|Home | About | Journals | Submit | Contact Us | Français|
Ribosome inactivating proteins (RIPs) like ricin, pokeweed antiviral protein (PAP) and Shiga-like toxins 1 and 2 (Stx1 and Stx2) share the same substrate, the α-sarcin/ricin loop, but differ in their specificities towards prokaryotic and eukaryotic ribosomes. Ricin depurinates the eukaryotic ribosomes more efficiently than the prokaryotic ribosomes, while PAP can depurinate both types of ribosomes. Accumulating evidence suggests that different docking sites on the ribosome might be used by different RIPs, providing a basis for understanding the mechanism underlying their kingdom specificity. Our previous results demonstrated that PAP binds to the ribosomal protein L3 to depurinate the α-sarcin/ricin loop and binding of PAP to L3 was critical for its cytotoxicity. Here, we used surface plasmon resonance to demonstrate that ricin toxin A chain (RTA) binds to the P1 and P2 proteins of the ribosomal stalk in Saccharomyces cerevisiae. Ribosomes from the P protein mutants were depurinated less than the wild-type ribosomes when treated with RTA in vitro. Ribosome depurination was reduced when RTA was expressed in the ΔP1 and ΔP2 mutants in vivo and these mutants were more resistant to the cytotoxicity of RTA than the wild-type cells. We further show that while RTA, Stx1 and Stx2 have similar requirements for ribosome depurination, PAP has different requirements, providing evidence that the interaction of RIPs with different ribosomal proteins is responsible for their ribosome specificity.
The plant toxin ricin produced by the castor bean (Ricinus communis) is a member of the ribosome inactivating proteins (RIPs) that have potent cytotoxicity to animal cells (Stirpe and Battelli, 2006). Ricin is an AB-toxin consisting of a catalytically active A chain associated with a cell binding B chain. The A chain of ricin (RTA) is an N-glycosidase that depurinates a universally conserved adenine (A4324 in the rat 28S rRNA and A2660 in the E. coli 23S rRNA) in the highly conserved α-sarcin/ricin loop (SRL) of the large rRNA (Endo and Tsurugi, 1987; 1988). The depurination of the SRL has been reported to interfere with the elongation factor 1-dependent binding of aminoacyl-tRNA to the ribosome, as well as the GTP-dependent binding of elongation factor 2 and inhibit protein synthesis at the translocation step (Montanaro et al., 1975; Osborn and Hartley, 1990). The B chain of ricin (RTB) is a galactose-specific lectin that is responsible for binding ricin to glycoproteins or glycolipids on the surface of cells to promote cellular attachment and subsequent endocytosis of ricin (Endo and Tsurugi, 1987; 1988; Endo et al., 1987). Reduction of the disulphide bond connecting the A and B subunits is required for the enzymatic activity of ricin (Sandvig and van Deurs, 1996). The crystal structures of both ricin and RTA have been solved (Katzin et al., 1991; Rutenber et al., 1991) and amino acids involved in catalysis have been identified (Ready et al., 1991; Kim and Robertus, 1992). We have previously shown that a single-amino acid mutation at Glu177 at the active site, E177K, resulted in loss of depurination activity and cytotoxicity of RTA in Saccharomyces cerevisiae (Li et al., 2007; Parikh et al., 2008).
The bacterial toxins, Shiga and Shiga-like toxins 1 and 2 (Stx1 and Stx2), as well as a type I RIP from plants, pokeweed antiviral protein (PAP), also depurinate the rRNA at the same position as ricin (Endo et al., 1988a). Although the SRL is the universal substrate for all RIPs, ribosomal proteins play an important role in making therRNA highly susceptible to attack by RIPs. The rat ribosome is depurinated by RTA with a Kcat nearly 105-fold greater than that measured using the naked 28S rRNA (Endo and Tsurugi, 1988). RTA depurinates the naked 23S rRNA from E. coli at the corresponding position, but not the ribosomes from E. coli (Endo et al., 1988b). Furthermore, the efficiency of ribosome inactivation among fungus, protozoan, plant, insect and prokaryotic ribosomes varies between the RIPs (Stirpe and Battelli, 2006). For example, ricin is 23 000 times more effective on rat liver ribosomes than on plant ribosomes, while bacterial ribosomes are resistant to ricin (Harley and Beevers, 1992). In contrast, PAP is equally active on ribosomes from all five kingdoms. These observations suggest that the differential sensitivity of ribosomes to RIPs that have identical rRNA substrate specificities may be due to differences in their ribosomal protein interactions (Hudak et al., 1999; Chan et al., 2007; McCluskey et al., 2008).
The RTA has been chemically cross-linked to ribosomal proteins P0 and L9 in human lung carcinoma cells (Vater et al., 1995). Trichosanthin, a type I RIP, has been shown to interact with the C-terminal tails of ribosomal stalk P proteins, P0, P1 and P2, as well as the mitotic checkpoint protein MAD2B in vitro (Chan et al., 2001; 2007). More recently, yeast two-hybrid and in vitro pull-down studies indicated that Stx1 interacts with the P proteins of the ribosomal stalk from human cells (McCluskey et al., 2008). We have shown that binding of PAP to ribosomal protein L3 was critical for depurination of the SRL and the reduction in viability (Hudak et al., 1999). We further demonstrated that expression of a truncated form of the ribosomal protein L3 (L3Δ), conferred trans-dominant resistance to PAP and produced toxin-resistant transgenic plants (Di and Tumer, 2005). Subsequent studies showed that the C-terminal end of P proteins was not needed for depurination of the Trypanosoma cruzi ribosomes by PAP (Ayub et al., 2008). Gelonin, another single-chain RIP was not able to inhibit the cross-linking of RTA to P0 and L9 in mammalian cells, indicating that it may not be binding to the same proteins as ricin (Vater et al., 1995). These studies indicate that interaction with the P proteins may not be a general feature of all RIPs; different RIPs may use different docking sites on the ribosome to access the SRL.
The ribosomal stalk is a lateral protuberance of the large ribosomal subunit that recruits translation factors to the ribosome (Gonzalo and Reboud, 2003; Diaconu et al., 2005) and is involved in GTPase activation by EF-Tu and EF-G (Mohr et al., 2002). The prokaryotic stalk is composed of L7/L12 protein dimers anchored to the ribosome through the L10 protein, which together with the L11 protein form the stalk base (Diaconu et al., 2005). The eukaryotic stalk structure is composed of phosphorylated P proteins, which have little sequence similarity to prokaryotic counterparts, L7/L12, but have similar organization and function (Gonzalo and Reboud, 2003). In S. cerevisiae, the P proteins consist of the 11 kDa proteins P1 and P2, representing eukaryotic orthologues of the prokaryotic L7/L12 protein, and P0, which is equivalent to the prokaryotic L10. The P proteins form a pentameric structure P0-(P1-P2)2 where the P1 and P2 proteins are in the form of heterodimers (P1α/P2β-P1β/P2α) that attach to P0 (Gonzalo et al., 2001; Guarinos et al., 2001; 2003; Tchorzewski et al., 2003a). The N-termini of the P1 proteins directly interact with P0, which docks on the ribosome and interacts with the rRNA through its N-terminal domain, forming the GTPase-associated centre (Gonzalo and Reboud, 2003). The C-termini of P proteins are exposed (Jose et al., 1995; Nusspaumer et al., 2000; Qiu et al., 2006). The P0 protein is essential for protein synthesis and cell viability in yeast (Santos and Ballesta, 1994). In contrast, yeast strains with deletions in either two (ΔP1 and ΔP2) or four (ΔP1ΔP2) genes encoding the acidic ribosomal phosphoproteins (P proteins) are viable, but grow more slowly than the wild-type strain (Remacha et al., 1992; 1995). The composition of the ribosomes is not affected in these mutants except for the missing stalk proteins, as shown by the recovery of ribosome function after the addition of the missing stalk proteins (Remacha et al., 1992; 1995).
Although several RIPs were shown to interact with the P proteins in vitro, the in vivo relevance of these interactions to the enzymatic activity and the cytotoxicity of RIPs has not been demonstrated. Furthermore, the role of the ribosome interactions in the kingdom specificity of the different RIPs is not well understood. Using the yeast P protein deletion mutants we present here the first in vivo evidence that the ribosomal stalk forms the docking site for RTA on the ribosome, allowing RTA to depurinate the SRL and reduce the viability of yeast cells. We further show that the interaction with the P proteins is not a general feature of all RIPs.
The wild-type yeast strain (W303) and three different isogenic strains with deletions in the P protein genes were used in this study. In D45 (ΔP2), the genes RPP2A and RPP2B, encoding P2α and P2β, respectively, were deleted, whereas in D67 (ΔP1) the genes RPP1A and RPP1B, encoding P1α and P1β, respectively, were deleted (Remacha et al., 1992). In the quadruple mutant, D4567 (ΔP1ΔP2), all four P protein genes were deleted (Remacha et al., 1995). As shown in Fig. 1A, immunoblotanalysis indicated that the ΔP1 mutant did not contain any P1 or P2 proteins on the ribosome but had free P2α and P2β proteins in the cytosol. In contrast, the ΔP2 mutant did not contain any P2 or P1 proteins on the ribosome or in the cytosol. Similarly, the ΔP1ΔP2 mutant did not contain any P proteins on the ribosome or in the cytosol. The ribosomal protein P0 was present on the ribosomes of these mutants at the same level as in the wild-type cells, indicating that it was not affected by the absence of the P proteins. The P0 was present only on the ribosomes and was not detected in the cytosol. These results are consistent with the previously reported studies, which indicated that the P1 proteins are degraded in the absence of P2 proteins, while P2 proteins are much more stable and consequently accumulate in the cytosol in the absence of P1 proteins (Fig. 1A and B) (Nusspaumer et al., 2000). Altogether, these results demonstrated that there were no differences in the stalk composition of the three different P protein deletion mutants.
Surface plasmon resonance was used to examine the interaction between RTA and the ribosomes from the P protein mutants. The N-terminal His-tagged RTA was immobilized on the target surface as the ligand and the N-terminal His-tagged enhanced green fluorescent protein (EGFP), which has a similar molecular weight as RTA, and was immobilized as the reference. Ribosomes isolated from the wild type, ΔP1, ΔP2 and ΔP1ΔP2 mutants were passed over the reference surface and the target surface as the analyte. The signal from the reference surface was subtracted from the signal from the target surface containing the His-tagged RTA to correct for non-specific binding. Under these conditions, we observed binding of the wild-type ribosomes to RTA, but not to EGFP (Fig. 2). The wild-type ribosomes bound to RTA with rapid association and dissociation. This interaction was abolished when ribosomes isolated from the P protein mutants were used. Ribosomes from the three different P protein mutants, ΔP1, ΔP2 and ΔP1ΔP2 showed similar binding profiles with markedly reduced association and almost no dissociation. Comparison of the sensorgrams at 5 nM ribosome concentration, shown in Fig. 2, with sensorgrams at higher ribosome concentra-tions (10 and 20 nM), shown in Fig. S1A and B, indicated similar interaction profiles except that the binding signals observed were higher at the higher ribosome concentrations. The slight differences observed in the association and dissociation profiles of the mutants were not due to the particular deletion, but possibly due to variation in the ribosome preparations or differences in the yeast metabolic conditions (Saenz-Robles et al., 1990). These results demonstrated that an intact ribosomal stalk is critical for the interaction of RTA with ribosomes.
As described above, binding of RTA to ribosomes that did not have an intact stalk structure was markedly reduced compared with the wild-type ribosomes. To determine if binding to the ribosomal stalk is critical for RTA to depurinate the 25S rRNA, we incubated purified ribosomes from the P protein mutants and the isogenic wild-type strain with RTA in vitro and examined depurination by dual-primer extension analysis at the different time points (Parikh et al., 2002). As shown in Fig. 3A, depurination of the rRNA from the wild-type strain was detected at 15 min after incubation with RTA and the extent of depurination increased with time. In contrast, depurination of the rRNA was reduced in the ΔP1, ΔP2 and ΔP1ΔP2 mutants. The ratio of the depurination band relative to the total amount of 25S rRNA (Depurination/25S) was quantified in Fig. 3B. The wild-type strain showed a time-dependent increase in the level of depurination by RTA. However, the level of depurination was reduced in the P protein mutants at each timepoint and was only ~67% of the wild type even after 1 h incubation with RTA (Fig. 3B). The depurination results agreed with the surface plasmon resonance binding analysis and showed that RTA could not depurinate the ribosomes without an intact stalk structure as efficiently as the wild-type ribosomes.
Previous results showed that ribosomes isolated from the P protein mutants did not bind RTA and showed reduced depurination by RTA in vitro. To determine if ribosome depurination was reduced in vivo, we examined the expression of RTA and the extent of rRNA depurination in the ΔP1, ΔP2 mutants and in the isogenic wild-type strain. The URA3 plasmids harbouring the wildtype preRTA (the precursor form of RTA) or the preRTA with a point mutation at the active site, E177K (Li et al., 2007), tagged with the V5 epitope (Southern et al., 1991) were transformed into the ΔP1 mutant. The preRTA and E177K on a LEU2 plasmid were transformed into the ΔP2 mutant. The isogenic wild-type yeast cells (W303) were transformed with the same plasmids as the control. Monoclonal antibody against the V5 epitope was used to detect RTA expression in the ΔP1 mutant (Fig. 4A), and polyclonal antibody against RTA was used to detect RTA expression in the ΔP2 mutant (Fig. 4B). Antibodies against the integral endoplasmic reticulum membraneprotein, dolichol-phosphate mannose synthase (Dpm1) were used as the loading control. Expression of the wildtype RTA was lower in the ΔP1 and ΔP2 mutants than the non-toxic active site mutant, E177K (Fig. 4A and B). The signal sequence targets the preRTA to the endoplasmic reticulum in yeast where it undergoes glycosylation as in the castor bean (Parikh et al., 2008). Therefore, multiple bands were observed due to the different glycosylated forms of RTA (Fig. 4A and B). The extent of rRNA depurination was determined by dual-primer extension analysis using total RNA isolated from these cells. Ribosomes were depurinated in the wild-type yeast cells expressing RTA but not in cells expressing the E177K (Fig. 4C and D). Ribosome depurination was reduced in the ΔP1 and the ΔP2 mutants expressing preRTA. As shown in Fig. 4E and F, the relative levels of depurination (Depurination/25S) in the ΔP1, ΔP2 mutants were reduced by 60% and 70%, respectively, relative to the wild-type yeast cells expressing RTA. These results demonstrated that the deletion of the ribosomal stalk components did not affect the RTA expression, yet it affected the level of ribosome depurination by RTA in vivo.
Ribosome depurination was reduced in vivo in the yeast mutants that did not contain an intact ribosomal stalk. Todetermine if the reduced ribosome depurination affected the sensitivity of yeast cells to RTA, we examined the viability of the wild-type cells and the ΔP1 and the ΔP2 mutants expressing RTA by plating cells on glucose plates after galactose induction for different times in liquid media. Yeast cells harbouring the empty vector or the non-toxic E177K were used as controls. As shown in Fig. 5, after 10 h of induction, RTA expression reduced the viability of the wild-type cells by almost three magnitudes compared with cells expressing E177K or harbouring the vector. In contrast, the ΔP1 or the ΔP2 mutant expressing RTA was more viable than the wildtype cells. Viability of the ΔP2 mutant expressing RTA, E177K or harbouring the vector was similar, while the ΔP1 mutant expressing RTA displayed a slight reduction in viability compared with the ΔP1 mutant expressing E177K or harbouring the vector. The increase in viability observed in the ΔP2 mutant correlated with the ribosome depurination results (Fig. 4), which showed that ribosomes were depurinated less in the ΔP2 mutant than in the ΔP1 mutant in vivo. These results demonstrated that the ΔP1 and the ΔP2 mutants that are disrupted in the ribosomal stalk were more resistant to RTA than the isogenic wild-type strain.
Previous results suggested that an intact ribosomal stalk is critical for RTA to depurinate the SRL and to reduce the viability of yeast cells. To determine if an intact ribosomal stalk is also required for ribosome depurination by other RIPs, we examined the depurination of ribosomes from the P protein mutants by a type I RIP, PAP and by other type II RIPs, Stx1 and Stx2. As Stx1 and Stx2 are AB-toxins, in which the A and the B subunits are linked by a disulphide bond, they were treated with dithiothreitol (DTT) to reduce the disulphide bond and to release the enzymatically active A chain before the ribosome treatment. PAP- and DTT-treated Stx1 and Stx2 were incubated with the purified ribosomes from the wild-type, ΔP1, ΔP2 and ΔP1ΔP2 mutants and the depurination of rRNA was examined by the dual-primer extension analysis (Parikh et al., 2002). As shown in Fig. 6A and B, PAP depurinated the ribosomes from the ΔP1, ΔP2 and ΔP1ΔP2 mutants and the isogenic wild-type cells at similar levels. However, ribosome depurination by Stx1A and Stx2A was reduced in the P protein mutants compared with the wild-type cells (Fig. 6A and B). The resultswith Stx1A were in agreement with the in vitro binding studies (McCluskey et al., 2008). These results indicated that both Stx1 and Stx2 require an intact ribosomal stalk to depurinate the SRL in vitro, while PAP utilizes a different mechanism.
Unlike ribosomes from the wild-type strain, the ribosomes isolated from the P protein mutants (ΔP1, ΔP2 and ΔP1ΔP2) contained only the P0 protein and not the acidic P1 or P2 proteins (Fig. 1), indicating that they had the expected stalk composition (Remacha et al., 1992; 1995). Previous studies demonstrated that ribosome function was fully recovered in the P protein mutants by the addition of the missing stalk proteins, indicating that the composition of the ribosome was not affected (Remacha et al., 1992; 1995). Moreover, the defective ribosomes did not show any significant alteration in the sucrose gradient mobility, providing evidence that they have not undergone important changes in the overall conformation (Fig. S2). Although there were differences in the relative proportion of the ribosome and the polysome peaks and half-mers were present in the three mutants, especially in the ΔP2 and ΔP1ΔP2, the overall pattern was similar to the parental strain and there were no detectable changes in the mobility of the different particles (Fig. S2). We observed a dynamic interaction between RTA and the wild-type ribosomes using surface plasmon resonance, characterized by rapid association and dissociation (Fig. 2). In contrast, ribosomes isolated from the P protein mutants showed markedly reduced binding with a very slow association and almost no dissociation, indicating that they bind to RTA in a different manner (Fig. 2). A similar pattern of association and dissociation was obtained with the three different deletion mutants, ΔP1, ΔP2 and ΔP1ΔP2, consistent with the immunoblot analysis, which indicated that the mutants had a similar stalk composition (Fig. 1). The kinetic analysis of the interaction between RTA and the wild-type ribosomes did not fit a simple 1:1 binding model, but was characterized by a different model, which will be described elsewhere (X.-P. Li et al., in preparation). These results indicated that the ribosomal stalk is the site where RTA interacts with the ribosome to localize its substrate.
The eukaryotic P0 protein is larger than the prokaryotic L10 due to the presence of a C-terminal extension, which shows sequence similarity to P1 and P2 (Diaconu et al., 2005). The last 13 amino acids at the C-terminus of the yeast and the human P proteins are almost identical (Fig. S3). In a previous study, the interaction of Stx1A with human P0, P1 and P2 proteins was examined by either pull-down experiments or with the yeast two-hybrid system and both Stx1A and RTA were shown to interact with the last 11-17 amino acids of the conserved C-termini of the P proteins in vitro (McCluskey et al., 2008). However in our study, RTA did not interact with the ribosomes that had only the P0 protein. The depurination of these ribosomes by RTA was reduced in vitro. Furthermore, ribosome depurination was reduced in vivo in yeastcells that did not have an intact ribosomal stalk, and these mutants were more resistant to RTA than the wild-type cells. These results showed that the presence of only P0 on the ribosome did not compensate for the absence of the P1 and P2 proteins. Analysis of the yeast stalk structure indicated that P0 could not fold properly in the absence of the P1α/P2β dimer (Krokowski et al., 2005). Therefore, the C-terminal tail of P0 may adopt a different conformation when assembled into the large subunit in the absence of P1 and P2, such that it is not accessible to RTA. We conclude from this study that the ribosomebound P0 is not sufficient to restore the interaction between RTA and the ribosome and the depurination of the SRL in the absence of P1 and P2.
We showed that in the absence of an intact ribosomal stalk, the cytotoxicity of RTA was reduced. Interestingly, the level of depurination in vivo was slightly higher in the ΔP1 mutant that contained free P2 proteins in the cytosol, than in the ΔP2 mutant, which did not contain free P1 or P2 (Fig. 4E and F). Furthermore, the ΔP2 mutant was more viable than the ΔP1 mutant in the presence of RTA (Fig. 5). These results indicated that the free P2 proteins in the cytosol did not protect against RTA, instead they appeared to increase the sensitivity to RTA. There are two possible explanations for this phenomenon. First, the free P2 protein may interact with RTA through its C-terminus in the cytoplasm. This P2-RTA complex may recognize the stalk, which only contains P0, allowing RTA to localize the SRL. Second, there might be free P2 in the vicinity of ribosomal stalk, even though it could not bind to P0 in the absence of P1 (Tchorzewski et al., 2003b; Krokowski et al., 2006; Garcia-Marcos et al., 2007). Unlike the other ribosomal proteins, the stalk components can exist in a cytoplasmic pool and can exchange with the ribosome bound polypeptides (Tsurugi and Ogata, 1985). Furthermore, P2 is able to form homodimers in solution (Zurdo et al., 1997). It is therefore possible that in the absence of P1 proteins, a small amount of P2 homodimers are formed that can weakly bind to the ribosomes. Hence, the free cytosolic pool of P2 or the P2 homodimers may help recruit RTA to the ribosome, leading to a higher level of depurination and cytotoxicity in the ΔP1 mutant than in the ΔP2 mutant (Figs (Figs44 and and5).5). Expression of a peptide corresponding to the last 17 amino acids of P1 and P2 proteins protected mammalian ribosomes from translation inhibition by Stx1A in vitro (McCluskey et al., 2008), indicating that in contrast to the intact P2 protein, the peptide corresponding to the C-terminus of the P proteins may bind to Stx1A, but may not be able to target it to the ribosome.
The sensitivity of translation machinery towards different RIPs is distinct among prokaryotes and eukaryotes. For example, RTA can depurinate eukaryotic but not prokaryotic ribosomes, whereas PAP is active against both types of ribosomes (Endo and Tsurugi, 1988). PAP has been shown to access the SRL by interacting with the ribosomal protein L3, an essential protein at the peptidyltransferase centre, highly conserved between the eukaryotes and prokaryotes (Hudak et al., 1999). Although the three dimensional structure of PAP and TCS are similar, PAP did not require the C-terminal end of the P proteins to depurinate the trypanosome ribosomes (Ayub et al., 2008). Blocking the C-termini of P proteins with a monoclonal antibody could prevent the inactivation of the ribosomes by TCS, but not by PAP (Ayub et al., 2008). We show here that PAP has similar activity towards the wild-type ribosomes and ribosomes from the P protein mutants in spite of the lack of an intact stalk (Fig. 6). These results provide evidence that PAP does not require an intact stalk structure to depurinate the SRL in vitro, consistent with its ability to bind to L3 to depurinate the eukaryotic ribosomes in vivo (Hudak et al., 1999). Yeast two-hybrid studies confirmed that PAP does not bind to the P proteins (R. Francisco, M. Remacha and J.P.G. Ballesta, unpublished).
L3 is located below the ribosomal stalk and the rRNA substrate for RTA, the SRL, has been localized in close proximity to the ribosomal protein L3 within the 50S subunit of Haloarcula marismortui ribosomes (Ban et al., 2000). Fig. S4 shows a comparison of the structures of the H. marismortui (Ban et al., 2000) and Escherichia coli 50S subunits (Schuwirth et al., 2005) around the SRL with several proximal ribosomal proteins highlighted. The L10e protein mapped in the crystal structure of H. marismortui (Ban et al., 2000) corresponds to bacterial L16 (Chandramouli et al., 2008). As shown in Fig. S4, the structure and the position of L3 are highly conserved between archaea and E. coli. Furthermore, the primary sequence of L3 protein is highly conserved in yeast, human and E. coli (Fig. S5A), especially in the region of L3 that is critical for binding of PAP to the ribosome (Hudak et al., 1999). Therefore, the similarities in the primary sequence and the 3-D structure of the L3 may be responsible for the sensitivity of both the eukaryotic and the prokaryotic ribosomes to PAP.
In contrast to L3, the components of the ribosomal stalk differ more in the primary sequence and structure between the prokaryotes and eukaryotes, even though they serve the same function (Gonzalo and Reboud, 2003). Sequence comparison of the P1 and P2 proteins among human, yeast and E. coli indicated that the P1 andP2 proteins are more highly conserved among the eukaryotes than between the eukaryotes and E. coli (Fig. S5B). Moreover, the conserved C-terminal region of P proteins that can interact with TCS (Chan et al., 2007), Stx1 and RTA in vitro (McCluskey et al., 2008) is missing in the E. coli L7/L12 (Diaconu et al., 2005). The stalk region containing L7/L12 and L10 was not fully modelled in the archaeal or the bacterial 50S structure (Fig. S4) due to its high mobility (Ban et al., 2000; Schuwirth et al., 2005). Recent analysis of the structural relationships among the ribosomal stalk proteins indicated that the archaeal and the bacterial stalk proteins are not structurally related (Grela et al., 2008). These observations suggest that the differences in the primary sequence and the architecture of the ribosomal stalk between the prokaryotic and eukaryotic ribosomes may be the source of the difference in the sensitivity of the eukaryotic, but not the prokaryotic ribosomes to ricin. In summary, we show here that the catalytic A subunit of ricin interacts with the P1 and the P2 proteins of the ribosomal stalk to localize its substrate in vivo and present evidence that the interaction of RIPs with different ribosomal proteins is the likely source of their kingdom specificity.
The S. cerevisiae strains that contained deletions in the P proteins were haploid strains, which carried two or four genetic markers depending on the different disruptions: D67 (ΔP1) (RPP1A::LEU2, RPP1B::TRP1), D45 (ΔP2) (RPP2A::URA3, RPP2B::HIS3) (Remacha et al., 1992) and D4567 (ΔP1ΔP2) (RPP1A::LEU2, RPP1B::TRP1, RPP2A:: URA3, RPP2B::HIS3) (Remacha et al., 1995). The wild-type strain was W303 (MATa ade2-1 trp1-1 ura3-1 leu2-3, 112 his-3-11,15 can1-100). Yeast strains were grown in either YPD medium or minimal SD medium containing 2% glucose. The preRTA used in this study contained the signal sequence (Li et al., 2007). An active site mutant of preRTA (E177K), which did not depurinate the SRL and was not toxic to yeast cells, was used as the negative control (Li et al., 2007). The preRTA and E177K under the control of the galactose-inducible GAL1 promoter in the vectors carrying the URA3 or LEU2 selection markers were transformed into the ΔP1 and the ΔP2 mutants based on the markers available in these strains. The preRTA (NT849) and E177K (NT859) in the YEp351-based vector, NT198, containing the LEU2 marker (Li et al., 2007), were transformed into the ΔP2 mutant. The 5′ gctctagagaattcaagatgaaaccggg and 3′ aaactgtgacgatggtggaggtgc primers were used to amplify the preRTA and E177K genes from NT849 and NT859 respectively. The amplified fragments were then cloned into pYES2.1/V5-His-TOPO vector (Invitrogen, Carlsbad, CA) with the URA3 marker and a V5 epitope and transformed into the ΔP1 mutant. The isogenic wild-type strain (W303) was transformed with each vector as the control. We were not able to examine the sensitivity of the quadruple mutant to RTA because this strain did not contain any available markers (Remacha et al., 1995).
Monomeric ribosomes were isolated from the yeast cells according to Algire et al. (2002) with some modifications. Yeast cells were grown to an OD595 = 0.8-1.5 at 30°C. Cell pellets were washed with ice-cold 0.9% KCl once and then with buffer A [20 mM HEPES-KOH, pH 7.6, 5 mM Mg(OAc)2, 50 mM KCl, 10% Glycerol]. The resulting pellets were re-suspended in buffer A containing 1 mg ml-1 Heparin, 1 mM PMSF, 1 mM DTT and with protease inhibitor cocktail (Sigma, 1:1000). The cell mixtures were ground in liquid nitrogen, transferred to centrifuge tube and centrifuged at 32 000 g for 20 min. The supernatant was transferred to 4 ml polycarbonate tubes containing 1 ml of buffer C cushion [50 mM HEPES-KOH, pH 7.6, 5 mM Mg(OAc)2,50 mMNH4Cl, 25% Glycerol] and centrifuged at 200 000 g for 2 h. The resulting pellets were re-suspended in buffer B [20 mM HEPES-KOH, pH 7.6, 20 mM Mg(OAc)2, 0.5 M KCl, 10% Glycerol] containing 1 mM GTP and 1 mM Puromycin and incubated at 30°C for 30 min. The mixtures were then centrifuged at 10 000 g for 10 min. The supernatant was applied onto 1 ml of buffer B (25% glycerol) cushion and spun down again at 200 000 g for 2 h as described above. Ribosome pellets were resuspended in buffer C. The concentration of ribosomes were calculated by using one OD260 = 20 pmol ml-1 of ribosomes. All centrifuge steps were performed at 4°C.
The Biacore 3000 (GE Healthcare, Piscataway, NJ, USA) was used to analyse the interaction between the ribosomes and the RTA. The N-terminal His-tagged RTA (Beiresources, Manassas, VA) was used as the ligand and the salt-washed monomeric ribosomes isolated from the P protein mutants or the isogenic wild-type strain were used as the analyte. The standard NTA chip running buffer (0.01 M HEPES, 0.15 M NaCl, 50 μM EDTA, 0.005% Surfactant P20, pH 7.4) containing 5 mM Mg(OAc)2 and 50 mM KCl was used. The sensor chips were first washed with the regeneration solution (0.01 M HEPES, 0.15 M NaCl, 0.35 M EDTA, 0.005% surfactant P20, pH 8.3) for 1 min followed by the running buffer. The running buffer containing 500 μM NiCl2 was injected for another 1 min followed by the running buffer to activate the NTA sensor chip. The His-tagged RTA prepared in the running buffer at 50 nM was injected at a flow rate of 20 μl min-1 to generate a resonance signal of 1500 RU. As a control, 30 nM of the N-terminal His-tagged EGFP prepared in the running buffer was immobilized as the reference at the same RU and the flow rate. Ribosomes at 5, 10 and 20 nM were passed through both the target and the reference surfaces at the flow rate of 30 μl min-1 for 3 min to monitor the association. The dissociation was monitored at the same flow rate in the running buffer for another 3 min. The signal from the reference surface was subtracted from the signal obtained from the surface containing the His-tagged RTA to correct for non-specific binding. After each ribosome binding,both sensor and reference surfaces were regenerated by injecting the regeneration solution for 1 min followed by injecting the 2 M NaCl solution in running buffer for another minute and then washing with the running buffer for 1 min at a flow rate of 100 μl min-1.
The RNA depurination assay was conducted according to the previously described dual-primer extension analysis (Parikh et al., 2002). For in vivo depurination, RTA expression was induced for 10 h. Total RNA was isolated as previously described (Parikh et al., 2002) and 2 μg of the RNA was incubated with 106 cpm of γ-32P-labelled primer (APEX2) to detect depurination and 2 × 105 cpm of labelled 25S primer. The APEX2 primer hybridized downstream of the depurination site on the SRL and generated a 73 nt primer extension product. The second primer (25S primer) hybridized near the 5′ end of the 25S rRNA and generated a 100 nt primer extension product (Parikh et al., 2002). Superscript II reverse transcriptase (Invitrogen, Carlsbad, CA) was used for the primer extension and the products were separated on the 6% polyacrylamide gel containing 7 M urea and quantified with the Phosphoimager (Molecular Dynamics, Sunnyvale, CA). For in vitro depurination by RTA, 20 pmol of ribosomes was incubated with 10 ng of RTA in the RIP buffer (60 mM KCl, 10 mM Tris-HCl pH 7.4, 10 mM MgCl2) at 30°C for 0, 15, 30 and 60 min. For in vitro depurination by other RIPs, 20 pmol of ribosomes were incubated with 10 ng of PAP or 26 ng of Stx1 or Stx2 in RIP buffer at 30°C for 30 min. DTT (30 μM) was added in RIP buffer to reduce the disulphide bond and to release the A chain from Stx1 and Stx2. Total RNA was extracted and used for dual-primer extension analysis (Parikh et al., 2002).
Total lysates were separated into the membrane and cytosolic fractions as previously described (Baykal and Tumer, 2007). The membrane fraction was used to detect the RTA expression. Expression of RTA was induced for 10 h, cells were centrifuged, washed with 50 mM Tris-HCl pH 7.5, 10 mM Na-Azide and 10 mM NaF, and incubated at 30°C for 20 min. Cells were then centrifuged and re-suspended in 20 mM Tris-HCl, pH 7.5, 1.2 M sorbitol, 50 mM KOAc and 150 U of lyticase to break the cell wall. After incubation for 45 min, cells were spun down and the cell debris was removed and the supernatant was centrifuged at 10 000 g for 15 min at 4°C. The pellets were then re-suspended in 20 mM Tris-HCl, pH 7.5, 0.25 M sorbitol, 50 mM KOAc and used as the membrane fraction.
The cytosolic fraction was isolated according to (Baykal and Tumer, 2007) with slight modifications. After breaking the yeast cells with glass beads in the lysis buffer [(50 mM Tris-HCl, pH 7.9, 5 mM EDTA, 1%(v/v) Triton X-100, 1 mM PMSF and proteinase inhibitor cocktail (Sigma, 1:1000)], cells were centrifuged down to remove cell debris. The supernatant was centrifuged down with 100 000 g for 1 h to isolate the cytosolic fraction.
For immunoblot analysis, the membrane and the cytosolic fractions were separated on a 12% SDS-PAGE. Monoclonal antibodies specific for P1β,P2α and P2β were used to detect the P1 and P2 proteins (Vilella et al., 1991), while polyclonal antibody specific for P0 was used to detect the P0 protein (Santos and Ballesta, 1994). Monoclonal antibody to L3 (gift of Dr. J.R. Warner) was used to detect the ribosomal protein L3 expression. Antibodies against dolichol-phosphate mannose synthase (Dpm1p; Invitrogen, Carlsbad, CA) and 3-phosphoglycerate kinase (Pgk1p; Invitrogen, Carlsbad, CA) were used to assess the purity of the membrane and cytoplasmic fractions respectively. The RTA expression was detected with either a polyclonal antibody against RTA (Li et al., 2007) or monoclonal antibody against the V5 epitope (Invitrogen, Carlsbad, CA). The RTA standard was purchased from Vector Laboratories (Burlingame, CA).
Yeast cells expressing RTA or E177K or harbouring the vector were grown in SD medium containing 2% glucose to an approximately OD595 = 0.3. Cells were transferred to medium containing 2% galactose to induce the expression of RTA. Aliquots were taken at 0, 4, 6 and 10 h post induction adjusted to OD595 = 0.1 and serial dilutions were plated on medium containing 2% glucose as previously described (Parikh et al., 2005).
We thank Dr. Bijal Parikh for comparison of the ribosome structure around the SRL (Fig. S4) and for critical reading of the manuscript, Jesus Revuelta for the polysome profiles (Fig. S2), Dr. Cheleste Thorpe for Stx1 and Stx2, Drs. Arturas Meskauskas and Jonathan Dinman for the ribosome isolation protocol, Dr. Jonathan Warner for the L3 monoclonal antibody and Beiresources for the N-terminal His-tagged RTA. This work was supported by National Institutes of Health Grant AI072425 (to N.E.T.) and Grant BFU2006-00365 from Ministerio de Educacion y Ciencia (to J.P.G.B.).
Additional supporting information may be found in the online version of this article.
Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.