The Cnr mutation is also thought to be a gain of function mutation, although the mechanism has yet to be understood . The Cnr mutation results from hypermethylation upstream of the gene near the promoter and has been shown to inhibit the genome-wide demethylation cascade associated with normal tomato ripening . Previously, these TFs were regarded as master regulators of ripening; however, given the new information about the nature of the mutations in Cnr, nor, and rin, it is less clear the precise roles the TFs are playing in ripening . The nor and rin mutants have been utilized in breeding for developing tomato hybrids with extended shelf life or extended field harvest depending on their purpose for the fresh market and processing tomato industries . Hybrids between elite varieties and the ripening mutants have a delayed ripening progression, but with the tradeoff of decreased fruit quality attributes, such as color, taste, and aroma . Although there are some publications dedicated to evaluating the physiological characteristics of mutant or hybrid fruit , up to this point, large plastic growing pots much of what we know about the ripening mutations is based on controlled greenhouse experiments with limited fruit and few ripening stages examined.
A complete dataset of phenotypic data produced from large-scale field trials evaluating fruit ripening and senescence is lacking to provide information relevant to breeding, particularly in the new context of the molecular mechanisms behind the nor and rin mutations. The Cnr mutant provides a unique opportunity to study the role of epigenetics in fruit ripening but is not used in breeding because the mutant phenotype is dominant. Cnr has been regarded as a ripening mutant due to its unique colorless phenotype and additional ripening defects . It has been suggested that Cnr fruit undergo normal growth and development ; however, fruit appear different from wild type even before ripening, with a smaller size, alterations in cell wall enzyme expression, and earlier chlorophyll degradation . To better utilize Cnr as a tool for studying fruit development and ripening, a broader understanding of the physiological and transcriptomic alterations in this mutant is necessary. These spontaneous single mutants need to be reevaluated as tools to understand the wide-ranging biological processes regulated by each TF. Previous literature has generally assumed that the mutations block ripening, resulting in similar processes affected . This study demonstrates that each mutant has a unique ripening phenotype, resulting from a combination of inhibited and delayed developmental processes.
We integrated phenotypic data with gene expression data and hormone measurements in the Cnr, nor, and rin mutants across ripening and senescence to characterize the extent and timing of the ripening defects. Tomatoes grown under field conditions were assessed for fruit traits over multiple seasons. We then performed a transcriptomic analysis to gain more definition of the timing in which mutant fruit deviated from WT in their development and to determine specific molecular functions altered in each mutant. Due to their pivotal role in regulating ripening, we focused on defects in hormone networks, including biosynthesis and accumulation. We analyzed the influence of each mutation on the expression of the other TF throughout ripening and senescence. Finally, to better understand the combined genetic effects of the mutants on fruit ripening, we generated homozygous double mutants of Cnr, nor, and rin and used phenotyping and transcriptional data to evaluate the relationships between the mutants.Tomato plants of c.v. ‘Alisa Craig’ and the isogenic ripening mutants Cnr, nor, and rin were grown in randomized plots under standard field conditions in Davis, CA, United States, during the 2016, 2017, 2018, and 2020 seasons. Fruit tagged at 10 days post-anthesis , which corresponds to 7 mm in fruit diameter, were harvested at stages equivalent to the WT fruit.
Fruit were sampled at the mature green , turning, red ripe , and overripe stages, corresponding to 37, 45, 50, and 57 dpa, respectively. The term “RR” is used throughout the manuscript to refer to the 50 dpa stage of all genotypes, even when the mutant fruit do not turn red. Fruit stages for each of the mutants were further validated by external color analysis . Double mutant fruit were generated through reciprocal crosses: Cnr × nor, nor × Cnr, Cnr × rin, rin × Cnr, nor × rin, and rin × nor. Fruit were selfed after the initial cross to generate an F2 segregating generation. The double mutants were initially selected in the F2 generation through genotyping and phenotyping. At least two additional generations after F2 were obtained through selfing to ensure the stability of the double mutations and to perform the experiments in this study. Three seasons of data were collected for the Cnr/nor fruit while only one season of data was collected for the rin/nor and Cnr/rin crosses.The mutant lines were genotyped for their respective mutations. For nor, the Phire Plant Direct PCR Kit was used to extract DNA and amplify the region of the gene containing the 2 bp mutation using the primers listed in Supplementary Table 1. The PCRs were run on a SimpliAmp Thermal Cycler with the following conditions denaturation: 99◦C for 5 min; 35 cycles of 98◦C for 5 s, 56◦C for 25 s, and 72◦C for 25 s; with a final extension of 72◦C for 1 min. The PCR products were purified using Wizard SV Gel and PCR Clean-Up System and then sequenced with Sanger technology to confirm the absence of the two nucleotides. For rin, the Phire Plant Direct PCR Kit was used to extract DNA and perform end-point PCRs using primers specific for the mutant and WT alleles . The following PCR conditions were used for the WT allele primers: denaturation 99◦C for 5 min; 35 cycles of 98◦C for 5 s, 55◦C for 25 s, and 72◦C for 25 s; with a final extension of 72◦C for 1 min. The PCR conditions for the mutant allele primers were: denaturation 98◦C for 5 min; 40 cycles of 98◦C for 5 s, 58◦C for 25 s, and 72◦C for 25 s; with a final extension of 72◦C for 1 min. The PCR products were visualized as bands using a 1% agarose gel. The Cnr epimutation was genotyped by bisulfite sequencing. Extracted DNA was treated with the Zymo Gold bisulfite kit . Bisulfite treated-DNA was PCR amplified for the CNR promoter region containing the methylation changes using the primers listed in Supplementary Table 1. The following PCR conditions were used: 94◦C for 2 min; 40 cycles of 94◦C for 30 s, 54◦C for 30 s, and 60◦C for 45 s, and a final extension of 60◦C for 10 min. The PCR products were then Sanger sequenced and compared to the same region amplified in untreated controls with primers . The following conditions were used to amplify the untreated DNA: 95◦C for 2 min; 35 cycles of 95◦C for 30 s, 56◦C for 30 s and 72◦C for 1 min, and a final extension of 72◦C for 10 min. To ensure mutants were homozygous for the locus, we confirmed the double mutants by allowing the plants to self for at least two additional generations and checking that the progeny were not segregating for any fruit phenotypes.Fruit trait data were collected across four field seasons . The genotypes, large plastic pots developmental stages, number of biological replicates, and number of field seasons used for fruit trait phenotyping can be found in Supplementary Table 2. One season of phenotyping was performed for Cnr/rin and rin/nor double mutant fruits for color, firmness, and ethylene. Three seasons of data were collected for the Cnr/nor double mutant fruit for ethylene and two seasons of data for color and firmness. Fruit were collected from multiple plots or harvests to capture environmental variability.
Fruit trait measurements were taken on the same day of harvest for all samples unless noted. Intact and halved fruit were imaged using the VideometerLab 3 facilitated by Aginnovation LLC1 . External color measurements were obtained from individual fruit with the CR-410 Chroma Meter and recorded in the L∗ a ∗b ∗ color space, where L∗ quantifies lightness, a∗ quantifies green/red color, and b∗ quantifies blue/yellow color. Principal component analysis of the color parameters was performed with the FactoMineR package and graphed with the FactoExtra package in R . Non-destructive firmness measurements were taken on the TA.XT2i Texture Analyzer using a TA-11 acrylic compression probe, a trigger force of 0.035 kg, and a test speed of 2.00 mm/sec with Exponent software . Firmness values are reported as kilograms force. The size was measured by taking the largest diameter of the fruit with a handheld caliper. Tomato juice was produced by pressing the fruit tissues with a juicer and filtering with cheesecloth to measure total soluble solids and titratable acidity . At least five biological replications of tomato juice were obtained from independent pools of 10–12 fruit from distinct plots in the field or at different harvest dates within the field season. TSS were measured as percent Brix with a Reichert AR6 Series automatic bench refractometer from the prepared juice with three technical replicates. Four grams of juice were diluted with water in 20 mL of deionized water to measure TA based on citric acid equivalents. Significant differences in fruit traits across genotypes and ripening stages were determined in R using Type I analysis of variance tests, followed by a post hoc test using the R package agricolae .On the day of harvest, the fruit pericarp tissues were dissected and flash-frozen in liquid nitrogen. Frozen tissues were then ground to a fine powder with the Retsch Mixer Mill MM 400. One gram of ground tissue was used for RNA extractions as described in Blanco-Ulate et al. . RNA concentrations were quantified with Nanodrop One Spectrophotometer and Qubit 3 . RNA integrity was then assessed on an agarose gel. Six biological replicates composed of 8–10 independent fruit were extracted per genotype and ripening stage from the 2016 and 2018 seasons.Four biological replicates each of Cnr/nor MG and RR fruit RNA were used to prepare cDNA libraries. cDNA libraries were prepared with Illumina TruSeq RNA Sample Preparation Kit v.2 from the extracted RNA. The quality of the barcoded cDNA libraries was assessed with the High Sensitivity DNA Analysis Kit in the Agilent 2100 Bioanalyzer and then sequenced on the Illumina HiSeq 4000 platform by the DNA Technologies Core at UC Davis Genome Center. Raw RNAseq data from WT, Cnr, nor, and rin at MG and RR were obtained from a published dataset by our group , GEO accession GSE148217, while raw RNAseq data from the immature stages of the ripening mutants were extracted from Lü et al. . The RNAseq datasets for the Cnr/nor double mutant were generated in this study. The raw sequencing reads from the different datasets were analyzed de novo following the bio-informatics pipeline described below. Ethylene production measurements were taken from MG, RR, and OR fruit on the day of harvest. At least five biological replicates of 5–7 fruit were used for the measurements. The genotypes, developmental stages, and number of biological replicates used for ethylene analysis in each field season can be found in Supplementary Table 2. Fruit were weighed and placed in 1 L airtight glass jars. Headspace gas was extracted from the sealed containers after 60 min and was injected into a Shimadzu CG-8A gas chromatograph . Sample peaks were measured against an ethylene standard. The rate of ethylene production was calculated from the peak, fruit mass, and incubation time. Frozen ground tissue prepared from the tomato fruit pericarp was lyophilized, weighed, and extracted in isopropanol:H2O:HCL1MOL with 100 l of internal standard solution as described in Casteel et al. . Abscisic acid and 1-aminocyclopropane-1-carboxylate were measured using liquid chromatography coupled totandem mass spectrometry and internal standards as described in Casteel et al. . The hormone concentrations were expressed as ng/g of dry weight. Four to six biological replicates composed of 8–10 fruit were used for these measurements for the 2017 season. Significant differences in hormone accumulation across genotypes and ripening stages were determined using Type I ANOVA in R, followed by an HSD test using the R package agricolae . In some cases, pairwise comparisons in hormone accumulation were also conducted by Student’s t-test in R.Fruit from the Cnr, nor, and rin mutants fail to acquire most ripening-associated traits that make them appealing for consumption. Yet, each mutant can be distinguished by their unique phenotypes .