The classic feature of Parkinson’s disease (PD) is the massive production and aggregation of the pathological form of $\alpha$-synuclein ($\alpha$-Syn) [1]. Patients with PD present with various clinical symptoms, including both motor and non-motor symptoms [2]. The benefits of current medications for PD are frequently only temporary, and they cannot stop the pathological process of neuronal degeneration; furthermore, long-term use is prone to cause terminal phenomena, on-off phenomena, and cardiovascular adverse reactions [3, 4]. Considering the multiplicity of pathological mechanisms and the uncontrollable irreversible loss of neurons, effective disease-modifying PD therapies are currently lacking [5]. Thus, elucidating the key molecular pathways of neuronal degeneration is vital for identifying the core pathogenesis of PD and developing targeted neuroprotective therapies and new interventional methods [6]. The fundamental mechanisms driving dopaminergic neuronal loss in the pathogenesis of PD remain unclear and constitute a major barrier to the development of disease-modifying interventions. Consequently, identifying novel molecular regulators that can attenuate neuronal deterioration is crucial and may accelerate the discovery of therapeutic targets.

Phosphodiesterase 4 (PDE4) regulates the cyclic adenosine monophosphate (cAMP) signaling pathway, and abnormal cAMP signaling activation can accelerate neuronal damage by promoting neuroinflammatory responses and oxidative stress; thus, PDE4 has been implicated in the pathogenic mechanisms of various neurodegenerative conditions [7, 8]. PDE4 inhibitors can improve neuroinflammation and oxidative stress by enhancing the cAMP signaling pathway and exhibit protective effects in various neurodegenerative disease models [9, 10]. PDE4 inhibitors are considered candidates for the treatment of PD because of their anti-inflammatory, antioxidant, and neuroplasticity-promoting properties [9]. Previous studies demonstrated that PDE4 inhibitors exert therapeutic benefits in PD models by safeguarding dopaminergic neurons and improving impaired spatial working memory; however, their exact molecular mechanisms are unclear [11, 12]. Currently used PDE4 inhibitors generally cause dose-limiting adverse reactions (especially gastrointestinal toxicity, such as nausea and vomiting), e.g., roipram, which was discontinued in a clinical trial for depression because of intolerable gastrointestinal side effects [7].

Roflupram (Roflu; also termed FFPM) is a potent PDE4 inhibitor (Fig. 1A). Roflu demonstrated superior selectivity for the PDE4A4, PDE4B2, and PDE4D4 subtypes compared with rolipram, along with notable anti-inflammatory and neuroprotective activities [13, 14]. Further, Roflu shows a low incidence of adverse effects such as vomiting, and it can cross the blood-brain barrier [13]. Roflu enhances autophagy and inhibits the activation of the inflammatory body of microglia, thereby reducing inflammatory responses and neural damage. Its advantages have been confirmed in neurodegenerative disease models, i.e., it can significantly improve cognitive dysfunction, and exert neuroprotective effects through the cAMP/protein kinase A (PKA)/cAMP-response element binding protein (CREB) signal transduction axis. Roflu also alleviates neuronal damage in brain ischemic injury models [9, 13, 15], suggesting its potential as a safe therapeutic intervention for stroke. However, the therapeutic potential of Roflu in PD requires further research.

Fig. 1.

Roflu increases the cell viability in the context of 6-OHDA-induced injury. (A) Structural diagram of Roflu. (B) The viability of cells treated with a gradient of Roflu concentrations (10–80 µM) for 48 h was determined by MTT assay. (C) The cytotoxicity of 6-OHDA (25–200 µM) following a 48 h exposure was assessed in SH-SY5Y cells using the MTT method. (D) A 1 h pre-incubation with Roflu (5–20 µM) was applied to cell cultures, and then exposed to 6-OHDA (100 µM) for 48 h. (E) SH-SY5Y cell viability was assessed by MTT after pretreatment with 20 µM Roflu for 1 h prior to exposure to 100 µM 6-OHDA for 48 h. Data are presented in the form of mean $\pm$ SD (n = 6). ^***p 0.001. ns: no difference. CTL, control; Roflu, Roflupram; 6-OHDA, 6-hydroxydopamine; MTT, 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; SD, standard deviation.

Cellular oxidative damage and stress play pivotal roles in the pathogenesis of PD. Oxidative stress is a state of stress caused by the overproduction of reactive oxygen species (ROS) [16, 17]. The high metabolic rate and the propensity for dopamine auto-oxidation render dopaminergic neurons particularly vulnerable in PD. Oxidative stress leads to a vicious cycle of mitochondrial dysfunction. The deficiency in mitochondrial complex I results in excessive ROS production, which is further exacerbated by a significantly compromised antioxidant defense system in the brains of PD patients, preventing cells from clearing excessive ROS. Abnormal aggregation of $\alpha$-Syn can occur because of protein oxidative damage, and aggregated $\alpha$-Syn inhibits mitochondrial functioning, thus exacerbating oxidative stress [18, 19]. 6-Hydroxydopamine (6-OHDA) exhibits highly selective toxicity toward catecholaminergic neurons, thus it has been used for generating animal PD models. 6-OHDA differs from dopamine by only one hydroxyl group and can be actively absorbed by cells via the dopamine transporter. Subsequently, dopaminergic neurons in the nigrostriatal pathway are selectively eliminated [20, 21].

This study evaluated the neuroprotective effects of Roflu, a PDE4 inhibitor, in a PD model. Previous study showed that PDE4 inhibitors exerted potential neuroprotective effects in PD, including 1-methyl-4-phenylpyridinium (MPP⁺)/1-methyl-4-phenyl-1,2,3, 6-tetrahydropyridine (MPTP), rotenone, and $\alpha$-Syn overexpression PD models [22, 23, 24]. The neuroprotection offered by PDE4 inhibitors is linked to improved regulation of autophagy and pathways involved in mitochondrial. However, it is unclear whether Roflu would exert a protective effect in the classic 6-OHDA model of PD, which is crucial for further validation of the role of Roflu in PD and its potential as a candidate drug. Here, we used a 6-OHDA-injured SH-SY5Y cell model to investigate the neuroprotective effects of Roflu and the involvement of a PDE4 target, showing that Roflu significantly increased cell viability and tyrosine hydroxylase (TH) expression levels induced by 6-OHDA while reducing lactate dehydrogenase (LDH) release. Furthermore, Roflu treatment promoted the recovery of mitochondrial membrane potential (MMP). We investigated PDE4B expression, which was lacking in previous studies, and found that Roflu downregulated 6-OHDA-induced PDE4B expression. PDE4B knockdown increased cell viability and reduced LDH release, whereas the protective effect of Roflu was antagonized by PDE4B overexpression. Our findings demonstrate that by inhibiting PDE4, Roflu alleviates 6-OHDA-induced oxidative damage, which confirms its considerable potential as a neuroprotective agent. This study identifies Roflu as a novel therapeutic candidate and provides a fresh mechanistic rationale for treating PD and other oxidative stress-related neurodegenerative disorders.

We found that specific inhibition of PDE4B by Roflu underlies its neuroprotective effects in a PD model established by 6-OHDA. Specifically, Roflu mitigated oxidative stress during the injury of SH-SY5Y cells, thereby reversing the degeneration of dopaminergic neurons. These findings highlight PDE4B inhibition as a potential therapeutic strategy for PD and deepen the understanding of the regulatory mechanism of the PDE4/cAMP signaling axis in neurodegenerative diseases, offering new candidate compounds and mechanism-based evidence for the targeted treatment of PD and related oxidative damage-induced neurodegenerative diseases. Although current mainstream drugs for PD, such as levodopa, can temporarily improve motor function, they cannot stop the pathological process of neuronal degeneration and are prone to cause adverse reactions, such as the wearing-off and on-off phenomena, when used for a long time [3]. Therefore, new intervention targets and drugs that can delay neuronal damage are needed. Conventional PDE4 inhibitors, such as rolipram, were discontinued in Phase II trials owing to adverse gastrointestinal reactions such as vomiting [7]. Compared with rolipram, Roflu shows higher selectivity for PDE4A4, PDE4B2, and PDE4D4 subtypes (IC₅₀ values for PDE4A4/B2/D4 subtypes are 3 to 10 fold lower than those of rolipram), and demonstrates minimal emetic potential along with favorable blood-brain barrier penetration in animal studies [13, 14], suggesting that its therapeutic window may be significantly better than that of rolipram. Roflu demonstrates significant advantages in safety, with no cytotoxicity observed within the concentration range of 10 to 80 µM in experiments. Additionally, Roflu has been proven to enhance the autophagic clearance of $\beta$-amyloid in Alzheimer’s disease models [15], suggesting that it cooperatively regulates abnormal aggregation of $\alpha$-Syn in PD [18]. This multi-target intervention is particularly important in complex neurodegenerative diseases.

Previous research showed that PDE4 inhibitors have potential neuroprotective effects in PD, including in MPP⁺/MPTP, rotenone, and $\alpha$-Syn overexpression PD models [22, 23, 24]. However, it was unclear whether Roflu would exert protective effects in the classic 6-OHDA-induced PD model. The 6-OHDA model is one of the most classic and widely used pharmacological induction models in PD research, particularly for simulating the pathological hallmarks of PD, the distinctive pathological changes in nigrostriatal dopaminergic neuronal loss [30, 31]. In the present study, an injury model of SH-SY5Y cells induced by 6-OHDA was employed to systematically assess the neuroprotective efficacy of Roflu. Our findings indicated that 6-OHDA treatment markedly decreased SH-SY5Y cell viability, promoted LDH release, and reduced the expression of TH. A key enzyme in dopamine synthesis, TH’s expression level directly reflects the functional state of dopaminergic neurons [27]. Roflu pretreatment significantly reversed these changes, indicating that Roflu can exert protective effects at multiple levels, such as cell survival, functional maintenance, and oxidative stress regulation. Thus, our results demonstrate that Roflu exerts antioxidant effect, thereby corroborating the established role of PDE4 inhibition in mitigating neural damage across diverse experimental models [22, 23, 24] and further supporting the potential of PDE4 inhibitors as neuroprotective agents. Roflu needs to be further examined in 6-OHDA and $\alpha$-Syn transgenic rodent models with regard to its protective effects on motor symptoms, dopamine content, and substantia nigra-striatum pathology, and to systematically evaluate the safety of long-term administration.

The core manifestations of oxidative damage in PD include an overproduction of ROS and a compromised antioxidant system (such as glutathione) [32]. Dopamine metabolism is a major source of ROS, while mitochondria, the cellular powerhouses, are also the primary site for its generation. Further amplify this oxidative burden, their dysfunction (especially the reduced activity of complex I) directly leads to an ATP synthesis disorder, causing neurons to be in an energy crisis which initiates a vicious cycle of sustained ROS overproduction [33]. Excessive ROS can directly attack and oxidatively damage lipids, proteins (e.g., $\alpha$-Syn), and DNA, thus damaging cell structure and function. Abnormal $\alpha$-Syn protein aggregates into Lewy bodies, potentially damaging mitochondria themselves and interfering with the phosphatase and tensin homolog PTEN-induced (PINK1)/Parkin pathway-mediated mitochondrial autophagy. Consequently, this results in the accumulation of dysfunctional mitochondria and creates a vicious cycle exacerbating mitochondrial dysfunction and oxidative damage [16, 34]. The selective degeneration of dopaminergic neurons is driven by a self-perpetuating cycle of mitochondrial impairment and persistent oxidative stress. In this study, 6-OHDA exposure resulted in significant ROS accumulation and MMP reduction, establishing oxidative injury as a key contributing factor in 6-OHDA-mediated neurotoxicity. These findings are consistent with those of a previous study demonstrating that PDE4 inhibitors attenuate MPP⁺-induced MMP loss in cells [22, 23, 24]. Roflu significantly decreased ROS levels induced by 6-OHDA and restored MMP, indicating that Roflu exerts protective effects at multiple levels, including cell survival and oxidative stress regulation. This observation aligns with reported antioxidant and mitochondrial protective effects of PDE4 inhibitors in other neurological injury models [9, 35, 36], underscoring their promise as neuroprotective agents.

Next, we investigated the connection between the neuroprotective effects of Roflu and PDE4. The PDE4B expression within the substantia nigra is the highest among all phosphodiesterases and is related to the function of the striatum and the pathology of PD [29]. We detected a marked upregulation of PDE4B expression in the 6-OHDA-treated SH-SY5Y cell model, and Roflu could markedly inhibit this trend. Moreover, knockdown of PDE4B simulated the effect of Roflu, whereas overexpression of PDE4B counteracted the improvement effect of Roflu on cell viability and LDH release and even reversed its regulatory effects on ROS and MMP. These findings collectively demonstrate that the neuroprotective mechanism of Roflu involved specifically through the inhibition of the PDE4B subtype and that PDE4B is the key target mediating its neuroprotective effect. This study addresses the lack of attention to PDE4B in previous studies, clarifying the role of PDE4B with regard to the protective effects of Roflu. This study confirmed that Roflu can exert neuroprotection in the PD model by attenuating oxidative stress. However, the specific mechanism of action still requires further research. The nuclear factor-erythroid 2-related factor 2 (Nrf2)/ heme oxygenase-1 (HO-1) pathway is the core regulatory hub for the cellular capacity to combat oxidative stress. Previous study have shown that inhibiting PDE4 can regulate it [10], which provides a direction for further exploration of more downstream molecular pathways.

Our study thus provides experimental evidence that Roflu exerts neuroprotective effects in a well-established 6-OHDA-induced PD model, complementing the previous data from different PD models such as MPP⁺/MPTP. These findings strengthen the universality of PDE4 inhibitors in various models of PD. Our results clearly identify PDE4B as the core target for Roflu in regulating oxidative damage, resolving the key controversy of “functional heterogeneity of PDE4 subtypes” in previous studies [22, 23, 24]. Given its high selectivity and favorable safety profile, Roflu shows significant promise as a drug candidate targeting of PD and other neurodegenerative diseases that share common pathological features.

All data generated or analysed during this study are included in the research article.

JHZ: Conceptualization, Methodology, Investigation, Funding acquisition, Writing—original draft. XHY: Conceptualization, Methodology, Investigation, Writing—original draft. JLX: Conceptualization, Methodology, Investigation. QMW: Data curation, Formal analysis, Software. ZML: Conceptualization, Funding acquisition, Project administration, Resources, Writing—review & editing. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

Not applicable.

Not applicable.

This work was supported by National Nature Science Foundation of China (Grant No. 82304434), GuangDong Basic and Applied Basic Research Foundation (Grant No. 2023A1515111199), China Postdoctoral Science Foundation (Grant No. 2022M713263), Scientific Research Starting Foundation for High-level Talents of Meizhou People’s Hospital (Grant No. KYQD202501, KYQD202502), PSM Guangdong Pharmaceutical Popular Science Research Foundation (Simcere Foundation) (Grant No. 2025KP178, 2025KP78), Shining Across China-Medicinal Research Fund (Grant No. Z04J2023E095), Social Development Science and Technology Plan Project of Meizhou (Grant No. 2024C0301079).

The authors declare no conflict of interest.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/JIN45490.