Identification and CRISPR/Cas9 Inactivation of the C1s Protease Responsible for Proteolysis of Recombinant Proteins Produced in CHO Cells.

Proteolysis associated with recombinant protein expression in Chinese Hamster Ovary (CHO) cells has hindered the development of biologics including HIV vaccines. When expressed in CHO cells, the recombinant HIV envelope protein, gp120, undergoes proteolytic clipping by a serine protease at a key epitope recognized by neutralizing antibodies. The problem is particularly acute for envelope proteins from clade B viruses that represent the major genetic subtype circulating in much of the developed world, including the US and Europe. In this paper, we have identified complement Component 1's (C1s), a serine protease from the complement cascade, as the protease responsible for the proteolysis of gp120 in CHO cells. CRISPR/Cas9 knockout of the C1s protease in a CHO cell line was shown to eliminate the proteolytic activity against the recombinantly expressed gp120. In addition, the C1s-/- MGAT1- CHO cell line, with the C1s protease and the MGAT1 glycosyltransferase knocked out, enabled the production of unclipped gp120 from a clade B isolate (BaL-rgp120) and enriched for mannose-5 glycans on gp120 that are required for the binding of multiple broadly neutralizing monoclonal antibodies (bN-mAbs). The availability of this technology will allow for the scale-up and testing of multiple vaccine concepts in regions of the world where clade B viruses are in circulation. Furthermore, the proteolysis issues caused by the C1s protease suggests a broader need for a C1s-deficient CHO cell line to express other recombinant proteins that are susceptible to serine protease activity in CHO cells. Similarly, the workflow described here to identify and knockout C1s in a CHO cell line can be applied to remedy the proteolysis of biologics by other CHO proteases.

Development of a Stable MGAT1- CHO Cell Line to Produce Clade C gp120 With Improved Binding to Broadly Neutralizing Antibodies.

The high rate of new HIV infections, particularly in Sub-Saharan Africa, emphasizes the need for a safe and effective vaccine to prevent acquired immunodeficiency syndrome (AIDS). To date, the only HIV vaccine trial that has exhibited protective efficacy in humans was the RV144 study completed in Thailand. The finding that protection correlated with antibodies to gp120 suggested that increasing the quality or magnitude of the antibody response that recognize gp120 might improve the modest yet significant protection (31.2%) achieved with this immunization regimen. However, the large-scale production of rgp120 suitable for clinical trials has been challenging due, in part, to low productivity and difficulties in purification. Moreover, the antigens that are currently available were produced largely by the same technology used in the early 1990s and fail to incorporate unique carbohydrates presented on HIV virions required for the binding of several major families of broadly neutralizing antibodies (bNAbs). Here we describe the development of a high-yielding CHO cell line expressing rgp120 from a clade C isolate (TZ97008), representative of the predominant circulating HIV subtype in Southern Africa and Southeast Asia. This cell line, produced using robotic selection, expresses high levels (1.2 g/L) of the TZ97008 rgp120 antigen that incorporates oligomannose glycans required for binding to multiple glycan dependent bNAbs. The resulting rgp120 displays a lower degree of net charge and glycoform heterogeneity as compared to rgp120s produced in normal CHO cells. This homogeneity in net charge facilitates purification by filtration and ion exchange chromatography methods, eliminating the need for expensive custom-made lectin, or immunoaffinity columns. The results described herein document the availability of a novel cell line for the large-scale production of clade C gp120 for clinical trials. Finally, the strategy used to produce a TZ97008 gp120 in the MGAT- CHO cell line can be applied to the production of other candidate HIV vaccines.

CRISPR/Cas9 gene editing for the creation of an MGAT1-deficient CHO cell line to control HIV-1 vaccine glycosylation.

Over the last decade, multiple broadly neutralizing monoclonal antibodies (bN-mAbs) to the HIV-1 envelope protein (Env) gp120 have been described. Many of these recognize epitopes consisting of both amino acid and glycan residues. Moreover, the glycans required for binding of these bN-mAbs are early intermediates in the N-linked glycosylation pathway. This type of glycosylation substantially alters the mass and net charge of Envs compared to molecules with the same amino acid sequence but possessing mature, complex (sialic acid-containing) carbohydrates. Since cell lines suitable for biopharmaceutical production that limit N-linked glycosylation to mannose-5 (Man5) or earlier intermediates are not readily available, the production of vaccine immunogens displaying these glycan-dependent epitopes has been challenging. Here, we report the development of a stable suspension-adapted Chinese hamster ovary (CHO) cell line that limits glycosylation to Man5 and earlier intermediates. This cell line was created using the clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR-associated protein 9 (Cas9) gene editing system and contains a mutation that inactivates the gene encoding Mannosyl (Alpha-1,3-)-Glycoprotein Beta-1,2-N-Acetylglucosaminyltransferase (MGAT1). Monomeric gp120s produced in the MGAT1- CHO cell line exhibit improved binding to prototypic glycan-dependent bN-mAbs directed to the V1/V2 domain (e.g., PG9) and the V3 stem (e.g., PGT128 and 10-1074) while preserving the structure of the important glycan-independent epitopes (e.g., VRC01). The ability of the MGAT1- CHO cell line to limit glycosylation to early intermediates in the N-linked glycosylation pathway without impairing the doubling time or ability to grow at high cell densities suggests that it will be a useful substrate for the biopharmaceutical production of HIV-1 vaccine immunogens.

Robotic selection for the rapid development of stable CHO cell lines for HIV vaccine production.

The production of envelope glycoproteins (Envs) for use as HIV vaccines is challenging. The yield of Envs expressed in stable Chinese Hamster Ovary (CHO) cell lines is typically 10-100 fold lower than other glycoproteins of pharmaceutical interest. Moreover, Envs produced in CHO cells are typically enriched for sialic acid containing glycans compared to virus associated Envs that possess mainly high-mannose carbohydrates. This difference alters the net charge and biophysical properties of Envs and impacts their antigenic structure. Here we employ a novel robotic cell line selection strategy to address the problems of low expression. Additionally, we employed a novel gene-edited CHO cell line (MGAT1- CHO) to address the problems of high sialic acid content, and poor antigenic structure. We demonstrate that stable cell lines expressing high levels of gp120, potentially suitable for biopharmaceutical production can be created using the MGAT1- CHO cell line. Finally, we describe a MGAT1- CHO cell line expressing A244-rgp120 that exhibits improved binding of three major families of bN-mAbs compared to Envs produced in normal CHO cells. The new strategy described has the potential to eliminate the bottleneck in HIV vaccine development that has limited the field for more than 25 years.