Sealing can arrest caries lesions. We aimed to evaluate if sealing effects and kinetics are bacterial-strain and sealing-material specific. Human dentin discs were mounted in a dual-chamber device. Caries lesions were induced chemically and contaminated with either Lactobacillus rhamnosus (LR) or Streptococcus sobrinus (SS). For (1) kinetics assessment, the initial bacterial load and the sealing period were varied, and lesions sealed using a self-etch adhesive and composite. For (2) comparing materials, six sealing protocols (#1-#6) were evaluated: 1# Self-etch adhesive plus composite placed without a liner, or #2 calcium hydroxide, or #3 mineral trioxide aggregate, or #4 Biodentine liners; #5 antibacterial adhesive plus composite; #6 glass ionomer cement. Pulpal fluid flow was simulated during sealing. The outcome was the number of surviving bacteria (CFU) per g dentin. For LR, bacterial survival increased significantly with increasing initial bacterial load and decreased with longer sealing periods. The relative reduction followed a first-order kinetics. More LR survived under calcium hydroxide or MTA than other materials (p < 0.001). For SS, nearly no bacteria survived sealing regardless of sealing period, initial bacterial load or sealing material. In conclusion, sealing effects and kinetics were strain- and material-specific.