OCULAR DELIVERY OF PEPTIDE AND PROTEIN DRUGS

IV. OCULAR DELIVERY OF PEPTIDE AND PROTEIN DRUGS

Peptides and proteins may be instilled into the eye for local/topical use. Instillation of a topical dose of a drug to the eye leads to absorption of a drug mainly through the conjunctival and corneal epithelia. For drugs meant for topical use, it must be minimally absorbed systemically as it can lead to undesirable side effects. Absorption into the systemic circulation may occur across the conjunctiva and sclera. However, for local delivery the cornea presents a significant barrier to the introcular penetration of peptide drugs in view of their high molecular weight and low lipophilicity. Lee et al. (75) reported that the penetration of inulin through the rabbit cornea was probably occurring via a paracellular route rather than a transcellular route.

Systemic absorption of peptide and protein drugs following topical administration to the eye could occur through contact with the conjunctival and nasal mucosae, the latter occurring as a result of drainage through the nasolacrimal duct. When systemic effects are desired, absorption through the conjunctival and nasal mucosae needs to be maximized. One also must consider other competing processes present in the ocular tissues. Of these processes, absorption by the avascular cornea is important, since a large portion of the drug thus absorbed is distributed to adjacent ocular tissues.

Ahmed and Patton (80) found that noncorneal (scleral) absorption accounted for about 80% absorption of inulin, a highly hydrophilic macro- molecule, into the iris-ciliary body. This observation is important, since most therapeutic peptides act locally in the iris-ciliary body, which is con-

Peptides and Proteins as Therapeutic Agents 503 tiguous with the sclera. Therefore, macromolecular drug absorption would

benefit from scleral absorption. Beside the transport barrier, another factor severely limiting the ocular absorption of peptide drugs is metabolism by ocular enzymes, specifically peptidases. Endopeptidases, like plasmin and collagenase, and exopepti- dases, like aminopeptidases, are present in the ocular fluids and tissues. The endopeptidase levels are usually low unless the eye is inflamed (81,82) or injured (83) and are of little concern relative to the stability of topically applied doses. Lee et al. (74) reported that within 5 minutes postinstillation, about 90% of leucine enkephalin and almost 100% of methionine enkepha- lin (pentapeptides) was recovered in the rabbit corneal epithelium in a hydrolyzed form. Therefore, aminopeptidase activity must be inhibited to facilitate ocular peptide absorption. Controlling these enzymes in the target tissues may not be practical given the fact that the same enzymes might be necessary for the homeostasis in the eye.

Cyclosporin A has been shown to improve the prognosis for corneal allograft rejection. It was found that when administered by nonocular routes in rabbits, it was detected in the systemic circulation but not in the ocular tissues (20,84,85). Also, topical administration of cyclosporin A did not produce any significant penetration within the eye beyond the cornea or the conjunctiva. This may be because cyclosporin A was bound to corneal and conjunctival epithelial cell membranes. Cyclosporin A eyedrops formu- lated in absolute ethanol did produce higher levels in intraocular tissues, which may be due to damage to corneal epithelium by alcohol.

Growth factors, especially epidermal growth factor (EGF), have been found to stimulate cell proliferation in the corneal epithelium, thus stimu- lating epithelialization during wound healing. Growth factors are mostly used in accelerating the wound-healing process, and it would be of great importance in corneal wounds since the cornea is an avascular organ. Many in vitro corneal preparations have been used to demonstrate the wound- healing process. Human EGF promotes endothelial wound healing (84). Many other growth factors also play a major role in corneal wound healing,

growth factor (PDGF). Basic fibroblast growth factor (bFGF) and insulin- like growth factor I (IGF-I) have been found in higher levels in patients suffering from diabetic retinopathy (88–90). IGF-I and bFGF can also induce fibrovascular changes in the retinal vessels.

A more practical strategy for circumventing the enzymatic barrier would be to administer peptide analogs that are resistant to the principal peptidases but possess equivalent biological activity [D-Ala 2 ]methionine enkephalinamide (DAMEA), which resists aminopeptidase-mediated clea- vage, falls in this category of peptide analogs (74). The permeation and

504 Dey et al. metabolic degradation of DAMEA in the albino rabbit cornea, conjunctiva,

and sclera has been studied (91). DAMEA was administered with and with- out peptidase inhibitors bestatin (aminopeptidase inhibitor) and SCH 39370 (enkephalinase inhibitor). It was found that sclera was the most permeable membrane to DAMEA, while cornea was almost impermeable to DAMEA.

10 cm/s across the cornea, con- junctiva, and sclera, respectively. When inhibitors were co-administered with DAMEA, the corneal permeability of intact DAMEA increased 15 times, conjunctival permeability increased 5.5 times, while scleral permeability remained practically unaltered.

The corneal and conjunctival penetration of 4-phenylazobenzyloxy- carbonyl-l-Pro-l-Leu-Gly-l-Pro-d-Arg (Pz-peptide) and its effect on the corneal and conjunctival penetration of hydrophilic solutes as well as on the ocular and systemic absorption of topically applied atenolol and pro- pranolol in the rabbit have been evaluated (92). The conjunctiva was 29 times more permeable than the cornea to 3 mM Pz-peptide. Conjunctival Pz-peptide transport was 1.7 times greater in the mucosal-to-serosal than in the opposite direction, whereas corneal Pz-peptide transport showed no directionality. The apparent permeability coefficients of Pz-peptide across the cornea and the conjunctiva increased over the 1–5 mM range, which suggests that Pz-peptide enhanced its own transport across both epithelial tissues. The cornea was more sensitive than the conjunctiva to the pene- tration-enhancement effect of Pz-peptide. Pz-peptide elevated the corneal transport of mannitol, fluorescein, and FD4 by 50, 57, and 106%, respec- tively, but it did not affect the conjunctival transport of mannitol and fluorescein. While Pz-peptide enhanced the ocular absorption of topically applied hydrophilic atenolol, it did not affect the ocular absorption of lipophilic propranolol. Interestingly, Pz-peptide did not affect the systemic

itate its own penetration across the cornea and the conjunctiva and increase the ocular absorption of topically applied hydrophilic but not lipophilic drugs, while not affecting the systemic absorption of either type of drug.

In addition, the presence of sites beyond the absorbing epithelia that are capable of degrading peptides and protein and the availability of multi- ple peptidases in a given site further decrease the absorption potential of such compounds. While the ocular route has been widely accepted for the use of topical application, its use in systemic delivery of peptides and pro- teins will be rather limited.

Peptides and Proteins as Therapeutic Agents 505