|Home | About | Journals | Submit | Contact Us | Français|
The use of extracts that are highly enriched in phytoestrogens, such as genistein, has become popular to promote various aspects of healthy aging, including maintenance of cognitive function. These compounds are promoted to menopausal women as safe, natural alternatives to traditional estrogen therapies, yet their safety and efficacy are poorly understood. Previous research in our lab found that once daily oral treatment of ovariectomized female Long-Evans (LE) rats with the soy phytoestrogen, genistein resulted in subtle deficits in performance on cognitive tasks assessing working memory and response inhibition/timing ability. The present study further modeled exposure of the menopausal woman to genistein by treating 14-month old ovariectomized female LE rats three times daily at a dose of genistein resulting in serum concentrations similar to those that could be achieved in humans consuming either a commercially available soy isoflavone supplement or a diet high in these phytoestrogens. Genistein (3.4 mg/kg) or sucrose control pellets were orally administered to animals daily, 30 minutes before behavioral testing, and again both 4 and 8 hours after the first treatment. The test battery consisted of a delayed spatial alternation task (DSA) that tested working memory and a differential reinforcement of low rates of responding (DRL) task that tested inhibitory control/timing. Genistein treatment impaired DSA performance relative to sucrose controls. Performance on the DRL task was largely unaffected by genistein treatment. Although the impairment measured on DSA was less pronounced than that we have previously reported following chronic treatment with 17β-estradiol, the pattern of the deficit was very similar to that observed with 17β-estradiol.
Soy extracts highly enriched in phytoestrogens are marketed as dietary supplements that will promote “healthy” aging by preventing or reducing various age-associated diseases and conditions, including cognitive decline. The potential therapeutic benefits derived from soy supplements containing isoflavones are presumed to be mediated through their estrogenic properties (Kostelac et al., 2003; Mueller et al., 2004; Suetsugi et al., 2003), and, because of this, many of these products are marketed to peri- and postmenopausal women as safe, natural alternatives to traditional hormonal therapies (McKee and Warber, 2005; Shifren and Schiff, 2010). However, the safety and efficacy of dietary supplements containing soy isoflavones are largely unknown. Past research suggests a variety of positive effects on the aging process following consumption of soy based dietary supplements (see Ferrari, 2004), but recent studies, including those using soy protein isolate, have been less convincing (see Lethaby et al., 2007; Prasain et al., 2010; Xiao, 2008).
Research assessing the effects of soy phytoestrogen diets with a mixed isoflavone content on cognition in ovariectomized young or middle-aged female rodents has revealed beneficial effects of either acute or chronic soy supplementation on hippocampally-mediated learning and memory tasks (Lee et al., 2009; Luine et al., 2006; Monteiro et al., 2008; Pan et al., 2000). These results parallel the effects of treatment with physiological doses of 17β-estradiol on these same tasks (Frick, 2009). Of note, these studies used mixtures of isoflavones, including genistein and daidzein, making it difficult to discern the effects of specific components of soy phytoestrogen diets on cognition. Further, several studies maintained the experimental animals on standard rodent chow or failed to report diet. Soy content of commercially based rodent diets can vary greatly from lot to lot, delivering relatively high (250-350 ug/g), and often quite variable levels of additional soy through the diet (Brown and Setchell, 2001; Thigpen et al., 2004, 2007), which could result in an elevated and inconsistent experimental dose of soy isoflavones received by the animals in these studies.
Genistein, of the major isoflavones, is the most extensively studied. Research has shown that genistein can bind both estrogen receptor (ER) α and ERβ, but unlike 17β-estradiol, has a higher binding affinity for ERβ (Kuiper et al., 1998). Genistein has also been found to bind the nonclassical ER, GPR30 (Maggiolini et al., 2004; Thomas and Dong, 2006), an estrogen-responsive G-protein coupled receptor (Prokai and Simpkins, 2007; Prossnitz et al., 2008), activation of which can restore ovariectomy-induced deficits in spatial memory, similar to effects of 17β-estradiol (Hammond et al., 2009). The sparse research targeting the effects of genistein on cognition reveals a mnemonic enhancing effect for hippocampus-sensitive tasks. Specifically, following both short-term and chronic (6 weeks) genistein treatment, spatial memory, as tested in the Morris water maze, was enhanced in young adult ovariectomized female rats (Alonso et al., 2010; Huang and Zhang, 2010; Xu et al, 2007). These effects are similar to those seen following ovariectomy and subsequent 17β-estradiol treatment (Frick, 2009). In contrast, genistein failed to improve the performance of aged rats (90-96 weeks) in a Morris water maze task (Alonso et al., 2010), suggesting an age-related change in the therapeutic potential of short-term genistein treatment on hippocampal tasks.
Recently our research group found a detrimental effect following an oral, once-a-day treatment of genistein (approximately 3.2 mg/kg/day) in ovariectomized female rats on an operant delayed spatial alternation (DSA) task (Neese et al., 2010a). This task requires rats to alternate their responses between two retractable levers, with variable delays occurring between opportunities to press. The detrimental effect of genistein was only seen in old rats and was limited to the last block of testing (Neese et al., 2010a); in contrast, 17β-estradiol treatment impaired performance on this same task in ovariectomized rats across a variety of ages including 3, 6, 12, and 18 month-olds (Wang et al., 2008, 2009). We also found a subtle effect of genistein treatment on an operant differential reinforcement of low (DRL) rates of responding schedule, a task which required animals to withhold a response for 15 seconds in order to receive a food reinforcer. Although not quite reaching statistical significance (p=0.055), genistein treatment produced a decrease in overall efficiency on the DRL schedule, similar to, but again less pronounced than our previous findings following 17β-estradiol treatment on this same task (Wang et al., 2008, 2011). Together, our findings suggest a detrimental effect of genistein treatment on the performance of both the DSA and DRL tasks in ovariectomized rats. This pattern of deficits is similar to what we have observed following 17β-estradiol treatment and is consistent with other research findings demonstrating parallel behavioral outcomes following genistein or 17β-estradiol treatment.
This study was conducted as a follow-up to our original study (Neese et al., 2010a) that used single daily genistein treatments. Genistein has a half-life of about four hours in rodents (Chang et al., 2000). While the once/day dosing used in our initial study resulted in blood concentrations one hour after treatment that were similar to blood concentrations in humans consuming typical high soy Asian diets (Aldercreutz et al., 1994), this once daily treatment failed to produce serum genistein levels that remained elevated throughout the day. The rats in the present study were treated three times daily, a treatment method that better models humans consuming an isoflavone-rich diet, where serum genistein levels have been shown to remain elevated throughout the day (Gardner et al., 2009). Additionally, as soy-based products are widely consumed by recently menopausal women (Kurzer, 2003; Nachtigal et al., 2005; Nieves, 2009) we used only middle-aged rats in this study. All rats were again subject to DSA and DRL testing (Neese et al., 2010a).
Thirty-two 14-month-old female Long-Evans rats were obtained from Harlan (Indianapolis, IN) and were maintained in facilities fully accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care (AAALAC). Rats were housed in a temperature and humidity controlled room (22°C, 40-55% humidity) on a 12-hour reverse light-dark cycle (lights off at 8:30 am). All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Illinois at Urbana-Champaign and were in accordance with the guidelines of the Public Health Service Policy on Humane Care and Use of Laboratory Animals (2002) and the Guidelines for the Care and Use of Mammals in Neuroscience and Behavioral Research (2003).
Rats were pair-housed in plastic cages (45×24×20cm) with corncob bedding. All rats were maintained on an AIN-93G soy-free diet (Harlan-Teklad, Madison, WI), with water available ad libitum. The diet was switched from standard rat chow (Harlan-Teklad 8604) to the AIN-93G on the day of ovariectomy surgery to ensure that the rats were not exposed to additional dietary genistein (or other estrogens) via the diet (Brown and Setchell, 2001; Thigpen et al., 2004, 2007). No change in food intake or body weight was observed following the change in diet. Upon arrival, estrous status was determined daily for two weeks prior to ovariectomy surgery to determine reproductive status prior to surgery, and again post-surgery to confirm ovariectomy. Briefly, sterile saline was flushed into the vaginal opening with an eye dropper and the collected fluid was placed on a microscope slide. Each slide was then stained with hematoxylin and eosin and estrous cycle phase was determined using a light microscope (Olympus BH-2). Slides were scored according to the four-stage cycle detailed in Yener et al., (2007).
Rats were food-restricted to and maintained at 85% of their individual free-feeding body weights one week after ovariectomy. During behavioral testing, rats were fed one hour after the daily test session was completed. Testing began two weeks following ovariectomy and occurred once daily, six days/week during the dark phase of the light cycle. The rats were equally divided into two treatment groups, a genistein-treated group (genistein incorporated into a sucrose pellet) and a sucrose-treated control group (sucrose pellet only). Both groups were treated three-times daily, with the initial treatment occurring 30 minutes prior to behavioral testing. Subsequent treatments occurred at both 4 and 8 hours after the initial treatment.
To achieve elevated blood genistein levels throughout the day, rats were treated three times daily with 97 mg fruit flavored sucrose pellets that contained 0.5% genistein (TestDiet, Richmond, IN, #1811494). Treatment began two days prior to the onset of operant training. The genistein content of the sugar pellets was tested using reverse phase HPLC with electrochemical detection (Penalvo et al., 2004), and they were found to contain ~96% of the targeted dose (0.485 mg). Tail vein blood samples were collected in pilot rats prior to the beginning of the study to confirm blood isoflavone levels following dosing with genistein. This was done to establish that levels remained elevated throughout the day, and subsequently dropped during the overnight washout period. Total genistein levels were determined in 10 μl aliquots of plasma by using LC/MS/MS (Twaddle et al., 2002). The detection limit for a 10 μl sample was 0.05 μM. Pilot research found serum genistein levels to peak at about one hour following treatment, and to decline by approximately one-half four hours later, rising again in a similar fashion following the second and third treatments, and falling back to negligible levels prior to the next day’s treatment (data not shown). Serum samples were acquired for the rats in the current study following the initial two days of treatment, again following DSA testing, and following DRL testing at the end of the study. Each sample was collected 30 minutes after treatment.
All rats were treated three times daily with 97 mg sucrose pellets. The sucrose-treated group weighed (mean±SEM in grams) 303.25±7.27 and the genistein-treated group weighed 329.44±8.37. The genistein-treated group received two 0.5% genistein containing pellets, which led to an average dose of 3.4 mg/kg/treatment. The average daily intake from the three treatments was 2.9 mg of genistein. Sucrose-treated rats received two pellets containing sucrose only. Each rat was removed from its home cage and given its sucrose pellets individually. All rats consumed the pellets within a few seconds.
Behavioral testing was conducted in 16 standard automated operant chambers (Med Associates Inc., St. Albans, VT) housed in sound-attenuated wooden boxes (interior dimensions: 55.9 cm wide, 38.1 cm high, 35.6 cm deep). All of the test chambers had the same features and dimensions: 21.6 cm high with a 29.2 cm × 24.8 cm stainless-steel grid floor that rested just above a tray filled with corncob bedding. Soy-free reward pellets (45-mg Formula P, P.J. Noyes Inc., Lancaster, NH) were dispensed through a pellet dispenser centered 2.5 cm above the floor on the operant panel. Retractable response levers were positioned symmetrically on both sides of the pellet dispenser, with a pair of stimulus cue lamps directly above each lever. The levers were 5.7 cm from midline and 7.0 cm above the floor and the cue lights were located 5.7 cm above the levers. Each chamber also contained a Sonalert tone generator, a white noise generator, and a house light located on the back wall. Experimental contingencies were programmed using the Med-State behavioral programming language (Med-Associates, Vermont).
Rats were trained to press the response levers by using an autoshaping program (Neese et al., 2010a, 2010b; Newland et al., 1986; Verma et al., 1996; Wang et al., 2008, 2009; Widholm et al., 2001, 2003). Autoshaping test sessions were terminated after 60 minutes elapsed or 100 reinforcers were delivered. Criterion for this condition was set at 100 lever presses within a single session. Most rats took 2-4 autoshaping sessions to achieve 100 reinforcers earned, except for one rat each from the sucrose and genistein-treated groups that required 7 or 11 sessions, respectively to achieve this criterion. Following autoshaping, the rats were exposed to a continuous reinforcement schedule in which the lever associated with reinforcement alternated following delivery of every fifth reinforcer in order to prevent the rats from developing a lever or side preference. This cycle of alternating levers terminated after 100 reinforcers were received or 60 minutes had elapsed. A performance criterion of 100 reinforcers earned for two consecutive sessions was established for this condition.
After lever press training, the rats were trained to alternate their responses prior to beginning testing on the variable delay DSA task (Gendle et al., 2004; Neese et al., 2010a, 2010b; Roegge et al., 2005; Wang et al., 2008, 2009; Widholm et al., 2003). The initial training schedule began with cued alternation (CA) training in which a cue light was illuminated to indicate the correct lever on each trial. Each correct cued lever press was reinforced. The lever associated with reinforcement was not altered following an incorrect response, i.e. the “correct” lever continued to be the lever opposite of the lever associated with reinforcement in the most recent “correct” trial. Levers were retracted and extended between trials. No delay was imposed between trials in this initial cued training phase. The time between retraction and extension of the levers was <0.15 seconds. Rats were trained to a criterion of one session above chance defined as >60% correct presses. Next, a noncued alternation (NCA) task was presented where the cue light no longer indicated the correct lever, and both cue lights were illuminated when the levers were extended. Correct responses still consisted of alternating right and left lever presses with the levers retracting and extending between presses. As for CA, the correct lever was again always the lever opposite of the one on the trial in which most recent “correct” lever press occurred. Again, no delay was imposed between opportunities to press. Each rat was trained for 10 sessions on the NCA phase.
The final alternation phase was the variable delay DSA task. Following each lever press, variable delays of 0, 3, 6, 9, or 18 seconds were imposed randomly between opportunities to press (Neese et al., 2010a, 2010b; Roegge et al., 2005; Wang et al., 2008, 2009; Widholm et al, 2003). There were 40 trials at each delay and a total of 200 trials per session. Delays were randomly balanced within each session and any specific delay was not presented on more than three consecutive trials. The correct lever was again always the lever opposite of the one on the trial in which most recent “correct” lever press occurred. Each animal was tested for 25 consecutive sessions.
Beginning on the day immediately following the completion of the DSA task, timing ability and inhibitory control were assessed using a DRL task (Sable et al., 2006, 2009; Neese et al., 2010a; Wang et al., 2008, 2011). Only one response lever (left) was used for the DRL schedule of responding. The lever remained extended throughout the session. The first two training sessions consisted of a fixed-ratio 1 schedule in which every response was reinforced for 200 trials, or 90 minutes, whichever occurred first. Animals did not have to meet a performance criterion before moving on to the next stage of training. Training sessions 3 and 4 consisted of a DRL-5 second schedule in which reinforcement was contingent upon at least a 5 second separation between responses. If a response occurred within the 5-second window, the response timer was reset. Training sessions 5 and 6 consisted of a DRL-10 second schedule. The rats were then tested for 30 sessions on a DRL-15 second schedule.
Following the completion of behavioral testing, all animals were given an overdose of CO2 and the uterine horn was dissected from the peritoneal cavity of all rats by separating the uterine horns from the underlying tissue, and excising the uterine body. Uterine horns were weighed immediately following removal.
The behavioral data were analyzed via repeated measures ANOVA using SPSS for Windows, Version 18.0. Linear analysis was conducted using JMP for Windows, Version 6.0. Treatment group was included in the analyses as the between subjects factor, α=0.05.
For CA, cumulative errors across all sessions served as the measure of learning, and were analyzed using between-subjects ANOVA. For NCA, the overall proportion correct across the ten sessions served as the primary measure of learning, and was analyzed using a 2 (treatment) × 10 (session) mixed ANOVA where session was a repeated measures factor.
For DSA, the proportion correct across the 25 test sessions was first averaged across blocks of five test sessions to produce five 5-session test blocks. The learning curves for proportion correct were then analyzed with trend analysis by linear contrasts in order to determine whether rate of improvement over subsequent sessions differed between the two treatment groups. Proportion correct at each delay across the 25 test sessions was also analyzed using a mixed 2 (treatment) × 5 (block) × 5 (delay) repeated measures ANOVA with block (1 - 5) and delay (0, 3, 6, 9, 18 sec) serving as repeated measures factors.
Error pattern analyses were also conducted to better understand potential behavioral changes in rats treated with genistein. Specifically, these analyses were completed to determine if genistein rats were more likely to respond incorrectly following a correct or an incorrect lever press. A “win-stay” error was defined as an incorrect response on the same lever that had been correct on the previous trial, whereby the rat responded correctly on the n-1 trial, but incorrectly on the nth trail. A “lose-stay” error was defined as an incorrect lever press on the same lever that had been incorrect on the previous trial, whereby the rat responded incorrectly on the n-1 trial as well as the nth trial. Lose-stay errors thus represent at least the second consecutive press on the lever no longer associated with reinforcement. The learning curves for errors committed were analyzed with trend analysis by linear contrasts between the two treatment groups for both win-stay and lose-stay errors. Win-stay and lose-stay errors were also analyzed separately using a mixed 2 (treatment) × 5 (block) repeated measures ANOVA with block (1 - 5) serving as a repeated measures factor. The latency to respond following a correct and an incorrect lever press were also analyzed separately using a mixed 2 (treatment) × 5 (block) repeated measures ANOVA with block (1 - 5) serving as a repeated measures factor.
For DRL, the ratio of reinforced to non-reinforced lever presses was the primary measure of learning. The data were first averaged to yield six blocks of five sessions each. The learning curves for the ratio of reinforced to non-reinforced lever presses were analyzed with trend analysis by linear contrasts between the two treatment groups. The ratio of reinforced to non-reinforced lever presses was also analyzed using mixed 2 (treatment) × 6 (block) ANOVA, where block (1-6) served as a repeated-measures factor.
Inter-response times (IRTs) representing the delay between lever presses were also examined. IRT distributions provide information as to the temporal location when responses “peak”. IRTs were divided into eight 2.5-sec intervals (e.g., 0<2.5 sec, 2.5<5.0 sec, etc.) with all IRTs longer than 17.5-sec falling in the last IRT interval. The IRT data from the initial DRL-15 testing block and the last DRL-15 testing block were analyzed separately using a 2 (treatment) × 8 (IRT interval) mixed ANOVA with IRT interval (1-8) as a repeated measures factor. This was done to determine the extent of response inhibition at the beginning of testing and to assess whether extended training shifted the pattern of responding to favor longer IRTs by the last block of testing.
Blood genistein levels were analyzed via a mixed 2 (treatment) × 3 (time) repeated measures ANOVA. Uterine horn weights were analyzed via one-way ANOVA with treatment as the between subjects factors.
Cycle status was defined as follows: regular cycle (4-5 day cycle), irregular cycle (6-12 day cycle), or extended estrus (periods of estrus lasting 3 or more days) (Fentie et al., 2004). Analyses found that the majority of rats had begun to cycle irregularly, with a small subset of animals in each treatment group entering into periods of extended estrus. On average, these cycle differences were spread evenly across both treatment groups. All rats stopped cycling post-ovariectomy, and remained in a phase resembling diestrus (Hubscher et al., 2005).
Treatment with genistein produced elevated serum genistein levels, an effect that was consistent across time. This was revealed by a significant main effect of treatment group, F(1,29)=74.652, p<0.001. As shown in Table 1, the genistein-treated rats had significantly elevated levels of serum genistein while the sucrose-treated control rats had negligible levels of genistein across the three sampling time points.
Treatment did not impact uterine horn weights. One-way ANOVA failed to uncover a significant difference between the genistein-treated group (mean±SEM in grams: 0.30±0.018) and sucrose-treated control group (0.28±0.014) on uterine horn weight, F(1,28)=0.626, p>0.05.
Genistein treatment had no effect on performance of the two DSA training phases, CA or NCA, p>0.05. A main effect of session was uncovered for NCA training, F(9,270)=51.417, p<0.001. As expected, all animals performed better across subsequent sessions of testing.
Genistein was found to impair performance on the DSA task. Trend analysis by linear contrasts of the learning curves represented in Figure 1A revealed a significant difference between the genistein-treated group (M=0.699) and the sucrose-treated control group (M=0.728), t(1)=2.10, p=0.038. The learning curve of the genistein-treated group had a flatter slope than that of the sucrose-treated control group, indicating an impairment in acquisition of the DSA task across the 25 sessions of testing. The main effect of treatment in the ANOVA also approached significance, F(1,30)=3.362, p=0.077 (Figure 1B), finding the genistein-treated group to perform worse overall than the sucrose-treated control group. Neither the block × treatment (F(4, 120)=1.691, p=0.199) nor the delay × treatment (F(4, 120)=1.120, p=0.339) interactions were significant. A block × delay effect was also uncovered, F(16,480)=16.682, p<0.001 (data not shown). The performance of both treatment groups improved across subsequent sessions of testing, except at the longest delay.
Genistein treatment was also found to influence error patterns during DSA testing. Trend analysis by linear contrasts of the error rates represented in Figure 2A revealed a difference between the genistein-treated group (M=113.41) and the sucrose-treated control group (M=98.27) on lose-stay errors that approached significance, t(1)=1.87, p=0.063. The main effect of treatment also approached significance for lose-stay errors, F(1,30)=3.889, p=0.058. As indicated in Figure 2B, the genistein-treated group committed more lose-stay errors during DSA testing than did the sucrose-treated control group. The treatment × error interaction for lose-stay errors was not significant, F(4,120)=1.791, p=0.184. Trend analysis by linear contrasts of the error rates represented in Figure 3 revealed a significant difference between win-stay error rates, t(1)=20.05, p=0.042. The genistein-treated group (M=185.05) committed win-stay errors at a higher rate, specifically seen across the later testing sessions, than did the sucrose-treated control group (M=170.91). The treatment × error interaction (F(4,120)=1.312, p=0.277) and the main effect of treatment for win-stay errors (F(1,30)=2.155, p=0.153) were not significant. A main effect of block was also uncovered for both lose-stay, F(4,120)=100.618, p<0.001, and win-stay errors, F(4,120)=64.297, p<0.001. As expected, both groups committed fewer errors of both types across subsequent sessions of testing.
Treatment was not found to influence latency to respond following either a correct F(1,30)=0.346, p=0.561, or an incorrect lever press, F(1,30)=0.094, p=0.762 (data not shown). Subsequent analysis also failed to uncover any significant main effects of block on latencies to respond following either a correct F(4,120)=1.351, p=0.265, or an incorrect lever press, F(4,120)=1.035, p=0.336, indicating that latencies to respond did not change across sessions of testing.
Two rats from the sucrose control group died shortly after DRL-15 testing began; therefore they are not included in the DRL-15 analyses. Overall, performance on the DRL task was not affected by treatment, as indicated by a lack of treatment induced effect on the main measure of learning, the ratio of reinforced to non-reinforced presses. Specifically, trend analysis by linear contrasts of the ratio of reinforced to non-reinforced presses represented in Figure 4 failed to reveal a difference in performance efficiency between the genistein-treated group M=0.3018) and the sucrose-treated control group (M =0.3671), t(1)=0.758, p=0.385, although the genistein-treated group did perform slightly worse than the sucrose control group. Omnibus ANOVA also failed to find a main effect of treatment on the ratio of reinforced to non-reinforced presses, F(1,28)=0.322, p=0.322. A main effect of block was uncovered, F(5,140)=27.850, p<0.001. The ratio of reinforced to non-reinforced lever presses improved across sessions of testing in both groups, suggesting that both groups learned the task.
Genistein treatment also failed to affect response patterns across 2.5 sec IRT bins in both the first 5 sessions of testing (initial testing block; Figure 5A), F(7,196)=0.595, p=0.548 and the last 5 sessions of testing (final testing block; Figure 5B), F(7,196)=0.225, p=0.754. As Figures 5A and 5B show, both treatment groups made, on average, equivalent responses in those bins associated with reward (>15.0 sec). A significant main effect of bin was uncovered for both the initial testing block, F(7,196)=22.537, p<0.001, and for the final testing block, F(7,196)=29.295, p<0.001. During the initial block of the task, the number of responses per IRT bin decreased as the IRT bin became longer. However, in contrast to the pattern during the initial block, excluding the first bin, the number of responses per IRT bin increased as the IRT became longer during the final block of testing, indicating that rats in both groups learned to withhold more of their responses over the course of testing.
The current study examined the effects of chronic dietary genistein supplements on behavioral tests designed to tap working memory and response inhibition. The study expanded upon our previous finding that chronic, once-a-day oral genistein exposure produced a deficit in DSA performance (Neese et al., 2010a). In that study we found old rats (22 months of age) treated with a single daily dose of approximately 3.2 mg/kg genistein to perform worse than both middle-aged and young rats treated with the same dose of genistein. We also found a subtle deficit in overall efficiency on an operant DRL schedule following genistein treatment. Because soy-based isoflavone-containing products are often marketed to recently menopausal women Kurzer, 2003; Nachtigal et al., 2005; Nieves, 2009), we focused the current study on middle-aged rats only.
To better model human intake of either a high soy diet (assuming three meals/day) or consumption of multiple daily doses of a soy-based supplement, we treated the middle-aged rats in this study three times per day. We found this treatment regimen to result in elevated serum levels of genistein throughout a twelve hour period that were slightly higher, but still similar to the levels measured in humans following consumption of three meals high in isoflavone content 90 mg isoflavones/day) or following ingestion of a soy supplement (mixed isoflavones, 192 mg isoflavones/day) (Gardner et al., 2009). Although serum genistein levels were elevated, uterine horn weights were not increased. This was not unexpected, as several studies using genistein doses similar to, or higher than ours (up to 20 mg/kg/day), also failed to measure a uterotrophic response (Bitto et al., 2009; Fanti et al., 1998; Moller et al., 2010; Sliwinski et al., 2009). This is also not unexpected given that the uterine horn is rich in ERα and genistein binds preferentially to ERβ.
Ovariectomized middle-aged female rats treated with genistein (approximately 3.4 mg/kg) three times per day in this study performed worse than sucrose-treated controls on an operant variable delay DSA task, a test of working memory. Linear analysis of the proportion correct revealed a flatter learning curve in the genistein treatment group, suggestive of an impairment in the acquisition of this task in relation to sucrose-treated controls. Omnibus ANOVA also revealed an impairment that approached significance in the overall proportion correct of genistein-treated rats. Linear analysis also found a flatter slope/slower rate of decline in the win-stay error rates of the genistein-treated group across the five blocks of testing sessions. An increase in lose-stay errors which approached significance was also uncovered by both linear analysis and ANOVA in the genistein-treated group. This deficit was similar to, but of a lesser magnitude (in both size and statistical significance) than that seen in 17β-estradiol treated middle-aged rats on the DSA task (Wang et al., 2009).
The DSA task was conducted to assess potential working memory changes in genistein-treated rats. Importantly, the length of the delay interval did not differentially affect performance of the genistein-treated group on the DSA task, results that do not support a change in working memory capacity. Rather, equivalent deficits across all delay intervals suggests a change in a non-mnemonic process, such as a shift in motivation (see Pontecorvo et al., 1996; Van Hest and Steckler, 1996) or an altered emotional response following genistein treatment (see Skirrow et al., 2009).
Latencies to respond following either a correct or an incorrect lever press did not differ between the genistein-treated and sucrose control groups. Thus it does not appear that genistein had an effect on motivation to complete the DSA task. An altered emotional response seems more likely (discussed below). Deficits were not present in the training phases of the task, but emerged when the delays were introduced suggesting that the genistein-treated rats may have experienced a heightened emotional reaction in response to variable delay periods and/or errors committed during this phase of testing.
Genistein-treated rats committed more win-stay errors than did the sucrose-treated rats during DSA testing, an effect revealed by the significant difference in the slopes of these error rates. Further, the overall number of lose-stay errors committed by the genistein-treated groups was also increased, a shift that approached significance in both rate and overall number of errors committed during DSA testing (see Fig 2A and 2B). This error type differs from the win-stay type as it is a response to a lever that had not been associated with reward on the previous press, and is indicative of response perseveration (see Martinez et al., 1988; Zornetzer et al., 1982). In summary, the genistein-treated rats committed more of both types of errors than did sucrose-treated rats during DSA testing.
Overall, the genistein-induced performance deficit during DSA testing relates to the inclusion of inter-trial delays, but this does not appear to be an effect on memory as the deficit does not vary by delay. Further, the error pattern analysis also fails to provide evidence that performance on a previous trial affected the outcome on the subsequent trial. Rather, this deficit appears related to other factors, such as frustration (i.e. a heightened emotional response) related to the inclusion of a varied delay interval (see Carpenter et al., 2002; Morgan et al., 2002) or to the higher error rate in genistein-treated rats.
The effects of genistein treatment in this study were limited to the DSA task. Performance on the DRL schedule was largely unaffected by genistein treatment. Both treatment groups exhibited a large amount of ‘bursting’ (responses in the <2.5 sec bin) during both the initial and final blocks of the DRL-15 schedule (Figures 5A and 5B), responses typical of DRL and other operant timing tasks (see Blough, 1963; Kirkpatrick and Church, 2003; Richards et al., 1993). Importantly, although both groups responded in shorter IRT bins during the initial block of testing, both the genistein-treated and sucrose-treated groups showed a higher frequency of responses in the longer IRT bins during the last block of testing. Although we did find a marginally significant main effect of once a day genistein treatment on the ratio of reinforced to non-reinforced lever presses in our previous work (Neese et al., 2010a), this effect was subtle, and not moderated by age. Still, the lack of an effect on DRL performance following multiple genistein treatment is somewhat surprising given that we did find chronic 17β-estradiol treatment to be detrimental to accurate performance on this task (Wang et al., 2008, 2011). The reasons for this lack of effect are not clear. As Figure 4 shows, the performance efficiency of the genistein-treated rats was slightly lower than the sucrose control treated group during most blocks of testing. Surprisingly, the performance of the sucrose control group dropped suddenly in the last two blocks of testing. This unexplained drop may have masked any effect, albeit subtle, that genistein might have had on the acquisition of this task. However, the lack of a treatment induced change in response patterns seen during the final block of IRT testing also fails to support a treatment effect.
It is important to note that some of the differences between the genistein-treated and sucrose-treated control groups on the DSA task only achieved p-values between 0.05 and 0.10, but we feel it is also important to describe effects that approach the preset α value (Cleophas, 2004; Morton, 2009). While adhering to a stringent α level protects against reporting an effect that may be spurious, this also reduces the power to detect a difference when there is one. This is particularly important in studies that address issues relevant to public health. We have discussed this issue in detail in a previous publication (see Schantz et al., 2004; pp. 673-6 and see Holson et al., 2008; Newman et al., 2008). With a compound like genistein that is consumed on a regular basis by so many people, the issue of Type II error—failing to detect a difference when in fact there is one—cannot be ignored. Thus, we felt it was important to present results that approached alpha. However, we remind readers that the deficits we observed are subtle and should be interpreted cautiously until they have been replicated.
This study extended our previous findings that genistein can negatively impact performance on a variable delay operant DSA task (Neese et al., 2010a), suggesting that multiple daily treatments result in impaired performance in middle-aged rats. Although the majority of studies testing behavior following genistein treatment found improvements in performance, rather than deficits (Alonso et al., 2010; Huang et al., 2010; Xu et al., 2007), these studies generally used hippocampally-mediated tasks reflecting what is typically seen following 17β-estradiol treatment on these same behavioral tasks (Frick, 2009). Our data also reflect this parallel, as we have published studies documenting a negative impact of both 17β-estradiol and DPN on DSA performance (Neese et al., 2010b; Wang et al., 2008, 2009). Our research group has recently also found that short-term genistein treatment impairs striatum-sensitive response-learning in ovariectomized young adult rats (Pisani et al., 2010), and the magnitude of this deficit was similar to that seen following 17β-estradiol treatment on the same task (Korol and Kolo, 2002; Zurkovsky et al., 2007). In contrast to the DSA task, genistein treatment failed to reduce performance efficiency on a DRL-15 schedule, results which do not parallel our previous research with 17β-estradiol on the DRL-15 schedule (Wang et al., 2008, 2011). Soy products, including those containing genistein, are being marketed as dietary agents to aide in healthy aging. Our results add to a growing body of evidence suggesting that there may be not only benefits, but also risks associated with the use of these isoflavones (Lethaby et al., 2007; Prasain et al., 2010; Xiao, 2008). Future research will be aimed at determining the underlying mechanisms of genistein action in the brain, and at extending these findings to other phytoestrogens to determine the cognitive effects of these compounds during the aging process.
This research was supported by National Institute on Aging Grants P01 AG024387 (SLS, WGH) and P50 AT006268 from ODS, NCAAM, and NCI (SLS, WGH). This research was also supported by NSF IOB 0520876 (DLK). Steven Neese also received support from National Institute of Environmental Health Sciences Grant T32 ES007326. The views presented in this article do not necessarily reflect those of the U.S. Food and Drug Administration.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of Interest Statement
The authors have no potential conflicts of interest to report.