Methodology for Benzodiazepine Receptor Binding Assays at Physiological Temperature Rapid Change in Equilibrium with Falling Temperature
R. M. DAWSON
Benzodiazepine receptors of rat cerebellum were assayed with [3Hl-labeled flunitrazepam at 37”C, and assays were terminated by filtration in a cold room according to one of three protocols: keeping each sample at 37°C until ready for filtration, taking the batch of samples (30) into the cold room and filtering sequentially in the order I-30, and taking the batch of 30 samples into the cold room and filtering sequentially in the order 30-I. The results for each protocol
were substantially different from each other, indicating that rapid disruption of equilibrium occurred as the samples cooled in the cold room while waiting to be filtered. Positive or negative cooperativity of binding was apparent, and misleading effects of y-aminobutyric acid on the affinity of diazepam were observed, unless
was kept at 37°C until just prior
Key Words: Benzodiazepine receptor; azepam; Temperature; Artifacts
Filtration assay; [3H]Flunitrazepam;
INTRODUCTION In assays of membrane-bound neurotransmitter receptors by radioligand-binding techniques, isolation of the receptor-ligand complex at the end of the assay is often achieved by filtration. It is necessary to rinse the filter with buffer to reduce the filter blank, but in so doing steps must be taken to minimize dissociation of the radiolabeled ligand from the receptor. These include rinsing as quickly as possible and using ice-cold buffer, since the rate of dissociation is less at lower temperatures. Consequently, it has been common practice in these and other laboratories (Freeman et al., 1985; Summers, 1980) to perform filtration in a cold room at 2-3”C, using a rinsing buffer that has been stored in the cold room. Thus assay tubes (up to 30 at a time) are incubated for a given time at 25 or 37”C, and the rack of samples is then transferred to the cold room for sequential filtration, a process that takes 78 min (Dawson, 1984). Valdes et al. (1985) and Mattera et al. (1985) used a similar technique in recent publications. The assumption is made that, during this brief time in the cold room, no significant change takes place to the receptor-ligand equilibrium due to decreasing temperature. The amount of radiolabeled ligand From the Defence Science and Technology Organization, Materials Research Laboratories, Ascot Vale, Victoria, Australia. Address reprint requests to: Defence Science and Technology Organization, Materials Research Laboratories, P.O. Box 50, Ascot Vale, Victoria 3032, Australia. Received October 3, 1985; revised and accepted March 7, 1986. 349 Journalof
0 1986 Elsevier Science Publishing
Co., Inc., 52 Vanderbilt
New York, NY 10017
R. M. Dawson
bound should neither decrease significantly, since dissociation is slowing down, nor increase, since for most receptors, the affinity of the ligand for the receptor does not increase with decreasing temperature. The assumption would not necessarily be valid, however, in the case of benzodiazepine receptors. Although both association and dissociation rates decrease with decreasing temperature (Speth et al., 1979), suggesting little change in equilibrium over the short term, the affinities of benzodiazepines for the receptor are considerably higher at lower temperatures (Braestrup and Squires, 1977; Speth et al., 1979). A noticeable increase in binding might, therefore, occur as the later samples in the batch sit in the cold room, depending on the rate of decrease of temperature and the rate of change of the receptor-ligand equilibrium. The experiments below were, therefore, designed to assess the extent of disturbance of equilibrium of the binding of a nonsaturating concentration of [3H]-labeled flunitrazepam ([3H]FNP) to rat brain benzodiazepine receptors which would be caused by transferring samples from a 37°C bath to a cold room for sequential filtration. Most studies of benzodiazepine receptor binding have not needed to consider this question, since such assays have usually been done at a temperature near 0°C to maximize binding (e.g., Braestrup et al., 1984). The results of the present study indicate that the above technique has a profound effect on the results for inhibition of L3HlFNP binding by diazepam and the effect of y-aminobutyric acid (GABA) on the inhibition curve. GABA is known to enhance the affinity of benzodiazepine receptor agonists for the receptor (Braestrup et al., 1984). METHODS [N-methy/-3H]Flunitrazepam was obtained from Amersham, Australia and New England Nuclear, with specific activities of 84 and 92.3 Ci/mmol, respectively. Similar results were obtained with the ligand from either source. The source of benzodiazepine receptor was rat cerebellum; two such cerebella (combined mass, 0.5-0.6 g) were homogenized in 10.0 ml Krebs phosphate buffer, pH 7.4 (Ultra-Turrax, setting 6.5, IO set) and washed twice by centrifugation at 17OOOg/20min. Assays were performed in the same buffer (1.0 ml) in Wasserman 5-ml serology tubes in batches of 30 using 0.1 ml homogenate in most cases, and [3H]FNP at a final concentration of 0.6 nM, which is well below a saturating concentration at 37°C (Speth et al., 1979). Nonspecific binding was evaluated using 3 PM diazepam. Assay tubes were prepared as follows: tubes l-3, total binding; tubes 4-6, nonspecific binding; tubes 7-15, diazepam (Roche), I-250 nM in increasing concentration; tubes 16-30, as for 1-15 but with the addition of IO ~.LMGABA. The tubes were incubated at 37°C for 60 min and then filtered sequentially in the order I-30 or 30-I by applying to prewetted GF/B glass microfiber filters (Whatman) under vacuum on a filtration manifold (Summers, 1980). Each tube was rinsed with 3-4 ml cold buffer, which was added to the filter, and the filter was rinsed with 3 x 5 ml cold buffer. The filtering and rinsing took place in a cold room at 2-4°C in all cases. Each tube took 0.2-0.3 min to process (Dawson, 1984). The filters were left on the filtration manifold under vacuum for a further 5-10 min, and then placed in scintillation vials. Then, 10.0 ml
toluene-0.4%, 2,5-diphenyloxazole was added to each vial, and the next day the samples were counted in a Beckman LS 5801 liquid scintillation system. Approximately 95% of the radioactivity was found free in the scintillant after overnight incubation (Dawson, 1984). Three methods of filtration were employed (all at 2-4°C) as follows: 1. Each sample was kept at 37°C until ready for filtration, i.e., all samples were filtered within 0.3 min of removal from the 37°C bath. Samples were filtered in the order I-30, although essentially the same results were obtained if filtration was in the reverse order. The contents of tubes I-30 were described previously. 2. The rack of samples was taken into the cold room and then filtered in the order I-30. 3. As for method 2, but samples were filtered in the order 30-I. For methods 2 and 3, the last samples to be filtered were out of the 37°C bath and in the cold room for 7-8 min before filtration, in contrast to method 1. RESULTS The results of the three methods of filtration (described above) are shown in Figure 1, where open symbols represent assays in the absence of GABA (tubes l15) and filled symbols represent assays in the presence of IO PM GABA (tubes 1630). Methods 1, 2, and 3 are represented by circles, triangles, and squares, respectively. Depending on the method of filtration, the results show substantial variation for assays in the absence of diazepam or in the presence of low concentrations of diazepam. Method 1 gives the expected result: progressive inhibition of specific binding with increasing concentration of diazepam and an enhancement of uninhibited specific binding by GABA (nearly twofold). By contrast, method 2 demonstrates apparent enhancement of binding by low concentrations of diazepam. This can be attributed to rapid disturbance of the binding equilibrium due to falling temperature (only 2-3 min after removing the rack of samples from the 37°C bath), and the increased binding of [3HlFNP has outweighed the small inhibitory effect of diazepam. Similarly, uninhibited specific binding in the absence of CABA is over three times as high for method 3 than it is for method 1; this is because the tubes measuring uninhibited binding are the last to be filtered with method 3. Further, the enhancement of binding by CABA is highest for method 2, in which the GABA samples have most opportunity to cool before filtration, and lowest for method 3, in which the GABA samples are filtered before the control samples. At higher concentrations of diazepam, corresponding to higher receptor occupancy by diazepam, the fall in temperature has little effect on the binding of r3H]FNP (Figure I). The artifactual binding curves resulting from methods 2 and 3 are reflected in Hill plots of the data of Figure 1, i.e., in plots of log Y/(1 - Y) versus log I where Y is the ratio of 13HlFNP bound in the presence of inhibitor of concentration I to that in its absence. The slope of such a plot is nH, the Hill coefficient, while Is0 is the concentration of inhibitor which occupies half of the receptors and for which log Y/(1 - Y) = 0 (Fields et al., 1978). Table 1 lists nH and Iso for Hill plots of the data
R. M. Dawson
8 DilaOLepam (MI
1. Specific binding of [3Hl-labeled fiunitrazepam to rat cerebellum benzodiazepine receptors as a function of the concentration of added diazepam. Open symbols, assays in absence of CABA (assay tubes 1-15); closed symbols, assays in presence of 10 uM CABA (assay tubes 16-30); (O,O), method 1 (samples kept at 37°C until filtration); (A, A), method 2 (filtration in cold room in order l-30); (Cl, n), method 3 (filtration in cold room in order 30-l). In each case, tubes 7-15 and 22-30 contain diazepam in increasing concentration. Each point is the mean of four determinations; the relative standard error was <5% in most cases and never exceeded 14%. The r3H]FNP bound is represented as the percentage of the value for uninhibited specific binding according to method 1.
of Figure 1. It can be seen that method 2 is characterised by nH > 1 in the presence or absence of CABA, indicating positive cooperativity. By contrast, the results for method 3 suggest negative cooperativity, since nH < 1. The preferred method 1 gives nH close to unity in the absence or presence of GABA, indicating a lack of cooperativity. Table 1 also shows the profound effect of the method of filtration on
Benzodiazepine TABLE 1
Values of nH and I50 for Hill Plotfb of the Data of Figure 1” - GABA
0.99 2 0.012
1.40 k 0.102
0.77 k 0.082
+fO PM CABA 150
78 k 7.4
101 t 3.3 17 + 2.2
1.11 i 0.031
40 +- 0.7
1.96 2 0.200
1.24 i 0.042 0.79 t 0.025
26 2 1.3 11 + 0.5
3.90 2 0.159 1.46 + 0.138
a Results are given as means 2 SEM. The concentration of L3HlFNP was 0.57-0.66 nM. b The Hill plot is a plot of log (Y)/(l - Y) versus log I where Y is the ratio of 13HlFNP bound in the presence of inhibitor (diazepam) of concentration I to that in its absence. The slope of such a plot is nH, and Iso is that value of I at which log (Y)/(l - Y) = 0 (Fields et al., 1978). c GABA shift = IQ, (- GABA)&, (i GABAI.
the Is0 values, and on the extent to which GABA increases the affinity of diazepam for the receptor.
DISCUSSION The results indicate that the benzodiazepine receptor-ligand equilibrium at 37°C shifts noticeably toward increased ligand binding within minutes of transferring samples to a cooler atmosphere, despite the fact that both association and dissociation rates are slower at lower temperatures, i.e., attainment of equilibrium should be less rapid than at 37X All the results shown in Figure 1 and Table 1 are consistent with this immediate shift of the equilibrium, including the relative insensitivi~ of binding to variations in the method of filtration when high concentrations of diazepam are present. Thus, binding of a benzodiazepine to the receptor would not be expected to change with temperature when binding to the receptor is already approaching saturation. If benzodiazepine receptor assays are to be performed at a physiological temperature, it is therefore essential to minimize the time between removing a sample from the incubation bath and separating free from bound ligand by filtration. The method of Valdes et al. (1985) for neuroleptic receptors and the standard method in these and other laboratories for cholinergic and adrenergic receptors (Freeman et al., 1985; Summers, 1980) are inappropriate in such cases. The artifacts described in the present paper may well occur also in other ligand-binding systems, and there is, therefore, a need to keep such a possibility in mind when planning ligand-binding experiments, particularly with novel ligands and/or receptors.
REFERENCES Braestrup C, Squires RF (1977) Specific benzodiazepine receptors in rat brain characterized by high-affinity [3H]diazepam binding. Proc Nat/ Acad Sci USA 74:3805-3809.
Braestrup C, Honore T, Nielsen M, Petersen EN, Jensen l_H (1984) Ligands for benzodiazepine receptors with positive and negative efficacy. Biochem Pharmacol33:859-862.
R. M. Dawson Dawson RM (1984) The affinity of L3Hlquinuclidinyl benzilate and some other radioligands for glass microfiber and cellulose filters in some common liquid scintillation solvents. Anal Biochem 139:493-501. Fields JZ, Roeske WR, Morkin E, Yamamura HI (1978) Cardiac muscarinic cholinergic receptors. J Biol Chem 253:3251-3258. freeman SE, Dawson RM, Culvenor A], Keeghan AM (1985) Interactions of amantadine with the cardiac muscarinic receptor. / MO/ Cell Car&o/ 17:9-21. Mattera R, Pitts BJR, Entman ML, Birnbaumer L (1985) Guanine nucleotide regulation of a mam-
malian myocardial muscarinic receptor system / Biol Chem 206:7410-7421. Speth RC, Wastek GJ, Yamamura HI (1979) Benzodiazepine receptors: temperature dependence of [3H]flunitrazepam binding. Life Sci 24:351-358. Summers RJ (1980) f3HlClonidine binding to n-adrenoceptors in membranes prepared from regions of guinea pig kidney: Alterations by monovalent and divalent cations. Br) Pharmaco/71:5763. Valdes JJ, Chester NA, Menking D, Shih T-M, Whalley C (1985) Regional sensitivity of neuroleptic receptors to subacute soman intoxication. Brain Res Bull 14:117-121.