ALBERT

Description: Mantle Cell Lymphoma (MCL) is a malignancy of mature B-cells. MCL has a poor prognosis and a limited response to traditional chemotherapy. Bortezomib (BZM), a new powerful inhibitor of the proteasome, can induce responses in up to 50% of relapsed MCL patients, suggesting that in at least half of the patients the lymphoma cells are intrinsically resistant to BZM or rapidly develop resistance during single agent therapy. To investigate possible mechanisms of BZM resistance, we cultured MCL cell lines continuously in sub-lethal concentrations of BZM that were then gradually increased. Resistance was slow to develop taking several months for truly resistant clones to grow out. We generated a bortezomib resistant (BR) clone of HBL-2 with an IC50 of 30nM compared to 5nM in the parental clone and several BR subclones of Jeko-1, the most resistant of which had an IC50 of 200nM compared to 3nM for the parental clone. All BR subclones also showed decreased sensitivity to three other proteasome inhibitors: MG-132, Lactacystin, and NLVS. The increase in IC50 to these drugs was between 3 and 8-fold, consistent with more off-target effects of these drugs compared to BZM. BAY11-7082, an inhibitor of NF-kB signaling, maintained its activity against the resistant cells. Resistance to BZM, once acquired, has remained stable over several months. This is remarkable because the resistant subclones grow significantly slower than the parental lines, even after having been removed from selection for extended periods of time. Consistent with slower cell proliferation, we found reduced Cyclin D1 protein expression in the BZM resistant Jeko clones; however, mRNA levels were comparable to the parental line, indicating that changes in Cyclin D1 protein translation and/or stability may be responsible for the decreased proliferation. BZM resistance has been associated with up-regulation of proteasome components and heat-shock proteins. Indeed, in the resistant HBL-2 subclone we found marked upregulation of two proteasome components (PSMA5 and PSMC1) and of Hsp70 by RT-PCR, but there was only a small change in Hsp70 protein expression. Nevertheless, upregulation of these genes could be part a more global gene expression response as seen with ER-stress and could thus reflect an adaptive change to BZM in the HBL-2 BR subclone. All three Jeko BR clones in contradistinction showed only minor changes in PSMA5, PSMC1 and Hsp70 mRNA expression and surprisingly had markedly reduced Hsp70 protein levels. Thus, in these subclones, BR resistance appears to correlate primarily with changes affecting cell cycle control. We conclude that resistance to BZM may be determined by several mechanisms that affect cell cycle control as well as expression of proteasome components and heat-shock proteins. While the slow development of resistance suggests adaptive changes, its persistence over time is more consistent with mutations or other genomic alterations that are not readily reversible. Ongoing studies aim to more precisely define the basis for BZM resistance in MCL.

Description: Bortezomib induces remissions in 30%-50% of patients with relapsed mantle cell lymphoma (MCL). Conversely, more than half of patients' tumors are intrinsically resistant to bortezomib. The molecular mechanism of resistance has not been defined. We generated a model of bortezomib-adapted subclones of the MCL cell lines JEKO and HBL2 that were 40- to 80-fold less sensitive to bortezomib than the parental cells. Acquisition of bortezomib resistance was gradual and reversible. Bortezomib-adapted subclones showed increased proteasome activity and tolerated lower proteasome capacity than the parental lines. Using gene expression profiling, we discovered that bortezomib resistance was associated with plasmacytic differentiation, including up-regulation of IRF4 and CD38 and expression of CD138. In contrast to plasma cells, plasmacytic MCL cells did not increase immunoglobulin secretion. Intrinsically bortezomib-resistant MCL cell lines and primary tumor cells from MCL patients with inferior clinical response to bortezomib also expressed plasmacytic features. Knockdown of IRF4 was toxic for the subset of MCL cells with plasmacytic differentiation, but only slightly sensitized cells to bortezomib. We conclude that plasmacytic differentiation in the absence of an increased secretory load can enable cells to withstand the stress of proteasome inhibition. Expression of CD38 and IRF4 could serve as markers of bortezomib resistance in MCL. This study has been registered at http://clinicaltrials.gov as NCT00131976.

Description: Introduction: BTK is involved in B-cell receptor (BCR) signal transduction and is an established target for the treatment of chronic lymphocytic leukemia (CLL) (Byrd, NEJM, 2013). ACP-196 is a novel, potent second generation BTK inhibitor, which binds covalently to Cys481 in the ATP-binding pocket of BTK. IC50determinations on nine kinases with a cysteine in the same position as BTK showed ACP-196 to be more selective than the first-in-class BTK inhibitor, ibrutinib (Covey, AACR, 2015). We present data evaluating the anti-tumor effects of ACP-196 in established murine models of CLL. Methods: Two distinct murine models were used for these studies. In the TCL1 adoptive transfer model, leukemic cells from Eμ-TCL1 transgenic mice were transplanted into C57BL/6 mice, resulting in a CD5+/CD19+ leukemia with peripheral blood, spleen and nodal involvement. ACP-196 treatment in drinking water (0.16 mg/mL) commenced when recipient mice had 〉 10% CD5+/CD19+ leukocytes in the peripheral blood. Mice were followed for survival. Separate cohorts were sacrificed for pharmacodynamic analyses after 1 and 4 weeks of treatment. In the second model, NSG mice received primary human CLL cells. The xenografted human CLL cells have comparable tumor biology (including active BCR signaling) to activated human lymph node resident CLL cells (Herman, Leukemia, 2013). PBMCs harvested from CLL patients were adoptively transferred at 1 x 108 cells per mouse. ACP-196 was initiated on day -1 (at the time of busulfan priming) at multiple doses ranging from 0.006 to 0.3 mg/mL in drinking water. Results: In the TCL1 model, treatment with ACP-196 showed 〉 90% BTK occupancy of BTK after 1 and 4 weeks of therapy. ACP-196 inhibited BCR signaling as shown by decreased autophosphorylation of BTK and reduction in surface expression of the BCR activation markers CD86 and CD69. After 1 week of ACP-196 inhibited BCR signaling as shown by a 6-fold reduction of autophosphorylation of BTK in the presence of anti-IgM, and surface expression of the BCR activation markers CD69 and CD86 were decreased by 47% and 57% respectively. Inhibition of BTK and downstream BCR activation was maintained through at least day 28 of treatment. Most notably, ACP-196 treatment resulted in a significant increase in survival compared with mice receiving vehicle (median 81 vs 59 days, respectively; P =0.02). In the NSG xenograft model, ACP-196 at the times examined did not cause a significant treatment-induced lymphocytosis in the patients evaluated (n=6). After 4 weeks of treatment with ACP-196, the NSG mice were sacrificed, and BCR signaling activity and tumor burden in the spleen were evaluated. ACP-196 treatment showed decreases in phosphorylation of PLCγ2 and ERK (P

Description: Emergency granulopoiesis refers to the increased production of neutrophils in bone marrow and their release into circulation induced by severe infection. Several studies point to a critical role for granulocyte colony-stimulating factor (G-CSF) as the main mediator of emergency granulopoiesis. However, the consequences of G-CSF stimulation on the transcriptome of neutrophils and their precursors have not yet been elucidated in humans. Here, we investigate the changes in mRNA and miRNA expression in successive stages of neutrophil development following in vivo administration of G-CSF in humans, mimicking emergency granulopoiesis. Blood samples were collected from healthy individuals after five days of G-CSF administration. Neutrophil precursors were sorted into discrete stages of maturation by flow cytometry and extracted RNA was subjected to microarray analysis. mRNA levels were compared to previously published expression levels in corresponding populations of neutrophil precursors isolated from bone marrow of untreated, healthy individuals. miRNA expression was investigated in the most mature cell population to determine G-CSF-induced changes in circulating neutrophils. G-CSF substantially affected mRNA and miRNA expression patterns, demonstrating significant impact on neutrophil development and function. 1110 mRNAs were differentially expressed more than 2-fold with G-CSF while the treatment induced changes in the levels of 73 miRNAs in the mature population. In addition, G-CSF treatment reduced the levels of four out of five measured granule proteins in mature neutrophils including hCAP-18, which was completely deficient in neutrophils from G-CSF-treated donors. Cell cycle analysis pointed towards an induced proliferative capacity of myelocytes. These results indicate that multiple biological processes are altered in order to satisfy the increased demand for neutrophils during G-CSF-induced emergency granulopoiesis. Disclosures No relevant conflicts of interest to declare.

Description: Acute Myeloid Leukemia (AML) is a form of cancer that affects the cells of the hematopoietic system. The currently applied WHO classification of genetically defined AML subtypes based on cytogenetic and molecular aberrations of known oncogenes and tumor suppressors allows to stratify patients into genetically defined low-, intermediate, and high-risk groups. Yet, this detailed genetic information has rarely resulted in the development of targeted therapies that are directed at the underlying genetic lesions, or even personalized therapies. With the notable exceptions of chronic myeloid leukemia (CML) and acute promyelocytic leukemia (APL), the prognosis for most AML subtypes treated with standard chemotherapy regimens remains for the most part poor with a median overall survival of approximately 18 months. The development of cancer may have both genetic and epigenetic causes, but what is common to all cancers is a unique gene-expression pattern (GEP) that drives the malignant phenotype and thus partly differs from that of normal cells representing the same type of cells/tissue and developmental stage. Significantly the unique GEP of cancer cells have been illustrated in a multitude of gene-expression profiling studies where mRNA were quantified using microarray or deep sequencing platforms. These technologies have been outstanding to group, diagnose and forecast AML subtypes based on differences in GEPs. However, they have only demonstrated a marginal role in the identification of potential therapeutic targets. This is mainly due to the fact that these studies have compared AMLs with other AMLs and not with corresponding normal cells of the same developmental stages. This is an obvious prerequisite if one wants to define the genes that drives the malignant phenotype of AML cells. Our hypothesis is that a comparison of the GEPs of AML cells with their closest normal counterparts within the hematopoietic hierarchy will provide a better understanding of AML biology, which we could then use to improve prognostication and as a starting point for the development of targeted therapeutic strategies including combinations of both new drugs or existing ones. To test this hypothesis, we FACS sorted hematopoietic stem and progenitors and generated micro-array based GEPs. Together with other normal hematopoietic cell GEPs derived from public repositories, we generated a gene expression-based map of the hematopoietic system (Frederik Otzen Bagger et al., HemaExplorer: a database of mRNA expression profiles in normal and malignant haematopoiesis., Nucleic acids research (2012); Frederik Otzen Bagger et al., HemaExplorer: a Web server for easy and fast visualization of gene expression in normal and malignant hematopoiesis., 119 Blood 6394–6395 (2012)) , using the first components of a principal component analysis (PCA). Once established, we mapped the AML-patient samples GEP (〉1300) into this PCA space to identify the normal populations closest to each individual AML GEP (see Figures 1A and 1B). Finally, gene-expression changes between individual AML samples and their corresponding individual normal counterparts were computed for further analyses, such as gene set enrichment and survival analysis.Figure 1(A, B) PCA plots of individual GEP profiles from two AML subtypes mapped to the normal landscape of hematopoietic differentiation. (C-E) Stratification of three independent patient cohorts with survival gene signature generated by comparison of cancer and direct normal counterpart.Figure 1. (A, B) PCA plots of individual GEP profiles from two AML subtypes mapped to the normal landscape of hematopoietic differentiation. (C-E) Stratification of three independent patient cohorts with survival gene signature generated by comparison of cancer and direct normal counterpart. Our method resulted in improved clinical prognostication of AML patients and enabled us to identify new potential therapeutic target genes and oncogenic pathways activity that are over-expressed in AML cells as compared to their normal counterparts. Similarly, using the connectivity-map database (a collection of genome-wide transcriptional expression data from cultured human cells treated with bioactive small molecules), we could screen substances that could reverse the aberrant gene-expression signatures; some of these drugs are already in use in AML treatment or in clinical trials, which strengthens the validity of our bioinformatics method. A singularly interesting perspective in this regard is the possibility to tailor individualized drug combinations potentially targeting cancer cells synergistically by combined inhibition of multiple oncogenic targets and pathways. Disclosures: No relevant conflicts of interest to declare.

Description: Key Points miRNA-130a is expressed in myeloblasts and promyelocytes and inhibits translation of CEBPE mRNA encoding transcription factor C/EBP-ε. Regulation of CEBPE mRNA by miRNA-130a is required for timed expression of secondary granule proteins and cell cycle exit.

Description: Abstract 3775 Olfactomedin 4 (OLFM4) was initially identified as a gene highly induced in myeloid stem cells by G-CSF treatment and independently as a gene highly expressed in colon cancers. OLFM4 was predicted in a bioinformatics analysis as associated with neutrophil specific granules. We analyzed the expression of OLFM4 mRNA in myeloid cells from normal human bone marrow and demonstrated that expression of OLFM4 mRNA is similar to the expression of LCN2 which codes for the specific granule protein NGAL (Figure 1), but distinct from expression of mRNA for myeloperoxidase and gelatinase which are marker proteins for azurophil granules and gelatinase granules, respectively. Subcellular fractionation of peripheral blood neutrophils demonstrated complete co-localization of OLFM4 with NGAL, and stimulation of neutrophils with fMLP or PMA resulted in co-release of NGAL and OLFM4, indirectly proving that OLFM4 is a genuine constituent of neutrophil specific granules. Figure 1. mRNA expression profiles for OLFM4 and LCN2 in populations enriched in myeloblasts/promyelocytes (MB/PM), myelocytes/metamyelocytes (MY/MM), banded cells/segmented cells (BC) and peripheral blood neutrophils (pb-PMN) normalized to ACTB. Figure 1. mRNA expression profiles for OLFM4 and LCN2 in populations enriched in myeloblasts/promyelocytes (MB/PM), myelocytes/metamyelocytes (MY/MM), banded cells/segmented cells (BC) and peripheral blood neutrophils (pb-PMN) normalized to ACTB. Interestingly, immunohistochemistry showed OLFM4 expression in only a subset of neutrophils (figure 2). We suspected that this might be dependent on the antibody, but two different commercial antibodies and an in-house antibody raised against a synthetic OLFM4 derived peptide, all polyclonal, showed similar patterns. Flow cytometry confirmed the existence of two populations of neutrophils, one expressing OLFM4 the other not. Figure 2. Immunohistochemistry of OLFM4 in neutrophils. Figure 2. Immunohistochemistry of OLFM4 in neutrophils. Immunohistochemistry of bone marrow cells showed that OLFM4 appears in myelocytes and is maintained in the cells during further maturation of the cells to segmented neutrophils. Again, only 30% of the neutrophil precursors from bone marrow stain positive for OLFM4 indicating, that different subsets of human neutrophils may exist. Disclosures: No relevant conflicts of interest to declare.

Description: Introduction Cancer emerges as a consequence of multiple genetic aberrations that ultimately cause global changes in gene expression driving the malignant phenotype. Hence, identification of aberrantly expressed genes in cancer cells, as compared to their corresponding normal cells, provides information on the biology of the cancer as well as potential targets for therapeutic interventions. Here, we report an access-free internet platform for visualization of mRNA expression profiles in acute myeloid leukemia (AML) patient samples as compared to normal bone marrow (BM) populations representing successive stages of differentiation along the myeloid differentiation pathway. The internet platform allows for quick visualization of user selected genes in AML patients in either human or murine hematopoietic stem cells (HSC), myeloid hematopoietic progenitor cells (HPC) and their progeny, as well as mature cells of innate and adaptive immune system. Significantly, users can upload own microarray data to compare gene expression in their samples of interest to those available at the internet platform. Overall, our internet platform represents a powerful tool for studies of normal hematopoietic development as well as aberrantly expressed genes in AML and potentially other hematological malignancies. The web tool will be made available at http://servers.binf.ku.dk/hemaexplorer/ Methods Raw Affymetrix microarray data from our own repository and public available databases were normalized and batch corrected to build an integrated gene expression database for a series of normal hematopoietic cells and AML patient samples, at progressing stages of differentiation, which can be visualized directly or compared to external samples added by the user. A complete database of internal and user-supplied samples is build at each run of the analysis, using uniform transformation and correction parameters adjusted to ensure full integrity and comparability across all samples. The microarray database includes 44 highly purified sorted human normal blood samples and BM populations including HSCs, myeloid HPCs and their progeny, as well as mature cells of the innate and adaptive immune system. 10 samples were from public sources. In addition, our database includes microarray data from multiple AML studies (〉 1000 samples across platforms) allowing for comparison of gene expression in WHO defined AML subclasses and normal hematopoietic cells. Results The main strength of the data-driven HemaExplorer 2.0 tool is the ability to quickly assess the main trends in how an AML sample diverges from normal cells during normal myeloid development. This can be accessed by specifying a gene of interest (Figure panel A-B. A user-supplied sample is marked in red in panel B) which provides interactive plots and hierarchical visualisations of gene expression where parameters including gene of interest, data source, and cell types can be selected and presented directly on the output figures usable for publications. Furthermore, a principal component analysis (PCA) plot can be performed, which allows for an unsupervised gene expression based mapping of user provided samples to the most closely related normal cell populations (figure panel C - here shown with cell lines commonly used in AML research). Discussion Several websites offer visualisation of gene expression in cells from the hematopoietic system, including Gene Expression Atlas (Nucl. Acids Res., 40, D1077–D1081) and our own HemaExplorer (Blood, 119(26), 6394-5 and Nucl. Acids Res. 41, D1034-D1039), without the possibility of external sample addition provided by the users. Gene Expression Commons (PLoS ONE 7(7), e40321) provides this option for public data, based on a common expression model. Here we offer the ability to add and compare unpublished microarray gene expression profiles to various types of normal blood and BM populations, in a ready-to-use platform normalized on single sample level, independently on the overall distribution of expression in a selected set of reference samples. Future efforts will include the ability to provide a standard differential expression analysis, comparing the user-supplied sample with a selection of the normal hematopoietic hierarchy. Furthermore, we aim to be able to offer a standalone version of the tool, which can be implemented internally on hospital servers, and expand the list of accepted microarray platforms. Disclosures: No relevant conflicts of interest to declare.

Description: Mantle cell lymphoma (MCL) is an aggressive and incurable B-cell lymphoma for which new treatment options are needed. Recent phase II clinical trials reported response to the proteasome inhibitor bortezomib (BZM) in up to 50% of pre-treated patients. Despite the successful use of BZM in the clinic, the precise molecular mechanisms underlying sensitivity or resistance to BZM in MCL remain largely unknown. To address this issue, we used U133A 2.0 microarrays to analyze gene expression in MCL cells from peripheral blood of 5 patients with previously untreated leukemic MCL. Samples were collected immediately before (0h) and at 3, 6, 24, and 72 hours after administration of BZM (1.5 mg/m2). After the blood collection at 72 hours, a second dose of BZM was given, and cells were collected 24 hours later. Two patients had major reductions in peripheral ALC already at 24h from dose 2 and normalized their blood counts by day 21 (sensitive), 1 patient had no change over a full course of 4 injections (resistant), and 2 patients had some decrease in ALC (intermediate). Genes differentially expressed with treatment were ranked according to the degree of correlation with time (Pearson). We used gene set enrichment analysis (GSEA) to detect distinct functional gene expression signatures; the most consistently up-regulated of which was a signature composed by proteasome and chaperone genes. To confirm and expand these findings, we exposed 10 MCL cell lines (7 sensitive, IC5010nM) to 10nM of BZM and analyzed gene expression at 1, 3, 6 and 24 hours. The proteasome signature was again dominant, and the majority of the up-regulated genes in both clinical and cell line samples shared binding motifs for the NRF, MAF, ATF and HSF families of transcription factors (TF). Thus genes up-regulated by BZM in vivo and in cell lines predominantly belonged to a functional response to oxidative and/or endoplasmic reticulum (ER) stress. Under physiologic conditions, this is thought to help restore homeostasis and protect from apoptosis. This response could therefore contribute to drug resistance or be a marker of an overwhelming insult before the cells undergo apoptosis. To address this issue, we investigated differences in response to BZM between sensitive and resistant cell lines. The proteasome signature was more strongly up-regulated in sensitive cells than in resistant cells, and the ER-stress response as measured by genes controlled by the NRF and MAF family of TFs was also more highly expressed in the sensitive group. Consistently, expression of HMOX1, which encodes a key enzyme in the antioxidant response, was increased by 32× at 24h in the sensitive group, but only by 4× in the resistant group; the expression of DDIT3, a transcription factor implicated in a pro-apoptotic response to ER-stress was 5.5-fold up-regulated in the sensitive cells but only 1.4-fold in the resistant cells. We conclude that in sensitive cells BZM induces an overwhelming ER-stress response with high expression of proteasome components and chaperone proteins that could serve as a predictor of response to BZM.

Description: Bortezomib, a potent inhibitor of the 26S proteasome, has remarkable activity against some lymphoid malignancies, particularly multiple myeloma and mantle cell lymphoma (MCL). The observed 30–50% clinical response rate to bortezomib as a single-agent in relapsed MCL is believed to be mediated mainly by inhibition of the NFkB signaling pathway. Recently, Hsp27 has been shown to confer resistance to bortezomib in a lymphoid cell line. However, much less is known about the mechanisms underlying anti-tumor activity of or resistance to bortezomib in MCL. To address these questions we studied 10 MCL cell lines with t(11;14)(q13;q32) as in vitro models. IC50 values were measured by the MTT cytotoxicity assay for both bortezomib and the BAY 11-7082 compound, a specific inhibitor of the NFkB pathway. The cell lines showed different profiles of response to bortezomib and were grouped according to their IC50 values as sensitive (S) (Granta 519, Jeko-1, SP-49, UPN-1), intermediate (In) (HBL-2, JVM-2, Z-138) and resistant (R) (Mino, NCEB-1, SP-53). The mean IC50 for bortezomib in the R group was 3 times higher than the mean IC50 of the S group. The cell lines also showed very different profiles of response to the NFkB-specific inhibitor BAY 11-7082, but grouped differently, suggesting that other mechanisms, in addition to the NFkB pathway, participate in the anti-neoplastic effect of proteasome inhibition in MCL. Resistance to both drugs had no correlation with P-glycoprotein activity as measured by the rodhamine efflux assay. We performed gene expression profiling on Affymetrix U133A 2.0 arrays of all 10 cell lines. Using GeneSpring software (Agilent), we normalized expression to the mean of the S group and identified 79 transcripts that showed a dose response behavior; that is, genes whose mean expression was 1.5x higher in the In versus the S group and 1.5x higher in the R versus the In group. Conversely, 55 transcripts followed the opposite trend, being down-regulated at least 1.5x in both comparisons. This list contained genes important in apoptosis control and in transmembrane transport, but was most notable for an overexpression of stress response genes in the R group. Notably, expression of Hsp70 was almost 10-fold higher in the R versus S group. We therefore analyzed expression of the HSP70 as well as HSP27 and HSP90 proteins by Western blotting. Hsp70 protein was highly expressed in all MCL cell lines and there was no discernable correlation with bortezomib resistance for any of the HSP-proteins analyzed. Other resistance mechanisms appear more important in MCL and we are currently testing further candidate genes for their ability to confer resistance to bortezomib in MCL.