Subacute ruminal acidosis (SARA) negatively affects the dairy industry by decreasing dry matter intake, milk production, profitability, and increasing culling rate and death loss. These consequences are related to a loss of barrier function in the rumen epithelial tissue associated with a reduction in ruminal pH. Six lactating Holstein cows were used in a replicated 2 × 2 Latin square design to determine the effects of SARA induction on the rumen epithelial tissue. Experimental periods were 10 d with d 1 to 3 for ad libitum intake of control diet, followed by 50% feed restriction on d 4, and ad libitum access on d 5 of the control diet (control) or control diet + 4.6 kg of a 50:50 wheat/barley pellet (challenge). Ruminal papillae biopsies were collected on d 1 and 6 of each period and stored at −80 C. Quantitative RT-PCR was used to determine expression genes related to barrier function with all reactions run in triplicate. Three reference genes (CMTM6, ERC1, and MRPL39) were selected due to stable expression across animals and times. Data were analyzed using the MIXED procedure of SAS with day, treatment, period, and square as fixed effects and cow as a random effect. Of the evaluated barrier function genes, the greatest relative abundance was observed for CLDN1 followed by CLDN4. Expression of CLDN1, CLDN4, and JAM2 was upregulated on d 6 (P < 0.1). Greater expression of TJP1/ZO-1 was observed for the control treatment (P = 0.086). A treatment × day effect was observed for OCLN (P = 0.07) as expression was upregulated on d 6 for the challenge treatment but downregulated for the control. No treatment effect (P > 0.1) was observed for CLDN1, CLDN4, JAM2, CAR, TLR2, TLR4, and IGFBP3. Toll-like receptors recognize bacterial components and are capable of upregulating barrier function of epithelial tissue. Both TLR2 and TLR4 were downregulated on d 6 (P = 0.05 and P = 0.1, respectively). Collectively, results suggest feed restriction and subsequent refeeding caused a greater effect on expression of barrier function genes than the additional starch in the challenge treatment. However the results may still be indicative of the rumen epithelium tissue response to SARA.