Rapid Publication - Semantic Scholar

5 downloads 0 Views 619KB Size Report
Erin E. Kershaw, Wendy K. Chung, Loraine Power-Kehoe, Melvin Chua, Louis A. Tartaglia, .... Future Leaders Award (S.C.C.); Hoffman LaRoche, Nutley,.
Rapid Publication Phenotype offatty Due to Gln269Pro Mutation in the Leptin Receptor (Lepr) Streamson C. Chua, Jr., David W. White, X. Sharon Wu-Peng, Shun-Mei Liu, Norichika Okada, Erin E. Kershaw, Wendy K. Chung, Loraine Power-Kehoe, Melvin Chua, Louis A. Tartaglia, and Rudolph L. Leibel

The ratfatty (fa) mutation produces profound obesity of early onset caused by hyperphagia, defective nonshivering thermogenesis, and preferential deposition of energy into adipose tissue. Genetic mapping studies indicate that fa and diabetes (db) are homologous loci in the rat and mouse genomes, respectively. It has been shown that db alleles carry mutations in the Lepr (leptin receptor) gene. This paper describes a point mutation in the fatty allele of Lepr. A nucleotide substitution at position 880 (A —»C) causes an amino acid substitution at position 269 (Gin —> Pro). The mutation generates a novel Msp I site that cosegregates with fa in 1,028 meioses examined in obese F2 progeny from two crosses (BNxl3M and WKYxl3M) and is still segregating in three rat colonies. PCR-based mutagenesis was used to introduce the fa mutation into the mouse Lepr cDNA. Transient transfection studies indicate that the mutant Lepr cDNA has greatly reduced binding of leptin (Lep) at the cell surface. These data are strong evidence that the single nucleotide substitution in the fa allele of Lepr (LeprSa) is responsible for the obese phenotype. Diabetes 45:1141-1143, 1996

O

bese Zucker rats (1) are homozygous for the fatty (fa) mutation in the Lepr (leptin receptor) gene (2). The same gene is mutated in the diabetes (db) mouse (2-4). The phenotypes of fa rats and db mice include severe obesity of early onset, extreme insulin resistance, and strain-specific susceptibility to diabetes (1,5,6). Lepr was originally described as an alternatively spliced single membrane-spanning receptor for leptin (Lep) with homology to the class I cytokine receptor family (7). In the db mouse, a G —» T point mutation generates a donor splice site, resulting in the insertion of a 106 nt exon fragment between Lys889 and Pro890 of the normal Lepr coding seFrom the Laboratory of Human Behavior and Metabolism (S.C.C., X.S.W.-P., S.-M.L, N.O., E.E.K., W.K.C., L.P.-K., M.C., R.L.L.), New York, New York; and Millennium Pharmaceuticals (D.W.W., L.A.T.), Cambridge, Massachusetts. Address correspondence and reprint requests to Dr. Rudolph L. Leibel, Laboratory of Human Behavior and Metabolism, Rockefeller University, 1230 York Ave., New York, NY 10021. E-mail: [email protected]. Received for publication 29 April 1996 and accepted in revised form 22 May 1996. AP, alkaline phosphatase; PCR, polymerase chain reaction; RFLV, restriction fragment length variant; RT, reverse transcription. DIABETES, VOL. 45, AUGUST 1996

quence (3,4). The insertion results in a premature stop codon for the splice variant, which is apparently responsible for signal transduction. The mutation in the dbPas mouse is due to a partial gene duplication (2). The nature of the Lepr mutations in the dbSJ mouse (8) and fak (corpulent) rat (6) are undetermined as yet. This report concerns the molecular basis for the fa mutation. RESEARCH DESIGN AND METHODS Sequence determination of rat Lepr cDNA. Brain and heart RNA were isolated by a modified guanidine HC1 method (9). Ten micrograms of total RNA was used to synthesize cDNA with 200 units of Superscript II Mo-MuLV reverse transcriptase and 100 ng of oligo dT, according to the manufacturer's recommendations. Lepr cDNA fragments were amplified using 2% of the cDNA product with Taq DNA polymerase and primers based on the murine Lepr cDNA sequence. Amplified fragments were purified from agarose gels with DEAE paper and sequenced with Taq DNA polymerase. Genotyping of polymorphisms. Genomic DNA isolation and polymerase chain reaction (PCR) amplification conditions have been previously reported (10). Leptin binding of Lepr cDNAs. A mutant mouse Lepr cDNA containing the Gln269Pro mutation was generated by overlap extension PCR using oligonucleotides containing the specified base changes (11). The construction was verified by sequence analysis. Conditions for transfections of COS7 cells and binding of an alkaline phosphatase (AP)-murine leptin fusion protein have been described (7). RESULTS AND DISCUSSION

A search for mutations within the Lepr gene in fatty if a/fa) rats was initiated with Northern blots. The major 5-kb isoform of Lepr mRNA was not different in size or amount between lean and obese (fa/fa) rats (data not shown). The next step was to determine, using reverse transcription (RT)-PCR, whether the 3.6-kb Lepr mRNA form, which is disrupted in the db mutation (3, 4), was expressed in fa/fa rats. A fragment containing nucleotides 789 to 3549 was amplified from both +/+ and fa/fa brain RNA. These results indicate that the apparent insertion/deletion restriction fragment length polymorphism initially reported for the fa mutation (2) does not disrupt the expression and sequence of Lepr in the fa/fa rat. These results indicate that the most likely mutation was a point mutation that altered the amino acid sequence of the Lepr protein. The entire 3.6-kb form of the Lepr cDNAs from +/+ and fa/fa rats, amplified as two overlapping fragments (nt 43 to 1020 and nt 789 to 3549), were sequenced. Some sequence differences between the wild-type (Brown Norway) and fa 1141

Lepr MUTATION IN fa

Humaim-Mouise-Ratt: P

Y

269 Q

V

K

Y

TAT CAG GTG 880

fatty:

Y

P

V

K

TAT COG GTG

cDNAs in the degenerate third base of codons (which did not result in amino acid substitutions) were seen. However, an A —» C transversion at nt 880 converted aa269 from glutamine (wild type) to proline (fa/fa) (Fig. 1). This difference was confirmed in both amplified cDNA fragments, since the altered base was in the region of overlap between the two fragments. The substitution occurs in a region of the extracellular domain that is conserved between mouse, rat, and human Lepr protein sequences (Fig. 1), suggesting that an evolutionarily conserved function is subserved by this segment. No other sequence variant that altered the Lepr amino acid sequence was observed in the Lepr cDNA of fa/fa rats. The nucleotide substitution generates an Msp I site that can be used as an restriction fragment length variant (RFLV) for determining gene dosage of fa. Using primers that flank the mutation (mObr8 and mObrll), it was found that Zucker rats from two colonies (Charles River and Vassar College)

FIG. 1. Nucleotide and conceptual translation of the fa allele of Lepr. The amino acid sequence of the segments conserved between mouse, rat, and human is shown above the nucleotide sequences for the BN and fa alleles of Lepr. The mutant sequences are shown in italics and underlined. The novel Msp I site is indicated by underlining. Nucleotide and amino acid numbering is according to Tartaglia et al. (7).

were still segregating the novel Msp I mutation (Fig. 2A, lanes d-h). Furthermore, Wistar diabetic fatty (WDF) rats maintained at Vassar College were also segregating the Msp I mutation (Fig. 2A, lanes i-j). Genotyping of rats with recombination events between/a and STRPs (simple tandem repeat polymorphisms) in two previously described crosses, BNxl3M and WKYxl3M (2,10), showed complete concordance of the Msp I site with the obese phenotype (Fig. 2B). To assess any functional consequences of the fa sequence variant, overlap extension PCR (11) was used to introduce the Gln269Pro mutation into the mouse Lepr cDNA (7). The mutant and wild-type cDNAs were transfected into COS cells, and binding to an AP-leptin fusion protein (AP-OB) was tested, The Gln269Pro mutation resulted in nearly a 10-fold reduction in total cell surface binding (Fig. 3A). However, Scatchard transformation of binding data assessed at multiple AP-OB concentrations (Fig. 3B) revealed that the

a b c d e f g h i

D

0 1142

12

FIG. 2. A: detection of the novel Msp I RFLV in Zucker and WDF rats. Genomic DNA was amplified with primers flanking the mutation (mObr8: TATGGAAGTCACAGATGATGG [sense primer] and mObrll: TCTTACAATTGTAGAATTCTC [antisense primer]) to generate a 112-bp product and digested with Msp I. The novel Msp I site in the fa allele is restricted to two fragments of 78 bp and 34 bp, while the wild-type allele bears no Msp I sites. The restricted amplification products were electrophoresed on an agarose gel (3% agarose + 2% low-melting-temperature agarose) and visualized with ethidium bromide staining and ultraviolet illumination. Lane a, BN ( + / + ) ; lane b, lean BNxl3M F l (fa/+); lane c, obese BNxl3M F2 (fa/fa); lanes d-f, lean 13M (+/+); lanes g-h, lean 13M (/a/+); lane i, lean WDF (/a/+); lane j , obese WDF (fa/fa); lane k, no added DNA blank. B: mapping of the Lepr Msp I mutation in obese BNxl3M and WKYxl3M F2 progeny. Genotypes for Lepr—Msp I were performed as in A. a, wild-type allele (BN or WKY); •, fa (13M) allele. The numbers underneath the haplotypes represent the number of chromosomes with the specified haplotype. DIABETES, VOL. 45, AUGUST 1996

S.C. CHUA AND ASSOCIATES

Note added in proof: Two other groups have identified the same missense mutation in Lepr in the fa Zucker rat (13,14).

1501

> 100-