Leptin-specific patterns of gene expression in white adipose tissue - PubMed (original) (raw)

. 2000 Apr 15;14(8):963-80.

Affiliations

PMID: 10783168
PMCID: PMC316534

Leptin-specific patterns of gene expression in white adipose tissue

A Soukas et al. Genes Dev. 2000.

Abstract

Leptin is a hormone that regulates body weight by decreasing food intake and increasing energy expenditure. ob/ob mice carry leptin mutations and are obese and hyperphagic. Leptin administration to lean and ob/ob mice activates a novel metabolic program that depletes adipose tissue. Although this response is physiologically distinct from that evident after food restriction, the molecular nature of these differences is as yet unknown. Expression monitoring of 6500 genes using oligonucleotide microarrays in wild-type, ob/ob, and transgenic mice expressing low levels of leptin revealed that differences in ambient leptin levels have dramatic effects on the phenotype of white adipose tissue. These data identified a large number of genes that are differentially expressed in ob/ob mice. To delineate the components of the transcriptional program specifically affected by leptin, the level of the same 6500 genes was monitored in wild-type and ob/ob mice at various times after leptin treatment or food restriction. A novel application of k-means clustering identified 8 clusters of adipose tissue genes whose expression was different between leptin treatment and food restriction in ob/ob mice and 10 such clusters in wild-type experiments. One of the clusters was repressed specifically by leptin in both wild-type and ob/ob mice and included several genes known to be regulated by SREBP-1/ADD1. Further studies confirmed that leptin decreases the levels of SREBP-1/ADD1 RNA and transcriptionally active SREBP-1/ADD1 protein in white adipose tissue. Future studies of the molecular basis for the apparent coordinate regulation of the other clusters of leptin-regulated genes may reveal additional mechanisms by which leptin exerts its weight-reducing effects.

PubMed Disclaimer

Figures

Figure 1

Northern blots of differentially expressed genes. Northern blots were used to confirm the data for the differences in expression on the microarrays. DNA fragments from several genes with the fold change and average difference values as indicated were used as probes. (A) A total of 20 μg of wild-type and ob/ob white adipose tissue total RNA was analyzed by Northern blotting as described (see Materials and Methods) for genes up-regulated in ob/ob mice. (B) Northern blotting also confirmed the down-regulation of genes relative to wild-type in ob/ob white adipose tissue. Fold change and average difference (hybridization to perfect match probe sets minus hybridization to mismatch probe sets) values derived from Affymetrix Genechip software are shown below the relevant blots. Fold change values calculated in an instance where a gene was undetectable in one sample are preceded by the symbol ∼ and the undetected average difference value is marked by an asterisk (*). Blots were probed with cyclophilin to confirm equal loading. These data confirmed that robust expression data could be extracted from analysis of the microarrays for those genes with average difference values >250.

Figure 1

Figure 2

Reduction in body mass and food intake after leptin treatment. Leptin caused a rapid reduction in body mass and food intake in both ob/ob (A) and wild-type mice (B). Pair feeding of ob/ob mice but not lean mice demonstrated a kinetically and statistically distinguishable pattern of weight loss from leptin treatment (green circles) [(*) P < 0.05 relative to pair-fed(red triangles)]. At all points except the wild-type day 0 measurement for body mass, leptin treatment was significantly different from PBS (blue squares) in body mass and food intake (P < 0.01). The time points used for microarray expression analysis are indicated (†). These were ob/ob, untreated ob/ob, days 2, 4, and 12 of leptin treatment or pair feeding, and day 4 of PBS treatment; wild-type, untreated, days 1, 2, 3, and 5 of leptin treatment or pair feeding, and day 3 of PBS treatment.

Figure 2

Figure 3

_K_-means cluster analysis of leptin treatment and pair feeding in ob/ob white adipose tissue. (Left) The complete output of _k_-means cluster analysis is shown with the fold change indicated colorimetrically (see legend, bottom). Individual genes are represented along the vertical axis, whereas the experimental conditions (i.e., treatment group, day) are represented horizontally. The 16 ob/ob clusters are labeled with letters (A_–_P). The separations between clusters are indicated by the alternating red and blue bar on the left side of the colorimetric output. (Right) The mean expression pattern for the genes in each cluster was calculated and normalized so that 1 unit represents maximal repression/induction. Instances in which significant differences distinguish leptin treatment from pair feeding in clusters A, C, H, I, L–O are indicated [(*) P < 0.001 relative to pair feeding].

Figure 4

Clusters of genes specifically repressed by leptin in ob/ob white adipose tissue. Leptin specifically repressed expression of genes in at least two kinetically distinguishable clusters. (A) Leptin selectively repressed genes in cluster 3L. Many of these genes play a role in fatty acid biosynthesis. Genes are ordered by distance from the mean expression level, with distance from the mean for each gene indicated in units of standard deviation. Pair feeding only weakly repressed genes within the cluster. The transcription factor SREBP-1, which has been shown to regulate many genes in the cluster, is also present, along with four copies of ATP-citrate lyase. This pattern of expression was confirmed on Northern blots of RNA from ob/ob PBS, leptin, or pair-fed-treated, wild-type, or ob/ob Tg white adipose tissue using a Spot14 probe (bottom). Spot14 is a representative SREBP-1-regulated gene in the cluster. β-actin was used to confirm equal loading. At all time points, mean expression level for the leptin-treated group differed significantly from pair feeding [(†) P < 0.005, (*) P < 0.001]. (B) In cluster 3A, leptin specifically repressed genes up-regulated in the ob/ob animal, whereas pair feeding had no effect. Differences in mean expression pattern of these genes between leptin treatment and pair feeding were highly significant at all time points, as indicated. Northern blotting with a probe for FK506 Binding Protein 51, a representative gene in the cluster, confirmed the observed pattern of expression (bottom). Cyclophilin was used as a loading control.

Figure 5

_K_-means cluster analysis of genes regulated by leptin treatment and pair feeding in wild-type white adipose tissue. (Left) The complete cluster output is shown with the fold change indicated colorimetrically (see legend, bottom). Individual genes are represented along the vertical axis, whereas the experimental conditions (i.e., treatment group, day) are represented horizontally. Thirteen clusters are shown (labeled A_–_M), with boundaries between clusters indicated by the alternating red and blue bar to the left of the complete colorimetric output. (Right) Mean expression pattern was calculated for each cluster and normalized so that 1 unit represents maximal repression/induction. Significant differences between leptin and pair feeding exist in clusters B_–_I, K, and M as indicated [(*) P < 0.001 relative to pair feeding].

Figure 6

Clusters of genes specifically repressed by leptin in wild-type white adipose tissue. Leptin specifically repressed two kinetically distinguishable clusters in lean white adipose tissue. (A) The genes in cluster 5K are repressed selectively by leptin and to a much lesser extent by pair feeding. Many of these genes play a role in fatty acid biosynthesis. Genes are ordered by distance from the mean expression level, with distance from the mean for each gene indicated in units of standard deviation. Two copies of the transcription factor SREBP-1, which controls the transcription of many genes in the cluster, are also present. Pair feeding produced significantly less of an effect on these same genes as indicated [(*) P < 0.001 relative to pair feeding]. Northern analysis of RNA from treated wild-type, white adipose tissue for a representative gene in the cluster, Spot14, demonstrates the selective, transient repression by leptin (bottom). β-actin was used as a loading control. (B) The genes in cluster 5G are repressed by leptin but with a different kinetic pattern than the genes in cluster 5K. This cluster of genes includes leptin itself. Pair feeding produced significantly less repression of these same genes as indicated. Northern analysis with a probe for leptin confirmed the efficacy of treatment and the observed pattern of expression (bottom). Cyclophilin was used as a control.

Figure 7

Repression of SREBP-1 RNA and transcriptionally active protein by leptin. (A) Northern blotting with a probe for SREBP-1 of RNA from leptin-treated ob/ob white adipose tissue is shown (top). Leptin significantly repressed SREBP-1 RNA at days 2, 4, and 12, whereas pair feeding decreased SREBP-1 RNA slightly only on days 2 and 4, with levels returning to baseline by day 12. In wild-type mice, leptin produced a transient depression in SREBP-1 RNA levels, as assessed by Northern blot, which correlated with the state of energy balance (bottom). SREBP-1 RNA was depressed during the period of leptin-induced negative energy balance, and returned when animals re-enter a state of energy balance at day 5 of leptin treatment. No such depression was observed in the pair-fed, wild-type samples. (B) Immunoblotting of leptin treated, ob/ob, white adipose tissue nuclear extracts (15 μg) revealed that leptin decreased the levels of the transcriptionally active, 68-kD form of SREBP-1. Pair feeding appeared to moderately induce levels of nuclear SREBP-1 protein relative to PBS-treated samples. Immunoblotting with anti-CREB antibody was used as a control.

References

1. Ahima RS, Prabakaran D, Mantzoros C, Qu D, Lowell B, Maratos-Flier E, Flier JS. Role of leptin in the neuroendocrine response to fasting. Nature. 1996;382:250–252. - PubMed
1. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K. Current protocols in molecular biology. New York, NY: Wiley and Sons; 1999.
1. Boizard M, Le Liepvre X, Lemarchand P, Foufelle F, Ferre P, Dugail I. Obesity-related overexpression of fatty-acid synthase gene in adipose tissue involves sterol regulatory element-binding protein transcription factors. J Biol Chem. 1998;273:29164–29171. - PubMed
1. Brown MS, Goldstein JL. The SREBP pathway: Regulation of cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell. 1997;89:331–340. - PubMed
1. ————— A proteolytic pathway that controls the cholesterol content of membranes, cells and blood. Proc Natl Acad Sci. 1999;96:11041–11048. - PMC - PubMed

Leptin-specific patterns of gene expression in white adipose tissue - PubMed (original) (raw)