Immunohistology of the sacroiliac joints in SpA

Sacroiliitis is considered as a key clinical and diagnostic sign of SpA. However, there are only few studies of the immunohistological features of sacroiliac joint biopsies, which can be explained by the technical difficulties required to gain access to these anatomically complex structures.

In the early work by Braun et al., biopsies from five patients with active AS (computed tomography-guided needle biopsy) were studied. Dense cellular infiltrates containing T-cells and macrophages were revealed in the synovial portion of sacroiliac joints in all patients. CD3+, CD4+, CD8+ and CD14+ cells were diffusely present, but in some areas accumulation of cells was observed. CD3+ cells and CD14+ cells were found very close to the areas of new bone formation. In situ hybridisation of the messenger RNA (mRNA) showed abundant expression of the tumour necrosis factor α (TNF α) in these infiltrates, but no message for interleukin 1β. Message for transforming growth factor β (TGF β) was clearly detected close to the edge of new bone formation, but no TGF β mRNA was found in the cellular infiltrates .

These data were confirmed and complemented later in the study by Francois et al. showing large number of TNF α-positive, interleukin 6 and interleukin 10-positive mononuclear cells located within the bone marrow in sacroiliac joints biopsy samples obtained from patients with early active disease, and TGF β-positive cells in patients with advanced AS . Moreover, in patients with early disease, accumulation of CD3+ cells within bone marrow with formation of aggregates was observed. Macrophages were also seen in the bone marrow and in the marrow spaces of the sclerotic iliac bone. CD68+ osteoclasts were seen in areas of enchondral ossification .

Important data on correlation between immunohistological findings and changes seen in magnetic resonance imaging (MRI) were published by Bollow et al. In their work, active inflammatory lesions in sacroiliac joints seen on MRI had a morphological substrate in the form of inflammatory cell (T-cells and macrophages) infiltrates in the biopsy samples obtained from patients with SpA by means of computed tomography-guided needle biopsy . The number of inflammatory cells (T-cells, macrophages and, to a lesser extent, B-cells) was significantly higher in patients with active sacroiliitis on MRI (signal enhancement >70%) in comparison to patients with moderate activity (enhancement 30–70%). Interestingly, in patients with advanced chronic changes of sacroilliitis on MRI (severe subchondral sclerosis, partial or total ankylosis), the number of inflammatory cells was significantly reduced in comparison to patients with less advanced changes .

Immunohistology of the spine in SpA

There are also only a few immunohistological studies involving spinal structures in patients with AS, which can also be explained by technical difficulties in obtaining samples. Nearly exclusively, bone material has been obtained from spinal surgery. In the study by Appel et al., histological samples of zygapophyseal joints from eight AS patients were obtained, who underwent spinal surgery because of the spinal ankylosis with hyperkyphosis and had MRI available for analysis. In this study, a correlation was found between interstitial bone marrow oedema and mononuclear cell infiltrates (including CD4+, CD8+ and CD20+ lymphocytes) observed in histological samples and bone marrow oedema seen on MRI . However, in only three of eight patients, inflammatory lesions were seen by both immunohistology and MRI. Interestingly, two of these three patients had the highest histological score for bone marrow oedema. These data suggest that a high degree of bone marrow inflammation is necessary to be detected by MRI .

Importantly, active spinal inflammation is still present in a substantial proportion of AS patients with a long-standing disease and advanced ankylosis. Appel H et al. analysed the zygapophyseal joints of AS patients, who underwent spinal surgery because of spinal ankylosis with hyperkyphosis. Zygapophyseal joints of six of the 16 AS patients demonstrated the presence of two or more CD3+T-lymphocyte aggregates (cluster of ≥50 cells), which was considered as a sign of active persistent inflammation . In these patients, interstitial CD4+, CD8+T-cells and CD20+B cells were more frequent in comparison to non-AS controls, as well as to AS patients without lymphocytic aggregates. Moreover, micro-vessel density was significantly higher in AS patients in comparison to non-AS controls, indicating activation of the neoangiogenesis as a result of a chronic inflammation. At the same time, the numbers of macrophages and CD68+ osteoclasts did not differ significantly between the AS and the non-AS subgroups. Thus, CD4+, CD8+ and CD20+ lymphocytes seem to play an important role in the development of acute inflammatory lesions (osteitis/bone marrow oedema) in patients with AS.

Neidhart et al. immunohistochemically analysed bony and soft-tissue specimens of the spine of 30 patients with AS and 20 with degenerative disc disease . In patients with AS, cathepsin K-positive and matrix metalloproteinase (MMP)1-positive mononuclear cells, fibroblast-like cells and cells attached to bone (especially at cites of bone remodelling) were observed more frequently in spinal structures of patients with AS in comparison to non-AS controls. Since cathepsin K and MMP1 are collagenolytic proteases, their active expression suggests an increased activity of osteoclasts. Increased osteoclastic activity in patients with SpA could be responsible for structural changes associated with osteodestruction (i.e., erosions, spondylitis anterior and spondylodiscitis) in spinal structures. However, these findings do not explain the mechanisms causing excessive new bone formation in SpA that is characteristic for the disease progression.

In recent years, the Wingless (Wnt) pathway has become of great scientific interest as a possible mechanism that could play a central role in excessive osteoproliferation in patients with AS. The Wnt family of glycoproteins is a major signalling pathway involved in cellular differentiation. Wnt signalling is essentially involved in the differentiation of mesenchymal stem cells to osteoblast precursor cells and their differentiation into mature osteoblasts . Fine-tuning of this pathway is undertaken by several regulatory pathways at both the extracellular and the nuclear level. In the first group, secreted inhibitors of the Wnt pathway such as dickkopf-1 (DKK-1) and sclerostin have been identified, whereas, in the second group, intracellular proteins such as axin and adenomatous polyposis coli among many others are involved . Both DKK-1 and sclerostin act as Wnt inhibitors and block bone formation. Inhibition of these molecules could activate osteoproliferation and this might be relevant to the pathogenesis of AS. For instance, blockade of DKK-1 led to fusion of sacroiliac joints in mice transgenic for TNF . Sclerostin, produced by osteocytes, inhibits bone formation by inhibition of bone morphogenic protein (BMP)-stimulated osteogenesis, and sclerostin blockade results in increase of bone mass in animal models .

Appel et al. studied the skeletal expression of sclerostin (zygapophyseal joint samples) and its relationship to radiographic progression in patients with AS . Sclerostin expression by osteocytes in AS was significantly lower in AS patients ( n = 5) as compared with normal controls ( n = 5, samples obtained during autopsy from subjects without spinal disease), patients with rheumatoid arthritis (RA) ( n = 4) and osteoarthritis ( n = 5): the mean percentage of sclerostin positive osteocytes was 14.7% versus 53.2%, 57.7% and 42.2%, respectively. Interestingly, low sclerostin expression in AS was not based on osteocyte death, because, in contrast to RA, only a minority of cortical bone lacunae in patients with AS were empty. The serum level of sclerostin was also significantly lower in AS patients compared with healthy control subjects. However, even more interesting, the serum level of sclerostin was significantly lower in patients demonstrating formation of new syndesmophytes over a 2-year period in comparison to patients without syndesmophyte growth . However, it is not fully clear whether the low sclerostin level causes syndesmophyte formation or whether it is a bystander effect. Nevertheless, sclerostin serum level could play a role as a predictor of new syndesmophyte formation in patients with AS.