Previous Article | Next Article ![]()
Applied and Environmental Microbiology, June 2004, p. 3346-3351, Vol. 70, No. 6
0099-2240/04/$08.00+0 DOI: 10.1128/AEM.70.6.3346-3351.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Department of Virology, Faculty of Health Sciences, Ben Gurion University of the Negev,1 Public Health Regional Laboratory,2 Southern Regional Health Office, Israel Ministry of Health, Beer Sheva, Israel3
Received 6 November 2003/ Accepted 29 February 2004
|
|
|---|
|
|
|---|
The seroprevalence of S. negevensis has been studied in groups of people of different ages, such as the entire population of a Negev kibbutz (n = 290) and healthy Bedouins of various ages (n = 198). The distribution of seropositivity by age indicated an early exposure to the infectious agent (5) and raised the question of the source of infection and mode of transmission. In the past, early age of acquisition of infection with Helicobacter pylori was associated with its presence in drinking water (18).
Since some newly described members of the Chlamydiales were in fact discovered as endosymbionts (endocytobionts) of free-living amoebae, the ability of S. negevensis to replicate in amoebae in the laboratory was also examined. It was found that not only were the organisms able to replicate successfully in Acanthamoeba polyphaga but they were also able to survive over long periods of time in amoebic cysts (12). In addition, other chlamydia-like microorganisms, such as members of the family Parachlamydiaceae, were shown to be able to grow successfully as endocytobionts of free-living amoebae (9, 11); some other Chlamydiales members, such as Chlamydophila (Chlamydia) pneumoniae, may infect these organisms but grow only slowly (1). Since free-living amoebae such as acanthamoebae may be found in many water sources, as well as in desert sand (22), it may be that the natural source of transmission of S. negevensis to humans is via such amoebae. Similar transmission of legionellae has been described previously (4), and indeed, amoebae are thought to play an important role in the natural history of infections with legionellae and possibly other intracellular microorganisms (10).
In this study, we examined the question of whether S. negevensis can be found in drinking water or in wastewater in the Negev. The development of a simple assay system for detection of the organism in such water samples is described, and evidence for the presence of S. negevensis in various water sources is presented.
|
|
|---|
Samples of tap water were taken from several neighborhoods in Beer Sheva, which obtains its water supply from both the national water carrier and from deep wells, that is, underground water reserves (aquifers) located in the area. The supply alternates between these two sources but not in a set pattern. Drinking water samples were also obtained from three different satellite communities near Beer Sheva, which receive their water from local deep wells. Drinking water brought to the area by the National Water Carrier normally has a turbidity measurement of <1 while the turbidity measurement of deep well water is usually <0.3. Drinking water is chlorinated to 0.1 to 0.5 ppm (or milligrams/liter) of free chlorine. Drinking water in the city is tested according to a municipal plan. In 2003, 684 tests were carried out. In one, 5 coliforms were detected, but upon retesting the source, none were found. All other samples were negative for coliforms.
Chlorination of water and determination of the survival of S. negevensis in chlorinated water.
Chlorine levels in water were determined by a pocket calorimeter (HACH Europe, S. A./N.V., Floriffoux, Belgium) used for routine field testing of chlorine levels. Sterile deionized water (containing 0.07 mg of free chlorine/liter) was exposed for 20 min to chlorination (0.5 mg of free chlorine/liter), generated with free chlorine reagent powder (cat. no. 21055-69; HACH) which was later neutralized with sodium thiosulfate (1 M). S. negevensis particles or S. negevensis in persistently infected amoebae precipitated in a microcentrifuge were suspended in chlorinated water immediately after its chlorination treatment and incubated at room temperature (RT). Control tubes received the same water without chlorine treatment. At the beginning of the incubation, the chlorine levels ranged from 0.48 to 0.52 mg of free chlorine/liter in the different experiments, and at the end of the 20-min incubation, they were 0.22 mg/liter. After neutralization of the chlorine, the contents of the experimental and control tubes were diluted in infection medium for titration on Vero cells, which was carried out by the plate immunoperoxidase assay as previously described (15). In parallel, samples were tested by membrane enzyme immunoassay (MEIA) (see below) to determine the effect of chlorination on antigen detection.
MEIA.
Cellulose nitrate filter membranes (0.45-µm pore size [Schleicher & Schuell] or 0.8-µm pore size [Pall Supor 800]) were used for filtration of control or tested water samples. Samples were mixed, and any coarse particles, if present, were removed by gravity drainage through several layers of viscose polypropylene material (this is a light gauze-like [but nonwoven] fabric composed of viscose and polypropylene fibers and used in clinics as disposable bed covers). Then a sample volume of 100 to 500 µl of reclaimed wastewater or 500 ml of drinking water was filtered (using a 96-place manifold system [S&S] with the 0.45-µm-pore-size membrane sheet or the 47-mm-diameter Pall Supor filter, respectively).
For each test, S. negevensis-infected Vero cell lysates or lysates of A. polyphaga persistently infected with S. negevensis (prepared as described below) served as controls. Twofold serial dilutions of these controls in distilled water (200-µl volumes) were filtered in duplicate through the manifold as calibration standards (Fig. 1B). Filters were processed as follows. They were frozen for 10 min at 70°C and incubated for 30 min at RT in 0.1% saponin in distilled water with vigorous shaking. The solution was removed, and the filter was frozen again at 70°C. The filter was then fixed for 15 min in a solution consisting of 30% methanol and 5% hydrogen peroxide, then fixed for 5 min in 70% ethanol, and fixed for 5 min in 96% ethanol. When the filter was almost dry, it was incubated for 45 min in a blocking solution containing 0.3% polyvinylpyrrolidone dissolved in TBS-Tween buffer, containing 10 mM Tris (pH 7.6), 150 mM NaCl, and 0.05% Tween-20, to which normal swine immunoglobulins (X-0906; DAKO), diluted 1:5,000, were added. The addition of swine immunoglobulins to the blocking buffer reduced the background due to nonspecific binding of rabbit or mouse immunoglobulins or of labeled secondary antibodies to biologic material on the filters.
![]() View larger version (65K): [in a new window] |
FIG. 1. Specificity of MEIA with respect to other bacteria commonly found in the respiratory tract. The MEIA was performed as described in Materials and Methods. Column A, 100 µl of various bacteria filtered in duplicate; column B, 100 µl of S. negevensis suspension in twofold dilutions.
|
The same procedure was employed for detection of amoebic antigens with mouse polyclonal anti-A. polyphaga (at 1:5,000) and HRP-conjugated, affinity-purified anti-mouse antibodies (115-035-003; Jackson), diluted 1:2,000.
Preparation of purified S. negevensis, Chlamydia trachomatis, and C. pneumoniae.
S. negevensis, C. trachomatis (L2/434/Bu), and C. pneumoniae (TW-183) were grown in Vero cells and purified by sucrose density gradient centrifugation as described previously (13, 17), except that the latter two organisms were grown in the absence of penicillin.
Preparation of positive controls for water studies.
Vero cells infected with S. negevensis were scraped off the growth surface with glass beads at 3 to 5 days postinfection; aliquots were frozen at 80°C in the presence of 50% fetal calf serum. Infectivity titers were determined by titration on Vero cells in the plate immunoperoxidase assay as previously described (15). A. polyphaga organisms persistently infected with S. negevensis were grown in PYG medium containing antibiotics as detailed previously (12). A. polyphaga organisms were counted, and their S. negevensis infectivity was determined by titration on Vero cells, as described previously (12). The infected amoebae served as positive controls in MEIA in those assays in which the presence of both S. negevensis and amoebic antigens were determined.
Hyperimmune sera used in assays.
Hyperimmune rabbit sera were prepared against S. negevensis, C. trachomatis, and C. pneumoniae as described previously for C. trachomatis and C. pneumoniae (7). Hyperimmune murine sera against A. polyphaga were produced by immunizing BALB/c mice with lysates of A . polyphaga grown in PYG.
List of bacterial strains and their American Type Culture Collection numbers.
A panel of eight standard bacterial strains obtained from the American Type Culture Collection were used in developing the present protocol of the MEIA. They were as follows: Branhamella (Moraxella) catarrhalis, ATCC 25238; Proteus vulgaris, ATCC 33420; Staphylococcus aureus, ATCC 25923; Escherichia coli, ATCC 25922; Staphylococcus epidermidis, ATCC 12228; Klebsiella pneumoniae, ATCC 13883; Haemophilus influenzae, ATCC 19418; Streptococcus pneumoniae, ATCC 49619. A loop of each of the bacteria in the panel was suspended in water (concentration, about 107/ml) and stored frozen at 70°C.
PCR assays.
DNA from drinking water samples was prepared as follows. A 500-ml volume of water sample was drained (by gravity) through a cellulose acetate filter (5-µm pore size, catalog no. 12342-47-N; Sartorius). The filter was incubated with Page's saline buffer (24) for 1 h. The solution containing the biological material was centrifuged for 20 min at 800 x g to sediment amoebae, and the precipitate was used to prepare DNA with the QIAamp kit (Qiagen). Assays for the detection of the S. negevensis genome by PCR were performed by using two sets of nested primers.
The first set consisted of primers ccF (CTT CGG GTT GTA AAG CAC TTT CGC) and ccR (CCC CGT CAA TTC TTT TGA GTT T), recognizing conserved chlamydial 16S ribosomal DNA (rDNA) sequences (433 to 457 and 933 to 945, respectively) and amplifying a 512-bp fragment, followed by nested ZpF and ZpR primers specific for S. negevensis Z (as previously described) (15), amplifying a 405-bp fragment.
A second set of nested primers consisted of AF and BR, amplifying a 1,099-bp fragment within the 23S rDNA of S. negevensis containing the large subunit intron (3), and IntD and IntR nested primers, amplifying a 338-bp fragment within this intron. If the DNA tested does not contain the intron, a 441-bp fragment is obtained with the AF and BR primers and no sequence is amplified by the IntF-IntR primer pair.
The use of the nested primers allowed increased sensitivity of the PCR assay when it was necessary; however, when DNA was abundant, primer set AF-BR, IntF-IntR, or ZpF-ZpR could also be used alone. PCR assays for the presence of amoebic sequences were carried out with the primers AmP1 and Amp2 described by Lai et al. (19) by using the cycling program described below for S. negevensis.
PCR amplifications were performed with Taq DNA polymerase (Takara Shuzo Co., Ltd., Kyoto, Japan). Amplification conditions were 1 cycle for 6 min at 94°C; 30 cycles of 60 s at 94°C, 60 s at 53°C, and 60 s at 73°C; and then 1 cycle of 10 min at 73°C (3).
When amplification was with the ccF-ccR primers, the same amplification program was used, except that annealing was performed at 58°C instead of 53°C. When nested primers were used, the material amplified with the first pair of primers was diluted 1:100, of which 1 µl was used in the second reaction mixture. The high dilution was needed to ensure the specificity of the assay.
|
|
|---|
Figures 1 and 2 demonstrate the specificity of the MEIA for Simkania with respect to a number of common bacterial respiratory pathogens (Fig. 1) and members of the Chlamydiaceae (Fig. 2). A volume of 100 µl was used for the assay. None of the panel of standard bacteria gave a positive signal with antibodies to S. negevensis and affinity-purified goat anti-rabbit HRP conjugates (Fig. 1), nor did C. trachomatis or C. pneumoniae antigens under the same conditions (Fig. 2).
![]() View larger version (64K): [in a new window] |
FIG. 2. Specificity of MEIA with respect to members of the Chlamydiaceae. Duplicate samples of 100-µl volumes of twofold dilutions of bacterial suspensions were applied to each of the membranes, as indicated. Sn, S. negevensis; Ct, C. trachomatis L2; Cpn, C. pneumoniae. Starting concentrations were as follows: S. negevensis, 4 x 104 organisms per dot; C. trachomatis L2, 2.5 x 104 organisms per dot; C. pneumoniae, 2.5 x 104 organisms per dot. Upper left panel, rabbit anti-S. negevensis serum used at 1:20,000; upper right panel, rabbit anti-C. trachomatis serum used at 1:20,000; lower left panel, rabbit anti-C. pneumoniae serum used at 1:20,000; lower right panel, serum diluent with secondary antibody only.
|
Effect of chlorination on S. negevensis particles and on S. negevensis in persistently infected amoebae.
In three separate experiments, the chlorinated suspension of Simkania particles had the same amount of infectivity as the control suspension that did not undergo chlorination. Control infectivity levels for the three experiments ranged from 1.2 x 104 to 2.8 x 105 IFU/ml. Survival values for S. negevensis particles ranged from 90 to 104% of the control untreated samples, with a mean of 96.8%. The variation obtained was within the experimental variation of titration. Similarly, S. negevensis in trophozoites of A. polyphaga were not sensitive to the treatment. In addition, chlorination treatment had no effect on antigen detection by MEIA (data not shown).
Survival in water of S. negevensis compared with C. trachomatis.
S. negevensis particles, S. negevensis-infected amoebae, and C. trachomatis L2 particles were exposed for various periods of time to sterile distilled water at RT, and both infectivity and antigen detection were monitored (Fig. 3). S. negevensis particle infectivity was remarkably preserved even after 7 days of incubation (ranging from 8 to 15% in various experiments). S. negevensis titers in infected amoebae even increased significantly under the same conditions. The infectivity of C. trachomatis L2 particles was reduced to 65% within 1 h and was completely destroyed by 5 h of exposure to distilled water. However, antigens were detected by MEIA in all samples tested at all time points. Even when infectivity was no longer present, no significant change in the level of antigen detection was observed (data not shown).
![]() View larger version (15K): [in a new window] |
FIG. 3. Residual infectivity of S. negevensis and C. trachomatis after incubation for various times in sterile distilled water at RT. Sn, S. negevensis; CtL2, C. trachomatis L2; Sn(A), S. negevensis in persistently infected A. castellani. The infectivity index is the ratio of infectivity at a given time point to the infectivity at time zero.
|
|
View this table: [in a new window] |
TABLE 1. Detection of S. negevensis and Acanthamoeba antigens in samples of drinking watera
|
|
View this table: [in a new window] |
TABLE 2. Presence of Simkania and amoebic antigens (as detected by MEIA) in samples of reclaimed wastewater and quality characteristics of the watera
|
![]() View larger version (45K): [in a new window] |
FIG. 4. Intron and amoebic amplicons detected by PCR assay of DNA prepared from water samples (lanes 2, 5, and 8) compared with DNA from purified cell culture-grown S. negevensis (lanes 1, 4, and 7), with the following primer pairs: AF-BR, lanes 1 and 2; intF-intR, lanes 4 and 5; amp1-amp2, lanes 7 and 8. M, marker (50- to 2,000-bp ladder; Bio-Rad).
|
![]() View larger version (21K): [in a new window] |
FIG. 5. Nested PCR for detection of 16S rDNA S. negevensis sequences in DNA prepared from water samples. The primers used were as follows: lanes 1 to 4, ccF-ccR; lanes 5 to 8, ZpF-ZpR (nested primers). Lanes 1 and 5: positive control; lanes 2 to 3 and 6 to 7: DNA prepared from two water samples; lanes 4 and 8, negative controls (no DNA).
|
|
|
|---|
Chlamydia-like organisms other than Simkania, members of the family Parachlamydiaceae, have been shown to be present in the environment and to be able to grow and survive in protozoa, such as Acanthamoeba castellani and Hartmanella (9, 11), and lately, to replicate in human monocyte-derived macrophages (8).
Because we are able to grow S. negevensis in human and simian cells lines, we are able to isolate it from human clinical samples. We have also succeeded in isolating it from natural water samples (unpublished data). However, since the methods described in this study detect bacterial antigens or genomes only and are not capable of determining whether the organisms are viable, further studies are needed to understand the significance of the detection of S. negevensis or S. negevensis-like microorganisms in water. Aspects of such studies could include determination of the possible correlation between a high seroprevalence of antibodies to S. negevensis in healthy population samples and the presence of the organisms in the drinking water or agricultural irrigation water used in the same area. Sera obtained from a number of groups of healthy pregnant women (n = 192) living in the same areas from which water samples were obtained for this study showed an average prevalence of antibodies to S. negevensis of 75% (range, 50 to 92% for the different groups) (unpublished data), as assayed by enzyme-linked immunosorbent assay (6). The serum samples were from Jewish and Bedouin women living in rural and urban locations. The seropositivity ranges are consistent with published data (5, 6).
If the organisms found in drinking water are indeed viable, it is possible that their detection is of relevance to infection of the respiratory tract, in that drinking water is also generally used for oral hygiene, bathing, and showering, which can result in the formation of inhalable aerosols. Also, organisms found in drinking water may infect the oropharyngeal epithelium, thus gaining access to the respiratory and gastrointestinal tracts. In an ongoing study in our laboratory, S. negevensis strains isolated from children with respiratory disease are being compared to strains that may be isolated from their households.
Although laboratory experiments indicate that S. negevensis is more resilient than other similar bacteria, such as Chlamydia, we are well aware that under environmental conditions, bacteria and their antigens may have a shorter survival time due to the presence of microorganisms and factors such as various chemicals in the water. Clearly, it would be advantageous to have simple surrogate markers for viability of environmental chlamydia-like bacteria, but these are not presently available. The possibility of transmission of potential pathogens in drinking water is clear, but reclaimed wastewater may be used for irrigation of crops and its pathogenic potential is also of concern. The significance of the ability of S. negevensis or S. negevensis-like microorganisms to survive in various water sources needs to be further studied to establish public health guidelines taking this into account. Future research will enable conclusions about the environmental, epidemiological, and health care implications of the very common existence of simkaniae in various water sources.
This study was supported by grant no. TA-Mou-99-C19-033 funded by the United States-Israel Cooperative Development Research Program, Bureau for Economic Growth, Agriculture, and Trade, U.S. Agency for International Development; by grant no. 4672/0 from the Office of the Chief Scientist of the Israel Ministry of Health, via the Keren Kayemet LeIsrael; and by a grant from the "Pinchas Sapir" fund of Mifal Hapayis.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»