Phenotypic and Genetic Variation for Rhizosphere Acidification, a Candidate Trait for pH Adaptability, in Deciduous Azalea (Rhododendron sect. Pentanthera)

Abstract

In breeding ornamental plants, selection for aesthetic qualities have predominated with less focus on improving certain physiological traits such as pH tolerance. Deciduous azaleas (Rhododendron sect. Pentanthera) are a prime example, where selections from the Woody Plant Breeding Program at the University of Minnesota have introduced 16 cold-hardy, ornamental plants but lack tolerance to high pH and calcareous soils common in many parts of the United States. High pH stress typically manifests in deciduous azalea as iron deficiency chlorosis; a yellowing of plant leaves due to reduced photosynthetic capacity. Chlorosis results in diminished aesthetic appeal and ultimately can lead to plant death. We present research on the quantitative trait variation for pH adaptability in deciduous azaleas; focusing on wild and cultivated populations, as well as novel phenotyping approaches to get a better picture of the phenotypic and genetic variation for pH adaptability in each. The Woody Plant Breeding program initiated its azalea breeding program in 1954 and has kept crossing records ever since. Unfortunately, the location and habitat details for many of the parents of popular cultivars are unavailable. We first categorize one North American species, Rhododendron viscosum (L.) Torr., which is commonly used as a parent in crosses for its white flowers, late season flowering and also for it’s potential to contribute stress adaptation. We developed 14 SSR loci from a sequenced transcriptome of a wild collected R. viscosum plant to help us estimate subpopulation differentiation and admixture throughout the southeastern U.S. range for the species for eventual comparison with trait additive variation. Population level Rst values (0.49) indicated a strongly differentiated population, and STRUCTURE analysis revealed limited admixture between subpopulations in the western part of the R. viscosum range. Next, we present the development of an in vitro phenotyping assay designed to expedite the Rhododendron breeding process and measure rhizosphere acidification, a candidate trait for pH adaptability. Increased iron solubility in soil and root tissue is critical to mitigating iron deficiency chlorosis, though iron solubility is limited in high pH and or calcareous soils. One way that plants make this iron reduction reaction more favorable is through rhizosphere acidification, resulting in lower soil pH near roots allowing newly soluble iron to be reduced and transported into root tissues. New image analysis methods using the technical programming language MATLAB (MathWorks, Inc.) were developed to measure and quantify rhizosphere acidification in tissue-cultured Rhododendron seedlings. Germination of immature Rhododendron seed in vitro was significantly enhanced by gibberellic acid (GA3) applications, resulting in progeny that were screenable for pH tolerance as early as 5 months after crossing as opposed to the usual 8 months. We were able to colorimetrically detect rhizosphere acidification for plants grown in tissue culture media, which correlated with root mass in the breeding populations tested. However, using root mass as the sole predictor of the variation in rhizosphere acidification failed to replicate the observed phenotypic variation across these breeding populations. This indicated that other sources of variation, possibly genetic, could explain a significant portion of the phenotypic variation in rhizosphere acidification. Finally, we employed our in vitro assay to phenotype progeny from two separate deciduous azalea mating designs: a factorial crossing scheme consisting of advanced selections from the UMN breeding program and half sib families from different R. viscosum subpopulations sampled throughout the southern United States. We also developed another MATLAB script to quantify growth rates and leaf color in seedlings of the factorial crossing scheme to correlate with rhizosphere acidification data for each family. Rhizosphere acidification had a narrow-sense heritability of 0.38 in the factorial design and 0.83 in the half-sib design when measured over three calcium carbonate liming rates. The additive variation for rhizosphere acidification was significant in the half sib design, though it was not so in the factorial design. The effect of an endophytic bacterium infection on rhizosphere acidification was found to be significant where present among half sib families derived from Arkansas and Texas subpopulations. Comparisons of trait to molecular variation within and among populations for rhizosphere acidification in R. viscosum revealed that rhizosphere acidification was not locally adapted, as the trait Qst value was 0.20, lower than our molecular differentiation estimate (Rst) of 0.49. Rhizosphere acidification quantified in vitro correlated positively with the growth rate of soil grown seedlings measured in the greenhouse (R2 = 0.49), with families possessing the highest rhizosphere acidification also having the highest growth rates. We conclude that increased rhizosphere acidification in deciduous azalea improves adaptability to high pH soils, and can be screened relatively early to expedite the breeding process of these popular woody ornamental plants.