PsychiatryResearch. 33: 19-29
19
Elsevier
Methodological Issues Platelet 3H-lmipramine James A. Severson,
in the Preparation Binding
Lon S. Schneider,
and Assay
of
and Eric R. Fredrickson
Received May 26, 1989; first revised version received September version received March 19, 1990; accepted April 8, 1990.
1.5, 1989; second revised
Abstract. Several methodological factors in the preparation of platelets and the determination of platelet 3H-imipramine (3H-IMI) binding were examined. The ionic composition of the assay significantly affected platelet 3H-IMI binding. Approximately 25% of the specific binding of 3H-IMI to intact platelet preparations was retained in the absence of sodium and chloride ions. The addition of sodium ions enhanced the specific binding of 3H-IMI, but the addition of chloride in the presence of sodium had a more pronounced effect, enhancing binding approximately five-fold over that observed with the addition of sodium. Sodium was the only cation tested that enhanced binding. Only halides enhanced binding in the presence of sodium with the following order of potency: Cl- > Br- > I- = F-. Ions increased the density of binding sites (B,,,,,) and did not affect the affinity of the binding sites for 3H-IMI. In the presence of sodium and chloride, the use of serotonin (5HT) to define nonspecific binding in saturation experiments resulted in lower binding densities (B,,,) than when desipramine was used to define nonspecific binding. The component of binding that was insensitive to 5HT was roughly equal to the B,, of 3H-IMI binding obtained in the absence of sodium and chloride using desipramine to define nonspecific binding. Overall, these data suggest that not all 3H-IMI binding that is displaced by desipramine is related to serotonergic mechanisms, and suggest that 5HT is a better choice than desipramine for the determination of the nonspecific binding of 3H-IMI. In addition, the binding of 3H-IMI to different platelet preparations was compared. The binding of 3H-IMI to intact platelets was less than that obtained using lysed platelet membranes when data were expressed per mg protein. The Coomassie Blue dye-binding method to determine platelet protein resulted in greater B,,,,, values than were obtained with the Folin phenol reagent method. The method of platelet preparation that is commonly used to prepare platelets for 3H-IMI binding resulted in similar binding values when compared to a method that prepares the entire platelet population. The results suggest that some, but not all, variations in laboratory methods used to prepare platelets and assay for platelet 3H-IMI binding may affect clinical studies examining this measure. Key Words.
Imipramine
binding,
platelets,
ion-dependence,
James A. Severson, Ph.D., is Manager,
receptor
binding.
Pharmaceutical Market Development, New Market Development, Amersham Corporation, Arlington Heights, IL. Lon S. Schneider, M.D., is Assistant Professor of Psychiatry, University of Southern California School of Medicine, Los Angeles, CA. Eric R. Fredrickson, B.S., was formerly Research Laboratory Technician, USC School of Medicine, and is now at the University of Kansas, School of Medicine, Kansas City, KS 66103. (Reprint requests to Dr. L.S. Schneider, Dept. of Psychiatry and the Behavioral Sciences, 1934 Hospital PI., Los Angeles, CA 90033, USA.) 0165-1781/90/$03.50
@ 1990 Elsevier Scientific
Publishers
Ireland
Ltd.
20
The binding of 3H-imipramine (3H-IMI) to platelets has received considerable attention as a potential biological marker for depression. In populations of clinically depressed individuals, the average density of the binding sites for 3H-IMI on platelets is reduced relative to populations of individuals who are not judged to be clinically depressed (reviewed by Langer et al., 1987; Beth et al., 1988). However, clinical conditions that differ from.depression also may have 3H-IMI binding that is reduced compared to control populations. For instance, adults with agoraphobia (Lewis et al., 1985), patients with Parkinson’s disease (Schneider et al., 1988a), patients with agitated Alzheimer’s disease (Schneider et al., 1988b), and children with nocturnal enuresis (Weizman et al., 1985) are reported to have reduced platelet JH-IMI binding. The usefulness of platelet 3H-IMI binding as a marker for depression is further complicated by the fact that not all laboratories that have studied 3H-IMI binding to platelets in depression have observed a reduction in the number of binding sites (Mellerup et al., 1982; Gentsch et al., 1985; Tang and Morris, 1985; Muscettola et al., 1986; Georgotas et al., 1987; Kanof et al., 1987). In addition to concern over the lack of consistency of findings in clinical depression is the seemingly wide variation in the values for the density of 3H-IMI binding sites from various laboratories. This suggests that laboratory-to-laboratory variation in techniques associated with JH-IMI binding may contribute to the variation in binding density reported in the literature, and also may contribute to the lack of consistency in findings of 3H-IMI binding in depression. The present report examines the effect of the method used to obtain platelets for 3H-IMI binding, the use of intact vs. lysed platelets, and the assay for platelet protein. In addition, it examines components of 3H-IMI binding that are regulated by ions and are sensitive and insensitive to displacement by serotonin (5-hydroxytryptamine, 5HT). Methods Blood Collection and Platelet Preparation. Whole blood was obtained from healthy volunteers who ranged in age from 25 to 35 years. Blood was collected by venipuncture into vacutainers containing EDTA and was allowed to sit for 1 to 2 hours at room temperature. Platelet-rich plasma was obtained by low-speed centrifugation (8Og, room temperature, 1 hour), and was then pipetted into a clean polypropylene centrifuge tube using a clean polypropylene Pasteur pipette and recentrifuged (21,OOOg, 4 “C, 5 min). The plasma supernatant was discarded and the platelet pellet stored at -70 ‘C. Platelets obtained from 10 ml of whole blood were thawed over ice in 2 ml of buffer I (50 mM Tris-HCl, pH 7.4, containing 150 mM NaCl and 20 mM EDTA, disodium salt) and were dispersed by gentle uptake and expulsion from a plastic Pasteur pipette. The resuspended pellet was centrifuged at (2 1,OOOg,4 o C, 5 min). The supernatant was discarded and the pellet resuspended in 2 ml of buffer I and recentrifuged. After the second wash in buffer I, the platelet pellet was resuspended in 2.0 ml of incubation buffer (50 mM Tris-HCl, pH 7.4, 120 mM NaCl, 5 mM KCl) and recentrifuged (4 ‘C, 21,OOOg, 5 min). The final platelet pellet was prepared for binding assays by resuspension with a plastic Pasteur pipette in 4.0 ml of cold incubation buffer. Platelets were prepared for ion-sensitive binding as described above with the exception that 50 mM Tris-acetate buffer (PH 7.4) without salts was substituted for incubation buffer in all washing and resuspensions.
21 lmipramine Binding Assay. 3H-IMI (73 Ci/mmole) was diluted in incubation buffer and was added to glass test tubes in 50 ,ul in duplicate. In binding saturation experiments, six 3H-IMI concentrations (0.1258 nkf) were used, and in single 3H-IMI concentration experiments, tubes contained 2 nM3H-IMI. Assay tubes received either desipramine (Sigma) to determine nonspecific binding (100 PM final concentration in 100 ,ul of incubation buffer) or 100 ,ul of incubation buffer for total binding. Platelet suspensions (100 ,~l) were added last for a final assay volume of 250 /11. For ion-sensitive binding, ions were added separately to assay tubes in 50~1 of Tris-acetate. Desipramine (100 PM assay concentration) and JH-IMI were diluted in Tris-acetate @H 7.4) and each was added in 50 ~1. Assays were initiated by the addition of 100 ~1 of platelets in Tris-acetate buffer. All 3H-IMI binding experiments were conducted at 20-40 /lg of protein (Bradford, 1976) per tube. Assay tubes were incubated at O-2 “C for 60 min and the assay was stopped by dilution with 5 ml of wash buffer (50 mM Tris-HCl, pH 7.4, 120 mkf NaCl, 5 mM KCl) and collection of platelets on glass fiber filters (S&S No. 32, Schleicher and Schuell, Keene, NH). Tubes were rinsed with an additional 5 ml of wash buffer and filtered. Filters were then washed twice with 5 ml of wash buffer. The radioactivity retained on each filter was determined by liquid scintillation spectroscopy. Protein Assays. The protein content of platelet preparations was determined using proteindye binding (Bradford, 1976). In some experiments, the Folin-phenol method (Lowry et al., 1951) also was used for comparison. Data Analysis. Binding isotherms were transformed to the relationship of bound/free vs. bound (Scatchard, 1949) and the binding capacity (B,,,,,) and equilibrium dissociation constant (&) were determined by least squares linear regression. JH-IMI displacement curves were analyzed using a four-parameter logistic model (ALLFIT; DeLean et al., 1978). The concentration of drug that displaced 50% of the specifically bound 3H-IMI (IC,,) obtained from this analysis was corrected for the concentration of JH-IMI (Cheng and Prusoff, 1973) for presentation as the inhibition constant (K,).
Results Ion-Sensitive Binding of lmipramine to Human Platelets. Platelets were prepared in Tris-acetate and assayed in duplicate in 2 nM 3H-IMI for total and nonspecific binding (100 PM desipramine). Sodium was added as NaCl at a final concentration of 300 mM. Approximately 75% of the 3H-IMI displaced by desipramine was sensitive to the addition of NaCl (mean + SD: with NaCl = 956 f 244 fmol/mg protein; without NaCl = 239 f 100; difference = 7 17 + 224, 16 paired experiments, t = 12.80, df = 15, p < 0.001). Determination of the Specificity of Cations. Platelets were prepared in Trisacetate and were tested for cationic dependency of desipraminedisplaceable 3H-IMI binding. Cations were added as 300 mM of the chloride salt. The only cation that was found to enhance the binding of 3H-IMI to platelets was sodium (Table 1); the addition of sodium as either NaCl or sodium acetate elevated binding. The effect of NaCl was greater than that of sodium acetate because of the additional effect of chloride on binding (see below). However, some chloride salts depressed the apparent specific binding of 3H-IMI.
22 Table 1. Effect of addition of ions on the binding of W-imipramine to platelets
Cations
Desipramine-displaceable bindlng (fmol/mg protein)
Anions
Desipramine-displaceable binding (fmoUmg protein)
None
194 j,
42
(9)
None
275 f
LiCl
150f
51
(6)
NaBr
736f144’(3)
KCI
126f
51
(6)
Nal
499 f 147’ (3)
NaCl
671 f 206’
(6)
NaF
459 f 110’ (6)
66 (9)
MgClz CaCb
59f
23’
(3)
NaHC03
323 f
57’ (3)
41 *
29’
(3)
KzHPO,
166f
55 (3)
ZnCh
47f
19’
(3)
Li2S04
195f
54 (3)
NaCl
926 f 240’ 16)
NaOAc
296 f
62’ (12)
Note. Platelets were prepared in 50 mM Tris-acetate, and Y-Limipramine binding was determined in duplicate for total and nonspeciffc binding using 2 mM 3H-imipramine and 100 @f desipramine to determine nonspecific binding. Cations were added at a concentration of 300 mM, and anions were added at a concentration of 150 mM in the presence of 300 mM sodium acetate to fulfill the requirement for sodium ion. Data are mean f SD of the number of determinations in parentheses
1.p < 0.05 vs. “none” in each group. Determination of the Specificity of Anions. Platelets were prepared in Trisacetate @H 7.4) and were tested for the anionic dependency of 3H-IMI binding in the presence of 300 mM sodium acetate because sodium was necessary to obtain the anion effect on binding (compare NaCl addition with the addition of sodium acetate in Table 1). Halides were the only anions that enhanced the binding of 3H-IMI in the presence of saturating amounts of sodium (Table 1). The order of effect of the halide salts was Cl- > Br- > I- = F-. Displacement of Platelet lmipramine Binding. In the presence of NaCl, displacement of 3H-IMI from platelets followed the pattern expected; displacement by desipramine was steep with a Ki of 43 nA4. 5HT was a much weaker displacing agent than desipramine with a K, of 1400 nM (Fig. 1). In the absence of NaCl, desipramine and 5HT were much weaker displacers. Desipramine was an approximately IO-fold weaker displacer (Ki = 513 nM) in the absence of NaCl, and 5HT did not significantly displace 3H-IMI in the absence of NaCl (Fig. 1). Dose Response to Addition of Ions. The addition of sodium ions as sodium acetate resulted in the dose-dependent elevation of 3H-IMI binding (Fig. 2A). Sodium ion increased desipramine-displaceable binding to approximately 150 fmol/mg protein above that obtained in the absence of ions. The concentration of sodium that elevated the binding by 50% was approximately 50 mM. In the presence of 300 mM sodium (as sodium acetate), the addition of chloride ion as NaCl enhanced the desipramine-displaceable binding of 3H-IMI to platelets to 800 fmol/ mg protein above that obtained with 300 mM sodium acetate alone (Fig. 2B). The concentration of chloride ion that elevated binding by 50% was approximately 30 mM.
23 Fig. 1. Displacement
of 3H-imipramine
from platelets
in the presence
and
0 10-S
10-8
lo*
10-7
10-S
104
CONCENTRATION Platelets were prepared in Tris-acetate for sodium&per&m binding and 11 concentrations each of desipramine (DMI) and serotonin (SHT) in the presence and absence of 300 mM sodium chloride were used to determine the effect of sodium on the concentration dependence of the displacement of %imipramine binding. Inhibition of JH-imipramine binding was anatyzed by iterative nonlinear regression using a 4-parameter logistic program and the ICY, corrected for the concentration of 3H-imipramine. Data are means of 3 independent experiments.
Fig. 2. Concentration dependence of the enhancement imipramine binding by sodium and chloride
A.
Na+ DEPENDENCE
B.
of platelet
3H-
Cl - DEPENDENCE
(3
~a+ CONC~RATION
(m.4)
Cl- CONCENTRATION
(mM)
Platelets were prepared in Tris-acetate and 3H-imipramine binding (2 nM) was determined in duplicate for total and nonspecific binding (100flM desipramine). (A) Sodium dependence was examined using increasing concentrations of sodium acetate (O-600 mM). (B) Chloride-dependent binding was examined using increasing concentrations of sodium chloride (O-400 mM) in the presence of 300 mM sodium acetate. Data are mean f SEM of the specific binding from 4 independent dose-response curves.
Effect of Ions and 5HT on Maximal Binding and Equilibrium Dissociation Constants. The influence of ions on 3H-IMI binding isotherms was tested. Platelets were prepared in Tris-acetate and the effect of ions on desipramine-displaceable binding was examined in the absence of ions, in the presence of sodium (as sodium acetate), and with sodium and chloride. In the absence of sodium and chloride, 3H-IMI binding was saturable with an equilibrium dissociation constant of 1.8 nM and a binding capacity of 373 fmol/mg protein (Table 2). The addition of sodium
24
Table 2. Effect of addition of ions and use of serotonin as a displacer on the bindina of 3H-imioramine to olatelets
Ion dependence iris-acetate
373 f
94
1.82 * 0.44
Tris + Na+
532 LII 74’
2.18 f 0.66
Tris + Na+ + Cl-
1964+
1382
1.86 zt 0.48
94
1.89 rt 0.92
Serotonin displacement Desipraminefor nonspecific binding Tris-acetate
373 It
Tris + NaCl
1847 + 1983
1.64 zk 0.38
1449 f 228“
1.91 f 0.54
Serotonin for nonspecific binding Tris + NaCl Serotonin-insensitive
bindinn
398f
116
1.76 k 0.48
Note. Binding isotherms were determined in duplicate for total and nonspecific binding at 6 concentrations of 3H-imipramine (0.125-E nU). Platelets were prepared in Tris-acetate and ion-dependent binding was conducted in the absence of ions (Tris-acetate), with 300 mM sodium acetate (Tris + Na’), and 300 mM sodium acetate and 150 mM NaCl (Tris t Na’ t Cl-). Nonspecific binding was determined using 100 &f desipramine (DMI). Serotonin displacement was conducted in the absence of ions (Tris only) and with 300 mM NaCl (Tris t NaCI). Nonspecific binding was determined using 100 &f serotonin and 100 &f DMI. Serotonin-insensitive binding was obtained by subtracting binding in the presence of 100fiM DMI from the binding in the presence of 1OOpM serotonin. Data are mean f SD of 4 independent experiments.
1.p < 0.05 vs. Tris-only. 2. p < 0.001 vs. Tris-only; p < 0.001 vs. Tris + Na. 3. p < 0.001 vs. Tris-only - DMI. 4. p < 0.05 vs. Tris + NaCl - DMI; p < 0.001 vs. serotonin-insensitive binding
increased the capacity of binding by approximately 40% with little change in the Kd. The addition of chloride in the presence of sodium more dramatically affected binding capacity; the B,,, for 3H-IMI binding was increased over 265% by addition of chloride, and by 525% over no ions. The Kd was not affected by the addition of ions. In addition, 5HT was tested as a displacer for saturable 3H-IMI binding. Binding assays were constructed so that binding isotherms were conducted in the presence and absence of Na and Cl using desipramine as the displacer, and in the presence of Na and Cl using 5HT as the displacer. Thus, it was possible by subtraction (tubes with 5HT added minus tubes with desipramine added) to obtain an isotherm for SHT-insensitive binding. In the absence of ions, the binding of 3H-IMI displaced by desipramine was saturable with a capacity of 373 fmol/mg protein (Table 2). The addition of sodium and chloride enhanced the binding of 3H-IMI to platelets almost five-fold. Using 5HT to define nonspecific binding in the presence of Na and Cl resulted in a Bmax that was 22% lower than that observed when desipramine was used to define nonspecific binding. The Kd for saturable JH-IMI binding was similar in assays in which 5HT or desipramine were used to define nonspecific binding. SHT-insensitive binding was not different from desipramine in the absence of Na and Cl.
25 of lmipramine Binding to Intact and Lysed Platelets. Platelets were prepared in buffer containing NaCl and were then lysed using the method of Raisman et al. (1981) for comparison with the binding to intact platelets using both the Bradford and Lowry protein assays to determine the protein content of platelet preparations. The saturable binding of 3H-IMI to lysed platelet preparations was 65% higher than the binding to intact platelets (Table 3). In addition, there was a significant increase in the Kd (lower binding affinity) of the binding of 3H-IMI to lysed platelets compared to the binding to intact platelets. The use of Lowry protein to express binding data resulted in a reduction in the apparent density of 3H-IMI binding. Overall, expression of B,,,,, data using Lowry protein reduced the B,,,,, by 29.9 f 12.7% (mean f SD, n = 24). Comparison
Table 3. Comparison preparations
of 3H-imSpramine binding
to intact and lysed platelet
Bradford protein
Lowry protein
3161 f 1166
2000 + 462’
1.69 f 0.73
1.69 zt 0.73
1910+55a2
1345 f 208’
1.18 + 0.172
1.18 f0.17
LY--J B,,,,, (fmol/mg protein)
Kd (nM) Intact B,,, &
(fmollmg protein) (nM)
Note. Platelet-rich plasma was prepared in Tris plus NaCl and KCI. The plasma was divided into equal volumes and one portion was processed to prepare intact platelets (see Methods) and the other portion was lysed (Raisman et al., 1961). Binding isotherms were conducted in duplicate at 6 concentrations of 3H-imipramine (0.1254 nA4) using 100 PM desipramine to determine nonspecific binding. Protein content of the intact and lysed platelet preparations was determined using both Coomassie Blue dye binding (Bradford, 1976) and the Folin phenol reagent (Lowry et al., 1951). Data are mean + SD of 12 individual experiments. 1. p < 0.01 vs. B,, using Bradford protein. 2. p < 0.01 vs. similar binding parameter from lysed platelets.
Comparison of Methods of Platelet Preparation. Platelet-rich plasma was processed to prepare the entire platelet population according to the method of Corash (1980) for comparison to the procedure described in Methods. In five parallel experiments, the B,,, of the binding of 3H-IMI to intact platelets was similar between the whole platelet population and the method used in this laboratory (Table 4). Discussion
Platelet 3H-IMI binding is regulated by sodium ions and anions: Na was the only cation tested that enhanced 3H-IMI binding and Cl was the most effective anion tested. Only halides enhanced the binding of 3H-IMI, and the order of halide specificity was similar to that observed in brain: Cl- > Br- > I- = F-. The effects of ions were shown to be concentration-dependent and to have a major effect on the measured density of 3H-IMI binding sites rather than to affect the affinity of the binding site for 3H-IMI. Furthermore, the effect of halides requires the presence of
26
Table 4. Comparison of methods of platelet DreDaration 3H-imipramine binding
Methods Single spin collection B,.,
(fmol/mg
protein)
Kd W)
1994 f
561
1.42 f
0.54
1869 f
561
Whole platelet population B max (fmol/mg
Kd WI
protein)
1.39 + 0.54
Note. Platelets were prepared from platelet-rich plasma by a one-step isolation procedure (see Methods) or by a method that yields the whole platelet population (Corash, 1980). Binding isotherms were constructed at 6 concentrations of 3H-imipramine (0.125-E nM) using 100 jAf desipramine to determine nonspecific binding. Platelet protein content was determined using protein-dye binding (Bradford, 1976). Data are mean + SD of 5 assays.
sodium; chloride salts of cations other than sodium were ineffective in increasing the density of 3H-IMI binding sites. In addition, desipramine did not distinguish between ion-sensitive and ion-insensitive binding, whereas 5HT did distinguish between these two populations of binding sites. The results obtained in this study for the ion-dependency of platelet 3H-IMI binding are similar to those reported previously for brain (Severson et al., 1986; Backstrom and Marcusson, 1987). In brain, ion-sensitive binding was found to be associated with 5HT neurons and SHT uptake. However, the proportion of ioninsensitive binding sites on platelets is less than in brain; platelet ion-insensitive binding of 3H-IMI represents between 10 and 25% of the desipramine-displaceable binding (Marcusson and Tiger, 1988; Hrdina, 1989; this study), whereas in brain approximately 50% of the binding is insensitive to ions. In addition, the use of 5HT to define nonspecific binding resulted in a binding density that was 78% of that obtained when desipramine was used. This is similar to the 75% of 3H-IMI binding that was found to be ion-sensitive. Together, these two approaches show that 25% of the specific binding of 3H-IMI to platelets is probably not related to 5HT uptake and, thus, represents sites extraneous to the measurements desired. Others have considered this component as part of the specific binding of 3H-IMI (Talvenheimo et al., 1979, 1983; Abbott et al., 1982; Rehavi et al., 1982). These data have immediate practical implications for studies of 3H-IMI to platelets. The data presented here clearly demonstrate that platelet JH-IMI binding is not absolutely dependent upon the presence of sodium ion. The present data suggest that 5HT is a better choice for the determination of nonspecific binding in studies of platelet 3H-IMI binding. It has been suggested that the binding of 3H-IMI to intact platelets is more consistent than binding to lysed platelet membranes because of inconsistencies in the degree of lysis of platelets (Fried1 et al., 1983). Our early preliminary studies confirmed these observations, and we have chosen to conduct platelet 3H-IMI
27 binding to intact platelets rather than the more commonly used lysed platelet membrane preparation. Several of our earlier studies using intact platelet preparations confirmed findings of decreased platelet 3H-IMI binding in depressed patients and extended these observations to elderly depressed patients (Schneider et al., 1985, 1986, 1988c), patients with Parkinson’s disease (Schneider et al., 1988~) and Alzheimer’s disease patients with agitated behavior (Schneider et al., 19886). Our studies of depression in the elderly have been confirmed by others using lysed platelet preparations (Nemeroff et al., 1988). These data suggest that not all variations in 3H-IMI binding may have a detrimental impact on the ability of the assay to detect significant population differences in platelet 3H-IMI binding. Studies presented here show that the density of 3H-IMI binding sites on an mg protein basis is greater in lysed platelets. This is most likely due to the removal of soluble proteins from the sample upon lysis and washing: 40% of platelet protein content is in the soluble fraction (Barber and Jamieson, 1970). For comparison purposes, in this study we also expressed binding densities per mg protein determined by the Lowry method (Lowry et al., 1951). We found that the Bradford method (Bradford, 1976) detects about 70% of the protein detected by the Lowry method in platelets. This contrasts with a 50% Bradford-to-Lowry ratio found in mouse brain (Severson et al., 1986). These differences presumably reflect differences in the mechanisms that the two methods use to detect protein. The Bradford method uses dye binding to arginine residues (Compton and Jones, 1985) whereas the Lowry method uses in part a reduction of tyrosine, tryptophan, and cysteine residues with a reaction of copper-protein complexes with the Folin reagent (Legler et al., 1985). The Lowry method also is sensitive to the amino acid composition of proteins (Lowry et al., 195 1; Legler et al., 1985). A result of the use of the Coomassie Blue method is that it increases binding densities by reducing the denominator used to express binding. However, the Bradford assay is less subject to interference by components of biological buffering systems than the Lowry method (Peterson, 1979). In addition, the system that we use for 3H-IMI binding uses less protein than that used by others, 20 to 4Opg protein per assay tube. We have found that higher protein concentrations reduce the apparent binding of 3H-IMI in a manner that suggests ligand depletion (Severson et al., 1986). The protein concentration chosen for our studies is at a concentration that is well within the linear response range for the 3H-IMI binding assay. Differences in platelet 3H-IMI binding between laboratories also could reflect incomplete collection of the entire platelet population, which could result in assaying 3H-IMI binding to platelet populations that differ in the density of binding sites (Arora and Meltzer, 1984). It is possible to isolate platelet populations that differ by size and buoyant density (Corash et al., 1977; Corash, 1980). However, the present data suggest that the binding of 3H-IMI to intact platelets is similar whether determined by a method designed to collect the entire platelet population or by less exhaustive platelet isolation procedures (Raisman et al., 1981; Schneider et al., 1986). In conclusion, the choice of assay conditions can significantly affect the data obtained in platelet 3H-IMI binding; protein concentrations, protein assays, ion
28 concentrations, the choice of agent used to define nonspecific binding, and the use of intact or lysed platelets can affect observed binding densities. While this study has characterized potential problems in 3H-IMI binding studies in depression, it will be necessary to test some assay variables in platelets obtained from depressed subjects to define how assay variables can affect the results obtained, and to determine their impact on the inconsistent results that are presently in the literature. References Abbott, W.M.; Briley, MS.; Langer, S.Z.; and Sette, M. Sodium shift of the inhibition of [sH]-imipramine binding by SHT and SHT-uptake blockers but not by tricyclic antidepressants. British Journal of Pharmacology, 76:295P, 1982. Arora, R.C., and Meltzer, H.Y. Imipramine binding in subpopulations of normal human blood platelets. Biological Psychiatry, 19:257-262, 1984. Backstrom, I.T., and Marcusson, J.O. 5-Hydroxytryptamine-sensitive [jH]imipramine binding of protein nature in the human brain: I. Characteristics. Brain Research, 425: 128-136, 1987. Barber, A.J., and Jamieson, G.A. Isolation and characterization of plasma membranes from human platelets. Journal of Biological Chemistry, 2456357-6365, 1970. Beth, P.; Eplov, L.; Gastpar, M.; Gentsch, C.; Mendlewicz, J.; Plenge, P.; Rielaert, C.; and Mellerup, E.T. WHO pilot study on the validity of imipramine platelet receptor binding sites as a biological marker of endogenous depression: A preliminary report on the initial evaluation phase of a World Health Organization collaborative study. Pharmacopsychiatry, 21:147-150, 1988. Bradford, M.M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry, 72:248-254, 1976. Cheng, Y.-C., and Prusoff, W.H. Relationship between the inhibition constant (K,) and the concentration of inhibitor which causes 50 per cent inhibition (Iso) of an enzymatic reaction. Biochemical Pharmacology, 2213099-3108, 1973. Compton, S.J., and Jones, C.G. Mechanism of dye response and interference in the Bradford protein assay. Analytical Biochemistry, 151:369-374, 1985. Corash, L. Platelet heterogeneity: Relevance to the use of platelets to study psychiatric disorders. Schizophrenia Bulletin, 6~254-257, 1980. Corash, L.; Tan, H.; and Gralnick, H.R. Heterogeneity of human whole blood platelet subpopulations: I. Relationship between buoyant density, cell volume, and ultrastructure. Blood, 49:71-87, 1977. DeLean, A.; Munson, P.J.; and Rodbard, D. Simultaneous analysis of families of sigmoidal curves: Application to bioassay, radioligand assay and physiological dose-response curves. American Journal of Physiology, 235:E97-E102, 1978. Friedl, W.; Propping, P.; and Week, B. 3H-Imipramine binding in platelets: Influence of varying proportions of intact platelets in membrane preparations on binding. Psychopharmacology, 80:96-99, 1983. Gentsch, C.; Lichtsteiner, M.; Gastpar, M.; Gastpar, G.; and Feer, H. 3H-Imipramine binding sites in platelets of hospitalized psychiatric patients. Psychiatry Research, 14: 177-187, 1985. Georgotas, A.; Schweitzer, J.; McCue, R.E.; Armour, M.; and Friedhoff, A.J. Clinical and treatment effects on 3H-clonidine and 3H-imipramine binding in elderly depressed patients. Life Sciences, 40:2137-2143, 1987. P.D. Differences between sodium-dependent and desipramine-defined Hrdina, [rH]imipramine binding in intact human platelets. Biological Psychiatry, 25:576-584, 1989. Kanof, P.D.; Coccaro, E.F.; Johns, C.A.; Siever, L.J.; and Davis, K.L. Platelet [rH]imipramine binding in psychiatric disorders. Biological Psychiatry, 22:278-286, 1987.
29 Langer, S.Z.; Galzin, A.M.; Poirier, M.F.; Loo, H.; Sechter, D.; and Zarifian, E. Association of [jH]imipramine and [jH]paroxetine binding with the 5HT transporter in brain and platelets: Relevance to studies in depression. Journal of Receptor Research, 7:499-521, 1987. Legler, G.; Muller-Platz, C.M.; Mentges-Hettkamp, M.; Pflieger, G.; and Julich, E. On the chemical basis of the Lowry protein determination. Analytical Biochemistry, 150:278-287, 1985. Lewis, D.A.; Noyes, R., Jr.; Coryell, W.; and Clancy, J. Tritiated imipramine binding to platelets is decreased in patients with agoraphobia. Psychiatry Research, 16: l-9, 1985. Lowry, O.H.; Rosebrough, N.J.; Farr, A.L.; and Randall, R.J. Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry, 193:265-275, 1951. Marcusson, J., and Tiger, G. [JH]Imipramine binding of protein nature in human platelets: Inhibition by 5-hydroxytryptamine and 5-hydroxytryptamine uptake inhibitors. Journal of Neurochemistry, 50:1032-1036, 1988. Mellerup, E.T.; Plenge, P.; and Rosenberg, R. 3H-Imipramine binding sites in platelets from psychiatric patients. Psychiatry Research, 7:221-227, 1982. Muscettola, G.; Di Lauro, A.; and Giannini, C.P. Platelet 3H-imipramine binding in bipolar patients. Psychiatry Research, 18:343-353, 1986. Nemeroff, C.B.; Knight, D.L.; Krishnan, R.R.; Slotkin, T.A.; Bissette, G.; Melville, M.L.; and Blazer, D.G. Marked reduction in the number of platlet-tritiated imipramine binding sites in geriatric depression. Archives of General Psychiatry, 45:919-923, 1988. Peterson, G.L. Review of the Folin phenol protein quantitation method of Lowry, Rosebrough, Farr and Randall. Analytical Biochemistry, 100:201-220, 1979. Raisman, R.; Sechter, D.; Briley, MS.; Zarifian, E.; and Langer, S.Z. High affinity -‘Himipramine binding in platelets from treated and untreated depressed patients compared to healthy volunteers. Psychopharmacology, 75:368-37 1, 198 1. Rehavi, M.; Skolnick, P.; and Paul, S.M. Solubilization and partial purification of the high affinity [jH]imipramine binding site from human platelets. FEBS Letters, 150:5 14-5 18, 1982. Scatchard, G. The attractions of proteins for small molecules and ions. Annals of the New York Academy of Sciences, 5 1:660-672, 1949. Schneider, L.S.; Chui, H.C.; Severson, J.A.; and Sloane, R.B. Decreased platelet 3Himipramine binding in Parkinson’s disease. Biological Psychiatry, 24:348-351, 1988~. Schneider, L.S.; Fredrickson, E.R.; Severson, J.A.; and Sloane, R.B. 3H-Imipramine binding in depressed elderly: Relationship to family history and clinical response. Psychiatry Research, 19:257-266, 1986. Schneider, L.S.; Severson, J.A.; Chui, H.C.; Pollock, V.E.; Sloane, R.B.; and Fredrickson, E.R. Platelet tritiated imipramine binding and MAO activity in Alzheimer’s disease patients with agitation and delusions. Psychiatry Research, 25:31 l-322, 19886. Schneider, L.S.; Severson, J.A.; and Sloane, R.B. Platelet 3H-imipramine binding in depressed elderly patients. Biological Psychiatry, 20:1234-1237, 1985. Schneider, L.S.; Severson, J.A.; Sloane, R.B.; and Fredrickson, E.R. Decreased platelet 3H-imipramine binding in primary major depression compared with secondary depression to medical illness in elderly outpatients. Journal of Affective Disorders, 15: 195-200, 1988~. Severson, J.A.; Woodward, J.J.; and Wilcox, R.E. Subdivision of mouse brain 3Himipramine binding based on ion dependence and serotonin sensitivity. Journal of Neurochemistry 46: 1743-I 754, 1986. Talvenheimo, J.; Nelson, P.J.; and Rudnick, G. Mechanism of imipramine inhibition of platelet 5-hydroxytryptamine transport. Journal of Biological Chemistry, 2541463 l-4635, 1979. Talvenheimo, J.; Fishkes, H.; Nelson, P.J.; and Rudnick, G. The serotonin transporterimipramine receptor: Different sodium requirements for imipramine binding and serotonin translocation. Journal of Biological Chemistry, 258:6115-6119, 1983. Tang, S.W., and Morris, J.M. Variation in human platelet 3H-imipramine binding. Psychiatry Research, 16:141-146, 1985. Weizman, A.; Care], C.; Tyano, S.; and Rehavi, M. Decreased high affinity 3H-imipramine binding in platelets of enuretic children and adolescents. Psychiatry Research, 14:39-46, 1985.
19
Elsevier
Methodological Issues Platelet 3H-lmipramine James A. Severson,
in the Preparation Binding
Lon S. Schneider,
and Assay
of
and Eric R. Fredrickson
Received May 26, 1989; first revised version received September version received March 19, 1990; accepted April 8, 1990.
1.5, 1989; second revised
Abstract. Several methodological factors in the preparation of platelets and the determination of platelet 3H-imipramine (3H-IMI) binding were examined. The ionic composition of the assay significantly affected platelet 3H-IMI binding. Approximately 25% of the specific binding of 3H-IMI to intact platelet preparations was retained in the absence of sodium and chloride ions. The addition of sodium ions enhanced the specific binding of 3H-IMI, but the addition of chloride in the presence of sodium had a more pronounced effect, enhancing binding approximately five-fold over that observed with the addition of sodium. Sodium was the only cation tested that enhanced binding. Only halides enhanced binding in the presence of sodium with the following order of potency: Cl- > Br- > I- = F-. Ions increased the density of binding sites (B,,,,,) and did not affect the affinity of the binding sites for 3H-IMI. In the presence of sodium and chloride, the use of serotonin (5HT) to define nonspecific binding in saturation experiments resulted in lower binding densities (B,,,) than when desipramine was used to define nonspecific binding. The component of binding that was insensitive to 5HT was roughly equal to the B,, of 3H-IMI binding obtained in the absence of sodium and chloride using desipramine to define nonspecific binding. Overall, these data suggest that not all 3H-IMI binding that is displaced by desipramine is related to serotonergic mechanisms, and suggest that 5HT is a better choice than desipramine for the determination of the nonspecific binding of 3H-IMI. In addition, the binding of 3H-IMI to different platelet preparations was compared. The binding of 3H-IMI to intact platelets was less than that obtained using lysed platelet membranes when data were expressed per mg protein. The Coomassie Blue dye-binding method to determine platelet protein resulted in greater B,,,,, values than were obtained with the Folin phenol reagent method. The method of platelet preparation that is commonly used to prepare platelets for 3H-IMI binding resulted in similar binding values when compared to a method that prepares the entire platelet population. The results suggest that some, but not all, variations in laboratory methods used to prepare platelets and assay for platelet 3H-IMI binding may affect clinical studies examining this measure. Key Words.
Imipramine
binding,
platelets,
ion-dependence,
James A. Severson, Ph.D., is Manager,
receptor
binding.
Pharmaceutical Market Development, New Market Development, Amersham Corporation, Arlington Heights, IL. Lon S. Schneider, M.D., is Assistant Professor of Psychiatry, University of Southern California School of Medicine, Los Angeles, CA. Eric R. Fredrickson, B.S., was formerly Research Laboratory Technician, USC School of Medicine, and is now at the University of Kansas, School of Medicine, Kansas City, KS 66103. (Reprint requests to Dr. L.S. Schneider, Dept. of Psychiatry and the Behavioral Sciences, 1934 Hospital PI., Los Angeles, CA 90033, USA.) 0165-1781/90/$03.50
@ 1990 Elsevier Scientific
Publishers
Ireland
Ltd.
20
The binding of 3H-imipramine (3H-IMI) to platelets has received considerable attention as a potential biological marker for depression. In populations of clinically depressed individuals, the average density of the binding sites for 3H-IMI on platelets is reduced relative to populations of individuals who are not judged to be clinically depressed (reviewed by Langer et al., 1987; Beth et al., 1988). However, clinical conditions that differ from.depression also may have 3H-IMI binding that is reduced compared to control populations. For instance, adults with agoraphobia (Lewis et al., 1985), patients with Parkinson’s disease (Schneider et al., 1988a), patients with agitated Alzheimer’s disease (Schneider et al., 1988b), and children with nocturnal enuresis (Weizman et al., 1985) are reported to have reduced platelet JH-IMI binding. The usefulness of platelet 3H-IMI binding as a marker for depression is further complicated by the fact that not all laboratories that have studied 3H-IMI binding to platelets in depression have observed a reduction in the number of binding sites (Mellerup et al., 1982; Gentsch et al., 1985; Tang and Morris, 1985; Muscettola et al., 1986; Georgotas et al., 1987; Kanof et al., 1987). In addition to concern over the lack of consistency of findings in clinical depression is the seemingly wide variation in the values for the density of 3H-IMI binding sites from various laboratories. This suggests that laboratory-to-laboratory variation in techniques associated with JH-IMI binding may contribute to the variation in binding density reported in the literature, and also may contribute to the lack of consistency in findings of 3H-IMI binding in depression. The present report examines the effect of the method used to obtain platelets for 3H-IMI binding, the use of intact vs. lysed platelets, and the assay for platelet protein. In addition, it examines components of 3H-IMI binding that are regulated by ions and are sensitive and insensitive to displacement by serotonin (5-hydroxytryptamine, 5HT). Methods Blood Collection and Platelet Preparation. Whole blood was obtained from healthy volunteers who ranged in age from 25 to 35 years. Blood was collected by venipuncture into vacutainers containing EDTA and was allowed to sit for 1 to 2 hours at room temperature. Platelet-rich plasma was obtained by low-speed centrifugation (8Og, room temperature, 1 hour), and was then pipetted into a clean polypropylene centrifuge tube using a clean polypropylene Pasteur pipette and recentrifuged (21,OOOg, 4 “C, 5 min). The plasma supernatant was discarded and the platelet pellet stored at -70 ‘C. Platelets obtained from 10 ml of whole blood were thawed over ice in 2 ml of buffer I (50 mM Tris-HCl, pH 7.4, containing 150 mM NaCl and 20 mM EDTA, disodium salt) and were dispersed by gentle uptake and expulsion from a plastic Pasteur pipette. The resuspended pellet was centrifuged at (2 1,OOOg,4 o C, 5 min). The supernatant was discarded and the pellet resuspended in 2 ml of buffer I and recentrifuged. After the second wash in buffer I, the platelet pellet was resuspended in 2.0 ml of incubation buffer (50 mM Tris-HCl, pH 7.4, 120 mM NaCl, 5 mM KCl) and recentrifuged (4 ‘C, 21,OOOg, 5 min). The final platelet pellet was prepared for binding assays by resuspension with a plastic Pasteur pipette in 4.0 ml of cold incubation buffer. Platelets were prepared for ion-sensitive binding as described above with the exception that 50 mM Tris-acetate buffer (PH 7.4) without salts was substituted for incubation buffer in all washing and resuspensions.
21 lmipramine Binding Assay. 3H-IMI (73 Ci/mmole) was diluted in incubation buffer and was added to glass test tubes in 50 ,ul in duplicate. In binding saturation experiments, six 3H-IMI concentrations (0.1258 nkf) were used, and in single 3H-IMI concentration experiments, tubes contained 2 nM3H-IMI. Assay tubes received either desipramine (Sigma) to determine nonspecific binding (100 PM final concentration in 100 ,ul of incubation buffer) or 100 ,ul of incubation buffer for total binding. Platelet suspensions (100 ,~l) were added last for a final assay volume of 250 /11. For ion-sensitive binding, ions were added separately to assay tubes in 50~1 of Tris-acetate. Desipramine (100 PM assay concentration) and JH-IMI were diluted in Tris-acetate @H 7.4) and each was added in 50 ~1. Assays were initiated by the addition of 100 ~1 of platelets in Tris-acetate buffer. All 3H-IMI binding experiments were conducted at 20-40 /lg of protein (Bradford, 1976) per tube. Assay tubes were incubated at O-2 “C for 60 min and the assay was stopped by dilution with 5 ml of wash buffer (50 mM Tris-HCl, pH 7.4, 120 mkf NaCl, 5 mM KCl) and collection of platelets on glass fiber filters (S&S No. 32, Schleicher and Schuell, Keene, NH). Tubes were rinsed with an additional 5 ml of wash buffer and filtered. Filters were then washed twice with 5 ml of wash buffer. The radioactivity retained on each filter was determined by liquid scintillation spectroscopy. Protein Assays. The protein content of platelet preparations was determined using proteindye binding (Bradford, 1976). In some experiments, the Folin-phenol method (Lowry et al., 1951) also was used for comparison. Data Analysis. Binding isotherms were transformed to the relationship of bound/free vs. bound (Scatchard, 1949) and the binding capacity (B,,,,,) and equilibrium dissociation constant (&) were determined by least squares linear regression. JH-IMI displacement curves were analyzed using a four-parameter logistic model (ALLFIT; DeLean et al., 1978). The concentration of drug that displaced 50% of the specifically bound 3H-IMI (IC,,) obtained from this analysis was corrected for the concentration of JH-IMI (Cheng and Prusoff, 1973) for presentation as the inhibition constant (K,).
Results Ion-Sensitive Binding of lmipramine to Human Platelets. Platelets were prepared in Tris-acetate and assayed in duplicate in 2 nM 3H-IMI for total and nonspecific binding (100 PM desipramine). Sodium was added as NaCl at a final concentration of 300 mM. Approximately 75% of the 3H-IMI displaced by desipramine was sensitive to the addition of NaCl (mean + SD: with NaCl = 956 f 244 fmol/mg protein; without NaCl = 239 f 100; difference = 7 17 + 224, 16 paired experiments, t = 12.80, df = 15, p < 0.001). Determination of the Specificity of Cations. Platelets were prepared in Trisacetate and were tested for cationic dependency of desipraminedisplaceable 3H-IMI binding. Cations were added as 300 mM of the chloride salt. The only cation that was found to enhance the binding of 3H-IMI to platelets was sodium (Table 1); the addition of sodium as either NaCl or sodium acetate elevated binding. The effect of NaCl was greater than that of sodium acetate because of the additional effect of chloride on binding (see below). However, some chloride salts depressed the apparent specific binding of 3H-IMI.
22 Table 1. Effect of addition of ions on the binding of W-imipramine to platelets
Cations
Desipramine-displaceable bindlng (fmol/mg protein)
Anions
Desipramine-displaceable binding (fmoUmg protein)
None
194 j,
42
(9)
None
275 f
LiCl
150f
51
(6)
NaBr
736f144’(3)
KCI
126f
51
(6)
Nal
499 f 147’ (3)
NaCl
671 f 206’
(6)
NaF
459 f 110’ (6)
66 (9)
MgClz CaCb
59f
23’
(3)
NaHC03
323 f
57’ (3)
41 *
29’
(3)
KzHPO,
166f
55 (3)
ZnCh
47f
19’
(3)
Li2S04
195f
54 (3)
NaCl
926 f 240’ 16)
NaOAc
296 f
62’ (12)
Note. Platelets were prepared in 50 mM Tris-acetate, and Y-Limipramine binding was determined in duplicate for total and nonspeciffc binding using 2 mM 3H-imipramine and 100 @f desipramine to determine nonspecific binding. Cations were added at a concentration of 300 mM, and anions were added at a concentration of 150 mM in the presence of 300 mM sodium acetate to fulfill the requirement for sodium ion. Data are mean f SD of the number of determinations in parentheses
1.p < 0.05 vs. “none” in each group. Determination of the Specificity of Anions. Platelets were prepared in Trisacetate @H 7.4) and were tested for the anionic dependency of 3H-IMI binding in the presence of 300 mM sodium acetate because sodium was necessary to obtain the anion effect on binding (compare NaCl addition with the addition of sodium acetate in Table 1). Halides were the only anions that enhanced the binding of 3H-IMI in the presence of saturating amounts of sodium (Table 1). The order of effect of the halide salts was Cl- > Br- > I- = F-. Displacement of Platelet lmipramine Binding. In the presence of NaCl, displacement of 3H-IMI from platelets followed the pattern expected; displacement by desipramine was steep with a Ki of 43 nA4. 5HT was a much weaker displacing agent than desipramine with a K, of 1400 nM (Fig. 1). In the absence of NaCl, desipramine and 5HT were much weaker displacers. Desipramine was an approximately IO-fold weaker displacer (Ki = 513 nM) in the absence of NaCl, and 5HT did not significantly displace 3H-IMI in the absence of NaCl (Fig. 1). Dose Response to Addition of Ions. The addition of sodium ions as sodium acetate resulted in the dose-dependent elevation of 3H-IMI binding (Fig. 2A). Sodium ion increased desipramine-displaceable binding to approximately 150 fmol/mg protein above that obtained in the absence of ions. The concentration of sodium that elevated the binding by 50% was approximately 50 mM. In the presence of 300 mM sodium (as sodium acetate), the addition of chloride ion as NaCl enhanced the desipramine-displaceable binding of 3H-IMI to platelets to 800 fmol/ mg protein above that obtained with 300 mM sodium acetate alone (Fig. 2B). The concentration of chloride ion that elevated binding by 50% was approximately 30 mM.
23 Fig. 1. Displacement
of 3H-imipramine
from platelets
in the presence
and
0 10-S
10-8
lo*
10-7
10-S
104
CONCENTRATION Platelets were prepared in Tris-acetate for sodium&per&m binding and 11 concentrations each of desipramine (DMI) and serotonin (SHT) in the presence and absence of 300 mM sodium chloride were used to determine the effect of sodium on the concentration dependence of the displacement of %imipramine binding. Inhibition of JH-imipramine binding was anatyzed by iterative nonlinear regression using a 4-parameter logistic program and the ICY, corrected for the concentration of 3H-imipramine. Data are means of 3 independent experiments.
Fig. 2. Concentration dependence of the enhancement imipramine binding by sodium and chloride
A.
Na+ DEPENDENCE
B.
of platelet
3H-
Cl - DEPENDENCE
(3
~a+ CONC~RATION
(m.4)
Cl- CONCENTRATION
(mM)
Platelets were prepared in Tris-acetate and 3H-imipramine binding (2 nM) was determined in duplicate for total and nonspecific binding (100flM desipramine). (A) Sodium dependence was examined using increasing concentrations of sodium acetate (O-600 mM). (B) Chloride-dependent binding was examined using increasing concentrations of sodium chloride (O-400 mM) in the presence of 300 mM sodium acetate. Data are mean f SEM of the specific binding from 4 independent dose-response curves.
Effect of Ions and 5HT on Maximal Binding and Equilibrium Dissociation Constants. The influence of ions on 3H-IMI binding isotherms was tested. Platelets were prepared in Tris-acetate and the effect of ions on desipramine-displaceable binding was examined in the absence of ions, in the presence of sodium (as sodium acetate), and with sodium and chloride. In the absence of sodium and chloride, 3H-IMI binding was saturable with an equilibrium dissociation constant of 1.8 nM and a binding capacity of 373 fmol/mg protein (Table 2). The addition of sodium
24
Table 2. Effect of addition of ions and use of serotonin as a displacer on the bindina of 3H-imioramine to olatelets
Ion dependence iris-acetate
373 f
94
1.82 * 0.44
Tris + Na+
532 LII 74’
2.18 f 0.66
Tris + Na+ + Cl-
1964+
1382
1.86 zt 0.48
94
1.89 rt 0.92
Serotonin displacement Desipraminefor nonspecific binding Tris-acetate
373 It
Tris + NaCl
1847 + 1983
1.64 zk 0.38
1449 f 228“
1.91 f 0.54
Serotonin for nonspecific binding Tris + NaCl Serotonin-insensitive
bindinn
398f
116
1.76 k 0.48
Note. Binding isotherms were determined in duplicate for total and nonspecific binding at 6 concentrations of 3H-imipramine (0.125-E nU). Platelets were prepared in Tris-acetate and ion-dependent binding was conducted in the absence of ions (Tris-acetate), with 300 mM sodium acetate (Tris + Na’), and 300 mM sodium acetate and 150 mM NaCl (Tris t Na’ t Cl-). Nonspecific binding was determined using 100 &f desipramine (DMI). Serotonin displacement was conducted in the absence of ions (Tris only) and with 300 mM NaCl (Tris t NaCI). Nonspecific binding was determined using 100 &f serotonin and 100 &f DMI. Serotonin-insensitive binding was obtained by subtracting binding in the presence of 100fiM DMI from the binding in the presence of 1OOpM serotonin. Data are mean f SD of 4 independent experiments.
1.p < 0.05 vs. Tris-only. 2. p < 0.001 vs. Tris-only; p < 0.001 vs. Tris + Na. 3. p < 0.001 vs. Tris-only - DMI. 4. p < 0.05 vs. Tris + NaCl - DMI; p < 0.001 vs. serotonin-insensitive binding
increased the capacity of binding by approximately 40% with little change in the Kd. The addition of chloride in the presence of sodium more dramatically affected binding capacity; the B,,, for 3H-IMI binding was increased over 265% by addition of chloride, and by 525% over no ions. The Kd was not affected by the addition of ions. In addition, 5HT was tested as a displacer for saturable 3H-IMI binding. Binding assays were constructed so that binding isotherms were conducted in the presence and absence of Na and Cl using desipramine as the displacer, and in the presence of Na and Cl using 5HT as the displacer. Thus, it was possible by subtraction (tubes with 5HT added minus tubes with desipramine added) to obtain an isotherm for SHT-insensitive binding. In the absence of ions, the binding of 3H-IMI displaced by desipramine was saturable with a capacity of 373 fmol/mg protein (Table 2). The addition of sodium and chloride enhanced the binding of 3H-IMI to platelets almost five-fold. Using 5HT to define nonspecific binding in the presence of Na and Cl resulted in a Bmax that was 22% lower than that observed when desipramine was used to define nonspecific binding. The Kd for saturable JH-IMI binding was similar in assays in which 5HT or desipramine were used to define nonspecific binding. SHT-insensitive binding was not different from desipramine in the absence of Na and Cl.
25 of lmipramine Binding to Intact and Lysed Platelets. Platelets were prepared in buffer containing NaCl and were then lysed using the method of Raisman et al. (1981) for comparison with the binding to intact platelets using both the Bradford and Lowry protein assays to determine the protein content of platelet preparations. The saturable binding of 3H-IMI to lysed platelet preparations was 65% higher than the binding to intact platelets (Table 3). In addition, there was a significant increase in the Kd (lower binding affinity) of the binding of 3H-IMI to lysed platelets compared to the binding to intact platelets. The use of Lowry protein to express binding data resulted in a reduction in the apparent density of 3H-IMI binding. Overall, expression of B,,,,, data using Lowry protein reduced the B,,,,, by 29.9 f 12.7% (mean f SD, n = 24). Comparison
Table 3. Comparison preparations
of 3H-imSpramine binding
to intact and lysed platelet
Bradford protein
Lowry protein
3161 f 1166
2000 + 462’
1.69 f 0.73
1.69 zt 0.73
1910+55a2
1345 f 208’
1.18 + 0.172
1.18 f0.17
LY--J B,,,,, (fmol/mg protein)
Kd (nM) Intact B,,, &
(fmollmg protein) (nM)
Note. Platelet-rich plasma was prepared in Tris plus NaCl and KCI. The plasma was divided into equal volumes and one portion was processed to prepare intact platelets (see Methods) and the other portion was lysed (Raisman et al., 1961). Binding isotherms were conducted in duplicate at 6 concentrations of 3H-imipramine (0.1254 nA4) using 100 PM desipramine to determine nonspecific binding. Protein content of the intact and lysed platelet preparations was determined using both Coomassie Blue dye binding (Bradford, 1976) and the Folin phenol reagent (Lowry et al., 1951). Data are mean + SD of 12 individual experiments. 1. p < 0.01 vs. B,, using Bradford protein. 2. p < 0.01 vs. similar binding parameter from lysed platelets.
Comparison of Methods of Platelet Preparation. Platelet-rich plasma was processed to prepare the entire platelet population according to the method of Corash (1980) for comparison to the procedure described in Methods. In five parallel experiments, the B,,, of the binding of 3H-IMI to intact platelets was similar between the whole platelet population and the method used in this laboratory (Table 4). Discussion
Platelet 3H-IMI binding is regulated by sodium ions and anions: Na was the only cation tested that enhanced 3H-IMI binding and Cl was the most effective anion tested. Only halides enhanced the binding of 3H-IMI, and the order of halide specificity was similar to that observed in brain: Cl- > Br- > I- = F-. The effects of ions were shown to be concentration-dependent and to have a major effect on the measured density of 3H-IMI binding sites rather than to affect the affinity of the binding site for 3H-IMI. Furthermore, the effect of halides requires the presence of
26
Table 4. Comparison of methods of platelet DreDaration 3H-imipramine binding
Methods Single spin collection B,.,
(fmol/mg
protein)
Kd W)
1994 f
561
1.42 f
0.54
1869 f
561
Whole platelet population B max (fmol/mg
Kd WI
protein)
1.39 + 0.54
Note. Platelets were prepared from platelet-rich plasma by a one-step isolation procedure (see Methods) or by a method that yields the whole platelet population (Corash, 1980). Binding isotherms were constructed at 6 concentrations of 3H-imipramine (0.125-E nM) using 100 jAf desipramine to determine nonspecific binding. Platelet protein content was determined using protein-dye binding (Bradford, 1976). Data are mean + SD of 5 assays.
sodium; chloride salts of cations other than sodium were ineffective in increasing the density of 3H-IMI binding sites. In addition, desipramine did not distinguish between ion-sensitive and ion-insensitive binding, whereas 5HT did distinguish between these two populations of binding sites. The results obtained in this study for the ion-dependency of platelet 3H-IMI binding are similar to those reported previously for brain (Severson et al., 1986; Backstrom and Marcusson, 1987). In brain, ion-sensitive binding was found to be associated with 5HT neurons and SHT uptake. However, the proportion of ioninsensitive binding sites on platelets is less than in brain; platelet ion-insensitive binding of 3H-IMI represents between 10 and 25% of the desipramine-displaceable binding (Marcusson and Tiger, 1988; Hrdina, 1989; this study), whereas in brain approximately 50% of the binding is insensitive to ions. In addition, the use of 5HT to define nonspecific binding resulted in a binding density that was 78% of that obtained when desipramine was used. This is similar to the 75% of 3H-IMI binding that was found to be ion-sensitive. Together, these two approaches show that 25% of the specific binding of 3H-IMI to platelets is probably not related to 5HT uptake and, thus, represents sites extraneous to the measurements desired. Others have considered this component as part of the specific binding of 3H-IMI (Talvenheimo et al., 1979, 1983; Abbott et al., 1982; Rehavi et al., 1982). These data have immediate practical implications for studies of 3H-IMI to platelets. The data presented here clearly demonstrate that platelet JH-IMI binding is not absolutely dependent upon the presence of sodium ion. The present data suggest that 5HT is a better choice for the determination of nonspecific binding in studies of platelet 3H-IMI binding. It has been suggested that the binding of 3H-IMI to intact platelets is more consistent than binding to lysed platelet membranes because of inconsistencies in the degree of lysis of platelets (Fried1 et al., 1983). Our early preliminary studies confirmed these observations, and we have chosen to conduct platelet 3H-IMI
27 binding to intact platelets rather than the more commonly used lysed platelet membrane preparation. Several of our earlier studies using intact platelet preparations confirmed findings of decreased platelet 3H-IMI binding in depressed patients and extended these observations to elderly depressed patients (Schneider et al., 1985, 1986, 1988c), patients with Parkinson’s disease (Schneider et al., 1988~) and Alzheimer’s disease patients with agitated behavior (Schneider et al., 19886). Our studies of depression in the elderly have been confirmed by others using lysed platelet preparations (Nemeroff et al., 1988). These data suggest that not all variations in 3H-IMI binding may have a detrimental impact on the ability of the assay to detect significant population differences in platelet 3H-IMI binding. Studies presented here show that the density of 3H-IMI binding sites on an mg protein basis is greater in lysed platelets. This is most likely due to the removal of soluble proteins from the sample upon lysis and washing: 40% of platelet protein content is in the soluble fraction (Barber and Jamieson, 1970). For comparison purposes, in this study we also expressed binding densities per mg protein determined by the Lowry method (Lowry et al., 1951). We found that the Bradford method (Bradford, 1976) detects about 70% of the protein detected by the Lowry method in platelets. This contrasts with a 50% Bradford-to-Lowry ratio found in mouse brain (Severson et al., 1986). These differences presumably reflect differences in the mechanisms that the two methods use to detect protein. The Bradford method uses dye binding to arginine residues (Compton and Jones, 1985) whereas the Lowry method uses in part a reduction of tyrosine, tryptophan, and cysteine residues with a reaction of copper-protein complexes with the Folin reagent (Legler et al., 1985). The Lowry method also is sensitive to the amino acid composition of proteins (Lowry et al., 195 1; Legler et al., 1985). A result of the use of the Coomassie Blue method is that it increases binding densities by reducing the denominator used to express binding. However, the Bradford assay is less subject to interference by components of biological buffering systems than the Lowry method (Peterson, 1979). In addition, the system that we use for 3H-IMI binding uses less protein than that used by others, 20 to 4Opg protein per assay tube. We have found that higher protein concentrations reduce the apparent binding of 3H-IMI in a manner that suggests ligand depletion (Severson et al., 1986). The protein concentration chosen for our studies is at a concentration that is well within the linear response range for the 3H-IMI binding assay. Differences in platelet 3H-IMI binding between laboratories also could reflect incomplete collection of the entire platelet population, which could result in assaying 3H-IMI binding to platelet populations that differ in the density of binding sites (Arora and Meltzer, 1984). It is possible to isolate platelet populations that differ by size and buoyant density (Corash et al., 1977; Corash, 1980). However, the present data suggest that the binding of 3H-IMI to intact platelets is similar whether determined by a method designed to collect the entire platelet population or by less exhaustive platelet isolation procedures (Raisman et al., 1981; Schneider et al., 1986). In conclusion, the choice of assay conditions can significantly affect the data obtained in platelet 3H-IMI binding; protein concentrations, protein assays, ion
28 concentrations, the choice of agent used to define nonspecific binding, and the use of intact or lysed platelets can affect observed binding densities. While this study has characterized potential problems in 3H-IMI binding studies in depression, it will be necessary to test some assay variables in platelets obtained from depressed subjects to define how assay variables can affect the results obtained, and to determine their impact on the inconsistent results that are presently in the literature. References Abbott, W.M.; Briley, MS.; Langer, S.Z.; and Sette, M. Sodium shift of the inhibition of [sH]-imipramine binding by SHT and SHT-uptake blockers but not by tricyclic antidepressants. British Journal of Pharmacology, 76:295P, 1982. Arora, R.C., and Meltzer, H.Y. Imipramine binding in subpopulations of normal human blood platelets. Biological Psychiatry, 19:257-262, 1984. Backstrom, I.T., and Marcusson, J.O. 5-Hydroxytryptamine-sensitive [jH]imipramine binding of protein nature in the human brain: I. Characteristics. Brain Research, 425: 128-136, 1987. Barber, A.J., and Jamieson, G.A. Isolation and characterization of plasma membranes from human platelets. Journal of Biological Chemistry, 2456357-6365, 1970. Beth, P.; Eplov, L.; Gastpar, M.; Gentsch, C.; Mendlewicz, J.; Plenge, P.; Rielaert, C.; and Mellerup, E.T. WHO pilot study on the validity of imipramine platelet receptor binding sites as a biological marker of endogenous depression: A preliminary report on the initial evaluation phase of a World Health Organization collaborative study. Pharmacopsychiatry, 21:147-150, 1988. Bradford, M.M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry, 72:248-254, 1976. Cheng, Y.-C., and Prusoff, W.H. Relationship between the inhibition constant (K,) and the concentration of inhibitor which causes 50 per cent inhibition (Iso) of an enzymatic reaction. Biochemical Pharmacology, 2213099-3108, 1973. Compton, S.J., and Jones, C.G. Mechanism of dye response and interference in the Bradford protein assay. Analytical Biochemistry, 151:369-374, 1985. Corash, L. Platelet heterogeneity: Relevance to the use of platelets to study psychiatric disorders. Schizophrenia Bulletin, 6~254-257, 1980. Corash, L.; Tan, H.; and Gralnick, H.R. Heterogeneity of human whole blood platelet subpopulations: I. Relationship between buoyant density, cell volume, and ultrastructure. Blood, 49:71-87, 1977. DeLean, A.; Munson, P.J.; and Rodbard, D. Simultaneous analysis of families of sigmoidal curves: Application to bioassay, radioligand assay and physiological dose-response curves. American Journal of Physiology, 235:E97-E102, 1978. Friedl, W.; Propping, P.; and Week, B. 3H-Imipramine binding in platelets: Influence of varying proportions of intact platelets in membrane preparations on binding. Psychopharmacology, 80:96-99, 1983. Gentsch, C.; Lichtsteiner, M.; Gastpar, M.; Gastpar, G.; and Feer, H. 3H-Imipramine binding sites in platelets of hospitalized psychiatric patients. Psychiatry Research, 14: 177-187, 1985. Georgotas, A.; Schweitzer, J.; McCue, R.E.; Armour, M.; and Friedhoff, A.J. Clinical and treatment effects on 3H-clonidine and 3H-imipramine binding in elderly depressed patients. Life Sciences, 40:2137-2143, 1987. P.D. Differences between sodium-dependent and desipramine-defined Hrdina, [rH]imipramine binding in intact human platelets. Biological Psychiatry, 25:576-584, 1989. Kanof, P.D.; Coccaro, E.F.; Johns, C.A.; Siever, L.J.; and Davis, K.L. Platelet [rH]imipramine binding in psychiatric disorders. Biological Psychiatry, 22:278-286, 1987.
29 Langer, S.Z.; Galzin, A.M.; Poirier, M.F.; Loo, H.; Sechter, D.; and Zarifian, E. Association of [jH]imipramine and [jH]paroxetine binding with the 5HT transporter in brain and platelets: Relevance to studies in depression. Journal of Receptor Research, 7:499-521, 1987. Legler, G.; Muller-Platz, C.M.; Mentges-Hettkamp, M.; Pflieger, G.; and Julich, E. On the chemical basis of the Lowry protein determination. Analytical Biochemistry, 150:278-287, 1985. Lewis, D.A.; Noyes, R., Jr.; Coryell, W.; and Clancy, J. Tritiated imipramine binding to platelets is decreased in patients with agoraphobia. Psychiatry Research, 16: l-9, 1985. Lowry, O.H.; Rosebrough, N.J.; Farr, A.L.; and Randall, R.J. Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry, 193:265-275, 1951. Marcusson, J., and Tiger, G. [JH]Imipramine binding of protein nature in human platelets: Inhibition by 5-hydroxytryptamine and 5-hydroxytryptamine uptake inhibitors. Journal of Neurochemistry, 50:1032-1036, 1988. Mellerup, E.T.; Plenge, P.; and Rosenberg, R. 3H-Imipramine binding sites in platelets from psychiatric patients. Psychiatry Research, 7:221-227, 1982. Muscettola, G.; Di Lauro, A.; and Giannini, C.P. Platelet 3H-imipramine binding in bipolar patients. Psychiatry Research, 18:343-353, 1986. Nemeroff, C.B.; Knight, D.L.; Krishnan, R.R.; Slotkin, T.A.; Bissette, G.; Melville, M.L.; and Blazer, D.G. Marked reduction in the number of platlet-tritiated imipramine binding sites in geriatric depression. Archives of General Psychiatry, 45:919-923, 1988. Peterson, G.L. Review of the Folin phenol protein quantitation method of Lowry, Rosebrough, Farr and Randall. Analytical Biochemistry, 100:201-220, 1979. Raisman, R.; Sechter, D.; Briley, MS.; Zarifian, E.; and Langer, S.Z. High affinity -‘Himipramine binding in platelets from treated and untreated depressed patients compared to healthy volunteers. Psychopharmacology, 75:368-37 1, 198 1. Rehavi, M.; Skolnick, P.; and Paul, S.M. Solubilization and partial purification of the high affinity [jH]imipramine binding site from human platelets. FEBS Letters, 150:5 14-5 18, 1982. Scatchard, G. The attractions of proteins for small molecules and ions. Annals of the New York Academy of Sciences, 5 1:660-672, 1949. Schneider, L.S.; Chui, H.C.; Severson, J.A.; and Sloane, R.B. Decreased platelet 3Himipramine binding in Parkinson’s disease. Biological Psychiatry, 24:348-351, 1988~. Schneider, L.S.; Fredrickson, E.R.; Severson, J.A.; and Sloane, R.B. 3H-Imipramine binding in depressed elderly: Relationship to family history and clinical response. Psychiatry Research, 19:257-266, 1986. Schneider, L.S.; Severson, J.A.; Chui, H.C.; Pollock, V.E.; Sloane, R.B.; and Fredrickson, E.R. Platelet tritiated imipramine binding and MAO activity in Alzheimer’s disease patients with agitation and delusions. Psychiatry Research, 25:31 l-322, 19886. Schneider, L.S.; Severson, J.A.; and Sloane, R.B. Platelet 3H-imipramine binding in depressed elderly patients. Biological Psychiatry, 20:1234-1237, 1985. Schneider, L.S.; Severson, J.A.; Sloane, R.B.; and Fredrickson, E.R. Decreased platelet 3H-imipramine binding in primary major depression compared with secondary depression to medical illness in elderly outpatients. Journal of Affective Disorders, 15: 195-200, 1988~. Severson, J.A.; Woodward, J.J.; and Wilcox, R.E. Subdivision of mouse brain 3Himipramine binding based on ion dependence and serotonin sensitivity. Journal of Neurochemistry 46: 1743-I 754, 1986. Talvenheimo, J.; Nelson, P.J.; and Rudnick, G. Mechanism of imipramine inhibition of platelet 5-hydroxytryptamine transport. Journal of Biological Chemistry, 2541463 l-4635, 1979. Talvenheimo, J.; Fishkes, H.; Nelson, P.J.; and Rudnick, G. The serotonin transporterimipramine receptor: Different sodium requirements for imipramine binding and serotonin translocation. Journal of Biological Chemistry, 258:6115-6119, 1983. Tang, S.W., and Morris, J.M. Variation in human platelet 3H-imipramine binding. Psychiatry Research, 16:141-146, 1985. Weizman, A.; Care], C.; Tyano, S.; and Rehavi, M. Decreased high affinity 3H-imipramine binding in platelets of enuretic children and adolescents. Psychiatry Research, 14:39-46, 1985.
