An immune-associated mitochondrial DNA variant with sex differences reveals a putative novel microprotein called MASL Authors: Ana R. Silverstein, Thalida E. Arpawong, Eileen M. Crimmins, Tae Jung Oh, Rong Lu, Melanie Flores, Hiroshi Kumagai, Junxiang Wan, Hemal H. Mehta, Kelvin Yen, Brendan Miller, Yeachan Lee, Anna Nogalska, Pinchas Cohen Published: 2026 ABSTRACT The use of mitochondrial wide association studies (MiWAS) to link mitochondrial DNA variants (mtSNPs) to phenotypes of interest has uncovered important connections between mitochondrial genes and human health. The recent introduction of a re-annotated mitochondrial genome that accounts for small open reading frames (sORFs) with protein coding potential suggests the existence of mitochondrial-derived microproteins, many of which remain uncharacterized. Thus, considering the re-annotated mitochondrial genome when conducting genomic analyses such as MiWAS facilitates the mapping of mtSNPs back to microprotein-encoding sORFs and uncovers interactions between mitochondrial microproteins and biological systems. Here, we employ MiWAS of venous blood samples from the Health and Retirement Study (HRS) and identify a mtSNP associated with sex-specific changes to immune composition. After accounting for re-annotation, we map the identified mtSNP back to a sORF that encodes a novel microprotein, termed MASL (Mitochondrial Associated Small d-Loop peptide). Complementary phenome-wide association studies (PheWAS) in HRS and and UK Biobank confirm interactions between this mtSNP and immune phenotypes of interest, and our targeted RNA-Seq method (mitoSNP-seq) elucidates sex-differences in gene expression and functional pathways potentially altered by this mtSNP that may be relevant to the associated microprotein. Early characterization of the MASL microprotein shows sex-differences in circulating MASL levels in human plasma, and sex-specific interactions when comparing male and female mice treated with synthesized MASL. Together, the results of this study not only contribute to our understanding of mitochondrial dynamics in immunity, but also provide early characterization of a novel mitochondrial-derived microprotein with sex-specific modulatory effects.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12979-026-00576-6. Main text Background Understanding the role of mitochondria in regulating immune function and activity is vital to improving human health and longevity. Mitochondria are not only the metabolic hub of eukaryotic cells, but also fundamental to the highly dynamic communication and response network that maintains the human immune system [1, 2]. Mitochondrial metabolism and changes in mitochondrial activity can play a role in immune cell maturity and function, and mitochondrial signaling molecules such as ROS and mtDNA can act as particularly active regulators of immunity. Mitochondrial complex III, a crucial component of mitochondrial metabolism, supports proper adaptive immune function [3]. The release of mtDNA into the cytosol, often triggered by ROS, initiates several signaling cascades involved in inflammatory regulation and immunity [4]. However, while a pattern of mitochondrial-immune crosstalk has been well established, much remains unknown. One relatively novel facet of mitochondrial biology that may be a key element to understanding mitochondrial-immune communication is microproteins. Advancements in whole-omics approaches reveal an abundance of protein-coding small open reading frames (sORFs) within larger genes [5]. Adapted approaches that target the unique differences in mitochondrial transcription machinery [6] introduce a re-annotated mitochondrial genome that contains hundreds of putative sORFs and a novel subclass of mitochondrial-encoded microproteins [7]. While several mitochondrial-encoded microproteins have been characterized as potent and diverse regulators of biological systems involved in health and disease [8], hundreds remain unexplored. Recent findings suggest that the mitochondrial-encoded microprotein MOTS-c functions as a novel regulator of host immune function, and serves as an adaptive mitochondrial-derived host defense peptide in response to immune signaling cues [9]. These findings point to the likely existence of a subset of mitochondrial-encoded microproteins with immunomodulatory effects that await identification. Here, we outline a framework for the discovery and early characterization of a novel immunomodulatory mitochondrial-derived microprotein, beginning with a targeted mitochondrial-genomic analysis technique termed MiWAS. The strategic use of the mitochondrial-wide association study (MiWAS), a tool used to assess associations between single-nucleotide polymorphisms in mtDNA (mtSNPs) and selected phenotypes of interest, has been instrumental in revealing interactions between mitochondrial genetics and human biology [10–13]. Several MiWAS analyses map mtSNPs back to microprotein-encoding sORFs [14–16], identifying novel microproteins that are prime candidates for detailed characterization. These findings also imply that specific mtDNA variants can induce functional alterations in microproteins encoded by corresponding sORFs, providing one possible mechanism of action for associated phenotypes observed in mtSNP carriers. Here, we leverage MiWAS analyses of venous blood samples from the Health and Retirement Study (HRS) to screen for mtSNPs associated with changes in immune cell composition. This screening identified a mtSNP linked to sex-specific immune modulation, which we mapped back to a sORF in the mitochondrial D-Loop. This sORF is predicted to encode a novel microprotein termed Mitochondrial-Associated- Small -d-Loop peptide, or MASL. We find that the associated mtSNP alters the amino acid sequence of MASL. Consequently, we hypothesize that mtSNP carriers produce a modified version of MASL which may have altered function and thus contribute to observed phenotypic differences. Acknowledging the inherent limitations of MiWAS, such as cohort-to-cohort genotype variation and the complexities of haplogroup architecture, we have complemented our study with phenome-wide association studies (PheWAS) in both HRS and UK Biobank to corroborate associations across different populations and analytical frameworks. To investigate functional pathways of the mtSNP and associated microprotein, we developed a targeted RNA-Seq technique in HRS subsamples, which we have termed mitoSNP-seq. In employing this method, we identify important sex-driven differences in transcript expression and molecular signaling, suggesting that MASL may regulate cell health in a sex-specific manner. Given these results, we treat MASL as a sex-dimorphic modulator of cell health during its preliminary characterization. Supporting this prediction, we observe sex- and -age-dependent variations in circulating human plasma levels of MASL, as well as unique immunomodulatory and sex-specific interactions in vivo. Collectively, our findings link a promising mtSNP and its corresponding microprotein to sex-specific immunomodulatory effects. Further characterization of this microprotein not only suggests an immunomodulatory role, but also introduces a sex-dimorphic regulatory explanation for its related function. Together, these results add to a growing body of work elucidating the role of mitochondrial microproteins as regulators of cell health, and provide novel insight into their significance in immune system modulation. Materials and methods Mitochondrial-wide association study in health and retirement study Our Mitochondrial-Wide Association Study (MiWAS) leveraged samples from the HRS 2016 Venous Blood Study (VBS). Specifically, we extracted measured percentages of neutrophils and lymphocytes from venous blood samples, extrapolated the neutrophil–lymphocyte ratio (NLR), and compared these parameters between groups. The HRS is a longitudinal, population-based survey containing results from over 37,000 participants over the age of 50, following health changes from 1992 onwards. In 2016, venous blood from a representative sample was collected. Venous blood sample data underwent pre-processing and quality control by HRS [17]; post-processed variables were cleaned using R (v4.3.1), by assessing normality of continuous variables, instituting variable transformation when necessary, and filtering outliers using a quantile cutoff. The self-identified ethnic-specific stratified cohort sample sizes were as follows: the non-Hispanic white (NHW) cohort included a total sample size of approximately 4,368, while the Hispanic cohort consisted of 816 individuals, and the non-Hispanic Black (NHB) cohort consisted of 981 individuals. Considering the smaller sample sizes of the Hispanic and NHB cohorts, we determined that we would not have strong enough power to extrapolate interpretable findings in MiWAS, and thus chose to leverage Hispanic and NHB samples in downstream PheWAS analyses once initial mtSNPs of interest were identified. HRS genotyping using the HumanOmni2.5 array included 256 mitochondrial SNPs (mtSNPs). Mitochondrial genotyping data were analyzed using PLINK (v2.0) [18] in bed/bim/fam format. Ancestry principal components were generated and included as covariates in statistical models to reduce issues of population stratification in association testing. Analyses were corrected for age, sex, ethnicity, BMI, and the first five nuclear principal components, while sex- and ethnicity-stratified analyses were corrected for all components minus sex and ethnicity, respectively. Phenotypes of interest consisted of continuous variables and thus linear regression models were used. A population level minor allele frequency (MAF) threshold of 1% was set, leaving a total of 85 mtSNPs qualified for analysis. Because we use SNP array technology to measure variation in mitochondrial SNPs, heteroplasmy could not be reliably evaluated in these analyses. However, considering heteroplasmy in WBC-derived mtDNA is typically below the limit of detection for GWAS assays, it is unlikely that undetected heteroplasmy significantly influenced our results. Given cohort characteristics, we applied a previously reported adaptive permutation approach to adjust for multiple testing [19, 20]. This approach was employed to reduce the likelihood of type I error associated with multiple testing, while also accommodating for the inherent features of mitochondrial genomic data. Permutation testing was performed with a maximum number of 47,373 permutations for the most significant mtSNP, and a family-wise error rate (FWER) cutoff of 0.05. All mtSNPs surpassing this empirical p-value threshold (p < 0.05) were considered statistically significant. STREGA reporting guidelines were considered during the design of this study [21]. Phenotype-wide association study in immune phenotypes To validate MiWAS results, PheWAS analyses were employed in both HRS and UK Biobank cohorts. Briefly, associations between the mtSNP G16393A variant and variables related to immune cell composition were tested in Hispanic, NHB, and NHW cohorts in HRS, and NHW cohorts in UK Biobank (this research has been conducted using the UK Biobank Resource under Application Number 66866). To derive NHW participants from UK Biobank, genotype and phenotype files were extracted and samples were subset based on self-identification as “Caucasian”, and participants who have opted out of data inclusion were removed, leaving a final sample set of 409,152participants. Phenotypic associations were analyzed using a general linearized model, and adjusted for age, sex, and the first five principal components to minimize confounding effects. Statistical significance was assessed using an FDR-corrected cutoff of p < 0.05, and trending results were determined using an unadjusted p < 0.05. Mito-SNP differential expression analysis (mitoSNP-Seq) RNA sequencing data from HRS participants were processed using a custom pipeline developed in R (v4.3.1). This pipeline utilized a differential-gene expression (DGE) design file to filter RNA-seq results based on defined selection criteria. The design file integrated phenotype data – including MASL mtSNP status and various covariates – with the most current version of the HRS RNAseq dataset (n = 3,738). The resulting DGE design file yielded a final analytic sample of 1,926 NHWs (1,088 female and 838 male; n = 38 and n = 21of mtSNP carriers, respectively) for downstream RNA-Seq analysis. A standard linear regression was performed for each gene assessing differential expression based on MASL mtSNP carrier status, and adjusted for batch effects, age, sex, and the first five nuclear principal components. P-values were adjusted for multiple testing using the Benjamini–Hochberg correction method to control for false discovery rate (FDR), implemented using the fdrtool package. Differentially expressed genes were plotted using EnhancedVolcano with cutoffs at an unadjusted p-value < 0.01 and an adjusted p-value < 0.05. Downstream GO enrichment analyses utilized a gene list filtered by an unadjusted p-value of 0.01, and significantly enriched terms were computed using a BH-adjusted threshold of p < 0.05. MASL protein structure prediction PEP-FOLD 4, a de novo peptide prediction server, was used to predict the structure of both MASL and the MASL V18I from respective amino acid sequences using a previously reported predictive algorithm [22]. PEP-FOLD 4 was selected due to its protein cutoff size of 40 amino acids, yielding highly specific peptide prediction models. A PDB output file was exported and visualized using the RCSB PDB Mol* 3D Viewer tool [23]. The protein domain tool IUPred3 was used to predict disordered protein and binding regions within the MASL amino acid sequence. DeepLoc 2.0 [24] was used to predict MASL protein localization. MASL antibody development Rabbit anti-MASL sera were produced by YenZym Antibodies (Brisbane, CA), followed by antigen-specific affinity purification. Samples were eluted and binding affinity to synthetic compound was tested before use. Cell culture HL-60 cells used in this study were purchased from ATCC (CCL-240). Per ATCC guidelines, cells were grown in IMDM with 20% FBS at 37 C with 5% CO2, and medium renewal every 2 to 3 days. For terminal differentiation to neutrophil-like cells (dHL-60 neutrophils), HL- 60 cells were passaged and incubated in IMDM media with 20% FBS and 1.25% DMSO for five days, upon which distinct morphological changes were observed, as previously published [25]. Immunocytochemistry Non-adherent HL-60 neutrophils at 50–70% confluence were centrifuged and resuspended in 1 mL of phosphate buffered saline (PBS). Cells were then centrifuged and resuspended in 4% paraformaldehyde to fix for 15 min at room temperature (RT). Smear was made on glass slides, air dried and then washed twice with PBS. Cells were then permeabilized using 0.1% triton X in PBS for 15 min at RT, and immediately blocked in 2% BSA PBS for one hour at RT. After incubation, MASL antibody, or a blocking mixture of MASL antibody and synthetic MASL peptide (at a ratio of 1:5) along with mitochondrial marker TOM 20 antibody (1:200), were added in a 0.1% BSA PBS mixture and incubated overnight at 4C. After overnight incubation, cells were washed with PBS and the Alexa-Fluor 488 goat (anti-rabbit) antibody in 0.1% BSA PBS was added (1:200) and incubated for 45 min at RT. Cells were washed with PBS and the nuclear-marker Hoechst (a dsDNA stain) was added (1:2000) prior to final wash and mounting. Fluorescent staining was imaged using the Leica STELLARIS confocal microscope. MASL detection with ELISA Human plasma samples were obtained from 79 subjects who underwent metabolic bariatric surgery at Seoul National University Bundang Hospital (IRB no. B1812513302). All subjects were Asian (51.5% male, 48.5% female), with a mean age of 40.1 years old and a mean BMI of 37.4 kg/m2. Blood samples were collected after overnight fasting prior to surgery, and were obtained again one year after surgery from a subpopulation (n = 67). Plasma samples were stored at −80 °C before analysis. Circulating levels of MASL were measured by in-house ELISA with a detection sensitivity of 50 pg/mL (Suppl. Figure S1). Briefly, plasma extraction occurred in 90% acetonitrile and 10% 1 N HCl. Extraction reagent was added to plasma at a 2:1 ratio in RIPA buffer, followed by gentle mixing and incubation at room temperature for 30 min. Mixtures were centrifuged and supernatants were removed and dried. Dried extracts were then reconstituted with 250μL of phosphate buffer (50 mM sodium phosphate, 150 mM sodium chloride, 0.5% Tween-20, pH 7.6) and prepped for ELISA analysis. To prepare ELISA, ninety-six-well microtiter plates were coated with rabbit anti-MASL antibody at 0.5 μg/well in 200 μl of 50 mM sodium bicarbonate buffer and incubated at room temperature for 3–4 h on a shaker followed by two washes with wash buffer and two washes with Superblock buffer (Pierce Chemicals, Rockford, IL). Next, standards, controls, or extracted samples of pre-titered anti-MASL antibody were added to the appropriate wells and incubated overnight. After incubation, plates were washed and incubated with streptavidin-HRP conjugate for 30 min at room temperature, then washed four additional times with wash buffer. After the final wash, 200 μl/well of 1-step ultra TMB was added and incubated for approximately ten minutes. The reaction was then stopped by the addition of 2 N H2SO4 and absorbance was measured on a plate spectrophotometer (Molecular Designs, Sunnyvale, CA) at 450 nm. We then compared MASL levels based on sex and age (< 50 years vs. ≥ 50 years). In addition, we used the data from baseline and one-year follow-up to divide the subjects into two groups based on the change of MASL levels: the MASL increased group and the MASL decreased group. Subcellular fractionation Cytosolic, nuclear, and mitochondrial fractions were prepared from cultured HL-60 cells. To extract nuclei, cells were washed in ice-cold DPBS and resuspended in fractionation buffer containing 10 mM HEPES pH 7.6, 3 mM MgCl2, 10 mM KCl, 5% (v/v) glycerol, 1% Triton-X100, and protease/phosphatase inhibitors for 15 min, followed by centrifugation for 5 min at 250 × g and 4 °C. The resulting supernatant was further centrifuged once more at 18,000 × g for 10 min at 4 °C to obtain a relatively pure cytoplasmic fraction. The original pellet prior to the 18,000 × g centrifugation was washed in 10 mM HEPES pH 7.6, 1.5 mM MgCl2, 10 mM KCl, and protease/phosphatase inhibitors and centrifuged at 250 × g and 4 °C. The washed pellet was then resuspended in RIPA lysis buffer and protease/phosphatase inhibitors, followed by three sonication periods of 5 s (separated by 10 s) with 30% amplitude on ice. The sonicated pellet was centrifuged at 18,000 × g for 10 min at 4 °C to obtain a relatively pure nuclear lysate. To extract mitochondria, cells were washed in ice-cold DPBS and resuspended in 2 ml hypotonic buffer containing 10 mM NaCl, 1.5 mM MgCl2, and 10 mM Tris–HCl pH 7.5 for 7.5 min. After the hypotonic incubation, cells were transferred to a glass homogenizer and homogenized by pressing straight down with the pestle 20 times, bursting the cells open while maintaining the integrity of mitochondria. Mitochondrial homogenization buffer (MHB) was then added to the 2 ml homogenized sample to achieve a 1X concentration (210 mM mannitol, 70 mM sucrose, 20 mM HEPES, and 2 mM EGTA). The homogenate was then transferred to a clean 5 ml tube and centrifuged at 17,000 × g for 15 min at 4 °C. The resulting pellet was washed in MHB buffer and centrifuged two more times, followed by a resuspension of the mitochondrial pellet in RIPA lysis buffer, and final centrifugation step of 14,000 × g for 10 min at 4 °C to obtain a relatively pure mitochondrial lysate. MASL ELISA was performed using the protocol described above. IP mouse injections Mouse care All animal work was approved by and performed in accordance with the University of Southern California Institutional Animal Care and Use Committee. A total of 24 male and 24 female C57BL/6N mice between 77–84 weeks were housed by sex, 3–4 per cage, in a facility with a 12-h light – 12-h dark cycle at 24C. Food (standard rodent diet, 10% kcal from fat) and water were available ad libitum. Mice were assigned groups to ensure equal weight distribution across treatments. MASL, MASL V18I (both resuspended in PBS), or PBS control were administered twice daily at 10 mg/kg via intraperitoneal injection for three weeks (n = 8/group). Body weight and food consumption were recorded every two days. Food intake was quantified by weighing the feed cage and calculating the change in overall pellet weight over time. Because food intake was measured at the cage level, we cannot determine why baseline food intake differed between groups (Suppl. Figure S5). Body composition was measured once at study initiation and once prior to study termination using an NMR minispec (Bruker, USA). Plasma collection After three weeks of injections, mice were sacrificed and samples were collected. Blood was immediately transferred to 4 mL K2 EDTA tubes and inverted 8–10 times. Samples were kept on ice prior to centrifugation for 10 min at 2000 RCF and 4 C until plasma and blood layers were completely separated. Plasma was collected and immediately frozen at −80C. Hematological analysis 100uL of fresh whole blood was immediately transferred to EDTA tubes. Tubes were inverted 8–10 times and kept on ice prior to hematology analysis within three hours of initial collection. For assessment of circulating immune cell levels in blood, the element HT5 veterinary hematology analyzer (Heska, CO) was used. Samples were aspirated and immune cell composition was measured. Measuring cytokine levels with ELISA Circulating levels of selected pro-inflammatory cytokines were measured using the 96-well plate based Mouse Pro-Inflammatory Panel I sandwich ELISA assay from Mesoscale Discovery (MD). Briefly, the plate was washed three times with Wash Buffer and 50uL of each prepared plasma sample was added along with calibrators and controls. The plate was sealed and incubated for 2 h at RT with gentle shaking. After incubation, the plate was again washed three times in Wash Buffer and detection antibody solution was added to each well and incubated with gentle shaking for 2 additional hours. The plate was again washed and Read Buffer T was added to each well prior to plate analysis. Statistical analysis All biological results were analyzed using independent two-tailed Student’s t-test, with Welch’s test correction in results with unequal variances where appropriate. A cutoff of p < 0.05 was considered statistically significant. Results Sex-differences in mtSNP associations with select immune measures We hypothesized that mtSNPs would associate with changes in immune phenotypes in HRS, and that they would do so in a sex-dependent manner. Previous work establishes that the human immune cell landscape is sex-specific [26]. Recently, the neutrophil–lymphocyte ratio (NLR) has become a factor of interest for morbidity and mortality prognosis and varies by sex [27]. Given these findings, we aimed to explore genotype–phenotype associations between mtSNPs and selected immune phenotypes (% lymphocytes, % neutrophils, and NLR), and to test whether these mtSNP associations differed between sexes. Of the selected 64 mtSNPs, only mtSNP G16393A (rs869031877) was significantly associated with changes in the percentages of lymphocytes (β = 2.1, SE = 0.81, P = 0.008) and neutrophils (β = −2.5, SE = 0.90, P = 0.005), as well as NLR (β = −0.28, SE = 0.14, P = 0.048), particularly among non-Hispanic whites (NHWs) (Fig. 1A-C). One additional mtSNP, A13781G was borderline significantly associated with changes in % lymphocytes and neutrophils, but not NLR. MtSNP G6261A was significantly associated with changes in % neutrophils and NLR, and mtSNP T10035C was significantly associated with changes in % neutrophils only (Suppl. Table ST1 and ST2). Given its significance across all three phenotypes of interest, we selected mtSNP G16393A for downstream analysis and comparison. By directly comparing immune cell levels between groups, NHW carriers of mtSNP G16393A showed a significant increase in the percentage of lymphocytes and a decrease in the percentage of neutrophils, resulting in a lower NLR ratio when compared to non-carriers (Fig. 1D-F). Given these findings, we next aimed to explore whether similar patterns would be observed between sexes.Fig. 1Mitochondrial SNP mtG16393A (rs869031877) is associated with changes in immune cell composition in carriers.—Solar plots (the MiWAS equivalent of a Manhattan Plot) depict significant associations between mtG16393A and 3 markers of immune function in Non-Hispanic Whites. mtG16393A is significantly associated with A % lymphocytes, (extracted from White Blood Cell counts) B % neutrophils, and C an extrapolated neutrophil–lymphocyte ratio (NLR). Bar plots showing D a significant increase in % lymphocytes E a significant decrease in % neutrophils and thus F a significant decrease in the NLR Mitochondrial SNP mtG16393A (rs869031877) is associated with changes in immune cell composition in carriers.—Solar plots (the MiWAS equivalent of a Manhattan Plot) depict significant associations between mtG16393A and 3 markers of immune function in Non-Hispanic Whites. mtG16393A is significantly associated with A % lymphocytes, (extracted from White Blood Cell counts) B % neutrophils, and C an extrapolated neutrophil–lymphocyte ratio (NLR). Bar plots showing D a significant increase in % lymphocytes E a significant decrease in % neutrophils and thus F a significant decrease in the NLR Considering the distinct sex-specific characteristics of immune function [26, 28], we hypothesized that males and females would show differences in associations between mtSNPs and immune markers, and thus stratified the NHW cohort by sex. Indeed, when stratified by sex, we observed a stronger association between mtSNP G16393A and selected immune phenotypes in females, and no significant association between mtSNP G16393A and selected phenotypes in males. Significant associations between selected phenotypes and additional mtSNPs were also observed in females (Fig. 2A-F), but not males (Fig. 2A-C, G-I). Again, mtSNP G16393A was the only mtSNP to be determined significantly associated with all three phenotypes of interest in females, wherein female mtSNP carriers exhibited significantly higher percentages of lymphocytes, lower percentages neutrophils, and a lower NLR than non-carriers (Fig. 2D-F). Of note, only one mtSNP, G1440A was found to be significantly associated with any selected phenotype in males, reaching the significance threshold in % lymphocytes and NLR, but not % neutrophils (Suppl. Figure S2).Fig. 2mtG16393A is associated with changes in levels of immune markers in a sex-specific manner.—Solar plots depict significant associations between mtG16393A and A % lymphocytes, B % neutrophils, and C the extrapolated neutrophil–lymphocyte ratio (NLR) in female NHW carriers. Bar plots show D a significant increase in % lymphocytes E a significant decrease in % neutrophils and F a significant decrease in the NLR in female carriers, while no significant change in G % lymphocytes, H % neutrophils and I NLR is observed in males mtG16393A is associated with changes in levels of immune markers in a sex-specific manner.—Solar plots depict significant associations between mtG16393A and A % lymphocytes, B % neutrophils, and C the extrapolated neutrophil–lymphocyte ratio (NLR) in female NHW carriers. Bar plots show D a significant increase in % lymphocytes E a significant decrease in % neutrophils and F a significant decrease in the NLR in female carriers, while no significant change in G % lymphocytes, H % neutrophils and I NLR is observed in males Phenome-wide association study (PheWAS) of immune phenotypes in HRS and UK Biobank highlight significance of mtSNP G16393A and selected immune markers specifically in NHWs We next employed a phenome-wide association study (PheWAS) in both the HRS and the UK Biobank to further validate MiWAS findings and confirm the associative relationship between mtSNP G16393A and select immune phenotypes across both cohorts. Using the HRS, we leveraged ethnically-diverse data to assess the relationship between mtSNP G16393A and immune phenotypes in an ethnicity-specific manner. Using a frequency assessment from 1000 Genomes, we estimated a 2.5% frequency of mtSNP G16393A in NHWs, a 2.5% frequency in Hispanics, and a 0.2% frequency in African Americans (Suppl. Figure S3). Considering the mtSNP frequency in NHWs and Hispanic populations, as well as the significance of mtSNP G16393A in selected immune phenotypes in NHWs, we conducted a mitochondrial PheWAS analysis to test whether mtSNP G16393A interacted with 14 other immune cell phenotypes in NHW and Hispanic participants (Suppl. Table ST3.2–3.3). Given the frequency of mtSNP G16393A is < 1% in African Americans, we did not conduct a PheWAS analysis in the HRS African American sample. No significant associations between mtSNP G16393A and all selected immune cell phenotypes were observed in Hispanics. In NHWs, using a lenient threshold of uncorrected p < 0.01, % neutrophils and % lymphocytes were once again the only trending immune cell phenotypes associated with mtSNP G16393A, further validating our observations from MiWAS. Consequently, we concluded that mtSNP G16393A is associated with changes in the immune cell landscape in NHW carriers, specifically through interactions with lymphocytes and neutrophils, while the mtSNP exhibits no immune-related effect in Hispanic carriers. We deemed these findings non-significant after adjusting for multiple testing, potentially due to insufficient statistical power. Therefore, we aimed to validate our observations in NHW populations using the “Caucasian” population in the UK Biobank. We specifically targeted the phenotypes “% Lymphocytes” and “% Neutrophils” and found a statistically significant relationship between lymphocytes, but not neutrophils, amongst the larger cohort. This is an important observation to consider when determining the mechanistic impact of mtSNP G16393A on immune function and composition (Suppl. Table ST3.4). Unique gene expression profile of mtSNP G16393A carriers Given the sex-specific interactions between mtSNP G16393A and immune composition, and the sex-dimorphic behavior of immune function, we hypothesized that mtSNP carriers would exhibit sex-dependent gene expression profiles. By characterizing these patterns, we sought to gain insight into the interactions between sex, immunity, and our identified mitochondrial mutation, and also aimed to identify underlying pathways that may contribute to the observed sex-specific features of this mtSNP. To explore this, we conducted a targeted differential gene expression analysis in a technique we have termed mitoSNP-seq. This technique leveraged RNA-Seq results from a subsample of HRS participants to compare the transcriptomic profiles of mtSNP G16393A carriers with those from non-carrier controls. We repeated this technique with sex-stratified cohorts to determine the sex-specific transcriptomic profiles of male and female carriers versus non-carrier controls. Collectively, in all NHW carriers, eight genes were significantly differentially expressed based on an FDR adjusted P-value of < 0.05, while over 800 genes surpassed the unadjusted p-value threshold of < 0.01 (Suppl. Table ST4). While many genes did not pass multiple-test correction, GO enrichment analysis of all genes surpassing an unadjusted p-value cutoff of < 0.01 revealed a distinct cluster of upregulated immune-specific genes, leading to prominent immune-related pathway enrichment. This subset of upregulated genes exposed a set of significantly enriched immune pathways across “biological pathway”, “molecular function”, and “cellular component” enrichment results (Suppl Table ST5; Suppl. Figure S4). Stratified analyses revealed similarities to MiWAS findings, in which a subset of immune-related genes and pathways were indeed significantly differentially expressed in females (Fig. 3B). However, we also observed significant differential expression of a large number of non-immune pathways in males (Fig. 3C). Pathway enrichment from filtered genes with an unadjusted p-value of < 0.01 confirmed enrichment of immune-specific pathways in females which were lost in males (Suppl. Table ST6). Enrichment included B cell immune response and immunoglobulin specific pathways in females, while enriched pathways in males were largely metabolic. These observations not only highlight the sex-specific nature of immune-related observations in mtSNP G16393A carriers but also suggest alternative regulation in males that may be linked to metabolism (Fig. 3D-I).Fig. 3Transcriptomic analyses reveal differences in transcript expression between male and female carriers. A Infographic explaining the mitoSNP-seq analysis tool. B Volcano plot showing the distribution of DEGs in female carriers and C volcano plot showing the distribution of DEGs in male carriers. Black dotted line represents an unadjusted p < 0.01, used for gene filtration for GO enrichment analysis. Red dotted line represents adjusted p < 0.05. GO Enrichment results for upregulated D biological process E molecular function and F cellular component pathways in female carriers. GO Enrichment results for upregulated G biological process H molecular function and I cellular component pathways in male carriers. All pathways survive an adjusted q < 0.05 Transcriptomic analyses reveal differences in transcript expression between male and female carriers. A Infographic explaining the mitoSNP-seq analysis tool. B Volcano plot showing the distribution of DEGs in female carriers and C volcano plot showing the distribution of DEGs in male carriers. Black dotted line represents an unadjusted p < 0.01, used for gene filtration for GO enrichment analysis. Red dotted line represents adjusted p < 0.05. GO Enrichment results for upregulated D biological process E molecular function and F cellular component pathways in female carriers. GO Enrichment results for upregulated G biological process H molecular function and I cellular component pathways in male carriers. All pathways survive an adjusted q < 0.05 mtSNP G16393A is located in a sORF encoding for a novel microprotein called MASL Through re-annotation of the mitochondrial genome, we identified a putative sORF-encoded microprotein in the regulatory (D-loop) region that overlaps with the location of mtSNP G16393A (Fig. 4A). The variant causes a valine-to-isoleucine substitution at the 18th amino acid of the predicted microprotein, termed MASL, encoding a mutant version of the microprotein that we have termed MASL V18I. Using PEP-FOLD 4 protein prediction software, we modeled the predicted 31 amino acid microprotein and the structural impact of the G16393A alteration. Based on IUPred prediction of disordered protein regions, we characterized MASL as a largely disordered microprotein with strong binding affinity at the N-terminus (Fig. 4D). While the V18I substitution did not dramatically alter the overall secondary structure of MASL, PEP-FOLD modeling predicted a less structured region at the site of the amino acid change (Fig. 4B). DeepLoc localization software predicted MASL distribution across multiple compartments of the cell, including the cytoplasm and the extracellular space, a finding that is compatible with detection of the microprotein in plasma (Fig. 4C). Using female-derived HL-60 cells, fluorescent staining demonstrated potent in vitro expression of MASL. This expression was predominantly localized near the nucleus, with additional intermittent signal across the cell in the cytoplasmic space (Fig. 4E). Of particular note, the use of Hoescht as a nuclear marker highlights distinct MASL localization to the nucleus, distinguished by the dark unstained shadow region indicative of the nucleolus. Consistent with these observations, ELISA analysis of subcellular fractions confirmed nuclear localization in HL-60 cells (Fig. 4F).Fig. 4Features of MASL microprotein and corresponding mtSNP variant. A Location of predicted microprotein and corresponding nucleotide and amino acid sequences. B Predicted model of microprotein structure of wildtype MASL and mtG16393A (V18I) variant. C Predicted localization and D predicted disordered regions of MASL wildtype. E Staining in HL-60 neutrophil immune cell line showing MASL expression (left) and successful blocking with MASL antibody (right). F ELISA of subcellular fractions in HL-60 neutrophils highlights nuclear localization. ELISA results showing MASL levels between G sexes and H age groups. I NLR between groups with increased versus decreased MASL levels. A p-value cutoff of p < 0.05 represents significance Features of MASL microprotein and corresponding mtSNP variant. A Location of predicted microprotein and corresponding nucleotide and amino acid sequences. B Predicted model of microprotein structure of wildtype MASL and mtG16393A (V18I) variant. C Predicted localization and D predicted disordered regions of MASL wildtype. E Staining in HL-60 neutrophil immune cell line showing MASL expression (left) and successful blocking with MASL antibody (right). F ELISA of subcellular fractions in HL-60 neutrophils highlights nuclear localization. ELISA results showing MASL levels between G sexes and H age groups. I NLR between groups with increased versus decreased MASL levels. A p-value cutoff of p < 0.05 represents significance MASL shows sex-dimorphic characteristics and may decline with age Given our observation that mtSNP G16393A may be sex-specific, we hypothesized that levels of the corresponding microprotein MASL may also be different between sexes. To test this, we measured levels of circulating MASL in 79 human plasma samples (33 males and 46 females). We found MASL levels were reduced in females, as well as patients over the age of 50 (Fig. 4 G,H). This observation is not surprising, given that levels of several other mitochondrial microproteins decline with age [29]. We also measured the association between levels of MASL and NLR. MASL levels were measured before and after bariatric surgery and patients were categorized by either an increase or decrease in MASL levels. The neutrophil–lymphocyte ratio was then compared between “increased” and “decreased” groups. When we compared NLR one year after surgery between the MASL increased group (in which MASL levels rose after surgery) and the MASL decreased group, the NLR was significantly lower in the MASL increased group compared to the decreased group (1.08 ± 0.43 vs. 1.47 ± 0.55. P = 0.044) (Fig. 4I). Since low MASL levels may be associated with higher NLR, and females have lower levels of MASL, it is plausible that our observed NLR increase in female mtSNP carriers could be related to MASL levels or MASL function altogether. To further test the sex-specific and immunomodulatory nature of MASL, male and female C57BL/6N mice at approximately 1.5 years old were treated with MASL or the corresponding variant for three weeks. Old mice were selected to further assess the role of MASL within an aging and potentially compromised immune environment. Over three weeks, mice received synthetic MASL, MASL V18I, or saline control via intraperitoneal injection. After sample collection, circulating immune cell and plasma cytokine levels were measured and compared between groups. Results revealed sex-specific effects on immunity, however many results contradicted our initial hypotheses based on MiWAS and mitoSNP-seq results. MASL significantly altered body composition in males (Fig. 5A,C), but not in females (Fig. 5B,D), with no significant effects on food intake for either sex (Suppl. Figure S5). Immunomodulatory effects also appeared to be stronger in males, marked by significant changes in a multitude of immune cells and circulating cytokine levels (Tables 1 and 2). Specifically, lymphocyte levels decreased significantly in males, while levels of neutrophils and other granulocytes were increased – opposite to the expected trends from MiWAS, wherein mtSNP carriers showed higher lymphocytes and lower neutrophils. In contrast, females displayed no significant change in circulating lymphocytes and only trending decreases in neutrophil levels in females, conflicting with previous data suggesting female-specific interactions of the microprotein-associated mtSNP. Furthermore, the comparative analysis between MASL and MASL V18I showed minimal differences in overall immune cell composition. However, mice treated with wild-type MASL exhibited changes in levels of certain inflammatory cytokines, such as IL-5 and KC/GRO, unlike those treated with MASL V18I. These results highlight a complex immunomodulatory action of MASL and the corresponding variant which challenge our previous thinking, exposing areas for future investigation.Fig. 5Body composition is altered in male – but not female – mice treated with MASL. Body composition results show muscle is A significantly increased in males treated with wildtype MASL or V18I but B is not significantly altered in females, and that fat is C significantly decreased in males treated with wildtype MASL but D not altered in females. A p-value cutoff of p < 0.05 represents significance. Of note, no comparisons between MASL and V18I were found to be statistically significant within groupsTable 1Immune cell composition of male and female C57BL6 mice treated with MASL, MASL V18I, or Control for 3 weeks. Values presented as mean ± SEM. Two-sided Student’s t-test was used to calculate significance between control and treatment groups. Significant values (P < 0.05) are bolded Cell TypeMalesFemalesMean ± SEMP (v Ctrl)P (v V18I)Mean ± SEMP (v Ctrl)P (v V18I)NEU %Control18.4875 ± 1.52Control53.075 ± 3.64MASL22.2638 ± 1.300.0570.82MASL45.125 ± 2.040.0770.534V18I23.2125 ± 2.130.093V18I47.6428 ± 3.520.306LYM %Control73.7375 ± 1.98Control23.725 ± 4.57MASL67.0375 ± 2.020.0330.66MASL27.0375 ± 4.590.6170.613V18I65.6125 ± 2.4110.021V18I23.3285 ± 5.590.957MONO %Control3.5375 ± 0.38Control7.5 ± 1.89MASL4.1625 ± 0.590.3890.93MASL10.5375 ± 5.180.5900.343V18I4.225 ± 0.300.178V18I4.986 ± 1.010.281EOS %Control3.0375 ± 0.40Control4.3125 ± 0.89MASL4.0625 ± 0.310.0630.79MASL5.35 ± 0.880.4220.075V18I3.8875 ± 0.550.235V18I11.014 ± 2.980.040BAS %Control1.2 ± 0.14Control11.3875 ± 1.47MASL2.1 ± 0.200.0030.17MASL11.95 ± 1.650.8030.763V18I3.0625 ± 0.640.013V18I13.03 ± 3.240.637All bolded entries meet P value thresholds of P<0.05Table 2Plasma cytokine levels in male and female C57BL6 mice. Values presented as mean ± SEM. Two-sided Student’s t-test was used to calculate significance between control and treatment groups, and between treatment groups. Significant values (P < 0.05) are boldedCytokineMaleFemaleMean ± SEMP (v Ctrl)P (v V18I)Mean ± SEMP (v Ctrl)P (v V18I)IFN-γ (pg/ml)Control0.4695 ± 0.07Control0.67 ± 0.16MASL0.5 ± 0.070.7240.665MASL0.57 ± 0.180.6700.979V18I0.55 ± 0.060.436V18I0.57 ± 0.120.632IL-10(pg/ml)Control11.87 ± 0.99Control51.54 ± 32.45MASL13.48 ± 0.840.2360.554MASL41.36 ± 17.240.7860.848V18I14.54 ± 0.660.079V18I37.32 ± 9.910.700IL-1β(pg/ml)Control0.98 ± 0.14Control0.95 ± 0.22MASL1.33 ± 0.080.0490.094MASL1.13 ± 0.060.4360.357V18I1.01 ± 0.140.741V18I1.45 ± 0.350.235IL-2(pg/ml)Control1.27 ± 0.21Control1.17 ± 0.11MASL1.28 ± 0.090.9640.002MASL1.07 ± 0.110.5520.022V18I1.66 ± 0.060.069V18I1.60 ± 0.180.058IL-4(pg/ml)Control0.10 ± 0.02Control0.12 ± 0.04MASL0.11 ± 0.030.7580.060MASL0.12 ± 0.030.9310.270V18I0.23 ± 0.050.025V18I0.19 ± 0.060.369IL-5(pg/ml)Control1.54 ± 0.16Control1.12 ± 0.14MASL2.65 ± 0.450.0360.027MASL2.17 ± 0.360.0160.781V18I1.54 ± 0.110.849V18I2.45 ± 1.000.179IL-6(pg/ml)Control18.54 ± 3.06Control25.27 ± 5.62MASL16.76 ± 2.000.6330.0168MASL27.85 ± 4.000.7140.081V18I35.62 ± 10.730.227V18I101.49 ± 41.610.074KC/GRO(pg/ml)Control108.13 ± 9.00Control102.99 ± 16.83MASL84.49 ± 4.310.0330.0003MASL140.22 ± 12.130.0940.690V18I112.75 ± 4.150.688V18I132.96 ± 13.000.192TNF-α(pg/ml)Control7.18 ± 1.02Control10.84 ± 2.58MASL7.76 ± 1.020.6070.416MASL10.64 ± 2.110.9530.565V18I6.88 ± 0.600.971V18I12.81 ± 3.120.632All bolded entries reach significance thresholds of P<0.05 Body composition is altered in male – but not female – mice treated with MASL. Body composition results show muscle is A significantly increased in males treated with wildtype MASL or V18I but B is not significantly altered in females, and that fat is C significantly decreased in males treated with wildtype MASL but D not altered in females. A p-value cutoff of p < 0.05 represents significance. Of note, no comparisons between MASL and V18I were found to be statistically significant within groups Table 1Immune cell composition of male and female C57BL6 mice treated with MASL, MASL V18I, or Control for 3 weeks. Values presented as mean ± SEM. Two-sided Student’s t-test was used to calculate significance between control and treatment groups. Significant values (P < 0.05) are bolded Cell TypeMalesFemalesMean ± SEMP (v Ctrl)P (v V18I)Mean ± SEMP (v Ctrl)P (v V18I)NEU %Control18.4875 ± 1.52Control53.075 ± 3.64MASL22.2638 ± 1.300.0570.82MASL45.125 ± 2.040.0770.534V18I23.2125 ± 2.130.093V18I47.6428 ± 3.520.306LYM %Control73.7375 ± 1.98Control23.725 ± 4.57MASL67.0375 ± 2.020.0330.66MASL27.0375 ± 4.590.6170.613V18I65.6125 ± 2.4110.021V18I23.3285 ± 5.590.957MONO %Control3.5375 ± 0.38Control7.5 ± 1.89MASL4.1625 ± 0.590.3890.93MASL10.5375 ± 5.180.5900.343V18I4.225 ± 0.300.178V18I4.986 ± 1.010.281EOS %Control3.0375 ± 0.40Control4.3125 ± 0.89MASL4.0625 ± 0.310.0630.79MASL5.35 ± 0.880.4220.075V18I3.8875 ± 0.550.235V18I11.014 ± 2.980.040BAS %Control1.2 ± 0.14Control11.3875 ± 1.47MASL2.1 ± 0.200.0030.17MASL11.95 ± 1.650.8030.763V18I3.0625 ± 0.640.013V18I13.03 ± 3.240.637All bolded entries meet P value thresholds of P<0.05 Immune cell composition of male and female C57BL6 mice treated with MASL, MASL V18I, or Control for 3 weeks. Values presented as mean ± SEM. Two-sided Student’s t-test was used to calculate significance between control and treatment groups. Significant values (P < 0.05) are bolded All bolded entries meet P value thresholds of P<0.05 Table 2Plasma cytokine levels in male and female C57BL6 mice. Values presented as mean ± SEM. Two-sided Student’s t-test was used to calculate significance between control and treatment groups, and between treatment groups. Significant values (P < 0.05) are boldedCytokineMaleFemaleMean ± SEMP (v Ctrl)P (v V18I)Mean ± SEMP (v Ctrl)P (v V18I)IFN-γ (pg/ml)Control0.4695 ± 0.07Control0.67 ± 0.16MASL0.5 ± 0.070.7240.665MASL0.57 ± 0.180.6700.979V18I0.55 ± 0.060.436V18I0.57 ± 0.120.632IL-10(pg/ml)Control11.87 ± 0.99Control51.54 ± 32.45MASL13.48 ± 0.840.2360.554MASL41.36 ± 17.240.7860.848V18I14.54 ± 0.660.079V18I37.32 ± 9.910.700IL-1β(pg/ml)Control0.98 ± 0.14Control0.95 ± 0.22MASL1.33 ± 0.080.0490.094MASL1.13 ± 0.060.4360.357V18I1.01 ± 0.140.741V18I1.45 ± 0.350.235IL-2(pg/ml)Control1.27 ± 0.21Control1.17 ± 0.11MASL1.28 ± 0.090.9640.002MASL1.07 ± 0.110.5520.022V18I1.66 ± 0.060.069V18I1.60 ± 0.180.058IL-4(pg/ml)Control0.10 ± 0.02Control0.12 ± 0.04MASL0.11 ± 0.030.7580.060MASL0.12 ± 0.030.9310.270V18I0.23 ± 0.050.025V18I0.19 ± 0.060.369IL-5(pg/ml)Control1.54 ± 0.16Control1.12 ± 0.14MASL2.65 ± 0.450.0360.027MASL2.17 ± 0.360.0160.781V18I1.54 ± 0.110.849V18I2.45 ± 1.000.179IL-6(pg/ml)Control18.54 ± 3.06Control25.27 ± 5.62MASL16.76 ± 2.000.6330.0168MASL27.85 ± 4.000.7140.081V18I35.62 ± 10.730.227V18I101.49 ± 41.610.074KC/GRO(pg/ml)Control108.13 ± 9.00Control102.99 ± 16.83MASL84.49 ± 4.310.0330.0003MASL140.22 ± 12.130.0940.690V18I112.75 ± 4.150.688V18I132.96 ± 13.000.192TNF-α(pg/ml)Control7.18 ± 1.02Control10.84 ± 2.58MASL7.76 ± 1.020.6070.416MASL10.64 ± 2.110.9530.565V18I6.88 ± 0.600.971V18I12.81 ± 3.120.632All bolded entries reach significance thresholds of P<0.05 Plasma cytokine levels in male and female C57BL6 mice. Values presented as mean ± SEM. Two-sided Student’s t-test was used to calculate significance between control and treatment groups, and between treatment groups. Significant values (P < 0.05) are bolded All bolded entries reach significance thresholds of P<0.05 Discussion The role of mitochondrial signaling in immunity presents one key area of interest for understanding the complex immune landscape as it relates to human health and disease. To fully capture the dynamic communication network involving mitochondria and immune cells, sex must be considered as an important underlying factor that shifts both mitochondrial and immune behavior. Within such a highly sensitive biological system, differences in immune cell composition and activity are key underlying contributors to various autoimmune and inflammatory conditions. When determining the mechanisms that drive the progression of many related health conditions, especially autoimmune disorders, sex-differences in immunity are a crucial factor [30]. Differences in mito-immune communication between sexes may offer additional complexity to the model, in which changes to mitochondrial machinery may influence immune activity. As mitochondrial genetics and associated microproteins continue to grow in importance within the context of cell health, it is highly likely that sex-dimorphic mitochondrial microproteins may be pivotal in regulating this system. In this report, we conducted a targeted MiWAS screening to determine associations between mtDNA variants and changes in immune cell composition while using sex-stratification to consider sex differences in immunity. In the process, we identified a novel mitochondrial microprotein called MASL, which we suggest may modulate immunity in a sex-dependent manner and may have a larger impact on sex-dimorphic system regulation overall. Initial MiWAS results demonstrated female-specific associations between a mtSNP (which alters the nucleotide at position 16,393 from a G to an A) and changes in immune cell composition. This mtSNP was also found to be associated with changes to overall metabolism-related gene expression in male carriers. During a screening of mtDNA sORFs, mtSNP G16393A was found to map to a sORF that encodes the putative microprotein MASL. The mtSNP G16393A alteration is predicted to cause a protein-level valine-to-isoleucine substitution at the 18th amino acid of MASL, encoding a version of MASL which we have termed MASL V18I. In this early characterization of MASL, we observe sex-specific interactions and a puzzling but prevalent immunomodulatory relationship that should continue to be explored. mtSNP G16393A is associated with changes in neutrophil and lymphocyte composition and upregulation of immune pathways in mtSNP carriers MiWAS and PheWAS results determined that mtSNP G16393A is significantly associated with altered lymphocyte and neutrophil levels specifically in NHWs, and especially in females. This remained true for our extrapolated neutrophil–lymphocyte ratio, which was reduced in mtSNP carriers. While NLR is an emerging immune marker related to morbidity and mortality, it is important to note that NLR remained in a healthy range for all participants [31]. Thus, while our findings related to this measure are important to consider especially within the context of sex-specific effects, we will focus more on observed differences in neutrophil and lymphocyte levels, which were used to define NLR. We observe significant increases in lymphocytes and decreases in neutrophils in mtSNP carriers and have speculated several explanations for this observation, the first being regulation of hematopoiesis. White blood cells, including lymphocytes and neutrophils, develop during a multi-stage stem cell differentiation process called hematopoiesis [32]. Given hematopoietic stem cells (HSCs) are highly sensitive to environmental stimuli, and rely on redox signaling to undergo maturation [33], we propose that this mitochondrial variant may indirectly influence the regulation of hematopoiesis, favoring lymphocyte differentiation and reducing neutrophil maturation in some capacity in carriers. Sex steroids offer an additional explanation that should be explored in the future. Sex steroids play an important role in regulating immune cell maturation and response pathways, and contribute to observed sex differences in immune function [34, 35]. Differences in sex-hormone signaling and pathway activation could play a part in exacerbating the variant effect on immune response, driving an association between G16393A and neutrophils, lymphocytes, and NLR in females only. Regardless, while the underlying mechanism has yet to be elucidated, the association between a mtDNA variant and changes in immune cell composition could indicate differences in mitochondrial signaling and immune cell maturation between carriers and non-carriers. To explore this relationship further, we carried out a targeted RNA-seq analysis to determine the gene expression profile of the mtSNP carriers. Our mitoSNP-seq analysis revealed potential underlying mechanisms for the observed differences in immune cell composition that are pronounced in females but not in males (Fig. 3). While no immune-related protein coding genes were significantly differentially expressed, we observed a distinct cluster of differentially expressed genes across all NHWs that were further distinguished after sex-stratification. Upon sex-stratification, 12 genes were distinctly differentially expressed in females, while 54 genes were differentially expressed in males. This observation points to a unique sex-dimorphic characteristic of mtSNP G16393A which we hypothesize translates to characteristics of the corresponding microprotein (Suppl. Table ST4.2–4.3). Further, GO enrichment analyses showed clear enrichment of immune-associated terms in female gene sets, particularly those involving immunoglobulin signaling and antigen binding, while enrichment of metabolism specific pathways was distinctly observed in male gene sets. Given the role of immunoglobulin signaling in female-specific immune modulation, enrichment of immunoglobulin specific terms in females, but not in males, is particularly noteworthy. These findings further support the importance of sex-associated immune signaling interactions in driving the effects of the G16393A alteration, especially in female carriers. Male carriers, on the other hand, may exhibit changes in energy metabolism that are not clearly related to one specific phenotype of interest, but may have larger consequences on overall cell health. Together, these observations demonstrate an important functional impact of the mtSNP and its associated microprotein, revealing sex-dependent immune and metabolic response pathways that should continue to be explored. Characterization and functional dynamics of the sORF encoded microprotein MASL in neutrophil activity and immune regulation Mitochondrial microproteins have previously been identified through initial application of MiWAS mtSNP associations [15, 36]. Previous work elucidates mitochondrial microproteins as potent regulators of cellular function, acting on behalf of the mitochondria to modulate biological systems within the context of many diseases [29, 36, 37]. Here, we introduce a novel mitochondrial microprotein to join this unique subclass of mitochondrial communicators. We predicted the region surrounding mtSNP G16393A encodes wild-type MASL, and therefore the alteration encodes a microprotein variant termed MASL V18I. MASL is a 31 amino acid microprotein encoded in the d-Loop region of the mitochondrial genome. Structural predictions of both microprotein variants indicate limited differences between the two, wherein wild-type MASL is a largely disordered protein that is predicted to localize to several cellular compartments (Fig. 4 B-D). Immunofluorescent staining of endogenous MASL in HL-60 neutrophils indicates non-uniform MASL distribution in both cytoplasmic and nuclear components (Fig. 4E), a finding that is further reinforced by the detection of MASL in nuclear fractions (Fig. 4F). This nuclear staining pattern may reflect euchromatin enrichment, suggesting proximity to actively transcribed DNA. These observations align with functional data that implicates the associated alteration (G16393A) in sex-specific transcriptional regulation. Although in silico nuclear localization scores were considerably lower compared to other components, staining patterns suggest that this prediction may reflect subnuclear specificity rather than an absence of nuclear localization entirely. Importantly, the use of HL-60 cells derived from a female donor provides important biological relevance to the sex-specific phenotypes observed in our human and mouse models. The distinct sex-specific associations of the G16393A variant prompted further investigation into whether MASL may be differentially regulated by sex. In human plasma, a higher NLR was associated with lower levels of MASL, and significantly lower levels of MASL were observed in females compared to males. Furthermore, while still in a healthy range, female carriers of mtSNP G16393A exhibited an elevated NLR. Together, these observations suggest that MASL may regulate neutrophil and lymphocyte activity, and thus lower or non-functional MASL may directly increase the NLR and impact overall immune function. While it is inconclusive how MASL may regulate the NLR, and specifically how that relates to sex, the combination of endogenous staining, sex-specific transcription patterns, and plasma detection outlines a biological framework that must continue to be explored. In mice, while sex-specific effects of MASL were noted, MASL seemingly induced more pronounced alterations in the immune cell composition of male mice rather than females, and also significantly altered the body composition of males, but not females. Given the enrichment of metabolism-associated pathways in male G16393A carriers, it is notable that body composition is uniquely shifted in males. However, such pronounced shifts in immune cell composition in males, but not females, was unexpected. Additionally, the MASL V18I variant exhibited effects on metabolism and immune cell composition that were comparable to the wild-type treatment group, suggesting that the variant may impair cell function through alternative mechanisms. These results present a fascinating contradiction between mtSNP G16393A and the synthesized microprotein that are difficult to reconcile from these observations alone. Males and females do inherently differ in immune response. Given that MASL may function as a mitochondrial signal that regulates cell maturity and activity, it is possible that immune cell signaling pathways found to be more prevalent in females may be more sensitive to disruptions in mitochondrial signaling, which mtSNP G16393A may induce. In contrast, the distinct composition of circulating immune cells in males may mount a stronger response to direct MASL treatment, offering one explanation for our observed confounding results. These findings are supported by differences in cytokine levels observed between groups. For instance, while IL-5, a cytokine that promotes granulocyte maturation [38], was increased in both males and females, increases in Il-1b (another critical regulator of granulopoiesis [39]) were distinct to males. This suggests that MASL may initiate a more robust signaling response to induce a granulocyte maturation in males that is notably absent in females. The MASL V18I variant induced only limited alterations to the cytokine response when compared to controls, while significant differences in the levels of several cytokines were observed between MASL V18I and MASL wild-type. This observation further suggests that wild-type MASL may be a more potent signaling effector, while the mitochondrial variant may have impaired signaling functionality that impacts immunity outside of immune composition and metabolism. However, these observations also imply that MASL may have alternative activation pathways in immune cells in males that do not require these cytokines. From these results, we can conclude an indirect regulatory impact of MASL in vivo, wherein these findings offer an area of interest that could support a deeper mechanistic understanding of MASL if explored. While these results highlight the complexity of the immunomodulatory landscape for which MASL acts, it is evident that follow-up studies will be essential to dissect these interactions more thoroughly and to determine the exact mechanism for which MASL and the corresponding mtSNP influence immune response. Limitations While this study highlights the discovery and early characterization of a novel immunomodulatory mitochondrial microprotein, there are limitations to this work that should be considered. First, observed mtSNP associations relied on secondary analyses from sequencing data in the Health and Retirement Study, and experimental validation was performed using ELISA. Experimental validation of mtSNP G16393A using PCR was not performed. In the future, validating the presence of mtSNP G16393A in the population by leveraging long-range PCR or mitochondrial enrichment of human samples could support additional confidence in the biological importance of this mtSNP. Also, it is important to consider the functional limitations of MiWAS in isolating the effects of specific mtSNPs within larger haplogroups [40]. However, by identifying and characterizing the putative microprotein that surrounds this locus, we provide a functional link that further validates the observed phenotypic effects of the alteration. Additionally, in mitoSNPseq analyses, GO enrichment tools were fed a gene list that consisted of genes that did not meet adjusted p-value cutoffs. Alternatively, we filtered genes using an unadjusted p < 0.01, specifically to highlight significantly enriched pathways which may still be of interest for understanding phenotypic observations of mtSNP G16393A. While this result is still of interest, specific genes which contribute to significantly enriched pathway outputs should be carefully interpreted. Finally, while we hypothesized we would observe specific immunomodulatory effects of MASL in vivo, our results did not directly support our hypothesized observations. Therefore, at this time we can only postulate an indirect functional impact of MASL and call for future and ongoing characterization of the peptide and its variants to further examine the dynamic relationship between MASL and the sexually dimorphic immune environment. Conclusions We identified a novel mitochondrial-derived microprotein that is characterized by dynamic, sex-dimorphic, and immunomodulatory properties. In a comprehensive screen of immune cell populations, the mitochondrial alteration located in the microprotein-encoding sORF was significantly associated with changes in neutrophil and lymphocyte composition, a phenotype that was notably more prominent in females. While downstream characterization of the associated microprotein, which we call MASL, can only indicate an indirect modulatory relationship with immunity, such findings highlight the importance of mitochondrial signaling in regulating the immune landscape and suggest three key outcomes. First, variants in the mitochondrial genome may be linked to immune dysregulation in carriers and may further contribute to differences in immune cell composition and function in a sex-dependent manner. Second, mitochondria may use a novel subclass of signaling molecules to partially regulate immune function; given the sensitivity of MASL levels to physiological factors in human plasma, and its demonstrated impact on immune signaling in vivo, MASL should be included in this subclass. Considering the regulation of both mitochondria and the immune system differs by sex, mitochondrial-derived microproteins such as MASL may also interact in a sex-specific manner. Although the mechanism of action of MASL currently remains unknown, observed associations with the related mtSNP (G16393A) highlight three potential mechanisms to explore in the future: 1) interactions with sex hormones and associated signaling cascades, 2) the regulation of immune cell differentiation and maturity, and 3) the modulation of these systems through gene transcription regulation. Finally, mitochondrial genome variants may lead to differences in health outcomes of mtSNP carriers by inducing functional changes in an encoded microprotein. While a biological framework for MASL activity has been outlined here, several underexplored features of this microprotein warrant further investigation. When screening for sequence similarity using BLASTp, we found that MASL shows no sequence homology with known proteins. However, it is important to note that the size (31 amino acids) and origin (the regulatory region of the mitochondrial genome) of MASL may limit its detectability in BLASTp. Future advancements in targeted mass spectrometry or mitochondrial-specific ribosome profiling could provide insight into the structural properties of this novel microprotein. Furthermore, detection of MASL in human plasma suggests signaling via secretion from the mitochondria. Thus, while MASL is encoded within mtDNA, plasma detection indicates a dynamic action outside of the mitochondrion, wherein MASL may act as a responsive messenger on behalf of mitochondria much like other MDPs. Although the exact mechanism of secretion remains unknown, it has been suggested that MDPs may be transported by mitochondrial-derived vesicles across cells and tissues to serve various modulatory functions [41]. Confirming the exact mechanism of secretion for MASL, as well as other MDPs, will be an important component in determining the direct mechanism of action of this microprotein. Collectively, these results highlight the identification and early characterization of a novel mitochondrial-derived microprotein, adding to a growing subclass of mitochondrial microproteins with potent biological consequences. Supplementary Information Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Supplementary Material 4. Supplementary Material 5. Supplementary Material 6. Supplementary Material 7. Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Supplementary Material 4. Supplementary Material 5. Supplementary Material 6. Supplementary Material 7.