The hypothalamic–pituitary–adrenal (HPA) axis is a pivotal neuroendocrine axis that modulates physiological homeostasis, stress responsivity and circadian function. The hypothalamus secretes corticotrophin-releasing hormone, which stimulates the anterior pituitary to release adrenocorticotropic hormone (ACTH). This ultimately signals the adrenal glands to release cortisol, which has widespread effects on the brain, affecting mood, behavior, cognition, and programming of the stress response (Viho et al., 2019).

Increasing evidence points to HPA axis dysregulation in PD. Heightened activity with elevated ACTH and cortisol levels has been reported in people with PD (Stypula et al., 1996). Cortisol levels also appear to be persistently elevated in blood (Hartmann et al., 1997; Breen et al., 2014; Costa et al., 2019) and saliva (Djamshidian et al., 2011; Skogar et al., 2011) of PD patients. Additionally, cortisol levels increase with aging and chronic stress, both of which are risk factors for PD (Lupien et al., 2009, 2018; Moffat et al., 2020). Epidemiological studies have revealed that higher job demands and expectations (Sieurin et al., 2018) and higher number of exposures to stressful events (Vlajinac et al., 2013) associate with an increased risk of development of PD.

Changes in stress and thus changes in cortisol can affect diverse symptoms of PD. Stress modulates motor system function since most parts of the motor system express glucocorticoid receptors, the primary receptor for cortisol (Metz, 2007). In PD mouse models, chronic stress exposure worsens motor deficits, aggravates neurodegeneration of the nigrostriatal system, and completely blocks compensatory recovery of motor tasks (Smith et al., 2008). In people newly diagnosed with PD, higher cortisol levels have been linked with greater deterioration in motor function (Haglin and Backman, 2016). Cortisol levels have also been associated with severity of depression (Seifried et al., 2013), prevalence of anxiety and anhedonia (van den Heuvel et al., 2020) and increased risk-taking in individuals with PD who have impulse compulsive behavior (Djamshidian et al., 2011).

Exercise acts both as a stressor and modifier of the neuroendocrine stress response (Hackney, 2006). Acute exercise, or single exercise session, activates the short-term stress response and increases salivary and plasma cortisol during the initial 30–60 min post-exercise (Wang et al., 2019; Dote-Montero et al., 2021). Moreover, Luger et al. have shown that this response appears to be directly proportional to the volume or intensity of exercise (Luger et al., 1987). An exercise intensity of 50–60% maximal oxygen uptake is the threshold before there is a change in cortisol, and this response amplifies when the intensity increases to 85%. In a study of elderly healthy men, cortisol released in response to acute vigorous exercise suppressed the subsequent cortisol response when presented with a psychosocial stressor, and this effect is dependent on exercise intensity (Caplin et al., 2021). Importantly, post-awakening salivary cortisol secretion and the cortical awakening response is significantly lower on the morning following an exercise session in healthy individuals (Anderson et al., 2023).

Chronic, or long-term, exercise training, defined as multiple sessions per week for duration ≥12 weeks in human studies, has different effects compared to acute exercise. Six months of high-intensity endurance exercise in women with mild cognitive impairment resulted in reduced plasma cortisol levels and improved executive cognitive function (Baker et al., 2010). A meta-analysis by Beserra et al. reports that in individuals with major depressive disorder, regular exercise is associated with decreased daytime cortisol levels but this is influenced by type of exercise and frequency (Beserra et al., 2018). Studies looking at the effects of exercise on cortisol levels in PD are sparse, but one study reports that 6 months of regular high-intensity treadmill exercise reduces cortisol secretion during the post-awakening period in people with PD (Smyth et al., 2019). Further studies are needed to clarify the acute and chronic effects of exercise on cortisol levels in individuals with PD and how this relates to motor and NMS of PD. Given initial studies showing an association between cortisol and mood, cognition, and motor burden in PD, it is possible that exercise-medicated regulation of cortisol may improve these symptoms.

Another key function of the neuroendocrine system is regulation of the circadian rhythm, orchestrated by the central biological clock – the suprachiasmatic nucleus (SCN). SCN is located within the hypothalamus and is attuned to the environmental 24 h cycle by projections from the retinohypothalamic tract, oscillators in peripheral tissues, and circulating neurotransmitters and hormones such as serotonin, cortisol, and melatonin (Saper et al., 2005; Dibner et al., 2010; Blume et al., 2019). SCN neurons adjust their circadian phase according to these photic and non-photic inputs and signal using humoral and autonomic nervous systems to the rest of the body. A key output pathway of the SCN is its projection to the pineal gland where melatonin is produced (Cajochen et al., 2003). Therefore, these input and output pathways are reciprocal. The neurohormone melatonin acts as a transmission signal to coordinate and stabilize the circadian rhythm and regulate hormone secretion, core temperature, cognition, and mood (Cardinali, 2021).

Increasing evidence points to circadian rhythm disruption being a key feature in PD, one that influences not just symptom expression but potentially also PD-associated neurodegeneration. Reduced SCN firing rate has been reported in PD mouse models, which may lead to overall decreased circadian output (Kudo et al., 2011). Breen et al. have reported hypothalamic volume loss and associated decreased melatonin output in people with PD (Breen et al., 2016). A cross-sectional study also confirmed significantly diminished amplitude and amount of melatonin secretion in individuals with PD compared to controls, which associated with excessive daytime sleepiness but not motor function (Videnovic et al., 2014). In a large cohort of newly diagnosed PD, the majority of participants showed abnormal sleep macro-architecture including increased sleep latency, reduced sleep efficiency, and reduced REM sleep, associating with lower circulating melatonin levels (Breen et al., 2014).

Accumulating evidence shows that melatonin has anti-apoptotic, anti-oxidant, anti-inflammatory, and free radical-scavenging properties that protect against mitochondrial dysfunction (Leon et al., 2004; Rodriguez et al., 2004; Esposito and Cuzzocrea, 2010; Singhal et al., 2011, 2012). An additional mechanism by which melatonin may exert its benefit is through stimulating autophagy and inhibiting aggregation of prion-like proteins such as α-synuclein that misfold and aggregate in PD (Jeong et al., 2012). Evidence for improvements in motor function have been observed with administration of 3-10 mg/kg of melatonin in multiple studies using rodent models of PD (Kim et al., 1998; Sharma et al., 2006; Gutierrez-Valdez et al., 2012; Zaitone et al., 2013). On the other hand, a study by Bassani et al. suggests that while administration of melatonin 10 mg/kg protects against dopaminergic loss and improves depression, it does not improve motor deficits in rats (Bassani et al., 2014).

In individuals with mild cognitive impairment, treatment with 3-9 mg of melatonin for 2 years has been shown to improve cognitive and depressive symptoms in addition to sleep (Jean-Louis et al., 1998; Furio et al., 2007; Cardinali et al., 2012). While 3–9 mg of melatonin improves sleep qualities and rapid eye movement (REM) sleep behavior disorder (RBD; Takeuchi et al., 2001; Medeiros et al., 2007), studies have failed to show an improvement in motor symptoms (Medeiros et al., 2007; Liguori et al., 2022). RBD appears in the prodromal phase of PD, preceding motor symptoms of PD by several years (Boeve et al., 2001; Iranzo et al., 2006; Tolosa et al., 2007). Studies have reported that >80% of individuals who initially developed RBD had progressed to develop PD, related synucleinopathy, or dementia within 10 years (Iranzo et al., 2013; Schenck et al., 2013). An interesting case report has described increase in dopamine transport density (as assessed by dopamine transporter scintigraphy) in an elderly male patient with RBD treated with 2 mg of melatonin daily (Kunz and Bes, 2017). After 6 months of treatment, his RBD disappeared. His dopamine transporter scan with binding ratios in the abnormal range before treatment showed improved binding ratios in the normal range 2 years after treatment with melatonin and remained normal even 4 years after the first scan. Although a case report provides only one patient’s response to melatonin therapy, it suggests that melatonin may have neuroprotective benefits, warranting further investigations.

There is growing interest in the use of exercise to synchronize the circadian rhythm, with studies showing that scheduled exercise can entrain the human sleep–wake rhythm. Vigorous wheel running exercise prevents the loss of rhythmicity induced from maintaining rats under constant light, and accelerates the emergence of a circadian pattern in rats moved to dim light (Lax et al., 1998). In healthy elderly men, a 3-month exercise training program induced a significant reduction in the fragmentation of their rest-phase rhythm (Van Someren et al., 1997). Exercise effects on improving the circadian rhythm clock are likely mediated by its effect on melatonin secretion. Exercise can have phase-shifting effects on melatonin secretion and lead to both acute and chronic increase in melatonin levels (Buxton et al., 1997). Exercise has also been shown to be beneficial in treatment of circadian rhythm disorders such as sleep, jet lag, and shift work disorder (Eastman et al., 1995; Shiota et al., 1996). Intensity and time of exercise also influences melatonin secretion, with melatonin levels increasing more with high-intensity exercise as compared to moderate-intensity exercise and morning exercise leading to a greater increase in melatonin at night compared to afternoon exercise (Kim and Kim, 2014; Carlson et al., 2019). Studies investigating whether exercise increases melatonin in people with PD are lacking. However, based on the known disruption in melatonin secretion in PD and the beneficial effects of melatonin administration in preclinical models of disease and in clinical trials of people with MCI and sleep disorders, exercise may be an essential way to restore melatonin levels and circadian rhythm in PD.

It is now well recognized that there is an association between type II diabetes mellitus and increased risk of developing PD (Cullinane et al., 2022). (Figure 1) In the periphery, insulin regulates glucose metabolism, but within the central nervous system, it has neuromodulatory, neuroprotective, and neurotrophic effects. The majority of insulin comes from pancreatic B-cells, and is transported across the blood–brain barrier; however insulin and the closely related insulin-like growth factor 1 (IGF-1) are also produced by pyramidal neurons in the cortex, hippocampus, and olfactory bulb (Ghasemi et al., 2013). Insulin receptors are abundantly found in the basal ganglia and substantia nigra (Figlewicz et al., 2003).

Insulin regulates dopaminergic transmission, maintenance of synapses, synaptic plasticity, and neuronal survival and growth (Chen et al., 2022). A marked loss of insulin receptor mRNA in the substantia nigra pars compacta and increased levels of insulin receptor phosphorylation, which deactivates insulin signaling, has been reported in PD (Moroo et al., 1994; Takahashi et al., 1996; Sekar and Taghibiglou, 2018). Moroo et al. have shown that in PD brains post mortem, there is almost total loss of insulin receptor-positive neurons in the substantia nigra while some dopaminergic neurons remain, suggesting that downregulation in the insulin receptor system may precede the death of dopaminergic neurons (Moroo et al., 1994). Both insulin deficiency and insulin resistance lead to decreased brain insulin signaling in PD and contribute to neuroinflammation, mitochondrial dysfunction, and oxidative stress (Cullinane et al., 2022). Induction of insulin signaling with IGF-1 and reversal of insulin resistance suppresses α-synuclein aggregation and toxicity (Kao, 2009).

In individuals with PD, insulin resistance associates with a more severe phenotype, accelerated disease progression, and increased risk of cognitive decline (Cereda et al., 2012; Kotagal et al., 2013; Malek et al., 2016; Mohamed Ibrahim et al., 2018). Population-based cohort studies have also shown that use of certain antidiabetic drugs (glitazones, glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase inhibitors), associate with a lower risk of developing PD (Brakedal et al., 2017; Brauer et al., 2020; Cullinane et al., 2022). The glucagon-like peptide-1 receptor agonist–exenatide–is also being explored as a potential disease modifying agent in PD (Cardoso and Moreira, 2020; Mulvaney et al., 2020).

Exercise has long been known to improve insulin sensitivity peripherally (Borghouts and Keizer, 2000; Sampath Kumar et al., 2019). In the brain, animal studies also point to exercise having a favorable impact on insulin signaling pathways (Park et al., 2005; Muller et al., 2011). In rodent models of insulin resistance and impairment in insulin receptor function in the hippocampus, treadmill exercise can rescue this decline in insulin signaling (Park et al., 2019). In rats with memory impairments, exercise raises insulin signaling in conjunction with improved cognitive function (Jeong et al., 2018).

In middle-aged sedentary individuals, exercise for 8 weeks was shown to increase brain insulin sensitivity following intranasal insulin administration (Kullmann et al., 2022). Honkala et al. demonstrated that sprint interval training, but not moderate-intensity continuous training, decreased insulin-stimulated glucose uptake in cortical gray matter and most brain regions in insulin-resistant, middle-aged adults (Honkala et al., 2018). Exercise also influences IGF-1 gene expression and protein levels in several brain regions, especially those involved in learning and cognition (Duzel et al., 2016). Exercise has been shown to induce increase in circulating IGF-1 in the periphery which can be transported through the blood–brain barrier into the brain (Schwarz et al., 1996). Further studies need to explore the effects of exercise on insulin sensitivity and IGF-1 levels in PD. Exercise-enhanced insulin signaling may be crucial in PD to counter the detrimental downstream effects of insulin resistance, such as neuroinflammation, mitochondrial dysfunction, and oxidative stress and a more severe PD phenotype.

Klotho was discovered as a master aging regulator when deficiency of the klotho protein in mice was found to severely shorten lifespan and prompt signs of premature aging while overexpression of klotho increased lifespan by ~30% (Kurosu et al., 2005; Zeldich et al., 2014). Additionally, klotho-insufficient mice develop degeneration of mesencephalic dopaminergic neurons in substantia nigra and ventral tegmentum area (Kosakai et al., 2011). Human relevance for klotho in longevity is supported by studies demonstrating that klotho levels decline with aging (Yamazaki et al., 2010; Semba et al., 2011) and klotho genetic variants are associated with lifespan in multiple aging populations (Arking et al., 2002, 2005). Mechanistically, klotho has multifactorial actions–it suppresses insulin and Wnt signaling (Utsugi et al., 2000; Liu et al., 2007), regulates ion channel clustering and transport (Chang et al., 2005), modulates N-methyl-d-aspartate receptor (NMDAR) signaling (Dubal et al., 2014) and promotes fibroblast growth factor (FGF) function (Urakawa et al., 2006). Klotho linked with FGF23 is important for regulation of calcium, phosphate, and vitamin D homeostasis (Tsujikawa et al., 2003; Razzaque, 2009; Huang and Moe, 2011; Hum et al., 2017; Erben, 2018; Kuro, 2018). Klotho overexpression also inhibits insulin/IGF-1 signaling as a mechanism to extend life (Kurosu et al., 2005). While insulin resistance increases with aging and thus is generally unfavorable, klotho’s role in inhibiting the insulin/IGF-1 pathway is to decrease lipid overload since lipid-laden cells are vulnerable to lipid-induced programmed cell death (Unger, 2006).

Mounting evidence now links klotho with enhanced cognition. In mouse models of aging and Alzheimer’s disease, systemic elevation of klotho boosts cognitive function and enhances long-term potentiation (Dubal et al., 2014, 2015; Masso et al., 2018; Zeng et al., 2019). Recently, subcutaneous delivery of low but not high dose klotho was shown to enhance cognition in aged rhesus monkeys (Castner et al., 2023). In humans, carrying a genetic variant of KLOTHO or having higher levels of klotho has been associated with better cognition or decreased risk of dementia in aging and in Alzheimer’s disease (Dubal et al., 2014; Yokoyama et al., 2015; Belloy et al., 2020; Ali et al., 2022; Kundu et al., 2022).

In PD mouse models, peripherally delivered klotho has been shown to result in improvement in cognitive and motor behavior and enhanced synaptic plasticity (Leon et al., 2017). Furthermore, klotho overexpression or exogenous klotho administration protects dopaminergic neurons against oxidative injury and alleviates astrogliosis, apoptosis, and oxidative stress (Baluchnejadmojarad et al., 2017; Leon et al., 2017). Recently, lower levels of cerebrospinal fluid klotho were reported in individuals with PD compared to healthy controls and associated with greater motor burden of disease and greater Hoehn and Yahr stage of disability (Zimmermann et al., 2021). Since aging is a major risk factor for PD, further studies are needed to investigate if klotho associates with clinical symptoms of PD or disease progression.

Given the known beneficial effects of exercise on health and longevity, it is not surprising that exercise is associated with changes in aging regulator klotho. Higher levels of klotho associate with superior lower extremity strength and functioning in older adults (Crasto et al., 2012; Semba et al., 2016; Saghiv et al., 2017). Klotho levels also tend to be higher in exercise-trained individuals compared to their untrained counterparts (Saghiv et al., 2017). In healthy middle-aged adults, endurance, or resistance exercise boosts plasma klotho levels both acutely and chronically (Matsubara et al., 2014; Santos-Dias et al., 2017; Tan et al., 2018; Amaro-Gahete et al., 2019). When looking at the underlying mechanisms of exercise, Jin et al. demonstrated that irisin, an exercise-derived myokine, enhances mortality and cognition in mice after cerebral ischemia through upregulation of klotho (Jin et al., 2021). This positive effect of exercise and irisin is abolished in klotho-knockout mice (Jin et al., 2021). Exercise-induced upregulation of klotho in rodents has also been linked to decreased production of reactive oxygen species leading to decreased oxidative stress (Ji et al., 2018).

Additional studies are needed to examine the effects of exercise on klotho in people with PD. With increasing evidence for klotho’s impact on longevity, cognition, and neuroprotection, there is fervent interest in development of klotho-boosting therapeutics. Exercise may be one way to enhance klotho levels that may help with PD symptoms and progression.

Vitamin D is a powerful neurosteroid that can pass the blood–brain barrier to play a significant role in the central nervous system. While in the periphery vitamin D is critical for bone metabolism and calcium regulation, its effects in the brain are vast. It plays a role in anti-inflammatory and anti-oxidant actions, intracellular calcium signaling, neurotransmitter release, and transcriptional regulation of >200 genes, including neurotrophins (Pignolo et al., 2022).

Vitamin D receptor concentration is richest in the substantia nigra and the hypothalamus (Eyles et al., 2005). Additionally, 1α-hydroxylase – the enzyme that converts vitamin D to its active form,1,25(OH)₂D₃ – is highly expressed in the substantia nigra, suggesting that vitamin D may be related to the pathogenesis of PD via loss of protection for vulnerable dopaminergic neurons in this brain region (Eyles et al., 2005). In their meta-analysis, Zhou et al. report that both 25-hydroxyvitamin D [25(OH)D] insufficiency (<30 ng/ml) and deficiency (<20 ng/mL) are significantly associated with an increased risk of PD (Zhou et al., 2019). A high prevalence (55–69%) of 25(OH)D deficiency has been noted in individuals with PD (Evatt et al., 2008, 2011). Higher vitamin D concentrations have also been shown to associate with better cognitive function and mood in individuals with PD (Peterson et al., 2013).

Peterson et al. demonstrate that treatment with vitamin D (30 mg/kg subcutaneously) in a rat model of PD attenuates behavioral deficits, decreases levels of oxidative stress and neuroinflammatory markers and prevents dopaminergic cell loss (Bayo-Olugbami et al., 2022). Administration of intraperitoneal vitamin D (1 μg/ml/kg per day) in rats has also been shown to reduce dopaminergic cell loss and increase expression of glial-derived neurotrophic factor (Sanchez et al., 2009). In a randomized, placebo-controlled trial, vitamin D supplementation at a dose of 1,200 international units per day for 12 months in PD participants with a certain vitamin D receptor genotype (Fok/CC) associated with better motor function, Hoehn & Yahr stage, and quality of life (Suzuki et al., 2013). Interestingly, the Fok/CC vitamin D receptor genotype has also previously been shown to be associated with less advanced cases of PD (Suzuki et al., 2012) and to have greater effects of vitamin D-dependent transcriptional activation as compared to the Fok/TT genotype. In another recent randomized placebo-controlled trial, Bytowoska et al. investigated the effects of vitamin D ranging in dose from 2,500 to 4,800 international units per day (based on body mass index) in patients with PD who had previously undergone deep brain stimulation surgery (Bytowska et al., 2023). They reported significant improvements in tests of ambulatory function and a downward trend, though not significant, in inflammatory marker C-reactive protein, in the group that received vitamin D supplementation for 12 weeks. As demonstrated in a recent review, C-reactive protein is a robust marker of inflammation in PD (Mehta et al., 2023). Given the potential benefits of vitamin D, improving vitamin D status in individuals with PD may be a valuable approach.

Vitamin D levels increase acutely after exercise, and these changes are further dependent on intensity of exercise, age, and sex (Maimoun et al., 2009; Sun et al., 2017; Dzik et al., 2022). In their meta-analysis, Zhang and Cao report that in most studies with chronic exercise (multiple sessions per week for >12 weeks), there is an increase in 25(OH)D in vitamin D-deficient individuals but not in vitamin-D sufficient individuals (Zhang and Cao, 2022). This positive effect has been more commonly reported in studies of endurance exercise, possibly due to lipolytic processes leading to release of vitamin D stored in adipose tissues. Exercise can also increase vitamin D receptor mRNA levels (Aly et al., 2016; Puangthong et al., 2021). Da Costa et al. report that vitamin D receptor expression is decreased in hemiparkinsonian rats, but this effect is attenuated with treadmill exercise (da Costa et al., 2022). Furthermore, they demonstrated that administering vitamin D3 with exercise improved behavioral outcomes, augmented dopamine levels, and rescued dopaminergic neurons as compared to exercise alone.