|Home | About | Journals | Submit | Contact Us | Français|
The effects of progesterone on memory are not nearly as well studied as the effects of estrogens. Although progesterone can reportedly enhance spatial and/or object recognition in female rodents when given immediately after training, previous studies have injected progesterone systemically, and therefore, the brain regions mediating this enhancement are not clear. As such, this study was designed to determine the role of the dorsal hippocampus in mediating the beneficial effect of progesterone on object recognition. Young ovariectomized C57BL/6 mice were trained in a hippocampal-dependent object recognition task utilizing two identical objects, and then immediately or 2 hrs afterwards, received bilateral dorsal hippocampal infusions of vehicle or 0.01, 0.1, or 1.0 μg/μl water-soluble progesterone. Forty-eight hours later, object recognition memory was tested using a previously explored object and a novel object. Relative to the vehicle group, memory for the familiar object was enhanced in all groups receiving immediate infusions of progesterone. Progesterone infusion delayed 2 hrs after training did not affect object recognition. These data suggest that the dorsal hippocampus may play a critical role in progesterone-induced enhancement of object recognition.
Much recent interest has been focused on the ability of sex-steroid hormones, such as estrogens and progestins, to modulate cognitive function due, in part, to conflicting reports from studies using estrogen plus progestin therapy as a treatment for memory loss in menopausal women. For example, several studies in post-menopausal women have demonstrated beneficial effects of estrogen plus progestin treatment on working memory (Duff & Hampson, 2000), verbal memory (Grigorova & Sherwin, 2006; Maki et al., 2001), episodic memory, and verbal fluency (Yonker et al., 2006). An additional study reported a trend for improvement in figural memory, but also a detrimental effect on verbal memory (Resnick et al., 2006). Consistent with the latter effect, other reports find no benefit of estrogen plus progestin treatment on global cognitive function (Rapp et al., 2003), declarative memory, or working memory (Wolf et al., 2005). Together, these data suggest a potentially complex interaction between estrogens and progestins with regard to mnemonic function that may depend on various factors including timing of treatment, specific hormone formulation, and type of memory tested. One way to help tease apart the interactions between the two hormones may be to determine how each modulates memory, both independently and in combination, in animal models of memory formation.
The primary endogenous sources of progesterone are the adrenals and gonads (Nelson, 2000). Although both estrogens and progestins are produced in the ovaries in response to stimulation from the brain, the vast majority of basic research in this area has focused on effects of estrogens, such as 17β-estradiol, rather than progestins, such as progesterone. As a result, much less is known about the role progesterone may play in modulating memory, both alone and in combination with estradiol. However, there is abundant evidence to suggest that both estradiol and progesterone alter the physiology of regions of the brain that are critical for learning and memory, such as the hippocampus. For example, estradiol and progesterone have both been demonstrated to modulate CA1 dendritic spine density in rats (Woolley & McEwen, 1993). Whereas estradiol reverses the decrease in synaptic spine density of pyramidal CA1 cells following ovariectomy, the effects of progesterone are more complex, leading to an initial increase in spine density during the first 2–6 hours after treatment, with a subsequent decrease over the next 18 hours (Woolley & McEwen, 1993). Further, estradiol and progesterone in female monkeys have both been shown to increase hippocampal synaptic proteins; estradiol alone increased syntaxin, synaptophysin, and spinophilin, whereas progesterone alone increased synaptophysin (Choi et al., 2003). Interestingly, co-administration of both estradiol and progesterone reversed these increases (Choi et al., 2003). Additionally, estradiol and progesterone can modulate intracellular signaling. For example, both estradiol and progesterone alone activate the extracellular signal-regulated kinase/mitogen-activated protein kinase (ERK/MAPK) pathway, and induce the nuclear translocation of phosphorylated ERK in hippocampal cultures (Nilsen & Brinton, 2002; Nilsen & Brinton, 2003).
Although several previous studies have examined the ability of estradiol and progesterone to modulate hippocampal-dependent memory in young ovariectomized rats and mice, these findings are inconsistent. For example, estradiol plus progesterone treatment improved spatial working memory in the T-maze (Gibbs, 2000) and spatial reference memory in the Morris water maze (Markham et al., 2002), reversed scopolamine-induced impairments of spatial working and reference memory in a radial arm maze (Tanabe, 2004), protected against colchicine-induced impairments in water maze performance (Vongher & Frye, 1999), and improved object recognition (Walf et al., 2006). In contrast, estradiol plus progesterone treatment has also been reported to impair or not affect spatial memory in the water maze in rats (Bimonte-Nelson et al., 2006; Chesler & Juraska, 2000), to impair footshock learning in mice (Farr et al., 1995), and to block the neuroprotective effects of estradiol against kainite lesions in rats (Rosario et al., 2006). Inconsistencies among these studies may be due to a number of methodological factors, as many aspects of experimental design differed among them. However, these discrepant findings may also suggest that attempting to understand how progesterone modulates memory by administering it with estradiol is not the most effective method of examining the role of this hormone in memory formation. Rather, examining the effects of progesterone alone on memory may shed more light on this subject.
The few studies that have examined the effects of progesterone alone on memory suggest an important effect of timing of administration. Chronic administration of progesterone prior to training (i.e., pre-training) impaired both spatial working memory and footshock avoidance learning in young ovariectomized rats and mice (Bimonte-Nelson et al., 2004; Farr et al., 1995). Acute pre-training progesterone injection had no effect on spatial memory in young ovariectomized rats (Chesler & Juraska, 2000; Sato et al., 2004). In contrast, single intraperitoneal injections of progesterone administered to young ovariectomized rats immediately after training (i.e., post-training) improved memory in Y-maze inhibitory avoidance and object recognition tasks (Frye & Lacey, 2000; Walf et al., 2006; Harburger et al., 2008). Systemic post-training progesterone injections also improved object recognition in ovariectomized middle-aged and aged mice (Lewis et al., 2008b). The discrepancy between the effects of pre- and post-training injections could stem from differences in task, dose, and/or side effects of progesterone treatment, given that this hormone has anxiolytic and analgesic effects (Bitran et al., 1991a; Bitran et al., 1991b; Frye & Duncan, 1994) which could influence non-mnemonic aspects of task performance (e.g., motor activity, motivation). As such, pre-training administration makes it more difficult to accurately interpret the effects of progesterone on memory, whereas administration of rapidly-metabolized forms of progesterone immediately after training allow for effects on memory consolidation to be observed on later testing in the absence of non-mnemonic confounds due to circulating progesterone.
The fact that the aforementioned studies administered progesterone systemically prevents specific localization of the modulatory effects of progesterone on the brain. The dorsal hippocampus is critical for several types of memory, including spatial and working memory, (Hock & Bunsey, 1998; Moser, Moser, and Andersen, 1993; Packard & McGaugh, 1996), and has been implicated more recently in object recognition memory (Baker & Kim, 2002; Broadbent et al., 2004). We have previously shown that estradiol infusion into the dorsal hippocampus enhances object recognition memory (Fernandez et al, 2008). Progesterone receptors are located in dorsal hippocampus (Guerra-Araiza et al., 2002; Guerra-Araiza et al., 2003; Kato et al., 1994), and progesterone can rapidly alter hippocampal morphology and cell signaling as described above (Choi et al., 2003; Nilsen & Brinton, 2002; Nilsen & Brinton, 2003; Woolley & McEwen, 1993). Further, progesterone administration has been demonstrated to influence performance in memory tasks mediated by the dorsal hippocampus (e.g., Frye & Lacey, 2000; Gibbs, 2000; Harburger et al., 2008). As such, the dorsal hippocampus may play a critical role in the mnemonic effects of progesterone.
Therefore, the present study was designed to examine the effects of dorsal hippocampal progesterone infusions on object recognition memory. Young ovariectomized mice were implanted bilaterally with guide cannulae aimed at the dorsal hippocampus. Mice were then trained in an object recognition task, immediately infused with vehicle or one of three doses of rapidly metabolized water-soluble progesterone, and then tested 48 hours later. This design was based on previous studies showing that systemic administration of water-soluble progesterone can enhance object recognition in young ovariectomized mice (Harburger et al., 2008). The version of the task used in this experiment is dependent on the hippocampus based on lesion (Clark et al., 2000) and pharmacological (Baker & Kim, 2002; Fernandez et al., 2008) studies. Based on the previous post-training studies discussed above, we hypothesized that intrahippocampal progesterone would enhance object recognition.
Subjects were 46 female C57BL/6 mice received from Taconic (Germantown, NY) at 6 weeks of age. Mice were group-housed, five per shoebox cage in a room with a 12:12 light/dark cycle (lights on at 7:00). All behavioral testing was conducted during the light phase, and animals had ad libitum access to food and water. All procedures were approved by the Institutional Animal Care and Use Committee of Yale University, and conformed to the guidelines established by the National Institute of Health Guide for the Care and Use of Laboratory Animals.
All mice were ovariectomized prior to treatment in order to eliminate endogenous sources of ovarian hormones. Surgery was conducted at least one week prior to testing as previously described (Lewis et al., 2008a). Briefly, mice were anesthetized with isoflurane gas (2% isoflurane in 100% oxygen) and placed into a sterotaxic apparatus (Kopf Instruments, Tujunga, CA) in preparation for cannula implantation. The ovaries and tips of the uterine horns were isolated and removed through bilateral dorsal incisions made at the level of the pelvis. The muscle wall was sutured and the skin closed with wound clips.
Cannula implantation took place immediately following ovariectomy as previously described (Lewis et al., 2008a). The skull was exposed through an incision in the scalp. Bregma and Lambda were aligned in the same horizontal plane, and small bilateral holes (1 cm in diameter) were drilled for placement of stainless steel guide cannulae (C232GC, 22 gauge, Plastics One, Roanoke, VA) with dummy cannulae (C232DC; Plastics One). Cannulae were directed toward dorsal hippocampus (1.7 mm posterior to bregma, +/− 1.5 mm lateral to midline, and 2.3 mm ventral to the surface of the skull), based on Paxinos and Franklin (1997). Cannulae were affixed to the skull with dental cement which also closed the wound.
Mice were allowed to recover for one week after surgery and received 30 mg/kg ibuprofen in the drinking water as an analgesic during recovery. After surgery, mice were housed singly for the remainder of the experiment.
Mice were randomly divided into four groups receiving vehicle (2-hydroxypropyl-β-cyclodextrin (HBC) complex; Sigma, St. Louis, MO; n = 12) or 0.01 (n = 11), 0.1 (n = 14), 1 (n = 9) μg of water-soluble progesterone (2-hydroxyproyl-β-cyclodextrin-encapsulated progesterone; Sigma) dissolved in physiological saline. Progesterone was purchased already encapsulated in HBC, and concentrations were based on the amount of progesterone, not the amount of HBC-progesterone, in solution. HBC enhances the water solubility of progesterone, and does not detrimentally affect the pharmacokinetic properties of steroid hormones (Brewster et al., 1995). Because this form of progesterone is metabolized within hours (Pitha et al., 1986), it is likely not in circulation during either phase of behavioral testing, thus, minimizing confounding effect of non-mnemonic performance factors (e.g., motivation, anxiety) on performance. Vehicle or progesterone was infused immediately following the sample phase of the object recognition task as described previously (Lewis et al., 2008a).
Briefly, dummy cannulae were replaced with injection cannulae (22 gauge; extending 0.8 mm beyond the tip of the guide cannula) that were attached to polyethylene tubing (PE50; Plastics One). This tubing was connected to a 10 μl Hamilton syringe controlled by a microinfusion pump (KDS 100, KD Scientific; New Hope, PA). After the injection cannula was inserted into the guide cannula, vehicle or one of three concentrations of progesterone (0.02, 0.2, or 2.0 μg/μl) was infused at a rate of 0.5 μl/min for 1 min at a volume of 0.5 μl/side of the hippocampus, resulting in doses of 0.01, 0.1, and 1.0 μg. Methylene blue infused into the entorhinal cortex using this protocol diffuses approximately 1 mm3 (Lewis & Gould, 2007). Although the diffusion properties of HBC-progesterone in the hippocampus may differ from those of methylene blue, these data suggest that infusions were likely limited to the dorsal hippocampus.
To demonstrate that the effects of progesterone on object recognition were limited to the first two hours after training, an additional set of mice was infused as described above with vehicle or 0.1 μg of progesterone two hours after training.
Mice were tested in a hippocampal-dependent version of the novel object recognition task as previously described (Frick & Gresack, 2003) one week after surgery. The testing chamber was a white open field box (approximately 58 cm long by 58 cm wide by 46 cm high). A camera mounted on the ceiling above the chamber was connected to a monitor, VCR, and computer outside of the room that were used to collect data. Mice were first habituated to the testing chamber by placing them in the empty chamber and allowing them to explore freely for five minutes. To control for general levels of activity prior to hormone infusion, locomotor activity was measured by recording crossings of a 5 × 3 grid superimposed over the chamber on the computer monitor. Crossings in the inner grids (inner crossing), outer grids (outer crossings), and all grids (total crossings) were recorded.
Sample phase training occurred the following day. Each mouse was placed in the testing chamber with two identical objects that were positioned near the northeast and northwest corners of the chamber. Mice were allowed to freely explore the objects until the total amount of exploration time for both objects equaled 30 sec. Maintaining a constant exploration time rather than a constant trial time in this fashion controlled for differences in propensity to explore between mice, and ensured that all mice experienced the objects for the same amount of time. This protocol has been shown to critically involve the dorsal hippocampus (Clark et al., 2000; Baker & Kim, 2002). Immediately after completion of the sample phase, mice were infused with vehicle or progesterone and then were returned to their home cages. The chamber and objects were cleaned with a 70% ethanol solution between mice.
Retention testing in the choice phase occurred 48 hours later. Generally, untreated young ovariectomized mice do not demonstrate memory after 48 hours (Lewis et al., 2008a). Thus, the 48 hour delay is appropriate for examining hormone-mediated enhancements in memory. Each mouse was placed back into the testing chamber along with one object previously seen in the sample phase (familiar) and one novel object. The position (northeast or northwest corner) and identity of the novel object was counterbalanced across mice. Mice were again allowed to explore until they had accrued 30 sec of object exploration. Time spent with the objects was recorded. Because mice innately tend to prefer novelty, a mouse that remembers the familiar object from the sample phase should spend significantly more time than chance (15 sec) exploring the novel object and less time than chance exploring the familiar object. Such a result would suggest a preference for the novel object. In addition, elapsed time to accumulate 30 seconds of object exploration was recorded to control for effects of progesterone infusion on non-mnemonic aspects of performance. Time spent exploring the object on either side of the chamber was also recorded in order to evaluate the possibility of side bias. Again, 70% ethanol was used to clean the chamber and objects between mice.
One-way ANOVAs were conducted on inner, outer and total grid crossings collected during habituation to examine pre-existing differences in locomotor activity.
To assess memory retention, one-way analyses of variance (ANOVAs) were conducted on time spent exploring the novel object during retention testing. Separate analyses were conducted on the data from the four dose-response groups and the data from the two groups testing effects of a two-hour delay (see below). A one-way ANOVA was also conducted on elapsed time during retention testing to measure effects of progesterone infusion on non-mnemonic aspects of task performance.
For all ANOVAs, significant main effects of Treatment were followed up with Fisher’s LSD post-hoc tests. All analyses were conducted using SPSS (SPSS Inc., Chicago, IL).
Cannula placements were examined after behavioral testing to ensure correct placement. Mice were cervically decapitated and brains were removed and stored in 10% formalin solution (Fisher Scientific, Pittsburgh, PA). A freezing microtome was used to cut coronal sections (60 μm thick) proximal to cannula tracts and mounted on PLUS slides (Fisher Scientific) and dried overnight. Cannula placements were verified under a light microscope. Because injection sites were within dorsal hippocampus for all mice (Figure 1), no mice were excluded from the data analyses.
Grid crossings during habituation are presented in Table 1. The main effect of Treatment was not significant for the inner, outer, or total grid crossings during the habituation phase of testing (inner, F(3, 37) = 1.038, p > 0.05; outer, F(3, 37) = 2.194, p > 0.05; total, F(3, 37) = 1.828, p > 0.05), indicating that the groups exhibited similar levels of general locomotor activity prior to progesterone infusion.
The main effect of Treatment was significant for time spent with the novel object during retention testing (F (3, 42) = 4.26, p = 0.01; Figure 2), indicating a significant effect of progesterone on object recognition. Post-hoc tests revealed that each progesterone-treated group explored the novel object significantly more than the vehicle group (0.01 μg, p = 0.006; 0.1 μg, p = 0.015; 1.0 μg, p = 0.004; Figure 2). These data suggest that bilateral dorsal hippocampal progesterone treatment enhances object recognition in young ovariectomized mice.
The main effect of Treatment was not significant for elapsed time to accumulate 30 s of exploration during retention testing (F(3, 42) = 1.082, p > .05), demonstrating that hormone treatment did not affect motivation to explore the objects. Elapsed time (± standard error of the mean) for each group was as follows: Vehicle: 522.1 ± 92.8; 0.01 μg: 395.3 ± 86.3; 0.1 μg: 567.3 ± 150.6; 1.0 μg: 738.0 ± 149.8.
To demonstrate that the effects of progesterone were limited to the period immediately following training, another set of mice received bilateral dorsal hippocampal infusions of vehicle or a moderate dose of progesterone (0.1 μg) 2 hrs after training. There was no difference in novel object exploration during choice between mice receiving delayed infusion of vehicle or of 0.1 μg progesterone (t (6) = 0.711, p > 0.05, Figure 3). These data suggest that the effects of the immediate progesterone treatments observed above occur within the first two hours after training.
The present data demonstrate that immediate post-training infusion of progesterone into the dorsal hippocampus enhances object recognition in young ovariectomized female mice, and that this effect occurs within two hours of infusion. Although peripheral post-training injections of progesterone have previously been shown to enhance object recognition throughout the lifespan in female mice using this object recognition paradigm (Harburger et al., 2008; Lewis et al., 2008b), the present data suggest that the dorsal hippocampus may be critical for the beneficial effects of progesterone on object recognition. Thus, this finding is an important step towards elucidating the neural mechanisms by which progesterone modulates memory.
The results of the present study are consistent with previous findings that post-training progesterone treatment enhances object recognition using a different testing protocol (Frye & Lacey, 2000; Walf et al., 2006), and fits particularly well with data from our laboratory using the same testing protocol showing that progesterone treatment immediately post-training enhances object recognition in young (Harburger et al., 2008), middle-aged, and aged (Lewis et al., 2008b) female mice. It is also consistent with beneficial effects of post-training progesterone on spatial memory consolidation in the Morris water maze in aged female mice (Lewis et al., 2008b). However, the beneficial effects of post-training hormone administration contrast with previous reports of detrimental effects (Bimonte-Nelson et al., 2004; Farr et al, 1995) or no effects (Chesler & Juraska, 2000; Sato et al., 2004) of pre-training progesterone administration in various learning and memory tasks including the Morris water maze, radial arm maze, and inhibitory avoidance. Given that all post-training studies to date have shown a beneficial effect of progesterone, whereas pre-training studies have not, it is likely that timing of administration is critical to the effect of this hormone. Thus, anxiolytic and analgesic effects (Bitran et al., 1991a; Bitran et al., 1991b; Frye and Duncan, 1994) of circulating progesterone and progestin metabolites may interfere with task performance if treatment is administered prior to training. However, such effects may still influence behavior during retention testing if they lead to long-term genomic alterations, a possibility that has yet to be fully addressed. The timing of progesterone treatment may also be important with regard to its effects on the brain. The present data, and previous reports, demonstrate that object recognition memory, in particular, is sensitive to modulation of the dorsal hippocampus (Clark et al., 2000; Baker & Kim, 2002; Fernandez et al., 2008). In the dorsal hippocampus, progesterone has a biphasic effect on CA1 synaptic spine density (Woolley & McEwen, 1993), such that progesterone increases spine density in the first 2–6 hours following treatment, and subsequently decreases spine density below baseline over the next 18 hours. Such biphasic morphological changes in the hippocampus could also account for the apparently conflicting reports of the effects of progesterone administration on hippocampal memory and underscore the sensitivity of such effects to timing of treatment.
Although the present data indicate that the progesterone-induced enhancement of object recognition is associated with the dorsal hippocampus, the specific cellular mechanisms underlying this enhancement remain unclear. As previously described, progesterone reportedly increases specific synaptic proteins (Choi, et al., 2003), and exerts a biphasic modulatory effect on CA1 dendritic spines (Woolley & McEwen, 1993). Additionally, progesterone, like estradiol, has been shown to modulate intracellular signaling in hippocampal cultures (Nilsen & Brinton, 2002; Nilsen & Brinton, 2003), specifically the ERK/MAPK pathway, leading to an increase in nuclear translocation of phosphorylated ERK. Interestingly, the effects of progesterone on the ERK phosphorylation can be blocked by both progesterone and estrogen receptor antagonists (Migliaccio, et al., 1998). Progesterone has also been shown to influence the cAMP/PKA (Collado et al., 1985) and PI-3K (Singh, 2001) pathways. Given the importance of these signaling pathways to hippocampal memory (Adams and Sweatt, 2002), activation of cell signaling in the dorsal hippocampus may be critical to the mnemonic effects of progesterone.
The receptor mechanisms through which progesterone affects memory are unclear, although there are a number of possibilities (Singh, 2005). Among these possibilities are nuclear progesterone receptors (PRA and PRB), which have been demonstrated to inhibit estradiol-induced transcription (Chalbos & Galtier, 1994), or to more recently described membrane-bound progesterone receptors, which have been demonstrated to alter both cAMP and MAPK activity in vitro (Zhu et al., 2003). Progesterone can modulate GABAA receptors (Reddy et al., 2005), which are common in the hippocampus and have been demonstrated to affect hippocampal memory tasks (Car, Luziemka-Leska, & Wisniewski, 1996; Mohler et al., 2008), although it is unclear if effects on behavior are due to specifically to a disruption of memory or of non-mnemonic aspects of task performance such as anxiety. Additionally, the effects of progesterone on memory could be the indirect result of one of the many progesterone metabolites (e.g. androgens, estrogens, allopregnanolone), which can independently influence memory (Melchior & Ritzmann, 1996; Silvers et al., 2003). As such, understanding how this complex hormone influences memory will prove challenging, given the many possible routes through which progesterone can modulate hippocampal physiology. Nevertheless, such an understanding should be possible to achieve given the consistency of post-training progesterone’s effects on memory.
In conclusion, the present data suggest that the dorsal hippocampus plays a critical role in progesterone-based enhancement of object recognition, which extends the previous literature on systemic post-training progesterone treatment. Identifying a specific brain region, such as the hippocampus, as critical for the beneficial mnemonic effects of progesterone is a necessary first step in understanding how this hormone modulates memory alone and in combination with estradiol. Future studies will undoubtedly reveal how progestins and estrogens interact in the hippocampus and related brain regions to modulate memory processes, allowing for more effective treatment of age-related cognitive decline, particularly in post-menopausal women.
This work was supported by Yale University and NIH grant AG022525 to K.M.F.
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.