Fermentative metabolism in grape berries: isolation and
characterization of pyruvate decarboxylase cDNA and analysis of
its expression throughout berry development
Etti Or *, Jenny Baybik, Avi Sadka, Aliza Ogrodovitch
Department of Tree Breeding,Institute of Horticulture,Agricultural Research Organization,The Volcani Center,POB6,
Bet Dagan50-250,Israel
Received 7 January 2000; received in revised form 29 February 2000; accepted 1 March 2000
Abstract
The involvement of pyruvate decarboxylase (PDC) in the control of alcohol production during ripening of fruit tissues under aerobic conditions has been only partially studied. Enzymological studies showed a significant increase in PDC activity during the ripening of oranges and pears, concurrently with the induction of ethanol production. In tomato, on the other hand, the induction of ethanol production and ADH gene expression after the onset of ripening was not accompanied by induction of PDC activity. The isolation of PDC cDNA from fruits has not yet been reported, nor has its expression pattern during fruit development. We report here the isolation of a cDNA clone encoding for a grape PDC and the characterization of its expression throughout berry development. The pattern of PDC gene expression throughout berry development, combined with earlier findings on constitutive PDC activity in the berry, may suggest that PDC is not the limiting factor for the production of ethanol in the berry, which is induced only after the onset of berry ripening. Alternatively, the induction of ADH gene expression, which occurs only after the onset of ripening in both tomatoes and grape berries, may serve as a regulator of ethanol production in response to a ripening-related cue. © 2000 Elsevier Science Ireland Ltd. All rights reserved.
Keywords:Fermentation; Pyruvate decarboxylase;Vitis6inifera
www.elsevier.com/locate/plantsci
1. Introduction
Under normal oxygen tension both alcohol pro-duction and alcohol dehydrogenase (ADH) mRNA levels are low in most plant tissues and they are induced only under anaerobic conditions [1]. However, ripening fruit tissues are an excep-tion. The expression of ADH mRNA, as well as the ethanol level, were significantly increased in fruits such as tomato, pear and grapes after the onset of ripening under normal oxygen tension [2 – 4].
In grape berries, and in other fruits accumulat-ing malic acid, the induction of ethanol produc-tion during ripening may be associated with increased levels of pyruvate within the fruit. The level of pyruvate increases as a consequence of the induction of malate decarboxylation by malic en-zyme at the onset of fruit ripening [5].
In the process of ethanol synthesis from pyru-vate, the conversion of pyruvate to acetaldehyde by pyruvate decarboxylase (PDC) precedes the reduction of acetaldehyde to ethanol by ADH [1]. Pyruvate is located at the branching point be-tween two alternative energy producing processes: its conversion to acetyl CoA by pyruvate dehydro-genase leads to respiration via the TCA cycle, whereas its conversion to acetaldehyde by
Accession number: AF195868
* Corresponding author. Fax: +972-3-9669583.
E-mail address:[email protected] (E. Or).
vate decarboxylase leads to fermentation. There-fore, PDC is considered to be a key enzyme in switching to fermentative metabolism and has been suggested to be the rate-limiting factor in ethanol production [6,7]. Accordingly, the expres-sion of the plant PDC genes that have been stud-ied so far is induced under anaerobic conditions, similarly to the case of ADH [6,7].
The involvement of PDC in the control of alco-hol production during ripening of fruit tissues has been only partially studied. Enzymological studies have shown a significant increase in PDC activity during the ripening of oranges and pears, in corre-lation with increased ethanol production [2,8]. However, neither the isolation of PDC cDNA from fruits, nor its pattern of expression during fruit development has been reported previously. We report here the isolation of a cDNA clone encoding for a grape PDC, and the characteriza-tion of its temporal expression during berry development.
2. Materials and methods
2.1. Plant material
The experiments were conducted with berries (Vitis 6inifera cv. Perlette grafted on 140-R root
stock) from a vineyard in the experimental station of the Volcani center, Bet Dagan, located in the central part of the coastal plain of Israel. Every week ten clusters were sampled, from berry set to ripening, during the summer of 1998. The berries were pooled; part of each sample was frozen in liquid nitrogen and stored at −80°C and part was used for monitoring the course of berry develop-ment at each sampling day. A total of 15 samples of five berries were taken for berry weight and diameter determination and three groups of 50 berries were used for the determination of total soluble solids (TSS) (%) and total titratable acidity (TA) (%), according to Ref. [9].
2.2. Extraction of RNA and DNA
Total RNA was extracted from the grape berries according to Ref. [10]. Genomic DNA was extracted from grape leaves according to Ref. [11]. Plasmid DNA was extracted by the alkaline lysis method, according to Ref. [12].
2.3. Screening of a bud cDNA library
A cDNA library was constructed from grape berries at various stages of development by means of thelZAPII cDNA library kit (Stratagene). The library was screened with 32P-labelled Tobacco
PDC2 cDNA clone [1] kindly provided by Profes-sor C. Kuhlemeier from the University of Bern, Switzerland. Screening was according to the man-ufacturer’s instructions. Following purification of the hybridizing plaques, a pBluescript plasmid containing the positive insert was excised from the recombinant phage according to the library kit instructions.
2.4. DNA sequencing
Both strands of GPDC were sequenced by the dideoxy chain termination method. Custom primers (BRL) were used in order to sequence both strands in their entirety. Sequence data were analyzed with the Genetics Computer Group (GCG) sequence analysis program (version 9.0).
2.5. Northern and Southern analyses
For Northern analysis, total RNA from grape berries was denaturated, fractionated by formalde-hyde/agarose gel electrophoresis, and transferred to a nylon membrane (Hybond N+, Amersham) according to Ref. [12]. Hybridization was per-formed with a 32P-labelled probe and a
comm-ercial hybridization buffer (Amersham). Hy-bridization and washing conditions were according to the buffer manufacturer’s instruc-tions.
For Southern analysis, grape genomic DNA was digested with various restriction endonucleases (which have no site within the GPDC1 cDNA), fractionated by agarose gel electrophoresis and transferred to a nylon membrane (Hybond N+, Amersham). Hybridization was performed with a
32P-labelled probe. The 265-bp probe was
am-plified by PCR from an intronless region in the GPDC1 gene, by means of two custom primers: the forward primer was 5%
-CCATTCTGCTTG-GCGCCCAAAGTG -3%, located between the nu-cleotides 581 and 604 in the GPDC1 cDNA, and the reverse primer was 5%
nucle-otides 846 and 869 in the GPDC1 cDNA. Hy-bridization buffer containing 50% formamide was prepared according to Ref. [12]. The hybridization
temperature was 42°C. The membrane was washed twice with 2× SSC/0.1% SDS for 20 min at room temperature, followed by a 15-min wash with 1×
SSC/0.1% SDS and a 15-min wash with 0.2×
SSC/0.1% SDS, both at 65°C.
3. Results and discussion
3.1. GPDC1 cDNA cloning and sequence analysis
A cDNA library, constructed with poly(A) mRNA from grape berries at various mixed devel-oping stages, was screened with a partial PDC cDNA from tobacco [1] and a 1918-bp cDNA clone was isolated (Fig. 1). According to se-quence analysis, this cDNA, designated GPDC1, contains a 1729-bp open reading frame, which encodes 575 amino acids, followed by a 189-bp 3%-UTR.
3.2. Comparison of GPDC1 with other plant
pyru6ate decarboxylases
Comparison of the deduced amino acid se-quence for GPDC1 with the sese-quence data bases (GenBank, EMBL and Swissprot) was carried out by means of the BLAST function of GCG soft-ware. This comparison revealed that the putative protein encoded by this cDNA was most closely related to PDC-1 from tobacco [1]. According to this comparison, the grape clone lacks the 5% end of the open reading frame, but it is predicted to be near full length (Fig. 2).
Further comparisons were conducted between the amino acid sequence of GPDC1 and those of a few plant PDC proteins from different classes, by mean of the pile-up function of GCG (Fig. 2). GPDC1 shows 87% identity to the tobacco PDC-1 [1], 86% identity to an Arabidopsis PDC-1 (acces-sion number u71121), 81% identity to PDC-2 from rice [13], 78% identity to PDC-3 from rice [14] and 77% identity to PDC-1 from rice [6].
The amino acids Glu191, Leu310, Leu337, Ala344,
Lys350, His360, Pro369, Leu370,Gln385, Ser391, Lys412
and Lys415are conserved among the GPDC1 gene
and the other PDC1 genes that were compared, although there is no evidence yet for functional differences among the various PDC genes, except for the PDC-2 gene from tobacco that is expressed only in the pollen [1].
Fig. 2. Pile-up alignment of the amino acid sequence of GPDC1 with those of other plant PDC genes. Gaps have been introduced for maximum alignment. Asterisks indicate the stop codon. Arrows above the sequence indicate invariant residues characteristic of PDC-1 genes. Residues identical to GPDC1 are shaded. Sequences are from GPDC1 [this work]; tobacco PDC-1 [1]; arabidopsis PDC-1 (accession number u71121); rice PDC-1 [6]; rice PDC-3 [14]; and rice PDC-2 [13].
3.3. Analysis of GPDC copy number in the Vitis
genome
Southern blot analysis with genomic DNA from
Vitis 6inifera cv. Perlette, digested with various
restriction enzymes, was carried out to determine
hybridiz-Fig. 2. (Continued)
ing bands in both lanes. While the more intense band probably represents GPDC1, the weakly hy-bridizing band is presumably a less closely related copy of the gene. These data suggest that there are two genes that code for PDC in the grape genome, similar to tobacco [1]. There is a theoretical possi-bility that under less stringent hybridization condi-tions more than two copies of grape PDC may be detected. In rice [15] and corn [16], for example, there are three copies of the PDC gene in the genome. However, the high identity among PDC sequences from different organisms, which sug-gests even higher identity within the same genome, throws doubt on the possibility that there are extra PDC copies within the grape genome that were not detected at all under the present hy-bridization conditions.
3.4. Expression pattern of GPDC during the
de6elopment of the berry
The expression of the PDC mRNA was investi-gated throughout berry development, from early immaturity to ripening. To select for berries in
representative developmental stages we first moni-tored the course of berry development and recorded changes in berry diameter, weight, total acidity and TSS during the growing season. Berry growth showed the double sigmoidal curve (data not shown), typical of grape development [17]. Acids accumulated until varaison, which occurred on June 3, followed by a decrease in acidity until maturity (Fig. 4A). Sugar started to accumulate at the onset of ripening (Fig. 4B). Based on these changes throughout development, berries were
se-Fig. 3. Southern blot analysis of genomic DNA from Vitis
6iniferacv. Perlette. Genomic DNA (10mg) was digested with
Fig. 4. Changes in titratable acid and soluble solids through-out cv. Perlette berry development. Every week ten clusters were sampled, from berry set to ripening, in the summer of 1998. At each sampling date the berries were pooled and subsamples of 50 berries were used for the determination of TSS (%) and total titratable acidity (%). Values are the average of three replications. Bars above the columns repre-sent the standard deviation.
under aerobic conditions showed that while the PDC transcript was present in the berry through-out its development, the expression of the ADH transcript was induced mainly after the onset of ripening, as shown in Fig. 5B and reported by Ref. [4].
While our data reflects the total steady state PDC RNA content, a theoretical possibility exists that both PDC genes are expressed in the berry, and that their individual patterns of expression are different from the sum of both. Yet, the presence of PDC transcript and activity in the berry throughout its development implies that
acetalde-Fig. 5. Expression pattern of GPDC and G7ADH in the developing berry. Northern blot analysis was conducted with the total RNAs (17.5 mg) from berries in several stages of development. Berries were selected from five time points, based on the changes in total acidity (TA), to represent the development process: (1) immature (6 May, 1.8% TA); (2) pre-varaison (20 May, 3.2% TA); (3) varaison (3 June, 3.7% TA); (4) post-varaison (17 June, 1.2% TA); (5) mature (1 July, 0.7% TA). RNA blots were probed with radiolabelled GPDC cDNA clone (1918 bp) and G7ADH cDNA clone (accession no. AF195867). The autoradiograms and the EtBr-stained gels were scanned, and quantitative analysis was carried out with the TINA software. Normalized values are presented. lected from five time points (immature,
pre-varaison, pre-varaison, post-varaison and mature) and analyzed for their PDC mRNA levels. According to this analysis PDC transcript exists in the fruit throughout its development (Fig. 5A). However, it can be recognized that towards the onset of ripen-ing (stage 3) there is some increase in the PDC transcript level, followed by a decrease from the post-varaison stage (stage 4) onwards.
The PDC expression pattern is well correlated with the PDC activity pattern, which was shown to be fairly constant throughout berry develop-ment, except for a short period of increased activ-ity after the onset of ripening [18].
hyde, the precursor for ADH activity, is constitu-tively available, independent of which PDC gene is responsible for it in each developmental stage. Consequently, the PDC activity data [18] and gene expression data (Fig. 5) may suggest that PDC is not the rate-limiting enzyme that switches on the metabolic machinery to produce ethanol in the grape berry during ripening, although it is most likely involved in the enhancement of ethanol pro-duction in the berries, as expected.
Likewise, the induction of ethanol production and ADH gene expression after the onset of tomato ripening was not accompanied by an in-duction of PDC activity [19]. However, under anaerobic condition both ADH and PDC activi-ties were induced in tomato. Therefore, it was suggested that the stimuli of ripening and anaero-biosis seem to act independently on the control of ethanol production in tomato [19].
In oranges and pears, on the other hand, ethanol production during ripening under aerobic conditions was accompanied by increases in both ADH and PDC activities [2,8,20], resembling the response of most plant tissues following exposure to anaerobiosis [6,7].
These data may lead to the assumption that there is more than one way to control ethanol production in fruits. In some fruits, such as pears and oranges, ethanol production may be regulated in a way resembling the mechanism reported for tissues under anaerobic conditions. According to this mechanism PDC availability may limit the ethanol production and the induction of PDC expression may be the trigger for the onset of fermentation, together with the induction of ADH expression or ahead of it.
In these fruits the possibility exists that ethanol production could result from reduced internal oxy-gen concentration, because of the increasing vol-ume of the fruit or the ripening climacteric. In this case it is reasonable to suppose that the control mechanism would be identical to that responding to the cues of anaerobiosis.
In some other fruits, such as tomato and, pre-sumably, also grape, there might be a different mode of regulation, controlled by the ripening process rather than by anaerobiosis cues. In these fruits, in which the PDC is more or less constitu-tively available, the trigger for ethanol production may be the induction of ADH expression. The cues which leads to its induction are as yet unknown.
Acknowledgements
The authors would like to thank Professor C. Kuhlemeier from the University of Bern in Switzerland, for the tobacco PDC cDNA probe. This work was supported by a grant from the Ministry of Science of Israel. This is a contribu-tion from the Agricultural Research Organizacontribu-tion, Volcani Center, Bet Dagan, Israel, No. 201/00, 2000 Series.
References
[1] M. Bucher, K.A. Brander, S. Sbicego, T. Mandel, C. Kuhlemeier, Aerobic fermentation in tobacco pollen, Plant Mol. Biol. 28 (1995) 739 – 750.
[2] C. Chervin, J.K. Truett, Alcohol dehydrogenase expres-sion and alcohol production during pear ripening, J. Am. Soc. Hort. Sci. 124 (1999) 71 – 75.
[3] T. Longhurst, E. Lee, R. Hinde, C. Brady, J. Speirs, Structure of the tomato ADH2 gene and Adh2 pseudo-genes, and a study of ADH2 gene expression in fruit, Plant Mol. Biol. 26 (1994) 1073 – 1084.
[4] P. Sarni-Manchado, C. Verries, C. Tesniere, Molecular characterization and structural analysis of one alcohol dehydrogenase gene (GV-Adh 1) expressed during ripen-ing of grapevine (Vitis6iniferaL.) berry, Plant Sci. 125 (1997) 177 – 187.
[5] H.P. Ruffner, Metabolism of tartaric and malic acids in
Vitis: a review — part B, Vitis 21 (1982) 346 – 358.
[6] M.A. Hossain, E. Hug, T.K. Hodges, Sequence of a cDNA fromOryza sati6a(L.) encoding pyruvate decar-boxylase 1 gene, Plant Physiol. 106 (1994) 799 – 800. [7] P. Lu, B.P. Davis, T.W. Jeffries, Cloning and
characteri-zation of two pyruvate decarboxylase genes fromPichia
stipitis CBS 6054, Appl. Environ. Microbiol. 64 (1998)
94 – 97.
[8] J.H. Bruemmer, B. Roe, Pyruvate dehydrogenase activity during ripening of ‘Hamlin’ oranges, Phytochemistry 24 (1985) 2105 – 2106.
[9] A. Sadka, E. Dahan, L. Cohen, K.B. Marsh, Aconitase activity and expression during the development of lemon fruit, Physiol. Plant. (in press).
[10] S. Chang, J. Puryear, J. Cairney, A simple and efficient method for isolating RNA from pine trees, Plant Mol. Biol. Repo. 11 (1993) 113 – 116.
[11] J. Steenkamp, I. Wild, A. Lourens, P. Van Helde, Im-proved method for DNA extraction fromVitis6inifera, Am. J. Enol. Vitic. 45 (1994) 102 – 106.
[12] J. Sambrook, E.F. Fritsch, T. Maniatis, A Laboratory Manual for Molecular Cloning, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989. [13] E. Hug, M.A. Hossain, T.K. Hodges, Cloning and
se-quencing of a cDNA encoding pyruvate decarboxylase 2 gene from rice, Plant Physiol. 109 (1995) 722.
rice genomic pyruvate decarboxylase gene that lacks introns: a pseudo gene?, Plant Physiol. 106 (1994) 1697 – 1698.
[15] M.A. Hossain, E. Hug, A. Grover, A.S. Dennis, W.J. Peacock, T.K. Hodges, Characterization of pyruvate decarboxylase genes from rice, Plant Mol. Biol. 31 (1996) 761 – 770.
[16] V.M. Peschke, M.M. Sachs, Multiple pyruvate decarboxy-lase genes in maize are induced by hypoxia, Mol. Gen. Genet. 240 (1993) 206 – 212.
[17] B.J. Coomb, Research on developing and ripening of the grape berry, Am. J. Enol. Vitic. 43 (1992) (1992) 101 – 110.
[18] J.S. Hawker, Changes in the activities of malic enzyme, malate dehydrogenase, phosphopyruvate carboxylase and pyruvate decarboxylase during the development of a non-climacteric fruit (the grape), Phytochemistry 8 (1969) 19 – 23.
[19] A.S. Chen, T. Chase, Alcohol dehydrogenase 2 and pyruvate decarboxylase induction in ripening and hypoxic tomato fruit, Plant Physiol. 31 (1993) 875 – 885.
[20] B. Roe, P.L. Davis, J. Bruemmer, Pyruvate metabolism of ‘Hamlin’ oranges, Phytochemistry 23 (1984) 713 – 717.
