Recombinant biomaterials: are they worth the trouble?

Abstract

Introduction Proteins play a crucial role in the field of biomaterials due to their inherent biocompatibility, bioactivity, and ability to interact with biological systems. As structural and functional components of natural tissues, proteins such as collagen, gelatin, fibrin, and silk fibroin are widely used in tissue engineering, drug delivery, and medical implants. However, the biological origin of these materials also raises questions related to batch-to-batch variability, the risk of disease transfer or ethical considerations (1). As a result, recombinant proteins are often proposed as a workaround since the tightly regulated expression of human proteins in e.g., microbial host systems rules out these risks. However, designing and expressing recombinant proteins is not straightforward and requires a good understanding of the necessary steps to translate a gene of interest into purified proteins that can be used as a biomaterial (2). The numerous interdependent experimental parameters make this field difficult to navigate for biomaterials scientists new to recombinant proteins. This contribution presents a systematic approach for the expression of elastin-like proteins (ELPs) that will help avoid common pitfalls and increase protein yield and purity.

Materials and Methods Plasmids encoding ELPs were generated based on a previously described method (3): short oligonucleotide fragments were fused together and ligated into a pUC18 cloning vector, after which the full construct was transferred to a pET15b expression vector used to transform the typical expression strain E. coli BL21(DE). Expression conditions reported in literature were applied, followed by isolating the ELPs from the cell lysate using immobilised metal affinity chromatography (IMAC) or inverse transition cycling (ITC).

Results Despite reproducing protocols from literature, overexpression of the ELPs was not observed in SDS-PAGE analyses. Moreover, no apparent difference between induced (+1mM IPTG) and non-induced cultures could be observed. Purifying the cell lysate resulted in minimal recovery of the protein of interest (PoI). To increase the expression yields alternative expression strains were considered together with adapting the experimental parameters, but this only resulted in minimal improvements. By starting from scratch and not relying on recommendations from literature, protein yield started to increase. By changing the sequence, the expression vector, replacing the expression strain and optimising the expression parameters via the Design of Experiments, expression yields comparable to literature reports were finally obtained.

Discussion The results indicate that expression parameters depend on the primary sequence of the PoI, making it difficult to reproduce experimental results and discouraging others from entering the field. Surprisingly, low reproducibility was also observed for expression vectors obtained from the Addgene repository and linked to the original scientific publication. This hints at an important yet unidentified experimental parameter currently not included in published method sections.

Conclusions It can be challenging to produce recombinant biomaterials, even for scientists familiar with recombinant DNA technology. However, many of these difficulties can be mitigated by applying a systematic approach tilting the balance in favour of the design freedom and unparallelled control that recombinant biomaterials offer, making recombinant biomaterials worth the trouble.