The autoimmune IgG subclass response defines the IgG deposition pattern in pemphigus patient skin

Pemphigus is an autoimmune blistering disease that is caused by autoantibodies to the extracellular domains of desmoglein (Dsg) 1 and/or 3 (Egami et al., 2020). Desmogleins are cadherin molecules that interconnect two half-desmosomes, ensuring stable binding between neighboring keratinocytes. When bound to desmoglein, these autoantibodies deposit in the stratified epithelial layers of skin and mucosa. Indirect immunofluorescence of healthy skin or mucosa sections shows a smooth and even desmoglein staining around cells, while direct immunofluorescence of the patient lesional skin shows granular IgG deposits to a greater or lesser extent ( Oktarina et al., 2011 Oktarina D.A.M. van der Wier G. Diercks G.F.H. Jonkman M.F. Pas H.H. IgG-induced clustering of desmogleins 1 and 3 in skin of patients with pemphigus fits with the desmoglein nonassembly depletion hypothesis. Br J Dermatol. 2011; 165: 552-562 Crossref PubMed Scopus (55) Google Scholar ). This varies from a subtle granular pattern in the basal layer to, in extreme cases, a total granular pattern throughout all epidermal layers. This granulation pattern is hypothesized to be caused by crosslinking of desmogleins. Using correlative electron microscopy it was observed that at the sites of the granules double-membrane structures had formed between cells in all forms of pemphigus ( Sokol et al., 2015 Sokol E. Kramer D. Diercks G.F.H. Kuipers J. Jonkman M.F. Pas H.H. et al. Large-scale electron microscopy maps of patient skin and mucosa provide insight into pathogenesis of blistering diseases. J Invest Dermatol. 2015; 135: 1763-1770 Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar ). In a minority of patients, however, clustering is absent, and the IgG is evenly deposited around the epithelial cells, which questions the crosslinking hypothesis. The autoimmune IgG response is dominated by IgG4 and can be accompanied by all of the other 3 subclasses ( Trémeau-Martinage et al., 1995 Trémeau-Martinage C. Oksman F. Bazex J Immunoglobulin G subclass distribution of anti-intercellular substance antibodies in pemphigus. Ann Dermatol Venereol. 1995; 122: 409-411 PubMed Google Scholar ). Of the four subclasses IgG4 is an outsider as it is the only one capable of Fab-arm exchange. Disulfide bonds between heavy chains are transiently reduced, resulting in half antibodies that randomly reform to new intact antibodies with other half-molecules of IgG4 ( Aalberse and Schuurman, 2002 Aalberse R.C. Schuurman J. IgG4 breaking the rules. Immunology. 2002; 105: 9-19 Crossref PubMed Scopus (384) Google Scholar ). It is estimated that about 99% of the IgG4 consists of these hybrid molecules ( Koneczny et al., 2017 Koneczny I. Stevens J.A.A. De Rosa A. Huda S. Huijbers M.G. Saxena A. et al. IgG4 autoantibodies against muscle-specific kinase undergo fab-arm exchange in myasthenia gravis patients. J Autoimmun. 2017; 77: 104-115 Crossref PubMed Scopus (90) Google Scholar ). These hybrid antibodies are no longer monospecific and bivalent but instead bispecific and monovalent. Such hybrid molecules therefore would be incapable of crosslinking desmoglein. IgG1, IgG2 and IgG3 are not capable of Fab-arm exchange therefore these antibodies remain bivalent and are capable of crosslinking. We investigated whether a difference in subclass response affects the IgG deposition in pemphigus patient skin. We hypothesized that a pure IgG4 response due to the molecules being monovalent would not crosslink desmogleins and therefore would lead to a smooth deposition pattern. We selected randomly fifteen skin biopsies with a partly granular IgG deposition pattern and added five more rare biopsies with a smooth IgG deposition pattern. This retrospective study was performed on redundant leftover material from diagnostics and in line with the guidelines of the University Medical Center Groningen ethics committee approved under study number 2021/298. According to institutional and national opt-out policies, separate written consent is not required. All biopsies were from both pemphigus vulgaris and pemphigus foliaceus patients, and we stained them for total IgG and for all the four subclasses. In all biopsies the IgG deposits were IgG4 positive, in addition two were also positive for IgG1, five for IgG2, seven for both IgG1 and IgG2 and in one biopsy all four subclasses were found. In the five biopsies with a smooth deposition, IgG4 was the only subclass found In the fifteen biopsies with a partly granular deposition, IgG1 and/or IgG2 and/or IgG3 were present in addition to IgG4. Typical examples are shown in Figure 1. Note that in the upper layers where IgG1 and IgG2 are not present the distribution of IgG is smooth again. The titers of IgG1 and IgG2 in contrast to IgG4 are not high enough to reach these layers. This is in line with IgG4 being the dominant response. Both IgG1 and IgG3 subclasses can effectively fix complement, whereas IgG2 and IgG4 have little or no complement fixation capacity. Therefore complement deposition can be expected if either IgG1 or IgG3 is present. As complement deposition will follow the distribution of IgG1 and IgG3 in patient skin it therefore is expected to be always present in a granular pattern. This is precisely what is found in daily practice. When IgG1 and/or IgG3 are absent no complement deposition is expected, and this is indeed seen in roughly fifteen to thirty percent of the cases. To study the effects of IgG subclass on cells we purified IgG from five sera that induced arrays by chromatography over protein G sepharose. Next we separated the IgG4 from the other subclasses by chromatography over CaptureSelect™ IgG4 (Hu) Affinity Matrix. After concentration we added the five IgG1, 2 and 3 mixtures and the five IgG4 samples to cells cultured under high calcium conditions. As shown in figure 2a adding IgG1, 2 and 3 resulted in a complete distortion of the Dsg3 distribution. Also, linear arrays formed perpendicular to the cell membrane (white arrows) as described before by Jennings et al. For IgG4 (Figure 2b) internalization was observed but the distribution of Dsg3 did not change. Figure 2IgG distribution after adding IgG subclass fractions to high calcium cultured cells. Show full caption a) After adding purified IgG1,2 and 3. Note the linear arrays (white arrows). b) After adding purified IgG4. White bar is 20 μm. View Large Image Figure Viewer Download Hi-res image a) After adding purified IgG1,2 and 3. Note the linear arrays (white arrows). b) After adding purified IgG4. White bar is 20 μm.