Biology subjects
Publications and source records attributed to Hammer, R..
Withdrawal StatementThe authors have withdrawn this preprint because the co-authors were not in agreement about posting the work in its current form. The authors request that this work not be cited. Please contact the corresponding author (Dr. Xunlei Kang, kangxu@health.missouri.edu) with questions.
Tumor-specific T-cells are key in combating cancer as shown in adoptive cell therapy with tumor infiltrating lymphocytes (TILs) and checkpoint inhibitor therapy. Studies in many types of cancer have shown that preexisting tumor reactive T-cells are not only tumor-but typically also patient-specific, requiring personalized treatment options. For viral infections, public T-cell receptors (TCRs) with substantial sequence homologies suggest shared immune-dominant targets in human leucocyte antigen (HLA)-matched individuals. We hypothesized that also in the complex TCR repertoires of tumors subsets of tumor-specific TCRs exist that can be found in different patients with identical or near identical TCRs. This paper presents a TCR-V(D)J-sequence clustering approach identifying clusters of tumor-specific common TCRs mainly from TILs of non-small cell lung cancer (NSCLC) patients. Using two TCR-clusters as examples, we show that T-cells engineered genetically to only express those common TCRs recognized HLA-matched allogeneic tumor cell lines in a cluster typical manner. Recognition of allogeneic tumors was dependent on the HLA allele inferred by the cluster and could be blocked by HLA antibodies. In addition to NSCLC, TCR repertoire analyses in pancreatic ductal adenocarcinoma and a smaller number of breast and colorectal cancer samples revealed TCRs highly homologous or even identical to NSCLC cluster TCRs. TCR-T cells expressing TCRs from these tumors assigned to a specific cluster recognized allogeneic tumor lines in the expected cluster-typical manner. These findings suggest a pan-cancer therapeutic potential of tumor-specific common TCRs.
Adoptive cell therapy (ACT) with TCR-engineered T-cells represents a promising alternative to TIL- or CAR-T therapies for patients with advanced solid cancers. Currently, selection of therapeutic TCRs critically depends on knowing the target antigens, a condition excluding most patients from treatment. Direct antigen-agnostic identification of tumor-specific T-cell clonotypes and TCR-T manufacturing using their TCRs can advance ACT for patients with aggressive solid cancers. We present a method to identify tumor-specific clonotypes from surgical specimens by comparing TCR{beta}-chain repertoires of TILs and adjacent tissue-resident lymphocytes. In seven NSCLC-patients, tumor-specific clonotypes were selected based on TIL-abundance and high tumor-to-nontumor frequency ratios. In two of the patients, we demonstrate that predicted tumor-specific clonotypes reacted against autologous tumors. In a third patient, we engineered TCR T-cells with four candidate tumor-specific TCRs that showed reactivity against the patients tumor and HLA-matched NSCLC cell lines. The TCR-T cells were then used to screen for candidate neoantigens and aberrantly expressed antigens. Three TCRs recognized recurrent driver-mutation KRAS Q61H-peptide ILDTAGHEEY presented by HLA-A*01:01. The TCRs were also dominant in a tumor relapse, one was found in cell free DNA. The finding of homologous TCRs in independent KRAS Q61H-positive cancers suggests a therapeutic opportunity for HLA-matched patients with KRAS Q61H-expressing tumors.
The processes that govern leukemia progression and remission are poorly understood. Our research reveals that the CXCL12 gradient, traditionally associated with LSC quiescence and survival, critically determines LSC localization and associated behavior. Specifically, CXCL12 guides LSCs to either the quiescent niche in the metaphysis, characterized by N-cadherin-expressing mesenchymal stromal cells (N-cad+ MSCs), or the proliferative niche in the central marrow (CM), marked by sinusoidal endothelial cells and associated stromal cells. We identified that the CXCL12 gradient is finely regulated by the interplay between dipeptidyl peptidase 4 (DPP4) on LSCs and glypican-3 (GPC3) on N-cad+ MSCs. DPP4 deactivates CXCL12, while GPC3 inhibits DPP4, resulting in a higher CXCL12 concentration in the metaphysis and a lower concentration in the CM. This differential gradient facilitates leukemia progression by promoting LSC quiescence and survival in the metaphysis versus relative proliferation and apoptosis in the CM. Depletion of Dpp4 from LSCs or Cxcl12 from N-cad+ MSCs disrupts this gradient, mobilizing LSCs from the metaphysis to the CM and significantly hindering leukemia development. Our findings redefine the role of CXCL12 in LSC behavior and provide a clearer understanding of leukemia progression. This novel insight highlights the potential for targeted therapeutic strategies that disrupt the CXCL12 gradient to treat minimal residual LSCs, offering a promising path toward a lasting cure for acute myeloid leukemia (AML).
ChatGPT has demonstrated its potential as a surrogate knowledge graph. Trained on extensive data sources, including open-access publications, peer-reviewed research articles and biomedical websites, ChatGPT extracted information on gene relationships and biological pathways. However, a major challenge is model hallucination, i.e., high false positive rates. To assess and address this challenge, we systematically evaluated ChatGPTs capacity for predicting gene relationships using GPT-3.5-turbo and GPT-4. Benchmarking against the KEGG Pathway Database as the ground truth, we experimented with diverse prompting strategies, targeting gene relationships of activation, inhibition, and phosphorylation. We introduced an innovative iterative prompt refinement technique. By assessing prompt efficacy using metrics like F-1 score, precision, and recall, GPT-4 was re-engaged to suggest improved prompts. A refined prompt, which combines a specialized role with explanatory text, significantly enhances the performance. Going beyond pairwise gene relationships, we also deciphered complex gene interplays, such as gene interaction chains and pathways pertinent to diseases like non-small cell lung cancer. Direct prompts showed limited success, but "least-to-most" prompting exhibited significant potentials for such network constructions. The methods in this study may be used for some other bioinformatics prediction problems.
In biomedical literature, biological pathways are commonly described through a combination of images and text. These pathways contain valuable information, including genes and their relationships, which provide insight into biological mechanisms and precision medicine. Curating pathway information across the literature enables the integration of this information to build a comprehensive knowledge base. While some studies have extracted pathway information from images and text independently, they often overlook the correspondence between the two modalities. In this paper, we present a pathway figure curation system named pathCLIP for identifying genes and gene relations from pathway figures. Our key innovation is the use of an image-text contrastive learning model to learn coordinated embeddings of image snippets and text descriptions of genes and gene relations, thereby improving curation. Our validation results, using pathway figures from PubMed, showed that our multimodal model outperforms models using only a single modality. Additionally, our system effectively curates genes and gene relations from multiple literature sources. A case study on extracting pathway information from non-small cell lung cancer literature further demonstrates the usefulness of our curated pathway information in enhancing related pathways in the KEGG database.
The long-lasting global COVID-19 pandemic demands timely genomic investigation of SARS-CoV-2 viruses. Here we report a simple and efficient workflow for whole genome sequencing utilizing one-step RT-PCR amplification on a microfluidic platform, followed by MiSeq amplicon sequencing. The method uses Fluidigm IFC and instruments to amplify 48 samples with 39 pairs of primers in a single step. Application of this method on RNA samples from both viral isolate and clinical specimens demonstrate robustness and efficiency of this method in obtaining the full genome sequence of SARS-CoV-2.