Short-term testosterone manipulations do not affect cognition or motor function but differentially modulate emotions in young and older male rhesus monkeys

Highlights

• Testosterone manipulations do not affect cognition or motor function but modulate emotions.

• High testosterone increases watching time of negative social scenes in young and old monkeys.

• High testosterone decreases anxiety to novel objects in older monkeys only.

Abstract

Human aging is characterized by declines in cognition and fine motor function as well as improved emotional regulation. In men, declining levels of testosterone (T) with age have been implicated in the development of these age-related changes. However, studies examining the effects of T replacement on cognition, emotion and fine motor function in older men have not provided consistent results. Rhesus monkeys (Macaca mulatta) are excellent models for human cognitive aging and may provide novel insights on this issue. We tested 10 aged intact male rhesus monkeys (mean age = 19, range 15–25) on a battery of cognitive, motor and emotional tasks at baseline and under low or high T experimental conditions. Their performance was compared to that of 6 young males previously tested in the same paradigm (Lacreuse et al., 2009; Lacreuse et al., 2010). Following a 4-week baseline testing period, monkeys were treated with a gonadotropin releasing hormone agonist (Depot Lupron, 200 μg/kg) to suppress endogenous T and were tested on the task battery under a 4-week high T condition (injection of Lupron + T enanthate, 20 mg/kg, n = 8) or 4-week low T condition (injection of Lupron + oil vehicle, n = 8) before crossing over to the opposite treatment. The cognitive tasks consisted of the Delayed Non-Matching-to-Sample (DNMS), the Delayed Response (DR), and the Delayed Recognition Span Test (spatial-DRST). The emotional tasks included an object Approach-Avoidance task and a task in which monkeys were played videos of unfamiliar conspecifics in different emotional context (Social Playbacks). The fine motor task was the Lifesaver task that required monkeys to remove a Lifesaver candy from rods of different complexity.

T manipulations did not significantly affect visual recognition memory, working memory, reference memory or fine motor function at any age. In the Approach-Avoidance task, older monkeys, but not younger monkeys, spent more time in proximity of novel objects in the high T condition relative to the low T condition. In both age groups, high T increased watching time of threatening social stimuli in the Social Playbacks. These results suggest that T affects some aspects of emotional processing but has no effect on fine motor function or cognition in young or older male macaques. It is possible that the duration of T treatment was not long enough to affect cognition or fine motor function or that T levels were too high to improve these outcomes. An alternative explanation for the discrepancies of our findings with some of the cognitive and emotional effects of T reported in rodents and humans may be the use of a chemical castration, which reduced circulating gonadotropins in addition to T. Further studies will investigate whether the luteinizing hormone LH mediates the effects of T on brain function in male primates.

Introduction

Human cognitive aging is characterized by declines in working memory, executive function, long-term memory and speed of processing (Park and Schwartz, 2000). These declines are substantial, develop as early as the 30s and affect everyone (Salthouse, 2010). Deficits in fine motor function also develop with age (Seidler et al., 2010, Smith et al., 1999), and can lead to serious impairments in quality of life and activities of daily living (Desrosiers et al., 1999). In contrast to the age-related declines in cognitive and motor function, emotional regulation has consistently been shown to significantly improve with age in humans (Carstensen et al., 2003, Carstensen et al., 2011).

With the aging population growing at an unprecedented rate worldwide (Kinsella and Wan, 2009), it is imperative to design interventions that would lessen age-related declines in cognitive and motor function. Because both estrogens and androgens affect cognition and motor function (Hampson, 2002) and decline with age in men (Harman et al., 2001) and women (Burger, 1996), hormone replacement therapy (HRT) has long been considered as such an intervention. Most of the work on this issue has focused on women’s health and has provided inconsistent and still highly debated results regarding the potential benefits of HRT on cognitive aging (Maki and Henderson, 2012, Sherwin and Henry, 2008). Fewer data are available regarding the effects of testosterone (T) on cognition in older men, but they also show mixed results. Whereas positive effects of T have been reported for spatial and verbal memory (Cherrier et al., 2001), spatial ability (Janowsky et al., 1994) and working memory (Janowsky et al., 2000), other studies found detrimental effects of T on verbal memory (Maki et al., 2007) or no effect of T on cognition (Emmelot-Vonk et al., 2008, Young et al., 2010). With regards to motor function, very limited data are available, with one recent report finding no effect of T on sequential movement in young or older adults (Siegel et al., 2008). Finally, growing evidence indicates that T increases amygdala reactivity (van Wingen et al., 2011) and has a robust influence on socioemotional processing, at least in young adults (Bos et al., 2012, Eisenegger et al., 2011). Surprisingly little is known about the effects of T on emotions in older adults, although antidepressant effects have been noted (Zarrouf et al., 2009).

Well-controlled investigations in appropriate animal models are clearly needed to shed a new light on the association between T and brain functioning in males. Studies in male rats have shown that T affects brain regions important for cognitive function: T increases the number of dendritic spines in the CA1 field of the hippocampus (Leranth et al., 2003) as well as the prefrontal cortex (Hajszan et al., 2007). In addition, T has a myriad of neurotrophic and neuroprotective effects, which have been implicated in a decreased risk of developing Alzheimer’s disease (Pike et al., 2006, Pike et al., 2008).

Cognitive studies in male rats have generally found that T administration was able to restore memory impairments induced by gonadectomy in a wide range of working memory tasks such as the Morris water maze (Sandstrom et al., 2006, Spritzer et al., 2011), radial arm maze (Gibbs and Johnson, 2008, Spritzer et al., 2008), Y maze (Hawley et al., 2013) or delayed matching-to-position task (Gibbs, 2005). The effects of T are complex, however, and appear dose-, duration- and task-dependent (Spritzer et al., 2011). For example, Spritzer et al (2011) indicated that T differentially affected working memory and reference memory depending on the test used, the dose used and the timing of T administration. There is a paucity of data on the effects of T on cognition in aged male rodents. In one study, T administered to supraphysiological levels increased working memory in aged male rats (Bimonte-Nelson et al., 2003). An earlier study, however, failed to find any beneficial effects of T on spatial learning in aged male rats tested in the Morris water maze (Goudsmit et al., 1990). It has generally been found that the cognitive effects of T in male rodents are not mediated by sensorimotor effects (Kritzer et al., 2001).

Besides its cognitive effects, T has been shown to modulate the reactivity of the hypothalamic–pituitary–adrenal axis (Bingham et al., 2011, Handa et al., 1994) and to have anxiolytic effects in a number of studies in male rodents (Aikey et al., 2002, Bitran et al., 1993, Edinger and Frye, 2004, Fernandez-Guasti and Martinez-Mota, 2005, Toufexis et al., 2005), cattle (Boissy and Bouissou, 1994) and ewes (Bouissou and Vandenheede, 1996). However, whether the anxiolytic properties of T are maintained in aged animals is not known.

Male rhesus monkeys are particularly well suited for investigating the effects of T on cognition, motor function and emotions because they share many characteristics with humans, including in the organization of the brain and neuroendocrine systems (Sorwell and Urbanski, 2013). They exhibit age-related declines in cognition (Baxter, 2001, Herndon and Lacreuse, 2002, Herndon et al., 1997), fine motor function (Lacreuse et al., 2005, Zhang et al., 2000), and T levels (Sorwell and Urbanski, 2013) that closely resemble those observed in humans. In addition, rhesus monkeys are highly social primates that share a vast repertoire of socioemotional behaviors with humans (Kalin and Shelton, 2003). However, there is a paucity of data on age-related changes in socioemotional behaviors. In a longitudinal study spanning the ages of 6 to 20 years, Suomi et al. (1996) reported that older rhesus monkeys retained the temperamental characteristics that they exhibited in early adulthood, but that the distribution of specific activities (e.g., agonism) changed across time. In a related macaque species (M. fascicularis), Veenema et al. (1997) found in contrast that older females withdrew from social interactions and showed increased arousal (as shown by increased frequencies of yawning, scratching, and body shaking), perhaps due to the inability to deal with complex social interactions. Thus, it remains unclear how emotions change with age in macaques and what influence, if any, T may exert on this pattern.

The goal of the present study was to examine the effects of low and high T conditions on cognitive, motor and emotional behavior in aged male rhesus monkeys. We tested 10 aged males in a battery of tasks assessing these functions and compared the results to those of 6 younger male rhesus monkeys tested in the same paradigms in prior studies from our laboratory (Lacreuse et al., 2009, Lacreuse et al., 2010). Based on the literature referenced above, we predicted that T would improve selective aspects of cognition, would have little or no effect on motor function, and would bias emotional processing towards negative stimuli.