Deglycosylated RBD produced in Pichia pastoris as a low-cost sera COVID-19 diagnosis tool and a vaccine candidate.

During the COVID-19 outbreak, numerous tools including protein-based vaccines have been developed. The methylotrophic yeast Pichia pastoris (synonymous to Komagataella phaffii) is an eukaryotic cost-effective and scalable system for recombinant protein production, with the advantages of an efficient secretion system and the protein folding assistance of the secretory pathway of eukaryotic cells. In a previous work, we compared the expression of SARS-CoV-2 Spike Receptor Binding Domain in P. pastoris with that in human cells. Although the size and glycosylation pattern was different between them, their protein structural and conformational features were indistinguishable. Nevertheless, since high mannose glycan extensions in proteins expressed by yeast may be the cause of a nonspecific immune recognition, we deglycosylated RBD in native conditions. This resulted in a highly pure, homogenous, properly folded and monomeric stable protein. This was confirmed by circular dichroism and tryptophan fluorescence spectra and by SEC-HPLC, which were similar to those of RBD proteins produced in yeast or human cells. Deglycosylated RBD was obtained at high yields in a single step, and it was efficient in distinguishing between SARS-CoV-2-negative and positive sera from patients. Moreover, when the deglycosylated variant was used as an immunogen, it elicited a humoral immune response ten times greater than the glycosylated form, producing antibodies with enhanced neutralizing power and eliciting a more robust cellular response. The proposed approach may be used to produce at a low cost, many antigens that require glycosylation to fold and express, but do not require glycans for recognition purposes.

Biochemical characterization of a GDP-mannose transporter from Chaetomium thermophilum.

Nucleotide Sugar Transporters (NSTs) belong to the SLC35 family (human solute carrier) of membrane transport proteins and are crucial components of the glycosylation machinery. NSTs are localized in the ER and Golgi apparatus membranes, where they accumulate nucleotide sugars from the cytosol for subsequent polysaccharide biosynthesis. Loss of NST function impacts the glycosylation of cell surface molecules. Mutations in NSTs cause several developmental disorders, immune disorders, and increased susceptibility to infection. Atomic resolution structures of three NSTs have provided a blueprint for a detailed molecular interpretation of their biochemical properties. In this work, we have identified, cloned, and expressed 18 members of the SLC35 family from various eukaryotic organisms in Saccharomyces cerevisiae. Out of 18 clones, we determined Vrg4 from Chaetomium thermophilum (CtVrg4) is a GDP-mannose transporter with an enhanced melting point temperature (Tm) of 56.9°C, which increases with the addition of substrates, GMP and GDP-mannose. In addition, we report-for the first time-that the CtVrg4 shows an affinity to bind to phosphatidylinositol lipids.

The broad range di- and tri-nucleotide exchanger SLC35B1 displays asymmetrical affinities for ATP transport across the ER membrane.

In eukaryotic cells, uptake of cytosolic ATP into the endoplasmic reticulum (ER) lumen is critical for the proper functioning of chaperone proteins. The human transport protein SLC35B1 was recently postulated to mediate ATP/ADP exchange in the ER; however, the underlying molecular mechanisms mediating ATP uptake are not completely understood. Here, we extensively characterized the transport kinetics of human SLC35B1 expressed in yeast that was purified and reconstituted into liposomes. Using [α32P]ATP uptake assays, we tested the nucleotide concentration dependence of ATP/ADP exchange activity on both sides of the membrane. We found that the apparent affinities of SLC35B1 for ATP/ADP on the internal face were approximately 13 times higher than those on the external side. Because SLC35B1-containing liposomes were preferentially inside-out oriented, these results suggest a low-affinity external site and a high-affinity internal site in the ER. Three different experimental approaches indicated that ATP/ADP exchange by SLC35B1 was not strict, and that other di- and tri-nucleotides could act as suitable counter-substrates for ATP, although mononucleotides and nucleotide sugars were not transported. Finally, bioinformatic analysis and site-directed mutagenesis identified that conserved residues K117 and K120 from transmembrane helix 4 and K277 from transmembrane helix 9 play critical roles in transport. The fact that SLC35B1 can promote ATP transport in exchange for ADP or UDP suggest a more direct coupling between ATP import requirements and the need for eliminating ADP and UDP, which are generated as side products of reactions taking place in the ER-lumen.

Conserved Glu-47 and Lys-50 residues are critical for UDP-N-acetylglucosamine/UMP antiport activity of the mouse Golgi-associated transporter Slc35a3.

Nucleotide sugar transporters (NSTs) regulate the flux of activated sugars from the cytosol into the lumen of the Golgi apparatus where glycosyltransferases use them for the modification of proteins, lipids, and proteoglycans. It has been well-established that NSTs are antiporters that exchange nucleotide sugars with the respective nucleoside monophosphate. Nevertheless, information about the molecular basis of ligand recognition and transport is scarce. Here, using topology predictors, cysteine-scanning mutagenesis, expression of GFP-tagged protein variants, and phenotypic complementation of the yeast strain Kl3, we identified residues involved in the activity of a mouse UDP-GlcNAc transporter, murine solute carrier family 35 member A3 (mSlc35a3). We specifically focused on the putative transmembrane helix 2 (TMH2) and observed that cells expressing E47C or K50C mSlc35a3 variants had lower levels of GlcNAc-containing glycoconjugates than WT cells, indicating impaired UDP-GlcNAc transport activity of these two variants. A conservative substitution analysis revealed that single or double substitutions of Glu-47 and Lys-50 do not restore GlcNAc glycoconjugates. Analysis of mSlc35a3 and its genetic variants reconstituted into proteoliposomes disclosed the following: (i) all variants act as UDP-GlcNAc/UMP antiporters; (ii) conservative substitutions (E47D, E47Q, K50R, or K50H) impair UDP-GlcNAc uptake; and (iii) substitutions of Glu-47 and Lys-50 dramatically alter kinetic parameters, consistent with a critical role of these two residues in mSlc35a3 function. A bioinformatics analysis revealed that an EXXK motif in TMH2 is highly conserved across SLC35 A subfamily members, and a 3D-homology model predicted that Glu-47 and Lys-50 are facing the central cavity of the protein.

The conundrum of UDP-Glc entrance into the yeast ER lumen.

UDP-Glc entrance into the endoplasmic reticulum (ER) of eukaryotic cells is a key step in the quality control of glycoprotein folding, a mechanism requiring transfer of a Glc residue from the nucleotide sugar (NS) to glycoprotein folding intermediates by the UDP-Glc:glycoprotein glucosyltransferase (UGGT). According to a bioinformatics search there are only eight genes in the Schizosaccharomyces pombe genome belonging to the three Pfam families to which all known nucleotide-sugar transporters (NSTs) of the secretory pathway belong. The protein products of two of them (hut1+ and yea4+) localize to the ER, those of genes gms1+, vrg4+, pet1+, pet2+ and pet3+ to the Golgi, whereas that of gms2+ has an unknown location. Here we demonstrate that (1) Δhut1 and Δgpt1 (UGGT null) mutants share several phenotypic features; (2) Δhut1 mutants show a 50% reduction in UDP-Glc transport into ER-derived membranes; (3) in vivo UDP-Glc ER entrance occurred in Δhut1Δyea4Δgms2 mutants and in cells in which Δhut1 disruption was combined with that of each of four of the genes encoding Golgi-located proteins. Therefore, disruption of all genes whose products localize to the ER or have an unknown location did not obliterate UDP-Glc ER entrance. We conclude that the hut1+ gene product is involved in UDP-Glc entrance into the ER, but that at least another as yet unknown NST displaying an unconventional sequence operates in the yeast secretory pathway. This conclusion agrees with our previous results showing that UDP-Glc entrance into the yeast ER does not follow the classical NST antiport mechanism.

The promiscuous phosphomonoestearase activity of Archaeoglobus fulgidus CopA, a thermophilic Cu+ transport ATPase.

Membrane transport P-type ATPases display two characteristic enzymatic activities: a principal ATPase activity provides the driving force for ion transport across biological membranes, whereas a promiscuous secondary activity catalyzes the hydrolysis of phosphate monoesters. This last activity is usually denoted as the phosphatase activity of P-ATPases. In the present study, we characterize the phosphatase activity of the Cu(+)-transport ATPase from Archaeglobus fulgidus (Af-CopA) and compare it with the principal ATPase activity. Our results show that the phosphatase turnover number was 20 times higher than that corresponding to the ATPase activity, but it is compensated by a high value of Km, producing a less efficient catalysis for pNPP. This secondary activity is enhanced by Mg(2+) (essential activator) and phospholipids (non-essential activator), and inhibited by salts and Cu(+). Transition state analysis of the catalyzed and noncatalyzed hydrolysis of pNPP indicates that Af-CopA enhances the reaction rates by a factor of 10(5) (ΔΔG(‡)=38 kJ/mol) mainly by reducing the enthalpy of activation (ΔΔH(‡)=30 kJ/mol), whereas the entropy of activation is less negative on the enzyme than in solution. For the ATPase activity, the decrease in the enthalpic component of the barrier is higher (ΔΔH(‡)=39 kJ/mol) and the entropic component is small on both the enzyme and in solution. These results suggest that different mechanisms are involved in the transference of the phosphoryl group of p-nitrophenyl phosphate and ATP.

Activation of Archaeoglobus fulgidus Cu(+)-ATPase CopA by cysteine.

CopA, a thermophilic ATPase from Archaeoglobus fulgidus, drives the outward movement of Cu(+) across the cell membrane. Millimolar concentration of Cys dramatically increases ( congruent with 800%) the activity of CopA and other P(IB)-type ATPases (Escherichia coli ZntA and Arabidopsis thaliana HMA2). The high affinity of CopA for metal ( congruent with 1 microM) together with the low Cu(+)-Cys K(D) (<10(-10)M) suggested a multifaceted interaction of Cys with CopA, perhaps acting as a substitute for the Cu(+) chaperone protein present in vivo. To explain the activation by the amino acid and further understand the mechanism of metal delivery to transport ATPases, Cys effects on the turnover and partial reactions of CopA were studied. 2-20 mM Cys accelerates enzyme turnover with little effect on CopA affinity for Cu(+), suggesting a metal independent activation. Furthermore, Cys activates the p-nitrophenyl phosphatase activity of CopA, even though this activity is metal independent. Cys accelerates enzyme phosphorylation and the forward dephosphorylation rates yielding higher steady state phosphoenzyme levels. The faster dephosphorylation would explain the higher enzyme turnover in the presence of Cys. The amino acid has no significant effect on low affinity ATP K(m) suggesting no changes in the E(1)<-->E(2) equilibrium. Characterization of Cu(+) transport into sealed vesicles indicates that Cys acts on the cytoplasmic side of the enzyme. However, the Cys activation of truncated CopA lacking the N-terminal metal binding domain (N-MBD) indicates that activation by Cys is independent of the regulatory N-MBD. These results suggest that Cys is a non-essential activator of CopA, interacting with the cytoplasmic side of the enzyme while this is in an E1 form. Interestingly, these effects also point out that Cu(+) can reach the cytoplasmic opening of the access path into the transmembrane transport sites either as a free metal or a Cu(+)-Cys complex.

Golgi and endoplasmic reticulum functions take place in different subcellular compartments of Entamoeba histolytica.

Entamoeba histolytica is a protozoan parasite that causes dysentery in developing countries of Africa, Asia, and Latin America. The lack of a defined Golgi apparatus in E. histolytica as well as in other protists led to the hypothesis that they had evolved prior to the acquisition of such organelle even though glycoproteins, glycolipids, and antigens have been detected, the latter of which react with antibodies against Golgi apparatus proteins of higher eukaryotes. We here provide direct evidence for Golgi apparatus-like functions in E. histolytica as well as for components of glycoprotein folding quality control. Using a combination of bioinformatic, cell biological, and biochemical approaches we have (a) cloned and expressed the E. histolytica UDP-galactose transporter in Saccharomyces cerevisiae; its K(m) for UDP-galactose is 2.9 microm; (b) characterized vesicles in an extract of the above protist, which transport UDP-galactose into their lumen with a K(m) of 2.7 microm;(c) detected galactosyltransferase activity(ies) in the lumen of the above vesicles with the K(m) for UDP-galactose, using endogenous acceptors, being 93 microm;(d) measured latent apyrase activities in the above vesicles, suggesting they are in the lumen; (e) characterized UDP-glucose transport activities in Golgi apparatus and endoplasmic reticulum-like vesicles with K(m)s for UDP-glucose of approximately 2-4 microm. Although the endoplasmic reticulum-like fraction showed UDP-glucose: glycoprotein glucosyltransferase activity, the Golgi apparatus-like fraction did not. This fraction contained other glucosyltransferases. Together, these studies demonstrate that E. histolytica has different vesicles that play a role in protein glycosylation and folding quality control, analogous to the above organellar functions of higher eukaryotes.

Loss of autoinhibition of the plasma membrane Ca(2+) pump by substitution of aspartic 170 by asparagin. A ctivation of plasma membrane calcium ATPase 4 without disruption of the interaction between the catalytic core and the C-terminal regulatory domain.

The plasma membrane calcium ATPase (PMCA) actively transports Ca(2+) from the cytosol to the extra cellular space. The C-terminal segment of the PMCA functions as an inhibitory domain by interacting with the catalytic core. Ca(2+)-calmodulin binds to the C-terminal segment and stops inhibition. Here we showed that residue Asp(170), in the putative "A" domain of human PMCA isoform 4xb, plays a critical role in autoinhibition. In the absence of calmodulin a PMCA containing a site-specific mutation of D170N had 80% of the maximum activity of the calmodulin-activated PMCA and a similar high affinity for Ca(2+). The mutation did not change the activation of the PMCA by ATP. Deletion of the C-terminal segment further downstream of the calmodulin-binding site led to an additional increase in the maximal activity of the mutant, which suggests that the mutation did not affect the inhibition because of this portion of the C-terminal segment. The calmodulin-activated PMCA was more sensitive to vanadate inhibition than the autoinhibited enzyme. In contrast, inhibition of the D170N mutant required higher concentrations of vanadate and was not affected by calmodulin. Despite its higher basal activity, the mutant had an apparent affinity for calmodulin similar to that of the wild type enzyme, and its rate of proteolysis at the C-terminal segment was still calmodulin-dependent. Altogether these results suggest that activation by mutation D170N does not involve the displacement of the calmodulin-binding autoinhibitory domain from the catalytic core and may arise directly from changes in the accessibility to the calcium-binding residues of the pump.

Pre-steady-state phosphorylation and dephosphorylation of detergent-purified plasma-membrane Ca2+-ATPase.

Pre-steady-state phosphorylation and dephosphorylation of purified and phospholipid-depleted plasma-membrane Ca(2+)-ATPase (PMCA) solubilized in the detergent polyoxyethylene 10 lauryl ether were studied at 25 degrees C. The time course of phosphorylation with ATP of the enzyme associated with Ca(2+), probably the true phosphorylation reaction, showed a fast phase (k(app) near 400 s(-1)) followed by a slow phase (k(app)=23 s(-1)). With asolectin or acidic phosphatidylinositol, the concentration of phosphoenzyme (EP) increased at as high a rate as before, passed through a maximum at 4 ms and stabilized at a steady level that was approx. half that without lipids. Calmodulin (CaM) did not change the rate of the fast phase, accelerated the slow phase (k(app)=93 s(-1)) and increased [EP] with small changes in the shape of the time course. Dephosphorylation was slow (k(app)=30 s(-1)) and insensitive to CaM. Asolectin accelerated dephosphorylation, which followed biexponential kinetics with fast (k(app)=220 s(-1)) and slow (k(app)=20 s(-1)) components. CaM stimulated the fast component by nearly 50%. The results show that the behaviour of the PMCA is complex, and suggest that acidic phospholipids and CaM activate PMCA through different mechanisms. Acceleration of dephosphorylation seems relevant during activation of the PMCA by acidic phospholipids.