Currently, there is a justified and intensive research effort to identify the presence of Parkinson's disease (PD) pathology prior to locomotor manifestation. The burgeoning aging population punctuates the priority of this research effort. Ideally, patients in the age group most at risk for PD (those in their sixties) could be monitored for reliable biomarkers of PD as part of a yearly checkup. If reliably predictive biomarkers or indicators are detected, therapeutic efforts could be initiated that, at the very least, could extend the period of life the patient may enjoy being free of eventual locomotor impairment. Preclinical studies suggest intervention with repurposed drugs could prove effective in this effort (Swanson et al. 2011; Parkinson Study Group 2013; Chotibut et al. 2014). While the search for a readily identifiable and reliable biomarker for PD pathology continues, promising candidates are emerging (Marek and Jennings 2009; Sharma et al. 2013). Certainly, an optimal biomarker would be one that not only has high reliability and specificity, but can be facilely obtained, such as from an assay of plasma. At present, a combination of imaging-based, CSF, and plasma markers may prove useful, particularly for predicting cognitive decline in PD (Mollenhauer et al. 2014). Furthermore, evidence of behavioral dysfunction in conjunction with other clinical observations in the periphery such as cardiac function, obtained by relatively non-invasive means such as imaging, could prove more definitive for evidence of PD pathology underway. At the core of these inquiries lies our need to understand the pathological process of PD. Here, we discuss impaired regulation of the catecholamines dopamine (DA) and norepinephrine (NE) as potentially significant markers of PD pathology, manifesting as impaired cardiac and cognitive function prior to locomotor impairment.
In this highlight, Goldstein and colleagues report evidence of impaired NE sequestration in cardiac tissue of PD patients. In 70% of the patients, NE loss exceeded 95%. The regulation of NE and its metabolites in the left ventricular apical tissue from the PD patients reflected that of genetically modified mouse cardiac tissue, wherein the expression of the vesicular monoamine transporter (VMAT2) was reduced (VMAT-Lo). These mice also presented with > 95% NE loss and had similar NE to NE metabolite ratios. The author concludes that this profile represents a sequestration-deamination shift in the cardiac nerve tissue, owing to a reduced storage capacity (loss of VMAT2) and increased deamination of the catecholamine (in this case NE) to the toxic aldehyde intermediate during catabolism. Perhaps the most compelling argument that this peripheral pathology is of significance to catecholamine regulation in the CNS is that these metabolite ratios were similar in the putamen of PD patients. In this brain region, catabolism profiles of DA resembled that seen for NE in human cardiac tissue (Goldstein et al. 2013, 2014).
Thus, could these similarities in catecholamine metabolite profiles between cardiac and CNS tissue in PD patients provide an angle of insight to identify a common underlying etiology, or at least an approach for the detection, of PD pathogenesis? A non-motor-based detection mechanism may be valuable for early intervention, particularly since compensatory mechanisms in DA biosynthesis may mask PD pathology, typically identified only when locomotor impairment is evident. In the 6-hydroxydopamine rat PD model, there is evidence of increased DA biosynthesis in the somatodendritic region of the nigrostriatal pathway, as DA tissue content is increased against remaining tyrosine hydroxylase (TH) recovered in substantia nigra (SN) (Salvatore 2014). This increase in TH activity may compensate for TH protein loss, and is speculated to mask eventual locomotor impairment. This study, like others, also reported that increased DA turnover (DOPAC:DA ratio) did not occur in the somatodendritic region and only occurred in striatum when DA loss exceeded 80% (Salvatore 2014). This turnover data in context of the Goldstein laboratory findings pose the question of whether there is accumulation of the aldehyde intermediate of DA catabolism (3,4-dihydroxyphenylacetaldehyde), or in the case of NE, 3,4-dihydroxyphenylglycolaldehyde prior to major loss of DA (or NE, in the case of the cardiac tissue). Thus, the findings of Goldstein and colleagues should prompt investigators to evaluate whether 3,4-dihydroxyphenylacetaldehyde or 3,4-dihydroxyphenylglycolaldehyde levels, rather than metabolite end products 3,4-dihydroxyphenylacetic acid or 3,4-dihydroxyphenylglycol, respectively, are increasing in ratio as dopaminergic or noradrenergic lesions progress.
Aldehyde intermediates are argued to contribute to the degeneration of DA and NE neurons. However, if these intermediates are derived from the parent catecholamines, it would be expected that their aldehyde intermediates would decrease, given the loss of the parent compound during PD progression. A reduction in VMAT2 expression, however, could be argued to play a role in permitting accumulation of aldehyde intermediates. One possible scenario could be that if DA or NE biosynthesis increases because of enhanced TH activity as the lesion progresses, the concurrent loss of VMAT2 (along with TH protein) could allow for the cytosolic accumulation of DA or NE to an extent that overwhelms the capacity of aldehyde dehydrogenase to reduce the aldehydes to carboxylic acid end products during DA or NE catabolism. VMAT2 expression decreases in both striatum and SN in PD and correlates with clinical severity (Okamura et al. 2010). Furthermore, VMAT-Lo mice also display progressive deficits in non-motor symptoms commonly reported in PD patients prior to locomotor impairment (Taylor et al. 2009) and, notably, VMAT2-Lo mice show loss of TH in the locus coeruleus (LC) at 18 months of age, 6 months prior to TH loss in the SN (at 24 months of age) (Taylor et al. 2014).
The similarities in metabolic profiles of peripheral NE and central NE and DA nigrostriatal pathways and the impact of VMAT2 loss in both LC and SN certainly reinforce the evidence that PD is a multisystem disorder. The earlier loss of VMAT2 in the LC versus the SN also argues that detection of NE-related or driven processes in the periphery and CNS could signify PD pathology before locomotor impairment manifests. Therefore, given the role for NE in cardiac and cognitive function, evidence of impaired NE regulation in cardiac tissue with impaired cardiac function in conjunction with cognitive disorders, depression, and rapid eye movement sleep disorder may be useful to detect early PD (Asahina et al. 2014; Sakakibara et al. 2014). In fact, the parkinsonian pre-motor period may include symptoms such as olfactory dysfunction, gastrointestinal issues, and altered sleep patterns (Stern and Siderowf 2010). Evidence also shows that dysfunction of the autonomic nervous system may increase with the disease stage of PD (Kim et al. 2014). Cardiac dysautonomia and neurodegeneration are associated with symptoms of fatigue, which is a common complaint of PD patients (Nakamura et al. 2011) and orthostatic hypotension, which is known to increase falls and related injuries in PD (Goldstein et al. 2005; Chaudhuri et al. 2011). Cardiac denervation is typically found in PD patients with orthostatic hypotension, which affects approximately 50% of PD cases (Braune et al. 1998; Goldstein et al. 2000; Takatsu et al. 2000; Taki et al. 2000). Cardiovagal dysfunction and hyposmia are present in early PD (Mizutani et al. 2014). Therefore, the approaches to monitor autonomic and cardiac function could prove valuable in diagnosing PD and evaluating disease-modifying strategies that could induce global neuroprotection. Animal models resembling cardiac nerve loss are emerging that could be used to further our understanding of these pathologies and their mechanisms, refine evaluation tools, and test novel therapies (see Joers and Emborg 2014 for review). Early detection of cardiac denervation may require sensitive biomarkers, such as in vivo imaging with regional analysis (Joers et al. 2012). In that regard, in their study Goldstein et al. used vesicular monoamine transporter 2-deficient (VMAT2-LO) mice to test their hypothesis that a shift from vesicular sequestration to deamination of cytoplasmic catecholamines in the residual nerves contributes to myocardial norepinephrine depletion in PD. This finding provides insight into a possible contributing factor to a common pathogenic pathway that perpetuates neuronal cell loss in PD.
Issues of cognitive performance are also associated with PD, even in the early stages. There is a link between the presence of anxiety and depression and the subsequent development of PD (Shiba et al. 2000; Leentjens et al. 2003). Notably the comorbidity of depression with PD is approximately 40%. Estimates of cognitive impairment are as high as 55% in newly diagnosed patients (Janvin et al. 2003). Early deficits include executive dysfunction, which is linked to the progression of decline to dementia (see Kehagia et al. 2010 for review) and ultimately mortality (Levy et al. 2002). Disbrow et al. (2013) used functional magnetic resonance imaging to show decreased cortical functional selectivity in networks subserving executive function. They compared movement initiation and inhibition in a group of medicated early-stage PD patients and controls (matched for age, mental status, and years of education). By comparing task trials matched for motor output they were able to isolate functional magnetic resonance imaging activation because of cognitive, as opposed to motor processes. Behavioral (reaction time and error rate) data were similar between the two groups. However, the PD group showed a failure to modulate activity in prefrontal and other cortical regions. Specifically, while only right Brodmann's areas 9 and 10 were active during movement initiation (regardless of the hand used) in the control group, activation was bilateral in the PD group. These findings in the cortex are in agreement with human and non-human primate electrophysiological recording data from the basal ganglia that also show decreased functional selectivity (Filion et al. 1988; Taha et al. 1996).
It is well known that significant nigrostriatal neuron loss occurs prior to the appearance of motor symptoms in PD, but loss of noradrenergic neurons in the LC also occurs in PD and this loss may precede nigrostriatal neuron loss (Rommelfanger and Weinshenker 2007; Del Tredici and Braak 2013). As the LC has widespread connections to the neocortex, hippocampus, and thalamus (Aston-Jones and Cohen 2005), the aforementioned cognitive dysfunction may also signify the potential utility of a biomarker sensitive to NE metabolic measures in detecting possible PD pathogenesis. The LC–frontal cortex norepinephrine system subserves arousal, attention, and the stress response, all functions which are abnormal in PD (Stern and Siderowf 2010). It has been proposed that the transitions between different tonic levels of LC neuronal activity play a role in the transition between behavioral states such as sleep, focused alert and exploratory attention (Aston-Jones and Cohen 2005). Thus, diminishing LC–frontal cortex norepinephrine system function may interfere with executive functions, such as the ability to redirect stimulus driven attention to rapidly change motor plans in the face of evolving environmental conditions (see Benarroch 2009 for review). Finally, it should be pointed out that there are dense dopaminergic inputs from the ventral tegmentum to frontal cortex with reciprocal connections with the LC, and stimulus response properties are similar for dopamine and norepinephrine (for review see Sara 2009). Therefore, impaired noradrenergic components in the neocortex could contribute to cognitive impairments in PD, in addition to impaired dopaminergic function. In summary, given that the LC may be affected by neurodegenerative processes earlier than the SN, and NE subserves the aforementioned cognitive functions, these behavioral signs, in addition to the cardiac and autonomic functions previously mentioned, signify the potential utility of a biomarker that can confirm evidence of impaired NE regulation in the detection of early-stage PD (Fig. 1).
One final thought to consider is the role of impaired catecholamine metabolism and storage as a contributing force in PD pathology (and detection prior to locomotor impairment) in relation to Braak staging. It has been argued that Lewy Body pathology in the LC may contribute to PD-related dementia (Del Tredici and Braak 2013). If the LC is subjected to degeneration prior to the SN, then is alpha synuclein accumulation always the primary cause? Goldstein and colleagues analyzed end-stage PD patients and reported that 30% of them did not have severe NE depletion in cardiac tissue (Goldstein et al. 2014). As the authors point out, this finding is not 100% congruent with expected cardiac denervation in end-stage PD. Instead, it suggests that there is pathological heterogeneity in patients with confirmed diagnoses of PD. Indeed, as noted earlier, cardiac denervation associated with orthostatic hypotension affects approximately 50% of PD cases, and only about 50% of early stage PD patients show executive dysfunction (Disbrow et al. 2012). These observations could on one hand suggest differences in susceptibility to alpha synuclein pathology. On the other hand, loss of DA neurons could be independent of alpha-synuclein-driven aggregation, a possibility recently supported in mice expressing wild-type synuclein at disease-relevant levels (Janezic et al. 2013). While the role of alpha synuclein pathology in PD progression is still being evaluated, understanding its association with, or contribution to, impairments to catecholamine metabolism that manifest during PD progression could yield valuable insights for understanding the pathological sequelae of PD.