Genetic and endocrine factors influencing juvenile-onset granulosa cell tumourigenesis

Smith, Kerri Nicole (2020) Genetic and endocrine factors influencing juvenile-onset granulosa cell tumourigenesis. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (49MB)

Abstract

Granulosa cell (GC) tumours are a rare ovarian neoplasm divided into adult and juvenile clinicopathologic subtypes. Whereas the adult-onset phenotype is attributed to an acquired mutation in the GC differentiation marker forkhead box L2 (FOXL2), the etiology of juvenile GC tumours is unknown. SWR/Bm (SWR) inbred strain female mice are spontaneously susceptible to the pubertal initiation of juvenile-onset GC tumours of the ovary. This heritable trait shares histological, endocrinological and malignant features with those GC tumours that appear in infants and young girls, and so the SWR strain is recognized as a model system for the identification of susceptibility genes and endocrine influences. Phenotypic mapping studies have consistently associated granulosa cell tumour susceptibility (Gct) loci with the GC tumour phenotype. The Gct1 locus on distal mouse chromosome (Chr) 4 is the driver in the SWR model, whereas the Gct4 locus on Chr X modifies tumour susceptibility in the presence of homozygous SWR-derived Gct1 (Gct1SW) alleles. The identity of the dehydroepiandrosterone (DHEA)- responsive Gct1 gene remains unresolved, although substantial evidence suggests the androgen receptor (Ar) gene shares identity with Gct4. The influence of these genes on tumour initiation is confined to the ovarian GCs over a short window of susceptibility, which coincides with pubertal endocrine stimulation and the activation of the first follicle wave from the ovarian medulla. Based on the genetic and endocrine evidence and using a combination of the SWR inbred strain, congenic sublines, and engineered constructs transferred to the SWR background, we sought to: 1) elucidate Gct1 identity using fine mapping, comprehensive genetic analyses, and in vivo candidate investigations; 2) delete the AR protein from the GC population to test its requirement for androgen-induced tumour initiation; and 3) lineage trace the GC tumour cell of origin using an inducible fluorescent reporter system, with GC-specificity conferred by an engineered construct that displaced a single Foxl2 allele. The Gct1 locus was resolved to 1.65 Mb and a candidate list of 18 unique annotated protein coding genes, non-coding RNA genes, and other processed transcripts with the addition of two nucleotide polymorphisms to a genetic maker panel for 12 previously phenotyped congenic mouse sublines. Complete genetic annotation of the SWR Gct1 interval by nextgeneration sequencing and copy number analysis identified a splice site variant (c.460-3C>T) in the dehydrogenase/reductase (SDR family) member 3 (Dhrs3) gene predicted to be highly pathogenic, whereas functional analyses did not reveal any candidates capable of initiating GC tumours in SWR females or GC tumour-resistant mice. No tumours arose in ovaries that were AR protein-negative in the GC population despite the retention of homozygous Gct1SW alleles, suggesting that the AR is required for GC tumour initiation. SWR females carrying the engineered Foxl2 allele were resistant to the androgen-induced GC tumour phenotype. Haploinsufficiency for Foxl2 therefore precluded the lineage tracing experiment, but suggests that the pathway for ovarian specification involving Foxl2 influences, or is influenced by, the activity of the Gct1 locus. Together, these genetic investigations support a two-stage tumour susceptibility model for Gct allele activity, in which establishment of GC tumour susceptibility requires the Gct1 driver locus and Foxl2 transcription factor in the developing ovary, and GC tumour initiation in a genetically susceptible juvenile ovary requires androgenic signalling through the AR. Although the complexities of GC tumour susceptibility in the SWR mouse model have yet to be fully unraveled, the validity of the genetic contribution is clear and potentially relevant for pediatric cases of juvenile-onset GC tumours.

Actions (login required)

View Item