• IMMUNE NETWORK
• /
• v.24 no.1
• /
• pp.9.1-9.21
• /
• 2024

The cytokine IL-7 plays critical and nonredundant roles in T cell immunity so that the abundance and availability of IL-7 act as key regulatory mechanisms in T cell immunity. Importantly, IL-7 is not produced by T cells themselves but primarily by non-lymphoid lineage stromal cells and epithelial cells that are limited in their numbers. Thus, T cells depend on cell extrinsic IL-7, and the amount of in vivo IL-7 is considered a major factor in maximizing and maintaining the number of T cells in peripheral tissues. Moreover, IL-7 provides metabolic cues and promotes the survival of both naïve and memory T cells. Thus, IL-7 is also essential for the functional fitness of T cells. In this regard, there has been an extensive effort trying to increase the protein abundance of IL-7 in vivo, with the aim to augment T cell immunity and harness T cell functions in anti-tumor responses. Such approaches started under experimental animal models, but they recently culminated into clinical studies, with striking effects in re-establishing T cell immunity in immunocompromised patients, as well as boosting anti-tumor effects. Depending on the design, glycosylation, and the structure of recombinantly engineered IL-7 proteins and their mimetics, recombinant IL-7 molecules have shown dramatic differences in their stability, efficacy, cellular effects, and overall immune functions. The current review is aimed to summarize the past and present efforts in the field that led to clinical trials, and to highlight the therapeutical significance of IL-7 biology as a master regulator of T cell immunity.

• IMMUNE NETWORK
• /
• v.22 no.6
• /
• pp.48.1-48.13
• /
• 2022

With the spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants, which are randomly mutated, the dominant strains in regions are changing globally. The development of preclinical animal models is imperative to validate vaccines and therapeutics against SARS-CoV-2 variants. The objective of this study was to develop a non-human primate (NHP) model for SARS-CoV-2 Delta variant infection. Cynomolgus macaques infected with Delta variants showed infectious viruses and viral RNA in the upper (nasal and throat) and lower respiratory (lung) tracts during the acute phase of infection. After 3 days of infection, lesions consistent with diffuse alveolar damage were observed in the lungs. For cellular immune responses, all macaques displayed transient lymphopenia and neutrophilia in the early stages of infection. SARS-CoV-2 Delta variant spike protein-specific IgM, IgG, and IgA levels were significantly increased in the plasma of these animals 14 days after infection. This new NHP Delta variant infection model can be used for comparative analysis of the difference in severity between SARS-CoV-2 variants of concern and may be useful in the efficacy evaluation of vaccines and universal therapeutic drugs for mutations.

• IMMUNE NETWORK
• /
• v.23 no.2
• /
• pp.15.1-15.17
• /
• 2023

Innate lymphoid cells (ILCs) are critical immune-response mediators. Although they largely reside in mucosal tissues, the kidney also bears substantial numbers. Nevertheless, kidney ILC biology is poorly understood. BALB/c and C57BL/6 mice are known to display type-2 and type-1 skewed immune responses, respectively, but it is unclear whether this extends to ILCs. We show here that indeed, BALB/c mice have higher total ILCs in the kidney than C57BL/6 mice. This difference was particularly pronounced for ILC2s. We then showed that three factors contributed to the higher ILC2s in the BALB/c kidney. First, BALB/c mice demonstrated higher numbers of ILC precursors in the bone marrow. Second, transcriptome analysis showed that compared to C57BL/6 kidneys, the BALB/c kidneys associated with significantly higher IL-2 responses. Quantitative RT-PCR also showed that compared to C57BL/6 kidneys, the BALB/c kidneys expressed higher levels of IL-2 and other cytokines known to promote ILC2 proliferation and/or survival (IL-7, IL-33, and thymic stromal lymphopoietin). Third, the BALB/c kidney ILC2s may be more sensitive to the environmental signals than C57BL/6 kidney ILC2s since they expressed their transcription factor GATA3 and the IL-2, IL-7, and IL-25 receptors at higher levels. Indeed, they also demonstrated greater responsiveness to IL-2 than C57BL/6 kidney ILC2s, as shown by their greater STAT5 phosphorylation levels after culture with IL-2. Thus, this study demonstrates previously unknown properties of kidney ILC2s. It also shows the impact of mouse strain background on ILC2 behavior, which should be considered when conducting research on immune diseases with experimental mouse models.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.119-126
• /
• 2024

In building information model (BIM), it is difficult to train an artificial intelligence (AI) model due to the lack of sufficient data about individual projects in an architecture firm. In this paper, we present a methodology to correctly train an AI neural network model based on a large language model (LLM) to predict the steel structure product weight ratios in BIM. The proposed method, with the aid of the LLM, can overcome the inherent problem of limited data availability in BIM and handle a combination of natural language and numerical data. The experimental results showed that the proposed method demonstrated significantly higher accuracy than methods based on a smaller language model. The potential for effectively applying large language models in BIM is confirmed, leading to expectations of preventing building accidents and efficiently managing construction costs.

• IMMUNE NETWORK
• /
• v.20 no.4
• /
• pp.33.1-33.19
• /
• 2020

We tested how adjuvants effect in a cancer vaccine model using an epitope derived from an autoactivation loop of membrane-type protease serine protease 14 (PRSS14; loop metavaccine) in mouse mammary tumor virus (MMTV)-polyoma middle tumor-antigen (PyMT) system and in 2 other orthotopic mouse systems. Earlier, we reported that loop metavaccine effectively prevented progression and metastasis regardless of adjuvant types and TH types of hosts in tail-vein injection systems. However, the loop metavaccine with Freund's complete adjuvant (CFA) reduced cancer progression and metastasis while that with alum, to our surprise, were adversely affected in 3 tumor bearing mouse models. The amounts of loop peptide specific antibodies inversely correlated with tumor burden and metastasis, meanwhile both T_H1 and T_H2 isotypes were present regardless of host type and adjuvant. Tumor infiltrating myeloid cells such as eosinophil, monocyte, and neutrophil were asymmetrically distributed among 2 adjuvant groups with loop metavaccine. Systemic expression profiling using the lymph nodes of the differentially immunized MMTV-PyMT mouse revealed that adjuvant types, as well as loop metavaccine can change the immune signatures. Specifically, loop metavaccine itself induces T_H2 and T_H17 responses but reduces T_H1 and T_reg responses regardless of adjuvant type, whereas CFA but not alum increased follicular T_H response. Among the myeloid signatures, eosinophil was most distinct between CFA and alum. Survival analysis of breast cancer patients showed that eosinophil chemokines can be useful prognostic factors in PRSS14 positive patients. Based on these observations, we concluded that multiple immune parameters are to be considered when applying a vaccine strategy to cancer patients.

• IMMUNE NETWORK
• /
• v.19 no.1
• /
• pp.7.1-7.11
• /
• 2019

Sjögren's syndrome (SS) is a chronic inflammatory autoimmune disorder that affects mainly salivary and lacrimal glands, but its cause remains largely unknown. Clinical data indicating that SS occurs in a substantial proportion of patients with lupus points to common pathogenic mechanisms underlying the two diseases. To address this idea, we asked whether SS develops in the lupus-prone mouse strain sanroque (SAN). Owing to hyper-activation of follicular helper T (Tfh) cells, female SAN mice developed lupus-like symptoms at approximately 20 wk of age but there were no signs of SS at that time. However, symptoms typical of SS were evident at approximately 40 wk of age, as judged by reduced saliva flow rate, sialadenitis, and IgG deposits in the salivary glands. Increases in serum titers of SS-related autoantibodies and numbers of autoantibody-secreting cells in cervical lymph nodes (LNs) preceded the pathologic manifestations of SS and were accompanied by expansion of Tfh cells and their downstream effector cells. Thus, our results suggest that chronic dysregulation of Tfh cells in salivary gland-draining LNs is sufficient to drive the development of SS in lupus-prone mice.

• IMMUNE NETWORK
• /
• v.20 no.5
• /
• pp.36.1-36.13
• /
• 2020

Hippo signaling pathways are evolutionarily conserved signal transduction mechanisms mainly involved in organ size control, tissue regeneration, and tumor suppression. However, in mammals, the primary role of Hippo signaling seems to be regulation of immunity. As such, humans with null mutations in STK4 (mammalian homologue of Drosophila Hippo; also known as MST1) suffer from recurrent infections and autoimmune symptoms. Although dysregulated T cell homeostasis and functions have been identified in MST1-deficient human patients and mouse models, detailed cellular and molecular bases of the immune dysfunction remain to be elucidated. Although the canonical Hippo signaling pathway involves transcriptional co-activator Yes-associated protein (YAP) or transcriptional coactivator with PDZ motif (TAZ), the major Hippo downstream signaling pathways in T cells are YAP/TAZ-independent and they widely differ between T cell subsets. Here we will review Hippo signaling mechanisms in T cell immunity and describe their implications for immune defects found in MST1-deficient patients and animals. Further, we propose that mutual inhibition of Mst and Akt kinases and their opposing roles on the stability and function of forkhead box O and β-catenin may explain various immune defects discovered in mutant mice lacking Hippo signaling components. Understanding these diverse Hippo signaling pathways and their interplay with other evolutionarily-conserved signaling components in T cells may uncover molecular targets relevant to vaccination, autoimmune diseases, and cancer immunotherapies.

• IMMUNE NETWORK
• /
• v.20 no.6
• /
• pp.51.1-51.17
• /
• 2020

Respiratory syncytial virus (RSV) causes severe pulmonary disease in infants, young children, and the elderly. Formalin inactivated RSV (FI-RSV) vaccine trials failed due to vaccine enhanced respiratory disease, but the underlying immune mechanisms remain not fully understood. In this study, we have used wild type C57BL/6 and CD4 knockout (CD4KO) mouse models to better understand the roles of the CD4 T cells and cellular mechanisms responsible for enhanced respiratory disease after FI-RSV vaccination and RSV infection. Less eosinophil infiltration and lower pro-inflammatory cytokine production were observed in FI-RSV vaccinated CD4KO mice after RSV infection compared to FI-RSV vaccinated C57BL/6 mice. NK cells and cytokine-producing CD8 T cells were recruited at high levels in the airways of CD4KO mice, correlating with reduced respiratory disease. Depletion studies provided evidence that virus control was primarily mediated by NK cells whereas CD8 T cells contributed to IFN-γ production and less eosinophilic lung inflammation. This study demonstrated the differential roles of effector CD4 and CD8 T cells as well as NK cells, in networking with other inflammatory infiltrates in RSV disease in immune competent and CD4-deficient condition.

• Journal of the Korea Society for Simulation
• /
• v.33 no.1
• /
• pp.19-26
• /
• 2024

The phase-type, PH, distribution is defined as the time to absorption into a terminal state in a continuous-time Markov chain. As the PH distribution includes family of exponential distributions, it has been widely used in stochastic models. Since the PH distribution is represented and generated by an initial probability vector and a generator matrix which is called the Markovian representation, we need to find a vector and a matrix that are consistent with given set of moments if we want simulate a PH distribution. In this paper, we propose an approach to simulate a PH distribution based on distribution function which can be obtained directly from moments. For the simulation of PH distribution of order 2, closed-form formula and streamlined procedures are given based on the Jordan decomposition and the minimal Laplace transform which is computationally more efficient than the moment matching methods for the Markovian representation. Our approach can be used more effectively than the Markovian representation in generating higher order PH distribution in queueing network simulation.

• Convergence Security Journal
• /
• v.23 no.4
• /
• pp.61-70
• /
• 2023

Clustering is an unsupervised learning method that involves grouping data based on features such as distance metrics, using data without known labels or ground truth values. This method has the advantage of being applicable to various types of data, including images, text, and audio, without the need for labeling. Traditional clustering techniques involve applying dimensionality reduction methods or extracting specific features to perform clustering. However, with the advancement of deep learning models, research on deep clustering techniques using techniques such as autoencoders and generative adversarial networks, which represent input data as latent vectors, has emerged. In this study, we propose a deep clustering technique based on deep learning. In this approach, we use an autoencoder to transform the input data into latent vectors, and then construct a vector space according to the cluster structure and perform k-means clustering. We conducted experiments using the MNIST and Fashion-MNIST datasets in the PyTorch machine learning library as the experimental environment. The model used is a convolutional neural network-based autoencoder model. The experimental results show an accuracy of 89.42% for MNIST and 56.64% for Fashion-MNIST when k is set to 10.