Abstract
β-Galactosidases (EC 3.2.1.23) constitute a widespread family of enzymes characterized by their ability to hydrolyze terminal, nonreducing β-d-galactosyl residues from β-d-galactosides. Several β-galactosidases, sometimes referred to as exo-galactanases, have been purified from plants and shown to possess in vitro activity against extracted cell wall material via the release of galactose from wall polymers containing β(1→4)-d-galactan. Although β-galactosidase II, a protein present in tomato (Lycopersicon esculentum Mill.) fruit during ripening and capable of degrading tomato fruit galactan, has been purified, cloning of the corresponding gene has been elusive. We report here the cloning of a cDNA, pTomβgal 4 (accession no. AF020390), corresponding to β-galactosidase II, and show that its corresponding gene is expressed during fruit ripening. Northern-blot analysis revealed that the β-galactosidase II gene transcript was detectable at the breaker stage of ripeness, maximum at the turning stage, and present at decreasing levels during the later stages of normal tomato fruit ripening. At the turning stage of ripeness, the transcript was present in all fruit tissues and was highest in the outermost tissues (including the peel). Confirmation that pTomβgal 4 codes for β-galactosidase II was derived from matching protein and deduced amino acid sequences. Furthermore, analysis of the deduced amino acid sequence of pTomβgal 4 suggested a high probability for secretion based on the presence of a hydrophobic leader sequence, a leader-sequence cleavage site, and three possible N-glycosylation sites. The predicted molecular mass and isoelectric point of the pTomβgal 4-encoded mature protein were similar to those reported for the purified β-galactosidase II protein from tomato fruit.
The most conspicuous and important processes related to postharvest quality of climacteric fruit are the changes in texture, color, taste, and aroma that occur during ripening. Because of the critical relationship that deleterious changes in texture have to quality and postharvest shelf life, emphasis has been placed on studying the mechanisms involved in the loss of firmness that occurs during tomato (Lycopersicon esculentum) fruit ripening. Although fruit softening may involve changes in turgor pressure, anatomical characteristics, and cell wall integrity, it is generally assumed that cell wall disassembly leading to a loss of wall integrity is a critical feature. The most apparent changes, in terms of composition and size, occur in the pectic fraction of the cell wall (for refs., see Seymour and Gross, 1996), which include increased solubility, depolymerization, deesterification, and a significant net loss of neutral, sugar-containing side chains (Huber, 1983; Fischer and Bennett, 1991; Seymour and Gross, 1996).
The best-characterized pectin-modifying enzymes are PG (EC 3.2.1.15) and PME (EC 3.1.1.11). Although PG and PME are relatively abundant and have substantial activity during tomato fruit ripening, softening still occurs, albeit with a slight delay, in fruit of transgenic plants in which PG (Smith et al., 1988) or PME (Tieman et al., 1992; Hall et al., 1993) gene expression and enzyme activity were significantly down-regulated. Moreover, overexpression of PG in nonripening mutant rin tomato fruit did not result in softening, even though depolymerization and solubilization of pectin was evident (Giovannoni et al., 1989).
Among the other known pectin modifications that occur during fruit development, one of the best characterized is the significant net loss of galactosyl residues that occurs in the cell walls of many ripening fruit (Gross and Sams, 1984; Kim et al., 1991; Seymour and Gross, 1996). Although some loss of galactosyl residues could result indirectly from the action of PG, β-galactosidase (exo-β[1→4]-d-galactopyranosidase; EC 3.2.1.23) is the only enzyme identified in higher plants capable of directly cleaving β(1→4)galactan bonds, and probably plays a role in galactan side chain loss (DeVeau et al., 1993; Carey et al., 1995; Carrington and Pressey, 1996). No endo-acting galactanase has yet been identified in higher plants.
The view that β-galactosidase is active in releasing galactosyl residues from the cell wall during ripening is supported by the dramatic increase in free Gal, a product of β-galactosidase activity (Gross, 1984), and a concomitant increase in β-galactosidase II activity in tomatoes during ripening (Carey et al., 1995). β-Galactosidases are generally assayed using artificial substrates such as p-nitrophenyl-β-d-galactopyranoside, 4-methylumbelliferyl-β-d-galactopyranoside, and X-Gal. However, it is clear that β-galactosidase II is also active against natural substrates such as β(1→4)galactan (Pressey, 1983; Carey et al., 1995; Carrington and Pressey, 1996). β-Galactosidase proteins have been purified and characterized in a number of other fruits including kiwifruit (Actinidia deliciosa; Ross et al., 1993), coffee (Coffea arabica; Golden et al., 1993), persimmon (Diospyros kaki; Kang et al., 1994), and apple (Malus domestica; Ross et al., 1994).
Carey et al. (1995) were able to purify one of the three previously identified β-galactosidases from ripening tomato fruit (Pressey, 1983), but only one (β-galactosidase II) was active against β(1→4)galactan. Even though they were able to identify putative β-galactosidase cDNA clones, none of the deduced amino acid sequences of the cDNAs matched the N-terminal sequence of the β-galactosidase II protein. Here we describe the cloning of a cDNA (pTomβgal 4) that apparently codes for β-galactosidase II. We also show that the gene corresponding to pTomβgal 4 is expressed in wild-type fruit during ripening and exhibits the expression pattern expected for β-galactosidase II in both wild-type and mutant fruit.
MATERIALS AND METHODS
Tomato (Lycopersicon esculentum Mill. cv Rutgers) plants were grown in a greenhouse using standard cultural practices. The ripening mutants ripening inhibitor (rin), nonripening (nor), and never-ripe (Nr) (Tigchelaar et al., 1978) were all in the cv Rutgers background. Flowers were tagged at anthesis and fruit were harvested according to the number of DPA or based on their surface color using ripeness stages, as previously described (Mitcham et al., 1989). For gene-expression studies, a variety of leaf, flower, and stem tissues were harvested from greenhouse-grown plants and roots were harvested from seedlings grown in basal tissue culture medium for 4 weeks after seed germination.
RNA Extraction
Fruits were processed immediately after harvest in the greenhouse by chilling on ice, excising the various tissues, and freezing them in liquid nitrogen. Tissue samples were ground using a mortar and pestle and stored at −80°C. RNA was extracted using the method described by Verwoerd et al. (1989). Poly(A+) RNA was purified from total RNA using oligo(dT) columns (Pharmacia2). RNA was quantified by measuring A260 using a dual-beam spectrophotometer.
RT-PCR
Degenerate primers were designed based on the highest shared deduced amino acid sequence identity we found between apple (Malus domestica; accession no. P48981), asparagus (Asparagus officinalis; accession no. P45582), and carnation (Dianthus caryophyllus; accession no. Q00662) β-galactosidase cDNA clones. The two primers used for the first reaction were BG5′E1 (WSNGGNWSNATHCAYTAYCC) and BG3′E (CCRTAYTCRTCNADNGGNGC). A second reaction was done on the products of the first reaction using BG5′I1 (ATHCARACNTAYGTNTTYTGG) and BG3′E. The degeneracy code for the primer sequences is N = a + t + c + g; H = a + t + c; B = t + c + g; D = a + t + g; V = a + c + g; R = a + g; Y = c + t; M = a + c; K = t + g; S = c + g; and W = a + t. The 5′ and 3′ primers corresponded to amino acids 72 to 78 and 321 to 315, respectively, of the apple clone. Amplification was with DNA polymerase (AmpliTaq, Perkin-Elmer) and standard PCR conditions using the cDNA made for the first cDNA library described below as a template (Ausubel et al., 1987). PCR products were separated in an agarose gel and fragments of the expected size (approximately 750 bp) were purified, cloned into pCRscript (Stratagene), and sequenced.
cDNA Library
Two cDNA libraries were constructed. The first comprised poly(A+) RNA isolated from breaker, turning, and pink fruit pericarp from cv Rutgers plants. The cDNA synthesis and library construction were done exactly according to the manufacturer's instructions for the ZAP-cDNA Gigapak II Gold cloning kit (Stratagene). First-strand cDNA synthesis was primed using a poly(dT) primer and inserts were directionally cloned into the Uni-Zap XR vector using EcoRI and XhoI restriction sites. The second library comprised poly(A+) RNA isolated from all fruit tissues (except seeds) from immature green, mature green, breaker, turning, pink, red-ripe, and overripe fruit of cv Rutgers plants. The cDNA synthesis and library construction was done exactly according to the manufacturer's instructions for the SuperScript Lambda System for cDNA synthesis and λ cloning (GIBCO-BRL). First-strand cDNA synthesis was primed using an oligo(dT) primer and cDNA inserts were directionally cloned into the λZipLox cloning vector using SalI and NotI restriction sites. Both libraries were amplified and maintained using the host strains provided by the manufacturers according to their instructions.
DNA and RNA Gel-Blot Analysis
Total RNA (20 μg/lane) was separated in a formaldehyde/Mops agarose gel, transferred to a Hybond-N+ nylon membrane (Amersham), fixed by incubating for 2 h at 80°C, hybridized overnight in a hybridization incubator (Robbins Scientific, Sunnyvale, CA) using a buffer described by Church and Gilbert (1984), washed to a final stringency of 0.1× SSC with 0.2% SDS at 65°C, and autoradiographed essentially as described by Ausubel et al. (1987). An RNA ladder standard (GIBCO-BRL) was used to estimate the lengths of the RNAs. Probes were synthesized using a random-priming kit with [32P]dATP as the label (Boehringer Mannheim). pTomβgal 4-specific probe was synthesized using an approximately 1-kb BamHI-XhoI fragment from the 3′ end of the cDNA insert (for northern-blot analysis) or a 0.5-kb fragment from the 3′-most end of the cDNA generated by PCR using the gene-specific primer 4–5F (GGTACAAGGCTACATTTAAC) and vector-specific primer T7 (for Southern-blot analysis). As a loading control, RNA blots were stripped and reprobed at a reduced hybridization and washing stringency using a soybean 26S rDNA fragment (Turano et al., 1997). For all hybridizations, [32P]dATP-labeled probe was diluted to 1 × 106 to 2 × 106 dpm/mL. DNA gel-blot analysis was done essentially as described by Smith and Fedoroff (1995), except that 3 μg of genomic DNA was used for each digest.
Sequence Analysis
Sequencing was done at the Iowa State University Sequencing Facility (Ames) using a PCR-based dideoxynucleotide terminator protocol and an automated sequencer (Applied Biosystems). The sequencing of both cDNA-insert strands was done by primer walking. Nucleotide and deduced amino acid sequence comparisons against the databases were done using BLAST searches (Altschul et al., 1990). Sequence data were analyzed and aligned using DNA Strider 1.2 (Marck, 1988) and MacDNAsis (Hitachi, San Bruno, CA) software.
Expression in Escherichia coli and β-Galactosidase Activity
The ORF of pTomβgal 4 was PCR amplified using oligonucleotides so that the signal peptide (amino acids 1–23) was removed and a BglII and an EcoRI restriction site was created at the 5′ and the 3′ end of the ORF, respectively. The 2.183-kb fragment was cloned into a BglII-plus-EcoRI-digested pFLAG-CTC vector (Kodak). The vector was transformed into the E. coli strain XL1-Blue MR (lacZ) (Stratagene). As a positive control for maximal β-galactosidase activity the vector pGEM (containing the E. coli lacZ gene fragment, Promega) was transformed into the strain DH5α (containing the lacIqZΔM15 cassette). Cultures were grown overnight to saturation in Luria-Bertani medium containing 0.4% Glc and 100 μg/mL ampicillin at 37°C.
The cultures were diluted 1:100 in Luria-Bertani medium containing 0.5 mm IPTG and 0.025% X-Gal and were grown in 250-mL Erlenmeyer flasks at 20°C with shaking at 150 rpm. β-Galactosidase activity was estimated by harvesting 1 mL of cells every 12 h. The cells were spun at full speed in a microcentrifuge (14,000 rpm) for 1 min, resuspended in 1 mL of water, and lysed by the addition of 50 μL of chloroform and vortexing. The lysate was spun at full speed for 2 min in a microcentrifuge, the supernatant was removed, 1 mL of dimethylformamide was added, and the tube was vortexed and sonicated for 10 s to solubilize any blue precipitate resulting from the cleavage of X-Gal. The tubes were again centrifuged at full speed for 5 min, and 750 μL of the supernatant was used to measure the A615.
RESULTS
Cloning
Degenerate primers were designed based on the most highly conserved regions of shared amino acid identity among three plant β-galactosidase cDNAs from apple, asparagus, and carnation. These primers were used in a RT-PCR protocol, and a single product of approximately 750 bp was amplified (see Methods). Several PCR-product clones were sequenced and three unique, putative tomato β-galactosidase clones were identified. One of the clones (RT-PCR2-1) was used to screen 106 plaques from a tomato fruit cDNA library at low stringency. Thirty positive cDNA clones were identified and partially sequenced. Complete sequencing and characterization of the RT-PCR and cDNA clones revealed the possibility of seven unique β-galactosidase genes (data not shown).
Sequence Characterization
One of the seven unique clone sequences was nearly identical, except for the 5′-untranslated region, to the previously published sequence of the tomato β-galactosidase cDNA clone pTomβgal 1 isolated from ripe cv Ailsa Craig fruit (Carey et al., 1995). The matching cDNA clone was named pTomβgal 10 (accession no. AF023847), and most likely corresponds to the same gene as pTomβgal 1, but differs because different tomato cultivars were used to isolate the cDNAs.
Sequence comparison of the N-terminal region of our putative β-galactosidase clones revealed that the deduced amino acid sequence of one cDNA (pTomβgal 4) most closely matched (28 of 30 amino acids) the partial N-terminal amino acid sequence of β-galactosidase II (TOMAA) that was purified from ripening tomato fruit and was shown to have exo-β(1→4)galactosidase activity against tomato cell wall preparations and galactan substrates in vitro (Fig. 1) (Carey et al., 1995). We suspect that the second- and third-to-last residues (KY) in the TOMAA sequence are incorrect (Fig. 1). In all of the plant β-galactosidase sequences published to date, the residues ST occur at these positions (Fig. 1). In addition, all of the other tomato β-galactosidase clones we have sequenced also contain the residues ST or conserved substitutions at these positions (not shown). Therefore, the deduced amino acid sequence of pTomβgal 4 probably codes for the exo-β(1→4)galactanase characterized by Carey et al. (1995).
Figure 1.
Multiple sequence alignment of the N-terminal amino acid sequence of β-galactosidase II protein from tomato fruit and the deduced amino acid sequences of various plant β-galactosidase cDNA clones. The two-amino acid mismatch between the partial N-terminal amino acid sequence of β-galactosidase II (TOMAA) and the deduced amino acid sequences discussed at length in Results is indicated by boldface type. An asterisk (*) below any amino acid indicates that that position is conserved in all sequences. The accession numbers for the cDNA clones are: pTomβgal 4, AF020390; pTomβgal 1, P48980; asparagus, P45582; apple, P48981; and carnation, Q00662.
The complete sequence of the cDNA insert of pTomβgal 4 is available in GenBank. The cDNA insert is 2554 nucleotides long and contains a single, long ORF predicted to start with the first in-frame ATG at nucleotide 64 and end with TAA at nucleotide 2238. This ORF codes for a 79-kD protein 725 amino acids long. The programs PSORT version 6.4 (Nakai and Kanehisa, 1992) and SignalP version 1.1 (Nielsen et al., 1997) were used to predict that the ORF contains a hydrophobic leader sequence that would be cleaved between the Ala and Ser residues at positions 23 and 24, respectively, and that the mature polypeptide has an extracellular location. The mature polypeptide contains three possible N-glycosylation sites at Asn-282, Asn-459, and Asn-713; however, the Asn at position 713 is unlikely to be glycosylated because of the Pro at position 714. The predicted molecular mass of the unglycosylated mature polypeptide was 75 kD, with a pI of 8.9.
The deduced amino acid sequence of pTomβgal 4 shared significant identity with all published plant β-galactosidase amino acid sequences in the database (Fig. 1). When the entire ORF of each β-galactosidase gene was compared with that of pTomβgal 4, the shared sequence identity was 64% for tomato pTomβgal 1, 68% for apple, 62% for asparagus, and 56% for carnation.
DNA Gel-Blot Analysis
Because the tomato β-galactosidases exist as a multigene family, gel blots were performed to test all seven of the putative β-galactosidase clones for possible cross-hybridization. When the full-length pTomβgal 4 was used as a probe, no cross-hybridization to the other six clones was detected under high-stringency hybridization and washing conditions (data not shown). A 0.5-kb fragment derived from the 3′ end of the pTomβgal 4 cDNA insert was used as a probe to determine the gene's copy number because the genomic DNA sequence of this gene is uncharacterized. When high-stringency conditions were used, only one band was observed for each restriction-enzyme digest, suggesting that the gene corresponding to pTomβgal 4 is present as a single copy (Fig. 2).
Figure 2.
Autoradiograph of a gDNA gel blot. The probe was synthesized using a 0.5-kb fragment from the 3′ end of the pTomβgal 4 cDNA insert. Three micrograms of gDNA was digested with BamHI (B), EcoRI (E), or HindIII (H) and loaded in each lane.
pTomβgal 4 Hybridizes to a Transcript Expressed in Ripening Fruit
Northern-blot analysis revealed where and when pTomβgal 4 transcript was present during fruit and plant development. To minimize any potential signal during the long exposures necessary for northern-blot analysis due to cross-hybridization, the probe used was synthesized using the 3′ one-third of the pTomβgal 4 insert. The 3′ ends of the various plant and putative tomato β-galactosidase clones have the lowest degree of shared sequence identity (see Fig. 1). pTomβgal 4-specific probe did not detect transcript in tomato fruit peel, pericarp, or columella tissues between 10 and 45 DPA, i.e. up to the mature green stage of development. However, transcript was detected at the breaker stage (Fig. 3A), the stage representing incipient coloration and the visible beginning of ripening. Transcript reached maximum accumulation at the turning stage and continued to be present during the later stages of fruit ripening, albeit at decreasing levels (Fig. 3A). At the turning stage of development, maximum transcript levels were detected in the peel and/or outermost region of the outer pericarp, and was present in all fruit tissues tested (Fig. 3B). Expression of the gene corresponding to pTomβgal 4 was not limited to fruit; transcript was detected in roots, stems, and flowers but not in leaves (Fig. 3C).
Figure 3.
Detection of pTomβgal 4 hybridization by RNA gel-blot analysis. All lanes were prepared using 20 μg of total RNA. Probes used are indicated to the right of each blot. A, RNA was isolated from total pericarp and peel tissue of fruit from 10 to 45 DPA and from breaker (Br), turning (Tr), pink (Pk), red-ripe (Rd), and overripe (OR) fruit. B, RNA for this blot was isolated from various tissues including the peel (Pl), outer pericarp (OP), inner pericarp/columella (IP), and locules not including seeds (Lo) of turning-stage fruit. C, RNA for this blot was isolated from flower (F), leaf (L), root (R), and stem (S) tissues. The blots in the upper panels of A and C were exposed to Kodak XAR film using an intensifying screen at −80°C for 4 d, and the blot in the upper panel of B was exposed to Kodak Biomax film using an intensifying screen at −80°C for 2 d.
β-Galactosidase II Gene Expression Is Attenuated in the Ripening Mutants nor, rin, and Nr
Carey et al. (1995) showed that β-galactosidase II activity increased 4-fold during ripening in wild-type cv Ailsa Craig fruit, whereas in the mutants rin and nor, enzyme activity remained at the basal level of normal, mature green fruit and did not change during ripening. They also showed that total β-galactosidase activity showed no marked ripening-related changes, and levels were similar in both wild-type and mutant fruit. We therefore concluded that if pTomβgal 4 coded for β-galactosidase II, then it should not detect an increase in gene transcript in the mutants rin and nor during the period corresponding to the red-ripe stages of normal fruit development. Indeed, when pTomβgal 4 was used as a probe, very little if any transcript was detected in RNA extracted from nor and rin fruit at either 45 or 50 DPA (Fig. 4).
Figure 4.
Autoradiograph of RNA gel blots of mutant fruit. Twenty micrograms of total RNA from wild-type (wt) and the ripening mutants nor, rin, and Nr fruit peel and pericarp was loaded into each lane. Fruit was harvested at the DPA indicated or at the turning (Tr) or red-ripe (Rd) stages of development. Probes used were synthesized using the clones indicated to the right of each blot. The blots were exposed to Kodak Biomax film using an intensifying screen at −80°C for 2 d.
pTomβgal 4 also failed to hybridize to any transcript in RNA isolated from fruit of Nr (Fig. 4). As a positive control, pTomβgal 10 was used as a probe for the same RNA gel-blot analysis (Fig. 4). Carey at al (1995) had shown that the pTomβgal 1 clone detected transcript in fruit of the mutants nor and rin 45 and 65 DPA. Because we suspect that pTomβgal 10 corresponds to the same gene as pTomβgal 1 but that they are from different cultivars, it should hybridize to transcript isolated from both nor and rin fruit 45 to 65 DPA. As expected, pTomβgal 10 did hybridize to transcript isolated from fruit of nor and rin plants 45 and 50 DPA (Fig. 4). pTomβgal 10 also detected transcript in RNA isolated from fruit of Nr plants (Fig. 4).
pTomβgal 4 Codes for a β-Galactosidase
The pTomβgal 4 ORF was cloned in-frame into the repressible/inducible bacterial expression vector pFLAG-CTC. The host strain XL1-Blue MR is a mutant strain containing neither endogenous β-galactosidase activity nor α-complementation. Induction of gene transcription by IPTG caused the immediate cessation of E. coli growth at 30 to 37°C; however, induction at 20°C did allow for some limited growth. When clones containing the pTomβgal 4 ORF were grown at 20°C and induced with IPTG, the cells slowly turned blue after 36 h of growth in medium containing the β-galactosidase substrate X-Gal (Fig. 5). If not induced with IPTG, no blue coloration was seen, even after extended growth in medium containing X-Gal. As an additional negative control, clones consisting of XL1-Blue MR transformed with the FLAG vector alone showed no β-galactosidase activity with or without IPTG induction, even after 7 d of growth (Fig. 5). As a positive control for maximal β-galactosidase (derived from E. coli β-galactosidase) activity, the cloning vector pGEM was transformed into the host strain DH5α. These results are shown in Figure 5.
Figure 5.
Detection of β-galactosidase activity from pTomβgal 4 expression in E. coli. Cells were harvested and extracts were prepared every 12 h and the A615 was measured. Cultures were grown with the addition of the chromogenic substrate X-Gal (open symbols) or X-Gal and the transcriptional inducer IPTG (closed symbols) in the medium. The vector used as a positive control for E. coli β-galactosidase activity was pGEM (▪), and the vector used as a negative control and for expression was pFLAG-CTC either without (○, •) or with (⋄, ♦) the pTomβgal 4 ORF.
DISCUSSION
Although it is apparent that a number of enzymes may be involved in the degradation of cell wall pectin during fruit ripening, it is important to identify each enzyme and the corresponding gene. It should then be possible to create null mutations for each gene and gene combination to understand the overall effect each gene product has on cell wall metabolism and its consequential effect, if any, on fruit softening. To meet part of this objective, we investigated the role of β-galactosidases in tomato during fruit ripening and softening, and describe the cloning of a β-galactosidase cDNA clone that most likely codes for a β(1→4)galactan-degrading enzyme and is expressed in ripening tomato fruit tissues.
Although Carey et al. (1995) isolated three β-galactosidase isozymes and several related cDNAs, it is not known why a cDNA coding for β-galactosidase II, the isozyme with known activity against native polysaccharide substrates, was never identified (G. Seymour, personal communication). We used one of the RT-PCR clones (2-1) to screen our own cDNA library; however, this RT-PCR clone did not share more sequence identity to pTomβgal 4 than did pTomβgal 1 (data not shown). It is possible that the choice of poly(A+) RNA that was used to construct the libraries was critical in the identification of β-galactosidase cDNA clones.
We believe that pTomβgal 4 is a cDNA derived from the transcript of a gene that codes for β-galactosidase II for three reasons. First, the deduced amino acid sequence of the highly conserved N-terminal portion of the expected mature pTomβgal 4 translation product matches almost exactly (28 of 30 amino acids) the N-terminal sequence of β-galactosidase II (Fig. 1). The two amino acids (KY) in the β-galactosidase II sequence that do not match the pTomβgal 4 deduced amino acid sequence are believed to be incorrect, since all plant β-galactosidase sequences in the database and four additional β-galactosidase-related cDNAs that were identified from tomato match the deduced amino acid sequence of pTomβgal 4 at these same two amino acid (ST) positions (Fig. 1).
Second, the transcript detected by pTomβgal 4 is present in normal ripening fruit at the same time that β-galactosidase II activity is detected (Fig. 3) (Carey et al., 1995). Moreover, little or no transcript was detected in fruit at 45 and 50 DPA from the mutants nor, rin, and Nr (Fig. 4). This observation also coincides with the data presented by Carey et al. (1995) that β-galactosidase II activity remained at levels equal to those in mature green fruit and did not increase in fruit from nor or rin plants 45 to 65 DPA. Carrington and Pressey (1996) recently reported that β-galactosidase II activity was detected in cv Rutgers fruit only after the turning stage of ripeness. The northernblot data in the present study suggest that maximum β-galactosidase II activity should occur only after the turning stage, assuming that mRNA levels predict extractable enzyme activity (Fig. 3).
Third, the apparent molecular mass of 77.9 kD and the pI of 8.9 for the mature protein predicted from the pTomβgal 4 sequence are similar to those determined for β-galactosidase II. Pressey (1983) estimated a molecular mass of 62 kD by gel-filtration column chromatography and a pI of 7.8 by IEF, and Carey et al. (1995) estimated a molecular mass of 75 kD by SDS-PAGE and a pI of 9.8.
To evaluate the role of the gene corresponding to pTomβgal 4 in tomato fruit ripening/softening, we have initiated gene-knockout studies. We are currently establishing transgenic tomato plant lines via Agrobacterium tumefaciens-mediated transformation, which are expressing pTomβgal 4 in the antisense orientation.
ACKNOWLEDGMENTS
The authors express their appreciation to Karen Green and J. Norman Livsey for excellent technical support, to Frank Turano for generously providing the 26S soybean rDNA clone, and to Luca Pelligrini for critically reviewing the manuscript.
Abbreviations:
- DPA
days postanthesis
- IPTG
isopropyl-β-d-thiogalactopyranoside
- ORF
open reading frame
- PG
polygalacturonase (endo-α1→4-d-galacturonan hydrolase)
- PME
pectin methylesterase
- RT
reverse transcriptase
- X-Gal
5-bromo-4-chloro-3-indoxyl-β-d-galactopyranoside
Footnotes
The accession number for the pTomβgal 4 sequence reported in this article is AF020390.
Use of a company or product name by the U.S. Department of Agriculture does not imply approval or recommendation of the product to the exclusion of others that may also be suitable.
LITERATURE CITED
- Altschul SF, Gish W, Miller W, Meyers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215:403–410. doi: 10.1016/S0022-2836(05)80360-2. [DOI] [PubMed] [Google Scholar]
- Ausubel F, Brent R, Kingston R, Moore D, Seidman J, Smith J, Struhl K, eds (1987) Current Protocols in Molecular Biology. John Wiley & Sons, New York
- Carey AT, Holt K, Picard S, Wilde R, Tucker GA, Bird CR, Schuch W, Seymour GB. Tomato exo-(1→4)-β-d-galactanase. Isolation, changes during ripening in normal and mutant tomato fruit, and characterization of a related clone. Plant Physiol. 1995;108:1099–1107. doi: 10.1104/pp.108.3.1099. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carrington CM, Pressey R. β-Galactosidase II activity in relation to changes in cell wall galactosyl composition during tomato ripening. J Am Soc Hortic Sci. 1996;121:132–136. [Google Scholar]
- Church GM, Gilbert W. Genomic sequencing. Proc Natl Acad Sci USA. 1984;81:1991–1995. doi: 10.1073/pnas.81.7.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- DeVeau EJ, Gross KC, Huber DJ, Watada AE. Degradation and solubilization of pectin by β-galactosidases purified from avocado mesocarp. Physiol Plant. 1993;87:279–285. [Google Scholar]
- Fischer RL, Bennett AB. Role of cell wall hydrolases in fruit ripening. Annu Rev Plant Physiol Plant Mol Biol. 1991;42:675–703. [Google Scholar]
- Giovannoni JJ, DellaPenna D, Bennett AB, Fischer RL. Expression of a chimeric polygalacturonase gene in transgenic rin (ripening inhibitor) tomato fruit results in polyuronide degradation but not fruit softening. Plant Cell. 1989;1:53–63. doi: 10.1105/tpc.1.1.53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Golden KD, John MA, Kean EA. β-Galactosidase from Coffea arabica and its role in fruit ripening. Phytochemistry. 1993;34:355–360. [Google Scholar]
- Gross KC. Fractionation and partial characterization of cell walls from normal and non-ripening mutant tomato fruit. Physiol Plant. 1984;62:25–32. [Google Scholar]
- Gross KC, Sams CE. Changes in cell wall neutral sugar composition during fruit ripening: a species survey. Phytochemistry. 1984;23:2457–2461. [Google Scholar]
- Hall LN, Tucker GA, Smith CJS, Watson CF, Seymour GB, Bundick Y, Boniwell JM, Fletcher JD, Ray JA, Schuch W and others. Antisense inhibition of pectin esterase gene expression in transgenic tomatoes. Plant J. 1993;3:121–129. [Google Scholar]
- Huber DJ. The role of cell wall hydrolases in fruit softening. Hortic Rev. 1983;5:169–219. [Google Scholar]
- Kang IK, Suh SG, Gross KC, Byun JK. N-terminal amino acid sequence of persimmon fruit β-galactosidase. Plant Physiol. 1994;105:975–979. doi: 10.1104/pp.105.3.975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kim J, Gross KC, Solomos T. Galactose metabolism and ethylene production during development and ripening of tomato fruit. Postharv Biol Technol. 1991;1:67–80. [Google Scholar]
- Marck C. DNA Strider: a “C” program for the fast analysis of DNA and protein sequences on the Apple Macintosh family of computers. Nucleic Acids Res. 1988;16:1829–1836. doi: 10.1093/nar/16.5.1829. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mitcham EJ, Gross KC, Ng TJ. Tomato fruit cell wall synthesis during development and senescence. In vivo radiolabeling of cell wall fractions using [14C]sucrose. Plant Physiol. 1989;89:477–481. doi: 10.1104/pp.89.2.477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nakai K, Kanehisa M. A knowledge base for predicting protein localization sites in eukaryotic cells. Genomics. 1992;14:897–911. doi: 10.1016/S0888-7543(05)80111-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nielsen H, Engelbrecht J, Brunak S, von Heijne G. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng. 1997;10:1–6. doi: 10.1093/protein/10.1.1. [DOI] [PubMed] [Google Scholar]
- Pressey R. β-Galactosidases in ripening tomatoes. Plant Physiol. 1983;71:132–135. doi: 10.1104/pp.71.1.132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ross GS, Redgwell RJ, MacRae EA. Kiwifruit β-galactosidase: isolation and activity against specific fruit cell-wall polysaccharides. Planta. 1993;189:499–506. [Google Scholar]
- Ross GS, Wegrzyn T, MacRae EA, Redgwell RJ. Apple β-galactosidase. Activity against cell wall polysaccharides and characterization of a related cDNA clone. Plant Physiol. 1994;106:521–528. doi: 10.1104/pp.106.2.521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Seymour GB, Gross KC. Cell wall disassembly and fruit softening. Postharvest News Info. 1996;7:45N–52N. [Google Scholar]
- Smith CJS, Watson CFS, Ray J, Bird CR, Morris PC, Schuch W, Grierson D. Antisense RNA inhibition of polygalacturonase gene expression in transgenic tomatoes. Nature. 1988;334:724–726. [Google Scholar]
- Smith DL, Fedoroff NV. LRP1, a gene expressed in lateral and adventitious root primordia of Arabidopsis. Plant Cell. 1995;7:735–745. doi: 10.1105/tpc.7.6.735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tieman DM, Harriman RW, Ramamohan G, Handa AK. An antisense pectin methylesterase gene alters pectin chemistry and soluble solids in tomato fruit. Plant Cell. 1992;4:667–679. doi: 10.1105/tpc.4.6.667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tigchelaar EC, McGlasson WB, Buescher RW. Genetic regulation of tomato fruit ripening. HortScience. 1978;13:508–513. [Google Scholar]
- Turano FJ, Thakkar SS, Fang T, Weisemann JM. Characterization and expression of NAD(H) dependent glutamate dehydrogenase genes in Arabidopsis thaliana. Plant Physiol. 1997;113:1329–1341. doi: 10.1104/pp.113.4.1329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Verwoerd TC, Dekker BMM, Hoekema A. A small-scale procedure for the rapid isolation of plant RNAs. Nucleic Acids Res. 1989;17:2362. doi: 10.1093/nar/17.6.2362. [DOI] [PMC free article] [PubMed] [Google Scholar]