The conserved enzyme aminolevulinic acid synthase (ALAS) initiates heme biosynthesis in certain bacteria and eukaryotes by catalyzing the condensation of glycine and succinyl-CoA to yield aminolevulinic acid. In humans, the ALAS isoform responsible for heme production during red blood cell development is the erythroid-specific ALAS2 isoform. Owing to its essential role in erythropoiesis, changes in human ALAS2 (hALAS2) function can lead to two different blood disorders. X-linked sideroblastic anemia results from loss of ALAS2 function, while X-linked protoporphyria results from gain of ALAS2 function. Interestingly, mutations in the ALAS2 C-terminal extension can be implicated in both diseases. Here, we investigate the molecular basis for enzyme dysfunction mediated by two previously reported C-terminal loss-of-function variants, hALAS2 V562A and M567I. We show that the mutations do not result in gross structural perturbations, but the enzyme stability for V562A is decreased. Additionally, we show that enzyme stability moderately increases with the addition of the pyridoxal 5\'-phosphate (PLP) cofactor for both variants. The variants display differential binding to PLP and the individual substrates compared to wild-type hALAS2. Although hALAS2 V562A is a more active enzyme in vitro, it is less efficient concerning succinyl-CoA binding. In contrast, the M567I mutation significantly alters the cooperativity of substrate binding. In combination with previously reported cell-based studies, our work reveals the molecular basis by which hALAS2 C-terminal mutations negatively affect ALA production necessary for proper heme biosynthesis.

G protein pathway suppressor 2 (GPS2) has been shown to play a pivotal role in human and mouse definitive erythropoiesis in an EKLF-dependent manner. However, whether GPS2 affects human primitive erythropoiesis is still unknown. This study demonstrated that GPS2 positively regulates erythroid differentiation in K562 cells, which have a primitive erythroid phenotype. Overexpression of GPS2 promoted hemin-induced hemoglobin synthesis in K562 cells as assessed by the increased percentage of benzidine-positive cells and the deeper red coloration of the cell pellets. In contrast, knockdown of GPS2 inhibited hemin-induced erythroid differentiation of K562 cells. GPS2 overexpression also enhanced erythroid differentiation of K562 cells induced by cytosine arabinoside (Ara-C). GPS2 induced hemoglobin synthesis by increasing the expression of globin and ALAS2 genes, either under steady state or upon hemin treatment. Promotion of erythroid differentiation of K562 cells by GPS2 mainly relies on NCOR1, as knockdown of NCOR1 or lack of the NCOR1-binding domain of GPS2 potently diminished the promotive effect. Thus, our study revealed a previously unknown role of GPS2 in regulating human primitive erythropoiesis in K562 cells.

Inherited iron metabolism defects are possibly missed or underdiagnosed in iron-deficient endemic settings because of a lack of awareness or a methodical screening approach. Hence, we systematically evaluated anemia cases (2019 to 2021) based on clinical phenotype, normal screening tests (high-performance liquid chromatography, α gene sequencing, erythrocyte sedimentation rate, C-reactive protein, and tissue transglutaminase), and abnormal iron profile by targeted next-generation sequencing (26-gene panel) supplemented with whole-exome sequencing, multiplex ligation probe amplification/mitochondrial DNA sequencing, and chromosomal microarray. Novel variants in ALAS2, STEAP3, and HSPA9 genes were functionally validated. A total of 290 anemia cases were screened, and 41 (14%) enrolled for genomic testing as per inclusion criteria. Comprehensive genomic testing revealed pathogenic variants in 23 of 41 cases (56%). Congenital sideroblastic anemia was the most common diagnosis (14/23; 61%), with pathogenic variations in ALAS2 (n = 6), SLC25A38 (n = 3), HSPA9 (n = 2) and HSCB, SLC19A2, and mitochondrial DNA deletion (n = 1 each). Nonsideroblastic iron defects included STEAP3-related microcytic anemia (2/23; 8.7%) and hypotransferrenemia (1/23; 4.3%). A total of 6 of 22 cases (27%) revealed a non-iron metabolism gene defect on whole-exome sequencing. Eleven novel variants (including variants of uncertain significance) were noted in 13 cases. Genotype-phenotype correlation revealed a significant association of frameshift/nonsense/splice variants with lower presentation age (0.8 months versus 9 years; P < 0.01) compared with missense variants. The systematic evaluation helped uncover an inherited iron defect in 41% (17/41) of cases, suggesting the need for active screening and awareness for these rare diseases in an iron-deficient endemic population.

Photodynamic therapy (PDT) is nowadays widely employed in cancer treatment. We sought to assess the efficacy of combining PDT with anti-programmed cell death protein 1 (PD1) and to investigate the associated mechanisms in nonsmall cell lung cancer (NSCLC). We established a xenograft tumor model in C57BL/6J mice using Lewis lung carcinoma (LLC) cells, recorded tumor growth, and quantified reactive oxygen species (ROS) levels using a ROS detection kit. Pathological changes were assessed through H&E staining, while immunofluorescence (IF) was used to determine the expression of CD8 and Foxp3. Transcriptomic analysis was conducted, analyzing differential expressed genes (DEGs) among control, PDT, and PDT combined with anti-PD1 (PDT+anti-PD1) groups. Functional enrichment analysis via Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) was performed. The Cancer Genome Atlas (TCGA) database was utilized to analyze the expression of aminolevulinate synthase gene (ALAS2), integrin alpha10 (ITGA10), ATP1A2, a disintegrin and metalloprotease 12 (ADAM12), and Lox1 in lung adenocarcinoma and adjacent tissues, with concurrent immune infiltration analysis. Quantitative real-time polymerase chain reaction and western blot were employed to measure mRNA and protein expression levels. Treatment with PDT combined with anti-PD1 significantly inhibited tumor growth and increased the number of CD8+ cells while decreasing Foxp3+ cells. Immune infiltration results presented ALAS2, ADAM12, and ITGA10 were associated with various types of T cells or macrophages. Additionally, the expression levels of EGFR, ERK, and PI3K/Akt were suppressed after PDT with anti-PD1 treatment. Our findings collectively suggest that PDT combined with anti-PD1 treatment could enhance the infiltration of CD8+ T cells, suppressing tumor growth, and this effect was associated with ALAS2, ITGA10, and ADAM12. The underlying mechanism might be linked to EGFR, ERK, and PI3K/Akt signaling. Overall, this study provides valuable insights into the application of PDT combined with anti-PD1 treatment in NSCLC.

Congenital sideroblastic anemia (CSA) is a group of disorders caused by different genetic mutations that result in low iron utilization and ineffective erythropoiesis. Current treatments are limited, and some patients do not respond to vitamin B6 therapy. Luspatercept is a novel erythropoietic maturation agent approved for adult β-thalassemia and Myelodysplastic syndromes with ring sideroblasts (MDS-RS) associated with ineffective erythropoiesis. Here we report 2 patients with CSA due to mutations in ALAS2 and SLC25A38 genes who became unresponsive after a period of treatment with vitamin B6 and iron chelators but achieved transfusion independence and a markedly reduced spleen after combination with luspatercept.

Cold agglutinin disease (CAD) is a rare B-cell lymphoproliferative disorder of the bone marrow, manifested by autoimmune hemolytic anemia caused by binding of monoclonal IgM autoantibodies to the I antigen. Underlying genetic changes have previously been reported, but their impact on gene expression profile has been unknown. Here, we define differentially expressed genes in CAD B cells. To unravel downstream alteration in cellular pathways, gene expression by RNA sequencing was undertaken. Clonal B-cell samples from 12 CAD patients and IgM-expressing memory B cells from 4 healthy individuals were analyzed. Differential expression analysis and filtering resulted in 93 genes with significant differential expression. Top upregulated genes included SLC4A1, SPTA1, YBX3, TESC, HBD, AHSP, TRAF1, HBA2, RHAG, CA1, SPTB, IL10, UBASH3B, ALAS2, HBA1, CRYM, RGCC, KANK2, and IGHV4-34. They were upregulated at least 8-fold, while complement receptor 1 (CR1/CD35) was downregulated 11-fold in clonal CAD B cells compared to control B cells. Flow cytometry analyses further confirmed reduced CR1 (CD35) protein expression by clonal CAD IgM+ B cells compared to IgM+ memory B cells in controls. CR1 (CD35) is an important negative regulator of B-cell activation and differentiation. Therefore, reduced CR1 (CD35) expression may increase activation, proliferation, and antibody production in CAD-associated clonal B cells.

High-throughput sequencing was used to screen expressing differences of miRNA, lncRNA, and mRNA in CD19+ B peripheral blood samples of newly diagnosed immune thrombocytopenia (ITP) patients and healthy controls. The study aimed to explore the regulatory role of ceRNA network in the pathogenesis of dysfunctional CD19 + B lymphocytes of ITP patients.
CD19+ B lymphocytes were isolated from peripheral blood samples of ITP patients and their healthy counterparts. High-throughput sequencing was used to screen for the expression of miRNA, lncRNA, and mRNA of ITP patients and healthy controls, which were analysed by the ceRNA network. Moreover, qPCR was used to verify the differential expression of miRNA, lncRNA, and mRNA in ITP patients and healthy controls. The correlation between differentially expressed miRNA, lncRNA, mRNA, and B lymphocyte subsets was also analysed.
The CD19+ B lymphocytes of 4 newly diagnosed ITP patients and 4 healthy controls were sequenced and analysed. There were 65 differentially expressed lncRNA and 149 mRNA forming a ceRNA network showed that 12 lncRNA and 136 differentially expressed mRNA were closely associated. Similarly, miR-144-3p, miR-374c-3p, and miR-451a were highly expressed in ITP patients, as confirmed by qPCR, which was consistent with the high-throughput sequence results. LOC102724852 and CCL20 were highly expressed in ITP patients, while LOC105378901, LOC112268311, ALAS2, and TBC1D3F were not as compared to healthy controls, which was consistent with the high-throughput sequence results. In addition, the expression of miR-374c-3p, LOC112268311, LOC105378901, and CXCL3 were correlated with the percentage of B lymphocyte subsets.
The ceRNA network of miRNA, lncRNA, and mRNA in peripheral CD19 + B lymphocytes plays an essential role in the pathogenesis of ITP.

Feature selection is a critical step for translating advances afforded by systems-scale molecular profiling into actionable clinical insights. While data-driven methods are commonly utilized for selecting candidate genes, knowledge-driven methods must contend with the challenge of efficiently sifting through extensive volumes of biomedical information. This work aimed to assess the utility of large language models (LLMs) for knowledge-driven gene prioritization and selection.
In this proof of concept, we focused on 11 blood transcriptional modules associated with an Erythroid cells signature. We evaluated four leading LLMs across multiple tasks. Next, we established a workflow leveraging LLMs. The steps consisted of: (1) Selecting one of the 11 modules; (2) Identifying functional convergences among constituent genes using the LLMs; (3) Scoring candidate genes across six criteria capturing the gene\'s biological and clinical relevance; (4) Prioritizing candidate genes and summarizing justifications; (5) Fact-checking justifications and identifying supporting references; (6) Selecting a top candidate gene based on validated scoring justifications; and (7) Factoring in transcriptome profiling data to finalize the selection of the top candidate gene.
Of the four LLMs evaluated, OpenAI\'s GPT-4 and Anthropic\'s Claude demonstrated the best performance and were chosen for the implementation of the candidate gene prioritization and selection workflow. This workflow was run in parallel for each of the 11 erythroid cell modules by participants in a data mining workshop. Module M9.2 served as an illustrative use case. The 30 candidate genes forming this module were assessed, and the top five scoring genes were identified as BCL2L1, ALAS2, SLC4A1, CA1, and FECH. Researchers carefully fact-checked the summarized scoring justifications, after which the LLMs were prompted to select a top candidate based on this information. GPT-4 initially chose BCL2L1, while Claude selected ALAS2. When transcriptional profiling data from three reference datasets were provided for additional context, GPT-4 revised its initial choice to ALAS2, whereas Claude reaffirmed its original selection for this module.
Taken together, our findings highlight the ability of LLMs to prioritize candidate genes with minimal human intervention. This suggests the potential of this technology to boost productivity, especially for tasks that require leveraging extensive biomedical knowledge.

UNASSIGNED: : Erythroid cells play important roles in hemostasis and disease. However, there is still significant knowledge gap regarding stress erythropoiesis.
UNASSIGNED: : Two single-cell RNAseq datasets of erythroid cells on GEO with accession numbers GSE149938 and GSE184916 were obtained. The datasets from two sources, bone marrow and peripheral blood were analyzed using Seurat v4.1.1, and other tools in R. QC metrics were performed, data were normalized and scaled. Principal components that capture the variation of the data were determined. In clustering the cells, KNN graph was constructed and Louvain algorithm was applied to optimize the standard modularity function. Clusters were defined via differential expression of features.
UNASSIGNED: We identified 9 different cell types, with a particular cluster representing the stress erythroids. The clusters showed differentially expressed genes as observed from the gene signature plot. The stress erythroid cluster differentially expressed some genes including ALAS2, HEMGN, and GUK1.
UNASSIGNED: The erythroid population was found to be heterogeneous, with a distinct sub-cell type constituting the stress erythroids; this may have important implications for our knowledge of steady-state and stress erythropoiesis, and the markers found in this cluster may prove useful for future research into the dynamics of stress erythroid progenitor cell differentiation.

Hereditary hemolytic anemia (HHA) refers to a heterogeneous group of genetic disorders that share one common feature: destruction of circulating red blood cells (RBCs). The destruction of RBCs may be due to membranopathies, enzymopathies, or hemoglobinopathies. Because these are genetic disorders, incorporation of next-generation sequencing (NGS) has facilitated the diagnostic process of HHA.
Genetic data from 29 patients with suspected hereditary anemia in a tertiary hospital were retrospectively reviewed to evaluate the efficacy of NGS on hereditary anemia diagnosis. Targeted NGS was performed with custom probes for 497 genes associated with hematologic disorders. After genomic DNA was extracted from peripheral blood, prepared libraries were hybridized with capture probes and sequenced using NextSeq 550Dx (Illumina, San Diego, CA, USA).
Among the 29 patients, ANK1 variants were detected in five, four of which were pathogenic or likely pathogenic variants. SPTB variants were detected in six patients, five of which were classified as pathogenic or likely pathogenic variants. We detected g6pd pathogenic and spta1 likely pathogenic variants in two patients and one patient, respectively. Whole-gene deletions in both HBA1 and HBA2 were detected in two patients, while only HBA2 deletion was detected in one patient. One likely pathogenic variant in PLKR was detected in one patient, and one likely pathogenic variant in ALAS2 was detected in another.
Here, NGS played a critical role in definitive diagnosis in 18 out of 29 patients (62.07%) with suspected HHA. Thus, its incorporation into the diagnostic workflow is crucial.