Identification of Secret Agent as the O-GlcNAc Tra
http://www.100md.com
病菌学杂志 2005年第15期
Department of Plant Molecular Genetics, Centro Nacional de Biotecnología (CSIC), Campus Universidad Autónoma de Madrid, 28049 Madrid, Spain
Department of Plant Biology and Plant Molecular Genetics Institute, University of Minnesota, Saint Paul, Minnesota 55108
ABSTRACT
Serine and threonine of many nuclear and cytoplasmic proteins are posttranslationally modified with O-linked N-acetylglucosamine (O-GlcNAc). This modification is made by O-linked N-acetylglucosamine transferases (OGTs). Genetic and biochemical data have demonstrated the existence of two OGTs of Arabidopsis thaliana, SECRET AGENT (SEC) and SPINDLY (SPY), with at least partly overlapping functions, but there is little information on their target proteins. The N terminus of the capsid protein (CP) of Plum pox virus (PPV) isolated from Nicotiana clevelandii is O-GlcNAc modified. We show here that O-GlcNAc modification of PPV CP also takes place in other plant hosts, N. benthamiana and Arabidopsis. PPV was able to infect the Arabidopsis OGT mutants sec-1, sec-2, and spy-3, but at early times of the infection, both rate of virus spread and accumulation were reduced in sec-1 and sec-2 relative to spy-3 and wild-type plants. By matrix-assisted laser desorption ionization-time of flight mass spectrometry, we determined that a 39-residue tryptic peptide from the N terminus of CP of PPV purified from the spy-3 mutant, but not sec-1 or sec-2, was O-GlcNAc modified, suggesting that SEC but not SPY modifies the capsid. While our results indicate that O-GlcNAc modification of PPV CP by SEC is not essential for infection, they show that the modification has a role(s) in the process.
INTRODUCTION
Dynamic modification of serine and threonine with O-linked ?-N-acetylglucosamine (O-GlcNAcylation) is widespread among nuclear and cytoplasmic eukaryotic proteins (33). The modification is essential for viability in both plants and animals (9, 23). O-GlcNAcylation is a regulatory modification that shares many common traits with protein phosphorylation (26). In addition, it has been documented for several proteins that phosphorylation and O-GlcNAcylation can occur at the same site. While the details of how O-GlcNAcylation functions in specific pathways remain to be elucidated, some general themes have emerged. The level of O-GlcNAcylation is dynamic and is influenced by hormonal signals, stress, and metabolic status, and the modification has roles in the regulation of transcription and protein synthesis and degradation. Disturbance of cellular processes regulated by O-GlcNAcylation appears to be involved in very important pathological disorders such as Alzheimer's disease (17) and diabetes (31).
The enzymes responsible for O-GlcNAc addition (O-GlcNActransferases [OGTs]) are highly conserved in plants and animals (33). OGT enzymes have an N-terminal tetratricopeptide repeat domain and a C-terminal catalytic domain. While animals have one OGT, Arabidopsis thaliana has two: Secret Agent (SEC) and SPINDLY (SPY) (9, 29). Genetic experiments have demonstrated that SEC and SPY have at least partly overlapping functions (9), but so far, there is very little information on their target proteins.
O-GlcNAc modifications have been found in some structural proteins from different animal viruses such as the cytomegalovirus basic phosphoprotein (8), the adenovirus fiber protein (20), and the baculovirus gp41 protein (34), as well as in the nonstructural rotavirus NS26 protein (6). The biological relevance of these modifications is yet unknown.
Plum pox virus (PPV) is a potyvirus that in nature infects fruit trees of the Prunus genus but that is also able to infect experimentally different herbaceous hosts (18). The messenger-polarity single-stranded genomic RNA of potyviruses is translated into a large polyprotein that is further processed by three virus-encoded proteases (21, 22). The potyviral genome is encapsidated in flexuous rod particles made up of 2,000 units of a single type of capsid protein (CP) located at the C end of the viral polyprotein (25). The N- and C-terminal regions of the potyviral CP are surface exposed and can be released from the virus particles by mild proteolysis with trypsin (24). The CP of PPV has been shown to be modified by O-GlcNAcylation and phosphorylation. O-GlcNAc-modified residues were mapped to the N-terminal portion of the protein (3). In this paper, we have identified SEC as the plant OGT involved in these modifications and investigated the relevance of SEC function for virus infection.
MATERIALS AND METHODS
Virus infection and purification. PPV isolate Rankovic (16) and the chimera PPV-NK-GFP (5) were used in these experiments. Plants were grown in a greenhouse maintained at 16 h of light with supplementary illumination and a 19 to 22°C temperature. Young Nicotiana benthamiana plants were mechanically inoculated by rubbing three leaves per plant with crude sap from PPV-infected plants. Wild-type A. thaliana ecotypes Columbia (Col-0) and Wassilewskija (WS) and the sec-1 (WS background), sec-2, and spy-3 (Col-0 background) mutants (9) were inoculated by infiltration with Agrobacterium tumefaciens C58C1 harboring pBINPPV-NK-GFP, which contains a full-length cDNA copy of the genome of PPV-NK-GFP cloned into pBin19 (C. Lucini, J. J. López-Moya, J. M. Alamillo, and J. A. García, unpublished results). Approximately 150 μl of the Agrobacterium cultures (optical density at 600 nm = 0.5) induced with acetosyringone were applied with a syringe to the undersides of three leaves of young Arabidopsis seedlings.
Although PPV accumulation appears to be quite similar in N. clevelandii, N. benthamiana, and Arabidopsis, the protocol previously described for PPV purification from leaves of N. clevelandii (16) was not suitable for purification from N. benthamiana or Arabidopsis, probably because of virus aggregation (data not shown). Thus, the original protocol was slightly modified for these hosts. In brief, 10 g of N. benthamiana- or Arabidopsis-infected leaves collected 3 weeks postinoculation was homogenized with 50 ml (Arabidopsis) or 20 ml (N. benthamiana) of 0.18 M McIlvain's citric acid-phosphate buffer, pH 7, containing 0.2% thioglycolic acid, 0.01 M sodium diethyldithiocarbamate, 0.5 M urea, and 3 mM EDTA first for 5 min with a mortar and pestle and then for 10 min with a Waring blender at low speed. Next, 50 ml (Arabidopsis) or 10 ml (N. benthamiana) of cold chloroform was added to the mixture and shaken in the Waring blender for another 5 min. The homogenate was centrifuged for 11 min at 5,000 x g, and the supernatant was centrifuged for 2 h at 82,500 x g. The pellet was resuspended in 3 ml (Arabidopsis) or 4 ml (N. benthamiana) of 10 mM McIlvain's citric acid-phosphate buffer, pH 7, containing 1 M urea and 0.2% ?-mercaptoethanol for 2 h and centrifuged for 10 min (Arabidopsis) or 5 min (N. benthamiana) at 1,500 x g. The supernatant was centrifuged for 1.5 h at 57,000 x g (Arabidopsis) or 82,000 x g (N. benthamiana). The resulting pellet was resuspended for 2 h in 0.5 ml (Arabidopsis) or 0.6 ml (N. benthamiana) of 0.1 M sodium borate buffer, pH 8.2, containing 10 mM EDTA, clarified by centrifugation for 5 min at 1,200 x g (Arabidopsis) or 1,500 x g (N. benthamiana), layered over a 3-ml cushion of 20% sucrose in the same buffer, and centrifuged at 72,000 x g for 2 h. The pellet was resuspended in 50 μl (Arabidopsis) or 100 μl (N. benthamiana) of 5 mM sodium borate buffer, pH 8.2, and clarified by centrifugation for 10 min at 4,800 x g (Arabidopsis) or 7,400 x g (N. benthamiana). All the purification steps were carried out at 4°C, and the purified virus was stored at –20°C.
Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) analysis of PPV CP digested with trypsin. Approximately 10 μg of purified PPV virions was digested with 2 ng of modified porcine trypsin (Promega) in a buffer containing 25 mM ammonium carbonate, pH 8, in a reaction volume of 10 μl. Digestion proceeded at 37°C for 20 min, and then it was rapidly stopped by adding 1 μl of 0.5% trifluoroacetic acid. Reaction products were desalted with a Zip-Tip reverse-phase C18 column (Millipore) and eluted in 5 μl of 70% aqueous acetonitrile and 0.1% trifluoroacetic acid.
About 0.4 μl of matrix solution (5 g/liter 2,5-dihidroxibenzoic acid in 33% aqueous acetonitrile and 0.1% trifluoroacetic acid) was deposited onto a 400-μm AnchorChip MALDI target (Bruker Daltonics, Bremen, Germany) and allowed to dry at room temperature. Then, 0.4 μl of the tryptic peptide mixture was added and again allowed to dry at room temperature. Peptide mass fingerprinting spectra were acquired with a MALDI-TOF Bruker Reflex IV mass spectrometer (Bruker Daltonics) equipped with a nitrogen laser (337 nm). Analyses were performed in reflector positive ion mode, accumulating 120 shots, with 400 ns of pulsed ion delayed extraction and an acceleration voltage of 20 kV and 1,600 V in the reflector detector. The mass spectra were externally calibrated using a mixture of peptide standards (angiotensin II, substance P, bombesin, somatostatine 28, and cytochrome C).
Processing of the spectra and data analysis were performed with Bruker Daltonics XTOF 5.1.1 and Biotools 2.1 software.
Assessment of spread and accumulation of PPV. Spreading of PPV-NK-GFP was assessed by visualizing green fluorescent protein (GFP) fluorescence under a Leica MZ FLIII fluorescence microscope with excitation and barrier filters of 480/40 nm and 510 nm, respectively. PPV accumulation was quantified by double-antibody sandwich indirect enzyme-linked immunosorbent assay (ELISA) using the REALISA kit (C. C. Durviz, S.L.). The samples were prepared by grinding the infected leaves with a mortar and pestle in 5 mM sodium phosphate buffer, pH 7.2 (2 ml/g), and storing them frozen at –20°C. A standard curve was obtained with different amounts of purified PPV virions diluted in the extract of healthy plants. The enzymatic reaction was developed with the Sigma FAST p-nitrophenyl phosphate tablet set system (Sigma), and the optical density at 405 nm of the reaction product was measured in a Titertek Multiskan MCC/340 (Labsystems) spectrophotometer. Statistical analysis of the data was carried out with PRISM 4 software (GraphPad).
RESULTS
The N-terminal region of PPV CP is O-GlcNAcylated in Nicotiana benthamiana and Arabidopsis thaliana. Mass spectrometry analyses have revealed the existence of O-GlcNAcylation at the N-terminal region of the CP of PPV virions purified from N. clevelandii (3). In order to assess whether this modification could also take place in other plant species, PPV was purified from another well-characterized PPV host, N. benthamiana, and from two different ecotypes of Arabidopsis, Col-0 and WS, which has been recently shown to be susceptible to PPV infection (unpublished results from several laboratories). PPV virions purified from infected plants of these species were partially digested with trypsin, and the resulting peptides were analyzed by MALDI-TOF. Similar to the previously published spectrum of PPV purified from N. clevelandii (3), the spectra of the three samples showed three ions corresponding to the partially digested tryptic peptide spanning residues 1 to 39 of the PPV CP (Fig. 1 and 2). One ion was the size expected from the unmodified peptide (m/z = 4,077 Da), and the other two ions were the sizes expected if the peptide was modified with one or two O-GlcNAc residues (m/z = 4,280 Da and 4,483 Da, respectively) (Fig. 2). These results indicate that similar O-GlcNAcylation of PPV CP appears to take place in different host plants.
PPV infection of sec plants is impaired relative to that of wild-type or spy plants. Arabidopsis has two OGTs: SPY and SEC. SEC protein is more similar to animal OGTs than to SPY, while SPY shows the same level of similarity with SEC and animal OGTs (9, 14). Both SEC (9) and SPY (28) have shown to have OGT activity. Two T-DNA insertion mutants of SEC have been described: sec-1 is in a WS background and the T-DNA is inserted into the tetratricopeptide repeat domain, and sec-2 is in a Col-0 background and the T-DNA is inserted into an intron adjacent to exons coding for the putative catalytic region of the protein (9). While sec-1 and sec-2 have no obvious phenotypes (9), spy plants have defects in a number of processes, including gibberellin and cytokinin responses, flowering, circadian regulation, and light inhibition of hypocotyls elongation (7, 15, 27, 30). spy-3 is in a Col-0 background and has a Gly-to-Ser substitution in the C-terminal region of the protein (14). Interestingly, homozygous sec-1/spy-3 and sec-2/spy-3 double mutants die during embryogenesis, indicating that SEC and SPY have an overlapping and essential function during embryogenesis (9).
In order to investigate the possible role of plant OGTs in the O-GlcNAcylation of PPV CP and in virus infection, sec-1, sec-2, and spy-3 mutant plants were inoculated with PPV-NK-GFP, a recombinant PPV expressing the GFP, which allows an easier monitoring of the infection (5). PPV caused in wild-type Col-0 and WS mild symptoms consisting of leaf chlorosis, rosette gathering, curling of cauline leaves, and shortening of the inflorescence stems (data not shown). Similar symptoms were observed in the three mutant plants inoculated with PPV-NK-GFP, especially late in the infection. Although nearly 100% of the inoculated plants became infected, examination of GFP localization in the leaves detected differences in the pattern of infection between the different genotypes (Fig. 3). Radiation of the infection from the major veins and the extent of infection of the lamina were reduced in both sec alleles relative to the corresponding wild type. At 12 days postinoculation (dpi), a similar number of leaves from spy-3 and wild-type plants were infected (Fig. 4) but the percentage of sec-1 and sec-2 plants with detectable GFP expression was slightly, but reproducibly, reduced. The sec-1 and sec-2 alleles had a larger effect on the spread of an infection within a leaf. Large GFP fluorescence areas were observed in a smaller proportion of leaves from sec-1 and sec-2 plants than in wild-type or spy-3 plants (Fig. 3 and 4). The differences in the extent of virus spread among the different genotypes had decreased by 19 dpi (Fig. 3 and 4).
In order to have a more quantitative assessment of the progress of PPV infection in the different genotypes, the amount of virus in the infected leaves was determined by ELISA. At 12 dpi, PPV accumulation was significantly lower in sec-1 and sec-2 plants than in their corresponding wild-type plants (Fig. 5). In contrast, virus titer was not affected in spy-3. In agreement with the experiments using GFP to monitor the infection, the reduction in virus accumulation in sec plants was less pronounced at 19 dpi (Fig. 5). The infectivity of the PPV virions produced in the wild-type and the mutant plants was assessed by a local lesion assay in Chenopodium foetidum (19). Virions produced in sec-1 and sec-2 plants appeared to be slightly less infectious than those produced in the corresponding wild-type Arabidopsis plants. However, due to the rather low linearity and accuracy of the local lesion assay, the differences were not statistically significant (data not shown).
The O-GlcNAcylation of the N-terminal region of PPV-CP is prevented by sec but not by spy mutations. PPV was purified from infected sec-1, sec-2, and spy-3 plants. The purified virions were subjected to partial trypsin digestion, and the resulting peptides were analyzed by MALDI-TOF. The spectrum of the spy-3 sample showed the presence of the peptide from amino acids 1 through 39 in the same three forms, nonglycosylated, mono-, and di-O-GlcNAcylated, found in the spectrum of virions purified from wild-type Arabidopsis (Fig. 2 and Table 1). In contrast, only nonglycosylated peptide from amino acids 1 through 39 was detected in samples prepared from virions purified from sec-1 and sec-2 plants (Fig. 2 and Table 1), indicating that SPY OGT or other plant proteins were not able to glycosylate the N-terminal region of PPV CP in the absence of the SEC protein.
The presence of acetylated forms (mass increase of 42 Da) of the nonglycosylated and O-GlcNAcylated variants of the peptide from amino acids 1 through 39 have been previously reported (3). These acetylated peptides were also detected in the virions purified from N. benthamiana and wild-type and mutant Arabidopsis (Fig. 2 and Table 1). A careful examination of the spectra also detected some signals that could correspond to phosphorylated (mass increase of 80 Da) and/or phosphorylated plus acetylated (mass increase of 80 plus 42 Da) forms of the nonglycosylated peptide from amino acids 1 through 39 in all the spectra (Fig. 2 and Table 1). Moreover, we also detected peptides with the size expected for the peptide from amino acids 1 through 39 mono-O-GlcNAcylated and phosphorylated among the tryptic peptides of virions purified from N. benthamiana, wild-type Arabidopsis Col-0 and WS, and spy-3 plants (Fig. 2 and Table 1). Interestingly, faint peaks that could correspond to a phosphorylated form of the di-O-GlcNAcylated peptide from amino acids 1 through 39 were found in the spectra of virions purified from N. benthamiana and wild-type Arabidopsis Col-0 (Fig. 2 and Table 1), suggesting that at least three Ser/Thr residues of the peptide from amino acids 1 through 39 can be modified at the same time.
DISCUSSION
O-GlcNAc modification has been shown to play very important regulatory roles in animals (32), and in agreement with this, deletion of the OGT gene of mouse was lethal (23). Loss of either SPY or SEC function alone does not cause lethality. However, sec spy double mutants died during embryogenesis, showing that OGT activity is also essential for plant viability and that SEC and SPY have at least one overlapping function (9). The extent of overlap in the function between SPY and SEC is unknown. While spy plants exhibit a number of obvious defects, including altered gibberellin and cytokinin responses, circadian rhythms, light responses, and flowering time (7, 15, 27, 30), these defects are not detectable or are much weaker in sec plants (9; L. Hartweck and N. E. Olszewski, unpublished results). These results can be explained if SPY and SEC have completely overlapping substrates but SPY is responsible for most of the O-GlcNAcylation or if SPY has some substrates that are not modified by SEC.
Very little is known of the OGT targets in plant cells. O-glycosylation with the terminal GlcNAc modification of several nuclear pore complex proteins of tobacco, including a protein that shows sequence similarity to bacterial aldose-1-epimerases, has been described previously (11), but the enzyme responsible for these modifications has not been identified. We have previously reported that the CP of a plant virus, PPV, was O-GlcNAc modified in N. clevelandii (3); now, we show that this modification is not specific for this host, since it takes place in a similar way in two other plant species, N. benthamiana and Arabidopsis (Fig. 2). More important, the MALDI-TOF analysis of PPV virions purified from sec and spy mutants indicates that SEC protein is the OGT responsible for the O-GlcNAc modification of the N-terminal region of PPV CP and that neither SPY nor other plant proteins is able to glycosylate this sequence (Fig. 2). However, we cannot rule out the possibility that SPY could be involved in the O-GlcNAc modification of other regions of PPV CP or other PPV proteins. These results support the hypothesis that SEC and SPY have partial functional independence. Whereas O-GlcNAc modification of the N-terminal region of PPV CP, like that of animal proteins, adds single sugar monomers (3; J. J. Pérez, S. Juárez, and J. A. García, unpublished results), the size of sugar chains of the glycosylated nuclear pore proteins of tobacco corresponds to more than five monosaccharides (10), suggesting that different enzymes could be involved in each of these modifications. Having in mind that SPY is less similar to mammalian OGTs than SEC, it is tempting to speculate that SPY could be involved in the second type of modification, which would be specific to plants.
The ability of PPV to infect sec-1 and sec-2 mutants (Fig. 3) demonstrates that O-GlcNAc modification by SEC OGT is not essential for PPV viability in Arabidopsis. However, the efficiency of PPV infection is markedly lower in these mutants than in the spy-3 mutant or in wild-type plants (Fig. 4 and 5), indicating that SEC activity plays an important role in infection. O-GlcNAc modifications at the segment from amino acids 1 through 39 of PPV CP are not essential, since deletions or substitutions that remove all Thr and Ser residues of this region have no perceptible effects on PPV infection in N. clevelandii (3, 4). Therefore, O-GlcNAc modification of other portions of PPV CP or other PPV proteins by SEC may play a role in the infection process. The present data do not allow us to discriminate whether some O-GlcNAc modifications are more important than others or the overall level of O-GlcNAcylation is the relevant factor. Moreover, we cannot rule out the possibility that the effect of sec mutation on PPV infection might be an indirect result of the action of SEC on a host protein. It is also important to point out that, although O-GlcNAc modification by SEC is not essential for the PPV infection of Arabidopsis, the possibility exists that O-GlcNAcylation plays a more important role in some plants (for instance, in the PPV natural woody hosts) than in others.
The role of O-GlcNAc modification in PPV infection is still an open question. It is well known that in mammals, O-GlcNAcylation and phosphorylation collaborate in the regulation of macromolecular interactions that control the activity of many cellular processes (26). PPV CP has been shown to contain phosphoserine and phosphothreonine residues (3). Although MALDI-TOF does not allow a precise analysis of protein phosphorylation, the MALDI-TOF spectra of PPV CP subjected to trypsin digestion showed signals that could correspond with phosphorylated peptides (Fig. 2 and Table 1). As expected, the presence of these putative phosphorylated peptides was not affected by the sec or spy mutations. Both nonglycosylated and O-GlcNAc-modified peptides appeared to be phosphorylated, in agreement with the fact previously observed in mammalian organisms that, although O-GlcNAcylation and phosphorylation are reciprocal modifications, a single polypeptide can contain simultaneously phosphate and O-GlcNAc residues (2). These data suggest that the O-GlcNAcylation of PPV CP, like that of those mammalian systems, could function in coordination with phosphorylation.
The genomic RNA is a central player in the most critical processes of viral infection: translation, replication, encapsidation, and cell-to-cell and long-distance movement. Thus, it is likely that allocation of RNA molecules to each of these processes is highly regulated. Ivanov and colleagues have demonstrated that phosphorylation of the CP of the potyvirus potato virus A reduces its ability to bind RNA and have suggested that this modification might regulate the formation and/or stability of virions and other viral ribonucleoproteins (12, 13). It is tempting to speculate that CP O-GlcNAc modification could also contribute to this regulatory mechanism. The fact that the effect of the sec alleles on PPV infection appears to be more apparent at early times (Fig. 3 through 5) suggests that the regulation of RNA distribution could be required at the beginning of the infection, when there could be a limiting amount of RNA molecules. An alternative possibility is that O-GlcNAc residues modifying PPV CP could be playing a role in virion stability similar to that proposed for the galactose and fucose residues that are O-linked to the N-terminal NAcSer of the Potato virus X CP, which has been shown to affect the water-absorbing capacity of the viral particles (1). It is striking that plant RNA viruses have adopted different O-glycosylation systems to modify their CPs; it will be interesting in the future to determine whether the role of SEC is PPV specific or it also modifies the CPs of other potyviruses or of viruses of other families.
The results described in this report identify a novel target for antiviral action. We have demonstrated that the O-GlcNAc modification of PPV takes place in several plant hosts and that viral propagation is compromised by sec mutations that do not noticeably affect plant growth. This could contribute to control the propagation of the virus in field conditions.
ACKNOWLEDGMENTS
We thank Elvira Domínguez for technical assistance and Juan Pablo Albar for helpful advice in MALDI-TOF analysis.
This work was supported by grants BIO2001-1434 and BIO2004-02687 from the Spanish MEC, QLK2-CT-2002-01050 and SP22-CT-2004 from the European Union to J.A.G., MCB-0112826 from the National Science Foundation, and DE-FG01-04ER04-02 from the U.S. Department of Energy to N.E.O.
REFERENCES
Baratova, L. A., N. V. Fedorova, E. N. Dobrov, E. V. Lukashina, A. N. Kharlanov, V. V. Nasonov, M. V. Serebryakova, S. V. Kozlovsky, O. V. Zayakina, and N. P. Rodionova. 2004. N-terminal segment of potato virus X coat protein subunits is glycosylated and mediates formation of a bound water shell on the virion surface. Eur. J. Biochem. 271:3136-3145.
Comer, F. I., and G. W. Hart. 2000. O-glycosylation of nuclear and cytosolic proteins. Dynamic interplay between O-GlcNAc and O-phosphate. J. Biol. Chem. 275:29179-29182.
Fernández-Fernández, M. R., E. Camafeita, P. Bonay, E. Méndez, J. P. Albar, and J. A. García. 2002. The capsid protein of a plant single-stranded RNA virus is modified by O-linked N-acetylglucosamine. J. Biol. Chem. 277:135-140.
Fernández-Fernández, M. R., J. L. Martínez-Torrecuadrada, J. I. Casal, and J. A. García. 1998. Development of an antigen presentation system based on plum pox potyvirus. FEBS Lett. 427:229-235.
Fernández-Fernández, M. R., M. Mouri?o, J. Rivera, F. Rodríguez, J. Plana-Durán, and J. A. García. 2001. Protection of rabbits against rabbit hemorrhagic disease virus by immunization with the VP60 protein expressed in plants with a potyvirus-based vector. Virology 280:283-291.
Gonzalez, S. A., and O. R. Burrone. 1991. Rotavirus NS26 is modified by addition of single O-linked residues of N- acetylglucosamine. Virology 182:8-16.
Greenboim-Wainberg, Y., I. Maymon, R. Borochov, J. Alvarez, N. Olszewski, N. Ori, Y. Eshed, and D. Weiss. 2005. Cross talk between gibberellin and cytokinin: the arabidopsis GA response inhibitor SPINDLY plays a positive role in cytokinin signaling. Plant Cell 17:92-102.
Greis, K. D., W. Gibson, and G. W. Hart. 1994. Site-specific glycosylation of the human cytomegalovirus tegument basic phosphoprotein (UL32) at serine 921 and serine 952. J. Virol. 68:8339-8849.
Hartweck, L. M., C. L. Scott, and N. E. Olszewski. 2002. Two O-linked N-acetylglucosamine transferase genes of Arabidopsis thaliana L. Heynh. have overlapping functions necessary for gamete and seed development. Genetics 161:1279-1291.
Heese-Peck, A., R. N. Cole, O. N. Borkhsenious, G. W. Hart, and N. V. Raikhel. 1995. Plant nuclear pore complex proteins are modified by novel oligosaccharides with terminal N-acetylglucosamine. Plant Cell 7:1459-1471.
Heese-Peck, A., and N. V. Raikhel. 1998. A glycoprotein modified with terminal N-acetylglucosamine and localized at the nuclear rim shows sequence similarity to aldose-1-epimerases. Plant Cell 10:599-612.
Ivanov, K. I., P. Puustinen, R. Gabrenaite, H. Vihinen, L. R?nnstrand, L. Valmu, N. Kalkkinen, and K. M?kinen. 2003. Phosphorylation of the potyvirus capsid protein by protein kinase CK2 and its relevance for virus infection. Plant Cell 15:2124-2139.
Ivanov, K. I., P. Puustinen, A. Merits, M. Saarma, and K. M?kinen. 2001. Phosphorylation down-regulates the RNA binding function of the coat protein of potato virus A. J. Biol. Chem. 276:13530-13540.
Jacobsen, S. E., K. A. Binkowski, and N. E. Olszewski. 1996. SPINDLY, a tetratricopeptide repeat protein involved in gibberellin signal transduction in Arabidopsis. Proc. Natl. Acad. Sci. USA 93:9292-9296.
Jacobsen, S. E., and N. E. Olszewski. 1993. Mutations at the SPINDLY locus of Arabidopsis alter gibberellin signal transduction. Plant Cell 5:887-896.
Laín, S., J. L. Riechmann, E. Méndez, and J. A. García. 1988. Nucleotide sequence of the 3' terminal region of plum pox potyvirus RNA. Virus Res. 10:325-342.
Liu, F., K. Iqbal, I. Grundke-Iqbal, G. W. Hart, and C. X. Gong. 2004. O-GlcNAcylation regulates phosphorylation of tau: a mechanism involved in Alzheimer's disease. Proc. Natl. Acad. Sci. USA 101:10804-10809.
López-Moya, J. J., M. R. Fernández-Fernández, M. Cambra, and J. A. García. 2000. Biotechnological aspects of plum pox virus. J. Biotechnol. 76:121-136.
López-Moya, J. J., and J. A. García. 2000. Construction of a stable and highly infectious intron-containing cDNA clone of plum pox potyvirus and its use to infect plants by particle bombardment. Virus Res. 68:99-107.
Mullis, K. G., R. S. Haltiwanger, G. W. Hart, R. B. Marchase, and J. A. Engler. 1990. Relative accessibility of N-acetylglucosamine in trimers of the adenovirus types 2 and 5 fiber proteins. J. Virol. 64:5317-5323.
Revers, F., O. Le Gall, T. Candresse, and A. J. Maule. 1999. New advances in understanding the molecular biology of plant/potyvirus interactions. Mol. Plant-Microbe Interact. 12:367-376.
Riechmann, J. L., S. Laín, and J. A. García. 1992. Highlights and prospects of potyvirus molecular biology. J. Gen. Virol. 73:1-16.
Shafi, R., S. P. Iyer, L. G. Ellies, N. O'Donnell, K. W. Marek, D. Chui, G. W. Hart, and J. D. Marth. 2000. The O-GlcNAc transferase gene resides on the X chromosome and is essential for embryonic stem cell viability and mouse ontogeny. Proc. Natl. Acad. Sci. USA 97:5735-5739.
Shukla, D. D., P. M. Strike, S. L. Tracy, K. H. Gough, and C. W. Ward. 1988. The N and C termini of the coat proteins of potyviruses are surface-located and the N-terminus contains the major virus-specific epitopes. J. Gen. Virol. 69:1497-1508.
Shukla, D. D., C. W. Ward, and A. A. Brunt. 1994. Genome structure, variation and function, p. 74-112. In D. D. Shukla, C. W. Ward, and A. A. Brunt (ed.), The potyviridae. CAB International, Cambridge, United Kingdom.
Slawson, C., and G. W. Hart. 2003. Dynamic interplay between O-GlcNAc and O-phosphate: the sweet side of protein regulation. Curr. Opin. Struct. Biol. 13:631-636.
Sothern, R. B., T. S. Tseng, S. L. Orcutt, N. E. Olszewski, and W. L. Koukkari. 2002. GIGANTEA and SPINDLY genes linked to the clock pathway that controls circadian characteristics of transpiration in Arabidopsis. Chronobiol. Biol. 19:1005-1022.
Thornton, T. M., L. Kreppel, G. W. Hart, and N. E. Olszewski. 1999. Genetic and biochemical analysis of Arabidopsis SPY, p. 445-448. In A. Altman, M. Ziv, and S. Izhar (ed.), Plant biotechnology and in vitro biology in the 21st century. Kluwer Academic Publishers, Dordrecht, The Netherlands.
Thornton, T. M., S. M. Swain, and N. E. Olszewski. 1999. Gibberellin signal transduction presents...the SPY who O-GlcNAc'd me. Trends Plant Sci. 4:424-428.
Tseng, T. S., P. A. Salomé, C. R. McClung, and N. E. Olszewski. 2004. SPINDLY and GIGANTEA interact and act in Arabidopsis thaliana pathways involved in light responses, flowering, and rhythms in cotyledon movements. Plant Cell 16:1550-1563.
Vosseller, K., L. Wells, M. D. Lane, and G. W. Hart. 2002. Elevated nucleocytoplasmic glycosylation by O-GlcNAc results in insulin resistance associated with defects in Akt activation in 3T3-L1 adipocytes. Proc. Natl. Acad. Sci. USA 99:5313-5318.
Wells, L., and G. W. Hart. 2003. O-GlcNAc turns twenty: functional implications for post-translational modification of nuclear and cytosolic proteins with a sugar. FEBS Lett. 546:154-158.
Wells, L., K. Vosseller, and G. W. Hart. 2001. Glycosylation of nucleocytoplasmic proteins: signal transduction and O-GlcNAc. Science 291:2376-2378.
Whitford, M., and P. Faulkner. 1992. A structural polypeptide of the baculovirus Autographa californica nuclear polyhedrosis virus contains O-linked N-acetylglucosamine. J. Virol. 66:3324-3329.(D. Chen, S. Juárez, L. Ha)
Department of Plant Biology and Plant Molecular Genetics Institute, University of Minnesota, Saint Paul, Minnesota 55108
ABSTRACT
Serine and threonine of many nuclear and cytoplasmic proteins are posttranslationally modified with O-linked N-acetylglucosamine (O-GlcNAc). This modification is made by O-linked N-acetylglucosamine transferases (OGTs). Genetic and biochemical data have demonstrated the existence of two OGTs of Arabidopsis thaliana, SECRET AGENT (SEC) and SPINDLY (SPY), with at least partly overlapping functions, but there is little information on their target proteins. The N terminus of the capsid protein (CP) of Plum pox virus (PPV) isolated from Nicotiana clevelandii is O-GlcNAc modified. We show here that O-GlcNAc modification of PPV CP also takes place in other plant hosts, N. benthamiana and Arabidopsis. PPV was able to infect the Arabidopsis OGT mutants sec-1, sec-2, and spy-3, but at early times of the infection, both rate of virus spread and accumulation were reduced in sec-1 and sec-2 relative to spy-3 and wild-type plants. By matrix-assisted laser desorption ionization-time of flight mass spectrometry, we determined that a 39-residue tryptic peptide from the N terminus of CP of PPV purified from the spy-3 mutant, but not sec-1 or sec-2, was O-GlcNAc modified, suggesting that SEC but not SPY modifies the capsid. While our results indicate that O-GlcNAc modification of PPV CP by SEC is not essential for infection, they show that the modification has a role(s) in the process.
INTRODUCTION
Dynamic modification of serine and threonine with O-linked ?-N-acetylglucosamine (O-GlcNAcylation) is widespread among nuclear and cytoplasmic eukaryotic proteins (33). The modification is essential for viability in both plants and animals (9, 23). O-GlcNAcylation is a regulatory modification that shares many common traits with protein phosphorylation (26). In addition, it has been documented for several proteins that phosphorylation and O-GlcNAcylation can occur at the same site. While the details of how O-GlcNAcylation functions in specific pathways remain to be elucidated, some general themes have emerged. The level of O-GlcNAcylation is dynamic and is influenced by hormonal signals, stress, and metabolic status, and the modification has roles in the regulation of transcription and protein synthesis and degradation. Disturbance of cellular processes regulated by O-GlcNAcylation appears to be involved in very important pathological disorders such as Alzheimer's disease (17) and diabetes (31).
The enzymes responsible for O-GlcNAc addition (O-GlcNActransferases [OGTs]) are highly conserved in plants and animals (33). OGT enzymes have an N-terminal tetratricopeptide repeat domain and a C-terminal catalytic domain. While animals have one OGT, Arabidopsis thaliana has two: Secret Agent (SEC) and SPINDLY (SPY) (9, 29). Genetic experiments have demonstrated that SEC and SPY have at least partly overlapping functions (9), but so far, there is very little information on their target proteins.
O-GlcNAc modifications have been found in some structural proteins from different animal viruses such as the cytomegalovirus basic phosphoprotein (8), the adenovirus fiber protein (20), and the baculovirus gp41 protein (34), as well as in the nonstructural rotavirus NS26 protein (6). The biological relevance of these modifications is yet unknown.
Plum pox virus (PPV) is a potyvirus that in nature infects fruit trees of the Prunus genus but that is also able to infect experimentally different herbaceous hosts (18). The messenger-polarity single-stranded genomic RNA of potyviruses is translated into a large polyprotein that is further processed by three virus-encoded proteases (21, 22). The potyviral genome is encapsidated in flexuous rod particles made up of 2,000 units of a single type of capsid protein (CP) located at the C end of the viral polyprotein (25). The N- and C-terminal regions of the potyviral CP are surface exposed and can be released from the virus particles by mild proteolysis with trypsin (24). The CP of PPV has been shown to be modified by O-GlcNAcylation and phosphorylation. O-GlcNAc-modified residues were mapped to the N-terminal portion of the protein (3). In this paper, we have identified SEC as the plant OGT involved in these modifications and investigated the relevance of SEC function for virus infection.
MATERIALS AND METHODS
Virus infection and purification. PPV isolate Rankovic (16) and the chimera PPV-NK-GFP (5) were used in these experiments. Plants were grown in a greenhouse maintained at 16 h of light with supplementary illumination and a 19 to 22°C temperature. Young Nicotiana benthamiana plants were mechanically inoculated by rubbing three leaves per plant with crude sap from PPV-infected plants. Wild-type A. thaliana ecotypes Columbia (Col-0) and Wassilewskija (WS) and the sec-1 (WS background), sec-2, and spy-3 (Col-0 background) mutants (9) were inoculated by infiltration with Agrobacterium tumefaciens C58C1 harboring pBINPPV-NK-GFP, which contains a full-length cDNA copy of the genome of PPV-NK-GFP cloned into pBin19 (C. Lucini, J. J. López-Moya, J. M. Alamillo, and J. A. García, unpublished results). Approximately 150 μl of the Agrobacterium cultures (optical density at 600 nm = 0.5) induced with acetosyringone were applied with a syringe to the undersides of three leaves of young Arabidopsis seedlings.
Although PPV accumulation appears to be quite similar in N. clevelandii, N. benthamiana, and Arabidopsis, the protocol previously described for PPV purification from leaves of N. clevelandii (16) was not suitable for purification from N. benthamiana or Arabidopsis, probably because of virus aggregation (data not shown). Thus, the original protocol was slightly modified for these hosts. In brief, 10 g of N. benthamiana- or Arabidopsis-infected leaves collected 3 weeks postinoculation was homogenized with 50 ml (Arabidopsis) or 20 ml (N. benthamiana) of 0.18 M McIlvain's citric acid-phosphate buffer, pH 7, containing 0.2% thioglycolic acid, 0.01 M sodium diethyldithiocarbamate, 0.5 M urea, and 3 mM EDTA first for 5 min with a mortar and pestle and then for 10 min with a Waring blender at low speed. Next, 50 ml (Arabidopsis) or 10 ml (N. benthamiana) of cold chloroform was added to the mixture and shaken in the Waring blender for another 5 min. The homogenate was centrifuged for 11 min at 5,000 x g, and the supernatant was centrifuged for 2 h at 82,500 x g. The pellet was resuspended in 3 ml (Arabidopsis) or 4 ml (N. benthamiana) of 10 mM McIlvain's citric acid-phosphate buffer, pH 7, containing 1 M urea and 0.2% ?-mercaptoethanol for 2 h and centrifuged for 10 min (Arabidopsis) or 5 min (N. benthamiana) at 1,500 x g. The supernatant was centrifuged for 1.5 h at 57,000 x g (Arabidopsis) or 82,000 x g (N. benthamiana). The resulting pellet was resuspended for 2 h in 0.5 ml (Arabidopsis) or 0.6 ml (N. benthamiana) of 0.1 M sodium borate buffer, pH 8.2, containing 10 mM EDTA, clarified by centrifugation for 5 min at 1,200 x g (Arabidopsis) or 1,500 x g (N. benthamiana), layered over a 3-ml cushion of 20% sucrose in the same buffer, and centrifuged at 72,000 x g for 2 h. The pellet was resuspended in 50 μl (Arabidopsis) or 100 μl (N. benthamiana) of 5 mM sodium borate buffer, pH 8.2, and clarified by centrifugation for 10 min at 4,800 x g (Arabidopsis) or 7,400 x g (N. benthamiana). All the purification steps were carried out at 4°C, and the purified virus was stored at –20°C.
Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) analysis of PPV CP digested with trypsin. Approximately 10 μg of purified PPV virions was digested with 2 ng of modified porcine trypsin (Promega) in a buffer containing 25 mM ammonium carbonate, pH 8, in a reaction volume of 10 μl. Digestion proceeded at 37°C for 20 min, and then it was rapidly stopped by adding 1 μl of 0.5% trifluoroacetic acid. Reaction products were desalted with a Zip-Tip reverse-phase C18 column (Millipore) and eluted in 5 μl of 70% aqueous acetonitrile and 0.1% trifluoroacetic acid.
About 0.4 μl of matrix solution (5 g/liter 2,5-dihidroxibenzoic acid in 33% aqueous acetonitrile and 0.1% trifluoroacetic acid) was deposited onto a 400-μm AnchorChip MALDI target (Bruker Daltonics, Bremen, Germany) and allowed to dry at room temperature. Then, 0.4 μl of the tryptic peptide mixture was added and again allowed to dry at room temperature. Peptide mass fingerprinting spectra were acquired with a MALDI-TOF Bruker Reflex IV mass spectrometer (Bruker Daltonics) equipped with a nitrogen laser (337 nm). Analyses were performed in reflector positive ion mode, accumulating 120 shots, with 400 ns of pulsed ion delayed extraction and an acceleration voltage of 20 kV and 1,600 V in the reflector detector. The mass spectra were externally calibrated using a mixture of peptide standards (angiotensin II, substance P, bombesin, somatostatine 28, and cytochrome C).
Processing of the spectra and data analysis were performed with Bruker Daltonics XTOF 5.1.1 and Biotools 2.1 software.
Assessment of spread and accumulation of PPV. Spreading of PPV-NK-GFP was assessed by visualizing green fluorescent protein (GFP) fluorescence under a Leica MZ FLIII fluorescence microscope with excitation and barrier filters of 480/40 nm and 510 nm, respectively. PPV accumulation was quantified by double-antibody sandwich indirect enzyme-linked immunosorbent assay (ELISA) using the REALISA kit (C. C. Durviz, S.L.). The samples were prepared by grinding the infected leaves with a mortar and pestle in 5 mM sodium phosphate buffer, pH 7.2 (2 ml/g), and storing them frozen at –20°C. A standard curve was obtained with different amounts of purified PPV virions diluted in the extract of healthy plants. The enzymatic reaction was developed with the Sigma FAST p-nitrophenyl phosphate tablet set system (Sigma), and the optical density at 405 nm of the reaction product was measured in a Titertek Multiskan MCC/340 (Labsystems) spectrophotometer. Statistical analysis of the data was carried out with PRISM 4 software (GraphPad).
RESULTS
The N-terminal region of PPV CP is O-GlcNAcylated in Nicotiana benthamiana and Arabidopsis thaliana. Mass spectrometry analyses have revealed the existence of O-GlcNAcylation at the N-terminal region of the CP of PPV virions purified from N. clevelandii (3). In order to assess whether this modification could also take place in other plant species, PPV was purified from another well-characterized PPV host, N. benthamiana, and from two different ecotypes of Arabidopsis, Col-0 and WS, which has been recently shown to be susceptible to PPV infection (unpublished results from several laboratories). PPV virions purified from infected plants of these species were partially digested with trypsin, and the resulting peptides were analyzed by MALDI-TOF. Similar to the previously published spectrum of PPV purified from N. clevelandii (3), the spectra of the three samples showed three ions corresponding to the partially digested tryptic peptide spanning residues 1 to 39 of the PPV CP (Fig. 1 and 2). One ion was the size expected from the unmodified peptide (m/z = 4,077 Da), and the other two ions were the sizes expected if the peptide was modified with one or two O-GlcNAc residues (m/z = 4,280 Da and 4,483 Da, respectively) (Fig. 2). These results indicate that similar O-GlcNAcylation of PPV CP appears to take place in different host plants.
PPV infection of sec plants is impaired relative to that of wild-type or spy plants. Arabidopsis has two OGTs: SPY and SEC. SEC protein is more similar to animal OGTs than to SPY, while SPY shows the same level of similarity with SEC and animal OGTs (9, 14). Both SEC (9) and SPY (28) have shown to have OGT activity. Two T-DNA insertion mutants of SEC have been described: sec-1 is in a WS background and the T-DNA is inserted into the tetratricopeptide repeat domain, and sec-2 is in a Col-0 background and the T-DNA is inserted into an intron adjacent to exons coding for the putative catalytic region of the protein (9). While sec-1 and sec-2 have no obvious phenotypes (9), spy plants have defects in a number of processes, including gibberellin and cytokinin responses, flowering, circadian regulation, and light inhibition of hypocotyls elongation (7, 15, 27, 30). spy-3 is in a Col-0 background and has a Gly-to-Ser substitution in the C-terminal region of the protein (14). Interestingly, homozygous sec-1/spy-3 and sec-2/spy-3 double mutants die during embryogenesis, indicating that SEC and SPY have an overlapping and essential function during embryogenesis (9).
In order to investigate the possible role of plant OGTs in the O-GlcNAcylation of PPV CP and in virus infection, sec-1, sec-2, and spy-3 mutant plants were inoculated with PPV-NK-GFP, a recombinant PPV expressing the GFP, which allows an easier monitoring of the infection (5). PPV caused in wild-type Col-0 and WS mild symptoms consisting of leaf chlorosis, rosette gathering, curling of cauline leaves, and shortening of the inflorescence stems (data not shown). Similar symptoms were observed in the three mutant plants inoculated with PPV-NK-GFP, especially late in the infection. Although nearly 100% of the inoculated plants became infected, examination of GFP localization in the leaves detected differences in the pattern of infection between the different genotypes (Fig. 3). Radiation of the infection from the major veins and the extent of infection of the lamina were reduced in both sec alleles relative to the corresponding wild type. At 12 days postinoculation (dpi), a similar number of leaves from spy-3 and wild-type plants were infected (Fig. 4) but the percentage of sec-1 and sec-2 plants with detectable GFP expression was slightly, but reproducibly, reduced. The sec-1 and sec-2 alleles had a larger effect on the spread of an infection within a leaf. Large GFP fluorescence areas were observed in a smaller proportion of leaves from sec-1 and sec-2 plants than in wild-type or spy-3 plants (Fig. 3 and 4). The differences in the extent of virus spread among the different genotypes had decreased by 19 dpi (Fig. 3 and 4).
In order to have a more quantitative assessment of the progress of PPV infection in the different genotypes, the amount of virus in the infected leaves was determined by ELISA. At 12 dpi, PPV accumulation was significantly lower in sec-1 and sec-2 plants than in their corresponding wild-type plants (Fig. 5). In contrast, virus titer was not affected in spy-3. In agreement with the experiments using GFP to monitor the infection, the reduction in virus accumulation in sec plants was less pronounced at 19 dpi (Fig. 5). The infectivity of the PPV virions produced in the wild-type and the mutant plants was assessed by a local lesion assay in Chenopodium foetidum (19). Virions produced in sec-1 and sec-2 plants appeared to be slightly less infectious than those produced in the corresponding wild-type Arabidopsis plants. However, due to the rather low linearity and accuracy of the local lesion assay, the differences were not statistically significant (data not shown).
The O-GlcNAcylation of the N-terminal region of PPV-CP is prevented by sec but not by spy mutations. PPV was purified from infected sec-1, sec-2, and spy-3 plants. The purified virions were subjected to partial trypsin digestion, and the resulting peptides were analyzed by MALDI-TOF. The spectrum of the spy-3 sample showed the presence of the peptide from amino acids 1 through 39 in the same three forms, nonglycosylated, mono-, and di-O-GlcNAcylated, found in the spectrum of virions purified from wild-type Arabidopsis (Fig. 2 and Table 1). In contrast, only nonglycosylated peptide from amino acids 1 through 39 was detected in samples prepared from virions purified from sec-1 and sec-2 plants (Fig. 2 and Table 1), indicating that SPY OGT or other plant proteins were not able to glycosylate the N-terminal region of PPV CP in the absence of the SEC protein.
The presence of acetylated forms (mass increase of 42 Da) of the nonglycosylated and O-GlcNAcylated variants of the peptide from amino acids 1 through 39 have been previously reported (3). These acetylated peptides were also detected in the virions purified from N. benthamiana and wild-type and mutant Arabidopsis (Fig. 2 and Table 1). A careful examination of the spectra also detected some signals that could correspond to phosphorylated (mass increase of 80 Da) and/or phosphorylated plus acetylated (mass increase of 80 plus 42 Da) forms of the nonglycosylated peptide from amino acids 1 through 39 in all the spectra (Fig. 2 and Table 1). Moreover, we also detected peptides with the size expected for the peptide from amino acids 1 through 39 mono-O-GlcNAcylated and phosphorylated among the tryptic peptides of virions purified from N. benthamiana, wild-type Arabidopsis Col-0 and WS, and spy-3 plants (Fig. 2 and Table 1). Interestingly, faint peaks that could correspond to a phosphorylated form of the di-O-GlcNAcylated peptide from amino acids 1 through 39 were found in the spectra of virions purified from N. benthamiana and wild-type Arabidopsis Col-0 (Fig. 2 and Table 1), suggesting that at least three Ser/Thr residues of the peptide from amino acids 1 through 39 can be modified at the same time.
DISCUSSION
O-GlcNAc modification has been shown to play very important regulatory roles in animals (32), and in agreement with this, deletion of the OGT gene of mouse was lethal (23). Loss of either SPY or SEC function alone does not cause lethality. However, sec spy double mutants died during embryogenesis, showing that OGT activity is also essential for plant viability and that SEC and SPY have at least one overlapping function (9). The extent of overlap in the function between SPY and SEC is unknown. While spy plants exhibit a number of obvious defects, including altered gibberellin and cytokinin responses, circadian rhythms, light responses, and flowering time (7, 15, 27, 30), these defects are not detectable or are much weaker in sec plants (9; L. Hartweck and N. E. Olszewski, unpublished results). These results can be explained if SPY and SEC have completely overlapping substrates but SPY is responsible for most of the O-GlcNAcylation or if SPY has some substrates that are not modified by SEC.
Very little is known of the OGT targets in plant cells. O-glycosylation with the terminal GlcNAc modification of several nuclear pore complex proteins of tobacco, including a protein that shows sequence similarity to bacterial aldose-1-epimerases, has been described previously (11), but the enzyme responsible for these modifications has not been identified. We have previously reported that the CP of a plant virus, PPV, was O-GlcNAc modified in N. clevelandii (3); now, we show that this modification is not specific for this host, since it takes place in a similar way in two other plant species, N. benthamiana and Arabidopsis (Fig. 2). More important, the MALDI-TOF analysis of PPV virions purified from sec and spy mutants indicates that SEC protein is the OGT responsible for the O-GlcNAc modification of the N-terminal region of PPV CP and that neither SPY nor other plant proteins is able to glycosylate this sequence (Fig. 2). However, we cannot rule out the possibility that SPY could be involved in the O-GlcNAc modification of other regions of PPV CP or other PPV proteins. These results support the hypothesis that SEC and SPY have partial functional independence. Whereas O-GlcNAc modification of the N-terminal region of PPV CP, like that of animal proteins, adds single sugar monomers (3; J. J. Pérez, S. Juárez, and J. A. García, unpublished results), the size of sugar chains of the glycosylated nuclear pore proteins of tobacco corresponds to more than five monosaccharides (10), suggesting that different enzymes could be involved in each of these modifications. Having in mind that SPY is less similar to mammalian OGTs than SEC, it is tempting to speculate that SPY could be involved in the second type of modification, which would be specific to plants.
The ability of PPV to infect sec-1 and sec-2 mutants (Fig. 3) demonstrates that O-GlcNAc modification by SEC OGT is not essential for PPV viability in Arabidopsis. However, the efficiency of PPV infection is markedly lower in these mutants than in the spy-3 mutant or in wild-type plants (Fig. 4 and 5), indicating that SEC activity plays an important role in infection. O-GlcNAc modifications at the segment from amino acids 1 through 39 of PPV CP are not essential, since deletions or substitutions that remove all Thr and Ser residues of this region have no perceptible effects on PPV infection in N. clevelandii (3, 4). Therefore, O-GlcNAc modification of other portions of PPV CP or other PPV proteins by SEC may play a role in the infection process. The present data do not allow us to discriminate whether some O-GlcNAc modifications are more important than others or the overall level of O-GlcNAcylation is the relevant factor. Moreover, we cannot rule out the possibility that the effect of sec mutation on PPV infection might be an indirect result of the action of SEC on a host protein. It is also important to point out that, although O-GlcNAc modification by SEC is not essential for the PPV infection of Arabidopsis, the possibility exists that O-GlcNAcylation plays a more important role in some plants (for instance, in the PPV natural woody hosts) than in others.
The role of O-GlcNAc modification in PPV infection is still an open question. It is well known that in mammals, O-GlcNAcylation and phosphorylation collaborate in the regulation of macromolecular interactions that control the activity of many cellular processes (26). PPV CP has been shown to contain phosphoserine and phosphothreonine residues (3). Although MALDI-TOF does not allow a precise analysis of protein phosphorylation, the MALDI-TOF spectra of PPV CP subjected to trypsin digestion showed signals that could correspond with phosphorylated peptides (Fig. 2 and Table 1). As expected, the presence of these putative phosphorylated peptides was not affected by the sec or spy mutations. Both nonglycosylated and O-GlcNAc-modified peptides appeared to be phosphorylated, in agreement with the fact previously observed in mammalian organisms that, although O-GlcNAcylation and phosphorylation are reciprocal modifications, a single polypeptide can contain simultaneously phosphate and O-GlcNAc residues (2). These data suggest that the O-GlcNAcylation of PPV CP, like that of those mammalian systems, could function in coordination with phosphorylation.
The genomic RNA is a central player in the most critical processes of viral infection: translation, replication, encapsidation, and cell-to-cell and long-distance movement. Thus, it is likely that allocation of RNA molecules to each of these processes is highly regulated. Ivanov and colleagues have demonstrated that phosphorylation of the CP of the potyvirus potato virus A reduces its ability to bind RNA and have suggested that this modification might regulate the formation and/or stability of virions and other viral ribonucleoproteins (12, 13). It is tempting to speculate that CP O-GlcNAc modification could also contribute to this regulatory mechanism. The fact that the effect of the sec alleles on PPV infection appears to be more apparent at early times (Fig. 3 through 5) suggests that the regulation of RNA distribution could be required at the beginning of the infection, when there could be a limiting amount of RNA molecules. An alternative possibility is that O-GlcNAc residues modifying PPV CP could be playing a role in virion stability similar to that proposed for the galactose and fucose residues that are O-linked to the N-terminal NAcSer of the Potato virus X CP, which has been shown to affect the water-absorbing capacity of the viral particles (1). It is striking that plant RNA viruses have adopted different O-glycosylation systems to modify their CPs; it will be interesting in the future to determine whether the role of SEC is PPV specific or it also modifies the CPs of other potyviruses or of viruses of other families.
The results described in this report identify a novel target for antiviral action. We have demonstrated that the O-GlcNAc modification of PPV takes place in several plant hosts and that viral propagation is compromised by sec mutations that do not noticeably affect plant growth. This could contribute to control the propagation of the virus in field conditions.
ACKNOWLEDGMENTS
We thank Elvira Domínguez for technical assistance and Juan Pablo Albar for helpful advice in MALDI-TOF analysis.
This work was supported by grants BIO2001-1434 and BIO2004-02687 from the Spanish MEC, QLK2-CT-2002-01050 and SP22-CT-2004 from the European Union to J.A.G., MCB-0112826 from the National Science Foundation, and DE-FG01-04ER04-02 from the U.S. Department of Energy to N.E.O.
REFERENCES
Baratova, L. A., N. V. Fedorova, E. N. Dobrov, E. V. Lukashina, A. N. Kharlanov, V. V. Nasonov, M. V. Serebryakova, S. V. Kozlovsky, O. V. Zayakina, and N. P. Rodionova. 2004. N-terminal segment of potato virus X coat protein subunits is glycosylated and mediates formation of a bound water shell on the virion surface. Eur. J. Biochem. 271:3136-3145.
Comer, F. I., and G. W. Hart. 2000. O-glycosylation of nuclear and cytosolic proteins. Dynamic interplay between O-GlcNAc and O-phosphate. J. Biol. Chem. 275:29179-29182.
Fernández-Fernández, M. R., E. Camafeita, P. Bonay, E. Méndez, J. P. Albar, and J. A. García. 2002. The capsid protein of a plant single-stranded RNA virus is modified by O-linked N-acetylglucosamine. J. Biol. Chem. 277:135-140.
Fernández-Fernández, M. R., J. L. Martínez-Torrecuadrada, J. I. Casal, and J. A. García. 1998. Development of an antigen presentation system based on plum pox potyvirus. FEBS Lett. 427:229-235.
Fernández-Fernández, M. R., M. Mouri?o, J. Rivera, F. Rodríguez, J. Plana-Durán, and J. A. García. 2001. Protection of rabbits against rabbit hemorrhagic disease virus by immunization with the VP60 protein expressed in plants with a potyvirus-based vector. Virology 280:283-291.
Gonzalez, S. A., and O. R. Burrone. 1991. Rotavirus NS26 is modified by addition of single O-linked residues of N- acetylglucosamine. Virology 182:8-16.
Greenboim-Wainberg, Y., I. Maymon, R. Borochov, J. Alvarez, N. Olszewski, N. Ori, Y. Eshed, and D. Weiss. 2005. Cross talk between gibberellin and cytokinin: the arabidopsis GA response inhibitor SPINDLY plays a positive role in cytokinin signaling. Plant Cell 17:92-102.
Greis, K. D., W. Gibson, and G. W. Hart. 1994. Site-specific glycosylation of the human cytomegalovirus tegument basic phosphoprotein (UL32) at serine 921 and serine 952. J. Virol. 68:8339-8849.
Hartweck, L. M., C. L. Scott, and N. E. Olszewski. 2002. Two O-linked N-acetylglucosamine transferase genes of Arabidopsis thaliana L. Heynh. have overlapping functions necessary for gamete and seed development. Genetics 161:1279-1291.
Heese-Peck, A., R. N. Cole, O. N. Borkhsenious, G. W. Hart, and N. V. Raikhel. 1995. Plant nuclear pore complex proteins are modified by novel oligosaccharides with terminal N-acetylglucosamine. Plant Cell 7:1459-1471.
Heese-Peck, A., and N. V. Raikhel. 1998. A glycoprotein modified with terminal N-acetylglucosamine and localized at the nuclear rim shows sequence similarity to aldose-1-epimerases. Plant Cell 10:599-612.
Ivanov, K. I., P. Puustinen, R. Gabrenaite, H. Vihinen, L. R?nnstrand, L. Valmu, N. Kalkkinen, and K. M?kinen. 2003. Phosphorylation of the potyvirus capsid protein by protein kinase CK2 and its relevance for virus infection. Plant Cell 15:2124-2139.
Ivanov, K. I., P. Puustinen, A. Merits, M. Saarma, and K. M?kinen. 2001. Phosphorylation down-regulates the RNA binding function of the coat protein of potato virus A. J. Biol. Chem. 276:13530-13540.
Jacobsen, S. E., K. A. Binkowski, and N. E. Olszewski. 1996. SPINDLY, a tetratricopeptide repeat protein involved in gibberellin signal transduction in Arabidopsis. Proc. Natl. Acad. Sci. USA 93:9292-9296.
Jacobsen, S. E., and N. E. Olszewski. 1993. Mutations at the SPINDLY locus of Arabidopsis alter gibberellin signal transduction. Plant Cell 5:887-896.
Laín, S., J. L. Riechmann, E. Méndez, and J. A. García. 1988. Nucleotide sequence of the 3' terminal region of plum pox potyvirus RNA. Virus Res. 10:325-342.
Liu, F., K. Iqbal, I. Grundke-Iqbal, G. W. Hart, and C. X. Gong. 2004. O-GlcNAcylation regulates phosphorylation of tau: a mechanism involved in Alzheimer's disease. Proc. Natl. Acad. Sci. USA 101:10804-10809.
López-Moya, J. J., M. R. Fernández-Fernández, M. Cambra, and J. A. García. 2000. Biotechnological aspects of plum pox virus. J. Biotechnol. 76:121-136.
López-Moya, J. J., and J. A. García. 2000. Construction of a stable and highly infectious intron-containing cDNA clone of plum pox potyvirus and its use to infect plants by particle bombardment. Virus Res. 68:99-107.
Mullis, K. G., R. S. Haltiwanger, G. W. Hart, R. B. Marchase, and J. A. Engler. 1990. Relative accessibility of N-acetylglucosamine in trimers of the adenovirus types 2 and 5 fiber proteins. J. Virol. 64:5317-5323.
Revers, F., O. Le Gall, T. Candresse, and A. J. Maule. 1999. New advances in understanding the molecular biology of plant/potyvirus interactions. Mol. Plant-Microbe Interact. 12:367-376.
Riechmann, J. L., S. Laín, and J. A. García. 1992. Highlights and prospects of potyvirus molecular biology. J. Gen. Virol. 73:1-16.
Shafi, R., S. P. Iyer, L. G. Ellies, N. O'Donnell, K. W. Marek, D. Chui, G. W. Hart, and J. D. Marth. 2000. The O-GlcNAc transferase gene resides on the X chromosome and is essential for embryonic stem cell viability and mouse ontogeny. Proc. Natl. Acad. Sci. USA 97:5735-5739.
Shukla, D. D., P. M. Strike, S. L. Tracy, K. H. Gough, and C. W. Ward. 1988. The N and C termini of the coat proteins of potyviruses are surface-located and the N-terminus contains the major virus-specific epitopes. J. Gen. Virol. 69:1497-1508.
Shukla, D. D., C. W. Ward, and A. A. Brunt. 1994. Genome structure, variation and function, p. 74-112. In D. D. Shukla, C. W. Ward, and A. A. Brunt (ed.), The potyviridae. CAB International, Cambridge, United Kingdom.
Slawson, C., and G. W. Hart. 2003. Dynamic interplay between O-GlcNAc and O-phosphate: the sweet side of protein regulation. Curr. Opin. Struct. Biol. 13:631-636.
Sothern, R. B., T. S. Tseng, S. L. Orcutt, N. E. Olszewski, and W. L. Koukkari. 2002. GIGANTEA and SPINDLY genes linked to the clock pathway that controls circadian characteristics of transpiration in Arabidopsis. Chronobiol. Biol. 19:1005-1022.
Thornton, T. M., L. Kreppel, G. W. Hart, and N. E. Olszewski. 1999. Genetic and biochemical analysis of Arabidopsis SPY, p. 445-448. In A. Altman, M. Ziv, and S. Izhar (ed.), Plant biotechnology and in vitro biology in the 21st century. Kluwer Academic Publishers, Dordrecht, The Netherlands.
Thornton, T. M., S. M. Swain, and N. E. Olszewski. 1999. Gibberellin signal transduction presents...the SPY who O-GlcNAc'd me. Trends Plant Sci. 4:424-428.
Tseng, T. S., P. A. Salomé, C. R. McClung, and N. E. Olszewski. 2004. SPINDLY and GIGANTEA interact and act in Arabidopsis thaliana pathways involved in light responses, flowering, and rhythms in cotyledon movements. Plant Cell 16:1550-1563.
Vosseller, K., L. Wells, M. D. Lane, and G. W. Hart. 2002. Elevated nucleocytoplasmic glycosylation by O-GlcNAc results in insulin resistance associated with defects in Akt activation in 3T3-L1 adipocytes. Proc. Natl. Acad. Sci. USA 99:5313-5318.
Wells, L., and G. W. Hart. 2003. O-GlcNAc turns twenty: functional implications for post-translational modification of nuclear and cytosolic proteins with a sugar. FEBS Lett. 546:154-158.
Wells, L., K. Vosseller, and G. W. Hart. 2001. Glycosylation of nucleocytoplasmic proteins: signal transduction and O-GlcNAc. Science 291:2376-2378.
Whitford, M., and P. Faulkner. 1992. A structural polypeptide of the baculovirus Autographa californica nuclear polyhedrosis virus contains O-linked N-acetylglucosamine. J. Virol. 66:3324-3329.(D. Chen, S. Juárez, L. Ha)