Brain derived neurotrophic factor keeps pattern electroretinogram from dropping after superior colliculus lesion in mice

Abstract

AIM

To determine if brain-derived neurotrophic factor (BDNF) could offer protention to retinal ganglion cells following a superior colliculus (SC) lesion in mice using pattern electroretinogram (PERG) and optical coherence tomography (OCT) as a measures of ganglion cell response and retinal health.

METHODS

Seven C57BL/6J mice with BDNF protection were tested with PERG and OCT before and after SC lesions.

RESULTS

Compared with baseline PERG, the amplitude of PERG decreased 11.7% after SC lesions, but not significantly (P>0.05). Through fast Fourier transform (FFT) analysis of the PERGs before and after SC lesions, it was found that dominant frequency of PERGs stayed unchanged, suggesting that the ganglion cells of the retina remained relatively healthy inspite of damage to the ends of the ganglion cell axons. Also, OCT showed no changes in retinal thickness after lesions.

CONCLUSION

It was concluded that BDNF is essential component of normal retinal and helps retina keeping normal function. While retina lack of BDNF, ex vivo resource of BDNF provides protection to the sick retina. It implies that BDNF is a kind therapeutic neurotrophic factor to retina neurodegeneration diseases, such as glaucoma, age related macular degeneration.

INTRODUCTION

Previous researches have provided compelling evidence that retinal ganglion cells (RGCs) generate pattern electroretinogram (PERG)^[1]. The PERG therefore permits the examination of the health of these output cells. PERG is therefore a useful experimental tool to evaluate potential treatments to preserve ganglion cell function in clinical conditions such as glaucoma and diabetic retinopathy.

Optical coherence tomography (OCT) is an optical signal acquisition and processing method. It captures micrometer-resolutionand three-dimensional images from within optical scattering media (e.g. biological tissue). In ophthalmology, it has been used widely as a diagnostic tool to examine the structure of whole retina in vivo, such as defining the thickness of the different layers of retina. Such measures have proved important in determining the changes with retinal damage or disease^[2]–^[5].

Brain-derived neurotrophic factor (BDNF) was first found by German neurobiochemist Barde et al^[6] in 1982^[7]. It acts on certain neurons of the central nervous system and the peripheral nervous system acting to support the survival of existing neurons, and encouraging the growth and differentiation of new neurons and synapses in development^[8]–^[13]. BDNF is also expressed in the retina but is also available to the retina via its production in the superior colliculus (SC). It was known that after SC lesions, the retina no longer is able to obtain BDNF from the SC and its functional levels decrease immediately and remain at low level at least for 4mo. RGCs inside retina would therefore have to produce additional BDNF to satisfy their needs^[2]. It was also shown that in early stages of glaucoma, in patients' serum, the concentration of BDNF is significantly lower than normal. Therefore, BDNF in the serum might be a useful biochemical marker for early detection of primary open angle glaucoma^[14]. All above information suggests that an ex vivo resource of BDNF might provide protection to retina following damage or disease.

In order to determine if there was interrupted signal from damaged area, i.e. lesioned SC. Fast Fourier transform (FFT) analysis of the PERGs before and after SC lesion was implemented. This kind of FFT analysis can manifest all types of frequency components in PERG wave^[15].

MATERIALS AND METHODS

Animals and Husbandry

This study was supported by Huaiyin Normal University. All experiments were conducted according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. C57BL/6J mice purchased from Jackson Laboratory (Bar Harbor, ME, USA) were raised using typical 12h light vs 12h dark cycle and were fed with mice rasing diet. For the present study, 7 mice, 4 months old, were used. PERG and OCT were recorded under ketamine/xylazine anesthesia.

Brain-derived Neurotrophic Factor Administration

Under ketamine/xylazine anesthesia, Hamiltom microsringe was used to inject progressively (total time 5min) 1 µL BDNF only into the left eye of each mouse. On the side surface of mouse eye, there is a natural line under which is posterior chamber of mouse eye resides. The microsringe needle just pierced the posterior chamber along the posterior side of this line. This procedure allowed the BDNF solution to diffuse to the retina. Human recombinant BDNF was used at a concentration of 200 ng/µL (Amgen, Thousand Oaks, CA, USA)^[8]. After injection, mice were allowed to recover and put back in their cages. One week later, surgery was performed to lesion the SC.

Surgical Procedure

Mice were weighed by putting in a cup and anesthetized with intraperitoneal injections (about 0.6 mL/kg) of a mixture of xylazine, 8.6 mg/mL, and ketamine, 42.8 mg/mL. Under sterile conditions, the bone was removed over the SC contralateral to the eye injected with BDNF. The coordinates for this surgery were 0.5 mm lateral to the central suture and 2.9 mm posterior to the bregma. A narrow column of cortical tissue was then aspirated to expose the SC^[16]. Then the superficial layers of the SC were also removed by aspiration. Mice were then allowed to recover from anesthesia and were put back in their cages. Surgical ablation of SC was considered a better than lesions done with neurotoxins, such as ibotenic acid or kainic acid^[17] so as to get rid of possible retrograde toxic effects to RGCs^[2].

Pattern Electroretinogram Recording

Before BDNF injections, a baseline PERG was recorded from 7 left eyes on 7 mice and then repeated after SC lesion, experimented PERG was recorded from 7 eyes with BDNF administrated (piercing of eye ball, any way, had negative impact of retina funcion and mouse health, it took some time to let retina function and mouse health recover to normal level). We chose time window one week later in order to allow for BDNF to reach the retina and have its full effect after the SC lesions. A detailed description of the PERG technique was reported before^[1],^[3],^[18]–^[20]. Here we briefly state the whole procedure. Mice were weighed and anesthetized with intraperitoneal injections (about 0.6 mL/kg) of a mixture of ketamine, 42.8 mg/mL, and xylazine, 8.6 mg/mL. After mice were anesthetized, they were then carefully fixed in a stereo-taxic frame that can make mice have no vision obstacle. The body of the animal was kept at a stable body temperature of 37°C by using a feedback controlled heating pad. The mice eyes were widely open, and because mice position was lower than the monitor, their optical axes were made laterally and upwardly to the center of the monitor's screen which contained visual stimuli. A PERG electrode (0.25-mm diameter silver wire was configured to a semicircular loop of 2 mm radius) was placed on the cornea surface by means of a micromanipulator. In order to prevent cornea from drying, a small drop of balanced saline solution was administrated every 30min. Reference and ground electrodes were needles inserted under the skin of the scalp on the back of the head and the tail, respectively. Visual stimuli were contrast-reversing (2 reversal/s) horizontal squarewave (100% contrast, mean luminance 50 cd/m², 0.05 cyc/°) generated by a coded graphic card (VSG Cambridge Research Systems, Rochester, UK) on a cathode-ray tube display (Sony Multiscan 500; Sony, Park Ridge, NJ, USA) whose center was aligned with the projection of the pupil. The pupils were not dilated with atropine, because the mouse eye has a large depth of focus^[12]–^[13],^[17]. At the viewing distance of 20 cm, the stimulus scope covered an area of 69×63°. Three to five continuous PERG responses to 600 contrast reversals each were recorded. The responses were overlapped to check for consistency and then averaged (1800-3000 sweeps). The PERG waveform consists a major positive peak at about 90 to 120ms followed by a slower negative wave with a broad trough at approximately 200 to 300ms^[2],^[19],^[21].

Fast Fourier Transform Analysis

FFT analysis was carried out using the FFT function (Mathsworks, R2010a) built in a Matlab code.

Spectral-Domain Optical Coherence Tomography

In vivo spectral domain (SD)-OCT (Bioptigen, Inc., Durham, NC, USA) images were scanned before surgery and at immediately time point after SC lesion. Three-dimensional images were obtained from a raster of 100 b-scans centered on the optic disk and spanning an area of 1.3×1.3 mm². Two obvious land marks were identified in b-scans of all experimental mice: the retinal surface, the border between the inner nuclear layer (INL) and the outer plexiform layer (OPL). Inner retina was defined as the distance between the retinal surface and the border between INL and OPL. A hollow cylinder shaped tissue (inner radius 0.2 mm, outer radius 0.65 mm.) surrounded the optic disk was analyzed. Retinal landmarks were identified by naked eyes on 100 b-scans covering the entire tissue blocks, the average inner retinal thicknesses were computed using custom software^[2].

RESULTS

Amplitude of Pattern Electroretinogram on Brain-derived Neurotrophic Factor Administrated Group Decreases Insignificantly Compared with Pattern Electroretinogram Baseline

Among 7 mice, the PERG amplitude baseline was (mean±SE) 13.91±0.69 µV, the PERG amplitude of BDNF administrated group was (mean±SE) 12.28±0.82 µV. Compared with baseline PERG, BDNF administrated group PERG amplitude decreased (11.7%) insignificantly (t-test, P=0.095). Previous research^[2] showed that PERG amplitude of no BDNF administrated group decreased significantly immediately after SC lesion (Figure 1).

Figure 1. PERG baseline amplitude was about 13.91 µV, the BDNF administrated group PERG amplitude was around 12.28 µV.

Compared with baseline PERG, BDNF administrated group PERG amplitude decreased insignificantly (P>0.05). Right panel showed that PERG amplitude of no BDNF administrated group decreased significantly immediately after SC lesion (this figure was from Yang et al^[2]). ^aP<0.05.

Dominant Frequency of Baseline Pattern Electroretinogram, Brain-derived Neurotrophic Factor Administrated Group Pattern Electroretinogram Unchanged

Through FFT analysis to baseline PERG and BDNF administrated group PERG, it was found that the dominant frequency of the baseline PERG was 3-4 Hz (average 3.57±0.20, mean±SE), and that of BDNF administrated group PERG was also 3-4 Hz (average 3.29±0.18, mean±SE). The dominant frequency of baseline PERG therefore remained unchanged in the BDNF administrated group (t-test, P=0.36; Figure 2).

Figure 2. The dominant frequency of baseline PERG is on average 3.57 Hz, and that of PERG with BDNF administration is on average 3.29 Hz.

Dominant frequency of baseline PERG, BDNF administrated group PERG stays unchanged (P=0.36).

No Loss of Optical Coherence Tomography Determined Inner Retinal Thickness After Superior Colliculus Lesion

SC lesion induced no changes in retinal thickness are summarized in Figure 3. Inner retina thickness before SC lesion was on average 95.01±0.43 µm (mean±SE). Inner retina thickness immediately after SC lesion was on average 94.25±0.45 µm (mean±SE). There were no significant changes in the inner retina before and immediately after SC lesion (t-test, P=0.27; Figure 3).

Figure 3. The distance between the two red allows in upper panel is defined as inner retina thickness.

This thickness did not change before and immediately after SC lesions. P=0.27.

DISCUSSION

PERG is a particular kind of electroretinogram (ERG) recorded in response to contrast modulation of special visual stimuli (for example, checkerboards or contrast-reversing gratings) whose characteristics are basically different from those of the traditional flash light induced ERG. According to previous data, PERG is related to the function of retinal ganglion cells and flash ERG is related to function of photoreceptors^[1]. A good many mouse models of optic nerve disease based on ERG researches such as glaucoma, optic neuritis have been developed to probe specific pathological mechanisms and the effect of neuroprotective therapies. The use of these models may be greatly promoted by the availability of non-invasive approaches which can monitor RGC function longitudinally such as the PERG^[1]–^[2]. In the current research, PERG was employed as an effective modality to assess the visual function of mice under both normal and experimental conditions.

According to anatomical data in mice, more than 95% of retina's RGC output goes to the contralateral SC^[2]. Our previous research revealed that after SC lesion, PERG of the eye contralateral to SC lesion decreased immediately (within 1h after SC lesion) and significantly^[2]. But, interestingly, administration BDNF in retina before the SC lesion showed that PERG measurements were within the normal range. PERG did not decrease significantly after SC lesions. Results suggest that BDNF provides effective protection in the short term to retinal RGCs under stress or sufferring damage. It was shown that BDNF binds at least two receptors on the surface of neurons that are capable of responding to this growth factor via the high-affinity tropomyosin-related kinase B (TrkB) receptor and the low-affinity receptor-p75NTR^[21]–^[23]. BDNF may also modulate the activity of various neurotransmitter receptors, including the alpha-7 nicotinic receptor^[22],^[24].

The dominant frequency is a stable characteristic of normal PERG under some conditions. It may change in animals with retinal neuropathy or if there is some other interrupted signal coming from damaged tissue. Before and after SC lesion in short time, the dominant frequency of the PERGs did not change. The lack of change in the dominant frequency that we find following BDNF injection also reinforces evidence that following removal of the normal source of BDNF from the SC, administering BDNF protects the retina at least in the short interval of time we examined. OCT data provides evidence that before and immediately after SC lesion, inner layer of retina was still of normal thickness. Previous data has shown that ten days after SC lesion, the inner retinal thickness is significantly smaller than normal and RGC layer also shrinks^[2].

The primary goal of this research is to test both the function and structure changes of retina (such as PERG and inner retina thickness) before and after SC lesion and the efficacy of BDNF's protection to retina. From the results of this research, it can be concluded that ex vivo of BDNF can be a useful therapeutic neurotrophic factor in treatment of some neurodegeneration diseases, such as glaucoma, age related macular degeneration.