Glycosylation and antiproliferative activity of hyperglycosylated IFN-α2 potentiate HEK293 cells as biofactories

A COVID-19 vaccine candidate based on SARS-CoV-2 spike protein and immune-stimulating complexes

Spike protein from SARS-CoV-2, the etiologic agent of the COVID-19 pandemic disease, constitutes a structural protein that proved to be the main responsible for neutralizing antibody production. Thus, its sequence is highly considered for the design of candidate vaccines. Animal cell culture represents the best option for the production of subunit vaccines based on recombinant proteins since they introduce post-translational modifications that are important to mimic the natural antigenic epitopes. Particularly, the human cell line HEK293T has been explored and used for the production of biotherapeutics since the products derived from them present human-like post-translational modifications that are important for the protein's activity and immunogenicity. The aim of this study was to produce and characterize a potential vaccine for COVID-19 based on the spike ectodomain (S-ED) of SARS-CoV-2 and two different adjuvants: aluminum hydroxide (AH) and immune-stimulating complexes (ISCOMs). The S-ED was produced in sHEK293T cells using a 1-L stirred tank bioreactor operated in perfusion mode and purified. S-ED characterization revealed the expected size and morphology. High N-glycan content was confirmed. S-ED-specific binding with the hACE2 (human angiotensin-converting enzyme 2) receptor was verified. The immunogenicity of S-ED was evaluated using AH and ISCOMs. Both formulations demonstrated the presence of anti-RBD antibodies in the plasma of immunized mice, being significantly higher for the latter adjuvant. Also, higher levels of IFN-γ and IL-4 were detected after the ex vivo immune stimulation of spleen-derived MNCs from ISCOMs immunized mice. Further analysis confirmed that S-ED/ISCOMs elicit neutralizing antibodies against SARS-CoV-2. KEY POINTS: Trimeric SARS-CoV-2 S-ED was produced in stable recombinant sHEK cells in serum-free medium. A novel S-ED vaccine formulation induced potent humoral and cellular immunity. S-ED formulated with ISCOMs adjuvant elicited a highly neutralizing antibody titer.

Changes in antibody binding and functionality after humanizing a murine scFv anti-IFN-α2: From in silico studies to experimental analysis

The structural and dynamic changes introduced during antibody humanization continue to be a topic open to new contributions. For this reason, the study of structural and functional changes of a murine scFv (mu.scFv) anti-rhIFN-α2b after humanization was carried out. As it was shown by long molecular dynamics simulations and circular dichroism analysis, changes in primary sequence affected the tertiary structure of the humanized scFv (hz.scFv): the position of the variable domain of light chain (VL) respective to the variable domain of heavy chain (VH) in each scFv molecule was different. This change mainly impacted on conformation and dynamics of the complementarity-determining region 3 of VH (CDR-H3) which led to changes in the specificity and affinity of humanized scFv (hz.scFv). These observations agree with experimental results that showed a decrease in the antigen-binding strength of hz.scFv, and different capacities of these molecules to neutralize the in vitro rhIFN-α2b biological activity. Besides, experimental studies to characterize antigen-antibody binding showed that mu.scFv and hz.scFv bind to the same antigen area and recognize a conformational epitope, which is evidence of docking results. Finally, the differences between these molecules to neutralize the in vitro rhIFN-α2b biological activity were described as a consequence of the blockade of certain functionally relevant amino acids of the cytokine, after scFv binding. All these observations confirmed that humanization affected the affinity and specificity of hz.scFv and pointed out that two specific changes in the frameworks would be responsible.

Development of a suitable manufacturing process for production of a bioactive recombinant equine chorionic gonadotropin (reCG) in CHO-K1 cells

Equine chorionic gonadotropin (eCG) is a heterodimeric glycoprotein hormone produced by pregnant mares that has been used to improve reproductive performance in different domestic species. Several strategies to produce the hormone in a recombinant way have been reported; nevertheless, no approach has been able to produce a recombinant eCG (reCG) with significant in vivo bioactivity or in sufficient quantities for commercial purposes. For this reason, the only current product available on the market consists of partially purified preparations from serum of pregnant mares (PMSG). Herein, we describe a highly efficient process based on third-generation lentiviral vectors as delivery method for the production of reCG in suspension CHO-K1 cells, with productivities above 20 IU 10⁶ cell^-1.d^-1 and 70% purification yields after one purification step. Importantly, reCG demonstrated biological activity in cattle, since around 30 μg of reCG were needed to exert the same biologic effect of 400 IU of PMSG in an ovulation synchronization protocol. The results obtained demonstrate that the developed strategy represents an attractive option for the production of reCG and constitutes an auspicious alternative for the replacement of animals as a source of PMSG.

Pharmacokinetics Versus In Vitro Antiproliferative Potency to Design a Novel Hyperglycosylated hIFN-α2 Biobetter

PURPOSE

IFN4N is a glycoengineered version of recombinant human interferon alpha 2 (rhIFN-α2) that was modified to exhibit four N-glycosylation sites. It shows reduced in vitro specific biological activity (SBA) mainly due to R23 mutation by N23. However, it has improved pharmacokinetics and led to a high in vivo antitumor activity in mice. In order to prepare a new IFN-based biobetter, this work compares the influence of glycosylation (affecting pharmacokinetics) with the in vitro antiproliferative SBA on the in vivo efficacy.

METHODS

Based on IFN4N, three groups of muteins were designed, produced, and characterized. Group A: variants with the same glycosylation degree (4N) but higher in vitro antiproliferative SBA (R23 restored); group B: muteins with higher glycosylation degree (5N) but similar in vitro antiproliferative activity; and group C: variants with improved glycosylation (5N and 6N) and in vitro antiproliferative bioactivity.

RESULTS

Glycoengineering was successful for improving pharmacokinetics, and R23 restoration considerably increased in vitro antiproliferative activity of new muteins compared to IFN4N. Hyperglycosylation was able to improve the in vivo efficacy similarly to or even better than R23 restoration. Additionally, the highest glycosylated mutein exhibited the lowest immunogenicity.

CONCLUSIONS

Hyperglycosylation constitutes a successful strategy to prepare a novel IFN biobetter.

Development of a high cell density transient CHO platform yielding mAb titers greater than 2 g/L in only 7 days

We developed a simple transient Chinese Hamster Ovary expression platform. Titers for a random panel of 20 clinical monoclonal antibodies (mAbs) ranged from 0.6 to 2.7 g/L after 7 days. Two factors were the key in obtaining these high titers. First, we utilized an extremely high starting cell density (20 million cells/ml), and then arrested further cell growth by employing mild hypothermic conditions (32°C). Second, we performed a 6-variable Design of Experiments to find optimal concentrations of plasmid DNA (coding DNA), boost DNA (DNA encoding the XBP1S transcription factor), transfection reagent (polyethylenimine [PEI]), and nutrient feed amounts. High coding DNA concentrations (12.5 mg/L) were found to be optimal. We therefore diluted expensive coding DNA with inexpensive inert filler DNA (herring sperm DNA). Reducing the coding DNA concentration by 70% from 12.5 to 3.75 mg/L did not meaningfully reduce mAb titers. Titers for the same panel of 20 clinical mAbs ranged from 0.7 to 2.2 g/L after reducing the coding DNA concentration to 3.75 mg/L. Finally, we found that titer and product quality attributes were similar for a clinical mAb (rituximab) expressed at very different scales (volumes ranging from 3 ml to 2 L).

Bifunctional GM-CSF-derived peptides as tools for O-glycoengineering and protein tagging

Rapid development of effective biotherapeutics has been a concern during the last couple decades. In our work we designed two novel peptide tags, GMOP and mGMOP, derived from the N-terminal region of human granulocyte and macrophage colony stimulating factor (hGM-CSF), which contain four and six potential O-glycosylation sites, respectively. These peptide tags were fused to the N-terminus of human interferon-α2b (hIFN-α2b), a therapeutic antiviral and antiproliferative protein rapidly cleared from circulation. Two new molecules were obtained which, consistently with the presence of O-glycans, showed higher molecular masses, more negatively charged isoforms, and higher sialic acid content compared to wild-type IFN. In vitro bioactivity of purified chimeras revealed a similar antiviral specific biological activity (SBA) compared to unmodified IFN. A reduction of antiproliferative SBA was only observed for mGMOP-IFN. Pharmacokinetic studies in rats showed a notable improvement in terminal half-life (t_1/2elim) (3.3 and 2.8 times-longer) and a marked reduction of the apparent clearance (CLapp, 3.7 and 4.1-fold lower for GMOP-IFN and mGMOP-IFN in comparison with native IFN, respectively). Furthermore, the in vitro thermal and plasma stability of both proteins was improved. Finally, a monoclonal antibody (mAb) that recognizes an N-terminal GM-CSF epitope was able to bind both chimeras in western blots and ELISAs. This demonstrates the potential of both peptides to behave as bifunctional tags to create novel long-acting biotherapeutics and to facilitate detection and purification.

Impact of glycan cloud on the B-cell epitope prediction of SARS-CoV-2 Spike protein

The SARS-CoV-2 outbreak originated in China in late 2019 and has since spread to pandemic proportions. Diagnostics, therapeutics and vaccines are urgently needed. We model the trimeric Spike protein, including flexible loops and all N-glycosylation sites, in order to elucidate accessible epitopes for antibody-based diagnostics, therapeutics and vaccine development. Based on published experimental data, six homogeneous glycosylation patterns and two heterogeneous ones were used for the analysis. The glycan chains alter the accessible surface areas on the S-protein, impeding antibody-antigen recognition. In presence of glycan, epitopes on the S1 subunit, that notably contains the receptor binding domain, remain mostly accessible to antibodies while those present on the S2 subunit are predominantly inaccessible. We identify 28 B-cell epitopes in the Spike structure and group them as non-affected by the glycan cloud versus those which are strongly masked by the glycan cloud, resulting in a list of favourable epitopes as targets for vaccine development, antibody-based therapy and diagnostics.

Glycomics and glycoproteomics of viruses: Mass spectrometry applications and insights toward structure-function relationships

The advancement of viral glycomics has paralleled that of the mass spectrometry glycomics toolbox. In some regard the glycoproteins studied have provided the impetus for this advancement. Viral proteins are often highly glycosylated, especially those targeted by the host immune system. Glycosylation tends to be dynamic over time as viruses propagate in host populations leading to increased number of and/or "movement" of glycosylation sites in response to the immune system and other pressures. This relationship can lead to highly glycosylated, difficult to analyze glycoproteins that challenge the capabilities of modern mass spectrometry. In this review, we briefly discuss five general areas where glycosylation is important in the viral niche and how mass spectrometry has been used to reveal key information regarding structure-function relationships between viral glycoproteins and host cells. We describe the recent past and current glycomics toolbox used in these analyses and give examples of how the requirement to analyze these complex glycoproteins has provided the incentive for some advances seen in glycomics mass spectrometry. A general overview of viral glycomics, special cases, mass spectrometry methods and work-flows, informatics and complementary chemical techniques currently used are discussed. © 2020 The Authors. Mass Spectrometry Reviews published by John Wiley & Sons Ltd. Mass Spec Rev.

Expression of glycosylated human prolactin in HEK293 cells and related N-glycan composition analysis

Prolactin (PRL) is a hormone produced by the pituitary gland with innumerable functions, such as lactation, reproduction, osmotic and immune regulation. The present work describes the synthesis of hPRL in human embryonic kidney (HEK293) cells, transiently transfected with the pcDNA-3.4-TOPO^® vector carrying the hPRL cDNA. A concentration of ~ 20 mg/L, including glycosylated (G-hPRL) and non-glycosylated (NG-hPRL) human prolactin, was obtained, with ~ 19% of G-hPRL, which is higher than that observed in CHO-derived hPRL (~ 10%) and falling within the wide range of 5–30% reported for pituitary-derived hPRL. N-Glycoprofiling analysis of G-hPRL provided: (i) identification of each N-glycan structure and relative intensity; (ii) average N-glycan mass; (iii) molecular mass of the whole glycoprotein and relative carbohydrate mass fraction; (iv) mass fraction of each monosaccharide. The data obtained were compared to pituitary- and CHO-derived G-hPRL. The whole MM of HEK-derived G-hPRL, determined via MALDI–TOF-MS, was 25,123 Da, which is 0.88% higher than pit- and 0.61% higher than CHO-derived G-hPRL. The main difference with the latter was due to sialylation, which was ~ sevenfold lower, but slightly higher than that observed in native G-hPRL. The “in vitro” bioactivity of HEK-G-hPRL was ~ fourfold lower than that of native G-hPRL, with which it had in common also the number of N-glycan structures.

Human thyroid-stimulating hormone synthesis in human embryonic kidney cells and related N-glycoprofiling analysis for carbohydrate composition determination

A strain of embryonic human kidney cells (HEK293) was transiently co-transfected with the expression vectors coding for the α- and β-subunits of human thyroid-stimulating hormone (hTSH), and, for the first time, a human cell-derived recombinant hTSH was synthesized and extensively characterized. The purification strategy involving two steps provided an overall yield of 55% and a purity level > 90%. The purified material (hTSH-HEK) was analyzed and compared to a CHO-derived recombinant preparation (hTSH-CHO) and to a pituitary-derived (hTSH-Pit) preparation. The three preparations showed an equivalent purity (> 95%) with a hTSH-HEK molecular mass 2.1% lower than that of hTSH-CHO and 2.7% higher than that of hTSH-Pit. Remarkable differences were found in the carbohydrate moiety, the lowest sialic acid content and highest fucose content being observed in hTSH-HEK. In vivo biological activity was confirmed for the three preparations, the hTSH-HEK bioactivity being 39 and 16% lower than those of hTSH-CHO and hTSH-Pit, respectively. The hTSH-HEK circulatory half-life (t _1/2) was also shorter than those of hTSH-CHO (1.5-fold) and hTSH-Pit (1.2-fold). According to these findings, HEK-293-derived hTSH can be considered to be useful for clinical applications, in view as well of its human origin and particular carbohydrate composition.