Neuregulin-1 protects against acute optic nerve injury in rat model

JNS-13922; No of Pages 10 Journal of the Neurological Sciences xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns

Neuregulin-1 protects against acute optic nerve injury in rat model Wei Yang a,1, Tao-Tao Liu a,1, Xiao-Bin Song b, Yan Zhang c, Zhao-Hui Li d, Qian Hao a,⁎, Zhi-Hua Cui a,⁎, Hong Lei Liu a, Chun Ling Lei a, Jun Liu a a

Department of Ophthalmology, The First Hospital of Jilin University, Changchun 130021, PR China Department of Emergency Surgery, Jilin Province People's Hospital, Changchun 130021, PR China Department of Otorhinolaryngology, Head and Neck Surgery, The First Hospital of Jilin University, Changchun 130021, PR China d Department of Ophthalmology, People's Hospital of Changchun City, Changchun 130021, PR China b c

a r t i c l e

i n f o

Article history: Received 9 January 2015 Received in revised form 9 June 2015 Accepted 14 July 2015 Available online xxxx Keywords: Neuregulin-1 Optic nerve injury Neuroprotective effects Apoptosis

a b s t r a c t Objectives: In this study, we employed a rat model and examined the expression pattern of neuregulin-1 (NRG-1) in optic nerve and retinal ganglion cells (RGCs) in response to optic nerve injury to understand the role of NRG-1 in conferring protection against acute optic nerve injury. Method: Forty-eight male rats were randomly divided into two groups, the sham-operation group (n = 24) and optic nerve injury group (n = 24). Flash visual evoked potentials (FVEP) and fundography images were acquired at different time points following optic nerve injury (2 h, 1 d, 2 d, 7 d, 14 d and 28 d). Semi-quantitative analysis of NGR-1 expression pattern was performed by immunohistochemistry (IHC) staining. In a related experiment, 100 male rats were randomly divided into NGR-1 treatment group (n = 60) (treated with increasing dose of NGR-1 at 0.5 μg, 1 μg and 3 μg), normal saline (NS) group (n = 20) and negative control group (n = 20). Optic nerve injury was induced in all the animals and in situ cell death was measured by detecting the apoptosis rates using TUNEL assay. Results: Fundus photography results revealed no detectable differences between the sham-operation group and optic nerve injury group at 2 h, 1 d, 2 d and 7 d. However at 2 weeks, the optic discs turned pale in all animals in the optic nerve injury group. NRG-1 expression increased significantly at all time points in the optic nerve injury group (P b 0.05), compared to the sham-operation group, with NRG-1 expression peaking at 14 d and gradually declining by 28 d. Statistically significant differences in amplitude and latency of P100 wave were also detected between the optic nerve injury and sham-operation group (P b 0.05). In related experiment, compared to NS group, treatment with 1 μg and 3 μg of recombinant human NRG-1 resulted in statistically significant FVEPP100 amplitude values (all P b 0.05). Further, compared to the NS group, ganglion cell apoptosis was dramatically reduced in the NRG-1 group at all time points and the reduction was statistically significant in 3 μg NRG-1 treatment group at 7 d, 14 d and 28 d (all P b 0.05). Conclusion: Our results strongly suggest that NRG-1 is highly effective in preserving normal optic nerve function and is essential for tissue repair following optic nerve injury. Thus, NRG-1 expression confers protection against acute optic nerve injury in a dose-dependent manner. © 2015 Published by Elsevier B.V.

1. Introduction The optic nerve, known as cranial nerve II, is derived from the optic cups and contains embryonic retinal ganglion cells (RGCs). Optic nerve transmits visual information from the retina to brain and optic nerve injury, such as optic nerve crush and transection, mainly depends on the extent of damage to RGCs [1]. Loss of RGCs occurs after optic nerve injury through retrograde apoptosis and

⁎ Corresponding authors at: Department of Ophthalmology, the First Hospital of Jilin University, No. 71 Xinmin Road, Changchun 130021, PR China. E-mail addresses: [email protected] (Q. Hao), [email protected] (Z.-H. Cui). 1 Wei Yang and Tao-Tao Liu are both considered as first authors.

this process inhibits neuro-regeneration and neurite outgrowth, causing irreversible visual impairment [2]. RGCs long axons extend to form the optic nerve and are the sole output neurons from the retina to the brain, therefore, axonal transport of RGC plays a key role in neuronal function and visual perception [3]. Experimental optic nerve injury is a widely accepted model for retinal ganglion cell injury and displays similar intrinsic and extrinsic apoptotic events leading to RGC death [4]. Optic nerve injury is mainly a central nervous system (CNS) damage involving the selective loss of RGCs, and is accompanied by other adverse events such as mitochondrial dysfunction and oxidative stress [5]. RGC axons have poor regenerative capacity and thus are prone to permanent neurodegenerative defects following optic nerve injury [6]. Multiple pathological conditions of the eye, such as trauma, inflammation, disease and degeneration,

http://dx.doi.org/10.1016/j.jns.2015.07.023 0022-510X/© 2015 Published by Elsevier B.V.

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023

2

W. Yang et al. / Journal of the Neurological Sciences xxx (2015) xxx–xxx

cause optic nerve damage. In view of its high prevalence, intense focus is directed towards finding effective treatments for glaucoma related optic nerve injuries. For example, suturing optic nerve to peripheral nerve segment stimulates RGC axonal regeneration [7]. Similarly, neurotrophic factors can also greatly influence the survival and differentiation of neuron, and stimulate cell proliferation and migration of oligodendrocyte precursor cells, supporting RGC survival, axonal regeneration and optic nerve repair [8]. Neuregulin-1 (NRG-1) is expressed as both transmembrane and secreted isoforms, and binds transmembrane NEU/ERBB2 receptors [9]. NRG-1 signaling in neuronal network in the cortex and hippocampus is critical for cognition, learning, memory and emotion [10]. NRG1 isoforms are produced by alternative splicing, generating significant protein diversity from a single gene, with wide range of biological functions essential for nervous system development [11,12]. NRG-1 isoforms contain epidermal growth factor-like domains that allow its interaction with erythroblastosis B (ErbB) tyrosine kinase receptors, including ErbB2, ErbB3 and ErbB4, and signaling mediated through these receptors are important in neural development, myelin formation, synaptic transmission and nerve injury repair [13]. For instance, type III NRG-1 isoform expressed in the axolemma stimulates Schwann cell proliferation and controls myelin thickness and axon ensheathment, both of which are critical in nerve repair [14]. Interestingly, overexpression of NRG-1 reduced synaptic plasticity and impaired sensory gating and inhibitory neurotransmission. On the other hand, loss of neuronal NRG-1 expression resulted in hypoactivity and reduced neurotransmission, indicating that NRG-1 expression levels are tightly regulated to maintain the delicate balance of excitatory and inhibitory neurotransmission [15]. Multiple sclerosis, traumatic brain injury, stroke and Alzheimer's disease are associated with severe neurological defects and neurodegeneration, and NRG-1 appears to confer neuroprotection in these disease settings [16,17]. Interestingly, the structure of the neural retinal tissues is similar to nervous tissue and NRG-1 may indeed have a protective role against optic nerve injury [18]. Optic nerve injury is a serious health concern and currently there are no available strategies to prevent degeneration. As a prelude to the development of novel approaches to prevent RGC death, we investigated NRG-1 expression pattern during optic nerve injury and repair in a rat model. In addition, we tested the effectiveness of exogenous addition of recombinant human NRG-1 in preventing optic nerve damage and promoting repair and functional recovery. 2. Materials and method

cavities once immediately following the acute optic nerve injury (n = 20 for each dose group); the control group (NS group) (n = 20) was injected with normal saline following acute optic nerve injury; acute optic nerve injury was not induced in all the animals in the negative control group (n = 20). Left eyes of all the experimental animals were operated while no operation was performed in the right eyes. Clarity of refractive media in both eyes was verified, with pupillary reflexes and no eye ground changes. The experimental animals were given standard feeding in dedicated rooms at the experimental animal center and the rats adapted well to the environment to reduce the impact of external factors on the study results. One drop of ofloxacin 0.3% eyedrops was given in the affected eye three times a day for 3 days prior to operation to clean conjunctival sacs. 2.3. Building the model of acute optic nerve injury of rat A 10% chloral hydrate (0.3 ml/100 g) was administered through intraperitoneal injection as general anesthesia and 2% lidocaine injection for local anesthesia. Following anesthesia, rats were placed in prone position on the surgery table and the epidermis was dissected under binocular microscope. The capsule of Tenon was opened by cutting the lateral rectus and blunt dissection was performed along or near the surface of bitamporal sclera until the optic nerve was clearly exposed. Next, pressure was applied to optic nerve for 30 s using medium-sized tweezer (with force of 90 g) at a precise point 2 mm behind the eyeball, and sutured. In general, the optic nerve injury was successfully induced in the animals, with no bleeding or lowering of blood count. The rats in the Sham-operation group underwent the same procedures without applying pressure with tweezer. Normal blood supply in the retina was verified by direct observation of all the animals under an ophthalmoscope. Ofloxacin 0.3% eyedrops was used 4 times a day after surgery, 3 times a day after 2 weeks, and twice a day after 3 weeks. Gentamicin 20,000 units were injected after the operation through intramuscular injection once every day for 3 days to prevent infection. 2.4. Treatment method The paracentesis of anterior chamber was performed by microsyringe within 30 min following acute optic nerve injury in the NRG-1 treatment group to release 10–15 μl of aqueous humor. Next, the vitreous body was punctured at 0.5–1 mm behind the corneal limbus and recombinant human (rHu) NRG-1 was injected in a 10 μl volume at different concentrations of 0.5 μg, 1 μg and 3 μg. An equal volume of normal saline was injected in the same manner into vitreous body. Rats were checked to verify good blood supply in the retina by direct observation with an ophthalmoscope.

2.1. Ethics statement Animal studies were conducted in strict accordance with the Laboratory Animal Welfare Act, the Guide for the Care and Use of Laboratory Animals (National Institutes of Health, publication number 78-23, revised 1978) and all procedures were approved by the Care of Experimental Animal Center of the First Hospital of Jilin University. All efforts were made to minimize the suffering of animals. 2.2. Experimental animals and division A total of 148 two-month old Sprague-Dawley (SD) male rats (SPF clean grade), weighing 180–200 g, were obtained from the experimental animal center of the First Hospital of Jilin University. All National Medical Animal Usage Standards were followed. In the first experimental setup, 48 rats were randomly divided into two groups, the sham-operation group (n = 24) and the optic nerve injury group (n = 24). In these two groups, 4 rats were designated to each experimental time point of 2 h, 1 d, 2 d, 7 d, 14 d and 28 d. In the second experimental setup, 100 rats were randomly divided into three groups as follows: in the NRG-1 treatment group (n = 60), different dose of NRG-1 protein (0.5 μg, 1 μg and 3 μg) was injected into the vitreous

2.5. Experimental observation and test Fundus photography and visual evoked potential examinations were performed on all rats postoperatively. Fundus photography was performed under general anesthesia 30 min after mydriasis by application of tropicamide-phenylephrine hydrochloride eyedrops (Mydrin P, Santen Pharmaceutical, Osaka, Japan). Flash visual evoked potential (FVEP) examination was conducted 20 min after chloral hydrate injection under the same degree of anesthesia. To fully dilate pupil, tropicamide-phenylephrine eyedrops was applied 3 times in 10 min (with the diameter of 5 mm), the corneal electrode impedance was placed into conjunctival sac for direct contact with cornea surface before 20 min. dark adaptation. Surface anesthesia was performed twice with Alcaine (Alcon Humacao, Puerto Rico) before placing the corneal electrode. A drop of pasting agent (methylcellulose) was added on the corneal electrode after it was cleaned and disinfected. The all-field flash stimulator was applied, with the cornea at 30 cm distance from the white flash of 1.28 cd/m2 light intensity, stimulating frequency of 1 Hz and bandwidth of 1–100 Hz. Each rat was measured 3 successive times and the average was calculated. The interval between two flashes was 5 s with the other eye covered with black cloth. The latent time and amplitude of FVEP P100 was observed.

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023

W. Yang et al. / Journal of the Neurological Sciences xxx (2015) xxx–xxx

3

Fig. 1. Fundus photography of rats in each group. A: Control group; B: normal group; C: 2 h after injury; D: one day after injury; E: two days after injury; F: seven days after injury; G: 14 days after injury; H: 28 days after injury.

2.6. Hematoxylin-eosin (HE) staining Rats in every group received intraperitoneal anesthesia by chloral hydrate at 1 d, 2 d, 7 d, 14 d and 28 d. Both eyeballs and the optic nerve were collected for HE staining. The eyeball specimens were dehydrated in gradient alcohol after fixation and immersed for 5 min in xylene before embedding in wax. The specimen were incubated in at 60 °C overnight. The specimens covered with wax were embedded in liquid paraffin, immersed in water and the specimens were cut into 4 μm thick slices. After deparaffinization and gradient rehydration, the slices were stained with hematoxylin for 10 min and washed with running water for 5 min. Next, depigmentation was performed in 1% hydrochloric acid alcohol, washed with running water for 5 min, washed with 50%, 70% and 80% alcohol for 3 min each, stained with 0.5% eosin for 5 min, washed with 95% and 100% alcohol for 3 min each and treated

with xylene for 12 min, xylene II for 2 min and xylene III for 2 min. The slices were sealed by neutral resin and observed under the light microscope. 2.7. Immunohistochemistry Immunohistochemistry (IHC) was employed to study NRG-1 protein expression pattern in the retinal tissue of sham-operation group and optic nerve injury group. Chloral hydrate overdose was injected into the peritoneal cavity to sacrifice the rats and the eyeballs were removed. The eyeball specimens were paraffin embedded and cut into 4 μm thick slices. The tissue slices were sealed for 10 min at room temperature with 3% H 2O 2 after dewaxing and hydration. After washing with distilled water once, antigen retrieval by thermal denaturation was performed under high temperature and pressure,

Fig. 2. Fundus photography of rats in each group after NRG-1 treatment. A: 14 days negative group; B: 14 days normal saline group; C: 14 days NRG-1 0.5 μg group; D: 14 days NRG-1 1 μg group; E: 14 days NRG-1 3 μg group.

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023

4

W. Yang et al. / Journal of the Neurological Sciences xxx (2015) xxx–xxx Amp:20k

Amp:20k

SLM:Single

Time:250 ms

Grid: X 250 ms/d

SLM:Single

Time:250 ms

Grid: X 25 ms/d

Y 100.0 uv/d

Y 20.0 uv/d

8r 1I

A

B

Fig. 3. Flash visual evoked potential shape of (A) the Sprague-Dawley rats in the normal control group, which showed that the incubation period and amplitude of waves P1, N1 and P2 were relatively stable; and of (B) acute optic nerve after injury of Sprague-Dawley rats, indicating a relatively flat, wild and irregular shape with longer incubation period and lower amplitude.

followed by washing tissues with phosphate-buffered saline (PBS) for 5 min. Next, a series of steps were performed, which included blocking with normal goat serum (Jackson Immuno Research) for 20 min, staining with primary antibodies at 37 °C for l h, washing twice with PBS for 5 min each, incubation with biotinylated second antibody at 37 °C for l.5 h, PBS wash for 5 min twice, and incubation with strept–avidin–biotin complex (SABC) at 37 °C for l.5 h, followed by washing with PBS for 5 min developing the reaction with diaminobenzidine (DAB), with hematoxylin counterstaining after washing with distilled water. Hydrochloric acid alcohol differentiation was performed and after step-dehydration and mounting the sample, the slides were sealed and the results observed under the light microscope. The staining pattern appearing as yellow granular staining or plasma membrane staining were scored as positive (nucleus was not stained). Semi-quantitative analysis of the immunohistochemical staining results was performed at a fixed microscope height, pre-selected exposure time and f-number of digital camera, and photographs in 4 different visual fields were selected randomly at 400 × magnification. Image-proplus 5.02 software was used to measure the number of the positive cells of NRG-1.

2.8. Apoptosis detection Detection of apoptosis in ganglion cells was achieved by in situ terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining. After dewaxing and hydration, the tissue specimens were treated with freshly prepared 3% acetic acid (PH 2.5) at room temperature for 10 min and washed with distilled water 3 times for 2 min. Freshly diluted proteinase K (0.01 m Tris buffered saline [TBS, PH 7.5], 1: 200) was added to the slices to digest the sample for 15 min at room temperature. Labeling buffer (20 μl) was added to the specimens.

The tissue slices were incubated with a mixture of 1 μl TdT, DIG-dUTP and 20 μl labeling buffer in a humidified box at 37 °C for 2 h. The slices were washed 3 times with 0.0lM TBS (PH7.5) for 2 min each wash and blocked for 30 min at room temperature. Biotinylated anti-digoxin antibodies were added at 1:100 dilution and at 50 μl per slice. After incubation for 30–60 min in a humidified box at 37 °C, the tissues were 3 times with 0.0lM TBS (PH7.5) for 2 min. Next, the tissue slices were incubated with SABC at 37 °C 1.5 h. Color development with DAB was performed after washing the slices with 0.01 M TBS (PH 7.5). Finally, the slides were sealed with clear-mount and the results observed under the light microscope. Images were taken at 400× magnification in 4 different visual fields selected randomly. Image-proplus 5.02 software was used to count ganglion cells in every frame to measure the number of TUNEL positive cells in the retinal tissue. 2.9. Statistical analysis Statistical analysis was performed using SPSS17.0 software. All data were presented as mean ± standard deviation. The data were analyzed by one-way analysis of variance (ANOVA), and t-test was for comparisons between groups. A P value of b 0.05 was considered statistically significant. 3. Results 3.1. Fundus photography The eye ground of normal SD rats was orange, with clear blood vessels in radial arrangement. The color of the optic disc was deeper than the surrounding area. Macula lutea was absent and there was no detectable optic cup and optic disc contraction. Further, difference in the eye

Table 1 Detailed values of amplitude and latency of wave P100 in dark adaptation of Flash visual evoked potential among the control and experimental groups at the time point of 2 h, 1 d, 2 d, 7 d, 14 d, and 28 d (n = 4, ½x  SD). The latency of P100 waves (ms) (n = 4)

Before modeling 2h 1d 2d 7d 14 d 28 d

The latency of P100 waves (μV) (n = 4)

Shan-operation group

Experimental group

t

P

Shan-operation group

Experimental group

t

P

37.9 ± 3.5 41.7 ± 4.7 39.6 ± 5.1 41.1 ± 3.7 42.8 ± 4.1 38.4 ± 3.9 41.9 ± 4.4

38.4 ± 4.0 81.9 ± 5.9 90.1 ± 7.7 86.7 ± 6.3 79.7 ± 4.7 68.4 ± 3.9 67.2 ± 4.1

0.188 10.660 10.940 12.480 11.830 10.880 8.413

0.857 b0.001 b0.001 b0.001 b0.001 b0.001 b0.001

25.6 ± 4.7 25.7 ± 4.9 26.1 ± 5.6 24.9 ± 5.3 24.0 ± 4.9 23.7 ± 3.7 23.9 ± 4.5

24.9 ± 3.9 4.9 ± 2.1 3.7 ± 1.9 8.8 ± 3.3 12.7 ± 2.5 12.9 ± 3.1 13.7 ± 3.9

0.229 7.803 7.576 5.157 4.108 4.475 3.426

0.826 b0.001 b0.001 0.002 0.006 0.004 0.014

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023

W. Yang et al. / Journal of the Neurological Sciences xxx (2015) xxx–xxx

5

Fig. 4. The differences in (A) latency and (B) amplitude of flash visual evoked potential shape P100 waves at each time point after surgery in the NS group and the NRG-1 group (n = 4). Note: NS, normal saline.

ground between the sham-operation group and the NS group were also absent. The photographic images of the eye ground at 2 h–7d after acute optic nerve crush injury in the sham-operation group did not display any difference from the NS group. By the second week, the optic disc turned pale and continued to look pale at fourth week (Fig. 1). No visible changes were observed in the NRG-1 treatment group when compared to the negative control group and the NS group between the time points of 2 h–7d. However, the optic discs in the negative group and NS group turned pale at second week, with no detectable changes observed in the NRG-1 treatment groups (Fig. 2). 3.2. Visual evoked potential (VEP) The latency and amplitude of P1, N1 and P2 waves in the shamoperation group were relatively stable (Fig. 3A) and individual

Amp:20k

SLM:Single

Time:500 ms

Grid: X 50 ms/d

Y 50.0 uv/d

differences in waveform, amplitude and latency of the P100 wave were as expected (Fig. 3B). Abnormal waveforms in rats appeared at 2 h after injury, presenting as flat, wide and irregular with longer latency and lower amplitude of the P100 wave. The amplitude slightly recovered a few days later but still was at 50% lower than normal. The amplitude and latency of the P100 waves in the two groups are shown in Table 1. The differences in the amplitude and latency of P100 wave after surgery were statistically significant at each time point (P b 0.05, Fig. 4). The amplitude of FVEP-P100 in the NRG-1 treatment group and NS group was flattened on the first day after injury, but recovered gradually, with the largest change on the 14d (P b 0.05) as shown in Fig. 5. The amplitude in NRG-1 treatment group was significantly higher than the NS group at each time point (all P b 0.05). Compared to NS group, the latency at each time point in the 0.5 μg NRG-1 group was shorter, but the difference was

Amp:20k

SLM:Single

Time:250 ms

Grid: X 25 ms/d

A Amp:20k

SLM:Single

Time:250 ms

Grid: X 25 ms/d

Y 20.0 uv/d

C

Y 10.0 uv/d

B Amp:20k

SLM:Single

Time:250 ms

Grid: X 25 ms/d

Y 20.0 uv/d

D

Fig. 5. Flash visual evoked potential shapes of rats in every group after operation. A: 14 days normal saline group; B: 14 days NRG-1 0.5 μg group; C: 14 days NRG-1 1 μg group; D: 14 days NRG-1 3 μg group.

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023

6

W. Yang et al. / Journal of the Neurological Sciences xxx (2015) xxx–xxx

Table 2 Incubation and amplitude of wave P100 in dark adaptation of Flash visual evoked potential among the control and experimental groups at the time point of 1 d, 2 d, 7 d, 14 d, and 28 d (n = 4, ½x  SD). The latency of P100 waves (ms) (n = 4)

1d 2d 7d 14 d 28 d

The latency of P100 waves (μV) (n = 4)

NS group

NRG-1 0.5 μg

NRG-1 1 μg

NRG-1 3 μg

NS group

NRG-1 0.5 μg

NRG-1 1 μg

NRG-1 3 μg

90.17 ± 9.98 83.70 ± 8.11 77.15 ± 7.17 69.90 ± 8.01 67.08 ± 7.01

89.70 ± 8.12 79.91 ± 7.78 74.65 ± 8.07 68.17 ± 7.22 65.34 ± 6.64

88.36 ± 9.15 65.53 ± 8.66ab 62.86 ± 7.78b 55.92 ± 6.75ab 48.49 ± 6.01ab

65.31 ± 8.45abc 51.55 ± 7.67abc 44.31 ± 6.13abc 40.07 ± 6.04abc 37.17 ± 6.28abc

4.67 ± 0.39 7.08 ± 0.51 7.56 ± 0.66 12.14 ± 1.01 13.55 ± 1.23

6.71 ± 0.56a 11.75 ± 1.05a 12.63 ± 1.14a 17.75 ± 1.21a 19.99 ± 1.23a

9.86 ± 0.78ab 17.91 ± 1.21ab 18.65 ± 1.25ab 23.31 ± 2.30ab 25.04 ± 2.33ab

13.11 ± 1.11abc 25.16 ± 2.11abc 26.33 ± 2.26abc 31.17 ± 2.99abc 32.90 ± 3.01abc

Note: a—comparing to the NS group, P b 0.05; b—compared to the NRG-1 0.5 μg group, P b 0.05; c—compared to the NRG-1 1 μg group, P b 0.05; NS, normal saline; NRG-1, Neuregulin-1.

not statistically significant (P N 0.05). At 1 d, the latency in 1 μg NRG1 group was shorter and the difference was statistically significant, compared with the other time points in the same group (P b 0.05). Importantly, the latency and amplitude observed in the 1 μg NRG-1 group at 14 d and 28 d showed statistically significant differences with the corresponding values in the NS group, (P b 0.01). Similarly, the VEP between 3 μg NRG-1 group and NS group after injury were statistically significant at every time point (P b 0.05). Notably, differences between 3 μg NRG-1 group and NS group at time points 2 d, 7 d, 14 d and 28 d were statistically significant (P b 0,01) as shown in Table 2, and Fig. 6. 3.3. HE staining All three layers of cells are observed in the retina of normal adult rats, namely RGCs layer, the double-cell layer, and the optic fiber layer. RGCs are found in regular and compact arrangement, with clear nucleus. Retina in the sham-operation group thickened with mild edema, but no changes were observed in the number and arrangement of nuclei. In the optic nerve injury group, edema in retina occurred 2 h after injury, with severe edema in the nerve fiber layer and inner plexiform layer, with visible thickness and pale staining. On 1 d and 2 d after injury, the retinal edema disappeared, but more severe changes were observed, including reduced number of RGCs, cellular degeneration, visible thinning of nerve fiber layer and disorderly arrangement of cells in the inner plexiform layer accompanied by a thinning of the outer nuclear layer. At 7 d and 14 d after injury, the outer layer of retina showed marked thinning and no additional changes were seen even at 28 d after injury (Fig. 7). Thus, in our study, most relevant changes were observed within the 14 d period. In the second experimental setup, following optic nerve injury in the NS group, a sparse arrangement of retinal RGCs were observed, with several RGCs undergoing nuclear fragmentation, chromatin decondensation, vacuolar degeneration and visible thinning of the nerve fiber layer. Similar morphological observations were also evident in the 0.5 μg NRG-1 group. On the other hand,

although both 1 μg and 3 μg NRG-1 treatment groups showed sparse and irregular arrangement of RGCs, Muller fiber was observed in both groups, with significant reduction in the severity of vacuolar degeneration and thinning of nerve fiber layer. Within 2w following optic nerve injury, the number of RGCs in the treatment group and the control group reduced gradually with time, with no differences observed between normal saline group and negative control group at each time point (P b 0.05). However, in the NRG-1 treatment group, the number of RGCs at each time point was greater than the negative control group and the difference was statistically significant (P b 0.05), with the exception of the 2 d time point of 0.5 μg NRG-1 group. The differences were pronounced in the 1 μg and 3 μg NRG-1 groups when compared with the negative control group (P b 0.01) (Table 3, Fig. 8). 3.4. Expression of NRG-1 in the experimental group and the sham-operation group Immunohistochemistry results detected NRG-1 protein expression in retinal tissues of both sham-operation and optic nerve injury groups at every time point. NRG-1 protein was mainly localized in RGC and the inner nuclear layer (Fig. 9). Compared to the sham-operation group, retinal NRG-1 expression in the optic nerve injury group markedly increased initially but subsequently NRG-1 expression levels declined (Table 4, Fig. 10). In the optic nerve injury group, NRG-1 expression peaked at 14 d after injury and gradually decreased to normal levels by the fourth week. Compared to the sham-operation group, NRG-1 expression increased after acute optic nerve injury at every time point at 2 h, 1 d, 2 d, 7 d and 14 d (P b 0.001), with peak expression observed at 14 d after injury. However, by the fourth week these differences were no longer statistically significant (P N 0.05). 3.5. Apoptosis detection results TUNEL staining showed that the number of apoptotic ganglion cells in the NRG-1 treatment groups decreased significantly compared to NS

Fig. 6. The differences of (A) latency and (B) amplitude of flash visual evoked potential shape P100 waves at each time point after surgery in the NS group, the NRG-1 0.5 μg group, the NRG1 1 μg group and the NRG-1 3 μg group (n = 4). Note: NS, normal saline.

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023

W. Yang et al. / Journal of the Neurological Sciences xxx (2015) xxx–xxx

7

Fig. 7. Immunohistochemical staining to detect NRG-1 expression at each time point. A: HE staining at time point of 7 d after injury in the experimental group (×10); B: HE staining at time point of 14 d after injury in the experimental group (×10); C: HE staining at time point of 7 d after injury of 10 times in the control group; D: HE staining at time point of 14 d after injury of 10 times in the control group.

Table 3 The number of retinal ganglion cells at the time point of 2 d, 7 d and 14 d (/4HP, ½x  SD).

2d 7d 14 d

NC group

NS group

NRG-1 0.5 μg group

NRG-1 1 μg group

NRG-1 3 μg group

62.19 ± 6.13 51.17 ± 5.87 33.28 ± 5.07

61.19 ± 5.98 50.15 ± 6.01 31.90 ± 4.87

64.75 ± 6.99 62.56 ± 6.12a 51.41 ± 5.29a

76.88 ± 7.27ab 73.87 ± 6.99ab 60.42 ± 7.01ab

87.38 ± 8.05abc 84.65 ± 7.54abc 70.64 ± 6.01abc

Note: a—Comparing to the NC group and NS group, P b 0.05; b—compared to the NRG-1 0.5 μg group, P b 0.05; c—compared to the NRG-1 1 μg group, P b 0.05; NC, normal control; NS, normal saline; NRG-1, Neuregulin-1.

group at each time point (Fig. 11). Apoptotic glial cells were also detected in the optic nerve tissue, with strong brown staining in our TUNEL assays (Fig. 12). Compared to the NS group, the number of apoptotic ganglion cells at each time point decreased in the NRG-1 treatment groups based on the TUNEL staining results (Fig. 13). The number of apoptotic cells within each group increased with time, peaking at 7 d, and gradually reduced, but still remaining high at 14 d, which was higher than 28 d (Table 5). Compared to the NS group, the number of apoptotic cells at every time point in the 0.5 μg NRG-1 group decreased, but this decrease was not statistically significant (P N 0.05). In the 1 μg NRG-1 group, except at 1 d, a significant difference in cell death was observed among the various time points (P b 0.05), especially at 7 d, 14 d and 28 d (P b 0.01). Overall, the results suggested that administration of exogenous recombinant human NRG-1 within 30 min of optic nerve injury dramatically reduces the RGC cell death and promotes functional recovery in a dose dependent manner. 4. Discussion In this study, an acute optic nerve injury rat model was employed to investigate the influence of NRG-1 expression levels on the severity of injury and functional recovery following optic nerve injury. In addition, we examined whether exogenous supplementation with varying concentrations of human recombinant NRG-1 could offer adequate protection against optic nerve injury and enhance tissue repair and functional recovery. Our results showed that the latency was extended and FVEP-P100 amplitude sharply fell in the first day following injury, but maximal recovery was seen with 14 d. FVEP is useful tool for recoding the local responses from visual field defects and can diagnose optic nerve disorders in patients, in conjunction with optical coherence tomography imaging [19]. FVEP is a superior in monitoring the progression optic nerve injury, an provides an accurate estimation of demyelinization and axonal degeneration [20]. Most importantly, in our experimental model of optic nerve injury, protection and functional recovery by NRG-1 treatment was clearly demonstrated by shorter latency and higher FVEP-P100 amplitude. The results strongly suggested that NRG-1 is involved in protection against apoptosis and promotes tissue repair following optic nerve injury. These results are further supported by our observations of the accompanying morphological changes by HE staining following

NRG-1 treatment. In addition, dramatic differences in amplitude and latency of P100 wave were found between optic nerve injury group, sham-operation group and the NS group, suggesting that P100 wave is a suitable index for reliable evaluation of neural conduction along the visual pathway. Interestingly, our study also showed that the endogenous NRG-1 expression level was elevated in the early phase after optic nerve injury, and at later phase, the NRG-1 expression levels declined, suggesting that NRG-1 expression levels are sensitive to injury and may be important in optic nerve repair and protection. Consistent with this, overexpression of NRG-1 was observed in a previous study following nerve injury and was associated with improved nerve regeneration [21]. It is argued that in the early phase following optic nerve injury, NRG-1 secretion/release is accelerated by the activation of neurons and glial cells as a mechanism to prevent oligodendrocyte cell death and maintain the ability of axonal regeneration and remyelination to regenerate the optic nerve [22]. It was proposed that NRG-1 expression decreases in the later phase after nerve injury possibly because other factors may substitute for NRG1 in tissue repair [14].

Fig. 8. The number of retinal ganglion cells (/4HP) at each point in the NC group, NS group, the NRG-1 0.5 μg group, the NRG-1 1 μg group and the NRG-1 3 μg group (n = 4). Note: NC, negative control; NS, normal saline.

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023

8

W. Yang et al. / Journal of the Neurological Sciences xxx (2015) xxx–xxx

Fig. 9. Immunohistochemical staining of NRG-1 at different time points after injury. A: The positive staining of NRG-1 at 2 weeks after injury in control group (×20); B: the positive staining of NRG-1 at 2 weeks after injury in experimental group (×20); C: the positive staining of NRG-1 at 4 weeks after injury in control group (×20); D: the positive staining of NRG-1 at 4 weeks after injury in experimental group (×20).

Table 4 Positive NRG-1 cell counting in retina at different time points after injury (cells/mm2, ½x  SD).

2h 1d 2d 7d 14 d 28 d

Sham-operation group

Experimental group

t

P

20.21 ± 1.86 25.65 ± 2.11 16.15 ± 3.17 18.75 ± 3.04 21.37 ± 3.22 22.21 ± 3.01

26.17 ± 2.01 41.47 ± 3.58 68.75 ± 5.55 89.90 ± 9.79 140.41 ± 12.07 24.24 ± 2.29

4.353 7.614 16.460 13.880 62.190 1.073

0.005 b0.001 b0.001 b0.001 b0.001 0.324

Fig. 10. The number of NRG-1 positive cells at each point after acute optic nerve injury. Compared to the sham-operation group, retinal NRG-1 expression in the experimental group first increased markedly and later, the expression declined.

Optic nerve injury is the most common neuro-ophthalmic disorder caused by craniocerebral trauma, inflammation, high intraocular pressure, and compression by growing tumor [23–25]. Death of RCGs due to optic nerve crush or optic nerve transection inhibits tissue regeneration and aggravates the optic nerve injury [26]. Our results argue that delivery of exogenous human recombinant NRG-1 enhances repair and is protective against optic nerve injury in a dose-dependent manner by inhibiting apoptosis of RGCs. RGC death, axonal degeneration and reduced regenerative capacity are seen in several neurodegenerative diseases of the CNS [27]. Damage to RGC, following optic nerve injury, induces RGC apoptosis and neurodegeneration, leading to visual loss [1]. NRG-1 is as important neurotrophic factor with crucial roles in stimulating axon regrowth in RGCs after optic nerve injury [28]. There is also evidence that NRG-1 expression has a protective effect during neurodevelopment and neuro-regeneration by influencing RGC survival and regeneration after optic nerve injury [29]. Our results support the view that NRG-1 plays a protective role and we believe that this is mainly due to its inhibition of RGC apoptosis. In our study, the anti-apoptosis role of NRG-1 in RGCs is further supported by the fact that, compared to the NS group, ganglion cell apoptosis was reduced in the NRG-1 treatment group at most time points, with most significant affects seen in the 1 μg and 3 μg NRG-1 groups. Our study found that NRG-1 expression pattern and functional correlations are strongly related to tissue repair, neuroprotection and functional recovery from optic nerve injury by inhibiting apoptosis of RGCs. However, there are limitations in this study that may influence the overall results. First, other relevant mechanisms, such as the role of NRG-1in anti-inflammation, were not considered relevant in this study. Second, the rat model of optic nerve injury is not an exact replication of the clinical defects in humans, which limits the scope of its clinical applications. Finally, delivery of exogenous recombinant human NRG-1 likely did not

Fig. 11. Retina TUNEL staining. A: Time point of 14 d in the NRG-1 0.5 μg group, (×20); B: time point of 14 d in the NRG-1 1 μg group, (×20); C: time point of 14 d in the NRG-1 3 μg group, 20 times; D: time point of 14 d in the NS group, 20 times; E: time point of 28 d in the NS group, 20 times.

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023

W. Yang et al. / Journal of the Neurological Sciences xxx (2015) xxx–xxx

9

Fig. 12. Optic TUNEL Staining. A: Time point of 14 d in the NRG-1 3 μg group (×20); B: Time point of 14 d in the NRG-1 1 μg group (×20); C: Time point of 14 d in the NS group (×20).

accurately capture the recovery/regeneration process and better genetic models may be necessary to address this issue with inducible expression systems. In conclusion, endogenous expression level of NRG-1 was highly elevated at early phases following optic nerve injury, and at later phases, tNRG-1 expression levels declined, suggesting that NRG-1 expression levels may play an important role in optic nerve repair and protection. Importantly, exogenous recombinant human NRG-1 was capable of dramatically enhancing the repair and protection in the optic nerve injury model by inhibiting RGCs apoptosis, and this protective effect was dose-dependent.

Competing interests The authors have declared that no competing interests exist.

Fig. 13. The apoptotic number of ganglion cells counting at each point (n = 4). Compared to the NS group, the apoptotic number of ganglion cells at each time point decreased in the NRG-1 treatment groups based on TUNEL staining. Note: NS, normal saline.

Table 5 The apoptotic number of ganglion cells counting at each point (n = 4, ½x  SD).

1d 2d 7d 14 d 28 d

NS group

NRG-1 0.5 μg group

NRG-1 1 μg group

NRG-1 3 μg

9.97 ± 2.08 25.51 ± 2.58 35.16 ± 2.94 28.75 ± 3.01 24.97 ± 2.96

9.25 ± 2.70 23.46 ± 3.00 34.75 ± 3.21 26.65 ± 2.55 24.12 ± 2.71

8.83 ± 1.86 13.86 ± 2.85a 20.16 ± 2.22a 15.22 ± 2.01a 14.01 ± 1.96a

4.44 ± 1.01ab 8.17 ± 1.27ab 12.73 ± 1.52ab 9.16 ± 2.01ab 7.25 ± 1.07ab

Note: a—Comparing to the NS group, P b 0.05; b—compared to the NRG-1 0.5 μg group, P b 0.05; c—compared to the NRG-1 1 μg group, P b 0.05; NS, normal saline; NRG-1, Neuregulin-1.

Acknowledgments We would like to acknowledge the reviewers for their helpful comments on this paper.

References [1] H.M. Schmitt, H.R. Pelzel, C.L. Schlamp, R.W. Nickells, Histone deacetylase 3 (HDAC3) plays an important role in retinal ganglion cell death after acute optic nerve injury, Mol. Neurodegener. 9 (2014) 39. [2] X.T. Yang, G.H. Huang, D.F. Feng, K. Chen, Insight into astrocyte activation after optic nerve injury, J. Neurosci. Res. 93 (2015) 539–548. [3] X. Xia, R. Wen, T.H. Chou, Y. Li, Z. Wang, V. Porciatti, Protection of pattern electroretinogram and retinal ganglion cells by oncostatin M after optic nerve injury, PLoS One 9 (2014) e108524. [4] K.T. Janssen, C.E. Mac Nair, J.A. Dietz, C.L. Schlamp, R.W. Nickells, Nuclear atrophy of retinal ganglion cells precedes the bax-dependent stage of apoptosis, Invest. Ophthalmol. Vis. Sci. 54 (2013) 1805–1815. [5] M.K. Giacci, L. Wheeler, S. Lovett, E. Dishington, B. Majda, C.A. Bartlett, et al., Differential effects of 670 and 830 nm red near infrared irradiation therapy: a comparative study of optic nerve injury, retinal degeneration, traumatic brain and spinal cord injury, PLoS One 9 (2014) e104565. [6] V. Pernet, M.E. Schwab, Lost in the jungle: new hurdles for optic nerve axon regeneration, Trends Neurosci. 37 (2014) 381–387. [7] L.I. Benowitz, Y. Yin, Optic nerve regeneration, Arch. Ophthalmol. 128 (2010) 1059–1064. [8] W.K. Wong, A.W. Cheung, S.W. Yu, O. Sha, E.Y. Cho, Hepatocyte growth factor promotes long-term survival and axonal regeneration of retinal ganglion cells after optic nerve injury: comparison with CNTF and BDNF, CNS Neurosci. Ther. 20 (2014) 916–929. [9] G. Gambarotta, G. Ronchi, S. Geuna, I. Perroteau, Neuregulin 1 isoforms could be an effective therapeutic candidate to promote peripheral nerve regeneration, Neural Regen. Res. 9 (2014) 1183–1185. [10] N. Wang, G.F. Zhang, X.Y. Liu, H.L. Sun, X.M. Wang, L.L. Qiu, et al., Downregulation of neuregulin 1-ErbB4 signaling in parvalbumin interneurons in the rat brain may contribute to the antidepressant properties of ketamine, J. Mol. Neurosci. 54 (2014) 211–218. [11] S. Britsch, The neuregulin-I/ErbB signaling system in development and disease, Adv. Anat. Embryol. Cell Biol. 190 (2007) 1–65. [12] D.A. Talmage, Mechanisms of neuregulin action, Novartis Found Symp., 2892008. 74–84 (discussion-93). [13] J. Newbern, C. Birchmeier, Nrg1/ErbB signaling networks in Schwann cell development and myelination, Semin. Cell Dev. Biol. 21 (2010) 922–928. [14] F.R. Fricker, A. Antunes-Martins, J. Galino, R. Paramsothy, F. La Russa, J. Perkins, et al., Axonal neuregulin 1 is a rate limiting but not essential factor for nerve remyelination, Brain 136 (2013) 2279–2297. [15] A. Agarwal, M. Zhang, I. Trembak-Duff, T. Unterbarnscheidt, K. Radyushkin, P. Dibaj, et al., Dysregulated expression of neuregulin-1 by cortical pyramidal neurons disrupts synaptic plasticity, Cell Rep. 8 (2014) 1130–1145. [16] L. Wu, S. Walas, W. Leung, D.B. Sykes, J. Wu, E.H. Lo, et al., Neuregulin1-beta decreases IL-1beta-induced neutrophil adhesion to human brain microvascular endothelial cells, Transl. Stroke Res. 6 (2014) 116–124. [17] T. Nickl-Jockschat, T. Stocker, A. Krug, V. Markov, R. Huang, F. Schneider, et al., A Neuregulin-1 schizophrenia susceptibility variant causes perihippocampal fiber tract anomalies in healthy young subjects, Brain Behav. 4 (2014) 215–226. [18] M.C. Ortega, A. Bribian, S. Peregrin, M.T. Gil, O. Marin, F. de Castro, Neuregulin-1/ ErbB4 signaling controls the migration of oligodendrocyte precursor cells during development, Exp. Neurol. 235 (2012) 610–620.

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023

10

W. Yang et al. / Journal of the Neurological Sciences xxx (2015) xxx–xxx

[19] D. Pastare, I. Kire, R. Erts, G. Laganovska, A. Millers, Evaluation of axonal optic nerve damage using visual evoked potentials and optical coherence tomography in patients with multiple sclerosis, Medicina (Kaunas) 49 (2013) 474–478. [20] E. Kantorova, P. Ziak, E. Kurca, M. Koysova, M. Hladka, K. Zelenak, et al., Visual evoked potential and magnetic resonance imaging are more effective markers of multiple sclerosis progression than laser polarimetry with variable corneal compensation, Front. Hum. Neurosci. 8 (2014) 10. [21] R.M. Stassart, R. Fledrich, V. Velanac, B.G. Brinkmann, M.H. Schwab, D. Meijer, et al., A role for Schwann cell-derived neuregulin-1 in remyelination, Nat. Neurosci. 16 (2013) 48–54. [22] P.A. Fernandez, D.G. Tang, L. Cheng, A. Prochiantz, A.W. Mudge, M.C. Raff, Evidence that axon-derived neuregulin promotes oligodendrocyte survival in the developing rat optic nerve, Neuron 28 (2000) 81–90. [23] J. Pasol, Neuro-ophthalmic disease and optical coherence tomography: glaucoma look-alikes, Curr. Opin. Ophthalmol. 22 (2011) 124–132. [24] J. Wang, R.J. Hamm, J.T. Povlishock, Traumatic axonal injury in the optic nerve: evidence for axonal swelling, disconnection, dieback, and reorganization, J. Neurotrauma 28 (2011) 1185–1198.

[25] C. Dai, P.T. Khaw, Z.Q. Yin, D. Li, G. Raisman, Y. Li, Structural basis of glaucoma: the fortified astrocytes of the optic nerve head are the target of raised intraocular pressure, Glia 60 (2012) 13–28. [26] S.Y. Bu, G.H. Yu, G.X. Xu, Expression of insulin-like growth factor 1 receptor in rat retina following optic nerve injury, Acta Ophthalmol. 91 (2013) e427–e431. [27] J.C. Koch, L. Tonges, U. Michel, M. Bahr, P. Lingor, Viral vector-mediated downregulation of RhoA increases survival and axonal regeneration of retinal ganglion cells, Front. Cell. Neurosci. 8 (2014) 273. [28] I. Mahar, S. Tan, M.A. Davoli, S. Dominguez-Lopez, C. Qiang, A. Rachalski, et al., Subchronic peripheral neuregulin-1 increases ventral hippocampal neurogenesis and induces antidepressant-like effects, PLoS One 6 (2011) e26610. [29] B. Mead, A. Logan, M. Berry, W. Leadbeater, B.A. Scheven, Intravitreally transplanted dental pulp stem cells promote neuroprotection and axon regeneration of retinal ganglion cells after optic nerve injury, Invest. Ophthalmol. Vis. Sci. 54 (2013) 7544–7556.

Please cite this article as: W. Yang, et al., Neuregulin-1 protects against acute optic nerve injury in rat model, J Neurol Sci (2015), http://dx.doi.org/ 10.1016/j.jns.2015.07.023