Efficient discovery of anti-inflammatory small-molecule combinations using evolutionary computing

The control of biochemical fluxes is distributed, and to perturb complex intracellular networks effectively it is often necessary to modulate several steps simultaneously. However, the number of possible permutations leads to a combinatorial explosion in the number of experiments that would have to be performed in a complete analysis. We used a multiobjective evolutionary algorithm to optimize reagent combinations from a dynamic chemical library of 33 compounds with established or predicted targets in the regulatory network controlling IL-1β expression. The evolutionary algorithm converged on excellent solutions within 11 generations, during which we studied just 550 combinations out of the potential search space of ~9 billion. The top five reagents with the greatest contribution to combinatorial effects throughout the evolutionary algorithm were then optimized pairwise. A p38 MAPK inhibitor together with either an inhibitor of IκB kinase or a chelator of poorly liganded iron yielded synergistic inhibition of macrophage IL-1β expression. Evolutionary searches provide a powerful and general approach to the discovery of new combinations of pharmacological agents with therapeutic indices potentially greater than those of single drugs.     

Synergistic killing of human small cell lung cancer cells by the Bcl-2-inositol 1,4,5-trisphosphate receptor disruptor BIRD-2 and the BH3-mimetic ABT-263

Small cell lung cancer (SCLC) has an annual mortality approaching that of breast and prostate cancer. Although sensitive to initial chemotherapy, SCLC rapidly develops resistance, leading to less effective second-line therapies. SCLC cells often overexpress Bcl-2, which protects cells from apoptosis both by sequestering pro-apoptotic family members and by modulating inositol 1,4,5-trisphosphate receptor (IP₃R)-mediated calcium signaling. BH3-mimetic agents such as ABT-263 disrupt the former activity but have limited activity in SCLC patients. Here we report for the first time that Bcl-2-IP₃ receptor disruptor-2 (BIRD-2), a decoy peptide that binds to the BH4 domain of Bcl-2 and prevents Bcl-2 interaction with IP₃Rs, induces cell death in a wide range of SCLC lines, including ABT-263-resistant lines. BIRD-2-induced death of SCLC cells appears to be a form of caspase-independent apoptosis mediated by calpain activation. By targeting different regions of the Bcl-2 protein and different mechanisms of action, BIRD-2 and ABT-263 induce cell death synergistically. Based on these findings, we propose that targeting the Bcl-2–IP₃R interaction be pursued as a novel therapeutic strategy for SCLC, either by developing BIRD-2 itself as a therapeutic agent or by developing small-molecule inhibitors that mimic BIRD-2.Lung cancer accounts for 12% of all new cancers worldwide and is a leading cause of cancer-related mortality in the United States.^{1, 2, 3} Although small cell lung cancer (SCLC) comprises only 15% of lung cancer cases,^{2, 3} it has an annual mortality rate approaching that of breast and prostate cancer.⁴ Compared with the more common non-small cell lung cancer (NSCLC), SCLC is more aggressive and associated with rapid development of metastasis.² Moreover, although SCLC is more responsive to chemotherapy and radiation therapy initially, it typically relapses quickly with treatment-resistant disease.² In contrast to dramatic advances in chemotherapy and personalized medicine in other malignancies, the life expectancy of SCLC patients has remained <2 years for decades and is <1 year for patients with extensive disease.^{5, 6} The lethality of SCLC is attributed in part to the development of resistance to standard combination chemotherapies, underscoring the need to develop novel therapeutic approaches based on understanding the molecular and cellular biology of SCLC.^{5, 6}Evasion from apoptosis is a major hallmark of cancer and a prominent factor underlying drug resistance in SCLC.³ Multiple mechanisms contribute to apoptosis resistance in SCLC, including elevated expression of the antiapoptotic Bcl-2 protein³ (Supplementary Figure S1). Tsujimoto and colleagues discovered elevated levels of Bcl-2 mRNA and protein in SCLC cells not long after their identification of Bcl-2 as the protein product of the bcl-2 gene in follicular lymphoma.^{7, 8} Subsequently, immunohistochemistry of 164 primary SCLC samples revealed 76% were positive for Bcl-2, a finding substantiated by microarray detection of increased BCL-2 mRNA levels in 84% of SCLC samples^{9, 10} and by genomic sequencing of circulating SCLC tumor cells.¹¹ Moreover, proteomic profiling documented that Bcl-2 is more highly expressed in SCLC than in NSCLC, reflecting the vastly different biology of these lung cancer subtypes.¹²The major known function of Bcl-2 is to bind and sequester BH3-only proteins such as Bim, preventing these proteins from inducing apoptosis.^{13, 14, 15} Therefore, a major investment has been made in targeting this interaction for cancer treatment. The interaction takes place in a hydrophobic groove on Bcl-2 and the therapeutic strategy for targeting this interaction has been to develop small molecules, BH3-mimetic agents, which bind in the hydrophobic groove and induce apoptosis by displacing the BH3-only proteins. This approach has been reviewed in detail.^{14, 15, 16}Among BH3-mimetic agents advancing through clinical trials for both hematological malignancies^{15, 17} and solid tumors¹⁸ are ABT-737 and its orally bioavailable derivative ABT-263 (Navitoclax). Reported studies of ABT-199, a selective inhibitor of Bcl-2, are at present limited to hematological malignancies.¹⁸ In screening a large number of cancer cell lines, the pioneering work of Oltersdorf et al.¹⁹ demonstrated potent single-agent activity of ABT-737 against cell lines representative of lymphoid malignancies and SCLC. Clinical trials of ABT-263, an orally bioavailable version of ABT-737, achieved overall response rates ranging from as high as 35% in relapsed/refractory chronic lymphocytic leukemia (CLL) and 22% in follicular lymphoma.¹⁷ Reported responses are generally less in solid tumors with the notable exception of SCLC.¹⁸ But even in SCLC, activity of ABT-263 is limited in comparison to hematological malignancies, with 1 of the 39 (3%) of patients achieving a partial response to ABT-263 and 9 of the 37 (23%) achieving stable disease in a phase I clinical trial.²⁰ This experience suggests a need to develop additional ways of targeting Bcl-2 for cancer treatment.A potential alternative therapeutic target for Bcl-2-positive malignancies involves interaction of Bcl-2 with the inositol 1,4,5-trisphosphate receptor (IP₃R), an IP₃-gated Ca²⁺ channel located on the endoplasmic reticulum (ER). Bcl-2 is located not only on the outer mitochondrial membrane but also on the ER, and at both of these locations, it functions as a potent inhibitor of apoptosis.^{21, 22, 23} ER-localized Bcl-2 interacts with IP₃Rs and inhibits apoptosis by preventing excessive IP₃R-mediated Ca²⁺ transfer from the ER lumen into the cytoplasm and nearby mitochondria.^{24, 25, 26} Notably, regions of Bcl-2 involved in binding BH3-only proteins and IP₃Rs are entirely different. Whereas BH3-only proteins and their BH3-mimetic counterparts bind in a hydrophobic groove composed of BH3 domains 1–3 of Bcl-2,^{13, 14} the BH4 domain of Bcl-2 is necessary for interaction with IP₃Rs.²⁷ To develop a peptide inhibitor of Bcl-2–IP₃R interaction, we identified the Bcl-2-binding region on the IP₃R and developed a small synthetic 20 amino-acid peptide corresponding to this region.²⁸ This peptide, when fused to the cell-penetrating peptide of HIV TAT, binds to the BH4 domain of Bcl-2 and functions as a decoy peptide, inhibiting Bcl-2–IP₃R interaction.^{29, 30} We currently refer to this peptide as BIRD-2 (Bcl-2-IP₃ Receptor Disruptor-2), having formerly named it TAT-IDP_DD/AA.³¹ By disrupting the Bcl-2–IP₃R interaction, BIRD-2 abrogates Bcl-2 control over IP₃R-mediated Ca²⁺ elevation and induces Ca²⁺-mediated apoptosis in primary human CLL cells²⁹ and diffuse large B-cell lymphoma cells.³² Notably, BIRD-2 does not kill normal cells, including human lymphocytes isolated from peripheral blood²⁹ and normal murine embryonic fibroblasts (F Zhong and C Distelhorst, unpublished data).The present investigation was undertaken to determine whether Bcl-2–IP₃R interaction is a potentially useful therapeutic target in SCLC. In support of this concept, we find the majority of SCLC lines tested are sensitive to BIRD-2-induced apoptosis and that BIRD-2 induces apoptosis in several ABT-263-resistant SCLC lines. BIRD-2, we find, lacks generalized cytotoxicity as it does not induce cell death in NSCLC lines or a normal lung epithelial line. On the other hand, we find that BIRD-2 and ABT-263 synergize in killing SCLC cells. These findings for the first time provide preclinical evidence of the potential value of targeting both antiapoptotic mechanisms of Bcl-2 for the treatment of SCLC.     

Regulation of haem biosynthesis in normoblastic erythropoiesis: role of 5-aminolaevulinic acid synthase and ferrochelatase

The development of haem biosynthetic enzyme activity during normoblastic human erythropoiesis was examined in seven patients. The first and last enzymes of the haem biosynthetic pathway, ALA synthase and ferrochelatase, were assayed by radiochemical/high performance liquid chromatographic (HPLC) methods. An assay for ferrochelatase activity in human bone marrow was developed. Enzyme substrates were protoporphyrin IX and ⁵⁹Fe²⁺ ions. ⁵⁹Fe-labelled haem was isolated by organic solvent extraction/sorbent extraction followed by reversed-phase HPLC. Optimal activity occurred at pH 7.3 in the presence of ascorbic acid, in darkness and under anaerobic conditions. Haem production was proportional to cell number and was linear with time to 30 min. The assay was sensitive to the picomolar range of haem production. ALA synthase and ferrochelatase activity was assayed in four highly purified age-matched erythroid cell populations. ALA synthase activity was maximal in the most immature erythoid cells and diminished as the cells matured with an overall five fold loss of activity from proerythroblast to late erythroblast development. Ferrochelatase activity was, however, more stable with less than a two fold change in activity observed during the same period of erythroid differentiation. Maximal activity occurred in erythroid fractions enriched with intermediate erythroblasts. These results support sequential rather than simultaneous appearance of these enzymes during normoblastic erythropoiesis. Quantitative analysis of relative enzyme activity however indicates that at all times during erythroid differentiation ferrochelatase activity is present in excess to that theoretically required relative to ALA synthase activity since ALA and haem are not produced in stoichiometric amounts. The lability of ALA synthase versus the stability and gross relative excess of ferrochelatase activity indicates a far greater role for ALA synthase in the regulation of erythroid haem biosynthesis than for ferrochelatase.     

Pharmacogenetic analysis of lithium-induced delayed aging in Caenorhabditis elegans

Lithium (Li(+)) has been used to treat mood affect disorders, including bipolar, for decades. This drug is neuroprotective and has several identified molecular targets. However, it has a narrow therapeutic range and the one or more underlying mechanisms of its therapeutic action are not understood. Here we describe a pharmacogenetic study of Li(+) in the nematode Caenorhabditis elegans. Exposure to Li(+) at clinically relevant concentrations throughout adulthood increases survival during normal aging (up to 46% median increase). Longevity is extended via a novel mechanism with altered expression of genes encoding nucleosome-associated functions. Li(+) treatment results in reduced expression of the worm ortholog of LSD-1 (T08D10.2), a histone demethylase; knockdown by RNA interference of T08D10.2 is sufficient to extend longevity ( approximately 25% median increase), suggesting Li(+) regulates survival by modulating histone methylation and chromatin structure.     

Response of Two Heat Shock Genes to Selection for Knockdown Heat Resistance in Drosophila Melanogaster

To identify genes involved in stress resistance and heat hardening, replicate lines of Drosophila melanogaster were selected for increased resistance to knockdown by a 39° heat stress. Two selective regimes were used, one with and one without prior hardening. Mean knockdown times were increased from ~5 min to >20 min after 18 generations. Initial realized heritabilities were as high as 10% for lines selected without hardening, and crosses between lines indicated simple additive gene effects for the selected phenotypes. To survey allelic variation and correlated selection responses in two candidate stress genes, hsr-omega and hsp68, we applied denaturing gradient gel electrophoresis to amplified DNA sequences from small regions of these genes. After eight generations of selection, allele frequencies at both loci showed correlated responses for selection following hardening, but not without hardening. The hardening process itself was associated with a hsp68 frequency change in the opposite direction to that associated with selection that followed hardening. These stress loci are closely linked on chromosome III, and the hardening selection established a disequilibrium, suggesting an epistatic effect on resistance. The data indicate that molecular variation in both hsr-omega and hsp68 contribute to natural heritable variation for hardened heat resistance.     

Chemokines and dendritic cells: a crucial alliance

Dendritic cells (DC) are bone marrow-derived professional antigen-presenting cells that function as sentinels of the immune system. Their importance in immunity resides in their unique ability to prime or tolerize T lymphocytes, thereby initiating or inhibiting immune responses. They reside in all tissues and organs and upon appropriate activation, migrate to secondary lymphoid organs to present antigen to T lymphocytes in the T cell zones. Because of this central role in T cell activation, there is a great deal of interest in using DC therapeutically to deliver positive or negative signals to the immune system. The DC system is critically dependent on the ability of DC at different stages of maturation to respond to a range of soluble and cell-bound signals, including members of the chemokine gene superfamily. This review will describe the interactions between DC and the chemokine system.