THE ISOLATION OF cDNA CLONES FROM CUCUMBER (Cucumis ...

CELLULAR & MOLECULAR BIOLOGY LETTERS

Volume 8, (2003) pp 421 – 438 http://www.cmbl.org.pl Received 12 February 2003 Accepted 30 April 2003 THE ISOLATION OF cDNA CLONES FROM CUCUMBER (Cucumis sativus L.) FLORAL BUDS COMING FROM PLANTS DIFFERING IN SEX ZBIGNIEW PRZYBECKI*, MAGDALENA EWA KOWALCZYK, EWA SIEDLECKA, EWA URABAŃCZYK-WOCHNIAK and STEFAN MALEPSZY Department of Plant Genetics, Breeding and Biotechnology, Faculty of Horticulture, Warsaw Agricultural University, Nowoursynowska 166, 02-787 Warsaw, Poland Abstract: In this study, we found flower cDNA clones which may be connected with the development of flower sex in cucumber. Two pairs of nearly-isogenic lines: gynoecious GY3 (FFMMGG) versus hermaphrodite HGY3 (FFmmGG) and monoecious B10 (ffMMGG) versus gynoecious 2gg (ffMMgg) were used for clone isolation. To obtain differentially-expressed clones, we applied the differential screening method. 454 clones from GY3 and 478 from B10 cDNA libraries were isolated. The results of RFLP analysis with 56 cDNA clones showed no clones which cosegregated with sex in cucumber. The 28 cDNA B10 and 33 cDNA GY3 clones isolated using the differential screening method were sequenced. Some of them seem to may play a role in cell differentiation or flower development. Among the 61 identified clones, 14 show high homology to plant proteins, although of unknown function. 11 show high homology to known proteins, and the possible function of some of them is discussed. For 3 clones, no significant similarity was found. The 31 clones displayed high homology to plant cDNA in EST database. The patterns of expression of five differential cDNA clones, 35GY3, 216GY3, 47GY3, 100B10 and 157B10, were analyzed in cucumber flower buds using in situ RT-PCR. The most interesting clone is 35GY3, because of its possible role in the inhibition of the development of male specific elements in the female cucumber flower Key Words: Cucumber, Cucumis sativus L., Sex Expression, Flower Development, Differential Screening, in situ RT-PCR. * Corresponding author, E-mail: [email protected] Abbreviations used: RFLP - restriction fragment length polymorphism; EST - expressed sequence tag; RT-PCR - reverse transcription polymerase chain reaction; FAA formaldehyde acetic acid

422

CELL. MOL. BIOL. LETT.

Vol. 8. No. 2. 2003

INTRODUCTION Numerous experiments have been performed to isolate the genetic factors of sex determination in flowering plants. Cucumber is a monoecious species in which sex expression has been extensively studied. Up to now, only two sex genetic factors have been isolated in plants: Ts2 in maize, described as alcohol dehydrogenase [1] and CS-ACS1, which is a good sex gene candidate, because of its tight linkage to the F locus in cucumber [2]. Among the various cucumber genotypes, it is possible to distinguish a wide range of types of flower: staminate, bisexual and pistillate [3]. The main genetic factors responsible for sex determination have been described, but the mechanism of their action remains unexplained. The current genetic model proposes that sex expression in cucumber is determined by three major genes. Female sex expression is determined by a single dominant gene, F (female), which is modified by In-F (Intensifier of female sex expression) and gy (gynoecious – described here as g) to enhance gynoecious sex expression [3, 4]. The F locus interacts with two other recessive loci m (andromonoecious) and a (androecious), the latter of which determines the degree of maleness expressed in the plants [4]. Young cucumber floral buds (1-2 mm in size) have uniformly differentiated organs of both sexes. All buds contain stamen and pistil primordia. In later stages, the sex organs become distinctly formed [5, 6]. In pistillate flowers, the delayed development of male organs is accompanied by the normal growth and development of ovaries. The opposite is true for staminate flowers; ovaries are delayed and stamens grow normally [3]. That moment is crucial for uniform buds because sex is differentiating at that stage. In this study, we tried to find genes which could be involved in the formation of the sexual phenotype in cucumber. It is understandable that a gene product which determines sex is expressed when the sex of the floral bud is not yet differntiated. We tried to find differences in pools of transcripts from two pairs of near isogenic lines in order to isolate differentially-expressed genes. We used normal and mutated isogenic lines and isolated the products of genes which were differentially expressed between them. We also present the results of the identification of differentially-expressed clones. To find candidate genes which are really responsible for sex determination, we performed RFLP analysis to check segregation between different sex phenotypes and differentially-expressed cDNA clones in cucumber. To observe the occurrence of transcripts of genes undergoing differential expression in different types of flowers in various cucumber lines, we used the in situ RT-PCR method. MATERIALS AND METHODS Plant material Two pairs of near isogenic lines – monoecious B10 (ffMMGG) line versus gynoecious 2gg (ffMMgg) line and gynoecious line GY3 (FFMMGG) versus


423

hermaphroditic line HGY3 (FFmmGG) – were used to examine sex gene expression in this study. To obtain floral buds, plants were cultivated in plastic pots in the greenhouse with a 16h/8h day/night photoperiod (25ºC-27ºC by day/18ºC-20ºC at night). Light intensity in the greenhouse was about 1500 µmol (quantum)m-2·s-1 of photosynthetically active radiation. The floral buds were collected from four lines (in the case of B10, the floral buds were collected up to the 6th node only). For RNA isolation, the floral buds were immediately frozen in liquid nitrogen and stored at -80ºC. A part of them were fixed in FAA for in situ RT PCR. Segregating populations Two F2 populations of cucumber that segregate for the alleles at the G locus (i.e. monoecious B10 versus gynoecious 2gg) and the M locus (gynoecious GY3 versus hermaphroditic HGY3) were used for the genetic analysis of previously isolated clones. 2gg was crossed to its near isogenic line B10; GY3 was crossed to its near isogenic line HGY3. Two populations of F1 plants were obtained. After self pollination of the F1 populations, we obtained the F2 population. 100 F2 of GY3 x HGY3 and 200 2gg x B10, 10 seeds of each F1 generation and also 10 seeds of each parental line were germinated on wet filter paper in a Petri dish at 28ºC in the dark for 2 days. The resulting seedlings were transferred to soil to a plastic tunnel. The sex of each flower up to node 25-30 on the main stem was assessed and the flowers were classified as male, bisexual or female, and analysed via the chi-square test [7]. Isolation of total RNA and poly (A+) mRNA Total RNA was extracted from floral buds (at the 1-2-mm stage of growth) using the Trizol method (Gibco), and then the RNA was purified by precipitation via the lithium chloride method [8]. Poly(A)+ RNA was isolated from the total RNA with Dynabeads oligo (dT)25 (Dynal). Construction of cDNA libraries and differential screening cDNA libraries were constructed using the UniZAPII-cDNA synthesis kit (Stratagene) with poly (A+) RNA prepared from the 1-2mm cucumber floral buds from two lines (B10 and GY3), according to the Stratagene instructions. The cDNA inserts were cloned in the Eco RI and XhoI cloning site of the ZAP II vector. The libraries densities were: B10 6 x 109 pfu/ml and GY3 3 x 109 pfu/ml. The libraries were screened for differentially-expressed clones using a highly sensitive differential screening method [9]. In the screening of the GY3 (tester) library, HGY-3 was used as a driver, while 2gg was used as the driver for the B10 (tester) library.. A two-round hybridization was performed. Negative clones were purified and the recombinant cDNAs excised from ZAP phages in Bluescript SK (-) using the biological rescue recommended by the supplier (Stratagene).

424


Vol. 8. No. 2. 2003

DNA sequencing and computer analysis cDNA clones were sequenced from the 5’ end by the dideoxy sequencing method [10] using a Big Dye Termination Kit (Perkin Elmer) and a DNA sequencer (model 377; Perkin Elmer). Partial sequence data were compared for homology with the GenBank + SwissProt + PIR + PDB + PRF database using BLAST network service and EST using TBLAST network service Isolation of DNA and bulked segregant analysis Genomic DNA was extracted from young leaves of the four parental lines (B10, 2gg, GY3 and HGY3), two F1 populations (2gg x B10 and GY3 x HGY3) and two F2 segregating populations, and used for Southern blot hybridization. DNA was extracted using a polysaccharide removal method [11]. Equal amounts of DNA were pooled from 20 plants from each group of the F2 population expressing the same sex phenotype. Samples of genomic DNA (5μg each) were digested separately with Eco RI and HinfI and separated by electrophoresis on 1% Tris-borate (1.2 V/cm for 16h). DNA was transferred to Hybond N+ following the standard procedure [12]. The cDNA inserts used as the probes were amplified by PCR. The primers were designed on the basis of vector sequence T7 (5’ gtaatacgactcactatagggc3’) and SK (5’cgctctagaactagtggatc3’). The reaction mixture was incubated in a thermocycler (Gene Amp PCR system 9700 PE Applied Biosystema) for 1 cycle: 95ºC 1’; followed by 35 cycles: 95ºC 30’’; 66ºC 30’’; 72ºC 1,30’’ and finally 4ºC. The PCR products were separated by electrophoresis on 1% agarose gels and isolated using a Qiaex II Gel Extraction Kit (Qiagen). The products were digested with EcoRI and XhoI restriction enzymes to remove vector sequences. The products after digestion were separated and isolated from 2% agarose gels with Qiaex II (Qiagen). The probes were labelled with 32P dCTP via a random priming (Roche) method. Hybridization membranes were analysed via autoradiography with Phosphor Screen Kodak using the FX molecular Imager System (BioRad). In situ RT-PCR The procedure of fixing plant material and in situ RT-PCR was performed according to Urbańczyk et al. [13]. Five pairs of primers were designed on the basis of nucleotide sequences of differential cDNA clones using the “Oligo 5.0” program: 35GY3 Forward primer TTCTCCGACTCACGTTCAACC, Reverse primer GGTTGCATTATCCTTCTGTAG; 216GY3 Forward primer CATTTCTCACCCTCTTATCCA, Reverse primer ATGAAGAAATGGATGTGCGAG; 157GY3 Forward primer GCCTAACTTCGCAGCGCTCAT, Reverse primer AGTGCAATTCCCGCCATGG; 100B10


425

Forward primer GCACTGCCCTGTCAACCTTGT, Reverse primer CCTTCAACTGCAAAACCTCCG; 47GY3 Forward primer TGTCCGAGTTTGATGATGTTC, Reverse primer GATATCGTTTTCAGCTCCCGG. The results of in situ RT-PCR were observed under a microscope (Olympus BX 60) and documented using a CCD-camera (Olympus Μ-CMAD-2) and analyzed using the Photometrics program for V Windows and Adobe Photoshop 6.0. RESULTS AND DISCUSSION Identification of cDNA clones from floral buds To elucidate the developmental processes involved in sex determination in cucumber, we performed a comparative study using two pairs of near isogenic lines. The differential screening method allowed the isolation of differentiallyexpressed cDNA clones. After a second round of hybridization, 454 clones from the GY3 cDNA library and 478 clones from the B10 cDNA library were isolated, and some of them were randomly chosen and sequenced. DNA sequences of the 61 cDNA clones were examined for homology with the EST and GenBank + SwissProt + PIR + PDB + PRF databases, in order to identify their biological function. The length of the clones ranged from 70-690 bases. The clones' accession number in the EST GenBank, sequence similarities and E-Value of the chosen clones to the corresponding genes, together with their GenBank accession number, are shown in Tabs. 1 and 2. Among the 28 identified cDNA B10 and 33 cDNA GY3 clones, 14 of them show high homology to plant proteins and two to bacterial proteins but of unknown function. 11 show high homology to known proteins and the possible function of some of them is discussed. For 3 clones, no significant similarity was found. 31 clones displayed a high homology to plant cDNA in the EST database. The genes which have been putatively or hypothetically identified encode proteins with different cell functions. Some of them (Tabs. 1 and 2) seem to play a role in cell differentiation or flower development. Three of the clones (133B10, 157B10 and 216GY3) show homology to the chaperonin protein family. They are similar to the chaperonin 60 beta chain precursor from potato and garden pea chloroplasts. The plant chloroplast chaperonins (cpn 60 and cpn 10) are key cellular components in numerous folding pathways leading to biologically active proteins [14]. Chaperonins are known as heat shock proteins, but the genes are now known to be induced by a wide variety of environmental or metabolic stresses. The term “heat shock protein” is a misnomer because many agents other than heat induce the expression of the heat shock protein gene; consequently “stress protein” is the preferred term. They can be also activated after hormone treatment. It is possible that such proteins are highly expressed in differentiating floral buds to take part in perception or signalling (ethylene, giberellin or auxin). Ubiquitin protein also belongs to the chaperonin family. One of the isolated clones, 420GY3, shows high homology to E2 ubiquitin


426

Vol. 8. No. 2. 2003

conjugating enzyme. Ubiquitin is a small 8-kd stress protein that may facilitate the targeting and removal of other proteins denatured during stress. Recent studies with Arabidopsis thaliana have revealed that the ubiqutin proteolytic system plays a central role in the auxin-response pathway, and may also be an important member of the GA-dependent signal transduction pathway [15]; in addition ubiquitin ligase is required for auxin response [16]. Aspects of auxin biology, such as auxin biosynthesis, perception and response, are still poorly understood [17]. The mechanism by which sex in cucumber is determined by auxins is not known. Investigations on sex in cucumber revealed that auxin, like ethylene, induces femaleness [18]. A number of experiments have been performed to analyze the role of ubiquitination in auxin signalling [19]. Tab. 1. The sequence similarity of B10 cDNA clones to the corresponding known genes in the database that may play a role in cell differentiation or flower development. Clone 39/B10 45/B10 81/B10 85/B10 95/B10 100/B10 122/B10 133/B10 138/B10 157/B10 166/B10 460B10

ACC. Noa

Descriptionb

BU791029 testa pericarp cDNA clone, Hordeum vulgare BU791030 putative glycine- and prolinerich protein, SP stapfianus BU791036 nucleolar protein, Cicer arietinum BU791039 Tomato shoot meristem cDNA, Lycopersicon esculentum BU791041 Brassinosteroid biosynthetic protein, Pisum sativum BU791043 expressed protein, A. thaliana BU791046 putative glycine- and prolinerich protein, Sporobolus stapfinus BU791048 chaperonin beta subunit protein, Pisum sativum BU791049 bZIP transcription factor protein, A.thaliana BU791050 chaperonin beta subunit protein, Pisum sativum BU791052 flower bud cDNA clone, A. thaliana BU791054 delta cop protein, Zea mays

EACC. Nod Length c Value compared 0.024 BG416265 125 2e-06 CAB61840

480

3e-53 CAA10127

480

8e-36

BG643739

264

2e-37

AF325121

552

2e-58

NP563718

401

4e-06 CAB61840

400

1e-28

489

PO8927

2e-23 CAC40022

467

3e-25

PO8927

588

0.82

AV534491

199

1e-20

AF216852

297

a

Accesion number of the cDNA clone, bA description of the best data match is given together with database accession number of homologous genesd, cSequence similarity

Ubiquitination-mediated proteolysis has emerged as being fundamentally as important as phosphorylation in terms of its involvement in diverse cellular events. In the search for candidates, the Aux/IAA family of transcription factors


427

are suspects. The Aux/IAA are a family of extremely short-lived nuclear proteins (6-80 -minute lifespans). Aux/IAA genes are induced by auxin and auxin also stimulates the rapid degradation of these proteins by increasing their interaction with SCF. The mechanism by which auxin influences SCF-mediated Aux/IAA turnover is not understood (F-box protein interacts with the Skpp1 and Cdc53 cullin proteins to form a ubiquitin ligase complex called SCF) [20]. Tab. 2. The sequence similarity of GY3 cDNA clones to the corresponding known genes in the database that may play a role in cell differentiation or flower development Clone 15/Gy3

ACC. Noa

Descriptionb

EACC. Nod Length Valuec compared 2e-79 BF051149 390

BU791055 Dev./immature green fruit cDNA clone, L. esculentum 35/Gy3 BU791059 hypothetical protein, 4e-08 NP192387 482 A. thaliana 47/Gy3 BU791062 2-oxoglutarate dehydro. 6e-34 NP201376 380 protein, A. thaliana 79/GY3 BU791064 developing caryopsis cDNA e-104 AL507226 571 clone, Hordeum vulgare 6e-87 AAB26162 510 87/Gy3 BU791067 ADPglucose pyrophosphorylase protein, E. coli 192/GY3 BU791079 tassel primordium cDNA clone, 0.12 BE123402 500 Zea mays 1e-07 TO7733 407 216/Gy3 BU791080 chaperonin 60 beta chain precursor protein, S. tuberosum 420/GY3 BU791087 E2 ubiquitin-conjugating 2e-81 NP566563 640 enzyme protein, A. thaliana a Accesion number of the cDNA clone, bA description of the best data match is given together with the database accession number of homologous genes d, cSequence similarity

Another interesting point to discuss is the presence of the COP9 signalosome protein, which probably has a role in the nucleoplasmic partitioning of ubiquitinated substrates [21]. It can also participate in light signalling. The Holm study [21] has found that the COP1 protein is investigated and the model proposed is that COP1 accumulates in the nucleus in the dark and interacts with light-dependent transcription factor, which is a positive regulator of photomorphogenesis. We have identified 138 B10 cDNA clone as a bZIP transcription factor from A. thaliana and also 260 B10 cDNA clone as a COP protein. Both of these cDNA clones are present in the male floral buds. Light has been shown to have an effect on sex determination in cucumber. Much of the light signalling appears to act directly on transcription factors. The positive signal for the photoreceptors received by downstream transcription factors are thus balanced by the ability of COP1 to target some of these transcription factors for degradation, allowing a very dynamic control of the transcriptional output [21].

428


Vol. 8. No. 2. 2003

Another clone, 95B10, shares a similarity with the brassinosteroid biosynthetic pathway. Brassinosteroids (BRs) are important hormones that regulate growth and development [22]. They take part in light and hormone signalling processes; interplay between these two process has been studied, but the mechanisms connecting these pathways are poorly understood [23]. Perhaps some BR proteins can act in sex differentiation between photoreceptor and signal transduction. Two cDNA clones, 122 B10 and 45B10, show a similarity to a putative glycineand proline-rich protein. The serine-proline rich proteins (male specific) were isolated by the Goldberg group [24]. These proteins are generally reported to be mostly cell wall proteins [25]. Many processes in photomorphogenesis, including cell expansion and differentiation, have been shown to be phytochrome regulated. It was suggested that the differential spatial accumulation of cell wall structural protein mRNAs is an important component in the process of photomorphogenesis [26]. We have also isolated a 47GY3 clone which displayed homology to component E1 2-oxoglutarate dehydrogenase from A. thaliana. The 2-oxoglutarate dehydrogenase complex (OGDC) occupies a central point in cellular metabolism within the tricarboxylic acid cycle (TCA). The TCA cycle in mitochondria catalyses the complete oxidation of organic acid to CO2 and the reduction of NAD(P) in the mitochondrial matrix. CO2 is known to stimulate ethylene production. Perhaps 2-oxoglutarate dehydrogenase in gynoecious plants is more highly expressed than in other types of flowers. There could be some link between ethylene and female sex occurrence [27, 28]. Similarly to chaperonin, ADP glucose pyrophosphorylase is also connected with heat stress. This enzyme might have a faster turnover rate under heat stress conditions compared with the other enzymes assayed. The 87GY3 cDNA clone shows a high similarity to ADP glucose pyrophosphorylase [29]. The clones described above may be involved in the same part of a light or hormone signalling cascade. Since light has a profound effect on cell growth and differentiation, it would be expected to be a major factor regulating cell wall structural protein gene expression. It is possible that five B10 cDNA clones (138B10, 260B10, 95B10, 122B10, 45B10) can be correlated with perception of light signals. The genes which may have chaperonin function (133B10, 157 B10, 216GY3) or E2 ubiquitin-conjunction enzymes (420Gy3) may take part in signalling or in response to stress factors including hormones and environmental factors. It is possible that they are somehow involved in flower differentiation in cucumber, as in Petunia [30] or Arabidopsis [31, 32]. The 2-oxoglutarate dehydrogenase (47GY3) and ADP glucose pyrophosphorylase (87 GY3) biochemical pathways can be connected to factors (hormones and environmental conditions) which influence sex in cucumber. The genes which have been putatively or hypothetically identified encode proteins with different cell functions, and they could also be specifically involved in flower development. Very little is know about the sequenced cDNAs


429

which share homology to cDNA from the EST database, even if it is a flower bud EST of A. thaliana, e.g. clone 166B10. SEGREGATION ANALYZIS It was possible that some of the isolated clones correspond to the G or M locus. To examine this possibility, two F2 segregating populations were used: 2gg x B10 and GY3 x HGY3 (Tab. 3). Tab. 3. Sex type segregation in two F2 populations

F2 population 2gg x B10

Types of plants monoecious gynoecious 156 47

GY3 x HGY3

gynoecious 79

hermaphrodite 25

Ratio

χ2

*P

3:1

0.0528 >0.5

3:1

0.3693 >0.5

*

P = χ2 probability with 1DF.

The F2 population (2gg x B10) segregated into two classes: plants producing only female flowers (gynoecious) and plants producing male and female flowers (monoecious) in a 1:3 ratio (Tab. 3). The monoecious groups display different types of occurrence of male and female flowers. To further investigate the groups which have the same typical sex pattern as the monoecious parental line, B10 was chosen. The F1 population displayed a monoecious phenotype. The F2 population (GY3 x HGY3) segregated into two classes: hermaphrodite (hermaphrodite) and female flowers (gynoecious) in a 1:3 ratio (Tab. 3). The F1 population were gynoecious. The results of the Southern analysis of the parental plants, F1 and F2 populations with 56 differentially expressed clones, showed no cDNA clone which cosegregated with sex in cucumber. This indicates that the analyzed clones are not products or good markers of sex determining genes in cucumber. The cDNA clones are differentially expressed between near isogenic lines in bisexual floral buds. They are most likely involved in an interactive mechanism in sex expression. There are a lot of reports about the isolation of transcripts that are correlated with sex determination [33, 34, 35, 36] in cucumber. Up till now, only the Trebitsh group [37] has found that CS-ACS1G is closely linked to the F locus and cosegregates with female sex. CS-ACS1G the differential screening strategy consisting of seeking gene products did not allow for the isolation of clones which are the sought genes connected with the M or G locus. An attempt to correlate locus M to ethylene receptor ERS was made by the Kahana group [38], but ERS segregates independently of M. The differential screening method which we chose was based on the manner of sex differentiation in cucumber. It allows the isolation of genes which are involved

430


Vol. 8. No. 2. 2003

in flower sex development but does not point to genes which are really sex determining genes. In situ RT-PCR analyzis Differential cDNA clones had been isolated and sequenced. As it turned out, they cannot be encoded by the genes determining sex in cucumbers. However, the occurrence of their products in differential flower libraries differing in sex indicates the possibility that they play a specific role in the development of specific types of flowers. To observe the expression patterns of isolated clones in developing flowers at different developmental stages, we performed in situ RTPCR. This method makes it possible to detect even single copies of mRNA in freshly frozen or paraffin embedded cells or tissues [39]. In our laboratory, the RT-PCR procedure in situ was optimized for flower buds [13]. Transcriptome analysis using in situ methods, does not only allows the determination of the presence, but also the distribution of a transcript. Five cDNA clones were used for the analysis: 35GY3, 47GY3, 216GY3, 100B10 and 157B10. Sequence amino acid comparison of the putative proteins encoded by these clones, and those from the database are presented in Fig. 1. Two of the clones (35GY3 and 100B10) seem to be interesting, as they are unknown. Three others may potentially affect flower differentiation, as discussed above. The sequence of clone 35GY3 shows homology to a hypothetical protein with an unknown function isolated from Arabidopsis thaliana. The distribution of the transcript of this clone appears to be very interesting. In 1-2mm flower buds of cucumber of the GY3 line (female line), a signal was observed in the layer of the ovule and in the perianth, as well as in the coloring of cell nuclei in the whole bud. In 3-4mm buds, a strong signal occurred in stamen primordia (Fig. 2A). Accumulation of the transcript was also noted in the placenta and petals. Moreover, stained cell nuclei were seen all over the bud. The localization of the signal in nuclei is not an artifact, as confirmed by the lack of a signal in the control where no primers were added (Fig. 2B) In buds over 4mm in size, a signal was observed in stamen primordia; a very strong signal was observed in the two external layers of the primordia cells and a slightly weaker one in the whole area of the stamen primordia. The signal in the stamen primordia was stronger in the 3-4mm buds. Staining of placenta cells, ovules and cell nuclei in the whole bud was also observed. The patterns of expression in the female buds of lines B10 (monoecious line) and 2gg (female line) were similar. No signals were found for buds of line HGY3 (hermaphroditic line). Similarly, no accumulation of transcript in the male buds was found. Signals of expression of clone 35GY3 were thus only observed in female cucumber flower buds, and were the strongest in a site where, as it seems, only the development of stamens should be inhibited. This suggests that the products of this gene could inhibit male elements in the female flower. However, this requires further investigation.


431

Fig. 1. Sequence amino acid comparison of putative cucumber proteins encoded by cDNA clones used in in situ RT-PCR analysis, and those from the database. In alignment with the display, the following symbols denote the degree of conservation observed in each column: "*" means that the residues or nucleotides in that column are identical in all the sequences in the alignment; ":" means that conserved substitutions were observed; "." means that semi-conserved substitutions were observed.

For clone 100B10, the signal was localized in flower buds of the GY3 line in corolla sepals, stamen primodia (3-4mm buds) and ovules (buds >4mm), and in line 2gg buds. In flower buds of the HGY3 line and in female buds of the B10 line, no signal of this clone was observed. In 3-4mm male buds, a transcript was observed in the pistil primordium and in >4mm, a stronger signal occurred in the whole of the pistil primordium (Fig. 3A).


432

Vol. 8. No. 2. 2003

cDNA of clone 100B10 was identified as a homologue of the expressed protein from A. thaliana, but homologous mRNA was also found in maturing pear fruit. So far, the function of this protein is not known. In these investigations, the pattern of its expression in male buds appears of interest because of the increase of the signal in inhibited pistil growth.

Fig. 2. In situ RT-PCR of 35GY3 clone analyzed in cross-sectioned 3-4 mm female buds of Cucumis sativus L. A. – reaction with specific 35GY3 primers; localization signals of transcripts in stamen primordium (sp), petal primordium (pp), petals (p) and cell nuclei (cn) in the whole bud; B. – control reaction without specific 35GY3 primers in the RTPCR mix.

cp

Fig. 3. In situ RT-PCR of 100B10 clone analyzed in cross-sectioned >-4 mm male buds of Cucumis sativus L. A. – reaction with specific 100B10 primers; localization signals of transcripts in the whole pistil primordium (cp); B. – control reaction without specific 100B10 primers in the RT-PCR mix.

Clones 216GY3 and 157B10 show homology to the chaperonin family. Clone 216GY3 is similar to the chaperonin 60 beta chain precursor from potato and garden pea chloroplasts. The pattern of weak expression for clone 216GY3 in the 1-2mm buds of the GY3 line is visible in stamen primordia, corolla and


433

sepals. In older buds, the signal of clone 216GY3 was located in stamen primordia, ovules and corolla sepals. In HGY3 buds, transcripts were distributed in the pistil, corolla sepals and anther sacs. In 1-2 mm male flower buds of the B10 line, a specific expression in the pistil primordium was noted (Fig. 4A). In 3-4mm, and >4mm buds, no signals were observed. By contrast, female flower lines B10 and 2gg showed expression in larger buds. The signals were observed in stamen primordia, ovules and corolla sepals. Thus, transcripts were observed rather in older female flowers. Moreover, strong signals were observed in hermaphroditic buds. The accumulation of large amounts of the transcripts of these genes in pistil primordia of the male flower, which will thus not develop, is interesting.

Fig. 4. In situ RT-PCR of 216GY3 clone analyzed in cross-sectioned 2 mm male buds of Cucumis sativus L. A. – reaction with specific 216GY3 primers; localization signals of transcripts in pistil primordium (cp); B. – control reaction without specific 216GY3 primers in the RT-PCR mix.

Clone 157B10, a putative protein binding a Rubisco subunit in buds of line GY3 and 2gg, did not give signals. In female buds of line B10, a characteristic fairly strong staining of the outermost layer of the cells of the oldest primordia was observed (Fig. 5A). Moreover, the closer it is to the placenta, the less the outer layer of cells of the ovules shows an intense accumulation of the transcript. A signal was also observed in stamen primordia and pistil styles (Fig. 5B). Chaperonins (clones 216GY3, 157B10) are a group of ubiquitous proteins which prevent the misfolding of new polypeptides whose synthesis has not yet been completed [40]. One of the first proteins identified as a chaperonin was the Rubisco subunit binding protein [41]; clone 157B10 shows homology to this protein. Chloroplast chaperonins (Cpn60α and Cpn60β), which show functional and structural similarity to the GroEL complex in E. coli, fold and form and assemble the prokaryotic Rubisco protein. They also play an important role in the tolerance of stress such as heat shock, probably by refolding proteins which have been damaged by stress [42]. The need for the appearance of specific proteins during the formation of a given sex may require the presence of specific proteins maintaining their structure. Hence the observed strong accumulation of

434


Vol. 8. No. 2. 2003

the clone 216GY3 transcript in this organ. On the other hand, for clone 157B10 strong mRNA accumulation was observed in the oldest ovules, a weaker one in younger ones and a lack of the transcript in the youngest ovules. Cpn60 is not only helpful in the formation of Rubisco, but it also assists in the folding of other proteins [43].

Fig. 5. In situ RT-PCR of 157B10 clone analyzed in cross-sectioned > 4 mm female buds of Cucumis sativus L. A. – reaction with specific 157B10 primers; localization signals of transcripts in pistil style (ps), stamen primordia (sp); B. – reaction with specific 157B10 primers; localization signals of transcripts in ovules (o).

This could suggest that chaperonins participate in the translocation and assembly of very different polypeptides [44]. Thus, they may also participate in the “cascade of sex expression”; on the other hand, research on the transgenesis of these genes indicates that the coding genes (at least chaperonin 60 beta from Arabidopsis) belong to the class of developmentally-regulated genes. When the construct Cpn promoter-GUS Cpn 60 β B3 was used in young flowers, expression was only limited to the calyx sepals, whereas in older flowers GUS activity was observed in stamen walls, the stigma and at the base of the pistil [45]. Experiments have shown that 60β transcripts are not only present in photosynthetically-active tissues, but in many other organs such as roots, stems and etiolated leaves. Cpn60 may thus act as general molecular chaperonins [46]. The expression of clone 47GY3 was not observed in hermaphroditic flowers (HGY3) or in male and female line B10. However, transcripts appeared in female buds (GY3) in corolla petals and stamen primordia (Fig. 6A). In female buds of the 2gg line, it is difficult to unequivocally determine the occurrence of clone 47GY3 mRNA. This clone shows similarity to the E1 component of 2-oxoglutarate dehydrogenase, which takes part in the tricarboxylic acid cycle [47] and catalyzed conversion of 2-oxoglutarate to succinyl-CoA and CO2. CO2 in turn is known to stimulate ethylene production. 2-oxoglutarate dehydrogenase may thus in some way participate in the formation of flowers of a defined type.


435

Fig. 6. In situ RT-PCR of 47GY3 clone analyzed in cross-sectioned 4 mm female buds of Cucumis sativus L. A. – reaction with specific 47GY3 primers; localization signals of transcripts in petals (p), stamens primordia (sp); B. – reaction without specific 47GY3 primers in RT-PCR mix.

The distribution of the transcripts of the analyzed genes appears interesting, especially that of 35GY3, as their functions ascertained earlier indicate that they could fulfill important roles during the formation of a defined type of flower. In this study, we reported on the isolation of cDNA clones which are differentially expressed between four lines differing in sex phenotype in cucumber. Sequencing of isolated gene products and construction of gene profiles expression could facilitate the understanding of sex and flower development in the cucumber plant. Isolation of full length sequences of cDNAs and a more detailed study of expression pattern would be required. However, the information that we have obtained in this study and the collection of differentially-expressed clones should be useful for further investigation of mystery of the mechanism of flower sex determination. Acknowledgements. We thank Dr. Tadeusz Rorat, D. Sc. (Institute of Plant Genetics, Polish Academy of Sciences, Poznań, Poland) for his helpful comments. This research was supported by the State Committee for Scientific Research. Research Project No. 5P06A00117. REFERENCES 1. DeLong, A., Calderon-Urrea, A. and Dellaporta, S.L. Sex determination gene TASSELSEED2 of maize encodes a short chain alcohol dehydrogenase required for stage-specific floral organ abortion. Cell 74 (1993) 757-768. 2. Trebitsh, T., Staub, J.E., O’Neil, S.D. Identification of a 1aminocycloprpane-1-carboxylic acid synthase gene linked to the female (F). Plant Physiol. 113 (1997) 987-995. 3. Kubicki, B. Investigation of sex determination in cucumber (Cucumis sativus L.). VII. Trimonoecism. Genet. Polon.10 (1969) 123-143.

436


Vol. 8. No. 2. 2003

4. Pierce, L.K. and Wehner, T.C. Review of genes and linkage group in cucumber. HortScience 25 (1990) 605-615. 5. Atsmon, D. and Galun, E. Morphogenesis study of staminate, pistillate and hermaphrodite flowers Cucumis sativus L. Phytomorphology 10 (1960) 110-115. 6. Nitsch, J.P., Kurtz E.B., Jr., Livermant, J. and Went, F.W. The development of sex expression in cucurbit flower. Am. J. Bot. 39 (1952) 32-43. 7. Pearson, K. On the criterion that a given system of deviation from the probable in the case of calculated system of variables is such that it can be reasonably supported to have arisen from random sampling, Philosophical Mag. J., Fifth series 50,London, Edinburgh, Dublin, pp 157-175. 8. Maens, M. and Messens, E. Phenol as grinding material in RNA preparation. Nucleic Acid Res. 20 (1992) 4374. 9. Rorat, T., Irzykowski, W. and Grygorowicz, W.J. Identification and expression of novel cold induced genes in potato (Solanum sogardianum). Plant Sci. 124 (1997) 69-78. 10. Sanger, F., Nicklen, S. and, Coulson, A.R. DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74 (1977) 5463-5467. 11. Michaels, S.D., John, M.C. and Amasino, R.M. Removal of polysaccharides from plant DNA by ethanol precipitation. Biotechnics 17 (1994) 274-276. 12. Sambrook, J., Fritsch, E.F. and Maniatis, T. Molecular cloning. A laboratory Manual. Cold Spring Harbour, NY: Cold Spring Harbour Laboratory Press 1989. 13. Urbańczyk-Wochniak, E., Filipecki, M. and Przybecki, Z. A useful protocol for in situ RT-PCR on plant tissues. Cell. Mol. Biol. Lett. 7 (1) (2002) 7-18. 14. Viitanen, P.V., Schmidt, M., Buchner, J., Suzuki, T., Vierling, E., Dickson, R., Lorimer, G.H., Gatenby A. and Soll, J. Functional characterization of the higher plant chloroplast chaperonins. J. Biol. Chem. 270 (1995) 1815818164. 15. Chen, X., Wang, B. and Wu, R. A gibberelin stimulated ubiquitin conjunction enzyme gene is involved in α-amylase gene expression in rice aleurone. Plant Mol. Biol. 29 (1995) 787-795. 16. Gray, W.M., del Pozo, C., Walker, L., Hobbie, L., Risseeuw, E., Banks, T., Crosby, W.L., Yang, M., Ma, H. and Estelle, M. Identification of an SCF ubiquitin-ligase complex required for auxin response in Arabidopsis thaliana. Genes Develop. 13 (1999) 1678-1691. 17. Gray, W.M. and Estelle, M. Function of the ubiquitin-proteasome pathway in auxin response.TIBS 25 (2000) 133-138. 18. Galun, E., Izhar, S. and Atsmon, D. Determination of relative auxin content in hermaphrodite and andromonoecious Cucumis sativus L. Plant Physiol. 40 (1972) 321-326. 19. Kepinski, S. and Leyser, O. Ubiquitination and auxin signalling: A degrading story. The Plant Cell (2002) 81-95. 20. Deshaies, R.J. SCF and cullin/Ring-H2-based ubiqutin ligase. Annu. Rev. Cell Dev. 15 (1999) 435-467.


437

21. Holm, M., Ma, L., Qu L. and Deng, X. Two interacting bZIP proteins are direct targets of COP1-mediated control of light-dependent gene expression. Genes Develop. 16 (2002) 1247-1259. 22. Noguchi, T., Fujioka, S., Choe, S., Takatsuto, S., Yoshida, S., Yuan, H. and Feldmann, K.A. Tax FE brassinosteroid-insensitive DWARF mutants of Arabidopsis accumulate brassinosteroids Plant Physiol. 121 (1999) 743752. 23. Neff, M.M., Nguyen, S.M., Malancharuvil, E.J., Fujioka, S., Noguchi, T., Seto Hideharu, S., Tsubuki, M., Honda, T., Takatsuto, S., Yoshida, S. and Chory, J. BAS1: A gene regulating brassinosteroid levels and light responsiveness in Arabidopsis. Proc. Natl. Acad. Sci. USA 96 (1999) 15316-15323. 24. Goldberg, A., Kahana, A., Silberstein, L. and Perl-Treves, R. Markers for cucumber male flower development isolated by differential display and differential hybridization, Conference – Quebec, 2000, S25 – 27. 25. Yasuda, E., Ebinuma, H. and Wabiko, H. A novel glycine-rich/hydrophobic 16kDA polypeptide genes from tobacco: similarity to proline-rich protein genes and its wound inducible and developmentally regulated expression. Plant Mol. Biol. 33 (1997) 667-678. 26. Sheng, J., Jeong, J. and Mehdy, M.C. Developmental regulation and hytochrome-mediated induction of mRNAs encoding proline-rich protein, glycine rich proteins and hydroxyproline rich proteins in Phaseolus vulgaris Proc. Natl. Acad. Sci. 90 (1993) 828-832 . 27. www.biochem.uwa.edu.au/AHM/MillarResL3.html 28. Millar, A.H., Hill, S.A. and Leaver, C.J. Plant mitochondrial dehydrogenase complex: Purification and characterization in potato. Biochem. J. 343 (1999) 327-334. 29. Greene, T.W. and Hannah, L.C., Enhanced stability of maize endosperm ADP-glucose pyrophosphorylase is gained through mutants that alter subunit interactions. Proc. Natl. Acad. Sci. USA 95 (1998) 13342-13347. 30. Van der Meer, I.M., Stam, M.E., van Tunen, A.J., Mol, J.N.M. and Stuitje, A.R. Antisense inhibition of flavonoid biosynthesis in Petunia anthers results in male sterility Plant Cell 4 (1992) 253-262. 31. Tsukaya, H., Takahashi, T., Naito, S. and Komeda, Y. Floral-organ specific and constitutive expression of an Arabidopsis thaliana heat shock HSP18,2: GUS fusion gene is retained even after homeotic conversion of flowers by mutation Mol. Gen. Genet. 237 (1993) 26-32. 32. Watts, F.Z., Butt, N., Layfield, P., Machuka, J., Burke, J.F. and Moore, A.L. Floral expression of a gene encoding an E2-related ubiquitin-conjugating protein from Arabidopsis thaliana Plant. Mol. Biol. 26 (1994) 445-451. 33. Kahana, A., Silberstein, L., Kessler, N., Goldstein, R. S. and Perl-Treves, R. Expression of ACC oxidase genes differs among sex genotypes and sex phases in cucumber. Plant Mol. Biol. 41 (1999) 517-528.

438


Vol. 8. No. 2. 2003

34. Yamasaki, S., Fujii, N. and Takahashi, H. The ethylene regulated sex expression of CS-ETR2 and CS-ERS genes in Cucumber Plants and their possible involvement with sex expression in flowers. Plant Cell Physiol. 41 (2000) 608-616. 35. Yamasaki, S., Fujii, N., Matsuura, S., Mizusawa, H. and Takahashi, H. The M locus and ethylene-controlled sex determination in Andromonoecious cucumber plants. Plant Cell Physiol. 42 (2001) 608-619. 36. Ando, S., Sato, Y., Kamachi, S. and Sakai, S. Isolation of a MADS-box gene (Eraf17) and correlation of its expression with the induction of formation of female flowers by ethylene in cucumber plants (Cucumis sativus L.) Planta 213 (2001) 943-952. 37. Trebitsh, T., Staub, J.E. and O’Neil, S.D. Identification of a 1aminocyclopropane-1-carboxylic acid synthase gene linked to the female (F). Plant Physiol. 113 (1997) 987-995. 38. Saraf-Leavy, T., Kahana, A., Kessler, N., Silberstein, L., Wang, Y., Gal-On, A. and Perl-Treves, R. Genes specifying synthesis and perception in cucumber sex-genotypes. Conference -Quebec 2000; S25-27. 39. Nuovo, G.J. In situ localization of PCR-amplified DNA and cDNA. Methods Mol. Biol. 123 (2000) 217-238 40. Hartl, F.U. Molecular chaperones in cellular protein folding. Nature 381 (1996) 571-580. 41. Bourraclough, R. and Ellis, R.J. Assembly of newly synthesized large subunits into ribulose biphosphate carboxylase in isolated pea chloroplasts. Biochem. Biophys. Acta 608 (1980) 19-31. 42. Gatenby, A.A. Synthesis and assembly of bacterial and higher plant Rubisco subunits in Escherichia coli. Photosynth. Res. 17 (1988) 145-157. 43. Lubben, T.H., Donaldson, G.K., Vittanen, P.V. and Gatenby, A.A. Several proteins imported into chloroplasts from stable complexes with the Gro-Elrelated chloroplast molecular chaperone. Plant Cell 1 (1989) 1223-1230. 44. Ellis, J. van der Vies, S. Molecular chaperones. Annu. Rev. Biochem 60 (1991) 321-347. 45. Zabaleta, E., Oropeza, A., Jimenez, B., Salerno, G, Crespi, M. and HerreraEstrella, L. Isolation and characterization of genes encoding chaperonin 60 β from Arabidopsis thaliana. Gene 111 (1992) 175-181. 46. Zabaleta, E., Assad, N, Oropeza, A., Salerno, G. and Herrera-Estrella, L. Expression of one of the members of the Arabidopsis chaperonin 60 β gene family is developmentally regulated and wound-repressible. Plant Mol. Biol. 24 (1994) 195-202. 47. Millar, A.H., Hill, S.A. and Leaver, C.J. Plant mitochondrial 2-oxoglutarate dehydrogenase complex: Purification and characterization in potato. Biochem. J. 343 (1999) 327-334.