Claudin-4 Upregulation in Acantholytic and Autoimmune-Mediated Bullous Disorders
and Bruce R. Smoller 1Round 1
Reviewer 1 Report
Comments and Suggestions for Authors
In this paper, the authors performed immunohistochemistry showing increased expression of claudin-4, a major component of tight junctions, in acantholytic and autoimmune bullous diseases. The expression of claudin-4 in Haily-Haily disease and Darier disease has already been reported by Raiko et al. Therefore, the novelty of this paper lies in the investigation of claudin-4 expression in autoimmune bullous diseases. However, the authors did not systematically examine the expression of tight junction component proteins as in the paper by Raiko et al. It seems to me that the paper only examined claudin-4 expression in other skin diseases that were not investigated by Raiko et al.
1) It is interesting that the expression of claudin-4 was negative in spongiotic dermatitis. It would be better to show the immunohistochemistry of spongiotic dermatitis as a negative control and emphasize that claudin-4 expression is helpful in differentiating acantholysis from spongiosis.
2) In the peri-lesional epidermis, claudin-4 was expressed in the granular layer, whereas it was expressed in the suprabasal layer in the blister of cutaneous pemphigus (Figure 1B) and in the basal layer in the blister of bullous pemphigoid (Figure 1D). Why did the expression area of cluadin4 change in these lesions?
3) Two investigators determined the intensity of claudin-4 with the naked eye. This method was not quantitative, and it is not reliable whether claudin-4 expression was actually elevated.
4) What do the authors think about claudin-4 positivity in warty dyskeratoma?
5) It is interesting that claudin-4 expression was not seen in the bullous areas of mucosal PV, but what about mucous membrane pemphigoid? Did the authors also examine claudin-4 expression in epidermolysis bullosa acquisita, dermatitis herpetiformis, etc.?
6) In Table 1, it would be helpful to show the number of cases examined and the number of positive cases for each disease.
Author Response
Dear reviewer,
Thank you so much for reviewing the manuscript and providing helpful comments!
1) It is interesting that the expression of claudin-4 was negative in spongiotic dermatitis. It would be better to show the immunohistochemistry of spongiotic dermatitis as a negative control and emphasize that claudin-4 expression is helpful in differentiating acantholysis from spongiosis.
I have added photos demonstrating a negative expression of claudin-4 in eczema in figure 1 and comments in the manuscript.
2) In the peri-lesional epidermis, claudin-4 was expressed in the granular layer, whereas it was expressed in the suprabasal layer in the blister of cutaneous pemphigus (Figure 1B) and in the basal layer in the blister of bullous pemphigoid (Figure 1D). Why did the expression area of cluadin4 change in these lesions?
The peri-lesional epidermis in all cases shows a normal staining pattern of claudin-4. Within the lesional areas of cutaneous PV (suprabasilar layer) and BP (basal layer) show overexpression of claudin-4. We suggested that significant upregulation of claudin-4 in the lesional areas might serve as a compensatory mechanism aimed at maintaining barrier function in affected skin regions. This upregulation in claudin-4 may serve to counteract the barrier disruption caused by the downregulation of claudin-1 in these disorders.
3) Two investigators determined the intensity of claudin-4 with the naked eye. This method was not quantitative, and it is not reliable whether claudin-4 expression was actually elevated.
I have included additional images depicting normal claudin-4 staining to facilitate a comprehensive comparison. Notably, the expression of claudin-4 in normal skin exhibits a comparatively weaker intensity when contrasted with its expression in disrupted areas in acantholytic entities and BP. In acantholytic lesions and Bullous Pemphigoid (BP), a striking observation emerges—claudin-4 expression in the disrupted areas appears significantly more robust than in non-lesional areas. Please see figures 1 and 2.
4) What do the authors think about claudin-4 positivity in warty dyskeratoma?
We suggested that significant upregulation of claudin-4 in the lesional areas might serve as a compensatory mechanism aimed at maintaining barrier function in affected skin regions. This upregulation in claudin-4 may serve to counteract the barrier disruption caused by the downregulation of claudin-1 in these disorders.
5) It is interesting that claudin-4 expression was not seen in the bullous areas of mucosal PV, but what about mucous membrane pemphigoid? Did the authors also examine claudin-4 expression in epidermolysis bullosa acquisita, dermatitis herpetiformis, etc.?
This is a great idea. We would consider performing claudin-4 in these entities for a different project.
6) In Table 1, it would be helpful to show the number of cases examined and the number of positive cases for each disease.
I added the number of cases in the table.
Reviewer 2 Report
Comments and Suggestions for Authors
This paper describes claudin-4 expression in acantholytic disorders via immunohistochemistry. Claudin-4 is important protein in tight junctions and its presence/absence plays a role in barrier function. It does seem like a bit of stretch to make some of the conclusions stated in this paper based on pattern of staining only. I would like more discussion of the limitations of this study, possible next steps or follow up studies should also be mentioned. The methods section needs to be more robust as do the conclusions.
page2 line 56 For the study cases, was diagnosis of PV, PF etc also confirmed with direct immunofluorescence or just formalin-fixed tissue?
page 2 line 74 is there an image of normal staining in normal skin to compare the study cases with?
page 2 line 84 - was claudin 4 downregulated in eczematous skin?
Page 3 - negative control picture would be helpful
page 4 line 112 - what is the reference for this?
page 4 line 122 - the paper cited seems to be looking at claudin 4 in skin treated with all-trans retinoic acid. How is this an acantholytic disorder?
page 5 line134 - epidermal should be epithelial
Author Response
Dear reviewer,
Thank you for reviewing our manuscript! Please see below for our point-by-point responses to your suggestions.
This paper describes claudin-4 expression in acantholytic disorders via immunohistochemistry. Claudin-4 is important protein in tight junctions and its presence/absence plays a role in barrier function. It does seem like a bit of stretch to make some of the conclusions stated in this paper based on pattern of staining only. I would like more discussion of the limitations of this study, possible next steps or follow up studies should also be mentioned. The methods section needs to be more robust as do the conclusions.
I have added the limitation of the study in the discussion part.
“It is crucial to acknowledge the limitations of our study, including the relatively small size of our current cohort and the utilization of a single antibody marker. To validate and generalize our findings, further extensive studies involving larger cohorts and multiple antibody markers are imperative.”
page2 line 56 For the study cases, was diagnosis of PV, PF etc also confirmed with direct immunofluorescence or just formalin-fixed tissue?
All PV, PF, and BP cases were diagnosed based on H&E and positive DIF study. I have added this information to the manuscript.
page 2 line 74 is there an image of normal staining in normal skin to compare the study cases with?
I added a photo demonstrating a normal staining pattern of claudin-4 in figure 1.
page 2 line 84 - was claudin 4 downregulated in eczematous skin?
Claudin-4 was not downregulated in eczematous skin. I have added this information to the manuscript.
“Significantly, claudin-4 staining in all eczema cases in our study revealed stronger expression in non-lesional skin (the stratum granulosum) than in lesional skin, aligning with the pattern identified in the 2015 study by Gruber et al.15 This observation implies an upregulation of Claudin-4 in non-lesional areas, highlighting a distinctive contrast in expression between non-lesional and lesional eczematous skin.”
Page 3 - negative control picture would be helpful
I have added a photo of negative control case in figure 1.
page 4 line 112 - what is the reference for this?
I have added the reference.
page 4 line 122 - the paper cited seems to be looking at claudin 4 in skin treated with all-trans retinoic acid. How is this an acantholytic disorder?
I removed the reference #12.
page 5 line134 - epidermal should be epithelial
It was corrected in the manuscript.
Reviewer 3 Report
Comments and Suggestions for Authors
Comment 1: In the Abstract you mention 22 cases (“we collected 22 cases…”) while under RESULTS you state, “We collected a total of 25 cases…..”. To summarize: n=22 (study set) PLUS n=13 (control/comparison set) sums up to 35 cases…… Please correct these numbers! [Looking at Figure 1 C-D and reading RESULTS, there is an obvious explanation: keratinocytes of the basal layer of the subepidermal blister in BP show positivity (n=3!). Fascinating finding, albeit BULLOUS PEMPHIGOID is a NON-ACANTHOLYTIC blistering disorder that does not really fit under the presented title of “acantholytic”. This finding nevertheless is most remarkable and deserves detailed discussion.]
Comment 2: Remarkably, only 1 antibody [anti-CLAUDIN-4] was used in this study, while the various CLAUDINS, OCCLUDIN and other associated proteins that play pivotal roles at the TJ were not applied. As these various proteins always play in concert, their roles may be more clearly understood in studies with multiple-antibody-panels. Therefore this “singular Claudin-4 approach” is one of the weak points of this study. Please comment briefly, why you opted for the one-shot-approach – and why this approach suffices.
Comment 3. Surprisingly, keratinocytes in eczema [spongiotic-vesicular dermatitis] were only weakly decorated by anti-claudin-3 AB. Similar and not-so-similar findings have been reported before (e.g. Gruber R et al (2015) 277-2789; Am J Pathol), “Claudin-4 immunointensity is significantly up-regulated in non-lesional atopic dermatitis (whereas OCCLUDIN immunointensity is unaltered).” Other authors (e.g. Li J (2029) 1789-1805 I J Molec Med) mention, “Claudin-4 upregulation may be a compensatory response for the disrupted barrier function caused by Claudin-1 downregulation…” Apparently Claudin-4 is modulated via ∆Np63, making p63 a negative regulator of Claudin-4 expression (Kojima T et al. 2017, 25-31). You already mentioned this is briefly or partially. To put your findings into clearer perspective, the manuscript might benefit from a more detailed discussion of the (postulated and speculated) etiopathogenetic underpinnings of Claudin-4 upregulation (vs. downregulation) in acantholytic vs. spongiotic-vesicular (eczematous) disorders. Also a short comment would help, regarding the upregulation of Claudin-4 in both inflammatory and non-inflammatory (and genetic) disorders.
Comment 4. Are there anatomical differences between keratinocytes of the mucosa and keratinocytes of the cornified skin? Are there specific ultrastructural differences at the TJ?
Comment 5. Explain Claudin-4 positivity of the non-acantholytic basal keratinocytes (at the roof-top) of the BP blisters.
Author Response
Dear reviewer,
Thank you for reviewing our manuscript! Please see below for our point-by-point responses to your suggestions.
Comment 1: In the Abstract you mention 22 cases (“we collected 22 cases…”) while under RESULTS you state, “We collected a total of 25 cases…..”. To summarize: n=22 (study set) PLUS n=13 (control/comparison set) sums up to 35 cases…… Please correct these numbers! [Looking at Figure 1 C-D and reading RESULTS, there is an obvious explanation: keratinocytes of the basal layer of the subepidermal blister in BP show positivity (n=3!). Fascinating finding, albeit BULLOUS PEMPHIGOID is a NON-ACANTHOLYTIC blistering disorder that does not really fit under the presented title of “acantholytic”. This finding nevertheless is most remarkable and deserves detailed discussion.]
I corrected the numbers in the revised manuscript.
“We collected a total of 35 cases from patients aged between 30 and 87 years, comprising 18 males and 17 females. The cases encompassed a variety of conditions: two cases of Hailey-Hailey disease, one case of Darier disease, three cases of Grover disease, one case of acantholytic acanthoma, two cases of warty dyskeratoma, 11 cases of pemphigus vulgaris (including five cutaneous PV and six mucosal PV), two cases of pemphigus foliaceus, three cases of bullous pemphigoid, five cases of eczematous conditions, and five unremarkable skin cases (take from margins of surgical excisions)”
I also added comments regarding BP in the manuscript.
“A noteworthy contribution of our study is the first-time report of claudin-4 overexpression in the keratinocytes surrounding blisters in an autoimmune-mediated subepidermal bullous disorder, bullous pemphigoid. The underlying pathogenesis of this observation remains unclear and warrants further exploration.
Comment 2: Remarkably, only 1 antibody [anti-CLAUDIN-4] was used in this study, while the various CLAUDINS, OCCLUDIN and other associated proteins that play pivotal roles at the TJ were not applied. As these various proteins always play in concert, their roles may be more clearly understood in studies with multiple-antibody-panels. Therefore this “singular Claudin-4 approach” is one of the weak points of this study. Please comment briefly, why you opted for the one-shot-approach – and why this approach suffices.
I have added the limitation of the study in the discussion part.
“It is crucial to acknowledge the limitations of our study, including the relatively small size of our current cohort and the utilization of a single antibody marker. To validate and generalize our findings, further extensive studies involving larger cohorts and multiple antibody markers are imperative.”
Comment 3. Surprisingly, keratinocytes in eczema [spongiotic-vesicular dermatitis] were only weakly decorated by anti-claudin-3 AB. Similar and not-so-similar findings have been reported before (e.g. Gruber R et al (2015) 277-2789; Am J Pathol), “Claudin-4 immunointensity is significantly up-regulated in non-lesional atopic dermatitis (whereas OCCLUDIN immunointensity is unaltered).” Other authors (e.g. Li J (2029) 1789-1805 I J Molec Med) mention, “Claudin-4 upregulation may be a compensatory response for the disrupted barrier function caused by Claudin-1 downregulation…” Apparently Claudin-4 is modulated via ∆Np63, making p63 a negative regulator of Claudin-4 expression (Kojima T et al. 2017, 25-31). You already mentioned this is briefly or partially. To put your findings into clearer perspective, the manuscript might benefit from a more detailed discussion of the (postulated and speculated) etiopathogenetic underpinnings of Claudin-4 upregulation (vs. downregulation) in acantholytic vs. spongiotic-vesicular (eczematous) disorders. Also a short comment would help, regarding the upregulation of Claudin-4 in both inflammatory and non-inflammatory (and genetic) disorders.
I expanded these in the manuscript as you suggested.
For spongiosis – I added “Significantly, claudin-4 staining in all eczema cases in our study revealed stronger expression in non-lesional skin (the stratum granulosum) than in lesional skin, aligning with the pattern identified in the 2015 study by Gruber et al.15 This observation implies an upregulation of Claudin-4 in non-lesional areas, highlighting a distinctive contrast in expression between non-lesional and lesional eczematous skin.”
For acantholysis – I repeated a literature search to learn more about etiopathogenesis and here is the best explanation we could come up with. “Upregulation of claudin-4 might serve as a compensatory mechanism aimed at maintaining barrier function in affected skin regions. This upregulation in claudin-4 may serve to counteract the barrier disruption caused by the downregulation of claudin-1 in these disorders.”
Comment 4. Are there anatomical differences between keratinocytes of the mucosa and keratinocytes of the cornified skin? Are there specific ultrastructural differences at the TJ?
I commented on this in the manuscript.
“Another study by Hatakeyama et al. investigating the expression patterns of adhesion molecules in the gingival mucosa reported the presence of claudin-1 in the intermediate portion of the uppermost epidermal layers, but only in a few cases. The authors proposed that TJs in the oral mucosa could resemble structures present in the uppermost epidermal layer, akin to the stratum granulosum in the skin epidermis.17 Contrasting this, our own study revealed a distinctive difference in claudin-4 expression. We noted the presence of claudin-4 exclusively within disrupted keratinocytes in acantholytic and bullous areas of cutaneous PV, whereas mucosal PV exhibited no such claudin-4 presence. These observations suggest the potential absence (or markedly diminished expression) of claudin-4 in the mucosa and the possibility that the EP417 antibody used for claudin-4 immunohistochemistry may not cross-reactively stain claudin-1. In a more recent study, it was revealed that the baseline expression of TJ genes differs between oral and skin epithelium, and these expression patterns change during wound healing. Notably, the knockdown of claudin-1 and occludin resulted in alterations in proliferation and migration in skin epithelium, while no such effects were observed in oral keratinocytes.18 In the skin epithelium, claudin-1 knockdown triggers claudin-4 overexpression, possibly compensating for the loss. However, in the oral mucosal epithelium, claudin-1 knockdown leads to reduced keratinocyte proliferation and migration, preventing claudin-4 overexpression.18”
Comment 5. Explain Claudin-4 positivity of the non-acantholytic basal keratinocytes (at the roof-top) of the BP blisters.
This is a normal expression of claudin-4 in normal skin as well as peri-lesional area of all entities in this study.
“Claudin-4 immunohistochemistry demonstrated weak-to-moderate expression in the keratinocytes located in the stratum granulosum, keratinocytes surrounding hair follicles, and adnexal glands (normal staining pattern) in all 35 cases.”
Round 2
Reviewer 1 Report
Comments and Suggestions for Authors
I am still not convinced by the quantitative method of immunostaining for claudin 4 in this paper, and this paper has the limitation of a small number of cases. However, the fact that this paper found increased membranous expression of claudin-4 in disrupted keratinocytes within acantholytic and/or bullous areas is interesting, even if the mechanism is unclear.
