Production of native recombinant proteins using a novel split intein affinity technology.

Affinity tags are frequently engineered into recombinant proteins to facilitate purification. Although this technique is powerful, removal of the tag is desired because the tag can interfere with biological activity and can potentially increase the immunogenicity of therapeutic proteins. Tag removal is complex, as it requires adding expensive protease enzymes. To overcome this limitation, split intein based affinity purification systems have been developed in which a CC-intein tag is engineered into a protein of interest for binding to a NC-intein peptide ligand fixed to a chromatographic support. Tag removal in these systems is achieved by creating an active intein-complex during protein capture, which triggers a precise self-cleavage reaction. In this work, we show applications of a new split intein system, Cytiva™ ProteinSelect™. One advantage of the new system is that the NC-intein ligand can be robustly produced and conjugated to large volumes of resin for production of gram scale proteins. SARS-CoV-2 spike protein receptor binding domain and a Bispecific T Cell Engager in this work were successfully captured on the affinity resin and scaled 10-fold. Another advantage of this system is the ability to sanitize the resin with sodium hydroxide without loosing the 10-20 g/L binding capacity. Binding studies with IL-1b and IFNAR-1 ECD showed that the resin can be regenerated and sanitized for up to 50 cycles without loosing binding capacity. Additionally, after several cycles of sanitization, binding capacity was retained for the SARS-CoV-2 spike protein receptor binding domain and a Bispecific T Cell Engager. As with other split intein systems, optimization was needed to achieve ideal expression and recovery. The N-terminal amino acid sequence of the protein of interest required engineering to enable the cleavage reaction. Additionally, ensuring the stability of the CC-intein tag was important to prevent premature cleavage or truncation. Controlling the hold time of the expression product and the prevention of protease activity prior to purification was needed. These results demonstrate the feasibility of the Cytiva™ ProteinSelect™ system to be used in academic and industrial research and development laboratories for the purification of novel proteins expressed in either bacterial or mammalian systems.

Improved mRNA affinity chromatography binding capacity and throughput using an oligo-dT immobilized electrospun polymer nanofiber adsorbent.

Increased demand for mRNA-based therapeutics and improved in vitro transcription (IVT) yields have challenged the mRNA purification platform. Hybridization-affinity chromatography with an immobilized oligo-deoxythymidilic acid (oligodT) ligand is often used to capture mRNA through base pairing with the polyadenylated tail. Commercially available oligodT matrices include perfusive cross-linked poly(styrene-divinylbenzene) 50 µm POROS™ chromatography resin beads and convective polymethacrylate CIMmultus® monolithic columns consisting of 2 µm interconnected channels. POROS™ columns may be limited by poor mass transfer for larger mRNAs and slow flowrates, while monoliths can operate at higher flowrates but are limited by modest binding capacity. To enable both high flowrates and binding capacity for mRNA of all lengths, prototype chromatography media was developed by Cytiva using oligodT immobilized electrospun cellulose nanofibers (Fibro™) with a 0.3-0.4 µm pore size. In this work, four polyadenylated mRNAs ranging from ∼1900-4300 nucleotides were used to compare the dynamic binding capacity (DBC) of Fibro™, POROS® and CIMmultus® columns as a function of residence time and binding buffer compositions. Fibro™ improved the DBC ∼2-4-fold higher than CIMmultus® and ∼2-13-fold higher than POROS™ across all residence times, mRNA length, and binding matrix compositions tested. CIMmultus® DBC was least dependent on residence time and mRNA size, while both Fibro™ and POROS™ DBC increased at slower flowrates and with shorter mRNA. Surprisingly, inverse size exclusion (ISE) experiments showed that POROS™ was not limited by diffusion and POROS™ along with CIMmultus® demonstrate higher mRNA permeation however the Fibro™ prototype is not in the final configuration. Lastly, IVT reaction products were subjected to purification and oligodT elution product yield, quality, and purity were consistent across the three matrices investigated. These results highlight the benefits of high DBC and equivalent product profiles offered by the oligodT Fibro™ prototype compared to commercially available oligodT media.

Effect of 3-5 monocyclizations of angiotensin II and 4-aminoPhe6-Ang II on AT2 receptor affinity.

The endogenous angiotensin II (Ang II) and the synthetic AT(2) selective agonist 4-aminoPhe(6)-Ang II respond very differently to identical cyclizations. Cyclizations of Ang II by thioacetalization, involving the 3 and 5 amino acid residue side chains, provided ligands with almost equipotent binding affinities to Ang II at the AT(2) receptor. In contrast, the same cyclization procedures applied on the AT(2) selective 4-aminoPhe(6)-Ang II delivered significantly less potent AT(2) receptor ligands, although the AT(2)/AT(1) selectivity was still very high. The fact that different structure-activity relationships are observed after imposing conformational restrictions on Ang II and 4-aminoPhe(6)-Ang II, respectively, suggests that the peptides, despite large similarities might adopt quite different backbone conformations when binding to the AT(2) receptor.

A frame shifted disulfide bridged analogue of angiotensin II.

N-(2-Mercaptoethyl)glycine [NMGly] was incorporated into the 3 and 5 positions of angiotensin II and oxidized to give the corresponding cyclized disulfide c[NMGly(3,5)]Ang II. The binding affinity to the angiotensin II receptor (AT(1)) of this conformationally constrained analogue, which is related to the potent Ang II agonist c[Hcy(3,5)]Ang II, was examined. The analogue had no affinity to the AT(1) receptor. Theoretical conformational analysis was performed to compare the conformational characteristics of model compounds of c[Hcy(3,5)]Ang II and the frame shifted analogue c[NMGly(3,5)]Ang II in an attempt to explain the lack of affinity.