Recently, a number of genes have been identified that are involved in the O-linked glycosylation of α-dystroglycan (αDG)1
, an important component of the dystrophin associated complex that anchors muscle fibres to the extracellular scaffold 
, these genes include fukutin
, fukutin related protein
. Mutations in these genes have been shown to result in the aberrant glycosylation of αDG resulting in a broad spectrum of congenital muscular dystrophies 
. Sequence analysis suggests that these genes encode for Type-II membrane proteins with putative glycosyltransferase activity, an observation consistent with the hypoglycosylation of αDG detected in patients carrying mutations in any of these genes 
. Subsequent studies have revealed that these proteins are normally located within the ER/Golgi complex in agreement with their proposed role in the glycosylation of αDG [2–5]
. Interestingly, a number of the mutations associated with these genes result in the miss-localisation within the cell 
suggesting that their retention within the ER/Golgi complex is vital for the appropriate glycosylation of αDG.
The retention of proteins, including glycosyltransferases, within the ER and Golgi is a highly dynamic process reliant on the tight regulation of both antero- and retrograde transport steps 
. Sequence analysis and biochemical analyses have demonstrated that retrograde transport is largely regulated by the receptor mediated recognition of specific peptide motifs on the extra-membranous domains of ER and Golgi resident proteins 
. In contrast, regulation of anterograde transport appears to be dependent on the shorter N-terminal transmembrane domain (TMD) that is typically found in ER and Golgi resident proteins 
. It has been proposed that this interaction between the shortened N-terminal TMD and the atypical lipid composition found within the ER and Golgi complex plays an important role in regulating anterograde trafficking [7,8]
. Indeed, in the context of muscular dystrophy, the N-terminal transmembrane domains of the proteins encoded by fukutin
and fukutin related protein
have been shown to be sufficient to retain the protein within the Golgi complex 
Although the role of the N-terminal TMD in the retention of these proteins is now known, a molecular understanding of this process remains to be elucidated. Several models based on lipid-mediated sorting and protein oligomerisation have been proposed [6,8,10,11]
. To understand at a molecular level how lipids regulate protein trafficking, we are studying the transmembrane domain of the putative glycosyltransferase linked to Fukuyama muscular dystrophy encoded by the gene fukutin
. In keeping with this family of proteins, the transmembrane domain of Fukutin has been demonstrated to be sufficient for the targeting of the protein to the Golgi apparatus 
. Using a combination of liquid, solid-state NMR methods and other biophysical techniques, we are investigating how the differences in lipid bilayer composition affect the structure, oligomeric state and lateral segregation of Fukutin’s transmembrane domain.
A pre-requisite to these studies is the introduction of NMR sensitive isotopes into the protein of interest. Until recently, solid-state NMR studies of transmembrane peptides and proteins have relied on the introduction of labels site-selectively using solid-phase peptide synthesis. This technique permits the preparation of peptides up to 50 residues in length in milligram quantities sufficient for biophysical characterisation. However, recent advances in both liquid and solid-state NMR methodology rely on the uniform incorporation of NMR sensitive isotopes within the protein or peptides, a process that is costly when using solid-phase peptide synthesis, as uniformly labelled amino acids are required.
To exploit these advances in NMR methodology in the study of how lipids regulate Fukutin-1 trafficking, we have sought to develop an efficient bacterial expression system which permits both uniform and selective and extensive labelling of the transmembrane domain of Fukutin-1 thereby avoiding the synthesis of uniformly labelled peptide [12,13]
. The expression and purification of transmembrane peptides and proteins is notoriously difficult due to their hydrophobicity and potential toxicity to the host cell 
. To overcome these difficulties a number of groups have utilized carrier proteins to aid solubility, reduce toxicity and in some cases target the protein/peptide to the bacterial membranes 
. However, it is our experience that after purification the yields obtained can be low as the desired peptides represent only a small fraction of the overall fusion protein and significant losses can be incurred during purification. Here we report on an efficient expression and purification method for the transmembrane domain of Fukutin-1 using simply a His6
tag for purification. Even in the absence of a classical bacterial membrane protein targeting sequence, the peptide is effectively targeted to the bacterial membrane presumably due to the intrinsic sequence and is expressed in quantities sufficient to support biophysical characterisation and introduction of isotopes for subsequent NMR studies.