Dissection of genetic architecture of nine hazardous component traits of mainstream smoke in tobacco (Nicotiana tabacum L.)

Manling Xu, Zhijun Tong, Chengting Jin, Qixin Zhang, Feng Lin, Dunhuang Fang, Xuejun Chen, Tianneng Zhu, Xiang‐Yang Lou, Bingguang Xiao, Haiming XuFrontiers in Plant Science 2024 · DOI

This study, published in Frontiers in Plant Science, provides a high-resolution look at the "genetic blueprint" behind the harmful chemicals found in tobacco smoke. By mapping the genetic architecture of nine hazardous traits across multiple environments, the researchers identified 19 novel genetic regions (QTLs) and several candidate genes that could allow breeders to develop tobacco varieties with significantly lower toxicity.

TL;DR

Researchers used a high-density genetic map and 271 recombinant inbred lines to identify 19 novel genetic regions (QTLs) controlling hazardous smoke components in tobacco. The study reveals that while environment plays a massive role in toxin levels, specific pleiotropic genes (like Nt21g04598.1) offer clear targets for breeding "harm-reduction" tobacco varieties.

What Problem Does This Paper Address?

Tobacco smoke is a complex mixture of over 8,000 compounds, many of which are toxic or carcinogenic. While public health efforts focus on cessation, the tobacco industry and agricultural scientists also pursue "harm reduction" by trying to breed tobacco plants that naturally produce fewer of these hazardous precursors.

The challenge is that these traits are "quantitative"—they are controlled by many genes acting together, and their expression changes based on the environment (soil, weather, etc.). Previous attempts to map these genes were limited by low-resolution genetic maps. This paper addresses this gap by using a high-density SNP-based map to pinpoint the specific genomic "addresses" (Quantitative Trait Loci, or QTLs) responsible for smoke toxicity.

Methods

The authors employed a sophisticated genetic mapping approach:

  • Population Construction: They created 271 Recombinant Inbred Lines (RILs) from a cross between two elite varieties: K326 (a common US flue-cured variety) and Y3 (a variety from Zimbabwe known for elite agronomic traits).
  • Multi-Environment Testing: The lines were grown over three years (2018–2020) in Shilin, China, to account for Genotype-by-Environment (GE) interactions.
  • Phenotyping: Mainstream smoke was analyzed for nine traits: Benzo[a]pyrene (B[a]P), Carbon Monoxide (CO), Crotonaldehyde (CRO), Hydrogen Cyanide (HCN), Ammonia (NH3), NNK, Phenol (PHE), Tar (TAR), and a composite Hazard Index (HI).
  • QTL Mapping: They used a Full QTL Model (via QTLNetwork 2.0) that accounts for additive effects, epistasis (gene-gene interactions), and environmental interactions.
  • Bioinformatics: To move from a genetic region to a specific gene, they used BLAST, SnpEff for variant annotation, and GO/KEGG enrichment analysis to identify candidate genes within the QTL intervals.

Datasets

  • Genomic Data: A high-resolution linkage map containing 46,324 markers (mostly SNPs and Indels) classified into 7,107 bins across 24 linkage groups.
  • Phenotypic Data: Three years of field trial data measuring the nine hazardous smoke components across the 271 RILs.
  • Reference Genome: The Nicotiana tabacum K326 reference genome was used for candidate gene prediction.

Key Insights and Ideas

  • The Environment Dominates: One of the most striking findings was that "interaction heritability" ($h^2_{ge}$) was over 80% for most traits. This means that a plant's genetic potential to produce toxins is highly sensitive to the environment. Breeding for stable low-toxin varieties requires testing across many locations.
  • Pleiotropy is a Shortcut: The researchers found that several QTLs for different toxins overlapped (e.g., the same region on Linkage Group 15 controlled both Phenol and Tar). This "pleiotropy" suggests that breeders can reduce multiple harmful chemicals simultaneously by targeting a single genetic locus.
  • Y3 as a Genetic Goldmine: Most alleles that decreased hazardous components came from the Y3 parent. This identifies Y3 as a critical genetic resource for future harm-reduction breeding programs.
  • Metabolic Links: A high genetic correlation (0.84) was found between Carbon Monoxide and Crotonaldehyde, suggesting these two disparate chemicals may share a common regulatory or combustion-related pathway.

Key Results

  • 19 Novel QTLs: Detected across eight of the nine traits. 17 of these showed significant additive effects (direct genetic control).
  • Candidate Genes: Six high-priority candidate genes were identified. Notably, Nt21g04598.1, Nt21g04600.1, and Nt21g04601.1 were predicted to have pleiotropic effects on both Phenol (PHE) and Tar (TAR).
  • Heritability: While the stable genetic heritability ($h^2_g$) was relatively low (11%–20%), the total heritability (including environmental interactions) exceeded 95%, indicating that nearly all variation in smoke toxicity can be explained if the environment is accounted for.
  • Superior Genotypes: The study designed "predicted superior genotypes" (GSLs) that combine the best alleles from both parents, providing a roadmap for creating a "low-hazard" tobacco line.

Limitations and Future Work

  • QTL Resolution: While the map is high-density, the identified QTL regions still contain multiple genes. Fine mapping is required to confirm which specific gene is the "causative" one.
  • Functional Validation: The candidate genes were identified via bioinformatics. Future work should use CRISPR/Cas9 gene editing or RNA interference (RNAi) to knock out these genes and prove they actually reduce smoke toxins.
  • Combustion Physics: The study focuses on chemical precursors in the leaf. However, smoke composition is also affected by the physical structure of the leaf and how it burns. Future models might need to integrate leaf porosity and burn rate as co-factors.

Summary updated

Abstract

Tobacco (<i>Nicotiana tabacum</i> L.) use is the leading cause of preventable death, due to deleterious chemical components and smoke from tobacco products, and therefore reducing harmful chemical components in tobacco is one of the crucial tobacco breeding targets. However, due to complexity of tobacco smoke and unavailability of high-density genetic maps, the genetic architecture of representative hazardous smoke has not been fully dissected. The present study aimed to explore the genetic architecture of nine hazardous component traits of mainstream smoke through QTL mapping using 271 recombinant inbred lines (RILs) derived from K326 and Y3 in multiple environments. The analysis of genotype and genotype by environment interaction (<i>GE</i>) revealed substantially greater heritability over 95% contributed mostly by <i>GE</i> interaction effects. We also observed strong genetic correlations among most studied hazardous smoke traits, with the highest correlation coefficient of 0.84 between carbon monoxide and crotonaldehyde. Based on a published high-density genetic map, a total of 19 novel QTLs were detected for eight traits using a full QTL model, of which 17 QTLs showed significant additive effects, six showed significant additive-by-environment interaction effects, and one pair showed significant epistasis-by-environment interaction effect. Bioinformatics analysis of sequence in QTL region predicted six genes as candidates for four traits, of which <i>Nt21g04598.1</i>, <i>Nt21g04600.1</i>, and <i>Nt21g04601.1</i> had pleiotropic effects on PHE and TAR.
BibTeX
@inproceedings{xu2024dissection,
  author = {Manling Xu and Zhijun Tong and Chengting Jin and Qixin Zhang and Feng Lin and Dunhuang Fang and Xuejun Chen and Tianneng Zhu and Xiang‐Yang Lou and Bingguang Xiao and Haiming Xu},
  title = {{Dissection of genetic architecture of nine hazardous component traits of mainstream smoke in tobacco (Nicotiana tabacum L.)}},
  booktitle = {Frontiers in Plant Science},
  year = {2024},
  doi = {10.3389/fpls.2024.1358953},
}
Figure