Search

Microbes are highly diverse and widely distributed organisms. They account for ~60% of Earth’s biomass and new predictions point for the existence of 1011 to 1012 species, which are constantly sharing genes through several different mechanisms. Genomic Islands (GI) are critical in this context, as they are large regions acquired through horizontal gene transfer. Also, they present common features like genomic signature deviation, transposase genes, flanking tRNAs and insertion sequences. GIs carry large numbers of genes related to specific lifestyle and are commonly classified in Pathogenicity, Resistance, Metabolic or Symbiotic Islands. With the advent of the next-generation sequencing technologies and the deluge of genomic data, many software tools have been developed that aim to tackle the problem of GI prediction and they are all based on the prediction of GI common features. However, there is still room for the development of new software tools that implements new approaches, such as, machine learning and pangenomics based analyses. Finally, GIs will always hold a potential application in every newly invented genomic approach as they are directly responsible for much of the genomic plasticity of bacteria.

Abstract Background Cellular senescence is a state in which mammalian cells enter into an irreversible growth arrest and altered biological functions. The senescence response in mammalian cells can be elicited by DNA-damaging agents. In the present study we report that the DNA-damaging agent N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) is able to induce senescence in the HCT-116 colon cancer cell line. Results Cells treated with lower concentrations of MNNG (0–25 microM) for 50 h showed a dose-dependent increase in G2/M phase arrest and apoptosis; however, cells treated with higher concentrations of MNNG (50–100 microM) showed a senescence-like G0/G1 phase arrest which was confirmed by increased expression of β-galactosidase, a senescence induced marker. The G2/M phase arrest and apoptosis were found to be associated with increased levels of p53 protein, but the senescence-like G0/G1 phase arrest was dissociated with p53 protein levels, since the p53 protein levels decreased in senescence-like arrested cells. We further, determined whether the decreased level of p53 was a transcriptional or a translational phenomenon. The results revealed that the decreased level of p53 protein in senescence-like arrested cells was a transcriptional phenomenon since p53 mRNA levels simultaneously decreased after treatment with higher concentrations of MNNG. We also examined the effect of MNNG treatment on other cell cycle-related proteins such as p21, p27, cyclin B1, Cdc2, c-Myc and max. The expression levels of these proteins were increased in cells treated with lower concentrations of MNNG, which supported the G2/M phase arrest. However, cells treated with higher concentrations of MNNG showed decreased levels of these proteins, and hence, may not play a role in cell cycle arrest. We then examined a possible association of the expression of APC protein and telomeric DNA signals with cellular senescence in MNNG-treated cells. We found that protein and mRNA levels of APC were drastically reduced in cells treated with higher concentrations of MNNG. The loss of APC expression might lead to chromosomal instability as well as microtubular disorganization through its dissociation with tubulin. In fact, the protein level of α-tubulin was also drastically decreased in senescence-like arrested cells treated with higher concentrations of MNNG. The levels of telomeric DNA also decreased in cells treated with higher concentrations of MNNG. Conclusions These results suggest that in response to DNA alkylation damage the senescence-like arrest of HCT-116 cells was associated with decreased levels of APC protein, microtubular organization, and telomeric DNA.

Edge devices equipped with computer vision must deal with vast amounts of sensory data with limited computing resources. Hence, researchers have been exploring different energy-efficient solutions such as near-sensor, in-sensor, and in-pixel processing, bringing the computation closer to the sensor. In particular, in-pixel processing embeds the computation capabilities inside the pixel array and achieves high energy efficiency by generating low-level features instead of the raw data stream from CMOS image sensors. Many different in-pixel processing techniques and approaches have been demonstrated on conventional frame-based CMOS imagers; however, the processing-in-pixel approach for neuromorphic vision sensors has not been explored so far. In this work, for the first time, we propose an asynchronous non-von-Neumann analog processing-in-pixel paradigm to perform convolution operations by integrating in-situ multi-bit multi-channel convolution inside the pixel array performing analog multiply and accumulate (MAC) operations that consume significantly less energy than their digital MAC alternative. To make this approach viable, we incorporate the circuit's non-ideality, leakage, and process variations into a novel hardware-algorithm co-design framework that leverages extensive HSpice simulations of our proposed circuit using the GF22nm FD-SOI technology node. We verified our framework on state-of-the-art neuromorphic vision sensor datasets and show that our solution consumes ~2× lower backend-processor energy while maintaining almost similar front-end (sensor) energy on the IBM DVS128-Gesture dataset than the state-of-the-art while maintaining a high test accuracy of 88.36%., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Kaiser, Datta, Wang, Jacob, Beerel and Jaiswal.)

The demand to process vast amounts of data generated from state-of-the-art high resolution cameras has motivated novel energy-efficient on-device AI solutions. Visual data in such cameras are usually captured in analog voltages by a sensor pixel array, and then converted to the digital domain for subsequent AI processing using analog-to-digital converters (ADC). Recent research has tried to take advantage of massively parallel low-power analog/digital computing in the form of near- and in-sensor processing, in which the AI computation is performed partly in the periphery of the pixel array and partly in a separate on-board CPU/accelerator. Unfortunately, high-resolution input images still need to be streamed between the camera and the AI processing unit, frame by frame, causing energy, bandwidth, and security bottlenecks. To mitigate this problem, we propose a novel Processing-in-Pixel-in-memory (P 2 M) paradigm, that customizes the pixel array by adding support for analog multi-channel, multi-bit convolution, batch normalization, and Rectified Linear Units (ReLU). Our solution includes a holistic algorithm-circuit co-design approach and the resulting P 2 M paradigm can be used as a drop-in replacement for embedding memory-intensive first few layers of convolutional neural network (CNN) models within foundry-manufacturable CMOS image sensor platforms. Our experimental results indicate that P 2 M reduces data transfer bandwidth from sensors and analog to digital conversions by [Formula: see text], and the energy-delay product (EDP) incurred in processing a MobileNetV2 model on a TinyML use case for visual wake words dataset (VWW) by up to [Formula: see text] compared to standard near-processing or in-sensor implementations, without any significant drop in test accuracy., (© 2022. The Author(s).)

High-quality 3D image recognition is an important component of many vision and robotics systems. However, the accurate processing of these images requires the use of compute-expensive 3D Convolutional Neural Networks (CNNs). To address this challenge, we propose the use of Spiking Neural Networks (SNNs) that are generated from iso-architecture CNNs and trained with quantization-aware gradient descent to optimize their weights, membrane leak, and firing thresholds. During both training and inference, the analog pixel values of a 3D image are directly applied to the input layer of the SNN without the need to convert to a spike-train. This significantly reduces the training and inference latency and results in high degree of activation sparsity, which yields significant improvements in computational efficiency. However, this introduces energy-hungry digital multiplications in the first layer of our models, which we propose to mitigate using a processing-in-memory (PIM) architecture. To evaluate our proposal, we propose a 3D and a 3D/2D hybrid SNN-compatible convolutional architecture and choose hyperspectral imaging (HSI) as an application for 3D image recognition. We achieve overall test accuracy of 98.68, 99.50, and 97.95% with 5 time steps (inference latency) and 6-bit weight quantization on the Indian Pines, Pavia University, and Salinas Scene datasets, respectively. In particular, our models implemented using standard digital hardware achieved accuracies similar to state-of-the-art (SOTA) with ~560.6× and ~44.8× less average energy than an iso-architecture full-precision and 6-bit quantized CNN, respectively. Adopting the PIM architecture in the first layer, further improves the average energy, delay, and energy-delay-product (EDP) by 30, 7, and 38%, respectively., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Datta, Kundu, Jaiswal and Beerel.)

Immune checkpoint blockade (ICB) has demonstrated an impressive outcome in patients with metastatic melanoma, yet, durable complete response; even with Ipilimumab/Nivolumab combo are under 30%. Primary and acquired resistance in response to ICB is commonly due to a tumor immune escape mechanism dictated by the tumor microenvironment (TME). Macrophage Migratory Inhibition Factor (MIF) has emerged as an immunosuppressive factor secreted in the TME. We have previously demonstrated that blockade of the MIF-CD74 signaling on macrophages and dendritic cells restored the anti-tumor immune response against melanoma. Here, we report that inhibition of the MIF-CD74 axis combined with ipilimumab could render resistant melanoma to better respond to anti-CTLA-4 treatment. We provide evidence that blocking the MIF-CD74 signaling potentiates CD8+ T-cells infiltration and drives pro-inflammatory M1 conversion of macrophages in the TME. Furthermore, MIF inhibition resulted in reprogramming the metabolic pathway by reducing lactate production, HIF-1α and PD-L1 expression in the resistant melanoma cells. Melanoma patient data extracted from the TCGA database supports the hypothesis that high MIF expression strongly correlates with poor response to ICB therapy. Our findings provide a rationale for combining anti-CTLA-4 with MIF inhibitors as a potential strategy to overcome resistance to ICB therapy in melanoma, turning a "cold" tumor into a "hot" one mediated by the activation of innate immunity and reprogramming of tumor metabolism and reduced PD-L1 expression in melanoma cells., (© 2020 The Author(s). Published with license by Taylor & Francis Group, LLC.)

Despite the clinical success of T-cell checkpoint blockade, most patients with cancer still fail to have durable responses to immunotherapy. The molecular mechanisms driving checkpoint blockade resistance, whether preexisting or evolved, remain unclear. To address this critical knowledge gap, we treated B16 melanoma with the combination of CTLA-4, PD-1, and PD-L1 blockade and a Flt3 ligand vaccine (≥75% curative), isolated tumors resistant to therapy, and serially passaged them in vivo with the same treatment regimen until they developed complete resistance. Using gene expression analysis and immunogenomics, we determined the adaptations associated with this resistance phenotype. Checkpoint resistance coincided with acquisition of a "hypermetabolic" phenotype characterized by coordinated upregulation of the glycolytic, oxidoreductase, and mitochondrial oxidative phosphorylation pathways. These resistant tumors flourished under hypoxic conditions, whereas metabolically starved T cells lost glycolytic potential, effector function, and the ability to expand in response to immunotherapy. Furthermore, we found that checkpoint-resistant versus -sensitive tumors could be separated by noninvasive MRI imaging based solely on their metabolic state. In a cohort of patients with melanoma resistant to both CTLA-4 and PD-1 blockade, we observed upregulation of pathways indicative of a similar hypermetabolic state. Together, these data indicated that melanoma can evade T-cell checkpoint blockade immunotherapy by adapting a hypermetabolic phenotype., (©2020 American Association for Cancer Research.)

Despite the success of immune checkpoint blockade against melanoma, many "cold" tumors like prostate cancer remain unresponsive. We found that hypoxic zones were prevalent across preclinical prostate cancer and resisted T cell infiltration even in the context of CTLA-4 and PD-1 blockade. We demonstrated that the hypoxia-activated prodrug TH-302 reduces or eliminates hypoxia in these tumors. Combination therapy with this hypoxia-prodrug and checkpoint blockade cooperated to cure more than 80% of tumors in the transgenic adenocarcinoma of the mouse prostate-derived (TRAMP-derived) TRAMP-C2 model. Immunofluorescence imaging showed that TH-302 drives an influx of T cells into hypoxic zones, which were expanded by checkpoint blockade. Further, combination therapy reduced myeloid-derived suppressor cell density by more than 50%, and durably reduced the capacity of the tumor to replenish the granulocytic subset. Spontaneous prostate tumors in TRAMP transgenic mice, which completely resist checkpoint blockade, showed minimal adenocarcinoma tumor burden at 36 weeks of age and no evidence of neuroendocrine tumors with combination therapy. Survival of Pb-Cre4, Ptenpc-/-Smad4pc-/- mice with aggressive prostate adenocarcinoma was also significantly extended by this combination of hypoxia-prodrug and checkpoint blockade. Hypoxia disruption and T cell checkpoint blockade may sensitize some of the most therapeutically resistant cancers to immunotherapy.

Purpose: Agonist antibodies targeting the T-cell costimulatory receptor 4-1BB (CD137) are among the most effective immunotherapeutic agents across preclinical cancer models. In the clinic, however, development of these agents has been hampered by dose-limiting liver toxicity. Lack of knowledge of the mechanisms underlying this toxicity has limited the potential to separate 4-1BB agonist-driven tumor immunity from hepatotoxicity. Experimental Design: The capacity of 4-1BB agonist antibodies to induce liver toxicity was investigated in immunocompetent mice, with or without coadministration of checkpoint blockade, via (i) measurement of serum transaminase levels, (ii) imaging of liver immune infiltrates, and (iii) qualitative and quantitative assessment of liver myeloid and T cells via flow cytometry. Knockout mice were used to clarify the contribution of specific cell subsets, cytokines, and chemokines. Results: We find that activation of 4-1BB on liver myeloid cells is essential to initiate hepatitis. Once activated, these cells produce interleukin-27 that is required for liver toxicity. CD8 T cells infiltrate the liver in response to this myeloid activation and mediate tissue damage, triggering transaminase elevation. FoxP3 + regulatory T cells limit liver damage, and their removal dramatically exacerbates 4-1BB agonist-induced hepatitis. Coadministration of CTLA-4 blockade ameliorates transaminase elevation, whereas PD-1 blockade exacerbates it. Loss of the chemokine receptor CCR2 blocks 4-1BB agonist hepatitis without diminishing tumor-specific immunity against B16 melanoma. Conclusions: 4-1BB agonist antibodies trigger hepatitis via activation and expansion of interleukin-27-producing liver Kupffer cells and monocytes. Coadministration of CTLA-4 and/or CCR2 blockade may minimize hepatitis, but yield equal or greater antitumor immunity. Clin Cancer Res; 24(5); 1138-51. ©2018 AACR ., (©2018 American Association for Cancer Research.)

CD40 agonists bind the CD40 molecule on antigen-presenting cells and activate them to prime tumor-specific CD8 + T cell responses. Here, we study the antitumor activity and mechanism of action of a nonreplicating adenovirus encoding a chimeric, membrane-bound CD40 ligand (ISF35). Intratumoral administration of ISF35 in subcutaneous B16 melanomas generates tumor-specific, CD8 + T cells that express PD-1 and suppress tumor growth. Combination therapy of ISF35 with systemic anti-PD-1 generates greater antitumor activity than each respective monotherapy. Triple combination of ISF35, anti-PD-1, and anti-CTLA-4 results in complete eradication of injected and noninjected subcutaneous tumors, as well as melanoma tumors in the brain. Therapeutic efficacy is associated with increases in the systemic level of tumor-specific CD8 + T cells, and an increased ratio of intratumoral CD8 + T cells to CD4 + Tregs. These results provide a proof of concept of systemic antitumor activity after intratumoral CD40 triggering with ISF35 in combination with checkpoint blockade for multifocal cancer, including the brain.

Coordinated manipulation of independent immune regulatory pathways in the tumor microenvironment-including blockade of T-cell checkpoint receptors and reversal of suppressive myeloid programs-can render aggressive cancers susceptible to immune rejection. Elevated toxicity associated with combination immunotherapy, however, prevents translation of the most efficacious regimens. We evaluated T-cell checkpoint-modulating antibodies targeting CTLA-4, PD-1, and 4-1BB together with myeloid agonists targeting either STING or Flt3 in the TRAMP-C2 model of prostate cancer to determine whether low-dose intratumoral delivery of these agents could elicit systemic control of multifocal disease. Intratumoral administration of the STING agonist cyclic di-GMP (CDG) or Flt3 Ligand (Flt3L) augmented the therapeutic effect of systemic triple checkpoint modulation and promoted the cure of 75% of mice with bilateral TRAMP-C2; however, when all agents were administered locally, only CDG mobilized abscopal immunity. Combination efficacy correlated with globally enhanced ratios of CD8 + T cells to regulatory T cells (Treg), macrophages, and myeloid-derived suppressor cells, and downregulation of the M2 marker CD206 on tumor-associated macrophages. Flt3L improved CD8 + T-cell and dendritic cell infiltration of tumors, but was diminished in efficacy by concomitant Treg expansion. Although intratumoral CDG/checkpoint therapy invokes substantial ulceration at the injection site, reduced CDG dosing can preserve tissue integrity without sacrificing therapeutic benefit. For high-order combinations of T-cell checkpoint antibodies and local myeloid agonists, systemic antibody administration provides the greatest efficacy; however, local administration of CDG and antibody provides substantial systemic benefit while minimizing the potential for immune-related adverse events. Cancer Immunol Res; 5(8); 676-84. ©2017 AACR ., (©2017 American Association for Cancer Research.)