10
Early diagnosis of tuberculous meningitis by detection of anti-BCG secreting cells in cerebrospinal fluid
A nitrocellulose immunospot assay to enumerate cells secreting anti-BCG antibodies was evaluated in the diagnosis of tuberculous meningitis. Among 25 Chinese patients with this disease diagnosed on clinical and cerebrospinal fluid (CSF) criteria, 24 had anti-BCG IgG antibody secreting cells in CSF, at a median value of 56 per 104 CSF cells. Among 6 patients examined within one week, 5 had antibody secreting cells in CSF, and all 19 patients examined 2-8 weeks after onset had such cells in CSF. Blood examined from 5 patients contained anti-BCG IgG or IgM antibody secreting cells, but usually at much lower numbers than in corresponding CSF, indicating that the specific antibody response is largely confined to CSF. Anti-BCG antibodies measured in parallel by ELISA were slightly raised in CSF in only 1 of the 6 patients examined within a week, whereas a good correspondence was seen between numbers of cells and antibody secreting antibody concentrations from the second week onwards. Detection of anti-BCG IgG antibody secreting cells has higher sensitivity and specificity than ELISA measurement of anti-BCG IgG antibodies, and represents a new, useful method for the early diagnosis of tuberculous meningitis.
Introduction The
prevalence
of tuberculous
meningitis
in
developing
these diseases, the specific B-cell response as documented in single cells is largely confined to the CSF. We report the application of this principle to the diagnosis of tuberculous
meningitis. Patients and methods Patients 16 female and 9 male Chinese patients aged 4-54 years (mean 31) had tuberculous meningitis. The diagnosis was based on a clinical picture of tuberculous meningitis and CSF abnormalities in the form of pleocytosis (which, depending on the duration of the disease, was dominated by polymorphonuclear leucocytes or mononuclear cells), increased CSF pressure, reduced glucose (less than 40 mg per 100 ml) and chloride concentrations (less than 110 mmol per 100 ml), and increased total protein. CSF only was available from 20 patients, and CSF and blood were obtained simultaneously from the remaining 5. Seven of the CSF samples were taken before treatment, and the remaining eighteen were obtained at various intervals after anti-tuberculous drugs were started. All patients improved clinically, but 1 died from a
complication of hydrocephalus. CSF was also obtained from 8 female and 4 male Chinese controls aged 8-54 years (mean 35) with other inflammatory neurological disorders (OIND): 4 had suppurative meningitis confirmed by culture of bacteria from CSF, 2 had cryptococcus meningitis verified by positive India-ink preparations, 1 had cryptococcus meningitis diagnosed on the basis of clinical and CSF features, and 5 had acute viral meningo-encephalitis of unknown cause. CSF and blood were also obtained from 10 patients with acute appendicitis undergoing spinal anaesthesia. No patient had neurological signs or symptoms, or abnormal CSF cell count, CSF/serum albumin ratio, or CSF IgG index.
countries, including China, remains high and the disease continues to have a high mortality. Tuberculous meningitis is difficult to diagnosis with certainty, especially in its early phase. The clinical presentation and cerebrospinal fluid (CSF) abnormalities vary, and smears for acid-fast bacilli yield few positive results. Usually growth of Mycobacterium tuberculosis is not recognised in culture media until after about 4 weeks, and the frequency of positive cultures in our hospital in China is only about 15%. To improve diagnosis, enzyme-linked immunosorbent assays (ELISA) have been developed to demonstrate anti-mycobacterial antibodies in CSF and serum, but the sensitivity and specificity of ELISA is low.1,2 A different approach for an early, sensitive, and specific diagnosis of tuberculous meningitis could be the detection and enumeration of cells from CSF-secreting specific antibodies.3,4 Adopting this approach, we have demonstrated that patients with Lyme neuroborreliosis constantly have cells secreting antibodies to Borrelia burgdorferi,5 and most patients with multiple sclerosis have cells secreting antibodies to myelin autoantigen.6 In both
Antigen BCG was obtained from the Shanghai Institute of Biological Products (SIBP), Shanghai, China, adjusted to a protein concentration of 10 ltg per ml, and stored at 4°C.
CSF and blood sampling CSF obtained by lumbar puncture was sampled in silicon-coated glass tubes. After cell counting by phase-contrast microscopy, the CSF was centrifuged at 1000 rpm for 10 min. The supernatant was discarded and the cell-pellet resuspended in RPMI 1640 tissue culture medium (Flow Laboratories, Irvine, UK) containing 15% fetal calf serum (FCS; SIBP, Shanghai) and antibiotics. This
Institute of Neurology, Shanghai Medical University, Shanghai, China, (Prof Chuan-Zhen Lu, MD, Jian Qiao MD, Tao Shen, MD) and Department of Neurology, Karolinska Institutet, Huddinge University Hospital, Stockholm, Sweden (Chuan-Zhen Lu and Prof H Link, MD). Correspondence to Prof
ADDRESSES:
Chuan-Zhen Lu, Department of Neurology, Karolinska Institutet, Huddinge University Hospital, S-141 86 Huddinge, Sweden.
11
TABLE I-ANTI-BCG ANTIBODY SECRETING CELLS OF IgG AND IgM ISOTYPES* IN CSF AND BLOOD FROM PATIENTS WITH TUBERCULOUS MENINGITIS
*No of cells per 10’ mononuclear cells
+ve=positive
suspension was centrifuged, washed twice in medium, and diluted to final cell concentration of 25-50 x 103 cells per ml. Peripheral blood was taken in heparinised glass tubes, and mononuclear cells were separated by density gradient centrifugation on Ficoll (Shanghai Reagent Factory 3), washed three times in phosphate buffered saline (pH 7-4 [PBS]) diluted in medium, counted, and adjusted to a final concentration of 106cells per ml. a
Enumeration of anti-BCG secreting cells
IgG and anti-BCG IgM
To detect cells secreting BCG antibodies of the IgG isotype, we used a solid-phase enzyme-linked immunospot assay and microtitre plates with nitrocellulose bottoms (‘Millititre HA’, Millipore, Bedford, USA).’’ 100 ul samples of BCG (1 ug/ml) in coating buffer (0-05 molfl carbonate buffer, pH 9-6) were added to individual wells, and plates kept overnight at 4°C. The optimum antigen concentration for coating was established in preliminary experiments on CSF cells from patients with clinically diagnosed tuberculous meningitis. After removal of the coating solution by suction through the nitrocellulose membranes and subsequent washings in PBS, the remaining protein binding-sites on the membranes were blocked with 10% FCS at 37°C for 1 h. The plates were then washed with PBS and dried. 200 ui samples containing 5-10 x 103 CSF cells or 20 x 104 peripheral blood lymphocytes (PBL) in medium were added to each well. All examinations were made in duplicate or triplicate, according to the numbers of cells available. After overnight incubation at 37°C in 5% CO2 and a humidified atmosphere the wells were emptied and washed five times with PBS. 100 ul of high-affinity purified biotinylated rabbit anti-human IgG (diluted 1 in 400) or IgM (I in 200) antiserum (SIBP, Shanghai) was added to appropriate wells. After 4 h the wells were washed with PBS, then incubated for 1 h with avidin-biotin peroxidase complex (SIBP; diluted 1 in 100). The enzyme reaction was developed as described.7 Red-brown coloured spots were counted. Values obtained were standardised to numbers of spots per 10’ CSF cells or PBL. We have shown earlier that the spots detected in this assay represent the secretion of
antibodies.
Duration of tuberculous
meningitis (weeks)
Fig 2-Anti-BCG IgG secreting cells per 10" CSF cells (upper), and anti-BCG IgG levels at optical density 492 nm in CSF (lower) from 23 tuberculous meningitis patients in relation to onset of symptoms.
ELISA to measure anti-BCG IgG
Polyvinyl chloride microtitre plates with 96 wells (Tientsin Plastic-Glass, Tientsin, China) were coated with 100 1 BCG (1 Ilg/ml) in 0-05 mol/1 carbonate buffer (pH 9-6). After 2 hours’ coating at 37°C or overnight at 4°C, the plates were washed three times with PBS-tween, blocked with 10% FCS for 1 h at 37°C, and then washed three times with PBS-tween. 100 1 samples of CSF diluted 1 in 10 or serum diluted 1 in 500 were added. After 2 h at 37°C the plates were washed with PBS-tween. 100 ul samples of biotinylated rabbit anti-human IgG antiserum diluted 1 in 1000 in PBS with 10% FCS were added, and the plates were kept at room temperature for 2 h. After washing and addition of avidin-biotin peroxidase complex diluted 1 in 100 in PBS with 10% FCS, the plates were incubated for 1 h, washed, and 100 (il of substrate was added (4 mg 0-phenylenediamine, 10 ml citrate phosphate buffer, and 15 11130% H202). The enzyme reaction was stopped when the positive control (CSF from a patient with clinically diagnosed tuberculous meningitis) showed an optical density (OD) at 492 nm of about 1 -0; the negative control (CSF from a patient with acute appendicitis) gave an OD value of less than 012. The OD at 492 nm for CSF diluted 1 in 10 from the 10 appendicitis patients was 0-019-0-095 (mean 0065, SD 0-017), yielding an upper reference value for anti-BCG IgG in CSF (mean + 3 SD) of12. Our upper reference value for anti-BCG IgG antibodies in serum (mean +3 SD) is 0-56, based on earlier results for 50 Chinese blood donors.
Results Anti-BCG IgG and IgM secreting cells Cells secreting anti-BCG IgG were present in CSF from all but 1 (96%) of the 25 patients with tuberculous meningitis (table I; fig 1). The number of cells varied between 8 and 450 per 104 CSF cells (median 56; mean 90). Of 5 patients examined in parallel for anti-BCG antibody secreting cells of the IgM isotype in CSF, 3 were positive
Fig 1-Anti-BCG IgG secreting cells per 104 cells in CSF and peripheral blood (PB) from 25 patients with tuberculous meningitis (TM), and per 10" CSF cells from 12 patients with other inflammatory neurological diseases (OIND).
(table I). Peripheral blood was examined for anti-BCG antibody secreting cells in 5 patients (tables I and II). Although all 5 had IgG antibody secreting cells in CSF, only 2 had positive blood specimens and the numbers were much lower. 4 of these 5 patients had very few cells secreting BCG antibodies
12
TABLE H—ANTi-BCG ANTIBODY SECRETING CELLS OF IgG AND IgM ISOTYPES* IN CSF AND BLOOD IN 5 PATIENTS WITH
TUBERCULOUS MENINGITIS
(TM)
*No of cells per 1 ()4 mononuclear cells. M= male, F= female.
isotypes we have shown that almost all patients with a clinical diagnosis of tuberculous meningitis had anti-BCG secreting cells in CSF. M tuberculosis and BCG have proved to have the same antigens, as tested by a reference antiserum.9 We therefore used BCG as antigen in our assays. The results in the 5 patients with tuberculous meningitis in whom we examined peripheral blood as well as CSF clearly show that the specific B-cell response in tuberculous meningitis is largely confined to the CSF. Thus if the immunospot assay is used for diagnosis of tuberculous meningitis, CSF should be examined first. Cells secreting anti-BCG antibodies of the IgM isotype were also detectable in CSF, but less frequently than and always in parallel with cells secreting specific antibodies of the IgG class. A raised IgA index reflecting intrathecal IgA synthesis has been reported in tuberculous meningitis,1O and investigation of the specific IgA antibody response would be of interest.
of the IgM isotype in peripheral blood and 1 did not have detectable cells in CSF obtained at the same time. We conclude that cells secreting specific antibodies in tuberculous meningitis are largely confined to the CSF, but are also demonstrable in peripheral blood at much lower
frequencies. Among 6 patients examined within a week after onset, 5 had anti-BCG IgG secreting cells in CSF (fig 2). The highest numbers of such cells were seen in specimens obtained during weeks 2 and 3 after onset. Thereafter, lower numbers were found. The duration of disease was not known in 2 patients. Of the 12 controls with OIND, 1 with cryptococcus meningitis had anti-BCG IgG secreting cells in CSF numbering 30 per 104 CSF cells.
Anti-BCG IgG measured by ELISA On the basis of an upper reference value ofO- 12 at OD 492 for CSF, 19 (76%) patients with tuberculous meningitis had raised anti-BCG IgG concentrations in CSF. Of the 6 patients examined within a week after onset of symptoms, only 1 had raised antibody titres (fig 2). Among the 4 patients examined during the second week, 3 had raised titres. The remaining patients had increased anti-BCG concentrations which were in most higher than those recorded during the second week after onset. There was no significant correlation between the numbers of anti-BCG IgG secreting cells and anti-BCG IgG titres measured by ELISA in the 23 CSF specimens obtained up to 8 weeks after the onset of symptoms of tuberculous meningitis nm
(fig 2). patient with cryptococcus meningitis who was positive for anti-BCG antibody secreting cells also had a raised anti-BCG IgG antibody titre in CSF. The patient died soon after examination, and a second specimen was not available, nor was necropsy done. No anti-BCG IgG was detectable in CSF in the other 11 OIND patients. The
Discussion An early diagnosis of tuberculous meningitis is highly relevant to the treatment and prognosis of the disease. A common difficulty in developing countries, including China, is to distinguish tuberculous meningitis from, for example, partly treated bacterial meningitis or fungal
meningitis. By use of an immunospot assay which allows enumeration of individual cells secreting specific antibodies of various
To improve the early diagnosis of tuberculous meningitis, several tests have been proposed, including ELISA and radioimmunoassay for the detection of mycobacterial antigen,1,11-14 and ELISA to detect antibodies to mycobacteria,15,16 but because of their lack of sensitivity and specificity they are not often used.2 Antibodies to mycobacteria can be seen in CSF from patients with various diseases that are accompanied by blood-brain barrier damage, including other inflammatory diseases of the nervous system, cerebrovascular diseases, and intracranial neoplasms.17,IS As DanieP pointed out, improvements in technology are necessary before clinical application of these or related methods can be achieved. The assay we used fulfils several of the prerequisites for a useful diagnostic test in tuberculous meningitis. The sensitivity is high (96%), and positive results are obtained in most patients examined within a week after onset. The specificity also seems to be high. The assay is simple and cost-effective, and results are available within 24 hours. Our results indicate that specific antibody secreting cells can be detected in CSF before specific antibodies. This sensitivity is a major advantage since an early diagnosis can be made. The sensitivity and specificity of the immunospot assay in this context needs to be confirmed by larger studies. We are using this assay routinely in the diagnosis of tuberculous meningitis in our hospital in Shanghai, and we believe it is a new and reliable method for the diagnosis of this disease, especially in its early phase. The detection of mycobacteria by the polymerase chain reaction (PCR) could also be useful for diagnosis.19 Comparative studies of the diagnostic accuracy of the immunospot assay and PCR in tuberculous meningitis are warranted. This study was supported by Shanghai Scientific Association, the Swedish Medical Research Council, and funds from Karolinska Institute. We thank Ms Yvonne Nilsson for secretarial help.
REFERENCES 1. Daniel IM. New
approaches to the rapid diagnosis of tuberculous meningitis. J Infect Dis 1987; 155: 659-02. 2. Daniel TM. Antibody and antigen detection for the immunodiagnosis of tuberculosis: why not? What more is needed? Where do we stand today? J Infect Dis 1988; 158: 678-80. 3. Link H. CSF IgG and its congeners. In: Thompson EJ, ed., Advances in CSF protein research and diagnosis. Lancaster: MTP Press, 1987: 49-88. 4. Link H, Baig S, Olsson T, Lolli F. Antibody producing cells in CSF: a new tool for evaluation of B cell response in MS. In: Confavreux C, Aimard G, Devic M, eds. Trends in European multiple sclerosis research. Amsterdam: Elsevier Science Publishers, 1988: 173-81.
13
5. Baig S, Olsson T, Link H. Predominance of Borrelia burgdorferi specific B cells in cerebrospinal fluid in neuroborreliosis. Lancet 1989; ii: 71-74. 6. Olsson T, Baig S, Höjeberg B, Link H. Antimyelin basic protein and antimyelin antibody-producing cells in multiple sclerosis. Ann Neurol 7.
1990; 27: 132-36. Kaplow LS. Substitute for benzidine in immunoperoxidase stains. Am J
Clin Pathol 1975; 63: 451. 8. Zachau A, Strigård K, Baig S, Höjeberg B, Olsson T. Distribution of plasma cells secreting antibodies against nervous tissue antigens during experimental allergic encephalomyelitis enumerated by a nitrocellulose immunospot assay. J Neurol Sci 1989; 91: 323-36. 9. Chaparas SD, Hendricks SR. Comparison of strains of BCG, 1: antigenic analysis and tuberculosis reactivity. Infect Immun 1973; 7: 777-80. 10. Kinnman J, Link H, Frydén A. Characterization of antibody activity in oligoclonal immunoglobulin G synthesized within the central nervous system in a patient with tuberculous meningitis. J Clin Microbiol 1981; 13: 30-35. 11. Sada E, Ruiz-Palacios GM, Vidal YL, Léon SP. Detection of mycobacterial antigens in cerebrospinal fluid of patients with
tuberculous meningitis by enzyme-linked immunosorbent assay. Lancet 1983; ii: 651-53. 12. Kadival GV, Mazarelo TMS, Chaparas SD. Sensitivity and specificity of enzyme-linked immunosorbent assay in the detection of antigen in
tuberculous 901-04.
meningitis cerebrospinal fluid. J Clin Microbiol 1986; 23:
Raja A, Machicao AR, Morrissey AB, Jacobs MR, Daniel TM. Specific detection of mycobacterium tuberculosis in radiometric cultures by using an immunoassay for antigen 5. J Infect Dis 1988; 158: 468-70. 14. Kadival GV, Samuel AB, Mazarelo TMS, Chaparas SD. Radioimmunoassay for detecting mycobacterium tuberculosis antigen in cerebrospinal fluids of patients with tuberculous meningitis. J Infect Dis 1987; 155: 608-11. 15. Hernandez R, Munoz O, Guiscafre H. Sensitive enzyme immunoassay for early diagnosis of tuberculous meningitis. J Clin Microbiol 1984; 20: 13.
533-35. 16. Chandramuki A, Allen PRJ, Keen M, Ivanyi J. Detection of mycobacterial antigens and antibodies in the cerebrospinal fluid of patients with tuberculous meningitis. J Med Microbiol 1985; 20: 239-47. 17. Prabhakar
S, Oommen A. ELISA using mycobacterial antigens as a diagnostic aid for tuberculous meningitis. J Neurol Sci 1987; 78:
203-11. 18. Watt G, Zarespe
G, Bautista S, Laughlin LW. Rapid diagnosis of tuberculous meningitis by using an enzyme-linked immunosorbent assay to detect mycobacterial antigen and antibody in cerebrospinal fluid. J Infect Dis 1988; 158: 681-86. 19. Bell J. The polymerase chain reaction. Immunol Today 1989; 10: 351-55.
Association within a family of a balanced autosomal translocation with major mental illness
282 pedigrees in the MRC Cytogenetics Registry, Edinburgh, with familial autosomal anomalies were examined for the presence of associated mental illness. In one large pedigree there were 23 cases of mental and/or behavioural disorders meeting Research Diagnostic Criteria. 34 of the 77 family members available for cytogenetic analysis carried a balanced translocation t(1:11) (q43,q21). Psychiatric diagnoses had been recorded for 16 of the 34 members with the translocation compared with only 5 of the 43 without it. The lod scores (against chance linkage of the translocation with mental illness) were greatest when the mental disorders in the phenotype were restricted to schizophrenia, schizoaffective disorder, recurrent major depression, and adolescent conduct and emotional disorders. Although the mental illness in this family may not be typical of that in the general population, the findings suggest that the q21-22 region of chromosome 11 may be a promising area to examine for genes predisposing to major mental
illness.
Introduction The identification of patients with rare associated cytogenetic anomalies has played an important part in the localisation of genes causing several human diseases. We present here details of a large Scottish family with a balanced translocation t(l: 11) (q43,q21) associated with many cases of major mental illness. Smith and colleagues reported a family in which manic-depressive illness cosegregated in
members with a translocation breakpoint in the same region of chromosome 11.
Subjects and methods The MRC Cytogenetics Registry, Edinburgh, contains individual family and annual follow-up clinical data on 282 pedigrees with familial autosomal anomalies identified by the unit over the past 20 years. Each pedigree was examined for the presence of associated psychiatric illness. We found several isolated associations but only one definite example of many cases of mental disorder cosegregating with a chromosomal anomaly within a single pedigree. The proband was initially ascertained in 1968 during a cytogenetic survey of boys at an allocation centre for admission of juvenile delinquents to Scottish borstals. He had an apparently balanced translocation between chromosome 1 and a chomosome in group C. The abnormality, which was present in four generations, could be traced through several branches, one of which also had a Robertsonian 13q-14q rearrangement. No phenotypic abnormalities were reported in the original description of the pedigree.2 However, the cumulative annual follow-up data from general practitioners over the intervening years showed that many family members had been referred to psychiatrists and/or admitted to mental hospitals. We reapproached the family and systematically inquired about the mental health of each member. 77 family members were included in the study, 58 living and 19 who had died. Case-notes
Department of Psychiatry, University of Edinburgh, Royal Edinburgh Hospital (D. St Clair, MRCPsych, D. Blackwood, MRCPsych, W. Muir, MRCPsych, M. Walker, MA) and
ADDRESSES:
Medical Research Council Human Genetics Unit, Western General Hospital (D. St Clair, W. Muir, A. Carothers, PhD, G. Spowart, BSc, Prof C. Gosden, PhD, Prof H. J. Evans, PhD), Edinburgh, UK. Correspondence to Dr D. St Clair, MRC Human Genetics Unit, Western General Hospital, Edinburgh EH4 2XU, UK.
Early diagnosis of tuberculous meningitis by detection of anti-BCG secreting cells in cerebrospinal fluid
A nitrocellulose immunospot assay to enumerate cells secreting anti-BCG antibodies was evaluated in the diagnosis of tuberculous meningitis. Among 25 Chinese patients with this disease diagnosed on clinical and cerebrospinal fluid (CSF) criteria, 24 had anti-BCG IgG antibody secreting cells in CSF, at a median value of 56 per 104 CSF cells. Among 6 patients examined within one week, 5 had antibody secreting cells in CSF, and all 19 patients examined 2-8 weeks after onset had such cells in CSF. Blood examined from 5 patients contained anti-BCG IgG or IgM antibody secreting cells, but usually at much lower numbers than in corresponding CSF, indicating that the specific antibody response is largely confined to CSF. Anti-BCG antibodies measured in parallel by ELISA were slightly raised in CSF in only 1 of the 6 patients examined within a week, whereas a good correspondence was seen between numbers of cells and antibody secreting antibody concentrations from the second week onwards. Detection of anti-BCG IgG antibody secreting cells has higher sensitivity and specificity than ELISA measurement of anti-BCG IgG antibodies, and represents a new, useful method for the early diagnosis of tuberculous meningitis.
Introduction The
prevalence
of tuberculous
meningitis
in
developing
these diseases, the specific B-cell response as documented in single cells is largely confined to the CSF. We report the application of this principle to the diagnosis of tuberculous
meningitis. Patients and methods Patients 16 female and 9 male Chinese patients aged 4-54 years (mean 31) had tuberculous meningitis. The diagnosis was based on a clinical picture of tuberculous meningitis and CSF abnormalities in the form of pleocytosis (which, depending on the duration of the disease, was dominated by polymorphonuclear leucocytes or mononuclear cells), increased CSF pressure, reduced glucose (less than 40 mg per 100 ml) and chloride concentrations (less than 110 mmol per 100 ml), and increased total protein. CSF only was available from 20 patients, and CSF and blood were obtained simultaneously from the remaining 5. Seven of the CSF samples were taken before treatment, and the remaining eighteen were obtained at various intervals after anti-tuberculous drugs were started. All patients improved clinically, but 1 died from a
complication of hydrocephalus. CSF was also obtained from 8 female and 4 male Chinese controls aged 8-54 years (mean 35) with other inflammatory neurological disorders (OIND): 4 had suppurative meningitis confirmed by culture of bacteria from CSF, 2 had cryptococcus meningitis verified by positive India-ink preparations, 1 had cryptococcus meningitis diagnosed on the basis of clinical and CSF features, and 5 had acute viral meningo-encephalitis of unknown cause. CSF and blood were also obtained from 10 patients with acute appendicitis undergoing spinal anaesthesia. No patient had neurological signs or symptoms, or abnormal CSF cell count, CSF/serum albumin ratio, or CSF IgG index.
countries, including China, remains high and the disease continues to have a high mortality. Tuberculous meningitis is difficult to diagnosis with certainty, especially in its early phase. The clinical presentation and cerebrospinal fluid (CSF) abnormalities vary, and smears for acid-fast bacilli yield few positive results. Usually growth of Mycobacterium tuberculosis is not recognised in culture media until after about 4 weeks, and the frequency of positive cultures in our hospital in China is only about 15%. To improve diagnosis, enzyme-linked immunosorbent assays (ELISA) have been developed to demonstrate anti-mycobacterial antibodies in CSF and serum, but the sensitivity and specificity of ELISA is low.1,2 A different approach for an early, sensitive, and specific diagnosis of tuberculous meningitis could be the detection and enumeration of cells from CSF-secreting specific antibodies.3,4 Adopting this approach, we have demonstrated that patients with Lyme neuroborreliosis constantly have cells secreting antibodies to Borrelia burgdorferi,5 and most patients with multiple sclerosis have cells secreting antibodies to myelin autoantigen.6 In both
Antigen BCG was obtained from the Shanghai Institute of Biological Products (SIBP), Shanghai, China, adjusted to a protein concentration of 10 ltg per ml, and stored at 4°C.
CSF and blood sampling CSF obtained by lumbar puncture was sampled in silicon-coated glass tubes. After cell counting by phase-contrast microscopy, the CSF was centrifuged at 1000 rpm for 10 min. The supernatant was discarded and the cell-pellet resuspended in RPMI 1640 tissue culture medium (Flow Laboratories, Irvine, UK) containing 15% fetal calf serum (FCS; SIBP, Shanghai) and antibiotics. This
Institute of Neurology, Shanghai Medical University, Shanghai, China, (Prof Chuan-Zhen Lu, MD, Jian Qiao MD, Tao Shen, MD) and Department of Neurology, Karolinska Institutet, Huddinge University Hospital, Stockholm, Sweden (Chuan-Zhen Lu and Prof H Link, MD). Correspondence to Prof
ADDRESSES:
Chuan-Zhen Lu, Department of Neurology, Karolinska Institutet, Huddinge University Hospital, S-141 86 Huddinge, Sweden.
11
TABLE I-ANTI-BCG ANTIBODY SECRETING CELLS OF IgG AND IgM ISOTYPES* IN CSF AND BLOOD FROM PATIENTS WITH TUBERCULOUS MENINGITIS
*No of cells per 10’ mononuclear cells
+ve=positive
suspension was centrifuged, washed twice in medium, and diluted to final cell concentration of 25-50 x 103 cells per ml. Peripheral blood was taken in heparinised glass tubes, and mononuclear cells were separated by density gradient centrifugation on Ficoll (Shanghai Reagent Factory 3), washed three times in phosphate buffered saline (pH 7-4 [PBS]) diluted in medium, counted, and adjusted to a final concentration of 106cells per ml. a
Enumeration of anti-BCG secreting cells
IgG and anti-BCG IgM
To detect cells secreting BCG antibodies of the IgG isotype, we used a solid-phase enzyme-linked immunospot assay and microtitre plates with nitrocellulose bottoms (‘Millititre HA’, Millipore, Bedford, USA).’’ 100 ul samples of BCG (1 ug/ml) in coating buffer (0-05 molfl carbonate buffer, pH 9-6) were added to individual wells, and plates kept overnight at 4°C. The optimum antigen concentration for coating was established in preliminary experiments on CSF cells from patients with clinically diagnosed tuberculous meningitis. After removal of the coating solution by suction through the nitrocellulose membranes and subsequent washings in PBS, the remaining protein binding-sites on the membranes were blocked with 10% FCS at 37°C for 1 h. The plates were then washed with PBS and dried. 200 ui samples containing 5-10 x 103 CSF cells or 20 x 104 peripheral blood lymphocytes (PBL) in medium were added to each well. All examinations were made in duplicate or triplicate, according to the numbers of cells available. After overnight incubation at 37°C in 5% CO2 and a humidified atmosphere the wells were emptied and washed five times with PBS. 100 ul of high-affinity purified biotinylated rabbit anti-human IgG (diluted 1 in 400) or IgM (I in 200) antiserum (SIBP, Shanghai) was added to appropriate wells. After 4 h the wells were washed with PBS, then incubated for 1 h with avidin-biotin peroxidase complex (SIBP; diluted 1 in 100). The enzyme reaction was developed as described.7 Red-brown coloured spots were counted. Values obtained were standardised to numbers of spots per 10’ CSF cells or PBL. We have shown earlier that the spots detected in this assay represent the secretion of
antibodies.
Duration of tuberculous
meningitis (weeks)
Fig 2-Anti-BCG IgG secreting cells per 10" CSF cells (upper), and anti-BCG IgG levels at optical density 492 nm in CSF (lower) from 23 tuberculous meningitis patients in relation to onset of symptoms.
ELISA to measure anti-BCG IgG
Polyvinyl chloride microtitre plates with 96 wells (Tientsin Plastic-Glass, Tientsin, China) were coated with 100 1 BCG (1 Ilg/ml) in 0-05 mol/1 carbonate buffer (pH 9-6). After 2 hours’ coating at 37°C or overnight at 4°C, the plates were washed three times with PBS-tween, blocked with 10% FCS for 1 h at 37°C, and then washed three times with PBS-tween. 100 1 samples of CSF diluted 1 in 10 or serum diluted 1 in 500 were added. After 2 h at 37°C the plates were washed with PBS-tween. 100 ul samples of biotinylated rabbit anti-human IgG antiserum diluted 1 in 1000 in PBS with 10% FCS were added, and the plates were kept at room temperature for 2 h. After washing and addition of avidin-biotin peroxidase complex diluted 1 in 100 in PBS with 10% FCS, the plates were incubated for 1 h, washed, and 100 (il of substrate was added (4 mg 0-phenylenediamine, 10 ml citrate phosphate buffer, and 15 11130% H202). The enzyme reaction was stopped when the positive control (CSF from a patient with clinically diagnosed tuberculous meningitis) showed an optical density (OD) at 492 nm of about 1 -0; the negative control (CSF from a patient with acute appendicitis) gave an OD value of less than 012. The OD at 492 nm for CSF diluted 1 in 10 from the 10 appendicitis patients was 0-019-0-095 (mean 0065, SD 0-017), yielding an upper reference value for anti-BCG IgG in CSF (mean + 3 SD) of12. Our upper reference value for anti-BCG IgG antibodies in serum (mean +3 SD) is 0-56, based on earlier results for 50 Chinese blood donors.
Results Anti-BCG IgG and IgM secreting cells Cells secreting anti-BCG IgG were present in CSF from all but 1 (96%) of the 25 patients with tuberculous meningitis (table I; fig 1). The number of cells varied between 8 and 450 per 104 CSF cells (median 56; mean 90). Of 5 patients examined in parallel for anti-BCG antibody secreting cells of the IgM isotype in CSF, 3 were positive
Fig 1-Anti-BCG IgG secreting cells per 104 cells in CSF and peripheral blood (PB) from 25 patients with tuberculous meningitis (TM), and per 10" CSF cells from 12 patients with other inflammatory neurological diseases (OIND).
(table I). Peripheral blood was examined for anti-BCG antibody secreting cells in 5 patients (tables I and II). Although all 5 had IgG antibody secreting cells in CSF, only 2 had positive blood specimens and the numbers were much lower. 4 of these 5 patients had very few cells secreting BCG antibodies
12
TABLE H—ANTi-BCG ANTIBODY SECRETING CELLS OF IgG AND IgM ISOTYPES* IN CSF AND BLOOD IN 5 PATIENTS WITH
TUBERCULOUS MENINGITIS
(TM)
*No of cells per 1 ()4 mononuclear cells. M= male, F= female.
isotypes we have shown that almost all patients with a clinical diagnosis of tuberculous meningitis had anti-BCG secreting cells in CSF. M tuberculosis and BCG have proved to have the same antigens, as tested by a reference antiserum.9 We therefore used BCG as antigen in our assays. The results in the 5 patients with tuberculous meningitis in whom we examined peripheral blood as well as CSF clearly show that the specific B-cell response in tuberculous meningitis is largely confined to the CSF. Thus if the immunospot assay is used for diagnosis of tuberculous meningitis, CSF should be examined first. Cells secreting anti-BCG antibodies of the IgM isotype were also detectable in CSF, but less frequently than and always in parallel with cells secreting specific antibodies of the IgG class. A raised IgA index reflecting intrathecal IgA synthesis has been reported in tuberculous meningitis,1O and investigation of the specific IgA antibody response would be of interest.
of the IgM isotype in peripheral blood and 1 did not have detectable cells in CSF obtained at the same time. We conclude that cells secreting specific antibodies in tuberculous meningitis are largely confined to the CSF, but are also demonstrable in peripheral blood at much lower
frequencies. Among 6 patients examined within a week after onset, 5 had anti-BCG IgG secreting cells in CSF (fig 2). The highest numbers of such cells were seen in specimens obtained during weeks 2 and 3 after onset. Thereafter, lower numbers were found. The duration of disease was not known in 2 patients. Of the 12 controls with OIND, 1 with cryptococcus meningitis had anti-BCG IgG secreting cells in CSF numbering 30 per 104 CSF cells.
Anti-BCG IgG measured by ELISA On the basis of an upper reference value ofO- 12 at OD 492 for CSF, 19 (76%) patients with tuberculous meningitis had raised anti-BCG IgG concentrations in CSF. Of the 6 patients examined within a week after onset of symptoms, only 1 had raised antibody titres (fig 2). Among the 4 patients examined during the second week, 3 had raised titres. The remaining patients had increased anti-BCG concentrations which were in most higher than those recorded during the second week after onset. There was no significant correlation between the numbers of anti-BCG IgG secreting cells and anti-BCG IgG titres measured by ELISA in the 23 CSF specimens obtained up to 8 weeks after the onset of symptoms of tuberculous meningitis nm
(fig 2). patient with cryptococcus meningitis who was positive for anti-BCG antibody secreting cells also had a raised anti-BCG IgG antibody titre in CSF. The patient died soon after examination, and a second specimen was not available, nor was necropsy done. No anti-BCG IgG was detectable in CSF in the other 11 OIND patients. The
Discussion An early diagnosis of tuberculous meningitis is highly relevant to the treatment and prognosis of the disease. A common difficulty in developing countries, including China, is to distinguish tuberculous meningitis from, for example, partly treated bacterial meningitis or fungal
meningitis. By use of an immunospot assay which allows enumeration of individual cells secreting specific antibodies of various
To improve the early diagnosis of tuberculous meningitis, several tests have been proposed, including ELISA and radioimmunoassay for the detection of mycobacterial antigen,1,11-14 and ELISA to detect antibodies to mycobacteria,15,16 but because of their lack of sensitivity and specificity they are not often used.2 Antibodies to mycobacteria can be seen in CSF from patients with various diseases that are accompanied by blood-brain barrier damage, including other inflammatory diseases of the nervous system, cerebrovascular diseases, and intracranial neoplasms.17,IS As DanieP pointed out, improvements in technology are necessary before clinical application of these or related methods can be achieved. The assay we used fulfils several of the prerequisites for a useful diagnostic test in tuberculous meningitis. The sensitivity is high (96%), and positive results are obtained in most patients examined within a week after onset. The specificity also seems to be high. The assay is simple and cost-effective, and results are available within 24 hours. Our results indicate that specific antibody secreting cells can be detected in CSF before specific antibodies. This sensitivity is a major advantage since an early diagnosis can be made. The sensitivity and specificity of the immunospot assay in this context needs to be confirmed by larger studies. We are using this assay routinely in the diagnosis of tuberculous meningitis in our hospital in Shanghai, and we believe it is a new and reliable method for the diagnosis of this disease, especially in its early phase. The detection of mycobacteria by the polymerase chain reaction (PCR) could also be useful for diagnosis.19 Comparative studies of the diagnostic accuracy of the immunospot assay and PCR in tuberculous meningitis are warranted. This study was supported by Shanghai Scientific Association, the Swedish Medical Research Council, and funds from Karolinska Institute. We thank Ms Yvonne Nilsson for secretarial help.
REFERENCES 1. Daniel IM. New
approaches to the rapid diagnosis of tuberculous meningitis. J Infect Dis 1987; 155: 659-02. 2. Daniel TM. Antibody and antigen detection for the immunodiagnosis of tuberculosis: why not? What more is needed? Where do we stand today? J Infect Dis 1988; 158: 678-80. 3. Link H. CSF IgG and its congeners. In: Thompson EJ, ed., Advances in CSF protein research and diagnosis. Lancaster: MTP Press, 1987: 49-88. 4. Link H, Baig S, Olsson T, Lolli F. Antibody producing cells in CSF: a new tool for evaluation of B cell response in MS. In: Confavreux C, Aimard G, Devic M, eds. Trends in European multiple sclerosis research. Amsterdam: Elsevier Science Publishers, 1988: 173-81.
13
5. Baig S, Olsson T, Link H. Predominance of Borrelia burgdorferi specific B cells in cerebrospinal fluid in neuroborreliosis. Lancet 1989; ii: 71-74. 6. Olsson T, Baig S, Höjeberg B, Link H. Antimyelin basic protein and antimyelin antibody-producing cells in multiple sclerosis. Ann Neurol 7.
1990; 27: 132-36. Kaplow LS. Substitute for benzidine in immunoperoxidase stains. Am J
Clin Pathol 1975; 63: 451. 8. Zachau A, Strigård K, Baig S, Höjeberg B, Olsson T. Distribution of plasma cells secreting antibodies against nervous tissue antigens during experimental allergic encephalomyelitis enumerated by a nitrocellulose immunospot assay. J Neurol Sci 1989; 91: 323-36. 9. Chaparas SD, Hendricks SR. Comparison of strains of BCG, 1: antigenic analysis and tuberculosis reactivity. Infect Immun 1973; 7: 777-80. 10. Kinnman J, Link H, Frydén A. Characterization of antibody activity in oligoclonal immunoglobulin G synthesized within the central nervous system in a patient with tuberculous meningitis. J Clin Microbiol 1981; 13: 30-35. 11. Sada E, Ruiz-Palacios GM, Vidal YL, Léon SP. Detection of mycobacterial antigens in cerebrospinal fluid of patients with
tuberculous meningitis by enzyme-linked immunosorbent assay. Lancet 1983; ii: 651-53. 12. Kadival GV, Mazarelo TMS, Chaparas SD. Sensitivity and specificity of enzyme-linked immunosorbent assay in the detection of antigen in
tuberculous 901-04.
meningitis cerebrospinal fluid. J Clin Microbiol 1986; 23:
Raja A, Machicao AR, Morrissey AB, Jacobs MR, Daniel TM. Specific detection of mycobacterium tuberculosis in radiometric cultures by using an immunoassay for antigen 5. J Infect Dis 1988; 158: 468-70. 14. Kadival GV, Samuel AB, Mazarelo TMS, Chaparas SD. Radioimmunoassay for detecting mycobacterium tuberculosis antigen in cerebrospinal fluids of patients with tuberculous meningitis. J Infect Dis 1987; 155: 608-11. 15. Hernandez R, Munoz O, Guiscafre H. Sensitive enzyme immunoassay for early diagnosis of tuberculous meningitis. J Clin Microbiol 1984; 20: 13.
533-35. 16. Chandramuki A, Allen PRJ, Keen M, Ivanyi J. Detection of mycobacterial antigens and antibodies in the cerebrospinal fluid of patients with tuberculous meningitis. J Med Microbiol 1985; 20: 239-47. 17. Prabhakar
S, Oommen A. ELISA using mycobacterial antigens as a diagnostic aid for tuberculous meningitis. J Neurol Sci 1987; 78:
203-11. 18. Watt G, Zarespe
G, Bautista S, Laughlin LW. Rapid diagnosis of tuberculous meningitis by using an enzyme-linked immunosorbent assay to detect mycobacterial antigen and antibody in cerebrospinal fluid. J Infect Dis 1988; 158: 681-86. 19. Bell J. The polymerase chain reaction. Immunol Today 1989; 10: 351-55.
Association within a family of a balanced autosomal translocation with major mental illness
282 pedigrees in the MRC Cytogenetics Registry, Edinburgh, with familial autosomal anomalies were examined for the presence of associated mental illness. In one large pedigree there were 23 cases of mental and/or behavioural disorders meeting Research Diagnostic Criteria. 34 of the 77 family members available for cytogenetic analysis carried a balanced translocation t(1:11) (q43,q21). Psychiatric diagnoses had been recorded for 16 of the 34 members with the translocation compared with only 5 of the 43 without it. The lod scores (against chance linkage of the translocation with mental illness) were greatest when the mental disorders in the phenotype were restricted to schizophrenia, schizoaffective disorder, recurrent major depression, and adolescent conduct and emotional disorders. Although the mental illness in this family may not be typical of that in the general population, the findings suggest that the q21-22 region of chromosome 11 may be a promising area to examine for genes predisposing to major mental
illness.
Introduction The identification of patients with rare associated cytogenetic anomalies has played an important part in the localisation of genes causing several human diseases. We present here details of a large Scottish family with a balanced translocation t(l: 11) (q43,q21) associated with many cases of major mental illness. Smith and colleagues reported a family in which manic-depressive illness cosegregated in
members with a translocation breakpoint in the same region of chromosome 11.
Subjects and methods The MRC Cytogenetics Registry, Edinburgh, contains individual family and annual follow-up clinical data on 282 pedigrees with familial autosomal anomalies identified by the unit over the past 20 years. Each pedigree was examined for the presence of associated psychiatric illness. We found several isolated associations but only one definite example of many cases of mental disorder cosegregating with a chromosomal anomaly within a single pedigree. The proband was initially ascertained in 1968 during a cytogenetic survey of boys at an allocation centre for admission of juvenile delinquents to Scottish borstals. He had an apparently balanced translocation between chromosome 1 and a chomosome in group C. The abnormality, which was present in four generations, could be traced through several branches, one of which also had a Robertsonian 13q-14q rearrangement. No phenotypic abnormalities were reported in the original description of the pedigree.2 However, the cumulative annual follow-up data from general practitioners over the intervening years showed that many family members had been referred to psychiatrists and/or admitted to mental hospitals. We reapproached the family and systematically inquired about the mental health of each member. 77 family members were included in the study, 58 living and 19 who had died. Case-notes
Department of Psychiatry, University of Edinburgh, Royal Edinburgh Hospital (D. St Clair, MRCPsych, D. Blackwood, MRCPsych, W. Muir, MRCPsych, M. Walker, MA) and
ADDRESSES:
Medical Research Council Human Genetics Unit, Western General Hospital (D. St Clair, W. Muir, A. Carothers, PhD, G. Spowart, BSc, Prof C. Gosden, PhD, Prof H. J. Evans, PhD), Edinburgh, UK. Correspondence to Dr D. St Clair, MRC Human Genetics Unit, Western General Hospital, Edinburgh EH4 2XU, UK.
