We have previously reported a physiologically relevant interaction between KCNQ1 (Q1) and KCNH2 (H2). While the H2 C-terminus has been suggested to play a role, so far, no more detailed information regarding the interaction site is available. The methods used in the study are cell culture, PCR for mutagenesis, patch clamp for ion current recordings, co-immunoprecipitation for determination of protein interaction. Co-expression of Q1 and H2 resulted in an increase of I H2 (tails after +50 mV; Q1 + H2, 36 ± 6 pA/pF; H2, 14 ± 2 pA/pF; n = 10; 12; P < 0.05). Upon expressing a non-conductive (dominant-negative) Q1-pore mutation (dnQ1), there was still an increase in I H2 (tails after +50 mV; H2 + dnQ1, 24 ± 4 pA/pF; n = 10; P < 0.05) making the pore region unlikely as an interaction site. Experiments using the KCNH2-pore blocking agent quinidine supported these findings. If Q1 and H2 formed hetero-tetramers, steric changes within the pore should change the quinidine half-inhibitory concentrations (IC50). However, I H2 sensitivity did not significantly change in the presence or absence of Q1 (IC50 341 ± 63 vs. 611 ± 293 nmol/L, respectively, P = n.s.), providing further evidence that the pore is not a likely H2-Q1 interaction site. To obtain further insights into the role of intra-cytoplasmic structures, we used both C- and N-terminally truncated mutant H2 proteins. Both H2 mutants co-immunoprecipitated with Q1, suggesting no specific role of C- or N-termini. Accordingly, rather than these, the transmembrane domains of the α-subunits appear relevant for the interaction. Our results largely exclude the formation of hetero-tetramers between H2 and Q1 comprising the pore region or H2 C- or N-termini.