Genetic insights into the relationship between immune cell characteristics and ischemic stroke: A bidirectional Mendelian randomization study

Ischemic stroke, a major contributor to global disability and mortality, is underpinned by intricate pathophysiological mechanisms, notably neuroinflammation and immune cell dynamics. Prior research has identified a nuanced and often paradoxical link between immune cell phenotypes and ischemic stroke susceptibility. The aim of this study was to elucidate the potential causal links between the median fluorescence intensity (MFI) and morphological parameters (MP) of 731 immune cell types and ischemic stroke risk.


INTRODUC TI ON
Ischemic stroke, marked by cerebral blood flow obstruction, is a major global health concern, leading to significant mortality and morbidity [1].This condition results from a confluence of genetic, environmental, and physiological influences [2].The immune system's involvement in ischemic stroke pathogenesis has recently gained considerable attention.Immune cells, known to drive inflammatory processes around ischemic events, can critically influence stroke outcomes [3].
Neuroinflammation plays a crucial role in the pathophysiology of ischemic stroke, characterized by a complex inflammatory cascade within the brain [4].This process is initiated by resident immune cells, augmented by peripheral immune components such as neutrophils, macrophages, and T lymphocytes [4,5].A pivotal aspect of this cascade is the disruption of the blood-brain barrier, neuronal damage and cerebral edema, which significantly intensifies inflammatory infiltration following ischemic stroke [6,7].A study underscored the association between elevated levels of specific inflammatory mediators, such as CD40 ligand (CD40L) and monocyte chemoattractant protein-1 (MCP-1), and the risk of long-term nonfatal vascular events in ischemic stroke patients [8].
These inflammatory responses are integral to the pathogenesis of ischemic stroke, contributing to the development of thrombosis and atherosclerosis.Furthermore, a direct correlation has been established between various inflammatory markers, including white blood cell count, and prognostic factors such as mortality, cognitive impairment, and functional status [9,10].These insights enhance our understanding of the pathophysiological mechanisms underlying ischemic stroke and aid in developing predictive and interventional strategies for its prognosis.A study by Tuttolomondo et al. highlights the significance of inflammation in ischemic stroke pathogenesis [11].This research suggests that early intervention with high-dose atorvastatin may improve inflammatory markers and enhance functional recovery in patients, indicating a potential therapeutic pathway targeting early and aggressive inflammation management in stroke cases.
Advancements in genetic research have enabled deeper exploration of the intricate relationship between immune cell diversity and ischemic stroke.Utilizing Mendelian randomization (MR), which employs genetic variants as instrumental variables [12], this study investigates the causal impact of immune cells on stroke risk.
MR provides a novel perspective in understanding genetic correlations between an extensive range of immune cells and ischemic stroke risk.
Previous research has underscored complex interactions between various immune cells and the central nervous system in the context of ischemic stroke.Ischemic strokes, the predominant stroke type, initiate a series of biological reactions involving both neuroglial and peripheral immune cells.These interactions significantly affect the progression of brain damage and stroke outcomes.Although neuroglial cell research has been extensive, studies on peripheral immune cells in ischemic stroke have been more limited and narrow in scope, resulting in incomplete insights into the peripheral immune system's role in stroke [13].The immune response post-stroke is multifaceted and contradictory.
Peripheral immune cells, such as neutrophils, T cells, B cells, dendritic cells, and macrophages, infiltrate the ischemic brain tissue and influence the progression of injury.Their dual roles involve both exacerbating and mitigating ischemic damage, including neuronal apoptosis, repair, differentiation, and neuroregeneration [14,15].Moreover, the complexity of immune responses and individual genetic differences pose challenges in understanding these associations.
Originally designed as an alternative to randomized controlled trials [16], MR provides robust causal evidence between exposures and outcomes via genetic variations [17].The strength of MR lies in its ability to identify causal relationships independent of confounding factors and reverse causality, due to the random allocation of genetic variations prior to disease onset [18].Through MR, this study aims to conduct an exhaustive analysis of various peripheral immune cell types and characteristics, seeking a more comprehensive understanding of the immunogenetic aspects of ischemic stroke.

Research design
Our study used a two-sample MR approach to examine the bidirectional causal link between 421 immunological cell traits and ischemic stroke.Using this method, we sought to determine the causal influence of immune traits on ischemic stroke.MR analysis adheres to three critical assumptions: (i) the genetic variants must be strongly associated with the exposure; (ii) the variants should be independent of confounding factors; and (iii) the variants must affect the outcome solely through the exposure (Figure 1).

Data sources
Data were sourced from a comprehensive genome-wide association study (GWAS), offering a detailed statistical summary of immunological traits (accession numbers GCST0001391 to GCST0002121) [19].This dataset includes 731 distinct immune

Data processing
In accordance with the foundational assumptions of MR, and drawing on recent literature, our study set the significance threshold for each immune trait's instrumental variables at 5 × 10-6 [20,21].The linkage disequilibrium (LD) among SNPs was calculated using the genomes of 1000 individuals from a European population as a reference.We selected SNPs with an LD coefficient (r 2 ) below 0.001 and independent SNPs located over 10,000 kilobases apart, aiming to identify SNPs in the European population with independent genetic effects.In our analysis, phenotypic F-statistics were computed for each instrumental variable to ascertain their strength and reduce potential biases arising from weak instruments.Instrumental variables demonstrating F-statistics below the threshold of 10 were systematically excluded from further consideration.

Statistical processing
All analyses were performed using R software, version 4.3.1.

Two-sample Mendelian randomization study and sensitivity analysis
Our study conducted two-sample MR analyses on various immune cell-related phenotypes including B cells, T cells, Tregs, and other immune cells.We employed random-effects inverse variance weighting (IVW), MR-Egger, and weighted median (WM) methods for these analyses.The primary outcomes were derived using the IVW method, with statistical significance set at p < 0.05.IVW gives more weight to variants with smaller variances, hence more precise estimates, by weighting a variant's effect on the outcome inversely to its variance.
This method prioritizes more reliable, lower variance estimates and uses the reciprocal of the outcome variance for weighting, excluding the intercept in regression fitting [22].For pleiotropy assessment, where a single gene influences multiple phenotypes, we utilized MR-Egger regression.This method includes an intercept evaluation to detect pleiotropy, with a significant intercept (p < 0.05) indicating notable horizontal pleiotropy [23].We also employed Cochran's Q statistic to examine heterogeneity, which measures the consistency of genetic variants' effects on phenotypes.A significant Cochran's Q (p < 0.05) suggests considerable heterogeneity in the effects of different genetic variants [24].For heterogeneous data without pleiotropic effects (p > 0.05), the WM approach was used [23].

Reverse Mendelian randomization analysis
Using the same MR methods, we investigated whether ischemic stroke causally affects immune cell traits and explored the potential for reverse causation.In this reverse MR analysis, ischemic stroke was considered the exposure factor, and various immune cell traits were treated as outcomes.

RE SULTS Forward Mendelian randomization analysis results
Table S1 provides SNP-related information.As shown in Figure 2 and Table S2, 19 immune phenotypes suggestive of a link with ischemic stroke were identified, all within MFI traits.No significant associations were found with MP.These phenotypes comprised five T-cellrelated, six Treg-related, six B-cell-related, one myeloid cell-related, and one granulocyte-related trait (Table S3).S4).

DISCUSS ION
This study represents a pioneering effort to systematically delineate the causal link between immune phenotypes and ischemic stroke.
Leveraging data from GWASs, we evaluated 421 immune cell phenotypes for their association with ischemic stroke.Our MR analysis pinpointed 23 immune cell phenotypes that potentially affect the risk of ischemic stroke.Out of these, eight phenotypes, excluding those exhibiting reverse causation, are believed to contribute to the onset of ischemic stroke, whereas 15 exhibit neuroprotective properties.B cells play a multifaceted role in the immune response to stroke, displaying both detrimental and restorative effects.After ischemic stroke, the compromised blood-brain barrier may permit B cells and other peripheral immune cells to enter ischemic brain tissue [25,26].
These cells are crucial in modulating the progression of ischemic brain injury and influencing the secondary inflammatory response post-stroke [15,27].However, studies by Schuhmann et al. indicate that targeting B cells does not significantly alter lesion volume or functional outcomes during the acute phase of experimental stroke [28].Our findings mirror this complexity, revealing that while some B-cell phenotypes, such as CD19 on CD24+ CD27+ B cells, CD19 on memory B cells, and CD26 on immature-mature B cells, offer neuroprotection, others such as absolute B-cell counts, CD27 on CD20-CD38-B cells, and CD38 on IgD+ CD24-B cells, are associated with increased stroke risk.This underscores the need for further research into the nuanced role of B-cell-related phenotypes in stroke.
their regulatory function in inflammation, a critical factor in stroke.
The inflammatory response following ischemic stroke is complex, contributing to both brain injury and repair, with T cells playing a pivotal role in balancing these outcomes.
Our findings robustly demonstrate that numerous immune characteristics instrumental in improving ischemic stroke are present in Tregs.Earlier studies indicate that Tregs initially mitigate acute brain injury by moderating the immune response, largely due to their immunosuppressive capabilities and interactions with other immune cells, such as neutrophils [39,40].On a molecular level, Tregs secrete cytokines including IL-10, TGFβ, and IL-35.These cytokines suppress the activity of other immune cells, thereby fostering an antiinflammatory environment that safeguards neural tissue [41][42][43].
Tregs employ surface molecules such as PD-L1, CTLA-4, and galectin-1 for essential direct cell-to-cell interactions, which are pivotal for their immunosuppressive role and in maintaining the integrity of the blood-brain barrier [44][45][46].Additionally, Tregs are capable of modulating astrocyte responses.This modulation involves diminishing the formation of reactive astrocytes and reducing inflammatory glial scarring, partly through bimodal protein activity and potential IL-10 factors, thus promoting a milieu conducive to neuronal survival [47][48][49].Moreover, Tregs secrete neurotrophic factors such as brainderived neurotrophic factor, which plays a crucial role in neural repair and regeneration following a stroke [50].
Our study distinctly identified that CD39 phenotypes across various immune cells exert a neuroprotective impact on ischemic stroke.Research demonstrates that CD39 catalyzes the conversion of extracellular adenosine triphosphate (ATP) to adenosine monophosphate (AMP), followed by CD73 transforming AMP into the immunosuppressive nucleoside adenosine (ADO) [51].This enzymatic process, with CD39 as a key regulator, serves as an immune modulator, transitioning the cellular milieu from an ATP-driven pro-inflammatory state to an ADO-predominant anti-inflammatory state, thereby playing a vital role in immune regulation [51].The absence of CD39 correlates with reduced liver insulin sensitivity and increased levels of pro-inflammatory cytokines, including IL-6, IL-1β, TNFα, and IFNγ, intensifying the inflammatory response [52,53].Our study leverages the MR approach to discern a potential causal relationship between immune phenotypes and ischemic stroke risk, moving beyond simple associative studies.MR significantly mitigates confounding factors by using genetic variants as instrumental variables, a crucial technique in overcoming common obstacles in observational studies [54].This method offers a comprehensive understanding of immunological mechanisms, specifically examining the impact of immune cells such as B cells, T cells, and Tregs on stroke risk.Importantly, it aids in identifying immune phenotypes that may increase or decrease stroke risk, providing vital insights to aid the development of targeted therapies.This approach is essential for enabling more precise and personalized strategies in stroke prevention and treatment.
This study has some limitations.As a pioneering study in mapping the entire causal relationship between the immune landscape and ischemic stroke, it is inherently exploratory.Our goal was to examine as many immune phenotypes as possible that could affect stroke onset, thereby setting a course for future research.

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Our findings further substantiate that CD39 presence in CD39+ CD8+ T cells and CD39+ activated CD4 Tregs is inversely associated with stroke occurrence.Notably, CD39 on CD39+ activated CD4 Tregs emerges as a pivotal immune element for neuroprotection in ischemic stroke.This research delineates specific immune characteristics, underscoring the potential for targeted therapy involving Tregs in future ischemic stroke interventions.
Consequently, we did not apply multiple testing corrections.The study is underpinned by an MR analysis, using SNP data sourced from the OPEN GWAS database.The aggregate nature of these data precludes access to individual-level details, which consequently limits our capacity to categorize ischemic stroke into more specific subtypes.Nevertheless, the extensive genetic variant coverage in our dataset establishes a robust basis for exploring the general association between immune cell characteristics and ischemic stroke risk.It is projected that future research will increasingly focus on the classification of ischemic stroke.Moreover, the issue of pleiotropy is addressed based on the statistical significance of pleiotropy tests alone, without excluding pleiotropic instrumental variables.While these limitations increase the risk of type I errors, they decrease the chance of type II errors.This approach allows for a more expansive and systematic exploration of the immune landscape in ischemic stroke, creating a pathway for future studies on precise targets.Ultimately, our study offers a detailed and systematic insight into the immune dynamics in ischemic stroke.It establishes a foundation for future precision medicine endeavors, focusing on specific immune phenotypes for stroke prevention and treatment.Our research marks a significant step in elucidating the intricate relationship between the immune system and ischemic stroke, providing crucial directions for subsequent investigations.AUTH O R CO NTR I B UTI O N S Shuai Hou: Writing -original draft; writing -review and editing; conceptualization; methodology; software; data curation; validation.Xia Deng: Writing -original draft; writing -review and editing; view and editing; visualization; formal analysis; project administration; supervision.Haiyan Yang: Writing -review and editing; supervision; formal analysis; project administration; funding acquisition.Chunping Wang: Writing -review and editing; supervision; formal analysis; project administration; funding acquisition.FU N D I N G I N FO R M ATI O N This research was supported by the funding of the Key Research and Development Program of Shandong Province (grant no.2023RKY07003), Yuan Du Scholars and Weifang Key Laboratory.