A large body of evidence suggests that hypoxia drives aggressive molecular features of malignant cells irrespective of cancer type. Non-Hodgkin lymphomas (NHL) are the most common hematologic malignancies characterized by frequent involvement of diverse hypoxic microenvironments. We studied the impact of long-term deep hypoxia (1% O2) on the biology of lymphoma cells. Only 2 out of 6 tested cell lines (Ramos, and HBL2) survived ≥ 4 weeks under hypoxia. The hypoxia-adapted (HA)b Ramos and HBL2 cells had a decreased proliferation rate accompanied by significant suppression of both oxidative phosphorylation and glycolytic pathways. Transcriptome and proteome analyses revealed marked downregulation of genes and proteins of the mitochondrial respiration complexes I and IV, and mitochondrial ribosomal proteins. Despite the observed suppression of glycolysis, the proteome analysis of both HA cell lines showed upregulation of several proteins involved in the regulation of glucose utilization including the active catalytic component of prolyl-4-hydroxylase P4HA1, an important druggable oncogene. HA cell lines demonstrated increased transcription of key regulators of auto-/mitophagy, e.g., neuritin, BCL2 interacting protein 3 (BNIP3), BNIP3-like protein, and BNIP3 pseudogene. Adaptation to hypoxia was further associated with deregulation of apoptosis, namely upregulation of BCL2L1/BCL-XL, overexpression of BCL2L11/BIM, increased binding of BIM to BCL-XL, and significantly increased sensitivity of both HA cell lines to A1155463, a BCL-XL inhibitor. Finally, in both HA cell lines AKT kinase was hyperphosphorylated and the cells showed increased sensitivity to copanlisib, a pan-PI3K inhibitor. In conclusion, our data report on several shared mechanisms of lymphoma cell adaptation to long-term hypoxia including: 1. Upregulation of proteins responsible for glucose utilization, 2. Degradation of mitochondrial proteins for potential mitochondrial recycling (by mitophagy), and 3. Increased dependence on BCL-XL and PI3K-AKT signaling for survival. In translation, inhibition of glycolysis, BCL-XL, or PI3K-AKT cascade may result in targeted elimination of HA lymphoma cells.

Cellular senescence has recently been recognized as a significant contributor to the poor prognosis of glioblastoma, one of the most aggressive brain tumors. Consequently, effectively eliminating senescent glioblastoma cells could benefit patients. Human ADP/ATP translocases (ANTs) play a role in oxidative phosphorylation in both normal and tumor cells. Previous research has shown that the sensitivity of senescent cells to mitochondria‐targeted senolytics depends on the level of ANT2. Here, we systematically mapped the transcript and protein levels of ANT isoforms in various types of senescence and glioblastoma tumorigenesis. We employed bioinformatics analysis, targeted mass spectrometry, RT‐PCR, immunoblotting, and assessment of cellular energy state to elucidate how individual ANT isoforms are expressed during the development of senescence in noncancerous and glioblastoma cells. We observed a consistent elevation of ANT1 protein levels across all tested senescence types, while ANT2 and ANT3 exhibited variable changes. Alterations in ANT protein isoform levels correlated with shifts in the cellular oxygen consumption rate. Our findings suggest that ANT isoforms are mutually interchangeable for oxidative phosphorylation and manipulating individual ANT isoforms could have potential for senolytic therapy.

Background Mantle cell lymphoma (MCL) is a chronically relapsing malignancy with deregulated cell cycle progression. We analyzed efficacy, mode of action, and predictive markers of susceptibility to palbociclib, an approved CDK 4/6 inhibitor, and its combination with venetoclax, a BCL2 inhibitor. Methods A panel of nine MCL cell lines were used for in vitro experiments. Four patient derived xenografts (PDX) obtained from patients with chemotherapy and ibrutinib-refractory MCL were used for in vivo proof-of-concept studies. Changes of the mitochondrial membrane potential, energy-metabolic pathways, AKT activity, and pro-apoptotic priming of MCL cells were evaluated by JC-1 staining, Seahorse XF analyser, genetically encoded fluorescent AKT reporter, and BH3 profiling, respectively. MCL clones with gene knockout or transgenic (over)expression of CDKN2A, MYC, CDK4 , and RB1 were used to estimate impact of these aberrations on sensitivity to palbociclib, and venetoclax. Results Co-targeting MCL cells with palbociclib and venetoclax induced cytotoxic synergy in vitro and in vivo. Molecular mechanisms responsible for the observed synthetic lethality comprised palbociclib-mediated downregulation of anti-apoptotic MCL1, increased levels of proapoptotic BIM bound on both BCL2, and BCL-XL and increased pro-apoptotic priming of MCL cells mediated by BCL2-independent mechanisms, predominantly palbociclib-triggered metabolic and mitochondrial stress. Loss of RB1 resulted in palbociclib resistance, while deletion of CDKN2A or overexpression of CDK4 , and MYC genes did not change sensitivity to palbociclib. Conclusions Our data strongly support investigation of the chemotherapy-free palbociclib and venetoclax combination as an innovative treatment strategy for post-ibrutinib MCL patients without RB1 deletion.

Besides many other mutations in known cancer driver genes, mantle cell lymphoma (MCL) is characterized by recurrent genetic alterations of important regulators of the phosphoinositol-3-kinase (PI3K) cascade including PIK3CA gains and PTEN losses. To evaluate the biological and functional consequences of these aberrations in MCL, we have introduced transgenic expression of PIK3CA (PIK3CA UP) and performed knockout/knockdown of PTEN gene (PTEN KO / KD) in 5 MCL cell lines. The modified cell lines were tested for associated phenotypes including dependence on upstream B-cell receptor (BCR) signaling (by an additional BCR knockout). PIK3CA overexpression decreased the dependence of the tested MCL on prosurvival signaling from BCR, decreased levels of oxidative phosphorylation, and increased resistance to 2-deoxy-glucose, a glycolysis inhibitor. Unchanged AKT phosphorylation status and unchanged sensitivity to a battery of PI3K inhibitors suggested that PIK3CA gain might impact MCL cells in AKT independent manner. PTEN KO was associated with a more distinct phenotype: AKT hyperphosphorylation and overactivation, increased resistance to multiple inhibitors (most of the tested PI3K inhibitors, BTK inhibitor ibrutinib, and BCL2 inhibitor venetoclax), increased glycolytic rates with resistance to 2-deoxy-glucose, and significantly decreased dependence on prosurvival BCR signaling. Our results suggest that the frequent aberrations of the PI3K pathway may rewire associated signaling with lower dependence on BCR signaling, better metabolic and hypoxic adaptation, and targeted therapy resistance in MCL.

Venetoclax (VEN), a B-cell lymphoma 2 (BCL2) inhibitor, has a promising single-agent activity in mantle cell lymphoma (MCL), acute lymphoblastic leukemia (ALL), and large BCLs, but remissions were generally short, which call for rational drug combinations. Using a panel of 21 lymphoma and leukemia cell lines and 28 primary samples, we demonstrated strong synergy between VEN and A1155463, a BCL-XL inhibitor. Immunoprecipitation experiments and studies on clones with knockout of expression or transgenic expression of BCL-XL confirmed its key role in mediating inherent and acquired VEN resistance. Of note, the VEN and A1155463 combination was synthetically lethal even in the cell lines with lack of expression of the proapoptotic BCL2L11/BIM and in the derived clones with genetic knockout of BCL2L11/BIM. This is clinically important because BCL2L11/BIM deletion, downregulation, or sequestration results in VEN resistance. Immunoprecipitation experiments further suggested that the proapoptotic effector BAX belongs to principal mediators of the VEN and A1155463 mode of action in the BIM-deficient cells. Lastly, the efficacy of the new proapoptotic combination was confirmed in vivo on a panel of 9 patient-derived lymphoma xenografts models including MCL (n = 3), B-ALL (n = 2), T-ALL (n = 1), and diffuse large BCL (n = 3). Because continuous inhibition of BCL-XL causes thrombocytopenia, we proposed and tested an interrupted 4 days on/3 days off treatment regimen, which retained the desired antitumor synergy with manageable platelet toxicity. The proposed VEN and A1155463 combination represents an innovative chemotherapy-free regimen with significant preclinical activity across diverse BCL2+ hematologic malignancies irrespective of the BCL2L11/BIM status.

Figure S1: Endocrine-resistant MCF-7 cells demonstrate sensitivity to TRAIL Figure S2: Endocrine-resistant CSCs demonstrate sensitivity to TRIAL. Figure S3: Primary-derived patient samples. Figure S4: Endocrine-resistant primary and PDX tumour cells are sensitive to TRAIL. Figure S5: Mechanism of endocrine-resistant tumour cell sensitivity to TRAIL.

(A) Comparison of the extent of apopotosis induced by HHT (after 24hrs exposure) in HHT-resistant (HHT-R) clones compared to the original cell lines (CTRL). (B) Western blot analysis demonstrates upregulation of BCL-XL protein in HHT-R clones compared to CTRL cell lines.

Large body of evidence suggests that hypoxia drives aggressive molecular features irrespective of cancer type. Non-Hodgkin lymphomas (NHL) are the most common hematologic malignancies characterized by widespread disease and frequent involvement of diverse hypoxic microenvironments (e.g., bone marrow, malignant effusions, large lymphoma masses). Impact of long-term hypoxia on the biology of lymphoma cells and its potential role in mediating drug resistance and disease relapses due to adaptive changes shaped by the hypoxic microenvironment remain only partially understood. In this study we analyzed impact of short- and long-term hypoxia on a panel of six lymphoma cell lines including diffuse large B-cell lymphoma (UPF4D, UPF8D), Burkitt lymphoma (Ramos, UPF9T), and mantle cell lymphoma (HBL2, MINO). Only 2 out of 8 tested lymphoma cell lines (Ramos, and HBL2) survived longer than 4 weeks under deep hypoxia (1% O2). The remaining tested lymphoma cell lines died by hypoxia-induced apoptosis (which could be inhibited by co-cultivation with pan-caspase inhibitor Z-VAD-FMK). The hypoxia-adapted (HA) Ramos and HBL2 cells had severely decreased proliferation rate accompanied by complex changes of the transcriptome, proteome, and metabolome. Seahorse analysis revealed significant inhibition of both oxidative phosphorylation and glycolytic pathways ( Figure 1). Of note, the whole-proteome profiling confirmed significant downregulation of proteins regulating oxidative phosphorylation, but significant upregulation of proteins regulating glycolysis. Sensitivity of HA cells to 2-deoxyglucose, an inhibitor of glycolysis was markedly increased ( Figure 1). The data suggest that under long-term deep hypoxia, lymphoma cells try to compensate for the decrease in ATP production from the oxidative phosphorylation process by boosting structural machinery of glycolysis, on which they became vitally dependent for survival. Despite the increased glycolytic machinary, however, the level of glycolysis decreases due to severe lack of oxygen. In our opinion, this is a clear representation of the Warburg effect, a hallmark of cancer based on the metabolic reprogramming of the cancer cells under hypoxia. Sensitivity of HA cells to a panel of the tested cytotoxic drugs (cytarabin, cisplatin, bortezomib) and targeted agents (venetoclax, S63545, A1155463, TRAIL, polatuzumab vedotin, IM-156, copanlisib) was either unchanged or increased (compared to sensitivity of the original cell lines cultured under normoxia). In contrast to so far published data, long-term hypoxia was not associated with acquired drug resistance in case of any of the tested agent. Of note, HA cells became significantly more sensitive to A1155463, an inhibitor of anti-apoptotic protein BCL-XL ( Figure 1). Unchanged levels of BCL-XL protein in HA cells and the respective normoxic cell lines (detected by western blotting) suggest that more complex mechanisms plausibly underlie the observed markedly increased sensitivity to A1155463. The whole-transcriptome and proteome analyses of both HA cell lines (compared to the parental cell lines cultured under normoxic conditions) detected among other complex changes signficant upregulation of P4HA1 (both mRNA, and protein), a prolyl hydroxylase involved in the stabilization of hypoxia-induced factor (HIF) 1 alpha, and putative oncogene associated with tumor aggressiveness. Transgenic (over)expression of P4HA1 in hypoxia-sensitive MINO cells effectively inhibited the hypoxia-induced apoptosis. Western blot analysis implemented on a panel of primary lymphoma cells revealed markedly higher expression of P4HA1 protein in the lymphoma cells obtained from malignant effusions (putative hypoxic microenvironment) compared to lymphoma cells obtained from leukemized blood (putative normoxic microenvironment). Our data suggest that P4HA1 positively impacts survival of lymphoma cells under hypoxia. In translation, P4HA1, BCL-XL, and structural proteins of the glycolytic pathway may represent novel drugable targets for more effective elimination of hypoxia-adapted lymphoma cells. Financial Support: Ministry of Health of the Czech Republic AZV NU23-03-00172, Grant Agency of the Czech Republic GA23-05377S, and National Institute for Cancer Research (EXCELES) LX22NPO5102.

Downregulation of MYC, cFLIP or BCL6 in OCI-Ly7 and BJAB DLBCL cell lines (by western blot) after exposure to HHT.

Western blot analysis showing protein expression profile of key regulators of apoptosis in 18 DLBCL cell lines (12 germinal center B-cell-like (GCB) and 6 activated B-cell-like (ABC) DLBCL cell lines). *UPF4D cell line (GCB origin) was derived in our laboratory.

Cell of origin (COO), BCL2 expression status and IC100 for ABT-199 (µM) and HHT (nM) are shown.

To investigate whether the expression level of BCL2 proteins detected by western blot is comparable with the expression level assessed by IHC analysis, we analyzed protein expression (by IHC) of BCL2, MCL1 and BCL-XL from subcutaneous lymphoma xenografts obtained from mice xenotransplanted with selected DLBCL cell lines with known protein expression (by western blot).

Viability of DLBCL cell lines cultured with different concentrations of HHT (nM) or ABT-199 (µM) was measured by WST-8 survival/proliferation assay for 7-14 days. Graphs indicate % of viable cells, i.e. % of maximal absorbance of untreated cells shown with dashed lines. Representative example of two independent experiments is shown. Standard deviations were < 5% for all measurements.

Shown are examples of different levels of semi-quantitative protein expression (0-3).

The graphs show the extent of apoptosis of DLBCL cell lines induced by different concentrations of HHT, ABT-199 and ABT-737 after 24hrs exposure to particular agents.

Supplemental Figure 1. Next-generation sequencing of PDXs and primary MCL cells Supplemental Figure 2. IHC analysis of primary MCL samples and murine xenografts Supplemental Figure 3. Array comparative genomic hybridization of 24 primary MCL samples Supplemental Figure 4. Sensitivity of HBL2 and MAVER-1 cells resistant to venetoclax to BCL-XL inhibitors WEHI-539 and A1155463 Supplemental Table 1. Complete list of protein coding variants Supplemental Table 2. FISH analyses of the established murine PDXs of MCL Supplemental Table 3. Baseline characteristics of patients Supplemental Table 4. IHC analysis of MCL samples and xenografts Supplemental Table 5. Tumor and spleen weights at the end of in vivo experiments

Proteins from the Bcl-2 family play an essential role in the regulation of apoptosis. However, they also possess cell death-unrelated activities that are less well understood. This prompted us to study apoptosis-unrelated activities of the Bax and Bak, pro-apoptotic members of the Bcl-2 family. We prepared Bax/Bak-deficient human cancer cells of different origin and found that while respiration in the glioblastoma U87 Bax/Bak-deficient cells was greatly enhanced, respiration of Bax/Bak-deficient B lymphoma HBL-2 cells was slightly suppressed. Bax/Bak-deficient U87 cells also proliferated faster in culture, formed tumours more rapidly in mice, and showed modulation of metabolism with a considerably increased NAD+/NADH ratio. Follow-up analyses documented increased/decreased expression of mitochondria-encoded subunits of respiratory complexes and stabilization/destabilization of the mitochondrial transcription elongation factor TEFM in Bax/Bak-deficient U87 and HBL-2 cells, respectively. TEFM downregulation using shRNAs attenuated mitochondrial respiration in Bax/Bak-deficient U87 as well as in parental HBL-2 cells. We propose that (post)translational regulation of TEFM levels in Bax/Bak-deficient cells modulates levels of subunits of mitochondrial respiratory complexes that, in turn, contribute to respiration and the accompanying changes in metabolism and proliferation in these cells.

Dear Editor, Cancer is a pathology still on the rise,1 with unmet need for efficient therapy, owing to factors such as considerable differences in mutational signature in the same patient in primary tumours and proximal/distal metastases, shown, for example, for renal cancer.2 What is needed then is an invariant target predominantly only affected by drugs in cancer cells. A thus far untested approach is targeting mitochondrial respiration using compounds from the group of mitocans,3 epitomised by mitochondrially targeted tamoxifen (MitoTam), that is, tamoxifen tagged with the mitochondrial vector triphenylphosphonium (TPP) (Figure S1A; see also Supporting Information for description of synthesis).4,5 This strategy is based on the premise that cancer cells differ from their non-cancerous counterparts,3 making them selectively vulnerable to TPP-tagged anti-cancer agents,6 and on the premise that mitochondrial function is vital for tumour progression.7,8 Wehave recently conducted Phase 1/1bMitoTamclinical trial for metastatic solid tumour patients, with all patients undergoing palliative therapy after exhaustion of established therapeutic regimens (MitoTam-01 trial; EudraCT 2017-004441-25). Although the Phase 1/1b clinical trial will be published in its entirety elsewhere, of the individual types of cancer, the greatest benefit was found for clear cell renal cancer patients represented here by two subjects (Tables S1 and S2). These patients underwent three and four rounds of MitoTam therapy, respectively, at 1 mg/kg three times per week followed by a week of rest, totalling four such cycles, with one patient showing tumour stabilisation and the other partial remission (Figure 1A). The trial revealed excellent safety profile of MitoTam, with only occasional grade 1 toxicity. The high efficiency for renal cancer was found to correlate with the highest level of MitoTam and its metabolites reached in kidneys (Figures 1B, S1B and S1C), being excreted via bile (Figure S1D).

Mantle cell lymphoma (MCL) is a rare chronically relapsing subtype of aggressive B-cell non-Hodgkin lymphoma characterized by the canonical chromosomal translocation t (11;14) and other molecular cytogenetic aberrations including overexpression of BCL2 protein. Venetoclax (VEN), a BCL2 inhibitor, has demonstrated activity in MCL both as a monotherapy and in combination with other targeted agents. Treatment with single agent venetoclax is, however, hampered by frequent development of drug resistance caused in large part by adaptive upregulation of other anti-apoptotic BCL2 family members. Immunodeficient NOD-SCID-gamma mice were xenografted subcutaneously with MCL cell lines and PDX cells established from patients with relapsed and refractory MCL. The therapy was given for two weeks. VEN was administered by oral gavage (100 mg/kg/d), A1155463 was administered intraperitoneally (10 mg/kg/d). Western blotting was implemented to evaluate expression levels of BCL2 proteins, immunoprecipitation was used to analyze the levels of BIM bound to BCL2 and BCL-XL before and after exposure to VEN. CRISPR-Cas9 was employed to derive clones with BCL-XL knock-out. Blood cell counts in mice on A1155463 +/- VEN therapy were analyzed using the Mindray BC-5300 Auto Hematology Analyzer. In this study, we analyzed molecular mechanisms of in vivo acquired resistance to VEN using a panel of several murine cell line-based xenografts (CDX) and patient-derived xenografts (PDX) of MCL. First, the mice xenotransplanted with VEN-sensitive MCL cells were subject to monotherapy with VEN until development of resistant tumors (VEN-R). Western blot analysis of VEN-R tumors revealed upregulation of BCL-XL in majority of VEN-R tumors compared to controls. Other changes included upregulation of MCL1 and downregulation of BIM proteins in several models. Immunoprecipitation experiments confirmed that BCL-XL indeed serves as a buffer for BIM released from BCL2 after exposure to VEN thereby blocking VEN-triggered apoptosis. Importantly, we demonstrated that the upregulation of BCL-XL caused not only VEN resistance, but also led to BCL-XL-specific pro-apoptotic priming of VEN-R lymphoma cells. In vitro, the combination of VEN and A1155463, a specific BCL-XL inhibitor, induced cytotoxic synergy on a panel of MCL cell lines and primary cells. CRISPR-Cas9-mediated BCL-XL knock-out resulted in marked sensitization to VEN-induced apoptosis in MCL cell lines. In vivo, pharmacological blockage of BCL-XL strongly increased sensitivity to VEN. Despite that A1155463 exerted limited anti-lymphoma activity as monotherapy on tested CDX and PDX models, its combination with VEN was synthetically lethal and exerted significantly enhanced anti-lymphoma activity. The efficacy of the VEN and A1155463 combination was highly effective even in mice bearing VEN-R tumors (tumors with acquired VEN resistance, Figure 1). It has been published that mechanisms responsible for BCL-XL upregulation include activation of NFkappaB signaling via CD40 and hypoxia. Upregulation of BCL-XL induced by these factors in turn caused VEN resistance in vivo. Indeed, we confirmed that upregulation of BCL-XL was observed already after engraftment of VEN-sensitive MCL cells in immunodeficient mice. MCL cells isolated ex vivo from the established CDX tumors were significantly more sensitive to BCL-XL inhibition compared to the corresponding in vitro growing cell lines. Our results thus confirmed increased BCL-XL dependence of MCL cells in vivo compared to in vitro. We suggest that these microenvironmental factors were also critical for the selection of VEN-resistant clones during the therapy with VEN. Historically, experimental therapy of patients with chronic lymphocytic leukemia with navitoclax, a combined inhibitor of BCL2 and BCL-XL, was rater disappointing. The plausible reasons included both low efficacy of navitoclax due to insufficient BCL2 inhibition, and a dose-limiting thrombocytopenia. We demonstrated that the thrombocytopenia associated with continued therapy with A1155463 could be managed by 4 days on / 3 days off treatment strategy, which cannot be applied in case of fixed dual BCL2/BCL-XL inhibitors like navitoclax, or AZD4320. In summary, the combined inhibition of BCL2 and BCL-XL with VEN and A1155463 is a highly effective experimental treatment strategy for R/R MCL with a potential translation to the clinical grounds. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal