The human skin organ culture model as an optimal complementary tool for murine pemphigus models

Abstract

German

Spanish

French

Pemphigus is a severe autoimmune bullous disease of the skin and/or mucous membranes caused by autoantibodies that mainly target the adhesion proteins desmoglein (Dsg) 3 and/or Dsg1. Clinically, pemphigus is characterized by flaccid blistering, leading to severe water and electrolyte loss. Before the introduction of corticosteroid treatment, the disease turned out to be fatal in many cases. Despite recent therapeutic improvements, treatment of pemphigus patients is centred on prolonged systemic immunosuppression and remains challenging. Current drug development for pemphigus has a strong focus on disease-causing B cells and autoantibodies and, more recently, also on modulating autoantibody-induced tissue pathology and keratinocyte signalling. This drug development requires reliable pre-clinical model systems replicating the pathogenesis of the human disease. Among those are neonatal and adult mouse models based on the transfer of Dsg3, Dsg1/3 or Dsg1-specific autoantibodies. To reduce the number of animal experiments, we recently established a standardized human skin organ culture (HSOC) model for pemphigus. This model reproduces the clinical phenotype of autoantibody-induced tissue pathology in pemphigus vulgaris. For induction of blistering, a recombinant single-chain variable fragment (scFv) targeting both Dsg1 and 3 is injected into pieces of human skin (obtained from plastic surgeries). Further characterization of the HSOC model demonstrated that key morphologic, molecular and immunologic features of pemphigus are being replicated. Thus, the pemphigus HSOC model is an excellent alternative to pemphigus animal model systems that are based on the transfer of (auto)antibodies.

Keywords

Desmoglein human skin organ culture pemphigus 3R animal experimentation in vitro model systems

Introduction

Pemphigus is a group of chronic, rare, life-threatening, autoimmune bullous diseases of the mucosa and the skin.^1,2 Although mortality has declined significantly with the advent of new therapies such as rituximab,^3–6 patients still suffer from significant morbidity, treatment-related disabilities that reduce the quality of life and increased mortality.^7–9 Despite the recent advances in treatment, a significant proportion of patients do not achieve remission, experience severe adverse events during treatment and/or suffer from relapse under minimal or discontinued therapy.^10,11

Autoantibodies targeting two desmosomal proteins, desmoglein (Dsg) 1 and/or Dsg3, are in the majority of cases responsible for the etiopathogenesis of pemphigus, which results in various disease manifestations.^1,2 Anti-Dsg1 antibodies are the cause of pemphigus foliaceous (PF) that solely manifests on the skin. Anti-Dsg3 antibodies lead to development of the mucosal-dominant phenotype of pemphigus vulgaris (PV), while mucocutaneous PV is observed when both Dsg3 and Dsg1 are targeted by autoantibodies. The correlation of the antibody profile with the clinical phenotype is explained by the desmoglein compensation hypothesis.¹² This explanation is based on the differential expression patterns of Dsg3 and Dsg1 in mucosal and cutaneous epidermis, and the notion that either Dsg3 or Dsg1 alone is sufficient to maintain cell–cell adhesion. Also, in less than 5% of pemphigus patients, autoantibodies targeting other autoantigens, such as desmocollin 3 (Dsc3), can impair epithelial adhesion.^13–16 Hence, Dsg3 and Dsg1 are the main targets of the autoantibody response in pemphigus as confirmed by a wide array of research.²

Cumulative in vivo and in vitro evidence from mice and cultured monolayer keratinocyte cultures attributes a crucial role to antibody-induced Dsg-mediated pathogenic signalling in the regulation of desmosomal adhesion in PV^1,17 which is now also inferred from studies on PF.¹⁸ Specifically, binding of anti-Dsg3 autoantibodies to extradesmosomal Dsg3¹ sterically hinders Dsg transadhesion¹⁹ and is intimately linked to outside-in signal transmission.^1,17 By analogy to integrin signalling, the latter aligns with the thermodynamic principle of ligand-induced receptor activation and signal transmission through conformational changes.²⁰ Consistently, peptide cross-linking of Dsg3 abrogates antibody-induced loss of transadhesion and also prevents signal activation and blistering.²¹ Moreover, pharmacological intervention preventing pathogenic signalling, unless affecting tissue integrity,²² also prevents blistering despite concurrent disruption of Dsg3 transadhesion by anti-Dsg3 autoantibodies.²¹ In a nutshell, besides typical mechanotransducing signals, intervening with epidermal growth factor receptor and Wnt signalling and a variety of downstream effectors has been shown to interfere with the conversion to proliferative cells, shrinking of desmosomes and loss of cell–cell adhesion upon autoantibody binding.^1,23–25 Remarkably, methylprednisolone, a commonly used glucocorticoid and immunosuppressant in pemphigus, therapeutically counteracts PV antibodies by upregulating adhesion molecules both transcriptionally and translationally in keratinocytes.²⁶ These findings emphasize that modifying autoantibody-induced signalling is an innovative approach to counteract the consequences of antibody binding in terms of lesion formation.

The development of novel therapies in pemphigus in principle focuses on two main research areas: i) the characterization of pathogenic antibodies, their targets and pathogenic signalling, and ii) aspects of anti-Dsg B cell and antibody production^27,28 as well as B and T cell-mediated immune tolerance.^29,30 This ongoing drug development requires reliable pre-clinical model systems replicating the pathogenesis of the human disease. Among those are neonatal or adult mouse models based on the transfer of Dsg3, Dsg1/3 or Dsg1-specific autoantibodies. To reduce the number of animal experiments, we recently established a standardized human skin organ culture (HSOC) model for pemphigus. Here, we review pemphigus mouse models in the context of how well they reproduce the disease and contrast the advantages/disadvantages to established in vitro models, including the HSOC model of pemphigus.

Pemphigus mouse models

Model 1: neonatal passive transfer model

The so termed ‘passive transfer mouse model’ to study pemphigus was developed by Grant Anhalt and colleagues in 1982.³¹ It is based on intraperitoneal injection of total purified immunoglobulin G (IgG) isolated from human PV patients’ sera into neonatal mice (see Figure 1). Within 18–72 h after IgG transfer, suprabasal blisters develop, recapitulating the clinical phenotype of the human disease. This confirmed earlier work obtained in an organ culture model that pemphigus autoantibodies are pathogenic, and that these antibodies are sufficient to induce blisters.^32,33 In addition to the initially used neonatal BALB/cJ mice, various other mouse strains (including C57Bl/6J strain) develop experimental pemphigus after intraperitoneal or subcutaneous injection of IgG targeting Dsg3, Dsg3/Dsg1 or Dsg1. Furthermore, experimental antibodies such as the monoclonal Dsg3 mouse antibody AK23^34,35 as well as patients’ antibodies cloned by phage display³⁶ similarly induce skin and/or mucosal pathology in this model and allow to specifically study Dsg3- or Dsg3/Dsg1-mediated keratinocyte signalling responses. Variations of this model also demonstrated pathogenicity of Dsg1 autoantibodies that are the cause of PF,³⁷ as well as the pathogenicity of Dsg3 autoantibodies obtained from patients with paraneoplastic pemphigus.³⁸

Figure 1.

Different pemphigus mouse models recapitulate distinct steps of pemphigus pathogenesis. In the majority of patients, the pathogenesis of pemphigus vulgaris (PV) is caused by autoantibodies that target desmoglein 3 (Dsg3) or Dsg3 and Dsg1 (Dsg3/1). The production of autoreactive T and B cells leads to the presence of autoantibodies, which bind to its autoantigen Dsg3 or Dsg3/1, leading to uncoupling of transadhesion through steric hindrance and intracellular signalling, and ultimately to blister formation. 1) The passive transfer of patient immunoglobulin G (IgG) into neonatal mice of various genetic backgrounds replicates the clinical phenotype of PV by human pemphigus IgG or experimental antibodies. The phenotype of pemphigus foliaceus (PF) has also been replicated in this model using anti-Dsg1 antibodies. 2) Using adult C57Bl/6J mice specifically allows the study of the PV phenotype in skin appendages and stem cells. 3) The SCID/hu mouse model with engrafted human skin and the injection of pemphigus IgG allows the study of the disease in human skin. 4) The active transfer disease model using autoantigen-knockout mice or Dsg3 hybridomas allows to study the loss of tolerance observed in pemphigus patients. 5) The immunization of humanized HLA-DBR1*04:02 transgenic mice with recombinant Dsg3 or Dsc3 recapitulates the effector phase of T and B cells. Image created with BioRender.

The neonatal mouse model of pemphigus has been used to investigate pemphigus pathogenesis and drug effects for over four decades. It has significantly contributed to our understanding of pemphigus pathogenesis. This includes the identification of signalling cascades in Dsg1/3-targeted keratinocytes,^1,17 of which many have been identified in human pemphigus patients’ skin and mucous membranes.^39–42 Comparable signalling data on Dsg1 have also been published.^19,43,44 Mechanistically, the neonatal mouse model also served to underpin the ‘compensation hypothesis’, stating that disruption of Dsg3 function by anti-Dsg3 patients’ antibodies can be compensated by Dsg1 expression in PV and vice versa in PF.¹²

Model 2: adult passive transfer model

Although the neonatal mouse model served to discover key pathogenic pathways in PV or PF, the morphological skin development in these animals is not terminated at birth, and they lack skin appendages such as hair follicles. Hence, to study pemphigus pathogenesis in adult skin and/or oral mucosa, the ‘adult (12-week-old) passive transfer mouse model’ was developed using C57Bl/6J or Rag2^–/– mice³⁵ (see Figure 1). This allows to study adult tissue pathology involving stem cell compartments in PV as Dsg3 antibodies affect basal cells in the epidermis and in mucous membranes, as well as in hair follicles, where stem cells are known to reside. Published and unpublished results now suggest that extradesmosomal Dsg3,¹ the major antigenic target in PV, is a crucial mechanosensing signalling node surveilling quiescence of stem cells in their compartments and of committed progenitors in G0 of the cell cycle³⁹ (Hariton, Schulze et al., unpublished). The pathological response to loss of Dsg3 transadhesion implicates fate conversion of these keratinocytes to proliferative cells, also observed in pemphigus patients,^39,41,42 aiming to trigger tissue repair and restore Dsg3 function.

Model 3: SCID/hu passive transfer model

An alternative approach to study PV and PF in mice by passive transfer is based on intraepidermal injection of patient IgG into human skin grafted onto immunodeficient SCID mice (see Figure 1).⁴⁵ Within 24–48 h after IgG injection, grafts develop intraepidermal blisters, resembling the corresponding human disease. Although this model allows to investigate mechanisms of anti-human Dsg1/3-induced human tissue pathology, the overall procedure is labour- and time-intensive. Furthermore, complex surgical intervention in the mice is required, which again needs highly specialized personnel. After skin transplantation, several weeks are required for engraftment before patient IgG can be injected. These disadvantages may explain why this model has not been used after its initial description in 2001.

Model 4: active transfer models

Whilst the above-described mouse models duplicate aspects of autoantibody-induced clinical tissue pathology, immunization of autoantigen-knockout mice with subsequent intravenous transfer of reactive splenocytes into immunodeficient Rag2^–/– mice was used to develop an ‘active autoimmune model’ in which the loss of tolerance to the autoantigens in pemphigus is experimentally reproduced⁴⁶ (see Figure 1). This approach was taken because immunization of wild-type mice expressing Dsg3 does not lead to in vivo pathogenic autoantibodies.^46–48 Within a few days, autoantibody production is noted in the Rag2^–/– recipient mice, which are maintained without further immunizations for six months. While lesions develop within 7–14 days after transfer, weight loss was evident mainly due to blisters in the oral mucosa, which was severe in some mice and caused their death. Hair loss was observed within one more week. By histology, IgG depositions were observed in skin and mucosal biopsies while suprabasal blisters were restricted to hair follicles and mucosa but no epidermal blisters were present. Overall, lesions were apparent in 83% of the mice.⁴⁶ A similar phenotype can be induced by adoptive transfer of Dsg3 hybridoma cell lines^34,49 or of naïve splenocytes isolated from Dsg3-deficient mice without immunization but reduced penetrance.^50,51

This model system has been instrumental in extending our understanding of mechanisms leading to the loss of tolerance in pemphigus,⁵² including the characterization of the T cell responses and the function of Tregs.^29,53 However, systemic steroid treatment, the mainstay of pemphigus patient therapy,⁵⁴ had no effect on the clinical phenotype in this model.⁵⁵. Overall, the immunization-induced disease models reflect (auto)immunity towards Dsg3 and possibly Dsc3.¹⁵

In addition to the study of mechanisms leading to autoantibody maturation, these models are also suited to studying molecular mechanisms of autoantibody-induced tissue pathology. However, if timing of the insult on the keratinocytes is of importance, passive transfer models should be used because of the lag phase in autoantibody production that may blur the exact onset of disease and initial binding sites of antibodies. The broad use of the model is further hindered by the labour-intensive protocol, the relatively low disease penetrance, and the non-response to a standard treatment (corticosteroids). Furthermore, the severe and rapid weight loss that occurs in most of the mice with oral lesions imposes a significant burden on the animals.

Model 5: HLA-DBR1*04:02 transgenic mouse model

Another model to study antibody development in pemphigus is the humanized PV mouse model. Eming et al.⁵⁶ described a protocol in which DBA/J1 mice transgenic for HLA-DRB1*04:02 and HLA-DRB1*03:02 (which is in a linkage disequilibrium with DRB1*04:02) and the human CD4 co-receptor, which were devoid of functional murine major histocompatibility complex class II (I-Aβ(−/−), were immunized with human Dsg3 (see Figure 1). After immunization, these mice mounted a robust IgG response against human Dsg3. Of note, there was no evidence for tissue-bound IgG deposits in the oral mucosa of the mice, and consequently no clinical phenotype was observed. However, anti-human Dsg3 antibodies generated in this model can be used to test for pathogenicity using an in vitro model of pemphigus. Furthermore, a similar model has recently been established for atypical pemphigus: HLA-DRB1*04:02 mice immunized with human Dsc3 developed pathogenic autoantibodies which are able to bind to human keratinocytes.⁵⁷

Critical appraisal of pemphigus mouse models

In conclusion, pemphigus mouse models have been invaluable to elucidate mechanisms of Dsg3 autoantibody generation. They further provide in depth insights into highly complex signalling mechanisms of tissue pathology and blister formation in pemphigus. More recently, use of pre-clinical pemphigus mouse models demonstrated the possibility to selectively deplete autoreactive B cells.⁴⁹ All of the described mouse models (see Figure 1) duplicate certain aspects of human pemphigus and, as mentioned above, have significantly shaped our current understanding of pemphigus pathogenesis. These models are, however, not without disadvantages (see Table 1^{15,34,35,40,45,46,49–51,56–73}); foremost that the situation in patients may not be fully reflected in murine models. Hence, to overcome some of these limitations, as well as to replace animal experimentation, several in vitro models of pemphigus have been developed. Below, these models are described in detail, with special emphasis on the suitability to replace (some) animal experimentation in pemphigus research.

Table 1.

Pemphigus models covering different aspects of the disease pathogenesis.

Model no.	Model	Suited for	Limitations
1	Neonatal passive transfer model	• Replication of clinical, histological, and ultrastructural PV and PF phenotype• Timely defined onset of phenotype after injection	• Morphological skin development not terminated at birth, and hence mice lack skin appendages• No screening of immunomodulating agents possible• Injections need to be continued to keep phenotype over several weeks
2	Adult passive transfer model^35,40	• Replication of histological PV phenotype in oral mucosa and hair follicles• Investigation of longitudinal regeneration mechanisms• Study of skin appendages and stem cells possible	• High maintenance• No evaluation of the acquired immune response possible• No screening of immunomodulating agents possible• Injections need to be refreshed to keep phenotype in subsequent hair cycles
3	SCID/hu passive transfer model⁴⁵	• Replication of histological PV phenotype in epidermis• Investigation of the PV blistering response in human tissue in complete organism	• Labour and time intensive• Surgical intervention
4	Active transfer model^{15,34,46,49–51}	• Replication of histological PV phenotype in oral mucosa and hair follicles• Investigation of effector phase of autoimmune response, especially activation and interaction of T and B cells• Investigation of loss of tolerance in mice• Long and stable Dsg3-specific autoantibody response	• Two different mouse strains required• Labour and time intensive• Low disease penetrance without immunization (17%)• Limited suitability when studying tissue pathology
5	HLA-DBR1*04:02 transgenic mouse model^56,57	• Anti-human Dsg3 and Dsc3 autoantibody production• Investigation of effector phase of autoimmune response, especially activation and interaction of T and B cells	• No clinical phenotype in mice
6	Monolayer keratinocyte culture (human, mouse, dog)⁵⁷^–⁶³	• Study of PV tissue pathology in target keratinocytes: intercellular adhesion and signalling responses in keratinocytes• Screening of compound libraries	• Only one cell type; no complex tissues, no immune cells• Time point of IgG stimulation after calcium elevation is critical
7	Human skin organ culture (HSOC) model^24,64–⁷³	• Intact human skin preserving all layers and desmosome expression patterns• Tissue contains immune cells• Replication of histological PV phenotype• Easy to replicate model when using injection of experimental antibodies with standardized production• Easy evaluation	• No blood flow• Dependent on plastic surgeries• Skin needs to be used preferentially from plastic surgery no later than 18 h after surgery• Our HSOC model was not yet tested in other laboratories

Dsc3: desmocollin; Dsg3: desmoglein 3; IgG: immunoglobulin G; PV: pemphigus vulgaris; PF: pemphigus foliaceous.

Pemphigus in vitro models

Model 6: monolayer keratinocyte cultures

Cultured keratinocytes have been invaluable to unravel morphological changes and molecular mechanisms of pemphigus tissue pathogenesis, even before the identification of the major PV antigen Dsg3 in 1991.⁷⁴ Primary epidermal monolayer keratinocytes isolated from skin of humans, mice and other mammals comprise stem and progenitor cells and recapitulate the early differentiation processes in the epidermis upon elevation of the calcium concentrations in the culture medium to around 1.2 mM.^75,76 Consistent with the clinical location of the lesions in the stratified epithelium,^1,2 first analyses of committed progenitor cells or early differentiated mouse and canine keratinocytes highlighted the expression of PV, but not PF, antigens.^58,59 Accordingly, incubation of these cells with PV IgG allowed to define major hallmarks of PV already three decades ago. Within a time frame of 12–24 h, these hallmarks are widening of intercellular spaces, internalization of PV antibodies into sub-membranous vesicles, keratin retraction and disrupted desmosomes followed by acantholysis (loss of intercellular adhesion).^60–62 Based on these studies, Jones and colleagues proposed in 1986 that the PV IgG antigen is a cell adhesion molecule of desmosomes.⁶³ The finding, however, that PV IgG also binds to the plasma membrane outside of desmosomes, revealed by indirect IgG labelling and electron microscopy on biopsies from bovine tongue epidermis, canine oesophagus or cultured epidermal mouse keratinocytes,^62,63,77 suggested already at the time that the PV antigen also has extradesmosomal functions.¹

Today, monolayer keratinocyte cultures remain a central tool in the quest of defining mechanisms of tissue-based pathologies leading to loss of cell–cell adhesion in pemphigus (see Figure 2^59,62,78). In this respect, both the choice of cell types and culture conditions are to be considered and will be discussed below.

Figure 2.

Keratinocyte dissociation assay. 1) Keratinocytes at 80% confluence are expanded (left) and enter the dissociation assay at 100% confluence (right). 2) Experimental outcome after applying mechanical stress: control sheet (left) and fragments generated after AK23 antibody incubation (right). 3) Overview of the keratinocyte dissociation assay. The keratinocyte dissociation assay, developed by Calautti et al.⁷⁸ to test the strength of E-cadherin-mediated cell–cell adhesion, was adapted to assess desmosomal adhesive strength for studies related to the autoimmune disease pemphigus vulgaris.^59,62 The approach relies on semi-controlled mechanical stress exerted on the monolayer sheet enzymatically detached by dispase to semi-quantitatively assess intercellular adhesive strength between keratinocytes. Note that after thawing, keratinocytes are cultured for one passage (I) and seeded at higher density at the next passage for the experiment (II). At 100% confluence, it is crucial to increase calcium to 1.2 mM to stabilize intercellular adhesion (III) and wait at least 6 h before exposing the keratinocytes to pemphigus vulgaris (PV) antibodies (IV). The cells are then detached with dispase II and (V) semi-controlled mechanical stress is applied by pipetting the sheet 10x up and down using a 1 ml pipette (VI). Finally, the keratinocytes are fixed (VII), stained (VIII) and fragments are counted using, for example, ImageJ (IX). Image created with BioRender.

Regarding the choice of cells, primary keratinocytes are better suited than transformed keratinocyte cell lines, and epidermal keratinocytes isolated from human skin are in principle preferred over those from mouse or other mammals. Specifically, immortalized cell lines have often acquired mutations as they have been passaged over long periods of time and may have lost the ability to coordinately activate specific cellular signalling networks. For example, early committed keratinocyte progenitors (6 h after calcium switch) revert to proliferative cells in presence of PV IgG or experimental antibodies.³⁹ An increased ratio of Ki67 positive (proliferative) basal cells has also been documented in human and canine PV patients.^41,42 However, transformed cells such as HaCaT cells, double knock-out for p53 besides chromosomal aberrations, might not fully recapitulate the in vivo situation, as p53 is critical for cell cycle exit.⁷⁹ Moreover, according to our investigations, primary mouse and canine epidermal keratinocytes exhibit a comparable response pattern to primary human keratinocytes when challenged with PV IgG^39,77 and hence can be used for specific approaches such as studies under knock-out conditions. Primary keratinocytes from all species are isolated using similar, amply described protocols but can also be commercially acquired; human foreskin and also adult human skin are good sources for these cells; the same accounts for skin from E17 to E18 mouse embryos or adult mice⁸⁰ as well as adult hairy or non-hairy canine skin.⁸¹ It is conceivable, however, that some of the patients’ IgG might not bind mouse or canine Dsg3 or Dsg1.

With regard to culture conditions, it cannot be overstated that controlled conditions concerning seeding density, antibody stimulation and time to differentiation are required to obtain reliable and reproducible results. Furthermore, chemically defined media without foetal calf serum (which introduces batch-to-batch variations including varying levels of IgG and hormones) are available today and yield superior reproducibility when studying morphological and signalling responses. Moreover, purified IgG from patients’ sera or defined experimental, pathogenic antibodies such as AK23³⁵ are recommended to avoid side effects from contaminating blood components which are usually not reaching the epidermis, at least not to the same extent as in cell culture. Last, the time point of IgG stimulation after calcium elevation is critical and needs harmonization; already historically some groups choose a 6 h time point before adding the antibodies to primary keratinocytes,⁷⁵ when intermediate filaments are visibly attached to the plasma membrane, while others used 24 h.⁸² As mentioned above, the phase after the calcium switch marks the transition between proliferative and fully committed progenitor cells during which desmosome precursors are being stabilized.⁸¹ The latter includes a continuous change in Dsg3 associated effector molecules, such as the recruitment of desmoplakin, likely interfering with the signalling network that Dsg3 is able to activate. In summary, cell types, media, and keratinocyte culture conditions are to be carefully defined and implemented when studying pemphigus pathology. If possible, we also encourage the confirmation of hypotheses on pathogenic events established in cell culture through validation on biopsies from uninvolved skin of PV patients.

Like in any other disease, the most important aspect in pemphigus is to demonstrate causality: causality of Dsg and adjunct autoantibodies, immune responses, morphological changes and signalling events. As the main phenotype in vitro, the loss of intercellular adhesion or acantholysis, is easily recapitulated and can be quantified, keratinocyte cultures are well suited to address these questions by including inhibitors and activators, siRNA knock-down, knock-out cells, et cetera. The main tool to define the extent of loss of adhesion is the keratinocyte dissociation assay, which applies semi-controlled mechanical stress on the monolayer sheet which was enzymatically detached by dispase. This assay was first developed by Calautti and collaborators to test the strength of E-cadherin-mediated cell–cell adhesion⁷⁸ and was then adapted by our group for analyses on pemphigus.^83,84 Over the many years since its implementation, experimental conditions have diverged and results are not necessarily compatible. Efforts are now being made to harmonize the keratinocyte dissociation assay as recently initiated by the group around Prof. Yazdi.⁸⁵

Besides loss of cell–cell adhesion, other in vitro read-outs have been routinely used in pemphigus which may or may not be causative for pathogenicity. Examples are antigen clustering, which was shown to represent a bystander effect using Fab fragments in mouse keratinocyte cultures⁸⁴ and mouse models,³¹ Dsg3 endocytosis, evidenced by surface biotinylation, with consequences on desmosome disassembly late in the process,⁸⁶ and finally keratin retraction, one of the most sensitive read-outs in early acantholysis observed by electron microscopy or immunofluorescence microscopy.^87,88 The latter is less striking when keratinocytes are exposed to AK23 or PV IgG comprising only anti-Dsg3 antibodies rather than antibodies targeting Dsg3 and Dsg1.

Taken together, research performed on monolayer epidermal keratinocyte cultures has been invaluable to define tissue pathology in pemphigus. This understanding has opened new routes to propose therapeutic interventions in PV to directly targeting keratinocyte signalling and preventing antibody-induced tissue pathogenesis. However, to potentially screen and pre-clinically validate novel drugs, a next generation of a reliable human skin model is required which comprises a full human stratified epithelium including the immune, dermal and adipose components.

Model 7: the HSOC model: a human tissue approach to pemphigus vulgaris

As outlined above, a reliable human model is allowing to expand the understanding of tissue pathology in pemphigus and, in the future, eventually tie in immunological aspects of peripheral immunity with regard to B or T cell maturation. Furthermore, an accurate and convenient model to screen novel therapeutic drugs is also missing. Hence, we developed and characterized a new pemphigus model which is reliable, manageable and concomitantly responds to the 3R principles:⁶⁴ a HSOC model for pemphigus implicating the use of a bivalent PV antibody cloned by phage display.⁸⁹ As the name indicates, the HSOC in vitro model consists of a full thickness human skin explant culture which reliably develops typical PV lesions within 12–24 h after injection of a well-defined, recombinant single chain variable fragment cloned from PV patients targeting Dsg3 and Dsg1 (e.g. PX43, also termed (D31)2/29); US Patent US 8,298,545 B2).^36,64,90

One of the first skin organ culture models to study blistering (acantholysis) in pemphigus was established in 1977.⁶⁵ Skin, removed under local anaesthesia from healthy volunteers, was placed on lens paper floating on the surface of culture medium containing foetal calf serum and pooled human pemphigus serum. The binding of IgG to the keratinocyte cell surface was confirmed by direct immunofluorescence microscopy and split formation after 24 h by conventional histology.⁶⁵ As mentioned above, this study suggested for the first time that acantholysis is independent of complement following heat inactivation of the serum.

In this type of organ culture, the pemphigus IgG must diffuse into the skin piece, which occurs within 6–12 h, hampering synchronized onset of tissue pathology. This method was in the following abundantly used to study morphological and ultrastructural changes in presence of serum, plasma or purified IgG from PV or PF patients, attesting high similarity between the organ culture model and the patient with regard to the acantholytic processes.^66–71 Moreover, proteases could be assigned to the process and a remarkable direct effect of methylprednisolone on keratinocytes was reported, as well as the presence of apoptotic factors in serum or plasma.

Later, this skin organ culture model was successfully modified by Van der Wier, Pas and Jonkman.⁷² They placed the skin in a transwell at the air–liquid interface where only the bottom of the skin biopsy was in contact with the solution containing IgG. A similar effect was obtained when several biopsies were submerged in a solution, simplifying harvesting. The authors further confirmed that cadherin expression was unchanged for up to 24 h in the skin explant as compared with the in vivo situation,⁷² suggesting that this model is suitable for studies on onset of tissue pathology in pemphigus.

The skin organ culture model developed by Waschke et al.⁹¹ utilized the skin from cadavers of the human body donor programme; the biopsies from donors without history of skin disease were acquired within 24 h after death. However, the expression of cadherins other than Dsg3 was not monitored. To better represent the situation in patients, an automated shear stress was, however, applied to the skin samples. Remarkably, instead of letting pemphigus IgG diffuse through the biopsy, the IgG was intradermally injected. Furthermore, for the first time, AK23, a monoclonal Dsg3 antibody,³⁵ was employed, requiring, however, co-injection of exfoliative toxin A (ETA)⁹² to digest Dsg1 and achieve acantholysis.⁹³ This model was recently also used to investigate mucosal PV where human epidermis was replaced by human oral mucosa.⁹³ The distinct advantage of this model is that AK23 alike mucosal-dominant PV IgG is sufficient to induce lesions.⁹⁴

Intradermal injection instead of diffusion has the advantage to better synchronize the exposure of the keratinocyte to the antibody and study molecular mechanisms of tissue pathology. Furthermore, using monoclonal, target-specific antibodies instead of polyclonal PV IgG from patients has several advantages: monoclonal antibodies have a well-characterized specificity, results achieved have a greater reproducibility and the availability of antibodies is not limited. Patient IgG preparations vary between individuals⁶⁴ and the heterogeneity in patient IgG leads to inconsistent split formation which is standardized with monospecific antibodies such as AK23³⁵ or PX43.⁶⁴

The most recent, robust and reliable HSOC model for PV, as judged from biostatistical consistency, has been developed by our group.⁶⁴ We introduced three major adaptations to the procedure: first, a cloned, bi-specific anti-Dsg 3 and Dsg1 single chain antibody variable fragment (scFv) called PX43 (also known as (D31)2/29³⁶) is used instead of IgG from pemphigus patients or the monospecific AK23 with ETA. The scFv was cloned by antibody phage display from an active pemphigus patient, is well defined and does not require the addition of ETA.^36,64,90 The use of a recombinant scFv allows standardized production in high amounts and high, consistent quality. Using PV patient IgG in turn can lead to a high variability in the extent of split formation, as pathogenic activities of IgGs vary substantially between patients. However, the scFv derived from phage display is produced from the same original stock in unlimited amounts. PV patients’ IgG is a finite source, and it can be difficult to impossible to get the same patient serum used in a previous HSOC. Second, the recombinant antibody directed against Dsg3/1 is injected into the skin between dermis and epidermis instead of being added to the culture medium (see Figure 3). Third, fresh human skin from plastic surgeries was utilized instead of skin from recently deceased individuals. This skin is not older than 18 h when injected with the scFv. Therefore, the only variation in the HSOC model comes from the skin donors. Key steps of this procedure can be viewed at https://www.youtube.com/watch?v=mLYwoJHqC34.

Figure 3.

The human skin organ culture for pemphigus. 1) Human full-thickness skin is acquired at the latest 18 h after plastic surgery. The fat is trimmed off and the skin cut into pieces of 1 cm × 1 cm size. The scFv or control antibodies are injected in a defined volume between dermis and epidermis and (2) the skin transferred into a transwell in a six-well plate. 3) Incubation of the injected skin pieces occurs under standard cell culture conditions (5% CO₂, 37°C) for up to 24 h. 4) After incubation, the skin piece can be separated into two or more pieces for (5) H&E staining, immunofluorescence staining or any other analysis. Image created with BioRender.

This HSOC model can also be modified for pre-clinical compound testing. Substances to be tested can be pre- or co-injected together with the scFv.²⁵ The therapeutic potential is then conveniently determined by comparing the percentage of epidermal split formation along the length of the skin section in conventional histology. Furthermore, prior knowledge of the therapeutic target is not necessary, allowing to also test compounds identified in pre-screening of libraries of active substances in monolayer cultures. If a desired effect is observed, new knowledge will further be obtained on pathways with functional relevance for the disease and for the treatment.²⁵ Moreover, after the high-throughput screen of a compound library, the substances can be tested afterwards in or on human skin, depending on the set-up. It is also possible to test the potential drugs by using creams for drug delivery to the target area. Besides screening drugs for first line treatment of pemphigus blisters, other possible set-ups could be to test the effect of PV antibodies on skin pigmentation, wound healing or the analysis of hair follicles.

It is important to note that the use of skin from different donors is not a disadvantage; quite the opposite is true: if the same experimental outcome can be seen although the skin donor varies, this emphasizes the power of the HSOC model. One limitation is the lack of a systemic immune response. The HSOC model does not have any blood flow and no connection to primary or secondary lymphoid organs. The immune response in the HSOC model for pemphigus comes from the still residual immune cells only. Another possible limitation might be the sex of the patient undergoing plastic surgery. So far, in our lab the majority of skin donors have been female with only few male donors (41 females vs. 1 male). This leads to a study bias which cannot be influenced by the study coordinators nor the experimenters but has to be considered when analysing the data.

The HSOC model for pemphigus can be further improved on optical coherence tomography with an automated detection of split formation. Moreover, the HSOC model is currently further validated in our research consortium by directly comparing it with keratinocyte cultures and mouse data based on transcriptomic, proteomic, phosphoproteomic and epigenomic alterations in response to autoantibody binding to Dsg3 and Dsg3/1. Taken together, the proposed HSOC model for pemphigus has several advantages over other pemphigus models (see Table 1): up to now, invaluable insights into pathomechanism of acantholysis have been gathered using this model.^64,65 It has already been used to test old and discover potentially new therapeutic drugs for pemphigus,²⁵ before mouse models have to be utilized. This effectiveness minimizes the number of mice used for drug testing prior to clinical trials. The fact that it is actual human skin with the correct architecture of all epidermal layers is a major advantage over analyses in monolayer keratinocyte cultures; it allows to study onset of acantholysis in the context of intact skin, including layer-specific changes in morphology or protein localization.⁹⁵ And finally, the role of the keratinocyte in shaping peripheral immunity could potentially be investigated in this model.

Conclusion

Human skin explant models are used for a variety of applications including the testing of topical therapeutics, and are attracting growing attention to replace animal testing.⁹⁶ In pemphigus research, there is no doubt that animal (mouse) models have played an essential role to elucidate mechanisms of peripheral tolerance resulting in anti-Dsg B cell production. Furthermore, the mouse models provided deep insights into a complex signalling network in keratinocytes underlying tissue pathology in pemphigus. We are certain that animal research will continue to provide critical insights into pemphigus pathogenesis and the pre-clinical development of novel therapeutic options. However, the priority pre-clinical use of a highly standardized, reproducible and complex in vitro model, the HSOC model for pemphigus, will significantly reduce the number of mouse experiments in such studies. Other advantages of the HSOC model compared with mouse models are the human origin, faster and cost-effective experiments (no mouse holding, less antibody used) and it has been set up to work with characterized antibodies. Hence, for pre-clinical studies investigating autoantibody-induced tissue damage in pemphigus, the HSOC model must be considered as the first-line option method. This will further allow the gathering of new insights when using methodologies at the frontier of science and novel screening applications.

While this manuscript was under consideration, another review on pemphigus disease models was published by Lotti and colleagues⁹⁷ which is complementary to this article. While Lotti et al.⁹⁷ gives an in-depth overview over mouse models used in pemphigus, the current review briefly mentions the in vivo models as a basis to focus on in vitro and ex vivo models with detailed insights into the experimental set-up, their use, and previous and current gain of knowledge in pemphigus as well as future opportunities.

Footnotes

Acknowledgement

We thank Dr John R Stanley, University of Pennsylvania, Philadelphia, PA, USA, for sharing reagents such as PX43 and corresponding sequences.

Data availability statement

Not applicable.

Declaration of conflicting interests

EJM is the president of the Board of the cell culture company CELLnTEC advanced cell systems AG; no financial payments were received. All other authors have no conflict of interest to declare.

Ethics statement

Our study did not require an ethical board approval because it did not contain human or animal trials.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Swiss National Science Foundation (Sinergia CRSII5_202301/1); the Cluster of Excellence Precision Medicine in Chronic Inflammation (EXC 2167) and the Research Training Group Autoimmune Pre-Disease (GRK 2633), all from the Deutsche Forschungsgemeinschaft (DFG); and the Schleswig-Holstein Excellence-Chair Program from the State of Schleswig Holstein.

ORCID iD

Veronika Hartmann

References

Di Zenzo

Amber

Sayar

, et al. Immune response in pemphigus and beyond: Progresses and emerging concepts. Semin Immunopathol 2016; 38: 57–74.

Hammers

Stanley

JR.

Mechanisms of disease: Pemphigus and bullous pemphigoid. Annu Rev Pathol 2016; 11: 175–197.

Shimanovich I. Baumann T. Schmidt E. Zillikens D. Hammers C M. Long-term outcomes of rituximab therapy in pemphigus. J Eur Acad Dermat Venereol 2020; 34(12): 2884–2889. 10.1111/jdv.16561.

Tovanabutra

Bax

Feng

, et al. Temporal outcomes after rituximab therapy for pemphigus vulgaris. J Invest Dermatol 2022; 142: 1058–1064.e7.

Schmidt

Kasperkiewicz

Joly

Pemphigus. Lancet 2019; 394: 882–894.

Werth

Joly

Mimouni

, et al. Rituximab versus mycophenolate mofetil in patients with pemphigus vulgaris. N Engl J Med 2021; 384: 2295–2305.

Ghodsi

Asadi

Ghandi

, et al. Family impact of pemphigus disease in an Iranian population using the Family Dermatology Life Quality Index. Int J Womens Dermatol 2020; 6: 409–413.

Ghodsi

Chams-Davatchi

Daneshpazhooh

, et al. Quality of life and psychological status of patients with pemphigus vulgaris using Dermatology Life Quality Index and General Health Questionnaires. J Dermatol 2012; 39: 141–144.

Ujiie

Rosmarin

Schön

, et al. Unmet medical needs in chronic, non-communicable inflammatory skin diseases. Front Med (Lausanne) 2022; 9: 875492.

10.

Daneshpazhooh

Zafarmand Sedigh

Balighi

, et al. Immunologic prediction of relapse in patients with pemphigus vulgaris (PV) in clinical remission. J Am Acad Dermatol 2016; 74: 1160–1165.

11.

Gheisari

Faraji

Dadras

, et al. Methylprednisolone pulse therapy plus adjuvant therapy for pemphigus vulgaris: An analysis of 10 years’ experience on 312 patients. Dermatol Ther 2019; 32: e13057.

12.

Mahoney

Wang

Rothenberger

, et al. Explanations for the clinical and microscopic localization of lesions in pemphigus foliaceus and vulgaris. J Clin Invest 1999; 103: 461–468.

13.

Chernyavsky

Amber

Agnoletti

, et al. Synergy among non-desmoglein antibodies contributes to the immunopathology of desmoglein antibody-negative pemphigus vulgaris. J Biol Chem 2019; 294: 4520–4528.

14.

Kalantari-Dehaghi

Anhalt

Camilleri

, et al. Pemphigus vulgaris autoantibody profiling by proteomic technique. PLoS One 2013; 8: e57587.

15.

Lotti

Atene

Marconi

, et al. Development of a desmocollin-3 active mouse model recapitulating human atypical pemphigus. Front Immunol 2019; 10: 1387.

16.

Sinha

Sajda

The evolving story of autoantibodies in pemphigus vulgaris: Development of the ‘Super Compensation Hypothesis’. Front Med (Lausanne) 2018; 5: 218.

17.

Müller

Williamson

Kolly

, et al. Outside-in signaling through integrins and cadherins: A central mechanism to control epidermal growth and differentiation? J Invest Dermatol 2008; 128: 501–516.

18.

Godsel

Roth-Carter

Koetsier

, et al. Translational implications of Th17-skewed inflammation due to genetic deficiency of a cadherin stress sensor. J Clin Invest 2022; 132: e144363.

19.

Heupel

W-M

Zillikens

Drenckhahn

, et al. Pemphigus vulgaris IgG directly inhibit desmoglein 3-mediated transinteraction. J Immunol 2008; 181: 1825–1834.

20.

Campbell

Humphries

MJ.

Integrin structure, activation, and interactions. Cold Spring Harb Perspect Biol 2011; 3: a004994.

21.

Spindler

Rötzer

Dehner

, et al. Peptide-mediated desmoglein 3 crosslinking prevents pemphigus vulgaris autoantibody-induced skin blistering. J Clin Invest 2013; 123: 800–811.

22.

Sayar

Rüegg

Schmidt

, et al. EGFR inhibitors erlotinib and lapatinib ameliorate epidermal blistering in pemphigus vulgaris in a non-linear, V-shaped relationship. Exp Dermatol 2014; 23: 33–38.

23.

Galichet

Borradori

Müller

EJ.

A new light on an old disease: Adhesion signaling in pemphigus vulgaris. J Invest Dermatol 2014; 134: 8–10.

24.

Schmitt

Waschke

Autoantibody-specific signalling in pemphigus. Front Med (Lausanne) 2021; 8: 701809.

25.

Burmester

IAK

Flaswinkel

Thies

C-S

, et al. Identification of novel therapeutic targets for blocking acantholysis in pemphigus. Br J Pharmacol 2020; 177: 5114–5130.

26.

Nguyen

Arredondo

Chernyavsky

, et al. Pemphigus vulgaris IgG and methylprednisolone exhibit reciprocal effects on keratinocytes. J Biol Chem 2004; 279: 2135–2146.

27.

Ellebrecht

Maseda

Payne

AS.

Pemphigus and pemphigoid: From disease mechanisms to druggable pathways. J Invest Dermatol 2021; 142(3 Pt B): 907–914. 10.1016/j.jid.2021.04.040. 34756581

28.

Didona

Maglie

Eming

, et al. Pemphigus: Current and future therapeutic strategies. Front Immunol 2019; 10: 1418.

29.

Iriki

Takahashi

Wada

, et al. Peripheral tolerance by Treg via constraining OX40 signal in autoreactive T cells against desmoglein 3, a target antigen in pemphigus. Proc Natl Acad Sci U S A 2021; 118: e2026763118.

30.

Mestrallet

Rouas-Freiss

LeMaoult

, et al. Skin immunity and tolerance: Focus on epidermal keratinocytes expressing HLA-G. Front Immunol 2021; 12: 772516.

31.

Anhalt

Labib

Voorhees

, et al. Induction of pemphigus in neonatal mice by passive transfer of IgG from patients with the disease. N Engl J Med 1982; 306: 1189–1196.

32.

Schiltz

Michel

Production of epidermal acantholysis in normal human skin in vitro by the IgG fraction from pemphigus serum. J Invest Dermatol 1976; 67: 254–260.

33.

Anhalt

Till

Diaz

, et al. Defining the role of complement in experimental pemphigus vulgaris in mice. J Immunol 1986; 137: 2835–2840.

34.

Tsunoda

Ota

Aoki

, et al. Induction of pemphigus phenotype by a mouse monoclonal antibody against the amino-terminal adhesive interface of desmoglein 3. J Immunol 2003; 170: 2170–2178.

35.

Schulze

Galichet

Sayar

, et al. An adult passive transfer mouse model to study desmoglein 3 signaling in pemphigus vulgaris. J Invest Dermatol 2012; 132: 346–355.

36.

Payne

Ishii

Kacir

, et al. Genetic and functional characterization of human pemphigus vulgaris monoclonal autoantibodies isolated by phage display. J Clin Invest 2005; 115: 888–899.

37.

Hacker-Foegen

Janson

Amagai

, et al. Pathogenicity and epitope characteristics of anti-desmoglein-1 from pemphigus foliaceus patients expressing only IgG1 autoantibodies. J Invest Dermatol 2003; 121: 1373–1378.

38.

Amagai

Nishikawa

Nousari

, et al. Antibodies against desmoglein 3 (pemphigus vulgaris antigen) are present in sera from patients with paraneoplastic pemphigus and cause acantholysis in vivo in neonatal mice. J Clin Invest 1998; 102: 775–782.

39.

Williamson

Raess

Caldelari

, et al. Pemphigus vulgaris identifies plakoglobin as key suppressor of c-Myc in the skin. EMBO J 2006; 25: 3298–3309.

40.

Luyet

Schulze

Sayar

, et al. Preclinical studies identify non-apoptotic low-level caspase-3 as therapeutic target in pemphigus vulgaris. PLoS One 2015; 10: e0119809.

41.

Williamson

Suter

Olivry

, et al. Upregulation of c-Myc may contribute to the pathogenesis of canine pemphigus vulgaris. Vet Dermatol 2007; 18: 12–17.

42.

Williamson

Hunziker

Suter

, et al. Nuclear c-Myc: A molecular marker for early stage pemphigus vulgaris. J Invest Dermatol 2007; 127: 1549–1555.

43.

Hammers

Stanley

JR.

Desmoglein-1, differentiation, and disease. J Clin Invest 2013; 123: 1419–1422.

44.

Harmon

Simpson

Johnson

, et al. Desmoglein-1/Erbin interaction suppresses ERK activation to support epidermal differentiation. J Clin Invest 2013; 123: 1556–1570.

45.

Zillikens

Schmidt

Reimer

, et al. Antibodies to desmogleins 1 and 3, but not to BP180, induce blisters in human skin grafted onto SCID mice. J Pathol 2001; 193: 117–124.

46.

Amagai

Tsunoda

Suzuki

, et al. Use of autoantigen-knockout mice in developing an active autoimmune disease model for pemphigus. J Clin Invest 2000; 105: 625–631.

47.

Fan

Memar

McCormick

, et al. BALB/c mice produce blister-causing antibodies upon immunization with a recombinant human desmoglein 3. J Immunol 1999; 163: 6228–6235.

48.

Boch

Dräger

Zillikens

, et al. Immunization with desmoglein 3 induces non-pathogenic autoantibodies in mice. PLoS One 2021; 16: e0259586.

49.

Ellebrecht

Bhoj

Nace

, et al. Reengineering chimeric antigen receptor T cells for targeted therapy of autoimmune disease. Science 2016; 353: 179–184.

50.

Holm

Markholst

Confirmation of a disease model of pemphigus vulgaris: Characterization and correlation between disease parameters in 90 mice. Exp Dermatol 2010; 19: e158-165.

51.

Aoki-Ota

Tsunoda

Ota

, et al. A mouse model of pemphigus vulgaris by adoptive transfer of naive splenocytes from desmoglein 3 knockout mice. Br J Dermatol 2004; 151: 346–354.

52.

Kasperkiewicz

Ellebrecht

Takahashi

, et al. Pemphigus. Nat Rev Dis Primers 2017; 3: 17026.

53.

Takahashi

Kouno

Nagao

, et al. Desmoglein 3-specific CD4+ T cells induce pemphigus vulgaris and interface dermatitis in mice. J Clin Invest 2011; 121: 3677–3688.

54.

Joly

Horvath

Patsatsi

, et al. Updated S2K guidelines on the management of pemphigus vulgaris and foliaceus initiated by the European Academy of Dermatology and Venereology (EADV). J Eur Acad Dermatol Venereol 2020; 34: 1900–1913.

55.

Takae

Nishikawa

Amagai

Pemphigus mouse model as a tool to evaluate various immunosuppressive therapies. Exp Dermatol 2009; 18: 252–260.

56.

Eming

Hennerici

Bäcklund

, et al. Pathogenic IgG antibodies against desmoglein 3 in pemphigus vulgaris are regulated by HLA-DRB1*04:02-restricted T cells. J Immunol 2014; 193: 4391–4399.

57.

Hudemann

Maglie

Llamazares-Prada

, et al. Human desmocollin 3-specific IgG antibodies are pathogenic in a humanized HLA Class II transgenic mouse model of pemphigus. J Invest Dermatol 2022; 142: 915–923.e3.

58.

Stanley

Yuspa

SH.

Specific epidermal protein markers are modulated during calcium-induced terminal differentiation.

J Cell Biol 1983; 96: 1809–1814.

59.

Wilkinson

Lee

Lillie

, et al. Ultrastructure of cultured canine oral keratinocytes. Am J Vet Res 1989; 50: 1161–1165.

60.

Woo

Hogan

Patel

, et al. Specificity and inhibition of the epidermal cell detachment induced by pemphigus IgG in vitro. J Invest Dermatol 1983; 81: 115s–121s.

61.

Jones

Arnn

Staehelin

, et al. Human autoantibodies against desmosomes: Possible causative factors in pemphigus. Proc Natl Acad Sci U S A 1984; 81: 2781–2785.

62.

Patel

Diaz

Anhalt

, et al. Demonstration of pemphigus antibodies on the cell surface of murine epidermal cell monolayers and their internalization. J Invest Dermatol 1984; 83: 409–415.

63.

Jones

Yokoo

Goldman

RD.

A cell surface desmosome-associated component: Identification of tissue-specific cell adhesion molecule.

Proc Natl Acad Sci U S A 1986; 83: 7282–7286.

64.

Burmester

IAK

Emtenani

Johns

J-G

, et al. Translational use of a standardized full human skin organ culture model in autoimmune blistering diseases. Curr Protoc Pharmacol 2019; 85: e56.

65.

Michel

CS.

An organ culture model for the study of pemphigus acantholysis. Br J Dermatol 1977; 96: 295–302.

66.

C-H

Michel

Schiltz

JR.

Epidermal acantholysis induced in vitro by pemphigus autoantibody. Am J Pathol 1978; 90: 18.

67.

Schiltz

Michel

Papay

Appearance of ‘pemphigus acantholysis factor’ in human skin cultured with pemphigus antibody. J Invest Dermatol 1979; 73: 575–581.

68.

Morioka

Naito

Ogawa

The pathogenic role of pemphigus antibodies and proteinase in epidermal acantholysis. J Invest Dermatol 1981; 76: 337–341.

69.

Swanson

Dahl

MV.

Methylprednisolone inhibits pemphigus acantholysis in skin cultures. J Invest Dermatol 1983; 81: 258–260.

70.

Jeffes

Kaplan

Ahmed

AR.

Acantholysis produced in vitro with pemphigus serum: hydrocortisone inhibits acantholysis, while dapsone and 6-mercaptopurine do not inhibit acantholysis. J Clin Immunol 1984; 4: 359–363.

71.

Schuh

Besch

Braungart

, et al. Protease inhibitors prevent plasminogen-mediated, but not pemphigus vulgaris-induced, acantholysis in human epidermis. Biol Chem 2003; 384: 311–315.

72.

Van der Wier

Pas

Jonkman

MF.

Experimental human cell and tissue models of pemphigus. Dermatol Res Pract 2010; 2010: 143871.

73.

Egu

Walter

Spindler

, et al. Inhibition of p38MAPK signalling prevents epidermal blistering and alterations of desmosome structure induced by pemphigus autoantibodies in human epidermis. Br J Dermatol 2017; 177: 1612–1618.

74.

Amagai

Klaus-Kovtun

Stanley

JR.

Autoantibodies against a novel epithelial cadherin in pemphigus vulgaris, a disease of cell adhesion. Cell 1991; 67: 869–877.

75.

Barrandon

Green

Three clonal types of keratinocyte with different capacities for multiplication.

Proc Natl Acad Sci U S A 1987; 84: 2302–2306.

76.

Watt

Green

Stratification and terminal differentiation of cultured epidermal cells. Nature 1982; 295: 434–436.

77.

Suter

Wilkinson

Dougherty

, et al. Ultrastructural localization of pemphigus vulgaris antigen on canine keratinocytes in vivo and in vitro. Am J Vet Res 1990; 51: 507–511.

78.

Calautti

Cabodi

Stein

, et al. Tyrosine phosphorylation and src family kinases control keratinocyte cell–cell adhesion. J Cell Biol 1998; 141: 1449–1465.

79.

Boukamp

Popp

Altmeyer

, et al. Sustained nontumorigenic phenotype correlates with a largely stable chromosome content during long-term culture of the human keratinocyte line HaCaT. Genes Chromosomes Cancer 1997; 19: 201–214.

80.

Caldelari

Müller

EJ.

Short- and long-term cultivation of embryonic and neonatal murine keratinocytes. In: Ward

Tosh

(eds) Mouse cell culture. Totowa, NJ: Humana Press, pp.125–138.

81.

Kolly

Suter

MM and

Müller

EJ.

Proliferation, cell cycle exit, and onset of terminal differentiation in cultured keratinocytes: Pre-programmed pathways in control of C-Myc and Notch1 prevail over extracellular calcium signals. J Invest Dermatol 2005; 124: 1014–1025.

82.

Osada

Seishima

Kitajima

Pemphigus IgG activates and translocates protein kinase C from the cytosol to the particulate/cytoskeleton fractions in human keratinocytes. J Invest Dermatol 1997; 108: 482–487.

83.

Caldelari

de Bruin

Baumann

, et al. A central role for the armadillo protein plakoglobin in the autoimmune disease pemphigus vulgaris. J Cell Biol 2001; 153: 823–834.

84.

De Bruin

Caldelari

Williamson

, et al. Plakoglobin-dependent disruption of the desmosomal plaque in pemphigus vulgaris. Exp Dermatol 2007; 16: 468–475.

85.

Schmidt

Feoktistova

Panayotova-Dimitrova

, et al. Pitfalls in the application of dispase-based keratinocyte dissociation assay for in vitro analysis of pemphigus vulgaris. Vaccines (Basel) 2022; 10: 208.

86.

Calkins

Setzer

Jennings

, et al. Desmoglein endocytosis and desmosome disassembly are coordinated responses to pemphigus autoantibodies. J Biol Chem 2006; 281: 7623–7634.

87.

Kitajima

Inoue

Yaoita

Effects of pemphigus antibody on the organization of microtubules and keratin-intermediate filaments in cultured human keratinocytes. Br J Dermatol 1986; 114: 171–179.

88.

Wilgram

Caulfield

Lever

WF.

An electron microscopic study of acantholysis in pemphigus vulgaris. J Invest Dermatol 1961; 36: 373–382.

89.

Hammers

Stanley

JR.

Antibody phage display: Technique and applications. J Invest Dermatol 2014; 134: 1–5.

90.

Hammers

Chen

Lin

, et al. Persistence of anti-desmoglein 3 IgG(+) B-cell clones in pemphigus patients over years. J Invest Dermatol 2015; 135: 742–749.

91.

Waschke J, Spindler V, Bruggeman P, Zillikens D, Schmidt G and Drenckhahn D. Inhibition of Rho A activity causes pemphigus skin blistering. The Journal of cell biology 2006; 175(5): 721–7. 10.1083/jcb.200605125.

92.

Amagai M, Matsuyoshi N, Wang ZH, et al. Toxin in bullous impetigo and staphylococcal scalded-skin syndrome targets desmoglein 1. Nat Med 2000; 6: 1275–1277.

93.

Saito

Stahley

Caughman

, et al. Signaling dependent and independent mechanisms in pemphigus vulgaris blister formation. PLoS One 2012; 7: e50696.

94.

Egu

Sigmund

Schmidt

, et al. A new ex vivo human oral mucosa model reveals that p38MAPK inhibition is not effective in preventing autoantibody-induced mucosal blistering in pemphigus. Br J Dermatol 2020; 182: 987–994.

95.

Emtenani

Ghorbanalipoor

Mayer-Hain

, et al. Pathogenic activation and therapeutic blockage of FcαR-expressing polymorphonuclear leukocytes in IgA pemphigus. J Invest Dermatol 2021; 141: 2820–2828.

96.

Neil

Brown

Williams

AC.

Human skin explant model for the investigation of topical therapeutics. Sci Rep 2020; 10: 21192.

97.

Lotti

Atene

Zanfi

, et al. In vitro, ex vivo, and in vivo models for the study of pemphigus. Int J Mol Sci 2022; 23: 7044.