A comprehensive model of tomato fruit ripening regulation by the transcription factors NOR-like1, NAC-NOR, and MADS-RIN

Plant Physiology, 2025, 198, kiaf291 https://doi.org/10.1093/plphys/kiaf291 Advance access publication 30 June 2025 Research Article A comprehensive model of tomato fruit ripening regulation by the transcription factors NOR-like1, NAC-NOR, and MADS-RIN Victor Aprilyanto, 1 Xiaowei Wang, 1 Rufang Wang, 2 Stan Kronenberg, 1 Feitse Bos, 1 Cristian Peña-Ponton, 1 Gerco C. Angenent, 1,3 Ruud A. de Maagd 3, * 1 Laboratory of Molecular Biology, Wageningen University, Wageningen 6700HB, The Netherlands 2 Institute of Facility Agriculture, Guangdong Academy of Agricultural Sciences, Guangzhou 510640, China 3 Bioscience, Wageningen Plant Research, Wageningen 6700AA, The Netherlands *Author for correspondence: [email protected] (R.A.d.M.) The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/General-Instructions) is: Ruud A. de Maagd. Abstract Tomato (Solanum lycopersicum) fruit ripening involves climacteric ethylene production, lycopene accumulation, texture softening, and flavor enhancement, a highly coordinated process accompanied by profound gene expression changes. To construct a comprehensive model of ripening regulation, we studied the effects on ripening phenotypes and underlying gene expression changes in combinations of knockout alleles of NON-RIPENING-like1 (NL1), NON-RIPENING (NAC-NOR, NOR), and RIPENING INHIBITOR (MADS-RIN). Thus, we demonstrated that the products of the putative paralogous transcription factor genes NL1 and NOR together orchestrate ripening initiation and progression through ethylene production. NL1, or the ethylene production that it induces, together with NOR, stimulates the gene expression of transcription factor MADS-RIN, which then becomes the major driver of all ripening processes studied here. NOR and, particularly, NL1 have relatively minor but discernable and clearly different quantitative contributions to the ripening progression after initiation. Thus, the comprehensive model establishes a hierarchy of gene expression events regulating the start and progression of fruit ripening. Received March 23, 2025. Accepted June 9, 2025. © The Author(s) 2025. Published by Oxford University Press on behalf of American Society of Plant Biologists. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Introduction Ripening is an important part of fleshy fruit development and increases the chances of seed dispersal by frugivores. To achieve this, ripening makes the fruit appear distinct from the other parts of the plant and its environment and palatable to the animals consuming it. A series of physiological changes occur during ripening, including sugar production, flavor and aroma biosynthesis, and softening (Giovannoni 2001; Li et al. 2021). Two types of fruit ripening are broadly distinguished based on the role of the gaseous hormone ethylene. Climacteric ripening is associated with a respiration peak and is usually followed by an ethylene burst, while neither happens in nonclimacteric ripening (Giovannoni 2004). Tomato (Solanum lycopersicum) is a climacteric fruit that produces a substantial amount of ethylene during ripening. Ethylene biosynthesis starts from the conversion of S-adenosyl-L-methionine (SAM) to 1-aminocyclopropane-1-carboxylic acid (ACC) by ACC SYNTHASE (ACS; Adams and Yang 1979; Boller et al. 1979). Next, ACC is oxidized to ethylene by ACC OXIDASE (ACO; Hamilton et al. 1991). Two ethylene production systems, 1 and 2, respectively, are known to occur in climacteric fruits (McMurchie et al. 1972; Barry et al. 2000). System 1 is active until the onset of ripening, and it is autoinhibitory, meaning that ethylene inhibits its own production, while system 2 takes over at the onset of ripening, is autocatalytic, and responsible for the ripening-related ethylene peak. Despite the presence of multiple ACS and ACO genes in tomato, only a few change expression during fruit development and ripening (Barry et al. 2000, 1996). Based on these expression changes, ACS1A and ACS6 are active in system 1, while ACS2 and ACS4 are active in system 2. Early discoveries of “ripening transcription factors” (ripening TFs) came from the observation of spontaneous ripening mutants, such as ripening inhibitor (rin; written as rin-s from here onwards), non-ripening (nor; written as nor-s from here onwards), and Colorless non-ripening (Cnr) (Robinson and Tomes 1968; Tigchelaar et al. 1973; Thompson et al. 1999; Wang et al. 2020). As their names suggest, the fruits from these mutants stay green or pale yellow and firm, whereas a wild-type fruit becomes red and soft. The production of ethylene is also severely diminished in these mutants. The rin mutant has a 2.6 kb deletion in the intergenic region between the tandem MADS-box (MCM1, AGAMOUS, DEFICIENS, and SRF) genes RIPENING INHIBITOR (MADS-RIN) and MACROCALYX (MADS-MC), which results in a rin-mc fusion gene that was thought to abolish the MADS-RIN function (Vrebalov et al. 2002). However, a re-evaluation via CRISPR/Cas9 knockout of both MADS-RIN (abbreviated as RIN from here onwards) and the mutant rin allele showed a milder ripening phenotype than the rin-s mutant. These observations led to the conclusion that rin-mc is a gain-of-function mutation that combines the DNA binding by RIN and the negative transcription regulation by MADS-MC (Ito et al. 2017, 2015; Li et al. 2020). Like rin-s, the fruit of nor-s mutant produces little ethylene and stays green and firm. The nor-s allele has a 2 bp deletion in the Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025 third exon of the NAC (NAM, ATAF, and CUC) gene NAC-NOR (NON-RIPENING; abbreviated as NOR from here onwards), which produces a C-terminally truncated protein, making it initially also regarded to be a loss-of-function mutation (Giovannoni et al. 2004). Using mutagenesis by CRISPR/Cas9, it was later shown that a true nor knockout mutant (nor-cr) fruit had a milder ripening phenotype (Wang et al. 2019b; Gao et al. 2020). Moreover, the knockout of the nor-s allele (nor-scr mutant) also partially restored ripening similar to the level of nor-cr, demonstrating that the truncated protein produced by nor-s is responsible for the nonripening phenotype and represents a (trans) dominant-negative mutation (Wang et al. 2019b). In correspondence with the phenotype, nor-cr fruit produces less ethylene and lycopene during ripening due, at least in part, to reduced expression of system 2 ethylene biosynthesis genes (e.g. ACS2 and ACO3) and carotenoid biosynthesis genes GERANYLGERANYLDIPHOSPHATE SYNTHASE 2 (GGPPS2) and PHYTOENE SYNTHASE 1 (PSY1), respectively (Wang et al. 2019b; Gao et al. 2020). Moreover, the fruit of nor-cr was significantly firmer, which is associated with a reduced expression of cell wall metabolism genes, e.g. POLYGALACTURONIDASE 2a (PG2a) and PECTATE LYASE (PL) (Gao et al. 2020). The reduced expression of these ripening-related genes in the nor knockout suggests that NOR might directly regulate them. Indeed, some of these genes were shown to be directly bound and activated by NOR due to the presence of NAC TF binding sites in the promoter region (Gao et al. 2020). NOR-like1 (abbreviated as NL1 from here onwards), previously also named NAC3 (Jing et al. 2018) or SNAC4 (Kou et al. 2016), is another NAC TF that has been shown to regulate tomato ripening (Gao et al. 2018). In addition to its highly similar protein sequence, NL1 shares functions similar to NOR in ripening regulation. The silencing or knockout of NL1 reduced the ripening ethylene production, color development, and fruit softening. Likewise, the expression of genes related to those traits was downregulated (Gao et al. 2018; Kou et al. 2018, 2016). Despite the similarity to NOR, NL1 has additional regulatory roles in seed and preripening fruit development (Han et al. 2014; Guan et al. 2023; Peng et al. 2023). Due to the similar ripening phenotypes caused by the knockout of each gene, NOR and NL1 are likely to coregulate ripening together, particularly the system 2 ethylene biosynthesis (Gao et al. 2018, 2020; Wang et al. 2019b). However, since there is currently little data on the effect of multiple TF knockouts on ripening, it is unknown whether these TFs regulate ripening independently, redundantly, additively, or synergistically. This study aims to determine the mode of ripening regulation by NOR and NL1, as well as their mutual interaction and that with the ripening regulator RIN. It addresses this question by comparing the phenotypes and ripening-related gene expression of combinations of nor and nl1 knockout alleles and the combination of rin and nor mutations. Results Generation of NAC-NOR and NOR-like1 knockout mutants Using CRISPR/Cas12a, we produced knockout mutations in NL1 using crRNAs targeting the first exon. We obtained biallelic (Δ4 bp/Δ6 bp) and heterozygous (Δ4 bp/+) T 0 mutants. In the next generation (T 1 ), we subsequently segregated transgene-free homozygous mutants containing the 2 distinct 4 bp deletion mutations. Both mutations cause a translation frameshift, after amino acids 29 and 28, respectively, and truncated proteins (Supplementary Fig. S1A). Next, we generated allelic combinations of nor nl1 double mutants by crossing nor-cr1 obtained from the previous study (Supplementary Fig. S1B; Wang et al. 2019b) with nl1-cr1 from this study (named as nor-cr1 nl1-cr1). The nor-cr1 nl1-cr1 double homozygous mutants failed to set fruits or produced abnormally small seedless fruits, as also seen by others (Li et al. 2025). Manual pollination with wild-type pollen resulted in normally developing, seeded fruits, indicating that the combined nor and nl1 mutations led to a male fertility defect. As reported in the paper by Li et al., this was due to pollen wall collapse and severely reduced viability and germination. The same paper reports that the single mutations had no such effects, which we confirmed. The fruits derived from manual pollination of nor-cr1 nl1-cr1 with wild-type pollen were used for further phenotyping. Color development is inhibited in the nor knockout and is lacking in the double knockout mutant In this study, we defined ripening initiation as the transition from the mature green (MG) to breaker stages that is indicated by an observable color change occurring at the pericarp stylar end. We then defined ripening progression as the phenotypic changes after ripening initiation. This distinction is important to specifically address the effect of nor and nl1 knockout mutations on ripening aspects. First, we compared the ripening initiation and progression between the wild type, nor-cr and nl1-cr single, and all allelic combinations of nor and nl1 double knockout mutants. The wild-type fruits reached the breaker stage around 53 d postanthesis (dpa), and from this point, the fruit became fully red over the next 7 d (Fig. 1, A and B). The change of color in ripening progression matches well with the changes in fruit pigments as measured by remittance spectroscopy, where chlorophyll content decreased, and lycopene content increased (Supplementary Fig. S2). Similar to what was reported earlier (Wang et al. 2019b), both nor-cr mutants were slightly delayed initiating ripening, but slowly progressed through ripening (Fig. 1, A and B). This was supported by a slower rate of both chlorophyll degradation and lycopene accumulation, resulting in an orange-colored fruit at 7 and 14 d postbreaker stages (Supplementary Fig. S2). In contrast, both nl1-cr mutants had a much longer delay of ripening initiation (Fig. 1A), as observed previously (Gao et al. 2018). During the ripening (from the breaker stage onwards), we observed no visible color difference to the wild type (Fig. 1B, Supplementary Fig. S2). Combinations of knockout alleles from the 2 genes revealed an allele dosage effect on ripening initiation and progression. The nor-cr1/+ nl1-cr1/+ double heterozygous mutant showed a delayed ripening initiation. Although visually similar, the nor-cr1/+ nl1-cr1/+ mutant accumulated significantly less lycopene at breaker + 7 d (Supplementary Fig. S2B). Leaving just a single functional NL1 copy (as in nor-cr1 nl1-cr1/+) resulted in yellow-colored fruits, which was a more severe effect compared to leaving a single functional NOR copy (as in nor-cr1/+ nl1-cr1) that still turned red albeit delayed (Fig. 1B). This was further confirmed by spectroscopy which showed higher lycopene accumulation for the latter mutant. Most strikingly, we did not observe any color change up to 142 dpa, and only general senescence at 340 dpa when all copies of NOR and NL1 were knocked out (as in nor-cr1 nor-like1-cr1) (Fig. 1B). Daily measurements on the fruit of this double homozygous mutant from 80 dpa to 7 d afterward showed no chlorophyll degradation or lycopene accumulation, further supporting that this mutant did not ripen at all. As there was neither ripening initiation or progression observed in this double mutant, we chose 2 | Plant Physiology, 2025, Vol. 198, No. 3 Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025 the fruits of 80 and 100 dpa, far beyond the time to breaker of any of the other mutants, for comparison with the MG stages (defined as 5 d prior to the breaker stage) of the wild type and other mutants. Since ripening in tomato starts from the inner tissues (Brecht 1987), we also compared the color difference between those (jelly, columella, and placenta) and pericarp at the MG and breaker stages. There was no development of red color in either tissue at the MG stage, while at the breaker stage, the inner tissue generally showed a more intense red color than the pericarp (Supplementary Fig. S3). A similar pattern was observed in the nor-cr mutants, although the color of the inner tissues and pericarp was lighter than that of the wild type. Interestingly, the breaker stage of nl1-cr mutants had their pericarp developing a more intense red color than the inner tissue, suggesting that the pericarp ripened earlier. In the double mutants, the allelic combinations followed the patterns of the respective single mutants. The nor-cr1/+ nl1-cr1 mutant exhibited color change in the pericarp earlier than in the inner tissue, whereas the nor-cr1 nl1-cr1/+ mutant barely developed red color in either the pericarp or inner tissue upon ripening. Finally, the nor-cr1 nl1-cr1 double homozygous mutant showed no color change in either tissue. In relation to lycopene production, we compared the expression of the genes upstream in the biosynthetic pathway, GGPPS2, and PSY1, in the pericarp tissue between the wild type and mutants. In the wild type, both GGPPS2 and PSY1 expression Figure 1. Fruit ripening phenotypes. A) Days-to-breaker of the wild type and mutants. The nor-cr1 nl1-cr1 double homozygous mutant was not included since the ripening initiation of this mutant could not be established. At least 25 fruits of each genotype were used in the comparison. For the box plot in panel (A), the center line represents the median; box limits indicate the upper and lower quartiles (Q3 and Q1, respectively); and whiskers extend to 1.5 times the interquartile range (IQR) from the quartiles. The asterisks indicate significant differences using a 2-tailed Student’s t-test at P<0.05 (*) and P<0.01 (**) between each mutant and the wild type. B) Fruit appearance from 30 d postanthesis (30 dpa) to breaker + 14 d (BR14) from the wild type and the nor-cr and nl1-cr single and double mutants. For the nor-cr1 nl1-cr1 double homozygous mutant, fruits were harvested at 30, 80, 100 dpa, while the fruits of 142 and 340 dpa were harvested at 100 dpa. Bar=1 cm. Gene expression by RT-qPCR for carotenoid biosynthesis pathway genes C) GGPPS2 and D) PSY1. For the bar graphs in panels (C) and (D), error bars represent the standard error (SE) of the mean, with sample sizes of n=3 biological replicates for each treatment. Asterisks indicate significant differences using a 2-tailed Student’s t-test at P<0.05 (*) and P<0.01 (**) between each mutant and wild type at the same developmental stages, except for the nor-cr1 nl1-cr1 double homozygous mutant, where both 80 and 100 dpa stages were compared to wild type MG. Abbreviations: 30 dpa, 30 d postanthesis; MG, mature green; BR, breaker; BR7, breaker + 7 d; BR14, breaker + 14 d. A model of tomato ripening regulation by 3 TFs | 3 Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025 increased during ripening, with peak expression at the breaker and 7 d afterward, respectively (Fig. 1, C and D). The knockout of NOR lowered the expression of both during ripening, explaining the orange phenotype of the mutant. This was in contrast with the knockout of NL1, which lowered GGPPS2 and PSY1 expression only at the MG and breaker stages but not at the later stages (Fig. 1, C and D). This matched the higher final lycopene content in the nl1 compared to nor knockout mutants. The combined knockout alleles of nor and nl1 showed an allele dosage effect on the expression of GGPPS2 and PSY1, with a more profound decrease in the mutant with decreasing functional NOR compared to decreasing functional NL1. This correlated well with the observed fruit appearance and measurement of color change during ripening. Lastly, there was no increase in GGPPS2 and PSY1 expression in the nor-cr1 nl1-cr1 double homozygous mutant at 80 or 100 dpa, matching the absence of ripening color in this mutant (Fig. 1, C and D). To summarize, both NOR and NL1 regulate the ripening process, with NL1 regulating the ripening initiation and NOR and NL1 together, with a bigger role for NOR, regulating ripening progression. The wild-type alleles of both genes act additively during ripening, as revealed by the allele dosage effect, and without an active allele of either of the two, ripening is inhibited altogether. NAC-NOR and NL1 regulate ethylene production additively Ethylene production is a prominent feature of climacteric ripening in tomato. In the wild type, fruit ethylene production was negligible at 30 dpa and at the MG stage but rose significantly toward the breaker stage and the next 7 d (Fig. 2A). The overall climacteric rise of ethylene production was reduced in nor-cr and nl1-cr single homozygous knockout lines, confirming their previously reported regulatory roles in ethylene production (Gao et al. 2018, 2020; Wang et al. 2019b). Despite the similarity, the dynamics of ethylene production differed between the mutants. In the nor-cr mutants, while not significantly different up to the breaker stage, the peak production at the breaker + 7 d stage in this mutant was much lower than the wild type. This was in contrast with nl1-cr mutant fruits which produced significantly less ethylene at the breaker stage, but rose to wild-type levels afterwards (Fig. 2A). In parallel with color development, we then observed a gradual decrease in peak ethylene production when more alleles of NOR and NL1 were knocked out. The mutant with 1 functional NOR allele (nor-cr1/+ nl1-cr1) produced more ethylene than that having 1 functional NL1 allele (nor-cr1 nl1-cr1/+). Like the single nl1-cr mutants, the nor-cr1/+ nl1-cr1 fruit also produced much less ethylene at the breaker stage. This underlines the similar yet not identical role of the 2 genes in regulating ethylene production, where NL1 impacts early ripening (the breaker stage) production and NOR regulates later ethylene production, with the latter having the major role. A residual role of NL1 in late ethylene production became apparent only in the nor-cr1 nl1-cr1/+ genotype, producing less ethylene than the nor-cr1 single mutant. Lastly, the absence of ethylene production in the double homozygous mutant (nor-cr1 nl1-cr1) suggests that ripening ethylene production is triggered by these 2 genes in an additive manner. To understand the dynamic of ethylene biosynthesis regulation at the molecular level, we compared the expression of system 2 ethylene biosynthesis genes between genotypes. The expression of ACS2 and ACS4 in the wild type was correlated with ethylene production, with expression rising more than 100-fold and 50-fold for ACS2 and ACS4, respectively, at breaker and peaking Figure 2. Ethylene production and the expression of system 2 ethylene biosynthesis genes in wild type and mutants at 5 developmental and ripening stages. A) Ethylene production. For the nor-cr1 nl1-cr1 double homozygous mutant, ethylene production was measured at 30, 80, and 100 dpa. At least 5 fruits per stage from each genotype were used for the measurements. For the box plot in panel (A), the center line represents the median; box limits indicate the upper and lower quartiles (Q3 and Q1, respectively); and whiskers extend to 1.5 times the interquartile range (IQR) from the quartiles. The expression of system 2 ethylene biosynthesis genes comprises B) ACS2, C) ACS4, and D) ACO1. For the bar graphs in panels (B) to (D), error bars represent the standard error (SE) of the mean, with sample sizes of n=3 biological replicates for each treatment. Asterisks indicate significant differences using a 2-tailed Student’s t-test at P<0.05 (*) and P<0.01 (**) between each mutant and the wild type at the same developmental stages, except for the nor-cr1 nl1-cr1 double homozygous mutant, where both 80 and 100 dpa stages were compared to MG wild type. 4 | Plant Physiology, 2025, Vol. 198, No. 3 Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025 7 d later (Fig. 2, B and C). Overall, ACS2 expression was more severely impacted by the mutations than ACS4 expression was. In nor-cr1, ACS2 expression was significantly lower in all ripening stages (from the breaker to breaker +14 d stages), in contrast to its expression in nl1-cr1, which was lower only at the MG stage. This pattern, albeit weaker, is also observed for ACS4, where nor-cr1 had a reduced expression at the breaker + 7 d stage. While the nor-cr1/+ nl1-cr1/+ double heterozygous mutant showed no significantly lower ACS2 and ACS4 expression (Fig. 2, B and C), further removal of an additional NOR allele had a bigger impact than that of an extra NL1 allele, which is consistent with the ethylene production (Fig. 2A). The contribution of NL1 in regulating ACS2 and ACS4 expression can be seen when comparing expression in the mutants with nonfunctional NOR alleles. Here, nor-cr1 nl1-cr1/+ had lower expression of ACS2 and ACS4 than the nor-cr1 mutants (Supplementary Table S1), showing that although small, NL1 contributes to the regulation of ethylene biosynthesis genes. Finally, both ACS2 and ACS4 were expressed at very low levels in the double homozygous mutant (Fig. 2, B and C). This suggests that the overall expression of ACS2 and ACS4 and resulting ethylene production (Fig. 2A) is controlled by NL1 at the onset of ripening and then by both NL1 and NOR during the progression of ripening. Additionally, ACO1 expression was also affected in the mutants. In the wild type, ACO1 expression peaked at the MG stage and went down at later stages of ripening. A similar pattern was also shown in the nor-cr1 mutant, although with a slower rate of decrease, which resulted in a higher expression than the wild type at the breaker + 14 d stage (Fig. 2D). Despite a similar decreasing expression trend in the nl1-cr1 mutant, the expression levels for each stage during ripening were significantly higher than the wild type, suggesting that the knockout shifts the ACO1 expression peak, reflecting the ripening delay in the mutants. When additional NOR and NL1 functional alleles were lost, ACO1 expression was further decreased. This indicates that NOR and NL1 also control ethylene production through ACO1 expression, although ACS2 and ACS4 expression correlates better with actual ethylene production. Substantial ripening restoration by external ethylene occurs only when a functional NOR allele is present The observation of reduced ethylene production and its biosynthetic gene expression before and during ripening in the nor and nl1 mutants suggests that ethylene production drives both ripening initiation and progression. Since the lack of ethylene precludes both the initiation and progression of ripening, we hypothesized that supplying external ethylene would restore these phenotypes in at least some of the mutants. To test this, we treated the wild type and mutant fruits with ethephon solution and observed the fruit color development over 14 d posttreatment (dpt). Most of the ethephon-treated fruits initiated ripening earlier than the mock-treated fruits, although ripening progression varied between mutants containing nor and nl1 knockout alleles (Fig. 3). In the nor-cr mutants, both ripening initiation and progression were visually similar between the mock and ethephon-treated fruits, confirming that a functional NOR and sufficient ethylene are required. This contrasted with the nl1-cr mutants, where ethephon treatment initiated earlier ripening and accelerated its progression (breaker at 4 dpt) compared to the mock treatment (breaker at 14 dpt), showing that a sufficient level of initial ethylene is required to initiate ripening and that this is, at least mainly, controlled by NL1 (Fig. 3). This was further supported by earlier ripening initiation in the fruits of both nor-cr1/+ nl1-cr1 and nor-cr1 nl1-cr1/+ double mutants in ethephon treatment. However, only the former was able to progress through ripening similar to normal. Finally, the ethephon treatment on nor-cr1 nl1-cr1 fruit did not change the fruit color even after 14 dpt, indicating that this mutant did not ripen even with externally supplied ethylene (Fig. 3). This suggests that ethylene production is controlled by NL1 as well as by NOR activity and that both are required for normal ripening. Taken together, we showed that a sufficient level of initial ethylene is required to initiate ripening and that NL1 controls this. After that, the continuation of ethylene production in ripening progression requires both NOR and NL1. Fruit firmness is differentially regulated by NAC-NOR and NL1 Fruit firmness decreases during ripening. The on-the-vine firmness of the wild-type fruit peaked at the MG stage and subsequently decreased during ripening (Supplementary Fig. S4A). The nor-cr mutant fruits were firm at the MG stage, but softened slower than the wild type during ripening, resulting in firmer fruits at the breaker + 14 d stage. In contrast, the nl1-cr mutants had much firmer fruits at the MG stage but softened faster during ripening, resulting in slightly firmer fruits at the breaker+ 14 d stage. Combining nor and nl1 knockout alleles produced an aggregate of what was observed in the single knockout mutants. The nor-cr1/+ nl1-cr1/+ double heterozygous had a higher firmness at MG and softened at a slightly slower rate compared to the wild type. From this point, further removal of the NL1 allele resulted in fruit with higher firmness at MG but softening at a rate similar to the wild type, while removal of the NOR allele gave a fruit with normal preripening firmness but softened slower. Lastly, the fruit of the nor-cr1 nl1-cr1 double homozygous mutant had a significantly higher preripening firmness that did not decrease throughout 100 dpa. Taken together, it can be concluded that fruit softening, controlled by NL1, occurs prior to ripening initiation and that it is accelerated under combined control by NL1 and NOR during ripening progression. We studied the expression of PG2a and PL which are associated with loss of viscosity and fruit softening (Wang et al. 2019a). PG2a is one of the most strongly induced genes during ripening, and its expression in the wild type increased to more than 1000-fold at the breaker + 7 d stage (Supplementary Fig. S4B). PL expression increased 50-fold, peaking at the same stage (Supplementary Fig. S4C). The expression of both genes in nor-cr1, but not for the nl1-cr1, was lower at the breaker + 7 d stage, indicating that NOR controls more of their expression than NL1 does. Further depletion of functional NOR alleles resulted in lower expression of PG2a and PL than that of an NL1 allele. Finally, no significant expression increase occurred in the double homozygous (nor-cr1 nl1-cr1) mutant at 80 and 100 dpa. By comparing the gene expression with the fruit firmness phenotype, we showed that the expression change of PG2a and PL was associated negatively with the fruit firmness during ripening but not before it. Although the higher PG2a and particularly PL expression could explain the lower fruit firmness in the nor-cr1 mutant at the MG stage, it could not satisfactorily explain the much higher fruit firmness in the mutants containing the nl1-cr1 mutation. This indicates that the change in fruit firmness, due to the activities of cell wall metabolism genes other than PG2a and PL, has occurred in the early fruit development, and NL1 regulates it. A model of tomato ripening regulation by 3 TFs | 5 Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025 NOR and NL1 additively regulate the expression of ripening TFs Considering that the knockouts of NOR, NL1, or both downregulate many ripening-related genes, we hypothesized that they may also affect the expression of other ripening TFs. To investigate this, we compared the expression of RIN, NOR, NL1, FUL1, and FUL2 in the single and double mutants of nor and nl1. RIN plays a significant role in regulating ripening, and its knockout causes a marked decrease in ethylene production and other ripening aspects (Ito et al. 2017; Li et al. 2020). It is also the most upregulated TF during ripening as far as tested here, more so than FUL1 and NOR and much more so than FUL2 and NL1 (Fig. 4A). Our analysis revealed that RIN expression was downregulated in both nor and nl1 single mutants, with the downregulation being more pronounced in the nor than in the nl1 knockout. In the allelic combinations of the double mutants, RIN expression was only reduced further in the complete knockout of NOR. Furthermore, there was no increase in RIN expression in the nor nl1 double homozygous mutant at 80 and 100 dpa, indicating that RIN expression is influenced by NOR and, to a lesser extent, by NL1. We also investigated the effect of the knockouts on the expression of NOR. Like RIN, NOR expression was downregulated in both nor and nl1 single mutants, with the downregulation being more pronounced in nor than in nl1 (Fig. 4B). Interestingly, at the MG stage, NOR expression was significantly lower only in mutants with partial or complete NL1 knockout, suggesting that NL1 regulates NOR. In the nor nl1 double homozygous mutant, NOR expression was further downregulated, with no significant increase observed at 80 and 100 dpa. These findings suggest that both NOR and NL1 additively regulate NOR transcription levels. Despite its high sequence similarity with NOR, the expression pattern of NL1 differed significantly from that of NOR (Fig. 4C). Interestingly, there were no significant changes in NL1 expression in the nor single mutant, suggesting that NOR does not regulate NL1 expression. However, NL1 transcripts were not detectable in mutants with a complete nl1 knockout, likely due to transcript degradation via the nonsense-mediated mRNA decay (NMD) pathway (Lykke-Andersen and Jensen 2015). This presumed degradation of NL1 transcripts prevented us from assessing the effect of the nl1 knockout on its own expression. Both FUL1 and FUL2 are TFs belonging to the MADS family, and during ripening, they interact with RIN to regulate downstream genes, including in the ethylene biosynthesis pathway (Bemer et al. 2012; Wang et al. 2019b). Despite their high similarity in protein sequence, the expression patterns of both FULs differed in the nor and nl1 mutants. The expression of FUL1 was significantly Figure 3. Comparison between the wild type and the single and double mutants of nor and nl1 fruits treated with water (mock; upper rows) and ethephon (lower row) from 0 to 14 dpt. Images were digitally extracted for comparison. 6 | Plant Physiology, 2025, Vol. 198, No. 3 Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025 downregulated in mutants containing a homozygous nor knockout (Fig. 4, D and E). Furthermore, FUL1 expression decreased further with the additional knockout of NL1, eventually resulting in no expression increase in the nor nl1 double homozygous mutant. These findings indicate that NOR and NL1 regulate FUL1 expression either directly or indirectly. RNA-seq analysis of gene expression prior to ripening The comparison of phenotypes and expression of several ripening-related genes among the mutants further supports the hypothesis that NL1 regulates ripening initiation and both NL1 and NOR additively regulate ripening progression. However, the relative contribution of both NL1 and NOR in these processes required further investigation. To identify which ripening genes are regulated by both NL1 and NOR and why knocking out both NOR and NL1 abolishes ripening completely, we did an RNA-seq of the pericarp tissue at the MG stage. In the case of nor-cr1 nl1-cr1 mutant, we used 80 dpa fruit as the stage comparable to the MG stage since it showed no signs of ripening. Pericarp tissue from at least 3 different fruits at the defined stages was used for each biological replicate. Detection of the differentially expressed genes (DEGs) was based on a cutoff threshold of |log2fold| >1 and FDR <0.05 (Supplementary Table S3). Venn diagrams were made to compare and assess overlap between mutants of numbers of upor downregulated DEGs (Fig. 5A). At MG, the nl1-cr1 had 583 upand 348 downregulated DEGs compared to the wild type, while the nor-cr1 had 482 upand 223 downregulated DEGs. These numbers varied significantly compared to the nor-cr1 nl1-cr1 double mutant, which, at 80 dpa, had 1,005 upand 1,399 downregulated DEGs. We observed that more of these DEGs were shared between the double mutant and nl1-cr1 compared to the nor-cr1, indicating that the NL1 knockout induced more profound transcriptomic changes than the NOR knockout at this stage. A GO enrichment from the downregulated DEGs showed that the nor-cr1 nl1-cr1 double homozygous mutant had an enrichment of “fruit ripening” GO term (GO:0009835), showing that the knockout of both NL1 and NOR downregulates the expression of most ripening-related genes (Fig. 5B; Supplementary Table S4). Interestingly, this GO term enrichment was shared only with the nl1-cr1 mutant (Supplementary Fig. S5; Supplementary Table S5), suggesting that knocking out NL1 is necessary to downregulate those genes and leading to the ripening initiation delay observed in the mutant. The absence of the “fruit ripening” GO term in the nor-cr1 mutant (Supplementary Fig. S6; Supplementary Table S6) indicates that the effect of nor-cr1 mutation only minimally affected the expression of ripening genes before ripening is initiated. Gene comparison within the “fruit ripening” GO term between the nl1-cr1 and nor-cr1 nl1-cr1 mutants revealed that most genes Figure 4. Expression of ripening regulating TF genes A) RIN, B) NOR, C) NL1, D) FUL1, and E) FUL2. For the bar graphs in all panels, error bars represent the standard error (SE) of the mean, with sample sizes of n=3 biological replicates for each treatment. Asterisks indicate significant differences using a 2-tailed Student’s t-test at P<0.05 (*) and P<0.01 (**) between each mutant and wild type at the same developmental stages, except for the nor-cr1 nl1-cr1 double homozygous mutant, where 80 and 100 dpa stages were compared to wild type MG. No statistical test was conducted for NL1 expression in the lines with complete nl1-cr1 knockout due to its very low expression. A model of tomato ripening regulation by 3 TFs | 7 Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025 associated with the former were subsets of those found in the latter (Fig. 5C). This included some of the most common ripeningrelated genes responsible for: ethylene biosynthesis (ACS2, ACS4, ACO6), color development (GGPPS2, PSY1, SGR1, ZISO), cell wall metabolism (PG2a, PL, CEL1, CEL2), and ripening-related demethylases DEMETER-LIKE 2 (DML2) and AlkB HOMOLOG 2 (ALKBH2). While the expression of some of these genes was downregulated in the nl1-cr1 mutant at the MG stage (Figs. 1, C, D, 2, B to D and 4B), other genes like ACO6, DML2, and ALKBH2 also followed the same trend (Fig. 5, D to F). In addition to the downregulated ripening-related genes in the nl1-cr1 mutant, more genes were downregulated in the nor-cr1 nl1-cr1 double homozygous mutant. Among the others are the ripening TFs, such as NOR, RIN, and NAC4, which are known to Figure 5. Most fruit ripening-related genes were downregulated in the nor-cr1 nl1-cr1 mutant at 80 dpa. A) Venn diagrams showing the upand downregulated DEG overlaps between nl1-cr1, nor-cr1, and nor-cr1 nl1-cr1 mutants at the MG stage. For nor-cr1 nl1-cr1, the DEGs from 80 dpa were used for comparison. B) GO enrichment on the downregulated genes of nor-cr1 nl1-cr1. C) Heatmap of the genes comprising the “fruit ripening” GO term (GO:0009835) compared between nl1-cr1, nor-cr1, and nor-cr1 nl1-cr1 mutants. RT-qPCR results of the several ripening genes compared between the wild type and mutants, comprising D) ACO6, E) DML2, F) ALKBH2, and G) NAC4. For the bar graphs in panels (D) to (G), error bars represent the standard error (SE) of the mean, with sample sizes of n=3 biological replicates for each treatment. Asterisks indicate statistical significance based on a 2-tailed Student’s t-test at P<0.05 (*) and P<0.01 (**) between mutants to wild type at the same developmental stage, except for the nor-cr1 nl1-cr1 double homozygous mutant, where the 80 and 100 dpa stages were compared to wild type MG. 8 | Plant Physiology, 2025, Vol. 198, No. 3 Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025 regulate ethylene biosynthesis during ripening (Fig. 5C) (Zhu et al. 2014; Ito et al. 2017; Wang et al. 2019b; Gao et al. 2020, 2021; Li et al. 2020). As shown in Fig. 4, A and B, the expression of both RIN and NOR, in the nor-cr1 nl1-cr1 double mutant, was very low at both 80 and 100 dpa. Since both TFs regulate the expression of system 2 ethylene biosynthesis genes, their downregulation in the double mutant explains the low ACS2 and ACS4 expression and, subsequently, the absence of ripening. Surprisingly, the expression of RIN was significantly higher in the nor-cr1 mutant at the MG stage, which was in line with the expression of RIN observed in the RT-qPCR (Figs. 5C and 4A). As RIN is the major regulator of system 2 ethylene (Ito et al. 2017; Li et al. 2020), this upregulation subsequently led to the increased expression of ACS2 and ACS4 (Fig. 2, B and C), and eventually to the ripening ethylene production. The higher ethylene production in some fruits (Fig. 2A); higher expression of ACS2, ACS4, and RIN (Figs. 2, B, C and 4A); and lower firmness at the MG stage of nor-cr1 fruits (Supplementary Fig. S4) demonstrate that these ripening phenomena preceded the color change in this mutant. Additionally, we observed no increase of NAC4 expression, which further supports the severely inhibited ethylene production in the double mutant (Fig. 5G). Taken together, the downregulation of the ripening TFs in addition to ethylene biosynthetic genes might explain the difference between the delayed ripening in nl1-cr1 and no ripening in nor-cr1 nl1-cr1 mutants. RIN and NOR regulate most of the ripening progression Both RNA-seq and RT-qPCR show that RIN is significantly downregulated in the single and double mutants of nor-cr1 and nl1-cr during ripening, suggesting that it is regulated by both NOR and NL1. While the role of RIN in ripening regulation is known (Ito et al. 2017; Li et al. 2020), it is also relevant to know its contribution relative to NOR and NL1 and its position in the ripening network. To investigate this further, we created the rin-cr1 knockout mutant using CRISPR/Cas9 and compared its phenotype and gene expression to that of the other mutants. The rin-cr1 mutant exhibited a 22 bp deletion in the third exon of RIN, which resulted in a 104 aa truncated protein that aligns with the wild-type RIN product for the first 93 residues, comparable to the rinG1 mutant (Ito et al. 2015) (Supplementary Fig. S1C). We also crossed rin-cr1 with both nor-cr mutants and segregated the homozygous double mutants to study the combined roles of RIN and NOR in ripening regulation. The rin-cr1 mutant developed yellow-colored fruits after the breaker stage that turned to slightly orange at the breaker +14 d stage (Fig. 6A), which was similar to the phenotype observed in the rin CRISPR mutants from previous studies (Ito et al. 2017, 2015). The fruits of rin-cr1 were strikingly different from nor-cr1 fruits, which developed a more orange-red color, confirming that the knockout of RIN results in more severe effects on ripening (Fig. 6A). When both mutations were combined, the resulting double mutants rin-cr1 nor-cr1 and rin-cr1 nor-cr2 changed color much slower than both single mutants, giving a pale-yellow color at the breaker + 14d stage (Fig. 6A). Although both rin-cr1 and nor-cr1 single mutants required a few more days to reach breaker (55 dpa, from 53 in the wild type), their combined effects of double homozygous mutations delayed it much further (to 65 dpa) (Fig. 6B). The rin-cr1 mutant produced substantially less ethylene than both wild type and the nor-cr1 mutant (Fig. 6C), indicating that RIN plays the dominant role in regulating ethylene production during ripening progression, which is in line with its more severe color phenotype. Although ethylene production was low in the rin-cr1 mutant at the breaker stage, we observed a slight increase over time, suggesting that while RIN has the dominant role, it coregulates ethylene production with other factors. Interestingly, we observed even lower ethylene levels with no further increase over time in both rin-cr1 nor-cr double mutants (Fig. 6C). This indicates that the residual rise in ethylene production in the rin-cr1 mutant might be due to NOR, showing that ripening-related ethylene production is coregulated by RIN and, to a much lesser extent and independent of RIN, by NOR. To see whether ethylene could induce and accelerate ripening in these mutants, we treated the fruits with ethephon solution and observed the color change up to 14 d afterward (Supplementary Fig. S7). The treated fruit of the rin-cr1 mutant ripened earlier and accumulated more color than the mock-treated fruit, indicating that the ethylene treatment could still progress ripening. In contrast, the mockor ethephontreated fruits of both double mutants initiated ripening with a similar delay as the color change shows, indicating that ethylene could not induce ripening in the absence of NOR. In addition to ethylene production, we also compared fruit softening during ripening. As shown in Fig. 6D, both nor-cr1 and rin-cr1 single mutants softened albeit slower than the wild type. Softening was significantly reduced in the double mutants as the fruit firmness did not decrease from the MG to breaker + 7 d stage and only slightly so at breaker + 14 d. Thus, although the regulation of fruit softening by RIN and NOR may well be through ethylene signaling, the slight difference in softening rate observed between these mutants does not reflect their large difference in ethylene production. A comparison of system 2 ethylene biosynthesis gene expression (ACS2 and ACS4) revealed differences between the single and double mutants. Although both single mutants had lower ACS2 expression compared to the wild type, the expression in the nor-cr1 was significantly lower than in rin-cr1 at breaker + 7 d (Fig. 6E; Supplementary Table S2). This was in contrast to ACS4 expression, which was downregulated more in rin-cr1 than in nor-cr1 (Fig. 6F). Moreover, the rin-cr1 nor-cr1 double mutant expressed ACS2 even lower than that of both single mutants but expressed ACS4 like the single rin-cr1 mutant (Supplementary Table S2). The latter shows that NOR and RIN additively regulate ACS2, while ACS4 is regulated primarily by RIN. The differential regulation of ACS2 and ACS4 also suggests why rin-cr1 had more impact on ethylene production than nor-cr1. Other ripening genes like those acting in carotenoid biosynthesis (GGPPS2 and PSY1) and cell wall metabolism (PG2a and PL) were affected similarly to ethylene biosynthesis. The expression of these genes was downregulated in the single mutants and even further in the double mutant (Supplementary Fig. S8, A to D; Supplementary Table S2). Interestingly, the expression of GGPPS2, PG2a, and PL was more downregulated in nor-cr1 than in rin-cr1 at the breaker + 7 d stage, confirming that they are also directly regulated by NOR as previously shown (Gao et al. 2020), in addition to being regulated by ethylene. To see whether the phenotype is correlated with the expression of ripening TF genes, we also compared the expression of RIN, NOR, and NL1 between the wild type and the mutants. The expression of RIN was downregulated in both nor-cr1 and rin-cr1 single mutants, indicating that its expression is indirectly regulated by ethylene or directly via transcriptional regulation by NOR and RIN (Fig. 6G). The expression of NOR was downregulated in the nor-cr1 knockout, while it was only shifted in the rin-cr1 single mutant (Fig. 6H). This was in contrast with the expression of NL1, which was hardly affected in all mutants (Fig. 6I). A model of tomato ripening regulation by 3 TFs | 9 Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025 Lang Z, Wang Y, Tang K, Tang D, Datsenka T, Cheng J, Zhang Y, Handa AK, Zhu J-K. Critical roles of DNA demethylation in the activation of ripening-induced genes and inhibition of ripeningrepressed genes in tomato fruit. Proc Natl Acad Sci U S A. 2017: 114(22):E4511–E4519. https://doi.org/10.1073/pnas.1705233114 Li H-L, Xiang L-T, Zhao X-D, Zhu B-Z, Zhu H-L, Qu G-Q, Luo Y-B, Gao Y, Jiang C-Z, Fu D-Q. Functional redundancy of transcription factors SlNOR and SlNOR-like1 is required for pollen development in tomato. Hortic Res. 2025:12(4):uhaf003. https://doi.org/10.1093/hr/ uhaf003 Li S, Chen K, Grierson D. Molecular and hormonal mechanisms regulating fleshy fruit ripening. Cells. 2021:10(5):1136. https://doi.org/ 10.3390/cells10051136 Li S, Zhu B, Pirrello J, Xu C, Zhang B, Bouzayen M, Chen K, Grierson D. Roles of RIN and ethylene in tomato fruit ripening and ripening-associated traits. New Phytol. 2020:226(2):460–475. https://doi.org/10.1111/nph.16362 Liu R, How-kit A, Stammitti L, Teyssier E, Rolin D, Mortain-bertrand A, Halle S, Liu M, Kong J, Wu C, et al. A DEMETER-like DNA demethylase governs tomato fruit ripening. Proc Natl Acad Sci U S A. 2015:112(34): 10804–10809. https://doi.org/10.1073/pnas.1503362112 Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2-ΔΔCT method. Methods. 2001:25(4):402–408. https://doi.org/10.1006/ meth.2001.1262 Lü P, Yu S, Zhu N, Chen Y-R, Zhou B, Pan Y, Tzeng D, Fabi JP, Argyris J, Garcia-Mas J, et al. Genome encode analyses reveal the basis of convergent evolution of fleshy fruit ripening. Nat Plants. 2018:4(10):784–791. https://doi.org/10.1038/s41477-018-0249-z Lykke-Andersen S, Jensen TH. Nonsense-mediated mRNA decay: an intricate machinery that shapes transcriptomes. Nat Rev Mol Cell Biol. 2015:16(11):665–677. https://doi.org/10.1038/nrm4063 McMurchie EJ, McGlasson WB, Eaks IL. Treatment of fruit with propylene gives information about the biogenesis of ethylene. Nature. 1972:237(5352):235–236. https://doi.org/10.1038/237235a0 Niu Q, Xu Y, Huang H, Li L, Tang D, Wu S, Liu P, Liu R, Ma Y, Zhang B, et al. Two transcription factors play critical roles in mediating epigenetic regulation of fruit ripening in tomato. Proc Natl Acad Sci U S A. 2025:122(15):e2422798122. https://doi.org/10.1073/pnas. 2422798122 Oeller PW, Lu MW, Taylor LP, Pike DA, Theologis A. Reversible inhibition of tomato fruit senescence by antisense RNA. Science. 1991:254:437–439. https://doi.org/10.1126/science.1925603 Peng Z, Li H, Liu G, Jia W, Fu D. NAC transcription factor NOR-like1 regulates tomato fruit size. Planta. 2023:258(1):9. https://doi.org/ 10.1007/s00425-023-04166-4 Pratt HK, Goeschl JD. Physiological roles of ethylene in plants. Annu Rev Plant Biol. 1969:20(1):541–584. https://doi.org/10.1146/ annurev.pp.20.060169.002545 Robinson R, Tomes M. Ripening inhibitor: a gene with multiple effects on ripening. Rep Tomato Genet Coop. 1968:18:36–37. Schouten RE, Farneti B, Tijskens LMM, Alarcón AA, Woltering EJ. Quantifying lycopene synthesis and chlorophyll breakdown in tomato fruit using remittance VIS spectroscopy. Postharvest Biol Technol. 2014:96:53–63. https://doi.org/10.1016/j.postharvbio. 2014.05.007 Slaman E, Lammers M, Angenent GC, de Maagd RA. High-throughput sgRNA testing reveals rules for Cas9 specificity and DNA repair in tomato cells. Front Genome Ed. 2023:5:1196763. https://doi.org/10. 3389/fgeed.2023.1196763 Thompson AJ, Tor M, Barry CS, Vrebalov J, Orfila C, Jarvis MC, Giovannoni JJ, Grierson D, Seymour GB. Molecular and genetic characterization of a novel pleiotropic tomato-ripening mutant. Plant Physiol. 1999:120(2):383–389. https://doi.org/10.1104/pp.120. 2.383 Tieman DM, Taylor MG, Ciardi JA, Klee HJ. The tomato ethylene receptors NR and LeETR4 are negative regulators of ethylene response and exhibit functional compensation within a multigene family. Proc Natl Acad Sci U S A. 2000:97(10): 5663–5668. https://doi.org/10.1073/pnas.090550597 Tigchelaar E, Tomes M, Kerr E, Barman R. A new fruit-ripening mutant, non-ripening (nor). Rep Tomato Genet Coop. 1973:23:33–34. van Roekel JS, Damm B, Melchers LS, Hoekema A. Factors influencing transformation frequency of tomato (Lycopersicon esculentum). Plant Cell Rep. 1993:12(11):644–647. https://doi.org/10.1007/ BF00232816 Vrebalov J, Ruezinsky D, Padmanabhan V, White R, Medrano D, Drake R, Schuch W, Giovannoni J. A MADS-box gene necessary for fruit ripening at the tomato ripening-inhibitor (rin) locus. Science. 2002:296:343–346. https://doi.org/10.1126/science.1068181 Wang D, Samsulrizal NH, Yan C, Allcock NS, Craigon J, Blanco-Ulate B, Ortega-Salazar I, Marcus SE, Bagheri HM, Perez-Fons L, et al. Characterization of CRISPR mutants targeting genes modulating pectin degradation in ripening tomato. Plant Physiol. 2019a:179: 544–557. https://doi.org/10.1104/pp.18.01187 Wang R, Angenent GC, Seymour G, de Maagd RA. Revisiting the role of master regulators in tomato ripening. Trends Plant Sci. 2020:25(3):291–301. https://doi.org/10.1016/j.tplants.2019.11.005 Wang R, Tavano ECDR, Lammers M, Martinelli AP, Angenent GC, de Maagd RA. Re-evaluation of transcription factor function in tomato fruit development and ripening with CRISPR/ Cas9-mutagenesis. Sci Rep. 2019b:9(1):1696. https://doi.org/10. 1038/s41598-018-38170-6 Weber E, Engler C, Gruetzner R, Werner S, Marillonnet S. A modular cloning system for standardized assembly of multigene constructs. PLoS One. 2011:6(2):e16765. https://doi.org/10.1371/ journal.pone.0016765 Werner S, Engler C, Weber E, Gruetzner R, Marillonnet S. Fast track assembly of multigene constructs using golden gate cloning and the MoClo system. Bioeng Bugs. 2012:3(1):38–43. https://doi. org/10.4161/bbug.3.1.18223 Zhang RX, Liu Y, Zhang X, Chen X, Sun J, Zhao Y, Zhang J, Yao JL, Liao L, Zhou H, et al. Two adjacent NAC transcription factors regulate fruit maturity date and flavor in peach. New Phytol. 2023:241(2): 632–649. https://doi.org/10.1111/nph.19372 Zhong S, Fei Z, Chen Y-R, Zheng Y, Huang M, Vrebalov J, McQuinn R, Gapper N, Liu B, Xiang J, et al. Single-base resolution methylomes of tomato fruit development reveal epigenome modifications associated with ripening. Nat Biotechnol. 2013:31(2):154–159. https:// doi.org/10.1038/nbt.2462 Zhou L, Tian S, Qin G. RNA methylomes reveal the m6A-mediated regulation of DNA demethylase gene SlDML2 in tomato fruit ripening. Genome Biol. 2019:20(1):156. https://doi.org/10.1186/ s13059-019-1771-7 Zhu M, Chen G, Zhou S, Tu Y, Wang Y, Dong T, Hu Z. A new tomato NAC (NAM/ATAF1/2/CUC2) transcription factor, SlNAC4, functions as a positive regulator of fruit ripening and carotenoid accumulation. Plant Cell Physiol. 2014:55(1):119–135. https://doi.org/10. 1093/pcp/pct162 Zude M. Comparison of indices and multivariate models to nondestructively predict the fruit chlorophyll by means of visible spectrometry in apple fruit. Anal Chim Acta. 2003:481(1): 119–126. https://doi.org/10.1016/S0003-2670(03)00070-9 16 | Plant Physiology, 2025, Vol. 198, No. 3 Downloaded from https://academic.oup.com/plphys/article/198/3/kiaf291/8178771 by Azienda Ospedaliera di Padova - Registro Tumori del Veneto user on 24 October 2025