Valproic acid-mediated neuroprotection and

Review For reprint orders, please contact: [email protected]

Valproic acid-mediated neuroprotection and neurogenesis after spinal cord injury: from mechanism to clinical potential

Spinal cord injury (SCI) is difficult to treat because of secondary injury. Valproic acid (VPA) is clinically approved for mood stabilization, but also counteracts secondary damage to functionally rescue SCI in animal models by improving neuroprotection and neurogenesis via inhibition of HDAC and GSK-3. However, a comprehensive review summarizing the therapeutic benefits and mechanisms of VPA for SCI and the issues affecting clinical trials is lacking, limiting future research on VPA and impeding its translation into clinical therapy for SCI. This article presents the current status of VPA treatment for SCI, emphasizing interactions between enhanced neuroprotection and neurogenesis. Crucial issues are discussed to optimize its clinical potential as a safe and effective treatment for SCI. Keywords: animal model • clinical trial • GSK-3 inhibitor • HDAC inhibitor • neurogenesis • neuroprotection • secondary injury • spinal cord injury • valproic acid

Background With an estimated global incidence between 250,000 and 500,000 people each year [1] , spinal cord injury (SCI) is a life-altering catastrophic event that can cause permanent neurological disabilities. To date, no fully restorative therapies have been found [2] . The difficulty of SCI treatment is closely associated with the prolonged period during which secondary injuries can occur, which includes the cascade of vascular, cellular and biochemical events that result in the progressive loss of cells (i.e., neurons, oligodendrocytes, astrocytes and precursor cells), axonal demyelination, and the establishment of an unfavorable environment that includes extensive inflammatory cells, glial scarring and molecular inhibitors [3,4] . Therapeutic strategies have been developed to counteract the secondary impairments and protect the survival of the remaining neurons (neuro protection). These strategies compensate for lost neurons and reconstruct the disrupted neuronal circuits by stimulating the proliferation of endogenous stem/progenitor cells, promoting neuronal differentiation

10.2217/RME.14.86 © 2015 Future Medicine Ltd

and maturation, and facilitating neurite outgrowth and synaptic integration (neurogenesis) [5–7] . Among the strategies that have been developed to restrict the extent of secondary injury, pharmacological treatment is considered to have the highest potential due to its ability to be immediately and non invasively administered after primary injury [2,4] . Valproic acid (VPA), an approved drug for mood stabilization, is receiving increased attention due to its interrelated therapeutic effects on neuroprotection and neurogenesis for the treatment of the injured spinal cord [8,9] . Current experimental evidence has demonstrated that VPA attenuates tissue damage and improves functional restoration in animal models of SCI by protecting the surviving neurons from secondary damage and stimulating neurogenesis, improving neuronal reinnervation and functional integration [10–14] . These therapeutic benefits of VPA have also been confirmed by abundant in vitro studies [15–18] . The underlying mechanism is associated with multiple interconnected signaling pathways mediated by the potent effects of VPA on HDAC inhibition

Regen. Med. (Epub ahead of print)

Tianci Chu‡,1, Hengxing Zhou‡,1, Lu Lu1, Xiaohong Kong2, Tianyi Wang1, Bin Pan1 & Shiqing Feng*,1 1 Department of Orthopaedics, Tianjin Medical University General Hospital, 154 Anshan Road, Heping District, Tianjin 300052, PR China 2 School of Medicine, Nankai University, 94 Weijin Road, Nankai District, Tianjin 300071, PR China *Author for correspondence: Tel.: +86 022 6036 3937 Fax: +86 022 8836 6139 shiqing.feng@ yahoo.com ‡ Authors contributed equally

part of

ISSN 1746-0751

Review Chu, Zhou, Lu et al. and GSK-3 inhibition [8,19] . As a short-chain fatty acid, VPA readily penetrates the blood–spinal cord barrier (BSCB) [20] , which is a prerequisite for treating injured spinal cord. Moreover, as a widely used clinical agent for decades, VPA has been proven to exhibit therapeutic benefits beyond mood stabilization at therapeutic serum concentrations (0.4–0.8 mM) with confirmed safety and tolerability [8,21] . These properties of VPA support its clinical use in pharmacological applications to treat SCI. However, relevant studies are sparse and have mainly evaluated the general effects of VPA on the treatment of SCI. To date, comprehensive reviews that summarize the ability of VPA to promote neuroprotection and neurogenesis after SCI are lacking, particularly with regard to in-depth descriptions of its mechanism and clinical potential, despite the urgent need for safe and effective therapies and the evident potential of VPA in the treatment of SCI [5,22] This dearth not only limits the directions and depth of VPA study but also impedes its translation into clinical use for SCI treatment. We therefore review the therapeutic effects of VPA on neuroprotection and neurogenesis that compensate for secondary damage in animal models of SCI. Meanwhile, we analyze the underlying mechanisms of VPA by linking the reported in vivo and in vitro evidence. Furthermore, based on the current status of VPA treatment, we present crucial issues that must be addressed to further develop VPA as a promising pharmacological candidate for the safe and effective treatment of SCI. VPA-induced neuroprotection & its mechanism A major obstacle in the treatment of SCI is the spread of cellular and tissue destruction due to BSCB breakdown, vascular dysfunction, inflammatory responses, excitotoxicity, apoptosis and autophagic cell death during the secondary injury period [4,22] . Worse still, this destruction can be aggravated by ongoing neuronal death and demyelination because the spared neural cells, especially neurons and oligodendrocytes, are highly vulnerable to injury-induced harmful factors [3,23] . Hence, neuroprotective interventions that counteract the subsequent pathophysiological processes are critical to the development of therapeutic strategies, and the amount of preserved spinal cord tissue is closely related to the degree of functional recovery after injury [4,22,24,25] . Although the amount of spinal cord tissue needed to maintain neurological functions in humans remains unknown, findings from animal models indicate that substantial functions are significantly preserved by axons from as few as 1.4–12% of the original neurons throughout the injury site [24,25] .

10.2217/RME.14.86


These findings are encouraging for SCI patients with little remaining neurological function below the injury level because even small improvements in the protection of spared neurons and axons against secondary destruction could significantly improve functional recovery [4] . It has been repeatedly demonstrated in animal models of SCI through the epigenetic modulation of relevant gene transcription that, as a potent HDAC inhibitor, VPA exhibits neuroprotective benefits by attenuating SCI-induced apoptosis, inflammation, neurotoxicity and autophagy during the secondary injury period. In addition, VPA upregulates prosurvival neurotrophic proteins, ameliorating the unfavorable environment and protecting the remaining neural cells from secondary damage (Table 1) [12,26] . Reduction of cell death

It has been shown that the administration of VPA at therapeutic doses attenuates demyelination and axonal loss and preserves the involved oligodendrocytes and neurons, leading to remarkable recovery of hindlimb activity in a rat transection model of SCI [28] . The mechanism was thought to prevent endoplasmic reticulum (ER) stress by reducing the progressive accumulation of the proapoptotic factor C/EBP homologous protein in the nucleus, which is possibly mediated by the HDAC inhibitory effect of VPA [28] . Similarly, VPA administration was found to protect motor neurons from oxidative stress and ER stress-induced death during the early period of SCI by inhibiting the release of cytochrome C and counteracting the activation of caspase-9 and caspase-12, which are possibly controlled by VPA-mediated HDAC inhibition [13] . Meanwhile, the SCI-triggered production of reactive oxygen species, the superoxide anion (O2-) and inducible nitric oxide synthase were also suppressed by VPA treatment, in association with the attenuation of reactive oxygen species-induced JNK activation and the downstream phosphorylation of Mcl-1 and Bim [13] . Additionally, VPA was reported to protect cultured neural progenitor cells from death under staurosporine- or hydrogen peroxide-stimulated conditions by the upregulation of antiapoptotic Bcl-xl expression and the downregulation of proapoptotic Bax expression via the NF-κB signaling pathway, which is possibly mediated by HDAC inhibition [35] . This in vitro evidence was further confirmed by injecting VPA into the developing rat CNS [35] . Moreover, VPA injection was found to limit the cell death induced by both excessive and defective autophagy during the stress response in a rat spinal cord contusion model [32] . This VPA-mediated decrease in autophagy was thought to be associated with HDAC inhibition, which possibly lowers the expression levels of Beclin-1 and LC3-II, biological

future science group

Valproic acid-mediated neuroprotection & neurogenesis after spinal cord injury

Review

Table 1. Valproic acid-mediated neuroprotection and neurogenesis in the treatment of spinal cord injury. Model

Administration paradigm

Effect and mechanism

Female Wistar rats; T10 contusion injury; 10 g × 25 mm impact

Intraperitoneal; 150 mg/kg twice daily for 1 day

Neurogenesis; promoted endogenous NSC proliferation; upregulated Nestin expression

Year/Ref. 2009/ [27]

Male ICR mice; T9 contusion injury; 90 kdyn impact

Intraperitoneal; initial 150 mg/kg 7 days po; 150 mg/kg daily for 1 week; combined with transplantation of embryonic mouse NSC 7 days po

Neurogenesis; promoted neuronal differentiation of NSCs via HDAC inhibition; reconstructed neuronal circuits by transplant-derived neurons in a relay manner; improved hindlimb motor function without CST axon re-extension

2010/ [10]

Female SD rats; T8 contusion injury

Intraperitoneal; initial 150 mg/kg 3 h po; 300 mg/kg daily or twice daily for 1 week, then daily for the next 2 weeks

Neuroprotection; reduced CHOP accumulation via HDAC inhibition; alleviated loss of oligodendrocyte, myelin and axons; increased tissue sparing; improved hindlimb recovery; no significance in electrophysiology

2011/ [28]

Female Wistar rats; T8 contusion injury

Intraperitoneal; initial 300 mg/kg immediately po; 300 mg/kg twice daily for 2 weeks

Neuroprotection; alleviated reduction of histone acetylation with increased Ac-H3 and Ac-H4; reduced apoptosis with upregulated protective proteins (e.g., Hsp70 and Bcl-2); improved locomotion recovery

2011/ [12]

Male SD rats; T9 clip contusion injury

Intraperitoneal; 200 mg/kg twice daily for 1 week

Neuroprotection; restored histone acetylation level via HDAC inhibition; restrained inflammatory reaction with decreased macrophage level; reduced cavity volume; improved hindlimb recovery

2012/ [29]

Female Wistar rats; T9 contusion injury; 10 g × 12.5 mm impact

Intraperitoneal; initial 300 mg/kg 8 h po; 300 mg/kg twice daily for 1 week

Neuroprotection; neurogenesis; promoted histone acetylation with upregulated Ac-H3; increased GDNF and BDNF expression via HDAC inhibition; attenuated Nogo-A inhibition on axonal growth of neurons; reduced apoptotic cells and lesion size; improved locomotion recovery

2012/ [11]

Male SD rats; T10 contusion injury; 10 g × 25 mm impact or 10 g × 50 mm

Subcutaneous; initial 150 or 300 mg/kg immediately po; 150 or 300 mg/kg twice daily for 5 days

Neuroprotection; reduced degradation of tight junction proteins via inhibition of MMP-9 expression and activity; attenuated BSCB disruption; prevented infiltration of neutrophils and macrophages; inhibited inflammatory mediators expression, apoptotic cell death, and caspase-3 activation; increased expression of Ac-H3, pAkt, HSP27, and HSP70; decreased expression of p53 and MMP-2; reduced lesion volume; improved functional recovery

2012/ [30]

Female SD rats; T12-L1 contusion injury; 10 g × 25 mm impact

Intraperitoneal; initial 100, 200 or 400 mg/kg 3 h po; 100, 200 or 400 mg/kg daily for 1 week

Neuroprotection; neurogenesis; HDAC inhibition; inhibited local inflammation and attenuated activation of macrophages/microglial cells at 400 mg/kg; increased tissue sparing and decreased cavity at 400 mg/kg; promoted neuronal dendrites growth with increased Ac-H3 and MAP2 expression; increased regional BDNF and GDNF levels at 400 mg/kg; improved locomotion function with 400 mg/kg as the most efficient dosage

2012/ [31]

Ac-H: Acetylated histone; BSCB: Blood–spinal cord barrier; CST: Corticospinal tract; ER: Endoplasmic reticulum; GABA: γ-Aminobutyric acid; ICR: Institute of Cancer Research; L: Lumbar; NSC: Neural stem cell; po: Postoperation; SD: Sprague–Dawley; T: Thoracic.


10.2217/RME.14.86

Review Chu, Zhou, Lu et al.

Table 1. Valproic acid-mediated neuroprotection and neurogenesis in the treatment of spinal cord injury (cont.). Model

Administration paradigm

Effect and mechanism

Female SD rats; T10 contusion injury

Intraperitoneal; initial 300 mg/kg immediately po; 300 mg/kg twice daily for 2 weeks

Neuroprotection; reduced autophagy with attenuated levels of LC3 and beclin-1; alleviated myelin sheath damage and neuronal loss; promoted tissue sparing; improve hindlimb recovery

Year/Ref. 2013/ [32]

Female SD rats; T10 contusion injury; 10 g × 50 mm impact

Intraspinal; initial 1.0 μl/h (500 ng/day) immediately po; 1.0 μl/h (500 ng/day) continuously for 3 days

Neuroprotection; inhibited P2X4R expression in activated microglia through p38 MAPKtriggered signaling; attenuated accumulation of microglia/macrophages and astrocytes; preserved spinal cord tissues and neuronal fibers; reduced gliosis; improved hindlimb locomotion

2013/ [33]

Male SD rats; T9 clip compression injury; 2-min compression with a closing force of 30 g

Intraperitoneal; initial 200 mg/kg immediately po; 200 mg/kg twice daily for 7 days

Neurogenesis; upregulation of Nestin and SOX2 expression; increased expression of spinal NSCs; possibly through the activation of GSK-3β signaling pathway

2013/ [34]

Male SD rats; T10 contusion injury; 10 g × 25 mm impact

Subcutaneous; initial 300 mg/kg immediately po; 300 mg/kg twice daily for 7 days

Neuroprotection; inhibited JNK activation; inhibited oxidative stress with attenuated increase of O2- and iNOS; inhibited ER stress-induced caspase-12 activation; suppressed cytochrome C release and caspase-9 activation; increased Bcl-2/Bax ratio and inhibited CHOP expression; attenuated motor neuron cell death

2014/ [13]

Male C57/BL6J mice; T7–9 hemisection injury

Intraperitoneal; initial 100 mg/kg 4 weeks po; 100 mg/kg twice daily for 4 weeks; combined with local lentiviral injection of SOX2 immediately po

Neuroprotection; neurogenesis; enhanced neuronal survival and maturation with features of GABAergic interneurons

2014/ [14]

Ac-H: Acetylated histone; BSCB: Blood–spinal cord barrier; CST: Corticospinal tract; ER: Endoplasmic reticulum; GABA: γ-Aminobutyric acid; ICR: Institute of Cancer Research; L: Lumbar; NSC: Neural stem cell; po: Postoperation; SD: Sprague–Dawley; T: Thoracic.

markers of autophagy that are induced after SCI [32,36–37] . By preventing pathological autophagy from exacerbating secondary cell death, VPA exerts potent neuroprotective benefits after SCI [37,38] . However, further studies are needed to reveal the in-depth mechanism of VPA-attenuated autophagy and its relationship with the other aforementioned neuroprotective roles of VPA. Attenuation of BSCB disruption

In addition to suppressing SCI-induced oxidative stress, ER stress, apoptosis, and autophagy, VPA has been reported to prevent the permeability and degradation of the BSCB, thereby attenuating the subsequent extensive inflammation during the secondary injury period after spinal cord contusion injury (Figure 1) [30] . BSCB disruption after injury is mainly caused by an upregulation of MMP-9 that degrades the basal components, such as the tight junction proteins occludin and ZO-1 [39] . The disruption of the barrier facilitates neutrophil and macrophage infiltration, vasogenic

10.2217/RME.14.86


edema and hemorrhage, all of which are tightly associated with the extent of cell death and tissue damage during secondary injury and exacerbate the degree of functional deficits [22,40] . Through HDAC inhibition, VPA attenuates the expression and activation of MMP-9, inhibits the production of inflammatory mediators (e.g., TNF-α, IL-β and IL-6), upregulates neuroprotective proteins (e.g., HSP70, HSP27 and pAkt), and downregulates apoptosis-associated factor p53, resulting in a remarkable preservation of neurons and oligodendrocytes, a reduction in the lesion volume, and improvement of functional restoration after SCI [39] . Furthermore, it has recently been reported that continuous intraspinal infusion of VPA can limit neurotoxicity (e.g., microglia overactivation and excitotoxicity) in addition to attenuating excessive inflammatory infiltration and reducing astrogliosis, which can lead to a significant preservation of the remaining neuronal fibers and a resultant improvement in hindlimb locomotion of rats after severe SCI [31,33] . The expression


Review

Valproic acid-mediated neuroprotection & neurogenesis after spinal cord injury

A

C

E

1 day

5 days

VPA (mg/kg) Sham

300

D

35 30

*

25

*

20 15 10 5 0

Sham Veh Veh

ED-1 (5 days)

MPO (1 day)

G

Evans blue (µg dye per g tissue)


40

VPA

150

300

Sham

30

F

25 20 15

*

10 5 0

150 300 VPA (mg/kg)

Veh

* Sham Veh

Veh

VPA

70 60 50 40

*

30 20 10 0

150 300 VPA (mg/kg)

Sham Veh

H

VPA

I 140

Vehicle VPA

120

Sham Veh VPA 110 kDa

ED-1

55 kDa

β-tubulin

100 J 80 * 60 40

*

20

Relative intensity of degraded MBP

150


B

Veh

Immunoreactive cells (percent of control)

Sham

1 day

VPA (mg/kg)

10 8 6

*

4 2

0 MPO-positive ED-1-positive neutrophils macrophages

0 Sham Veh

VPA

Figure 1. Valproic acid attenuates blood–spinal cord barrier disruption and blood infiltration after spinal cord injury. After spinal cord injury (SCI; 25 or 50 gm-cm), rats were treated with VPA (150 or 300 mg/kg). Blood–spinal cord barrier permeability was measured by using Evans blue dye and blood infiltration was assessed by counting the MPO or ED-1-labled cells at 1 and 5 days or western blot for ED-1 at 5 days after injury (n = 5/group). Representative spinal cord showing Evans blue dye permeabilized into moderately injured (25 gm-cm) spinal cord at (A) 1 and (C) 5 days and quantification of the amount of Evans blue at (B) 1 and (D) 5 days after injury. (E) Evans blue injected spinal cord from Veh or VPA (300 mg/kg)-treated rats at 1 day after moderately severe SCI (50 gm-cm). (F) Quantification of the Evans blue extravasation. (G) Photomicrographs from MPO-labeled neutrophils or ED-1-labeled macrophages in animals injected with Veh (left panel) or VPA (300 mg/kg, right panel). The representative sections were obtained at 2 mm rostral to lesion epicenter. Scale bar: 50 μm. (H) Quantitative analysis of MPO- or ED-1-positive cells. (I) Western blot of ED-1 at 5 days after injury. (J) Densitometric analyses of western blots show that VPA (300 mg/kg) significantly inhibited ED-1 expression when compared with that observed in Veh control at 5 days after injury. Data represent mean ± standard deviation. *p