Search

Despite the remarkable success of immunotherapy in the treatment of melanoma, resistance to these agents still affects patient prognosis and response to therapies. Beta-2-microglobulin (β2M), an important subunit of major histocompatibility complex (MHC) class I, has important biological functions and roles in tumor immunity. In recent years, increasing studies have shown that B2M gene deficiency can inhibit MHC class I antigen presentation and lead to cancer immune evasion by affecting β2M expression. Based on this, B2M gene defect and T cell-based immunotherapy can interact to affect the efficacy of melanoma treatment. Taking into account the many recent advances in B2M-related melanoma immunity, here we discuss the immune function of the B2M gene in tumors, its common genetic alteration in melanoma, and its impact on and related improvements in melanoma immunotherapy. Our comprehensive review of β2M biology and its role in tumor immunotherapy contributes to understanding the potential of B2M gene as a promising melanoma therapeutic target.

Uveal melanoma (UM) is the most frequent primary intraocular malignancy in adults, characterized by melanin depositions in melanocytes located in the uveal tract in the eyes. Differentiation of melanin species (eumelanin and pheomelanin) is crucial in the diagnosis and management of UM, yet it remains inaccessible for conventional histology. Here, we report that femtosecond time-resolved pump-probe microscopy could provide label-free and chemical-specific detection of melanin species in human UM based on their distinct transient relaxation dynamics at the subpicosecond timescale. The method is capable of delineating the interface between melanoma and paracancerous regions on various tissue conditions, including frozen sections, paraffin sections, and fresh tissues. Moreover, transcriptome sequencing was conducted to confirm the active eumelanin synthesis in UM. Our results may hold potential for sensitive detection of tumor boundaries and biomedical research on melanin metabolism in UM.

Abstract Background The MYCN amplification is a defining hallmark of high-risk neuroblastoma. Due to irregular oncogenes orchestration, tumor cells exhibit distinct fatty acid metabolic features from non-tumor cells. However, the function of MYCN in neuroblastoma fatty acid metabolism reprogramming remains unknown. Methods Gas Chromatography-Mass Spectrometer (GC-MS) was used to find the potential target fatty acid metabolites of MYCN. Real-time PCR (RT-PCR) and clinical bioinformatics analysis was used to find the related target genes. The function of the identified target gene ELOVL2 on cell growth was detected through CCK-8 assay, Soft agar colony formation assay, flow Cytometry assay and mouse xenograft. Chromatin immunoprecipitation (ChIP) and Immunoprecipitation-Mass Spectrometer (IP-MS) further identified the target gene and the co-repressor of MYCN. Results The fatty acid profile of MYCN-depleted neuroblastoma cells identified docosahexaenoic acid (DHA), an omega-3 polyunsaturated fatty acid with anti-tumor activity, significantly increased after MYCN depletion. Compared with MYCN single-copy neuroblastoma cells, DHA level was significantly lower in MYCN-amplified neuroblastoma cells. RT-PCR and clinical bioinformatics analysis discovered that MYCN interfered DHA accumulation via ELOVL fatty acid elongase 2 (ELOVL2) which is a rate-limiting enzyme of cellular DHA synthesis. Enforced ELOVL2 expression in MYCN-amplified neuroblastoma cells led to decreased cell growth and counteracted the growth-promoting effect of MYCN overexpression both in vitro and vivo. ELOVL2 Knockdown showed the opposite effect in MYCN single-copy neuroblastoma cells. In primary neuroblastoma, high ELOVL2 transcription correlated with favorable clinical tumor biology and patient survival. The mechanism of MYCN-mediated ELOVL2 inhibition contributed to epigenetic regulation. MYCN recruited PRC1 (Polycomb repressive complex 1), catalysed H2AK119ub (histone 2A lysine 119 monoubiquitination) and inhibited subsequent ELOVL2 transcription. Conclusions The tumor suppressive properties of DHA and ELOVL2 are repressed by the MYCN and PRC1 jointly, which suggests a new epigenetic mechanism of MYCN-mediated fatty acid regulation and indicates PRC1 inhibition as a potential novel strategy to activate ELOVL2 suppressive functions.

As a promising in vivo tool for cancer research, zebrafish have been widely applied in various tumor studies. The zebrafish xenograft model is a low-cost, high-throughput tool for cancer research that can be established quickly and requires only a small sample size, which makes it favorite among researchers. Zebrafish patient-derived xenograft (zPDX) models provide promising evidence for short-term clinical treatment. In this review, we discuss the characteristics and advantages of zebrafish, such as their transparent and translucent features, the use of vascular fluorescence imaging, the establishment of metastatic and intracranial orthotopic models, individual pharmacokinetics measurements, and tumor microenvironment. Furthermore, we introduce how these characteristics and advantages are applied other in tumor studies. Finally, we discuss the future direction of the use of zebrafish in tumor studies and provide new ideas for the application of it.

Recent studies show that deep neural networks (DNNs) are extremely vulnerable to elaborately designed adversarial examples. Adversarial training, which uses adversarial examples as training data, has been proven to be one of the most effective methods of defense against adversarial attacks. However, most existing adversarial training methods use adversarial examples relying on first-order gradients, which perform poorly against second-order adversarial attacks and make it difficult to further improve the robustness of the model. In contrast to first-order gradients, second-order gradients provide a more accurate approximation of the loss landscape relative to natural examples. Therefore, our work focuses on constructing second-order adversarial examples and utilizing them for training DNNs. However, second-order optimization involves computing the Hessian inverse, which typically consumes considerable time. To address this issue, we propose an approximation method that transforms the problem into optimization within the Krylov subspace. Compared with the Euclidean space, the Krylov subspace method typically does not require storing the entire matrix. It only needs to store vectors and intermediate results, avoiding explicitly calculating the complete Hessian matrix. We approximate the adversarial direction by a linear combination of Hessian-vector products in the Krylov subspace to reduce the computation cost. Because of the non-symmetrical Hessian matrix, we use the generalized minimum residual to search for an approximate polynomial solution of the matrix. Our method efficiently reduces computational complexity and accelerates the training process. Extensive experiments conducted on the MNIST, CIFAR-10, and ImageNet-100 datasets demonstrate that our adversarial learning using second-order adversarial samples outperforms other first-order methods, leading to improved model robustness against various attacks. [ABSTRACT FROM AUTHOR]