Optimization of DsRed production in Escherichia coli: effect of ribosome binding site sequestration on translation efficiency

DsRed-Express, a popular reporter protein, cannot be expressed in Escherichia coli using a consensus ribosome binding site (RBS) potentially due to basepairing in the RBS that inhibits translation initiation. Saturation mutagenesis was used to probe for a gene sequence that minimized basepairing in the RBS while maintaining the same spectral properties and maturation characteristics as DsRed-Express. The new DsRed, designated here as RFP(EC) for E. coli optimized red fluorescent protein, fluoresces 2.5 times greater than DsRed-Express when expressed from the same vector.     

The C-terminal domain of the Xenopus cyclin-dependent kinase inhibitor, p27Xic1, is both necessary and sufficient for phosphorylation-independent proteolysis

Cell cycle progression is regulated by cyclin-dependent kinases (CDKs), cyclins, and CDK inhibitors. In the frog, Xenopus laevis, the CDK inhibitor p27(Xic1) (Xic1) inhibits DNA synthesis by negatively regulating CDK2-cyclin E. Using the frog egg extract as a model system for the study of Xic1, studies have demonstrated that Xic1 protein levels are regulated by nuclear ubiquitination and proteolysis. To characterize the molecular mechanism that regulates Xic1 turnover, we have identified the minimal sequences of Xic1 that are necessary and sufficient for its nuclear ubiquitination and degradation. Using deletion mutagenesis, our studies indicated that the C-terminal 50 amino acids of Xic1 are critical for its proteolysis beyond a role in nuclear transport. Replacement of the Xic1 C terminus with the SV40 nuclear localization sequence resulted in the nuclear localization of Xic1 but not its ubiquitination or degradation. Our deletion studies also indicated that the CDK2-cyclin binding domain of Xic1 is important for its efficient retention in the nucleus. Further deletion analyses identified at least 3 lysine residues within the Xic1 C terminus that are targeted for specific ubiquitination. Importantly, our studies demonstrated that the Xic1 C-terminal 50 amino acids can serve as a nuclear degradation signal when fused to a stable heterologous nuclear protein. Moreover, a 30-amino-acid region within the C terminus of Xic1 can serve as a nuclear ubiquitination signal. To address the role of phosphorylation on Xic1 turnover, all the potential phosphorylation sites within the C-terminal 50 amino acids of Xic1 were mutated to alanine to prevent possible phosphorylation. This resulted in a Xic1 protein that was nevertheless degraded in a manner similar to wild-type Xic1, suggesting that phosphorylation of Xic1 is not critical for its nuclear ubiquitination or proteolysis.     

Genetic tools for reliable gene expression and recombineering in <Emphasis Type="Italic">Pseudomonas putida</Emphasis>

Pseudomonas putida is a promising bacterial host for producing natural products, such as polyketides and nonribosomal peptides. In these types of projects, researchers need a genetic toolbox consisting of plasmids, characterized promoters, and techniques for rapidly editing the genome. Past reports described constitutive promoter libraries, a suite of broad host range plasmids that replicate in P. putida, and genome-editing methods. To augment those tools, we have characterized a set of inducible promoters and discovered that IPTG-inducible promoter systems have poor dynamic range due to overexpression of the LacI repressor. By replacing the promoter driving lacI expression with weaker promoters, we increased the fold induction of an IPTG-inducible promoter in P. putida KT2440 to 80-fold. Upon discovering that gene expression from a plasmid was unpredictable when using a high-copy mutant of the BBR1 origin, we determined the copy numbers of several broad host range origins and found that plasmid copy numbers are significantly higher in P. putida KT2440 than in the synthetic biology workhorse, Escherichia coli. Lastly, we developed a λRed/Cas9 recombineering method in P. putida KT2440 using the genetic tools that we characterized. This method enabled the creation of scarless mutations without the need for performing classic two-step integration and marker removal protocols that depend on selection and counterselection genes. With the method, we generated four scarless deletions, three of which we were unable to create using a previously established genome-editing technique.     

Genetic heterogeneity of plasma lipoproteins in the mouse: control of low density lipoprotein particle sizes by genetic factors

In order to assess the genetic control of sizes and concentrations of mouse plasma low density (LDL) and high density lipoproteins (HDL), we used gel permeation fast protein liquid chromatography (FPLC) and nondenaturing gradient polyacrylamide gel electrophoresis to measure the particle sizes of LDL and HDL. Using chromatography we also quantified LDL-cholesterol and HDL-cholesterol concentrations in plasma and used them as indexes of plasma concentrations of the respective particles among 10 inbred strains (AKR/J, BALB/cByJ, C3H/HeJ, C57BL/6J, C57BL/6ByJ, C57L/J, DBA/1LacJ, 129/J, NZB/BINJ, SWR/J) and three sets of recombinant inbred (RI) strains (AKXL/TyJ, BXH/TyJ, CXB/ByJ) of mice. HDL had a dichotomous distribution among the 10 inbred strains. One group had large HDL particle sizes and high HDL-cholesterol concentrations. Another group had smaller HDL particles and lower HDL-cholesterol concentrations, and HDL sizes and HDL-cholesterol concentrations were significantly correlated. In the RI strains, HDL sizes and HDL-cholesterol cholesterol concentrations clearly segregated with one or another of the progenitor strains, and RI strain distributions showed a strong linkage to the apolipoprotein (apo) A-II gene (Apoa-2). In contrast, LDL-cholesterol concentrations and particle sizes on FPLC did not show dichotomous distributions among the 10 inbred strains. In RI strains, the configuration of the LDL FPLC profiles and LDL-cholesterol concentrations did resemble one or another of the progenitors in the majority of cases, but LDLs of several RI strains resembled neither progenitor strain in profile configuration, and LDL-cholesterol concentrations were both greater and smaller than those of progenitor strains. However, LDL particle diameters (as judged by peaks of LDL-cholesterol profiles) did segregate with progenitors in 29/33 (88%) of RI strains suggesting that a major gene may affect LDL size. In attempting to identify a major LDL-size determining gene, we compared apoB gene restriction fragment length polymorphisms (RFLPs) to the distributions of peak LDL sizes in RI strains. Concordance rates of peak LDL sizes to apoB gene polymorphisms were 18/22 (82%) for the EcoRV RFLP, 5/7 (71%) for HindIII RFLP, and 23/29 (79%) for both (range of P values 0.90-0.95). Thus we could not unequivocally implicate the apoB gene in determining the size of LDL particles. In summary, the genetic control of LDL sizes is more complicated than is the case for HDL; however, the differences in LDL size among these strains of mice may be controlled by a major, as yet unidentified, gene.     

Dual Catalytic Activity of Hydroxycinnamoyl-Coenzyme A Quinate Transferase from Tomato Allows It to Moonlight in the Synthesis of Both Mono- and Dicaffeoylquinic Acids

Tomato (Solanum lycopersicum), like other Solanaceous species, accumulates high levels of antioxidant caffeoylquinic acids, which are strong bioactive molecules and protect plants against biotic and abiotic stresses. Among these compounds, the monocaffeoylquinic acids (e.g. chlorogenic acid [CGA]) and the dicaffeoylquinic acids (diCQAs) have been found to possess marked antioxidative properties. Thus, they are of therapeutic interest both as phytonutrients in foods and as pharmaceuticals. Strategies to increase diCQA content in plants have been hampered by the modest understanding of their biosynthesis and whether the same pathway exists in different plant species. Incubation of CGA with crude extracts of tomato fruits led to the formation of two new products, which were identified by liquid chromatography-mass spectrometry as diCQAs. This chlorogenate:chlorogenate transferase activity was partially purified from ripe fruit. The final protein fraction resulted in 388-fold enrichment of activity and was subjected to trypsin digestion and mass spectrometric sequencing: a hydroxycinnamoyl-Coenzyme A:quinate hydroxycinnamoyl transferase (HQT) was selected as a candidate protein. Assay of recombinant HQT protein expressed in Escherichia coli confirmed its ability to synthesize diCQAs in vitro. This second activity (chlorogenate:chlorogenate transferase) of HQT had a low pH optimum and a high K_m for its substrate, CGA. High concentrations of CGA and relatively low pH occur in the vacuoles of plant cells. Transient assays demonstrated that tomato HQT localizes to the vacuole as well as to the cytoplasm of plant cells, supporting the idea that in this species, the enzyme catalyzes different reactions in two subcellular compartments.The importance of plant-based foods in preventing or reducing the risk of chronic disease has been widely demonstrated (Martin et al., 2011, 2013). In addition to vitamins, a large number of other nutrients in plant-based foods promote health and reduce the risk of chronic diseases; these are often referred to as phytonutrients. The presence of phytonutrients in fruit and vegetables is of significant nutritional and therapeutic importance, as many have been found to possess strong antioxidant activity (Rice-Evans et al., 1997). Phenolics are the most widespread dietary antioxidants and caffeoylquinic acids, such as chlorogenic acid (CGA), dicaffeoylquinic acids (diCQAs), and tricaffeoylquinic acids (triCQAs), play important roles in promoting health (Clifford, 1999; Niggeweg et al., 2004). CGA limits low density lipid oxidation (Meyer et al., 1998), diCQAs possess antihepatotoxic activity (Choi et al., 2005), and triCQAs reduce the blood Glc levels of diabetic rats (Islam, 2006). diCQA derivatives have been shown to protect humans from various kinds of diseases; diCQAs suppress melanogenesis effectively (Kaul and Khanduja, 1998), show anti-inflammatory activity in vitro (Peluso et al., 1995), and exhibit a selective inhibition of HIV replication (McDougall et al., 1998). The physiological effects of caffeoylquinic acid derivatives with multiple caffeoyl groups are generally greater than those of monocaffeoylquinic acids, perhaps because the antioxidant activity is largely determined by the number of hydroxyl groups present on the aromatic rings (Wang et al., 2003; Islam, 2006). Furthermore, both diCQAs and triCQAs may function as inhibitors of the activity of HIV integrase, which catalyzes the insertion of viral DNA into the genome of host cells (McDougall et al., 1998; Slanina et al., 2001; Gu et al., 2007).CGA is the major soluble phenolic in Solanaceous crops (Clifford, 1999) and the major antioxidant in the average U.S. diet (Luo et al., 2008), while different isomers of diCQAs have been identified in many crops such as coffee (Coffea canephora), globe artichoke (Cynara cardunculus), tomato (Solanum lycopersicum), lettuce (Lactuca sativa), and sweet potato (Ipomoea batatas; Clifford, 1999; Islam, 2006; Moco et al., 2006, 2007; Moglia et al., 2008). In tomato, CGA accounts for 75% and 35% of the total phenolics in mature green and ripe fruit, respectively, amounting to 2 to 40 mg 100 g^–1 dry weight (DW), although levels decline after ripening and during postharvest storage (Slimestad and Verheul, 2009). diCQAs and triCQAs also accumulate in tomato fruit (diCQAs, approximately 2 mg 100 g^–1 DW; and triCQAs, 1–2 mg 100 g^–1 DW; Chanforan et al., 2012).Three pathways (Villegas and Kojima, 1986; Hoffmann et al., 2003; Niggeweg et al., 2004) have been proposed for the synthesis of CGA: (1) the direct pathway involving caffeoyl-CoA transesterification with quinic acid by hydroxycinnamoyl-Coenzyme A:quinate hydroxycinnamoyl transferase (HQT; Niggeweg et al., 2004; Comino et al., 2009; Menin et al., 2010; Sonnante et al., 2010); (2) the route by which p-coumaroyl-CoA is first transesterified with quinic acid via hydroxycinnamoyl-Coenzyme A transferase (HCT) acyltransferase (Hoffmann et al., 2003; Comino et al., 2007), followed by the hydroxylation of p-coumaroyl quinate to 5-caffeoylquinic acid, catalyzed by C3′H (p-coumaroyl-3-hydroxylase; Schoch et al., 2001; Mahesh et al., 2007; Moglia et al., 2009); and (3) the use of caffeoyl-glucoside as the acyl-donor (Villegas and Kojima, 1986). In tomato, the synthesis of CGA involves transesterification of caffeoyl-CoA with quinic acid by HQT (Niggeweg et al., 2004).To date, it is not clear whether diCQAs are derived directly from the monocaffeoylquinic acids (such as CGA) through a second acyltransferase reaction involving an acyl-CoA or not, although their structural similarity provides good a priori evidence supporting this hypothesis. Recently the in vitro synthesis of 3,5-diCQA from CGA and CoA by HCT from coffee has been reported (Lallemand et al., 2012). By contrast, in sweet potato, an enzyme that catalyzes the transfer of the caffeoyl moiety of CGA to another molecule of CGA, leading to the synthesis of isochlorogenate (3,5-di-O-caffeoylquinate), has been described, but the corresponding gene has not been identified (Villegas and Kojima, 1986).We report a chlorogenate:chlorogenate transferase (CCT) activity leading to the synthesis of diCQAs in tomato fruits and describe how alternative catalysis, by a single enzyme, leads to the production of both CGA and diCQA in different cellular compartments.