Nonselective inhibition of the epigenetic transcriptional regulator BET induces marked lymphoid and hematopoietic toxicity in mice.

Bromo and extra terminal (BET) proteins (BRD2, BRD3, BRD4 and BRDT) are epigenetic transcriptional regulators required for efficient expression of growth promoting, cell cycle progression and antiapoptotic genes. Through their bromodomain, these proteins bind to acetylated lysine residues of histones and are recruited to transcriptionally active chromatin. Inhibition of the BET-histone interaction provides a tractable therapeutic strategy to treat diseases that may have epigenetic dysregulation. JQ1 is a small molecule that blocks BET interaction with histones. It has been shown to decrease proliferation of patient-derived multiple myeloma in vitro and to decrease tumor burden in vivo in xenograft mouse models. While targeting BET appears to be a viable and efficacious approach, the nonclinical safety profile of BET inhibition remains to be well-defined. We report that mice dosed with JQ1 at efficacious exposures demonstrate dose-dependent decreases in their lymphoid and immune cell compartments. At higher doses, JQ1 was not tolerated and due to induction of significant body weight loss led to early euthanasia. Flow cytometry analysis of lymphoid tissues showed a decrease in both B- and T-lymphocytes with a concomitant decrease in peripheral white blood cells that was confirmed by hematology. Further investigation with the inactive enantiomer of JQ1 showed that these in vivo effects were on-target mediated and not elicited through secondary pharmacology due to chemical structure.

Off-target immune cell toxicity caused by AG-012986, a pan-CDK inhibitor, is associated with inhibition of p38 MAPK phosphorylation.

AG-012986 is a pan-CDK (cyclin-dependent kinase) inhibitor that has in vitro and in vivo antitumor properties but was stopped in development due in part to rapid bone-marrow-independent white blood cell toxicity in preclinical studies and the potential for acute and delayed immunosuppression in humans. Because peripheral lymphocytes are largely nonproliferating, it was hypothesized the toxicity of AG-012986 was due to an off-target mechanism and not driven by the intended pharmacology. We show the toxicity mechanism in primary human immune cells is caspase driven. T-cells treated with AG-012986 and acutely stimulated through the T-cell receptor exhibited decreased toxicity while still maintaining cell division inhibition. This indicated that the pharmacology of AG-012986 functioned as expected but the toxicity had now been decoupled through activation. Induced phosphorylation of p38 and IL-2 production was impaired with AG-012986. Thus, AG-012986 could cause apoptosis of T-cells by targeting upstream kinases in the p38 Mitogen-activated protein kinase (MAPK) pathway and impairing cellular survival.

Energy-efficient image compression for resource-constrained platforms.

One of the most important goals of current and future sensor networks is energy-efficient communication of images. This paper presents a quantitative comparison between the energy costs associated with 1) direct transmission of uncompressed images and 2) sensor platform-based JPEG compression followed by transmission of the compressed image data. JPEG compression computations are mapped onto various resource-constrained platforms using a design environment that allows computation using the minimum integer and fractional bit-widths needed in view of other approximations inherent in the compression process and choice of image quality parameters. Advanced applications of JPEG, such as region of interest coding and successive/progressive transmission, are also examined. Detailed experimental results examining the tradeoffs in processor resources, processing/transmission time, bandwidth utilization, image quality, and overall energy consumption are presented.

Regulation of gene expression in mast cells: micro-rNA expression and chromatin structural analysis of cytokine genes.

Despite deriving from two different compartments of the immune system (myeloid and lymphoid respectively), Th2 cells and mast cells produce the same panel of cytokines, interleukin (IL)4, IL5 and IL13. We have compared the chromatin structure of the RAD50/IL13/IL4 locus in Th2 cells and mast cells. Th2 and mast cells display strong overlap in their patterns of DNase I hypersensitivity throughout this locus, except that the first intron of the IL13 gene (MCHS) is DNase I hypersensitive only in mast cells and the conserved non-coding sequence (CNS)-1 in the IL4/IL13 intergenic region is DNase I hypersensitive only in Th2 cells (explaining why cytokine expression is impaired in Th2 cells but not in mast cells of CNS-1-deleted mice). We have also examined the role of micro-RNAs (miRNAs) in the development and activation of mast cells and T cells. miRNAs are 21- to 25-nucleotide small RNAs that regulate gene expression posttranscriptionally by targeting protein-coding mRNAs. Using oligonucleotide arrays to analyse miRNA expression in murine T cells and mast cells, we have identified distinctive cell type-specific patterns of miRNA expression as well as changes related to differentiation and cell activation. We are studying the biological functions of selected miRNAs.

Precision-aware self-quantizing hardware architectures for the discrete wavelet transform.

This paper presents designs for both bit-parallel (BP) and digit-serial (DS) precision-optimized implementations of the discrete wavelet transform (DWT), with specific consideration given to the impact of depth (the number of levels of DWT) on the overall computational accuracy. These methods thus allow customizing the precision of a multilevel DWT to a given error tolerance requirement and ensuring an energy-minimal implementation, which increases the applicability of DWT-based algorithms such as JPEG 2000 to energy-constrained platforms and environments. Additionally, quantization of DWT coefficients to a specific target step size is performed as an inherent part of the DWT computation, thereby eliminating the need to have a separate downstream quantization step in applications such as JPEG 2000. Experimental measurements of design performance in terms of area, speed, and power for 90-nm complementary metal-oxide-semiconductor implementation are presented. Results indicate that while BP designs exhibit inherent speed advantages, DS designs require significantly fewer hardware resources with increasing precision and DWT level. A four-level DWT with medium precision, for example, while the BP design is four times faster than the digital-serial design, occupies twice the area. In addition to the BP and DS designs, a novel flexible DWT processor is presented, which supports run-time configurable DWT parameters.

Retina expression and cross-species validation of gene silencing by PF-655, a small interfering RNA against RTP801 for the treatment of ocular disease.

PURPOSE: PF-655, a synthetic 19-mer siRNA, targeting the RTP801 gene is currently in clinical trials for the treatment of wet age-related macular degeneration and diabetic macular edema. Preclinical studies have shown a dose-related suppression of RTP801 expression in rat disease models. Investigative studies were conducted with PF-655 to validate the Dutch-Belted rabbit as a biologically relevant species for gene silencing to support nonclinical ocular toxicity and continual dosing studies. METHODS: Cross-species comparison and DNA sequencing was done to determine the level of homology between PF-655 and rabbit RTP801. Human (HEK 293) and rabbit (SIRC cornea) cell lines were stimulated with CoCl(2) to mimic hypoxic stress (an inducer of RTP801 expression) and treated with PF-655. Taqman-polymerase chain reaction and immunoblot analysis were performed to gauge RTP801 expression in cell culture and rabbit retinas. RESULTS: Sequence analysis showed a 1-base mismatch in the PF-655 targeting site from genomic DNA of Dutch-Belted rabbit and the SIRC cell line, a cornea cell derived from the New Zealand White rabbit. HEK and SIRC CoCl(2)-stressed cells induced RTP801 expression 10-20-fold above control conditions. Treatment with 20 or 100 nM PF-655 showed a decrease in gene expression, 40%-50% relative to appropriate controls. RTP801 mRNA was detectable in primary rabbit retina tissues, with cycle threshold values showing a large linear range for the assay. CONCLUSION: These results support our investigation into cross-species validation of gene suppression by a therapeutic siRNA designed to a human gene. The SIRC cell line was utilized as a surrogate to test the degree of RTP801 gene silencing induced by PF-655 in vitro. With a 1-base mismatch, the level of silencing in a rabbit ocular cell line was comparable to that of a human cell line. Sequence analysis and expression data confirmed the relevance of the RTP801 target gene in rabbits and the utility of this species as a relevant animal model. Additionally, our work outlines a tractable method that validates relevant larger non-rodent species for ophthalmic drug testing.

Mechanistic investigation of imatinib-induced cardiac toxicity and the involvement of c-Abl kinase.

The Bcr-abl tyrosine kinase inhibitor imatinib mesylate is the frontline therapy for chronic myeloid leukemia. Imatinib has been reported to cause congestive heart failure and left ventricular contractile dysfunction in patients and cardiomyopathy in rodents, findings proposed to be associated with its pharmacological activity. To investigate the specific role of Abelson oncogene 1 (c-Abl) in imatinib-induced cardiac toxicity, we performed targeted gene inhibition of c-Abl by RNA interference in neonatal cardiomyocytes (NCMs). Suppression of c-Abl did not lead to cytotoxicity or induction of endoplasmic reticulum (ER) stress. To further dis associate c-Abl from imatinib-induced cardiac toxicity, we designed imatinib structural analogs that do not have appreciable c-Abl inhibition in NCMs. The c-Abl inactive analogs induced cytotoxicity and ER stress, at similar or greater potencies and magnitudes as imatinib. Furthermore, combining c-Abl gene silencing with imatinib and analogs treatment did not significantly shift the cytotoxicity dose response curves. Imatinib and analogs were shown to accumulate in lysosomes, likely due to their physicochemical properties, and disrupt autophagy. The toxicity induced by imatinib and analogs can be rescued by bafilomycin A pretreatment, demonstrating the involvement of lysosomal accumulation in cardiac toxicity. The results from our studies strongly suggest that imatinib induces cardiomyocyte dysfunction through disruption of autophagy and induction of ER stress, independent of c-Abl inhibition.

Chromatin-based regulation of cytokine transcription in Th2 cells and mast cells.

Th2 cells and mast cells are major sources of IL4, IL5 and IL13, cytokines that mediate immunity against parasites and are also central players in the pathophysiology of asthma, allergy and atopic disease. We asked whether Th2 cells and mast cells, which belong to the lymphoid and myeloid lineages, respectively, use different cis-acting regulatory regions to transcribe the cytokine genes. Comparison of DNase I hypersensitivity patterns at the RAD50/IL4/IL13 locus revealed that most hypersensitive sites (HSs) are common to Th2 and mast cells, but two regions [conserved non-coding sequence (CNS) 1 and mast cell HSs] show cell type-specific differences. CNS-1, one of the most highly conserved CNS regions in the RAD50/IL13/IL4 locus, displays two strong DNase I HSs in Th2 cells but is not DNase I hypersensitive in mast cells, explaining a previous finding that deletion of CNS-1 impairs cytokine expression in Th2 cells but not in mast cells. Conversely, two constitutive HSs (mast cell HSs) in the first intron of the IL13 gene are present in mast cells but not in Th2 cells; these sites develop early during mast cell differentiation and may have a role in maintaining accessibility of the IL13 locus to high-level transcription in stimulated cells.

Molecular analysis of a locus control region in the T helper 2 cytokine gene cluster: a target for STAT6 but not GATA3.

The linked IL-4 and IL-13 cytokine genes, which are activated and silenced in T helper (Th) 2 and Th1 cells, respectively, are flanked by the equivalently expressed RAD50 and KIF3A genes. A scan of DNase I hypersensitivity and DNA methylation across approximately 100 kb of the KIF3A/IL-4/IL-13/RAD50 cluster revealed differences in chromatin structure between Th1 and Th2 cells at the 3' end of the RAD50 gene, a region previously shown to contain a locus control region (LCR) regulating Th2-specific expression of IL-4 and IL-13. Naive CD4 T cells did not exhibit any DNase I hypersensitivity in this region, but stimulation under either Th1 or Th2 conditions caused rapid development of three hypersensitive sites. An additional hypersensitive site developed rapidly only under Th2 conditions, through a mechanism dependent on signal transducers and activators of transcription 6 (STAT6) but not GATA3. Our data point to a physical separation in the actions of STAT6 and its downstream effector GATA3 during Th2 differentiation: STAT6 directly remodels the RAD50 LCR, whereas GATA3 acts only in the vicinity of the IL-4 gene. We suggest that the RAD50 LCR has a complex and dual role in Th1 and Th2 differentiation, communicating early T cell antigen receptor and cytokine signals to the IL-4/IL-13 locus in both differentiating cell types.

An epigenetic view of helper T cell differentiation.

Antigen and cytokine receptor signals act in synergy to direct the differentiation of CD4+ T cells. These signals initiate reciprocal activation and silencing of the interferon-gamma (IFN-gamma) and interleukin 4 (IL-4) cytokine gene loci, changes that are heritably maintained in the resulting T helper type 1 (T(H)1) or T(H)2 cells and their progeny. Early, unpolarized transcription and chromatin remodeling of the poised cytokine genes of naive T cells is followed by consolidation and spreading of epigenetic changes and the establishment of self-reinforcing transcription factor networks. Recent studies have begun to elucidate the molecular mechanisms that establish and maintain polarized cytokine gene expression, and thus the cellular identity of differentiated helper T cells.

A distal enhancer in the interferon-gamma (IFN-gamma) locus revealed by genome sequence comparison.

Large-scale cross-species DNA sequence comparison has become a powerful tool to identify conserved cis-regulatory modules of genes. However, bioinformatic analysis alone cannot reveal how an evolutionarily conserved region regulates gene expression: whether it functions as an enhancer, silencer, or insulator; whether its function is cell-type restricted; and whether biologically relevant transcription factors bind to the element. Here we combine bioinformatics with wet-lab techniques to illustrate a general and systematic method of identifying functional conserved regulatory regions of genes. We applied this approach to the interferon-gamma (IFN-gamma) gene. Comparison of human and mouse IFN-gamma reveals a highly conserved non-coding sequence located approximately 5 kb 5' of the transcription start site. This region coincides with constitutive and inducible DNase I hypersensitivity sites present in IFN-gamma-producing Th1 cells but not in Th2 cells that do not produce IFN-gamma. Histone methylation at the 5' conserved non-coding sequences indicates a more accessible chromatin structure in Th1 cells compared with Th2 cells. This element binds two transcription factors known to be essential for IFN-gamma expression: nuclear factor of activated T cells, an inducible transcription factor, and T-box protein expressed in T cells, a cell lineage-restricted transcription factor. Together, these findings identify a highly conserved distal enhancer in the IFN-gamma cytokine locus and validate our approach as a successful method to detect cis-regulatory elements.

Regulation of Ig class switch recombination by NF-kappaB: retroviral expression of RelB in activated B cells inhibits switching to IgG1, but not to IgE.

Mutant NF-kappaB-deficient B cells from knockout mice lacking RelA, p105/p50 or the transactivation domain of c-Rel exhibit distinct and selective cell-intrinsic defects in their ability to undergo class switch recombination (CSR) to specific Ig isotypes. This isotype-specific requirement for particular NF-kappaB transcription factors in B cells activated to undergo CSR is intriguing because the NF-kappaB composition in B cells is also highly regulated and can vary significantly depending upon how B cells are activated. These studies prompted us to test by retroviral transduction of normal B cells whether changes in the NF-kappaB composition in activated B cells could modulate cytokine-driven CSR. RelB, RelA, c-Rel, p50 and p52 were first expressed in lipopolysaccharide-activated primary B cells and then induced by cytokine addition to undergo CSR to IgG1, IgE, IgG2a, IgG2b or IgA. Surprisingly, only retroviral expression of RelB altered CSR, resulting in a 3-fold decrease in CSR to IgG1 induced by IL-4. This effect was isotype specific as RelB expression did not affect CSR to IgE within the same culture or to other isotypes tested. The transactivation domain of RelB was required for inhibition of CSR to IgG1. Expression of p50-RelB or p52-RelB dimers joined covalently by a flexible peptide linker also specifically inhibited IgG1 CSR. RelB-mediated inhibition of IgG1 CSR was associated with a decrease in germline gamma1 transcription, but not with changes in proliferation as assayed by CFSE labeling. Thus, RelB complexes can specifically inhibit CSR to IgG1, but not IgE, in activated, primary B cells.

Th2 lineage commitment and efficient IL-4 production involves extended demethylation of the IL-4 gene.

The relation of CpG methylation to gene silencing is well established, but the contribution of DNA demethylation to gene expression during cell differentiation remains unclear. We show that the IL-4 locus undergoes a complex series of methylation and demethylation steps during T helper cell differentiation. The 5' region of the IL-4 locus is hypermethylated in naive T cells and becomes specifically demethylated in Th2 cells, whereas a highly conserved DNase I-hypersensitive region at the 3' end shows the converse behavior, being hypomethylated in naive T cells and becoming methylated during Th1 differentiation. 5' demethylation is not required for chromatin remodeling or primary transcription of the IL-4 gene but is strongly associated with efficient, high-level induction of IL-4 transcripts by differentiated Th2 cells.

Bioinformatics for the 'bench biologist': how to find regulatory regions in genomic DNA.

The combination of bioinformatic and biological approaches constitutes a powerful method for identifying gene regulatory elements. High-quality genome sequences are available in public databases for several vertebrate species. Comparative cross-species sequence analysis of these genomes shows considerable conservation of noncoding sequences in DNA. Biological analyses show that an unexpectedly high number of the conserved sequences correspond to functional cis-regulatory regions that influence gene transcription. Because research biologists are often unfamiliar with the bioinformatic resources at their disposal, this commentary discusses how to integrate biological and bioinformatic methods in the discovery of gene regulatory regions and includes a tutorial on widely available comparative genomics programs.

Deletion of a conserved Il4 silencer impairs T helper type 1-mediated immunity.

Helper T cell differentiation involves silencing as well as activation of gene expression. We have identified a conserved silencer of the gene encoding interleukin 4 (Il4) marked by DNase I hypersensitivity (HS IV) and permissive chromatin structure in all helper T cells. Deletion of HS IV increased Il4 and Il13 transcription by naive T cells and led to T helper type 2 skewing in vitro. HS IV controlled Il4 silencing during T helper type 1 differentiation, as HS IV-deficient T helper type 1 cells that expressed interferon-gamma also produced abundant interleukin 4 in vitro and in vivo. Despite mounting a vigorous interferon-gamma response, HS IV-deficient mice were more susceptible to Leishmania major infection than were wild-type littermate control mice, showing a critical function for Il4 silencing in T helper type 1-mediated immunity.