The Methylation Analysis of the Glucose-Dependent Insulinotropic Polypeptide Receptor (GIPR) Locus in GH-Secreting Pituitary Adenomas

Abstract

The glucose-dependent insulinotropic polypeptide receptor (GIPR) is aberrantly expressed in about one-third of GH-secreting pituitary adenomas (GH-PAs) and has been associated with a paradoxical increase of GH after a glucose load. The reason for such an overexpression has not yet been clarified. In this work, we aimed to evaluate whether locus-specific changes in DNA methylation patterns could contribute to this phenomenon. By cloning bisulfite-sequencing PCR, we compared the methylation pattern of the GIPR locus in GIPR-positive (GIPR⁺) and GIPR-negative (GIPR^-) GH-PAs. Then, to assess the correlation between Gipr expression and locus methylation, we induced global DNA methylation changes by treating the lactosomatotroph GH3 cells with 5-aza-2'-deoxycytidine. Differences in methylation levels were observed between GIPR⁺ and GIPR^- GH-PAs, both within the promoter (31.9% vs. 68.2%, p < 0.05) and at two gene body regions (GB_1 20.7% vs. 9.1%; GB_2 51.2% vs. 65.8%, p < 0.05). GH3 cells treated with 5-aza-2'-deoxycytidine showed a ~75% reduction in Gipr steady-state level, possibly associated with the observed decrease in CpGs methylation. These results indicate that epigenetic regulation affects GIPR expression in GH-PAs, even though this possibly represents only a part of a much more complex regulatory mechanism.

Keywords: 5-aza-2′-deoxycytidine; CpG methylation; GH-secreting pituitary adenomas; acromegaly; glucose-dependent insulinotropic polypeptide receptor (GIPR).

Figure 1

The relative expression of GIPR in GH-PAs was evaluated via ddPCR. (A) Steady-state level of GIPR (upper panel) and HMBS (lower panel) in GH-PAs (p1–p9) and a normal pituitary gland (NPit). One-dimensional-dot plot display of mono-color droplet fluorescence intensity. Colored and black dots indicate each droplet positive (blue for GIPR, green for HMBS) or negative for PCR assay, respectively. For each sample, droplets are depicted according to the event (number of droplets read during reading; x-axis) and its FAM (Blue) or VIC (green) fluorescence intensity (Ch1/Ch2 Amplitude; y-axis). The setting of the threshold for droplet positivity was performed manually. (B) The normalized ratio of GIPR to HMBS in all samples. The error bars associated with each point represent Poisson’s 95% confidence intervals. #, gsp⁺.

Figure 2

CpG sites in the GIPR locus of the HumMeth450 BeadChip were investigated by cloning-based BSP. (A) GIPR locus as annotated in the UCSC Genome Browser (GRCh37/hg 19 Assembly, genomic region chr19:46,169,919–46,187,927). Red CpG sites were investigated during this study. (B) The promoter and the gene body are reported as pink and pale blue shading boxes, respectively. The numbers below the graph refer to the percentage of methylation observed in each CpG analyzed in GIPR⁺ and GIPR⁻ somatotropinomas. (C) The percent methylation of the various CpG groups was compared between GIPR⁺ and GIPR⁻ GH-PAs. Values and error bars represent the mean of the percent methylation of CpGs within each cluster and the standard deviation of the data, respectively; significance: * p < 0.05.

Figure 3

Effects of 5-Aza-dC treatment in GH3 cells. The effect of 5-Aza-dC on cell viability is reported in panel (A) as dose–response curves, 48 h after treatment. Cells were treated with 5-Aza-dC (1 μM and 5 μM) and the expression of Gipr (B) and Prl (C) were evaluated. Values and error bars represent the mean of three independent experiments in at least three replicates and the standard deviation of the data, respectively. ** p < 0.001 in treated vs. non-treated (NT) cells.

Figure 4

Direct BSP analysis of CpG sites in GH3 cells. (A) Schematic representation of the Gipr gene (UCSC Genome Browser RGSC 6.0/rn6, genomic region chr1:80,063,066–80,074,019), comprising the promoter (pink box), gene body (pale blue shading box), and the six regions investigated. In the bottom part of the panel, each row represents a single sample (i.e., NT and 5-Aza-dC) and each circle represents a single CpG in the PCR product. The percentage of CpG methylation is reported in different shades of gray. (B) Representative electropherograms of Gipr locus sequencing of bisulfite-converted DNA from treated (5-Aza-dC, 5 μM) and non-treated cells (NT). The triangles highlight the CpGs in which a methylation change has occurred. ** p < 0.001; * p < 0.05 in treated vs. NT cells.