QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: jpet.108.136895v1 325/3/910    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Whiteaker, P. Articles by McIntosh, J. M. Search for Related Content PubMed PubMed Citation Articles by Whiteaker, P. Articles by McIntosh, J. M.

CELLULAR AND MOLECULAR

Synthesis and Characterization of ¹²⁵I--Conotoxin ArIB[V11L;V16A], a Selective 7 Nicotinic Acetylcholine Receptor Antagonist

Division of Neurobiology, Barrow Neurological Institute, Phoenix, Arizona (P.W.); Institute for Behavioral Genetics, University of Colorado, Boulder, Colorado (M.J.M., A.C.C.); and Departments of Biology (S.C., C.D., J.M.M.) and Psychiatry (J.M.M.), University of Utah, Salt Lake City, Utah

Received for publication January 19, 2008
Accepted March 4, 2008.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
The 7 nicotinic acetylcholine receptors (nAChRs) are widely expressed both in the central nervous system (CNS) and periphery. In the CNS, ¹²⁵I--bungarotoxin is commonly used to identify 7 nAChRs specifically. However, -bungarotoxin also interacts potently with 1^* and 910 nAChRs, two receptor subtypes in peripheral tissues that are colocalized with the 7 subtype. [³H]Methyllycaconitine is also frequently used as an 7-selective antagonist, but it has significant affinity for 6^* and 910 nAChR subtypes. In this study, we have developed a highly 7-selective -conotoxin radioligand by iodination of a naturally occurring histidine. Both mono- and diiodo derivatives were generated and purified (specific activities were 2200 and 4400 Ci mmol^-1, respectively). The properties of the mono- and diiodo derivatives were very similar to each other, but the diiodo was less stable. For monoidodo peptide, saturation binding to mouse hippocampal membranes demonstrated a K_d value of 1.15 ± 0.13 nM, similar to that of ¹²⁵I--bungarotoxin in the same preparations (0.52 ± 0.16 nM). Association and dissociation kinetics were relatively rapid (k_obs for association at 1 nM was 0.027 ± 0.007 min^-1; k_off = 0.020 ± 0.001 min^-1). Selectivity was confirmed with autoradiography using 7-null mutant tissue: specific binding was abolished in all regions of 7^-/- brains, whereas wild-type mice expressed high levels of labeling and low nonspecific binding. ¹²⁵I--conotoxin ArIB[V11L; V16A] should prove useful where 7 nAChRs are coexpressed with other subtypes that are also labeled by existing ligands. Furthermore, true equilibrium binding experiments could be performed on 7 nAChRs, something that is impossible with ¹²⁵I--bungarotoxin.

Despite this successful track record, it has become apparent that -Bgt has some disadvantages. For example, -Bgt also binds with similar affinity to 8^* (Gotti et al., 1994) and 910 nAChRs (Elgoyhen et al., 2001) as it does to the 1 muscle-type and 7 neuronal nAChRs already mentioned. Neuronal nAChR expression and effects outside of the brain/CNS are areas of increasing interest, and 7 nAChR gene and/or protein expression has been shown to overlap with that of the -Bgt-sensitive 910 subtype in cochlea (Morley et al., 1998), dorsal root ganglion (Haberberger et al., 2004), keratinocytes (Nguyen et al., 2000), and lymphocytes (Peng et al., 2004). In addition, injured muscle coexpresses muscletype (1^*) and 7 nAChR subtypes (Martyn and Richtsfeld, 2006), both of which bind -Bgt with high affinity. Furthermore, the exceptionally slow binding kinetics of ¹²⁵I--Bgt to 7 nAChRs (Salvaterra and Mahler, 1976) makes true equilibrium binding studies impractical.

Accordingly, alternatives to -Bgt in identifying 7 nAChRs have been sought. The norditerpenoid alkaloid methyllycaconitine (MLA) isolated from Delphinium sp. is a competitive antagonist for 7 nAChRs, at which it has approximately nanomolar affinity and more useful kinetic properties than -Bgt (Davies et al., 1999). It is unfortunate that MLA also interacts with 6^* nAChRs with relatively high affinity (Salminen et al., 2005). In the central nervous system, MLA-sensitive 6β2^* nAChRs are concentrated in the dopaminergic projections of the substantia nigra and ventral tegmental area, and the optic tract (Gotti et al., 2005). These regions also contain 7 nAChRs (Clarke et al., 1985; Pauly et al., 1989; Quik et al., 2003), complicating efforts to identify 7 nAChRs specifically using MLA. 910 nAChRs are difficult to express heteromerically, but chimeric subunits (containing the N-terminal ligand binding domain of each subunit, fused in each case to the C-terminal of a 5-hydroxytryptamine_3A subunit) express well, and they also bind [³H]MLA with nanomolar affinity (Baker et al., 2004), suggesting that identification of peripheral 7 nAChRs with MLA may also suffer problems of specificity.

In a previous study (Whiteaker et al., 2007), we identified an -conotoxin, -CtxArIB, with some selectivity toward 7 nAChRs. Using a guided amino acid substitution strategy, we eventually produced a pair of modified peptides (-CtxArIB[V11L;V16A] and -CtxArIB[V11L;V16D]) with similar to unmodified affinity for 7 nAChRs, but with much improved 7 selectivity. It is noteworthy that both peptides also exhibited relatively rapid association and dissociation kinetics. In the current study, we describe the synthesis and characterization of a radioligand based on -CtxArIB (¹²⁵I--CtxArIB[V11L;V16A]). This peptide exhibits saturable binding, possesses similar affinity for 7 nAChRs as does ¹²⁵I--Bgt, selectively labels 7 nAChRs, and has relatively rapid association and dissociation kinetics. As such, it is possible to facilitate specific identification of 7 nAChRs in previously hard-to-study contexts, and to enable the performance of true equilibrium binding experiments at 7 nAChRs.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Preparation of Membranes, Mouse Brain Sections. Mice were bred at the Institute for Behavioral Genetics (University of Colorado, Boulder, CO), and they were housed five per cage. Male mice (C57BL/6J, and mice engineered to lack 7 nAChR subunit gene expression; Franceschini et al., 2002) were used when they were 60 to 120 days old. The vivarium was maintained on a 12-h light/dark cycle, and mice were given free access to food and water. All procedures used in this study were approved by the Animal Care and Utilization Committee of the University of Colorado.

To prepare brain membranes, each mouse was sacrificed by cervical dislocation, and the brain was removed and placed on an ice-cold platform. Tissue was collected from the hippocampus, olfactory tubercle, striatum, superior colliculus, thalamus, and midbrain. Individual hippocampal or thalamic samples were homogenized in ice-cold hypotonic buffer (14.4 mM NaCl, 0.2 mM KCl, 0.2 mM CaCl₂, 0.1 mM MgSO₄, and 2 mM HEPES, pH 7.5) using a glass-Teflon tissue grinder (Marks et al., 1998). Particulate fractions were collected by centrifugation at 25,000g for 15 min at 4°C (Eppendorf 5417 C centrifuge; Eppendorf North America, New York, NY). The pellets were resuspended in fresh homogenization buffer, incubated on ice for 10 min, and then harvested by centrifugation as described above. Each pellet was washed twice more by resuspension/centrifugation before storage (in pellet form under homogenization buffer) at -70°C until use. Pooled olfactory tubercle, striatal, and superior colliculus tissue from pairs of mice was homogenized, and then it was stored in the same way. Midbrain samples were pooled from 24 mice, and then they were homogenized as described for hippocampi, before being divided into 24 individual aliquots and stored at -70°C until needed.

Frozen Torpedo californica electroplax tissue was obtained from Aquatic Research Consultants (San Pedro, CA). Tissue was thawed on ice, and then it was homogenized (3 x 5 s at full speed) using a Polytron homogenizer (Brinkmann Instruments, Westbury, NY) in ice-cold isotonic binding buffer (144 mM NaCl, 2 mM KCl, 2 mM CaCl₂, 1 mM MgSO₄, and 20 mM HEPES, pH 7.5) supplemented with bovine serum albumin [0.1% (w/v)] and protease inhibitors (5 mM EDTA, 5 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, and 10 µg/ml each of aprotinin, leupeptin trifluoroacetate, and pepstatin A; protease inhibitor buffer) to minimize proteolysis. Membranes were collected by centrifugation at 25,000g for 30 min at 4°C, and the resulting pellets were rehomogenized as described above. The resulting membranes were washed twice more by resuspension in protease inhibitor buffer and centrifugation before storage (in pellet form under homogenization buffer) at -70°C until use.

Membranes were also prepared from Xenopus oocytes injected with 12.5 ng of each cRNA encoding rat 9 and 10 nAChR subunits (clones generously provided by A. B. Elgoyhen, Universidad de Buenos Aires, Buenos Aires, Argentina). After incubation at 17°C for 4 to 5 days, transfected oocytes were frozen using liquid nitrogen (typically in batches of 40) and stored at -70°C until required. The approach used was a slight modification of that described by Parker et al. (1998). Oocytes were thawed, suspended in 2 ml of distilled water, and homogenized (16 s at full speed) using a Polytron homogenizer (Brinkmann Instruments). The resulting homogenate was then centrifuged gently at 2000g for 2 min to separate it into three layers (lipids, top; membrane fragments, middle; and large fragments and pigment granules, bottom). Membrane fragments were collected using a Pasteur pipette, and they were used for binding assays.

For mouse brain sections, mice of 7^+/+ and 7^-/- genotype were killed by cervical dislocation, and the brains were removed from the skull and rapidly frozen by immersion in isopentane (-35°C; 10 s). The frozen brains were wrapped in aluminum foil, packed in ice, and stored at -70°C until sectioning. Coronal tissue sections (14 µmin thickness) were obtained using a Leica (Nussloch, Germany) CM 1850 cryostat/microtome, and thaw-mounted onto SuperFrost glass slides (Fisher Scientific, Pittsburgh, PA). Sections were stored, desiccated, at -70°C until use.

Choice of -CtxArIB[V11L;V16A], His-Iodination. In our previous study, we identified an 7 nAChR subtype-selective peptide, -CtxArIB. This peptide, although moderately selective, also showed some affinity for 6/3β2β3 nAChRs (6/3β2β3:7 affinity ratio = 3.56). Accordingly, we generated a series of increasingly more 7-selective derivatives. Of these derivatives, the two most selective were -CtxArIB[V11L;V16A] (functional IC₅₀ value at 7 = 0.356 nM; 6/3β2β3:7 affinity ratio = 337) and -CtxArIB[V11L;V16D] (functional IC₅₀ value at 7 = 1.09 nM; 6/3β2β3:7 affinity ratio = 760; Whiteaker et al., 2007). We chose -CtxArIB[V11L;V16A] as a starting point for the generation of a radiolabeled, 7-selective -CtxArIB derivative because it is highly 7-selective, but it has higher affinity than -CtxArIB[V11L;V16D].

-CtxArIB and each of its derivatives contain a single histidine residue at position 15 (Whiteaker et al., 2007), which represents a potential iodination site. Before attempting radioiodination at this native residue, we first cold-iodinated -CtxArIB[V11L;V16A] to see whether this affected the affinity of -CtxArIB[V11L;V16A] for 7 nAChRs.

Synthesis of -CtxArIB[V11L;V16A] was performed as described previously (Whiteaker et al., 2007). To iodinate the peptide, 5 nmol of -CtxArIB[V11L;V16A] was dissolved in 10 µl of 0.25 mM Tris, pH 8.2. To this was added 15 µl of 0.4 mM NaI or 5 mCi of Na¹²⁵I (volume, 15 µl; specific activity, 2200 Ci/mmol; PerkinElmer Life and Analytical Sciences, Boston, MA). The iodination reaction was initiated by the addition of 12.5 µl of freshly prepared 4 mM chloramine-T (p-toluene sulfonochloramide). The water-soluble chloramine-T provides a more easily controlled iodination reaction than the alternative insoluble reagent iodogen. The reaction proceeded at room temperature for 40 min, and then it was quenched by the addition of 65 µl of freshly prepared 0.5 M ascorbic acid. The pH of the reaction mixture was further lowered by the addition of 0.8 ml of 0.1% trifluoroacetic acid (TFA). Monoiodinated and diiodinated peptides were separated from unmodified peptide by reversed-phase high-performance liquid chromatography using an analytical C18 column (Grace Vydac, Hesperia, CA). Buffer A was 0.1% TFA, and buffer B was 0.092% TFA, 60% acetonitrile. The gradient was 10 to 50% buffer B over 40 min. Flow rate was 1 ml/min, and the absorbance was monitored at 220 nM. For purification of radioactive peptide a solution of 2.5% sodium thiosulfate and 0.2% potassium iodide in 1 N NaOH was added to the waste collection beaker to trap unreacted ¹²⁵I. Fractions containing mono- and diiodinated peptide were collected in siliconized tubes containing 10 µlof 20 mg/ml lysozyme to minimize absorption to the tubes. Using the above-mentioned chromatographic conditions, the unmodified peptide elutes at approximately 30.8% B, the monoiodinated peptide elutes at approximately 31.5% B, and the diiodinated peptide elutes at approximately 33.8% B. Final yield for the cold reaction was approximately 2 nmol of monoiodo peptide and 1 nmol of diiodo peptide, and for the hot reaction, final yield was 0.5 nmol of monoiodo peptide and 0.3 nmol of diiodo peptide. Mono- and diiodination of -CtxArIB[V11L;V16A] was verified by matrix-assisted laser desorption ionization time-of-flight mass spectrometry at the Salk Institute for Biological Studies (San Diego, CA) under the direction of Jean Rivier.

Measurement of His-Iodination (Nonradioactive) Effects on 7 Affinity. The effects of His iodination (using nonradioactive iodide) on -CtxArIB[V11L;V16A] affinity for 7 subtype nAChRs were determined using inhibition binding experiments. Peptide concentration ranges were as follows: -CtxArIB[V11L;V16A], 0.3 nM to 10 µM; I₁--CtxArIB[V11L;V16A], and I₂--CtxArIB[V11L;V16A], 0.03 nM to 1 µM. Protocols followed were very similar to those described previously (Whiteaker et al., 2007). Affinity for 7 nAChRs was measured by displacement of ¹²⁵I--Bgt binding to hippocampal membranes. Incubations were performed at 22°C for 4 h in 1.2-ml polypropylene tubes arranged in a 96-well format. Hippocampal membranes were incubated with 2 nM monoiodinated -Bgt (¹²⁵I₁--Bgt, 2000 Ci mmol^-1; GE Healthcare, Piscataway, NJ) in a total volume of 30 µl of protease inhibitor buffer to minimize proteolysis. Displacement of ¹²⁵I--Bgt binding was assessed in duplicate. Total (no peptide added) and nonspecific binding (defined using 10 µM -cobratoxin) was also measured in duplicate. After the initial incubation, 1 ml of isotonic binding buffer, supplemented with bovine serum albumin [0.1% (w/v)] was added to each tube, and the incubation was continued for a further 5 min at 22°C. This dilution and further incubation step allow some of the nonspecific ¹²⁵I₁--Bgt binding to dissociate, but it has no measurable effect on specific binding, increasing the signal-to-noise ratio of the binding assay. Binding reactions were terminated by filtration using a 96-place manifold (Inotech Biosystems, Lansing, MI). Particulate fractions were collected onto single layers of Inotech 0.75-µm retention glass fiber filters that had been soaked in 5% nonfat skim milk dissolved in isotonic binding buffer. Samples were washed six times, and all filtration and washing steps were conducted in a 4°C cold room, using ice-cold buffer. Bound radioligand was quantified by gamma counting at 83 to 85% efficiency, using a Packard Cobra counter (PerkinElmer Life and Analytical Sciences).

¹²⁵I--CtxArIB[V11L;V16A] Association and Dissociation Kinetics. First, the association kinetics of specific ¹²⁵I₁--CtxArIB[V11L;V16A] and ¹²⁵I₂--CtxArIB[V11L;V16A] binding to hippocampal membranes was measured. Hippocampal tissue was chosen because it expresses high concentrations of 7 nAChRs, is relatively large, and is simple to dissect. To minimize nonspecific binding and to reduce the possibility of labeling putative lower affinity receptor populations, a low (1 nM) radioligand concentration of ¹²⁵I--CtxArIB[V11L;V16A] was used. Incubations proceeded for 2 to 180 min in a total volume of 30 µl of isotonic binding buffer supplemented with protease inhibitors in 96-well polypropylene plates. At each time point, total (no added drug) and nonspecific (in the presence of 10 µM -cobratoxin) binding was measured in triplicate. Binding reactions were terminated by filtration using a 96-place manifold (Inotech Biosystems). Particulate fractions were collected onto single layers of Inotech 0.75-µm retention glass fiber filters that had been soaked in 5% nonfat skim milk dissolved in isotonic binding buffer. Alternative filter preparations were piloted [filters soaked in binding buffer alone, or soaked in 0.5% (w/v) polyethylenimine], but these preparations produced higher nonspecific binding, and they were not used (data not shown). Samples were washed six times, in a 4°C cold room, using ice-cold buffer. Bound radioligand was quantified by gamma counting at 83 to 85% efficiency, using a Packard Cobra counter (PerkinElmer Life and Analytical Sciences).

Dissociation kinetics was measured similarly. Hippocampal membranes were first incubated with 1 nM radioligand for 2 h (long enough for equilibrium binding to be closely approached, as revealed by the association kinetics experiments), in the same configuration as described for the association experiments. Total (no added drug) and nonspecific binding (in the presence of 10 µM -cobratoxin) triplicates were set up for each dissociation time point. Dissociation was initiated by the addition of 10 µlof40 µM -cobratoxin to each well, and it was allowed to proceed for 2 to 180 min before sample collection, washing, and radioactivity counting as described for the association experiments.

¹²⁵I--CtxArIB[V11L;V16A] Saturation Binding. Saturation binding experiments were performed for ¹²⁵I₁--CtxArIB[V11L;V16A] and ¹²⁵I₂--CtxArIB[V11L;V16A], using mouse hippocampal membranes as a source of 7 nAChRs. For comparison, saturation of ¹²⁵I₁--Bgt binding was assessed in parallel. A slightly modified version of the ¹²⁵I₁--Bgt binding assay described previously was used. Incubations were performed at 22°C in 1.2-ml polypropylene tubes arranged in a 96-well format, but they proceeded for 2 h. Membranes were incubated with a range of radioligand concentrations (approximately 0.04–4 nM) in a total volume of 30 µl of binding buffer supplemented with bovine serum albumin and protease inhibitors. Total (no peptide added) and nonspecific binding (defined using 10 µM -cobratoxin) was determined in triplicate, for each concentration of each ligand. After the initial incubation, 1 ml of isotonic binding buffer, supplemented with bovine serum albumin [0.1% (w/v)] was added to each tube, and the incubation was continued for a further 2 min at 22°C. The previously performed dissociation binding experiments indicated that this volumetric expansion step allows some of the nonspecific radioligand binding to dissociate, but it has no measurable effect on specific binding, increasing the signal-to-noise ratio of the binding assay. Binding reactions were then terminated by filtration, washed, and counted as described previously.

¹²⁵I--CtxArIB[V11L;V16A] Autoradiography, Effect of 7 Genotype. The preceding experiments demonstrated that ¹²⁵I--CtxArIB[V11L;V16A] binding to mouse brain membranes was saturable, showed relatively rapid kinetics, and it seemed to occur only at 7 nAChRs. To address this issue further, autoradiography was performed using sections from both wild-type and 7^-/- mouse brains.

Autoradiographic procedures were similar to those described previously for ¹²⁵I--CtxMII (Whiteaker et al., 2000b). Before incubation with ¹²⁵I₁--CtxArIB[V11L;V16A] or ¹²⁵I₂--CtxArIB[V11L;V16A], four series of sections from each mouse were incubated in binding buffer [144 mM NaCl, 1.5 mM KCl, 2 mM CaCl₂, 1 mM MgSO₄, 20 mM HEPES, and 0.1% bovine serum albumin (w/v), pH 7.5] plus 1 mM phenylmethylsulfonyl fluoride (to inactivate endogenous serine proteases) at 22°C for 15 min. For all binding reactions, the standard binding buffer was supplemented with bovine serum albumin [0.1% (w/v)], 5 mM EDTA, 5 mM EGTA, and 10 µg/ml each of aprotinin, leupeptin trifluoroacetate, and pepstatin A to protect the ligand from endogenous proteases. A series of sections was then incubated with 1 nM ¹²⁵I₁--CtxArIB[V11L;V16A] for 2 h at 22°C, whereas a second series was incubated under the same conditions with 1 nM ¹²⁵I₂--CtxArIB[V11L;V16A]. The remaining series were then used to determine nonspecific labeling by the two radioligands, in the presence of 10 µM -cobratoxin. After incubation, the slides were washed as follows: 30 s in binding buffer plus 0.1% (w/v) bovine serum albumin (22°C), 30 s in binding buffer plus 0.1% (w/v) bovine serum albumin (0°C), 5 s in 0.1x binding buffer plus 0.01% (w/v) bovine serum albumin (twice at 0°C), and twice at 0°C for 5 s in 5 mM HEPES, pH 7.5.

Sections were initially dried with a stream of air and then by overnight storage (22°C) under vacuum before exposure to Super Resolution phosphorimaging screens (PerkinElmer Life and Analytical Sciences; 2–4 day exposure) for image capture. Images were collected using a Packard Cyclone system (PerkinElmer Life and Analytical Sciences).

¹²⁵I--CtxArIB[V11L;V16A] Lot Lasting Tests. To test the stability of ¹²⁵I₁--CtxArIB[V11L;V16A] and ¹²⁵I₂--CtxArIB[V11L; V16A], saturation binding parameters were measured in a series of assays performed at weekly intervals. At each time point, for each radioligand, peak signal-to-noise ratio, binding affinity (K_d), and maximal binding (B_max) were measured. Saturation binding assays were performed as described previously for hippocampal membranes, but using midbrain membranes. Midbrain tissue was used in this case because this is a large brain region (providing sufficient tissue for this extended series of experiments), and it contains a moderate density of 7 nAChRs, permitting the collection of large (in terms of absolute counts) amounts of both specific and nonspecific binding from the same samples. In contrast, hippocampal membranes yielded very few nonspecific counts at very low ligand concentrations, making accurate determination of the nonspecific signal difficult.

I₁--CtxArIB[V11L;V16A] nAChR Subtype Selectivity Tests. It was apparent from the preceding autoradiographic experiments that moderate concentrations of ¹²⁵I₁--CtxArIB[V11L;V16A] only bind to 7 nAChRs. However, it was still necessary to investigate whether iodination resulted in increased affinity for non-7 nAChR subtypes. According, the affinity of I₁--CtxArIB[V11L;V16A] at native muscle (1β1) and neuronal (4β2^*, 6β2^*, and β4^*) subtype nAChRs was assessed using a series of inhibition binding assays. In addition, functional IC₅₀ was measured at 910 nAChRs heterologously expressed in Xenopus oocytes (because no well characterized native source of this nAChR subtype was available to us in a form that could be used for inhibition binding assays).

Inhibition of ¹²⁵I--Bgt binding to Torpedo membranes was used to measure the affinity of -CtxArIB[V11L;V16A] and its derivatives for muscle-type 1β1 nAChRs. Identical assay conditions were used as those described for ¹²⁵I--Bgt binding to hippocampal membranes.

The interaction between -CtxArIB[V11L;V16A] and its derivatives and 4β2^* nAChRs was probed using displacement of [³H]cytisine (5 nM, 21.2 Ci mmol^-1; PerkinElmer Life and Analytical Sciences) binding to mouse thalamic membrane preparations. Incubations were performed at 22°C for 1 h in polystyrene 96-well plates. A total volume of 100 µl of isotonic binding buffer supplemented with protease inhibitors was used for incubations. -CtxArIB[V11L; V16A] concentrations of 0.3 nM to 10 µM were used, and blanks were defined using 100 µM(-)-nicotine tartrate. Binding reactions were terminated by filtration onto single layers of Inotech 0.75-µm retention glass fiber filters that were soaked in 0.5% polyethylenimine for [³H]cytisine assays. Samples were washed as described for ¹²⁵I--Bgt binding. After addition of 1 ml of Budget Solve Scintillation Fluid (Research Products International, Mt. Prospect, IL) to each sample, bound radioligand was quantitated by liquid scintillation counting (at 45% efficiency), using a Packard 1600 TR liquid scintillation spectrometer.

Affinity for ¹²⁵I--CtxMII-binding nAChRs (6β2^* subtype; Gotti et al., 2005) was assessed using pooled membranes from mouse olfactory tubercle, striatum, and superior colliculus. The conditions were the same as described in Salminen et al. (2005) and very similar to those described previously for ¹²⁵I--Bgt binding, with the following modifications. ¹²⁵I--CtxMII (2200 Ci mmol^-1) was used at 0.5 nM, initial incubations proceeded for 2 h, and incubations were continued for 4 min after the dilution step. Bound radioligand was quantitated by gamma counting, as described previously.

The affinity of -CtxArIB[V11L;V16A] for native β4^* nAChRs was measured using inhibition of ¹²⁵I-epibatidine (200 pM; 2200 Ci mmol^-1; PerkinElmer Life and Analytical Sciences) binding in the presence of the β2^* nAChR-selective agonist A85380 [GenBank] (10 nM, unlabeled), as described in Whiteaker et al. (2000a). The inferior colliculus, interpeduncular nucleus, medial habenula, and olfactory bulbs contain relatively high proportions of β4^* nAChRs compared with other brain regions (Whiteaker et al., 2000a), so membranes from these regions were used in the present study. A total incubation volume of 30 µl of isotonic binding buffer supplemented with protease inhibitors was used, and incubations proceeded at 22°C for 2 h in polystyrene 96-well plates. Nonspecific binding was again defined in the presence of 100 µM(-)-nicotine tartrate, and binding reactions were terminated and washed as described for [³H]cytisine.

Finally, the affinity of -CtxArIB[V11L;V16A] for 910 nAChRs was measured using inhibition of ¹²⁵I-epibatidine (5 nM) binding to transfected Xenopus oocyte membranes. The specific activity of commercially available ¹²⁵I-epibatidine (PerkinElmer Life and Analytical Sciences) was reduced 10-fold (to 220 Ci mmol^-1) by isotopic dilution with nonradioactive iodoepibatidine (a kind gift from K. J. Kellar, Georgetown University, Washington, DC). The total incubation volume was 30 µl using isotonic binding buffer supplemented with protease inhibitors and 0.1 mM phenylmethylsulfonyl fluoride. Incubations proceeded at 22°C for 1 h in polystyrene 96-well plates. Nonspecific binding was defined in the presence of 1 µM unlabeled -Bgt, and binding reactions were terminated and washed as described for ¹²⁵I-epibatidine.

Data Analysis. A more detailed account of how the data analysis was performed is given in Supplemental Material. Because much of the analysis was performed using standard models, these equations are not given here, but they are provided in the Supplemental Material for readers who want to follow our approach more thoroughly. A less commonly used approach was used to determine the signal-to-noise ratio of the ligand binding assays, and this approach is presented in full here.

For binding association kinetics, data were fit to a single exponential increase to maximal model (eq. 1 in Supplemental Fig. 1), providing an observed (apparent) association rate. For binding dissociation kinetics, data were fit to a single exponential decrease model (eq. 2 in Supplemental Fig. 1). These kinetic parameters were also used, together with the ligand concentration used, to calculate K_d values for binding as the quotient of the apparent association and dissociation rates (eq. 3 in Supplemental Fig. 1).

Saturation of specific binding was fit using the Hill equation (eq. 4 in Supplemental Fig. 1), whereas nonspecific binding was modeled as a linear increase with ligand concentration (eq. 5 in Supplemental Fig. 1). Combining these representations allowed the signal-to-noise (specific-to-nonspecific binding) ratio to be described:

(1)

where B_max is specific binding at saturation, [L] is the ligand concentration, n_H is the Hill coefficient of specific binding, K_d is the affinity constant of specific binding, m is the slope of increasing nonspecific binding as ligand concentration [L] increases, and c is nonspecific binding at [L] = 0 (counter background).

Differentiation of eq. 1 in terms of x gives eq. 2:

(2)

The value of eq. 2 is zero at the value of [L] corresponding to peak signal-to-noise ratio. This value of [L] was determined for individual experiments by least-squares data fitting of eq. 2 using the parameters previously obtained from fitting specific binding to the Hill equation and nonspecific binding to a simple linear increase with concentration. In turn, the peak signal-to-noise ratios for individual experiments were then calculated by entering the relevant value of [L] into eq. 1.

Data fits were performed using nonlinear, least-squares curve fitting in SigmaPlot version 9.0.1 statistical analysis (t tests; ANOVA) was performed using SPSS version 15.0.1 (SPSS Inc., Chicago, IL). Both software packages were purchased from Systat Software, Inc. (Point Richmond, CA).

	Results

Top Abstract Materials and Methods Results Discussion References

 
Effects of -CtxArIB[V11L;V16A] His-Iodination. Both mono- and diiodinated -CtxArIB[V11L;V16A] derivatives were generated using nonradioactive iodination at the single histidine residue of the parent compound. This allowed the effects of iodination on -CtxArIB[V11L;V16A] affinity for 7 nAChRs to be assessed. Iodinated products were high-performance liquid chromatography purified (Supplemental Fig. 2), and toxin iodination was confirmed with matrix-assisted laser desorption time-of-flight mass spectrometry. The monoisotopic masses (in daltons) were as follows: monoiodo peptide, 2436.88 calculated, 2436.8 observed; and diiodo peptide, 2562.78 calculated, 2562.8 observed. As shown in Fig. 1, histidine iodination mildly increased affinity at 7 nAChRs as measured by inhibition of ¹²⁵I₁--Bgt (high specific activity, monoiodinated -Bgt) binding to mouse hippocampal membranes. This increase in affinity seemed to be progressive, and the diiodo derivative had significantly higher affinity for 7 nAChRs than did the parent compound (see figure legend for statistics). These K_i values may be marginally higher than would be predicted from functional data [-CtxArIB[V11L;V16A] functional IC₅₀ = 0.356 nM, corresponding to a predicted binding K_i value of 2.25 nM (Whiteaker et al., 2007), compared with 10.5 nM as measured here]. This presumably reflects the effects of the exceptionally slow dissociation kinetics of ¹²⁵I--Bgt, which make measuring true equilibrium affinity constants impossible.

View larger version (27K): [in this window] [in a new window]   Fig. 1. Effects of -CtxArIB[V11L;V16A] His-iodination. Nonradioactive mono- and diiodinated derivatives were generated by cold-iodination at H15 of -CtxArIB[V11L;V16A]. Peptides were tested for their ability to inhibit 125I1--Bgt (2 nM) binding to hippocampal membranes. Affinity values were as follows: native Ki = 10.5 ± 2.19 nM (filled squares), monoiodo Ki = 4.92 ± 2.32 nM (open circles); diiodo Ki = 1.96 ± 0.43 nM (filled circles); n = 3 in each case. One-way ANOVA indicated that diiodo -CtxArIB[V11L;V16A] had significantly higher 7 nAChR potency than the parent peptide (F[2,6] = 5.48, p = 0.044). Lines were generated by nonlinear least-squares curve fitting. r2 values were 0.996, 0.999, and 0.989 for the summary fits to data from native, monoiodo, and diiodo peptides, respectively.

View larger version (19K): [in this window] [in a new window]   Fig. 2. 125I--CtxArIB[V11L;V16A] association and dissociation kinetics. The association and dissociation time courses of specific 125I1--CtxArIB[V11L;V16A] and 125I2--CtxArIB[V11L;V16A] (1 nM in each case) labeling of hippocampal membranes were observed at 22°C. A, association kinetics. Bmax was measured for both radioligands, and it was statistically identical [Bmax monoiodo = 22.5 ± 3.4 fmol mg-1 (protein), diiodo = 24.1 ± 2.0 fmol mg-1; n = 3 for both, p = 0.75 by t test]. kobs values were also indistinguishable (kobs monoiodo = 0.027 ± 0.007 min-1, diiodo = 0.019 ± 0.003 min-1; n = 3 for both, p = 0.33 by t test). Lines were generated by nonlinear least-squares curve fitting. r2 values were 0.989 and 0.976 for the summary fits to data from monoiodo and diiodo peptides, respectively. B, dissociation kinetics. Initial binding (Bo) was measured for both radioligands, and it was statistically identical [Bo monoiodo = 30.4 ± 1.5 fmol mg-1 (protein), diiodo = 31.1 ± 1.7 fmol mg-1; n = 3 for both, p = 0.78 by t test]. koff values were significantly quicker for 125I1- versus 125I2--CtxArIB[V11L;V16A], however (koff monoiodo = 0.020 ± 0.001 min-1, diiodo = 0.0115 ± 0.003 min-1; n = 3 for both, p = 0.0039 by t test). Lines were generated by nonlinear least-squares curve fitting. r2 values were 0.997 and 0.991 for the summary fits to data from monoiodo and diiodo peptides, respectively.

View larger version (18K): [in this window] [in a new window]   Fig. 3. 125I--CtxArIB[V11L;V16A] saturation binding. A, total and nonspecific binding to mouse hippocampal membranes was measured using a range of 125I--CtxArIB[V11L;V16A] concentrations, and compared with that of 125I1--Bgt. Nonspecific binding was defined by addition of 10 µM -cobratoxin. Each point is the mean ± S.E.M. of three individual determinations. Lines were generated by nonlinear least-squares curve fitting. For total binding, r2 values were 0.999 for the summary fits for all three peptides tested. For nonspecific binding, r2 values were 0.992 for 125I1--CtxArIB[V11L;V16A] and 0.999 for the summary fits to data from 125I2--CtxArIB[V11L;V16A] and 125I1--Bgt. B, specific binding to mouse hippocampal membranes was determined as the difference between specific and nonspecific binding for each radioligand. Binding parameters (Bmax, nH, and Kd) were calculated by fitting individual determinations to the Hill equation. Bmax values were statistically identical for each compound [72.7 ± 6.8, 70.0 ± 4.0, and 69.7 ± 14.0 fmol mg-1 (protein) for 125I1--CtxArIB[V11L;V16A], 125I2--CtxArIB[V11L;V16A], and 125I1--Bgt, respectively; mean ± S.E.M., n = 3; one-way ANOVA gives F[2,6] = 0.031, p = 0.969], as were nH values (1.03 ± 0.08, 1.14 ± 0.10, and 1.23 ± 0.11; one-way ANOVA gives F[2,6] = 1.11, p = 0.388). Whereas the affinities of 125I1--CtxArIB[V11L;V16A] and 125I2--CtxArIB[V11L; V16A] specific binding to hippocampal membranes were indistinguishable from each other, that of 125I1--Bgt was significantly higher (Kd values 1.15 ± 0.13, 0.93 ± 0.13, and 0.52 ± 0.16 nM, respectively; one-way ANOVA gives F[2,6] = 5.11, p = 0.045, Tukey's HSD post hoc test confirms that 125I1--Bgt Kd value is lower). Lines were generated by nonlinear least-squares curve fitting. r2 values were 0.999, 0.996, and 0.998 for the summary fits for 125I1--CtxArIB[V11L;V16A], 125I2--CtxArIB[V11L;V16A], and 125I1--Bgt, respectively. C, specific-to-nonspecific ratios were also calculated for each compound, as described under Materials and Methods. Neither the ligand concentration at which peak signal-to-noise ratio occurred nor the ratio itself differed significantly between the radioligands, as assessed by one-way ANOVA (F[2,6] = 0.904, p = 0.454 and F[2,6] = 0.209, p = 0.817, respectively). Lines were generated by nonlinear least-squares curve fitting. r2 values were 0.971, 0.967, and 0.876 for the summary fits for 125I1--CtxArIB[V11L;V16A], 125I2--CtxArIB[V11L;V16A], and 125I1--Bgt, respectively.

View larger version (56K): [in this window] [in a new window]   Fig. 4. 125I--Bgt and 125I--CtxArIB[V11L;V16A] autoradiography, effect of 7 genotype. Autoradiography was performed using sections from both wild-type and 7-/- mouse brains, for 125I--Bgt (0.5 nM) and both 125I1--CtxArIB[V11L;V16A] and 125I2--CtxArIB[V11L;V16A] (1 nM). Although the greater specific activity of 125I2--CtxArIB[V11L;V16A] compared with 125I1--CtxArIB[V11L;V16A] resulted in more intense labeling with the former ligand, the patterns of labeling for all three radioligands were extremely similar in wild-type sections (left column). In 7-/- sections, in contrast, each radioligand produced only a low, uniform, nonspecific binding pattern (right column). Abbreviations for brain regions: A, amygdala; Cx1, layer I of cortex; Cx6, layer VI of cortex; Hp, hippocampus; ZI, zona incerta.

¹²⁵I--CtxArIB[V11L;V16A] Association and Dissociation Kinetics.

View larger version (16K): [in this window] [in a new window]   Fig. 5. 125I--CtxArIB[V11L;V16A] lot lasting tests. Saturation binding experiments were performed on mouse midbrain membranes at weekly intervals using 125I1--CtxArIB[V11L;V16A] and 125I2--CtxArIB[V11L; V16A]. At each time point, for each radioligand, peak signal-to-noise ratio, Kd, and Bmax were measured as described under Materials and Methods. All points represent the mean ± S.E.M. of two separate determinations. A, signal-to-noise ratio deteriorated over time for both ligands. One-way ANOVA indicated that assay signal-to-noise ratio changed significantly over time. For 125I1--CtxArIB[V11L;V16A], F[11,23] = 7.87, p = 0.001. Tukey's HSD post hoc test indicates that values in weeks 10 to 12 are significantly lower than in week 1 at significance level p < 0.05. For 125I2--CtxArIB[V11L;V16A], F[11,23] = 19.2, p < 0.001. Tukey's HSD post hoc test indicates that values in weeks 3 to 12 are significantly lower than in week 1 at significance level p < 0.05). Signal-to-noise ratio dropped more rapidly for 125I2--CtxArIB[V11L;V16A] (time constant = 0.200 weeks-1) than for 125I1--CtxArIB[V11L;V16A] (time constant = 0.076 weeks-1). Lines were generated by nonlinear least-squares curve fitting. r2 values were 0.793 and 0.975 for the summary fits for 125I1--CtxArIB[V11L;V16A] and 125I2--CtxArIB[V11L;V16A], respectively. B, affinity of both ligands remained unchanged over 12 weeks. Although a trend to higher Kd value over time was observed for 125I2--CtxArIB[V11L;V16A], one-way ANOVA indicated that no significant affinity changes occurred (for 125I1--CtxArIB[V11L;V16A], F[11,23] = 0.992, p = 0.502; for 125I2--CtxArIB[V11L;V16A], F[11,23] = 1.53, p = 0.239). Lines were generated by nonlinear least-squares curve fitting. r2 values were 0.504 and 0.357 for the summary fits for 125I1--CtxArIB[V11L;V16A] and 125I2--CtxArIB[V11L;V16A], respectively. C, maximal binding values changed over time for 125I2--CtxArIB[V11L; V16A] but not 125I1--CtxArIB[V11L;V16A]. The Bmax values measured with 125I1--CtxArIB[V11L;V16A] remained constant over 12 weeks (one-way ANOVA shows F[11,23] = 2.32, p = 0.082), but they fell significantly in the same period for 125I2--CtxArIB[V11L;V16A] (one-way ANOVA shows F[11,23] = 5.55, p = 0.002; Tukey's HSD post hoc test indicates that values in weeks 10 to 12 are significantly lower than in week 1 at significance level p <0.05). Lines were generated by nonlinear least-squares curve fitting. r2 values were 0.089 and 0.057 for the summary fits for 125I1--CtxArIB[V11L;V16A] and 125I2--CtxArIB[V11L;V16A], respectively.

View larger version (26K): [in this window] [in a new window]   Fig. 6. Inhibition binding at further nAChR subtypes -CtxArIB[V11L; V16A]. Nonradioactive mono-(open points) and diiodinated (filled points) derivatives were generated by cold-iodination at H15 of -CtxArIB[V11L; V16A]. Inhibition of ligand binding at non-7 nAChR subtypes was assessed using the same concentrations of -CtxArIB[V11L;V16A] His-iodinated derivatives as shown in Fig. 1. Subtypes probed were as follows: 1β1 (2 nM 125I--Bgt binding to Torpedo electric organ membranes; triangles), 4β2* (5 nM [3H]cytisine binding to thalamic membranes; diamonds), 6β2* (0.5 nM 125I--CtxMII binding to pooled olfactory tubercle, striatal and superior colliculus membrane; hexagons), 910 (5 nM 125I-epibatidine binding to transfected Xenopus oocyte membranes; circles), and β4* (0.2 nM 125I-epibatidine + 10 nM A85380 [GenBank] binding to either inferior colliculus, interpeduncular nucleus, or olfactory bulb membranes; the latter is shown here; inverted triangles). Points represent mean ± S.E.M. of three to four separate determinations. Lines are shown only to connect data points for clarity; no r2 values were generated for these data as a result.

¹²⁵I--CtxArIB[V11L;V16A] Saturation Binding.

¹²⁵I--CtxArIB[V11L;V16A] Autoradiography, Effect of 7 Genotype. The previous experiments demonstrated that ¹²⁵I--CtxArIB[V11L;V16A] binding to mouse brain membranes was both saturable and showed relatively rapid kinetics. The inhibition binding experiments also seemed to indicate that iodinated -CtxArIB[V11L;V16A] derivatives retained the ability of the parent compound to discriminate 7 from other nAChR subtypes. However, it was a remote possibility that ¹²⁵I--CtxArIB[V11L;V16A] might also label non-nAChR targets. To address this issue, autoradiography was performed using sections from both wild-type and 7^-/- mouse brains. An example is shown in Fig. 4, at the level of the hippocampus. In wild-type sections, the binding pattern (layered cortical binding, and intense labeling of hypothalamic nuclei, amygdala, and hippocampus) was typical of 7-specific nAChR labeling in mouse brain sections, as shown previously using ¹²⁵I--Bgt (Franceschini et al., 2002). As would be expected, given the greater specific activity of ¹²⁵I₂--CtxArIB[V11L;V16A] compared with ¹²⁵I₁--CtxArIB[V11L;V16A], the labeling was more intense with the former ligand. For both ligands, deletion of the nAChR 7 gene resulted in the complete loss of specific labeling across all brain regions (above a low, even level of nonspecific binding).

¹²⁵I--CtxArIB[V11L;V16A] Lot Lasting Tests. The rates of deterioration of ¹²⁵I₁--CtxArIB[V11L;V16A] and ¹²⁵I₂--CtxArIB[V11L;V16A] were monitored using saturation binding experiments, performed at weekly intervals. The following parameters were measured at each time point, for each radioligand: peak signal-to-noise ratio, K_d, and B_max.

As illustrated in Fig. 5, these parameters were most stable for ¹²⁵I₁--CtxArIB[V11L;V16A]. Both radioligands exhibited a drop in signal-to-noise performance over the time period investigated (12 weeks; Fig. 5A), but for ¹²⁵I₁--CtxArIB[V11L;V16A] performance only deteriorated significantly versus initial levels in weeks 10 to 12. In contrast, ¹²⁵I₂--CtxArIB[V11L;V16A] suffered from a significant loss of performance versus initial levels beginning in week 3, and it suffered further deterioration in the following weeks. For both ligands, the deterioration of signal-to-noise ratio was fit well using a single exponential decay model. The decay rate for ¹²⁵I₂--CtxArIB[V11L;V16A] signal-to-noise ratio (0.200 weeks^-1) was approximately 2.5 times faster than that for ¹²⁵I₁--CtxArIB[V11L;V16A] (0.0761 week^-1).

In contrast, little change in affinity was seen for either ligand during extended testing (Fig. 5B), with ¹²⁵I₁--CtxArIB[V11L;V16A] K_d values remaining essentially constant throughout the 12-week investigation. The K_d values measured for ¹²⁵I₂--CtxArIB[V11L;V16A] showed an upward trend (apparent lowering of affinity) with time, but this never reached significance. To some extent, this lack of significance may reflect the difficulty of measuring accurate K_d values at later time points, due to deteriorating assay performance. Regardless, apparent changes in K_d values were less dramatic than those seen for signal-to-noise ratio.

B_max measured using ¹²⁵I₁--CtxArIB[V11L;V16A] saturation binding remained similar, even after 12 weeks (Fig. 5C). In contrast a slight, but significant decrease in apparent B_max was observed when using ¹²⁵I₂--CtxArIB[V11L;V16A], with values in weeks 10 to 12 being lower than in week 1.

I--CtxArIB[V11L;V16A] Affinity at Non-7 nAChR Subtypes. The affinity of iodinated -CtxArIB[V11L;V16A] derivatives was also tested at non-7 nAChR subtypes, to determine whether they retained the 7 nAChR selectivity of the noniodinated parent compound. As shown in Fig. 6, iodinated derivatives of -CtxArIB[V11L;V16A] did not significantly inhibit binding at a variety of other known nAChR subtypes, including several that also bind -Bgt or MLA with high affinity. There may have been some slight inhibition of binding to 6β2^* and 910 nAChRs at very high (micromolar) concentrations, but both I--CtxArIB[V11L;V16A] derivatives retain a >100-fold preference for 7 over 6β2^* and 910 nAChRs, making them highly selective ligands.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
As outlined under Introduction, ¹²⁵I--Bgt, although a useful and informative tool, suffers from significant problems of specificity when used outside of the mammalian CNS, and exceptionally slow kinetics. In this series of studies, we describe the synthesis and characterization of mono- and diiodinated versions of CtxArIB[V11L;V16A]. These conotoxin derivatives, particularly ¹²⁵I₁-CtxArIB[V11L;V16A], address the problems associated with the use of ¹²⁵I--Bgt, providing greater specificity and much faster kinetics.

Earlier studies had provided two highly 7-selective (>300-fold versus other nAChR subtypes) derivatives of CtxArIB as potential lead compounds (-CtxArIB[V11L; V16A] and -CtxArIB[V11L;V16D]; Whiteaker et al., 2007). We chose the former on the basis of its higher 7 affinity (0.36 versus 1.09 nM IC₅₀ value versus 7 nAChR function; Whiteaker et al., 2007), which would reduce the potential impact of any iodination-induced affinity loss. All of the -CtxArIB derivatives that we had previously characterized contained a single histidine residue at position 15 (H15), providing a single site for iodination. This is in contrast to -CtxMII, which hosts two naturally occurring histidines. Thus, histidine iodination of -CtxMII would produce a confusing mix of iodination products (a problem solved by addition of an N-terminal tyrosine and selective iodination at this newly introduced residue; Whiteaker et al., 2000b). Initial concerns that iodination at H15 would result in a loss of 7 affinity proved unfounded. In fact, iodination at this position mildly increased affinity for 7 nAChRs, with a progressive increase in affinity from unlabeled through monoiodinated to diiodinated H15 -CtxArIB[V11L;V16A]. Because addition of bulky, relatively nonpolar iodine atom(s) at the H15 position was well tolerated (or even advantageous), this site may be suitable for the introduction of substituents with similar properties (photoactivatable or fluorescent labels, biotin).

Kinetics experiments illustrated a further advantage of ¹²⁵I--CtxArIB[V11L;V16A] over ¹²⁵I--Bgt, that of relatively rapid association and dissociation. The dissociation rate of ¹²⁵I--Bgt from 7 nAChRs has been measured as 0.00074 min^-1 at 20°C (Salvaterra and Mahler, 1976), which is approximately 26 and 16 times slower than the rates measured for ¹²⁵I₁--CtxArIB[V11L;V16A] and ¹²⁵I₂--CtxArIB[V11L;V16A], respectively. In contrast, the true association rates (k_on) of both ¹²⁵I--CtxArIB[V11L;V16A] ligands were measured as 7.5 x 10⁶ M^-1 min^-1, which is only 3.3 times faster than that reported for the association of ¹²⁵I--Bgt to 7 nAChRs (2.3 x 10⁶ M^-1 min^-1; Salvaterra and Mahler, 1976). In practical terms, the faster dissociation of ¹²⁵I--CtxArIB[V11L;V16A] compared with ¹²⁵I--Bgt is largely responsible for the lower 7 nAChR affinity of ¹²⁵I--CtxArIB[V11L;V16A] derivatives compared with ¹²⁵I--Bgt, and it makes equilibrium binding assays with ¹²⁵I--CtxArIB[V11L;V16A] derivatives feasible. The benefits of being able to perform true equilibrium binding assays can be seen when comparing the K_i values calculated for I--CtxArIB[V11L;V16A] derivatives using ¹²⁵I--Bgt inhibition binding (4.92 ± 2.32 and 1.96 ± 0.43 nM for mono- and diiodo derivatives, respectively) to the K_d values observed from direct ¹²⁵I--CtxArIB[V11L;V16A] saturation binding (1.15 ± 0.13 and 0.93 ± 0.13 nM, respectively). The saturation binding values are closer to those calculated using kinetic analysis (2.67 and 1.51 nM).

In hippocampal saturation binding experiments ¹²⁵I₁--CtxArIB[V11L;V16A], ¹²⁵I₂--CtxArIB[V11L;V16A], and ¹²⁵I₁--Bgt all bound to the same number of sites, as would be expected if each ligand was binding specifically to a single nAChR population (7 nAChRs in this case). The saturation binding experiments also showed that both ¹²⁵I₁--CtxArIB[V11L;V16A] and ¹²⁵I₂--CtxArIB[V11L;V16A] have lower nonspecific binding than ¹²⁵I₁--Bgt at equivalent concentrations (with ¹²⁵I₁--CtxArIB[V11L;V16A] seeming to have the cleanest binding of all in this comparison). Despite this, maximal signal-to-noise ratios were similar for each radioligand, with increased nonspecific binding being approximately canceled out by higher affinity along the series ¹²⁵I₁--CtxArIB[V11L;V16A] to ¹²⁵I₂--CtxArIB[V11L;V16A] to ¹²⁵I₁--Bgt. Thus, the real-world assay performance of each of these radioligands (when newly synthesized) is very similar.

As assessed by 7 affinity, binding kinetics, and assay performance, the two ¹²⁵I--CtxArIB[V11L;V16A] derivatives are quite similar to each other when newly synthesized. However, significant differences were uncovered when the deterioration of ¹²⁵I₁--CtxArIB[V11L;V16A] and ¹²⁵I₂--CtxArIB[V11L;V16A] lots was compared. From this point of view, ¹²⁵I₁--CtxArIB[V11L;V16A] is clearly a better radioligand because its signal-to-noise ratio performance drops more slowly, and no changes in B_max or K_d were measured over a 12-week period, in contrast to the situation with ¹²⁵I₂--CtxArIB[V11L;V16A]. Nevertheless, in certain well defined circumstances (very low 7 nAChR expression in which extremely high specific activity would be needed to obtain a measurable signal, and when the radioligand could be used within 2 weeks of being synthesized), ¹²⁵I₂--CtxArIB[V11L; V16A] could be a very useful alternative to ¹²⁵I₁--CtxArIB[V11L;V16A]. It is not known exactly why ¹²⁵I₂--CtxArIB[V11L;V16A] undergoes more rapid and more extensive deterioration than does ¹²⁵I₁--CtxArIB[V11L;V16A]. Part of the explanation may simply be the higher radioactive concentration and thus radiolysis rate in the ¹²⁵I₂--CtxArIB[V11L;V16A] solution (both peptides were stored at similar concentrations, giving the ¹²⁵I₂--CtxArIB[V11L;V16A] stock approximately double the radioactive concentration compared with the ¹²⁵I₁--CtxArIB[V11L;V16A] stock). In addition, studies of diiodo-insulin illustrate a further mechanism by which the assay performance of diiodo peptides may suffer rapid deterioration. Decay of the first radioiodine on diiodo-insulin molecules results in a mix of decay products, including free radioiodide and radiolabeled polymers that exhibit low receptor affinity and high nonspecific binding (Pérez-Maceda et al., 1982). Accumulation of either or both of these products could be responsible for the relatively rapid deterioration of ¹²⁵I₂--CtxArIB[V11L;V16A] samples seen in this study.

It was also possible that iodination of -CtxArIB[V11L; V16A] could result in a loss of the selective interaction of the parent compound with 7 nAChRs. Again, this concern proved unfounded, as assessed using two separate approaches. First, as illustrated in Fig. 6, the mono-iodinated -CtxArIB[V11L;V16A] derivative failed to interact with other nAChR subtypes (1β1, 4β2^*, 6β2^*, β4^*, and 910) with anything like the same affinity seen at 7 nAChRs. Second, all ¹²⁵I--CtxArIB[V11L;V16A] binding to mouse brain sections was abolished by nAChR 7 gene deletion. Thus, both the mono- and diiodo versions of ¹²⁵I--CtxArIB[V11L;V16A] retain the exceptional 7 selectivity of the parent compound, which is greater than that of -Bgt.

	Acknowledgements

 
We thank J. E. Rivier and R. Kaiser (Salk Institute for Biological Sciences) for their assistance with mass spectrometry.

	Footnotes

 
This work was supported by National Institutes of Health Grants DA12242 (to P.W.), MH53631 and GM48677 (to J.M.M.), and DA-15663 (to A.C.C.).

Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.

doi:10.1124/jpet.108.136895.

ABBREVIATIONS: -Bgt, -bungarotoxin; nAChR, nicotinic acetylcholine receptor; CNS, central nervous system; MLA, methyllycaconitine; -Ctx, -conotoxin; I₁, monoiodo; I₂, diiodo; TFA, trifluoroacetic acid; ANOVA, analysis of variance; HSD, honestly significant difference; A85380 [GenBank] , [3-(2(S)-azetidinylmethoxy)pyridine].

The online version of this article (available at http://jpet.aspetjournals.org) contains supplemental material.

Address correspondence to: Dr. J. Michael McIntosh, University of Utah, 257 S. 1400 East, Salt Lake City, UT 84112. E-mail: mcintosh.mike{at}gmail.com

	References

Top Abstract Materials and Methods Results Discussion References

 

Baker ER, Zwart R, Sher E, and Millar NS (2004) Pharmacological properties of 9 10 nicotinic acetylcholine receptors revealed by heterologous expression of subunit chimeras. Mol Pharmacol 65: 453-460.[Abstract/Free Full Text]
Changeux JP, Kasai M, and Lee CY (1970) Use of a snake venom toxin to characterize cholinergic receptor protein. Proc Natl Acad Sci U S A 67: 1241-1247.[Abstract/Free Full Text]
Chen DN and Patrick JW (1997) The -bungarotoxin-binding nicotinic acetylcholine receptor from rat brain contains only the 7 subunit. J Biol Chem 272: 24024-24029.[Abstract/Free Full Text]
Clarke PBS (1992) The fall and rise of neuronal alpha-bungarotoxin binding-proteins. Trends Pharmacol Sci 13: 407-413.[CrossRef][Medline]
Clarke PBS, Schwartz RD, Paul SM, Pert CB, and Pert A (1985) Nicotinic binding in rat-brain–autoradiographic comparison of [³H]acetylcholine, [³H]nicotine, and [¹²⁵I]-bungarotoxin. J Neurosci 5: 1307-1315.[Abstract]
Davies ARL, Hardick DJ, Blagbrough IS, Potter BVL, Wolstenholme AJ, and Wonnacott S (1999) Characterisation of the binding of [³H]methyllycaconitine: a new radioligand for labelling 7-type neuronal nicotinic acetylcholine receptors. Neuropharmacology 38: 679-690.[CrossRef][Medline]
Elgoyhen AB, Vetter DE, Katz E, Rothlin CV, Heinemann SF, and Boulter J (2001) 10: a determinant of nicotinic cholinergic receptor function in mammalian vestibular and cochlear mechanosensory hair cells. Proc Natl Acad Sci U S A 98: 3501-3506.[Abstract/Free Full Text]
Franceschini D, Paylor R, Broide R, Salas R, Bassetto L, Gotti C, and De Biasi M (2002) Absence of alpha 7-containing neuronal nicotinic acetylcholine receptors does not prevent nicotine-induced seizures. Mol Brain Res 98: 29-40.[Medline]
Gotti C, Hanke W, Maury K, Moretti M, Ballivet M, Clementi F, and Bertrand D (1994) Pharmacology and biophysical properties of 7 and 7--8 alpha-bungarotoxin receptor subtypes immunopurified from the chick optic lobe. Eur J Neurosci 6: 1281-1291.[CrossRef][Medline]
Gotti C, Moretti M, Clementi F, Riganti L, McIntosh JM, Collins AC, Marks MJ, and Whiteaker P (2005) Expression of nigrostriatal 6-containing nicotinic acetylcholine receptors is selectively reduced, but not eliminated, by β3 subunit gene deletion. Mol Pharmacol 67: 2007-2015.[Abstract/Free Full Text]
Haberberger RV, Bernardini N, Kress M, Hartmann P, Lips KS, and Kummer W (2004) Nicotinic acetylcholine receptor subtypes in nociceptive dorsal root ganglion neurons of the adult rat. Auton Neurosci 113: 32-42.[CrossRef][Medline]
Marks MJ, Smith KW, and Collins AC (1998) Differential agonist inhibition identifies multiple epibatidine binding sites in mouse brain. J Pharmacol Exp Ther 285: 377-386.[Abstract/Free Full Text]
Martyn JAJ and Richtsfeld M (2006) Succinylcholine-induced hyperkalemia in acquired pathologic states–etiologic factors and molecular mechanisms. Anesthesiology 104: 158-169.[CrossRef][Medline]
Morley BJ, Li HS, Hiel H, Drescher DG, and Elgoyhen AB (1998) Identification of the subunits of the nicotinic cholinergic receptors in the rat cochlea using RT-PCR and in situ hybridization. Mol Brain Res 53: 78-87.[Medline]
Nguyen VT, Ndoye A, and Grando SA (2000) Novel human 9 acetylcholine receptor regulating keratinocyte adhesion is targeted by pemphigus vulgaris autoimmunity. Am J Pathol 157: 1377-1391.[Abstract/Free Full Text]
Parker MJ, Beck A, and Luetje CW (1998) Neuronal nicotinic receptor β2 and β4 subunits confer large differences in agonist binding affinity. Mol Pharmacol 54: 1132-1139.[Abstract/Free Full Text]
Pauly JR, Stitzel JA, Marks MJ, and Collins AC (1989) An autoradiographic analysis of cholinergic receptors in mouse-brain. Brain Res Bull 22: 453-459.[CrossRef][Medline]
Peng HS, Ferris RL, Matthews T, Hiel H, Lopez-Albaitero A, and Lustig LR (2004) Characterization of the human nicotinic acetylcholine receptor subunit 9 (CHRNA9) and 10 (CHRNAIO) in lymphocytes. Life Sci 76: 263-280.[CrossRef][Medline]
Pérez-Maceda B, Linde S, Sonne O, and Gliemann J (1982) [¹²⁵I]-labeled diiodoinsulins–binding affinities, biologic potencies, and properties of their decay products. Diabetes 31: 634-640.[Medline]
Quik M, Sum JD, Whiteaker P, McCallum SE, Marks MJ, Musachio J, McIntosh JM, Collins AC, and Grady SR (2003) Differential declines in striatal nicotinic receptor subtype function after nigrostriatal damage in mice. Mol Pharmacol 63: 1169-1179.[Abstract/Free Full Text]
Salminen O, Whiteaker P, Grady SR, Collins AC, McIntosh JM, and Marks MJ (2005) The subunit composition and pharmacology of -conotoxin MII-binding nicotinic acetylcholine receptors studied by a novel membrane-binding assay. Neuropharmacology 48: 696-705.[CrossRef][Medline]
Salvaterra PM and Mahler HR (1976) Nicotinic Acetylcholine-receptor from rat-brain–solubilization, partial-purification, and characterization. J Biol Chem 251: 6327-6334.[Abstract/Free Full Text]
Whiteaker P, Christensen S, Yoshikami D, Dowell C, Watkins M, Gulyas J, Rivier J, Olivera BM, and McIntosh JM (2007) Discovery, synthesis, and structure activity of a highly selective 7 nicotinic acetylcholine receptor antagonist. Biochemistry 46: 6628-6638.[CrossRef][Medline]
Whiteaker P, Jimenez M, McIntosh JM, Collins AC, and Marks MJ (2000a) Identification of a novel nicotinic binding site in mouse brain using [¹²⁵I]-epibatidine. Br J Pharmacol 131: 729-739.[CrossRef][Medline]
Whiteaker P, McIntosh JM, Luo SQ, Collins AC, and Marks MJ (2000b) ¹²⁵I--conotoxin MII identifies a novel nicotinic acetylcholine receptor population in mouse brain. Mol Pharmacol 57: 913-925.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: jpet.108.136895v1 325/3/910    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Whiteaker, P. Articles by McIntosh, J. M. Search for Related Content PubMed PubMed Citation Articles by Whiteaker, P. Articles by McIntosh, J. M.