Recovery of visual evoked potentials after regeneration of cut retinal ...

303

Restorative Neurology and Neuroscience 23 (2005) 303–312 IOS Press

Recovery of visual evoked potentials after regeneration of cut retinal ganglion cell axons within the ascending visual pathway in adult rats Peter Heiduschka 1, Dietmar Fischer2 and Solon Thanos ∗ Department of Experimental Ophthalmology, University Eye Hospital M u¨ nster, Domagkstr. 15, D – 48149 Münster, Germany

Received 9 March 2005 Revised 13 June 2005 Accepted 13 June 2005

Abstract. Purpose: Following optic nerve damage, retinal ganglion cells (RGCs) fail to regenerate their axons and soon undergo apoptosis. However, many RGCs survive axotomy and regenerate lengthy axons after a lens injury (LI). If the cut optic nerve is re-sutured, RGC axons grow into the distal part of the optic nerve and reach their natural targets within the thalamus and midbrain. In this study, we check time-dependence and extent of restoration of flash visual evoked potentials (FVEPs) to examine the functional relevance of the regenerated retinogeniculate pathway. Methods: The optic nerve in adult rats was cut and re-sutured. The lens was injured transsclerally using a pointed glass capillary. FVEPs were measured starting at the time point of surgery, and then repeatedly up to an age of several months. Results: Detectable FVEPs appeared approximately ten weeks after the surgery, and their amplitudes increased during the next months to reach eventually 15–40% of their values before surgery. Conclusions: Partial restoration of FVEPs indicates that some regenerating RGC axons have “bridged” the distance between the eye and the central targets forming a functional re-connection of the corresponding RGC with thalamic target neurones to elicit recordable activation of the visual cortex. Keywords: Rat retina, axotomy, optic nerve regeneration, lens injury, visual evoked potentials, functional recovery

1. Introduction

1 Present address: Novartis Pharma AG, NIBR, WKL-127.1.09, Klybeckstr., Gate 15, CH-4002 Basel, Switzerland, E-mail: peter. [email protected]. 2 Present address: University of Ulm, Experimental Neurology, Festpunkt O25, room 541, Albert-Einstein-Allee 11, D-89081 Ulm, Germany. E-mail: [email protected]. ∗ Corresponding author: S. Thanos, Department of Experimental Ophthalmology, University Eye Hospital Münster, Domagkstraße 15, D – 48149 Münster, Germany. Tel.: +49 251 8356915; Fax: +49 251 8356916; E-mail: [email protected].

Neurones of the central nervous system (CNS) fail to regenerate their axons spontaneously after a lesion, and their parent cell bodies degenerate retrogradely. In analogy, retinal ganglion cells (RGC) are not able to regenerate affected axons after the optic nerve is injured, and undergo apoptotic cell death [3,13,21]. The major obstacles for the successful regeneration of cut RGC axons are the inhibitory properties of the optic nerve myelin, the gliotic scar at the site of lesion, and the insufficient survival ability of the le-

0922-6028/05/$17.00  2005 – IOS Press and the authors. All rights reserved

304

P. Heiduschka et al. / Recovery of visual evoked potentials after regeneration of cut retinal ganglion cell axons

sioned RGC [1,2,4–6,24,25]. However, rescue of lesioned RGC and even regeneration of their axons can be achieved by suitable pharmacological and/or surgical measures, most important of them being grafting peripheral nerve at the optic nerve stump (for a review, see [14]). Mansour-Robaey et al. [19] reported neuroprotective effects in rat eyes being associated with an intravitreal injection of vehicle solution. Similarly, RusselakisCarneiro et al. [22] reported about neuroprotective effects on damaged RGC of saline injections into the cat eye after optic nerve lesion. These effects are probably due to an inflammatory reaction in the vitreous body, which has been recently reported to exert strong neuroprotective effects on axotomised RGCs [10,16]. In these studies, an inflammatory reaction in the eye causing an activation of macrophages was initiated by a lens injury (LI). In addition to neuroprotective effects, a LI stimulates the regeneration program of cut RGC axons in vitro and in vivo [10,12,16,29] into a peripheral nerve graft. After a LI, approximately one third of all RGC grow their axons over several millimetres into the distal part of the crushed optic nerve [12,16,29]. In the meantime, LI has been used as a regeneration stimulus in a study about protein tyrosine phosphatase expression after optic nerve crush [18]. A review about the influence of lenticular factors on the post-traumatic fate of neurones was recently given [26]. We have recently shown that RGCs are able to regenerate their axons over much longer distances if the optic nerve was clearly cut and re-sutured rather than crushed. Several of the regenerating axons reached the pre-chiasmatic regions three weeks after surgery, and their natural targets within the thalamus and midbrain after about five weeks post-surgery [11]. In order to examine whether axons reaching the thalamic relay nuclei were able to re-establish a functional connection, we measured flash visual evoked potentials (FVEPs) in some of the operated animals three months after surgery. Indeed, a FVEP could be recorded in animals with completely cut and re-sutured optic nerves and simultaneous lens injury, whereas no response to optic stimulus could be detected in animals without lens injury [11]. It was now of interest to examine long-term re-establishment of the functional connection. To study these connections, rats received optic nerve cut and re-suture surgery concomitantly with lens-puncture. The FVEPs were monitored over several moths post-surgery.

2. Materials and methods 2.1. Surgical procedures for optic nerve regeneration Experiments were performed with male and female adult rats of the Sprague-Dawley strain, weighing 200– 230 g and aged between 8 and 10 weeks. All experiments were carried out in accordance with the ARVO Statement for “The Use of Animals in Ophthalmic and Vision Research”, and were approved by the local Committee for Animal Care. For microsurgery, the animals received an intraperitoneal injection of a mixture of 0.2–0.3 ml ketamine sulphate (50–60 mg/kg, ParkeDavis) and 0.1 ml xylazine (10–15 mg/kg, Medistar) per 200 g body weight. For anaesthesia during electrophysiology, the animals received an intraperitoneal injection of 0.42 mg chloral hydrate/kg body weight, diluted in physiological saline. To axotomise RGCs, the head of the anaesthetised animal was positioned and fixed in a self-constructed head-holder. The left optic nerve was surgically exposed in its intraorbital segment that spans 2–3 mm beyond the eye cup, as has been described previously [27]. Part of the lacrimal gland and the dorsal extraocular muscles were removed to gain access to the optic nerve. Two threads, positioned nasally and laterally in the meninges of the nerve (10.0 silk, Ethicon, Hamburg, Germany), held it in place and prevented traction when the nerve was cut. An incision was made in the meninges perpendicular to the axonal orientation, extending over one-third of the dorsal optic nerve. The optic nerve was completely cut with microscissors at a distance of 0.5 mm behind the eye. The cut ends of the distal and proximal nerve stumps were realigned by tightening the two threads together. This method ensured that no gap remained between the nerve stumps, and that the nerve fascicles were topographically aligned. The entire procedure was carried out within 30–40 min. All microsurgery was performed with the aid of a surgical microscope (Zeiss, OPMI 19). The fundus of the retina was examined immediately after surgery to ensure that the vasculature was intact. In order to study beneficial effects of lens injury on axonal growth, a glass capillary with a tip diameter of 20 µm was inserted through the sclera and retina about 2 mm above the optic nerve in the animals of the experimental group [10]. The capillary penetrated the vitreous body perpendicular to the optic nerve, injuring the lens. Injury of the lens was confirmed by transcorneal visualisation immediately after surgery and by monitoring the slow opacification, which developed in the


days and weeks after lesion was made. No neurotrophic drugs or other substances were injected into the vitreous body or into the lens in this study. 2.2. Functional assessment of visual performance To assess restoration of visual function, flash visualevoked potentials (FVEP) were recorded repeatedly beginning after surgery for four to six months in a grounded Faraday box. Two stainless-steel screws with a shaft diameter of 1.17 mm (Fine Science Tools, Heidelberg, Germany) served as measuring electrodes and were inserted symmetrically through the skull into the visual cortices (2.5 mm laterally to the midline, 2 mm anterior to the lambda), penetrating the cortex approximately 0.5 mm. The electrodes were connected directly to shielded co-axial cables leading to an amplifier. The silver wire reference electrode was placed subcutaneously onto the skull anterior to the eyes. The optical stimulus was a flash of light (250 lx), produced by a fast rechargeable photographic flash (Unomat, Reutlingen, Germany) and covering almost the whole visual field except the lower ventral part. Signals were amplified with an ISO-DAM8 multi-channel amplifier (WPI, Sarasota, USA) at frequencies between 0.1 and 500 Hz and at an amplification factor of 1000. The FVEP was recorded directly after the flash for 200 ms at a sampling rate of 2000 Hz. Responses to ten flashes with intervals of 60 s were averaged. Averaging was performed by a computer coupled to the amplifier, with the MP 100 data acquisition system (WPI, Sarasota, USA). The MP 100 system provided the signal that triggered the relay for the flash through an analogue output. Averaged signals of each group were compared with respect to both latency after flash onset and amplitude. In order to assure that measured signals really represent cortical response to optical stimuli instead of some straying electrical pulses due to flash light discharge, additional measurements were performed in each session where both the flash light and the animal were covered with black opaque cardboard thus suppressing optical stimulus. In none case, any influence of electrical discharge pulse on the recorded waveform could be seen. A crucial point is to decide whether a FVEP could be deduced from an actual waveform or not, because several downward deflections could have their origin in spontaneous cortical activity. We therefore confined the search for a possible FVEP to a latency between 20 and 50 ms. It was important to see a relatively

305

steep slope at the beginning of the deflection in order to consider it a FVEP to be. The amplitude was then determined by setting the baseline just at the beginning of the deflection and calculating the difference between the baseline and the deepest point of the waveform’s deflection. Several experimental groups were established. Nonaxotomised rats, rats with axotomy but without resuturing, and rats with axotomy and re-suturing, but without lens injury, served as controls. In animals with axotomy, re-suturing and lens injury, the contralateral optic nerve was either left intact or cut to ensure that any input from the other eye did not interfere with the recording.

3. Results 3.1. Control rats without surgery The first series of experiments addressed the question whether chronically inserted stainless steel screws would be suitable for long term measurements. For this purpose, FVEPs were recorded in non-operated control animals (n = 2). Figure 1(a) shows a typical series of FVEPs recordings in a control animal, demonstrating that FVEPs can reliably be recorded with the applied method over a long period of time exceeding one year. Although the shapes of the curves showed some degree of variations, the general shape remained similar, and the amplitudes remained in the same range (Fig. 1b). Different amplitudes were recorded between the left and the right cortex in both animals. However, the different amplitudes remained stable and were fairly reproducible over the time of experiment. 3.2. Control rats with left optic nerve cut As expected, the FVEPs recorded on the contralateral side were significantly reduced when one of the two optic nerves was cut (n = 2). However, they did not disappear completely, because of the ipsilateral projection of few axons from the remaining intact eye (approximately 5%). Figure 2(a) shows one example of a series of FVEP measurements. The FVEP recorded in the right visual cortex decreased clearly after cutting the left optic nerve. The amplitudes remained in the same range over four months (Fig. 2b). Similarly, FVEP amplitudes markedly decreased in another animal after cut of the left optic nerve (Fig. 2c).

306

a


left cortex - right eye

100 µV 50 ms

right cortex - left eye 0

6

9

13

18

25

35

51

62

time after electrode implantation [weeks] left cortex - right eye right cortex - left eye

amplitude [%]

b

200 180 160

160

140

140

120

120

100

100

80

80

60

60

40

40

20

20

0

c

amplitude [%]

0 0

4

8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 time [weeks]

0

4

8

12

16

20

time [weeks]

Fig. 1. a) Repeated recordings of FVEP in a control rat without further surgery over a time scale of 14 months. Dotted lines represent recordings with covered flash light as described in Materials and Methods section. b) Change of FVEP amplitude during time of experiment deduced from the curves shown in 1a. c) Second example of FVEP recordings in a non-operated animal.

3.3. Control rats with cut and re-sutured left optic nerve without LI In a further group, the left optic nerve was cut and subsequently re-sutured (n = 3). In this group, LI was not performed in order to exclude that FVEP recovery occurs without support of regeneration. To avoid a contribution of the ipsilaterally projecting fibres from the remaining intact right optic nerve (Figs 2, 3a, b), it was also cut during a second surgery. As shown in Fig. 3(a,b), both FVEPs disappeared immediately after cut of the right optic nerve thus indicating that the residual rightside FVEP was transmitted through the ipsilateral retinofugal projection of the right eye. If both optic nerves of the animal were cut at the beginning of the experiment, no VEP could be recorded during the following seven months (Fig. 3c). Repeated recordings with covered flash light source did not yield

any response, too (Fig. 3c), thus indicating that the responses recorded throughout the control groups were flash-light-evoked. 3.4. FVEP recovery after re-suturing and LI This group of animals was treated like the control group of Fig. 3, but received lens-puncture in addition to surgery to stimulate axonal growth within the optic nerve (n = 7). Figure 4 shows a series of VEP measurements obtained in a rat of this group with cut and re-sutured left optic nerve and injured lens. Schematic drawings on top of Fig. 4 in correlation with the recordings depict the stages of the experiment. As expected, intact FVEPs could be recorded bilaterally before surgery (Fig. 4a). After surgery, i.e. cut and re-suturing of the left optic nerve and LI, the amplitude measured in the affected


a

307


100 µV 50 ms

0

5

11

18

time after left optic nerve cut [weeks] left cortex - right eye right cortex - left eye

right cortex - left eye

b

amplitude [%]

amplitude [%]

c

160 140

120

120

100

100

80

80

60

60

40

40

20

20

0

0 0

4

8

12

16

20

time [weeks]

0

4

8

12

16

20

24

time [weeks]

Fig. 2. a) Repeated recordings of FVEP in a control rat before (left curves) and after cutting of the left optic nerve over a time scale of 4 months. b) Change of FVEP amplitude during the time of the experiment. c) Second example of FVEP recordings in of an animal with the left optic nerve cut.

right visual cortex was much smaller (Fig. 4b) and reminiscent of the control group shown in Fig. 3. During the following time, the amplitude measured in the right visual cortex continuously increased between the 9th and 19th weeks after surgery (Fig. 4c). To exclude a contribution of the right optic nerve with its ipsilateral projection, this nerve was then cut in order to eliminate its influence on the FVEP (Fig. 4d). The FVEP was diminished completely in the left visual cortex, and the small evokable response remained intact in the right visual cortex, therefore indicating the contribution of the pretreated left optic nerve (Fig. 4d). This small FVEP remained stable over several weeks and was still recordable 27 weeks after initial surgery (Fig. 4d,e) thus indicating that the regenerating left optic nerve has formed connections.

In the other rats in this group (n = 6), both optic nerves were cut at the same time in order to eliminate the ipsilateral input from the right eye. In these cases, any appearance of a newly emerging FVEP would be attributed to regenerating RGC axons from the treated left optic nerve. Before any surgery, normal FVEPs could be recorded from both sides. The FVEPs disappeared immediately after surgery (Fig. 5a). Three months later, FVEPs were clearly recordable from both cortices (Fig. 5). The amplitude was continually increasing in both sides, but was higher in the right cortex that is contralateral to the treated left eye. Restored FVEP could be recorded bilaterally in four out of six operated animals. The amplitudes in the right visual cortex varied between 15% and 40% of the values measured before surgery,

308


a


100 µV 50 ms


9

20

32

time after surgery [weeks] left cortex - right eye right cortex - left eye

amplitude [%]

b

140 120 100 80 60 40 20 0 0

4

8

12

16

20

24

28

32

time [weeks]


c 100 µV 50 ms


8

16

31

time after surgery [weeks] Fig. 3. a) Repeated recordings of FVEP in a rat after cut and re-suturing of the left optic nerve without LI over a time scale of 7 months. b) Monitoring of FVEP amplitude during time of experiment showed no changes until the right optic nerve was also cut. c) Repeated recordings of VEP in a rat after cut of both optic nerves and re-suturing of the left optic nerve without LI over a time scale of 7 months. Dotted lines represent recordings with covered flash light.

and the amplitudes measured in the left visual cortex, if visible, were between 20% and 30% of the value before surgery.

4. Discussion The rat optic nerve was used here as a model to study long-term stabilisation of functional contacts between projection neurons of the retina and their central targets. The principal finding of the present study is that once

RGCs are stimulated to grow axons towards the brain, the axons are capable to form stable connections as seen with recording of FVEPs. This finding points to chronic partial repair of the retinofugal projection in rats. Regeneration of axons within the CNS is a particular challenge, because its myelin is thought to inhibit axonal stumps from growth, which leads to growth cone retraction and collapse [9,20,24]. Major inhibitory molecules are the Nogo protein (see [15] for a review), the myelin-associated glycoprotein [17], the


a

b

c

309

d


100 µV 50 ms

right cortex - left eye 0 1

4

9

13

19

20

27

time [weeks]

e

left cortex - right eye right cortex - left eye

amplitude [%] 160 140 120 100 80 60 40 20 0 0

4

8

12

16

20

24

28

time [weeks]

Fig. 4. a-d) Repeated recordings of FVEP in a rat after cut and re-suturing of the left optic nerve with lens injury over a time scale of 7 months. For details, see the text. e) Change of FVEP amplitude during time of experiment.

oligodendrocyte-myelin glycoprotein [28] and proteoglycans of the extracellular matrix [25]. In conformity with the inhibitory environment of the optic nerve [7], no regeneration of RGC axons can be observed except short sproutings, when the optic nerve is cut and resutured without neuroprotection and growth promotion caused by a LI. This is confirmed in the present study by the absence of re-established FVEP in the corresponding group of rats. In contrast to this, a growthpromoting stimulus by a LI allowed the RGC axons to grow in sufficient quantity in order to re-establish functional contacts. This was seen in both groups of rats which had received lens injury and either simultaneous or delayed right-side optic nerve cuts and back suture. Although the mechanisms of lens-stimulated growth of axons still needs to be elucidated, the punctured lens [10,11,16] exerts indirect effects through stimulation of macrophages to an inflammatory reac-

tion, as reviewed in [26], and release of several proteins and other compounds. Among these proteins, a neurite promoting factor of 14 kDa seems to be of special efficacy [29]. Simultaneously to the stimulation of macrophages, also activation of resident microglia, astrocytes and Mu¨ ller cells is found after LI [16]. This indirect stimulation does not exclude, however, direct effects on the retinal cells culminating in helping RGCs to overcome the inhospitable environment of the optic nerve distal to the lesion. Fischer et al. [11] discussed lenticular crystallins as candidate factor to exert neuroprotective and neurite-promoting effects. Irrespective of which of the mechanisms is operating in this model, once axons cross the scar at the site of injury they can traverse over a number of millimetres throughout the ascending visual pathway. Re-establishment of FVEPs occurs with a delay of about 10 weeks after optic nerve surgery. This finding

310

a



100 µV 50 ms

4

10

16

26

37

time after surgery [weeks] right cortex - left eye

b

amplitude [%]

c

100 90 80 70 60 50 40 30 20 10 0 0

4

8

12 16 20 24 28 32 36 40

amplitude [%] 100 90 80 70 60 50 40 30 20 10 0 0

4

8

time [weeks]

d

amplitude [%]

0

4

time [weeks]

e

100 90 80 70 60 50 40 30 20 10 0 8

12 16 20 24 28 32 36

12 16 20 24 28 32 36 40 time [weeks]

amplitude [%] 100 90 80 70 60 50 40 30 20 10 0

left cortex - right eye right cortex - left eye

0

4

8

12 16 20

time [weeks]

Fig. 5. a) Repeated recordings of FVEP in a rat with both optic nerves cut, re-suturing of the left optic nerve and lens injury before (left curves) and after surgery over a time scale of 8 months. Dotted lines represent recordings with covered flash light. b-e) Changes of FVEP amplitudes during time of experiment in four different animals.

is consistent with former studies that showed that regenerating axons within peripheral nerve grafts need about 5 to 6 weeks to bridge the distance between eye and visual centres (see [14] for review). In the model used in our study, retinal axons arrive at the thalamus and midbrain about 5 weeks after optic nerve surgery as shown by anterograde axonal tracing [11]. The time elapsed

between fibre arrival and recovery of FVEP may be the time essential for plastic processes of synapse formation which are responsible for eliciting FVEPs. Reestablishing FVEPs is attributed to a sufficient number of axons regenerating throughout the pathway. Although not quantified in the present study, the fibres represent a sub-population of cut axons, as also seen in


the reduced amplitudes of FVEPs. The relatively stable amplitudes over months indicate, however, that no or little regressive events take place after establishment of connections. Finally, direction of the axons towards the contralateral brain (group with delayed cut of the right optic nerve) or towards both sides of the brain (group with simultaneous cut of both nerves) resulted in similar bilateral amplitudes. This finding indicates the ability of axons to innervate both sides of the brain when arriving there. In fact, this finding is also in line with the results of anterograde staining of the RGC, where several RGC were found to project their regenerating axons to the ipsilateral SC [11]. The extent of recovery varied from 15% to 40% of initial amplitude and is explained by the fact that only a portion of RGCs can regenerate axons. From various studies in the past, it is generally accepted that about 10% of the retinofugal axons are able to elicit FVEPs [8,11,23], and this is likely to be the order of magnitude in the present study. Recovered FVEP could only be recorded in four of six animals demonstrating that appropriate adaptation of the cut ends of the optic nerve may have failed in those two cases where no FVEP could be recorded. The novelty of the present study is the long-term improvement of FVEPs elicited by regenerating axons within the rat retinofugal pathway. Although the amplitudes of FVEPs are reduced, they seem to be stable over several months, indicating permanent synaptic contacts between axons arriving and local retinoreceptive neurons.

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

Acknowledgements This study was supported by the grant no. 01 Ko 9805/7 from the German Federal Ministry for Education and Research (BMBF) as well as by a grant from the BMBF (no. 01 KS 9604/0) the German Research Foundation (DFG) (grant Th 386/10-2) and the Interdisciplinary Centre of Clinical Research M u¨ nster (IZKF Projects No. F5 and F7).

[14] [15]

[16]

[17]

[18]

References [1]

[2]

A.J. Aguayo, D.B. Clarke, T.N. Jelsma, P. Kittlerova, H.C. Friedman and G.M. Bray, Effects of neurotrophins on the survival and regrowth of injured retinal neurons, Ciba Found. Symp. 196 (1994), 135–144. C. Bandtlow, T. Zachleder and M.E. Schwab, Oligodendrocytes arrest neurite growth by contact inhibition, J. Neurosci. 10 (1990), 3837–3848.

[19]

311

M. Berkelaar, D.B. Clarke, Y.C. Wang, G.M. Bray and A.J. Aguayo, Axotomy results in delayed death and apoptosis of retinal ganglion cells in adult rats, J. Neurosci. 14 (1994), 4368–4374. M. Berry, Post-injury myelin-breakdown products inhibit axonal growth; an hypothesis to explain the failure of axonal regeneration in the mammalian central nervous system, Bibl. Anat. 23 (1982), 1–11. M. Berry, J. Carlile and A. Hunter, Peripheral nerve explants grafted into the vitreous body of the eye promote the regeneration of retinal ganglion cell axons severed in the optic nerve, J. Neurocytol. 25 (1996), 147–170. M. Berry, J. Carlile, A. Hunter, W.L. Tsang, P. Rosutrel and J. Sievers, Optic nerve regeneration after intravitreal peripheral nerve implants: trajectories of axons regrowing through the optic chiasm into the optic tracts, J. Neurocytol. 28 (1999), 721–741. G. Campbell, J.K. Holt, H.R. Shotton, P.N. Anderson, S. Bavetta and A.R. Lieberman, Spontaneous axonal regeneration after optic nerve injury in adult rat, NeuroReport 10 (1999), 3955–3960. S. Eitan, A. Solomon, V. Lavie, E. Yoles, D.L. Hirschberg, M. Belkin and M. Schwartz, Recovery of visual response of injured adult rat optic nerves treated with transglutaminase, Science 264 (1994), 1764–1768. J.W. Fawcett, J. Rokos and I. Bakst, Oligodendrocytes repel axons and cause axonal growth cone collapse, J. Cell. Sci. 92(Pt. 1) (1989), 93–100. D. Fischer, M. Pavlidis and S. Thanos, Cataractogenic lens injury prevents traumatic ganglion cell death and promotes axonal regeneration both in vivo and in culture, Invest. Ophthalmol. Vis. Sci. 41 (2000), 3943–3954. D. Fischer, P. Heiduschka and S. Thanos, Lens-injurystimulated axonal regeneration throughout the optic pathway of adult rats, Exp. Neurol. 172 (2001), 257–272. D. Fischer, Z. He and L.I. Benowitz, Counteracting the Nogo receptor enhances optic nerve regeneration if retinal ganglion cells are in an active growth state, J. Neurosci. 24 (2004), 1646–1651. E. Garcia-Valenzuela, W. Gorczyca, Z. Darzynkiewicz and S.C. Sharma, Apoptosis in adult retinal ganglion cells after axotomy, J. Neurobiol. 25 (1994), 431–438. P. Heiduschka and S. Thanos, Restoration of the visual pathway, Progr. Ret. Eye Res. 19 (2000), 577–606. D. Hunt, R.S. Coffin and P.N. Anderson, The Nogo receptor, its ligands and axonal regeneration in the spinal cord; a review, J. Neurocytol. 31 (2002), 93–120. S. Leon, Y. Yugin, J. Nguen, N. Irwin and L.I. Benowitz, Lens injury stimulates axon regeneration in the mature rat optic nerve, J. Neurosci. 20 (2000), 4615–4626. M. Li, A. Shibata, C. Li, P.E. Braun, L. McKerracher, J. Roder, S.B. Kater and S. David, Myelin-associated glycoprotein inhibits neurite/axon growth and causes growth cone collapse, J. Neurosci. Res. 46 (1996), 404–414. B. Lorber, M. Berry, W. Hendriks, J. den Hertog, R. Pulido and A. Logan, Stimulated regeneration of the crushed adult rat optic nerve correlates with attenuated expression of the protein tyrosine phosphatases RPTP(, STEP, and LAR, Mol. Cell. Neurosci. 27 (2004), 404–416. S. Mansour-Robaey, D.B. Clarke, Y.-C. Wang, G.M. Bray and A.J. Aguayo, Effects of ocular injury and administration of brain-derived neurotrophic factor on survival and regrowth of axotomized retinal ganglion cells, Proc. Nat. Acad. Sci. USA 91 (1994), 1632–1636.

312 [20]

[21]

[22]

[23]

[24] [25]

P. Heiduschka et al. / Recovery of visual evoked potentials after regeneration of cut retinal ganglion cell axons S.J. Moorman and R.M. Gould, Differentiating oligodendrocytes inhibit neuronal growth cone motility in different ways, J. Neurosci. Res. 50 (1997), 791–797. H.A. Quigley, R.W. Nickells, L.A. Kerrigan, M.E. Pease, D.J. Thibault and D.J. Zack, Retinal ganglion cell death in experimental glaucoma and after axotomy occurs by apoptosis, Invest. Ophthalmol. Vis. Sci. 36 (1995), 774–786. M. Russelakis-Carneiro, L.C. Silveira and V.H. Perry, Factors affecting the survival of cat retinal ganglion cells after optic nerve injury, J. Neurocytol. 25 (1996), 393–402. J. Sautter and B.A. Sabel, Recovery of brightness discrimination in adult rats despite progressive loss of retrogradely labelled retinal ganglion cells after controlled optic nerve crush, Eur. J. Neurosci. 5 (1993), 680–690. M.E. Schwab, J.P. Kapfhammer and C.E. Bandlow, Inhibitors of neurite growth, Ann. Rev. Neurosci. 16 (1993), 565–595. C.C. Stichel and H.W. Müller, The CNS lesion scar: New

[26]

[27]

[28]

[29]

vistas on an old regeneration barrier, Cell Tissue Res 294 (1998), 1–9. T. Stupp and S. Thanos, Can lenticular factors improve the posttrauma fate of neurons? Progr. Ret. Eye Res. 24 (2005), 241–257. M. Vidal-Sanz, G.M. Bray, M. Villegas-Pérez, S. Thanos and A.J. Aguayo, Axonal regeneration and synapse formation in the superior colliculus by retinal ganglion cells in the adult rat, J. Neurosci. 7 (1987), 2894–2909. K.C. Wang, V. Koprivica, J.A. Kim, R. Sivasankaran, Y. Guo, R.L. Neve and Z. He, Oligodendrocyte-myelin glycoprotein is a Nogo receptor ligand that inhibits neurite outgrowth, Nature 417 (2002), 941–944. Y. Yin, Q. Cui, Y. Li, N. Irwin, D. Fischer, A.R. Harvey and L.I. Benowitz, Macrophage-derived factors stimulate optic nerve regeneration, J. Neurosci. 23 (2003), 2284–2293.