Journal of Experimental Botany

This Article Abstract FREE Full Text (PDF) E-letters: Submit a response Alert me when this article is cited Alert me when E-letters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (20) Request Permissions Disclaimer Google Scholar Articles by Feurtado, J. A. Articles by Bewley, J. D. Search for Related Content PubMed PubMed Citation Articles by Feurtado, J. A. Articles by Bewley, J. D. Agricola Articles by Feurtado, J. A. Articles by Bewley, J. D. Social Bookmarking          What's this?

Journal of Experimental Botany, Vol. 52, No. 359, pp. 1239-1249, June 1, 2001
© 2001 Oxford University Press

Original Papers

The cloning and characterization of -galactosidase present during and following germination of tomato (Lycopersicon esculentum Mill.) seed

J. Allan Feurtado¹, Mitali Banik and J. Derek Bewley²

Department of Botany, University of Guelph, Guelph, Ontario N1G 2W1, Canada

Received 2 August 2000; Accepted 19 February 2001

	Abstract

Top Abstract Introduction Materials and methods Results and discussion References

 
-Galactosidase (EC 3.2.1.22) is present in the embryo, micropylar and lateral endosperm of seeds of tomato during and following germination. Its activity is unchanged even when germination of the seeds is prevented by an osmoticum. It is also present in the developing and mature dry seed. A cDNA clone for tomato seed -galactosidase (LeaGal) has been isolated and the characteristics of the protein deduced; the predicted molecular mass of the mature enzyme is 39.8 kDa, with a pI of 4.91. The tomato -galactosidase has a high homology (>62%) at the amino acid level with that of other plant -galactosidases. A hydrophobic signal peptide region is identified which is indicative that the enzyme enters the lumen of the endoplasmic reticulum during its translation, prior to its export to the protein body or cell wall, the presumed sites of its substrates. Using amino acid alignment and phylogenetic analysis, key amino acids have been identified, and relationships to other -galactosidases inferred. Southern hybridization analyses show that the enzyme is derived from a single gene (for which a partial sequence has been obtained) and yet there are at least three different isoforms within the seed; post-translational modifications are thus presumed to occur. From Northern hybridization studies it is evident that -galactosidase transcripts are present in the lateral and micropylar endosperm during and following germination, and also to a lesser extent in the embryo.

Key words: -Galactosidase, tomato seed, germination, gene expression.

	Introduction

Top Abstract Introduction Materials and methods Results and discussion References

 
Tomato seeds exhibit ‘coat-enhanced’ dormancy, where the embryo is non-dormant but is mechanically restrained by the surrounding endosperm (Groot and Karssen, 1987). The cell walls of the endosperm are rich in mannans with an appreciable amount of galactose and glucose being present as well, which is suggestive of there being galactomannans or galactoglucomannans as a wall constituent (Groot et al., 1988). Several studies have shown that endo-ß-mannanase (EC 3.2.1.78), the key enzyme involved in the breakdown of mannan polymers, increases in tomato seeds during germination in the embryo and micropylar endosperm (Nonogaki et al., 1992; Nomaguchi et al., 1995; Toorop et al., 1996; Voigt and Bewley, 1996; Still and Bradford, 1997). Weakening of the micropylar endosperm is a prerequisite for radicle emergence and while endo-ß-mannanase may contribute to this event, its presence seems not to be essentially linked to the completion of germination (Toorop et al., 1996; Still and Bradford, 1997; Bewley, 1997). Later, after germination, an increase in activity of several isoforms of this enzyme occurs in the lateral endosperm, as the seedling becomes established (Toorop et al., 1996; Voigt and Bewley, 1996).

Degradation of cell wall galactomannan requires the presence of three enzymes working co-operatively, endo-ß-mannanase, ß-mannoside mannohydrolase (ß-mannosidase, exo-ß-mannanase, EC 3.2.1.25) and -galactosidase (EC 3.2.1.22) (Reid and Meier, 1973). Endo-ß-mannanase has been studied extensively, but while the other two hydrolases have been reported to be present in tomato seeds (Groot et al., 1988; Hilhorst and Downie, 1996) very little is known about their location and activity during and following germination. Yet these enzymes could be intricately linked in their activity to that of endo-ß-mannanase in the mobilization of the endosperm cell walls. The aleurone layer of fenugreek (Trigonella foenum-graecum), for example, concurrently synthesizes and secretes endo-ß-mannanase and -galactosidase (Reid, 1985) during mobilization of the galactomannan-rich endosperm cell walls. In lucerne (Medicago sativa) and guar (Cyamopsis tetragonoloba) seeds -galactosidase action has been suggested to be required before the other two enzymes can hydrolyse cell-wall galactomannans (McCleary and Matheson, 1975). In contrast, in germinated lettuce seeds -galactosidase can only release galactose from the products of prior endo-ß-mannanase activity (Leung and Bewley, 1983).

This study reports on the timing and location of -galactosidase activity during and after germination of the tomato seed, and also during development. cDNA and genomic clones have been obtained for the enzyme, and the characteristics of the derived protein determined. Its pattern of activity is unlike that of endo-ß-mannanase or ß-mannoside mannohydrolase (Feurtado, 1999) in that it is present virtually unchanged throughout the seed during germination and early seedling growth.

	Materials and methods

Top Abstract Introduction Materials and methods Results and discussion References

 
Seed germination
Tomato seeds (Lycopersicon esculentum Mill. cv. Glamour, 1996 and 1998 harvest years, Stokes Seeds Ltd., St Catharines, Ontario) were stored at 4 °C over silica gel until used. Seeds were imbibed on two layers of Whatman No. 1 filter paper with approximately 5 ml double-deionized H₂O (ddH₂O) or PEG 8000 (Sigma) and left in the dark to germinate at 25 °C.

Developing seeds
Tomato fruits were either obtained from The Greenhouse, Brampton, ON, or grown in greenhouses at the University of Guelph under natural lighting conditions. Fruits were staged according to their degree of ripeness (Bewley et al., 2000). Seeds were removed from the fruits at each stage, rinsed in 2% HCl overnight, washed in ddH₂O, and dried slowly on Whatman No. 1 filter paper on the laboratory bench before storage at 5 °C until required. -Galactosidase activity was also assayed using seeds at some of the stages immediately following their removal from the fruit, and identical results were obtained.

Galactomannan-degrading enzymes: extraction from tomato seed parts
Fifteen seeds were dissected to separate the embryo, lateral endosperm, and micropylar endosperm (Toorop et al., 1996) at 12 h intervals throughout germination and early seedling growth. Each was ground in an ice-cold mortar in 200 µl 0.1 M Hepes/0.5 M NaCl buffer (pH 8). It had previously been determined that extraction in this buffer was better than using McIlvaine buffer pH 5 (McIlvaine, 1921). Extraction in the presence or absence of salt was equally efficient. The extract was centrifuged at 21 000 g for 10 min at 4 °C. The supernatants collected were all adjusted to 0.5 ml volume, or 0.7 ml for lateral endosperms prior to enzyme assays. Prior to -galactosidase assays, a 10% aqueous solution of polyethylenimine (PEI, Sigma) pH 8 (made from 50% aqueous stock PEI) was added to enzyme extracts to a final concentration of approximately 2%. Extracts were mixed and then centrifuged at 4 °C at 21 000 g for 5 min to clear. For developing seeds, the above protocol was used, except that 10 intact seeds were extracted in 1.5 ml buffer.

PEG 8000 solutions of the desired osmotic potential were prepared according to the equation of Michel and Kaufmann (Michel and Kaufmann, 1973). An MPa of -0.5 was used since it produced a water potential just sufficient to inhibit radicle protrusion. Tomato seeds were dissected and enzymes extracted as above.

-Galactosidase assays
The -galactosidase assay was adapted from Leung and Bewley (Leung and Bewley, 1981). The modification involved using 96-well microtitre-assay plates. The assay mixture consisted of 75 µl McIlvaine buffer pH 4.5 (0.1 M citric acid, 0.2 M Na₂HPO₄.7H₂O, 1:0.81 ratio), 15 µl substrate (10 mM p-nitrophenyl--D-galactopyranoside (Sigma) dissolved in McIlvaine pH 5 buffer (1:1.1 ratio), and 60 µl enzyme extract. After incubation at 37 °C for 15 min, reactions were stopped by adding 75 µl 0.2 M Na₂CO₃. The yellow colour produced was measured at OD₄₀₅ in a microplate reader (Molecular Devices Corp., Sunnyvale, CA). Control experiments involved adding enzyme extracts after the stop solution had been added. These controls were used as the zero calibration reading. The molar extinction coefficient for p-nitrophenol was taken as 18 400 (Reid and Meier, 1973) to calculate the amount of p-nitrophenol released in units of pmol min^-1 seed part^-1.

Isolation of a cDNA clone
A cDNA library, prepared using germinated gib-1-mutant tomato seeds treated with GA as a starting material, was obtained from Dr KJ Bradford (University of California, Davis, CA). For screening the cDNA library, the manufacturer's instructions were followed with limited modifications (Zap Express cDNA Gigapack III Gold Cloning Kit, Stratagene). Replica filters were placed between Whatman 3 MM paper and autoclaved for 3 min. The membranes were fixed plaque-side down on a UV transilluminator, using the analytical setting, for 3–4 min. Replica filters (Hybond-N nylon membranes, Amersham Pharmacia Biotech) were screened by hybridization using DIG-labelled coffee bean (Coffea arabica) -galactosidase cDNA. Drs Jack Goldstein and Alex Zhu of the New York Blood Center, New York, NY, provided the coffee bean -galactosidase cDNA. The probe was prepared using the DIG High Prime DNA Labelling and Detection Starter Kit II (Roche). Positive clones were rescued into the pBK-CMV phagemid vector (Stratagene) and isolated using the QIAprep Spin Miniprep Kit (Qiagen). Putative positives were sequenced using a Perkin Elmer 377 (ABI Prism, v.3.2) DNA sequencer. Sequence analysis was carried out using Generunner and World Wide Web resources including BLAST searches of the National Center for Biotechnology Information (NCBI) database (Altschul et al., 1990) and Baylor College of Medicine (BCM) sequence analysis tools (Smith et al., 1996). Both strands of the longest clone, designated LeaGal (GenBank accession number AF191823), were sequenced.

Amino acid alignment and phylogenetic analysis
The deduced LeaGal protein sequence was first aligned with previously isolated dicotyledonous -galactosidases (or -galactosidase-like proteins) coffee bean (Zhu and Goldstein, 1994; AAA33022), pinto bean (Phaseolus vulgaris) (Davis et al., 1997; AAA73964), guar (Overbeeke et al., 1989; P14749), soybean (Glycine max) (Davis et al., 1996; AAA73963), antbush (Senna occidentalis) (CAA03733), and Arabidopsis thaliana (CAB87430), using the BCM Multiple Alignment Program (MAP) (Huang, 1994). Further analysis of the LeaGal protein and its dicot cohorts with other eukaryotic -galactosidases (or -galactosidase-like proteins) was accomplished through phylogenetic analysis. Nineteen -galactosidases representing each eukaryotic genus extractable from NCBI, together with an Escherichia coli -galactosidase (Aslanidis et al., 1989; P16551) used as a hypothetical outgroup, were aligned using MAP. In addition to the dicot species, the species and accession numbers were: Aspergillus niger (de Vries et al., 1999; CAB46229), Caenorhabditis elegans (T24018), Drosophila melanogaster (Adams et al., 2000, AAF52871), Homo sapiens (Kornreich et al., 1989; NP_000160), Mortierella vinacea (Shibuya et al., 1997; BAA33931), Mus musculus (Ohshima et al., 1995; P51569), Penicillium simplicissimum (CAA08915), Saccharomyces cerevisiase (Turakainen et al., 1991; JQ1021), Schizosaccharomyces pombe (CAA93244), Torulaspora delbrueckii (Oda and Fukunaga, 1999; BAA86883), Trichoderma reesei (Hypocrea jecorina) (CAA93244), and Zygosaccharomyces cidri (Turakainen et al., 1994; AAA35280). A phylogenetic tree was constructed using the PHYLIP software package, v.3.5c (Felsenstein, 1993). The aligned sequences were used to create a distance matrix that was analysed by the Fitch–Margoliash (Fitch and Margoliash, 1967) method. To ensure the best possible tree, global rearrangements and five randomizations (jumbles) of the species input order were performed. Further analysis included subjecting 100 bootstrapped replications to the same method, which resulted in the same tree configuration. Final adjustments included rooting the tree to an E. coli -galactosidase as the hypothetical outgroup.

Extraction of RNA and northern blot analysis
Tomato seeds imbibed for 24, 36, 48, 60, and 72 h were dissected into embryos, micropylar and lateral endosperms (Toorop et al., 1996). Approximately 1500 parts of each were collected, frozen in liquid N₂ and ground to a fine powder before extraction of total RNA using the hot phenol/LiCl extraction buffer method (Verwoerd et al., 1989). The powder was suspended in 15–20 ml 80 °C phenol:buffer solution (1:1 ratio of phenol to 100 mM TRIS-Cl pH 8, 100 mM LiCl, 10 mM EDTA, and 1% SDS). The solution was precipitated overnight with 4 M LiCl and then again with 3 M sodium acetate (pH 5.2) and cold absolute ethanol (both at -20 °C); the resulting pellets were dissolved in DEPC-treated ddH₂O. Twenty µg seed part total RNA were fractionated in a 1.2% agarose gel in formaldehyde and transferred to a nylon membrane (Zeta-Probe, BioRad) primarily as described earlier (Sambrook et al., 1989). Following transfer, the membrane was fixed using an UV transilluminator for 3–4 min, then briefly soaked in 2x SSC to remove any residual 20x SSC. The membrane was prehybridized in prewarmed solution (1.5 ml 50x Denhardt's (1966) solution, 2.5 ml 20x SSPE, 0.5 ml 10% SDS, 1 ml denatured salmon sperm DNA (10 mg ml^-1), and 4.5 ml formamide) for 4–6 h. It was probed using 1 ml LeaGal cDNA (-³²P-labelled, Random Primed cDNA Labelling Kit, Roche) in 9 ml prewarmed hybridization solution (1 ml 50x Denhardt's, 2.5 ml 20x SSPE, 0.5 ml 10% SDS, 0.5 ml denatured salmon sperm DNA (10 mg ml^-1), and 4.5 ml formamide). After prehybridization and hybrization for approximately 24 h at 42 °C the membrane was washed with 2x SSC and 0.1% SDS, 1x SSC and 0.1% SDS, 0.5x SSC and 0.1% SDS, and 0.1x SSC and 0.1 SDS for 20 min each at 42 °C, before being dried briefly on Whatman 3 MM and wrapped in plastic before exposure to X-ray film at –60 °C (Kodak X OMAT AR).

DNA extraction and Southern blot analysis
Genomic DNA from 5 g tomato leaves was extracted using a modified CTAB method (Murray and Thompson, 1980). Tomato leaves were ground in liquid N₂ before being lyophilized for 24 h. An equal volume of extraction buffer (50 mM TRIS-HCl pH 8, 0.7 M NaCl, 10 mM EDTA, 1% [w/v] CTAB, and 1% [v/v] ß-mercaptoethanol) was added to the lyophilized powder and mixed gently before incubating at 56 °C for 15 min. The solution was cooled at room temperature before adding 1 vol. chloroform:octanol (24:1 v/v). After clearing for 5 min at 21000 g, the aqueous phase was mixed gently with 0.1 vol. 10% (w/v) CTAB before adding 1 vol. chloroform:octanol (v/v). After clearing at 21 000 g for 5 min, 1.2 vol. 50 mM TRIS-Cl pH 8, 10 mM EDTA, and 1% w/v CTAB were added to the aqueous phase before mixing gently. After standing for 30 min and a clearing spin for 10 min at 21 000 g, 0.5 vol. 1 M NaCl was added to dissolve the pellet. Following RNase treatment, the DNA was precipitated overnight in cold absolute ethanol at –20 °C, pelleted, and rinsed in 70% ethanol. The resultant DNA was dissolved in 50–100 µl sterile ddH₂O. Thirty µg DNA were digested with restriction enzyme endonucleases before being separated in a 1% agarose gel. The gel was soaked in depurination solution (0.25 M HCl) for 10 min with gentle shaking, in denaturation buffer (0.5 M NaOH, 1.5 m NaCl) for 2 washes of 15 min, and in neutralization buffer (0.5 M TRIS-HCl, pH 7.2, 1.5 M NaCl, and 1 mM EDTA) for 2 washes of 15 min. The transfer of DNA to a Zeta-Probe nylon membrane (BioRad) primarily followed Sambrook et al. (Sambrook et al., 1989). Following transfer, the membrane was fixed over a UV transilluminator and washed in 2x SSC before prehybridization. The prehybridization solution consisted of 6x SSC, 20 mM NaH₂PO₄, 5x Denhardt's, 0.4% SDS, and 250 µg denatured salmon sperm DNA. The membrane was prehybridized for 3 h at 65 °C before addition of -³²P-labelled tomato LeaGal full-length probe (Random Primed cDNA Labeling Kit, Roche) and hybridization for approximately 18 h at 65 °C. After hydridization the membrane was washed in 2x SSC, 0.1% SDS and 1x SSC, 0.1% SDS for 20 min each at 65 °C. The membrane was blotted dry on Whatman 3 MM paper before covering with plastic and exposing to X-ray film at -60 °C (Kodak X OMAT AR).

Preparation and screening of a tomato genomic library
Genomic DNA was prepared from tomato leaves as above. About 100 µg genomic DNA was partially digested with Hind III and separated on a 1% low melting point agarose gel. DNA ranging from 9–23 kb bandsize was cut out with the gel which was digested with Gelase enzyme (Epicenter Technologies, Madison, WI). DNA was precipitated with ammonium acetate (pH 7.0) to a final concentration of 2.5 M and 4 vol. ethanol. The DNA pellet was recovered by centrifugation, washed with 70% ethanol and dissolved in TE buffer (pH 7.5).

To prepare the cloning vector about 5 mg DasHII (Stratagene) was digested with HindIII and XhoI. The HindIII site was used for cloning. Vector arms were dephosphorylated with shrimp alkaline phosphatase. After extraction with phenol:chloroform and precipitation the pellet was dissolved in 10 mM TRIS-HCl, pH 7.5. The partially digested DNA and DashII DNA were ligated using T4 ligase. The ligated DNA was packaged using Gigapack III gold packaging extract (Stratagene). The library was plated out on lawns of E. coli XLI-blue (P2) cells and screened by hybridization of membrane filter plaques replicas (Sambrook et al., 1989) using the 1540 bp tomato -³²P-labelled LeaGal as a probe. Following plaque purification of positive clones, phage DNA was isolated using a Wizard lambda Preps DNA Purification System (Promega, Madison, WI, USA). A 6.5 kb HindIII fragment was subcloned into the corresponding site of pBluscript II (SK)+ (Stratagene) for sequence analysis.

Isoelectric focusing of tomato -galactosidase
Tomato seeds were imbibed for 24 h. The embryo and endosperm were separated and ground in extraction buffer containing 100 mM TRIS-HCl, pH 8.0, 150 mM NaCl, and protease inhibitor cocktail (Sigma). The extract was centrifuged for 10 min at 21 000 g in a microfuge and the supernatant concentrated and desalted with 50 mM TRIS-HCl, pH 8.0 by ultrafiltration using a Centricon-10 (Amicon Inc, Beverly, MI).

Ten per cent (w/v) polyacrylamide isoelectric focusing gels were cast between 125x260 mm glass plates with a 5 mm spacer. Ampholytes with a pH range 3–10 (BioRad) were used. The gels were prefocused at 500 V for 20 min, 1000 V for 7 min, 1500 V for 7 min, and 2000 V for 10 min. Then the samples were loaded on the strips on the basic side. The gel was focused for 30 min at 2000 V and then the strips were removed and the gels focused for another 90 min. The protein from the gel was transferred to a nitrocellulose membrane (Trans-Blot transfer medium, Bio-Rad) by non-electrophoretic diffusion (Chandra Sekhar and DeMason, 1990). Following transfer the blot was incubated with 3 mM -galactosidase substrate (6-bromo-2-naphthyl--D-galactopyranoside, Sigma) containing 0.04% (w/v) Fast Garnet GNC (Sigma) in 100 mM sodium acetate buffer, pH 5.5 and incubated at 35 °C. Within 20 min of incubation the coloured bands were evident. The reaction was stopped by placing the membrane in 6% acetic acid.

Following focusing, the pH gradient was measured by cutting an empty lane into 1 cm² pieces and incubating the pieces in 500 µl deionized water with shaking. Isoelectric points (pIs) were determined by measuring the distance from each isoform to the basic side using the pH gradient as a standard. Another 1 cm² gel slice containing protein was used to measure -galactosidase activity using p-nitrophenyl -D-galactopyranoside (Sigma) as a substrate.

	Results and discussion

Top Abstract Introduction Materials and methods Results and discussion References

 
Changes in -galactosidase in tomato seeds
Two previous studies have shown the presence of -galactosidase in tomato seeds. A small (<20%) increase in activity in isolated endosperms of gib-1-mutant seeds, incubated in GA, but not in water was reported (Groot et al., 1988). However, activity was already high at the start of incubation. It has also been shown that activity is lower in endosperms of cv. Moneymaker wild type than its ABA-deficient sit^w mutant at two time points during germination, but there was no difference in activity between the time points for either genotype (Hilhorst and Downie, 1996). To define the presence of -galactosidase in tomato seeds further, its activity was examined both during and following germination of the cv. Glamour, and in different locations in the seed (Fig. 1A). The first seeds to complete germination exhibited radicle emergence after 36 h from the start of imbibition, and germination of the population was completed by 72 h (Fig. 1A).

View larger version (81K): [in this window] [in a new window]   Fig. 1. Germination (curve, shaded) and -galactosidase activity in the embryo, micropylar and lateral endosperm (bar graphs) of cv. Glamour tomato seeds. Enzyme activity is expressed on (A) a per seed part basis and (B) a per gram fresh weight (gfwt) basis. 48NG, 48 h ungerminated seed, 48G, 48 h germinated seed. Error bars are the standard deviation of six replicates from three different extractions.

 
-Galactosidase activity was present in the embryo and endosperm at 12 h from the start of imbibition (Fig. 1). On a per seed part basis, activity was more or less constant during the first 72 h in all seed parts, with most being in the lateral endosperm, and least in the micropylar tip endosperm region (Fig. 1A). At 48 h, the germinated and ungerminated seeds were assayed separately, and enzyme activities were the same in both. When the data were expressed on a fresh weight basis, activity was high in the micropylar tip, as expected given the small size of this region, and about the same as in the larger lateral endosperm (Fig. 1B). Upon completion of germination the radicle penetrated the micropylar endosperm, and while much of it was destroyed as a consequence there were still remnants of this region that could be dissected and assayed. Some overlap of this tissue with the lateral region is possible, since the micropylar and lateral regions are only approximate and there is no clear line of demarcation between the two. Overall, though, this activity in the micropylar region accounted for only about 15–20% of the total in the endosperm (Fig. 1A); the small increase in this region (on a fresh weight basis) after 60 h (Fig. 1B) was not significant. Embryo activity was lowest on a fresh weight basis, and decreased following germination, likely due to the increase in fresh weight as it grew into a seedling (Fig. 1B). Presumably the -galactosidase in the embryo was involved in the mobilization of the raffinose oligosaccharides that are present therein (Hilhorst and Downie, 1996).

-Galactosidase thus appears to be a constitutive enzyme whose activity is not specific to germination or the post-germinative mobilization of the galactomannan reserves of the endosperm during seedling growth. To establish this further, tomato seeds were imbibed and prevented from germinating by placing them in a solution of -0.5 MPa PEG. This had no consistent effect on the amount of -galactosidase activity over an 84-h period in any of the seed parts (Fig. 2) nor on a fresh weight basis, except that there was no decline at the later times in the embryo, since it did not complete germination nor subsequently increase in fresh weight (not shown). Incubating the seeds in 5 or 10 µm ABA also prevented germination, but there was no increase or decrease in -galactosidase activity during a 72 h imbibition compared with the water controls (not shown). Hence -galactosidase activity is unlike that of endo-ß-mannanase (Toorop et al., 1996; Voigt and Bewley, 1996) in that it is present early in germination in all regions of the seed, and especially in the lateral endosperm region. Endo-ß-mannanase does not increase in the lateral endosperm unless germination has been completed. Further, inhibition of germination by osmotica or ABA prevents this subsequent endo-ß-mannanase production (Toorop et al., 1996).

View larger version (31K): [in this window] [in a new window]   Fig. 2. -Galactosidase activity in the embryo, micropylar and lateral endosperms of ungerminated cv. Glamour tomato seeds during imbibition in -0.5 MPa polyethylene glycol. Activity is expressed on a per seed part. Error bars are the standard deviation of four replicates from two extractions at each time point.

 

-Galactosidase activity in developing seeds
Since -galactosidase was already definitely present in mature seeds during early imbibition, it was determined when, during development, this activity appeared. -Galactosidase was present in mature seeds taken from stage 7 over-ripe fruits, in similar amounts to that present during germination (Figs 1, 3). During development, activity increased in seeds taken at an early stage of fruit ripening (stage 2, breaker/turning stage) and remained constant from stage 4 (dark orange/red stage) onwards, again at a comparable level to that in germinating seeds. The initial increase in -galactosidase activity at stage 2 coincided with the commencement of synthesis of the seed storage proteins (not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 3. -Galactosidase activity in tomato seeds isolated from fruits at different stages of ripening and then dried. Error bars are the standard deviation of six replicates from two separate experiments.

 

Characterization of an -galactosidase cDNA and its deduced protein
As part of this objective to identify and characterize the enzymes associated with galactomannan breakdown, a cDNA clone for -galactosidase was obtained from the tomato seed. The longest isolated clone obtained, designated LeaGal (GenBank accession number AF191823), is 1540 bp in length, has an open reading frame which encompasses nucleotides 99 to 1325 and has a G+C content of 40.5%. There are 5' and 3' untranslated regions of 98 and 215 bp, respectively, and a TAA stop codon at nucleotide 1326. A comparison of the cDNA sequence with others in GenBank pointed out striking homologies with the previously isolated dicotyledonous -galactosidase cDNAs. From this it was concluded that the LeaGal cDNA encodes a tomato -galactosidase.

The entire -galactosidase protein, as deduced from the cDNA sequence, contains 409 amino acids (aa) (Fig. 4) and has a molecular mass of 44.9 kDa, and a pI of 5.27. On the assumption, however, that the mature protein begins at aa 46, after a signal peptide, the enzyme is 364 aa in length, is 39.8 kDa in mass and has a pI of 4.91. This pI is closer to that observed using isoelectric focusing (see Fig. 7). A signal peptide of 15 aa has been reported in coffee bean (Zhu and Goldstein, 1994). In guar, an unusually long pre-pro-sequence of 47 aa has been found (Overbeeke et al., 1989). The authors suggest that the first 24 residues serve as a signal peptide, that there is a signal-peptidase-processing site between residues 24 and 25, and that further processing takes place by peptidases to form the mature protein. Similarly, soybean and pinto bean have long pre-pro-sequences of 59 and 62 aa, respectively, based on aa analysis and N-terminal sequencing of mature recombinant proteins (Davis et al., 1996, 1997). In tomato, the first 22 aa are hydrophobic, as revealed by a Kyte and Doolittle hydropathy profile (Kyte and Doolittle, 1982), suggesting this is part of a signal peptide. In fact, the SignalP WWW server (Nielsen et al., 1997) predicts a cleavage site between aa 22 and 23 (VYA-RL). However, the mature protein, based on the aa alignment data, seems to begin at aa 46 (Fig. 4). This would make the N-terminal sequence of immature -galactosidase of tomato a similar length to that of guar (Overbeeke et al., 1989).

View larger version (91K): [in this window] [in a new window]   Fig. 4. Amino acid sequence alignment of the tomato (Lycopersicon esculentum), coffee bean (Coffea arabica), guar (Cyamopsis tetragonolobus), soybean (Glycine max), Arabidopsis thaliana, antbush (Senna occidentalis), and pinto bean (Phaseolus vulgaris). The alignment was performed using MAP and shaded using Boxshade 3.21. Amino acids which are the same are shaded in black, while similar ones are shaded in grey. The beginning of the mature protein is marked with an arrow. Possible important catalytic residues are marked with an asterisk; horizontal bars indicate residues conserved in eukaryotic genera examined in the phylogenetic analysis.

 

View larger version (47K): [in this window] [in a new window]   Fig. 7. Isoelectric focusing of -galactosidase extracted from 24 h-imbibed tomato whole seeds, embryos and endosperms (micropylar and lateral). Proteins were transferred onto a nitrocellulose membrane and stained for enzyme activity with 6-bromo-2-naphthyl--D-galactopyranoside. Enzyme activity in gel slices (measured using p-nitrophenyl substrate) occurred in the same places as its activity appeared on the nitrocellulose membrane.

 
The presence of a signal peptide in the tomato -galactosidase may destine it for one of two locations: the cell wall region, where galactomannan is located, and/or protein bodies, where the enzyme has been found in soybean and lupin cotyledons (Plant and Moore, 1982; Herman and Shannon, 1985) and in the embryo and endosperm of data palm (Phoenix dactylifera) (Chandra Sekhar and DeMason, 1990). A current model for -galactosidase action has the enzyme synthesized during seed development and sequestered within protein bodies (Obendorf, 1997). Also at this time raffinose-series oligosaccharides and galactosyl cyclitols, substrates for -galactosidase, are synthesized and stored generally in the cytoplasm (Obendorf, 1997). In mature cotton and lupin seeds raffinose-type oligosaccharides are present mainly outside the protein bodies (Muller and Jacks, 1983; Plant and Moore, 1983). During germination, raffinose-type oligosaccharides are transported from the cytosol to the protein bodies, which have now become a vacuole for oligosaccharide hydrolysis (Obendorf, 1997). Whether this model of having -galactosidase active in protein bodies is widely applicable in germinating/growing seeds remains to be determined, but the presence of a signal peptide on the tomato -galactosidase does allow it to be transported into the ER, and thence to a vacuolar (protein body) or extracellular site.

Alignment of the tomato -galactosidase enzyme with other dicotyledonous -galactosidases shows that the protein is conserved among species (Fig. 4). Overall aa identities of 71.6, 69.2, 69.1, 67.3, 67.1, and 56.1% occur between the tomato -galactosidase and those from coffee bean (Zhu and Goldstein, 1994), pinto bean (Phaseolus vulgaris) (Davis et al., 1997), antbush (CAA03733), guar (Overbeeke et al., 1989), soybean (Glycine max) (Davis et al., 1996) and Arabidopsis (CAB87430), respectively. Coffee bean and guar -galactosidase share approximately 80% homology even though their signal peptides have little similarity (Zhu and Goldstein, 1994). This appears to be the case for all dicot -galactosidases: very high homology in the mature protein with apparent variation in signal peptides (when only the mature proteins are compared, similarities range from 78.8% for coffee and 62% for Arabidopsis) (Fig. 4). Through site-directed mutagenesis experiments, it has been shown that Trp-31 and Tyr-108 (61 and 138 in tomato) are critical for -galactosidase activity in coffee bean (Zhu et al., 1995, 1996). Additional site-directed mutagenesis experiments that substituted every other aa at Trp-31 revealed that only in the presence of aromatic rings (Phe and Tyr) was there a significant amount of enzyme activity. This supported the hypothesis of Mathew and Balasubramaniam (Mathew and Balasubramaniam, 1987), that Trp-31 affects the pK_a of a neighbouring carboxylic acid group (Maranville and Zhu, 2000). Asp-319 in guar (317 in tomato) has also been suggested to play a role in the active site of the enzyme (Overbeeke et al., 1989). All three of these aa, Asp-317, Trp-61, and Tyr-138 in tomato, are conserved in the dicotyledonous -galactosidases (Fig. 4) and all eukaryotic -galactosidases examined in the phylogenetic analysis (not shown).

Analysis of LeaGal and its dicot -galactosidase cohorts with other eukaryotic -galactosidases produced three distinct clades representing dicot, fungal, and metazoan taxonomic groups (Clades A, B, and C in Fig. 5). Further groupings within the dicot (A) and fungal (B) species, also seem to cluster based on taxonomy (Soltis and Soltis, 2000). For example, guar, pinto bean, antbush, and soybean are grouped into the family Fabaceae of the Rosidae. LeaGal groups closer to coffee since they both belong to the sub-class Asteridae. In the fungi, Z. cidri, T. delbruekii, and S. cerevisiae are yeasts (sub-group Hemiascomycetes of the Ascomycota) and belong to the family Saccharomycetaceae; T. reesei, P. simplicissimum, and A. niger are in the sub-group Euascomycetes of the Ascomycota, while S. pombe is grouped with the Archiascomycetes of the Ascomycota (Berbee and Taylor, 1999; Wheeler et al., 2000). M. vinaceae belongs to a different phylum, the Zygomycota (Fig. 5).

View larger version (48K): [in this window] [in a new window]   Fig. 5. Relationship of LeaGal to other reported eukaryotic -galactosidases. Phylogenetic analysis of 20 aa sequences lodged in GenBank produced three major clades representing plant (dicot), fungal and metazoan -galactosidases and, in general, inferred taxonomic relationships between the species. The tree was constructed using the distance approach and Fitch–Margoliash algorithm in PHYLIP (Fitch and Margoliash, 1967; Felsenstein 1993). Branch lengths represent genetic distances and are proportional to the number of amino acid substitutions.

 
Outside the dicot clade, the predicted tomato enzyme is closest to M. vinacea with an aa similarity of 42.3% (mature protein 45.1%) while that of the T. reesei is the least similar at 34% (35.8%) (E. coli outgroup 14.6%). In total, 40 conserved residues in the eukaryotic -galactosidases could be identified through sequence alignment, including a characteristic -galactosidase signature G-[LIVMFY]-x(2)-[LIVMFY]-x-[LIVM]-D-D-x-W-x(3,4)-R-[DNSF] (Hofman et al., 1999), suggesting these areas are important for enzyme function (Fig. 4; not shown).

View larger version (90K): [in this window] [in a new window]   Fig. 6. Southern blot analysis of DNA isolated from mature tomato leaves following digestion with several restriction enzymes and gel separation. Probed with -32P-labelled LeaGal -galactosidase cDNA.

 

Genomic sequence, Southern and northern blot analyses of -galactosidase
Southern hybridization analysis of tomato genomic DNA cut with several restriction enzymes indicated that there is only one gene present for tomato -galactosidase (Fig. 6). Screening of a genomic library with LeaGal cDNA as the probe yielded a positive genomic clone. Restriction digestion and Southern blot analysis showed it to be a 6.2 kb HindIII fragment. The fragment was subcloned into pBluescript SK (+) for sequencing (GenBank accession number AF289080). Sequence analysis of the 5.7 kb partial LeaGal gene sequence confirmed that the exon sequences were identical to the LeaGal cDNA sequence, as would be expected if there is a single gene for -galactosidase. The gene sequence lacks the codons for the 29 amino acids encoded by the 3' end, however, because of the presence of a HindIII site in the intron closest to the 3' end. The putative transcription and TATA box sites start at nucleotides 630 and 594, respectively, from the 5' end of the gene. The region encoding the mature enzyme consists of 14 exons interrupted by 14 introns, the exon boundaries being identified by reference to the LeaGal cDNA sequence. The introns range in size from 88 bp to 1281 bp.

The number of genes in other species has not been determined, nor is there any other genomic sequence available for the plant enzyme. However, in alfalfa, guar, carob (Ceratonia siliqua), soybean, date, coffee bean, pinto bean, lupin, and white clover (Trifolium repens), from which the enzyme has been purified, there are at least two isoforms present (Courtois and Petek, 1966; McCleary and Matheson, 1974; Williams et al., 1978; Itoh et al., 1979; Plant and Moore, 1982; Chandra Sekhar and DeMason, 1990; Dhar et al., 1994), although, with the exception of the date seed, lupin and soybean, their location within the seed has not been determined. The reported molecular masses of the isoforms are varied (17–57 kDa), but most approximate that of a 40 kDa mature enzyme as deduced from the cDNA sequences. The protein aggregates as tetramers in soybean, mung bean and lentil, with a molecular mass of 160 kDa (Del Campillo et al., 1981; Del Campillo and Shannon, 1982; Dey et al., 1983), even though the component polypeptides are active.

The -galactosidase enzyme from tomato has not been purified, but in the seeds alone there are two major isoforms (pI 4.6 and 4.8) and one minor (pI 5.0), all of which are present in the same proportions in both the endosperm and embryo (Fig. 7). Since there is only one gene for the enzyme, the isoforms apparently arise by post-translational modification, perhaps by changes at the C-terminus or variation in glycosylation. In the seed, both raffinose and galactomannan are present but their distribution between different tissues is not known (Groot et al., 1988; Hilhorst and Downie, 1996; Lin et al., 1998). Raffinose has been reported to be present in the embryo and galactomannans in the endosperm cell wall. At least one isoform could function in the embryo to hydrolyse raffinose which, in the whole seed at least, is known to decline following imbibition, while the other(s) could remove the unit galactose side chains from galactomannan in the endosperm.

The tomato -galactosidase cDNA was used for northern analyses to investigate gene expression in 24, 36, 48, 60, and 72 h imbibed seeds. Blots of the total RNA from the various seed parts during imbibition revealed a fairly constant expression of the -galactosidase gene (Fig. 8). -Galactosidase mRNA transcripts were lowest in the embryo (Fig. 8D), while expression was higher in the micropylar and lateral endosperm (Fig. 8B). Expression was similar in these two endosperm regions and increased slightly in both during the latter stages of imbibition, and especially at 72 h (Fig. 8B). Since -galactosidase activity is fairly constant during and following germination, presumably the transcripts are being used to replace enzyme that is being turned over. This may proceed at a faster pace in the endosperm regions as cell walls are being hydrolysed prior to the tissue itself being degraded.

View larger version (87K): [in this window] [in a new window]   Fig. 8. Northern blot analysis of total RNA isolated from the embryo, micropylar endosperm and lateral endosperm of tomato seeds at different times up to 72 h during and following germination. The blot (B, D) was probed with -32P-labelled LeaGal -galactosidase cDNA before exposure to X-ray film (5 h for the endosperm regions, 11 h for the embryo). Approximate equal loading of total RNA (20 µg) in each lane of an agarose gel is shown by ethidium bromide staining (A, C). WS, whole seed.

 
In conclusion, the cDNA for -galactosidase has been isolated from tomato seeds, and its genomic clone, and an extensive comparison with this enzyme has been conducted in other plants, as well as other eukaryotic genera. From a physiological standpoint, it is evident that -galactosidase in the tomato seed is present in the embryo and endosperm throughout germination. It is synthesized during seed development and is present in the mature dry seed; it does not change in activity in either the embryo or endosperm prior to or following radicle emergence. Its activity is also unaffected when radicle extension is inhibited, demonstrating that its activity is not contingent upon completion of germination. Similar high and constant -galactosidase activity is present in the embryo and endosperm of imbibed lettuce seeds, although a small increase is induced by red light (Leung and Bewley, 1981, 1983). In fenugreek, while -galactosidase activity is high in the embryo during and following germination, it is produced in, and secreted by, the aleurone layer of the endosperm following germination (Reid and Meier, 1973). -Galactosidase activity increases in intact seeds of several other legumes (tissue not specified), mostly following germination, at a time when both galactomannans and oligosaccharides are being mobilized (McCleary and Matheson, 1974). It is generally assumed that -galactosidase in embryos during, and perhaps following germination is involved in the hydrolysis of raffinose-series oligosaccharides as an early source of carbohydrate prior to the mobilization of the major hemicellulose or starch reserves (Bewley and Black, 1978), whereas that in the endosperm is involved in the hydrolysis of wall-associated galactomannan reserves (Bewley, 1997). In tomato seeds, -galactosidase is present before any increase in endo-ß-mannanase (Toorop et al., 1996) or ß-mannoside mannohydrolase (Feurtado, 1999) activity, and hence is available to participate in reserve mobilization in conjunction with these enzymes.

-Galactosidase activity is also present in ripening tomato fruits (Watkins et al., 1988; Feurtado, 1999). In addition, a search of the GenBank EST library revealed three overlapping EST clones isolated from tomato ovary tissue (GenBank accession numbers AI898528, AI899122 and AI484856) and two EST clones from tomato roots (BE451601, BF098394) that are almost identical to the LeaGal cDNA sequence, suggesting that the enzyme is present in many tissues. This is also consistent with the suggestion that -galactosidase is not specific to galactomannan or raffinose mobilization during germination and early seedling growth, but plays a wider role in the plant, hydrolysing terminal galactose residues in various tissues. In view of the different tissues and regions in which -galactosidase is active, and its different substrates, it is perhaps surprising there is only one gene for the tomato enzyme. However, there are at least three isoforms in the seed, allowing for isoforms with specificity for the oligomer-associated (cytoplasmic raffinose-series sugars) and polymer-associated (cell-wall galactomannans) substrates. The presence of a putative signal peptide is indicative that the enzyme enters the lumen of the endoplasmic reticulum during its synthesis, which is consistent with a requirement for packaging of the enzyme into secretory vesicles for it to reach the sites where its substrates are present, either the protein body or the cell wall.

	Acknowledgments

 
We are most grateful to Dr KJ Bradford, UC Davis for the tomato seed cDNA library, and Drs J Goldstein and A Zhu, New York Blood Center for the coffee bean -galactosidase cDNA. Sandy Reid and Sarah Knight provided valuable assistance during this study. This work was completed by JAF in partial fulfillment of his MSc thesis in the Department of Botany, University of Guelph during which the guidance of Dr WE Rauser was much appreciated. The work was supported by a Natural Sciences and Engineering Research Council of Canada grant A2210 to JDB.

	Notes

 
² To whom correspondence should be addressed. Fax: +1 519 767 1991. E-mail: dbewley{at}uoguelph.ca

¹ Present address: Department of Biological Sciences, Simon Fraser University, Burnaby, BC V5A 1S6, Canada.

	References

Top Abstract Introduction Materials and methods Results and discussion References

 
Adams MD et al. (196 co-authors). 2000. The genome sequence of Drosophila melanogaster. Science 287, 2185–2195.[Abstract/Free Full Text]

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. Journal of Molecular Biology 215, 403–410.[Web of Science][Medline]

Aslandis C, Schmid K, Schmitt R. 1989. Nucleotide sequences and operon structure of plasmid-borne genes mediating uptake and utilization of raffinose in Escherichia coli. Journal of Bacteriology 171, 6753–6763.[Abstract/Free Full Text]

Berbee ML, Taylor JW. 1999. Fungal phylogeny. In: Oliver RP, Schweizer M, eds. Molecular fungal biology. Cambridge, UK: Cambridge University Press, 21–77.

Bewley JD. 1997. Breaking down the walls—a role for endo-ß-mannanase in release from seed dormancy? Trends in Plant Science 2, 464–469.

Bewley JD, Banik M, Bourgault R, Feurtado JA, Toorop P, Hilhorst HWM. 2000. Endo-ß-mannanase activity increases in the skin and outer pericarp of tomato fruits during ripening. Journal of Experimental Botany 51, 529–538.[Abstract/Free Full Text]

Bewley JD, Black M. 1978. Physiology and biochemistry of seeds in relation to their germination, Vol. 1. Development, germination and growth. Berlin, Springer-Verlag.

Chandra Sekhar KN, DeMason DA. 1990. Identification and immunocytochemical localization of -galactosidase in resting and germinating date palm (Phoenix dactylifera L.) seeds. Planta 181, 53–61.

Courtois JE, Petek F. 1966. -Galactosidase from coffee beans. Methods in Enzymology 8, 565–670.

Davis MO, Hata DJ, Johnson SA, Jones DE, Harmata MA, Evans ML, Walker JC, Smith DS. 1997. Cloning, expression and characterization of a blood group B active recombinant -D-galactosidase from pinto bean (Phaseolus vulgaris). Biochemistry and Molecular International Biology 42, 453–467.

Davis MO, Hata DJ, Johnson SA, Walker JC, Smith DS. 1996. Cloning, expression and characterization of a blood group B active recombinant -D-galactosidase from soybean (Glycine max). Biochemistry and Molecular Biology International 39, 471–485.[Web of Science][Medline]

Del Campillo E, Shannon LM. 1982. An -galactosidase with hemagglutinin properties from soybean seeds. Plant Physiology 69, 628–631.[Abstract/Free Full Text]

Del Campillo E, Shannon LM, Hankins CN. 1981. Molecular properties of the enzyme phytohemagglutinin of mung beans. Journal of Biological Chemistry 256, 7177–7180.[Abstract/Free Full Text]

Denhardt DT. 1966. A membrane-filter technique for the detection of complementary DNA. Biochemical and Biophysical Research Communications 23, 641–646.[Web of Science][Medline]

de Vries RP, van der Broeck HC, Dekkers E, Manzanares P, de Graaff LH, Visser J. 1999. Differential expression of three -galactosidase genes and a single ß-galactosidase gene from Aspergillus niger. Applied and Environmental Microbiology 65, 2453–2460.[Abstract/Free Full Text]

Dey PM, Del Campillo EM, Lezica RP. 1983. Characterization of a glycoprotein -galactosidase from lentil seeds (Lens culinaris). Journal of Biological Chemistry 258, 923–929.[Abstract/Free Full Text]

Dhar M, Mitra M, Hata J, Butnariu O, Smith D. 1994. Purification and characterization of Phaseolus vulgaris -D-galactosidase isozymes. Biochemistry and Molecular Biology International 34, 1055–1062.[Web of Science][Medline]

Felsenstein J. 1993. PHYLIP (Phylogeny Inference Package) version 3.5c. (Distributed by the author.) Department of Genetics, University of Washington, Seattle, USA.

Feurtado JA. 1999. -Galactosidase and ß-mannoside mannohydrolase in tomato seeds (Lycopersicon esculentum Mill. cv. Glamour) and fruit (L. esculentum Mill. cv. Trust). MSc thesis, Department of Botany, University of Guelph, Canada.

Fitch WM, Margoliash E. 1967. Construction of phylogenetic trees. Science 155, 279–284.[Free Full Text]

Groot SPC, Karssen CM. 1987. Gibberellins regulate seed germination in tomato by endosperm weakening: a study with gibberellin-deficient mutants. Planta 171, 525–531.

Groot SPC, Kieliszewska-Rokicka B, Vermeer E, Karssen CM. 1988. Gibberellin-induced hydrolysis of endosperm cell walls in gibberellin-deficient tomato seeds prior to radicle protrusion. Planta 174, 500–504.[Web of Science]

Herman EM, Shannon LM. 1985. Accumulation and subcellular localization of -galactosidase-hemagglutinin in developing soybean cotyledons. Plant Physiology 77, 886–890.[Abstract/Free Full Text]

Hilhorst HWM, Downie B. 1996. Primary dormancy in tomato (Lycopersicon esculentum cv. Moneymaker): studies with the sitiens mutant. Journal of Experimental Botany 47, 89–97.

Hofmann K, Bucher P, Falquet L, Bairoch A. 1999. The PROSITE database, its status in 1999. Nucleic Acids Research 27, 215–219.[Abstract/Free Full Text]

Huang X. 1994. On global sequence alignment. Computer Applications in the Biosciences 10, 227–235.[Abstract/Free Full Text]

Itoh T, Shimura S, Adachi S. 1979. -Galactosidase from alfalfa seeds (Medicago sativa L.). Agricultural and Biological Chemistry 43, 1499–1504.

Kornreich R, Desnick RJ, Bishop DF. 1989. Nucleotide sequence of the human -galactosidase A gene. Nucleic Acids Research 17, 3301–3302.[Free Full Text]

Kyte J, Doolittle RF. 1982. A simple method for displaying the hydropathic character of a protein. Journal of Molecular Biology 157, 105–132.[Web of Science][Medline]

Leung DWM, Bewley JD. 1981. Red-light- and gibberellic-acid-enhanced -galactosidase activity in germinating lettuce seeds, cv. Grand Rapids. Control by the axis. Planta 152, 436–441.

Leung DWM, Bewley JD. 1983. A role for -galactosidase in the degradation of the endosperm cell walls of lettuce seeds, cv. Grand Rapids. Planta 157, 274–277.

Lin T-P, Yen W-L, Chien C-T. 1998. Disappearance of desiccation tolerance of imbibed crop seeds is not associated with the decline of oligosaccharides. Journal of Experimental Botany 49, 1203–1212.[Abstract/Free Full Text]

Maranville E, Zhu A. 2000. Assessment of amino-acid substitutions at tryptophan 16 in -galactosidase. European Journal of Biochemistry 267, 1495–1501.[Web of Science][Medline]

Mathew CD, Balasubramaniam K. 1987. Mechanism of action of -galactosidase. Phytochemistry 13, 1747–1757.

McCleary BV, Matheson NK. 1974. -D-Galactosidase activity and galactomannan and galactosylsucrose oligosaccharide in germinating legume seeds. Phytochemistry 13, 1747–1757.

McCleary BV, Matheson NK. 1975. Galactomannan structure and ß-mannanase and ß-mannosidase activity in germinating legume seeds. Phytochemistry 14, 1187–1194.

McIlvaine TC. 1921. A buffer solution for colorimetric comparison. Journal of Biological Chemistry 49, 183–186.[Free Full Text]

Michel BE, Kaufmann MR. 1973. The osmotic potential of polyethylene glycol 6000. Plant Physiology 51, 914–916.[Abstract/Free Full Text]

Muller LL, Jacks TJ. 1983. Intracellular distribution of free sugars in quiescent cottonseed. Plant Physiology 71, 703–704.[Abstract/Free Full Text]

Murray MG, Thompson WF. 1980. Rapid isolation of high molecular weight plant DNA. Nucleic Acids Research 8, 4321–4325.[Abstract/Free Full Text]

Nielsen H, Engelbrecht J, Brunak S, von Heijne G. 1997. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Engineering 10, 1–6.[Abstract/Free Full Text]

Nomaguchi M, Nongaki H, Morohashi Y. 1995. Development of galactomannan-hydrolyzing activity in the micropylar endosperm tip of the tomato seed prior to germination. Physiologia Plantarum 94, 105–109.

Nonogaki H, Matsushima H, Morohashi Y. 1992. Galactomannan hydrolyzing activity develops during priming in the micropylar endosperm tip of tomato seeds. Physiologia Plantarum 85, 167–172.

Obendorf RL. 1997. Oligosaccharides and galactosyl cyclitols in seed desiccation tolerance. Seed Science Research 7, 63–74.

Oda Y, Fukunaga M. 1999. Isolation and characterization of MELt gene from Torulaspora delbrueckii IFO 1255. Yeast 15, 1797–1801.[Web of Science][Medline]

Ohshima T, Murray GJ, Nagle JW, Quirk JM, Kraus MH, Barton NW, Brady RO, Kulkarni AB. 1995. Structural organization and expression of the mouse gene encoding -galactosidase A. Gene 166, 277–280.[Web of Science][Medline]

Overbeeke N, Fellinger AJ, Toonen MY, van Wassenaar D, Verrips CT. 1989. Cloning and nucleotide sequence of the -galactosidase cDNA from Cyamopsis tetragonoloba (guar). Plant Molecular Biology 13, 541–550.[Web of Science][Medline]

Plant AR, Moore KG. 1982. -D-Mannosidase and -D-galactosidase from protein bodies of Lupinus angustifolius cotyledons. Phytochemistry 21, 985–989.

Plant AR, Moore KG. 1983. The protein, lipid and carbohydrate composition of protein bodies from Lupinus angustifolius seeds. Phytochemistry 22, 2359–2363.

Reid JSG. 1985. Galactomannans. In: Dey PM, Dixon RA, eds. Biochemistry of storage carbohydrates in green plants. London: Academic Press, 265–288.

Reid JSG, Meier H. 1973. Enzymatic activities and galactomannan mobilization in germinating seeds of fenugreek (Trigonella foenum-graecum L. Leguminosae). Secretion of -galactosidase and ß-mannosidase by the aleurone layer. Planta 112, 301–308.

Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning—a laboratory manual, 2nd edn. New York: Cold Spring Harbor Laboratory Press.

Shibuya H, Kobayashi H, Sato T, Kim WS, Yoshida S, Kaneko S, Kasamo K, Kusakabe I. 1997. Purification, characterization and cDNA cloning of a novel -galactosidase from Mortierella vinacea. Bioscience, Biotechnology and Biochemistry 61, 592–598.

Smith RF, Wiese BA, Wojzynski MK, Davison DB, Worley KC. 1996. BCM search launcher—an integrated interface to molecular biology database search and analysis services available on the world wide web. Genome Research 6, 454–462.[Abstract/Free Full Text]

Soltis DE, Soltis PS. 2000. Contributions of plant molecular systematics to studies of molecular evolution. Plant Molecular Biology 42, 45–75.[Web of Science][Medline]

Still DW, Bradford KJ. 1997. Endo-ß-mannanase activity from individual tomato endosperm caps and radicle tips in relation to germination rates. Plant Physiology 113, 21–29.[Abstract]

Toorop PE, Bewley JD, Hilhorst HWM. 1996. Endo-ß-mannanase isoforms are present in the endosperm and embryo of tomato seeds, but are not essentially linked to the completion of germination. Planta 200, 153–158.

Turakainen H, Hankaanpaa M, Korhola M, Aho S. 1994. Characterization of MEL genes in the genus Zygosaccharomyces. Yeast 10, 733–745.[Web of Science][Medline]

Turakainen H, Korhola M, Aho S. 1991. Cloning, sequencing and chromosomal location of a MEL gene from Saccharomyces carlsbergensis NCYC396. Gene 101, 97–104.[Web of Science][Medline]

Verwoerd TC, Dekker BMM, Hoekma A. 1989. A small-scale procedure for the rapid isolation of plant RNAs. Nucleic Acids Research 17, 2362.[Free Full Text]

Voigt B, Bewley JD. 1996. Developing tomato seeds when removed from the fruit tissue produce multiple forms of germinative and post-germinative endo-ß-mannanase. Responses to desiccation, abscisic acid and osmoticum. Planta 200, 71–77.

Watkins CB, Haki JM, Frenkel C. 1988. Activities of polygalacturonase, -D-mannosidase and -D- and ß-D-galactosidases in ripening tomato. HortScience 23, 192–194.

Wheeler DL, Chappey C, Lash AE, Leipe DD, Madden TL, Schuler GD, Tatusova TA, Rapp BA. 2000. Database resources of the National Center for Biotechnology Information. Nucleic Acids Research 28, 10–14.[Abstract/Free Full Text]

Williams J, Villarroya H, Petek F. 1978. -Galactosidases II, III and IV from seeds of Trifolium repens. Purification, physiochemical properties and mode of galactomannan hydrolysis in vitro. Biochemical Journal 175, 1069–1077.[Web of Science][Medline]

Zhu A, Goldstein J. 1994. Cloning and functional expression of DNA encoding coffee bean -galactosidase. Gene 140, 227–231.[Web of Science][Medline]

Zhu A, Monohan C, Wang ZK. 1996. Trp-16 is essential for the activity of -galactosidase and -N-acetylgalactosaminidase. Biochimica et Biophysica Acta 1297, 99–104.[Medline]

Zhu A, Wang ZK, Goldstein J. 1995. Identification of tyrosine 108 in coffee bean -galactosidase as an essential residue for enzyme activity. Biochimica et Biophysica Acta 1247, 260–264.[Medline]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				A. Itai, K. Ishihara, and J. D. Bewley Characterization of expression, and cloning, of {beta}-D-xylosidase and {alpha}-L-arabinofuranosidase in developing and ripening tomato (Lycopersicon esculentum Mill.) fruit J. Exp. Bot., December 1, 2003; 54(393): 2615 - 2622. [Abstract] [Full Text] [PDF]

				B. Mo and J. D. Bewley The relationship between {beta}-mannosidase and endo-{beta}-mannanase activities in tomato seeds during and following germination: a comparison of seed populations and individual seeds J. Exp. Bot., November 1, 2003; 54(392): 2503 - 2510. [Abstract] [Full Text] [PDF]

				J. C. Pennycooke, M. L. Jones, and C. Stushnoff Down-Regulating {alpha}-Galactosidase Enhances Freezing Tolerance in Transgenic Petunia Plant Physiology, October 1, 2003; 133(2): 901 - 909. [Abstract] [Full Text] [PDF]

				C.-T. Wu and K. J. Bradford Class I Chitinase and {beta}-1,3-Glucanase Are Differentially Regulated by Wounding, Methyl Jasmonate, Ethylene, and Gibberellin in Tomato Seeds and Leaves Plant Physiology, September 1, 2003; 133(1): 263 - 273. [Abstract] [Full Text] [PDF]

				F. Chen, P. Dahal, and K. J. Bradford Two Tomato Expansin Genes Show Divergent Expression and Localization in Embryos during Seed Development and Germination Plant Physiology, November 1, 2001; 127(3): 928 - 936. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) E-letters: Submit a response Alert me when this article is cited Alert me when E-letters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (20) Request Permissions Disclaimer Google Scholar Articles by Feurtado, J. A. Articles by Bewley, J. D. Search for Related Content PubMed PubMed Citation Articles by Feurtado, J. A. Articles by Bewley, J. D. Agricola Articles by Feurtado, J. A. Articles by Bewley, J. D. Social Bookmarking          What's this?

Online ISSN 1460-2431 - Print ISSN 0022-0957

Copyright © 2005 Society for Experimental Biology

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics