The KtrAB system of Vibrio alginolyticus is a sodium-dependent potassium transport system. KtrB, the membrane integral and K+ translocating subunit of the KtrAB complex, belongs to a superfamily of K+ transporter (SKT). These proteins are likely to have evolved from simple K+ channels of the M1PM2 type like KcsA by multiple gene duplication and gene fusion. They share a so called fourfold M1PM2-motif, in which two transmembrane helices (M1 and M2) are connected by a p-loop (P), which folds half back into the membrane. Comparing members of this superfamily with the K+ channel KcsA for structural predictions a striking amino acid sequence in helix M2C was found. In VaKtrB the first part of this helix, M2C1, consists of 12 hydrophobic amino acids and is expected to form an α-helix. The following very flexible and hydrophilic part, M2C2, with many glycines and small, partly polar amino acids is not supposed to have a helical conformation. By contrast, the last part, M2C3, shows a partial amphipathic and α-helical character, followed by three positive charged amino acids (R341, K343, K344) which are consistent with the "positive inside rule" and should be localized in the cytoplasm. Due to these findings Durell and Guy in 1999 hypothesised two possible folding models for segments PC and M2C but till now the conformation of this part remains unclear. In this thesis the role of the M2C region was studied in more detail. Point and partial to complete deletions in M2C2 led to a huge increase in Vmax for K+ transport while the affinity for potassium and the sodium transport properties were unaffected. Together with some PhoA-fusion studies which indicated that M2C2 forms a flexible structure within the membrane these data were interpreted to mean that M2C2 forms a flexible gate controlling K+ translocation at the cytoplasmic side of KtrB. This hypothesis was confimed by EPR measurements of single and double spin-labeled cysteine variants of KtrB. It was shown that M2C2 forms a loop inside the cavity of the protein. Upon the addition of K+ ions M2C2 residue T318R1 moved both with respect to M2B residue D222R1 and to M2C3 residue V331, but not with respect to M2C1 residue M311R1. Other residues within M2B, M2C1 and M2C3 did not move with respect to each other. With the help of a rotamer library analysis the measured distances were used to propose two new models for the structure of the M2C2 gate inside the KtrB protein in a closed conformation in the absence of K+ ion and in an open conformation in the presence of K+ ions. Since a flexible gate like M2C2 is missing in potassium channels, it is interpreted to be a transporter-specific structure. In the context of the analysis of the role of M2C2 in purified and reconstituted KtrB by biochemical and biophysical approaches a protocol for the overproduction, purification and reconstitution of natively folded, active protein was developed. In addition, results obtained from static light scattering measurements are shown in order to gain information about the oligomeric state of single subunits as well as of the assembled KtrAB complex.