The heat-susceptible cucumber inbred line “02-8” produced by our laboratory was used for all experiments in this study. DNA sequences of the CaM gene were found in the NCBI database and Cucumber Genome Database¹. A fragment of the CsCaM3 gene was amplified by PCR with the P1 forward primer (5′-CTCTAGAATGGCGGATCAGCTCACCGA-3′) and P2 reverse primer (5′-CGGATCCCCTTAGCCATCATGACCTTAAC-3′); underlined sequences show the XbaI and BamHI restriction sites.

Evolutionary analyses were conducted in MEGA6 (Tamura et al., 2013). The evolutionary history was inferred using the neighbor-joining method (Saitou and Nei, 1987). The optimal tree with a summed branch length of 0.02022066 is shown (next to the branches). The evolutionary distances were computed using the Poisson correction method (Zuckerkandl and Pauling, 1965) and are within the units of the number of amino acid substitutions per site. The analysis involved eight amino acid sequences. All positions containing gaps and missing data were eliminated. There were a total of 149 positions in the final dataset.

The expression vector pBI-CsCaM3 (Supplementary Figure S1) with the CaMV35S promoter and nptII gene was constructed and transferred to the Agrobacterium tumefaciens strain EHA105 by the freeze–thaw method. The transformation of cucumber was conducted as described by Cao et al. (2011). The first filial generation of the transgenic plants (OE-1, OE-2, and OE-3) were used in all experiments conducted at the three-leaf stage. The identification of transgenic cucumber plants was by Southern and northern blots. The hybridization was conducted using a DNA labeling and detection kit (Boehringer Mannheim, Mannheim, Germany), and the dosage of DNA and total RNA were 10 μg and 12 μg, respectively. A fragment of the 35S promoter was PCR amplified using the 35S-F forward primer (5′-ATAGTGGGATTGTGCGTCA-3′) and 35S-R reverse primer (5′-GCACCTACAAATGCCATCA-3′) and was then used as a probe for Southern blotting. The resulting fragment of CsCaM3 was used as the probe for northern blotting. The heat treatment conditions for all experiments were under 43 and 25°C during the day and night, respectively.

The expression levels of CsCaM3 were analyzed at the three-leaf stage, with three biological and technical replicates. From the tissues of first top leaf, stem, and root of cucumber inbred line “02-8” seedlings were extracted total RNA with TRIzol reagents (Invitrogen, Carlsbad, CA, United States) after 43°C treatment for 12 h, and RNA was extracted from the first top leaf of seedlings after 0, 6, 12, and 24 h of temperature and hormone treatments. Solutions of 5 μM abscisic acid (ABA) and 100 μM salicylic acid (SA) were sprayed onto the leaves of seedlings under a 43°C temperature treatment. RNA was converted into cDNA with a kit (Genstar, Beijing, China). All qRT-PCR analyses were conducted by gene-specific primers (Supplementary Table S1) and a SYBR Premix Ex Taq kit (TaKaRa, Dalian, China). The RT-PCR protocol consisted of one cycle at 94°C for 3 min; 28 cycles of 94°C for 1 min, 55°C for 45 s, and 72°C for 90 s; and one cycle of 72°C for 7 min. The 2^-ΔCt method (Livak and Schmittgen, 2001) was used to assess the relative gene expression.

Amplification of CsCaM3 was conducted by PCR with the sub-Cam3F forward primer (5′-ATGGCGGATCAGCTCACCGA-3′) and the sub-Cam3R reverse primer (5′-CTTGGCCATCATGACCTTCA-3′), which does not contain a stop codon. In order to control CsCaM3-GFP vectors with a CaMV 35S promoter, the PCR products were fused into the pEZS-NL-GFP vector in-frame with the sequence of green fluorescent protein (GFP). The vector was transferred into A. tumefaciens strain EHA105, and the leaf epidermis of Nicotiana benthamiana leaves were injected with the transformed EHA105 (Na et al., 2016). GFP fluorescence of the transformed plants was observed after two days in darkness using a laser scanning confocal microscope (Leica TCS SP2, Leica Microsystems, Wetzlar, Germany).

The CsCaM3 gene was selected for localizing mRNA in the leaf and root tissues of cucumbers grown under HT stress (i.e., 43°C). Digoxigenin-labeled RNA probes were hybridized to the cucumber tissues of young leaf and root as described by Ma et al. (2012). The gene fragment was labeled using the DIG RNA Labeling Kit (SP6/T7; Roche, Shanghai, China). The fragments of probes were amplified from cDNA using gene-specific primers including SP6 and T7 RNA polymerase-binding sites (CsCaM3-SP6, 5′-ATTTAGGTGACACTATAGAAGATGGCGGATCAGCTCACCG A-3′; CsCaM3-T7, 5′-TAATACGACTCACTATAGGGAGACC TTAGCCATCATGA CCTTAAC-3′).

The measurement of chlorophyll fluorescence under minimal fluorescence (F_o) was measured without actinic light. Then, the leaf was exposed to a saturation pulse to measure maximal fluorescence (F_m). The optimal quantum yield of photosystem II (PSII) (F_v/F_m, F_v = F_m -F_o) was derived from measured F_o and F_m values, where F_v was the difference between F_o and F_m. Actinic light was turned on after saturation pulse, and the fluorescent signals reached the steady state slowly with intermittent saturation pulses. F_s determined the steady-state fluorescence yield, and F_m′ was the resulting light-adapted maximum fluorescence. In this process, photochemical quenching of open PSII was measured, revealing photo-protective non-photochemical quenching (NPQ) and other heat dissipation mechanisms. qP is the fraction of open PSII reaction centers, characterized as the coefficients of photochemical fluorescence quenching (Genty et al., 1989; van Kooten and Snel, 1990). Electron transport rate (ETR) and effective quantum yield of PSII (Φ_PSII) are valuable measurements in various plant stress studies. The F_v/F_m, Φ_PSII, NPQ, and qP values were determined from the same individual using the following equations: Φ_PSII = (F_m′ -F_s)/F_m′, NPQ = (F_m/F_m′) - 1, qP = (F_m′ -F_s)/(F_m′ -F_o), and ETR = Φ_PSII × PFD (photon flux density) × 0.5. Images were obtained using a Chlorophyll fluorescence imager (IMAGING-PAM, WALZ, Germany). Total chlorophyll, chlorophyll a, and chlorophyll b contents were assessed using a spectrophotometric method (Porra et al., 1989).

The malondialdehyde (MDA) content and peroxidase (POD) activities were tested according to the method described by Velikova et al. (2000). Superoxide dismutase (SOD) activity was measured using nitro blue tetrazolium (NBT) methods (Attar et al., 2006). Catalase (CAT) activities were measured using the method described by Dong et al. (2014).

Accumulations of superoxide and hydrogen peroxide (H₂O₂) were visually analyzed with NBT and 3,3-diaminobenzidine (DAB) staining using the methods described by Driever et al. (2009). Measurement of reactive oxygen species (ROS) was conducted using 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) staining. Leaf epidermal strips were peeled from cucumber leaves without chemical treatment. The leaf epidermal strips were floated on a solution of 50 μM DCFH-DA (Sigma Chemicals, St. Louis, MO, United States) for 30 min before treatment. ROS accumulation was observed by fluorescence microscopy using a Leica MZ FL III (excitation, 450 ± 490 nm; barrier 520 ± 560 nm).

The viability of root tissue was assessed using the fluorescein diacetate/propidium iodide (FDA/PI) staining mixture method as described by Pan et al. (2001). The tips of roots (about 2.5 cm in length) were gently incubated in a FDA/PI mixture (containing FDA 12.5 g mL^-1 and PI 5.0 g mL^-1) for 10 min and then washed with distilled water. Images were observed with a fluorescent microscope (Olympus BX41; Olympus Corporation, Tokyo, Japan) under blue light excitation (460–495 nm) and emission at 510 nm and analyzed using a photographic apparatus (Olympus CamediaC-5060; Olympus, Tokyo, Japan) and analySIS software (Soft Imaging Solutions GmbH, Münster, Germany).

Stomata were measured as described previously (Zhu et al., 2016). ABA content was measured as previously described (Kara et al., 1997). Enzyme-linked immunosorbent assay kits and 96-well (12 × 8) plates used in this study were purchased from the Crop Chemical Control Center of China Agricultural University. All assays and measurements were conducted in triplicate.

The NCBI database and Cucumber Genome Database (see text footnote 1) were searched to find DNA sequences of CaM genes. From these sequences, primers were designed to PCR amplify the CaM gene CsCaM3 from cucumber total cDNA, resulting in the isolation of CsCaM3, a 450-bp ORF encoding 149 amino acids. The amino acid sequence of CsCaM3 showed high similarity to CsCaM (XM_011655459.1) from cucumber, CmCaM (XM_008466393) from melon, AtCaM3 AY091301) from Arabidopsis, JrCaM7 (XM_018967153.1) from Juglans regia, StCaM (JX576246.1) from Solanum tuberosum, NtCaM7 (XM_016640697.1) from Nicotiana tabacum, BoCaM (XM_013776345) from Brassica oleracea, and SlCaM3 (NM_001321494) from Solanum lycopersicum (Figure 1).

The expression pattern of CsCaM3 was assessed in root, stem, and leaf tissues from cucumber inbred line “02-8.” The expression level of the gene was highest in leaves and lowest in roots (Figure 2A). ABA and HT stress treatment increased the expression of CsCaM3, but SA did not (Figures 2B–D). These results concurred with the results of situ hybridization of tissues was observed in leaf and root tissues, and the hybridization signals subjected to 43°C high-temperature stress treatment were stronger than that in the control group (Figures 2E–H). Subcellular location results indicated that CsCaM3 proteins are located in the cell membrane and nuclei (Figure 3). Taken together, CsCaM3 is likely involved in the regulation of HT stress tolerance in cucumber.

In order to identify the function of CsCaM3, the CsCaM3-overexpression vector (i.e., pBI-CsCaM3) was constructed, from about 1000 explants used for transformation, four transgenic plants were obtained (Supplementary Figure S2), which were identified by genomic Southern and northern blotting (Figures 4A,B). The hybridization signal of CsCaM3 in transgenic plants was determined to be stronger relative to the wild-type (WT) plants (Figure 4B). Three T1 transgenic plants (OE-1, OE-2, and OE-3) were obtained by selfing and used to assess heat-tolerance. The conditions of heat treatment were 43 and 25°C under day and night, respectively. More than 30 seedlings of each strains were used. After heat treatment for 5 days, more than 30% of WT plants were dead, but only 5% of the T1 transgenic plants were dead. After 10 days of HT treatment, all WT plants were dead, while about 60% of the T1 transgenic plants still survived (Figures 4C,E,F). The growth potential of transgenic plant radicles was stronger than that of WT plants (Figure 4D). Overexpression of CsCaM3 may thus have improved the heat tolerance of cucumber plants.

In order to identify the sensitivity of transgenic cucumber plants to HT stress, the viability of root tips was assessed. The root tip viabilities of both WT and transgenic cucumber plants were decreased under the 43°C HT treatment, but there were more dead cells in WT plants relative to the transgenic plants (Figure 5A). Moreover, the heat resistance of transgenic plants was stronger than that of WT plants, and HTs affected the root viability of the plants.

The transgenic and WT plants were treated at 43°C for three days, after which their chlorophyll fluorescence parameters were measured. The chlorophyll fluorescence parameters of both were significantly affected, but all parameters of the WT plants were lower than those of the transgenic plants (Table 1). The damage to the photosynthesis systems of the WT plants was more serious than to that to the transgenic plants under HT stress (Figure 5B), suggesting overexpression of CsCaM3 could increase heat-resistance of cucumber plants.

The contents of overall ROS, H₂O₂ and superoxide in the leaves of transgenic plants were lower than those of the WT plants, but under normal conditions, the ROS content of both the transgenic and WT did not significantly differ (Figure 6). Compared to the control plants, the chlorophyll contents of both the WT and transgenic plants were not remarkably different. However, the activities of SOD, POD, and CAT of the transgenic plants were significantly higher than those of the WT plants, and the MDA contents of WT plants were significantly higher than those of the transgenic plants (Table 2). These results again demonstrate that overexpression of CsCaM3 in cucumber could increase the activities of antioxidant enzymes and decrease MDA content and oxidative stress damage.

The CLH, CBR1, PAO, and RCCR genes are related to the chlorophyll catabolic pathway in plants, and their expression reflects the process of photosynthesis. Therefore, the expression levels of these four genes were measured. The transcript levels of the four genes substantially differed between the WT and transgenic plants. The expressions of CBR1, PAO, and RCCR in transgenic plants were higher than those of WT plants, but the expression of CLH in transgenic plants was lower than that of the control plants under the 43°C temperature treatment for 3 days. At the same time, the expression levels of seven genes (CsSOD, CsPOD, CsCAT, CscAPX, AOX, HSP70, and HSP90) in the transgenic plants were markedly elevated and were higher than those of the WT plants. The expression levels of CsABI1, CsABI2, and CsABI3, which are involved in the ABA catabolic pathway in plants, were significantly increased compared with the WT plants (Figure 7). These results suggest that overexpression of CsCaM could affect the transcription of genes not only in the chlorophyll pathway, but also in the ABA catabolic pathway.

After 3 days of the 43°C temperature treatment, the ABA content of the transgenic and WT plants was assayed. The ABA content of all transgenic plants was significantly higher than that of the WT plants (Table 3). On the other hand, the opening degree of stomata of WT plants leaves was greater than that of the transgenic plants under the temperature stress treatment (Figure 8).