Genotypes in Smokers: Correlations with Smoking Behavior

Peter Bauer1*, Susan Collins2, and Anil Batra2
1Department of Medical Genetics, University of Tübingen, Germany; 2Clinic for Psychiatry and Psychotherapy, University of Tübingen, Germany

*Corresponding author

Introduction

Smoking behavior is influenced by both genetic and environmental factors. Moreover, smoking initiation and smoking persistence have a heritability of at least 50%. A large body of work has been dedicated to associating multiple genetic markers of neurobiological pathways previously linked to addiction and smoking such as the central dopaminergic, serotoninergic, and nicotinergic pathways with different aspects of smoking behavior. In order to investigate the feasability of future genetic analyses, we genotyped 14 SNPs in 288 samples of addicted smokers using the LightCycler® 480 System with HybProbe assays.

Materials and Methods

Blood samples were taken from 272 volunteers who had given informed consent for analysis of target genes presumably involved in nicotine addiction. DNA was prepared using either a manual high-salt method (236 samples) or the MagNA Pure Compact Instrument with the Nucleic Acid Isolation Kit I (36 samples).

PCR primers and HybProbe probes for the analysis of 14 different SNPs (Figure 1) were designed by Tib Molbiol (Berlin). A BioRobot 8000 platform was programmed to perform PCR setups using the LightCycler® 480 Genotyping Master in 384-well microtiter plates. Details of the PCR reaction are given elsewhere [1]. In brief, we used 10 pmol of each primer and 3 pmol of each HybProbe probe (anchor and sensor). PCR volumes were 10 µl, and we amplified 55 cycles with a touchdown PCR profile (lowering the annealing temperature from 65°C in the first cycle to 55°C in the 10th cycle in 1°C/cycle steps).

Optimizing genotyping assays

Overall, we achieved robust genotyping for 12 of 14 genotyping assays with our standard approach (Figure 1), relying mainly on a touchdown PCR protocol and asymmetric amplification (preferentially amplifying the PCR strand complementary to the anchor and sensor probes). For two assays (COMT and MAOA), we had to take additional measures in order to get good genotyping quality.

COMT V158M polymorphism

The standard approach gave very uneven melting peaks for the mismatch (low temperature melting) and the perfect match (high temperature melting) alleles (Figure 2). Changes in primer or HybProbe concentrations did not improve these patterns. At this stage, we were able to distinguish only between homozygous BB genotypes (in purple) in all other genotypes (AA and AB; yellow and blue samples) and thus had lost discrimination power. Finally, we decided to move the unlabeled forward primer 50 bases upstream and, by this simple measure, greatly improved the robustness of the genotyping assay. Re-genotyping gave us reliable genotypes for 268 samples (1.5% technical failures).

MAOA polymorphism

In this case, again, melting curves for the high temperature melting peak were polymorphic and did not allow accurate grouping of genotypes (Figure 2). A skewed melting profile did not reliably distinguish between AB genotypes (yellow traces) and BB genotypes (red traces). There, we observed much better melting profiles in 20 µl volume assays and therefore re-genotyped a technically identical assay in a doubled PCR volume. Likewise, we were able to reliably genotype all samples for MAOA (1.9% technical failure probably due to bad DNA quality).

Results and Discussion

We have genotyped 14 SNPs in a cohort of 272 smokers. Overall performance of the system was high with drop-out rates below 3%. Three SNPs (SLC6A4, HTR2A, RAPGEF) were mis-annotated in the public databases and did not show heterozygosity although the PCR performance was very robust. MAOA and COMT genotyping had to be improved by shifting the unlabeled forward primer or increasing the reaction volume from 10 µl to 20 µl.

Eleven SNPs were polymorphic in our cohort. For these 11 SNPs, genotype counts did not significantly deviate from Hardy-Weinberg equilibrium. Therefore, the accuracy of the genotyping system is high. Moreover, we validated SNPs for ANKK1 by an independent method (RFLP; data not shown) giving exactly identical genotypes. The success rate for genotyping in this study was more than 95% (Table 1), with no evidence for a technical performance failure of the LightCycler® 480 Instrument.

Conclusions

Our genotyping data indicate that analyses of genetic variation in the dopaminergic system are feasible using the LightCycler® 480 Instrument with HybProbes. This new high-throughput technology has enabled more reliable genotyping and thus more accurate analyses of genetic variants and smoking behavior. Future studies will report on these clinically relevant associations.

 

 

 

References

1. Walter et al. (2006) Biochemica 2:8–11

Acknowledgements

We thank IS Mäckle-Jentsch and Claudia Bauer for their excellent technical assistance.

 

This article was originally published in Biochemica 3/2007, pages 10-12. ©Springer Medizin Verlag 2007

Facts, background information, dossiers
  • addiction
  • Weinberg
  • TIB Molbiol
More about Roche Diagnostics
Your browser is not current. Microsoft Internet Explorer 6.0 does not support some functions on Chemie.DE