Applied and Environmental Microbiology, April 2003, p . 1898-1903, Vol . 69, No . 4

Effect of Particles on the Recovery of Cryptosporidium Oocysts from Source Water Samples of Various Turbidities

Yao Yu Feng, Say Leong Ong,^* Jiang Yong Hu, Lian Fa Song, Xiao Lan Tan, and Wun Jern Ng

Center for Water Research, Department of Civil Engineering, National University of Singapore, Singapore 119260, Singapore

Received 19 August 2002/ Accepted 6 January 2003

Detection of Cryptosporidium oocysts in raw water sources is considered an important component in the control of Cryptosporidium in drinking water supplies . Various methods have been developed to detect C . parvum in both raw source waters and finished drinking waters . For example, the USEPA has established an Information Collection Requirement Rule (ICR) method to detect Cryptosporidium oocysts present in source waters (15) . The ICR method, which replaced the American Society for Testing and Materials P229 method, has been criticized for being difficult to perform, yielding highly variable results, and providing different responses in a variety of water matrices (17) . With the Safe Drinking Water Act Amendments of 1996, the USEPA revised the ICR method and proposed draft method 1622 . This proposed method was validated in January 1999 . Method 1622 includes the following main steps: (i) oocyst capture and concentration from a volume of water by filtration, elution, and centrifugation; (ii) oocyst purification by immunomagnetic separation (IMS); (iii) staining of the IMS product by fluorescent-antibody (FA) staining and 4',6-diamidino-2-phenylindole (DAPI) counterstaining; and (iv) examination and confirmation of oocysts by epifluorescence and differential interference contrast microscopy (16) .

Although USEPA method 1622 is a widely used assay for detection of Cryptosporidium oocysts, this method still has drawbacks, such as low levels of oocyst recovery and extremely variable results (1, 2, 11, 12) . Connell et al . (4) studied oocyst recovery from 430 samples collected from 87 source waters . They noted that the levels of oocyst recovery from 97.5% of spiked water samples ranged from less than 10% to as high as 80% . They further noted that the chances of obtaining levels of oocyst recovery in the range from 20 to 70% were almost equal . Turbidity was among the possible factors which affected oocyst recovery in the IMS procedure . However, there has not been unanimous agreement concerning turbidity as some studies have shown that turbidity has little effect on oocyst recovery in the IMS procedure (11), while other studies have indicated that samples with high turbidity can result in low levels of oocyst recovery (2, 13) . It has even been reported there was an optimum turbidity of 500 nephelometric turbidity units (NTU) which could result in the highest level of oocyst recovery (1) . Despite the many studies that have been conducted on method 1622, the effects of turbidity are still not well understood . By comparing the levels of oocyst recovery from surface water and reagent water with an Envirocheck capsule filter (Pall Gelman Sciences, Ann Arbor, Mich.) as recommended in method 1622, Simmons et al . (12) found that the rates of recovery of oocysts from surface water were significantly lower and also more variable than the rates of recovery from reagent water . They also found that the rates of recovery of oocysts from surface water increased significantly when another kind of filter from their laboratory was employed instead of the Envirocheck capsule filter . The findings of these authors suggested that turbidity might affect the performance of method 1622 by affecting the filtration procedure . In view of uncertainties concerning the impact of turbidity on method 1622, the objectives of this study were to identify the critical steps of method 1622 and to determine whether turbidity does indeed have an effect on oocyst recovery .

Enumeration of oocyst stock suspensions.
The suspensions of C . parvum oocysts were enumerated by placing 10 replicates of 10-µl aliquots on glass well microscope slides and staining the preparations with fluorescein isothiocyanate (FITC)-conjugated anti-Cryptosporidium sp . monoclonal antibody (MAb) (Waterborne) . The slides were then examined by using an epifluorescence microscope as described in USEPA method 1622 .

Water matrices used to assess C . parvum oocyst recovery following FA staining, IMS, centrifugation, and filtration.
Ten-microliter portions of an oocyst suspension were used to spike the solutions for each step employed in method 1622 to investigate which steps were the crucial steps that could affect oocyst recovery . To recover seeded oocysts following IMS, 10 ml of reagent water was used . For the recovery studies with seeded oocysts following filtration, 10-liter portions of different water matrices were used . These matrices included reservoir water, reclaimed water, microfiltration (MF) filtrate, and tap water . The tap water was obtained from the laboratory . The reclaimed water and MF filtrate were collected from a dual-membrane-based water reclamation facility in which secondary effluent from a domestic wastewater treatment plant was further treated by chlorination, MF, reverse osmosis, and UV disinfection . The reservoir water was collected from a local freshwater reservoir, and all reservoir water samples were grab samples . The turbidities of all samples were measured before use with a HACH 2100P turbidimeter (Hach Co., Loveland, Colo.) . To study the effect of reservoir water suspended particles, particles were obtained by filtering 10 liters of reservoir water through an Envirochek capsule filter (Pall Gelman Sciences) . The trapped particles were then eluted with elution buffer and dried in an oven at 100°C . The dried particles were then added into 10-liter tap water samples to produce particle-spiked tap water . Other types of water matrices were also prepared by adding different sizes and masses of Silica Gel 60 (Merck, Darmstadt, Germany) to tap water to generate various levels of turbidity .

Filtration, IMS, and FA staining of seeded oocysts.
Ten-liter water samples were filtered at a flow rate of 2.0 liters/min through Envirochek capsule filters (Pall Gelman Sciences) . Oocysts were eluted from the capsule filters with elution buffer and wrist action agitation, as specified in method 1622 . Eluants were collected in 250-ml conical-bottom centrifuge tubes, and the oocysts were concentrated by centrifugation at 1,500 x g for 15 min (Eppendorf 5810; Eppendorf, Hamburg, Germany) . The resulting pellet volumes were recorded . Supernatant was carefully aspirated from each tube until the volume above the pellet was 5 ml . Reagent water was then added so that the pellet volume was 5% or less in the 10-ml samples subjected to IMS . An anti-Cryptosporidium IMS kit (Dynal Inc., Lake Success, N.Y.) was utilized to separate the oocysts from other interfering particulate matter by the IMS protocol as described in method 1622 . Samples were transferred to well slides (Waterborne) and stained with FITC-conjugated anti-Cryptosporidium sp . MAb (Waterborne) . The slides were then examined by using an epifluorescence microscope as described in method 1622 .

Epifluorescence microscopy.
An Olympus BX51 fluorescence microscope equipped with a blue filter block (excitation wavelength, 490 nm; emission wavelength, 510 nm) was used to detect FITC-conjugated MAb-labeled oocysts at a magnification of x200 . The presence of oocysts was confirmed at a magnification of x400 by using a UV filter block (excitation wavelength, 400 nm; emission wavelength, 420 nm) for visualization of DAPI, and the internal morphology of oocysts was determined by Nomarski differential interference contrast microscopy .

Measurement of particle size distribution.
An LS 230 particle size analyzer with a Small Volume Module Plus and Coulter LS control software (Coulter, Miami, Fla.) was used to measure the particle size distribution in reservoir water . The particle size range analyzed was from 0.04 to 2,000 µm .

Effect of water matrices on oocyst recovery.
In order to study whether the water matrix had an effect on oocyst recovery, 10-liter portions of various water samples were spiked with predetermined numbers of oocysts . Preliminary studies had shown that none of the water matrices used contained Cryptosporidium oocysts before spiking . The results obtained in this series of studies are summarized in Table 1 . The levels of recovery ranged from 8.4 to 16.2%, from 9.0 to 22.9%, and from 13.4 to 22.0% for the reclaimed water, tap water, and MF filtrate, respectively . These values were not significantly different from one another (P > 0.271, as determined by a t test) . However, the levels of recovery from the most turbid reservoir water samples (range, 79.8 to 92.0%; mean, 85.0%) were much higher than those from the reclaimed water, tap water, and MF filtrate . The t test indicated that the levels of recovery from reservoir water were significantly higher than the levels of recovery from the three other water matrices (P < 0.001) . The high levels of recovery from reservoir water samples were obtained with samples that had relatively high turbidity . As discussed above, filtration had an adverse effect on oocyst recovery . This suggested that recovery of oocysts associated with the filtration stage might be influenced by the turbidity of the water matrix .

 
Effect of particles on oocyst recovery in reservoir water.
Experiments were carried out to investigate if suspended particles could be responsible for the high levels of recovery from reservoir water samples . To facilitate this study, similar quantities of oocysts were spiked into four different water matrices . The first matrix was the actual reservoir water, while the second was prepared by filtering reservoir water through the capsule filter recommended in method 1622 . The third matrix was tap water, which served as the control, while the last matrix was tap water spiked with particles obtained from the capsule filter used to filter the reservoir water . Table 2 summarizes the average results . After the particles were removed from the reservoir water, the mean level of oocyst recovery was found to decrease drastically from 82.4 to 16.4% . This observation suggested that the suspended particles present in the reservoir water contributed to the high level of oocyst recovery associated with reservoir water . This was verified by the increase in the level of oocyst recovery from 15.9% in tap water to 70.8% when the extracted particles (obtained from reservoir water) were added to tap water . Possibly because of a loss of particles during preparation, the spiked tap water had a turbidity of 4.5 NTU, compared to the turbidity of 5 NTU for the original reservoir water . The lower turbidity could have contributed to the slightly lower level of recovery from the former sample .

 
Effect of turbidity on oocyst recovery from reservoir water samples.
It was observed that the turbidities of reservoir water samples were different for different collections, ranging from 1.8 to 40 NTU . Rain was found to be the key factor causing high turbidity, and this was especially true for the sample with a turbidity of 40 NTU . The effect of turbidity on oocyst recovery from reservoir water samples was investigated, and the results are summarized in Table 3 . A t test indicated that the levels of recovery from reservoir water samples with turbidities other than 40 NTU were significantly higher than the levels of recovery from tap water (P < 0.005) . However, no significant difference between the level of recovery from reservoir water with a turbidity of 40 NTU and the level of recovery from tap water (P = 0.182) was found . As the turbidity of reservoir water increased from 1.8 to 5 NTU, the level of oocyst recovery increased significantly, from 63.3% (standard deviation, ±8.1) to 85.0% (standard deviation, ±5.2%) (P = 0.008, as determined by a t test), although the difference between the levels of recovery from samples with turbidities of 1.8 and 3.0 NTU was not significant (P = 0.344, as determined by a t test) . When the turbidity increased further from 5 to 40 NTU, the level of oocyst recovery decreased significantly, from 85.0% (standard deviation, ±5.2%) to 24.9% (standard deviation, ±8.4%) (P < 0.001, as determined by a t test) . The highest mean level of recovery was obtained at a turbidity of 5 NTU, and the value was significantly higher than the levels of recovery for all the other turbidities investigated (P < 0.025, as determined by a t test) . This finding suggested that a moderate degree of turbidity enhances oocyst recovery . However, increasing the turbidity beyond the threshold value caused the efficiency of recovery to decline . This finding agreed with previous findings which showed that high turbidity was not conducive for oocyst detection (12) . The results obtained in this study also suggested that a moderate amount of suspended particles (i.e., a moderate degree of turbidity) enhanced oocyst recovery, probably because the oocysts adhered to the suspended particles, making them easy to capture .

 
Experimental verification of the particle effect.
As the suspended particles present in the reservoir water were likely to consist of a mixture of particles of various sizes, it was useful to investigate if particle size and particle size distribution affected the efficiency of recovery . The particle size distribution in reservoir water with a turbidity of 5 NTU is shown in Fig . 1 . We found that most of the suspended particles were in size range from 1 to 400 µm . However, the relative volumetric fractions varied considerably with respect to particle sizes .

 
To study the effect of particle size, silica gel particles in four size ranges from 5 to 400 µm were introduced into different tap water samples . The results obtained are shown in Table 4 . We found that the level of recovery decreased as the size of the added particles increased . A t test indicated that the levels of recovery from samples with particles that were not larger than 63 µm were significantly higher than the levels of recovery from tap water (P < 0.011) . However, no significant differences were detected for the levels of recovery from samples containing silica particles larger than 63 µm (P > 0.158) . The highest mean level of recovery was 51.5% (standard deviation, ±6.5%), when the size of the added silica particles was in the range from 5 to 40 µm . This value was significantly higher than the levels of recovery associated with the other silica particle sizes (P < 0.005, as determined by a t test) . Figure 1 shows that although the volumetric fraction of particles in the size range from 5 to 40 µm was low, the particle count for this fraction accounted for 70 to 80% of all the particles present (after calculation) . Table 4 shows that in order to obtain a turbidity of 5 NTU, the masses needed for the different size ranges of silica were quite different . For example, only 0.71 g of silica particles in the size range from 5 to 40 µm was needed to obtain a turbidity of 5 NTU in 10 liters of tap water, while 10.4 g or more of silica particles was needed when larger particles were used .

 
In order to study the effect of particle concentration on oocyst recovery, different masses of silica particles in the size range from 5 to 40 µm were added to 10-liter tap water samples to obtain different concentrations of particles in the water samples . The results obtained are summarized in Table 5 . The levels of recovery from samples spiked with silica particles were significantly higher than the levels of recovery from tap water samples (P < 0.004, as determined by a t test) . The efficiency of recovery improved from 37.4% (standard deviation, ±3.0%) to 90.6% (standard deviation, ±3.7%) when the amount of silica added was increased from 0.35 to 1.42 g, and this increase was statistically significant (P < 0.013, as determined by a t test) . However, the efficiency of recovery deteriorated significantly when 2.16 g or more of silica particles was added to a 10-liter tap water sample (P < 0.021, as determined by a t test) . This suggested that while a moderate amount of silica particles enhanced oocyst recovery, there was a threshold concentration above which the recovery efficiency was adversely affected . The highest level of recovery was obtained when 1.42 g of silica was added to 10 liters of tap water, and the corresponding water turbidity was 10 NTU .

 
We noted that the filtration step resulted in a significant decrease in the level of oocyst recovery . We found that of the different water matrices investigated, reservoir water gave the highest average level of recovery, 85.0% (standard deviation, ±5.2%) . We suggest that the suspended particles present in the reservoir water enhanced oocyst recovery . A turbidity of 5 NTU for the reservoir water resulted in the highest level of oocyst recovery . As reservoir water samples with different turbidities have distinct particle size distributions, it is reasonable to believe that particle size and concentration could be the key factors affecting oocyst recovery . Using silica to verify this hypothesis, we found that addition of particles in the size range from 5 to 40 µm to obtain a turbidity of 10 NTU resulted in the highest level of oocyst recovery . This size range corresponded to the major particle count fraction of suspended particles identified in reservoir water samples . When different amounts of silica particles in the size range from 5 to 40 µm were added to 10-liter tap water samples, the sample with a turbidity of 10 NTU yielded a higher level of recovery than the water samples with turbidities ranging from 2.5 to 7.5 NTU . Further increases in the amount of silica particles added (i.e., increases in the turbidity to 15 to 20 NTU) led to decreases in the levels of oocyst recovery . This confirmed that particle size distribution and particle concentration affected oocyst recovery .

The filtration stage includes filtration, elution, and centrifugation . It is possible that oocysts could adhere to particles which are larger than the oocysts (diameter, 4 to 6 µm) . Because of this attachment, oocysts could then be more readily retained by the filter and therefore could subsequently be recovered during the elution and centrifugation steps . Pellets resulting from centrifugation were easily obtained when silica was added to tap water, while this was not true when no silica was added . However, when there were too many particles or the particles were too big, it was not easy to elute the particles from the filter . Consequently, some particles were retained on the filter, thereby reducing the recovery efficiency of the elution process . We found in this study that when silica particles in the size range from 5 to 40 µm were added to tap water to increase the turbidity to more than 10 NTU or when particles larger than 40 µm were added to tap water to obtain a turbidity of 5 NTU, various amounts of particles remained on the filter after the elution process . Furthermore, a large amount of particles may also affect the IMS step, as reported by Stanfield et al . (13) and Campbell and Smith (2) . This was also noted from microscopic observations made in this study . We also noticed that most of the oocysts were entrapped on the background debris (from reservoir water or water to which silica was added) present on the slides (Fig . 2 and 3) . In contrast, background debris was not present on the slides when tap water was used . As observed in this study, when silica particles in the size range from 5 to 40 µm were added to tap water to increase the turbidity to a level that did not exceed 15 NTU, the typical image of FITC staining appeared, as shown in Fig . 3, and we found that the oocysts could be embedded in a thin layer of background debris present on a slide . However, the oocysts were not blocked by the background . A similar phenomenon also occurred with DAPI-stained preparations . In contrast, when silica particles in the size range from 5 to 40 µm were added to tap water to increase the turbidity to more than 20 NTU or when particles larger than 40 µm were added to tap water to obtain a turbidity of 5 NTU, the corresponding background particles were present in large thick clumps, oocysts were found to be embedded within the clumps, and weak fluorescence from FITC and DAPI staining could be observed behind the clumps . It is therefore reasonable to believe that some of the oocysts were blocked from view and therefore could not be counted .

 
In this study, high levels of recovery (95.1% ± 3.4%) following the IMS step were obtained for spiked reagent water . Similar experiments were performed by Bukhari et al . (1), as well as by Campbell and Smith (2) . These authors reported levels of recovery of 77.4% ± 5.6% for deionized water (data after calculation) and 89.2% ± 30.4% for water with no turbidity (data after calculation), respectively . The values obtained in this study were, however, higher and more stable . We also noted in this study that there was an optimal turbidity in terms of oocyst recovery when silica particles in the size range from 5 to 40 µm were added to tap water samples . Deviating from this optimal turbidity led to a decrease in the efficiency of recovery in the filtration step and possibly also in the IMS step . The turbidities of the IMS concentrate increased from 2,520 to 5,400 and 7,200 NTU when the turbidities of the water were increased from 2.5 to 10 and 20 NTU, respectively . The highest level of oocyst recovery was obtained when the water turbidity was 10 NTU . It was possible that when the turbidity of the water was more than 10 NTU, the recovery following both the filtration stage and IMS was affected . Stanfield et al . (13) reported that the levels of recovery following IMS for treated water were 96.1% ± 9.5%, while for raw water they were only 48.9% ± 14.2% . Campbell and Smith (2) noted that an increase in turbidity from 0 to 60 NTU caused the levels of recovery following IMS to decrease from 89.2% ± 30.4% to 62.4% ± 32.4% . A further increase in turbidity to 611 NTU caused the levels of recovery following IMS to decrease drastically to 29.0% ± 34.1% . However, it should be pointed out that the negative effect of higher turbidity in nature on IMS is still speculative as contradictory conclusions have been reported . For example, Rochelle et al . (11) reported that the levels of recovery following IMS were 79.1% ± 12.4% when the turbidities of water concentrate samples were between 210 and 11,480 NTU .

In this study, silica particles were used to adjust the water turbidity because they could readily represent the turbidity caused by inorganic particles . In addition, they could also simulate the effects of particles present in reservoir water conveniently since a full range of silica particles of the sizes typically present in reservoir water was easily obtainable .

In conclusion, results obtained in this study indicated that filtration was the most important step in terms of oocyst recovery and that suspended particles present in reservoir water could help improve oocyst recovery . In addition, we found that the recovery was affected by particle size and particle concentration . We propose that adding particles to a water matrix is a possible approach to improve oocyst recovery when method 1622 is used for oocyst detection .

1. Bukhari, Z., R . M . McCuin, C . R . Fricker, and J . L . Clancy. 1998 . Immunomagnetic separation of Cryptosporidium parvum from source water samples of various turbidities . Appl . Environ . Microbiol . 64:4495-4499.
2. Campbell, A., and H . Smith. 1997 . Immunomagnetic separation of Cryptosporidium oocysts from water samples: round robin comparison of techniques . Water Sci . Technol . 35(11-12):397-401.
3. Clark, D . P. 1999 . New insights into human cryptosporidiosis . Clin . Microbiol . Rev . 12:554-563.
4. Connell, K., J . Scheller, K . Miller, and C . C . Rodgers. 2000 . Performance of methods 1622 and 1623 in the ICR supplemental surveys . In Proceedings of the AWWA Water Quality Technology Conference, Salt Lake City, Utah . American Water Works Association, Denver, Colo . [On CD-ROM.]
5. Korich, D . G., J . R . Mead, M . S . Madore, N . A . Sinclair, and C . R . Sterling. 1990 . Effects of ozone, chlorine dioxide, chlorine, and monochloramine on Cryptosporidium parvum oocyst viability . Appl . Environ . Microbiol . 56:1423-1428.
6. Kramer, M . H., B . L . Herwaldt, G . F . Craun, R . L . Calderon, and D . D . Juranek. 1996 . Surveillance for waterborne-disease outbreaks—United States, 1993-1994 . Morb . Mortal . Wkly . Rep . CDC Surveill . Summ . 45:1-33.
7. LeChevallier, M . W., W . D . Norton, and R . G . Lee. 1991 . Giardia and Cryptosporidium spp . in filtered drinking water supplies . Appl . Environ . Microbiol . 57:2617-2621.
8. MacKenzie, W . R., N . J . Hoxie, M . E . Proctor, M . S . Gradus, K . A . Blair, D . E . Peterson, J . J . Kazmierczak, D . G . Addiss, K . R . Fox, J . B . Rose, and J . P . Davis. 1994 . A massive outbreak in Milwaukee of Cryptosporidium infection transmitted through the public water supply . N . Engl . J . Med . 331:161-167.
9. Marshall, M . M., D . Naumovitz, Y . Ortega, and C . R . Sterling. 1997 . Waterborne protozoan pathogens . Clin . Microbiol . Rev . 10:67-85.
10. Moss, D . M., S . N . Bennett, M . J . Arrowood, S . P . Wahlquist, and P . J . Lammie. 1998 . Enzyme-linked immunoelectrotransfer blot analysis of a cryptosporidiosis outbreak on a United States Coast Guard cutter . Am . J . Trop . Med . Hyg . 58:110-118.
11. Rochelle, P . A., D . L . Ricardo, A . Johnson, M . H . Stewart, and R . L . Wolfe. 1999 . Evaluation of immunomagnetic separation for recovery of infectious Cryptosporidium parvum oocysts from environmental samples . Appl . Environ . Microbiol . 65:841-845.
12. Simmons, O . D., M . D . Sobsey, C . D . Heaney, F . W . Schaefer, and D . S . Francy. 2001 . Concentration and detection of Cryptosporidium oocysts in surface water samples by method 1622 using ultrafiltration and capsule filtration . Appl . Environ . Microbiol . 67:1123-1127.
13. Stanfield, G., E . Carrington, F . Albinet, B . Compagnon, N . Dumoutier, B . Hambsch, A . Lorthioy, G . Medema, H . Pezoldt, M . R . Roubin, A . Lohman, and T . Whitmore. 2000 . An optimized and standardized test to determine the presence of the protozoa Cryptosporidium and Giardia in water . Water Sci . Technol . 41(7):103-110.
14. Steiner, T . S., N . M . Thielman, and R . L . Guerrant. 1997 . Protozoal agents: what are the dangers for the public water supply? Annu . Rev . Med . 48:329-340.
15. U.S . Environmental Protection Agency. 1996 . EPA information collection rule microbial laboratory manual . Publication EPA/600/R-95/178 . Office of Research and Development, Washington, D.C.
16. U.S . Environmental Protection Agency. 1999 . Method 1622: Cryptosporidium in water by filtration, immunomagnetic separation, and fluorescent antibody . Publication EPA-821-R-99-001 . Office of Water, Washington, D.C.
17. U.S . Environmental Protection Agency. 2001 . Method 1623: Cryptosporidium and Giardia in water by filtration, immunomagnetic separation, and fluorescent antibody . Publication EPA-821-R-01-025 . Office of Water, Washington, D.C.
18. Venczel, L . V., M . Arrowood, M . Hurd, and M . D . Sobsey. 1997 . Inactivation of Cryptosporidium parvum oocysts and Clostridium perfringens spores by a mixed-oxidant disinfectant and by free chlorine . Appl . Environ . Microbiol . 63:1598-1601.

Free Online Full-text Article

What Is MIC?, What Is Molecular Biology?, What Is Anthrax?, What Is Genetic Engineering?, What Is Biofilter?, e, Microbiology, n, Bacteria, s, Bacteriology, i, Microbe, i, Microorganism, e, Escherichia coli, n, Suspension cells, n, Escherichia coli, e, Yeasts, n, Biodegradation, s, Escherichia coli, i, Antibiotics, a, Microorganism, s, Bacteriological, s, Escherichia coli, s, Antimicrobial, r, Streptococci, r, Escherichia coli, i, Salmonella typhimurium, a, Prokaryotes, i, Bacteria, a, S. cerevisiae, e, S. cerevisiae, i, S. cerevisiae, r, Staphylococcus, o, Bacteriological




 

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005