1

ONLINE SUPPLEMENT COMMON POLYMORPHISMS IN THE CYP11B1 AND CYP11B2 GENES:

EVIDENCE FOR A DIGENIC INFLUENCE ON HYPERTENSION *Samantha Alvarez-Madrazo1, *Scott M. MacKenzie1, Eleanor Davies1, Robert Fraser1, Wai-Kwong Lee1, Morris Brown2, Mark J. Caulfield3, Anna F. Dominiczak1, Martin Farrall4, Mark Lathrop5, Thomas Hedner6, Olle Melander7, Patricia B. Munroe3, Nilesh Samani8, Paul M. Stewart9, Björn Wahlstrand6, John Webster10, Colin N.A. Palmer11, Sandosh Padmanabhan1, John M.C. Connell11.

*Joint first authors 1. Institute of Cardiovascular and Medical Sciences, College of Medical,

Veterinary and Life Sciences, University of Glasgow, Glasgow G12 8TA, U.K.

2. Clinical Pharmacology Unit, University of Cambridge, Addenbrookes Hospital, Cambridge CB2 2QQ, U.K.

3. Clinical Pharmacology, William Harvey Research Institute, Barts and the London Medical & Dental School, Queen Mary University of London, Charterhouse Square, London EC1M 6BQ, U.K.

4. Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford OX3 7BN, U.K.

5. Centre National de Genotypage, Rue Gaston Cremieux, Evry, Paris 91057, France.

6. Institute of Medicine, The Sahlgrenska Academy, University of Gothenburg, PO-BOX 417, SE 405 30, Gothenburg, Sweden.

7. Department of Clinical Sciences, Lund University, Malmö 20502, Sweden. 8. Glenfield Hospital, University of Leicester, Groby Road, Leicester LE3

9QP, U.K. 9. Centre for Endocrinology, Diabetes and Metabolism, School of Clinical

and Experimental Medicine, University of Birmingham, Edgbaston, Birmingham B15 2TT, U.K.

10. Medicine & Therapeutics, Aberdeen Royal Infirmary, Foresterhill, Aberdeen AB9 2ZB, U.K.

11. Medical Research Institute, School of Medicine, University of Dundee, Dundee DD1 9SY, U.K.

Short title: CYP11B1/CYP11B2 DIGENIC INFLUENCE ON HYPERTENSION Corresponding author: John Connell

Address: School of Medicine, University of Dundee, Ninewells Hospital, Dundee DD1 9SY

Tel: 01382 632673 Fax: 01382 644267 email: j.m.connell@dundee.ac.uk

2

Supplemental Methods Rationale for polymorphism selection in the CYP11B1/CYP11B2 locus SNPs in the CYP11B1/CYP11B2 locus were selected for inclusion in this study either because they had been reported as functional in previous studies (e.g. rs1799998), or because they were identified as tagSNPs in our pilot study. This pilot study was designed to dissect in detail the polymorphic variations of this locus in Caucasian subjects. We analysed 79 Caucasian unrelated offspring (50 hypertensives and 29 normotensives) from the BRIGHT transmission disequilibrium test (TDT). Our aim was to estimate linkage disequilibrium (LD), haplotype structure and haplotype tagSNPs (htSNPs) for the CYP11B locus. As SNPs are abundant in this locus and high-throughput genotyping was available, in order to allow fine mapping of a causal variant, it was decided that a combination of a selection of potential causal variants (-344, -1889 and -1859) and a set of several markers (htSNPs) capturing most of the information in the region would be appropriate. The most informative regions across the CYP11B1/CYP11B2 locus were identified by reviewing LD plots using the HapMap CEU (HapMap Release #20 and #22). However, sequencing of this region is difficult because of the similarity between these genes and, therefore, the HapMap coverage of this region was incomplete. Barr et al. previously sequenced the CYP11B1 and CYP11B2 genes extensively in 26 normotensives subjects from the MONItoring of trends and determinants in CArdiovascular disease (MONICA) IV study (Barr et al. 2007). The HapMap data for Caucasians combined with the 26 normotensive subjects from the MONICA study were used to select a set of SNPs that would capture the variability in the region (see Figure S1). In order to have 90% coverage of this locus, 15 SNPs in the CYP11B1/CYP11B2 were selected for genotyping (see Table S1). Four individuals were excluded because their genotype call rate was <50%. As there is significant LD in the locus and there is currently no consensus for selecting haplotype blocks or their boundaries, two blocks were chosen to ensure the best selection of htSNPs: Block 1 comprised SNPs in the CYP11B1 plus the intergenic region and Block 2 comprised SNPs in the CYP11B2 plus the intergenic region. For Block 1, by using rs5301 in the 3’UTR region, rs6410 in exon 1 and rs142570922 in the CYP11B1 promoter, 100% of alleles were captured with r2>0.8. For Block 2, rs4736354 in the intergenic region, rs4546 in exon 3, the Intron2 polymorphism and rs1799998 in the promoter of the CYP11B2 gene were used. Therefore, these two blocks predicted 90% coverage of the most common haplotypes and the remaining SNPs. In addition to these criteria, the performance and cost of the genotyping assays, the interruption of LD over short distances with the presence of a conversion and the strength of association in the discovery (BRIGHT) cohort were also considered, leading to a final selection of six htSNPs in the two cohorts (see Table S3).

3

Genotyping PCR of genomic DNA for Direct Sequencing For the discovery stages of this study, the rs1799998 SNP was genotyped by direct sequencing. A 232bp fragment encompassing the rs1799998 SNP was amplified using 25ng template DNA in a 25µl reaction volume and 1.25U Thermo-Start DNA Polymerase (ABGene, Surrey UK), according to manufacturer’s protocol. Oligonucleotide primers SF1F and SF1R (Table S2) were added at a final concentration of 10µM. The reactions were subjected to the following PCR conditions: a single cycle at 94°C for 15min, followed by 30 cycles at 94°C for 30s, 68°C for 30s, 72°C for 1min and then incubated at 72°C for 7min.

A 600bp fragment in intron 2 of CYP11B2, encompassing the Intron2 polymorphism, was amplified using 50ng template DNA in a 25µl reaction volume and 0.625U Thermo-Start DNA Polymerase (ABGene, Surrey UK), according to manufacturer’s protocol. Oligonucleotide primers ICTAQMANF and ICTAQMANR were added at a final concentration of 20µM. The reactions were subjected to the following PCR conditions: a single cycle at 95°C for 15min, followed by 45 cycles at 95°C for 15s, 62°C for 30s, 72°C for 2min and then incubated at 72°C for 7min.

A CYP11B1 promoter fragment including rs142570922 and rs149845727 was amplified using a nested PCR to achieve specificity, as optimised by Barr et al. In the first step, the same amount of template, reaction volume and 1.25U Thermo-Start DNA Polymerase as for rs1799998 was used. Oligonucleotide primers (concentration 10µM) B1 5’UTR and B1Prom260 were added. The cycling conditions were: 95°C for 15 minutes followed by 35 cycles of 95°C for 30s, 60°C for 30s and 72°C for 3 minutes, finishing at 72°C for 7 minutes. Then the PCR products were diluted 1:10 in nuclease free water, using 1µL PCR product as a template. 1µL diluted template DNA was amplified in a 25µl reaction volume using 0.625 U Thermo-Start DNA Polymerase (ABGene, Surrey UK). Oligonucleotide primers B1 5’7-32 and B1 369-393 at 10µM were added. The amplification conditions were 95°C for 15 minutes followed by 20 cycles of 95°C for 15s, 60°C for 30s and 72°C for 1 minutes, finishing at 72°C for 7 minutes; a PCR product of 387bp was obtained. Sequencing reactions All PCR products were cleaned for sequencing using AMPure (Agencourt, USA) according to the manufacturer’s protocol. 10µL of cleaned PCR product was used for sequencing by the BigDye Terminator v3.1 kit (Applied Biosystems, USA) according to the manufacturer’s protocol, along with the corresponding sequencing primers. The rs1799998 products were sequenced with SF1F at a concentration of 3.2µM and cycled at 45s at 96°C, 25s at 50°C, 4min 60°C for 25 cycles. The CYP11B2 Intron2 products were sequenced with primer INTCONR(B1B2) and the CYP11B1 promoter products with primer B1PROM 250-229AS, both at 3.2µM concentration, cycled for 45s at 96°C and 4min at 60°C for 25 cycles.

Sequencing products were cleaned using CleanSEQ (Agencourt, USA) according to the manufacturer’s protocol. DNA sequences were analysed using a 3730 DNA Analyzer (ABI, USA) and visualized using SeqScape

4

version 2.1.1. Samples were excluded when the chromatogram was judged of poor quality following two attempts at genotyping. See Table S1 for primer sequences. Validation of TaqMan® assay for SNP rs1799998 The TaqMan® assay (Applied Biosystems, UK) for rs1799998 was made available after the discovery phase had been completed and was adopted for the second (Swedish) study. To confirm that the results for this assay were consistent with those obtained by direct sequencing, 90 BRIGHT samples were re-genotyped by Taqman. All genotyping calls obtained by Taqman assay matched those obtained by direct sequencing (data not shown).

In vivo Expression Studies Preparation of human adrenal cDNA Total RNA was isolated from frozen adrenal tissue using the RNeasy® Lipid Tissue Mini kit (QIAGEN, Crawley, U.K.), according to the standard protocol, the resulting RNA being DNased using the Turbo DNA-free kit (Applied Biosystems) and quantified using the NanoDrop spectrophotometer. Genomic DNA was isolated from frozen tissue using the DNeasy Blood & Tissue kit (QIAGEN) according to the standard protocol. Samples were genotyped for key polymorphisms at the CYP11B1 and CYP11B2 by a combination of direct sequencing and Taqman assays. On the basis of these results, 5 tissue samples were found to be heterozygous for Haplotypes 1 and 2, as defined in Figure 1A. Total adrenal RNA from these samples was reverse transcribed using the ImPromII Reverse Transcription System (Promega, Southampton, U.K.), according to the standard protocol, using 2mM MgCl2, 0.5µg random primers and 0.5µg total RNA template. Negative control reactions omitting reverse transcriptase were also performed. Measurement of relative levels of Haplotype1- and Haplotype2-derived CYP11B1 and CYP11B2 cDNA A summary of this method is provided in Figure S3. Briefly, 2µL human adrenal cDNA was subjected to PCR of either CYP11B1 or CYP11B2 using FastStart Taq (Roche, Mannheim, Germany) according to the standard protocol. PCR of CYP11B1 cDNA used primers B1ASPCRF and B1ASPCRR, while CYP11B2 amplification used primers B2ASPCRF and B2ASPCRR (See Table S1). Resulting PCR product was cloned into the pCR®4-TOPO® plasmid vector, which was then used to transform One Shot® TOP10 Competent Cells, according to the standard chemical transformation protocol of the TOPO TA Cloning Kit for Sequencing (Invitrogen, Paisley, U.K.). These cells were spread on warmed LB agar plates containing 50µg/mL kanamycin and incubated overnight at 37°C. The next day, 96 colonies resulting from each transformation were picked randomly and suspended in water. Colony plasmid DNA was amplified using the PlasmidAmp kit (QIAGEN, Crawley, U.K.) according to the standard protocol, then sequenced using the BigDye Terminator v3.1 Cycle Sequencing Kit and the M13F primer. Sequences resulting from CYP11B1 amplification were classified as belonging to either Haplotype1 or Haplotype2 categories on the basis of the rs6410 SNP, while positive CYP11B2 sequences were classified on the basis of the rs4546 SNP (Figure 1A).

5

Negative samples either contained no readable sequence or contained sequence corresponding to genomic DNA, presumably as a result of incomplete DNasing of the original adrenal RNA sample. The resulting numbers of positive Haplotype1 and Haplotype2 colonies were subjected to a one sample t-test. Method validation To ensure there was no inherent bias in the above method, validating experiments were performed using plasmids in place of adrenal cDNA. These plasmids, a gift from Professor Perrin White, contained CYP11B1 or CYP11B2 cDNA in a pGL3 backbone; as supplied, the plasmids were found to correspond to CYP11B1 Haplotype1 and to CYP11B2 Haplotype2, so were named pB1Hap1 and pB2Hap2, respectively. The opposite forms, pB1Hap2 and pB2Hap1, were generated using the original plasmids, the QuikChange® Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA), and specific SDM primers (Table S1). Resulting plasmids were purified using the QIAfilter Plasmid Midi kit (QIAGEN, Crawley, U.K.) and sequence was confirmed using the BigDye Terminator v3.1 Cycle Sequencing Kit and the M13F primer, as above. Using these plasmids as template, we found that the RT-PCR primers designed for the separate amplification of the highly-homologous CYP11B1 or CYP11B2 cDNAs were specific. Furthermore, by mixing together varying amounts of pB1Hap1 and pB1Hap2, or of pB2Hap1 and pB2Hap2, then subjecting this template DNA to the appropriate protocol, as above, we could demonstrate that the resulting numbers of Hap1- and Hap2-containing cell colonies was not preferentially biased towards one particular form (see Figure S4).

6

Supplemental Tables Table S1. Characteristics of polymorphisms genotyped in the CYP11B1 and CYP11B2 genes in the Pilot BRIGHT TDT. SNPs are presented in chromosomal order and their location within each gene is also indicated. Position base pair (bp) derived from NCBI www.ncbi.nlm.nih.gov/projects/SNP; Assembly GRCh37p5, Genome Build 37.3. Alleles M/m: Major allele/minor allele; MAF, minor allele frequency; Obs. Het, observed heterozygosity; HWE p, Hardy-Weinberg equilibrium p-value.

Variationno.1 8 CYP11B1 3’UTR rs4736312 143953937 A/C 0.49 0.45 0.512 8 CYP11B1 3’UTR rs5301 143952273 A/G 0.43 0.40 0.133 8 CYP11B1 Intron 5 rs6471570 143957405 C/A 0.49 0.44 0.384 8 CYP11B1 Intron 2 rs28551743 143959412 G/A 0.46 0.40 0.145 8 CYP11B1 Intron 2 rs5895733 143959478-143959479 -/AT 0.46 0.48 1.006 8 CYP11B1 Exon 2 rs5283 143960597 C/T 0.44 0.32 0.0047 8 CYP11B1 Exon 1 rs6410 143961005 A/G 0.49 0.45 0.458 8 CYP11B1 Promoter rs149845727 (-1859) 143963087 C/T 0.47 0.45 0.519 8 CYP11B1 Promoter rs142570922 (-1889) 143963117 A/C 0.49 0.43 0.3510 8 Intergenic rs4736354 143980466 G/C 0.41 0.40 0.2011 8 Intergenic rs6471580 143986701 A/G 0.42 0.38 0.0712 8 CYP11B2 Intron 6 rs5895735 143994419-143994420 -/G 0.44 0.38 0.0613 8 CYP11B2 Exon 3 rs4546 143996553 C/T 0.43 0.38 0.0714 8 CYP11B2 Intron 2 Intron 2 143996983-82 Wt/Conv 0.50 0.47 0.7715 8 CYP11B2 Promoter rs1799998 (-344) 143999600 T/C 0.45 0.51 1.00

Chr Gene Location SNP Position bp MAF Obs.Het HWE pAlleles M/m

7

Table S2: Primer sequences used for direct sequencing and in vivo expression studies.

Primer Method Sequence 5’–3’

SF1F PCR & Sequencing GTG TCA GGG CAG GGG GTA

SF1R PCR AGG CGT GGG GTC TGG ACT

ICTAQMANF PCR GAT GGC ATG AAG CAC AAA GCT

ICTAQMANR PCR CCT TGG GCG ACA GCA CA

B1 5’UTR PCR TCC TTC GCA TCC CTT GTA AGT T

B1Prom260 PCR CTT GGA TTA TTC ATC TCC TTG CAA GG

B1 5’7-32 PCR GCA TCC CTT GTA AGT TGG ATT CCT AA

B1 369-393 PCR AAG CAT TCC CTT TGA AAA CTG GTA C

INTCONR(B1B2) Sequencing GTG TTC GAG CTG CAG CCT TTT C

B1PROM 250-229AS Sequencing AAG TCA AAT TGT CTC TGT TTG

pCMVB1 SDMHap2F In vivo Expression TTC CAG GAA CTG GGG CCC ATT TTC

AGG TAC GAT TTG GGA GGA GC

pCMVB1 SDMHap2R In vivo Expression GCT CCT CCC AAA TCG TAC CTG AAA

ATG GGC CCC AGT TCC TGG AA

pCMVB2 SDMHap1F In vivo Expression GTG GCC AGG GAC TTC TCC CAG GCC

CTG AAG AAG AAG GTG CTG C

pCMVB2 SDMHap1R In vivo Expression GCA GCA CCT TCT TCT TCA GGG CCT

GGG AGA AGT CCC TGG CCA C

B1ASPCRF In vivo Expression CCT GCA CCT GGA AGT A

B1ASPCRR In vivo Expression CAC ACC ATG CCT GC

B2ASPCRF In vivo Expression TGC ATC CCT GCA GGA TGA T

B2ASPCRR In vivo Expression CCG CTC TCC AAA AAG AGC T

M13F In vivo Expression GTA AAA CGA CGG CCA G

8

Table S3: Characteristics of the polymorphisms genotyped in the CYP11B1 and CYP11B2 genes in the BRIGHT and Swedish Cases and Controls. Shown are the most informative SNPs genotyped in 2 Caucasian population studies at the CYP11B1/CYP11B2 locus. SNPs are presented in chromosomal order and their location within each gene is also indicated. Position base pair (bp) derived from NCBI www.ncbi.nlm.nih.gov/projects/SNP; Assembly GRCh37p5, Genome Build 37.3. Status: ALL, includes the Case Control subjects; HTN, includes only the hypertensive cases genotyped (n=1643 BRIGHT, n=1612 Swedish); NT, includes only normotensive members (n=1697 BRIGHT, n=1317 Swedish). MAF, minor allele frequency; Obs. Het, observed heterozygosity; HWE p, Hardy-Weinberg equilibrium p-value.

Chromosome Gene Status MAF HWE p MAF HWE p

8 CYP11B1 Exon 1 rs6410 143961005 ALL A 0.481 0.49 0.53 0.46 0.48 0.04HTN 0.51 0.58 0.49 0.56NT 0.48 0.14 0.47 0.02

8 CYP11B1 Promoter rs149845727 143963087 ALL C 0.476 0.50 0.80 0.467 0.53 1.0 X 10-3

(-1859 C/T) HTN 0.52 0.19 0.56 3.53 X 10-7

NT 0.48 0.10 0.49 0.448 CYP11B1 Promoter rs142570922 143963117 ALL A 0.451 0.49 0.47 0.447 0.52 2.2 X 10-3

(-1889 A/C) HTN 0.52 0.10 0.55 7.92 X 10-6

NT 0.46 0.01 0.49 0.748 CYP11B2 Exon 3 rs4546 143996553 ALL T 0.435 0.49 0.78 0.429 0.53 2.88 X 10-6

HTN 0.48 0.54 0.56 1.76 X 10-9

NT 0.50 0.84 0.49 0.788 CYP11B2 Intron 2 Intron2 143996983-82 ALL Conv 0.467 0.49 0.19 0.455 0.53 1.0 X 10-3

HTN 0.51 0.32 0.56 1.91 X 10-7

NT 0.46 0.01 0.48 0.438 CYP11B2 Promoter rs1799998 143999600 ALL C 0.435 0.50 0.26 0.433 0.53 3.05 X 10-6

(-344 T/C) HTN 0.49 0.72 0.56 2.67 X 10-9

NT 0.51 0.20 0.5 0.78

BRIGHT Swedish

Location SNP Position bp Minor allele Obs.Het Obs.Het

A"

B"

10

Figure S2: Linkage disequilibrium plots for BRIGHT and SWEDISH populations, showing D’ and R2 values in each cell.

11

Figure S3: Flowchart summarising the method developed for measuring relative levels of transcripts originating from the alternative haplotype forms of either CYP11B1 (i.e. B1Hap1 or B1Hap2) or CYP11B2 (i.e. B2Hap1 or B2Hap2).

PCR of B1Hap1 and B1Hap2 cDNA using

common CYP11B1-specific primers that amplifies both forms with equal efficiency

Resulting mixture of B1Hap1 and B1Hap2 PCR products cloned into vector

Resulting B1-containing colonies picked and

assigned to Hap1 or Hap2 group by sequencing

Isolation of total RNA from heterozygous (Hap1/Hap2) adrenal tissue

Cells transformed with resulting B1-containing

vectors

Colonies grown on selective medium

RNA reverse-transcribed to cDNA

Ratio of B1Hap1:B1Hap2 colonies calculated; this

ratio is equal to the ratio of B1Hap1:B1Hap2 mRNA in

the original tissue

PCR of B2Hap1 and B2Hap2 cDNA using

common CYP11B2-specific primers that amplifies both forms with equal efficiency

Resulting mixture of B2Hap1 and B2Hap2 PCR products cloned into vector

Resulting B2-containing colonies picked and

assigned to Hap1 or Hap2 group by sequencing

Cells transformed with resulting B2-containing

vector

Colonies grown on selective medium

Ratio of B2Hap1:B2Hap2 colonies calculated; this

ratio is equal to the ratio of B1Hap1:B1Hap2 mRNA in

the original tissue

CYP11B1 method

CYP11B2 method

12

Figure S4: Verification of haplotype-specific measurement method. Plasmids containing cDNA corresponding to CYP11B1 (A) or CYP11B2 (B) Haplotype1 and Haplotype2 sequences were mixed together in varying, precise proportions and subjected to the method for measuring relative levels of haplotype-specific cDNAs. Plots represent observed Haplotype1-containing colony numbers as a percentage of total successfully genotyped colonies, versus the expected number of colonies based on the original plasmid mixture. Data are presented as the mean of three technical replicates (each genotyping a batch of 48 colonies) ± standard error of the mean. (A) Colonies resulting from mixes of CYP11B1 Haplotype1 and Haplotype2 plasmids; (B) Colonies arising from mixes of CYP11B2 Haplotype1 and Haplotype2 plasmids.

Expected Percentage of Haplotype1 Colonies

!"

#$"

$!"

%$"

&!!"

!" #$" $!" %$" &!!"

!"

#$"

$!"

%$"

&!!"

!" #$" $!" %$" &!!"Expected Percentage of Haplotype1 Colonies

Obs

erve

d P

erce

ntag

e of

Hap

loty

pe1

Col

onie

s

Obs

erve

d P

erce

ntag

e of

Hap

loty

pe1

Col

onie

s

A B