In our previous studies, a 3-week drought treatment decreased tuber yield in a half-sib family of Katahdin-derived potato cultivars (Sołtys-Kalina et al. 2016; Plich et al. 2020). In the present experiments, a 2-week period of drought stress was probably too short to cause statistically significant differences in the mean tuber yield per plant in the cultivars Cayuga, Dalila, Katahdin, Pontiac, Sebago, and Seneca (Table S1). However, this stress significantly reduced average tuber weight (ATW) in the cultivars Katahdin and Pontiac. ATW was significantly lower (p value ≤ 0,001) in the drought-stressed plants (R, recovered plants) of the cultivars Katahdin (26% decrease) and Pontiac (31% decrease) in relation to the control plants (C) (Fig. 1a). This may imply a decrease in tuber marketability for the cultivars Katahdin and Pontiac. Tubers harvested from C and R plants of the cultivar Katahdin are shown in Fig. 1b.
The canopy response to drought was analysed based on the JIP test, measuring fluorescence parameters that indicate the efficiencies and fluxes of electrons and energy around PSI and PSII in plants subject to stress events (Boguszewska-Mańkowska et al. 2020). Twelve JIP parameters showed statistically significant differences between C and R plants (Table S2). In Katahdin, the value of the JIP parameter PItotal reached 122% of that in the control, whereas a decrease in that parameter was observed in other cultivars. PItotal is used as an indicator of drought (Viljevac Vuletić et al. 2022, and references therein). Principal component analysis (PCA) performed based on the main components, PC1 and PC2, separated R plants of the cultivars Pontiac and Katahdin as showing most distinct in responses to drought (Fig. 1c). Based on ATW and PCA, Katahdin and Seneca (both late-maturing cultivars) were selected as models for the comparisons of the organelle/nuclear DNA ratio, cell cycle progression and gene expression.
In plants, the degradation and recycling of unnecessary cytosolic structures maintain cellular homeostasis under stress conditions (Cao et al. 2021). In our study, R plants of the cultivar Seneca showed a plastid/nuclear DNA ratio (pt/nucDNA) that was 1.96-fold (P < 0.001) lower than that of the C plants. For the mitochondrial/nuclear DNA (mt/nucDNA) ratio, the corresponding value for Seneca was 1.37-fold (P < 0.001) lower. No significant differences were observed in Katahdin (Fig. 1d). We postulate that the drought-induced autophagic machinery was expressed more efficiently in the cultivar Seneca.
There is a close association between cell cycle progression and plant adaptation to drought stress (Qi and Zhang 2020). The balance between cell cycle progression and adaptation (e.g., growth inhibition) to dynamic environmental conditions is critical for adaptive success. In the leaves of Katahdin and Seneca that had recovered from drought, cell cycle progression was evaluated in the middle part of the lamina and compared to the reaction in control plants. In the cultivar Seneca, cell cycle progression was modulated in response to drought stress, whereas no significant changes in the cell cycle were observed in Katahdin (Fig. 1e). These modifications were related to a decreasing frequency of nuclei in S phase (54.3%) and an increased frequency of nuclei in G0/G1 phase (17.5%). A decrease in the frequency of nuclei in G2 phase (33.3%) was also noted (Fig. 1f). The cell cycle arrest in G0/G1 phase observed in the cultivar Seneca can be interpreted as a component of plant adaptation to drought. This delay in cell cycle progression allows the repair of DNA damage and the initiation of programmed cell death or senescence to avoid unfavourable changes in progeny cells in response to stress treatment.
Differentially expressed genes (DEGs) were identified in the cultivars Katahdin and Seneca. DEGs (P value < 0.05) were selected by comparing the corresponding C and R samples (Tables S3-S6). We identified higher numbers of both up- and downregulated genes in the leaves of Seneca (1665 and 1106, respectively) than in the leaves of Katahdin (416 and 155, respectively) (Figure S1). PCA showed that the transcriptional profile was more dependent on the plant cultivar than on the treatment. PC1 clearly separated the cultivars and explained 58% of the variance, while PC2 explained 24% of the variance and separated the control and drought-stressed plants. PCA indicated a closer relationship of Katahdin plants grown under control and drought conditions compared to Seneca plants, for which the grouping of plants representing different experimental conditions was more prominent (Fig. 1g). Subsequently, we focused on cell cycle-related genes (Fig. S2). Among these genes, three were significantly upregulated only in Katahdin, three only in Seneca, and one in both cultivars (Fig. 1h), while no downregulated genes were revealed. Two of the seven DEGs were classified as D-type cyclins (CYCD6-1, CYCD3-2), two were classified as A-type cyclins (CYCA1-1, CYCA2-1) and the remaining gene CYCB1-2 belonged to the B-type cyclins. These genes are downstream targets of cyclin-dependent kinases (CDKs) that coordinate cell cycle progression (Carneiro et al. 2021; Qi and Zhang 2020; Saidi and Hajibarat 2021). In plants, MYB3R4 transcription factors regulate the expression of G2/M-specific genes (Carneiro et al. 2021). In our study, the transcript level of one MYB3R4 gene was upregulated in R plants of the cultivar Katahdin (Fig. 1h). Retinoblastoma-related protein plays a multifunctional role in cell cycle regulation. The proteins encoded by retinoblastoma-related (RBR) genes are negative regulators of the G1/S transition (Banerjee et al. 2020). Here, the RBR homologue RBL901 was one of four genes related to the regulation of the cell cycle that showed significantly higher expression in R plants of Seneca than in the corresponding C plants, while in Katahdin, its expression was not significantly changed by drought stress (Fig.1h). It should be emphasized that this gene presented the highest expression among all DEGs encoding proteins controlling cell cycle progression (Fig. 1h). In summary, we report for the first time that cell cycle progression in the leaves of drought-stressed potato plants likely plays a significant role in the final tuber yield distribution. We found that the level of the RBL901 transcript was related to the cell cycle progression data, where the increase in the frequency of nuclei in G0/G1 phase was accompanied by a decreased frequency of nuclei in S phase. We postulate that RBL901 might be a novel candidate factor involved in the regulation of the potato response to drought stress.