Search

Cell surface molecules that can act as viral receptors may exert an important selective pressure on RNA viral quasi-species. Coxsackie-Adenovirus Receptor and Decay Accelerating Factor (DAF, CD55) have been identified as receptors for Coxsackie B virus. In studies of viral replication using different strains of Coxsackievirus serotype 4 (CBV-4), it was found that despite lack of expression of these cell surface molecules on Chinese Hamster Ovary (CHO) cells and despite their common use as negative controls in Coxsackie B virus receptor assays, two strains were able to replicate, one (V89-4557) without cytopathic effect (CPE), and the other (T318) with strong CPE. A weak signal was obtained using antibodies against enterovirus, visualized with FITC-conjugated antibodies, when the Coxsackievirus B4 strain V89-4557 was inoculated on CHO cells compared to no signal with the non-replicating Coxsackievirus B4 strain VD2921, indicating some degree of binding of the former to the cells.

Monoclonal antibodies that interact with the decay accelerating factor (DAF, CD55), the lymphocyte homing receptor (CD44) or the intercellular adhesion molecule I (ICAM- 1) were found to inhibit the replication of different strains of Coxsackievirus serotype B4 (CBV-4) to various extent. By adding antibodies to CD55 the replication of two (V345 and VD2921) of seven strains in HeLa cells, three (V89-4557, VD2921 and T318) of seven in A549-10C cells and one (VD2921) of five strains in RD cells was blocked totally. Consequently, the replication of one strain (VD2921) was blocked in all cells indicating that this strain uses CD55 as a receptor or as a co-receptor on all cell lines and is unable to use another cell surface protein. The binding of this strain to the cell surface was inhibited by the antibodies to CD55. None of the CBV-4 strains was blocked totally by adding antibodies to CD44 to HeLa and A549-10C cells, whereas in RD cells the replication of one (T318) of the CBV-4 strains was blocked totally. The antibodies to ICAM-1 did not inhibit totally the replication of any strain in HeLa and RD cells, but it blocked totally the replication of one strain (CBV-4-E) in A549-10C cells. In HeLa and A549-10C cells the degree of replication correlated highly with the degree of cytopathic effect (CPE). In RD cells, four of the strains replicated without CPE. The adding of antibodies to the integrin alpha(v)beta(3) led to slightly enhanced replication of three of the CBV-4 strains in all cell lines. It is concluded that the receptor usage by different strains of CBV-4 varies not only within the same cells but also between different cell lines.

The aim was to study whether different strains of Coxsackievirus B4 (CBV-4) are able to infect human pancreatic islet cells in vitro and cause morphological and functional damages. Isolated islets maintained in tissue culture were infected with five well- characterised strains of CBV-4. Aliquots of the culture medium were analysed with regard to virus replication and insulin content. Infected and uninfected islets were examined by light microscopy to determine the degree of virus-induced cytopathic effect (CPE). The results showed that the islet cells were susceptible to infection by all the strains of CBV-4 although the outcome of the infection differed. The virus titres obtained at 48 and 72 hours post infection differed significantly between all the CBV-4 strains (p<0.001), indicating different ability to replicate in islet cells. Pronounced to weak CPE, which was partly due to the origin (donor) of the islets, was induced by four of the five CBV-4 strains. One strain (VD2921) replicated without causing CPE despite high virus titres. One (V89-4557) of the CBV-4 strains always revealed pronounced CPE. Infection by this strain also caused functional impairment that significantly affected insulin response to high glucose at 48 hours post infection (p<0.001). Replication of another CBV-4 strain (JVB) in the islet cells significantly increased the release of insulin compared to non-infected control cells (p<0.001) indicating damage of the beta-cells leading to leakage of insulin.

A persistent infection of rhabdomyosarcoma (RD) cells by Coxsackie B4 virus (CBV-4) was established. The persistently infected RD (piRD) cells have been maintained for over 130 passages (30 months) and have released virus continuously without cellular destruction. The production of infectious virus declined three times during the study. After the first decline (third week post infection) a viral variant with a littered cytopathic effect (CPE) and a marked delayed replication cycle on Green Monkey Kidney (GMK) cells, replaced the original viral population. 100-fold diluted cell cultures were recovered from the piRD cells at the 48(th) and 104(th) passage. All 96 cultures from the former whereas 72% from the second dilution showed virus production when tested on GMK cells. Using a streptavidin/biotin immune-staining assay all piRD cells were positively stained. Test for ts mutants showed that the persistence of the CBV-4 strain was not dependent upon incubation temperature and addition of the antiviral compound disoxaril did not cure the piRD cells.

Considerable differences in antibody responses measured by capture-IgM RIA and neutralization tests (NT) were seen in children with newly diagnosed type I (insulin-dependent) diabetes mellitus (IDDM) when five different strains of Coxsackie B4 virus (CBV-4) were used. The IgM positivity of the 160 patients varied between 3.7 and 10.0% (P < 0.05) and with the use of all strains, IgM was found in 13.2%. Matched controls showed a significantly lower frequency (P < 0.05), but there were no apparent differences between the strains, although no IgM was found against two strains. In the NT different results were also obtained in the IDDM children with use of the five strains. However, these results differed considerably from those of the RIA, indicating that different epitopes on the virus were involved. Serum specimens from 75 patients with aseptic meningitis from whom enteroviruses had been isolated, also showed varying IgM frequencies, ranging from 13.3 to 18.7%, but the differences between the strains were different to those found in the IDDM patients. We speculate that only certain strains of a serotype of CBV may be used to distinguish IDDM pathogenesis from that of other diseases.

In a population-based setting, we traced serum samples collected at time of birth from 55 mothers whose children later developed insulin-dependent diabetes (IDDM) and matched them pairwise to control subjects who gave birth at the same hospital during the same month. The sera were analysed for IgM antibodies to coxsackie B virus serotypes 2, 3 and 4 (CBV-2, 3 and 4) using a type-specific mu-antibody-capture radioimmunoassay. Despite a decreased power due to the close matching by time of birth we found a significantly higher frequency of CBV-3 IgM at delivery in mothers whose children later became diabetic compared to their matched control subjects. When using the presence of CBV-3 IgM as a risk factor the Mantel-Haenszel odds ratio estimate (95% confidence limits) was 2.57 (1.02; 7.31), p = 0.043. For CBV-2 and CBV-4, respectively no significant difference was found between mothers of patients and control subjects. According to the odds ratio estimate for CBV-3 and the proportion of exposed mothers among patients estimated in this study the aetiological fraction for this risk determinant would be 27%. In conclusion, this study indicates that children of mothers who expressed CBV IgM at delivery are at increased risk for developing childhood onset IDDM. A fetal infection with CBV similar to rubella virus may initiate autoimmunity or cause persistent infection that may lead to progressive beta-cell destruction.

Five strains of Coxsackie B4 virus and one of Echo 11 virus were tested with regard to their ability to replicate in pancreatic mouse beta-cells and interfere with the alterations of the cytoplasmic Ca2+ concentration ([Ca2+]i) induced by glucose. All strains except one both multiplied and caused cytopathic effect. In a control group 68% of the beta-cells responded to 11 mM glucose with large amplitude oscillations of [Ca2+]i. After inoculation with the infectious strains these oscillations appeared in only 5% of the beta-cells, whereas the non-infectious strain did not modify the glucose effect on [Ca2+]i. Despite the virus interference with the glucose response, [Ca2+]i was increased after depolarization with excessive extracellular K+ and the oscillations were induced in most beta-cells when glucose was combined with the insulin-releasing sulfonylurea tolbutamide.

To study the possible role of Coxsackie B virus serotypes 1-5 (CBV 1-5) as an etiological factor in miscarriage, 97 women with miscarriage were tested for the presence of CBV-IgM by radioimmunoassays (RIAs) and compared with 113 control women undergoing voluntary interruption of pregnancy in the same gestational week. Of the 80 women with miscarriage before the 13th week of gestation, 34 (42%) had CBV-IgM, whereas 18 of 100 (18%) corresponding control women had these antibodies, a statistically significant difference (P < 0.001). There was no significant difference in CBV-IgM frequency between the women with miscarriage from the 13th through the 27th week and their controls. IgM against CBV 2 predominated, followed by IgM against CBV 5, CBV 4, CBV 3, and CBV 1. Two strains of CBV 5 were used in the RIAs, one (M147) isolated from a woman with miscarriage included in the study and one (V136) isolated in 1971 from a patient with meningitis. When the former strain was used, 13 women with miscarriage and seven control women had IgM, but with use of V136 only two women in each group were IgM positive. CBV 5 was isolated from the placental tissue from more women with miscarriage (6 of 28; 21%), than control women (2 of 21; 10%), but the difference was not statistically significant. No other viruses, except cytomegalovirus from a woman with miscarriage, were isolated.

IgM antibodies to Coxsackie B virus serotypes 1-5 (CBV 1-5) were studied by radioimmunoassay (RIA) of blood samples from women who had undergone a spontaneous abortion or given birth to a stillborn child. Results were compared with those of controls comprising healthy pregnant women who had not suffered such experiences. Among 48 women with abortions, 16 (33%) had CBV-IgM, while among the controls only three of 37 (8%) had these antibodies, a statistically significant difference (P less than 0.025). Among 21 women with a stillborn child, 11 (52%) were CBV-IgM positive, while the corresponding proportion in their controls was four of 18 (22%), a difference which is not statistically significant. It is concluded that CBV may be an important causative agent in spontaneous abortions and possibly also in stillbirths.

Thirty-five children with newly-diagnosed Type 1 (insulin-dependent) diabetes mellitus and their 47 siblings were investigated for the presence of IgM antibodies to Coxsackie B virus serotypes 1-5 (CBV 1-5) with the aid of mu-antibody-capture radioimmunoassays. When a high cut-off value was used, 16 (46%) diabetic children and 16 (34%) siblings showed CBV-IgM. Of the siblings of diabetic patients positive for CBV-IgM, 11 (44%) were CBV-IgM-positive; the corresponding figure for the siblings of negative patients was five (26%). With a lower cut-off value, leading to a "borderline titre", the frequency of IgM positivity increased in both the patients and siblings. When the borderline titres were included, the number of IgM-positive patients was 19 (54%) and the corresponding number of siblings was 29 (62%). Of the siblings of positive patients, 27 (93%) showed CBV-IgM, and of the siblings of the negative patients, two (11%) were CBV-IgM-positive. Sixteen (89%) siblings of IgM-negative patients remained negative. Regarding the serotypes of CBV to which IgM was directed, CBV 4 was most frequent, followed by serotypes CBV 3, CBV 2, CBV 5 and CBV 1. There was a striking similarity between the individual diabetic child and his or her sibling(s) concerning this specificity of IgM. It is concluded that within most families with a newly-diagnosed diabetic child positive for CBV-IgM the same serotype(s) of the virus circulates and that the intrafamilial spread of virus is considerable. The results strongly indicate that the IgM detected was CBV-specific and caused by a recent or current CBV infection.(ABSTRACT TRUNCATED AT 250 WORDS)

By using radioimmunoassay (RIA) for detection of IgM antibodies to Coxsackie B viruses (CBV), the occurrence of these antibodies was investigated in patients with sarcoidosis and asbestos-related lesions. Sixty-one per cent of the patients with sarcoidosis, all patients with benign asbestos pleural effusion, and 67% of those with diffuse asbestos-related pleural thickening showed CBV-IgM. Patients with healed sarcoidosis or pleural plaques were all negative, and among the "healthy" controls seven per cent had CBV-IgM. Thus, there was a high frequency of CBV-IgM in patients with sarcoidosis and in those with asbestos-related diseases. Since the titres could be the effect of an unspecific polyclonal stimulation of the B cells, sera were tested for antibodies to rubella and cytomegalovirus, but without any remarkable results.

IgM antibodies to Coxsackie B virus (CBV) have recently been observed in the serum of a relatively high proportion of children with newly diagnosed insulin-dependent diabetes mellitus (IDDM). In the present study, 108 IDDM patients below the age of 15 years, diagnosed during the period 1976 to 1985, were investigated at the onset of their disease by mu-antibody-capture radioimmunoassay (RIA) of IgM against seven different enterovirus antigens, namely virions of CBV serotypes 1-5 (CBV 1-5) and procapsids of CBV 3 and CBV 5. As has been shown the RIAs with virions give type-specific or narrow type-specific reactions, whereas procapsids react with IgM against both homotypic and heterotypic enteroviruses. The annual frequency of IgM against virions varied between 15 and 76% (mean 38%). IgM against CBV3 and CBV2 predominated, but IgM against the other serotypes was also observed. When procapsids were used as antigen, the frequency of IgM varied between 11 and 86% (mean 63%). With virions and procapsids, the corresponding variation was 44-100% (mean 70%). The total number of patients exhibiting virion-IgM was 41, whereas procapsid-IgM alone [indicating an infection with Coxsackie A virus (CAV) and/or echo virus (EV)] was detected in 36 patients. For 2 of the years, samples of serum from control groups were included. These showed a significantly lower frequency of IgM in both the virion and the procapsid RIAs. It is concluded that not only infection with CBV but also that with other enteroviruses, such as CAV and/or EV, may be involved in the pathogenesis of IDDM in children.

The effects of animal sera used at various concentrations in dilution buffers for radioimmunoassays (RIAs) of human enterovirus-IgM were studied. The origin of the sera had no impact on the titres, but adverse effects on virus-specificity tests were noted when sera from some species were used. In attempts to block the binding of 35S-labelled virus by adding unlabelled virus, sera from cow, horse and lamb and from swine and man could usually not be used; instead of a blocking effect, an increase in bound labelled virus was noted in most cases. When fetal or newborn calf serum or sera from chicken were used, this phenomenon did not occur. The factor(s) causing the enhancement of virus binding could not be identified, but it was evident that it was not present in all sera from the same species and it was very probable that immunoglobulins were not involved.

All new cases of insulin dependent diabetes mellitus (IDDM) in children below 15 years of age were recorded prospectively during a 21-year period 1964-1984 in a defined uptake area with a relatively constant child population. The total number of children recorded was 222-111 boys and 111 girls. The number of new cases varied between 4 cases in 1968 and 20 in 1984; in 1983 seventeen new cases were recorded. Specific IgM antibodies against Coxsackie B virus (CBV), types 1-5 were measured by a reverse radioimmunoassay (RIA) technique in 24 consecutive patients collected during the period March 1982-January 1984, some of whom represented the recent period of a very high incidence of diabetes. Sixteen patients (67%) exhibited CBV IgM responses, strongly suggesting a current or recent CBV infection. The titres declined during the first few months of diabetes and seemed to be absent after the first half-year period. Among age-matched non-diabetic children scheduled for elective procedures during the same period, no cases with CBV-IgM antibodies were detected. Only three of the 16 IgM-RIA-positive patients showed a significant rise in the neutralising antibody titre against the same Coxsackie B type. It is concluded that CBV may play a pathogenetic role in induction of IDDM, and possibly more frequently so during periods with a high incidence of diabetes, at least in children below 15 years of age.

A reverse radioimmunoassay (RIA) of antibodies to enteroviruses, previously developed for the detection of IgM antibodies to Coxsackie B1 (CB1) and B3 (CB3) and to Echo 11 (E11) and 30 (E30) viruses, was extended in the present study for the detection of IgM antibodies to Coxsackie B2 (CB2), B4 (CB4), and B5 (CB5) viruses and of IgG antibodies to CB1-CB5, E11, and E30 viruses. After standardisation of the assays and application to a collection of serum specimens from patients with proven enterovirus infections, specimens from patients with diagnosed or suspected acute myo- and/or pericarditis (myopericarditis group), and control specimens from patients with nonenterovirus infections, were studied, as well as from apparently healthy subjects. Of the patients with enterovirus infections, 29 of 30 (97%) were positive in the IgM RIA and 19 of 25 (76%) in the IgG RIA. In the myopericarditis group, 18 of 37 (49%) patients showed Coxsackie B (CB) virus-specific IgM titres and 9 of 37 (24%) CB virus-specific IgG titres. In the control specimens very few positive responses were detected. The RIAs appeared to be type specific or at least predominantly type specific, provided that the amount of labeled virus was carefully standardised. The sensitivity of the RIAs seemed to be rather high for IgM but low for IgG. In the neutralisation (NT) test no significant rise or fall in titre against CB viruses was demonstrated in the myopericarditis group. It is concluded that the reverse IgM RIA may be valuable for studies of the role of CB viruses in acute myo- and/or pericarditis.

Twenty-four consecutive children with newly diagnosed insulin-dependent (type I) diabetes mellitus (IDDM) were investigated for a history of infectious disease. Thirteen of the 24 (54%) patients reported symptoms of acute infection within two months before diabetes was diagnosed. The mean age was 8.5 years and 15 (63%) of the patients were girls. No clear seasonal variation in onset was seen. Coxsackie B (CB)-virus-specific IgM responses were detected by reverse radioimmunoassay (RIA) in 16 of the 24 (67%) patients on the day of diagnosis of IDDM. The highest titre was usually recorded at that time, but with some the highest titre was found with a second serum obtained three to seven weeks after diagnosis. Thereafter the titres declined, and after six months IgM was detected only in a few patients. Thirteen patients displayed monotypic IgM responses, whereas three patients showed ditypic responses. Among the former, IgM was recorded against Coxsackie B4 (CB4) in four, B5 (CB5) in three, B1 (CB1) in two, B2 (CB2) in two, and B3 (CB3) in two patients. The ditypic responses were against CB2 and CB3, CB3 and CB4, and CB5. No CB-virus-specific IgM was detected in sera, found during the same period, from age-matched nondiabetic children without evidence of infection. In neutralisation (NT) tests, antibodies to the homotypic virus were found in 12 of the 16 diabetic patients showing CB-virus-specific at the time of diagnosis. A significant rise in NT titre was demonstrated in three of these patients. No significant clinical difference was noted between IgM positive and IgM negative patients.(ABSTRACT TRUNCATED AT 250 WORDS)

When fresh guinea pig serum is added to chicken red blood cells (RBC) agglutinated by E. rhusiopathiae, the RBC lyse. This indicates that the E. rhusiopathiae-RBC complex is capable of inducing an alternative pathway of complement activation.

A predominantly type-specific mu-capture radioimmunoassay (RIA) of IgM antibodies to Coxsackie B1-B5 (CB1-CB5) viruses was previously described (Frisk et al., 1984). The present study is concerned with the specificity of this assay, using as antigen different strains of one serotype (CB5) and procapsids of two serotypes (CB3 and CB5). Eight strains of CB5 virions were tested against acute and/or convalescent sera from 10 patients from whom CB5 had been isolated. Seven patients' sera were tested against their own strain. The frequency of IgM-positive patients varied from 9 of 10 (90%) to 5 of 10 (50%). In three cases the highest titres were obtained with the patients' own strain. When sera from patients with other enterovirus infections were tested against the CB5 strains, heterotypic titres were obtained to a certain extent (0-15.6%). The strains giving a high frequency of homotypic titres varied concerning heterotypic reactions. It is concluded that the choice of strain is important if a high frequency of homotypic titres with no or only a few heterotypic reactions is to be obtained. When procapsids were used as antigen, both homotypic and heterotypic titres were seen to a large extent. All patients with homotypic IgM against CB3 or CB5 virions showed IgM against the CB3 or the CB5 procapsids, respectively. When sera from patients with other enterovirus infections were tested, IgM was found in 54 of 93 patients (58%) with use of the CB3 procapsid and in 52 of 87 patients (60%) with the CB5 procapsid. It was often not the same patients who showed IgM against the two different procapsids. When both procapsids were used, IgM positivity was found in 62 of 81 patients (77%) with other enterovirus infections. It is concluded that the use of two or more procapsids in combination is of value for the diagnosis of a recent or current enterovirus infection.

Antibody responses to varicella-zoster virus (VZV) deoxythymidine kinase (dTK) and herpes simplex virus (HSV) dTK in homologous and heterologous infections were studied. Antibodies blocking the enzymatic activity of VZV-dTK appeared late after varicella and decreased more or less in parallel with the decreasing complement fixing [CF] titre. In herpes zoster, on the other hand, antibodies to VZV-dTK appeared soon after infection. Antibodies against HSV dTKs appeared long after primary infection, but they were subsequently present in all other HSV-CF positive sera. In recurrent HSV, all acute sera were already HSV-dTK antibody positive, and three of nine persons showed an increase in titer between their acute and convalescent sera. Blocking antibodies to VZV-dTK appeared rapidly in specimens from three of 18 individuals positive by an immunofluorescence VZV-immunity test during HSV infection, whereas all other specimens remained devoid of blocking antibodies against VZV-dTK. A rise in antibody titre against HSV-dTK during VZV infections was observed in serum specimens from three of 13 HSV-CF positive patients, whereas an antibody response against HSV-dTK was not found in HSV-CF negative individuals in connection with VZV infections. The relevance of the sporadic increase in the titres of antibodies against heterologous viral dTKs is discussed.

An improved method for the detection of deoxythymidine kinase (TK) in human sera is reported. The method which utilizes 125I-iododeoxyuridine (IdUrd) as a substrate was used to measure TK in sera from patients with different diseases. Sera collected during the acute stage of infectious mononucleosis were found to contain elevated levels of TK, in most cases 10-40 times the normal value. The serum TK activity disappeared gradually and reached a normal level within 4 weeks. Sera from patients with other viral infections contained in most cases normal serum TK levels except in connection with measles, rubella, varicella, herpes simplex virus and cytomegalovirus infections. Additional studies revealed that sera from patients with different types of advanced lymphomas, acute leukemias, chronic granulocytic leukemia and lung cancer of the small-cell type with metastases, contained high TK levels which fluctuated in parallel with alterations in activity of the disease. The TK activity in sera from patients with both mononucleosis and tumor disease was characterized by electrophoresis and by its ability to utilize cytidine triphosphate as the phosphate donor. The results showed that the serum TK has the same properties as the human cytosolar TKI, except in connection with varicella.

A method for the absorption of false radioimmunoassay (RIA) IgM titres against herpes simplex virus (HSV) and cytomegalovirus (CMV) is presented. The serum specimens were absorbed by a mixture of protein A-Sepharose and protein A-Sepharose saturated with normal human gamma globulin (PAS/IgG). The detection of rheumatoid factor of IgM class (IgM-RF) as well as antinuclear antibodies (ANA) of both IgM and IgG class by solid-phase RIA is also described, and their role in the false IgM results was studied. It was found that the PAS/IgG absorption removed 50-90% of both IgM-RF and total IgG. The reduction of IgM-ANA clustered at 50-90% or nothing, whereas the reduction of IgG-ANA was approximately 50%. The studies with HSV and CMV antigens indicated that the removal of false IgM titres was more effective than the removal of each of these four factors. It was concluded that the IgM-RF titres alone were not sufficiently high to explain the false IgM results, but the ANA activity probably contributed.

Sera from essentially all Swedish children aged 0-14 years with Type 1 (insulin-dependent) diabetes mellitus with onset during an autumn period (October-December 1985) and a late spring period (May-June 1986) were selected. In all, 98 patients were analysed for IgM antibodies against coxsackie B virus serotypes 1 through 5 by a mu-antibody capture radio immunoassay technique. Sera from 94 referent children matched for age, sex and residential area, collected during the same period, were also analysed. During the autumn period, 10 out of 67 (15%) diabetic children were IgM positive while 14 out of 75 (19%) of the healthy referent children demonstrated positivity. During the late spring period only one out of 31 (3%) children with diabetes and two out of 19 (10%) referent children were IgM positive. In the diabetic patients, five were coxsackie B2 positive while coxsackie B1, 3, 4 and 5 were represented by one to three patients each. Eight referent children were coxsackie B4 positive, six were B3 positive and two B2 positive, while no referent children were positive against coxsackie B1 and 5. During these two periods in late 1985 and early 1986 these data demonstrate no evidence of increased antibody frequency against coxsackie B virus 1 through 5 at the onset of childhood diabetes in Sweden.

Indirect radioimmunoassays (RIAs) of IgM and IgG antibodies to enteroviruses have been developed, using coxsackieviruses B1 and B3, and echoviruses 11 and 30. The titres of IgM and IgG were assayed in paired sera from patients infected with one of these viruses or coxsackieviruses A7, A9, A16, B2, B4, B5 or echoviruses 4, 17, or 25. Both IgM and IgG were found in almost all serum pairs with each of the four viruses used as an antigen, and there were no certain differences between titres obtained with homologous and heterologous antigens. The convalescent phase specimens contained significantly higher titres compared with the acute phase specimens, the difference being most pronounced for IgG. Of the specimens from patients with nonenterovirus infections, a relatively high percentage contained IgM and IgG against enterovirus antigen. However, no increases in titres were seen between acute and convalescent specimens. When specimens from younger patients, aged 2 days to 22 months, without evidence of enterovirus infections, were assayed with enterovirus antigen, the frequency of IgM titres was seen to increase with age. Almost all specimens from newborns were negative, whereas the specimens from 12- to 22-month-old children showed a high frequency of IgM titres. In specimens from patients aged 2 days to 8 months, the ratio between IgM and IgG titres increased with age, probably due to a loss of maternal IgG. The IgG titres in specimens from 8.5- to 22-month-old children were similar to the titres of specimens from the patients with nonenterovirus infections. A reverse IgM assay was also developed, using the same viruses and serum specimens as for the indirect assays. In contrast to the indirect IgM assay, the reverse IgM assay was apparently type specific, provided that the amount of labeled virus was carefully standardized. The reverse IgM RIA detected and identified antibody responses better than the neutralization test. Attempts to develop a reverse IgG assay were promising concerning the specificity, but the sensitivity was low.

Thirty-six consecutive paediatric patients (0-16 years old) with recently contracted juvenile diabetes (IDDM) during 1982-84 were included in the study. Sera were assayed for recent or current Coxsackie B virus (CBV) infection using a specific and sensitive IgM RIA. Eighteen patients (50%) had IgM against CBV 1-5. The patients were also assayed for restriction fragment length polymorphism (RFLP) patterns with DNA probes coding for HLA-DR and DQ beta chains. The CBV-positive patients (n = 18) had either RFLP patterns associated with HLA-DR 3 or 4 or HLA-DQ patterns III or IV beta. Two of the CBV negative patients had neither HLA-DR 3 nor DR 4 and four of them had neither DQ patterns III nor IV. Eleven out of 18 CBV-positive patients had HLA-DQ III and DR 3 (61%) versus 5 out of 18 (28%) of the CBV-negative patients. All 11 patients with serology positive for CBV 2, 3, and 5 had HLA-DR 4 and DQ IV patterns. This was significantly (P less than 0.01) different from all five CBV 4-positive patients, who in contrast all had HLA-DR 3 or HLA-DQ III patterns. CBV 1-positive patients (n = 2) all had HLA-DR 3, 4, and HLA-DQ III, IV patterns. Thus CBV 4 seems to be significantly associated with a different host genetic constitution from at any rate CBV 2, 3, and 5, and possibly CBV 1.

IgM antibodies to coxsackievirus type B 1-5 (CB 1-5) have recently been observed in sera from children with newly diagnosed insulin-dependent diabetes mellitus (IDDM). In the present study IDDM patients below 15 years of age diagnosed between 1978 and 1984 inclusive in two different areas (counties) of Sweden, Uppsala and Linköping, were studied at the onset of their disease. The incidence of IDDM per 100,000 children below the age of 15 years varied between 11.1 and 40.8, with a mean of 26.3 in Uppsala and 23.0 in Linköping. CB-IgM was determined by a mu-antibody capture radioimmunoassay (RIA) and was found in the sera of 40% (range 15-76%) of the Uppsala patients and 30% (0-57%) of the Linköping patients. The IDDM onset peaked in January, August and October, and these months also displayed the largest numbers of CB-IgM-positive cases. There seemed to be a relation between IDDM incidence and CB-IgM positivity, but not during the whole period. IgM antibodies against CB2 and CB3 predominated during the first part of the study period, whereas during the last two years IgM antibodies against all five serotypes were demonstrated. It is concluded that if CB virus infection is involved in the pathogenesis of IDDM in children, this may be the case in less than half of the patients, though with seasonal and annual peaks and nadirs and also changes of serotypes that may reflect the natural epidemiology of these viruses.