Eczematoid lesions and extensive molluscum contagiosum in a child with novel <scp> <i>CARMIL2</i> </scp> defect

Combined immunodeficiencies (CIDs) are a group of inborn errors of immunity characterized by defects in development and function in both B and T cell immunity. Patients with CIDs have increased susceptibility to opportunistic infections; however, the disease severity is lower compared to severe combined immunodeficiency (SCID).1 CARMIL2 (capping protein regulator and myosin 1 linker 2) deficiency is one of the recently described autosomal recessive forms of CID2. Herein we describe a child with CARMIL2 deficiency with eczematoid lesions, and extensive molluscum contagiosum with a novel splice-site mutation. A 4-year-old male child, third born to a second-degree consanguineously married couple from the Northern part of India (Jammu and Kashmir) presented to us with recurrent episodes of itching, and redness over the face and trunk since 6 months of age. He also had one episode of pneumonia at 2 years of age requiring hospital admission and intravenous antimicrobials. On clinical examination, he had eczematous lesions in face, trunk, both flexor and extensor aspects of upper and lower limbs. He also had multiple molluscum lesions over the face, scalp, and back (Figure 1). Onychomycosis was noted in the fingers of lower and upper limbs. He had right sided club foot. Other systemic examination was unremarkable. Clinical differential diagnoses of autosomal recessive Hyper-IgE syndrome (DOCK8 deficiency) and Wiskott–Aldrich syndrome were considered and evaluated for the same. Blood investigations revealed anemia (Hb–88 g/L), leukocytosis with eosinophilia (18.58 × 109/L, neutrophils 35%, lymphocytes 52%, monocytes 7%, eosinophils 6%), and normal platelet counts. He also had elevated inflammatory parameters at the time of evaluation (erythrocyte sedimentation rate-32 mm/h. and C-reactive protein-46.8 mg/L). Serology for human immunodeficiency virus was negative. Immunological investigations revealed a mild reduction in CD4+ T lymphocytes and a reduction in CD4/CD8 ratio (Table S1). The percentage of CD4 + CD127-CD25+ cells was low (2.7%) compared to the age-matched control (6.9%). The absolute counts of T-regulatory cells (CD4 + CD127-CD25+FOXP3+) cells were also reduced in the patient compared to the age-matched control (Figure 2A). (A) Flow-cytometry analysis of T-regulatory cells (CD4+25+127-FOXP3+) showing reduced counts in patients compared to healthy control. (B) Flow-cytometric analysis of Lymphocyte proliferation assay using carboxyfluorescein succinimidyl ester (CFSE) dye showing reduced CD3+ T cell proliferation in patient in response to both phytohemagglutinin (PHA) and CD3/CD28 stimulation. Peripheral blood mononuclear cells (PBMCs) were isolated and the assay was done using carboxyfluorescein diacetate succinimidyl ester (CSFE) staining and stimulation with phytohemagglutinin (PHA) and CD3/CD28 beads. Cells were acquired through a flow-cytometer and analysis was done by Kaluza software. The assay showed reduced CD3+ T cell proliferation in patient in response to both phytohemagglutinin (PHA) and CD3/CD28 stimulations (Figure 2B). A whole-exome sequencing done by Next Generation Sequencing showed a novel homozygous splice-site defect in CARMIL2 (CARMIL2:NM_001013838:exon21:c.2082+1G>A, p.Val642Lysfs*5). The variant is confirmed by Sanger sequencing and the parents are found to be heterozygous carriers for the same variant. The variant allele frequency in gnomAd is 0. The semi-automated ACMG classification of the variant is pathogenic (P3_Strong) according to GeneBe software. The CADD phred score was 33. MPA impact predicted “high splice” and the score was 10. SPiP predicted that the variant has a 98.4% risk of altering the splice site. RT-PCR analysis and sequencing of the CARMIL2 transcripts of the patient confirmed the aberrant splicing resulting in skipping of exon-21 (Figure 3). The child underwent cryotherapy for the molluscum lesions, and he was started on topical moisturizers for the eczema. He was also commenced on oral cotrimoxazole and itraconazole prophylaxis and replacement intravenous immunoglobulin therapy. At second year follow-up, he was keeping in good health; however, few molluscum lesions in the face persisted. The parents were counseled for hematopoietic stem cell transplantation. The CARMIL family includes three human paralogs: CARMIL-1, CARMIL-2, and CARMIL-3. CARMIL-2, previously known as RLTPR (RGD motif, leucine-rich repeats, tropomodulin domain, and proline-rich containing murine homolog to human CARMIL2) is a regulator of actin capping protein. The protein is expressed in the cytoplasm, especially in cells of skin, lymphoid tissue, and gastro-intestinal system. In immunological cells, it is essential for CD-28 mediated T cell co-stimulation, modulation of signaling in B, NK, and some myeloid cells, and coordination of actin polymerization in the immunological synapse.2 CARMIL-2 also maintains cellular polarity by modulating microtubules and intermediate filaments.3 Clinical phenotype in patients with CARMIL2 deficiency is broad ranging from skin manifestations in the form of allergic eczema, recurrent bacterial, viral, and fungal infections, very early-onset inflammatory bowel disease without overt immunodeficiency, and eosinophilic gastrointestinal disorders. Very rarely, Epstein–Barr virus (EBV) associated smooth muscle tumor with growth retardation is also observed. Sorte et al. described childhood-onset molluscum contagiosum, warts, and dermatitis in four patients from three Norwegian family and all patients had a homozygous missense founder mutation in CARMIL2.4 Our patient had predominant skin manifestations in the form of early-onset of eczema, cheilitis, extensive molluscum contagiosum, and onychomycosis. Flow cytometry of the affected patients shows a reduction in Tregs and CD4+ T cells in peripheral blood. Wang et al. and Sorte et al. demonstrated the importance of CARMIL-2 in activation of CD4+ T cell, but not of CD8+ T lymphocytes.4, 5 The depletion of Tregs and defective CD20 co-stimulation probably results in allergic and hypersensitivity disorders in these patients.2 Dysfunction of CD28-CARMIL2-CARD11 pathway also leads to impaired development of Th1/Th17 and hyperactivation of Th-2 lymphocytes. Defective memory B-cell formation with poor specific antibody response has also been reported in these patients. Our index child had a decreased proportion of Tregs and CD3 + CD4+ T lymphocytes and reduced CD3+ T cell proliferation in patient in response to PHA and CD3/CD28 stimulation. We could not do an immunoblot or flow cytometry to show the reduced CARMIL2 protein expression in our patient and this is the limitation of our report. In a cohort of 15 patients with CARMIL2 defect, Kolukisa et al. reported missense, frameshift, and splice-site variants 11, 3, and 1 patient, respectively.6 In this cohort, R385T mutation was reported in three distinct families. The clinical presentation is variable even within the members of same family, irrespective of protein expression and mutation. Therefore, the authors concluded that there was no strong genotype–phenotype relationship in CARMIL2 defect. A recent large multi-centric cohort of 89 CARMIL2 deficiency patients revealed that splice-site and missense variants are common in CARMIL2 defect.7 Moreover, they also showed that phenotype is completely penetrant by the age of 10 years.7 To conclude, we report a novel homozygous spice-site variant (c.2082+1G>A) in CARMIL2 in a child who presented with eczema and extensive molluscum contagiosum. CIDs such as CARMIL2 deficiency must be considered and actively worked up in patients with early-onset eczema and skin viral infections. Sathish Kumar Loganathan: Data curation; writing – original draft; methodology; writing – review and editing. Sumit Goel: Investigation; writing – review and editing; formal analysis. Sanjib Mondal: Investigation; writing – original draft; supervision. Daniel Leung: Investigation; methodology; validation; software; formal analysis. Koon Wing Chan: Data curation; formal analysis; methodology. Yu Lung Lau: Methodology; formal analysis; data curation; supervision; resources. Pandiarajan Vignesh: Conceptualization; investigation; funding acquisition; writing – original draft; writing – review and editing; formal analysis. Amit Rawat: Writing – review and editing; project administration; resources. We acknowledge the support from pediatric subspecialities in the Department of Pediatrics, Advanced Pediatric Centre, PGIMER, Chandigarh, and other departments of PGIMER, Chandigarh that helped in the diagnosis and clinical management of the cases. We thank the junior residents of Department of Pediatrics who have assisted in the clinical management. We also thank faculty members, research fellows, and laboratory technicians of the Pediatric Immunology laboratory, Allergy Immunology Unit, Advanced Pediatrics Centre, PGIMER, Chandigarh who have assisted in the laboratory experiments. We especially thank Dr. Jhumki Das who substantially helped us in the laboratory experiments. P.V. received funding from the Indian Council of Medical Research, Government of India, for carrying out the project entitled, “Clinical and Molecular Evaluation of Inborn Errors of Immunity in a tertiary care centre of North India” (No. 33/29–2/2019-TF/Rare/BMS). All authors declare there are no potential conflicts of interest. This study was performed in line with the principles of the Declaration of Helsinki. The Departmental Review Board of the Advanced Pediatrics Centre approved this manuscript (No. 58–22, dt. 26.07.2022). Informed consent was obtained from the parents of the child included in this report. All authors have given final consent for publication. Parents consented for the publication of the clinical details of their child in an academic journal. The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/pai.70003. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request. Table S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.