Transcranial direct current stimulation (tDCS) has been used as a useful interventional brain stimulation technique to improve unilateral upper-limb motor function in healthy humans, as well as in stroke patients. Although tDCS applications are supposed to modify the interhemispheric balance between the motor cortices, the tDCS after-effects on interhemispheric interactions are still poorly understood. To address this issue, we investigated the tDCS after-effects on interhemispheric inhibition (IHI) between the primary motor cortices (M1) in healthy humans. Three types of tDCS electrode montage were tested on separate days; anodal tDCS over the right M1, cathodal tDCS over the left M1, bilateral tDCS with anode over the right M1 and cathode over the left M1. Single-pulse and paired-pulse transcranial magnetic stimulations were given to the left M1 and right M1 before and after tDCS to assess the bilateral corticospinal excitabilities and mutual direction of IHI. Regardless of the electrode montages, corticospinal excitability was increased on the same side of anodal stimulation and decreased on the same side of cathodal stimulation. However, neither unilateral tDCS changed the corticospinal excitability at the unstimulated side. Unilateral anodal tDCS increased IHI from the facilitated side M1 to the unchanged side M1, but it did not change IHI in the other direction. Unilateral cathodal tDCS suppressed IHI both from the inhibited side M1 to the unchanged side M1 and from the unchanged side M1 to the inhibited side M1. Bilateral tDCS increased IHI from the facilitated side M1 to the inhibited side M1 and attenuated IHI in the opposite direction. Sham-tDCS affected neither corticospinal excitability nor IHI. These findings indicate that tDCS produced polarity-specific after-effects on the interhemispheric interactions between M1 and that those after-effects on interhemispheric interactions were mainly dependent on whether tDCS resulted in the facilitation or inhibition of the M1 sending interhemispheric volleys.