Overcoming the Refractory Expression of Secreted Recombinant Proteins in Mammalian Cells through Modification of the Signal Peptide and Adjacent Amino Acids

Abstract

The expression and subsequent purification of mammalian recombinant proteins is of critical importance to many areas of biological science. To maintain the appropriate tertiary structure and post-translational modifications of such proteins, transient mammalian expression systems are often adopted. The successful utilisation of these systems is, however, not always forthcoming and some recombinant proteins prove refractory to expression in mammalian hosts. In this study we focussed on the role of different N-terminal signal peptides and residues immediately downstream, in influencing the level of secreted recombinant protein obtained from suspension HEK293 cells. Using secreted alkaline phosphatase (SEAP) as a model protein, we identified that the +1/+2 downstream residues flanking a heterologous signal peptide significantly affect secreted levels. By incorporating these findings we conducted a comparison of different signal peptide sequences and identified the most productive as secrecon, a computationally-designed sequence. Importantly, in the context of the secrecon signal peptide and SEAP, we also demonstrated a clear preference for specific amino acid residues at the +1 position (e.g. alanine), and a detrimental effect of others (cysteine, proline, tyrosine and glutamine). When proteins that naturally contain these “undesirable” residues at the +1 position were expressed with their native signal peptide, the heterologous secrecon signal peptide, or secrecon with an additional alanine at the +1 or +1 and +2 position, the level of expression differed significantly and in an unpredictable manner. For each protein, however, at least one of the panel of signal peptide/adjacent amino acid combinations enabled successful recombinant expression. In this study, we highlight the important interplay between a signal peptide and its adjacent amino acids in enabling protein expression, and we describe a strategy that could enable recombinant proteins that have so far proved refractory to expression in HEK293 cells, to be produced in sufficient quantities to answer important biological questions.

A plant protein signal sequence improved humoral immune response to HPV prophylactic and therapeutic DNA vaccines

“…The secretion function has been conserved from bacteria through eukaryotes and, though diverse in length/composition, ss share a conserved tripartite structure: 20-30 residues in length with a basic "N domain," a 7-13 residue hydrophobic "H domain" and a slightly polar "C domain." 18 In addition to this common structure, conservation of residues in particular positions of the mammalian signal peptide sequences is emerging. In this sense, our results are corroborated by very recent findings showing that in mammalian ss, small neutral amino acids are prevalent in ¡1 (last amino acid of the signal peptide) and ¡3 positions, whereas bulky aromatic amino acids are often found at ¡2 position and proline residues are virtually absent from all positions from ¡3 to C1 (first amino acid of the mature peptide) positions.…”

Whole synthetic pathway engineering of recombinant protein production

“…To determine predicted cleavage sites and relative "signal peptide-ness" (D-scores), each synthetic signal peptide was screened in silico using SignalP 4.1 (www.cbs.dtu.dk/services/SignalP ;Petersen, Brunak, von Heijne, & Nielsen, 2011). Güler-Gane et al (2016) recently showed that the secretion efficiency of recombinant proteins can be generally increased by inserting two alanine residues at the signal peptidemature protein junction, and accordingly this design feature was added to all five synthetic components (see Supporting Information Table S1 for sequences). Güler-Gane et al (2016) recently showed that the secretion efficiency of recombinant proteins can be generally increased by inserting two alanine residues at the signal peptidemature protein junction, and accordingly this design feature was added to all five synthetic components (see Supporting Information Table S1 for sequences).…”

“…The performance of signal peptide parts is typically unpredictable as the design rules governing their function are poorly defined (Brockmeier et al, 2006;Haryadi et al, 2015;Obst, Lu, & Sieber, 2017). For this example, such screening was unnecessary as previous studies have already identified an appropriate signal peptide for the specific context of SEAP production in mammalian cells (Barash et al, 2002;Güler-Gane et al, 2016). Moreover, it is not known why signal peptide function is highly protein-specific, a phenomenon that is likely determined by amino acid sequences at the signal peptide-product N-terminal junction.…”

“…The four components with the highest D-score values were selected for in vitro testing, alongside secrecon. Güler-Gane et al (2016) recently showed that the secretion efficiency of recombinant proteins can be generally increased by inserting two alanine residues at the signal peptidemature protein junction, and accordingly this design feature was added to all five synthetic components (see Supporting Information Table S1 for sequences).…”

“…Accordingly, it is not possible to rationally select or design signal peptides to maximize the translocation rate of specific proteins. However, trial and error testing has identified parts that exhibit robust performance with diverse protein partners, and panels of these constructs can be screened to find optimal signal peptide-product combinations (Güler-Gane et al, 2016;Haryadi et al, 2015;Kober, Zehe, & Bode, 2013). For this example, such screening was unnecessary as previous studies have already identified an appropriate signal peptide for the specific context of SEAP production in mammalian cells (Barash et al, 2002;Güler-Gane et al, 2016).…”