Tuberculosis, caused by Mycobacterium tuberculosis(MDR-TBMtb), is an infectious chronic disease that has plagued humanity for thousands of years (Furin et al., 2019). According to the 2024 WHO Global Tuberculosis Report, there were an estimated 10.8 million (95% CI: 10.1–11.7 million) new cases of tuberculosis worldwide in 2023, a small increase from 10.7 million in 2022 although still much higher than 10.4 million in 2021 and 10.1 million in 2020. The report highlights that drug-resistant TB continues to be a public health crisis, with approximately 400,000(95% UI: 360 000–440 000) new cases of Multidrug-Resistant Tuberculosis (MDR-TB) or rifampicin-resistant TB in 2023 (Organization W H, 2024). Undoubtedly, tuberculosis remains a major global health problem. Therefore, early diagnosis and treatment to curb the spread of tuberculosis are urgently needed. Recent advances in tuberculosis immunotherapy have opened new avenues for treatment. The M72/AS01E vaccine has shown 54% efficacy in preventing pulmonary TB in latently infected adults (Van Der Meeren et al., 2018). Cytokine-based therapies, including Interleukin-2 (IL-2) and Interferon-gamma (IFN-γ) supplementation, have demonstrated improved outcomes in MDR-TB patients (Grahmann and Braun, 2008; Tan et al., 2017). Additionally, nanotechnology-based drug delivery systems—such as liposomal formulations and polymeric nanoparticles—enable targeted, sustained, and specific delivery and may enhance TB therapy (Nair et al., 2023). Effective host immune responses can prevent the progression of tuberculosis in most cases, although some individuals may develop latent tuberculosis infection(LTBI). When the host immune function is compromised, latent tuberculosis may progress to active tuberculosis. Poor adherence to treatment regimens, improper use of anti-tuberculosis drugs, incorrect prescriptions by physicians, or poor drug quality can contribute to the development of drug-resistant tuberculosis. Some studies have found that in high-prevalence tuberculosis environments, there is still a small subset of individuals who possess innate resistance to developing latent Mycobacterium tuberculosis infection upon exposure (Stein et al., 2018). Therefore, it is important to gain a deeper understanding of host immune defense mechanisms against Mycobacterium tuberculosis and the current progress in tuberculosis immunotherapy. Single-cell sequencing allows for comprehensive analysis of immune cells at the single-cell level, and this technology can provide valuable information for tuberculosis diagnosis and precision treatment. This review aims to summarize the research progress of single-cell sequencing in tuberculosis immunotherapy, critically assess current limitations and methodological challenges, and propose future research priorities for clinical translation.

The host’s immune recognition, response, and regulation of Mtb determine the occurrence, development, and outcome of tuberculosis. Upon infecting the host, Mtb induces a series of innate and adaptive immune responses,while employing mechanisms to evade immune surveillance (Kilinç et al., 2021).

The emergence of single-cell genomics represents a turning point in cell biology. For the first time, we can determine the expression levels of each gene in the genome in thousands of individual cells within a single experiment. Sequencing is a core technology of genomics. Traditional second-generation sequencing technology typically provides the average transcriptome date of groups of cells, which can mask specific information among individual cells. As research progresses, it has become increasingly apparent that no two cells are identical, whether in structure or function. Therefore, traditional sequencing methods have failed to fully reveal the differences between cells. Since its advent in 2009, single-cell sequencing has become an important tool for studying the differences in cell populations and the evolutionary relationships of cells within organisms (Xue et al., 2015; Tang et al., 2019). Its core processes include the preparation of single-cell suspensions, single-cell isolation, nucleic acid amplification, high-throughput sequencing, and data analysis. Among these, single-cell isolation and nucleic acid amplification are crucial steps that directly affect the quality of subsequent data and the interpretation of results.

Single-cell sequencing technology can analyze the genomic, transcriptomic, and epigenomic maps of individual cells, revealing cellular heterogeneity, which is essential for understanding the pathogenesis of tuberculosis. However, obtaining high-quality single-cell samples from tuberculosis-infected tissues still faces many challenges. Firstly, during the sample preparation process, ensuring the activity, integrity, and low agglomeration rate of cells is vital. The processing methods for different types of samples vary, making it difficult to guarantee consistency in preparation. Additionally, technical variability may lead to batch effects, impacting data comparisons among different studies. Therefore, standardizing cross-laboratory protocols and processes is particularly necessary to ensure the comparability and reliability of data across different laboratories.

Tuberculosis has traditionally been classified into two major categories: LTBI and ATB. Clinical observations and research have revealed that hosts exhibit unique, complex, and individualized characteristics. Different hosts may share some common clinical manifestations of pulmonary tuberculosis, such as chronic cough, recurrent low -grade fever, night sweats, and fatigue;However the severity of these symptoms can vary significantly. It is worth noting that the risk of patients with latent tuberculosis infection progressing to active tuberculosis also shows significant individual differences. These observations highlight the heterogeneous manifestations and progression of the same disease in different hosts. Granuloma, as a key pathological diagnostic marker of tuberculosis, is characterized by a classic morphological structure: a caseous necrotic center surrounded by immune cells and fibroblasts. Research has found that granulomas in patients with active TB and LTBI exhibit morphological heterogeneity (O’garra et al., 2013). The formation of granulomas is a complex immune process that may either control and contain the infection (Russell et al., 2009) or promote the proliferation and spread of Mtb (Flynn, 2004). Differences in the composition of immune cells within granulomas may lead to completely different effects on Mtb—either promoting or inhibiting its growth. Variations in interaction patterns between Mtb and host immune cells may also result in different outcomes for granulomas. In addition, differences among Mtb strains may alter the progression trajectory of host granulomas, and the various states of granulomas may affect the activity and concentration of antibiotics. Therefore, the multi-level heterogeneity from the host to tissues and then to cells has a significant and unique impact on the occurrence and development of tuberculosis. With advancements in technology, fine analysis at the single-cell level has become possible. Single-cell sequencing technology has significantly enhanced the capacity for single-cell analysis (Wen and Tang, 2018). Through this technology, researchers can explore the heterogeneity, function, and interrelationships of tuberculosis immune cells, identify known, newly discovered, and rare tuberculosis pathogen targets, discover new therapeutic targets and disease biomarkers, characterize specific immune cell subsets in immunotherapy, and evaluate the impact of immunotherapy on cell subsets and cytokines in tuberculosis patients. This approach provides an in-depth understanding of drug sensitivity, drug resistance, and treatment modalities. Single-cell sequencing offers a more intuitive and comprehensive perspective, helping us understand tuberculosis immunotherapy and its latest developments.

Given that granulomas formed in zebrafish are extremely similar to those in humans, the zebrafish M. marinum model has become an effective system for studying the host and pathogen genetics of mycobacterial infection and granuloma formation (Davis et al., 2002). Cronan et al. (2021) found that atypical type 2 immune responses coordinated the formation and epithelialization processes of tuberculosis granulomas through scRNA-seq analysis of zebrafish granulomas and research on macaques infected with Mtb. High-throughput single-cell genomic analysis methods offer significant opportunities for defining the cell types, phenotypic states, and intercellular interactions that constitute granulomas, as well as for a deeper understanding of their dynamic changes (Prakadan et al., 2017). The Berit Carow team used ISS transcriptomics analysis and multiple immune markers on tissue sections to compare the immune microenvironment of granulomas in patients with tuberculosis and sarcoidosis. The study found that the formation of tertiary immune structures is a common feature of granulomas in tuberculosis patients, providing a new perspective for understanding the immune complexity of granulomas (Carow et al., 2023). Gideon et al. (2022) conducted multimodal analyses—including positron emission tomography-computed tomography (PET-CT) imaging, scRNA-seq, immunohistochemistry, and flow cytometry—on macaque lung granulomas, revealing cellular correlations related to tuberculosis control. Their research indicated that the cell ecosystem is enriched with type 1–17 and cytotoxic T cells, which are involved in pro-inflammatory signaling networks with multiple cell populations. In this cellular ecosystem, Mtb may be controlled or continue to survive and reproduce, providing new insights for developing host immune targets for tuberculosis vaccines and treatment strategies. The Winchell CG team determined through flow cytometry and single-cell RNA sequencing ligand-receptor analysis that interleukin-15 (IL-15) signaling is a key driver of cytotoxic T cells. Their research supports the idea that CD8(+) lymphocytes are crucial for the early control of Mtb infection, indicating that multifunctional cytotoxic responses may serve as potential vaccine targets (Winchell et al., 2023). This series of studies not only reveals the complex immune mechanisms underlying granuloma formation but also provides an important theoretical basis and potential strategies for precise immune interventions in tuberculosis.

With the further development of single-cell technology, these studies have provided key insights into single-cell immune responses. Single-cell immune profiling analysis enhances our understanding of responses at the single-cell level and facilitates the discovery of new diagnostic biomarkers and therapeutic targets. Single-cell RNA has shown that type I interferon (IFN-I)-responsive cells exhibit defects in their response to IFN-γ, which is crucial for controlling Mtb infection. Mrs. Koto’s research found that interstitial macrophages (IM) and plasmacytoid dendritic cells (pDC) are the primary producers of type I interferons, with pDC located near human Mtb granulomas (Kotov et al., 2023). Feng Y analyzed the characteristics of γδ T cell subsets in peripheral blood mononuclear cells (PBMCs), tuberculous pleural exudate (TPE), and pleural effusion from tuberculosis patients using single-cell sequencing. This study identified a subset of Vδ2 T cells with strong effector functions and high expression of FCGR3A, highlighting the functional diversity of γδ T cells in tuberculosis infection (Feng et al., 2025). The research by Villani et al. defined the heterogeneity of intermediate monocyte subpopulations through single-cell RNA sequencing indicating that these early identified subsets are highly heterogeneouss (Villani et al., 2017). Mtb can regulate the polarization of macrophage (Mily et al., 2020). By identifying the responses of different macrophage population to infection, we can better understand how immune cell response to Mtb infection. CHEN Q and colleagues performed single-cell sequencing on the alveolar lavage fluid of three patients with active pulmonary tuberculosis and found that alveolar macrophage subsets with increased proportions in these patients may initially polarize to M1 type and subsequently transform to M2 type. Additionally, there was a significant increase in the number and intensity of cell communications related to alveolar macrophages in pulmonary tuberculosis patients (Chen et al., 2022). SHARP J D and colleagues (Sharp et al., 2016) used chromatin immunoprecipitation sequencing (CHIP-Seq) to study the SigH binding sites that directly regulate the genes and operons of Mycobacterium tuberculosis, demonstrating the expansion of the SigH regulator and its regulatory response capabilities to various emergency situations. This work also underscores the significant value of a whole-genome approach in understanding the regulation of bacterial genes. It is noteworthy that immunomodulation-related genes are differentially expressed in tuberculosis patients. SONG J and colleagues (Song et al., 2023) established a competitive endogenous RNA(ceRNA) regulatory network using whole transcriptome sequencing and analyzed the regulatory non-coding RNAs involved in the pathological processes of tuberculosis. They identified has-miR-106a-5p, has-miR-17-5p, and has-miR-2355-5p as potential diagnostic biomarkers.Together, these studies provide new therapeutic targets for tuberculosis treatment and may drive innovations in treatment and prevention strategies.

Through single-cell sequencing technology, we can reveal the composition of T cell populations and immune cell subpopulations, thereby enhancing our understanding of their responses to Mtb infection. Wang Y and colleagues (Wang et al., 2023) utilized single-cell RNA transcriptome and T-cell/B-cell receptor (TCR/BCR) sequencing to unveil systemic immune dysregulation phenomena in patients with severe tuberculosis. CAI Y and colleagues (Cai et al., 2020) compared RNA-seq datasets from healthy controls (HC), individuals with LTBI, and TB patients, observing that natural killer (NK) cell subsets (CD3-CD7+GZMB+) were gradually depleted from HC to LTBI and TB stages. Their findings confirmed that changes in the frequency of this NK cell subgroup could effectively distinguish between TB patients, LTBI individuals, and HC. Additionally, Cai Y and colleagues (Cai et al., 2022) revealed the diversity of human T cell functions in tissues infected with tuberculosis and the specific amplification of T cell receptor (TCR) sequences within these tissues through single-cell sequencing technology. This knowledge about the composition and function of T cells at the infection site may assist in vaccine development. MUSVOSVI M and colleagues (Musvosvi et al., 2023) analyzed specific sequences of Mycobacterium tuberculosis in two longitudinal cohorts using single-cell and block TCR sequencing, alongside the GLIPH2 algorithm. This study provided a preliminary list of TCR specificities and established a comprehensive TCR sequence database, serving as a valuable tool for identifying candidate tuberculosis vaccine antigens. Finally, XU Y and colleagues (Xu et al., 2022) conducted a comprehensive analysis using both array and single-cell RNA-seq methodologies, finding that the expression of the immune-related hub gene ADM in peripheral blood could serve as a novel biomarker for differentiating tuberculosis from LTBI and HC. These findings provide promising ideas and insights for developing new treatment strategies for tuberculosis.

Through single-cell sequencing technology, we can gain a deeper understanding of the differences in gene expression among cells, thereby inferring the interactions within gene regulatory networks. This is particularly significant for understanding the potential responses of tuberculosis patients to immunotherapy. Kong et al. (2021) investigated the effects of secondary immune stimulation using bacterial lipopolysaccharide (LPS) and conducted single-cell transcriptomic sequencing before and after BCG vaccination. They found that BCG vaccination could reduce systemic inflammation and identified 75 genes that were affected by the LPS response, including inflammatory mediators with enhanced expressions, such as CCL3 and CCL4.This study elucidated the molecular mechanisms underlying BCG’s beneficial immune effects, highlighting the combination of a lower inflammatory state in circulation with an enhanced response of innate immune cells to reinfection. The changes associated with the training effects of BCG-induced circulating monocytes can offer valuable insights for the future enhancement of vaccine development.

Not all patients can benefit from current immunotherapy, and those who do respond to the treatment may develop drug resistance over time. Therefore, analyzing the mechanisms of drug resistance in immunotherapy is crucial for the effective application of these therapies and the formulation of individualized treatment strategies. Single-cell analysis holds promise for revealing the underlying causes of drug resistance and facilitating the development of tailored treatment plans. LEUNG K and colleagues (Leung et al., 2017) demonstrated that SMRT sequencing exhibited high stability in identifying clinical isolates of multidrug-resistant tuberculosis (MDR-TB). Through SMRT sequencing and comparative genomic analysis, researchers were able to identify mutations associated with the gradual development of drug resistance and the growth adaptability of MDR-TB in response to anti-tuberculosis drugs. The comparative genomic analysis also revealed new mutations at the rv0888, lpdA, and cobM gene loci, which may explain the differences in antibiotic resistance and growth patterns between the two MDR-TB resistant strains. Furthermore, Sehgal et al. (2021) utilized dynamic high-resolution scRNA-seq to investigate the fundamental mechanisms of immune escape in persistent anti-PD-1 responses within a functional model of immunotherapy. This approach provides an important reference for analyzing drug resistance in other treatment modalities or in conjunction with combination immunotherapy.

Immunotherapy offers hope to tuberculosis patients. The discovery of new targets and biomarkers is expected to shorten the treatment cycle for tuberculosis, enhance therapeutic efficacy, and reduce drug resistance. However, research on evaluating the effectiveness of tuberculosis immunotherapy through single-cell sequencing remains relatively scarce. As our understanding of the characteristics of immune cells, intercellular signal transduction, and changes in cytokines deepens, it will become increasingly possible to clarify the responses and prognoses of tuberculosis patients to immunotherapy in a more intuitive and effective manner.

The heterogeneity of tuberculosis and the complexity of the host microenvironment lead to varying treatment responses and clinical outcomes. Advances in single-cell sequencing technology have provided an important tool for analyzing the heterogeneity of immune cells and the drug resistance associated with tuberculous meningitis (TBM). The application of this technology, coupled with the continuous development of spatial multi-omics platforms and comprehensive analysis of large-scale datasets, has significantly enhanced our understanding of tuberculosis heterogeneity and awareness of drug resistance. Furthermore, these advancements have contributed to innovations in diagnostic technologies for tuberculosis patients and the development of new drugs.

However, the application of single-cell technology faces several challenges and limitations. For instance, solid tissues must be processed into single-cell suspensions prior to sequencing, a complex and challenging procedure. Additionally, the binding activity of transcription factors cannot yet be measured at the single-cell level, and many fundamental statistical analysis issues, such as handling batch effects and technical variability, remain inadequately addressed (Trapnell, 2015). Although external RNA spike-in is the best method for quantifying technical noise, it has some limitations, such as the need to accurately calibrate the amount of external RNA spike-in added based on the total amount of endogenous RNA, and the inability to estimate the random loss of RNA molecules due to the effects of cell lysis efficiency (Kolodziejczyk Aleksandra et al., 2015). It is also noteworthy that some single-cell subtypes may not express major histocompatibility complex (MHC) Class II molecules in their resting state and can only be defined by non-RNA molecules or low-abundance transcripts, which may lead to these cells being overlooked in analyses (Villani et al., 2017). Moreover, technical issues related to RNA sequencing, such as library preparation, sequencing depth, and the influence of biological states, may result in data loss, as well as increased costs and the risk of numerous sequencing artifacts (Navin, 2014; Fittall and Van Loo, 2019). Standardizing protocols between different laboratories is key to addressing the current methodological consistency issues. Standardization can eliminate result variability caused by differences in experimental conditions, operational procedures, and data analysis methods. By establishing uniform experimental processes and standards, researchers can ensure the reproducibility and reliability of their results. Furthermore, this consistency can facilitate comparisons across studies, thereby promoting progress and collaboration in scientific research.

Effective drug targets identified in vitro need to be translated into effective therapeutic targets in vivo (Bellerose et al., 2020). Therefore, conducting in vivo analyses of drug targets is particularly important. By combining existing technologies, we should maximize the advantages of single-cell sequencing and integrate single-cell and spatial transcriptomics to gain a deeper understanding of the three-dimensional structure of granulomas. It is important to track immune responses during treatment to identify markers associated with drug resistance and develop targeted treatment strategies based on individual patient immune characteristics. Furthermore, optimizing the delivery mechanisms of nanomedicines through single-cell analysis is also an important objective. Overcoming these limitations and implementing precision medicine will provide new hope for tuberculosis patients.

In conclusion, single-cell sequencing technology has demonstrated significant potential in tuberculosis research, enabling us to gain a deeper understanding of the heterogeneity and drug resistance of immune cells involved in tuberculosis. With continuous advancements in this technology and the integration of spatial multi-omics platforms, we will be better equipped to analyze the complexity of tuberculosis and lay the groundwork for developing new diagnostic tools and therapeutic drugs. Despite this potential, challenges remain in the form of sample processing complexity, technical limitations, and high costs. Future research should focus on addressing these issues to achieve the goals of precision medicine and personalized treatment through technological innovation and interdisciplinary collaboration, ultimately providing more effective treatment options for tuberculosis patients and strengthening our overall response to this global health threat.