Knutson, Gaylord J.

Abstract Imatinib mesylate (imatinib, Gleevec®, Novartis, Basel, Switzerland) inhibits T cells in vitro and in vivo (Dietz et al., Blood104: 1094–1099, 2004; Cwynarski et al., Leukemia18: 1332–1339, 2004). The drug blocks T cell cycle progression rather uniquely as it neither inhibits expression of CD69, an early marker of T cell activation, nor induces apoptosis. To characterize the molecular effects of imatinib leading to this mode of T-cell inhibition, we measured the changes in transcriptome (by Affymetrix U133 chips), proteome and phosphoproteome (by Western blotting, differential phosphoprotein expression and mass spectrometry). We found that phytohemagglutinin activated T cells pre-treated with imatinib had reduced expression of 983 transcripts and increased expression of 271 transcripts when compared to untreated PHA activated T cells by the factor of 1.5 or more (p<0.05). Among the prominently down-regulated transcripts were granzyme B, CTLA-4 and IL-2-receptor α-chain (CD25), all characteristic of activated T cells, as well as cyclins D2 and D3 and cyclin-dependent kinases 3, 4 and 7, the molecules regulating cell cycle progression. Among the up-regulated transcripts were Kruppel-like transcription factors 2 and 7, and p27, a finding compatible with the observed cell cycle inhibition. Furthermore, we selected and identified 30 proteins from 2-D gels that were up-regulated and/or hyperphosphorylated in imatinib treated activated T cells. Among these were four heterogeneous ribonucleoproteins, three lamins and γ-actin, all components of the nucleoskeleton at the interface of chromatin and inner nuclear membrane and involved in replication and transcription (Herrmann and Foisner, Cell. Mol. Life Sci.60: 1607–1612, 2003; Shumaker et al. Curr. Opinion Cell Biol.15: 358–366, 2003). Thus, imatinib-borne interference with T cell signal transduction affects the nuclear structure indicating for the first time that nucleoskeleton structural changes are associated with T cell activation status.

Lineage-specific growth factors mobilize peripheral blood progenitor cells (PBPC) and accelerate hematopoietic recovery after high-dose chemotherapy. Recombinant human thrombopoietin (rhTPO) may further increase the progenitor-cell content and regenerating potential of PBPC products. We evaluated the safety and activity of rhTPO as a PBPC mobilizer in combination with granulocyte colony-stimulating factor (G-CSF) in 29 breast cancer patients treated with high-dose chemotherapy followed by PBPC reinfusion. Initially, patients received escalating single doses of rhTPO intravenously (IV) at 0.6, 1.2, or 2.4 μg/kg, on day 1. Subsequent patients received rhTPO 0.6 or 0.3 μg/kg on days −3, −1, and 1, or 0.6 μg/kg on days −1 and 1. G-CSF, 5 μg/kg IV or subcutaneously (SC) twice daily, was started on day 3 and continued through aphereses. Twenty comparable, concurrently and identically treated patients (who were eligible and would have been treated on protocol but for the lack of study opening) mobilized with G-CSF alone served as comparisons. CD34+ cell yields were substantially higher with the first apheresis following rhTPO and G-CSF versus G-CSF alone: 4.1 × 106/kg (range, 1.3 to 17.6) versus 0.8 × 106/ kg (range, 0.3 to 4.2), P = .0003. The targeted minimum yield of 3 × 106CD34+ cells/kg was procured following a single apheresis procedure in 61% of the rhTPO and G-CSF–mobilized group versus 10% of G-CSF–mobilized patients (P = .001). In rhTPO and G-CSF mobilized patients, granulocyte (day 8 v 9, P= .0001) and platelet recovery (day 9 v 10, P= .07) were accelerated, and fewer erythrocyte (3 v 4,P = .02) and platelet (4 v 5, P = .02) transfusions were needed compared with G-CSF–mobilized patients. Peripheral blood platelet counts, following rhTPO and G-CSF, were increased by greater than 100% and the platelet content of PBPC products by 60% to 110% on the first and second days of aphereses (P < .0001) with the greatest effect seen with repeated dosing of rhTPO at 0.6 μg/kg. rhTPO is safe and well tolerated as a mobilizing agent before PBPC collection. Mobilization with rhTPO and G-CSF, in comparison to a comparable, nonrandomized G-CSF–mobilized group of patients, decreases the number of apheresis procedures required, may accelerate hematopoietic recovery, and may reduce the number of transfusions required following high-dose chemotherapy for breast cancer.

Abstract Lineage-specific growth factors mobilize peripheral blood progenitor cells (PBPC) and accelerate hematopoietic recovery after high-dose chemotherapy. Recombinant human thrombopoietin (rhTPO) may further increase the progenitor-cell content and regenerating potential of PBPC products. We evaluated the safety and activity of rhTPO as a PBPC mobilizer in combination with granulocyte colony-stimulating factor (G-CSF) in 29 breast cancer patients treated with high-dose chemotherapy followed by PBPC reinfusion. Initially, patients received escalating single doses of rhTPO intravenously (IV) at 0.6, 1.2, or 2.4 μg/kg, on day 1. Subsequent patients received rhTPO 0.6 or 0.3 μg/kg on days −3, −1, and 1, or 0.6 μg/kg on days −1 and 1. G-CSF, 5 μg/kg IV or subcutaneously (SC) twice daily, was started on day 3 and continued through aphereses. Twenty comparable, concurrently and identically treated patients (who were eligible and would have been treated on protocol but for the lack of study opening) mobilized with G-CSF alone served as comparisons. CD34+ cell yields were substantially higher with the first apheresis following rhTPO and G-CSF versus G-CSF alone: 4.1 × 106/kg (range, 1.3 to 17.6) versus 0.8 × 106/ kg (range, 0.3 to 4.2), P = .0003. The targeted minimum yield of 3 × 106CD34+ cells/kg was procured following a single apheresis procedure in 61% of the rhTPO and G-CSF–mobilized group versus 10% of G-CSF–mobilized patients (P = .001). In rhTPO and G-CSF mobilized patients, granulocyte (day 8 v 9, P= .0001) and platelet recovery (day 9 v 10, P= .07) were accelerated, and fewer erythrocyte (3 v 4,P = .02) and platelet (4 v 5, P = .02) transfusions were needed compared with G-CSF–mobilized patients. Peripheral blood platelet counts, following rhTPO and G-CSF, were increased by greater than 100% and the platelet content of PBPC products by 60% to 110% on the first and second days of aphereses (P &lt; .0001) with the greatest effect seen with repeated dosing of rhTPO at 0.6 μg/kg. rhTPO is safe and well tolerated as a mobilizing agent before PBPC collection. Mobilization with rhTPO and G-CSF, in comparison to a comparable, nonrandomized G-CSF–mobilized group of patients, decreases the number of apheresis procedures required, may accelerate hematopoietic recovery, and may reduce the number of transfusions required following high-dose chemotherapy for breast cancer.

Several laboratories report factor VIII coagulant (VIII: C) from human, bovine and porcine sources interact with Willebrand factor (WF). Bovine coagulant may be S-S bonded to WF: the human and porcine VIII:C may reversibly bind to WF. This interaction has been used by various laboratories to help purify the VIII:C. The VIII:C from all 3 species are activated by Ila and inactivated in EDTA. Our work has involved the porcine proteins bound by WF and anti-VIII:C antibodies. Human antibodies to human VIII:C bind porcine plasma proteins which on SDS-PAGE band at apparent MW 160K, 130K and 80K. These data are consistent with the human VIII:C (Switzer and McKee (JCI 60:819, 1977)). Their human material appears to contain S-S linked chains, porcine chains separate in SDS alone. Thrombin treatment results in loss of high molecular weight proteins and an appearance of activation/deactivation products at MW = 80K and 73K and lower. Two murine monoclonal anti-porcine VIII:C antibodies partially inactivate VIII:C while a third doesn’t. This latter antibody, when immobilized, removes activity from solutions of VIII:C dissociated from WF but not from solutions of undissociated VIII:C. The VIII:C immobilized through this antibody is still active, can be removed by organic solvent in active form, and contains the same protein chains as those bound to the polyclonal human anti- VIII:C. The monoclonal antibody VIII:C complex is inactivated and releases the 130K + 160K chains in EDTA. The 80K MW chain can be removed separately. Based on relative staining intensity in SDS gels, the porcine VIII:C at this stage of purity may be heterogeneous, (130K + 80K and 160K + 80K), with the chains associated through Ca++. The 80K MW chain, may bind the VIII:C to porcine WF. The site of this interaction does not seem to be critical to the measurement of VIII:C activity in vitro.

Abstract Tumor-associated hypoxia has been associated with malignant progression, metastasis and resistance to therapy. We studied the role of oxygen tension (pO2) in modulating the properties of cultured human prostate cancer (CaP) cells. In air, where oxygen tension (pO2) is approximately 20 kPa, cultured CaP cells expressed lower levels of transcripts associated with cancer and urologic disease; reducing pO2 to 2 kPa made these levels comparable to those in surgical CaP specimens suggesting that hypoxically cultured cells, rather than normoxically cultured cells, express genes at levels akin to those in tumor cells in situ. In human primary CaP tissue, transcripts of hypoxia-associated genes DLG7, CCNB1 and HMMR were overexpressed and associated with Gleason score and disease prognosis; this implies the potential prognostic value of these transcripts. While hypoxia affected the CaP cell proteome mostly quantitatively, it did affect expression of tumor–associated antigens and suggested the potential therapeutic relevance of the hypoxic antigenic landscape. Overall, our results suggest that hypoxia modifies cellular properties of cultured CaP cells towards a phenotype more similar to tumor cells in situ. Introduction of pO2 as a variable can constitute a tool for the identification of more effective prognostic biomarkers and development of better therapy for CaP and, possibly, other tumors. Citation Format: Christian R. Gomez, Stanimir Vuk-Pavlovic, Farhad Kosari, Jan Marie Munz, Claire Schreiber, Gaylord Knutson, Cristine Charlesworth, R. Jeffrey Karnes, John Cheville, George Vasmatzis. Prognostic value of hypoxia-associated genes in prostate cancer [abstract]. In: Proceedings of the AACR Special Conference on Advances in Prostate Cancer Research; 2012 Feb 6-9; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2012;72(4 Suppl):Abstract nr C43.

Abstract Tumor-associated hypoxia facilitates malignant progression, metastasis and poor prognosis in prostate cancer (CaP); however, effects of oxygen tension (pO2) in CaP are just beginning to be investigated. To establish mechanisms whereby hypoxia enhances malignant properties and survival of CaP and to identify pO2-regulated tumor-associated macromolecules, we propagated CaP cell lines LnCaP, VCaP and DU145 in hypoxia (pO2=2 kPa) and compared them with normoxically grown cells (pO2=20 kPa); hypoxic cells proliferated faster and secreted more vascular endothelial growth factor (p&lt;0.05). Transcriptome studies revealed different gene expression in cells grown in hypoxia relative to those in normoxia. Interestingly, in hypoxic cells transcripts associated with cancer and urologic disease were overexpressed in comparison to normoxic cells (p&lt;0.001) suggesting an association of low pO2 and aggressive features of CaP. Further, we analyzed the transcriptome in primary CaP cells isolated by laser-capture microdissection (LCM) and whole CaP tissue. We compared the data to those from CaP cells cultured at pO2=20 kPa or pO2=2 kPa. Notably, hypoxia increased transcript levels for pyruvate dehydrogenase kinase isozyme 1, nuclear prelamin A recognition factor, glucose phosphate isomerase, and glyceraldehyde-3-phosphate dehydrogenase in all three cell lines (p&lt;0.05) to levels comparable to those found in primary bulk tissue and LCM isolated cells (p&lt;0.005). This finding suggests that gene expression in hypoxically cultured cells is more akin to that in tumor cells in situ than are cells grown normoxically. By 2D-gel electrophoresis, we found that change in pO2 affected the proteome mostly quantitatively (i.e., by change in spot intensity). Our results suggest that hypoxia affects transcriptome and proteome in CaP cells. In addition, we screened 20 patient sera and 20 healthy control sera for spontaneous antibodies cross-reactive with VCaP cells and found that the sera reacted with numerous proteins, some previously reported to elicit an immune response in CaP patients (e.g., nucleoporin 62 and transitional endoplasmic reticulum ATPase). By this method we detected numerous novel autoantigens (under validation). Currently we are clarifying the pO2 effects on the relationship of transcriptome, proteome and tumor-associated antigens in CaP cells. Support: DOD PC094680 (CRG), Minnesota Partnership for Biotechnology and Medical Genomics, Mayo Clinic Prostate SPORE 5P50CA091956 (FK); Mrs. Adelyn L. Luther, Singer Island, Florida and Mayo Clinic Cancer Center (SV-P). Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr LB-308. doi:10.1158/1538-7445.AM2011-LB-308

Abstract In early clinical trials, vaccination with allogeneic cancer cells enhanced the overall survival in some patients with malignancies. Because the administered vaccine cells were cultured at pO2=20 kPa, it is unclear whether they provided an adequate antigen match to tumor cells in situ where pO2 is generally much lower. Thus, we postulate that hypoxically grown vaccine cells will provide a better antigen match to tumors in situ. We are testing this hypothesis by studying the effects of hypoxia on prostate cancer (CaP) cells. LnCaP and VCaP cells were cultured at pO2=2 kPa or 20 kPa and analyzed by 2-D gel electrophoresis. The difference in pO2 affected the proteome mostly quantitatively (i.e., by change in spot intensity). Using a threshold of fivefold change, we found that hypoxia decreased the levels of thirteen proteins and increased of four in VCaP cells. These results are in line with the reports showing that hypoxia affects expression only of a small fraction of total cellular protein and that the content of total protein is not altered significantly. To determine if the hypoxia-dependent changes affected immune reactivity of the cells, we took advantage of the spontaneous autoantibodies against tumor-associated antigens (TAAs). Hence, we compared the binding to lysates of LnCaP and VCaP cells of antibodies pooled from CaP patients (n=25), healthy controls (n=25), rheumatoid arthritis (n=17), colorectal cancer (n=10) and lung cancer (n=10). CaP patient sera plasma reacted with numerous spots, some even in control groups and thus considered nonspecific. CaP sera plasma specifically bound to seven spots; four were hypoxia-specific. All selected spots were excised from the gel, trypsin-digested, and identified by MALDI-TOF mass spectrometry as heat shock 70 kDa protein 4; protein disulfide isomerase A3; heterogeneous nuclear ribonucleoprotein L and leucine-rich repeat-containing protein 47. With the exception of the latter, the proteins have been identified or validated as TAAs before, but apparently none in CaP. CaP patient antibodies bound to lysates of LnCaP cells grown at pO2=2 kPa, but less than to such VCaP cell lysates. Our results confirmed the findings that hypoxia affects the proteome mostly quantitatively. Lysates of hypoxic CaP cells revealed novel potential TAAs suggesting the potential immunologic relevance of antigens in the hypoxic proteome. Thus, pO2 control can provide a tool for the development of more effective immunotherapy for CaP and possibly other cancers. Support: DOD PC094680 (CRG), PCF Creativity Award (CRG), Mayo Clinic Prostate SPORE 5P50CA091956 (SV-P); Mrs. Adelyn L. Luther, Singer Island, Florida and Mayo Clinic Cancer Center (SV-P). Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr C223.

Abstract Imatinib mesylate (STI571, imatinib) inhibited DNA synthesis in primary human T cells stimulated with allogeneic mature dendritic cells or phytohemagglutinin (PHA) but did not induce apoptosis. The values for the concentration that inhibits 50% (IC50) of T-cell proliferation stimulated by dendritic cells and PHA were 3.9 μM and 2.9 μM, respectively, that is, within the concentration range found in patients treated with imatinib mesylate. Interestingly, imatinib mesylate did not inhibit expression of T-cell activation markers CD25 and CD69, although it reduced the levels of activated nuclear factor-κB (NF-κB) and changed phosphorylation or protein levels of Lck, ERK1/2, retinoblastoma protein, and cyclin D3. When T cells were washed free of imatinib mesylate, they proliferated in response to PHA, demonstrating that inhibition is reversible. Treatment with imatinib mesylate led to accumulation of the cells in G0/G1 phase of the cell cycle. The in vitro observations were confirmed in vivo in a murine model of delayed-type hypersensitivity (DTH). In mice treated with imatinib mesylate, DTH was reduced in comparison to sham-injected controls. However, the number of splenic T cells was not reduced showing that, similarly to in vitro observations, imatinib mesylate inhibited T-cell response, but did not cause apoptosis. These findings indicate that long-term administration of high-dose imatinib mesylate might affect immunity.