Age-dependent effects of social isolation on mesolimbic dopamine release.

In humans, social isolation is a known risk factor for disorders such as substance use disorder and depression. In rodents, social isolation is a commonly used environmental manipulation that increases the occurrence of behaviors related to these disorders. Age is thought to influence the effects of social isolation, but this predictive relationship is not well-understood. The present study aimed to determine the effects of social isolation on mesolimbic dopamine release at different developmental age points in mice. The experimental ages and their corresponding comparison to human age stages are as follows: 1 month = adolescence, 4 months = mature adulthood, 12 months = middle adulthood, and 18 months = older adult. Mice were socially isolated for 6 weeks during these developmental stages, then in vivo fixed potential amperometry with recording electrodes in the nucleus accumbens was used to measure stimulation-evoked dopamine release, the synaptic half-life of dopamine, dopamine autoreceptor functioning, and the dopaminergic response to cocaine. Isolation altered dopamine functioning in an age-dependent manner. Specifically, isolation increased dopamine release in the adult ages, but not adolescence, potentially due to increased inhibitory effects of dopamine autoreceptors following adolescent social isolation. Regarding the cocaine challenge, isolation increased dopaminergic responses to cocaine in adolescent mice, but not the adult mice. These findings have implications for clinical and experimental settings. Elucidating the relationship between age, social isolation, and neurochemical changes associated with substance use disorder and depression may lead to improvements in preventing and treating these disorders.

Introduction

Social isolation has been repeatedly associated with poor mental health and increased prevalence of substance use disorder (SUD) and depression (see Global Council on Brain Health 2017; Holt-Lunstad et al. 2015), and both objective and subjective types of isolation have increased in prevalence due to the COVID-19 pandemic (Loades et al. 2020). Although time will further elucidate the impact of pandemic-induced isolation on mental health, early studies showed particularly detrimental mental health effects on adolescent and elderly populations (Berg-Weger and Morley 2020; Loades et al. 2020). In the current study, we used mice to examine the effects of social isolation on neurochemical functioning related to SUD and depression in different age groups.

In rodents, social isolation is a commonly used environmental manipulation to increase the expression of stress hormones, drug-seeking, and the occurrence of behaviors related to anxiety and depression (Brenes et al. 2008; Fone and Porkess 2008; Ieraci et al. 2016; Kokare et al. 2010; Walker et al. 2019). In these studies, social isolation was applied during adolescence or young adulthood, developmental stages in which the brain is still plastic and considered more vulnerable to environmental stressors (Arakawa 2018). Social isolation during adulthood has been shown to induce similar behaviors related to substance use, anxiety, and depression, but often to a lesser degree than isolation during adolescence (Fone and Porkess 2008; Ieraci et al. 2016; Zorzo et al. 2019). Although several studies suggest that late adulthood may be another age at which rodents and humans become more vulnerable to the detrimental effects of social isolation (Arranz et al. 2009; Sullens et al. 2021), no related studies have compared isolation-induced effects in multiple developmental stages that span from adolescence to late adulthood.

The mesolimbic dopamine system is one modulator of behaviors related to SUD and depression. The mesolimbic dopamine pathway consists of dopamine cell bodies in the ventral tegmental area (VTA) that project to limbic nuclei, most notably the nucleus accumbens (NAc) (Baik 2013; Ikemoto and Panksepp 1999; Wise 2008). Social isolation during adolescence or young adulthood has been shown to increase dopamine release, dopamine uptake rates, dopamine transporter (DAT) expression levels, and psychostimulant-induced dopamine transmission (Han et al. 2012; Yorgason et al. 2013, 2016). Isolation-induced hyperdopaminergic profiles fit with behavioral studies that show increased self-administration and conditioned place preference to substances of abuse, such as ethanol, morphine, cocaine, and other psychostimulants (Ding et al. 2005; Fone and Porkess 2008; McCool and Chappell 2009; Whitaker et al. 2013; Schenk et al. 1987; Zakharova et al. 2009). On the other end of the spectrum, reduced functioning in this pathway is associated with anhedonia and decreased motivation, both symptoms associated with depression (Perona et al. 2008; Scheggi et al. 2018; Wise 2008). Accordingly, several antidepressants function either fully or partly as dopamine agonists. A functional mesolimbic dopamine system is crucial for the healthy regulation of reward-seeking, motivation, and mood.

The present study aimed to determine the effects of social isolation on dopamine release at different developmental age points. Dopamine functioning is known to alter with age alone, and dopamine neurons are thought to be particularly vulnerable to the aging process. Dopamine receptor expression in the NAc reaches its highest point during mid-adolescence, declining during adulthood, following an inverted U-shape of rising and falling (Burke and Miczek 2013; Karkhanis et al. 2019). Similarly, dopamine release and neuronal activity also seem to fit this developmental inverted U-shape with dopamine release and activity being the greatest in young adulthood (Huang et al. 1995; Pitts et al. 2020; Santiago et al. 1993; Stamford 1989). Little is known about the effects of social isolation on dopamine functioning across these ages, especially during mid to late adulthood. Such information is particularly important given the commonality of reduced social interaction during old age and health-related quarantines (Banerjee and Rai 2020; Singh and Misra 2009).

In the current study, mice from four age groups (adolescence, young adulthood, middle-aged adulthood, and old age) were placed in one of two housing conditions (group-housed or socially isolated) for 6 weeks. See Fig. 1 for a comparison of mouse and human age ranges. In vivo fixed potential amperometry was used to measure stimulation-evoked dopamine efflux in the NAc of anesthetized mice. Measures included quantification of dopamine release and two mechanisms that govern extracellular dopamine: dopamine transporter (DAT) and dopamine autoreceptor (DAR) functioning. Also, to examine the effects of aging and social isolation on the dopaminergic response to a commonly researched psychostimulant, all mice received an in-test injection of cocaine, which serves as a dopamine agonist by inhibiting DAT. Overall, we expected that social isolation would have the greatest effects on dopamine functioning when applied during adolescence and late adulthood. Such studies are important for elucidating the relationship between social isolation, aging, and neurochemical changes associated with SUD and depression. An improved understanding of these contributing factors may help in prevention and treatment of these disorders.

Fig. 1

A comparison between the rates at which mice and humans age. Mouse ages are based on the developmental rates of C57BL/6J mice with the assumption of healthy aging in a laboratory setting. Created with BioRender.com

Methods

All experiments were approved by the Institutional Animal Care and Use Committee at the University of Memphis and conducted in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals. Efforts were made to reduce the number of animals used and to minimize pain and discomfort.

Animals and housing conditions

Fifty-five male C57BL/6J mice were obtained from Jackson Laboratory. All mice were housed in a temperature-controlled room at 21 °C with a 12 h light: 12 h dark cycle, with lights on at 0600. Food and water were available ad libitum. Mice were group-housed, 2–5 per cage, until 1, 4, 12, or 18 months. Age of mice roughly translates to human age stages as follows: 1 month = adolescence, 4 months = mature adulthood, 12 months = middle adulthood, and 18 months = older adult (see Fig. 1) (Flurkey et al. 2007; Hagan 2017; Lester 2021). At the appropriate age, mice were either isolated to a cage by themselves or were allowed to remain group-housed. The number of mice per experimental group were as follows: 1 month group-housed n = 7 and isolated n = 7, 4 months group-housed n = 7 and isolated n = 8, 12 months group-housed n = 6 and isolated n = 7, 18 months group-housed n = 6 and isolated n = 7. All mice were in transparent cages on racks in the same room so sensory cues were not limited, but they were inhibited from social isolation (for standard isolation methods see Walker et al. 2019). Mice remained in these housing conditions for 6 weeks, which is a commonly applied isolation period in related research (Yorgason et al. 2013, 2016).

Surgery and dopamine recordings

Dopamine recordings took place at the conclusion of the housing condition period. Mice were permanently anesthetized using urethane (1.5 g/kg, i.p.) and mounted in a stereotaxic frame (David Kopf Instruments), with body temperature maintained at approximately 37 °C. A stimulating electrode (SNE-100, MicroProbes) was inserted into the left VTA (coordinates: AP − 3.3 mm from bregma, ML + 0.3 mm from midline, DV − 4.0 from dura; Paxinos and Franklin 2001). A stainless-steel auxiliary and Ag/AgCl reference electrode combination was positioned in contact with cortical tissue on the contralateral side of the brain (− 2.0 mm from bregma). Finally, a carbon fiber recording electrode with an active recording surface of 500 µm (length) by 7 µm (o.d.) was implanted into the left NAc (coordinates: AP + 1.5 mm from bregma, ML + 1.0 mm from midline, and DV − 4.0 mm from dura) (Paxinos and Franklin 2001). All amperometric recordings were made within a Faraday cage to increase signal to noise ratio. A fixed potential (+ 0.8 V) was applied via the auxiliary electrode, and dopamine oxidation currents were continuously monitored at 10,000 samples per second via an electrometer filtered at 50 Hz (ED401 e-corder and EA162 picostat, eDAQ Inc.). Following surgical set-up, the initial stimulation protocol consisted of 20 monophasic 0.5 ms duration pulses (800 µAmps) at 50 Hz every 30 s to establish a dopamine response. Electrical stimulations were delivered to the stimulating electrode by an optical isolator and programmable pulse generator (Iso-Flex and Master-8, AMPI). In vivo fixed potential amperometry is confirmed as a reliable and valid measure of stimulation-evoked dopamine release in the NAc, providing a temporal resolution suitable to quantify phasic dopamine release, reuptake rates, and DAR functioning (Dugast et al. 1994; Holloway et al. 2018; Lester et al. 2010; Suaud-Chagny et al. 2002).

DAR sensitivity was assessed by applying paired test pulses (T1 and T2, 10 pulses each at 50 Hz, 10 s between T1 and T2) to the VTA every 30 s. Five sets of conditioning pulses (1, 5, 10, 20, and 40; 0.5 ms pulse duration at 15 Hz) were delivered prior to T2 such that there was 0.3 s between the end of the conditioning pulse train and the initiation of T2. These stimulation parameters are identical to those of Fielding et al. (2013), Holloway et al. (2018), and Mittleman et al. (2011) and similar in concept to studies by Benoit-Marand et al. (2000) and Kennedy et al. (1992). Autoreceptor-mediated inhibition of evoked dopamine efflux was calculated as the change in the amplitude of T2 with respect to T1 for each set of conditioning pulses (T2/T1 × 100). Low-to-high DAR functioning is represented as low-to-high percent inhibition of evoked dopamine efflux. In other words, increased DAR functioning results in more inhibition of dopamine release, observed as a reduced amplitude of T2 relative to T1.

Following the autoreceptor test, stimulation parameters were reset to 20 pulses at 50 Hz every 30 s. Baseline levels of VTA stimulation-evoked dopamine were monitored for 5 min before the drug challenge (cocaine administration, 10 mg/kg, i.p.). Stimulation parameters and dopamine recordings continued for 1-h post-injection. Dopamine release was quantified as the magnitude of the response and dopamine uptake, an indication of DAT functioning, was measured by the synaptic half-life of dopamine (the time required for 50% decrease from the maximum evoked increase to the pre-stimulus baseline level) (Benoit-Marand et al. 2000; Estes et al. 2019; Holloway et al. 2018; Mittleman et al. 2011). Given that cocaine is a DAT inhibitor, specifically target proteins that regulate dopamine uptake kinetics and not necessarily release, analysis of changes in dopamine half-life is commonly used as an indication of the influence of a DAT inhibitor (Holloway et al. 2018; Mittleman et al. 2011; Siciliano et al. 2014). Thus, the synaptic half-life of dopamine will be extracted every 10 min following the cocaine injection. Following dopamine recordings, in vitro electrode calibrations were conducted by exposing each carbon fiber recording electrode to a series of known solutions of dopamine concentrations (0.2, 0.4, 0.8, and 1.2 uM) via a flow injection system (Dugast et al. 1994; Prater et al. 2018), allowing conversion of raw current data (nAmp) to dopamine concentration (μM).

Histology

At the conclusion of each amperometric recording, a direct anodic current (100 µAmps for 10 s) was applied to the stimulating electrode to mark electrode placement. Mice were then euthanized with intracardial urethane (0.345 g/ml), and brains were removed and stored in a solution of 10% formalin with 0.1% potassium ferricyanide for at least 1 week then 30% sucrose for at least 1 additional week. Coronal sectioning of each brain was performed using a cryostat at − 20 °C, and electrode placements were identified using a light microscope and recorded on coronal diagrams (Paxinos and Franklin 2001).

Drugs

All chemicals used were obtained from MilliporeSigma. Urethane (U2500) and cocaine hydrochloride (C5776) were dissolved in 0.9% saline. Dopamine hydrochloride (H8502) was dissolved in phosphate-buffered saline (PBS) at pH 7.4.

Data analysis

Baseline dopamine release and half-life

Two-way between-subjects analysis of variances (ANOVAs) were used to determine the effect of age at time of isolation (1, 4, 12, and 18 months old) and housing (isolated and group-housed) on baseline dopamine release and the baseline synaptic half-life of dopamine. When appropriate, independent t tests were used to compare means between housing conditions in each age group.

Dopamine autoreceptor functioning

Autoreceptor-mediated inhibition of evoked dopamine release was expressed as percentage change between test stimulations (T1 and T2) for each set of pre-pulses. A mixed three-way ANOVA was used to determine the effect of age and housing (between-subjects factors) on DAR functioning across the different pre-pulse settings (within-subjects factor). When appropriate, independent t tests were used to compare autoreceptor-mediated dopamine release between housing conditions in each age group.

Dopaminergic response to cocaine

The synaptic half-life of dopamine following the cocaine injection was expressed as percent change of baseline dopamine half-life (with baseline half-life being 100%). The timing of dopamine efflux following drug administration has been associated with abuse liability, with drugs that increase dopamine release quicker being more reinforcing (Volkow and Morales 2015). Thus, we were interested in determining whether age or housing altered the dopaminergic response over time. A mixed three-way ANOVA was used to determine the effects of age and housing (between-subjects factors) on percent change of dopamine half-life following cocaine administration over the 1-h recording period in 10 min intervals (within-subject factor). To assess differences at the peak effect time of cocaine, a two-way between-subjects ANOVA was used to determine the effect of age and housing on percent change in dopamine half-life at 20 min post-injection. When appropriate, independent t tests were used to compare percent change in dopamine half-life at 20 min post-injection between housing conditions in each age group.

Results

Baseline dopamine release and half-life

The tips of the stimulating and recording electrodes were positioned within the anatomical boundaries of the VTA and NAc core, respectively (Fig. 2). Baseline stimulation-evoked dopamine release and synaptic half-life were assessed in each mouse prior to cocaine administration. Regarding baseline dopamine release, no main effect of age was observed [F(3,47) = 0.74, p = 0.533, ηp2 = 0.05]; however, a main effect of housing was observed [F(1,47) = 9.15, p = 0.004, ηp2 = 0.17], with isolated mice displaying greater dopamine release than group-housed mice. There was not a significant interaction between age and housing on baseline dopamine release [F(3,47) = 2.24, p = 0.096, ηp2 = 0.13]; however, the effect size for this interaction was large enough to warrant further exploration of these effects (Cohen 1988). Follow-up tests concluded that in the youngest age group (isolated at 1 month old), no differences were observed between isolated and group-housed mice, t(12) = 0.58, p = 0.570, ηp2 = 0.03. However, housing did alter baseline dopamine release in the young adult mice (isolated at 4 months old) and the middle adult mice (isolated at 12 months old), with isolated mice displaying increased dopamine release concentrations compared to group-housed mice (young adults: t(13) =  − 2.60, p = 0.022, ηp2 = 0.34; middle-age adults: t(11) = − 2.42, p = 0.034, ηp2 = 0.35). In the older mice (isolated at 18 months), no differences in baseline dopamine release were observed between isolated and group-housed mice, t(11) =  − 1.04, p = 0.319, ηp2 = 0.09 (Fig. 3a, b).

Fig. 2

Representative coronal sections of the mouse brain (adapted from the atlas of Paxinos and Franklin 2001), with gray shaded areas indicating the placements of (A) stimulating electrodes in the ventral tegmental area (VTA) and amperometric recording electrodes in the nucleus accumbens (NAc). Numbers correspond to mm from bregma

Fig. 3

Baseline (pre-cocaine) dopamine release and half-life. Profiles indicate representative responses from each age and housing group (a). Significant differences in mean (± SEM) dopamine release for housing were observed, with isolated (Iso) mice displaying increased dopamine release compared to group-housed mice in the young adult and middle-aged adult groups (b). No significant differences in mean (± SEM) dopamine half-life were observed between age or housing groups (c). *Indicates p < 0.05

Regarding baseline dopamine half-life, no main effects of age or housing were observed [age: F(3,47) = 1.16, p = 0.337, ηp2 = 0.07; housing: F(1,47) = 1.23, p = 0.273, ηp2 = 0.03], and there was not a significant interaction between age and housing [F(3,47) = 0.56, p = 0.643, ηp2 = 0.04]. Thus, none of the experimental manipulations altered the time required for the stimulation-evoked dopamine to be cleared from the synapse (Fig. 3a, c).

Dopamine autoreceptor functioning

Autoreceptor-mediated inhibition of evoked dopamine release was expressed in terms of percent change of dopamine release between test stimulations (T2/T1 × 100) for each set of conditioning pre-pulses. A lower percentage indicates greater autoreceptor-mediated inhibition of dopamine release. As expected, there was a significant main effect of the number of pre-pulses on autoreceptor-mediated dopamine release [F(6, 270) = 250.28, p < 0.001, ηp2 = 0.85], with indications that autoreceptor-mediated dopamine release decreased as the number of pre-pulses increased. In other words, as the number of pre-pulses increases, DAR functioning increases also, resulting in greater inhibition of dopamine release. Neither age nor housing altered this pattern of autoreceptor-mediated dopamine release across the conditioning pre-pulses (pre-pulses × age: F(18,270) = 0.46, p = 0.971, ηp2 = 0.03; pre-pulses × housing: F(6,270) = 0.21, p = 0.972, ηp2 = 0.01) However, there was a significant three-way interaction between age, housing, and number of pre-pulses [F(18,270) = 1.713, p = 0.037, ηp2 = 0.10], indicating that isolation altered DAR functioning differently depending on the age of the mice. Follow-up analyses assessed differences in autoreceptor-mediated dopamine release between group-housed and isolated mice in each age group. The only significant differences were found in the youngest age group (isolated at 1 month old), with isolated mice displaying reduced dopamine release following 40 and 80 pre-pulses compared to group-housed mice (40 pre-pulses: t(10) = 3.47, p = 0.006, ηp2 = 0.55; 80 pre-pulses: t(10) = 4.21, p = 0.002, ηp2 = 0.64). These differences indicate greater DAR functioning in the isolated mice at this age group (Fig. 4).

Fig. 4

Autoreceptor-mediated inhibition of dopamine release. Stimulation parameters were set to include two test stimulations (T1, T2) and a varying number of conditioning pre-pulses (pp). Two example responses are depicted (a, b). Greater decreases in dopamine release (% of T2/T1) indicates increased autoreceptor functioning. The only significant effect of isolation on mean (± SEM) differences in autoreceptor functioning were found in adolescent age mice (c). Isolation did not alter dopamine autoreceptor functioning in the other age groups (d–f). *Indicates p < 0.05

Dopaminergic response to cocaine

The synaptic half-life of dopamine was measured in 10 min intervals following the cocaine challenge and was converted into percent change, with baseline representing 100%. As expected, there was a significant main effect of time post injection on percent change in dopamine half-life [F(6,270) = 77.26, p < 0.001, ηp2 = 0.63]. Neither age nor housing altered the percent change in dopamine half-life over time following the cocaine injection [time × age: F(18,270) = 1.09, p = 0.361, ηp2 = 0.07; time × housing: F(6,270) = 0.92, p = 0.481, ηp2 = 0.02], and there was no significant three-way interaction between time, age, and housing [F(18,270) = 0.72, p = 0.786, ηp2 = 0.05]. Thus, neither age nor housing nor the interaction of the two variables altered the timing of the dopaminergic response to cocaine.

Next, we compared group differences in the dopaminergic response to cocaine at the time of cocaine’s peak effect (20 min post injection). A significant main effect of age was observed [F(3,45) = 3.52, p = 0.022, ηp2 = 0.19], with the following percent changes in dopamine half-life averaged across housing groups: adolescent M = 222.41%, SEM = 15.81; young adult M = 220.65%, SEM = 14.41; middle-age adult M = 175.22%, SEM = 8.25; old adult M = 186.67%, SEM = 16.12. No main effect of housing was observed on percent change in dopamine half-life 20 min post cocaine [F(1,45) = 0.80, p = 0.376, ηp2 = 0.02]; however, a significant interaction between age × housing was observed [F(3,45) = 3.10, p = 0.036, ηp2 = 0.17]. Follow-up analyses were conducted to compare percent change in dopamine half-life at 20 min post injection between housing conditions from each age group. In the youngest group (isolated as adolescents at 1 month old), the percent change in dopamine half-life following cocaine was significantly greater in isolated mice compared to group-housed mice [t(12) = − 2.30, p = 0.040, ηp2 = 0.31], indicating that social isolation during adolescence resulted in a greater dopaminergic response to cocaine. No significant differences in dopamine half-life following cocaine were observed between isolated and group-housed mice in the other age groups [young adult: t(13) = − 0.30, p = 0.771, ηp2 = 0.01; middle-aged adult: t(11) = − 1.91, p = 0.083, ηp2 = 0.25; older adult: t(9) = 1.82, p = 0.103, ηp2 = 0.27] (Fig. 5).

Fig. 5

Dopaminergic response to cocaine. Neither age nor housing altered the pattern of percent change in dopamine half-life over the 1-h recording period following cocaine administration (a–d). Profiles indicate example responses from group-housed and isolated (Iso) adolescent mice 20 min post cocaine, with the gray line representing pre-drug (e). A significant interaction between age and housing was observed in mean (± SEM) percent change in dopamine half-life at 20 min post injection. In adolescents, isolated mice displayed a greater dopaminergic response to cocaine compared to group-housed mice (f). *Indicates p < 0.05

Discussion

Social isolation is a prevalent public health risk factor that, although certainly not new, has been cited as part of a “loneliness epidemic of modern society” within the last twenty years (Hämmig 2019; Killeen 1998). Social isolation is a risk factor for many psychiatric disorders including SUD and depression, with studies indicating particular vulnerabilities in adolescent and elderly populations (Berg-Weger and Morley 2020; Loades et al. 2020). The present study aimed to examine the effect of social isolation on mesolimbic dopamine functioning in mice across 4 age groups. The mesolimbic dopamine system modulates behaviors related to reward, motivation, and mood (Baik 2013; Berg-Weger and Morley 2020; Ikemoto and Panksepp 1999; Wise 2008), and dysfunction of this pathway is implicated in SUD and depression. Overall, we found that age by itself altered the measured aspects of dopamine functioning and that age also influenced the effects of social isolation on mesolimbic dopamine functioning.

Age effects

The mesolimbic dopamine system is thought to alter with development/aging; however, in the present study, no main effect of age was observed on baseline (pre-cocaine) phasic release. Endogenous dopamine release and neuronal activity have been shown to exhibit an inverted U-shape pattern across the lifespan of rodents and humans, with dopamine release and activity being the greatest towards the end of adolescence and declining with old age (Burke and Miczek 2013; Karkhanis et al. 2019; Huang et al. 1995; Pitts et al. 2020; Santiago et al. 1993; Stamford 1989). The present study measured VTA stimulation-evoked dopamine release. Electrical stimulation negates dopamine release differences that are due to changes in neuronal activity; therefore, the present findings suggest that previously observed age-dependent differences in dopamine transmission are related more to neuronal activity levels or firing rate rather than properties of dopamine release. Another possibility is that our study design missed the adolescent peak in dopamine release, as the youngest group of mice were 9 weeks old at the time of dopamine recordings (placed in housing conditions at 3 weeks old and remained for 6 weeks). The end of adolescence in mice is defined by researchers as being between post-natal day 55–65 (7–9 weeks old) (see Brust et al. 2015). Thus, our “adolescent” group was adolescent at the time of housing condition onset but was transitioning into young adulthood by the time of dopamine recordings. Comparing dopamine responses with mice in the earlier stages of adolescence would address these questions.

A main effect of age was observed in the percent change in dopamine half-life following cocaine, with the adolescent and young adults displaying a greater dopaminergic response to cocaine compared to the older age groups. These results fit with previous studies that have shown enhanced behavioral sensitization to and preference for cocaine and amphetamine in adolescent mice (Caster et al. 2007; Kameda et al. 2011; Schramm-Sapyta et al. 2004). Age-related differences in dopaminergic responses to psychostimulants are inconsistent in the literature. Psychostimulants have been shown to elicit less (Cao et al. 2007; Laviola et al. 2001), more (Stansfield and Kirstein 2005), or equal (Camarini et al. 2008) dopamine in adolescent rodents compared to adults. Discrepancies are likely due to the differences in applied research techniques and the type of dopamine transmission being quantified (phasic vs tonic dopamine). It should also be noted that given the purpose of this project separate analyses of non-isolated mice were not run. Instead, main effects of age and interactive effects of age were assessed. It is possible that the inclusion of the isolation factor masked potential age-effects.

Housing effects at each age group

Adolescent mice

In the present study, social isolation altered dopamine functioning in an age-dependent manner. Stressors such as social isolation are thought to have greater long-term impacts on neural functioning during developmental stages of increased plasticity, with adolescence being a potential sensitive period (Fuhrmann et al. 2015; McEwen 2017; Ver Hoeve et al. 2013). In the adolescent mice, social isolation did not alter baseline (pre-cocaine) dopamine release. These results fit with other studies showing similar baseline dopamine levels in isolated and group-housed rodents (Fabricius et al. 2010; Karkhanis et al. 2014; Wang et al. 2011); however, in some studies, social isolation during adolescence increased dopamine release (Han et al. 2012; Yorgason et al. 2013, 2016). In the present study, the isolated adolescent mice displayed increased DAR functioning compared to the group-housed controls. DARs, which are generally comprised of D2 type receptors, regulate dopamine release by inhibiting dopamine synthesis and inducing hyperpolarization (Cubeddu and Hoffman 1982; Mercuri et al. 1997; Wolf and Roth 1990). Adolescent social isolation has been shown to increase the number of D2 receptors (Han et al. 2012; King et al. 2009), supporting our findings of increased DAR functioning in isolated adolescents. This observed increase in DAR functioning may have counteracted an increase in baseline dopamine release in the adolescent isolated mice.

All mice received an injection of cocaine (10 mg/kg, ip) during NAc dopamine recordings. In adolescent mice, social isolation increased the dopaminergic response to cocaine, as measured by the percent change in the synaptic half-life of dopamine at the peak effect time of cocaine (20 min post injection). These results were expected as others have also shown that social isolation during adolescence leads to a greater effect on extracellular dopamine levels following psychostimulant administration (Fulford and Marsden 2002; Jones et al. 1992; Yorgason et al. 2016) and increased self-administration of psychostimulants (Bozarth et al. 1989; Ding et al. 2005; Howes et al. 2000; Schenk et al. 1987). Karkhanis et al. (2019) suggest that this increased release following dopamine agonists may be due to higher levels of tyrosine hydroxylase, the rate limiting enzyme for dopamine synthesis, in rodents isolated during adolescence. Overall, our findings suggest that social isolation during adolescence increases DAR functioning, which may be a compensatory mechanism to inhibit excess dopamine release, but that the cocaine challenge overpowered the inhibitory control of DARs, resulting in an increased dopaminergic response to cocaine in isolated adolescents compared to group-housed controls. Initial exposure to cocaine has been shown to decrease D2 receptor signaling (Navarro et al. 2013), and repeated cocaine administration attenuates DAR-mediated inhibition of dopamine release (Jones et al. 1996; Pierce et al. 1995). Thus, the effects of cocaine are dependent on D2 functioning, and the present data suggests that mice with more sensitive DAR (i.e., dopamine systems more tightly controlled by DAR), such as the isolated adolescents, may respond more dramatically to cocaine.

The current study did not measure the duration of isolation-induced effects, but recent evidence suggests that social isolation and other stressful experiences during adolescence may result in long-term, even permanent changes in neural function (Bendersky et al. 2021; Burke et al. 2017). In the current study, the mice were isolated for 10–12 weeks, which is a relatively long isolation period, but it is possible that we could have seen more dopaminergic phenotypes develop had we tested later into adulthood. Most studies examining the effects of isolation during adolescence do not resocialize the rodents before testing; however, some effects of adolescent isolation, such as altered novelty-seeking and VTA activity, have been shown to persist long-term, even after resocialization (Einon and Morgan 1977; Whitaker et al. 2013).

Adult mice

The adult mice used in this study ranged in ages from 4 to 18 months at the onset of the housing condition. Young and middle-aged adult mice (4 and 12 months, respectively) exhibited similar isolation-induced dopaminergic phenotypes. Isolation in these age groups increased baseline dopamine release but did not alter the synaptic half-life of dopamine, DAR functioning, or the dopaminergic effect of cocaine. Although dopamine autoreceptor functioning was enhanced in socially isolated adolescent mice, young and middle-aged adult mice did not exhibit this effect. Therefore, the increased inhibitory control on dopamine release seen in adolescents was not present in the adult age cohort. Few studies have examined the effect of social isolation on dopamine release in adult mice. Yorgason et al. (2013) found that social isolation during adulthood does not alter NAc dopamine signaling, while Gomes et al. (2019) found that environmental stressors in adulthood may instead induce a depression-like hypodopaminergic state. The findings of Gomes et al. (2019) are particularly interesting given the inverted U-shape of the development of the dopamine system with receptor expression and activity declining during adulthood (Burke and Miczek 2013). Furthermore, social isolation, especially during late adulthood, may increase the vulnerability of neurons to age-related degeneration (Zigmond and Smeyne 2019).

In the present study, the old mice (isolated at 18 months) did not exhibit isolation-induced baseline dopamine release, DAR functioning, or dopaminergic response to cocaine. Although no phenotypes were significantly different between isolated and group-housed older mice, the pattern across phenotypes appears to be trending in a direction opposite than that of the adolescent mice. Isolated old mice seemed to display a reduced dopaminergic response to cocaine, although not at the significant level (p = 0.075). Similar to the results of Gomes et al. (2019), an isolation-induced hypodopaminergic state may result in reduced motivation, lack of internal drive for rewards, and increased occurrence of depressive episodes. Older populations have already shown to exhibit this vulnerability to social isolation (Perona et al. 2008; Scheggi et al. 2018; Wise 2008), and more research is needed to understand the underlying neural mechanisms. It is important to note that the social isolation paradigm used in the present study was not complete isolation. Mice were in ventilated, transparent cages on racks with other mice; therefore, theoretically, mice could see, hear, and smell other mice even while being “isolated” in a cage with no cage-mates. While this isolation may be enough to alter dopaminergic functioning on adolescent and young adult mice, it may not provide enough of a stressor to alter the dopaminergic system once it has fully matured. Social defeat stress exposure in adulthood has been shown to alter NAc dopamine functioning (Deal et al. 2018); thus, varying degrees of social stress may alter this system differently.

Conclusions

Isolation altered dopamine release measurements in an age-dependent manner. Specifically, isolation increased dopamine release in the adult ages, but not adolescence, potentially due to increased inhibitory effects of DARs following social isolation during adolescence. Regarding the cocaine challenge, isolation increased the dopaminergic response to cocaine in adolescent mice, but not the adult mice. Thus, in some measurements, isolation seemed to have the opposite effect in adolescents compared to the old mice. The mechanisms controlling these differences may provide insight on the relationship between age, social interaction, and behaviors related to motivation and reward. It is important to note that the mice in this study were all male. Further studies are needed to assess these interactive effects of age and isolation in females. Previous studies suggest that dopamine receptor populations fluctuate at different rates across development in males vs females (Andersen et al. 1997; Trainor 2011) and that social isolation may be more anxiogenic (Trainor 2011) or more anxiolytic (Guo et al. 2004) in female rodents compared to males. Others, however, have suggested that males and females do not differ in basal or drug-induced dopamine levels (Egenrieder et al. 2019) and that males and females are affected similarly by isolation (see Mrackova 2020). Understanding the influence of social isolation on dopaminergic functioning across the lifespan is important for both clinical and experimental settings. Animal researchers should consider mouse age and housing conditions when examining dopamine-related behaviors and neural functioning. Elucidating the relationship between age, social isolation, and neurochemical changes associated with SUD and depression may lead to improvements in preventing and treating these disorders.