Near-infrared spectra (NIRS) from plant tissues can be used to predict traits owing to their relationship to internal biochemical states, shaped by both environmental and genetic components. Here, we tested the use of NIRS as predictors of budbreak the following year. We measured NIRS on leaf and bud tissue, collected at several dates during the growing season, of 240 dessert apple cultivars in 2021 and 2022. NIRS collected in 2021 and budbreak of 2022 were used to train partial least squares (PLSR) models, then tested using NIRS of 2022 to predict budbreak in 2023. A GWAS using these predictions identified a QTL, previously associated to budbreak in apple, indicating a significant genetic component was maintained in the predictions. Our results demonstrate the potential of NIRS to predict future developmental stages, such as budbreak, by detecting the metabolic states that precede them and could aid in genetic studies of difficult-to-measure traits.

The timing of floral budbreak in apple has a significant effect on fruit production and quality. Budbreak occurs as a result of a complex molecular mechanism that relies on accurate integration of external environmental cues, principally temperature. In the pursuit of understanding this mechanism, especially with respect to aiding adaptation to climate change, a QTL at the top of linkage group (LG) 9 has been identified by many studies on budbreak, but the genes underlying it remain elusive. Here, together with a dessert apple core collection of 239 cultivars, we used a targeted capture sequencing approach to increase SNP resolution in apple orthologues of known or suspected A. thaliana flowering time-related genes, as well as approximately 200 genes within the LG9 QTL interval. This increased the 275 223 SNP Axiom® Apple 480 K array dataset by an additional 40 857 markers. Robust GWAS analyses identified MdPRX10, a peroxidase superfamily gene, as a strong candidate that demonstrated a dormancy-related expression pattern and down-regulation in response to chilling. In-silico analyses also predicted the residue change resulting from the SNP allele associated with late budbreak could alter protein conformation and likely function. Late budbreak cultivars homozygous for this SNP allele also showed significantly up-regulated expression of C-REPEAT BINDING FACTOR (CBF) genes, which are involved in cold tolerance and perception, compared to reference cultivars, such as Gala. Taken together, these results indicate a role for MdPRX10 in budbreak, potentially via redox-mediated signaling and CBF gene regulation. Moving forward, this provides a focus for developing our understanding of the effects of temperature on flowering time and how redox processes may influence integration of external cues in dormancy pathways.

Trehalose 6-phosphate (T6P) signalling regulates carbon use and allocation and is a target to improve crop yields. However, the specific contributions of trehalose phosphate synthase (TPS) and trehalose phosphate phosphatase (TPP) genes to source- and sink-related traits remain largely unknown. We used enrichment capture sequencing on TPS and TPP genes to estimate and partition the genetic variation of yield-related traits in a spring wheat (Triticum aestivum) breeding panel specifically built to capture the diversity across the 75,000 CIMMYT wheat cultivar collection. Twelve phenotypes were correlated to variation in TPS and TPP genes including plant height and biomass (source), spikelets per spike, spike growth and grain filling traits (sink) which showed indications of both positive and negative gene selection. Individual genes explained proportions of heritability for biomass and grain-related traits. Three TPS1 homologues were particularly significant for trait variation. Epistatic interactions were found within and between the TPS and TPP gene families for both plant height and grain-related traits. Gene-based prediction improved predictive ability for grain weight when gene effects were combined with the whole-genome markers. Our study has generated a wealth of information on natural variation of TPS and TPP genes related to yield potential which confirms the role for T6P in resource allocation and in affecting traits such as grain number and size confirming other studies which now opens up the possibility of harnessing natural genetic variation more widely to better understand the contribution of native genes to yield traits for incorporation into breeding programmes.

The domestication and breeding of crops has been a major achievement for mankind enabling the development of stable societies and civilisation. Crops have become more productive per unit area of cultivated land over the course of domestication supporting a current global population of 7.8 billion. Food security crops such as wheat and maize have seen large changes compared with early progenitors. Amongst processes that have been altered in these crops, is the allocation of carbon resources to support larger grain yield (grain number and size). In wheat, reduction in stem height has enabled diversion of resources from stems to ears. This has freed up carbon to support greater grain yield. Green revolution genes responsible for reductions in stem height are known, but a unifying mechanism for the active regulation of carbon resource allocation towards and within sinks has however been lacking. The trehalose 6-phosphate (T6P) signalling system has emerged as a mechanism of resource allocation and has been implicated in several crop traits including assimilate partitioning and improvement of yield in different environments. Understanding the mode of action of T6P through the SnRK1 protein kinase regulatory system is providing a basis for a unifying mechanism controlling whole-plant resource allocation and source-sink interactions in crops. Latest results show it is likely that the T6P/SnRK1 pathway can be harnessed for further improvements such as grain number and grain filling traits and abiotic stress resilience through targeted gene editing, breeding and chemical approaches.

Morphological variation among the viviparous sea snakes (Hydrophiinae), a clade of fully aquatic elapid snakes, includes an extreme “microcephalic” ecomorph that has a very small head atop a narrow forebody, while the hind body is much thicker (up to three times the forebody girth). Previous research has demonstrated that this morphology has evolved at least nine times as a consequence of dietary specialization on burrowing eels, and has also examined morphological changes to the vertebral column underlying this body shape. The question addressed in this study is what happens to the skull during this extreme evolutionary change? Here we use X-ray micro-computed tomography and geometric morphometric methods to characterize cranial shape variation in 30 species of sea snakes. We investigate ontogenetic and evolutionary patterns of cranial shape diversity to understand whether cranial shape is predicted by dietary specialization, and examine whether cranial shape of microcephalic species may be a result of heterochronic processes. We show that the diminutive cranial size of microcephalic species has a convergent shape that is correlated with trophic specialization to burrowing prey. Furthermore, their cranial shape is predictable for their size and very similar to that of juvenile individuals of closely related but non-microcephalic sea snakes. Our findings suggest that heterochronic changes (resulting in pedomorphosis) have driven cranial shape convergence in response to dietary specializations in sea snakes.

Genomic selection (GS) can be effective in breeding for quantitative traits, such as yield, by reducing the selection cycle duration. Speed breeding (SB) uses extended photoperiod and temperature control to enable rapid generation advancement. Together, GS and SB can synergistically reduce the breeding cycle by quickly producing recombinant inbred lines (RILs) and enabling indirect phenotypic selection to improve for key traits, such as height and flowering time, prior to field trials. In addition, traits measured under SB (SB traits) correlated with field-based yield could improve yield prediction in multivariate GS. A 193-line spring wheat (Triticum aestivum L.) training population (TP), tested for grain yield in the field in multiple environments, was used to predict grain yield of a 350-line selection candidate (SC) population, across multiple environments. Four SB traits measured on the TP and SC populations were used to derive principal components, which were incorporated into multivariate GS models. Predictive ability was significantly increased by multivariate GS, in some cases being twice as high as univariate GS. Based on these results, an efficient breeding strategy is proposed combining SB and multivariate GS using yield-correlated SB traits for yield prediction. The potential for early indirect SB phenotypic selection for targeted population improvement prior to trials was also investigated. Plant height and flowering time showed strong relative predicted efficiency to indirect selection, in some cases as high as direct field selection. The higher selection intensity and rate of generation turnover under SB may enable a greater rate of genetic gain than direct field phenotyping.

The genetic improvement of modern wheat varieties has been very successful throughout the history of breeding; yet, future wheat production remains challenging. While annual yield increases need to be doubled over the next few decades, the global production trends in all major wheat growing regions indicate a yield plateau. To overcome this, innovative strategies that efficiently integrate modern technologies in breeding programs are required. Using simulations based on real wheat data sets, we exemplify how genomic selection and "speed breeding," a novel rapid generation advancement technology, can be combined to substantially reduce the length of the breeding cycle and maximize genetic gain per unit time. We outline the opportunities and challenges associated with the fusion of these breeding tools and reinforce the importance of integrating novel genetic diversity in breeding programs to achieve sustainable long-term genetic gain.

Understanding processes in sources and sinks that contribute to crop yields has taken years of painstaking research. For crop yield improvement, processes need to be understood as standalone mechanisms in addition to how these mechanisms perform at the crop level; currently there is often a chasm between the two. Fundamental mechanisms need to be considered in the context of crop ideotypes and the agricultural environment which is often more water limited than carbon limited. Different approaches for improvement should be considered, namely is there genetic variation? Or if not, could genetic modification, genome editing, or alternative approaches be utilized? Currently, there are few examples where genetic modification has improved intrinsic yield in the field for commercial application in a major crop. Genome editing, particularly of negative yield regulators as a first step, is providing new opportunities. Here we highlight key mechanisms in source and sink, arguing that for large yield increases integration of key processes is likely to produce the biggest successes within the framework of crop ideotypes with optimized phenology. We highlight a plethora of recent papers that show breakthroughs in fundamental science and the promise of the trehalose 6-phosphate signalling pathway, which regulates carbohydrate allocation which is key for many crop traits.

'Speed breeding' (SB) shortens the breeding cycle and accelerates crop research through rapid generation advancement. SB can be carried out in numerous ways, one of which involves extending the duration of plants' daily exposure to light, combined with early seed harvest, to cycle quickly from seed to seed, thereby reducing the generation times for some long-day (LD) or day-neutral crops. In this protocol, we present glasshouse and growth chamber-based SB approaches with supporting data from experimentation with several crops. We describe the conditions that promote the rapid growth of bread wheat, durum wheat, barley, oat, various Brassica species, chickpea, pea, grass pea, quinoa and Brachypodium distachyon. Points of flexibility within the protocols are highlighted, including how plant density can be increased to efficiently scale up plant numbers for single-seed descent (SSD). In addition, instructions are provided on how to perform SB on a small scale in a benchtop growth cabinet, enabling optimization of parameters at a low cost.

The growing human population and a changing environment have raised significant concern for global food security, with the current improvement rate of several important crops inadequate to meet future demand [1]. This slow improvement rate is attributed partly to the long generation times of crop plants. Here we present a method called ‘speed breeding’, which greatly shortens generation time and accelerates breeding and research programs. Speed breeding can be used to achieve up to 6 generations per year for spring wheat (Triticum aestivum), durum wheat (T. durum), barley (Hordeum vulgare), chickpea (Cicer arietinum), and pea (Pisum sativum) and 4 generations for canola (Brassica napus), instead of 2-3 under normal glasshouse conditions. We demonstrate that speed breeding in fully-enclosed controlled-environment growth chambers can accelerate plant development for research purposes, including phenotyping of adult plant traits, mutant studies, and transformation. The use of supplemental lighting in a glasshouse environment allows rapid generation cycling through single seed descent and potential for adaptation to larger-scale crop improvement programs. Cost-saving through LED supplemental lighting is also outlined. We envisage great potential for integrating speed breeding with other modern crop breeding technologies, including high-throughput genotyping, genome editing, and genomic selection, accelerating the rate of crop improvement.

The selection of reference genes for RT-qPCR requires the measurement of gene expression across experimental samples. A reference gene with a low variance in its expression for one set of samples may not be suitable for other experiments sampled across different developmental stages, tissue types and seasons. The latter are often typical of postharvest gene expression experiments. In this study we initially used 61 mRNA-seq data sets from different tissues, stages of development and stress conditions of apple (Malus x domestica) to select 10 potential reference genes. These were tested using RT-qPCR, reference gene selection software, and 156 mRNA-seq datasets from a second independent laboratory. From these results we have selected four potential reference genes for use in postharvest and other apple gene expression experiments. Use of this mRNA-seq data method may facilitate selection of reference genes in other crop species for which these data are available.

To accurately measure gene expression using PCR-based approaches, there is the need for reference genes that have low variance in expression (housekeeping genes) to normalise the data for RNA quantity and quality. For non-model species such as Malus x domestica (apples), previously, the selection of reference genes relied on using homology to reference genes in model species. In this study, a genomics approach was used to identify apple genes with low variance in expression in 217 messenger RNA (mRNA)-seq data sets covering different tissues, during fruit development, and treated with a range of different stress conditions. Ten potential reference genes were chosen for validation by quantitative PCR (qPCR) over 29 different tissue types and treatments. From the combined mRNA-seq and qPCR results, three potential reference genes are proposed that can be used as good controls for PCR based expression studies. The three genes show homology to lipid transfer proteins, phytochrome protein phosphatase and the ubiquitination pathway. With the progression of research away from non-model species, this approach provides a robust method for selecting candidate genes for use as reference genes in qPCR.

Information is scant on the mineral nutrition of plants undergoing partial rootzone drying (PRD). Researchers applied PRD to ‘Petopride’ tomato in a glasshouse by alternating irrigation to one side of the plant with half of the water in control (C). Roots in PRD treatment had higher magnesium (Mg) and copper (Cu) than C roots. Leaves in PRD treatment had lower P and higher K than C leaves. Minerals in leaves of C and PRD were within literature reported sufficiency ranges. Fruit of the PRD treatment had higher nitrogen (N) and zinc (Zn), but lower phosphorus (P), calcium (Ca), manganese (Mn) and boron (B) than C fruit. Mineral nutrition of fruit was most affected by PRD compared to that of leaf and root. Fruit of the PRD treatment had a higher incidence of blossom end rot (BER) than C fruit. Incidence of BER notwithstanding, although PRD affected the mineral nutrition of the tomato plant, there was no evidence that the lower yield in PRD was due to poor mineral nutrition.

For any given genotype, the environment in which an apple is grown can influence the properties of the fruit considerably. While there has been extensive research on the mechanism of the genetic control of fruit quality traits, less effort has been made to investigate the way that these genetic mechanisms interact with the environment. To address this issue, we employed a large 'Royal Gala' × 'Braeburn' population of 572 seedlings replicated over sites in three climatically diverse apple-growing regions in New Zealand. Phenotyping for traits including fruit maturation timing, firmness and dry matter content was performed at each of these three sites for a single growing season (2011), and at two sites (Motueka and Hawke's Bay) for two seasons (2009 and 2010). The phenotype data collected over 2 years at two sites enabled the detection of 190 quantitative trait loci (QTL) that controlled these traits regardless of year or growing location, as well as some chromosomal loci that influenced the traits in a single given environment or year. For those loci that were environmentally stable over three sites, there was an interdependency of fruit maturation date, dry matter content and storage potential within this population, with two regions on Linkage Groups (LGs) 10 and 16 strongly contributing. If these loci were used in a marker-assisted selection programme to select for progeny bearing firmer fruit, this would have the unintentional consequence of selecting, high dry matter content, later maturing apples. In addition, a further 113 new QTLs with a smaller effect were identified, some of which were exhibited only in a single growing environment, demonstrating the underlying complexity of control of traits determining fruit quality, in addition to the need for being aware of environmental effects when developing new apple varieties.

The Malling 9 (M.9) dwarfing rootstock is widely used in apple breeding and commercial cultivation to shorten the juvenile period, reduce vegetative growth and increase flowering of the scion. A segment of M.9 stem (interstock) or M.9 bark grafted into a compound tree can cause significant dwarfing of the scion, suggesting that the dwarfing signal may be vascular derived. To better understand how the M.9 rootstock alters the growth and development of the scion, we compared gene expression in vascular-enriched tissue from dwarfing and vigorous rootstocks. RNA sequencing indicated that key flowering genes were upregulated in M.9 relative to a vigorous rootstock, Malling 793 (M.793). An in-depth analysis of the apple FT/TFL1 gene family identified four new members: MdMFTa, MdMFTb, MdBFTa and MdBFTb. Quantitative RT-PCR analysis confirmed the higher expression of MdFT1/2, MdBFTa/b, MdCO, MdGI, and MdSOC1 in two different dwarfing rootstocks (M.9 and Malling 27 (M.27)) relative to M.793. Both MdFT1/2, and MdBFTa/b were expressed at higher levels in multiple dwarfing rootstock accessions relative to more vigorous genotypes. In perennial species, FT promotes flowering, and has additional roles in accelerating the transition from juvenility to maturity, and regulating cycles of seasonal growth and termination. Apple dwarfing rootstocks reduce the juvenile phase and promote both flowering and early shoot termination. Our work supports a role for MdFT in promoting flowering and earlier shoot termination. We suggest that upregulation of a suite of flowering genes including MdFT, and possibly MdBFT, in the vasculature is part of the underlying mechanism of apple dwarfing rootstocks. Genes involved with response to biotic and abiotic stress and disease were also upregulated in the M.9 rootstock, suggesting that stress, possibly mediated by JA and ABA signalling, also plays a role in the M.9-induced phenotype.

Fast establishment of the canopy in young trees, followed by reduced vegetative vigour and precocity are desirable traits in fruit production. Severe heading (cutting back the primary axis of the tree after the first year of growth) to induce branching is a nursery practice to increase early fruit yield. Our aim was to provide a systematic study of the responses of young pear trees to severe heading. We used an experimental system with two scion genotypes and three rootstocks to create trees with contrasting branching habits and vigour. The trees’ trunks were headed and a single bud was allowed to outgrow in the following season. Architectural analysis was used to quantify the development of regenerated trees. In the first year after heading, the growth of the primary axes and, depending on the scion genotype, the sylleptic branching of regenerating trees, were invigorated. In the second year, the percentage of budbreak was also increased, but the shoot growth was greatly reduced. Axis propensity to flower in spring of the third year of growth was increased. The new insights into the effects of heading on tree aging and flowering will be used for guiding the best approaches to managing young pear trees.

Improved knowledge of rootstock effects on pear (Pyrus communis L.) tree development is required before early assessment of rootstock breeding populations can be improved. Two cultivars, 'Doyenné du Comice' and 'Concorde', were grafted on Pyrus calleryana Decne. (vigorous), Quince BA29 (semi-vigorous) and Quince C (semi-dwarfing) rootstocks. Growth of the compound trees was studied over 2 years after grafting and flowering was recorded in the spring of the third year. Using architectural analysis, annual shoot types common to all treatments and closely connected to the patterns of extension of preformed and neoformed metamers were identified and the differences among rootstock vigour treatments were quantified by proportions of these shoot types. Rootstock affected node neoformation, which was highest in the treatment with P. calleryana. The extent of sylleptic branching varied among the treatments, whereas the budbreak along the primary-axis in the second year of growth was unaffected, hence, the number of proleptic secondary axes was largely determined the primary-axis node number developed in year one. Spring flowering first occurred in the third year of tree growth and its intensity was influenced by rootstock. Quince C, the least vigorous rootstock, produced the highest number of floral buds. Flowering was delayed in young pear trees compared with apple, as shown in previous studies, so we conclude flowering does not play such a pivotal role in secondary axes development and early tree dwarfing by rootstock as has been observed in apple.

There is a need to develop novel breeding approaches to increase the rate of genetic gain for wheat yield due to rising demand and climate volatility. Genomic selection (GS) can increase genetic gain in quantitative traits, such as yield, principally through a shorter selection cycle. ‘Speed breeding’ (SB) could further accelerate the breeding cycle by using extended photoperiod in the glasshouse, enabling rapid generation advance. An integrated breeding approach could include fast inbred line development, phenotyping of yield secondary traits for multivariate GS and indirect phenotypic selection for important traits prior to field trials in a target environment. The objective of this thesis was to explore a comprehensive breeding strategy that combines the aforementioned tools, aimed at increasing the rate of genetic gain in wheat. Firstly, to investigate the potential for SB application across major crop species, a sample of wheat, barley, canola and chickpea cultivars were evaluated under either 22-hour light (SB) or natural, diurnal lighting in a temperature-controlled glasshouse to quantify and compare the effects on development. Wheat and barley were also grown at high densities to investigate potential single seed descent (SSD) programs. Time to anthesis was significantly reduced for all species relative to the natural photoperiod conditions. Coupled with harvesting seed two weeks post-anthesis, SB could enable up to six generations of wheat per year. Grain and spike number were similar between lighting treatments and germination percentage was higher in grain of SB-grown plants at two weeks after flowering. These findings indicated that SB had potential to be used at multiple stages of a breeding program. In order to identify traits that could be measured under SB (SB traits) that were correlated to field-based yield and could be used in multivariate GS, an explorative study was carried out on a 135-line, bi-parental (SeriM82 x Hartog, SxH) bread wheat population. Days to anthesis (SB-DTA), plant height (SB-height), spike length and flag leaf length were measured under SB and narrow-sense heritability and genetic correlation with field-based yield was determined from previously performed field trials in three environments that varied with respect to extent and timing of water deficit. Field-based traits were also measured, including DTA, plot height and two senescence indices. The heritability of most SB and field-based traits were higher than yield heritability and SB-DTA and SB-height showed strong genetic correlations with yield (up to -0.80 and -0.66 respectively) in the irrigated environment. Further investigation with a larger, more diverse population was carried out using a previously developed 256-line, multi-reference bread wheat population, and data from two previous rain-fed yield trials. Multivariate models incorporating the field-based senescence indices significantly increased yield prediction ability although the addition of SB traits did not improve prediction above that of univariate models. Yield heritability in these field trials was relatively high and using cross-validation meant a 20% reduction in training population (TP) size, which may have contributed to this result. To validate the proposed breeding strategy of combining GS and SB in a real breeding scenario, a new population of 350 selection candidates (SC) was created by inter-crossing 20 individuals from the SxH population, followed by five generations of selfing under SB. Development of this near-homozygous population was completed in 15 months. A 193-line subset of the SxH population with yield data from five field trials was used as the TP to predict yield of SC lines. These predictions were validated in three yield trials; one irrigated and two rain-fed trials performed across two locations. Four SB traits measured on both the TP and SC population were used to derive principal components (SB-PCs), which were incorporated into multivariate GS models. Predictive ability was significantly increased by multivariate GS with SB-PCs, in some cases being twice the ability of univariate GS, except when predicting yield at Warwick, a higher water-stress location. The genetic correlations between the SB-PCs and yield from the Warwick SC trial were lower than those with Gatton, a location that experienced higher rainfall, illustrating the impact of environmental differences between secondary traits and yield at the target location on prediction ability. In fact, the addition of SB-PCs appeared to have a negative impact on ability to predict yield in Warwick. These findings could be related to the difficulties of predicting into water deficit environments, such as Warwick, compared to environments with more water availability, such as in Gatton and SB conditions. An additional use for SB traits during population development, indirect phenotypic selection for important field-based traits, was explored. The genetic correlations of SB-DTA to field-based time to anthesis and SB-height to plot height indicated indirect phenotypic selection under SB, prior to field trials, could rapidly condition populations for a target environment. Lower selection accuracy, compared to field-based phenotypic selection, could be compensated for by high generation turnover and result in a higher rate of genetic gain. This strategy could benefit wheat breeding programs by increasing, and in some cases, doubling, the rate of genetic gain through rapid development of improved selections.