Novel Insights: A Novel PHIP Variant in a Family with Severe Early-Onset Obesity

Abstract

Introduction

Severe childhood obesity can be caused by pathogenic variants in several genes involved in monogenic and syndromic obesity. Recently, heterozygous variants in pleckstrin homology domain interacting protein (PHIP) have been identified in patients with obesity as part of Chung-Jansen syndrome.

Case Presentation

The index patient is a 5-year-old boy with severe obesity since 1 year of age, developmental delay, facial dysmorphism, and behavior problems. Whole-exome sequencing identified a novel missense variant in PHIP (c.3182C>A, p.Ala1061Glu) in the index patient. Further genetic testing in family members revealed segregation of the same PHIP variant in the brother and mother, who both presented with severe childhood obesity and developmental delay or learning difficulties. The PHIP missense variant was predicted pathogenic by multiple in silico tools and affects a highly conserved residue.

Conclusion

Early-onset obesity may be monogenic. Our finding expands the spectrum of disease-causing variants in PHIP and demonstrates variable intrafamilial clinical expressivity and severity. Screening for PHIP variants should be included in genetic testing in patients with severe early-onset obesity.

Established Facts

• Severe early-onset obesity can be caused by pathogenic variants in multiple genes involved in monogenic and syndromic obesity.

• Heterozygous variants in pleckstrin homology domain interacting protein (PHIP) have recently been identified in patients with obesity as part of Chung-Jansen syndrome.

Novel Insights

• We describe a novel PHIP missense variant that demonstrated variable intrafamilial phenotypic expression with severe childhood-onset obesity in all affected family members.

• This case highlights the importance of genetic testing in children with severe obesity. Screening for PHIP variants should be included as part of genetic testing in severe early-onset obesity.

• Liraglutide could be a potential treatment option for patients with severe obesity caused by PHIP variants

Introduction

Genetic obesity can be caused by pathogenic variants in single genes, copy number variants, or methylation defects. Monogenic obesity has an estimated prevalence of 5–10% in patients with severe childhood obesity [1–4]. Endocrine Society guidelines for pediatric obesity recommend screening for genetic forms of obesity in children with severe obesity before 5 years of age and who have clinical features of syndromic obesity and/or family history of severe obesity [5]. Finding genetic causes of severe obesity is of great importance to increase knowledge of underlying mechanisms, identify targets for potential treatments, allow genetic counseling, and support patients against the social stigmata of obesity.

Heterozygous variants in pleckstrin homology domain interacting protein (PHIP) are associated with Chung-Jansen syndrome (OMIM#617991), a rare genetic disorder characterized by developmental delay, intellectual disability/learning difficulties, obesity, dysmorphic features, and behavioral problems. In recent years, an increasing number of reports of patients with pathogenic variants in PHIP have been published [6–17].

PHIP is a family member of DDB1-CUL4-associated factor proteins, and it is involved in neurodevelopmental processes, E3 ubiquitination, and regulation of the insulin and insulin-like growth factor signaling pathway [8, 18, 19]. In addition, PHIP has shown to be involved in energy homeostasis as PHIP enhances transcription of POMC, a neuropeptide involved in appetite regulation [11].

We present a boy with very early-onset severe obesity, developmental delay, facial dysmorphism, and behavioral problems. Whole-exome sequencing (WES) revealed a novel missense PHIP variant in the index patient and in two family members with severe obesity and developmental delay or learning disabilities.

Case Presentation

Patient 1 (II-1)

The index patient is a 5-year-old boy born at 40 weeks of gestation with birth length 50 cm (−0.8 SD), birth weight 3605 g (−0.5 SD), and head circumference 36 cm (+0.3 SD). He presented with severe obesity already at the age of 1 year with length 81.4 cm and weight 16.0 kg (weight-for-length +43%). At the age of 3.6 years, his height was 108.8 cm (+2.0 SD) and weight 32.0 kg (body mass index [BMI] 27 kg/m² [+7.4 SD], ISO-BMI 74, and weight-for-height 72%). Presently at 5.4 years, his height was 120.5 cm (+1.3 SD) and weight 34.1 kg (BMI 23.5 kg/m² [+4.6 SD], ISO-BMI 43, and weight-for-height +48.4%). Figure 1 presents the growth charts of the index patient. He had severe delay in speech and language development. From the age of 3 years, he received weekly speech therapy and from the age of 4 years weekly occupational therapy to support development of skills that require motor precision and motor planning. Due to heavy nocturnal snoring and mouth breathing, a polysomnography was conducted, which did not reveal any significant sleep-related breathing disorder or impact on sleep quality. Hearing and vision were normal. He had behavioral problems. He presented with facial dysmorphism; a short nose, anteverted nares, prominent forehead, and thick earlobes. He had brachydactyly and 5th finger clinodactyly. The testicles were located distally in the inguinal canal, and the liver exhibited slight fat on abdominal ultrasound examination. Fasting glucose level was 5.8 mmol/L, indicating prediabetes. Insulin levels, lipid profile, and liver tests were normal. In further studies, thyroid tests, salivary cortisol at night, array-CGH, U-oligosaccharides, U-glycosaminoglycans, and U-organic acids were conducted with normal results.

Fig. 1.

Growth charts of patient 1 (index patient). a Height- and weight-for-age. b Weight-for-height (%).

Patient 2 (II-3)

The index patient’s brother is 15 years of age and was born at 39+2 weeks of gestation with a birth length of 52.5 cm (+1.3 SD), a birth weight of 3,960 g (+1.2 SD), and a head circumference of 37 cm (+1.6 SD). He developed severe obesity at 2 years of age with a height of 95 cm (+1.9 SD) and a weight of 20.4 kg (BMI 22.6 kg/m² [+4.2 SD], ISO-BMI 50.5, and weight-for-height 40.6%). Presently at 15 years of age, his height was 177.1 cm (+0.6 SD) and weight 136.5 kg (BMI 43.5 kg/m² [+4.3 SD], ISO-BMI 41, and weight-for-height 112%), and waist circumference 130 cm (waist-to-height ratio 0.73). Figure 2 presents his growth charts. He suffered from constipation during childhood. At 3 years of age, he was diagnosed with severe dysphasia. He had difficulties with cognitive skills. He had gross motor delay and toe-walking from early stages. Achilles tendon lengthening surgery was performed at the age of 6 years with good treatment response. He had problems with balance and coordination. Hearing was normal. He presented with short nose, anteverted nares, and prominent eyebrows. He had normal chromosomes and he was investigated for Prader-Willi syndrome with negative results. Thyroid hormone levels were normal. He was diagnosed with impaired glucose tolerance and fatty liver at 15 years of age. He had normal insulin levels. His lipid profile showed high triglyceride (1.8 mmol/L) and low HDL (0.99 mmol/L). Liraglutide treatment was commenced at the age of 15 years. Thus far, he has lost 8.2 kg (6%) of his weight in the first 5 months with liraglutide treatment and glucose tolerance and triglyceride levels have normalized.

Fig. 2.

Growth charts of patient 2 (brother of index patient). a Height- and weight-for-age. Liraglutide treatment was started at 15.3 years of age. b Weight-for-height (%).

Patient 3 (I-2)

The index patient’s mother presented with severe obesity already in childhood. She had learning difficulties and reading problems in school. She had a short nose and anteverted nares. Her height was 160 cm (−0.9 SD). She had gestational diabetes during pregnancy with the elder son (patient 2). She underwent gastric bypass at 30 years of age with a preoperative BMI of 52 kg/m². Eleven years after surgery, her BMI was 35 kg/m² and total weight loss (%TWL) was 32%.

Genetic Analyses and Results

This study was carried out at Children’s Hospital, Helsinki University Hospital, and Folkhälsan Research Center in Helsinki, Finland. This study is part of a research project investigating genetic causes of metabolic bone diseases, including growth disturbances. Written informed consent was obtained from all adult participants. In case of minors, the children gave an assent and a written informed consent was obtained from the parents. Ethical approval for this study was obtained from the Research Ethics Committee of the Hospital District of Helsinki and Uusimaa. WES was performed at Blueprints Genetics (Espoo, Finland). WES included sequence and copy number variation (CNV) analyses. Median read depth for the analysis was 191-fold, and 99.79% of target nucleotides were covered with >20-fold read depth. Sequence reads were aligned to reference human genome (GRCh37/hg19). Burrows-Wheeler Aligner software was used for read alignment and variant calling was performed using GATK.

CNV analysis did not detect any known disease-causing or novel CNVs that were considered pathogenic. A novel heterozygous missense variant NM_017934.7 (PHIP) c.3182C>A, p.Ala1061Glu was identified in the index patient (II-1), the affected brother (II-3), and the affected mother (I-2). This variant was not found in the unaffected father (I-1) or in the unaffected siblings (II-2 and II-4) (Fig. 3). The identified variant has not been observed in gnomAD or 1000 Genomes and has not been reported in ClinVar (version 2023-03-31) or HGMD Professional (version 1.2024, accessed on 2024-06-01). This variant was predicted disease causing/damaging by MutationTaster2, SIFT, M-CAP, MVP, Provean, and PrimateAI and has a CADD score of 26.6. This variant affects a highly conserved residue (PhyloP 100-way score of 7.5). The variant is located in the pleckstrin-homology interacting domain (PBR), a region that mediates the interaction with insulin-receptor substrate (IRS-1). Analysis of the mutant protein by HOPE indicated that the mutant residue is bigger, more hydrophilic and introduces a negative charge compared to the smaller, neutral charged wild-type residue. MutScore was 0.832 and AlphaMissense prediction was likely pathogenic (score 0.938). All these supported the variant to be disease-causing. The PHIP missense variant was confirmed by Sanger sequencing.

Fig. 3.

Pedigree of the family with novel PHIP variant p.Ala1061Glu. Square, male; circle, female. Filled symbols represent affected individuals.

Discussion

We describe a novel heterozygous missense variant (c.3182C>A, p. Ala1061Glu) in PHIP in three family members with severe childhood obesity and variable degree of developmental delay and learning difficulties. The PHIP missense variant was predicted pathogenic by multiple in silico tools and affects a highly conserved residue.

In recent years, there have been an increasing number of reports of pathogenic PHIP variants in patients with Chung-Jansen syndrome (OMIM#617991), a rare autosomal dominant disorder characterized by developmental delay, intellectual disability/learning difficulties, obesity, dysmorphic features, and behavioral abnormalities. The previously described PHIP variants include missense, splicing, nonsense, frameshift, and larger deletions [6–17]. PHIP has a gnomAD pLI score of 1 and missense Z-score of 5.14, indicating intolerance to loss-of-function and missense variation. So far, no mutational hot spots have been found [10].

Notably, our 3 patients had very early-onset obesity, the index patient presented with severe obesity already at 1 year of age, his brother at 2 years of age and their mother reported that she suffered from obesity already in childhood. The index patient and his brother both had facial dysmorphism and developmental delay. In addition, the index patient presented with hypotonia, brachydactyly, and 5th finger clinodactyly and the brother had problems with balance and coordination. The mother reported learning difficulties and reading problems in school. Our patients demonstrated intrafamilial phenotypic variability. The phenotype of our patients is consistent with the previously reported phenotype in patients with PHIP variants. The largest cohort of patients with Chung-Jansen syndrome has been reported by Sudnawa et al. [17]. They described 47 patients with pathogenic/likely pathogenic PHIP variants, most of them de novo. The most common clinical characteristics included developmental delay (85%), hypotonia (78%), visual problems (66%), obesity/overweight (56%), attention-deficit/hyperactivity (51%), constipation (49%), and anxiety (47%). They observed obesity/overweight from the age of 7 years and an increasing frequency of obesity/overweight with increasing age [17]. A study by Kampmeier et al. [14] with 23 patients showed developmental delay in 96%, intellectual disability in 91%, behavioral problems in 87%, and obesity/overweight in 70% of their patients. One-third of their patients presented with obesity/overweight at 5 years of age and they observed an increase in obesity/overweight during puberty. In addition, they reported craniofacial dysmorphic features in all patients. The most common dysmorphic facial features were large ears/earlobes, anteverted nares, and prominent eyebrows. In another study, 74% of 23 individuals with PHIP variants presented with overweight and all patients had developmental delay/intellectual disability [8]. Marenne et al. [11] reported that all patients in their study presented with severe obesity, but not all had developmental delay and they suggested that some patients with PHIP variants may present with obesity alone.

The biological functions of PHIP are still incompletely understood. PHIP plays an important role in several processes including neurodevelopment, E3 ubiquitination, and insulin and insulin-like growth factor signaling [8, 18, 19]. Morgan et al. [20] demonstrated that PHIP is a histone H3 lysine 4 (H3K4) methylation binding protein and a chromatin-associated cullin ring ligase-4 (CRL4) ubiquitin substrate receptor and is required for CRL4 recruitment to chromatin. Loss of PHIP impairs CRL4 interaction with chromatin. Tirado-Class investigated two PHIP missense variants using cell-based assays and found that the variants disrupted replication fork stability and genome integrity [15]. A recent study investigated genome-wide DNA methylation profiles in affected individuals with PHIP variants. The authors reported a novel DNA methylation episignature for Chung-Jansen syndrome. Interestingly, they also found similarities between the DNA methylation patterns of patients with PHIP variants and that of clinically and molecularly related Börjeson-Forssman-Lehmann syndrome (caused by PHF6 variants) and White-Kernohan syndrome (caused by DDB1 variants). This episignature may provide an additional tool to improve the interpretation of genetic variants and molecular diagnosis [21].

In addition, PHIP has also been shown to be involved in the expression of POMC and Marenne et al. [11] demonstrated that pathogenic PHIP variants repressed POMC expression. POMC is part of the leptin-melanocortin pathway and has an important role in the regulation of appetite. Figure 4 presents the main elements of leptin-melanocortin pathway. Leptin increases the production of POMC, which is processed to α-MSH. α-MSH stimulates MC4R and leads to reduced food intake and increased energy expenditure. Pathogenic variants in genes involved in the leptin-melanocortin pathway can lead to monogenic obesity. Biallelic variants in POMC cause early-onset obesity, adrenal insufficiency, and red hair pigmentation. Setmelanotide has been approved for the treatment of POMC deficiency [22]. Future studies are needed to investigate whether patients with pathogenic PHIP variants could benefit from existing targeted treatments. Modulating the pathways that control recruitment of PHIP to chromatin has also been suggested as a potential therapeutic target in Chung-Jansen syndrome [20].

Fig. 4.

Leptin-melanocortin pathway. LEPR, leptin receptor; TUB, Tubby bipartite transcription factor; SH2B1, Src homology 2 B adapter protein; POMC, Pro-opiomelanocortin; PHIP, pleckstrin homology domain interacting protein; PCSK1, proprotein convertase subtilisin/kexin type 1; CPE, carboxypeptidase E; α-MSH, α-melanocyte-stimulating hormone; MRPA2, melanocortin 2 receptor accessory protein 2; MC4R, melanocortin-4 receptor; SIM1, single-minded homolog 1; BDNF, brain-derived neurotrophic factor; NTRK2, neurotrophic tyrosine kinase receptor type 2.

The effect of bariatric surgery on long-term weight loss in patients with PHIP variants is unclear. The index patient’s mother had undergone gastric bypass with a %TWL 32% 11 years after surgery. The weight loss of our patient is higher compared to mean %TWL of patients without gene variants in the leptin-melanocortin pathway 10 years after gastric bypass [23]. The index patient’s brother started liraglutide therapy at 15 years of age with initially good response regarding weight loss (6%) during the short follow-up of 5 months. The weight loss in our patient is greater than that observed in a clinical trial of liraglutide treatment in adolescents aged 12–18 years [24]. More studies on patients with PHIP variants are needed to evaluate the long-term weight loss effects of bariatric surgery and liraglutide treatment.

We acknowledge the lack of functional studies as a limitation of our study. Functional studies are necessary to determine the pathogenicity of the variant. Further studies are needed to investigate the molecular mechanisms in PHIP-related obesity to develop treatment strategies for these patients.

In conclusion, our report highlights the importance of considering a genetic cause in early-onset obesity. The findings expand the molecular spectrum of disease-causing variants in PHIP and demonstrate variable phenotypic expression. Screening for PHIP variants should be included as part of genetic testing in patients with severe early-onset obesity.

Acknowledgments

We thank the family for participation in this study.

Statement of Ethics

This study protocol was reviewed and approved by the Hospital District of Helsinki and Uusimaa Regional Committee on Medical Research Ethics (Approval No. HUS 404/2018). Written informed consent was obtained from participants and their parents to participate in the study and for the publication of the details of their medical case.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was supported by Sigrid Jusélius Foundation, Novo Nordisk Foundation, Academy of Finland, Folkhälsan Research Foundation, Foundation for Paediatric Research, Finska Läkaresällskapet, Päivikki ja Sakari Sohlberg Foundation, and Stiftelsen Dorothea Olivia, Karl Walter och Jarl Walter Perkléns minne.

Author Contributions

O.M. and P.L.: study design, genetic data analysis, and drafting manuscript. N.V., K.A., and J.K.: clinical characterization of the patients. All authors edited and approved the final manuscript.

Funding Statement

This study was supported by Sigrid Jusélius Foundation, Novo Nordisk Foundation, Academy of Finland, Folkhälsan Research Foundation, Foundation for Paediatric Research, Finska Läkaresällskapet, Päivikki ja Sakari Sohlberg Foundation, and Stiftelsen Dorothea Olivia, Karl Walter och Jarl Walter Perkléns minne.

Data Availability Statement

The data that support the findings of this study are not publicly available due to information that could compromise the privacy of the study subjects. Further inquiries can be directed to the corresponding author.