Supplementary File 1. Supplementary Methods and Results

Supplementary File 1. Supplementary Methods and Results

1. nAChR subunit gene candidate SNPs.

We selected ten pre-synaptic nAChR subunit genesfor resequencing as the most probable candidates for contribution to variation in nicotine dependence. We identified certain nAChR SNPs to ensure that amplicon design included coverage of these SNPs, if in coding or splice site regions, or to develop additional amplicons to capture these individual SNPs (genotyping by sequencing), based on recent findings (Table 1).

Table 1: nAChR subunit gene candidate SNP list

rsID / nAChR gene / Reference
rs2072660 / CHRNB2 / (Conti et al., 2008)
rs2072661 / CHRNB2 / (Conti et al., 2008)
rs4950 / CHRNB3 / (Zeiger et al., 2008)
rs4953 / CHRNB3 / (Zeiger et al., 2008)
rs6474413 / CHRNB3 / (Bierut et al., 2007)
rs13277254 / CHRNB3 / (Bierut et al., 2007)
rs2304297 / CHRNA6 / (Zeiger et al., 2008)
rs6495308 / CHRNA5 / (Berrettini et al., 2008)
rs16969968 / CHRNA5 / (Saccone et al., 2007)
rs578776 / CHRNA3 / (Bierut et al., 2007)
rs1051730 / CHRNA3 / (Saccone et al., 2007)
rs3813567 / CHRNB4 / (Saccone et al., 2007)
rs2236196 / CHRNA4 / (Hutchison et al., 2007)
rs6122429 / CHRNA4 / (Hutchison et al., 2007)

2. DNA, quantification and whole genome amplification.

DNA was extracted from saliva obtained from participants of a smoking cessation intervention (Nishita et al., 2009), and was quantified using the Quant-iT™ PicoGreen® dsDNA reagent (Invitrogen). Samples with <900ng of DNA were whole genome amplified (WGA) using the REPLI-g method (Qiagen) with a minimum of 100 ng of DNA input, according to the manufacturer’s protocol, prior to sequencing.

3. Primer design, sample preparation and 454 sequencing.

Primer design, sample preparation, and sequencing followed the protocol methodology supplied by Roche under Application Note No.5 “Amplicon sequencing”, the GS FLX Amplicon DNA Library Preparation Method Manual, the GS FLX emPCR Method Manual, the GS FLX Sequencing Method Manual, and the Genome Sequencer FLX Operator's Manual for the 454 Genome Sequencer System, with some slight modifications described below.

a. Amplicon design for 454 sequencing was attempted for the coding regions and splice sites for the candidate gene of 10 nAChR subunit genes, the intergenic region between CHRNA3 and CHRNA5, the 3’ UTRs of CHRNA3 and CHRNA5, and the 16 SNPs listed in Table 1. In addition, for quality control purposes, 10 amplicons containing 10 QC SNPs and a single Y chromosome amplicon were also designed (data not shown).Primers were designed using Primer3 software with general parameters of: maximum GC content less than 50%; primer size between 18 – 22 base pairs (bp); and amplicon size between 150 – 270 bp. Exceptions include 3 primers allowing GC content as high as 70% and primer length as short as 16 bp, and 12 amplicons were as long as 350 bp.

After this initial round of design, the amplicons were tested for successful amplification on a set of DNA samples, by separating the resulting PCR products on a 0.8% agarose gel and looking for the expected fragment size. It was determined that 10 amplicons were not performing successfully from the gel test. For these failing amplicons, primers were redesigned using relaxed criteria with respect to amplicon length and specificity. These redesigned primers were not tested for amplification. A summary of the final amplicon coverage for each gene or SNP can be found in Table 2 (data not shown for QC SNPs).

For each of the amplicon-specific primer pairs, sixteen different versions were designed, where to the 5’ end of each primer pair was added one of 16 unique 3-basepair tags. Each forward primer and reverse primer also contained the “A” and “B” 19-mer sequence required for the 454 sequencing method, as described in the Application Note No.5.

b. Each DNA sample was amplified with the set of amplicons using a set of primer pairs tagged with one of the 16 unique 3-bp tags in singleplex PCR reactions. The 6 µl PCR reaction conditions were as follows: 1X Advantage® HF 2 Polymerase Mix (Clontech), 1X HF 2 PCR Buffer (Clontech), 1X HF dNTP Mix (Clontech), 4 ng of DNA, and 1 µM of each primer. Thermocycling conditions: 95˚C 5m, (95˚C 10s, 58˚C 30s, 65˚C 60s) 10X with a decrease of 0.5˚C in the annealing temperature each cycle, then (95˚C 2s, 58˚C 45s, 65˚C 1.15m) 23X, 65˚C 15m, 4˚C hold. Following amplification, each PCR reaction was quantified using PicoGreen®, and then normalized based on molarity. Following normalization, all PCR products for a single DNA sample were pooled and purified using the NucleoFast 96 PCR Plates (Clontech) according to manufacturer’s instructions. The pooled, purified products for each DNA sample were quantified using PicoGreen® and normalized to a concentration of 4 ng/µl. Then, up to 16 different samples’ products, each amplified using a different unique 3-bp tagged primer set, were multiplexed prior to continuation of the protocol at the Ampure bead purification step as described in the GS FLX Amplicon DNA Library Preparation Method Manual. All further procedures followed the FLX 454 manuals without modification.

Table 2: Amplicons yielding PCR products of the expected size.

Gene or rsID / Target Length (bases) / AmpliconCount / Amplicon Tested* / Bases Covered / Bases Uncovered / Percent Covered
CHRNB2 / 1509 / 13 / 7 / 1086 / 423 / 72.0
CHRNA2 / 1590 / 12 / 11 / 1387 / 203 / 87.2
CHRNB3 / 1377 / 11 / 11 / 1377 / 0 / 100.0
CHRNA6 / 1485 / 11 / 11 / 1461 / 24 / 98.4
CHRNA7 / 1509 / 6 / 6 / 625 / 884 / 41.4
CHRNA5 / 1407 / 12 / 12 / 1392 / 15 / 98.9
CHRNA3/A5 / 3350 / 16 / 16 / 2818 / 532 / 84.1
CHRNA3 / 1518 / 15 / 14 / 1423 / 95 / 93.7
CHRNB4 / 1497 / 14 / 13 / 1417 / 80 / 94.7
CHRNB1 / 1506 / 18 / 0 / 1501 / 5 / 99.7
CHRNA4 / 1884 / 14 / 9 / 1136 / 748 / 60.3
rs2072660 / 1 / 1 / 1 / 1 / 0 / 100.0
rs2072661 / 1 / 1 / 1 / 1 / 0 / 100.0
rs4950 / 1 / 1 / 1 / 1 / 0 / 100.0
rs6474413 / 1 / 1 / 1 / 1 / 0 / 100.0
rs13277254 / 1 / 1 / 1 / 1 / 0 / 100.0
rs2304297 / 1 / 1 / 1 / 1 / 0 / 100.0
rs6495308 / 1 / 0 / 0 / 0 / 1 / 0.0
rs3813567 / 1 / 1 / 1 / 1 / 0 / 100.0
rs6122429 / 1 / 1 / 0 / 1 / 0 / 100.0
rs2236196 / 1 / 1 / 0 / 1 / 0 / 100.0

In addition, the bases represented by rs1051730, rs16969968, rs4953 and rs578776 were covered though not explicitly designed against because they were within a bigger design target, i.e., within a nAChR exon.

Table 3: Final 454 sequencing coverage statistics

Gene or rsID / Target Length (bases) / N Amplicons / Observed Final Bases Covered (>5x on >80% of samples) / % Final Bases Covered (>5X) / Observed Final Bases Covered (>10x on >80% of samples) / % Final Bases Covered (>10X) / Observed Common Variants (MAF ≥5%) / Observed Rare Variants (1%< MAF <5%) / Observed Rare Variants (MAF <1%)
CHRNB2 / 1509 / 13 / 697 / 46.19 / 326 / 21.6 / 0 / 0 / 3
CHRNA2 / 1590 / 12 / 1410 / 88.68 / 1280 / 80.5 / 3 / 2 / 9
CHRNB3 / 1377 / 11 / 1229 / 89.25 / 1229 / 89.25 / 0 / 2 / 4
CHRNA6 / 1485 / 11 / 1485 / 100 / 1250 / 84.18 / 0 / 0 / 2
CHRNA7 / 1037 / 6 / 450 / 43.39 / 405 / 39.05 / 1 / 0 / 2
CHRNA5 / 1245 / 10 / 885 / 71.08 / 885 / 71.08 / 1 / 1 / 5
CHRNA3/CHRNA5 / 3641 / 20 / 2967 / 81.49 / 2967 / 81.49 / 7 / 1 / 16
CHRNA3 / 1389 / 13 / 1307 / 94.1 / 1073 / 77.25 / 3 / 3 / 8
CHRNB4 / 1497 / 14 / 1442 / 96.33 / 1412 / 94.32 / 0 / 4 / 11
CHRNB1 / 1506 / 18 / 1115 / 74.04 / 1057 / 70.19 / 1 / 0 / 5
CHRNA4 / 1808 / 14 / 0 / 0 / 0 / 0 / 5 / 0 / 8
rs2072660 / 1 / 1 / 1 / 100 / 1 / 100 / 1 / 0 / 0
rs2072661 / 1 / 1 / 1 / 100 / 1 / 100 / 1 / 0 / 0
rs4950 / 1 / 1 / 1 / 100 / 1 / 100 / 1 / 0 / 0
rs6474413 / 1 / 1 / 1 / 100 / 1 / 100 / 1 / 0 / 0
rs13277254 / 1 / 1 / 1 / 100 / 1 / 100 / 1 / 0 / 0
rs2304297 / 1 / 1 / 1 / 100 / 1 / 100 / 1 / 0 / 0
rs3813567 / 1 / 1 / 0 / 0 / 0 / 0 / 1 / 0 / 0
rs2236196 / 1 / 1 / 0 / 0 / 0 / 0 / 1 / 0 / 0
rs6122429 / 1 / 1 / 0 / 0 / 0 / 0 / 1 / 0 / 0

4. Sanger Resequencing.

DNA extracted from saliva samples was amplified using Phire Hot Start DNA Polymerase (manufactured by Finnzymes, distributed by NEB) and primers developed by Weiss et al (Weiss et al., 2008) for CHRNB2 Exons 1 and 5 and CHRNA4 Exons 2, 5 and 6 (regions with <10X read depth from 454 sequencing). Five primers were redesigned after a continued lack of success, allowing sequencing of Exon 2 in the forward direction and of Exon 5 in the 3’ and 5’ regions (Table 5). Amplification of CHRNA4 Exons 1, 3 and 4 was attempted with no success. The 20µl PCR mix included 50ng DNA, 1.0U Phire™ Hot Start DNA Polymerase, 1x Phire™ Reaction Buffer, 200µM dNTPs (Applied Biosystems), 0.5µM each primer, 1x BSA and 3% DMSO (final concentrations). The PCR conditions were 1 cycle at 98°C for 30 sec, 30 cycles at 98°C for 5 sec, 65°C for 5 sec and 72°C for 20 sec followed by 1 cycle of 72°C for 1 min. PCR products were purified using ExoSAP-IT® (USB Corporation) with purification conditions of 1 cycle at 37°C for 30 mins followed by 1 cycle at 85°C for 15 mins. Cycle sequencing used Big Dye™ Terminator v3.1(Applied Biosystems) in 1/16 reactions using 5 pmol/µl of primer and 1µl of purified PCR product. Cycle sequencing reactions were cleaned using the Ethanol/Sodium Acetate/EDTA method outlined by Applied Biosystems in the Big Dye™ instructions. Sequencing was performed in formamide using a 3130xl Genetic Analyzer (Applied Biosystem). Unsuccessful samples were repeated using wgaDNA and the same PCR and cycle sequencing conditions. In the case ofCHRNA4 Exons 2 and 5 neither wgaDNA nor genomic DNA gave good Sanger sequence. These individuals were subsequently successfully sequenced using genomic DNA and 1/8 Big Dye cycle sequencing reactions. Variant detection was performed using Mutation Surveyor (v3.24 SoftGenetics), RefSeq NM_000748.2 for CHRNB2 and NM_000744.5 for CHRNA4 and reference files generated in Mutation Surveyor using the RefSeq and the “generate synthetic chromatograms” function. The Normal Allele (NM) score calculated by Mutation Surveyor for the SNPs found by Sanger sequencing is presented in Table 6. This score is calculated using the height of expected allele in sample file (the expected allele is taken from the reference file), the deviation,which measures peak shape of fourbases on either side of the center of each peak, and the combined contribution of peak height and peak shape. A low NM score indicates a large difference between the reference file and the sample file (0 indicates a homozygote alternate and heterozygotes are generally between 0.5 and 4, David Hulce, Mutation Surveyor, pers. comm.), and a large NM score (close to 100) shows a high level of similarity between the reference file and sample file. Any score falling in the middle of the range indicates a large amount on noise or error. A new SNP was reported only when found in both reads per subject and when it was in a region covered by the reads from all other subjects.

In the case of five SNPs (rs56175056, ss196003120, rs45442394, rs3827020, ss196003134), it was possible to compare genotypes from 454 to Sanger resequencing. Overall, mean (SD) levels of missing, concordant and discordant genotype data were 16.3% (6.9%), 74.5% (19.9%) and 9.2% (13.6%), with a total of 1800 genotypes available in each dataset to evaluate concordance. Only the two common SNPs exhibited discordant genotype data, with 96% of discordant genotypes representing major allele homozygote/heterozygote differences. In the case of two SNPs (rs2072660 and rs2072661), it was possible to compare genotype results from 454 to Taqman from a previous project. 860 genotypes were examinedand there was concordance between the two methods in all genotypes, except for one individual for whom rs2072660 was called as a heterozygote by Taqman and as a homozygote for Allele 2 by 454.
Table 4: Primers for PCR (P) or Sanger (S) sequencing of CHRNB2 and CHRNA4

Primer Name / Primer Sequence / P, S / Target Region / Source
304_CHRNB2_1_L / CAGGTAGACCCTCAGTGGGA / P / CHRNB2 Exon 1 / Weiss et al 2008
305_CHRNB2_1_R / CTTCTTCCCTTCCAAGCCTC / P / CHRNB2 Exon 1 / Weiss et al 2008
328_CHRNB2_1_F / AACAGCCCTTCTTCCACCTC / S / CHRNB2 Exon 1 / Weiss et al 2008
329_CHRNB2_1_R / AATCCCTTCTTCCACCTTCC / S / CHRNB2 Exon 1 / Weiss et al 2008
310_CHRNB2_5c_L / AAGGTGTCACACAGGCCTTC / P / CHRNB2 Exon 5 / Weiss et al 2008
311_CHRNB2_5c_R / CATCTTCCACCGACTCCAAT / P / CHRNB2 Exon 5 / Weiss et al 2008
334_CHRNB2_5a_F / CTGGGTTGATGGGTAAGGAG / S / CHRNB2 Exon 5 5' / Weiss et al 2008
335_CHRNB2_5a_R / CACGATCTTGGAGATGAGCA / S / CHRNB2 Exon 5 5' / Weiss et al 2008
336_CHRNB2_5b_F / TATGACTTCATCATTCGCCG / S / CHRNB2 Exon 5 center / Weiss et al 2008
337_CHRNB2_5b_R / CCACTGGTGCAGGCTCAG / S / CHRNB2 Exon 5 center / Weiss et al 2008
338_CHRNB2_5c_F / CTCGTCGGCAAGTACCTCAT / S / CHRNB2 Exon 5 3' / Weiss et al 2008
339_CHRNB2_5c_R / CAACACAGGCCTGACTCACA / S / CHRNB2 Exon 5 3' / Weiss et al 2008
301_CHRNA4_2_L / TGCATCTGAAACACTCCTCG / P / CHRNA4 Exon 2 / Weiss et al 2008
300_CHRNA4_2_R / CCTGGTTCCTATCCTGCCTT / P / CHRNA4 Exon 2 / Weiss et al 2008
392_CHRNA4_E2_F / GCTCATGTGAGGAGCTGAAA / S / CHRNA4 Exon 2 / This study
317_CHRNA4_2_R / ACTTGGAGTGTTCCCACGTC / S / CHRNA4 Exon 2 / Weiss et al 2008
390_CHRNA4_E5_F / GGGTAATCACCAGCAGCAGT / P / CHRNA4 Exon 5 / This study
387_CHRNA4_E5_R / ACTGTGTCCCAAAGCCAGAG / P / CHRNA4 Exon 5 / This study
441_CHRNA4_E5_S_F / GTCTCACACCCTTCGCTCTC / S / CHRNA4 Exon 5 5' / This study
322_CHRNA4_5a_R / ATGGTGTAGAAGAGCGGCAG / S / CHRNA4 Exon 5 5' / Weiss et al 2008
323_CHRNA4_5b_F / CATCACCTATGCCTTCGTCA / S / CHRNA4 Exon 5 center / Weiss et al 2008
324_CHRNA4_5b_R / CACGGAGAGGGCTGGTCT / S / CHRNA4 Exon 5 center / Weiss et al 2008
325_CHRNA4_5c_F / TGGAAGCTGAGAAAGCCAG / S / CHRNA4 Exon 5 3' / Weiss et al 2008
442_CHRNA4_E5_S_R / CCTGGATTACACACCAGGAAG / S / CHRNA4 Exon 5 3' / This study
295_CHRNA4_6_L / aGCTCCATGTGTGCTGCTTT / P / CHRNA4 Exon 6 / Weiss et al 2008
294_CHRNA4_6_R / aGGAAGGCAGAGTCCTGGTC / P / CHRNA4 Exon 6 / Weiss et al 2008
326_CHRNA4_6_F / GAAGGTGGGAGGTGGAGAAG / S / CHRNA4 Exon 6 / Weiss et al 2008
327_CHRNA4_6_R / TGCTGAGAGACTGGGAGGAG / S / CHRNA4 Exon 6 / Weiss et al 2008

Table 5: Sanger sequencing coverage statistics

Gene Region / Average % of read region covered by 2x / % Samples covered at 2x
CHRNA4 Exon 2 / 80.90 / 93.08
CHRNA4 Exon 5 5' / 70.77 / 87.28
CHRNA4 Exon 5 center / 86.12 / 92.19
CHRNA4 Exon 5 3' / 80.56 / 79.46
CHRNA4 Exon 6 / 96.19 / 97.54
CHRNB2 Exon 1 / 81.77 / 98.66
CHRNB2 Exon 5 5' / 85.85 / 98.44
CHRNB2 Exon 5 center / 85.55 / 98.88
CHRNB2 Exon 5 3' / 87.96 / 98.66

Table 6: Mean Sanger sequencing NM scores for CHRNB2 and CHRNA4SNPs

SNP ID / Homozygote NM_Score / Heterozygote NM_Score
CHRNA4
ss196003137 / n/a* / 1.44
rs56328826 / n/a / 1.2
ss196003135 / n/a / 0.76
ss196003134 / n/a / 1.33
ss196003133 / n/a / 1.38
rs3827020 / 0.03 / 1.69
rs45442394 / 0.3 / 0.7
ss196003130 / n/a / 0.74
rs1044397 / 0.02 / 1.26
rs1044396 / 0.01 / 1.23
ss196003127 / n/a / 0.68
rs45569837 / n/a / 1.33
ss196003125 / n/a / 1.59
rs55915440 / 0.06 / 0.56
rs2229960 / 0.04 / 0.86
rs2229959 / 0.01 / 0.68
ss196003121 / n/a / 0.39
ss196003120 / n/a / 0.82
rs56175056 / n/a / 0.88
ss196003118 / n/a / 1.75
rs1044394 / 0.03 / 8.92
rs1044393 / 0.03 / 0.82
ss196003115 / n/a / 0.74
ss196003114 / n/a / 0.99
ss196003113 / n/a / 1.21
ss196003159 / n/a / 0.97
ss196003158 / n/a / 0.68
ss196003157 / n/a / 1.14
rs6090384 / 0.03 / 0.7
rs2273505 / 0.02 / 1.22
rs2273506 / 0.04 / 1.28
ss196003109 / n/a / 0.91
CHRNB2
ss196003138 / n/a / 1.66
rs12062049 / n/a / 1.03
ss196003140 / n/a / 1.38
rs56205574 / n/a / 1.27
rs2280781 / 0.19 / 0.78
ss196003143 / n/a / 1.06
ss196003144 / n/a / 2.04
ss196003145 / n/a / 0.85
ss196003146 / n/a / 1.15
ss196003147 / n/a / 0.85
rs55685423 / n/a / 0.78
rs55857552 / n/a / 1.05
ss196003150 / n/a / 0.51
rs4845378 / 0.08 / 1.26
rs56146528 / n/a / 1.48

*not observed

Figure Legend

Supplementary Figure 1. 454 and Sanger Resequencing Amplicons and Depth. 454 icons: light gray <5x, medium gray ≤ 10x, dark gray ≤ 20x , black > 20X. Sanger icons: gray 1-2X, black > 2x.

References

Berrettini W, Yuan X, Tozzi F, Song K, Francks C, Chilcoat H et al (2008). Alpha-5/alpha-3 nicotinic receptor subunit alleles increase risk for heavy smoking. Mol Psychiatry13: 368-73.

Bierut LJ, Madden PA, Breslau N, Johnson EO, Hatsukami D, Pomerleau OF et al (2007). Novel genes identified in a high-density genome wide association study for nicotine dependence. Hum Mol Genet16: 24-35.

Conti DV, Lee W, Li D, Liu J, Van Den Berg D, Thomas PD et al (2008). Nicotinic acetylcholine receptor beta2 subunit gene implicated in a systems-based candidate gene study of smoking cessation. Hum Mol Genet17: 2834-48.

Hutchison KE, Allen DL, Filbey FM, Jepson C, Lerman C, Benowitz NL et al (2007). CHRNA4 and tobacco dependence: from gene regulation to treatment outcome. Arch Gen Psychiatry64: 1078-86.

Nishita DM, Jack LM, McElroy M, McClure JB, Richards J, Swan GE et al (2009). Clinical trial participant characteristics and saliva and DNA metrics. BMC Med Res Methodol9: 71.

Saccone SF, Hinrichs AL, Saccone NL, Chase GA, Konvicka K, Madden PA et al (2007). Cholinergic nicotinic receptor genes implicated in a nicotine dependence association study targeting 348 candidate genes with 3713 SNPs. Hum Mol Genet16: 36-49.

Zeiger JS, Haberstick BC, Schlaepfer I, Collins AC, Corley RP, Crowley TJ et al (2008). The neuronal nicotinic receptor subunit genes (CHRNA6 and CHRNB3) are associated with subjective responses to tobacco. Hum Mol Genet17: 724-34.