Single-cell transcriptomics reveals continuous phenotypic spectrum of macrophages in inflammatory bowel disease
Background: Limitations of the Dichotomous M1/M2 Static Classification System and the Bottleneck of Macrophage Heterogeneity Data in Inflammatory Bowel Disease (IBD) R&D
Inflammatory bowel disease (IBD), encompassing Crohn's disease and ulcerative colitis, is a chronic mucosal inflammatory disorder resulting from the complex interplay of genetic predisposition, intestinal microbial dysbiosis, epithelial barrier disruption, and aberrant immune activation. Current IBD therapeutic development pipelines have relied on a static guideline that classifies macrophages into a simplistic dichotomy of pro-inflammatory M1 and anti-inflammatory M2. However, this reductionist framework inherently lacks the ability to capture the continuous spectrum of macrophage subsets operating within the intestinal mucosal microenvironment—inflammatory, regulatory, reparative, and fibrosis-associated functional groups—thereby creating a critical blind spot. Specifically, the noise generated by the loss of spatial coordinate information during tissue dissociation, the translational gap arising from mouse-human interspecies transcriptomic differences, and the inability to in silico control the feedback flux generated after administration of TNF-α/IL-23/JAK-STAT axis inhibitors (e.g., infliximab, risankizumab, upadacitinib) have repeatedly led to failures in achieving effective mucosal healing concentrations, thus creating a data barrier. Even in the IBD market, where global multinational pharmaceutical companies such as AbbVie, Takeda, and Johnson & Johnson invest over $20 billion annually, clinical bottlenecks persist, with primary non-response rates of 30-40% and secondary loss rates of 13-24% per year. This is directly related to the resolution limitations of existing bulk omics baselines, which fail to capture the multidimensional tensor structure of macrophage plasticity.
Discovery: Implementation of Single-Cell/Spatial Transcriptome Multimodal Algorithms and Demonstration of Intestinal Macrophage Subset Resolution Tensor Synchronization
The simultaneous operation of 10x Genomics Chromium-based scRNA-seq and Visium/MERFISH/CosMx SMI spatial transcriptomic platforms successfully deconstructed intestinal mucosal macrophages into at least 6-8 functional clusters, including CD68⁺CD163⁺ resident, TREM2⁺ lipid metabolism-associated, SPP1⁺ fibrosis-promoting, and FCN1⁺ monocyte-derived inflammatory subsets. By in silico preemptive calculation of the binding free energy landscape of transcription factors (IRF5, STAT6, NF-κB p65, PPARG) between each subset using differential equation-based kinetics (RNA velocity, scVelo), the phenotypic transition trajectory from normal mucosa to active inflammation to remission repair was topologically elucidated. The application of batch effect removal algorithms such as Harmony and scVI eliminated technical variations between multi-institutional cohorts, demonstrating a resolution that surpasses the existing M1/M2 binary model. In particular, TREM2⁺SPP1⁺ double-positive subsets function as fibrosis-inflammation crosstalk hubs, and their critical role in determining the topological variation of downstream transcriptional networks of the Wnt/β-catenin and Notch signaling pathways was cross-validated in multiple independent datasets (Human Cell Atlas IBD Consortium, CEDARS cohort).
Establishment of a Model for Fine-Grained Layered Regulation of Macrophage Phenotype Transition Pathways and Reversible Mucosal Immune Homeostasis
A multi-omics-based patient molecular phenotype precision stratification was established by integrating scRNA-seq, spatial transcriptomics, and cytokine proteomics matrices through tensor decomposition. This allows for the stratification of patients with the same IBD diagnosis at the molecular level into fibrosis-dominant, inflammation-dominant, and mixed types, and enables the allocation of optimized macrophage-targeted therapeutic strategies—chimeric antigen receptor macrophages (CAR-M, Carisma Therapeutics CT-0508 lead), adoptive macrophage transfer, receptor-targeted therapies (CSF1R inhibitor pexidartinib, anti-CD47/SIRPα axis), and nanoparticle-based drug delivery systems (PLGA/liposome nanocarrier for macrophage-selective endocytosis)—for each stratum. By simultaneously implementing the rate-limiting step constants of IL-10/TGF-β anti-inflammatory signal upregulation and TNF-α/IL-1β pro-inflammatory signal downregulation through CRISPR-Cas9-based gene editing and regulation of microbiome metabolites (e.g., short-chain fatty acids such as butyrate and propionate), a backbone is established for reversibly and autonomously regulating effective mucosal homeostasis even under aberrant immune stress conditions. An intestinal organoid-macrophage co-culture platform and biomaterial-assisted scaffold (decellularized ECM, hydrogel) technology are integrated as a preclinical validation infrastructure for this layered model.
Prospects: Establishment of a Programmable Mucosal Immunology Standard and Implementation of Next-Generation IND Digital Governance
The declaration of a complete reset of IBD R&D governance from a static, post-hoc symptomatic system to an AI-powered, multidimensional tensor-based programmable infrastructure is entering the implementation phase. Following Pfizer's JAK inhibitor tofacitinib, second-generation selective JAK1 inhibitors (upadacitinib, filgotinib), anti-IL-23p19 antibodies (guselkumab Phase 3 GALAXI/QUASAR, mirikizumab), and S1P receptor modulators (ozanimod, etrasimod) are expanding the pipelines of global multinational pharmaceutical companies. By linking single-cell resolution macrophage subset mapping data to high-throughput screening (HTS) stage genetic gradient correction coefficients, a computational moat is established to zero out batch-to-batch phenotypic transition variations. This platform meets the FDA/EMA digital healthcare companion diagnostic (CDx) specifications—macrophage subset biomarker panel (TREM2, SPP1, CD163 quantitative cutoffs)—and functions as a master asset that maximizes patient selection precision in the IND (Investigational New Drug) regulatory approval framework, structurally reducing clinical trial failure rates and disruptively shortening approval timelines. The 2025 release of Carisma Therapeutics' CAR-M platform Phase 1 data in solid tumors and the establishment of a real-time integrated pipeline of Oxford Nanopore long-read sequencing and spatial transcriptomics demonstrate that this governance transition has already passed an irreversible tipping point.
Inflammatory bowel disease (IBD), which mainly includes Crohn's disease and ulcerative colitis, is a chronic inflammatory disorder of the gastrointestinal tract characterized by recurrent episodes of intestinal inflammation. The development of IBD is influenced by multiple factors, including genetic predisposition, intestinal dysbiosis, epithelial barrier impairment, and abnormal immune activation. Among innate immune cells, macrophages are key regulators of intestinal immune homeostasis and are involved in inflammatory responses, tissue remodeling, and mucosal repair. Their ability to adopt different functional states in response to local environmental signals has made them an important focus of current therapeutic research in IBD. Traditionally, macrophages have been classified into pro-inflammatory M1 and anti-inflammatory M2 phenotypes. However, recent findings from single-cell transcriptomic and spatial analyses suggest that intestinal macrophages represent a far more diverse and dynamic population than this simplified classification implies. Multiple macrophage subsets with inflammatory, regulatory, reparative, and fibrosis-associated functions coexist within the intestinal microenvironment and contribute differently to disease progression and tissue healing. These observations highlight the importance of developing more selective and targeted macrophage-based therapeutic approaches. In this review, we discuss the current understanding of macrophage plasticity and its role in the pathogenesis of IBD. Particular attention is given to newer macrophage-targeted therapeutic strategies, including adoptive macrophage transfer, engineered macrophages, receptor-targeted therapies, nanoparticle-based delivery systems, microbiome modulation, microbial metabolite regulation, gene-editing approaches, organoid technologies, and biomaterial-assisted platforms. We also examine the contribution of macrophages to epithelial regeneration, mucosal healing, fibrosis, and intestinal
The elucidation of multidimensional macrophage plasticity in the intestinal microenvironment in this study goes beyond theoretical mucosal immunology mechanisms and directly translates into practical applications for global pharmaceutical pipelines and next-generation precision medicine therapeutic business lines.
First, by instantaneously scanning macrophage subset-specific phenotypic transition kinetics in the clinical setting using a single-cell transcriptomic AI scanning algorithm, the temporal gap in therapeutic decision-making for anti-TNF-unresponsive patients is eliminated, and a protective barrier for mucosal healing rates is established.
At the same time, by linking to open-source single-cell databases aggregated in the Human Cell Atlas IBD Consortium and GEO/ArrayExpress, a companion diagnostic (CDx) panel interface is realized that allows for virtual simulation of confounding variables in clinical trial design and real-time reverse calculation of effective docking concentrations for CSF1R and CD47/SIRPα axis-targeted antibodies.
Furthermore, when multinational companies conduct large-scale regulatory clinical trials for next-generation macrophage-targeted therapies, linking TREM2 and SPP1 expression quantitative values as correction coefficients maximizes the probability of obtaining FDA/EMA regulatory approval for clinical trial protocols and cGMP commercial manufacturing, functioning as a backbone infrastructure.