Search

Phosphorylation is one of the most important post-translational modifications, playing a crucial role in regulating many cellular processes, including transcription, cytoskeletal rearrangement, cell proliferation, differentiation, apoptosis, and signal transduction. However, to date, little work has been carried out on the phosphoproteome in CHO cells. In this study we have carried out a large scale differential phosphoproteomic analysis of recombinant CHO cells following a reduction of culture temperature (temperature shift). The reduction of culture temperature during the exponential phase of growth is commonly employed by the biopharmaceutical industry to increase product yield; however, the molecular mechanisms of temperature shift in CHO cells remain poorly understood. We have identified 700 differentially expressed phosphopeptides using quantitative label-free LC-MS/MS phosphoproteomic analysis in conjunction with IMAC and TiO 2 phosphopeptide enrichment strategies, following a reduction in temperature from 37 to 31 °C. Functional assessment of the phosphoproteomic data using gene ontology analysis showed a significant enrichment of biological processes related to growth (e.g., cell cycle, cell division), ribosomal biogenesis, and cytoskeleton organization, and molecular functions related to RNA binding, transcription factor activity, and protein serine/threonine kinase activity. Differential phosphorylation of two proteins, ATF2 and NDRG1, was confirmed by Western blotting. This data suggests the importance of including the post-translational layer of regulation, such as phosphorylation, in CHO "omics" studies. This study also has the potential to identify phosphoprotein targets that could be modified using cell line engineering approaches to improve the efficiency of recombinant protein production.

In global proteomic analysis, it is estimated that proteins span from millions to less than 100 copies per cell. The challenge of protein quantitation by classic shotgun proteomic techniques relies on the presence of missing values in peptides belonging to low-abundance proteins that lowers intraruns reproducibility affecting postdata statistical analysis. Here, we present a new analytical workflow MvM (missing value monitoring) able to recover quantitation of missing values generated by shotgun analysis. In particular, we used confident data-dependent acquisition (DDA) quantitation only for proteins measured in all the runs, while we filled the missing values with data-independent acquisition analysis using the library previously generated in DDA. We analyzed cell cycle regulated proteins, as they are low abundance proteins with highly dynamic expression levels. Indeed, we found that cell cycle related proteins are the major components of the missing values-rich proteome. Using the MvM workflow, we doubled the number of robustly quantified cell cycle related proteins, and we reduced the number of missing values achieving robust quantitation for proteins over ∼50 molecules per cell. MvM allows lower quantification variance among replicates for low abundance proteins with respect to DDA analysis, which demonstrates the potential of this novel workflow to measure low abundance, dynamically regulated proteins.

Lysine crotonylation on histones is a recently identified post-translational modification that has been demonstrated to associate with active promoters and to directly stimulate transcription. Given that crotonyl-CoA is essential for the acyl transfer reaction and it is a metabolic intermediate widely localized within the cell, we postulate that lysine crotonylation on nonhistone proteins could also widely exist. Using specific antibody enrichment followed by high-resolution mass spectrometry analysis, we identified hundreds of crotonylated proteins and lysine residues. Bioinformatics analysis reveals that crotonylated proteins are particularly enriched for nuclear proteins involved in RNA processing, nucleic acid metabolism, chromosome organization, and gene expression. Furthermore, we demonstrate that crotonylation regulates HDAC1 activity, expels HP1α from heterochromatin, and inhibits cell cycle progression through S-phase. Our data thus indicate that lysine crotonylation could occur in a large number of proteins and could have important regulatory roles in multiple nuclei-related cellular processes.

The filamentous bacterium Streptomyces coelicolor modulates polar growth and branching by phosphorylating the cytoskeletal protein DivIVA. Previous MALDI-TOF analysis of DivIVA showed that a large 7.2 kDa tryptic peptide was multiply phosphorylated. To aid localization of the phosphorylation sites, we introduced additional tryptic cleavage sites into DivIVA, and the resulting phosphopeptides were analyzed by LC-MS/MS. Phosphopeptide isomers could be separated chromatographically, but because of overlapping elution and spectrum quality, site assignment by standard software tools was ambiguous. Because fragment ions carrying the phosphate group are essential for confident localization, large numbers of spectra were collected using targeted LC-MS/MS, and a special script was developed for plotting the elution of site-determining fragments from those spectra under the XIC of the parent ions. Where multiple phosphopeptide isomers were present, the elution of the site-determining y-ions perfectly coincided with the elution of the corresponding phosphopeptide isomer. This method represents a useful tool for user inspection of spectra derived from phosphopeptide isomers and significantly increases confidence when defining phosphorylation sites. In this way, we show that DivIVA is phosphorylated in vivo on five sites in the C-terminal part of the protein (T304, S309, S338, S344, and S355). The data have been deposited to the ProteomeXchange Consortium with identifier PXD000095.

Here we report the use of capillary isoelectric focusing under native conditions for the separation of protein complex isoforms and subcomplexes. Using biologically relevant HIS-tag and FLAG-tag purified protein complexes, we demonstrate the separations of protein complex isoforms of the mammalian target of rapamycin complex (mTORC1 and 2) and the subcomplexes and different phosphorylation states of the Dam1 complex. The high efficiency capillary isoelectric focusing separation allowed for resolution of protein complexes and subcomplexes similar in size and biochemical composition. By performing separations with native buffers and reduced temperature (15 degrees C) we were able to maintain the complex integrity of the more thermolabile mTORC2 during isoelectric focusing and detection (<45 min). Increasing the separation temperature allowed us to monitor dissociation of the Dam1 complex into its subcomplexes (25 degrees C) and eventually its individual protein components (30 degrees C). The separation of two different phosphorylation states of the Dam1 complex, generated from an in vitro kinase assay with Mps1 kinase, was straightforward due to the large pI shift upon multiple phosphorylation events. The separation of the protein complex isoforms of mTORC, on the other hand, required the addition of a small pI range (4-6.5) of ampholytes to improve resolution and stability of the complexes. We show that native capillary isoelectric focusing is a powerful method for the difficult separations of large, similar, unstable protein complexes. This method shows potential for differentiation of protein complex isoform and subcomplex compositions, post-translational modifications, architectures, stabilities, equilibria, and relative abundances under biologically relevant conditions.

Many cellular processes are driven by protein complexes. Although the identification of protein components in such complexes has become almost a routine matter, accurate determination of their stoichiometry within a protein complex is still a challenge. We have established a method to determine the stoichiometries of protein complexes using absolute quantification (AQUA) with the help of synthetic standard peptides in combination with multiple reaction monitoring (MRM). Our approach is exemplified by the analysis of the human spliceosomal hPrp19/CDC5L complex, which consists of seven individual proteins and plays a crucial role in the assembly of the fully catalytically active spliceosome during pre-mRNA splicing. We evaluated several conditions for complete hydrolysis of the protein complex and found that the denaturing conditions under which hydrolysis is performed are absolutely crucial for accurately determining protein stoichiometries within this complex. In addition, we tested the suitability of different AQUA peptides and further compared different MS techniques to read out the relative signal intensities that were then used in absolute quantification. Our analyses revealed that dependent on the denaturing conditions different stoichiometries within the complex were obtained. The most consistent results were obtained by enzymatic hydrolysis in the presence of acetonitrile in combination with MRM.

Ribonucleotide reductases (RNRs) catalyze the conversion of nucleoside 5'-diphosphates to the corresponding deoxynucleotides supplying the dNTPs required for DNA replication and DNA repair. Class I RNRs require two subunits, alpha and beta, for activity. Humans possess two beta subunits: one involved in S phase DNA replication (beta) and a second in mitochondrial DNA replication (beta' or p53R2) and potentially DNA repair. Gemcitabine (F(2)C) is used clinically as an anticancer agent, and its phosphorylated metabolites target many enzymes involved in nucleotide metabolism, including RNR. The present investigation with alpha (specific activity of 400 nmol min(-1) mg(-1)) and beta' (0.6 Y./beta'2 and a specific activity of 420 nmol min(-1) mg(-1)) establishes that F(2)CDP is a substoichiometric inactivator of RNR. Incubation of this alpha/beta' with [1'-(3)H]-F(2)CDP or [5-(3)H]-F(2)CDP and reisolation of the protein by Sephadex G-50 chromatography resulted in recovery 0.5 equiv of covalently bound sugar and 0.03 equiv of tightly associated cytosine to alpha2. SDS-PAGE analysis (loaded without boiling) of the inactivated RNR showed that 60% of alpha migrates as a 90 kDa protein and 40% as a 120 kDa protein. Incubation of [1'-(3)H]-F(2)CDP with active site mutants C444S/A, C218S/A, and E431Q/D-alpha and the C-terminal tail C787S/A and C790S/A mutants reveals that no sugar label is bound to the active site mutants of alpha and that, in the case of C218S-alpha, alpha migrates as a 90 kDa protein. Analysis of the inactivated wt-alpha/beta' RNR by size exclusion chromatography indicates a quaternary structure of alpha6beta'6. A mechanism of inactivation common with halpha/beta is presented.

The mitotic protein kinase Plk1 catalyzes events associated with centrosome maturation, kinetocore function, spindle formation, and cytokinesis and is a target for anticancer drug design. It is composed of a N-terminal kinase domain and a C-terminal polo-box domain (PBD). The PBD domain serves to localize the kinase on cognate phosphorylated substrates, and this binding relieves the inhibition of the kinase by the PBD. Similar to many protein kinases, Plk1 is activated by phosphorylation on a threonine residue, Thr210, in the activation segment. In this work, we describe expression in Escherichia coli cells and purification of full-length Plk1 in quantities suitable for structural studies and use this material for quantitative characterization of the activation events with the substrate translationally controlled tumour protein (TCTP). The presence of the PBD-binding phosphopeptide enhances phosphorylation by the activating Ste20-like kinase (Slk). Native Plk1 exhibits a basal catalytic efficiency k cat/ K(M) of 9.9 x 10 (-5) s (-1) microM (-1). Association with a polo-box-binding phosphopeptide increased the catalytic efficiency by 11x largely through an increase in k(cat) with no change in K(M). Phosphorylation by Slk increases catalytic efficiency by 202x with a 2.3-fold reduction in K(M) and 88-fold increase in k(cat). Phosphorylation and the presence of the PBD-binding phosphopeptide result in an increase in catalytic efficiency of 1515x with a 2.3-fold decrease in K(M) and a 705-fold increase in k(cat) over the unmodified Plk1. Knowledge of kinase regulatory mechanisms and the structures of the Plk1 individual domains has allowed for a model to be proposed for these activatory events.

Minichromosomal maintenance proteins (MCMs) are considered to be the replicative helicase. Methanobacterium thermoautotrophicum has a single MCM gene (mtMCM). The crystal structure of the mtMCM N-terminal region is a double hexamer. Structure-guided sequence alignment indicates a structural conservation of this fragment across archaeal and eukaryotic MCMs. The mtMCM structure was successfully used to analyze a Saccharomyces cerevisiae MCM5 mutant, called BOB1, which contains a single residue change from Pro to Leu and bypasses a kinase normally required for initiation of DNA replication. A domain-push model was proposed to explain the BOB1 bypass activity. Here we investigate the effects of BOB1 mutation on the biochemical activities of mtMCM. Surprisingly, the BOB1 mutation (P62L) had a major effect on the helicase activity but had no significant impact on DNA binding and ATPase activities. These results will contribute to a more detailed understanding of the BOB1 bypass activity and other aspects of DNA replication control.

Vav proteins are Rho GTPase-specific guanine nucleotide exchange factors (GEFs) that are distinguished by the tandem arrangement of Dbl homology (DH), Pleckstrin homology (PH), and cysteine rich domains (CRD). Whereas the tandem DH-PH arrangement is conserved among Rho GEFs, the presence of the CRD is unique to Vav family members and is required for efficient nucleotide exchange. We provide evidence that Vav2-mediated nucleotide exchange of Rho GTPases follows the Theorell-Chance mechanism in which the Vav2.Rho GTPase complex is the major species during the exchange process and the Vav2.GDP-Mg(2+).Rho GTPase ternary complex is present only transiently. The GTPase specificity for the DH-PH-CRD Vav2 in vitro follows this order: Rac1 > Cdc42 > RhoA. Results obtained from fluorescence anisotropy and NMR chemical shift mapping experiments indicate that the isolated Vav1 CRD is capable of directly associating with Rac1, and residues K116 and S83 that are in the proximity of the P-loop and the guanine base either are part of this binding interface or undergo a conformational change in response to CRD binding. The NMR studies are supported by kinetic measurements on Rac1 mutants S83A, K116A, and K116Q and Vav2 CRD mutant K533A in that these mutants affect both the initial binding event of Vav2 with Rac1 (k(on)) and the rate-limiting dissociation of Vav2 from the Vav2.Rac1 binary complex (thereby influencing the enzyme turnover number, k(cat)). The results suggest that the CRD domain in Vav proteins plays an active role, affecting both the k(on) and the k(cat) for Vav-mediated nucleotide exchange on Rho GTPases.

Protein phosphorylation is a major mechanism of post-translational protein modification used to control cellular signaling. A challenge in phosphoproteomics is to identify the direct substrates of each protein kinase. Herein, we describe a chemical strategy for delivery of a bio-orthogonal affinity tag to the substrates of an individual protein kinase. The kinase of interest is engineered to transfer a phosphorothioate moiety to phosphoacceptor hydroxyl groups on direct substrates. In a second nonenzymatic step, the introduced phosphorothioate is alkylated with p-nitrobenzylmesylate (PNBM). Antibodies directed against the alkylated phosphorothioate epitope recognize these labeled substrates, but not alkylation products of other cellular nucleophiles. This strategy is demonstrated with Cdk1/cyclinB substrates using ELISA, western blotting, and immunoprecipitation in the context of whole cell lysates.

Regulation of histone gene transcription at the G1/S phase transition via the Site II cell cycle control element is distinct from E2F-dependent mechanisms operative at the growth factor-related restriction point. E2F-independent activation of histone H4 gene expression combines contributions of several promoter factors, including HiNF-M/IRF2 and the HiNF-D/CDP-cut complex which contains pRB, CDK1, and cyclin A as non-DNA binding subunits. Mutational analyses suggest additional rate-limiting factors for Site II function. Using sequence-specific Site II DNA affinity chromatography, we identified a 45 kDa protein (KIAA0005 or BZAP45) that is embryonically expressed and phylogenetically conserved. Based on amino acid sequence analysis, BZAP45 contains a unique decapeptide that is part of a putative leucine-zipper protein with a nucleotide (ATP or GTP) binding fold. Bacterial expression of a full-length cDNA produces a 45 kDa protein. Binding studies reveal that highly purified BZAP45 does not interact with Site II, suggesting that BZAP45 function may require partner proteins. Forced expression of BZAP45 strongly stimulates H4 promoter (nt -215 to -1)/CAT reporter gene activity. Deletion analyses and point mutations indicate that BZAP45 enhances H4 gene transcription through Site II. Thus, BZAP45 is a novel regulatory factor that contributes to transcriptional control at the G1/S phase transition.

Cdc25 is a dual-specificity phosphatase that catalyzes the activation of the cyclin-dependent kinases, thus causing initiation and progression of successive phases of the cell cycle. Although it is not significantly structurally homologous to other well-characterized members, Cdc25 belongs to the class of well-studied cysteine phosphatases as it contains their active site signature motif. However, the catalytic acid needed for protonation of the leaving group has yet to be identified. To elucidate the role and identity of this key catalytic residue, we have performed a detailed pH-dependent kinetic analysis of Cdc25B. The pK(a) of the catalytic cysteine was found to be 5.6-6.3 in steady state and one-turnover burst experiments using the small molecule substrates p-nitrophenyl phosphate and 3-O-methylfluorescein phosphate. Interestingly, Cdc25B does not exhibit the typical bell-shaped pH-rate profile with small molecule substrates seen in other cysteine phosphatases and indicative of the catalytic acid because it lacks pH dependence between 6.5 and 9. Reactions of Cdc25B with the natural substrate Cdk2-pTpY/CycA, however, did yield a bell-shaped pH-rate profile with a pK(a) of 6.1 for the catalytic acid residue. Recent structural studies of Cdc25 have suggested that Glu474 [Fauman, E. B., et al. (1998) Cell 93, 617-625] or Glu478 [Reynolds, R. A., et al. (1999) J. Mol. Biol. 293, 559-568] could function as the catalytic acid in Cdc25B. Using site-directed mutagenesis and truncation experiments, however, we found that neither of these residues, nor the unstructured C-terminus, is responsible for the observed pH dependence. These results indicate that the catalytic acid does not appear to lie within the known structure of Cdc25B and may lie on its protein substrate.

Ataxia telangiectasia mutated (ATM)- and Rad3-related protein (ATR) is a phosphatidylinositol-kinase (PIK)-related kinase that has been implicated in the response of human cells to multiple forms of DNA damage and may play a role in the DNA replication checkpoint. The purification of an ATR complex allowed identification of chromodomain-helicase-DNA-binding protein 4 (CHD4) as an ATR-associated protein by tandem mass spectrometric sequencing. CHD4 (also called Mi-2beta) is a component of a histone-deacetylase-2 (HDAC2)-containing complex, the nucleosome remodeling and deacetylating (NRD) complex. Endogenous ATR, CHD4, and HDAC2 are shown to coimmunoprecipitate, and ATR and HDAC2 coelute through two biochemical purification steps. Other members of the NRD complex, HDAC1, MTA1, and MTA2, are also detectable in ATR immunoprecipitates. ATR's association with CHD4 and HDAC2 suggests that there may be a linkage between ATR's role in mediating checkpoints induced by DNA damage and chromatin modulation via remodeling and deacetylation.

Previously we showed that the yeast proteins Spt16 (Cdc68) and Pob3 are physically associated, and interact physically and genetically with the catalytic subunit of DNA polymerase alpha, Pol1 [Wittmeyer and Formosa (1997) Mol. Cell. Biol. 17, 4178-4190]. Here we show that purified Spt16 and Pob3 form a stable, abundant, elongated heterodimer and provide evidence that this is the functional form of these proteins. Genetic interactions between mutations in SPT16 and POB3 support the importance of the Spt16-Pob3 interaction in vivo. Spt16, Pob3, and Pol1 proteins were all found to localize to the nucleus in S. cerevisiae. A portion of the total cellular Spt16-Pob3 was found to be chromatin-associated, consistent with the proposed roles in modulating chromatin function. Some of the Spt16-Pob3 complex was found to copurify with the yeast DNA polymerase alpha/primase complex, further supporting a connection between Spt16-Pob3 and DNA replication.

Guanine nucleotide exchange factors (GEFs) activate Ras proteins by stimulating the exchange of GTP for GDP in a multistep mechanism which involves binary and ternary complexes between Ras, guanine nucleotide, and GEF. We present fluorescence measurements to define the kinetic constants that characterize the interactions between Ras, GEF, and nucleotides, similar to the characterization of the action of RCC1 on Ran [Klebe et al. (1995) Biochemistry 34, 12543-12552]. The dissociation constant for the binary complex between nucleotide-free Ras and the catalytic domain of mouse Cdc25, Cdc25(Mm285), was 4.6 nM, i.e., a 500-fold lower affinity than the Ras.GDP interaction. The affinities defining the ternary complex Ras. nucleotide.Cdc25(Mm285) are several orders of magnitude lower. The maximum acceleration by Cdc25(Mm285) of the GDP dissociation from Ras was more than 10(5)-fold. Kinetic measurements of the association of nucleotide to nucleotide-free Ras and to the binary complex Ras. Cdc25(Mm285) show that these reactions are practically identical: a fast binding step is followed by a reaction of the first order which becomes rate limiting at high nucleotide concentrations. The second reaction is thought to be a conformational change from a low- to a high-affinity nucleotide binding conformation in Ras. Taking into consideration all experimental data, the reverse isomerization reaction from a high- to a low-affinity binding conformation in the ternary complex Ras. GDP.Cdc25(Mm285) is postulated to be the rate-limiting step of the GEF-catalyzed exchange. Furthermore, we demonstrate that the disruption of the Mg2+-binding site is not the only factor in the mechanism of GEF-catalyzed nucleotide exchange on Ras.