Mice carrying mutations in the fatty liver dystrophy (fld) gene have features of human lipodystrophy1, a genetically heterogeneous group of disorders characterized by loss of body fat, fatty liver, hypertriglyceridemia and insulin resistance2,3,4. Through positional cloning, we have isolated the gene responsible and characterized two independent mutant alleles, fld and fld2J. The gene (Lpin1) encodes a novel nuclear protein which we have named lipin. Consistent with the observed reduction of adipose tissue mass in fld and fld2J mice, wild-type Lpin1 mRNA is expressed at high levels in adipose tissue and is induced during differentiation of 3T3-L1 pre-adipocytes. Our results indicate that lipin is required for normal adipose tissue development, and provide a candidate gene for human lipodystrophy. Lipin defines a novel family of nuclear proteins containing at least three members in mammalian species, and homologs in distantly related organisms from human to yeast.
The initial description of the BALB/cByJ-fld mutant mouse (hereafter fld) focused on two features for which the mutation was named: presence of a triglyceride-filled fatty liver and a progressive neuropathy affecting peripheral nerves5,6,7,8. Subsequently, an independent mutant strain, C3H/HeJ-fld2J (hereafter fld2J), with the same phenotype and a mutation allelic to fld was identified (Mouse Genome Database, http://www.informatics.jax.org). Our further characterization indicates that in addition to fatty liver and neuropathy, fld and fld2J mutants have diminished adipose tissue depots with 50–90% reductions in white and brown fat pad mass (Fig. 1a, and data not shown). Adipocytes in the affected tissue appear immature, with sparse lipid droplets (Fig. 1 b,c). Similar reductions in adipose tissue mass and cellular lipid content have been observed in transgenic mice expressing a transcription factor that interferes with adipocyte differentiation9. In these mice, it has been shown that reduced serum leptin levels resulting from diminished adipose tissue mass is responsible for the development of fatty liver, hypertriglyceridemia and insulin resistance10. Reduced leptin levels were also recently documented in human patients with familial partial lipodystrophy11. Analogously, fld mice have reduced plasma leptin levels ( Fig. 1d) and insulin resistance1, suggesting that the primary defect in fld and fld2J mice is impaired adipose tissue development, with other phenotypic features (that is, fatty liver, hypertriglyceridemia and insulin resistance) occurring as secondary manifestations of the mutation.
We used a positional cloning strategy to isolate the gene responsible for the fld phenotype. Of three genes identified in the fld critical region12, one of these, Lpin1 (formerly Kiaa0188), is mutated in both the fld and fld2J strains. Characterization of Lpin1 revealed a complex rearrangement in the genome of fld mice compared with their wild-type littermates. Southern-blot analysis using exonic DNA probes from the 5′ and 3′ ends of the gene indicated the presence of a deletion and a duplication, respectively (Fig. 2a). PCR-scanning of Lpin1 uncovered additional alterations in fld DNA, with primer sets flanking the deleted and duplicated regions failing to amplify fld DNA ( Fig. 2b). When forward or reverse primers from opposite ends of the gene were paired, however, PCR products were generated from fld DNA, indicating an internal inversion. Based on the integration of these results and additional Southern-blot, PCR amplification and DNA sequence data (data not shown), we determined that the fld mutant allele contains three gross abnormalities (Fig. 2c): (i) the deletion of approximately 2 kb resulting in the elimination of exons 2 and 3, including the translation initiation site; (ii) the inversion of a large genomic region covering more than 40 kb and containing most of the coding sequence; and (iii) duplication of a 0.5-kb segment of the 3′ UTR, with one copy at the position it occurs in the wild-type allele and a second copy carried along with the inverted region. The 5′ and 3′ ends of the fld allele remain intact (Fig. 2c, and data not shown), indicating that the mutation is confined to Lpin1 . The molecular mechanisms leading to the fld allele are unknown, although the coincidence of chromosomal breakpoints at the deletion/inversion and inversion/duplication junctions indicates a concerted event rather than a series of independent steps13. Moreover, because no homologous sequences were identified at the breakpoint regions, we propose a mechanism involving non-homologous recombination.
In contrast to fld, PCR scanning of genomic DNA from the fld2J strain revealed no gross abnormalities in Lpin1 (data not shown). We therefore sequenced Lpin1 cDNA and genomic DNA prepared from tissues from fld2J and wild-type littermates, and detected a single point mutation resulting in a Gly84Arg substitution (Fig. 2d). Gly84 is invariably conserved in homologous sequences from diverse organisms including human, Drosophila melanogaster , Caenorhabditis elegans, Arabidopsis thaliana, Saccharomyces cerevisiae, Schizosaccharomyces pombe and Plasmodium falciparum (Fig. A, see http://genetics.nature.com/supplementary_info/). These data indicate that the nonconservative mutation in the fld2J allele has a functional consequence on lipin activity.
Using northern-blot analysis of total RNA from wild-type mouse tissues, we determined that Lpin1 mRNA is prominently expressed in adipose tissue, skeletal muscle and testis (Fig. 3a, left). Lower expression was also detected in kidney, lung, brain and liver (the latter was detectable only with poly(A)+ RNA). In addition to the approximately 5-kb transcript present in these tissues, we observed an abundant transcript of approximately 3 kb in testis. Because peripheral neuropathy is one of the phenotypic manifestations of the fld mutation, we examined Lpin1 expression in peripheral nerves of wild-type mice by RT–PCR (Fig. 3a, right). Lpin1 RNA was undetectable in sciatic nerve even after extensive PCR amplification (45 cycles), whereas the peripheral nerve marker myelin P2 was readily amplified (40 cycles). This finding indicates that neuropathy in the mutant mice may be a secondary effect of lipin deficiency in other tissues. Tissues from mice homozygous for the fld and fld2J mutations had aberrant Lpin1 mRNA levels. We detected no Lpin1 mRNA in fld adipose tissue, whereas elevated mRNA levels were observed in fld2J adipose tissue (Fig. 3b). Increased expression in fld2J tissue may be the consequence of transcriptional feedback regulation of Lpin1 due to impaired lipin function. Consistent with impaired adipocyte development in fld mice, Lpin1 mRNA expression was induced on differentiation of 3T3-L1 pre-adipocytes (Fig. 3c), coincident with expression of adipsin, a marker of mature adipocytes14.
Lpin1 encodes a novel gene product of 891 amino acids with no similarity to previously characterized proteins or protein domains, although motif searches revealed the presence of a putative nuclear localization signal (NLS; Fig. 4a). Subcellular localization of a lipin–GFP fusion protein revealed that wild-type lipin is predominantly nuclear in HEK293 (Fig. 4b) and 3T3-L1 cells (data not shown). In contrast, mutant lipin2J–GFP protein occurs predominantly in the cytoplasm (Fig. 4b), indicating that the mutation alters the subcellular localization of this protein. Because the Gly84Arg mutation lies outside the NLS, it may affect the function of the nuclear import signal indirectly by altering protein conformation and/or binding of a factor involved in nuclear translocation. Our results imply that nuclear localization is required for normal lipin function.
Through database searches, we identified several mouse and human EST and genomic sequences with similarities to Lpin1. These include two Lpin1-related mouse genes (Lpin2 and Lpin3) and three human homologs (LPIN1, LPIN2 and LPIN3), indicating that Lpin genes form a family in both organisms (Fig. 4a ). The mouse and human genes form three orthologous pairs as determined by sequence comparison and genomic mapping (Fig. 4c). Multiple sequence alignment of lipin-related proteins revealed two strongly conserved regions, which we designate amino-terminal and carboxy-terminal lipin (NLIP and CLIP) domains (Fig. 4a; and Fig. A, see http://genetics.nature.com/supplementary_info/). In addition to mouse and human proteins, we recognized several predicted proteins with NLIP and CLIP domains from a broad range of eukaryotic organisms ( Fig. 4a), indicating that a lipin-like ancestral gene appeared early during evolution.
The identification of lipin has revealed a new factor required for normal adipose tissue development and metabolism. Elucidation of the molecular function of lipin will likely lead to new insights into these processes. The existence of at least two additional Lpin1-related genes suggests that members of this novel protein family may have roles in diverse tissues and cellular processes. The human ortholog of Lpin1, LPIN1, is a potential candidate gene for lipodystrophy, a heterogeneous group of disorders with unknown genetic determinants, except LMNA, responsible for Dunnigan-type familial partial lipodystrophy15,16. Studies showing linkage of the chromosome 2p21 region harboring LPIN1 to fat mass17 and serum leptin levels17,18,19 support this possibility.
We obtained BALB/cByJ-+/fld and C3H/HeJ-+/ fld2J mice from The Jackson Laboratory and bred them to produce the phenotypically mutant (fld/fld, fld2J/ fld2J) and wild-type (+/+ and +/fld, +/fld2J) mice used here. Mice were fed a standard laboratory chow diet (Purina 5001) and maintained in a 14:10-h light-dark cycle. All animal experiments were performed according to guidelines established in the “Guide for the Care and Use of Laboratory Animals.”
Adipose tissue and leptin analysis.
Epididymal fat pads were sectioned (4 μm) and stained with hematoxylin and eosin. For leptin measurements, we obtained blood from the retro-orbital sinus and prepared plasma by centrifugation. Leptin levels were determined with a mouse leptin RIA kit (Linco).
Mouse genomic DNA was prepared using DNAzol (Gibco/BRL). Blots containing restriction-digested DNA (5 μg/lane) were hybridized to a 257-bp 5′ probe corresponding to exons 2 and 3 derived from exon trapping12, or a 225-bp 3′ probe corresponding to part of the 3′ UTR and generated by PCR (5′–TACGCAGGGACACATTTCCA–3′ and 5′–GAGAGATGCAGCTGCGTCA–3′). We hybridized filters at 65 °C in sodium phosphate (0.5 M, pH 7.0), 7% SDS and 1% BSA, and washed at 65 °C to a final stringency of 0.1×SSC/0.1% SDS. Hybridization signals were detected by phosphorimaging.
We prepared total RNA from flash-frozen mouse tissues and 3T3-L1 cells using TRIzol (Gibco/BRL) and loaded 20 μg per lane in 1.2% agarose/formaldehyde gels. Poly(A)+ RNA from liver was prepared using PolyATract reagents (Promega) and 2 μg were analyzed. We performed hybridizations and washes as for Southern blots, except that the temperature was 63 °C. The probe used for northern blots corresponded to nt 1–1,264 of Lpin1 cDNA. The adipsin (Adn) cDNA probe has been described20.
PCR, RACE and DNA sequencing.
PCR amplification of genomic DNA was performed by cycling 1 min at 94 °C, 45 s at 55 °C, 1–2 min at 72 °C for 30–32 cycles. Primers for amplification of the inversion breakpoints were as follows: p1, 5′–CCCTTGAGCACGTTCACA–3′; p2, 5′–CTGATCGTTGTCAGTCTCT–3′; p3, 5′–GGTTGTGGGGACCCTGGA–3′; p4, 5′–GCCTGCTGCAGATGCGTT–3′. RACE cloning of full-length cDNAs for Lpin2 and Lpin3 was performed using brain and 7-day embryo Marathon-Ready cDNA (Clontech), respectively. PCR products were TA-cloned into pCR2.1 (Invitrogen), and sequenced using the Amplicycle sequencing kit (Perkin Elmer) and an ABI model 373A sequencer. For RT–PCR studies, we prepared RNA from pooled sciatic nerve from three wild-type mice, aged three weeks. cDNA was synthesized from total RNA (2 μg) as described5,6,7,8, and amplified with primers known to span one or more introns in Lpin1 (5′–GCCAGGTGTTTGTGACGGTGA–3′, 5′–GCTGGCTTTCCATTCTCGCA–3′) or myelin P2 (5′–GTGAGCACTTCGATGACTACA–3′, 5′–ATCCAGCAGCGTTCTCCTTAT–3′). Products resulting after 45 cycles (lipin, 334 bp) or 40 cycles (myelin P2, 263 bp) were analyzed by agarose gel electrophoresis.
Nuclear localization studies.
The entire coding region of Lpin1 cDNA was amplified from cDNA of wild-type and fld2J mice using the primers 5′–GCTCGAATTCAGACAATGAA TTACGTGGGGCAGCT–3′ and 5′–CGTGCAGTCGACGCTGAGGCTGAATG CATGTCCTGGT–3′, and cloned as an EcoRI/SalI fragment into the pEGFP-N1 vector (Clontech). Cells were transfected using FuGENE 6 transfection reagent (Roche). After 48 h, cells were fixed with 4% paraformaldehyde in PBS and observed with a Zeiss Axiovert confocal fluorescence microscope.
We maintained HEK293 and 3T3-L1 cell lines in DMEM containing 10% fetal bovine serum (basal medium). To induce adipocyte differentiation, 3T3-L1 cells were grown to confluence (day 0) and then cultured for 3 d in differentiation medium consisting of basal medium plus insulin (10 μg/ml), dexamethasone (2 μg/ml) and methylisobutylxanthine21 (0.5 mM). On days 4–7, cells were maintained in basal medium with insulin.
Radiation hybrid mapping.
We mapped Lpin2 and Lpin3 using a mouse-hamster radiation hybrid panel (Research Genetics)22. Oligonucleotide primer pairs derived from the 3′ UTR of each gene were as follows: Lpin2, 5′–GGCGAGACCCAATCCCTGA–3′ and 5′–GGGTCTTCCTCTGTAAGA–3′; Lpin3, 5′–CCTGGCTTGAGCTTGCCTT–3′ and 5′–CCCACGGCATGCATCTTCT–3′.
GenBank accession numbers.
Lpin1, AF180471; Lpin2, AF286723; Lpin3, AF286724; LPIN1, Q14693; LPIN2 , Q92539; LPIN3, AL132654; Drosophila homolog, AAF59125; C. elegans homolog, CAA16154; S. cerevisiae homolog, P32567; S. pombe homolog, CAB52577; A. thaliana homolog, AAF23287; P. falciparum homolog, CAB10579.
Note: Supplementary information is available on the Nature Genetics web site (http://genetics.nature.com/supplementary_info/).
Reue, K., Xu, P., Wang, X.-P. & Slavin, B.G. Adipose tissue deficiency, glucose intolerance, and increased atherosclerosis result from mutation in the mouse fatty liver dystrophy ( fld ) gene. J. Lipid Res. 41, 1067–1076 (2000).
Seip, M. & Trygstad, O. Generalized lipodystrophy, congenital and acquired (lipoatrophy). Acta Paediatr. Scand. Suppl. 413, 2–28 (1996).
Senior, B. & Gellis, S.S. The syndrome of total lipodystrophy and partial lipodystrophy. Pediatrics 33, 593–612 (1964).
Dörfler, H., Rauh, G. & Bassermann, P. Lipoatrophic diabetes. Clin. Invest. 71, 264–269 (1993).
Langner, C.A. et al. The fatty liver dystrophy ( fld ) mutation. A new mutant mouse with a developmental anormality in triglyceride metabolism and associated tissue-specific defects in lipoprotein lipase and hepatic lipase activities . J. Biol. Chem. 264, 7994– 8003 (1989).
Langner, C.A., Birkenmeier, E.H., Roth, K.A., Bronson, R.T. & Gordon, J.I. Characterization of the peripheral neuropathy in neonatal and adult mice that are homozygous for the fatty liver dystrophy ( fld ) mutation. J. Biol. Chem. 266 , 11955–11964 (1991).
Rehnmark, S., Giometti, C.S., Slavin, B.G., Doolittle, M.H. & Reue, K. The fatty liver dystrophy mouse: microvesicular steatosis associated with reduced fatty acid oxidation and altered expression levels of peroxisome proliferator-regulated proteins. J. Lipid Res. 39, 2209–2217 ( 1998).
Klingenspor, M., Xu, P., Cohen, R.D., Welch, C. & Reue, K. Altered gene expression pattern in the fatty liver dystrophy mouse reveals impaired insulin-mediated cytoskeleton dynamics. J. Biol. Chem. 274, 23078–23084 ( 1999).
Shimomura, I. et al. Insulin resistance and diabetes mellitus in transgenic mice expressing nuclear SREBP-1c in adipose tissue: model for congenital generalized lipodystrophy. Genes Dev. 12, 3182– 3194 (1998).
Shimomura, I., Hammer, R.E., Ikemoto, S., Brown, M.S. & Goldstein, J.L. Leptin reverses insulin resistance and diabetes mellitus in mice with congenital lipodystrophy. Science 401, 73–76 ( 1999).
Hegele, R.A., Cao, H., Huff, M.W. & Anderson, C.M. LMNA R482Q mutation in partial lipodystrophy associated with reduced plasma leptin concentration . J. Clin. Endocrinol. Metab. 85, 3089– 3893 (2000).
Péterfy, M., Phan, J., Oswell, G.M., Xu, P. & Reue, K. Genetic, physical and transcript map of the fld region on mouse chromosome 12. Genomics 62, 436–444 (1999).
Okano, H., Aruga, J., Nakagawa, T., Shiota, C. & Mikoshiba, K. Myelin basic protein gene and the function of antisense RNA in its repression in myelin-deficient mutant mouse. J. Neurochem. 56, 560–567 ( 1991).
Ailhaud, G., Grimaldi, P. & Negrel, R. Cellular and molecular aspects of adipose tissue development . Annu. Rev. Nutr. 12, 207– 233 (1992).
Cao, H. & Hegele, R.A. Nuclear lamin A/C R482Q mutation in Canadian kindreds with Dunnigan-type familial partial lipodystrophy. Hum. Mol. Genet. 9, 109–112 (2000).
Shackleton, S. et al. LMNA, encoding lamin A/C, is mutated in partial lipodystrophy . Nature Genet. 24, 153– 156 (2000).
Comuzzie, A.G. et al. A major quantitative trait locus determining serum leptin levels and fat mass is located on human chromosome 2. Nature Genet. 15, 273–275 ( 1997).
Rotimi, C.N. et al. The quantitative trait locus on chromosome 2 for serum leptin levels is confirmed in African-Americans. Diabetes 48, 643–644 (1999).
Hager, J. et al. A genome-wide scan for human obesity genes reveals a major susceptibility locus on chromosome 10. Nature Genet. 20, 304–308 (1998).
Spiegelman, B.M., Frank, M. & Green, H. Molecular cloning of mRNA from 3T3 adipocytes. Regulation of mRNA content for glycerophosphate dehydrogenase and other differentiation-dependent proteins during adipocyte development. J. Biol. Chem. 258, 10083–10089 (1983).
Student, A.K., Hsu, R.Y. & Lane, M.D. Induction of fatty acid synthetase synthesis in differentiating 3T3-L1 preadipocytes. J. Biol. Chem. 255, 4745–4750 (1980).
McCarthy, L.C. et al. A first-generation whole genome-radiation hybrid map spanning the mouse genome. Genome Res. 7, 1153– 1161 (1997).
Thompson, J.D., Higgins, D.G. & Gibson, T.J. CLUSTALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680 (1994).
Perriere, G. & Gouy, M. WWW-query: an on-line retrieval system for biological sequence banks. Biochimie 78, 364– 369 (1996).
Rowe, L.B., Sweet, H.O., Gordon, J.I. & Birkenmeier, E.H. The fld mutation maps near to but distinct from the Apob locus on mouse Chromosome 12. Mamm. Genome 7, 555–557 (1996).
Deloukas, P. et al. A physical map of 30,000 human genes. Science 282, 744–746 (1998).
We thank L. Rowe and the late E. Birkenmeier for a set of (BALB/c-fld×CAST)F2 DNA samples; T. Beischlag for the Arnt-GFP construct; B. Slavin for adipose tissue sections; Q. Han for assistance in the cell culture; and D. Scott for assistance with fluorescence microscopy. This work was supported by grant HL24841 from the National Institutes of Health.
Alignments of the NLIP and CLIP domains showing positions of residues that are identical (black background) or similar (gray background) in at least 90% of the sequences. The asterisk above the NLIP sequence denotes the position of the Gly84Arg mutation in fld2J mice. (PPT 392 kb)
About this article
Cite this article
Péterfy, M., Phan, J., Xu, P. et al. Lipodystrophy in the fld mouse results from mutation of a new gene encoding a nuclear protein, lipin. Nat Genet 27, 121–124 (2001). https://doi.org/10.1038/83685
Neuromuscular Disorders (2020)
Variation in the Lipin 1 Gene Is Associated with Birth Weight and Selected Carcass Traits in New Zealand Romney Sheep
Phosphatidic acid metabolism regulates neuroendocrine secretion but is not under the direct control of lipins
IUBMB Life (2020)
Substrate channeling in the glycerol-3-phosphate pathway regulates the synthesis, storage and secretion of glycerolipids
Biochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids (2020)
Biochemical Journal (2020)