Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The free passage of both endogenous and exogenous molecules into the brain is restricted by the presence of both the BBB, located at the cerebrovascular endothelium, and the blood-CSF barrier, found at the level of the choroid plexuses and arachnoid membrane. Analgesia, produced by delta opioid receptors, is thought to be a centrally mediated event, so only those agonists that can cross these barriers and enter the brain can produce an antinociceptive effect (Frederickson et al., 1981; Ward and Takemori, 1983; Galligan et al., 1984; Shook et al., 1987). The ability to enter the CNS is thus an essential consideration in the design of opioid analgesic drugs (Abbruscato et al., 1996).

The delta opioid receptor-selective, enzymatically stable peptide [D-penicillamine^2,5]enkephalin (DPDPE) recently acquired special significance, because a saturable uptake system for this analgesic into the CNS has been identified (Williams et al., 1996). This latter study utilized a bilateral in situ brain perfusion technique and measured a time-dependent CNS uptake of [³H]DPDPE over a period of 30 min; blood-to-brain and blood-to-CSF unidirectional rate constants, K_in, were calculated as 1.46 ± 0.31 and 0.99 ± 0.29 µl · min¹ · g¹, respectively. Furthermore, it was determined that the saturable uptake system for [³H]DPDPE was likely to be found at the BBB, the blood-CSF barrier playing only a minor role in the brain uptake of this enkephalin analog.

Independent saturable transport systems for the entry of endogenous peptides, amino acids, hexoses, nucleosides and monocarboxylic acids into the CNS have previously been identified (Davson and Segal, 1995). Furthermore, it is known that certain exogenous compounds can also enter the brain via these systems; for example, a dynorphin-like analgesic peptide, E-2078, has been shown to use absorptive-mediated endocytosis to enter the CNS (Terasaki et al., 1989), and the antitumor amino acid analog acivicin has been shown to be transported across the BBB by the large neutral amino acid carrier (Williams et al., 1990; Takada et al., 1991; Chikhale et al., 1995). Interestingly, leucine-enkephalin, which can enter the brain by a saturable uptake system (Zlokovic et al., 1989), failed to inhibit the CNS entry of [³H]DPDPE, which suggests that there are separate saturable uptake systems for DPDPE and the endogenous enkephalins (Williams et al., 1996).

In order to provide the information necessary to develop DPDPE as a clinical analgesic, as well as DPDPE analogs with a greater ability to enter the CNS, the aim of this present study was to characterize further the uptake of [³H]DPDPE into the CNS, in terms of saturation kinetics and specificity.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

In situ brain perfusion technique. The experimental protocol was approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Arizona. The bilateral brain perfusion method has recently been described by Williams et al. (1996) and will be only briefly summarized here. Adult Sprague-Dawley rats weighing 250 to 350 g were anaesthetized (sodium pentobarbital; 64.8 mg · kg¹) and heparinized (10,000 U · kg¹). Both common carotid arteries were cannulated with fine silicone tubing connected to a perfusion system. The perfusion fluid consisted of a protein containing Ringer (NaCl 117.0 mM, KCl 4.7 mM, MgSO₄ (3H₂O) 0.8 mM, NaHCO₃ 24.8 mM, KH₂PO₄ 1.2 mM, CaCl₂ (6H₂O) 2.5 mM, D-glucose 10 mM, dextran (MW 70,000) 39 g · l¹ and bovine serum albumin 1 g · l¹). Trace amounts of Evans blue were added to the perfusion medium for visualization. The perfusion fluid was gassed with 5% CO₂ in O₂ (PO₂ 640-730 mm Hg), warmed to 37°C, filtered and debubbled before entering the animal at a flow rate of 3.1 ml · min¹. With the start of perfusion, the right and left jugular veins were sectioned to allow drainage of the perfusion medium. The perfusion pressure was continuously monitored and ranged from 85 to 96 mm Hg. [³H]DPDPE and [¹⁴C]-sucrose, the vascular space marker, in the absence and presence of inhibitors, were then infused via a slow-drive syringe pump into the inflowing mammalian Ringer for 20 min. After the set perfusion time, a cisterna magna CSF sample (approximately 50 µl) was taken, and perfusion was terminated by decapitation. The brain was removed, choroid plexuses excised and the brain homogenized. Brain tissue samples (approximately 50 mg wet wt.), together with the CSF and 100-µl perfusate samples, were prepared for radioactive counting (Beckman LS 5000 TD counter; Beckman Instruments Inc., Fullerton, CA) by the addition of 1 ml of tissue solubilizer (TS-2; Research Products Inc., Mount Pleasant, IL). After solubilization, 30 µl of glacial acetic acid was added to eliminate chemiluminescence and, together with 4 ml of Budget Solve Liquid Scintillation Cocktail (Research Products Inc.), made possible accurate liquid scintillation counting. The ³H and ¹⁴C activities were converted from counts per minute to disintegrations per minute with the use of internal stored quench curves.

Calculation of kinetic constants. The amount of radioactivity in the brain and CSF, C_Tissue (dpm · g¹ or dpm · ml¹), was then expressed as a percentage of that in the artificial perfusate, C_pl (dpm · ml¹), and termed R_Tissue %.

(1)

Unidirectional rate constants, K_in (µl · min¹ · g¹), were determined by single time-point analysis, as previously described (Zlokovic et al., 1986; Williams et al., 1996), where T is the time in minutes:

(2)

Blood-to-brain unidirectional transfer constants determined in this manner were corrected for vascular space by subtracting [¹⁴C]sucrose (R_Brain) from [³H]DPDPE (R_Brain).

1

1

(3)

1

1

(4)

1

1

(5)

(6)

(7)

1

1

1

1

Labeled substances. [³H₂]Tyr¹-DPDPE (18.1 Ci/mmol) was provided by Multiple Peptide Systems (San Diego, CA) under the direction of the National Institute on Drug Abuse (NIDA). Radiochemical purity was found to be 97% for [³H]DPDPE as determined by rpHPLC analysis (Williams et al., 1996). [¹⁴C]Sucrose (0.44 Ci/mmol) was purchased from NEN Research Products (Boston, MA).

Unlabeled substances. DPDPE and ICI 174,864 (allyl²-Tyr-AIB-Phe-OH) were provided by Multiple Peptide Systems under the direction of NIDA. Poly-L-lysine hydrobromide (up to MW 5000) was obtained from Bachem Bioscience Inc. (King of Prussia, PA). Porcine insulin and 2-aminobicylo-[2,2,1]-heptane-2-carboxylic acid (BCH) were purchased from Sigma Chemical Company (St. Louis, MO).

Data analysis. For all experiments, the data were presented as means ± S.E.M. Student's t test was used for the comparison of two means, and statistical significance was taken as P < .05.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

The uptake of [³H]DPDPE into the CNS was studied under conditions in which the brain was perfused with various concentrations (1 to 100 µM) of unlabeled DPDPE (fig. 1). These experiments revealed a saturable uptake of [³H]DPDPE into the brain. At a concentration of 10 µM unlabeled DPDPE in the perfusion medium, the uptake of [³H]DPDPE into the brain was inhibited by approximately 23%. Brain uptake of [³H]DPDPE was not significantly inhibited further with either 50 or 100 µM unlabeled DPDPE. The CSF uptake of [³H]DPDPE was not significantly affected by the presence of 1 to 100 µM unlabeled DPDPE (fig. 1). The half-saturation constant K_m, the maximal transport rate V_max and the diffusion constants K_d were determined from the vascular brain perfusion data are shown in table 1.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   The relative uptake of [3H]DPDPE into the brain () and CSF () measured as a function of unlabeled DPDPE concentration. Uptake has been divided into saturable and nonsaturable components. The proportion of the brain uptake of [3H]DPDPE that represents vascular space is also shown. Values are mean ± S.E.M.

Table 2 shows the effect of the large neutral amino acid transport inhibitor BCH on the uptake of [³H]DPDPE into the brain and CSF. Although BCH did not significantly inhibit the entry of [³H]DPDPE into the CNS, it did significantly increase the entry of [³H]DPDPE into the brain. The effects of increasing concentrations of poly-L-lysine on the brain uptake of [¹⁴C]sucrose are shown in figure 2. This histogram illustrates that poly-L-lysine, at concentrations  5 µM, causes loss of BBB integrity as measured by the vascular space marker [¹⁴C]sucrose. Although 0.3 µM poly-L-lysine had no significant effect on cerebrovascular space, it was found to increase significantly the uptake of [³H]DPDPE into the brain (P < .01) (table 2). It should also be noted that at poly-L-lysine concentrations  5 µM, there was a visible extravasation of Evans blue into the brain substance, which was not observed in the control or 0.3 µM poly-L-lysine experiments. Evans blue, in effect, measures albumin permeability because it binds tightly to albumin under physiological conditions (Rubin et al., 1991).

View larger version (24K): [in this window] [in a new window]   Fig. 2.   Histogram illustrating the effect of poly-L-lysine concentration on the brain uptake of [14C]sucrose, a marker molecule that under normal conditions remains predominantly within the cerebrovascular space.

Figure 3 illustrates the effect of the delta opioid receptor antagonist ICI 174,864 and the pancreatic hormone insulin on the uptake of [³H]DPDPE into the CNS. However, although ICI 174,864 and insulin did not inhibit the uptake of [³H]DPDPE, insulin did significantly increase uptake into the brain (P < .01) and CSF (P < .05). The [¹⁴C]sucrose/cerebrovascular space values obtained in the presence of 50 µM ICI 174,864 (2.93 ± 0.37%) or 10 µM insulin (2.32 ± 0.14%) were not significantly different from control values (1.98 ± 0.33%).

View larger version (32K): [in this window] [in a new window]   Fig. 3.   The effect of ICI 174,864, a delta opioid receptor antagonist, and insulin on the uptake into the brain and CSF of [3H]DPDPE. Data are mean ± S.E.M. (bars) for 3 to 4 animals. Values have been corrected for [14C]sucrose/vascular space. **P < .01; *P < .05 by Student's t test.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The kinetic analysis of [³H]DPDPE uptake into the CNS was determined by using six concentrations of unlabeled DPDPE in the range 1 to 100 µM (fig. 1). The results of this study indicated that the brain uptake of [³H]DPDPE consisted of both saturable and nonsaturable components, which could be described by Michaelis-Menten type kinetics with K_m = 45.5 ± 27.6 µM, V_max = 51.1 ± 13.2 pmol · min¹ · g¹ and K_d = 0.6 ± 0.3 µl · min¹ · g¹ (Table 1). These kinetic constants reflect the presence of a saturable uptake system for [³H]DPDPE with a relatively large affinity and a low capacity. Interestingly, although leucine-enkephalin failed to inhibit significantly the CNS uptake of [³H]DPDPE (table 2; Williams et al., 1996), leucine-enkephalin has previously been found to enter the brain by a saturable uptake system with a relatively large K_m value, 34 to 41 µM, and a low capacity, 0.14 to 0.16 nmol · min¹ · g¹ (Zlokovic et al., 1989). In contrast to brain uptake, the CSF entry of [³H]DPDPE was found to be purely nonsaturable with a diffusion constant of 0.9 ± 0.1 µl · min¹ · g¹ (fig. 1). Several peptides have been shown to cross the BBB and/or the blood-CSF barrier via passive diffusion, e.g., thyrotropin-releasing hormone (Zlokovic et al., 1985; 1988).

DPDPE is currently undergoing Phase 1 clinical trials to determine its therapeutic plasma concentration range. However, it is known that DPDPE, at a concentration of 60 mg/kg, can elicit significant analgesia within 5 min of being administered i.v. in the mouse (Weber et al., 1991). It is also known that approximately 4% of [³H]DPDPE administered i.v. can be found in the blood after 5 min. If we suppose that 4% of a 60-mg/kg dose of DPDPE remains in the blood, then this will result in a total blood concentration in a 30-g mouse of 0.072 mg. A mouse has approximately 2 ml of blood, i.e., 0.036 mg/ml or 56 µM. The K_m value determined in the present study was 46 µM, so if 0.036 mg/ml was a therapeutic blood concentration, then both the saturable and the nonsaturable uptake systems characterized in the present study for [³H]DPDPE would be important for the brain entry of this analgesic.

The antitumor amino acid analog, acivicin, has been shown to be transported across the BBB by the large neutral amino acid carrier (Williams et al., 1990; Takada et al., 1991; Chikhale et al., 1995). Although DPDPE (MW 645.8) is considerably larger than acivicin (MW 178.6), it is possible that the N-terminal tyrosine of DPDPE enables it to use this large neutral amino acid carrier at the BBB. However, BCH, a nonmetabolizable analog that is known to be specific for this transporter, failed to inhibit the uptake of [³H]DPDPE into either the brain or the CSF (table 2). This indicates that [³H]DPDPE does not utilize the large neutral amino acid transporter to enter the CNS and confirms our previous work, which indicates that [³H]DPDPE crossed the BBB intact and was not metabolized to produce [³H]tyrosine (Williams et al., 1996). It is unclear why there was a significant increase in the uptake of [³H]DPDPE into the brain in the presence of BCH, but one could hypothesize that the transport mechanism for [³H]DPDPE was stimulated by the saturation of the large neutral amino acid transport mechanism.

Peptides have been found to enter the CNS by several saturable mechanisms, including absorptive-mediated endocytosis (Kumagai et al., 1987), receptor-mediated transport (Pardridge et al., 1985) and carrier-mediated transport (Banks and Kastin, 1987). In an attempt to clarify further the saturable mechanism by which [³H]DPDPE enters the CNS, we examined uptake in the presence of a variety of substances (table 2; fig. 3).

It is known that the luminal and abluminal endothelial cell membranes of the BBB have anionic sites (Vorbrodt, 1989), and it has been suggested that these sites restrain vesicle and channel formation for transport into or through the endothelial cell (Hardebo and Kåhrström, 1985). In addition, the negative surface charge prevents the adhesion of anionized substances on the intima (Danon and Skutelsky, 1976), so cationized macromolecules are more capable of nonspecific adsorption onto the cell membrane (Griffin and Giffels, 1982). The isoelectric point (pI) of a compound therefore plays an important role in its uptake by brain capillaries (Nagy et al., 1983; Vorbrodt et al., 1995) and in its passage across the choroid plexus (Griffin and Giffels, 1982). For example, positively charged peptides, such as cationized albumin, have been shown to use absorptive-mediated endocytosis to cross the BBB (Kumagai et al., 1987). Furthermore, in contrast to cationized ferritin, native (anionic) ferritin has been shown to have a limited passage across the choroid plexus epithelium (Van Deurs et al., 1981). DPDPE (pI ~ 6.5) is slightly negatively charged at physiological pH, so it would seem unlikely to use absorptive-mediated endocytosis to enter the CNS (Baldwin and Chien, 1984; Shen et al., 1992).

Poly-L-lysine (3-300 µM) has been shown to inhibit the binding of E-2078, ebiratide and cationized albumin, to isolated brain capillaries (Kumagai et al., 1987; Terasaki et al., 1989; Shimura et al., 1991). However, in vivo intracarotid artery infusions of 5 µM poly-L-lysine have been shown to produce significant Evans blue/albumin leakage, a result that suggests loss of BBB integrity (Westergren and Johansson, 1993). In the present study, poly-L-lysine, at concentrations in the perfusion medium ranging from 5 to 300 µM, also caused visible extravasation of Evans blue, as well as an increase in [¹⁴C]sucrose/vascular space, compared with control groups (fig. 2). In contrast, 0.3 µM poly-L-lysine in the perfusion medium did not significantly affect the visible extravasation of Evans blue or the [¹⁴C]sucrose/vascular space. The polycation protamine sulphate has also been shown to affect the extravasation of a vascular space marker in a dose-dependent manner (Nagy et al., 1983). Furthermore, poly-L-lysine at picomolar concentrations has been shown not to affect the passage of albumin or horseradish peroxidase across mouse fibroblast membranes in vitro (Shen and Ryser, 1978). Interestingly, in the presence of 0.3 µM poly-L-lysine, the uptake of [³H]DPDPE into the brain did increase approximately 3-fold (table 2). This would suggest that 0.3 µM poly-L-lysine significantly affects the cellular uptake of [³H]DPDPE, not its passage through interendothelial junctions. Previous studies have suggested that the focal points of this increase in barrier permeability are the endothelial tight junctions (Nagy et al., 1983), and in the present study, this may be so at poly-L-lysine concentrations of 5 to 300 µM. Poly-L-lysine has also been shown to result in glomerular epithelial cell swelling (Seiler et al., 1975), and it may be that endothelial/epithelial cell death is responsible for the loss of BBB integrity (Nagy et al., 1983). Although it is known that polycation infusions result in neutralization of the negative surface charge (Nagy et al., 1983), how 0.3 µM poly-L-lysine increases the blood-brain passage of [³H]DPDPE has not been established. It may be due to the opening up of hydrophilic channels in the endothelial cell membrane (Nagy et al., 1983) or to stimulation of transendothelial vesicular transport (Westergaard, 1980). However, the results of the present study do suggest that poly-L-lysine is not in competition with [³H]DPDPE. It is likely that if absorptive-mediated endocytosis is responsible for the poly-L-lysine-mediated increased passage of [³H]DPDPE, it is not the saturable uptake system identified in the absence of poly-L-lysine (fig. 1).

Receptors have been shown to mediate the endocytosis of insulin, insulin-like growth factors and transferrin across the BBB (Pardridge et al., 1985; Duffy et al., 1988; Fishman et al., 1987), and insulin receptors are known to be expressed at the blood-CSF barrier (Baskin et al., 1986). Figure 3 illustrates the absence of an effect by the delta opioid receptor antagonist ICI 174,864 at a concentration above that of the K_m value for saturable DPDPE transport. This would suggest that delta opioid receptors do not mediate the saturable uptake of this delta opioid receptor-selective peptide, [³H]DPDPE, through the BBB (Terasaki et al., 1989; Zlokovic et al., 1989). In addition, the uptake of [³H]DPDPE into the brain and CSF was not significantly decreased in the presence of insulin, even at a concentration of 10 µM, which can occupy the receptors (Shimura et al., 1991). This suggests that the receptor-mediated endocytosis system responsible for the passage of insulin into the CNS is not involved in the saturable uptake of [³H]DPDPE into the brain. Furthermore, it has been hypothesized that different saturable mechanisms of peptide transport could be manifested in the different rate and saturation constants for individual peptides and proteins (Zlokovic et al., 1990; Shimura et al., 1991; Poduslo et al., 1994). The kinetic constant (K_m) determined for the brain flux of [³H]DPDPE was in the low micromolar range (table 1), which would also appear to reflect carrier-mediated, rather than receptor-mediated, [³H]DPDPE uptake (Zlokovic et al., 1990).

It is of interest that insulin was found to increase significantly the uptake of [³H]DPDPE into the brain and CSF (fig. 3), while not affecting [¹⁴C]sucrose/vascular space. It is known that insulin affects the transmembrane transport of different substances, particularly glucose, into numerous different kinds of cells (Ayre, 1989). Previous reports have provided evidence for insulin potentiation of drug uptake by cells; for example, insulin has been shown to increase the cytotoxic effect of methotrexate, an anticancer agent, in human breast cancer cells in vitro (Alabaster et al., 1981). Although the authors suggested that the increased cytotoxicity was related to an increase in the sensitivity of the cells to methotrexate that resulted from a metabolic modification by insulin, a later proposal suggested that insulin had induced changes in cellular lipid synthesis and membrane lipid profile, resulting in changes in membrane fluidity and enhanced methotrexate transport (Schilsky et al., 1981). The enhanced cytotoxicity was suggested to be related to an increased capacity of the cells to accumulate free intracellular methotrexate. Another study showed that insulin increased the passage across the BBB of the deoxyribonucleoside analog azidodeoxythymidine (Ayre et al., 1989). It has been proposed that "through an unidentified interaction with its specific receptors on brain capillary endothelial cells, insulin facilitates the passage of drug molecules across the BBB into the brain" (Ayre, 1989). The results of the present study indicate that insulin can also alter the brain uptake of the analgesic [³H]DPDPE. Thus this study provides further evidence for the insulin-potentiation theory (fig. 3). The significance of insulin as a therapeutic modality in the alleviation of pain could prove invaluable if we could safely control its known side effects.

In conclusion, the present study has demonstrated that an enzymatically stable enkephalin analog, DPDPE, can enter the brain via both saturable and nonsaturable uptake mechanisms, which can be described by Michaelis-Menten type kinetics with a K_m value of 45.5 ± 27.6 µM, a V_max value of 51.1 ± 13.2 pmol · min¹ · g¹ and a K_d value of 0.6 ± 0.3 µl · min¹ · g¹. In addition, the CSF uptake of [³H]DPDPE could not be self-inhibited (K_d = 0.9 ± 0.1 µl · min¹ · g¹). Further experiments revealed that [³H]DPDPE was not entering the CNS via the large neutral amino acid transporter, by binding to opioid receptors or by the leucine-enkephalin uptake system. It would also appear that although [³H]DPDPE is not in competition with either poly-L-lysine or insulin to enter the CNS, both substances do have significant effects on the uptake of [³H]DPDPE, and this may have valuable clinical implications. It is still unclear which saturable uptake mechanism is responsible for the entry of [³H]DPDPE into the CNS, but overall the results suggest a carrier-mediated transport system, which may be valuable for other peptide and nonpeptide drugs.