A. Svejgaard

Increased frequency of homozygosity of abnormal mannan-binding-protein alleles in patients with suspected immunodeficiency

A low plasma concentration of mannan-binding protein (MBP) impairs opsonisation and phagocytosis. Three different mutations in the MBP gene have a dominant effect on MBP concentration. We investigated the frequency of the abnormal MBP alleles in 228 unrelated patients suspected of various non-HIV-related immunodeficiencies. The frequency of heterozygotes for the abnormal alleles was not different from that in the background population (36.0% and 37.4%, respectively). By contrast, the frequency of homozygotes for the abnormal alleles was significantly increased (8.3% and 0.8%, respectively; p = 0.0017). This finding implies that homozygotes for abnormal MBP alleles are predisposed to recurrent infections.

HLA-D and -DR antigens in genetic analysis of insulin dependent diabetes mellitus

SummaryThree groups of patients with insulin-dependent diabetes mellitus, ascertained by different procedures, were investigated for HLA-A, B, C and D antigens (n=164), and a subset (n=93) for HLA-DR. Both HLA-D/DR3 and D/DR4 were strongly positively associated and D/DR2 was negatively associated with insulin-dependent diabetes. HLA-DR4 was found to be a better marker for insulin-dependent diabetes than Dw4. The HLA-B associations (B8, B15 and B18) were clearly secondary to the increases of HLA-D/DR3 and D/DR 4. The HLA associations did not differ between familial and isolated cases indicating that these two groups may well have a common genetic background. Based on analysis of HLA-haplotype sharing in affected sibling pairs, a simple dominant model of inheritance could be ruled out, and a simple recessive model was found unlikely. The relative risks for the HLA-Dw3,4 and HLA-DR3,4 phenotype were 21.2 and 44.4 respectively and exceeded those of both the HLA-Dw3 and HLA-DR3 (5.6 and 4.3) as well as the HLA-Dw4 and DR4 (10.1 and 10.5) phenotypes. This argues against an intermediate genetic model but further studies are needed to clarify whether there is more than one susceptibility gene for insulin-dependent diabetes mellitus within the HLA-system.

Association of mannose-binding-lectin deficiency with severe atherosclerosis

which CNTF could be seen in the skin. The optical density of CNTF immunoreactivity in ALS patients (mean 11·0 [SD 1·0]) was significantly higher than in controls (8·0 [0·3] p<0·001). There was no relation between the optical density and the presence of dysphagia, weight loss, muscular atrophy, loss of active movement, and bedridden state in ALS patients or controls. Loss of CNTF has been proposed as a possible mechanism for the sporadic form of ALS, 5 but few studies have looked at CNTF expression in patients with ALS. Our results suggest that the changes of skin CNTF in ALS are likely to be related to the disease process. These results imply that increased CNTF in skin may play a neurotropic role and may help to explain why decubitus formation is rare in ALS.

Association between HLA-DR2 and Production of Tumour Necrosis Factor α and Interleukin 1 by Mononuclear Cells Activated by Lipopolysaccharide

The production of tumour necrosis factor (TNF) and interleukin 1 (IL‐1) by lipopolysaccharide‐activated mononuclear cells from 39 healthy donors was studied in vitro by bioassay and ELISA. The donors were typed for HLA‐A, ‐B, ‐C, ‐DR, and ‐DP antigens. There was no detectable production of TNFβ (lymphotoxin). The intracellular levels of bioactive TNFα were minimal or undetectable in all cases. Cells from HLA‐DR2+ individuals secreted significantly lower amounts of TNFα than cells from HLA‐DR2− donors [2 ng/ml (1.5–4.4) and 7.5 ng/ml (3.9–8.3) respectively (medians 25–75%); P<0.01]. The difference disappeared if the cells were preactivated for 2 days with 1000 U/ml of recombinant gamma interferon (rIFN‐γ). In some individuals, the TNFα response increased considerably after IFN‐γ priming, in particular in those possessing the HLA‐DR2 antigen. In contrast, there was no detectable difference in the production of IL‐1β between the donors, and the IL‐1β response decreased significantly after rIFN‐γ priming in HLA‐DR2+ individuals [2.3 ng/ml (1.1–8.4) versus 7.2 ng/ml (5–7.9); P<0.05] and in HLA‐DR2− individuals [3 ng/ml (1.1–5.3) versus 5.7 ng/ml (3.9–7.5); P<0.01]. There was no correlation between the TNFα and IL‐1 responses and any of the other HLA‐DR, ‐DP, or ‐B antigens. There was a significant positive correlation between the levels of TNFα measured by ELISA and by cytotoxicity assay. However, the TNFα‐containing supernatants from 9 out of 37 individuals appeared to contain inhibitor(s) of the biological activity of TNFα. The presence of inhibitor(s) was not associated with any HLA antigens.

HL-A in psoriasis vulgaris and in pustular psoriasis--population and family studies.

A collaborative study of HL‐A typing was done on 156 unrelated patients with psoriasis vulgaris and thirty‐one patients with pustular psoriasis. Highly significant increases of HL‐A13 and HL‐A17 were found in psoriasis vulgaris: about half of the patients carried either of these antigens as compared to about 12% of the controls. The effect of HL‐A13 and 17 is confined almost exclusively to psoriasis with onset before the age of 35 years. In contrast, none of these antigens was increased in pustular psoriasis, indicating that pustular and common psoriasis are different aetiological entities. Family studies in psoriasis vulgaris showed that psoriasis can develop in relatives lacking the HL‐A haplotype following psoriasis in the other family members. Hence, it seems most likely that HL‐A represents only part of the genetic basis of psoriasis vulgaris. This disease is probably of a polygenic threshold character, and HL‐A13 and 17 decreases the threshold making the expression of psoriasis more likely in individuals carrying these determinants in addition to other probably non‐HL‐A linked psoriasis determinants. This concept is supported by the observation of a higher incidence of a positive family history in HL‐A13 and/or 17 positive patients than in patients lacking these antigens. Studies of antistreptolysin O and antistreptococcal hyaluronidase titres and of triggering streptococcal infections yielded no clue as to the pathogenetic effect of HL‐A. The frequencies of HL‐A13 and 17 were not significantly increased in twenty‐one patients with psoriatic arthritis, but HL‐A27 may be increased in these patients. This study shows the importance of HL‐A in the subdivision of diseases and stresses the importance of family studies for clarifying the associations between HL‐A and disease.

RECIPIENT LYMPHOCYTE SENSITIVITY TO METHYLPREDNISOLONE AFFECTS CADAVER KIDNEY GRAFT SURVIVAL

In 42 recipients of mixed-lymphocyte-culture (MLC) incompatible cadaver kidneys and conventional immunosuppressive treatment (azathioprine and steroids) the concentrations of methylprednisolone suppressing the in-vitro response of pretransplant lymphocytes to phytohaemagglutinin by 50% (ED50) were determined. 1-year graft survival was significantly higher in 21 recipients with methylprednisolone ED50 values below the median than in 21 patients with higher than median ED50s (86% v 29%; p less than 0.0002). Thus, the steroid sensitivity of recipients strongly influences the survival of MLC-mismatched kidneys. In 42 transplant recipients treated with cyclosporin and steroids, the effect of steroid sensitivity was also apparent (1-year graft survival 76% and 57% for recipients with low and high ED50, respectively), though not significant. Determination of the sensitivity to steroids may be valuable in determining which recipients can be given HLA-DR-mismatched kidneys and may serve as a guideline for determining the dose of steroids to be used.

NcoI restriction fragment length polymorphism (RFLP) of the tumour necrosis factor (TNF alpha) region in primary biliary cirrhosis and in healthy Danes

The restriction fragment length polymorphism of the human tumour necrosis factor (TNFα) region was investigated by means of 20 different restriction enzymes and a human TNFα cDNA probe. Only one of the enzymes, NcoI, revealed a polymorphic pattern consisting of fragments of 10.5 and 5.5 kb, which behaved as alleles. In a panel of 108 random, healthy, unrelated Danes, the phenotype frequencies of the 10.5 and 5.5 kb bands were 0.94 and 0.53 and the gene frequencies were 0.71 and 0.29, respectively. The test for Hardy‐Weinberg equilibrium showed no significant deviation from the expected values. The 5.5 kb band was strongly positively associated with HLA‐DR3, HLA‐B8, and HLA‐A1. In 22 patients with primary biliary cirrhosis a significantly (corrected P= 0.024) decreased frequency of the 10.5 kb fragment was found. Additional studies are in progress to substantiate this association.

The germline repertoire of T-cell receptor beta chain genes in patients with relapsing/remitting multiple sclerosis or optic neuritis.

We have studied the germline repertoire of T-cell receptor β chain genes in 20 patients with monosymptomatic optic neuritis (ON), which may mark the onset of MS, and in 37 patients with relapsing/remitting (R/R) MS; we compared this repertoire to that of 99 unselected, unrelated healthy Danes

RELATIVE IMPORTANCE OF FOUR AND LA LOCI IN DETERMINING MIXED-LYMPHOCYTE REACTION

Abstract The one-way mixed-lymphocyte-culture test (M.L.C.) was done between an unrelated person and seven siblings, one of whom was a recombinant between the LA and Four series of the HL-A system. The single LA and Four antigen differences could be tested in four different combinations each. Isolated LA-antigen differences always failed to stimulate, while the isolated Four antigen differences gave unequivocally positive responses, even though the Four antigens mismatched, HL-A7 and FJH (W27), cross-react serologically. These findings suggest that M.L.C. reactivity is elicited by the Four antigens or depends on disparity on a third locus closely linked to the Four locus. As the M.L.C. reaction may be an in-vitro model of allograft rejection, these results call for an analysis of kidney-transplant data in respect of the clinical importance of LA versus Four antigen incompatibilities.

Crossing-over within the HL–A System

GENETIC information for the human HL–A transplantation antigens seems to be located at two closely linked mutational sites, which are often called LA, and Four1. A considerable number of genetic factors, behaving as multiple alleles, belong to each of the two sites1–4. It is still uncertain whether these are separate but closely linked “genetic units”, or whether tney belong to the same “unit” as families with unequivocal recombinants. Batchelor and Chapman5 and Ward et al.6 found a few cross-overs, but they were questioned by Walford7, for they did not relate to well known HL–A specificities.