| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology, Diabetes and Nutrition (M.F., F.S.C., A.R.S.), University of Maryland School of Medicine, and Geriatrics Research and Education Clinical Center (A.R.S.), Baltimore Veterans Administration Medical Center, Baltimore, Maryland 21201; Division of Endocrinology (R.K.), Cook County Hospital, Chicago, Illinois 60612; Center of Health Sciences (M.d.F.P.B., M.G.D.N.S., J.B.-N.), Federal University of Rio Grande do Norte, Natal, Rio Grande do Norte, Brazil 59010-180; and Endocrine Service (S.V., B.L.W.), Hospital das Clinicas, University of Sao Paulo, Brazil 05403-900
Address all correspondence and requests for reprints to: Mao Fu, Division of Endocrinology, Diabetes and Nutrition, University of Maryland School of Medicine, 660 West Redwood Street, Baltimore, Maryland 21201. E-mail: mfu{at}medicine.umaryland.edu.
| Abstract |
|---|
|
|
|---|
C, and C636A (Phe189X). In three siblings with Brunzell syndrome, we identified a splice site mutation (IVS42A
G) in AGPAT2, showing that AGPAT2 mutations can also cause Brunzell syndrome. Eighteen CGL patients from 15 families from the same region of northeastern Brazil were homozygous for a frameshift mutation (669insA of AF05149) in Gng3lg. Despite having the same mutation, the subjects had widely divergent clinical manifestations. In our subjects, there did not appear to be any distinguishing clinical characteristics between CGL subjects with AGPAT2 or Gng3lg mutations with the exception of mental retardation in carriers of Gng3lg. In summary, mutations in AGPAT2 and Gng3lg are approximately equally represented in CGL; despite harboring the same Gng3lg mutation, subjects may have widely divergent clinical manifestations, suggesting modifying influences of other genes and/or environment; and Brunzell syndrome may be caused by a mutation in AGPAT2. | Introduction |
|---|
|
|
|---|
Two loci (BSCL1 and BSCL2) linked to CGL have been mapped to chromosomes 9q34 and 11q13, respectively (9, 10). Positional cloning of BSCL1 revealed mutations in AGPAT2, which encodes the 1-acylglycerol-3-phosphate O-acyltransferase 2 (11). This 278-amino-acid protein belongs to the family of acyltransferases and catalyzes the acylation of lysophosphatidic acid to form phosphatidic acid, a key intermediate in the biosynthesis of triacylglycerol and glycerophospholipids (12). Positional cloning of BSCL2 by Magre et al. (10) disclosed mutations in Gng3lg (also named seipin), which is homologous to the murine guanine nucleotide-binding protein (G protein)
3-subunit-linked gene, a gene of unknown function.
Despite the identification of these genes, several questions remain. What is the relative prevalence of AGPAT2 and Gng3lg mutations in subjects with CGL? Are there any distinguishing clinical characteristics between CGL subjects with mutations in AGPAT2 compared with Gng3lg? How consistent is the phenotype in CGL subjects with the same mutation? To begin to answer these questions, we studied these two genes in CGL subjects from 26 families. Our findings suggest that mutations in AGPAT2 and Gng3lg are approximately equally represented in CGL. There do not appear to be any obvious distinguishing phenotypic characteristics in subjects with mutations in one gene or the other with the exception that mental retardation appears to be associated with AGPAT2 mutations but not Gng3lg mutations. Furthermore, despite harboring the same Gng3lg mutation, subjects may have widely divergent clinical manifestations, suggesting modifying influences of other genes and/or environment. Finally, we identified a mutation in AGPAT2 in three siblings with Brunzell syndrome (OMIM no. 272500) (13), a related syndrome characterized by generalized lipodystrophy and systemic angiomatosis. These findings implicate mutations in AGPAT2 as a cause of Brunzell syndrome.
| Subjects and Methods |
|---|
|
|
|---|
We studied 30 affected CGL subjects from 26 families (CGL-F1 to CGL-F26) as well as three affected siblings from a family (B-F1) with Brunzell syndrome. The 18 affected members of families CGL-F1 through CGL-F15 were all from a geographically localized region of Serido, a county of Rio Grande do Norte State in northeastern Brazil. Probands from other families were, to our knowledge, not related to one another. All subjects with CGL had a generalized form of lipodystrophy with near absence of adipose tissue at birth or beginning in early infancy and muscular hypertrophy. Most patients presented with acanthosis nigricans, hepatomegaly, insulin resistance/diabetes, and hypertriglyceridemia. Their clinical characteristics are summarized in Table 1
and Fig. 1
. Additional clinical characteristics of the probands from CGL-F16, -F17, -F22, -F25, and -F26, and B-F1 have been reported previously (14, 15, 16, 17, 18). This project was approved by the Medical Ethics Committee of the University of Rio Grande do Norte. Written informed consent was obtained from all subjects.
|
|
Genomic DNA was extracted from peripheral blood cells using standard protocols (QIAquick; QIAGEN, Santa Clarita, CA). All exons of Gng3lg and AGPAT2 with adjacent intron-exon junctions were screened for mutations in families CGL-F1, CGL-F2, CGL-F16 through -F26, and B-F1. Exons 111 and the surrounding intronic sequences of Gng3lg (GenBank accession no. AP001458) were amplified by PCR using previously described primers (10). PCR conditions were 95 C for 2 min, followed by 30 cycles at 95 C for 45 sec, 60 C for 45 sec, and 72 C for 1 min, with a final extension at 72 C for 7 min. Intronic primers for PCR amplification of the six exons and exon-intron boundaries of AGPAT2 (GenBank accession no. AL590226) were designed using the Primer3 program (primer sequences available from the authors upon request). PCR conditions were 95 C for 2 min, followed by 35 cycles at 95 C for 45 sec, 56 C for 45 sec, and 72 C for 1 min, with a final extension at 72 C for 7 min. All PCR products were sequenced in both directions on an ABI 377 or ABI 3700 DNA sequencer and analyzed with Sequence Analysis 3.2 software (Applied Biosystems Division/PerkinElmer, Foster City, CA).
Once identified by sequence analysis, a rapid genotyping assay was used to screen for the adenosine insertion at codon 669 of exon 4 of Gng3lg. Five microliters of exon 4 PCR product, which were generated with upstream primer 5'-TTGTGTGTCAAGGGTCCTCA-3' and downstream primer 5'-AAAACAAGACCCCCACATCA-3', were digested with 7.5 U of HpaI restriction endonuclease (New England BioLabs, Beverly, MA) for 3 h at 37 C. The fragments were separated on a 2% agarose gel and visualized after staining with ethidium bromide. The adenosine insertion at codon 669 creates a unique HpaI restriction site, which generates two DNA fragments of 178 and 109 bp.
| Results |
|---|
|
|
|---|
|
We also detected five polymorphisms in Gng3lg in both affected and unaffected family members: in intron 1 at +62-+64delGGG; in intron 2 at 8 C
A; in intron 5 at +69A
G; in the 5' flanking region at 49T
C; and in exon 9 at A1288G (silent variant).
Sequence analysis of AGPAT2 in both CGL-affected individuals from family CGL-F19 revealed a homozygous A
T substitution at nucleotide 712 in exon 5 (Fig. 3A
, GenBank accession no. NM_006412), generating a TAG termination codon predicting a truncated protein 215 amino acids in length (the full-length AGPAT2 protein is 278 amino acids long). An affected individual from Brazil (CGL-F20; Table 1
) was a compound heterozygote for mutations predicting a premature termination codon (636C
A; Phe189X; Fig. 3B
) and a splice-site mutation (IVS3-1G
C; Fig. 3C
). Another affected individual from Brazil (CGL-F21) was homozygous for a deletion of exons 3 and 4, predicting a frameshift mutation and premature termination codon. Finally, a homozygous splice-site mutation (IVS42A
G) predicting a frameshift and a premature termination codon (Gln196fsX228) was found in AGPAT2 of an affected female subject (CGL-F22; Table 1
) as well as in three affected siblings in a family with Brunzell syndrome (B-F1; Table 1
and Fig. 3D
).
|
G; in the 3' untranslated region (UTR) at 1139C
T; in the 3'UTR at 1321G
C; and in the 3'UTR at 1420 C
T. We were not able to detect any mutations in Gng3lg or AGPAT2 in four subjects (CGL-F23 through -F26).
| Discussion |
|---|
|
|
|---|
3-subunit-linked gene (Gng3lg) in all BSCL2-linked families (10).
Our studies in 33 subjects from 26 CGL families and one Brunzell syndrome family represents one of the largest studies of CGL reported to date. We found four mutations in AGPAT2 and five mutations in Gng3lg, which explained the CGL phenotype in all but four subjects. Three of the mutations we found in AGPAT2 are novel, two of which predict premature chain termination and a truncated protein: A712T (Lys215X) and 636C
A (Phe189X). Figure 4
summarizes all mutations in AGPAT2 (A) and Gng3lg (B) identified to date.
|
A Brazilian female (CGL-F16) with CGL and severe insulin resistance (17), unrelated to Brazilian families CGL-F1-F15, had a Gng3lg mutation (645insAA). The same mutation was found in another Brazilian case by Magre et al. (10). Subject CGL-F17 lived in Canada, was of Indian origin, and had 980delC, which was previously described in a CGL case from India (10). The study subject CGL-F18 carried the diagnosis of acquired generalized lipodystrophy. This Lebanese subject had a 368delGTATC mutation in Gng3lg previously described in subjects with lipodystrophy of Lebanese origin (10). Thus, these findings are more likely to reflect a founder effect, although we cannot rule out the possibility of de novo mutations in Gng3lg.
We found three novel missense mutations in AGPAT2, two in a Brazilian female (CGL-F20), who was found to be a compound heterozygote (IVS3-1G
C and C636A, Phe189X), and one in two affected female siblings from family CGL-F19 (A712T, Lys215X). Brazilian subject CGL-F21 was a female who had a 317588del of AGPAT2, which also is most likely to have arisen in Portugal because this same mutation was described in a subject from Portugal, and European settlers from this region of Brazil are known to have originated from Portugal (11).
We did not find mutations in either Gng3lg or AGPAT2 in four subjects. Affected members of family CGL-F23, in addition to lipodystrophy, had neurodegenerative disorder and congenital cataracts. The proband from consanguineous pedigree CGL-F24 also had idiopathic pulmonary fibrosis and premature puberty. An 18-ar-old white female (CGL-F25) with lipodystrophy, diabetes, multiple xanthomas, bilateral cataracts, hemorrhagic pancreatitis, and cardiomyopathy also did not have mutations in Gng3lg or AGPAT2. Van Maldergem et al. (19) also identified three families from a total 44 families for which they did not find a mutation in either Gng3lg or AGPAT2. It is possible that these subjects may have mutations in regions of Gng3lg or AGPAT2 not studied (i.e. regulatory elements in flanking regions or introns) or may have mutations in yet-to-be identified genes that cause related lipodystrophic syndromes.
Reports of cases with generalized lipodystrophy indicate phenotypic variation. Subjects with Brunzell syndrome typically have congenital generalized lipodystrophy with cystic angiomatosis of the long bones (13). It has been suggested that Brunzell syndrome might have the same genetic etiology as BSCL (http://www.ncbi.nlm.nih.gov/entrez/dispomim.cgi?id=272500). However, others have suggested that Brunzell syndrome could be a separate entity (20). Two sisters (subjects B-F1-1 and B-F1-2) from an African-American pedigree had congenital generalized lipodystrophy, cystic angiomatosis of long bones, and primary amenorrhea, attributed to polycystic ovaries. The brother (subject B-F1-3), in whom systemic angiomatosis was not detected, and both affected sisters carried the same splice site mutation (IVS42A
G) in AGPAT2, showing directly that Brunzell syndrome is a clinical and genetic variant of CGL. Van Maldergem et al. (19) also recently reported a mutation in Gng3lg in a subject with Brunzell syndrome.
Our studies of a relatively large number of CGL subjects with mutations in both AGPAT2 and Gng3lg provided the opportunity to compare clinical characteristics of subjects with each genetic etiology. Although definitive conclusions cannot be made, it appears that the two genetic etiologies of CGL have very similar, albeit variable, clinical characteristics. One notable exception is that mental retardation appears to be associated with Gng3lg mutations but not AGPAT2 mutations (Table 1
). A similar observation was made by Van Maldergem et al. (19), suggesting that Gng3lg plays a role in brain development or function, although we cannot rule out the possibility that the mental retardation was due to consanguinity and homozygosity for one or more mutations at other loci.
In summary, these studies have led to several novel findings. First, in our relatively large sample set, mutations in AGPAT2 and Gng3lg were approximately equally prevalent in patients with CGL. Second, there do not appear to be any major distinguishing phenotypic characteristics between subjects with AGPAT2 or Gng3lg mutations, with the possible exception of mental retardation, which appears to be associated with Gng3lg mutations but not AGPAT2 mutations. Third, studies of 18 subjects from the same locale in Brazil with the same mutation in Gng3lg show that there is great phenotypic variability in clinical manifestations, suggesting modifying effects of other genes or environmental factors. Finally, we show that Brunzell syndrome may be caused by mutations in AGPAT2. Future functional studies of Gng3lg and AGFAT2 will undoubtedly unveil greater insights into the clinical spectrum of lipodystrophy disorders as well as adipose tissue biology.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations: BSCL, Berardinelli-Seip syndrome; CGL, congenital generalized lipodystrophy; OMIM, Online Mendelian Inheritance in Man; UTR, untranslated region.
Received March 19, 2003.
Accepted August 30, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. A. Hegele, T. R. Joy, S. A. Al-Attar, and B. K. Rutt Thematic review series: Adipocyte Biology. Lipodystrophies: windows on adipose biology and metabolism J. Lipid Res., July 1, 2007; 48(7): 1433 - 1444. [Abstract] [Full Text] [PDF] |
||||
![]() |
J R Porter and T G Barrett Monogenic syndromes of abnormal glucose homeostasis: clinical review and relevance to the understanding of the pathology of insulin resistance and {beta} cell failure J. Med. Genet., December 1, 2005; 42(12): 893 - 902. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y.-H. Yu and H. N. Ginsberg Adipocyte Signaling and Lipid Homeostasis: Sequelae of Insulin-Resistant Adipose Tissue Circ. Res., May 27, 2005; 96(10): 1042 - 1052. [Abstract] [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 |