Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Tobacco smoking is strongly associated with the development of cardiovascular disease and is the number one preventable cause of death in the United States of America. Specifically, smoking has been associated with both hemorrhagic and nonhemorrhagic stroke (Gill et al., 1989). Smoking cessation therapies include abstinence ("cold turkey"), nicotine replacement, lobeline, clonidine, and certain antidepressant drugs. Yet, in spite of the availability of several treatments for smoking cessation, failure rates remain as high as 70%, even with the most effective protocols (Haustein, 2000). Recently developed quaternary ammonium analogs of nicotine may hypothetically have therapeutic benefits as smoking cessation agents, as suggested by their ability to act as subtype-selective nicotinic receptor antagonists and to inhibit nicotine-evoked dopamine release from superfused brain slices (Crooks et al., 1995; Dwoskin et al., 2000; Wilkins et al., 2002).

Blood-brain barrier (BBB) permeability of synthetic quaternary ammonium nicotine analogs may be limited, due to their physicochemical properties, since these molecules contain a positively charged N-n-alkylpyridinium moiety. The BBB is comprised of brain capillary endothelial cells connected by tight junctions circumferentially surrounding the cell margin (Butt et al., 1990), which prevents such charged molecules from entering the brain by passive permeation. The BBB presents drug permeation restrictions similar to a continuous cell membrane, allowing lipid-soluble molecule transport across the membrane, whereas molecules that are hydrophilic, charged, protein bound, or of large molecular weight have restricted permeation (Smith, 1990). Attempts have been made to increase brain drug delivery across the BBB either by increasing a drug's lipid solubility or by causing a temporary "opening" of the BBB (Greig, 1989) using osmotic methods or specific solutes including RMP-7. Additional methods used to augment brain drug delivery include direct brain or cerebrospinal fluid injection, intracarotid infusion to maximize brain arterial concentrations, inhibition of active removal from brain, or blocking drug metabolism (Smith, 1993). Consequently, assessment of drug delivery to the brain and the development of novel strategies to deliver such therapeutic agents to the CNS are of paramount importance in the design and discovery of drugs as CNS therapeutic agents.

The BBB choline transporter may be used to overcome restricted permeation of positively charged quaternary ammonium compounds across the BBB (Metting et al., 1998). Choline enters the CNS via a native nutrient BBB transport system (Diamond, 1971; Cornford et al., 1978; Metting et al., 1998; Allen and Smith, 2001). This transport system may be an effective strategy as a CNS delivery vector for drugs that have positively charged quaternary ammonium grouping within their structure, such as the quaternized ellipticines that are cytotoxic against isolated human brain tumor cells (Vistica et al., 1994). It has already been demonstrated that lymphoblast choline transporters are effective in delivering nitrogen mustard alkylating agents intracellularly (Goldenberg and Begleiter, 1980). These latter studies suggest that choline transporters can be used opportunistically to deliver drugs across cell membranes. As such, the BBB choline transporter may offer a similar opportunity for the transport of charged molecules and may have wide application for the delivery of therapeutics to treat CNS disorders.

In the present article, the ability of synthetic choline and nicotine analogs to inhibit BBB choline transport was evaluated, providing initial information on the ability of this transporter to deliver potential drug candidates, such as N-n-octylnicotinium iodide (NONI), to brain, as well as providing information about the choline transporter pharmacophore. The results suggest that charged, polar compounds can be designed that have high affinity for the BBB choline transporter and that can be delivered via this transporter to the CNS.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Subjects. Male Fischer-344 rats (220-330 g) were obtained from Charles River Laboratories, Inc. (Wilmington, MA) and were used for all experiments described herein. All studies were approved by the Animal Care and Use Committee and were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

In Situ Rat Brain Perfusion Technique. The in situ rat brain perfusion technique was used to evaluate BBB transporter binding and uptake of nicotine and choline analogs into brain. Analog access to brain via the BBB choline transporter was evaluated indirectly by assessment of analog-induced inhibition of [³H]choline uptake into brain. Promising candidates were evaluated directly by determining the CNS distribution parameters of the ³H-labeled analog (i.e., [³H]NONI) and by determining the ability of natural substrates for the choline transporter to inhibit uptake of the ³H-labeled analog.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Structures of nicotine, lobeline, choline, and the analogs of choline and nicotine.

Chemicals. [³H-Methyl]choline chloride (~80 Ci/mmol), [¹⁴C]sucrose (4.75 mCi/mmol), and [³H]diazepam (79.5 Ci/mmol) were obtained from PerkinElmer Life Sciences (Boston, MA). [³H]NONI (10 mCi/mmol) was obtained from Sibtech (Newington, CT; random tritium labeling with >90% of the tritium label incorporated in the N-n-octyl chain). All radiolabeled compounds were dried completely before use to eliminate possible contaminants, including [³H]H₂O. Unlabeled choline chloride, lobeline hydrochloride, S-()-nicotine, and components of the physiologic buffer (NaCl, NaHCO₃, KCl, NaH₂PO₄, CaCl₂, MgSO₄, and d-glucose) were obtained from Sigma-Aldrich (St. Louis, MO). Unlabeled NONI, NMNI, NBNI, NDNI, NMPI, and NOPI were prepared as previously described (Crooks et al., 1995; Xu et al., 2002).

Calculations. Concentrations of tracer in brain and perfusion fluid were expressed as disintegrations per minute per gram of brain or disintegrations per minute per milliliter of perfusion fluid, respectively. Blood-brain barrier transport was determined using the initial uptake method, as previously described (Takasato et al., 1984, 1989; Smith, 1996). In preliminary experiments, linear and unidirectional [³H]NONI uptake into brain was determined by perfusion with [³H]NONI (13.8 µM) for a 5- to 60-s period. Unidirectional uptake transfer constants (K_in) were calculated from the following relationship to the linear portion of the uptake curve.

(1)

where Q* is the quantity of [³H]tracer in brain (disintegrations per minute per gram) at the end of perfusion, C* is the perfusion fluid concentration of [³H]NONI (disintegrations per minute per milliliter), T is the perfusion time (s), and V_o is the extrapolated intercept (T = 0 s; "vascular volume" is measured in milliliters per gram). After determination of a perfusion time that allowed an adequate amount of tracer to pass into brain and yet remained in the linear uptake zone, K_in was determined in single time point experiments as K_in = [Q*  V_oC*]/C*T] (Takasato et al., 1984). When [³H]choline uptake was evaluated, vascular tracer was removed in most experiments by a brief intravascular wash (15 s) with tracer-free perfusion fluid. K_in values were converted to apparent cerebrovascular permeability-surface area products (PA) using the Crone-Renkin equation (Smith, 1989).

(2)

where F is cerebral perfusion fluid flow. In all instances, PA differed by <2% from K_in because F exceeded K_in by >40-fold.

(3)

Statistics. Data presented are generated from the frontal cerebral cortex with comparable data seen in parietal and occipital cortical regions, as well as the hippocampal, striatal, thalamic, cerebellum, and midbrain regions (data not shown). Inhibition of brain [³H]choline uptake by choline analogs and [³H]NONI brain uptake values were expressed as means ± S.E.M. for n = 3-5 independent determinations for each compound evaluated. Data were analyzed by analysis of variance with Bonferroni correction for multiple comparisons (Instat; GraphPad Software, San Diego, CA). Differences between the means were considered significant at p < 0.05. Power analyses were calculated using GraphPad StatMate version 1.01i (GraphPad Software).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Inhibition of [³H]Choline Brain Uptake. Binding of choline and nicotine analogs to the BBB choline transporter was determined by evaluation of their ability to inhibit [³H]choline uptake into brain. These studies were conducted to provide insight into the potential ability of these compounds not only to bind to the transporter but also to determine their potential ability to be transported across the BBB from the vasculature compartment into brain. Baseline [³H]choline PA values (1.26 ± 0.04 × 10³ ml/s/g) obtained in the present experiments were similar to previously published values (Allen and Smith, 2001; Lockman et al., 2001). Given that nicotine and lobeline are currently used as smoking cessation therapies, both were assessed for their ability to inhibit choline uptake into brain at a single concentration of 250 µM. Although nicotine did not significantly reduce choline PA (PA_i ~1.10 ± 0.06 × 10³ ml/s/g), lobeline did inhibit brain [³H]choline uptake with an apparent K_i value of 393 ± 51 µM, which is within an order of magnitude of the K_m for choline (Table 1). It is important to note that from previous studies using the capillary depletion method evaluating brain [³H]choline uptake that no significant endothelial cell association of choline occurred (Allen and Smith, 2001). Furthermore, washout studies using the capillary depletion method indicated that no differences in endothelial cell sequestration or binding occurred. Taken together, this uptake represents true [³H]choline brain penetration.

[³H]NONI Brain Uptake. Given the initial K_i values for NONI, experiments were performed to verify that this compound gains access to brain via the BBB choline carrier. The brain distribution parameters of [³H]NONI were evaluated using the rat brain perfusion method. Uptake of [³H]NONI (1 µCi/ml) into brain was evaluated from 0 to 60 s in the absence of unlabeled NONI. Brain/perfusion fluid ratios (i.e., volume of distribution or "space") were plotted as a function of time and are illustrated in Fig. 2. The transfer coefficient value (K_in) for [³H]NONI uptake was determined to be 1.59 ± 0.14 × 10³ ml/s/g, calculated according to the slope of the compound accumulating in brain versus time (Smith, 1989). An uptake time of 45 s was chosen as within the linear portion of the brain uptake curve to evaluate [³H]NONI brain uptake in the presence of unlabeled NONI. Unlabeled NONI (250 µM) in the perfusion fluid resulted in 46% inhibition of [³H]NONI brain uptake. The inhibition suggests saturable kinetic parameters associated with transporter-mediated BBB transport is present for NONI (Fig. 3).

View larger version (9K): [in this window] [in a new window]   Fig. 2.   Linear time course of [3H]NONI (1 µCi/ml) uptake into rat brain. Calculated transfer coefficient is related to slope values (Smith, 1996); Kin ~1.59 × 102 (r2 = 0.985). Data are frontal cortex brain perfusion concentration ratios (i.e., brain distribution spaces) versus time. Each point is the mean ± S.E.M. (n = 3).

View larger version (9K): [in this window] [in a new window]   Fig. 3.   Inhibition of brain [3H]NONI (1 µCi/ml) PA coefficient (control) by the addition of 250 µM NONI to the perfusion fluid on frontal cortex. Data are the mean ± S.E.M. (n = 6-8). , indicates significant difference (p < 0.05) from control.

3

3

4

View larger version (11K): [in this window] [in a new window]   Fig. 4.   Inhibition of brain cortex [3H]NONI PA coefficient (control) by the addition of unlabeled choline (500 µM and 5 mM). Data are the mean ± S.E.M. (n = 4-8). , indicates significant difference (p < 0.05) from control.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Pharmacological intervention in neurological diseases is often limited by poor access of therapeutic agents into the CNS. Approaches to successful therapy in CNS diseases must consider the BBB as a potential impediment. Numerous methods have been used with limited success to deliver polar drugs to the CNS (Pardridge, 1997, 1998). Use of antibodies, drug lipidization, and prodrug development have been used with some success. Use of small molecules that directly target transport proteins to overcome BBB restrictions eliminates the need for the drug to be biotransformed in brain and linking to antibodies. Lipidization can result in an undesirable pharmacokinetic profile (Greig, 1989). We have proposed to use BBB transport proteins such as the choline transporter as brain drug delivery vectors for polar drugs (Metting et al., 1998); this approach has been successful with regard to the amino acid transporter and the transport of L-DOPA and gabapentin into brain (Smith, 1993). In this respect, the present study focused on the BBB choline transporter for drug delivery because this transporter may offer the advantage of delivering cholinergic and nicotine-like drugs to brain (Metting et al., 1998). Of significance in this article is that the BBB choline transporter efficiently transports choline, a small charged molecule with minimal passive permeation, to the brain. Evaluation of this transporter system may afford a possible brain entry portal for similarly charged therapeutic molecules. To date, this system has not been established as a vector for such CNS delivery.

To use a BBB transporter protein as a CNS drug delivery vector, multiple factors must be considered. These factors include: 1) kinetic availability to transport physiologic molecules, 2) structural binding requirements of the transporter, 3) therapeutic compound manipulation so that the compound binds but also remains active in vivo, and 4) actual transport of the molecule (not just binding to the transporter) into brain. This discussion will address each issue with regard to the novel nicotine analogs studied in this study and will demonstrate delivery of NONI to brain via the BBB choline transporter.

In vivo choline transport at the BBB has been demonstrated to be carrier-mediated and saturable (Cornford et al., 1978; Allen and Smith, 1999; Allen and Smith, 2001). Endogenous choline plasma levels are less than the calculated K_m for the transporter, and thus, with respect to availability of the transporter, it is not saturated under physiological conditions. These characteristics, coupled with the high transport capacity of the choline transporter and the adequate blood-to-brain transfer rate, suggests this carrier has the necessary characteristics to deliver therapeutic molecules to brain (Smith, 1993).

Structural binding requirements of the choline transporter have been extensively characterized in multiple tissues using binding inhibition studies (for a review, see Lockman and Allen, 2002). Previous studies reveal that the principal structural moiety in the choline molecule required for recognition by the transporter binding site is the quaternary ammonium group, which is believed to interact with a corresponding anionic group on the transporter binding site. Simon et al. (1975) demonstrated this relationship and observed that simple quaternary ammonium cations, such as tetraethylammonium and tetramethylammonium, were inhibitors of choline uptake in striatal synaptosomes. Furthermore, when longer alkyl moieties were substituted for the N-methyl and N-ethyl groups, increased binding affinity to the choline transporter was observed (Dowdall et al., 1976). The increased affinity of these modified compounds was hypothesized to result from a hydrophobic interaction with the lipid membrane (Lerner, 1989). The present results with the N-n-alkylnicotinium and N-n-alkylpyridinium analogs are consistent with these reports (Table 1) and demonstrate that: 1) N-n-alkylazaaromatic compounds containing long-chain n-alkyl groups and, which are charged at physiologic pH, bind significantly to the choline transporter; 2) the hydroxyl group of the choline molecule is not an absolute requirement for binding to the choline transporter; and 3) when the 3-hydroxy group in the choline molecule is removed, replacement of one of the N-methyl groups with a longer N-n-alkyl group results in compounds with good affinity for the choline transporter and appears to completely overcome the hydroxy group requirement that has been previously observed (Allen et al., 1996b). As indicated in Table 1, the transporter affinity of these compounds is close to that of choline itself suggesting comparable binding. As these compounds are required in much lower concentrations in brain to exert effects, the potential to deliver these agents to brain in sufficient quantities to be therapeutic seems likely. Taken together, these findings suggest the N-n-alkyl chains may be interacting with a proximal lipophilic domain and facilitating binding and/or transport.

The N-n-alkylnicotinium compounds evaluated in this study were previously examined for their ability to inhibit nicotine-evoked dopamine release from superfused rat striatal slices (Crooks et al., 1995; Dwoskin and Crooks, 2001; Wilkins et al., 2002). Importantly, these nicotine analogs were observed to selectively inhibit the 362* neuronal nicotinic acetylcholine receptor, which has been suggested to mediate nicotine-evoked dopamine release from its presynaptic terminals in striatum. Specifically, given the ability of NONI to inhibit the effect of nicotine on dopaminergic systems in vitro (IC₅₀ ~1.1 µM; Crooks et al., 1995; Wilkins et al., 2002) and the calculated apparent K_i at the BBB choline transporter (Table 1), the ability of NONI to access the brain via active transport through the BBB choline carrier was further explored.

The results of the present study show that NONI, the N-n-octyl derivative of nicotine, is transported into brain via the BBB choline transporter. This is a significant finding because from structural and physicochemical considerations, one would predict that the ability of NONI to penetrate the BBB and enter the CNS would be poor, due to the charged, polar nature of the compound. [³H]NONI brain/perfusion fluid ratios were plotted versus time (Fig. 2), and the calculated transfer coefficient value based upon slope values (Smith, 1989) was determined to be 1.59 × 10³ ml/s/g. These data suggest that, despite NONI being a charged quaternary ammonium compound, significant brain penetration occurs that is greater than what would be expected from passive permeation alone, indicating an active transport process.

To further confirm that [³H]NONI penetration into brain is an active process, unlabeled NONI (250 µM) was added to the perfusion fluid. If [³H]NONI were to enter the brain via active transport, unlabeled NONI would theoretically compete with the transport of the radiolabeled component, and thus reduce [³H]NONI brain uptake. Indeed, Fig. 3 illustrates that [³H]NONI brain uptake was reduced by ~46% in the presence of 250 µM unlabeled NONI. These data provide further evidence that [³H]NONI is not entering brain by passive permeation alone. To determine the role of the BBB choline transporter in [³H]NONI brain uptake, unlabeled choline was added to the perfusion buffer. As with the previous experiment, if [³H]NONI enters the brain via the BBB choline transporter, the presence of unlabeled choline should diminish [³H]NONI brain entry. Given that the K_m of choline for the transporter is ~45 µM (Allen and Smith, 2001), the unlabeled choline concentrations would be sufficient to saturate the transporter and effectively block the transport of [³H]NONI via this carrier. In fact, the presence of 500 µM choline tended to decrease uptake by ~25%, whereas 5 mM significantly reduced uptake to ~50% of control (Fig. 4). Taken together, the present results indicate that the BBB choline transporter has a substantive role to play in the active transport of [³H]NONI into brain. It is important to note that, although the K_i value for NONI is ~49 µM, the amount of choline required for inhibition is higher. The concentrations of NONI required to inhibit the effect of nicotine on dopaminergic systems, however, is significantly lower, in the range of 1.1 µM as noted above. As such, delivery of sufficient NONI to brain to cause an effect can occur with these physiologic conditions. Furthermore, considering both brain distribution and the previous demonstration that NONI inhibits nicotine-evoked dopamine release by selectively interacting with the nicotinic receptor subtype mediating this effect, it is likely that NONI may have significant potential as a CNS therapeutic agent, and if it can be shown that NONI has a good brain distribution profile, it may have a significant use in smoking cessation therapy and stroke prevention. The current observations that the charged, polar N-n-alkylnicotinium analogs studied herein not only inhibited the effect of nicotine on dopamine systems in the CNS but also gained access to the brain from the periphery, in spite of potential physicochemical limitations, is a key finding in the development of NONI as a potential smoking cessation agent.

Previously, we have identified derivatives of lobeline and isoarecolone that bind to the BBB basic amine transporter (Metting et al., 1998). In light of the findings that NONI is transported in to brain via the BBB choline transporter, further studies are warranted on lobeline and isoarecolone derivatives to determine whether the BBB choline transporter can act as vector-mediated uptake system to also deliver these drugs into the brain.

This article provides, for the first time, evidence that the BBB choline transporter can be used as a brain drug delivery vector for nicotine and choline analogs. Furthermore, the results suggest that other structurally related charged quaternary ammonium compounds containing lengthy N-n-alkyl groups may also exhibit increased brain distribution via active choline transport.