Protein Engineering for Cardiovascular Therapeutics: Untapped Potential for Cardiac Repair

Abstract

Numerous new and innovative approaches for repairing damaged myocardium are currently under investigation, with several encouraging results. In addition to the progression of stem cell-based approaches and gene therapy/silencing methods, evidence continues to emerge that protein therapeutics may be used to directly promote cardiac repair and even regeneration. However, proteins are often limited in their therapeutic potential by short local half-lives and insufficient bioavailability and/or bioactivity, and many academic laboratories studying cardiovascular diseases are more comfortable with molecular and cellular biology compared with protein biochemistry. Protein engineering has been employed broadly to overcome weaknesses traditionally associated with protein therapeutics and has the potential to specifically enhance the efficacy of molecules for cardiac repair. Yet protein engineering as a strategy has not yet been employed in the development of cardiovascular therapeutics to the degree that it has in other fields. In this review, we discuss the role of engineered proteins in cardiovascular therapies to date. Further, we address the promise of applying emerging protein engineering technologies to cardiovascular medicine and the barriers that must be overcome to enable the ultimate success of this approach.

An explosion of interest in new therapies for heart repair has occurred recently. Most notably, clinical trials of cell-based cardiovascular therapy have garnered much attention and preliminary results seem encouraging^{1, 2}. However, the mechanism of action of cell-based therapies remains unclear, and many details with regard to characterization, quality control and delivery of cells remain to be worked out prior to widespread therapeutic application. Nonetheless, the reported efficacy of cardiac cell-based therapies in human trials has generated interest in additional therapeutic development pathways. Indeed, the identification of the release of paracrine-acting proteins as one mechanism by which cell-based therapies in the heart may improve cardiac function³ has spurred renewed interest in protein-based therapeutic approaches for cardiac repair.

Although protein therapeutics are increasingly becoming mainstream in numerous fields including cancer and inflammatory diseases – 9 of the 20 top selling drugs in 2012 were proteins⁴ – proteins have a relatively low penetration in the cardiovascular market. This is not to suggest a lack of progress in the cardiovascular therapeutic development, as small molecule drugs such as statins^{5, 6} and anticoagulants⁷ have been huge successes. Rather, as overall new drug approvals shift predominantly to proteins, protein therapeutic development represents an area of opportunity for cardiovascular medicine. Protein therapies allow for more targeted interventions – with well-defined regulatory pathways, improved scalability, and potentially reduced overall cost – compared to cell-based approaches. Furthermore, proteins have much larger surface areas than the small molecules of typical orally bioavailable drugs, and thus many of the molecular targets that are probably not “druggable” with small molecules can be targeted with proteins. A classic example of this concept is that no clinically successful small molecule activates the insulin receptor, and therefore Type I diabetes patients must inject insulin.

With regard to proteins for cardiovascular therapy, much effort has been put into enhancing cardiac microvasculature formation. More recently, the potential of proteins to induce cardiomyocyte proliferation has been identified⁸, increasing optimism that protein-based approaches can be effective for cardiac regeneration. However, protein therapeutics often have limitations, including poor bioavailability, undesirable pharmacokinetics and biodistribution profiles, and off-target effects^9–12. One approach to overcome these deficiencies is protein engineering, a strategy that may not only render a therapeutic concept feasible, but may generate the intellectual property necessary for development in patients.

The term “protein engineering” is a general one that covers many techniques for modifying proteins, including the use of molecular display technologies to enable targeted delivery^{13, 14}, structural modification of proteins to impart enhanced properties such as resistance to enzymatic degradation via rational design¹⁵ or directed evolution¹⁶, fusion of proteins to polymers or other protein domains to promote immune evasion^{17, 18}, incorporation of desired properties or functional groups via non-canonical amino acids^{19, 20}, and numerous others. Despite this promise and widespread application in fields such as oncology, engineered protein therapeutics have so far failed to become a major part of the cardiologist’s toolkit. Here we review the use of protein therapeutic approaches in the heart to date, with emphasis on therapies that have reached the clinical trial stage. We then highlight emerging and promising techniques in protein engineering and discuss how, when combined with new discoveries in molecular cardiology, these approaches might lead to a new generation of therapeutics for repairing the heart.

Protein Engineering for Cardiovascular Therapeutics to Date

Though none have yet achieved widespread clinical use, numerous engineered proteins have been applied to cardiovascular diseases in clinical trials. Many other native protein therapeutic approaches have also been applied toward cardiac repair, and several of these may benefit from protein engineering. Some of the earliest and most high profile examples of protein engineering technology for cardiovascular therapy have been clinical failures. However, other examples have proven more successful, and ongoing clinical trials offer exciting possibilities.

Tumor Necrosis Factor (TNF) Antagonists

One of the most prominent, and ultimately unsuccessful, examples of translation of protein engineering technology for cardiovascular therapy involved the TNF antagonist etanercept (Enbrel®). Developed based on research by Beutler and colleagues²¹, etanercept is a dimeric fusion protein comprising the Fc region of human Immunoglobulin G1 and the soluble human TNF Receptor 2 (TNFR2) domain that is approved for treatment of various forms of arthritis. Following the description of overexpression of TNF in heart failure²², it was hypothesized that TNF inhibition could be an effective treatment modality. However, despite initial promising results in carefully performed studies, both pre-clinically^{23, 24} and in patients^{25, 26}, two large scale clinical trials revealed that not only did etanercept not show clinical benefit in chronic heart failure (CHF) patients, but it may in fact increase the risk of CHF-associated morbidity and mortality²⁷. Moreover, the chimeric monoclonal antibody infliximab (Remicade®), a TNF inhibitor that works through a different mechanism than etanercept^{28, 29}, also did not improve the clinical outcome from heart failure patients, with high doses showing profound adverse effects³⁰. It should be noted that both etanercept and infliximab are efficacious against various forms of arthritis, and infliximab is highly effective against Crohn’s disease. Thus protein engineering-based antagonism of TNF has proven to be a case of a successful product development pathway that has so far been unsuccessfully applied to cardiovascular therapy despite compelling pre-clinical investigation.

Natriuretic Peptides

The natriuretic peptides offer an example of a promising application of protein engineering to cardiovascular therapy. Atrial natriuretic peptide (ANP) and B-type natriuretic peptide (BNP), also known as brain natriuretic peptide and basic natriuretic peptide, have similar activities and are well established as useful biomarkers in the guidance of cardiovascular therapy^{31, 32}. In addition, exogenous administration of BNP has been shown to improve multiple parameters in patients with systolic heart failure³³. However, BNP as a therapeutic (nesritide) may be limited due to its hypotensive effects ³⁴. C-type natriuretic protein (CNP) causes a lower hypotensive response than ANP and BNP, owing to its lack of activity on arteries due to differential receptor expression³⁵. Additionally, a related peptide isolated from snake venom, Dendroaspis Natriuretic Peptide (DNP), acts similarly to ANP and BNP but is highly potent and resistant to enzymatic degradation³⁶. In order to leverage the advantageous properties of CNP and DNP, Burnett and colleagues synthesized a chimeric natriuretic peptide that comprises domains of both (CD-NP)³⁷. Initial results from clinical trials for this engineered protein as a heart failure therapy were promising³⁸ and a follow-up trial may shed more light on the potential of CD-NP for clinical use³⁹.

Insulin-like Growth Factor-1 (IGF-1)

IGF-1 is a protein with a similar molecular structure to insulin that has been shown to provide protection from the progression of heart failure in mice⁴⁰. In humans, low serum levels of IGF-1 are associated with an increased risk of ischemic heart disease⁴¹. However the undesirable side effects of IGF-1 systemic delivery are well noted and include increased risk of diabetic retinopathy and cancer^42–44. Thus, as of early 2013, there were only two active clinical trials examining IGF-1 (Mecasermin®) as a cardiovascular therapy^{45, 46}. As an attempt to overcome these effects by promoting local delivery, Tokunou et. al. engineered an IGF-1 fusion with the heparin-binding (HB) domain of heparin-binding epidermal growth factor to make HB-IGF, which proved effective in stimulating chondrocyte biosynthesis⁴⁷. Additionally, Hubbell and colleagues engineered a variant of IGF-1 with increased immobilization capacity within fibrin that improved smooth muscle cell proliferation⁴⁸, introducing the possibility of a co-factorial local delivery approach. Notably, an IGF-1 modified to enable interaction with self-assembling peptides for cardiac delivery has demonstrated efficacy in improved cardiac function following MI^{49, 50}. Thus, though not yet applied in clinical trials, engineered variants of IGF-1 may yield therapeutics for cardiovascular therapy.

Stromal cell-derived Factor-1α (SDF-1)

One protein under active clinical investigation for cardiac regeneration – although not currently as a protein therapy – is SDF-1. SDF-1 is a chemokine that plays important roles in angiogenesis and leukocyte trafficking⁵¹. The discovery that SDF-1 induces stem cell homing to the heart following injury^{52, 53} spurred interest in its therapeutic application. However, SDF-1 is proteolytically cleaved by both matrix metalloproteinase-2 (MMP-2)^{54, 55} and dipeptidyl peptidase IV⁵⁶, and thus the likelihood of retained bioactivity in the myocardium following injury – a highly inflammatory environment – is low. For this and other reasons, the only active clinical trial of SDF-1 for cardiac therapy employs plasmid delivery^{57, 58}, which offers the potential for prolonged SDF-1 expression but is also limited by issues of safety and unpredictability common to gene therapy approaches. Protein engineering applied to SDF-1 offers an alternative; we developed a protease-resistant form of SDF-1 by mutating a single amino acid within the MMP-2 cleavage site¹⁵. This protease-resistant SDF-1 successfully induced endothelial progenitor cell recruitment to the heart following MI that resulted in improved cardiac function¹⁵ and led to increased angiogenesis and improved ventricular function following onset of myocardial ischemia⁵⁹. Protein engineering efforts to improve delivery and tissue retention of SDF-1 have also been reported⁶⁰, as has the creation of a polypeptide analog of SDF-1 that induced improved recovery after MI compared to the native protein⁶¹. Future synergy of these and other protein engineering strategies may allow for a therapeutic approach that overcomes the limitations of the native SDF-1 protein.

Granulocyte Colony-Stimulating Factor (G-CSF)

G-CSF is a glycoprotein that selectively induces a reduction of SDF-1 and an increase in the SDF-1 receptor CXCR4 in the bone marrow⁶². A majority of clinical trials involving G-CSF in the heart have focused on its use as an adjunct to cell therapy^63–65, but others have assessed its potential as a primary therapy^66–69 as G-CSF has been shown to prevent deleterious remodeling after MI⁷⁰. A potential next step in these active lines of therapeutic investigation could involve the use of fusion proteins incorporating G-CSF moieties, as improved functionality relating to delivery⁷¹ and stability⁷² of G-CSF have already been reported.

Interleukin (IL) Receptor Antagonists

Interleukins are proteins that induce pleiotropic effects in a wide variety of cell types⁷³. In the heart, IL-1 is a potent mediator of inflammation during ischemia/reperfusion injury⁷⁴ and elevated plasma levels of IL-6 have been linked with increased risk of future MI⁷⁵, thus their inhibition is of therapeutic interest. IL-1 has a naturally occurring competitive inhibitor for its receptor (IL-1R), the IL-1 related protein IL-1 receptor antagonist (IL-1Ra), also known as anakinra⁷⁶. Both forced overexpression and injections of anakinra have been shown to reduce apoptosis following heart injury in rodents^{74, 76}, and this drug has been applied in clinical trials with promising results⁷⁷. In addition, IL-6 inhibition has been pursued via antagonism of its receptor (IL-6R or CD126) by an engineered, humanized monoclonal antibody, tocilizumab. Encouragingly, a human case report of tocilizumab administration resulting in successful improvement of cardiac dysfunction associated with multicentric Castleman disease by has been published⁷⁸, and a clinical trial investigating the efficacy of tocilizumab in MI is ongoing⁷⁹. Thus interleukin receptor antagonists may represent a promising path forward for protein engineering applied to cardiovascular therapy.

Erythropoietin (EPO)

A glycoprotein that regulates red blood cell production, EPO has well-documented cardioprotective properties and works via multiple mechanisms^{80, 81}. Pre-clinical studies in both rat⁸² and rabbit⁸³ acute MI models showed that EPO administration improves cardiac contractility and hemodynamic parameters, prompting human clinical trials. Initial findings indicated that EPO, while failing to improve left ventricular ejection fraction, may reduce heart failure in acute MI patients⁸⁴. However, a more recent trial has cast doubt on the potential for EPO as a therapy for heart failure, indicating that it may lead to increased infarct size⁸⁵. Owing to these results, focus has shifted towards examining the efficacy of low doses of EPO⁸⁶, an approach that might be aided by applying protein engineering techniques to improve potency and stability of EPO, as has already been demonstrated⁸⁷.

Neuregulin-1β (NRG)

NRG is an essential regulator of cardiovascular development and plays an important role in cardiovascular disease⁸⁸. The cardioprotective potential of NRG has been well established in pre-clinical models^{89, 90}, and enthusiasm for future therapeutic application of NRG in heart failure increased upon the report of its potential ability to induce cardiomyocyte proliferation⁸. Clinical trials with the epidermal growth factor (EGF)-like domain of NRG in heart failure have demonstrated safety and efficacy^{91, 92}, and another NRG molecule – glial growth factor-2 (GGF-2) – is currently in clinical trials for heart failure⁹³ and has shown efficacy in a pre-clinical model⁹⁴. The nature of the interactions between the cognate receptor of NRG in the heart, ErbB4, and its preferred signal induction partner, ErbB2, may provide an opportunity for exploitation via protein engineering. Our group hypothesized that ErbB receptor interactions, which are prerequisite to signaling, could be biased away from the most commonly induced partnerships by receptor-ligand affinity interactions imposed by a bivalent ErbB ligand⁹⁵. Bivalent NRG (NN) was shown to induce differential signaling compared to NRG⁹⁵ and has superior cardioprotective efficacy compared to the EGF-like domain of NRG in a mouse model of doxorubicin-induced cardiomyopathy⁹⁶. Further exploration of this strategy may reveal a benefit of protein engineering for NRG in cardiac repair and regeneration.

Vascular Endothelial Growth Factor (VEGF) and Fibroblast Growth Factor (FGF)

Due to their well-characterized mitogenic effects on endothelial cells in many contexts, VEGF⁹⁷ and FGF⁹⁸ – as well as their splice isoforms and variants – have been examined extensively for therapeutic vascularization in the heart over the last ~25 years^99–101. However, negative outcomes of early clinical trials with both FGF¹⁰² and VEGF¹⁰³ have stalled progress toward clinical translation of therapeutic vascularization approaches, with common limitations such as short half-life and systemic side effects cited as reasons for the failures. Pre-clinical protein engineering studies of improved stability of angiogenic growth factors through a variety of techniques are plentiful^104–110, and the potential for local delivery of VEGF via fusion with a collagen-binding domain has been reported¹¹¹. Yet, there are still no protein drugs for therapeutic vascularization of the heart used in clinical practice. Strategies that combine protein engineering with nanotechnology¹⁰⁹ or other methods to improve delivery^{49, 112–115} may hold promise to enable protein-based therapies for therapeutic vascularization in the heart in the future.

Additional Therapeutic Avenues

Beyond what is mentioned above, additional ongoing cardiovascular therapeutic efforts may benefit from protein engineering. Like NRG, periostin has also been reported to induce cardiomyocyte proliferation¹¹⁶ and thus could be of further therapeutic interest. Growth differentiating factor 11 (GDF11) has recently been identified as a mediator of the reversal of age-related cardiac hypertrophy¹¹⁷ and could open a new area of exploration for protein therapeutics for the heart and other organs. Glucagon-like peptide-1-based therapies have been effective pre-clinically¹¹⁸, and the cyclic peptide approach currently employed¹¹⁹ may lend itself to refinement through protein engineering. Biased G protein-coupled receptor ligands hold intriguing promise¹²⁰ and represent a template for future protein engineering strategies designed to promote selective receptor activation, such as has already been done for the ErbB receptor system⁹⁵. Adipokines are peptides or proteins secreted by adipocytes that can have beneficial or detrimental effects on the cardiovascular system depending on when and for how long they are exposed to it (reviewed in ^{121, 122}). Protein engineering techniques that impart enhanced control over tissue localization and delivery of this class of peptides/proteins may prove critical in enabling their eventual therapeutic application. Further, intravenous immunoglobulin administration may be effective against certain forms of heart failure¹²³. Engineering of proteins has perhaps been most extensively applied to immunoglobulin molecules, and so any effects observed in ongoing clinical trials¹²⁴ could potentially be augmented through robust protein engineering approaches. Antibody-based therapeutic approaches as well as other protein therapies for immunomodulation continue to emerge as potential treatments for atherosclerosis¹²⁵. More recently, mimetic peptides and monoclonal antibodies have been developed to inhibit PCSK9 to lower cholesterol levels¹²⁶. All of the aforementioned approaches have promise, and the further application of protein engineering techniques to augment these therapeutic strategies could potentially enhance their efficacy.

Therapeutic Frontiers in Protein Engineering

Despite the implementation of some protein therapeutics for cardiovascular therapy, many limitations still remain. The ability to target delivery of a protein, or any drug, directly to the heart would minimize the required dose as well as undesirable off-target effects; but this ability remains mostly beyond our current reach. In general, proteins have often been considered poor drug candidates drugs owing to low oral bioavailability, a lack of long-term stability and other characteristics, and this applies to proteins for cardiovascular therapy as well. Indeed, a friend of the authors’ – an organic chemist currently employed in drug discovery by a major pharmaceutical company – revealed that many drug discovery professionals with chemistry training are taught a common mantra: “proteins are not drugs”. The biotechnology revolution has helped to change this perception, and protein engineering is a significant reason why, with fields such as oncology brimming with new engineered protein therapeutics. However, this revolution has, for the most part, occurred outside of the realm of cardiovascular medicine. The existing cardiovascular therapeutics created via protein engineering noted above demonstrate only a fraction of the potential of the field. Innovative and exciting protein engineering approaches are currently being developed, with many already having been applied to non-cardiovascular therapeutics. Application of these burgeoning technologies to cardiovascular medicine – for example, to enable heart-specific targeted drug delivery, to enhance protein stability and circulation times, or to promote direct stimulation of specific pathways to promote cardiomyogenesis via a therapeutic protein – may enhance the utility of protein-based cardiac therapy.

Molecular Display

Molecular display encompasses techniques that present molecules – typically peptides or proteins – on the surface of a cell, virus, or other host entity. This approach enables a linkage of genotype and phenotype of the displayed molecule, as its coding information is hybridized with that of the carrier. Thus, selective enrichment is possible through successive propagation steps under specified conditions – a process known as panning – and large polypeptide libraries can be screened for desired properties or interactions (Figure 1).

Figure 1.

Molecular Display. A protein or peptide with specific properties can be derived via molecular display following multiple rounds of selection. Polypeptide libraries can be made or purchased and either inserted into molecular hosts such as phage or yeast or appended to carrier molecules such as mRNA or ribosomes. The schematic above is representative of phage or cell-surface display, where proteins are displayed on the surfaces of organisms that are subsequently exposed to various substrates, which serve as selection criteria. In the example above, protein-mediated binding of the organism with a specific substrate followed by washing away of non-specific binders and elution enables amplification of those organisms displaying proteins with the desired properties. Repetitive screening with additional or more stringent selection criteria enables ultimate isolation of an optimized protein sequence based on phenotype, which can then be linked with genotype following extraction of genetic material from the organism. Thus, molecular display facilitates discovery of information needed to design a protein with specific properties evolved to match nearly any chosen selection criteria.

The concept of molecular display arose following George Smith’s seminal report of phage display, the display of a foreign protein on the surface of filamentous phage¹³. Since then, numerous platform approaches for molecular display have been developed to enable high throughput screening of protein interactions. Technologies such as cell-surface display, ribosomal display, mRNA display and others have allowed for directed evolution of proteins to refine activity and stability. Phage display is the most commonly employed and is popular in the biotechnology industry, having been used in the development of numerous protein drugs¹²⁷, especially antibodies¹²⁸. Beyond directed evolution, phage display has been employed for panning of biological and/or tissue samples – known as biopanning – to select for peptide ligands that enable tissue-specific homing¹²⁹ and/or enhanced tissue retention¹³⁰ of drugs or drug carriers.

In addition to phage display, other display technologies have emerged and may have relevance for the development of cardiovascular therapies in the near future. Invented by Wittrup and colleagues¹⁴, yeast surface display enables, amongst other things, high throughput quantitative library screening via fluorescent-activated cell sorting (FACS) and has been used to evolve extremely high affinity antibodies¹³¹ and peptides¹³² that have facilitated molecular targeting, for example, to promote high-resolution vascular imaging¹³³. Cell-free protein evolution methods such as ribosome display¹³⁴ or mRNA display¹³⁵ are not encumbered by transformation or expression limitations and also allow rapid evolution of high-affinity binding proteins. Overall, molecular display technologies are generally mature methods for selection and refinement of protein characteristics that have significant potential to promote cardiovascular protein therapies.

Engineered Protein Scaffolds

Antibodies have been the most successful protein therapeutics to date. However, antibodies have weaknesses as therapeutics, including limited tissue penetration due to their large size. To address this and other issues, numerous ‘engineered protein scaffolds’ (EPS) have been developed. For the purposes of protein engineering, EPS comprise a minimal polypeptide framework that can be based on either an immunoglobulin or a non-immunoglobulin molecule¹³⁶. The framework is typically monomeric without disulfide bonds or glycosylation sites and is usually highly stable and readily soluble¹³⁷. EPS are also typically able to be easily expressed in a host organism and have surface-accessible residues that allow incorporation of sequence diversity and subsequent selection via molecular display.

Numerous EPS have moved beyond the initial development stage towards clinical use¹³⁷. Amongst the most well-developed are: Adnectins¹³⁸, which are derived from type III fibronectin domains; Anticalins¹³⁹, derived from lipocalin; Kunitz domains¹⁴⁰, derived from Kunitz-type protease inhibitors; DARPins¹⁴¹, derived from ankyrin repeat proteins; Avimers¹⁴², derived from the A-domain of low density lipoprotein receptor; and many more. Ecallantide (Kalbitor®), a rationally designed Kunitz domain, was approved in 2009 for clinical use to treat hereditary angioedema¹⁴³ and could potentially be used in cardiothoracic surgery as a replacement for aprotinin. Many other EPS are being tested for myriad applications¹³⁷, and as molecular targets for cardiac repair and regeneration continue to be defined, EPS have the potential to play an important role in future development of cardiovascular therapeutics.

Non-Canonical Amino Acids (ncAAs)

The concept of incorporating non-canonical amino acids (also referred to as unnatural amino acids, non-natural amino acids, and artificial amino acids) site-specifically into proteins offers stunning potential: an avenue to attach any chemical functional group of interest to a protein, allowing for coupling with other proteins or molecules such as fluorophores or small molecule drugs using straightforward, efficient, and tunable chemical reactions. Although several variations exist, the general approach involves aminoacylation of a tRNA that has an incorporated suppressor anticodon with a ncAA. This chemical modification facilitates site-specific ncAA incorporation during mRNA translation at a nonsense codon site, which can be inserted (site-specifically) in a gene of interest using standard gene synthesis or molecular biology techniques¹⁴⁴ (Figure 2). This development of ncAA technology was pioneered by Schultz and colleagues¹⁹, and this group and many others have contributed to important advances that have made it robust for imparting desired functionality to proteins for therapeutic applications. For example, ncAA incorporation enables site-specific PEGylation of therapeutic proteins¹⁴⁵ to enhance evasion of the immune system and promote longer circulation times. This approach also allows for insertion of molecular staples – hydrocarbon linkers that can stabilize protein secondary structure in part by physically preventing entropic structural relaxation – into proteins to improve stability¹⁴⁶. This strategy could be extended via expressed protein ligation techniques¹⁴⁷ to fuse other engineered proteins for similar purposes¹⁴⁸. Using a related process developed by Tirrell and colleagues, residue-specific ncAA incorporation, Interferonβ-1b was PEGylated for improved pharmacological properties^{149, 150}. ncAA site-specific incorporation has further been utilized to generate bispecific antibodies¹⁵¹ and thus could be adapted to the creation of bivalent NRG molecules⁹⁵ that might stimulate cardiomyogenesis. As ncAA technology continues to evolve, its potential for application towards cardiovascular therapy, which is already promising, should only be enhanced.

Figure 2.

Incorporation of non-canonical amino acids (ncAA) into proteins. ncAA incorporation into proteins has been shown via multiple methods. The schematic above contrasts canonical amino acid (AA) incorporation into proteins with site-specific ncAA incorporation. Canonical AA incorporation proceeds following aminoacylation catalyzed by an endogenous aminoacyl-tRNA synthetase, which charges tRNA with a cognate amino acid. ncAA aminoacylation is possible via incorporation of a heterologous, orthogonal (i.e. not cross-reactive with host cell machinery) tRNA:synthetase pair into the cell responsible for protein production. Due to this incorporation, an orthogonal ncAA can be inserted site-specifically in response to a specific codon (typically a stop codon). A given ncAA typically has similar structure to a canonical AA except that a desired, atypical chemical functional group is included such that it is accessible to participate in a reaction. Thus, once translated, a protein incorporating a ncAA is readily modifiable with molecules such as polyethylene glycol (PEG), which improves circulation and limits immediate renal clearance, or with a molecule such as a fluorescent probe that facilitates imaging, as well as many other potential possibilities.

Enabling Methods for Drug Delivery

Protein engineering techniques have begun to be used for the development of proteins and peptides that enable more effective and efficient small molecule, protein and nucleic acid delivery across biological barriers and/or targeted to specific tissues. This is a critical consideration, as effective cardiovascular protein therapies will almost certainly require targeted or localized delivery to the heart. Some of these strategies have already been mentioned or implied, such as the functionalization of a therapeutic protein or a micro- or nanocarrier with a targeting sequence derived by molecular display or from an engineered protein scaffold. This general concept has also been applied to enable gene delivery with enhanced tissue specificity¹⁵². In addition, Dowdy and coworkers developed the concept of protein transduction domain (PTD)-containing fusion proteins for protein delivery¹⁵³. In this method, the protein of interest is fused with a PTD – sometimes also referred to as a cell-penetrating peptide – such as that derived from the human immunodeficiency virus TAT protein¹⁵³. This allows, amongst other things, proteins to cross biological barriers that would normally restrict them due to size and charge considerations¹⁵⁴ (Figure 3). PTDs have also been used in gene delivery¹⁵⁵. A conceptually similar fusion protein approach whereby the protein of interest is fused with an antibody or peptide that is transported across a biological barrier, such as the blood-brain barrier, by receptor-mediated transport has also been explored¹⁵⁶ and is sometimes referred to as the “molecular Trojan horse” approach¹⁵⁷. Some of these techniques have already been applied to facilitate targeted delivery to the myocardium¹⁵⁸, and continued expansion of these and other efforts¹⁵⁹ will likely prove critical to the ultimate success of protein-based cardiovascular therapies.

Figure 3.

Protein engineering methods to enable drug delivery. Numerous techniques within the realm of protein engineering could be employed to enhance delivery of small molecules, proteins and nucleic acids to the heart. Examples shown above include the targeting of a nanocarrier to a cell type of interest (e.g. cardiomyocytes, endothelial cells, etc.) by a peptide molecule derived via molecular display and the enhancement of cellular internalization of a therapeutic entity via fusion with a protein transduction domain (PTD).

Enabling Technologies for Cell-Based Therapy

In addition to providing an alternative for cell-based cardiac therapy, protein engineering can also be used to augment this therapeutic approach. Cardiac cell transplantation has been limited by poor localization of injected cells¹⁶⁰, an ~90% death rate for cells within one week of the injection or implantation¹⁶¹, and, in the case of stem cells, uncontrolled cell proliferation or differentiation following transplantation¹⁶². Some of these limitations have been overcome in the most recent clinical trials^{1, 2}, however the efficiency and efficacy of cell-based cardiac therapy can still be improved. Our group has employed a therapeutic approach utilizing molecularly designed self-assembling peptide scaffolds combined with engineered proteins to create a microenvironment for improved engraftment and regenerative potential of transplanted cells for myocardial regeneration^{15, 49, 113}. In this case, self-assembly occurs through electrostatic interactions between heterospecific, complementary amino acid sequences; a peptide scaffold is engineered to express one sequence and the therapeutic protein entity is engineered to express the complement. This approach allows for nearly any chosen growth factor(s) to be controllably displayed or delivered along with cells, which may bind or otherwise interact with the peptide scaffold. In this way, differentiation and/or engraftment of transplanted cells can be improved. Moreover, the starting materials are easy to synthesize and purify on a large scale. This and other methods for enabling cell-based therapies for the heart provide yet another therapeutic frontier for cardiovascular protein engineering.

CONCLUSION

Protein engineering is a powerful approach with untapped potential for cardiovascular therapeutic development. As previously stated, about half of new drugs are proteins⁴, and many of these therapeutics were created via protein engineering. The application of this technology to therapeutic development should only increase in the future, and yet no engineered protein drugs are currently being routinely applied for treatment of cardiovascular diseases, perhaps owing to deficiencies in our current understanding of the molecular mechanisms of cardiac repair and regeneration, which is still evolving (the authors direct the reader to several extensive recent reviews on the topic^163–166). From the standpoint of therapeutic development, two phenomena that contribute to myocardial repair and regeneration have been the focus of much of the effort: vascularization and cardiomyogenesis. While the fundamental processes that constitute vascularization – angiogenesis, arteriogenesis, and vasculogenesis – are relatively well defined, the mechanistic underpinnings of cardiomyogenesis remain the subject of controversy. Hopefully, advances in the molecular cardiology field and resolution of fundamental controversies within it will facilitate a leap forward for cardiovascular medicine through increased utilization of engineered protein therapeutics.

Non-standard Abbreviations and Acronyms

CHF

chronic heart failure

HB

heparin-binding

MI

myocardial infarction

mRNA

messenger ribonucleic acid

PEG

poly(ethylene glycol)

TAT

trans-activator of transcription

tRNA

transfer ribonucleic acid