Regulation of hindbrain Pyy expression by acute food deprivation, prolonged caloric restriction, and weight loss surgery in mice

Mouse studies.

Mice were maintained on a 12:12-h light-dark cycle (lights on at 0700). All in vivo studies were performed in accordance with the United Kingdom Home Office Animal Procedures Act (1986) and the UCL Animal Ethics Committee. C57Bl/6 mice were purchased from Charles River (Margate, UK). Timed matings were set up between adult C57BL6/J mice. Females were inspected daily for plugs with the day of plug detection considered as E0.5. Embryos from three pregnant females were dissected for each stage (E9.5 and E13.5) and eight tissue blocks covering the mesencephalon to myelencephalon (E9.5) or brain stem (E13.5) were pooled for gene expression analysis. In addition, brain stems were dissected from P2 and P13 pups and P70 adults.

The generation of Pyy-Cre transgenic mice was described previously (52). Cre-positive offspring were back-crossed to C57BL6/J for 10 generations. In the resulting Pyy-Cre transgenic mice, the expression of Cre recombinase is under the control of the regulatory regions located in the first untranslated exon of the murine Pyy gene. ROSA26R-YFP reporter mice, in which Yfp expression is blocked by a loxP flanked STOP cassette, were obtained from Jackson Laboratories (Bar Harbor, ME). Crossing of YFP reporter mice to Pyy-Cre transgenic mice resulted in the generation of PYY-Cre;ROSA-EYFP mice, in which Cre-mediated excision of the STOP cassette limits Yfp expression only to Pyy-expressing cells. Homozygotic mice with a deletion of the gene encoding Npy and their wild-type (WT) littermates were derived by heterozygous breeding (22). The generation and breeding of PyyKO mice was described previously (8).

Dietary studies.

Six-month-old male C57BL6/J mice (n = 20) were singly housed for 1 wk and then randomized in two groups: mice in the first group were maintained on a normal chow diet (n = 10); those in the second group were calorie restricted for 4 wk (CR; n = 10). CR was carried out by a step-down regimen (16, 55), and mice received 80% of their control group's (Control diet) food. Food pellets were given at the onset of the dark phase (1800). At the end of the study, mice were culled during the first 3 h of the light phase after an overnight fast (14 h, between 1800 and 0800). After decapitation, brain stems were dissected, snap-frozen, and stored at −80°C until further use.

Surgical procedures.

Four-week-old male C57BL6/J mice were fed a high-fat diet (HFD) for 16 wk. Macronutrient composition and fat content of the control diet were 70% carbohydrate; 10% fat, and 20% protein; macronutrient composition and fat content of HFD were 35% carbohydrate, 45% fat, and 20% protein. They were then transported to facilities where they underwent surgery. Following 1 wk of acclimatization, mice were divided into two weight-matched groups to undergo either the EGA procedure (n = 8) or sham operation (n = 8) (60). In the EGA procedure, a laparotomy was performed, and the pyloric sphincter was ligatured, followed by an anastomosis between midjejunum and stomach, resulting in the exclusion of the duodenum and the proximal jejunum from the alimentary tract (60). Sham-operated mice underwent the same duration of anesthesia and a laparotomy was performed. Postoperatively, mice were fed the HFD and culled 10 days after the surgery, following an overnight fast. Tissue harvesting was performed as described above. Hormonal assays in mice exposed to different dietary regimens and EGA procedure were performed as described previously (16).

Transcardiac perfusion, immunohistochemistry, and in situ hybridization.

Transcardiac perfusion was accomplished using a peristaltic pump with 4% paraformaldehyde (PFA). After the perfusion, brains were removed from the skulls and stored in the same PFA solution at 4°C overnight. Fixed brains were then cryoprotected in 30% sucrose at 4°C overnight and stored at −80°C until sectioning. Immunohistochemistry (IHC) and in situ hybridization (ISH) were performed on 30-μm coronal sections as previously described (18). Rabbit anti-peptide YY (Peninsula Laboratories, San Carlos, CA) and rabbit anti-5HT (Immunostar, Wisconsin) antibodies were used to detect PYY and serotonin [5-hydroxytryptamine (5-HT)] neurons. DIG-labeled ISH riboprobe was generated using Pyy mouse sequence. Imaging was performed with an Olympus BX51 microscope combined with SimplePCI software. Confocal microscopy was performed on a Bio-Rad MRC1000 microscope. For phospho-signal transducer and activator of transcription 3 (pSTAT3) immunostaining, 24-h-fasted 3-mo-old PYY-Cre;ROSA-EYFP mice were injected with 5 μg/kg ip recombinant mouse leptin (R&D Systems, Abingdon, UK) or vehicle and perfused 45 min later. Brain stem and hypothalamus sections were stained for pSTAT3 as previously described (54, 54) using rabbit pSTAT3 (Tyr⁷⁰⁵) antibody (Cell Signaling Technology, UK).

Gene expression analysis.

The effect of peripheral PYY-(3–36) administration on central Pyy expression was assessed in C57BL6/J mice. Twelve-week-old male mice (n = 20) were divided into two weight-matched groups to receive an injection of either saline or PYY-(3–36) (50 ng/g ip; Bachem, UK). All injections were started at 0800 following an overnight fast, and mice were culled 6 h post-injection. Tissue harvesting was performed as described above.

For gene expression analysis, total RNA was extracted from frozen tissues using TRIzol reagent, and 2 μg of RNA was reverse transcribed to cDNA. Real-time quantitative PCR was performed on an ABI Prism 7900HT system (Applied Biosystems, Foster City, CA) using predesigned assays. Various housekeeping genes were tested for their expression stability across different samples; glyceraldehyde-3-phosphate dehydrogenase and ubiquitin C (Applied Biosystems) were found to be most stable and thus used as endogenous controls. Relative expression of the gene was determined by comparing its expression to that of the endogenous control. Data evaluation was carried out using SDS v. 2.1 software.

Peptide extraction and chromatographic and RIA procedures.

Fed and 24-h-fasted mice were used for the assessment of PYY isoforms in the brain stem. For peptide extraction, harvested tissues were boiled (95°C for 10 min) in 10× their volume of 1 M acetic acid. The extraction solutions were homogenized and centrifuged. The supernatant was loaded onto a 360-mg C₁₈ SepPak cartridge (Waters, Milford, MA), which had been activated with acetonitrile (ACN) and equilibrated with 0.12% trifluoroacetic acid (TFA). After loading of the supernatant, the SepPak was washed with 0.12% TFA and eluted with 80% ACN-water solution containing 0.1% TFA, and dried SepPak fractions were then dissolved in 100 μl of water prior to loading of the column. Reverse-phase HPLC was carried out on a Shimadzu Class VP System using an Alltech Alltime C₁₈ column (15 × 4.6 mm). The solvent systems used were: (A) 0.1% (vol/vol) TFA in water and (B) 0.1% (vol/vol) TFA in ACN. The solvent flow rate was 1 ml/min, and the elution was carried out with an initial gradient of 29% solvent B over the first minute followed by a gradient of 29–31% solvent B over the next 50 min. Brain stem extracts from fed and 24-h-fasted mice (n = 3 per group) were injected onto the column, and based on the retention times of the standards, 250-μl fractions were collected between 5 and 25 min and tested for the presence of PYY-like immunoreactivity (PYY-LI) using a rat/mouse total PYY RIA (Millipore,Watford, UK).

Statistical analysis.

All group data are expressed as means ± SE unless otherwise indicated. Variations in Pyy and receptor mRNA levels were assessed using Student's t-test or one-way ANOVA. Post hoc tests were performed in case a significant effect was detected (the Bonferroni correction was used when the equality of variances assumption held, and the Dunnett t3 correction was used otherwise). For all statistical analyses, P < 0.05 was considered significant. Data were analyzed using SPSS v. 14.0.

Distribution of PYY neurons in mouse brain stem and validation of PYY-Cre;ROSA-EYFP mouse.

In agreement with the studies undertaken by Glavas et al. (27), ISH on mouse brain stem sections indicated only one population of Pyy-expressing neurons scattered throughout the Gi region (Fig. 1, A and B). In addition, Yfp-expressing neurons were present in the Gi region of PYY-Cre;ROSA-EYFP mice (Fig. 1C). Nearly all PYY immunoreactive cells expressed EYFP, confirming that the expression of the PYY-Cre transgene occurred in all PYY cells as expected. Approximately 70% of EYFP-expressing cells stained for PYY, indicating that a fraction of the EYFP-positive cells that arose from PYY cells no longer expressed the peptide (Fig. 1, D and E). PYY IHC on Gi sections from Pyy-null mice (PyyKO) did not reveal any PYY-ir.

Because lineage tracing experiments showed that some descendants of PYY cells discontinued producing the peptide, we examined the developing hindbrain for Pyy mRNA expression. More specifically, we investigated brain stem Pyy expression at E9.5 and E13.5 (n = 4 per age) and at P2, P13, and P70 (n = 6 per age). Pyy mRNA was already present at E9.5 in the part of the brain covering mesencephalon to myelencephalon, and its level changed significantly at various developmental stages [F (5,28) = 34.10, P = 0.0001; Fig. 2A]. Pyy expression increased from day E9.5 and reached a peak at P2 with a significant increase between E13.5 and P2 (Dunnett's post hoc P = 0.0001). From P2 onward, Pyy expression started to decrease, and by P70 its expression had decreased by 69% compared with P2 (P = 0.0001; Fig. 2A). The presence of Pyy mRNA and protein at the P2 stage was further confirmed by ISH and IHC undertaken on brain stem sections from NpyKO mice (Fig. 2, B and C).

Due to the concerns of cross-reactivity of the PYY antibody with NPY, we investigated the distribution of PYY-ir fibers in mice lacking NPY (NpyKO). We observed PYY-ir fibers scattered throughout the DVC, including the nucleus of the solitary tract (NTS), dorsal motor nucleus of the vagus (10N), and the hypoglossal nucleus (12N) (Fig. 3, A–F). Taken together, our data indicate the presence of PYY-expressing cell bodies in the Gi region of the hindbrain and positive fibers throughout the DVC and 12N.

Raphe magnus (RMg), raphe obscurus (Rob), and raphe pallidus (RPa) of the brain stem raphe nuclei are located in close proximity to the Gi region (59). 5-HT IHC on hindbrain sections from PYY-Cre;ROSA-EYFP mice showed that PYY neurons lie just above the RMg and RPa and lateral to Rob cell bodies (Fig. 4A). Single scan and serial z-stack images obtained at 2-μm intervals demonstrated close appositions between 5-HT and PYY cell bodies (Fig. 4B). 5-HT fibers were also in close apposition with PYY neurons, suggestive of possible synaptic contacts (Fig. 4C). Furthermore, ∼8% of the PYY cell bodies in the RMg and Rob nuclei costained with 5-HT antibody (Fig. 4, D–F).

Changes in brain stem Pyy and Y-receptor mRNA levels in response to food deprivation and prolonged CR.

Because of the peripheral expression of the Y4r (15) and studies indicating a lack of involvement of y6r in feeding regulation and absence of a functional y6r in humans and rats (12, 15, 46), we investigated the expression of brain stem Pyy, Y1r, Y2r, and Y5r under different feeding conditions and following weight loss surgery. The expression of Pyy and the three receptors in male C57BL/6 mice was unaltered by overnight fasting (data not shown). However, 24-h fasting led to a significant reduction in Pyy expression [t(28) = 3.38, P = 0.002] but did not alter Yr expression (Fig. 5A). Four weeks of CR caused a significant decrease in body weight [Fig. 5B; t(17) = −8.18, P = 0.0001]. Whereas fasting peripheral acyl-ghrelin and PYY levels were unaltered (data not shown), circulating leptin levels were significantly decreased in response to CR [Fig. 5C; t(4.44) = 10.24, P = 0.001]. Brain stem Pyy and Y1r expressions were reduced in the CR mice [t(15.64) = 3.29, P = 0.005; t(17) = 3.49, P = 0.003, respectively; Fig. 5D] and were positively correlated (r = 0.498, P = 0.03). Neither Y2r nor Y5r expression was altered by 4 wk of CR. Body weight loss correlated negatively with both Pyy (r = −0.561, P = 0.015) and Y1r expression (r = −0.576, P = 0.012). Furthermore, Y1r expression correlated positively with fasting leptin levels (r = 0.686, P = 0.002).

Changes in brain stem Pyy and Y-receptor expression following EGA surgery.

We (16) recently reported that EGA procedure resulted in a significant reduction in body weight and fasting circulating leptin levels, whereas acyl-ghrelin and total PYY levels were markedly elevated. Brain stem Pyy and Y1r expression decreased significantly in response to EGA procedure [t(11.86) = 3.76, P = 0.003; t(13) = 2.20, P = 0.046, respectively; Fig. 5E] and were positively correlated (r = 0.753, P = 0.001). Both Pyy and Y1r expression correlated with body weight loss (r = −0.526 P = 0.044; r = −0.533 P = 0.041, respectively), circulating leptin (r = 0.623, P = 0.022; r = 0.546, P = 0.044), and PYY levels (r = −0.627, P = 0.012; r = −0.574, P = 0.025). Y2r and Y5r expression was not affected by EGA surgery.

In view of the relationship between circulating leptin and PYY levels and central Pyy expression, we investigated whether intraperitoneal leptin and PYY-(3–36) administration acted on brain stem PYY neurons. Peripheral leptin administration following 24-h fasting induced pSTAT3 staining in the arcuate nucleus, ventromedial hypothalamic nucleus, and dorsomedial hypothalamic nucleus of the hypothalamus (data not shown). However, no pSTAT3 staining was observed in brain stem PYY neurons, suggesting that leptin does not act directly on PYY neurons in 24-h-fasted mice (data not shown). TaqMan real-time PCR assay showed no effect of peripheral PYY-(3–36) injection on brain stem Pyy mRNA levels in mice following an overnight fast (mRNA expression in AU: Saline, 2.19 ± 0.08; PYY, 2.10 ± 0.07).

Assessment of brain stem PYY-(1–36) and PYY-(3–36) levels using HPLC/RIA.

All chromatographic procedures were performed in NpyKO mice to prevent false positive results due to cross-reactivity of NPY with the PYY antibody. Therefore, prior to HPLC, we determined Pyy expression in the brain stems of NpyKO and WT mice to ensure that the expression was unaffected by Npy deletion; Pyy mRNA levels were similar in both groups (data not shown). Following the injection of a mixture containing both peptides, the first peak, corresponding to PYY-(3–36), was detected at 16 min and the second peak, corresponding to PYY-(1–36), at 18 min. The ratio of PYY-(1–36) to PYY-(3–36) in brain stem extracts from fed and 24-h-fasted mice was significantly different [3.26 ± 0.29 fed vs. 1.33 ± 0.33 fasted group; t(2.05) = 6.48, P = 0.002]. Figure 6, A and B, shows data from one representative chromatogram from fed and 24-h-fasted mice. Under fed conditions, PYY-(1–36) was the dominant form in the brain stem, whereas under fasted conditions PYY-LI in fractions corresponding to both isoforms was similar. This was due both to an increase in PYY-(3–36)-LI and to a decrease in PYY-(1–36)-LI in the fasted state (Fig. 6C). The high SEM in the fed group is due to one mouse having lower PYY-(3–36) and PYY-(1–36)-LI. However, the ratio of the two isoforms was similar to that of the other two mice.