a4-Fucosyltransferase is regulated during flower development ...

Download PDF
1 downloads 0 Views 197KB Size Report
Comment
GlcNAc in a(1,3)-, a(1,4)-linkages (Staudacher et al.,. 1999). Xyloglucan ... by b3-galactosyltransferase and a4-fucosyltransferase onto GlcNAc residues of ...
Journal of Experimental Botany, Vol. 53, No. 373, pp. 1429– 1436, June 2002

a4-Fucosyltransferase is regulated during flower development: increases in activity are targeted to pollen maturation and pollen tube elongation Caroline Joly, Renaud Le´onard, Abderrahman Maftah and Catherine Riou-Khamlichi1 Glycobiologie et Biotechnologie EA3176, Institut des Sciences de la Vie et de la Sante´, Universite´ de Limoges, 123 avenue Albert Thomas, 87060 Limoges Cedex, France Received 22 November 2001; Accepted 11 February 2002

Abstract a4-Fucosylation represents a final step of protein N-glycosylation. a4-fucosylated N-glycans are thought to be involved in cell-to-cell communication and recognition in primates and plants. Nevertheless, in the plant life cycle, the function of a4-fucosylation remains largely unknown. To gain an insight into the role of a4-fucosylation during development, the study focused on tobacco flowers. It is shown that an increase in a(1,4)fucosyltransferase (Fuc-T) activity is only observed during anther development, whereas it remains at a constant but low level (around 20 pmol Fuc h 1 mg 1 protein) in the gynoecium and perianth. At least a 4-fold higher activity is detected in mature pollen grains. These data suggest that a(1,4)Fuc-T activity is regulated during anther development. Furthermore, a(1,4)Fuc-T activity could be required during pollen tube elongation where the activity level peaks at 350 pmol h 1 mg 1 protein. Based on enzyme profile and cycloheximide effects on pollen germination and activity, it is hypothesized that the gene encoding a4-Fuc-T could be regulated late during pollen development. A potential role of a4-fucosylation during pollen tube elongation is also discussed. Key words: Elongation, fucosylation, pollen, tobacco.

Introduction In both plant and mammalian cells, N-glycosylation is involved in protein maturation. Processing of N-linked glycans occurs along the secretory pathway from the 1

endoplasmic reticulum through the Golgi apparatus. N-glycans seem to be involved in the correct folding of polypeptides, stability and transport to their proper destination (Hammond and Helenius, 1995). Arabidopsis mutants deficient in correct N-glycosylation patterns and, therefore, lacking complex N-glycans are mostly embryo lethal, unable to develop normally or fail to breed (von Schaewen et al., 1993; Boisson et al., 2001; Lukowitz et al., 2001). Also, N-glycans might also be involved in programmed cell death (Lindholm et al., 2000). Another critical aspect of plant N-glycans is their potential involvement in allergenic immune responses in mammals when complex carbohydrate structures are localized on pollen grains or in foodstuffs (GarciaCasado et al., 1996; Wilson et al., 2001). In plants and mammals, fucosyltransferases (Fuc-T) are a specific group among the large Golgi-resident glycosyltransferase families. They are mainly involved in the maturation of complex type N-glycans. Plant Fuc-T catalyse the transfer of fucose from GDP-fucose to GlcNAc in a(1,3)-, a(1,4)-linkages (Staudacher et al., 1999). Xyloglucan a2-Fuc-T exist in plants and catalyse transfer of fucose to galactose in xyloglucans which are involved in the polysaccharide–cellulose network of primary cell walls (Faik et al., 2000). The Lewisa motif is synthesized by transfer of galactose and fucose residues by b3-galactosyltransferase and a4-fucosyltransferase onto GlcNAc residues of terminal lactosamine, respectively. The Lewisa motif synthesized by the a4-Fuc-T has been found in unrelated species across the plant kingdom. Nevertheless, ambiguity about the presence of Lewisa still remains for the Cruciferae family, where the Lewisa motif was not detected in Arabidopsis plants, but was characterized in papaya fruit (Fitchette et al., 1999; Wilson et al., 2001). In animals, this motif seems to be exclusively

To whom correspondence should be addressed. Fax: q33 555 457 653. E-mail: [email protected]

ß Society for Experimental Biology 2002

1430

Joly et al.

present in primates (Blancher and Socha, 1997). In humans, Lewisa structures are involved in blood group determinism and in cell-to-cell interactions (Staudacher et al., 1999). In plants, a4-fucosylation may be involved in cell-to-cell communication anduor recognition (FitchetteLaine´ et al., 1997; Melo et al., 1997). Several immunolocalization studies revealed that this glycotope was associated with extracellular soluble or membrane bound-glycoproteins, but these glycoproteins were not fully characterized (Fitchette et al., 1999). Therefore, Lewisa-containing glycans could be considered as markers of protein transport (Fitchette et al., 1999). However, all these studies were mainly performed on cell suspension cultures or root tips and not during a specific plant developmental programme. In flowering plants, pollen development represents a good model to study cell communication and signalling as well as cell elongation. Two categories of genes encoding pollen-specific proteins have been described according to their expression patterns during pollen development. Early genes are only expressed during microsporogenesis and might encode structural proteins (Mascarenhas, 1990, 1993). Late genes are expressed after microspore first mitosis during pollen maturation and pollen tube elongation. The regulation of late genes suggests that pollen late proteins might be key components involved in pollen germination anduor during pollen tube elongation (Mascarenhas, 1990, 1993). So far, several pollen late proteins are potentially N-glycosylated (Rogers et al., 1992; Tebbutt et al., 1994; Ogawa et al., 1996; Witting et al., 2000). Moreover, glycosylation seems to play a key role during pollen tube growth. An inhibitor of glycosylation, tunicamycin, which blocks the connection of glycan chains to peptides, inhibited pollen tube growth, suggesting an involvement of N-glycosylated proteins in the elongation process (Capkova et al., 1997). Only one N-glycoprotein Cry j I from japanese cedar (Cryptomeria japonica) pollen is a4-fucosylated (Ogawa et al., 1996). This protein has a pectate lyase enzyme activity and could be involved in cell wall elaboration during germination and pollen tube growth (Taniguchi et al., 1995). a4-Fucosylation could be involved in male gametophyte development, which represents a particular aspect of plant reproductive development. It could also contribute to biological activity anduor the correct targeting of proteins. In order to define the role(s) of a4-fucosylation during reproductive development, a(1,4)Fuc-T activity was analysed during tobacco flower development. It was shown that a(1,4)Fuc-T activity is developmentally regulated with peaks of activity, firstly, during pollen maturation and, secondly, during pollen tube elongation. a4-Fucosylation could be involved in male gametophyte development which represents a particular aspect of plant reproductive development.

Materials and methods Plant material

Tobacco (Nicotiana tabacum L. var Xanthi) plants were grown in a greenhouse at 25 8C with a 16 h photoperiod. Microspores and pollen grains were observed under light microscopy. Their developmental stages were chosen on morphological criteria such as diameter. Diameter measurement was performed using Adobe PhotoShop 5 software from pictures.

In vivo pollen tube growth assay

A 46 mm unopened flower was used for hand-pollination experiments. Its anthers were cut off and 48 h later the emasculated flower was fully opened and hand-pollinated on its exudating stigma. Pollen grains germinate on the stigma and pollen tubes elongate firstly into the stigma and thus, into the style.

In vitro pollen tube growth Anthers were isolated from flowers at anthesis and pollen grains were collected. Pollen grains germination occured in liquid germination medium (3 mM HBO3, 1.7 mM Ca(NO3)2, 10% sucrose, pH 6.6), in darkness at 25 8C. Cycloheximide was purchased from Sigma (St Louis, MO).

Protein extraction

Plant tissues were ground in liquid nitrogen. The resulting powder was placed into an extraction buffer (20 mM MES, 0.1% Triton X100, 15% glycerol), vigorously vortexed and centrifuged at 12 000 g for 3 min. Anthers were wounded in order to release microspores or pollen grains which were transferred to the extraction buffer described above. Homogenates were vortexed vigorously followed by several rounds of freezing in liquid nitrogen and then centrifuged at 4 8C at 12 000 g for 2 min. The protein content of supernatants was determined using the Bio-Rad reagent (Bio-Rad, Hercules, CA) and BSA as a standard (Bradford, 1976). The same protein supernatants were used for in vitro enzymatic assay of a(1,4)fucosyltransferase and dot blot analyses. Pollen protein extraction was prepared as described previously (Capkova et al., 1994). Pollen grains were homogenized in 50 mM TRIS-HCl pH 6.8, 10% sucrose and 1% (vuv) mercaptoethanol buffer. The soluble cytosolic proteins were obtained by centrifugation at 13 000 g for 15 min at 4 8C. Wall and membrane-bound proteins were extracted from the pellet with the previous buffer supplemented with 1% SDS. Extraction of pollen coat proteins was performed as described earlier using diethyl ether as the organic solvent (Yih Bih et al., 1999).

In vitro enzymatic assay of a4-fucosyltransferase Activity was performed on crude extracts for 2 h at 37 8C in a 60 ml volume reaction containing 25 mM sodium cacodylate (pH 6.5), 5 mM ATP, 20 mM MnCl2, 10 mM a-L-fucose, 3 mM GDP w14Cx-fucose (310 mCi mmol 1; Amersham Pharmacia Biotech, UK), 0.1 mM type I N-acetyl-lactosamine acceptor, and 15 mg protein. In crude plant extracts, only type I acceptor reveals a(1,4)Fuc-T activity (Faik et al., 2000). The reaction was stopped by adding 3 ml water and the mixture was then filtered on a hydrophobic Sep-Pak C18 cartridge (Waters Millipore,

a4-Fucosyltransferase is regulated during pollen development

1431

Bedford, MA) that retains the acceptor (either fucosylated or not). Free GDP w14Cx-fucose was washed off with 10 ml of water. The acceptor was eluted with 5 ml of ethanol, collected into scintillation vials and the fucosylated reaction product was quantified after addition of 2 vols of Biodegradable Counting Scintillant (Amersham Pharmacia Biotech, UK) by a liquid scintillation beta counter (Liquid scintillation analyser, Tri-Carb-2100TR, Packard). Presence of Lewisa motifs by dot blot analysis

The blots were performed by using a chemiluminescence blotting substrate (POD) system according to the manufacturer’s instructions (Boeringher Mannheim). Proteins were deposited onto a nitrocellulose membrane (Hybond-C extra membrane, Amersham, UK). After blocking in 1% blocking solution (vuv) containing (50 mM TRIS base, pH 7.5, 150 mM NaCl) TRIS-buffered saline TBS solution for 2 h at room temperature, the membrane was incubated with rabbit primary anti-sycamore laccase Lewisa antibody (generous gift from Dr L Faye, Mont St Aignan, France) in a 1 : 2000 dilution for 1 h at room temperature. It was then washed twice into TBS buffer containing 0.1% Tween 20 followed by two washes into 0.5% blocking solution (vuv). The membrane was incubated with a 1 : 1000 dilution secondary antibody conjugated with horseradish peroxidase (Dako, Denmark) and thoroughly washed in TBS buffer containing 0.1% Tween 20. The immunoreactive proteins were revealed by a chemiluminescence reaction. The membrane was then exposed to X-ray films (Biomax Film, Kodak) for different exposure times. Human saliva was used as a positive control and bovine proteins as the negative control (Yazawa and Furukawa, 1980; Oulmouden et al., 1997).

Results a(1,4)Fuc-T activity is detected at a low level in tobacco flowers

a(1,4)Fuc-T activity was determined in sepals, petals, stamens, and pistils taken at three stages of flower development based on bud length (Fig. 1). A relatively constant basal a(1,4)Fuc-T activity (approximately 20 pmol Fuc h 1 mg 1 protein) was detected in sepals, petals and non-pollinated pistils isolated from very young flower buds to opened flowers (Fig. 1). Therefore, in growing tissues, a4-fucosyltransferase seems to function as a housekeeping enzyme involved in N-glycoproteins maturation and in primary metabolism. Nevertheless, a 3-fold increase in a(1,4)Fuc-T activity was observed in stamens at anthesis (Fig. 1). a(1,4)Fuc-T activity specifically increases in anthers

Five successive stages based on flower bud length are described: stage A (11–20 mm), stage B (22–39 mm), stage C (43–45 mm), stage D (46 mm), and stage E (46 mm). By contrast to stage D, stage E corresponds to the fully opened flower. In order to define these particular stages precisely, anthers and pistils were also weighed fresh and measured. These criteria were also confirmed by morphological data described previously (Koltunow et al.,

Fig. 1. a(1,4)Fuc-T activity in tobacco flower organs. a(1,4)Fuc-T activity was determined in sepals, petals, stamens, and pistils taken at three stages of flower development. Each stage is characterized by the flower bud length: 12, 28 and 46 mm, respectively. Each point corresponds to an average of three independent experiments "SD. Three flower buds were used to perform one experiment. Scale bar: 10 mm.

1990). These stages correspond to cellular events occurring during male and female sporophyte and gametophyte development (Koltunow et al., 1990; De Martinis and Mariani, 1999). a(1,4)Fuc-T activity was thus more accurately followed during anthers and pistils development (Fig. 2A, B). In anthers, a weak a(1,4)Fuc-T activity was detected at stage A (approximately 15 pmol Fuc h 1 mg 1 protein). Two significative increases in a(1,4)Fuc-T activity were measured during anther development (Fig. 2A). The first activity burst between stages A and B was concomitant with an increase in anther weight. It corresponded to the degradation of connective tissue in the stomium region and the beginning of microgametogenesis. The second burst was observed between stages C and D (around 90 pmol Fuc h 1 mg 1 protein) (Fig. 2A). After stage D, the loss of anther weight was linked to its dehiscence (Fig. 2A). For stages D and E, a basal activity was measured in anthers empty of pollen grains (approximately 20 pmol Fuc h 1 mg 1 protein). During ovule development (stages A–C) and megagametogenesis (stage D), a basal a(1,4)Fuc-T activity was detected (approximately 15 pmol Fuc h 1 mg 1 protein) while pistil continuously gained in weight (Fig. 2B). By contrast to the results shown in Fig. 1, in fully opened flowers, a 3-fold increase in activity was measured in the pollinated pistil (Fig. 2B).

1432

Joly et al. Table 1. Measurements of a(1,4)Fuc-T activity level during pollen development Diameter was determined on 300 microspores and pollen grains as described in Materials and methods. a(1,4)Fuc-T activity was performed on microspores or pollen grains isolated from 15 anthers. Results are the means"SD of three independent experiments.

Free microspores Binucleate microspores Mature pollen grains

Fig. 2. a(1,4)Fuc-T activity in sexual organs. Flower development stages are based on bud length: stage A: 11–20 mm, stage B: 22–39 mm, stage C: 43–45 mm, stage D: 46 mm, stage E: 46 mm. In contrast to stage D, stage E corresponds to fully opened flowers. (A) a(1,4)Fuc-T activity during in anther developement. For each stage, fresh weight was determined for five anthers. (B) a(1,4)Fuc-T activity in pistil development. For each stage, fresh weight was determined for one pistil. a(1,4)Fuc-T activity assays were performed on anthers and pistils taken from three flower buds. For each stage, activity and weight were determined three times independently and SD was calculated.

Regarding the high activity detected at stage E in both anthers and pollinated pistils, it is suggested that a(1,4)Fuc-T activity is mainly associated with pollen development. a(1,4)Fuc-T activity specifically increases during late pollen maturation

Isolated microspores, pollen grains and the corresponding sporophytic tissues were isolated from stages A to E. A basal a(1,4)Fuc-T activity was detected in free microspores (Table 1). Activity clearly increased in binucleate microspores and reached a maximal level in mature pollen grains (approximately 120 pmol Fuc h 1 mg 1 protein) (Table 1). In order to correlate this high activity to the presence of Lewisa motifs in pollen proteins, dot blot analyses from free and binucleate microspores and mature pollen grains

Diameter (mm)

a(1,4)Fuc-T activity (pmol Fuc h 1 mg 1 protein)

22.85"1.58 29.65"1.47 38.40"2.25

34.5"4.1 77.3"4.9 120.1"6.5

Fig. 3. Immunodetection of Lewisa motifs on microspores and pollen grains. (A) Lewisa epitopes during microgametogenesis. Proteins were isolated from free microspores (lane 3), binucleate microspores (lane 4) and mature pollen grains (lane 5). (B) Distribution of Lewisa epitopes in pollen grains. Lanes 3 to 5 correspond to cytosolic, membrane-bound and pollen coat proteins, respectively. For each dot blot, bovine protein extract was loaded as a negative control (lane 1) and human saliva as a positive control (lane 2). 250 ng of proteins were loaded per lane. Anti-plant Lewisa antibodies were revealed by peroxidase assay as described in Materials and methods. The exposure time for autoradiography was 30 s.

were performed (Fig. 3A). Increased amounts of Lewisa motifs were detected from free microspores to mature pollen grains (Fig. 3A). Since the same protein amounts were loaded, it can be concluded that the appearance of Lewisa structures was clearly correlated with the increase in a(1,4)Fuc-T activity during pollen development. In addition, a dot blot analysis was performed on different fractions of pollen proteins in order to localize Lewisa epitopes on mature pollen grains (Fig. 3B). Although Lewisa motifs were found in the cytosol, higher amounts of this motif were detected in the pollen coat, wall and membranes suggesting a role for a4-fucosylated proteins during the establishment of these structures.

a4-Fucosyltransferase is regulated during pollen development

A specific increase in a(1,4)Fuc-T activity is observed during pollen germination and pollen tube elongation

In vivo germination pollen assays were performed on hand-pollinated stigmas. a(1,4)Fuc-T activity was then followed in non-pollinated or hand-pollinated stigmas. A relatively constant and basal activity was always detected in non-pollinated stigmas and styles (Fig. 4). By contrast, a significative increase in activity was observed in these tissues after pollination (3-fold after 2 h and 5-fold after 5 h) (Fig. 4). Both results strongly suggest that an increase in a(1,4)Fuc-T activity is linked to germination events and pollen tube elongation. To assess pollen tube growth directly, in vitro germination assays were performed. Consequently, a time-course of in vitro pollen tube growth was performed and a(1,4)Fuc-T activity was measured at the different points. While a(1,4)Fuc-T activity remained relatively constant in germinating pollen grains between 0 h and 5 h culture, a 3-fold increase in activity was found in a 12 h culture (Table 2). The activity remained at this high level even after 20 h culture (Table 2). This last result suggests that higher amounts of a4-fucosylated N-glycoproteins are likely to be synthesized during pollen tube elongation. This hypothesis was tested by immunodetection of Lewisa motifs on the same amount of proteins from mature pollen grains and pollen tubes from a 12 h and 20 h culture, respectively. The dot blot analysis revealed an increase in the Lewisa motif amounts from 0 to 20 h (Fig. 5). Both high levels of a(1,4)Fuc-T activity and Lewisa structures are well correlated and suggest that a4-fucosylated proteins are required in pollen tube elongation.

synthesized enzymes during pollen tube elongation, an in vitro pollen grains germination assay was performed in the presence of 150 mM cycloheximide (Chx) (Table 2). At this concentration, Chx is an inhibitor of de novo protein synthesis (Riou-Khamlichi et al., 2000). In order to check the Chx effect, percentages of germination and pollen tube elongation were determined after 5 h and 12 h treatment. 83% of pollen grains germinated after 5 h Chx treatment and only 8% of pollen grains presenting a normal pollen tube length, was detected after 12 h Chx treatment (Table 2). It is concluded that 150 mM Chx has no effect on pollen germination but strongly inhibits pollen tube elongation. This result was also in agreement with previous work (Mascarenhas, 1993). Between 0 h and 5 h of Chx treatment, a(1,4)Fuc-T activity remained constant around 110 pmol h 1 mg 1 protein similar to the level determined in untreated pollen grains (Table 2). This result suggests that this enzyme may have a low turnover or is stable as described for late specific pollen proteins (Ylstra and McCormick, 1999). By contrast, a very low increase in activity was observed

Table 2. Determination of a(1,4)Fuc-T activity during pollen tube growth with or without cycloheximide Pollen grain germination assays were conducted for 5, 12 and 20 h in liquid germination medium. A treatment with 150 mM cycloheximide was performed on germinated pollen grains for 5 h and 12 h. a(1,4)FucT activity was determined in germinating pollen grains and pollen tubes in both germination conditions. Results are the means"SD of three independent experiments, each performed on pollen grains isolated from 15 anthers. 300 germinating pollen grains or pollen tubes were observed and counted to determine percentage of germination (a) and percentage of up-to-3 cm forms (b).

Increase in a(1,4)Fuc-T activity is due to protein neosynthesis during pollen tube elongation

To test whether the 3-fold increase in a(1,4)Fuc-T activity (346 pmol Fuc h 1 mg 1 protein) was due to newly

a(1,4)Fuc-T activity (pmol Fuc h 0

5h

12 h

1

mg

1

protein) 20 h

Untreated 115"10 109"5 (94%) (a) 364"19 (89%) (b) 363"13 150 mM 115"10 110"8 (83%) (a) 171"15 (8%) (b) nda cycloheximide a

Fig. 4. a(1,4)Fuc-T activity in hand-pollinated female tissues. Handpollination was performed as described in Materials and methods. a(1,4)Fuc-T activity was followed and determined from 0 to 20 h in nonor hand-pollinated stigmas and styles. Three stigmas and styles were used per experiment. Results are the means"SD of three independent experiments.

1433

nd: Not done.

Fig. 5. Immunodetection of Lewisa motifs during pollen tube elongation. Protein extract was isolated from mature pollen grains (lane 3) and pollen tubes after 12 h and 20 h culture, lane 5 and lane 6, respectively. Lane 4 corresponds to germinating pollen grains treated with 150 mM cycloheximide for 12 h. Bovine protein extract was loaded as a negative control (lane 1) and human saliva as a positive control (lane 2). 125 ng of proteins were loaded per lane. Anti-plant Lewisa antibodies were revealed by peroxidase assay as described in Materials and methods. The exposure time for autoradiography was 30 s.

1434

Joly et al.

on 12 h Chx-treated pollen grains (around 171 pmol h 1 mg 1 protein) (Table 2). This low increase could be due to the presence of the 8% of well-elongated pollen tubes (Table 2). Furthermore, an increase in the amount of Lewisa motifs was not observed within 12 h of Chx-treated pollen tubes (Fig. 5). In the presence of Chx, no additional amount of Lewisa is found, strongly suggesting an inhibition of the de novo synthesis of a4-fucosyltransferase. It is concluded that, during pollen tube elongation, a new pool of a4-Fuc-T enzyme is required.

microgametogenesis. The crucial event after mitosis I corresponds to the beginning of intine elaboration (McCormick, 1993). A pollen grain has two cell walls: an inner layer, intine, which is largely composed of pectocellulose and an outer network, exine, which is rich in sporopollenin and covered with tryphine (Lord, 2000). The tapetum plays a determinant role in exine deposition whereas the microspore itself controls intine production. It is suggested that these increases in a4-Fuc-T activity, generated by maturing pollen grains, could be linked to intine elaboration and maturation.

Discussion

A specific and high a(1,4)Fuc-T activity seems to be required during pollen tube elongation

In plants, Lewisa harbouring N-glycans were associated with extracellular soluble or membrane-bound glycoproteins (Fitchette et al., 1999; Bakker et al., 2001). For the first time the variations of a(1,4)Fuc-T activity during tobacco flower development have been investigated. It is postulated that this activity during pollen development and pollen tube elongation could reflect the localization of a4-fucosylated substrates and, therefore, their potential involvement in the male reproductive process. A positive correlation between an increase in the amount of Lewisa motifs and an increase in a(1,4)Fuc-T activity during pollen maturation and pollen tube elongation is shown. a(1,4)Fuc-T could be a housekeeping enzyme

Cellular events such as division, elongation or expansion occur in the corolla and reproductive tissues during flower development. a(1,4)Fuc-T activity always remained detectable at a constitutive and basal level in the perianth, non-pollinated pistils and empty anthers. This suggests that a(1,4)Fuc-T activity is basically required during tobacco flower growth and is implicated in N-glycoprotein maturation and in primary metabolism. Thus, it might be suspected that a4-Fuc-T simply acts as a housekeeping enzyme as described for another glycosyltransferase (Wenderoth and von Schaewen, 2000). Strikingly, peaks of a(1,4)Fuc-T activity were only associated with male gametophyte development whereas no significant variations were detected in the female gametophyte. These data suggest that a(1,4)Fuc-T could be considered as a housekeeping enzyme during ovule development. Nevertheless, further immunolocalizations are necessary to confirm the presence or absence of specific Lewisa containing N-glycoproteins in the pistil. a(1,4)Fuc-T activity is up-regulated during microgametogenesis

A 2-fold increase in a(1,4)Fuc-T activity was detected in binucleate microspores and also during late pollen grain maturation suggesting a role for a4-Fuc-T during

During germination, intine is destabilized implying invagination and pollen tube emergence at pectin-rich aperture. After 5 h of pollen germination, a(1,4)Fuc-T activity increases by about 3-fold strongly suggesting that new a4-fucosylated proteins are needed during pollen tube expansion. It is proposed that a(1,4)Fuc-T activity might be up-regulated on demand and that N-glycoproteins harbouring such complex carbohydrates could play an essential role from germination to whole pollen tube elongation. Indeed, complex glycans are probably involved in such a developmental process because the Arabidopsis thaliana cgl homozygous mutant which lacks complex glycans is male sterile (Von Schaewen et al., 1993). Taken together, these data and those reported in previous studies support the idea that the pollen grain represent a unique developmental and highly specialized structure in which essential functions could be mediated by glycoproteins with complex N-glycans. Furthermore, N-glycoproteins involved in cell wall elaboration are essential for tobacco pollen tube growth (Capkova et al., 1997). Gene encoding a4-Fuc-T might potentially belongs to the pollen late genes family

Proteins encoded by late genes seem to play an important role during pollen maturation anduor pollen tube elongation because mature pollen grains store large amounts of these proteins and mRNAs in readiness for pollen germination (Mascarenhas, 1990, 1993). At least, a 4-fold increase in a(1,4)Fuc-T activity was measured in mature pollen grains compared to that detected in free microspores. Moreover, in nongerminating quiescent pollen grains taken 72 h after anthesis, a(1,4)Fuc-T activity was still high and similar to that detected in mature pollen grains at anthesis (data not shown). Thus, this high level of activity could be linked to a pre-existing pool of enzyme which might accumulate during late pollen development. An additional 3-fold increase in activity was observed in well-elongated pollen tubes after 12 h culture. Before 12 h culture, this new

a4-Fucosyltransferase is regulated during pollen development

peak was not detected, but a(1,4)Fuc-T activity level remained as important and constant as in mature pollen grains. Furthermore, a similar situation was also observed in cycloheximide-treated germinating pollen grains. Taken together, all these results are in favour of an accumulation of an active enzyme form during late pollen maturation ready to be used during germination and pollen tube elongation. Cycloheximide treatment of germinating pollen grains suggests that the important increase in a(1,4)Fuc-T activity was linked to de novo synthesized enzymes from a pool of mRNAs used between 5 h and 12 h culture. According to published literature and to this study’s different results, which strongly suggest the existence of two pools of active enzyme and mRNAs in pollen grains, the gene encoding a4-Fuc-T might belong to the late pollen gene family. Therefore, it is suspected that a4-fucosylated proteins could also be regulated as late pollen proteins. Most proteins encoded by late genes are involved in pectin metabolism during pollen intine and pollen tube cell wall elaboration. Furthermore, these proteins are often N-glycosylated (Tebbut et al., 1994; Ogawa et al., 1996; Witting et al., 2000). Cry j I allergenic protein from japanese cedar is a4-fucosylated and has a pectate lyase enzyme activity (Taniguchi et al., 1995; Ogawa et al., 1996). Two tobacco late pollen pectate lyases g10 and Nt59 are potentially N-glycosylated (Rogers et al., 1992; Kulikauskas and McCormick, 1997). As g10 deduced polypetide sequence is highly similar to Cry j I, but not to Nt59 (data not shown), further work is necessary to determine the involvement of the g10 enzyme during tobacco pollen development and, subsequently, its putative glycosylation profile. In the plant kingdom, the functions of Lewisacontaining N-glycoproteins remain largely unknown. This study’s results implicate, for the first time, a(1,4)Fuc-T activity during pollen development and it might be relevant to an involvement of complex N-glycans in this process. Moreover, a4-Fuc-T is the first N-glycosyltransferase shown to be involved in a fundamental aspect of plant development. Further investigations should help to characterize and localize a4-fucosylated candidates in order to understand their function during pollen development and pollen tube elongation.

Acknowledgements We thank Dr Loı¨c Faye and Michel Carlue´ for their critical reading of the manuscript, Pascal Guillaume for his technical assistance. This work was performed within the French ‘GT-rec’ network and supported in part by grants MENRT (ACC SV No 9514111) and CNRS (PCV program). Caroline Joly was supported by a PDZR fellowship.

1435

References Bakker H, Bardor M, Molthoff JW, Gomord V, Elbers I, Stevens LH, Jordi W, Lommen A, Faye L, Lerouge P, Bosch D. 2001. Galactose-extended glycans of antibodies produced by transgenic plants. Proceeding of the National Academy of Sciences, USA 98, 2899–2904. Blancher A, Socha WW. 1997. The ABO, Hh and Lewis blood group in humans and non-human primates. In: Blancher A, Klein, Socha WW, eds. Molecular biology and evolution of blood group and MHC antigens in primates. Berlin, Heidelberg: Springer-Verlag, 30–92. Boisson M, Gomord V, Audran C, Berger N, Dubreucq B, Granier F, Lerouge P, Faye L, Caboche M, Lepiniec L. 2001. Arabidopsis glucosidase I mutants reveal a critical role of N-glycan trimming in seed development. The EMBO Journal 20, 1010–1019. Bradford MM. 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248–254. Capkova V, Zbrozek J, Tupy J. 1994. Protein synthesis in tobacco pollen tubes: preferential synthesis of cell-wall 69 kDa and 66 kDa glycoproteins. Sexual Plant Reproduction 7, 57–66. Capkova V, Fidlerova A, van Amstel T, Croes AF, Mata C, Schrauwen JAM, Wullems GJ, Tupy J. 1997. Role of N-glycosylation of 66 and 69 kDa glycoproteins in wall formation during pollen tube growth in vitro. European Journal of Cell Biology 72, 282–285. De Martinis D, Mariani C. 1999. Silencing gene expression of the ethylene-forming enzyme results in a reversible inhibition of ovule development in transgenic tobacco plants. The Plant Cell 11, 1061–1071. Faik A, Bar-Peled M, DeRocher AE, Zeng W, Perrin RM, Wilkerson C, Raikhel NV, Keegstra K. 2000. Biochemical characterization and molecular cloning of an a-1,2-fucosyltransferase that catalyses the last step of cell wall xyloglucan biosynthesis in pea. Journal of Biological Chemistry 275, 15082–15089. Fitchette AC, Cabanes-Mancheteau M, Marvin L, Martin B, Satiat-Jeunemaitre B, Gomord V, Crooks K, Lerouge P, Faye L, Hawes C. 1999. Biosynthesis and immunolocalization of Lewis a-containing N-glycans in the plant cell. Plant Physiology 121, 333–343. Fitchette-Laine´ AC, Gomord V, Cabanes M, Michalski J-C, Saint Macary M, Foucher B, Cavelier B, Hawes C, Lerouge P, Faye L. 1997. N-glycans harboring the Lewis a epitopes are expressed at the surface of plant cells. The Plant Journal 12, 1411–1417. Garcia-Casado G, Sanchez-Monge R, Chrispeels MJ, Armentia A, Salcedo G, Gomez L. 1996. Role of complex asparagine-linked glycans in the allergenicity of plant glycoproteins. Glycobiology 6, 471–477. Hammond C, Helenius A. 1995. Quality control in the secretory pathway. Current Opinion in Cell Biology 7, 523–529. Koltunow AM, Truettner J, Cox KH, Wallroth M, Goldberg RB. 1990. Different temporal and spatial gene expression patterns occur during anther development. The Plant Cell 2, 1201–1224. Kulikauskas R, McCormick S. 1997. Identification of the tobacco and Arabidopsis homologues of the pollenexpressed LAT59 gene of tomato. Plant Molecular Biology 34, 809–814. Lindholm P, Kuittinen T, Sorri O, Guo D, Merits A, Tormakangas K, Runeberg-Roos P. 2000. Glycosylation of

1436

Joly et al.

phytepsin and expression of dad1, dad2 and ost1 during onset of cell death in germinating barley scutella. Mechanisms of Development 93, 169–173. Lord E. 2000. Adhesion and cell movement during pollination: chercher la femme. Trends in Plant Science 5, 368–373. Lukowitz W, Nickle TC, Meinke DW, Last RL, Conklin PL, Somerville CR. 2001. Arabidopsis cyt1 mutants are deficient in a mannose-1-phosphate guanylyltransferase and point a requirement of N-linked glycosylation for cellulose biosynthesis. Proceeding of the National Academy of Sciences, USA 98, 2262–2267. Mascarenhas JP. 1990. Gene activity during pollen development. Annual Review of Plant Physiology and Plant Molecular Biology 41, 317–338. Mascarenhas JP. 1993. Molecular mechanisms of pollen tube growth and differentiation. The Plant Cell 5, 1303–1314. McCormick S. 1993. Male gametophyte development. The Plant Cell 5, 1265–1275. Melo NS, Nimtz M, Conradt HS, Fevereiro PS, Costa J. 1997. Identification of the human Lewis a carbohydrate motif in a secretory peroxidase from a plant cell suspension culture (Vaccinium myrtillus L.). FEBS Letters 415, 186–191. Ogawa H, Hijikata A, Amano M, Kojima K, Fukushima H, Ishizuka I, Kurihara Y, Matsumoto I. 1996. Structures and contribution to the antigenicity of oligosaccharides of Japanese cedar (Cryptomeria japonica) pollen allergen Cry j I: relationship between the structures and the antigenic epitopes of plant N-linked complex-type glycans. Glycoconjugate Journal 13, 555–566. Oulmouden A, Wierinckx A, Petit JM, Costache M, Palcic MM, Mollicone R, Oriol R, Julien R. 1997. Molecular cloning and expression of a bovine alpha(1,3)-fucosyltransferase gene homologous to a putative ancestor gene of the human FUT3-FUT5-FUT6 cluster. Journal of Biological Chemistry 272, 8764–8773. Riou-Khamlichi C, Menges M, Healy SJ, Murray JAH. 2000. Sugar control of the plant cell cycle: differential regulation of the Arabidopsis D-type gene cyclin expression. Molecular Cell Biology 20, 4513–4521. Rogers HJ, Harvey A, Lonsdale DM. 1992. Isolation and characterization of the tobacco gene with homology to pectate

lyase which is specifically expressed during microsporogenesis. Plant Molecular Biology 20, 493–502. Staudacher E, Altmann F, Wilson IBH, Ma¨rz L. 1999. Fucose in N-glycans: from plant to man. Biochimica et Biophysica Acta 1473, 216–236. Tebbutt SJ, Rogers HJ, Lonsdale DM. 1994. Characterization of a tobacco gene encoding a pollen-specific polygalacturonase. Plant Molecular Biology 25, 283–297. Taniguchi Y, Ono A, Sawatani M, Nanba M, Kohno K, Usui M, Kurimoto M, Matuhasi T. 1995. Cry j I, a major allergen of japanese cedar pollen, has a pectate lyase enzyme activity. Allergen 50, 90–93. von Schaewen A, Sturm A, O’Neill J, Chrispeels MJ. 1993. Isolation of a mutant Arabidopsis plant that lacks Nacetylglucosaminyltransferase I and is unable to synthesize Golgi-modified complex N-linked glycans. Plant Physiology 102, 1109–1118. Wenderoth I, von Schaewen A. 2000. Isolation and characterization of plant N-acetylglucosaminyltransferase I (GntI ) cDNA sequences. Functional analyses in the Arabidopsis cgl mutant and in the antisense plants. Plant Physiology 123, 1097–1108. Wilson IBH, Zeleny R, Kolarich D, Staudacher E, Stroop CJM, Kamerling JP, Altmann F. 2001. Analysis of Asn-linked glycans from vegetable foodstuffs: widespread occurrence of Lewisa, core a1,3-linked fucose and xylose substitutions. Glycobiology 11, 261–274. Witting FRA, Knuiman B, Derksen J, Capkova V, Twell D, Schrauwen JAM, Wullems GJ. 2000. The pollen-specific gene Ntp303 encodes a 69 kDa glycoprotein associated with the vegetative membranes and the cell wall. Sexual Plant Reproduction 12, 276–284. Yazawa S, Furukawa K. 1980. alpha-L-Fucosyltransferases related to biosynthesis of blood group substances in human saliva. Journal of Immumogenetic 7, 137–148. Yih Bih F, Wu SSH, Ratnayake C, Walling LL, Nothnagel EA, Huang AHC. 1999. The predominant on the surface of maize pollen is an endoxylanase synthesized by a tapetum mRNA with a long 59 leader. Journal of Biological Chemistry 274, 22884–22984. Ysltra B, McCormick S. 1999. Analysis of mRNA stabilities during pollen development and in BY2 cells. The Plant Journal 20, 101–108.