/min for up to 45 min did not appear to change the titer. Moderate mixing can be beneficial by providing additional opportunity for contact between the phage and the bacteria. Comparison of filtration and soft-agar methods. In a series of experiments designed to compare the accuracy and precision of both methods on the same sample, 2-liter mixtures of E. coli B and the T2 phage in Tryptose Phosphate Broth were mechanically mixed for 3 min at 45 rev/min. Samples were taken randomly throughout the TABLE 2. Examination offiltrate offiltration method Determination Titer of stock solution (11/ml) Soft-agar method Filtration method Analysis of mixture before filtration Analysis of filtrate after filtration FIG. 4. Virus detection by use of membrane filter. The dark background is a continuous layer of Escherichia coli B. The light spots on the layer of bacteria are plaques formed by the growth of the TR phage. Since a known volume of solution was filtered, the phage count per milliliter can be easily determined. container. Typical results (Table 1) indicate that the filtration method produced titers that were lower, by a factor of approximately 2, than those obtained by the soft-agar method. Examination of the filtrate. The results in Table 1 indicate that the filtration method is not as accurate as the soft-agar method. These lower titers could result if inactivation of the phage occurred during the filtration step or if some of the phage did not adsorb onto the bacteria and passed through the filter with the filtrate. The latter possibility was shown to be the more probable when samples of the filtrate were tested for the phage (Table 2). The sum of the results of the soft-agar method after filtration (1.6 X 11) and the filtration method before filtration (.37 X 11) are close enough to the results of the soft-agar method before filtration (2.8 X 11) to infer that inactivation of the phage was not the major cause of the low results obtained with the filtration method. Effect of phage concentration.. The soft-agar method can not be used to determine the phage titer when phage are present in low concentration. A sample volume of.1 ml is used in this method. Therefore, the lower limit of detection with the soft-agar method is 1 phage per.1 ml or 1 phage per ml. The filtration method does not have

4 18 LOEHR AND SCHWEGLER APPL. MlCROBIOL. this drawback. Any volume of a mixture of E. coli B and the phage can be filtered in an effort to obtain sufficient plaques to establish the phage concentration. Sample volumes of 1, 2, 5, and 1 ml routinely have been used in our laboratory. When different sample volumes are filtered, care must be taken to have sufficient but not excessive amounts of bacteria in the sample. The filtration method can determine phage in a concentration of at least 1 phage per 1 ml. DIscussIoN When bacterial and phage growth are not inhibited and when there are sufficient but not excessive amounts of bacteria on the filter, the filtration method can be used for the detection and enumeration of the phage. Figure 4 illustrates typical plaques obtained by the filtration method. The dark background is a continuous layer of E. coli B. The light spots are plaques formed by the T2 phage. In calculating the titer of the stock phage solution, it was assumed that each plaque originated from only one phage particle. Discrete plaques occur when the phage adsorb onto growing cells, invade, reproduce, and cause cell lysis. Inoculation of E. coli B from a stock slant into Tryptose Phosphate Broth, followed by incubation from 1 to 12 hr, provides sufficient growing bacteria for the filtration and soft-agar methods. Incubation of the bacteria for longer than the above period increases the proportion of dead bacteria to which the phage could also adsorb, thus decreasing the phage titer. The results shown in Fig. 1 are similar to those observed by Delbriick (194), who found that the physiological state of the host bacteria will significantly affect resultant numbers of plaques formed. The latent period is a function of temperature of the phage-bacterial solution as well as of the physiological age of the bacteria (Delbruick, 194; Ellis and Delbriuck, 1939). Phage growth curves at different temperatures and with hosts of different physiological ages have been shown to be qualitatively the same but quantitatively quite different. Under optimal cultural conditions, the minimal latent period for the T2 phage has been reported to be about 21 min (Luria, 1953). The results of experiments reported herein (Fig. 2) are comparable to those obtained by Ellis and Delbriick (1939) at similar temperatures. The latent period is longer when the phage-bacteria mixture is held at lower temperatures. The data in Fig. 3 indicate that the latent period of the T2 phage in these experiments was at least 45 min. This is an advantage for the analyst using either method since it allows ample time for mixing and sampling. The soft-agar method is more accurate than the filtration method (Table 1). In the soft-agar method the entire contents of a.1-ml sample are added to a concentrated E. coli B solution and plated. Therefore, all phage particles, whether or not they are immediately adsorbed, have the opportunity to adsorb, reproduce, and be represented by a plaque. In the filtration method, only those phage particles that are adsorbed onto the bacteria remain on the membrane filter and create a plaque. Unadsorbed phage pass through the filter with the filtrate. The inorganic and organic compounds that are present in the liquid greatly influence the adsorption of the phage to bacteria. Tryptose Phosphate Broth was used in these phage experiments as dilution and maintenance media because it contains the optimal concentration of organic and inorganic cofactors. The filtration method is applicable to solutions which contain low concentrations of phage since any quantity of solution can be filtered to obtain sufficient plaque counts. With this method, a quantity of a 1-hr culture of E. coli B is added to a solution containing the phage and is mixed; suitable samples are then filtered through a membrane filter. After filtration, the filter is placed on an absorbent pad previously saturated with M-Endo Broth and incubated at 37 C. After 24 hr, the resultant plaques are counted and, since the volume that was filtered is known, the phage concentration can be estimated. The quantity of E. coli B added to the phage solution should be enough so that the sample taken for assay will contain about 19 to 11 bacteria. Because the filtration method requires only equipment normally available in a bacteriological laboratory, and because the method is simple, does not require a highly skilled technician, and can be done in a short period of time, the filtration method appears to have significant potential as a routine procedure for the examination of waters for bacteriophage. The coliform organisms are used to measure the virological quality of water. The final verdict on whether the filtration method and the coliphage can be used to measure the virological quality of water will only come after investigation of the relationship of coliphage to coliform organisms and specific enteric viruses in polluted and potable water. Research along these lines is being continued. LITERATURE CITED ADAMS, M. H Bacteriophages. Interscience Publishers, Inc., New York. AMERICAN PUBLIC HEALTH AssoCIATION Standard methods for the examination of water and waste water. 11th ed., p American Public Health Association, Inc., New York. CLARKE, N. A., G. BERG, P. W. KABLER, AND

5 VOL. 13, 1965 FTRATION METHOD FOR PHAGE DETECTION 19 S. L. CHANG Human enteric viruses in water: source, survival and removability, p In Advances in Water Pollution Research, vol. 2. Pergamon Press, New York. CLARKE, N. A., AND S. L. CHANG Enteric viruses in water. J. Am. Water Works Assoc. 51: COMMITTEE ON PUBLIC HEALTH ACTIVITIES Coliform organisms as an index of water safety -Progress Report. J. Sanit. Eng. Div., Am. Soc. Civil Engrs. 87: DELBRUYCK. M The growth of bacteriophage and lysis of the host. J. Gen. Physiol. 23: ELLIS, E. L., AND M. DELBRUCK The growth of bacteriophage. J. Gen. Physiol. 22: GDERMIESTER, E., AND H. WATANABE An investigation of the occurrence of bacteriophage in surface waters. Gesundh. Ingr. 55:241. HUEBNER, R. J Seventy newly recognized viruses in man. Public Health Rept. U.S. 74: KELLY, S. M., AND W. W. SANDERSON Density of enteroviruses in sewage. J. Water Pollution Control Federation 32: LURIA, S. E General virology. John Wiley & Sons, Inc., New York. MELNICK, J. L., J. EMMONS, E. M. OPTON, AND J. H. COFFEY Coxsackie viruses from sewage-methodology including an evaluation of the grab sample and gauze pad collection procedures. Am. J. Hyg. 69: MOSLEY, J. W Epidemiologic aspects of viral agents in relation to waterborne diseases. Public Health Rept. U.S WEIBEL, S. R., F. R. DIXON, R. B. WEIDNER, AND L. J. MCCABE Waterborne-disease outbreaks, J. Am. Water Works Assoc. 56: WEY, J. S., T. D. Y. CHIN, C. R. GRAVELLE, AND S. ROBINSON Enterovirus in sewage during a poliomyelitis epidemic. J. Water Pollution Control Federation 34: Downloaded from on July 19, 218 by guest