loading page

QTL x environment interactions for panicle traits in switchgrass (Panicum virgatum)
  • +10
  • Li Zhang,
  • Xiaoyu Weng,
  • Kathrine Behrman,
  • Jason Bonnette,
  • John Reilley,
  • Francis Rouquette,
  • Phillip Fay,
  • Yanqi Wu,
  • Felix Fritschi,
  • Robert Mitchell,
  • David Lowry,
  • Arvid Boe,
  • Thomas Juenger
Li Zhang
The University of Texas at Austin
Author Profile
Xiaoyu Weng
The University of Texas at Austin
Author Profile
Kathrine Behrman
The University of Texas at Austin
Author Profile
Jason Bonnette
The University of Texas at Austin
Author Profile
John Reilley
USDA National Resources Conservation Service
Author Profile
Francis Rouquette
Texas A&M University Agricultural Research and Extension Center at Overton
Author Profile
Phillip Fay
USDA-ARS Grassland Soil and Water Research Laboratory
Author Profile
Yanqi Wu
Oklahoma State University Stillwater
Author Profile
Felix Fritschi
University of Missouri System
Author Profile
Robert Mitchell
USDA-ARS Wheat Sorghum and Forage Research
Author Profile
David Lowry
Michigan State University
Author Profile
Arvid Boe
South Dakota State University
Author Profile
Thomas Juenger
The University of Texas at Austin
Author Profile

Abstract

Panicle traits exhibit quantitative variation controlled by genes, the environment, and their interaction. In switchgrass, a perennial biofuel crop, identification of quantitative trait loci (QTL) and QTL x E interactions controlling panicle architecture could aid breeding efforts and cultivar development by impacting seed productivity. In this study, we evaluate the genetic architecture of panicle traits including panicle length, primary branching number, and secondary branching number in an outcrossing switchgrass population grown across ten field sites in the central United States. We evaluated pleiotropic relationships between panicle traits and flowering time, tiller production and biomass. We also identified environmental factors correlated with QTL x E interactions and potential candidate genes underlying panicle trait QTL in switchgrass. Overall, our multi-environment mixed QTL model detected 18 QTL for panicle traits. Twelve of the QTL exhibited consistent effects (i.e., no QTL x E), and most (4 of 6) of the effects with QTL x E exhibited condition-specific effects. Many of the QTL x E effects were associated with yearly mean temperature and photoperiod. Panicle QTL co-localized with previously identified flowering time QTL and candidate genes associated with flowering, supporting a pleiotropic model of panicle development based on shared developmental genetics and responses to environmental signals.