Patients and their families were recruited at the Genodermatoses unit, department of dermatology at the American University of Beirut Medical Center (AUBMC, Beirut, Lebanon) under the ethical approval Institutional Review Board (IRB) from AUBMC (Protocol Number: IRB: DER.MK.01). Approval and written informed consents were obtained from patients or their parents (if minors) for obtaining blood sample and pictures. The sibling did not participate, and his genetic data was not available. Clinical phenotypes were provided by the referring physician. Peripheral blood samples were collected from the participating individuals and kept at 4 degrees Celsius. DNA was extracted within 1 hour after blood collection from specimens and stored at 4°C.

Whole Exome Sequencing was performed for the patient and her parents at Macrogen Laboratory (https://dna.macrogen.com/)(Seoul, South Korea). 4 μg DNA of each sample was processed. One-hundred fifty-one base pair(bp-paired-end) reads were sequenced using the Illumina NovaSeq6000 platform. The library preparation was performed according to the manufacturer’s protocol (Twist Human core exome). The Phred quality value was assigned to Q30, which means that the base call accuracy was 99.9%. Supplementary Table S1 shows the raw data statistics and quality scores. The generated FASTQ files were mapped to the Human GRCh37/hg19 reference assembly. Annotation and filtering processes for curated potential variants were performed as previously described (15).

Punch biopsies were obtained from the lesion of the affected patient (MK384) and controls (MK391 and MK392), individuals without skin autoimmune or chronic inflammatory diseases who are around the same age as the patient. Samples were processed directly for total RNA extraction using the Invitrogen Life Technologies TRIzol reagent following the manufacturer’s protocol. RNA quantification was performed using the Nanodrop 1000 spectrophotometer (Thermo Scientific) at AUB. All samples, from both the patient and controls, were processed in technical duplicates, and the fragments per kilobase of exon per million mapped fragments (FKPM) values represent the mean of these duplicates. Further experimental details were performed as previously described (16).

Htseq-count was used to summarize the read counts at the gene level. The differentially expressed genes (DEGs) were identified using DESeq2 after filtering out genes with low counts in all samples from the expression matrix. A gene was considered differentially expressed if its adjusted P-value (false discovery rate or FDR) was less than 0.01, and the absolute Log2 fold change was greater than or equal to 1, which corresponds to a 2-fold change threshold. Differential expression analyses compared the gene-expression profiles between the patient and the controls. Data analysis was conducted using iDEP.91 for generating heatmaps, scatterplots, and boxplots. Hierarchical clustering with the top 1000 most variable genes used a correlationbased distance metric, with linkage determined by an average cut-off score of 4. Gene expression values were centered by subtracting their mean value. Additionally, k-mean clustering with four clusters was performed using the top 2000 variable genes to identify biological processes. Gene normalization was achieved using the mean-center pathway database for GO biological processes within iDEP. 91.

The lists of DEGs were imported to the Ingenuity Pathway Analysis (IPA) software package (Ingenuity Systems, Redwood City, CA) to identify top enriched canonical pathways, upstream regulators, mechanistic networks, disease and functions’ annotations. The analysis followed a methodology similar to the description reported by (17). In brief, the probe sets were mapped to the Ensemble and HUGO gene symbols within IPA software. Probe sets that did not map to any HUGO/Ensemble genes were discarded. Pathways were ranked according to p-values, with the most significantly impacted pathways displayed at the top. The analysis included genes with log2 fold changes of < -4.4 and >+ 4.7 and FDR less than or equal to 0.05. These criteria were used to identify genes with the most pronounced differential expression, which were subsequently selected for functional annotation.

Peripheral blood mononuclear cells (PBMCs) were isolated using a Ficoll-Paque PLUS gradient (GE Healthcare; IL, USA). Anti-human antibodies for surface makers, with isotype-matched controls, were used to stain CD3 (UCHT1) and CD14 (M5E2) from Biolegend (CA, USA), and IL-17RA (W23-251) from BD Biosciences (NJ, USA). In brief, T cells were identified by gating on the lymphocytes in the forward vs. side scatter plots using the CD3 marker, a pan marker for T lymphocytes. Monocytes were identified by gating on the monocytes in the forward vs. side scatter and using the CD14 marker, which is expressed on all monocytes and is part of the TLR4 signaling complex. Neutrophils were not specifically gated using a marker; instead, granulocytes, primarily composed of neutrophils, were gated in the forward vs. side scatter plots. Cells were sorted using a BD FACS Aria cell sorter (BD Biosciences, NJ, USA), and analyzed with FlowJo software (Tree Star Inc.; OR, USA).

The vectors were generated at Yamaguchi University Graduate School of Medicine, Japan by Dr. Yutaka Shimomura. Tags were introduced into the C-terminus. Using skin cDNA from a healthy control individual, expression vectors were constructed for an N-terminal HA-tagged IL17RA (IL17RA-HA), IL17RA-Y391*-HA, and Il17RC-FLag. For wild-type (Wt) IL17RA, the coding sequences were amplified by polymerase chain reaction (PCR) using a forward primer (IL17RA-F-EcoRI: 5’-CTCTGAATTCTCAGAACGTTCGTTCGCTGC-3’) and a reverse primer (IL17RA-Wt-R-HA-SalI: 5’AAAAGTCGACTAGGCGTAGTCGGGCACGTCGTAGGGGTATGCACTGGGCCCCTCTGACT-3’) which introduced the HA-tag. The amplified PCR product was digested with EcoRI & SalI enzymes and subcloned into the pCXN2.1 vector at EcoRI and XhoI sites.

For the IL17RA-Y391*-HA vector, PCR was performed using the forward primer to amplify its sequence (IL17RA-F-EcoRI: 5’-CTCTGAATTCTCAGAACGTTCGTTCGCTGC-3’) and reverse primer (IL17RA-Y391*-R-HA-KpnI: 5’AAAGGTACCTAGGCGTAGTCGGGCACGTCGTAGGGGTAGAGGGGGTGGTCGGCT-3’) that introduced the reverse primer. The amplified product was digested with EcoRI and KpnI enzymes and subcloned into the mammalian pCXN2.1 vector at EcoRI and KpnI sites. The mutation c.1173C > G (p. Tyr391Ter) was introduced into the pCXN2.1- HA-IL17RA-Wt vector using the Quick-Change site-directed mutagenesis kit (Agilent Technologies). Similar to the cloning of IL17RA, cDNA sequences encoding the intracellular domain (IC) of IL17RC and IL17RD were amplified by PCR using forward and reverse primers. For IL17RC-Flag, the forward primer (IL17RC-F-EcoRI:5’CTCGAATTCGCCACCATGCCTGTGCCCTGGTTCTT-3’) and reverse primer (IL17RC-R-Flag-NheI): 5’AAAGCTAGCTACTTATCGTCGTCATCCTTGTAATCAGTCCCGTCCCCCGCCCCA-3’) were used. The HA-tag nucleotide sequences were introduced into the reverse primer. For all tags, it was introduced into the C-terminus. The PCR products were digested with EcoRI and NheI enzymes and subcloned into pCXN2.1 vector at EcoRI and NheI sites (pCXN2.1-IL17RC/D-HA).

Human embryonic kidney (HEK293T) cells were used according to the manufacturer’s instructions. 0.6 μg of vectors encoding N-terminal HA-tagged IL17RA (Wt or p. Tyr391Termutant [Mut]) or an empty pCXN2.1 vector, were transfected into wells.

For immunofluorescence (IIF) studies, HEK293T cells were transfected with expression vectors for IL17RA-Wt-HA, IL17RA-Y391*-HA, or an empty vector (50 ng each). Cells were stained with rabbit polyclonal anti-HA antibody (diluted 1:1,000; Abcam) and mouse monoclonal antipan-cadherin antibody (diluted 1:1,000; Abcam). For Co-IP assays, HEK293T cells were transfected with IL17RA-HA and IL17RC-Flag, as well as IL17RA-HA and IL17RD-Flag (1.0 μg each), using Lipofectamine 2000 (Life Technologies). Immunoprecipitation was performed using mouse anti-DDDDK (flag) agarose gels (MBL International). Cell culture, transfection, IIF, and Co-IP were performed as previously described (16).

The proband II.1 is a 5-year-old female child, the offspring of first-cousin consanguineous Lebanese parents ( Figure 1A ). She presented to the dermatologist at the age of one year with psoriasis on her scalp but no hair loss ( Figure 1B ), along with recurrent erythematous. At the time of presentation, she was not receiving any medical treatment; yet, she had persistent oral mucosal candidiasis, confirmed by a KOH smear showing candida albicans growth ( Table 1 ). She did not have any other infections, and her immune workup showed unremarkable findings, her White Blood Cell(WBC) count and Automated Differential were within the reference range [Neutrophils[0.8-7.2x10(9)/L], Lymphocytes [1.3-8x10(9)/L], Monocytes [0.1-1.1x10(9)/L], Eosinophils [0-0.7x10(9)/L], Basophils [0-0.2x10(9)/L]. The patient’s parents and brother underwent clinical assessment, which did not reveal any features of psoriasis or oral candidiasis, prompting a genetic workup.

The whole exome sequencing results of the patient and her parents yielded an average of 85,000 variants per sample. After applying a stringent filtering approach (see material and methods), 1,882 variants were retained for further analysis. Segregating with the phenotypes and assuming a recessive mode of inheritance, only homozygous variants were included in the first filter, yielding 206 homozygous variants ( Supplementary Figure S1 , see methods). Amongst these, a novel homozygous variant was detected on chromosome 22 (NM_014339.6 c.1173C>G, g.23439C>G) mapping to the IL17RA gene, resulting in a premature stop codon (p. Tyr391Ter). This variant was found in a homozygous form in the proband and in a heterozygous manner in her parents ( Figures 1C, D ) (ClinVar#SCV002520326). The variant is classified as pathogenic according to the American College of Medical Genetics and Genomics (ACMG) classification (18), with a Combined Annotation Dependent Depletion (CADD) score of 36 (19). It is predicted to have a deleterious effect with a tree vote score of (184|16), which may cause nonsensemediated decay NMD (-476 AA/more than 10% missing) according to MutationTaster (20). This variant is absent from gnomAD (21), ExAC (22), and our in-house database of 300 Lebanese exomes. While additional variants in HLA-DRB1 and NLRP3 genes ( Table 1 , Figure 1E ) may influence the phenotype(s), we focused on this variant to gain insights into its consequences on IL17RA protein function, as it is implicated in both phenotypes observed in the affected child. Of note, potentially pathogenic variants in genes previously associated with psoriasis such as PSORS1-13, MHC, I, TNF, and TRAF3IP2 (23–26), and genes implicated in CMC, including STAT1, AIRE, IL17F, IL17RC, and RORC (27–29) were absent in our patient.

To further elucidate the molecular mechanisms underlying both phenotypes and ascertain the diagnosis, RNA-seq was conducted on lesional skin biopsies from the indexed patient (II.1) and two non-lesional skin samples from healthy controls. The data distribution and sample correlation of the FPKM values are shown in Supplementary Figures S2A-C . We detected 1,136 differentially expressed genes (DEGs) with 967 upregulated and 169 genes downregulated when defined by a fold change (FCH) >2 and a false discovery rate (FDR) <0.01 ( Supplementary Table S2 , Supplementary Figures S2D-F ). Tables 2A, B show the top 15 upregulated and downregulated DEGs, respectively. The top upregulated genes include those encoding the S100 calcium-binding proteins (S100A8, S100A7/A, and S100A9) which exhibited the most significant increase in expression with a log2 fold change of 8.17. Additionally, proteolysis regulation molecules (SERPINB4, SERPINB3, PI3), matrix metalloproteinases (MMP1 and MMP3) and chemokines (CCL18) were highly upregulated. Conversely, the top downregulated genes include PPAR-fatty acids and lipid metabolism-related genes such as FADS2, which exhibited the most substantial decrease in expression, along with FABP7, ACSBG1, ELOVL3, PM20D, and DGAT2L6, keratinocyte related genes KRT79 and KRT2, Serpin Family A Member 12(SERPINA12), and (C1ORF68).

We identified 39 enriched pathways with an adjusted p-value (adj.p-value) of <0.05 ( Figure 2A ; Supplementary Table S3 ). Among the top 15 enriched pathways ( Table 3 ), T cell receptor signaling (adj.p-value 1.26E-07), chemokine signaling (adj.p-value 9.54E-07), cytokine-cytokine receptor (adj.p-value 1.55E-06), Staphylococcus aureus infection (adj.p-value 2.86E-06), and notably the IL17 signaling pathway (adj.p-value 1.97E-06) were upregulated ( Figure 2B ).

Interestingly, we found that the most significant canonical pathway is IL17A in Psoriasis [pvalue: 2.65E-08, -log (p-value: 7.58)] amongst 16 canonical pathways identified by IPA with a p-value of ≤ 0.05 ( Figure 2B ; Supplementary Table S4 ).

IL17R emerged as the master upstream regulator (p-value 3.29E-10) with a predicted activation in the pathway indicated by a z-score of 2, among 566 upstream regulators with a of p-value ≤ 0.05. ( Figures 2C, D ; Supplementary Table S5 ).

In terms of diseases categories, dermatological diseases (with a p-value between 1.44E-17 and 4.21E-02) were among the second top categories out of 73 identified ( Figures 2D, E ; Supplementary Table S6 ). Among these, 877 disease function annotations were identified with a p-value ≤ 0.05 ( Supplementary Table S7 ). Strikingly, under the “dermatological diseases” category, psoriasis emerged as the top annotated disease function with a p-value of 2.94E-12, involving 9 corresponding molecules ( Supplementary Table S7 ). We conducted additional analysis to identify and annotate molecules and biological functions that are common to psoriasis, fungal infection (associated with CMC), and inflammatory response. This analysis is detailed in Figure 2G and Supplementary Table S8 .

Moreover, to gain better insight into this increased IL17 pathway activity, we quantified the presence of the protein in the patient’s PBMCs.

We used flow cytometry on PBMCs from the patient, her sibling, and her parents, using a specific monoclonal antibody against the IL17RA protein. Results showed that the patient (II.1) expresses normal levels of IL17RA on her T lymphocytes, monocytes, and neutrophils ( Figure 3 ). This suggests that the mutation produces a stable truncated protein, which may be capable of homodimerizing and potentially enhancing the IL17RA downstream pathway. These findings prompted us to assess the structure and function of the mutated protein in vitro.

To assess the impact of the Y391* variant on the structural and functional properties of IL17RA, site-directed mutagenesis was performed on human IL17RA cDNA cloned into an HA-tagged expression vector. The generated plasmid was transfected into HEK293 and HaCat cells to examine the localization of the mutant protein and compared to the wild-type IL17RA (Wt).

Immunofluorescence studies revealed that both Wt-IL17RA protein and mutant protein IL17RA Y391* were localized to the cell membrane in both HEK293 and HaCat cells ( Supplementary Figures S2A, B ). To assess the physical interaction with IL17RC/D, Co-immunoprecipitation (IP) assays were performed. The results showed that the mutant p.Y391* retained its capability to bind to both IL17RC and IL17RD proteins. This finding confirms a potential deregulated signaling pathway that could contribute to the underlying psoriatic phenotype ( Figures 4A, B ).