Co-translational control of protein complex formation: a fundamental pathway of cellular organization?

Analyses of proteomes from a large number of organisms throughout the domains of life highlight the key role played by multiprotein complexes for the implementation of cellular function. While the occurrence of multiprotein assemblies is ubiquitous, the understanding of pathways that dictate the formation of quaternary structure remains enigmatic. Interestingly, there are now well-established examples of protein complexes that are assembled co-translationally in both prokaryotes and eukaryotes, and indications are that the phenomenon is widespread in cells. Here, we review complex assembly with an emphasis on co-translational pathways, which involve interactions of nascent chains with other nascent or mature partner proteins, respectively. In prokaryotes, such interactions are promoted by the polycistronic arrangement of mRNA and the associated co-translation of functionally related cell constituents in order to enhance otherwise diffusion-dependent processes. Beyond merely stochastic events, however, co-translational complex formation may be sensitive to subunit availability and allow for overall regulation of the assembly process. We speculate how co-translational pathways may constitute integral components of quality control systems to ensure the correct and complete formation of hundreds of heterogeneous assemblies in a single cell. Coupling of folding of intrinsically disordered domains with co-translational interaction of binding partners may furthermore enhance the efficiency and fidelity with which correct conformation is attained. Co-translational complex formation may constitute a fundamental pathway of cellular organization, with profound importance for health and disease.

Butyrophilin 3A1 binds phosphorylated antigens and stimulates human γδ T cells.

Human T cells that express a T cell antigen receptor (TCR) containing γ-chain variable region 9 and δ-chain variable region 2 (Vγ9Vδ2) recognize phosphorylated prenyl metabolites as antigens in the presence of antigen-presenting cells but independently of major histocompatibility complex (MHC), the MHC class I-related molecule MR1 and antigen-presenting CD1 molecules. Here we used genetic approaches to identify the molecule that binds and presents phosphorylated antigens. We found that the butyrophilin BTN3A1 bound phosphorylated antigens with low affinity, at a stoichiometry of 1:1, and stimulated mouse T cells with transgenic expression of a human Vγ9Vδ2 TCR. The structures of the BTN3A1 distal domain in complex with host- or microbe-derived phosphorylated antigens had an immunoglobulin-like fold in which the antigens bound in a shallow pocket. Soluble Vγ9Vδ2 TCR interacted specifically with BTN3A1-antigen complexes. Accordingly, BTN3A1 represents an antigen-presenting molecule required for the activation of Vγ9Vδ2 T cells.

The structural basis for autonomous dimerization of the pre-T-cell antigen receptor.

The pre-T-cell antigen receptor (pre-TCR), expressed by immature thymocytes, has a pivotal role in early T-cell development, including TCR ß-selection, survival and proliferation of CD4(-)CD8(-) double-negative thymocytes, and subsequent αß T-cell lineage differentiation. Whereas αßTCR ligation by the peptide-loaded major histocompatibility complex initiates T-cell signalling, pre-TCR-induced signalling occurs by means of a ligand-independent dimerization event. The pre-TCR comprises an invariant α-chain (pre-Tα) that pairs with any TCR ß-chain (TCRß) following successful TCR ß-gene rearrangement. Here we provide the basis of pre-Tα-TCRß assembly and pre-TCR dimerization. The pre-Tα chain comprised a single immunoglobulin-like domain that is structurally distinct from the constant (C) domain of the TCR α-chain; nevertheless, the mode of association between pre-Tα and TCRß mirrored that mediated by the Cα-Cß domains of the αßTCR. The pre-TCR had a propensity to dimerize in solution, and the molecular envelope of the pre-TCR dimer correlated well with the observed head-to-tail pre-TCR dimer. This mode of pre-TCR dimerization enabled the pre-Tα domain to interact with the variable (V) ß domain through residues that are highly conserved across the Vß and joining (J) ß gene families, thus mimicking the interactions at the core of the αßTCR's Vα-Vß interface. Disruption of this pre-Tα-Vß dimer interface abrogated pre-TCR dimerization in solution and impaired pre-TCR expression on the cell surface. Accordingly, we provide a mechanism of pre-TCR self-association that allows the pre-Tα chain to simultaneously 'sample' the correct folding of both the V and C domains of any TCR ß-chain, regardless of its ultimate specificity, which represents a critical checkpoint in T-cell development. This unusual dual-chaperone-like sensing function of pre-Tα represents a unique mechanism in nature whereby developmental quality control regulates the expression and signalling of an integral membrane receptor complex.

Dissecting specificity in the Janus kinases: the structures of JAK-specific inhibitors complexed to the JAK1 and JAK2 protein tyrosine kinase domains.

The Janus kinases (JAKs) are a pivotal family of protein tyrosine kinases (PTKs) that play prominent roles in numerous cytokine signaling pathways, with aberrant JAK activity associated with a variety of hematopoietic malignancies, cardiovascular diseases and immune-related disorders. Whereas the structures of the JAK2 and JAK3 PTK domains have been determined, the structure of the JAK1 PTK domain is unknown. Here, we report the high-resolution crystal structures of the "active form" of the JAK1 PTK domain in complex with two JAK inhibitors, a tetracyclic pyridone 2-t-butyl-9-fluoro-3,6-dihydro-7H-benz[h]-imidaz[4,5-f]isoquinoline-7-one (CMP6) and (3R,4R)-3-[4-methyl-3-[N-methyl-N-(7H-pyrrolo[2,3-d]pyrimidin-4-yl)amino]piperidin-1-yl]-3-oxopropionitrile (CP-690,550), and compare them with the corresponding JAK2 PTK inhibitor complexes. Both inhibitors bound in a similar manner to JAK1, namely buried deep within a constricted ATP-binding site, thereby providing a basis for the potent inhibition of JAK1. As expected, the mode of inhibitor binding in JAK1 was very similar to that observed in JAK2, highlighting the challenges in developing JAK-specific inhibitors that target the ATP-binding site. Nevertheless, differences surrounding the JAK1 and JAK2 ATP-binding sites were apparent, thereby providing a platform for the rational design of JAK2- and JAK1-specific inhibitors.

Crystal structures of the Lyn protein tyrosine kinase domain in its Apo- and inhibitor-bound state.

The Src-family protein-tyrosine kinase (PTK) Lyn is the most important Src-family kinase in B cells, having both inhibitory and stimulatory activity that is dependent on the receptor, ligand, and developmental context of the B cell. An important role for Lyn has been reported in acute myeloid leukemia and chronic myeloid leukemia, as well as certain solid tumors. Although several Src-family inhibitors are available, the development of Lyn-specific inhibitors, or inhibitors with reduced off-target activity to Lyn, has been hampered by the lack of structural data on the Lyn kinase. Here we report the crystal structure of the non-liganded form of Lyn kinase domain, as well as in complex with three different inhibitors: the ATP analogue AMP-PNP; the pan Src kinase inhibitor PP2; and the BCR-Abl/Src-family inhibitor Dasatinib. The Lyn kinase domain was determined in its "active" conformation, but in the unphosphorylated state. All three inhibitors are bound at the ATP-binding site, with PP2 and Dasatinib extending into a hydrophobic pocket deep in the substrate cleft, thereby providing a basis for the Src-specific inhibition. Analysis of sequence and structural differences around the active site region of the Src-family PTKs were evident. Accordingly, our data provide valuable information for the further development of therapeutics targeting Lyn and the important Src-family of kinases.

Relationship between mitochondrial DNA damage and photoreceptor death in developing and adult retina, assessed in normal and degenerative rat strains.

In this study, we used Real-Time PCR to study the correlation of mtDNA deletions and photoreceptor death by apoptosis in one normal (SD) and two different degenerative (RCS and P23H) rat strains. Our results show that, in the SD and RCS strains, mtDNA deletion frequency increased and fell during neonatal life, correlating with rates of photoreceptor death during the critical period of photoreceptor development, and into adulthood. Results suggest that mitochondrial damage occurs in close association with photoreceptor death, in the normal (SD) and fast degenerative (RCS) retinas. The lack of a similar association was observed in the slowly degenerative P23H-3 strain.

The unstructured C-terminus of the tau subunit of Escherichia coli DNA polymerase III holoenzyme is the site of interaction with the alpha subunit.

The tau subunit of Escherichia coli DNA polymerase III holoenzyme interacts with the alpha subunit through its C-terminal Domain V, tau(C)16. We show that the extreme C-terminal region of tau(C)16 constitutes the site of interaction with alpha. The tau(C)16 domain, but not a derivative of it with a C-terminal deletion of seven residues (tau(C)16Delta7), forms an isolable complex with alpha. Surface plasmon resonance measurements were used to determine the dissociation constant (K(D)) of the alpha-tau(C)16 complex to be approximately 260 pM. Competition with immobilized tau(C)16 by tau(C)16 derivatives for binding to alpha gave values of K(D) of 7 muM for the alpha-tau(C)16Delta7 complex. Low-level expression of the genes encoding tau(C)16 and tau(C)16triangle up7, but not tau(C)16Delta11, is lethal to E. coli. Suppression of this lethal phenotype enabled selection of mutations in the 3' end of the tau(C)16 gene, that led to defects in alpha binding. The data suggest that the unstructured C-terminus of tau becomes folded into a helix-loop-helix in its complex with alpha. An N-terminally extended construct, tau(C)24, was found to bind DNA in a salt-sensitive manner while no binding was observed for tau(C)16, suggesting that the processivity switch of the replisome functionally involves Domain IV of tau.

The 2.7 A crystal structure of the autoinhibited human c-Fms kinase domain.

c-Fms, a member of the Platelet-derived Growth Factor (PDGF) receptor family of receptor tyrosine kinases (RTKs), is the receptor for macrophage colony stimulating factor (CSF-1) that regulates proliferation, differentiation and survival of cells of the mononuclear phagocyte lineage. Abnormal expression of c-fms proto-oncogene is associated with a significant number of human pathologies, including a variety of cancers and rheumatoid arthritis. Accordingly, c-Fms represents an attractive therapeutic target. To further understand the regulation of c-Fms, we determined the 2.7 A resolution crystal structure of the cytosolic domain of c-Fms that comprised the kinase domain and the juxtamembrane domain. The structure reveals the crucial inhibitory role of the juxtamembrane domain (JM) that binds to a hydrophobic site immediately adjacent to the ATP binding pocket. This interaction prevents the activation loop from adopting an active conformation thereby locking the c-Fms kinase into an autoinhibited state. As observed for other members of the PDGF receptor family, namely c-Kit and Flt3, three JM-derived tyrosine residues primarily drive the mechanism for autoinhibition in c-Fms, therefore defining a common autoinhibitory mechanism within this family. Moreover the structure provides an understanding of c-Fms inhibition by Gleevec as well as providing a platform for the development of more selective inhibitors that target the inactive conformation of c-Fms kinase.

Multiple oligomeric forms of Escherichia coli DnaB helicase revealed by electrospray ionisation mass spectrometry.

The Escherichia coli DnaB protein (DnaB(6)) is the hexameric helicase that unwinds genomic DNA so it can be copied by the DNA replication machinery. Loading of the helicase onto DNA requires interactions of DnaB(6) with six molecules of its loading partner protein, DnaC. Nano-electrospray ionisation mass spectrometry (nanoESI-MS) of mutant proteins was used to examine the roles of the residues Phe102 (F102) and Asp82 (D82) in the N-terminal domain of DnaB in the assembly of the hexamer. When the proteins were prepared in 1 M ammonium acetate containing magnesium and adenosine triphosphate (ATP) at pH 7.6, both hexameric and heptameric forms of wild-type and F102W, F102E and D82N mutant DnaBs were observed in mass spectra. The spectra of the D82N mutant also showed substantial amounts of a decameric species and small amounts of a dodecamer. In contrast, the F102H DnaB mutant was incapable of forming oligomers of order higher than the hexamer. Thus, although Phe102 is not the only determinant of hexamer assembly, this residue has a role in oligomerisation. NanoESI mass spectra were obtained of mixtures of DnaB(6) with DnaC. The DnaB(6)(DnaC)(6) complex (calculated M(r) 481 164) was observed only when the two proteins were present in equimolar amounts. The data are consistent with cooperative assembly of the complex. ESI mass spectra of mixtures containing DnaC and ATP showed that DnaC slowly hydrolysed ATP to ADP as indicated by ions corresponding to DnaC/ATP and DnaC/ADP complexes. These experiments show that E. coli DnaB can form a heptameric complex and that nanoESI-MS can be used to probe assembly of large (>0.5 MDa) macromolecular complexes.

Stabilization of native protein fold by intein-mediated covalent cyclization.

A mutant version of the N-terminal domain of Escherichia coli DnaB helicase was used as a model system to assess the stabilization against unfolding gained by covalent cyclization. Cyclization was achieved in vivo by formation of an amide bond between the N and C termini with the help of a split mini-intein. Linear and circular proteins were constructed to be identical in amino acid sequence. Mutagenesis of Phe102 to Glu rendered the protein monomeric even at high concentration. A difference in free energy of unfolding, DeltaDeltaG, between circular and linear protein of 2.3(+/-0.5) kcal mol(-1) was measured at 10 degrees C by circular dichroism. A theoretical estimate of the difference in conformational entropy of linear and circular random chains in a three-dimensional cubic lattice model predicted DeltaDeltaG=2.3 kcal mol(-1), suggesting that stabilization by protein cyclization is driven by the reduced conformational entropy of the unfolded state. Amide-proton exchange rates measured by NMR spectroscopy and mass spectrometry showed a uniform, approximately tenfold decrease of the exchange rates of the most slowly exchanging amide protons, demonstrating that cyclization globally decreases the unfolding rate of the protein. The amide proton exchange was found to follow EX1 kinetics at near-neutral pH, in agreement with an unusually slow refolding rate of less than 4 min(-1) measured by stopped-flow circular dichroism. The linear and circular proteins differed more in their unfolding than in their folding rates. Global unfolding of the N-terminal domain of E.coli DnaB is thus promoted strongly by spatial separation of the N and C termini, whereas their proximity is much less important for folding.

Comparison of gene expression between left atria and left ventricles from non-diseased humans.

We examine the reliability and accuracy of gene array technology in analyzing differences in gene expression between human non-diseased left atrium and left ventricle. We have used cDNA gene arrays and validated those data by carefully designed quantitative real-time polymerase chain reaction (PCR). We have identified pitfalls using cDNA gene array technology based on comparisons with other gene array studies and with changes reported for the levels of expression of the genes corresponding to these cDNAs. The high error rate reported here underscores the cautionary comments reported by others in this field.

In vivo protein cyclization promoted by a circularly permuted Synechocystis sp. PCC6803 DnaB mini-intein.

A synthetic Synechocystis sp. PCC6803 DnaB split mini-intein gene was constructed for the in vivo cyclization of recombinant proteins expressed in Escherichia coli. The system was used to cyclize the NH(2)-terminal domain of E. coli DnaB, the structure of which had been determined previously by NMR spectroscopy. Cyclization was found to proceed efficiently, with little accumulation of precursor, and the product was purified in high yield. The solution structure of cyclic DnaB-N is not significantly different from that of linear DnaB-N and it unfolds reversibly at temperatures approximately 14 degrees C higher. Improved hydrogen bonding was observed in the first and last helices, and the length of the last helix was increased, while the 9-amino acid linker used to join the NH(2) and COOH termini was found to be highly mobile. The measured thermodynamic stabilization of the structure (Delta Delta G approximately 2 kcal/mol) agrees well with the value estimated from the reduced conformational entropy in the unfolded form. Simple polymer theory can be used to predict likely free energy changes resulting from protein cyclization and how the stabilization depends on the size of the protein and the length of the linker used to connect the termini.