The transcription factor sterol regulatory element binding protein (SREBP)-1c is intimately involved in the regulation of lipid and glucose metabolism. To investigate whether mutations in this gene might contribute to insulin resistance, we screened the exons encoding the aminoterminal transcriptional activation domain in a cohort of 85 unrelated human subjects with severe insulin resistance. Two missense mutations (P87L and P416A) were found in single affected patients but not in 47 control subjects. However, these variants were indistinguishable from the wild-type in their ability to bind DNA or to transactivate an SREBP-1 responsive promoter construct. We also identified a common intronic single nucleotide polymorphism (C/T) located between exon 18c and 19c. In a case-control study of 517 U.K. Caucasian case subjects and 517 age- and sex-matched control subjects, the T-allele at this locus was significantly associated with type 2 diabetes in men (odds ratio = 1.42 [1.11–1.82], P = 0.005) but not women. In a separate population-based study of 1,100 Caucasians, carriers of the T-allele showed significantly higher levels of total and LDL cholesterol (P < 0.05) compared with wild-type individuals. In summary, we have conducted the first study of the SREBP-1c gene as a candidate for human insulin resistance. Although the rare mutations identified were functionally silent in the assays used, we obtained some evidence, which requires conformation in other populations, that a common variant in the SREBP-1c gene might influence diabetes risk and plasma cholesterol level.
Type 2 diabetes is characterized by peripheral insulin resistance, increased hepatic gluconeogenesis, loss of glucose-induced insulin secretion and, in most patients, obesity (1). Recent data suggest that dysregulated fatty acid metabolism might be a key unifying mediator of these disparate phenomena (2–4).
Sterol regulatory element binding proteins (SREBPs) are transcription factors crucial in the regulation of fatty acid and cholesterol metabolism. To date, three SREBP isoforms have been identified: SREBP-1a and SREBP-1c, derived from a single gene (SREBF1) through alternative promoter usage, and SREBP-2, encoded by a separate gene (SREBF2). The aminoterminal segment of the SREBPs contains an acidic transactivation domain and a basic helix-loop-helix leucine zipper (bHLH-Zip) region that mediates protein dimerization and DNA binding. SREBPs are embedded in the membrane of the endoplasmic reticulum as 120-kDa precursor proteins. Following fatty acid or cholesterol depletion, 68-kDa aminoterminal fragments (mature SREBP) are cleaved proteolytically from these precursor proteins and migrate into the nucleus (5). There they activate different target genes, encoding key enzymes of fatty acid and cholesterol metabolism. Of the two isoforms, SREBP-1c is the predominant transcript in most organs, including liver and adipose tissue, of adult animals (6). The mouse isoform of SREBP-1c, also known as adipocyte differentiation and determination factor–1, was identified as a factor promoting differentiation of cultured preadipocytes (7).
Dysregulation of SREBP-1 expression or function has been demonstrated in humans (8) and a number of animal models of insulin resistance and diabetes. Thus, overexpression of a constitutively active form of SREBP-1c in adipose tissue leads to lipodystrophic diabetes in mice (9), genetic deletion of SREBP-1 rescues the fatty liver phenotype of ob/ob mice (10), and manipulation of SREBP-1 levels in pancreatic β-cells can modulate the effects of fatty acids on insulin secretion (11,12). There have been no reports of genetic studies of the SREBF1 gene in relation to human diabetes phenotypes; however, it has been suggested that the mutation in lamin A that leads to the Dunnigan Köbberling type of autosomal dominant face sparing lipodystrophy might disrupt normal interaction between lamins and SREBPs at the nuclear membrane (13).
To identify whether variations in the SREBF1 gene contribute to insulin resistance and type 2 diabetes, we screened exon 1c-8, encoding mature SREBP-1c, in a cohort of 85 unrelated human subjects with a variety of syndromes of severe insulin resistance (14). Two mutations resulting in an amino acid substitution, P87L and P416A, were identified in patients but not in control subjects. The P87A mutant is localized within the proline-leucine-glycine–rich domain, and the P416A mutant neighbors the bHLH-Zip domain (Fig. 1A). Since relatives of these two patients were not available for testing, we performed in vitro experiments in order to investigate whether these mutations might be implicated in the development of insulin resistance in the affected subjects.
Undifferentiated 3T3-L1 preadipocytes without endogenous expression of SREBP-1c were transfected with either wild-type mature SREBP-1c or the different mutants, and the ability to activate transcription was measured in a promoter reporter gene assay. The human LDL receptor promoter was used as a target sequence because it has been shown to be activated by SREBP-1c in previous reports (15). These experiments revealed that the mutants do not alter the ability of SREBP-1c to activate transcription (Fig. 1B). Since the P416A mutant is localized close to the bHLH-Zip domain, we further investigated whether the mutations influence DNA binding. Electromobility gel shift experiments were performed using in vitro translated wild-type and mutant mature SREBP-1c proteins and a radiolabelled, double-stranded DNA fragment containing sterol regulatory element (sre)-1, a known SREBP cis element (5). In these experiments, similar amounts of protein resulted in a similar shift of DNA fragments, indicating that these mutants do not influence the binding of SREBP-1c to sre-1 (Fig. 1C). Comparable results were found using a DNA fragment containing an e-box motif, a second SREBP binding site (data not shown).
We next examined whether common single nucleotide polymorphisms (SNPs) in the whole SREBF1 gene are associated with type 2 diabetes or phenotypes related to this disease. Searching an internal database for SNPs showing an allele frequency >10% resulted in the identification of an intronic polymorphism (C/T) between the exons 18c and 19c. Genotyping of a Caucasian population-based case-control study indicated that the T-allele at this locus was significantly associated with type 2 diabetes in men, whereas no association was found in women (Table 2). Sexual dimorphisms have often been reported in the relation between lipid parameters and fat distribution to insulin resistance and type 2 diabetes. It has been shown that men are more likely to gain visceral abdominal fat and deep subcutaneous abdominal fat, which correlate to fasting insulin levels (16–19). In these fat depots, lipogenic enzyme activity has been demonstrated to be lower than in subcutaneous fat, which is expected to result in a flux of free fatty acids to the portal vein affecting liver insulin sensitivity (20). Since SREBP-1c regulates the expression of lipogenic enzymes such as acetyl-CoA-carboxylase and fatty acid synthase (5), this might suggest a potential mechanism by which the described polymorphism (or other polymorphisms in linkage disequilibrium with it) could influence insulin sensitivity predominantly in men.
Recent reports have suggested that SREBP-1 could be implicated in the pathogenesis of β-cell failure in type 2 diabetic patients due to glucolipotoxicity (11,12). The results of the present genetic study might support this idea because we identified an association of a common polymorphism with type 2 diabetes in the case-control study but no rare mutations likely to cause severe insulin resistance in the severe insulin resistance cohort.
To examine whether the intronic C/T polymorphism might be associated with alterations in metabolic parameters related to type 2 diabetes, we further genotyped 1,100 unrelated U.K. Caucasian subjects from the Isle of Ely study, a prospective population-based study of the etiology of type 2 diabetes (21). We found no evidence for an association of the T-allele with altered glucose or insulin levels, but carriers of the T-allele had significantly higher levels of total and LDL cholesterol (Table 3) with a similar pattern for both sexes. Since the human LDL receptor is a known SREBP-1 target, this supports the idea that the described polymorphism might influence activation of transcription by SREBP-1c either directly or through linkage disequilibrium with an unidentified functional variant. No significant association was found with BMI, fasting triglycerides or fasting nonesterified fatty acids, 30-min insulin and 2-h glucose levels in an oral glucose tolerance test, waist-to-hip ratio, or systolic and diastolic blood pressure.
In summary, we have conducted the first examination of SREBP-1c as a candidate gene for human insulin resistance and detected two novel missense mutations. While these did not affect function, at least in the assays used, we did identify a common SNP in SREBP-1c that showed some evidence for association with diabetes risk and plasma cholesterol levels. These findings will require confirmation in other populations.
RESEARCH DESIGN AND METHODS
Molecular screening and population genetics.
Genomic DNA from subjects was randomly preamplified in a 50 μl primer extension preamplification (PEP) reaction using methods previously described (22). PEP-amplified DNA was diluted eightfold in water, and 4 μl of this dilution was subjected to further amplification of the SREBF1 gene using gene specific primers (Table 1). Twenty primer pairs were required to span the entire coding region of exon 1c-8. The PCR was performed using a 9:1 mixture of AmpliTaq Gold (Roche) and Pfu-Turbo (Promega) polymerase. After PCR amplification, PCR products were subjected to either single-stranded conformation polymorphism analysis or denaturating high-performance liquid chromatography, and products showing abnormal patterns were subsequently sequenced using ABI big-dye terminator (Perkin Elmer).
Severe insulin resistance cohort.
A cohort of human subjects with severe insulin resistance was collected in Cambridge, U.K. The inclusion criteria for this cohort were 1) a fasting insulin >100 pmol/l or an insulin requirement >200 units/day, 2) acanthosis nigricans, 3) and a BMI <33 kg/m2. In the present study, a total of 85 subjects from this cohort were screened for mutations in exon 1c-8 of SREBP-1. After identification of genetic variants in the severe insulin resistance cohort, a control population consisting of 47 subjects from different ethnic groups (Caucasian, African American, Hispanic, and Asian) was screened for the presence of these variants. Cambridge Local Research Ethics Committee approval was obtained, and informed consent was received from subjects before participation.
Case-control study cohort.
The Cambridgeshire Case Control Population was used. It consists of 517 subjects with type 2 diabetes and 517 matched control subjects and has been described in detail previously (23).
Ely cohort.
The Ely population is a collection of 1,100 samples obtained from the Isle of Ely Study, a prospective population-based cohort study of the etiology and pathogenesis of type 2 diabetes and related metabolic disorders in the U.K. (21). It is an ethnically homogeneous Caucasian population in which phenotypic data have been recorded at the outset and after 4.5 years. All subjects were aged between 40 and 65 years at baseline. The cohort was recruited from a population sampling frame with a high response rate (74%), making it representative of the general population for this area in eastern England.
Methods for genotyping.
Genotyping was performed using an adaptation of the fluorescence polarization template–directed incorporation method described by Chen, Levine, and Kwok (24). In short, PEP-amplified DNA samples were PCR amplified in 8-μl reactions with primers flanking the variant site; unincorporated dNTP and remaining unused primer were degraded by exonuclease I and shrimp alkaline phosphatase at 37°C for 45 min before the enzymes were heat inactivated at 95°C for 15 min. At the end of the reaction, the samples were held at 4°C. Single-base primer extension reactions were performed as previously described (24), and allele detection was performed by measuring fluorescence polarization on an LJL Analyst fluorescent reader (Molecular Devices). The PEP protocol was specifically developed and tested to ensure that allele bias was not introduced during the amplification process. A minimum of 12% internal replicate samples within each population (case-control and Ely subjects) were included in all genotyping tests to assess genotyping accuracy.
Plasmid cloning and mutagenesis.
The cDNA of the aminoterminal domain of human SREBP-1a (SREBP-1a-NT) was kindly provided by F. Foufelle (25). SREBP-1c-NT (aa 1–480) was cloned by PCR using SREBP-1a-NT as a template and the following primers: 5′-ATGGATTGCACATTCGAAGACATGCTTCAGCTTATC-3′ and 5′-GTCAGGCTCCGAGTCACTGCC-3′. The cycler conditions were denaturation (1 min at 95°C), annealing (1 min at 55°C), and elongation (2 min at 72°C) (32 cycles). PCR products were then ligated into pGEM-T-Easy Cloning Vector (Promega), and after sequence conformation cDNA was subcloned into the EcoRI site of pcDNA3.1. Mutagenesis was performed by using Quick Change Site Directed Mutagenesis Kit (Promega).
Transient transfection and promoter reporter gene assay.
3T3-L1 cells were maintained in 10% Dulbecco’s modified Eagle medium, neonatal calf serum, 2 mmol/l l-glutamine, 100 units/ml penicillin, and 0.1 mg/ml streptomycin at 5% CO2 and 37°C. For transfection experiments using Fugene-Reagent (Boehringer), cells were seeded in 24-well plates at 70% confluents. We used 1 μg of the reporter-plasmid (pGL-LDL-receptor-promoter, kindly provided by D. Müller-Wieland [15]) and 25 ng of pcDNA3.1-SREBP-1c-NT (wild-type or mutants). To correct for transfection efficiency, 10 ng per well of Renilla luciferase (pRL-CMV; Promega) were cotransfected. Equal amounts of plasmid DNA per well were obtained by adding appropriate amounts of pcDNA3.1. Forty-eight hours after transfection, cells were lysed by passive lysis buffer, and a dual luciferase reporter assay (Promega) was performed.
Electromobility gel shift assay.
Double-stranded oligonucleotides were prepared by mixing equal amounts of the complementary single-stranded DNAs containing a sre-1 (5′-GATCCTGATCACCCCACTGAGGAG-3′), heating to 70°C for 15 min, and cooling down to room temperature. The annealed oligonucleotides were labeled with 32P in the presence of [γ32P-ATP] and polynucleotide kinase (Amersham Pharmacia). SREBP-1c proteins were translated in vitro using 1 μg pcDNA3.1-SREBP-NT (wild-type or mutants) and reticulocyte lysate (TNT-Quick Coupled System; Promega) in a total volume of 50 μl. Of this reaction, 7 μl were then incubated for 10 min at room temperature in a total volume of 20 μl of 15 mmol/l HEPES (pH 7.9), 1 mmol/l EDTA, 1 mmol/l dithiothreitol, 1 mmol/l MgCl2, 10% glycerin, and 2 μg dIdC/dAdT. Subsequently, 100,000 cpm of the [γ32P-ATP]-labeled DNA fragment were added, followed by an incubation at 20°C for 20 min. The samples were separated on a 5% polyacrylamide gel in 0.5× TBE at 10 mA.
Rare mutations in mature SREBP-1c in human subjects with severe insulin resistance. A: Localization of the identified mutations in relation to functional domains. B: Promoter reporter gene experiment in order to investigate effect of mutants on transactivation. Preconfluent 3T3-L1 cells were cotransfected with a reporter plasmid carrying the coding sequence of the firefly luciferase under control of the human LDL-receptor promoter and expression plasmids encoding the aminoterminal domain (aa 1–480) of wild-type and mutant SREBP-1c (SREBP-1c-NT). Transfection efficiency was normalized by cotransfection of renilla-luciferase under control of a constitutive active promoter. Results were expressed as fold stimulation compared with MOCK-transfected cells. C: Electromobility gel shift assay in order to investigate effect of mutants on DNA binding. Wild-type and mutant SREBP-1c-NT were translated in vitro and incubated with a radiolabelled, double-stranded DNA fragment containing an sre-1. Five percent polyacrylamide gel electrophoresis, autoradiography.
Rare mutations in mature SREBP-1c in human subjects with severe insulin resistance. A: Localization of the identified mutations in relation to functional domains. B: Promoter reporter gene experiment in order to investigate effect of mutants on transactivation. Preconfluent 3T3-L1 cells were cotransfected with a reporter plasmid carrying the coding sequence of the firefly luciferase under control of the human LDL-receptor promoter and expression plasmids encoding the aminoterminal domain (aa 1–480) of wild-type and mutant SREBP-1c (SREBP-1c-NT). Transfection efficiency was normalized by cotransfection of renilla-luciferase under control of a constitutive active promoter. Results were expressed as fold stimulation compared with MOCK-transfected cells. C: Electromobility gel shift assay in order to investigate effect of mutants on DNA binding. Wild-type and mutant SREBP-1c-NT were translated in vitro and incubated with a radiolabelled, double-stranded DNA fragment containing an sre-1. Five percent polyacrylamide gel electrophoresis, autoradiography.
Sequences for primers for molecular scanning of exon 1c-8 of the human SREBF1 gene
| Exon | Amplicon | Forward primer | Reverse primer |
|---|---|---|---|
| 1c | 1 | GCCTTGACAGGTGAAGTCGGC | GGTTGGAGGTAGCCCTCCACT |
| 2 | 1 | AGTCTCTGCCTCACTCACGC | CAGGAGGTGGAGACAAGCTG |
| 2 | GGCCTATTTGACCCACCCTA | TCTTCCTTGATACCAGGCCC | |
| 3 | TGCCACATTGAGCTCCTCTC | ACCTGTAGAGAAGCCTCCCG | |
| 4 | CAGCTCCACCCCTGTGTTAG | CCCCAGTTTCTAAAGCCTGC | |
| 3 | 1 | GGGAGAGGGAGGTCTGTGTC | TAAGCTGTGTGTCTGGGCTG |
| 4 | 1 | ACTGACAGTCACAGCTGCCC | CCGTCTGTCTTCATGGCTGT |
| 2 | TTCATCAAGGCAGACTCGCT | TGTGTGTAGACCCCACTCCC | |
| 5 | 1 | GGGAGTGGGGTCTACACACA | GGAGCGGTAGGCTTCTCAAT |
| 2 | ACTGGTCGTAGATGCGGAGA | CAGTTGCCCCTGATCTGTTC | |
| 3 | TTTGCTAGGGCTCTCCAACC | AGAAGAGGCCAGACTGGAGC | |
| 6 | 1 | GCCATCGACTACATTCGCTT | AGGCACCAGGTCTCTTTCAG |
| 2 | CTTGCTAGGGCTCTCCAACC | GTTTCTGGTTGCTGTGTTGC | |
| 7 | 1 | GCGTCCTGGGCTAGCTTTAG | CTCCACCTCAGTCTTCACGC |
| 2 | GGGAACACAGACGTGCTCAT | CCTGTCATGAGGCTCAGAGG | |
| 8 | 1 | AATAACTGAGGCCTGGAGCC | GACGCTGGTGGTATCTGAGG |
| 2 | CCACAGTAAGGTGGGCAGAA | GTATCTGAGGGGCTGGGAAG | |
| 3 | CTCGTCTTCCTCTGCCTGTC | AGTGTCCCTCCCAAAGATGC |
| Exon | Amplicon | Forward primer | Reverse primer |
|---|---|---|---|
| 1c | 1 | GCCTTGACAGGTGAAGTCGGC | GGTTGGAGGTAGCCCTCCACT |
| 2 | 1 | AGTCTCTGCCTCACTCACGC | CAGGAGGTGGAGACAAGCTG |
| 2 | GGCCTATTTGACCCACCCTA | TCTTCCTTGATACCAGGCCC | |
| 3 | TGCCACATTGAGCTCCTCTC | ACCTGTAGAGAAGCCTCCCG | |
| 4 | CAGCTCCACCCCTGTGTTAG | CCCCAGTTTCTAAAGCCTGC | |
| 3 | 1 | GGGAGAGGGAGGTCTGTGTC | TAAGCTGTGTGTCTGGGCTG |
| 4 | 1 | ACTGACAGTCACAGCTGCCC | CCGTCTGTCTTCATGGCTGT |
| 2 | TTCATCAAGGCAGACTCGCT | TGTGTGTAGACCCCACTCCC | |
| 5 | 1 | GGGAGTGGGGTCTACACACA | GGAGCGGTAGGCTTCTCAAT |
| 2 | ACTGGTCGTAGATGCGGAGA | CAGTTGCCCCTGATCTGTTC | |
| 3 | TTTGCTAGGGCTCTCCAACC | AGAAGAGGCCAGACTGGAGC | |
| 6 | 1 | GCCATCGACTACATTCGCTT | AGGCACCAGGTCTCTTTCAG |
| 2 | CTTGCTAGGGCTCTCCAACC | GTTTCTGGTTGCTGTGTTGC | |
| 7 | 1 | GCGTCCTGGGCTAGCTTTAG | CTCCACCTCAGTCTTCACGC |
| 2 | GGGAACACAGACGTGCTCAT | CCTGTCATGAGGCTCAGAGG | |
| 8 | 1 | AATAACTGAGGCCTGGAGCC | GACGCTGGTGGTATCTGAGG |
| 2 | CCACAGTAAGGTGGGCAGAA | GTATCTGAGGGGCTGGGAAG | |
| 3 | CTCGTCTTCCTCTGCCTGTC | AGTGTCCCTCCCAAAGATGC |
Relationship between genotype C/T in intron 18c and diabetes in a case-control study in a U.K. Caucasian population
| Genotype frequencies | Allele frequencies | Hardy-Weinberg test | OR (95% CI) | P | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CC | CT | TT | total | C | T | ||||||||
| Men | |||||||||||||
| Control | 174 | 122 | 22 | 318 | 0.7390 | 0.2610 | 0.922 | 1.42 (1.11–1.82) | 0.005 | ||||
| Type 2 diabetes | 139 | 143 | 34 | 316 | 0.6661 | 0.3339 | 0.757 | ||||||
| Total | 313 | 265 | 56 | 634 | 0.7027 | 0.2973 | 0.993 | ||||||
| Women | |||||||||||||
| Control | 78 | 81 | 15 | 174 | 0.6810 | 0.3190 | 0.346 | 1.05 (0.76–1.47) | 0.748 | ||||
| Type 2 diabetes | 74 | 81 | 16 | 171 | 0.6696 | 0.3304 | 0.356 | ||||||
| Total | 152 | 162 | 31 | 345 | 0.6754 | 0.3246 | 0.188 | ||||||
| Total | |||||||||||||
| Control | 252 | 203 | 37 | 492 | 0.7185 | 0.2815 | 0.658 | 1.27 (1.04–1.55) | 0.014 | ||||
| Type 2 diabetes | 213 | 224 | 50 | 487 | 0.6674 | 0.3326 | 0.427 | ||||||
| Total | 465 | 427 | 87 | 979 | 0.6931 | 0.3069 | 0.431 | ||||||
| Genotype frequencies | Allele frequencies | Hardy-Weinberg test | OR (95% CI) | P | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CC | CT | TT | total | C | T | ||||||||
| Men | |||||||||||||
| Control | 174 | 122 | 22 | 318 | 0.7390 | 0.2610 | 0.922 | 1.42 (1.11–1.82) | 0.005 | ||||
| Type 2 diabetes | 139 | 143 | 34 | 316 | 0.6661 | 0.3339 | 0.757 | ||||||
| Total | 313 | 265 | 56 | 634 | 0.7027 | 0.2973 | 0.993 | ||||||
| Women | |||||||||||||
| Control | 78 | 81 | 15 | 174 | 0.6810 | 0.3190 | 0.346 | 1.05 (0.76–1.47) | 0.748 | ||||
| Type 2 diabetes | 74 | 81 | 16 | 171 | 0.6696 | 0.3304 | 0.356 | ||||||
| Total | 152 | 162 | 31 | 345 | 0.6754 | 0.3246 | 0.188 | ||||||
| Total | |||||||||||||
| Control | 252 | 203 | 37 | 492 | 0.7185 | 0.2815 | 0.658 | 1.27 (1.04–1.55) | 0.014 | ||||
| Type 2 diabetes | 213 | 224 | 50 | 487 | 0.6674 | 0.3326 | 0.427 | ||||||
| Total | 465 | 427 | 87 | 979 | 0.6931 | 0.3069 | 0.431 | ||||||
ORs are for the association of the T-allele with type 2 diabetes.
Relationship between genotype C/T in intron 18c and metabolic parameters in a U.K. Caucasian population
| CC | CT | TT | P | |
|---|---|---|---|---|
| n (m/f) | 194/251 | 146/192 | 31/32 | |
| Cholesterol | ||||
| Total (mmol/l) | 6.12 (6.03–6.22) | 6.25 (6.14–6.36) | 6.33 (6.07–6.58) | 0.046 |
| HDL (mmol/l) | 1.45 (1.42–1.48) | 1.43 (1.40–1.47) | 1.45 (1.36–1.53) | 0.584 |
| LDL (mmol/l) | 4.04 (3.95–4.13) | 4.18 (4.08–4.28) | 4.23 (4.00–4.46) | 0.024 |
| CC | CT | TT | P | |
|---|---|---|---|---|
| n (m/f) | 194/251 | 146/192 | 31/32 | |
| Cholesterol | ||||
| Total (mmol/l) | 6.12 (6.03–6.22) | 6.25 (6.14–6.36) | 6.33 (6.07–6.58) | 0.046 |
| HDL (mmol/l) | 1.45 (1.42–1.48) | 1.43 (1.40–1.47) | 1.45 (1.36–1.53) | 0.584 |
| LDL (mmol/l) | 4.04 (3.95–4.13) | 4.18 (4.08–4.28) | 4.23 (4.00–4.46) | 0.024 |
Data are means (95% CI) adjusted for age, sex, and BMI. P value for multiplicative model.
M.L. and I.B. contributed equally to this work.
Article Information
This work was supported in part by grants from the Deutsche Forschungsgemeinschaft, Germany (M.L.), the Wellcome Trust, London, U.K., (A.V.P. and S.O.R) and the Medical Research Council, London, U.K. (S.O.R. and N.J.N.)
