SnoRNA Snord116 (Pwcr1/MBII-85) deletion causes growth deficiency and hyperphagia in mice

Abstract

Prader-Willi syndrome (PWS) is the leading genetic cause of obesity. After initial severe hypotonia, PWS children become hyperphagic and morbidly obese, if intake is not restricted. Short stature with abnormal growth hormone secretion, hypogonadism, cognitive impairment, anxiety and behavior problems are other features. PWS is caused by lack of expression of imprinted genes in a approximately 4 mb region of chromosome band 15q11.2. Our previous translocation studies predicted a major role for the C/D box small nucleolar RNA cluster SNORD116 (PWCR1/HBII-85) in PWS. To test this hypothesis, we created a approximately 150 kb deletion of the > 40 copies of Snord116 (Pwcr1/MBII-85) in C57BL/6 mice. Snord116del mice with paternally derived deletion lack expression of this snoRNA. They have early-onset postnatal growth deficiency, but normal fertility and lifespan. While pituitary structure and somatotrophs are normal, liver Igf1 mRNA is decreased. In cognitive and behavior tests, Snord116del mice are deficient in motor learning and have increased anxiety. Around three months of age, they develop hyperphagia, but stay lean on regular and high-fat diet. On reduced caloric intake, Snord116del mice maintain their weight better than wild-type littermates, excluding increased energy requirement as a cause of hyperphagia. Normal compensatory feeding after fasting, and ability to maintain body temperature in the cold indicate normal energy homeostasis regulation. Metabolic chamber studies reveal that Snord116del mice maintain energy homeostasis by altered fuel usage. Prolonged mealtime and increased circulating ghrelin indicate a defect in meal termination mechanism. Snord116del mice, the first snoRNA deletion animal model, reveal a novel role for a non-coding RNA in growth and feeding regulation.

Figure 1. Derivation of Snord116 (Pwcr1/MBII-85) deletion mice.

A. Genomic region and structure of targeting constructs (not to scale). Top: Location of BACs used for making constructs. Middle: Map of Prader-Willi syndrome critical region (Pwcr) on mouse chromosome 7C. Gray thin line with arrowhead, large transcript originating from the Snrpn promoter; short black bars, coding exons for Snrpn on the left, and for Ube3a on the right; short gray bars, non-coding exons; high black bars, snoRNAs (not representing accurate numbers of elements in clusters, there are at least 41 copies of Snord116). Bottom: Left, upstream construct; right, downstream construct. Thick lines represent mouse genomic sequence, and thin lines vector sequence. Black arrowheads, loxP sites; white arrowheads, Frt1 sites; grey arrowheads, Frt 5 sites. Neo, Puro, TK and DT boxes indicate location of PGK-Neo^r, PGK-Puro^r , HSV-TK and diphtheria toxin selection markers. Positions and direction of the genotyping primers PA, PB, PC and PD are indicated. B. Southern blots of targeted ES cell clones with probes from outside the targeting constructs. Lanes 1 and 2, NcoI digests were hybridized with probe UL located upstream of the upstream targeting construct. Lane 1, wild-type ES cells (WT); lane 2, a recombinant ES cell clone (B41) targeted with the upstream construct. The WT allele generates a 19.5 kb band, and the recombinant fragment is 9.5 kb. Lanes 3-6, Stu1 digested DNA was probed with probe DR located downstream of the downstream targeting construct. Lanes 3 and 5, clone B41; lane 4, C1, a recombinant clone after transfecting B41 with the downstream targeting construct; lane 6, clone D1, derived from C1 after Cre recombinase-mediated recombination. The WT allele, represented by the upstream targeted clone B41, generates a band of >16 kb, and the recombinant fragments are 6.5 kb in C1 (a doubly-targeted 2-loxP clone) and 10 kb in D1 (a 1-loxP clone with the ∼150 kb deletion), respectively. C. Genotyping of 1-loxP ES cells by multiplex PCR with primers PA, PC and PD. The WT band is 435 bp, the amplification product of primers PC and PD; and the deleted allele is 327 bp, a product of amplification between PA and PD. Lanes 1 and 2 represent two clones after Cre mediated deletion (D1 and D5); Lane 3, wild-type ES cells; lane 4, doubly targeted 2-loxP ES cell clone (C1). H₂O: no template negative control; MW: molecular weight marker. D. Quantitative real-time RT-PCR study for expression of gene elements in PWS critical region. DNaseI treated total brain RNA from 4-week old mice (n = 9 WT, 7 Del) was used as template for qRT-PCR assays. Each sample was assayed in triplicate. Gapdh expression was used as internal control. The levels for each sample are normalized to the average of the WT samples. Error bars, standard error of the mean (SEM). Asterisks indicate significant differences; for Snord64, P<0.0005; Snord116, P<1 ×10⁻⁷, and Snord115, P<0.01.

Figure 2. Growth and somatic development of Snord116del mice.

A. Daily weight curves for males (left) and weekly weight curves for males (middle) and females (right). The female weight curve in the first month (not shown) overlaps with the male curve. Error bars, standard deviations. Significant weight differences between Snord116del mice and their WT littermates ranged from P<0.05 in young pups, after postnatal day 2 (P2), to P<1 ×10⁻⁸ after 6 months (n = 15∼20 in each sex/genotype group). Insert photo: Snord116del (top) and WT (bottom) pups at P14. B. Comparison of body and organ weights in Snord116del mice and WT littermates at P5 and P13 (P5: n = 27 WT, 24 Del; P13: n = 8 WT, 9 Del). Error bar: SEM. At P5, all differences were significant at P<0.003 except for brain at P<0.02. At P13, all differences were significant at P<0.0001 except for stomach, P = 0.05. Bd: body, Br: brain, Lv: liver, St: stomach. C. Levels of Igf1 transcripts in liver and IGF-I protein in serum are reduced in deletion mice. Left: Liver RNA from 4-week old (n = 9 WT, 7 Del) and 8-week old (n = 8 WT, 8 Del) mice were assayed by qRT-PCR with Gapdh levels as internal control. WT levels were set at 1. Error bar: SEM. Asterisks: P<0.05. Right: Serum levels of IGF-I protein from the same group of mice were measured by ELISA. Asterisks: P<0.0005. D. Hematoxylin and eosin stained sagittal sections of the pituitary at E12.5 and E14.5 (left WT, middle Del) and coronal sections at E18.5 (right) revealed normal pituitary size and structure. E. Immunohistochemical staining for growth hormone (somatotrophs) revealed no detectable differences in WT (top) and Del (bottom) mice at 4 weeks of age.

Figure 3. Neurobehavioral tests of Snord116del and WT littermates.

A. Elevated plus maze. Mice were tested at 3 months of age (n = 15, both genotypes). Left panel: Snord116del mice spent more time in the closed arms (P<0.05). Right panel: mutant mice made more entries to the closed arms (P<0.02). B. Wire-hanging test. At 5 months of age, Snord116del mice were able to hang significantly longer than the WT littermates (P<5 ×10⁻⁵, n = 17, both genotypes). At 2 months of age, the difference was not significant (n = 13 WT, 9 Del). C. Rotarod test of balance and motor learning. Mice were tested at 2 months (n = 18 WT, 10 Del). Two trials were done on each day for 6 consecutive days. Asterisks indicate significant differences between WT and deletion mice (P<0.05). All error bars are SEM.

Figure 4. Feeding and energy homeostasis regulation of Snord116del mice and their WT littermates.

A. Daily food intake and response to High-Fat (HF) diet. The intake of rodent chow during the dark (black bars) and light cycles (open bars) were measured for five consecutive days at 7 weeks of age. Then, the mice were placed on HF diet, and their intake was measured daily for five consecutive days. Food intake was normalized to average body weight during the five days of measurement. The daily intake on regular chow diet was not different between WT (W) and Snord116del (D) mice. On HF diet, the daily intake of the deletion males (HD) was significantly lower than that of their WT littermates (HW) (P<0.005). This was entirely due to decreased daytime intake. The deletion females on HF (HD) had lower intake in the dark (P<0.05), which was compensated by increased daytime intake (P<0.05), so that their total daily intake was not different from their WT littermates (HW). (Males, n = 16 WT, 16 Del; females n = 15 WT, 11 Del). B. Deletion mice gain less weight on HF diet. Eight-week old mice were put on HF for 4 months, and their body weight was measured twice per week and plotted as the ratio to their weight prior to HF diet. Differences between male WT (MW, n = 16) and male Snord116del (MD, n = 16) mice were significant at P<0.01 except for the first week on HF. Differences between female WT (FW, n = 15) and female deletion (FD, n = 11) mice were significant at P<0.05 after 7 weeks on HF. C. Deletion mice have less body fat on regular and HF diet. Dual energy X-ray absorptiometry was used to measure body composition of 5-months and 9-months old mice on rodent chow, and 6-months old mice that had been on HF diet for 4 months. The differences between WT and Snord116del mice are all significant, except for the 5-months old males on chow. Males, at 5 months (n = 13 WT, 9 Del; P = 0.07); at 9 months (n = 17, both genotypes, P<2 ×10⁻⁷); at 6 months after 4 months on HF diet (n = 16, P<0.0001). Females at 5 months (n = 12, both genotypes, P<0.02), at 9 months (n = 14 WT, 10 Del; P<0.007) and at 6 months after 4 months on HF diet (n = 15 WT, 10 Del; P<0.02). D. Compensatory food intake after fasting. Mice at 3 months of age were fasted for 24 hr, and the ratio of their cumulative food intake to their average pre-fasting daily intake was computed. M1 and M2, cumulative intake ratios of males in the first 24 and 48 hr (n = 13 WT, 9 Del). F1 and F2, cumulative intake ratios of females in the first 24 and 48 hr (n = 12). The deletion mice were able to compensate within 48 hr for the missed feeding. The deletion males over-compensated compared to their WT littermates (P<0.02). E. Maintenance of body temperature in cold environment. The core body temperature of 4 months old Snord116del (n = 9) and WT (n = 13) males in a 4°C cold room was measured with an anal probe. The mutants were better able to maintain their core temperatures over extended time. Asterisks indicate P<0.05.

Figure 5. Glucose and insulin tolerance tests of male mice.

A. Glucose tolerance test (GTT) for mice on regular diet (RD) at 6 months of age. Mice were injected with 1 g/kg glucose at time zero after 16 hr overnight fasting and serum glucose was measured for up to 2 h. n = 13 WT, 9 Del). Deletion mice were better able to clear the glucose that WT, P<0.05 at 30 min, and P<0.001 at 60 and 90 min time points). B. Glucose tolerance test for mice after 4 months on HF diet (n = 16) starting from 8 weeks of age (P<0.01 for all time points). C. Insulin tolerance test (ITT) for mice on regular diet (RD) at 6 months of age. Wild type (n = 13) and mutant (n = 9) mice were injected with 1.6 U/kg insulin at time zero after 5 hr fasting from the onset of the light cycle (P<0.05 at 60, 90 and 120 min). D. Insulin tolerance test for 6-months old mice after 4 months of HF diet. WT mice (n = 15) and deletion mice (n = 16) were injected with 2 U/kg insulin (P<0.02 at time 0, and P<0.0002 at all other time points). Deletion males were more sensitive to insulin than WT.

Figure 6. Snord116del mice are hyperphagic and have altered metabolism.

A. Daily food intake during 6 days in Oxymax metabolic chambers was normalized to body weight (n = 8 WT, 7 Del, all points P<0.05). Error Bars: SEM. B. Average feeding bout duration of the same group of mice as in (A) during each 12 hr cycle. D, dark cycles; L, light cycles. Asterisks indicate P<0.05. C. Daily food intake at different ages. Daily food intake was averaged over 5–7 consecutive days and normalized to body weight of each individual. At 7 week, males n = 16 both genotypes; females, n = 15 WT, 11 Del; at 3 months, males n = 13 WT, 9 Del, P<0.04; females n = 12 WT, 12 Del; at 6–7 months, (males n = 15 both genotypes, P<2 ×10⁻⁶; females n = 14 WT, 16 Del, P<3 ×10⁻⁶, and at 10 months males n = 21 WT, 19 Del, P<3 ×10⁻⁵; females n = 13 both genotypes, P<2 ×10⁻⁵. D. Oxygen consumption rates (averaged over 2 hr time periods) on the second and third day in the metabolic chamber (n = 15 both genotypes). Bars at bottom indicate light and dark cycles. Asterisks, P<0.05. E. Respiratory exchange rate (RER) averaged over 2 hr time periods for the same group of mice as in (D) during the same time in the metabolic chamber. Asterisks, P<0.05. F. Weight change in mice food-restricted to 80% their normal daily intake. Males (n = 11 both genotypes) and females (n = 13 both genotypes) at 11 months of age. Asterisks indicate significant difference between WT and Del females with P<0.05. G. Snord116del mice have increased ghrelin levels. Plasma ghrelin levels were measured in ad lib fed mice at 11 months of age (n = 12 both genotypes; P<0.002) and 24 hr fasted mice (n = 13 both genotypes ; P<0.0002)