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

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.056002v1 307/2/682    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 HighWire Citing Articles via Google Scholar Google Scholar Articles by Lindner, M. D. Articles by Gribkoff, V. K. Search for Related Content PubMed PubMed Citation Articles by Lindner, M. D. Articles by Gribkoff, V. K.

BEHAVIORAL PHARMACOLOGY

An Assessment of the Effects of Serotonin 6 (5-HT6) Receptor Antagonists in Rodent Models of Learning

Bristol-Myers Squibb Pharmaceutical Research Institute, Neuroscience Biology, Wallingford, Connecticut

Received June 23, 2003; accepted August 6, 2003.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Antagonists of serotonin 6 (5-HT6) receptors have been reported to enhance cognition in animal models of learning, although this finding has not been universal. We have assessed the therapeutic potential of the specific 5-HT6 receptor antagonists 4-amino-N-(2,6-bis-methylamino-pyrimidin-4-yl)-benzenesulfonamide (Ro 04-6790) and 5-chloro-N-(4-methoxy-3-piperazin-1-yl-phenyl)-3-methyl-2-benzothiophenesulfonamide (SB-271046) in rodent models of cognitive function. Although mice express the 5-HT6 receptor and the function of this receptor has been investigated in mice, all reports of activity with 5-HT6 receptor antagonists have used rat models. In the present study, receptor binding revealed that the pharmacological properties of the mouse receptor are different from the rat and human receptor: Ro 04-6790 does not bind to the mouse 5-HT6 receptor, so all in vivo testing included in the present report was conducted in rats. We replicated previous reports that 5-HT6 receptor antagonists produce a stretching syndrome previously shown to be mediated through cholinergic mechanisms, but Ro 04-6790 and SB-271046 failed to attenuate scopolamine-induced deficits in a test of contextual fear conditioning. We also failed to replicate the significant effects reported previously in both an autoshaping task and in a version of the Morris water maze. The results of our experiments are not consistent with previous reports that suggested that 5-HT6 antagonists might have therapeutic potential for cognitive disorders.

In addition, there is suggestive evidence that atypical antipsychotics may attenuate cognitive deficits in patients with schizophrenia, perhaps through their action as 5-HT6 receptor antagonists. Atypical antipsychotics have very high affinities for 5-HT6 receptors and block stimulation of adenylyl cyclase activity produced by serotonin (Sebben et al., 1994). They enhance extracellular levels of glutamate in the frontal cortex (Daly and Moghaddam, 1993), and chronic treatment with atypical antipsychotics decreases 5-HT6 receptor expression in the hippocampus (Frederick and Meador-Woodruff, 1999). Most patients with schizophrenia have cognitive deficits (Meltzer and McGurk, 1999), and atypical anti-psychotics attenuate these cognitive deficits (Purdon et al., 2000). Typical antipsychotics such as haloperidol do not mediate their effects through the 5-HT6 receptor (Bourson et al., 1995, 1998; Bentley et al., 1999; Frederick and Meador-Woodruff, 1999), nor do they attenuate cognitive deficits in schizophrenia patients (Purdon et al., 2000). Together, these results may suggest that the cognitive effects of atypical antipsychotics may be mediated by their action as 5-HT6 receptor antagonists.

Finally, several studies have reported that specific 5-HT6 receptor antagonists improve learning and memory in animal models. Analogs of the selective 5-HT6 receptor antagonist Ro 04-6790 attenuated scopolamine-induced deficits in a passive avoidance task (Bos et al., 2001). Ro 04-6790 also increased acquisition and consolidation in normal young rats in an operant autoshaping task, and it attenuated scopolamine-induced deficits in this task (Meneses, 2001). The selective 5-HT6 receptor antagonists Ro 04-6790, SB-271046, and SB-357134 all increased retention of a spatial mapping Morris water maze task in normal young rats (Rogers and Hagan, 2001; Woolley et al., 2001; Stean et al., 2002).

In contrast to all the studies that support the therapeutic potential of 5-HT6 receptor antagonists, Russell and Dias (2002) reported that they were unable to replicate any of the therapeutic effects of 5-HT6 receptor antagonists. The objective of the studies included in the present report was to assess the therapeutic potential of 5-HT6 receptor antagonists for enhancing cognitive function in rodent models. Both rats and mice express the 5-HT6 receptor, and both species have been used to investigate the function of this receptor. Initially, we planned to assess 5-HT6 receptor antagonists in both mouse and rat models of cognitive behavior. However, although mice have been used to assess the therapeutic potential of 5-HT6 receptor antagonists (Bourson et al., 1998), all positive effects with 5-HT6 receptor antagonists have been detected in rats, and our own initial work with mice failed to detect therapeutic effects with 5-HT6 receptor antagonists (data not shown). Examination of the literature revealed that four critical residues have been identified for ligand binding to the 5-HT6 receptor, and one of these four residues is different in the mouse receptor compared with rat and human receptor (Boess et al., 1998; Kohen et al., 2001). Therefore, the objective of our first experiment was to determine whether 5-HT6 receptor antagonists bind to the mouse receptor. This study revealed that the pharmacological properties of the mouse receptor are different from the rat and human receptor, so we subsequently conducted all efficacy tests in rats. In addition, because most of the published literature demonstrating the therapeutic potential of 5-HT6 receptor antagonists was produced with Ro 04-6790 or SB-271046, we focused our efforts on attempting to replicate the initial preclinical experiments using these two standards.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Animals. These studies were conducted in an animal care facility certified by the American Association for Accreditation of Laboratory Animal Care, and all experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee before the studies were initiated. All rats were from Harlan (Indianapolis, IN): male Sprague-Dawley, 285 ± 1.8 g, N = 297; male hooded Long-Evans, 282 ± 4.8 g, N = 69; and male Wistar, 360 ± 4.7 g, N = 160. Rats were housed in polycarbonate cages in a temperature-controlled room with a 12:12-h light/dark cycle. All rats were housed three to four per cage, and food and water was available ad libitum, except for the Wistar rats that were used in the food reward autoshaping task, which were singly housed and maintained on a restricted diet, 12 to 15 g/day of standard rat chow, until their body weights were approximately 85% of ad libitum fed rats, at which time behavioral testing was initiated.

Cloning and Transfection of Human, Rat, and Mouse 5-HT6 Receptors. The mouse and rat 5-HT6 receptors were obtained by polymerase chain reaction amplification using whole rat (adult male Sprague-Dawley) or mouse (adult male BALB/c) brain cDNA (BD Biosciences Clontech, Palo Alto, CA) followed by TA cloning, insert excision and purification, and finally cloning into the pCDNA3.1 vector (Invitrogen, Frederick, MD). The mouse insert was amplified using 0.4 µM each of the following primers: 5'-ATGGTTCCAGAGCCCGGCCCTGTCAAC-3' and 5'-TCAGTTCATGGGGGAACCAAGTGGATGCTG-3'. The rat insert was amplified using 0.4 µM each of the following primers: 5'-ATGGTTCCAGAGCCAGGCCCTGTCAAC-3' and 5'-CTCCAATGGCCAGCTCTTGACCTGGTCA-3'. The completed vectors were transformed into DH5 F-competent cells (Invitrogen) and a large-scale preparation of DNA was prepared using the Qiafilter plasmid maxi kit (QIAGEN, Valencia, CA) for sequence confirmation followed by transfection. The human 5-HT6 receptor was obtained from David R. Sibley (National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD). It was subcloned into pCDNA3.1 as well.

The constructs were transiently transfected into human embryonic kidney 293 cells using LipofectAMINE Plus (Invitrogen). Briefly, 1.5 x 10⁷ cells were plated into 150-mm tissue culture plates 24 h before transfection. Cells were transfected using 8 µg of DNA, 40 µl of Plus reagent, and 60 µl of LipofectAMINE reagent per plate according to manufacturer's instructions. After a 3-h transfection incubation, cells were fed fresh medium and were harvested 48 h later. Control transfections with a fluorescent protein demonstrated an 80% transfection efficiency.

Cells were rinsed, scraped up, and homogenized with a Polytron. The cell homogenate was centrifuged for 30 min at 30,000g. The pellet was resuspended in 50 mM Tris, pH 7.2, 1 mM EDTA plus mammalian protease inhibitor (Sigma-Aldrich, St. Louis, MO). Protein concentration was determined using bicinchoninic acid (Pierce Chemical, Rockford, IL).

5-HT6 Receptor Binding Studies. The protocol was adapted from a previous report (Boess et al., 1997). The B_max values for the human, rat, and mouse membrane preparations were 18,000, 11,400, and 4600 fmol/mg, respectively. The membranes (2, 5, and 15 µg of protein for the human, rat, and mouse receptors, respectively) were added to [³H]d-lysergic acid diethylamide (LSD) (84 Ci/mmol; Amersham Biosciences Inc., Piscataway, NJ) to start the binding assay in 50 mM Tris, 2 mM MgCl₂, pH 7.4. The [³H]LSD was diluted in 0.4% bovine serum albumin, 200 µM ascorbic acid and then diluted 1:10 in the samples to give a final concentration of 2 nM. Final concentrations of compounds ranged from 10^-6 to 10^-10 M and were dissolved in dimethyl sulfoxide with a final concentration of 1% dimethyl sulfoxide. Nonspecific binding was determined using 10 µM clozapine. The 96-well plates were shaken for 10 min at room temperature on a titer plate shaker (Lab Line Instruments, Melrose Park, IL) and then incubated for 1 h at 37°C. The samples were filtered through GF/B membranes (Whatman, Maidstone, UK) pre-treated with 0.5% polyethylenimine. The samples were immediately washed five times with 1.5 ml of ice-cold 20 mM Tris, pH 7.4. They were counted in a Microbeta Trilux 1450 scintillation counter (PerkinElmer Wallac, Turku, Finland).

Yawning/Stretching/Chewing. Rats were given vehicle injections and placed in individual, transparent chambers for 1 h each day for 4 d before the test day, to habituate them to the observation chambers and testing procedure. On the test day, rats were placed in the observation chambers immediately after drug administration and observed continuously for yawning, stretching, and chewing behaviors from 30 to 90 min after drug or vehicle injections (Ro 04-6790, 30.0 mg/kg; SB-271046, 30.0 mg/kg; physostigmine, 0.1 mg/kg). Average number of yawns, stretches, and vacuous chewing movements during the 1-h observation period were recorded as reported previously (Sleight et al., 1998; Bentley et al., 1999).

Conditioned Fear. Rats were first placed in individual sound-attenuating test chambers (MED Associates, St. Albans, VT) for a 6-min conditioning session, which consisted of 2 min of habituation, a tone/footshock pairing (30 s, 87-dB tone; 2 s, 2.5-mA footshock), followed by 2 min with no tone or shock, another tone/footshock pairing, and a final minute of no stimuli presentation. Control rats were dosed with vehicle and placed in the fear-conditioning box for 6 min but with no tone or shock. Rats were then removed from the test box and returned to their home cages. Percentage of time freezing was measured 24 h later in a 7-min contextual memory test in the same chambers with no tone or shock. Freezing behavior was quantified with Freezeview image analysis software (Actimetrics, Evanston, IL) using filter 25 and bout length of 0.75 s. SB-271046 solutions (1.0, 10.0, and 30.0 mg/kg p.o.) and scopolamine (1.0 mg/kg i.p.) were coadministered 1.5 h before the conditioning session. Ro 04-6790 (30.0 mg/kg i.p.) and scopolamine (1.0 mg/kg i.p.) were coadministered 30 min before the conditioning session. All drugs were administered before the conditioning session only.

Autoshaping. Ro 04-6790 was tested in an autoshaping procedure as reported previously (Meneses, 2001). On the 1st day of testing, rats were habituated to operant chambers (Coulbourn Instruments, Allentown, PA) by filling the food trough with 50 food pellets (45 mg/pellet). As soon as all the pellets were eaten, each rat was given 10 trials. During a trial, rats were presented with a retractable lever for 8 s, and the cage was illuminated by a house light located at the top of the cage directly above the lever. After 8 s, the lever was retracted, the house light was extinguished, and a 45-mg food pellet was delivered to the food trough. If the rat pressed the lever while it was extended, it was immediately retracted, the pellet was delivered, and the light was extinguished. After pellet delivery, there was a 60-s intertrial interval. The next day, each rat was tested by giving them 20 trials without the initial 50-pellet habituation period. Ro 04-6790 (0, 1, 5, or 10 mg/kg i.p., 1 ml/kg, in sterile saline) was administered immediately after training, and scopolamine (0.17 mg/kg i.p.) was administered 10 min after the training session, as reported previously (Meneses, 2001). The dependent measure was the percentage of bar presses during the test session. To overcome potential floor effects, another experiment was conducted in which rats were given additional daily test sessions, first using the same procedures as described above, and then under slightly different conditions to further facilitate the rate of acquisition, in which the lever remained extended for 30 s, and was retracted for 10 s, for 50 trials each day, as reported previously (Andrews et al., 1995).

Morris Water Maze. Ro 04-6790 was tested in the Morris water maze as reported previously (Woolley et al., 2001). Briefly, rats were administered saline vehicle or Ro 04-6790 at the optimal dose (30 mg/kg i.p., 1 ml/kg) 30 min before daily acquisition training. During acquisition training rats received three, 90 s trials per day for 3 days, to swim to a hidden platform, with 20 s on the platform at the end of each trial. They were then given one probe trial per day, without drug, 7, 10, and 14 days after the end of acquisition training. During probe trials, the hidden platform was removed, rats were allowed to swim for 60 s, and swimming duration within a 10-cm annulus around the former platform location was quantified. One experiment was conducted with Sprague-Dawley rats and another experiment was conducted with Long-Evans hooded rats.

SB-271046 was also tested in the Morris water maze as reported previously (Rogers and Hagan, 2001). Briefly, rats were dosed with vehicle or SB-271046 at the optimal dose (10 mg/kg p.o., 2 ml/kg in 1% methylcellulose) 2 h before acquisition and probe test sessions. Rats were tested in four 60-s acquisition trials on day 1, and six acquisition trials per day on days 2 to 5. Probe trials were conducted immediately after the last acquisition trial, and again 4, 7, and 10 days after the end of acquisition training. Latency to reach the platform was recorded during acquisition trials, and the percentage of time in the target quadrant was the dependent measure during probe trials.

Drugs. Ro 04-6790 and SB-271046 were synthesized at Bristol-Myers Squibb Co. (Wallingford, CT) and the structures confirmed using standard analytical methods. Scopolamine hydrobromide and physostigmine salicylate were purchased from Sigma-Aldrich. All compounds were dissolved in sterile saline and administered i.p. at 1 ml/kg, except SB-271046 in the conditioned fear and Morris water maze experiments, where it was suspended in 1% methyl cellulose and administered by oral gavage at 1 to 2 ml/kg.

Statistics. Data analyses were conducted with SAS. In the text and all figures, data are presented as means ± S.E.M. The yawning, stretching and chewing, conditioned fear and autoshaping data were analyzed with planned contrasts between the main control group and each treatment group, including repeated measures where appropriate. The data in the last autoshaping experiment was analyzed with a 2 x 2 analysis of variance, including Ro 04-6790 and scopolamine as factors in the analysis. In the Morris water maze experiments, differences between groups were analyzed separately for each trial during acquisition, as reported previously (Woolley et al., 2001). For probe trials with SB-271046, potential differences between groups in duration of time spent swimming in the target quadrant was analyzed separately for each probe trial, as reported previously (Rogers and Hagan, 2001). For probe trials with Ro 04-6790, performance was analyzed separately for each group and for each probe trial: duration of time in each quadrant was compared with the duration of time spent swimming in the target quadrant, as reported previously (Woolley et al., 2001).

	Results

Top Abstract Materials and Methods Results Discussion References

 
5-HT6 Receptor Binding Studies. Receptor binding studies were performed using recombinant receptors from human, rat, and mouse. Clozapine and methiothepin were shown to have similar K_i values at the 5-HT6 receptor from all three species. All the compounds tested, including SB-271046, Ro 04-6790, and clozapine, had potencies in the rat receptor that were similar to their potencies in the human receptors; however, a different pharmacology was observed for the mouse receptor (Fig. 1; Table 1). SB-271046 had a 4-fold lower affinity at the mouse receptor than at the rat receptor, and although Ro 04-6790 had a K_i of 23.07 nM in rat receptors, it failed to inhibit LSD binding in the mouse receptor up to 1 µM (Table 1). We do not know whether these changes are manifest in all mouse strains, but based on these results, all subsequent in vivo experiments were conducted in rats.

View larger version (12K): [in this window] [in a new window]   Fig. 1. Ro 04-6790, SB-271046, and clozapine inhibit [3H]LSD binding to human embryonic kidney 293 cell membranes expressing the human, rat, or mouse receptors.

Yawning/Stretching/Chewing. In an experiment with the optimal dose of SB-271046, physostigmine produced a significant increase in chewing behaviors, F(1,29) = 6.27, p = 0.02; and a trend toward increased yawning, F(1,29) = 2.19, p = 0.15. SB-271046 produced a significant increase in stretching behavior, F(1,29) = 6.55, p = 0.01 (Fig. 2). In an experiment examining the effects of the optimal dose of Ro 04-6790, physostigmine produced a significant increase in chewing, F(1,21) = 4.44, p = 0.04, and a trend toward increased yawning, F(1,21) = 3.21, p = 0.09; whereas Ro 04-6790 produced a significant increase in the number of stretches, F(1,21) = 8.72, p = 0.008, and vacuous chewing, F(1,21) = 4.44, p = 0.04 (Fig. 3).

View larger version (26K): [in this window] [in a new window]   Fig. 2. Yawning/stretching/chewing from 30 to 90 min after vehicle, physostigmine (0.1 mg/kg) or SB-271046 (30 mg/kg), N values = 10 to 11.

View larger version (20K): [in this window] [in a new window]   Fig. 3. Yawning/stretching/chewing from 30 to 90 min after vehicle, physostigmine (0.1 mg/kg), or Ro 04-6790 (30 mg/kg), N values = 8.

Conditioned Fear. In the conditioned fear experiment with Ro 04-6790, there were significant differences between 1) the no-shock controls and the shocked, vehicle-treated controls, F(1,75) = 32.52, p = 0.0001; 2) the scopolamine-treated group and the vehicle group, especially during the later time points; the treatment x minute interaction between those groups was statistically significant, F(6,450) = 2.18, p = 0.04; 3) the no-shock controls versus the scopolamine-treated group, F(1,75) = 15.20, p = 0.0002; and 4) the Ro 04-6790 group versus the scopolamine-treated group, F(1,75) = 5.58, p = 0.02 (Fig. 4A). Ro 04-6790 at 30 mg/kg reduced the amount of freezing beyond that seen in animals treated with scopolamine only, an effect which is consistent with impaired performance.

View larger version (33K): [in this window] [in a new window]   Fig. 4. Percentage of time spent freezing in 7-min conditioned fear test. SCOP, scopolamine; No-Shock Vehicle Controls, control rats not given any shock during conditioning session. A, Ro 04-6790 (30 mg/kg) did not attenuate scopolamine-induced deficits, N values = 19 to 20. B, SB-271046 did not attenuate scopolamine-induced deficits at any dose (1.0, 10.0, and 30.0 mg/kg), N values = 20 to 24.

In the conditioned fear experiment with SB-271046, there were significant differences between 1) the no-shock controls and the shocked, vehicle-treated controls, F(1,126) = 51.65, p = 0.0001; 2) the scopolamine-treated group and the vehicle group, F(1,126) = 5.18, p = 0.02; 3) the no-shock controls and the scopolamine-treated group, F(1,126) = 24.66, p = 0.0001; and 4) the SB-271046 30-mg/kg treated group and the scopolamine-treated group, F(1,126) = 6.93, p = 0.01 (Fig. 4B). Again, the effect of SB-271046 at 30 mg/kg was in the direction of impaired performance, relative to the group treated with scopolamine alone. Lower doses of SB-271046 did not produce significant effects, but even at 1.0 and 10.0 mg/kg, the trend was in the direction of impaired performance, relative to animals treated with scopolamine alone.

Autoshaping. In the first autoshaping experiment with Ro 04-6790, there were no significant increases in bar pressing in animals with any dose of Ro 04-6790, F values (1,73) < 1.27, p values > 0.26 (Fig. 5A). In the second autoshaping experiment examining the ability of Ro 04-6790 to attenuate the effects of scopolamine, dosing rats with 0.17 mg/kg scopolamine after the first session did not significantly reduce bar pressing 24 h later, and Ro 04-6790 did not increase bar-pressing rate on its own or in animals dosed with scopolamine, F values (1,43) < 1.0, p values > 0.40 (Fig. 5B). In the third autoshaping experiment, rats were repeatedly tested with scopolamine (0.17 mg/kg) and Ro 04-6790 administered after each session, using the dose of Ro 04-6790 (5 mg/kg) that previously produced peak effects (Meneses, 2001). To facilitate acquisition of bar pressing and overcome any potential floor effects, an additional session was conducted using the Meneses procedure, and five more sessions with 50 trials per day, with the duration of lever extension increased from 8 to 30 s. Rats did begin to acquire this task: bar pressing increased over days, and a scopolamine deficit was eventually evident, the day x scopolamine interaction was statistically significant, F(6,318) = 2.34, p = 0.03 (Fig. 6). Ro 04-6790 (5 mg/kg) did not facilitate acquisition or attenuate the scopolamine-related deficit, in fact, there was a trend for Ro 04-6790 to impair performance in both vehicle and scopolamine-treated rats, but that trend was not significant.

View larger version (24K): [in this window] [in a new window]   Fig. 5. Percentage of bar presses in the autoshaping task on the test day. All drugs were administered immediately after the conditioning trial, 24 h before the test trial. A, treatment groups were as follows: 0, vehicle (N = 13), or 1.0, 5.0, or 10.0 mg/kg Ro 04-6790 (N values = 21 to 22). B, Ro 04-6790 (5.0 mg/kg) combined with vehicle or with scopolamine (0.17 mg/kg), scopolamine alone, or vehicle alone (N values = 12).

View larger version (16K): [in this window] [in a new window]   Fig. 6. Percentage of bar presses in the autoshaping task with repeated test sessions. All drugs were administered after each test session. Scopolamine (0.17 mg/kg) significantly reduced acquisition of bar pressing, but Ro 04-6790 (5 mg/kg) only produced a nonsignificant trend in the direction of reduced bar pressing (N values = 12).

Morris Water Maze. In the test of Ro 04-6790 in both albino Sprague-Dawley and hooded Long-Evans rats, latencies to reach the target platform during acquisition trials improved over trials (Fig. 7). In the albino Sprague-Dawley rats, there was a significant difference between the vehicle and Ro 04-6790 only on the second trial of acquisition training, in which the drug treatment group had longer latencies than the vehicle-treated group, F(1,28) = 6.79, p = 0.01 (Fig. 7A). In the hooded Long-Evans rats, there was a significant difference between the vehicle and Ro 04-6790 only on the fourth trial of acquisition training, in which the drug-treatment group had longer latencies than the vehicle-treated group, F(1,27) = 6.89, p = 0.01 (Fig. 7B).

View larger version (24K): [in this window] [in a new window]   Fig. 7. Swim latencies during acquisition training in the Morris water maze with vehicle-treated or Ro 04-6790 (30 mg/kg), N values = 14-15. Asterisks indicate trials where Ro 04-6790-treated rats performed significantly different from vehicle-treated controls. A, albino Sprague-Dawley rats. B, hooded Long-Evans rats.

Analyses of probe trial data revealed no evidence that Ro 04-6790 increased retention of the target location in either albino Sprague-Dawley rats or hooded Long-Evans rats (Fig. 8). Although the hooded Long-Evans rats had more of a preference for the target annulus during the first probe trial, both albino Sprague-Dawley rats and hooded Long-Evans rats spent more time swimming in the target annulus on the first probe trial, 7 days after the end of acquisition training, regardless of whether they were in the vehicle or Ro 04-6790 group. The Ro 04-6790-treated Sprague-Dawleys may not have had quite as strong a preference for the target location as the vehicle-treated rats during the first probe trial, because they failed to discriminate between the target quadrant and one of the adjacent quadrants (Fig. 8A, top left). In the second probe trial, 10 days after the end of acquisition training, vehicle-treated Sprague-Dawley rats had no preference for the target annulus, whereas the Ro 04-6790-treated rats swam in the target annulus more than in one of the adjacent quadrants, but the trend was just the opposite in hooded Long-Evans rats, where only the vehicle-treated group swam more in the target annulus than in one of the adjacent quadrants. On the third probe trial, 14 days after the end of acquisition training, there was no preference to swim in the target annulus in any of the groups. In fact, the hooded Long-Evans rats treated with Ro 04-6790 actually exhibited a preference to swim in quadrants other than where the target had been located (Fig. 8B, bottom right).

View larger version (34K): [in this window] [in a new window]   Fig. 8. Swim duration in target annulus and equivalent annulus areas in other quadrants during probe trials with vehicle or Ro 04-6790 (30 mg/kg), N values = 14 to 15. Probe trials were conducted 7, 10, or 14 days after the end of acquisition training. Asterisks indicate significant differences between duration in the target annulus and each of the other quadrants, as determined by planned contrasts, p values 0.05. A, albino Sprague-Dawley rats. B, hooded Long-Evans rats.

Hooded Long-Evans rats rapidly acquired the Morris water maze task to asymptotic levels of performance, but SB-271046 did not affect latencies to find the Morris water maze, F values (1,38) < 1.4, p values > 0.25 (Fig. 9A). Rats also exhibited evidence that they were using spatial mapping strategies, because they swam almost 50% of the time in the target quadrant during the first probe trial immediately after the last acquisition trial (Fig. 9B). The spatial mapping strategy extinguished over repeated probe trials, as the rats spent less and less time swimming near the former target location. There was a trend for the vehicle-treated rats to spend more time swimming in the target quadrant on the third probe trial, and a trend for the SB-271046-treated rats to spend more time swimming in the target quadrant on the fourth probe trial, but there were no statistically significant differences between the vehicle-treated and the SB-271046-treated groups on any of the probe trials, F values (1,38) < 1.75, p > 0.20.

View larger version (21K): [in this window] [in a new window]   Fig. 9. Morris water maze testing with vehicle or SB-271046 (10 mg/kg), N values = 20. A, swim latencies during acquisition training. B, percentage of time spent swimming in the target quadrant during probe trials immediately after the end of acquisition training (0), or 4, 7, or 10 days after the end of acquisition training.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The results of the present experiments demonstrate, for the first time, that there are differences in the pharmacological properties of the mouse 5-HT6 receptor, relative to the human and rat receptor. Although the mouse, rat, and human receptors are fairly homologous (>84% identical), four critical residues had been identified for ligand binding to the 5-HT6 receptor, and one of these four residues is different in the mouse than in the rat and human receptor (Boess et al., 1998; Kohen et al., 2001). Site mutations of the rat receptor that alter the one residue that is unique to the mouse receptor, significantly affect ligand binding (Boess et al., 1998), but those site mutations were not identical to the mouse receptor sequence, and receptor binding had not been reported previously with the mouse 5-HT6 receptor. Receptor binding studies included in the present report show that both Ro 04-6790 and SB-271046 have high affinity to rat and human receptors, but Ro 04-6790 did not bind to the mouse receptor, and SB 271046 had a lower affinity at the mouse receptor than at the rat or human receptor. These results make it clear that mice should not be used to assess the therapeutic potential of 5-HT6 receptor antagonists unless the compounds are shown to bind to the mouse receptor, and the results of previous studies assessing compounds such as Ro 04-6790 in mice need to be reevaluated (Bourson et al., 1998).

Based on the results of the receptor binding experiments, all behavioral assessments included in the present report were conducted in rats. Our results replicated the finding that 5-HT6 receptor antagonists produce stretching, a behavioral syndrome mediated by cholinergic facilitation (Bourson et al., 1995; Sleight et al., 1996, 1998; Bentley et al., 1999). We detected stretching with both Ro 04-6790 and with SB-271046. Previous studies had not seen increased stretching with SB-271046 except to accentuate the effects produced by the acetylcholinesterase inhibitor physostigmine [J. Stretton, unpublished data; cf. Routledge (1999); Reavill and Rogers (2001)]. Our tests with SB-271046 may have been more sensitive to stretching behavior because we habituated our rats to the testing procedure on numerous occasions before conducting the testing. These habituation periods reduce activity levels during the test, which make it easier to observe stretching if it occurs.

In contrast to the stretching behavior, which is consistent with facilitation of cholinergic neurotransmission, none of our efforts to assess the therapeutic potential of 5-HT6 receptor antagonists on measures of cognitive function detected any positive effects. For example, the results of our experiment with Ro 04-6790 in the conditioned fear task were not consistent with the positive effects reported with analogs of Ro 04-6790 in the passive avoidance task (Bos et al., 2001). In another experiment, we were unable to detect significant, positive effects with SB-271046 in the conditioned fear test. Even after closely replicating the methods used previously with Ro 04-6790 and SB-271046 in an autoshaping task and in the Morris water maze (Meneses, 2001; Rogers and Hagan, 2001; Woolley et al., 2001), we failed to replicate any of the positive results reported in those studies. For example, we did not see evidence of improved acquisition or retention with Ro 04-6790 in the autoshaping task. Meneses reported bar-pressing rates of 10% for the vehicle-treated control group on the test day, and in our experiments, vehicle-treated controls pressed the bar on 6 to 7% of the 20 trials, which is within the expected range of Meneses' experiments, but it is so low that there are potential floor effects that may reduce the sensitivity of this test. However, even with repeated testing, no significant effects were detected for Ro 04-6790 in this task, either in normal rats or in rats with scopolamine-induced deficits. In addition, even if increased bar-pressing rates had been detected in this task, additional studies would need to be conducted to rule out potential nonspecific effects such as increased activity levels and/or disinhibition or impulsivity.

Likewise, we saw no evidence of increased retention with Ro 04-6790 in the Morris water maze using the same procedures reported previously (Woolley et al., 2001). The animals reached asymptotic levels during acquisition, they showed evidence of spatial mapping during retention trials, and performance declined with repeated probe trials, but there was no evidence that Ro 04-6790 improved acquisition or retention. If anything, Ro 04-6790 actually impaired performance during acquisition trials. It is not clear what the critical difference is between our experiments and the previous experiment that reported positive results. However, it is interesting to note that in the previous Morris water maze experiment, the group treated with Ro 04-6790 performed significantly better than the vehicle-treated controls during acquisition trials, but this difference seemed to be due to the fact that the vehicle-treated controls suddenly performed worse than expected during the last three acquisition trials. Whatever the cause, if the difference in performance between the vehicle-treated controls and the group dosed with Ro 04-6790 is due to uncharacteristically poor performance among the vehicle-treated controls, this should not be interpreted as evidence that the drug improved performance.

Finally, we were also unable to replicate the positive effects reported for SB-271046 in the Morris water maze, although we replicated as precisely as possible the methods used previously (Rogers and Hagan, 2001). Our results suggest that cognition enhancing effects of Ro 04-6790 and SB-271046 are not reliable. In addition, we agree with previous criticisms suggesting that the effects reported previously were not necessarily evidence of therapeutic potential (Russell and Dias, 2002). For example, the effect in both previous Morris water maze studies may have been attributed to perseveration, rather than due to increased retention (Russell and Dias, 2002). In other words, even if we had replicated the results of the previous studies showing prolonged searching for the previous target location, it would not be appropriate to conclude that this effect was evidence of improved cognitive function; it could also be attributed to perseveration or impaired cognitive function and additional experiments would have to be run to rule out that possibility.

Consistent with the results of the present experiments, other investigators have also reported difficulty in replicating the positive effects of Ro 04-6790 and SB-271046 in the Morris water maze (Russell and Dias, 2002), and in other models of cognitive function. Chronic i.c.v. administration of antisense oligonucleotides to the 5-HT6 receptor did not affect performance in a conditioned fear task (Yoshioka et al., 1998), and 5-HT6 receptor knockouts had no effect in a novel object recognition test (Martin et al., 1998). Not only are the therapeutic effects of 5-HT6 receptor antagonists in preclinical models of cognitive function in question, but virtually every piece of evidence that supports the therapeutic potential of 5-HT6 receptor antagonists for cognitive deficits is complicated by at least one report that fails to support it. For example, although some studies have suggested that 5-HT6 receptor antagonists enhance cholinergic and glutamatergic transmission, one study failed to detect an increase in hippocampal extracellular acetylcholine levels after administration of Ro 04-6790 (Shirazi-Southall et al., 2002), and another study failed to detect increases in glutamate release from frontal cortex after systemic or direct application of SB-271046 (Russell and Dias, 2002). The fact that hippocampal extracellular acetylcholine levels were elevated after administration of clozapine but not Ro 04-6790 (Shirazi-Southall et al., 2002) also suggests that atypical antipsychotics might not enhance cognitive function in patients with schizophrenia through their action on 5-HT6 receptors. Recent studies also suggest that there are no differences in 5-HT6 receptor binding or receptor densities in schizophrenic patients (East et al., 2002), and find neither an association between 5-HT6 receptor polymorphisms and susceptibility to schizophrenia (Shinkai et al., 1999; Ohmori et al., 2001) nor an association between 5-HT6 receptor polymorphisms and response to clozapine in schizophrenic patients (Masellis et al., 2001). Another study reported that there were no significant differences in genotypic or allelic distribution of 5-HT6 receptors among Alzheimer's disease patients and controls, which suggests that these polymorphisms probably do not represent major genetic risk factors for Alzheimer's disease (Thome et al., 2001; Orlacchio et al., 2002).

Several previous studies have reported linear dose-response curves with more and more robust, positive effects of Ro 04-6790 up to the highest dose tested, 30 mg/kg (Sleight et al., 1998; Bentley et al., 1999; Woolley et al., 2001). Ro 04-6790 also attenuated scopolamine-induced rotations with a maximal effect at 30 mg/kg (Bourson et al., 1998); therefore, we tested Ro 04-6790 at this optimal dose of 30 mg/kg during observations for stretching, in the conditioned fear task, and in the Morris water maze. Likewise, several studies have reported that consistent, robust effects were obtained with SB-271046 at 10 mg/kg (Dawson et al., 2000, 2001; Routledge et al., 2000), including tests for retention in the Morris water maze (Rogers and Hagan, 2001), so we tested SB-271046 at this optimal dose in the Morris water maze. Large sample sizes were tested with these optimal doses, and when we were unable to detect significant effects with Sprague-Dawley rats in the Morris water maze, we repeated the study with Long-Evans rats. The autoshaping task was also conducted with a very large sample size and testing was continued until we could be sure that the lack of significance was not due to a floor effect. All these decisions were made in an attempt to maximize the probability of either detecting a beneficial effect on cognitive function, or of replicating the results of previously published studies that reported statistically significant effects with 5-HT6 receptor antagonists.

Despite our best efforts to detect significant therapeutic effects on measures of cognitive function, it is impossible to prove that a treatment is inactive or that it has no therapeutic efficacy, and no matter how many studies are conducted, it is always possible that additional studies might still uncover some potential efficacy. For example, it is possible that the use of different doses or different dosing times, or the use of an experimental design that was not confounded with extinction effects, may have detected therapeutic effects of 5-HT6 receptor antagonists on measures of cognitive function. We would emphasize that our efforts to demonstrate efficacy and/or to replicate previous positive results were fairly extensive, and that we made every effort to maximize our chances of detecting therapeutic effects. However, we do not conclude from the present experiments that 5-HT6 receptor antagonists have no therapeutic efficacy. Instead, the negative results of the present experiments simply raise questions about the reliability and validity of the therapeutic potential of 5-HT6 receptor antagonists. For results to be accepted as valid, they must be reliable, and we would simply argue that the therapeutic potential of 5-HT6 receptor antagonists cannot be accepted as valid until they can be shown to produce therapeutic effects reliably.

The results of the present experiments suggest that 5-HT6 receptor antagonists may not have therapeutic potential for cognitive disorders. Alternatively, there may be some differences between the studies, which are critical for producing and detecting positive and potentially therapeutic effects. It is clearly impossible to prove that a treatment has no therapeutic potential, and replicating the procedures used in previous reports does not constitute an exhaustive assessment of the therapeutic potential of this target. However, knowledge about the reliability and robustness of the results in preclinical studies would help to more accurately assess the therapeutic potential of novel compounds and the predictive validity of the models. If a treatment such as 5-HT6 receptor antagonism ultimately does or does not have efficacy in the clinic, determining the differences between the preclinical studies which produced positive and negative results may allow us to determine which tests and approaches appear to have greater predictive validity.

	Acknowledgements

 
We acknowledge Lynn Balanda for performing the cell transfections and Dr. Cen Xu for help with the pharmacology.

	Footnotes

 
DOI: 10.1124/jpet.103.056002.

ABBREVIATIONS: 5-HT6, serotonin 6; LSD, d-lysergic acid diethylamide; Ro 04-6790, 4-amino-N-(2,6-bis-methylamino-pyrimidin-4-yl)-benzenesulfonamide; SB-271046, 5-chloro-N-(4-methoxy-3-piperazin-1-yl-phenyl)-3-methyl-2-benzothiophenesulfonamide; SB-357134-A, N-(2,5-dibromo-3-fluorophenyl)-4-methoxy-3-piperazin-1ylbenzenesulfonamide hydrochloride.

Address correspondence to: Dr. Mark D. Lindner, Bristol-Myers Squibb Pharmaceutical Research Institute, Department of Neuroscience, 5 Research Parkway, Wallingford, CT 06492. E-mail: mark.lindner{at}bms.com

	References

Top Abstract Materials and Methods Results Discussion References

 

Andrews JS, Jansen JHM, Linders S, Princen A, and Broekkamp CLE (1995) Performance of four different rat strains in the autoshaping, two-object discrimination and swim maze tests of learning and memory. Physiol Behav 57: 785-790.[CrossRef][Medline]

Bentley JC, Bourson A, Boess FG, Fone KC, Marsden CA, Petit N, and Sleight AJ (1999) Investigation of stretching behaviour induced by the selective 5-HT6 receptor antagonist, Ro 04-6790, in rats. Br J Pharmacol 126: 1537-1542.[CrossRef][Medline]

Boess FG, Monsma FJ Jr, Meyer V, Zwingelstein C, and Sleight AJ (1997) Interaction of tryptamine and ergoline compounds with threonine 196 in the ligand binding site of the 5-hydroxytryptamine6 receptor. Mol Pharmacol 52: 515-523.[Abstract/Free Full Text]

Boess FG, Monsma FJ Jr, and Sleight AJ (1998) Identification of residues in transmembrane regions III and VI that contribute to the ligand binding site of the serotonin 5-HT6 receptor. J Neurochem 71: 2169-2177.[Medline]

Bos M, Sleight AJ, Godel T, Martin JR, Riemer C, and Stadler H (2001) 5-HT(6) receptor antagonists: lead optimisation and biological evaluation of N-aryl and N-heteroaryl 4-amino-benzene sulfonamides. Eur J Med Chem 36: 165-178.[CrossRef][Medline]

Bourson A, Boess FG, Bos M, and Sleight AJ (1998) Involvement of 5-HT6 receptors in nigro-striatal function in rodents. Br J Pharmacol 125: 1562-1566.[CrossRef][Medline]

Bourson A, Borroni E, Austin RH, Monsma FJ Jr, and Sleight AJ (1995) Determination of the role of the 5-HT(6) receptor in the rat brain: a study using antisense oligonucleotides. J Pharmacol Exp Ther 274: 173-180.[Abstract/Free Full Text]

Daly DA and Moghaddam B (1993) Actions of clozapine and haloperidol on the extracellular levels of excitatory amino acids in the prefrontal cortex and striatum of conscious rats. Neurosci Lett 152: 61-64.[CrossRef][Medline]

Dawson LA, Nguyen HQ, and Li P (2000) In vivo effects of the 5-HT(6) antagonist SB-271046 on striatal and frontal cortex extracellular concentrations of noradrenaline, dopamine, 5-HT, glutamate and aspartate. Br J Pharmacol 130: 23-26.[CrossRef][Medline]

Dawson LA, Nguyen HQ, and Li P (2001) The 5-HT(6) receptor antagonist SB-271046 selectively enhances excitatory neurotransmission in the rat frontal cortex and hippocampus. Neuropsychopharmacology 25: 662-668.[CrossRef][Medline]

East SZ, Burnet PW, Leslie RA, Roberts JC, and Harrison PJ (2002) 5-HT6 receptor binding sites in schizophrenia and following antipsychotic drug administration: autoradiographic studies with [125I]SB-258585. Synapse 45: 191-199.[CrossRef][Medline]

Frederick JA and Meador-Woodruff JH (1999) Effects of clozapine and haloperidol on 5-HT6 receptor mRNA levels in rat brain. Schizophr Res 38: 7-12.[CrossRef][Medline]

Kohen R, Fashingbauer LA, Heidmann DE, Guthrie CR, and Hamblin MW (2001) Cloning of the mouse 5-HT6 serotonin receptor and mutagenesis studies of the third cytoplasmic loop. Brain Res Mol Brain Res 90: 110-117.[Medline]

Martin GR, Eglen RM, Hamblin MW, Hoyer D, and Yocca F (1998) The structure and signalling properties of 5-HT receptors: an endless diversity? Trends Pharmacol Sci 19: 2-4.[CrossRef][Medline]

Masellis M, Basile VS, Meltzer HY, Lieberman JA, Sevy S, Goldman DA, Hamblin MW, Macciardi FM, and Kennedy JL (2001) Lack of association between the T C 267 serotonin 5-HT6 receptor gene (HTR6) polymorphism and prediction of response to clozapine in schizophrenia. Schizophr Res 47: 49-58.[CrossRef][Medline]

Meltzer HY and McGurk SR (1999) The effects of clozapine, risperidone and olanzapine on cognitive function in schizophrenia. Schizophr Bull 25: 233-255.

Meneses A (2001) Effects of the 5-HT(6) receptor antagonist Ro 04-6790 on learning consolidation. Behav Brain Res 118: 107-110.[CrossRef][Medline]

Monsma FJ Jr, Shen Y, Ward RP, Hamblin MW, and Sibley DR (1993) Cloning and expression of a novel serotonin receptor with high affinity for tricyclic psychotropic drugs. Mol Pharmacol 43: 320-327.[Abstract]

Ohmori O, Shinkai T, Hori H, and Nakamura J (2001) Novel polymorphism in the 5'-upstream region of the human 5-HT6 receptor gene and schizophrenia. Neurosci Lett 310: 17-20.[CrossRef][Medline]

Orlacchio A, Kawarai T, Paciotti E, Stefani A, Orlacchio A, Sorbi S, George-Hyslop PH, and Bernardi G (2002) Association study of the 5-hydroxytryptamine(6) receptor gene in Alzheimer's disease. Neurosci Lett 325: 13-16.[CrossRef][Medline]

Purdon SE, Jones BD, Stip E, Labelle A, Addington D, David SR, Breier A, and Tollefson GD (2000) Neuropsychological change in early phase schizophrenia during 12 months of treatment with olanzapine, risperidone, or haloperidol. The Canadian Collaborative Group for Research in Schizophrenia. Arch Gen Psychiatry 57: 249-258.[Abstract/Free Full Text]

Reavill C and Rogers DC (2001) The therapeutic potential of 5-HT6 receptor antagonists. Curr Opin Investig Drugs 2: 104-109.[Medline]

Rogers DC and Hagan JJ (2001) 5-HT6 receptor antagonists enhance retention of a water maze task in the rat. Psychopharmacology 158: 114-119.[CrossRef][Medline]

Routledge C (1999) Proceedings of the British Pharmacological Society Meeting, joint meeting with the Portuguese Society of Pharmacology. 8-9 April 1999. Abstracts; 21P.

Routledge C, Bromidge SM, Moss SF, Price GW, Hirst W, Newman H, Riley G, Gager T, Stean T, Upton N, et al. (2000) Characterization of SB-271046: a potent, selective and orally active 5-HT(6) receptor antagonist. Br J Pharmacol 130: 1606-1612.[CrossRef][Medline]

Ruat M, Traiffort E, Arrang JM, Tardivel-Lacombe J, Diaz J, Leurs R, and Schwartz JC (1993) A novel rat serotonin (5-HT6) receptor: molecular cloning, localization and stimulation of cAMP accumulation. Biochem Biophys Res Commun 193: 268-276.[CrossRef][Medline]

Russell MG and Dias R (2002) Memories are made of this (perhaps): a review of serotonin 5-HT(6) receptor ligands and their biological functions. Curr Top Med Chem 2: 643-654.[CrossRef][Medline]

Sebben M, Ansanay H, Bockaert J, and Dumuis A (1994) 5-HT6 receptors positively coupled to adenylyl cyclase in striatal neurones in culture. Neuroreport 5: 2553-2557.[Medline]

Shinkai T, Ohmori O, Kojima H, Terao T, Suzuki T, and Abe K (1999) Association study of the 5-HT6 receptor gene in schizophrenia. Am J Med Genet 88: 120-122.[CrossRef][Medline]

Shirazi-Southall S, Rodriguez DE, and Nomikos GG (2002) Effects of typical and atypical antipsychotics and receptor selective compounds on acetylcholine efflux in the hippocampus of the rat. Neuropsychopharmacology 26: 583-594.[CrossRef][Medline]

Sleight AJ, Boess FG, Bos M, Levet-Trafit B, Riemer C, and Bourson A (1998) Characterization of Ro 04-6790 and Ro 63-0563: potent and selective antagonists at human and rat 5-HT6 receptors. Br J Pharmacol 124: 556-562.[CrossRef][Medline]

Sleight AJ, Monsma FJ Jr, Borroni E, Austin RH, and Bourson A (1996) Effects of altered 5-HT₆ expression in the rat: functional studies using antisense oligonucleotides. Behav Brain Res 73: 245-248.[CrossRef][Medline]

Stean TO, Hirst WD, Thomas DR, Price GW, Rogers D, Riley G, Bromidge SM, Serafinowska HT, Smith DR, Bartlett S, et al. (2002) Pharmacological profile of SB-357134: a potent, selective, brain penetrant and orally active 5-HT(6) receptor antagonist. Pharmacol Biochem Behav 71: 645-654.[CrossRef][Medline]

Thome J, Retz W, Baader M, Pesold B, Hu M, Cowen M, Durany N, Adler G, Henn FA, and Rosler M (2001) Association analysis of HTR6 and HTR2A polymorphisms in sporadic Alzheimer's disease. J Neural Transm 108: 1175-1180.[CrossRef]

Tsai SJ, Chiu HJ, Wang YC, and Hong CJ (1999a) Association study of serotonin-6 receptor variant (C267T) with schizophrenia and aggressive behavior. Neurosci Lett 271: 135-137.[CrossRef][Medline]

Tsai SJ, Liu HC, Liu TY, Wang YC, and Hong CJ (1999b) Association analysis of the 5-HT6 receptor polymorphism C267T in Alzheimer's disease. Neurosci Lett 276: 138-139.[CrossRef][Medline]

Vogt IR, Shimron-Abarbanell D, Neidt H, Erdmann J, Cichon S, Schulze TG, Muller DJ, Maier W, Albus M, Borrmann-Hassenbach M, et al. (2000) Investigation of the human serotonin 6 [5-HT6] receptor gene in bipolar affective disorder and schizophrenia. Am J Med Genet 96: 217-221.[CrossRef][Medline]

Woolley ML, Bentley JC, Sleight AJ, Marsden CA, and Fone KC (2001) A role for 5-HT(6) receptors in retention of spatial learning in the Morris water maze. Neuropharmacology 41: 210-219.[CrossRef][Medline]

Yoshioka M, Matsumoto M, Togashi H, Mori K, and Saito H (1998) Central distribution and function of 5-HT6 receptor subtype in the rat brain. Life Sci 62: 1473-1477.[CrossRef][Medline]

This article has been cited by other articles:

				P. Svenningsson, E. T. Tzavara, H. Qi, R. Carruthers, J. M. Witkin, G. G. Nomikos, and P. Greengard Biochemical and Behavioral Evidence for Antidepressant-Like Effects of 5-HT6 Receptor Stimulation J. Neurosci., April 11, 2007; 27(15): 4201 - 4209. [Abstract] [Full Text] [PDF]

				R. Lydic, R. Garza-Grande, R. Struthers, and H. A. Baghdoyan Nitric oxide in B6 mouse and nitric oxide-sensitive soluble guanylate cyclase in cat modulate acetylcholine release in pontine reticular formation J Appl Physiol, May 1, 2006; 100(5): 1666 - 1673. [Abstract] [Full Text] [PDF]

				J. C. Talpos, L. S. Wilkinson, and T. W. Robbins A comparison of multiple 5-HT receptors in two tasks measuring impulsivity J Psychopharmacol, January 1, 2006; 20(1): 47 - 58. [Abstract] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.056002v1 307/2/682    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 HighWire Citing Articles via Google Scholar Google Scholar Articles by Lindner, M. D. Articles by Gribkoff, V. K. Search for Related Content PubMed PubMed Citation Articles by Lindner, M. D. Articles by Gribkoff, V. K.