Transmembrane delivery of specific recombinant proteins for reprogramming of somatic cells and disease therapy

The anionic plasma membrane is generally refractory to passive extracellular-to-cytoplasmic transit of proteins. A number of highly regulated endocytotic processes specify which extracellular proteins gain access to the cytosol, and which are excluded. Whilst critical for normal biological function in the natural setting, this property of the cellular plasma membrane represents major impedance to delivery of therapeutic proteins to cells, particularly delivery of large and/or charged proteins. The need to circumvent this membrane impermeability for research, or treatment of disease, led to development of various polycationic peptides (collectively termed cell penetrating/transduction peptides; CPP’s) that are capable of transmembrane transfer without disruption to the lipid bilayer (reviewed Sawant & Torchilin, 2010, and references therein). These were commonly devised from viral surface proteins or viral-host protein-protein interactions shown to be important to infection. The vast majority of CPP’s described to date (eg. Tat peptide, Penetratin; reviewed Deshayes et al., 2005) are non-selective in nature; ie. transmembrane transduction is achieved in most/all cell types. Interaction of CPP’s with lipid raft components/cell surface glycoproteins (often ubiquitously expressed across cell types) mediates cytoplasmic translocation via macropinocytosis, clathrin-mediated endocytosis, and/or caveolae/ lipid raft-mediated endocytosis, often in a concentration-specific manner (Duchardt et al., 2007, and references therein). Since host cell glycoproteins are ubiquitously expressed, many CPPs are non-selective in nature; ie. translocation occurs in most/all cell types (e.g. Tat peptide, Penetratin peptide; reviewed Deshayes et al., 2005). In a landmark study, Takahashi et al., (2006) forced expression of four key transcription factors (Oct4, Sox2, Klf4 and cMyc) in somatic cells to reprogram them to pluripotent, colony-forming phenotype that resemble embryonic stem cells (ES cells) by various criteria. Although this presents an opportunity to derive patient-specific stem cells for human disease therapy, elucidation of the molecular events that characterize the adoption of the pluripotent phenotype is required before their clinical applicability could be realized. I utilized a non-selective CPP to deliver key recombinant proteins to somatic cells in vitro, aimed at deciphering the temporal and molecular events that characterize early somatic cell reprogramming. Specifically, I investigated the concentration and temporal requirements of cMyc in repression of lineage associated genes (,eg. Thy1 in fibroblasts), a requisite biological event that precedes adoption of the pluripotent phenotype (Heffernan et al., 2011, submitted; Chapter 3). I describe construction of recombinant protein expression vectors incorporating (i) an arginine-rich basic domain (49-RKKRRQRRR-57) of HIV trans-activating transcriptional activator (Tat) protein (for transduction across cellular membranes), and (ii) mouse cMyc protein (denoted pTATmcMyc). Purification of semi-soluble/particulate pTAT-mcMyc recombinant protein preceded experiments highlighting contributions of cMyc and other reprogramming factors in repressing Thy1 in fibroblasts, suggesting a ‘cMyc-mediated’ and ‘default (cMyc absent)’ mechanism of Thy1 repression (Chapter 3). In chapters 4 & 5, I propose a theoretical framework for the treatment of multiple sclerosis(MS), a disease characterized by neural demyelination in the central nervous system (CNS). Conceptually, in vivo administration of fusion protein incorporating nonselective CPPs (as outlined Chapter 3) may treat disease that manifests across numerous/all cell types. However the full therapeutic potential of CPP’s will be realized when cell selective CPP’s are devised for cell-specific delivery of therapeutic proteins in vivo.Chapter 4 outlines preliminary development of a glial cell-specific CPP (gCPP), modeled on arenaviral infection of glia, for targeted delivery of therapeutic peptides. A screen of putative gCPPs in vitro highlighted one gCPP (termed ‘TD2.2) that effectively translocated to human glial cells (immature and matured oligodendrocytes, and astrocytes), yet appeared largely incapable of translocating to a non-glial (human) cell line. This tentatively demonstrated glial cell-selectively of the TD2.2 peptide sequence. Time course, sectional confocal microscopy provided further visual evidence for transduction of TD2.2 to human oligodendrocytes in vitro (Chapter 4). Myelin Associated Glycoprotein (MAG) is an oligodendrocyte-derived, periaxonal protein that regulates neural-glial cell signaling, structural/spatial integrity of myelin and Nodes of Ranvier and maintains glialaxonal cell interactions (Yang et al., 1996; Dashiell et al., 2002; Nguyen et al., 2009). The proteolytic cleavage of the periaxonal (extracellular) component of MAG by matrix metalloproteases (MMPs) results in loss of physical and molecular interactions of neural and glial cells, thus contributing to the progressive demyelination and axonal loss characteristic of MS (Sato et al., 1984; Moller et al., 1987; Tang et al., 1997; Stebbins et al., 1997; Milward et al., 2008). The mobile, digested product of MAG is also thought to represent a circulatory auto-antigen, further exacerbating disease. Chapter 5 of this thesis outlines theoretical design and construction of a mutated MAG protein (MAGMUT) capable of evading MMP mediated proteolysis. Following construction of protein expression vectors for expression (in E.coli) of histidinetagged, wildtype MAG (MAGWT) and MAGMUT recombinant protein, technical difficulties were encountered with induction and/or protein purification. In addition, MMP7 digestion experiments with oligodendrocytes expressing retrovirally delivered transgenes comprising EGFP-MAGWT and EGFPMAGMUT fusion protein were also somewhat inconclusive. Thus, validation of the ability of the MAGMUT sequence to evade MMP7-mediated proteolysis in vitro could not be conclusively drawn. In addition to a final discussion of experimental results and a proposal of future research directions, Chapter 6 addresses philosophical concerns if iPS technology and alternative strategies for deriving therapeutic cells (ie. transdifferentiation). Strategies to maximize persistence of recombinant protein in circulation, and functionality of recombinant protein in the cytosol, that could be adopted for in vivo validation of recombinant proteins (Chapters 4 & 5) are also discussed. To conclude, the myriad of possibilities in recombinant protein design, and relative ease of purification for screening of putative peptides, highlight the therapeutic potential of this technology for treatment of human disease. However, the full potential of this technology for human disease therapy will only be realized with concurrent development of strategies for targeted cellular delivery