Local adaptation, genetic diversity and key environmental interactions in a collection of novel red clover germplasm populations

Angus Donald Heslop ^1* Sai K Arojju ² Rainer Hofmann ³ John L Ford ⁴ Zulfi Z Jahufer ⁵ Anna C Larking ⁶ Rachael Ashby ⁷ Charles Hefer ¹ Ken G Dodds ⁷ A Saei ⁶ Jessica O'connor ⁷ Andrew G Griffiths ⁶

• ¹ AgResearch Ltd, Christchurch, New Zealand
• ² Radiata Pine Breeding Company, Christchurch, New Zealand
• ³ Faculty of Agriculture and Life Sciences, Lincoln University, Lincoln, Canterbury, New Zealand
• ⁴ PGG Wrightson Seeds Limited, Palmertson North, New Zealand
• ⁵ School of Agriculture and Food Sciences, Faculty of Science, The University of Queensland, Gatton, Queensland, Australia
• ⁶ AgResearch Limited, Palmerston North, New Zealand
• ⁷ AgResearch Ltd, Mosgiel, New Zealand

Red clover (Trifolium pratense L.) is known for its large taproot, nitrogen fixation capabilities and production of forage high in protein and digestibility. It has the potential to strengthen temperate pastural systems against future adverse climatic events by providing higher biomass during periods of water deficit. Being outcrossing and self-incompatible, red clover is a highly heterozygous species. If evaluated and utilized correctly, this genetic diversity can be harnessed to develop productive, persistent cultivars. In this study, using multi-location, multi-year field trials and genotyping-by-sequencing (GBS), we were able to assess genetic diversity and genetic relationships of 92 geographically diverse red clover germplasm populations. Through the development of integrated linear mixed models based on genomic, phenotypic, and environmental information we assessed variance components and genotype-by-environment (G x E) interactions for eight physiological and morphological traits. Key interactions between environmental variables and plant performance were also evaluated using a common garden site at Lincoln. We found that the genetic structure of the 92 populations was highly influenced by country of origin. The expected heterozygosity within populations ranged between 0.08 and 0.17 and varied with geographical origin. For the eight physiological and morphological traits measured there was high narrow-sense heritability (h2 > 0.70). The influence of environmental variables, such as mean precipitation, temperature and isothermality of the original collection locations, on plant and trait performance in the local field trials was also highlighted. Along with the identification of genes associated with these bioclimatic variables that could be used as genetic markers for selection in future breeding programs. Our study identifies the importance of diverse germplasm when adding genetic variation into breeding programs. It also identifies efficient evaluation methods and key climatic variables that should be considered when developing adaptive red clover cultivars.