The spectrum of germline VHL variants in Iranian patients with Retinal Capillary Hemangioblastoma and genotype-phenotype analysis

Abstract

Backgrounds: Retinal capillary hemangioblastoma (RCH) although may also be observed sporadically, is the most frequent and earliest manifestation of the Von Hippel-Lindau disease (VHL). This is the first report of different types of variants spectrum and genotype-phenotype correlations in the Iranian families with RCH.

Methods: In this Prospective Observational Case Series study, 17 families with RCH were included. PCR was performed to amplify 3 exons of the VHL gene. Afterwards, Sanger sequencing was performed on all the PCR products. For the detection of VHL copy number variations, MLPA was used.

Results: Our study identified 10 different types of predominantly pathogenic VHL variants. Missense mutations were the most common variants located on the α domain of the VHL protein (pVHL) in exon 3. Also, the pathogenic variants in RCH patients with CNS hemangioblastoma (CHB) mostly (72.7%) were located on the α domain. In a family, Proband’s sister had a different mutation with more severe clinical manifestations and a poor prognosis.

Conclusion: α domain of VHL may play a potential role in the pathogenesis of RCH.  Our findings suggest that the genotype-phenotype characteristics in those variants in α- domain may predispose patients to RCH with CHB. 

Background

Retinal capillary hemangioblastomas (RCH) is a benign tumor that represents the initial manifestation in roughly half of Von Hipple Lindau (VHL) patients [1]. They may also occur sporadically without systemic involvement [2, 3]. RCH increases with age and reaches a 95% probability in VHL patients over 80 years [4]. The incidence of RCH has been estimated at approximately 1 in 73080 live births [5].

VHL disease is an autosomal- dominant tumor syndrome and highly penetrant condition (exceed 90% by 60 years of age) characterized by malignant and benign tumors [2, 6, 7] including RCH, CNS, clear cell renal carcinoma (RCC), endolymphatic sac tumor (ELST), pheochromocytomas (PHEO), and as well as pancreatic (PC) and kidney cysts (KC) [8]. Eighty percent of all cases are familial and 20% due to de novo mutations [9]. VHL incidence has been estimated at approximately 1 in 36,000 live births. VHL patients have been categorized into types 1 and 2. Type 1 (low risk) includes CHB, RCH, PC, RCC and neuroendocrine tumors, whereas type 2 (high risk) includes PHEO, CHB and RCH [4].

The disease is characterized by over 900 germline mutations in the VHL tumor suppressor gene, consistent with the Knudson "two-hit" hypothesis, located on chromosome 3p25-26 [10, 11]. This gene contains 3 exons and encodes 2 main VHL protein (pVHL) products: pVHL30 and pVHL19. pVHL30 degrades hypoxia-inducible factors (HIF), which are formed in response to hypoxia [12]. Loss of VHL function results in aberrant upregulation of HIF, which could induce high levels of vascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF) expressions. Subsequently, overproduction of VEGF and PDGF stimulates angiogenesis, growth and cellular proliferation, then contributing to promoting the growth of RCH and other tumors associated with VHL disease [13–16]. The pVHL has an α domain which directly binds elongin C and β domain binds to the HIF-α [15, 16]. Roughly 60–70% of VHL mutations cause truncation of pVHL by NSS mutations (nonsense, small deletions/insertions and splice mutations) and large/complete deletions [7].

There is no data on VHL variants and genotype-phenotype correlation in Iranian families with RCH. Therefore, in the present study, we tried to investigate the VHL variants spectrum and evaluated their genotype - phenotypes correlations.

Methods

In this Prospective Observational Case Series study, 14 families with RCH were included in the VHL genetic testing. All tests were conducted in the Medical Genetics department– Rasoul Akram Hospital in Tehran, Iran. The diagnosis of RCH was based on clinical findings. Informed consent was provided for each patient who participated in this study.

Three ml of blood samples were collected from all the patients in EDTA anticoagulant tubes. The DNA was extracted by using a modified salting-out method. The quality and quantity of extracted DNA were measured spectrophotometrically at wavelength 260nm and 280nm. Isolated DNA was stored at − 20°C until use. All coding exons and the corresponding intron/exon boundaries of the VHL gene were amplified using gene-specific oligonucleotide primers. The primer sequences for exon1, exon 2 and exon 3 and Size product and Tm are shown in Table 1.

PCR was conducted in a 25 µL mixture comprising 200 ng of genomic DNA, 10 pm of each primer, 12.5 µl of 2× Master Mix (Fermentas, USA) containing Taq DNA polymerase, MgCl2, dNTPs, and reaction buffers.

Sequencing was performed using the Sanger method on the ABI 3730 sequencer (Bioneer, South Korea). The resulting sequences were analyzed using Mutation Survey, Varsome software and CADD score. We used the MLPA-VHL kit (P016-C2, MRC-Holland, Amsterdam, and The Netherlands) to look for VHL copy number variations, according to the manufacturer's protocol.

Results

In this study, 10 different mostly pathogenic VHL germline mutations were identified in 17 families (Table 2). Analysis of sequences of 17 families represented missense mutations were the most common (20/35, 57.1%), followed by large/complete deletions (11/35, 31.4%), splice site variants (3/35, 8.6%) and frameshift (1/35, 2.8%) (Table 2). Amino acid substations were located in 2 domains, α and β, while there was no mutation in the codons from 1 to 69 (Fig. 2). Individuals carrying amino acid substitutions in α and β -domain were 17 cases and 7 cases, respectively. Six deletions, especially occurring at α- domain and 4 deletions concentrated in β- domain and 1 deletion were located in 2 domains.

The frequency of missense mutations was in exon 1 (17.1%) and exon 3 (40%), the large/complete deletion was in exon 1(2.8%), 2 (11.4%) and 3 (20%) (Table 3). The 2 splice site variants of missense changes were placed in the last base of exon 2 (5.7%) and another was placed in the canonical splice donor for exon 2 (2.8%), and there was 2.8% frameshift in exon 2. Eight males and 12 females had missense mutations, 6 males and 5 females had large/complete deletions, 2 males and 1 female had splice site variants, and 1 male had frameshift mutations.

Table 3. Distribution of genetic mutations in the VHL gene in RCH patients

Overall, the most common variants involved in exon 3 locate α- domain of pVHL. Interestingly, the most frequently observed mutations in our patients were c.208G > A (p.E70K) which occurred in 4 families (23.5%). However, mutations c.563T > G (L188R) and c.467A > G (Y156C) were presented with the highest incidence in 8th and 10th family.

Based on ACMG classification, 23 patients had pathogenic criteria and 1 patient had likely pathogenic criteria.

Genotype-phenotype correlation

Six individuals carrying the c.208G > A (p.E70K) mutation showed RCH or CHB; 5 patients had only RCH and the other patient manifested RCH and CHB. Eighth family carrying the c.563T > G (p. L188R) missense mutation manifested RCH, CHB or KC; 2 patient had only RCH and the other patient manifested RCH, CHB and KC. Tenth family carrying the c.467A > G (Y156C) missense mutation manifested only RCH. Six patients carrying the large/complete deletion manifested RCH and CHB, 4 patients had only RCH and 1 patient manifested RCH and PCC (Table 1).

The pathogenic missense mutations c.208G > A (p.E70K) were placed within the HIF-α binding site (residues 65–117) and 2 splice site variants were in Chaperonin-containing tailless complex polypeptide 1 (CCT) (residues 100– 155), where VHL needs to be coupled correctly with elongin BC (Fig. 2) [17, 18]. PCC was found in 2 subjects in the family number 11; Proband was carrying a heterozygous deletion in the exon 1. Her sister has a different mutation (a missense mutation c.499C > T (p.R167W) in the exon 3) with more severe clinical manifestations and a poor prognosis. The pedigree of the investigated this family was shown in Fig. 1.

With a comparison between these two groups RCH + VHL + and RCH + VHL-, the following results were obtained (Table 3):

Statistically significant different was not observed between the age distribution of RCH + Variant + and RCH + variant – that were 32.7 ± 6.42 and 34.5 ± 11.90, respectively (P = 0.501). Also, there was not a statistically significant difference in the sex distribution between RCH + Variant - and RCH + variant - (P = 0.603) (Table 4).

Table 4. Distribution of age and sex in RCH patients.

Twenty-four RCH patients carried amino acid substitutions and deletions located in α domain and 12 RCH patients carried amino acid substitutions and deletions located in β –domain, 68.6% vs. 34.3%, respectively.

In addition, it was shown that 11 CHB in RCH patients carried amino acid substitutions and deletions located on α domain, and 4 CHB in RCH patients carried amino acid substitutions and deletions located in β –domain, 72.7% vs. 36.3%, respectively. There was not a statistically significant difference between RCH patients with and without CHB (P = 0.473).

Discussion

RCH is a benign tumor that may occur solitary or as a VHL syndrome manifestation [1, 2]. This is the first study describing different types of variants and genotype-phenotype analysis in the Iranian VHL patients with RCH.

Our analyses indicated that missense mutations in the VHL gene were the most common type which is confirmed by Nordstrom-O’Brien et al, Wong et al and Jee-Soo Lee et al. [8, 19, 20].

RCH patients with CHB was the most presenting phenotype (n = 14), followed by PCC and KC (n = 2), PC (n = 1). Pradeep Mettu et al. demonstrated that mutation in the α domain have a higher risk of RCH than β domain [21]. We also observed that in patients with RCH (regardless of VHL syndrome) and CHB manifestation, there is a higher chance of mutation in the α domain than the β domain. However, Sheng-Jie Liu and et al. described that missense mutation in HIF -α binding site increases the risk of CHB manifestation [22].

Several studies represented that the causative factor of hemangioblastoma is dysregulated HIF-α [2, 8] HIF-α binding site mutation can disrupt the binding of pVHL to HIF-α transcription factor and subsequently its degradation. Therefore, it can promote angiogenesis factors that have been proposed to account for the notable vascularity often seen in the tumors associated with the VHL syndrome. The HIF-α binding site mutation in our data was c.208G > A (E70K) which was previously reported to reduce HIF-degradation [23–25].

The CCT complex is correlated exactly with exon 2 and is often found mutated or deleted in the VHL disease or solitary RCC [17]. The CCT complex binding site mutations in our data include V155M, V155L splice variants and D126Ffs*31 frameshift mutation. These variants in CCT complex binding site manifested pathogenic classification.

Interestingly, the high incidence of c.208G > A (E70K) mutation represented in several studied families and its limited scope of manifestations (either CHB or RCH) was supported by the findings of Jee-Soo Lee et al and Hwang et al..[20, 26] However, mutations c.563T > G (L188R) and c.467A > G (Y156C) were presented with the highest incidence each in a family.

Our results showed that about 23.5% of families with VHL gene mutation had no familial history, which are probably due to inheritance from a mosaic parent or can be sporadic germline VHL mutations [8]. It seems that cases with solitary RCH should be considered for VHL gene evaluation in other family members. Sgambati et al. [27] observed that in asymptomatic parents the rate of mosaicism was 4.8% of de novo VHL patients. Decker et al. showed that may occur up to 20% of de novo mutations. Recently, Wu P et al. [28] in China reported a more prevalence of de novo VHL germline mutations (56.3%). Our data showed that all of the splice site variants were in RCH patients with CHB. Notably, we observed that VHL-related tumors in patients with no familial history were lower compared with patients with a family history of VHL (Table 2).

In one family of our samples (family #11), we had two different mutations in two members, that both of them categorized as VHL type 2A. Proband had a deletion with a history of RCH, CHB and the other patient had a point mutation with symptoms of RCH, CHB, and PCC (Fig. 1, B). In 2019 Hongwei Shen et al. reported one family with hereditary spherocytosis with two different point mutations in siblings that one of them was de novo mutation and another one transmitted from parents [29]. The parents of our proband weren’t accessible for evaluation.

Conclusion

The present study evaluated Iranian VHL patients with RCH disease. We describe a spectrum of VHL variants and genotype-phenotype correlations according to alterations in pVHL. Our results suggest insight into the genotype-phenotype correlation in those amino acid substitutions in α- domain may predispose patients to RCH and CHB.

List Of Abbreviations

Retinal capillary hemangioblastoma (RCH)

Von Hippel-Lindau disease (VHL)

VHL protein (pVHL)

CNS hemangioblastoma (CHB)

Cclear cell renal carcinoma (RCC)

Endolymphatic sac tumor (ELST)

Pheochromocytomas (PHEO)

Pancreatic (PC)

Kidney cysts (KC)

Hypoxia-inducible factors (HIF)

Vascular endothelial growth factor (VEGF)

Platelet-derived growth factor (PDGF)

Chaperonin-containing tailless complex polypeptide 1 (CCT)

Declarations

Funding

This work supported by grant form research deputy of eye research center, Tehran, Iran.

Conflict of interest

The authors declare that they have no conflict of interest.

Code availability

Ethics code: 32903

Ethics approval

The study protocol was approved by the institutional ethics committee of Iran University of Medical Sciences (Iran) (Ethics code: 32903).

Consent to participate

Informed consent was obtained from the parents or guardians of all patients who participated in the study.

Consent for publication

All authors and patients

Availability of data and materials

All datasets generated for this study are included in the article.

Acknowledgments

This work supported by grant form research deputy of eye research center, Tehran, Iran.

Author’s contributions

M.N. designed the study and finalized the manuscript; F.A. wrote the main manuscript text, performed genetic analysis, genetic tests and interpretation of data ; S.T. designed the study and assisted in the genetic analyses; A.S. designed the study; R.M. commented on and revised the manuscript; R.K. collected samples; Z.A. performed genetic tests; G.KH. Supervised the study and commented on the manuscript. All authors read and approved the final version.

References

1. Kreusel K-M, Bechrakis NE, Krause L, Neumann HP, Foerster MH. Retinal angiomatosis in von Hippel–Lindau disease: a longitudinal ophthalmologic study. Ophthalmology. 2006;113(8):1418–24.
2. Gossage L, Eisen T, Maher ER. VHL, the story of a tumour suppressor gene. Nature Reviews Cancer. 2015;15(1):55–64.
3. Decker J, Neuhaus C, Macdonald F, Brauch H, Maher ER. Clinical utility gene card for: von Hippel–Lindau (VHL). European Journal of Human Genetics. 2014;22(4):572-.
4. Ruppert MD, Gavin M, Mitchell KT, Peiris AN. Ocular Manifestations of von Hippel-Lindau Disease. Cureus. 2019;11(8).
5. Mishra P, Pal S, Kanaujia V, Sharma K. Retinal capillary hemangioblastoma associated with retinochoroidal coloboma in Von Hippel-Lindau disease. Indian journal of ophthalmology. 2019;67(5):688–90.
6. Shuin T, Yao M, Shinohara N, Yamasaki I, Tamura K. Clinical status of Von Hippel-Lindau disease associated pheochromocytoma in Japan: a national epidemiologic survey. Nihon Hinyokika Gakkai zasshi The japanese journal of urology. 2012;103(3):557–61.
7. Maher ER, Neumann HP, Richard S. von Hippel–Lindau disease: a clinical and scientific review. European Journal of Human Genetics. 2011;19(6):617–23.
8. Nordstrom-O'Brien M, van der Luijt RB, van Rooijen E, van den Ouweland AM, Majoor‐Krakauer DF, Lolkema MP, et al. Genetic analysis of von Hippel‐Lindau disease. Human mutation. 2010;31(5):521–37.
9. Ordookhanian C, Kaloostian PE, Ghostine SS, Spiess PE, Etame AB. Management strategies and outcomes for VHL-related craniospinal hemangioblastomas. Journal of Kidney Cancer and VHL. 2017;4(3):37.
10. Wittebol-Post D, Hes F, Lips C. The eye in von Hippel-Lindau disease. Long-term follow-up of screening and treatment: recommendations. Journal of internal medicine. 1998;243(6):555–61.
11. Nielsen SM, Rhodes L, Blanco IG, Chung WK, Eng C, Maher E, et al., editors. Von Hippel-Lindau disease: genetics and role of genetic counseling in a multiple neoplasia syndrome2016: American Society of Clinical Oncology.
12. Stebbins CE, Kaelin WG, Pavletich NP. Structure of the VHL-ElonginC-ElonginB complex: implications for VHL tumor suppressor function. Science. 1999;284(5413):455–61.
13. Iliopoulos O, Levy AP, Jiang C, Kaelin WG, Goldberg MA. Negative regulation of hypoxia-inducible genes by the von Hippel-Lindau protein. Proceedings of the National Academy of Sciences. 1996;93(20):10595-9.
14. Kim WY, Kaelin WG. Role of VHL gene mutation in human cancer. Journal of clinical oncology. 2004;22(24):4991–5004.
15. Lonser RR, Glenn GM, Walther M, Chew EY, Libutti SK, Linehan WM, et al. von Hippel-Lindau disease. The Lancet. 2003;361(9374):2059–67.
16. Tarade D, Ohh M. The HIF and other quandaries in VHL disease. Oncogene. 2018;37(2):139–47.
17. Feldman DE, Thulasiraman V, Ferreyra RG, Frydman J. Formation of the VHL–elongin BC tumor suppressor complex is mediated by the chaperonin TRiC. Molecular cell. 1999;4(6):1051–61.
18. Melville MW, McClellan AJ, Meyer AS, Darveau A, Frydman J. The Hsp70 and TRiC/CCT chaperone systems cooperate in vivo to assemble the von Hippel-Lindau tumor suppressor complex. Molecular and cellular biology. 2003;23(9):3141–51.
19. Wong WT, Agrón E, Coleman HR, Reed GF, Csaky K, Peterson J, et al. Genotype-phenotype correlation in von Hippel-Lindau disease with retinal angiomatosis. Archives of ophthalmology. 2007;125(2):239–45.
20. Lee J-S, Lee J-H, Lee KE, Kim JH, Hong JM, Ra EK, et al. Genotype-phenotype analysis of von Hippel-Lindau syndrome in Korean families: HIF-α binding site missense mutations elevate age-specific risk for CNS hemangioblastoma. BMC medical genetics. 2016;17(1):1–8.
21. Mettu P, Agrón E, Samtani S, Chew EY, Wong WT. Genotype-phenotype correlation in ocular von Hippel-Lindau (VHL) disease: the effect of missense mutation position on ocular VHL phenotype. Investigative ophthalmology & visual science. 2010;51(9):4464–70.
22. Liu S-J, Wang J-Y, Peng S-H, Li T, Ning X-H, Hong B-A, et al. Genotype and phenotype correlation in von Hippel–Lindau disease based on alteration of the HIF-α binding site in VHL protein. Genetics in Medicine. 2018;20(10):1266–73.
23. Clifford SC, Cockman ME, Smallwood AC, Mole DR, Woodward ER, Maxwell PH, et al. Contrasting effects on HIF-1α regulation by disease-causing pVHL mutations correlate with patterns of tumourigenesis in von Hippel-Lindau disease. Human molecular genetics. 2001;10(10):1029–38.
24. Miller F, Kentsis A, Osman R, Pan Z-Q. Inactivation of VHL by Tumorigenic Mutations That Disrupt Dynamic Coupling of the pVHL· Hypoxia-inducible Transcription Factor-1α Complex. Journal of Biological Chemistry. 2005;280(9):7985–96.
25. Cingoz S, van der Luijt R, Kurt E, Apaydin M, Akkol I, Ozgen MH. A novel missense mutation (N78D) in a family with von Hippel–Lindau disease with central nervous system haemangioblastomas, pancreatic and renal cysts. Familial cancer. 2013;12(1):111–7.
26. Hwang S, Ku CR, Lee JI, Hur KY, Lee M-S, Lee C-H, et al. Germline mutation of Glu70Lys is highly frequent in Korean patients with von Hippel–Lindau (VHL) disease. Journal of human genetics. 2014;59(9):488–93.
27. Sgambati M, Stolle C, Choyke P, Walther M, Zbar B, Linehan W, et al. Mosaicism in von Hippel–Lindau disease: lessons from kindreds with germline mutations identified in offspring with mosaic parents. The American Journal of Human Genetics. 2000;66(1):84–91.
28. Wu P, Zhang N, Wang X, Ning X, Li T, Bu D, et al. Family history of von Hippel–Lindau disease was uncommon in Chinese patients: suggesting the higher frequency of de novo mutations in VHL gene in these patients. Journal of human genetics. 2012;57(4):238–43.
29. Shen H, Huang H, Luo K, Yi Y, Shi X. Two different pathogenic gene mutations coexisted in the same hereditary spherocytosis family manifested with heterogeneous phenotypes. BMC medical genetics. 2019;20(1):1–7.