A sensor complements the steric gate when DNA polymerase ϵ discriminates ribonucleotides.

The cellular imbalance between high concentrations of ribonucleotides (NTPs) and low concentrations of deoxyribonucleotides (dNTPs), is challenging for DNA polymerases when building DNA from dNTPs. It is currently believed that DNA polymerases discriminate against NTPs through a steric gate model involving a clash between a tyrosine and the 2'-hydroxyl of the ribonucleotide in the polymerase active site in B-family DNA polymerases. With the help of crystal structures of a B-family polymerase with a UTP or CTP in the active site, molecular dynamics simulations, biochemical assays and yeast genetics, we have identified a mechanism by which the finger domain of the polymerase sense NTPs in the polymerase active site. In contrast to the previously proposed polar filter, our experiments suggest that the amino acid residue in the finger domain senses ribonucleotides by steric hindrance. Furthermore, our results demonstrate that the steric gate in the palm domain and the sensor in the finger domain are both important when discriminating NTPs. Structural comparisons reveal that the sensor residue is conserved among B-family polymerases and we hypothesize that a sensor in the finger domain should be considered in all types of DNA polymerases.

Enhanced polymerase activity permits efficient synthesis by cancer-associated DNA polymerase ϵ variants at low dNTP levels.

Amino acid substitutions in the exonuclease domain of DNA polymerase ϵ (Polϵ) cause ultramutated tumors. Studies in model organisms suggested pathogenic mechanisms distinct from a simple loss of exonuclease. These mechanisms remain unclear for most recurrent Polϵ mutations. Particularly, the highly prevalent V411L variant remained a long-standing puzzle with no detectable mutator effect in yeast despite the unequivocal association with ultramutation in cancers. Using purified four-subunit yeast Polϵ, we assessed the consequences of substitutions mimicking human V411L, S459F, F367S, L424V and D275V. While the effects on exonuclease activity vary widely, all common cancer-associated variants have increased DNA polymerase activity. Notably, the analog of Polϵ-V411L is among the strongest polymerases, and structural analysis suggests defective polymerase-to-exonuclease site switching. We further show that the V411L analog produces a robust mutator phenotype in strains that lack mismatch repair, indicating a high rate of replication errors. Lastly, unlike wild-type and exonuclease-dead Polϵ, hyperactive variants efficiently synthesize DNA at low dNTP concentrations. We propose that this characteristic could promote cancer cell survival and preferential participation of mutator polymerases in replication during metabolic stress. Our results support the notion that polymerase fitness, rather than low fidelity alone, is an important determinant of variant pathogenicity.

Structural evidence for an essential Fe-S cluster in the catalytic core domain of DNA polymerase ϵ.

DNA polymerase ϵ (Pol ϵ), the major leading-strand DNA polymerase in eukaryotes, has a catalytic subunit (Pol2) and three non-catalytic subunits. The N-terminal half of Pol2 (Pol2CORE) exhibits both polymerase and exonuclease activity. It has been suggested that both the non-catalytic C-terminal domain of Pol2 (with the two cysteine motifs CysA and CysB) and Pol2CORE (with the CysX cysteine motif) are likely to coordinate an Fe-S cluster. Here, we present two new crystal structures of Pol2CORE with an Fe-S cluster bound to the CysX motif, supported by an anomalous signal at that position. Furthermore we show that purified four-subunit Pol ϵ, Pol ϵ CysAMUT (C2111S/C2133S), and Pol ϵ CysBMUT (C2167S/C2181S) all have an Fe-S cluster that is not present in Pol ϵ CysXMUT (C665S/C668S). Pol ϵ CysAMUT and Pol ϵ CysBMUT behave similarly to wild-type Pol ϵ in in vitro assays, but Pol ϵ CysXMUT has severely compromised DNA polymerase activity that is not the result of an excessive exonuclease activity. Tetrad analyses show that haploid yeast strains carrying CysXMUT are inviable. In conclusion, Pol ϵ has a single Fe-S cluster bound at the base of the P-domain, and this Fe-S cluster is essential for cell viability and polymerase activity.

Structural consequence of the most frequently recurring cancer-associated substitution in DNA polymerase ε.

The most frequently recurring cancer-associated DNA polymerase ε (Pol ε) mutation is a P286R substitution in the exonuclease domain. While originally proposed to increase genome instability by disrupting exonucleolytic proofreading, the P286R variant was later found to be significantly more pathogenic than Pol ε proofreading deficiency per se. The mechanisms underlying its stronger impact remained unclear. Here we report the crystal structure of the yeast orthologue, Pol ε-P301R, complexed with DNA and an incoming dNTP. Structural changes in the protein are confined to the exonuclease domain, with R301 pointing towards the exonuclease site. Molecular dynamics simulations suggest that R301 interferes with DNA binding to the exonuclease site, an outcome not observed with the exonuclease-inactive Pol ε-D290A,E292A variant lacking the catalytic residues. These results reveal a distinct mechanism of exonuclease inactivation by the P301R substitution and a likely basis for its dramatically higher mutagenic and tumorigenic effects.

Multivalent Interactions of Human Primary Amine Oxidase with the V and C22 Domains of Sialic Acid-Binding Immunoglobulin-Like Lectin-9 Regulate Its Binding and Amine Oxidase Activity.

Sialic acid-binding immunoglobulin-like lectin-9 (Siglec-9) on leukocyte surface is a counter-receptor for endothelial cell surface adhesin, human primary amine oxidase (hAOC3), a target protein for anti-inflammatory agents. This interaction can be used to detect inflammation and cancer in vivo, since the labeled peptides derived from the second C2 domain (C22) of Siglec-9 specifically bind to the inflammation-inducible hAOC3. As limited knowledge on the interaction between Siglec-9 and hAOC3 has hampered both hAOC3-targeted drug design and in vivo imaging applications, we have now produced and purified the extracellular region of Siglec-9 (Siglec-9-EC) consisting of the V, C21 and C22 domains, modeled its 3D structure and characterized the hAOC3-Siglec-9 interactions using biophysical methods and activity/inhibition assays. Our results assign individual, previously unknown roles for the V and C22 domains. The V domain is responsible for the unusually tight Siglec-9-hAOC3 interactions whereas the intact C22 domain of Siglec-9 is required for modulating the enzymatic activity of hAOC3, crucial for the hAOC3-mediated leukocyte trafficking. By characterizing the Siglec-9-EC mutants, we could conclude that R120 in the V domain likely interacts with the terminal sialic acids of hAOC3 attached glycans whereas residues R284 and R290 in C22 are involved in the interactions with the active site channel of hAOC3. Furthermore, the C22 domain binding enhances the enzymatic activity of hAOC3 although the sialic acid-binding capacity of the V domain of Siglec-9 is abolished by the R120S mutation. To conclude, our results prove that the V and C22 domains of Siglec-9-EC interact with hAOC3 in a multifaceted and unique way, forming both glycan-mediated and direct protein-protein interactions, respectively. The reported results on the mechanism of the Siglec-9-hAOC3 interaction are valuable for the development of hAOC3-targeted therapeutics and diagnostic tools.

Dual targeted poplar ferredoxin NADP(+) oxidoreductase interacts with hemoglobin 1.

Previous reports have connected non-symbiotic and truncated hemoglobins (Hbs) to metabolism of nitric oxide (NO), an important signalling molecule involved in wood formation. We have studied the capability of poplar (Populus tremula × tremuloides) Hbs PttHb1 and PttTrHb proteins alone or with a flavin-protein reductase to relieve NO cytotoxicity in living cells. Complementation tests in a Hb-deficient, NO-sensitive yeast (Saccharomyces cerevisiae) Δyhb1 mutant showed that neither PttHb1 nor PttTrHb alone protected cells against NO. To study the ability of Hbs to interact with a reductase, ferredoxin NADP(+) oxidoreductase PtthFNR was characterized by sequencing and proteomics. To date, by far the greatest number of the known dual-targeted plant proteins are directed to chloroplasts and mitochondria. We discovered a novel variant of hFNR that lacks the plastid presequence and resides in cytosol. The coexpression of PttHb1 and PtthFNR partially restored NO resistance of the yeast Δyhb1 mutant, whereas PttTrHb coexpressed with PtthFNR failed to rescue growth. YFP fusion proteins confirmed the interaction between PttHb1 and PtthFNR in plant cells. The structural modelling results indicate that PttHb1 and PtthFNR are able to interact as NO dioxygenase. This is the first report on dual targeting of central plant enzyme FNR to plastids and cytosol.

Evolution, three-dimensional model and localization of truncated hemoglobin PttTrHb of hybrid aspen.

Thus far, research on plant hemoglobins (Hbs) has mainly concentrated on symbiotic and non-symbiotic Hbs, and information on truncated Hbs (TrHbs) is scarce. The aim of this study was to examine the origin, structure and localization of the truncated Hb (PttTrHb) of hybrid aspen (Populus tremula L. × tremuloides Michx.), the model system of tree biology. Additionally, we studied the PttTrHb expression in relation to non-symbiotic class1 Hb gene (PttHb1) using RNAi-silenced hybrid aspen lines. Both the phylogenetic analysis and the three-dimensional (3D) model of PttTrHb supported the view that plant TrHbs evolved vertically from a bacterial TrHb. The 3D model suggested that PttTrHb adopts a 2-on-2 sandwich of α-helices and has a Bacillus subtilis -like ligand-binding pocket in which E11Gln and B10Tyr form hydrogen bonds to a ligand. However, due to differences in tunnel cavity and gate residue (E7Ala), it might not show similar ligand-binding kinetics as in Bs-HbO (E7Thr). The immunolocalization showed that PttTrHb protein was present in roots, stems as well as leaves of in vitro -grown hybrid aspens. In mature organs, PttTrHb was predominantly found in the vascular bundles and specifically at the site of lateral root formation, overlapping consistently with areas of nitric oxide (NO) production in plants. Furthermore, the NO donor sodium nitroprusside treatment increased the amount of PttTrHb in stems. The observed PttTrHb localization suggests that PttTrHb plays a role in the NO metabolism.

Structural comparison of the active site channels in rodent and primate vascular adhesion protein-1.

In this study, we have made homology models of mouse, rat, and monkey vascular adhesion protein-1 (VAP-1) to reveal basis for the species-specific ligand recognition of VAP-1. Based on the structural comparisons, rodent VAP-1s have a narrower active site channel than primate VAP-1s. The variable residues in mouse and rat VAP-1, Phe447 from arm I and the polar residues from the first α-helix of the D3 domain together with C-terminal residues are likely to affect ligand recognition and binding.

Structure of collagen receptor integrin α(1)I domain carrying the activating mutation E317A.

We have analyzed the structure and function of the integrin α(1)I domain harboring a gain-of-function mutation E317A. To promote protein crystallization, a double variant with an additional C139S mutation was used. In cell adhesion assays, the E317A mutation promoted binding to collagen. Similarly, the double mutation C139S/E317A increased adhesion compared with C139S alone. Furthermore, soluble α(1)I C139S/E317A was a higher avidity collagen binder than α(1)I C139S, indicating that the double variant represents an activated form. The crystal structure of the activated variant of α(1)I was solved at 1.9 Å resolution. The E317A mutation results in the unwinding of the αC helix, but the metal ion has moved toward loop 1, instead of loop 2 in the open α(2)I. Furthermore, unlike in the closed αI domains, the metal ion is pentacoordinated and, thus, prepared for ligand binding. Helix 7, which has moved downward in the open α(2)I structure, has not changed its position in the activated α(1)I variant. During the integrin activation, Glu(335) on helix 7 binds to the metal ion at the metal ion-dependent adhesion site (MIDAS) of the ß(1) subunit. Interestingly, in our cell adhesion assays E317A could activate collagen binding even after mutating Glu(335). This indicates that the stabilization of helix 7 into its downward position is not required if the α(1) MIDAS is already open. To conclude, the activated α(1)I domain represents a novel conformation of the αI domain, mimicking the structural state where the Arg(287)-Glu(317) ion pair has just broken during the integrin activation.

Identification of two imidazole binding sites and key residues for substrate specificity in human primary amine oxidase AOC3.

Human membrane primary amine oxidase (hAOC3; also known as vascular adhesion protein-1, VAP-1) is expressed upon inflammation in most tissues, where its enzymatic activity plays a crucial role in leukocyte trafficking. We have determined two new structures of a soluble, proteolytically cleaved form of hAOC3 (sAOC3), which was extracted from human plasma. In the 2.6 Å sAOC3 structure, an imidazole molecule is hydrogen bonded to the topaquinone (TPQ) cofactor, which is in an inactive on-copper conformation, while in the 2.95 Å structure, an imidazole molecule is covalently bound to the active off-copper conformation of TPQ. A second imidazole bound by Tyr394 and Thr212 was identified in the substrate channel. We furthermore demonstrated that imidazole has an inhibitory role at high concentrations used in crystallization. A triple mutant (Met211Val/Tyr394Asn/Leu469Gly) of hAOC3 was previously reported to change substrate preferences toward those of hAOC2, another human copper-containing monoamine oxidase. We now mutated these three residues and Thr212 individually to study their distinct role in the substrate specificity of hAOC3. Using enzyme activity assays, the effect of the four single mutations was tested with four different substrates (methylamine, benzylamine, 2-phenylethylamine, and p-tyramine), and their binding modes were predicted by docking studies. As a result, Met211 and Leu469 were shown to be key residues for substrate specificity. The native structures of sAOC3 and the mutational data presented in this study will aid the design of hAOC3 specific inhibitors.

Comparison of GFL-GFRalpha complexes: further evidence relating GFL bend angle to RET signalling.

Glial cell line-derived neurotrophic factor (GDNF) activates the receptor tyrosine kinase RET by binding to the GDNF-family receptor alpha1 (GFRalpha1) and forming the GDNF(2)-GFRalpha1(2)-RET(2) heterohexamer complex. A previous crystal structure of the GDNF(2)-GFRalpha1(2) complex (PDB code 2v5e) suggested that differences in signalling in GDNF-family ligand (GFL) complexes might arise from differences in the bend angle between the two monomers in the GFL homodimer. Here, a 2.35 A resolution structure of the GDNF(2)-GFRalpha1(2) complex crystallized with new cell dimensions is reported. The structure was refined to a final R factor of 22.5% (R(free) = 28%). The structures of both biological tetrameric complexes in the asymmetric unit are very similar to 2v5e and different from the artemin-GFRalpha3 structure, even though there is a small change in the structure of the GDNF. By comparison of all known GDNF and artemin structures, it is concluded that GDNF is more bent and more flexible than artemin and that this may be related to RET signalling. Comparisons also suggest that the differences between artemin and GDNF arise from the increased curvature of the artemin ;fingers', which both increases the buried surface area in the monomer-monomer interface and changes the intermonomer bend angle. From sequence comparison, it is suggested that neuturin (the second GFL) adopts an artemin-like conformation, while persephin has a different conformation to the other three.

The structure of the conserved neurotrophic factors MANF and CDNF explains why they are bifunctional.

We have solved the structures of mammalian mesencephalic astrocyte-derived neurotrophic factor (MANF) and conserved dopamine neurotrophic factor (CDNF). CDNF protects and repairs midbrain dopaminergic neurons in vivo; MANF supports their survival in culture and is also cytoprotective against endoplasmic reticulum (ER) stress. Neither protein structure resembles any known growth factor but the N-terminal domain is a saposin-like lipid-binding domain. MANF and CDNF may thus bind lipids or membranes. Consistent with this, there are two patches of conserved lysines and arginines. The natively unfolded MANF C-terminus contains a CKGC disulphide bridge, such as reductases and disulphide isomerases, consistent with a role in ER stress response. The structure thus explains why MANF and CDNF are bifunctional; neurotrophic activity may reside in the N-terminal domain and ER stress response in the C-terminal domain. Finally, we identified three changes, (MANF)I10-->K(CDNF), (MANF)E79-->M(CDNF) and (MANF)K88-->L(CDNF), that may account for the biological differences between the proteins.

The structure of the glial cell line-derived neurotrophic factor-coreceptor complex: insights into RET signaling and heparin binding.

Glial cell line-derived neurotrophic factor (GDNF), a neuronal survival factor, binds its co-receptor GDNF family receptor alpha1 (GFR alpha 1) in a 2:2 ratio and signals through the receptor tyrosine kinase RET. We have solved the GDNF(2).GFR alpha 1(2) complex structure at 2.35 A resolution in the presence of a heparin mimic, sucrose octasulfate. The structure of our GDNF(2).GFR alpha 1(2) complex and the previously published artemin(2).GFR alpha 3(2) complex are unlike in three ways. First, we have experimentally identified residues that differ in the ligand-GFR alpha interface between the two structures, in particular ones that buttress the key conserved Arg(GFR alpha)-Glu(ligand)-Arg(GFR alpha) interaction. Second, the flexible GDNF ligand "finger" loops fit differently into the GFR alphas, which are rigid. Third, and we believe most importantly, the quaternary structure of the two tetramers is dissimilar, because the angle between the two GDNF monomers is different. This suggests that the RET-RET interaction differs in different ligand(2)-co-receptor(2)-RET(2) heterohexamer complexes. Consistent with this, we showed that GDNF(2).GFR alpha1(2) and artemin(2).GFR alpha 3(2) signal differently in a mitogen-activated protein kinase assay. Furthermore, we have shown by mutagenesis and enzyme-linked immunosorbent assays of RET phosphorylation that RET probably interacts with GFR alpha 1 residues Arg-190, Lys-194, Arg-197, Gln-198, Lys-202, Arg-257, Arg-259, Glu-323, and Asp-324 upon both domains 2 and 3. Interestingly, in our structure, sucrose octasulfate also binds to the Arg(190)-Lys(202) region in GFR alpha 1 domain 2. This may explain how GDNF.GFR alpha 1 can mediate cell adhesion and how heparin might inhibit GDNF signaling through RET.