Identification of a Cowden syndrome patient with a novel PTEN mutation and establishment of patient-derived induced pluripotent stem cells

Cowden syndrome (CS) is an autosomal dominant inherited disorder characterized by multiple hamartomas in various organs such as the mucosa, skin, and gastrointestinal tract. Patients with CS are at high risk for breast and thyroid cancers. Phosphatase and tensin homolog deleted on chromosome 10 (PTEN) is a tumor suppressor gene that negatively regulates the AKT pathway, and PTEN mutations are known to be the major causes of this syndrome. However, the pathogenesis of this syndrome has not been clarified. Here, we present a case of a Japanese woman with multiple oral polyps, breast cancer, and thyroid cancer who was clinically diagnosed with CS. We obtained DNA and RNA samples from the patient’s peripheral blood mononuclear cells (PBMCs) and buccal mucosa tumor. Next-generation sequencing revealed novel germline mutations (c.1020delT and c.1026G > A) in exon 8 of PTEN. Sanger sequencing identified no PTEN transcript from the mutant allele. Furthermore, CS-specific induced pluripotent stem cells (CS-iPSCs) were established from PBMCs of the patient under feeder- and serum-free culture. Compared with healthy PBMCs and iPSCs, both of the CS-derived PBMCs and CS-iPSCs exhibited significantly reduced expression of the PTEN transcript. The transcriptional variant, PTENδ, was increased in CS-iPSCs, suggesting that it may be the cause of the disease. Supplementary Information The online version contains supplementary material available at 10.1007/s11626-021-00637-8.


Introduction
Cowden syndrome (CS) (MIM 158,350) is a disease that causes hamartoma neoplasms in multiple organs, including the skin, mucous membranes, and gastrointestinal tract (Eng, 1997). It is a rare autosomal dominant genetic disorder that occurs in 1 per 250,000 people. Throughout their lifetime, CS patients have an increased risk of developing breast and thyroid cancer, specifically 50-85% and 30-40%, respectively (Tan et al., 2012). Because of the high risk of developing cancer, the National Comprehensive Cancer Network (NCCN) clinical practice guidelines in the U.S.A. recommend CS patients undergo careful medical testing and genetic counseling (NCCN.org, 2019). It is known that most cases of CS result from a mutant phosphatase and tensin homolog deleted on chromosome 10 (PTEN) allele (MIM 601,728) (Eng, 2000). PTEN is well known as a major tumor suppressor gene. The PTEN protein is a multifunctional phosphatase that regulates the AKT/mTOR pathway and plays an important role in the regulation of cell behavior, including growth, survival, and migration (Milella et al., 2015). PTEN mutations are detected in many malignant neoplasms. However, since there are few studies on the relationship between PTEN mutations and the phenotype of CS, the pathogenesis of this disease remains unclear.
Since Takahashi and Yamanaka reported the establishment of induced pluripotent stem cells (iPSCs) in mice (Takahashi and Yamanaka, 2006), iPSCs have been established from patients with many hereditary diseases and have been used to elucidate their pathogenesis and develop therapies. In the field of oncology, the use of iPSCs with the genetic backgrounds of patients with hereditary tumor syndromes may lead to the elucidation of not only the development of hereditary tumors but also the mechanism of somatic carcinogenesis. We have previously reported the induction of iPSCs from dental pulp and peripheral blood mononuclear cells (PBMCs) of patients with several genetic disorders under feeder-, serum-free culture conditions (Yamasaki et al., 2014;Hamada et al., 2020aHamada et al., , 2020b. In the present study, we found two novel PTEN exon 8 mutations in the same allele in a Japanese CS patient. In addition, we established disease-specific iPSCs from PBMCs of CS patients and investigated the expression of the PTEN gene and PTENδ splice variant in CS-PBMCs and CS-iPSCs.

Materials and methods
Ethics statement This study was approved by the Ethics Committee of Human Genome-Gene Analysis Research at Hiroshima University (approval number: hi-58 and hi-72).
All animal experiments in this study strictly followed the protocol approved by the Institutional Animal Care and Use Committee of Hiroshima University (approval number: A-11-140).

Phenotype of the affected individual
In 2014, a 60-yrold Japanese woman presented to our Department of Oral Maxillofacial Surgery at Hiroshima University Hospital with multiple oral polyps and pain in the buccal mucosa upon biting. She had first noticed oral polyps at the age of 25 yr. Eventually, these polyps extended to the whole mouth, and she sought care in otolaryngology. She had an extensive medical history: uterine fibroid at age 24, systemic scleroderma at age 37, thyroid cancer at age 41, a gastric lymphatic polyp at age 43, breast cancer at age 45, rheumatoid arthritis at age 53, and pericarditis pleurisy at age 55. She was treated for these diseases and received follow-up care in the hospital. However, she was never diagnosed with a syndrome, and there was no evident family history related to her condition.
The patient had a height of 158 cm, weight of 61.8 kg, and a body mass index of 24.7. Her head circumference was 58 cm (97th percentile), and she had macrocephaly. Many small mucosal papules appearing as "cobblestones" were present on the tongue, upper lower lip, palate, gingiva, and buccal mucosa. Each tumor was 2-8 mm in diameter, elastic, soft, and with a smooth surface. Endoscopic examination revealed a papillomatous lesion in the upper and lower gastrointestinal tract. Her face was dotted with trichilemmomas and hyperkeratotic papules (Supplementary data 1A-E). Histological examination of the excised buccal tumor tissue confirmed warty mucosal proliferation (Supplementary data 1F-G). Neither acral keratosis nor intellectual disability was observed. We clinically diagnosed her with CS based on the syndrome-testing criteria adapted by NCCN.

Genetic analysis
We obtained written informed consent from the patient for genetic testing. Her DNA was extracted from PBMCs using a QIAamp® DNA mini kit (Qiagen, Valencia, CA), according to the manufacturer's instructions. PBMC samples were prepared by centrifugation in a Histopaque-1077 (Sigma Aldrich, St. Louis, MO) density gradient.
Mutation analysis using next-generation sequencing (NGS) To detect pathogenic mutations, targeted sequencing was performed with the MiSeq System and the TruSight One Panel (Illumina, San Diego, CA), which is designed to comprehensively sequence more than 4800 disease-associated genes, according to the manufacturer. We analyzed these data using Illumina VariantStudio 3.0 and the Integrative Genomics Viewer (Broad Institute, Cambridge, MA).

Reverse transcription-PCR (RT-PCR)
Total RNA from PBMCs and the buccal mucosa tumor was prepared using TRIzol® Reagent (Thermo Fisher Scientific, Waltham, MA) and the Illustra RNAspin Mini RNA Isolation Kit (GE Healthcare, Chicago, IL), following the manufacturer's instructions. Isolated RNA was treated with amplification-grade DNase I (Invitrogen, Waltham, MA). RNA was reverse transcribed using the SuperScript VILO cDNA Synthesis Kit (Thermo Fisher Scientific) to synthesize cDNA. A PCR assay was performed using the OnePCR SuperMix (GenedireX Inc., New Taipei City, Taiwan). The products were electrophoresed on a 1.5% agarose gel, and the PCR bands were visualized on the ChemiDoc Touch Imaging System (Bio-Rad, Hercules, CA). Quantitative PCRs (RT-qPCRs) for each sample were carried out on an AriaMx realtime PCR system (Agilent Technologies, Santa Clara, CA), using Brilliant III Ultra-Fast SYBR Green QPCR Master Mix (Agilent Technologies). The expression level of each 1 3 gene was normalized to the amount of β-actin. The primers used in these experiments are shown in Table 1. The cycle program for product amplification was 1 cycle of 95 °C for 10 min (hot-start activation), followed by 40 cycles of 95 °C for 30 s (denaturation), 55 °C for 1 min (annealing), and 72 °C for 1 min (extension).

Establishment and characterization of CS-PBMC-derived CS-iPSC Establishment and maintenance of CS-iPSC under
feeder-and serum-free conditions. The establishment of CS-iPSCs from CS-PBMCs (Takahashi et al., 2007) was performed according to a method previously reported under feeder-and serum-free conditions (Hamada et al., 2020a). After culturing PBMCs in serum-free RD6F medium (Sato et al., 1987) supplemented with IL-2 (CELEUK, Takeda Pharm., Osaka, Japan) for 6 d, the cells were infected with Sendai virus vector at MOI = 6 and reseeded on Laminin-E8 in serum-free hESF6 medium supplemented with FGF2, heparin, and activin A (Yamasaki et al., 2014). After 14 d, the embryonic stem (ES) cell-like colonies emerged, colonies were picked, and the cells passaged. Characterization was performed as previously described (Yamasaki et al., 2014).
Alkaline phosphatase (ALP) staining On day 25 after Sendai virus infection, ALP staining was performed according to the manufacturer's protocol, using a Fast Red substrate kit (Nichirei Biosciences Inc., Tokyo, Japan). The induction efficiency was calculated as the ratio of the number of ALPpositive colonies to the number of seeded PBMCs.

Western blot analysis
The iPSCs were added to RIPA buffer containing protease phosphatase inhibitor, and the cells were disrupted with an ultrasonic homogenizer (TOMY SEIKO, Tokyo, Japan). After centrifugation at 13,000 × g for 15 min at 4℃, the supernatant was collected and the protein concentration was measured with a Pierce BCA Protein Assay kit (Thermo Fisher Scientific), according to the manufacturer's instructions. A total of 25 µg of PTEN, phosphorylated PTEN, AKT, and β-actin, and 75 µg of phosphorylated AKT protein were loaded onto a 10% SDS-PAGE gel for electrophoresis. After transferring to PDVF membrane, each antibody was used to detect the appropriate protein (Table 2).   Bio-Plex AKT signaling assay Twenty micrograms of total protein extracted from each iPSCs was reacted with antibody beads specific for phosphorylated forms of IRS-1, PTEN, AKT, GSK3α/β, BAD, mTOR, p70 S6 kinase, and S6 ribosomal protein and incubated at room temperature for 16 h. Biotinylated antibody was added to the beads for 30 min, followed by addition of streptavidin conjugated phycoerythrin (PE) for 10 min. Phosphoproteins of AKT-Signal Pathway molecules were analyzed by a BioPlex200 system (Bio-Rad).

Statistical Analysis
The data are presented as the mean ± standard error of the mean (SEM) from independent experiments. Statistical analysis included the Student's t test. A p value < 0.05 was considered statistically significant.

Results
Novel PTEN genetic mutations were detected by NGS analysis NGS analysis revealed a deletion of base position 1020 T and a mutation of base position 1026 G to A in the coding sequence of the PTEN gene (Fig. 1A). In addition, c.1026 + 32 T > G mutation was observed. All of these mutations were found on the same read, indicating that they were present in the same allele. The mutations were confirmed by Sanger sequencing (Fig. 1B). Moreover, germline mutations of the gene-associated mismatch repair system were detected (Table 3).

High induction efficiency of CS-iPSCs and their characterization After Sendai virus vector infection, ES cell-like colonies appeared in approximately
14 d and could be cultured. The induction efficiency of CS-iPSCs was significantly higher, averaging 0.24%, compared to 0.05% for normal human-derived iPSCs (WT-iPSCs) ( Fig. 2A, B). CS-iPSCs expressed various undifferentiated marker genes and showed the ability to differentiate into three germ layers in vitro and in vivo ( Fig. 2D-G). However, the expression levels of undifferentiated marker genes and the histological morphology of teratomas were not significantly different from those of WT-iPSCs (Fig. 2G, H, I). The iPSCs are free from mycoplasma contamination. Short tandem repeat (STR) analysis results showed that CS-iPSCs were identical to CS patient PBMC (Supplementary data 2A). Karyotype analysis revealed that CS-iPSCs were 46, XX (Supplementary data 2B).

PTEN expression decreased and AKT signalingrelated gene expression increased
Gene expression of PTEN was significantly decreased in CS-iPSCs, compared with WT-iPSCs (Fig. 3A). We also observed a reduction in PTEN gene expression in CS-PBMCs (Fig. 3B). The mRNA extracted from CS-PBMCs and CS-iPSCs was sequenced by the Sanger method using primers set from exon 7 to exon 9 of PTEN (Table 1), but the transcript from the mutant allele was not observed (Fig. 3C). Analysis of the phosphorylation levels of AKT signaling-related proteins by western blotting and Bio-Plex systems showed that phosphorylated PTEN was decreased, while the phosphorylation levels of AKT, mTOR, and GSK3α/β were increased (Fig. 3D, E).

Expression of PTENδ derived from the mutant allele increased in CS-iPSC
To analyze the expression of PTEN δ, the PTEN splicing variant, we designed primers for PTEN exon 7 and PTEN intron 8, sequenced them by the Sanger method, and analyzed their expression levels using RT-qPCR. Interestingly, the expression of PTEN δ was significantly upregulated in CS-iPSCs (Fig. 4A), and there was no expression from the wild-type allele (Fig. 4B).

Discussion
In this case study, the patient had numerous polyps in the oral mucosa, trichilemmomas on the face, and macrocephaly. She also had a medical history of breast cancer and thyroid cancer. Although acral keratosis, autism spectrum disorder, and developmental delay were not found, her symptoms met the typical clinical criteria of CS (Milella et al., 2015). These symptoms led us to search for mutations of PTEN. First, we examined the whole exon region of PTEN by the single-strand conformation polymorphism method. However, we found no results suggesting mutation. NGS analysis revealed two germline mutations (c.1020delT and c.1026G > A) at the end of PTEN exon 8. Since these mutations were located at the end of exon 8 and did not cause remarkable changes in the higher-order structure of the DNA, they could not be detected by the Polymerase Chain Reaction-Single-Strand Conformation Polymorphism method. Evaluation of the splicing sites showed that splicing was likely affected by these mutations (Table 4).
Recently, Chen et al.. reported that some variants of PTEN result in pathogenic exon skipping, alternative splicing, or use of cryptic splice sites in CS (Chen et al., 2017). It is known that mRNAs containing incorrect translation termination codons, caused by errors during transcription and replication or splicing, are selectively degraded by the nonsense-mediated mRNA decay pathway (Behm-Ansmant et al., 2007). Reduction in PTEN gene expression and loss of the transcript from the mutant allele may indicate that this mutation causes intron retention and aberrant mRNA degradation by the NMD pathway. Moreover, analysis of PTEN gene expression in buccal mucosal tumors of patients showed that it did not cause obvious homozygous loss. These findings support that the pathogenesis was caused by  haploinsufficiency as has been reported (Di Cristofano et al., 1999Cristofano et al., , 2001Salmena et al., 2008). In addition, an intron mutation (c.1026 + 32 T > G) has been suggested to affect the splicing silencer site in the exon, but this mutation has a high allele frequency, and according to ClinGen PTEN Expert Panel Specifications, it has a benign stand-alone effect (Mester et al., 2018). The advantages of using iPSCs to study diseases of hereditary tumor syndromes include solving the problem of an insufficient supply of samples from specimens, as well as producing cells that cannot be established as cell lines by differentiating into specific cells. Furthermore, it may be useful for tracing the mechanism of tumorigenesis. The CS-iPSCs established in this study showed reduced PTEN gene expression and protein expression, compared to WT-iPSCs. The CS-iPSCs also showed high induction efficiency. However, the effects on the AKT pathway due to PTEN reduction seemed to be corrected downstream, and we did not observe the phenotype such as hamartoma or neoplastic growth that this patient developed in clinical practice. It may be that haploinsufficiency of PTEN alone is not sufficient for the development of CS, and additional factors are required.
There are several variants of PTEN, including a variant containing intron 8 (PTEN δ) (Sharrard and Maitland, 2000). The function of PTEN δ is reported to be similar to that of PTEN (Breuksch et al., 2018), but the details are unknown. We analyzed the expression level and nucleotide sequence of PTEN δ by RT-qPCR and Sanger sequencing using the primers designed for exon 7 and intron 8. Gene expression of PTEN δ was low in WT-PBMCs and CS-PBMCs, making it difficult to compare the two. However, compared to WT-iPSCs, the expression of PTEN δ in CS-iPSCs was predominantly high, and the expression from the wild-type allele was suppressed. This suggests that abnormal translation from intron 8 causes structural changes in the C-terminus of PTEN δ (Fig. 4C) and aberrant PTEN δ may be involved in the pathogenesis of this disease.
Using NGS, we also discovered other germline genetic variants associated with tumor development. MLH1 is a component of intracellular DNA mismatch repair (MMR) and is also known to be involved in genomic instability and Lynch syndrome. V384D of MLH1 is a polymorphism found in approximately 2.5% of East Asians, and this variant is known to have reduced MMR activity. The V384D variant of MLH1 is frequently found in breast cancer tissues, suggesting that haploinsufficiency of MLH1 may modulate malignant tumor progression (Lee et al., 2019). The MSH6 gene is also involved in genomic stability, and it has been suggested that G39E may be associated with breast cancer (Lee et al., 2014), but no conclusion has been reached yet. Furthermore, BRCA2 is known as a causative gene of hereditary breast and ovarian cancers, but mutant M784V has been reported to be neutral (Spearman et al., 2008). The relationship between these mutations and PTEN has not been fully elucidated, but mutations in these genes may support the patient's medical history.

Conclusion
Although there are many reported cases of Cowden syndrome, a few studies have focused on its pathogenesis. In this study, we report a novel mutation of PTEN and the establishment of the first CS-iPSCs, and analyzed the expression of the PTEN gene. In recent years, with the progress of cancer genome medicine, the identification of hereditary tumor-associated genes such as PTEN has been progressing. However, further studies on pathogenic mutations and their pathogenesis are required to establish a comprehensive therapeutic system for hereditary tumors. Using CS disease-specific iPSCs, as shown in this study, will provide further new insights for disease research and therapeutic drug discovery.
1 3 the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.