INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The diagnosis of syphilis depends on clinical findings, examination of lesions for treponemes, and/or serologic tests. Serologic tests are divided into nontreponemal and treponemal tests. Nontreponemal tests are useful for screening, while treponemal tests are used as confirmatory tests. In practice, two or three assays are performed in series for the serologic diagnosis of syphilis (22, 28). First, sera are screened in a microflocculation assay, such as the Venereal Disease Research Laboratory test (VDRL), using nontreponemal antigens (25). Reactive sera are then tested for antibodies specific for Treponema pallidum antigens by using the fluorescent treponemal antibody absorption assay (FTA-ABS) or the microhemagglutination assay (MHA-TP) (12, 22, 28). Since the mid-1980s, the Western blot assay (WB), set up with boiled sodium dodecyl sulfate extracts of T. pallidum, has been introduced as an alternative to the other confirmatory methods (5, 13, 15). In addition, among the treponemal tests, several enzyme immunoassays have been described and evaluated, showing sensitivities and specificities similar to those of FTA-ABS and MHA-TP (9, 11, 23, 37).

Currently, the T. pallidum immobilization test (TPI), the first treponemal antibody test developed (by Nelson and Mayer in 1949 [26]), is limited to mostly research purposes, since it is complex and technically difficult. The test is based on the ability of the patient's antibody to bind to the surface of T. pallidum (2) and to immobilize living spirochetes in the presence of complement, as observed by dark-field microscopy. TPI is accepted as a specific test for syphilis, though it is less sensitive than the newer treponemal tests (30).

Previous observations from our laboratory showed the high specificity of surface immunofluorescence assay (SIFA) performed with live Borrelia burgdorferi spirochetes for the detection of serum-specific antibodies against surface antigens of the Lyme disease agent (7). In this study, we investigated the sensitivity and specificity of SIFA both for the detection of specific antibodies against T. pallidum and for the serologic diagnosis of syphilis.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Study groups. One hundred patients (74 males and 26 females) between 21 and 68 years of age (mean age, 37.2 years) suffering from syphilis were studied. The diagnosis, based on clinical and laboratory data, was early syphilis for 46 patients (i.e., primary syphilis for 26 patients and secondary syphilis for 20 patients), early latent syphilis for 50 patients, and late syphilis for 4 patients. Sixteen patients with primary syphilis and three with secondary syphilis were human immunodeficiency virus (HIV)-positive.

Source of T. pallidum. Treponema pallidum subsp. pallidum (Nichols strain), originally purchased from the Statens Serum Institut, Copenhagen, Denmark, was maintained in adult male New Zealand White rabbits by testicular passage (24). Infected rabbits were individually housed, maintained at 16 to 18°C, and given antibiotic-free food and water ad libitum. Each treponemal suspension was prepared from infected rabbit testicles from 10 to 14 days after inoculation (24).

VDRL. A commercial kit was used by following the manufacturer's instructions (Behring AG, Marburgan der Lahn, Germany). The results were read under light microscopy at a ×100 magnification.

FTA-ABS. Serum samples were tested by a fluorescent T. pallidum-specific assay (Trepo Spot IF, BioMerieux SA, Marcy l'Etoile, France) by using microscope slides coated with T. pallidum (Nichols strain) whole antigen. Sera were preliminarily screened by following the plus (+) score system proposed by the manufacturer. Afterwards, all the samples scored as 1+ were retested by using serial twofold dilutions starting from 1/5 up to 1/40. Since the 1+ sera were positive at a dilution of <1:20, they were considered negative according to the previously reported data of Larsen and coworkers (21). The specimens scoring above the 1+ level were retested by diluting samples from 1:10 up to 1:640. A reaction was considered positive when at least 80% of the spirochetes gave a brilliant green fluorescence at a serum dilution of 1:20.

MHA-TP. Quantitative MHA-TP testing (MHA-TP, Fujirebio, Tokyo, Japan) was performed in accordance with the manufacturer's instructions, which had an established cutoff of a 1:80 dilution.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and WB. Treponemal suspensions obtained from infected rabbit testicles were diluted in sample buffer to 10⁹ organisms per ml, as determined by dark-field microscopy examination at a magnification of ×400. The separation of T. pallidum polypeptides was performed with a Laemmli buffer system (19) by using a 12% acrylamide gel as described elsewhere (6). The WB procedure of Towbin et al. (35) was performed as described elsewhere (6). After electrophoretic transfer, the blots were incubated for 12 h at room temperature with sera diluted 1:100 in phosphate-buffered saline containing 0.05% (vol/vol) Tween 20. Antigen-antibody complexes were detected with peroxidase-labeled rabbit antiserum to total human immunoglobulin (Ig) (Dako, Copenhagen, Denmark) diluted 1:500 in phosphate-buffered saline and with 4-chloro-1-naphthol (Bio-Rad, Hercules, Calif.) as the enzyme substrate.

TPI. TPI was performed by following the guidelines of the Centers for Disease Control and Prevention (36). Briefly, treponemes were extracted from rabbit testis in basal medium (36) under anaerobic conditions and were incubated with serum samples (with or without added complement) in anaerobic jars at 35°C for 16 to 18 h. All sera were tested in duplicate to ensure the validity of the procedure. A test was considered negative if the difference between the average number of motile treponemes, determined by dark-field microscopy (magnification, ×400), of the two "test" tubes and the "control" tube (i.e., the tube without complement added) was 20% or less. A test was considered positive if the difference between the average count of motile spirochetes in the two test tubes and the control tube was 50% or greater. In order to be validated, in the control tube at least 70% active motile treponemes had to be read. Specimens were always tested in a blind fashion.

SIFA. Serum samples were tested by SIFA as previously reported (7, 32), with slight modifications. Sera were inactivated at 56°C for 1 h, diluted 1:20 in a suspension of 100% motile treponemes extracted from infected rabbit testicles (10⁸ bacteria/ml) in the medium described by Stamm and Bassford (34), and incubated under an atmosphere of 95% N₂-5% CO₂ at 37°C for 1 h. After the incubation period, the viability of treponemes was checked again by dark-field microscopy. In order to validate the test, at least 99% of treponemes had to be motile. The suspension was then pelleted by centrifugation at 10,000 × g for 20 min, and washed three times in phosphate-buffered saline. The treponemes were diluted to a final concentration of 10⁷ bacteria/ml, spotted (10 µl) onto microscope slides, air dried, and fixed with cold acetone for 10 min. Igs bound to T. pallidum surface-exposed antigens were detected by using fluorescein-conjugated rabbit anti-human total Ig (Dako), diluted 1:40 in phosphate-buffered saline. The results were read under UV light microscopy at a magnification of ×400.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Antibodies against T. pallidum surface-exposed antigens were easily detectable in sera from syphilitic patients by SIFA. Positive sera induced a bright fluorescence of treponemes that appeared slightly more slender than bacteria stained by FTA-ABS (Fig. 1).

View larger version (141K): [in this window] [in a new window]   FIG. 1.   Photomicrograph of T. pallidum stained by SIFA. Living spirochetes were incubated with T. pallidum-positive (A) and -negative (B) human serum samples. Magnification, ×320. (C and D) T. pallidum stained by SIFA and by FTA-ABS, respectively, at a higher magnification (×800). The spirochetes stained by SIFA appear slightly more slender than those stained by FTA-ABS.

The results obtained in the various stages of syphilitic infection by SIFA are reported in Table 1 and compared with the results obtained by the other serologic methods. SIFA, WB (Fig. 2), and VDRL were the most sensitive methods in the early stage of syphilis, since 45 out of 46 serum samples were positive by these methods; the serum sample negative by these tests was from a patient showing a chancre lesion from which treponemes were identified by dark-field microscopy. It is interesting that 6 out of 16 HIV-infected patients with primary syphilis and 1 out of 3 HIV-infected patients with secondary syphilis were positive by SIFA, WB, and VDRL but negative by FTA-ABS and MHA-TP. All (n = 50) serum samples from patients with early latent syphilis were positive by SIFA, WB, FTA-ABS, and VDRL, whereas 92% (46 of 50) and 72% (36 of 50) of the sera were positive by MHA-TP and TPI, respectively. Late-syphilis sera were positive by all the tests. Tables 2 to 5 report the comparison of results obtained by SIFA with results obtained by the other treponemal tests (WB, FTA-ABS, MHA-TP, and TPI) in testing specimens from syphilis patients, healthy donors, Lyme disease patients, and leptospirosis patients and in testing BFP samples. The sensitivity and specificity of SIFA and of the other methods were as follows. SIFA and WB were 99% sensitive (99 of 100 positive specimens) and specific (140 of 140 negative specimens, including 50 BFP reactors, 50 healthy blood donors' samples, 20 samples of patients with Lyme disease, and 20 samples of patients suffering from leptospirosis). FTA-ABS showed a sensitivity and a specificity of 90 and 89%, respectively (90 of 100 positive and 125 of 140 negative specimens). MHA-TP showed a sensitivity of 84% (84 of 100 positive specimens) and a specificity of 98.5% (138 of 140 negative specimens). Finally, TPI showed a sensitivity of 52% (52 of 100 positive specimens) and a specificity of 100% (140 of 140 negative specimens).

View larger version (27K): [in this window] [in a new window]   FIG. 2.   WB analysis of human sera with T. pallidum whole-cell preparation. (A) Sample from a positive serum pool (lane 1) and two serum samples from patients with early latent syphilis (lanes 2 and 3). These samples clearly recognized four T. pallidum bands of 15.5, 17.0, 44.5, and 47.0 kDa. These bands are considered diagnostic for T. pallidum infection when at least three of them are detected (5, 27). (B) Lanes 1 and 2 were probed with two sera defined as BFP reactors since they were positive by VDRL but negative by MHA-TP and TPI. Also, WB was negative since the sera recognized only two of the four above-mentioned T. pallidum bands (the 47.0- and 44.5-kDa bands). The apparent molecular masses of individual T. pallidum proteins are shown in the middle.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The reliability of immunological reactions between antibodies and surface-exposed antigens of the spirochete B. burgdorferi that are caused by using living organisms and SIFA has been previously shown by members of our group (7) and subsequently confirmed by Cox et al. (8). In this study, we compared SIFA for the detection of antibodies against T. pallidum with other treponemal serologic tests. The sensitivity of SIFA and WB was 99%; in contrast, FTA-ABS and MHA-TP were less sensitive, with values of 90 and 84%, respectively. In fact, FTA-ABS and MHA-TP failed to recognize antitreponemal antibodies in 7 out of 19 specimens from HIV-infected patients suffering from primary or secondary syphilis, in contrast with SIFA, WB, and VDRL. The reliability of syphilis serology in patients with HIV infection has been questioned (10, 17, 33), and the need for a good test for this group of patients is well known, because of the false-negative results that can be obtained by using the classic treponemal serologic methods. TPI was the least sensitive method, showing a sensitivity of 52%, in agreement with previous reported data (30). The specificity of SIFA was analyzed by testing sera from several control groups. None of these sera was found to be reactive by SIFA (specificity, 100%). Similar results were obtained by WB and TPI. MHA-TP showed high specificity (98.5%), whereas FTA-ABS was less specific (89%), since several BFP samples and sera from patients with Lyme disease gave false-positive reactions. These data are in agreement with those reported by other authors (3, 4, 14, 18, 20).

When the SIFA-positive sera were tested by WB, all of them clearly recognized at least three out of the four T. pallidum bands with apparent molecular masses of 15.5, 17.0, 44.5, and 47.0 kDa, which are considered diagnostic for T. pallidum infection (5, 27). In addition, the sera recognized several other variable bands. It is known that none of the four T. pallidum bands considered diagnostic for T. pallidum infection is surface exposed (1, 27). Therefore, the antibodies reactive in SIFA were not those involved in the recognition of these T. pallidum antigens by WB. On the other hand, it is unlikely that the rare outer membrane proteins of T. pallidum described by Blanco et al. (1) and by Radolf et al. (29) could be reliably identified by WB, at least under our experimental conditions, due to their scanty representation on the bacterial surface (1, 29). In fact, it has been shown that T. pallidum surface proteins can be reliably detected by WB only by using enriched outer membrane protein preparations (1, 29). Notwithstanding this, the specificity of SIFA was demonstrated in our study by comparing the T. pallidum SIFA results with T. pallidum WB results obtained in detecting other diagnostic, albeit not surface-exposed, T. pallidum proteins.

In conclusion, the results obtained in the present study demonstrate that antibodies reactive with the surface-exposed antigens of live T. pallidum organisms are highly specific for syphilitic infection, showing no equivocal reactivities with the various control sera used. This observation confirms previous data regarding the specificity of antibodies reactive with the surface antigens of the spirochete B. burgdorferi in patients suffering from Lyme disease (7). In contrast with B. burgdorferi SIFA, which proved sensitive only for late Lyme disease (7), the T. pallidum SIFA was highly sensitive for both early and late syphilis. These results therefore suggest the possibility of using the T. pallidum SIFA as a confirmatory test in the serologic diagnosis of syphilis.