| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Departments of Clinical Biochemistry, Medicine and Surgery, University of Cambridge, Addenbrooke’s Hospital, Hills Road, Cambridge, CB2 2QQ, United Kingdom
Address all correspondence and requests for reprints to: Dr. Antonio Vidal-Puig, Department of Clinical Biochemistry, University of Cambridge, Box 232, Level 4, Addenbrooke’s Hospital, Hills Road, Cambridge, United Kingdom CB2 2QR. E-mail: ajv22{at}hermes.cam.ac.uk
Abstract
Familial partial lipodystrophy-Dunnigan variety (FPLD) is an autosomal dominant form of lipodystrophy resulting in a loss of sc fat from the trunk and limbs with retention of fat in the visceral depots, face, and neck. Specific point mutations in the gene encoding the nuclear lamina proteins, lamins A and C, have been established to cause this syndrome. We undertook studies to determine which members of the lamin family were expressed in human fat cells, to examine the effect of differentiation state on lamin A and C expression in human preadipocytes, and to test the hypothesis that regional variation in lamin A/C expression might underlie the stereotyped anatomical pattern of FPLD. Lamins A, C, and B1, but not B2, were expressed in sc mature human adipocytes. Subcutaneous preadipocytes expressed all four lamins, with lamin A and C expression increasing with ex vivo differentiation. Consistent with previously reported resistance to ex vivo differentiation, omental preadipocytes did not show an increase in lamin A or C mRNA under these conditions. Lamin A/C mRNA levels were similar in isolated mature adipocytes and preadipocytes from omental, sc, and neck sites. However, lamin C was consistently lower, and the ratio of lamin A/C mRNA was higher in sc mature adipocytes compared with omental mature adipocytes. We conclude that the depot-specific pattern of lamin A/C expression does not provide clues to the mechanism of FPLD. Nonetheless, these studies provide new information regarding the expression of lamin isoforms in normal human adipose cells, which will inform future studies of the molecular pathogenesis of FPLD.
DUNNIGAN-TYPE familial partial lipodystrophy (FPLD; OMIM 151660) is an autosomal dominant disorder characterized by a progressive loss of sc adipose tissue from the trunk, gluteal region, and extremities. In contrast with these sites, adipose tissue in intraabdominal sites is retained, and there is usually an excessive accumulation of adipose tissue on the face and neck (1, 2, 3). The disorder is associated with severe insulin resistance and hypertriglyceridemia (4). Several groups have recently reported that mutations in the nuclear lamina protein lamin A/C are the cause of this syndrome (5, 6, 7, 8).
Lamins are intermediate filament proteins that lie on the nucleoplasmic side of the nuclear envelope (9). Although originally thought to be purely structural in function, other proposed roles of lamin proteins include the maintenance of nuclear stability, the regulation of nuclear import/export, and the regulation of DNA and RNA synthesis. The lamins are usually classified into two forms (A type and B type) with distinct biochemical and structural properties (10). In humans, two genes encoding B-type lamins have been described (11, 12), and a single gene, LMNA, encodes A-type lamins (13). LMNA encodes at least three splice isoforms of A-type lamins, lamin A, lamin C (13), and lamin A
10 (14). Lamin A is initially synthesized as an immature polypeptide of 664 residues. Lamin C differs due to an alternative splice site in exon 10, which results in the addition of five lamin C-specific amino acids before an early termination (13). Lamin A10 is identical to lamin A, except that it lacks the region coded by exon 10 (14). The lamin A protein comprises an N-terminal globular domain, a central 
-helical domain, and a globular C-terminal domain that contains both a nuclear localization signal and an isoprenylation motif in its last four residues (for review, see Ref. 10). All FPLD-associated mutations occur in the C-terminal globular domain of the protein, either at exon 8 (5, 6, 7, 8, 15), which is common to both lamin A and lamin C, or at exon 11 (8, 15), which is specific to lamin A. Mutations at other sites in LMNA are responsible for at least three other distinct inherited diseases that are not characterized by abnormalities of fat (16, 17, 18, 19, 20, 21, 22).
Key questions raised by the discovery that specific lamin A/C mutations result in FPLD include 1) how can genetic variants in a widely expressed nuclear lamina protein result in a phenotype largely restricted to adipose tissue; and b) what is the basis for the highly stereotyped anatomical pattern of fat loss seen in FPLD? As a step toward addressing the first question, we believed it was important to determine which forms of lamin are expressed in human preadipocytes and adipocytes and what, if any, alterations in lamin expression might occur during adipocyte differentiation. In response to the second question we hypothesized that different fat depots might express different amounts of lamin isoforms in either adipocytes or preadipocytes and that this might underlie regional differences in the susceptibility of certain fat depots to the presence of a mutant lamin. To test this hypothesis we have examined lamin isoform expression in human adipose cells from depots that are differentially affected in FPLD.
Materials and Methods
Acquisition of human tissue
Omental (OM), sc, and neck (NK) adipose tissue biopsies were obtained from patients undergoing elective open surgery, typically total abdominal hysterectomies for omental and sc samples and thyroid surgery for neck samples. A total of 10 patients (9 female and 1 male) contributed sc and OM samples, and 5 patients (all female) gave NK samples. None of the patients had diabetes, hyperthyroidism, or severe systemic illness. Cambridge local research ethics committee approval was obtained, and all patients gave their informed consent.
Preadipocyte and adipocyte cell isolation
Unless stated otherwise, all reagents were obtained from Sigma (Poole, UK). Whole adipose tissue was taken under sterile conditions during surgery, and the delay between the biopsy of the tissue and arriving in the laboratory was less than 4 h. All OM samples used in experiments were paired with the sc sample from the same patient and vice versa. NK had to be obtained separately due to surgical procedures. Mature adipocytes were isolated as described previously (23). After the adipocytes were removed, preadipocytes were isolated by resuspending the cell pellet in 2 ml lysis buffer (0.15 M NH4Cl, 10 mM KHCO3, and 0.1 M EDTA). This was incubated for 2 min at room temperature and then centrifuged at 1500 x g for 5 min at 4 C. The cell pellet was resuspended in serum-containing medium (SCM), placed into a 25-cm2 tissue culture flask, and incubated at 37 C under humidified air containing 5% CO2.
Preadipocyte cell culture media and protocols
SCM consisted of DMEM nutrient mixture/Ham’s F-12 containing 2 mM L-glutamine, 10% FBS, 0.1 mg/ml streptomycin, and 100 U/ml penicillin. Differentiation medium was DMEM nutrient mixture/Ham’s F-12 containing 33 µM biotin, 17 µM pantothenic acid, 10 µl/ml transferrin, 0.2 nM T3, 0.1 µM cortisol, 0.5 µM bovine insulin, 0.1 mg/ml streptomycin, 100 U/ml penicillin, and 1 µl/ml dimethylsulfoxide. Modified differentiation medium (MDM) was as described above, except containing 10-7 M rosiglitazone (SmithKline Beecham, Harlow, UK) and 10-7 M LG100268 (Ligand Pharmaceuticals, Inc., San Diego, CA) given in dimethylsulfoxide to a final concentration of 1 µl/ml.
Preadipocyte cells were given SCM every 2–3 d until 1 wk after reaching confluence. At this point (d 0) flasks were given differentiation medium or MDM medium. For the first 2 d of differentiation the medium also contained 250 µM isobutylmethylxanthine. Cells were taken for analysis on d 5, 10, and 15 after the introduction of the differentiation or MDM.
RNA isolation
Total RNA was isolated using the RNeasy system (QIAGEN, Crawley, UK) following the manufacturer’s protocol, quantified spectroscopically at 260 nm using a GeneQuant Nucleotide calculator (Amersham Pharmacia Biotech, Uppsala, Sweden), and checked for integrity on a 1% Tris-Borate-EDTA gel using ethidium bromide staining. RNA was then snap-frozen and stored at -80 C until needed.
RNA analysis
For specific probe templates a cDNA probe for lamin A and lamin C was generated from skeletal muscle cDNA (CLONTECH Laboratories, Inc., Palo Alto, CA) as follows. A region matching bases 1482–1819 of the lamin A cDNA (GenBank accession no. L12401) was amplified by PCR (Table 1
). This probe covers a region of 217 bases common to both lamin A and lamin C with an extra 121 bases that are lamin A specific. Therefore, this probe allows the detection of lamin A and lamin C as protected bands of 338 and 217 bases, respectively. PCR products for lamin B1, lamin B2, and emerin were also amplified (Table 1). The PCR products were ligated into the pGEM-T Easy vector (Promega Corp., Cambridge, UK) and used to transform JM109 Supercompetant cells (Promega Corp.). The orientation of the ligation was verified by direct sequencing. To correct for RNA loading, a pre-made template for human cyclophilin was used, which protects a region of 109 bases (Ambion, Inc., Austin, TX).
|
The template for the antisense probe was prepared by linearizing the plasmid containing the lamin insert with the restriction endonuclease SpeI (for lamin A/C and lamin B2) or NcoI (for lamin B1 and emerin). These reactions were then treated with proteinase K at a final concentration of 200 µg/ml. The templates were purified using phenol/chloroform extraction and ethanol precipitation. Antisense probes and reference probes were transcribed from the linearized template DNA using bacteriophage T7 or Sp6 RNA polymerases (Stratagene, La Jolla, CA) using previously published protocols (24, 25). Protected bands were visualized by autoradiography and quantified by phosphorimager analysis using MacBas V2.2 software (Fuji Photo Film Co., Ltd., Tokyo, Japan).
Quantitation of lamin A/C and lamin C mRNA expression by real-time semiquantitative RT-PCR (Taqman)
Total RNA was isolated from mature adipocytes as described above, and then 100 ng were reverse transcribed for 1 h at 37 C in a 20-µl reaction containing 1 x RT buffer (50 mM Tris-HCl, 75 mM KCl, 3 mM MgCl2, and 10 mM dithiothreitol), 100 ng random hexamers, 1 mM dNTPs, and 100U Moloney murine leukemia virus reverse transcriptase (Promega Corp.). A reaction containing 500 ng skeletal muscle total RNA was also included as a standard. After first strand cDNA synthesis, this standard was serially diluted 1:2 in deoxyribonuclease-free water to generate a standard curve for the PCR analysis.
Oligonucleotide primers and Taqman probe were designed using Primer Express, version 1.0 (Perkin-Elmer Corp. PE Applied Biosystems, Foster City, CA) and sequences from the GenBank database. Two sets of probe and primer were designed (Table 1
). For total lamin A/C mRNA, the region amplified was common to both lamin A and lamin C. For the lamin C-specific set, the PCR product overlapped the region of alternate splicing for lamin C, so that the forward primer was common to lamin A and lamin C, whereas the reverse primer was lamin C specific. The lamin C probe overlapped the splice site (see Table 2 for sequences). The Taqman probes were labeled at the 5'-end with the reporter dye 6-carboxy-fluorescein and at the 3'-end with the quencher 6-carboxy-tetramethyl-rhodamine. Oligonucleotide primers and Taqman probes for GAPDH, the internal control, were purchased from Perkin-Elmer Corp.
PCR was carried out in quadruplicate for each sample on an ABI 7700 sequence detection system (Perkin-Elmer Corp.). Each 25-µl reaction contained 2 µl first strand cDNA, 1 x PCR master mix, 300 nM forward and 300 nM reverse primers, and 150 nM Taqman probe. All reactions were carried out using the following cycling parameters: 50 C for 2 min, 95 C for 10 min, followed by 40 cycles of 95 C for 15 sec, 60 C for 1 min. For relative quantitation of the samples, we used the comparative threshold cycle (Ct) method (separate tubes). A validation is performed to ensure that two targets (e.g. lamin A/C and glyceraldehyde-3-phosphate dehydrogenase) are amplified with the same efficiency. One of the targets in the pair can then be used as a calibrator from which to standardize the other target, so readings from different samples can be compared.
Western blotting
Total cell protein was extracted in 300 µl RIPA buffer [1 x PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% SDS containing Complete, Mini Protease Inhibitor cocktail (Roche Molecular Biochemicals, Mannheim, Germany)]. Cells were then incubated on ice for 30 min, vortexing periodically. The lysed cells were centrifuged at 12,000 rpm for 30 min at 4 C, and the supernatant was transferred to a new tube. The total protein concentration was measured at 595 nm using a Coomassie blue reagent kit (Pierce Chemical Co., Rockford, IL) with BSA as a standard. Protein samples were mixed with 5 x loading buffer, boiled for 5 min, and then separated by SDS-PAGE. Proteins were then electroblotted onto Immobilon-P polyvinyldifluoride membranes (Millipore Corp., Bedford, MA). The membranes were then washed in TBS-Tween (Tris-buffered saline with 0.05% Tween 20) and blocked in a solution of TBS-Tween containing 5% powdered milk. Lamin A and lamin C proteins were detected with mouse monoclonal antihuman primary antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) diluted 1:250. For aP2, a rabbit antihuman antibody was used at a 1:10,000 dilution (gift from Dr. D. Bernlohr, University of Minnesota, St. Paul, MN). An appropriate horseradish peroxidase-conjugated secondary antibody (DAKO Corp., Glostrup, Denmark) was used, and bands were visualized by enhanced chemiluminescence (Amersham Pharmacia Biotech).
Statistical analysis
ANOVA post-hoc and paired t tests were carried out on RPA and Western blotting data using StatView version 4.5 (Abacus Concepts, Inc., Berkeley, CA). All data are presented as the mean ± SE.
Results
Expression of lamins in human adipose cells
RPAs specific for lamin A, lamin C, lamin B1, and lamin B2 were developed and used to examine gene expression in human preadipocytes and adipocytes. mRNAs encoding all four lamins were detectable in preadipocytes. After normalizing to cyclophilin levels, the order of mRNA abundance was lamin B1 > A = C > B2. In isolated human adipocytes lamin B2 mRNA was undetectable (Fig. 1), but the other three lamin mRNAs were all present in the order B1 > A = C. mRNA encoding the lamin-interacting protein, emerin, was also expressed in adipocytes and preadipocytes (data not shown).
|
RPAs specific for each isoform were used to examine whether preadipocyte differentiation was associated with changes in lamin A or C mRNA levels. To maximize differentiation, cells were treated with the conventional differentiation cocktail to which rosiglitazone and LG100268, agonists of PPAR
and RXR (26, 27), respectively, were added. The behavior of preadipocytes obtained from sc, OM, and NK sites were compared. Lamin A and lamin C mRNA levels in preadipocytes derived from sc, OM, and NK depots and grown to confluence were quantified. The lamin A/C ratio was close to unity for preadipocytes from all three sites (0.82 ± 0.03 vs. 0.97 ± 0.07 vs. 0.97 ± 0.06 for NK, OM, and sc, respectively; P = NS). No significant difference in either lamin A or C mRNA expression was seen between preadipocytes from the three sites on d 0 of differentiation (Fig. 2A).
|
, top panels) and C (Fig. 2B, bottom panels), respectively, which were maximal on d 10. In NK preadipocytes, differentiation produced a similar, but less pronounced, trend. Consistent with previous observations (28), differentiation of OM preadipocytes was very limited, as assessed by morphology and aP2 protein expression (data not shown). Neither lamin A nor C mRNA changed in abundance over the 15-d differentiation protocol in these cells (Fig. 2B). The ratio of lamin A to lamin C mRNA remained at approximately 1:1 during differentiation of preadipocytes from all depots.
We explored whether alterations in the mRNA level were associated with changes in the expression of lamin A and C proteins. Cell lysates from preadipocytes at different stages of differentiation were immunoblotted with an antibody that recognizes both lamin A and lamin C, allowing measurement of total lamin A/C (Fig. 3A). Consistent with the increase in lamin A and C mRNA in preadipocytes derived from these regions, by 10 d of differentiation lamin A/C protein increased approximately 8- and 3.5-fold in sc and NK preadipocytes, respectively (Fig. 3B). Surprisingly, given the apparent lack of effect of the differentiation mix on lamin mRNA expression in OM preadipocytes, lamin A/C protein was increased by approximately 2.5-fold in these cells.
|
Lamin A mRNA levels were similar between OM and sc isolated mature adipocytes (Fig. 4A). In contrast, lamin C mRNA was considerably higher in OM than sc adipocytes. Overall, the lamin A/lamin C ratio was higher in the sc than the OM depot (OM vs. sc, 1.213 ± 0.109 vs. 2.708 ± 0.420; P = 0.0064; Fig. 4B).
|
). There were no differences in the levels of lamin A or lamin C protein between sc and OM depots (Fig. 4C). In all samples examined, lamin C protein was the more abundant isoform, being present at approximately twice the level of lamin A (ratio of lamin C/lamin A, sc vs. OM, 2.340 ± 0.200 vs. 2.126 ± 0.334; P = NS). This contrasts with mRNA abundance, suggesting that lamin C mRNA may be more efficiently translated into protein than lamin A.
Real-time semiquantitative RT-PCR was designed to be able to measure lamin A/C and C isoforms in adipocytes obtained from neck biopsies compared with OM and sc. The RT-PCR technique provided independent conformation of the higher level of lamin C expression in OM vs. sc adipocytes. Levels of lamin A/C mRNA were similar among NK, OM, and sc adipocytes. For lamin C, values in NK adipocytes were closer to those in sc than OM adipocytes (Fig. 4D).
Discussion
The causative gene for the syndrome FPLD has recently been found to be LMNA (5, 6, 7), the gene encoding A-type nuclear lamins. Interestingly, mutations at other points of this gene are thought to be responsible for three different disorders, conduction system disease (CMD1A; OMIM 115200) (17, 22, 29), the autosomal dominant form of Emery-Dreifuss muscular dystrophy (EDMD2; OMIM 150330 and 181350) (16, 18, 20), and limb girdle muscular dystrophy type 1B (LGMD1B; OMIM 119001) (19). In addition, an X-linked version of Emery-Dreifuss muscular dystrophy (EDMD1; OMIM 310300) is caused by mutations in another nuclear matrix protein, emerin, a protein known to interact with lamin A (30).
Although lamins appear to be necessary for cell survival and are expressed in the early embryo (31), the precise constituents of the lamina vary somewhat between cell types (32, 33). As lamins may have a role in the control of transcription (34, 35, 36, 37, 38, 39), the composition of the adipocyte nuclear lamina may impinge on cell-specific gene expression. We found that lamin A, lamin C, lamin B1, and emerin mRNAs are all highly expressed in confluent preadipocytes. Lamin B2 was only weakly expressed in preadipocytes and was not seen in mature adipocytes. This adipocyte nuclear lamina repertoire contrasts with other cells, such as skeletal muscle and cardiomyocytes, where lamin B2 is the prevalent B-type lamin, and lamin B1 is absent (40). This new information regarding the range of lamins expressed in human fat cells provides an essential framework for attempts to understand the pathophysiology of FPLD.
As one possible mechanism underlying lipodystrophy involves the failure of preadipocyte differentiation, we explored the effects of differentiation on the expression of lamin isoforms in human preadipocytes. In sc preadipocytes both lamin A/C mRNA and protein were markedly increased during 10 d of differentiation. As cell number does not increase during this time, this result suggests that the adipocyte differentiation process per se is likely to involve the accumulation of lamin A/C within each cell. FPLD is a dominantly inherited process, but mice heterozygous for an LMNA deletion do not appear to develop lipodystrophy (41). Thus, it appears likely that the mutant lamins cause FPLD either through a dominant negative interference with the function of normal lamin or through a toxic effect of the mutant lamin. If the latter mechanism is operating, it is tempting to speculate that an increase in mutant lamin expression that occurs during the differentiation process might arrest this process and be centrally involved in the development of lipodystrophy. Formal proof of this concept will require examination of the effects of mutant lamins expressed in adipocytes. Of note, in this regard, preadipocytes from the omental depot, which is spared in FPLD, show much lower increases in lamin A/C mRNA and protein when exposed to the same differentiation cocktail. However, this may simply reflect the previously described generalized resistance of cells from this depot to differentiation ex vivo.
The pattern of fat loss in FPLD is remarkably stereotyped. Intraabdominal adipose tissue is preserved, adipose tissue is increased on the neck and face, and sc adipose tissue elsewhere is lost. In this regard there is a growing body of literature attesting to the presence of consistent intrinsic differences in the biochemical properties (42), gene expression profile (43, 44), and cell biology (45) of fat cells from particular anatomical depots. For example, we and others have reported that leptin gene expression is markedly lower in adipocytes derived from OM vs. sc depots (44, 46, 47). Also, as mentioned above, preadipocytes from intraabdominal sites respond poorly to the prodifferentiation actions of PPAR agonists (28). Although there is widespread acceptance of the major differences between sc and intraabdominal adipocytes, there is also a growing literature indicating functional differences between adipocytes from different sc areas (48, 49). The recognition of different patterns of adipose loss in patients diagnosed with familial partial lipodystrophy syndrome has suggested the possibility of different varieties. Indeed, whereas in the Dunnigan variety fat is mainly lost from both the extremities and trunk, in the Kobberling variety the loss of adipose tissue is restricted to the extremities, with normal amounts of sc fat elsewhere (50). Thus, future studies should evaluate potential differences in lamin A/C gene expression among these depots.
We tested the hypothesis that the peculiar pattern of lipodystrophy seen in FPLD might relate to natural differences in the extent of lamin A/C expression between adipose depots. Although the ratio of lamin A/C mRNA differed consistently between omental and sc adipocytes, this difference was not found at the protein level. Additionally, the pattern of lamin A/C RNA expression in adipocytes from the neck was more similar to sc than OM adipocytes. Thus, we conclude that the anatomical pattern of lipodystrophy in FPLD is unlikely to simply reflect regional variation in the expression of the lamin A/C gene itself. We therefore hypothesize that a protein(s) that specifically interacts with the mutant lamin is expressed in a regionally determined manner. Attempts to identify such proteins are ongoing.
In summary, we have described, for the first time, the repertoire of lamins expressed in human preadipocytes and adipocytes and have demonstrated that lamin A/C expression increases with preadipocyte differentiation. Although we found consistent site-related differences in the expression of lamin A/C RNAs, these are unlikely to account for the pattern of fat loss seen in FPLD. Nevertheless, the information obtained from these studies will be invaluable for future studies attempting to explain the relationship between specific mutations in lamin A/C and FPLD.
Acknowledgments
We thank all of the patients and surgeons at Addenbrooke’s Hospital (Cambridge, UK) who helped to make this project possible. We also thank Dr. D. Bernlohr of the University of Minnesota for providing the aP2 antibody.
Footnotes
This work was supported by grants from the Medical Research Council of the United Kingdom (to C.J.L.), Deutsche Forschungsgemeinschaft (to D.B.), and the Wellcome Trust (to S.O. and A.V.P.).
Abbreviations: CMD1A, Conduction system disease; EDMD, Emery-Dreifuss muscular dystrophy; FPLD, familial partial lipodystrophy (Dunnigan variety); LA, lamin A; LC lamin C; LGMD1B, limb girdle muscular dystrophy type 1B; LMNA, human gene encoding lamin A/C; MDM, modified differentiation medium; NK, neck; OM, omental; RPA, ribonuclease protection assay; SCM, serum-containing medium.
Received June 20, 2001.
Accepted November 5, 2001.
References
release from human adipose tissue in vitro. J Endocrinol 163:33–38[Abstract]
2 promoter. Biochem Biophys Res Commun 240:99–103[CrossRef][Medline]
and stimulates adipogenesis. Genes Dev 9:2350–2363 have depot-specific effects on human preadipocyte differentiation. J Clin Invest 100:3149–3153[Medline]
and transforming growth factor-ß1. Eur J Endocrinol 141:436–442[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  G. Duque As a Matter of Fat: New Perspectives on the Understanding of Age-Related Bone Loss IBMS BoneKEy, April 1, 2007; 4(4): 129 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Lelliott, M. Lopez, R. K. Curtis, N. Parker, M. Laudes, G. Yeo, M. Jimenez-Linan, J. Grosse, A. K. Saha, D. Wiggins, et al. Transcript and metabolite analysis of the effects of tamoxifen in rat liver reveals inhibition of fatty acid synthesis in the presence of hepatic steatosis FASEB J, July 1, 2005; 19(9): 1108 - 1119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. B. Savage, P. R. Murgatroyd, V. K. Chatterjee, and S. O'Rahilly Energy Expenditure and Adaptive Responses to an Acute Hypercaloric Fat Load in Humans with Lipodystrophy J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1446 - 1452. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Garg Acquired and Inherited Lipodystrophies N. Engl. J. Med., March 18, 2004; 350(12): 1220 - 1234. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
