**2. Mechanism of Action of BoNT-A**

BoNT is a neurotoxin protein, which comprises a 50-kDa light chain and a 100 kDa heavy chain linked by a disulfide bond [17]. Seven serotypes of BoNT has been identified, and the most commonly used type in medicine is BoNT-A [17]. BoNT enters the presynaptic neuron membrane through binding of the heavy-chain C-terminal to the synaptic vesicle protein (SV2) [18]. After toxin endocytosis, the disulfide bond of BoNT is cleaved. The light-chain protein, which is the true active moiety, is then linked to the synaptosomal nerve associated protein 25 (SNAP-25) [18]. SNAP-25 is a protein with essential function for the binding of vesicles to the cell membrane and signal transduction. By binding the light-chain protein of BoNT-A to SNAP-25 and other SNAP families, BoNT-A inhibits neurotransmitters' exocytosis from the vesicles; hence, the affected neuromuscular junctions become paralyzed [18].

A clinical study confirmed SV2 and SNAP-25 immunoreactive fibers are distributed over the suburothelial and muscular layers instead of the urothelium in human bladder [19]. SV2 or SNAP-25 protein is not expressed within the urothelial or muscular cells [19]. The SV2 are expressed more abundantly in the cholinergic and parasympathetic fibers, as compared to the less than half expression to the sensory and sympathetic nerves. These findings suggest that the parasympathetic nerves are the main target of BoNT-A action in the human urinary bladder [19]. Other clinical studies associated with animal models demonstrated the SV2 expression in the human and rat bladder mucosae, as well as synaptosomal nerve-associated protein 23 (SNAP-23) and SNAP-25 in the urothelial cells and mucosa (differed in intensity) from the rat and human bladder [20]. SNAP-23 is a homologous target membrane SNAP receptor (t-SNARE) and is structurally and functionally similar to SNAP-25. SNAP-23 may be cleaved by BoNT-A, but human SNAP-23 is more resistant to botulinum [21,22]. The distribution pattern of SNAP-23 is different from that of SNAP-25: SNAP-23 is expressed mainly within the superficial or apical layer of urothelial layer, while SNAP-25 is detected throughout the urothelial layer [20]. SNAP-23 also interacts to multiple vesicle-associated membrane protein and syntaxin [23]. Since the urothelium is considered both a barrier as well as a significant signal transduction gate, the release of other neurotransmitters such as glutamate, adenosine triphosphate (ATP), neurotrophins, or NO may be affected after BoNT-A injection [24,25].

In clinical studies, BoNT-A inhibits the release of acetylcholine (Ach) and other neurotransmitters at the neuromuscular junction in human striated muscle [26]. Further neural modulating effects are observed by influencing the α and γ motor neurons after BoNT-A treatment (the α and γ motor neurons innervate the extrafusal and the intrafusal muscle fibers, respectively) [26]. Intravesical BoNT-A administration results in SNAP-25 cleavage, which inhibits the vesicular noradrenaline release. This action may prevent the α- and β3-adrenoceptors activation, and the reaction additionally affects the bladder neck contracture and detrusor relaxation [27]. A clinical study with receptor analysis conducted after BoNT-A injection treatment to the human bladders with neurogenic detrusor overactivity (NDO) showed significant reduction of the M2 and M3 muscarinic receptors as well as the P2X purinoceptor 2 (P2X2) and P2X purinoceptor 3 (P2X3) in the muscle layer [28]. This indicates that BoNT-A hinders DO through both sensory and motor features.

In addition, the ATP receptor P2X3 is critical for peripheral pain responses and afferent pathways controlling the urinary bladder volume reflexes [29–31]. In an animal model study, P2X3-null mice presented a marked hyporeflexia of the urinary bladder [29]. This result indicates that ATP plays an important role in mediating bladder fullness sensation and is crucial in the pathophysiology of OAB. In clinical studies, the human bladder, P2X3, and the transient receptor potential vanilloid subfamily-1 (TRPV1) are observed in the suburothelial layer [30]. A clinical study on the receptor profiles in biopsies from NDO or idiopathic DO patients showed a decreased P2X3 and TRPV1 immunoreactivity in the sensory nerve fiber after BoNT-A intravesical injection treatment [31]. The degree of decrease in the TRPV1 and P2X3 immunoreactivity is found to be correlated to clinical improvement (reduction of frequency and urgency status) [31]. Clinical studies with intravesical BoNT-A injections demonstrated significant inhibition of ATP and neurotrophin release and an increase of nitric oxide (NO) release from the human urothelial cells [24,30].

Animal models have shown possible mechanisms of action of BoNT-A injection treatment in interstitial cystitis/bladder pain syndrome (IC/BPS). In the isolated rat bladder model of acute injury and chronic inflammation, a significant amount of reduced calcitonin gene-related peptide and substance P from the afferent nerve terminals is observed [32,33]. The results suggest that BoNT-A injection treatment is a solution method of neurogenic inflammation in patients with IC/BPS [32,33]. TRPV1 inflammatory sensitization is found to play a vital role in inflammatory pain mediation [34]. Some proinflammatory agents (e.g., nerve growth factor, ATP, and IGF-I) sensitize rat nociceptors by promoting the recruitment of TRPV1 channels to the neuronal surface [34]. In preclinical studies, BoNT-A injection into oocytes expressing TRPV1 was found to block the TRPV1 membrane translocation by affecting protein kinase C (PKC) signaling [35]. The inhibition of the inflammatory sensitization of TRPV1 receptors by BoNT-A may also describe the therapeutic effects of BoNT-A injection to medication refractory IC/BPS. In an animal model study, BoNT-A has been shown to inhibit the ATP release from the urothelium in chronic bladder inflammation [36]. In clinical studies, reduction of the nerve growth factor and brain-derived neurotrophic levels in patients with IC/BPS after intravesical BoNT-A injection has demonstrated an analgesic effect [37,38]. BoNT-A conducted direct analgesic effects through exocytosis suppression of sensory neurotransmitters over the peripheral nociceptive neurons. However, indirect analgesic effects seemed also present with decreased spinal cord neuronal activity and with prevention of central sensitization as verified in some other clinical studies [39,40].

In a preclinical study for LUTS that related to prostate enlargement, BoNT-A injection has been reported to induce prostate atrophy and activate the apoptotic pathway in rats, which may result in reduced sympathetic stimulation of the prostate [41]. BoNT-A injection in a rat model revealed prostate weight reduction and reduced level of tyrosine hydroxylase-positive sympathetic nerve fibers and synaptophysin-positive cells in the epithelium [42,43]. In clinical studies, human prostate injection of BoNT-A has demonstrated apoptotic activity at the epithelial and stromal components of the prostate [44]. The reaction subsequently reduced the anatomical obstruction [44]. A clinical study comparing intraprostatic BoNT-A and normal saline injections demonstrated a significant contractile function reduction while maintaining relaxation response by presenting reduced prostatic urethral pressure response to intravenous norepinephrine and electrostimulation [45].

#### **3. BoNT-A Treatment in OAB and DO**

OAB is a clinical syndrome characterized by urinary urgency, usually accompanied by frequency and nocturia, with or without urinary incontinence, in the absence of urinary tract infection or other pathology [46]. Treatment of OAB is typically initiated with behavioral therapy, followed by oral medications including antimuscarinics or beta-agonists [47]. A large-scale study showed 46.2% of OAB patients discontinued medical treatment due to poor response or less effective as expected after treatment [48]. DO was defined as a urodynamic observation characterized by involuntary detrusor contraction during the bladder-filling phase [49]. DO is usually associated with symptoms of urgency, which is defined as a complaint of sudden, compelling desire to pass urine that is difficult to defer [50]. DO has been noticed in those with disturbances in the nerve, detrusor muscular, or urothelial levels [50]. Ach plays a key role in mediating bladder contraction through muscarinic receptors and detrusor muscle, and mediating ATP through purinergic receptors (P2X) stimulation has also been associated with bladder contraction [30].

In the central nervous system (CNS), the prefrontal cerebral cortex, the L-region of the pontine micturition center, and the lumbar spinal cord play the essential role of detrusor contraction inhibition [50]. A spinal cord lesion above the lumbosacral cord level may cause inhibitory pathway dysfunction, which further disturbs the voluntary control of micturition and results in NDO [50]. The sacral spinal reflex is known to be mediated by the unmyelinated C-fibers and is active in patients with SCI and NDO [50]. For idiopathic detrusor overactivity (IDO), although there is no specific central or peripheral neurologic dysfunction, increased expression of P2X2, TRPV1, and muscarinic receptors over the urothelium have been found in a clinical study [51].

Several clinical studies have demonstrated BoNT-A's efficacy in urgency and urinary incontinence reduction [10]. In the efferent nerves, BoNT-A injection temporarily inhibits the detrusor muscle contraction by blocking the Ach release via cleaving SNAP-25 from both the preganglionic and postganglionic nerve [24]. BoNT-A also blocks ATP release from the purinergic efferent nerves in the detrusor muscle [52]. In the afferent nerves, BoNT-A injection markedly reduced urothelial ATP release [53,54]. NO inhibits the afferent nerve conduction in the bladder detrusor and BoNT-A injection facilitates increased NO release from the urothelium [54]. In summary, BoNT-A injection has effects involving the efferent and afferent nerve pathways.
