Upload
vuongtuong
View
218
Download
1
Embed Size (px)
Citation preview
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 1
NF-B determines axonal re- and degeneration by cell-specific balance of RelA and p50 1
subunits in the adult CNS 2
3
Ronny Haenold1, Falk Weih1, Karl-Heinz Herrmann2, Karl-Friedrich Schmidt3,I, Katja Krempler4, 4
Christian Engelmann1, Klaus-Armin Nave5, Jürgen R. Reichenbach2, Sigrid Löwel3,I, Otto W. Witte4, and 5
Alexandra Kretz4 6
7
1 Leibniz Institute for Age Research Fritz Lipmann Institute, Beutenbergstr. 11, 07745 Jena, Germany 8
2 Friedrich Schiller University of Jena Medical School, Institute of Diagnostic and Interventional Radiology, 9
Medical Physics Group, Philosophenweg 3, 07743 Jena, Germany 10
3 Friedrich Schiller University of Jena, Institute of General Zoology and Animal Physiology, Erbertstr. 1, 07743 11
Jena, Germany 12
4 Hans Berger Department of Neurology, Jena University Hospital, Erlanger Allee 101, 07747 Jena, Germany 13
5 Max Planck Institute for Experimental Medicine, Department of Neurogenetics, Hermann-Rein-Str. 3, 37075 14
Göttingen, Germany 15
16
I Present address of S. Löwel and K.-F. Schmidt: Georg August University, Bernstein Focus Neurotechnology 17
(BFNT) and Johann Friedrich Blumenbach Institute for Zoology and Anthropology, Berliner Str. 28, 37073 18
Göttingen, Germany. 19
20
Corresponding author: Ronny Haenold, Leibniz Institute for Age Research Fritz Lipmann Institute, 21
Beutenbergstr. 11, 07745 Jena, Germany, E-mail: [email protected], Phone: +49 3641 656050, Fax: +49 3641 22
656335 23
24
Running title: Adult axoneogenesis by cellular balance of RelA and p50 25
Keywords: anaphase-promoting complex, axonal regeneration, Cdh1, manganese-enhanced MRI (MEMRI), NF-26
B, p50, RelA/p65, Wallerian degeneration 27
© 2014. Published by The Company of Biologists Ltd.Jo
urna
l of C
ell S
cien
ceA
ccep
ted
man
uscr
ipt
JCS Advance Online Article. Posted on 23 May 2014
2
Abbreviations: 28
APC anaphase-promoting complex 29
CAII carbonic anhydrase II 30
Cdh1 co-activator protein for APC 31
CNR contrast-to-noise ratio 32
CNS central nervous system 33
CRALBP cellular retinaldehyde-binding protein 34
CSPG chrondroitin sulfate proteoglycans 35
CTX cholera toxin B subunit 36
EMI1 early mitotic inhibitor 1 37
β-gal β-galactosidase 38
Id2 inhibitor of DNA binding 2 39
i.vit. intravitreally 40
MBP myelin basic protein 41
MEMRI manganese-enhanced MRI 42
MG Müller glia 43
MRI magnetic resonance imaging 44
NF-B nuclear factor-B 45
NLS nuclear localization signal 46
ODC oligodendrocyte 47
ON optic nerve 48
ONI optic nerve injury 49
RGC retinal ganglion cell 50
ROI region of interest 51
SCI spinal cord injury 52
TUJ-1 β-III tubulin 53
54
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 3
Summary 55
NF-B is dually involved in neurogenesis and brain pathology. Here, we addressed its role in 56
adult axoneogenesis by generating mutations of RelA and p50 heterodimers of canonical NF-B. 57
In addition to activation in astrocytes, optic nerve axonotmesis caused a hitherto unrecognized 58
RelA induction in growth inhibitory oligodendrocytes. Intraretinally, RelA was induced in 59
severed retinal ganglion cells and inferred in bystander Muller glia. Cell type-specific deletion of 60
transactivating RelA in neurons and/or macroglia considerably stimulated axonal regeneration in 61
a distinct and synergistic pattern. In contrast, deletion of the p50 suppressor subunit promoted 62
spontaneous and post-injury Wallerian degeneration. Growth effects mediated by RelA deletion 63
paralleled a downregulation of growth inhibitory Cdh1 and upregulation of the endogenous Cdh1 64
suppressor EMI1. Pro-degenerative loss of p50, however, stabilized retinal Cdh1. In vitro, RelA 65
deletion elicited opposing, pro-regenerative shifts in active nuclear and inactive cytoplasmic 66
moieties of Cdh1 and Id2. The involvement of NF-B and cell cycle regulators such as Cdh1 in 67
regenerative processes of non-replicative neurons presents novel options regarding how 68
molecular reprograming might be executed to stimulate adult axoneogenesis and treat CNS 69
axonopathies. 70
71
72
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
4
Introduction 73
The transcription factor nuclear factor-B (NF-B) is ubiquitously expressed and is critical for 74
various neuropathologies (Kaltschmidt and Kaltschmidt, 2009). In the nervous system, it 75
establishes its role by subunit- and cell type-specific activation and post-translational 76
modifications on the RelA and p50 subunits. Moreover, tissue maturation, its activation in 77
peripheral (PNS) versus central (CNS) nervous system, and the context of injury influence 78
cellular NF-B functions. Studies on neonatal sympathetic and sensory neurons have indicated 79
that NF-B RelA and p50 can either promote or inhibit axogenesis during post-natal 80
development (Gutierrez and Davies, 2011; Gutierrez et al., 2008). Overexpression of a dominant-81
negative form of the NF-B inhibitor IBα in astrocytes indeed limited loco-regional damage 82
after spinal cord injury (SCI) and further stimulated axon sprouting and functional recovery 83
(Brambilla et al., 2005; 2009). However, the significance of individual NF-B subunits, their 84
activation in separate cell types, and their impact on axonal regeneration and Wallerian 85
degeneration currently remain undefined. Intriguingly, a putative involvement of NF-B in the 86
repulsive feature of white matter substances (Chen et al., 2000) and the anti-growth program 87
exerted by oligodendrocytes (ODC) has not yet been investigated. More importantly, stimulus-88
dependent axo-nuclear transport of NF-B as demonstrated by FRAP analysis of cultivated 89
hippocampal neurons using a RelA-GFP reporter (Meffert et al., 2003) may trigger a cell-90
intrinsic pro-/anti-regenerative program in axons themselves. 91
NF-B-induced gene expression is modulated by nuclear translocation of either complexes 92
containing transcriptionally active RelA or homodimers of the transcriptionally inactive p50 93
subunit. While interference with the upstream kinases IKKα/β/γ or overexpression of IBα 94
results in inhibition of any dimer of the classical NF-B cascade, subunit-specific knockouts 95
shift the balance between the individual moieties activated. Thus, ablation of either RelA or p50 96
can propagate dual or even opposing effects, as exemplified for post-ischemic infarct volumes in 97
murine stroke models (Inta et al., 2006; Li et al., 2008; Zhang et al., 2005). 98
In the present study, we modulated the balance between RelA and p50 subunits specifically in 99
neurons and macroglia of mutant mice using the Cre/loxP system, or by insertion of a pGK-neo 100
cassette, and investigated cell type-specific roles of individual NF-B subunits in adult 101
axoneogenesis. We show that selective suppression of RelA induction in neurons and macroglia 102
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 5
(RelACNSKO) or in oligodendrocytes (RelAODCKO) and astrocytes (RelAASTKO) alone differentially 103
and synergistically increased axonal regeneration. In contrast, upregulation of NF-B activity by 104
ubiquitous p50 deficiency prompted Wallerian degeneration. The divergent effects of RelA and 105
p50 on axon integrity were reflected in a subunit-specific regulation of the ubiquitin E3 ligase 106
adaptor protein Cdh1, which was suppressed under axoneogenesis and upregulated under 107
Wallerian degeneration. On the subcellular level, RelA-deficient neuronal cultures exhibited an 108
inactivating shift in Cdh1 protein from the nucleus to the cytoplasm and a reciprocal nuclear 109
accumulation of the pro-regenerative Cdh1 substrate Id2 (inhibitor of DNA binding 2). In 110
summary, balanced levels of the NF-B subunits RelA and p50 and their orchestration with cell 111
cycle regulators such as Cdh1 may contribute to define the growth potential of injured CNS fiber 112
networks. 113
114
Results 115
Visual system development is not impaired in RelACNSKO and p50KO mice 116
Conditional neuro-ectodermal deletion of RelA was achieved by nestin-Cre-based recombination 117
of homozygous floxed relA alleles (RelACNSKO). In these RelACNSKO mice, robust reduction of 118
RelA protein in the retina and optic nerve (ON) was confirmed by immunoblotting (data not 119
shown). Reduced canonical NF-B activity in the CNS of RelACNSKO mice, in the absence of a 120
compensatory upregulation of alternative NF-B subunits, has recently been demonstrated by 121
reduced expression of the NF-B target gene IB (Kretz et al., 2013). Histopathological 122
analysis revealed normal layer architecture of transversal H&E-stained retina (Fig. 1A) and an 123
obviously unimpaired relation between retinal ganglion cells (RGC) and Muller glia (MG), as 124
shown in retinal whole mount preparations labeled for the RGC marker β-tubulin (TUJ-1) and 125
the MG marker cellular retinaldehyde-binding protein (CRALBP) (Fig. 1B). Myelin sheath 126
formation and structural myelin basic protein (MBP) content in naïve ON of RelACNSKO mice 127
were inconspicuous as assessed by electron microscopy (Fig. 1C) and immunoblotting (not 128
shown). RGC numbers (P=0.88) in the ganglion cell layer (GCL) and oligodendrocyte (ODC) 129
densities (P=0.43) as well as myelin sheath dimensions (g-ratio; P=0.99) in the ON were 130
indistinguishable from controls (Fig. 1D), thus excluding developmental or apoptotic cell loss by 131
RelA deletion in neurons and macroglia. Further, as shown in Fig. 1E, in vivo parameters for 132
visual acuity and contrast sensitivity were indistinguishable among RelACNSKO, RelAfl/fl;+/+ and 133
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
6
RelA+/+;tg/+ (RelAflox allele-negative, Cre recombinase-positive) mice (P>0.05 for each compared 134
condition), thus assuring that transgene insertion alone or loss of RelA did not impair 135
functionality of the visual projection. 136
Similarly, mice with homozygous deletion of p50 (p50KO) have been described to develop 137
normally (Sha et al., 1995). However, starting at an age of six months, they become susceptible 138
to precocious neural degeneration (Lu et al., 2006) and enhanced apoptotic cell death of the 139
visual and acoustic systems (Lang et al., 2006; Takahashi et al., 2007). 140
141
Constitutive NF-B activity is increased in p50KO mice 142
To explore the consequences of transcriptionally inhibitory p50 deletion on canonical NF-B 143
activity, p50KO mice were crossed with the B-lacZ reporter line. In this line, the lacZ gene 144
encoding for β-galactosidase (β-gal) is driven by multiple B binding sites (Schmidt-Ullrich et 145
al., 1996). Compared to B-lacZ control mice with intact p50 gene expression, double transgenic 146
B-lacZ;p50KO mice revealed a four- to five-fold increase in X-Gal-positive cells in the retina 147
(P<0.02; Fig. 1F) at just three months of age, thus confirming that loss of p50 enhances 148
constitutive NF-B activity. No signal was detected in age-matched p50KO mice devoid of the 149
lacZ gene (Fig. 1F). Thus, deletion of either NF-B subunit in RelACNSKO and p50KO mice can 150
dynamically shift transcription towards suppression or activation. 151
152
Optic nerve injury induces NF-B in different cell types 153
Optic nerve injury (ONI) induced strong and topic NF-B activation in cells of superficial retinal 154
layers and in the epicenter of the squeezed ON, as detected in respective whole mount 155
preparations of B-lacZ reporter mice (X-Gal panels in Fig. 2). Retinal numbers of X-Gal-156
positive cells significantly increased by three- and 18-fold at three and 10 days after ONI, 157
respectively, as compared to naïve specimens (P<0.01; graph in Fig. 2A). As shown by 158
immunohistochemistry, NF-B-dependent induction of β-gal was located to nuclei of TUJ-1-159
positive RGC (Fig. 2B, ONI panels; inset in bottom line). Within the ON, the majority of β-gal-160
positive nuclei could be ascribed to ODC expressing the marker protein carbonic anhydrase II 161
(CAII; Fig. 2D, ONI panels; inset in bottom line). Specificity of the CAII signal was confirmed 162
by lack of immunoreactivity in the non-myelinated head of the naïve ON (Fig. 2D, dotted line in 163
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 7
naïve panel). Thus, ONI strongly induced NF-B activity at the site of axonal damage as well as 164
in the soma of axotomized RGC. 165
Activation of the classical NF-B pathway for target gene expression involves nuclear 166
translocation of RelA, its phosphorylation on serine residues, and exposure of a nuclear 167
localization signal (NLS). Using a Ser536 phospho-specific RelA antibody, we detected 168
substantial RelA phosphorylation in extracts of the ON within 3 h after ONI (Fig. 2C, top; n=3). 169
Notably, pSer536 was not present in the naïve ON despite of its robust RelA content (Fig. 2C 170
top, loading control). The specificity of the pSer536 immunosignal was confirmed by emergence 171
of an equivalent band following TNF-treatment of hippocampal neurons (Fig. 2C, top). Since 172
post-lesional occurrence of RelA phosphorylation was almost abolished in RelACNSKO, but 173
preserved in wild type mice (Fig. 2C, bottom), its activation should derive from axons and/or 174
macroglia of the ON lesion site. 175
NLS-RelA signal in the injured ON, but not in the naïve ON, co-localized with the nuclear β-gal 176
immunoreactivity observed in CAII-positive ODC of B-lacZ reporter mice, indicating 177
perilesional cytoplasm-to-nucleus translocation of RelA in the majority of β-gal-expressing ODC 178
(Fig. 2E, left panel). Such NLS-RelA signal was absent in the nuclei of perilesional CAII-179
positive ODC of RelACNSKO mice (Fig. 2E, right panel), confirming the specificity of RelA 180
activation in ODC. Additionally, we explored nuclear translocation of RelA in the OLN-93 181
oligodendrocytic cell line by confocal laser microscopy. As detected by the activation-specific 182
NLS-RelA antibody, nuclei of non-stimulated control cells were free of RelA signal (Fig. 2F, left 183
panel). Following application of TNF, a prototypical cytokine released during tissue damage, 184
NLS-directed immunoreactivity became strikingly evident in DAPI-positive nuclei of OLN-93 185
cells (Fig. 2F, right, Z-stack). According results were achieved using an antibody directed 186
against the C-terminus of the RelA protein, thus confirming the specificity of the NLS-RelA 187
reactivity (not shown). Collectively, solid steady-state RelA expression in ODC (not shown) 188
together with induced β-gal reactivity in ODC (Fig. 2D) indicated post-lesional RelA activation 189
in ODC. 190
191
RelA and p50 subunit-dependent RGC survival after injury 192
NF-B activation studies performed on B-lacZ;p50KO reporter mice revealed a strong induction 193
of NF-B at the ON injury site, as demonstrated by intense blue color reaction in the X-Gal 194
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
8
assay. Analysis of corresponding retina revealed an unexpected reduction in the total number of 195
X-Gal-positive cells when analyzed as soon as 9 dpi (Fig. 3A; n=3), arguing for enhanced lesion-196
induced RGC death under p50 deficiency. Since NF-B has been shown to either inhibit or 197
propagate cell death, we investigated post-lesional RGC survival in RelACNSKO and p50KO mice 198
on the base of TUJ-1 immuno-reactivity. Neuro-glial deletion of RelA modestly supported RGC 199
rescue after ONI by 7% (P<0.05), compared to controls. In contrast, loss of p50 strongly severed 200
RGC death, thus resulting in a decline by 55% of control levels (Fig. 3B; P<0.01). 201
Histologically, retinal atrophy suggested an overall increased susceptibility of CNS neurons to 202
harmful events in p50KO animals. 203
204
Axonal regeneration is stimulated in RelACNSKO mice, whereas p50 deletion triggers Wallerian 205
degeneration 206
In RelACNSKO mice, anterograde cholera toxin B subunit (CTX) tracing (suppl. Fig. 1) revealed 207
dense bundles of regenerating axons growing into the ON and towards the injury site four weeks 208
after ONI (Fig. 4A, top). Robust trans-lesional axon growth beyond the scar and into the distal 209
ON stump was achieved in the majority of ON specimens (81%; n=16). Although not implicated 210
in software-based quantifications due to their low percentage (<5%, not shown), individual axons 211
spanned a growth distance of 2–3 mm. In control specimens, a substantially lower number of 212
newly generated RGC axons reached the ON head, which almost regularly missed intra- or trans-213
lesional growth aspects (20%; n=10; Fig. 4A, middle). RelA+/+;tg/+ animals similarly showed 214
repressed growth responses (data not shown), thus ruling out a phenotype by the transgene itself. 215
Examination of corresponding retina in each group excluded impaired labeling quality to be the 216
cause of lower ingrowth results in controls; however, in RelACNSKO mice, intraretinal fascicular 217
degeneration appeared attenuated. Generally, our histological post-mortem results correlated well 218
with in vivo analysis of ON regeneration using contrast-enhanced MRI (MEMRI; suppl. results 219
and suppl. Fig. 2). The strong crush technique applied in order to avoid fiber sparing 220
corresponded to the pronounced axonal dieback as well as to the long latency and still rather 221
limited distance of regeneration behind the lamina cribrosa (suppl. Fig. 2). 222
Software-based quantitative analysis of longitudinal ON sections detected on average a five-fold 223
increase in the extent of axonal regeneration in RelACNSKO mice compared to controls (RelAflox: 224
85.91±23.95×103 µm2 versus RelACNSKO: 458.82±66.65×103 µm2; P<0.001; Fig. 4B). Considering 225
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 9
the maximum growth responses in both groups, an even greater stimulation (10-fold) was 226
attained. Further characterization revealed trans-lesional growth areas to be 17-fold (RelAflox: 227
1.67±0.53×103 µm2 versus RelACNSKO: 29.10±5.57×103 µm2; P<0.01; Fig. 4C) and trans-lesional 228
distances to be four-fold (RelAflox: 122.55±0.53 µm versus RelACNSKO: 487.04±59.57 µm; 229
P<0.001; Fig. 4D) increased in RelACNSKO mice compared to controls. Comparable spaces of 230
approximately 240 µm between the optic disc and the lesion site in animals of both groups 231
(RelAflox: mean 223.68 µm versus RelACNSKO: mean 263.57 µm; P=0.07; Fig. 4E) excluded 232
artifacts by biased position of the injury site. Further, co-localization of GAP-43 with the CTX 233
tracer (Fig. 5A) and the emergence of GAP-43-enhanced rudimentary growth cones (Fig. 5B) 234
emphasized the axoneogenetic process in RelACNSKO mice. Intraretinally, RGC elongated MAP2-235
positive dendritic-like arbors, which grew into deeper retinal layers towards the optic disc (Fig. 236
5B, right). Such a growth pattern was exclusive to RelACNSKO mice, thus suggesting fundamental 237
growth reprograming with high axonal and dendritic plasticity at the expense of a defined 238
polarized growth shape. 239
In contrast to the growth improvement in RelACNSKO mice, injured ON of p50KO animals 240
demonstrated negligible regeneration, which was – although not significant – even lower than in 241
controls (Fig. 4A, bottom; 4B). Of note, p50KO animals are known to develop normally and 242
possess physiological RGC counts at young mature ages, however evolve age-dependent 243
alterations in axon-myelin structure (Lu et al., 2006; Takahashi et al., 2007). Now assuming 244
spontaneous destabilization of axonal integrity in p50KO mice, we applied the degeneration 245
marker Fluoro Jade and compared the ON cytoskeleton in naïve and lesioned young RelACNSKO 246
and RelAflox mice with young and aged naïve p50KO mice. RelACNSKO and control animals 247
remained devoid of any staining under naïve conditions and were indistinguishable regarding the 248
extent of fiber degeneration two weeks after axonotmesis (Fig. 5C). At a susceptibility age of 10 249
months, however, naïve p50KO mice displayed drastic fiber disintegration and neurofilament 250
breakdown, which was not discernable in age-matched naïve wild type (not shown) or naïve 251
RelACNSKO mice (Fig. 5C, furthest right). Specificity of the assay was confirmed by negative 252
signals in brain regions adjacent to the intracerebral part of the degenerating optic tract (Fig. 5C, 253
furthest right). Further, toxic axonopathy induced by intraocular TNF application (2 ng), resulted 254
in a similar staining pattern as under axonotmesis four weeks after treatment (not shown). Since 255
RGC quantifications in p50KO mice revealed a drastic decline relative to control levels (Fig. 3B), 256
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
10
growth failure in this case is suggested to be also due to reinforced retinal atrophy including 257
RGC decimation following ONI. These observations highlight the functional polarity of RelA 258
and p50 in axonal de- and regeneration. Whilst p50KO mice display a reduced capacity for repair 259
of endogenous or injury-induced fiber damage, fiber integrity and regeneration are stabilized and 260
promoted in RelACNSKO mice. 261
262
The pro-regenerative effect of RelA deletion is cell type dependent 263
The unexpectedly high expression of RelA in naïve ODC raised the possibility of a hitherto 264
unrecognized growth-relevant RelA function executed by interaction between neurons and 265
growth inhibitory ODC. To explore whether the pro-regenerative effect in RelACNSKO mice was 266
due to deletion of RelA in RGC or rather in ODC, we deleted RelA specifically in ODC by 267
CNP1 promoter-driven Cre recombinase expression (RelAODCKO). Activity of CNP1-Cre 268
recombination in the ON was verified in Rosa26R;CNP1-Cre reporter mice using a colorimetric 269
assay (Fig. 6A). Robust deletion of RelA protein in the ON of RelAODCKO mice was further 270
confirmed by immunoblotting (data not shown). In RelAODCKO mice, outgrowth of newly 271
generated axons four weeks after ONI was significantly enhanced over controls by about nine-272
fold (RelAflox: 17.59±9.00×103 µm2 versus RelAODCKO: 155.53±46.00×103 µm2; P<0.05; Fig. 6B), 273
suggesting that RelA ablation in ODC contributes to neutralize the common white matter-derived 274
growth failure. Since the slow kinetic of Wallerian degeneration in the CNS adds to regenerative 275
failure, we investigated the progress of ONI-dependent myelin degradation by MBP protein 276
detection. In superordinate RelACNSKO constructs implicating RelA loss in ODC, structural MBP 277
decline under Wallerian degeneration was much stronger than in controls, which may explain 278
stimulated axogenesis in RelAODCKO mice (Fig. 6C; n=3). Since in RelAODCKO mice absolute 279
growth dimensions as for expanse and length remained below those in RelACNSKO mice (Fig. 4B), 280
it is suggestive that additional cell populations add to achieve maximum growth responses under 281
neuro-glial RelA deletion. 282
We further investigated axoneogenesis in mice with conditional RelA deletion selectively in 283
astrocytes (RelAASTKO). GLAST-mediated Cre targeting also to retina-specific astrocytes, i.e. 284
MG, was verified using animals with a combined tamoxifen- (TAM) inducible GLAST-285
CreERT2/loxP system (Slezak et al., 2007) and Rosa26-YFP reporter expression (YFP;GLAST-286
CreERT2; Srinivas et al., 2001). In their retinae, obviously all YFP-positive cells identified as 287
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 11
MG by their long processes and strong CRALBP co-expression (not shown) showed a profound 288
increase in YFP reporter signal in response to TAM (Fig. 6D, arrows). Following ONI in 289
RelAASTKO mice, a 19-fold increase in elaborated axons emerged compared to controls (RelAflox: 290
17.59±9.00×103 µm2 versus RelAASTKO: 383.31±57.18×103 µm2; P<0.01; Fig. 6E), which 291
exceeded growth stimulation in RelAODCKO mice but remained below the absolute growth values 292
observed in RelACNSKO mice (Fig. 4B). Since chondroitin sulfate proteoglycans (CSPG) secreted 293
by scar-forming astrocytes are chemically repulsive and mechanically impenetrable growth 294
inhibitors, we investigated post-lesional neurocan secretion in RelACNSKO mice. Neurocan 295
production within the lesion epicenter was less dense and compacted than in controls 10 days 296
(Fig. 6F) and four weeks (not shown) after ONI. Consecutively, axons elaborated from 297
RelACNSKO mice grew straightly into and beyond the scar, whereas in controls most of them were 298
arrested when confronted with the scar region (Fig. 6G, arrows), suggesting RelA-dependent 299
expression of repulsive scar constituents. Importantly, invasion of F4/80+ CD11b+ microglia and 300
macrophages into the injury region appeared indiscernible in either line (not shown). Therefore, 301
a difference mediated by immune-privileged CNS functions seems unlikely. 302
In summary, supreme post-lesional axonal regeneration observed in RelACNSKO mice cannot be 303
explained by RelA deletion in oligodendrocytes (RelAODCKO) or astroglia (RelAASTKO) alone, but 304
argues for further cell type-specific, in particular neuron-intrinsic molecular mechanisms. 305
306
RelA-associated regeneration parallels cell type-specific Cdh1 suppression 307
Therefore, we examined whether the axonal growth regulator Cdh1, which was found expressed 308
in the naïve mature cortex and retina (Fig. 7A), where it localizes to RGC (Fig. 7B), is a target of 309
RelA-mediated growth suppression. This hypothesis was reinforced by the fact that highest 310
levels were found in myelin-enriched projecting fibers of optic and sciatic nerves (Fig. 7A). 311
Following ONI, Cdh1 levels declined by 50% in the retina of regeneration-competent RelACNSKO 312
mice but rose to 182% in non-regenerating RelAflox mice compared to uninjured controls (Fig. 313
7C). In contrast, Cdh1 was upregulated to 220% and 225%, respectively, in young adult p50KO 314
mice after ONI as well as in the spontaneously degenerating retina of aged 10-month-old p50KO 315
mice (Fig. 7C). Assuming that neuronal Cdh1 content defines the threshold for re- and 316
degeneration and that Cdh1 levels are modulated by pro- or anti-regenerative inputs from ODC 317
and astrocytes, Cdh1 levels should be equally, but gradually suppressed in RelAODCKO and 318
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
12
RelAASTKO mice, compared to RelACNSKO mice. Accordingly, Cdh1 levels were found reduced in 319
all growth-stimulating knockout lines, with strongest suppression by 47% in the best 320
regenerating RelACNSKO and RelAASTKO mice, respectively, as compared to 20% reduction in 321
RelAODCKO mice (Fig. 7D). In addition, the physiological inhibitor of Cdh1, EMI1 (early mitotic 322
inhibitor 1), was upregulated (40%) in naïve retina of RelACNSKO mice, whilst downregulated 323
(30%) in aged and non-regenerating p50KO mice (Fig. 7E). ONI further induced a two- to three-324
fold upregulation of EMI1; however, this induction was independent of the genotype (273% in 325
RelAflox versus 221% in RelACNSKO). 326
These data show that Cdh1 levels are negatively correlated with successful regrowth in mature 327
neurons and support the notion that the APCCdh1 cascade is involved in the NF-B-dependent 328
ambivalent regulation of axonal restoration and degradation. 329
330
RelA deletion causes an inactivating shift in subcellular Cdh1 localization 331
We next examined RelA-associated post-translational modifications of Cdh1. Since 332
phosphorylation and subsequent shift of stabilized Cdh1 from the nucleus to the cytoplasm 333
indicates its inactivation (Huynh et al., 2009) we investigated its subcellular distribution in wild 334
type and RelA-deficient hippocampal neurons (HN) during axogenesis (Fig. 7F-H). According to 335
naïve RGC in vivo (Fig. 7B) hippocampal neurons of controls showed a common mixed nuclear 336
and cytoplasmic Cdh1 content (Huynh et al., 2009), however with a nuclear dominance (Fig. 7F-337
H). In the absence of RelA, a two-fold relative increase in cytoplasmic Cdh1 occurred (RelAflox: 338
19.4±1.8% versus RelACNSKO: 40.8±3.7%; P<0.001; Fig. 7F-H). Neurons with a cytoplasmic 339
Cdh1 preponderance above 70% were almost exclusively restricted to cultures of RelACNSKO mice 340
(Fig. 7H, top). Strong staining against the axo-neuronal marker TUJ-1 indicated neuronal 341
viability and axonal vitality in both conditions (Fig. 7F, top) irrespective of their RelA-dependent 342
Cdh1 content. 343
As a further target for RelA-mediated Cdh1 regulation, we investigated protein levels of HLH-344
related, pro-regenerative Id2. Since Id2 is a nuclear target, we looked for a pro-nuclear shift of 345
active Id2 in HN of RelACNSKO with reduced Cdh1 content. Immunocytochemical and subcellular 346
analysis revealed a predominant cytoplasmic localization of Id2 in wild type HN with 347
characteristic accumulation in the axon hill (Fig. 7I, arrowhead and inset) displaying a mean 348
extranuclear signal of 59.2±4.1% (Fig. 7J). In contrast, in RelACNSKO neurons, Id2 349
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 13
immunoreactivity was concentrated to DAPI-positive nuclei (Fig. 7I), whilst the mean 350
percentage of extranuclear signal was reduced to 21.9±2.0% (Fig. 7J). Thereby, average nuclear 351
content was augmented by about 100% (RelAflox: 40.8% versus RelACNSKO: 78.1%; P<0.001), 352
suggesting that reduced Cdh1 activity stabilized nuclear Id2 steady-state levels. Since total 353
protein content may not reflect this regulation (Kim et al., 2006), immunocytochemistry was 354
preferred over immunoblot analysis of whole cell lysates. Further studies will aim to identify 355
growth-responsive target genes regulated by this RelA/p50-EMI1-APCCdh1-Id2 pathway. 356
357
Discussion 358
There is accumulating evidence that RelA is critical for axon formation during embryonic neural 359
development (Gavaldà et al., 2009). In cervical superficial ganglia, enhanced site-specific 360
Serin536 phosphorylation of RelA in the presence of p50 impairs expansion of neurite length and 361
complexity (Gutierrez et al., 2008), whereas RelA suppression by overexpression of either p50 or 362
a dominant-negative IBα super-repressor in newborn hippocampal neurons results in complete 363
growth arrest (Imielski et al., 2012). It has been suggested that modification of both IBα and 364
activated RelA determines a functional switch from growth inhibition to growth promotion 365
(Gavaldà et al., 2009; Gutierrez et al., 2008). Moreover, as recently exemplified for hippocampal 366
neurogenesis, the balance between transactivation-competent and -incompetent NF-B subunits 367
may also be crucial for axogenesis (Imielski et al., 2012). However, such previous experiments 368
were based on in vitro analysis of premature PNS and newborn hippocampal neurons. At present, 369
the relevance of NF-B in the mature post-lesional CNS for structural restoration is undefined 370
and subunit- and cell type-specific features of NF-B activation remain unclear. In this study, we 371
investigated the importance of NF-B on axonal re- and degeneration (i) in mature neurons, (ii) 372
of the CNS, (iii) in vivo, (iv) after axonal injury, (v) in a cell-autonomous manner, and (vi) under 373
consideration of the interdependence between the CNS-dominant NF-B subunits RelA and p50. 374
Apart from induction of p50 and RelA already demonstrated for severed RGC (Choi et al., 1998; 375
Takahashi et al., 2007), ONI elicited a previously unrecognized activation of RelA in macroglia. 376
To the best of our knowledge, this is the first time that ODC as prototypical CNS growth 377
inhibitors have been shown to induce RelA by injury. Thus, RelA in ODC may play a critical 378
role in the cell-autonomous regulation of axonal de- and remyelination following axonal injury. 379
This might depend on the type of injury, since in a cuprizone model, ablation of IKK2 in 380
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
14
astrocytes – but not in ODC – was sufficient to prevent toxic demyelination (Raasch et al., 381
2011). Further, according to Brambilla et al. (2005), we could confirm that reactive astroglia of 382
the scar region are a growth-relevant source of NF-B activation. In their study, NF-B was 383
inhibited in astrocytes by GFAP-dependent overexpression of an IBα super-repressor. Whereas 384
this approach does not discriminate between subunit-specific roles, we now have specified RelA 385
as one of the subunits involved in astroglial scar formation and growth suppression. Thus, in 386
addition to the well-characterized role of NF-B in immune cells for neuron-axonal integrity 387
(Emmanouil et al., 2009), our data emphasize the significance of RelA for axonal renewal by its 388
regulation in CNS-intrinsic neuro-ectodermal neurons and macroglia. Inhibition of such RelA 389
activation either in astrocytes or ODC, or in neurons and macroglia together, elicited a graded, 390
cell type-specific stimulation of axon regrowth. The robust growth stimulation in either case of 391
RelA depletion – displaying a five- to 19-fold relative increase over controls and an absolute 392
augmentation in RelACNSKO mice that exceeded that in RelAASTKO >> RelAODCKO mice – points to 393
multimodal positive effects exerted synergistically by suppression of tonic growth inhibitors 394
from myelin, glial scar, and severed RGC and axons. The growth stimulation in RelAODCKO mice 395
coincided with pronounced degradation of MBP protein in the lesioned ON, suggesting 396
accelerated Wallerian degeneration in RelAODCKO mice and, thus, reconstitution of a more 397
permissive post-injury milieu. This is in accordance with previously described RelA-dependent 398
activation of the MBP promoter in response to TNF stimulation (Huang et al., 2002) and may 399
have implications for de- and remyelination in various myelin-related disorders. 400
The growth-promoting influence of ODC-specific RelA deletion was quantitatively exceeded by 401
its loss in astrocytes. Mechanistically, enhanced regeneration in RelAASTKO mice correlated with a 402
diminished production of CSPG at the lesion site, suggesting that loss of RelA in astrocytes 403
facilitates penetrating growth events through the glial scar and restitutes a target-directed growth 404
orientation. Brambilla and colleagues identified that expression of a dominant-negative form of 405
IBα in scar-forming astrocytes improves functional recovery following SCI (Brambilla et al., 406
2005; 2009). Here, our GLAST-Cre model suggests further benefit from RelA inhibition not 407
only in astrocytes of the ON, but also in MG of associated retina. Due to the most comprehensive 408
deletion mechanism in the nestin-Cre line, deviations from basal RGC densities calculated for 409
naïve RelACNSKO mice were not expected for RelAODCKO or RelAASTKO mice and thus are unlikely 410
to be the reason for the gradual differences in regeneration. That the most effective regeneration 411
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 15
was observed in RelACNSKO mice implicates neuron-intrinsic RelA effects when reciprocal neuro-412
glial communication is present. Ongoing studies on RGC/neuron-specific knockout mice will 413
further elucidate temporal and spatial neuro-glial interactions. 414
The histologically evident increase in axonal outgrowth in RelACNSKO mice was confirmed by 415
preceding tract-specific, contrast-enhanced MRI techniques performed in vivo (suppl. results and 416
suppl. Fig. 2; Haenold et al., 2012; Fischer et al., 2014). Ongoing long-term studies using 417
repetitive MRI of the visual projection may delineate whether these axons become connected 418
with midbrain targets, as recently claimed by Benowitz’s group under cAMP- and PTEN-419
regulated oncomodulin effects (De Lima et al., 2012). 420
Growth modulation by NF-B was highly subunit specific since abrogation of the RelA-binding 421
partner p50 did not show any pro-regenerative effect but rather destabilized axonal integrity. 422
Moreover, naïve p50KO animals exhibited signs of precocious atrophy as early as 10 months after 423
birth, followed by severe cytoskeletal disintegration and functional visual impairments (will be 424
shown elsewhere). Such detrimental consequences are in line with the spontaneous degenerative 425
process recently described for aging p50-deficient animals (Lang et al., 2006; Takahashi et al., 426
2007). Collectively, the balance between p50 and RelA appears necessary to stabilize axons on 427
the structural and functional level. In the absence of RelA, p50 cannot compensate for the lack of 428
transcriptional regulation by RelA. Apart from such loss-of-function, growth promotion may be 429
released by novel dimerization partners of p50, such as Bcl-3 or c-Rel. We are currently 430
generating neuro-glia-specific RelA;c-Rel double-knockout mice to establish whether the pro-431
regenerative response is further enhanced or diminished. In contrast, in p50KO mice the lack of its 432
suppressor function might release pro-apoptotic/pro-degenerative gene expression profiles. 433
Whether reactive NF-B induction in p50KO mice – as shown for retina – also occurs in ODC 434
and astrocytes of the ON and thus contributes to modulated growth responses will be addressed 435
in future studies. 436
437
Transcriptional changes controlled by RelA and p50, which influence axogenesis and axonal 438
degeneration are still undefined. Interestingly, inhibition of Cdh1, a co-activator for the E3 439
ubiquitin ligase anaphase-promoting complex/cyclosome (APC/C), in embryonic primary 440
cultures can enhance axon elaboration and override growth suppression by myelin factors 441
(Konishi et al., 2004). In cycling cells, Cdh1 promotes ubiquitylation and degradation of cell 442
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
16
cycle regulators such as cyclins and Cdh1 itself. Transcriptional repression of Cdh1 by 443
upregulation of the Smad-interacting protein-1 (SIP1/ZEB2) influences growth arrest and 444
senescence, e.g. upon TNF-mediated NF-B activation (Chua et al., 2007; Katoh and Katoh, 445
2009). Notably, APCCdh1 expression has been shown to remain particularly high in nuclei of 446
post-mitotic neurons. Here, analysis of Cdh1 in naïve CNS tissue indicated expression in the 447
retina and cortex as well as in myelin-enriched axonal projections, thus complementing 448
previously described Cdh1 expression in astrocytes (Herrero-Mendez et al., 2009). In accordance 449
with its anti-growth function, retinal Cdh1 levels were suppressed in growth permissive 450
RelACNSKO mice following ONI. In the different cell type-specific RelA knockout lines, the best 451
pro-regenerative effect paralleled the greatest Cdh1 decline, whilst Cdh1 levels increased in 452
growth-incompetent retinae of p50KO mice. Therefore, our data suggest that Cdh1 is a CNS-453
specific downstream target of RelA and possibly of p50, which can regulate growth responses by 454
balancing intranuclear Cdh1 levels. In support of this molecular interaction, the F-box only 455
protein EMI1 that physiologically functions as a negative Cdh1 regulator was reversely governed 456
and upregulated in growth-conditioned RelACNSKO mice, but downregulated in growth-457
incompetent p50KO mice. 458
These observations make Cdh1 and its coupling to NF-B activity a strong candidate for 459
integrating extra- and intra-neuronal growth impulses in the mature CNS, followed by either a 460
positive or negative growth response. The reduced and predominantly cytoplasmic, i.e. inactive, 461
Cdh1 protein content in RelACNSKO mice supports the proposal of a RelA-dependent Cdh1 462
inactivation with consecutive growth stimulatory function. In contrast, stabilized Cdh1 levels in 463
retinae of RelAflox mice coincided with enhanced nuclear Cdh1 accumulation. As a further line of 464
evidence for this pathway, the Cdh1 substrate Id2 was stabilized in nuclei of RelACNSKO neuronal 465
cultures. Increment of steady-state levels of Id2 either by overexpression of an ubiquitylation-466
resistant D-box mutant or by Cdh1 knockdown has been reported to stimulate axonal outgrowth 467
of immature neurons and enhances regeneration of dorsal root ganglion neurons after SCI 468
(Lasorella et al., 2006; Yu et al., 2011). As for Cdh1, pro-regenerative effects may only in part 469
depend on ubiquitylation and proteasomal degradation essential for cell cycle control, e.g. by 470
cyclins, but implement CNS-specific regulatory mechanisms. Studies on fractionated cell 471
extracts will further specify the role of Id2 in growth control. 472
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 17
In summary, our results support a novel mechanism that controls axonal self-renewal and 473
Wallerian degeneration in the adult CNS in a dual manner by highly subunit- and cell type-474
specific regulation of a RelA/p50-EMI1-APCCdh1-Id2 cascade. Whilst recent data on the 475
influence of NF-B or Cdh1 on axonal growth were acquired from cultured and immature 476
neurons, our study addresses open questions on their growth regulatory capacity in vivo and in 477
the mature CNS and extends the current knowledge on NF-B and Cdh1 to the pathophysiology 478
of neurodegenerative and traumatic CNS diseases. 479
480
Materials and Methods 481
Transgenic mice 482
For regeneration studies, 14–18-week-old mice with a homozygous Cre/loxP-based deletion of 483
relA alleles (RelAfl/fl;tg/+) and floxed littermate controls (RelAfl/fl;+/+, designated RelAflox) were 484
used. Conditional RelAflox mice were a kind gift from R. M. Schmidt (Technical University 485
Munich, Germany) (Algül et al., 2007). The nestin-Cre mouse line was obtained from The 486
Jackson Laboratory (Bar Harbor, Maine, USA). Animals were backcrossed to a C57Bl/6 (B6) 487
background for at least 10 generations. Oligodendrocyte-specific RelA deletion was achieved by 488
CNP1 promoter-driven Cre activity in RelAODCKO mice (Lappe-Siefke et al., 2003) and verified 489
in Rosa26R;CNP1-Cre double transgenic reporter animals carrying a lox-STOP-lox cassette 490
(Soriano, 1999). In these animals, the loci of Cre expression can be monitored by lacZ-sensitive 491
colorimetric assay. For astrocyte-specific RelA ablation (RelAASTKO), a GLAST-driven and 492
tamoxifen-inducible CreERT2/loxP system was employed (Slezak et al., 2007). Deletion was 493
induced as published elsewhere (Slezak et al., 2007). Constitutive knockout animals lacking the 494
p50 subunit (p50KO) by insertion of a pGK-neo cassette into exon 6 on a B6 background (Sha et 495
al., 1995), and B6 wild type controls were applied at an age of three and 10 months. Pan-NF-B 496
activation was assessed using the B-lacZ reporter line (Schmidt-Ullrich et al., 1996). 497
Animals were kept under controlled conditions in a pathogen-free environment and provided 498
with food and water ad libitum. All animal interventions were performed under deep anesthesia 499
and in accordance with the European Convention for Animal Care and Use of Laboratory 500
Animals and approved by the local ethics committee. 501
502
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
18
Acute CNS injury and tracer applications 503
The ON was squeezed immediately behind the posterior eye pole with small, tilted forceps for 10 504
s. Eyes which developed ocular pathology/ischemia (e.g. corneal opaqueness, retinal atrophy, 505
congestive bleeding, cataract, lipofuscin deposits) were excluded. In this model of axonotmesis, 506
preserved myelin sheaths serve as guidance structures for outgrowing axons (Misantone et al., 507
1984). 508
ON fiber regeneration was evaluated by a dual anterograde tracing protocol: two to three days 509
prior to ONI, 2 μl of FITC-conjugated CTX (CTX-488; 1 µg/µl; C-22842, Molecular 510
Probes/MoBiTec, Goettingen, Germany) were intravitreally (i.vit.) injected, followed by i.vit. 511
application of complementary Cy3-labeled CTX (2 µl of CTX-594; 1 µg/µl; C-22841, Molecular 512
Probes/MoBiTec) four weeks after ONI and two to three days prior to histological dissection 513
(suppl. Fig. 1). Atraumatic i.vit. injections were performed with a 5-µl Hamilton syringe 514
connected to a 34 G needle in the infero-temporal circumference approximately 1 mm distal of 515
the corneo-scleral circumference, thereby sparing scleral vessels. To monitor needle insertion 516
and liquid inoculation and to avoid lens puncture, microinjections were accomplished under a 517
binocular microscope (Zeiss, Jena, Germany). 518
519
Immunohistochemistry and immunocytochemistry 520
NF-B RelA expression was evaluated using RelA (polyclonal C-20, 1:500, Santa Cruz, 521
Heidelberg, Germany), phospho-specific RelA (monoclonal pSer-536 clone 93H1, 1:200-1:500, 522
Cell Signaling, Frankfurt, Germany) and NLS-specific RelA (monoclonal MAB 3026 clone 523
12H11, 1:250, Millipore, Darmstadt, Germany) antibodies. Axonal de- and regeneration were 524
assessed by co-stainings against phosphorylated neurofilaments (SMI-31, monoclonal, 1:500, 525
Sternberger Monoclonals, Lutherville, ML, USA), MAP-2 (monoclonal, 1:500, Chemicon), and 526
GAP-43 (polyclonal, 1:250, Chemicon). For cellular co-localization, antibodies against RGC 527
(TUJ-1, monoclonal, 1:250, Covance, Munich, Germany), MG (CRALBP, monoclonal, 1:250, 528
Abcam, Cambridge, MA, USA; GLAST, polyclonal, 1:250, Santa Cruz), astrocytes (GFAP, 529
polyclonal, 1:500, DAKO, Hamburg, Germany), and ODC (CAII, polyclonal, 1:500, Santa Cruz) 530
were used. Reporter expression of EGFP and β-galactosidase were assessed with antibodies 531
against GFP (polyclonal, 1:100, Santa Cruz) and β-gal (polyclonal, 1:500, Chemicon). 532
Inflammatory responses (not shown) were evaluated with antibodies against CD11b 533
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 19
(monoclonal, 1:500, Serotec, Düsseldorf, Germany), Iba1 (polyclonal, 1:250, Wako Chemicals, 534
Richmond, VA, USA) and F4/80 (polyclonal, 1:500, Dianova, Hamburg, Germany). Glial scar 535
formation was investigated applying antibodies against neurocan (monoclonal, 1:250, Abcam) 536
and GFAP (polyclonal, 1:500, DAKO). Cdh1 and Id2 in cultured hippocampal neurons were 537
detected using anti-Fizzy-related (monoclonal, 1:500, Novus Biologicals, Cambridge, UK; 538
polyclonal, 1:500, Invitrogen, Darmstadt, Germany) and anti-Id2 (polyclonal, 1:500, Santa Cruz) 539
antibodies. Suitable secondary antibodies were used to visualize primary antibody binding. 540
Photomicrographs were captured with Zeiss AxioVision 2 (Zeiss, Jena, Germany) and 541
AxioImager microscopes (software AxioVision 4.8; Zeiss). 542
543
X-Gal and Fluoro Jade B staining protocols 544
ON and retinae of B-lacZ reporter mice were fixed in 2% PFA/0.2% glutaraldehyde in PBS 545
(4°C) for 10 min, washed, and incubated in X-Gal staining solution (Roche, Mannheim, 546
Germany) for 24 h at 37°C. Samples of liver and thymus were taken as positive controls. 547
Longitudinal ON sections were immersed in 1% sodium hydroxide in 80% ethanol for 5 min, 548
followed by incubation in 70% ethanol and distilled water for 2 min, respectively. For 549
background suppression, slides were incubated in 0.06% potassium permanganate for 10 min. 550
The staining solution was prepared from a 0.01% stock of Fluoro Jade B (Chemicon) in 0.1% 551
acetic acid and freshly used at a final concentration of 0.0004%. After a 20-min incubation, 552
slides were rinsed in distilled water, dried at 50°C, cleared with xylene, and covered with 553
ImmuMountTM (Shandon, Pittsburgh, PA, USA). 554
555
Electron microscopy 556
Animals were transcardially perfused with ice-cold PBS and freshly prepared fixative containing 557
1% PFA, 3% glutaraldehyde, 0.5% acrylaldehyde, and 0.05 M CaCl2 in 0.1 M cacodylate buffer 558
(pH 7.3). ON were fixed at 4°C overnight and processed for ultrafine sectioning on an EMTP 559
tissue processor (Leica, Wetzlar, Germany). Specimens were rinsed six times (15 min/rinse) in 560
0.1 M cacodylate buffer and postfixed in 1% osmium and 1% potassium hexacyanoferrate II in 561
0.1 M cacodylate buffer at 4°C for 1 h. Following rinsing in cacodylate buffer and distilled 562
water, ON were dehydrated in acetone (30, 50, 70, 90, 95%; 30 min at each step; Merck, 563
Darmstadt, Germany), then twice in 100% acetone (45 min each time). Samples were transferred 564
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
20
to acetone-resin (EPON) composites and embedded for polymerization. Ultrathin sections of 50-565
nm thickness were cut with Reichert Ultracut S knifes (Leica) and 35° diamond blades (Diatome, 566
Hatfield, PA, USA). Ultramicrographs were captured by a transmission electron microscope 567
(JEM 1400, Jeol, Ehing, Germany) under 80 kV using a CCD camera (Orius SC 1000, Gatan 568
Munich, Germany). 569
570
Quantification of RGC/ODC and axonal regeneration 571
Four weeks after ONI, retinae were fixed in 4% PFA/PBS (pH 7.4), flattened on glass slides, and 572
subjected to immunohistochemical procedures using the RGC marker TUJ-1. RGC densities 573
were assessed at 3–4/6 radial eccentricity under a fluorescence microscope (AxioImager, Zeiss). 574
Fluorescence microscopy was also used to quantify ODC in 10-µm longitudinal cryosections of 575
the naïve ON after labeling with CAII antibodies. 576
Serial quantification of growth responses in anterogradely traced ON was performed using the 577
AutMess imaging program (Zeiss) supplemented by the AxioVision LE module under a 40× 578
objective (AxioImager, Zeiss). Since the undiluted CTX tracer displays defined invariable signal 579
intensity, the threshold for signal detection was similar for all groups and specimens. Fields of 580
regeneration (ROI) were automatically identified and summed to define the total regenerative 581
area in each slice. The total regenerative area of each ON was determined by summing the ROIs 582
of all the slices. 583
584
Immunoblotting 585
Lysates of retinae and ON were exposed to SDS-PAGE and processed for immunoblotting 586
according to standard protocols. The following antibody concentrations were used: RelA 587
(polyclonal C-20, 1:1,000, Santa Cruz), phospho-RelA (monoclonal pSer-536 clone 93H1, 588
1:1,000, Cell Signaling), NLS-RelA (monoclonal MAB 3026 clone 12H11, 1:500, Millipore), 589
MBP (polyclonal, 1:200, Millipore), Id2 (polyclonal, 1:500, Santa Cruz), Cdh1 (Fizzy-related, 590
monoclonal, 1:500, Novus Biologicals), and EMI1 (monoclonal, 1:500, Invitrogen). β-Actin 591
(polyclonal, 1:10,000, Abcam) was applied as loading control. Results were repeated at least 592
three times for three different specimens. 593
594
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 21
In vitro experiments 595
For immunocytochemistry on primary neurons, cultures of E16 embryonic RelACNSKO and 596
RelAflox hippocampi were stained with SMI-31, Cdh1, and Id2 antibodies and counterstained with 597
DAPI. Our protocol on embryonic, genotype-specific neuronal cultures from transgenic mice is 598
available on request. Intranuclear-to-cytoplasmic switch of the signals was semi-quantitatively 599
assessed using a Zeiss AxioImager microscope and AxioVision imaging software (Zeiss). 600
OLN-93 cells were grown as monolayers in sterile DMEM medium supplemented with 10% 601
fetal bovine serum for 5 days at 37°C, 5% CO2 and controlled humidity. To induce 602
differentiation, serum was reduced to 0.5% for 5 days. Immuno-stainings with NLS-RelA or 603
RelA (C-20) antibodies (see above) were performed 30 min after TNF (20 ng/ml; Sigma, 604
Germany) or PBS treatment. 605
606
Optometric measurement 607
Visual acuity and contrast sensitivity were investigated making use of the optokinetic reflex in a 608
virtual-reality optomotor device. Freely moving animals were subjected to moving sine wave 609
gratings of various spatial frequencies and contrasts. Gratings were varied up to the detection 610
threshold of reflexive head tilting (Prusky et al., 2004). Animals in the knockout and control 611
groups were analyzed at identical time points within the circadian rhythm. 612
613
Statistical analysis 614
Statistical analyses were performed using the Student’s t-test for single comparisons, followed 615
by post-hoc test calculation. Data are presented as mean±standard error (s.e.m.). For each 616
experiment, individual n numbers are given separately. Results reaching P≤0.05 were considered 617
statistically significant (P<0.05, *; P<0.01, **; P<0.001, ***; P> 0.05, #). 618
619
Acknowledgements 620
We appreciate the provision of the OLN-93 cell line by C. Richter-Landsberg. We thank S. 621
Tausch for technical and M. Baldauf for histological assistance. We are grateful to K. Buder for 622
the support on ELMI and S. Schmidt and I. Krumbein for MRI advice. 623
624
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
22
Funding 625
R.H. is supported by the VELUX Foundation, Switzerland. A.K. was supported by the IZKF 626
Jena. 627
628
Author Contributions 629
R.H. and A.K. organized the study and prepared the manuscript. K.-H.H. developed the MEMRI 630
protocol. K.K. and K.-F.S. conducted the functional animal tasks. C.E. performed cell culture 631
experiments. K.-A.N. provided the Cre line for the creation of ODC specific mouse mutants. 632
F.W., O.W.W., S.L., and J.R.R. supervised and financed the study and helped with data 633
interpretation. 634
635
References 636
Algül, H., Treiber, M., Lesina, M., Nakhai, N., Saur, D., Geisler, F., Pfeifer, A., Paxian, S. 637
and Schmid, R. M. (2007). Pancreas-specific RelA/p65 truncation increases susceptibility of 638
acini to inflammation-associated cell death following cerulein pancreatitis. J. Clin. Invest. 639
117, 1490-1501. 640
Berkowitz, B. A., Roberts, R., Oleske, D. A., Chang, M., Schafer, S., Bissig, D. and 641
Gradianu, M. (2009). Quantitative mapping of ion channel regulation by visual cycle activity 642
in rodent photoreceptors in vivo. IOVS 50, 1880-1885. 643
Brambilla, R., Bracchi-Ricard, V., Hu, W. H., Frydel, B., Bramwell, A., Karmally, S., 644
Green, E. J. and Bethea, J. R. (2005). Inhibition of astroglial nuclear factor kappaB reduces 645
inflammation and improves functional recovery after spinal cord injury. J. Exp. Med. 646
202,145-156. 647
Brambilla, R., Hurtado, A., Persaud, T., Esham, K., Pearse, D. D., Oudega, M., Bethea and 648
J. R. (2009). Transgenic inhibition of astroglial NF-kappa B leads to increased axonal sparing 649
and sprouting following spinal cord injury. J. Neurochem. 110, 765-778. 650
Chen, M. S., Huber, A. B., van der Haar, M. E., Frank, M., Schnell, L., Spillmann, A. A., 651
Christ A. F. and Schwab, M. E. (2000). Nogo-A is a myelin-associated neurite outgrowth 652
inhibitor and an antigen for monoclonal antibody IN-1. Nature 403, 434-439. 653
Choi, J.-S., Kim-Yoon, S. and Joo, C.-K. (1998). NF-B activation following optic nerve 654
transaction. Korean J. Ophthalmol. 12, 19-24. 655
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 23
Chua, H. L., Bhat-Nakshatri, P., Clare, S. E., Morimiya, A., Badye, S. and Nakshatri, H. 656
(2007). NF-B represses E-cadherin expression and enhances epithelial to mesenchymal 657
transition of mammary epithelial cells: potential involvement of ZEB1 and ZEB2. Oncogene 658
26, 711-724. 659
De Lima, S., Koriyama, Y., Kurimoto, T., Oliveira, J. T., Yin, Y., Li, Y., Gilbert, H. Y., 660
Fagiolini, M., Martinez, A. M. and Benowitz, L. (2012). Full-length axon regeneration in 661
the adult mouse optic nerve and partial recovery of simple visual behaviors. Proc. Natl. Acad. 662
Sci. U.S. A. 109, 9149-54. Erratum in (2012): Proc. Natl. Acad. Sci. U.S.A 10933, 13465. 663
Emmanouil, M., Taoufik, E., Tseveleki, V., Vamvakas, S. S., Tselios, T., Karin, M., 664
Lassmann, H. and Probert, L. (2009). Neuronal I kappa B kinase beta protects mice from 665
autoimmune encephalomyelitis by mediating neuroprotective and immunosuppressive effects 666
in the central nervous system. J. Immunol. 183, 7877-7889. 667
Fischer, S., Engelmann, C., Herrmann, K.-H., Reichenbach, J.R., Witte, O.W., Weih, F., 668
Kretz, A. and Haenold, R. (2014). In vivo imaging of optic nerve fiber integrity by contrast-669
enhanced MRI in mice. J. Vis. Exp., e51274, doi:10.3791/51274 (in press 2013). 670
Gavaldà, N., Gutierrez, H. and Davies, A. M. (2009). Developmental switch in NF-kappaB 671
signalling required for neurite growth development. Development 136, 3405-3412. 672
Gutierrez, H. and Davies, A. M. (2011). Regulation of neural process growth, elaboration and 673
structural plasticity by NF-B. Trends in Neurosci. 34, 316-325. 674
Gutierrez, H., O'Keeffe, G. W., Gavaldà, N., Gallagher, D. and Davies, A. M. (2008). 675
Nuclear factor kappa B signaling either stimulates or inhibits neurite growth depending on the 676
phosphorylation status of p65/RelA. J. Neurosci. 28, 8246-8256. 677
Haenold, R., Herrmann, K.-H., Schmidt, S., Reichenbach, J. R., Schmidt, K.-F., Löwel, S., 678
Witte, O. W., Weih, F. and Kretz, A. (2012). Magnetic resonance imaging of the mouse 679
visual pathway for in vivo studies of degeneration and regeneration in the CNS. Neuroimage 680
59, 363-376. 681
Herrero-Mendez, A., Almeida, A., Fernández, E., Maestre, C., Moncada, S. and Bolaños, 682
J. P. (2009). The bioenergetic and antioxidant status of neurons is controlled by continuous 683
degradation of a key glycolytic enzyme by APC/C-Cdh1. Nat. Cell Biol. 11, 747-752. 684
Huang, C. J., Nazarian, R., Lee, J., Zhao, P. M., Espinosa-Jeffrey, A. and de Vellis, J. 685
(2002). Tumor necrosis factor modulates transcription of myelin basic protein gene through 686
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
24
nuclear factor kappa B in a human oligodendroglioma cell line. Int. J. Dev. Neurosci. 20, 289-687
296. 688
Huynh, M. A., Stegmüller, J., Litterman, N. and Bonni, A. (2009). Regulation of Cdh1-APC 689
function in axon growth by Cdh1 phosphorylation. J. Neurosci. 29, 4322-4327. 690
Imielski, Y., Schwamborn, J. C., Lüningschrör, P., Heimann, P., Holzberg, M., Werner, H., 691
Leske, O., Püschel, A. W., Memet, S., Heumann, R., Israel, A., Kaltschmidt, C. and 692
Kaltschmidt, B. (2012). Regrowing the adult brain: NF-B controls functional circuit 693
formation and tissue homeostasis in the dentate gyrus. PLoS One 7, e30838. 694
Inta, I., Paxian, S., Maegele, I., Zhang, W., Pizzi, M., Spano, P., Sarnico, I., Muhammad, S., 695
Herrmann, O., Inta, D., Liou, H.-C., Schmid, R. M. and Schwaninger, M. (2006). Bim 696
and Noxa are candidates to mediate the deleterious effect of the NF-B subunit RelA in 697
cerebral ischemia. Neurobiol. Dis. 26, 12896-12903. 698
Kaltschmidt, B. and Kaltschmidt, C. (2009). NF-B in the nervous system. Cold Spring Harb 699
Perspect Biol. 1, a001271. 700
Katoh, M. and Katoh, M. (2009). Integrative genomic analyses of ZEB2: Transcriptional 701
regulation of ZEB2 based on SMADs, ETS1, HIF1alpha, POU/OCT, and NF-kappaB. Int. J. 702
Oncol. 34, 1737-1742. 703
Kim, N. S., Kim, H. J., Koo, B. K., Kwon, M. C., Kim, Y. W., Cho, Y., Yokota, Y., 704
Penninger J. M. and Kong, Y. Y. (2006). Receptor activator of NF-kappaB ligand regulates 705
the proliferation of mammary epithelial cells via Id2. Mol. Cell. Biol. 26, 1002-1013. 706
Konishi, Y., Stegmüller, J., Matsudal, T., Bonni, S. and Bonni, A. (2004). Cdh1-APC 707
controls axonal growth and patterning in the mammalian brain. Science 303, 1026-1030. 708
Kretz, A., Herrmann, K.H., Fischer, S., Engelmann, C., Witte, O.W., Reichenbach, J.R., 709
Weih, F., Haenold, R. (2013) Dysfunctional NF-κB and brain myelin formation. Eur. J. 710
Hum. Genet. 22, 724-725. 711
Lang, H., Schulte, B. A., Thou, D., Smythe, N., Spicer, S. S., and Schmiedt, R. A. (2006). 712
Nuclear factor кB deficiency is associated with auditory nerve degeneration and increased 713
noise-induced hearing loss. J. Neurosci. 26, 3541-3550. 714
Lappe-Siefke, C., Goebbels, S., Gravel, M., Nicksch, E., Lee, J., Braun, P. E., Griffiths, I. 715
R., and Nave, K.-A. (2003). Disruption of Cnp1 uncouples oligodendroglial function in 716
axonal support and myelination. Nature Gen. 33, 366-374. 717
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 25
Lasorella, A., Stegmüller, J., Guardavaccaro, D., Liu, G., Carro, M. S., Rothschild, G., de 718
la Torre-Ubieta, L., Pagano, M., Bonni, A. and Lavarone, A. (2006). Degradation of id2 719
by the anaphase-promoting complex couples cell cycle exit and axonal growth. Nature 27, 720
471-474. 721
Li, J., Lu, Z., Li, W. L., Yu, S. P. and Wei, L. (2008). Cell death and proliferation in NF-B 722
p50 knockout mouse after cerebral ischemia. Brain Res. 1230, 281-289. 723
Lu, Z. Y., Yu, S. P., Wei, J. F. and Wei, L. (2006). Age-related neural degeneration in nuclear-724
factor kappaB p50 knockout mice. Neuroscience 139, 965-978. 725
Meffert, M. K., Chang, J. M., Wiltgen, B. J., Fanselow, M. S. and Baltimore, D. (2003). NF-726
kappa B functions in synaptic signaling and behavior. Nat. Neurosci. 6, 1072-1078. 727
Misantone, L. J., Gershenbaum, M. and Murray, M. (1984). Viability of retinal ganglion 728
cells after optic nerve crush in adult rats. J. Neurocytol. 13, 449-465. 729
Prusky, G. T., Alam, N. M., Beekman, S. and Douglas, R. M. (2004). Rapid quantification of 730
adult and developing mouse spatial vision using a virtual optomotor system. IOVS 45, 4611-731
4616. 732
Richter-Landsberg, C. and Heinrich, M. (1996). OLN-93: a new permanent oligodendroglia 733
cell line derived from primary rat brain glial cultures. J. Neurosci. Res. 45, 161-73. 734
Raasch, J. N., Zeller, G., van Loo, G., Merkler, D., Mildner, A., Erny, D., Knobeloch, K. 735
P., Bethea, J. R., Knust, A. M., Del Turco, D., Deller, T., Blank, T., Priller, J., Brück, 736
W., Pasparakis, M. and Prinz, M. (2011). IkappaB kinase 2 determines oligodendrocyte 737
loss by non-cell-autonomous activation of NF-kappaB in the central nervous system. Brain 738
134, 1184-1198. 739
Sandvig, I., Thuen, M., Hoang, L., Olsen, Ø., Sardella, T. C., Brekken, C., Tvedt, K. E., 740
Barnett, S. C., Haraldseth, O., Berry, M. and Sandvig, A. (2012). In vivo MRI of olfactory 741
ensheathing cell grafts and regenerating axons in transplant mediated repair of the adult rat 742
optic nerve. NMR Biomed. 25, 620-31. 743
Schmidt-Ullrich, R., Mémet, S., Lilienbaum, A., Feuillard, J., Raphaël, M. and Israel, A. 744
(1996). NF-kappaB activity in transgenic mice: developmental regulation and tissue 745
specificity. Development 122, 2117-2128. 746
Sha, W. C., Liou, H. C., Tuomanen, E. I. and Baltimore, D. (1995). Targeted disruption of the 747
p50 subunit of NF-B leads to multifocal defects in immune responses. Cell 80, 321-330. 748
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
26
Slezak, M., Göritz, C., Niemiec, A., Frisén, J., Chambon, P., Metzger, D. and Pfrieger, F. 749
W. (2007). Transgenic mice for conditional gene manipulation in astroglial cells. Glia 55, 750
1565-1576. 751
Soriano, P. (1999). Generalized lacZ expression with the ROSA26 Cre reporter strain. Nature 752
21, 70-71. 753
Srinivas, S., Watanabe, T., Lin, C. S., William, C. M., Tanabe, Y., Jessell, T. M., and 754
Costantini, F. (2001). Cre reporter strains produced by targeted insertion of EYFP and ECFP 755
into the ROSA26 locus. BMC Dev. Biol. 1, 4. 756
Takahashi, Y., Katai, N., Taniguchi, S.-I. and Hayashi, T. (2007). Development of 757
spontaneous optic neuropathy in NF-Bp50-deficient mice: requirement for NF-Bp50 in 758
ganglion cell survival. Neuropathol. Applied Neurobiol. 33, 692-705. 759
Thuen, M., Olsen, O., Berry, M., Pedersen, T. B., Kristoffersen, A., Haraldseth, O., 760
Sandvig, A. and Brekken, C. (2009). Combination of Mn(2+)-enhanced and diffusion tensor 761
MR imaging gives complementary information about injury and regeneration in the adult rat 762
optic nerve. J. Magn. Reson. Imaging 29, 39-51. 763
Yu, P., Zhang, Y. P., Shields, L. B., Zheng, Y., Hu, X., Hill, R., Howard, R., Gu, Z., Burke, 764
D. A., Whittemore, S. R., Xu, X. M. and Shields, C. B. (2011). Inhibitor of DNA binding 2 765
promotes sensory axonal growth after SCI. Exp. Neurol. 231, 38-44. 766
Yu, X., Wadghiri, Y. Z., Sanes, D. H. and Turnbull, D. H. (2005). In vivo auditory brain 767
mapping in mice with Mn-enhanced MRI. Nature Neuroscience 8, 961-968. 768
Zhang, W., Potrovita, I., Tarabin, V., Herrmann, O., Beer, V., Weih, F., Schneider, A. and 769
Schwaninger, M. (2005). Neuronal activation of NF-B contributes to cell death in cerebral 770
ischemia. J. Cereb. Blood Flow Metab. 25, 30-40. 771
772
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 27
Figure Legends 773
Fig. 1. Visual system characterization under modulated RelA activity. 774
(A) H&E staining of transversal retina showed unsuspicious layer morphology in RelACNSKO 775
mice. Scale bar: 100µm (B) Co-labeling of retinal whole mounts with the RGC marker TUJ-1 776
(green) and the MG marker CRALBP (red) exhibited a normal RGC and MG ratio and cell 777
contacts in superficial NFL/GCL in RelACNSKO mice as compared to controls (n=3). Scale bar: 50 778
μm. (C) Electron microscopy displayed unimpaired myelin sheaths insulating ON axons in 779
RelACNSKO mice as compared to controls (n=5/6). Scale bar in left row: 150 nm; scale bar in right 780
row: 25 nm. (D) Impaired cell differentiation and survival were ruled out by similar naïve RGC 781
densities (P=0.88), ODC numbers (P=0.43), and g-ratios (P=0.99) in the ON of RelACNSKO mice 782
and controls (n=5/6 for each group). (E) Functional parameters revealed physiological values for 783
visual acuity (P=0.21) and contrast sensitivity (P=0.61) in three-month-old RelACNSKO mice 784
(n=5). (F) Increased retinal κB-lacZ reporter activity in B-lacZ;p50KO mice compared to 785
controls indicated RelA/NF-B hyperactivity due to p50 deletion (P<0.02; n=5). Scale bar: 200 786
μm. NFL, nerve fiber layer; GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner 787
nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; PR, photo receptor layer; 788
ODC, oligodendrocyte. 789
790
Fig. 2. ONI activates NF-B in a cell-autonomous manner. 791
(A) ONI in B-lacZ reporter mice increased the number of β-gal-positive cells in the retina by 792
three- (1-3 dpi) and 18-fold (4-10 dpi) thereafter. dpi, days post injury (n=3). (B) Micrographics 793
of retinal whole mount preparations depicting increased numbers of X-Gal-positive cells (blue) 794
in superficial retinal layers after ONI (X-Gal panel). Scale bar: 200 μm. Transversal retinal 795
sections indicated pan-NF-B activation in the GCL (ONI panel), and co-localization of the β-796
gal signal (red) with the TUJ-1 staining (green) evidenced NF-B activation in DAPI-positive 797
(blue) RGC nuclei (purple; inset, 40× objective) (n=4). Scale bar: 80 μm. (C) Immunoblot 798
analysis with total RelA (C-20) and phospho-specific (Ser536) RelA antibodies demonstrated 799
expression of RelA in naïve ON, and activation within 3 h after ONI in WT animals (top; n=3). 800
TNF-treated hippocampal neurons served as a positive control. Ser536 phosphorylation of RelA 801
was almost completely abolished in RelACNSKO mice (bottom; n=3). (D) Enzymatic X-Gal assay 802
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
28
of ON in toto preparation revealed NF-B activation in the lesion epicenter (n=6). Scale bar: 100 803
μm. Co-localization of the β-gal signal (red; ONI panel) with the ODC marker CAII (green) in 804
longitudinal ON sections confirmed NF-B activation in DAPI-stained nuclei (blue) of ODC 805
(purple; inset, 40× objective) (n=3). Scale bar: 30 μm. Note the absence of CAII reactivity in the 806
non-myelinated ON head (dotted line in naïve panel). Scale bar: 100 μm. (E) Left: Co-807
localization (yellow) of NLS-RelA-positive cells (green) with β-gal signals (red) in B-lacZ 808
reporter mice showed that RelA was the dominant NF-B subunit activated by ONI in ODC 809
(n=4). Right: No NLS-RelA signal was detectable in nuclei of CAII-positive ODC of RelACNSKO 810
mice. Scale bar: 30 μm. (F) TNF-induced activation of RelA in oligodendrocytic OLN-93 cells 811
as detected by NLS-RelA antibody. Z-stack analysis confirmed nuclear localization of RelA in 812
stimulated cells. Scale bar: 20 μm. 813
814
815
Fig. 3. RelA and p50 subunits differentially control RGC survival after ONI. 816
(A) ONI performed on B-lacZ;p50KO reporter mice strongly induces B-dependent X-Gal 817
staining (blue color) at the ON injury site (right panel), which was lacking at the naïve site (left 818
panel). Asterisk, lesion site. Scale bar: 200 μm. Photomicrographs of corresponding flat-mounted 819
retinae showing substantial numbers of X-Gal-positive RGC under naïve conditions, and a 820
drastic reduction at 9 dpi. Scale bar: 200 μm. (B) RelA deletion exerts moderate, but significant 821
anti-apoptotic effects on deafferentiated RGC four weeks after ONI (n=9 for RelACNSKO; n=7 for 822
RelAflox). In contrast, p50 ablation drastically severed RGC death (n=3). 823
824
825
Fig. 4. Axonal regeneration is enhanced in RelACNSKO mice. 826
(A-D) Axonal regeneration as detected by CTX-594-positive fiber ingrowth into the ON was 827
greatly stimulated in RelACNSKO mice (n=16) as compared to controls (n=10). (A) Trans-lesional 828
fiber elongation was frequently observed in RelACNSKO mice, but rarely in controls. Asterisks, 829
lesion site. In p50KO mice (n=6), axon ingrowth into the ON head was even lower than in 830
controls. Scale bar: 100 μm. (B) Quantification of total growth area within individual ON of 831
RelACNSKO versus p50KO and control mice. Gender-related effects were not evident. (C) Selective 832
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 29
assessment of trans-lesional growth areas. (D) Absolute trans-lesional axon lengths in RelACNSKO 833
mice versus controls. (E) Measurement of pre-lesional distances indicated comparable positions 834
of the injury site in RelACNSKO mice and controls, thus excluding artifacts in real growth 835
distances. 836
837
Fig. 5. Wallerian degeneration is severed in p50KO mice. 838
(A) CTX-594 (red) co-localized with GAP-43 (green) in elongating axons of RelACNSKO mice. 839
Note the strongest GAP-43 signal in axon tips (arrows) beyond the injury site (asterisks). Scale 840
bar: 100µm (B) Left: In RelACNSKO mice, newly elaborated axons developed rudimentary growth 841
cones at their GAP-43-positive tips (green). Scale bar: 15 μm. Right: In retinae counterstained 842
with GFAP for anatomic orientation, the appearance of aberrant MAP2-positive processes in 843
RelACNSKO mice indicated a pro-regenerative influence on RGC dendrites too (dashed box with 844
magnification). Scale bar: 30 μm. (C) ONI induced Wallerian degeneration in RelACNSKO mice 845
and RelAflox controls (n=3), as indicated by Fluoro Jade (FJ)-reactive axon bulges. However, 846
p50KO mice showed strong spontaneous fiber degeneration at a susceptible age of 10 months, as 847
demonstrated by intense FJ staining (n=3). Specificity of the signal was indicated by lack of FJ 848
reactivity in naïve ON of 10-month-old RelACNSKO mice and outside the injured optic tract (right 849
panel). Scale bar: 50 μm. 850
851
Fig. 6. RelA differentially modulates axonal regeneration in RelACNSKO, RelAODCKO, and 852
RelAASTKO mice. 853
(A) CNP1-Cre activity in the ON of RelAODCKO mice was confirmed by its intense blue staining 854
in Rosa26R;CNP1-Cre reporter mice. Scale bar: 100 μm. (B) RelA deletion selectively in ODC 855
(RelAODCKO) significantly enhanced axonal regeneration. (C) Accelerated degradation of white 856
matter MBP in RelACNSKO mice four weeks after ONI (two independent runs using pooled protein 857
of three animals). (D) Deletion of RelA in MG (arrows), i.e. specialized astrocytes of the retina, 858
was confirmed by YFP induction in YFP;GLAST-CreERT2 reporter mice after tamoxifen (TAM) 859
treatment. Scale bar: 50µm. (E) Pro-regenerative effects of RelA deletion in astrocytes 860
(RelAASTKO) exceeded the growth response in RelAODCKO mice. (F) Growth-promoting effects 861
paralleled a reduced production of repulsive neurocan at the injury site (n=3). Asterisks denote 862
scar region. Scale bar: 250 µm. (G) In RelACNSKO mice, regrowing axons penetrated the scar 863
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
30
region more easily (asterisks), as shown by CTX labeling (n=16 for RelACNSKO; n=10 for 864
controls). Arrows delineate preferred growth direction of newly built axons. Scale bar: 150 µm. 865
866
Fig. 7. RelA controls Cdh1 levels and its nucleo-cytoplasmic shuttling. 867
(A-D) Cdh1 acts in neurons to control RelA-dependent axon growth. (A) Expression of Cdh1 868
was increased in naïve optic and sciatic nerves and, to a lesser extent, in cortex and retina (n=3; 869
pooled protein from three animals). Calculations indicate optic density relative to actin loading 870
control. (B) Within the naïve retina, Cdh1 was localized to RGC and displayed a nuclear (n; 871
arrow), cytoplasmic (c; arrowhead), or mixed (n/c; asterisk) distribution. Scale bar, 50µm. (C) 872
Cdh1 upregulation in wild type retina four weeks after ONI was suppressed in RelACNSKO mice 873
(n=4; three samples). In contrast, Cdh1 was upregulated in growth-incompetent p50KO mice four 874
weeks after ONI as well as in progeroid retina at the age of 10 months (n=3; two samples). (D) In 875
growth-conditioned retina, Cdh1 was found suppressed as a function of the cell type targeted by 876
RelA deletion. The degree of Cdh1 decline in RelACNSKO RelAASTKO > RelAODCKO mice 877
correlated with absolute growth responses (n=4; three samples). (E) The Cdh1 inhibitor EMI1 878
was antagonistically upregulation in naïve RelACNSKO mice. ONI caused no further induction 879
(n=3; three samples). Corresponding to increased Cdh1 levels, EMI1 declined with aging in 880
p50KO mice (n=3; two samples). (F-H) RelA defines nucleo-cytoplasmic shuttling of Cdh1 in 881
cultured hippocampal neurons (HN). (F) Co-staining for Cdh1 and nuclear DAPI in HN (4 div) 882
revealed predominant nuclear (active) location of Cdh1 in wild type cells (left, purple nuclei; 883
arrowhead), whereas RelA deletion caused a shift of Cdh1 to a cytoplasmic (inactive) location 884
(right, blue nuclei; arrow). Solid neuro-neuritic TUJ-1 staining indicated viability of either 885
culture (top panel; n=3). Scale bar for TUJ-1: 30 µm; scale bar for Cdh1/DAPI: 10 µm. (G) 886
Nucleo-cytoplasmic shift of Cdh1 in RelA-deficient HN was assessed by cellular 887
immunofluorescence absorption profiles. Both graphs indicate the preponderance of 888
extranuclear, inactive Cdh1 in RelACNSKO as compared to RelAflox. (H) Quantitative analysis of 889
average cytoplasmic Cdh1 signal distribution (analyzed cell number: RelAflox n=122; RelACNSKO 890
n=131). Analysis of cytoplasmic-to-nuclear signal relation revealed a shift from high cell 891
fractions (77.1%) with low cytoplasmic Cdh1 signal (0–30%) in controls to fractions (48.1%) 892
with dominant cytoplasmic Cdh1 signal (31–70%) in RelACNSKO populations. (I) Representative 893
images of cytoplasmic Id2 signal distribution in RelAflox and RelACNSKO HN (3.5 div) revealing 894
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Haenold et al., Adult axoneogenesis by cellular balance of RelA and p50
JOCES/2013/140731 - Revised manuscript for Journal of Cell Science 31
predominant cytoplasmic accumulation of Id2 in the axon hill (left; arrowhead and inset) of wild 895
type cells, whereas RelA deletion caused its nuclear (active) concentration (right; arrow and 896
inset). Scale bar: 10 µm. (J) Quantification of subcellular signal distribution evidenced a shift 897
from dominant cytoplasmic Id2 in controls (59.2%) to a preponderance of nuclear Id2 in 898
RelACNSKO populations (78.1%) (analyzed cell number: RelAflox n=90; RelACNSKO n=130). 899
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t
Jour
nal o
f Cel
l Sci
ence
Acc
epte
d m
anus
crip
t