Abstract
The lipid bilayer of a cell presents a significant barrier for the delivery of many molecular imaging reagents into cells at target sites in the body. Protein translocation domains (PTDs) are peptides that breach this barrier. Conjugation of PTDs to imaging agents can be utilized to facilitate the delivery of these agents through the cell wall, and in some cases, into the cell nucleus, and have potential for in vitro and in vivo applications. PTD imaging conjugates have included small molecules, peptides, proteins, DNA, metal chelates, and magnetic nanoparticles. The full potential of the use of PTDs in novel in vivo molecular probes is currently under investigation. Cells have been labeled in culture using magnetic nanoparticles derivatized with a PTD and monitored in vivo to assess trafficking patterns relative to cells expressing a target antigen. In vivo imaging of PTD-mediated gene transfer to cells of the skin has been demonstrated in living animals. Here we review several natural and synthetic PTDs that have evolved in the quest for easier translocation across biological barriers and the application of these peptide domains to in vivo delivery of imaging agents.
Introduction
As molecular therapies are developed for a broader range of diseases, the demand for information that guides therapeutic decisions will change in nature and increase in complexity. Whereas clinical diagnostic and therapeutic measurements are currently made using imaging modalities that reveal changes at the level of organs and tissues, the development of molecular therapeutic approaches will require the use of imaging strategies that provide information at the cellular and molecular level with unprecedented specificity for a given disease. At the same time, new techniques will be required in the laboratory as research in molecular therapeutics moves from primarily cell culture assays to in vivo measurements of efficacy and target specificity. A significant problem in the development of these approaches is the inefficient delivery of many types of detectable molecular markers into tissues and cells within the body.
Whether directly labeling a molecular target or labeling a biological event using a reporter gene, molecular imaging agents must traverse the formidable barrier presented by the lipid bilayer of a cell. This barrier limits the intracellular delivery of a wide range of macromolecules including peptides and proteins, molecules that hold much potential as specific imaging agents. One strategy for breaching this barrier has been the use of directed translocation through the cell membrane using protein transduction domains. Peptide sequences comprising these domains have been utilized to facilitate the introduction of macromolecules into cells both in vitro and, possibly most importantly for imaging, in vivo.
The term “protein transduction domain” is based on the fact that the first such peptide was derived from a viral sequence (human immunodeficiency virus [HIV]) and “transduction” has traditionally been used to refer to gene transfer mediated by viruses or components of viruses. Given the large number of nonviral proteins that contain these domains and the wide use of these transporter peptides for the delivery of various cargo-types across the cell membrane, we will refer to these domains as protein “translocation” domains (PTDs). For completeness, it is important to note that these domains have also been referred to as cell penetrating peptides (CPPs). More generally, because the translocation or penetrating ability of these agents is demonstrably not restricted to peptides, they have been more broadly classified as molecular transporters, a term that neither restricts the nature of the cargo nor the structure enabling cell entry.
PTDs consist of short, typically basic, peptide sequences of 7–34 amino acids (aa) in length that are capable of crossing the cell membrane either alone or attached to a molecular cargo [1–3]. Nonbasic PTDs have also been reported, suggesting that a range of structures can be used to enable uptake, and that a number of uptake mechanisms are possible. Modifications to the peptide backbone have been made to improve the efficiency of translocation across biological barriers, and innovative advances in cargo release mechanisms from PTDs provide the ability to achieve targeted delivery of “free” cargo when necessary. PTDs have been shown to enable or enhance the translocation of a variety of attached cargo molecules including small molecules, peptides, metals, proteins, and DNA across tissue, cell membranes and, in some cases, the nuclear envelope [4]. There is evidence that transporter peptides have the potential to facilitate the introduction of imaging reporter genes directly into cells within the body where they can be expressed and used to tag cells. In addition, imaging reporter probes guided by PTDs could be more widely distributed throughout tissues, and new routes of reporter probe administration (e.g., transcutaneous or transalveolar) could be explored. In this review, we will focus on the use of PTDs in in vivo imaging strategies with the goals of understanding the current status of peptide-mediated translocation and demonstrating the potential of these domains in molecular imaging.
Protein Translocation Domains: Form and Function
PTDs occur naturally in proteins, such as the widely studied PTD of the HIV transcriptional activator Tat protein, and have been produced through design and synthesis, often inspired by the natural sequences [5,6]. The minimum transduction domain of the HIV Tat protein consists of aa residues 47–57 or 49–57 [7]. However, shorter Tat peptides have been shown to mediate translocation across cell membranes with decreased efficiency relative to the 9–10 aa minimal domain [8]. Antennapedia transcription factor (Antp) is a 60 aa homeodomain from the fruit fly (Drosophilia melanogaster) with a translocation domain spanning aa residues 43–58. This region is the third helix of the Antp homeodomain, also known as penetratin. Penetratin has been shown to translocate peptides of up to 100 aa in length [4,9–11]. Viral protein 22 (VP22) from herpes simplex virus (HSV) is a major component of the viral tegument and besides its function as a structural protein of the virus it has a complex pattern of expression and subcellular localization. In addition to its functions in the expressing cells, VP22 has the capability of intercellular spread [12].
Based on the naturally occurring PTD sequences, many synthetic peptides have been generated in an attempt to improve translocation across membranes and to identify more practical PTDs for use in imaging and therapy. PTDs typically include basic residues, such as arginine or lysine [13], and/or hydrophilic residues. Morris et al. [14] produced a 21-residue peptide carrier (Pep-1) containing lysine and hydrophilic residues and used it to deliver functional enzymes and antibodies intracellularly. Futaki et al. [15] utilized a series of arginine-rich peptides derived from 14 RNA-and DNA-binding proteins, including Tat (48–60), and found that all were able to translocate across cell membranes. Comparisons between the relative level of cellular uptake of arginine homopolymers versus Tat PTDs are difficult to interpret given the dependence of uptake on specific cargos or specific cell types in which the peptide translocation is attempted [4,16]. A major advantage of homooligomers of arginine is that they can be assembled in a significantly more cost-effective fashion through a segment doubling strategy [17]. This lowers the cost for various uses and thereby expands the range of applications that could draw on this approach.
PTD Chemistry in the Context of Molecular Imaging
In order to be effective, a peptide delivery system must form stable complexes, evade excretion and metabolism, avoid entrapment in extracellular matrices and intracellular compartments, carry its payload to the target site, and release it if necessary [18]. Several characteristics of the PTDs determine their ability to meet these demands as required for use in vivo. The mechanisms of membrane translocation can vary for each class of PTDs [14,19,20], and the efficiency of this process is a function of cell type and cargo type [21–23]. A given PTD can also operate through more than one mechanism. The lack of a single generalized mechanism for PTD translocation is a consequence of the variety of membrane barriers and the multiple modes of entry possible for a given barrier.
Residue Number, Type, and Charge
Although receptor-mediated uptake has been considered as a mechanism of PTD uptake, the range of structural variations that exhibit uptake suggests for many, if not most, basic domains a mechanism based on electrostatics and hydrogen bonding driven by a polar gradient across the membrane. This is consistent with the observation that arginine and specifically guanidinium groups (two hydrogen bonds) perform better than lysines and more specifically ammonium groups (one hydrogen bond) in facilitating cellular uptake. According to this mechanism, as the arginine oligomer approaches the cell surface interaction between cell surface charges, the guanidnium groups produce a complex whose association is related to the number of arginine groups [4,6,24]. As this complex moves into the less polar plasma membrane, the association of guanidinium groups with charged partners strengthens. The resultant complex of arginines and anionic counterions is neutral and sufficiently nonpolar for passage through the lipid bilayer. As noted above, the strength of this complex increases as it moves from the polar extracellular milieu into the nonpolar membrane. Further passage and release from the inner leaf of the plasma membrane weakens the association of the complex as it emerges in the more polar cytoplasmic milieu. This uptake mechanism would be driven overall by the membrane potential.
Many aspects of the above mechanism are consistent with observations for a range of PTDs [1,8,15]. Arginine residues appear to play a key role in the translocation process for many PTDs, and substitution of even one of these basic residues in such PTDs may result in a 70–90% reduction in translocation efficiency relative to the full-length domain [5]. Too few basic residues would not form strong associated complexes with cell surface charged residues. Too many could produce strong complexes that are less likely to release from the inner leaf of the membrane. In addition to aa residue type and number, chirality of the PTD sequence has been shown to affect uptake of several different classes of PTD–cargo complexes independent of the rate of decomposition of the
As stated above, the exact mechanism of PTD membrane translocation depends on the structure of the PTD itself, therefore certain alterations or modifications in the PTD primary structure may be useful in some applications of molecular imaging while they may not be relevant given others. For example, cargo delivery by those transporters utilizing mechanisms resembling endocytosis may benefit from key substitutions in the aa chain, allowing cargo escape from the endosome into the cytosol [13]. Such alterations would not benefit PTDs utilizing a mechanism based on electrostatics and hydrogen bonding driven by a polar gradient across the membrane. Thus, no blanket statement may be made regarding the utility of specific modifications across all PTDs. Instead, each PTD and its application must be matched, and the PTD backbone must be tailored using principles of hydrophobicity, steric hindrance, and mechanism of translocation.
Beyond Primary Structure
Secondary structures of the PTD and the relative location of charged residues appear to affect the efficiency of translocation of certain PTDs. Again, the presence and the magnitude of these effects depend on the underlying mechanism of translocation. For example, there is some evidence that the ability of Tat, Antp, and VP22 to form alpha-helical conformations could play a role in the translocating ability of these PTDs [2,10]. Ho et al. [16] improved on the transduction efficiency of the Tat (47–87) domain by strengthening its alpha-helical content and optimizing the placement of three arginine residues along a single face of the helix. Such modifications could be important in optimizing delivery of molecular imaging probes.
Flexibility of the PTD may provide advantages in translocation as well. Wender et al. [5] designed a polyguanidine peptoid derivative with a 1,4-backbone spacing of side chains of arginine oligomers with an oligo-glycine backbone lacking stereogenic centers and found that flexibility of conformation and sterically unencumbered straight chain alkyl spacing groups were very important for cellular uptake. Because cellular uptake is a dynamic process rather than a binding event, flexibility rather than preorganized structure is beneficial.
The PTD–Cargo Linkage and Clearance
The entity linking the PTD to its cargo is another important factor in the process of delivering imaging agent cargo across biological membranes. PTDs may be connected to their cargo through protein fusion, or via noncovalent or covalent linkages. When the PTD is part of the fusion protein, it would seem likely that the PTD would stay in the intracellular compartment along with its cargo. However, investigators have demonstrated evidence of PTD separation from its cargo after delivery of a fusion protein into the cytoplasm. Such degradation has been observed at varying rates depending on the specific PTD–cargo fusion protein [1].
Spacer groups within the linkage between the PTD and its imaging agent cargo may be varied to allow certain conformations of the PTD–cargo complex, or to provide points of detachment between the PTD and its cargo. Detachment strategies may be based on the intrinsic instability of the spacer group or biochemistry within the target cell, providing the opportunity to deliver cargo to specific tissues or cell types. Linkages can be designed to be susceptible to cleavage under changes in pH, in the presence of specific enzymes, or under photolytic conditions. There are a plethora of linkage possibilities between the PTD and its imaging agent cargo including disulfide bonds, esters, amides, phosphatates, hydrophobic interactions, or ionic types of interactions [27]. Chen et al. [28], for example, used a disulfide linkage between an arginine oligomer and an active peptide to enable delivery and release of the active peptide into cardiomyocytes and intact heart.
The kinetics of clearance of the transduced peptides from the cell and eventually the body are a function of the stability of the PTD–cargo complex. Engineered stability may allow tailoring of the PTD–cargo complex for the desired half-life [10]. Improved stability and increased efficiency of imaging agent cargo delivery may be realized by sequence inversion of the PTD sequence or synthesis of the PTD using D-amino acids [20], both of which protect the peptides from protease degradation. Production of a retro-inversion form of the Tat domain (57–48) demonstrated an apparent several-fold increase in uptake compared with the native Tat sequence [4].
The Cargo
Delivery of a variety of cargo types has been tested with PTDs. For example, a number of PTD–protein fusions have been evaluated; these have been composed of PTDs and β-galactosidase, human papillomavirus protein 16 (HPV-16), cyclin-dependent kinase, caspase-3, enhanced green fluorescent protein (eGFP), Cu- and Zn-superdismutase, Cre-recombinase, p53, PEA-15, CRAC, and human glutamate dehydrogenase (GDH) [1,3,29–34] (Table 1). The Tat domain has been the sequence most extensively used in molecular imaging to transport imaging agents into cells for cell labeling and subsequent trafficking studies in vivo as well as to study the biodistribution of the PTDs themselves. In these studies, superparamagnetic iron oxide nanoparticles were derivatized with Tat [35] or Tat was noncovalently linked to the radionuclides technetium 99m, rhenium [46], and indium-Ill using the bifunctional chelator diethylenetriamine pentaacetic acid (DTPA) [20,36,37].
Protein Transduction Domains and Cargo
Effects of the structure of the engineered molecular imaging cargo may also be relevant to efficiency of translocation and biocompatability. For example, Tat fusion proteins appear to undergo conformational changes as they are transferred across the cell membrane. Proteins linked to a Tat PTD are more efficiently translocated than native proteins [11,38,39]. This may, in part, explain why Tat-mediated protein translocation is unidirectional across the cell membrane [23]. The efficiency of some PTD–cargo complexes may rely less on their streamlined nature and more on other aspects of their conformational relationship. When the polyguanidine peptoid derivative (Pep-1) was used to translocate large proteins across cell membranes, for example, it appeared that there were many Pep-1 molecules accompanying/complexing a single cargo molecule [14]. In fact, arborization of multiple Tat molecules around a rhodamine-labeled plasmid has been shown to increase plasmid uptake, a finding that may improve the efficiency of reporter gene delivery in molecular imaging [40]. In addition to translocation efficiency, the size and homogeneity of the PTD–cargo complex determines its toxicity, ability to activate complement, solubility, blood compatibility, stability, and antigenicity, all critical factors in determining the success of any imaging strategy [18].
Cellular Characteristics Affecting PTD Translocation Efficiency
The translocation efficiency of PTDs varies between cell types and several-fold differences in efficiency of a single PTD have been reported when different cell lines were tested [26]. The reason for these differences in permeability of the biological membrane is not completely clear, however, varied composition of the cell surface constituents has been reported and linked to differences in translocation. For example, heparan sulfate (HS) proteoglycans are structural cellular membrane components that appear to be required for surface binding and translocation of some PTDs both in specific types of cell culture and in vivo [41,42]. Tyagi et al. [43] found that the affinity of the Tat-HS binding was proportional to the size of the heparin oligosaccharide and the content of arginine in the transduction domain. Other cell surface constituents such as glycosoaminoglycans (GAGs) may also contribute to successful translocation of certain PTDs through biological membranes but may not be critical. The dependence of PTD translocation efficiency on cell surface binding varies with the PTD.
Using PTDs to Translocate Macromolecules—Applications in Molecular Imaging
Radionuclide imaging modalities such as single photon emission computed tomography (SPECT) or positron emission tomography (PET), magnetic resonance imaging (MRT), and optical imaging modalities including bioluminescence and fluorescence imaging, have been shown to have the sensitivity and specificity that can provide cellular and molecular data in vivo [44–46]. Direct linking of a detectable tracer or probe to the molecule or cell of interest may provide sufficient information to reveal the position and relative abundance of the molecule or cell in the body but this type of labeling approach cannot provide sustained or long-term observations.
Alternatively, a reporter gene approach can be used to monitor cell trafficking, gene expression patterns, or other biological processes, over long periods of time. In this approach, cells are transduced with a reporter gene either in culture or in vivo. Depending on the choice of promoter (constitutive, inducible, or cell-type specific), this can provide spatiotemporal information about the expression of a gene or a labeled cell population. The reporter gene product itself may be directly detectable as with the fluorescent proteins (e.g., GFP), or after exogenous addition of a substrate or probe as in the use of luciferin and luciferase (Luc), or positron-labeled uracil, or acycloguanosine derivatives, as in the case of the HSV thymidine kinase (HSV-TK); a PET-detectable reporter protein [47–51]. The obvious advantage of integrating reporter genes into the genome of the target cells is that the reporter gene propagates with cell division and allows detection of the cell and its progeny, enabling long-term detection. In addition, by linking the expression of multiple reporter genes, a set of in vivo imaging modalities can be used coincidently, or sequentially, and in vivo imaging can be linked to ex vivo assays through the multifunctional reporters [52].
The processes of direct labeling or labeling via reporter genes possess inherent challenges, and achieving an equal distribution of imaging probes in vivo can be a formidable task. Additionally, low efficiency of gene transfer using viral or nonviral methods can limit the effectiveness of reporter gene imaging strategies. The requirement of in vivo delivery of a detection probe for reporter genes approaches (e.g., luciferin in bioluminescence imaging or FHBG (4-[18F]-fluoro-3-hydroxymethylbutyl)-guanine ([18F]-FHBG) in PET imaging) to localize the reporter protein imposes additional constraints because of the biodistribution of these agents.
PTDs offer some solutions to these problems: PTD-mediated translocation has been demonstrated for a diverse collection of biologically active molecules including proteins of up to 120 kDa molecular weight. Examples of proteins that have been delivered are listed in Table 1. In addition, liposomes (200 nm), and other small- or intermediate-sized molecules such as antisense DNA and siRNA have also been translocated across biological barriers while linked to a PTD [4]. These have largely been for the purpose of improving therapeutic efficacy in the fields of oncology [53–55], neurology [39,56], dermatology [24], cardiovascular medicine [57], and infectious disease [10,32]. Application of PTDs to imaging of potential therapeutic targets has been a more recent development. In addition to the PTD-mediated intracellular delivery of imaging reporter genes and probes, the imaging community has utilized iron oxide particles (40 nm) derivatized with PTDs in MRI and PTD chelates labeled with isotopes in nuclear medicine. However, the full potential of PTDs for delivery of imaging agents is currently being actively investigated.
PTDs to Address Issues in Direct Labeling or Delivery of Intact Imaging Probes
By introducing an externally detectable molecule into the cytoplasm, PTDs may offer the ability to directly label cells for cell trafficking studies without the limitations of cell membrane composition or special linking agents which arise in traditional direct cell-labeling techniques. Many investigators have used PTDs to carry fluorescent labels into cells in culture [5]. In addition, the Tat PTD can facilitate intracellular delivery of 40-nm superparamagnetic iron oxide (CLIO) nanoparticles into lymphocytes obtained from T-cell receptor transgenic animals that target a specific antigen [35]. Using this cell-labeling technique, the distribution of the labeled T cells relative to cells expressing the target antigen was monitored in vivo using MRI (Figure 1). The process of cell labeling with CLIO nanoparticles derivatized with Tat was shown in this study not to adversely affect normal cell activation or function [58,59]. Hematopoetic and neural progenitor cells have also been tracked in vivo using Tat-derivatized CLIO nanoparticles with implications for monitoring stem cell therapies in the future [60]. By attaching higher numbers of Tat peptides on each CLIO particle, cellular uptake has been shown to be enhanced [61]. Other researchers have recently proven cellular uptake of a gadolinium MRI contrast agent conjugated to polyarginine [62]. Novel chelation strategies have been key in advancing cell labeling, particularly in MRI and radioisotope imaging applications [36,63].
Three-dimensional magnetic resonance image of a mouse demonstrating an in vivo cell trafficking study using T lymphocytes labeled, ex vivo, with iron oxide particles highly derivatized with Tat (CLIO-HD). The two thighs of the mouse are shown and higher concentrations of CLIO-HD-labeled ovalbumin (OVA)-specific CD8+ T cells (red and green) are seen in the region of a B16 melanoma, overexpressing the OVA antigen, in the left flank (left side of the image) relative to background signal (blue) seen in the region of a B16F0 melanoma (no OVA expression) in the right flank (right side of image) [59]. (Courtesy of R. Weissleder.)
PTDs may offer the ability to rapidly distribute molecules ***systemically and may be effective in addressing the important problem in molecular imaging of rapid delivery of molecular probes with a resulting homogeneous distribution. Biologically active molecules are observed intracellularly within seconds to minutes following administration of PTD–cargo complexes in to cell cultures [10,16] and the Tat PTD has been shown to improve the distribution of biologically active cytoplasmic reporter proteins when Tat is administered systemically as a fusion protein or a cross-linked chimera [64]. Although not yet fully characterized, PTDs may improve the temporal resolution and the biodistribution of imaging agents in in vivo imaging applications and could lead to novel diagnostic tools and accurate measurements of therapeutic outcome.
Apart from their abilities to assist in translocation of molecules across the cell membrane and systemic distribution of molecules within the body, PTDs have also been touted and specifically designed to improve the delivery of drugs across tissue barriers including the skin, oral, and alveolar tissues and could potentially change the way imaging agents are administered [65]. Cyclosporin and protein kinase C agonists have been shown to be biologically active following PTD-mediated transport across the skin [24,66]. New administration routes for imaging agents could allow patient self-administration of “personalized” imaging agents facilitating repeated imaging throughout therapy with minimal patient discomfort and maximum circulation times.
PTDS to Address Issues in Reporter Gene Imaging
Many modifications have been made to a variety of PTDs in an attempt to optimize their ability to carry genetic materials into cells. For example, Tung et al. [40] have found that arborization of at least eight Tat peptides provides enhanced transfection of multiple cell lines. Gene transfer experiments using PTDs in animal models have included the successful introduction of antisense-oligonucleotides and DNA for triple helix therapy [15,20]. The intracellular expression of VP22-PTD-protein fusion DNA leads to the subsequent expression of the fusion protein, and spreading of these fusion proteins to neighboring cells. Examples of PTD-mediated transfer of reporter genes include the use of arginine-rich peptides such as HIV-1 Tat (48–60), HIV-1 Rev (34–50), and FHV coat protein (35–49) and oligo-arginines of 4–16 residues and transfection of COS-7 cells with luciferase-coding plasmid [15,67]. Siprashveli et al. [67] have imaged the delivery of the luciferase reporter gene to cells in the skin of living mice by cysteine-flanked, internally spaced arginine-rich (CFIS-R) PTDs and utilized in vivo measurements of bioluminescence to localize delivery and measure expression levels (Figure 2).
Bioluminescent images demonstrating PTD-mediated delivery of the luciferase reporter gene. In order to enhance plasmid DNA delivery by arginine-containing PTDs, a PTD with a core of interspersed residues of arginine, histidine, and glycine flanked by cysteine residues was constructed. Naked DNA (2 μg) or the cysteine-flanked internally spaced arginine-rich (CFIS-R) peptides carrying DNA (CG(RHGH)2RGC/DNA complexes) were injected intradermally. A single injection of the peptide oligomer-DNA complex was performed while two separate injections of naked DNA were carried out in a separate animal. The enzyme substrate luciferin was administered 48 hr later, followed by bioluminescent imaging (5 min after substrate administration). Pseudocolor images, representing bioluminescent signal intensity, were superimposed on gray-scale reference images of mice to better localize the signal [27]. Images demonstrated luciferase expression in mouse skin indicating successful in vivo gene transfer using the CFIS-R peptide that was augmented over the level of gene transfer achieved by naked DNA alone.
Several critical factors need to be taken into account in the application of PTDs to reporter gene introduction and imaging. For example, the linkage between the imaging reporter gene and its PTD carrier is important in determining the delivery and later biologic availability of the gene. Methods of attaching DNA to PTDs have included reversible DNA condensation with cationic lipids, cationic polymers, polyarginine peptides or polylysine peptides, PTDs containing cysteine-flanked, internally spaced arginine peptides, and a combination of a PTD and a DNA-containing liposome via an amphiphilic single-terminus reactive PEG derivative [15,37,67,68]. Park et al. synthesized PEGylated-peptides and glycopeptides, which formed stable cross-linked DNA condensates via disulfide bonds upon binding to plasmid DNA [69–72]. Blessing et al. [73] used cationic detergents with a single sulfhydryl group, which dimerized after binding DNA.
Stability of the PTD-nucleotide cargo complex has been shown to be a prime determinant of the efficiency of translocation across the cell membrane using the above PTD strategies. This could be a particularly difficult factor to address in the development of reporter gene imaging strategies since exact conditions of the local environment are difficult to control in vivo. Several cross-linking strategies are available, depending on the PTD utilized [74,75]. Adami et al. [76] demonstrated an inverse relationship between cross-linking and gene expression in cultured cells, however, this relationship is presumed to be quite different in the case of in vivo applications where increased cross-linking stability could allow improved delivery of PTD-gene complexes to target tissues. Escape of the cargo from nuclease digestion will be critical for delivery and expression of nucleic acids as therapies and imaging agents and PTDs can be engineered to protect their cargoes. Larger polymers bind more tightly to DNA and are thus advantageous to minimize degradation within the circulation [76].
Effective delivery and expression of DNA requires that the cargo be transduced into the nucleus, and various strategies may be employed to increase nuclear delivery of imaging reporter genes. The Tat protein itself is capable of delivering macromolecules to the nucleus [8,77]. In addition, because of HIV-Tat functionality and ability to cause HIV terminal repeat-driven RNA synthesis, it may be able to provide amplified RNA synthesis of an attached nucleotide sequence [5]. Intranuclear delivery of PTDs other than Tat has been observed [78]. Incorporation of a specific nuclear localization sequence can provide enhanced localization to the nucleus. These sequences may be chemically linked to a linear dsDNA fragment, or, as was demonstrated with the SV40 T-antigen (Tag) nuclear localization sequence, a DNA alkylating agent can be used to produce a covalent bond between plasmid DNA and a nuclear localization signal [20,79,80].
A feature of retroviral-mediated gene transfer is the integration of the DNA into the host cell genome, which is critical to long-term imaging of transduced cells. Integration following PTD-mediated DNA transfer is a low-frequency, random event, and if integration is desired for imaging or therapy, additional steps will have to be taken. One strategy of improving integration is the use of site-specific recombination triggered by the reporter gene-PTD complex itself [81]. Jo et al. [82] used a recombinant fusion protein consisting of Cre protein with the MTS protein transduction domain from Karposi Sarcoma Virus FGF-4 to introduce Cre-recombinase intracellularly and demonstrated sequence-specific recombination. By facilitating DNA recombination from a plasmid DNA into the host chromosome, Cre-recombinase delivered via linkage to a PTD could facilitate the incorporation of imaging or combined imaging/therapeutic reporter genes into the host cell, thus providing amplification of desired gene expression. Recombinases other than Cre have been developed. For example, following its expression in mammalian cells, the R4 integrase, a site-specific, unidirectional recombinase derived from the genome of phage R4 of Streptomyces parvulus, provides site-specific recombination between attB and attP recognition sites cloned on an extrachromasomal vector. In addition, Olivares et al. [83] and Sclimenti et al. [84] have demonstrated R4-mediated integration of plasmid DNA into the human genome using various techniques. Although demonstrated for targeting genes to specific sites in cells grown, or maintained, in culture, this approach is only now being used for in vivo gene targeting [85].
Potential Pitfalls in the Use of PTDs in Imaging
Any molecular imaging agent must be highly sensitive to subtle changes in the biology and chemistry of the target tissue but cannot interfere with the biological processes themselves [48]. Thus, if PTDs are to be utilized to deliver imaging agents, it is critical that these agents are relatively inert in vivo. Generally, smaller PTDs are less likely to have unfavorable interactions with biological systems while large cationic polymers may be toxic or precipitate [86]. The possibility of an immunogenic response by virally derived PTDs has been proposed [20], although the immunogenicities of the majority of PTDs are largely unknown [11]. If necessary, there have been strategies developed with the intent of avoiding an immunogenic response to agents utilizing PTDs. These include mild immune suppression of the recipient, use of very short PTDs or readily degradable PTDs, utilization of PTDs without anchor residues or TCR recognition motifs to avoid major histocompatibility complex presentation, and alternation between transduction domains in a given patient to avoid the reintroduction of any given peptide sequence [10,11,15].
In vivo use of imaging agents linked to PTDs is in its infancy, and little is known about the exact effects that the PTD–cargo complex will have on the final biodistribution of imaging probes and reporter genes. The biodistribution of superparamagnetic particles derivatized with Tat (Tat-CLIO) is reported to be similar to other nanoparticles, but Tat-CLIO is more rapidly cleared from the blood and shows differences in intraparenchymal hepatic distribution [87]. Bullok et al. [88] synthesized D-Tat-peptides labeled with fluorescein-5-maleimide (FM) at the C-terminus and 99m-Tc(CO)3 at the N-terminus and demonstrated concentration-dependent uptake of this conjugate within human Jurkat cells. Biodistribution studies of 99m-Tc(CO)3Tat-peptide complexes in Balb/c mice were performed by these investigators and suggested that the organ uptake and clearance of the Tat-peptide PTDs depended not only on the PTD itself, but on the chelated cargo imaging molecule as well. Rapid blood clearance rates were also observed. These observations emphasize the need for more extensive research into the exact effects of PTDs and their imaging cargo on the targeting of tissues in vivo.
Summary and Conclusions
Molecular therapeutic applications will require new diagnostic, dosing, monitoring, and outcomes measurements, and within the field of molecular imaging we must develop the tools to meet these demands. The potential applications of protein translocation domains in molecular imaging are vast, from addressing issues of labeling cells for trafficking studies to probe distribution to offering a method of reporter gene delivery into cells. Likewise, the tools and techniques of molecular imaging will have much to offer to the future investigation of PTDs, as these tools will provide noninvasive assays that avoid tissue sampling, and the need for fixation, a process which may introduce serious artifacts [89]. The opportunities offered by PTDs for diagnosis, therapy, and outcome measures are numerous, and, as the mechanisms of translocation and the factors that influence translocation efficiency are revealed, additional rational modifications to these peptides will be possible, and a class of reagents for in vivo delivery will be created.
