Gas-phase electrophoresis on a nano-Electrospray Gas-phase Electrophoretic Mobility Molecular Analyzer (nES GEMMA) separates single-charged, native analytes according to the surface-dry particle size. A volatile electrolyte, often ammonium acetate, is a prerequisite for electrospraying. Over the years, nES GEMMA has demonstrated its unique capability to investigate (bio-)nanoparticle containing samples in respect to composition, analyte size, size distribution, and particle numbers. Virus-like particles (VLPs), being non-infectious vectors, are often employed for gene therapy applications. Focusing on adeno-associated virus 8 (AAV8) based VLPs, we investigated the response of these bionanoparticles to pH changes via nES GEMMA as ammonium acetate is known to exhibit these changes upon electrospraying. Indeed, slight yet significant differences in VLP diameters in relation to pH changes are found between empty and DNA-cargo-filled assemblies. Additionally, filled VLPs exhibit aggregation in dependence on the applied electrolyte's pH, as corroborated by atomic force microscopy. In contrast, cryogenic transmission electron microscopy did not relate to changes in the overall particle size but in the substantial particle's shape based on cargo conditions. Overall, we conclude that for VLP characterization, the pH of the applied electrolyte solution has to be closely monitored, as variations in pH might account for drastic changes in particles and VLP behavior. Likewise, extrapolation of VLP behavior from empty to filled particles has to be carried out with caution.

Over the past decades, virus-like particle (VLP)-based gene therapy (GT) evolved as a promising approach to cure inherited diseases or cancer. Tremendous costs due to inefficient production processes remain one of the key challenges despite considerable efforts to improve titers. This review aims to link genome-scale metabolic models (GSMMs) to cell lines used for VLP synthesis for the first time. We summarize recent advances and challenges of GSMMs for Chinese hamster ovary (CHO) cells and provide an overview of potential cell lines used in GT. Although GSMMs in CHO cells led to significant improvements in growth rates and recombinant protein (RP)-production, no GSMM has been established for VLP production so far. To facilitate the generation of GSMM for these cell lines we further provide an overview of existing omics data and the highest production titers so far reported.

Adeno-associated virus (AAV)-based virus-like particles (VLPs) are thriving vectors of choice in the biopharmaceutical field of gene therapy. Here, a method to investigate purified AAV serotype 8 (AAV8) batches via a nanoelectrospray gas-phase mobility molecular analyzer (nES GEMMA), also known as an nES differential mobility analyzer, is presented. Indeed, due to AAV’s double-digit nanometer scale, nES GEMMA is an excellently suited technique to determine the surface-dry particle size termed electrophoretic mobility diameter of such VLPs in their native state at atmospheric pressure and with particle-number-based detection. Moreover, asymmetric flow field-flow fractionation (AF4, also known as AFFFF) and atomic force microscopy (AFM) techniques were employed as orthogonal techniques for VLP characterization. In addition, AF4 was implemented to size-separate as well as to enrich and collect fractions of AAV8 VLPs after inducing analyte aggregation in the liquid phase. Bionanoparticle aggregation was achieved by a combination of heat and shear stress. These fractions were later analyzed with nES GEMMA (in the gas phase) and AFM (on a solid surface). Both techniques confirm the presence of dimers, trimers, and putative VLP oligomers. Last, AFM reveals even larger AAV8 VLP aggregates, which were not detectable by nES GEMMA because their heterogeneity combined with low abundance was below the limit of detection of the instrument. Hence, the combination of the employed orthogonal sizing methods with the separation technique AF4 allow a comprehensive characterization of AAV8 VLPs applied as vectors.

Abstract Virus‐like particles (VLPs) are proteinaceous shells derived from viruses lacking any viral genomic material. Adeno‐associated virus (AAV) is a non‐enveloped icosahedral virus used as VLP delivery system in gene therapy (GT). Its success as vehicle for GT is due to its selective tropism, high level of transduction, and low immunogenicity. In this study, two preparations of AAV serotype 8 (AAV8) VLPs either carrying or lacking completely genomic cargo (i.e., non‐viral ssDNA) have been investigated by means of a native nano‐electrospray gas‐phase electrophoretic mobility molecular analyzer (GEMMA) (native nES GEMMA) and native nano‐electrospray ionization quadrupole reflectron time‐of‐flight mass spectrometry (MS) (native nESI QRTOF MS). nES GEMMA is based on electrophoretic mobility principles: single‐charge nanoparticles (NPs), that is, AAV8 particle, are separated in a laminar sheath flow of dry, particle‐free air and a tunable orthogonal electric field. Thus, the electrophoretic mobility diameter (EMD) of a bio‐NP (i.e., diameter of globular nano‐objects) is obtained at atmospheric pressure, which can be converted into its MW based on a correlation. First is the native nESI QRTOF. MS's goal is to keep the native biological conformation of an analyte during the passage into the vacuum. Subsequently, highly accurate MW values are obtained from multiple‐charged species after deconvolution. However, once applied to the analysis of megadalton species, native MS is challenging and requires customized instrumental modifications not readily available on standard devices. Hence, the analysis of AAV8 VLPs via native MS in our hands did not produce a defined charge state assignment, that is, charge deconvolution for exact MW determination was not possible. Nonetheless, the method we present is capable to estimate the MW of VLPs by combining the results from native nES GEMMA and native ESI QRTOF MS. In detail, our findings show a MW of 3.7 and 5.0 MDa for AAV8 VLPs either lacking or carrying an engineered genome, respectively.

Post-translational glycosylation determines the pharmacodynamic and pharmacokinetic properties of therapeutic proteins. While desialylation of FVIII/VWF (asialo-FVIII/VWF) has a significantly reduced half-life compared to normal FVIII/VWF complex (Sodetz et al 1977), we introduced additional sialic acids on recombinant (r) FVIII to obtain the opposite: prolonged survival of rFVIII in the circulation. Additional sialic acid residues were introduced to rFVIII by covalently binding polysialic acid (PSA) to full length (FL) rFVIII. The resulting drug candidate BAX 826, polysialylated human rFVIII, is manufactured from octocog alfa which is expressed in Chinese Hamster Ovary Cells by a plasma/albumin free cell culture method and is the active substance in Baxalta´s licensed product ADVATE.

Background & Methods Recombinant factor VII (rFVII), the precursor molecule for recombinant activated FVII (rFVIIa), is, due to its need for complex post translational modifications, produced in mammalian cells. To evaluate the suitability of a human cell line in order to produce rFVII with post-translational modifications as close as possible to pdFVII, we compared the biochemical properties of rFVII synthesized in human embryonic kidney-derived (HEK)293 cells (HEK293rFVII) with those of rFVII expressed in Chinese hamster ovary (CHO, CHOrFVII) and baby hamster kidney (BHK, BHKrFVII) cells, and also with those of plasma derived FVII (pdFVII), using various analytical methods. rFVII was purified from selected production clones derived from BHK, CHO, and HEK293 cells after stable transfection, and rFVII isolates were analyzed for protein activity, impurities and post-translational modifications. Results & Discussion The analytical results showed no apparent gross differences between the various FVII proteins, except in their N-linked glycosylation pattern. Most N-glycans found on rFVII produced in HEK293 cells were not detected on rFVII from CHO and BHK cells, or, somewhat unexpectedly, on pdFVII; all other protein features were similar. HEK293rFVII glycans were mainly characterized by a higher structural variety and a lower degree of terminal sialylation, and a high amount of terminal N-acetyl galactosamines (GalNAc). All HEK293rFVII oligosaccharides contained one or more fucoses (Fuc), as well as hybrid and high mannose (Man) structures. Conclusions From all rFVII isolates investigated, CHOrFVII contained the highest degree of sialylation and no terminal GalNAc, and CHO cells were therefore assumed to be the best option for the production of rFVII.

IntroductionClearance mechanisms for rFVII (or the active enzymerFVIIa) and rFIX are influenced by post-translationalmodifications, especially N-glycosylation. This should beconsidered when choosing a recombinant expressionsystem in view of the varying ability of frequently usedcell lines to perform modifications similar to humanproteins.Differences in the pharmacokinetic properties ofrecombinant FVIIa versus plasma-derived (pd)FVII ordesialylated rFVIIa are known for human FVII(a). AsialorFVIIa clears quickest, whereas pdFVII, having a higherdegree of sialylation, is cleared to a lesser extent [1]. Inthe case of FIX, the degrees ofserine phosphorylationand tyrosine sulfation in the activation peptide havebeen postulated to influence pharmacokinetic behavior,especially in vivo recovery [2].We chose CHO, BHK and HEK293 cells for expres-sion of rFVII to compare post-translational proteinmodifications, and HEK293-derived cell lines to generatehighly phosphorylated and sulfated rFIX for in vivo stu-dies. rFIX from the same clone and production run waspurified using two different down-stream processes: Thefirst to enrich high phosphorylated and sulfated protein,the second to purify total rFIX at high yield. TheseHEK293-derived rFIX isoforms were compared withCHOrFIX and pdFIX in a pharmacokinetic study in FIXknock-out mice.Materials and methods

The striking feature of carbohydrates is their constitutional, conformational and configurational diversity. Biology has harnessed this diversity and manipulates carbohydrate residues in a variety of ways, one of which is epimerization. RmlC catalyzes the epimerization of the C3' and C5' positions of dTDP-6-deoxy-D-xylo-4-hexulose, forming dTDP-6-deoxy-L-lyxo-4-hexulose. RmlC is the third enzyme of the rhamnose pathway, and represents a validated anti-bacterial drug target. Although several structures of the enzyme have been reported, the mechanism and the nature of the intermediates have remained obscure. Despite its relatively small size (22 kDa), RmlC catalyzes four stereospecific proton transfers and the substrate undergoes a major conformational change during the course of the transformation. Here we report the structure of RmlC from several organisms in complex with product and product mimics. We have probed site-directed mutants by assay and by deuterium exchange. The combination of structural and biochemical data has allowed us to assign key residues and identify the conformation of the carbohydrate during turnover. Clear knowledge of the chemical structure of RmlC reaction intermediates may offer new opportunities for rational drug design.

BACKGROUND:Isoelectric focusing (IEF) of alpha(1)-proteinase inhibitor (A1PI) shows that commercial products and plasma have different glycoisoform band patterns. Those in Aralast (Grifols Biologicals) reflect an anodal shift of glycoisoforms, which has caused concern. The protein, including glycoproteomic analyses, and structural features of A1PI products were investigated by state-of-the-art techniques.STUDY DESIGN AND METHODS:Batches from Aralast, Prolastin (Bayer), and Zemaira (Aventis Behring LLC) were analyzed by high-resolution IEF and high-performance size-exclusion chromatography (HP-SEC). Preparative separated isoforms from IEF were further purified by chromatography and subjected to mass spectrometry for sequence analyses, peptide mapping, and glycosylation analysis. Deamidation was quantified by enzymatic isoaspartate detection. Multiple sequence alignments and structural bioinformatics analyses were performed.RESULTS:In HP-SEC, Prolastin had the highest aggregate content at approximately 30 percent. Isoforms from all products purified by high-resolution IEF were sequenced with an amino acid coverage of more than 98 percent. Deamidation of Asn116 and Asn314 in A1PI was to found to some extent in all products and confirmed quantitatively by enzymatic analysis. There were no signs of methionine oxidation. Cys232 was found to be cysteinylated in A1PI in Prolastin and Aralast as in plasma, but not in Zemaira. All products showed truncation of the C-terminal lysine. Intact A1PI concentrates contained mainly diantennary, disialylated and smaller amounts of triantennary, trisialylated N-glycans. The percentage of fucosylation was similar in all products. Site-specific glycan analysis revealed bands M6 contained only diantennary glycans, whereas the more acidic bands M4 and M2 also carried triantennary structures. The most acidic isoforms, M2 in Prolastin and Zemaira and M0 in Aralast, additionally exhibited tetraantennary N-glycans.CONCLUSION:Protein chemical characterization of A1PI showed that all A1PI products to some extent differ from A1PI circulating in human plasma. Bioinformatic analysis indicated that removal of C-terminal Lys394 and cysteinylation of Cys232 are unlikely to affect structure and/or function of A1PI but cysteinylation may influence interaction between A1PI and its physiologic ligands. Aralast, Prolastin, and Zemaira contain the same set of N-glycans in the same ratios as those in normal human plasma A1PI. Tri- and tetraantennary structures are responsible for the partitioning into IEF isoforms, with the migration shift of Aralast not being due to any difference in the N-glycosylation, but to the partial loss of the C-terminal lysine.

The cell surface of Aneurinibacillus thermoaerophilus DSM 10155 is covered with a square surface (S)-layer glycoprotein lattice. This S-layer glycoprotein, which was extracted with aqueous buffers after a freeze-thaw cycle of the bacterial cells, is the only completely water-soluble S-layer glycoprotein to be reported to date. The purified S-layer glycoprotein preparation had an overall carbohydrate content of 19%. Detailed chemical investigations indicated that the S-layer O-glycans of previously established structure accounted for 13% of total glycosylation. The remainder could be attributed to a peptidoglycan-associated secondary cell wall polymer. Structure analysis was performed using purified secondary cell wall polymer-peptidoglycan complexes. NMR spectroscopy revealed the first biantennary secondary cell wall polymer from the domain Bacteria, with the structure alpha-L-Glc p NAc-(1-->3)-beta-L-Man p NAc-(1-->4)-beta-L-Gal p NAc-(1-->3)-alpha-L-Glc p NAc-(1-->3)-beta-L-Man p NAc-(1-->4)-beta-L-Gal p NAc-(1-->3)-alpha-L-Glc p NAc-(1-->4)-[alpha-L-Glc p NAc-(1-->3)-beta-L-Man p NAc-(1-->4)-beta-L-Gal p NAc-(1-->3)-alpha-L-Glc p NAc-(1-->3)-beta-L-Man p NAc-(1-->4)-beta-L-Gal p NAc-(1-->3)-alpha-L-Glc p NAc-(1-->3)]-beta-L-Man p NAc-(1-->3)-alpha-L-Glc p NAc-(1-->3)-beta-L-Man p NAc-(1-->3)-alpha-L-Glc p NAc-(1-->3)-alpha-L-Glc p NAc-(1-->O)-PO(2)(-)-O-PO(2)(-)-(O-->6)-MurNAc- (where MurNAc is N -acetylmuramic acid). The neutral polysaccharide is linked via a pyrophosphate bond to the C-6 atom of every fourth N -acetylmuramic acid residue, in average, of the A1gamma-type peptidoglycan. In vivo, the biantennary polymer anchored the S-layer glycoprotein very effectively to the cell wall, probably due to the doubling of motifs for a proposed lectin-like binding between the polymer and the N-terminus of the S-layer protein. When the cellular support was removed during S-layer glycoprotein isolation, the co-purified polymer mediated the solubility of the S-layer glycoprotein in vitro. Initial crystallization experiments performed with the soluble S-layer glycoprotein revealed that the assembly property could be restored upon dissociation of the polymer by the addition of poly(ethylene glycols). The formed two-dimensional crystalline S-layer self-assembly products exhibited the same lattice symmetry as observed on intact bacterial cells.

dTDP-6-deoxy-L-lyxo-4-hexulose reductase (RmlD) catalyzes the final step in the conversion of dTDP-D-glucose to dTDP-L-rhamnose in an NAD(P)H- and Mg2+-dependent reaction. L-rhamnose biosynthesis is an antibacterial target. The structure of RmlD from Salmonella enterica serovar Typhimurium has been determined, and complexes with NADH, NADPH, and dTDP-L-rhamnose are reported. RmlD differs from other short chain dehydrogenases in that it has a novel dimer interface that contains Mg2+. Enzyme catalysis involves hydride transfer from the nicotinamide ring of the cofactor to the C4′-carbonyl group of the substrate. The substrate is activated through protonation by a conserved tyrosine. NAD(P)H is bound in a solvent-exposed cleft, allowing facile replacement. We suggest a novel role for the conserved serine/threonine residue of the catalytic triad of SDR enzymes.

ABSTRACT The steps involved in the biosynthesis of the ADP- l - glycero -β- d - manno- heptose (ADP- l -β- d -heptose) precursor of the inner core lipopolysaccharide (LPS) have not been completely elucidated. In this work, we have purified the enzymes involved in catalyzing the intermediate steps leading to the synthesis of ADP- d -β- d -heptose and have biochemically characterized the reaction products by high-performance anion-exchange chromatography. We have also constructed a deletion in a novel gene, gmhB (formerly yaeD ), which results in the formation of an altered LPS core. This mutation confirms that the GmhB protein is required for the formation of ADP- d -β- d -heptose. Our results demonstrate that the synthesis of ADP- d -β- d -heptose in Escherichia coli requires three proteins, GmhA (sedoheptulose 7-phosphate isomerase), HldE (bifunctional d -β- d -heptose 7-phosphate kinase/ d -β- d -heptose 1-phosphate adenylyltransferase), and GmhB ( d , d -heptose 1,7-bisphosphate phosphatase), as well as ATP and the ketose phosphate precursor sedoheptulose 7-phosphate. A previously characterized epimerase, formerly named WaaD (RfaD) and now renamed HldD, completes the pathway to form the ADP- l -β- d -heptose precursor utilized in the assembly of inner core LPS.

dTDP-D-glucose 4,6-dehydratase (RmlB) was first identified in the L-rhamnose biosynthetic pathway, where it catalyzes the conversion of dTDP-D-glucose into dTDP-4-keto-6-deoxy-D-glucose. The structures of RmlB from Salmonella enterica serovar Typhimurium in complex with substrate deoxythymidine 5′-diphospho-D-glucose (dTDP-D-glucose) and deoxythymidine 5′-diphosphate (dTDP), and RmlB from Streptococcus suis serotype 2 in complex with dTDP-D-glucose, dTDP, and deoxythymidine 5′-diphospho-D-pyrano-xylose (dTDP-xylose) have all been solved at resolutions between 1.8 Å and 2.4 Å. The structures show that the active sites are highly conserved. Importantly, the structures show that the active site tyrosine functions directly as the active site base, and an aspartic and glutamic acid pairing accomplishes the dehydration step of the enzyme mechanism. We conclude that the substrate is required to move within the active site to complete the catalytic cycle and that this movement is driven by the elimination of water. The results provide insight into members of the SDR superfamily.

ABSTRACT The glycan chains of the surface layer (S-layer) glycoprotein from the gram-positive, thermophilic bacterium Aneurinibacillus (formerly Bacillus ) thermoaerophilus strain DSM 10155 are composed of l -rhamnose- and d - glycero - d - manno -heptose-containing disaccharide repeating units which are linked to the S-layer polypeptide via core structures that have variable lengths and novel O-glycosidic linkages. In this work we investigated the enzymes involved in the biosynthesis of thymidine diphospho- l -rhamnose (dTDP- l -rhamnose) and their specific properties. Comparable to lipopolysaccharide O-antigen biosynthesis in gram-negative bacteria, dTDP- l -rhamnose is synthesized in a four-step reaction sequence from dTTP and glucose 1-phosphate by the enzymes glucose-1-phosphate thymidylyltransferase (RmlA), dTDP- d -glucose 4,6-dehydratase (RmlB), dTDP-4-dehydrorhamnose 3,5-epimerase (RmlC), and dTDP-4-dehydrorhamnose reductase (RmlD). The rhamnose biosynthesis operon from A. thermoaerophilus DSM 10155 was sequenced, and the genes were overexpressed in Escherichia coli . Compared to purified enterobacterial Rml enzymes, the enzymes from the gram-positive strain show remarkably increased thermostability, a property which is particularly interesting for high-throughput screening and enzymatic synthesis. The closely related strain A. thermoaerophilus L420-91 T produces d -rhamnose- and 3-acetamido-3,6-dideoxy- d -galactose-containing S-layer glycan chains. Comparison of the enzyme activity patterns in A. thermoaerophilus strains DSM 10155 and L420-91 T for l -rhamnose and d -rhamnose biosynthesis indicated that the enzymes are differentially expressed during S-layer glycan biosynthesis and that A. thermoaerophilus L420-91 T is not able to synthesize dTDP- l -rhamnose. These findings confirm that in each strain the enzymes act specifically on S-layer glycoprotein glycan formation.