Falls in older adult populations are a significant cause of morbidity and mortality (Cameron et al., 2012) with non-fatal injuries initiating a vicious cycle leading to a fear of falling, social isolation, loss of independence, deconditioning, and a significantly greater use of health care services (Stevens et al., 2006; Williams et al., 2006).

In Parkinson's Disease (PD), gait disturbances such as decreased stride length and gait variability are associated with increased risk of falls. Balance and gait deficits in PD are frequently refractory to therapy (Azevedo Coste et al., 2014; Perez-Lloret et al., 2014) and may be actually worsened by pharmacological and surgical interventions (Bloem et al., 1996), making falls a significant source of morbidity in PD (Schaafsma et al., 2003). Freezing of gait (FOG) is a syndrome normally seen in advanced PD and can occur when subjects are either on or off medication. FOG may be partly due to a failure to adequately scale amplitudes for the intended movement (Chee et al., 2009) and/or defective motor programming setting by the Supplemental Motor Area (SMA) and its maintenance by the basal ganglia, leading to a mismatch between intention and automation (Chee et al., 2009).

Cognitive and motor function must be carefully integrated to execute gait. Dysfunction of the basal ganglia in PD results in impaired motor control of skilled voluntary movements (Magrinelli et al., 2016) and movements become excessively slow and underscaled in size (Benecke et al., 1987). Biochemically, imbalance in multiple neurotransmitters (including but not limited to dopamine, acetylcholine, and GABA) is seen not only in basal ganglia and motor structures, but also limbic circuitries (Ondo and Hunter, 2003; Perez-Lloret and Barrantes, 2016). Balance disturbance and falls in PD may be more related to disruption in cholinergic rather than dopaminergic neurotransmission (Bohnen et al., 2009).

A key part of the subcortical cholinergic system is the pedunculopontine nucleus (PPN), which appears critically involved in gait disturbances in PD (Mazzone et al., 2005; Androulidakis et al., 2008; Acar et al., 2011), as PPN neuronal loss is evident in PD (Rinne et al., 2008) (Note that although we refer to the “PPN” throughout this manuscript, at the resolution of the imaging used here, it would perhaps be more accurate to refer to this region as the “mesencephalic locomotor region” as it likely includes the cuneiform nucleus—however, we use PPN as this terminology is consistent with much prior literature (e.g., Mazzone et al., 2005; Androulidakis et al., 2008; Acar et al., 2011).

Connectivity to/from the PPN appears critical for FOG in PD (Fling et al., 2014). Structural deficits in connectivity are evident between basal ganglia-PPN and other tracts in FOG (Fling et al., 2014; Vercruysse et al., 2015). DTI tractography obtained with 3T MR imaging in PD patients with FOG has demonstrated asymmetrically decreased connectivity between the PPN and the SMA, compared to PD subjects without FOG (Fling et al., 2014). FOG is also associated with diffuse white matter damage involving major cortico-cortical, corticofugal motor, and several striatofrontal tracts with DTI (Vercruysse et al., 2015). In addition to structural/anatomical connectivity, advanced neuroimaging techniques have enabled the studies of functional connectivity, which refers to the statistical temporal dependences between anatomically separated brain regions, to reveal the functional communication in the brain. Functional imaging studies (e.g., fMRI) have reported increased activity or altered connectivity during gait visualization in the midbrain locomotion centers between FOG episodes (Hanakawa et al., 1999), possibly reflecting compensatory mechanisms which might be overwhelmed with stress by turning or multitasking (Shine et al., 2013). Moreover, resting state functional magnetic resonance imaging (rs-fMRI) has allowed the inference of functional connectivity by measuring the level of spontaneous co-activation between fMRI time courses of brain regions recorded during rest. In vivo functional connectivity studies with rs-fMRI suggest that FOG patients may have significantly altered connectivity between PPN-SMA (Fling et al., 2014), which might reflect a maladaptive compensatory mechanism.

While the PPN has most often been investigated in PD in the context of FOG, it is unclear if altered PPN activity is present in non-FOG PD patients. Surgical targeting of the PPN is usually reserved for people with FOG resulting in significant impairment. Yet, even in early stages of the disease, there are a number of ways in which Parkinsonian gait is different from controls. While mildly affected PD patients can usually perform simple straight-line walk tasks without difficulty, they experience difficulties with turning, and when performing simultaneous motor or cognitive tasks (dual tasks), and/or crossing obstacles (Camicioli et al., 1998; Bond and Morris, 2000). They may have an abnormal gait pattern characterized by a shortened stride length, increased stride variability, and reduced walking speed (Morris et al., 2001; Buckley et al., 2008).

Ways to modulate PPN activity and connectivity have proven elusive. Acetylcholinesterase inhibitors may affect the PPN but such effects are likely to be modest. PPN Deep Brain Stimulation (DBS) has been shown to (inconsistently) improve gait difficulties in PD (Mazzone et al., 2005; Peppe et al., 2010; Acar et al., 2011; Hamani et al., 2011; Tykocki et al., 2011; Wilcox et al., 2011). However, the PPN tends to be spatially diffuse and is difficult to visualize on standard T1-weighted images, making electrode placement for DBS therapy difficult. Another potential way to modulate PPN is through the vestibular system, as PPN neurons tend to be highly vestibular-responsive (Aravamuthan and Angelaki, 2012). Galvanic vestibular stimulation (GVS) is a non-invasive technique that activates vestibular afferents to the thalamus and also the basal ganglia (Stiles and Smith, 2015) which in turn are directed to the PPN (Visser and Bloem, 2005), possibly explaining why GVS may positively impact posture/standing balance in PD (Kataoka et al., 2016). A few studies have demonstrated that noisy GVS improved postural and balance responses (Pal et al., 2009; Samoudi et al., 2015) as well as motor deficits in PD (Yamamoto et al., 2005; Pan et al., 2008; Lee et al., 2015a,b). These studies have speculated that noisy vestibular input may have improved information flow through the basal ganglia via stochastic facilitation (SF). SF is a phenomenon observed in a non-linear system where stochastic biological noise paradoxically increases sensitivity of a system to detect a weak stimulus possibly resulting in functional benefits (Mcdonnell and Ward, 2011). In addition to noisy stimuli, sinusoidal stimuli have been suggested as a means to activate steady-state, as opposed to transient balance responses that would be induced with pulsed stimuli (Latt et al., 2003). Sinusoidally oscillating stimuli may also activate irregular vestibular afferents (Gensberger et al., 2016) relying on voltage dependent K-channels (Eatock and Songer, 2011). While a couple of fMRI studies have shown sinusoidal GVS modulated activations in various brain regions (Della-Justina et al., 2014; Lee et al., 2016), GVS's influence on the PPN has not yet been investigated. A recent study in healthy older adults (n = 20) found that noisy GVS resulted in sustained reduction in Centre of Pressure (COP) parameters, such as velocity, and Root Mean Square (RMS) (Fujimoto et al., 2016). The mechanisms of this reduction was speculated to be on the basis of induced synaptic plasticity in the vestibular nuclei and the flocculus of the cerebellum, but effects on the PPN were not considered (Fujimoto et al., 2016).

Given the non-invasive, and potentially portable nature of GVS, we wished to determine if PPN connectivity could be modulated in mildly-affected PD subjects who may demonstrate reduced stride length for example, and thus may be at increased risk for falls, but did not exhibit FOG. Thus, in this study, we investigated whether or not functional connectivity between the PPN and other brain regions could be reliably assessed, whether or not these connections were significantly modulated by GVS, and if the connectivity was modulated in a stimulus-specific manner. Since a priori knowledge about functional connectivity to/from the PPN at the spatial and temporal resolution afforded by fMRI is unknown, this was essentially an exploratory approach. Careful care was taken to detect robust activation from PPN structures by analyzing the data in native space (without registration to a template) and utilizing subject-specific weightings of voxels within the PPN region. We demonstrated that PPN connectivity is sensitive to vestibular stimulation in PD in a stimulus-dependent manner.

Balance impairment remains a vexing problem in PD and is associated with considerable morbidity. Pharmacological (particularly dopaminergic) and surgical interventions have had varying degrees of success. Development of novel therapies has also been challenging because the pathophysiology and neuropathological substrates underlying gait disturbances are incompletely known.

To the best of our knowledge, we have shown for the first time that is possible with GVS to non-invasively modulate the functional connectivity in PD subjects between cortical/subcortical ROIs and the PPN—a structure critical for normal supraspinal control of locomotion. This demonstrated alteration in functional connectivity complements previous work examining anatomic connectivity patterns. Anatomically, the PPN has been shown to have connections with various areas such as the vestibular nuclei (Aravamuthan and Angelaki, 2012), deep cerebellar nuclei (Hazrati and Parent, 1992), premotor, supplemental motor (SMA) and primary motor cortices (Aravamuthan et al., 2007), frontal eye fields (Matsumura et al., 2000), thalamic nuclei and basal ganglia nuclei (Charara et al., 1996). Widespread projections involving the PPN include direct glutamatergic inputs from the motor cortex, and GABAergic inputs from substantia nigra pars reticulata (SNr), globus pallidus internus (GPi), subthalamic nucleus (STN), and deep nuclei of cerebellum. Ascending efferent projections target GPi, substantia nigra pars compacta (SNc), and thalamus. Descending efferent projections connect to pontine, medullary reticular formation, and the spinal cord vital for control of muscle tone and locomotion. Additionally, the PPN appears to be important in the initiation, acceleration, deceleration, and termination of locomotion through connections to the basal ganglia and higher cortical regions (Lee et al., 2000).

Our results indicate significant differences in the PPN functional connectivity between PD subjects and controls. This is particularly relevant when noting that none of our PD subjects had FOG, given their relatively mild disease. However, even mild disease is associated with altered gait and our results suggest that PPN functional connectivity patterns change even in the early stages of disease course.

We have shown that both GVS stimuli patterns (noisy and sinusoidal) augment overall deficient PPN connectivity in PD. Our results are consistent with previous studies demonstrating GVS activation of vestibular afferents to basal ganglia (Visser and Bloem, 2005; Stiles and Smith, 2015), which are also directed to the midbrain locomotion network (Peppe et al., 2010). PET studies in humans have also shown activation in the putamen in response to vestibular stimulation (Bottini et al., 1994).

We found differences in PPN network connectivity depending upon the type of stimulus used. Noisy GVS significantly decreased the functional connectivity between the left PPN and right-pallidum in PD. The PPN receives strong inhibitory, GABAergic inputs from the BG nuclei (GPi, STN, and SNr), which have disrupted connectivity in PD (Pahapill and Lozano, 2000). In particular, previous studies suggest that the PPN may be the principal target of pallidal outflow, since more than 80% GPi neurons were found to send axonal branches to both the PPN and thalamus in monkeys (Harnois and Filion, 1982). In PD, the inhibitory GABAergic synaptic activities from the GPi to the PPN is abnormally overactive, which may underlie the akinesia and the gait problems seen in the PD (Pahapill and Lozano, 2000). Taken together, these studies demonstrate the important role of the PPN and pallidal connectivity in motor and gait dysfunctions of PD. We demonstrated that noisy GVS significantly decreased connectivity between the left PPN and right-pallidum in PD (but not normal controls), which suggests that potential benefits of GVS on balance in PD may be partly mediated through attenuation of overactive pallidal inputs to the PPN.

Noisy GVS also decreased connectivity between the right PPN and the left cerebellar cortex. A DWI study found the connectivity of the PPN region with the cerebellum, thalamus, pallidum, and STN (Muthusamy et al., 2007). The cerebellum functions that help control of movement, coordination, and posture (Saab and Willis, 2003) are speculated to be associated with the existence of the pathway with the PPN (Muthusamy et al., 2007). The fact that prior studies have suggested a beneficial effect of noisy GVS (Fujimoto et al., 2016) may indicate that hyperactive PPN-cerebellar connections are partly “normalized” with GVS.

We also found that GVS increased the functional connectivity between the left inferior parietal cortex and the right PPN (with noisy stimuli) and left PPN (with sinusoidal stimuli). Previous studies in non-human primates examining cortical inputs to the pontine nuclei have supported the anatomical and functional relationships between the PPN and the inferior parietal cortex (May and Andersen, 1986; Martinez-Gonzalez et al., 2011). Like the PPN, the left inferior parietal cortex is involved in gait as well as visuospatial information processing, motor planning, and preparation (Caspers et al., 2012). Imagining normal gait activates the left inferior parietal lobule, in addition to the precuneus and bilateral dorsal premotor cortex, the left dorsolateral prefrontal cortex, and the right posterior cingulate cortex (Malouin et al., 2003). The left inferior parietal cortex appears to be especially related to FOG. Gray matter volume in the left inferior parietal region is significantly reduced in PD subjects with FOG patients compared to both PD subjects without FOG and healthy controls (Kostić et al., 2012). Thus, our result that noisy GVS increased the connectivity between the PPN and the left inferior parietal cortex might suggest GVS could improve gait difficulties in PD by augmenting the connectivity.

In the current study, one of the advantages is that we performed analyses keeping each subject's data in their original space without warping the data to a common template. We are frankly skeptical of fMRI studies suggesting robust activation from brainstem structures (e.g., PPN) when data are spatially transformed to a template, given the significant registration errors that can occur to small brainstem nuclei during whole-brain registration (Ng et al., 2009). In addition, we utilized PLS to find the combination of PPN voxels on a subject-by-subject basis that maximally corresponded with other ROIs. In effect, the first column of Q in Equation (2) represents a subject-specific spatial filter to “focus” the activity that maximally covaried with other ROIs.

It is interesting that many of the abnormal connectivities that we detected are lateralized, when balance might be considered a “midline” function. Balance control and gait are asymmetrical in patients with PD, and gait asymmetries have been linked to the pathophysiology of FOG (Boonstra et al., 2014). The symptoms of PD generally show an asymmetric onset and progression and it has been proposed that this may lead to a degree of “unbalanced” motor function, such that FOG is triggered by a breakdown in the bilateral co-ordination underlying the normal timing of gait (Plotnik et al., 2005).

There are a few limitations in our study. We examined a relatively small number of PD patients. However, by carefully selecting the PPN voxels via PLS on a subject-by-subject basis, we expect that we have significantly enhanced our effect size, and thus increasing our statistical power and thus increasing our statistical power. PLS is one of the most widely used blind source separation (BSS) approaches which have largely benefited the neuroscience studies (Ziegler et al., 2013; Chen et al., 2016, 2018; Smith and Nichols, 2018). In the future studies, we are interested to further explore the effective voxel selections using such data-driven approaches. We have shown GVS induced changes in PPN connectivity in people with mild to moderate PD but not in healthy controls. We speculate an “inverted-U” shape of effectiveness of GVS as a function of disease severity: in controls, GVS had minimal effect, in early/moderate PD, it had some effect (shown here), and in severe disease, degeneration in the PPN itself may prevent modulation of its connectivity. Future work is required to further investigate the relationship between disease severity and PPN connectivity and determine the behavioral significance of this altered PPN connectivity. We do note that we found a negative correlation between UPDRS scores and overall PPN connectivity, yet still found robust modulation of connectivity across all of our subjects.

The relation between behavioral gait measures (e.g., gait variability) and ultimate falls risk—the most important issue for people with PD—is an active area of research. Conceivably, previously-described GVS improvements in balance may not ultimately translate into reduced fall risk but such determination would require a prospective trial in the future. We have focused on the PPN because of its possible therapeutic implications, but gait disturbances in PD likely involve several cortical and subcortical structures. For example, PD patients have decreased activity of the SMA during gait (Hanakawa et al., 1999), and PD individuals have diminished pre-movement electroencephalographic potentials originating from the SMA prior to step initiation (Smith et al., 2012). Future studies to assess connectivity changes modified by GVS in other supra-spinal locomotion centers including SMA/pre-SMA may help to guide the development of optimal stimuli on a subject-specific basis.

This study was carried out in accordance with the recommendations of TCPS2, UBC Clinical Research Ethics Board with written informed consent from all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the UBC Clinical Research Ethics Board.

This study is ideated by MM. SL, LK, and DK conducted the experiment and data acquisition. SG and FB contributed to the data preprocessing. JC performed data analysis and manuscript writing. Interpretation of the results was performed by SL, JC, AL, FB, and MM. MM, ZW, and AL contributed to the project supervision and manuscript finalization. All authors have approved the final manuscript.