in5ml of LB broth and grown again at37◦C,150 rpm for3h.The dilutions of culture were made at this point and used for template preparation.

Template preparation

The LB grown cultures containing109c.f.u.were se-rially diluted with sterile deionized water to give105 to101cells per0.5ml.Dilutions were confirmed by spotting on EMB/SS/TCBS agar plates,respectively. The diluted samples of the required cell concentra-tions were pooled and combined in a single tube to give1.5ml of total volume.The mixtures of these cells were harvested by centrifugation at20000×g for30min.The cell pellets were washed twice with 1.0ml of sterile saline solution(0.9%NaCl in deion-ized water)and once with sterile deionized water and pelleted at20000×g for30min.The pellets were suspended in8.5µl sterile deionized water.In a PCR tube,8.5µl of cell suspension was mixed with1µl 10×TE(100mM Tris,5mM EDTA,pH8.0)and 0.5µl proteinase K(20mg ml−1);to give afinal con-centration of10µg proteinase K in1×TE.The cells were lysed to release the target DNA by incubating at 37◦C for15min,60◦C for10min andfinally,the proteinase K was denatured at98◦C for10min The lysed samples were used directly as template in the M-PCR or kept frozen at−20◦C until needed. Primers used

The target pathogens were selected since they are known to cause water and food borne diseases. Following pathogens and their target loci were used in the study. E.coli is primarily a habi-tant of human gut and hence,a useful indica-tor of fecal contamination.PCR primers that were used in the detection of E.coli targeted the lamB locus(Bej et al.1990).The reported primers BL4910(5 -CTGATCGAATGGCTGCCAGGCTCC-3 )and BR5219(5 -CAACCAGACGATAGTTATCA-CGCA-3 )were used to amplify a309bp frag-ment.The following invA and ctxA primers have been designed using the LaserGene software,DNAS-TAR.Salmonellae one of the etiological agents for typhoid and gastrointestinal infection were detected with the invA locus.This protein is a component of the sec-independent secretion apparatus,which is necessary for full virulence of the bacterium as demonstrated by in vitro studies(Clark et al.1996, Jones et al.1994).Salmonella spp.primers were designed from the invA gene(Galan et al.1992,GenBank accession no.M90846).Primer InvA F,5 -CCTGATCGCACTGAATATCGTACTG-3 ,is located between201and225bp,and primer InvA R,5 -GACCATCACCAATGGTCAGCAGG-3 ,is located between776and799bp;and could amplify a598bp product.The ctx operon of Vibrio spp.was chosen as target locus for inclusion in the M-PCR protocol as it is known to be associated with pathogenicity(Shi-rai et al.1991).A435bp product is amplified from Vibrio spp.using the primers derived from the ctxA se-quence(GenBank accession no.D30053).The primer ctxA F-1,5 -CTCAGACGGGATTTGTTAGGCACG-3 located between306and329bp,and ctxA R,5 -GATCTTGGAGCATTCCCACAACC-3 as antisense primer,located between719–741bp,which could am-plify a435bp product.All the primers purchased in the study were synthesized by Gibco BRL,USA. PCR amplification

The reactions were performed in the Perkin Elmer Gene Amp system9600.Different programs used in the optimization of the M-PCR are listed in Table1. The time required between the two temperature steps as gradient is defined as the ramp.The reaction con-ditions were used as described earlier for duplex PCR for Salmonella and Vibrio in afinal reaction volume of 25µl(Kapley et al.2000).

Results

Factors that influence the amplification of multi-target loci have been investigated with specific reference to amplification of enteropathogens E.coli,Salmonella and Vibrio spp.The major factors involved in product amplification are discussed below.

Effect of the thermocycling program Thermocycling programs used in developing the mul-tiplex PCR assay to detect three pathogens are shown in Table1.The PCR product profile for different ther-mocycling programs is shown in Figure1.Template concentration used in the reaction for all the programs was103c.f.u.A single-step PCR reaction with an an-nealing temperature of55◦C did not amplify all three loci and resulted in amplification of many non-specific products(Figure1,lane2).Similarly,by varying the temperature between45–60◦C,no difference was ob-served(data not shown).Program B,the double-step program,also did not yield the PCR products for the1915 Table1.Development of temperature programs for optimization of M-PCR.

Program A35cycles of94◦C for60s,45/55/60◦C for90s and72◦C for60s

Program B

Step115cycles of94◦C for30s,45◦C for90s and72◦C for20s

Step220cycles of94◦C for20s,60◦C for45s and72◦C for30s

Program C

Step15cycles of96◦C for30s,60◦C for60s,72◦C for30s with ramp of170s

Step25cycles of95◦C for30s,55◦C for60s,72◦C for30s,with ramp of170s

Step320cycles of94◦C for20s,51◦C for30s,72◦C for45s,with a ramp of60s

Program D

Step15cycles of95◦C for30s,45◦C for10s,72◦C for10s,with a ramp of90s

Step25cycles of95◦C for30s,51◦C for30s,72◦C for10s,with a ramp of70s

Step35cycles of95◦C for30s,54◦C for30s,72◦C for10s,with a ramp of60s

Step45cycles of95◦C for30s,57◦C for60s,72◦C for10s,with a ramp of50s

Step510cycles of94◦C for10s,60◦C for60s,72◦C for15s,with a ramp of80s

1916

Fig.2.Demonstration of sensitivity and optimization of primer ra-tios in the M-PCR using thermocycling program D.Lanes1,7,9 and14show the100bp standard ladder from Gibco BRL,USA. Lanes2to6demonstrate the sensitivity of M-PCR with template concentration derived from E.coli,Salmonella and Vibrio spp.;lane 2,105c.f.u.;lane3,104c.f.u.;lane4,103c.f.u.;lane5,102 c.f.u.;lane6,101c.f.u.;lane9,amplification product using103 c.f.u.Pseudomonas putida as template;lanes10to13show the different primer mixing ratios tried.Lane10had0.25:0.25:0.25, lane11had0.25:0.5:0.5,lane12had0.25:0.5:0.25and lane13 had0.25:1.0:1.0ratios of lamB:invA:ctxA inµM concentration per reaction,respectively.

organism using the following primer concentrations. The primer mixing ratios used were as follows in the different lanes seen in Figure2,viz.,lane10had 0.25:0.25:0.25,lane11had0.25:0.5:0.5,lane12had 0.25:0.5:0.25and lane13had0.25:1.0:1.0ratios of lamB:invA:ctxA inµM concentration per reaction,re-spectively.All primer concentrations tried showed the amplification of the three target loci but with different degrees of amplification of each product.

The preference in amplification of products varies with the primer concentration even when the target template,as number of cells per reaction,was kept constant as shown in Figure2,lanes10–13.The lamB product was observed to be more dominant than invA and ctxA with all the primer ratios tried.Equimo-lar primer ratio does not amplify all three-target loci evenly,as seen in Figure2,lane10.When only the concentration of invA primers was increased to0.5µM there was no significant change observed(lane12);it also had no effect on non-specific amplification.The primer concentration which led to the best possible option for all the products to amplify equally as seen in lane11was lamB at0.25µM;invA and ctxA at 0.5µM.Increasing the concentration of invA and ctxA to1µM gave rise to non-specific amplification as shown in lane13.

The specificity of the primers was tested by using 103c.f.u.of Pseudomonas putida as the template and using the primer ratio described for lane11.This re-action did not give non-specific extension of primers even after loading complete reaction volume,as shown in Figure2,lane8;it signifies that the primers are specific to the target template.Similarly,the addition of the same non-specific template did not alter the PCR pattern when tried with basic composition of reaction as used for lane11(data not shown).

Sensitivity of the M-PCR

Template concentration ranging from105to101c.f.u. of each organism was tested to determine the sensi-tivity of the M-PCR.Target templates were amplified using the optimum primer concentration as described for lane11.The amplification products can be seen in Figure2,lanes2to6.The results show that the sensitivity of the reaction is102c.f.u.The lane with 101c.f.u.of each organism did not show any product. This could probably be due to the fact the cells could not be recovered in the initial process of template preparation.

Discussion

M-PCR,since it wasfirst reported,has been discussed in detail regarding the interaction of its reaction com-ponents to amplify one conserved locus from different templates or different loci from the same template (Chamberlain et al.1988,Franck et al.1998,Hene-garieu et al.1997,McGregor et al.1996,Tsen& Jian1998,Wang et al.1997).In this paper we de-scribe the importance of the thermocycling steps and role of multi-step temperature cycles,which ensure the specific annealing of primers having different melt-ing temperatures.The reaction mixture also had the heterogeneous template derived through a simple total DNA extraction protocol.We have described the or-ganism specific PCR,which has been optimized based on a single step PCR protocol(Purohit et al.1997). The single step cycle,when extended to even two loci,resulted only in smearing as non-specific exten-sion.The annealing temperatures tried were45◦C, 55◦C and60◦C.All three gave the same result that is shown in Figure1,lane2.The program earlier re-ported for Salmonella and Vibrio duplex(Kapley et al. 2000),which uses a three-step program,also gave only lamB product with thick smearing covering faint ctxA1917

product.This may be due to additional template load, which reduces the primer specificity(data not shown). Program B was designed with15cycles of amplifi-cation at45◦C.The remaining cycles were carried out at60◦C to increase the specificity of the M-PCR as shown in Table1.This program gave a band for E.coli hidden in the smearing due to non-specific products.The results suggested that the dense zone of non-specific products between300bp to700bp might be due to uncontrolled extension of specifically annealed primers.This prompted us to use the ramp in the next program designed,as was used before in du-plex PCR(Kapley et al.2000).The three-step program C was tried to address this non-specific extension.The reasoning here was to reduce the non-specific smear-ing by carrying out the amplification of thefirst few cycles at higher temperature,which should result in amplification of specific target products.This method did amplify the lamB and invA locus,however the ctxA could not be amplified because either the anneal-ing time or the temperature was not optimum in this method.

After trying different combinations the best-fit conditions resulted in program D,which amplified all the three products with sensitivity up to102c.f.u.The descending temperature for annealing starting at60◦C to45◦C,i.e.,step5to step1failed to amplify both invA and ctxA.Program D started withfive cycles hav-ing annealing temperature of45◦C to co-amplify the three targeted genes.The annealing temperature was raised to51◦C in step2,and thereafter the annealing temperature was raised at the rate of3◦C for every step in thefive steps of program D.The same pro-gram when operated without considering the ramp and with the usual time required between the temperature steps by thermocycler yielded lighter product bands covered with smearing.The ramp,with slower tem-perature gradient than0.3◦C s−1gave non-specific extension for thefirst four steps.The last step,how-ever,did require a0.15◦C s−1gradient since it uses 60◦C as annealing temperature(data not shown). The ramp provides the condition where the specific primers stay on to the template even after a tempera-ture gradient.The same conditions remove the poorly annealed primers,thereby reducing the mis-priming in the extension cycles as shown in Figure2.

Primer ratios also affect M-PCR,besides the an-nealing temperatures and extension time.Initial re-sults showed that using equimolar concentrations of 0.25µM primers with a high number of template gave good results with105c.f.u.(data not shown).How-ever,these primer concentrations were not efficient to equally amplify all the three PCR products when template concentration was lowered to103,as seen in Figure2,lane10.By adjusting the primer ratio as described earlier,we found that even with a low template concentration(c.f.u.103),each target product was equally generated.The ratio of1.0µM primers for Salmonella and Vibrio gave additional bands with 103and above template levels.

The optimization of temperature cycles for M-PCR protocol in this study yielded a protocol for the de-tection of three water-borne pathogens.The depleting reserves of clean and potable water is a major con-cern for the developing countries.A WHO report in 1996states that the developing world faces a high risk from cholera and typhoid infections(WHO/EOS/96); hence,we have chosen Salmonella and Vibrio as tar-get organisms with E.coli as an indicator of fecal contamination.We have earlier reported the duplex amplification of Salmonella and Vibrio using the phoE primers.Even though these primers have been tested for132different Salmonellae,we have not included them in this multiplex reaction,since the amplifica-tion product(365bp)does not clearly resolve on the gel along with the lamB product(309bp).Primers specific for Vibrio spp.amplified a region of the ctx operon that encodes for cholera enterotoxin,a major virulence determinant of V.cholerae01.Primer ctxAF and ctxAR give an amplification product of485bp (Kapley et al.2000),while ctxAF-1,used in this study, and ctxAR gave a435bp product.The use of either primer set did not effect the sensitivity of the program. This suggests that the developed thermocycling proto-col D can accommodate minorfluctuations in primer types.For the multiplex protocol,we have used the 435bp-amplification product of the ctxA gene since it is clearly distinguishable with lamB and invA on agarose gels.

Acknowledgement

This work was supported by the Department of Biotechnology,Ministry of Science and Technology, New Delhi,India.

References

Bej AK,Steffan RJ,DiCesare J,Haff L,Atlas RM(1990)Detection of coliform bacteria in water by polymerase chain reaction and gene probes.Appl.Environ.Microbiol.56:307–314.1918

Chamberlain JS,Gibbs RA,Ranier JE,Nguyen PN,Caskey CT (1988)Deletion screening of the Duchenne muscular dystrophy locus via multiplex DNA amplification.Nucl.Acids Res.16: 11141–11156.

Clark CG,MacDonald LA,Ginocchio CC,Galan JE,Johnson RP (1996)Salmonella typhimurium InvA expression probed with a monoclonal antibody to the C-terminal peptide of InvA.FEMS Microbiol.Lett.136:263–268.

Franck SM,Bodworth BT,Moon HW(1998)Multiplex PCR for en-terotoxigenic,attaching and effacing,and shiga toxin-producing Escherichia coli strains from claves.J.Clin.Microbiol.36: 1795–1798.

Galan JE,Ginocchio C,Costeas P(1992)Molecular and functional organization of the Salmonella invasion gene,invA:homology of InvA to members of a new protein family.J.Bacteriol.174: 4338–4349.

Henegarieu O,Heerema NA,Dlouhy SR,Vance GH,V ogt PH (1997)Multiplex PCR:critical parameters and step by step protocol.BioTechniques23:504–511.

Jones BD,Ghori N,Falkow S(1994)Salmonella typhimurium initiates murine infection by penetrating and destroying the spe-cialized epithelial M cells of the Peyer’s patches.J.Exp.Med.

180:15–23.

Kapley A,Lampel K,Purohit HJ(2000)Development of duplex PCR for the detection of Salmonella and Vibrio in drinking water.

World J.Microbiol.Biotech.16(in press).

Kreader CA(1996)Relief of amplification inhibition in PCR with bovine serum albumin or T4gene32protein.Appl.Environ.

Microbiol.62:1102–1196.McGregor DP,Forster S,Steven J,Adair J,Leary SEC,Leslie DL, Harris WJ,Titball RW(1996)Simultaneous detection of mi-croorganisms in soil suspension based on PCR amplification of bacterial16S rRNA fragments.BioTechniques21:463–471. Purohit HJ,Kapley A,Khanna P(1997)Detection of en-teropathogens by PCR.Indian J.Clin.Biochem.12:111–114. Saiki RK(19)The design and optimization of the PCR.In:Erlich HA,ed.PCR Technology.USA:Stochton Press,pp.7–16. Shirai H,Nishibuchi M,Ramamurthy T,Bhattacharya SK,Pal SC, Takeda Y(1991)Polymerase chain reaction for detection of the cholera entertoxin operon of Vibrio cholerae.J.Clin.Microbiol.

29:2517–2521.

Tsen HY,Jian LZ(1998)Development and use of multiplex PCR system for the rapid screening of heat labile toxin I,heat labile toxin II and shiga-like toxin I and I genes of Escherichia coli in water.J.Appl.Microbiol.84:585–593.

Wang RF,Cao WW,Cerniglia CE(1997)A universal protocol for PCR detection of13species of foodborne pathogens in food.J.

Appl.Microbiol.83:727–736.

Way JS,Josephson KL,Pillai SD,Abbaszadegan M,Gerba CP, Pepper IL(1993)Specific detection of Salmonella spp.by mul-tiplex polymerase chain reaction.Appl.Environ.Microbiol.59: 1473–1479.

World Health Organization(1996)‘Water supply and sanitation sec-tor monitoring report’.Sector status as of31December1994.

WHO/EOS/96.15.Geneva,Switzerland.下载本文