The objective of this study was to evaluate microbial leakage by means of genome counts, through the implant-abutment interface in dental implants with different Morse taper abutments. Fifty-six samples were prepared and divided in four groups: CMC TB (14 Cylindrical Implants-14 TiBase Abutments), CMX TB (14 Conical Implants-14 TiBase Abutments), CMX PU (14 Conical Implants-14 Universal Abutment) and CMX U (14 Tapered Implants-14 UCLA Abutments). Assemblies had their interface submerged in saliva as the contaminant. Samples were subjected either to thermomechanical cycling (2 × 10 mechanical cycles with frequency of 5 Hz and load of 120 N simultaneously with thermal cycles of 5-55 °C) or thermal cycling (5-55 °C). After cycling, the contents from the inner parts of assemblies were collected and analyzed using the Checkerboard DNA-DNA hybridization technique. Significant differences in the total genome counts were found after both thermomechanical or thermal cycling: CMX U > CMX PU > CMX TB > CMC TB. There were also significant differences in individual bacterial counts in each of the groups ( < 0.05). Irrespective of mechanical cycling, the type of abutment seems to influence not only the total microbial leakage through the interface, but also seems to significantly reflect differences considering individual target species.