INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Mycoplasma capricolum subspecies capripneumoniae is the causative agent of contagious caprine pleuropneumonia (CCPP), a significant disease of goats in Africa, the Middle East, and western Asia, with mortality rates being up to 80% in susceptible herds. While clinical disease has so far been reported in 38 countries, only 11 countries have isolated the causative organism, principally because M. capricolum subsp. capripneumoniae is difficult to culture (26). M. capricolum subsp. capripneumoniae is a mycoplasma of the "M. mycoides cluster," a taxonomic grouping of six closely related mycoplasmas which are all pathogenic in ruminants. Serological cross-reactivities between members of the M. mycoides cluster (3), together with similarities in the clinical diseases that they cause, all play a part in hindering accurate diagnosis of CCPP (19).

A number of serological tests currently exist, but most are difficult to use in situ, lack specificity, or require resources unavailable in many countries affected by the disease. These include the complement fixation test (CFT; the prescribed test for international trade [17, 19]), passive hemagglutination (17), competitive enzyme-linked immunosorbent assay (ELISA) (27), and the latex agglutination test (LAT; with beads coated with M. capricolum subsp. capripneumoniae capsular polysaccharide [CPS]) (20). Antigen detection tests include immunoblotting (7) and sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis with mycoplasmas concentrated from pleural fluid (28). In addition, a PCR test for the detection of M. capricolum subsp. capripneumoniae has been developed (8). However, all these tests exhibit certain limitations in either specificity, sensitivity, ease of application, cost, or the requirement for specialist equipment or expertise.

The antibody-detection LAT with M. capricolum subsp. capripneumoniae-derived CPS-coated beads is now routinely applied in Kenya (19). While the test is inexpensive and simple to perform, animals in the early or acute stages of the disease (prior to the appearance of circulating antibodies or when high levels of circulating antigen may eclipse the immune response [17, 29]) may be misdiagnosed as negative for CCPP. In this respect, the CFT exhibits clear limitations: in published reports, CCPP in between 80 and 100% of acute-phase animals experimentally infected with M. capricolum subsp. capripneumoniae was not detected by the CFT (10, 17, 19), demonstrating the potential limitations of antibody detection as a sole diagnostic technique. Furthermore, significant complement fixation titers in infected goats were not observed until 21 days postinfection (17). This could be significant in the field, since the incubation period for CCPP can be as little as 6 days, with death in acute cases following only 2 days later (16). Thus, at an early stage in a field outbreak false-negative results based upon serological diagnosis are a real possibility. In contrast, positive reactions may also occur with circulating antibodies long after an M. capricolum subsp. capripneumoniae infection has been cleared, thus giving an inaccurate diagnosis of the likely infective state of an animal.

The development of a LAT which detects circulating M. capricolum subsp. capripneumoniae antigen should allow the diagnosis of a current infection, since antigen has been detected within 3 to 9 days of the onset of clinical signs following infection with M. capricolum subsp. capripneumoniae (23). Used in conjunction with the current antibody-detection LAT, this should allow effective field screening at all stages of infection. LATs are very rapid, are relatively inexpensive, and can be used in situ by untrained personnel. The aim of the work described in this paper was to investigate the potential of an LAT with anti-M. capricolum subsp. capripneumoniae immunoglobulin-coated latex microspheres to diagnose clinical and subclinical M. capricolum subsp. capripneumoniae infection in goats.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Mycoplasma strains and growth conditions. The mycoplasma strains used in this study are shown in Table 1. All strains were grown in Mycoplasma Experience (ME) broth and agar medium (Mycoplasma Experience, Reigate, United Kingdom). To measure the titer of broth cultures used in agglutination reactions, serial dilutions were made in fresh medium to measure color-changing units or, alternatively, were plated onto solid medium followed by colony counting (18). Sodium azide (0.05%) was added to broth cultures at the time of sampling, and cultures were subsequently stored at 4°C prior to agglutination reactions.

Antisera. M. capricolum subsp. capripneumoniae polyclonal antiserum (F38) was raised in rabbits following vaccination with whole sonicated F38 type strain, while a membrane fraction antiserum (R37) was raised against a Triton Tx114-phase (30) fraction of F38. Both antisera have previously been described in detail (13). The immunoglobulin G (IgG) fraction was purified on a protein A-Sepharose column by a standard methodology (5) and was eluted at a final concentration of 14.5 mg/ml in phosphate-buffered saline (PBS). A negative control rabbit IgG fraction (R46) was similarly purified at a final concentration of 7.5 mg/ml.

Purification of CPS. CPS was purified from spent medium by a modification (14) of a previously published method (21). Mycoplasmas were removed by centrifugation, the pH of the cleared supernatant was adjusted to 5.0 with glacial acetic acid, and then the cleared supernatant was heated to 100°C for 30 min, adjusted back to pH 7.5 with NaOH, filtered through Whatmann 3MM paper, and then precipitated with 2 volumes of ethanol. The precipitate was resuspended in 0.1 volume of distilled H₂O, and RNase A and DNase I and II (Sigma) were added to a final concentration of 1 µg/ml (each). This mixture was left for 2 h at 37°C, and then SDS was added to 0.5%, followed by the addition of proteinase K (Sigma) to a final concentration of 1 mg/ml. The mixture was left for 24 h at 45°C and was then dialyzed against running tap water for 3 days with a 15-kDa molecular mass cutoff membrane. The dialysate was then extracted once with phenol-chloroform (50:50), extracted once with chloroform, ethanol precipitated, rinsed in 80% ethanol, and dialyzed overnight against several changes of distilled H₂O. CPS was further purified by gel filtration twice through a Sephacryl S-300 column, where it was eluted early, immediately after the void volume. No protein could be detected in the purified CPS by a variety of assays (Bradford assay, Lowry assay, and following silver staining of SDS-polyacrylamide gels). The amount of carbohydrate present was measured by the phenol-sulfuric acid method (4).

Measurement of CPS antibody titer. ELISA analysis was used to estimate the M. capricolum subsp. capripneumoniae CPS antibody titer in the IgG fraction used to coat the latex particles (to allow standardization between different antibody batches). Microtiter plates (Greiner) were coated with a saturating amount of antigen (0.1 ml of antigen solution at a concentration of 10 µg/ml in PBS left overnight at 4°C). Antigen was either purified CPS or whole sonicated M. capricolum subsp. capripneumoniae that was pelleted out of the growth medium, washed three times in PBS (supplemented with 5% [wt/vol] glucose), and then resuspended in PBS prior to sonication (three 20-s bursts). The plates were then blocked in a 5% solution of dry skim milk in PBS plus 0.05% Tween 20 (PBST; Sigma) for 2 h at room temperature. Primary antibody (either anti-M. capricolum subsp. capripneumoniae IgG or a negative control IgG) was standardized at an initial concentration of 1 mg/ml in PBS and was then diluted 1:100 in blocking solution. Twofold serial dilutions in blocking buffer were then made across the plate (1:100 down to 1:102,400). Incubations were done in a final volume of 0.1 ml. Appropriate negative primary and secondary controls were included. Sodium azide was added to a final concentration of 0.05% (wt/vol), and the plates were incubated overnight at room temperature. Four 5-min washes in PBST were performed, and then secondary antibody (pig anti-rabbit horseradish peroxidase conjugate; DAKO A/S, Glostrup, Denmark) was added at a 1:1,000 dilution in blocking solution. After incubation at room temperature for 1 h, the plates were washed four times in PBST and developed with o-phenylenediamine dichloride (Sigma). The optical density was measured at 492 nm with an IEMS plate reader (Labsystems).

Coating of microspheres. A 10% solution of 0.8-µm-diameter polystyrene latex microspheres (LB8; Sigma) was used in the binding reaction. Prior to use, the bead suspension was vigorously vortexed to ensure even distribution of the beads and to break up any large particles. To 0.15 ml of beads (15 mg) was added 2.5 mg of polyclonal anti-M. capricolum subsp. capripneumoniae IgG, and the volume was made up to 1 ml with PBS. (For experiments in which the ratio of antibody to beads was altered, the volume of beads and the final reaction volume were kept constant, while the amount of added antibody was varied.) Negative control beads were produced by an identical protocol but with IgG purified from a nonimmunized rabbit. The mixture was incubated for 120 min at room temperature on an end-to-end shaker and was then centrifuged at 15,000 rpm (20,000 × g) for 3 min in a microcentrifuge. The supernatant was removed and was stored at 4°C for ELISA analysis, and the beads were resuspended in 1 ml of PBST. The mixture was centrifuged as described above, the supernatant was removed, and the beads were resuspended in a final volume of 1.0 ml of PBS supplemented with 0.05% sodium azide as a preservative. Resuspension of the beads was normally achieved by vortexing, although a brief sonication (5 s) was found to be useful in difficult cases. For preabsorption of polyclonal anti-M. capricolum subsp. capripneumoniae IgG with CPS, 10 µg of purified CPS was added to 100 µg of IgG in a reaction volume of 0.1 ml of PBS. A positive control tube contained only IgG. The mixture was left to incubate at room temperature for 30 min, followed by centrifugation at 15,000 rpm (20,000 × g) in a microcentrifuge. The supernatant was added to 15 µl of beads (15 mg) and treated as described above, prior to resuspension in a final volume of 0.1 ml of PBS. The beads were then tested in agglutination reactions.

Measurement of unbound antibody. ELISA analysis was used to measure the amount of unbound antibody following microsphere coating and thus estimate the binding capacity of the latex microspheres for IgG. A standard curve was produced by diluting anti-M. capricolum subsp. capripneumoniae IgG to concentrations of 100, 50, 10, 5, and 1 µg/ml in PBST, and 0.1 ml of each aliquot was added (in triplicate) to 96-well microtiter plates. Similarly, aliquots of the bead supernatant were diluted in PBS in the range of 1:10 to 1:100 and were added in triplicate to the wells. Appropriate controls (including primary antibody only and secondary antibody only) were included on the plates. The plates were left overnight at 4°C, washed twice in PBST, washed once in distilled H₂O, and then blocked in a 5% solution of dry skim milk in PBST for 3 h at 37°C. The plates were then washed twice in PBST and once in distilled H₂O, and secondary antibody (goat anti-rabbit horseradish peroxidase conjugate; DAKO A/S) was added at a 1:2,000 dilution in PBST supplemented with 5% dry skim milk. After incubation at room temperature for 1 h, the plates were washed in PBST and developed with o-phenylenediamine dichloride (Sigma). The optical density was measured at 492 nm with an IEMS plate reader (Labsystems). The standard curve was found to be linear in the range of 0.1 to 10 µg of IgG per well, and dilutions of bead supernatant which gave readings within this range were used to calculate the amount of IgG remaining in the supernatant compared to that originally added to the latex microspheres.

LATs. LATs were performed with (i) purified CPS diluted in negative goat serum, medium, or PBS; (ii) field and experimentally infected goat serum samples; and (iii) mycoplasma suspensions grown in ME liquid medium, with serial dilutions made in the same medium. Agglutination reactions were performed in triplicate by the same procedure: 30 µl of test solution was aliquoted onto a glass microscope slide, and 5 µl of the bead suspension (containing 75 µg of beads coated with 12.5 µg of anti-M. capricolum subsp. capripneumoniae IgG) was added with a pipette. The beads were thoroughly mixed, and the slide was incubated for 1 to 10 min on a rocking shaker with gentle agitation. The key to good agglutination reactions appeared to be side-to-side movement of the reaction solution across the slide. Beads coated with the negative IgG were mixed with the positive test solution as a negative control, while beads coated with anti-M. capricolum subsp. capripneumoniae IgG were also mixed with antigen-free test solution (negative goat serum or ME growth medium) to test for autoagglutination. Agglutination (when seen against a dark background) was most obvious when the reaction mixture was still wet, although it could still be clearly ascertained when it was dry in the case of strong positive reactions. In borderline cases, increasing the reaction time by adding distilled H₂O (10 to 15 µl) after 10 to 15 min or by conducting the reaction in a humid atmosphere to prevent desiccation of the mixture was found to be beneficial. It is also possible to perform agglutination reactions in standard 96-well microtiter plates (Greiner). In this case, 6 µl of beads was added to 60 µl of test solution in a well, the plate was sealed with tape or an adhesive plastic cover (BIS Ltd., Lancs, United Kingdom), and the plate was subjected to gentle agitation for 15 to 90 min prior to visualization under a low-power microscope or magnifying glass.

CFT. The CFT was performed according to the Office International des Epizooties manual of standards (19) with antigen and reagents supplied by CIRAD-EMVT.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Definition of agglutinating activity. Following the mixing of anti-M. capricolum subsp. capripneumoniae coated latex microspheres with M. capricolum subsp. capripneumoniae antigen-positive solution, an agglutination reaction occurred, the rapidity and extent of which depended upon the amount of antigen present. For the purposes of this study a four-point scale (exhibited pictorially in Fig. 1A) was adopted and was defined as follows: +++, heavy flocculent precipitates forming within 1 min (clear background); ++, heavy flocculent precipitates taking 1 to 5 min to form (clear background); +, light flocculent precipitate against mostly clear background (5 to 15 min), +/, light flocculent precipitate against cloudy homogeneous background after extended incubation (15 to 30 min); , no precipitate, cloudy homogeneous background.

View larger version (52K): [in this window] [in a new window]   FIG. 1.   Visual definition of agglutination reactions. (A) Agglutination reactions photographed while still wet with CPS in solution. Agglutination activity is defined from +++ (uppermost reaction) to  (lower reaction). (B) Agglutination reactions photographed after drying by using pure CPS solution in PBS serially diluted from 120 to 0 ng of CPS ml1. The clarity of the agglutination reaction is reduced upon desiccation, while at 1.9 and 0.9 ng ml1 reactions that were still wet were positive, although this is not apparent in the figure.

Binding capacity of latex microspheres for anti-M. capricolum subsp. capripneumoniae polyclonal IgG. Variable amounts of IgG (0.01 to 1.0 mg) were incubated with 1.5 mg of latex microspheres in a reaction volume of 0.15 ml. Following centrifugation, the cleared supernatant was tested for the presence of IgG by ELISA analysis. No significant amounts of IgG remained in the supernatant in the range of 0.01 to 0.1 mg of added IgG per 1.5 mg of beads, suggesting that the binding capacity of the beads had not yet been reached. A small amount of unbound antibody (0.04 ± 0.01 mg) was observed following incubation of the latex beads with 0.5 mg of IgG, and a significant amount of free antibody (0.3 ± 0.004 mg) was observed when 1.0 mg of IgG was used in the binding reaction. The results suggest that the binding capacity of the beads is in the range of 0.46 to 0.7 mg of IgG per 1.5 mg of beads (0.33 to 0.5 mg of IgG per mg of beads).

Agglutinating activity of latex microspheres coated with different amounts of anti-M. capricolum subsp. capripneumoniae IgG. Beads coated with different amounts of anti-M. capricolum subsp. capripneumoniae IgG (0.001 to 1.0 mg of IgG per 1.5 mg of beads) were tested in agglutination reactions with M. capricolum subsp. capripneumoniae-positive growth medium (titer, 5 × 10⁸ CFU per ml). Beads coated with higher levels of IgG (0.1 to 1.0 mg of IgG per 1.5 mg of beads) gave consistently optimum agglutination reactions. Beads coated with a lower, nonsaturating level of IgG (0.01 to 0.1 mg) gave more variable results: only two of four preparations gave positive reactions by the LAT, even with high levels of M. capricolum subsp. capripneumoniae antigen in the test solution. For this reason a higher (but nonsaturating) level of anti-M. capricolum subsp. capripneumoniae IgG was used to coat the latex microspheres to produce the test for general usage (0.25 mg of IgG per 1.5 mg of beads in a final volume of 0.1 ml). Control IgG-coated beads (prepared with IgG purified from a nonimmunized rabbit) gave negative agglutination reactions with both positive and negative media, while M. capricolum subsp. capripneumoniae-specific IgG-coated beads gave negative agglutination reactions with negative media.

Effect of excess anti-M. capricolum subsp. capripneumoniaeIgG on agglutination activity. Since the LAT detects circulating M. capricolum subsp. capripneumoniae antigen, the presence of M. capricolum subsp. capripneumoniae antibodies in a serologically positive animal might reduce the sensitivity of the test by sequestering free antigen. The effect of excess anti-M. capricolum subsp. capripneumoniae IgG on the antigen-detection LAT was tested by the addition of 5 or 1 µl of anti-M. capricolum subsp. capripneumoniae IgG (concentration, 14.5 mg per ml) to M. capricolum subsp. capripneumoniae-positive growth medium (titer, 5 × 10⁸ CFU per ml) either (i) 10 min before mixing or (ii) immediately following mixing of the medium with 5 µl of coated microspheres in an LAT. Preabsorption of M. capricolum subsp. capripneumoniae antigen-positive medium with 5 µl of anti-M. capricolum subsp. capripneumoniae IgG completely removed agglutination activity, while simultaneous addition attenuated, but did not completely remove, agglutination activity. Preabsorption with 1 µl of anti-M. capricolum subsp. capripneumoniae IgG reduced agglutination activity, while simultaneous addition had no observable affect. Thus, very high levels of anti-M. capricolum subsp. capripneumoniae IgG in an infected goat might, in theory, affect the sensitivity of LAT, although in practice this did not appear to be a problem (see below).

Sensitivity of latex microspheres at detecting M. capricolum subsp. capripneumoniae in medium. M. capricolum subsp. capripneumoniae strain F38 was cultured in ME medium to a titer of 5 × 10⁸ CFU per ml (growth was stopped at the mid-logarithmic phase to prevent overgrowth and cell death, which could give an underestimate of the true amount of M. capricolum subsp. capripneumoniae antigen present). The culture was serially diluted in fresh medium and was tested in an LAT (Table 2). Clear agglutination activity was observed down to a dilution of 1:4,096, which is equivalent to 3.6 × 10³ mycoplasmas per agglutination reaction (0.03 ml) or a titer of 1.2 × 10⁵ mycoplasmas per ml. This is the same degree of sensitivity previously reported for an antigen-detection LAT directed at M. pneumoniae (15). Similar sensitivity levels were observed with all strains of M. capricolum subsp. capripneumoniae tested. When the medium was cleared of mycoplasmas prior to testing by LAT (by centrifugation at 15,000 × g for 30 min), no difference in sensitivity was observed, strongly suggesting that a soluble antigen rather than (or in addition to) the mycoplasmas themselves was responsible for the agglutinating activity. If the mycoplasma pellet was resuspended in fresh ME medium following centrifugation, a positive agglutination reaction was observed at an approximately 100-fold reduced sensitivity, presumably due to shedding of the CPS coat during resuspension (following a further round of centrifugation, this medium was still positive). No agglutination was observed with negative IgG-coated beads or anti-M. capricolum subsp. capripneumoniae IgG-coated beads placed in M. capricolum subsp. capripneumoniae-negative medium.

Effect of preabsorption with CPS on agglutinating activity. M. capricolum subsp. capripneumoniae has been reported to produce large quantities of extracellular CPS, which is shed into the growth medium (21). It seemed likely that the soluble antigen recognized by the test was CPS. Anti-M. capricolum subsp. capripneumoniae IgG was preabsorbed with purified CPS to remove CPS-specific antibodies and was then used to coat the beads, as described above. All agglutination activity was now lost from these beads, not only with purified CPS in solution (66 µg per ml) but also with M. capricolum subsp. capripneumoniae culture medium (titer, of 5 × 10⁸ CFU per ml) and previously LAT-positive serum samples from CCPP-positive animals. Nonpreabsorbed beads continued to give a positive reaction in these tests. To ensure that CPS had not simply displaced IgG from the latex particles, preabsorbed beads were incubated with goat anti-rabbit IgG. A strong agglutination reaction indicated that the beads were still coated with rabbit IgG. Similarly, incubation of the preabsorbed beads with rabbit anti-M. capricolum subsp. capripneumoniae IgG did not result in agglutination, as would be expected if the beads were coated with M. capricolum subsp. capripneumoniae CPS. These findings strongly suggest that CPS is the major or single antigen recognized by the test.

Sensitivity of latex microspheres at detecting pure CPS in solution. Twofold serial dilutions of M. capricolum subsp. capripneumoniae CPS in M. capricolum subsp. capripneumoniae-negative goat serum were made, and aliquots were incubated (in triplicate) with microspheres coated with different amounts of anti-M. capricolum subsp. capripneumoniae IgG. The results are shown in Table 3 and Fig. 1B. Beads coated with 0.25 mg of IgG per 1.5 mg of beads were slightly more sensitive than beads coated with smaller amounts and were able to reproducibly detect amounts as as low as 2 to 4 ng of CPS in a reaction volume of 0.03 ml (equivalent to a concentration of 60 to 120 ng of CPS per ml). This level of sensitivity was noted for CPS dissolved in M. capricolum subsp. capripneumoniae-negative goat serum, ME growth medium, and PBS. The reaction time in PBS was considerably shortened, however, and a higher level of background agglutination (exhibited as a grainy background) was noted. The maximum concentration of CPS tested by LAT was 0.5 mg/ml, which still gave a strong positive signal. This finding suggests that saturation of antibody-coated microspheres due to excessive CPS levels in infected goats is unlikely to occur.

Measurement of CPS-specific antibody titer in M. capricolum subsp. capripneumoniae IgG fraction. To enable reproducibility between different batches of antisera, the titer of the CPS-specific IgG used to coat the beads was measured by ELISA. Two ELISAs were used: one to measure the anti-CPS titer (with purified CPS used as the antigen) and the second to measure the overall anti-M. capricolum subsp. capripneumoniae titer (with whole sonicated M. capricolum subsp. capripneumoniae used as the antigen). The titer was expressed per milligram of IgG. Thus, the specific anti-CPS titer could be estimated and taken into account (if necessary) for different batches of serum.

View larger version (26K): [in this window] [in a new window]   FIG. 2.   ELISA data showing antibody titers of various sera against purified CPS antigen (A) and whole M. capricolum subsp. capripneumoniae antigen (B). Serum was initially diluted 1:100 and was then serially diluted twofold down to 1:102,400. The linear portion of each graph was used to estimate the point of intersection with the x axis (titer of F38-positive serum against whole antigen, 1:9,600; R37-positive serum against whole antigen, 1:2,400; titer of F38-positive serum against CPS, 1:12,800; titer of R37-positive serum against CPS, 1:76,800). OD 492, optical density at 492 nm.

Specificity of anti-M. capricolum subsp. capripneumoniae LAT. The LAT was tested against a variety of mycoplasmal species known to be evolutionarily closely related to M. capricolum subsp. capripneumoniae (including all members of the M. mycoides cluster) and mycoplasmas likely to be found as associated pathogens (such as M. ovipneumoniae) (Table 1). All mycoplasmal species were grown in ME liquid medium to the mid-logarithmic phase, with titers in the range of 5 × 10⁷ to 8 × 10⁸ per ml. The LAT was specific for M. capricolum subsp. capripneumoniae and bovine serogroup 7, with all strains testing positive. Similar sensitivity levels were noted for all the strains examined (dilutions in the range of 10² to 10³ were positive). No agglutination was noted for any other mycoplasmal species tested, even in undiluted growth medium. The specificity of the LAT exactly mirrors that of the CPS-specific monoclonal antibody WM-25 (2, 23), providing further evidence that CPS is the M. capricolum subsp. capripneumoniae antigen recognized by the LAT.

Testing of LAT with sera from animals experimentally infected with M. capricolum subsp. capripneumoniae. The LAT was tested over time with sera from animals experimentally infected with M. capricolum subsp. capripneumoniae (Table 4). Blood samples were not available at 7 days postinfection, but a positive reaction was observed in all three animals 14 days postinfection (8 to 11 days following the onset of clinical signs). The animal which exhibited the most intense clinical signs also exhibited the longest positive reaction by LAT (goat 1855, which exhibited a persistent cough from days 5 to 22 postinfection). All three animals exhibited a significant humoral immune response (IgG and IgM) against M. capricolum subsp. capripneumoniae from week 2 onward (by ELISA and immunoblotting) (data not shown). It is unclear why the signal from the antigen-detection LAT increased in intensity at 8 weeks postinfection for animals 1855 and 1864, but it may be related to the fact that the relative levels of anti-CPS IgG and IgM affected test sensitivity.

Correlation with CFT. Experimentally infected animals 1855, 1864, and 1884 were all negative for CCPP by CFT, although they were positive by a variety of other criteria (immunoblotting, ELISA, clinical signs [12], and antibody-detection LAT). This finding suggested that CFT may be less sensitive than the other methods of diagnosis, particularly in the case of low-level infections, such as those observed for the experimentally infected goats. Similar findings have been reported previously following infection of goats with a high-passage strain of M. capricolum subsp. capripneumoniae, in which only 7 of 20 infected animals exhibited a positive titer by CFT (11). The LAT was used to test 57 field serum samples obtained from CCPP-affected areas in Eritrea (T. Tekleghiorghis, Prog. Abstr. 3rd Workshop Mycoplasmas of Ruminants of the COST Action 826) to determine the level of correlation with the results of CFT (Table 5). An exact correlation in diagnosis was noted for 38 of 57 samples tested (67% overall correlation). Twenty-six serum samples were negative by both CFT and LAT, while 12 serum samples were positive by both tests. Seven serum samples were CFT positive and LAT negative, while 12 serum samples were LAT positive but CFT negative, which would agree with the premise that the antigen-detection LAT may be more sensitive than CFT. However, an exact correlation is unlikely since CFT detects a serological response, in contrast to LAT, which detects M. capricolum subsp. capripneumoniae antigen and which is therefore likely to be more sensitive at the early stages of infection (CPS antigen has been detected in the serum of experimentally infected animals within 4 to 9 days [23]) or 8 to 11 days [this study], while significant complement fixation titers were not observed until 21 days postinfection [17]).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

An antigen-detection LAT was produced by coating polystyrene latex microspheres with the IgG fraction of M. capricolum subsp. capripneumoniae-specific rabbit hyperimmune antiserum. Experimental findings are consistent with the fact that the M. capricolum subsp. capripneumoniae CPS is the soluble antigen recognized by the test. Spent M. capricolum subsp. capripneumoniae growth medium cleared by centrifugation still gave a positive agglutination reaction (with no decrease in sensitivity), indicating that a soluble antigen rather than the whole organism is responsible. Purified CPS in solution caused strong and rapid agglutination, while preabsorption of anti-M. capricolum subsp. capripneumoniae IgG with CPS prior to coating of latex microspheres removed all agglutinating activity when the beads were subsequently used in a LAT. In addition, the specificity of the LAT exactly matched that of a CPS-specific monoclonal antibody (WM-25) which only recognizes M. capricolum subsp. capripneumoniae and bovine serogroup 7 from M. mycoides cluster mycoplasmas (2, 22). The LAT was tested against a bank of 52 serum samples from goat herds located at a variety of locations in the United Kingdom (a country which has never experienced CCPP), with all samples returning a negative result in agglutination reactions. Latex microspheres coated with anti-M. capricolum subsp. capripneumoniae IgG in the range of 0.1 to 1.0 mg of IgG per 1.5 mg of microspheres were found to be effective in the LAT, and the sensitivity of the test was found to be 2 ng of CPS, or the equivalent of 3.6 × 10³ CFU M. capricolum subsp. capripneumoniae per reaction mixture, in a volume of 0.03 ml (equivalent to 125 ng of CPS per ml, or 1.2 × 10⁵ M. capricolum subsp. capripneumoniae CFU/ml). This is a level of sensitivity similar to that reported previously for an M. pneumoniae antigen-detection LAT, for which a detection limit of 2 × 10⁵ CFU was reported (15). However, it should be noted that the amount of CPS produced by M. capricolum subsp. capripneumoniae may vary in vivo; thus, these sensitivity values must be treated with caution, particularly since soluble CPS rather than whole mycoplasma appears to be the main target for the LAT. A 67% correlation between the results of LAT and the results of CFT for the diagnosis of CCPP was noted with field sera from goat herds naturally infected with M. capricolum subsp. capripneumoniae, with the main discrepancy in diagnosis being apparently due to the increased sensitivity of the LAT compared to that of CFT; while the LAT detected CCPP in our three experimentally infected goats, CFT failed to do so (although the animals were positive by both ELISA and immunoblotting). This apparently poor sensitivity of CFT at detecting subclinical CCPP has been noted before (11).

An effective antigen-detection test should be useful in identifying animals at the earliest stages of CCPP, prior to the development of a measurable immune response. This could be significant in the field, since the incubation period for CCPP can be as little as 6 days, with death in acute cases following only 2 days later (16). Previous work has indicated that CFT is unable to detect infection in such animals (thought to be due to the fact that high levels of CPS eclipse the immune response) (10, 17, 19), and even for animals with nonacute cases of CCPP CFT appears to be relatively insensitive prior to about 21 days postinfection (17, 20). In contrast, M. capricolum subsp. capripneumoniae CPS antigen has been detected in M. capricolum subsp. capripneumoniae-infected goats 4 to 9 days following the onset of clinical signs (23), suggesting that a suitable antigen-detection test should allow early diagnosis of CCPP. In our hands, using experimentally infected animals, the LAT was able to detect CPS at the first bleed (14 days postinfection), 8 to 11 days following the onset of clinical signs. The data presented in this paper suggest that the period over which the antigen-detection LAT remains effective may extend over several weeks; thus, the test may not be restricted solely to diagnosis at the primary stages of infection. The observed correlation of the results of LAT with those of CFT offers good evidence that the antigen-detection LAT should be able to diagnose both acute and nonacute cases of CCPP over an extended period.

Interestingly, CCPP (unique among diseases caused by mycoplasmas of the M. mycoides cluster) does not exhibit a bacteriemic phase and appears to be lung specific (16, 26). Detection of M. capricolum subsp. capripneumoniae antigen in the sera of infected goats (as reported in this paper and by others [23]) suggests either a hitherto undetected systemic phase to the infection or, perhaps more likely, that soluble antigen (CPS) is entering the circulation from the infected lung. Using both PCR and direct culture we were unable to detect M. capricolum subsp. capripneumoniae in the sera of infected animals, in support of the latter hypothesis (data not shown). The role (if any) of CPS in disease pathogenesis and the mechanism by which it enters the circulation remain to be determined, however.

A diagnostic test based upon the detection of carbohydrate rather than protein or DNA could offer certain advantages, since polysaccharides are the end product of multienzyme synthesis pathways and random mutations at the DNA level are less likely to affect the overall antigenic structure of the molecule. This is in contrast to a protein- or PCR-based antigen detection system, in which the change of a single nucleotide could result in an altered epitope and destroy the binding of a monoclonal antibody or prevent primer extension during PCR (24). Similarly, the likelihood that antigenic variation between mycoplasmal strains will affect the ability to diagnose CCPP is much reduced (a problem with protein-specific monoclonal antibodies, because a minority recognize epitopes in all strains of M. capricolum subsp. capripneumoniae and none of the antibodies is completely specific [27]). From the work described in this paper (Table 1) and that of others (2, 22), there is no evidence for antigenic variation of CPS between strains of M. capricolum subsp. capripneumoniae. The observed cross-reactivity between the CPSs of M. capricolum subsp. capripneumoniae and bovine serogroup 7 is unlikely to present problems in field diagnosis since both species appear to be host specific (goats and cattle, respectively) (26). There has been a single isolated incident of the isolation of bovine serogroup 7 from a goat (1), but since this animal was also infected with Pasteurella multocida, evidence for a significant pathogenic role for bovine serogroup 7 in goats appears to be slight. Apart from this, all current tests for the detection of M. capricolum subsp. capripneumoniae (with the exception of the PCR test [8]) also show cross-reactivity with bovine serogroup 7, including a monoclonal antibody-based ELISA (27).

For ease of duplication, the CPS antigen-detection LAT was produced by using a simple rabbit polyclonal hyperimmune antiserum. Use of the purified IgG fraction was found to result in an accurate and sensitive test, although a higher titer of CPS-specific antiserum was available (produced against an M. capricolum subsp. capripneumoniae membrane fraction). Production of a CPS-monospecific antiserum was not successful due to the lack of response in immunized animals (13). While a LAT based upon CPS monoclonal antibodies might give greater specificity, the costs and ease of production are likely to be considerably higher, and it is unlikely that specificity would be increased (all M. capricolum subsp. capripneumoniae CPS monoclonal antibodies produced to date also cross-react with bovine serogroup 7 [2]). A good polyclonal antiserum has the additional advantage that it should detect several epitopes, which might increase test sensitivity and reduce the susceptibility of a test to antigenic variations between isolates.

A test which allows rapid and inexpensive primary screening prior to confirmatory laboratory diagnosis should be useful in the field. In particular, such a test used in conjunction with the current antibody-detection LAT (20) (which may be expected to exhibit optimum sensitivity at a later stage of infection) could allow the rapid and inexpensive screening of herds (using latex beads of different colors to differentiate the tests). In contrast to other diagnostic tests, no expensive equipment or reagents (e.g., ELISA plate readers, PCR machines, gel electrophoresis equipment, or enzymes) are required and results can be obtained in situ within minutes. Requirements are minimal: glass slides (which are inexpensive and reusable) and disposable plastic Pasteur pipettes (with a predetermined drop size). The test appears to be more sensitive than CFT (which cannot effectively detect acute cases [10, 19]), is active at an earlier stage of infection, and is also very specific since it detects a conserved soluble antigen (CPS) rather than the whole organism. The stability of the beads after storage for 1 month at 37°C was unaffected (data not shown). The test should allow rapid testing of herds in the field, an important attribute for a disease like CCPP, in which death can occur well before the results of laboratory-based diagnosis are available to aid in field treatment strategies.