BDNF promotes activation of astrocytes and microglia contributing to neuroinflammation and mechanical allodynia in cyclophosphamide-induced cystitis

Interstitial cystitis/bladder pain syndrome (IC/BPS) is an enigmatic chronic inflammatory disease of the bladder. It is characterized by bladder pain accompanied with frequency, urgency, nocturia, and sterile urine, excluding typical urinary tract infections [1‐3]. Chronic pain symptom affects the patients’ quality of life severely. A community-based study in the USA reported a high prevalence rate of BPS, 2.7–6.5% in women, and 2–4% in men [4, 5]. These rates typically vary between 6 and 32% in different countries [6]. Although a variety of theories have been proposed, including bladder urothelial defects or dysfunction, mast cell activation and autoimmunity, the etiology and pathophysiology remain largely unexplored [2, 7].

Our previous study demonstrated that neuroinflammation in the spinal dorsal horn (SDH) might be associated to mechanism of interstitial cystitis, and contribute to mechanical allodynia [8, 9]. Neuroinflammation is a complex and well-coordinated process consisting of various glial cells in the central nervous system (CNS) and peripheral immune cells [10]. Astrocytes and microglia are two important glial cells that participate actively during the neuroinflammatory process by releasing proinflammatory cytokines such as interleukin-1β (IL-1β), interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α), etc. [11, 12]. Mitogen-activated protein kinases (MAPK) is a family of serine/threonine kinases that include p38 and Jun N-terminal kinase (JNK) signaling pathways; they are activated and phosphorylated in astrocytes and microglia, participating in the release of proinflammatory cytokines [13, 14]. A large amount of evidence suggests that neuroinflammation can lead to and aggravate pathological pain [15]. Previous studies have reported that BDNF-TrkB signaling can affect pain modulation by regulating neuroinflammation in some pathological pain models [16, 17].

Brain-derived neurotrophic factor (BDNF) is a major neurotrophic factor in the CNS, and plays an important role in learning and memory formation [18]. Also, BDNF plays a crucial role in the occurrence and development of neuroinflammation [11]. BDNF is regulated by neuronal activity and stored in dense-core synaptic vesicles at the terminals of the neurons [19]. After a nerve injury or inflammation in the spinal cord, dorsal root ganglia, or other related areas, BDNF is overexpressed and released from neurons [20]. The increased release of BDNF contributes to synaptic plasticity and central sensitization, thereby participating in the development of chronic pain [21, 22]. Tyrosine-protein kinase B (TrkB) is a high-affinity BDNF receptor [19], which is reportedly expressed in neurons, microglia, and astrocytes [23]. BDNF mediates its action through various signaling pathways triggered by the activation of the TrkB receptor.

However, whether and how BDNF-TrkB signaling in the SDH regulates neuroinflammation and mechanical allodynia in cystitis remains unknown. In this investigation, we present novel evidence showing how astrocytes and microglia activated to increase neuroinflammation and mechanical allodynia in our CYP-induced cystitis model through BDNF-TrkB-p38/JNK signaling.

As an important urinary neurotrophic factor, BDNF is detected to be abnormally elevated in some lower urinary tract diseases, such as overactive bladder [32], stress urinary incontinence [33], and interstitial cystitis [34]. BDNF plays a specific role in urinary frequency and urgency without affecting capacity [35]. It has been reported that intrathecal injection of exogenous BDNF could shorten the urination interval and make urinary cycle disorder in a rat model; treatment with ANA-12, however, resulted in the opposite effect [36]. As nerve growth factor (NGF), BDNF is also a major neurotrophic factor in the CNS and plays a crucial role in the occurrence and development of neuroinflammation [11].

The conjecture that BDNF may promote activation of astrocytes and microglia to contribute to aggravating neuroinflammation and mechanical allodynia of CYP-induced cystitis through BDNF-TrkB signaling is the first time to be proposed. In the present study, we demonstrated that BDNF colocalized mainly in neurons, which coincides with precious reports that BDNF is regulated by neuronal activity and stored in dense-core synaptic vesicles at the terminals of these neurons [19]. The co-stained neurons were mainly located in Laminate II to IV according to Rexed [37] and Steiner [38], which indicated that BDNF is mainly synthesized and released by neurons located in Laminate II to IV after receiving the nociceptive signal triggered by CYP. Furthermore, we also detected that TrkB receptors expressed in neurons, microglia, and astrocytes, which was consistent with the results reported [23]. Meanwhile, the expression of BDNF, TrkB, p-TrkB, Iba1, GFAP, p-p38, p-JNK, IL-1β, and TNF-α in the L6-S1 SDH were all upregulated. ANA-12 is a potent and highly selective TrkB antagonist, which can cross the blood-brain-barrier and exert central TrkB blockade without compromising neuron survival [39]. Treatment with ANA-12 could reverse upregulation significantly and attenuate mechanical allodynia. Besides, intrathecal injection of exogenous BDNF could further promote activation of astrocytes and microglia, aggravate neuroinflammation, and mechanical allodynia. In summary, our study has validated the conjecture proposed previously; in our CYP-induced cystitis rat model, the neurons received noxious signals and released BDNF, which promoted activation of astrocytes and microglia to release proinflammatory factors through BDNF-TrkB-p38/JNK signaling pathway, thus, contributing to neuroinflammation and mechanical allodynia (Fig. 7).

In our study, we administered ANA-12 (0.5 mg/kg) at three different time periods to study its effects, and achieve good results. Further, 0.5 mg/kg of ANA-12 is a recognized concentration, which has been used in many studies [24, 25]. We tried three different concentrations of ANA-12; the result showed that 1.0 mg/kg of ANA-12 was more effective than the other two concentrations used and 0.1 mg/kg of ANA-12 had almost no therapeutic effect. Although the effect of ANA-12 (0.5 mg/kg) to reduce the decrease of the mechanical threshold was not as significant as ANA-12 (1.0 mg/kg), their final treatment effect is no different. Studies have shown that high doses of ANA-12 may be the induction of cell death [39]. Therefore, the therapeutic effect of ANA-12 (0.5 mg/kg) is more moderate. K252a is a TrkB receptor antagonist that has been used in many studies as well [26]; however, it can also inhibit the activation of phosphorylase kinase, protein kinase A (PKA), protein kinase C (PKC) [40], or even antagonize TrkA and TrkC receptors [41].

BDNF-TrkB signaling modulates the pain process by regulating neuroinflammation which has been reported in other pathological pain models as well [16, 17]. Glial cells play an important role in the development of neuroinflammatory and neuropathic pain. Astrocytes and microglia participate in the development of pathological neuroinflammatory processes by releasing proinflammatory cytokines (IL-1β, TNF-α) [11, 12]. Nonetheless, there is little evidence indicating whether there is a link between BDNF-TrkB signaling and activation of astrocytes and microglia. In our previous studies, we observed that the activation of astrocytes and microglia contribute to mechanical allodynia in CYP-induced cystitis model [8, 9]. As far as we know, Iba1 and GFAP are the main markers of microglia and astrocytes, respectively. Moreover, OX-42 and p38 are two indirect markers for the activation of microglia [42]. In the present study, the expression change of OX-42 and p-p38 were consistent with Iba1, both reflected an increase in activation of microglia. Astrocytes are the first glial cells identified as activated after peripheral nerve injury [43]. Astrocytes play an important role in the development and maintenance of some forms of chronic pain [44] and participate in chemotherapy-evoked behavioral changes [45]. Microglia plays a major role in chronic pain [46]. CYP-induced cystitis is considered a complete experimental model of cystitis [8, 9, 34]. Through systemic intraperitoneal injection of CYP 25 mg/kg every 3 days, we established a chronic cystitis model [9]. Antagonizing TrkB receptor could inhibit the activation of astrocytes and microglia significantly in the cystitis model. Besides, injection of exogenous BDNF could further promote the activation of microglia and astrocytes significantly. And the results of immunofluorescence double staining showed that TrkB receptors expressed in microglia and astrocytes. These results indicated that BDNF-TrkB signaling may be involved in the activation of astrocytes and microglia, which might be an important discovery for the study of astrocytes and microglia.

There are several signaling pathways reportedly implicated in the inflammatory responses regulating proinflammatory cytokines and chemokine production, including the nuclear factor kappaB (NF-κB) pathway, the MAPK pathway, and the activated protein 1 (AP-1) pathway [47]. MAPK is a family of serine/threonine kinases which include p38, JNK, and Erk [48]. Previous studies have reported that JNK [13] and p38 [14] are involved in the activation of both astrocytes and microglia. In our study, the expression of p-p38 and p-JNK were upregulated in the SDH of our CYP-induced cystitis model, suggesting the activation of p38/JNK signaling pathway. In addition, ASK1-dependent p38/JNK was reported as a downstream molecule of the BDNF signaling pathway [49]. BDNF-TrkB signaling regulates the progress of p38/JNK signaling. It is noteworthy to mention that BDNF intervenes with other signaling pathways, such as SP-NK1R signaling and Glu-NMDA signaling, which are two classic signaling pathways in neuroinflammation and pathological pain [50, 51]. Moreover, BDNF-TrkB signaling can affect the expression of GluN2B on neurons, which plays an important role in synaptic plasticity and hyperalgesia [52].

Interestingly, in the present study, we discovered that treatment with ANA-12 not only restrained the activation of TrkB but also downregulated the expression of BDNF, compared to the CYP + DMSO group. Previous research has found that, after activation, astrocytes and microglia can also produce and release BDNF except proinflammatory factors [53, 54]. It has also been reported that TNF-α can bind to TNFR1 on astrocytes and microglia to promote the release of BDNF [11, 55]. TNFR1 is also present on neurons, and binding with TNF-α can also stimulate the neuron to release BDNF [56]. Therefore, we hypothesized that with the release of TNF-α reduced or inhibited after the TrkB receptor antagonized, the “positive feedback” was restrained.

Nonetheless, there were some limitations to our study. When the ANA-12 treatment was applied, we mainly detected the changes occurring at the protein and tissue level in the L6-S1 SDH that was harvested from the rats treated every other day after CYP injection; however, we did not detect those changes in other two groups. In addition, we used OX-42 in the immunofluorescence assay to show microglial activity in the SDH since we could not find an available Iba1 antibody. Although OX-42 is an activated microglia marker, it may ignore changes in the ratio of the M1 and M2 phenotypes, and thus, fail to demonstrate complete activation of the microglia [57]. Further study is required to find a solution to these problems.

Our results provide evidence that BDNF promotes activation of astrocytes and microglia to release TNF-α and IL-1β contributing to aggravating the neuroinflammation and mechanical allodynia through BDNF-TrkB-p38/JNK signaling in CYP-induced cystitis. Moreover, antagonizing TrkB receptors can alleviate neuroinflammation and mechanical allodynia significantly. These observations may provide a novel strategy for the study of astrocytes and microglia and treating cystitis.