Chondrocyte antigen expression, immune response and susceptibility to arthritis.

The association of HLA-B27 with certain forms of arthritis implies a role for MHC class I-restricted T cells in the arthritic process. Our aim was to study CD8(+) T cell responses towards specific antigens localized in joint tissue. Known determinants were introduced into chondrocytes of transgenic (TG) mice, under the control of the cis-regulatory sequences of the human type II collagen gene (COL2A1). Two Escherichia coli beta-galactosidase (beta-gal)-expressing lines were derived (CIIL73 and CIIL64) as well as two lines (CIINP) expressing influenza A virus nucleoprotein (NP). Expression of the antigens could be demonstrated in cartilaginous tissues. The TG lines showed variable degrees of responsiveness towards the transgene-introduced antigens; whilst 75% of CIIL73 mice had an impaired cytotoxic T lymphocyte (CTL) response towards beta-gal, the response in CIIL64 mice was essentially normal. However, both lines displayed normal proliferative and antibody responses to beta-gal. A reduced CTL response was seen to NP in the CIINP lines in approximately 65% of the animals. In spite of the persistence of T cell responses to the transgene antigens in these lines, induction of CTL responses alone has so far failed to induce clinical signs of arthritis. Interestingly, some animals expressing beta-gal were susceptible to arthritis following challenge with type II collagen alone, whilst their non-TG littermates and TG mice from other lines remained unaffected. As beta-gal is expressed by E. coli, a component of the normal gut flora, this suggests a possible role for gut-derived immune responses. We believe these lines could form the basis of a model for studying links between intestinal inflammation and arthritis.

An integrated multiplex-PCR and PCR-RFLP typing system for markers associated with seronegative arthritides.

The system was designed with emphasis on the identification HLA-B alleles and genotypes associated or potentially associated with seronegative arthritides. By using a combination of multiplex SSP and PCR-RFLPs, the assays can be economically performed on a large range of sample sizes in diagnosis and epidemiology. 24 HLA-B alleles and subtypes can be discriminated, including options for PCR-RFLP or sequence specific amplification of the allele groups B27 and B60 (B*4001 and B*4007). In addition, the internal control carries central MHC polymorphisms, which can help to identify HLA extended halplotypes. False negatives, caused by preferential amplification of the internal control under suboptimal PCR conditions, were prevented by employing new, optimized PCR buffer. Four of the HLA-B primers were pooled into a multiplex reaction whose products were subtyped by digestion with seven restriction endonucleases. Specificity and sensitivity were verified in a panel of 68 homozygous cell lines and 200 heterozygous samples. An HLA-B*27-B*40 hybrid allele was observed in 3 out of 95 B*27-positive individuals from Berlin, Germany. Such an allele could be mistyped by some published assays as a B*27/B*40 heterozygote, a genotype reported to confer an increased risk for ankylosing spondylitis.

Allelic polymorphism of two multifunctional regions in the central human MHC: tenascin X, XB-S and YB, and their duplicated fragments XA and YA.

Two highly polymorphic sequences have been discovered in the complement C4 region of the human major histocompatibility complex (MHC). They are part of a duplicated unit of overlapping genes, transcribed in opposite directions and containing the sequences of tenascin X (XB), XB-S, XA, YB and YA. Fragments of 1014 bp and 894 bp were co-amplified by polymerase chain reaction (PCR) and digested with two different restriction enzymes. This preliminary study provides evidence for more than five different alleles each of XA (YA) and XB (XB-S, YB), and at least 11 XA-XB haplotypes. Their association with extended HLA-B-DR haplotypes, C4 complotypes, and C4 region restriction fragment length polymorphisms (RFLP) is discussed. X gene polymorphisms are a complement region marker system that should be uniquely suited for PCR-based typing methods. They could become a useful addition to HLA class I and class II markers in the mapping of candidate genes for MHC-associated diseases, including the X and Y genes themselves.

TNFB polymorphisms characterize three lineages of TNF region microsatellite haplotypes.

The human major histocompatibility complex (MHC) contains a variety of genes, many of which are highly polymorphic and of immunological importance. A database of MHC extended haplotypes was used to integrate experimental, cell line, and population data. Three alleles of the human TNF-beta (lymphotoxin-alpha) gene were identified, named TNFB*1SL, TNFB*2LL, and TNFB*1LS, each representing a different lineage in the evolution of TNF region haplotypes. Lower variability in the length of the associated microsatellite alleles indicates that *1SL characterizes the youngest of the three haplotype lineages. Microsatellite haplotypes in the two older lineages show evidence for a coevolution of alleles through concerted expansions. Genetic predispositions to high and low TNF-alpha (cachectin) responses seem to have evolved independently in more than one lineage. The literature data suggest different, or even opposite, associations concerning the regulation of TNF-alpha in macrophages and lymphoid cells. Microsatellite ud may be the most informative marker for studies of the associations of individual TNF region markers with secretion levels, immunity, and disease.

Strategy for controlling preferential amplification and avoiding false negatives in PCR typing.

The use of the PCR method for routine testing has increased dramatically during recent years. Most assays involve co-amplification either of an internal control, of several alleles at a given locus or of a variety of bands produced by low-stringency primer annealing. In such multiplex reactions, certain products will often amplify preferentially. Amplimers that are more sensitive can be outcompeted under suboptimal PCR conditions, leading to assignment of false negatives. Optimization of PCR parameters such as temperature steps, relative concentrations of primers and their annealing temperature do not alone ensure against false negatives when caused by stable double-stranded DNA (dsDNA) regions in the amplified sequence. A two-step strategy to solve this problem is presented in this paper: (i) titration of the PCR with NaCl as a model inhibitor to establish the critical range within which false negatives occur; (ii) titration of the PCR with a dsDNA-destabilizing additive under false-negative-inducing conditions until the relative amplification efficiencies of co-amplified fragments are adjusted. Betaine is introduced as a novel and efficient cosolute. These measures to achieve reliable PCR typing of a difficult target should be useful for many qualitative and quantitative multiplex PCR applications.

Reactive arthritis-associated bacteria can stimulate lymphocyte proliferation in non-exposed individuals and newborns.

In reactive arthritis (ReA) a specific T cell response to the triggering bacterial antigen is present in the synovial fluid, while in paired peripheral blood T cells the response is markedly reduced. The proliferative response to ReA-associated bacteria in the peripheral blood of ReA patients was compared with that seen in the blood of healthy adults, who denied exposure to these microbes, and in the umbilical cord blood of newborns, who have clearly not been exposed to bacterial antigen. Peripheral blood mononuclear cells (PBMC) from non-exposed adults and those from umbilical cord blood proliferated to ReA-associated bacteria, whilst little response was seen in ReA PBMC. The response was MHC class II-restricted, required processing of the bacterial antigen, was seen in both CD45RO+ and CD45RA+ subsets, and was not oligoclonal. These T cell responses are similar to those previously demonstrated in non-exposed individuals to malaria, leishmania and trypanosoma antigen, and may reflect the existence of 'natural' T cell immunity to ReA-associated bacteria. The lack of such responses in ReA peripheral blood may suggest that such 'natural' responses may restrict the dissemination or progression of infection.