Analysis of the in planta antiviral activity of elderberry ribosome-inactivating proteins Frank Vandenbussche 1 , Stijn Desmyter 1 , Marialibera Ciani 2 , Paul Proost 3 , Willy J. Peumans 1 and Els J. M. Van Damme 4 1 Laboratory for Phytopathology and Plant Protection, Katholieke Universiteit Leuven, Belgium; 2 Dipartimento di Patologia sperimentale, Universita ` di Bologna, Italy; 3 Rega Institute, Department of Microbiology and Immunology, Katholieke Universiteit Leuven, Belgium; 4 Department of Molecular Biotechnology, Gent University, Belgium Although the type-2 ribosome-inactivating proteins (SNA-I, SNA-V, SNLRP) from elderberry (Sambucus nigra L.) are all devoid of rRNA N-glycosylase activity towards plant ribosomes, some of them clearly show polynucleotide– adenosine glycosylase activity towards tobacco mosaic virus RNA. This particular substrate specificity was exploited to further unravel the mechanism underlying the in planta antiviral activity of ribosome-inactivating proteins. Trans- genic tobacco (Nicotiana tabacum L. cv Samsun NN) plants expressing the elderberry ribosome-inactivating proteins were generated and challenged with tobacco mosaic virus in order to analyze their antiviral properties. Although some transgenic plants clearly showed antiviral activity, no clear correlation was observed between in planta antiviral activity of transgenic tobacco lines expressing the different ribosome- inactivating proteins and the in vitro polynucleotide– adenosine glycosylase activity of the respective proteins to- wards tobacco mosaic virus genomic RNA. However, our results suggest that the in planta antiviral activity of some ribosome-inactivating proteins may rely on a direct mech- anism on the virus. In addition, it is evident that the working mechanism proposed for pokeweed antiviral protein cannot be extrapolated to elderberry ribosome-inactivating proteins because the expression of SNA-V is not accompanied by induction of pathogenesis-related proteins. Keywords: elderberry; polynucleotide–adenosine glycosylase activity; ribosome-inactivating protein; Sambucus nigra; viral protection. Ribosome-inactivating proteins (RIPs; EC 3.2.2.22) are a heterogeneous family of structurally and evolutionary related plant proteins sharing a common functional domain that catalytically removes a specific adenine residue from a highly conserved, surface-exposed stem-loop structure found in the large rRNA of prokaryotic and eukaryotic ribosomes [1,2]. At present, they are subdivided on the basis of the structure of the genes and the corresponding proteins into holo-RIPs and chimero-RIPs [3]. Whereas holo-RIPs consist exclusively of a single catalytically active protomer of either one (classical type-1 RIPs) or two smaller polypeptide chains (e.g. maize RIP b-32), chimero-RIPs are built up of chimeric protomers with an N-terminal catalytically active domain arranged in tandem with a structurally and functionally unrelated C-terminal domain (classical type-2 and type-3 RIPs). Biochemical and molecular studies have shown that the elderberry tree expresses a complex mixture of type-2 RIPs and/or lectins in virtually all tissues. In agreement with the chronological order of their discovery, these Sambucus nigra agglutinins (SNAs) are numbered SNA-I to SNA-V. The first elderberry lectin was identified in bark tissue and described as a NeuAc(a-2,6)Gal/GalNAc-specific agglutinin (called SNA-I) [4,5]. Although already discovered in 1984, SNA-I was recognized as a type-2 RIP only when the corresponding gene was cloned in 1996 [6]. Besides SNA-I, elderberry bark contains a second NeuAc(a-2,6)Gal/Gal- NAc-specific agglutinin which shares 77% sequence simi- larity with SNA-I (but has a different oligomeric organization) and was called SNA-I¢ [7]. The second lectin identified in elderberry bark was a GalNAc-specific agglu- tinin, called SNA-II with no obvious relation to SNA-I [5]. However, molecular cloning of a GalNAc-specific elder- berry bark type-2 RIP (called SNA-V) later revealed that SNA-II consists of subunits that correspond to slightly truncated B chains of this genuine type-2 RIP. Both SNA-V and SNA-II are derived from a single precursor, through differential processing [8]. It should be mentioned here that SNA-V (or a very closely related paralog) was first described by Girbes et al. [9] as nigrin b. After the identification of the bark lectins SNA-I and SNA-II, two other elderberry lectins called SNA-III and SNA-IV were isolated from seeds and Correspondence to E. J. M. Van Damme, Department of Molecular Biotechnology, Ghent University, Coupure Links 653, 9000 Ghent, Belgium. Fax: + 32 9264 6219, Tel.: + 32 9264 6086, E-mail: ElsJM.VanDamme@UGent.be Abbreviations: PAG, polynucleotide–adenosine glycosylase; PAP, pokeweed antiviral protein; PR, pathogenesis-related; RIP, ribosome- inactivating protein; SNA, Sambucus nigra agglutinin; SNLRP, Sambucus nigra lectin-related protein; TMV, tobacco mosaic virus. Enzyme: Ribosome-inactivating protein, rRNA N-glycosylase (EC 3.2.2.22). (Received 16 January 2004, revised 24 February 2004, accepted 27 February 2004) Eur. J. Biochem. 271, 1508–1515 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04059.x fruits, respectively. Molecular cloning revealed that SNA- IV, which is the most abundant fruit protein, is a dimeric GalNAc-specific lectin encoded by a truncated type-2 RIP gene (with a major deletion comprising almost the whole A chain) [10]. Besides the different SNAs, the bark of elderberry contains an additional protein, SNLRP, which is both structurally and evolutionary closely related to the other elderberry type-2 RIPs but possesses a B chain which, because of several amino-acid substitutions in the sugar- binding sites, is devoid of carbohydrate-binding activity [11]. Initially, RIPs were thought to act exclusively on ribosomes or rRNA through their rRNA N-glycosylase activity. However, using a highly sensitive HPLC-fluores- cence-based method, Barbieri et al. [12] showed that several RIPs release more than one adenine residue from rRNA. Moreover, they found that saporin L1, a type-1 RIP from the leaves of soapwort (Saponaria officinalis L.), is capable of removing multiple adenine residues from various nucleic acid substrates including herring sperm DNA, mammalian DNA, genomic viral RNA, rRNA and poly(A) [13–15]. Additional testing of 61 RIPs revealed that all of them extensively deadenylate herring sperm DNA and that several are active towards viral genomic RNA [16–20]. On the basis of these findings it has been suggested that polynucleotide–adenosine glycosylase (PAG) activity may be responsible for the potent antiviral activity of RIPs (against both animal and plant viruses). However, it is still unclear whether the results obtained in these in vitro studies can be extrapolated to the complex environment of acell[16]. Although the antiviral activity of RIPs against plant viruses is well documented, the underlying mechanism(s) has not yet been elucidated. In principle, three possible expla- nations can be put forward [3,21]. First, RIPs may act directly on the viral nucleic acids through their PAG activity. Secondly, RIPs may act directly on the host by selectively killing the infected cells, thus preventing the virus from replicating and spreading to neighboring cells. Finally, RIPs may act indirectly through activation of the plant’s defence system. As most RIPs with a documented strong in planta antiviral activity [e.g. pokeweed antiviral protein (PAP), trichosanthin] are able to depurinate both viral nucleic acids and plant ribosomes, it has been difficult to assess how the PAG activity contributes to RIP-mediated protection. To further unravel the mode of action of RIPs, the in planta antiviral activity of a set of type-2 RIPs (SNA-I, SNA-V, SNLRP) from elderberry (S. nigra L.) which are all devoid of rRNA N-glycosylase activity towards plant ribosomes [9] but strikingly differ from each other with respect to their PAG activity towards tobacco mosaic virus (TMV) RNA, was analyzed using a TMV/tobacco (Nicotiana tabacum L. cv. Samsun NN) model system. In addition to the genuine type-2 RIP, the elderberry lectin SNA-IV, which is consid- ered a type-2 RIP without an A chain, was included in the study as a negative control. A comparison of the protection offered by the different ectopically expressed elderberry RIPs did not reveal a clear correlation between in planta antiviral activity and PAG activity towards TMV genomic RNA. However, it is evident that the working mechanism suggested for PAP cannot be extrapolated to elderberry RIPs because expression of the latter is not accompanied by induction of pathogenesis-related (PR) proteins. Materials and methods Plasmid constructions All manipulations were performed according to standard techniques [22]. The pGK vector was constructed by replacing the b-glucuronidase gene from the plant transfor- mation vector pGPTV-KAN [23] with the expression cassette of the pFF19 vector [24]. The coding sequences of the various RIPs/lectins were amplified by PCR to engineer appropriate restriction sites. The restricted PCR products were inserted between the cauliflower mosaic virus 35S promoter and polyadenylation signal of the linearized and dephosphorylated pFF19 vector. After confirmation of the sequence by dideoxy sequencing [25], inserts were subcloned into the expression cassette of the pGK plant transforma- tion vector. The resulting plasmids, pGKsnaI,pGKsnaIV, pGKsnaV and pGKsnlrp, were transferred into Agrobacte- rium tumefaciens GV3101 by electroporation [26]. Transformation of N. tabacum Transformation of tobacco (N. tabacum L. cv. Samsun NN) was performed using the leaf disc cocultivation method [27]. Transgenic shoots were selected on Murashige-Skoog medium supplemented with 0.1 mgÆL )1 a-naphthalene acetic acid, 1 mgÆL )1 6-benzylaminopurine, 100 mgÆL )1 cefotaxime, 100 mgÆL )1 carbenicillin and 100 mgÆL )1 kana- mycin (Duchefa Biochemie BV, Haarlem, the Netherlands). Transformed plants were kept in a culture room or a greenhouse at 22 °C, 50% relative humidity, and a 16 h photoperiod until use. Molecular analysis of transformants Total genomic DNA was isolated as described by Goode & Feinstein [28]. The presence of the transgenes was investi- gated with PCR using two internal primers derived from the N-terminal and C-terminal sequence of the various RIPs/ lectins. Only the PCR-positive plants were further analysed at the RNA and protein level. Total RNA was isolated as described by Eggermont et al. [29], dissolved in RNase-free water, and quantified spectro- photometrically. Approximately 30 lgtotalRNAwas denatured with glyoxal/dimethyl sulfoxide and separated on a 1.2% (w/v) agarose gel. After electrophoresis, RNA was capillary blotted on to Hybond-N + membranes (Amersham Biosciences, Uppsala, Sweden). Membranes were first probed with random-primer-labeled cDNAs encoding the different RIPs/lectins or PR proteins (PR-1, PR-2, PR-3, proteinase inhibitor II). Subsequently the membranes were reprobed with a random-primer-labeled cDNA fragment complementary to the 3¢ end of the tobacco 25S rRNA. To estimate the expression levels of the different RIPs/ lectins, transgenic lines (T 1 generation) were analysed for recombinant (r)RIP/lectin content by Western blot densi- tometry. To minimize variation, 10 selfed plants of each line were grown under identical conditions (22 °C, 50% relative humidity, 16 h photoperiod). When plants reached the six- leaf stage, the third and fourth leaf were pooled, lyophilized, and ground using mortar and pestle. Total protein from 50 mg lyophilized leaf material was extracted in 100 m M Ó FEBS 2004 Antiviral activity of ribosome-inactivating proteins (Eur. J. Biochem. 271) 1509 Hepes (pH 7.6). The protein concentration was determined using the Bio-Rad protein assay kit (Bio-Rad, Hercules, CA, USA) using BSA as standard [30]. For Western blot analysis,  2.5 lg(SNA-V),7.5 lg (SNLRP) or 15 lg (SNA-I, SNA- IV) total protein was separated by electrophoresis on a 15% (w/v) SDS/polyacrylamide gel and transferred to an Immo- bilon-P membrane (Millipore, Bedford, MA, USA) using a Trans-blot SD semidry transfer cell (Bio-Rad). Immuno- detection of the recombinant proteins was performed as described by Desmyter et al. [31] using affinity-purified polyclonal rabbit antibodies raised against native SNA-I, SNA-II or SNLRP as the primary antibody. Purification of recombinant proteins rRIPs were isolated using a combination of classical protein purification techniques and affinity chromatography (except for SNLRP, which possesses no sugar-binding activity). For the purification of recombinant SNA-I, leaves of SNA-I transformants were homogenized in a solution of 1 gÆL )1 ascorbic acid using a Waring blender. After centrifugation at 3000 g for 10 min, 1 gÆL )1 CaCl 2 was added to the supernatant, and the pH adjusted to 9.0 with 0.5 M NaOH. The extract was cleared by centrifugation at 8000 g for 10 min and filtered through Whatman 3MM filter paper. The filtrate was subsequently adjusted to pH 2.8 with 0.5 M HCl and loaded on to a column (2.6 cm · 10 cm; 50 mL bed volume) of S Fast Flow (Amersham Biosciences, Uppsala, Sweden) equilibrated with 20 m M acetic acid. After loading, the column was washed with 20 m M acetic acid until the A 280 fell below 0.01, and the bound proteins were eluted with 300 mL 0.1 M Tris/HCl (pH 8.7) contain- ing 0.5 M NaCl. This partially purified protein fraction was adjusted to pH 7.0 with 0.5 M NaOH and loaded on a column (1.6 cm · 5 cm; 10 mL bed volume) of fetuin– Sepharose 4B equilibrated with 0.2 M NaCl. After the columnhadbeenwashedwith0.2 M NaCl, the bound lectin was desorbed with 20 m M 1,3-diaminopropane (pH 9.0) andstoredat)20 °C until use. For the purification of recombinant SNA-IV and SNA- V, leaves of SNA-IV and SNA-V transformants were homogenized in 20 m M citric acid (pH 3.0) using a Waring blender. The crude extract was cleared by centrifugation at 3000 g for 10 min and filtered through Whatman 3MM filter paper. (NH 4 ) 2 SO 4 was added to a final concentration of 1.5 M and the pH adjusted to 7.0 with 0.5 M NaOH. The extract was then applied to a column (2.6 cm · 10 cm; 50 mL bed volume) of galactose–Sepharose 4B equilibrated with 1.5 M (NH 4 ) 2 SO 4 . After the column had been washed with 1.5 M (NH 4 ) 2 SO 4 until the A 280 fell below 0.01, bound proteins were desorbed with 0.2 M galactose in 1.5 M (NH 4 ) 2 SO 4 . The affinity-purified proteins were loaded on a column (1.6 · 10 cm; 20 mL bed volume) of phenyl- Sepharose (Amersham Biosciences) equilibrated with 1.5 M (NH 4 ) 2 SO 4 . After the column had been washed with 1.5 M (NH 4 ) 2 SO 4 , the protein was desorbed with 20 m M 1,3- diaminopropane (pH 9.0). Fractions (2.5 mL each) were collected and analysed by SDS/PAGE. Peak fractions containing the RIPs/lectins were pooled and stored at )20 °C until use. For the purification of rSNLRP, leaves from SNLRP transformants were homogenized in 50 m M acetic acid using a Waring blender. After adjustment of the pH to 3.0 with 0.5 M HCl, the homogenate was centrifuged at 3000 g for 10 min and filtered through Whatman 3MM filter paper. The cleared filtrate was then loaded on a column (2.6 cm · 10 cm; 50 mL bed volume) of S Fast Flow (Amersham Biosciences) equilibrated with 20 m M acetic acid. After the column had been washed with 100 mL 25 m M sodium formate (pH 3.8), the bound proteins were desorbed with a linear gradient (500 mL) of increasing NaCl concentration (from 0 to 0.5 M ). Fractions (5 mL each) were collected and analysed by SDS/PAGE. Peak fractions containing SNLRP were pooled, concentrated by freeze-drying, and loaded on a column (2.6 cm · 70 cm; 350 mL bed volume) of Sephacryl 100 equilibrated with 10 m M Tris/HCl (pH 7.5) for gel filtration. Finally, frac- tions (2.5 mL each) were collected and analysed with SDS/ PAGE. Peak fractions containing SNLRP were pooled and stored at )20 °C until use. Protein sequencing Purified recombinant proteins were analysed by SDS/ PAGE and transferred to ProBlott TM membranes (Applied Biosystems, Foster City, CA, USA) using a Trans-blot SD semidry transfer cell. Proteins were excised from the blots and sequenced on a Procise 491 cLC protein sequencer (Applied Biosystems). Enzyme assay The N-glycosylase activity of crude tobacco extracts or recombinant proteins towards rRNA was determined as described previously using rabbit reticulocyte ribosomes as substrate [32]. The PAG activity of the native RIPs was determined by measuring the adenine released from the substrate as described by Brigotti et al. [33]. Reactions were performed in Eppendorf tubes containing 1 pmol RIP and 5 lg substrate (TMV genomic RNA) in 50 lL PAG buffer (50 m M sodium acetate, 100 m M KCl, pH 4.0). After 40 min incubation at 30 °C, the adenine released was quantified by LC/MS on a Waters Alliance/ZQ apparatus (Waters Corporation, Milford, MA, USA). TMV bioassays Based on the molecular analysis, five transgenic lines per construct (SNA-I, SNA-IV, SNA-V, SNLRP) were ran- domly chosen to study their level of protection against TMV infection. Wild-type and transgenic (T 1 -generation) tobacco plants were grown to the six-leaf stage in a culture room at 22 °C under a 16 h photoperiod. Plants were mechanically inoculated by rubbing the upper two fully expanded leaves with a virus suspension [1 lgÆmL )1 TMV in 0.1 M KH 2 PO 4 , 2% (w/v) polyvinylpyrrolidone, pH 7.2] in the presence of carborundum. After infection, plants were maintained in a greenhouse at 22 °C, 50% relative humidity, and a 16 h photoperiod. Four days after infection the total number of lesions on both leaves was counted. To minimize variation, a complete randomised block design was used. Experiments were repeated three times. Before analysis, the dataset was transformed to the logarithmic scale. The transformed data 1510 F. Vandenbussche et al.(Eur. J. Biochem. 271) Ó FEBS 2004 were analysed using analysis of variance, and means were compared using a Tukey’s HSD multiple range test (a ¼ 0.05). Induction of PR proteins The possible constitutive expression of PR proteins in transgenic plants expressing the different RIPs/lectins was checked by Northern blot analysis. Total RNA was isolated, separated by agarose gel electrophoresis, and blotted. Blots were hybridized with specific probes for PR-1, PR-2, PR-3 and proteinase inhibitor II [34]. Results and discussion PAG activity towards TMV genomic RNA To corroborate the possible relation between the PAG and antiviral activity of RIPs, native elderberry RIPs were tested for their depurinating activity towards TMV genomic RNA (Table 1). Of the RIPs tested, both SNA-V and SNLRP exhibited weak PAG activity towards TMV genomic RNA, whereas SNA-I showed no activity under the same experi- mental conditions. This complete lack of PAG activity for SNA-I is most probably due to its intrinsically lower enzymatic activity. As previously shown, SNA-I exhibits a markedly lower PAG activity on different nucleic acid substrates than other type-2 RIPs (possibly as a result of its complex tetrameric structure) [35,36]. As none of the elderberry RIPs display enzymatic activity towards plant ribosomes in rRNA N-glycosylase activity assays [9], they are ideal candidates for assessing whether the PAG activity is involved in RIP-mediated protection. Expression of SNA-I, SNA-IV, SNA-V and SNLRP in transgenic tobacco Tobacco leaf discs were transformed with A. tumefaciens GV3101 harboring the various pGKrip/lectin vectors. The RIP/lectin-expressing plants are hereafter referred to as RIP/n where ÔRIPÕ denotes the RIP/lectin expressed and ÔnÕ the number. No ÔvisibleÕ phenotypic aberrations were observed in any of the RIP/lectin-expressing transformants, indicating that these elderberry type-2 RIPs exert no cytotoxic effects in planta. All plants obtained after transformation were analysed by PCR to confirm the presence of the RIP/lectin coding sequence in the tobacco genome. Only those plants that were positive in the PCR analysis were withheld for further analysis. Expression of the RIP/lectin transgenes was examined by analysing all transgenic lines (T 0 generation) at the RNA and protein level. Transcription products of the predicted size were detected in most transformants but never in wild-type plants (data not shown). Western blot analysis of crude leaf extracts confirmed that most of these transformants contained immunoreactive bands. As only a limited number of transformants could be analysed in the viral bioassay, five lines per construct were randomly chosen for a more detailed analysis. To estimate the expression levels of the RIPs/lectins, transgenic lines (T 1 generation) were analysed for rRIP/lectin content by Western blot densitometry (Fig. 1, Table 2). The highest expression levels were observed for the SNA-V lines, in which the RIP accounted for 0.8–5.0% of the total soluble leaf protein. All other lines exhibited markedly lower expression levels of recombinant RIPs/lectins, varying between 0.03% and 1.7% of the total soluble protein. To verify whether the rRIPs are enzymatically active, crude extracts from four different RIP/lectin-expressing transformants (SNA-I/4, SNA-IV/3, SNA-V/4 and SNLRP/2) were analysed for the presence of rRNA N-glycosylase activity. The SNA-IV transformant was included as a negative control. As shown in Fig. 2A, the characteristic Endo fragment was only detected in the SNA-V and SNLRP transformants (lanes 6 and 8), but not in the SNA-I and SNA-IV transformants (lanes 2 and 4). The lack of enzymatic activity in the SNA-I Table 1. RIP-catalysed release of adenine (pmol per pmol RIP per 40 min) from TMV genomic RNA. RIP Adenine released SNA-I 0.0 – trace SNA-V 0.45 ± 0.12 SNLRP 0.57 ± 0.16 a a Statistically significantly at P<0.02. Fig. 1. Western blot analysis of total soluble protein from wild-type and transgenic tobacco plants. Approximately 15 lg (SNA-I, SNA-IV), 7.5 lg (SNLRP) or 2.5 lg (SNA-V) total soluble protein was separ- ated in a 15% (w/v) SDS/polyacrylamide gel and transferred to a nylon membrane. Membranes were probed with polyclonal rabbit antibodies raised against the different RIPs/lectins. Sample loading was as fol- lows:lane–,wild-type;lane+,purenativeRIP/lectin;lanes1–5,RIP/ 1–5; lane 6–12, serial dilution (400–6.25 ng) of native pure RIP/lectin. Table 2. Expression levels (% of total soluble protein) of recombinant elderberry RIPs/lectins in transgenic tobacco plants in the different line numbers. RIP 1 2345 SNA-I 0.2 0.4 0.4 0.4 0.4 SNA-IV 0.03 1.0 1.7 0.4 0.4 SNA-V 4.3 0.8 3.7 5.0 2.4 SNLRP 0.3 0.8 0.1 0.8 0.6 Ó FEBS 2004 Antiviral activity of ribosome-inactivating proteins (Eur. J. Biochem. 271) 1511 transformants was presumably due to the lower expression levels in these transformants and the intrinsically lower enzymatic activity of SNA-I. Purification and characterization of recombinant proteins Recombinant SNA-I, SNA-IV, SNA-V and SNLRP were purified from leaves of transgenic tobacco plants, and their purity confirmed by SDS/PAGE and Western blot analysis. Staining of the gels with Coomassie Brilliant Blue yielded virtually identical migration patterns for the native and recombinant proteins (Fig. 3), indicating that the rRIPs/ lectins were electrophoretically pure. Moreover, the pres- ence of high-molecular-mass bands in the lanes with unreduced rSNA-I suggests that it adopts the same [A-s-s-B-s-s-B-s-s-A] 2 structure as native SNA-I. However, the presence of a faster migrating band in unreduced rSNA- I suggests that the intermolecular disulfide bridge formation between the two [A-s-s-B] pairs is less efficient in tobacco than in elderberry. Upon reduction, the high-molecular- mass bands of both rSNA-I and SNA-I disappeared giving rise to two polypeptides of nearly identical mass. All other RIPs (SNA-V, SNLRP) yielded a banding pattern typical of type-2 RIPs (i.e. characterized by the presence of two distinct polypeptides of  30 and 35 kDa, respectively), indicating that they are composed of [A-s-s-B] protomers. As antibodies to SNA-II were used to detect SNA-V, no antibodies directed against the A chain were present and accordingly only the B chain of SNA-V could be visualized in the Western blot analysis. In contrast with elderberry bark where  95% of the SNA-V precursor is converted into SNA-II and only 5% in SNA-V, the SNA-V-expressing tobacco plants contained little or no rSNA-II, indicating that in tobacco the precursor of rSNA-V is exclusively converted into the genuine type-2 RIP SNA-V. The molecular structure of the recombinant RIPs/lectins was further analysed by gel filtration on a Superose 12 column. As expected, the native and recombinant proteins were eluted at the same position (data not shown). Furthermore N-terminal sequencing of the recombinant proteins con- firmed that tobacco cells successfully recognize and cleave the signal peptide as well as the internal linker peptide at exactly the same positions as in the parent plant (data not shown). Analysis of the rRNA N-glycosylase activity using rabbit ribosomes as a substrate demonstrated that both purified rSNA-V and rSNLRP exhibited strong enzymatic activity (Fig. 2B, lanes 6 and 8) whereas no activity could be detected for the purified rSNA-IV (lane 4), which served as a negative control. However, in contrast with the results of the enzymatic assays with crude extracts, a clear signal was also observed for purified rSNA-I (lanes 2). These results clearly show that all rRIPs are enzymatically active. Similar results wererecentlyalsoreportedforSNA-IfandSNA-I¢ [37,38]. Fig. 2. rRNA N-glycosylase activity of crude extracts from five different RIP/lectin-expressing transformants (A) or purified recombinant RIPs/ lectins (B) towards rabbit reticulocyte ribosomes. The arrow indicates the position of the Endo fragment released from the rRNA. (–) and (+) indicate no treatment and aniline treatment, respectively. Sample loading was as follows: lanes 1 and 2, SNA-I/4; lanes 3 and 4, SNA-IV/ 3; lanes 5 and 6, SNA-V/4; lanes 7 and 8, SNLRP/2. Fig. 3. SDS/PAGE (left hand panel) and Western blot (right hand panel) analysis of purified native and recombinant RIPs/lectins under nonreducing (A) and reducing (B) conditions. Sample loading was as follows:lane1,SNA-I;lane2,rSNA-I;lane3,SNA-IV;lane4,rSNA- IV;lane5,SNA-V;lane6,rSNA-V;lane7,SNLRP;lane8,rSNLRP. Molecular mass markers were phosphorylase b (97 kDa), albumin (66 kDa), ovalbumin (45 kDa), carbonic anhydrase (30 kDa), trypsin inhibitor (20 kDa) and lysozyme (14 kDa). 1512 F. Vandenbussche et al.(Eur. J. Biochem. 271) Ó FEBS 2004 The results presented here and elsewhere thus indicate that tobacco is not only capable of correctly processing and assembling monomeric (SNLRP) but also dimeric (SNA-I¢, SNA-V) and even tetrameric (SNA-I, SNA-If) type-2 RIPs. TMV bioassays To assess whether and, if so, to what extent the elderberry RIPs provide in planta protection against viruses (non)trans- genic control plants and RIP-expressing tobacco plants (T 1 generation) were challenged with TMV. For each RIP, five transgenic lines were randomly chosen to be analysed in the bioassays. As SNA-IV lacks the catalytically active A chain, it is expected to confer no protection against viruses and hence serves as a negative control. To minimize variation, a complete randomised block design was used, with each block consisting of a nontransgenic control plant (wild- type), a transgenic control plant (empty-vector) and five RIP-expressing transgenic plants (RIP/1–5). Three inde- pendent experiments were set up for each RIP. Experimen- tal data were combined to calculate the mean number of lesions, which were compared using a Tukey’s HSD multiple range test (a ¼ 0.05). Significant differences between control and RIP-expressing plants were only observed for SNA-V (SNA-V/1, SNA-V/3, SNA-V/4) (Table 3). Interestingly, a direct correlation was observed between the expression level of rSNA-V and the level of protection against TMV in the different SNA-V lines. Only the three lines with the highest protection levels showed a significant reduction in lesion numbers compared with control plants. Although these results show that the type-2 RIP SNA-V is capable of conferring local protection against TMV infection, the level of protection is markedly lower than that of type-1 RIPs. On average, the three highest expressing SNA-V lines showed a reduction in lesion numbers of 39%. Using the same bioassay, Desmyter et al. [31] showed that IRIP, a type-1 RIP from Iris hollandica L., provides high levels of protection for transgenic tobacco plants, with an average reduction in lesion numbers of 75%. Moreover, all five IRIP-expressing lines exhibited similar protection levels. The observation that SNA-V provides protection only at very high expression levels (> 3.7% of the total soluble protein) suggests that the antiviral mechanism of SNA-V differs from that of IRIP. Like most type-1 RIPs, IRIP exhibits rRNA N-glycosylase activity towards both animal and plant ribosomes [39], and accordingly is capable of depurinating in planta part of the ribosomes in IRIP- expressing tobacco on TMV infection (as could be demon- strated by an in planta depurination assay) [40]. Thus, the strong in planta antiviral activity of IRIP is presumably due to its direct depurinating activity towards both viral nucleic acids and plant ribosomes and hence relies on a direct effect on both the invading viruses and the infected cells. As SNA- V does not act on plant ribosomes, the elderberry RIP can only act on the virus itself and not on the infected cells, which may explain why it provides less protection against viral infection than IRIP or other type-1 RIP. Although both SNA-V and SNLRP exhibited similar depurination rates on TMV genomic RNA (Table 1), only the highest SNA-V-expressing lines showed a significant reduction in TMV lesion numbers. However, as mentioned above, the SNA-V lines showed markedly higher expression levels than any of the other RIP-expressing tobacco lines. These results suggest that in planta depurination of viral nucleic acids requires high cellular RIP concentrations. Similar results were recently reported for the cap-specific depurination by the type-1 RIP PAP from pokeweed [41]. Using a filter-binding assay, the authors demonstrated that PAP has a nearly fourfold lower affinity for capped RNAs than for rRNA. As a consequence, PAP is only expected to significantly interact with capped RNAs at high cellular concentrations and/or high capped RNA levels. It is likely therefore that the ectopically expressed SNLRP exerts no protection in planta because its expression level remains below the threshold concentration required for activity. Analysis of PR protein expression in transgenic tobacco plants As according to previous reports the ectopic expression of PAP and various PAP mutants resulted in constitutive expression of some PR proteins [42,43], it seemed worth Table 3. Susceptibility of wild-type and transgenic tobacco plants to TMV infection. The mean number of lesions was calculated for each RIP separately and compared using a Tukey’s HSD multiple range test (a ¼ 0.05). Inhibition percentages were calculated by the following formula: 100 · (number of lesions wild-type – number of lesions transformant)/number of lesions wild-type. When the number of lesions of the transformant was as high or higher than the wild-type value, the inhibition percentage was set to zero. Tobacco line Number of plants Number of lesions Inhibition (%) Wild-type 26 151 ± 67 A 0 Empty vector 26 140 ± 47 A 7 SNA-I/1 26 113 ± 56 A 25 SNA-I/2 25 133 ± 68 A 12 SNA-I/3 26 140 ± 69 A 7 SNA-I/4 26 122 ± 57 A 19 SNA-I/5 25 159 ± 63 A 0 Wild-type 24 149 ± 60 A 0 Empty vector 24 152 ± 27 A 0 SNA-IV/1 24 158 ± 56 A 0 SNA-IV/2 24 145 ± 69 A 3 SNA-IV/3 24 131 ± 63 A 12 SNA-IV/4 24 112 ± 37 A 25 SNA-IV/5 24 123 ± 36 A 17 Wild-type 28 146 ± 77 A 0 Empty vector 28 141 ± 62 A 3 SNA-V/1 28 88 ± 40 B 40 SNA-V/2 28 160 ± 81 A 0 SNA-V/3 28 83 ± 47 B 43 SNA-V/4 28 97 ± 57B 34 SNA-V/5 26 185 ± 90 A 0 Wild-type 24 208 ± 128 A 0 Empty vector 24 167 ± 99 A 20 SNLRP/1 24 196 ± 94 A 6 SNLRP/2 24 197 ± 115 A 5 SNLRP/3 24 179 ± 93 A 14 SNLRP/4 24 237 ± 98 A 0 SNLRP/5 24 198 ± 88 A 5 Ó FEBS 2004 Antiviral activity of ribosome-inactivating proteins (Eur. J. Biochem. 271) 1513 while to check whether the same phenomenon occurs in tobacco plants that synthesize one of the elderberry type-2 RIPs/lectins. Therefore, the presence of mRNAs encoding PR-1, PR-2, PR-3 and proteinase inhibitor II was verified by Northern blot analysis. In none of the transgenic plants could any of these mRNAs be detected (data not shown), indicating that constitutive expression of neither the type-2 RIPs SNA-V, SNA-I and SNLRP nor the lectin SNA-IV was accompanied by an increase in acidic or basic PR proteins. Similar results were recently also reported for transgenic tobacco plants expressing IRIP or SNA-I¢ [31,37,38]. In spite of enhanced protection levels against TMV infection, none of these plants constitutively accu- mulated PR proteins. The results presented here and elsewhere thus clearly indicate that the proposed mode of action of PAP (i.e. through activation of the plant’s defence system) cannot be generalized to other RIPs. Acknowledgements This work was supported in part by grants from the Katholieke Universiteit Leuven and DG6 Ministerie voor Middenstand en Landbouw-Bestuur voor Onderzoek en Ontwikkeling. P. P. is a Postdoctoral Fellow of the Fund for Scientific Research-Flanders. The work performed in Bologna was partially supported by Progetto Strategico Oncologia n.74 (DD 19Ric, 09/01/02) from Ministero Istruzione Universita ` e Ricerca, Italy. The TMV strain was a gift from Professor M. Ho ¨ fte (Laboratory of Phytopathology, Ghent University, Belgium). References 1. Endo, Y., Mitsui, K., Motizuki, M. & Tsurugi, K. (1987) The mechanism of action of ricin and related toxic lectins on eu- karyotic ribosomes. The site and the characteristics of the mod- ification in 28 S ribosomal RNA caused by the toxins. J. Biol. Chem. 262, 5908–5912. 2. Endo, Y. & Tsurugi, K. (1987) RNA N-glycosidase activity of ricin A-chain. Mechanism of action of the toxic lectin ricin on eukaryotic ribosomes. J. Biol. Chem. 262, 8128–8130. 3. Van Damme, E.J.M., Hao, Q., Chen, Y., Barre, A., Vandenbus- sche, F., Desmyter, S., Rouge ´ , P. & Peumans, W.J. (2001) Ribo- some-inactivating proteins: a family of plant proteins that do more than inactivate ribosomes. Crit. Rev. Plant Sci. 20, 395–465. 4. Broekaert, W.F., Nsimba-Lubaki, M., Peeters, B. & Peumans, W.J. (1984) A lectin from elder (Sambucus nigra L.) bark. Bio- chem. J. 221, 163–169. 5. Kaku, H., Peumans, W.J. & Goldstein, I.J. (1990) Isolation and characterization of a second lectin (SNA-II) present in el- derberry (Sambucus nigra L.) bark. Arch. Biochem. Biophys. 277, 255–262. 6. Van Damme, E.J.M., Barre, A., Rouge ´ ,P.,VanLeuven,F.& Peumans, W.J. (1996) The NeuAc (alpha-2,6) -Gal/GalNAc- binding lectin from elderberry (Sambucus nigra) bark, a type-2 ribosome-inactivating protein with an unusual specificity and structure. Eur. J. Biochem. 235, 128–137. 7. Van Damme, E.J.M., Roy, S., Barre, A., Citores, L., Mostafa- pous, K., Rouge ´ , P., Van Leuven, F., Girbes, T., Goldstein, I.J. & Peumans, W.J. (1997) Elderberry (Sambucus nigra)barkcontains two structurally different Neu5Ac(alpha2,6)Gal/GalNAc-binding type 2 ribosome-inactivating proteins. Eur. J. Biochem. 245, 648–655. 8. Van Damme, E.J.M., Barre, A., Rouge ´ ,P.,VanLeuven,F.& Peumans, W.J. (1996) Characterization and molecular cloning of Sambucus nigra agglutinin V (nigrin b), a GalNAc-specific type-2 ribosome-inactivating protein from the bark of elderberry (Sam- bucus nigra). Eur. J. Biochem. 237, 505–513. 9. Girbes, T., Citores, L., Ferreras, J.M., Rojo, M.A., Iglesias, R., Munoz, R., Arias, F.J., Calonge, M., Garcia, J.R. & Mendez, E. (1993) Isolation and partial characterization of nigrin b, a non- toxic novel type 2 ribosome-inactivating protein from the bark of Sambucus nigra L. Plant Mol. Biol. 22, 1181–1186. 10. Van Damme, E.J.M., Roy, S., Barre, A., Rouge ´ ,P.,VanLeuven, F. & Peumans, W.J. (1997) The major elderberry (Sambucus nigra) fruit protein is a lectin derived from a truncated type 2 ribosome- inactivating protein. Plant J. 12, 1251–1260. 11. Van Damme, E.J.M., Barre, A., Rouge ´ ,P.,VanLeuven,F.& Peumans, W.J. (1997) Isolation and molecular cloning of a novel type 2 ribosome-inactivating protein with an inactive B chain from elderberry (Sambucus nigra)bark.J. Biol. Chem. 272, 8353–8360. 12. Barbieri, L., Ferreras, J.M., Barraco, A., Ricci, P. & Stirpe, F. (1992) Some ribosome-inactivating proteins depurinate ribosomal RNA at multiple sites. Biochem. J. 286, 1–4. 13. Barbieri, L., Gorini, P., Valbonesi, P., Castiglioni, P. & Stirpe, F. (1994) Unexpected activity of saporins. Nature (London) 372,624. 14. Barbieri, L., Valbonesi, P., Gorini, P., Pession, A. & Stirpe, F. (1996) Polynucleotide:adenosine glycosidase activity of sapor- in-L1: effect on DNA, RNA and poly (A). Biochem. J. 319, 507– 513. 15. Barbieri,L.,Valbonesi,P.,Govoni,M.,Pession,A.&Stirpe,F. (2000) Polynucleotide:adenosine glycosidase activity of saporin- L1: effect on various forms of mammalian DNA. Biochim. Biophys. Acta 1480, 258–266. 16. Barbieri, L., Valbonesi, P., Bonora, E., Gorini, P., Bolognesi, A. & Stirpe, F. (1997) Polynucleotide:adenosine glycosidase activity of ribosome-inactivating proteins: effect on DNA, RNA and poly (A). Nucleic Acids Res. 25, 518–522. 17. Bolognesi, A., Polito, L., Olivieri, F., Valbonesi, P., Barbieri, L., Battelli, M.G., Carusi, M.V., Benvenuto, E., Del Vecchio, B.F., Di Maro, A., Parente, A., Di Loreto, M. & Stirpe, F. (1997) New ribosome-inactivating proteins with polynucleotide:adenosine glycosidase and antiviral activities from Basella rubra L. & Bou- gainvillea spectabilis Willd. Planta 203, 422–429. 18. Di Maro, A., Valbonesi, P., Bolognesi, A., Stirpe, F., De Luca, P., Siniscalco, G.G., Gaudio, L., Delli, B.P., Ferranti, P., Malorni, A. & Parente, A. (1999) Isolation and characterization of four type-1 ribosome-inactivating proteins, with polynucleotide:adenosine glycosidase activity, from leaves of Phytolacca dioica L. Planta 208, 125–131. 19. Rajamohan, F., Venkatachalam, T.K., Irvin, J.D. & Uckun, F.M. (1999) Pokeweed antiviral protein isoforms PAP-I, PAP-II, and PAP-III depurinate RNA of human immunodeficiency virus (HIV)-1. Biochem. Biophys. Res. Commun. 260, 453–458. 20. Bolognesi, A., Polito, L., Lubelli, C., Barbieri, L., Parente, A. & Stirpe, F. (2002) Ribosome-inactivating and adenine polynucleo- tide glycosylase activities in Mirabilis jalapa L. tissues. J. Biol. Chem. 277, 13709–13716. 21. Peumans, W.J., Hao, Q. & Van Damme, E.J.M. (2001) Ribo- some-inactivating proteins from plants: more than RNA N-gly- cosidases? FASEB J. 15, 1493–1506. 22. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 23. Becker, D., Kemper, E., Schell, J. & Masterson, R. (1992) New plant binary vectors with selectable markers located proximal to the left T-DNA border. Plant Mol. Biol. 20, 1195–1197. 24. Timmermans, M., Maliga, P., Vieira, J. & Messing, J. (1990) The pFF plasmids: cassettes utilising CaMV sequences for expression of foreign genes in plants. J. Biotechnol. 14, 333–334. 1514 F. Vandenbussche et al.(Eur. J. Biochem. 271) Ó FEBS 2004 25. Sanger, F., Nicklen, S. & Coulson, A.R. (1977) DNA sequencing with chain-terminating inhibitors. Proc. Natl Acad. Sci. USA 74, 5463–5467. 26. Shen, W.J. & Forde, B.G. (1989) Efficient transformation of Agrobacterium spp. by high voltage electroporation. Nucleic Acids Res. 17, 8385. 27. Horsch, R.B., Fry, J.E., Hoffmann, N.L., Eichholtz, D., Rogers, S.G. & Fraley, R.T. (1985) A simple and general method for transferring genes into plants. Science 227, 1229–1231. 28. Goode, B.L. & Feinstein, S.C. (1992) ÔSpeedprepÕ purification of template for double-stranded DNA sequencing. Biotechniques 12, 374–375. 29. Eggermont, K., Goderis, I.J. & Broekaert, W.F. (1996) High- throughput RNA extraction from plant samples based on homogenisation by reciprocal shaking in the presence of a mixture of sand and glass beads. Plant Mol. Biol. Rep. 14, 273–279. 30. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. 31. Desmyter, S., Vandenbussche, F., Hao, Q., Proost, P., Peumans, W.J. & Van Damme, E.J.M. (2003) Type-1 ribosome-inactivating protein from iris bulbs: a useful agronomic tool to engineer virus resistance? Plant Mol. Biol. 51, 567–576. 32. Van Damme, E.J.M., Hao, Q., Charels, D., Barre, A., Rouge ´ ,P., Van Leuven, F. & Peumans, W.J. (2000) Characterization and molecular cloning of two different type 2 ribosome-inactivating proteins from the monocotyledonous plant Polygonatum multi- florum. Eur. J. Biochem. 267, 2746–2759. 33. Brigotti, M., Alfieri, R., Sestili, P., Bonelli, M., Petronini, P.G., Guidarelli,A.,Barbieri,L.,Stirpe,F.&Sperti,S.(2002)Damage to nuclear DNA induced by Shiga toxin 1 and ricin in human endothelial cells. FASEB J. 16, 365–372. 34. Ward, E.R., Uknes, S.J., Williams, S.C., Dincher, S.S., Wieder- hold, D.L., Alexander, D.C., Ahl-Goy, P., Me ´ traux, J P. & Ryals, J.A. (1991) Coordinate gene activity in response to agents that induce systemic acquired resistance. Plant Cell 3, 1085–1094. 35. Battelli, M.G., Barbieri, L., Bolognesi, A., Buonamici, L., Valbonesi, P., Polito, L., Van Damme, E.J.M., Peumans, W.J. & Stirpe, F. (1997) Ribosome-inactivating lectins with poly- nucleotide:adenosine glycosidase activity. FEBS Lett. 408, 355–359. 36. Barbieri, L., Ciani, M., Girbes, T., Liu, W Y., Van Damme, E.J.M., Peumans, W.J. & Stirpe, F. (2004) Enzymatic activity of toxic and non-toxic type 2 ribosome-inactivating proteins. FEBS Lett. in press. 37. Chen, Y., Vandenbussche, F., Rouge ´ ,P.,Proost,P.,Peumans, W.J. & Van Damme, E.J.M. (2002) A complex fruit-specific type-2 ribosome-inactivating protein from elderberry (Sambucus nigra)is correctly processed and assembled in transgenic tobacco plants. Eur. J. Biochem. 269, 2897–2906. 38. Chen, Y., Peumans, W.J. & Van Damme, E.J.M. (2002) The Sambucus nigra type-2 ribosome-inactivating protein SNA-I¢ exhibits in planta antiviral activity in transgenic tobacco. FEBS Lett. 516, 27–30. 39. Hao,Q.,VanDamme,E.J.M.,Hause,B.,Barre,A.,Chen,Y., Rouge ´ , P. & Peumans, W.J. (2001) Iris bulbs express type 1 and type 2 ribosome-inactivating proteins with unusual properties. Plant Physiol. 125, 866–876. 40. Desmyter, S. (2002) Study of the antiviral activity of IRIP, a type- 1 ribosome-inactivating protein from Iris hollandica.PhDThesis, Katholieke Universiteit Leuven, Belgium. 41. Hudak, K.A., Bauman, J.D. & Tumer, N.E. (2002) Pokeweed antiviral protein binds to the cap structure of eukaryotic mRNA and depurinates the mRNA downstream of the cap. RNA 8, 1148–1159. 42. Zoubenko, O., Uckun, F., Hur, Y., Chet, I. & Tumer, N. (1997) Plant resistance to fungal infection induced by nontoxic pokeweed antiviral protein mutants. Nat. Biotechnol. 15, 992–996. 43. Zoubenko, O., Hudak, K. & Tumer, N.E. (2000) A non-toxic pokeweed antiviral protein mutant inhibits pathogen infection via a novel salicylic acid-independent pathway. Plant Mol. Biol. 44, 219–229. Ó FEBS 2004 Antiviral activity of ribosome-inactivating proteins (Eur. J. Biochem. 271) 1515 . Analysis of the in planta antiviral activity of elderberry ribosome-inactivating proteins Frank Vandenbussche 1 , Stijn. specificity was exploited to further unravel the mechanism underlying the in planta antiviral activity of ribosome-inactivating proteins. Trans- genic tobacco