UV Dose-Response of Acinetobacter baumannii in Water

Michael R. Templeton1, Marina Antonakaki2, and Michael Rogers3

Department of Civil and Environmental Engineering, Imperial College London,

South Kensington campus, London, United Kingdom SW7 2AZ.

(Same address, telephone, and fax for all authors.)

1 Corresponding Author E-mail: ; Phone: +44 (0)207 594 6099; Fax: +44 (0)207 594 6124.

2 E-mail: .

3 E-mail: .

Key words: drinking water; disinfection; ultraviolet disinfection; waterborne pathogens; Acinetobacter baumannii.

Submitted for publication as a Short Communication in

Environmental Engineering Science

13 February 2008

Revision submitted 16 May 2008

Second revision submitted 8 July 2008


ABSTRACT

UV doses were applied to water samples containing Acinetobacter baumannii using a low-pressure UV collimated beam device. This produced a UV dose-response profile for A. baumannii from which the UV doses that are required for various levels of log inactivation of A. baumannii were interpolated. A UV dose of 1.8 mJ∙cm-2 achieved 2-log (99%) inactivation, 3.3 mJ∙cm-2 achieved 3-log (99.9%) inactivation, and 4.8 mJ∙cm-2 achieved 4-log (99.99%) inactivation. The UV dose-response of A. baumannii is similar to the previously reported dose-response profiles of other waterborne bacteria such as E. coli O157:H7 and protozoan pathogens such as Cryptosporidium parvum. The results of this study suggest that UV disinfection can be an effective control strategy for A. baumannii. Point-of-use UV disinfection (i.e. at the tap) may therefore be a candidate technology to reduce the risk of infection of immuno-compromised individuals from exposure to water containing A. baumannii.

Key words: drinking water; disinfection; ultraviolet disinfection; waterborne pathogens; Acinetobacter baumannii.


INTRODUCTION

There have been numerous reports concerning the persistence of low levels of various types of bacteria in tap water, among which are bacteria of the genus Acinetobacter (LeChevallier et al., 1980; Bifulco et al., 1989). There are at least 30 different species of Acinetobacter however Acinetobacter baumannii is of greatest clinical significance (Stewart, 1999). Acinetobacters are generally of low virulence but are being increasingly associated with hospital-acquired infections (Stewart, 1999). The most common Acinetobacter infections include pneumonia, bacteraemia, wound infections, and urinary tract infections (Bergogne-Berezin et al., 1996). The risk posed by A. baumannii that can persist in tap water may be especially high for immuno-compromised individuals, such as hospital patients and infants (Anaissie et al., 2002; Glasmacher et al., 2003). Certain types of A. baumannii are also multi-drug resistant (Armstrong et al., 1981; Corbella et al., 1998; Urban et al., 2003; Nemec et al., 2004). This pathogen has also received special attention recently in light of outbreaks amongst military personnel at medical facilities in Iraq (CDC, 2004).

Acinetobacter is readily found in the environment including drinking water and surface water, soil, sewage, and various types of foods (Baumann, 1968; Stewart, 1999). Acinetobacter species have been detected in surface water (Baumann, 1968), groundwater (Quevedo-Sarmiento et al., 1986), natural mineral water (Stewart, 1999), wells (Golas, 2000), and distributed drinking water (LeChevallier et al., 1980). Surveys have reported Acinetobacter spp. in 93, 96, 50, 42, and 38% of soil, surface water, estuaries, thermal springs and groundwater samples, respectively (Stewart, 1999). They are environmentally persistent and tolerate both wet and dry conditions (Jawad et al., 1996). They can survive for months on clothing and bedclothes, bed rails, ventilators and other surfaces, including sinks and doorknobs; transmission in hospitals can be especially difficult and important to control (Bergogne-Berezin et al., 1996). Another important means of transmission of Acinetobacter is by water supply; this is because Acinetobacter is often part of the profile of heterotrophic bacteria found on granular activated carbon and sand filters, in pipe biofilms, and point-of-use water treatment devices (Stewart, 1999). The relative importance of water supply as a source of infection from opportunistic pathogens such as A. baumannii in hospitals relative to other routes of exposure (e.g. person-to-person contact) is uncertain, however it has been argued that tap water can potentially be an important route of exposure that needs to be addressed (Anaissie et al., 2002).

The objective of this study was to determine the UV dose-response of A. baumannii in water to assess whether UV disinfection could be an effective control strategy for this bacterium. While the UV dose-response profiles of many other bacterial, viral, and protozoan pathogens have been reported in the literature (Table 1), there was no data for the UV sensitivity of A. baumannii in water at the time of this study.

EXPERIMENTAL PROTOCOLS

Acinetobacter baumannii (NCTC 12156, obtained from the National Collection of Type Cultures, Health Protection Agency, London, UK) was grown and enumerated using Leeds Acinetobacter Medium (LAM) as described elsewhere (Jawad et al., 1994). Where a liquid culture was required, the medium was prepared with the omission of agar and phenol red. Samples for exposure to UV light were prepared by mixing 2 ml of stationary phase liquid culture of A. baumannii into 200 ml of sterile phosphate buffered water (pH 7), in order to generate a working solution with a cell density of 104-105 CFU∙ml-1. This cell density was selected to allow high levels of inactivation (i.e. 4-log) to be quantified by ensuring enough countable survivors following exposure. It was not intended to represent actual cell densities in aquatic environments or in tap water. (A potential extension of this study is to investigate inactivation at lower cell densities, although such a study would require enrichment and sub-culturing of environmental isolates which may also have effects on the UV-sensitivity of the bacteria.) For enumeration of the bacteria following UV exposure, spread plates were prepared and incubated at 36°C for 24 hours.

A UV collimated beam apparatus containing a low-pressure UV lamp (254 nm emission) was used to apply UV doses. The UV intensity (also called ‘fluence rate’) distribution across the exposure surface was measured with a calibrated radiometer (IL1700 with SED240 sensor, International Light, Peabody, MA, USA) and was 0.248 mW∙cm-2 at the centre of the exposure surface. Samples (20 ml withdrawn from the working solution) were contained in 9 cm diameter Petri dishes while stirred by a 1-cm long magnetic stir bar during the UV exposures. The samples were transferred to and from the Petri dishes using 10 ml wide-tip sterile pipettes. The volume of inoculated water was exposed for different periods of time under the UV light to achieve a range of UV doses (ranging from 6 to 29 seconds for 1 to 5 mJ∙cm-2). The UV dose (also called ‘fluence’) calculation followed standard methods described in detail elsewhere (Bolton and Linden, 2003). The UV absorbance at 254 nm of the phosphate buffered solution with bacteria added was 0.21-0.22 cm-1, measured using a standard UV spectrophotometer. UV doses were initially applied at 10 to 25 mJ∙cm-2 however these doses resulted in total inactivation of A. baumannii (i.e. no countable survivors following UV exposure). Subsequently, UV doses were applied at 1, 1.5, 2, and 5 mJ∙cm-2 in order to generate a UV dose-response profile. The dose of 5 mJ∙cm-2 was the highest UV dose at which countable survivors could be measured following UV exposure (i.e. at least 20 colony forming units on the plate without dilution). At least three replicate exposures were conducted at each UV dose.

Bacteriophage MS2, a standard surrogate organism in UV disinfection studies with a well-known UV dose-response profile (Templeton et al., 2005; USEPA, 2006), was used as a benchmark for additional confirmation of the UV doses being applied by the collimated beam. The methods for growth and enumeration of phage MS2 have been described elsewhere (e.g. USEPA, 2001; Templeton et al., 2005). A UV dose of 40 mJ∙cm-2 is expected to result in approximately 2.5-log inactivation of phage MS2 in clear, particle-free water (Templeton et al., 2005). At 40 mJ∙cm-2 the average inactivation of phage MS2 was 2.6-log (n = 3), which validated the UV dose exposure method.

RESULTS

The UV dose-response relationship for A. baumannii over the experimental UV dose range is shown in Figure 1. There was a good linear fit over this dose range (R2 = 0.9875). The UV doses required to achieve various levels of log inactivation of A. baumannii were interpolated from the linear relationship and are summarised in Table 1. Table 1 also compares the UV dose-response of A. baumannii to that of other organisms reported in previous studies. A. baumannii is similar in UV sensitivity to other bacteria such as E. coli O157:H7 and protozoan pathogens such as Cryptosporidium parvum, but it is more sensitive than waterborne viruses.

Despite the good linear fit, variability was observed in the dose-response data at 5 mJ∙cm-2. The variability in inactivation at this dose may be associated with having approached the colony counting limits (i.e. fewer countable survivors on the zero dilution plate). The variability may also have been due to so-called “tailing” of the dose-response relationship. Tailing is a common occurrence in disinfection studies that has been attributed to factors such as sub-populations possessing phenotypic resistance to the disinfectant or partial shielding of a small number of cells by dead cells or cell aggregates (Stewart and Olson, 1996). Also, the non-zero y-intercept suggests that there may be a “shoulder” (i.e. a curvature) to the UV dose-response profile at UV doses below the range tested in this study (i.e. < 1 mJ∙cm-2). Practical difficulties in precisely achieving such low UV doses, requiring very short exposure times under the UV collimated beam (i.e. only a few seconds), meant that UV doses < 1 mJ∙cm-2 were not investigated.

DISCUSSION

UV disinfection guidelines in various jurisdictions recommend a design UV dose in the range of 40 mJ∙cm-2 for municipal water treatment (ONORM, 2001; DVGW, 2003) and this is also a typical design dose that is recommended for point-of-use UV disinfection (NSF-ANSI, 2004). While the UV dose applied in practice typically incorporates other factors of relevance to full-scale, flowing systems and safety factors, the UV dose range applied in practice is well above the dose range tested here; therefore, it is expected that UV disinfection should be an effective means of inactivating A. baumannii based on the dose-response profile determined in this study. Point-of-use UV disinfection should therefore be considered as a potential treatment technology to protect vulnerable individuals from A. baumannii in tap water, such as in hospital settings.

This research could be extended in several ways. For example, Acinetobacter species have exhibited increased resistance to chemical disinfectants when grown under certain conditions that favour cell aggregation (Wolfe et al., 1985). Cell aggregation has also been shown to have the potential to impact the UV dose-response of other organisms (Blatchley et al., 2001; Mamane-Gravetz and Linden, 2005). Aggregation was not monitored in this study.

The impact of other water quality variables (e.g. the presence of corrosion particles originating from pipe surfaces) should also be considered, since the phosphate buffered water in this study was particle-free and was intended to represent relatively high-quality treated drinking water (except with no chlorine residual). This is a common approach that has been used in previous dose-response studies (e.g. Shin et al., 2001; Templeton et al., 2005) and was deemed to be acceptable for the purpose of determining the UV dose-response of A. baumannii without confounding factors. Particle-association of microorganisms has been shown to have the potential to influence UV dose-response (Parker and Darby, 1995; Emerick et al., 1999; Templeton et al., 2006); this remains to be investigated in the context of the dose-response of A. baumannii.

Environmental isolates (e.g. from a hospital tap water) may possess different disinfectant resistance compared to reference strains obtained from culture collections (Wojcicka et al., 2007) and this should be investigated in the context of A. baumannii. However, the difficulty in doing so lies in obtaining water samples with high enough concentrations of environmental isolates to be able to conduct disinfection experiments. Also, this study considered only one strain of A. baumannii, and while previous UV disinfection studies have shown that there is usually little variation in the UV dose-response among strains of the same species type (Sommer et al. 2000), this remains to be confirmed for different strains of A. baumannii.

Lastly, some studies have reported slightly different UV dose-response behaviour of microorganisms when exposed to medium pressure versus low pressure UV lamps (Zimmer and Slawson, 2002) – this remains to be investigated in the context of A. baumannii.

SUMMARY

UV light is an effective disinfectant for controlling Acinetobacter baumannii, which was observed to have similar UV dose-response characteristics as other waterborne bacteria reported in previous studies. A UV dose of 4.8 mJ∙cm-2 was able to achieve 4-log (99.99%) inactivation of A. baumannii. This suggests that point-of-use UV disinfection, applied at typical UV doses used in practice (e.g. 40 mJ∙cm-2), should be considered as a candidate technology for protecting vulnerable individuals (e.g. hospital patients) from tap water that may be a source of A. baumannii.

ACKNOWLEDGMENTS

This research was supported in part by a donation from Trojan Technologies (London, ON, Canada). The authors also acknowledge the assistance of Dr. Geoff Fowler and Ms. Xiyu Phoon, both of Imperial College London, for constructing the UV collimated beam apparatus and assisting with the bacteriophage MS2 analysis, respectively.

REFERENCES

ANAISSIE, E.J., PENZAK, S.R. and DIGNANI, C. (2002). The hospital water supply as a source of nosocomial infections: a plea for action. Arch. Intern. Med. 162, 1483-1492.

ARMSTRONG, J. L., SHIGENO, D.S., CALOMIRIS, J.J. and SEIDLER, R.J. (1981). Antibiotic-resistant bacteria in drinking water. Appl. Environ. Microbiol. 42, 277–283.

BAUMANN, P. (1968). Isolation of Acinetobacter from soil and water. J. Bacteriol. 96, 39-42.

BERGOGNE-BEREZIN, E., JOLY-GUILLOU, M.-L. and TOWNER, K.J. (1996). Acinetobacter: Microbiology, Epidemiology, Infections, Management. CRC Press Inc, Boca Raton, FL.

BIFULCO, J.M., SHIREY, J.J. and BISSONNETTE, G.K. (1989). Detection of Acinetobacter spp. in rural drinking water supplies. Appl. Environ. Microbiol. 55, 2214-2219.

BLATCHLEY, E.R., DOUMANTIER, N., HALABY, T.N., LEVI, Y. and LAINE, J.M. (2001). Bacterial responses to ultraviolet irradiation. Wat. Sci. Technol. 43, 179-186.

BOLTON, J. R. and LINDEN, K.G. (2003). Standardization of methods for fluence (UV dose) determination in bench-scale UV experiments. J. Environ. Eng. 129, 209-215.