Comparative evaluation of the enzyme immunoassay and the immunofluorescence test for the detection of chlamydial group antibodies in human sera

Zbl. Bakt. Hyg., I. Abt, Orig. A 254, 435-440 (1983) Ocular Microbiology Section, Dr . Rajendra Prasad Centre, All India Institute of Medical Sciences, New Delhi-l10029, India

Comparative Evaluation of the Enzyme Immunoassay and the Immunofluorescence Test for the Detection of Chlamydial Group Antibodies in Human Sera Vergleichende Beurteilung des Enzym-Immunoassays und des Immunfluoreszenztests zur Ermittlung von Chlamydien-GruppenAntikorpern in menschlichen Seren ARVIND RAI and V. M. MAHAJAN

With 1 Figure' Received December 28,1982

Abstract The usefulness of an enzyme-linked immunosorbent assay (ELISA) in detecting chlarnydial group antibodies in human sera was evaluated by comparing the results with a standard micro-immunofluorescence test (micro-IF). Of the 176 samples tested, 94.9% sera gave concordant results, 23.9% being positive and 71% negative by both the tests. ELISA detected specific antibodies in 33.3% and 13.5% sera from poultry associated and unassociated individuals respectivel y while corresponding figures by IF test were 36.3% and 16.2% (p < 0.01). The IF and ELISA titres of five immune serum samples from monkeys when compared with conventional complement fixation (CF) test were found to be consistently higher. Though the two tests were found to be parallel in sensitivity and specificity (r = 0.94), IF had a marginal superiority over ELISA, while the latter could be performed more easily and read objectively.

Zusammenfassung Die Nutzbarkeit des ELISA-Tests zur Feststellung von Chlamydien-Gruppen-Antikorpern in menschlichen Seren wurde durch Vergleich der Ergebnisse mit denen eines Standard-Mikro-Immunfluoreszenztests (Mikro-IF) beurteilt. Von den 176 untersuchten Proben ergaben 94,4% iibereinstimmende Ergebnisse. 23,9% waren in beiden Tests positiv und 71% in beiden negativ. Mittels ELISA wurden spezifische Antikorper bei Hiihnern sowie zusammenlebenden und nicht zusammenlebenden Einzelpersonen ermittelt, wahrend die entsprechenden Werte im IF-Test 33,3 bzw. 16,2% waren (p < 0,01). Beim Vergleich mit der konventionellen Komplementbindungsreaktion waren die IF- und ELISA-Titer von 5 Immunserumproben von Affen durchgangig hoher, Obwohl eine Parallelitat der beiden 29 Zbl. Bakt. Hyg., 1. Abt, Orig. A 254

436

A.Rai and V.M.Mahajan

Tests nach ihrer Empfindlichkeit und Spezifirat festgestellt wurde (r = 0,94), war der IFdem ELISA-Test leicht iiberlegen, wogegen sich der letztere leichter durchfiihren und objektiver ablesen lieK

Introduction Until the development of the more sensitive micro-IF test the complement-fixation (CF) test was and indeed, still is widely used in the diagnosis of two important human chlamydial infections, lymphogranuloma venereum (LGV) and psittacosis (7, 8). Both the tests are however, not much useful in the diagnosis of trachomainclusion conjunctivitis or the genital-tract infections caused by these agents (6, 8, 9). The ELISA test has been developed more recently (4,5) and there is a need for further studies to establish its usefulness in the diagnosis and epidemiology of chlamydial infections. The present study compares the ELISA with the standard micro-IF test. Since the main objective of the study was to assess the efficacy of the two tests in detecting chlamydial group antibodies, the antigen was prepared from a strain of Chlamydia trachomatis having group antigen common to C. psittaci (2, 7, 8).

Materials and Methods Antigen . Stock C. trachomatis (serotype E) grown in yolk-sacs by inoculating 0.2 ml amounts containing 105-10. essential lethal dose (ELD5o ) (5), were harvested and homogenised. The homogenate was spun at 500 x g for 10 min to remove host cell debris and the chlamydiaIelementary bodies (EB) were pelleted by centrifuging mid phase at 20,000 x g for 20 min. The final pellet rich in chlamydiae was resuspended in 0.1 M phosphatebuffered saline (PBS). Optimal antigen dilution for ELISA and IF tests was determined using known positive and negative sera obtained from our previous study (5). Antigen from uninfected yolk-sacs was similarly made. Sera. A total of 176 human sera were tested of which 102 were from individuals dealing with chicken poultry and the remaining 74 (representing the control group) from a matched population, not-associated with poultry. In addition, five known positive (CF titres 1;64 1;256) serum samples for the use in standardization were procured from our earlier study (5). All the sera were stored at - 70 °C till used. IF test. The micro-IF test (9) was done employing 3-6% infected yolk-sac suspensions as antigen. Two-fold serum dilutions were added to the antigen slide and incubated for 30 to 40 min at 37 °C and subsequently washed before flooding with optimally diluted fluorescein-isothiocyanate conjugated anti-human IgG (Kallested Lot No. F204 183-2). The slides were examined using a Reichert fluorescent microscope, Austria. Appropriate controls were included. ELISA technique. The method has been described earlier (1,5). In brief, polystyrene micro-ELISA plates (Dynatech, Microtitre M29 AR) were coated with 200 pI of optimally diluted (1: 80) chlamydial antigen in 0.06 M bicarbonate buffer pH 9.6, followed by overnight incubation at 4°C. The plates were washed thrice in phosphate-buffered saline pH 7.2 containing 0.05% Tween-20 (PBS/T). Subsequently, the plates were incubated for 1 hr at 37 °C with 200 pI of two-fold serum dilutions made in PBS/T containing 0.5% bovine serum albumin (BSA). Unbound antibody was washed-out as described above. To detect specifically bound IgG, 200 pI (1: 8,000 diluted) anti-human IgG peroxidase (Miles, U. K.) conjugate was added to the wells for 45 min at 37 °C. After further washing, the bound peroxidase was visualised by incubating the plates with 200 pI of freshly prepared chromogenic substrate solution (40 mg Orrho-phenylenediarnine per 100 ml of phosphate - citrate

Chlamydial Group Antibodies

437

buffer pH 5.0, containing 40 pi of 30% H 20 2 ) . The plates were incubated in dark for 20 to 30 min. The reaction was terminated by adding 50 pi of 5N H 2S04 to each well and results were noted by visual estimation of the intensity of colour developed. Appropriate controls were included during each assay.

Results Table 1. Comparison of IF test with ELISA for the detection of chlamydial antibodies in poultry associated and unassociated population Type of Population

Serum samples Number %

Chlamydial antibody by IF ELISA

Poultry associated 102 sera

32 2 5 63

31.4 1.9 4.9 61.8

+

Control group 74 sera

10 0 2 62

13.5 0.0 2.7 83.8

+ +

+ +

+ +

+

1:2048 r:0·94

1:1024 IJl UJ

•

1:512

g 1:256

~~

o

1:64D

(A

1:32-

.S1' ::J

I

1:128

UJ

1:16 ~1:16

,8 ,0 .0° ° Ie

•

I '&.,

<1:32 1:32

°t

• ••

I

f

°•

•

ft

•

..

1:64 t28 1:256 1:512 1:1024 l2048 1:4096 CC'flRESPONDIN3 IF T1TRES

Fig. 1. Correlation between ELISA and IF titres of sera from poultry associated individuals (e), control population (0) and five selected sera (~) which were tested repeatedly on different days to show the reproducibility of ELISA test. Bars (e) indicate the range of ELISA titres obtained when these sera were tested repeatedly. (r) represents Spearman's rank correlation coefficient values between IF and ELISA.

12/74

37/102

16.2

10/74

13.5

36.3 33.3

32 = 65) NO

4 3

3

6

10

NO

IF values Chi-Square (x 2 ) = 8.5 df = 1 P < 0.01

« 32 = 62) ELISA « 16 = 64)

IF

«

ELISA 16 = 68) 6

«

IF

1

2

9

8

1 :64

2

4

6

7

1:128

1 :16

IF /ELISA

1:32

Antibody titres by IF or ELISA

% Positivity

34/102

Number of Sera Positive/Total IF ELISA

Spearm an's correlation Coefficient (r) betw een IF and ELISA r = 0.94 NO = Not done

Control group

Poultry associated population

Type of population

1

J'l

1

3

4

1:256

0

0

2

2

1 :1024

0

0

0

1

1:2048

ELISA value s Chi-Square (x 2 ) = 8.9 df = 1 P < 0.01

0

1

2

4

1:512

NO

0

NO

1

1 :4096

Table 2. Comparative sero-positivities and correlation of corresponding titres of chlamydiaI antibodies in poultry associated ind ividuals and the control group by IF te st and ELISA

$>l

::s

iT ~.

$>l

~

~

:<

0-

::s

$>l

~.

;xl

?>

~ ..... 00

ChlamydiaI Group Antibodies

439

Of the total 176 sera, 167 (94.9%) gave concordant results, 42 (23.9%) were positive and 125 (71%) negative for chlamydial antibodies by both tests (Spearman's rank correlation co-efficient (r) = 0.94) (Table 1). Out of the 9 samples which gave discordant results, seven were ELISA negative but were positive by IF test. There were only two sera negative by IF test but positive by ELISA (Fig. 1). ELISA test was started at a dilution of 1: 16 and IF at 1: 32 because many known negative sera gave false reaction at higher serum concentrations. A starting dilution of 1: 16 in ELISA was found satisfactory in our earlier study too (5). ELISA detected chlamydial antibodies in 33.3% and 13.5% of sera from poultry associated and control populations respectively (Chi-square = 8.9; p < 0.01) as against 36.3% and 16.2% detected by IF test in similar categories (Chi-square 18.5; p < 0.01) (Table 2). The ELISA titres of positive sera were found in the range of 1: 16-1 :2048 whereas IF titres varied between 1: 32-1: 4096 (Table 2). The ELISA and IF titres of five positive monkey immune sera were always four to eight times higher than those of CF titres. While the CF titres of these sera ranged between 1: 64-1: 256, the ELISA and IF titres varied from 1:256-1:2048. The reproducibility of ELISA test was verified by the finding that no more than one tube difference in titres was obtained when some of the sera were tested repeatedly on different days (Fig. 1 and Table 3). Table 3. The ELISA titres of 5 different sera tested on four different days Sera

IF

titres

ELISA titres on four different days 1 2 3

1 2 3 4 5

1:256 1:1024 1:2048 1:2048 1:2048

1:256 1:512 1:1024 1:512 1:1024

1:128 1:256 1:512 1:1024 1:2048

1:256 1:256 1: 1024 1:256 1: 1024

4

1:256 1:512 1:1024 1:1024 1:1024

Discussion The findings of the present study revealed a strong correlation between ELISA and IF tests, 94.9% sera yielding concordant results. The IF test was able to detect a higher number of positives with higher titres. A statistical analysis of the results indicated a marginal superiority of IF over ELISA. Of the total 176 sera from the two populations, ELISA detected chlamydial antibodies in 25% (33.3% in poultry contacts and 13.5% in controls). The IF on the other hand, yielded positivity in 27.8% sera (36.3% and 16.2% in poultry associated and unassociated groups, respectively). Though both populations were apparently healthy, a high seropositivity among poultry associated individuals was probably due to an active or past chlamydiosis which they might have acquired through infected birds. Previous studies from different parts of the world have indicated a prevalence of CF antibodies in 0.4%-14% sera from clinically normal individuals though these figures varied much in different IF reports (2,7,8,9). A higher positivity in both the groups observed in this study probably resulted from the employment of more sensitive serologic techniques. However, a discrepancy in 5.1 % of the sera suggested that

440

A.Rai and V.M.Mahajan

none of the two tests is fool proof. An advantage of ELISA is that it does not necessarily require spectrophotometer to measure optical density and the readings can be taken objectively as well, while IF readings cannot be taken without the aid of a fluorescent microscope.

Acknowledgements. We thank Dr. Shiv Cbaran Gupta, Assistant Professor of Virology of the department of Veterinary Microbiology, Haryana Agricultural University, Hissar for constant help in collection of serum samples from poultry associated persons.

References

1. Charan, S., Arvind Rai, and V.M. Maha;an: Comparison of enzyme-linked immunosorbent assay and haemagglutination-inhibition test for the detection of Newcastle disease virus antibodies in human sera . J. Clin. Path. 34 (1981) 90-92 2. Grayston, ]. T . and S.P. Wang : New knowledge of chlamyd iae and the diseases they cause. J. infect. Dis. 132 (1975) 81-105 3. Ismail, A. S. et al.: Demonstration of group-specific precipitation antibodies in chlamydial infections of poultry. Bed. Munch. tierarztl, Wschr. 88 (1975) 21-24 4. Lewis, V.]., W.L. Thacker , and S.A.Mitochell: Enzyme-linked immunosorbent assay for chlamydial antibodies. J. Clin. Microbiol. 6 (1977) 507-510 5. Rai, A., S. Cbaran, and V.M . Maha;an : Enzyme-linked immunosorbent assay (ELISA) for the detection of chlamydial antibodies. Ind. J. Med . Res. 76 (1982) 393-397 6. Schachter, ]. and C.R.Dawson: Comparative efficacy of various diagnostic methods for chlamydial infections, p. 337, ed. D. Hobson and K. K. Homes. Nongonococcal Urethritis and Related Infections. American Society for Microbiology, Washington D. C. (1977) 7. Schachter,]. : Chlamydial infections. New Engl. J. Med. 298 (1978) 540-549 8. Taylor-Rob inson, D. and B.]. Thomas : The role of C. trachomatis in genital-tract and associ ated diseases. J. Clin. Path. 33 (1980) 205-233 9. Wang, S. P. and ]. T. Grayston: Human serology in Chlamydia trachomatis infection with micro-immunofluorescence test. J. Infect. Dis. 130 (1974) 388-397 Dr. V. M. Mahaian, Ocular Microbiology Section, Dr . Rajendra Prasad Centre, All India Institute of Medic al Science, New Delhi - 110029, India