Targeting and post‐translational processing of human α1‐antichymotrypsin in BY‐2 tobacco cultured cells†

The post‐translational processing of human α₁‐antichymotrypsin (AACT) in Bright Yellow‐2 (BY‐2) tobacco cells was assessed in relation to the cellular compartment targeted for accumulation. As determined by pulse‐chase labelling experiments and immunofluorescence microscopy, AACT sent to the vacuole or the endoplasmic reticulum (ER) was found mainly in the culture medium, similar to a secreted form targeted to the apoplast. Unexpectedly, AACT expressed in the cytosol was found in the nucleus under a stable, non‐glycosylated form, in contrast with secreted variants undergoing multiple post‐translational modifications during their transit through the secretory pathway. All secreted forms of AACT were N‐glycosylated, with the presence of complex glycans as observed naturally on human AACT. Proteolytic trimming was also observed for all secreted variants, both during their intracellular transit and after their secretion in the culture medium. Overall, the targeting of human AACT to different compartments of BY‐2 tobacco cells led to the production of two protein products: (i) a stable, non‐glycosylated protein accumulated in the nucleus; and (ii) a heterogeneous mixture of secreted variants resulting from post‐translational N‐glycosylation and proteolytic processing. Overall, these data suggest that AACT is sensitive to resident proteases in the ER, the Golgi and/or the apoplast, and that the production of intact AACT in the plant secretory pathway will require innovative approaches to protect its structural integrity in vivo. Studies are now needed to assess the activity of the different AACT variants, and to identify the molecular determinants for the nuclear localization of AACT expressed in the cytosol.     

Understanding the function of bacterial and eukaryotic thiolases II by integrating evolutionary and functional approaches

Acetoacetyl-CoA thiolase (EC 2.3.1.9), commonly named thiolase II, condenses two molecules of acetyl-CoA to give acetoacetyl-CoA and CoA. This enzyme acts in anabolic processes as the first step in the biosynthesis of isoprenoids and polyhydroxybutyrate in eukaryotes and bacteria, respectively. We have recently reported the evolutionary and functional equivalence of these enzymes, suggesting that thiolase II could be the rate limiting enzyme in these pathways and presented evidence indicating that this enzyme modulates the availability of reducing equivalents during abiotic stress adaptation in bacteria and plants. However, these results are not sufficient to clarify why thiolase II was evolutionary selected as a critical enzyme in the production of antioxidant compounds. Regarding this intriguing topic, we propose that thiolase II could sense changes in the acetyl-CoA/CoA ratio induced by the inhibition of the tricarboxylic acid cycle under abiotic stress. Thus, the high level of evolutionary and functional constraint of thiolase II may be due to the connection of this enzyme with an ancient and conserved metabolic route.     

Nucleocytoplasmic transit of human α1‐antichymotrypsin in tobacco leaf epidermal cells†

Recently, we have observed a nuclear localization for human α₁‐antichymotrypsin (AACT) expressed in the cytosol of transgenic Bright Yellow‐2 (BY‐2) tobacco cultured cells (see accompanying paper: Benchabane, M., Saint‐Jore‐Dupas, C., Bardor, M., Faye, L., Michaud, D. and Gomord, V. (2008a) Targeting and post‐translational processing of human α₁‐antichymotrypsin in BY‐2 tobacco cultured cells. Plant Biotechnol. J. doi: 10.1111/j.1467‐7652.2008.00382.x). In the present article, we assess whether the intrinsic DNA‐binding activity of AACT can explain its nuclear localization, and whether this same activity has an impact on its protease inhibitory potency and stability in planta. An engineered form of AACT with no DNA‐binding activity, rAACTΔK, was compared with the wild‐type polypeptide, rAACT, in terms of chymotrypsin inhibitory potency, stability in planta and distribution in tobacco cells. In accordance with available data reporting distinct sites for protease inhibition and DNA binding, rAACT and rAACTΔK showed similar antichymotrypsin activity, similar to the activity of native AACT purified from human plasma. As observed for AACT in BY‐2 tobacco cells, a green fluorescent protein (GFP)‐AACT fusion transiently expressed in the cytosol of tobacco leaf epidermal cells was detected mainly in the nucleus by confocal laser microscopy. By contrast, rAACTΔK expressed as a GFP fusion showed a balanced distribution between the cytosol and the nucleus, similar to the distribution pattern of free GFP exhibiting no DNA‐binding affinity. In line with immunodetection data showing higher accumulation levels for GFP‐AACT in tobacco leaf cells, rAACTΔK was more susceptible than rAACT to tryptic digestion in the presence of DNA. Overall, these observations suggest the following: (i) a retention effect of DNA on AACT in the nucleus; and (ii) a stabilizing effect of the AACT–DNA interaction on rAACT challenged with non‐target proteases, which, possibly, may be useful in protecting this protein in plant expression platforms.     

Amino acid-coded tagging approaches in quantitative proteomics

To improve the efficiency, accuracy, reproducibility, throughput and proteome coverage of mass spectrometry-based quantitative approaches, both in vitro and in vivo tagging of particular amino acid residues of cellular proteins have been introduced to assist mass spectrometry for global-scale comparative studies of differentially expressed proteins/modifications between different biologically relevant cell states or cells at different pathological states. The basic features of these methods introduce pair-wise isotope signals of each individual peptide containing a particular type of tagged amino acid (amino acid-coded mass tagging) that originated from different cell states. In this review, the applications of major amino acid-coded mass tagging-based quantitative proteomics approaches, including isotope-coded affinity tag, isobaric tags for relative and absolute quantification (iTRAQ?) and stable isotope labeling by amino acids in cell culture are summarized in the context of their respective strengths/weakness in identifying those differentially expressed or post-translational modified proteins regulated by particular cellular stress on a genomic scale in a high-throughput manner. Importantly, these gel-free, in-spectra quantitative mechanisms have been further explored to identify/characterize large-scale protein–protein interactions involving various functional pathways. Taken together, the information about quantitative proteome changes, including multiple regulated proteins and their interconnected relationships, will provide an important insight into the molecular mechanisms, where novel targets for diagnosis and therapeutic intervention will be identified.     

Comparative Proteomic Analysis Identifies Key Metabolic Regulators of Gemcitabine Resistance in Pancreatic Cancer

Pancreatic adenocarcinoma (PDAC) is highly refractory to treatment. Standard-of-care gemcitabine (Gem) provides only modest survival benefits, and development of Gem resistance (GemR) compromises its efficacy. Highly GemR clones of Gem-sensitive MIAPaCa-2 cells were developed to investigate the molecular mechanisms of GemR and implemented global quantitative differential proteomics analysis with a comprehensive, reproducible ion-current–based MS1 workflow to quantify ～6000 proteins in all samples. In GemR clone MIA-GR8, cellular metabolism, proliferation, migration, and ‘drug response’ mechanisms were the predominant biological processes altered, consistent with cell phenotypic alterations in cell cycle and motility. S100 calcium binding protein A4 was the most downregulated protein, as were proteins associated with glycolytic and oxidative energy production. Both responses would reduce tumor proliferation. Upregulation of mesenchymal markers was prominent, and cellular invasiveness increased. Key enzymes in Gem metabolism pathways were altered such that intracellular utilization of Gem would decrease. Ribonucleoside-diphosphate reductase large subunit was the most elevated Gem metabolizing protein, supporting its critical role in GemR. Lower Ribonucleoside-diphosphate reductase large subunit expression is associated with better clinical outcomes in PDAC, and its downregulation paralleled reduced MIAPaCa-2 proliferation and migration and increased Gem sensitivity. Temporal protein-level Gem responses of MIAPaCa-2 versus GemR cell lines (intrinsically GemR PANC-1 and acquired GemR MIA-GR8) implicate adaptive changes in cellular response systems for cell proliferation and drug transport and metabolism, which reduce cytotoxic Gem metabolites, in DNA repair, and additional responses, as key contributors to the complexity of GemR in PDAC. These findings additionally suggest targetable therapeutic vulnerabilities for GemR PDAC patients.     

New insight into the evolution of aquaporins from flowering plants and vertebrates: orthologous identification and functional transfer is possible

Aquaporins (AQPs) represent a family of channel proteins that transport water and/or small solutes across cell membranes in the three domains of life. In all previous phylogenetic analysis of aquaporin, trees constructed using proteins with very low amino acid identity (<15%) were incongruent with rRNA data. In this work, restricting the evolutionary study of aquaporins to proteins with high amino acid identity (>25%), we showed congruence between AQPs and organismal trees. On the basis of this analysis, we defined 19 orthologous gene clusters in flowering plant species (3 PIP-like, 7 TIP-like, 6 NIP-like and 3 SIP-like). We described specific conserved motifs for each subfamily and each cluster, which were used to develop a method for automatic classification. Analysis of amino acid identity between orthologous monocotyledon and dicotyledon AQPs from each cluster, suggested that PIPs are under high evolutionary constraint. The phylogenetic analysis allowed us the assignment of orthologous aquaporins for very distant animal lineages (tetrapods-fishes). We also demonstrated that the location of all vertebrate AQPs in the ortholog clusters could be predicted by comparing their amino acid identity with human AQPs. We defined four AQP subfamilies in animals: AQP1-like, AQP8-like, AQP3-like and AQP11-like. Phylogenetic analysis showed that the four animal AQPs subfamilies are related with PIP-like, TIP-like, NIP-like and SIP-like subfamilies, respectively. Thus, this analysis would allow the prediction of individual AQPs function on the basis of orthologous genes from Arabidopsis thaliana and Homo sapiens.