Meegan Batch process solar disinfection is an efficient means of disinfecting drinking water contaminated with Shigella dysenteriae

ims:  The mortality and morbidity rate caused by Shigella dysenteriae type I infection is increasing in the developing world each year. In this paper, the possibility of using batch process solar disinfection (SODIS) as an effective means of disinfecting drinking water contaminated with Sh. dysenteriae type I is investigated.

Methods:  Phosphate-buffered saline contaminated with Sh. dysenteriae type I was exposed to simulated solar conditions and the inactivation kinetics of this organism was compared with that of Sh. flexneri, Vibrio cholerae and Salmonella typhimurium.

Significance:  Recovery of injured Sh. dysenteriae type I may be improved by plating on medium supplemented with catalase or pyruvate. Sh. dysenteriae type I is very sensitive to batch process SODIS and is easily inactivated even during overcast conditions. Batch process SODIS is an appropriate intervention for use in developing countries during Sh. dysenteriae type I epidemics.
Introduction

After virtually disappearing at the beginning of the 20th century, epidemic Shigella dysenteriae type I reappeared in 1968 in Central America and later in Asia and Africa (Mata et al. 1970; World Health Organisation 1988; Tuttle et al. 1995). Today, bacillary dysentery is endemic throughout the world with 150 million cases and almost 600 000 deaths occurring annually (Sansonetti 1999). About 95% of these cases occur in developing countries where water quality and sanitation is less than adequate. The low infective dose (thought to be as little as 10 cells; Sansonetti 1999) together with the emergence of antimicrobial resistant strains has made it increasingly difficult to control both the spread and treatment of this organism. In 1987, Sh. dysenteriae type I strains resistant to all commonly available anti-microbial agents were isolated in Bangladesh (Munshi et al. 1987). Strains resistant to trimethorprim–sulphamethoxazole and ampicillin were isolated in both Africa and Asia (Frost et al. 1981; Central Statistics Office 1991).

Small improvements in water supply and sanitation facilities in poor communities have a lower impact on diarrhoea caused by pathogens of low infective dose such as Sh. dysenteriae type I compared with pathogens of high infective dose, e.g. Vibrio cholerae (Esrey et al. 1985). Zeng-sui et al. (1989) reported that provision of good quality drinking water supplies reduces the transmission of viral hepatitis A, cholera and acute watery diarrhoea but does not influence the incidence of bacillary dysentery. It is clearly desirable to control all potential routes of transmission of Sh. dysenteriae type I. Batch process solar disinfection (SODIS), which takes advantage of the most abundant source of energy in many of these regions, natural sunlight, may provide additional possibilities for control of bacillary dysentery.

The SODIS technique consists of filling transparent bottles with drinking water and exposing them to full sunlight for up to 8 h with the subsequent inactivation of microbial and viral pathogens (Acra et al. 1989; Sommer et al. 1997; Kehoe et al. 2001). The bactericidal effect of sunlight is due to optical and thermal processes and a strong synergistic effect occurs at temperatures exceeding 45°C (McGuigan et al. 1998). In addition to direct u.v. killing, sunlight is absorbed by endogenous (e.g. cytochromes) and exogenous (e.g. humic substances) photosensitizers that then react with oxygen producing highly reactive oxygen molecules such as hydrogen peroxide (H2O2), singlet oxygen and superoxides which exert a bactericidal effect (Whitelam and Codd 1986; Farr and Kogoma 1991). As a result, oxygen levels within the container should be at a maximum (Reed et al. 2000; Kehoe et al. 2001). Most bacterial strains produce catalase in response to hydrogen peroxide. However, Sh. dysenteriae type I does not produce a catalase that is detected by standard methods and thus may be more sensitive to batch process SODIS. Bogosian et al. (2000) noted that H2O2-sensitive cells of V. vulnificus produced during starvation were recovered by growth on medium supplemented with catalase or pyruvate but not by growth on standard medium. We supplemented medium with catalase or pyruvate to achieve maximum plating efficiency.

We show that sublethally solar injured Sh. dysenteriae type I may be recovered on medium supplemented with either catalase or pyruvate but not on standard medium. Sh. dysenteriae type I is extremely sensitive to SODIS with inactivation occurring even during overcast conditions.
Materials and methods

The following bacterial strains were used in this experiment: Sh. dysenteriae type I ATCC 13313, V. cholerae 8021 serovar 01, classical biotype, Ogawa serotype purchased from the NCTC Collection, Colindale, London, UK, Salmonella typhimurium C5Nxr as described by Smith et al. (2000). Sh. flexneri (M90 T; Franzon et al. 1990) was kindly donated by P.J. Sansonetti, Institut Pasteur, Paris, France. All experiments involving Sh. dysenteriae type I were performed in a class 3 containment laboratory in correspondence with EU regulations.

All bacterial strains were inoculated (single colony) in 100 ml of sterile nutrient broth (Oxoid CM67) and incubated at 37°C for 18 h to obtain a stationary phase culture. Cells were harvested by centrifugation at 855 g for 10 min and washed three times with HPLC analytical reagent sterile water to completely remove the nutrients. Finally, the pellet was resuspended in sterile phosphate-buffered saline (PBS), pH 7·3 (Oxoid; BR14) to a final concentration of 106 CFU ml−1. These organisms were found to be more unstable when maintained in water than PBS (Kehoe 2001). By resuspending these cells in PBS we aimed to expose them to solar irradiation in their most stable environment. The solar simulation apparatus described by McGuigan et al. (1998) was used. The irradiating light source was a 150 W Xenon arc lamp (model 66057/68806 Oriel Ltd., Stratford, CT, USA) fitted with a rear reflector and u.v. collecting optics. The light from the lamp was passed through an Air Mass 1.0 heat-absorbing solar filter (model KG2, Melles-Griot, Cambridge, UK), which closely approximates the incident solar irradiation expected at sea level on the equator. The complete continuous output spectrum of this system is given in McGuigan et al. (1998). A low optical irradiance of 42 mW cm−2, corresponding to an overcast day in Kenya (Joyce et al. 1996) was simulated and the water temperature was maintained at 42°C for Sh. dysenteriae type I and Sh. flexneri while Salm. typhimurium and V. cholerae were exposed to higher levels of irradiation in order to obtain inactivation within 8 h (42 and 45°C respectively and 87 mW cm−2). Sh. dysenteriae type I and Sh. flexneri inactivation occurred at such a high rate under these high optical irradiances that it was necessary to reduce the optical irradiance to 42 mW cm−2 for calculation of inactivation kinetics.

In the field trials of the SODIS technique described by Conroy et al. (2001) test subjects placed their SODIS bottles on the roof of their dwelling or kept them inside their dwelling in a darkened area, at room temperature. Bottles were exposed on the roof of Maasai huts and reached water temperatures of between 40 and 55°C or were kept indoors in the shade where water temperatures were similar to room temperature. Consequently, in our experiments a control solution was left in the dark at room temperature throughout the procedure. Test samples were maintained at the intermediate water temperature of 42°C to ensure that thermal inactivation processes did not predominate. Volumes of 100 μl were taken from each bottle of the control and irradiated groups at the beginning of each experiment and at each sampling interval. These volumes were diluted in a series of 10-fold dilutions and plated in triplicate on either standard plate count agar (SPCA; Oxoid CM 463) or agar supplemented with either catalase or pyruvate (see below) and the CFU/ml were calculated by the method of Hoben and Somasegeram (1982) following incubation at 37°C for 18 h. First-order solar decay constants (kJ−1) were calculated from the slope of the regression line Ln(Nt/N0) vs cumulative dose in kJ, where N0 is the number of viable bacteria in CFU/ml at time zero and Nt is the number of viable bacteria in CFU/ml at time, t. Plotting values as a function of cumulative dose as opposed to time allowed comparison between all four organisms studied taking into consideration the differing optical intensity and temperature. This measurement also takes into account, water volume and dimensions of solar reactors. Each experiment was repeated at least three times. Exact statistical tests were implemented in StatXact 5 (STATCON, Witzenhausen, Germany). First-order decay constants were compared using analysis of variance with general scores. Catalase (EC 1.11.1.6, from bovine liver; Sigma, C-9322) solutions were prepared by dilution in ice-cold phosphate buffer (10 mm, pH 7). Solutions were immediately filter sterilized with 0·2 μm membrane filters (Sarstedt, Nümbrecht, Germany, 83.1826.001) and 0·5 ml aliquots aseptically transferred to the surface of a standard agar plate. Quantities of 406, 812, 1700, 2445, 3260 units catalase were applied to plates and that concentration which gave optimum plating efficiency was determined. Catalase plates were prepared approx. 1 h prior to sampling. A solution of catalase, which had been boiled for 10 mins, acted as a control.

Pyruvate plates were prepared by addition of sodium pyruvate (Sigma, p-8574) directly to the medium before autoclaving. The following concentrations were examined for plating efficiency and the optimum determined; 0·03, 0·05, 0·07, 0·1 and 0·25%. Glacial acetic acid (0·03%), a by-product of H2O2 degradation by pyruvic acid acted as a control (Zelitch 1972; Elstner and Heupel 1976).
Results

The solar inactivation behaviour of the four bacteria differed considerably (Fig. 1; Table 1). Sh. dysenteriae type I is significantly more sensitive to SODIS than either Sh. flexneri, V. cholerae or Salm. typhimurium (P = 0·015). No change in culturability was noted in the dark control microcosms over the course of exposure. A 6-log reduction in CFUs of Sh. dysenteriae type I was observed after just 1·5-h exposure to simulated overcast conditions at equatorial latitudes (Fig. 2). Six-hour exposure is required to inactivate a similar concentration of Sh. flexneri. The order of sensitivity to batch process SODIS is: Sh. dysenteriae type I > Sh. flexneri > Salm. typhimurium > V. cholerae. An optical dose of approx. 6 kJ is required to inactivate 106Sh. dysenteriae type I/ml while approx. 24 and >60 kJ is required to inactivate 106Salm. typhimurium/ml and 106V. cholerae/ml respectively. To put these figures in perspective, an optical dose of 60 kJ would be achieved in approx. 100 min under a standard equatorial solar irradiance of 100 mW cm−2.

Figure 1. Solar inactivation of Sh. dysenteriae type I, Shigella flexneri, Salmonella typhimurium and Vibrio cholerae plated on standard plate count agar (SPCA) expressed in terms of cumulative u.v. dose received (300–400 nm)
image
Table 1. Representative decay constants (kJ−1), in terms of cumulative u.v. dose received (300–400 nm), for solar disinfected Sh. dysenteriae type I, Sh. flexneri, V. cholerae 01 and Salm. typhimurium plated on standard plate count agar (SPCA), medium supplemented with pyruvate or medium supplemented with catalase. R2 values in parentheses Decay constants (kJ−1)
SPCA Pyruvate Catalase
Sh. dysenteriae type I 3·055 (0·942) 1·61 (0·900) 1·191 (0·987)
Sh. flexneri 0·462 (0·895) 0·435 (0·913) 0·314 (0·997)
Salm. typhimurium 0·168 (0·952) 0·171 (0·969) 0·171 (0·986)
V. cholerae 0·076 (0·917) 0·074 (0·910) –

Figure 2. Solar inactivation of Sh. dysenteriae type I, exposed to a solar irradiance of 42 mW cm−2 and a water temperature of 42°C plated on standard plate count agar (SPCA) (•) or medium supplemented with catalase (406 units/plate) (○) or pyruvate (0·05%) (bsl00072)
image

The optimum concentration of catalase and pyruvate is 406 units per plate and 0·05% respectively (data not shown). The inactivation of Sh. dysenteriae type I on SPCA and plates supplemented with 0·05% pyruvate and 406 units catalase are presented in Fig. 2 and Table 1. These results show that when grown on standard plates, 6 log units of Sh. dysenteriae type I appear to be completely inactivated after 1·5-h exposure. However, when this sample was plated on medium supplemented with pyruvate or catalase, almost 104 CFU ml−1 were culturable.

Comparisons of decay constants for Sh. flexneri, Salm. typhimurium and V. cholerae when plated on standard agar and supplemented plates are presented in Table 1 and supplementation of medium with either catalase or pyruvate had little effect on the culturability of these organisms.
Discussion

Shigella dysenteriae type I is sensitive to batch process SODIS. When plated on SPCA 106Sh. dysenteriae type I/ml are inactivated after 1·5-h exposure to simulated equatorial overcast conditions. As shown in Table 1, Salm. typhimurium and V. cholerae have significantly lower decay constants. MacKenzie et al. 1992 reported that solar treatment of drinking water to prevent and control the spread of cholera is effective only under selected conditions, possibly related to altitude and intensity of ultraviolet radiation. However, children under 6 years of age drinking solar disinfected water were protected from V. cholerae infection during an outbreak in rural Kenya (Odds Ratio, 0·12; 95% CI, 0·02–0·65) (Conroy et al. 2001). This suggests that drinking solar disinfected water during a Sh. dysenteriae type I outbreak would protect against infection transmitted by that route.

When grown on supplemented medium it takes almost three times longer for Sh. dysenteriae type I to become nonculturable (see Fig. 2 and Table 1) but inactivation is still occurring at a much higher rate when compared with Salm. typhimurium and V. cholerae although Sh. dysenteriae type I is only exposed to overcast conditions. Pyruvate neutralizes peroxides by a direct nonenzymatic decarboxylation reaction (Mallet et al. 2002). However, it is also thought to act as an important metabolic fuel. Therefore, the improved plating efficiency observed when irradiated Sh. dysenteriae type I is plated on supplemented medium may be due to either or a combination of these factors. However, similar increases in plating efficiencies were noted when catalase was added to the medium. Catalase enzymatically decomposes hydrogen peroxide and is not thought to act as an energy reserve. Supplementation of the medium with boiled catalase had no effect on the plating efficiency of irradiated Sh. dysenteriae type I also suggesting that catalase does not act as an energy reserve. Since catalase and pyruvate have a similar effect on the plating efficiency of Sh. dysenteriae type I and catalase appears to exert its effect by enzymatic decomposition of peroxide then pyruvate is likely to act through neutralization of peroxides rather than acting as an energy reserve. We recommend the supplementation of recovery medium with pyruvate as it may be added to the agar prior to autoclaving and is thus evenly distributed throughout the medium. Catalase, on the other hand, is very unstable at room temperature and it is therefore difficult to predict the shelf life of the plates.

Although supplementation of medium seems to have the greatest impact on the plating efficiency of Sh. dysenteriae type I, such cells are also more susceptible to SODIS and therefore will be inactivated at a faster rate than other species. In addition, we have previously shown that viable bacterial cells (Salm. typhimurium) which were exposed to solar conditions but still culturable on standard plates are less infective than nonexposed viable cells when administered via the intraperitoneal route (Smith et al. 2000). Subsequent studies showed that such culturable but irradiated bacteria were also less infective when administered via the oral route (Kehoe 2001).

Shigella dysenteriae type I is inactivated by batch process SODIS even during equatorial overcast conditions. Batch process SODIS is therefore an appropriate intervention for developing countries during Sh. dysenteriae type I endemics even where adequate sanitation is provided as improvements in water quality and sanitation have little impact on the epidemiology of this organism because of the low infective dose. Studies testing the efficacy of solar/u.v. disinfection should incorporate pyruvate into bacteriological medium.
Acknowledgements

Sincere thanks to P.J. Sansonetti, Institut Pasteur, Paris, France for providing the Sh. flexneri strain used in this project. We thank Ronán Conroy for assistance with the statistical analysis. This project was funded by Royal College of Surgeons in Ireland Research Committee and Enterprise Ireland/British Research Council Research Travel Scheme.
References

Acra, A., Jurdi, M., Mu'allem, H., Karahagopian, Y. and Raffoul, A. (1989) Water Disinfection by Solar Radiation: Assessment and Application. Ottawa, Ont: International Development Research Centre.
Bogosian, G., Aardema, N.D., Bourneuf, E.V., Morris, P.J. and O'Neil, J.P. (2000) Recovery of hydrogen peroxide sensitive culturable cells of Vibrio vulnificus gives the appearance of resuscitation from a viable but nonculturable state. Journal of Bacteriology 182, 5070–5075.
CrossRef,
PubMed,
Web of Science®
Central Statistics Office (1991) Women and men in Zambia. Facts and Figures 11. Zambia: Central Statistics Office.
Conroy, R., Elmore-Meegan, M., Joyce, T., McGuigan, K. and Barnes, J. (2001) Solar disinfection of drinking water protects against cholera in children aged under 6. Archives of Disease in Childhood 85, 293–295.
CrossRef,
PubMed,
Web of Science®
Elstner, E. and Heupel, A. (1976) Formation of hydrogen peroxide by isolated cell walls from horseradish (Armoracia lapathifolia Gilib.). Planta 130, 175–180.
CrossRef,
ChemPort,
Web of Science®
Esrey, S.A., Feachem, R.G. and Hughes, J.M. (1985) Interventions for the control of diarrhoeal diseases among young children: improving water supplies and excreta disposal facilities. World Health Organisation 63, 757–772.
PubMed,
ChemPort,
Web of Science®
Farr, S. and Kogoma, T. (1991) Oxidative stress responses in Escherichia coli and Salmonella typhimurium. Microbiological Reviews 55, 561–585.
PubMed
Franzon, V., Arondel, J. and Sansonetti, P. (1990) Contribution of superoxide dismutase and catalase activities to Shigella flexneri pathogenesis. Infection and Immunity 58, 529–535.
PubMed,
Web of Science®
Frost, J., Rowe, B., Vandepitte, J. and Threlfall, E. (1981) Plasmid characterisation in the investigation of an epidemic caused by multiply resistant Shigella dysenteriae type I in Central Africa. Lancet 2, 1074–1076.
CrossRef,
PubMed,
ChemPort,
Web of Science®
Hoben, H.J. and Somasegeram, O. (1982) Comparisons of pour, spread and drop plate methods for enumeration of Rhizobium spp. in inoculants made from presterilised peat. Applied and Environmental Microbiology 62, 399–402.
Joyce, T., McGuigan, K., Elmore-Meegan, M. and Conroy, R. (1996) Inactivation of fecal bacteria in drinking water by solar heating. Applied and Environmental Microbiology 62, 399–402.
PubMed,
Web of Science®
Kehoe, S.C. (2001) Batch process solar disinfection of drinking water: process and pathogenicity. PhD thesis , Royal College of Surgeons in Ireland, Dublin, Ireland. November 2001.
Kehoe, S.C., Joyce, T.M., Ibrahim, P., Gillespie, J.B., Shahar, R.A. and McGuigan, K.G. (2001) Effect of agitation, turbidity, aluminium foil reflectors and container volume on the inactivation efficiency of batch process solar disinfectors. Water Research 35, 1061–1065.
CrossRef,
PubMed,
Web of Science®
McGuigan, K.G., Joyce, T.M., Conroy, R.M., Gillespie, J.B. and Elmore-Meegan, M. (1998) Solar disinfection of drinking water contained in transparent plastic bottles: characterizing the bacterial inactivation process. Journal of Applied Microbiology 84, 1138–1148.
Direct Link:
Abstract
PDF(356K)
MacKenzie, T.D., Ellison, R.T. and Mostow S.R. (1992) Sunlight and cholera. Lancet 340, 367.
CrossRef,
PubMed,
ChemPort,
Web of Science®
Mallet, R.T, Squires, J.E., Bhatia, S. and Sun, J. (2002) Pyruvate restores contractile function and antioxidant defenses of hydrogen peroxide-challenged myocardium. Journal of Molecular and Cellular Cardiology 34, 1173–1184.
CrossRef,
PubMed,
ChemPort,
Web of Science®
Mata, I., Gangarosa, E., Caceres, A., Perera, D. and Mejicanos, M. (1970) Epidemic Shiga bacillus dysentery in Central America. 1. Etiologic investigations in Guatemala, 1969. Journal of Infectious Disease 122, 170–180.
PubMed,
Web of Science®
Munshi, M., Sack, D., Haider, K., Ahed, Z. and Rahaman, M. (1987) Plasmid mediumted resistance to nalidixic acid in Shigella dysenteriae. Lancet ii, 419–421.
CrossRef
Reed, R., Mani, S. and Meyer, V. (2000) Solar photo-oxidative disinfection of drinking water: preliminary field observations. Letters in Applied Microbiology 30, 432–436.
Direct Link:
Abstract
Full Article (HTML)
PDF(210K)
References
Sansonetti, P. (1999) Shigella plays dangerous games. ASM News 65, 611–617.
Web of Science®
Smith, R., Kehoe, S., McGuigan, K. and Barer, M. (2000) Effects of simulated solar disinfection on infectivity of Salmonella typhimurium. Letters in Applied Microbiology 31, 284–288.
Direct Link:
Abstract
Full Article (HTML)
PDF(100K)
References
Sommer, B., Marion, A., Solarte, M., Dierolf, C., Valiente, C., Mora, D., Rechsteiner, R., Setters, P. et al. (1997) SODIS – an emerging water treatment process. Journal of Water Supply: Research and Technology – Aqua 46, 127–137.
Web of Science®
Tuttle, J., Ries, A., Chimab, R., Perera, C., Bean, N. and Griffin, P. (1995) Antimicrobial-resistant epidemic Shigella dysenteriae type I in Zambia: modes of transmission. Journal of Infectious Disease 171, 371–375.
PubMed,
Web of Science®
Whitelam, G. and Codd, G. (1986) Damage to microorganisms by light. Special Publications of the Society for General Microbiology 17, 129–169.
ChemPort
World Health Organisation (1988) Guidelines for the Control of Epidemics due to Shigella dysenteriae Type I, 12. Geneva: WHO.
Zelitch, I. (1972) The photoxidation of hyloxylate by envelope-free spinach chloroplasts and its relation to photorespiration. Archives of Biochemistry and Biophysics 150, 698–707.
CrossRef,
PubMed,
ChemPort,
Web of Science®
Zeng-sui, W., Shepard, D., Yun-cheng, Z., Cash, R., Ren-jie, Z., Zhen-xing, Z. and Fu-min, S. (1989) Reduction of enteric infectious disease in rural China by providing deep-well tap water. Bulletin of the World Health Organization 67, 171–180.
PubMed