Alan J. Schafer

Candidate Gene Association Study in Type 2 Diabetes Indicates a Role for Genes Involved in β-Cell Function as Well as Insulin Action

Type 2 diabetes is an increasingly common, serious metabolic disorder with a substantial inherited component. It is characterised by defects in both insulin secretion and action. Progress in identification of specific genetic variants predisposing to the disease has been limited. To complement ongoing positional cloning efforts, we have undertaken a large-scale candidate gene association study. We examined 152 SNPs in 71 candidate genes for association with diabetes status and related phenotypes in 2,134 Caucasians in a case-control study and an independent quantitative trait (QT) cohort in the United Kingdom. Polymorphisms in five of 15 genes (33%) encoding molecules known to primarily influence pancreatic β-cell function—ABCC8 (sulphonylurea receptor), KCNJ11 (KIR6.2), SLC2A2 (GLUT2), HNF4A (HNF4α), and INS (insulin)—significantly altered disease risk, and in three genes, the risk allele, haplotype, or both had a biologically consistent effect on a relevant physiological trait in the QT study. We examined 35 genes predicted to have their major influence on insulin action, and three (9%)—INSR, PIK3R1, and SOS1—showed significant associations with diabetes. These results confirm the genetic complexity of Type 2 diabetes and provide evidence that common variants in genes influencing pancreatic β-cell function may make a significant contribution to the inherited component of this disease. This study additionally demonstrates that the systematic examination of panels of biological candidate genes in large, well-characterised populations can be an effective complement to positional cloning approaches. The absence of large single-gene effects and the detection of multiple small effects accentuate the need for the study of larger populations in order to reliably identify the size of effect we now expect for complex diseases.

Digenic inheritance of severe insulin resistance in a human pedigree

Impaired insulin action is a key feature of type 2 diabetes and is also found, to a more extreme degree, in familial syndromes of insulin resistance. Although inherited susceptibility to insulin resistance may involve the interplay of several genetic loci, no clear examples of interactions among genes have yet been reported. Here we describe a family in which five individuals with severe insulin resistance, but no unaffected family members, were doubly heterozygous with respect to frameshift/premature stop mutations in two unlinked genes, PPARG and PPP1R3A these encode peroxisome proliferator activated receptor γ, which is highly expressed in adipocytes, and protein phosphatase 1, regulatory subunit 3, the muscle-specific regulatory subunit of protein phosphatase 1, which are centrally involved in the regulation of carbohydrate and lipid metabolism, respectively. That mutant molecules primarily involved in either carbohydrate or lipid metabolism can combine to produce a phenotype of extreme insulin resistance provides a model of interactions among genes that may underlie common human metabolic disorders such as type 2 diabetes.

Familial partial lipodystrophy associated with compound heterozygosity for novel mutations in the LMNA gene

To the Editor: Dunnigan-Köbberling syndrome or familial partial lipodystrophy (FPL) is an inherited form of partial lipodystrophy characterised by selective loss of subcutaneous limb and gluteal fat, and excess facial fat deposition [1]. Some authors have suggested that FPL be subclassified into Dunnigan and Köbberling subtypes [2], the key difference being the loss of subcutaneous truncal fat in the Dunnigan subtype; this depot is preserved or increased in Köbberling FPL. Although the molecular mechanisms responsible for the unusual fat distribution seen in FPL remain unknown, the majority of cases are the result of heterozygous mutations in the LMNA gene (dominantnegative mutations in PPARG account for some of the remainder). Lamin A/C is principally a structural nuclear envelope protein which, like other intermediate filament proteins, consists of a central α-helical coiled-coil rod domain flanked by two globular domains. To date, the majority of mutations associated with FPL have been confined to exons 8 and 11 of LMNA, both of which contribute to the carboxy-terminal immunoglobulin-like globular domain. Structural modelling of this region suggests that the exon 8 FPL mutations occupy a discrete superficial patch on this globular domain [3] and are likely to alter protein–protein interactions either between lamin proteins themselves or between lamin proteins and other nuclear elements. As LMNA mutations are currently the most common explanation for inherited forms of partial lipodystrophy, we routinely sequence exons 8 to 12 in all probands with partial lipodystrophy. Three unrelated Caucasian pedigrees harbouring a novel heterozygous LMNA serine-583-leucine (S583L; nt-1748 [C→T]) mutation in exon 11 were identified (Table 1). The S583L variant was associated with a stereotyped pattern of partial lipodystrophy in which subcutaneous limb and gluteal fat was lost but subcutaneous abdominal, facial and neck fat was preserved (Fig. 1). This Köbberling-type FPL phenotype was apparent in kindreds A, B and C, and is consistent with that previously described in carriers of a mutation in the preceding residue (R582H) [4]. It was suggested that the fact that the R582H mutation is lamin A specific (exon 11 is not transcribed in lamin C), as opposed to exon 8 mutations which affect the structure of lamins A and C, might account for the subtle Köbberling-type FPL phenotype seen in carriers of this mutation [4]. However, this explanation is not supported by two reports describing carriers of an R584H mutation with typical Dunnigan-type FPL [5, 6]. Three patients in kindred C (subjects CI, CIV and CVI) were notable for a lack of subcutaneous truncal as well as subcutaneous limb fat (Fig. 1). They were subsequently found to be compound heterozygotes for S583L and a threonine-528-methionine (T528M; nt-1547 [C→T]) LMNA mutation in exon 9. Carriers of the T528M mutation alone (subjects CII and CV) were not clinically lipodystrophic, although both had slightly increased fasting triglycerides and subject CII was noted to be diabetic and hypertensive at the time of screening. Interestingly, a more substantial amino acid change involving a charge change at this residue (threonine-528-lysine) is known to produce a form of EmeryDreifuss muscular dystrophy [7]. None of the T528M carriers reported muscular weakness or palpitations. Results of clinical evaluation, creatine kinase measurement and electrocardiography were also unremarkable in these subjects. Acanthosis nigricans and features of the metabolic syndrome were variably present in both single heterozygotes and compound heterozygotes (Table 1). Neither the S583L nor the T528M LMNA variant was present in 100 unrelated control subjects (200 alleles). Serine 583 and threonine 528 are highly conserved residues (identical in human, rat, mouse, chick and Xenopus). The S583L mutation is bracketed by two known mutations in exon 11, R582H [4] and R584H [5, 6], which are only expressed in lamin A. Although the T528M mutation is not expected to alter the structure of the globular C-terminal domain substantially, it may modify the phenotypic impact of the S583L mutation. The notion that a mutation without a clinically discernable phenotype might, in combination with a second mutation, produce a pathological phenotype is supported by the observations made in kindreds with an arginine-527-histidine (R527H) mutation. R527H heterozygotes were said to be phenotypically “normal”, whilst R527H homozygotes develop mandibulo-acral dysplasia [8], a complex phenotype including partial lipodystrophy and insulin resistance. Compound heterozygosity has been reported in one other subject with FPL [5]. In that particular case, the subject was a compound heterozygote for a R482Q and V440M mutation and was reported to have the typical Dunnigan pattern of lipodystrophy accompanied by very severe insulin resistance, diabetes and aggressive cardiovascular disease. The single mutation carriers in this family had a similar pattern of lipodystrophy but less severe metabolic consequences. In summary, we have identified two novel LMNA mutations. The S583L variant was associated with a stereotyped pattern of “atypical” partial lipodystrophy, in which subcutaneous abdominal fat was preserved, in three unrelated kindreds (Köbberling-type FPL). One of these kindreds also harboured the T528M variant, and compound heterozygotes manifest typical Dunnigan-type FPL. These observations add to the remarkably complex genotype–phenotype relationships associatDiabetologia (2004) 47:753–763

9 Sex Determination and Its Pathology in Man

Publisher Summary This chapter focuses on the molecular pathology of sex determination. Significant progress has been made in the identification and cloning of genes involved in sex determination. The advances have primarily resulted from the interdisciplinary analysis of individuals with altered or abnormal sexual development. This in turn has clarified the molecular basis for the wide variety of sex-reversal syndromes. The demarcation between sex determination and sex differentiation is becoming less distinct, particularly with the discovery of a gene steroidogenic factor-1 ( SF -1) which is involved at multiple levels of sexual development—in gonadal genesis, anti-Mullerian hormone (AMH) activation, and steroidogenesis. Many of the genes currently known to be involved in sex determination also have critical functions in other developmental processes— SF -1 in adrenal gland development, WT -1 in kidney formation, and SOX 9 in bone morphogenesis. These diverse roles emphasize the developmental complexities of embryonic differentiation and provide an insight into the intricacies of sex determination.

PGC-1alpha genotype modifies the association of volitional energy expenditure with [OV0312]O2max.

UNLABELLED Sedentary lifestyles are increasingly common and result in low cardiorespiratory fitness ([OV0312]O2max), a well-established predictor of early mortality and coronary heart disease (CHD). Adaptation in [OV0312]O2max after exercise training varies considerably between people. Because there are hereditary components to fitness, it is likely that genetic factors explain some of this variability. PPARGC1 (PGC-1alpha) coactivates genes involved in energy transduction and mitochondrial biogenesis. Transgenic mouse data demonstrate that overexpression of PGC-1alpha mRNA increases endurance capacity by transformation of nonoxidative to oxidative skeletal muscle tissue. Other murine studies demonstrate that exercise increases PGC-1alpha mRNA expression. PURPOSE To explore whether a candidate polymorphism in the PGC-1alpha gene modifies the association between physical activity energy expenditure (PAEE) and predicted [OV0312]O2max ([OV0312]O2max.pred). METHOD We examined whether the Gly482Ser polymorphism of PGC-1alpha modified the relationship between objectively measured PAEE and [OV0312]O2max.pred in a population-based sample of 599 healthy middle-aged people. PAEE was assessed using individual calibration with 4 d of heart rate monitoring. [OV0312]O2max.pred was measured during a submaximal exercise stress test on a bicycle ergometer. RESULTS Homozygosity at the Ser482 allele was found in 12.7% of the cohort, whereas 38.9% and 48.4% carried the Gly482Gly and Gly482Ser genotypes, respectively. The association between PAEE and [OV0312]O2max.pred (mL x kg(-1) x min(-1)) was strongest in people homozygous for the Ser482 allele (beta = 12.03; P < 0.00001) compared with carriers of the Gly allele (beta = 5.61; P < 0.00001). In a recessive model for the Ser482 allele, the interaction between PAEE and genotype on [OV0312]O2max.pred (L x min(-1)) was highly significant (P = 0.009). CONCLUSION Our results indicate that Ser482 homozygotes may be most capable of improving cardiorespiratory fitness when physically active, and that Gly482Ser may explain some of the between-person variance previously reported in cardiorespiratory adaptation after exercise training.

Genetic variants of insulin receptor substrate-1 (IRS-1) in syndromes of severe insulin resistance. Functional analysis of Ala513Pro and Gly1158Glu IRS-1

Aims To define further the role of IRS‐1 mutations in human syndromes of severe insulin resistance.

Differentiating campomelic dysplasia from Cumming syndrome

Cumming syndrome (CS) (OMIM# 211890) and campomelic dysplasia (CD) (OMIM #114290) are two skeletal dysplasias that present with limb shortness and bent long bones. In the two cases of CS that our group reported previously [Dibbern et al., 1998], we subsequently identified SOX9 mutations, which to date have only been reported in CD cases, prompting us to re-examine the manifestations of these cases. CD, first described in 1971 [Maroteaux et al., 1971], is generally a neonatally lethal condition characterized by short, bowed long bones, macrocephaly, a flat face, hypertelorism, depressednasal bridge,micrognathia, cleft palate, lowset ears, hypoplastic trachea, a small chest, hypoplastic scapulae, small iliac wings, hypoplastic cervical vertebrae, kyphoscoliosis in survivors, non-mineralized pedicles, and frequently male to female sex reversal. Less common findings include congenital heart defects, hydronephrosis, absence of the olfactory bulbs and tracts, and hydrocephalus [Houston et al., 1983; Mansour et al., 1995, 2002; Taybi and Lachman, 1996; Mortier et al., 1997;Gorlin et al., 2001; Spranger et al., 2002]. Loss of function mutations of a single allele of theSOX9 gene on chromosome17 has been shown to cause CD [Foster et al., 1994; Wagner et al., 1994; Kwok et al., 1995; Schafer et al., 1995, 1996; Cameron et al., 1996;Meyer et al., 1997;McDowall et al., 1999; Friedrich et al., 2000; Thong et al., 2000; Preiss et al., 2001; Sock et al., 2003]. CS was first described in 1986 in a 27-week-gestation male fetuswith cervical lymphocele, polycystic dysplasia of kidneys, pancreas, and liver, short intestines, and polysplenia. Radiographically, there were bowed long bones, a small chest, and ossificationdefects of the vertebraewith anormal scapulae and pelvis [Cumming et al., 1986]. Subsequent reports of CS included two cases by Urioste et al. [1991], one by Ming et al. [1997], two by Dibbern et al. [1998], and affected sibs by Perez del Rio et al. [1999]. Other than the initial report, all the other caseswere female. These caseshave all had campomelia of long bones, a small chest, and a cystic hygroma. With the exception ofCase 1 ofDibbern et al. [1998], all hadat least onemulticystic organ with or without polysplenia. The gene for CS is not known, but SOX9was excluded in one case [Ming et al., 1997]. Case 1 (Registry #95–130) was a 28-week-gestation apparently female fetuswith severe polyhydramnios, a small chest, a large cervical lymphocele, multiple renal parenchymal cysts, and cystic dilatation of the calyceal system and pelvis of the right kidney. Radiographs showedmesomelia of the long bones with campomelia of femora, tibiae, and ulnae. The fibulaewere hypoplastic. The scapulae were hypoplastic and the spine had flat vertebral bodies with coronal clefts. The iliumwas tall and narrow. The acetabular roof was hypoplastic, and the ischium and pubis were poorly ossified. The genetic sex of the baby was determined to be male by PCR amplification of genomic DNA using SRY and AMELX sequences as markers for the Y and X chromosomes, respectively. Case 2 (Registry #93–409) was a 21-week-gestation apparently female fetus with midface hypoplasia, micrognathia, a large cystic lymphocele, and small bell-shaped chest. Necropsy did not show cysts of the internal organs but there was a complex congenital heart defect (ventricular septal defect, double-outlet right ventricle, and coarctation of the aorta). Radiographs revealed rhizo-, meso-, and acromelia with campomelia of the long bones. The fibulae were hypoplastic. The vertebraewereflatwithout coronal clefts. The iliawere tall and wide, and acetabular roofs poorly ossified. The scapulae were not well seen on the available radiographs. The karyotype was normal 46,XY. DNA samples from these two cases were screened for SOX9 mutations. In Case 1, genomic DNA was amplified using PCR and single-stranded conformation polymorphism (SSCP) mutation analysis of the entire SOX9 open reading frame and exon-intron boundaries as described previously [Kwok et al., 1995]. In Case 2, the PCR products were sequenced directly, as described [Wagner et al., 1994]. Case 1 was found to have a heterozygous c891a transversion by sequencing following SSCP, and the DNA change was subsequently confirmed by restriction digest analysis. This change results in a nonsensemutation, which alters a tyrosine to a premature stop (Y297X) and is predicted to result in a truncated protein of 296 amino acid residues compared to 509 normally. Case 2 had a heterozygous t1528c mutation that changes the TGA stop codon to a CGA arginine codon. This mutation is expected to result in a mutant protein with an additional 50 residues. The specific mutations identified here have not been reported before nor identified in our laboratories (G.S., I.B., andA.J.S., unpublished). Pre-mature stop codonsas inCase 1have been identified in several CDcases [Foster et al., 1994; Wagner et al., 1994; Meyer et al., 1997]. A mutation similar to that in Case 2, a frameshift at codon 507 resulting in a mutant protein extended by 70 residues, was described in a classic 46,XXCD casewho died shortly after birth andwho had multiple small cysts of the ovaries [Kwok et al., 1995]. In some cases, mutant SOX9 protein, or its mRNA, has been shown to be unstable [Meyer et al., 1997], whichmay also be the case for themutant protein inCase 2, or the additional protein residues may otherwise interfere with normal protein function. Our cases had many findings in common with CS: campomelia of prenatal onset, cystic hygroma, and a small chest. Case 1 also had a cleft palate and multicystic kidneys. Case 2 had a complex congenital heart defect. The neck in CD is usually described as short with redundant skin. Cystic hygroma and cystic kidneys have rarely been reported in CD [Houston et al., 1983;Mansour et al., 1995; Gorlin et al., 2001]. In contrast to our two cases that had short, irregular chonGrant sponsor: NIH; Grant number: 5P01-HD22657; Grant sponsor: DFG; Grant number: Sche 194/15-1.