Correction
9 Mar 2026: Wang L, Li P, Zeng P, Xie D, Gao M, et al. (2026) Correction: Dosage suppressors of gpn2ts mutants and functional insights into the role of Gpn2 in budding yeast. PLOS ONE 21(3): e0344594. https://doi.org/10.1371/journal.pone.0344594 View correction
Figures
Abstract
Gpn2 is a highly conserved protein essential for the assembly of RNA polymerase II (RNAPII) in eukaryotic cells. Mutations in Gpn2, specifically Phe105Tyr and Leu164Pro, confer temperature sensitivity and significantly impair RNAPII assembly. Despite its crucial role, the complete range of Gpn2 functions remains to be elucidated. To further explore these functions, we conducted large-scale multicopy suppressor screening in budding yeast, aiming to identify genes whose overexpression could mitigate the growth defects of a temperature-sensitive gpn2 mutant (gpn2ts) at restrictive temperatures. We screened over 30,000 colonies harboring plasmids from a multicopy genetic library and identified 31 genes that rescued the growth defects of gpn2ts to various extents. Notably, we found that PAB1, CDC5, and RGS2 reduced the drug sensitivity of gpn2ts mutants. These findings lay a theoretical foundation for future studies on the function of Gpn2 in RNAPII assembly.
Citation: Wang L, Li P, Zeng P, Xie D, Gao M, Ma L, et al. (2024) Dosage suppressors of gpn2ts mutants and functional insights into the role of Gpn2 in budding yeast. PLoS ONE 19(12): e0313597. https://doi.org/10.1371/journal.pone.0313597
Editor: Jorge Perez-Fernandez, Universidad de Jaen, SPAIN
Received: June 4, 2024; Accepted: October 28, 2024; Published: December 6, 2024
Copyright: © 2024 Wang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This work was supported by grants from the National Key Research and Development Program of China (2023YFD1401500 to F.Z.), the National Natural Science Foundation of China (no. 32070574 to F.Z.), the Natural Science Foundation of Hebei Province (C2024204178 to F.Z.), and Operation Expenses for Universities' Basic Scientific Research of Hebei Province (KY2024043 to F.Z.).
Competing interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Introduction
Gpn2 is a conserved member of the Gpn-loop GTPase family, which includes essential proteins Gpn1 (Npa3) and Gpn3. This family, originating from a single precursor gene in archaea, exhibits significant sequence divergence yet retains distinct functional roles [1]. In yeast, functional defects in Npa3, Gpn2, or Gpn3, as well as the knockdown of Gpn1 or Gpn3 in human cells, result in the cytoplasmic accumulation of RNA polymerase II (RNAPII) subunits Rpb1 and Rpb3. This suggests a direct involvement of these proteins in RNAPII assembly [2–8]. Particularly, Gpn2 has been identified as critical for the assembly of the Rpb3 subcomplex during RNAPII biogenesis in budding yeast [9]. However, the full spectrum of Gpn2 functions remains unclear.
In previous studies, Gpn2 was identified as a key factor in RNAPII assembly, directly interacting with the Rpb12 and Rpb3 subunits, as well as the Rba50 assembly factor [9]. At the same time, genetic screening further confirmed the association of Gpn2 with Rpb12, Rpb3 and Rba50 [9]. The gpn2ts mutant demonstrates temperature sensitivity and is susceptible to the transcription inhibitor mycophenolic acid (MPA) [9]. Furthermore, the Gpn2-Rba50 complex is conserved and essential for cellular functional viability. Deficiencies in Gpn2 and Rba50 activity disrupt the assembly of the Rpb3 subcomplex, adversely affecting subsequent stages in the assembly pathway. Notably, our findings reveal that Gpn2 promotes and regulates the integration of Rpb2 with Rba50. The Gpn2-Rba50 complex also plays a similar role in the assembly of RNAP III [10].
The power of unraveling novel functional relationships through genetic interactions, such as gene-gene or epistatic interactions, is well-documented. Suppression is a quintessential genetic interaction that facilitates the discovery of novel interactions and the identification of new genes [11, 12]. Multicopy genetic screening stands as one of the most effective high-throughput methods for identifying dosage suppressors [13, 14]. In yeast, multicopy libraries are commonly utilized to isolate genes that confer suppression phenotypes through elevated expression levels [15]. This screening method can simultaneously identify multiple genes related to a query gene, thereby unveiling new gene functions and providing insights into the roles of previously uncharacterized genes. Such multicopy genetic screens have proven successful in budding yeast, where it was determined that RPB12 and RBA50 act as suppressors for gpn2ts mutants. This genetic evidence demonstrates that Gpn2, in conjunction with Rpb12 and Rba50, directly influences assembly of the Rpb3 subcomplex [9]. Similarly, our studies have shown that RBA50 and GPN3 serve as multicopy genetic suppressors for npa3ts mutants [9].
Based on previous research that identified suppressors such as Rpb12, Rba50, and other GPNs [9], we sought to further elucidate the essential functions of Gpn2 by conducting a systematic large-scale genetic screen for additional suppressors of gpn2ts mutants [16]. In this study, we report the identification of 31 suppressor genes that enhance the growth of gpn2ts and reveal multiple roles for Gpn2 in budding yeast. Based on the mechanism of genetic suppression, we propose potential functions for previously unidentified suppressors, such as Pab1, Cdc5, and Rgs2, which may act as chaperonins in RNAPII assembly. Additionally, we conducted drug sensitivity assays, which demonstrated that PAB1, CDC5, and RGS2 can mitigate the effects of DNA-damaging drugs and transcriptional inhibitors on gpn2ts at 32°C. Furthermore, at 34°C, PAB1, CDC5, and RGS2 each can compensate for gpn1ts, with CDC5 reducing the sensitivity of gpn1ts to 4-nitroquinoline oxide (4-NQO) tablets. Through fluorescence localization of Rpb1, we observed that high expression of CDC5 can compensate for the nuclear dispersion observed in the gpn2ts mutant. These findings provide a theoretical foundation for in-depth exploration of the regulatory mechanisms by which Gpn2 is involved in RNAPII assembly.
Materials and methods
Yeast strains and media
The gpn2ts strains (W303-1a, gpn2-del::gpn2-F105Y L164P-13myc:URA3) were utilized for the multicopy genetic suppressor screen assessing temperature sensitivity at 34°C [9]. Derivative strains were used to evaluate suppression levels and to test sensitivity to DNA-damaging agents. All manipulations were performed using the Saccharomyces cerevisiae W303-1a strain (MATa; ade2-1; ura3-1; his3-11, 15; trp1-1; leu2-3, 112; can1-100). The RPB1 with a C-terminal GFP tag was generated at their genomic loci by integration of the PCR-generated cassettes amplified from genomic DNA of the Yeast GFP Clone Collection, using oligonucleotides complementary to the sequences flanking the endogenous loci [9].
Yeast-peptone-dextrose (YPD) is a rich medium that contains 1% yeast extract, 2% peptone, and 2% glucose. Synthetic dropout (SD) is a synthetic medium supplemented with amino acids. Yeast cells were cultured at specified temperatures in YPD, except for those carrying pRS425 (2 μm)-based multicopy constructs, which were grown in leucine-free SD medium (SD-Leu-, 0.67% yeast nitrogen base [YNB], 2% glucose, 0.02% tryptophan, 0.02% histidine, 0.02% adenine, and 0.02% uracil). Mycophenolic acid, hydroxyurea, methyl methane sulphonate, and 4-nitroquinoline oxide (Sigma) were added at the indicated concentrations. For ten-fold serial dilution assays, cultures grown exponentially at 25°C were spotted onto plates in ten-fold serial dilutions and incubated at specified temperatures for three to four days before imaging [10].
Isolation of multicopy suppressors of gpn2ts
Approximately 1×108 log-phase gpn2ts cells were transformed with 5 μg of plasmid DNA from a 2 μm-based multicopy yeast genomic DNA library. This library, partially digested with Sau3AI, was constructed in the pRS425 vector (2 μm ARS LEU2) with an average insert size of approximately 10 kb [16]. After transformation, cells were plated on SD-Leu- plates and incubated at 25°C for one day, followed by 34°C for five days. The transformation efficiency was approximately 1×104 transformants per μg of DNA. Plasmids from the colonies that survived at 34°C were isolated and transformed into the E. coli DH5α strain for amplification. To verify that the suppression function was attributable to the plasmid and not a chromosomal mutation in the original yeast transformant, plasmids were retransformed into the gpn2ts strain. The inserts were sequenced using the T7 and T3 primers. Subsequently, individual genes were subcloned into pRS425 and retested to confirm their role in the suppression of gpn2ts.
Fluorescence microscopy
For the fluorescence localization of Rpb1-GFP, cells harboring pRS425-based constructs were initially cultured in the SD-Leu- medium at 25°C. These cultures were then shifted to 34°C for a period of 3 hours. Subsequently, the selected culture was harvested and fixed using 70% ethanol. After fixation, the cells were washed and imaged using a Nikon fluorescent inverted microscope [9].
Results
Genetic screening for multicopy suppressors of the gpn2ts mutant
The rapid degradation of Gpn2 proved non-viable, underscoring its role as an essential protein [4]. Gpn2 contains a GTP-binding domain ranging from amino acids 7 to 261 and is highly conserved across species from archaea to yeast and humans. This evolutionary conservation suggests that Gpn2 may be pivotal in central biological processes (Fig 1A).
A: The phylogenetic tree was generated based on the multiple sequence alignment of full-length GPN2 amino acid sequences using the neighbor-Joining method in MEGA6 with 1000 bootstraps. Pink bars indicate domains predicted using the Pfam website (https://pfam.xfam.org). B: Molecular model of the yeast Gpn2 GTPase domain protein dimer. The GPN2 model was simulated by Zhanglab (https://zhanglab.ccmb.med.umich.edu). Conserved G3 domains (green) and the Phe105 (red) and Leu164 (blue) mutation sites of gpn2ts are shown. C: The gpn2ts mutant was transformed with multicopy plasmids pRS425 and pRS425-GPN2. The W303-1a (WT) strain was transformed with pRS425. These strains were spotted in tenfold serial dilutions on SD-Leu- plates and incubated at the indicated temperatures for four days. D: The indicated strains were spotted in ten-fold serial dilutions on YPD plates containing mycophenolic acid (MPA), hydroxyurea (HU), methyl methane sulfonate (MMS), or 4-nitroquinoline oxide (4-NQO), and incubated for four days at 25°C.
Through random mutagenesis of GPN2, we isolated a mutant, gpn2ts, featuring a double mutation (Phe105Tyr and Leu164Pro). Phe105 is situated near the G3 motif, a crucial GTPase-interacting domain essential for GTP binding and hydrolysis, whereas Leu164 resides within the conserved residue loop [9] (Fig 1B). This functionally compromised Gpn2 variant shows temperature sensitivity (Fig 1C) and increased susceptibility to genotoxic stresses such as hydroxyurea (HU), methyl methane sulphonate (MMS), and 4-nitroquinoline oxide (4-NQO) (Fig 1D). GPN2 was initially characterized as a chromosomal instability (CIN) mutant with defects in sister chromatid cohesion [5, 17]. These findings reinforce the hypothesis that GPN2 is indispensable for maintaining genomic integrity.
To delineate the specific biological functions of Gpn2, we performed a large-scale search for multicopy suppressor genes that could mitigate the temperature sensitivity of GPN2 mutants at restrictive temperatures (refer to the Materials and Methods section) (Fig 2).
The strategy for isolating multicopy suppressors is described in the ‘Materials and Methods’ section.
We screened over 30,000 colonies, isolating 100 potential suppressors. Plasmids extracted from these colonies were retransformed into DH5α, confirming 60 candidates. These plasmids were sequenced to identify the genomic regions responsible for suppression. Many contained overlapping genomic fragments, collectively representing 31 unique genomic regions. Further subcloning experiments pinpointed specific genes linked to the suppressive phenotype. We identified 31 such genes that varied in their ability to correct the growth defects of gpn2ts and their sensitivity to the transcription inhibitor MPA (Fig 3). These genes span various functions, pathways, and cellular locations, including DNA repair and RNA polymerase biogenesis (Table 1).
Tenfold serial dilutions of exponentially growing WT cells (W303-1a) carrying pRS425 mock plasmid and gpn2ts mutants carrying the indicated constructs were spotted onto SD-Leu- plates with or without MPA and incubated at the indicated temperatures for four days.
Multicopy genetic suppressors of gpn2ts
The suppressor genes identified can be categorized into different functional groups based on their known functions:
Genes encoding molecular chaperones.
LHS1. Lhs1 acts as a molecular partner to hsp70 within the endoplasmic reticulum lumen, playing a crucial role in peptide translocation and folding. It is regulated by the unfolded protein response pathway [18, 19]. Lhs1 has a homolog, GRP170, and recent research indicates that Lhs1 selectively recognizes misassembled forms of the episodic sodium channel (α ENaC) via its transmembrane domain [20, 21].
NAS2. Nas2, a member of the SLC13 gene family, is a Na+-coupled transporter originally isolated from high endothelial venule endothelial cells. It is known for transporting sulfates in conjunction with Na+ and also mediates the transport of oxygen anions from micronutrients such as selenium and chromium [22]. Additionally, Nas2 is an evolutionarily conserved partner in the assembly of the 19S proteasome regulatory granule (RP) base subcomplex [23]. Recent studies have shown that Nas2 plays a pivotal role in the assembly of six different ATPase subunits (Rpt1 to Rpt6) in the proteasome, facilitating one of the final steps in their assembly [24].
HSP150. Hsp150, an O-mannosylated heat shock protein, is part of the Pir protein family. It plays a crucial role in cell wall stability and is induced by heat shock, oxidative stress, and nitrogen limitation [25–27]. These chaperones may facilitate the proper folding of Gpn2 or interact with RNA polymerase subunits as assembly chaperones. They could represent novel RNAPII assembly factors, similar to Hsc82/Hsp82 and the R2TP/prefoldin-like complex involved in RNA polymerase assembly [28].
Genes affecting nuclear transport/splicing.
PAB1. Pab1 is the primary poly(A)-binding protein in yeast, playing a pivotal role in mRNA stability, nuclear export, and translation initiation [29, 30]. Its deletion affects various processes, including translation initiation, translation termination, and mRNA decay [31]. Furthermore, Pab1 has been recognized as a key marker for stress granules, which form through phase separation in response to physiological stress, resulting in the creation of hydrogels outside the cell [32]. We hypothesize that the formation of stress granules serves as a self-protective mechanism for cells.
NUP60. Nup60 is an FG-nucleoporin component of the central core of the nuclear pore complex, playing a crucial role in nucleocytoplasmic transport and maintaining the permeability barrier of the nuclear pore complex (NPC) [33, 34]. This study suggests that Gpn2, known as an RNAPII assembly factor, may also influence the nuclear transport of RNAPII, as evidenced by Gpn1 and Gpn3 acting as multicopy suppressors for the gpn2ts mutant [3, 35–38].
PRP46. Prp46, a member of the NineTeen Complex (NTC), is vital for pre-mRNA splicing [39, 40]. It may compensate for defective Gpn2 by altering splice-site recognition or promoting alternative splicing [41].
Transcription-/translation-related genes.
TFC3. Tfc3 is a subunit of the RNA polymerase III transcription initiation factor complex (TFIIIC) in Saccharomyces cerevisiae. Composed of six distinct subunits, TFIIIC binds to RNA polymerase III promoters, initiating the assembly of the transcription complex. The largest yeast subunit, tau138, encoded by TFC3, targets the B-box promoter element. Mutations in TFC3 disrupt tRNA and 5S RNA synthesis, underscoring its critical role in transcription [42–44].
SRB5 and MED6. Srb5 and Med6 are subunits of the RNAPII mediator complex, crucial for associating with core polymerase subunits to form the RNAPII holoenzyme, essential for transcriptional regulation [45, 46]. Additionally, the abundance of Med6 protein increases in response to DNA replication stress, highlighting its role in cellular stress responses [47, 48].
SGN1. SGN1 encodes a cytoplasmic RNA-binding protein featuring an RNA recognition motif (RRM). Overexpression of SGN1 has been shown to suppress the temperature-sensitive growth phenotype in specific alleles of TIF4631, TIF4632, and PAB1, all of which are involved in translational initiation. These findings imply that SGN1 may play a role in mRNA translation [49].
The suppressive effect of these genes likely influences transcription and translation processes. Overexpression may stabilize mRNA or protein levels of gpn2ts, thereby preserving their translation and the accumulation of residual active peptides [49]. These observations also raise the possibility that Gpn2 may directly or indirectly impact transcription or mRNA translation.
Ribosome-related genes.
RPL26A. Rpl26a, part of the ribosomal 60S subunit, binds to 5.8S rRNAs. While the deletion of RPL26A has minimal effects, overproduction of the yeast ribosomal protein Rpl26 fails to integrate into ribosomes, indicating that protein L26 is not essential for ribosome assembly and function [50, 51]. Additionally, RPL26A and RPL14A act as weak suppressors of a conditional mutation in the basal transcription factor TFIIIC, suggesting a supportive role in transcription under constrained conditions [52].
MRPL27. Mrpl27 is a component of the mitochondrial ribosomal large subunit. Mutants with disrupted MRPL27 genes are unable to grow on non-fermentable carbon sources, demonstrating that MRPL27 is crucial for mitochondrial functionality in yeast [53].
Protein regulation-related genes.
CDC34. CDC34 encodes a ubiquitin-conjugating enzyme (E2), serving as the catalytic subunit of the SCF ubiquitin-protein ligase complex. This complex is responsible for targeting proteins for degradation, including those implicated in cell cycle progression. Elevated levels of Cdc34 under DNA replication stress conditions suggest its crucial role in managing genotoxic stress [54, 55].
RPN10. Rpn10, a proteasome polyubiquitin receptor, features a ubiquitin-interacting motif that selectively binds to polyubiquitin chains, playing a vital role in intracellular proteolysis [56]. As part of the proteasome regulatory particle, Rpn10 primarily coordinates with the ubiquitin-type DSK2 receptor to form the Dsk2-proteasome, facilitating its function [57].
RGS2. Rgs2 serves as a negative regulator of glucose-induced cAMP signaling, in addition to its role as a general regulator of G protein signaling (RGS). It directly activates the GTPase activity of the heterotrimeric G protein alpha subunit, Gpa2p [58, 59]. Rgs2 may also regulate Gpn2, a GTPase from the GPN family, acting as a GTPase activating protein (GAP). As a multifunctional RGS protein, RGS2 regulates various signaling pathways linked to G proteins. Research interest in RGS2 extends beyond cancer to include implications in cardiovascular and cerebrovascular diseases.
Genes affecting genome stability.
CDC73. Cdc73 is a key component of the Paf1p complex and interacts with RNA polymerases I and II to modulate their activities. This protein is involved in transcription elongation and regulates ATR activity. Overexpression of CDC73 mRNA promotes carcinogenesis [60–63]. Additionally, studies indicate that Cdc73 plays a crucial role in maintaining genomic stability by mediating telomere homeostasis [64].
CDC5. Cdc5, a Polo-like kinase essential for the mitotic cell cycle, regulates the nuclear shape and the expansion of the nuclear envelope during mitosis. It plays a crucial role in the cellular response to DNA damage by attenuating the DNA damage checkpoint by reducing Rad53 hyperphosphorylation, allowing cells to adapt to DNA damage [65–68]. CDC5 also interacts with promoters of genes encoding microRNAs (MIRs) and DNA-dependent RNAPII, influencing MIR promoter activity and RNAPII occupancy. This interaction suggests that high expression of CDC5 may stabilize RNAPII, thereby mitigating the temperature sensitivity of gpn2ts [69].
RFC1. Rfc1 is a subunit of the heteropentameric replication factor C (RFC), a DNA-binding protein and ATPase essential in DNA repair processes. RFC binds to cell cycle checkpoint proteins, initiating signal transduction downstream of DNA damage checkpoints, and is involved in mismatch repair and the excision repair of damaged DNA [8, 70]. Although RFC1 has minimal impact on gpn2ts mutants under restricted temperatures, high expression of RFC1 at 32°C can alleviate the growth defects of these mutants, suggesting it may indirectly reduce gpn2ts sensitivity.
NHP6A. Nhp6A is a high-mobility group (HMG) protein that binds to DNA. It facilitates the maintenance of genome stability and plays important roles in transcription and DNA replication [71, 72]. In yeast, the NHP6A/B proteins are crucial for initiating RNA polymerase II transcription. Complementation of these proteins may not only enhance RNA polymerase II transcription but also partially compensate for the deficiencies observed in gpn2ts mutants [73].
In budding yeast, mutations in GPN2 have been shown to increase the frequency of chromosome fragment loss, potentially influencing the establishment of sister chromatid cohesion [17]. The defective Gpn2 protein disrupts the assembly of RNAPII, which may directly bind to the single-stranded DNA at the termini of DNA breaks, subsequently transcribing RNA and forming an R-loop [5, 74, 75]. This study demonstrates that overexpression of cell cycle regulatory genes such as CDC34, along with DNA repair-related genes, can mitigate the growth defects associated with gpn2ts mutants. The cell cycle significantly influences the choice of pathway for DNA double-strand break repair [76]. These findings suggest that GPN2 plays a role in DNA repair processes and impacts genome stability. Given that genomic instability is a hallmark of cancer, GPN2 may also have potential implications in cancer development.
Additional genes. This category encompasses three members of the seripauperin multigene family (PAU1, PAU3, and PAU23), primarily encoded in subtelomeric regions and negatively regulated by oxygen levels (Fig 4D) [77]. It is hypothesized that mutations in GPN2 may influence intracellular oxygen content. Additionally, this category includes genes of unknown function, such as EMC10, YMR272W-B, YER085C, ANS1, YNR075C-A, JJJ2, YBR071W, and YLR036C. Our study aims to provide valuable insights into their roles, potentially advancing functional research.
A-E: Network representation of interactions between suppressors and RNAPII subunits or among suppressors, visualized using Cytoscape. Known physical interactions are grouped by function, with connecting edge thickness and intensity indicating interaction strength (data from STRING, https://string-db.org). Node colors: red, suppressors; yellow, GPN2; all other nodes (black and gray in panels A-E) represent RNAPII subunits.
Possible suppression mechanisms
The potential suppression mechanisms of dosage suppressors of gpn2ts primarily include: 1) regulating or stabilizing the mutant protein; 2) engaging downstream factors that circumvent the need for the mutant protein. For instance, the RGS2 gene, which encodes a known GTPase-activating protein (GAP), may regulate or stabilize Gpn2 activity, aligning with the first mechanism. The chaperone proteins Hsp150, Nas2, and Lhs1 are crucial for the proper folding of newly synthesized polypeptides. Their overexpression may enhance the stability of either the Gpn2 mutant protein or essential subunits of RNAPII, fitting into the first suppression category. Additionally, transcription- and translation-related genes, including TFC3 and SGN1, suggest that increasing the levels of mutant GPN2 mRNA or protein can suppress the observed temperature-sensitive (TS) phenotype, also supporting the first explanation. Lastly, interaction network analysis places the Gpn2-centered pathway in parallel with the Cdc34 pathways, aligning with the second suppression mechanism (Fig 4A).
CDC5 is a suppressor of gpn1ts and gpn2ts
Multicopy genetic screening, a high-throughput method, is employed to identify dosage suppressors [13], enabling the simultaneous discovery of multiple genes related to a query gene, aiding in the discovery of new gene functions and providing insights into the functions of unknown genes. In our study, we selected the nuclear transporter Pab1 [29], the genome stability-related protein Cdc5 [67], and the G protein signaling regulator Rgs2 to investigate their roles in the assembly of RNAPII [59]. Our findings indicate that the overexpression of PAB1, CDC5, and RGS2 mitigates the sensitivity of gpn2ts strains to temperature fluctuations, DNA damage agents, and transcription inhibitors (Fig 5A). However, this suppression is limited to semi-restrictive temperatures, as these genes cannot prevent death at lethal temperatures. Additionally, experiments conducted on gpn1ts strains, which are sensitive to both high and low temperatures, revealed that overexpression of CDC5 reduced their sensitivity to low temperatures, DNA damage drugs, and transcriptional repression drugs. Similarly, the overexpression of PAB1, CDC5, and RGS2 decreased the sensitivity of gpn2ts strains to semi-limiting temperatures (Fig 5B). High expression of these genes substantially compensated for the growth defects of gpn2ts in the presence of DNA-damaging and transcription-inhibitory drugs (Fig 5B). Given that the Gpn family proteins were initially characterized in chromosomal instability (CIN) mutants [28], it is plausible that PAB1, CDC5, and RGS2 play crucial roles in regulating Gpn1 and Gpn2 activities that are essential for maintaining genome stability and transcriptional regulation.
A: Overexpression of PAB1, CDC5, and RGS2 reduces the sensitivity of gpn2ts strains to DNA-damaging agents. Tenfold serial dilutions of logarithmic-phase cultures were plated on SD-Leu- plates containing hydroxyurea (HU), methyl methane sulfonate (MMS), 4-nitroquinoline oxide (4-NQO), or mycophenolic acid (MPA) at the indicated concentrations and temperatures, and incubated for four days. B: Similarly, overexpression of PAB1, CDC5, and RGS2 reduces the sensitivity of gpn1ts strains to DNA-damaging agents. C: Overexpression of CDC5 reduces the fluorescence diffusion of Rpb1 in gpn2ts strains. Logarithmic-phase cultures were grown in SD-Leu- medium at 34°C for three hours, fixed with ethanol, and observed for fluorescence.
Previous studies have demonstrated that Gpn2 is involved in RNAPII assembly [9]. To delve deeper into the functions of PAB1, CDC5, and RGS2 in relation to this process, we examined their impact on gpn2ts mutants. Our findings revealed that elevated expression of CDC5 mitigated the fluorescence dispersion of Rpb1, the largest subunit of RNAPII, indicating an effective stabilization or localization effect, whereas RGS2 and PAB1 showed no such influence (Fig 5C). Although overexpression of CDC5 does not completely restore the nuclear localization of Rpb1, it underscores the close association between Cdc5 and Gpn2, suggesting a potential role of CDC5 in modulating RNAPII assembly dynamics influenced by Gpn2.
Discussion
Through our high-throughput screening, we identified a range of genes whose overexpression can suppress the defects observed in gpn2ts strains. The comprehensive nature of the screen allowed us to identify a substantial number of candidate genes, each of which was subsequently confirmed through meticulous re-cloning and re-testing processes. Notably, several genomic regions, such as CDC34, were isolated multiple times independently, bolstering our confidence in the reliability and significance of our findings concerning genes that interact genetically with GPN2.
Further analysis of the suppressor interaction network revealed five distinct modules, illustrating diverse genetic interactions. Some suppressors were found to interact with subunits of RNAPII (Fig 4). Specifically, one module (Fig 4A) contains regulatory proteins; Fig 4B and 4C focus on transcription and translation-related factors. Additionally, other modules include members of the seripauperin multigene family [78] (Fig 4D) and chaperone proteins (Fig 4E). These interactions were delineated from manually curated clusters within the physical protein-protein interaction (PPI) database, providing valuable insights into the functions of these genes.
Our findings enable us to infer the functions of previously uncharacterized genes such as EMC10 and JJJ2. EMC10 is shown to interact with CDC34 in the PPI database (Fig 4A). Given that CDC34 also acts as a dosage suppressor of gpn2ts, we propose that EMC10 may function as a regulator of Cdc34. Overexpression of EMC10 may thus modulate Cdc34 activity to compensate for the defective GPN2. JJJ2, featuring a J-domain homologous to the E. coli DnaJ molecular chaperone [79], interacts with the chaperone protein Lhs1 (Fig 4E). This suggests that Jjj2, along with Hsp150, Nas2, and Lhs1, may function as novel RNAPII assembly factors, similar to the roles of Hsp90/Hsp82 and the R2TP/prefoldin-like complex in the assembly of RNA polymerases [28]. In addition, we identified Rgs2, a GTPase-activating protein (GAP) that may regulate Gpn2 through its GAP activity.
Previous studies have demonstrated genetic and physiological interactions among the three members of the Gpn family, where GPN1 and GPN3 can suppress gpn2ts mutants [9]. In this experiment, CDC5 suppresses both gpn2ts and gpn1ts mutants (Fig 5A and 5B). Furthermore, it complements the fluorescence diffusion of Rpb1 in the nucleus (Fig 5C), suggesting that CDC5 may also play an important role in transcriptional regulation. CDC5 is an essential gene, and its overexpression causes phenotypes such as cell cycle arrest, actin cytoskeleton abnormalities, and other defects. As a protein kinase, it plays a critical regulatory role [66, 67]. Whether a reduction in Cdc5 enzyme activity can compensate for GPN mutants remains an open question, and its connection with the GPN family warrants further investigation.
There appear to be numerous ways in which the temperature-sensitive (TS) phenotype can be bypassed. Our results suggest that RNA polymerase activity, ribosome biogenesis, and DNA repair are linked to the functional defect in gpn2ts mutants. Additionally, enhanced transcription, translation, and nuclear export can bypass the TS phenotype, possibly by influencing effector proteins. We believe that the findings from this multicopy suppressor screening will offer valuable insights into the functions of Gpn2 and its associated biological pathways.
Acknowledgments
We thank all members of the laboratory for their contributions. We also appreciate the assistance in sequencing from our biotechnology collaborators.
References
- 1. Gras S, Chaumont V, Fernandez B, Carpentier P, Charrier-Savournin F, Schmitt S, et al. Structural insights into a new homodimeric self-activated GTPase family. EMBO reports. 2007; 8, 569–575. PMCID: PMC2002535. pmid:17468740
- 2. Carré C, Shiekhattar R. Human GTPases associate with RNA polymerase II to mediate its nuclear import. Mol Cell Biol. 2011; 31: 3953–3962. PMCID: PMC3187354. pmid:21768307
- 3. Forget D, Lacombe AA, Cloutier P, Al-Khoury R, Bouchard A, Lavallée-Adam M, et al. The protein interaction network of the human transcription machinery reveals a role for the conserved GTPase RPAP4/GPN1 and microtubule assembly in nuclear import and biogenesis of RNA polymerase II. Mol Cell Proteomics. 2010; 9: 2827–2839. pmid:20855544.
- 4. Liu X, Xie D, Hua Y, Zeng P, Ma L, Zeng F. Npa3 interacts with Gpn3 and assembly factor Rba50 for RNA polymerase II biogenesis. FASEB J. 2020; 34: 15547–15558. pmid:32985767.
- 5. Minaker SW, Filiatrault MC, Ben-Aroya S, Hieter P, Stirling PC. Biogenesis of RNA polymerases II and III requires the conserved GPN small GTPases in Saccharomyces cerevisiae. Genetics. 2013; 193: 853–864. pmid:23267056.
- 6. Niesser J, Wagner FR, Kostrewa D, Mühlbacher W, Cramer P. Structure of Gpn-loop GTPase Npa3 and implications for RNA polymerase II assembly. Mol Cell Biol. 2015; 36: 820–31. pmid:26711263.
- 7. Reyes-Pardo H, Barbosa-Camacho AA, Pérez-Mejía AE, Lara-Chacón B, Salas-Estrada LA, Robledo-Rivera AY, et al. A nuclear export sequence in Gpn-loop GTPase 1, an essential protein for nuclear targeting of RNA polymerase II, is necessary and sufficient for nuclear export. Biochim Biophys Acta. 2012; 1823: 1756–1766. pmid:22796641.
- 8. Sakato M O’Donnell M, Hingorani MM. A central swivel point in the RFC clamp loader controls Pcna opening and loading on DNA. J Mol Biol. 2012; 416: 163–75. pmid:22197374.
- 9. Zeng F, Hua Y, Liu X, Liu S, Lao K, Zhang Z, et al. Gpn2 and Rba50 directly participate in the assembly of the Rpb3 subcomplex in the biogenesis of RNA polymerase II. Mol Cell Biol. 2018; 38: e00091–18. pmid:29661922.
- 10. Xie D, Zhao X, Ma L, Wang L, Li P, Cheng H, et al. Rba50 and Gpn2 recruit the second largest subunits for the assembly of RNA polymerase II and III. Int J Biol Macromol. 2022; 204: 565–575. pmid:35176321.
- 11. Casselton L, Zolan M. The art and design of genetic screens: filamentous fungi. Nat Rev Genet. 2002; 3: 683–97. pmid:12209143.
- 12. Liang C, Weinreich M, Stillman B. Orc and Cdc6p interact and determine the frequency of initiation of DNA replication in the genome. Cell. 1995; 81: 667–76. pmid:7774008.
- 13. Gazy I, Liefshitz B, Bronstein A, Parnas O, Atias N, Sharan R, et al. A genetic screen for high copy number suppressors of the synthetic lethality between elg1Δ and srs2Δ in yeast. G3 (Bethesda). 2013; 3: 917–26. pmid:23704284.
- 14. Kondou Y, Higuchi M, Matsui M. High-throughput characterization of plant gene functions by using gain-of-function technology. Annu Rev Plant Biol. 2010; 61: 373–93. pmid:20192750.
- 15. Ramer SW, Elledge SJ, Davis RW. Dominant genetics using a yeast genomic library under the control of a strong inducible promoter. Proc Natl Acad Sci USA. 1992; 89: 11589–93. pmid:1454852.
- 16. Zeng F, Quintana DG. High-copy yeast library construction and high-copy rescue genetic screen in Saccharomyces cerevisiae. Methods Mol Biol. 2021; 2196: 77–83. pmid:32889714.
- 17. Ben-Aroya S, Coombes C, Kwok T, O’Donnell KA, Boeke JD, Hieter P. Toward a comprehensive temperature-sensitive mutant repository of the essential genes of Saccharomyces cerevisiae. Mol Cell. 2008; 30: 248–58. pmid:18439903.
- 18. Baxter BK, James P, Evans T, Craig EA. SSI1 encodes a novel Hsp70 of the Saccharomyces cerevisiae endoplasmic reticulum. Mol Cell Biol. 1996; 16: 6444–56. pmid:8887673.
- 19. Craven RA, Egerton M, Stirling CJ. A novel Hsp70 of the yeast ER lumen is required for the efficient translocation of a number of protein precursors. EMBO J. 1996; 15: 2640–50. pmid:8654361.
- 20. Buck TM, Jordahl AS, Yates ME, Preston GM, Cook E, Kleyman TR, et al. Interactions between intersubunit transmembrane domains regulate the chaperone-dependent degradation of an oligomeric membrane protein. Biochem J. 2017; 474: 357–376. pmid:27903760.
- 21. Sukhoplyasova M, Keith AM, Perrault EM, et al. Lhs1 dependent ERAD is determined by transmembrane domain context. Biochem J. 2023; 480: 1459–1473. pmid:37702403.
- 22. Miyauchi S, Srinivas SR, Fei YJ, Gopal E, Umapathy NS, Wang H, et al. Functional characteristics of Nas2, a placenta-specific Na+-coupled transporter for sulfate and oxyanions of the micronutrients selenium and chromium. Placenta. 2006; 27: 550–9. pmid:16129486.
- 23. Saeki Y, Toh-E A, Kudo T, Kawamura H, Tanaka K. Multiple proteasome-interacting proteins assist the assembly of the yeast 19S regulatory particle. Cell. 2009; 137: 900–13. pmid:19446323.
- 24. Sekaran S, Park S. The penultimate step of proteasomal ATPase assembly is mediated by a switch dependent on the chaperone Nas2. J Biol Chem. 2023; 299: 102870. pmid:36621624.
- 25. Ezaki B, Gardner RC, Ezaki Y, Kondo H, Matsumoto H. Protective roles of two aluminum (Al)-induced genes, HSP150 and SED1 of Saccharomyces cerevisiae, in Al and oxidative stresses. FEMS Microbiol Lett. 1998; 159: 99–105. pmid:9485599.
- 26. Kapteyn JC, Van Egmond P, Sievi E, Van Den Ende H, Makarow M, Klis FM. The contribution of the O-glycosylated protein Pir2p/Hsp150 to the construction of the yeast cell wall in wild-type cells and beta 1,6-glucan-deficient mutants. Mol Microbiol. 1999; 31: 1835–44. pmid:10209754.
- 27. Russo P, Simonen M, Uimari A, Teesalu T, Makarow M. Dual regulation by heat and nutrient stress of the yeast HSP150 gene encoding a secretory glycoprotein. Mol Gen Genet. 1993; 239: 273–80. pmid:8510655.
- 28. Boulon S, Pradet-Balade B, Verheggen C, Molle D, Boireau S, Georgieva M, et al. HSP90 and its R2tp/Prefoldin-like cochaperone are involved in the cytoplasmic assembly of RNA polymerase II. Mol Cell. 2010; 39: 912–924. pmid:20864038.
- 29. Brambilla M, Martani F, Bertacchi S, Vitangeli I, Branduardi P. The Saccharomyces cerevisiae poly (A) binding protein (Pab1): Master regulator of mRNA metabolism and cell physiology. Yeast. 2019; 36: 23–34. pmid:30006991.
- 30. Brune C, Munchel SE, Fischer N, Podtelejnikov AV, Weis K. Yeast poly(A)-binding protein Pab1 shuttles between the nucleus and the cytoplasm and functions in mRNA export. RNA. 2005; 11: 517–31. pmid:15769879.
- 31. Mangkalaphiban K, Ganesan R, Jacobson A. Direct and indirect consequences of PAB1 deletion in the regulation of translation initiation, translation termination, and mRNA decay. bioRxiv [Preprint]. 2023. 2023.05.31.543082. pmid:37398227.
- 32. Riback JA, Katanski CD, Kear-Scott JL, Pilipenko EV, Rojek AE, Sosnick TR, et al. Stress-triggered phase separation is an adaptive, evolutionarily tuned response. Cell. 2017; 168:1028–1040.e19. pmid:28283059.
- 33. Denning D, Mykytka B, Allen NP, Huang L, Al Burlingame, Rexach M. The nucleoporin Nup60p functions as a Gsp1p-GTP-sensitive tether for Nup2p at the nuclear pore complex. J Cell Biol. 2001; 154: 937–50. pmid:11535617.
- 34. Powrie EA, Zenklusen D, Singer RH. A nucleoporin, Nup60p, affects the nuclear and cytoplasmic localization of ASH1 mRNA in S. cerevisiae. RNA. 2011; 17: 134–44. pmid:21036941.
- 35. Cristóbal-Mondragón GR, Lara-Chacón B, Santiago Á, De-la-Rosa V, González-González R, Muñiz-Luna JA, et al. Fret-based analysis and molecular modeling of the human Gpn-loop GTPases 1 and 3 heterodimer unveils a dominant-negative protein complex. FEBS J. 2019; 286: 4797–4818. pmid:31298811.
- 36. Dekker C, Stirling PC, McCormack EA, Filmore H, Paul A, Brost RL, et al. The interaction network of the chaperonin Cct. EMBO J. 2008; 27:1827–39. pmid:18511909.
- 37. Di Croce L. Regulating the shuttling of eukaryotic RNA polymerase II. Mol Cell Biol. 2011; 31: 3918–20. pmid:21844223.
- 38. Wild T, Cramer P. Biogenesis of multisubunit RNA polymerases. Trends Biochem Sci. 2012; 37: 99–105. pmid:22260999.
- 39. Albers M, Diment A, Muraru M, Russell CS, Beggs JD. Identification and characterization of Prp45p and Prp46p, essential pre-mRNA splicing factors. RNA. 2003; 9: 138–50. pmid:12554883.
- 40. Ben-Yehuda S, Dix I, Russell CS, McGarvey M, Beggs JD, Kupiec M. Genetic and physical interactions between factors involved in both cell cycle progression and pre-mRNA splicing in Saccharomyces cerevisiae. Genetics. 2000; 156: 1503–17. pmid:11102353.
- 41. van Leeuwen J, Pons C, Boone C, Andrews BJ. Mechanisms of suppression: The wiring of genetic resilience. Bioessays. 2017; 39: https://doi.org/10.1002/bies.201700042 pmid:28582599.
- 42. Rozenfeld S, Thuriaux P. Genetic interactions within TFIIIC, the promoter-binding factor of yeast RNA polymerase III. Mol Genet Genomics. 2001; 265: 705–10. pmid:11459191.
- 43. Lefebvre O, Carles C, Conesa C, Swanson RN, Bouet F, Riva M, et al. TFC3: gene encoding the B-block binding subunit of the yeast transcription factor IIIC. Proc Natl Acad Sci U S A. 1992; 89: 10512–6. pmid:1279682.
- 44. Lefebvre O, Rüth J, Sentenac A. A mutation in the largest subunit of yeast TFIIIC affects tRNA and 5S RNA synthesis. Identification of two classes of suppressors. J Biol Chem. 1994; 269: 23374–81. pmid:8083243.
- 45. Kornberg RD. Mediator and the mechanism of transcriptional activation. Trends Biochem Sci. 2005; 30: 235–9. pmid:15896740.
- 46. Mukundan B, Ansari A. Novel role for mediator complex subunit Srb5/Med18 in termination of transcription. J Biol Chem. 2011; 286: 37053–7. pmid:21921038
- 47. Lee YC, Min S, Gim BS, Kim YJ. A transcriptional mediator protein that is required for activation of many RNA polymerase II promoters and is conserved from yeast to humans. Mol Cell Biol. 1997; 17: 4622–32. pmid:9234719.
- 48. Tkach JM, Yimit A, Lee AY, Riffle M, Costanzo M, Jaschob D, et al. Dissecting DNA damage response pathways by analysing protein localization and abundance changes during DNA replication stress. Nat Cell Biol. 2012; 14: 966–76. pmid:22842922
- 49. Winstall E, Sadowski M, Kuhn U, Wahle E, Sachs AB. The Saccharomyces cerevisiae RNA-binding protein Rbp29 functions in cytoplasmic mRNA metabolism. J Biol Chem. 2000; 275: 21817–26. pmid:10764794.
- 50. Sung MK, Porras-Yakushi TR, Reitsma JM, Huber FM, Sweredoski MJ, Hoelz A, et al. A conserved quality-control pathway that mediates degradation of unassembled ribosomal proteins. Elife. 2016; 5: e19105. pmid:27552055.
- 51. Babiano R, Gamalinda M, Woolford JL Jr, de la Cruz J. Saccharomyces cerevisiae ribosomal protein L26 is not essential for ribosome assembly and function. Mol Cell Biol. 2012 Aug; 32(16): 3228–41. pmid:22688513.
- 52. Dieci G, Ruotolo R, Braglia P, Carles C, Carpentieri A, Amoresano A, et al. Positive modulation of RNA polymerase III transcription by ribosomal proteins. Biochem Biophys Res Commun. 2009; 379: 489–93. pmid:19116144.
- 53. Graack HR, Grohmann L, Kitakawa M. The nuclear coded mitoribosomal proteins YmL27 and YmL31 are both essential for mitochondrial function in yeast. Biochimie. 1991; 73: 837–44. pmid:1764528.
- 54. Goebl MG, Goetsch L, Byers B. The Ubc3 (Cdc34) ubiquitin-conjugating enzyme is ubiquitinated and phosphorylated in vivo. Mol Cell Biol. 1994; 14: 3022–9. pmid:8164658.
- 55. Williams KM, Qie S, Atkison JH, Salazar-Arango S, Alan Diehl J, Olsen SK. Structural insights into E1 recognition and the ubiquitin-conjugating activity of the E2 enzyme Cdc34. Nat Commun. 2019; 10: 3296. pmid:31341161.
- 56. Chandra A, Chen L, Madura K. Synthetic lethality of rpn11-1 rpn10Δ is linked to altered proteasome assembly and activity. Curr Genet. 2010; 56: 543–57. pmid:20941496.
- 57. Zuin A, Bichmann A, Isasa M, Puig-Sàrries P, Díaz LM, Crosas B. Rpn10 monoubiquitination orchestrates the association of the ubiquilin-type DSK2 receptor with the proteasome. Biochem J. 2015; 472: 353–65. pmid:26450923.
- 58. Kehrl JH, Sinnarajah S. RGS2: a multifunctional regulator of G-protein signaling. Int J Biochem Cell Biol. 2002; 34: 432–8. pmid:11906816.
- 59. Versele M, de Winde JH, Thevelein JM. A novel regulator of G protein signalling in yeast, Rgs2, downregulates glucose-activation of the cAMP pathway through direct inhibition of Gpa2. EMBO J. 1999; 18: 5577–91. pmid:10523302.
- 60. Landsverk HB, Sandquist LE, Sridhara SC, Rødland GE, Sabino JC, de Almeida SF, et al. Regulation of ATR activity via the RNA polymerase II associated factors CDC73 and PNUTS-PP1. Nucleic Acids Res. 2019; 47: 1797–1813. pmid:30541148.
- 61. Mueller CL, Jaehning JA. Ctr9, Rtf1, and Leo1 are components of the Paf1/RNA polymerase II complex. Mol Cell Biol. 2002; 22: 1971–80. pmid:11884586.
- 62. Zhang Y, Sikes ML, Beyer AL, Schneider DA. The Paf1 complex is required for efficient transcription elongation by RNA polymerase I. Proc Natl Acad Sci USA. 2009; 106: 2153–8. pmid:19164765.
- 63. Zheng HC, Gong BC, Zhao S. The clinicopathological and prognostic significances of CDC73 expression in cancers: a bioinformatics analysis. Oncotarget. 2017; 8: 95270–95279. pmid:29221126.
- 64. Melnick AF, Mullin C, Lin K, McCarter AC, Liang S, Liu YE, et al. Cdc73 protects notch-induced T-cell leukemia cells from DNA damage and mitochondrial stress. Blood. 2023;142: 2159–2174. pmid:37616559.
- 65. Botchkarev VV Jr, Haber JE. Functions and regulation of the polo-like kinase Cdc5 in the absence and presence of DNA damage. Curr Genet. 2018; 64: 87–96. pmid:28770345.
- 66. Coutelier H, Xu Z, Morisse MC, Lhuillier-Akakpo M, Pelet S, Charvin G, et al. Adaptation to DNA damage checkpoint in senescent telomerase-negative cells promotes genome instability. Genes Dev. 2018; 32: 1499–1513. pmid:30463903.
- 67. Vidanes GM, Sweeney FD, Galicia S, Cheung S, Doyle JP, Durocher D, et al. CDC5 inhibits the hyperphosphorylation of the checkpoint kinase Rad53, leading to checkpoint adaptation. PLoS Biol. 2010; 8: e1000286. pmid:20126259.
- 68. Walters AD, May CK, Dauster ES, Cinquin BP, Smith EA, Robellet X, et al. The yeast polo kinase Cdc5 regulates the shape of the mitotic nucleus. Curr Biol. 2014; 24: 2861–7. pmid:25454593.
- 69. Zhang S, Xie M, Ren G, Yu B. CDC5, a DNA binding protein, positively regulates posttranscriptional processing and/or transcription of primary microRNA transcripts. Proc Natl Acad Sci USA. 2013; 110:17588–17593. pmid:24101471.
- 70. Yao N, Coryell L, Zhang D, Georgescu RE, Finkelstein J, Coman MM, et al. Replication factor C clamp loader subunit arrangement within the circular pentamer and its attachment points to proliferating cell nuclear antigen. J Biol Chem. 2003; 278: 50744–53. pmid:14530260.
- 71. Giavara S, Kosmidou E, Hande MP, Bianchi ME, Morgan A, d’Adda di Fagagna F, et al. Yeast Nhp6A/B and mammalian Hmgb1 facilitate the maintenance of genome stability. Curr Biol. 2005; 15:68–72. pmid:15649368.
- 72. Kassavetis GA, Steiner DF. Nhp6 is a transcriptional initiation fidelity factor for RNA polymerase III transcription in vitro and in vivo. J Biol Chem. 2006; 281: 7445–51. pmid:16407207.
- 73. Paull TT, Carey M, Johnson RC. Yeast HMG proteins NHP6A/B potentiate promoter-specific transcriptional activation in vivo and assembly of preinitiation complexes in vitro. Genes Dev. 1996; 10: 2769–81. pmid:8946917.
- 74. Tresini M, Warmerdam DO, Kolovos P, Snijder L, Vrouwe MG, Demmers JA, et al. The core spliceosome as target and effector of non-canonical ATM signalling. Nature. 2015; 523: 53–8. pmid:26106861.
- 75. Yasuhara T, Kato R, Hagiwara Y, Shiotani B, Yamauchi M, Nakada S, et al. Human Rad52 promotes xpg-mediated r-loop processing to initiate transcription-associated homologous recombination repair. Cell. 2018; 175: 558–570.e11. pmid:30245011.
- 76. Symington LS, Gautier J. Double-strand break end resection and repair pathway choice. Annu Rev Genet. 2011; 45: 247–71. pmid:21910633.
- 77. Rachidi N, Martinez MJ, Barre P, Blondin B. Saccharomyces cerevisiae PAU genes are induced by anaerobiosis. Mol Microbiol. 2000; 35: 1421–30. pmid:10760143.
- 78. Viswanathan M, Muthukumar G, Cong YS, Lenard J. Seripauperins of Saccharomyces cerevisiae: a new multigene family encoding serine-poor relatives of serine-rich proteins. Gene. 1994; 148: 149–53. pmid:7926827.
- 79. Cyr DM, Langer T, Douglas MG. DnaJ-like proteins: molecular chaperones and specific regulators of Hsp70. Trends Biochem Sci. 1994; 19:176–81. pmid:8016869.