Abstract
Chilling damage can cause major reductions in cucumber (Cucumis sativus L.) yield. Cucumber plants can withstand a chilling event (i.e., tolerance and susceptibility), in which response is dictated primarily by maternally inherited plastid genomes or by the biparental contribution of a nuclear factor. To examine the modes of inheritance, exact reciprocal backcross cucumber populations (BC1–5), were created by crossing ‘Chipper’ (chilling-tolerant plastid, susceptible nucleus) and line North Carolina State University (NCSU) M29 (chilling-susceptible plastid, susceptible nucleus). These progeny and their parents were subjected to chilling stress [5.5 h at 4 °C in 270 μmol·m−2·s−1 photosynthetic photon flux (PPF) cool white lighting] at the first true-leaf stage. The chilling response of individuals possessing either NCSU M29 or ‘Chipper’ cytoplasm in any generational comparison was not significantly different (P > 0.05) from that of the maternal plastid source (susceptible or tolerant). Moreover, lines within a plastid type did not differ significantly (P > 0.05) in chilling response despite unequal nuclear contributions demonstrating the absence of nuclear additive or dosage effects originating in ‘Chipper’ or NCSU M29. Additionally, line NC-76, previously identified as a nuclear source of chilling tolerance, performed intermediate to ‘Chipper’ and NCSU M29 in chilling response under these stress conditions. The F1 progeny derived from crossing both BC5 plastidic response types (susceptible and tolerant) with NC-76 (paternal parent) performed comparable to their plastid donors and were significantly different (P < 0.0001) from one another despite their heterozygous nuclear nature resulting from the contribution of the nuclear chilling-tolerant factor contributed by NC-76. The response of tolerant and susceptible BC5 lines (i.e., ‘Chipper’ plastid in the NCSU M29 background and NCSU M29 plastid in ‘Chipper’ background, respectively) was reversible by crossing BC progeny with an alternate chilling-response plastid type. It is concluded that under these chilling conditions, plastid effects determine tolerance or susceptibility in the cucumber germplasms examined.
Associating stress physiology with a specific genotype that can be manipulated in an applied breeding program is frequently a complex task (Kreps et al., 2002), because related biochemical pathways are often shared in the form of “co-tolerances” (Choi et al., 2002; Foolad et al., 1999; Iturriaga et al., 2006; Kant et al., 2008; Komori and Imaseki, 2005; Sabehat et al., 1998; Takahashi et al., 2005). However, despite such complications, the study and classification of genetic factors conferring tolerance to abiotic stressors such as chilling temperatures are an important agricultural goal of many plant improvement programs.
Chilling response can have a direct effect on yield, particularly in crops grown outside their recommended climatic zones (Lynch, 1990). For example, cucumber is a tropical plant often sown in temperate regions where temperatures often fluctuate dramatically in the early spring as seedlings germinate and emerge. Cucumber is very sensitive to chilling, and recovery period from such challenges is relatively long (Chung et al., 2003; Staub and Bacher, 1997). Symptoms associated with chilling injury (CI) in cucumber seedlings include poor germination and emergence, seedling death, and yellow or necrotic tissue. Mature fruit may experience surface pitting, tissue deterioration, yellowing, water soaking, and increased levels of both water loss and greater susceptibility to pathogens after a chilling challenge. Maternal (Chung et al., 2003) and nuclear (Kozik and Wehner, 2008) inheritance patterns have been proposed as genetic explanations of response to chilling temperatures (less than 12 °C) in cucumber.
Complex nuclear–plastid intercommunications are present within plant cells (Goldschmidt-Clermont, 1998; Gray and Greenberg, 1992; Mayfield, 1990; Rodermel, 2001; Susek and Chory, 1992), and as such, it is likely that both entities have the capacity for affecting some component of chilling response. It remains unclear in many plant species to what degree each interacts with the other or, more importantly, where the abiotic stress response originates for particular developmental stages and what environmental conditions elicit adverse effects. The maternally inherited chilling response in cucumber documented by Chung et al. (2003) was followed by a report by Kozik and Wehner (2008) suggesting that there is also a nuclear factor that influences chilling response. Not only did the temperature and duration of chilling differ between the studies of Chung et al. (2003) and Kozik and Wehner (2008), but also the genetic background of the accessions evaluated varied [i.e., ‘Chipper’ (tolerant) and Gy14 (susceptible) in the former and NC-76 (tolerant), ‘Chipper’ (slightly tolerant), and Gy14 (susceptible) in the latter]. Moreover, neither of these studies considered the consequence of additive or dosage effects contributed by cytoplasmic or nuclear factors on chilling response.
Chung et al. (2007) identified putative plastidic single nucleotide polymorphisms (SNPs) (bp locations: 4813, 56561, and 126349) associated with CI in cucumber (susceptibility and tolerance). However, the role of the SNPs as genetic factors controlling specific biochemical pathways has not been defined. Given the differing modes of inheritance proposed for CI (Chung et al., 2003; Kozik and Wehner, 2008) and the lack of understanding of potential cytoplasmic and nuclear interactions during chilling in cucumber (Chung et al., 2007), a study was designed to: 1) elucidate the expression of the maternally inherited chilling response using exact reciprocal backcross generations created from crosses between ‘Chipper’ and NCSU M29 (susceptible); 2) determine whether the effects of maternal factors for chilling response are influenced by nuclear genes for chilling tolerance using F1 progeny that combined both the plastidic components as well as the nuclear source of chilling tolerance from NC-76 (Kozik and Wehner, 2008), and; 3) determine the effects of “re-introgression” of a plastid response type into its original nuclear background. Not only will this study provide a better understanding of the genetics of CI, but it will allow for the development of chilling tolerant germplasm for cucumber improvement.
Materials and Methods
Germplasm.
Three cucumber lines were chosen for experimentation based on both their previously determined response to chilling temperatures (Kozik and Wehner, 2008; Smeets and Wehner, 1997; Wehner, 1982) and their potential for development of commercially acceptable germplasm. Although NCSU M29 (P2) is chilling-susceptible, both ‘Chipper’ (P1) and NC-76 (P3) are chilling-tolerant based on these previous studies. These lines were obtained from the U.S. Department of Agriculture–Agricultural Research Service cucumber breeding program in Madison, WI (‘Chipper’ and NCSU M29, hereafter designated as CH and M29, respectively) and the NCSU cucurbit breeding program in Raleigh, NC (NC-76).
Experimental progeny.
An initial exact reciprocal cross was made between two (one each) CH (P1) and M29 (P2) plants by hand pollination in a greenhouse at the University of Wisconsin (UW), Madison. These and their derived advanced backcross progenies were used in three experiments to examine various hypotheses concerning the effects on chilling response of different plastid types. Strategic comparisons were made because highly inbred recurrent parental plants were self-pollinated each generation, providing pollen for each successive generation of crossing.
Random F1 progeny were successively backcrossed to their paternal genetic donor [i.e., BC1-BC5, Expt. 1 (Fig. 1A)]. Advanced reciprocal BC5 progeny, resulting from the initial crosses between CH and M29 [CH × M29 (CBC5, synonym P4) and M29 × CH (MBC5, synonym P5)], were crossed to NC-76 (P3) to create F1 progeny [Expt. 2 (Fig. 1B)]. In this case, F1 progeny possessing both nuclear and plastidic chilling response factors were created using CBC5 [P4 (line possessing the tolerant CH plastid with theoretically greater than 98% of M29 nuclear background)] and MBC5 [P5 (line possessing the susceptible M29 plastid with theoretically greater than 98% of CH nuclear background)] as maternal parents in matings with NC-76 as the pollen donor. In addition, CBC5 and MBC5 were crossed to M29 and CH (maternal parents), respectively, to create re-introgression F1 progeny (i.e., introduction of the original tolerant or susceptible plastid type into advanced backcross progeny possessing the original nuclear material) [Expt. 3 (Fig. 1C)]. This allowed for a single mating to provide progeny for a direct comparison (e.g., CBC5 × M29 versus M29 × CBC5) of chilling response with plastidic factors as the sole variable.
A diagrammatic representation of three cucumber chilling experiments conducted under controlled conditions [5.5 h (0800 to 1330 hr) at 4 °C (immediate and constant) where the light level was 270 μmol·m−2·s−1 photosynthetic photon flux (PPF) supplied by only cool white lamps]. Expt. 1 (A) tested responses for exact reciprocal backcross populations (BC1–5) created from cucumber lines ‘Chipper’ (CH) and NCSU M29 (M29). Expt. 2 (B) evaluated chilling response of combined plastid and nuclear tolerant progeny resulting from a combination of both nuclear and plastid sources of cucumber chilling tolerance (NC-76 and CH, respectively). Expt. 3 (C) tested the response of F1 “re-introgression” progeny, derived in two cytoplasmic backgrounds (CH and M29), to directly assess hypothesized tolerant versus susceptible plastidic genotypes.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 136, 4; 10.21273/JASHS.136.4.256
Pre-chilling growing conditions.
Parents and experimental progeny, from all generations for the three experimental setups, were simultaneously and completely randomized, assigned numbers, and sown in peat pots (185 cm3) containing a sterilized 1 sand:1 peatmoss mixture. Each parental line/hybrid/generation was represented by six individuals in each of four replications (i.e., 24 individuals). The seeds were germinated in the greenhouse at 29 °C (daylight conditions)/24 °C (dark conditions) with an applied bottom heat of 35 °C to the germination flats. Seedlings, in the cotyledon stage, were then transported into a controlled environmental chamber in the UW–Madison Biotron (Madison, WI). Seedlings were grown at 22 °C (light)/18 °C (dark) under a 9-h photoperiod at a light level of 600 μmol·m−2·s−1 PPF (0800 to 1700 hr) supplied by cool-white fluorescent and incandescent lamps according to Chung et al. (2003). Relative humidity (RH) was held at 60%. Seedlings were watered, at soil level, twice per day to saturation and fertilized with 20N–8.8P–16.6K fertilizer (Peter's; Scotts, Marysville, OH) once per week.
Chilling protocol and damage ratings.
Seedlings with first-true leaves emerged and fully opened with no remaining adaxial leaf curl were subjected to a chilling treatment of 5.5 h (0800 to 1330 hr) at 4 °C (immediate and constant) where the light level was 270 μmol·m−2·s−1 PPF supplied by only cool white lamps according to Chung et al. (2003). This provided a uniform physiological age for chilling damage assessment.
During chilling treatment, RH was 52%. Before chilling, seedlings were watered at the soil level only to the extent of eliminating dryness and were provided the standard watering on conclusion of the chilling protocol. After chilling, plants were returned to pretreatment conditions, and leaf damage was quantified by visual rating 5 d after the chilling treatment.
Visual ratings on a total of 288 individuals [136 tolerant plastid-type (CH maternal source) and 152 susceptible plastid-type (M29 maternal source)] were made using a previously determined scale ranging from 0 to 9, in which 0 = no damage, 1 to 2 = trace injury, 3 to 4 = slight injury, 5 to 6 = moderate damage, 7 to 8 = advanced damage, and 9 = plant death (Smeets and Wehner, 1997).
Data analysis.
Visual rating data were subjected to analysis of variance (ANOVA) using the PROC GLM procedure in SAS (Version 9; SAS Institute, Cary, NC). Generation least squares means, sd, and their attending 95% confidence limits (CLs) were calculated for each experiment (three). Strategically designed parent and progeny performance comparisons were made using a t test analysis (P < 0.01).
Notation of plastidic types.
Parents and cross-progeny were given the plastid chilling sensitivity designation of s for susceptible and t for tolerant, in which sensitivities are presented as prefixes in describing generation plastid status and original maternal source (e.g., tCF1 signifying CH-tolerant plastid type in the F1 generation). Additionally, the annotation, - and --, is used here to designate plastid and nuclear constitution of a genotype, respectively. For example, backcross CBC5 progeny (putative chilling-tolerant) are designated as t/chch, whereas MBC5 progeny (putative chilling-susceptible) are given the notation of s/chch. Neither CH nor M29 is known to have tolerant nuclear genes, and, thus, their nuclear contribution is annotated as chch. In contrast, the chilling-tolerant nuclear genome of NC-76 is denoted as ChCh and Chch when in the homozygous and heterozygous state, respectively. Because the notation s defines a susceptible plastid type, the mating, CBC5 × NC-76, creates an F1 with a plastid and nuclear notation, t/Chch, whereas progeny of a MBC5 × NC-76 mating are designated s/Chch.
Results
The lack of adequate seed in progeny generations precluded multiple chilling evaluations and, thus, a single completely randomized experiment without blocking was designed to include a large number of plants (experimental unit). This design evaluated a larger sampling per entry (average sample per entry = 23) than previously described in cucumber chilling experiments using the same conditions as those described here [average sample size per entry = six individuals (Smeets and Wehner, 1997) and 17 individuals (Chung et al., 2003)] as well as providing an opportunity for data collection across all entries simultaneously. This methodology also provided a precise assessment of damage ratings (i.e., increased discrimination of damage rating classes) while minimizing the effects of environment (i.e., a single post-chilling treatment recovery site). Additionally, all parents and cross-progeny were simultaneously compared under identical conditions in all specific cross-comparisons across all experimental categories.
Expt. 1: exact reciprocal populations.
Expt. 1 tested whether backcross progeny (i.e., BC1 to BC5) exhibited chilling tolerance equivalent to their respective maternal parent and whether chilling response during backcrossing was altered by the introgression of nuclear factors resident in the susceptible and tolerant parental lines (Fig. 1A). The mean chilling ratings of tolerant and susceptible parental plastid sources CH (P1) and M29 (P2) were 2.5 (1.0 sd, 95% CL = 2.0 to 3.1) and 5.4 (1.7 sd, 95% CL = 4.8 to 6.0), respectively (Table 1). Although reciprocal cross-progenies of each generation [i.e., F1 (P1 × P2 and P2 × P1) and BC1–5 (BC1P2 – BC5P2 and BC1P1 – BC5P1)] differed (P < 0.0001), progeny performance within any one population was not different. Thus, although no differences were detected between chilling responses of cross-progeny plants containing the same maternally inherited plastid type, the response of parental CH and M29 lines differed from progeny (F1 and BC1–5) in which they were the paternal recurrent nuclear donor. This indicated that their respective additive nuclear contributions did not influence chilling response.
Mean, sd, and 95% confidence limits of chilling (4 °C for 5.5 h at 270 μmol·m−2·s−1 PPF) injury ratings in parental cucumber lines (‘Chipper’ and NCSU M29) and their reciprocal F1, BC1, BC2, BC3, BC4, and BC5 progeny (Expt. 1).z
Analysis of maternal effects indicated that the performance of CH-derived backcross progeny (136) did not differ from their plastid donor parent (P = 0.44) (Table 1). Likewise, the chilling response of M29-derived progeny (152) did not differ from M29 [P = 0.76 (susceptible plastid)]. However, CH- and M29-derived progeny in alternate cytoplasmic backgrounds [i.e., CH versus F1 and BC1–5 (having the M29-type plastid type)] and [M29 versus F1 and BC1–5 (having the CH-type plastid type)] differed (P < 0.0001) in their chilling response. Thus, the chilling response observed in specific cross-progeny (CH- or M29-plastid background) persisted across generations (F1 to BC5) despite an increasing nuclear contribution [(1/2)n] of the paternal recurrent donor with each backcross and an equivalent decrease in nuclear contribution of the maternal (non-recurrent) parent.
Expt. 2: cumulative plastid and nuclear chilling response
Expt. 2 tested whether the chilling response of F1-generated plastid types (s = susceptible and t = tolerant) differed in the presence of the nuclear NC-76 Ch chilling tolerance gene (Fig. 1B). Although the mean chilling ratings of tolerant (CH) and susceptible (M29) lines were 2.5 (1.0 sd, 95% CL = 2.0 to 3.1) and 5.3 (1.7 sd, 95% CL = 4.8 to 6.0), respectively, the mean chilling response rating of the tolerant line NC-76 [P3 (nuclear source)] was 3.7 (1.1 sd, 95% CL = 3.1 to 4.4) (Table 2; Fig. 2).
Mean, sd, and 95% confidence limits of chilling (4 °C for 5.5 h at 270 μmol·m−2·s−1 PPF) injury ratings in parental cucumber lines (‘Chipper’, NCSU M29, and NC76) and their derived F1 progeny created by advanced backcross and parental matings (BC5 × NC76) (Expt. 2).z
Chilling responses of cucumber parental lines and F1 progeny possessing both nuclear and plastidic chilling tolerance, in which NC-76, Ch = nuclear tolerance, and ‘Chipper’ and NCSU M29 (M29) possessing plastidic tolerance (t) and susceptibility (s), respectively; and F1 progeny are tolerant (A; CBC5 × NC-76) and susceptible (B; MBC5 × NC-76). Plants were chilled for 5.5 h (0800 to 1330 hr) at 4 °C (immediate and constant) where the light level was 270 μmol·m−2·s−1 photosynthetic photon flux (PPF) supplied by only cool white lamps.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 136, 4; 10.21273/JASHS.136.4.256
The F1 progeny derived from crossing both BC5 plastidic response types (susceptible and tolerant) with NC-76 (paternal tolerant parent) performed similarly to their plastid donors and were significantly (P < 0.0001) different from one another despite the heterozygous nuclear nature resulting from the contribution of the nuclear tolerant factor present in NC-76 (Table 2). The mean chilling response ratings of CBC5 and MBC5 backcross progeny were 2.7 (1.3 sd, 95% CL = 2.1 to 3.3) and 4.3 (1.6 sd, 95% CL = 3.7 to 4.9), respectively. The mean chilling response rating of CF1 progeny [tCF1 (P4 × P3)], possessing the tolerant plastid in an ≈50% M29 (P2):50% NC-76 (P3) nuclear background, was 2.2 (0.6 sd, 95% CL = 1.6 to 2.8). In contrast, MF1 progeny [sMF1 (P5 × P3)], with a susceptible plastid in an ≈50% CH (P1):50% NC-76 (P3) nuclear background, responded negatively to chilling, yielding a mean rating of 5.5 (1.9 sd, 95% CL = 4.9 to 5.9).
Line NC-76 demonstrated an intermediate chilling response when compared with CH and M29 and the reciprocal progeny populations of Expt. 1 (P < 0.0001 for all comparisons) (Table 2; Fig. 2). Comparative analyses of CH (P1) and NC-76 (P3) to their tCF1 progeny revealed differences between NC-76 and tCF1 (P = 0.0004), but not between CH and tCF1 (P = 0.42) in chilling response. Although the chilling response of M29 did not differ from sMF1 (P = 0.75), the response of sMF1 differed from NC-76 (P < 0.0001). Likewise, chilling response comparisons between the two plastidic types (s and t) of contrasting F1 progeny (i.e., tCF1 and sMF1) were predictably different (P < 0.0001). Thus, the chilling response of F1 progeny heterozygous for the nuclear chilling resistance factor contributed from NC-76 was similar to their maternal tCH and sM29 parent lines (Table 2). In contrast, the response of F1 progeny derived from intercrosses between tolerant and susceptible parents differed despite being heterozygous (Ch/ch) for the NC-76 dominant tolerant nuclear chilling allele.
Expt. 3: re-introgression crosses.
Expt. 3 tested whether the chilling response of advanced backcross progeny having unique chilling plastid types [i.e., tCBC5 (CH plastid with theoretically greater than 98% of M29 nuclear background) and sMBC5 (M29 plastid with theoretically greater than 98% of CH nuclear background)] is affected by plastid exchange (re-introgression) (Fig. 1C). This was accomplished by crossing CBC5 and MBC5 (paternal parents) to M29 (s) and CH (t), respectively, to produce re-introgression F1 progeny. Because neither CH nor M29 carries the dominant chilling tolerant nuclear gene (Ch) resident in NC-76, the progeny comparisons here did not consider the effect of Ch. The mean chilling response ratings for these parental lines and backcross progeny were reported previously [Expt. 3 (Table 3)].
Mean, sd, and 95% confidence limits of chilling injury (4 °C for 5.5 h at 270 μmol·m−2·s−1 PPF) in parental cucumber lines (‘Chipper’ and NCSU M29) and re-introgression lines (Expt. 3).z
The mean chilling response rating of re-introgression F1 progeny possessing CH-tolerant cytoplasm (CH × MBC5) was 2.8 (1.0 sd, 95% CL = 2.2 to 3.3) (Table 3). In contrast, the mean rating of its counterpart re-introgression F1 progeny containing M29-susceptible cytoplasm was 4.9 (1.6 sd, 95% CL = 4.3 to 5.5). The chilling response of parental lines used to derive each of the re-introgression progeny differed (P < 0.0001) from each other (e.g., P1 versus P5 and P2 versus P4). In contrast, the chilling response of re-introgression CHF1 and M29F1 progeny did not differ from their maternal sources [i.e., CH (P1) and M29 (P2), respectively]. Thus, re-introduction (or re-introgression) of the original plastid type (i.e., the s and t types of M29 and CH, respectively) as examined in re-introgression F1 progeny restores the chilling response to that equal to parental CH and M29 cultigens.
Discussion
Abiotic stresses such as chilling temperatures often occur in unpredictable patterns, making it in many cases nearly impossible to apply substantial preventative management practices for crop protection. Thus, the use of germplasm that is resistant or tolerant to abiotic stress has become an attractive managerial tool for minimizing profit loss. However, breeding for such stress resistance is often complex, long-term, and labor-intensive. In cucumber, the nuclear and maternal factors that condition chilling response are manifested under differing environmental conditions (Chung et al., 2003; Kozik and Wehner, 2008). As previously reported, chilling response of cucumber seedlings was assessed in controlled environments under different conditions, and, thus, inter-experimental comparisons are difficult. The influences of maternal and nuclear factors were examined here under uniform experimental conditions of Chung et al. (2003) to provide the first report of their respective roles in chilling response at the seedling stage.
Injury to cucumber after a chilling challenge can vary depending on chilling environment and as such is an important consideration for the interpretation of results between experiments (Staub and Wehner, 1996). The chilling response of CH and M29 was originally evaluated by Smeets and Wehner (1997), who stated that genetic variation for chilling response in cucumber could be quantified under specific test conditions (i.e., 270 μmol·m−2·s−1 PPF at 4 °C for 7 h). Chung et al. (2003) subsequently used a chilling duration of 5.5 h at 4 °C to identify the maternal factors associated with CI using the same rating system as that of Smeets and Wehner (1997). Smeets and Wehner (1997) found CH to be chilling-tolerant (average damage rating = 5.7) and M29 susceptible (average damage rating = 7.6). The average chilling damage rating of CH differed between Smeets and Wehner (1997) and Chung et al. (2003) (mean = 2.0)]. Additionally, line M29 was evaluated by Smeets and Wehner [1997 (mean = 7.6)], but not by Chung et al. (2003). However, Smeets and Wehner (1997) found the relative CI of line Gy14 (mean = 7.7) to be comparable to M29. The mean chilling response of line Gy14 as estimated by Chung et al. (2003) was 6.1. Although common cultigens in each of these studies were found to perform similarly in response to chilling stress, mean damage ratings were, on average, slightly less in Chung et al. (2003). Likewise, because the conditions in our study are those set forth by Chung et al. (2003), the mean damage ratings of CH (2.5) and M29 (5.4) as examined here more closely correspond with the results of that study than those reported in Smeets and Wehner (1997).
Multiple genetic factors that control response to abiotic stress can add to the complexity of the interpretation of challenges such as low temperatures. Recently, Kozik and Wehner (2008) evaluated chilling in cucumber by altering experimental conditions from that of Chung et al. (2003) and Smeets and Wehner (1997) and characterized a nuclear component (Ch) controlling chilling response resident in NC-76. Although maintaining chilling temperature and duration as defined by Smeets and Wehner (1997), Kozik and Wehner (2008) increased the light intensity from 270 to 500 μmol·m−2·s−1 PPF and assessed damage 14 d post-treatment instead of 5 d as was the case in both earlier studies. Under this modified chilling treatment, parental lines and cross-progeny were classified as resistant (average rating = 0 to 2), moderately resistant (average rating = 3 to 6), and susceptible (average rating = 7 to 9) such that CH, which was previously classified as resistant (5.7) by Chung et al. (2003), was designated as susceptible (numerical damage ratings not reported). Likewise, although line M29 (not tested by Chung et al., 2003) would be classified as susceptible by Kozik and Wehner (2008) and Smeets and Wehner (1997) if damage ratings were averaged over those experiments (mean = 7.6), NC-76, homozygous (ChCh), is classified highly resistant to chilling stress. Although no distinct numerical damage rating was applied to NC-76, data taken collectively (i.e., same visual rating scale) from the studies of Chung et al. (2003), Kozik and Wehner (2008), and Smeets and Wehner (1997) suggest that the numerical chilling damage rating attribution of highly resistant NC-76 should range between 0 to 1 and 0 to 2, respectively.
Chilling damage rating differences among inbred lines (i.e., CH and M29) across experiments (Chung et al., 2003; Kozik and Wehner, 2008; Smeets and Wehner, 1997) indicate that chilling damage is environmentally dependent. These historic controlled environment outcomes suggest that the assessment of plant reaction to chilling temperature is dependent on testing location (i.e., field versus controlled environment), methodology (i.e., duration and intensity of stress), and quantification (i.e., visual assessment of stress damage). Furthermore, it is likely that any differences observed in chilling response between these historic experiments and those reported here are the result of dissimilarities in test methodology rather than chilling damage quantification, because all experiments used the visual damage rating standards set forth in Smeets and Wehner (1997).
Assessing chilling response differences is critical for deployment of novel germplasm, especially if genetic control originates from two sources (nuclear and cytoplasmic) such as cucumber. For instance, Kozik and Wehner (2008) found CH and Gy14 seedlings to be equally susceptible to chilling (4 °C/7 h/500 μmol·m−2·s−1 PPF) as the susceptible M29 and identified a nuclear component to chilling challenge whose effect was deemed to be greater than the plastid influence. However, in the studies reported here, chilling response data from exact reciprocal populations designed to assess the role of both plastid and nuclear factors indicated that such populations did not differ from that of their plastid parental contributors (Tables 1 and 2), supporting the results of Chung et al. (2003). Moreover, these response types (susceptible and tolerant) persisted and remained constant in backcross generations in which increasing dosages of paternal nuclear alleles were introduced into progeny. The chilling damage recorded in reciprocal cross-progeny examined here supports the observations of susceptible nuclear genotypes types made by Kozik and Wehner (2008) because the chilling response detected between backcross generations possessing varying dosages of CH and M29 nuclear DNA did not differ.
Multiple biochemical pathways are associated with the repair of CI, which affect the genetic design of improved ideotypes (Kudoh and Sonoike, 2002). The experimental differences (i.e., chilling duration and light intensity) between Kozik and Wehner (2008) and Chung et al. (2003)/Smeets and Wehner (1997) could elicit genotype-dependent physiological responses (Scheibe et al., 2005). For instance, plastidic photosystems are damaged by both intense light and/or low temperatures (i.e., chilling and freezing) (Jansen et al., 1999; Kudoh and Sonoike, 2002; Nishiyama et al., 2001), in which photon fluxes as low as 5 μmol·m−2·s−1 PPF can cause an increase in total loss of function to photosystem II (PSII), especially at levels below light saturation (Jansen et al., 1999). It is likely that the light levels used by Kozik and Wehner (2008), which were almost twice that used by Chung et al. (2003) and that reported here, resulted in substantial damage to PSII. Such damage likely affected the efficiency of PSII repair, which is directly dependent on the surrounding environmental conditions in which both high light and low temperature decrease efficiency of these repair operations (Govindachary et al., 2004, 2007; Kanervo et al., 2007).
Other photosystem-related responses have been detected in cucumber in response to chilling temperatures. Kudoh and Sonoike (2002) investigated photosystem I (PSI) damage in cucumber under conditions (4 °C for 5 h, light levels at 190 μmol·m−2·s−1 PPF) similar to Chung et al. (2003) and that imposed here. They found that visible symptoms (e.g., tissue chlorosis) attributable to chilling were the consequence of photoinhibition of PSI. Specifically, PPF densities during chilling cause a chlorophyllic decrease, not PPF applied in normal temperatures (Kudoh and Sonoike, 2002).
Plastids are the first and most severely affected organelle in the plant cell when plants are challenged with chilling stress (Fowler and Limin, 2001). Therefore, plastidic responses might be more rapidly influenced by small changes in chilling temperatures than those conditioned by nuclear genes. This might be a plausible explanation for the germplasm chilling differences observed by Chung et al. (2003) and Kozik and Wehner (2008) and those reported here. For instance, the experimental chilling conditions imposed by Kozik and Wehner (2008) might cause a plastidic response that is beyond its maximum biochemical and/or mechanistic capacity to condition a tolerance response [threshold of internal damage (Scheibe et al., 2005)]. Moreover, if both the plastid and nucleus contribute to chilling response, then as one becomes inoperable, the other could provide a physiological mechanism for a tolerant response. Likewise, genotypic differences in plastid or nuclear constitution (e.g., one organelle tolerant and the other susceptible) could elicit similar responses under a given set of environmental conditions (i.e., t/chch and s/Ch_).
The direct influence of the Ch gene on seedlings during chilling was examined here by evaluating F1 progeny derived from crossing the nuclear Ch donor, NC-76, with CBC5 and MBC5, two lines whose chilling response is determined by their respective plastome genotypes. These F1s were tested at the less severe conditions of Chung et al. (2003) than those of Kozik and Wehner (2008) to evaluate any potential augmentation of Ch on plastid-conferred chilling response. Although Kozik and Wehner (2008) defined NC-76 as possessing a dominant factor for chilling tolerance, it is only partially tolerant to chilling conditions in the heterozygous state (Ch/ch). Line NC-76 (s/ChCh) was strategically crossed with CBC5 (t/chch) and MBC5 (s/chch) to create F1 progeny possessing contrasting plastid types but similar (i.e., Ch/ch) nuclear genotypes. This allowed for the critical evaluation of chilling responses (i.e., tolerant versus susceptible) attributable solely to plastidic or nuclear sources. Placement of the tolerant CH plastid into a the susceptible nuclear background of M29 resulted in backcross progeny whose only tolerant contribution is found in the plastome because the genetic contributions of the mitochondria are inherited paternally in cucumber (Havey, 1997).
Based on findings of Kozik and Wehner (2008) and the data presented here (Tables 1 to 3; Figs. 1 and 2), it can be deduced that NC-76 possesses a chilling-tolerant nuclear factor and a susceptible plastid type (s/ChCh). That is, all parental lines, hybrids (F1), and backcross progeny possessing the tolerant plastid responded similarly to chilling challenge regardless of their nuclear constitution. In fact, under the chilling conditions described here, the average chilling response of NC-76 was intermediate between that of parental lines CH and M29. It is, therefore, hypothesized that the homozygous state of the nuclear tolerance gene (Ch) resident in NC-76 augments a susceptible plastid type (s) to provide a slightly more tolerant response than that observed in M29 (s/chch) under the chilling conditions used here. With similar logic, both CBC5 × NC-76 and MBC5 × NC-76 F1 progeny would predictably be tolerant [(t/Chch) and/or (s/Chch)] to chilling temperatures if the dominant Ch allele contributed more significantly than the plastid. However, F1 progeny (heterozygous Chch) differed in response and were influenced by plastid type and not nuclear content. Thus, it must be concluded that, under the environmental conditions set forth in this study, the maternal plastid contribution provided is more predictive than the nuclear contribution in determining the chilling phenotype of the cross-progeny examined here (Table 2; Fig. 2).
Identification of plastomic control of chilling response requires experimental conditions that would allow the plastid to remain operable during and after (recovery) the chilling episode. Given previous reports (Kudoh and Sonoike, 2002) and the results of this study, it is hypothesized that differences in genotypic response to chilling temperatures (i.e., between tolerant cucumber CH and susceptible cucumber genotypes (e.g., Gy14, M29) (Chung et al., 2003; Kozik and Wehner, 2008; Smeets and Wehner, 1997; this study) is the result of the presence/absence of a tolerant plastidic genotype (t versus s) or a plastidic incapacitation related to severe chilling conditions (i.e., intense light and extended low temperature exposure). This hypothesis could be tested by evaluating the chilling response of an advanced backcross progeny line (NC-76 recurrent parent) originating from an original CH × NC-76 mating. Such cross-progeny would contain both plastidic (CH, tolerant plastid) and homozygous dominant nuclear (NC-76, nuclear tolerant Ch) chilling-tolerant components. Individuals with this genotype (t/ChCh) would presumably be chilling-tolerant under the experimental conditions described by Kozik and Wehner (2008) as well as those used by Chung et al. (2003) and Smeets and Wehner (1997).
The effect of Ch on chilling response was negligible in cross-progeny under the chilling conditions applied here. In contrast, plastid type was directly associated with chilling response in cross-progeny (Tables 1 to 3). For instance, the CI of CH and M29 differed from individuals with nearly identical nuclear constitutions (MBC5 and CBC5, respectively) but with contrasting plastid type. Moreover, the re-introduction of the original plastid type by crossing M29 and CBC5 and CH with MBC5 restored the chilling response type observed in parental types (i.e., CH and M29). Thus, it may be concluded that a simple, yet effective, form of breeding CI tolerance into elite cultivars (e.g., backcrossing) may be effective for the introduction of plastomes conferring a CI-tolerant phenotype. Because cucumber plastomes are inherited maternally, introgression of such factors would be a relatively rapid process during plant improvement.
Literature Cited
Choi, D., Rodriguez, E.M. & Close, T.J. 2002 Barley Cbf3 gene identification, expression pattern, and map location Plant Physiol. 129 1781 1787
Chung, S., Gordon, V.S. & Staub, J.E. 2007 Sequencing cucumber (Cucumis sativus L.) chloroplast genomes identifies difference between chilling-tolerant and -susceptible cucumber lines Genome 50 215 225
Chung, S., Staub, J.E. & Fazio, G. 2003 Inheritance of chilling injury: A maternally inherited trait in cucumber J. Amer. Soc. Hort. Sci. 128 526 530
Foolad, M.R., Hyman, J.R. & Lin, G.Y. 1999 Relationships between cold- and salt-tolerance during seed germination in tomato: Analysis of response and correlated response to selection Plant Breed. 118 49 52
Fowler, D.B. & Limin, A.E. 2001 Impact of cold stress. 7 May 2009. <http://www.plantstress.com/Articles/cold_i/cold_i.htm>
Goldschmidt-Clermont, M. 1998 Coordination of nuclear and chloroplast gene expression in plant cells Intl. Rev. Cytol. 177 115 180
Govindachary, S., Bigras, C., Harnois, J., Joly, D. & Carpentier, R. 2007 Changes in the mode of electron flow to photosystem I following chilling-induced photoinhibition in a C3 plant, Cucumis sativus L Photosynth. Res. 94 333 345
Govindachary, S., Bukhov, N.G., Joly, D. & Carpentier, R. 2004 Photosystem II inhibition by moderate light under low temperature in intact leaves of chilling-sensitive and -tolerant plants Physiol. Plant. 121 322 333
Gray, K.M. & Greenberg, E.P. 1992 Physical and functional maps of the luminescence gene cluster in an autoinducer-deficient Vibrio fischeri strain isolated from a squid light organ J. Bacteriol. 174 4384 4390
Havey, M.J. 1997 Predominant paternal transmission of the mitochondrial genome in cucumber J. Hered. 88 232 235
Iturriaga, G., Cushman, M.A.F. & Cushman, J.C. 2006 An EST catalogue from the resurrection plant Selaginella lepidophylla reveals abiotic stress-adaptive genes Plant Sci. 170 1173 1184
Jansen, M.A.K., Mattoo, A.K. & Edelman, M. 1999 D1-D2 protein degradation in the chloroplast Eur. J. Biochem. 260 527 532
Kanervo, E., Suorsa, M. & Aro, E. 2007 Assembly of protein complexes in plastids 283 313 Bock R. Topics in current genetics. Springer-Verlag Heidelberg, Germany
Kant, P., Gordon, M., Kant, S., Zolla, G., Davydov, O. & Heimer, Y.M. 2008 Functional-genomics-based identification of genes that regulate Arabidopsis responses to multiple abiotic stresses Plant Cell Environ. 31 697 714
Komori, T. & Imaseki, H. 2005 Transgenic rice hybrids that carry the rf-1 gene at multiple loci show improved fertility at low temperature Plant Cell Environ. 28 425 431
Kozik, E.U. & Wehner, T.C. 2008 A single dominant gene Ch for chilling resistance in cucumber seedlings J. Amer. Soc. Hort. Sci. 133 225 227
Kreps, J.A., Wu, Y., Chang, H.S., Zhu, T., Wang, X. & Harper, J.F. 2002 Transcriptome changes for Arabidopsis in response to salt, osmotic, and cold stress Plant Physiol. 130 2129 2141
Kudoh, H. & Sonoike, K. 2002 Irreversible damage to photosystem I by chilling in the light: Cause of the degradation of chlorophyll after returning to normal growth temperature Planta 215 541 548
Lynch, D.V. 1990 Chilling injury in plants: The relevance of membrane lipids 17 34 Katterman F. Environmental injury to plants Academic Press New York, NY
Mayfield, S.P. 1990 Chloroplast gene regulation: Interaction of the nuclear and chloroplast genomes in the expression of photosynthetic proteins Curr. Opin. Cell Biol. 2 509 513
Nishiyama, Y., Yamamoto, H., Allakhverdiev, S., Inaba, M., Yokota, A. & Murata, N. 2001 Oxidative stress inhibits the repair of photodamage to the photosynthetic machinery EMBO J. 20 5587 5594
Rodermel, S. 2001 Pathways of plastid-to-nucleus signaling Trends Plant Sci. 6 471 478
Sabehat, A., Weiss, D. & Lurie, S. 1998 Heat-shock proteins and cross-tolerance in plants Physiol. Plant. 103 437 441
Scheibe, R., Backhausen, J.E., Emmerlich, V. & Holtgrefe, S. 2005 Strategies to maintain redox homeostasis during photosynthesis under changing conditions J. Expt. Bot. 56 1481 1489
Smeets, L. & Wehner, T. 1997 Environmental effects on genetic variation of chilling resistance in cucumber Euphytica 97 217 225
Staub, J.E. & Bacher, J. 1997 Cucumber as a processed vegetable In: Processing vegetables: Science and technology IV Technomic Lancaster, PA
Staub, J.E. & Wehner, T.C. 1996 Noninfectious disorders: Temperature stress 66 87 Zitter T.A., Hopkins D.L. & Thomas C.E. Compendium of cucurbit diseases Part II APS Press St. Paul, MN
Susek, R.E. & Chory, J. 1992 A tale of two genomes: Role of a chloroplast signal in coordinating nuclear and plastid genome expression Funct. Plant Biol. 19 387 399
Takahashi, R., Benitez, E.R., Funatsuki, H. & Ohnishi, S. 2005 Soybean maturity and pubescence color genes improve chilling tolerance Crop Sci. 45 1387 1393
Wehner, T.C. 1982 Genetic variation for low-temperature germination ability in cucumber Cucurbit Genet. Coop. Rpt. 5 16 17