Persistent Activation of mRNA Translation by Transient Hsp90 Inhibition.

The heat shock protein 90 (Hsp90) chaperone functions as a protein-folding buffer and plays a role promoting the evolution of new heritable traits. To better understand how Hsp90 can affect mRNA translation, we screen more than 1,600 factors involved in mRNA regulation for physical interactions with Hsp90 in human cells. The mRNA binding protein CPEB2 strongly binds Hsp90 via its prion domain. In a yeast model, transient inhibition of Hsp90 results in persistent activation of a CPEB translation reporter even in the absence of exogenous CPEB that persists for 30 generations after the inhibitor is removed. Ribosomal profiling reveals that some endogenous yeast mRNAs, including HAC1, show a persistent change in translation efficiency following transient Hsp90 inhibition. Thus, transient loss of Hsp90 function can promote a nongenetic inheritance of a translational state affecting specific mRNAs, introducing a mechanism by which Hsp90 can promote phenotypic variation.

Protein-only mechanism induces self-perpetuating changes in the activity of neuronal Aplysia cytoplasmic polyadenylation element binding protein (CPEB).

Neuronal cytoplasmic polyadenylation element binding protein (CPEB) plays a critical role in maintaining the functional and morphological long-lasting synaptic changes that underlie learning and memory. It can undergo a prion switch, but it remains unclear if this self-templating change in protein conformation is alone sufficient to create a stable change in CPEB activity: a robust "protein-only" biochemical memory. To investigate, we take advantage of yeast cells wherein the neuronal CPEB of Aplysia is expressed in the absence of any neuronal factors and can stably adopt either an active or an inactive state. Reminiscent of well-characterized yeast prions, we find that CPEB can adopt several distinct activity states or "strains." These states are acquired at a much higher spontaneous rate than is typical of yeast prions, but they are extremely stable--perpetuating for years--and have all of the non-Mendelian genetic characteristics of bona fide yeast prions. CPEB levels are too low to allow direct physical characterization, but CPEB strains convert a fusion protein, which shares only the prion-like domain of CPEB, into amyloid in a strain-specific manner. Lysates of CPEB strains seed the purified prion domain to adopt the amyloid conformation with strain-specific efficiencies. Amyloid conformers generated by spontaneous assembly of the purified prion domain (and a more biochemically tractable derivative) transformed cells with inactive CPEB into the full range of distinct CPEB strains. Thus, CPEB employs a prion mechanism to create stable, finely tuned self-perpetuating biochemical memories. These biochemical memories might be used in the local homeostatic maintenance of long-term learning-related changes in synaptic morphology and function.

Architecture of the ribosome-channel complex derived from native membranes.

The mammalian Sec61 complex forms a protein translocation channel whose function depends upon its interaction with the ribosome and with membrane proteins of the endoplasmic reticulum (ER). To study these interactions, we determined structures of "native" ribosome-channel complexes derived from ER membranes. We find that the ribosome is linked to the channel by seven connections, but the junction may still provide a path for domains of nascent membrane proteins to move into the cytoplasm. In addition, the native channel is significantly larger than a channel formed by the Sec61 complex, due to the presence of a second membrane protein. We identified this component as TRAP, the translocon-associated protein complex. TRAP interacts with Sec61 through its transmembrane domain and has a prominent lumenal domain. The presence of TRAP in the native channel indicates that it may play a general role in translocation. Crystal structures of two Sec61 homologues were used to model the channel. This analysis indicates that there are four Sec61 complexes and two TRAP molecules in each native channel. Thus, we suggest that a single Sec61 complex may form a conduit for translocating polypeptides, while three copies of Sec61 play a structural role or recruit accessory factors such as TRAP.

Membrane-protein integration and the role of the translocation channel.

Most eukaryotic membrane proteins are integrated into the lipid bilayer during their synthesis at the endoplasmic reticulum (ER). Their integration occurs with the help of a protein-conducting channel formed by the heterotrimeric Sec61 membrane-protein complex. The crystal structure of an archaeal homolog of the complex suggests mechanisms that enable the channel to open across the membrane and to release laterally hydrophobic transmembrane segments of nascent membrane proteins into lipid. Many aspects of membrane-protein integration remain controversial and poorly understood, but new structural data provide testable hypotheses. We propose a model of how the channel recognizes transmembrane segments, orients them properly with respect to the plane of the membrane and releases them into lipid. We also discuss how the channel would prevent small molecules from crossing the lipid bilayer while it is integrating proteins.

Cooperation of transmembrane segments during the integration of a double-spanning protein into the ER membrane.

While membrane insertion of single-spanning membrane proteins into the endoplasmic reticulum (ER) is relatively well understood, it is unclear how multi-spanning proteins integrate. We have investigated the cotranslational ER integration of a double-spanning protein that is derived from leader peptidase. Both transmembrane (TM) segments are inserted into the membrane by the Sec61 channel. While the first, long and hydrophobic TM segment (TM1) inserts into the lipid bilayer on its own, the second, shorter TM anchor (TM2) collaborates with TM1 during its integration. TM1 diffuses away from the Sec61 complex in the absence of TM2, but is close to Sec61 when TM2 arrives inside the channel. These data suggest that the exit of a weak TM segment from the Sec61 channel into the lipid phase can be facilitated by its interaction with a previously integrated strong and stabilizing TM anchor.