Rob Fowkes

Studying Cat (Felis catus) Diabetes: Beware of the Acromegalic Imposter

Naturally occurring diabetes mellitus (DM) is common in domestic cats (Felis catus). It has been proposed as a model for human Type 2 DM given many shared features. Small case studies demonstrate feline DM also occurs as a result of insulin resistance due to a somatotrophinoma. The current study estimates the prevalence of hypersomatotropism or acromegaly in the largest cohort of diabetic cats to date, evaluates clinical presentation and ease of recognition. Diabetic cats were screened for hypersomatotropism using serum total insulin-like growth factor-1 (IGF-1; radioimmunoassay), followed by further evaluation of a subset of cases with suggestive IGF-1 (>1000 ng/ml) through pituitary imaging and/ or histopathology. Clinicians indicated pre-test suspicion for hypersomatotropism. In total 1221 diabetic cats were screened; 319 (26.1%) demonstrated a serum IGF-1>1000 ng/ml (95% confidence interval: 23.6–28.6%). Of these cats a subset of 63 (20%) underwent pituitary imaging and 56/63 (89%) had a pituitary tumour on computed tomography; an additional three on magnetic resonance imaging and one on necropsy. These data suggest a positive predictive value of serum IGF-1 for hypersomatotropism of 95% (95% confidence interval: 90–100%), thus suggesting the overall hypersomatotropism prevalence among UK diabetic cats to be 24.8% (95% confidence interval: 21.2–28.6%). Only 24% of clinicians indicated a strong pre-test suspicion; most hypersomatotropism cats did not display typical phenotypical acromegaly signs. The current data suggest hypersomatotropism screening should be considered when studying diabetic cats and opportunities exist for comparative acromegaly research, especially in light of the many detected communalities with the human disease.

Glial Cells Missing 1 regulates equine Chorionic Gonadotrophin beta subunit via binding to the proximal promoter

Equine chorionic gonadotrophin (eCG) is a placental glycoprotein critical for early equine pregnancy and used therapeutically in a number of species to support reproductive activity. The factors in trophoblast that transcriptionally regulate eCGβ-subunit (LHB), the gene which confers the hormones specificity for the receptor, are not known. The aim of this study was to determine if glial cells missing 1 regulates LHB promoter activity. Here, studies of the LHB proximal promoter identified four binding sites for glial cells missing 1 (GCM1) and western blot analysis confirmed GCM1 was expressed in equine chorionic girdle (ChG) and surrounding tissues. Luciferase assays demonstrated endogenous activity of the LHB promoter in BeWo choriocarcinoma cells with greatest activity by a proximal 335u2009bp promoter fragment. Transactivation studies in COS7 cells using an equine GCM1 expression vector showed GCM1 could transactivate the proximal 335u2009bp LHB promoter. Chromatin immunoprecipitation using primary ChG trophoblast cells showed GCM1 to preferentially bind to the most proximal GCM1-binding site over site 2. Mutation of site 1 but not site 2 resulted in a loss of endogenous promoter activity in BeWo cells and failure of GCM1 to transactivate the promoter in COS-7 cells. Together, these data show that GCM1 binds to site 1 in the LHB promoter but also requires the upstream segment of the LHB promoter between −119u2009bp and −335u2009bp of the translation start codon for activity. GCM1 binding partners, ETV1, ETV7, HOXA13, and PITX1, were found to be differentially expressed in the ChG between days 27 and 34 and are excellent candidates for this role. In conclusion, GCM1 was demonstrated to drive the LHB promoter, through direct binding to a predicted GCM1-binding site, with requirement for another factor(s) to bind the proximal promoter to exert this function. Based on these findings, we hypothesize that ETV7 and HOXA13 act in concert with GCM1 to initiate LHB transcription between days 30 and 31, with ETV1 partnering with GCM1 to maintain transcription.

Feline hypersomatotropism and acromegaly tumorigenesis: a potential role for the AIP gene

Acromegaly in humans is usually sporadic, however up to 20% of familial isolated pituitary adenomas are caused by germline sequence variants of the aryl-hydrocarbon-receptor interacting protein (AIP) gene. Feline acromegaly has similarities to human acromegalic families with AIP mutations. The aim of this study was to sequence the feline AIP gene, identify sequence variants and compare the AIP gene sequence between feline acromegalic and control cats, and in acromegalic siblings. The feline AIP gene was amplified through PCR using whole blood genomic DNA from 10 acromegalic and 10 control cats, and 3 sibling pairs affected by acromegaly. PCR products were sequenced and compared with the published predicted feline AIP gene. A single nonsynonymous SNP was identified in exon 1 (AIP:c.9T > G) of two acromegalic cats and none of the control cats, as well as both members of one sibling pair. The region of this SNP is considered essential for the interaction of the AIP protein with its receptor. This sequence variant has not previously been reported in humans. Two additional synonymous sequence variants were identified (AIP:c.481C > T and AIP:c.826C > T). This is the first molecular study to investigate a potential genetic cause of feline acromegaly and identified a nonsynonymous AIP single nucleotide polymorphism in 20% of the acromegalic cat population evaluated, as well as in one of the sibling pairs evaluated.