Method for Salmonella Concentration from Water at ph 3.5,

Size: px

Start display at page:

Download "Method for Salmonella Concentration from Water at ph 3.5,"

Isaac Chandler
6 years ago
Views:

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 1979, p. 1-6 0099-2240/79/07-0001/06$02.00/0 Vol. 38, No. 1 Method for Salmonella Concentration from Water at ph 3.5, Using Micro-Fiber Glass Filters J. C. BLOCK` AND D.

1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 1979, p /79/ /06$02.00/0 Vol. 38, No. 1 Method for Salmonella Concentration from Water at ph 3.5, Using Micro-Fiber Glass Filters J. C. BLOCK` AND D. ROLLAND2 Institut National de la Sante et de la Recherche Medicale, Research Unit 95,1 and Laboratoire d'hygiene et de Recherche en Sante Publique, Faculte de Medecine, Vandoeuvre-lks-Nancy, France Received for publication 4 April 1979 A method is described for the concentration of Salmonella from water. As is done with enterovirus, Salmonella bacteria were concentrated from water in two steps: by ph 3.5 adsorption on and ph 9.5 elution from 8-,um porosity micro-fiber glass filter tubes. This method worked in less than 30 min, and Salmonella typhimurium was inactivated only slightly in spite of rapid ph variations (ph 3.5 to 9.5). It was demonstrated that the retention by the filters stems from two phenomena: a low retention in the micro-fiber glass labyrinth for small filtered volumes, and a high retention by adsorption at ph 3.5 for any filtered volume (experiments done with 15- and 80-liter samples). Addition in tap water of trivalent ions like Al" did not increase Salmonella adsorption. In most of the trials, Salmonella recovery varied from 42 to 93%. Preliminary field investigations indicate that enterovirus and Salmonella may both be concentrated from the same water sample by this procedure. Several methods have been described for the isolation of Salmonella from water, such as the gauze pads method (12), the diatomaceous earth filtration method (2), and the plate membrane filtration method (13). Recently, a high-volume sampling technique was reported, which allows for an improved quantitative approach (11). However, some disadvantages were encountered with the high-volume sampling technique, in which a Balston type AA (0.2,um) cartridge filter is used. Both filter clogging by suspended matter and slow flow rate contributed to the long processing time necessary for sampling large quantities of water. The current study was concerned with the application to Salmonella of a virus concentration method (6, 8, 15), which uses Balston type C filter tubes (8 am) and acidification of the water samples to ph 3.5. The larger porosity permits a faster filtration of water, and the acidification to ph 3.5 improves the retention of the Salmonella by adsorption to the filters. This paper reports preliminary investigations of Salmonella typhimurium inactivation at different ph values and gives results of the Salmonella adsorption on, and recovery by, alkaline elution from 8-,um porosity micro-fiber glass filters. MATERIALS AND METHODS Bacteria. A strain of S. typhimurium, previously isolated from water and identified by Prof. Le Minor, Institut Pasteur, was used. For each experiment, the bacteria were grown in nutrient broth (Institut Pasteur). After incubation for 18 h at 37 C, the bacterial suspension was centrifuged at 5,000 x g for 5 min. The cellular pellet was washed with sterile 0.8% NaCl solution and recentrifuged. The washing procedure was repeated two times, and finally the bacteria cells were resuspended in 10 ml of sterile 0.8% NaCl solution. Optical densities of appropriate dilutions of cell suspensions were measured at 650 nm. The cell concentration was calculated by reference to a standard curve relating optical density and cell concentration. The suspension concentration usually was 109 bacteria per ml. Depending on the experiment, this stock culture was diluted l0-5 or 106 before use. Salmonella titration. The estimation of Salmonella density in water samples before and after filtration and in the concentrate was carried out by two different procedures. (i) MF technique. The membrane filter (MF) technique was used for water samples containing only Salmonella and no other bacteria. Briefly, a 1- or 50- ml volume of the sample was passed through a sterile membrane filter (porosity, 0.45,Lm; diameter, 47 mm; Sartorius) under partial vacuum. To neutralize the ph of each sample (ph 3.5), the filter was rinsed by filtration of 20 to 30 ml of sterile neutral water. Then the membrane filter was removed and placed on the sterile medium agar (nutrient agar; Institut Pasteur). After incubation of the membrane filter for 24 h at 37 ± 0.5 C, the Salmonella colonies were counted, and the bacterial density was calculated in terms of number of Salmonella per 50 ml. (ii) MPN technique. The most-probable-number (MPN) technique was used for samples containing Salmonella and saprophytic bacteria collected from tap water. Briefly, the samples were diluted (10-fold dilutions) in selenite enrichment broth by the method

2 BLOCK AND ROLLAND of Leifson (Merck). One milliliter of each appropriate dilution of sample was inoculated into selenite enrichment broth (three tubes per dilution; 1 ml per tube).

2 2 BLOCK AND ROLLAND of Leifson (Merck). One milliliter of each appropriate dilution of sample was inoculated into selenite enrichment broth (three tubes per dilution; 1 ml per tube). After a 24-h incubation at 37 ± 0.50C, each tube was submitted to a confirmed test. For this purpose, 0.1 ml of medium from each tube was streaked on Hektoen enteric agar (Oxoid) plates. The plates were incubated at 37 ± 0.5 C for 24 to 48 h. The presence of Salmonella was indicated by typical colonies (green colonies with a black center) which were confirmed by biochemical assays (o-nitrophenyl-f8-d-galactopyranoside negative, H2S positive, glucose positive) and serological agglutinations. The number of Salmonella resulting from multipleportion decimal dilution platings was computed as the combination of positive findings and recorded in terms of the MPN according to the MacGrady tables. Water supply and conditioning. The water used in the experiments was Nancy, France, tap water. The characteristics of the finished water during the period of experimentation were as follows: temperature, 160C; turbidity, 1 formazine unit; ph 7.5; biocarbonates, 65 mg/liter; total hardness, 10 F; chloride, 10 mg/liter; calcium, 25 mg/liter; and magnesium, 8 mg/liter. When sterilized water was used, it was either autoclaved (120 C, 20 min) or chlorinated (10 mg of free chlorine per liter for 18 h) and then neutralized by sodium thiosulfate. The chlorination increased the water ph to 8 to 8.5. Attempts to enhance the Salmonella adsorption were made by the addition of HCI (technical grade; final ph, 3.5), either with or without 5 x 10-' M AlCl3. The water sample volumes varied from 15 to 90 liters. Preliminary investigations. To measure the effect of ph on Salmonella inactivation, experiments were carried out in sterile tap water at ph 7.5, in sterile tap water acidified to ph 3.5, and in a protein solution of 3% beef extract (Difco), ph 9.5. Each of nine bottles (three per ph treatment, 1 liter per bottle) was inoculated with approximately 103 bacteria and stirred gently during the entire experiment. Samples of 50 ml were taken from each bottle at 10, 20, 30, and 40 min, and the Salmonella were enumerated by the MF technique. The inactivation effect of ph 3.5 followed by ph 9.5 was also measured. Each of three bottles (1 liter of sterile water, ph 3.5, per bottle) was inoculated with approximately 103 Salmonella and stirred gently during the entire experiment. Two samples of 50 ml were taken from each bottle every 10 min for 1 h and filtered according to the MF technique. One membrane of each sampling was rinsed with sterile neutral water. The other membrane was first rinsed with 3% beef extract solution, ph 9.5, and then covered with 20 ml of this solution. After 5 min, the beef extract solution was filtered off, and the membrane was rinsed with sterile water. The Salmonella were then enumerated by the MF technique. Salmonella filter systems and sampling apparatus. Epoxy-fiber glass filter tubes 63 mm in length (type C; Balston Inc., Lexington, Mass.) were used as the Salmonella filters in conjunction with a type clear plastic filter holder. The filters were composed of borosilicate glass microfibers bonded with epoxy resin. The retention efficiency for liquids was APPL. ENVIRON. MICROBIOL. 98% for 8-[Lm particles. Both filter and holder were placed in a sampling apparatus. A schematic representation of this is shown in Fig. 1. A pump (Masterflex no ; Bioblock) forced the water sample through the filter tube. The flow rate was kept at 1 to 1.5 liters/ min. Two sampling ports (Fig. 1) were used for taking water samples before and after filtration. Before use, the system was disinfected by a chlorine solution (1,000 mg of free chlorine per liter for 1 h), and then neutralized with sodium thiosulfate. Salmonella-concentrating procedure. The Salmonella-concentrating procedure was done in two steps: filtration followed by elution. Figure 2 gives a schematic representation of a concentration run. (i) Filtration. After the addition of the Salmonella, the water was or was not adjusted (according to the assay) to ph 3.5 and was or was not made up to 5 x 10'- M A1C13. The water samples were pumped through the filter (outside to inside flow). Every 5 min, water samples were taken before and after the filter and submitted to the Salmonella analysis by both MF and MPN techniques. (ii) Elution. After the water was processed, the Salmonella retained by the filters were eluted by passing 250 ml of 3% beef extract solution (ph 9.5) through the filter (inside to outside flow). This protein solution was passed very slowly one time through the filter, and then immediately submitted to Salmonella analysis by the MPN technique. RESULTS S. typhimurium inactivation at different ph values. Some experiments were conducted to determine whether the filtration ph (ph 3.5) and the elution ph (ph 9.5) could affect Salmonella survival and, consequently, the method efficiency (Fig. 3). At ph 3.5 and 9.5, the Salmonella count decreased slowly so that 90 and 91%, respectively, of the bacteria were recovered after 30 min. Moreover, an examination of the standard s shows that the tests (ph 3.5 or 9.5) and their controls (ph 7.5) overlap with respect to mean number of Salmonella recovered. One complementary investigation was carried out reproducing the rapid variations of ph (filtration at ph 3.5 followed by elution at ph 9.5) (Fig. 4). The results were calculated in terms of Salmonella recovery compared with a control of ph 7.5. More Salmonella were recovered at ph 3.5 than after rapid variations of ph (ph 3.5 to 9.5) even in the first few minutes (84 and 85%, respectively). After 30 min, the Salmonella count was decreased so that 80% of the bacteria were recovered at ph 3.5 and only 58% were recovered at ph 3.5 followed by 9.5. Evaluation of Balston filters for Salmonella retention. The results of a series of experimental runs comparing the Salmonella retention by Balston fiber glass filters (8-t,m porosity) at ph 7.5 and 3.5 are given in Table 1. At

3 VOL. 38, 1979 SALMONELLA CONCENTRATION FROM WAT'ER 3 -FILTER HOLDER HCI offand AIC13 PUMP -FILTER TUBE SAMPLING SAMPLING PORT 1 PORT 2 WATER SAMPLE FIG. 1. Salmonella filter system and sampling apparatus..1 \ *---* H 3. _-H 35 "i_ byph. 'suss~ ~ ~~~~ - XI~~~~~~~ -t.j FIG. 2. Schematic representation of the Salmonella-concentrating procedure. 0 s---a ph 3.5 (_r) o ~~ph ~~~~ 5.A (beest s ataeh.ss) T I ae (U"kt ) FIG. 3. S. typhimurium inactivation at different ph values. ph 7.5, mean Salmonella retention varied, according to the volume of water filtered, from 85 to 40% (significant difference). At ph 3.5, mean Salmonella retention did not vary significantly when the filtered volume of water was increased. Consequently, for 12 liters filtered, the results suggest that Salmonella retention was greater and more constant at ph 3.5 than at ph TIME (MhnufIs) FIG. 4. S. typhimurium inactivation at ph 3.5 followed by ph 9.5. The retention capacity of the Balston filters was studied by filtration of larger samples (82 liters) (Fig. 5). The results are expressed in terms of Salmonella retention calculated from initial Salmonella concentration. At ph 8.4, the Salmonella retention decreased from 76 to 5% as the filtered volume increased from 7 to 82 liters. On the other hand, at ph 3.5 the Salmonella retention by fiber glass filters varied from 69 to 67%, without significantly decreasing as the filtered water volume increased. A graphic integration of the curves (Fig. 5) indicated that ph 3.5 water filtration permitted 54% more Salmonella retention than did filtration at ph 8.4. Nine experimental runs were conducted to determine the combined effect of H+ ions (ph 3.5) and A1" ions (AlCl3 final concentration, 5 X 1o-4 M) on Salmonella retention by Balston fiber glass filters (Table 2). For 15 liters filtered at ph 7.8, the average Salmonella retention was 79%. Acidification at ph 3.5 with or without Al3" increased the Salmonella retention by a small but statistically significant amount. However, AlCl3 addition did not produce any significant increase in Salmonella retention.

4 4 BLOCK AND ROLLAND TABLE 1. Salmonella retention by Balston filters (8,um) at ph 7.5 and 3.5 No. of Salmo- % of Salmonella retained for: Water ph Expt no. nella in input 4 liters 8 liters 12 liters (15 liters)' filtered filtered filtered , , , Mean Standard , , , , , Mean Standard a Salmonella enumeration was by the MF technique. ph of the _.ter APPL. ENVIRON. MICROBIOL. 100 z 2I-U z 2 50.j FILTERED VOLUME (Lits) FIG. 5. S. typhimurium retention on Balston filters at ph 7.8 and 3.5. Recovery of Salmonella from water using fiber glass filters. Salmonella retained by the filters were recovered by passing 3% beef extract solution through the filter (inside to outside flow) (Table 3). Because the enumeration was done by the MPN technique, the bacteria recovered in the eluate must be regarded as an estimation rather than an exact number. Consequently, the range of variation can only be estimated. The percent bacterial recovery varied from 4 to 78%, from 64 to 93%, and from 16 to 87% for trials run at ph 7.8, ph 3.5, and ph 3.5 plus AlCl3, respectively. DISCUSSION Bacteria, like viruses, have an isoelectric point (10, 12). For this reason, attempts have been made to apply the virus concentration method (adsorption and elution with fiber glass filters) (6, 8) to Salmonella. In testing the feasibility of this method, it was necessary to determine the survival of Salmonella at ph 3.5 and 9.5, to investigate the Salmonella retention capacity of 8-,um porosity Balston filters, and to determine the recovery of Salmonella by acid adsorption on, and alkaline elution from, micro-fiber glass filters. Experiments have demonstrated that S. typhimurium was only slightly inactivated after 30 min at ph 3.5. This can be correlated with the survival of Salmonella in contaminated acid foods (3). However, a rapid ph variation from 3.5 to 9.5 showed a larger inactivation, one of at least 40%. Such an observation depends on the Salmonella strain. Other studies performed with S. typhosa did not indicate any inactivation after 7 h at ph 10 (1). Nevertheless, the Salmonella filtration and elution must be carried out

VOL. 38, 1979 SALMONELLA CONCENTRATION FROM WATER 5 quickly, preferably in less than 30 min. Balston filter tubes used in these experiments have a retention efficiency of 98% for 8-,um particles.

5 VOL. 38, 1979 SALMONELLA CONCENTRATION FROM WATER 5 quickly, preferably in less than 30 min. Balston filter tubes used in these experiments have a retention efficiency of 98% for 8-,um particles. However, for small water samples (less than 5 liters) and for a 1-liter/min flow rate, it appears that Salmonella (0.6 to 0.8 um) are retained by the filters (approximately 80%). The thickness of the filters, an indirect route through the epoxy-fiber glass, and even some blind alleys (as in a labyrinth) could explain these results. TABLE 2. Salmonella retained by Balston filters (8,pm) at ph 7.8, ph 3.5, andph 3.5 plus AICl3 No. of Sal- % of Sal- Water ph Expt no. monella in input (15 tonella liters)a retained , , , Mean 78.9 Standard , , , Mean 87.2 Standard AlC , , , Mean 88.2 Standard 7.9 asalmonella enumeration was by the MF technique. TABLE 3. Recovery of S. typhimurium from tap water using Balston filters (8 pm) % of Sal- No. of Salmo- monella Water ph Expt no. nella added in recovered 15 litersa by beef extract elution , , , , , , AiC , , , 'Salmonella enumeration was by the MPN technique. Still, for water samples larger than 5 liters, the retention efficiency of the filter decreased rapidly (less than 5% retention for 80-liter samples). Acidification of the water to ph 3.5 solved this problem; at ph 3.5, few Salmonella pass through the filter tubes (no more than 40% of the bacterial population). Consequently, the Salmonella retention by 8-,um porosity Balston filters stems from two phenomena: (i) retention in the epoxymicro-fiber glass labyrinth which becomes rapidly ineffective, and (ii) retention due to adsorption. At ph 3.5, which is below the Salmonella isoelectric point (approximately 4.5 according to Lamanna et al. [10]), the superficial charge of the bacteria cells becomes positive. Under the same conditions, the micro-fiber glass of the filter develops a negative charge (9) and so permits Salmonella adsorption. This adsorption phenomenon has been used previously for a selective separation of Escherichia coli and S. typhimurium (5). Also, for enterovirus concentration, some investigators have improved the adsorption by adding A1Cl3 to a final concentration of 5 x 10-4 M (4, 7, 8, 15). In our experiments, addition in tap water of the trivalent ion A13+ did not increase Salmonella adsorption. Elution of Salmonella from the filters is the more difficult step of the method. Beef extract solution, ph 9.5, is used to reverse the superficial charge of the bacteria, thus permitting desorption. Moreover, this alkaline protein solution is passed through the filter from inside to outside because some Salmonella are trapped in the epoxy-micro-fiber glass labyrinth. In most of the trials, beef extract eluted from 42 to 93% of the Salmonella. In a few trials, the Salmonella recovery was very poor (4 to 16%). These low recoveries were a result of the fragility of the filters (no more than 25 lb/in2 from inside to outside) and of the viscosity of the beef extract solution; therefore, another elution technique was tested. Preliminary data indicate that cutting and homogenizing the filters with beef extract solution would improve recovery (unpublished data). Our first field investigations have permitted indigenous Salmonella recovery (manuscript in preparation). However, the overall findings do not allow extrapolation of the results to natural waters, because the presence of various substances (e.g., mineral and organic matter, other bacteria) may affect Salmonella adsorption. Moreover, the superficial charge of bacteria changes with the age of the cells (14), and so the adsorption behavior also changes. Nevertheless, it is expected that enterovirus and Salmonella will both be concentrated from the same water sample by this procedure.

6 BLOCK AND HOLLAND ACKNOWLEDGMENTS This work was carried out in the Laboratoire d'hygiene et de Recherche en Sante Publique (Director, J. M. Foliguet). We acknowledge L. H. Pottenger for rewriting the paper and J.

6 6 BLOCK AND HOLLAND ACKNOWLEDGMENTS This work was carried out in the Laboratoire d'hygiene et de Recherche en Sante Publique (Director, J. M. Foliguet). We acknowledge L. H. Pottenger for rewriting the paper and J. Bara for typing it. LITERATURE CITED 1. Berg, G., P. V. Scarpino, and D. Berman Survival of bacteria and viruses in natural waters, p In Proceedings of the National Symposium on Quality Standards for Natural Waters, Ann Arbor, Mich. 2. Brezenski, F. T., and R. Russomanno The detection and use of salmonellae in studying polluted tidal estuaries. J. Water Pollut. Control Fed. 41: Buttiaux, R., H. Beerens, and A. Tacquet Manuel de techniques bacteriologiques, 4th ed. Flammarion Paris Publishers, Paris. 4. Farrah, S. R., S. M. Goyal, C. P. Gerba, C. Wallis, and J. L. Melnick Concentration of poliovirus from tap water onto membrane filters with aluminium chloride at ambient ph levels. Appl. Environ. Microbiol. 35: Hall, A. N., S. D. Hogg, and G. 0. Phillips Gradient elution of Salmonella typhimurium and Escherichia coli strains from a DEAE-cellulose column. J. Appl. Bacteriol. 41: Hill, W. F., W. Jakubowski, E. W. Akin, and N. A. Clarke Detection of virus in water: sensitivity of APPL. ENVIRON. MICROBIOL. the tentative standard method for drinking water. Appl. Environ. Microbiol. 31: Homma, A., M. D. Sobsey, C. Wallis, and J. L. Melnick Virus concentration from sewage. Water Res. 7: Jakubowski, W., J. C. Hoff, N. C. Anthony, and W. F. Hill, Jr Epoxy-fiberglass adsorbent for concentrating viruses from large volumes of potable water. Appl. Microbiol. 28: Kessick, M. A., and R. A. Wagner Electrophoretic mobilities of virus adsorbing filter materials. Water Res. 12: Lamanna, C., F. M. Malette, and L. N. Zimmerman Basic bacteriology. Its biological and chemical background, 4th ed. The Williams & Wilkins Co., Baltimore. 11. Levin, M. A., J. R. Fischer, and V. J. Cabelli Quantitative large-volume sampling technique. Appl. Microbiol. 28: Moore, B Detection of paratyphoid carriers in towns by means of sewage examination. Bull. Health Lab. Sew. 7: Presnell, M. W., and W. H. Andrews Use of the membrane filter and a filter aid for concentrating and enumerating indicator bacteria and Salmonella from estuarine waters. Water Res. 10: Sherbert, G. V The biophysical characterisation of the cell surface. Academic Press, Inc., London. 15. Sobsey, M. D., C. Wallis, M. Henderson, and J. L. Melnick Concentration of enteroviruses from large volumes of water. Appl. Microbiol. 26:

Influence of Salts on Virus Adsorption to Microporous Filters

Influence of Salts on Virus Adsorption to Microporous Filters APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 2000, p. 2914 2920 Vol. 66, No. 7 0099-2240/00/$04.00 0 Copyright 2000, American Society for Microbiology. All Rights Reserved. Influence of Salts on Adsorption