Work

Polygalacturonases: Many Genes in Search of a Function1

Year: 1998
Type: review
Abstract: Pectins are a major component of the plant cell wall and comprise one of the two major coextensive networks in which cellulose microfibrils are embedded (Carpita and Gibeaut, 1993). Pectic polysaccharides exist in the cell wall as either “smooth” regions of a linear copolymer of α-(1–4)-linked GalUA or “hairy” regions that have attached α-(1–2)-linked rhamnosyl residues that may be substituted with Ara- and Gal-rich side chains. The pectin structure is further elaborated by divalent cation cross-linkages and possible esterification to other cell wall polymers. Because of the contribution of both ionic and covalent linkages, the structure of pectin may be modified by the ionic strength of the apoplast, by enzymes that modify the charge of the GalUA residues, or by enzymes that cleave either the α-(1–4)-linked GalUA backbone or side chains of the hairy pectin regions. Because plant cells undergo dramatic changes in shape and developmentally regulated episodes of cell separation, in which the pectin network is systematically disassembled, pectin metabolism is critical to many developmental processes. A wide range of enzymes are known to catalyze aspects of pectin modification and disassembly. The best characterized areexo- and endo-PGs, pectate lyase, pectin methylesterase, and β-galactosidase, which has been proposed to have the capacity to reduce the apparent molecular size of pectic polymers by cleaving neutral side chain residues (De Veau et al., 1993). In addition, it is likely that there are as-yet-undiscovered enzymes that may play critically important roles in cleaving covalent cross-linkages that tether pectins to other structural networks within the cell wall. Because of the extensive study of PG-mediated pectin disassembly, this review summarizes what is known about the complexity and structure of genes encoding plant PGs and their role in developmental processes. PGs were first identified over 35 years ago and have been suggested to be involved in the disassembly of pectin that accompanies many stages of plant development, particularly those that require cell separation. For example, PG activity has been shown to be associated with organ abscission (Taylor et al., 1990; Bonghi et al., 1992), pod and anther dehiscence (Meakin and Roberts, 1991), and pollen grain maturation and pollen tube growth (Pressey and Reger, 1989; Pressey, 1991). PG activity has also been detected in rapidly growing tissues, indicating that it may be involved in cell expansion (Pressey and Avants, 1977;Sitrit et al., 1996). Although it is clear that PG participates in many plant developmental processes, the majority of research has focused on PG in ripening fruit, abscission zones, or pollen. Molecular cloning and the modification of PG gene expression in transgenic plants have provided new insights into the physiological function of this enzyme. In addition, it is now clear that PGs are encoded by relatively large gene families in plants and that they are expressed in a wider range of developmental contexts than previously appreciated. Changes in cell wall structure are thought to underlie fruit softening, and ripening-associated cell wall disassembly has been examined in a number of fruit species, especially tomato (Lycopersicon esculentum) (Fischer and Bennett, 1991). Pectins, hemicelluloses, and possibly the amorphous regions of cellulose undergo structural modifications during ripening (Huber, 1983, 1984; O'Donoghue et al., 1994). Pectin disassembly is particularly extensive and is associated with the later stages of ripening and with fruit deterioration in the overripe stages (Dawson et al., 1992; Huber and O'Donoghue, 1993). We recently described the temporal sequence of cell wall polymer disassembly in very rapidly ripening melon (Cucumis melo cv Charentais). In this ripening system, the initiation and early stages of fruit softening were accompanied by a decrease in the molecular size of hemicellulosic polysaccharides, most notably of a tightly bound fraction of xyloglucan. Later in ripening, pectin disassembly was found to be extensive in the deteriorative overripe stages (Rose et al., 1998). These results suggest that fruit softening of Charentais melon is the consequence of the disassembly of two distinct networks, with the initiation of xyloglucan disassembly occurring earlier in ripening. Late softening and tissue deterioration are associated with extensive pectin disassembly. In addition, disassembly of pectins may increase the pore size of this network, resulting in the cell wall swelling that is seen in many fruit during the late stages of softening (Redgwell et al., 1997), or in increased accessibility of the substrate to enzymic action. Correlated with the solubilization and depolymerization of pectins in cv Charentais melon was an increase in pectin-degrading enzyme activity and the appearance of three PG mRNAs, one of which was demonstrated by heterologous expression in Aspergillus oryzae to encode an endo-PG (Hadfield et al., 1998). Overall, these results suggest that PG-mediated pectin disassembly occurs after the early stages of fruit softening and is likely to contribute significantly to the overripe stages of softening and deterioration. PG-dependent pectin disassembly has been most extensively studied in ripening tomato, and the development of molecular genetic techniques has provided a direct means of determining the contribution of PG activity to ripening-associated fruit softening. The suppression of PG gene expression in wild-type tomato and the ectopic expression of PG in the ripening-impaired pleiotropic mutant ripening inhibitor(rin) showed that PG-mediated pectin depolymerization was not necessary for normal ripening and softening (Sheehy et al., 1988;Smith et al., 1988; Giovannoni et al., 1989). Compressibility of transgenic tomato fruit with altered levels of PG gene expression. A, Compressibility of PG antisense (▪) and control (○) tomato fruits. Fruit were harvested at the breaker stage and the compressibility was determined at the time points indicated. The difference in compressibility is statistically significant at P < 0.001. Data are from Langley et al. (1994). B, Compressibility of wild-type (○), rin (▵), and transgenic rin expressing PG under control of the E8 promoter (▪) tomato fruit. Fruit were harvested 35 d after anthesis and exposed to propylene to induce the transgene, and compressibility was determined at the indicated time points. Graph was redrawn from Giovannoni et al. (1989). In a converse experiment, a chimeric gene consisting of the PG structural gene under control of the ethylene-inducible promoter E8 was introduced into mutant rin tomato fruit that normally fail to ripen or soften and that do not express PG. In the transgenic fruit, PG gene expression, enzyme activity, and pectin solubilization and depolymerization were restored to near wild-type levels. However, these fruit did not soften and were not altered in other ripening parameters such as pigment accumulation or ethylene production, indicating that PG-mediated pectin disassembly is not sufficient for normal ripening or softening to occur (Fig. 1B) (Giovannoni et al., 1989). However, the PG-complemented rin fruit were more susceptible to pathogen attack than were control fruit (Bennett et al., 1993). Similarly, PG antisense fruit showed a decrease in pathogen susceptibility compared with control fruit (Kramer et al., 1992). Effect of antisense PG gene expression on overripe tomato fruit. Left, Control fruit expressing normal levels of PG; right, PG antisense fruit expressing < 1% wild-type level of PG. Control and antisense fruit were harvested on the same day and stored until control fruit advanced to the overripe stage and showed visible signs of deterioration. Photo courtesy of Calgene, Inc. Analysis of cell walls from transgenic fruit with altered levels of PG activity have also elucidated the mechanisms of pectin disassembly in vivo. The suppression of pectin depolymerization but not of pectin solubilization in transgenic PG antisense fruit indicates that these two processes have distinct enzymic determinants, with depolymerization but not solubilization being primarily due to the action of PG. In contrast, pectins in PG-complemented rin fruit were both solubilized and depolymerized. Taken together, the data suggest that PG activity is both necessary and sufficient for pectin depolymerization but that it may be one of multiple, redundant pectin-solubilizing activities, all of which are suppressed in the rin mutant. PG activity is also high in fruit other than tomato, such as avocado (Persea americana) and peach (Prunus persica). In peach three distinct activities have been identified, two of which cleave the substrate by an exo mode of action, and one that hydrolyzes the pectin backbone in an endofashion (Downs et al., 1992). In contrast to tomato, in whichexo-PG activity is present during early fruit development and remains constant during ripening, exo-PG activities in peach are ripening regulated (Pressey, 1987; Downs et al., 1992).Endo-PG activity in peach is associated with the freestone character, and a genetic linkage between freestone andendo-PG has recently been identified (Lester et al., 1996). Avocado fruit have high levels of PG activity that are temporally correlated to solubilization and depolymerization of polyuronides during fruit ripening (Huber and O'Donoghue, 1993). In contrast, a number of fruit have been reported to lack endo-PG activity, including strawberry (Fragaria ananassa) (Huber, 1984), apple (Malus domestica) (Bartley, 1978), persimmon (Diospyros kaki) (Cutillas-Iturralde et al., 1993), and melon (McCollum et al., 1989). In many of these fruit, ripening-associated pectin depolymerization does not occur, and pectin solubilization may be catalyzed by other enzymes. Alternatively, PG activity may be present but at as-yet-undetectable levels. For example, apple pectins are disassembled at the late stages of ripening (Knee, 1973), and recently, under rigorous examination, endo-PG activity and PG mRNA accumulation have been detected but the levels are much lower than that observed in tomato (Wu et al., 1993). In strawberry, three different PG activities were detected and partially purified, two of which were exo-PGs and one of which was endo acting (Nogata et al., 1993). In melon (McCollum et al., 1989; Rose et al., 1998) and persimmon (Cutillas-Iturralde et al., 1993), pectins are extensively depolymerized during ripening but PG activity is low or undetectable. It has been suggested that in melon, pectin depolymerization results from β-galactosidase activity because cell wall pectins extracted from preripe fruit undergo a downshift inMr when treated with a partially purified β-galactosidase extract (Ranwala et al., 1992). However, the characterization of ripening-regulated mRNAs that encode functional PGs in melon suggests that pectin disassembly during the late stages of fruit softening may be at least in part PG dependent (Hadfield et al., 1998). It is important to note that even very low levels of PG may be sufficient to catalyze extensive pectin disassembly. This became apparent in transgenic antisense tomato fruit, in which reductions in PG activity of 80% had little impact on pectin structure, indicating that in tomato PG is present in at least 5-fold excess (Smith et al., 1990). It is therefore possible that the disassembly of pectin is PG dependent, even in fruit with very low levels of PG activity. During plant development various organs undergo programmed senescence and are shed from the parent plant (Hadfield and Bennett, 1997). Abscission is the consequence of the dissolution of cell walls between adjacent cells in the separation layer consisting of 5 to 50 rows of small, isodiametric cells, with the fracture usually occurring in the plane of the middle lamella. The cells flanking the fracture plane remain largely intact after abscission has taken place, except for the nonliving cells of the vascular system, which are ruptured due to physical forces (Sexton and Roberts, 1982). During separation pectins are solubilized, the middle lamella swells, fenestrates, and disappears, and the microfibrillar network becomes disorganized. During abscission there is an increase in cell secretion, presumably to accommodate the increase in secretion of cell wall-modifying enzymes. Inhibitors of protein and RNA synthesis markedly delay abscission, implying that de novo protein and RNA synthesis are required for cell separation (Sexton and Roberts, 1982). EGase activity increases dramatically during abscission, and multiple EGase genes are expressed in the abscission region (Lashbrook et al., 1994; Del Campillo and Bennett, 1996). Both the activity of EGase and the rate of abscission are suppressed by auxin and accelerated by ethylene, supporting a role for EGase in the abscission process. Antisense suppression of one EGase isoform resulted in a partial reduction in flower abscission (Lashbrook et al., 1998). PG activity also increases during abscission of both tomato flower andSambucus nigra leaf explants (Tucker et al., 1984; Taylor et al., 1993). In contrast to EGase activity, which is distributed throughout the abscission zone, PG activity is restricted to the distal portion of the zone tissue (Roberts et al., 1989). PG activity has also been detected in flower and fruit abscission zones in a variety of species, but the relative predominance of EGase or PG varies depending on the subtending organ. For example, in peach EGase activity is high in leaf-abscission zones but much lower in fruit-abscission zones, whereas PG activity is not detected in leaf-abscission zones but is present at very high levels in fruit-abscission zones (Bonghi et al., 1992). The difference in hydrolytic activities between the two types of abscission zones is of potential practical use because it may allow genetic alteration of abscission of specific plant organs. In tomato a fruit PG-specific antibody and a cDNA corresponding to the ripening-regulated gene did not react with protein or mRNA isolated from leaf-abscission zones. In addition, PG activity in abscission zones was unaffected by the presence of an antisense fruit PG gene that suppressed PG activity in fruit by 99% (Taylor et al., 1990). These results strongly suggested that the PG in leaf-abscission zones is distinct from the fruit-ripening-associated PG. This has been confirmed by the recent identification of three PG genes that are expressed in both tomato leaf- and flower-abscission zones, and that are divergent from the fruit-ripening-associated PG gene (Kalaitzis et al., 1997). The three abscission PGs differ from each other in sequence and in their temporal patterns of expression. TAPG4 is divergent from and expressed much earlier than either TAPG1 or TAPG2. It is possible that the divergent PGs have different substrate specificities or a different mode of action, i.e. endo- versus exo-cleavage, or they may correspond to redundant activities under the control of different regulatory factors. Pod dehiscence is similar to abscission in that it occurs at a precise location in the pod and results from cell wall disassembly and separation of a specialized layer of unlignified cells called the dehiscence zone. Like abscission, the middle lamella is broken down and pectin is lost during cell separation, and the process is accompanied by an increase in EGase activity (Meakin and Roberts, 1991). Although PG enzyme activity has not been definitively demonstrated in dehiscence zones, cDNAs have recently been identified that encode putative PGs expressed specifically in the dehiscence zone during the cell-separation process, suggesting a role for PG in pod dehiscence (Jenkins et al., 1996; Petersen et al., 1996). Germination and growth of the pollen tube through the pistil occurs rapidly, and many of the biochemical events that occur during pollen maturation prepare the pollen for this coming developmental event (Mascarenhas, 1990). High levels of exo-PG activity have been detected in the pollen of many species, including maize (Zea mays) and other grasses (Pressey and Reger, 1989) and a large number of trees (Pressey, 1991). The activity measured in pollen appears to be exclusively exo-PG; the pollen of some species contain extraordinarily high levels of exo-PG activity, with Eastern cottonwood (Populus deltoides) pollen having levels of PG activity 12 times higher than that of tomato fruit (Pressey, 1991). The functional role of PG in pollen tubes may be to degrade the walls of the stylar cells to allow penetration of the pollen tube or to provide wall precursors for tube growth. Alternatively, pollen tube PG may be acting on its own wall to facilitate growth (Brown and Crouch, 1990). The addition of pectinase to the growth medium of pollen tubes germinating in vitro stimulates their growth (Mascarenhas, 1990). It is possible that oligogalacturonide signaling is involved in the guidance of the pollen tube to the ovule, and that exo-PG may facilitate the rapid turnover of oligogalacturonide-signaling by to a size that is as suggested for other and In purified cottonwood PG were for with a of of (Pressey, 1991), which is in the range of for that as in other et al., 1992). A number of genes are expressed in pollen and have been on the of expression. genes are expressed the first and late genes are expressed The encoded by early genes are to be and involved in cell wall synthesis and whereas late genes may encode for pollen maturation and pollen tube growth (Mascarenhas, 1990). In many species, including and cDNAs with sequence to PG have been identified and as late genes (Brown and Crouch, 1990; and 1992; et al., et al., 1994). The mRNA of pollen PGs is in germinating and growing pollen the anther PG expression was detected to the suggesting its in anther dehiscence et al., 1993). In also detected the presence of PG mRNA in the and in some cells of the to pollen and in the anther and of maize at a later stage of of PG in maize to maize exo-PG the presence in the walls of cells modifications such as in the or expansion in the growing pollen tubes et al., 1993), indicating roles in pollen development and in Although the expression and activity of PG is high in pollen and fruit, it is also found throughout the suggesting that PG may have a For example, PG activity and mRNA accumulation have been detected in germinating and (Pressey and Avants, et al., 1996). In the level of exo-PG activity was during the most rapid of in the most rapidly growing region of the PG activity has also been detected in other growing plant tissues, including and tomato and and the of and (Pressey and Avants, In germinating tomato an mRNA encoding a PG was to the of the in the vascular tissue of the et al., 1996). The of PG mRNA and protein to the vascular in a number of growing and 1992; et al., et al., suggests that PG may be involved in and disassembly of the cell wall. PG activities associated with distinct of plant development are encoded by families with regulated in of their and temporal expression. It is also possible that divergent PG gene encode enzymes that may differ in their biochemical such as mode of or substrate genes encoding PG have been described in a number of species, including tomato et al., et al., 1997), melon (Hadfield et al., et al., et al., maize and 1992), peach (Lester et al., (Brown and Crouch, and et al., suggesting that families encoding PGs are in the plant The of PG gene are relatively divergent from one For example, the sequence of the tomato fruit PG is to a zone but to a melon fruit PG indicating that the sequence of the PG gene families to the of the major many of the PG a relatively high level of over their there are regions that are all of the from plants and particularly in the portion of the and have been to some residues within this region et al., 1990; et al., 1996; et al., 1996). of these sequence have as the for to PG gene families from plant species by The presence of three in the part of PG have by the of a that a that is all PGs to and be to the sequence a PG (Lester et al., 1994; et al., 1996; Petersen et al., 1996; et al., 1998). This has resulted in the identification of families with distinct temporal and patterns of gene expression. PG genes are expressed in pollen and their mRNAs are to each In pollen PGs are more than 99% to each with occurring primarily in the regions and 1992), suggesting that the activities of PGs expressed in maize pollen are PG genes presumably encode for this is and is on the to exo-PG activity in pollen A number of of the are pollen and it has been suggested that these a functional of an exo mode of PG action et al., 1994). Although it is to use these sequence to PGs as be until more rigorous biochemical is to the activity corresponding to PGs of This require heterologous expression of plant PG cDNAs in a that does not express The PG protein from tomato fruit extensive and including of the and an adjacent to the and Bennett, and at the (Sheehy et al., of the PGs to have a sequence that the protein to the of the and presumably to the cell wall. The presence of an appears to be present in one The function of the is not known but may the protein in an it is being to its or it may the protein to a specific location within the cell wall and Bennett, of the of the three major PG et al., A and be on expression patterns because both genes that are expressed during fruit ripening and during abscission or The presence of a these two with all of having a of A are to encode a of to the results in the of an with the same three indicating that the is not the for the of these into their own The functional of the of PGs into three major is and as more is the and expression of PGs in a variety of developmental the functional of these Although PGs have been studied in to fruit ripening, the of large families encoding PGs that are expressed in a wide range of different and developmental stages as much more than enzymes. wall modifications are associated with stage of development, and it is that there is a of cell enzymes that are expressed in a number of these including fruit ripening, and cell The of multiple genes to similar in each developmental the for of gene expression by a number of developmental and and for of the biochemical function of each distinct gene We are to Rose for critical of this and to for
Language: English
Subfield: Plant Science
Open Access status: green
OpenAlex ID: W2057046266
FWCI: 1.52
Cites: 62
Cited by: 461
Related to: 10