Detection of Plasmid Chlamydia trachomatis Serovars Two ...Indianapolis, Ind.), 66mMTris(pH8.6), 6.6mMMgCl2, 1.7 mM(NH4)2SO4, 10% (vol/vol) dimethyl sulfoxide, and 10 mM 2-mercaptoethanol,

Vol. 56, No. 8APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Aug. 1990, p. 2494-24980099-2240/90/082494-05$02.00/0Copyright X) 1990, American Society for Microbiology

Detection of Plasmid DNA from All Chlamydia trachomatisSerovars with a Two-Step Polymerase Chain Reaction

DAVID WELCH,1t CHAO-HUNG LEE,' AND STEVEN H. LARSEN'*Department of Microbiology and Immunology' and Department of Pathology,'

Indiana University School of Medicine, Indianapolis, Indiana 46202

Received 20 February 1990/Accepted 31 May 1990

A polymerase chain reaction was used to amplify a 137-base-pair sequence of DNA from a Chlamydiatrachomatis plasmid. Various parameters of the polymerase chain reaction were explored, and it was found thattwo short steps per reaction cycle were sufficient to achieve 1012-fold amplification in less than 1 h. By use ofthis procedure, 10-18 g of a sequence of plasmid DNA, representing the amount of that sequence found in oneC. trachomatis bacterium, was amplified to the point where it was clearly visible on an ethidium bromide-stained polyacrylamide gel under UV light. DNA from intact cells from each of the 15 serovars of C. trachomatiscould also be amplified for visualization. With this procedure, the presence or absence of C. trachomatis DNA ina sample could be established in less than 1.5 h. The speed and extreme sensitivity of this detection procedure maymake it a useful method for the detection of C. trachomatis, and similar techniques should be possible for anytype of bacteria.

Chiamydia trachomatis is best identified by tissue culturemethods that are difficult and time consuming (1). Directfluorescent-antibody tests and enzyme-linked immunosor-bent assays have recently been developed, but they haveonly 60 to 90% of the sensitivity and specificity that tissueculture methods do. Hence, a highly sensitive method ofdetection is desirable.

Immunological studies have classified C. trachomatis into15 serovars: A through K, Da, and L, through L3, each withdistinct immunological properties (29). Although each sero-var possesses a unique genome (22), all carry about 10 copiesof a common plasmid that is approximately 7.5 kilobase pairsin length (10, 12, 20, 22). The conservation of this plasmidmakes it an obvious target for the development of nucleic-acid-based methods of detection. Although several protocolshave been developed that successfully detect the C. tracho-matis plasmid DNA (6, 9, 10, 21), these methods require atleast 10 pg of DNA (10, 21) and are labor intensive. Inaddition, the requirement of working with radioactive iso-topes is not optimal. Recently, a procedure called a poly-merase chain reaction (PCR) was developed that can amplifysubpicogram quantities of DNA to the point where they canbe detected without radioactive hybridization probes.PCR is a method for the rapid amplification of specific

sequences of DNA which uses two oligonucleotide primerscomplementary to opposite strands of the sequence to beamplified (18, 27). The DNA in a sample is denatured andcooled to allow annealing of the primers to the single-stranded template. A thermostable DNA polymerase fromThermus aquaticus, Taq DNA polymerase (4), then cata-lyzes primer-directed synthesis of the DNA (26). In thepresence of an excess of enzyme, primers, and deoxynucle-oside triphosphates, repeated cycles of denaturation andpolymerization result in exponential amplification of thetarget sequence until the number of target molecules synthe-sized exceeds the number of enzyme molecules in the

* Corresponding author.t Present address: Department of Biochemistry and Molecular

Biology, Harvard University, Cambridge, MA 02138.

reaction or a substrate becomes limiting (17). PCR with TaqDNA polymerase has been used to amplify small quantitiesofDNA until they were visible directly on an electrophoreticgel (3, 13, 15, 25-27).PCR has been applied to the detection of Chlamydia spp.

by two laboratories to date. Dutilh et al. (7) used primers forthe major outer membrane protein gene target that recog-nized all but serovar J. Griffais and Thibon (8) used a PCRtest that resulted in the amplification of 8 of 200 clinicalspecimens. Of these eight specimens, six were also positiveby culture and the other two fell into a group of fourspecimens that were doubtful by culture. They do not reporton how many serovars their system will amplify.

This paper reports the development of a two-step PCR (17)designed to amplify a portion of the C. trachomatis plasmid.Important aspects of the methodology are detailed. Thistechnique amplified DNA from all chlamydial serovars andreadily detected as few as 10 target molecules, the amountfound in a single C. trachomatis bacterium. A 1012-foldamplification could be achieved in 1 h, generating enoughDNA to be readily detected by gel electrophoresis.

MATERIALS AND METHODS

DNA isolation, sequencing, and synthesis. PCR conditionswere tested with pSL125, a derivative of pBR327 containingthe entire plasmid of the L2 serovar strain 434 cloned into thevector plasmid pBR327 (10). pSL125 DNA was isolated byalkali extraction (2). To isolate total DNA from Percoll(Sigma Chemical Co., St. Louis, Mo.)-purified elementarybodies of C. trachomatis, 50 jig of cells from each serovarwas suspended in 50 RI ofTE buffer (10 mM Tris hydrochlo-ride [pH 8.0], 1 mM EDTA) and centrifuged for 1 min at14,000 x g. The pellet was then suspended by vortexing in100 RI of TE buffer containing 100 mM NaCl. Sodiumdodecyl sulfate was added to 1%, and the DNA was phenol-chloroform extracted, ethanol precipitated, and suspendedin 20 ,ul of TE buffer.Approximately 200 base pairs (bp) of the chlamydial DNA

in pSL125 were sequenced by the dideoxynucleotide se-quencing method (31) by using Sequenase (U.S. Biochemical

2494

on May 11, 2021 by guest

http://aem.asm

.org/D

ownloaded from

C. TRACHOMATIS DETECTION BY PCR 2495

Corp., Cleveland, Ohio). Using these sequence data, oligo-nucleotide primers were then synthesized with a DNAsynthesizer (Model 381A; Applied Biosystems, Inc., FosterCity, Calif.).PCR conditions. General conditions for reaction mixtures,

unless specified otherwise, were as follows: 100 ag ofpSL125 (which is equivalent to 10 copies of the plasmid), 60pmol of each appropriate primer, 660 ,uM of each deoxynu-cleoside triphosphate 1 U of Taq DNA polymerase (PerkinElmer-Cetus, Norwalk, Conn.), 100 pug of nuclease-freebovine serum albumin (Boehringer Mannheim Biochemicals,Indianapolis, Ind.), 66 mM Tris (pH 8.6), 6.6 mM MgCl2, 1.7mM (NH4)2SO4, 10% (vol/vol) dimethyl sulfoxide, and 10mM 2-mercaptoethanol, in a final volume of 30 ,ul andcapped with 15 ,ul of mineral oil.

Reactions were carried out on a Perkin Elmer-CetusThermocycler or a Coy TempCycler (Coy Laboratory Prod-ucts, Ann Arbor, Mich.) with similar results. The Thermo-cycler and TempCycler are programmable heat blocks thatrequire additional time to raise and lower temperaturesbetween cycle steps; therefore, the real time of each PCR islonger than the sum of each cycle step. The cycle length andtemperature were varied to determine optimal conditions.

After amplification was completed, 10 p.l of each reactionmixture was electrophoresed in a 5% acrylamide gel (5) andstained with 0.5 p.l of ethidium bromide per ml, and the gelwas photographed under UV light.Because of the extreme sensitivity of this amplification

technique, care was taken to avoid contamination by pipet-tors or airborne particles. Different rooms and pipettorswere used for making reaction mixtures than were used formaking amplified materials. It was also important to runnegative controls which lacked template DNA for all tests.

Restriction enzyme digestions. Digestion of amplified prod-ucts was achieved by adding 10 p.l of PCR product to 2 p.l of500 mM Tris hydrochloride (pH 8.0), 100 mM MgCl2, and 4U of the restriction endonuclease AluI (Bethesda ResearchLaboratories, Inc., Gaithersburg, Md.) in a total volume of20 p.J. The reaction mixture was incubated at 37°C for 2 h andstopped by adding EDTA to a final concentration of 10 mM.

RESULTS

DNA sequencing and primers for synthesis. The primers forPCR were based on the sequence data shown in the legend toFig. 1. Because of the extremely low G+C content of theplasmid (35% for the segment sequenced), primers weredesigned to be complementary to regions that are relativelyhigh in G+C. Primers s249, s250, and s260 contain 25 baseswith a G+C content of 40 to 44% (see Fig. 1 legend). Primers32 bases long with identical 5' ends to s249 and s260 werealso synthesizedPCR detection of plasmid DNA. As few as 10 copies of

plasmid could be readily detected (Fig. 2A, lanes 1 through3). PCR primers 25 bases in length were sufficient forannealing/elongation to occur at a temperature of 64°C (Fig.2A, lanes 1 and 2), whereas primers 32 bases long allowedannealing/elongation at a temperature of 72°C (Fig. 2A, lane3). The small reduction in total yield in lane 3 is apparentlythe result of experiment-to-experiment variability since anyone of the pairs of primers gave the same results on average.The exact concentration of Mg2+ has been reported to

significantly influence the PCR efficiency (11). We found thebuffer system reported here (basically similar to that given inreference 13) to be relatively insensitive to small changes inMg2+ concentration. Concentrations between 6.6 and 10

A

s249 Alu s260B I

FIG. 1. Plasmid map and target sequences. The 10-kilobasehybrid plasmid pSL125 was constructed from the plasmid of C.trachomatis serovar L2 strain LGV434 and plasmid pBR327 (7). TheA target sequence, flanked by primers s249 and s260, spans 137 bp.The larger target region, B, spans 160 bp with the followingsequence: CTlTGCGCACAGACGATCTATTTTTTGCATCCAATCAATCAGATTTCCTTTCGCATTAAAAAAAGACAGAATAAAGAAACCAAAATTCTAATCACATTTCCTATCAGCTTAATGGAGGAGTTGCAAAAATACACTTGTGGGAGAAATGGGAGAGTATTTGTTTC. Primer s249 is identical to the first 25 bases listed,primer s260 is complementary to bases 137 through 113, and primers250 is complementary to the last 25 bases listed.

mM all seemed to work satisfactorily (Fig. 2B). When thedimethyl sulfoxide concentration was varied, concentrationsbetween 5 and 10% were found to be satisfactory (data notshown). Yields were significantly reduced at dimethyl sul-foxide concentrations below 5% and at 15%.

Verification of the PCR product. To verify that amplifiedDNA products represented the correct target sequences, the

A B

Ml 2 3 CM 1 2 3 4 5 CM

- 194-118-194

-118

FIG. 2. Amplification of the chlamydial plasmid by PCR. Eachreaction underwent 45 cycles of amplification with 10 copies ofpSL125 template and a denaturation step of 1 s at 95°C. (A) Lane 1used primers s249 and s250 to produce a 160-bp fragment; lane 2used primers s249 and s260 to produce a 137-bp fragment; lane 3used the longer primers that had 5' termini coincident with s249 ands260 to produce the same 137-bp fragment seen in lane 2. Theannealing/elongation steps were at 64°C for 30 s (lanes 1 and 2) or at72°C for 30 s (lane 3). (B) The relative independence of the Mg2lconcentration in the PCR buffer system is shown. The reactionmixtures were identical to those shown in lane 2 of panel A exceptthat the Mg2+ concentrations were 4, 5, 6.6, 8, and 10 mM in lanes1 through 5, respectively. (A and B) Lanes M contain HaeIII-digested CPX174 DNA size marker; relevant band sizes are shown inthe margin. Lanes C show negative control reactions which lackedtemplate DNA.

VOL. 56, 1990

on May 11, 2021 by guest

http://aem.asm

.org/D

ownloaded from

APPL. ENVIRON. MICROBIOL.

A B C

194-

118.

72-

D E

-194

-118-'72

FIG. 3. Verification of target sequences. PCR conditions were 23cycles of 1 s at 95°C and 30 s at 64°C. A 10-ng amount of pSL125template was used. Lane A contains the AluI-digested 137-bp targetsequence A (see Fig. 1); lane B contains the undigested sequence A;lane C contains the negative control reaction mixture which lackedtemplate DNA, undigested; lane D contains the 160-bp targetsequence B (Fig. 1) digested with AluI; lane E contains the undi-gested sequence B. The positions of fragments of known length areindicated at either side.

appropriate primers were used to generate the sequenceslabeled A and B shown in Fig. 1. Half of each amplificationproduct was digested with AluI and compared with both theundigested products and the HaeIII-digested OX174 sizemarker.

Digestion of target sequence A (Fig. 1) with Alul shouldyield 103- and 34-bp fragments, whereas Alul-digested se-quence B (Fig. 1) should yield the same 103-bp fragment anda 57-bp fragment. Figure 3 shows the results of the diges-tions: the 103- and 34-bp bands are clearly visible in lane Aand the 103- and 57-bp fragments are visible in lane D. AluIdigestions are not quite complete, probably because ofmaterials present in the PCR buffer. Additional confirmationwas derived by probing dot blots of the amplified materialwith a biotinylated 30-base probe that binds to the middle ofa 137-bp fragment (14). Although control reactions (to whichtemplate was not added) were not detected, the PCR prod-ucts did hybridize with the probe (data not shown).Template DNA preparation by various extraction methods.

A less-labor-intensive method for obtaining DNA from C.trachomatis than that described in Materials and Methodswould greatly facilitate future PCRs with Chlamydia spp. or

other bacteria. Several techniques were tested to reveal thesimplest way to obtain amplifiable DNA from whole elemen-tary bodies of Chlamydia spp. Figure 4 shows the results ofboiling the samples with or without reducing agents and theresults obtained by simply adding whole cells to the PCRmixture, which reaches a temperature of 95°C itself. Allmethods, including adding whole cells directly to the PCRmixture, result in satisfactory amplification of target DNA.

Detection of chlamydial serovars. Because the direct addi-tion of intact cells to the PCR mixture was by far simplerthan the DNA extraction methods, it was used for amplifi-cation of all chlamydial serovars. All 15 serovars of C.

I Ila llb lic III M

1994118

FIG. 4. PCR of template DNA obtained by various DNA extrac-tion methods. Reaction conditions were identical to those in Fig.2A, lane 2. Lane I contains DNA extracted from 1 ,ug of elementarybodies as described in Materials and Methods; for lanes hIa throughIlc, 1 ,ug of whole bacteria, serovar E, had been boiled for 5 min inTE buffer alone (lane Ila), TE buffer with 40 mM dithiothreitol (laneIlb), or TE buffer with 1% 2-mercaptoethanol (lane IIc). Lane IIIcontains 1 ,ig of whole bacteria of serovar E added directly to thePCR mixture. Lane M contains HaeIII-digested 4X174 DNA.

trachomatis were amplified by this system as demonstratedby the appearance of 137-bp fragments (Fig. 5). A sample ofguinea pig inclusion conjunctivitis agent, i.e., a strain ofChlamydia psittaci, showed variable results: amplificationresulted in some experiments (Fig. 5) but not in others.Neisseria gonorrhoeae DNA (Fig. 5, lane N) did produce asmear of very short oligonucleotides but nothing in the137-bp region. The nature of DNA in the smear is notknown; however, material tested after 25 cycles yielded novisible product. Negative controls containing Klebsiellapneumonia, Pseudomonas aeruginosa, Salmonella enteridi-tis, Serratia marcescens, Staphylococcus aureus, Staphylo-coccus epidermidis, Streptococcus salivarius, human,mouse, and pBR327 plasmid DNAs yielded no amplificationproduct in the region of 137 bp on the gel. In fact, only twogave any visible product at all: very faint bands at 600 bp forS. epidermidis and 650 bp for S.enteriditis (data not shown).

DISCUSSION

We have developed a procedure to amplify a small portionof the plasmid of all C. trachomatis serovars with a highdegree of specificity. This amplification technique is a two-step PCR which can detect 10 copies of the plasmid, thesame number present in a single C. trachomatis bacterium,and amplify a sequence of that plasmid until it is clearlyvisible on an ethidium bromide-stained polyacrylamide gel.To establish this procedure, a number of experimentalparameters were tested for rapid amplification with a mini-mum of mispriming or misextension (11).

Reaction conditions. Relatively long primers allowed an-nealing to occur at 64 to 72°C, at which temperatures the TaqDNA polymerase (28) produced excellent results (Fig. 2).The particular buffer system used gave amplification of allserovars without the usual need to adjust the exact Mg2"concentration. We have now used this buffer system withover 20 pairs of primers and various templates, and all have

2496 WELCH ET AL.

on May 11, 2021 by guest

http://aem.asm

.org/D

ownloaded from

C. TRACHOMATIS DETECTION BY PCR 2497

E m < m m u aw I IL I_

234 _t

1188

FIG. 5. Amplification of DNA from all C. trachomatis serovars. A 1-Rg amount of elementary bodies was added to each sample andamplified for 30 cycles as described in Fig. 2A, lane 2. Lane S contains 6 ng of pSL125 template, lane P contains 1 jig of whole cells of guineapig inclusion conjunctivitis agent, and lane N contains 1 jig of N. gonorrhoeae DNA. Lanes marked M contain HaeIII-digested 4X174 DNA.

worked with 6.6 mM MgCl2 present. Thus, this system maybe particularly useful for anyone testing a new system sinceit precludes the need to try various MgCl2 concentrations. Itis notable that dimethyl sulfoxide is needed for this reactionto demonstrate good efficiency.The above conditions allowed the total reaction time to be

reduced by two-thirds of what previously published proce-

dures required. Also, we have been able to increase thenumber of cycles from approximately 30 to 35 cycles to 55cycles with excellent results (11).

Efficiency of the two-step PCR. The target sequence foramplification (approximately 1% of the total plasmid) has amass of approximately 10-19 g. Ten copies of this sequencewere amplified to 2.7 jLg (as determined by serial twofolddilutions analyzed by gel electrophoresis [27]), which is atleast a 1012-fold increase. According to the equation (1 + x)"= Y given by Saiki et al. (27) to determine amplificationefficiency (where x is the efficiency of the PCR, n is thenumber of cycles, and Y is the total yield), this amplificationprocedure has an average efficiency of approximately 89%per cycle. This is a considerable improvement over pastprocedures with efficiencies ranging from 35 to 66% (3, 17,23, 24, 30). Additionally, this 1012-fold amplification is higherthan any we have seen reported.

Conclusions. The ability of this procedure to amplify all C.trachomatis serovars supports restriction enzyme analysisand DNA hybridization data that suggest that the plasmids ofthese serovars are closely related (22). The amplification ofthe guinea pig inclusion conjunctivitis agent was unexpectedbecause of the finding that the plasmids of C. trachomatisand C. psittaci do not show strong DNA hybridization witheach other (12). However, there is a great deal of variationamong the plasmids of C. psittaci (16). Our results mayindicate that the plasmid of this particular C. psittaci strain ismore closely related to the C. trachomatis plasmid thanothers are.The value of this particular technique is greatly improved

by the ability to use whole cells added directly to the PCRmixture without the need to lyse the cells in a prior step or bychemical methods of DNA purification. Chlamydial organ-isms are known to contain no peptidoglycan but apparentlyare held together by disulfide linkages of proteins in theirouter membranes (19). The reduction of these proteins is notnecessary for the detection of the bacterial DNA by PCRsince only a few bacterial cells in a reaction without reducingagent are sufficient for detection to occur.

The development of a two-step PCR allows extremely highamplifications with very low background, making possiblethe direct amplification and visualization of DNA whenstarting with only attogram quantities. Because of the elim-ination of an entire step of the PCR and the abbreviation ofothers, this method also reduces the total time of the PCR onthe DNA template present in one bacterial cell to just 51 min(about one-third the average published time). Even whentime for gel electrophoresis is added, the entire detectionprocedure requires less than 1.5 h.

ACKNOWLEDGMENTS

We thank B. E. Batteiger for providing intact chlamydial cells; A.Roman, R. C. Bockrath, and D. Nathans for helpful suggestions;and J. A. Tischfield for the use of his equipment.

This work was supported by the Phi Beta Psi Sorority.

ADDENDUM IN PROOF

K. Kim and B. Kwon of our institution have recentlyfound that they were unable to amplify a specific DNA targetwith any of three sets of primers at any Mg2+ concentrationin a widely used buffer system (11). However, when theyadded 10% dimethyl sulfoxide to the reaction mixture, theresults were excellent. It may be important that this targetDNA had a G+C content of 65 mol%.

LITERATURE CITED1. Batteiger, B. E., and R. B. Jones. 1987. Chlamydial infections.

Infect. Dis. Clin. N. Am. 1:55-81.2. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction

procedure for screening recombinant plasmid DNA. NucleicAcids Res. 7:1513-1523.

3. Chehab, F. F., M. Doherty, S. Cai, Y. W. Kan, S. Cooper, andE. M. Rubin. 1987. Detection of sickle cell anaemia and thalas-saemias. Nature (London) 329:293-294.

4. Chien, A., D. B. Edgar, and J. M. Trela. 1976. Deoxyribonucleicacid polymerase from the extreme thermophile Thermus aquat-icus. J. Bacteriol. 127:1550-1557.

5. Danna, K. J., and D. Nathans. 1972. Bidirectional replication ofsimian virus 40 DNA. Proc. Natl. Acad. Sci. USA 69:3097-3100.

6. Dean, D., C. R. Pant, and P. O'Hanley. 1989. Improved sensi-tivity of a modified polymerase chain reaction amplified DNAprobe in comparison with serial tissue culture passage fordetection of Chlamydia trachomatis in conjunctival specimensfrom Nepal. Virology 12:133-137.

7. Dutilh, B., C. Bebear, P. Rodriguez, A. Vekris, J. Bonnet, and

I- N m

VOL. 56, 1990

on May 11, 2021 by guest

http://aem.asm

.org/D

ownloaded from

APPL. ENVIRON. MICROBIOL.

M. Garret. 1989. Specific amplification of a DNA sequencecommon to all Chlamydia trachomatis serovars using the poly-merase chain reaction. Res. Microbiol. 140:7-16.

8. Griffais, R., and M. Thibon. 1989. Detection of Chlamydiatrachomatis by the polymerase chain reaction. Res. Microbiol.140:139-141.

9. Hyypia, T., A. Jaalava, S. H. Larsen, P. Terho, and V. Hukka-nen. 1985. Detection of Chlamydia trachomatis in clinicalspecimens by nucleic acid spot hybridization. J. Gen. Micro-biol. 131:975-978.

10. Hyypia, T., S. H. Larsen, T. Stahlberg, and P. Terho. 1984.Analysis and detection of chlamydial DNA. J. Gen. Microbiol.130:3159-3164.

11. Innis, M. A., and D. H. Gelfand. 1990. Optimization of PCRs, p.3-12. In M. A. Innis, D. H. Gelfand, J. J. Sninsky, and T. J.White (ed.), PCR protocols. Academic Press, Inc., New York.

12. Joseph, T., F. E. Nano, C. F. Garon, and H. D. Caldwell. 1986.Molecular characterization of Chlamydia trachomatis and Chla-mydia psittaci plasmids. Infect. Immun. 51:699-703.

13. Kogan, S. C., M. Doherty, and J. Gitschier. 1987. An improvedmethod for prenatal diagnosis of genetic diseases by analysis ofamplified DNA sequences. N. Engl. J. Med. 317:985-990.

14. Leary, J. J., D. J. Brigati, and D. C. Ward. 1983. Rapid andsensitive colorimetric method for visualizing biotin-labeledDNA probes hybridized to DNA or RNA immobilized onnitrocellulose: bio-blots. Proc. Natl. Acad. Sci. USA 80:4045-4049.

15. Liang, W., and J. P. Johnson. 1988. Rapid plasmid insertamplification with polymerase chain reaction. Nucleic AcidsRes. 16:3579.

16. McClenaghan, M., A. J. Herring, and I. D. Aitkin. 1984.Composition of Chlamydia psittaci isolates by DNA restrictionenzyme analysis. Infect. Immun. 45:384-389.

17. Mullis, K. B. 1989. The polymerase chain reaction: why itworks, p. 237-243. In H. A. Erlich, R. Gibbs, and H. H.Kazazian (ed.), Polymerase chain reaction. Cold Spring HarborPress, Cold Spring Harbor, N.Y.

18. Mullis, K. B., and F. A. Faloona. 1987. Specific synthesis ofDNA in vitro via a polymerase-catalyzed chain reaction. Meth-ods Enzymol. 155:335-350.

19. Newhall, W. J., V, and R. B. Jones. 1983. Disulfide-linkedoligomers of the major outer membrane protein of Chlamydiae.J. Bacteriol. 154:998-1001.

20. Palmer, L., and S. Falkow. 1986. A common plasmid of Chla-mydia trachomatis. Plasmid 16:52-62.

21. Pao, C. C., S.-S. Lin, T.-E. Yang, Y.-K. Soong, P.-S. Lee, andJ.-Y. Lin. 1987. Deoxyribonucleic acid hybridization analysisfor the detection of urogenital Chlamydia trachomatis infectionsin women. Am. J. Obstet. Gynecol. 156:195-199.

22. Peterson, E. M., and L. M. de la Maza. 1988. Restrictionendonuclease analysis of DNA from Chlamydia trachomatisbiovars. J. Clin. Microbiol. 26:625-629.

23. Ou, C.-Y., S. Kwok, S. W. Mitchell, D. H. Mack, J. J. Sninsky,J. W. Krebs, P. Feorino, D. Warfield, and G. Schochetman.1988. DNA amplification for direct detection of HIV-1 in DNAof peripheral mononuclear cells. Science 239:295-297.

24. Powell, L. M., S. C. Wallis, R. J. Pease, Y. H. Edwards, T. J.Knott, and J. Scott. 1987. A novel form of tissue-specific RNAprocessing produces apolipoprotein B48 in intestine. Cell 50:831-840.

25. Saiki, R. K., T. L. Bugawan, G. T. Horn, K. B. Mullis, andH. A. Erlich. 1986. Analysis of enzymatically amplified B-globinand HLA-DQa DNA with allele-specific oligonucleotide probes.Nature (London) 324:163-166.

26. Saiki, R. K., D. H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi,G. T. Horn, K. B. Mullis, and H. A. Erlich. 1988. Primer-directed enzymatic amplification of DNA with a thermostableDNA polymerase. Science 239:487-491.

27. Saiki, R. K., S. Scharf, F. Faloona, K. B. Mullis, G. T. Horn,H. A. Erlich, and N. Arnheim. 1985. Enzymatic amplification ofB-globin genomic sequences and restriction site analysis fordiagnoses of sickle cell anemia. Science 230:1350-1354.

28. Wahl, G. M., S. L. Berger, and A. R. Kimmel. 1987. Molecularhybridization of immobilized nucleic acids: theoretical conceptsand practical considerations. Methods Enzymol. 152:399-407.

29. Wang, S. P., and J. T. Grayston. 1970. Immunological relation-ship between genital tract TRIC, lymphogranuloma venereum,and related organisms in the new microtitre indirect immuno-fluorescence test. Am. J. Ophthalmol. 70:367-374.

30. Wong, C., C. E. Dowling, R. K. Saiki, R. G. Higuchi, H. A.Erlich, and H. H. Kazazian. 1987. Characterization of B-thalas-saemia mutations using direct genomic sequencing of amplifiedsingle copy DNA. Nature (London) 330:384-386.

31. Zehang, H., R. Scholl, J. Browse, and C. Somerville. 1988.Double stranded DNA sequencing as a choice for DNA se-quencing. Nucleic Acids Res. 16:1220.

2498 WELCH ET AL.

on May 11, 2021 by guest

http://aem.asm

.org/D

ownloaded from