Re: 세로토닌 --＞ 도파민 방출 자극.. 과잉되면 흥분성 증가, 자율신경 장애, 불안초조 등

[문형철]

파킨슨 환자 우울증, 양극성 장애, 불면 등의 증상

세로토닌 부족때문

ssri(세로토닌 재흡수 억제제) 사용 --> 세로토닌 과잉

 --> 신경, 근육 흥분성 증가(tremer, 과반사, 과긴장 hypertonic), 

 --> 자율신경 기능장애(발한, 빈맥, 기립성 저혈압 등)

 --> mental state 변화(불안, 초조 심하면 섬망)

serotonin projection to dopamin neuron --> 도파민 방출 자극

ORIGINAL RESEARCH article

Front. Syst. Neurosci., 20 July 2021

Volume 15 - 2021 | https://doi.org/10.3389/fnsys.2021.682990

Increasing Serotonin to Reduce Parkinsonian Tremor

Daniele Caligiore1*

Francesco Montedori1

Silvia Buscaglione2

Adriano Capirchio1

• 1Computational and Translational Neuroscience Laboratory, Institute of Cognitive Sciences and Technologies, National Research Council, Rome, Italy
• 2Neurophysiology and Neuroengineering of Human-Technology Interaction Research Unit (NeXT), Campus Bio-Medico University, Rome, Italy

While current dopamine-based drugs seem to be effective for most Parkinson's disease (PD) motor dysfunctions, they produce variable responsiveness for resting tremor. This lack of consistency could be explained by considering recent evidence suggesting that PD resting tremor can be divided into different partially overlapping phenotypes based on the dopamine response. These phenotypes may be associated with different pathophysiological mechanisms produced by a cortical-subcortical network involving even non-dopaminergic areas traditionally not directly related to PD.

In this study, we propose a bio-constrained computational model to study the neural mechanisms underlying a possible type of PD tremor: the one mainly involving the serotoninergic system. The simulations run with the model demonstrate that a physiological serotonin increase can partially recover dopamine levels at the early stages of the disease before the manifestation of overt tremor. This result suggests that monitoring serotonin concentration changes could be critical for early diagnosis. The simulations also show the effectiveness of a new pharmacological treatment for tremor that acts on serotonin to recover dopamine levels. This latter result has been validated by reproducing existing data collected with human patients.

현재의 도파민 기반 약물은

대부분의 파킨슨병(PD) 운동 기능 장애에 효과가 있는 것으로 보이지만,

안정 시 떨림에 대한 반응이 다양합니다.

이러한 일관성의 부족은

도파민 반응에 따라

PD 안정 시 진전이 부분적으로 겹치는 여러 가지 표현형으로 나눌 수 있다는

최근의 증거를 고려하면 설명할 수 있습니다.

이러한 표현형은

전통적으로 PD와 직접 관련이 없는 비도파민 영역까지 포함하는

피질-피질하 네트워크에 의해 생성되는

다양한 병리 생리학적 메커니즘과 관련이 있을 수 있습니다.

이 연구에서는 주로 세로토닌계와 관련된 본태성 진전의 신경 메커니즘을 연구하기 위해 생물학적 제약이 있는 계산 모델을 제안합니다.

이 모델로 실행한 시뮬레이션은

생리적 세로토닌 증가가

본태성 진전 증상이 나타나기 전 질병의 초기 단계에서

도파민 수치를 부분적으로 회복시킬 수 있음을 보여줍니다.

이 결과는

세로토닌 농도 변화를 모니터링하는 것이

조기 진단에 중요할 수 있음을 시사합니다.

이 시뮬레이션은

또한 세로토닌에 작용하여 도파민 수치를 회복하는

새로운 진전 치료제의 효과를 보여줍니다.

이 후자의 결과는

실제 환자를 대상으로 수집한 기존 데이터를 재현하여 검증되었습니다.

1. Introduction

Resting tremor is one of the most disabling features of Parkinson's disease (PD). People affected by resting tremor exhibit uncontrollable movements involving a body part, for example, the arm, when at rest. Tremor tends to decrease or stop when they deliberately move the affected part of the body (Deuschl et al., 2000; Kalia and Lang, 2015). Like other PD motor features, tremor is thought to result primarily from the death of dopamine-producing cells in the substantia nigra pars compacta (SNc), an area in the midbrain mainly targeting the striatum. This area is the main input gate of the basal ganglia, subcortical nuclei critical to managing motor behavior. Thus, a consistent reduction of striatal dopamine levels causes malfunctioning of the basal ganglia circuits that, in turn, may contribute to the emergence of tremor at rest (Pare et al., 1990; Wichmann and DeLong, 1999; Dovzhenok and Rubchinsky, 2012).

Based on this evidence, drug therapies for tremor often aim at recovering dopamine levels (Deuschl et al., 2000; Caligiore et al., 2019). However, while these approaches seem to produce amelioration for most PD motor dysfunctions, they generate variable responsiveness for resting tremor (Helmich et al., 2012; Wu and Hallett, 2013; Connolly and Lang, 2014). The lack of dopamine-based therapies consistency with tremor could be explained by considering that based on the dopamine response, PD resting tremor can be divided into various partially overlapping phenotypes that may be associated with different pathophysiological mechanisms (Zach et al., 2020). These mechanisms might be produced by a cortical-subcortical network involving areas traditionally not directly related to PD. Several data support this framework (Obeso et al., 2010; Caligiore et al., 2016, 2017b; Helmich, 2018).

The membrane properties of the basal ganglia cells support pacemaking (Surmeier et al., 2005) but do not produce tremor oscillations in healthy basal ganglia circuits. In contrast in PD, tremor-related activity (i.e., neural activity in the tremor frequency band, correlated with the tremor movement) was observed in the globus pallidus (Hutchison et al., 1997) and subthalamic nucleus (STN) (Levy et al., 2000), in the thalamus (Thal) (Lenz et al., 1994), and in the cortex (Timmermann et al., 2003). Lesions in different parts of the basal ganglia-thalamo-cortical network [in the cortex (Deuschl et al., 2000), globus pallidus, and Thal (Mitchell and Ostrem, 2020), in the STN (Alvarez et al., 2005)] suppress tremor. The fact that breaking the loop at multiple sites leads to the same effect, tremor suppression, suggests that the loop itself, more than any of its parts, contributes to tremor generation.

The tremor network can also involve the dorsal raphe nucleus (DRN), which could modulate the release and concentration of dopamine through the serotonergic projections it sends to the striatum (Bara-Jimenez et al., 2005; Di Matteo et al., 2008; Politis and Niccolini, 2015). It has been recently shown that aside from the dopaminergic dysfunction, in PD, there is a progressive loss of serotonergic terminals that has a slower pace and begins earlier than the dopaminergic one. This serotonergic dysfunction has been associated with the development of both non-motor and motor symptoms (De La Fuente-Fernndez et al., 2001; Lindgren et al., 2010; Politis and Niccolini, 2015) including tremor (Jankovic, 2018; Pasquini et al., 2018).

Overall, the vast range of conditions provoking or relieving tremor supports the involvement of a complex brain network, including both cortical and subcortical areas. Various dysfunctions within this network can produce different types of tremors. One mainly due to a direct dopaminergic system deficiency and for which dopamine-based actions work. Others involving the dopaminergic system and other brain areas, for which dopamine-based treatments are less effective (Pasquini et al., 2018; Zach et al., 2020).

Starting from this system-level perspective, we propose a bio-constrained computational model to study the neural mechanisms underlying a possible type of PD tremor: the one mainly involving the serotonergic system. The model reproduces the interactions between basal ganglia direct (DP) and indirect pathways (IPs), primary motor cortex (M1) and Thal (Marreiros et al., 2013; Hintzen et al., 2018), and the critical role of the DRN, the brain region producing serotonin, on the basal ganglia circuits (Padovan-Neto et al., 2020). More in detail, the model reproduces the excitatory and inhibitory influences between these brain areas through an ordinary differential equations system (Reed et al., 2013). Moreover, it drives the movement of a simulated 2D anthropomorphic robotic arm, reproducing the main features of the human arm (Katayama and Kawato, 1993; Caligiore et al., 2014).

The simulations run with the model demonstrate that serotonin could affect dopamine levels to maintain homeostasis between the DP and IP signals (Di Giovanni et al., 2002; Guiard et al., 2008; Reed et al., 2013; Sgambato-Faure and Tremblay, 2018). When there is a dopaminergic dysfunction, the model shows that the serotonin-based compensation mechanism is not able by itself to bring the system back to physiological dopamine levels. This result agrees with recent studies (Roussakis et al., 2016; Jiménez-Sánchez et al., 2019). However, the model suggests that the compensation mechanism can nevertheless mitigate the progression of the disease in the very early stages, before the manifestation of overt tremor. The model also shows that serotonergic dysfunctions affect striatal dopamine release, indirectly contributing to the emergence of a serotonin-related tremor phenotype. Overall, these results could be critical to devising new early diagnosis strategies based on serotonin monitoring (Wilson et al., 2019).

The model allowed us to study the effect of a possible new drug therapy that builds on the serotonin-compensatory role to recover dopamine levels in simulated patients. This drug produces a boosting of the serotonin physiological compensatory effect that contributes to reduce tremor. This latter result has been validated by reproducing existing data collected with human patients, showing that a serotonin increase could produce a tremor decrease (Qamhawi et al., 2015).

1. 소개

휴식 중 떨림은

파킨슨병(PD)의 가장 장애적인 증상 중 하나입니다.

본태성 진전 환자는

안정 시 팔과 같은 신체 부위에서 통제할 수 없는 움직임을 보입니다.

떨림은 영향을 받는 신체 부위를 의도적으로 움직일 때 감소하거나 멈추는 경향이 있습니다(Deuschl et al., 2000; Kalia and Lang, 2015).

다른 PD 운동 특징과 마찬가지로

진전은 주로 선조체를 주로 표적으로 하는

중뇌 영역인 흑질 흑질(SNc)의 도파민 생성 세포가 죽어서

발생하는 것으로 생각됩니다.

이 영역은

운동 행동을 관리하는 데 중요한

피질하핵인 기저핵의 주요 입력 게이트입니다.

따라서

선조체 도파민 수치의 지속적인 감소는

기저핵 회로의 오작동을 유발하여

안정 시 진전의 출현에 기여할 수 있습니다

(Pare et al., 1990; Wichmann and DeLong, 1999; Dovzhenok and Rubchinsky, 2012).

이러한 증거에 근거하여

떨림에 대한 약물 치료는

종종 도파민 수치를 회복하는 것을 목표로 합니다(Deuschl 외, 2000; Caligiore 외, 2019).

그러나

이러한 접근 방식은

대부분의 PD 운동 기능 장애를 개선하는 것처럼 보이지만,

안정 시 떨림에 대한 다양한 반응성을 생성합니다

(Helmich et al., 2012; Wu and Hallett, 2013; Connolly and Lang, 2014).

도파민 기반 치료법이

떨림과 일관성이 없는 것은

도파민 반응에 따라

본태성 진전이 서로 다른 병리 생리학적 메커니즘과 연관될 수 있는

부분적으로 겹치는 다양한 표현형으로 나눌 수 있다는 점을 고려하면

설명할 수 있습니다(Zach et al., 2020).

이러한 메커니즘은

전통적으로 PD와 직접적으로 관련이 없는 영역을 포함하는

피질-피질하 네트워크에 의해 생성될 수 있습니다.

여러 데이터가

이 프레임워크를 뒷받침합니다(Obeso et al., 2010; Caligiore et al., 2016, 2017b; Helmich, 2018).

기저핵 세포의 막 특성은

심장 박동을 지원하지만(Surmeier 등, 2005),

건강한 기저핵 회로에서는 떨림 진동을 일으키지 않습니다.

이와 대조적으로

본태성 진전에서는

떨림 관련 활동(즉, 떨림 주파수 대역의 신경 활동, 떨림 운동과 상관관계가 있는 신경 활동)이

구상체(Hutchison 등, 1997)와 시상하핵(STN)(Levy 등, 2000),

시상(Thal)(Lenz 등, 1994),

피질(Timmermann 등, 2003)에서 관찰되었습니다.

기저핵-탈라모-피질 네트워크의

다른 부분[피질(Deuschl et al., 2000),

구상체 및 탈(Thal)(Mitchell and Ostrem, 2020), STN(Alvarez et al., 2005)]의 병변은

떨림을 억제합니다.

여러 부위에서 루프를 끊어도

떨림 억제라는 동일한 효과가 나타난다는 사실은

루프 자체가 다른 어떤 부분보다 떨림 생성에 기여한다는 것을 시사합니다.

떨림 네트워크는

또한 등측 시상핵(DRN)을 포함할 수 있는데,

이는 선조체로 보내는 세로토닌 투사를 통해

도파민의 방출과 농도를 조절할 수 있습니다

(Bara-Jimenez 등, 2005; Di Matteo 등, 2008; Politis and Niccolini, 2015).

최근에는

도파민성 기능 장애 외에도

PD에서 도파민성보다 속도가 느리고

더 일찍 시작되는

세로토닌성 단자의 점진적인 손실이 있음이

밝혀졌습니다.

이러한

세로토닌 기능 장애는

떨림을 포함한 비운동 및 운동 증상(De La Fuente-Fernndez 외., 2001; Lindgren 외., 2010; 폴리티스와 니콜리니, 2015)의 발생과 관련이 있습니다(Jankovic, 2018; 파스퀴니 외., 2018).

전반적으로 떨림을 유발하거나 완화하는 광범위한 조건은

대뇌 피질과 피질하 영역을 포함한

복잡한 뇌 네트워크의 관여를 뒷받침합니다.

이 네트워크 내의 다양한 기능 장애는

다양한 유형의 떨림을 유발할 수 있습니다.

하나는

주로 직접적인 도파민 시스템 결핍으로 인한 것으로

도파민 기반 작용이 작용합니다.

도파민 기반 치료의 효과가 떨어지는

도파민 시스템 및 기타 뇌 영역과 관련된 다른 원인도 있습니다(Pasquini 등, 2018; Zach 등, 2020).

이러한 시스템 수준의 관점에서 출발하여,

우리는 주로

세로토닌계와 관련된 본태성 진전의 신경 메커니즘을 연구하기 위해

생물학적 제약이 있는 계산 모델을 제안합니다.

이 모델은

기저핵 직접 경로(DP)와 간접 경로(IP),

일차 운동 피질(M1)과 탈(Thal) 간의 상호작용을 재현하고(Marreiros 외, 2013; Hintzen 외, 2018),

기저핵 회로에서 세로토닌을 생성하는 뇌 영역인

DRN의 중요한 역할을 재현합니다(Padovan-Neto 외, 2020).

좀 더 자세히 설명하자면,

이 모델은 일반 미분 방정식 시스템을 통해

이러한 뇌 영역 간의 흥분 및 억제 영향을 재현합니다(Reed et al., 2013).

또한 시뮬레이션된 2D 의인화된 로봇 팔의 움직임을 구동하여

인간 팔의 주요 특징을 재현합니다(Katayama and Kawato, 1993; Caligiore et al., 2014).

이 모델로 실행한 시뮬레이션은

세로토닌이 도파민 수치에 영향을 주어

DP와 IP 신호 사이의 항상성을 유지할 수 있음을 보여줍니다

(Di Giovanni 외, 2002; Guiard 외, 2008; Reed 외, 2013; Sgambato-Faure와 Tremblay, 2018).

이 모델은

도파민 기능 장애가 있을 때

세로토닌 기반 보상 메커니즘만으로는

시스템을 생리적 도파민 수치로 되돌릴 수 없음을 보여줍니다.

이 결과는

최근의 연구 결과와 일치합니다(Roussakis 외., 2016; Jiménez-Sánchez 외., 2019).

그러나 이 모델은 그럼에도 불구하고

보상 메커니즘이 명백한 떨림이 나타나기 전인

초기 단계에서 질병의 진행을 완화할 수 있음을 시사합니다.

이 모델은 또한

세로토닌 기능 장애가

선조체 도파민 방출에 영향을 미쳐

세로토닌 관련 진전 표현형의 출현에 간접적으로 기여한다는 것을 보여줍니다.

전반적으로 이러한 결과는

세로토닌 모니터링에 기반한

새로운 조기 진단 전략을 고안하는 데 중요할 수 있습니다(Wilson et al., 2019).

이 모델을 통해

시뮬레이션 환자의 도파민 수치를 회복하기 위해

세로토닌 보상 역할을 기반으로 하는

가능한 신약 치료의 효과를 연구할 수 있었습니다.

이 약물은 세로토닌의 생리적 보상 효과를 강화하여 떨림을 줄이는 데 기여합니다. 이 후자의 결과는 인간 환자를 대상으로 수집한 기존 데이터를 재현하여 세로토닌이 증가하면 떨림이 감소할 수 있음을 보여줌으로써 검증되었습니다(Qamhawi et al., 2015).

2. Materials and Methods2.1. Model Architecture

The model architecture reproduces the dynamical interaction between six components (Figure 1): M1, Thal, DRN, SNc, DP, and IPs. DP reproduces the overall activity of the striatum and the internal portion of the globus pallidus, whereas IP reproduces the activity of the circuit involving the external globus pallidus (GPe) and the STN (Smith et al., 1998; Gurney et al., 2001; Haber, 2003).

Figure 1

Figure 1. Model architecture including primary motor cortex (M1), thalamus (Thal), dorsal raphe nucleus (DRN), substantia nigra pars compacta (SNc); basal ganglia direct pathway (DP); basal ganglia indirect pathway (IP). The connections linking different components can be excitatory (arrows) or inhibitory (lines ending with dot). The dashed lines indicate the effects of serotonergic projections from DRN to SNc, DP, and IP. Through these connections, serotonin modulates the dopamine release in the system.

The model components communicate through excitatory and inhibitory connections. M1 receives excitatory input from the Thal (Guillery and Sherman, 2002; Bosch-Bouju et al., 2013; Planetta et al., 2013; Stephenson-Jones et al., 2016) and sends excitatory output to DP (striatum) and IP (STN) (Bolam et al., 2000; Caligiore et al., 2019; Paraskevopoulou et al., 2019). M1 also produces an inhibitory effect on DRN (through other cortical areas not reproduced here for simplicity, e.g., prefrontal cortex) (De Simoni et al., 1987; Monti, 2011; Lopes et al., 2019; Sargin et al., 2019). DRN receives excitatory signals from SNc and sends serotonergic projections to SNc (inhibitory), DP, and IP (both excitatory). Through these connections, it modulates the dopamine (DA) release to DP and IP (Vertes, 1991; Di Matteo et al., 2008). Thal receives input from the basal ganglia DP (excitatory connections) and IP (inhibitory connections) (Alexander et al., 1986; Smith et al., 1998; Middleton and Strick, 2000; Gerfen and Surmeier, 2011; Caligiore et al., 2013). SNc produces a DA release that has an excitatory effect on DP and an inhibitory effect on IP (Cachope and Cheer, 2014).

2.2. Model Equations

The following differential equations system simulates the activity of the model components and their dynamical interactions:

⋅M1=a1Thal⋅Thal−τM1⋅M1    (1)𝑀1·=𝑎1𝑇ℎ𝑎𝑙·𝑇ℎ𝑎𝑙-𝜏𝑀1·𝑀1    (1)

⋅Thal=a2Ex+a2DP⋅DP−a2IP⋅IP−τThal⋅Thal    (2)𝑇ℎ𝑎𝑙·=𝑎2𝐸𝑥+𝑎2𝐷𝑃·𝐷𝑃-𝑎2𝐼𝑃·𝐼𝑃-𝜏𝑇ℎ𝑎𝑙·𝑇ℎ𝑎𝑙    (2)

⋅DRN=a3Ex−a3M1⋅M1+a3SNc⋅SNc−τDRN⋅DRN    (3)𝐷𝑅𝑁·=𝑎3𝐸𝑥-𝑎3𝑀1·𝑀1+𝑎3𝑆𝑁𝑐·𝑆𝑁𝑐-𝜏𝐷𝑅𝑁·𝐷𝑅𝑁    (3)

⋅5−HT=a4DRN⋅DRN−τ5−HT⋅5−HT    (4)5-𝐻𝑇·=𝑎4𝐷𝑅𝑁·𝐷𝑅𝑁-𝜏5-𝐻𝑇·5-𝐻𝑇    (4)

⋅SNc=a5Ex−a5DRN⋅DRN−τSNc⋅SNc    (5)𝑆𝑁𝑐·=𝑎5𝐸𝑥-𝑎5𝐷𝑅𝑁·𝐷𝑅𝑁-𝜏𝑆𝑁𝑐·𝑆𝑁𝑐    (5)

 ⋅DA=G⋅5−HT⋅SNc−τDA⋅DA    (6) 𝐷𝐴·=𝐺·5-𝐻𝑇·𝑆𝑁𝑐-𝜏𝐷𝐴·𝐷𝐴    (6)

⋅DP=a7Ex+a7DA⋅DA−τDP⋅DP    (7)𝐷𝑃·=𝑎7𝐸𝑥+𝑎7𝐷𝐴·𝐷𝐴-𝜏𝐷𝑃·𝐷𝑃    (7)

⋅IP=a8Ex−τIP⋅IP+αIP⋅sin(fIP⋅t)DA    (8)𝐼𝑃·=𝑎8𝐸𝑥-𝜏𝐼𝑃·𝐼𝑃+𝛼𝐼𝑃·sin(𝑓𝐼𝑃·𝑡)𝐷𝐴    (8)

The equations reproduce: the influence of Thal on M1 (Equation 1); the effects of the DP and IP on Thal activity (Equation 2); the modulation of DRN by M1 and SNc (Equation 3); the serotonin (5-HT) release from the DRN (Equation 4); the influence of DRN on SNc activity (Equation 5); the DA release from SNc in response to 5-HT concentration (Equation 6); the influence of DA on medium spiny neuron (msn) in the DP (Equation 7); and the influence of the GPe-STN loop and DA on IP (Equation 8).

We obtained this system by critically manipulating the equations proposed by Reed et al. (2013). In particular, we introduced two innovations critical to investigate the role of serotonin in PD tremor: (i) a more detailed IP equation (Equation 8) to simulate the oscillatory behavior affecting the STN-GPe loop (Nambu et al., 2002; Stanford, 2003; Surmeier et al., 2005); (ii) a bio-constrained 2D dynamic arm, allowing to record the effects of serotonin increase on overt tremor (Qamhawi et al., 2015). Although the reciprocal connections linking GPe and STN can produce and maintain low-frequency rhythms (≤5Hz) (Plenz and Kital, 1999; Terman et al., 2002), they do not produce tremor oscillations in healthy basal ganglia circuits. In PD, the DA loss induces an excessive synchronization of the GPe-STN oscillatory activity, contributing to generate and propagate abnormal tremor-related oscillations over the striatal-thalamo-cortical system (Nini et al., 1995; Guillery et al., 1998; Boraud et al., 2005; Gatev et al., 2006; Özkurt et al., 2011).

These equations allow studying the modulation between the various components of the model with simple linear functions, except for the oscillatory term (Equation 8), and for the term that represents the DA release in the striatum (Equation 6). Moreover, DA has an excitatory effect on the DP activity (Equation 7) and an inhibitory effect on IP (Equation 8). These two different types of dopaminergic modulations reflect what happens in the real basal ganglia (Cachope and Cheer, 2014; Fiore et al., 2015). Finally, each equation has a decay term, ensuring that the variable (firing rates or concentrations) goes to zero in the absence of input.

2.3. Dynamic Arm

The model controls a simulated 2D dynamic arm (Flash and Hogan, 1985) implemented with the realistic biomechanical parameters proposed in Katayama and Kawato (1993) (as shown in this study for more details on the arm equations) (Figure 2). We numerically integrated the dynamical arm equations using a Runge-Kutta fourth-order method with an integration step set to 0.01. The arm shoulder is anchored at [0.0, 0.0] coordinates of the workspace. The range of variation of the shoulder (β) and elbow (α) angles are [−60.0°, 150.0°] and [0.0°, 180.0°], respectively. The arm movements are driven by M1 signals obtained in three different model conditions: before DA dysfunction (pre-lesion), after DA dysfunction (post-lesion), and after serotonergic treatment (post-treatment). Minimum and maximum values of M1 activity are suitably scaled to get, respectively, minimum and maximum values of the elbow and shoulder angles.

Figure 2

Figure 2. Representation of the dynamic arm used to test the model motor behavior. The task space is a horizontal plane in which the arm is free to move according to the inputs received from M1.

We used a proportional derivative controller (PDcontroller) to capture some muscular visco-elastic properties of the human arm (Caligiore et al., 2014). This mathematical model computes the torques vector (T) necessary to move the arm joints, starting from the desired angles supplied by M1 (Ades), the current angular position (Acur), and the current angular velocity (⋅Acur𝐴𝑐𝑢𝑟·):

T=Kp(Ades−Acur)−Kd⋅Acur    (9)𝑇=𝐾𝑝(𝐴𝑑𝑒𝑠-𝐴𝑐𝑢𝑟)-𝐾𝑑𝐴𝑐𝑢𝑟·    (9)

where Kp is the vector containing the values of the PD proportional constant for the shoulder and elbow joints (shoulder = 20.0 Nm/rad; elbow = 10 Nm/rad), whereas Kd is the vector containing the values of the PDcontroller damping constant (shoulder = 1.5 Nm/rad; elbow = 1.0 Nm/rad). These values were chosen taking into account human arm visco-elastic properties (An et al., 1981).

2.4. Simulation Settings

The model was developed using the Python programming language. The code is available here https://github.com/ctnlab/serotonin_PD_tremor_model.

2.4.1. Tuning the Model Parameters

Table 1 summarizes the values of the model parameters and their biological meaning. The values of the model parameters were set to obtain a steady-state for the variables of the equation similar to those found in experimental observations (Feldman et al., 1997; Segovia et al., 1997; Jones et al., 1998; Knobelman and Lucki, 2001; Mahon et al., 2006; Ohara et al., 2007; Shimamoto et al., 2013) (see Table 2). In particular, the parameters settings were done through an automatic optimization procedure, namely a “genetic algorithm” (Fortin et al., 2012; Caligiore et al., 2017a). The algorithm went through optimization cycles called “generations”. The parameters search started from the values used in Reed et al. (2013). For each generation, a “population” of candidate solutions corresponding to different possible sets of model parameters was “evolved” to (1) minimize the mean absolute percentage error (APE) between the model steady-state values and those drawn from the experimental observations. A low value of such error indicated that the activity exhibited by the model was similar to the data collected from the real subjects; (2) minimize the oscillation in the healthy model (meanPhys); (3) maximize the oscillation in the damaged model (meanPark). Each “generation” explores 30 candidate parameter populations, and hence models (during 300 generations of the genetic algorithm), to get a meanAPE = 0.003 (fitness = 0.9969), a value that guaranteed a good matching between the simulated and the real data; meanPhys = 0.0053 (fitness = 0.9947) to guarantee a low value of physiological oscillation; and (meanPark = 0.0051) to predict oscillation amplitude observed in PD subjects (fitness = 0.9949). The parameters shown in Table 1 are the best ones obtained with this procedure.

Table 1

Table 1. Parameters of the equations.

Table 2

Table 2. Variables of the equations.

2.4.2. Simulation Trials

The model equations are numerically integrated with an integration step Δt = 0.1 (i.e., one simulation step corresponds to 0.1 s of real-time). The simulation runs for 300 s (3,000 simulation steps). At the beginning of the simulation, the shoulder and elbow angles of the arm are set to the starting position (α = 0°; β = 0°). Within the simulation time, we monitor the changes in the model components (i.e., activity of brain areas and neuromodulators concentrations) in five-time windows that we called “simulation trials”: the first one lasts 75 s and it is used to simulate the behavior of the health system; the second one (75, 125) s is used to study the effects of the DA lesion (occurring at 75 s); the other three remaining trials (the first two lasting 50 s each, the last 75 s) are used to investigate the effects of increasing doses of the simulated serotonin-based treatment.

2.4.3. Experiment Repetitions

We collected data from 20 different simulated subjects, obtained by running the model with distinct seeds of a pseudo-random number generator, which in turn causes different values of a noise signal acting on the model. In particular, we added to the initial values of the variables of the equations (Table 2), a small random number drawn from a normal distribution having mean zero and SD set to the initial value times, a number increasingly chosen in the range (0.01, 0.02) (0.01 for the subject 1, 0.02 for the subject 20, with an increasing step set to 0.0005). At the same time, we added to the values of the parameters of the equations (Table 1), a small random number drawn from a normal distribution having mean zero and SD set to the value of the parameter times, a number increasingly chosen in the range (0.01, 0.02) (increasing step still set to 0.0005). In this way, we exploited the capacity given by the simulation approach to producing many subjects with low cost to have higher statistical reliability of the results.

3. Results

This section presents the results obtained with the model and directed to (i) support the 5-HT role in maintaining homeostasis between DP and IP by affecting DA release (Reed et al., 2013; Sgambato-Faure and Tremblay, 2018); (ii) demonstrate that 5-HT dysfunctions can indirectly contribute to the emergence of tremor by producing a DA decrease; (iii) study the effect of a new pharmacological treatment for tremor that acts on 5-HT to recover DA levels. This latter result was validated by reproducing data collected with human patients showing that 5-HT enhancement could reduce tremor (Qamhawi et al., 2015).

3.1. Serotonin Compensates Dopamine Loss

To investigate how 5-HT affects DA release, we introduced an SNc malfunctioning. At the beginning of the simulation, the model has no dysfunctions. After 75 s, we introduced an SNc impairment by increasing the value of τSNc (+25%) (Equation 5). This action produces an SNc activity decrease approximating the effects of a neural loss. At the same time, we increase the value of αIP (+11%) (Equation 8), simulating an efficacy decrease of the DA receptors that regulate the oscillatory component of IP.

The SNc spike frequency decreases, leading to a reduction of DA concentration (Figures 3A,B). This loss produces a chain of effects: Thal activity decreases (Figure 3C) because DP activity goes down whereas IP activity goes up (Figure 3D); M1 activity decreases (Figure 3C) because it directly depends on Thal activity. Lower activation of M1 produces a lower inhibition of DRN that, as a consequence, increases its activity (Figure 3B). Finally, higher activity of DRN supports an increase of 5-HT release that, in turn, partially recovers the DA level (Figure 3A).

Figure 3

Figure 3. Relationship between activities of model components and changes in DA and 5-HT concentrations. The dotted vertical line represents the timing of the lesion in the dopaminergic circuits (τSNc + 25%, αIP + 11%). The four graphs represent: (A) the concentration levels of 5-HT (brown line) and DA (pink line); (B) the spike frequency of DRN (light red line) and SNc (light blue line); (C) the spike frequency of Thal (orange line) and M1 (purple line); (D) the spike frequency of IP (light green line) and DP (dark green line).

3.2. Serotonin Dysfunctions Produce Dopamine Loss

Starting from a healthy system, we produced a 5-HT impairment by gradually increasing the τDRN parameter, approximating the effects of a DRN neural loss (Equation 3). Figure 4 shows that as we increase the degree of the simulated 5-HT impairment, the DA concentration in the striatum drops down. As a result, the oscillatory component of IP increase (Equation 8), feeding the emergence of an overt tremor (Figure 5). This outcome supports the hypothesis that 5-HT dysfunctions might indirectly contribute to the emergence of tremor.

Figure 4

Figure 4. Dopamine concentration vs. 5-HT release impairments. DAMAGE1 = τDRN + 40%; DAMAGE2 = τDRN + 60%; DAMAGE3 = τDRN + 80%.

Figure 5

Figure 5. Oscillation induced by 5-HT release impairments. DAMAGE1 = τDRN + 40%; DAMAGE2 = τDRN + 60%; DAMAGE3 = τDRN + 80%.

The results shown in Figures 4, 5 were statistically validated through a repeated-measures ANOVA with the condition (HEALTH, DAMAGE 1, DAMAGE 2, and DAMAGE 3) as a within-subjects factor. The Bonferroni post hoc test was also conducted on significant interactions, and the Greenhouse-Geisser correction was used in the case of violations of sphericity. The ANOVA revealed that there is a statistically significant difference between DA concentrations in different conditions [F(1,016−19,299) = 61,683,499, Cohen's f = 0, p < 0.001, power = 1]. Moreover, the post hoc analysis showed that the health condition (i.e., with no DRN damage) is significantly different from the other three conditions, with a greater DA concentration (level of DA concentration: 2,723 vs. 2,435 vs. 2,313 vs. 2,204 nM, respectively). Furthermore, the DA concentration decreases with the damage entity, the greater the DRN damage, the lower the DA concentration. Similarly, the ANOVA showed that there is a statistically significant difference between the tremor rate in different conditions [F(1,011−19,204) = 6,967,198, Cohen's fr = 0, p < 0.001, power = 1]. The post hoc analysis confirmed that the health condition is significantly different from the other three conditions, the tremor, in this case, is reduced with respect to the DRN damage conditions. Furthermore, the greater the DRN damage, the greater the tremor.

3.3. A New Treatment for Tremor Based on Serotonin Increase

In this section, we show that it is possible to build on the neural mechanisms underlying the physiological influence of 5-HT on DA release (sections 3.1, 3.2), to test a new treatment for tremor based on a 5-HT increase. Figure 6 shows the effects of this treatment at the level of neural activity, whereas Figure 7 shows the effects on tremor.

Figure 6

Figure 6. Relationship between activities of model components and changes in DA and 5-HT concentrations after SNc damage and selective serotonin inhibitors (SSRIs) simulated treatment. The first dotted vertical line (black) represents the timing of the lesion in the dopaminergic circuits (τSNc + 70%, αIP + 333%). The other three dotted vertical lines (gray scale) indicate the beginning of different serotonin-based drug (SSRIs) treatment periods with different doses of the treatment (τ5−HT − 20%, τ5−HT − 30%, τ5−HT − 40%). Color code and graphs displacement as Figure 3.

Figure 7

Figure 7. Oscillation amplitude of the bio-mimetic arm before and after the DA impairment (τSNc + 70%, αIP + 333%) and following SSRIs treatment (τ5−HT − 20%, τ5−HT − 30%, τ5−HT − 40%).

Starting from a healthy model, after 75s we introduced an SNc impairment by increasing the value of τSNc (+70%) (Equation 5) and by increasing the value of αIP (+333%) (Equation 8). Since these impairments are stronger than those produced in section 3.1, they produce a stronger DA loss (compare Figures 6A,B vs. Figures 3A,B). In this case, the physiological compensatory mechanisms involving 5-HT in between 75 and 125s only allow to partially recover the DA loss, and it is not strong enough to avoid tremor (case “SNc DAMAGE” on Figure 7).

Then we boost the physiological influence of 5-HT on DA release by gradually reducing the value of τ5−HT (Equation 4) between 125 and 300 s. This parameter could be a proxy of the selective serotonin inhibitors (SSRIs) effects on serotonin transporters, which carry 5-HT in the DRN pre-synaptic neurons (Serretti et al., 2006; Dogan et al., 2008; Murphy and Lesch, 2008). Thus, a gradual decrease of τ5−HT produces an increase of 5-HT release simulating the effects of a treatment based on a gradual increase of SSRIs. More in detail, we simulated three subsequent increases of SSRIs which produce the following chain of effects: Thal activity gradually increases (Figure 6C) because DP activity gradually goes up whereas the IP activity goes down (Figure 6D); as a consequence, M1 activity increases (Figure 6C) because it directly depends on Thal activity; a greater activation of M1 produces greater inhibition of DRN that, as a consequence, decreases its activity (Figure 6B); a lower activity of DRN supports a lower inhibition of SNc (Equation 5) that, in turn, recovers the DA level reaching similar levels observed before the lesion (Figure 6A). The reduction of IP oscillatory activity together with the increase of M1 activity produce a gradual decrease of tremor (Figure 7).

Data were analyzed with repeated-measures ANOVA with the condition (HEALTH, SNc DAMAGE, TRMT 1, TRMT 2, and TRMT 3) as a within-subjects factor. The Bonferroni post hoc test was also conducted on significant interactions, and the Greenhouse-Geisser correction was used in the case of violations of sphericity. The ANOVA revealed that the main effect of SSRIs treatment [F(1,192−22,654) = 3,475,671, Cohen's f = 0, p < 0.001, power = 1] is significant. The post hoc test showed a significant increase of tremor oscillation after the damage (0.005 vs. 0.032 m), furthermore, the treatment improves the tremor, in particular, increasing the dose decreases the tremor (0.024 vs. 0.019 vs. 0.016 m).

4. Discussion

In this study, we proposed a system-level computational model to investigate the role of the serotonergic system in PD tremors. The results shown in Figure 3 suggest that at the very early stage of the disease, a physiological increase in 5-HT concentration, which indirectly affects DA release, could partially recover a low reduction of DA. Even though this compensatory mechanism is not able by itself to bring DA back to physiological levels, it contributes to maintaining DA at values that do not produce overt tremor in the patient (Figure 3D). Figure 4 further supports the critical role 5-HT has in the modulation of DA concentration by showing that 5-HT impairments produce DA loss and the emergence of tremor (Figure 5). These results agree with empirical data showing that high levels of 5-HT could promote DA release from SNc dopaminergic projections (Blandina et al., 1989; Bonhomme et al., 1995; De Deurwaerdère et al., 2002). Interestingly, the tremor entity (amplitude of oscillations) produced by the 5-HT dysfunction is lower than the tremor entity obtained by the impairment of the dopaminergic system (Figure 5 vs. Figure 7). The 5-HT dysfunction, indeed, only acts on SNc DA release. By contrast, the tremor entity shown in Figure 7 is due to both the SNc DA release dysfunction (τSNc) and the efficacy reduction of D2 DA receptors that regulate the oscillatory component of IP (αIP). For this reason, the 5-HT-based treatment only partially contributes to recover the IP health activity (Figure 6D). Hence, the model suggests that 5-HT impairments alone contribute to the emergence of weak overt tremor symptoms, like the one that one could show at the very beginning of the disease progression. To have large tremor oscillations, it is also critical to consider the IP D2 DA receptors responsiveness dysfunctions. This result agrees with data we obtained using a spiking neural network model to study the differences between tremor and patients with akinetic PD (Caligiore et al., 2019) and agrees with experiments suggesting the involvement of both 5-HT and DA dysfunctions in the progression of resting tremor (Pasquini et al., 2018).

The results summarized in Figures 6, 7 demonstrate the effectiveness of a possible treatment based on the gradual increase of SSRIs on 5-HT transporters carrying 5-HT on DRN (Serretti et al., 2006; Dogan et al., 2008; Murphy and Lesch, 2008). Figure 6 shows that the physiological increase of 5-HT concentration operating before the simulated treatment (75, 125) s produces a compensatory mechanism that increases the DRN activity to partially recover DA levels. By contrast, the treatment (125, 300) s leads to a DRN reduction, producing a higher DA level recovery and a tangible reduction of overt tremor (Figure 7).

Interestingly, the simulations show that the SSRIs treatment triggers different network dynamics than those involved in the 5-HT physiological compensatory mechanisms operating before the treatment. The physiological compensation could be biologically explained by a negative feedback mechanism mainly due to increased activation of the 5-HT1A autoreceptors present in the axonic terminals of 5-HT DRN neurons (Hajós et al., 1995). This mechanism produces an increase in DRN activity (Figures 3B, 6B up to 125 s). By contrast, the simulations run with the model to test the effect of the SSRIs treatment suggest that the negative feedback between the concentration of 5-HT and the DRN activity has to follow a different dynamical trajectory, producing a decrease in DRN activity to get a high DA level recovery. The reduction of the 5-HT degradation constant (τ5−HT) causes an increase in the 5-HT extracellular levels, which leads to an increased DA release by SNc. The effect is an excitatory modulation of DP and an inhibitory modulation of IP. This dynamic produces an increase in the firing rate of Thal and, in turn, of M1 with consequent reduction of the DRN activity.

Parkinson's disease tremor treatments based on dopamine medication (e.g., Levodopa administration) lead to a widespread and non-specific increase of DA concentration in the SNc (Connolly and Lang, 2014; Lee and Yankee, 2021). By contrast, the treatment with SSRIs proposed in this study would allow acting in specific ways on the area of interest. In this case, we increase the intrinsic concentration of 5-HT. By directing the drug delivery toward specific cellular markers (Huang et al., 2019), it is possible to inhibit only the SSRIs of our interest, allowing a more focused and not widespread increase in neurotransmitter concentration (Valentino and Commons, 2005; Wang et al., 2005; Rao and Nanda, 2009; Dankoski et al., 2016).

Despite these encouraging results, the current version of the model mainly represents a tool to test the effects of neurotransmission modulation rather than test new treatments based on 5-HT manipulation. Recent influential data demonstrate that 5-HT loss could precede PD motor symptoms and could be critical to study the PD pathophysiology and to develop novel therapeutic strategies (Wilson et al., 2019). However, evidence obtained from clinical research found several hurdles that, to date, have impaired the clinical development of serotonergic drugs (Huot et al., 2017; Muñoz et al., 2020). For example, the studies conducted in patients with PD and animal models mainly focussed only on 5-HT1A and 5-HT2A receptors, but the 5-HT system consists of many receptors (at least 14). It is not yet clear, what is the role of these other receptors. With 5-HT1A stimulation, it seems not easy to separate any benefit from interaction with L-DOPA anti-Parkinsonian action. In addition, the use of 5-HT1A receptor partial agonists also seems marred by a potentially detrimental effect on L-DOPA anti-Parkinsonian action, suggesting that such drugs may have a narrow therapeutic window (Bezard et al., 2013).

Hence, modulation of the 5-HT system is a promising way to address several manifestations of PD, but further investigation is required to clarify mechanisms of neurotransmitter interactions and to determine optimal compounds and doses for effective therapies producing the maximal benefit with minimal adverse events (Huot et al., 2017; Muñoz et al., 2020). New data from clinical research will support the design of an updated version of the model to make more consistent the simulations of the potential role of 5-HT for early diagnosis and therapy.

5. Conclusions

This article proposes a computational model to study how serotonin could affect dopamine release within the basal ganglia and how serotonin dysfunctions indirectly contribute to the emergence of tremor. The computer simulations run with the model suggest that serotonin changes monitoring could help in early PD diagnosis. They also demonstrate the effectiveness of possible new pharmacological treatments for tremor acting on serotonin to recover dopamine levels. However, evidence obtained from clinical research found several hurdles that, to date, have impaired the clinical development of serotonergic drugs (Huot et al., 2017; Muñoz et al., 2020). Therefore, more clinical evidence is needed to confirm‎ or disprove the results obtained with the model.

The work focuses on the role of serotonin in Parkinsonian tremor, for this reason, studies how serotonin affects dopamine release through the meso-striatal circuits, including the SNc. Other studies on depression, mild cognitive impairment, and Alzheimer's disease, instead, have investigated the interaction between serotonin and dopamine release through circuits involving the meso-cortico-limbic network, including the ventral tegmental area (Martorana and Koch, 2014; Silvetti et al., 2019; Wang et al., 2019; Caligiore et al., 2020). Future studies on the relationship between these two different ways the serotonin affects dopamine could be critical to understanding the system-level neural mechanisms underlying the comorbidities between PD, depression, mild cognitive impairment, and Alzheimer's disease (Marsh, 2013; Caligiore et al., 2016, 2017b; Cummings et al., 2019).

Data Availability Statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://github.com/ctnlab/serotonin_PD_tremor_model.

Author Contributions

DC: conceptualization, formal analysis, investigation, methodology, resource, validation, software, visualization, writing - original draft, writing - review and editing, funding acquisition, project administration and supervision. FM: data curation, formal analysis, investigation, writing - review and editing, visualization, validation and software. SB: formal analysis, investigation, methodology, resource and software. AC: data curation, formal analysis, investigation, methodology, resource, writing - review and editing, visualization, validation and software. All authors contributed to the article and approved the submitted version.

Funding

This research was supported by the ERASMUS + project ARIS (www.aris-project.eu), Grant Agreement 2019-1-BE01-KA202-050425, by the FETPROACT-EIC IM-TWIN project, Grant Number 952095, and by the Advanced School in Artificial Intelligence (www.as-ai.org).

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

We thank Flora Giocondo for helping in statistical analysis.

References

Alexander, G., DeLong, M. R., and Strick, P. L. (1986). Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Ann. Rev. Neurosci. 9, 357–381. doi: 10.1146/annurev.ne.09.030186.002041

PubMed Abstract | CrossRef Full Text | Google Scholar

Alvarez, L., Macias, R., Lopez, G., Alvarez, E., Pavon, N., Rodriguez-Oroz, M. C., et al. (2005). Bilateral subthalamotomy in parkinson's disease: initial and long-term response. Brain 128, 570–583. doi: 10.1093/brain/awh397

PubMed Abstract | CrossRef Full Text | Google Scholar

An, K. N., Hui, F. C., Morrey, B. F., Linscheid, R. L., and Chao, E. Y. (1981). Muscles across the elbow joint: a biomechanical analysis. J. Biomech. 14, 659–669. doi: 10.1016/0021-9290(81)90048-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Bara-Jimenez, W., Bibbiani, F., Morris, M. J., Dimitrova, T., Sherzai, A., Mouradian, M. M., et al. (2005). Effects of serotonin 5-HT1A agonist in advanced Parkinson's disease. Mov. Disord. 20, 932–936. doi: 10.1002/mds.20370

PubMed Abstract | CrossRef Full Text | Google Scholar

Bezard, E., Tronci, E., Pioli, E. Y., Li, Q., Porras, G., Björklund, A., et al. (2013). Study of the antidyskinetic effect of eltoprazine in animal models of levodopa-induced dyskinesia. Mov. Disord. 28, 1088–1096. doi: 10.1002/mds.25366

PubMed Abstract | CrossRef Full Text | Google Scholar

Blandina, P., Goldfarb, J., Craddock-Royal, B., and Green, J. P. (1989). Release of endogenous dopamine by stimulation of 5-hydroxytryptamine3 receptors in rat striatum. J. Pharmacol. Exper. Ther. 251(3):803–809.

PubMed Abstract | Google Scholar

Bolam, J. P., Hanley, J. J., Booth, P. A. C., and Bevan, M. D. (2000). Synaptic organisation of the basal ganglia. J. Anat. 196, 527–542. doi: 10.1046/j.1469-7580.2000.19640527.x

CrossRef Full Text | Google Scholar

Bonhomme, N., De Deurwaèrdere, P., Le Moal, M., and Spampinato, U. (1995). Evidence for 5-HT4 receptor subtype involvement in the enhancement of striatal dopamine release induced by serotonin: a microdialysis study in the halothane-anesthetized rat. Neuropharmacology 34, 269–279. doi: 10.1016/0028-3908(94)00145-I

PubMed Abstract | CrossRef Full Text | Google Scholar

Boraud, T., Brown, P., Goldberg, J. A., Graybiel, A. M., and Magill, P. J. (2005). “Oscillations in the basal ganglia: the good, the bad, and the unexpected”, in The Basal Ganglia VIII, eds J. Paul Bolam, C. A. Ingham, and P. J. Magill (Boston, MA: Springer), 1–24.

Google Scholar

Bosch-Bouju, C., Hyland, B. I., and Parr-Brownlie, L. C. (2013). Motor thalamus integration of cortical, cerebellar and basal ganglia information: implications for normal and parkinsonian conditions. Front. Comput. Neurosci. 7:163. doi: 10.3389/fncom.2013.00163

PubMed Abstract | CrossRef Full Text | Google Scholar

Cachope, R., and Cheer, J. F. (2014). Local control of striatal dopamine release. Front. Behav. Neurosci. 8:188. doi: 10.3389/fnbeh.2014.00188

PubMed Abstract | CrossRef Full Text | Google Scholar

Caligiore, D., Helmich, R. C., Hallett, M., Moustafa, A. A., Timmermann, L., Toni, I., et al. (2016). Parkinson's disease as a system-level disorder. NPJ Parkinsons Dis. 2:16025. doi: 10.1038/npjparkd.2016.25

CrossRef Full Text | Google Scholar

Caligiore, D., Mannella, F., Arbib, M. A., and Baldassarre, G. (2017a). Dysfunctions of the basal ganglia-cerebellar-thalamo-cortical system produce motor tics in Tourette syndrome. PLoS Comput. Biol. 13:e1005395. doi: 10.1371/journal.pcbi.1005395

PubMed Abstract | CrossRef Full Text | Google Scholar

Caligiore, D., Mannella, F., and Baldassarre, G. (2019). Different dopaminergic dysfunctions underlying parkinsonian akinesia and tremor. Front. Neurosci. 13:550. doi: 10.3389/fnins.2019.00550