disease
Yu, H.
Citation
Yu, H. (2007, June 21). Therapeutic and imaging potential of peptide agents in cardiocascular disease. Retrieved from https://hdl.handle.net/1887/12090
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/12090
7 General discussion and summary
Summary
Cardiovascular diseases, such as myocardial infarction and stroke, are the major cause of death in Western society and the principal clinical manifestation of atherosclerosis. Although percutaneous angioplasty has been established as a safe and effective intervention for the treatment of symptomatic atherosclerosis, restenosis considerably reduces its therapeutic efficacy. Now multiple inflammatory processes were shown to contribute to the development of atherosclerosis and restenosis, both pathophysiologies are regarded as inflammatory diseases. Although our understanding to atherosclerosis has advanced in the last few decades, currently available imaging techniquesfor the diagnosis of cardiovascular diseases are rather nonselective, often invasive and yield very limited information on the plaque composition. In addition, traditional immunosuppressants as a therapy for the treatment of restenosis are accompanied by severe side-effects and nephrotoxicity. More selective and less toxic drug candidates possessing both anti-inflammatory and antiproliferative characteristics are highly desired in the treatment and prevention of restenosis. In this thesis, we aimed to develop novel peptide ligands by a combined phage display and bioorganic synthesis strategy and to investigate the targeted potential of developed peptide leads for the diagnosis and treatment of atherosclerosis and restenosis. The thesis comprises two parts. The first part focuses on the development, by combinatorial phage display and organic synthesis, of novel peptide leads as therapeutic and targeted imaging agents for atherosclerosis. The second part focuses on validating the potential of a novel peptide inhibitor of NFAT for the treatment of restenosis. Furthermore, rational optimization of VIVIT peptide was carried out in order to improve pharmaceutical characteristics including the potency, pharmacokinetics and stability.
In Chapter 1, we generally introduced the pathology underlying the initiation and the progression of atherosclerosis and restenosis. We have selected two targets, scavenger receptor A type I (SR-AI) and CD40 protein as candidate for therapeutic intervention and diagnosis of atherosclerosis because both cell surface receptors are instrumental in the ontogenesis of disease. SR-AI plays a key role in innate immune response and intracellular cholesterol accumulation that precedes the foam cell formation, allowing us to monitor a hallmark of disease development in atherosclerosis. Its substrate, oxLDL will upon internalization aggravate the disease by releasing entrapped bioactive lipids, by eliciting an immune response to
MHC-II presented oxLDL derived antigens and by causing oxidative stress in the host cell. In Chapter 3, we describe the successful identification of a novel peptide ligand of SR-AI by peptide phage display, which was showed to be specific SR-AI antagonist. A synthetic peptide sequence encoded by the major SR-AI binding phage clone (PP1) was able to displace ligand binding to SR-AI at an IC50 of 29 μM and the minimal essential motif required for SR-AI interaction was defined.
After docking to a streptavidin scaffold, peptides were effectively internalized by macrophages in a SR-AI dependent manner. The enriched phage pool and streptavidin immobilized PP1 exhibited a similar in vivo biodistribution profile in mice with marked accumulation in hepatic macrophages. Importantly, PP1 affected a significant 2-fold increase of 125I-streptavidin uptake by aortic artery lesions in ApoE-/- mice, which was not observed for the control. The potential of the peptide lead in SR-AI directed imaging of atherosclerotic lesions was established. The second candidate CD40 plays a pluripotent role as co-stimulatory molecule in adaptive immune responses at all stages of the atherosclerotic process. In Chapter 4, affinity selection of phage displayed peptide libraries led to the identification of specific CD40 binding peptide. A peptide competition ELISA was set up to identify the essential motif of the peptide selective for human CD40.
Tetramerization of the CD40 binding peptide increases its avidity and biologic potency, which is in agreement with the presumed trimeric interaction between CD40 and CD40L. Moreover, peptide encoding phage were able to accumulate in inflamed joints in a murine rheumatoid arthritis model establishing its imaging potential. Therefore, CD40 selective binding peptide may hold promise as targeting device for imaging agent or for the delivery of genes to the inflammatory active vulnerable plaque.
In the second part, we turned our attention to the inflammatory process itself which underlies atherosclerosis and restenosis. Restenosis, a hyperplastic disease with an inflammatory component, is at least partly regulated by calcineurin/NFAT signaling. We opted for VIVIT, a more selective and less toxic inhibitor of NFAT function than the conventional calcineurin inhibitors CsA/FK506, as therapeutic candidate in antirestenotic therapy. In Chapter 2, we reviewed the regulatory pathways in calcineurin/NFAT signaling and its importance to cardiovascular disorders including restenosis. The recent advances, application and perspective of VIVIT as an immunosuppressant in the treatment of cardiovascular diseases were delineated as well. Our focus was then to map the role of NFAT in atherosclerosis/restenosis and to optimize the potency and facilitate the delivery of VIVIT analogs for the development of a new generation immunosuppressants. In Chapter 5, we evaluated the potential of synthetic peptide inhibitor of NFAT, VIVIT, as superior drug candidate than CsA in anti-immune and anti-proliferative therapy. VIVIT appears to be a specific and potent inhibitor of NFAT activation and thus of NFAT mediated proliferation and inflammation. Therefore, synthetic VIVIT therapy will not be accompanied by non-NFAT mediated side effects on
calcineurin signaling and constitutes a promising lead in anti-restenotic therapy. In addition, we define the key role of NFAT which interweaves together with MEK- ERK as a positive cross-talk to trigger vSMC hyperplasia. In chapter 6 we described the rational design and successful identification of a selective bipartite peptide inhibitor of NFAT with nanomolar potency, showing superior selectivity for NFAT and reduced toxicity over CsA. We pursued alanine scan and truncation studies to pinpoint the minimal essential motif of VIVIT as HPVIVIT. When conjugated this motif to an Inhibitors of NFAT-calcineurin association (INCA) analogue, a non-toxic bipartite compound named MCV was designed, which potently inhibited NFAT activation with an IC50 of 2 nM. Moreover, we delineated the accurate binding cleft of the MCV compounds on calcineurin in a docking model in order to elucidate the underlying mechanism for the tremendous affinity gain. MCV compounds with improved potency, optimal chemical features, superior selectivity and low toxicity, constitute excellent candidates in the generation of novel, selective anti-inflammatory and anti-proliferative drugs.
In general, phage display is a powerful tool for new drug discovery and the designed peptides selective for SR-AI and CD40 are of great potential as contrast agents for the diagnosis of atherosclerotic plaques. In addition, we have firmly established that synthetic VIVIT peptide displays a more favorable profile than that of CsA as a better drug candidate for the prevention of vSMC hyperplasia and restenosis. The follow-up designed bipartite peptide with improved potency, chemical features and superior selectivity renders this class of NFAT inhibitors excellent candidates for the treatment of autoimmune diseases as well as cardiovascular disorders such as atherosclerosis, myocardiac hypertrophy and restenosis.
7.1 Phage display and new drug discovery
Phage display allows for efficient screening of peptide/antibody on a variety of targets such as antibodies, enzymes, cell-surface receptors, mRNA, whole cell layers which are all prominent targets in drug discovery. Peptides derived by phage display are generally small in size (7-25 amino acids), nonimmunogenic and easy to be synthesized for large scale production. Peptide phage library is based on iterative selection of randomized peptide sequences, which especially fit for selection of novel unnatural ligands. Principal drawbacks of peptide phage display are the lack of potency and the bias towards selection of binding epitopes.
Moreover, receptor such as SR-AI or CD40 trimer, which will reduce the chance of successful selection for potent peptide inhibitors in comparison with the high affinity of antibodies derived by phage display. These limitations can be overcome by the combination of rational optimization, combinatorial organic synthesis and design of peptide oligomers either through synthesis or by docking the biotinylated peptide to a streptavidin scaffold to mimic the tetrameric binding which will generally enhance the binding affinity of the peptide lead. Another powerful tool is
to employ the combinatorial approach to perform side group modifications which has been established as an efficient method to improve the peptide affinity. A further issue that needs to be tackled is the unfavorable pharmacokinetics of peptides with enzymatic instability, rapid clearance and poor oral bioavailability.
Lead peptides selected from phage library have only rarely been applied in direct clinical use without modifications. In this regard, D-peptide, retro-reverso D- peptide, conjugation to uptake enhancers can help to circumvent these problems.
Peptidomimetics, compounds containing non-peptidic structural elements mimicking the biological action of the natural parent peptides, will display a more favorable pharmacokinetics and enhanced bioavailability for eventually clinical use.
Phage display derived antibodies tend to display high potency and affinity towards the targets. However, immunogenic issue arisen from the lack of humanized antibodies will preclude their clinical use. Nanobodies, therapeutic proteins derived from llama containing the variable domain of heavy chain of heavy-chain antibody (VHH), were recently developed which possess full antibody function but lack light chains1. They are much smaller in size than the original heavy-chain antibodies but harbor the full antigen-binding capacity. Nanobodies are of high target specificity, low immunogenicity, enzymatic stability, and often exhibit a preference for receptor-ligand binding clefts. Therefore, llama derived antibody phage display will be a promising alternative to the traditional antibody/peptide phage display in new drug development.
7.2 Peptide targeting to SR-AI
Scavenger receptors SR-AI and CD36 play a major and redundant role in modified LDL uptake and foam cell formation. SR-AI has a broad ligand binding profile Previous in vitro and mouse model studies have shown that if SR-AI is knockout or knockdown, there are no large adverse effects on atherosclerosis. The reason is that other scavenger receptors will compensate for the loss of modified LDL uptake in SR-AI deficiency. Therefore, selective SR-AI binding peptide PP1 is probably better fit for non-invasive targeted imaging than direct therapeutic use in atherosclerosis. Similarly, CD36 antagonists might not be an ideal target for the inhibition of atherosclerosis as well. However, since SR-AI and CD36 have been identified to be the major receptors responsible for modified LDL uptake, simultaneous inhibition of SR-AI and CD36 function might be beneficial in the treatment of foam cell formation and atherosclerosis. In this thesis, we have successfully employed peptide phage display for the design of peptide ligands selective for SR-AI which proved to be useful as homing device for contrast agents for diagnostic imaging of atherosclerotic plaque. Incorporation of selective SR-AI ligands as homing device for contrast agents is expected to improve the sensitivity of atherosclerotic plaque imaging. This could be particularly relevant to advanced atherosclerotic plaques which are characterized by a thin fibrous cap, a large necrotic lipid core and many infiltrated macrophages in the rupture-prone shoulder
region. SR-AI peptide aided imaging of macrophage rich vulnerable plaques may represent a novel approach in the diagnosis of vulnerable plaques in order to prevent acute clinical complications.
7.3 Peptide targeting to CD40
Interruption of the CD40 signaling was shown to be beneficial for the prevention and amelioration of atherosclerosis. The affinity of CD40 selective binding peptide NP31 is surprisingly high (400 nM) which might results from the firm binding characteristics of constrained peptides. Preliminary studies showed that NP31 did not affect the CD40L-induced expression of costimulatory molecules on human dendritic cells, but interfered with cytokine production such as CD40L induced IL- 6 and IL-12 production (data not shown). The peptide was therefore assumed to be a partial antagonist for CD40 signaling. There are three possibilities to explain the partial CD40 inhibition by NP31. First, NP31 binds to an inert area instead of the crucial binding pocket of CD40/CD40L interaction. Secondly, CD40L forms trimers before interaction with CD40. The selection of our peptide antagonists for CD40 is based on a CD40 recombinant monomeric protein template. It is likely that the trimeric conformational structure is crucial for efficient CD40 ligation.
Thirdly, NP31 as a peptide may be too small to efficient block ligand-receptor binding and therefore the CD40 signaling. Streptavidin-conjugated biotinylated NP31 was shown to be a more potent binder of CD40 than NP31 monomers. This will provide us with more information on whether NP31 indeed acts as an (ant) agonist of CD40 signaling.
Furthermore, Antibody phage display will lead to more bulky and potent binders/inhibitors of CD40. Actually, several CD40L specific antibodies have been raised and tested in a pre-clinical stage. However, due to their severe toxicity, the antibody therapy against CD40L had been suspended. Llama derived antibody phage display might form an alternative tool for future anti-CD40L therapy. In addition, cell based selection by phage display had also been proved feasible2,3. Therefore, CD40L or CD40 overexpressing cells can be used for peptide or antibody selection. This system will optimally mimic the trimeric characteristics of CD40 (L) under physiological conditions. One issue we need to point out here is that antibody therapy against CD40/CD40L will powerfully influence the immune system and will undoubtedly produce severe side effects. Peptide inhibitors of CD40 as partial antagonists might be better candidates than CD40 blocking antibodies in targeted imaging and therapeutic approaches for the treatment of autoimmune disease and atherosclerosis. NP31 peptide will therefore be valuable for further optimization and characterization.
7.4 VIVIT peptide as a therapy to restenosis
Therapeutic agents with high affinity, good bioavailability, metabolic stability, and low toxicity are highly preferred in new drug discovery. Although the potency of selective NFAT inhibitors, such as VIVIT, had been tremendously improved as
shown in Chapter 6, several aspects of the bipartite peptide inhibitor, MCV, still need to be addressed for optimal pharmacological features. Moreover, the underlying mechanism of MCV binding to NFAT requires further elucidation.
Firstly, the pharmacokinetics and the efficacy of the MCV peptide, which was demonstrated to be a potent NFAT inhibitor in vitro, need to be addressed in vivo to validate its clinical application and efforts in that direction are currently underway. The general reactivity of the maleimido group of MCV to sulfhydryl groups may attenuate the MCV peptide potency in vivo. This issue can be overcome by replacing the maleimido group with milder cysteine reactive constituents, such as iodoacetamide (IM). Secondly, poor in vivo stability of peptide drugs can be circumvented by D-isomer analogues or retro-inverso D- peptide with maintained side chain topology of L-peptide. The rapid clearance of peptide can be improved by incorporating them into liposome or matrigel carrier for targeted delivery or local administration4-6. Thirdly, the efficiency of MCV peptide entry into cells needs to be characterized and improved. Injection of FITC labeled MCV peptides into mice will shed light on peptide cell entry and its vivo kinetics. Cell-permeable analogs, including the 11R- or penetratin-tagged peptides will facilitate peptide cell entry and improve in vivo kinetics. Recently we have obtained the IL-2-promoter sensor mice which will be a suitable mouse model to monitor the kinetics of MCV peptide and NFAT inhibition in vivo. Fourthly, the mechanism underlying the affinity gain of MCV peptide in comparison to VIVIT itself needs further elucidation. Reduced maleimido (succinimide) derivatives conjugated HPVIVIT will be helpful in the identification of the importance of the sulfhydryl group for MCV binding and will enable us to confirm the covalent reaction between MCV and calcineurin as predicted in Chapter 6. In addition, C266S mutagenesis of the recombinant calcineurin expression plasmid has been successfully constructed which will assist us to explore the binding mechanism of MCV peptide.
VIVIT was shown to be more selective and less toxic than CsA/FK506 in NFAT targeted immunosuppression which may be applicable in the cardiovascular system for the treatment of restenosis and hypertrophy. CsA and FK506 were both shown to be potent inhibitors of calcineurin/NFAT signaling. How can immunosuppressants be useful as therapeutics for the treatment of cardiovascular disease? In part atherosclerosis and in particularly restenosis can be regarded as calcineurin/NFAT dependent disease involved in cell inflammation and hyperplasia.
Calcineurin is ubiquitously expressed but its expression varies in different cell types7. The hypothesis is that the dose-response curve of calcineurin inhibition by CsA/FK506 obeys the law of mass. For example, it is expected that 10 nM of FK506 would inhibit 70% of calcineurin activity in lymphocytes which have
~5000 calcineurin molecule per cell but only 4% in cardiac cells which have about 200,000 calcineurin molecules per cell8. Therefore, 1 μM of FK506 will be required for the prevention of hypertrophy which is too high a dose for humans and will lead to toxicity. Since the downstream effector of calcineurin, NFAT has been
identified to be the crucial regulator of myocardial and SMC hyperplasia, targeting NFAT function by VIVIT might not only selectively inhibit the hyperplasia but also have a profound effect on cells that are highly resistant to CsA/FK506. In addition, the inhibition of calcineurin/NFAT signaling will not be influenced by variances of calcineurin or cyclophilin/FKBP12 expression in different cell types.
Therefore, we envision that potent NFAT inhibitor, MCV, will be equally potent in the treatment of transplantation, hypertrophy or restenosis at a very low dose and will generate far less toxicity in nonimmune cells than CsA/FK506.
Finally, the exact contribution of NFAT plays in restenosis compared with other signaling pathway such as MEK/ERK is still not clear. While calcineurin/NFAT and MAPK signaling pathways serve as pivotal regulators in growth response, as suggested in Chapter 5, they represent a small component in the whole signaling network in the control of cell survival or death. The predominant involvement and mutual crosstalk of these two pathways in the cardiovascular system need further scrutiny.
References
1. Muyldermans S. Single domain camel antibodies: current status. J Biotechnol 2001;74:277-302.
2. Michon IN, Hauer AD, der Thusen JH, Molenaar TJ, Van Berkel TJ, Biessen EA, Kuiper J. Targeting of peptides to restenotic vascular smooth muscle cells using phage display in vitro and in vivo. Biochim Biophys Acta 2002;1591:87-97.
3. Watters JM, Telleman P, Junghans RP. An optimized method for cell-based phage display panning. Immunotechnology 1997;3:21-29.
4. Benkirane N, Guichard G, Van Regenmortel MH, Briand JP, Muller S. Cross-reactivity of antibodies to retro-inverso peptidomimetics with the parent protein histone H3 and chromatin core particle. Specificity and kinetic rate-constant measurements. J Biol Chem 1995;270:11921-11926.
5. Ruvo M, Fassina G. End-group modified retro-inverso isomers of tripeptide oxytocin analogues: binding to neurophysin II and enhancement of its self-association properties.
Int J Pept Protein Res 1995;45:356-365.
6. Bartnes K, Hannestad K, Guichard G, Briand JP. A retro-inverso analog mimics the cognate peptide epitope of a CD4+ T cell clone. Eur J Immunol 1997;27:1387-1391.
7. Klee CB, Draetta GF, Hubbard MJ. Calcineurin. Adv Enzymol Relat Areas Mol Biol 1988;61:149-200.
8. Crabtree GR. Generic signals and specific outcomes: signaling through Ca2+, calcineurin, and NF-AT. Cell 1999;96:611-614.
