Probenecid Treatment Enhances Retinal and Brain Delivery of N-4-Benzoylaminophenylsulfonylglycine, An Anionic Aldose Reductase Inhibitor

Abstract

Anion efflux transporters are expected to minimize target tissue delivery of N-[4-(benzoylaminophenyl)sulfonyl]glycine (BAPSG), a novel carboxylic acid aldose reductase inhibitor, which exists as a monocarboxylate anion at physiological conditions. Therefore, the objective of this study was to determine whether BAPSG delivery to various eye tissues including the retina and the brain can be enhanced by probenecid, a competitive inhibitor of anion transporters. To determine the influence of probenecid on eye and brain distribution of BAPSG, probenecid was administered intraperitoneally (120 mg/kg body weight; i.p.) 20 minutes prior to BAPSG (50 mg/kg; i.p.) administration. Drug disposition in various eye tissues including the retina and the brain was determined at 15 min, 1, 2 and 4 hr after BAPSG dose in male Sprauge-Dawley rats. To determine whether probenecid alters plasma clearance of BAPSG, influence of probenecid (120 mg/kg; i.p.) on the plasma pharmacokinetics of intravenously administered BAPSG (15 mg/kg) was studied as well. Finally, the effect of probenecid co-administration on the ocular tissue distribution of BAPSG was assessed in rabbits following topical (eye drop) administration. Following pretreatment with probenecid in the rat study, retinal delivery at 1 hr was increased by about 11 fold (2580 vs 244 ng/gm; p<0.05). Further, following probenecid pretreatment, significant BAPSG levels were detectable in the brain (45 ± 20 ng/gm) at 1 hr, unlike controls where the drug was not detectable. Plasma concentrations, plasma elimination half-life, and total body clearance of intravenously administered BAPSG were not altered by i.p. probenecid pretreatment. In the topical dosing study, a significant decline in BAPSG delivery was observed in the iris-ciliary body but no significant changes were observed in other tissues of the anterior segment of the eye including tears. Thus, inhibition of anion transporters is a useful approach to elevate retinal and brain delivery of BAPSG.

1. Introduction

Similar to brain drug delivery, drug delivery to the retina is a challenge, due to the presence of biological barriers (1). Most compounds penetrate the retina of the eye poorly after systemic administration (2). Several approaches such as intravitreal (3) and periocular (4) injections are being investigated for achieving therapeutically effective levels of drugs in the treatment of retinal or vitreous disorders. However, for treating chronic disorders, these approaches are less convenient than noninvasive systemic delivery approaches such as oral dosing. Therefore, understanding the underlying mechanisms that are responsible for the poor availability of the drugs following systemic administration would enhance the development of strategies for improving the non-invasive delivery of such molecules. Following systemic administration, cornea, lens and retinal tissue levels of a drug depend on its ability to cross blood-aqueous barrier (BAB) and blood-retinal barrier (BRB), respectively. However, the BRB determines the retinal drug disposition of the systemically administered drugs. The BAB is more permeable than the BRB (5, 6) – this is especially true for small molecular weight drugs. The BRB is located at two levels. The outer BRB, located in the retinal pigment epithelium (RPE) and the inner BRB, is found in the endothelial membranes of the blood vessels of the retina. The BRB plays a critical role in the homeostatis of the neural retina by limiting the entry of xenobiotics into the extravascular spaces of the retina and by preventing the loss of essential solutes. Barza et al. (7) determined the kinetics of intravitreally injected carbenicillin, an organic anion antimicrobial, in rabbits following concomitant intraperitoneal administration of probenecid, an organic anion transport inhibitor. Probenecid increased the vitreous half-life of carbenicillin from 5 to 13 hr. Similar observations were made in Rhesus monkeys, wherein, probenecid increased the vitreous half-life of carbenicllin and cefazolin from 10 to 20 hr and from 7 to 30 hr, respectively (8). These findings suggest that a probenecid-sensitive active transport system is present in the blood-retinal barrier to actively remove organic acids such as penicillins and cephalosporins.

An anionic aldose reductase inhibitor (ARI) suitable for treating diabetic retinopathy and neuropathies is a drug of our interest. Among a series of ARIs, we identified that N-4-[(benzoylaminophenyl)sulfonyl]glycine (BAPSG) has high aldose reductase inhibitory activity and permeability across ocular barriers such as cornea and conjunctiva (9) (10). Such aldose reductase inhibitory activity is of value in treating diabetic complications. Furthermore, it was demonstrated that BAPSG has ARI activity in cultured retinal pigment epithelial (ARPE-19) cells and in vivo (11). BAPSG is likely to be safe, due to the lack of hydantoin moiety in its structure and due to its high selectivity for aldose reductase compared to other similar oxido-reductase families of enzymes (12). BAPSG is a low molecular weight (MW = 334.35) drug that is anionic at physiological conditions (pKa = 3.35; log P = 1.09).

In our laboratory, an anionic efflux transporter, multidrug-resistance associated protein (MRP1), was identified and characterized in the human retinal pigment epithelial cells (11). Further, it was also shown that the delivery of BAPSG, can be enhanced by the presence of a putative MRP1 inhibitor, probenecid.

Since anionic solute efflux systems are present in the blood-retinal barrier to remove drugs from inside of the eye to the blood and because BAPSG is a substrate for such transporters, the purpose of this study was to determine whether pretreatment with probenecid enhances BAPSG delivery to the retina. Systemic pretreatment with probenecid and systemic administration of BAPSG as opposed to intravitreal administration was assessed in a rat model since our long-term goal is to establish a systemic approach for retinal delivery of BAPSG. Since some of the barrier functions are similar between retina and the brain, we assessed BAPSG delivery to the brain as well. Additionally, to determine the value of inhibiting anion efflux transporters in enhancing BAPSG delivery from an eye drop to the tissues of the anterior segment of the eye, we investigated the influence of probenecid co-administration in an eye drop on BAPSG delivery in tears, cornea, aqueous humor, lens, and iris-ciliary body.

2. Materials and methods

2.1. Animal studies

All animal studies were conducted during the authors’ tenure at the University of Nebraska Medical Center after approval by the IACUC and according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.

2.2. Influence of probenecid on BAPSG delivery: Systemic administration

The influence of intraperitoneal (i.p.) pretreatment with probenecid, an inhibitor of organic anion efflux transporters (13, 14), on the eye and brain tissue distribution of BAPSG was examined in male Sprauge Dawley rats (SASCO, Wilmington, MA) weighing about 200 gm. Further, the influence of probenecid pretreatment on the plasma pharmacokinetics of intravenously administered BAPSG was also assessed.

BAPSG formulations were prepared just before dosing the animals. For intraperitoneal pharmacokinetic studies, an appropriate amount of BAPSG was placed in phosphate buffer saline (pH, 7.4) and sonicated for 5 minutes to obtain a suspension. The dosing volumes for these studies ranged from 500 µl to 1000 µl. Probenecid was formulated as a solution (50 mg/ml) in 8.6% sodium carbonate solution for injection. For intravenous pharmacokinetic studies, BAPSG solutions were made in saline containing 10% DMSO and 10% propylene glycol, and the injection volume did not exceed 100 µl.

Twenty minutes following intraperitoneal administration of probenecid (120 mg/kg), BAPSG (50 mg/kg) was administered intraperitoneally. At various times (15 min, 1, 2 and 4 hr) thereafter, 200 µl of the blood was collected via the tail vein followed by euthanasia of each rat with a lethal dose (150 mg/kg) of sodium pentobarbital (Sleepaway, Fort Dodge, IA). The eye and brain tissues were removed at the above time points, rinsed with normal saline, and rapidly blotted. From the eye, cornea, lens, and retina were carefully isolated. The tissues were stored at −70 °C until analysis. All samples were analyzed using a HPLC assay.

2.3. Influence of probenecid on the plasma pharmacokinetics of BAPSG in rats

To determine whether the effect of probenecid on the tissue distribution of BAPSG is due to its effect at the blood-ocular barriers or due to its effect on the systemic clearance of the drug, the effect of probenecid treatment on the plasma pharmacokinetics of BAPSG was studied. In this study, BAPSG was administered at a dose of 15 mg/kg via a tail vein injection, either alone or following treatment with probenecid (125 mg/kg; i.p.) at 20 min a priori. Serial blood samples (200 µl) were collected and the plasma was isolated for drug analysis.

2.4. Influence of probenecid on BAPSG delivery: Topical eye drop administration

Following a single topical dose, in vivo uptake of BAPSG in rabbits was determined in the presence of probenecid,. The eye drop formulation was prepared as follows- twenty-five microliters of 11.4 mg of probenecid in 1 ml of ethanol was added to 1 ml of BAPSG suspension in phosphate buffer saline (pH 7.4), to obtain a final concentration of 1 mM probenecid (0.28 µg/ml) and 5 mg/ml of BAPSG. Thirty microliters of the formulation was instilled into the conjunctival cul-de-sac of each eye. At 30, 60, 120 and 180 min post-dosing, the rabbits were sacrificed with a marginal ear vein injection of 150 mg/kg sodium pentobarbital (Sleepaway, Fort Dodge, IA). The ocular tissues were collected and processed as described above.

Probenecid treatments performed as described above in rat and rabbit models were compared with plain drug treatments at the same drug doses reported by our group in a previous study (15).

2.5. Drug analysis

Plasma and tissue samples were analyzed using a HPLC assay as per the methods described previously (15). The lower limit of quantification (LOQ) of the assay was 25 ng/ml and the limit of detection (LOD) was 1 ng. Linear standard graphs of BAPSG were obtained in the concentration range of 25 ng/ml to 100 µg/ml, and the relative standard deviation between the assays was less than 7%. Concentrations below LOQ were reported as non-detectable and considered as ‘zero’ for the pharmacokinetic data analysis.

2.6. Statistical methods

The study data are expressed as arithmetic mean ± s.d. (n=3 to 6). The means between the groups were compared using analysis of variance and specific comparisons were made using Tukey’s pos-hoc analysis (SPSS, version 11.5). Differences were considered statistically significant at P<0.05.

3. Results

3.1. Influence of probenecid pretreatment (i.p.) on the ocular disposition of BAPSG following systemic (i.p.) administration

Retina, cornea, lens and brain exposure of BAPSG, expressed as AUC_0–4hr based on drug amounts per gram (gm) tissue, was in the order: cornea > retina ≫ lens ≫ brain in both probenecid pretreated and untreated control groups. In control animals, peak BAPSG levels were observed at 1 hr after dosing in all the ocular tissues except in brain (15), where BAPSG was undetectable beyond 15 min. Following pre-treatment with probenecid, retinal delivery of BAPSG at 1 hr increased by about 11-fold (2.58 vs. 0.24 µg/gm; p<0.05) (Figure 1). The AUC_0–4hr (µg-hr/gm) estimated for the retina was 2.4-fold higher in the presence of probenecid (4.11 ± 0.77 vs. 1.7 ± 0.7; p<0.05). Also, the retina to plasma AUC_0–4hr ratio of BAPSG was increased from 0.08 to 0.26 in the presence of probenecid. However, there was no significant increase in the BAPSG levels at T_max (1 hr) or AUC_0–4hr in the lens and cornea following probenecid pretreatment (Table 1). Interestingly, BAPSG levels of 0.045 ± 0.02 µg/gm were detectable in brain at 1 hr following probenecid pretreatment, while being undetectable in plain BAPSG group. BAPSG was not detectable in brain beyond 1 hr in probenecid treatment group as well. BAPSG concentrations in brain at 15 min in the absence and presence of probenecid were 0.37 ± 0.28 µg/gm and 0.21 ± 0.23 µg/gm, respectively.

Figure 1.

Tissue concentrations of BAPSG at 1-hr following intraperitoneal administration (50 mg/kg) with or without probenecid treatment in a rat model. Probenecid (125 mg/kg; i.p.) was administered 20 min prior to the administration of BAPSG. Data is expressed as mean ± s.d. for n = 3 rats. *p<0.05 vs. retinal levels of BAPSG in the ‘BAPSG alone’ group. BAPSG alone data was from Sunkara et al. (15).

Table 1.

Influence of probenecid (125 mg/kg; i.p.) on the ocular tissue levels (at 1-hr) and tissue delivery (AUC_0–3hr) of BAPSG (50 mg/kg; i.p.) in a rat model. Data is expressed as mean ± s.d. for n = 3 rats.

	BAPSG Concentration (µg/gm) at Tmax (1 hr)		AUC0–4hr (µg-hr/gm)
Tissue	BAPSG alone	BAPSG + Probenecid	BAPSG alone	BAPSG + Probenecid
Retina	0.24 ± 0.05	2.5 ± 0.82*	1.73 ± 0.7	4.11 ± 0.77#
Lens	0.025 ± 0.019	0.06 ± 0.04	0.55 ± 0.45	0.20 ± 0.17
Cornea	1.94 ± 0.88	3.45 ± 0.93	5.05 ± 2.34	9.18 ± 5.42
Brain	ND	0.042 ± 0.02	an.d.	bn.d.

ND = Non-detectable

^an.d. = Could not be determined since BAPSG was not detectable after 15 min

^bn.d. = Could not be determined since BAPSG was not detectable at 2 and 4 hr

^*p<0.05 vs. BAPSG alone (BAPSG concentration)

^#p<0.05 vs. BAPSG alone (AUC_0–4hr)

In this study we observed that intraperitoneal administration of BAPSG after pretreatment with probenecid results in plasma BAPSG exposure (AUC_0–4hr) of 18.2 µg-hr/ml compared to 21.7 µg-hr/ml reported previously in the absence of probenecid pretreatment (15). Thus, following intraperitoneal administration, similar BAPSG exposure was evident in the plasma with or without probenecid pretreatment. Further, lack of influence of probenecid on BAPSG plasma clearance was assessed following intravenous administration of BAPSG.

3.2. Effect of probenecid pretreatment (i.p.) on plasma pharmacokinetics of intravenously administered BAPSG

Figure 2 and Table 2 show the effect of probenecid on the plasma concentration profiles and the pharmacokinetic parameters of BAPSG, respectively. Probenecid did not alter the plasma profiles and various pharmacokinetic parameters including AUC, clearance, mean residence time, and volume of distribution of BAPSG (p<0.05).

Figure 2.

Plasma concentration profiles of BAPSG following intravenous (15 mg/kg) administration with or without probenecid treatment in a rat model. Probenecid (125 mg/kg; i.p.) was administered 20 min prior to the administration of BAPSG. Data is expressed as mean ± s.d. for n = 4 – 6 rats. BAPSG alone data was from Sunkara et al. (15).

Table 2.

Plasma pharmacokinetic parameters of BAPSG (15 mg/kg) following intravenous administration with or without probenecid treatment in a rat model. Probenecid (125 mg/kg) was administered intraperitoneally 20 min prior to the administration of BAPSG. Data is presented as mean ± s.d. for n = 4 – 6 rats.

Parameter	BAPSG Alone	BAPSG + Probenecid
AUC0-∞ (µg-hr/ml)	9.22 ±2.24	10.51 ± 4.7
CLT (L/hr)	0.26 ± 0.07	0.25 ± 0.12
MRT (hr)	0.65 ± 0.23	0.88 ± 0.26
Vd/F (L)	0.25 ± 0.06	0.23 ± 0.12
Co (µg/ml)	14.54 ± 9.0	16.90 ± 14.2

Key: C_o – Concentration at zero-time

AUC_0→∞ – Area under the curve from zero to infinity

V_d – Volume of distribution

Cl_T – Total body clearance (Dose/ AUC_0→∞)

MRT – Mean residence time (AUMC_0→∞/ AUC_0→∞)

3.3. Effect of probenecid coadministration with BAPSG in an eye drop on the disposition of BAPSG in the tissues of the anterior segment of the eye

The AUC_0–3hr and C_max of BAPSG in cornea, conjunctiva, and aqueous humor did not significantly increase following co-administration of BAPSG with probenecid (Table 3 and Figure 3). The tear BAPSG versus time profiles were also not significantly different between probenecid and control groups. In iris-ciliary body, however, about 6-fold decrease in AUC and 3-fold decrease in C_max were observed in the presence of probenecid.

Table 3.

Pharmacokinetic parameters of BAPSG (150 µg/eye) following topical administration with or without probenecid co-administration in a rabbit model. Probenecid (1 mM) was formulated and administered along with BAPSG. Data is presented as mean ± s.d. for n = 4 eyes^a.

	Cornea		Conjunctiva		Aqueous humor		Iris-ciliary Body
Parameter	BAPSG Alone	BAPSG + Probenecid	BAPSG Alone	BAPSG + Probenecid	BAPSG Alone	BAPSG + Probenecid	BAPSG Alone	BAPSG + Probenecid
AUC0–3hr(µg-min/gm)	399.5 ± 65.7	516.7 ± 96.34	75.38 ± 9.07	78.89 ± 23.45	5.56 ± 0.99	9.54 ± 4.0	12.06 ± 4.96	2.13* ± 0.7
Cmax (µg/gm)	2.2 ± 0.48	2.7 ± 0.36	0.95 ± 0.48	0.49 ± 0.14	0.027 ± 0.003	0.027 ± 0.05	0.09 ± 0.03	0.03* ± 0.016
Tmax (min)	57.9 ± 17.75	68.9 ± 14.61	27.24 ± 26.5	59.13 ± 19.0	73.77 ± 14.08	68.3 ± 45.8	n.d.	n.d.

Key: T_max – Time to reach maximum concentration (C_max)

AUC _0–3hr – Area under the curve from zero to 3 hr

n.d. – Could not be determined

^aTear parameters are not presented as 0-hr concentration could not be estimated; profiles are shown in Figure 5

^*p<0.05 vs. BAPSG alone (Iris-ciliary body)

Figure 3.

Tear concentration of BAPSG versus time profiles following topical administration of BAPSG (150 µg/eye) in rabbits with or without 1 mM probenecid. Data is expressed as mean ± s.d. for n = 4 eyes. Key: (◦) without probenecid; (•) with probenecid. BAPSG alone data was from Sunkara et al. (15).

4. Discussion

In this study, we report for the first time that systemic retinal delivery of an anionic aldose reductase inhibitor can be elevated by pretreatment with a competitive inhibitor of anion transporters. Our findings can potentially be extended towards the development of systemic approaches that are non-invasive to the globe of the eye in enhancing retinal drug delivery. In the current study, we specifically demonstrated that systemic pretreatment with probenecid, a competitive inhibitor of anion transporters, can increase the retinal and brain delivery of an anionic aldose reductase inhibitor, BAPSG, when administered via systemic route in a rat model. Probenecid did not alter plasma pharmacokinetics including clearance of BAPSG. In rabbit eye drop studies, also performed as a part of this study, a similar advantage was not evident in enhancing BAPSG delivery to the rabbit anterior segment eye tissues following co-administration of the drug with probenecid.

With intraperitoneal probenecid pretreatment followed by intraperitoneal injection of BAPSG in a rat model, we observed about 11-fold increase in retinal concentration at 1 hour and a 2.4-fold increase in retinal AUC, suggesting the value of probenecid in enhancing retinal BAPSG delivery. Further, while drug levels were below detection limits in the brain at 1 hour in the absence of probenecid, probenecid pretreatment resulted in significant drug delivery at 1 hour after BAPSG dosing. Thus, the delivery of anionic BAPSG to the retina as well as the brain was elevated by probenecid. One explanation for such enhanced drug delivery is a reduction in the plasma clearance of the drug due to inhibition of anion transporters in clearance organs such as the kidney. However, our plasma pharmacokinetic studies indicated that intraperitoneal pretreatment with probenecid does not significantly elevate plasma AUC of BAPSG or reduce its plasma clearance. Thus, a change in plasma pharmacokinetics can be ruled out as a cause of the observed BAPSG delivery enhancement to the retina and the brain. A possible explanation for the observation is inhibition of anion efflux transporters by probenecid, leading to reduced clearance of BAPSG from the retina and the brain. This we believe is the most likely reason for the results observed in this study.

Drug within the retina can be removed via the anterior part of the eye or the posterior part of the eye (16). The possible clearance pathways in the posterior region include drug removal by retinal blood vessels or transfer across retinal pigment epithelium followed by clearance via choriocapillaries. Potential transporters responsible for drug removal into or towards blood vessels include anion transporters, Oat3 and MRP1. Organic anion transporter protein, Oat3, has been recently suggested to be localized on the abluminal side of the rat retinal blood vessel endothelial cells (17). Further, evidence exists for MRP1 and other anion efflux transporters in the retinal pigment epithelium (11, 18). While we observed enhanced retinal and brain level of BAPSG following systemic drug and probenecid administrations, other reports indicated reduced anion drug clearance from the vitreous following probenecid co-treatment. Hosoya et al. (17), recently demonstrated that rat vitreal removal rates of three anionic drugs, p-aminohippuric acid, benzylpenicillin, and 6-mercaptopurine, can be reduced by probenecid (17). Thus, anion transporters such as Oat3 and MRPs might have contributed to enhanced BAPSG levels in the retina and the brain following probenecid treatment in this study.

Anionic efflux transporters were previously implicated in the brain disposition of several anionic solutes (19–22), and it was shown that inhibiting such efflux systems enhances drug delivery to the brain. Although inhibition of anion efflux transporters at the level of retina and brain can explain BAPSG delivery enhancement, we must consider the possibility of competition for plasma protein binding between probenecid and BAPSG, leading to release of greater quantities of free BAPSG in the presence of probenecid. Such elevation in free drug quantities can enhance drug diffusion or transport to the target tissues since unbound form of drug can readily utilize either simple diffusion or other specialized mechanisms of anion transport, unlike the drug-protein complex. Indeed, we previously observed that BAGPSG is highly bound to plasma proteins with a free fraction of 0.12% and 11.9% in plasma at 10 and 50 µg/mL total drug concentrations, respectively (15). Thus, release of free drug from plasma proteins is another possible explanation for our observations. However, if plasma free drug fraction is elevated, we should have observed a uniform elevation in drug levels in all the tissues assessed. However, this was not the case in the other tissues assessed, the lens and the cornea. Also, if there is a substantial change in the plasma free fraction of the drug, some changes in drug volume of distribution and/or clearance can be anticipated. This again, was not the case in this study. Thus, it appears that a change in retina and brain tissue efflux transporter activity is the likely predominant contributor to the observed enhancement in BAPSG delivery.

In this study we did not perfuse the vasculature including the retinal and brain blood vessels prior to estimation of tissue drug levels. Therefore, some drug levels in the tissue entrapped blood must have contributed to the observed tissue drug levels. However, the observed differences between treatments must be originating from tissue level differences since blood levels are not expected to be different between the treatments.

Unlike the above retinal and brain drug delivery enhancement in the rat study, in the rabbit eye drop study we did not observe an increase in BAPSG delivery to the tissues of the anterior segment including tears, cornea, conjunctiva, and aqueous humor. Since anion efflux transporters including MRPs have been identified in the corneal epithelium (23, 24), and because BAPSG can be effluxed by such transporters (11), an explanation for the lack of any difference in vivo with probenecid treatment includes either inadequate inhibition of such transporters or low activity of such transporters in vivo in the cornea and conjunctiva. Interestingly, we observed reduced BAPSG levels in the iris-ciliary body in the probenecid treatment group compared to the plain drug treatment. The underlying mechanisms have yet to be investigated. While there is evidence indicating active organic anion uptake into the iris or ciliary body from the aqueous humor side, the sensitivity of the involved transporter systems to probenecid is equivocal (25–27).

Following topical administration, therapeutic levels of the drug reaches the anterior segment of the eye including cornea, aqueous humor and lens, but generally not to the posterior segment that includes vitreous, retina and choroid (28, 29). However, drug delivery to the retina can be achieved by periocular and intraocular injections (3, 4, 30, 31). Since ARI therapy is intended for long-term, systemic administration would be a convenient and safe approach. However, following systemic administration, the blood- retinal and blood-aqueous barriers restrict the movement of molecules into the retina and anterior segment, respectively (1, 28), with the blood-retinal barrier being more restrictive. Consistent with this, following systemic administration of BAPSG, we observed greater drug levels in the cornea as opposed to the retina. The restrictive nature of blood-retinal barrier can be overcome in part by the organic anion transport inhibitors as described in this study.

The impact of this study is two-fold. One, retinal as well as brain delivery of anionic drugs such as BAPSG can be elevated following systemic pretreatment with an inhibitor for anion transporters. Second, the observed elevation in BAPSG delivery is particularly relevant for treating diabetic retinopathy. Aldose reductase inhibitors such as BAPSG are useful in treating multiple diabetic complications including retinopathy, cataracts, nephropathy, corneal epitheliopathy, and neuropathies (32). Thus, their purpose might be best served when administered by systemic routes. However, drug delivery to retina is low from this mode of administration (33). Use of anion transport inhibitors as demonstrated in this study might be useful in elevating retinal delivery of BAPSG. The value of aldose reductase inhibition in the brain has yet to be established. It is noteworthy that cognitive dysfunctions are now being identified as a complication of type 1 as well as type 2 diabetes (34). Further, there is evidence for elevated activity of aldose reducatase in the brain of diabetic animals (35). If ARIs are determined to be useful in treating such complications of the brain as well, the findings of this study in enhancing brain delivery of BAPSG would find an additional clinical application.

5. Conclusions

Efflux drug transporters are known to play a key role in the tissue accumulation of xenobiotics (36). Several experimental models including cell culture and knock-out animal models have been developed to study the role of these efflux transporters to explain drug resistance in cancer (37), poor oral absorption (6) and poor drug accumulation in the brain (38, 39). In this project, an important link between organic anion efflux transporter activity and anionic ARI accumulation in the retina and the brain was demonstrated. The findings of the present study conclusively show that treatment with an inhibitor of anionic efflux transporter, probenecid, enhances the retinal and brain accumulation of BAPSG. Our data also indicated that the increase in the retinal and brain levels is not due to the change in the systemic clearance of BAPSG upon probenecid treatment. However, BAPSG distribution to the lens and cornea was not effected by probenecid treatment, suggesting that probenecid is not altering the blood-aqueous barrier for ARI transport. Co-administration of BAPSG with probenecid following topical administration did not alter BAPSG disposition to the ocular tissues. Thus, inhibition of anion efflux transporters is a useful approach to elevate retinal and brain concentrations of an anionic aldose reductase inhibitor.