The function of membrane proteins is inextricably linked to the proper packing and assembly of their independently helical transmembrane (TM) segments. Here we examined whether an externally added TM peptide analogue could specifically inhibit the function of the membrane protein from which it is derived by competing for native TM helix packing sites, thereby producing a non-functional peptide-protein complex. This hypothesis was tested using Lys-tagged peptides synthesized with sequences corresponding to the three TM segments of the homotrimeric Escherichia coli diacylglycerol kinase (DGK). The peptide corresponding to wild-type DGK TM-2 inhibited the protein's enzymatic activity in a dose-dependent manner through formation of an inactive pseudo-complex, whereas peptides derived from TM-1 and TM-3 were benign toward DGK structure/function. Also, substitution of a conserved residue (Glu-69) within the TM-2 peptide abolished these effects, demonstrating the strict sequence requirements for TM-2-mediated association. This strategy, coupled with the practical advantages of the water solubility of Lys-tagged TM peptides, may constitute an attractive approach for the design of therapeutic membrane protein modulators even in the absence of a high resolution structure.The folding and oligomerization of integral membrane proteins can be divided into two energetically distinct steps (1). In this "two-stage" model, transmembrane (TM) 1 segments first fold into stable ␣-helices in the lipid bilayer. Upon establishment of the proper secondary structure elements, the nativelike spatial arrangement of side chains facilitates high affinity helix-helix association (2-5) mediated primarily by van der Waals packing and inter-helical H-bonding. The specificity of these interactions is crucial for proper membrane protein folding, as emphasized by the strict sequence requirements of residues at the helical interfaces of self-associating TM segments from such proteins as glycophorin A (6, 7), the influenza A M2 proton channel (8), phospholamban (9), synatobrevin (10), and the major coat protein from M13 bacteriophage (3,11,12).Here, we hypothesize that synthetic peptides composed of the TM segment sequences from a membrane protein target can specifically prevent proper protein folding by competing for native helix-helix packing sites. Conceptually analogous studies performed on the single-spanning dimeric glycophorin A (13) and two multi-spanning G-protein coupled receptors (14, 15) illustrated the ability of synthetic peptides to disrupt transmembrane helix-helix interactions. Furthermore, we sought to inquire whether appropriately designed TM segment peptides derived from a multi-spanning membrane protein could not only interact with the intact protein but also inhibit its function.The practical use of TM peptides in this context has been limited by the challenges associated with purifying and characterizing these inherently hydrophobic species. However, previous work in our laboratory has established the fact that flanking TM segm...