Parkinson’s disease and osteoporosis: basic and clinical implications

Abstract

Introduction:

Parkinson’s disease (PD) is the second most frequent neurodegenerative disease. Lewy bodies, the hallmark of this disease due to an accumulation of α-synuclein, leads to loss of dopamine-regulated motor circuits, concomitantly progressive immobilization and a broad range of nonmotor features. PD patients have more hospitalizations, endure longer recovery time from comorbidities and exhibit higher mortality than healthy controls. Although often overlooked, secondary osteoporosis has been reported frequently and is associated with a worse prognosis.

Areas covered:

In this review we discuss the pathophysiology of PD from a systemic perspective. We searched on PubMed articles from the last 20 years in PD, both clinical features and bone health status. We discuss possible neuro/endocrine mechanisms by which PD impacts the skeleton, review available therapy for osteoporotic fractures and highlight evidence gaps in defining skeletal co-morbid events.

Expert opinion:

Future research is essential to understand the local and systemic effects of dopaminergic signaling on bone remodeling and to determine how pathological α-synuclein deposition in the central nervous system might impact the skeleton. It is hoped that a systematic approach to the pathogenesis of this disease and its treatment will allow the informed use of osteoporotic drugs to prevent fractures in PD patients.

1. Introduction

Worldwide, the elderly population continues to expand at an unparalleled rate. Approximately, 8.5% of the total population is now over 65 years and is projected to reach nearly 17% of the world’s population by 2050 [1]. This aging trend presents several public health and socio-economic challenges including an increased risk for neurodegenerative and bone diseases. Parkinson’s disease (PD) is the second most frequently reported neurodegenerative disease and it has an estimated prevalence of 1% in subjects over the age of 60. Moreover, the socioeconomic impact of PD is also growing [2]. As PD progresses, neurological degeneration leads to wide-spread lesions in the brain along with balance impairment and an increased risk of falls. The systemic nature of PD seriously impairs life quality; symptoms comprise movement impairment and non-motor related dysfunction, dementia, gastrointestinal disturbances, body composition alterations, falls and a markedly increased risk of fracture as well as low bone mineral density [3,4]. The vast majority of research into the pathophysiology of PD has focused on the mechanisms of PD development in the substanta nigra and its neurologic implications; however, little is known about how PD impacts comorbidities, particularly, its association with bone mass and osteoporosis, a major complication of long-standing PD. In this article, we discuss the fundamental relationship between PD and bone health, with an eye towards future preventive therapies.

2. Pathophysiology of Parkinson’s disease

PD is a systemic neurodegenerative disease. It encompasses a series of comorbidities that contribute to the complexity of the disease, impairing life quality of patients. Symptoms span well recognized locomotor dysfunction including resting tremors, rigidity and bradykinesia to lesser known- non-motor features such as autonomic dysfunction, cognitive/neurobehavioral abnormalities, sleep and sensory disorders. Additionally, PD patients often experience weight loss and a greater risk of fracture which increases as the disease progresses with or without treatment [5].

2.1. Motor alterations and disease progression

The underlying pathological processes in PD progresses slowly by degeneration of a particular population of neurons in the mesencephalon. An important but not exclusive pathophysiological feature in PD is the loss of the dopaminergic neurons in the substantia nigra pars compacta resulting in disorganization of the basal ganglia (BG) circuits [6]. The pathological hallmark of PD is the formation of intracellular aggregates of α-synuclein in the cytoplasm in these dopaminergic neurons, known as Lewy bodies or if located in neurites, Lewy neurites [6,7]. These neurons are affected specifically in their cytoskeleton organization, which impairs their function and ultimately leads to cell death and alteration of the nigrostriatal circuit [6]. During the early stages of PD, the disease affects the dorsal motor nucleus of the vagal nerve, olfactory bulbs and nucleus, then the locus coeruleus, and later, the substantia nigra and ultimately cortical areas. The damage to the substantia nigra has been shown to be consistently accompanied by substantial extranigral lesions that cause impairment of the limbic system, telencephalic cortical function and autonomic regulatory mechanisms [8]. Overall, these alterations result in 4 classical cardinal (motor) features of PD that a clinical diagnosis includes: 1) bradykinesia, or slowness in movement, 2) rest tremor, 3) rigidity and 4) postural instability [3]. PD patients also experience non-motor alterations that contribute to quality of life deterioration. These are often expressed earlier than locomotor alterations, even as early as 20 years prior diagnosis.

2.2. Sleep disorders and sensory abnormalities

Patients with PD often present diverse sleep disorders. Most common alterations include insomnia, REM sleep behavior disorder (RBD), daytime sleepiness and restless legs syndrome, but frequently, pain and cramps or desire to urinate often interrupt their sleep, as well [9,10,11]. Sleep disturbances have been attributed to thalamic atrophies in PD patients and dysregulation of circadian rhythm [12,13]. It is still uncertain what are the mechanistic (s), alterations in the brainstem circuits that lead to dysfunction of glutamatergic, cholinergic and GABAergic signals in the dorsal pons within the brainstem, and how these may explain the symptom complex [14]. Clinical studies have shown that loss of REM sleep atonia is positively associated with neurodegenerative symptoms such as loss of olfactory function at early stages of neurological progression, suggesting these indicators can be used as early markers of PD [15].

2.3. Autonomic alterations in PD

PD patients commonly show signs of constipation and gastrointestinal tract dysfunction during the progression of the disease. There are considerable data supporting the presence of Lewy bodies in the myenteric plexus of the intestine at early stages of the disease; this location is consistent with the fact that these regions are typically vagal terminals [16]. Analysis of immunohistochemical staining of colonic biopsies have revealed in 4 out of 5 PD patients, the presence of phosphorylated-α-synuclein in the submucosal neurites of the enteric nervous system; remarkably there were no positive staining in control subjects [17]. Although, more evidence is required to fully characterize the pathophysiology of the gastrointestinal alterations in PD, the current findings suggest that the presence of Lewy bodies in the myenteric plexus may be an early sign of peripheral involvement in PD [16]. Other autonomic alterations can lead to manifestations such as orthostatic hypotension, sweating, bladder and erectile dysfunction [18, 19]. It should be noted that part of these, such as gastrointestinal dysfunction and imbalance could contribute to the higher risk of falls and subsequent fractures.

2.4. Body composition and disease progression

Clinical studies show that PD patients often experience weight loss, sometimes preceding the PD diagnosis. Chen et al. showed that the average weight of PD patients was stable prior to the disease diagnosis, but then decreased with disease progression. Surprisingly, these patients increase their food intake, but this does not lead to a concomitant weight gain [4]. PD patients were found to be 4-fold more likely to lose 10 lbs than healthy individuals during the eight years after first diagnosis. This weight change correlated with severity of disease leading to greater nutritional risk than healthy controls [20]. Indeed, accumulating evidence suggests a negative correlation between BMI in PD patients and severity of the disease, aging, comorbidities and higher daily levodopa dosing [21, 22]. These findings imply that weight loss in PD correlates closely with health-related quality of life and neurologic progression, an indication of the systemic nature of this disease.

2.5. Cognitive and neurobehavioral alterations

PD patients often present signs of cognitive impairment and they are at a higher risk of dementia aside from the risk of several comorbidities such as depression, anxiety and hallucinations [23, 24, 25]. Obsessive compulsive behavior and impulse behavior are attributed to the dopamine dysregulation, but these are still not well understood [26,27].

3. Bone strength, quality (and loss) in Parkinson’s disease patients

PD and osteoporosis are two common chronic disorders, affecting a substantial portion of the elderly population. Osteoporosis is characterized by low bone mass and deterioration of bone micro-architecture, which leads to an increased risk of fracture [28, 29]. Fractures, most often due to falls, cause significant morbidity and mortality in a substantial portion of the elderly population. It has been established through observational and longitudinal studies that PD patients have a higher risk of osteoporosis and low bone mineral density when compared to controls [30]. Interestingly the levodopa dosage in patients has been negatively associated with BMD in PD, particularly relative to spine and femoral neck sites [22]. Zhao et al performed a meta-analysis of the risk of osteoporosis and bone mineral density in individuals with PD [31]. The authors found that PD patients were at higher risk for osteoporosis (OR=1.18, 95% CI= [1.09, 1.27]) than healthy controls and have lower BMD levels than healthy controls overall. Particularly, PD patients have a lower hip, lumbar spine and femoral neck BMD than healthy controls.

The prevalence of PD is greater in males than females by a ratio of nearly 2:1. Yet, analysis -by gender- reveal that female PD patients had lower BMD than controls, while there were no significant differences in male PD patients compared to controls. However, male PD patients are at higher risk for osteoporotic fractures than female patients. This could be due to several reasons such as differences in vitamin D levels, frequency of falls, and reduced physical activity. In regards to the former, Hagenau et al., analyzed a total of 394 studies including subjects from diverse ethnicities and found that women with PD tended to have higher mean serum 25(OH)D (56±1.6 nmol/l) than men (50±2.6 nmol/l, p= 0.05) [32]. Gao et al showed in a cross-sectional study of 54 patients with PD and 59 healthy age-matched controls that PD patients had significantly lower BMD than in healthy controls, particularly in the lumbar spine. However, this study revealed that BMD scores of the spine, femoral neck, and hip were lower in females than in males in the healthy group and that within the PD group, BMD in the hip was significantly lower in females compared to males, contrary to Zhao et al 2012. Again, this study revealed a strong negative correlation between BMD in the spine, neck, and hip and severity of PD. In addition, there were negative correlations between BMD (T-score and Z-score) and scores for PD severity such as Webster, Unified Parkinson’s Disease Rating Scale for activities of daily living and motor activities (UPDRS II and III) as well as Hoehn-Yahr (H&Y) stage, and a positive correlation between BMD and Schwab and England (S&E) Scale activities of daily living score. Thus, in general, individuals with the lowest bone mass also had the most severe disease. Indeed, Gao suggested that low BMD in the spine, femoral neck, and hip may reflect the severity of PD and could be used as a surrogate marker not only of fracture risk but of disease progression [22].

The mechanisms by which PD patients develop osteoporosis it is still an area that is understudied. However, it is well known that sympathetic innervation is fundamental for integrating skeletal homeostasis with body composition principally by suppressing bone formation and increasing bone resorption via enhanced RANKL production; thus, it is fair to speculate that PD dopaminergic alterations also could translate to comprised sympathetic innervation to the skeleton [33,34]. It has been shown that animals treated with 6-hydroxydopamine and β-adrenergic agonists impairs osteoblast activity in mouse calvaria and that the hematopoietic elements in the bone marrow respond to sympathetic innervation by increasing outflow of neutrophils as well as increasing sclerostin from osteocytes [35,36, 37]. A recent study done by Handa et al., showed that dopamine receptors (Drd) are expressed in osteoblasts (Drd3 and Drd4) and osteoclasts (Drd1 and Drd3) and influence bone homeostasis [38]. In vitro data showed that the dopamine receptor agonist – levodopa- inhibit osteoclast differentiation and bone formation by decreasing osteoblast mineralization capacity. Animal models of PD responded to levodopa treatment by reducing their bone formation rate and increased serum levels of homocysteine. Independently, an in vivo model of neurodegeneration of dopaminergic neurons, showed increased in bone resorption and suppressed bone formation [38]. In the context of the PD, neurodegeneration occurs while patients are under treatment with levodopa, thus, these results suggest both factors occur simultaneously and contribute independently to bone loss [38]. Thus, it is conceivable that PD autonomic disturbances or the treatment of the disease itself could affect bone remodeling through enhanced adrenergic signaling leading to further uncoupling of bone formation to resorption. However, there are no basic or translational studies to test these hypotheses.

Muscle mass and activity also influence overall bone health. Sarcopenia and immobilization can impair bone formation in PD patients. It has been shown that exercise promotes bone remodeling by mechanosensory interactions between muscle and bone [39]. It has been reported that loading exercises are marginally beneficial in respect to bone mineral density and microstructure. In addition, regular physical exercise may suppress sclerostin production, thereby enhancing bone formation. Studies have shown that greater lean mass was directly related to higher BMD at the femoral neck during aging and smaller muscle area was associated with reduced cortical thickness and fractures [40,41]. Men with lower relative appendicular skeletal muscle mass (RASM) (<7.26 kg/m²) had significantly lower BMD compared to the ones with higher RASM, and these subjects showed a higher risk of osteoporosis than individuals with normal RASM. A cross sectional study of 104 PD patients aged >65 years old showed 55.8% of subjects had sarcopenia compared to 8.2% who were sarcopenic in the control group. Sarcopenia and frailty were positively correlated with greater motor alterations, higher frequency of falls and disease progression [42]. Thus, it seems intuitive that strategies to enhance muscle mass could prevent bone loss in PD patients, although once again there is a gap in evidence to support that tenet [40].

It has long been established that hormones are key regulators of bone homeostasis and alterations in these endocrine factors can strongly influence bone mineral density. Growth hormone is secreted by the pituitary and acts through IGF-1 to stimulate osteoblast and osteoclast proliferation and activity [43]. ACTH, another classic endocrine hormone, stimulates cortisol secretion and can bind to melanocortin receptor family, the MC2R that is expressed in osteoblasts and promote differentiation and proliferation [44]. Cortisol has several negative effects on bone by blocking calcium absorption and directly suppressing bone formation [45].

There is a limited amount of research on endocrine alterations in PD, but those few showed that untreated PD patients present hypothalamic disturbances that could led to decreases in hypothalamic releasing factors causing reduced plasma concentration of growth hormone, adrenocorticotropic (ACTH) and cortisol compared to controls [46]. Indeed, LBs have been found in the hypothalamus as early as Braak stage 3 of PD [47,48]. Dopamine can affect the anterior pituitary gland. PD patients with decreased levels of dopamine can have a direct effect on anterior pituitary gland secretion [49]. These findings suggest that PD may impact bone health by alterations in the pituitary-hypothalamic axis.

Low vitamin D has been associated in cross sectional studies with osteoporosis in PD. One study showed that PD patients presented more frequently with hypovitaminosis D and secondary hyperparathyroidism [50, 51, 52] than healthy controls, either as a consequence of limited UV light exposure and/or malnutrition. On the other hand, immobilization due to severe PD could increase serum calcium from greater resorption and this would suppress PTH and subsequently renal 1,25 dihydroxyvitamin D production [53]. Loss of sense of smell and appetite, and autonomic disfunction in the gastrointestinal tract of PD patients have been hypothesized as causative for the development of constipation and malabsorption in these patients. The latter, in turn, could lead to lower vitamin D levels from malabsorption. Notwithstanding, there are no studies that directly link vitamin D deficiency as a result of limited sun exposure and malnutrition to low bone mineral density or fractures in PD.

4. Parkinson’s disease, falls and neuroendocrine mediators

PD is a prime example of a progressive neurological condition where falls are prevalent, presumably because many risk factors converge in this disorder; however, the extent and severity of this problem is not well understood. Several studies have shown a positive association between PD and risk of falls. A meta-analysis involving 69,387 subjects showed that PD patients had an increased risk of fracture of 2–3 fold (HR hazard ratio = 2.66, 95% CI confidence interval: 2.10–3.36) compared to controls. In particular, there is a 4-fold increased risk of hip fracture compared to healthy patients, independently of gender [54].

A cross-sectional study of PD patients and early patients with PD have shown that falls are an increasing problem as the disease progresses. Compared with age- and sex-matched controls, patients with PD are more frequent fallers, and this is related to symptoms associated with the disease severity. Coinciding with the progressive nature of the disease, it was noted there was a 10-fold higher frequency of fallers in a cross-sectional cohort of PD patients compared with the group of patients with newly diagnosed PD. In addition, motor complications and disease severity were significantly associated with frequency of falling [2]. On a similar note, Silva de Lima et al study showed that the incidence rate of any type of fall was found to be higher in PD patients than controls (2.1 vs. 0.7 falls/person; p< 0.0001); moreover, the ‘new fall’ incidence rate after enrollment was 1.8 times higher for self-reported PD patients than controls (95% confidence interval, 1.6–2.0). Others have reported the median range for falling for non-treated PD patients was as high as 4–6 times within the 20 weeks duration of the study and can reach to an average of 20.8 falls per year in recurrent fallers [55,56]. These findings highlight PD as a prime “falling disease” [57].

Falls increase fracture risk and are the most common reason for emergency hospital admissions in PD patients [58]. Frequent fallers can be defined as patients that have fallen more than 5 times over a 6 months period and some have suggested a possible association between fall frequency and alterations in cortical areas in the brain [59]. Some have suggested that neurodegeneration leads to a cortical volume reduction and thus, a reduction in functional connectivity due to dopamine deficiency resulting in cortical inhibition through the basal ganglial network [60]. Fallers exhibit significantly longer disease duration, lower gray matter volume in the right superior temporal gyrus (STG), the right supramarginal gyrus (SMG), and part of the inferior parietal lobule (IPL), compared with non-frequent fallers. Furthermore, there is a significant linear correlation between fall frequency and gray matter volume reduction in the right IPL and right STG [59]. In addition, among patients with a high incidence of fracture, PD subjects show the most serious complications and a greater risk of mortality (2–3 fold HR increase) than those without the disease [61]. Hip fractures are a major cause of morbidity and mortality among PD patients. Importantly, hip replacement surgery in these individuals is associated with longer hospital stays, poorer mobility, and greater mortality [62]. Thus, understanding the physiological and cellular mechanisms relating the pathophysiology of PD to low bone mass and fractures is critical to developing targeted therapies.

A possible mechanism connecting PD to falls relies on the locomotor alteration they develop throughout their disease. As mentioned above, loss of nigrostriatal dopamine circuits causes postural instability, thus, greater risk of falls. Clinical evidence shows that severe dopamine loss in the basal ganglia is associated with discoordination and often increases slips and stops of movement over dynamic surfaces, that, in attempt to correct them, result in falls [63,64]. Moreover, frequent falls could also be explained by the decrease of muscular activity leading to muscular atrophy or sarcopenia in patients. PD severity as noted previously is associated with sarcopenia (odds ratio 2.30; 95% confidence interval 1.15–4.58) [65]. Accumulating evidence point to a strong connection between bone and muscle, and it includes endocrine and paracrine signals between them. It has previously been shown that increases in muscle growth promotes bone formation, possibly through IGF-I secretion from the muscle; IGF-1 produces a hypertrophic anabolic signaling affecting bone mass acquisition as well as muscle mass [66,67]. Serum IGF-1 declines with age, and has been identified as a one marker of osteoporotic risk in young women [68,69]. However, the association between levels of IGF-1 in the context of PD is controversial: IGF-1 is low in some PD patients although this may be due to multiple factors including nutritional aspects, age and coincident co-morbidities [Fig. 1, 70]. Others have shown higher IGF-1 serum levels in PD patients compared to controls and these are inversely correlated with the Unified Parkinson Disease Rating Scale for motor function (UPDRS-III) score at early stages of the disease suggesting that IGF-1 acts as a neuroprotective peptide as a compensatory mechanism to protect against further degeneration of neurological functions [71,72]. Despite these data, it remains unclear whether IGF-I can serve as an early biomarker of PD, while lower serum IGF-1 at later stages of PD could be due to the loss of muscle mass, nutritional deficiencies, concomitant morbidities and neurological damage.

Figure 1. Loss of nigrostriatal circuits in Parkinson’s Disease (PD).

Aggregation of α-synuclein leads to loss of dopaminergic neurons in substantia nigra. Disruption of nigrostriatal circuits leads to motor and nonmotor alterations. PD patients often show signs of sarcopenia that may contribute to bone deterioration. Endocrine and nutritional factors converge to overall bone impairment.

5. Animal models of Parkinson’s disease and their implications for understanding skeletal morbidity

There are several genetic animal models of PD that recapitulate some of the most important aspects of the disease, such as motor symptoms, Lewy body formation, nonmotor symptoms and disease progression. However, there is no mouse model that fulfills all these characteristics; each has advantages and limitations although these can provide great insights into the cellular and molecular mechanisms underlying the disease. A model of PD in mice is the overexpression of human alpha-synuclein under the Thy1 promoter driving its expression in nervous system. Its manifestations include olfactory and digestive alterations similar to how patients present at the first stages of the disease [73]. Other commonly used genetic models of PD are mutant mice with 3 independent single mutations, A53T, A30P, and E46K, forced expression of α-synuclein by viral infections and neurodegeneration induced by neurotoxin treatments [74,75,76]. The primary focus in the field has been to improve understanding of the cellular and molecular mechanisms of the neurological damage and its progression, for therapy and prevention. However, the molecular mechanisms explaining how neurodegeneration leads to a variety of the systemic alterations remain understudied. Therefore, the pre-clinical literature focusing on the impact of PD in bone health is very limited. Models of the modified LRRK2 gene have shown an increased risk of genetic and idiopathic PD [77]. But to the date are no reports of bone health in these patients. Nevertheless, Berwick et al. found that LRRK2 knock out mice showed higher β-catenin levels in the brain, improved bone microarchitecture, an increase in tibial cortical bone and stronger bones, mediated by activation of canonical Wnt signaling in bone [78].

More recently, α-synuclein has been shown to be expressed in several tissues [38, 79]. Close to ~80% of erythroid cells within the bone marrow express α-synuclein [80]. It is also found in the blood of mice and humans, thus providing some evidence that α-synuclein could be playing an endocrine role via the circulatory system, marrow niche and ultimately affecting bone remodeling [81]. We previously reported that α-synuclein is a hub gene in bone homeostasis [79]. Specifically, we generated a co-expression network consisting of 53 gene modules using expression profiles from intact and ovariectomized (OVX) mice from a panel of mouse inbred strains. The expression of four modules was altered by OVX, including one whose expression was decreased by OVX across all strains. This one module was enriched for genes involved in the response to oxidative stress, a process known to be involved in OVX-induced bone loss. Additionally, these genes were co-expressed in human bone marrow. Alpha synuclein was one of the most highly connected “hub” genes in that module. We subsequently characterized mice deficient in Snca and observed a 40% reduction in OVX-induced bone loss. Furthermore, protection was associated with the altered expression of that specific network module. In summary, the results of this study suggest that α-synuclein regulates bone network homeostasis and ovariectomy-induced bone loss. Surprisingly, we found that α-synuclein was expressed in both osteoblasts and osteoclasts at different stages of maturation in vitro [Fig. 2, 79]. Moreover, gene deletion of α-synuclein partially protected against ovariectomy-induced bone loss, including greater bone volume fraction, more trabecular number and thickness, and less weight gain [79]. Pathological α-synuclein dynamics in bone tissue could explain part of the impaired skeletal status in PD as well as other comorbidities. Notwithstanding, these results shed light that disturbances in α-synuclein in the pathophysiology of PD might extend beyond central and peripheral neurological systems.

Figure 2. Bone remodeling and dopamine.

Osteoblasts, osteocytes and osteoclasts are the major regulators of bone homeostasis. Alteration of dopamine at central levels may be affecting bone cell signaling transduction through dopamine receptors due to decrease in dopamine levels. PD patients show higher circulating levels of α-synuclein, presence of Lewy bodies in the midbrain and in enteric terminal nerves. We postulate that α-synuclein aggregates could translate to alterations in cellular dynamic of α-synuclein in bone cells as possible a contributing mechanism to bone impairment in PD patients.

6. Osteoporosis treatments for Parkinson’s disease patients

Despite the well-established evidence that PD patients are at a high risk of fracture and twice the risk of mortality from fractures, only few receive treatment [61]. Fractures in general occur because of falls and low BMD. Therapies to target either or both factors are fundamental to improve life quality and decrease mortality. As discussed previously levels of vitamin D have been positively correlated with BMD but its positive effects have only been demonstrated in deficient states, thus, supplementation benefits remain controversial [52,81]. There are studies suggesting vitamin D therapy can improve muscular function, reaction time, balance and coordination in older people; this potentially could protect against falls [82,83]. The positive effect of Vitamin D was greater in osteoporotic patients with hypovitaminosis D and co-administered calcium [84]. However, more recent meta-analyses have shown that vitamin D had no effect on overall fractures, hip fractures or falls, at low or high dose suggesting that vitamin D supplementation lacks clinical benefit and, could possibly have a harmful effect at very high levels [85,86]. However, PD patients are exposed often to malnutrition, thus, considering vitamin D and calcium supplementation at low dose may be of use in selected patients.

Among possible pharmacological treatment alternatives for osteoporosis in PD patients, antiresorptive treatments such as bisphosphonates are the most common. Bisphosphonates are pyrophosphate analogues that reduce bone resorption by directly inhibiting osteoclast activities [87]. The oral agents include alendronate, ibandronate and risedronate; all show increased bone mineral density and a markedly decreased risk of fracture [88]. Intravenous zoledronic acid, administered once per year, is the most effective bisphosphonate to date decreasing bone resorption markers and increasing BMD by 4–5% in the spine and ~3% values in the femoral neck when compared to control [89]. Fractures are reduced by 70% in the spine and ~40% in the hip along with improvement of ~6% of BMD in total hip, lumbar spine and femoral neck [90]. Sustainable decreases in serum bone resorption markers at 6 and 12 months of infusion were demonstrated. Denosumab is another anti-resorptive agent that works through a monoclonal antibody that binds to RANKL, thereby inhibiting osteoclast maturation. It has been shown to be effective in reducing bone resorption markers and increased total hip and lumbar spine BMD [91]. However, denosumab must be administered every 6 months and there is a rebound increase in bone resorption, and potentially more vertebral fractures if the medication is stopped. Teriparatide, is an anabolic agent first approved by the FDA in 2001 as an osteoporosis treatment targeting bone formation [92]. Intermittent PTH administration (i.e. daily injections) has been shown to effectively improve BMD and fracture risk [93]. Another anabolic drug, PTHrp analog (Abaloparatide) has been shown to prevent fractures and increase BMD in postmenopausal osteoporosis and is non-inferior to teriparatide [94,95].

To the date there are no data whether these drugs can prevent osteoporotic fractures in PD. However, there is a new and innovative trial to address this issue. The TOPAZ (NCT03924414) NIH sponsored trial will also address barriers to treatment of patients with PD by providing rigorous evidence about whether zoledronic acid (ZA) reduces fracture risk in patients with PD, simplifying treatment by giving ZA at home without extra medical visits and BMD testing. Since adherence is low with oral bisphosphonates, and one infusion of ZA may prevent bone loss for at least 2 years, this trial has clinical implications [61,96,97]. It should demonstrate how a home-based fracture prevention strategy can reach older PD patients who would otherwise not receive treatment to reduce their high risk of fractures.

7. Conclusion

PD is one of the most frequently diagnosed neurological disease. It is both a systemic and progressive disease, leading to impaired quality of life and a number of comorbidities. PD patients fall more frequently due to balance and postural changes, as well as potential alterations in muscle strength. This coupled with lower bone mineral density than healthy age-and sex-matched controls that appears to be multi-factorial in etiology, places PD patients at a much greater risk of fractures. Preventing falls and future fractures with anti-resorptive agents, diet, and targeted therapies to improve muscle performance, including exercise, should be a major priority for treating men and women with PD.

8. Expert opinion

PD is certain to impacts skeletal homeostasis. But it is unclear precisely how the disease changes bone remodeling and whether this occurs in a cell autonomous or cell non-autonomous manner. We have shown that the progression of PD impacts tissues outside the brain, including the gastrointestinal tract, muscle tone and the sympathetic nervous system. This points to potential endocrine mediators that arise from diseased tissue in PD. In addition, because sympathetic neurons and their transmitters, particularly dopamine, can impact bone remodeling, it is conceivable that the deleterious skeletal effects are both endocrine and neurocrine in origin. To understand those actions, it is reasonable to turn back to mouse models for further insights. For example, does α- synuclein, which is expressed in both osteoblasts and osteoclasts, directly impact bone formation or resorption?. Targeted genetic deletions in mice are likely to provide direct answers to these questions, and such studies are currently underway. Similarly, more refined over-expression models that focus on body composition changes may provide insights into how α-synuclein impacts body composition. For example, there are preliminary data from our lab suggesting that α- synuclein is also expressed in adipose tissue and may have a detrimental effect on adipocyte metabolism. Also, it is worth noting that in mice with the global deletion of α-synuclein, there was less weight gain after ovariectomy than in the wild type controls [79].

Here we discussed mice models of PD as a tool to study the neurological aspects of PD and also encourage their usage to improve the understanding of early PD symptoms and unveil markers of early disease stages. It remains unknown the cellular mechanisms that regulates the interplay between Lewy bodies pathogenesis and skeletal state in PD patients. If we understood how Lewy bodies affects body composition, we might be able to intervene and prevent some of the co-comorbidities associated with aging. Frailty is associated with a greater risk of hip fractures, muscle weakness and falls. In addition, insulin resistance and Type 2 Diabetes Mellitus (T2DM) are more common in PD. Alternative approaches that treat these metabolic conditions may also help in improving motor function, as recently noted from a trial in England with exenatide, a GLP-1 agonist used to treat T2DM [98]. Other approaches that reduce damage to the mitochondria, or senolytic agents used to prevent senescence might become relevant.

We are aware of the weaknesses in PD and osteoporosis field. Osteoporosis is often underdiagnosed and untreated, and compliance to treatment can be challenging, especially if patient experiment intolerance to medications and gastrointestinal dysfunction. Osteoporosis therapy is still a developing field and PD drugs potential interactions with osteoporosis medicine are currently understudied. One of our clinical concerns is the putative correlation between the dose of L-dopa and the rate of bone loss. If true that dopamine agonists can suppress bone formation, then anti-osteoporosis treatments will be absolutely necessary to prevent fractures. More epidemiologic and longitudinal studies are needed to help understand that potential relationship, and we would maintain that animal models are required to more fully understand the pharmacologic implications of what has been considered a life sustaining therapy.

Lastly, a key weakness is the lack of our understanding of the normal function of α-synuclein, the most common etiologic factor in PD and Lewy Body Dementia. It is well understood that misfolding of α-synuclein can cause neurologic damage but its role in lipid metabolism still needs to be defined both in the central nervous system and adipocytes.

In sum, we call for more bench and translational studies to address the gaps in our knowledge of the non-neurological effects of PD. This will ultimately lead to the development of a rational approach for therapy that addresses whole body impacts and quality of life. Treatments that alter the course of PD are on the horizon and depend on research into the fundamental defects in the nervous system and the rest of the body. We know that there is cytotoxicity as a result of increased reactive oxygen species generated by damaged mitochondria. A goal for further research is developing the model system that test interventional approaches.

We expect for the next five years that new modalities related to several approaches; i.e. stem cell transplants, or viral vectors carrying DNA that can reprogram neuronal cells, or block aggregation and misfolding of α-synuclein to replace or supplement current therapeutics with L-dopa and its derivatives.

Article Highlights.

• Nonmotor symptoms in PD often occur earlier than motor dysfunction and can raise the possibility of clinically significant disease. PD patients are at high risk for fractures due to low bone mineral density (BMD), sarcopenia, and fall frequency.

• Exercise and nutritional interventions in PD patients have direct effects on bone health. However, dopaminergic-nigrostriatal circuits disruption lead to progressive alterations in neuro/endocrine signaling in PD patients that could affect directly or indirectly skeletal health by alterations in pituitary-hypothalamic axis, growth hormone, ACTH and IGF-1 signaling.

• Osteoporotic therapies such as zoledronic acid (ZA), PTH and PTHrp analogs should be considered as potential fracture preventive therapies in PD patients.

• More research is needed to better understand the implications of α-synuclein aggregations in non-neural tissue and their impact on the skeleton.