**4. Selection of stay‐green genotypes in sunflower**

Given the importance of the stay-green genotypes selection for studies of this complex char‐ acter, we performed a screening and selection of contrasting sunflower genotypes associ‐ ated with early leaf senescence process evaluating 135 different genotypes from the INTA Sunflower Breeding Program (INTA Manfredi Sunflower Germplasm Collection) [61] growing under field conditions through a physiological, cytological and molecular approach.

Physiological measurement of growing cycle, anthesis time, number of leaves, plant size, evo‐ lution of total/dry leaf, green leaf area at anthesis time and SPAD was performed to select pairs of contrasting genotypes with very similar plant architecture, phenology and leaf area until anthesis, but with different senescence rate. This analysis allowed us to select 10 geno‐ types that were further evaluated, allowing the identification of two contrasting senescence inbred lines, R453 (early senescence genotype) and B481-6 (putative stay-green genotype) [23] (**Figure 3**).

by both approaches, highlighting them as potentially robust biomarkers associated with leaf senescence in sunflower. This complementary use of network and BioSignature Discoverer analysis of transcriptomic and metabolomic data provided a useful tool for identifying candi‐ date genes and metabolites, which may have a role in the triggering and development of the

**Figure 2.** Integrated network of hubs metabolites and transcription factors detected by WGCNA [39]. 24 selected hubs metabolites and 82 TFs statistically significant during senescence were correlated (p-value < 0.0001) and visualized in

Transcriptomic analysis in combination with metabolic profile is a comprenhensive tool for the study of leaf senescence. These results suggest a complex regulatory network underlying this process. Thus, the identification of regulatory networks based on expression profiling is an important starting point for the detection of new key genes involved in the triggering of

Given the importance of the stay-green genotypes selection for studies of this complex char‐ acter, we performed a screening and selection of contrasting sunflower genotypes associ‐ ated with early leaf senescence process evaluating 135 different genotypes from the INTA Sunflower Breeding Program (INTA Manfredi Sunflower Germplasm Collection) [61] growing

under field conditions through a physiological, cytological and molecular approach.

leaf senescence process.

34 Senescence - Physiology or Pathology

the senescence process in this crop.

**4. Selection of stay‐green genotypes in sunflower**

Cytoscape [60] by degree (node size and color) and edge weight (edge size and color).

**Figure 3.** Phenotypic analysis of sunflower genotypes under field experiment. (A) and (B) early senescence genotype R453. (C) and (D) Stay-green genotype B481-6. Images (A) and (C) correspond to 15 days after anthesis and images (B) and (D) 30 days after anthesis.

At the physiological level, green leaf area evolution (GLA) is an indirect measurement of photosynthetically active leaf area, and its decrease has been reported as product of active chloroplast degeneration and chlorophyll degradation.

R453 and B481-6 displayed similar GLA evolution until anthesis. Then, GLA decreased abruptly in R453 and faster than the B481-6 genotype. GLA decline in B481-6, on the other hand, was gradual and reached complete senescence at 200 ºCdAE later (**Figure 4A**). In sunflower, and many other monocarpic species, this senescence symptom is evident after anthesis, during grain-filling period, and is mainly due to source-sink relationships estab‐ lished at this stage of development [5, 20, 36, 62]. Radiation interception at the canopy level showed similar patterns with an early decrease in the early senescence genotype (**Figure 4C**). In grain crops, a delay in leaf senescence should have a positive impact on grain yield [3, 26]. Yield components were evaluated displaying significant differences in yield, with higher seed weight in the stay-green genotype (**Figure 4B**). These observations are in agreement with the expected for this trait suggesting a type B stay-green pheno‐ type [28]. Moreover, photosynthesis measurement was performed 15 days after anthesis showing higher photosynthesis rate in the stay-green genotype (**Figure 4D**), supporting this finding.

**Figure 4.** Physiological analysis of contrasting genotypes. R453 in orange and B481-6 in green colors. (A) Green leaf area (GLA); (B) yield; (C) radiation interception and (D) photosynthesis. ºCdAE indicates thermal time after emergence. Red line in (A) and (C) indicates anthesis time for both genotypes.

To confirm the senescence phenotype of the selected genotypes, we also performed cytological and molecular analysis. As senescence involves programed Cell Death (PCD), nuclear DNA degradation associated with PCD can be detected in situ by TUNEL assay. In this sense, we analyzed mesophyll cells nuclei of both genotypes 10 days after anthesis. R453, with premature senescence phenotype, also showed TUNEL-positive nuclei, whereas B481-6, with stay-green phenotype, had TUNEL-negative nuclei (**Figure 5**). Mesophyll is the most photosynthetically active tissue of higher plants, having cells with high chloroplast and chlorophyll content and the cells in this tissue are affected firstly during senescence [9, 62, 63]. TUNEL-positive nuclei detection indicates that senescence process has already started in the early senescence genotype.

The gene expression pattern of the senescence-associated candidate transcription factors was evaluated for differences in the timing of senescence pathways activation between the early senescence and stay-green genotype [21, 22]. HaNAC01, HaNAC03 and HaNAC05 transcrip‐ tion factors were evaluated at two different times, 5 days after anthesis (DAA) and 15 days after anthesis (**Figure 6**). R453 showed higher expression levels of the three NAC transcription factor than B481-6, and its expression increased by 15 days post-anthesis.

Altogether, these findings highlight these genotypes as interesting potential candidates for fur‐ ther analysis of leaf senescence in sunflower. The B481-6 genotype showed a stay-green phe‐ notype, also evidenced by cytological and molecular analysis and an increase of seed weight,

At the physiological level, green leaf area evolution (GLA) is an indirect measurement of photosynthetically active leaf area, and its decrease has been reported as product of active

R453 and B481-6 displayed similar GLA evolution until anthesis. Then, GLA decreased abruptly in R453 and faster than the B481-6 genotype. GLA decline in B481-6, on the other hand, was gradual and reached complete senescence at 200 ºCdAE later (**Figure 4A**). In sunflower, and many other monocarpic species, this senescence symptom is evident after anthesis, during grain-filling period, and is mainly due to source-sink relationships estab‐ lished at this stage of development [5, 20, 36, 62]. Radiation interception at the canopy level showed similar patterns with an early decrease in the early senescence genotype (**Figure 4C**). In grain crops, a delay in leaf senescence should have a positive impact on grain yield [3, 26]. Yield components were evaluated displaying significant differences in yield, with higher seed weight in the stay-green genotype (**Figure 4B**). These observations are in agreement with the expected for this trait suggesting a type B stay-green pheno‐ type [28]. Moreover, photosynthesis measurement was performed 15 days after anthesis showing higher photosynthesis rate in the stay-green genotype (**Figure 4D**), supporting

**Figure 4.** Physiological analysis of contrasting genotypes. R453 in orange and B481-6 in green colors. (A) Green leaf area (GLA); (B) yield; (C) radiation interception and (D) photosynthesis. ºCdAE indicates thermal time after emergence. Red

line in (A) and (C) indicates anthesis time for both genotypes.

chloroplast degeneration and chlorophyll degradation.

36 Senescence - Physiology or Pathology

this finding.

**Figure 5.** TUNEL assays of selected contrasting genotypes. (A) and (C): Nuclei visualization by DAPI staining. (B): TUNEL-positive nuclei in mesophyll cells of premature senescence genotypes (R453). (D): TUNEL-negative nuclei in stay-green genotype (B481-6) [23].

**Figure 6.** qPCR analysis of NAC transcription factors candidate genes. Relative transcript levels are shown as the ratio (log2 scale) between the expression in the early senescence genotype (R453) in relation to the stay-green genotype (B481-6) in two different times 5 days after anthesis (DAA) and 15 days after anthesis [23].

which makes this genotype a potential candidate for the functional stay-green phenotype in comparison with R453 genotype.
