(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



A protein–protein interaction map reveals that the Coxiella burnetii effector CirB inhibits host proteasome activity [1]

['Mengjiao Fu', 'State Key Laboratory Of Pathogen', 'Biosecurity', 'Beijing Institute Of Microbiology', 'Epidemiology', 'Academy Of Military Medicine Sciences', 'Fengtai', 'Beijing China', 'Yuchen Liu', 'State Key Laboratory Of Proteomics']

Date: 2022-08 

Coxiella burnetii is the etiological agent of the zoonotic disease Q fever, which is featured by its ability to replicate in acid vacuoles resembling the lysosomal network. One key virulence determinant of C. burnetii is the Dot/Icm system that transfers more than 150 effector proteins into host cells. These effectors function to construct the lysosome-like compartment permissive for bacterial replication, but the functions of most of these effectors remain elusive. In this study, we used an affinity tag purification mass spectrometry (AP-MS) approach to generate a C. burnetii-human protein-protein interaction (PPI) map involving 53 C. burnetii effectors and 3480 host proteins. This PPI map revealed that the C. burnetii effector CBU0425 (designated CirB) interacts with most subunits of the 20S core proteasome. We found that ectopically expressed CirB inhibits hydrolytic activity of the proteasome. In addition, overexpression of CirB in C. burnetii caused dramatic inhibition of proteasome activity in host cells, while knocking down CirB expression alleviated such inhibitory effects. Moreover, we showed that a region of CirB that spans residues 91–120 binds to the proteasome subunit PSMB5 (beta 5). Finally, PSMB5 knockdown promotes C. burnetii virulence, highlighting the importance of proteasome activity modulation during the course of C. burnetii infection.

As the causative agent of Q fever, C. burnetii colonizes host cells by transferring effector proteins into the host cytoplasm through its Dot/Icm secretion system to construct a replicative vacuole. The function of effectors remains largely unknown. Here, we performed a large-scale AP-MS screen to analyze the interactions among C. burnetii effectors and human proteins. These analyses found that CirB functions as an inhibitor of host proteasome activity, revealing that proteasome activity is important for intracellular survival of C. burnetii. Our data have laid the foundation for future exploring the molecular mechanisms underlying the roles of C. burnetii effectors in its virulence and for the identification of novel potential drug targets for the development of novel therapeutic treatment for C. burnetii infection.

In this study, using a combination of AP-MS and bioinformatics analysis, we identified PPIs between C. burnetii T4SS effectors and host proteins. Furthermore, we revealed that CBU0425 (designated Coxiella effector for intracellular replication B, CirB [ 6 ]) interacted with PSMB5 to inhibit the hydrolytic activity of the proteasome. This C. burnetii-human PPI map has advanced our understanding of the complex interactions between C. burnetii and its host, and the role of proteasomes in C. burnetii infection was first revealed.

The proteasome is one of the most important components for protein degradation in host cells. The activity of this abundant protein complex is tightly regulated and essential for the turnover of host proteins in the cytoplasm and nucleus [ 28 ] and has been shown to degrade thousands of short-lived and regulatory proteins, as well as damaged and misfolded proteins, to regulate various cellular functions [ 29 – 32 ]. The 20S core particle of the proteasome, which is composed of four stacked heptameric rings formed by α 7 β 7 β 7 α 7 subunits, is a barrel-shaped cylinder in which protein degradation occurs [ 33 ]. The proteolytically active subunits β1 (PSMB6), β2 (PSMB7), and β5 (PSMB5) harbor caspase-like, trypsin-like, and chymotrypsin-like catalytic activities, respectively [ 34 ]. In pathogen-invaded host cells, proteasomes cleave viral or bacterial antigens to generate peptides for MHC class I presentation, facilitating the clearance of infected cells [ 35 ]. Accordingly, pathogens have evolved numerous strategies to hijack proteasomes and ensure their intracellular replication. For example, L. pneumophila exploits host proteasomes to target unnecessary bacterial proteins for degradation by expressing ubiquitin ligases [ 36 , 37 ]. However, the involvement of the proteasome in C. burnetii infection has not yet been reported.

Since the functions of most identified C. burnetii effector proteins have not been elucidated, identifying the interaction between C. burnetii effectors and host proteins will provide important leads for our understanding of C. burnetii pathogenesis and for combating infections. Previously, yeast two-hybrid (Y2H) assays had been performed to screen host proteins interacting with C. burnetii effectors [ 19 ]. However, the PPIs identified in this screen only covered approximately 20% of the reported C. burnetii effectors, and the interactions between the other effectors and host proteins remain to be identified. Among the various high-throughput experimental methods available, mass spectrometry (MS) is a powerful approach with high coverage and accuracy for identifying new pathogen-host PPIs and for elucidating the mechanisms of pathogen replication, uncovering new functions of pathogen proteins, and revealing host pathways involved in infection [ 20 – 22 ]. Several host pathways required for a number of important pathogens including HIV [ 23 ], Zika virus [ 24 ], severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) [ 25 ], Chlamydia trachomatis [ 26 ], and Mycobacterium tuberculosis [ 27 ], have been systematically identified using affinity purification-mass spectrometry (AP-MS). Similarly, studies aiming to explore the mechanisms by which C. burnetii effectors manipulate host cells using this unbiased method are necessary and meaningful.

Coxiella burnetii is a Gram-negative bacterium responsible for the worldwide zoonotic disease Q fever, which manifests with acute or chronic symptoms. The Dot/Icm type IV secretion system (T4SS) is considered a key virulence determinant of C. burnetii, which transfers effector proteins into host cells to promote bacterial survival and replication [ 1 , 2 ]. To date, more than 150 C. burnetii effectors have been identified using Legionella pneumophila as a surrogate [ 2 – 6 ]. Notably, the C. burnetii T4SS effector AnkG inhibits host cell apoptosis by interacting with the host proteins p32, importin-α1 and RNA helicase 21 [ 7 , 8 ]. In addition to AnkG, the effectors CaeA and CaeB prevent mitochondrial-dependent (intrinsic) apoptosis [ 9 ]. Three effectors, CetCb2, CetCb4, and Cem9, modulate the mitogen-activated protein kinase (MAPK) pathway in yeast [ 10 ]. Three effectors, including Coxiella vacuolar proteins (Cvp) A, CvpB, and CvpF, locate on the membrane of Coxiella-containing vacuoles (CCVs) and promote intracellular replication by interacting with host targets through distinct mechanisms [ 11 – 14 ]. Additionally, the effectors Cig57 and CirA subvert membrane trafficking through interactions with FCHO2 and Rho GTPases, respectively [ 15 , 16 ]. In recent studies, NopA and CinF have been reported to inhibit the innate immune signaling pathway by perturbing the nuclear import of transcription factors and dephosphorylating IκB, respectively [ 17 , 18 ]. The diverse host pathways involved in C. burnetii infection have revealed that, in addition to the most widely studied autophagy-lysosome and apoptosis-related pathways, many previously understudied pathways in host cells may play important roles in C. burnetii pathogenesis.

Results

Generation of a C. burnetii-human PPI network To date, more than 150 effector proteins of C. burnetii have been identified, and some of these effectors have been reported to manipulate diverse host signaling pathways to promote C. burnetii replication, but the functions of the majority of these effectors remain elusive [1,2,38]. To further explore the biological functions of these effectors, an AP-MS approach was used to identify potential physical interactions between C. burnetii effectors and host proteins. Eighty genes encoding effectors from the C. burnetii Nine Mile phase II (NMII) strain identified from 2010 to 2018 [3–6,12,39–42] were selected and individually cloned into a pQM02 vector with an N-terminal mCherry and C-terminal twin-Strep tag (pQM02-cbu), and 53 of them were expressed at high levels in human cervical cancer (HeLa) cells (Fig 1A). These ectopically expressed effectors showed different subcellular localizations, with some colocalizing with LAMP1-positive vesicles (S1A Fig). To perform the AP-MS, effector proteins were enriched in human embryonic kidney (HEK-293T) cells with Strep-Tactin Sepharose beads and identified by liquid chromatography with tandem mass spectrometry (LC-MS/MS) (Fig 1B). The results were then processed using the method shown in Fig 1C. Raw data were searched using Proteome Discoverer, and the Pearson’s correlation coefficients (PCC) between three biological replicates of each effector was calculated separately. Only data from two replicates with the highest PCC were used for further analysis. Furthermore, only the data from the replicates with a PCC greater than 0.6 were used for interaction identification. Effectors with a low correlation or with a low identified protein number were rid of for MS samples. Thus, the final 106 raw data of 53 effectors were selected from 216 runs, including controls. The number of proteins identified in the mCherry-Strep controls ranged from 17 to 640, with a median of 83. To obtain a high confidence PPI network, a high-quality negative control is needed. Two highest spectral counts of all 28 HEK-293T cell experiments in the CRAPome database were used as controls. Therefore, 53 effectors and 2 CRAPome controls were first scored with the SAINTexpress algorithm. Host interactions with a Bayesian false discovery rate (BFDR) less than 0.01 (in 19,761 potential interactions) were subjected to further filtering. Then, the proteins that appeared most frequently in all human samples (443 proteins occurring in the top 5th percentile, S1 Table) were arbitrarily excluded. All these proteins appeared in at least 87 experiments in all 343 datasets from CRAPome. The C. burnetii-human PPI network contained 17,831 PPIs consisting of 53 T4SS effectors and 3,480 host proteins. The statistical figure of the total protein number identified by MS and the PPI number involving each effector protein were shown in Fig 1D. The effectors were classified according to the subcellular localization we identified in this study or reported previously [5,7,43]. The PCC of biological replicates of each effector ranged from 0.66 to 0.97 (S1B Fig). The number of interactions ranged from 3 to 1,955. Then, we selected 26 PPIs of interest derived from AP-MS experiments for further study and verified 13 of them in a co-immunoprecipitation (co-IP) assay (S2 Fig). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. AP-MS analysis of C. burnetii effectors. (A) PQM02-cbu plasmids were transfected into HeLa cells, and the expression of mCherry-Strep-tagged effectors was observed under a fluorescence microscope (magnification, ×600, bar = 20 μm). (B-C) Schematic diagram of affinity tag purification (B) and MS data analysis (C). (D) Number of host proteins and PPIs that identified by the AP-MS approach based on the subcellular location of effectors. (E) The Jaccard index of C. burnetii effectors. https://doi.org/10.1371/journal.ppat.1010660.g001 The Jaccard coefficients of C. burnetii effectors were calculated to evaluate the similarity between host proteins (Fig 1E), and the results showed that effectors could be classified into 5 clusters according to the similarity of their binding partners. However, we found that the clusters of these effectors did not exactly match the subcellular localization previously reported. This difference may have been a result of the protein expression profiles of different cell lines or a considerable number of cross-subcellular localization of protein interactions. Another potential explanation for this difference is the nonbiologically relevant interactions between proteins from different cellular compartments in lysed cells [44]. Nevertheless, the PPI network would provide a comprehensive reference for the in-depth study of uncharacterized effectors.

Systematic analysis of the protein interaction map To further define the function of these C. burnetii effectors, Gene Ontology and pathway enrichment analyses (Reactome) of each effector were performed separately. Cell component annotations of host interactors were noted first. As shown in Fig 2A, the endoplasmic reticulum, lysosome, nucleus, ribosome and vesicle were the main organelles in which these effectors were localized. As expected, the significantly enriched terms of host interactors were autophagy, proteasome and vesicle-related (Fig 2B). The effector proteins (CBU0021, CBU0937, CBU1823, CBU1825 and CBU2013) were mainly associated with autophagy. Protein catabolism mediated by the ubiquitin-dependent proteasome was another enriched term that was hit by more than 15 C. burnetii effectors. Additionally, some of these C. burnetii effectors exhibited a clear preference for the process involved in vesicle trafficking. Signaling pathways, mainly the MAPK, NF-κB and AKT pathways, were also enriched (Fig 2C). Surprisingly, the noncanonical, but not the canonical, NF-κB pathway was noticeably enriched, suggesting a potential role for certain effectors in modulating host inflammatory response through this signaling pathway. In addition, the main interactions enriched in functional pathways, such as vesicles (S3A Fig), NF-κB pathway (S3B Fig) and apoptosis (S3C Fig), were also analyzed. Among the enriched effectors, CirB (CBU0425) was mainly enriched in the process of proteasome-mediated degradation, differing from effectors that play multiple roles, as indicated by their enrichment in the cellular component and biological process categories. Furthermore, the interaction map revealed that CirB interacts with the majority subunits of the 20S core proteasome (Fig 2D), suggesting its role in modulating proteasome function. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Systematic analyses of the C. burnetii-host PPI network. (A-C) Functional characterization of clustered PPIs. (A) Cellular components, (B) Biological processes, (C) Reactome pathways. (D) Interactions between CirB and host proteins. https://doi.org/10.1371/journal.ppat.1010660.g002

Ectopically expressed CirB inhibits the activity of the host proteasome Although C. burnetii has been reported to manipulate diverse host processes and signaling pathways, no information showing the relationship between C. burnetii and the host proteasome has been reported. In our study, we first found that the hydrolytic activity of the proteasome decreased during C. burnetii infection, while cell viability was not significantly affected (Figs 3A and S4A), and the expression level of IκBα, which was mainly degraded by the proteasome, was significantly higher than that in the uninfected group (Fig 3B). As revealed by the PPI network, the proteasome subunits were targeted by several C. burnetii effectors; hence, we wondered whether any of these effectors was involved in inhibiting proteasome activity during infection. As shown in Fig 3C, only ectopically expressed CirB significantly inhibited proteasome activity, and this inhibitory effect presented a dose-dependent manner in the case of CBU1751 as a control (Figs 3D and S4B and S4C). In THP1-CirB (CirB overexpressing) cells, proteasome activity was also significantly lower than that in control cells (Fig 3E). To investigate proteasome activity in vitro, proteasomes of cells overexpressing CirB or a control plasmid were purified and subjected to a functional assay using Suc-LLVY-AMC as a substrate. Whereas proteasomes isolated from control cells possessed high hydrolytic activity (S4D Fig), proteasomes similarly purified from cells overexpressing CirB exhibited significantly lower activity (Fig 3F). Moreover, GST-CirB purified from E. coli also inhibited substrate degradation by the 20S proteasome (Fig 3G). Collectively, these results indicated that CirB inhibits the activity of the host proteasome. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. CirB inhibits proteasome activity. (A) THP-1 cells were differentiated with PMA (200 nM) in a 96-well plate and infected with NMII strains at different MOIs or left uninfected. At two days post-infection, proteasome activity was measured using a proteasome-Glo cell-based assay. (B) Differentiated THP-1 cells were infected with the NMII strain at an MOI of 100. At the indicated time points post-infection, cells were lysed and the expression of IκBα, Tubulin and Com1 was detected using Western blotting with the indicated antibodies. A sample collected from cells treated with MG132 (10 μM, 8 hours) on day 2 post-infection served as a control. (C) Plasmids expressing the indicated effector proteins were transfected into HeLa cells. Twenty-four hours later, the proteasome activity of the cells was measured as described above. The cells treated with bortezomib (10 μM, 8 hours) and MG132 (10 μM, 8 hours) served as positive controls. (D) Different amount of CirB, 100 ng of CBU1751, or 100 ng of the control plasmid was transfected into HeLa cells, respectively. Then, the proteasome activity was measured 24 hours later. (E) Normal THP-1, THP1-NC or THP1-CirB cells were differentiated with 200 nM PMA. Proteasome activity of the cells were measured 48 hours later. (F) PQM02-CirB or control plasmids were transfected into 293T or HeLa cells, proteasomes were enriched from cells, and the hydrolytic activity of proteasomes degrading the Suc-LLVY-AMC substrate was detected in vitro. (G) The 1mM of GST-CirB or GST proteins expressed in prokaryotes was added to the system containing 20S proteasome, 0.1% SDS and Suc-LLVY-AMC substrate in Tris-HCl buffer. The fluorescence was measured at an excitation wavelength of 390 nm and an emission wavelength of 460 nm. Data are representative of three independent experiments and bars represent the mean ± SD. *, p < 0.05, **, p < 0.01, and ***, p < 0.001. https://doi.org/10.1371/journal.ppat.1010660.g003

CirB is involved in inhibiting proteasome activity during C. burnetii infection CirB has been shown to be a T4SS effector of C. burnetii based on the fact that it can be translocated into host cells by L. pneumophila or C. burnetii in a Dot/Icm-dependent manner [5,6]. In this study, we validated the translocation of CirB by C. burnetii by performing a β-lactamase-based translocation assay (S5A and S5B Fig). To further explore the role of CirB in modulating proteasome activity during C. burnetii infection, we first attempted to generate a cirB deletion mutant through homologous recombination [45] but were unable to obtain such mutants despite extensive efforts. Therefore, we sought to employ alternative strategies to alter the expression of CirB in C. burnetii and detect the subsequent effects of this altered expression on host proteasome activity. The cirB gene was cloned into a pJB-Kan-3×FLAG plasmid and introduced into the NMII strain via electroporation to generate a strain that stably overexpressing CirB (NMIIpJB-CirB, Fig 4A). The growth characteristic of NMIIpJB-CirB in acidified citrate cysteine medium-2 (ACCM-2) was not different from that of the control strain NMIIpJB (S5C Fig), while NMIIpJB-CirB showed better intracellular replication in infected THP-1 cells (Fig 4B). Regarding proteasome activity, NMIIpJB-CirB infection caused a detectable inhibition of proteasome activity compared to strain NMIIpJB (Fig 4C). Meanwhile, the degradation of IκBα in cells infected with NMIIpJB-CirB was also inhibited due to the overexpression of CirB (Fig 4D). Using a modified CRISPRi system (S5D Fig), we successfully constructed strain NMIIpdCas9-sgcirB, and the mRNA levels of dcas9, cirB and single guide RNA (sgRNA) targeting cirB were verified using qRT-PCR. As shown in Fig 4E, the expression of dcas9 and sgRNA was highly induced and the mRNA level of cirB was decreased by approximately 90% compared to that of the control strains NMII and NMIIpdCas9. Moreover, CirB expression in THP-1 cells infected with NMIIpdCas9-sgcirB was almost completely suppressed, as determined by Western blotting (Fig 4F). The growth of NMIIpdCas9-sgcirB in THP-1 cells or ACCM-2 was not significantly different from that of the control strains (Figs 4G and S5E), but NMIIpdCas9-sgcirB infection partially alleviated the inhibition of proteasome activity induced by C. burnetii infection (Fig 4H). The primers used for detection are listed in S5F Fig. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Overexpression or knockdown of CirB in NMII affects proteasome activity. (A) Differentiated THP-1 cells were infected with NMII, NMIIpJB or NMIIpJB-CirB at an MOI of 100. Three days later, cells were lysed, and the expression of CirB or Com1 was examined using Western blotting with the indicated antibodies. (B) Differentiated THP-1 cells were infected with NMIIpJB or NMIIpJB-CirB at an MOI of 20. At different time points post-infection, the total DNA was extracted, and the GE of C. burnetii was quantitated using qPCR. Fold change in GE was calculated by comparing the GE at different days post-infection to day 0. Data are representative of three independent experiments and bars represent the mean ± SD, **, p < 0.01. (C) Differentiated THP-1 cells were infected with the indicated C. burnetii strains at an MOI of 100 for 2 days. Then, proteasome activity of the infected cells was measured. (D) Differentiated THP-1 cells were infected with NMIIpJB or NMIIpJB-CirB at an MOI of 100. At different times post-infection, the expression of IκBα and CirB were examined using the indicated antibodies. The expression of Tubulin and Com1 was used as the internal control for host cells and C. burnetii respectively. (E) Differentiated THP-1 cells were infected with NMII, NMIIpdCas9 or NMIIpdCas9-sgcirB at an MOI of 100. Two days post-infection, total RNA was extracted and the mRNA levels of sgcirB, dcas9 and cirB were quantitated using qRT-PCR. The relative gene expression was calculated as follows: the mRNA levels of sgcirB, dcas9 and cirB in NMIIpdCas9-or NMIIpdCas9-sgcirB-infected cells/that in NMII-infected cells. The expression of rpoB was used as an internal control. (F) Differentiated THP-1 cells were infected with NMII, NMIIpJB-dCas9 or NMIIpdCas9-sgcirB at an MOI of 100. Two days later, cells were lysed and the expression of CirB, dCas9 and DotB were examined using Western blotting with the indicated antibodies. (G) Differentiated THP-1 cells were infected with NMIIpdCas9 or NMIIpdCas9-sgcirB at an MOI of 20. At different time points post-infection, the total DNA was extracted and the GE of C. burnetii was quantitated using qPCR. (H) Differentiated THP-1 cells were infected with different strains at an MOI of 100. Two days later, the proteasome activity of infected cells was measured. Data are representative of three independent experiments and bars represent the mean ± SD. *, p < 0.05, **, p < 0.01, and ***, p < 0.001. https://doi.org/10.1371/journal.ppat.1010660.g004

CirB interacts with PSMB5 CirB was identified to interact with several functional subunits of the human 20S proteasome, including PSMB5, which displays a chymotrypsin-like activity. We therefore focused on the interaction of CirB and PSMB5 and sought to identify the key domain critical for this interaction. HIS-PSMB5 could be pulled down with GST-CirB expressed in prokaryotes in vitro (Fig 5A), and overexpressed CirB also interacted with endogenous PSMB5 in HeLa cells (Fig 5B), which confirmed the interaction between these proteins. Furthermore, in NMII-infected cells, overexpressed CirB colocalized with endogenous PSMB5 with another two effector proteins served as controls (Fig 5C). Because CirB lacks an obvious known PPI domain or motif, we constructed truncation mutants of CirB and PSMB5 to determine the key regions responsible for their interaction. CirB interacted with all PSMB5 truncation mutants, except for PSMB5Δ3, while the binding between CirB and PSMB5Δ4 or Δ5 was weaker, indicating that PSMB5 interacts with CirB through both amino acid regions 89–176 and 177–263 (S6A and S6B Fig). Simultaneously, amino acids 1–152 of CirB interacted with PSMB5, which was verified by co-IP and laser confocal microscopy (Figs 5D and S6C). We further eliminated portions of the CirB amino acid 1–152 sequence to generate 6 truncation mutants and ultimately identified CirB amino acids 91–120 as the key region mediating the interaction of CirB with PSMB5 (Figs 5E, 5F and S6D). As proline-and arginine-rich peptides have been reported to be flexible allosteric modulators of the proteasome and to exert the inhibitory effects [46,47], we replaced the RRRP sequence (amino acid residues 111–114 of CirB) in the CirBΔ10 truncation mutant to GGGA and thus generated CirBΔ10m. As shown in Fig 5G and 5H, the interaction of PSMB5 and CirB Δ10 was not abolished but was reduced when the RRRP sequence was substituted. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. CirB interacts with the host protein PSMB5. (A) pGEX-CirB, pGEX-GST or pET32a-PSMB5 was separately expressed in E. coli BL21. The purified GST-CirB or GST was co-incubated with GST Sepharose beads, then HIS-PSMB5 was subjected to the GST-conjugated beads. After eluting, samples were loaded for Western blotting. (B) HEK-293T cells were transfected with pQM02-CirB or the corresponding control plasmid. Then, the cells were lysed and proteins were immunoprecipitated with an anti-Strep antibody, and the intracellular PSMB5 was immunoblotted with an anti-PSMB5 antibody. (C) HeLa cells were transfected with mCherry-tagged CirB, CBU1780 or CBU1387 before C. burnetii infection at an MOI of 100, respectively. Two days later, cells were fixed, and endogenous PSMB5 was probed with an anti-PSMB5 and a corresponding goat anti-rabbit Alexa Fluor 488 antibody (green). C. burnetii was stained with anti-C. burnetii serum, followed by a Cy5 conjugated goat anti-mouse IgG H&L antibody (white). The nucleus was stained with DAPI (blue). Images were captured using a confocal microscope (magnification, ×600, bar = 20 μm). (D-E) FLAG-PSMB5 and HA-tagged CirB truncation mutants were co-transfected into 293T cells, and the cell lysates were immunoprecipitated with an anti-FLAG antibody and immunoblotted with an anti-HA or anti-FLAG antibody. Schematic diagram of the truncation mutants of CirB is shown in the upper panel. Sections marked in red represent the truncation mutants responsible for the interaction. (F) Schematic diagram showing the regions of PSMB5 and CirB interaction. (G) The diagram of the mutation sites in CirBΔ10m. (H) PRK5-PSMB5 and pCMV-CirBΔ10, pCMV-CirBΔ10m or pCMV-HA were co-transfected into 293T cells, and the cell lysates were immunoprecipitated with anti-FLAG antibody and immunoblotted with anti-FLAG or anti-HA antibody. https://doi.org/10.1371/journal.ppat.1010660.g005

A region that spans residues 91 to 120 of CirB is sufficient to inhibit proteasome activity To determine the CirB region that mainly exerted the inhibitory effect on proteasome activity, CirB and its truncation mutants were individually transfected into HeLa cells, and proteasome activity was assessed at 24 hours post-transfection. CirBΔ5 and CirBΔ10 exerted a significant inhibitory effect on proteasome activity (Fig 6A), and their overlapping sequence was amino acids 91–120 of CirB, which was also the region required for PSMB5 binding, as explained above. The predicted structure of CirB obtained by AlphaFold2 [48] was visualized using Web3dmodel (https://web3dmol.net/). As shown in Fig 6B, the 91–120 amino acid region of CirB, which is critical for the interaction of CirB and PSMB5, is marked in green. In the predicted structure, this region appears to fold independent of other regions of the protein, suggesting that residues 91–120 of CirB are important for CirB function. We thus synthesized a peptide of CirB 91-120aa (designated CirB-30aa) and measured its impact on proteasome activity. In Tris-HCl buffer, CirB-30aa significantly inhibited 20S proteasome-mediated substrate degradation in a dose-dependent manner (Fig 6C). We next hypothesized that the proline and arginine residues in CirB-30aa might be the key residues driving the inhibitory effects of CirB-30aa. A CirB-30aa mutant peptide (in which RRRP was replaced with GGGA, designated CirB-30aa-mut) and a 30 aa peptide of CBU1751 88-113aa (designated CBU1751-30aa) were synthesized and used as controls to confirm the inhibitory effect of CirB-30aa (S6E Fig). The results of the verification experiment showed that CirB-30aa exerted an inhibitory effect on proteasome activity in a dose- and time-dependent manner, while CirB-30aa-mut and CBU1751-30aa failed to affect proteasome activity (Fig 6D). To determine whether CirB-30aa functioned as an allosteric inhibitor of 20S proteasome activity, atomic force microscopy (AFM) imaging was used to detect conformational changes of the 20S proteasome with or without exposure to CirB-30aa. Purified 20S proteasome exhibited a cylindrical shape with an average width of 13.81 nm in the field of 300×300nm, and the addition of CirB-30aa clearly caused an increase in the 20S proteasome size, with the width of the proteasome increasing to 21.52 nm (Fig 6E and 6F). However, CirB-30aa-mut exerted no obvious effect on the size of the 20S proteasome. These results suggested that the 91–120 aa of CirB is the functional region critical for CirB-mediated inhibition of proteasome activity. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 6. The 91–120 aa region is the key domain of CirB that inhibits proteasome activity. (A) Plasmids expressing CirB or CirB truncation mutants, as well as the control plasmid, were transfected into HeLa cells. Twenty-four hours later, the proteasome activity of cells was measured. Data are representative of three independent experiments and bars represent the mean ± SD. **, p < 0.01, and ***, p < 0.001. (B) The predicted 3D protein structure of CirB protein. The sequence marked in green was the 91–120 amino acid region of CirB, and the amino acids “RRRP” were indicated with a red arrow. (C) Different concentrations of synthetic CirB-30aa were added to the mixture containing 20S proteasome, 0.1% SDS and Suc-LLVY-AMC substrate in Tris-HCl buffer. The fluorescence was measured at an excitation wavelength of 390 nm and an emission wavelength of 460 nm. (D) The effects of CirB-30aa, CirB-30aa-mut and CBU1751-30aa peptides on proteasome activity were examined, and the fluorescence was measured every 5 min. (E) The effects of CirB-30aa on proteasomes were observed with AFM. The purified and unmodified human 20S proteasomes and 20S proteasomes treated with CirB-30aa or CirB-30aa-mut were electrostatically attached to the mica substrate, and areas ranging from 0.09 (right panel) to 1 μm2 (left panel) were scanned. Upper panel: representative 3D images of a 1 μm× 1 μm field of 20S proteasomes in height mode. Lower panel: representative images of 20S proteasomes scanned using AFM. (F) The average size of the proteosome was altered by CirB-30aa. The width of purified and unmodified human 20S proteasomes and 20S proteasomes treated with CirB-30aa or CirB-30aa-mut was measured and calculated. n = 50. Data are representative of two independent experiments and bars represent the mean ± SD of 50 proteasomes. *, p < 0.05, and **, p < 0.01. https://doi.org/10.1371/journal.ppat.1010660.g006

[END]
---
[1] Url: https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1010660

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/