Combination and other strategies for colon targeting

Combination and other strategies for colon targeting

Physiological differences between the colon and the small intestine, such as intraluminal pressure and the level of hydration, can also be utilized to design a colon-targeted DDS. For example, Takada and colleagues uti-lized the higher intraluminal pressure in the colon and its low hydration state as a trigger mechanism for drug release. To utilize this as a trigger for drug release, the authors prepared liquid-filled hard gelatin capsules coated with an insoluble ethyl cellulose film. The drug was dissolved in a water soluble or insoluble semisolid base, such as PEG 1000, which liquifies at body temperature. After oral administration of the capsule, it behaves as a flexible membrane balloon with encapsulated drug, thus maintaining integrity during small intestinal transit. On reaching the colon, reabsorption of water leads to increased viscosity of the contents of the ethyl cellulose balloon, leading to its fragility and disintegration under higher pressure. The authors identified the thickness of the water-insoluble ethyl cellulose membrane as the key factor that controls drug release. Using this system, the authors demonstrated targeted delivery to the human colon using caffeine as a model drug and glycyrrhizin in dogs.

In addition to targeted drug release in the colon, the dosage form may incorporate a bioadhesive polymer to prolong the duration of time the dos-age form stays in the colon. The polymers that can be used for this purpose include polycarbophils, polyurethanes, and poly(ethylene oxide—propylene oxide) copolymers. Utilizing this strategy, Kakoulides et al. synthesized azo cross-linked bioadhesive acrylic polymers. The cross-linking prevents hydra-tion and swelling in the upper intestinal tract. On degradation of azo bonds in the large intestine, hydrogel swelling and bioadhesion was expected to lead to drug release and prolonged residence in the colonic environment.

Similarly, Gao et al. synthesized a conjugate of bioadhesive polymer N-(2-hydroxypropyl)methacrylamide (HPMA) and the drug 9-aminocamp-tothecin (9-AC) via a spacer containing a combination of an aromatic azo bond and a 4-aminobenzylcarbamate group. The spacer was designed to release the drug by azo bond cleavage in the colonic microenvironment. In subsequent studies, the authors observed colon targeting in mouse53 and rat54 models for the treatment of colon cancer. After oral administration of equal doses of the polymer conjugate or free 9-AC to mice, colon-specific release of 9-AC produced high local concentrations with the mean peak concentra-tion of 9-AC in cecal contents, feces, cecal tissue, and colon tissue being 3.2, 3.5, 2.2, and 1.6-fold higher, respectively. Therefore, the authors anticipated higher antitumor efficacy of the polymer conjugate due to prolonged colon tumor exposure to higher and more localized drug concentrations.

Combination strategies for colon-specific drug delivery commonly utilize a combination of pH and colonic microflora-based strategies. For example, Kaur and Kim prepared prednisolone beads with multiple coating layers for colonic delivery of the anti-inflammatory compound. The authors coated prednisolone on nonpareil beads followed by a hydrophobic coat of Eudragit® RL/RS; followed by a layer containing chitosan, suc-cinic acid, and Eudragit® RL/RS; followed by an outermost enteric coat layer (Figure 15.7a). In vitro experiments showed absence of drug release in simulated gastric and intestinal fluids, followed by drug release in the

Figure 15.7 Colonic drug targeting. Combination strategy for colonic drug targeting using an oral solid dosage form: (a) design of the targeted drug delivery system. Predniosolone (PDS, drug) was coated on nonpareil beads (1st layer), followed by a hydrophobic coat of Eudragit® RS/RL polymers (2nd layer), which was fol-lowed by a layer of Eudragit® RS/RL polymers in combination with chitosan and succinic acid (3rd layer), and the outermost enteric coating layer of Eudragit® L 100 (4th layer), (b) In vitro drug release from the system as a function of pH, succinic acid (SA) content in the formulation, and β-glucosidase content in the dissolution medium. The formulation dissolution was carried out in the gastric fluid for the first 2 h, followed by the small intestinal fluid for next 5 h, and the pathological colonic fluid for the last 7 h, and (c) plasma drug concentration after oral administration of powder, enteric coated, or colon targeted drug delivery systems in rats. (Modified from Kaur, K., Kim, K., Int. J. Pharm., 366, 140, 2009. With Permission.)

pathological colonic fluid with rate dependence on the presence of suc-cinic acid in the formulation and the presence of the enzyme β-glucosidase (Figure 15.7b). The authors proposed a combination mechanism of drug release that involved pH-triggered enteric dissolution of the outer-most layer, followed by chitosan and Eudragit® swelling in the presence of succinic acid, and biodegradation of chitosan by the colonic bacteria. Organic acid interacts with the amine groups in Eudragit® and chitosan polymers, leading to increased permeability of the coating and facilitated drug release at the colonic site. On oral administration of this formulation to male Sprague-Dawley rats, significant delay in the time to maximum plasma drug concentration (T_max) was obtained compared to both unmodi-fied powder and enteric-coated tablet formulations, thus indicating colonic targeting (Figure 15.7c).

Kidney-targeted drug delivery

Kidney-targeted drug delivery is quite promising to improve drug efficacy and safety in the treatment of renal diseases.1 Renal targeting is valuable to avoid extrarenal side effects of drugs used in the treatment of kidney diseases or to optimize the intrarenal distribution of a drug candidate, thus increasing its therapeutic index. Although renal drug delivery is not well studied, it highlights the challenges and opportunities inherent in develop-ing a targeted DDS. Among the drugs used for the treatment of kidney diseases are anti-inflammatory and antifibrotic compounds. Specific drug delivery to the kidney may also be helpful during shock, renal transplanta-tion, ureteral obstruction, diabetes, renal carcinoma, and other diseases such as Fanconi and Bartter’s syndrome. In addition, renal targeting can be helpful for drugs that would otherwise be rapidly metabolized and inac-tivated before reaching the kidney and to overcome or minimize the effects of pathological conditions, such as proteinuria, on drug distribution to the target site.

Cellular drug targets

Three cellular drug targets have been identified within the kidney— (1) proximal tubular cells, (2) mesangial cells, and (3) fibroblasts. Nephron, the functional unit of the kidney, consists of a renal corpuscle and a renal tubule. The renal corpuscle is responsible for blood filtration. It consists of the glomerulus and the Bowman’s capsule. The renal tubule consists of proximal and distal convoluted tubules interconnected by the loop of Henle. After blood filtration through the glomerulus, the proximal con-voluted tubule is responsible for pH regulation and reabsorption of salts and organic solutes from the filtrate. The luminal surface of the proximal tubular cells has a brush-border epithelium, with densely packed microvilli, which help increase the luminal surface area.

Mesangium, or the mesangial tissue, constitutes the inner layer of glom-erulus, within the basement membrane of the renal corpuscle. It surrounds the glomerular arteries and arterioles both within (intraglomerular) or outside (extraglomerular) the glomerulus. The glomerular epithelium is fenestrated and there is no basement membrane between the glomerular capillaries and the mesangial cells. Hence, mesangial cells are separated from the capillary lumen by only a layer of endothelial cells. Mesangial cells are phagocytic in nature and secrete an amorphous, basement mem-brane-like material, known as the mesangial matrix. These cells generate inflammatory cytokines and are involved in the uptake of macromolecules.

Fibroblasts synthesize ECM and collagen. Excessive production and accumulation of the ECM lead to fibrosis. Renal fibrosis is the underlying process that leads to the progression of chronic kidney disease to end-stage renal disease. It involves changes in the renal vasculature, glomeruloscle-rosis, and tubulointerstitial fibrosis. Of these, tubulointerstitial fibrosis is considered to be the most consistent predictor of an irreversible loss of renal function and progression to end-stage renal disease. The accumulation of ECM components in fibrotic disease is attributed to the activation of resi-dent interstitial fibroblasts. Therefore, targeted drug delivery to renal fibro-blasts has been attempted. For example, Kushibiki et al. used cationized gelatin to complex an enhanced green fluorescent protein (EGFP) express-ing plasmid, which was injected into the left kidney of mice through the ureter. The authors observed significant EGFP expression in the fibroblasts residing in the renal interstitial cortex. Similarly, Xia et al. reported the delivery of siRNA targeted against heat shock protein 47 (HSP47) using cationized gelatin microspheres to the mice kidneys with tubulointersti-tial fibrosis. The authors observed that the cationized gelatin microspheres enhanced and prolonged the antifibrotic effect of the siRNA.

Of these cell types, the proximal tubular cells have been the target of most drug-delivery strategies. They are metabolically the most active cells in the kidney and are involved with the transport and metabolism62 of several organic and inorganic substrates. Consequently, they have spe-cific transporter receptors on their luminal and basolateral membranes for substrate exchange between the blood and the urine. These transport and metabolic functions of the proximal tubular epithelial cells are utilized for drug targeting.

Particulate systems

The lack of basement membrane in the glomerular capillaries makes mesangial cells in close contact with the bloodstream, being separated from the capillary lumen by only a layer of endothelial cells. The mesangial cells, therefore, can be targeted using particulate carrier systems that may not filter through the glomeruli. Tuffin et al. used OX7-coupled immu-noliposomes to target renal mesangial cells. The authors coupled OX7 monoclonal antibody F(ab′)₂ fragments, directed against the mesangial cell expressing Thy1.1 antigen, on the surface of doxorubicin-loaded immu-noliposomes. The authors observed specific targeting to rat mesangial cells in vitro and in vivo on intravenous administration. Administration of doxorubicin-encapsulated immunoliposomes resulted in significant glo-merular damage, with low damage to other parts of the kidney and other organs. The targeted localization was not observed with free drug or lipo-somes, and immunoliposome localization was blocked by coadministration of free antibody fragments.

In a later study, the authors attempted to correlate the biodistribution of these immunoliposomes with the tissue distribution of the antigen. The Thy1.1 antigen showed high expression in rat glomeruli, brain cortex and striatum, and thymus; and moderate expression in the collecting ducts of the kidney, lung, and spleen. The biodistribution of immunoliposomes did not correlate well with the tissue distribution of Thy1.1 antigen, with the highest levels seen in the spleen, followed by lungs, liver, and kidney. Within the kidney, specific localized delivery to the mesangial cells was observed, which was sensitive to competition with the unbound OX7 monoclonal antibody fragments. The authors concluded that the absence of endothelial barriers and high target antigen density are important factors governing tissue localization of immunoliposomes.

An application of drug targeting to glomerular endothelial cells to reduce systemic side effects of drug therapy was reported by Asgeirsdottir et al., who used immunoliposomes to target glomerular endothelial cells in mice. Glomerulonephritis, a spectrum of inflammatory diseases specifically affecting renal glomeruli, is characterized by the activa-tion of proinflammatory pathways, resulting in glomerular injury and proteinuria. These disorders are frequently treated with glucocorticoids, such as dexamethasone, in combination with cytotoxic agents, such as cyclophosphamide, as anti-inflammatory and immunosuppressive agents. These drugs, however, present serious extrarenal side effects including an increase in blood glucose levels with dexamethasone. Asgeirsdottir et al. coupled monoclonal rat anti-mouse E-selectin antibody, MES-1, to the surface of liposomes. The selection of this antibody was designed to tar-get glomerular endothelial cells in glomerulonephritis, wherein endothelial cell expression of inflammation-related cell-adhesion molecules, such as E-selectin and VCAM-1, is upregulated. The authors obtained site-specific delivery of immunoliposome encapsulated anti-inflammatory agent dexa-methasone and observed reduction in glomerular proinflammatory gene expression with no effect on blood glucose levels.

In addition to liposomes, nanoparticles have been utilized for drug targeting to the mesangial cells. For example, Manil et al. used isobutyl-cyanoacrylate nanoparticles for targeting the antibiotic actinomycin D to rat mesangial cells. Compared to the free drug, the uptake of drug-loaded nanoparticles in the whole kidneys was over two fold at both 30 and 120 min after intravenous injection. Similar or higher uptake ratios were obtained for isolated rat glomeruli, but not for tubules. The glomerular uptake of nanoparticles was even higher in rats with experimental glomerulonephritis. Mesangial cell targeting was indicated by in vitro experiments, which dem-onstrated fivefold higher uptake by mesangial cells than the epithelial cells. In a separate study, Guzman et al. also obtained about twofold higher in vitro uptake of drug-loaded nanoparticles in rat mesangial cells using polycapro-lactone as the polymeric carrier and digitoxin as the drug candidate.

The prodrug approach

Prodrugs are drug conjugates designed to modify the physicochemical and/or biopharmaceutical properties of the drug candidate. Their derivatization is bioreversible and is designed to improve drug properties with respect to solubility, stability, permeability, presystemic metabolism, and targeting. Prodrugs retain the advantages of low molecular weight compounds such as low immunogenicity and feasibility of oral administration. Renal speci-ficity of prodrugs would depend on the renal-specific metabolism and/or uptake of the promoiety. For this purpose, amino acid prodrugs, which can be activated by kidney-specific enzymes, have been evaluated for renal targeting.

Amino acid prodrugs have advantages of biodegradability in addition to receptor-mediated uptake, which can help in both oral absorption and organ or tissue-specific targeting. For example, valine prodrugs of acyclo-vir and ganciclovir showed 3–5 times higher bioavailability than the par-ent compounds. Enhanced oral absorption of amino acid prodrugs is attributed to carrier-mediated intestinal uptake via transporters. For organ and tissue-specific drug targeting, the l-glutamate transport system has been commonly utilized.

Prodrug design for renal targeting is aimed at utilizing the kidney-specific enzymes. The proximal tubular cells contain high levels of metabolizing enzymes in the cytosol (such as l-amino acid decarboxylase, β-lyase, and N-acetyl transferase) and at the brush border (such as γ-glutamyl transpep-tidase). Examples of renal-targeted prodrugs include the γ-glutamyl pro-drugs of l-dopa and sulfamethoxazole.

Gludopa (γ-l-glutamyl-l-dopa) is a kidney-specific dopamine prodrug. Cummings et al. reported its pharmacokinetic and tissue distribution in rats. Gludopa was metabolized primarily in the liver and kidney, with dopamine being the major kidney metabolite. The pharmacokinetics of gludopa in healthy human volunteers indicated urinary dopamine excre-tion in parallel with urinary levodopa excretion, supporting the view that levodopa was the precursor of urinary dopamine. Based on these results, Boateng et al. indicated that gludopa may be useful in conditions where renal effects of dopamine are indicated. However, Lee noted the limitations in clinical practice posed by its low oral bioavailability in humans.

Kidney-specific delivery of parent compounds after IV administration of γ-l-glutamyl (G) and N-acetyl-γ-l-glutamyl (AG) prodrugs of p-nitroaniline, sulfamethoxazole, and sulphamethizole was investigated by Murakami et al. in rats. The authors observed higher plasma stability of AG over G prodrugs for all compounds. The concentration of parent compounds was higher in the kidney than the pulmonary and hepatic tissue for all compounds, with markedly increased kidney distribution of AG prodrugs of p-nitroaniline and sulfamethoxazole. The activation of AG prodrugs requires the action of two enzymes—deacylation by N-acylamino acid deacylase and hydrolysis by γ-glutamyl transpeptidase, whereas G prodrugs can be activated by the action of γ-glutamyl transpeptidase alone. When biodistribution of G pro-drugs of sulfamethoxazole was studied in mice, relatively high concentra-tions of sulfamethoxazole were found in nonrenal tissues as well indicating rapid kinetics of enzymatic cleavage of G prodrugs even in tissues with low γ-glutamyl transpeptidase activity. However, kidney selective accumula-tion was obtained after the administration of AG prodrugs. Drieman et al. hypothesized that the renal selectivity of AG prodrugs of sulfamethoxazole was due to a carrier-mediated transport followed by intracellular conversion of the prodrug to the active compound.

Effective utilization of the prodrug strategy requires intensive investiga-tion of the role of variables such as the linker groups and the promoiety modifications. This results in an inherent complexity in prodrug design and utilization for organ or tissue-targeted drug delivery.

Bioconjugation approaches

Bioconjugation of a drug to a carrier that is significantly larger than the molecular size of the drug allows the biopharmaceutical properties of car-rier to dominate the absorption and biodistribution of the conjugate. In the case of renal drug targeting to the proximal tubular cells, the conjugates would need to be filtered through the glomerular capillaries and reabsorbed by the tubular cells. Particles with a hydrodynamic diameter below 5–7 μm are rapidly filtered through the glomerulus.

For this purpose, the carriers that naturally accumulate in the prox-imal tubular cells can be used as drug carriers. These include the low (less than about 30 kDa) molecular weight proteins (LMWPs), such as lysozyme, aprotinin, and cytochrome C. They are readily filtered through the glomerulus but selectively reabsorbed by the proximal tubular cells (Figure 15.8a). Thus, LMWP–drug conjugates that are stable in the plasma but cleaved within the proximal tubular cells after endosomal/lyso-somal uptake can be used as effective vehicles for drug targeting. Drugs may be conjugated to LMWPs directly using the lysine amino groups or with a spacer. Relatively large size of the LMWPs allows the pharmaco-kinetic properties of the LMWPs to override those of the drug candidate in the LMWP–drug conjugates. This approach, however, is limited by the requirement of parenteral administration and potential immunogenicity of the conjugates.

The internalization of proteins in the proximal convoluted tubule epithe-lium cells is mediated via the multiligand megalin and cubilin receptors. These cells have very high endocytic activity. After endocytosis, the protein is degraded in the lysosomes, wherein the attached drug may be released (Figure 15.8b). Lysozyme has been used as a renal carrier for the non-steroidal anti-inflammatory drug (NSAID) naproxen (Figure 15.8c), the acetylcholinesterase (ACE) inhibitor compound captoril, and the nephroprotective compound triptolide (Figure 15.8d).

Naproxen is a carboxylic acid group-bearing compound that could be conjugated to the amine group in the lysozyme directly by an amide bond or through lactic acid spacer by an ester bond. The biodistribution and degradation of these conjugates was compared with lysozyme and naproxen by themselves in rats. Drug conjugation did not affect the renal uptake or degradation of lysozyme in the rat kidney. The pharmacokinetic profile of the conjugates was similar to that of lysozyme, but markedly different from the drug. The drug was rapidly taken up by and degraded in the kid-ney with no detectable levels in the plasma (Figure 15.8c). Similar results were obtained when captopril was conjugated with lysozyme through a spacer utilizing disulfide linkage. Targeting this ACE inhibitor to the kid-ney was hypothesized to prevent attenuation of renoprotective (antiprotein-uric) efficacy of captopril under high sodium concentrations. The drug was efficiently targeted to the kidney with the rapid release of the drug.

Triptolide is an immunosuppressive and anti-inflammatory natural compound with low water solubility and significant toxicity. Renal target-ing of triptolide–lysozyme conjugate linked through succinyl residue was investigated in rats. The authors obtained significantly higher targeting efficiency of the drug conjugate to the kidney with reversal of disease pro-gression in renal ischemia-reperfusion injury rat model, lower hepatotox-icity, and no effect on immune and genital systems, compared to the free drug (Figure 15.8d). These results demonstrated the potential therapeutic benefits of renal drug targeting.

In addition to the use of LMWPs as drug targeting ligands, their receptor-mediated uptake can also be utilized to mitigate renal toxicity of drugs. For example, endocytosis by proximal tubular cells is responsible for the renal accumulation and toxicity of aminoglycoside antibiotics, such as gentami-cin, which is a substrate of the megalin receptors. Watanabe et al. found that coadministration of cytochrome C competes with receptor-mediated renal uptake of gentamicin, thus reducing its renal accumulation in rats. However, the required dose of cytochrome C was quite high; the authors tested the relative efficacy of peptide fragments in reducing the renal accu-mulation of gentamicin. Three peptide fragments derived from actin-regulating proteins were identified that reduced the renal accumulation of gentamicin without affecting its plasma concentration-time profile.

In addition to the exploitation of LMWPs for modulating the pharma-cokinetics and biodistribution of drugs by utilizing their physiological disposition to modify drug biodistribution, drugs and enzymes can also be targeted to the renal proximal tubular epithelial cells by their surface modification. For example, Inoue et al. modified the enzyme superoxide dismutase (SOD), which disproportionates the superoxide free radical into oxygen and hydrogen peroxide—thus reducing free radical and oxidative stress in the cells. Intravenously administered Cu, Zn–SOD is rapidly removed from the circulation with a half-life of about 5 min and appears intact in the urine, thus indicating that it is filtered through the glomeru-lus. The authors conjugated hexamethylene diamine (AH) to SOD. The conjugate (AH–SOD) was rapidly filtered through the glomeruli but bound apical plasma membranes of proximal tubular cells followed by localized action in these cells. The authors observed more than 80% of the radio-activity derived from AH–SOD localized in the kidney at 30 min after injection, most of which was localized in the proximal tubular cells. In vitro kinetic studies revealed that the specific binding of AH–SOD to apical surface of the tubular cells was attributable to AH.

Polymeric carriers have also been described for renal drug targeting. These include the anionized derivatives of polyvinylpyrrolidone (PVP), low molecular weight N-(hydroxypropyl) methylacrylamide (HPMA), and low molecular weight chitosan. The use of synthetic polymers requires surface modification and derivatizing groups for optimum renal accumula-tion. For example, PVP by itself does not accumulate in the tubular epithe-lial cells, but on copolymerization with maleic acid, it selectively distributed into the kidneys on IV injection in mice. When anionized derivatives of PVP were prepared, the plasma clearance of these derivatives decreased with increasing size of anionic groups. In addition, even though the clear-ance of carboxylated PVP and sulfonated PVP from the blood was similar, renal accumulation of carboxylated PVP was several fold higher than that of sulfonated PVP. In summary, these studies demonstrate not only the potential for renal targeting of drugs where it may be beneficial but also the potential to prevent accumulation in the kidney for drugs that have renal toxicity.