‘Momordica charantia’ introducing a new rootstock for grafted cucumber under lowtemperature stress

Cucumber is a sensitive vegetable to low temperatures. Grafting veg­ etables on different rootstocks can decrease the harmful effects of environ­ mental stresses, including low­temperature stress. An experiment was per­ formed to evaluate grafting cucumbers on different rootstocks at low tempera­ tures. Cucumber growth and yield and photosynthesis traits were examined. Treatments were the optimum temperature (25±2°C), and cold temperature (15±3°C, Ts), and rootstocks, were Momordica charantia (Rmo), Cucurbita max‐ ima (Rma), non­grafted (Rn) and self­ grafted (Rs) with 4 replications. Shoot fresh and dry weight, chlorophyll, RWC, transpiration, decreased with tempera­ ture stress. The number of female flowers, electrolyte leakage, photosynthesis, stomatal conductance increased with Ts. First fruit emergence per plant, N, P, K, Mg concentration decreased with Ts stress. Transpiration, female flower, RWC, and stomata conductance, N, P, K, Ca, and phenol increased in Rma and Rmo. Mg was at the highest concentration in Rma and Na in Rn. All in all, using Rmo as well as Rma is recommended for rootstock as it causes more reproduc­ tive growth.


Introduction
Recently, the greenhouse culture of plants has widely expanded (Yan et al., 2013). Cucumbers (Cucumis sativus) are commonly grown in the greenhouse. The optimal temperature for aerial growth and roots are about 2823/1815°C (day/night) and 20°C, respectively (Ikeda and Kawashiro, 2005). Thus, cucumbers are coldsensitive vegetables. Low temperature stress can occur in plastic culture or nonequipped green houses, which are very common in Iran. It may also affect field culture cucumbers and spring culture cucumbers. Low temperatures have differ ent deleterious effects on cucumbers, including affecting the nutrient uptake of plants (Pregitzer and King, 2005), and inducing greater H 2 O 2 , MDA, and sol uble sugar content in cucumbers, causing damage to plants and inhibiting plant growth (Qiuyan et al., 2013).
There are several ways to control lowtempera ture damage to plants, including controlling root temperatures by warming nutrient solution or soil. Still, these ways require excess energy (Willits and Peet, 1998). Finally, excess chemical nutrient applica tions for promoting growth can prevent damage to plants, which unfortunately involves environmental risks. Therefore, vegetable grafting can be a proper way to increase plant resistance to environmental stresses in addition to expanding its yield, quality, and growth. Furthermore, if any energyefficient way is introduced, which can raise the growth rate of plants in greenhouse conditions at lower tempera tures, vast amounts of energy and money will be saved. Vegetable grafting has been used to increase plant tolerance under environmental stresses, includ ing high temperatures (Rivero et al., 2003), low tem peratures (Rivero et al., 2003), salinity stresses (Estan et al., 2005), drought stresses (Bhatt et al., 2002) and heavy metal stresses (Edelstein et al., 2005).
Fig leaf gourd and bur cucumbers (Sicyos angulatus) are used for increasing resistance against low temperature stresses for cucumber rootstocks (Venema et al., 2008). These rootstocks increase plant resistance using different ways. First, lowtem perature stresses decrease Co 2 assimilation in cucum bers and this reduction is improved by grafting (Zhou et al., 2009). Secondly, a previous study done by Zhou et al. (2007) showed that fig leaf gourd root stocks enhanced vegetative growth and yield of cucumbers under lowtemperature stresses.
There is a hypothesis stating that grafted plant responses to temperature stresses are related to scion species (Venema et al., 2005) (Rouphael et al., 2008;Kumar et al., 2015), low soil temperatures (Tachibana, 1982), salinity (Huang et al., 2010), improve NaCl and CaCl 2 tolerance in Cucumber (Colla et al., 2013), Al toxicity in cucumber (Rouphael et al., 2016). Different rootstocks were used in normal and stress conditions to improve yield and growth traits like dif ferent kinds of squash (Yang et al., 2006), (Massai et al., 2004), (Rouphael et al., 2008), Fig leaf guard (Tachibana, 1982, pumpkins and bottle gourd and P360 (Cucurbita maxima Duch.×Cucurbita moschata Duch.) (Colla et al., 2010 and. To the best of our knowledge, few studies have investigated the use of Momordica charantia as a rootstock for cucurbits, or more explicitly grafting, to examine the possibility of cucumber cultivation at low temperatures in the whole growth period. Hence, the goal of the present study was the use of Momordica charantia and Cucurbita maxima as rootstocks and Cucumis sativus var. DavosII, which is a common variety of cucumbers in Iran as a scion under lowtemperature stress.

Experimental design and plant preparation
This experiment was conducted in a plastic green house in the Department of Horticulture Science at the Isfahan University of Technology, Isfahan, Iran. An investigation was arranged as a combined analysis involving data from two locations, simultaneously collected, based on CRD with 4 replications. Treatments were optimum (25±2°C, Tc) and low tem peratures (15±3°C, Ts). Cucumbers (Cucumis sativus var. DavosII) were grafted to Momordica charantia (Rmo), Cucurbita maxima (Rma). Nongrafted (Rn) and selfgrafted (Rs) consider as a control. Scion seeds had been cultivated 10 days before rootstock seeds in cocopeat: perlite 1:1. Scion plants and root stocks were cut beneath and above the first true leaves, respectively. Holeinsertion grafting was used and grafted plants were transferred to a recovery greenhouse with high relative humidity. Plants were kept for 2 weeks in recovery conditions and gradually adapted to normal greenhouse conditions. Grafted plants were transferred to 5kg pots, including soil. Irrigation was used when the plant needed it. Chemical fertilizer NPK (20, 20, 20) 2 mg/pot was applied every 10 days. Plants were conducted to wire above the greenhouse, and there was no use of any pesticide.

Plant growth and fruit properties
The male and female flower, node numbers, and shoot numbers were counted during the experiment; root and internode length was determined with a ruler. The time and node of male and female flower emergence were recorded. Fruit diameter, fruit firm ness, and TSS were measured with a caliper (Mitutoyo Corp, Japan), Pentameter (model OSKI 10576), and portable Refractometer (PAL 1 Brix, Japan), respectively (Raeisi et al., 2014).
Shoots were excised from the roots using a steel blade and then fresh weights of roots and shoots were measured. All the samples were ovendried at 70˚C for 48 hours and the dry weights were estimat ed. During the experiment and finally 124 days after transplanting, fruits were harvested and washed using tap water and were weighed by an analytical balance.

SPAD value and photosynthesis trait assay
Chlorophyll content was measured using a chloro phyll meter (SPAD502 plus, Minolta, Japan). Fv/Fm was measured by chlorophyll fluorescence (OS 30, USA) after 3 weeks. Photosynthetic properties were determined from the youngest fullyexpanded leaf for 3 replications per treatment by Portable photo synthesis systems for gas exchange and chlorophyll fluorescence measurements (LICOR6400, USA) from 10:00 to 11:00 AM on a clear day (without clouds). The measurements were conducted with photosyn thetically active radiation (PAR) intensity of 1000 μmol m 2 s 1 and CO 2 concentration of 350 μmol·mol. Mesophyll conductance (μmol m 2 s 1 ) was calculated by dividing the photosynthetic rate by the substom atal CO 2 level (Ahmadi and Siosemardeh, 2005).

Antioxidant activity
Antioxidant activity was measured and expressed as gallic acid (equivalents l of gallic acid/g) with UV VIS spectrophotometer (Shimadzu UV160AJapan). Three mg of sample were dissolved in 5 mL methanol stock, and 1.4 ml of this solution was blended with 0.6 mL of DPPH solution. After 30 min, the absorbance of the solution was recorded at 515 nm by the spectrophotometer (UV 160A Shimadzu Corp., Kyoto, Japan) against methanol as a blank. The 0.2 mM of DPPH solution in methanol was used as a stock of DPPH for the determination of the free radi cal scavenging activity of the samples (Koleva et al., 2002).

Phenolic content
Total phenolic content was determined using the FolinCiocalteu method. The absorbance was mea sured at 725 nm with a spectrophotometer (UV 160A Shimadzu Corp., Kyoto, Japan). The results were expressed in gallic acid equivalents (mg/100 g fresh weight) using gallic acid (00.1 mg/mL) standard curve (Singleton et al., 1965).

Proline
Proline accumulation was determined using the method proposed by Bates et al. (1973). After the extraction of toluene, the clear phase was recovered and spectrophotometrically estimated at 520 nm using toluene as a blank. Purified proline was used for standardization (050 mg/mL) and was expressed as mol proline g 1 fresh weight.

Electrolyte leakage
Electrolyte leakage (EL) was measured using an electrical conductivity meter employing the method described by Lutts et al. (1995)

Relative water content
Relative water content (RWC%) was determined using ten 7 mmdiameter leaf discs. The leaf discs of each treatment were weighed (FW). They were then hydrated until saturation was reached (constant weight) for 48 h at 5°C in darkness (TW). The leaf discs were dried in an oven at 105°C for 24 h (DW). Relative water content was calculated according to the following equation (Filella et al., 1998): The determination of total nitrogen in the leaf samples was based on the Kjeldahl method (Estan et al., 2005). The concentrations of K, Ca, Mg, and P were measured (Shield Torch System, Agilent 7500a). Meanwhile, phosphorus was estimated by the vanadomolybdo phosphoric acid colorimetric method at 460 nm (Estan et al., 2005). P was colori metrically determined using a spectrophotometer (UV 160A Shimadzu Corp., Kyoto, Japan).

Statistical analysis
All data were analyzed using twoway ANOVA (Statistix 8 software) (Tallahassee FL, USA) and the means were compared for the significance by the least significant difference (LSD) test at P<0.05.

Analysis of variance of temperature and rootstock on some characteristics of cucumber
Temperature effect shoot fresh and dry weight, chlorophyll index, and the number of the female flower, photosynthetic rate, transpiration, stomatal conductance, RWC, EL, phenol content, and all nutri ent concentration (Table 1 a , b). Rootstock changes all cucumber characteristics except for average fruit weight, fruit diameter, TSS, firmness, mesophyll con ductance, proline, and antioxidant (Table 1 a, b). The interactive effect of temperature and rootstock showed that all measured parameters change signifi cantly (Table 1 a, b).

The main effect of temperature on cucumber
Shoot fresh and dry weight, chlorophyll, RWC, transpiration, decreased with temperature stress. The number of the female flower, EL, photosynthesis, and stomatal conductance increased with Ts (Table 2 a). First fruit emergence per plant, N, P, K, Mg con centration decreased with Ts stress (Table 2 b). Shoot fresh and dry weight and fresh root weight was high Table 2 a The effect of temperature on some characteristics of cucumber Tc= optimum temperature, Ts= low temperature. Within a column means followed by the same letter are not significantly different at P<5% according to LSD test. Table 2 b The effect of temperature on some characteristics of cucumber Tc= optimum temperature; Ts= low temperature. Within a column means followed by the same letter are not significantly different at P<5% according to LSD test. Table 1 a Analysis of variance of the effect of temperature and rootstock on some characteristics of cucumber NS= no significant, * significant at 5% and ** significant at 1%.

The main effect of rootstock on cucumber
The highest El and photosynthesis were observed in Rmo. The RWC and stomata conductance increased in Rma and Rmo, however, transpiration increased in Rs, Rmo, and Rma (Table 3 a). N, P, K, Ca, and phenol increased in Rma and Rmo. Mg was at the highest concentration in Rma and Na in Rn (Table   3 b). Rma has the highest fruit number, fruit yield, although, first fruit emerges in Rmo (Table 4).
The interaction effect of temperature and cucumber grafting on some growth characteristics of cucumber K concentration was the highest in Rmo in both temperatures and P concentration increased in Rma×Ts. The highest N, K and Ca, concentrations were seen in Rma×Ts. P concentration increased in Rma and Rmo at Ts. Mg increased in Rs×Tc and Na in Rs×Ts (Table 5).
Shoot fresh weights decreased in Rn and Rs at low temperatures (Ts), but shoot dry weights did not sig nificantly change in Rma and Rmo at both tempera tures (Fig. 1). Shoot dry weights followed the same trend as the fresh shoot weight was the highest in Rn, Rs, Rma at Tc (Fig. 1). Root fresh weights decreased in Rn, Rs, and Rma at Ts compared with Tc, but did not significantly change in Rmo at both temperatures, although the fresh root weight was the lowest in Rmo at Tc. The highest root fresh weight was seen in Rma×Tc (Fig. 2 a).
SPAD value was the highest in Rs, Rma, and Rmo at Tc (Fig. 2 b). The interaction effect of the tempera  ture on grafted cucumber did not significantly affect the time of the emergence of the first female flowers (data did not show). The highest and the lowest num ber of the male flower were seen in Rs×Tc and Rma, respectively. The female flower was affected by tem perature variations in different rootstocks. The female flower increased in the Rn and Rma at Ts and did not change in Rs and Rmo at Tc and Ts (Fig. 3). Fruit number was the highest in Rma×Tc, although the same results were seen in Rmo×Tc and Rmo and Rma at Ts (Fig. 4 a). Fruit yield was the highest in Rma×Tc; the same result was statically seen in Rma ×Ts (Fig. 4 b).
First, fruit emergence time significantly decreased in grafted plants at this temperature (Fig. 5). The highest fruit weight was in Rma×Tc; fruit diameter did not significantly change between treatments (Fig.  6). TSS increased in Rn and Rmo at Ts (Fig. 7 a). Firmness decreased in Rmo×Ts (Fig. 7 b) results, Goreta et al. (2008) found that shoot weight reduc tion in watermelon grafted onto Cucurbita maxima Duch.× Cucurbita moschata Duch was less) than non grafted plants under salt stresses. The grafted cucum ber plants showed less change in shoot and root weights under Cu stress conditions and this might be due to the lower accumulation of Cu in leaves through their squash rootstocks (Rouphael et al., 2008).
The lowest male flower was seen in Rma at both temperatures (Fig. 3). Conversely, the highest num ber of female flowers was observed in Rma at Ts and Tc compared with other rootstocks at each tempera ture. On the other hand, at the optimum tempera ture (Tc), each rootstock had a lower female flower Table 5 The interaction effect of temperature and cucumber grafting on some elements concentration (%) of leaves showed the highest male flower. Van der Ploeg and Heuvelink (2005) reported that low temperatures reduced the tomato fruit set through poor pollen quality and increased the period between anthesis and fruit maturity resulting in lower fruit yields. The result of Khah et al. (2006) showed that tomatoes cv. Big Red grafted onto cv. Heman and Primavera pro duced more fruit at low temperatures compared to nongrafted plants in greenhouse conditions. In our study, fruit number decreased at Ts and grafting had a beneficial effect on fruit number, but did not influ ence fruit yields. In agreement with our study, com mercial tomato grafting was not able to improve the reduction of tomato yields under low light stresses by shading (Krumbein and Schwarz, 2013). The economic increase of yield imparted by select vigorous commercial rootstocks (Kyriacou et al., 2017) like increasing tomato yields was observed by    grafting tomatoes onto Kagemusia and Helper's root stocks by Lee et al. (2007). SantaCruz et al. (2001) grafted tomato cv. Moneymaker onto Pera and observed that growth and yield of tomato increased under salt stress conditions. Similar results were observed by Estan et al. (2005), who found that tomato yield increased through grafting under salt stresses and also reported that fruit yield in hetro grafted plants increased more compared with self grafted plants. Ruiz et al. (1997) believed that increase yield and growth of the grafted plants might be due to an increase in the nutrient and water uptake fed by vigorous rootstock. Huang et al. (2010) reported that cucumbers grafted had a higher fruit yield than nongrafted plants under salt stresses. Colla et al. (2010) found that increasing CO 2 assimila tion by cucumber grafting increased fruit yields.

The interaction effect of temperature and cucumber grafting on some physiological characteristics and nutrient concentration of cucumber
The photosynthesis rate decreased in Rn and Rs at Ts compared with Tc. However, it did not significantly change in Rma and Rmo (Fig. 8 a). Transpiration decreased at Ts in all rootstocks compared with Tc ( Fig. 8 b). Stomata conductance gradually increased in Rs, Rma, and Rmo at Ts compared with Tc and reached the highest in Rmo×Ts (Fig. 8 c). Mesophyll conductance did not significantly change in each rootstock at Tc and Ts, except for ungrafted cucum bers in which it increased at Ts compared with Tc ( Fig. 8 d).
El and RWC were generally the lowest and the highest at TS, respectively. the highest EL was seen in Rmo×Ts. The maximum RWC, at Ts, was observed in Rn, Rma, and Rmo ( Fig. 9 a, b). Antioxidant increased in Rn and Rs at Ts and did not change in other treat Fig. 7 The interaction effect of temperature and cucumber grafting on TSS (a) and firmness (b). Tc= optimum tempe rature, Ts= low temperature; Rn = nongrafted (Cucumis sativus var. DavosII); Rs= self grafted; Rma= Cucurbita maxima; Rmo= Momordica charantia. ments (Fig. 10 a). Phenol content increased in all plants at Ts and reached the highest point in Rmo×Ts (Fig. 10 b). Proline did not significantly change among treatments (Fig. 10 c). Increasing antioxidant content with using root stock was reported by Rouphael et al. (2018); this may help grafted plant to have a better function under different stress.
The photosynthesis rate and SPAD value showed that the SPAD value did not change in Rmo. Thus, photosynthesis did not change. However, the same result was not seen in Rma. Therefore, it seems that another mechanism interfered with the stability of photosynthesis in Rma in this condition rather than changes of chlorophyll content (Fig. 6 a and 7a).
Some researchers demonstrated that the photo synthesis rate was influenced by leaf area, stomatal conductance, and chlorophyll content. In our study, lowtemperature stress caused a reduction in SPAD value and resulted in photosynthesis reduction. On the other hand, the photosynthesis in grafted cucum bers having high SPAD value and stomata conduc tance improved under lowtemperature stresses. Davis et al. (2008) reported that grafting increased photosynthesis under salt stress conditions too.
Grafted cucumbers showed higher photosynthesis rates, stomatal conductance, and intercellular CO 2 under salt stresses compared with nongrafted plants (Yang et al., 2006). Massai et al. (2004) and Moya et al. (2002) reported that grafting improved photosyn thesis under salt stress conditions, too. Rootstock induced changes in stomatal development, as reduced transpiration relates to lower stomatal den sity in grafted plants (Kumar et al., 2017). Zhou et al. (2009) reported that increasing the photosynthesis rate by grafting plants onto a differ ent rootstock might be due to a decrease in the ROS concentration in the leaves. Beside, grafting increased water and nutrient uptake of the plant and could enhance its photosynthesis (MartinezBallesta Fig. 9 The interaction effect of temperature and cucumber grafting on El (a) and RWC (b) Tc= optimum temperature, Ts= low temperature, Rn = nongrafted (Cucumis sativus var. DavosII); Rs= (self grafted); Rma= Cucurbita maxima, Rmo= Momordica charantia.  (2010) reported that melon grafting had high CO 2 assimilation due to increasing stomata conductance and Ci and resulted in an increase in photosynthesis. The photosynthesis rate and the fresh shoot weight of Rmo and Rma did not significantly change at both temperatures. Thus, low temperatures could not affect the photosynthesis rate in Rmo and Rma. On the other hand, decreasing the fresh weight was seen at Ts in Rn due to a decrease of photosynthesis rate in these rootstocks (Fig. 7).
Overall, low stomata and mesophyll conductance of ungrafted cucumbers which have adverse effects on water relation, mineral nutrient uptake and trans port (Kumar et al., 2017), carbohydrate and hormone relationship, photosynthesis and respiration rates result in a decline in yields (Kozlowiski, 1984;Barrick and Noble, 1993;Bacanamwo and Purcell, 1999). These results were in line with our results at low temperature stresses. Still, when cucumbers were grafted onto Karella they were kept under proper conditions for more efficient photosynthesis, and consequently, a better yield was produced.
The result of the study showed that salinity stress es reduced the K concentration in melons and cucumbers and grafted plants had higher nutrient concentrations than nongrafted plants. Many root stocks are capable of increasing the uptake and translocation of nutrients (Kumar et al., 2017). In agreement with our result, Colla et al. (2013) report ed that cucumber grafting had no influence on K and P concentrations in the leaves, root, and fruit under salt stresses. Our result showed that the highest K concentration was in Rmo at both temperatures and P concentration increased by grafting at Tc.

Conclusions
It seems that the photosynthetic traits of grafted cucumbers were not affected mainly by Rmo and Rma at stress conditions. On the other hand, hor monal changes or nutrient uptake of these root stocks seemingly caused lesser root and additional vegetative growth but stimulated more fruit and yields. The effect of these rootstocks on productivity growth resulted in more male flowers than female ones, which was predictable according to the produc tive growth model of Cucurbitaceae in which male flowers and then female flowers emerged in the bush. Accordingly, if this experiment lasted more, it might result in more female flowers, suggesting that further investigations might be needed. All in all, using Rmo as well as Rma is recommended after test ing the fruit quality and the economic yield.