| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Molecular Targeting and Polymer Toxicology Group, School of Pharmacy, University of Brighton, Brighton, UK; and
Cancer Research UK, Tumour Cytokine Biology Group, Wolfson Digestive Diseases Centre, University Hospital, Nottingham, UK
1Correspondence: Molecular Targeting and Polymer Toxicology Group, School of Pharmacy, University of Brighton, Brighton BN2 4GJ, UK. E-mail: s.m.moghimi{at}brighton.ac.uk;
 |
ABSTRACT |
|---|
 TOP ABSTRACT BACKGROUNDNANOSIZED TECHNOLOGIES FOR...INNER SPACETARGETINGTOXICITY ISSUESTHE FUTURE OF NANOMEDICINEREFERENCES |
|---|
Key Words: nanotechnology • nanosized drug delivery systems • nanoparticles • medical imaging • gene therapy • nanofibers • macrophage • endothelium • intracellular delivery • extravasation • toxicity
|
BACKGROUND |
|---|
|
TOPABSTRACTBACKGROUNDNANOSIZED TECHNOLOGIES FOR...INNER SPACETARGETINGTOXICITY ISSUESTHE FUTURE OF NANOMEDICINEREFERENCES |
|---|
|
NANOSIZED TECHNOLOGIES FOR MEDICAL IMAGING AND TARGETED DRUG DELIVERY |
|---|
|
TOPABSTRACTBACKGROUNDNANOSIZED TECHNOLOGIES FOR...INNER SPACETARGETINGTOXICITY ISSUESTHE FUTURE OF NANOMEDICINEREFERENCES |
|---|
. Colloidal gold, ironoxide crystals, and quantum dots (QDs) semiconductor nanocrystals are examples of nanoparticles, whose size is generally in the region of 1–20 nm, and have diagnostic applications in biology and medicine (Appendix 2, page 324). Gold nanoparticles have application as quenchers in fluorescence resonance energy transfer measurement studies (2
, 3)
. For example, the distance-dependent optical property of gold nanoparticles has provided opportunities for evaluation of the binding of DNA-conjugated gold nanoparticles to a complementary RNA sequence (3)
. Iron oxide nanocrystals with superparamagnetic properties are used as contrast agents in magnetic resonance imaging (MRI), as they cause changes in the spin-spin relaxation times of neighboring water molecules, to monitor gene expression or detect pathologies such as cancer, brain inflammation, arthritis, or atherosclerotic plaques (4
5
6
7
8
9
10
11)
. QDs can label biological systems for detection by optical or electrical means in vitro and to some extent in vivo (12
13
14
15
16
17
18
19)
. The fluorescence emission wavelength (from the UV to the near-IR) of QDs can be tuned by altering the particle size, thus these nanosystems have the potential to revolutionize cell, receptor, antigen, and enzyme imaging (19
20
21
22)
. Indeed, a recent report demonstrated the use of QDs for tracking metastatic tumor cell extravasation (20)
. Their large surface area-to-volume ratio offers potential for designing multifunctional nanosystems. Undoubtedly, application of such multi-wavelength optical nanotools may eventually aid our understanding of the complex regulatory and signaling networks that govern the behavior of cells in normal and disease states.
Nanovehicles and drug carriers
In addition, there are numerous engineered constructs, assemblies, architectures, and particulate systems, whose unifying feature is the nanometer scale size range (from a few to 250 nm). These include polymeric micelles, dendrimers, polymeric and ceramic nanoparticles, protein cage architectures, viral-derived capsid nanoparticles, polyplexes, and liposomes (Appendix 2) (23
24
25
26
27
28
29
30
31
32
33
34
35
36
37)
. First, therapeutic and diagnostic agents can be encapsulated, covalently attached, or adsorbed on to such nanocarriers. These approaches can easily overcome drug solubility issues, particularly with the view that large proportions of new drug candidates emerging from high-throughput drug screening initiatives are water insoluble. But some carriers have a poor capacity to incorporate active compounds (e.g., dendrimers, whose size is in the order of 5–10 nm). There are alternative nanoscale approaches for solubilization of water insoluble drugs too (38)
. One approach is to mill the substance and then stabilize smaller particles with a coating; this forms nanocrystals in size ranges suitable for oral delivery, as well as for intravenous injection (38
, 39)
. Thus, the reduced particle size entails high surface area and hence a strategy for faster drug release. Pharmacokinetic profiles of injectable nanocrystals may vary from rapidly soluble in the blood to slowly dissolving. Second, by virtue of their small size and by functionalizing their surface with synthetic polymers and appropriate ligands, nanoparticulate carriers can be targeted to specific cells and locations within the body after intravenous and subcutaneous routes of injection (23
24
25
, 31
, 35
, 36
, 40
41
42)
. Such approaches, may enhance detection sensitivity in medical imaging, improve therapeutic effectiveness, and decrease side effects. Some of the carriers can be engineered in such a way that they can be activated by changes in the environmental pH, chemical stimuli, by the application of a rapidly oscillating magnetic field, or by application of an external heat source (24
, 43
44
45)
. Such modifications offer control over particle integrity, drug delivery rates, and the location of drug release, for example within specific organelles. Some are being designed with the focus on multifunctionality; these carriers target cell receptors and delivers simultaneously drugs and biological sensors (46)
. Some include the incorporation of one or more nanosystems within other carriers, as in micellar encapsulation of QDs; this delineates the inherent nonspecific adsorption and aggregation of QDs in biological environments (16)
. In addition to these, nanoscale-based delivery strategies are beginning to make a significant impact on global pharmaceutical planning and marketing (market intelligence and life-cycle management) (23
, 25)
.
|
INNER SPACE |
|---|
|
TOPABSTRACTBACKGROUNDNANOSIZED TECHNOLOGIES FOR...INNER SPACETARGETINGTOXICITY ISSUESTHE FUTURE OF NANOMEDICINEREFERENCES |
|---|
Clearance mechanisms and opportunities for targeting
The network of blood and lymphatic vessels investing the body provides natural routes for the distribution of nutrients, clearing of unwanted materials, and delivery of therapeutic agents. Superficially, however, this network appears to provide little in the way of obvious controlled and specific access to tissues, and the science of these processes has been scant. Regardless of these limitations, nanoparticulate systems provide possibilities for access to cell populations and body compartments. When injected intravenously, particles are cleared rapidly from the circulation and predominantly by the liver (Kupffer cells) and the spleen (marginal zone and red pulp) macrophages (24)
. This site-specific, but passive, mode of clearance is a facet of the immune cells’ primary scavenging role for particulate invaders and self-effete products. Opsonization, which is surface deposition of blood opsonic factors such as fibronectin, immunoglobulins, and complement proteins, often aid particle recognition by these macrophages (24)
. However, size and surface characteristics of nanoparticles both play an important role in the blood opsonization processes and clearance kinetics (24
, 47
, 48)
. Larger particles (200 nm and above) are more efficient at activating the human complement system and are hence cleared faster from the blood by Kupffer cells than their smaller counterparts. This is a reflection of geometric factors and surface dynamics on the initial assembly of proteins involved in complement activation (24
, 48)
. The binding of blood proteins and opsonins to nanoparticles differ considerably in amount and in pattern depending on surface properties such as the presence and type of functional groups and surface charge density (24
, 47
, 48)
. Thus, differential opsonization may account for differences in clearance rates and macrophage sequestration of nanoparticles. This is particularly important with the view that macrophages are heterogeneous with respect to phenotype and physiological function, even within the same tissue. Hence, a particular population of phagocytes may employ one predominant recognition mechanism. The dynamic process of protein adsorption together with deposition of a variety of opsonic factors onto the surface of nanoparticles may indicate an arrangement based on a recognition hierarchy, or cooperativity, among macrophage receptors for clearance. For example, a specific macrophage receptor might recognize the earliest changes associated with a particle surface in the blood, whereas other receptors might recognize particles at a later stage thus ensuring complete removal from the circulation. These issues have not received detailed attention but their understanding could potentially open strategies for design and surface manipulation of nanosystems that target selective macrophage subpopulations.
Small particle size also means large surface area. This may pose problems in terms of aggregation of primary nanoparticles in the biological environment, which subsequently determines the effective particle size and hence clearance kinetics. Indeed, dendrimers and QDs are well known to flocculate in biological media. Another case is interaction between certain lipid-based nanosystems and lipoproteins leading to dramatic size changes. For instance, some hydrophobic components of Cremophor EL micelles incorporate rapidly into HDL and LDL leaving the remaining components to form large lipid structures in plasma that assembles C3 convertases (49)
. Thus, inherent in nanoparticle design is surface manipulation and engineering. Indeed, precision surface engineering with synthetic polymers can resolve aggregation and afford control over nanoparticle interaction and fate with biological systems. For example, polydentate phosphine coating or surface modification with various hydrophilic polymers renders QDs soluble, disperse, and stable in serum (17
18
19
, 50)
. There are numerous examples in the literature where the surface of nanocarriers is carefully assembled with projected "macromolecular hairs" made from poly(ethyleneglycol) or other related polymers (24)
. This strategy suppresses macrophage recognition by an array of complex mechanisms, which collectively achieve reduced protein adsorption and surface opsonization. Here, the efficiency of the process is dependent on polymer type, their surface stability, reactivity, and physics (e.g., surface density and conformation) (24
, 51
, 52)
. Indeed, this stealth-like behavior is similar to strategies developed by pathogenic microorganisms to fend off immune detection. Suppression of opsonization events is an important first step in enhancing passive retention of nanoparticles at sites and compartments other than macrophages in contact with the blood, and simply is a reflection of long circulatory profile of such surface-manipulated nanoparticles. In this instance, particle escape from the vasculature is normally restricted to sites where the capillaries have open fenestration, as in the sinus endothelium of the liver, or when the integrity of the endothelial barrier is perturbed by inflammatory processes or by tumor growth; the latter is the result of dysregulated angiogenesis (24
, 53
, 54)
. In liver, the size of fenestrae in the sinus endothelium can be as large as 150 nm; in tumor capillaries they vary greatly but rarely exceed 300 nm. Prolonged circulation properties are ideal for slow or controlled release of therapeutic agents in the blood to treat vascular disorders. Long circulating particles may have application in vascular imaging too (e.g., detection of vascular bleeding or abnormalities) or even act as artificial nanoscale red blood cells. But surface engineering is not the only way for keeping nanoparticles away from the macrophages. Recent advances in synthetic polymer chemistry afford precise control over the architecture and polydispersity of polymers, polymer-conjugates, and block copolymers. Some of these novel materials can form sterically stabilized nanoscale self-assembling structures with macrophage-evading properties (28
, 29)
.
If confinement to the vascular system is necessary, then splenic filtration processes must be borne in mind. Splenic filtration at interendothelial cell slits is predominant. This is particularly true for rigid or nondeformable particles whose size exceeds the width of the cell slits (200–250 nm) (24)
. Otherwise, opportunities are there for gaining efficient access to splenic red-pulp compartments with nanoparticles (Fig. 1
).
|
Recent developments in molecular biology have begun to reveal the wealth of information contained within blood and lymphatic vessels, and in particular that on the lumenal surface of endothelial cells. Molecular signatures related to particular vascular and lymphatic beds and types of endothelial cells have been identified, providing landmarks for circulating cells and molecules (55
56
57
58
59)
. These same signatures have now been exploited to direct therapeutic and diagnostic entities to selected pathological vessels, particularly those of cancer (19
, 60
, 61)
. This, however, requires assembly of the appropriate targeting ligands on nanocarriers and long circulating nanosystems, but the ultimate characteristics such as ligand density, spacing and conformation are dependent on ligand and particle properties (curvature and surface reactivity). These modifications determine the extent of particle stability and aggregation in vivo as well as the efficiency of receptor binding and follow up events, such as the mode of particle internalization (e.g., receptor-mediated endocytosis as opposed to lipid raft-dependent macropinocytosis) and associated signaling processes. Perhaps, the predetermined chemical architecture of dendrimetric systems could be taken into advantage for optimizing ligand conformation. Nevertheless, such entities may be viewed as tiny submarines that navigate capillaries in search of signature molecules expressed by the target, a process often refereed to as active targeting. This concept is applicable for targeting of nanosytems to other accessible cells within the vasculature (e.g., blood cells and macrophages) or even to cells and scaffolding structures after extravasation from the vasculature.
Interstitial injection of nanoparticles, however, may be the preferred choice particularly if the target is a specific lymph node or a group of them located regionally (62
, 63)
. In lymphatic capillaries, numerous endothelial cells overlap extensively at their margins and they lack adhesion mechanisms at many points. Immediately after interstitial injection many of the overlapped endothelial cells are separated and thus passageways are provided between the interstitium and lymphatic lumen through which particles are conveyed to the nodes via the afferent lymph (62)
. In lymph nodes, macrophages of medullar sinuses and paracortex are mainly responsible for particle capture from the lymph (62)
. The fate of interstitially injected nanoparticles is controlled by their size and surface characteristics (62
63
64)
. The size of the particles must be large enough to prevent their rapid leakage into the blood capillaries; particles in the range of 30 to 100 nm usually satisfy this criterion, whereas particles larger than 100 nm move very slowly and are susceptible to clearance by interstitial macrophages. Surface characteristics can control the extent of particle aggregation at interstitial sites, hence drainage kinetics and lymph node retention may be optimized by prior surface manipulation (64)
. For instance, hydrophilic nanoparticles, as opposed to their hydrophobic counterparts, repulse each other and interact poorly with the ground substance of the interstitium and drain rapidly into the initial lymphatics (Fig. 2
).
|
Very small particles (1–20 nm), particularly when they are long circulatory, can slowly extravasate from the vasculature into the interstitial spaces, from which they are transported to lymph nodes by way of lymphatic vessels (62)
. Depending on the species, particles may leave the blood pool through the permeable endothelium in lymph nodes, thus the extent of nodal vascularization and blood supply become important. Nevertheless, these modes of particle movement from the blood and interstitial sites to the lymph nodes provide intriguing opportunities for diagnostic imaging where, essentially, only an enhancement of signal over background is required; the physicochemical properties of QDs (50)
and superparamagnetic iron oxide nanocrystals (8)
makes them ideal for such purposes.
|
TARGETING |
|---|
|
TOPABSTRACTBACKGROUNDNANOSIZED TECHNOLOGIES FOR...INNER SPACETARGETINGTOXICITY ISSUESTHE FUTURE OF NANOMEDICINEREFERENCES |
|---|
24
25
26)
. For example, although most microorganisms are killed by macrophages, many pathogenic organisms have developed means for resisting macrophage destruction following phagocytosis. In certain cases, the macrophage lysosome and/or cytoplasm is the obligate intracellular home of the microorganism, examples include Toxoplasma gondii, various species of Leishmania, Mycobacterium tuberculosis, and Listeria monocytogenes. Passive targeting of nanoparticulate vehicles with encapsulated antimicrobial agents to infected macrophages is therefore a logical strategy for effective microbial killing (23
, 24
, 65
66
67)
. The endocytic pathway will direct the carrier to lysosomes where pathogens are resident. Degradation of the carrier by lysosomal enzymes releases drug into the phagosome-lysosome vesicle itself or into the cytoplasm either by diffusion or by specific transporters, depending on the physicochemical nature of the drug molecule. Approved formulations for human subjects are limited to lipid-based nanosytems (100–200 nm) with entrapped amophotericin B (Amp-B), and are recommended for treatment of visceral leishmaniasis or confirmed infections caused by specific fungal species (23
, 65
, 68)
. This mode of targeting has significantly reduced the required clinically effective quantity of Amp-B for treatment, achieving therapeutic drug concentrations in the infected macrophages. Other beneficial effects include significant reduction in nephrotoxicity, a common side effect associated with Amp-B administration, and proinflammatory cytokine release (68
, 69)
. To further alleviate nephrotoxicity, combination strategies may be sought. One possibility for exploration is liposome emulsification, as this may further stabilize Amp-B association with the carrier. Others have used multifunctional carriers to deliver antimicrobials to macrophages (67)
. For example, intravenous injection of tuftsin-bearing liposomes to infected animals have not only resulted in delivery of liposome-encapsulated drugs to the macrophage phagolysosomes, but also in the nonspecific stimulation of liver and spleen macrophage functions against parasitic, fungal and bacterial infections (67)
. The latter effect is due to the binding of tuftsin to its receptor, which further incites macrophage antimicrobial responses.
Endocytic delivery is a route for macrophage destruction. Recently nanocarrier-mediated macrophage suicide (delivery of macrophage toxins) has proved to be a powerful approach in removing unwanted macrophages in gene therapy and other clinically relevant situations such as autoimmune blood disorders, T cell-mediated autoimmune diabetes, rheumatoid arthritis, spinal cord injury, sciatic nerve injury, and restenosis after angioplasty (70
71
72
73
74
75)
. Alternatively, microbial-induced macrophage apoptosis strategies may be exploited to design and surface-engineer nanoparticles with macrophage-killing properties. One example is through targeting and selective activation of Toll-like receptor 4 (76)
.
Macrophages and dendritic cells play critical roles in determining immunogenicity and the generation of appropriate immune responses. Systems, such as numerous polymeric and ceramic nanospheres, nanoemulsions, liposomes, protein cage architectures, and viral-derived nanoparticles act as powerful adjuvants if they are physically or covalently associated with protein antigens (33
, 77
78
79)
. After endocytic uptake of nanoparticles, macrophages partially degrade the entrapped antigens and channel peptides into the MHC molecules (class I or II) for processing and presentation. Thus, there is considerable potential for nanoparticulate adjuvants for the development of new-generation vaccines made either recombinantly or from synthetic peptide antigens that are less or nonimmunogenic in their own right. Genetic immunization with nanoparticles has also received attention but the majority of attempts are based on cationic systems to allow DNA compaction (80
, 81)
.
Recent advances in cell biology have provided a wealth of information regarding the structure, recognition properties, and signaling functions of a variety of macrophage/dendritic cells receptors, particularly those that affect immunogenicity or adjuvanticity. Harnessing these receptors as therapeutic targets may prove a better strategy for antigen delivery and targeting with particulate nanocarriers. For example, in addition to endocytic and phagocytic roles, the mannose receptor plays an important role in antigen recognition and transport within lymphoid organs (82)
. Antigen coupling to oxidized mannan leads to rapid recognition by the mannose receptor, followed by internalization, efficient MHC class I presentation to CD8+ cells, and a preferential T1 response. It is the presence of the aldehydes, and not Schiff bases, that cause rapid entry into the class I pathway (82)
. Conversely, after reduction processes, internalization and presentation to MHC class I molecules are diminished by 1000-fold; the outcome is class II presentation and a T2 immune response. The soluble form of mannose receptor, or recombinant probes containing the cysteine-rich domain of the mannose receptor, may also be used for antigen transport. The N-terminal cysteine-rich domain of the mannose receptor is a counter-receptor for sulfated glycoprotein cysteine-rich domain ligand expressed by subpopulations of myeloid cells in secondary lymphoid organs (e.g., murine splenic metallophilic macrophages, subcapsular sinus macrophages in lymph nodes, follicular dendritic cells and migratory dendritic cells) (83)
. Indeed, the lectin activity of the cysteine-rich domain was recently exploited for targeting of acidic glycan-decorated proteins to metallopilic macrophages (83)
. Dectin-1 is another mannan-dependent receptor with phagocytic activity, which is expressed by macrophages, neutrophils and dendritic cells (84)
. This receptor recognizes ß-glucans from fungi and yeasts and may potentially mediate immunostimulatory effects of ß-glucans. Dectin-1 contains an immunoreceptor tyrosine-based activation motif in its cytoplasmic tail, thus ligand recognition may lead to activation of signaling pathways in myeloid cells. Dendritic cell receptors such as DEC-205 and DEC-SIGN have been implicated in antigen internalization and presentation to T cells (85
86
87)
. For example, DEC-205 can recycle and enhance antigen presentation via MHC class II-positive lysosomal compartments. An interesting approach would be to design and evaluate nanoparticles that target a combination of antigens and activators to immature dendritic cells in vivo for stimulating immune responses, or other appropriate agents to silence immunity. For instance, direct activation and dendritic cell migration may be achieved by specific targeting of Toll-like receptor ligands through dendritic cell specific molecules.
Endocytic loading of macrophages with nanotechnology-based contrast agents is an intriguing approach for disease detection; here the surrounding parenchyma, and not the pathology, will change in intensity. For example, after extravasation from the vasculature and lymph node arrival, intravenously injected paramagnetic iron oxide nanocrystals are sequestered by resident lymph node macrophages, and their intracellular accumulation shortens the spin relaxation process of nearby protons detectable by magnetic resonance imaging. In magnetic resonance imaging, those node regions accumulating iron oxide crystals appear dark relative to surrounding tissues. Indeed, this development has facilitated distinction between normal and tumor-bearing nodes or reactive and metastatic nodes, which is a desirable approach since a single intravenous injection provides access to a large number of lymph nodes (62)
. Recently, using this approach very small metastases (less than 2 mm in diameter) within normal-sized lymph nodes in patients with prostate cancer were identified (8)
. This is important, since microscopic tumor deposits are below the detection threshold of other advanced imaging techniques. Superparamagnetic nanoconstructs have aided visualization of vascular pathologies in arthritis (9)
and atherosclerotic plaques (10
, 11)
; this is a reflection of increased endothelial permeability and access to resident macrophages. Similarly, capture of near-infrared fluorescent type II quantum dots by lymph node macrophages has allowed sentinel lymph node mapping and visualization of deeply located lymph nodes (50)
.
Endothelium as a target
The concept of targeting to the blood vessels is an attractive one, particularly with the view that the endothelium plays an important role in a number of pathological processes including cancer (dysregulated angiogenesis), inflammation, oxidative stress and thrombosis. Indeed, a number of studies have demonstrated a level of control of arrest and distribution of passively targeted nanoparticles by specific endothelial cells, and these were linked to the surface properties of the carrier. For instance, early studies of polystyrene nanoparticles, designed to minimize Kupffer cell uptake, indicated exclusive arrest by the bone marrow sinus lining endothelial cells in rabbits (88)
. Arrest was followed by receptor-mediated internalization. Another example is the localization of intravenously injected polysorbate 80-coated nanoparticles to murine and rat blood-brain endothelial cells (89
, 90)
. Recent studies have shown that cationic liposomes within 1 h of entering the circulation, are internalized into endosomes and lysosomes of endothelial cells in a characteristic organ- and vessel-specific manner (91)
. These patterns seem to bear no relationship to the morphological characteristics of the endothelium associated with a particular site, but probably reflect vessel-specific expression of receptors for which such particles, or their surface-associated blood proteins, are ligands.
But recent dramatic progress in the development of a human vascular map, in particular through the application of cell and molecular biological tools such as serial analysis of gene expression (SAGE), subtractive proteomic mapping, and in vivo phage display, is generating yet another level of possibilities for specific targeting of drugs and biological agents (55
56
57
58
59)
. For example, SAGE examines the spectrum of mRNA species within a cell or tissue and allows rapid comparison with other cell types or tissues. Phage, however, behaves as a nanomachine; it can be engineered to display numerous peptides on its surface and after injection one can select peptides that make the phage home to a given target. Recently, Arap et al. (56)
described the first attempt at mapping the vasculature of a human patient using a phage display technique. The authors surveyed 47,160 sequences that localized to different organs within the patient, and determined that in many cases these sequences were similar to those of known ligands for endothelial cell-surface proteins, thus validating this technique for identifying novel endothelial targets in humans. This approach is now being used in a number of ways to target therapeutic agents, particularly to the vasculature of solid tumors. Examples include integrins 
vß3, vß5 and 5ß1, which are up-regulated in angiogenic endothelial cells and play a role in the process of angiogenesis (55
, 58
, 92
, 93)
. They bind with high affinity to sequences containing a characteristic RGD (Arg-Gly-Asp) motif, which seems to be central to anti-integrin approaches. Indeed, in vivo phage display studies by Assa-Munt et al. (94)
have led to the development of a cyclic nonapeptide RGD-4C, which avidly binds to the integrins vß3 and vß5. Coupling of RGD-4C to doxorubicin yielded a compound significantly more effective than doxorubicin alone, and with less side effects to the heart and liver, the main sites of doxorubicin toxicity (95)
. Zitzmann et al. (96)
have shown that the homing peptide RGD-4C binds to human breast cancer cells grown as xenografts in nude mice, as well as to endothelial cells, but not to nonendothelial normal cells. Inserting RGD sequences into adenovirus surface protein has been used to affect the tropism of the viral gene therapy vectors for targeting purposes (97
, 98)
. Other discovered binding sequences, such as the HWGF peptide motif, can bind to matrix metalloproteases-2 and -9, and dramatically enhances adenoviral trpoism to the endothelial and smooth muscle cells of large blood vessels (99)
. The hexapeptide NGR is another homing molecule derived from in vivo phage display studies (95)
. This peptide specifically binds to aminopeptidase N on angiogenic endothelial cells and when coupled to the cytokine TNF, it produces 8- to 10-fold enhancement of the therapeutic efficacy of doxorubicin and melphalan against murine tumors (100)
. Phage display studies have further revealed the ability of LyP-1, a cyclic nonapeptide, to recognize tumor lymphatic vessels (as well as tumor cells in certain tumors) and translocate to the nucleus (101)
. Although some of the described antiangiogeneic attempts are promising, it appears that local destruction of endothelial cells committed to an angiogenic program will not be sufficient to destroy the tumor. A significant proportion of endothelial cells will be recruited from the bone marrow pool as well as the circulating blood, and endothelial cells lost through treatment or natural processes, will be replaced on demand (102)
. Therefore future therapeutic approaches must acknowledge this population. By the same token, this new population, which will ultimately differentiate into mature endothelial cells, must by definition express a unique set of cell surface markers, consistent with its progenitor role, which identifies it and facilitates targeting.
Nevertheless, several studies have now combined the specificity of endothelial molecular markers with nanoparticles. For example, as a novel anti-angiogenic strategy targeted at solid tumors, some investigators have used a synthetic analog of vß3 to target therapeutic genes complexed with cationic nanoparticles at tumor-associated endothelial cells (103)
. Similar approaches have now been extended for site-specific imaging with vß3-targeted paramagnetic nanoparticles. This attempt detected and characterized early angiogenesis induced by minute solid tumors with magnetic resonance imaging (104)
. This is a valuable tool with which to phenotypic categorization and patient selection as well as track the effectiveness of antitumor treatment regimens. Similarly, selective targeting of peptide-coated QDs to blood and lymphatic vessels in tumors have been demonstrated (19)
. Another interesting approach was the ability of NGR motif-decorated liposomes to specifically attack tumors by shutting down their blood supply (105)
.
Other important targets expressed on the endothelium for therapeutic drug delivery include cell adhesion molecules (CAM). Intercellular cell adhesion molecule-1 (ICAM-1) and platelet endothelial cell adhesion molecule-1 (PECAM-1) are perhaps the most studied of the CAMs as targets. Antibodies against ICAM-1 and PECAM-1, which could theoretically be used to deliver therapeutic cargoes to endothelial cells are not taken up. However, anti-ICAM-1 and anti-PECAM-1 nanoparticles are readily endocytosed, by a unique but poorly understood pathway (106)
. Such anti-CAM nanoparticles have been shown to successfully deliver a variety of cargoes to pulmonary and cardiac endothelium in vivo (107
108
109)
. Endothelial leukocyte cell adhesion molecule-1 and vascular adhesion molecule-1, which are moderately expressed on the vascular endothelium of tumors having marked leukocytic infiltration (e.g., solid Hodgkin’s tumors, breast carcinoma and non-small cell lung carcinoma), have served as targets for therapeutic interventions with nano-constructs (110)
.
Extravasation: targeting of solid cancers
The development of "stealth" technologies has provided opportunities for passive accumulation of intravenously injected nanoparticles (20–150 nm) in pathological sites expressing "leaky" vasculature by extravasation (24)
. Although, attempts have included delivery of drugs and imaging agents with different nanoscale technologies to the underlying parenchyma of injured arteries and rheumatoid arthritis, the majority of efforts are concentrated on solid tumors (23
24
25
, 28
, 29
, 40
, 42
, 111
112
113
114)
.
As a result of perfusion heterogeneity, the spatial distribution of stealth nanoparticles in solid tumors is heterogeneous and unpredictable. As has been elegantly demonstrated by Jain (115)
structural and functional abnormalities of blood and lymphatic vessels within solid tumors impede efficient delivery of not only systemic nanoparticles, but macromolecules. Already compromised by abnormal hydrostatic pressure gradients, compressive mechanical forces generated by tumor cell proliferation cause intratumoral vessels to compress and collapse (116)
. Tumor-specific cytotoxic therapy, reducing tumor cell number, may result in more efficient delivery, by decompressing these same vessels (116)
; however, this enhanced perfusion could provide a route for metastasis. Distribution, organization and relative levels of collagen, decorin, and hyaluronan impede the diffusion of macromolecules and nanoparticles in tumors (117)
. Thus, diffusion of macromolecules and nanoparticles will vary with tumor types, anatomical locations, and possibly by factors that influence extracellular matrix composition and/or structure.
In spite of these limitations, there are regulatory approved formulations of long circulating liposomes with entrapped doxorubicin for management/treatment of AIDS-related Kaposi’s sarcoma, refractory ovarian cancer, and metastatic breast cancer (23
, 118)
. These formulations exhibit favorable pharmacokinetics when compared with the free drug, for example the area under the curve after a dose of 50 mg/m2 doxorubicin encapsulated in stealth liposomes is 
300-fold greater than that of free doxorubicin. Clearance and volume of distribution are reduced by at least 250- and 60-fold, respectively (118)
. As a result of these promising pharmacokinetic profiles, stealth doxorubicin containing liposomes are currently undergoing additional early- to late-phase clinical trials.
A number of engineering issues must be considered when applying particulate stealth systems for cancer drug delivery. First, the carrier must have a high drug loading capacity and remain stable within the vasculature with minimum drug loss. This criterion is met with the current regulatory approved stealth liposome formulation of doxorubicin. Here, doxorubicin is loaded actively by an ammonium sulfate gradient (as doxorubicin sulfate) yielding highly stable liposomes with high contents of doxorubicin aggregates (119)
. Second, it has been widely established that the majority of extravasated particulate systems, such as liposomes, do not interact with target cancer cells (111)
. They are often distributed heterogeneously in perivascular clusters that do not move significantly. The process of particle extravasation must be followed by the efflux of drug from the carrier, resulting in target exposure (being tumor cells, tumor-associated macrophages, components of tumor vasculature, or extracellular mediators such as proangiogenic proteases) to drug molecules. Here, the drug must be released at a rate that maintains free drug levels in the therapeutic range. However, the rate of drug release from liposomes depends on drug type and the encapsulation method (119)
. For example, stealth liposomes with entrapped cisplatin (where cisplatin is loaded passively) lack antitumor activity despite accumulation in tumor interstitial sites (120)
; this is simply due to the poor and extremely slow release of cisplatin from extravasated liposomes. This is in contrast to effective antitumor property of the same liposomal lipid composition but with entrapped doxorubicin. It is suggested that nonspecific chemical disruption, or collapse of the liposomal pH gradient, which was initially used to load liposomes actively with doxorubicin, triggers doxorubicin release (perhaps, in both monomeric and aggregated forms) from intact vesicles in a slow manner (119)
. If this proposed mechanism is true, then some of the antitumor effects of liposomal doxorubicin may be ascribed to antiangiogenesis since these vesicles accumulate at tumor interstitial sites closed to proangiogeneic vessels. Thus, the therapeutic potential of liposomal doxorubicin in animal models with drug-resistance tumors is worthy of exploration. Nonetheless, the most prominent feature of liposomal doxorubicin formulations has been a decrease in many of the side effects associated with free doxorubicin or doxorubicin encapsulated in macrophage-prone nanocarriers rather than an increase in potency (111
, 118)
.
The issue of drug release from nanocarriers remains central to cancer chemotherapy. Therefore, application of a number of solid and polymeric nanoparticles for cancer drug delivery must be viewed cautiously since drug molecules may not be released from extravasated nanoparticles at sufficient rates. To date, the most effective approaches are described with liposomes and to some extent with polymeric micelles, although the latter constructs have a low encapsulation volume. There are a number of biochemical-based advances that can trigger drug release from accumulated liposomes at interstitial sites. One attractive approach is enzyme-mediated liposome destabilization. For example, local concentration of secretory phospholipase A2 (sPLA2) is highly elevated in inflammatory and cancer tissues (121
, 122)
. The ester linkage in the sn-2 position of phospholipid is rapidly hydrolyzed by sPLA2. This leads to production of a fatty acid and a lysolipid, which have a synergistic membrane perturbing and permeabilizing effect (123)
. The activity of sPLA2 is much higher toward aggregated phospholipids (liposomes and micelles) than lipid monomers. sPLA2 can destabilize sterically protected liposomes bearing the substrate (121)
. So, by choosing an appropriate lipid composition one may control the extent of vesicle destabilization by sPLA2 (124)
. For example, stealth vesicles with entrapped cisplatin but composed of a nontoxic sPLA2 sensitive lipid prodrug (an ether lipid) were extremely effective in killing cancer cells in the presence of sPLA2, whereas the corresponding control formulation without sPLA2 sensitive lipids expressed insignificant cytotoxicity. Future efforts may consider combination strategies with sPLA2 sensitive and insensitive stealth liposomes, thus generating both rapid and prolonged interstitial drug release profiles in pathologies with elevated sPLA2 activity.
There are several approaches, which describe active targeting of nanoparticles to cancer cells and related extracellular elements, but with these methods the delivery part is still passive. One interesting method was in vivo detection and imaging of tumor-associated matrix metalloproteinase-7 (matrilysin) activity with a dendrimer-based fluorogenic substrate, which served as a selective "proteolytic beacon" for matrilysin (125)
. Others have used active targeting in combination with stealth technology. For example, tumor-specific monoclonal antibodies, and particularly internalizing epitopes, or ligands, such as folic acid, have been attached to the distal end of poly(ethyleneglycol) chains expressed on the surface of long circulating or macrophage-evading nanocarriers (24
, 40
41
42
, 111
, 119
, 126
, 127)
. Folate-targeting is an interesting approach and offers a number of advantages over monoclonal antibodies. The folate receptor expression is restricted to human ovarian, endometrial, colorectal and lung cancers; the receptor is generally absent in most normal human tissues with the exception of a few sites (41)
. First, folate is nonimmunogenic. Second, folic acid-decorated nanocarriers are rapidly internalized by folate receptor-bearing cancer cells and internalization is associated with rapid receptor recycling to the plasma membrane (41)
. This receptor-mediated internalization approach is believed to bypass cancer cell multidrug-efflux pumps (127)
. But, a major problem is the extent of drug release from internalized nanocarriers. For example, in an acidic compartment such as endosome/lysosome local conditions may not favor rapid doxorubicin release from the ammonium sulfate gradient loaded doxorubicin liposomes (bearing in mind that the drug reaches acidic cellular compartments as doxorubicin sulfate salt, and the drug concentration in the liposome aqueous phase is far above the drug solubility product) (119)
.
Nanoparticles for cytoplasmic drug delivery
Breaching of the endosomal membrane is particularly important for priming MHC class I-restricted cytotoxic T lymphocyte responses, for survival of genetic materials against nuclease degradation in the lysosomal compartment, or for those drugs that must reach cytoplasm in sufficient quantities (as for treatment of cytoplasmic infections or reaching nuclear receptors) after endocytic delivery with nanoparticulate carriers. Here, there are advances in particle engineering too. For instance, nanoparticles made from poly(DL-lactide-co-glycolide) can escape the endo-lysosomal compartment within minutes of internalization in intact form and reach the cytoplasm (44)
. The mechanism of rapid escape is by selective reversal of the surface charge of nanoparticles from the anionic to the cationic state in endo-lysosomes, thus resulting in a local particle-membrane interaction with subsequent cytoplasmic release. Another impressive approach for cytoplasmic delivery of nanoparticles is their surface manipulation with short peptides known as protein transduction domains such as HIV-1 TAT protein transduction domain (TAT PTD), which is a short basic region comprising residues 48–57, or heterologous recombinant TAT-fusion peptides (128
129
130)
. The electrostatic interaction between the cationic TAT PTD and negatively charged cell-surface constituents, such as heparan sulfate proteoglycans and glycoproteins containing sialic acids, is a necessary event before internalization (131)
. After this ionic interaction, cellular uptake occurs by lipid raft-dependent macropinocytosis in a receptor-independent manner; this is followed by a pH drop and destabilization of integrity of the macropinosome vesicle lipid bilayer, which ultimately results in the release of TAT-cargo into the cytosol (131)
. This mode of entry may further suggest the avidity of TAT PTD for glycophosphoinositol-anchored glycoproteins, which are present in lipid rafts, or binding to cholesterol membrane constituents that trigger macropinocytosis. An important feature of macropinosomes is that they do not fuse into lysosomes to degrade their contents. Although, these approaches have the potential to deliver and release drugs cytoplasmically for a sustained therapeutic effect in conditions such as cancer and stroke, possible cytotoxicity arising from the carrier components cannot be ruled out and warrants detailed investigation (132)
.
A number of gene transfer approaches are also based on destabilization of internalized vesicles via the surface charge effect. For instance, a wide range of cationic particles and synthetic polycations in linear, branched, or dendrimer form have been used to condense DNA, antisense oligonucleotides, and small interfering RNAs into nanostructures amenable to cellular internalization via endocytosis (34
, 133
, 134)
. For example, poly(ethylenimine)s are believed to act as proton sponges; they buffer the low pH in the endosomes and potentially induce membrane rupture, resulting in the release of polycation/nucleic complexes into the cytoplasm (134)
.
Other examples include pH-sensitive microparticles (e.g., spray-dried particles made from a combination of phospholipids and the pH-sensitive polymethylacrylate (Eudragit E100) and pH-sensitive and fusogenic liposomes, which can efficiently smuggle their cargo into cytosol (43
, 135)
. Vesicles that are pH-sensitive are assembled from a combination of unsaturated phosphatidylethanolamine and mildly acidic amphiphiles; this composition confers bilayer stability at neutral pH or above but destabilizes and becomes fusion-competent at the acidic pH of late endosomes (43)
. Alternatively, by mimicking the characteristics of certain toxins or viruses, cytoplasmic delivery of agents can be enhanced. This requires co-encapsulation of bacterial pore-forming toxins (e.g., listeriolysin O) or fusion peptides found in the envelope of glycoproteins of certain viruses in particulate vehicles (136)
. Indeed, an influenza vaccine based on the latter principle is currently available for human use (79)
. This vaccine is administered parenterally and is well tolerated in children, young adults and the elderly. The influenza strains chosen are dependent on the yearly recommendations of the World Health Organization. The delivery system is comprised of unilamellar vesicles of 150 nm in diameter, but intercalated into their lipid bilayer are viral components, which include neuraminidase and hemagglutinnin (HA) glycoproteins. The mode of action of these virus-like liposomes (virosomes) is dependent on HA glycoproteins, the major antigens of the influenza virus. The HA is composed of two subunits, HA1 and HA2. The first subunit has high affinity for sialic acid present on the surface of antigen presenting cells thus facilitating virosome binding. The HA2 subunit is a fusion peptide and is activated at low pH (5.0). Hence, in late endosomes, where pH is acidic, the virosome becomes fusion-competent; this process releases entrapped antigens into the cytosol for subsequent processing and presentation (79)
. Similarly, Wadia et al. (131)
have generated a fusogenic influenza dTAT-HA2 peptide; this complex specifically enhances macropinosome escape.
|
TOXICITY ISSUES |
|---|
|
TOPABSTRACTBACKGROUNDNANOSIZED TECHNOLOGIES FOR...INNER SPACETARGETINGTOXICITY ISSUESTHE FUTURE OF NANOMEDICINEREFERENCES |
|---|
138
139)
. For instance, exposure of human keratinocytes to insoluble single-wall carbon nanotubes (see also Appendix 1) was associated with oxidative stress and apoptosis (139)
. The issue of toxicity becomes even more serious for intravenously injected nanoparticles, as size partly determines tissue distribution. Thus, what is the ultimate fate of nanocarriers and their constituents in the body, and particularly those which are not bio-degradable such as functionalized carbon nanotubes and coating agents such as poly(ethyleneglycol)? Can these constituents or their degradation products exert untoward immunological and pharmacological activities? Can polymeric vectors used for gene delivery as well as other polymer-based biomaterials interfere with cellular machineries or induce altered gene expression? If so, what are the long-term consequences? Finally, to what extent can we translate cellular and immunological toxicity results observed in animal models to humans, as there are distinct intra- and interspecies variation.
Cell death and altered gene expression
Recent evidence is drawing attention to some of the above questions, but investigation in this avenue of research is scant. For example, though much has been made of the promise of cadmium selenide QDs in imaging, little is known about their metabolism and potential deleterious effects. However, cadmium selenide QDs are lethal to cells under UV irradiation, as this releases highly toxic cadmium ions (140)
. Some polymeric micelles depending on the nature of their monomer constituents, can induce cell death via apoptosis or necrosis, or both (141
, 142)
. Differential gene expression has been reported in certain cells after cisplatin delivery with polymeric micelles when compared with that of free cisplatin treatment (143)
. Degradation products arising from poly(L-lactic acid) particles show cytotoxicity, at least, to immune cells (144)
, thus raising concern over their application for sustained cytosolic drug release. Certain polymeric constituents used in nanoparticle design and engineering act as inhibitors of P-glycoprotein efflux pumps expressed in polarized endothelial cells that form the exterior of the blood-brain barrier (145)
, and could potentially interfere with transport of a number of modulators and homeostatic mediators in the central nervous system (146)
. Multidrug resistant tumors express these pumps, and recent studies have demonstrated that the inhibition of such pumps by synthetic polymers is due to cell sensitization as well as ATP depletion (147)
. Remarkably, these mechanisms were correlated with low expression of ATP binding cassette genes after polymer treatment, indicating that these polymers have the ability to modulate gene expression but the molecular basis of these events remains unknown (147)
. During the past few years we have witnessed a surge in the development and creation of polymeric self-assemblies and nanofibers that act as scaffolds for cell attachment, proliferation, and encapsulation, which are intended to be used as synthetic replacements for biological tissues (148
149
150)
. The effect of such engineered polymeric structures on gene expression therefore is critical in tissue engineering and cell therapy, as these materials may initiate a number of unexpected effects. Conversely, cell-material interaction may identify a host of materials effects that offer new levels of control over cell behavior. Indeed, these predictions have been highlighted by recent developments in nanoliter scale synthesis of arrayed biomaterials, which are providing rapid and more insight into stem cell-material interactions (150)
. For example, some polymers supported high levels of stem cell differentiation into epithelial-like cells, whereas other polymers helped stem cell growth in the absence of certain growth factors.
Cell death and gene therapy
A very clear warning is evident from the poor success in human gene therapy with viruses. Although, viral vectors are extremely efficient delivery systems for nucleic acids, they can induce severe immunotoxicity as well as inadvertent gene expression changes after random integration into the host genome (151
, 152)
. These issues have generated a surge in design and engineering of synthetic polycationic nonviral gene transfer systems. However, the polycationic nature of the gene-delivery vehicles can induce immediate or delayed cytotoxicity by mechanisms involving necrosis as well as apoptosis (153
, 154)
. Necrosis may occur as a result of membrane destabilization or pore formation after interaction between the cationic components of the delivery system with cell surface proteoglycans and negatively charged proteins in cytoskeleton, such as actin. In the case of Jurkat T cells the apoptotic mechanism appears to be due to polycation-mediated release of Bcl-2-sensitive proteins such as cytochrome c from the mitochondrial intermembrane space and altered mitochondrial functions (Fig. 3
). However, different cationic materials, and depending on their molecular weights and polydispersity, may initiate apoptosis at different times and by different mechanisms or modes. The effect of these materials on cell death may depend on cell nature, mitochondrial content and the extent mitochondrial heterogeneity. Nevertheless, cytotoxic gene-delivery systems may compromise transcription and translation processes and potentially limit protein expression. In protocols, which attempt to restore gene function, for instance in metabolic disorders, such toxicity issues take on even greater importance. In addition to these, cDNA microarray expression profiling studies have recently revealed marked changes in the expression of cell proliferation, differentiation and proapoptotic genes in human epithelial cells, after treatment with cationic formulations (155)
. This raises further concern as to whether such delivery systems could adversely influence the desired effects of the delivered genetic agents. For instance cationic carriers may exacerbate, attenuate or even mask the effects of delivered nucleic acids. Thus, gene transfer/therapy represents an important area where smart macromolecular design and engineering is critical to achieving a successful outcome in the near future and could benefit through recent advances in high-throughput approaches to polymer design and screening (156
, 157)
. Such approaches may lead to understanding of the molecular basis of interaction between cationic polymers and mitochondrial and nuclear membrane as well as cationic polymers and BCL-2 family of proteins comprising inhibitors and inducers of apoptosis.
|
Pseudoallergy and idiosyncratic reactions
Finally, another potential pitfall associated with nanocarrier infusion into human subjects is the generation of non-IgE-mediated signs of hypersensitivity (158)
. These reactions are idiosyncratic and are believed to be secondary to complement activation, and presumably are a reflection of an individual’s immune cell sensitivity to complement-derived mediators (158
, 159)
. Hypersensitivity can be ameliorated by slowing the rate of infusion or by patient premedication, and often fails to appear on repeat administration of the nanocarrier. Idiosyncratic reactions occur after infusion of stealth systems, such as poly(ethylene glycol)-grafted liposomes (158
, 160)
. Refined surface engineering may eventually eliminate such side effects, for example by better polymer design, linkage modification, controlling the conformation and packing of grafted polymers and/or by introducing complement regulatory proteins or inhibitors on to the nanoparticle surface (24
, 47
, 52)
. However, the ultimate goal is to understand the molecular mechanism of complement activation-related pseudoallergy, which operates in a small population of individuals. Future developments in immunogenomics and predictive gene-derived toxicogenomic may eventually provide new methods for assessing an individual’s sensitivity to nanomedicines and hence reduce the risk of immune-mediated side effects.
|
THE FUTURE OF NANOMEDICINE |
|---|
|
TOPABSTRACTBACKGROUNDNANOSIZED TECHNOLOGIES FOR...INNER SPACETARGETINGTOXICITY ISSUESTHE FUTURE OF NANOMEDICINEREFERENCES |
|---|
