Involvement of ethylene biosynthesis and signalling in fruit set and early fruit development in zucchini squash (Cucurbita pepo L.)

RESEARCH ARTICLE Open Access Involvement of ethylene biosynthesis and signalling in fruit set and early fruit development in zucchini squash (Cucurbita pepo L.) Cecilia Martínez 1 , Susana Manzano 1 , Zoraida Megías 1 , Dolores Garrido 2 , Belén Picó 3 and Manuel Jamilena 1* Abstract Background: We have identified a kind of parthenocarpy in zucchini squash which is associated with an incomplete andromonoecy, i.e. a partial conversion of female into bisexual flowers. Given that andromonoecy in this and other cucurbit species is caused by a reduction of ethylene production in the female flower, the associated parthenocarpic development of the fruit suggested the involvement of ethylene in fruit set and early fruit development. Results: We have compared the production of ethylene as well as the expression of 13 ethylene biosynthesis and signalling genes in pollinated and unpollinated ovaries/fruits of two cultivars, one of which is parthenocarpic (Cavili), while the other is non-parthenocarpic (Tosca). In the latter, unpollinated ovaries show an induction of ethylene biosynthesis and ethylene signal transduction pathway genes three days after anthesis, which is concomitant with the initiation of fruit abortion and senescence. Fruit set and early fruit development in pollinated flowers of both cultivars and unpollinated flowers of Cavili is coupled with low ethylene biosynthesis and signalling, which would also explain the partial andromonoecy in the parthenocarpic genotype. The reduction of ethylene production in the ovary cosegregates with parthenocarpy and partial andromonoecy in the selfing progeny of Cavili. Moreover, the induction of ethylene in anthesis (by ethephon treatments) reduced the percentage of bisexual parthenocarpic flowers in Cavili, while the inhibition of ethylene biosynthesis or response (by AVG and STS treatments) induces not only andromonoecy but also the parthenocarpic development of the fruit in both cultivars. Conclusions: Results demonstrate that a reduction of ethylene production or signalling in the zucchini flower is able to induce fruit set and early fruit development, and therefore that ethylene is actively involved in fruit set and early fruit development. Auxin and TIBA treatments, inducing fruit set and early fruit development in this species, also inhibit ethylene production and the expression of ethylene biosynthesis and response genes. A model is presented that discusses the crosstalk between ethylene and auxin in the control of fruit set and early fruit development in zucchini squash. Keywords: Cucurbita pepo, Fruit set, Parthenocarpy, Ethylene, Gene expression, Auxin Background Despite its molecular simplicity, ethylene regulates a number of developmental and physiological processes [1,2], including leaf and flower abscission, ripening of climacteric fruit and biotic and abiotic stresses. In the species of the Cucurbitaceae family ethylene controls sexual expression and is the main determinant of sexual phenotypes [3-5]. Thus, the ethylene biosynthesis genes CmACS7 and CsACS2 of melon and cucumber, respectively, regulate the arrest of stamen development in female flowers of monoecious cultivars, and their loss of function mutations lead to the conversion of female into bisexual flowers, and therefore the transformation of monoecious into andromonoecious cultivars [5-7]. In Cucurbita pepo, ethylene also regulates the sexual expression of monoecious cultivars, controlling both the precocity and the number of female flowers [4,5,8,9]. In * Correspondence: [email protected] 1 Departamento de Biología y Geología, Agrifood Campus of International Excellence (ceiA3), Universidad de Almería, La Cañada de San Urbano s/n, 04120 Almería, Spain Full list of author information is available at the end of the article © 2013 Martínez et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Martínez et al. BMC Plant Biology 2013, 13:139 http://www.biomedcentral.com/1471-2229/13/139 fact, the application of blocking agents of ethylene production (AVG) or perception (STS) is able to delay female flowering and reduce the number of female flowers per plant, but also to induce a conversion of female into bisexual flowers [4]. Likewise, female flowers of zucchini produce much more ethylene than male flowers throughout their development and maturation up to anthesis [4]. After pollination and fertilization, fruit set and fruit development is dependent on cell division and expansion promoted by hormones such as gibberellin (GAs), auxin and cytokinin [10-12]. Auxin is the determinant of fruit set, as has been demonstrated by the study of mutants or transgenic lines for ARF or IAA/Aux multigene families in tomato and Arabidopsis [10-12]. However, they seem to be closely related to gibberellin which is able to trigger fruit initiation without changes in auxin signalling genes [13,14]. In the Cucurbitaceae family, fruit set and development depend mainly on auxin [15]. The application of auxins induces parthenocarpic fruit set and development in cucumber [16], although the application of cytokinin also activates cell divisions in fruits [17], whereas brassinosteroid increase fruit set [10,18]. In zucchini auxin is also shown to be the most effective hormone to induce parthenocarpic fruit development [19], and this growth regulator is commonly applied to promote fruit set and growth in greenhouse production of this vegetable crop. Ethylene has been related with floral organ senescence and abscission after pollination. Pollination induces ethylene production in the ovaries and petals, and this ethylene appears to be responsible for coordinating ovary growth and petal senescence [20-23]. The implication of ethylene in fruit set and development was not studied in depth until a few years ago. Recent studies have shown an interconnection between early ovule abortion and the size of the silique in Arabidopsis ethylene mutants [24]. Meanwhile, pollination and gibberellin treatments are responsible for downregulating ethylene biosynthesis and signalling genes in tomato immediately after fruit set [12,22]. Since C. pepo morphotype zucchini has a large inferior ovary (about 6–8 cm long at anthesis), it is an ideal species suitable to study fruit set and early fruit development. We have recently observed that a reduction of ethylene in female flowers of zucchini can not only promote the development of stamens in the flower, converting female into bisexual flowers, but also induce the parthenocarpic fruit development in absence of pollination and fertilization [25]. This is also true for some cultivars of zucchini squash grown under high temperature conditions [25]. To study the role of ethylene in fruit set and early fruit development in zucchini squash, the present paper compares the production of ethylene and the expression of ethylene biosynthesis and response genes between pollinated and unpollinated ovaries of a non-parthenocarpic cultivar, as well as between a parthenocarpic and a non-parthenocarpic cultivar of zucchini squash. Moreover, we analysed fruit growth rates as well as ethylene production and the expression of ethylene genes in response to auxin and TIBA treatments. Results indicate that ethylene is directly involved in fruit set and early fruit development in this species. These two developmental processes require a low level of ethylene production and signalling within the few days after pollination, and the loss of pollination and fertilization is accompanied by an induction of ethylene biosynthesis and signalling 3 days after anthesis, concomitantly with fruit abortion. Results Different approaches have been used to determine the implication of ethylene in zucchini fruit set and early fruit development. Firstly we have determined ethylene production and the expression of 13 ethylene biosynthesis and signalling genes in pollinated and unpollinated ovaries and fruits of the non-parthenocarpic cultivar Tosca. Secondly, we have compared ethylene production and the expression of ethylene genes in the fruits of two contrasting cultivars for parthenocarpy: Tosca and Cavili. Finally, the same two cultivars were used to study the crosstalk between ethylene and auxins in the control of these developmental processes. Ethylene production and ethylene genes expression in pollinated and unpollinated ovaries/fruits of zucchini It has been shown that the fruits of many zucchini cultivars can initiate their growth in the absence of pollination and hormone application. In fact, in the nonparthenocarpic cv. Tosca the growth rate of pollinated and unpollinated fruits were very similar for the first three days (Figure 1A), which highlights the natural parthenocarpy of this species. After the third day, however, most unpollinated fruits aborted, and some of them grew at a significantly slower rate than pollinated fruits (Figure 1A). In the first 5 DPA, the profiles of ethylene production in pollinated and unpollinated fruits were very dissimilar (Figure 1B). While ethylene production decreased slightly in pollinated fruits, in unpollinated fruits it increased sharply 3 DPA (Figure 1B). These results indicate that the decrease in the growth rate of unpollinated fruits that occurs at 3 DPA is correlated with a burst of ethylene in the fruit, and that the maintenance of fruit growth likely requires a low level of ethylene. To detect putative genes regulating ethylene biosynthesis during fruit set and early fruit development, we have analysed the expression of six ACS genes and one ACO gene from C. pepo (Additional file 1: Table S1, Martínez et al. BMC Plant Biology 2013, 13:139 Page 2 of 14 http://www.biomedcentral.com/1471-2229/13/139 Figure 2). CpACS1 was one of the first ACS genes isolated from plants [3]; CpACS2, and CpACS4 to CpACS6 are four unigenes identified by new generation sequencing (NGS) of cDNA from different squash tissues [26; available at Cucurbigene.net]; and CpACS7 and CpACO1 have been isolated by PCR strategy with degenerated primers (unpublished). Expression was studied through qPCR at anthesis (0 DPA) and 3 DPA (Figure 2). CpACS1 was not expressed during this period (Figure 2). At anthesis, the expression of the other 6 genes was very low, except for CpACS4, which could explain the basal production of ethylene at this point. This basal expression was maintained in pollinated fruits at 3 DPA (Figure 2). Nevertheless, in unpollinated fruits the expression of the five ACS genes and CpACO1 was upregulated at 3 DPA. The highest induction was observed in CpACS4,CpACS6 and CpACS7 (Figure 2), which appear to be the main gene responsible for the ethylene produced at this point. To determine the molecular mechanisms behind the action of ethylene in fruit set and early fruit development, the relative expressions of six additional ethylene perception and signalling genes were also studied in pollinated and unpollinated fruits over 5 DPA. As occurred for CpACSs and CpACO1, other genes involved in ethylene perception such as CpETR1 and CpERS1, as well as in ethylene signalling (CpCTR1,CpCTR2,CpEIN3.1 and CpEIN3.2), also showed an expression profile similar to that of ethylene production (Figure 3). In pollinated fruits, expression levels remained low during early development, while in non-pollinated fruits, all perception and response genes were significantly upregulated at 3 DPA (Figure 3), coinciding with the peak of ethylene and the abortion of fruit growth (Figure 1). A b bb a a a a a a 0 50 100 150 200 250 012345 Fruit lenght (mm) Days post anthesis (DPA) pollinated unpollinated a B aaaa aa b b b a 0 5 10 15 20 25 30 012345 Ethylene production (nLgr-1 FW) Days post anthesis (DPA) Figure 1 Evolution of the size and ethylene production in pollinated and unpollinated ovaries/fruits of zucchini cv. Tosca. (A) Fruit size. (B) Ethylene production. Different letters indicate significant differences between the means at each data point between pollinated and unpollinated ovaries/fruits (p ≤0.05; n = 12 for fruit length, and n = 4 for ethylene production). 0 5 10 15 20 25 30 35 03030303030303 CpACS1 CpACS2 CpACS4 CpACS5 CpACS6 CpACS7 CpACO1 Relative expression Days post anthesis (DPA) pollinated unpollinated Figure 2 Relative expression of 7 ethylene biosynthesis genes in pollinated and unpollinated ovaries/fruits of zucchini cv. Tosca at anthesis (0 DPA) and 3 DPA. In pollinated flowers, expression at the day of anthesis was determined 8 h after hand pollination. Each data point represents the mean of 3 replicates with 3 fruits each. Error bars indicate SE. Martínez et al. BMC Plant Biology 2013, 13:139 Page 3 of 14 http://www.biomedcentral.com/1471-2229/13/139 Ethylene is actively involved in fruit set and early fruit development in zucchini In order to ascertain whether the differences in ethylene production between pollinated and unpollinated fruits at 3 DPA are the consequence of fruit set or fruit abortion, or rather a signal that is actively involved in fruit set and development, we have compared ethylene production and signalling in one parthenocarpic and one non-parthenocarpic variety, as well as in a segreganting population derived from the parthenocarpic one, and determined the effects of ethylene releasing and blocking agents on fruit set and early fruit development. The hybrid cultivar Cavili is able to develop parthenocarpic fruits of commercial size in absence of pollination or hormonal treatments [25]. The parthenocarpy of this cultivar is associated with an incomplete andromonoecy, i.e. a partial conversion of female into bisexual flowers, and a delay in floral organ maturation (Figure 4). Ovarybearing flowers of this cultivar can be classified into female flowers with no stamen development (Figure 4A), 0 5 10 15 20 25 30 35 0235 Relative Expression Days post anthesis (DPA) CpCTR1 0 5 10 15 20 25 30 35 0235 Relative expression Days post anthesis (DPA) CpCTR2 0 2 4 6 8 10 12 14 0235 Relative expression Days post anthesis (DPA) CpEIN3.1 0 1 2 3 4 5 6 0235 Relative expression Da y s post anthesis (DPA) CpEIN3.2 0 5 10 15 20 25 30 0235 Relative expression Days post anthesis (DPA) CpETR1 0 20 40 60 80 100 120 140 160 180 0235 Relative Expression Days post anthesis (DPA) CpERS1 0 1 2 3 4 5 6 7 0235 Relative expression Days post anthesis (DPA) CpACS4 pollinated unpollinated Figure 3 Relative expression of ethylene biosynthesis (CpACS4), perception (CpETR1 and CpERS1) and signalling (CpCTR1,CpCTR2, CpEIN3.1 and CpEIN3.2) genes over 5 DPA in pollinated and unpollinated ovaries/fruits of cv. Tosca. Each data point represents the mean of 3 replicates with 3 ovaries/fruits each. Error bars indicate SE. Martínez et al. BMC Plant Biology 2013, 13:139 Page 4 of 14 http://www.biomedcentral.com/1471-2229/13/139 or bisexual flowers showing a certain degree of stamen development (Figure 4B), but never reaching the size of stamen in male flowers (Figure 4C). At the same stage of development, bisexual flowers always showed a higher ovary and fruit size (Figure 4D). We compared the longitudinal growth rate of ovaries/fruits between female and bisexual flowers for a total of 22 days, starting with floral buds of about 4 mm in length (Figure 4E and F). The growth rate of ovary length in bisexual flowers deviated from that of female flowers at twelve days, immediately after anthesis of female flowers (Figure 4F). By this time petals of bisexual flowers were still immature and closed but their ovaries kept growing at a much faster rate than those of female flowers. Many of the ovaries in bisexual flowers reached a commercial size before anthesis. In fact, many of the bisexual flowers did not reach anthesis in the 24 days of study, and in others anthesis was delayed with respect to female flowers because of a lower growth rate of petals (Figure 4F). These results indicate that the parthenocarpy of this cultivar is not only correlated with stamen development, but also with a lower growth rate of petals, which delays maturation of petals and anthesis. Given that male flowers require twice as long as female ones to mature and reach anthesis (Figure 4F), it is likely that the delay in the maturation of bisexual flowers is associated with their masculinisation, i.e. the presence of stamens. To assess whether parthenocarpy and partial andromonoecy had the same genetic regulation, we phenotyped the F2 population derived by self-pollination of the F1 hybrid Cavili, which segregates for the two traits. A complete cosegregation between the two traits has been found in the F2 generation. Of a total of 95 plants, 23 were completely monoecious, and produced only female flowers,while72werepartially andromonoecious and produced both female and bisexual flowers (Table 1). Although the number of bisexual flowers in the latter varied from 20% to 100%, all of them C aaaaa b c a bbba aaaaa a F 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 1012141618202224 Petals length (mm) Time (days) anthesis bisexual flowers anthesis female flowers anthesis male flowers aaa a a a aaaaaa b b b b bE 0 50 100 150 200 250 300 350 400 0 2 4 6 8 101214161820 Ovary length (mm) Time (days) Cavili Female Cavili Bisexual Cavili Male ABD 50 mm 10 mm 10 mm 10 mm Figure 4 Growth rates of ovaries/fruits and petals in female, bisexual and male flowers of the parthenocarpic cultivar Cavili. (A) Female flower. (B) Bisexual flower with a partial development of stamen. (C) Male flower. (D) Ovary size of female and bisexual flowers with the same time of development. Flowers were tagged when they were 4 mm in length (time 0), and allowed to grow for 20 days in the case of female and bisexual flowers, and 24 days in the case of male flowers. (E) Comparison of ovary/fruit longitudinal growth rates between female and bisexual flowers. (F) Comparison of petal longitudinal growth rates of female, bisexual and male flowers. Note that petal growth is delayed in bisexual flowers, but not as much as in male flowers. Anthesis in the bisexual flowers is delayed even more than in male flowers. Statistical analysis was performed using the LSD method (p ≤0.05; n = 15). Martínez et al. BMC Plant Biology 2013, 13:139 Page 5 of 14 http://www.biomedcentral.com/1471-2229/13/139 developed into parthenocarpic fruits, which reached commercial size even before anthesis (Table 1). The 3:1 segregation ratio (χ 2 = 0,014, p value = 0,91) indicated that the partial andromonoecy and parthenocarpy of Cavili appears to be controlled by at least one dominant gene. Sex determination and female flower maturation in zucchini is known to be regulated by ethylene in the earliest stages of flower development [4,8,9]. Therefore, the parthenocarpy of Cavili could be the result of a reduction of ethylene in female flowers. We have found that the unpollinated Cavili ovaries/fruits produce significantly less ethylene than those of the non-parthenocarpic Tosca during the days immediately after anthesis (Figure 5B). The production of ethylene was also measured in the ovaries of 25 female and 25 bisexual sample flowers derived from the F2 population of Cavili. Results indicated that the higher growth rate of fruits in bisexual flowers cosegregated with a significant reduction of ethylene production in the ovary at 3 DPA (Table 1). All the analyzed ACS genes, including CpACS4, CpACS6 and CpACS7, which were those that mainly regulate ethylene production in unpollinated ovaries in the days immediately after anthesis, showed no significant differential expression between Cavili and Tosca unfertilized ovaries at 3DPA or showed higher expression in Cavili (Figure 5C). Only the expression of CpACO1 was lower in Cavili (Figure 5C). It appears therefore that the reduction in ethylene production observed in the unpollinated ovary of the parthenocarpic cultivar Cavili is not regulated at the level of transcription. For ethylene perception and response genes, only the transcripts of CpERS1 showed a lower accumulation of transcripts in Cavili (Figure 5C), suggesting that perception of ethylene could also be altered in this parthenocarpic cultivar. To confirm whether a reduction in ethylene production during the development of female flowers was enough to induce the parthenocarpic development of the zucchini ovary, we determined the effects of ethylene releasing and blocking agents on early fruit development. Control plants of Tosca produced no bisexual parthenocarpic flowers, while those of Cavili produced 60% (Figure 5D and Additional file 2: Figure S1). The application of ethephon significantly reduced the production of bisexual parthenocarpic flowers in Cavili, while the application of the ethylene blocking agents STS and AVG increased the production of bisexual parthenocarpic flowers not only in the parthenocarpic Cavili, but also in Tosca (Figure 5D and Additional file 2: Figure S1). These results demonstrate that ethylene is actively involved in fruit set and early fruit development in zucchini squash, and indeed a reduction of either ethylene biosynthesis or signalling in the developing female flower of zucchini can not only inhibit the arrest of stamens, promoting the conversion of female into bisexual flower, but also to induce the parthenocarpic development of the fruit. Effects of auxins on fruit growth rates and ethylene biosynthesis and signalling It is known that external application of auxins and TIBA, the latter an inhibitor of auxin polar transport, can induce fruit set and early fruit growth in different species. In zucchini, synthetic auxins are commonly used to stimulate the parthenocarpic development of fruit in off-season greenhouse production. We have studied the effects of NAA + NAAmide and TIBA (applied at anthesis in the ovary) on fruit development and ethylene production in the cultivars Cavili and Tosca. In the parthenocarpic cv. Cavili neither auxins nor TIBA were able to alter the longitudinal growth rate of the fruit (Figure 6A), suggesting that the ovaries of this cultivar could have a high concentration of auxins. However, in the fruits of the non-parthenocarpic cv. Tosca both treatments promoted longitudinal growth of the fruit, although TIBA was more effective than NAA + NAAmide (Figure 6C). Ethylene production in control and treated fruits was negatively correlated with early fruit growth rate. In Cavili, ethylene decreased progressively throughout the first 4 DPA in both control and treated fruits (Figure 4B), while in Tosca the only fruits where ethylene was induced at 3 DPA were the unpollinated control fruits, which were those which showed the lowest growth rate and finally aborted (Figure 6D). The effect of NAA and TIBA on the expression of ethylene genes was studied in fruits at 3 DPA, when differences in ethylene production were evident between pollinated and unpollinated fruits, and between parthenocarpic and non-parthenocarpic cultivars. In concordance with ethylene production data, ACS and ACO genes were downregulated by NAA and TIBA in the Table 1 Fruit size and ethylene production of female and bisexual flowers among monoecious and partially andromonoecious plants in the selfing progeny of Cavili (F2 population) F2 segregation (No. plants) Flower phenotype Ovary length at anthesis (mm) Ethylene production at 3 DPA (nL/gr FW) Monoecious (23 plants) Female 78.10±4.23 a 10.90±2.94 a Partially andromonoecious (72 plants) Female 77.43±1.00 a 10.90±2.94 a Bisexual 139.07±4.75 b 2.51±1.23 b Ovary size and ethylene production data were obtained from 25 replicates for each flower phenotype. Different letters within the same column indicate significant differences between female and bisexual flowers (t-analysis, p<0,01). Martínez et al. BMC Plant Biology 2013, 13:139 Page 6 of 14 http://www.biomedcentral.com/1471-2229/13/139 fruits of both Tosca and Cavili at 3 DPA (Figure 7), indicating that auxins regulate negatively the production of ethylene in the fruit during the days immediately after anthesis. TIBA treatment was more effective than treatment with NAA + NAAmide in reducing the expression of ethylene biosynthesis genes (Figure 7). Moreover, the downregulation of CpACS4 and CpACS7, the genes which contribute most to the production of ethylene in unpollinated fruits during the days immediately after anthesis, was higher than that observed for CpACS5,CpACS6 and CpACO1 (Figure 7). The expression of the ethylene receptor CpETR1 was similar in control ovaries of Tosca and Cavili, and was significantly reduced in response to both treatments in Cavili, and in response to TIBA in Tosca. The expression of the other receptor gene CpERS1 was higher in the non-parthenocarpic cultivar Tosca, and it was downregulatedbyNAAandTIBAinthiscultivar (Figure 7). Regarding ethylene response genes, the expression of CpCTR1 was also downregulated by TIBA in the two cultivars (Figure 7), but not by NAA; and although the hormonal treatments did not significantly change the expression of CpEIN3.1 and CpEIN3.2 in Tosca (Figure 7), in Cavili NAA downregulated the expression of both, whereas TIBA only reduced the expression of the former. In conclusion, many of the analysed ethylene biosynthesis, perception and signalling genes were downregulated by NAA and TIBA, two treatments that induce the growth rate of the zucchini fruit in the days immediately after anthesis. b b b b b a a a a a 0 50 100 150 200 250 300 350 400 02468 Fruit length (mm) Days post anthesis (DPA) A Cavili Tosca a C a A ab B b C 0 10 20 30 40 50 60 70 80 90 100 Tosca Cavili Percentage of bisexual flowers/ parthenocarpic fruit Control Etephon STS AVG D B aa aa a b b 0 2 4 6 8 10 12 14 16 18 20 01234 Ethylene production 6 h (nL gr-1 FW) Days post anthesis (DPA) Cavili Tosca C 0 1 2 3 4 5 6 7 Relative expression 3DPA Cavili Tosca Figure 5 Involvement of ethylene in the parthenocarpic development of Cavili fruit. (A) Comparison of ovary/fruit growth between the parthenocarpic cv. Cavili (bisexual and unpollinated flowers) and the non-parthenocarpic cv. Tosca (female and unpollinated flowers). (B) Comparison of ethylene production in unpollinated ovaries/fruits of Cavili and Tosca (C) Expression of ethylene biosynthesis, perception and response genes in unpollinated ovaries of Cavili and Tosca at 3 DPA. (D) Effects of ethylene releasing (ethephon) and blocking reagents (AVG and STS) on the percentage of bisexual parthenocarpic flowers in Cavili and Tosca. Different letters indicate statistical differences between treatments within each cultivar (p ≤0.05; n = 15; lowercase Tosca; uppercase, Cavili). Martínez et al. BMC Plant Biology 2013, 13:139 Page 7 of 14 http://www.biomedcentral.com/1471-2229/13/139 Discussion Fruit set and early fruit development is correlated with a downregulation of ethylene biosynthesis and signalling genes Ethylene production and expression data from 13 genes covering ethylene biosynthesis and signal transduction pathway in C. pepo have demonstrated that the loss of pollination in a non-parthenocarpic genotype of this species is accompanied by a sharp increase in ethylene biosynthesis and signalling in the ovary at 3 DPA. This is true for the 5 ACS-like genes showing expression in the ovary (one of the analysed genes, CpACS1, showed no expression in the ovary during this time of development) and CpACO1, as well as for two ethylene receptors (CpETR1 and CpERS1), two CTR1-like genes (CpCTR1 and CpCTR2) and two EIN3-like genes (CpEIN3.1 and CpEIN3.2). This ethylene signalling in the ovary few days after anthesis is associated with a decline in growth and finally with the abortion of fruit. Pollinated flowers, on the other hand, maintain a very low level of ethylene biosynthesis and signalling in the fruit during the days immediately after anthesis, concomitantly with fruit set and development. This lower ethylene signalling is also observed in parthenocarpic fruits of the zucchini cultivar Cavili, as well as in the parthenocarpic plants of the segregant selfing progeny of Cavili. Taken as a whole, these data indicate that the production of ethylene in the ovary few days after anthesis is negative correlated with fruit set and growth. Although the function of ethylene in fruit set and early fruit development has not been studied in depth, it has been reported that the inhibition of ethylene production or response by external treatments with AVG, STS or 1-methylcyclopropene (1-MCP), induces fruit set in pear a a a a a A 0 50 100 150 200 250 300 350 400 02468 Fruit length (mm) Days post anthesis (DPA) Control NAA+NAAmide TIBA a a a C b b b a a c c c 0 50 100 150 200 250 300 350 400 02468 Fruit length (mm) Days post anthesis (DPA) Control NAA+NAAmide TIBA aa b b a D a a 0 5 10 15 20 25 1234 Ethylene production (nL gr-1 FW) Days post anthesis (DPA) a ab a a B a a a ab 0 2 4 6 8 10 12 1234 Ethylene production (nL gr-1 FW) Days post anthesis (DPA) Tosca Cavili Fruit length Ethylene Production Figure 6 Effects of NAA + NAAmide and TIBA on fruit growth rates and the evolution of ethylene production in Cavili and Tosca. (A) Fruit growth rates in Cavili. (B) Ethylene production in Cavili. (C) Fruith growth rates in Tosca. (D) Ethylene production in Tosca. Ovaries were treated at anthesis. Different letters at the same time point indicate significant differences between treatments (p ≤0.05; n = 15 for ovary growth and n = 4 for ethylene production). Martínez et al. BMC Plant Biology 2013, 13:139 Page 8 of 14 http://www.biomedcentral.com/1471-2229/13/139 0 0.5 1 1.5 2CpACS7 Relative expression 3 DPA 0 0.5 1 1.5 2CpEIN3.1 0 0.5 1 1.5 2CpCTR1 0 1 2 3 4 5CpERS1 0 1 2 3CpACO1 0 0.5 1 1.5 2CpACS4 C NAA TIBA 0 0.5 1 1.5 2CpACS5 0 0.5 1 1.5 2CpACS6 Tosca ToscaCavili Cavili 0 0.5 1 1.5 2 2.5 3CpEIN3.2 0 0.5 1 1.5 2CpETR1 Figure 7 Effect of NAA and TIBA on the expression of ethylene genes. Flowers were treated at anthesis, and the expression of ethylene biosynthesis, perception and signalling genes was determined in the ovaries/fruit at 3 DPA, when ethylene production is induced in unpollinated ovaries. Each data point represents the mean of 3 replicates with 3 ovaries/fruits each. Error bars indicate SE. Martínez et al. BMC Plant Biology 2013, 13:139 Page 9 of 14 http://www.biomedcentral.com/1471-2229/13/139