Spb4 is a component of pre-60S ribosomal complexes.
We have previously found Spb4 in high-molecular-mass complexes of a size overlapping 60S/80S peaks in cell extracts by sucrose gradient fractionation (9
). Due to the fact that Spb4 is required for 60S r-subunit biogenesis, these complexes are most likely pre-60S r-particles. Indeed, Spb4 has been identified as a component of several purified pre-60S r-particles by the TAP method (5
). To confirm its association with pre-60S r-particles, we tagged Spb4 with a C-terminal TAP cassette (see Materials and Methods). Growth of the strain expressing Spb4-TAP as a sole source of Spb4 was indistinguishable from that of the wild-type isogenic counterpart (A). Moreover, the Spb4-TAP-expressing strain led to polysome and pre-rRNA processing profiles virtually identical to those of the wild-type strain (B and C). Thus, the Spb4-TAP fusion protein is fully functional. We then performed a two-step TAP purification of TAP-tagged Spb4 (see Materials and Methods). The strain W303-1A, which expresses a nontagged Spb4 protein, was used as a negative control. Coprecipitated proteins were separated by SDS-PAGE and stained with Coomassie blue (). Mass spectrometry identification of the different polypeptides revealed many proteins, including about 50 ribosome biogenesis factors (; see Tables S1 to S3 in the supplemental material). The majority of these factors have been described as components of pre-60S r-particles and implicated in the production of 60S r-subunits. The remaining factors are components of 90S preribosomal particles (see Table S1). Some proteins not involved in ribosome biogenesis were also found associated with Spb4-TAP (see Tables S2 and S3). Some of these proteins, such as translation factors, chaperones, or histones, are found frequently in TAP complexes (see Table S2), and therefore the specificity of their interaction with Spb4 is uncertain. Likewise, TAP complexes can be contaminated by r-proteins (see Table S2); thus, it is also unclear whether those we identified are the result of a contamination or are truly components of the purified preribosomal complexes.
Fig. 1. The Spb4-TAP fusion protein is fully functional. (A) The W303-1A strain harboring the empty YCplac111 vector (wild type) and the YDK37-1B strain containing the YCplac111-SPB4 (SPB4) or pTAPC111-SPB4 (SPB4-TAP) plasmid were grown in liquid SD-Leu medium. (more ...)
Fig. 2. Protein composition of the Spb4-TAP and Spb4(R360A)-TAP complexes. Strain YDK37-1B/ pTAPC111-SPB4, which expresses a TAP-tagged Spb4 fusion protein as a sole source of Spb4, was grown in YPD at 30°C to an OD600 of around 0.8. Strain W3031-A/pAS24-NTAP/2xFLAG-DNSPB4, (more ...)
To determine which pre-rRNA species are associated with Spb4-TAP, we performed precipitation experiments using IgG-conjugated Sepharose with total cell extracts of the Spb4-TAP strain and the nontagged control strain (see Materials and Methods). Coprecipitated RNAs were analyzed by Northern blotting. Consistent with the previous proteomic results, two of the most efficiently coprecipitated pre-rRNAs with Spb4-TAP are the 27SA2 and 27SB pre-rRNAs, which are components of early pre-60S r-particles (A; see Fig. S2 in the supplemental material). A modest enrichment of 35S and 32S pre-rRNAs, which are the components of 90S preribosomal particles, was also found (A; see Fig. S2). In contrast, 7S pre-rRNAs, which are components of intermediate nuclear pre-60S r-particles, and mature 25S rRNAs, which define late pre-60S r-particles and mature 60S r-subunits, were not significantly coprecipitated over the background levels obtained using untagged strain extracts (A; see Fig. S2). On the 40S r-subunit maturation pathway, only background levels of 20S pre-rRNA and 18S rRNA were found (A).
Fig. 3. Wild-type Spb4 is associated mainly with early pre-60S ribosomal particles, but dominant-negative Spb4(R360A) also coenriches 90S preribosomal particles. (A) Immunoprecipitation experiments were carried out using IgG-Sepharose and whole-cell extracts (more ...)
To further study the stage in the 60S r-subunit maturation pathway at which Spb4 associates with preribosomal particles, we screened for the presence of HA-tagged Spb4 following purification of GFP-tagged proteins specific for 90S (Pwp2/Utp1 and Nop58/Nop5) (10
), early, intermediate, late, and cytoplasmic pre-60S r-particles (Ssf1, Nop7/Yph1, Rix1/Ipi2, Arx1, and Kre35/Lsg1) (14
), and mature 60S r-subunits (P0) (54
). Purified particles were analyzed by SDS-PAGE, and HA-Spb4 was detected by Western blotting using anti-HA antibodies. Antibodies against other selected preribosomal factors (Has1 and Mrt4) and 60S r-subunits (L1) were also used to define the pre-60S particles at the different stages of their maturation. As shown in A, HA-Spb4 was strongly enriched in the Nop7-GFP- and Rix1-GFP-containing particles, present to some extent in 90S (Pwp2-GFP and Nop58-GFP), and late pre-60S r-particles, and practically absent from cytoplasmic (Arx1-GFP and Lsg1-GFP) and mature (P0-GFP) 60S r-subunits. As previously described, Nop7 is associated with a larger number of different pre-60S particles, although enriched within early E2
and intermediate nucleolar particles (see Fig. S2 in the supplemental material) (24
). Rix1, which is part of a protein complex together with Ipi1, Ipi3, and Rea1 (17
), stably associates with intermediate nucleolar/nucleoplasmic pre-60S particles (see Fig. S2) (33
). Moreover, we found that HA-Spb4 is absent from pre-40S r-particles (Tsr1 and Nob1) (15
) (data not shown).
Fig. 4. Wild-type Spb4 associates with pre-60S ribosomal particles, and the dominant-negative Spb4(R360A) protein stalls in 90S preribosomal particles. (A) The indicated GFP-tagged bait proteins were affinity purified by using the GFP-Trap_A procedure (see Materials (more ...)
Altogether, these results indicate that Spb4 is predominantly associated with early nucle(ol)ar pre-60S r-particles containing 27SB pre-rRNAs. A fraction of Spb4 is also present in earlier 66S r-particles containing 27SA2 and in 90S preribosomal particles containing either 35S or 32S pre-rRNA. Thus, Spb4 could bind 90S preribosomal particles and remain associated with pre-60S r-particles following cleavages of the 35S pre-rRNA at sites A0 to A2. Spb4 apparently efficiently dissociates from pre-60S particles after 27SB pre-rRNA processing.
The dominant-negative Spb4(R360A)-TAP protein stalls in 90S preribosomal particles.
To further address the properties of binding of Spb4 to preribosomal particles, we converted to alanine the conserved second arginine residue within motif VI (356-HRCGR
TGR-363; the second arginine is underlined) of Spb4-TAP. It has been previously shown for different RNA helicases that mutation of this arginine drastically reduced the ATPase activities of the mutated enzymes in vitro
(for examples, see references 45
; reviewed in reference 7
). In vivo
, this mutation is lethal and results in a DN phenotype when overexpressed in a wild-type strain (for examples, see references 26
, and 40
; reviewed in reference 7
). Since RNA helicases seem to interact transiently with their substrates, it has been hypothesized that DN versions of these enzymes would sequester these substrates and would not be released from the large ribonucleoprotein complexes to which they bind (2
). This scenario has been suggested for DN mutants of Prp2 (12
), Prp16 (58
), Prp22 (58
), and Prp43 (34
To test whether the Spb4(R360A)-TAP protein could confer a DN phenotype, we placed its corresponding allele under the control of the inducible GAL
promoter and expressed it in the wild-type strain. This allele could not restore the viability of an spb4
-null strain (data not shown). As shown in A, the overexpression of Spb4(R360A)-TAP in galactose-containing medium led to a strong inhibition of growth of the wild-type strain, while that of Spb4-TAP caused no change in the growth phenotype. Western blot analysis confirmed that both Spb4(R360A)-TAP and Spb4-TAP were comparably expressed in galactose-containing medium (B). Next, we tested the ability of the mutant Spb4(R360A)-TAP to affect the ribosome maturation pathway. The galactose-inducible TAP-tagged wild-type and DN versions of Spb4 were expressed in the wild-type strain, and polysome profile analyses were performed. The profile of the wild-type strain harboring the GAL::SPB4
construct was completely normal when grown in SD-Leu (C). However, despite Spb4 having a role in 60S r-subunit biogenesis (9
), induction of GAL::SPB4
for 12 h in SGal-Leu led to a mild 40S r-subunit deficit as revealed by an obvious increase in free 60S versus 40S r-subunits and a decrease in polysome levels (C). An induction for a longer time period of 24 h resulted in a further enhancement of this deficit (data not shown).
Fig. 5. Expression of the Spb4(R360A)-TAP protein leads to a dominant-negative growth phenotype and a deficit in 40S ribosomal subunits. (A) The W303-1A strain was transformed with plasmid YCplac111 (wild type), pAS24-NTAP/2xFLAG-SPB4 (GAL::SPB4-TAP), or pAS24-NTAP/2xFLAG-DNSPB4 (more ...)
To assess whether this detected 40S r-subunit shortage was due to a biogenesis defect, we first analyzed pre-rRNA processing by Northern hybridization at various time points after induction by galactose of either the Spb4-TAP or the Spb4(R360A)-TAP protein in the wild-type strain. As shown in , induction of DN Spb4(R360A)-TAP but not that of wild-type Spb4-TAP resulted in a clear time-dependent accumulation of 35S, 32S, and 27SB pre-rRNAs. Interestingly, the levels of the aberrant 23S RNA species did not increase. The levels of the other pre-rRNAs and those of mature rRNAs are not significantly altered upon induction of DN Spb4(R360A)-TAP compared to the induction of wild-type Spb4-TAP (). We also assessed the levels of 25.5S and the different 27S pre-rRNAs by primer extension analysis. As shown in Fig. S3 in the supplemental material, the steady-state levels of 25.5S, 27SA2, and 27SA3 slightly increased upon overexpression of Spb4(R360A)-TAP; consistent with the results obtained by Northern hybridization, the steady-state levels of both 27SBL and 27SBS pre-rRNAs significantly increased upon overexpression of Spb4(R360A)-TAP.
Fig. 6. Expression of the dominant-negative Spb4(R360A)-TAP protein inhibits pre-rRNA processing. YDK37-1A cells harboring YCplac111-SPB4 (SPB4) were grown at 30°C in SD-Leu medium or transferred to SGly/Lac-Leu and then shifted to SGal-Leu (Gal) for (more ...)
Finally, to determine whether or not the overexpression of Spb4(R360A)-TAP impairs nuclear export of preribosomal particles, we analyzed the localization of the L25-eGFP and S3-eGFP r-subunit protein reporters upon induction by galactose of either the Spb4-TAP or the Spb4(R360A)-TAP protein in the wild-type strain. Our results indicate that the localization of neither reporter was affected upon induction of either wild-type or DN Spb4 proteins (see Fig. S4A in the supplemental material). As a positive control for nuclear retention of r-protein reporters, we depleted the cells of the Spb4 and Fal1 proteins (see Fig. S4B). In all cases, the tested r-protein reporters did incorporate into preribosomal particles (see Fig. S5) (data not shown).
Altogether, these analyses suggest that DN Spb4(R360A)-TAP would be trapped into 90S preribosomal particles, delaying their further maturation. Alternatively, the ribosome maturation defects observed would indirectly arise as the consequence of the inefficient recycling of trans-acting factors of 90S preribosomal particles that could not dissociate from the defective pre-60S r-particles that accumulate in the DN mutant. To distinguish between these two possibilities, we investigated the protein and RNA compositions of preribosomal particles purified with TAP-tagged DN Spb4(R360A) after overexpression from a GAL promoter ( and ). As shown in Table S1 in the supplemental material, we identified only a few components of pre-60S r-particles, which were purified only using Spb4(R360A)-TAP as bait (e.g., Cgr1, Dbp10, Drs1, Mak5, Mak11, Nop4, Rsa4, and Ssf2). However, most coenriched pre-60S factors were also found in the affinity purification of wild-type Spb4 (see Table S1). Interestingly, Spb4(R360A)-TAP also copurified several components of RNA polymerase I and about 20 components of 90S preribosomal particles (see Table S1). Thus, we decided to determine which pre-rRNA species associate with Spb4(R360A)-TAP. To do this, RNA was extracted from TAP-purified complexes of both galactose-induced Spb4(R360A)-TAP and Spb4-TAP and analyzed by Northern blotting. As shown in , the profile of RNAs precipitated with overexpressed Spb4-TAP is overall similar to that of RNAs precipitated with Spb4-TAP expressed from its cognate promoter (compare A and B). Interestingly, 35S and 32S pre-rRNA-containing preribosomal particles were precipitated substantially more efficiently with Spb4(R360A)-TAP than with the TAP-tagged wild-type Spb4 fusion protein; in contrast, 27SB pre-rRNA was precipitated much less efficiently with Spb4(R360A)-TAP than with the TAP-tagged wild-type Spb4 fusion protein (B). In agreement with these results, DN HA-Spb4(R360A) protein is present in two peaks of high molecular mass as determined by sucrose gradient fractionation, one of them corresponding to 66S and 90S preribosomal particles (see Fig. S6B, lanes 13 to 16, in the supplemental material). Moreover, the DN HA-Spb4(R360A) protein was enriched within 90S preribosomal particles purified using Pwp2-GFP and Nop58-GFP as baits (C). Affinity purifications from GAL:HA-SPB4 cells yielded similar coenrichment of the wild-type HA-Spb4 protein in those 90S preribosomal complexes containing Pwp2-GFP or Nop58-GFP (B). However, while substantial wild-type HA-Spb4 protein is still detected in the pre-60S r-particles purified using Nop7-GFP or Rix1-GFP (B), practically background levels of DN HA-Spb4(R360A) protein are detected following Nop7-GFP or Rix1-GFP purifications (C, compare lanes Nop7-GFP and Rix1-GFP with lane None). Altogether, these results suggest that the DN mutant Spb4(R360A)-TAP is readily incorporated into 90S preribosomal particles but cannot properly dissociate from them. This likely impedes the correct processing of 35S pre-rRNA, thus producing a negative effect on the 40S r-subunit biogenesis pathway. Concomitantly, processing of 27SB pre-rRNA is also inhibited.
Conclusions and perspectives.
In this work, we have studied the physical environment of Spb4 by an affinity purification procedure. As expected from our previous functional analysis (9
) and different global interaction data sets (60
), Sbp4 associates mainly with pre-60S complexes. These complexes seem to be a heterogeneous mixture of distinct early nucle(ol)ar pre-60S r-particles (i.e., Nop7-purified complex) containing predominantly 27SB pre-rRNAs but also 27SA pre-rRNAs. Little association with pre-60S r-particles containing 7S pre-rRNAs was found, suggesting that the release of Spb4 from pre-60S r-particles might occur concomitantly with cleavage of 27SB pre-rRNAs. A small fraction of 35S pre-rRNA also coprecipitates with Spb4-TAP; therefore, Spb4 could bind at the level of 90S preribosomal particles. The association with the latter particles was more obvious when the DN Spb4(R360A) protein was analyzed, strongly suggesting that the mutant Spb4 binds to the 90S preribosomal particles but is inefficiently released from them. The copurification of several RNA polymerase I components with Spb4-TAP but more clearly with DN Spb4(R360A)-TAP protein strongly suggests that the association of Spb4 with 90S preribosomal particles may be cotranscriptional. Alternatively, it is also plausible that Spb4 associates mainly with pre-60S ribosomal particles, and the weak association between Spb4 and 90S preribosomal particles that we detected might reflect its premature and unproductive binding. Under wild-type conditions, the Spb4 protein could easily dissociate from these 90S preribosomal particles, most likely following ATP hydrolysis; however, the DN Spb4 protein, which is expected to have no ATPase activity, would have a reduced ability to dissociate from the 90S preribosomal particles in which it is engaged, leading to a jammed maturation pathway.
At first glance, the pre-rRNA processing defects due to the overexpression of the Spb4(R360A) protein mirrored those previously observed upon Spb4 mutation (spb4
) or depletion (9
). However, polysome profile analysis suggested that the inhibition of 35S pre-rRNA processing in the DN mutant is stronger than that upon mutation or depletion of Spb4, since a 40S r-subunit shortage was detected in the first case while a reduction in the levels of 60S r-subunits was detected in the second one (9
). Strikingly, this inhibition was not accompanied by the accumulation of aberrant 23S pre-rRNA. It has been clearly demonstrated that the RNA helicases Prp43 and Has1 are required for the synthesis of both 40S and 60S r-subunits (6
). Prp43 is present within 90S, pre-40S, and pre-60S r-particles (4
). Has1 associates with 90S and pre-60S r-particles (13
). Whether or not the particular pre-rRNA processing defects observed for the DN Spb4(R360A) protein reflect a dual function of Spb4 in both 40S and 60S r-subunit biogenesis needs further investigation. Thus, it must be addressed whether or not Spb4 could act twice during ribosome biogenesis, first as part of the 90S preribosomal particles, which could be a more transient interaction, and then again as part of the pre-60S particles, which would be a more stable interaction.
RNA helicases may require cofactor proteins for optimal function in vivo
. These cofactors could stimulate or inhibit the ATPase and helicase activities and confer substrate specificity and/or modify the affinity of the RNA helicase for its substrate (discussed in references 3
). To date, cofactors have been reported for only two out of the 19 RNA helicases involved in yeast ribosome biogenesis, Dpb8 and Prp43 (23
). These cofactors, named Esf2 for Dpb8 and Pfa1 for Prp43 (23
), stimulate the RNA-dependent ATPase activities of their respective RNA helicase partners in vitro
. Pfa1 also stimulates the RNA helicase activity of Prp43 in vitro
). Strikingly, only Pfa1 has been demonstrated to interact directly with Prp43 (35
). This interaction is stoichiometric, RNA independent, and salt resistant (35
). Unfortunately, no polypeptide was copurified in stoichiometric amounts with TAP-tagged wild-type and DN versions of Spb4. Further studies are needed to determine whether Spb4 has a direct cofactor(s). We firmly believe that the combination of biochemical and genetic approaches will allow the identification of such a partner(s).