Calcium involvement in Synaptic Plasticity
It has been shown that high intensity rTMS increases calcium (Ca2+) within neurons (Banerjee et al. 2017). Calcium plays an important messaging role in the initiation and regulation of LTP and LTD (Brini et al. 2014), therefore the previously discussed effects of the stimulation could be attributed to its effect on this ion. Calcium signals differ in size and purpose depending on their source and location in the neuron (Goldberg et al. 2003). Post-synaptic calcium enters through VGCCS, NMDA and AMPA receptors. Internally, calcium release from the endoplasmic reticulum (ER) also plays an important role in the initiation and maintenance of LTP and LTD (Brini et al. 2014). Its mechanism of action is through the activation of various proteins which can result in receptor activation and trafficking, transcriptional regulation, and intracellular structural changes (Brini et al. 2014). Two proteins of specific interest because of their relationship to LTP and LTD are calmodulin dependent kinase II (CAMKII) and calcineurin.
CAMKII and Calcineurin
One specific way calcium can regulate LTP and LTD is through activation of calmodulin dependent kinases and phosphatases. When calcium binds to the fast buffer, calmodulin (Faas et al. 2011) it can activate specific proteins. Two of these major synaptic proteins are CAMKII and calcineurin (also known as protein phosphatase 2B). Specifically, in the cortex and hippocampus, CAMKII is thought to contribute to LTP (Giese et al. 1998) and calcineurin to LTD (Zeng et al. 2001). CAMKII has an important role in the development of LTP. Knockdown and inhibition studies show LTP is perturbed and there is impaired learning and memory in animals with altered CaMKII function (Giese et al. 1998; Murakoshi et al. 2017; Rossetti et al. 2017). Though its primary form of activation is through binding of Ca2+/calmodulin complexes it can stay in an active phosphorylated state if two adjacent subunits are bound (Bossuyt & Bers 2013). CAMKII can become activated from calcium entering through NMDA receptors or being released from ER stores (Lu et al. 2014). Activated CAMKII has spatially specific roles, this allows for spine specific LTP changes. It binds to the NMDA subunit NR2B, which is necessary to induce LTP (Strack & Colbran 1998), contributes to the trafficking and conductance of AMPA receptors (Appleby et al. 2011; Kristensen et al. 2011; Opazo et al. 2010) and responds to VGCC calcium influxes by transporting calmodulin to the nucleus to activate cAMP response element binding (CREB) (Ma et al. 2014; Wheeler et al. 2008; Yan et al. 2016). It is also involved in maintaining LTP due to its interaction with NMDA (Barcomb et al. 2016; Rossetti et al. 2017; Sanhueza et al. 2011; Yamagata et al. 2009).
Calcineurin involvement at the synapse is related to LTD and the suppression of synaptic activity. When calcineurin is supressed LTD is inhibited (Mulkey et al. 1994; Zeng et al. 2001) and LTP and short term memory are enhanced (Ikegami et al. 1996; Malleret et al. 2011; Zeng et al. 2001). Conversely, when calcineurin expression is enhanced, LTP is diminished (Winder et al. 1998). Interestingly, it also plays a role in supressing inhibitory circuits when LTP is induced in excitatory circuits (Lu et al. 2000) When activated, calcineurin can directly affect the functions of intracellular proteins and can have indirect influences through modulation of downstream signalling cascades. For mechanisms directly related to LTD, activated calcineurin can modulate the phosphorylation state of AMPA receptors and cause endocytosis (Beattie et al. 2000; Tavalin et al. 2002) . It also affects the expression of potassium voltage gated channels which regulate the excitability of the post synaptic membrane (Lin et al. 2011). These mechanisms lead to the initiation of LTD or diminish LTP.
Effects of Low Intensity rTMS
Low intensity rTMS (LI-rTMS), which will be explored in this thesis, can also be used to produce unique effects making it an interesting alternative to HI-rTMS. As LI-rTMS does not generate the same depolarisation dependent mechanisms as HI-rTMS, but still induces plasticity (Rodger et al. 2012), it is important to understand these different mechanisms of action. Recent research investigating LI-rTMS with in vitro and in vivo models has begun to elucidate some of the mechanisms. Acute LI-rTMS stimulation released calcium from ER stores in neurons (Grehl et al. 2015) and increased BDNF in specific brain areas (Makowiecki et al. 2014). It is likely that this calcium rise could affect various neuronal processes (Figure 1.4), however these have yet to be explored experimentally. In addition, chronic LI-rTMS stimulation has been shown to improve abnormal connections within the brain without affecting normal brain function (Makowiecki et al. 2014; Poh et al. 2018; Rodger et al. 2012).
Other forms of low intensity stimulation that do not depolarise neurons could share similar mechanisms with LI-rTMS. Transcranial direct current stimulation (tDCS) and extremely low frequency electromagnetic fields (ELF-EMF) are both low intensity forms of stimulation but they have profound effects on the brain. Rather than activating neurons directly to fire, they may cause changes in the capacity for plasticity; metaplasticity.
Transcranial direct current stimulation (tDCS), which involves the application of weak electrical currents to the brain increases cortical excitability (Nitsche & Paulus 2001). Depending on the direction of the current it is hypothesised to cause neurons to hyper-or hypo-polarise affecting their excitability (Stagg & Nitsche 2011). TDCS has already been trialled as a treatment for various neuropsychiatric disorders and as a neurorehabilitation tool (Sánchez-Kuhn et al. 2017; Zhao et al. 2017a). However, one of the limitations of tDCS is the lack of focality, as the current must pass between two electrodes which are generally placed on either side of the head (Woods et al. 2016). Extremely low frequency electromagnetic fields are the signals produced by electrical appliances and their effect on the brain has also been studied. The intensity of these stimulations is comparable to LI-rTMS (<20mT) but the delivery method lacks focality and stimulations use very high frequency (>30Hz) (Guerriero & Ricevuti 2016). This type of stimulation has a clinical effect in altering pain thresholds (Choleris et al. 2002; Ghione et al. 2005). Interestingly, research has shown a similar ELF-EMF may have a similar mechanism of action to LI-rTMS as it increases calcium within both neuronal (Li et al. 2014b; Luo et al. 2014) and non-neuronal cells (Carson et al. 1990). In some cases a non-focal stimulation is appropriate for the type of treatment needed. But as targeted treatment with tDCS or ELF-EMF is not feasible, LI-rTMS can be considered an appealing alternative as it has been shown to have promising effects altering brain plasticity in a region dependent manner (Makowiecki et al. 2014).
Treatment Resistant Depression
People with depression who do not respond to antidepressant medication are termed as having treatment resistant depression (Hendrie & Pickles 2012). The percentage of nonresponse to antidepressants is sometimes estimated as high as 50% (Fava 2003). Specific forms of depression including atypical and melancholic, are also associated with higher treatment resistance (Fava 2003). In these patients, direct neuromodulation is a promising alternative therapy (Holtzheimer & Mayberg 2012). Electroconvulsive therapy (electrical shocks to the brain to induce seizures) has been effective in alleviating this condition but has the major side effect of amnesia (Khalid et al. 2008). As a result, other less severe methods of neuromodulation, including rTMS, have been explored (Holtzheimer & Mayberg 2012).
rTMS for the Treatment of Depression
In 2008 rTMS was approved by the FDA for use in patients with TRD (George, Taylor & Short 2013). A recent meta-analysis has shown as high as 75% of participants had positive responses, defined as a ≥50% reduction in scores on the Hamilton Depression Rating scale, and remission occurring in as high as 50% of patients in another study (Berlim et al. 2014), proving it to be an effective alternative treatment for depression. The theory of the antidepressant ability of rTMS is related to restabilising balance in the limbic system, by targeting specific frontal cortex areas, such as the dorsolateral prefrontal cortex, to affect downstream connected brain structures such as the amygdala (Baeken & De Raedt 2011).
There exists a lack of consistent stimulation parameters for the treatment of depression (Table 1.1). Often, either the left or right dorsolateral prefrontal cortex (DLPFC) is targeted, but that is where similarities between studies end. For treatment of the left DLPFC, high frequency stimulation is used, though this can range from 5-20Hz. A high intensity stimulation is also used but can range from 80-120% of the resting motor threshold. The total number of pulses per session, which has shown to be important for efficacy (Teng et al. 2017), ranges from 800-3000 and the total duration of treatment from two to seven weeks. A few studies have compared treatment parameters directly to define the most appropriate regime (Bakker et al. 2015; Brunelin 2012; Schulze et al. 2018) but there still lacks a clear consensus. This has resulted in rTMS treatment not being as effective as it could be if the treatment parameters were optimised. Chapter 2 of my thesis will focus on intensity to assess how altering this parameter can affect the outcome of the stimulation.
Memory Decline in Alzheimer’s Disease
Those with AD experience a progressive loss in cognitive functions over the course of the disorder. Generally, the first stage of Alzheimer’s disease is mild cognitive impairment (MCI). This manifests as small interferences such as linguistic, attentional, and visual deficits, as well as impairments in abstract reasoning (Carlesimo & Oscar-Berman 1992). Mild cognitive impairment is not always indicative of AD and can often be a precursor to other forms of dementia. However, interferences in particular forms of memory can provide indications that AD is developing (Chan et al. 2016). As the disease progresses memory functions that involve medial temporal lobe structures such as declarative and working memory are strongly affected while non-declarative memories such as procedural memory are relatively unchanged until the very late stages of the disease (Gold & Budson 2008; Karantzoulis & Galvin 2011).
Episodic, semantic and spatial memory loss are the most well-defined losses in AD. Episodic memory is the remembering of autobiographical details of one’s life and the temporal and spatial context of these moments (Tromp et al. 2015). The deterioration of this memory type is considered one of the first indicators of AD as deficits appear from very early on in the disease (Karantzoulis & Galvin 2011). Semantic memory is the memory of facts and AD patients experience a progressive loss in semantic memory function over the course of the disease (Hodges & Patterson 1995). Those with AD have difficulty in a variety of semantic knowledge tests, including naming objects, places and people (Adlam et al. 2006; Montembeault et al. 2017). Very early in AD, disorientation and getting lost are well documented (Vlček & Laczó 2014), demonstrating a loss in spatial memory. Spatial memory on its own is hard to define as it often involves both components of episodic and semantic memory (Moscovitch et al. 2005). Spatial memory deficits are specific to this particular dementia and can be used to predict the transition into AD from MCI (Bird et al. 2010; Chan et al. 2016; Pengas et al. 2012).
Table of contents :
AUTHORSHIP DECLARATION: CO-AUTHORED PUBLICATIONS
LIST OF FIGURES
LIST OF TABLES
CHAPTER 1: INTRODUCTION
1.1 Transcranial Magnetic Stimulation
1.2 Repetitive Transcranial Magnetic Stimulation
1.2.2 Coil Shape
1.3 Effects of High-Intensity rTMS
1.3.1 Synaptic Plasticity
1.3.2 Influence on Regulatory Molecules
1.3.4 Effects of Low Intensity rTMS
1.4 Therapeutic Uses of rTMS
2.1.1 Monoamine Changes
2.1.2 Neuroendocrine Imbalance
2.1.3 Decreased Neurogenesis
2.1.4 Other Neurobiological Causes
2.2 Current Treatments
2.3 Treatment Resistant Depression
2.4 rTMS for the Treatment of Depression
3. ALZHEIMER’S DISEASE
3.1 Memory Decline in Alzheimer’s Disease
3.2.2 Amyloid Plaques
3.2.3 Neurofibrillary Tangles
3.2.4 Calcium Dysregulation
3.2.6 Convergence of Theories: Synaptic Loss
3.3 Current Treatments
3.4 rTMS for the Treatment of Alzheimer’s Disease
4. ANIMAL MODELS
4.1 Animal Models of rTMS
4.1.1 Coil Size
4.1.3 rTMS in Animal Models of Neurological Disorder
4.2 Animal Models of Depression
4.2.1 Olfactory Bulbectomy
4.2.2 rTMS and Animal Depression Models
4.3 Animal Models of Alzheimer’s Disease
4.3.1 FAD mutations
4.3.2 PS1M146V KI mice
4.3.3 rTMS in Animal models of AD
5. THESIS SCOPE AND AIMS
CHAPTER 2: MEDIUM- AND HIGH-INTENSITY RTMS REDUCES PSYCHOMOTOR AGITATION WITH DISTINCT NEUROBIOLOGIC MECHANISMS
MATERIALS AND METHODS
SUPPLEMENTAL METHODS AND RESULTS
CHAPTER 3: FREQUENCY DEPENDENT EFFECTS OF LOW INTENSITY RTMS IN A MURINE MODEL OF ALZHEIMER’S DISEASE
MATERIALS AND METHODS
CHAPTER 4: DISCUSSION
ALTERING THE PARAMETERS OF RTMS CAN INFLUENCE THE EFFECT IN PRECLINICAL MODELS
Summary – rTMS parameters
ANIMAL MODELS ARE ESSENTIAL TO ASSESS SPECIFIC ASPECTS OF NEUROLOGICAL CONDITIONS
OB model of agitated treatment resistant depression
PS1M146V model of Alzheimer’s disease
Comparison of animal models
LIMITATIONS IN ANIMAL STUDIES OF RTMS
EVIDENCE OF EFFECTIVE TAILORED TREATMENTS
CONCLUSION AND FUTURE DIRECTION