Energy homeostasis and gastrointestinal endocrine differentiation do not require the anorectic hormone peptide YY

Abstract

The gastrointestinal hormone peptide YY is a potent inhibitor of food intake and is expressed early during differentiation of intestinal and pancreatic endocrine cells. In order to better understand the role of peptide YY in energy homeostasis and development, we created mice with a targeted deletion of the peptide YY gene. All intestinal and pancreatic endocrine cells developed normally in the absence of peptide YY with the exception of pancreatic polypeptide (PP) cells, indicating that peptide YY expression was not required for terminal differentiation. We used recombination-based cell lineage trace to determine if peptide YY cells were progenitors for gastrointestinal endocrine cells. Peptide YY(+) cells gave rise to all L-type enteroendocrine cells and to islet partial differential and PP cells. In the pancreas, approximately 40% of pancreatic alpha and rare beta cells arose from peptide YY(+) cells, suggesting that most beta cells and surprisingly the majority of alpha cells are not descendants of peptide YY(+)/glucagon-positive/insulin-positive cells that appear during early pancreagenesis. Despite the anorectic effects of exogenous peptide YY(3-36) following intraperitoneal administration, mice lacking peptide YY showed normal growth, food intake, energy expenditure, and responsiveness to peptide YY(3-36). These observations suggest that targeted disruption of the peptide YY gene does not perturb terminal endocrine cell differentiation or the control of food intake and energy homeostasis.

FIG. 1.

Targeted replacement of the mouse Pyy gene. A. Maps of the wild-type Pyy gene containing four exons, the targeting vector, and the disrupted alleles (before and after excision of PGK-neo^r). The lacZ gene was engineered to replace the Pyy gene. The 5′ and 3′ external probes used for Southern blotting are depicted by a black bars. (RI = EcoRI; APA = ApaI), and the PCR primers used for genotyping are shown as arrows. B. Southern analysis of ApaI digested DNA from Pyy^+/− intercrosses using the 3′ probe shown in panel A. Sizes of the wild-type, targeted neo⁻ and targeted neo⁺ alleles are 2, 5, and 7 kb, respectively. DNA correctly targeted at the 5′ end revealed a 5-kb EcoRI fragment with the 5′ probe (not shown). C. Reverse transcription-PCR of RNA from the colon of control, heterozygous, and null mice showing the presence of Pyy transcripts in Pyy^+/+ and Pyy^+/− mice but not in Pyy^−/− mice. 18S rRNA transcripts served as a positive control for reverse transcription-PCR in the null mice. D and E. The colon (D) and pancreas (E) of Pyy^−/− mice stained for β-galactosidase activity and PYY immunoreactivity. Note blue nuclear staining for β-galactosidase activity arising from expression of the targeted allele and the absence of brown staining for PYY. F. Colocalization of GLP-1 immunoreactivity (brown, cytoplasmic staining) with nuclear β-galactosidase activity (blue) in the colon of Pyy^−/− mice. G. Chromogranin A immunostained cells (brown) at the islet periphery colocalize with β-galactosidase in Pyy^−/− mice. H and I. All β-galactosidase stained cells in the colon (H) and islet mantle (I) in Pyy^+/− mice coexpress PYY (brown). Scale bar (in panel I): D, 110 μm; E and F 55 μm; G and H, 84 μm; I, 50 μm.

FIG. 2.

Weight gain in Pyy^+/+, Pyy^+/−, and Pyy^−/− mice. Pyy^+/+, Pyy^+/−, and Pyy^−/− littermates were weighed weekly from weaning until 6 months of age. Average weights are shown (n ≥ 11); analysis of variance of weights at all time points showed no significant difference between Pyy^+/+, Pyy^+/−, and Pyy^−/− mice.

FIG. 3.

Basal feeding and food intake in response to PYY administration in Pyy^+/+, Pyy^+/− and Pyy^−/− females. A. Basal cumulative and periodic food intake in Pyy^+/+, Pyy^+/−, and Pyy^−/− female mice. B. Cumulative and periodic food intake following intraperitoneal administration of PYY_3-36 (20 μg/100 g body weight) or vehicle (phosphate-buffered saline) in Pyy^+/+, Pyy^+/−, and Pyy^−/− female mice. For panels A and B, data are represented as means ± standard error of the mean; n = 4 to 8 mice/group. *, P < 0.05; **, P < 0.01; ***, P < 0.001 for phosphate-buffered saline- versus PYY-treated mice.

FIG. 4.

Oxygen consumption (VO₂) and respiratory exchange ratio (RER) in Pyy^+/+, Pyy^+/− and Pyy^−/− females. A. Basal VO₂ and respiratory exchange ratio in Pyy^+/+, Pyy^+/−, and Pyy^−/− female mice, n = 5 or 6 mice/group. B. VO₂ and respiratory exchange ratio following intraperitoneal administration of PYY_3-36 (20 μg/100 g body weight) in Pyy^+/+, Pyy^+/−, and Pyy^−/− female mice, n = 4 to 6 mice/group.

FIG. 5.

Endocrine differentiation in Pyy^−/− mice. A. Colon from a Pyy^−/− mouse showing normal distribution of chromogranin A (CGA)-immunopositive cells (arrows). B. Pyy^−/− mice show similar numbers of enteroendocrine cells staining for chromogranin A in the ileum and colon compared to wild-type mice. Results are expressed as mean number of stained cells counted per unit intestinal mucosal length ± standard error of the mean. C to H. Distribution of different endocrine cell types in the colon and pancreas of Pyy^−/− and Pyy^+/− mice. Tissues were examined for β-galactosidase activity by X-Gal staining (blue nuclear staining) and for hormones by immunoperoxidase staining (brown cytoplasmic staining). C. Serotonin-expressing cells were present in normal numbers and did not colocalize with β-galactosidase in Pyy^−/− colon. D. Normal islet architecture with insulin-expressing β cells in the islet core in Pyy^−/− mice. E and F. A subset of glucagon (E) and somatostatin (F) cells colocalized with β-galactosidase at the islet periphery in Pyy^−/− mice similar to their coexpression with PYY in normal mice. G. PP-immunopositive cells were absent from islets of Pyy^−/− mice. H. Cells staining for PP (brown) were readily identified in the pancreas of Pyy^+/− mice and colocalized with staining for β-galactosidase activity (blue). I. PP transcripts were not detected by reverse transcription-PCR of pancreatic RNA extracted from Pyy^−/− mice. In contrast, PP transcripts were detected in RNA from Pyy^+/− and Pyy^+/+ mice. Scale bar (in panel H): A, 17 μm; C, 22 μm; D to H, 25 μm.

FIG. 6.

Expression of a Pyy-hGH transgene is directed to PYY-expressing cells. A. Structure of the Pyy-hGH BAC transgene. The hGH gene was inserted into the first exon of the Pyy gene within the Pyy bacterial artificial chromosome (BAC) using homologous recombination in E. coli. B to D. Double immunofluorescent staining in the pancreas of Pyy-hGH mice for hGH (B, red), and PYY (C, green). Cells expressing both hGH and PYY appear yellow in the merged image (D). The arrow denotes a very rare PYY-negative/hGH-positive cell. E to G. Double immunofluorescent staining in the colon of Pyy-hGH mice for hGH (E, red), and PYY (F, green). The merged image (G) shows that hGH staining is only seen in PYY-expressing cells, which appear yellow. H. Double immunofluorescent staining in the pancreas of Pyy-hGH mice for hGH (green), and insulin (red). PYY expression is seen in a single rare β cell (arrow). Scale bar (in panel G): B to D, 98 μm; E to G, 50 μm; H, 85 μm.

FIG. 7.

Most pancreatic β cells and a significant fraction of glucagon-expressing α cells do not arise from PYY-expressing precursors. A. Descendants of PYY cells were marked by crossing Pyy Cre transgenic mice to the ROSA26 Cre indicator strain, which results in activation of lacZ expression in all PYY cells and all progeny of PYY-expressing cells. B. Double immunofluorescent staining for Cre (green) and PYY (red) in the pancreas of Pyy CRE mice. The arrow denotes a double-positive cell that is magnified in the inset. Cre is localized to the nucleus, and PYY stains the cytoplasm. Cre expression was restricted to PYY-expressing cells. C. Double immunofluorescent staining in the pancreas of Pyy Cre × RosA26 mice for β-galactosidase (green) and glucagon (red). Approximately 40% of α cells coexpressed β-galactosidase and appeared yellow, indicating that the majority do not arise from PYY-expressing cells. D to F. Double immunofluorescent staining in the pancreas of Pyy CRE × RosA26 mice for β-galactosidase (D, green) and insulin (E, red). An arrow denotes a rare double-positive cell in panel F. G and H. Most somatostatin-expressing ∂ cells arise from cells that express PYY, as shown by double immunofluorescent staining for somatostatin (G, red) and β-galactosidase (H, green) in Pyy Cre × RosA26 mice. I and J. All pancreatic polypeptide cells (I, red) appear to immunostain for β-galactosidase (J, green) in Pyy Cre × RosA26 mice, indicating that they are all related to PYY-expressing cells. For panels G to J, arrows indicate double-positive cells. Scale bar (in panel F): B, 28 μm; C to J, 50 μm.

FIG. 8.

PYY-expressing cells give rise to L-type enteroendocrine cells in the colon but not to serotonin-expressing enteroendocrine cells. Descendants of PYY cells in the colon of Pyy-Cre × ROSA26 mice were identified by double immunostaining for β-galactosidase and the indicated hormone. A. Double immunofluorescent staining in the colon of Pyy-Cre mice for PYY (green cytoplasmic staining) and Cre (red, nuclear staining) shows that Cre expression was restricted to PYY-expressing cells. B to D. Double immunofluorescent staining in the colon of Pyy-Cre × Rosa26 mice for β-galactosidase (B, green) and PYY (C, red). Double-stained cells appear yellow in the merged image (D). As expected, all L-type enteroendocrine cells express PYY at some stage of development. E. Serotonin-expressing enteroendocrine cells do not arise from PYY-expressing precursors. Double immunofluorescent staining in the colon of β-galactosidase (green) and serotonin (5HT, red) shows that serotonin cells do not express β-galactosidase. Scale bar (in panel E): A, 82 μm; B to D, 58 μm; E, 50 μm.