In order to provide a suitable expression platform for those proteins that require post-translational modifications such as glycosylation or disulfide bond formation, we generated pFastBac/HBM-CT-TOPO (Figure 2A)—a vector that facilitates the translocation of the cloned gene products into the endoplasmic reticulum. The vector harbors the Honey Bee Melittin (HBM) secretion signal coding sequence, which has been placed next to the topoisomerase binding site, upstream of the gene of interest (Figure 2A).
To validate the effect of the HBM signal sequence, we first generated recombinant expression baculovirus particles by cloning the emerald green fluorescent protein (EmGFP) gene into pFastBac/HBM-CT-TOPO (pFastBac/CT-TOPO and pFastBac/ NT-TOPO were used as controls), and used the generated particles to infect SF21 insect cells.
EmGFP with a HBM secretion signal sequence had a distinguished localization pattern compared to N- or C-terminal his-tagged EmGFP which was evenly distributed throughout the cytoplasm and nucleus in infected cells. The HBM-EmGFP fusion protein localized primarily in the periphery area of the cells and had a granular appearance, suggesting that the protein is in the process of secretion (Figure 2B).
To confirm this hypothesis, cellular and extracellular fractions were analyzed by Western blots. Although EmGFP from pFastBac/HBM-CT-TOPO and pFastBac/ CT-TOPO expressed at comparable levels (Figure 2C, lanes 3 and 4), only EmGFP expressed from pFastBac/HBM-CT-TOPO was found on the medium (Figure 2C, lanes 7 and 8), confirming that the HBM secretion signal sequence is able to promote secretion EmGFP. On a side note, the EmGFP levels produced by pFastBac/NT-TOPO were so high that leakage into the extracellular medium was unavoidable (Figure 2C, lanes 1 and 6).
To assess the relative secretion efficiency of proteins fused to the HBM signal sequence, we replaced the original signal sequence of two ordinarily secreted proteins by the HBM sequence, and we examined the presence of these protein products in the extracellular medium. We used human coagulation factor IX (F9) and erythropoietin (EPO) as models. Their corresponding genes were cloned into pFastBac/CT-TOPO (including their natural signal sequence coding region) and into pFastBac/HBM-CT-TOPO (excluding their signal sequence coding region).
Four different proteins were, therefore, expressed: wild type F9, wild type EPO, and their corresponding mature peptides fused to the HBM signal sequence. Virus particles were obtained and used to infect SF21 and High Five cells. Cell culture media were collected 24h, 48h, and 72h post-infection, and the secreted proteins were analyzed by Western blot.
Proteins were secreted to comparable levels regardless of what signal sequence the protein was fused to (Figure 2D). The relative electrophoretic mobility of the different proteins suggests that the signal sequence was correctly cleaved (Figure 2D). The amount of EPO and F9 in the culture media increased up to 8 and 2 mg/L respectively after three days post-infection. The results confirm that the HBM signal sequence can function as efficiently as the wild type signal sequences of EPO and F9 to promote the secretion of proteins validating the yeast-derived HBM signal peptide as an efficient protein secretion vehicle in insect cells.