Advertisement
Research LetterDermatologyGenetics
Open Access | 
10.1172/JCI195809
Germline inactivation of tumor suppressor BAP1 is associated with white spotting
Ching-Ni Njauw,1 Zhenyu Ji,1 David I. Latoni,2 Jose Mari Villa-Gonzalez,1,3 Shelley McCormick,4 Raj Kumar,5 Dmitrii Usoltsev,6 Mykyta Artomov,6 Boyi Gan,7 and Hensin Tsao1,4
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by
Njauw, C.
in:
PubMed
|
Google Scholar
|
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by Ji, Z. in: PubMed | Google Scholar
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by Latoni, D. in: PubMed | Google Scholar
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by Villa-Gonzalez, J. in: PubMed | Google Scholar
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by McCormick, S. in: PubMed | Google Scholar
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by Kumar, R. in: PubMed | Google Scholar
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by Usoltsev, D. in: PubMed | Google Scholar
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by Artomov, M. in: PubMed | Google Scholar
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by Gan, B. in: PubMed | Google Scholar
1Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, Massachusetts, USA.
2Department Dermatology, Univ Puerto Rico School of Medicine, San Juan, Puerto Rico.
3Department Dermatology, Cruces University Hospital, Bizkaia, Spain.
4Mass General Brigham Cancer Institute, Boston, Massachusetts, USA.
5Division Hematology/Oncology, Department Medicine, MGH, Boston, Massachusetts, USA.
6The Steve and Cindy Rasmussen Institute for Genomic Medicine, and Dept Pediatrics, Nationwide Children’s Hospital, Columbus, Ohio, USA.
7Department Experimental Radiation Oncology, MD Anderson Cancer Center, Houston, Texas, USA.
Address correspondence to: Hensin Tsao, Massachusetts General Hospital, Edwards 211, 50 Blossom Street, Boston, Massachusetts 02114, USA. Email: htsao@mgh.harvard.edu.
Authorship note: CNN and ZJ are co–first authors.
Find articles by Tsao, H. in: PubMed | Google Scholar
Published January 2, 2026 - More info
J Clin Invest. 2026;136(1):e195809. https://doi.org/10.1172/JCI195809.
© 2026 Njauw et al. This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
To the Editor: The BAP1 tumor predisposition syndrome (BAP1-TPDS), caused by germline variants in BAP1, increases susceptibility to multiple malignancies, including cutaneous melanoma (CM) and uveal melanoma (1). While BAP1 functions as a putative melanoma tumor suppressor, its full role in melanocyte survival and proliferation is poorly understood. We describe a white spotting phenotype in patients with germline BAP1 variants. Case 1 is a female patient with a germline p.Asp236Glyfs*7 variant who developed CM and breast cancer at ages 49 and 52, respectively (Figure 1A). She had a white forelock since birth but lacked eye or hearing problems. Direct visual and Wood’s light inspection of 13 more BAP1 variant carriers identified another 7 patients with chronic poliosis or depigmentation (Figure 1A).
Figure 1BAP1 inactivation and pigmentation. (A) Patient details and images of white spotting (demonstrated by Wood’s light in Cases 3,4,5,7). (B) Top genes correlated with BAP1 expression in normal human melanocytes (NHMs), the TCGA_SCKM and Lund cohorts, as ranked by Spearman coefficient, were subjected to Gene Ontology Biological Processes (GO-BP), GO-Cellular Compartment (GO-CC), and Human Phenotype Ontology (HPO) analysis; all GO and HPO analyses exhibited FDR < 0.05. (C) The 500 genes most correlated with BAP1 expression were subjected to HOMER analysis. Pigmentation-related transcription factors, MITF and TFE3, are highlighted in blue. (D) Suppression of Bap1 by sh(BAP1) in 3 melanoma cell lines leads to decreases in SOX10 and MITF, reduced cellular proliferation (Day 0 = 1; P values determined by t test on Day 5), and pigmentation loss. (E) Skin phenotypes of various Bap1 genotypes. Animal 1 (Bap1fl/fl) and Animal 2 (Tyr:CreA/Bap1fl/WT) have no visible phenotype. Animals 3–5 (Tyr:CreA/Bap1fl/fl) showed depigmentation of the fore- and hindpaws (red circles) while animals 4 and 5 had white spotting on the midabdomen. (F) Melanocyte quantitation (Sox10+ cells per hair follicle (HF)) shows significant reductions in the number of Sox10+ melanocytes (red arrows) in the depigmented abdominal area, ear, and dorsal paw of Tyr:CreA/Bap1fl/fl mouse compared with the Bap1fl/fl mouse (2-tailed t test).
To establish a biological link, we examined transcriptomes from normal human melanocytes (NHM; n = 308 samples from 35 patients) and tumors in the TCGA_SKCM (n = 363 samples) and the Lund melanoma cohorts (n = 214 samples) to determine if correlations between BAP1 expression levels and melanocyte lineage specifying genes exist (Supplemental Table 1 and Supplemental Figure 1A; supplemental material available online with this article; https://doi.org/10.1172/JCI195809DS1). Among the piebaldism and Waardenburg syndrome genes (Supplemental Figure 1B), SOX10 and BAP1 expression exhibited the highest correlation in all 3 datasets, while KIT and MITF levels also correlated with BAP1 levels, albeit weaker (Supplemental Figure 1B).
We then ranked all genes by their BAP1 correlation coefficient and performed Gene Ontology (GO) analysis (Figure 1B). The GO biological process terms most positively correlated with BAP1 expression were PIGMENTATION (Figure 1B and Supplemental Table 1) and RESPONSE TO IL-4 for NHM, PIGMENT METABOLIC PROCESS and PIGMENTATION for the TCGA_SKCM samples and PIGMENTATION and ENDOSOMAL TRANSPORT for the Lund specimens. PIGMENT GRANULE ranked among the top 2 GO-cellular compartment terms in all 3 datasets. Additionally, pigmentation disorders also ranked at the top of the Human Phenotype Ontology database (Figure 1B) among BAP1-correlated genes for NHM and melanoma tumors. HOMER analysis of the top 500 genes most correlated with BAP1 (Figure 1C) showed motif enrichments for MITF and TFE3, transcription factors regulating melanocyte lineage specification (2), in all 3 cohorts. BAP1 levels were also highly correlated with a curated set of 110 known or putative MITF targets (Supplemental Figure 1C and Supplemental Figure 2).
Functionally, we previously showed (3) that BAP1 depletion reduced melanoma cell viability and tumorigenicity, paradoxical to its putative suppressor role. With DepMap, we confirm that CRISPR-mediated deletion of BAP1 in 96 melanoma lines is associated with a loss of melanoma viability (Supplemental Figure 3). We subsequently depleted BAP1 in 3 melanoma cell lines and observed marked reductions in the protein levels of Sox10 and Mitf, cellular growth, and cellular pigmentation (Figure 1D and Supplemental Figure 4). Using Tyr:CreA with Bap1fl/fl mice, we also tested if melanocyte-specific BAP1 deletion can lead to a white spotting phenotype in vivo. No notable skin phenotypes were identified in either parent strains or the Tyr:CreA/Bap1fl/+ mice. However, all Tyr:CreA/Bap1fl/fl had depigmented paws (Figure 1E; red circle), and 3 of 4 had a white abdominal spot (Figure 1E). Compared with Bap1fl/fl mice, Tyr:CreA/Bap1fl/fl animals exhibited significant follicular melanocyte dropout in the depigmented zone (Figure 1F), ear, and dorsal foot; no tumors developed in any of the mice.
Nontumorigenic phenotypes have not been well characterized in the BAP1-TPDS. We provide a patient-to-mouse description of a nontumor phenotype for the BAP1-TPDS, and our evidence suggests that BAP1 may impinge on the migration, differentiation, survival, and melanization of follicular melanocytes. While the case series is small, the extreme rarity of the BAP1-TPDS precludes large-scale analysis, such as genotype-phenotype correlations. A pigmentation phenotype has not been described in previous murine models of Bap1 inactivation, as Bap1–/– embryos do not survive past day E9.5 (4). Heterozygous Bap1+/– are more susceptible to myeloid transformation (4) and mesotheliomas (5). Suppression of Bap1 in Xenopus during development also disrupts cell identity and ectodermal, mesodermal, and neural crest abnormalities (6). Thus, BAP1 may regulate melanocyte migration, identity, and pigmentation. Lastly, in tuberous sclerosis, hypopigmented macules and tumors occur with tuberin loss and mTOR activation, phenotypes reversed with topical rapamycin (7), thereby raising the possibility for rapamycin in BAP1-associated neoplasia.
Copyright © 2025 American Society for Clinical Investigation
ISSN: 0021-9738 (print), 1558-8238 (online)