Membrane spanning sequences

As in the Ca -ATPase of sarcoplasmic reticulum, the predicted number of membrane spanning sequences in PMCAl and PMCA2 is even, with both N- and C-terminus located on the cytoplasmic side, but their actual number is uncertain and may be 10 or less [30]. 

Human TNF-a is initially synthesized as a 233 amino acid polypeptide that is anchored in the plasma membrane by a single membrane-spanning sequence. This TNF pro-peptide, which itself displays biological activity, is usually proteolytically processed by a specific extracellular metallo-protease. Proteolytic cleavage occurs between residues 76 (Ala) and 77 (Val), yielding the mature (soluble) 157 amino acid TNF-a polypeptide. Mature human TNF-a appears to be devoid of a carbohydrate component, and contains a single disulfide bond. 

G-protein linked receptors, sometimes referred to as seven membrane spanning serpentine receptors include, as their name suggests, seven membrane-spanning sequences. Sites of receptor/G-protein interaction are within cytoplasmic and possibly membrane spanning sequences. Recent evidence indicates that die carboiqrl terminal portion of the third intracellular loop and the carbojqrl terminal portion of the receptor are the likely sites of interaction.(Bimbaumer, Bimbaumer, 1995). 

Figure 9.8 MDRl gene product P-glycoprotein. The protein has 12 membrane spanning sequences both ends of the protein are localized at the interior surface of the membrane. NBD, nucleotide-binding domain TMD, transmembrane domain (from Ref. [37]).

As pointed out by Majerus (1992), none of these three PI-PLC isoforms contains a membrane-spanning sequence. Since the PI substrates for these PI-PLC s are in the membrane bilayers, PI-PLC must bind to membranes before hydrolyzing Pis. The translocation of PI-PLC-71 (induced by its tyrosine phosphorylation) from the cytosol to the cellular membrane (Rhee and Choi, 1992a) may be an example of such a binding. 

All K channels are tetrameric molecules. There are two closely related varieties of subunits for K channels, those containing two membrane-spanning helices and those containing six. However, residues that build up the ion channel. Including the pore helix and the inner helix, show a strong sequence similarity among all K+ channels. Consequently, the structural features and the mechanism for ion selectivity and conductance described for the bacterial K+ channel in all probability also apply for K+ channels in plant and animal cells. 

The L and the M subunits are firmly anchored in the membrane, each by five hydrophobic transmembrane a helices (yellow and red, respectively, in Figure 12.14). The structures of the L and M subunits are quite similar as expected from their sequence similarity they differ only in some of the loop regions. These loops, which connect the membrane-spanning helices, form rather flat hydrophilic regions on either side of the membrane to provide interaction areas with the H subunit (green in Figure 12.14) on the cytoplasmic side and with the cytochrome (blue in Figure 12.14) on the periplasmic side. The H subunit, in addition, has one transmembrane a helix at the car-boxy terminus of its polypeptide chain. The carboxy end of this chain is therefore on the same side of the membrane as the cytochrome. In total, eleven transmembrane a helices attach the L, M, and H subunits to the membrane. 

The removal of released DA from the synaptic extracellular space to facilitate its intraneuronal metabolism is achieved by a membrane transporter that controls the synaptic concentration. This transporter has been shown to be a 619 amino-acid protein with 12 hydrophobic membrane spanning domains (see Giros and Caron 1993). Although it has similar amino-acid sequences to that of the NA (and GABA) transporter, there are sufficient differences for it to show some specificity. Thus DA terminals will not concentrate NA and the DA transporter is blocked by a drug such as nomifensine which has less effect on NA uptake. Despite this selectivity some compounds, e.g. amphetamine and 6-OHDA (but not MPTP), can be taken up by both neurons. The role of blocking DA uptake in the central actions of cocaine and amphetamine is considered later (Chapter 23). 

Fig. 1. Amino acid sequence homology between the neonatal fast-twitch and slow-twitch skeletal muscle forms of the Ca -ATPase. The sequence of the slow Ca -ATPase is shown above the neonatal fast-twitch form, with nonhomologous amino acids indicated by asterisks. The sequence of the slow ATPase is shifted to the right by one residue at residue 505 to allow realignment after the difference in sequence length. Ml-MlO, membrane spanning regions S1-S5, stalk sectors Tl, T2, major tryptic cleavage sites P,...

Aligned sequences of 16 members of the sugar transporter family. Residues which are identical in 5=50% of the 16 sugar-transporter sequences (excluding the quinate transporter (qa-y), the citrate transporter (CIT), the tetracycline transporter (pBR322) and lac permease (LacY)) are highlighted, and recorded below the sequences as CONSERVED . The locations of predicted membrane-spanning helices are indicated by horizontal bars. The sequences were taken from the references cited in the text. 

Fig. 3. (A) Model of the proposed pore forming part of K channel subunits. Segments S5 and S6 are possibly membrane-spanning helices. The helices are connected by a hydrophobic segment H5 which may be tucked into the lipid bilayer [48]. H5 is flanked by two proline residues P. Adjacent to these proline residues are amino acid side chains ( ) important for external TEA binding [45,46]. Approximately halfway between these two proline residues are amino acid side chains ( ) affecting internal TEA binding [46,47] and K channel selectivity [48]. (B) Mutations are indicated which affect in Shaker channels external TEA (TEAe) or internal TEA (TEA,) binding. Concentrations of TEA for half block of the wild-type and mutant K channels are given at the right-hand side of the corresponding sequence. Data have been compiled from [45-47].

---