agrp Is Required for Normal Somatic Growth in Larval Zebrafish

(A–D) Representative fish at 3 dpf following injection at day 0 with morpholino oligonucleotides designed to inhibit expression of each of the zebrafish agouti proteins (A). Body lengths (jaw to tail fin) of injected fish were measured using a micrometer at 5 dpf, showing a dose-responsive inhibition of growth following morpholino blockade of agrp expression (B). Body lengths corresponding to fish injected at day 0 were measured at 14 dpf, showing partial restoration of length (C). Body lengths at 5 dpf following injection of zygotes with 5 ng morpholino oligonucleotide are indicated, plus 30 pg capped agrp RNA, where indicated, showing partial rescue of growth (D). Bars indicate mean ± SEM. n = 30 (B), 10 (C), and 18–23 (D). Statistical significance was tested by one-way ANOVA followed by Tukey′s post test (p < 0.05; p < 0.01; p < 0.001). See also Figure S1

mc4r Is Required for Suppression of Somatic Growth by Morpholino Oligonucleotide Blockade of agrp

(A–D) Bar diagram showing the size of wild-type and mutant single mc4r coding alleles in three lines identified by TILLING (Sanger Institute) (A). The maximal potential protein coding segment of each receptor mutant is as follows: sa0122, 18 aa; sa0148, 30 aa; sa0149, 185 aa. mc4r mutations do not affect early larval growth (B). Offspring from heterozygote matings of sa0122, sa0148, and sa0149 were measured at 5 dpf, then genotyped by PCR. Numbers of fish analyzed in each group from left to right are 29, 47, 20 (sa0122); 26, 40, 27 (sa0148); and 25, 49, 19 (sa0149). Representative injected (5 ng agrp MO) or uninjected control sa0149 fish at 5 dpf with genotype and experimental treatment are indicated (C). Body lengths of fish treated as indicated (n = 24–36) were measured with a micrometer at 5 dpf, showing no effect of morpholino blockade of agrp on growth in the sa0149 mc4r/ fish (D). Bars indicate mean ± SEM. Results were analyzed by one-way ANOVA followed by Tukey′s post test (p < 0.001). See also Figure S2

agrp Regulates Expression of Multiple Pituitary Hormones in Zebrafish

(A–F) Whole-mount in situ hybridization of gh (A and B) and pomca (C and D) in fish injected with 2.5 ng standard control (A and C) or agrp MO oligonucleotides (B and D). Scale bars: 100 µM. Relative expression levels of gh (growth hormone), ghrh (growth hormone-releasing hormone), sst1 (somatostatin 1), sst2 (somatostatin 2), pomca (pro-opiomelanocortin a), pit1 (pituitary-specific positive transcription factor 1), sml a (somatolactin a), sml b, tsh (thyroid-stimulating hormone), trh (thyrotropin-releasing hormone), crh (corticotrophin-releasing hormone), fshb (follicle-stimulating hormone), lhb (luteinizing hormone), igf1a (insulin-like growth factor 1a), and igf1b were analyzed by Q-PCR in 30–40 4 dpf fish injected with 2.5 ng standard control or agrp MO oligonucleotides (E and F). mRNA expression was normalized to ef1a mRNA. Results were expressed as mean + SEM, and statistical analysis was done by unpaired t test (p < 0.05; p < 0.001; NS, not significant). See also Figure S3

Hypothalamic AgRP- and α-MSH-Expressing Neurons Project to the Pituitary

Horizontal view of larval zebrafish brain at 5 dpf illustrating the pituitary and underlying hypothalamus. Large white oval approximates the extent of the pituitary, and small yellow oval indicates a zone within the proximal pars distalis containing dense α-MSH-immunoreactive (ir) fibers (red) and AgRP-ir fibers (green). Large arrows indicate hypothalamic AgRP-ir fiber bundles, medium arrows indicate hypothalamic AgRP-ir cell bodies, and thin arrows indicate AgRP-ir fibers projecting from hypothalamus into the pituitary. Inset is an enlargement from the PPD showing parallel AgRP-ir and α-MSH-ir neuronal fibers. PI, pars intermedia; PPD, proximal pars distalis; RPD, rostral pars distalis; H, hypothalamus. Scale bars: main image, 10 µm, inset = 1 µm.

Morpholino Oligonucleotides Blocking agrp Splicing Reduce Normal Somatic Growth in Larval Zebrafish.

(A) Bar diagram showing the targets and predicted consequences of two morpholino oligonucleotides designed to block splicing of the agrp mRNA. (B) Efficiency of each antisense MO was confirmed by RT-PCR at 2dpf. Red arrows show predicted longer form of agrp RNA with disruption of the first splice donor (E1I1, left panel) and shorter form of agrp RNA resulting from blockade of first splice acceptor (I1E2, right panel), as seen on agarose gel from agrp E1I1 exon donor morphant and agrp I1E2 exon acceptor morphant, respectively. Predicted protein sequences of WT and morphant AgRP proteins are indicated. (C) Body lengths of fish injected with standard control MO, E1I1 MO and I1E2 MO were measured with a micrometer at 5dpf. Bars indicates mean±s.e.m. Results were analyzed by one way ANOVA followed by Tukey post test. (n=36, 33 and 42. ***, p<0.001).

Role of MC4R, AgRP and POMC in Zebrafish Growth.

(A-C) mc4r-/- fish exhibit increased growth by 42 dpf. Fish from a sa0149 mc4r +/- intercrossing were allowed to grow to adulthood. Body lengths (jaw to trunk terminus, tail fin excluded) of juveniles were measured at 42 days post fertilization (A). Female (B) and male (C) adult zebrafish were also measured at 8 months. Bars indicate mean±s.e.m. Results were analyzed by one way ANOVA followed by Tukey post test. (N=5-53, *P<0.05;***P<0.001). (D) Partial rescue of agrp MO inhibition of somatic growth by co-injection of pomca MO. Body lengths at 5 dpf following injection of wild type zygotes with 2 ng MO indicated, plus 0.4ng pomca ATG antisense MO showing partial rescue of growth. Bars indicates mean±s.e.m. n= 42, 42, 39. Statistical significance tested by one way ANOVA followed by Tukey post test. (***, p<0.001). (E-F) Developmental expression of melanocortin receptor and ligand genes. Relative expression levels of (E) mc1r, mc3r, mc4r, mc5ra, and mc5rb, and (F) agrp and pomca were analyzed by QPCR. ~200 WT zebrafish zygotes were raised and sacrificed at days indicated for RNA extraction and cDNA synthesis. 4dpf group was normalized to 100% and all gene expression was normalized to the house-keeping gene, ef1a or β-actin. Results are expressed as mean + s.e.m.

No Effect of agrp MO in the mc4r-/- Background

(A-D) Whole mount in situ hybridization of gh (A-B) and pomca (C-D) in uninjected sa0149 wild type fish (A and C) or un-injected sa0149 mc4r -/- fish (B and D) at 4dpf. Scale bar: 100µm. (E) Relative expression levels of agrp, mc4r, gh, ghrh, pomca, pomcb were analyzed by Q-PCR. Offspring from sa0149 sibling wild type and sa0149 mc4r-/- matings were raised at standard condition. 30 uninjected embryos from each condition were divided into 3 groups and sacrificed at 4dpf for RNA extraction and cDNA synthesis. Wild type group was normalized to 100% and all gene expression was normalized to the house-keeping gene, ef1a. Results are expressed as mean + s.e.m., and statistical analysis was done by unpaired t-test (Not Significant for all genes). (F-I) Whole mount in situ hybridization of gh (F-G) and pomca (H-I) in mc4r -/- fish injected with 2.5ng standard control (F and H) or agrp morpholino oligonucleotides (G and I). Scale bar: 100µm.

Expression of mc4r RNA in Zebrafish Pituitary.

RT-PCR amplification of mc4r and ef1a from zebrafish pituitary. Pituitary was carefully dissected and 1.9µg total RNA was extracted from 42 one year old WT adults. 10ng was used for each reaction as described in Experimental Procedures.