The construction of biological membranes

Abstract

The Nobel Prize in physiology and medicine was in 1999 given for experiments allowing in vitro analysis of complicated biological processes – in particular for work on how proteins cross and become integrated into membrane/lipid bilayers. This work started with the formulation of the ‘signal hypothesis’ in the early 1970s. It stated that proteins have a signal that guides them to and determines whether they are completely translocated across the endoplasmic reti-culum (ER) membrane (secretory proteins) or inserted into the ER membrane (type I, II and multispan transmembrane proteins). Since lipids are hydrophobic and most proteins are hydrophilic, it was proposed that the ER membranes contain ‘translocons’ that provide an aqueous channel for the passage of proteins. Such translocons have been isolated; the major components in mammals of these channels are called Sec61p and Tram. However, while this answered how proteins cross the ER, information on how proteins such as receptors become inserted into the hydrophobic membranes remained obscure. Further progress was made possible by an elegant refinement of the in vitro system, leading to the formulation of the signal hypothesis. Indeed, one would like to probe whether the protein, during its insertion into the membrane, is in contact with other proteins (Sec61/Tram) or with lipids; to address this question, site-specific photocrosslinking was developed and used1xThe protein-conducting channel in the membrane of the endoplasmic reticulum is open laterally toward the lipid bilayer. Martoglio, B. et al. Cell. 1995; 81: 207–214Abstract | Full Text PDF | PubMed | Scopus (218)See all References1: by introducing a stop-codon at a desired position in the membrane protein that can be suppressed by tRNA charged with a photoactivatable alanine, such experiments became possible.Using site-specific photocrosslinking, Heinrich et al.2xThe Sec61p complex mediates the integration of a membrane protein by allowing lipid partitioning of the transmembrane domain. Heinrich, S.U. et al. Cell. 2000; 102: 233–244Abstract | Full Text | Full Text PDF | PubMedSee all References2 studied how a single-span transmembrane (TM) protein is inserted into lipid bilayers. The experiments suggest that the TM segment, immediately upon insertion into the translocon, passively partitions into the lipid membrane while the protein chain is still undergoing elongation and is attached to the ribosome; in support of a lipid-partitioning process is the finding that insertion of one or two positive charges into the TM segment significantly lowered the amount of protein that partitioned into the membrane. It is suggested that the changed behaviour in the presence of charges might be used to insert multispan TM proteins properly into the membrane. The experiments also suggest that the hydrophilic translocon is a dynamic structure that is open towards the membrane bilayer. Sec61p did not only appear as a major component of the translocon but also served to shield the hydrophobic TM segment from the polar head groups of the lipid bilayer; thus, if detergent miscelles without polar head groups were used, direct interaction between the TM segment and lipids was observed – albeit not leading to membrane insertion.Our understanding of how multispanning TM proteins become inserted into membranes relies on approaches using indirect photo- crosslinking and gel analysis. Although these techniques have resulted in tremendous advances, it would be very interesting if it were possible to visualize the process directly because we all know that: ‘I believe it, if I see it.’ Considering the sophisticated in vitro techniques already applied in this area, such direct experiments should become possible in the not too distant future.