Background: Molecular markers associated with relevant agronomic traits could significantly reduce the time and cost involved in developing new sugarcane varieties. Previous sugarcane genome-wide association analyses (GWAS) have found few molecular markers associated with relevant traits at plant-cane stage. The aim of this study was to establish an appropriate GWAS to find molecular markers associated with yield related traits consistent across harvesting seasons in a breeding population. Sugarcane clones were genotyped with DArT (Diversity Array Technology) and TRAP (Target Region Amplified Polymorphism) markers, and evaluated for cane yield (CY) and sugar content (SC) at two locations during three successive crop cycles. GWAS mapping was applied within a novel mixed-model framework accounting for population structure with Principal Component Analysis scores as random component. Results: A total of 43 markers significantly associated with CY in plant-cane, 42 in first ratoon, and 41 in second ratoon were detected. Out of these markers, 20 were associated with CY in 2 years. Additionally, 38 significant associations for SC were detected in plant-cane, 34 in first ratoon, and 47 in second ratoon. For SC, one marker-trait association was found significant for the 3 years of the study, while twelve markers presented association for 2 years. In the multi-QTL model several markers with large allelic substitution effect were found. Sequences of four DArT markers showed high similitude and e-value with coding sequences of Sorghum bicolor, confirming the high gene microlinearity between sorghum and sugarcane.

Sugarcane is known for its highly complex genetics and more knowledge is needed for better use and conservation of genetic materials. In order to identify genotypes and to assess genetic diversity, diverse data sets such as morphological and molecular markers are used as a general approach. To evaluate the usefulness of different markers, important sugarcane genotypes in Argentina were characterized by AFLP, SSR and morphological traits. All genotypes characterized were grouped in one main cluster in dendrograms using two independent softwares. Interestingly, local genotypes grouped together with USA varieties and no clear genetic differentiation could be found probably due to intensive germplasm exchange between these breeding programs. The molecular markers tested were useful for genetic diversity assessment as well as for genotype identification. These markers should be included in the internationally established characters for sugarcane variety protection as they give a better view on whole genome complexity. Additionally, genetic similarities obtained from molecular markers will provide more accurate information to breeders than the pedigree method, especially when considering the asymmetric genetic inheritance of sugarcane. Morphological traits are valuable tools to identify genotypes since they reflect external resemblance more than genetic relatedness. When they were combined with molecular markers the dendogram obtained revealed genetic relationships and the genetic diversity was better estimated. In summary, both methods appear to be useful, complementing each other and should be used together to assist sugarcane breeders in estimating genetic diversity, electing parents for crossings, identifying superior lines and to protect intellectual property rights.

No abstract

Sugarcane commercial variety RA 87-3 was transformed with a genetic construct harboring the epsps gene from Agrobacterium strain CP4 conferring tolerance to glyphosate and nptII gene for kanamycin selection. Transformed lines were multiplied in greenhouse, and herbicide tolerance was evaluated using different concentrations (3, 4, 8 and 16 l/ha) of glyphosate (Helm 48 % p/v). All herbicidetolerant (HT) lines were field tested to confirm glyphosate tolerance and perform preliminary evaluations of phenotypic resemblance to parental cultivar. All transformed lines maintained herbicide tolerance, but many showed phenotypic changes and/or growth aberrations. Ten HT lines, showing close growth resemblance to RA 87-3, were analyzed using nine compulsory morphologic markers proposed by the International Union for the Protection of New Varieties of Plants (UPOV) and 339 molecular markers. Out of the ten HT lines tested, six showed minor morphologic and genetic variations and were selected for field testing over two vegetative crop cycles (plant cane and first ratoon) at two production areas in Argentina. The six field-tested HT lines were found to be almost indistinguishable when comparing agronomic and industrial characteristics and chemical composition. Stable heritance of the CP4 epsps gene and glyphosate tolerance throughout different clonal generations were confirmed by RT-qPCR and Southern blot. Taking into account all results, two out of the six lines tested were selected for a possible commercial release. Our study confirms the utility of genetic transformation as a complementary tool to classical breeding procedures and highlights the usefulness of UPOV traits together with molecular markers for early selections of transgenic events that closely resemble their parental genotype.

Comparative multi-environment trials (METs) of sugarcane genotypes are frequently conducted using a randomized complete-block design (RCBD) within environments. However, blocking does not always ensure spatial variation control because of differential competition for resources among neighboring genotypes. Heterogeneity within trials may also cause between-trial heterocedasticity. This work aims to evaluate different linear mixed models (LMMs) that enable the analysis of spatial correlation and residual heterogeneity among trials for both tons of cane per hectare (TCH) and sucrose content (SC%) in three series of multi-environmental trials conducted to evaluate advanced sugarcane clones. A total of 16 sugarcane trials conducted at different sites and in different crop cycles (age) were analyzed. Individual (age×site combination) and multi-environment analyses were performed. For SC%, the classic RCBD analysis within trial was adequate. For TCH, the anisotropic autoregressive model of order 1 (AR1×AR1) was the best to compare genotype means in most trials, allowing gain in information equivalent, on average, to the addition of 1.6 replicates to the original design. In the case of multi-environment analysis, the AR1×AR1 within-trial with among-trial heteroscedasticity was the best model to compare variety means, both for TCH and SC%. The results showed how a 53 54 S. Ostengo et al. more appropriate mixed model would help avoid commission of judgment errors in sugarcane variety recommendations.

Sugarcane brown rust, caused by Puccinia melanocephala, is a severe foliar disease that occurs in almost all countries where sugarcane is grown. The main control strategy is the use of resistant cultivars. The aim of this work was to identify molecular markers linked to genomic regions associated with a novel brown rust resistance source in sugarcane. An F1 progeny of 300 clones was obtained from a cross between TUC 00–36 and RA 87–3, highly susceptible and highly resistant to brown rust, respectively. A total of 60 F1 clones with extreme phenotype, either highly susceptible or highly resistant to brown rust, were selected. This “pooled tail” population was tested for reactions to brown rust under natural infection in the field during two crop seasons and under artificial infection in the greenhouse. Whole‐genome profiling was performed by DArT‐seq technology. Phenotypic data under both conditions and 23,299 single‐nucleotide polymorphisms (SNPs) obtained from genotyping were analysed to identify markers linked to the resistance trait. Single mapping analyses and subsequent multiple regression showed that 34 SNP markers were significantly linked to resistance alleles. These SNPs jointly explained 69% and 66% of the total phenotypic variation (R2) observed for field and controlled conditions, respectively. The mapping of the 34 SNP sequences revealed that 19 markers aligned to the sugarcane genome, whereas 12 markers aligned to the sorghum genome, all grouped on chromosome 5 with some functional annotations related to vegetal defence response. These marker loci could contribute to the development of molecular tools for molecular marker‐assisted breeding.

Sugarcane breeding programs incorporate foreign material to broaden the genetic base, expanding the gene pool. In South America, the Inter-university Network for the Development of the Sugarcane Industry (RIDESA) and Estación Experimental Agroindustrial Obispo Colombres (EEAOC) sugarcane breeding programs from Brazil and Argentina, respectively, have never exchanged materials. In that sense, the knowledge of the genetic diversity and population structure among sugarcane genotypes of both germplasm banks, determined in a reliable way through their molecular profiles, will provide valuable information to select the best parental accessions for crossing aimed at the efficient introgression of desirable alleles. For that, the aim was to determine the genetic diversity and population structure of 96 Saccharum commercial hybrids from RIDESA and EEAOC sugarcane breeding programs by using TRAP, SSR and markers related to disease resistance (e.g. Bru1 and G1). Genetic structure was determined through genetic similarity analysis, analysis of molecular variance (AMOVA), Multidimensional scaling (MDS), and a Bayesian method. Average PIC values were 0.25 and 0.26, Ho values were 0.24 and 0.28, and He values were 0.25 and 0.28, for TRAP and SSR primers, respectively. Genetic similarity, MDS, and analysis of structure revealed that Brazilian and Argentinean genotypes clustered in two groups clearly differentiated, whereas AMOVA suggested that there is more variability within programs than between them. Regarding Bru1 markers, Brazilian genotypes showed high frequency of haplotype 1 (71.4%) whereas Argentinean genotypes showed high frequency of haplotype 4 (80.8%); haplotypes 1 and 4 are indicated for the presence and absence of the brown rust resistance gene (Bru1), respectively. Respecting the G1 marker, most of the evaluated genotypes (60.4%) showed the presence of the fragment, in a similar proportion for genotypes of both programs. In conclusion, the exchange of materials, at least the most diverse genotypes, between RIDESA and EEAOC breeding programs will allow extending the genetic base of their germplasm banks, and the knowledge of genetic diversity will help breeders to better manage crosses, increasing the probability of obtaining more productive varieties.

Sucrose accumulation curves represent the maturity profile of sugarcane cultivars, which is considered as a character of interest for the selection of genotypes in breeding programs. However, variations due to the environment (E) and interaction between genotype and environment (G×E) may be confused with the effect of genotype (G) and hinder the selection process of promising clones. The objective of this study was to identify a group of accumulation curves with high intra-group genotypic variability in the sucrose accumulation process throughout several E. This group is then used to select genotypes according to their maturity profile. A protocol is presented whereby the following statistical tools are integrated: (i) classification of nonlinear accumulation curves according to parameters associated with the beginning of the maturity process, sucrose accumulation rate and the time elapsed until the accumulation rate decreases, (ii) estimation of the genotypic contribution to intragroup variability of each accumulation curve parameter within each group, and (iii) identification of the group of accumulation curves with the higher contribution of genotypic variability to total variance of sucrose accumulation parameters. The novelty of the work lies in the sequence of analytical steps to identify information useful to select genotypes according to their maturity profile. The protocol involves estimating parameters of nonlinear models for fitting maturity curves in multi-environment trials, clustering of curves according to the sucrose accumulation parameters and estimation of variability due to G, E, and G×E within each cluster to identify the group with characteristic genotypic curves. Its implementation is illustrated using 175 sucrose accumulation curves of nine sugarcane clones evaluated in different crop cycles (first and second ratoons) and several environments (7 to 50 for each clone) in Tucumán, Argentina. The proposed protocol allows identifying sucrose accumulation curves that exhibit a high genotypic variance, thus facilitating the selection of the best clones.