Evaluation of artificial signal peptides for secretion of two lysosomal enzymes in CHO cells

Enzyme replacement therapy (ERT) is a scientifically rational and clinically proven treatment for lysosomal storage diseases. Most enzymes used for ERT are purified from the culture supernatant of mammalian cells. However, it is challenging to purify lysosomal enzymes with sufficient quality and quantity for clinical use due to their low secretion levels in mammalian cell systems. To improve the secretion efficiency of recombinant lysosomal enzymes, we evaluated the impact of artificial signal peptides on the production of recombinant lysosomal enzymes in Chinese hamster ovary (CHO) cell lines. We engineered two recombinant human lysosomal enzymes, N-acetyl-alpha-glucosaminidase (rhNAGLU) and glucosamine (N-acetyl)-6-sulfatase (rhGNS), by replacing their native signal peptides with nine different signal peptides derived from highly secretory proteins and expressed them in CHO K1 cells. When comparing the native signal peptides, we found that rhGNS was secreted into media at higher levels than rhNAGLU. The secretion of rhNAGLU and rhGNS can, however, be carefully controlled by altering signal peptides. The secretion of rhNAGLU was relatively higher with murine Ig kappa light chain and human chymotrypsinogen B1 signal peptides, whereas Ig kappa light chain signal peptide 1 and human chymotrypsinogen B1 signal peptides were more effective for rhGNS secretion, suggesting that human chymotrypsinogen B1 signal peptide is the most appropriate for increasing lysosomal enzyme secretion. Collectively, our results indicate that altering signal peptide can modulate the secretion of recombinant lysosome enzymes and will enable lysosomal enzyme production for clinical use.

Automated summary

Enzyme replacement therapy is a scientifically rational and clinically proven treatment for lysosomal storage diseases, but improving the secretion efficiency of recombinant lysosomal enzymes requires altering signal peptides to modulate secretion levels.