Essentially Leading Antibody Production: An Investigation of Amino Acids, Myeloma, and Natural V-Region Signal Peptides in Producing Pertuzumab and Trastuzumab Variants

Abstract

Boosting the production of recombinant therapeutic antibodies is crucial in both academic and industry settings. In this work, we investigated the usage of varying signal peptides by antibody V-genes and their roles in recombinant transient production, systematically comparing myeloma and the native signal peptides of both heavy and light chains in 168 antibody permutation variants. We found that amino acids count and types (essential or non-essential) were important factors in a logistic regression equation model for predicting transient co-transfection protein production rates. Deeper analysis revealed that the culture media were often incomplete and that the supplementation of essential amino acids can improve the recombinant protein yield. While these findings are derived from transient HEK293 expression, they also provide insights to the usage of the large repertoire of antibody signal peptides, where by varying the number of specific amino acids in the signal peptides attached to the variable regions, bottlenecks in amino acid availability can be mitigated.

Keywords: antibody families; antibody leaders; essential amino acid; logistic regression; myeloma; non-essential amino acid; recombinant production; signal peptide.

Figure 1

A schematic of workflow in the present study. Yellow arrows indicate experiment sequences while green arrows indicate data flow for constructing the predictive model.

Figure 2

Production rates (%) of recombinant antibodies containing various heavy and light chain pairings. (A, B) Recombinant Pertuzumab (A) and Trastuzumab (B) variants with the various SPs grouped according to VH families. (C, D) Recombinant Pertuzumab and Trastuzumab variants with the various SPs grouped according to Vκ families. Recombinant Pertuzumab and Trastuzumab variants produced with native (blue) or the ’IgE’ (orange) SPs respectively, are shown. In all experiments, the recombinant wild-type Pertuzumab (POK PG1) or Trastuzumab (HOK HG1) was used (shown in the first column) for normalization. The production rate of the variants utilizing the ‘IgE’ SP (23) was used as the reference for the production rate.

Figure 3

Recombinant antibody production rates of wild-type Vκ1|VH3 Pertuzumab (blue) and Trastuzumab (orange) using the IgE, Vκ1 and native SPs in %.

Figure 4

Recombinant antibody production (%) with IgE, Vκ1 and mutated Vκ1 SPs, (Vκ1 P18R and Vκ1 P18S) against the wild-type Vκ1|VH3 Pertuzumab (blue) and Trastuzumab (orange) models.

Figure 5

Production in % (in bars, left axis) and amino acid counts (line, right axis) charts of recombinant Pertuzumab variants. The green arrows/boxes refer to an increase in production level while the red arrows depict a decrease in production level.

Figure 6

Production in % (in bars, left axis) and amino acid counts (line, right axis) charts of recombinant Trastuzumab variants. The green arrows/boxes refer to an increase in production level while the red arrows depict a decrease in production level.

Figure 7

Quantification comparison of transient recombinant antibody production between EAA spiked and non-spiked for POK PG1 and HOK HG1 constructs.

Figure 8

Logistic regression model to predict the recombinant antibody production rates using amino acid counts. The prediction is evaluated using Area under the ROC (AUC) in triplicates, each with two independents (A, B) testing datasets. The “dummy” classifier is used as a baseline control.