Uncovering low-dimensional, miR-based signatures of acute myeloid and lymphoblastic leukemias with a machine-learning-driven network approach

Complex phenotypic differences among different acute leukemias cannot be fully captured by analyzing the expression levels of one single molecule, such as a miR, at a time, but requires systematic analysis of large sets of miRs. While a popular approach for analysis of such datasets is principal component analysis (PCA), this method is not designed to optimally discriminate different phenotypes. Moreover, PCA and other low-dimensional representation methods yield linear or non-linear combinations of all measured miRs. Global human miR expression was measured in AML, B-ALL, and TALL cell lines and patient RNA samples. By systematically applying support vector machines to all measured miRs taken in dyad and triad groups, we built miR networks using cell line data and validated our findings with primary patient samples. All the coordinately transcribed members of the miR-23a cluster (which includes also miR-24 and miR-27a), known to function as tumor suppressors of acute leukemias, appeared in the AML, B-ALL and T-ALL centric networks. Subsequent qRT-PCR analysis showed that the most connected miR in the B-ALL-centric network, miR-708, is highly and specifically expressed in B-ALLs, suggesting that miR-708 might serve as a biomarker for B-ALL. This approach is systematic, quantitative, scalable, and unbiased. Rather than a single signature, our approach yields a network of signatures reflecting the redundant nature of biological signaling pathways. The network representation allows for visual analysis of all signatures by an expert and for future integration of additional information. Furthermore, each signature involves only small sets of miRs, such as dyads and triads, which are well suited for in depth validation through laboratory experiments. In particular, loss-and gain-of-function assays designed to drive changes in leukemia cell survival, proliferation and differentiation will benefit from the identification of multi-miR signatures that characterize leukemia subtypes and their normal counterpart cells of origin.

MiR-27a functions as a tumor suppressor in acute leukemia by regulating 14-3-3θ

MicroRNAs (miRs) play major roles in normal hematopoietic differentiation and hematopoietic malignancies. In this work, we report that miR-27a, and its coordinately expressed cluster (miR-23a∼miR-27a∼miR-24-2), was down-regulated in acute leukemia cell lines and primary samples compared to hematopoietic stem-progenitor cells (HSPCs). Decreased miR-23a cluster expression in some acute leukemia cell lines was mediated by c-MYC. Replacement of miR-27a in acute leukemia cell lines inhibited cell growth due, at least in part, to increased cellular apoptosis. We identified a member of the anti-apoptotic 14-3-3 family of proteins, which support cell survival by interacting with and negatively regulating pro-apoptotic proteins such as Bax and Bad, as a target of miR-27a. Specifically, miR-27a regulated 14-3-3θ at both the mRNA and protein levels. These data indicate that miR-27a contributes a tumor suppressor-like activity in acute leukemia cells via regulation of apoptosis, and that miR-27a and 14-3-3θ may be potential therapeutic targets.

Abstract 3385: Expression of CD22, a late B lymphoid antigen, does not distinguish leukemia stem cells from the bulk population in acute lymphoblastic leukemia cases

Introduction: Pediatric ALL is still fatal in ∼20% of cases, motivating development and clinical trials of novel antineoplastic agents in recurrent ALL such as CD22 monoclonal antibody and immunotoxins. CD22 is expressed on most ALL cases, but not every cell in every ALL case is detectably CD22+. In normal B lymphocyte development, CD22 is first expressed on the cell membrane at the early B to proB cell stage, at the time of D-J gene rearrangement. Taken together, these 2 facts beg the question of whether most or all of the presumed earliest cells in ALL cases, the leukemia stem cells, might be CD22-; if so, CD22 monoclonal antibody treatment might miss ALL stem cells. We therefore investigated CD22 expression on pediatric ALL stem cells. Methods: Studies were performed using secondary xenotransplants of five cases of pediatric precursor-B ALL. Cells (>90% human leukemia cells from spleens of immunodeficient mice) were stained with monoclonal antibodies and fluorescence-activated cell sorted (FACS). Titered doses of human ALL cells (100 -100000 cells/mouse; n=5 mice/group) from (1) unsorted, (2) mock sorted, (3) CD22high, and (4) CD22low populations, excluding non-viable and mouse cells, were separately transplanted by intravenous injection into NSG (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ) mice. NSG mice lack functional T, B cells, and NK cells, and are also deficient in innate immunity. Time to leukemia was determined from injection to onset of clinical signs of leukemia. Leukemia was confirmed by autopsy (splenomegaly) and FACS-analysis for human ALL. Pre- and post-sorted and pre- and post-transplanted cells were evaluated via multicolor FACS-analysis, using CD22, a viability dye, lymphoid (CD10, CD19), myeloid (CD13, CD33), stem-progenitor (CD34), and mouse cell (mouse CD45) markers. Results: Time to leukemia decreased in an inverse transplanted cell dose dependent manner in all 4 experimental groups (CD22high, CD22low, unsorted and mock sorted populations of each case), but was not statistically different among the four groups in any of the tested ALL cases. LSC quantitation is ongoing, based on frequency of mice with leukemias at the lowest transplanted cell doses. The heterogeneity of antigen expression in the leukemia cell populations was not different in cells analyzed before versus after NSG transplantation. Discussion: In these five cases of pediatric precursor-B ALL, the level of expression of CD22 did not, in evidence obtained to date, distinguish leukemia stem cells (as defined by NSG mouse-engrafting capacity) from the bulk population of ALL cells. Specifically, there is no suggestion from our data that leukemia stem cells were enriched in the CD22low populations, which provides further evidence against a hierarchical model of ALL stemness. Furthermore, targeting CD22 would not be predicted to differentially spare leukemia stem cells.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3385. doi:1538-7445.AM2012-3385

Mice deficient in MIM expression are predisposed to lymphomagenesis

Missing in metastasis (MIM) is a member of newly emerged inverse Bin-Amphiphysin-Rvs (BAR) domain protein family and a putative metastasis suppressor. Although reduced MIM expression has been associated with bladder, breast and gastric cancers, evidence for the role of MIM in tumor progression remains scarce and controversial. Herein we characterized a MIM knockout mouse strain and observed that MIM-deficient mice often developed enlarged spleens. Autopsy and histological analysis revealed that nearly 78% of MIM(-/-) mice developed tumors with features similar to diffuse large B lymphoma during a period from 1 to 2 years. MIM(-/-) mice also exhibited abnormal distribution of B cells in lymphoid organs with decrease in the spleen but increase in the bone marrow and the peripheral blood. Furthermore, the bone marrow of MIM(-/-) mice contained a higher percentage of pre-B2 cells but fewer immature B-cells than wild-type mice. In response to CXCL13, a B-cell chemokine released from splenic stromal cells, MIM-deficient B-cells did not undergo chemotaxis or morphological changes in response to the chemokine and also did not internalize CXCR5, the receptor of CXCL13. Microarray analyses demonstrated that MIM is the only member of the I-BAR domain family that was highly expressed in human B cells. However, low or absent MIM expression was common in either primary B-cell malignancies or established B-cell acute lymphocytic leukemia or lymphomas. Thus, our data demonstrate for the first time an important role for MIM in B-cell development and suggest that predisposition of MIM-null mice to lymphomagenesis may involve aberrant interactions between B lineage cells and the lymphoid microenvironment.

Effects of HMGN variants on the cellular transcription profile

High mobility group N (HMGN) is a family of intrinsically disordered nuclear proteins that bind to nucleosomes, alters the structure of chromatin and affects transcription. A major unresolved question is the extent of functional specificity, or redundancy, between the various members of the HMGN protein family. Here, we analyze the transcriptional profile of cells in which the expression of various HMGN proteins has been either deleted or doubled. We find that both up- and downregulation of HMGN expression altered the cellular transcription profile. Most, but not all of the changes were variant specific, suggesting limited redundancy in transcriptional regulation. Analysis of point and swap HMGN mutants revealed that the transcriptional specificity is determined by a unique combination of a functional nucleosome-binding domain and C-terminal domain. Doubling the amount of HMGN had a significantly larger effect on the transcription profile than total deletion, suggesting that the intrinsically disordered structure of HMGN proteins plays an important role in their function. The results reveal an HMGN-variant-specific effect on the fidelity of the cellular transcription profile, indicating that functionally the various HMGN subtypes are not fully redundant.

Developmental function of HMGN proteins

High mobility group N (HMGN) proteins are the only nuclear proteins known to specifically recognize the generic structure of the 147-bp nucleosome core particle. Both in vitro and in vivo experiments demonstrate that HMGN proteins are involved in epigenetic regulation by modulating chromatin structure and levels of posttranslational modifications of nucleosomal histones. Expression of HMGN proteins is developmentally regulated, and the loss or overexpression of these proteins can lead to developmental abnormalities. This review will focus on the role and on the possible molecular mechanism whereby HMGN proteins affect cellular differentiation and development.

Cell cycle-dependent binding of HMGN proteins to chromatin

Throughout the cell cycle, the histones remain associated with DNA, but the repertoire of proteins associated with the chromatin fiber continuously changes. The chromatin interaction of HMGNs, a family of nucleosome binding proteins that modulates the structure and activity of chromatin, during the cell cycle is controversial. Immunofluorescence studies demonstrated that HMGNs are not associated with chromatin, whereas live cell imaging indicated that they are present in mitotic chromosomes. To resolve this controversy, we examined the organization of wild-type and mutated HMGN1 and HMGN2 proteins in the cell nucleus by using immunofluorescence studies, live cell imaging, gel mobility shift assays, and bimolecular fluorescence complementation (BiFC). We find that during interphase, HMGNs bind specifically to nucleosomes and form homodimeric complexes that yield distinct BiFC signals. In metaphase, the nucleosomal binding domain of the protein is inactivated, and the proteins associate with chromatin with low affinity as monomers, and they do not form specific complexes. Our studies demonstrate that the mode of binding of HMGNs to chromatin is cell cycle dependent.

Tyk2 tyrosine kinase expression is required for the maintenance of mitochondrial respiration in primary pro-B lymphocytes

Tyk2, a member of the Jak family of protein tyrosine kinases, is critical for the biological actions of alpha/beta interferon (IFN-alpha/beta). Although Tyk2(-/-) mice are phenotypically normal, they exhibit abnormal responses to inflammatory challenges in a variety of cells isolated from Tyk2(-/-) mice. The reported phenotypic alterations in both Tyk2-null cells and mice are consistent with the possibility that the expression of this tyrosine kinase may regulate mitochondrial function. We report here that Tyk2-null pro-B cells are markedly deficient in basal oxygen consumption and exhibit a significant decrease in steady-state cellular ATP levels compared to wild-type cells. Tyk2-null cells also exhibit impaired complex I, III, and IV function of the mitochondrial electron transport chain. Reconstitution of Tyk2-null pro-B cells with either the wild type or a kinase-inactive mutant of Tyk2 restores basal mitochondrial respiration. By contrast, the kinase activity of Tyk2 is required for maintenance of both complex I-dependent mitochondrial respiration as well as induction of apoptosis in cells incubated with IFN-beta. Consistent with the role of Tyk2 in the regulation of tyrosine phosphorylation of Stat3, expression of a constitutively active Stat3 can restore the mitochondrial respiration in Tyk2-null cells treated with IFN-beta. Finally, Tyk2(-/-) mice show decreased exercise tolerance compared to wild-type littermates. Our results implicate a novel role for Tyk2 kinase and Stat3 phosphorylation in mitochondrial respiration.

Sox6 cell-autonomously stimulates erythroid cell survival, proliferation, and terminal maturation and is thereby an important enhancer of definitive erythropoiesis during mouse development

Erythropoiesis, the essential process of hematopoietic stem cell development into erythrocytes, is controlled by lineage-specific transcription factors that determine cell fate and differentiation and by the hormone erythropoietin that stimulates cell survival and proliferation. Here we identify the Sry-related high-mobility-group (HMG) box transcription factor Sox6 as an important enhancer of definitive erythropoiesis. Sox6 is highly expressed in proerythroblasts and erythroblasts in the fetal liver, neonatal spleen, and bone marrow. Mouse fetuses and pups lacking Sox6 develop erythroid cells slowly and feature misshapen, short-lived erythrocytes. They compensate for anemia by elevating the serum level of erythropoietin and progressively enlarging their erythropoietic tissues. Erythroid-specific inactivation of Sox6 causes the same phenotype, demonstrating cell-autonomous roles for Sox6 in erythroid cells. Sox6 potentiates the ability of erythropoietin signaling to promote proerythroblast survival and has an effect additive to that of erythropoietin in stimulating proerythroblast and erythroblast proliferation. Sox6 also critically facilitates erythroblast and reticulocyte maturation, including hemoglobinization, cell condensation, and enucleation, and ensures erythrocyte cytoskeleton long-term stability. It does not control adult globin and erythrocyte cytoskeleton genes but acts by stabilizing filamentous actin (F-actin) levels. Sox6 thus enhances erythroid cell development at multiple levels and thereby ensures adequate production and quality of red blood cells.

Unexpected role of ceruloplasmin in intestinal iron absorption

Ferroxidases are essential for normal iron homeostasis in most organisms. The paralogous vertebrate ferroxidases ceruloplasmin (Cp) and hephaestin (Heph) are considered to have nonidentical functions in iron transport: plasma Cp drives iron transport from tissue stores while intestinal Heph facilitates iron absorption from the intestinal lumen. To clarify the function of Cp, we acutely bled Cp-/- mice to stress iron homeostasis pathways. Red cell hemoglobin recovery was defective in stressed Cp-/- mice, consistent with low iron availability. Contrary to expectations, iron was freely released from spleen and liver stores in Cp-/- mice, but intestinal iron absorption was markedly impaired. Phlebotomy of wild-type mice caused a striking shift of Cp from the duodenal epithelium to the underlying lamina propria, suggesting a critical function of Cp in basolateral iron transport. Regulated relocalization of intestinal Cp may represent a fail-safe mechanism in which Cp shares with Heph responsibility for iron absorption under stress.

Circadian sensitivity to the chemotherapeutic agent cyclophosphamide depends on the functional status of the CLOCK/BMAL1 transactivation complex

The circadian clock controls many aspects of mammalian physiology, including responses to cancer therapy. We find that wild-type and circadian mutant mice demonstrate striking differences in their response to the anticancer drug cyclophosphamide (CY). While the sensitivity of wild-type mice varies greatly, depending on the time of drug administration, Clock mutant and Bmal1 knockout mice are highly sensitive to treatment at all times tested. On the contrary, mice with loss-of-function mutations in Cryptochrome (Cry1-/-Cry2-/- double knockouts) were more resistant to CY compared with their wild-type littermates. Thus, both time-of-day and allelic-dependent variations in response to chemotherapy correlate with the functional status of the circadian CLOCK/BMAL1 transactivation complex. Pharmacokinetic analysis of plasma concentration of different CY metabolites shows that, in contrast to the traditional view, circadian variations in drug sensitivity cannot be attributed to the changes in the rates of CY metabolic activation and/or detoxification. At the same time, mice of different circadian genotypes demonstrate significant differences in B cell responses to toxic CY metabolites: B cell survival/recovery rate was directly correlated with the in vivo drug sensitivity. Based on these results, we propose that the CLOCK/BMAL1 transcriptional complex affects the lethality of chemotherapeutic agents by modulating the survival of the target cells necessary for the viability of the organism.

Anemia and impaired stress-induced erythropoiesis in aceruloplasminemic mice

Ceruloplasmin (Cp) is an abundant, copper-containing plasma protein with an important role in iron homeostasis. Patients with hereditary Cp deficiency have iron deposits in liver and other organs, consistent with impaired iron flux. The mild anemia reported in some patients suggests a possible role for Cp in iron delivery to red cell precursors during erythropoiesis. To investigate this function of Cp, we determined the hematologic parameters in Cp-deficient mice under normal conditions and after erythropoiesis-inducing stress. Cp(-/-) mice have below normal hematocrit, red cell hemoglobin and volume, and serum iron. Red cell number and turnover and reticulocyte counts were identical in Cp(-/-) and Cp(+/+) mice. Thus, Cp(-/-) have mild microcytic, hypochromic anemia consistent with normal red cell formation but defective iron availability. Cp(-/-) and Cp(+/+) mice subjected to phenylhydrazine-induced hemolytic anemia exhibited identical decreases in hematologic parameters, but Cp(-/-) mice showed diminished recovery after removal of the stress. Administration of purified human Cp or iron-saturated transferrin to Cp(-/-) mice partially restored hemoglobin formation in reticulocytes. The mild anemia in Cp(-/-) mice and the diminished response to stress may reflect inefficient recycling of iron between the reticuloendothelial and erythropoietic systems. Our findings suggest a role for Cp in erythropoiesis by providing sufficient iron to the erythroid tissue and that the requirement for Cp is raised after erythropoietic stress.

Sphenopalatine ganglion block for the treatment of myofascial pain of the head, neck, and shoulders

BACKGROUND AND OBJECTIVES

This study examined the effectiveness of sphenopalatine ganglion block (SPGB) for myofascial pain syndrome of the head, neck, and shoulders using a double-blind, placebo-controlled, crossover study design with comparison to an internal standard consisting of trigger point injections (TPI).

METHODS

Patients (n = 23) were randomly assigned to receive either: (1) SPGB with 4% lidocaine, then TPI with 1% lidocaine, and finally SPGB with saline placebo or (2) SPGB with saline placebo, then TPI with 1% lidocaine, and finally SPGB with 4% lidocaine. Each respective treatment within each protocol was given sequentially at 1-week intervals for both groups. Prior to the first treatment, all patients assessed their average intensity of pain and pain at that particular moment using a visual analog pain scale. Pain intensity and pain relief were reassessed 30 minutes after each treatment and at 6 hours, 24 hours and 1 week using visual analog pain and pain relief scales. Pain intensity and pain relief data were transformed into natural logarithm units, and the statistical significance of SPGB with 4% lidocaine versus SPGB with placebo, SPGB with 4% lidocaine versus TPI, and TPI versus SPGB with placebo were tested by mixed-model analysis of variance. The magnitude of the differences in pain intensity and pain relief ratings were also compared via computation of 95% confidence intervals.

RESULTS

The analgesic effect of SPGB with 4% lidocaine was no better than placebo. Mixed-model analysis of variance revealed improved analgesia with administration of TPIs as compared to SPGB with 4% lidocaine and placebo over the entire week of observations (pain relief scores).

CONCLUSIONS

This study suggests that SPGB with 4% lidocaine is no more efficacious than placebo and less efficacious than administration of standard trigger point injections in the treatment of myofascial pain of the head, neck, and shoulders.

The analgesic response to intravenous lidocaine in the treatment of neuropathic pain

This study was performed in order to determine concentration-effect, and graded and quantal dose-response relationships for the clinical administration of intravenous (IV) lidocaine to patients with neuropathic pain. Thirteen patients were administered 500 mg of IV lidocaine at a rate of 8.35 mg/min over 60 min. Visual analog pain scores and venous blood samples were obtained concomitantly at 10 min intervals for 60 min. Blood samples were also obtained for determination of serum and serum water lidocaine concentrations at the onset of analgesia and at the time complete pain relief was attained. Lidocaine concentrations were determined by gas chromatography. Graded dose-response curves were prepared individually and for the group as a whole, and a quantal dose-response curve was prepared for the entire group. The dose-response relationship for IV lidocaine was characterized by large increases in pain relief for concomitant minimal increases in dosage. The difference between the ED50 (372.0 mg) and the ED90 (416.5 mg) was 44.5 mg of lidocaine (5.3 min of infusion). The concentration-effect relationship was also steep with pain scores abruptly decreasing over a range of 0.62 microgram/mL of lidocaine. Interestingly, the free concentration of lidocaine had no better correlation with the onset of analgesia or the attainment of complete analgesia than the serum concentration of lidocaine. This suggests that the mechanism of analgesia to IV lidocaine may not be based upon a conventional concentration-effect relationship. In conclusion, the results of this study suggest that the analgesic response to IV lidocaine is best characterized by a precipitous "break in pain" over a narrow dosage and concentration range.