847

than insulin sensitivity is the primary defect in familial type II diabetes. Lancet 1986; ii: 360-64. 20. Polonsky KS, Given BD, Hirch LJ, et al. Abnormal patterns of insulin secretion in NIDDM. N Engl J Med 1988; 318: 1231-39. 21. Ward WK, Olgiano DC, McKnight B, Halter JB, Porte D. Diminished B-cell secretory capacity in patients with NIDDM. J Clin Invest 1984; 74: 1318-28. 22. Temple RC, Carrington CA, Luzio SD, et al. Insulin deficiency in NIDDM. Lancet 1989; i: 293-95. 23. Cahill GF. Beta cell deficiency, insulin resistance or both? N Engl J Med 1988; 318: 1268-69. 24. Miller GJ, Kotecha S, Wilkinson WH, et al. Dietary and other characteristics relevant to coronary heart disease in men of Indian, West Indian and European descent in London. Atherosclerosis 1988; 70: 63-72. 25. Thompson R, Cruickshank JK. Dietary fatty acid intakes, P:S ratios and

risk of coronary heart disease in women of different ethnic groups. Clin Sci 1990; 78 (suppl 22): 29-30. 26. Clark A, Wells CA, Buley ID, et al. Increased &agr;-cells, reduced &bgr;-cells and exocine fibrosis. Quantitative changes in the pancreas in type II diabetes. Diabetes Res 1988; 9: 155-59. 27. Stout R. Insulin and atheroma. A 20-year perspective. Diabetes Care 1990; 13: 631-44. 28. Hughes LO, Cruickshank JK, Parry J, Raftery ER. Disturbances of insulin in British Asian and white men surviving a first myocardial infarction. Br Med J 1989; 299: 537-41. 29. Modan M, Halkin H, Almog S, et al. Hyperinsulaemia: a link between hypertension, obesity and glucose intolerance. J Clin Invest 1985; 75: 809-17. 30. Haffner S, Stem MP, Hazuda HP, Pugh J, Patterson JK. Hyperinsulinaemia in a population at high risk of NIDDM. N Engl J Med 1986; 315: 220-24.

Diagnosis of Chlamydia trachomatis eye infection in Tanzania by polymerase chain reaction/enzyme immunoassay

Detection of

Chlamydia trachomatis eye infection is largely unsatisfactory by standard laboratory methods. A polymerase chain reaction/enzyme immunoassay (PCR-EIA) that had previously been successful for diagnosis of genital C trachomatis infection was compared with direct antibody immunofluorescence (DFA) for detection of the organism in conjunctival scrapes. 234 Tanzanian children aged 1-7 years living in a village that had had no previous trachoma control programme were classified clinically as having no sign of trachoma (0) n=97, follicular trachoma (TF) n=100, or intense inflammatory trachoma with or without TF (TI ± TF) n=37. PCR-EIA detected C trachomatis in 24%, 54%, and 95% of subjects, respectively, compared with elementary body (EB) detection by DFA of 1%, 28%, and 60%, respectively. Overall prevalence of chlamydial eye infection was 22% by DFA compared with 48% by PCR-EIA. Of subjects with chlamydial DNA at pretreatment, 103 (92%) had no detectable chlamydial DNA at the end of 4 weeks of ocular tetracycline. The findings show that PCR-EIA is likely to affect trachoma diagnosis and epidemiology because of the increased sensitivity for detection of C trachomatis in all clinical groups; the less stringent requirements for specimen collection and transport make this method suitable for field use. Moreover, the semi-quantitative aspect of PCR-EIA may be useful for monitoring a decrease in chlamydial DNA after treatment. Introduction Trachoma is the leading infectious cause of blindness ip the world. About 500 million people are affected, of whom 7 million are blind.! Serious sequelae are thought to be due to repeated infection with Chlamydia trachomatis and possible incomplete clearance of the organism. A reservoir of infection in the community provides a source of antigenic

stimulus for an immunopathogenic host response and is a source for transmission of infection.2-4 In cross-sectional epidemiological studies from trachoma-endemic areas, conventional laboratory diagnosis of C trachomatis eye infection has been made by tissue culture (TC), direct antibody immunofluorescence (DFA), and, more recently, enzyme immunoassay (EIA).5-7 The detection of C trachomatis by these methods often correlates poorly with clinical follicular and inflammatory trachoma, and this organism is usually not detected when there are no features of eye disease?-9 DNA hybridisation and the polymerase chain reaction (PCR) for DNA amplification have been used for detection of C trachomatis cervical infection, and chlamydial DNA was found in culturenegative/DFA-negative specimens by PCR.1O-12 Additionally, PCR has shown increased sensitivity for detection of other pathogenic microorganisms over conventional methods.13,14 A laboratory test with increased sensitivity over C trachomatis culture and DFA without losing specificity would be useful in trachoma diagnosis, epidemiology, and control because such a test would confirm equivocal cases or detect symptomless infection. Moreover, evaluation of the presence of chlamydial nucleic acid and disease when DFA or culture are negative may give some added insight into pathogenesis. Previously, we have coupled PCR with liquid, nonisotopic hybridisation and an EIA for detection of DNARNA hybrids; we found that PCR-EIA was a sensitive, specific, and semi-quantitative method for diagnosis of HIV-1 infection and genital C trachomatis infection.15,10 Here, we compare PCR-EIA with DFA for diagnosis of C trachomatis in ocular specimens from subjects living in a trachoma-endemic area. ADDRESSES: Departments of Pediatrics (L. Bobo, PhD, R. Viscidi, MD) and Medicine (T Quinn, MD), Division of Infectious Diseases, and Dana Center for Preventive Ophthalmology, Wilmer Institute (B. Munoz, MS, S. West, PhD), Johns Hopkins School of Medicine, Baltimore, Maryland 21205, USA; Laboratory of Immunoregulation, National Institute of Allergy and Infectious Diseases, Bethesda, Maryland, USA (T. Quinn); and Kongwa Trachoma Project, Tanzania (H. Mkocha, BS). Correspondence to Dr L. Bobo.

848

Patients and methods Patients As part of a larger study on trachoma control, conjunctival were taken from children living in the Kongwa region of Tanzania in a village that had no previous trachoma control programme. 234 households with children aged 1-7 years were randomly selected from the village and 1 index child per household was enrolled. A trained ophthalmic nurse examined the eyes using a x 2-5 loupe.16 Subjects were grouped according to the simplified World Health Organisation (WHO) trachoma grading scheme-ie, 0, no sign of trachoma; TF, trachomatous inflammation (follicular); and TI ± TF, trachomatous inflammation (intense) with or without TF.17 This survey was part of a prospective study correlating mass treatment with clinical disease and prevalence of C trachomatis. Parents were asked whether they wished their children to be part of the study before the status of trachoma was known and the study was explained as part of a trachoma control programme. Verbal consent obtained from parents in accordance with cultural practices in this area was approved by the Johns Hopkins Hospital Joint Committee on Clinical Investigation. scrapes

Samples The tarsal conjunctiva of one eye was sampled with a ’Dacron’ swab, after which it was rolled on a slide for DFA and then placed in PCR holding buffer (10 mmoljl "tris", pH 8-3,mmol/1 edetic acid [EDTA], 1% ’Nonidet P-40’/’Tween 20’ [volume/volume]). Methanol-fixed slides

stained with monoclonal-fluoroscein isothiocyanate antibody conjugate to the major outer membrane protein (MOMP) (Syva, Palo Alto, California). Smears were considered positive if 5 or more elementary bodies (EB) per slide were seen. All specimens, irrespective of epithelial cell count, were evaluated by DFA and PCR-EIA. The swab for PCR-EIA was broken off into 500 1 of holding buffer and kept at room temperature or 4°C. A 200 III aliquot was lysed with proteinase K (100 Ilgjml) at 55°C for 1 h, and enzyme was inactivated by boiling for 3 min. Samples were then chilled on ice and concentrated to 100 III with spin microdialysis (Millipore Ultrafree UFC, Bedford, Massachusetts). A swab was taken from the practitioner’s hands before specimen collection and tested by PCR-EIA. were

PCR-EtA

amplification method that combines a PCR with of PCR products in an EIA for biotinylated hybrids (PCR-EIA) was used to detect a conserved region of the MOMP gene. This method has been described in detail elsewhere.1O Briefly, a 50 ul lysate of target cellular material was amplified through 30 cycles (1 min denaturation at 94°C,1 min primer annealing at 55°C, and 1 min primer extension at 72 °C) in a 100 III PCR reaction (buffer [10 mmol/1 "tris", pH 8-3, 50 mmol/1 KC1, 25 mmol/1 MgCl2, and 0-01% gelatin]; 0-2 mmol/1 each of dATP, dGTP, dCTP, and dTTP; 25 U Taq DNA polymerase; and 05 mol/1 of primers for conserved MOMP nucleotide sequences of C trachomatis [5’-GATAGCCAGCACAAAGAGAGCTAA-3’ sense] and [5’-CTTTGTTTTCGACCGTGTTTTGCAAACAGATGTGAA-3’-antisense]). A sample of PCR product was hybridised in solution for 30 min at 78°C to a biotinylated RNA probe that was complementary to nucleotide sequences internal to the above primers. The nucleic acid mixture was incubated in A gene

measurement

COMPARISON OF PCR-EIAAND DFA FOR DETECTION OF CHLAMYDIA TRACHOMA TIS BY CLINICAL GROUP

*% of each clinical group. t2 subjects negative by PCR-EIA

B."!lnlCal group of Correlation fluorescence values with clinical group.

n = total no of positive subjects in each category. Boxes = distribution of fu between first and third quartiles for positive samples Vertical lines (bar) extending from the boxes5th and 95th percentiles for data distribution and points external to the bars are at extremes of data. Dashed lines= median fu for each clinical group.

antibiotin antibody-coated wells of a microtitre plate and hybrids bound to the solid phase were detected by reaction with a &bgr;-d-galactosidase conjugated monoclonal antibody specific for DNA-RNA hybrids. The amount of solid-phase enzyme was assessed by measurement of fluorescent product from the reaction of enzyme with 4-methylumbelliferyl-p-d-galactoside; results were expressed as fluorescence units (fu). A positive cut-off was set at 5 SDs above the mean, which was determined from the same pool of previously TC-negative/DFA-negative ocular specimens for each PCR-EIA run. The cut-off value for each run was subtracted from the mean of each sample determination. Dilutions of C trachomatis serovar E (Bour) equivalent to DNA from 1, 10, 100, and 1000 inclusion forming units (IFU) were included in each run as a positive control for PCR-EIA. To confirm the results from the first test, 60 positive samples were retested in a second PCR-EIA on separate lysates with primers that spanned the four hypervariable regions of MOMP. For this PCR, we looked for an 871 bp band on a silver-stained 75% polyacrylamide gel which had undergone electrophoresis for 2 h at 53 mA. The following precautions were taken to minimise DNA contamination by the laboratory: specimen processing was done in a biological safety cabinet, dedicated positive-pressure pipettes or negative-pressure pipettes with pipette tips containing filters were used, gloves were changed frequently, and pipettes and hood surfaces were decontaminated with 1 mol/1 HC 1. PCR products for EIA were handled in another room with a standard set of pipettes. 5 historically negative clinical specimens and reagent blanks were included in each PCR run as a check for adequate control measures.

Results

Overall, PCR-EIA detected chlamydial infection in 49% of subjects from all clinical groups compared with 22% by DFA (table). When a cut-off of 50 or more epithelial cells was used as a criterion for adequacy of specimens, 16 samples would have been disallowed, 5 of which had 5 or more EBs per slide. Although the proportion of positive DFA was lower in samples with less than 50 epithelial cells than in those with 50 or more epithelial cells, this difference was not significant (Fisher’s two-tailed test). Median fu for positive subjects by PCR-EIA increased with extent of inflammation (figure). Of the 112 samples positive by PCR-EIA, 65 were PCR-EIA positive but negative by DFA. 42 (65%) of the 65 samples were from individuals with TF or TI. The other

849

35% were subjects with a clinical diagnosis of 0. Conversely, of the 122 samples negative by PCR-EIA, 4 were positive by DFA. When these 4 samples were diluted (1/10) with water, 3 were positive for chlamydial DNA. The results with the undiluted samples suggest the presence of an inhibitor of the PCR reaction. This conclusion is further supported by the observation that these 3 samples completely inhibited PCR when spiked with chlamydial DNA equivalent to 100 IFU as judged by EIA and gel electrophoresis. Every person in the village was treated with topical tetracycline in accordance with the WHO recommendations for individuals in a hyperendemic area. 103 (92 %) of the 112 subjects who were PCR-EIA positive at pretreatment had no detectable C trachomatis DNA at the end of 4 weeks of ocular tetracycline. Median fu of samples from the remaining 9 subjects decreased from 1463 before treatment to 1058 at the end of treatment. 122 subjects had no detectable chlamydial DNA during both time periods. Overall, 225 (96%) of 234 subjects had no detectable chlamydial DNA at the end of the treatment period. As part of a continuing study to test sequence variation in isolates, a subset of 66 PCR-EIA-positive samples were also tested with another primer pair spanning the four hypervariable regions of MOMP and results were in complete agreement with the first PCR. The sample from the field examiner’s hands taken before the subject’s specimen was placed in buffer was negative.

Discussion We have shown that PCR-EIA detected chlamydial infection more often than did DFA in all clinical groups of trachoma. That 65% of the subjects who were antigen negative and DNA positive had clinical disease suggests that these subjects may indeed have been infected with chlamydia at levels below the detection ability of DFA. Additionally, 24% of subjects with no sign of trachoma were PCR-EIA positive. Clinical grading is subject to error and it is possible that less obvious clinical disease may have been present in these subjects.19 Although increased detection was seen by PCR-EIA, 3 of 4 samples positive by DFA but negative by PCR-EIA demonstrated the presence of an inhibitor to DNA amplification which could be removed by dilution. We did not regard these samples as positive because all specimens were not checked for inhibition. Inhibitors for PCR have previously been reported in clinical specimens.2o PCR-EIA detection of C trachomatis was very specific. Although DNA cross-contamination of specimens with PCR products or entire DNA genomes can affect the specificity, inadvertent contamination of specimens was unlikely since amplifications of internal controls (both buffer and clinical) were consistently negative. Also, results of the first PCR-EIA were confirmed on separate lysates of selected positive samples in a DNA amplification of the hypervariable region of the MOMP gene. Chlamydial DNA contamination in the field was also unlikely because a sample from the practitioner’s hands taken before the specimen was placed in PCR buffer was negative. That 225 (96%) of 234 subjects had no detectable DNA by PCR-EIA while on tetracycline suggests that false-positive reactions due to amplification of non-target DNA characteristic to the host or the resident microbial flora did not interfere with specificity. Previous studies on the diagnosis of C trachomatis in cervical specimens showed that PCR-EIA did not detect other microbial species tested apart from C pneumoniae.

10

Although we did not set out to assess the efficacy of ocular tetracycline treatment, the semi-quantitative nature of PCR-EIA indicated that 96% of all subjects had no detectable chlamydial DNA by PCR-EIA on the last day of treatment. The reasons why chlamydial DNA was detected in 9 subjects (despite a decrease in DNA levels) at the end of 4 weeks of treatment are that these individuals might not have completed the full regimen of tetracycline over the month, or, less likely, C trachomatis might have been resistant to tetracycline. Additional analyses by PCR-EIA and DFA are in progress to see whether these subjects are a continuing reservoir for transmission of infection and are at risk for sustained infection when treatment is discontinued. Inadequate specimen collection, storage, and transport can adversely affect the eventual outcome of laboratory testing, and these factors may be especially difficult to optimise for field studies in developing countries. Specimen handling was less demanding for PCR-EIA than for DFA. Refrigeration of fixed slides for DFA is ideal but this can be difficult to maintain if slides are to be assessed at a distant site. However, we found that the sensitivity of PCR-EIA was unaffected when specimens were stored at room temperature both in the field and during transport. At present, the cost and complexity of PCR make it an unlikely diagnostic tool for developing countries. However, DFA also is not a simple test for the diagnosis of trachoma and requires a fluorescence microscope, expensive bulbs, and well-trained technicians. The findings from our study suggest that PCR-EIA is useful for the detection of C trachomatis in trachoma, and that the association of C trachomatis DNA with TF and TI is greater than previously recorded. Additionally, we detected chlamydial DNA in 24% of clinically normal eyes. Future studies should be aimed at determining both the pathobiological and epidemiological importance of C trachomatis DNA in antigen-negative subjects. We thank Ms Farifteh Firoozmand and Mr Dan King for laboratory assistance. This study was supported by the Edna McConnell Clark Foundation; Public Health Service grants 2 P01 AI 16959-09 and N01 AI 52579 from the National Institute of Allergy and Infectious Disease; and funds from the Blindness Prevention Program of the World Health Organisation. This work was carried out under the auspices of the National Prevention of Blindness Committee on Tanzania. We thank Dr G. L. Upunda, Regional Medical Officer, and Dr B. B. 0. Mmbaga, Assistant Medical Officer of Ophthalmology, Dodoma, for support. We also thank Mr Matthew Lynch, Mr Andrew Kyongoya, and the rest of the Kongwa Trachoma Project team.

REFERENCES 1. Dawson

CR, Jones BR, Tarizzo ML. Guide to trachoma control in prevention of blindness. Geneva: World Health

programmes for the Organisation, 1981.

Grayston JT, Wang S-P, Yeh L-J. Importance of reinfection in the pathogenesis of trachoma. Rev Infect Dis 1985; 7: 717-25. 3. Taylor HR, Johnson SL, Prendergast RA, Schacter J, Dawson CR, Silverstein AM. An animal model of trachoma. II. The importance of repeated infection. Invest Ophthalmol Vis Sci 1982; 23: 507-15. 4. Watkins NG, Hadlow WJ, Moos AB, Caldwell HD. Ocular delayed hypersensitivity: a pathogenetic mechanism of chlamydial conjunctivitis in guinea pigs. Proc Natl Acad Sci 1986; 83: 7480-84. 5. Schachter JS, Dawson CR, Hoshiwara I, Daghfous T, Banks J. The use of cycloheximide treated cells for isolating trachoma agent under field 2.

conditions. Bull WHO 1978; 56: 629-32. 6. Daroucar S, Woodland RM, Jones BR, Houshmand A, Farahmandian HA. Comparative sensitivity of fluorescent antibody staining of conjunctival scrapings and irradiated McCoy cell culture for the diagnosis of hyperendemic trachoma. Br J Ophthalmol 1980; 64: 276-78. 7. Mabey DWC, Robertson JN, Ward ME. Detection of Chlamydia trachomatis by enzyme immunoassay in patients with trachoma. Lancet 1987; ii: 1491-92. 8. Wilson MC, Millan-Velasco F, Telsch JM, Taylor HR. Direct-smear

850

fluorescent antibody cytology as a field diagnostic tool for trachoma. Arch Ophthalmol 1986; 104: 688-90. 9. Taylor HR, Rapoza PA, West S, et al. The epidemiology of infection in trachoma. Invest Ophthalmol Vis Sci 1989; 30: 1823-33. 10. Bobo L, Coutlee F, Yolken RH, Quinn T, Viscidi RP. Diagnosis of Chlamydia trachomatis cervical infection by detection of amplified DNA with an enzyme immunoassay. J Clin Microbiol 1990; 28: 1968-73. 11. Peterson EM, Oda R, Alexander JR. Molecular techniques for the detection of Chlamydia trachomatis. J Clin Microliol 1989; 27: 2359-63. 12. Claas HCJ, Wagenvoort JHT, Niesters HGM, et al. Diagnostic value of the polymerase chain reaction for Chlamydia detection as determined in a follow-up study. J Clin Microbiol 1991; 29: 42-45. 13. Ulrich PP, Bhat A, Seto B, et al. Enzymatic amplification of hepatitis type B virus DNA in serum compared with infectivity testing in chimpanzees. J Infect Dis 1989; 160: 37-43. 14. Shankar P, Manjunath N, Mohan KK, et al. Rapid diagnosis of tuberculous meningitis by polymerase chain reaction. Lancet 1991; 337: 5-7.

15. Coutlee F, Yang B, Bobo L, et al. Enzyme immunoassay for detection of hybrids between PCR-amplified HIV-1 DNA and a RNA probe: PCR-EIA. Aids Res Human Retro 1990; 6: 775-84. 16.

Taylor HR, West SK, Katala S, Foster A. Trachoma: evaluation ofa new grading system in the United Republic of Tanzania. Bull WHO 1987;

17.

Thylefors B, Dawson CR, Jones BR, West SK, Taylor HR. A simple system for the assessment of trachoma and its complications. Bull

65: 485-88.

WHO 1987; 65: 477-83.

Stephens RS, Mullenbach G, Sanchez-Pescador R, Agabian N. Sequence analysis of the major outer membrane protein gene from Chlamydia trachomatis serovar L2. J Bacteriol 1986; 168: 1277-82. 19. West SK, Taylor HR. Reliability of photographs for grading trachoma in field studies. Br J Ophthalmol 1990; 74: 12-13. 20. Wilde J, Eiden J, Yolken R. Removal of inhibitory substances from human fecal specimens for detection of group A rotaviruses by reverse transcriptase and polymcrase chain reactions. J Clin Microbiol 1990; 28: 18.

1300-07.

SHORT REPORTS Antithrombin III and arterial disease

Cross-sectional studies suggest that both low and high antithrombin III levels are associated with the risk of arterial disease, principally ischaemic heart disease (IHD). The prospective relation between antithrombin III and subsequent death from arterial disease has been investigated in 893 men in the Northwick Park Heart Study. Antithrombin III levels

directly correlated with high rather than low levels of factor VII activity and of plasma fibrinogen.

were

There were more deaths from arterial disease in the low and high thirds of the antithrombin III distribution than in the middle third.

There is good evidence that high levels of procoagulatory clotting factors promote the onset of ischaemic heart disease (IHD).1 Thus we might expect low levels of potential defence mechanisms against thrombosis, such as antithrombin III, likewise to be associated with increased risk and this is certainly true of venous thrombosis. Antithrombin III inactivates thrombin, and factors Xa, IXa, XIa, and XIIa (but not VIIa2), by binding with the target enzyme to form a proteolytically inactive complex.3,4 It thus plays a major part in the neutralisation of procoagulant activity. However, the relations of antithrombotic factors to the risk of arterial thrombosis seem paradoxical. In the case of antithrombin III, vegetarians, despite their lower than average risk of IHD, have significantly lower levels than non-vegetarians.s Two studies6,7 have shown higher antithrombin III values in postmenopausal women than in premenopausal women of the same age, though the incidence of IHD is higher in the former. Levels may also be higher in diabetic than in non-diabetic subjects.8 Yue et al9 found a gradient of rising antithrombin III values from those at low risk of IHD, via those at intermediate risk or

with chronic IHD, to those with acute myocardial infarction. Findings at recruitment to the Northwick Park Heart Study (NPHS)lO also suggested higher rather than lower values in those who had previously experienced IHD. By contrast, other workersl1-13 have reported lower concentrations in those with a history of infarction or other manifestations of arterial disease. These apparently contradictory findings could be partly explained by the postulate that inability to increase antithrombin III levels in some may directly contribute to subsequent events and that antithrombin III values in others may rise as a compensatory defence mechanism. However, the studies summarised have so far all been cross-sectional, so that it has not been possible to relate antithrombin III levels to later clinical events. This preliminary report describes the prospective relation between antithrombin III concentrations and subsequent mortality from arterial disease in men in NPHS. NPHS has been described in detail elsewhere.14,15 In 1980,

during the follow-up round of examinations that began in 1978, a chromogenic substrate method for determining antithrombin III activity in citrated plasma was introduced6 and was used until follow-up was completed in 1984. Mean antithrombin III levels are less than 100% because the first standard used had a high antithrombin III value.6 Factor VII activity (VIle) and fibrinogen levels were also measured. 14 "Entry values" for these factors in this report are those obtained during the re-examinations in 1980-84. Of the 1511 white men aged between 40 and 64 at recruitment in 1972-78 and, on average, 6 years older at follow-up, antithrombin III was measured in 893. By the end of 1990, 41 of these men had died of IHD and 6 of other arterial causes (eg, stroke and aortic aneurysm) making a total of 47 cardiovascular deaths. There were 42 non-cardiovascular deaths (mainly cancer). DISTRIBUTION OF DEATHS BY CAUSE ACCORDING TO LOW (L), MIDDLE (M), AND HIGH (H) THIRDS OF ANTITHROMBIN III DISTRIBUTION

Cox analysis, risk of death relative to 1 00 in M (95% confidence interval) for IHD deaths All cardiovascular deaths L 1 78 (0 73 to 4 33) L 1 75 (0 76 to 4 04) L 2 66* (1-18 to 6-00) H265*(1 10 to 6 34)

*p=0 02

enzyme immunoassay.

Detection of Chlamydia trachomatis eye infection is largely unsatisfactory by standard laboratory methods. A polymerase chain reaction/enzyme immunoas...
617KB Sizes 0 Downloads 0 Views