Proteasome Subunits Differentially Control Myeloma Cell Viability and Proteasome Inhibitor Sensitivity Chang-Xin Shi1, Yuan Xiao Zhu1, Laura A

Published OnlineFirst June 19, 2020; DOI: 10.1158/1541-7786.MCR-19-1026

MOLECULAR CANCER RESEARCH | CANCER GENES AND NETWORKS

Proteasome Subunits Differentially Control Myeloma Cell Viability and Proteasome Inhibitor Sensitivity Chang-Xin Shi1, Yuan Xiao Zhu1, Laura A. Bruins1, Cecilia Bonolo de Campos1, William Stewart1, Esteban Braggio1, and A. Keith Stewart1,2

ABSTRACT ◥ We generated eight multiple myeloma cell lines resistant to drug-resistant multiple myeloma cell lines by low level expression bortezomib; five acquired PSMB5 mutations. In 1,500 patients such of mutated PSMB6 lacking splicing function. Loss of PSMB8 mutations were rare clinically. To better understand disruption of and PSMB9 was neither lethal nor restored bortezomib sensitiv- proteasomes on multiple myeloma viability and drug sensitivity, we ity. Significant codependency of PSMB5, PSMB6, and PSMB7 systematically deleted the major proteasome catalytic subunits. expression was observed. We demonstrated elevated levels of Multiple myeloma cells without PSMB5 were viable. Drug-resistant, PSMB6 and 7, but not 8 and 9, in some, but not all, serial patient PSMB5-mutated cell lines were resensitized to bortezomib by samples exposed to proteasome inhibitors. In summary, we PSMB5 deletion, implying PSMB5 mutation is activating in its show PSMB6 and PSMB7, but not PSMB5, to be essential for drug resistance function. In contrast, PSMB6 knockout was lethal to multiple myeloma cell survival, this dependency is structural multiple myeloma cell lines. Depleting PSMB6 prevented splicing of and that upregulation or activating mutation of PSMB5, 6, and 7 the major catalytic subunits PSMB5, PSMB7, PSMB8, and PSMB10; confers proteasome inhibitor resistance, while depletion confers however, PSMB6 engineered without splicing function or catalytic sensitivity. activity, also restored viability, inferring the contribution of PSMB6 to proteasome structure to be more important than functional Implications: These findings support modulation of PSMB5, activity. Supporting this, bortezomib sensitivity was restored in PSMB6, or PSMB7 expression as a new therapeutic strategy.

Introduction b-subunits, as well their immunocounterpart are expressed as a proprotein, whose catalytic functions are protected by a propeptide. Multiple myeloma is a plasma cell malignancy characterized by These subunits are activated upon removing their propeptide by the accumulation of plasma cells in the bone marrow (1). Treatment autocatalytic processing (14). Bortezomib directly inhibits the func- outcomes have improved significantly following the introduction tional activity of PSMB5, PSMB6, PSMB8, and PSMB9 (15). It has long of the proteasome inhibitors, bortezomib, carfilzomib, and ixazo- been believed that PSMB5, and not PSMB6 and PSMB7, plays an mib, into routine clinical practice (2–7). Proteasome inhibitors used important role in cell viability, because a PSMB5 p.Thr1Ala muta- in combination with other compounds, such as immunomodulatory tion is lethal in yeast (16–19). However, subsequent publications drugs and steroids achieve an initial response in 90%–100% of have demonstrated that significant inhibition of protein degrada- patients (8–10). Despite the initial response, almost all patients tion can only be achieved when both, PSMB5 and either PSMB6 or eventually relapse and become refractory to multiple anti-multiple PSMB7, sites are inhibited (20, 21). Furthermore, proteasome myeloma therapies, including proteasome inhibitors (11). inhibitor cytotoxicity does not correlate with inhibition of chymo- The proteasome complex 26S consists of the 20S proteolytic core trypsin-like sites alone, and the coinhibition of either PSMB6 or and one or two 19S regulatory complexes, which function as ubiquitin PSMB7 sites are required for PSMB5-specific inhibitors to achieve receptors to recognize substrates, deubiquitinate enzymes, and facil- maximum cytotoxicity (22). itate substrate unfolding and translocation into the 20S proteasome. Recently, we employed CRISPR targeting of 19,052 human genes to The 20S core has three b subunits that catalyze peptide hydrolysis, identify unbiased targets that contribute to bortezomib resistance (23). PSMB5 (b5), PSMB6 (b1), and PSMB7 (b2), with a chymotrypsin, Of 20 targets conferring drug resistance, the proteasome regulatory caspase, and trypsin-like activity, respectively (12). Upon exposure to subunit, PSMC6, was the only validated gene to reproducibly confer IFNg, these three subunits are replaced by their corresponding homol- bortezomib resistance in our experimental model. We confirmed that ogous, the immunoproteasome subunits PSMB8 (b5i), PSMB9 (b1i), inhibition of the chymotrypsin-like proteasome activity by bortezomib and PSMB10 (b2i), respectively (13). These three constitutive catalytic was significantly reduced in cells lacking PSMC6. We investigated other members of the PSMC group individually (PSMC1–5) and found fi 1Department of Hematology, Mayo Clinic in Arizona, Scottsdale, Arizona. 2Center that de ciency in each of those subunits also imparts bortezomib for Individualized Medicine, Mayo Clinic, Rochester, Minnesota. resistance. More recently, we have demonstrated an increased incidence Note: Supplementary data for this article are available at Molecular Cancer Research Online (http://mcr.aacrjournals.org/). of acquired proteasomal subunit mutations in relapsed multiple myeloma compared with newly diagnosed disease, underpinning Corresponding Author: A. Keith Stewart, Mayo Clinic in Arizona, 13400 East Shea Blvd, Scottsdale, AZ 85259. Phone: 507-284-5511; Fax: 507-293-2379; a potential role of point mutations in the clonal evolution of E-mail: [email protected] multiple myeloma, including mutations in 19S proteasome subunits in 4% of patients (N ¼ 895, IA10 release of the CoMMpass study; Mol Cancer Res 2020;XX:XX–XX https://themmrf.org). doi: 10.1158/1541-7786.MCR-19-1026 Furthermore, we functionally characterized four somatic PSMB5 2020 American Association for Cancer Research. mutations (three of them affecting the proteasome inhibitor–binding

AACRJournals.org | OF1

Shi et al.

pocket) in primary multiple myeloma cells identified in a patient under phosphate method was used for packaging lentiviral vectors (28, 29). prolonged proteasome inhibition (24). Nevertheless, PSMB5 muta- Briefly, 293T cells were split into 100 mm tissue culture dishes tions are very uncommon even in relapsed multiple myeloma. Because (15 mL total volume medium, no more than 70% confluent before subunit mutations appears to explain only a fraction of proteasome transformation). Transformation was performed 24 hours later. The inhibitor resistance seen in patients, we were interested in further 2 mL calcium phosphate-DNA coprecipitation solution was made exploring the role of proteasome subunit mutation and its role in with 100 mL2.5mol/LCaCl2 and plasmid DNA (18.34 mgfor bortezomib sensitivity and resistance. pSPAX2, 7.34 mg for pMD2.G, and 24.34 mg for lentivector), diluted with 1/10 TE, pH 8.0 to the final total volume of 1 mL. Next, one volume of 2 Ca/DNA solution was added to one volume of 2 Materials and Methods HEPES. After 1 minute, 1.5 mL of the mixed solution was added to Cell lines the cell culture medium. The next morning the medium was Eight human multiple myeloma cell lines expressing Cas9 were used changed with 12 mL fresh DMEM and 10% FBS. The viruses were in this study: RPMI-8226, FR4, U266 and JJN3(23), KMS12PE, XG-1, harvested 20 hours after medium change. For infection of multiple MM1.S, and OCI-MY5, which were obtained from Dr. Leif Bergsagel’s myeloma cell lines, 1 106 cells were added into 6-well plates with 6 laboratory (Mayo Clinic, Scottsdale, AZ) from 2014 to 2016. An initial mL polybrene (5 mg/mL), 1.5 mL lentivirus, and fresh medium to genetic analysis of these lines (copy-number variation analysis) estab- the total volume of 3 mL. Four hours later, 3 mL of fresh medium lished a baseline, identifying fingerprint (developed by Drs. Leif was added. Next morning the RPMI medium was changed. Bergsagel and Jonathan Keats, Translational Genomics Research Institute, Phoenix, AZ). The identity of cell lines was confirmed using Construction of the PSMB5 and PSMB6 mutants copy-number variation analysis each time samples are removed from The wild-type PSMB5 was engineered to have a synonymous liquid nitrogen storage for propagation. All cell lines taken from liquid mutation in the sgRNA-targeted region (protospacer adjacent nitrogen were thawed and maintained in RPMI1640 media, supple- motif sequence CCA was mutated to TCA, which is required for mented with 5% of sterile FCS and antibiotics. All cell lines were sgRNA PSMB5CrA CAAGTCCGAAAAACCCGCGC targeting); two maintained for 4 to 8 weeks (less than 20 passages) before starting additional mutations in PSMB% were introduced (PSMBM1 and experiment and they were tested negative for Mycoplasma at the PSMB5M2). The PSMB5 mutation 1 has G-1A, T1A, K32A, beginning and during the experiments (using Mycoplasma Detection and K33A mutations, resulting in no self-splicing function or catalytic Kit from Lonza). Bortezomib resistance was developed by a stepwise activity; the PSMB5M2 mutation was similar to mutation 1 but increase in bortezomib concentration. with no propeptide, resulting in no catalytic activity. We also constructed a lentivector expressing wild-type PSMB6, in which the Targeted sequencing protospacer adjacent motif sequence TGG was mutated to AGG We performed DNA targeted sequencing in eight multiple myeloma (CCCGGAGCCTCCAATGGCAA) such that PSMB6 sgRNA 10 can- cell lines using Semiconductor-Sequencing Technology (IonTorrent not target it. In addition, two versions of mutated PSMB6 (M1 and M2) PGM) as per the manufacturer’s protocol (25). All coding regions of were constructed. The PSMB6 M1 has a 1G and 1T, which were the proteasome subunits were amplified in 200 bp amplicons by mutated to –1A and 1A, and a 33K, mutated to 33A. These mutations multiplex PCR using customized oligos (Ion AmpliSeq Designer). were directly synthesized by Integrated DNA Technologies and cloned Libraries were templated and enriched using IonOneTouch2 and IonO- into a lentivector. neTouch ES automated systems, respectively. Samples were sequenced using the 318 Chip (Thermo Fisher Scientific). Raw data were aligned Western blot analysis and indexed in BAM and BAI files using the IonTorrent suite. Variants We used a standard Western blot technique. Equal amounts of were called using IonTorrent Somatic VariantCaller and low stringency protein were subjected to SDS-PAGE gels followed by transfer to settings (Thermo Fisher Scientific). VCF files were annotated using a polyvinylidene difluoride membranes. Membranes were probed with standalone application developed by the Mayo Clinic Bioinformatics primary antibodies overnight and then washed and incubated with Program called BioR (Biological Annotation Data Repository; ref. 26). horseradish peroxidase–conjugated secondary antibodies. Detection Somatic variants with a Mapping Quality <20, read depth <10,or was performed by the enhanced chemical luminescence method. found in <1% of reads were removed. To remove germline mutations, When probing for multiple targets or validating equal amount of common variants were eliminated on the basis of the minor allele protein load, stripping and reprobing a single membrane were con- frequencies (>0.01%) available in one of the following germline variant ducted. In case, if protein reprobed was absent, the same samples were databases: 1000 Genomes Project, ExAC, and ESP6500, unless present usually rerun to validate the result. HA antibody (catalog no. MMS- in known CLL/MBL mutation hotspots or in Catalogue of Somatic 101P Clone 16B12) was from Covance. Antibodies against proteasome Mutations in Cancer. Finally, variants of significant interest were visually subunits PSMB5 (catalog no. BML-PW8895), PSMB6 (catalog no. inspected using Integrative Genomics Viewer (27). BML-PW8140), PSMB7 (catalog no. BML-PW8145), PSMB8 (catalog no. BML-PW8840), PSMB9 (catalog no. BML-PW8150), PSMB10 Single-guide RNAs construction, lentivector packaging, and (catalog no. BML-PW8350), and PSMA1 (a6) (catalog no. BML- infection PW8100) were purchased from ENZO Life Science. PSMB2 (b4) was All single-guide RNAs (sgRNA) were synthesized by Integrated purchased from Proteintech. DNA Technologies. sgRNAs were cloned into lentivector pCDHPuroCr, as described previously (23). sgRNA sequences are MTT assay shown in Supplementary Data. Packaging plasmids psPAX2 Cellular viability of the multiple myeloma cell lines was determined (Addgene plasmid # 12260) and pMD2.G (Addgene plasmid # by the MTT Assay (Sigma-Aldrich) following the manufacturer's 12259) were gifts from Dr. Didier Trono, Laboratory of Virology instructions. Briefly, cells were seeded at a concentration of 2 104 and Genetics, EPFL Switzerland. A modified Graham calcium– cells/well in 90 mL culture medium and 10 mL medium with drug was

OF2 Mol Cancer Res; 2020 MOLECULAR CANCER RESEARCH

Multiple Myeloma Cell Survival Depends on PSMB6 and PSMB7

þ added. cell cultures were incubated for 72 hours at 37 C and 5% CO2. above), we began a deeper analysis by CRISPR depleting PSMB5 in the After the incubation period, 10 mL of the MTT labeling reagent was RPMI-8226R cell line (Fig. 1B, top) using sgRNAs (Supplementary added (final concentration 0.5 mg/mL) to each well. The microplate Fig. S2). Three subclones (clone 1.1, 1.2, and 1.3) had no detectable fi was incubated for 4 hours in a humidi ed atmosphere (37 C, 5% CO2). PSMB5 expression, both in denaturing and native gels (Supplementary Solubilization solution (100 mL) was added into each well. The plate Fig. S2). No significant effect on cell growth was observed in cells was allowed to stand overnight in the incubator in a humidified without PSMB5 (Fig. 1B, middle). Depleting PSMB5 did, however, atmosphere (37 C, 5% CO2). Complete solubilization of the purple restore bortezomib sensitivity comparable with the parental cell line formazan crystals was checked and the absorbance of the samples was (Fig. 1B, bottom). Next, we introduced a new mutation (p.Ala108Thr; measured using a microplate (ELISA) reader. The wavelength used to refs. 32–35) deliberately restoring bortezomib resistance (Fig. 1C) and measure absorbance of the formazan product by the ELISA reader was confirming that each of two different identified PSMB5 mutations 550 nm. The reference wavelength was 650 nm. All MTT assay results conferred proteasome inhibitor resistance. The six most efficient were the average of triplicate wells on 96-well plates. SD was calculated sgRNAs targeting PSMB5 (Supplementary Fig. S3) were used for fi by Excel. IC50 was calculated by PRISM software. further con rmatory studies in a second PSMB5-mutant, bortezomib-resistant cell line, KMS11R, which also remained viable after mRNA sequencing PSMB5 knockout. Five of the sgRNAs significantly restored KMS11R þ Multiple myeloma bone marrow CD138 cells were harvested sensitivity to bortezomib (Supplementary Fig. S4). PSMB5 was also and total RNA was prepared using RNeasy plus using the protocols successfully depleted in two bortezomib-sensitive cell lines RPMI-8226 of the manufacturer (Qiagen). mRNA sequencing experiments (clone 11.2, 11.4, and 11.8 seen in Supplementary Fig. S2) and OCI- were performed and analyzed as described previously (30) at the MY5 cells (Supplementary Fig. S5). Surprisingly then, and in contrast Medical Genome Facility in Mayo Clinic (Rochester, MN). RNA- to published literature in yeast, removal of PSMB5 did not reduce sequencing data were processed using Mayo RNA-Seq Bioinfor- viability in four different multiple myeloma cell lines, including matics workflow MAP-RSeq (31), including read alignment bortezomib-resistant and -sensitive multiple myeloma cell lines. Our against human hg19 genome build (by TopHat v2.0.12) and read in vitro data, thus, confirm that PSMB5 mutations confer bortezomib count quantification for genes (by featureCount v1.4.4). We ana- resistance, that complete loss in PSMB5 expression is both tolerated lyzed differential expression of PSMB5, PSMB6, PSMB7, PSMB8, and restores bortezomib sensitivity, and infer that induced PSMB5 PSMB9, and PSMB10 between pre- and post-bortezomib treatment mutations are a gain of function or confer stability of expression. samples. Patient bone marrow samples were obtained by written informed consent from the patients. The studies were conducted Absence of PSMB6 is lethal in multiple myeloma in accordance with Declaration of Helsinki and that the After confirming that PSMB5 knockout was not lethal in multiple studies were approved by Mayo Clinic Institutional Review Board myeloma cell lines (while the mutation is lethal in yeast), we explored (Rochester, MN). the role of other proteasome subunits on viability and drug sensitivity. After infection and Western blot screening of RPMI-8226, six of 24 guides RNAs targeting PSMB6 were selected (Supplementary Results Fig. S6). All six sgRNAs efficiently depleted PSMB6 (Fig. 2A) and in Generation and sequencing of bortezomib-resistant cell lines contrast to PSMB5, all of them had a lethal effect ( Fig. 2B). While still Eight multiple myeloma cell lines (RPMI-8226, FR4, KMS11, viable, we then examined the expression of PSMB5, PSMB7, PSMB8, XG-1, MM1S, U266, JJN3, and OCI-MY5) sensitive to bortezomib PSMB9, and PSMB10 in the PSMB6-deficient cells, and showed that (IC50 <5 nmol/L) were selected for relative clonal resistance through expression and splicing of each of these proteins, with the exception of stepwise increases of bortezomib concentration (posttreatment IC50 PSMB9, was significantly inhibited (Fig. 2A). To exclude off-target ranging from 13 to 38 nmol/L; Fig. 1A). We defined bortezomib CRISPR effects, we infected RPMI-8226R cells with a lentivirus resistance in cell lines as an IC50 of resistant cells more than five expressing either the guide RNA–resistant WT PSMB6, two function- times higher than the parental cells. We screened all 49 proteasome ally deficient PSMB6 mutants (M1 and M2; Fig. 2C), or empty vector subunits (Supplementary Table S1) in these isogenic cell line (control). After the establishment of stable cell lines expressing guide- pairs by targeted sequencing, and found PSMB5 mutations in five resistant PSMB6 WT, M1 and M2, or empty vector, we reinfected these of the eight lines (RPMI-8226R, KMS11R, FR4R, XG-1R, and MM1. stable cell lines with sgRNA targeting PSMB6, and in this instance SR; Fig. 1A). Specifically, a p.Met104Val mutation was found successfully rescued the lethal phenotype (Fig. 2D). Restoration of cell in four cell lines (RPMI-8226R, FR4R, KMS11R, and XG-1R), viability correlated with PSMB6 expression, including the mutated and a p.Thr80Ala in MM1.S. XG-1R also acquired a mutation in PSMB6 M1 and M2 lacking critical catalytic or splicing function PSMD9 (p.Glu54Lys), a proteasome 19S subunit. No mutations (Fig. 2E). This experiment was repeated in a second bortezomib- were found in the remaining three bortezomib-resistant cell lines, resistant cell line, KMS11R, and a similar toxicity of PSMB6 knockout but two of them (U266 and JJN3) exhibited higher protein expres- was observed (Supplementary Fig. S7). To exclude mutation in PSMB5 sion levels of PSMB5 and PSMB6 when compared with parental, as a confounding factor, the toxicity of PSMB6 knockout was also drug-sensitive cell lines (Supplementary Fig. S1). Therefore, we confirmed in a bortezomib-sensitive, unmutated OCI-MY5 cell line identified a potential resistance mechanism secondary to protea- (Supplementary Fig. S8). These results demonstrate the importance of some subunit mutation or increased proteasome subunit expression PSMB6, highlighting that these human multiple myeloma cell lines in seven of eight bortezomib-resistant cell lines. depend on PSMB6 structurally, rather than functionally.

PSMB5-depleted cells are viable and confer bortezomib Sensitivity to bortezomib is restored by functionally deﬁcient sensitivity PSMB6 expression Because PSMB5 mutations were present in ﬁve of eight bortezomib- In an attempt to develop multiple myeloma cell lines that survive resistant cell lines, but rare in primary patients (see Introduction after PSMB6 deletion, we chose bortezomib-resistant KMS11R, which

AACRJournals.org Mol Cancer Res; 2020 OF3

Shi et al.

AC1.2

Cell lines Mutations IC50 (resistant) IC50 (parent) RPMI8226 PSMB5 Met104Val 34 3.5 1 FR4 PSMB5 Met104Val 35 2.8 KMS11 PSMB5 Met104Val 14 2.3 0.8 XG-1 PSMB5 Met104Val, PSMD9 Glu54Lys 38 2.6 1+B5 MM1.S PSMB5 Thr80Ala 34 1.9 0.6 U266 Non 17 2.7 1+B5M JJN3 Non 37 4.2 0.4 My5 Non 13 2.3 MTT % control 1+NS 0.2 C 1 2 3 4 5 6 C+NS 0 B B5 BTZ nmol/L 1 0.06 0.36 0.11 0.48 0.09 0.28 0 5 10 15 20 25 30 35 β-Acn 1.2 1.0 1.23 1.0 1.0 1.0 0.97 1.03 1

1.2 0.8 5+B5 1 5+B5M 0.8 0.6 0.6 5+NS 0.4 0.4 0.2 C+NS MTT % of control MTT % control 0 0.2

0 BTZ nmol/L 0 5 10 15 20 25 30 35 1.2 1 2 1 3 1+B5 1+B5M 1+NS 5+B5 5+B5M 5+NS C+ C- 8226 0.8 4 B5 0.6 5 6 0.97 1.12 0.26 1.56 1.39 0.03 1.80 1.61 1.0 0.4 Control 1 β-Acn 0.2 Control 2 MTT % control 0 1.07 1.22 1.33 1.22 1.46 1.37 1.33 1.05 1.0 0 5 10 15 20 25 30 35 BTZ nmol/L -0.2 -0.4

Figure 1. Sensitivity to bortezomib (BTZ) is restored by CRISPR depletion of PSMB5. A, Eight resistant multiple myeloma cell lines were developed using a stepwise increase in

bortezomib concentration. Genome mutations were analyzed using a targeted sequencing panel, which included all 49 proteasome subunits. An IC50 was obtained for both parental and isogenic, resistant cell lines (72-hour treatment with bortezomib). B, One of the bortezomib-resistant multiple myeloma cell lines, RPMI-8226R, was depleted of PSMB5 using CRISPR guide RNA. One to six are PSMB5-depleted clones as detected by Western blot analysis with % reductions noted (top). C and control represent the RPMI-8226R cell line (bortezomib resistant). Control 2 is the bortezomib-sensitive parental RPMI-8226 cells. MTT assay of the depleted and control cell lines for cell viability (middle). Drug response of PSMB5 depleted and controls following treatment with bortezomib (bottom). C, This represents a rescue experiment using lentiviral reintroduction of PSMB5 performed on PSMB5-depleted clones 1 and 5 B. B5, cells manipulated with a lentivector expressing wild-type PSMB5; B5M, a lentivector expressing PSMB5 with mutation at Ala108Thr; NS, empty lentivector used as a nonspeciﬁc control; Cþ, RPMI-8226R cells maintained with 20 nmol/L bortezomib; and C, RPMI-8226R cells maintained without bortezomib. Reintroduction of wild-type or mutant PSMB5 restored drug resistance. Expressed wild-type or mutated PSMB5 detected by Western blot analysis (bottom). Protein expression was detected by Western blotting and analyzed using ImageJ software (the numbers below the blots in B and C were normalized to b-actin and untreated control).

has a PSMB5 mutation (p.Met104Val). We selected two sgRNAs (Cr10 However, both “PSMB6 low” clones exhibited restored sensitivity to and Cr15) of six sgRNAs targeting PSMB6 for further cloning bortezomib (Fig. 3E). This demonstrates that a PSMB5 mutation (Fig. 3A) and identified two viable clones (10.3 and 15.2), which causing bortezomib resistance is dependent on the natural expression expressed significantly lower PSMB6 than the control cells, but of PSMB6 and that deletion of certain amino acids resulting in sufficient to maintain cell viability. A PSMB6 loss-of-splicing function lowering of PSMB6 levels can reverse resistance. This also highlights was observed in clone 10.3 (Fig. 3B). Sequencing data revealed that that PSMB6 amino acids 160Ala and 161Ile play an important role for clone 10.3 had acquired an in-frame deletion with amino acid change appropriate splicing. in PSMB6, leading to the replacement of amino acids (160Ala and 161Ile) by 160Val (Fig. 3C). Furthermore, clone 15.2 had a deletion of Stability of the PSMB5 and PSMB6 proteasome subunits is PSMB6 181Met (Fig. 3C). The amino deletion and low expression of codependent PSMB6 resulted in no significant decline in cell growth potential for Because multiple myeloma cell lines survived without PSMB5, this clone 15.2, while clone 10.3 had a slower growth rate (Fig. 3D). raised the question of which proteasome subunit compensates for the

OF4 Mol Cancer Res; 2020 MOLECULAR CANCER RESEARCH

Multiple Myeloma Cell Survival Depends on PSMB6 and PSMB7

A Cr.1 Cr.3 Cr.4 Cr.10 Cr.11 Cr.15 Control C -1 1 33 G T K B6 B6WT 0.13 0.36 0.30 0.29 0.27 0.11 1.0 -1 1 33 A A A B5 B6M1 0.33 0.42 0.39 0.45 0.44 0.18 1.0 1 33 A A B7 B6M2

0.28 0.35 0.36 0.18 0.24 0.16 1.0 D B8 1.4 1.2 0.73 0.70 0.66 0.64 0.61 0.60 1.0 1 B9 0.8 0.6 1.02 0.97 0.86 0.75 0.76 0.67 1.0 0.4 0.2 B10 0 0.47 0.54 0.66 0.56 0.55 0.28 1.0 MTT % (NS1+NS2) control β-Acn 0.83 0.92 1.32 1.18 1.14 0.87 1.0 B E 1.2 B6 1 1.0 0.82 1.11 1.23 0.15 0.76 0.81 0.97 0.8 0.6 GAPDH 0.4 1.0 1.0 0.95 1.16 1.0 1.20 1.27 1.20 B6M1+Cr.10 B6WT+Cr.10 B6M2+Cr.10 NS1+Cr.10 B6WT+NS2 B6M1+NS2 B6M2+NS2 NS1+NS2 MTT % control 0.2 0

Figure 2. PSMB6 is essential for multiple myeloma cell survival. A, Depletion of PSMB6 with six different sgRNAs (Cr.1, Cr.3, Cr.4, Cr.10, Cr.11, and Cr.15) demonstrated on day 4 that splicing for PSMB5, PSMB7, PSMB8, and PSMB10 was significantly inhibited (the top band is the proteasome subunit proprotein and the bottom band is the mature proteasome subunit). B, The same panel of cell lines was analyzed using an MTT assay on day 8 after transfection (96-well plate with triplicate wells for each sample for 72 hours of bortezomib treatment). C, A schematic presentation of the mutations introduced in PSMB6. B6WT, PSMB6 wild- type protein; B6M1, PSMB6 with mutations at G-1A, T1A, and K33A; and B6M2, PSMB6 with mutations at T1A and K33A without the propeptide. The mature PSMB6 is characterized by an open box and the dark box marks the PSMB6 prosequence. B6WT, B6M1, and B6M2 were modified to resistant PSMB6 Cr10 targeting [protospacer adjacent motif sequence TGG was mutated to AGG for PSMB6 sgRNA 10 (CCCGGAGCCTCCAATGGCAA), resulting in nontargeting of it by sgRNA 10]. D, The toxicity of PSMB6 depletion was rescued by expressing either PSMB6 wild-type or PSMB6 mutants. RPMI-8226R, bortezomib-resistant cell lines expressing wild-type PSMB6 had reintroduced stable expression of PSMB6 WT, M1, and M2 described in C, and was then infected with sgRNA- targeting PSMB6 (Cr10). The infected cells were aliquoted into 96-well at day 4 and MTT reagent was added at day 8 after infection. Control NS1, RPMI-8226R cell line infected with an empty lentivector; Control NS2, a lentivector expressing nonspecificsgRNA.B5–B10 represents PSMB5–PSMB10. E, PSMB6 expression was measured on the cell lines from D at day 4 after infection. Protein expression was detected by Western blotting and analyzed by ImageJ software [numbers below the blots in A and E were normalized to b-actin (A)andGAPDH(E) and untreated control].

loss of PSMB5. Consequently, we checked the expression of five other by removing prosequence following PSMB5 deletion. PSMB10 was proteasome catalytic subunits (PSMB6, PSMB7, PSMB8, PSMB9, and expressed at higher levels than the control cells. PSMB10) and one noncatalytic subunit (PSMB2 as a control) on We ran a native gel to check 20S proteasome protein expression multiple myeloma cell lines lacking PSMB5 (Fig. 4A). After PSMB5 levels (Fig. 4B). PSMB10 was the only protein expressed at higher deletion, PSMB6 and PSMB7 expression levels were significantly levels than the control cells. Expression of both PSMB6 and PSMB7 lowered than in control cells. No significant change in PSMB8 and was lower, while there was no significant change in PSMA1 expression PSMB9 expression was observed in different RPMI-8226 cell sub- (one of the a subunits used as a control). clones, even though PSMB8 was expected to significantly compensate The dependence of PSMB6 and PSMB7 expression on PSMB5 for the loss of PSMB5, because PSMB8 replaces PSMB5 upon stim- prompted us to query the corollary: is the dependence of PSMB5 ulation by IFNg. PSMB9 was the only protein with a significant and PSMB7 expression based on PSMB6 presence. The same two influence in maturation (significant pro-PSMB9 band was observed) clones (10.3 and 15.2) from Fig. 3B expressing a lower level of

AACRJournals.org Mol Cancer Res; 2020 OF5

Shi et al.

A Cr.1 Cr.3 Cr.4 Cr.10 Cr.11 Cr.15 C D 1.2 B6 1 0.54 0.54 0.18 0.32 0.27 0.14 1.0 0.8 β-Acn 0.6 1.05 1.12 1.19 0.99 1.09 1.16 1.0 0.4 MTT % control B 10.1 10.2 10.3 10.4 15.1 15.2 15.3 C 0.2

B6 0 Clone 10.3 Clone 15.2 Control 1.02 1.28 0.09 1.19 1.22 0.19 0.93 1.0 β-Acn 0.64 0.60 0.69 0.69 0.66 0.80 0.96 1.0 E 1.2

C 1 160 180 B6del 10.3 A I 0.8 G M 6del 15.2 0.6 B6 15.3 0.4 Control

MTT % DMSO control Control 2 V G 0.2 Clone 10.3 Clone 15.2 0 0 5 10 15 20 BTZ nmol/L

Figure 3. Sensitivity to bortezomib (BTZ) is restored by even low level PSMB6 expression. A, The KMS11R bortezomib-resistant cell line was targeted by sgRNAs (Cr 1, 3, 4, 10, 11, and 15) against PSMB6 (B6). B, Further subcloning was performed (clone 10.1, 10.2, 10.3, and 10.4 were from KMS11R cells targeted by sgRNA Cr.10 and clone 15.1, 15.2, and 15.3 were from KMS11R cells targeted by sgRNA Cr.15). This identified clone 10.3, which expressed low level unspliced PSMB6 and clone 15.2, which expressed low levels of PSMB6. C, Sanger sequencing revealed that clone 10.3 had acquired Ala (A)160 and Ile (I)161 replaced by Val (V)160. Clone 15.2 had a Met (M) 181 deletion identified. D, The multiple myeloma cell lines expressing residual PSMB6 grow normally (clone 15.2), while clone 10.3 grows more slowly than parental cells, but both remained viable in contrast to cells completely deficient in PSMB6. E, Clones 10.3 and 15.2 were sensitive to bortezomib treatment. Controls included KMS11R (bortezomib resistant) and control 2 KMS11 (bortezomib sensitive). Protein expression was detected by Western blotting and analyzed by ImageJ software (the numbers below the blots in A and B were normalized to b-actin and untreated control).

mutated PSMB6 were employed. We detected expression of PSMB5 restore PSMB6, PSMB7, and PSMB8 expression to parental cell levels. and PSMB7 in these two clones, but at significantly lower expression PSMB9 levels were partially restored, although remaining lower than than the control cells, with significant amounts of the proprotein of the control and in its unspliced form (Fig. 5B). Expression levels of PSMB5 and PSMB7 appearing, indicating that splicing of these two PSMB5 M2 (without propeptide and lacking catalytic function) were proteins was also significantly inhibited (Fig. 4C). Here, we, there- lower than the WT and M1, and the PSMB5 deletion phenotype was fore, demonstrate that expression of PSMB5 and PSMB6 is code- not rescued (Fig. 5B). Native gel detection indicated that only a low pendent and that expression of PSMB7 is highly dependent on both level of PSMB5 was incorporated into 20S proteasome (Fig. 5C). This PSMB5 and PSMB6. result show that propeptide of PSMB5 is important for expression and incorporation into 20S proteasome. We repeated this experiment by Stability of PSMB5 and PSMB6 does not require normal infecting KMS11R cells with WT PSMB6 and functionally deficient functional activity mutants. In this case, all PSMB6 variants restored the expression of Next, we addressed whether this mutual dependence was based on PSMB5 to parental cell levels and partial rescue was seen for PSMB7 the proteasome's catalytic function or structure. We constructed (Fig. 5D). We found that RPMI-8226 cells lacking PSMB5 had lentivectors expressing PSMB5 WT, M1, and M2 (Fig. 5A). We relatively higher ubiquitinated protein accumulation, but no signifi- selected two subclones from sgRNA, PSMB5-depleted RPMI-8226 cant difference was seen between RPMI-8226 cells without PSMB5 and and reinfected with guide-resistant PSMB5 WT, M1, and M2. Unsur- the cells rescued by PSMB5 WT, M1, and M2 (Fig. 5E), which indicate prisingly, reexpression of wild-type PSMB5 restored the expression of that it has no significant influence on the function of human 20S PSMB6, PSMB7, PSMB8, PSMB9, and PSMB10 to the parental cell proteasome without PSMB5. These results indicate that PSMB5 and level (Fig. 5B). PSMB5 M1 was expressed as an unspliced form, as PSMB6 expression are codependent, but do not require catalytic or expected (Fig. 5B). However, the unspliced PSMB5 M1 was still able to splicing function of the protein.

OF6 Mol Cancer Res; 2020 MOLECULAR CANCER RESEARCH

Multiple Myeloma Cell Survival Depends on PSMB6 and PSMB7

AB1.1 1.2 1.3 C1 11.2 11.4 11.8 C2 B5 1.1 1.2 1.3 C1 11.2 11.4 11.8 C2 0.04 0.06 0.06 0.86 0.05 0.05 0.07 1.0 B5 0.06 0.03 0.03 0.81 0.07 0.16 0.21 1.0 B6 B6 0.36 0.27 0.25 0.89 0.30 0.22 0.56 1.0 0.28 0.26 0.48 0.82 0.31 0.30 0.34 1.0 B7 B7 0.15 0.35 0.38 1.64 0.63 0.14 0.09 1.0 0.42 0.29 0.35 0.66 0.28 0.33 0.31 1.0 B8 B8 0.98 0.92 0.81 0.75 0.83 0.81 0.78 1.0 2.0 1.4 1.97 1.09 1.62 1.38 0.86 1.0 B9 0.70 0.67 0.65 0.82 0.56 0.69 0.60 1.0 B9 B10 1.17 0.77 0.86 0.60 0.65 0.56 0.39 1.0 0.84 1.49 1.51 0.83 1.73 1.98 1.39 1.0 B10 B2 1.52 1.20 1.40 1.08 1.42 1.19 1.25 1.0 0.47 0.44 0.44 0.95 0.73 0.68 0.75 1.0 A1 GAPDH 0.82 0.97 0.90 1.03 0.80 0.82 0.85 1.0 1.23 1.32 1.38 0.98 1.13 1.11 1.11 1.0

C 10.1 10.2 10.3 10.4 15.1 15.2 15.3 C B5

1.39 1.69 0.42 1.42 1.38 0.56 1.21 1.0

B7 1.44 2.09 0.77 1.40 1.37 0.49 1.13 1.0 β-Acn 0.51 0.48 0.62 0.62 0.60 0.73 0.89 1.0

Figure 4. High interdependence was observed between expression of PSMB5 and PSMB6. A, Expression of PSMB6 and PSMB7 was significantly reduced in multiple myeloma cell lines lacking the PSMB5 gene. Clones 1.1, 1.2, and 1.3 are subclones derived from clone 1 of bortezomib-resistant RPMI-8226R CRISPR-depleted PSMB5 and clones 11.2, 11.4, and 11.8 are subclones from clone 11 of bortezomib-sensitive RPMI-8226–depleted PSMB5. C1, parental bortezomib-sensitive RPMI-8226; C2, parental bortezomib-resistant RPMI-8226R. B, Incorporation of PSMB6 and PSMB7 was significantly reduced in multiple myeloma cell lines lacking the PSMB5 gene. The same panel of lysates from A was examined for protein expression using a native gel. C, Expression of PSMB5 and PSMB7 was significantly reduced when multiple myeloma cell lines expressed low levels of mutated PSMB6. The panel of lysates from Fig. 3B was used for Western blot detection. B2, PSMB2. B5–B10 represents PMBB5– PSMB10. Protein expression was detected by Western blotting and analyzed by ImageJ software (numbers below the blots in A–C were normalized to GAPDH (A), PSMA1 (B), and b-actin (C) and untreated control].

PSMB7 depletion is lethal to multiple myeloma cell lines Increased expression levels of PSMB5, 6, and 7 in patients after Because PSMB7 expression is highly dependent on the expression of bortezomib treatment PSMB5 and PSMB6, we proposed that expression of PSMB5 and Recently, we finished RNA-sequencing for 450 patients with mul- PSMB6 would also be highly dependent on PSMB7. Nine of 12 PSMB7 tiple myeloma (unpublished data). Among these, 10 patients with sgRNAs (Supplementary Table S2) were significantly toxic to the paired samples were collected before and after multiple rounds of RPMI-8226R cell line (Fig. 6A), and with one exception correlated chemotherapy treatment (Supplementary Data). Eight of these had to the corresponding PSMB7 knockout efficiency (Fig. 6B). As received proteasome inhibitors. mRNA sequencing results indicated expected, expression of both PSMB5 and especially PSMB6 were that 2 (patient 1 ID: 6496153 and patient 2 ID: 5493955) of the 8 reduced significantly following PSMB7 depletion (Fig. 6C). PSMB8 patients had notable and significantly increased expression levels of was the only protein among the immunoproteasome subunits with a PSMB5, 6, and 7, but not PSMB8, 9, and 10 (Fig. 7). PSMB5, 6, and 7 significant reduction in its expression. Unlike the PSMB6 knockout, from 1 patient increased 1.5-, 2.0-, and 2.0-fold correspondingly in constitutive and immunoproteasome subunit splicing were not influ- late-stage disease compared with pretreatment. PSMB5, 6, and 7 from enced after PSMB7 knockout. patient 2 increased 5.3-, 1.9-, and 1.7-fold correspondingly in late-stage

AACRJournals.org Mol Cancer Res; 2020 OF7

Shi et al.

A -1 1 32 33 C C EV WT M1 M2 EV WT M1 M2 G T K K B5WT B5

-1 1 32 33 1.0 0.10 0.81 0.92 0.18 0.06 0.89 1.23 0.25 A A A A B5M1 A1

1 32 33 1.0 1.14 1.06 0.78 1.06 0.95 0.90 0.66 1.22 A A A B5M2 D B6 B Clone 11.4 Clone 11.8 1.0 0.84 1.28 1.48 0.12 0.94 1.03 1.34 B5

C EV WT M1 M2 EV WT M1 M2 1.0 1.07 1.22 1.12 0.32 098 0.94 0.81 B5 B7 1.0 0.04 0.64 0.63 0.38 0.09 1.0 0.93 0.71 1.0 1.20 1.18 0.90 0.22 0.77 0.61 0.53 B6 GAPDH 1.0 0.49 0.70 0.66 0.51 0.55 0.98 0.93 0.58 1.0 0.90 0.81 0.98 0.71 0.97 0.97 0.84 B6M1+Cr.10 NS1+NS2 B6WT+NS2 B6M1+NS2 B6WT+Cr.10 B6M2+NS2 NS1+Cr.10 B7 B6M2+Cr.10 1.0 0.34 0.75 0.64 0.56 0.34 0.93 0.74 0.55 B8 1.0 0.24 0.44 0.72 0.60 0.71 0.78 0.77 0.73 E C EV WT M1 M2 EV WT M1 M2 B9 1.0 0.48 0.72 0.53 0.56 0.53 1.0 0.76 0.69 B10 1.0 0.39 0.41 0.36 0.35 0.31 0.38 0.45 0.46 β-Acn 1.0 1.69 1.58 1.46 1.35 1.34 1.0 1.12 0.94

Figure 5. Expression of PSMB5 and PSMB6 is mutually dependent on structure, but not function. A, Schematic presentation of mutations introduced into PSMB5. B5 WT, PSMB5 wild-type protein; B5 M1, PSMB5 with mutations at G-1A, T1A, K32A, and K33A. The PAM sequence is mutated in the sgRNA target region. B5 M2, PSMB5 with mutations at T1A, K32A, and K33A without expression of the propeptide. The mature part of PSMB5 is characterized by an open box and the dark box marks the PSMB5 prosequence. B, Expression of PSMB6 and PSMB7 is restored by the expression of wild-type or guide-resistant mutated B5 M1 but not B5 M2. Cis for RPMI- 8226R control cells. EV, empty lentivector. C, Mutated PSMB5 with unspliced prosequence (PSMB5 M1) can be efﬁciently incorporated into the 20S proteasome. The panel of lysates from B was used for native gel detection. D, Expression of PSMB5 and PSMB7 is restored by the expression of wild-type or mutated PSMB6 (B6 M1 and B6 M2). The cell lysate from Fig. 2E was used for Western blot assay. E, Ubiquitination of multiple myeloma cell lines lacking PSMB5 by reintroducing expression of wild-type and mutated PSMB5. The same panel of cell lysate as in B was used to detect the expression of ubiquitination. The protein expression was detected by Western blotting and analyzed by ImageJ software [numbers below the blots in B–D were normalized to b-actin (B), PSMA1 (C), and GAPDH (D) and untreated control].

disease compared with pretreatment. On the other hand, in 6 of the these immunoproteasome subunits may not overcome bortezomib patients an increase was not observed. resistance.

Absence of the immunoproteasome subunits PSMB8 and PSMB9 does not restore sensitivity to bortezomib in multiple Discussion myeloma cell lines Our results indicate that alterations in PSMB5 (via mutation or PSMB8 and PSMB9 are also targets of proteasome inhibitors overexpression) are the predominant mechanism for in vitro induced bortezomib and carﬁlzomib. We designed three sgRNAs for PSMB8 bortezomib resistance, consistent with previous reports (32, 33, 35–37). or PSMB9 (Supplementary Table S3). We did not observe any Depleting PSMB5 in PSMB5-mutant cell lines restored sensitivity to signiﬁcant correlation in bortezomib sensitivity with or without the bortezomib and introducing the frequently reported PSMB5 muta- PSMB8 and PSMB9 subunits in human multiple myeloma cell lines tion (p.Ala108Thr) into PSMB5-depleted lines restored resistance (Supplementary Fig. S9), indicating that inhibition of the activity of to bortezomib, further establishing PSMB5 mutations as a major

OF8 Mol Cancer Res; 2020 MOLECULAR CANCER RESEARCH

Multiple Myeloma Cell Survival Depends on PSMB6 and PSMB7

A C 1.2 C Cr.1 Cr.2 Cr.3 Cr.4 Cr.8 Cr.9 Cr.10 Cr.11 B7 1 1.0 0.12 0.13 0.10 0.12 0.05 0.10 0.14 0.08 0.8 B5 0.6 1.0 0.50 0.64 0.50 0.52 0.45 0.58 0.67 0.39 0.4 MTT % control B6 0.2 1.0 0.25 0.25 0.14 0.10 0.09 0.23 0.24 0.08 0 C B8 Cr.2 Cr.3 Cr.5 Cr.9 Cr.1 Cr.4 Cr.7 Cr.8 Cr.6 Cr.11 Cr.10 Cr.12 1.0 0.76 0.55 0.56 0.57 0.41 0.43 0.67 0.48

B9 B C Cr.1 Cr.2 Cr.3 Cr.4 Cr.5 Cr.6 1.0 1.03 0.95 0.98 0.98 1.06 0.87 1.29 1.07 B7 1.0 0.15 0.22 0.06 0.07 1.01 0.82 B10 1.0 0.99 1.06 0.72 0.81 0.53 0.83 1.18 0.78 β-Acn 1.0 1.10 0.91 1.08 1.26 0.82 1.20 β-Acn 1.0 1.07 1.10 1.07 1.22 1.07 1.39 0.87 1.01

C Cr.7 Cr.8 Cr.9 Cr.10 Cr.11 Cr.12

B7 1.0 0.87 0.06 0.20 0.12 0.19 0.13

β-Acn 1.0 1.03 1.08 1.34 1.13 1.40 1.07

Figure 6. PSMB7 depletion is lethal to multiple myeloma cells. A, MTT assay for RPMI-8226R cells infected with lentivector expressing sgRNAs (Cr.1–Cr.12) targeting PSMB7. On day 4 after infection, the infected cells were aliquoted for an MTT assay (triplicate wells), which was read on day 8. C, control cell line RPMI-8226R. B, The panel from A was examined for the expression of PSMB7 by Western blotting. C, The expression of PSMB5–PSMB10 (B5–B10) was detected by Western blotting from RPMI-8226R cells targeted by eight sgRNAs against PSMB7. The protein expression was detected by Western blotting and analyzed by ImageJ software (numbers below the blots in B and C were normalized to b-actin and untreated control). contributor to bortezomib resistance in multiple myeloma cell lines, stantially exposed to proteasome inhibitors, a higher frequency of and consistent with the report by Hess and colleagues, wherein PSMB5 mutation will be discovered. PSMB5 of K562 cells mutated by CRISPR technology (p.Arg78Asn, Nevertheless, because PSMB5 mutations are generally uncommon p.Ala79Gly, p.Ala79Thr, and p.Ala108Val) but expressing physio- in primary patients analyzed to date in large sequencing efforts, we logic level of PSMB5 developed resistance to bortezomib (38). took a deeper dive into proteasome subunit deletion employing Different proteasome inhibitor–resistant mechanisms may well CRISPR. Our results demonstrate that deletion of PSMB6 and PSMB7, exist in multiple myeloma cell lines, as in a report by Soriano and but not PSMB5, is lethal for human multiple myeloma cell lines. Both colleagues (39), which concluded that proteasome inhibitor– bortezomib-resistant and -sensitive human multiple myeloma cell adapted cells were largely independent of proteasome activity and lines are able to survive in the absence of PSMB5. This result changes that resistance was due to metabolic adaptions. our knowledge about the role PSMB5 on human cell survival, at With respect to primary patient samples, in Barrio and collea- least in the context of malignant transformation. The yeast gues (24), we reported the first example of a drug-resistant patient with PSMB5 homologous gene, pre2, has been reported as an essential multiple PSMB5 mutations. The same mutation (p.Met104Val) gene. By mutating pre2,butnotpre3 (homologous to the human that appeared in the four cell lines in our analysis was also noted PSMB6) or pup1 (homologous to PSMB7), yeast growth defects are in Barrio and colleagues’ report, although mutated to a different found (16, 17, 19, 40). This might be explained by the immuno- amino acid (p.Met104Ile, position with prosequence ¼ p.Met45Ile proteasome subunit PSMB8 (absent in yeast), compensating for the without prosequence). Two closely located PSMB5 amino acids muta- loss of PSMB5 in human cells. Our result in multiple myeloma cells tions (p.Thr80Ala in our study and VS p.Ala86Pro and p.Ala79Thr in is in contrast to the findings reported by Hirano and colleagues (41), Barrio and colleagues’ study) were also reported in Barrio colleagues’ in which depleting PSMB5 led to cytotoxicity in 293 cell lines, which study. The patient in this instance had been treated with multiple could reflect off-target activity of short hairpin RNA (shRNA) or regimens containing bortezomib. We would predict that in future deep alternatively that nontransformed 293 cells are, in contrast to sequencing studies (as opposed to exome) of relapsed patients sub- transformed malignant cells, dependent on PSMB5.

AACRJournals.org Mol Cancer Res; 2020 OF9

Shi et al.

90 Figure 7. ﬁ 80 Signi cant increase in expression levels of PSMB5, 6, and 7 in patients with late-stage disease. mRNA sequencing results 70 revealed signiﬁcantly increased PSMB5, 6, and 7 levels when early and late stages of disease were examined, while levels of PSMB8, 9, 60 and 10 did not change. P1, patient one; P2, patient two; RPKM, reads per kilobase of transcript, per million mapped reads. 50 P1 (pretreatment)

RPKM 40 P1 (end Stage) 30 20 10 0 PMSB5 PSMB6 PSMB7 PSMB8 PSMB9 PSMB10

40 P2 (pretreatment)

RPKM 30 P2 (end Stage)

0 PMSB5 PSMB6 PSMB7 PSMB8 PSMB9 PSMB10

Our results demonstrate that expression of the three constitutive and colleagues (43) reported that PSMB10 requires PSMB9 for effi- proteasome subunits PSMB5, PSMB6, and PSMB7 is highly codepen- cient incorporation into preproteasomes, and preproteasomes con- dent. This dependence is reliant on the structure, but not the function, taining PSMB9 and 10 require PSMB8 for efficient maturation. The of PSMB5 and PSMB6. PSMB6 plays an essential role in the 20S authors proposed a cooperative model for proteasome assembly, with proteasome biogenesis. In the absence of PSMB6, the splicing of all one major pathway leading to immunoproteasomes containing all other proteasome catalytic subunits, with the exception of PSMB9, is three IFNg-inducible catalytic b subunits (PSMB8, PSMB9, and inhibited, indicating that these subunits cannot be incorporated into PSMB10), another major pathway leading to constitutive proteasomes the 20S core without PSMB6 expression. This phenotype was not containing all three constitutive catalytic b subunits (PSMB5, PSMB6, observed by deletion of PSMB5 and PSMB7. Our current knowledge of and PSMB7), and a minor pathway possibly leading to mixed catalytic the incorporation of b-subunits is mostly derived from shRNA subunits. Our results demonstrate the existence of the pathway leading knockdown from a single b subunit experiment (41). In that study, to the constitutive proteasome assembly. Our results also indicate a the authors proposed that addition of b subunits to the a ring occurred cross-talk between the two main pathways, because splicing of the in an orderly manner, starting from b2 (PSMB7), followed by b3, b4, PSMB9 was inhibited without expression of PSMB5 and splicing of b5 (PSMB5), b6, b1 (PSMB6), and finally b7. However, our results PSMB8 and PSMB10 was inhibited by depletion of PSMB6. highlight that PSMB6 plays a central role in the incorporation of b Both, bortezomib and carfilzomib target the constitutive protea- subunits. Therefore, we propose a new model for constitutive proteasome and the immunoproteasome indiscriminately. It has been spec- some assembly (Supplementary Fig. S 10) wherein PSMB6 is the first ulated that adverse events attributed to proteasome inhibitors, such as subunit to be added to a ring, PSMB7 is added to the a ring peripheral neuropathy and gastrointestinal effects, might be caused by immediately after PSMB6 (or same time with PSMB6), PSMB5 and targeting of the constitutive proteasome. Thus, developing immuno- other proteasome subunits are incorporated into the 20S proteasome proteasome-specific drugs was proposed as a new strategy for multiple later (we did not delete noncatalytic b subunit in this study, so the order myeloma therapy (44–46). However, our results show that depletion of addition of noncatalytic b needs future study), because multiple of the immunoproteasome subunits PSMB8 and PSMB9 was nonrel- myeloma cell lines cannot survive without PSMB6 and PSMB7, but can evant to bortezomib resistance, showing no influence on the growth survive without PSMB5. of multiple multiple myeloma cell lines. Mice completely lacking Immunoproteasome subunits PSMB9 and PSMB10 have been immunoproteasome subunits (PSMB8, PSMB9, and PSMB10) are reported as mutually required for incorporation into the 20S protea- viable, fertile, and appeared healthy, only displaying major alterations some core, independently of PSMB8 (42). In a separate study, Griffin in antigen presentation (47). However, despite this promising

OF10 Mol Cancer Res; 2020 MOLECULAR CANCER RESEARCH

Multiple Myeloma Cell Survival Depends on PSMB6 and PSMB7

therapeutic index, our studies suggest that targeting the immunopro- PSMB5, PSMB6, and PSMB7 was highly codependent. Importantly, teasome in multiple myeloma might not represent an ideal therapeutic the physical structure, but not the catalytic function, is required for this strategy. dependence. We analyzed 10 patients with paired bone marrow samples collected before treatment and at later stage posttreatment disease by mRNA Disclosure of Potential Conflicts of Interest sequencing. Eight of these patients were treated with multiple drugs No potential conflicts of interest were disclosed. including bortezomib and developed multiple drug–resistant diseases including to resistance to bortezomib. Two of these patients had Authors’ Contributions elevated expression levels of PSMB5, PSMB6, and PSMB7, but not in vivo C.-X. Shi: Conceptualization, formal analysis, validation, investigation, PSMB8, PSMB9, and PSMB10. Because mutation of PSMB5 is methodology, writing-original draft. Y.X. Zhu: Formal analysis, validation, rare, elevated expression levels of PSMB5, PSMB6, and PSMB7 might investigation, methodology. L.A. Bruins: Methodology. C. Bonolo de Campos: contribute significantly to bortezomib resistance in the clinic. This, Formal analysis. W. Stewart: Data curation, formal analysis, validation, rather, preliminary observation requires larger datasets. methodology. E. Braggio: Formal analysis, investigation, writing-review and Because expression of PSMB5, PSMB6, and PSMB7 is codependent editing. A.K. Stewart: Conceptualization, supervision, funding acquisition, writing-review and editing. and because low level expression of PSMB6 and knockout of PSMB5 can overcome bortezomib resistance caused by PSMB5 mutation, we Acknowledgments propose that inducing the downregulation or otherwise inhibiting the We thank Dr. Leif Bergsagel for providing human MM parent cell lines. We expression of either PSMB5, PSMB6, or PSMB7 might be a novel thank Dr. Didier Trono for providing packaging plasmids psPAX2 (Addgene strategy to overcome human multiple myeloma resistance to protea- plasmid # 12260) and pMD2.G (Addgene plasmid # 12259). We thank the Mayo some inhibitors. Clinic Medical Genome Facility for providing targeting sequencing. This study In summary, our current and previous studies demonstrated that was supported by U54 grant 4U54CA224018, the Mayo Clinic Multiple Myeloma PSMB5 mutation is a dominant mechanism in multiple myeloma cell SPORE, and the Multiple Myeloma Foundation. A.K. Stewart was supported by NIH (1U54224018). lines induced to be resistant to bortezomib in vitro. However, such in vivo mutations are rare . Alterations in PSMB expression is perhaps The costs of publication of this article were defrayed in part by the payment of page in vivo an alternate and rational explanation for induced resistance . charges. This article must therefore be hereby marked advertisement in accordance In that regard our studies show that, PSMB5, PSMB6, and PSMB7, with 18 U.S.C. Section 1734 solely to indicate this fact. but not PSMB8 and PSMB9, are highly relevant for bortezomib sensitivity in multiple myeloma. Absence of PSMB6 or PSMB7, but Received October 16, 2019; revised March 2, 2020; accepted June 16, 2020; not PSMB5, was lethal in multiple myeloma cell lines. Expression of published first June 19, 2020.

References 1. Rajkumar SV, Kumar S. Multiple myeloma: diagnosis and treatment. Mayo Clin 12. Finley D. Recognition and processing of ubiquitin-protein conjugates by the Proc 2016;91:101–19. proteasome. Annu Rev Biochem 2009;78:477–513. 2. Shah JJ, Orlowski RZ. Proteasome inhibitors in the treatment of multiple 13. Rock KL, Goldberg AL. Degradation of cell proteins and the gener- myeloma. Leukemia 2009;23:1964–79. ation of MHC class I-presented peptides. Annu Rev Immunol 1999;17: 3. Moreau P, Richardson PG, Cavo M, Orlowski RZ, San Miguel JF, Palumbo A, 739–79. et al. Proteasome inhibitors in multiple myeloma: 10 years later. Blood 2012;120: 14. Budenholzer L, Cheng CL, Li Y, Hochstrasser M. Proteasome structure and 947–59. assembly. J Mol Biol 2017;429:3500–24. 4. Laubach JP, Schlossman RL, Mitsiades CS, Anderson KC, Richardson PG. 15. Berkers CR, Verdoes M, Lichtman E, Fiebiger E, Kessler BM, Anderson KC, et al. Thalidomide, lenalidomide and bortezomib in the management of newly Activity probe for in vivo profiling of the specificity of proteasome inhibitor diagnosed multiple myeloma. Expert Rev Hematol 2011;4:51–60. bortezomib. Nat Methods 2005;2:357–62. 5. Rosenthal A, Kumar S, Hofmeister C, Laubach J, Vij R, Dueck A, et al. A phase Ib 16. Friedman H, Snyder M. Mutations in PRG1, a yeast proteasome-related gene, study of the combination of the aurora kinase inhibitor alisertib (MLN8237) and cause defects in nuclear division and are suppressed by deletion of a mitotic bortezomib in relapsed multiple myeloma. Br J Haematol 2016;174:323–5. cyclin gene. Proc Natl Acad Sci U S A 1994;91:2031–5. 6. Kumar SK, Berdeja JG, Niesvizky R, Lonial S, Laubach JP, Hamadani M, et al. 17. Chen P, Hochstrasser M. Biogenesis, structure and function of the yeast 20S Safety and tolerability of ixazomib, an oral proteasome inhibitor, in com- proteasome. EMBO J 1995;14:2620–30. bination with lenalidomide and dexamethasone in patients with previously 18. Heinemeyer W, Fischer M, Krimmer T, Stachon U, Wolf DH. The active sites of untreated multiple myeloma: an open-label phase 1/2 study. Lancet Oncol the eukaryotic 20 S proteasome and their involvement in subunit precursor 2014;15:1503–12. processing. J Biol Chem 1997;272:25200–9. 7. Reeder CB, Reece DE, Kukreti V, Mikhael JR, Chen C, Trudel S, et al. Long-term 19. Groll M, Heinemeyer W, Jager S, Ullrich T, Bochtler M, Wolf DH, et al. The survival with cyclophosphamide, bortezomib and dexamethasone induction catalytic sites of 20S proteasomes and their role in subunit maturation: a therapy in patients with newly diagnosed multiple myeloma. Br J Haematol mutational and crystallographic study. Proc Natl Acad Sci U S A 1999;96: 2014;167:563–5. 10976–83. 8. Stewart AK, Richardson PG, San-Miguel JF. How I treat multiple myeloma in 20. Kisselev AF, Callard A, Goldberg AL. Importance of the different proteolytic sites younger patients. Blood 2009;114:5436–43. of the proteasome and the efficacy of inhibitors varies with the protein substrate. 9. Kouroukis TC, Baldassarre FG, Haynes AE, Imrie K, Reece DE, Cheung MC. J Biol Chem 2006;281:8582–90. Bortezomib in multiple myeloma: systematic review and clinical considerations. 21. Oberdorf J, Carlson EJ, Skach WR. Redundancy of mammalian proteasome Curr Oncol 2014;21:e573–603. beta subunit function during endoplasmic reticulum associated degradation. 10. Guerrero-Garcia TA, Gandolfi S, Laubach JP, Hideshima T, Chauhan D, Biochemistry 2001;40:13397–405. Mitsiades C, et al. The power of proteasome inhibition in multiple myeloma. 22. Britton M, Lucas MM, Downey SL, Screen M, Pletnev AA, Verdoes M, et al. Expert Rev Proteomics 2018;15:1033–52. Selective inhibitor of proteasome's caspase-like sites sensitizes cells to specific 11. Laubach J, Garderet L, Mahindra A, Gahrton G, Caers J, Sezer O, et al. inhibition of chymotrypsin-like sites. Chem Biol 2009;16:1278–89. Management of relapsed multiple myeloma: recommendations of the interna- 23. Shi CX, Kortum KM, Zhu YX, Bruins LA, Jedlowski P, Votruba PG, et al. CRISPR tional myeloma working group. Leukemia 2016;30:1005–17. genome-wide screening identifies dependence on the proteasome subunit

AACRJournals.org Mol Cancer Res; 2020 OF11

Shi et al.

PSMC6 for bortezomib sensitivity in multiple myeloma. Mol Cancer Ther 2017; 37. Ruckrich T, Kraus M, Gogel J, Beck A, Ovaa H, Verdoes M, et al. Characterization 16:2862–70. of the ubiquitin-proteasome system in bortezomib-adapted cells. Leukemia 24. Barrio S, Stuhmer T, Da-Via M, Barrio-Garcia C, Lehners N, Besse A, et al. 2009;23:1098–105. Spectrum and functional validation of PSMB5 mutations in multiple myeloma. 38. Hess GT, Fresard L, Han K, Lee CH, Li A, Cimprich KA, et al. Directed evolution Leukemia 2019;33:447–56. using dCas9-targeted somatic hypermutation in mammalian cells. Nat Methods 25. Rothberg JM, Hinz W, Rearick TM, Schultz J, Mileski W, Davey M, et al. An 2016;13:1036–42. integrated semiconductor device enabling non-optical genome sequencing. 39. Soriano GP, Besse L, Li N, Kraus M, Besse A, Meeuwenoord N, et al. Proteasome Nature 2011;475:348–52. inhibitor-adapted myeloma cells are largely independent from proteasome 26. Kocher JP, Quest DJ, Duffy P, Meiners MA, Moore RM, Rider D, et al. The activity and show complex proteomic changes, in particular in redox and energy Biological Reference Repository (BioR): a rapid and flexible system for genomics metabolism. Leukemia 2016;30:2198–207. annotation. Bioinformatics 2014;30:1920–2. 40.HeinemeyerW,GruhlerA,MohrleV,MaheY,WolfDH.PRE2,highly 27. Robinson JT, Thorvaldsdottir H, Winckler W, Guttman M, Lander ES, Getz G, homologous to the human major histocompatibility complex-linked et al. Integrative genomics viewer. Nat Biotechnol 2011;29:24–6. RING10 gene, codes for a yeast proteasome subunit necessary for chrymo- 28. Graham FL, van der Eb AJ. A new technique for the assay of infectivity of human tryptic activity and degradation of ubiquitinated proteins. J Biol Chem 1993; adenovirus 5 DNA. Virology 1973;52:456–67. 268:5115–20. 29. Jordan M, Schallhorn A, Wurm FM. Transfecting mammalian cells: optimiza- 41. Hirano Y, Kaneko T, Okamoto K, Bai M, Yashiroda H, Furuyama K, et al. tion of critical parameters affecting calcium-phosphate precipitate formation. Dissecting beta-ring assembly pathway of the mammalian 20S proteasome. Nucleic Acids Res 1996;24:596–601. EMBO J 2008;27:2204–13. 30. Zhu YX, Shi CX, Bruins LA, Jedlowski P, Wang X, Kortum KM, et al. Loss of 42. Groettrup M, Standera S, Stohwasser R, Kloetzel PM. The subunits MECL-1 and FAM46C promotes cell survival in myeloma. Cancer Res 2017;77:4317–27. LMP2 are mutually required for incorporation into the 20S proteasome. 31. Kalari KR, Nair AA, Bhavsar JD, O'Brien DR, Davila JI, Bockol MA, et al. MAP- Proc Natl Acad Sci U S A 1997;94:8970–5. RSeq: Mayo analysis pipeline for RNA sequencing. BMC Bioinformatics 43. Griffin TA, Nandi D, Cruz M, Fehling HJ, Kaer LV, Monaco JJ, et al. Immu- 2014;15:224. noproteasome assembly: cooperative incorporation of interferon gamma (IFN- 32. Oerlemans R, Franke NE, Assaraf YG, Cloos J, van Zantwijk I, Berkers CR, et al. gamma)-inducible subunits. J Exp Med 1998;187:97–104. Molecular basis of bortezomib resistance: proteasome subunit beta5 (PSMB5) 44. Ettari R, Zappala M, Grasso S, Musolino C, Innao V, Allegra A. gene mutation and overexpression of PSMB5 protein. Blood 2008;112:2489–99. Immunoproteasome-selective and non-selective inhibitors: a promising 33. Franke NE, Niewerth D, Assaraf YG, van Meerloo J, Vojtekova K, van Zantwijk approach for the treatment of multiple myeloma. Pharmacol Ther 2018; CH, et al. Impaired bortezomib binding to mutant beta5 subunit of the 182:176. –92 proteasome is the underlying basis for bortezomib resistance in leukemia cells. 45. SantosRLA,BaiL,SinghPK,MurakamiN,FanH,ZhanW,etal.Structure Leukemia 2012;26:757–68. of human immunoproteasome with a reversible and noncompetitive inhib- 34. Lu S, Yang J, Song X, Gong S, Zhou H, Guo L, et al. Point mutation of the itor that selectively inhibits activated lymphocytes. Nat Commun 2017;8: proteasome beta5 subunit gene is an important mechanism of bortezomib 1692. resistance in bortezomib-selected variants of Jurkat T cell lymphoblastic lym- 46.KuhnDJ,HunsuckerSA,ChenQ,VoorheesPM,OrlowskiM,Orlowski phoma/leukemia line. J Pharmacol Exp Ther 2008;326:423–31. RZ. Targeted inhibition of the immunoproteasome is a potent strategy 35. Ri M, Iida S, Nakashima T, Miyazaki H, Mori F, Ito A, et al. Bortezomib-resistant against models of multiple myeloma that overcomes resistance to con- myeloma cell lines: a role for mutated PSMB5 in preventing the accumulation of ventional drugs and nonspecific proteasome inhibitors. Blood 2009;113: unfolded proteins and fatal ER stress. Leukemia 2010;24:1506–12. 4667–76. 36. Balsas P, Galan-Malo P, Marzo I, Naval J. Bortezomib resistance in a myeloma 47. Kincaid EZ, Che JW, York I, Escobar H, Reyes-Vargas E, Delgado JC, et al. Mice cell line is associated to PSMbeta5 overexpression and polyploidy. Leuk Res 2012; completely lacking immunoproteasomes show major changes in antigen pre- 36:212–8. sentation. Nat Immunol 2011;13:129–35.

OF12 Mol Cancer Res; 2020 MOLECULAR CANCER RESEARCH

Proteasome Subunits Differentially Control Myeloma Cell Viability and Proteasome Inhibitor Sensitivity

Chang-Xin Shi, Yuan Xiao Zhu, Laura A. Bruins, et al.

Mol Cancer Res Published OnlineFirst June 19, 2020.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-19-1026

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2020/06/19/1541-7786.MCR-19-1026.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/early/2020/08/01/1541-7786.MCR-19-1026. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.