Molecular Changes Induced by Carbon Dioxide Laser in Hailey‐Hailey Disease: A Potential Mechanism Underlying Treatment Efficacy

ABSTRACT

Introduction

Hailey‐Hailey disease (HHD) is a rare genodermatosis caused by mutations in the ATP2C1 gene that codes for SPCA1, a calcium transporter in the epidermis. HHD impairs quality of life, and no curative treatment exists.

Methods

To confirm the efficacy and safety of CO₂ laser in HHD, we conducted a randomized, prospective, controlled study that included 10 patients with histologically confirmed HHD. A 4‐mm punch biopsy was taken from unaffected skin and the laser‐treated area before treatment. CO₂ laser treatment was performed under sedation, and after 6 months, new biopsies were taken at the same sites. Samples were used to extract RNA and analyze potential changes in gene expression.

Results

Laser treatment provided a significant reduction in the lesioned area (p < 0.002). Gene expression analysis revealed reduced ATP2C1 expression in lesional skin and high expression of genes related to keratinocyte proliferation and IL‐17‐dependent inflammation. Interestingly, in affected skin after laser treatment, there was an enhancement of ATP2C1 with a reduced expression of genes related to keratinocyte proliferation and IL‐17‐dependent inflammation.

Conclusions

CO₂ laser is safe and effective in HHD. A plausible explanation could be molecular changes after treatment, especially ATP2C1 enhancement and IL‐17‐dependent inflammation decreased.

1. Introduction

Hailey‐Hailey disease (HHD), also known as familial benign chronic pemphigus, is a rare autosomal dominant genodermatosis characterized by the presence of flaccid blisters primarily affecting skin folds. This chronic, relapsing condition significantly impacts the quality of life of patients [1, 2].

HHD is caused by heterozygous mutations in the ATP2C1 gene, which encodes the secretory pathway Ca²⁺/Mn²⁺‐ATPase 1 (SPCA1) protein, responsible for the transport of calcium (Ca²⁺) and manganese (Mn²⁺) ions. To date, the most widely accepted molecular theory regarding the cause of the disease is haploinsufficiency. In this model, the patient carries a germline mutation in one allele of the gene, and the functioning copy is ultimately insufficient to compensate and prevent the phenotype from manifesting [3, 4] Triggers such as heat, sweating, friction, or trauma could lead to a second molecular hit, explaining why lesions typically appear in skin folds and less frequently in other body areas [5].

Under normal conditions, a Ca²⁺ gradient exists in the epidermis, with concentrations up to four times higher in the outer layers compared to the basal layer. Elevated extracellular Ca²⁺ levels regulate key processes in keratinocytes, including differentiation, adhesion, motility, and lipid secretion. However, this gradient is disrupted during physiological skin aging and in certain acantholytic genodermatoses, including HHD [6, 7, 8, 9].

SPCA1, a member of the type II Ca²⁺ ATPase family, is a transmembrane protein localized in the Golgi apparatus. Its proper function is crucial for sequestering Ca²⁺ and Mn²⁺ within the Golgi. SPCA1 plays a vital role in synthesizing desmosomal proteins, which are essential for keratinocyte adhesion in the epidermis [10, 11, 12]. Dysfunction of SPCA1 is implicated in suprabasal acantholysis, a hallmark feature of HHD. In vitro studies using SPCA1‐deficient keratinocyte cultures have shown delayed translocation of desmosomal proteins such as desmoplakin and desmoglein‐3, which may explain the acantholytic phenotype observed in HHD [13, 14, 15]. Notably, SPCA1 accounts for approximately 70% of Ca²⁺ transport in the epidermis, primarily in the suprabasal layers. The remaining 30% is mediated by sarco/endoplasmic reticulum Ca²⁺ ATPase 2 (SERCA2), encoded by the ATP2A2 gene and localized in the sarcoplasmic reticulum of the basal epidermal layer [5, 16] Mutations in ATP2A2 lead to Darier disease (DD), another genodermatosis with clinical and histological features similar to HHD [17, 18].

Diagnosing HHD is complex, as histology is not specific, direct immunofluorescence is negative, and autoantibody detection is not useful for diagnosis or follow‐up [19, 20, 21, 22]. Treatment remains challenging, and multiple topical and systemic therapies have been reported with limited success, including antibiotics for secondary infections, botulinum toxin and anticholinergics to minimize sweating, and immunomodulators to control the inflammatory response [23, 24, 25, 26, 27, 28, 29, 30]. These therapies generally provide temporary symptomatic relief but do not alter the natural course of the disease. Surgical intervention is reserved for refractory, localized disease and involves excision of the affected area. Although some successes have been reported, aesthetic and functional sequelae limit its use. On the contrary, the carbon dioxide (CO₂) laser offers controlled depth impact reducing the risk of pathological scarring. This technique is reproducible, less operator‐dependent, and associated with shorter recovery times [30, 31, 32]. Although CO₂ laser treatment does not alter the natural course of HHD due to its systemic and germline nature, in recent years, several studies—including our own—have reported promising outcomes in patients with HHD refractory to conventional therapies, showing medium‐term improvement in CO₂ laser–treated areas [32, 33]. Nevertheless, the lack of large‐scale, randomized controlled trials evaluating long‐term efficacy and safety limits its availability. To address this knowledge gap, we designed a study to evaluate laser efficacy and explore the underlying pathophysiological mechanisms driving its therapeutic effects.

2. Methods

A prospective cohort study was conducted at Clínica Universidad de Navarra, Pamplona (Spain), between September 2021 and May 2024. The project was approved by the hospital's ethics board committee, ID 2020.240. The study included 10 patients with HHD, with molecularly and histologically confirmed bilateral involvement of skin folds and refractory disease.

A total of 12 patients were initially enrolled in the study, comprising six women and six men, aged between 37 and 81 years (mean: 60; standard deviation: 15). Two patients were excluded due to loss to follow‐up, resulting in a final cohort of 10 participants. All patients had a family history of HHD and had presented with cutaneous lesions for at least 5 years. Among them, two were siblings, and two were a father–daughter pair; the remaining six subjects were unrelated. ATP2C1 variants were detected in all 10 probands.

Prior to enrollment, all patients had received topical corticosteroids and topical antibiotics (100%), and nearly all had been treated with systemic antibiotics (90%). Six patients (60%) had previously used topical calcineurin inhibitors, three (30%) had undergone botulinum toxin infiltrations, and two (20%) had received methotrexate. There was a single case of previous use of azathioprine, doxepin, cyclosporine, apremilast, and adalimumab (10% each)

The groin was the most frequently affected area (90%), followed by the perineum (70%), genitals (50%), axillae, antecubital fossae, and buttocks (30%). A washout period of previous treatment was established: 4 months for patients treated with photodynamic therapy or botulinum toxin, 1 month for topical or oral immunosuppressants, and 2 weeks for antibiotics and/or antifungals.

2.1. Statistical Analysis

The validated application ImitioMeasure was used for measuring the affected area and calculating its reduction. We performed the statistical analyses using Stata version 14 (StataCorp LP, 2015, Stata Statistical Software: Release 14, College Station, Texas). Quantitative variables were summarized using the mean with standard deviation (SD). The Wilcoxon matched‐pairs test was used to evaluate the intervention efficacy compared to control. All statistical tests were 2‐sided. p < 0.05 were considered statistically significant.

2.2. Methodology

Prior to laser treatment, a 4‐mm diameter punch biopsy was taken from healthy skin unaffected by the disease (Biopsy 0), and another 4‐mm punch biopsy from the affected area where the laser treatment would be applied (Biopsy 1). Both lesional and non‐lesional biopsies were taken from comparable anatomical regions to reduce transcriptional background noise due to skin‐site specificity.

The biopsies were collected under sterile conditions and preserved in liquid nitrogen, stored at −80°C. The treatment was performed under anesthetic sedation. We used the Lumenis AcuPulse CO₂ ablative laser with a wavelength of 10,600 nm, operating with the following parameters: F200 scanner, 8‐mm spot size, 4‐mm depth, 0.50‐s duration, and 24 watts of power. Three passes were applied to each area until a yellowish color and slight bleeding were achieved, indicating that the papillary dermis had been reached [32]. Immediately afterward, medium‐potency topical corticosteroids (betamethasone) and a petroleum jelly gauze were applied to the treated area. After 48 h, the dressing was removed, and patients were instructed to perform daily cleansing with water and antiseptic soap until complete re‐epithelialization, which occurred within 4 to 6 weeks.

2.3. RNA Extraction and Gene Analysis

Six months after the baseline visit, two additional 4‐mm punch biopsies were taken: one from the previously treated area (Biopsy 2) and another one from healthy and untreated skin located close to the treated area (Biopsy 3) (Figure 1).

FIGURE 1.

Biopsy protocol and CO₂ laser treatment. (a) Patient with bilateral inguinal involvement. Two biopsies are obtained on one side, labeled as 0 (clinically unaffected skin adjacent to the lesions) and 1 (lesional skin scheduled for CO₂ laser treatment). (b–d) CO₂ laser treatment of one affected side; the contralateral side is treated during a subsequent visit. (e) Clinical appearance at 6‐month follow‐up, showing complete resolution of lesions. At this time, two additional biopsies are obtained, labeled as 2 (previously affected, laser‐treated skin) and 3 (adjacent untreated healthy skin) for subsequent gene‐expression analysis.

After obtaining all the specimens, RNA was extracted from 4 mm punch skin biopsies. Briefly, tissue samples were homogenized using the Bead Ruptor 12 (Omni), and RNA was extracted using automated platforms, either Maxwell system (Promega) employing magnetic beads and/or with TRIzol reagent (Invitrogen). Following purification, RNA concentration and purity were assessed using a Nanodrop spectrophotometer (Z‐5600), and RNA integrity was evaluated with the Bioanalyzer 2100 system (Agilent) to determine RIN and DV200 values. All RNA samples were stored at −80°C until further use.

Transcriptomic profiling was performed using Clariom S human microarrays (Affymetrix, Santa Clara, CA) following the GeneChip WT Pico protocol (ThermoFisher Scientific). A total of 10–40 ng of input RNA per sample was used, and procedures were carried out in accordance with the manufacturer's instructions (GeneChip 3′ IVT Pico Kit Manual Workflow User Guide, P/N 703308). Hybridization and washing steps were performed using the GeneChip Hybridization Oven 640 and GeneChip Fluidics Station 450 (protocol FS450‐0007). Arrays were scanned using the GeneChip Scanner 3000 system following the manufacturer's guidelines (“GeneChip Expression Wash, Stain and Scan” protocol, P/N 702731).

Changes in gene expression were analyzed using Affymetrix arrays and computational biology analysis. We postprocess the transcriptomics data performing a downstream analysis using in‐house functions developed in MATLAB (MathWorks) [34] We equalized the data and stabilized them through the log₂ transform of the data plus one—to avoid the undefined values of log₂; of zero; calculated the average values for each of the two groups—Biopsy 2 and 3—of three biological replicates each; selected the Differentially Expressed Genes (DEGs) whose absolute difference in mean values between the two groups was less than the selection threshold θ_DEG = 1 of fold change (FC) in the log₂ scale; and selected the statistically significant DEGs using unpaired Student's t‐test with equal variances and with a significance threshold of α _DEG = 0.05. Genes whose expression varied when comparing Biopsy 0 (unaffected skin before treatment) and Biopsy 3 (unaffected skin 6 months post‐treatment) were excluded from the final analysis, as no laser was applied to the unaffected skin. Gene expression changes between Biopsies 1 and 2 (affected area before and after treatment) were interpreted as modifications resulting from the treatment. These changes were subsequently validated using quantitative polymerase chain reaction (qPCR). B2M, PGK1, and GADPH were used as housekeeping genes. Pre‐dispensed plates with custom‐designed target probes (TaqMan MGB) from Thermo Fisher Scientific (Applied Biosystems TaqMan Low Density Array, TLDA) were used.

Custom‐designed, preloaded TaqMan MGB probe plates (ThermoFisher Scientific; Applied Biosystems TaqMan Low Density Array, TLDA) were used for quantitative PCR analysis. The format consisted of 24 assays per 384‐well plate, enabling technical duplicates for each sample and analysis of 8 samples per plate.

Plates were run on a QuantStudio 12 K Flex Real‐Time PCR System (ThermoFisher Scientific), and data were analyzed using the online Relative Quantification Software v4.3 (Applied Biosystems). RNA reverse transcription was performed using the SuperScript IV VILO Master Mix with ezDNase enzyme (Ref. 11,766,050, Invitrogen; ThermoFisher Scientific) with input amounts of 100, 500, or 1000 ng of RNA, depending on availability. PCR reactions were conducted using the TaqMan Fast Advanced Master Mix (ThermoFisher Scientific) according to the manufacturer's protocol.

Out of the 24 TaqMan assays, three targeted endogenous control genes (BM2, GAPDH, PGK1) for gene expression normalization. GAPDH was excluded due to high variability. Relative gene expression was calculated using the 2^–ΔΔCt method.

3. Results

The mean affected area was 65 cm² (SD 43) at baseline, decreasing to 6 cm² (SD 5) after laser treatment and 6 months of follow‐up, and further reducing to 4 cm² (SD 4) and 3 cm² (SD 5) after one and 2 years, respectively. The extent of the lesions showed a statistically significant reduction at 6, 12, and 24 months compared with baseline (p < 0.002) (Figure 2).

FIGURE 2.

Efficacy of laser therapy in a patient with cervical involvement. (a, b) Baseline presentation showing erythematous‐scaly patches on both sides of the neck. (c, d) Resolution of lesions observed on both sides following laser therapy, with sustained improvement over a two‐year follow‐up period. (e) Basal active lesions at the chest and near the neck. (f) Resolution of treated lesions near the neck but persistence of untreated lesions at the chest and at other locations as the back (g).

Following treatment, all patients exhibited pigmentary changes in the treated areas—seven with hypopigmentation and three with hyperpigmentation. No cases of pathological scarring were observed. One patient developed a localized infection at the treatment site despite appropriate wound care; this resolved completely after 1 week of oral ampicillin without sequelae.

No patient experienced recurrence of the disease within the treated areas, although three developed new lesions at the periphery. These cases were managed with topical corticosteroids, and in one instance, due to persistent activity and at the request of the patient, an additional laser session was performed using the same parameters, with no recurrence observed to date.

Global changes in gene expression among healthy skin, lesional skin before treatment, and previously affected skin after treatment—together with their confirmation by quantitative polymerase chain reaction (PCR)—are shown in the heat map (Figure 3). Higher gene‐expression levels are depicted in red, whereas lower levels appear in blue. A detailed analysis of differential expression across the subgroups is presented below.

FIGURE 3.

Heatmap showing changes in gene expression when comparing healthy skin with skin affected by HHD before and after CO₂ laser treatment from transcriptomics microarrays. The color bar codifies the gene expression in a log₂ scale. Higher gene expression corresponds to redder color, while bluer color represents a lower gene expression intensity.

A decreased expression of the ATP2C1 gene was observed in affected compared with healthy skin. This reduction was accompanied by the upregulation of several genes associated with keratinocyte proliferation (DEFB103A, EPGN, GJB6, KLK6, KRT6A, LCE3D, S100A7, S100A9, S100A12, SERPINB4, TCN1, and SPRR2A) and inflammation (DEFB103A, KRT6A, PI3, LCN2, TCN1, S100A7, S100A9, and S100A12) (Figure 4).

FIGURE 4.

Gene expression analysis in healthy versus affected skin. Bar graph illustrating relative quantification (RQ) changes in gene expression using qPCR. Affected skin shows reduced ATP2C1 expression and increased expression of genes involved in keratinization and inflammatory processes compared to healthy skin.

Subsequent analysis of genes showing statistically significant expression changes between affected skin biopsies collected before and after treatment revealed increased expression of ATP2C1 and DEFB103A following laser application. In contrast, post‐treatment decreases were observed for KLK6, KRT6A, LCE3D, LCN2, PI3, S100A7, S100A9, S100A12, GJB6, EPGN, SERPINB4, SPRR2A, and TCN1 (Figure 5).

FIGURE 5.

Gene expression changes in affected skin before and after CO₂ laser treatment. Bar graph showing relative quantification (RQ) of gene expression analyzed by qPCR. Post‐treatment, affected skin demonstrates increased expression of ATP2C1 and DEFB103A genes, alongside decreased expression of genes associated with keratinization and inflammatory pathways.

These findings suggest that the gene expression profile of treated skin more closely resembles that of healthy skin. Notably, no changes in ATP2A2 expression were detected when comparing healthy and affected skin either before or after treatment.

4. Discussion

Most treatments for HHD are based on isolated case reports, often lacking long‐term follow‐up data, with recurrence being the norm. In this study, treatment with CO₂ laser demonstrated long‐term efficacy and a favorable safety profile.

Several theories have been proposed to explain the mechanism of laser action. Some authors suggest that fibrosis of the treated skin strengthens the epidermal surface, reducing the likelihood of developing lesions despite exposure to triggers. However, this explanation is not entirely plausible since the laser acts on the papillary dermis but not at the reticular dermis, where fibrosis occurs [32, 35]. Additionally, no cases of pathological scarring were observed in our series, either clinically or histologically. Moreover, confocal microscopy studies performed after CO₂ laser treatment in HHD have evidenced neoangiogenesis but not fibrosis [35, 36].

On the other hand, some authors argue that epidermal normalization post‐treatment likely results from epidermal regeneration from unaffected keratinocytes within adnexal structures, which might not carry the mutation [33, 37]. However, this hypothesis is debatable, given that the germline mutation causing the disease affects all body cells. A slight variation of this hypothesis that we find more plausible, consists of re‐epithelialization occurring from stem cells of the follicular epithelium that are not reached by the laser. These cells carry the mutation in the ATP2C1 gene but may express ATP2A2 normally. Under normal conditions, ATP2A2 is localized in the basal layer of the epidermis and at the follicular infundibulum. Although it is responsible for only a portion of the Ca²⁺ gradient, it may be able to compensate and restore the epidermal architecture and Ca2+ suprabasal gradients. In fact, in this study, no changes in ATP2A2 expression were found in any of the biopsies, suggesting that its expression is consistent and does not interfere with the pathogenesis of HHD.

Finally, based on our experience, laser‐induced long‐term improvements in HHD patients may be associated with molecular changes affecting ATP2C1 gene expression and other genes involved in Ca²⁺ metabolism, keratinocyte differentiation, and inflammation. In this study, a statistically significant decrease in ATP2C1 expression was found in lesional skin when compared to healthy biopsies, which indicates that lower rates of expression of ATP2C1 could be related to lesion appearance. Interestingly, an increment in ATP2C1 expression was observed in affected skin following laser application, accompanied by the disappearance of the lesions. Thus, this higher expression of ATP2C1 could potentially lead to the restoration of ionic gradients, proper formation and conformation of desmosome components and finally, to epidermal homeostasis.

Additionally, an increase in the expression of genes related to keratinocyte proliferation and skin inflammation was detected in affected skin compared to healthy skin, suggesting a probable causal role. It is well established that low expression of ATP2C1 induces acantholysis. Consequently, the skin initiates compensatory mechanisms to restore the epidermal barrier, including keratinocyte hyperproliferation, which leads to the premature ascent of undifferentiated cells to the suprabasal layers and contributes to the HHD phenotype. Interestingly, laser treatment reduced the expression of these genes to levels comparable to those found in healthy skin.

Finally, an overexpression of genes associated with skin inflammation was observed in biopsies taken from lesional epidermis. The acantholysis characteristic of HHD could facilitate the penetration of pathogens, triggering an inflammatory response and modifying the microbiome. Once again, laser treatment resulted in a lower expression of these inflammation‐related genes. This effect could be secondary to the increased expression of ATP2C1, as previously mentioned, which facilitates the proper assembly of desmosomal components, avoids acantholysis, and prevents pathogen entry. Furthermore, in the affected skin after treatment, an increase in the expression of the DEFB103A gene was observed, a gene primarily involved in restoring and maintaining the integrity of the skin physical barrier.

Thus, unlike other treatments that only provide temporary improvement in HHD by targeting cofactors such as sweating, infections, or inflammation, CO₂ laser therapy offers long‐lasting effects on treated skin by eliminating mutant keratinocytes that disrupt epidermal structure and trigger inflammatory responses. Importantly, CO₂ laser therapy does not restore the function of the mutated HHD keratinocytes themselves; rather, the more plausible explanation is that these aberrant cells are ablated and subsequently replaced by keratinocytes that do not harbor such mutations, thereby re‐establishing normal epidermal architecture.

Nonetheless, further experiments in healthy keratinocytes would be necessary to determine whether CO₂ laser exposure can induce the expression of ATP2C1/2 in non‐diseased cells.

Finally, among the genes related to the inflammation process in the skin, we found high expression of Th17‐related genes (S100A7, S100A9 and S100A12) in lesional biopsies. Interestingly, their expression decreased after laser intervention. On one hand, the dominance of Th17 could be a direct effect of mutations in ATP2C1 in immune cells, driving a Th17 bias in T cells, as has been described for psoriasis or DD [38, 39] On the other hand, the altered epidermal homeostasis and a disrupted microbiome may be indirectly responsible for the increased expression of Th17‐related genes in the skin. In fact, some authors argue that it could be a transversal mechanism of disease that connects different entities with skin barrier dysfunction such as atopic dermatitis, psoriasis, DD and HHD. Moreover, recent evidence has shown that selective blockade of IL‐17 in DD with monoclonal antibodies controls flares and induces disease remission, with sustained results after 1 year of follow‐up [39, 40]. Based on the findings described in this study, we consider there is sufficient evidence to explore the use of IL‐17 inhibitors in HHD. Furthermore, high expression of this cytokine has been rarely reported in the pemphigus group [41, 42, 43]. Therefore, it would be of interest to investigate its role in the acantholysis observed in other diseases with similar pathophysiological and phenotypical features, potentially representing a common link among acantholytic disorders—whether autoimmune or genodermatoses. However, the involvement of Th17 cells remains hypothetical, and further studies involving detailed immunophenotyping and cytokine profiling are warranted to confirm these preliminary findings.

To the best of our knowledge, this is the first project to evaluate gene expression modification in HHD after treatment through RNA extraction using expression microarrays (Affymetrix) and subsequent qPCR validation. We consider this a robust method to understand the pathophysiology of this type of genodermatosis and to advance the treatment of these complex diseases.

5. Limitations

The small sample size of the study, influenced by recruitment challenges associated with the rarity of HHD, represents a significant limitation. However, similar studies analyzing gene expression have been conducted with comparable or even smaller sample sizes. Furthermore, technical challenges during RNA extraction may have affected the quality of the data, potentially limiting the ability to detect all gene expression changes or accurately quantify their levels.

6. Conclusions

In conclusion, CO₂ laser therapy has proven to be a safe and effective treatment, achieving long‐term remission in patients with HHD. While the exact mechanism by which the laser induces remission remains unclear, the normalization of the treated epidermis based on ATP2A2 expression, along with molecular changes in genes related to skin conformation and inflammation, represent plausible explanatory hypotheses. Finally, the use of IL‐17 inhibitors could be explored as a potential targeted therapy.

Funding

The project was funded by the State Research Agency (Spain), reference (PID2020‐114340RA‐I00).

Ethics Statement

The Institutional Review Board of the hospital approved this study and the rules of the Publication Ethics Committee (COPE) of the journal have been followed.

Consent

Consent for the publication of recognizable patient photographs or other identifiable material was obtained by the authors and this information may be publicly available.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus Edgar et al. 2002 and are accessible through GEO Series accession number (GSE272359) (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE272359).