**Background:** The oral mucosa is a unique barrier site exposed to diverse microbiota and pathogens, yet CD8+ resident memory T cell (TRM) biology in this tissue is poorly understood. This study aimed to characterize the distribution, ontogeny, and immunostimulatory functions of oral CD8+ TRM, and to develop a method for their in vivo depletion.
**Methods:** Naive OT-I or P14 CD8+ T cells were adoptively transferred into C57BL/6J mice, followed by systemic infection with VSV-ova or LCMV-Arm. A viral-prime epitope-pull (VPEP) model was developed: 5–6 days after infection, the oral surfaces were swabbed with cognate peptide (SIINFEKL or gp33) dissolved in a slurry of fine dental pumice and contraceptive gel containing nonoxynol-9 (N-9). TRM residency was confirmed by parabiosis surgery. CD103+ TRM were depleted in vivo using anti-CD103 antibodies conjugated to Saporin toxin (103-SAP) in CD103-/- hosts. Bulk RNA sequencing was performed on buccal mucosa 12 hours after oral peptide re-exposure. Immunofluorescence and Ce3D microscopy were used to visualize TRM distribution.
**Key Results:** Systemic viral infections generated memory CD8+ T cells in the oral mucosa, and VSV-ova/OT-I cells outperformed LCMV-Arm/P14 cells in magnitude and TRM marker expression (CD103+, CD69+, Ly6Clo). VPEP yielded ~4-fold more memory OT-I T cells in the oral mucosa compared to systemic infection alone (Fig. 3B). Parabiosis confirmed that CD103+ cells were resident (Fig. 2B–C). VPEP-elicited TRM were maintained for at least 10 months (Fig. 3D). TGFβR signaling was required for CD103 expression (Fig. 3E). TRM were broadly distributed in tongue, buccal mucosa, gingiva, palate, and periodontium, often near taste buds and salivary ducts (Fig. 4). Oral TRM reactivation with SIIN peptide induced a ~4-fold increase in total oral OT-I T cells, largely due to recruitment of circulating CD103neg Ly6Chi cells (Fig. 5B). RNAseq revealed 847 significantly upregulated genes (Log2 fold change >1, FDR <0.05) in SIIN- vs. gp33-treated buccal mucosa, including T cell effector molecules (Ifng, Prf1, Gzma, Gzmb, Gzmc), chemokines, TLRs, keratin/mucus genes, complement, and periodontitis-related genes (Mmp8, Mmp9, Mmp3, Mmp25, Tnfsf10, Il1b) (Figs. 6–7). In vivo depletion of CD103+ TRM with 103-SAP reduced CD103+ TRM by >90% (Fig. 8A) and abrogated the majority of transcriptional changes: only 177 genes were significantly upregulated in SIIN-103 vs. gp33, compared to 529 in SIIN vs. gp33 (Fig. 8D). Histologically, TRM depletion eliminated OT-I T cell clusters and reduced pan-leukocyte infiltration upon antigen re-exposure (Fig. 8I–L).
**Clinical Implications:** This study establishes that oral CD8+ TRM can be locally augmented and are potent mediators of local immunity and inflammation. The VPEP model provides a framework for developing mucosal vaccines targeting the oral cavity. The demonstration that CD103+ TRM depletion mitigates inflammatory responses suggests that targeting TRM could be a therapeutic strategy for oral inflammatory diseases such as periodontitis, oral lichen planus, or Sjogren's syndrome. Additionally, TRM reactivation may be exploited to enhance antitumor immunity in oral cancers. The findings also highlight the potential for TRM to influence taste and salivary gland function, with implications for chemosensory and autoimmune disorders.