Long pre-mRNA depletion and RNA missplicing contribute to

Transcription

articles
© 2011 Nature America, Inc. All rights reserved.
Long pre-mRNA depletion and RNA missplicing
contribute to neuronal vulnerability from loss of
TDP-43
Magdalini Polymenidou1,2,6, Clotilde Lagier-Tourenne1,2,6, Kasey R Hutt2,3,6, Stephanie C Huelga2,3,
Jacqueline Moran1,2, Tiffany Y Liang2,3, Shuo-Chien Ling1,2, Eveline Sun1,2, Edward Wancewicz4, Curt Mazur4,
Holly Kordasiewicz1,2, Yalda Sedaghat4, John Paul Donohue5, Lily Shiue5, C Frank Bennett4, Gene W Yeo2,3 &
Don W Cleveland1,2
We used cross-linking and immunoprecipitation coupled with high-throughput sequencing to identify binding sites in 6,304
genes as the brain RNA targets for TDP-43, an RNA binding protein that, when mutated, causes amyotrophic lateral sclerosis.
Massively parallel sequencing and splicing-sensitive junction arrays revealed that levels of 601 mRNAs were changed (including
Fus (Tls), progranulin and other transcripts encoding neurodegenerative disease–associated proteins) and 965 altered splicing
events were detected (including in sortilin, the receptor for progranulin) following depletion of TDP-43 from mouse adult brain
with antisense oligonucleotides. RNAs whose levels were most depleted by reduction in TDP-43 were derived from genes with
very long introns and that encode proteins involved in synaptic activity. Lastly, we found that TDP-43 autoregulates its synthesis,
in part by directly binding and enhancing splicing of an intron in the 3′ untranslated region of its own transcript, thereby
triggering nonsense-mediated RNA degradation.
Amyotrophic lateral sclerosis (ALS) is an adult-onset disorder in which
premature loss of motor neurons leads to fatal paralysis. Most cases
of ALS are sporadic, with only 10% of affected individuals having a
familial history. A breakthrough in understanding ALS pathogenesis
was the discovery that TDP-43, which in the normal setting is primarily nuclear, mislocalizes and forms neuronal and glial cytoplasmic
aggregates in ALS1,2, frontotemporal lobar degeneration (FTLD), and
in Alzheimer’s and Parkinson’s diseases (reviewed in ref. 3). Dominant
mutations in TDP-43 were subsequently identified as being causative in
sporadic and familial ALS cases and in rare individuals with FTLD4–7.
At present, it is unresolved as to whether neurodegeneration is a result
of a loss of TDP-43 function or a gain of toxic property or a combination
of the two. However, a notable feature of TDP-43 pathology is TDP43 nuclear clearance in neurons containing cytoplasmic aggregates,
consistent with pathogenesis being driven, at least in part, by a loss of
TDP-43 nuclear function1,7.
Several lines of evidence suggest an involvement of TDP-43 in
multiple steps of RNA metabolism, including transcription, splicing or
transport of mRNA3, as well as microRNA metabolism8. Misregulation
of RNA processing has been described in a growing number of neurological diseases9. The recognition of TDP-43’s role in neurodegeneration
and the recent identification of ALS-causing mutations in FUS/TLS10,11,
another RNA/DNA binding protein, has reinforced a crucial role for
RNA processing regulation in neuronal integrity. However, a comprehensive protein-RNA interaction map for TDP-43 and identification
of post-transcriptional events that may be crucial for neuronal survival
remain to be established.
A common approach for identifying specific RNA-binding protein targets or aberrantly spliced isoforms related to disease has been
through selection of candidate genes. However, recent advances in
DNA sequencing technology have provided powerful tools for exploring transcriptomes at remarkable resolution12. Moreover, cross-linking,
immunoprecipitation and high-throughput sequencing (CLIP-seq or
HITS-CLIP) experiments have found that a single RNA-binding protein
can have previously unrecognized roles in RNA processing and affect
many alternatively spliced transcripts13–15. We used these approaches
to identify a comprehensive TDP-43 protein-RNA interaction map
in the CNS. After depletion of TDP-43 in vivo, RNA sequencing and
splicing-sensitive microarrays were used to determine that TDP-43 is
crucial for maintaining normal levels and splicing patterns of >1,000
mRNAs. The most downregulated of these TDP-43–dependent RNAs
have pre-mRNAs with very long introns that contain multiple TDP43–binding sites and encoded proteins related to synaptic activity. The
nuclear TDP-43 clearance widely reported in TDP-43 proteinopathies1,7
will lead to a disruption of this role on long RNAs that are preferentially
expressed in brain, thereby contributing to neuronal vulnerability.
1Ludwig Institute for Cancer Research, University of California at San Diego, La Jolla, California, USA. 2Department of Cellular and Molecular Medicine, University of
California at San Diego, La Jolla, California, USA. 3Stem Cell Program and Institute for Genomic Medicine, University of California at San Diego, La Jolla, California,
USA. 4Isis Pharmaceuticals, Carlsbad, California, USA. 5RNA Center, Department of Molecular, Cell and Developmental Biology, Sinsheimer Labs, University of
California, Santa Cruz, California, USA. 6These authors contributed equally to this work. Correspondence should be addressed to D.W.C. ([email protected]) or
G.W.Y. ([email protected]).
Received 20 December 2010; accepted 14 February 2011; published online 27 February 2011; doi:10.1038/nn.2779
nature neuroscience advance online publication
1
RESULTS
Protein-RNA interaction map of TDP-43 in mouse brain
We used CLIP-seq to identify in vivo RNA targets of TDP-43 in adult
mouse brain (Supplementary Fig. 1a). After ultraviolet irradiation to
stabilize in vivo protein-RNA interactions, we immunoprecipitated
TDP-43 with a monoclonal antibody16 that had a higher immunoprecipitation efficiency than any of the commercial antibodies tested
(Supplementary Fig. 1b). Complexes representing the expected molecular weight of a single molecule of TDP-43 bound to its target RNAs
were excised (Fig. 1a) and sequenced. We also observed lower mobility
protein-RNA complexes whose abundance was reduced by increased
nuclease digestion. Immunoblotting of the same immunoprecipitated
samples before radioactive labeling of the target RNAs revealed that
TDP-43 protein was a component of both the ~43-kDa complex and
more slowly migrating complexes (Fig. 1a).
We performed two independent experiments and obtained 5,341,577
and 12,009,500 36-bp sequence reads. Mapped reads of both experiments were predominantly in protein-coding genes, with ~97% of them
oriented in the direction of transcription, confirming little DNA contamination. The positions of mapped reads from both experiments were
highly consistent, as exemplified by TDP-43 binding on the semaphorin
3F transcript (Fig. 1b).
We used a cluster-finding algorithm with gene-specific thresholds that
accounted for pre-mRNA length and variable expression levels15,17 to
Figure 1 TDP-43 binds distal introns of premRNA transcripts through UG-rich sites in vivo.
(a) Autoradiograph of TDP-43-RNA complexes
trimmed by different concentrations of
micrococcal nuclease (MNase, left). Complexes
in red box were used for library preparation and
sequencing. Immunoblot showing TDP-43 in
~46 kDa and higher molecular weight complexes
dependent on ultraviolet (UV) crosslinking
(right). (b) Example of a TDP-43–binding site
(CLIP cluster) on Semaphorin 3F defined
by overlapping reads from two independent
experiments surpassing a gene-specific
threshold. (c) University of California Santa Cruz
(UCSC) Genome Browser screenshot of neurexin 3
intron 8 (mm8; chr12:89842000–89847000)
displaying three examples of TDP-43–binding
modes. The right-most CLIP cluster represents
a canonical binding site coinciding GU-rich
sequence motifs, whereas the left-most cluster
lacks GU-rich sequences and a region containing
multiple GU repeats showed no evidence of
TDP-43 binding. The second CLIP cluster
(middle purple-outlined box) with weak binding
was found only when relaxing cluster-finding
algorithm parameters. (d) Flow-chart illustrating
the number of reads analyzed from both CLIPseq experiments. (e) Histogram of Z scores
indicating the enrichment of GU-rich hexamers
in CLIP-seq clusters compared with equally sized
clusters, randomly distributed in the same premRNAs. Sequences and Z scores of the top eight
hexamers are indicated. Pie charts enumerate
clusters containing increasing counts of (GU)2
compared with randomly distributed clusters
(P ≈ 0, χ2 = 21,662). (f) Pre-mRNAs were
divided into annotated regions (top).
Distribution of TDP-43 (left) or previously
published Argonaute-binding sites21 as a control
(right) showed preferential binding of TDP-43 in
distal introns.
2
identify TDP-43–binding sites from clusters of sequence reads (Fig. 1b),
with a conservative threshold (the number of reads mapped to a cluster
had to exceed the expected number by chance at P < 0.01). This stringent
definition will miss some true binding sites, but was intentionally chosen to
identify the strongest bound sites while minimizing false positives. Indeed,
additional probable binding sites could be identified by inspection of reads
mapped to specific RNAs (see Fig. 1c). Moreover, similarly defined clusters
from the low mobility complexes (Supplementary Fig. 1b) showed a 92%
overlap with those from the monomeric complexes (Fig. 1a), consistent
with the reduced mobility complexes comprising multiple TDP-43s (or
other RNA-binding proteins) bound to a single RNA.
Genome-wide comparison of our replicate experiments revealed
(Supplementary Fig. 1c) that the vast majority (90%) of TDP-43–
binding sites in experiment 1 overlapped with those in experiment 2
(compared with an overlap of only 8% (P ≈ 0, Z = 570) when clusters
were randomly distributed across the length of the pre-mRNAs containing them). Combining the mapped sequences yielded 39,961 clusters,
representing binding sites of TDP-43 in 6,304 annotated protein-coding
genes, approximately 30% of the murine transcriptome (Fig. 1d). We
computationally sampled reads (in 10% intervals) from the CLIP
sequences and found a clear logarithmic relationship (Supplementary
Fig. 1d), from which we calculated that our current dataset contains
~84% of all TDP-43 RNA targets in the mouse brain. Comparison with
the mRNA targets identified from primary rat neuronal cells18 by RNA
a
b
MNase
UV: +
α–TDP-43
+ +
+
MNase
α–TDP-43 IgG
+ + + – +
IgG
+
1 kb
CLIP reads
experiment 1
CLIP reads
experiment 2
80
58
46
30
CLIP reads
combined
25
17
7
CLIP cluster
RNA
c
Protein
(62 reads)
Sema3F
d
2 kb
CLIP experiment 1
5,341,577 reads
CLIP reads
combined
CLIP experiment 2
12,009,500 reads
Combine reads
17,351,077 reads
Trim and map
3,473,413 reads
CLIP clusters
97.50% sense
UGUGUG
Cluster definition
e
12
8
TDP-43 clusters Randomly distributed
GUGU count clusters GUGU count
Motif Z score
augugu 158
uguagu 156
guaugu 146
ugugua 143
uaugug 138
guguau 137
9%
9%
10%
10%
57%
15%
12%
0
1
2
Motif Z score
ugugug
462
gugugu
453
100
Z score
300
3
f
Distal
intron
5’ UTR
46%
21%
11%
4
0
Direction of
transcription
in genes
39,961 clusters in
6,304 genes
Nrxn 3 intron 8
log2 (motif count)
articles
Proximal
intron
Not annotated
3’ UTR
2 kb Exon 2 kb
2% 1%
3%
1%
30%
>3
63%
TDP-43 CLIP
binding
4% 3%
28%
24%
27%
14%
Ago CLIP
binding
advance online publication nature neuroscience
articles
intronic clusters being >2 kb from the nearest exon-intron boundary
(Fig. 1f). This number rose to 82% for clusters >500 bases from the
nearest exon-intron boundary (Supplementary Fig. 2b). Such distal
intronic binding is in sharp contrast with published RNA-binding maps
for tissue-specific RNA binding proteins involved in alternative splicing, such as Nova or Fox214,15. The same analysis on published data in
mouse brain for the Argonaute proteins17,21, which are recruited by
microRNAs to the 3′ ends of genes in metazoans17,21, showed a substantially different pattern of binding. Only 24% (or 30%) of Argonaute
clusters resided within 2 kb (or 500 bases) of the nearest exon-intron
boundary, whereas 28% were in 3′ untranslated regions (3′ UTRs)
(Fig. 1f and Supplementary Fig. 2b). This prominent concentration
of Argonaute binding near 3′ ends is in stark contrast with the uniform
distribution of TDP-43–binding sites across the length of pre-mRNAs
(Supplementary Fig. 2c).
TDP-43 binds GU-rich distal intronic sites
Sequence motifs enriched in TDP-43–binding sites were determined by
comparing sequences in clusters with randomly selected regions of similar
sizes in the same protein-coding genes. Z score statistics revealed that the
most significantly enriched hexamers consisted of GU-repeats (Z > 450),
consistent with published in vitro results20, or a GU-rich motif interrupted
by a single adenine (Z = 137–158) (Fig. 1e). The majority (57%) of clusters contained at least four GUGU elements, as compared with only 9% RNAs altered after in vivo TDP-43 depletion in mouse brain
when equally sized clusters were randomly placed in the same pre-mRNAs To identify the contribution of TDP-43 in maintaining levels and
(Fig. 1e). Furthermore, the number of GUGU tetramers correlated with splicing patterns of RNAs, we injected two antisense oligonucleotides
the strength of binding, as estimated by the relative number of reads in each (ASOs) directed against TDP-43 and a control ASO with no target in
cluster per gene compared with all clusters in other genes (Supplementary the mouse genome into the striatum of normal adult mice (Fig. 2a).
Fig. 2a). Nevertheless, genome-wide analysis revealed that GU-rich Striatum is a well-defined structure that is amenable to accurate dissecrepeats were neither necessary nor sufficient to specify a TDP-43–binding tion and isolation, with TDP-43 expression levels being comparable to
site. One example is the left-most binding site in neurexin 3 (Fig. 1c), other brain regions. Stereotactic injections of ASOs that target TDP-43,
which does not have a GU motif, whereas a GU-rich motif 2-kb upstream control ASO or saline were performed in three groups of age- and sexof it was not bound by TDP-43. In fact, only ~3% of all transcribed matched adult C57BL/6 mice and were tolerated with minimal effects
300 nucleotide stretches containing more than
a
TDP-43 mRNA
three GUGU tetramers contained TDP-43 clusStereotactic injection of ASOs
AAAAAAAAAA
in the striatum of normal mice
ters by CLIP-seq, indicating that TDP-43 target
TDP-43–specific ASO
RNA-seq
genes cannot be identified by simply scanning
RNA
qRT-PCR
extraction
nucleotide sequences for GU-rich regions.
RT-PCR
RNase H
Although the vast majority (93%) of TDPAAAAAAAAAA
43 sites were in introns, we identified a bindProtein
ing preference of TDP-43 with most (63%)
Immunoblot
extraction
A
AA AA
Dissection of
the striatum
A
TDP-43 pre-mRNA
degradation
b
c
Control oligo
TDP-43 KD
Control oligo
342,150,462 reads
72 nt long (n = 4)
Saline
TDP-43
120
100
80
60
40
20
0
120
100
80
60
40
20
0
94.56% sense
qRT-PCR
e
2
4
6
8
Expression control (log2 RPKM)
10
qRT-PCR
100
80
60
40
20
0
Meg3/Gtl2
Rian
Malat1/Neat2
Xist
120
100
80
60
40
20
0
e
C
o
ol ntr
ig o
o l
TD
ol Pig 43
o
2
120
Saline
Control
RNA-seq
TDP-43 KD
Percentage Meg3/Gtl2
TDP-43
f
140
lin
6
4
11,852 expressed
genes
Saline
Sa
TDP-43 KD
Downregulated genes (239)
Upregulated genes (362)
0
94.33% sense
0
8
0
103,514,455 reads
Direction of
transcription in genes
Percentage ncRNA expression
10
127,725,000 reads
Densitometric
analysis
Control oligo
d
TDP-43 KD
294,180,255 reads
72 nt long (n = 4)
Trimmed
and mapped
GAPDH
Percentage Percentage
TDP-43
TDP-43
mRNA
protein
Figure 2 In vivo depletion of TDP-43 in mouse
brain with ASOs. (a) Strategy for depletion of
TDP-43 in mouse striatum. TDP-43–specific
or control ASOs were injected into the striatum
of adult mice. TDP-43 mRNA was degraded
via endogenous RNase H digestion, which
specifically recognizes ASO–pre-mRNA (DNA/
RNA) hybrids. Mice were killed after 2 weeks
and striata were dissected for RNA and protein
extraction. (b) Semi-quantitative immunoblot
demonstrating substantial depletion of
TDP-43 protein to 20% of controls in TDP-43
ASO–treated mice compared with saline- and
control ASO–treated mice. qRT-PCR showed
similar TDP-43 depletion at the mRNA level.
KD, knockdown. (c) Flow-chart illustrating the
number of reads sequenced and aligned from
the RNA-seq experiments. (d) Differentially
regulated genes identified by RNA-seq analysis.
Scatter-plot revealed the presence of 362 and
239 genes (diamonds) that were significantly
up- (red) or down- (green) regulated after
TDP-43 depletion. RNA-seq analysis confirmed
that TDP-43 levels were reduced to 20% of
control animals. (e) Normalized expression
(based on RPKM values from RNA-seq) of four
noncoding RNAs that were TDP-43 targets and
were downregulated after TDP-43 depletion.
(f) Quantitative RT-PCR validation of noncoding
RNA Meg3/Gtl2.
Expression TDP-43 knockdown
(log2 RPKM)
immunoprecipitation (RIP, an approach with the serious caveat that
the absence of cross-linking allows re-association of RNAs and RNAbinding proteins after cell lysis, as previously documented19) revealed
2,672 of the genes with CLIP-seq clusters in common. As expected from
our CLIP-seq analysis in whole brain, we found strong representation
of neuronal (see below) and glial mRNA targets, including glutamate
transporter 1 (Glt1), myelin-associated glycoprotein (Mag) and myelin
oligodendrocyte glycoprotein (Mog).
3
articles
Relative mRNA expression (fold change)
Cluster count
Mean cluster count in introns
Total intron length (kb)
b
on the survival of the animals. Mice were killed a
40
600 200
Upregulated
Downregulated
2
after 2 weeks and total RNA and protein from
Saline
Control oligo
35
500 150
TDP-43 KD
striata were isolated (Fig. 2a). Samples treated
30
1.5
100
with TDP-43 ASO showed a significant and
400
25
reproducible reduction of TDP-43 RNA and
50
300
20
1
protein to approximately 20% of normal levels
0
15
when compared with controls (P < 3 × 10–5;
200
10
0.5
Fig. 2b).
100
5
To explore the effects of TDP-43 downregulation on its target RNAs, we converted
0
0
0
Nrxn3 Park2 Nlgn1 Fgf14 Kcnd2 Cadps Efna5 Chat
Most
Most
Not
downregulated
poly-A–enriched RNAs from four biological
upregulated
regulated
replicates of TDP-43 or control ASO-treated, c
1.0 Mouse exons
Mouse introns
Human introns
as well as three saline-treated animals, to
P < 6 x 10
P < 5.3 x 10
0.9 P < 0.0872
0.8
cDNAs and sequenced them in a strand0.7
specific manner22, yielding an average of >25
0.6
0.5
million 72-bp reads per library. The number
0.4
of mapped reads per kilobase of exon, per milEnriched in brain
0.3
Not enriched in brain
lion mapped reads (RPKM) for each annotated
0.2
Random subset
0.1
Not in random subset
protein-coding gene was determined to estab0
12
10
10
10
10
10
10
10
10
10
lish a metric of normalized gene expression .
Median exon length (nt)
Median intron length (nt)
Median intron length (nt)
Hierarchical clustering of gene expression
values for the independent samples revealed Figure 3 Binding of TDP-43 on long transcripts enriched in brain sustains their normal mRNA levels.
high correlation (R2 = 0.96) between biologi- (a) Correlation between RNA-seq and CLIP-seq data. Genes were ranked on their degree of regulation
cal replicates of each condition (TDP-43 and after TDP-43 depletion (x axis) and the mean number of intronic CLIP clusters found in the next 100
control ASO/saline) (Supplementary Fig. 3a). genes from the ranked list were plotted (y axis, green line). Similarly, the mean total intron length for
the next 100 genes was plotted (y axis, red line). The cluster count for each upregulated gene (red dots)
Notably, all control ASO–treated samples were
and each downregulated gene (green dots) was plotted using the same ordering (inset). (b) qRT-PCR
clustered together, as were the samples from for selected downregulated genes with long introns (except for Chat) revealed a significant reduction
TDP-43 ASO–treated animals, consistent with of all transcripts when compared with controls (P < 8 × 10−3). Error bars represent s.d. calculated
TDP-43 reduction having an appreciable effect in each group for 3–5 biological replicates. (c) Cumulative distribution plots comparing exon length
(left) or intron length (middle) across mouse brain tissue–enriched genes (388 genes) and non-brain
on regulation of gene expression.
Reads of each treatment group were com- tissue–enriched genes (15,153 genes). Genes enriched in brain had significantly longer median intron
−6
bined, yielding greater than 100 million length compared with genes not enriched in brain (right, solid red line and black lines, P < 6.2 × 10
by two-sample Kolmogorov Smirnov goodness-of-fit hypothesis test), whereas a random subset of
uniquely mapped reads per condition (Fig. 2c). 388 genes showed no difference in intron length (dashed lines). Similar analysis across human brain
Approximately 70% (11,852) of annotated pro- tissue–enriched genes (387 genes) and non-brain tissue–enriched genes (17,985 genes) also showed
tein-coding genes in mouse satisfied at least 1 significantly longer introns in brain enriched genes (solid red and black lines, P < 5.3 × 10−6), whereas a
RPKM in either condition. Statistical compari- random subset of 387 genes showed no difference in intron length (dashed lines).
son revealed that 362 genes were significantly
upregulated and 239 downregulated on reduction of TDP-43 protein Of the RNAs containing TDP-43 clusters, 96% were unaffected by TDP(P < 0.05; Fig. 2d and Supplementary Tables 1 and 2). We found that 43 depletion, suggesting either that other RNA-binding proteins comTDP-43 itself was downregulated by RPKM analysis to 20% of the levels pensate for TDP-43 loss or that the remaining 20% of TDP-43 protein
in control treatments, consistent with quantitative RT-PCR (qRT-PCR) suffices to regulate these transcripts. For the 239 RNAs downregulated
measurements (Fig. 2b). RNAs unique to neurons (including doublecor- after TDP-43 depletion, a striking enrichment of multiple TDP-43 bindtin, the neuron-specific beta-tubulin and choline O-acetyl transferase ing sites was observed (Fig. 3). In fact, the 100 most downregulated
(Chat)) or glia (including glial fibrillary acidic protein, myelin binding genes contained an average of ~37 TDP-43–binding sites per pre-mRNA
protein, Glt1 and Mag) were highly represented in the RNA-seq data, con- and 12 genes had more than 100 clusters (Fig. 3a and Supplementary
firming assessment of RNA levels in multiple cell types, as expected.
Table 2). We did not observe this bias for multiple TDP-43 binding
Of the set of ~242 literature-curated murine noncoding RNAs23, sites if we randomized the order of genes (Supplementary Fig. 4a),
expression of 4 increased and expression of 55 decreased by more than if we ordered them by their expression levels (RPKM) in either treattwofold on TDP-43 depletion (P < 10−5; Supplementary Table 3). ment (Supplementary Fig. 4b,c), or if the Argonaute-binding sites were
Malat1/Neat2, Xist, Rian and Meg3 are noncoding RNA examples that plotted on genes ranked by their expression pattern on TDP-43 deplewere both decreased (Fig. 2e,f) and bound by TDP-43, consistent with tion (Supplementary Fig. 4d). Furthermore, this trend was significant
for TDP-43 clusters found in introns (Fig. 3a), but not in exons, 5′ or
TDP-43 having a direct role in regulating their expression.
3′ UTRs (Supplementary Fig. 4e–g).
To address whether TDP-43 binding enrichment in the downTDP-43 binding to long pre-mRNAs sustains their levels
RNA-seq and CLIP-seq datasets were integrated by first ranking all regulated genes could be attributed to intron size, we performed the
11,852 expressed genes by their degree of change on TDP-43 reduc- same analysis on the ranked list, but calculated total (Fig. 3a) or mean
tion compared with control treatment. For each group of 100 consec- (Supplementary Fig. 4h) intron size instead of cluster counts. We
utively ranked genes (starting from the most upregulated gene), we found that the most downregulated genes after TDP-43 reduction
determined the mean number of TDP-43 clusters. No enrichment in had exceptionally long introns that were more than sixfold longer
TDP-43 clusters in the upregulated genes was identified, indicating that (average of 28,707 bp; median length of 11,786 bp) than unaffected
their upregulation was likely an indirect consequence of TDP-43 loss. or upregulated genes (average of 4,532 bp, median length of 2,273 bp,
–6
Frequency
–6
2
4
3
4
2
3
4
2
3
4
articles
Table 1 The fraction of downregulated, but not upregulated TDP-43 targets increased with intron size
Mean intron
length (kb)
Expressed
genes
TDP-43 targets
Downregulated
genes
Downregulated
TDP-43 targets
Upregulated
genes
Upregulated
TDP-43 targets
0–1
2,566
510 (20%)
31
8 (26%)
100
7 (7%)
1–10
8,022
4,485 (56%)
80
47 (59%)
252
56 (32%)
10–100
1,238
1,027 (83%)
109
104 (95%)
10
3 (30%)
>100
26
26 (100%)
19
19 (100%)
0
0 (0%)
Total
11,852
6,048 (51%)
239
178 (74%)
362
66 (18%)
Number of TDP-43 targets among expressed, downregulated or upregulated genes, grouped by their mean intron length as indicated. The percentages in parentheses represent the
fraction of TDP-43 targets found in each group upon TDP-43 depletion.
P < 4 × 10−18 by t test). Again, this correlation of downregulation with
intron size was not observed for any of the control conditions mentioned above (Supplementary Fig. 4a–g). Indeed, the enrichment of
TDP-43 binding can be largely attributed to intron size differences,
as the number of TDP-43–binding sites per kilobase of intron length
(cluster density; Supplementary Fig. 4i) was only slightly increased
(P < 0.022) for downregulated versus unaffected or upregulated genes
(0.072 sites per kb downregulated genes, 0.059 sites per kb other genes).
Dividing all mouse protein-coding genes into four groups on the basis
of mean intron length (<1 kb, 1–10 kb, 10–100 kb and >100 kb) confirmed that the fraction of TDP-43 targets increased (from 20% to
100%) with intron size (Table 1). Indeed, 83% of genes that contained
average intron lengths of 10–100 kb and all 26 genes that contained
>100-kb-long introns were direct targets of TDP-43.
A highly significant fraction (74%) of all downregulated genes were
direct targets of TDP-43 in comparison with genes that were unchanged
(52%, P < 0.001) or upregulated (18%, P < 10−17) on TDP-43 depletion. Notably, all 19 downregulated genes with >100-kb-long introns
were direct TDP-43 targets. In strong contrast, no genes in the same
intron length category were upregulated on TDP-43 depletion and only
30% of upregulated genes with 10–100-kb-long introns were TDP-43
targets (Table 1). The crucial role of TDP-43 in maintaining the mRNA
abundance of long intron-containing genes was also reflected by the
downregulation after TDP-43 depletion of ~10% of genes with >10-kblong introns, the large majority (123 of 128, 96%) of which are direct
TDP-43 targets.
Gene ontology analysis showed that TDP-43 targets whose expression was downregulated after TDP-43 depletion were highly enriched
for synaptic activity and function (Supplementary Figs. 5 and 6 and
Supplementary Table 4). Notably, several genes with long introns targeted by TDP-43 have crucial roles in synaptic function and have been
implicated in neurological diseases, such as subunit 2A of the NMDA
receptor (Grin2a), the ionotropic glutamate receptor 6 (Grik2/GluR6),
the calcium-activated potassium channel alpha (Kcnma1), the voltagedependent calcium channel (Cacna1c), and the synaptic cell-adhesion
molecules neurexin 1 and 3 (Nrxn1, Nrxn3) and neuroligin 1 (Nlgn1).
We analyzed a compendium of expression array data from different
mouse organs and human tissues and found that genes preferentially
expressed in brain have significantly longer introns (P < 6 × 10−6), but
not exons (Fig. 3c). The length of these genes was not correlated with
the size of the respective proteins and the prevalence of long introns
was largely conserved between the corresponding mouse and human
genes. Although binding of TDP-43 in long introns can be explained
by the increased likelihood to contain UG repeats, the conservation
through evolution of this particular gene structure (Supplementary
Fig. 7) suggests that these exceptionally long introns contain important
regulatory elements.
To validate the RNA-seq results, we analyzed a selection of brainenriched TDP-43 targets containing long introns. Genome browser
views of neurexin 3 (Nrxn3), Parkin 2 (Park2), neuroligin 1 (Nlgn1),
fibroblast growth factor 14 (Fgf14), potassium voltage-gated channel
subfamily D member 2 (Kcnd2), calcium-dependent secretion activator (Cadps) and ephrin-A5 (Efna5) revealed a scattered distribution of
multiple TDP-43–binding sites across the full length of the pre-mRNA
(Supplementary Fig. 7a), consistent with the results of the global analysis (Fig. 3a). qRT-PCR verified the TDP-43–dependent reduction of
all the long transcripts that we tested (Fig. 3b). Chat had a median
intron size of <10 kb, with TDP-43 clusters restricted to a single intronic
site (Supplementary Fig. 7a). Nevertheless, qRT-PCR confirmed the
RNA-seq result of a significant reduction in Chat levels after TDP-43
depletion (P = 0.005; Fig. 3b).
Only 18% of the upregulated genes were direct targets of TDP43 (Table 1) and gene ontology analysis revealed an enrichment for
genes involved in the inflammatory response (Table 2), suggesting
that their differential expression is an indirect consequence of TDP43 loss. However, of the 66 upregulated RNAs that contained CLIPseq clusters, 29% harbored TDP-43–binding site(s) in their 3′ UTR,
a percentage that is twofold higher than that of downregulated genes
(Supplementary Fig. 8). This suggests that TDP-43 represses gene
expression when bound to 3′ UTRs.
Table 2 Main functional categories enriched within the TDP-43–regulated genes
GO
Downregulated genes (239)
categories
Molecular function
Cellular component
Biological process
GO term
GO:0006811
Ion transport
Upregulated genes (362)
Corrected P value
GO term
Corrected P value
2.31 × 10–8
GO:0006952
Defense response
2.28 × 10–20
10–7
GO:0007268
Synaptic transmission
1.79 ×
GO:0006954
Inflammatory response
4.97 × 10–12
GO:0045202
Synapse
5.33 × 10–8
GO:0000323
Lytic vacuole
1.70 × 10–10
GO:0005886
Plasma membrane
2.35 × 10–13
GO:0005764
Lysosome
1.70 × 10–10
GO:0005216
Ion channel activity
1.49 × 10–11
GO:0004197
Cysteine-type
endopeptidase activity
6.00 × 10–4
GO:0022803
Passive transmembrane 7.06 × 10–12
transporter activity
GO:0003950
NAD+ ADPribosyltransferase activity
1.11 × 10–2
Examples of the most significantly enriched Gene Ontology (GO) terms in the list of up- or downregulated genes on TDP-43 knockdown, as annotated. Downregulated genes,
the majority of which were direct targets of TDP-43, were enriched for synaptic activity and function. In contrast, upregulated genes, most of which were not bound by TDP-43,
encoded primarily inflammatory and immune response–related proteins. The P value indicated was corrected for multiple testing using the Benjamini-Hochberg method.
5
a
Constitutive
all exons
cassette
EST/mRNA
defined
Included
all exons
excluded
RNA-seq
defined
P < 3 x 10–6
P < 2 x 10–3
Included
array exons
excluded
Junction array
defined
P < 10–3
0
b
10
20
30
Percent of exons with TDP-43 binding within 2 kb
2 kb
Inclusion
76% (480 reads)
TDP-43 KD
Control
CLIP clusters
RNA-seq
control
19% (117 reads)
RNA-seq
TDP-43 KD
24% (147 reads)
Sort1
81% (498 reads)
Exclusion
c
Human exons orthologous to mouse
Human EST/mRNA
defined cassette exons
Human constitutive
exons
Human, other types
of alternative splicing
d
Exons included on
Exons excluded on
TDP-43 depletion in mouse TDP-43 depletion in mouse
3%
7%
12%
85%
36%
3
2
1
0
3
2
1
0
Control KD
Sort1
Sema3f
0.4
1.5
1
0.5
0
3
2
1
0
Pdp1
2
1
0
Ddef2
2
1
0
Mpzl1
69%
Exons excluded on TDP-43 depletion
Control KD
Kcnd3
All mouse exons
on junction array
9%
22%
57%
Exons included on TDP-43 depletion
0.2
TDP-43 mediates alternative splicing of
0
its mRNA targets
Atp2b2
8
Although TDP-43–binding sites were enriched
4
in distal introns (Fig. 1f), 11% (21,041 out of
0
Eif4h
190,161) of all mouse exons, including both
0.8
constitutive and alternative exons, contained
0.4
TDP-43–binding site(s) in a 2-kb window
0
Dnajc5
extending from the 5′ and 3′ exon-intron
boundaries (Fig. 4a). Compared with all exons,
TDP-43 clusters were significantly enriched (P < 8 × 10−3) around exons
with transcript evidence for either alternative inclusion or exclusion
(that is, cassette exons). Of the 8,637 known mouse cassette exons,
15.1% contained TDP-43–binding sites in the exon or intron within
2 kb of the splice sites. A splice index score for all exons, a measure
similar to the ‘percent spliced in’ (or ψ) metric24, was determined by
the number of reads that mapped on exons as well as reads that mapped
at exon junctions (Fig. 4b). This analysis resulted in the identification
of 203 cassette exons that were differentially included (93) or excluded
(110) (P < 0.01) when TDP-43 was depleted. Notably, sortilin 1, the gene
encoding the receptor for progranulin25,26, had the highest splice index
score, with exon 18 exclusion requiring TDP-43 (Fig. 4b). Included exons
(P < 3 × 10−6) and, to a lesser extent, excluded exons (P < 2 × 10−3) identified by RNA-seq were significantly enriched (~2.7-fold and ~2.0-fold,
respectively) for TDP-43 binding when compared to all mouse exons
(Fig. 4a). Only 33% of RNA-seq–verified TDP-43–regulated cassette
exons had previous expressed sequence tag (EST)/mRNA evidence for
alternative splicing, demonstrating that our approach identified previously unknown alternative splicing events.
6
P < 8 x 10–3
12
8
4
0
Ratio inlcusion to exclusion
Figure 4 TDP-43 mediates alternative splicing
regulation of its RNA targets. (a) Schematic
representation of different exon classes defined
by EST and mRNA libraries, RNA-seq or splicingsensitive microarray data as indicated (left).
Right, bar plot displaying the percentage of
exons that contain TDP-43 clusters within 2 kb
upstream and downstream of the exon-intron
junctions. (b) Example of alternative splicing
change on exon 18 of sortilin 1 analyzed using
RNA-seq reads mapping to the exon body (left).
Arrows depict increased density of reads in the
TDP-43 knockdown samples compared with
controls. 76% of the spliced-junction reads
in the TDP-43 knockdown samples supported
inclusion versus only 19% in the control
oligo–treated samples (right). (c) Comparison
of mouse cassette exons detected by splicingsensitive microarrays to conserved exons in
human orthologous genes. 85% and 57% of
human exons corresponding to the excluded
and included mouse exons, after TDP-43
depletion, respectively (left and middle pie
charts), contained EST and mRNA evidence for
alternative splicing. As a control, the percentage
of human exons orthologous to all mouse exons
represented on the splicing-sensitive array
and that have alternative splicing evidence
are shown in the right pie chart. (d) Semiquantitative RT-PCR analyses of selected targets
confirmed alternative splicing changes in TDP-43
knockdown samples compared with controls.
Right, representative acrylamide gel pictures
of RT-PCR products from control or knockdown
adult brain samples. Quantification of splicing
changes from three biological replicates per
group (left); error bars represent s.d.
Ratio inlcusion to exclusion
articles
3
2
1
0
Control KD
4
2
0
4
2
0
Control KD
2
Ppp3ca
Dclk1
0
15
10
5
0
Dst
Tsn
4
4
2
0
4
2
Rbm33
ex3
ex2&3
0
Poldip3
Control (n = 3)
TDP-43 KD (n = 3)
Kcnip2
As an independent method of identifying TDP-43–regulated exons,
RNAs from the same ASO-treated animals were analyzed on customdesigned splicing-sensitive Affymetrix microarrays27. Using a conservative statistical cutoff, we detected 779 alternatively spliced events that
significantly changed on TDP-43 depletion (Supplementary Fig. 9).
Notably, included (P < 10−3), but not excluded, exons (P < 0.3) were significantly enriched for TDP-43 binding (~1.8-fold and ~1.3-fold, respectively) when compared with the unchanged exons on the microarray
(Fig. 4a), similar to the trend seen by RNA-seq. The combined RNA-seq
and splicing-sensitive microarray data defined a set of 512 alternatively
spliced cassette exons whose splicing was affected by loss of TDP-43. The
majority of human orthologs of these murine exons (85% of those with
excluded and 57% with included exons) had prior EST/mRNA evidence
for alternative splicing (Fig. 4c).
Semi-quantitative RT-PCR on selected RNAs validated splicing alterations with more inclusion or exclusion after TDP-43 reduction (Fig. 4d
and Supplementary Fig. 10). Notably, varying the extent of TDP-43
downregulation (between 40–80%) correlated with the magnitude of
splicing changes (Supplementary Fig. 11). However, the majority of
articles
b
5 kb
120
P < 4 x 10 –3
WT
80
GAPDH
60
100
80
60
40
20
0
40
20
0
d
WT TDP-43 Tg
Tet-on: 24 h
GFP–TDP-43: – +
+
GFP-TDP-43
Endogenous
TDP-43
30
25
UGUGUG
1
2
3
Tardbp
WT TDP-43 Tg
48 h
–
58
46
CLIP clusters
TDP-43 Tg
myc–TDP-43
end. TDP-43
100
CLIP reads
combined
c
Percentage
endogenous
TDP-43 protein
a
Percentage endogenous
TDP-43 mRNA
altered splicing events observed on TDP-43
depletion do not have TDP-43 clusters within
2 kb of the splice sites, implicating longerrange interactions or indirect effects of TDP43 through other splicing factors. Consistent
with this latter hypothesis, we identified TDP43 binding on pre-mRNAs of RNA-binding
proteins, including Fus/Tls, Ewsr1, Taf15,
Adarb1, Cugbp1, RBFox2 (Rbm9), Tia1, Nova
1 and 2, Mbnl, and neuronal Ptb (or Ptbp2).
After TDP-43 depletion, mRNA levels of
Fus/Tls and Adarb1 were reduced and exon 5
in the Tia1 transcript was more included,
whereas exon 10 in Ptbp2 was more excluded
(Supplementary Table 5).
GAPDH
TDP-43 autoregulation through binding
on its 3′ UTR
e isoform 3–specific primers f
g
P < 3 x 10
We found TDP-43–binding sites in an alterP < 2 x 10
180
TDP-43 3’ UTR 3
Unrelated 3’ UTR
40
natively spliced intron in the 3′ UTR of the
160
Short TDP-43 3’ UTR
P < 10
TDP-43 pre-mRNA (Fig. 5a). This binding
140
Long TDP-43 3’ UTR
30
100 P < 10
120
does not coincide with a long stretch of UG
P < 10
80
100
20
repeats, suggesting a lower strength of bindP < 10
60
80
ing (Supplementary Fig. 2a), consistent with a
10
40
60
recent report28. TDP-43 mRNAs spliced at this
40
20
0
20
0
site (Fig. 5a) are predicted to be substrates for
myc–TDP-43–HA – + – +
Tet-on: 48 24 48
0
nonsense-mediated RNA decay (NMD), a prosiRNA UPF1 – – + +
GFP–TDP-43: – +
+
No co-transfection
RFP
myc–TDP-43–HA siRNA UPF1
cess that targets mRNAs for degradation when
exon-junction complexes (EJCs) deposited Figure 5 Autoregulation of TDP-43 through binding on the 3′ UTR of its own transcript. (a) CLIP-seq
during splicing, located 3′ of the stop codon, reads and clusters on TDP-43 transcript showing binding mainly in an alternatively spliced part of the
are not displaced during the pioneer round 3′ UTR, lacking long uninterrupted UG repeats. (b) qRT-PCR showing ~50% reduction of endogenous
of translation29. In contrast, TDP-43 mRNAs TDP-43 mRNA in transgenic mice overexpressing human myc–TDP-43 not containing introns and
3′ UTR. (c) Immunoblots confirming the reduction of endogenous TDP-43 protein (upper panel)
with an unspliced 3′ UTR would not have such to 50% of control levels (by densitometry, lower panel) in response to human myc–TDP-43
a premature termination codon and should overexpression. (d) Immunoblot showing reduction of endogenous TDP-43 protein in HeLa cells on
escape NMD. This TDP-43 binding implies tetracycline induction of a transgene encoding GFP-myc–TDP-43–HA (annotated GFP–TDP-43). The
autoregulatory mechanisms reminiscent to ~30-kDa product accumulating after 48 h (red arrow) was immunoreactive with four TDP-43–specific
those reported for other RNA-binding pro- antibodies. (e) qRT-PCR using primers spanning the junctions of TDP-43 isoform 3 (upper panel),
teins30,31. Indeed, expression in mice of a TDP- showed ~100-fold increase in response to tetracycline induction of GFP–TDP-43. TDP-43 isoform 3
was present at very low levels before tetracycline induction. (f) Luciferase assays showing significant
43–encoding transgene without the regulatory reduction of relative fluorescence units (RFUs) in cells expressing renilla luciferase under the control
3′ UTR (E.S., S.-C.L. and D.W.C., unpublished of long TDP-43 3′ UTR when co-transfected with myc–TDP-43–HA. Cells treated with siRNA against
observations) led to significant reduction of UPF1 showed a significant increase of RFUs when expressing Luciferase with the long TDP-43
endogenous TDP-43 mRNA and protein 3′ UTR, but not in controls. (g) qRT-PCR scoring levels of endogenous TDP-43 isoform 3 in HeLa cells
transiently transfected with myc–TDP-43–HA, UPF1 siRNA or both simultaneously. TDP-43 isoform 3
(P < 4 × 10–3; Fig. 5b,c) in the CNS.
To identify the molecular basis of this mech- was increased in response to elevated TDP-43 protein levels, blocking of NMD (by UPF1 siRNA) and
there was a synergistic effect in the combined condition.
anism, we generated HeLa cells in which GFPmyc–TDP-43–HA mRNA lacking introns and
3′ UTR was transcribed from a single copy, tetracycline-inducible trans- sites) of the TDP-43 3′ UTR downstream of the stop codon of a renilla
gene inserted at a predefined locus by site-directed (Flp) recombinase16. luciferase gene (Supplementary Fig. 12a). Both unspliced and spliced
After 24 or 48 h of GFP-myc–TDP-43–HA induction, a substantial 3′ UTRs were determined to be present in brain RNAs from mouse and
reduction of endogenous TDP-43 protein was observed, accompanied human CNSs (Supplementary Fig. 12a). Both variants, as well as an
by accumulation of a shorter, ~30-kDa product (Fig. 5d) recognized unaltered luciferase reporter, were transfected into HeLa cells along with
by four different TDP-43–specific antibodies. Although this ~30-kDa plasmids driving either increased TDP-43 expression or red fluorescent
band could be derived from the transgene encoding TDP-43, it was protein (RFP; Supplementary Fig. 12b). Increased levels of TDP-43 pronot recognized by antibodies to myc or HA and its size is compatible tein led to a significant reduction of luciferase produced from the gene
with the endogenous TDP-43 isoform 3. Using qRT-PCR with primers carrying the long, intron-containing TDP-43 3′ UTR when compared with
spanning the exon junctions of TDP-43 isoform 3, we found a ~100-fold the short or unrelated 3′ UTR (P < 10–4; Fig. 5f). Moreover, co-transfection
increase of the spliced isoform 3 on overexpression of GFP-myc–TDP- of the reporters with siRNAs targeting UPF1 (Supplementary Fig. 12c),
43–HA protein (Fig. 5e).
an essential component that marks an NMD substrate for degradaTo test whether TDP-43 drives splicing of its pre-mRNA through bind- tion32, enhanced luciferase produced by the intron-containing 3′ UTR
ing to its 3′ UTR, we cloned a long unspliced version (containing the TDP- by ~1.5-fold, indicating that UPF1-dependent degradation of this mRNA
43–binding sites) and a short spliced version (without TDP-43–binding occurred (Fig. 5f). Lastly, the endogenous spliced isoform 3 of TDP-43
–2
–4
–4
Fold change endogenous
isoform 3 TDP-43 mRNA
–6
–12
Percentage of control
(normalized to RFU)
Fold change endogenous
isoform 3 TDP-43 mRNA
–3
7
articles
a
CLIP reads
combined
CLIP clusters
RNA-seq
Control
RNA-seq
TDP-43 KD
was increased, not only on elevated TDP-43 expression (by transient transfection), but also on blocking of NMD, with a synergistic effect in the
combined conditions (Fig. 5g).
8
KD/control = 0.7
Fus/Tls
Mammalian
conservation
c
120
Control oligo
TDP-43 KD
Control oligo
TDP-43 KD
100
FUS/TLS
80
GAPDH
60
40
20
0
Saline Control TDP-43
oligo
KD
d
2 kb
CLIP reads
combined
CLIP clusters
RNA-seq
Control
RNA-seq
TDP-43 KD
KD/control = 3
120
80
40
0
e
Fold change Grn mRNA
b
Percentage
FUS/TLS protein
TDP-43 regulates expression of disease-related transcripts
TDP-43 protein bound and directly regulated a variety of transcripts
involved in neurological diseases (Supplementary Table 6 and
Supplementary Figs. 8 and 13), including Fus/Tls and Grn (Fig. 6),
which encode FUS/TLS and progranulin, mutations in which cause
ALS10,11 or FTLD33,34, respectively. TDP-43 bound to the 3′ UTR of
Fus/Tls mRNA and to introns 6 and 7, all of which are highly conserved
between mammalian species (Fig. 6a). Gene annotation and the presence of RNA-seq reads in these introns were consistent with either
an alternative 3′ UTR or intron retention. Fus/Tls mRNA and protein
were reduced to approximately 40% of their normal levels (Fig. 6b,c).
Progranulin mRNA, on the other hand, was markedly increased by
~3–6-fold compared with controls (Fig. 6d,e).
CLIP-seq data also confirmed TDP-43 binding to two RNAs that were
previously reported to be associated with TDP-43: histone deacetylase
6 (Hdac6)35 and low–molecular weight neurofilament subunit (Nefl)36.
Our RNA-seq data indicate that HDAC6, which functions to promote
the degradation of polyubiquitinated proteins, was reduced on TDP-43
depletion (Supplementary Fig. 14a,b), albeit to a lesser degree in vivo
than previously reported in cell culture35. It has been known for many
years that Nefl mRNA levels are reduced in degenerating motor neurons
from individuals with ALS37. We identified TDP-43 clusters in the 3′ UTR
of Nefl (Supplementary Fig. 14c). In addition, RNA-seq data confirmed
that the mouse Nefl 3′ UTR was longer than annotated and Nefl mRNA
levels were slightly reduced on TDP-43 depletion (Supplementary
Fig. 14d). Multiple TDP-43–binding sites were also present in the premRNA from the Mapt gene encoding tau, whose mutation or altered
splicing of exon 10 has been implicated in FTD38. However, neither the
levels nor the splicing pattern of Mapt RNA were affected by TDP-43
depletion (Supplementary Table 6). Moreover, we identified multiple
TDP-43 intronic binding sites in the Hdh transcript (Supplementary
Fig. 13), which encodes huntingtin, the protein whose polyglutamine
expansion causes Huntington’s disease in humans39, accompanied
by cytoplasmic TDP-43 accumulations40. Moreover, Hdh levels were
decreased in mouse brain on TDP-43 depletion (Supplementary Table 6).
In contrast, we found no evidence for direct binding or TDP-43 regulation of the Sod1 transcript (Supplementary Table 6), whose aberrant
splicing in familial ALS cases41,42 has raised the possibility that SOD1
missplicing may be involved in the pathogenesis of sporadic ALS.
5 kb
Percentage
FUS/TLS mRNA
Figure 6 TDP-43 regulates expression of FUS/TLS and progranulin.
(a) CLIP-seq reads and clusters on Fus/Tls transcript showing TDP-43
binding in introns 6 and 7 (also annotated as an alternative 3′ UTR) and
the canonical 3′ UTR. RNA-seq reads from control or TDP-43 knockdown
samples (equal scales) showed a slight reduction in Fus/Tls mRNA in the
TDP-43 knockdown group and expression values from RNA-seq confirmed
that Fus/Tls mRNA was reduced to 70% of control on TDP-43 reduction.
(b) qRT-PCR confirmed Fus/Tls mRNA downregulation on TDP-43 depletion.
s.d. was calculated within each group for three biological replicas. (c) Semiquantitative immunoblot (upper panel) revealed a slight, but consistent,
reduction of FUS/TLS protein in ASO-treated mice to ~70% of control
levels, as quantified by densitometric analysis (lower panel). (d) CLIP-seq
reads and clusters on progranulin transcript (Grn) showing a sharp binding
in the 3′ UTR. RNA-seq reads from control or TDP-43 knockdown samples
(equal scales) showed a significant increase in progranulin mRNA in the
TDP-43 knockdown group compared with controls. (e) qRT-PCR confirmed
the statistically significant increase (P < 3 × 10−4) in progranulin mRNA in
samples with reduced TDP-43 levels when compared with controls. s.d. was
calculated in each group for three biological replicas.
7
6
5
4
3
2
1
0
Saline Control TDP-43
oligo
KD
Grn
DISCUSSION
TDP-43 is a central component in the pathogenesis of an ever-increasing list of neurodegenerative conditions. We created a genome-wide
RNA map of >39,000 TDP-43–binding sites in the mouse transcriptome and determined that levels of 601 mRNAs and splicing patterns
of 965 mRNAs were altered following TDP-43 reduction in the adult
nervous system. Thus, although earlier efforts have implicated TDP43 as a splicing regulator of a few candidate genes20,43,44, our RNA-seq
and microarray results establish that TDP-43 regulates the largest set
(512) of cassette exons thus far reported, demonstrating its broad role in
articles
alternative splicing regulation. We also found that TDP-43 was required
for maintenance of 42 non-overlapping, noncoding RNAs.
These findings also provide a direct test for how the nuclear loss
of TDP-43 widely reported in the remaining motor neurons in ALS
autopsy samples1 may contribute to neuronal dysfunction, independent
of potential damage from TDP-43 aggregates. The results of our experiments using in vivo reduction of TDP-43 coupled with RNA sequencing
indicate that TDP-43 is crucial for sustaining levels of 239 mRNAs,
including those encoding synaptic proteins, the choline acetyltransferase, and the disease-related proteins FUS/TLS and progranulin. A
substantial proportion of these pre-mRNAs are directly bound by TDP43 at multiple sites in exceptionally long introns, a feature that we found
most prominently in brain-enriched transcripts (Fig. 3), thereby identifying one component of neuronal vulnerability from TDP-43 loss.
A plausible model for the role of TDP-43 in sustaining the levels of
mRNAs derived from long pre-mRNAs is that TDP-43 binding in long
introns prevents unproductive splicing events that would introduce
premature stop codons and thereby promote RNA degradation. Our
results thus identify a previously unknown conserved role for TDP-43 in
regulating a subset of these long intron-containing brain-enriched genes.
None of our evidence eliminates the possibility that TDP-43 affects RNA
levels by additional mechanisms, such as through transcription regulation or by facilitation of RNA polymerase elongation, similar to what has
been shown for another splicing regulator, SC35 (ref. 45).
FUS/TLS is another RNA-binding protein whose mutation causes
ALS10,11 and, in some rare cases, FTLD-U. Similar to TDP-43, FUS/
TLS aggregation has been observed in different neurodegenerative
conditions, including Huntington’s disease and spinocerebellar ataxia
(reviewed in ref. 3). We have now shown that Fus/Tls mRNA is a direct
target of TDP-43 and its level is reduced on TDP-43 depletion (Fig. 6),
thereby identifying a previously unknown FUS/TLS dependency on
TDP-43. The latter is also true for two additional disease-relevant proteins, progranulin and its proposed receptor sortilin. Progranulin levels
were sharply increased on TDP-43 reduction and splicing of sortilin was
altered. In fact, TDP-43 directly binds and regulates the levels and splicing patterns of transcripts implicated in various neurologic diseases46
(Supplementary Table 6 and Supplementary Fig. 13), consistent with
a broad role of TDP-43 in these conditions3.
Finally, in contrast with a recent study that reported an autoregulatory
mechanism for TDP-43 that is independent of pre-mRNA splicing28,
our results indicate that TDP-43 acts as a splicing regulator to reduce its
own expression level by binding to the 3′ UTR of its own pre-mRNA.
Although there may be additional mechanisms beyond NMD28, we found
that TDP-43 enhanced splicing of an alternative intron in its own 3′ UTR,
thereby autoregulating its levels through a mechanism that involves splicing-dependent RNA degradation by NMD (Fig. 5). TDP-43 autoregulation occurs in the mammalian CNS, as seen by a significant reduction of
endogenous TDP-43 mRNA and protein in response to the expression of
a TDP-43 transgene lacking the regulatory intron, as we found and others
have reported previously47,48. Both spliced and unspliced TDP-43 RNAs
were found in human and mouse brain, consistent with autoregulation
at normal TDP-43 levels that substantially attenuates TDP-43 synthesis
through production of unstable, spliced RNA.
TDP-43-dependent splicing of its 3′ UTR intron as a key component
of a TDP-43 autoregulatory loop could participate in a feedforward
mechanism enhancing the cytoplasmic TDP-43 aggregates that are hallmarks of familial and sporadic ALS. Following an initiating insult (for
example, one that traps some TDP-43 in initial cytoplasmic aggregates),
the reduction in nuclear TDP-43 levels would decrease splicing of its 3′
UTR, which would in turn produce an elevated pool of stable TDP-43
mRNA. Repeated translation of that TDP-43 mRNA would increase
synthesis of new TDP-43 in the cytoplasm whose subsequent co-aggregation into the initial complexes would drive their growth. Disrupted
autoregulation, by any event that lowers nuclear TDP-43, thus provides
a mechanistic explanation for what may be a critical, intermediate step
in the molecular mechanisms underlying age-dependent degeneration
and death of neurons in TDP-43 proteinopathies.
METHODS
Methods and any associated references are available in the online version of the paper at http://www.nature.com/natureneuroscience/.
Accession codes. Microarray CEL files and sequenced reads have been deposited
at the Gene Expression Omnibus database repository and the NCBI Short Read
Archive, respectively. All our data (microarray, RNA-seq and CLIP-seq) are
combined as a ‘SuperSeries entry’ under one accession number GSE27394.
Note: Supplementary information is available on the Nature Neuroscience website.
ACKNOWLEDGMENTS
The authors would like to thank members of B. Ren’s laboratory, especially Z. Ye,
S. Kuan and L. Edsall for technical help with the Illumina sequencing and
U. Wagner for helpful discussions, K. Clutario and J. Boubaker for technical help,
as well as all of the members of the Yeo and Cleveland laboratories, M. Ares Jr
for generous support, and the neuro-team of ISIS Pharmaceuticals for critical
comments and suggestions on this project. M.P. is the recipient of a Human
Frontier Science Program Long Term Fellowship. C.L.-T. is the recipient of the
Milton-Safenowitz post-doctoral fellowship from the Amyotrophic Lateral
Sclerosis Association. D.W.C. receives salary support from the Ludwig Institute
for Cancer Research. S.C.H. is funded by a National Science Foundation Graduate
Research Fellowship. This work was supported by grants from the US National
Institutes of Health (R37 NS27036 and an American Recovery and Reinvestment
Act Challenge grant) to D.W.C and partially by grants from the US National
Institutes of Health (HG004659 and GM084317) and the Stem Cell Program at the
University of California, San Diego to G.W.Y.
AUTHOR CONTRIBUTIONS
M.P., C.L.-T., J.M. and T.Y.L. performed the experiments. K.R.H., S.C.H. and T.Y.L.
conducted the bioinformatics analysis. S.-C.L. developed the monoclonal TDP-43–
specific antibody used for CLIP-seq and the tetracycline-inducible GFP–TDP-43–
expressing HeLa cells. S.-C.L. and E.S. generated the transgenic myc–TDP-43 mice.
J.P.D. and L.S. conducted the preliminary splice-junction microarray analyses. M.P.,
C.L.-T., E.W., C.M., Y.S., C.F.B. and H.K. conducted the antisense oligonucleotide
experiments. M.P., C.L.-T., K.R.H., G.W.Y. and D.W.C. designed the experiments.
M.P., C.L.-T., K.R.H., S.C.H., G.W.Y. and D.W.C. wrote the paper.
COMPETING FINANCIAL INTERESTS
The authors declare no competing financial interests.
Published online at http://www.nature.com/natureneuroscience/.
Reprints and permissions information is available online at http://www.nature.com/
reprintsandpermissions/.
1. Neumann, M. et al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and
amyotrophic lateral sclerosis. Science 314, 130–133 (2006).
2. Arai, T. et al. TDP-43 is a component of ubiquitin-positive tau-negative inclusions
in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Biochem.
Biophys. Res. Commun. 351, 602–611 (2006).
3. Lagier-Tourenne, C., Polymenidou, M. & Cleveland, D.W. TDP-43 and FUS/TLS: emerging roles in RNA processing and neurodegeneration. Hum. Mol. Genet. 19, R46–R64
(2010).
4. Gitcho, M.A. et al. TDP-43 A315T mutation in familial motor neuron disease. Ann.
Neurol. 63, 535–538 (2008).
5. Kabashi, E. et al. TARDBP mutations in individuals with sporadic and familial amyotrophic lateral sclerosis. Nat. Genet. 40, 572–574 (2008).
6. Sreedharan, J. et al. TDP-43 mutations in familial and sporadic amyotrophic lateral
sclerosis. Science 319, 1668–1672 (2008).
7. Van Deerlin, V.M. et al. TARDBP mutations in amyotrophic lateral sclerosis with
TDP-43 neuropathology: a genetic and histopathological analysis. Lancet Neurol. 7,
409–416 (2008).
8. Buratti, E. et al. Nuclear factor TDP-43 can affect selected microRNA levels.
FEBS J. 277, 2268–2281 (2010).
9. Cooper, T.A., Wan, L. & Dreyfuss, G. RNA and Disease. Cell 136, 777–793 (2009).
10.Kwiatkowski, T.J. Jr. et al. Mutations in the FUS/TLS gene on chromosome 16 cause
familial amyotrophic lateral sclerosis. Science 323, 1205–1208 (2009).
9
articles
11.Vance, C. et al. Mutations in FUS, an RNA processing protein, cause familial amyotrophic lateral sclerosis type 6. Science 323, 1208–1211 (2009).
12.Mortazavi, A., Williams, B.A., McCue, K., Schaeffer, L. & Wold, B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat. Methods 5, 621–628 (2008).
13.Ule, J. et al. CLIP identifies Nova-regulated RNA networks in the brain. Science 302,
1212–1215 (2003).
14.Licatalosi, D.D. et al. HITS-CLIP yields genome-wide insights into brain alternative
RNA processing. Nature 456, 464–469 (2008).
15.Yeo, G.W. et al. An RNA code for the FOX2 splicing regulator revealed by mapping
RNA-protein interactions in stem cells. Nat. Struct. Mol. Biol. 16, 130–137 (2009).
16.Ling, S.C. et al. ALS-associated mutations in TDP-43 increase its stability and promote
TDP-43 complexes with FUS/TLS. Proc. Natl. Acad. Sci. USA 107, 13318–13323
(2010).
17.Zisoulis, D.G. et al. Comprehensive discovery of endogenous Argonaute binding sites
in Caenorhabditis elegans. Nat. Struct. Mol. Biol. 17, 173–179 (2010).
18.Sephton, C.F. et al. Identification of neuronal RNA targets of TDP-43-containing
ribonucleoprotein complexes. J. Biol. Chem. 286, 1204–1215 (2011).
19.Mili, S. & Steitz, J.A. Evidence for reassociation of RNA-binding proteins after cell
lysis: implications for the interpretation of immunoprecipitation analyses. RNA 10,
1692–1694 (2004).
20.Buratti, E. et al. Nuclear factor TDP-43 and SR proteins promote in vitro and in vivo
CFTR exon 9 skipping. EMBO J. 20, 1774–1784 (2001).
21.Chi, S.W., Zang, J.B., Mele, A. & Darnell, R.B. Argonaute HITS-CLIP decodes
microRNA-mRNA interaction maps. Nature 460, 479–486 (2009).
22.Parkhomchuk, D. et al. Transcriptome analysis by strand-specific sequencing of
complementary DNA. Nucleic Acids Res. 37, e123 (2009).
23.Pang, K.C. et al. RNAdb—a comprehensive mammalian noncoding RNA database.
Nucleic Acids Res. 33, D125–D130 (2005).
24.Wang, E.T. et al. Alternative isoform regulation in human tissue transcriptomes. Nature
456, 470–476 (2008).
25.Hu, F. et al. Sortilin-mediated endocytosis determines levels of the frontotemporal
dementia protein, progranulin. Neuron 68, 654–667 (2010).
26.Carrasquillo, M.M. et al. Genome-wide screen identifies rs646776 near sortilin as a regulator of progranulin levels in human pasma. Am. J. Hum. Genet. 87, 890–897 (2010).
27.Du, H. et al. Aberrant alternative splicing and extracellular matrix gene expression in
mouse models of myotonic dystrophy. Nat. Struct. Mol. Biol. 17, 187–193 (2010).
28.Ayala, Y.M. et al. TDP-43 regulates its mRNA levels through a negative feedback loop.
EMBO J. 30, 277–288 (2011).
29.Le Hir, H., Moore, M.J. & Maquat, L.E. Pre-mRNA splicing alters mRNP composition:
evidence for stable association of proteins at exon-exon junctions. Genes Dev. 14,
1098–1108 (2000).
30.Wollerton, M.C., Gooding, C., Wagner, E.J., Garcia-Blanco, M.A. & Smith, C.W.
Autoregulation of polypyrimidine tract binding protein by alternative splicing leading
to nonsense-mediated decay. Mol. Cell 13, 91–100 (2004).
10
31.Sureau, A., Gattoni, R., Dooghe, Y., Stevenin, J. & Soret, J. SC35 autoregulates its
expression by promoting splicing events that destabilize its mRNAs. EMBO J. 20,
1785–1796 (2001).
32.Perlick, H.A., Medghalchi, S.M., Spencer, F.A., Kendzior, R.J. Jr. & Dietz, H.C.
Mammalian orthologs of a yeast regulator of nonsense transcript stability. Proc. Natl.
Acad. Sci. USA 93, 10928–10932 (1996).
33.Baker, M. et al. Mutations in progranulin cause tau-negative frontotemporal dementia
linked to chromosome 17. Nature 442, 916–919 (2006).
34.Cruts, M. et al. Null mutations in progranulin cause ubiquitin-positive frontotemporal
dementia linked to chromosome 17q21. Nature 442, 920–924 (2006).
35.Fiesel, F.C. et al. Knockdown of transactive response DNA-binding protein (TDP-43)
downregulates histone deacetylase 6. EMBO J. 29, 209–221 (2010).
36.Strong, M.J. et al. TDP43 is a human low molecular weight neurofilament (hNFL)
mRNA-binding protein. Mol. Cell. Neurosci. 35, 320–327 (2007).
37.Bergeron, C. et al. Neurofilament light and polyadenylated mRNA levels are decreased
in amyotrophic lateral sclerosis motor neurons. J. Neuropathol. Exp. Neurol. 53,
221–230 (1994).
38.Hutton, M. et al. Association of missense and 5′-splice-site mutations in tau with the
inherited dementia FTDP-17. Nature 393, 702–705 (1998).
39.MacDonald, M.E. et al. A novel gene containing a trinucleotide repeat that is expanded
and unstable on Huntington’s disease chromosomes. Cell 72, 971–983 (1993).
40.Schwab, C., Arai, T., Hasegawa, M., Yu, S. & McGeer, P.L. Colocalization of transactivation-responsive DNA-binding protein 43 and huntingtin in inclusions of Huntington
disease. J. Neuropathol. Exp. Neurol. 67, 1159–1165 (2008).
41.Valdmanis, P.N. et al. A mutation that creates a pseudoexon in SOD1 causes familial
ALS. Ann. Hum. Genet. 73, 652–657 (2009).
42.Birve, A. et al. A novel SOD1 splice site mutation associated with familial ALS revealed
by SOD activity analysis. Hum. Mol. Genet. 19, 4201–4206 (2010).
43.Mercado, P.A., Ayala, Y.M., Romano, M., Buratti, E. & Baralle, F.E. Depletion of TDP
43 overrides the need for exonic and intronic splicing enhancers in the human apoA-II
gene. Nucleic Acids Res. 33, 6000–6010 (2005).
44.Dreumont, N. et al. Antagonistic factors control the unproductive splicing of SC35
terminal intron. Nucleic Acids Res. 38, 1353–1366 (2009).
45.Lin, S., Coutinho-Mansfield, G., Wang, D., Pandit, S. & Fu, X.D. The splicing factor
SC35 has an active role in transcriptional elongation. Nat. Struct. Mol. Biol. 15,
819–826 (2008).
46.Tollervey, J.R. et al. Characterizing the RNA targets and position-dependent splicing
regulation by TDP-43. Nat. Neurosci. advance online publication, doi:10.1038/nn.2778
(27 February 2011).
47.Xu, Y.F. et al. Wild-type human TDP-43 expression causes TDP-43 phosphorylation,
mitochondrial aggregation, motor deficits, and early mortality in transgenic mice.
J. Neurosci. 30, 10851–10859 (2010).
48.Igaz, L.M. et al. Dysregulation of the ALS-associated gene TDP-43 leads to neuronal
death and degeneration in mice. J. Clin. Invest. 121, 726–738 (2011).
ONLINE METHODS
CLIP-seq library preparation and sequencing. Brains from 8-week-old female
C57Bl/6 mice were rapidly dissociated by forcing through a cell strainer with a
pore size of 100 µm (BD Falcon) before ultraviolet crosslinking. CLIP-seq libraries were constructed as previously described15, using a custom-made mouse
monoclonal antibody to TDP-43 (ref. 16, 40 µl of ascites per 400 µl of beads
per sample). Libraries were subjected to standard Illumina GA2 sequencing
protocol for 36 cycles.
Generation of transgenic mice. All animal procedures were conducted in accordance with the guidelines of the Institutional Animal Care and Use Committee
of the University of California. cDNA containing N-terminal myc-tagged fulllength human TDP-43 were amplified and digested by SalI and cloned into the
XhoI cloning site of the MoPrP.XhoI vector (ATCC #JHU-2). The resultant
MoPrP.XhoI-myc-hTDP-43 construct was then digested upstream of the minimal Prnp promoter and downstream of the Prnp exon 3 using BamHI and NotI
and cloned into a shuttle vector containing loxP flanking sites. The final construct was then linearized using XhoI, injected into the pro-nuclei of fertilized
C56Bl6/C3H hybrid eggs and implanted into pseudopregnant female mice.
Stereotactic injections of antisense oligonucleotides. We anesthetized
8–10-week-old female C57Bl/6 mice with isofluorane. Using stereotaxic guides,
3 µl of ASO solution, corresponding to a total of 75 µg or 100 µg ASOs, or saline
buffer was injected using a Hamilton syringe directly into the striatum. Mice
were monitored for any adverse effects for the next 2 weeks until they were killed.
The striatum and adjacent cortex area were dissected and frozen at –80 °C in
1 ml Trizol (Invitrogen). Trizol extraction of RNA and protein was performed
according to the manufacturer’s instructions.
RNA quality and RNA-seq library preparation. RNA quality was measured using
the Agilent Bioanalyzer system according to the manufacturer’s recommendations. RNA-seq libraries were constructed as described previously22. We used 8 pM
of amplified libraries for sequencing on the Illumina GA2 for 72 cycles.
RT and qRT-PCR. cDNA of total RNA extracted from striatum was generated
using oligodT and Superscript III reverse transcriptase (Invitrogen) according
to the manufacturer’s instructions. To test candidate splicing targets, RT-PCR
amplification using between 24 and 27 cycles were performed from at least three
mice treated with a control ASO and three mice with treated with a TDP-43 ASO.
Products were separated on 10% polyacrylamide gels followed by staining with
SYBR gold (Invitrogen). Quantification of the different isoforms was performed
with ImageJ software (US National Institutes of Health). Intensity ratio between
products with included and excluded exons were averaged for three biological
replicates per group.
qRT-PCR for mouse Tardbp and Fus/Tls were performed using the Express
One-Step SuperScript qRT-PCR kits (Invitrogen) and the thermocycler ABI
Prism 7700 (Applied Biosystems). cDNA synthesis and amplification were performed according to the manufacturer’s instruction using specific primers and
5′ FAM, 3′ TAMRA-labeled probes. The cyclophilin gene was used to normalize
the expression values.
qRT-PCR for all other genes tested were performed with 3–5 mice for each
group (treated with saline, control ASO or ASO to TDP-43) and two technical
replicates using the iQ SYBR green supermix (Bio-Rad) on the IQ5 multicolor
real-time PCR detection system (Bio-Rad). The analysis was done using the IQ5
optical system software (Bio-Rad; version 2.1). Expression values were normalized to at least two of the following control genes: Actb, Actg1 and Rsp9. Expression
values were expressed as a percentage of the average expression of the saline
treated samples. Inter-group differences were assessed by two-tailed Student’s
t test. Primers for RT-PCR and qRT-PCR were designed using Primer3 software
(http://frodo.wi.mit.edu/primer3/) and sequences are available on request.
Immunoblots. Proteins were separated on custom-made 12% SDS page gels
and transferred to nitrocellulose membrane (Whatman) following standard
protocols. Membranes were blocked overnight in Tris-buffered saline Tween-20
(TBST) and 5% non-fat dry milk (wt/vol) at 4 °C, incubated for 1 h at 20 °C with
primary antibodies and then again with horseradish peroxidise–linked secondary antibodies to rabbit or mouse (GE Healthcare) in TBST with 1% milk. For
primary antibodies, we used rabbit antibody to FUS/TLS (Bethyl Laboratories,
nature neuroscience
cat #A300-302A; 1:5,000), rabbit antibody to TDP-43 (Proteintech, cat #10782;
1:2,000), rabbit antibody to TDP-43 (Aviva System Biology, cat #ARP38942_T100;
1:2,000), custom-made mouse antibody to TDP-43 (1:1,000)16, custom-made
rabbit antibody to RFP raised against the full-length protein (1:7,000), mouse
DM1α antibody to tubulin (1:10,000) and mouse antibody to GAPDH (Abcam,
cat #AB8245; 1:10,000).
Cells, cloning and luciferase assays. HeLa Flp-In cells expressing GFP-myc–
TDP-43–HA were generated as previously described16. Isogenic cell lines were
grown at 37 °C and 5% CO2 in Dulbecco’s modified Eagle medium (DMEM),
supplemented with 10% tetracycline-free fetal bovine serum (vol/vol) and penicillin/streptomycin. Expression of GFP-myc–TDP-43–HA was induced with
1 mg ml–1 tetracycline for 24–48 h.
To assess the mechanism of TDP-43 autoregulation, we cloned the proximal
part of mouse TDP-43 3′ UTR into the psiCHECK-2 vector (Promega), which
contains a Renilla luciferase and a Firefly luciferase reporter expression cassettes.
The following primers were used to amplify 1.7 kb of TDP-43 3′ UTR using
cDNA from mouse brain: 5′-aaa CTC GAG cag gct ttt ggt tct gga
aa-3′ and 5′-aaa gcg gcc gcA CCA TTT TAG GTG CGG TCA C-3′. We
obtained two products of 1.7 kb and 1.1 kb, corresponding, respectively, to an
unspliced and a spliced isoform of TDP-43 3′ UTR (Supplementary Fig. 13a).
Both products were purified on 1% agarose gels and cloned independently in
the psiCHECK-2 vector using NotI and XhoI restriction sites located 3′ to the
Renilla luciferase translational stop. Given that binding sites for TDP-43 mainly
lie in the alternative intron, the spliced isoform was used as a control to assess
the effect of TDP-43 protein on its own RNA.
Human myc-TDP-43-HA cDNA (a generous gift from C. Shaw, King’s College
London) was cloned into mammalian expression vector, pCl-neo (Promega).
RFP was cloned in the vector pcDNA3 (Invitrogen).
250 ng of psiCHECK-2 vector with or without TDP-43 3′ UTR and 250 ng of the
vector expressing TDP-43 or RFP were co-transfected in Hela cells using Fugene 6
transfection reagent (Roche) in 12-wells cell culture plates. Luciferase assays were
performed 48 h after transfection using the Dual-Luciferase Reporter 1000 assay
system (Promega) according to the manufacturer’s instructions. Five independent
experiments were performed and 20 µl of lysate were used in duplicate for each
condition. RFU for Renilla luciferase were normalized to firefly luciferase to control for transfection efficiency. Duplicates were averaged and each condition was
expressed as a percentage of the samples without transfection of TDP-43 or RFP
cDNA. Inter-group differences were assessed by two-tailed Student’s t test.
Mouse gene structure annotations. The mouse genome sequence (mm8) and
annotations for protein-coding genes were obtained from the UCSC Genome
Browser. Known mouse genes (knownGene containing 31,863 entries) and
known isoforms (knownIsoforms containing 31,014 entries in 19,199 unique isoform clusters) with annotated exon alignments to the mouse genomic sequence
were processed as follows. Known genes that were mapped to different isoform
clusters were discarded. All mRNAs aligned to mm8 that were greater than 300
nucleotides were clustered together with the known isoforms. For the purpose
of inferring alternative splicing, genes containing fewer than three exons were
removed from further consideration. A total of 1.9 million spliced ESTs were
mapped onto the 16,953 high-quality gene clusters to identify alternative splicing events. Final annotated gene regions were clustered together so that any
overlapping portion of these databases was defined by a single genomic position.
Promoter regions were arbitrarily defined as 1.5 kb upstream of the transcriptional start site of the gene and intergenic regions as unannotated regions in
the genome. To identify 5′ and 3′ UTRs, we relied on the coding annotation in
UCSC Genome Browser known genes that we extended 1.5 kb downstream or
upstream the start and stop codons, respectively.
Comparison to human alternatively spliced exons. To find evidence for conserved
alternative splicing patterns in humans, we used the UCSC LiftOver tool to obtain
the corresponding human coordinates of the mouse exons that had evidence for
differential splicing either from RNA-seq or splicing-sensitive microarray data.
These human genome coordinates were then compared with human gene structure
annotations constructed analogously to mouse annotations as described above to
determine whether the exon in the human ortholog was alternatively or constitutively spliced based on transcript evidence in human EST or mRNA databases49.
doi:10.1038/nn.2779
Computational identification of CLIP-seq clusters. CLIP-seq reads were
trimmed to remove adaptor sequences and homopolymeric runs, and mapped
to the repeat-masked mouse genome (mm8) using the bowtie short-read alignment software (version 0.12.2) with parameters -q -p 4 -e 100 -y -a -m 10 --best
--strata (-q for fastq file, -p for cpus, -e for mismatch threshold, -y to do a sensitive search, -a to report all alignments, --best for ordered output, --strata report
only the best group of alignments), incorporating the base-calling quality of the
reads. To eliminate redundancies created by PCR amplification, we considered
all reads with identical sequence to be a single read. Significant clusters were
calculated by first determining read number cutoffs using the Poisson distribution,
λ k e−λ
f ( k ; λ) =
, where λ is the frequency of reads mapped over an interval of
k!
nucleotide sequence, k is the number of reads being analyzed for significance and
f(k;λ) return the probability that exactly k reads would be found. For any desired
P value, P cutoff, a read number cutoff was calculated by summing the probabilities for finding k or more tags, and determining the minimum value of k that
satisfies i = k such that f(k;λ) > P cutoff. The frequency λ was calculated by dividing the total number of mapped reads by the number of non-overlapping intervals
present in the transcriptome. The interval size was chosen on the basis of the
average size of the CLIP product, which includes only the selected RNA fragments, but not any ligated adapters (150 bp). A global and a local cutoff were
determined using the whole transcriptome frequency or gene-specific frequency,
respectively. The gene-specific frequency was the number of reads overlapping
that gene divided by the pre-mRNA length. A sliding window of 150 bp was used
to determine where the read numbers exceed both the global and local cutoffs.
At each significant interval, we attempted to extend the region by adding in the
next read, ensuring continued significance at the same P cutoff.
Log curve analysis to determine target saturation of CLIP-derived clusters. We
used a curve-fitting approach with various sampling rates to estimate the number
of CLIP-seq reads required to discover additional CLIP clusters. The target rate was
calculated by determining the new set of CLIP-derived gene targets found at each
step-wise increase in the number of sequenced reads. A scatter plot of targets found
versus sampling rate was fitted with a log curve, which was then used to extrapolate
the number of targets expected to be found by increasing read counts.
Splicing-sensitive microarray data analysis. Microarray data analysis was performed as previously described, selecting significant, high-quality events using
a q value > 0.05 and an absolute separation score (sepscore) > 0.5 (ref. 27). The
equation for sepscore = log2[TDP-43 depleted (Skip/Include)/ Control treated
(Skip/Include)]. For each replicate set, the log2 ratio of skipping to inclusion was
estimated using robust least-squares analysis. Previously published work using
similar cutoffs have validated about 85% of splicing events by RT-PCR27.
Events on the array were defined in the mm9 genome annotation, and for
proper comparison to RNA-seq and CLIP-seq data, cassette events were converted
to the mm8 genome annotation using the UCSC LiftOver tool. If an event did not
exactly overlap an mm8 annotated exon, it was left out of further analyses.
Genomic analysis of CLIP clusters. Two biologically independent TDP-43 CLIPseq libraries were generated and sequenced on the Illumina GA2. Subjecting reads
from each library to our cluster-identification pipeline described above defined
15,344 and 30,744 clusters for CLIP experiments 1 and 2, respectively. A gene
was considered to be a TDP-43 target that overlapped in both experiments if it
contained an overlapping cluster. We generated ten randomly distributed cluster
sets and compared each to the original clusters. To compute the significance of
the overlap, we calculated the standard or Z-score as follows: (percent overlap in
the two experiments – mean percent overlap in ten randomly distributed cluster
sets)/ s.d. of percent overlap in the randomly distributed cluster sets. A P value
was computed from the standard normal distribution and assigned significance if
it was lower than P < 0.01. To generate the final set of TDP-43 CLIP-seq clusters,
unique reads from both experiments were combined and subjected to our clusteridentification pipeline. Overall, clusters were at most 300 bases in length, with the
median of 142 bases. As a comparison to a published HITS-CLIP/CLIP-seq dataset of a RNA binding protein in mouse brain, we downloaded Ago-HITS-CLIP
reads (Brain[A–E]_130_50_fastq.txt) from http://ago.rockefeller.edu/rawdata.
php and subjected the combined 1,651,104 reads to our cluster-identification and
generated 33,390 clusters in 7,745 genes21. To identify enriched motifs in cluster
regions, Z scores were computed for hexamers as previously published15.
doi:10.1038/nn.2779
Functional annotation analysis. We used the Database for Annotation,
Visualization and Integrated Discovery (DAVID Bioinformatic Resources 6.7;
http://david.abcc.ncifcrf.gov/). For all genes downregulated or upregulated on
TDP-43 depletion, a background corresponding to all genes expressed in brain
was used.
Transcriptome and splicing analysis. Strand-specific RNA-seq reads each from
control oligonucleotide, saline- and TDP-43 oligonucleotide–treated animals
were generated and ~50% mapped uniquely to our annotated gene structure
database using the bowtie short-read alignment software (version 0.12.2, with
parameters -m 5 -k 5 --best --un --max -f) incorporating the base-calling quality
of the reads. To eliminate redundancies created by PCR amplification, all reads
with identical sequence were considered single reads. The expression value of
each gene was computed by RPKM. Each RNA-seq sample was summarized by
a vector of RPKM values for every gene and pair-wise correlation coefficients
were calculated for all replicates using a linear least-squares regression against
the log RPKM vectors. Hierarchical clustering revealed that the three treated
conditions clustered into similar groups. The reads in each condition were combined to identify genes that were significantly up- and downregulated on TDP-43
depletion. Local mean and s.d. were calculated for the nearest 1,000 genes, as
determined by log average RPKM values between knockdown and control and a
local Z score was defined. The resulting Z scores were used to assign significantly
changed genes (Z > 2 upregulated, Z < –2 downregulated), as well as ranking
the entire gene list for relative expression changes. Specific parameters, such as
intron length or TDP-43–binding sites, were plotted for the next 100 genes.
Exons with canonical splice signals (GT-AG, AT-AC, GC-AG) were retained,
resulting in a total of 190,161 exons. For each protein-coding gene, the 50 bases
at the 3′ end of each exon were concatenated with the 50 bases at the 5′ end of
the downstream exon producing 1,827,013 splice junctions. An equal number
of ‘impossible’ junctions was generated by joining the 50-base exon junction
sequences in reverse order. To identify differentially regulated alternative cassette exons, we employed a modification of a published method24. In short, the
read count supporting inclusion of the exon (overlapping the cassette exon and
splice junctions including the exon) were compared with the read count supporting exclusion of the exon (overlapping the splice junction of the upstream and
downstream exon. For a splice junction read to be enumerated, we required that
at least 18 nucleotides of the read aligned and 5 bases of the read extended over
the splice junction with no mismatches 5 bases around the splice junction. For
the TDP-43 and control ASOs comparison, we constructed a 2 × 2 contingency
table using of the counts of the reads supporting the inclusion and exclusion of
the exon, in two conditions. Every cell in the 2 × 2 table had to contain at least
five counts for a χ2 statistic to be computed. At P < 0.01, 110 excluded and 93
included single cassette exons were detected to be differentially regulated by
TDP-43. As an estimate of false discovery, we observed that ~20 single cassette
exons were detected by using the impossible junction database.
Pre-mRNA features and tissue-specificity analysis. Affymetrix microarray data
representing 61 mouse tissues were downloaded from the Gene Expression
Omibus repository (http://www.ncbi.nih.gov/geo) under accession number
GSE1133 (ref. 50). Probes on the microarrays were cross-referenced to 15,541
genes in our database using files downloaded from the UCSC Genome Browser
(knownToGnf1m and knownToGnfAtlas2). The expression value for each gene
was represented by the average value of the two replicate microarray experiments
for each tissue. To identify genes enriched in brain, we grouped 13 tissues as
‘brain’ (substantia nigra, hypothalamus, preoptic, frontal cortex, cerebral cortex,
amygdala, dorsal striatum, hippocampus, olfactory bulb, cerebellum, trigeminal,
dorsal root ganglia, pituitary) and the remaining 44 tissues as ‘non-brain’,
excluding the five embryonic tissues (embryonic days 10.5, 9.5, 8.5, 7.5 and 6.5).
− µnon-brain )
(µ
, where
For each gene, the t statistic was computed as t = brain
2
2
s brain
+ s non-brain
mbrain(sbrain) and mnon-brain(snon-brain) were the average (s.d.) of the gene expression values in brain and non-brain tissues, respectively. At a t statistic value
cutoff of ≥1.5, 388 genes were categorized as being brain enriched. Concurrently,
at a cutoff of <1.5, 15,153 genes were categorized as being non-brain enriched.
Random sets of 388 genes were selected from the 15,541 genes as controls for
the brain-enriched set. To determine whether pre-mRNA features were significantly different in the set of brain-enriched genes (or randomly chosen genes)
nature neuroscience
‘brain’ (temporal lobe, globus pallidus, cerebellum peduncles, cerebellum, caudate nucleus, whole brain, parietal lobe, medulla oblongata, amygdala, prefrontal
cortex, occipital lobe, hypothalamus, thalamus, subthalamic nucleus, cingulated
cortex, pons, fetal brain), and the remaining 62 tissues as ‘non-brain’. At the same
t statistic cutoffs, 387 and 17,985 genes were categorized as brain enriched and
non-brain enriched, respectively. Random sets of 387 genes were selected from
the 18,372 genes as controls for the brain-enriched set.
49.Yeo, G.W., Van Nostrand, E.L. & Liang, T.Y. Discovery and analysis of evolutionarily
conserved intronic splicing regulatory elements. PLoS Genet. 3, e85 (2007).
50.Su, A.I. et al. A gene atlas of the mouse and human protein-encoding transcriptomes.
Proc. Natl. Acad. Sci. USA 101, 6062–6067 (2004).
compared with non-brain–enriched genes, we performed the two-sample
Kolmogorov-Smirnov goodness-of-fit hypothesis test, which determines
whether the distribution of features were drawn from the same underlying continuous population. Cumulative distribution plots of pre-mRNA features were
generated to illustrate the differences.
Affymetrix microarray data representing 79 human tissues were downloaded
from the Gene Expression Omibus repository under the same accession number
GSE1133 (ref. 50). Probes on the human microarrays were cross-referenced to
18,372 genes in our database using files downloaded from the UCSC Genome
Browser (knownToGnfAtlas2 – hg18). The same analysis as done for the mouse
array data was repeated for these human array data. We grouped 17 tissues as
nature neuroscience
doi:10.1038/nn.2779
PolymenidouetalSUPPLEMENTARYINFORMATION
Long pre‐mRNA depletion and RNA missplicing contribute to neuronal
vulnerabilityfromlossofTDP‐43
Polymenidou M*, Lagier‐Tourenne C*, Hutt KR*, Huelga SC, Moran J, Liang TY, Ling SC, Sun E, Wancewicz E,
MazurC,KordasiewiczH,SedaghatY,DonohueJP,ShiueL,BennettCF,YeoGWandClevelandDW
Supplementary Figure 1 TDP‐43 CLIP‐seq method, reproducibility and saturation of binding sites. (a) Schematic
representation of cross‐linking, immunoprecipitation and high‐throughput sequencing protocol (CLIP‐seq). (b)
Autoradiographs of CLIP‐seq
experiments performed in
parallel using two different
antibodies, a commercially
available rabbit polyclonal
TDP‐43
antibody
(Aviva
Systems Biology, Cat. No.
ARP38942_T100) left and an
in‐house monoclonal antibody
recognizing an epitope within
aminoacids251‐414ofhuman
TDP‐4316. Under identical
conditionsandexposuretimes,
our monoclonal antibody
showedamuchhigherpotency
of immunoprecipitating TDP‐
43‐RNAcomplexes.Theorange
box depicts the excised low
mobility complexes used for
the cDNA libraries for
comparison to the monomeric
TDP‐43
complexes.
(c)
Reproducibility of CLIP‐seq
experimentsonagenome‐wide
scale. Clusters identified in
independent
CLIP‐seq
experiments were considered
tooverlapif1baseofacluster
in one experiment extended to
a cluster in the other
experiment(leftpanel).Agene
containing an overlapping
cluster was considered to
overlap (right panel). Venn
diagrams revealed statistically
significant overlap between
clusters and genes in replicate
CLIP‐seq
libraries,
with
randomly distributed clusters
shown as a comparison (lower panel). (d) Gene target saturation chart plotting gene targets found by cluster‐finding on
increasingsamplingrates(5%intervals).DatafollowsalogarithmicexpansionwithanR2valueof0.98.
natureneuroscience
Nature Neuroscience: doi:10.1038/nn.2779
1
Polymenidou et al
SUPPLEMENTARY INFORMATION
Supplementary Figure 2 Cluster strength and gene-location biases of TDP-43 binding. (a) Box plots of cluster strength (by
χ² analysis) versus GUGU-cluster bins (counts of non-overlapping GUGU tetramer found within the cluster). By t-test
comparison to all clusters, bins of 11-15 and 16+ GUGUs were significantly enriched (p < 0.01 and p < 10-37, respectively). On
each box, the central mark is the median, the edges of the box are the 25th and 75th percentiles, the whiskers extend to the
most extreme data points not considered outliers, and outliers are plotted individually (b) Pre-mRNAs were divided into
annotated 5′ and 3′ untranslated regions, exons, proximal introns (<0.5kb from a splice junction), distal introns (>0.5kb
from a splice junction) and unannotated regions. Pie-charts revealed that the majority of TDP-43 binding sites were located
in distal intronic regions of pre-mRNA (left panel). For comparison, previously published Argonaute (Ago) binding sites in
mouse brain21 are displayed as analogous pie-charts (right panel). (c) CLIP binding sites of TDP-43 in the two replicates,
Ago, and random TDP-43 clusters across genes. The fraction of CLIP clusters was plotted depending on the relative gene
position from the 5’ to the 3’ ends of each target gene.
Supplementary Figure 3 RNA-seq biological replicas. Correlation
between RNA-seq results obtained from mice subjected to different
ASO treatments. Heatmap generated from Cluster3 and Matlab, using
linear least-squares regression correlation coefficients of the pair-wise
comparison between RPKM values of all genes from each experiment.
“K” is TDP-43 knockdown samples, “C” is control oligo-, “S” is salinetreated samples, and numbers represent biological replicates.
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
2
Polymenidou et al
Supplementary Figure 4 Correlation between RNA-seq and CLIP-seq data using various parameters. (a) Genes were
ranked randomly instead of upon their degree of regulation after TDP-43 depletion and the average TDP-43 CLIP clusters
found in introns of the next 100 genes in the ranked list were plotted (green line). Similarly, the median of total intron
length for the next 100 genes was
plotted (blue line). (b) Genes were
ranked by ascending expression
(RPKM) values in the samples
treated with control ASO. (c) Genes
were
ranked
by
ascending
expression (RPKM) values in the
samples treated with TDP-43 ASO.
(d) Genes were ranked upon their
degree of regulation after TDP-43
depletion and average Ago clusters
found in introns of the next 100
genes (green line) or the total intron
length for the next 100 genes (blue
line) were plotted. (e) Genes were
ranked upon their degree of
regulation after TDP-43 depletion
and average clusters found in 5’UTR
for the next 100 genes (green line) or
the total 5’UTR length for the next
100 genes (blue line) were plotted.
(f) Genes were ranked upon their
depletion and average clusters found
in 3’UTR for the next 100 genes
(green line) or the total 3’UTR length
for the next 100 genes (blue line)
were plotted. (g) Genes were ranked
upon their degree of regulation after
TDP-43 depletion and average
clusters found in exons for the next
100 genes (green line) or the total
exon length for the next 100 genes
(blue line) were plotted. No
correlation between CLIP and RNAseq data was observed (a-f). (h)
Genes were ranked upon their
depletion and average clusters found
in introns for the next 100 genes
(green line) or the average intron
length for the next 100 genes (blue
line) were plotted. (i) Genes were
ranked upon their degree of
regulation after TDP-43 depletion
and average cluster density (cluster count/intron length) for the next 100 genes was plotted.
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
3
PolymenidouetalSUPPLEMENTARYINFORMATION
Supplementary Figure 5 The calcium signaling pathway is affected by TDP‐43 depletion. (a) The calcium‐signaling
pathwayannotatedbyKEGG(KyotoEncyclopediaofGenesandGenomes)wasfoundtobesignificantusingthefunctional
annotationtoolDAVID.Red‐borderedgenegroupscontainTDP‐43targetsthataredownregulateduponTDP‐43depletion.
(b)ListofdownregulatedTDP‐43targetscontributingtocalcium‐signalingpathway
Nature
Neuroscience: doi:10.1038/nn.2779
natureneuroscience
4
Polymenidou et al
Supplementary Figure 6 The long-term potentiation pathway is affected by TDP-43 depletion. (a) The long-term
potentiation pathway annotated by KEGG was found to be significant using the functional annotation tool DAVID. Redbordered gene groups contain TDP-43 targets that are downregulated upon TDP-43 depletion. (b) List of downregulated
TDP-43 targets contributing to long-term potentiation pathway.
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
5
Polymenidou et al
Supplementary Figure 7 Representative examples of genes with long introns and multiple TDP-43 binding sites. (a) Most
transcripts in this group (mean intron length >10kb) displayed multiple intronic TDP-43 binding sites as determined by
CLIP-seq (purple bars above each gene). In contrast, Chat is an example of a downregulated gene with short intron length
(mean intron length <10kb) that shows a single TDP-43 binding site. The number of clusters per gene (n) is in purple on the
left. Note that each cluster represents a binding cluster defined by a collection of overlapping reads as shown for Park2
(inset). (b) Examples of conservation of gene structures in human neurexin 3 and parkin 2 showing similar genic
architectures as their mouse orthologues.
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
6
Polymenidou et al
Supplementary Figure 8 TDP-43 binding on the 3′UTR of RNAs represses gene expression. (a) Normalized expression
(based on RPKM values from RNA-seq) of selected up-regulated genes upon TDP-43 depletion that also contain TDP-43
binding regions in their 3’UTR. (b) Enrichment for binding within the 3’UTR among upregulated TDP-43 targets, compared
to unchanged genes. *Upregulated with intronic binding versus unchanged with intronic binding (p<1.8x10-6),
**Upregulated with 3’UTR binding versus unchanged with 3’UTR binding (p<8.2x10-3), ***Upregulated with 3’UTR binding
versus downregulated with 3’UTR binding (p < 1.4 x 10-2). P-values calculated by chi-square.
Supplementary Figure 9 Splicing changes
upon TDP-43 depletion identified by exonjunction arrays. Distribution of alternative
splicing events upon TDP-43 depletion
detected
using
splicing-sensitive
microarrays. Of the 326 cassette events
defined by Affymetrix on mm9 annotation,
287 were annotated by our stringent gene
structure annotations in mm8, and were
used for comparisons to RNA-seq and CLIPseq data.
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
7
Polymenidou et al
Supplementary Figure 10 Examples of alternative splicing regulation by TDP-43. RNA targets with exons included (a) or
excluded (b) upon TDP-43 knockdown (same as in Fig. 4d). Semi-quantitative RT-PCR of selected targets showing splice
changes in samples with TDP-43 knockdown compared to controls. Schematic representation of assessed exons with the
changed exon(s) colored orange (left). The exon number of the gene is indicated below each box. The red arrows depict the
position of the primers used for RT-PCR. Purple boxes indicate TDP-43 binding sites as defined by CLIP-seq. Representative
acrylamide gel pictures of RT-PCR products from control or knockdown adult brain samples, as indicated (right panel).
Quantification of splicing changes from three biological replicas per group is shown in the middle panel.
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
8
Polymenidou et al
Supplementary Figure 11 Splicing changes correlate with the level of TDP-43 knockdown. (a) Acrylamide gels with semiquantitative RT-PCR samples showing sortilin 1 (Sort1) splicing changes in mice with various levels of TDP-43 reduction
(shown as percentages above each bar). (b) Significant correlation of TDP-43 levels to Sort1 splicing changes (correlation
value = -0.905, n=27). (c) Acrylamide gels with semi-quantitative RT-PCR samples showing Kcnip splicing changes in mice
with various levels of TDP-43 reduction (shown as percentages above each bar on the lowest panel). (D-E) Significant
correlation of TDP-43 levels to Kcnip splicing changes (correlation value = 0.91 for exon 3 and 0.83 for exon 2 and 3, n=12).
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
9
Polymenidou et al
Supplementary Figure 12 TDP-43 3’UTRs used in luciferase assays and control immunoblots. (a) Two alternative 3’UTRs
of TDP-43 were amplified from mouse or human cDNA using primers whose approximate locations are shown by the small
red arrows below the schemes on the right. The first 1.7kb or 1.1kb of each long (unspliced) or short (spliced) 3’UTR
respectively were then cloned into a renilla luciferase reporter vector. The latter also contains the firefly luciferase gene that
is expressed independently from renilla luciferase and serves as a transfection normalization control. (b) Representative
immunoblots of HeLa cell lysates used for luciferase assays in Fig. 5F showing increased expression of TDP-43 in samples
transfected with a myc-hTDP-43-HA expressing vector. For control, cells were transfected with a vector expressing an
unrelated protein, namely red fluorescence protein (RFP), whose levels were confirmed by immunoblot (second panel). (c)
Representative immunoblots of HeLa cell lysates used for Fig. 5F,G showing decreased levels of UPF1 in samples transfected
with UPF1 siRNA.
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
10
Polymenidou et al
Supplementary Figure 13 TDP-43 binding and regulation of Hdac6 and Nefl transcripts. (a) CLIP-seq reads and clusters on
Hdac6 transcript showing intronic TDP-43 binding. RNA-seq reads from control or TDP-43 knockdown samples (equal
scales) show a slight decrease in Hdac6 mRNA in the TDP-43 knockdown group. (b) Quantitative RT-PCR confirmed the
slight decrease in Hdac6 mRNA upon TDP-43 depletion. Standard deviation was calculated within each group for 3 biological
replicas. (c) CLIP-seq reads and clusters on Nefl mRNA showing two distinct TDP-43 binding sites within the 3’UTR. Our
RNA-seq reads indicate that the 3’UTR of mouse Nefl is longer than reported in UCSC database (light blue bar indicates the
extention). RNA-seq reads from control or TDP-43 knockdown samples show a slight decrease in Nefl mRNA in the TDP-43
knockdown group, in accordance with previous report.
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
11
Polymenidou et al
Supplementary Figure 14 TDP-43 binding patterns on transcripts encoding for selected disease-related proteins. The
number of clusters per gene (n) is in purple on the left. Note that each cluster represents a binding site defined by a
collection of overlapping reads as shown for ataxin 1 (inset).
nature
neuroscience
Nature
Neuroscience:
doi:10.1038/nn.2779
12
Polymenidou et al
Supplementary Table 1 Genes upregulated upon TDP-43 depletion in adult mouse brain
Number of CLIP-seq clusters
RNA-seq
Total
in
5'UTR
in
introns
Oasl2
0
0
0
0
0
29.97
2.86
10.49
11.66
Ifit1
0
0
0
0
0
23.26
2.11
11.03
11.23
Gene symbol
Protein
Refseq/mRNA
identifier
BC034361
Oasl2 protein
Interferon-induced protein with
NM_008331
tetratricopeptide repeats 1 (IFIT-1)
NM_010501
Ifit3
NM 198095
Bst2
DAMP-1 protein
5830458K16Rik
Receptor transporting protein 4
NM 023386
Signal transducer and activator of
AK046517
Stat1
transcription 1
Glycoprotein (transmembrane) nmb
AK079220
Gpnmb
Lectin, galactoside-binding, soluble, 3
NM_011150
Lgals3bp
binding
Interferon induced transmembrane protein
Ifitm3
NM_025378
3
Transgelin 2
BC049861
Tagln2
AK077243
Ifit3
Interferon inducible protein 1
AK002545
Irgm
Vim
Vimentin
NM 011701
C3
Complement component 3
NM 009778
Gbp2
Guanylate nucleotide binding protein 2
NM 010260
Cd5l
CD5 antigen-like
NM 009690
Cd44
CD44 antigen isoform b
NM 001039150
Serine (or cysteine) proteinase inhibitor,
Serping1
NM_009776
clade
NM 009779
C3ar1
Complement component 3a receptor 1
in
RPKM in
in 3'UTR
exons
TDP-43 KD
RPKM in
Ratio
Z score
control KD/control
0
0
0
0
0
20.03
3.07
6.53
8.15
0
0
0
0
0
0
0
0
0
0
16.05
10.03
3.21
1.44
5.00
6.98
6.96
6.86
0
0
0
0
0
20.59
4.52
4.56
6.81
0
0
0
0
0
70.72
16.86
4.20
6.62
0
0
0
0
0
73.81
17.96
4.11
6.61
0
0
0
0
0
30.74
8.52
3.61
6.58
0
0
0
0
0
15.17
3.62
4.19
6.18
1
0
0
0
1
9.91
1.72
5.75
6.15
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
14.53
111.40
8.71
7.49
6.39
12.36
3.68
31.33
1.57
1.33
0.98
2.87
3.95
3.56
5.56
5.61
6.50
4.30
6.02
5.92
5.91
5.89
5.76
5.69
0
0
0
0
0
14.77
3.96
3.73
5.65
0
0
0
0
0
16.68
4.72
3.53
5.40
AF220014
0
0
0
0
0
8.96
1.91
4.69
5.39
NM 007807
AK171067
NM 175628
NM 029803
NM 008175
AK046736
NM 007631
NM 010705
NM 007669
NM 011909
NM 010259
NM 011163
BC003792
NM 009263
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
5.49
22.69
11.32
4.29
70.17
11.26
33.11
12.07
26.03
4.11
6.55
11.99
11.28
35.34
0.86
7.54
2.76
0.55
23.12
2.78
11.12
3.31
9.17
0.57
1.40
3.26
2.94
12.35
6.36
3.01
4.10
7.76
3.04
4.05
2.98
3.65
2.84
7.25
4.67
3.68
3.84
2.86
5.32
5.29
5.25
5.22
5.22
5.20
5.19
5.11
5.05
5.05
5.05
5.05
5.00
4.93
AK159572
0
0
0
0
0
44.59
16.67
2.67
4.53
Cd68
Cytochrome b-245, beta polypeptide
CD84 antigen
Alpha-2-macroglobulin
Interferon stimulated gene 12(b1)
Granulin
Novel TRAF-type zinc finger protein
Cyclin D1
Galectin 3
Cyclin-dependent kinase inhibitor 1A
Ubiquitin specific protease 18
Guanylate nucleotide binding protein 1
Protein kinase, interferon-inducible double
Apolipoprotein B mRNA editing enzyme
Secreted phosphoprotein 1
TYRO protein tyrosine kinase binding
protein
Lectin, galactose binding, soluble 9
Schlafen 8
Cystatin F (leukocystatin)
Insulin-like growth factor I precursor (IGF-I)
(Somatomedin)
Transglutaminase 1, K polypeptide
Fc receptor, IgE, high affinity I, gamma
DEAD (Asp-Glu-Ala-Asp) box polypeptide
58
Interferon dependent positive acting
Complement component 1, q
subcomponent, A chain
H2-K1 protein
B aggressive lymphoma
CD52 antigen
H-2 class I histocompatibility antigen, D-B
alpha chain precursor (H- 2D(B))
Macrosialin precursor (CD68 antigen)
BC021637
0
0
0
0
0
38.54
14.73
2.62
4.51
S100a6
S100 calcium binding protein A6 (calcyclin)
NM_011313
0
0
0
0
0
11.39
3.44
3.31
4.50
Trim30
Cybb
Cd84
A2m
Ifi27
Grn
Fbxo39
Ccnd1
Lgals3
Cdkn1a
Usp18
Gbp1
Eif2ak2
Apobec1
Spp1
Tyrobp
Lgals9
Slfn8
Cst7
Igf1
Tgm1
Fcer1g
Ddx58
Isgf3g
C1qa
H2-K1
Parp9
Cd52
H2-D1
Rsad2
B2m
Fcgr2b
Slfn2
Arpc1b
Iigp2
Ly9
Sn
Slfn9
Dtx3l
NM 205820
Lyzs
Cxcr4
Slfn5
Msn
Lilrb4
Parp14
Cd48
Tripartite motif protein 30 (Down regulatory
protein of interleukin 2 receptor)
NM_011662
0
0
0
0
0
37.90
13.18
2.87
4.90
NM 010708
NM 181545
NM 009977
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
16.47
5.31
12.31
5.34
0.97
3.60
3.09
5.45
3.42
4.90
4.84
4.83
AY878193
0
0
0
0
0
7.26
1.73
4.20
4.80
NM 019984
NM 010185
0
0
0
0
0
0
0
0
0
0
4.50
33.93
0.69
12.36
6.48
2.75
4.80
4.72
4.70
AK155930
0
0
0
0
0
6.74
1.66
4.07
NM 008394
0
0
0
0
0
15.30
5.00
3.06
4.68
NM_007572
0
0
0
0
0
137.39
50.81
2.70
4.66
BC080756
NM 030253
NM 013706
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
31.08
6.64
10.40
11.60
1.68
2.87
2.68
3.94
3.62
4.66
4.56
4.53
Viral hemorrhagic septicemia virus(VHSV)
NM_021384
induced
NM 009735
Beta-2-microglobulin
Low affinity immunoglobulin gamma Fc
NM_010187
region receptor II precursor
NM 011408
Schlafen 2
Actin related protein 2/3 complex, subunit
AK171364
1B
Highly similar to Mus musculus interferonAK128991
gamma induced GTPase
Ly9 protein
BC095921
Sialoadhesin
NM 011426
Schlafen 9
NM 172796
Weakly similar to B-lymphoma- and BALBC085093
associated protein (Rhysin 2)
NM 205820
Toll-like receptor 13
Lysozyme C type M precursor
AK159276
C-X-C chemokine receptor type 4
NM 009911
Schlafen 5
NM 183201
Moesin
NM 010833
Leukocyte immunoglobulin-like receptor,
NM 013532
Poly (ADP-ribose) polymerase family,
NM_001039530
member 14
NM 007649
CD48 antigen
Nature Neuroscience:
doi:10.1038/nn.2779
nature neuroscience
0
0
0
0
0
4.28
0.70
6.12
4.49
0
0
0
0
0
139.63
54.00
2.59
4.45
0
0
0
0
0
11.96
3.85
3.11
4.42
0
0
0
0
0
7.60
2.09
3.64
4.41
0
0
0
0
0
11.69
3.81
3.07
4.32
0
0
0
0
0
8.11
2.26
3.59
4.32
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
9.30
3.25
3.48
2.77
0.55
0.60
3.37
5.88
5.81
4.20
4.18
4.15
0
0
0
0
0
8.50
2.56
3.32
4.11
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7.47
111.97
6.74
8.63
33.37
5.25
2.23
46.71
1.94
2.60
14.04
1.29
3.36
2.40
3.47
3.32
2.38
4.07
4.11
4.11
4.09
4.08
4.08
4.07
0
0
0
0
0
6.08
1.65
3.70
4.05
0
0
0
0
0
8.11
2.49
3.26
4.04
13
Polymenidou et al
Edg3
Ly86
AK029112
Ftl1
AK141670
Plscr2
Ctss
Lcn2
Tlr2
Unc93b1
Cd274
Cd9
Apoc1
Hexa
Igfbp5
Anxa2
Glipr1
Pdpn
Psmb8
Endothelial differentiation, sphingolipid
NM 010101
Lysosomal-associated protein
NM_010686
transmembrane 5
NM 030707
Macrophage scavenger receptor 2
Cellular repressor of E1A-stimulated genes
NM_011804
1
Epithelial membrane protein 1
NM 010128
Heme oxygenase (decycling) 1
NM 010442
Zinc finger protein 288
BC056446
Triggering receptor expressed on myeloid
NM_031254
cells
NM 007651
CD53 antigen
Angiogenin, ribonuclease A family, member
NM_007447
1
AK079888
MS4A7 protein homolog
ATP-binding cassette sub-family A member
NM_013454
1
NM 022325
Cathepsin Z preproprotein
E030016B05 product:embryonic epithelial
AK086962
gene 1
NM 199146
Hypothetical protein LOC209387
Transporter 1, ATP-binding cassette, subNM_013683
family
AK152638
Cathepsin D
Oncostatin M receptor beta
AB015978
Polymeric immunoglobulin receptor 3
AF251703
Cytochrome b-245, alpha polypeptide
NM 007806
TAP binding protein isoform 1
NM 001025313
NM_009777
subcomponent, B chain
NM 008533
CD180 antigen
NM 008332
Lectin, galactose binding, soluble 1
NM 008495
Glial fibrillary acidic protein
AK140151
Melanoma differentiation associated
NM_027835
protein-5
NM 144830
Chemokine (C-X-C motif) ligand 5
NM 009141
Solute carrier family 15, member 3
NM 023044
NM_007574
subcomponent, gamma
Lymphocyte antigen 86
NM 010745
Cybrd1 protein
AK029112
Ferritin light chain 1
NM 010240
Lysosomal acid lipase 1
AK141670
Phospholipid scramblase 2
NM 008880
Cathepsin S
AK028366
Lipocalin 2
NM 008491
NM 011905
Unc93 homolog B
NM 019449
CD274 antigen
NM 021893
CD9 antigen
NM 007657
Apolipoprotein C-I
NM 007469
Hexosaminidase A
AK159091
Insulin-like growth factor binding protein 5
NM 010518
Anxa2 protein
BC004659
GLI pathogenesis-related 1 (glioma)
NM 028608
Glycoprotein 38 precursor (GP38)
BC026551
Proteasome subunit beta type 8 precursor
BC013785
Hspb6
Laptm5
Msr2
Creg1
Emp1
Hmox1
Zbtb20
Trem2
Cd53
Ang1
Ms4a7
Abca1
Ctsz
Slc43a3
AI451617
Tap1
Ctsd
Osmr
Cd300lf
Cyba
Tapbp
C1qb
Cd180
Ifit2
Lgals1
Gfap
Ifih1
Tmem106a
Cxcl5
Slc15a3
0
0
0
0
0
10.24
3.29
3.11
0
0
0
0
0
53.57
23.32
2.30
4.03
4.03
0
0
0
0
0
60.70
25.61
2.37
4.00
0
0
0
0
0
25.09
11.02
2.28
3.99
0
0
27
0
0
0
0
0
27
0
0
0
0
0
0
5.99
12.84
10.94
1.59
4.95
3.79
3.76
2.60
2.89
3.99
3.96
3.95
0
0
0
0
0
46.13
19.85
2.32
3.92
0
0
0
0
0
20.62
9.15
2.25
3.84
0
0
0
0
0
21.94
9.78
2.24
3.79
0
0
0
0
0
5.29
1.41
3.74
3.78
6
0
6
0
0
32.69
14.78
2.21
3.73
0
0
0
0
0
55.45
25.30
2.19
3.73
0
0
0
0
0
4.91
1.29
3.82
3.73
0
0
0
0
0
4.37
1.02
4.28
3.72
0
0
0
0
0
4.36
1.03
4.22
3.68
2
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
333.96
8.11
3.16
12.34
18.91
152.95
2.73
0.65
4.92
8.34
2.18
2.97
4.86
2.51
2.27
3.68
3.67
3.67
3.67
3.66
2
0
2
0
0
110.40
50.84
2.17
3.65
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
9.19
9.25
9.55
98.80
3.20
3.26
3.44
46.64
2.87
2.84
2.77
2.12
3.63
3.61
3.56
3.54
0
0
0
0
0
5.25
1.57
3.35
3.53
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4.01
2.79
6.74
0.94
0.60
2.34
4.25
4.63
2.88
3.52
3.49
3.49
0
0
0
0
0
82.06
39.22
2.09
3.48
0
0
0
1
0
0
0
0
1
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
16.67
4.58
33.68
18.56
3.30
107.38
4.10
5.61
20.40
4.75
44.66
3.18
34.64
128.27
7.60
3.45
15.04
3.36
7.32
1.24
16.07
8.58
0.75
52.02
1.03
1.75
9.91
1.37
21.83
0.74
16.94
63.50
2.86
0.84
6.93
0.81
2.28
3.70
2.10
2.16
4.41
2.06
3.99
3.21
2.06
3.47
2.05
4.29
2.04
2.02
2.65
4.09
2.17
4.14
3.47
3.47
3.45
3.45
3.44
3.42
3.42
3.40
3.38
3.38
3.33
3.32
3.32
3.32
3.29
3.29
3.29
3.29
Heat shock protein, alpha-crystallin-related NM_001012401
0
0
0
0
0
7.11
2.68
2.65
3.23
Ras-related protein Rab-7b
Transmembrane protein 10
Tripartite motif protein 25
Histocompatibility 2, T region locus 23
Ms4a6c protein
AXL receptor tyrosine kinase
MHC class I Q4 beta-2-microglobulin (Qb1)
Tubulin beta-5 chain (Beta-tubulin class-V)
homolog
AK030688
NM 153520
AK169562
NM 010398
BC062247
NM 009465
0
1
0
0
0
1
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
4.86
14.00
6.39
12.19
2.10
20.86
1.51
6.33
2.35
5.55
0.52
10.40
3.22
2.21
2.72
2.20
4.08
2.01
3.23
3.23
3.22
3.21
3.20
3.20
AK037574
0
0
0
0
0
7.52
2.85
2.64
3.19
BC008225
0
0
0
0
0
2.19
0.54
4.03
3.16
Tnfrsf1a
Tumor necrosis factor receptor superfamily
NM_011609
0
0
0
0
0
10.35
4.37
2.37
3.15
Lcp1
NM 008879
0
0
0
0
0
10.43
4.40
2.37
3.14
AK088605
0
0
0
0
0
8.84
3.48
2.54
3.14
NM 201226
2
0
0
1
1
35.00
17.95
1.95
3.13
NM_026819
0
0
0
0
0
42.45
21.35
1.99
3.13
Cd22
Lymphocyte cytosolic protein 1
Glycoprotein galactosyltransferase alpha
1, 3
Leucine rich repeat containing 47
Dehydrogenase/reductase (SDR family)
member 1
CD22 antigen
Arhgdib
Rho, GDP dissociation inhibitor (GDI) beta
C1qg
5430435G22Rik
Tmem10
Trim25
H2-T23
Ms4a6c
Axl
H2-Q1
Tubb6
Ggta1
Lrrc47
Dhrs1
Slc14a1
Trim21
Nfe2l2
Fabp7
BC089618
Sfrp5
D12Ertd647e
Capg
Pik3ap1
Timp1
Urea transporter, erythrocyte (Urea
transporter B) (UT-B)
Tripartite motif protein 21
Nuclear factor erythroid 2 related factor 2
Fatty acid binding protein 7, brain
Novel protein similar to extracellular
proteinase inhibitor Expi
Secreted frizzled-related sequence protein
5
Hypothetical protein LOC52668 isoform 2
Gelsolin-like capping protein
Phosphoinositide-3-kinase adaptor protein
1
Metalloproteinase inhibitor 1 precursor
doi:10.1038/nn.2779
nature neuroscience
AK171589
0
0
0
0
0
4.48
1.36
3.29
3.12
NM_007486
0
0
0
0
0
19.72
9.76
2.02
3.09
AK153891
0
0
0
0
0
12.80
6.01
2.13
3.09
AK141294
U20532
NM 021272
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3.70
17.09
10.95
1.03
8.45
4.83
3.59
2.02
2.27
3.08
3.08
3.08
BC089618
0
0
0
0
0
2.71
0.69
3.93
3.06
NM_018780
0
0
0
0
0
3.18
0.83
3.81
3.05
NM 194066
NM 007599
0
0
0
0
0
0
0
0
0
0
16.90
3.46
8.51
0.93
1.99
3.70
3.04
3.03
NM_031376
0
0
0
0
0
5.62
1.99
2.83
3.03
BC008107
0
0
0
0
0
1.86
0.50
3.70
3.01
14
Polymenidou et al
Cd63
1810029B16Rik
Ctsc
Gusb
Stat2
NM 007653
NM 025465
NM 009982
BC004616
AF206162
NM_030150
0
0
0
0
0
2.49
0.69
3.61
2.76
Cd74
Cd63 antigen
Cathepsin C preproprotein
Beta-glucuronidase structural
Stat2 protein
member 12
Transcobalamin 2
Fc receptor, IgG, high affinity I
Colony stimulating factor 2 receptor, beta
2, low-affinity (granulocyte-macrophage)
Complement component 5, receptor 1
Cd151 protein
CKLF-like MARVEL transmembrane
domain-containing protein 3 (Chemokinelike factor superfamily member 3)
Epithelial stromal interaction 1 isoform b
Peripheral benzodiazepine receptor
Ribosomal protein L39
Scotin isoform 1
B-cell leukemia/lymphoma 3
Calponin 3, acidic
Interferon gamma inducible protein 47
Adipose differentiation related protein
Tumor necrosis factor receptor
superfamily,
Lipoma HMGIC fusion partner-like 2
Interferon-induced protein 35
Epithelial membrane protein 3
Transcription factor 7-like 2
Hypothetical Peptidase family
M20/M25/M40 containing protein
Transglutaminase 2, C polypeptide
Legumain
Uncoupling protein 2
Elastase 1, pancreatic
DNA segment, Chr 11, Lothar
Hennighausen 2,
Ia-associated invariant chain
NM 010545
0
0
0
0
0
7.68
3.34
2.30
2.73
Tgfbr2
Transforming growth factor, beta receptor II
NM_009371
0
0
0
0
0
10.18
4.75
2.14
2.73
Rac2
Apoe
Olig2
Anxa3
Apod
Slc11a1
Cxcl16
Rab31
Fcgr3
BC016112
P2ry13
Ctsh
1810009M01Rik
RAS-related C3 botulinum substrate 2
Apolipoprotein E
Oligodendrocyte transcription factor 2
Annexin A3
Apolipoprotein D
Solute carrier family 11
Chemokine (C-X-C motif) ligand 16
Rab31 protein
Fcgr3 protein
Col20a1 protein
Purinergic receptor P2Y13
Cathepsin H
LR8 protein
NM 009008
NM 009696
NM 016967
AK169423
NM 007470
NM 013612
NM 023158
BC013063
BC052819
BC016112
NM 028808
NM 007801
NM_023056
0
0
0
0
0
0
0
9
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
9
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4.47
556.59
13.58
9.95
84.02
6.06
4.46
36.69
26.35
2.62
11.97
13.22
23.57
1.58
314.21
6.98
4.68
47.53
2.55
1.58
20.69
15.06
0.76
6.17
6.76
13.38
2.83
1.77
1.94
2.13
1.77
2.38
2.83
1.77
1.75
3.43
1.94
1.96
1.76
2.72
2.71
2.71
2.71
2.70
2.70
2.70
2.70
2.69
2.68
2.68
2.67
2.66
Mcam
Cell surface glycoprotein MUC18 precursor
AB035509
Parp12
Tcn2
Fcgr1
Csf2rb2
C5r1
Cd151
Cmtm3
Epsti1
Bzrp
0610039P13Rik
Rpl39
Scotin
Bcl3
Cnn3
Ifi47
Adfp
Tnfrsf1b
Lhfpl2
Ifi35
Emp3
Tcf7l2
AK128969
Tgm2
Lgmn
Ucp2
Ela1
D11Lgp2e
Rps27l
Abhd4
Zfp36l1
Ctsl
Bcl2a1b
Ctsb
0610011I04Rik
Dbi
Prg1
Icam1
Cd109
Ccl3
Waspip
Npc2
Aqp4
Rps14
Parp3
Trim14
Tspan2
Cyr61
Aebp1
Tnfaip2
Dnase1l1
Tnc
Anxa5
Ephx1
AK153119
Tlr7
Mmp2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
26.25
6.39
10.99
11.23
14.11
14.29
2.51
4.96
5.15
6.79
1.84
2.55
2.21
2.18
2.08
2.99
2.99
2.98
2.98
2.97
NM_172893
1
0
1
0
0
7.20
2.87
2.51
2.97
NM 015749
NM 010186
0
0
0
0
0
0
0
0
0
0
19.96
6.63
10.54
2.64
1.89
2.52
2.96
2.96
AK155626
0
0
0
0
0
2.04
0.55
3.70
2.95
NM 007577
BC012236
0
0
0
0
0
0
0
0
0
0
3.23
13.80
0.88
6.75
3.67
2.05
2.95
2.93
BC003230
0
0
0
0
0
7.46
3.07
2.43
2.92
0
0
0
0
1
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
1.90
4.33
9.16
11.50
22.93
2.60
26.55
1.77
5.59
0.52
1.42
3.95
5.32
12.42
0.71
14.74
0.52
2.19
3.62
3.06
2.32
2.16
1.85
3.68
1.80
3.40
2.55
2.91
2.88
2.88
2.87
2.87
2.85
2.85
2.83
2.83
NM
NM
NM
NM
NM
NM
NM
NM
NM
178825
009775
028752
026055
025858
033601
028044
008330
007408
NM_011610
0
0
0
0
0
7.31
3.11
2.35
2.83
NM 172589
NM 027320
NM 010129
AJ223070
1
0
0
9
0
0
0
0
1
0
0
9
0
0
0
0
0
0
0
0
11.88
3.02
2.29
5.23
5.89
0.84
0.63
1.95
2.02
3.57
3.66
2.69
2.82
2.82
2.82
2.82
AK128969
NM
NM
NM
NM
009373
011175
011671
033612
Ribosomal protein S27-like
NM 026467
Abhydrolase domain containing 4
NM 134076
Zinc finger protein 36, C3H type-like 1
NM 007564
Cathepsin L precursor
AK159511
B-cell leukemia/lymphoma 2 related
NM_007534
protein A1b
NM 007798
Cathepsin B preproprotein
Hepatocellular carcinoma-associated
NM_025326
antigen 112
NM 001037999
Diazepam binding inhibitor isoform 1
Proteoglycan 1, secretory granule
NM 011157
Intercellular adhesion molecule
NM 010493
Hypothetical Alpha-macroglobulin receptor
domain/Terpenoid cylases/Protein
BC052443
prenyltransferases structure containing
protein
Chemokine (C-C motif) ligand 3
NM 011337
Wiskott-Aldrich syndrome protein
NM_153138
interacting
Niemann Pick type C2
NM 023409
Aquaporin-4 (AQP-4) (WCH4) (MercurialBC024526
insensitive water channel) (MIWC)
Ribosomal protein S14
NM 020600
NM_145619
member 3
AK148837
Tripartite motif-containing 14
Hypothetical CD9/CD37/CD63 antigens
AK020982
containing protein
Cysteine rich protein 61
NM 010516
AEBP1 protein
BC082577
Tumor necrosis factor, alpha-induced
AK154405
protein 2
BC023246
Deoxyribonuclease I-like 1 precursor
Tenascin C (Hexabrachion)
AK085667
Annexin A5
NM 009673
Ephx1 protein
BC057857
Hypothetical protein
AK153119
Toll-like receptor 7 precursor
AK154906
Matrix metalloproteinase 2
NM 008610
doi:10.1038/nn.2779
nature neuroscience
0
0
0
0
0
0
0
0
0
0
4.50
1.50
3.00
2.81
0
4
0
0
0
0
0
0
0
4
0
0
0
0
0
0
0
0
0
0
7.86
49.09
11.71
2.64
3.34
27.55
5.81
0.74
2.35
1.78
2.01
3.57
2.81
2.81
2.79
2.79
0
0
0
0
0
9.44
4.35
2.17
2.66
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7.61
38.83
17.26
62.27
3.38
21.78
9.32
35.74
2.25
1.78
1.85
1.74
2.65
2.65
2.64
2.64
0
0
0
0
0
2.30
0.67
3.44
2.63
1
0
0
0
1
174.71
100.36
1.74
2.63
0
0
0
0
0
10.51
5.16
2.04
2.63
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
55.28
8.49
4.49
31.49
3.85
1.70
1.76
2.20
2.65
2.62
2.61
2.60
0
0
0
0
0
2.38
0.70
3.39
2.59
0
0
0
0
0
3.94
1.33
2.96
2.59
2
0
2
0
0
12.07
6.37
1.90
2.59
0
0
0
0
0
43.30
24.85
1.74
2.59
1
0
1
0
0
35.71
20.55
1.74
2.59
1
0
1
0
0
21.63
12.39
1.75
2.57
0
0
0
0
0
5.57
2.37
2.35
2.57
0
0
0
0
0
2.07
0.64
3.23
2.57
1
0
0
0
1
44.17
25.55
1.73
2.57
0
0
0
0
0
0
0
0
0
0
5.91
8.49
2.59
3.94
2.28
2.16
2.57
2.56
0
0
0
0
0
2.85
0.87
3.27
2.56
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1.97
2.68
21.03
18.49
2.37
6.81
2.85
0.62
0.82
12.04
10.10
0.71
3.09
0.88
3.19
3.28
1.75
1.83
3.33
2.20
3.24
2.56
2.55
2.55
2.54
2.54
2.53
2.53
15
Polymenidou et al
Psmb9
Pqlc3
Mgst1
Ms4a6d
Zc3hav1
Zfp488
Tmem86a
Maff
Hexb
1110007F12Rik
AK051656
Rps26
Plek
Prkcd
AK048742
Pllp
Ppgb
Mpzl1
Aldh1a1
Ninj1
Nek6
BC013712
Olfml3
Atf3
Anxa4
Tspan4
Rpl32
Dbp
Clic1
Col5a3
Vat1
Lyn
Proteosome (prosome, macropain)
subunit, beta
Similar to mannose-P-dolichol utilization
defect 1 protein homolog
Microsomal glutathione S-transferase 1
MS4A6D protein
Zinc finger protein 488
V-maf musculoaponeurotic fibrosarcoma
oncogene
Hexosaminidase B
Protein KIAA1949 homolog
40S ribosomal protein S26
Pleckstrin
Protein kinase C, delta
Similar to Chromodomain-helicase-DNAbinding protein
NM_013585
0
0
0
0
0
2.14
0.66
3.22
2.53
AK138680
0
0
0
0
0
2.42
0.73
3.30
2.53
NM 019946
NM 026835
AK143568
NM 001013777
NM 026436
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
10.11
4.62
3.61
4.02
10.94
4.95
1.80
1.22
1.46
5.60
2.04
2.57
2.95
2.76
1.95
2.52
2.52
2.51
2.51
2.50
NM_010755
0
0
0
0
0
1.67
0.56
2.96
2.50
NM 010422
NM 197986
AK051656
BC100456
NM 019549
NM 011103
0
0
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
75.76
1.49
7.93
27.29
29.57
9.26
44.83
0.53
3.68
16.03
17.32
4.51
1.69
2.81
2.16
1.70
1.71
2.05
2.50
2.48
2.48
2.48
2.48
2.47
AK048742
3
0
3
0
0
4.75
1.90
2.49
2.47
NM_026385
0
0
0
0
0
32.01
18.77
1.71
2.47
Protective protein for beta-galactosidase NM 001038492
Myelin protein zero-like 1
AK090278
Aldehyde dehydrogenase family 1,
NM_013467
subfamily A1
Ninjurin 1
NM 013610
NIMA (never in mitosis gene a)-related
NM_021606
expressed
NM 001033308
Olfactomedin-like 3
NM 133859
Activating transcription factor ATF3b
BC019946
Annexin A4
AK032703
Tetraspanin 4
NM 053082
NM 172086
D site albumin promoter binding protein
NM 016974
Chloride intracellular channel 1
NM 033444
Adipocyte-specific protein 6
AB040491
Vesicle amine transport protein 1 homolog
NM 012037
0
5
0
0
0
5
0
0
0
0
43.83
13.64
26.11
7.37
1.68
1.85
2.46
2.46
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
Tyrosine-protein kinase Lyn (EC 2.7.1.112)
0
0
0
0
0
0
0
1
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
5
0
0
0
0
0
0
0
0
0
0
0
1
0
4
0
0
0
0
0
0
1
0
0
0
0
0
0
0
Transmembrane 4 superfamily member 11
BC031547
NM 011309
S100 calcium binding protein A1
Plasminogen activator, urokinase
NM 008873
Peroxiredoxin 1
NM 011034
Colony stimulating factor 1
NM 007778
NM 126166
G protein-coupled receptor 17
NM 001025381
Integrin beta-5 precursor
BC058246
Protease, serine, 18
NM 011177
Transferrin
NM 133977
Lipopolysaccharide inducible C-C
Ccrl2
NM_017466
chemokine
BC082538
P4ha3
P4ha3 protein
Apobec3
Apolipoprotein B editing complex 3
AK153719
Gns
Glucosamine (N-acetyl)-6-sulfatase
NM 029364
Tcirg1
T-cell, immune regulator 1
NM 016921
Parvg
Parvin, gamma
NM 022321
AK042457
Flavin containing monooxygenase 1
AK042457
Serinc5
Serine incorporator 5
NM 172588
Gap junction epsilon-1 protein (ConnexinGje1
AK147913
29)
Lxn
Latexin
NM 016753
A930021H16Rik
Weakly similar to MICAL-like 2
AK151086
Serine caroboxypeptidase 1
AK153485
Scpep1
Sox9
SRY-box containing gene 9
NM 011448
Nupr1
P8 protein
NM 019738
Pros1
Protein S (alpha)
AK145501
Sfpi1
Transcription factor PU.1
L03215
CASP8 and FADD-like apoptosis regulator
NM_207653
Cflar
isoform
Interleukin-10 receptor beta chain
AK160991
Il10rb
precursor
AF076623
Nes
Nes protein (Fragment)
S100a13
NM 009113
Proteasome (prosome, macropain) 28
Psme1
NM_011189
subunit
NM 146023
Evi2b
Ecotropic viral integration site 2b
Naglu
Alpha-N-acetylglucosaminidase
NM 013792
Glipr2
GLI pathogenesis-related 2
NM 027450
Tyrosine-protein phosphatase non-receptor
AK132509
Ptpn6
type 6
BC060276
BC060276
Filamin C (Gamma-filamin) (Filamin 2)
BC046309
Novel protein
BC046309
Sparc
Secreted acidic cysteine rich glycoprotein
AK148670
Apob48r
Apolipoprotein B48 receptor
NM 138310
Cd300d
MAIR-Iib
AB091768
Tgif
TG-interacting factor
NM 009372
Id3
Inhibitor of DNA binding 3
NM 008321
Litaf
LPS-induced TN factor
NM 019980
Hypothetical SAND/Sp100 containing
AK150373
5830484A20Rik
protein
S100a1
Plau
Prdx1
Csf1
Tlr3
Gpr17
Itgb5
Klk6
Trf
0
0
0
0
0
18.51
10.33
1.79
2.45
0
0
0
0
0
8.57
4.14
2.07
2.44
0
0
10.89
5.66
1.92
2.43
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2.82
15.03
2.40
4.72
5.11
23.83
23.96
8.39
1.45
26.33
0.90
8.31
0.76
1.92
2.16
14.34
14.46
4.02
0.53
15.78
3.12
1.81
3.18
2.46
2.37
1.66
1.66
2.09
2.74
1.67
2.43
2.43
2.43
2.42
2.42
2.42
2.42
2.41
2.41
2.41
6.83
3.17
2.15
2.39
40.59
4.84
40.08
12.14
11.10
24.29
21.94
8.26
135.42
24.20
2.04
23.94
6.65
5.84
14.61
13.15
4.00
82.88
1.68
2.37
1.67
1.82
1.90
1.66
1.67
2.07
1.63
2.39
2.38
2.38
2.38
2.36
2.36
2.35
2.35
2.34
1.39
0.52
2.67
2.34
4.46
5.37
45.72
8.29
5.93
6.68
37.22
1.83
2.44
28.30
4.07
2.79
3.22
22.46
2.44
2.20
1.62
2.04
2.13
2.07
1.66
2.33
2.32
2.32
2.31
2.31
2.31
2.31
2
0
2
0
0
96.07
59.23
1.62
2.31
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
20.88
1.97
15.18
20.07
8.30
11.06
5.42
12.57
0.67
8.64
12.18
4.12
5.90
2.49
1.66
2.93
1.76
1.65
2.01
1.87
2.18
2.31
2.30
2.30
2.30
2.30
2.30
2.29
1
0
0
0
1
13.76
7.83
1.76
2.29
0
0
0
0
0
7.84
3.84
2.04
2.28
0
0
0
0
0
0
0
0
0
0
7.21
12.91
3.52
7.24
2.05
1.78
2.28
2.28
0
0
0
0
0
9.19
4.67
1.97
2.27
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
14.44
8.57
3.49
8.24
4.29
1.30
1.75
2.00
2.68
2.27
2.27
2.27
0
0
0
0
0
5.80
2.76
2.10
2.26
0
0
3
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
2.56
4.31
119.54
1.61
9.78
2.94
16.04
18.08
0.87
1.78
74.49
0.59
5.12
1.03
9.47
10.54
2.94
2.42
1.60
2.70
1.91
2.86
1.69
1.72
2.26
2.26
2.26
2.25
2.25
2.25
2.25
2.24
0
0
0
0
0
1.42
0.55
2.58
2.24
Hcls1
Hematopoietic cell specific Lyn substrate 1
NM_008225
0
0
0
0
0
5.13
2.27
2.26
2.23
Pabpc1
Poly A binding protein, cytoplasmic 1
Ubiquitin-specific protease 2 isoform Usp245
NM 008774
0
0
0
0
0
55.02
34.31
1.60
2.22
NM_198091
1
0
1
0
0
22.60
13.99
1.62
2.22
Usp2
doi:10.1038/nn.2779
nature neuroscience
16
Polymenidou et al
Protein tyrosine phosphatase, receptor
type, C
Sdc4 protein
Clusterin
NM_011210
2
0
1
0
1
4.38
1.86
2.36
2.21
Sdc4
Clu
BC002312
NM 013492
3
1
0
0
0
0
0
0
3
1
46.40
259.47
29.47
163.59
1.57
1.59
2.21
2.20
H2-Eb1
Histocompatibility 2, class II antigen E beta
NM_010382
0
0
0
0
0
2.45
0.85
2.87
2.20
Ikbke
Inhibitor of kappaB kinase epsilon
NM 019777
0
0
0
0
0
1.67
0.62
2.70
2.20
Slc7a2
Solute carrier family 7 (cationic amino acid
NM_007514
4
0
0
0
4
18.08
10.65
1.70
2.20
0
0
0
0
0
29.58
18.32
1.61
2.20
0
0
0
0
0
3.69
1.40
2.64
2.20
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5.58
2.92
10.56
2.69
1.05
5.80
2.08
2.80
1.82
2.19
2.19
2.19
0
0
0
0
0
1.38
0.54
2.54
2.19
0
0
0
0
0
5.33
2.52
2.12
2.19
Ptprc
Lamp2
H2-Aa
Tor3a
Arl11
Hspb8
Stat5a
Rnf122
Tgfbi
Fgl2
Erbb2ip
Slc6a6
Ptgds2
Rgma
Ehd4
Lats2
Add3
Csf2rb1
9230117N10Rik
AK046834
Ctdsp1
BC023356
Gm2a
Rhoc
Chi3l1
Vcam1
Rpl14
Anpep
Ly6e
Pbxip1
Cldn11
Cpne3
Sema6a
Hpse
Il1a
Ltbr
S100a4
Syngr2
Cd83
Pycard
Dap
Lysosomal membrane glycoprotein 2
NM_001017959
isoform 1
Histocompatibility 2, class II antigen A,
NM_010378
alpha
BC071216
Torsin family 3, member A
ADP-ribosylation factor-like 11
NM 177337
Heat shock protein 20-like protein
NM 030704
NM_011488
transcription
Hypothetical RING finger containing
AK171823
protein
Transforming growth factor-beta-induced
AK155828
protein ig-h3 precursor (Beta ig-h3)
Fibrinogen-like protein 2
NM 008013
Erbb2 interacting protein isoform 2
NM 021563
Solute carrier family 6 (neurotransmitter
AK162776
transporter, taurine), member 6
Prostaglandin D2 synthase 2,
NM_019455
hematopoietic
RGM domain family, member A
NM 177740
EH domain containing protein MPAST2
NM 133838
LATS2B
AY015061
Adducin 3 (gamma)
AK144174
Colony stimulating factor 2 receptor, beta
AK170199
1, low-affinity (granulocyte-macrophage)
NM 133775
Interleukin 33
Hypothetical ATP-dependent DNA ligase
AK046834
containing protein
CTD (carboxy-terminal domain, RNA
NM_153088
polymerase II
Transcription factor SOX-10 (SOX-21)
BC023356
(Transcription factor SOX-M)
NM 010299
GM2 ganglioside activator protein
Ras homolog gene family, member C
NM 007484
Chitinase 3-like 1
NM 007695
Vascular cell adhesion molecule 1
AK030195
NM 025974
Alanyl (membrane) aminopeptidase
NM 008486
Lymphocyte antigen Ly-6E precursor
BC005684
Pre-B-cell leukemia transcription factor
NM 146131
Claudin 11
NM 008770
Copine III
NM 027769
Sema domain, transmembrane domain
(TM), and cytoplasmic domain,
AK042751
(semaphorin) 6A
Heparanase
NM 152803
Interleukin 1 alpha
NM 010554
Lymphotoxin B receptor
NM 010736
NM 011311
Synaptogyrin 2
NM 009304
CD83 antigen
BC107344
PYD and CARD domain containing
NM 023258
Death-associated protein
NM 146057
0
0
0
0
0
3.85
1.52
2.54
2.19
0
15
0
0
0
14
0
1
0
0
2.20
52.87
0.76
33.12
2.88
1.60
2.18
2.18
10
0
7
0
3
65.13
41.37
1.57
2.17
0
0
0
0
0
8.10
4.14
1.95
2.17
0
0
3
14
0
0
0
0
0
0
2
14
0
0
0
0
0
0
1
0
21.16
8.51
8.99
38.18
13.15
4.41
4.72
23.73
1.61
1.93
1.90
1.61
2.17
2.16
2.16
2.15
0
0
0
0
0
2.66
0.94
2.84
2.15
0
0
0
0
0
35.00
22.20
1.58
2.14
1
0
1
0
0
6.80
3.39
2.01
2.14
0
0
0
0
0
18.40
11.18
1.65
2.13
0
0
0
0
0
19.60
12.35
1.59
2.10
0
0
0
0
0
1
0
0
1
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
44.63
8.55
7.43
12.23
13.16
3.27
66.62
19.28
121.77
12.09
28.75
4.50
3.80
7.14
7.69
1.25
43.05
12.04
78.83
7.11
1.55
1.90
1.95
1.71
1.71
2.63
1.55
1.60
1.54
1.70
2.10
2.10
2.10
2.10
2.09
2.09
2.09
2.09
2.08
2.07
4
0
4
0
0
10.85
6.10
1.78
2.07
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2.46
2.04
5.40
3.08
6.68
8.06
3.35
5.44
0.91
0.76
2.68
1.19
3.44
4.24
1.35
2.73
2.72
2.69
2.01
2.60
1.94
1.90
2.48
1.99
2.07
2.06
2.05
2.05
2.05
2.05
2.04
2.04
Trp53inp1
Transformation related protein 53 inducible
NM_021897
0
0
0
0
0
9.11
4.94
1.84
2.04
Cotl1
Stom
Pdlim4
Npl
Sgpl1
Coactosin-like protein
Stom protein
Reversion induced LIM gene
N-acetylneuraminate pyruvate lyase
Sphingosine phosphate lyase 1
Voltage-dependent calcium channel
gamma-5
Hypothetical Actin-crosslinking proteins
containing protein
T-cell acute lymphocytic leukemia 1
Adaptin-ear-binding coat-associated
protein 2
Bcl2-associated athanogene 3
Mannosidase alpha class 2B member 2
Vesicle-associated membrane protein 8
Interferon consensus sequence binding
protein 1
transcription
Stress-associated endoplasmic reticulum
protein
Cyclin G1
Loss of heterozygosity, 11, chromosomal
region
BC010249
BC003789
NM 019417
NM 028749
NM 009163
3
0
0
0
1
0
0
0
0
0
3
0
0
0
1
0
0
0
0
0
0
0
0
0
0
42.36
9.75
5.47
5.27
22.27
27.47
5.41
2.73
2.60
14.29
1.54
1.80
2.00
2.03
1.56
2.04
2.04
2.04
2.04
2.04
NM_080644
3
0
3
0
0
7.26
3.80
1.91
2.03
AK090002
1
0
1
0
0
4.32
1.94
2.22
2.03
NM 011527
0
0
0
0
0
1.36
0.57
2.39
2.02
NM_025383
1
0
1
0
0
18.09
11.08
1.63
2.02
NM 013863
NM 008550
NM 016794
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
6.88
16.04
9.46
3.62
9.90
5.26
1.90
1.62
1.80
2.02
2.01
2.01
BC005450
1
0
1
0
0
5.86
3.00
1.95
2.01
Cacng5
4930429M06Rik
Tal1
Necap2
Bag3
Man2b2
Vamp8
Irf8
Stat3
D3Ucla1
Ccng1
Loh11cr2a
doi:10.1038/nn.2779
nature neuroscience
NM_011486
3
0
3
0
0
20.13
12.93
1.56
2.01
NM_030685
1
0
0
0
1
29.42
18.99
1.55
2.01
NM 009831
0
0
0
0
0
35.70
23.34
1.53
2.00
NM_172767
1
0
0
0
1
11.14
6.60
1.69
2.00
17
Polymenidou et al
Supplementary Table 2 Genes downregulated upon TDP-43 depletion in adult mouse brain
CLIP-seq number of clusters
Gene symbol
Protein
Kl
Park2
Fgf14
Sdk1
Efna5
Kcnip4
T-cell lymphoma breakpoint associated
target 1
TAR DNA binding protein isoform 5
Calcium-dependent secretion activator 1
CUB and Sushi multiple domains 1
Ctnna2 protein
Hypothetical carbonic anhydrase structure
containing protein
Calsyntenin 2
Neuregulin 3
Adenylate cyclase 2
Potassium voltage-gated channel, Shalrelated
Low density lipoprotein-related protein 1B
(deleted in tumors)
Calcium channel, voltage-dependent,
alpha2/delta
Neurexin III
Hyaluronan and proteoglycan link protein 4
precursor
Klotho
Parkin
Fibroblast growth factor 14 isoform b
Sidekick homolog 1
Ephrin A5 isoform 2
Kv channel interacting protein 4
Cntnap2
Tmem28
Sntg1
Hs6st3
AK134882
Kcnj3
Hdh
Arc
Rims1
Klf10
Nlgn1
Kcnk4
Folr1
Slc17a8
Tcba1
Tardbp
Cadps
Csmd1
Ctnna2
Car10
Clstn2
Nrg3
2900006F19Rik
Adcy2
Kcnd2
Lrp1b
Cacna2d3
Nrxn3
in
in 5'UTR
introns
RNA-seq
Refseq/mRNA
identifier
in
RPKM in
in 3'UTR
exons
TDP-43 KD
RPKM in
Ratio
Z score
control KD/control
Total
NM_001025286
58
0
57
1
0
3.16
18.06
0.18
-8.17
NM_001003898
BC079679
NM_053171
BC079648
5
56
154
115
0
0
0
0
0
56
152
115
0
0
1
0
5
0
1
0
10.17
7.62
2.51
8.80
43.61
23.80
9.43
22.44
0.23
0.32
0.27
0.39
-6.91
-5.90
-5.39
-4.88
AK012302
52
1
51
0
0
6.05
15.81
0.38
-4.67
NM_022319
NM_008734
NM_028387
NM_153534
38
132
36
15
0
0
0
0
38
131
36
15
0
0
0
0
0
1
0
0
7.70
2.75
2.94
8.57
19.32
8.15
8.52
20.60
0.40
0.34
0.34
0.42
-4.59
-4.47
-4.35
-4.25
NM_019697
78
0
77
0
1
14.60
34.39
0.42
-4.18
AK136118
87
0
87
0
0
2.59
6.87
0.38
-4.08
NM_009785
63
0
63
0
0
12.00
25.88
0.46
-4.05
BC060719
178
0
178
0
0
16.25
37.23
0.44
-3.93
AK034300
0
0
0
0
0
9.71
20.10
0.48
-3.86
NM_013823
AF250294
NM_207667
NM_177879
NM_010109
AK148685
0
39
85
29
21
109
0
0
0
0
0
0
0
39
85
29
21
109
0
0
0
0
0
0
0
0
0
0
0
0
2.53
0.63
6.99
0.51
3.90
3.98
6.33
2.18
14.99
1.54
9.33
9.30
0.40
0.29
0.47
0.33
0.42
0.43
-3.84
-3.78
-3.76
-3.69
-3.69
-3.58
Contactin associated protein-like 2 isoform b
NM_025771
116
0
116
0
0
11.00
21.03
0.52
-3.52
Transmembrane protein 28
Syntrophin, gamma 1
Heparan sulfate 6-O-sulfotransferase 3
Potassium inwardly-rectifying channel J3
Huntington disease gene homolog
Activity regulated cytoskeletal-associated
Rims1 protein
Kruppel-like factor 10
Neuroligin 1
TRAAK
Folate receptor 1
Vesicular glutamate transporter-3
Similar to potassium channel protein erg
long isoform
weakly similar to Keratin 8
Choline acetyltransferase
Regulator of G-protein signaling 6
Slit homolog 3
Glutamate receptor, metabotropic 7
Protein tyrosine phosphatase, receptor type,
N polypeptide 2
NM_173446
NM_027671
NM_015820
AK134882
NM_008426
NM_010414
NM_018790
BC055294
NM_013692
NM_138666
NM_008431
NM_008034
NM_182959
79
23
80
2
21
10
0
6
0
102
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
79
23
79
2
21
9
0
6
0
101
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
6.21
2.01
4.14
2.47
6.57
8.52
25.40
15.32
6.55
8.55
0.76
0.67
1.08
12.27
4.88
8.95
5.55
12.70
16.55
50.51
29.50
12.13
16.28
2.07
1.81
2.84
0.51
0.41
0.46
0.44
0.52
0.51
0.50
0.52
0.54
0.53
0.37
0.37
0.38
-3.49
-3.43
-3.35
-3.33
-3.32
-3.29
-3.29
-3.25
-3.19
-3.19
-3.16
-3.16
-3.16
AK034003
23
0
23
0
0
2.90
5.92
0.49
-3.12
AK171114
NM_009891
NM_015812
NM_011412
NM_177328
0
2
37
40
78
0
0
0
0
0
0
2
37
40
77
0
0
0
0
1
0
0
0
0
0
0.56
0.97
1.68
4.10
8.82
1.39
2.59
4.03
8.39
16.33
0.40
0.37
0.42
0.49
0.54
-3.11
-3.11
-3.10
-3.10
-3.09
AK138655
64
0
63
0
1
18.06
34.41
0.52
-3.08
Lsamp
Limbic system-associated membrane protein
AK140043
71
0
70
0
1
15.54
28.93
0.54
-3.07
Opcml
Opioid binding protein/cell adhesion
Hypothetical Glycosyl transferase, family 2
containing protein
Islet cell autoantigen 1
Procollagen, type VIII, alpha 2
KCNQ5 potassium channel
Mitogen-activated protein kinase 11
Similar to BA93M14.1 (Glypican 5)
Zeta-sarcoglycan (Zeta-SG) (ZSG1)
Astrotactin 2 isoform a
Potassium inwardly-rectifying channel,
subfamily J, member 6
Myosin, light polypeptide 4
Cystine knot-containing secreted protein
Natriuretic peptide receptor 3 isoform b
Carboxyl-terminal PDZ ligand of neuronal
nitric oxide synthase homolog
Kin of IRRE-like 2
Vesicle-associated membrane protein 1
Thyrotropin-releasing hormone degrading
Glial cell line derived neurotrophic factor
Adrenergic receptor, alpha 1d
Weakly similar to uridine phosphorylase
TAFA1 protein
Neurexin-1-alpha precursor (Neurexin Ialpha)
Hypothetical Proline-rich region
profile/Serine-rich region profile containing
protein
NM_177906
116
0
115
0
1
18.55
35.28
0.53
-3.06
BC111102
23
0
23
0
0
0.63
1.54
0.41
-3.00
NM_010492
NM_199473
AK147264
NM_011161
AK084223
AK136779
NM_019514
NM_130880
16
0
36
0
24
56
72
30
0
0
0
0
0
1
0
0
16
0
36
0
24
55
72
30
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4.83
0.61
14.96
5.36
3.87
0.74
5.08
4.47
9.23
1.46
26.18
9.86
7.69
1.85
9.45
8.80
0.52
0.42
0.57
0.54
0.50
0.40
0.54
0.51
-2.99
-2.98
-2.97
-2.97
-2.96
-2.95
-2.95
-2.95
Hapln4
Kcnh7
AK171114
Chat
Rgs6
Slit3
Grm7
Ptprn2
4930431L04Rik
Ica1
Col8a2
Kcnq5
Mapk11
Gpc5
Sgcz
Astn2
Otud7
Kcnj6
Myl4
Sostdc1
Npr3
AK138471
Kirrel2
Vamp1
Trhde
Gfra2
C630028F04Rik
Adra1d
Upp2
C630007B19Rik
Nrxn1
AB125594
doi:10.1038/nn.2779
nature neuroscience
AK044218
28
0
27
0
1
3.28
6.37
0.52
-2.92
NM_010858
NM_025312
NM_001039181
0
0
2
0
0
0
0
0
2
0
0
0
0
0
0
2.83
0.75
0.64
5.55
1.82
1.52
0.51
0.41
0.43
-2.91
-2.90
-2.89
AK138471
38
0
38
0
0
5.60
9.97
0.56
-2.88
NM_172898
BC049902
NM_146241
NM_008115
NM_172885
NM_013460
BC027189
NM_182808
0
1
17
6
43
0
23
56
0
1
0
0
0
0
0
0
0
0
17
6
43
0
23
56
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0.86
19.15
2.82
9.38
3.15
0.99
0.77
1.93
2.12
34.73
5.43
16.06
5.93
2.30
1.77
4.10
0.41
0.55
0.52
0.58
0.53
0.43
0.43
0.47
-2.85
-2.83
-2.81
-2.80
-2.79
-2.77
-2.76
-2.76
AB093249
162
0
162
0
0
18.66
33.28
0.56
-2.76
AB125594
45
0
43
0
2
9.06
15.71
0.58
-2.76
18
Polymenidou et al
Ptprt
Grm8
Siglech
Ryr2
Nrn1
A830018L16Rik
Smyd3
Tssc1
AK161480
Trpc3
Gabra3
Aqp1
Plcb1
Kcnk4
Kcnma1
Protein tyrosine phosphatase, receptor type,
T
Glutamate receptor, metabotropic 8
SIGLEC-like protein
Ryr2 protein
Neuritin 1
VEST-1 protein homolog
SET and MYND domain containing 3
Tumor suppressing subtransferable
candidate 1
Transient receptor potential cation channel
Gamma-aminobutyric acid (GABA-A)
receptor
Aquaporin 1
Plcb1 protein
Potassium channel subfamily K member 4
Calcium-activated potassium channel alpha
subunit 1
NM_021464
74
0
74
0
0
8.13
14.29
0.57
-2.75
NM_008174
NM_178706
BC043140
NM_153529
AK135047
NM_027188
21
0
41
2
8
39
0
0
0
0
0
0
21
0
41
2
8
39
0
0
0
0
0
0
0
0
0
0
0
0
1.18
6.14
1.82
17.81
15.04
4.41
2.67
10.50
3.83
31.03
24.75
8.20
0.44
0.59
0.47
0.57
0.61
0.54
-2.75
-2.74
-2.73
-2.73
-2.73
-2.72
NM_201357
6
0
6
0
0
4.51
8.29
0.54
-2.72
AK161480
NM_019510
0
3
0
0
0
3
0
0
0
0
0.55
3.69
1.18
6.83
0.46
0.54
-2.71
-2.71
NM_008067
2
0
2
0
0
15.39
25.51
0.60
-2.71
NM_007472
BC058710
BC061075
0
59
0
0
0
0
0
59
0
0
0
0
0
0
0
0.67
24.54
0.57
1.44
44.03
1.21
0.46
0.56
0.47
-2.68
-2.68
-2.67
U09383
96
0
96
0
0
8.34
14.36
0.58
-2.66
Cacna1c
Calcium channel, voltage-dependent, L type,
NM_009781
77
0
77
0
0
5.57
9.56
0.58
-2.65
Slc10a4
E030025D05Rik
Ttr
E030025D05Rik protein
Transthyretin
Zinc transporter 3 (ZnT-3) (Solute carrier
family 30 member 3)
Potassium voltage-gated channel,
Two cut domains-containing homeodomain
protein
Kin of IRRE-like protein 3 precursor (Kin of
irregular chiasm-like protein 3)
Coagulation factor V
Similar to neurula-specific ferrodoxin
reductase-like protein homolog
Adenylate cyclase 1
Glutamate receptor, ionotropic, NMDA2A
(epsilon)
NM_173403
AK173298
NM_013697
0
20
0
0
1
0
0
18
0
0
0
0
0
1
0
1.24
6.06
44.88
2.70
10.10
80.56
0.46
0.60
0.56
-2.65
-2.62
-2.62
Slc30a3
Kcns3
4933439F18Rik
Satb2
Kirrel3
F5
2810401C16Rik
Adcy1
A130090K04Rik
Grin2a
Cacna1e
Calcium channel, voltage-dependent, R type
BC066199
0
0
0
0
0
5.66
9.55
0.59
-2.60
NM_173417
NM_025757
0
1
0
1
0
0
0
0
0
0
0.58
29.33
1.20
51.30
0.48
0.57
-2.60
-2.59
NM_139146
5
0
5
0
0
6.79
10.98
0.62
-2.58
AK045373
74
0
74
0
0
6.02
9.99
0.60
-2.58
NM_007976
0
0
0
0
0
0.94
2.08
0.45
-2.57
AK158809
3
0
2
0
1
10.73
17.93
0.60
-2.57
NM_009622
NM_001033391
17
3
0
0
12
3
0
0
5
0
38.58
2.68
68.45
4.95
0.56
0.54
-2.56
-2.56
NM_008170
83
0
83
0
0
2.19
4.23
0.52
-2.56
NM_009782
40
0
34
6
0
8.43
14.18
0.59
-2.55
81
8
0
0
81
8
0
0
0
0
16.54
7.73
27.53
12.71
0.60
0.61
-2.54
-2.53
59
1
57
0
1
7.07
11.27
0.63
-2.53
3
0
3
0
0
0.61
1.24
0.49
-2.53
39
0
39
0
0
0.57
1.16
0.49
-2.52
2
3
0
0
2
3
0
0
0
0
10.21
5.59
16.56
9.29
0.62
0.60
-2.51
-2.51
28
0
28
0
0
35.79
62.69
0.57
-2.51
0
0
0
0
0
0.52
1.06
0.50
-2.50
28
0
28
0
0
20.31
34.26
0.59
-2.48
26
62
110
34
0
0
0
0
26
62
109
34
0
0
1
0
0
0
0
0
2.37
1.14
18.54
2.34
4.38
2.38
30.84
4.32
0.54
0.48
0.60
0.54
-2.48
-2.47
-2.47
-2.47
52
0
52
0
0
17.00
27.85
0.61
-2.46
20
0
20
0
0
15.37
23.88
0.64
-2.46
3
0
3
0
0
7.38
11.76
0.63
-2.45
0
2
2
4
4
0
0
0
0
0
0
2
2
4
3
0
0
0
0
0
0
0
0
0
1
15.86
10.60
8.94
7.34
18.66
24.95
16.93
14.67
11.59
30.56
0.64
0.63
0.61
0.63
0.61
-2.45
-2.45
-2.42
-2.41
-2.41
63
0
63
0
0
9.67
15.47
0.62
-2.41
0
75
0
0
0
0
0
74
0
0
0
0
0
1
0
2.49
10.39
1.45
4.46
16.43
2.83
0.56
0.63
0.51
-2.40
-2.40
-2.39
Fibroblast growth factor 12 isoform b
NM_010199
Weakly similar to copine VII
AK014396
Weakly similar to heterogeneous nuclear
BC052358
0710005M24Rik
ribonucleoproteins C1/C2 (hnRNPC1/C2)
NM_007733
Col19a1
Procollagen, type XIX, alpha 1
Hypothetical calcium-binding EGF-like
AK018073
AK018073
domain containing protein homolog
G protein-coupled receptor 26
NM_173410
Gpr26
Cpne9
Similar to copine-like protein KIAA1599
AK044080
Hypothetical EF-hand/Phorbol
esters/diacylglycerol binding
AK083222
Dgkb
domain/Cytochrome c family heme-binding
site containing protein
Leucine-rich repeat-containing protein 17
BC030317
BC030317
precursor
Calcium/calmodulin-dependent protein
NM_009793
Camk4
kinase IV
NM_011855
Odz1
Odd Oz/ten-m homolog 1
Wwox
WW-domain oxidoreductase
NM_019573
Igsf4d
Weakly similar to BK134P22.1
AK046800
AK138164
NB-2 homolog
AK138164
Glutamate receptor, ionotropic, kainate 2
NM_010349
Grik2
(beta)
NM_001033399
Gfod1
Glutamate receptor, ionotropic, NMDA2C
NM_010350
Grin2c
(epsilon)
BC014845
Klc2
Klc2 protein
Ttc15
Weakly similar to CGI-87 PROTEIN
AK031763
L42339
Sodium channel 3
L42339
Rtn4r
Nogo receptor
NM_022982
Syt13
Synaptotagmin XIII
NM_030725
Weakly similar to Hypothetical band 4.1
AK147370
Pdzd10
family containing protein
NM_021306
Ecel1
Damage-induced neuronal endopeptidase
Dpp10
Dipeptidyl peptidase 10
NM_199021
Fbxw8
F-box protein FBX29
AK015338
Collagen and calcium binding EGF domains
NM_178793
Ccbe1
1
NM_053195
Slc24a3
Snrp70
U1 small nuclear ribonucleoprotein 70 kDa
NM_009224
Podxl2
Podocalyxin-like 2
NM_176973
A930001M12Rik
NM_177175
Gpr158
G protein-coupled receptor 158 isoform b
NM_175706
Zfp180
Zfp180 protein
BC063741
Cbln2
Cerebellin 2 precursor protein
NM_172633
Stx1a
Syntaxin 1A
NM_016801
Neural cell adhesion molecule 2 precursor
AF001287
Ncam2
(N-CAM 2) (RB-8 neural cell adhesion
molecule)
AK145822
Fosl2
Fos-like antigen 2
Fgf12
Cpne4
doi:10.1038/nn.2779
nature neuroscience
4
0
3
0
1
0.61
1.18
0.52
-2.37
17
2
1
0
20
3
0
1
0
0
0
0
0
1
0
0
17
1
1
0
20
2
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
13.17
69.49
12.22
0.85
29.70
16.63
3.66
24.86
20.23
117.85
18.69
1.68
48.92
26.55
6.17
41.33
0.65
0.59
0.65
0.51
0.61
0.63
0.59
0.60
-2.36
-2.36
-2.35
-2.34
-2.34
-2.34
-2.33
-2.33
37
0
37
0
0
7.79
12.23
0.64
-2.33
1
0
1
0
0
6.83
10.53
0.65
-2.32
19
Polymenidou et al
Pcdh7
Pde1a
Tnfrsf25
Apba2bp
Kcns2
BC052055
Alas2
D14Ertd171e
Magi2
Galnt9
Nr4a3
Cntn4
Atp2b3
Kcnj9
Large
Accn1
Rasgrf2
Xkr4
Cabp1
Scube1
Ephb6
AK147314
Wnt9a
2900046G09Rik
Kcnab3
Atrnl1
Kcnmb4
Nr4a1
Phospho1
Bai3
Reps2
Mamdc1
Emx1
Slit2
Rgs7
Rims4
Osbpl10
Slc5a7
Tbr1
Rfx3
Lynx1
Itpka
Irxl1
Sema5b
Hnt
Prkag1
Chrm3
3110035E14Rik
2610044O15Rik
Rit2
AK142702
Egr4
Mical2
Sstr3
Ldlr
Hcn1
E330039K12Rik
Lrrn6c
BC107398
Sidt1
Oprd1
Insig1
Mib1
B230343H07Rik
Fxyd7
Unc5d
Me3
Cacna1a
S69381
Protocadherin 7
NM_018764
Phosphodiesterase 1A, calmodulinNM_001009978
dependent
BC017526
Tnfrsf25 protein
Amyloid beta A4 protein-binding family A
AK013520
member 2-binding protein (X11L-binding
protein 51) (mXB51)
NM_181317
K+ voltage-gated channel, subfamily S, 2
NM_182636
Aminolevulinic acid synthase 2, erythroid
NM_009653
CAZ-associated structural protein
NM_177814
Activin receptor interacting protein 1
AK147530
UDP-N-acetyl-alpha-DNM_198306
galactosamine:polypeptide
Orphan nuclear receptor NR4A3 (Orphan
BC068150
nuclear receptor TEC) (Translocated in
extraskeletal chondrosarcoma)
AK163520
Contactin 4
Plasma membrane calcium ATPase 3
NM_177236
G protein-activated inward rectifier
potassium channel 3 (GIRK3) (Potassium
channel, inwardly rectifying subfamily J
BC065161
member 9) (Inwardly rectifier K(+) channel
Kir3.3)
like-glycosyltransferase
NM_010687
Accn1 protein
BC038551
RAS protein-specific guanine nucleotideAK158472
releasing factor 2
NM_001011874
XK-related protein 4
calcium binding protein 1
NM_013879
Scube1 protein
BC066066
Eph receptor B6
NM_007680
Delta kalirin-7 homolog
AK147314
Wnt9a protein
BC066165
NM_133778
Voltage-gated potassium channel beta-3
NM_010599
subunit (K(+) channel beta-3 subunit) (Kvbeta-3)
NM_181415
Attractin-like 1
Calcium activated potassium channel beta 4
NM_021452
Nuclear receptor subfamily 4, group A,
NM_010444
member 1
Phosphatase, orphan 1
NM_153104
Brain-specific angiogenesis inhibitor 3
BC032251
RalBP1 associated Eps domain containing
AK147418
protein
Hypothetical Phenylalanine-rich region
AK163637
profile containing protein
NM_010131
Empty spiracles homolog 1
Slit homolog 2 protein precursor (Slit-2)
AK220505
Regulator of G protein signaling 7
NM_011880
Regulating synaptic membrane exocytosis 4
NM_183023
Oxysterol-binding protein-like protein 10
NM_148958
Solute carrier family 5 (choline transporter)
NM_022025
T-box brain gene 1
AK158835
Regulatory factor X, 3
NM_011265
Ly6/neurotoxin 1
NM_011838
Inositol 1,4,5-trisphosphate 3-kinase A
NM_146125
NM_177595
Semaphorin-5B precursor (Semaphorin G)
BC052397
Neurotrimin precursor
BC023307
AMP-activated protein kinase subunit
AK157795
gamma 1
NM_033269
Cholinergic receptor, muscarinic 3, cardiac
NM_178399
AK137318
Ras-like without CAAX 2
NM_009065
AK142702
Early growth response 4
NM_020596
Flavoprotein oxidoreductase MICAL2
NM_177282
Somatostatin receptor 3
NM_009218
Low density lipoprotein receptor
AK170264
Hyperpolarization-activated, cyclic
NM_010408
Hypothetical homeodomain-like structure
AK220342
containing protein
NM_175516
Leucine rich repeat neuronal 6C
BC107398
SID1 transmembrane family, member 1
NM_198034
Opioid receptor, delta 1
NM_013622
Insulin induced gene 1
NM_153526
Ubiquitin ligase mind bomb
NM_144860
RIKEN cDNA B230343H07
NM_001037906
FXYD domain-containing ion transport
NM_022007
regulator
NM_153135
Netrin receptor Unc5h4
Hypothetical protein FLJ34862 (Malic
AK133593
enzyme) homolog
Calcium channel, voltage-dependent, P/Q
AK052625
type, alpha 1A subunit
Potassium voltage-gated channel subfamily
S69381
C member 3
doi:10.1038/nn.2779
nature neuroscience
0
0
0
0
0
8.97
14.37
0.62
28
0
28
0
0
11.79
17.86
0.66
-2.32
-2.32
1
0
1
0
0
1.64
3.11
0.53
-2.32
0
0
0
0
0
6.64
10.29
0.65
-2.31
0
0
0
62
62
0
0
0
0
0
0
0
0
61
62
0
0
0
0
0
0
0
0
1
0
6.22
4.33
0.53
17.51
15.73
9.68
7.20
1.00
27.80
23.87
0.64
0.60
0.53
0.63
0.66
-2.30
-2.30
-2.30
-2.30
-2.30
9
0
9
0
0
5.48
8.80
0.62
-2.30
0
0
0
0
0
4.32
7.17
0.60
-2.30
37
2
0
0
37
2
0
0
0
0
6.24
15.15
9.63
23.00
0.65
0.66
-2.29
-2.28
0
0
0
0
0
14.13
21.28
0.66
-2.28
38
110
0
0
38
110
0
0
0
0
17.42
5.29
27.47
8.55
0.63
0.62
-2.28
-2.28
1
0
1
0
0
5.30
8.57
0.62
-2.27
42
2
4
1
33
1
0
0
0
0
0
0
0
0
42
2
4
1
33
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3.75
11.29
3.63
13.80
38.00
1.75
24.10
6.24
17.56
5.88
20.81
63.06
3.26
39.57
0.60
0.64
0.62
0.66
0.60
0.54
0.61
-2.27
-2.26
-2.26
-2.26
-2.25
-2.25
-2.25
0
0
0
0
0
4.85
7.87
0.62
-2.25
37
4
0
0
37
4
0
0
0
0
11.96
8.43
17.88
12.95
0.67
0.65
-2.24
-2.23
0
0
0
0
0
26.01
42.41
0.61
-2.23
0
91
0
0
0
91
0
0
0
0
1.90
13.83
3.41
20.65
0.56
0.67
-2.22
-2.21
-2.21
4
0
4
0
0
25.57
41.37
0.62
74
1
73
0
0
5.78
8.96
0.65
-2.21
0
10
55
5
27
0
2
30
0
0
5
1
127
0
0
1
0
0
0
0
0
0
0
0
0
0
0
10
54
5
27
0
0
30
0
0
5
1
127
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0.99
2.05
9.76
3.62
3.10
3.50
15.44
3.30
62.35
17.25
1.79
3.97
21.16
1.89
3.63
15.02
5.80
5.03
5.60
23.26
5.27
101.97
26.73
3.27
6.42
33.56
0.53
0.56
0.65
0.62
0.62
0.62
0.66
0.63
0.61
0.65
0.55
0.62
0.63
-2.20
-2.20
-2.20
-2.20
-2.20
-2.20
-2.19
-2.19
-2.19
-2.19
-2.19
-2.18
-2.18
0
0
0
0
0
2.04
3.59
0.57
-2.18
42
6
1
10
0
0
14
0
1
23
17
0
0
0
0
0
0
0
0
0
42
5
1
10
0
0
14
0
1
23
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
18.85
40.66
9.02
12.49
0.84
6.65
20.36
2.43
3.78
4.69
29.40
66.20
13.90
18.40
1.54
9.92
31.81
4.08
6.03
7.45
0.64
0.61
0.65
0.68
0.55
0.67
0.64
0.60
0.63
0.63
-2.17
-2.17
-2.16
-2.16
-2.16
-2.15
-2.15
-2.15
-2.14
-2.13
19
0
19
0
0
10.34
15.48
0.67
-2.13
1
1
13
0
0
15
19
0
0
0
0
0
0
0
1
1
13
0
0
12
19
0
0
0
0
0
2
0
0
0
0
0
0
1
0
2.24
1.04
3.38
7.33
9.29
14.45
3.48
3.84
1.93
5.31
10.66
14.22
21.16
5.48
0.58
0.54
0.64
0.69
0.65
0.68
0.64
-2.13
-2.13
-2.12
-2.12
-2.12
-2.12
-2.11
-2.11
1
0
1
0
0
4.44
7.04
0.63
45
0
45
0
0
1.48
2.62
0.57
-2.11
2
0
2
0
0
5.74
8.73
0.66
-2.10
35
0
35
0
0
20.73
32.14
0.65
-2.10
0
0
0
0
0
19.96
30.98
0.64
-2.10
20
Polymenidou et al
Neurotractin isoform a
NM_001039094
Similar to CUB and sushi multiple domains
AK144913
protein 2
ST6
NM_011372
St6galnac3
Serpinb8
Serine protease inhibitor 8
AK048230
Rnf152
RING finger protein 152
AK035832
AI894139
Weakly similar to zinc finger protein ZNF65
AK133549
Unc13a
Weakly similar to Munc- 13
BC058348
Dlgap1
Disks large-associated protein 1 (DAP-1)
AK163689
Klf16
Kruppel-like factor 16
NM_078477
Sodium/calcium exchanger 1 precursor
NM_011406
Slc8a1
(Na(+)/Ca(2+)-exchange protein 1)
Early growth response 2
NM_010118
Egr2
Dhcr24
24-dehydrocholesterol reductase
NM_053272
UDP-N-acetyl-alpha-DNM_173030
Galnt13
galactosamine:polypeptide
Mesenchymal stem cell protein DSC54
AK046456
AK046456
homolog
NM_172803
Dock4
Dedicator of cytokinesis 4
BC025575
NM_199200
Ncald
Neurocalcin delta
NM_134094
Neurofilament triplet M protein (160 kDa
AK051696
Nef3
neurofilament protein)
NM_009717
Neurod6
Neurogenic differentiation 6
Sema7a
Semaphorin 7A
NM_011352
Syt16
Synaptotagmin 14-like
AK137129
Rims3
AK122226
Sodium channel, voltage-gated, type I, beta
AK028238
AK028238
polypeptide
NM_025404
Arfl4
ADP-ribosylation factor 4-like
Hypothetical P-loop containing nucleotide
AK034089
triphosphate hydrolases structure containing
AK034089
protein
BC103530
Cckbr
Gastrin/cholecystokinin type B receptor
Itpr1
Inositol 1,4,5-triphosphate receptor 1
AK049015
BC089626
Protein C6orf142 homolog
BC089626
Eda
Ectodysplasin A (EDA protein homolog)
AJ243657
Protein kinase, cGMP-dependent, type I
NM_001013833
Prkg1
alpha
U7 snRNA-associated Sm-like protein
Lsm11
NM_028185
Lsm11
3-hydroxy-3-methylglutaryl-Coenzyme A
Hmgcr
NM_008255
reductase
NM_177056
A230078I05Rik
5730507A09Rik
NM_183087
Potassium voltage gated channel, ShawNM_001025581
Kcnc2
related
Negr1
92
0
91
1
0
28.02
44.28
0.63
AK144913
2
0
2
0
0
2.04
3.48
0.59
-2.09
60
2
10
1
15
83
0
0
0
0
0
2
1
0
60
2
10
1
13
82
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1.31
1.45
0.64
3.61
40.18
25.05
14.71
2.32
2.56
1.11
5.65
64.15
39.32
21.28
0.56
0.57
0.58
0.64
0.63
0.64
0.69
-2.09
-2.09
-2.08
-2.08
-2.08
-2.06
-2.06
36
0
36
0
0
8.61
12.73
0.68
-2.06
0
0
0
0
0
0
0
0
0
0
3.64
6.85
5.59
10.00
0.65
0.69
-2.06
-2.06
26
0
26
0
0
9.46
14.28
0.66
-2.06
20
0
20
0
0
1.63
2.86
0.57
-2.05
47
0
34
0
0
0
46
0
34
0
0
0
1
0
0
14.46
17.20
35.45
20.85
25.89
55.75
0.69
0.66
0.64
-2.05
-2.05
-2.03
doi:10.1038/nn.2779
nature neuroscience
-2.09
0
0
0
0
0
43.91
69.36
0.63
-2.03
0
1
17
7
0
0
0
0
0
1
17
7
0
0
0
0
0
0
0
0
9.52
26.85
10.45
22.53
14.29
41.81
15.29
34.57
0.67
0.64
0.68
0.65
-2.02
-2.02
-2.02
-2.02
3
0
3
0
0
35.01
55.05
0.64
-2.02
0
0
0
0
0
14.88
21.39
0.70
-2.02
6
0
6
0
0
1.05
1.86
0.57
-2.02
0
31
9
1
0
1
0
0
0
28
8
1
0
0
1
0
0
2
0
0
7.29
69.56
3.45
0.71
10.40
109.54
5.26
1.20
0.70
0.63
0.66
0.59
-2.02
-2.01
-2.01
-2.01
46
0
46
0
0
0.77
1.31
0.59
-2.01
0
0
0
0
0
0.99
1.72
0.57
-2.01
2
0
1
1
0
13.27
18.85
0.70
-2.01
0
2
0
0
0
2
0
0
0
0
7.82
4.64
11.29
7.12
0.69
0.65
-2.00
-2.00
27
0
26
1
0
8.75
12.79
0.68
-2.00
21
Polymenidou et al
Supplementary Table 3 ncRNAs with changes in expression levels upon TDP-43 depletion in adult mouse brain
RNAdb
index
LIT3360
LIT1822
LIT1602
LIT1604
LIT3442
LIT3440
LIT3441
LIT1799
LIT1623
LIT1749
LIT3409
LIT3410
LIT3411
LIT3368
LIT3422
LIT3369
LIT3366
LIT3371
LIT2029
LIT2028
LIT3352
LIT3351
LIT3362
LIT1787
LIT1651
LIT1387
LIT1111
LIT1671
LIT3367
LIT3323
LIT1806
LIT1647
name
reverse delta5-desaturase RNA
(Revdelta5ase)
Kruppel-type zinc-finger protein
ZIM3-like mRNA, complete
sequence
Tsix gene
domesticus antisense RNA
from the Xist locus, complete
sequence
C030002C11Rik gene
D630033A02Rik gene
2210403K04Rik gene
clone MBI-109 miscellaneous
RNA, partial sequence
Rian mRNA, imprinted gene
Mirg mRNA
Otx2 opposite strand (Otx2OS)
RNA, variant isoform
RIKEN cDNA 2810037O22
gene
Dleu2 mRNA, partial sequence,
alternatively spliced
retinal noncoding RNA 3
(RNCR3)
RIKEN cDNA 3010002C02
gene
(RNCR1)
vault RNA, complete sequence
vault-associated RNA
sequence
metastasis associated lung
adenocarcinoma transcript 1
(non-coding RNA), mRNA
(cDNA clone IMAGE:3582796)
metastasis associated lung
adenocarcinoma transcript 1
(non-coding RNA), mRNA
(cDNA clone IMAGE:3601939)
Hepcarcin RNA, complete
sequence
clone MBI-32 miscellaneous
clone MBII-386 miscellaneous
U22 snoRNA host gene (UHG)
gene, complete sequence
empty spiracles 2 opposite
strand (EMX2OS)
Nespas mRNA
RIKEN cDNA 2610100L16
gene
RNA component of
mitochondrial RNAase P (Rmrp)
on chromosome 4.
Chr
Start
Stop
RPKM in
strand
TDP-43 KD
RNA-seq
RPKM in
RPKM in
control
saline
Change
chr19 10249682
10250370
-
3.00
0.12
0.12
up
chr7
6559771
6580748
-
0.62
0.01
0.01
up
chrX
99634235
99687675
+
4.45
0.08
0.11
up
chrX
99634235
99687675
+
4.45
0.08
0.11
up
chr1 196724832 196738628
chr10 122388136 122388891
chr11 75277842 75282884
+
+
2.96
0.64
1.29
8.63
1.72
3.58
6.81
1.95
3.11
down
down
down
chr12 90196061
90196315
+
0.19
2.00
1.95
down
chr12 110093968 110100034
chr12 110182787 110197261
+
+
6.43
2.12
20.01
7.39
17.72
6.70
down
down
chr14 47591238
47715298
+
0.01
0.02
0.01
down
chr14 47592109
47641765
+
0.01
0.02
0.01
down
chr14 47595527
47783406
+
0.01
0.02
0.01
down
chr14 59340093
59342302
+
4.89
15.56
18.00
down
chr14 60557141
60636460
-
0.39
0.88
1.02
down
chr14 63542224
63547029
+
11.49
42.49
35.15
down
chr16
8767567
+
6.62
13.75
18.15
down
8766673
chr17 85515685
85526691
-
0.61
1.41
2.15
down
chr18 36927839
36927979
+
2.96
8.92
3.67
down
chr18 36927840
36927980
+
2.96
8.92
3.67
down
chr19
5537030
5801973
-
3.61
13.25
10.70
down
chr19
5625873
5801971
-
5.29
19.76
15.91
down
chr19
5795689
5802671
-
123.89
522.39
428.33
down
chr19
5796492
5796672
-
97.17
415.23
430.62
down
chr19
5801468
5801548
-
53.60
219.41
222.85
down
chr19
8791184
8793485
+
2.73
7.76
7.07
down
59511885
-
0.08
0.21
0.21
down
chr2 173909368 173909996
-
0.46
1.10
1.68
down
chr3
17982103
17991742
+
2.07
4.94
7.09
down
chr4
43513885
43514159
-
11.64
40.17
29.24
down
chr4 108974503 108974582
-
15.05
38.10
3.63
down
chr4 131581604 131581673
+
2.12
8.30
8.88
down
chr19 59478353
doi:10.1038/nn.2779
nature neuroscience
22
Polymenidou et al
LIT1313
LIT1314
LIT1808
LIT1345
LIT3370
LIT3450
LIT1225
LIT1226
LIT2020
LIT2015
LIT3416
LIT3432
LIT1675
LIT1355
LIT1629
LIT1703
LIT3324
LIT3329
LIT1010
LIT1761
LIT1009
LIT1760
LIT1008
LIT1006
LIT1003
LIT1001
RNA transcript from U17 small
nucleolar RNA host gene
U17HG gene
makorin 1 pseudogene mRNA,
partial sequence
(RNCR2)
Evf1 RNA, full length
Mit1/Lb9 mRNA, partial
sequence
molossinus Mit1/Lb9 gene,
partial sequence
Y3 scRNA gene, partial
sequence.
Y1 scRNA gene, partial
sequence
ROSA 26 transcription 1 mRNA
sequence
UBE3A-ATS transcript
containing exon L
Ipw gene
Ipw mRNA, partial sequence
Pwcr1 mRNA, complete
sequence
KvLQT1-AS allele (LIT1)
7SK class III RNA gene
inactive X specific transcripts
(Xist) on chromosome X,
transcript variant 1
inactive X specific transcripts
(Xist) on chromosome X,
transcript variant 2
Jpx mRNA, partial sequence
Ftx noncoding RNA, exons 6-7
chr4 131624009 131625709
-
1.42
3.89
3.02
down
chr4 131624010 131625709
-
1.42
3.89
3.02
down
chr4 148151418 148151511
-
17.86
42.33
13.45
down
chr5
89913208
89913907
-
0.48
1.69
1.18
down
chr5
112453533 112456585
-
9.34
37.31
18.35
down
chr6
6771701
6811661
-
0.35
1.27
1.82
down
chr6
30736982
30738860
+
12.42
34.37
39.93
down
chr6
30738334
30738754
+
13.80
39.68
46.32
down
chr6
47711224
47711289
+
741.35
2636.11
1178.13
down
chr6
47717676
47717749
-
703.99
2681.21
1067.28
down
chr6
113036182 113042974
-
1.12
2.43
2.52
down
chr7
59091716
59094070
-
1.32
4.60
3.67
down
chr7
chr7
59483753
59487027
59544353
59544354
-
0.03
0.39
0.10
1.50
0.08
1.23
down
down
chr7
59739300
59744327
-
0.07
0.26
0.12
down
chr7
chr9
143103111 143107841
77960986 77961317
-
0.23
37.46
0.70
124.99
0.62
80.77
down
down
chrX
99663092
99685952
-
12.17
37.86
38.29
down
chrX
99663092
99685952
-
12.17
37.86
38.29
down
chrX
chrX
chrX
chrX
chrX
chrX
chrX
chrX
99696298
99696350
99696350
99696383
99696384
99809719
99810272
99810325
99696574
99705674
99699246
99705680
99705680
99817912
99826413
99812065
+
+
+
+
+
-
1.64
0.42
0.33
0.39
0.39
0.19
0.19
0.17
5.92
1.04
0.98
0.94
0.94
0.42
0.38
0.47
4.94
0.28
3.13
0.90
0.90
0.63
0.50
4.06
down
down
down
down
down
down
down
down
doi:10.1038/nn.2779
nature neuroscience
23
Polymenidou et al
Supplementary Table 4 Enriched Gene Ontology terms in TDP-43 target RNAs that are downregulated upon TDP-43 depletion in adult brain
Gene
Ontology
Category
Molecular
Function
Cellular
Component
Biological
Process
Enriched Gene Ontology Term
GO:0005216
GO:0022838
GO:0015267
GO:0022803
GO:0005261
GO:0022836
GO:0046873
GO:0005509
GO:0022843
GO:0005244
GO:0022832
GO:0005262
GO:0022834
GO:0015276
GO:0031420
GO:0030955
GO:0005267
GO:0005249
GO:0005246
GO:0005245
GO:0016247
GO:0043167
GO:0046872
GO:0043169
GO:0008066
GO:0005886
GO:0031224
GO:0045202
GO:0044456
GO:0016021
GO:0031225
GO:0044459
GO:0030054
GO:0042734
GO:0045211
GO:0034703
GO:0005887
GO:0043005
GO:0034702
GO:0031226
GO:0014069
GO:0006811
GO:0007268
GO:0030001
GO:0019226
GO:0007267
GO:0050877
GO:0006812
GO:0044057
GO:0006816
GO:0015674
GO:0050804
GO:0046903
GO:0048878
GO:0051969
GO:0006873
GO:0031644
GO:0055082
GO:0032940
GO:0006813
GO:0050801
GO:0006887
GO:0019725
GO:0030182
GO:0015672
GO:0006836
GO:0042391
GO:0007610
GO:0048666
GO:0031175
GO:0007186
ion channel activity
substrate specific channel activity
channel activity
passive transmembrane transporter activity
cation channel activity
gated channel activity
metal ion transmembrane transporter activity
calcium ion binding
voltage-gated cation channel activity
voltage-gated ion channel activity
voltage-gated channel activity
calcium channel activity
ligand-gated channel activity
ligand-gated ion channel activity
alkali metal ion binding
potassium ion binding
potassium channel activity
voltage-gated potassium channel activity
calcium channel regulator activity
voltage-gated calcium channel activity
channel regulator activity
ion binding
metal ion binding
cation binding
glutamate receptor activity
plasma membrane
intrinsic to membrane
synapse
synapse part
integral to membrane
anchored to membrane
plasma membrane part
cell junction
presynaptic membrane
postsynaptic membrane
cation channel complex
integral to plasma membrane
neuron projection
ion channel complex
intrinsic to plasma membrane
postsynaptic density
ion transport
synaptic transmission
metal ion transport
transmission of nerve impulse
cell-cell signaling
neurological system process
cation transport
regulation of system process
calcium ion transport
di-, tri-valent inorganic cation transport
regulation of synaptic transmission
secretion
chemical homeostasis
regulation of transmission of nerve impulse
cellular ion homeostasis
regulation of neurological system process
cellular chemical homeostasis
secretion by cell
potassium ion transport
ion homeostasis
exocytosis
cellular homeostasis
neuron differentiation
monovalent inorganic cation transport
neurotransmitter transport
regulation of membrane potential
behavior
neuron development
neuron projection development
G-protein coupled receptor protein signaling pathway
doi:10.1038/nn.2779
nature neuroscience
Number of
listed genes
in term
Total number
of genes in
term
21
21
21
21
18
19
19
26
11
12
12
8
8
8
10
8
8
7
4
4
4
46
45
45
4
61
77
21
15
67
12
33
17
6
9
7
13
12
8
13
5
24
15
19
16
16
20
19
13
10
10
9
11
14
9
12
9
12
10
9
12
8
13
13
11
7
8
12
11
10
12
164
165
169
169
115
138
146
391
67
94
94
34
46
46
103
63
65
52
8
12
13
2204
2157
2177
17
1375
2367
220
143
2262
106
831
272
21
70
48
205
182
76
214
37
360
124
228
157
165
284
275
118
72
92
70
118
205
73
152
76
156
103
85
168
66
209
221
158
52
79
200
170
141
207
% of listed
Corrected
genes in
p-value*
term
12.8
12.7
12.4
12.4
15.7
13.8
13.0
6.6
16.4
12.8
12.8
23.5
17.4
17.4
9.7
12.7
12.3
13.5
50.0
33.3
30.8
2.1
2.1
2.1
23.5
4.4
3.3
9.5
10.5
3.0
11.3
4.0
6.3
28.6
12.9
14.6
6.3
6.6
10.5
6.1
13.5
6.7
12.1
8.3
10.2
9.7
7.0
6.9
11.0
13.9
10.9
12.9
9.3
6.8
12.3
7.9
11.8
7.7
9.7
10.6
7.1
12.1
6.2
5.9
7.0
13.5
10.1
6.0
6.5
7.1
5.8
6.64E-12
3.74E-12
3.97E-12
3.97E-12
7.19E-12
9.42E-12
2.13E-11
1.03E-09
5.31E-07
1.20E-06
1.20E-06
6.74E-06
5.34E-05
5.34E-05
1.94E-04
3.87E-04
4.42E-04
1.09E-03
2.00E-03
7.09E-03
8.62E-03
8.83E-03
1.06E-02
1.25E-02
1.61E-02
1.32E-11
2.20E-09
5.97E-08
7.11E-06
1.10E-05
6.04E-05
2.46E-04
4.13E-04
4.61E-04
4.62E-04
2.64E-03
3.37E-03
4.33E-03
4.16E-03
4.01E-03
3.85E-02
1.23E-06
1.12E-06
1.19E-06
1.50E-06
2.38E-06
3.25E-06
9.59E-06
1.34E-05
8.28E-05
5.95E-04
5.56E-04
5.54E-04
5.46E-04
6.00E-04
7.05E-04
7.09E-04
7.96E-04
8.34E-04
1.37E-03
1.35E-03
1.81E-03
1.96E-03
3.21E-03
3.47E-03
3.34E-03
4.63E-03
4.86E-03
5.44E-03
6.04E-03
5.91E-03
24
Polymenidou et al
GO:0042592
homeostatic process
GO:0051899
membrane depolarization
GO:0003001
generation of a signal involved in cell-cell signaling
GO:0006874
cellular calcium ion homeostasis
GO:0055074
calcium ion homeostasis
GO:0055066
di-, tri-valent inorganic cation homeostasis
GO:0050905
neuromuscular process
GO:0007626
locomotory behavior
GO:0048812
neuron projection morphogenesis
GO:0006875
cellular metal ion homeostasis
GO:0048667
cell morphogenesis involved in neuron differentiation
GO:0055065
metal ion homeostasis
GO:0007611
learning or memory
GO:0007612
learning
GO:0007628
adult walking behavior
*The p-value indicated is corrected for multiple testing using the Benjamini-Hochberg method.
doi:10.1038/nn.2779
nature neuroscience
15
5
6
6
6
7
5
8
8
6
8
6
6
5
4
329
26
48
52
54
82
33
113
114
57
115
59
60
38
19
4.6
19.2
12.5
11.5
11.1
8.5
15.2
7.1
7.0
10.5
7.0
10.2
10.0
13.2
21.1
8.13E-03
1.27E-02
1.56E-02
2.19E-02
2.53E-02
2.75E-02
2.76E-02
2.77E-02
2.84E-02
2.82E-02
2.84E-02
3.14E-02
3.30E-02
3.94E-02
4.33E-02
25
Polymenidou et al
Supplementary Table 5 TDP-43 binding and regulation on transcripts encoding for proteins involved in RNA metabolism
in
introns
in
exons
in 3'UTR
RPKM in
TDP-43 KD
RPKM in
control
309
1
308
0
0
52.07
53.16
0.98
-0.02
More exclusion
(chr16:7307308-7307361)
Adenosine deaminase, RNA-specific, B1 NM_001024839
14
0
14
0
0
21.21
27.16
0.78
-1.24
No
Senataxin
NM 198033
Ataxin-2
NM 009125
CUG triplet repeat, RNA-binding protein
NM_017368
1
ELAV
NM 010485
ELAV-like protein 2 (Hu-antigen B)
BC049125
ELAV-like protein 3
NM 010487
ELAV-like 4 isoform a
NM 010488
Elongation protein 3 homolog
AK088457
Ewing sarcoma homolog
AK147909
Fus/TLS protein
BC011078
FUS interacting protein
AK169071
Muscleblind-like 1
NM 020007
Muscleblind-like 2
AK164372
Neuro-oncological ventral antigen 1
NM 021361.1
Neuro-oncological ventral antigen 2
NM 001029877.
1
18
0
0
1
18
0
0
0
0
13.72
18.71
12.97
15.96
1.06
1.17
0.04
0.58
No
No
17
0
17
0
0
34.87
37.92
0.92
-0.35
No
2
17
7
8
3
8
4
3
19
14
15
7
0
0
0
0
0
0
0
0
0
0
0
0
2
17
7
8
3
6
0
1
19
14
13
6
0
0
0
0
0
0
3
0
0
0
0
0
0
0
0
0
0
2
1
2
0
0
2
1
17.20
9.71
35.32
14.14
15.61
44.86
92.88
16.64
39.54
68.83
7.62
15.04
16.79
11.13
35.20
14.16
17.54
48.66
129.56
17.47
34.50
66.40
16.97
36.85
1.02
0.87
1.00
1.00
0.89
0.92
0.72
0.95
1.15
1.04
0.45
0.41
-0.09
-0.91
0.05
-0.23
-0.81
-0.31
-1.46
-0.46
0.69
0.24
3
0
3
0
0
21.64
21.98
0.98
-0.17
Protein
A2bp1/Fox1
Ataxin-2 binding protein 1
Cugbp1
Elavl1
Elavl2
Elavl3
Elavl4
Elp3
Ewsr1
Fus
Fusip1
Mbnl1
Mbnl2
Nova1
Nova2
AY659958
Ratio
Z
KD/control score
Splicing changes
Change upon TDP-43
depletion (exon
coordinates)
in
5'UTR
Gene symbol
Adarb1
(ADAR2)
Als4
Atxn2
Refseq/mRNA
Total
identifier
RNA-seq
Ptbp2
Polypyrimidine tract binding protein 2
NM_019550
Raly
hnRNP-associated with lethal yellow
NM_023130
9
0
9
0
0
17.42
19.52
0.89
-0.73
Rbm9/Fox2
Sfrs9
Fox-1 homolog
Splicing factor, arginine/serine rich 9
TAF15 RNA polymerase II, TATA box
binding protein (TBP)-associated factor,
68 kDa
Cytotoxic granule-associated RNA
binding protein
AK044929
NM 025573
32
1
0
0
32
1
0
0
0
0
45.73
13.12
54.32
12.28
0.84
1.07
-0.72
0.08
No
No
No
No
No
No
No
No
No
No
No
No
More exclusion
(chr3:119716615-119716649)
More exclusion
(chr2:154551351-154551399)
No
No
AK011843
5
0
5
0
0
26.82
28.03
0.96
-0.22
No
NM_011585
6
0
6
0
0
26.46
21.48
1.23
0.92
More inclusion
(chr6:86384734-86384767)
Taf15
Tia1
Coordinates on mm8 mouse genome. *Human orthologue gene.
doi:10.1038/nn.2779
nature neuroscience
26
Polymenidou et al
Supplementary Table 6 TDP-43 binding and regulation on transcripts associated with disease
Gene symbol
Adarb1 (ADAR2)*
Als2
Als4
Ang1
App
Ar
Atxn1
Atxn10
Atxn2
Atxn7
Cacna1c
Chmp2b
Dctn1
Elp3
Fbxo7
Fgf14
Fig4
Fmr1
Fus/Tls
Fxn
Gria2
Gria3
Grik2
Grin1
Grin2a
Grin2c
Grn
Hdac6
Hdh
Ighmbp2
Kcnma1
Protein
Refseq/mRNA
identifier
Adenosine deaminase, RNANM_001024839
specific, B1
Alsin
AK020484
Senataxin
NM 198033
Angiogenin
NM 007447
Amyloid beta precursor protein
AK148439
NM_013476
Androgen receptor
Ataxin-1
NM 009124
Ataxin-10
NM 016843
Ataxin-2
NM 009125
Ataxin-7
NM 139227
Calcium channel, voltageNM_009781
dependent, L type
Chromatin modifying protein 2B
NM 026879
Dynactin 1
AK157867
Elongation protein 3 homolog
AK088457
F-box only protein 7
BC032153
Fibroblast growth factor 14
NM_207667
isoform b
Sac domain-containing inositol
NM_133999
phosphatase 3
Fragile X mental retardation
NM_008031
protein 1
Fus/Tls protein
BC011078
Frataxin
BC058533
Glutamate receptor, ionotropic,
NM_001039195
AMPA2 (alpha 2)
NM_016886
AMPA3 (alpha)
NM_010349
kainate 2 (beta)
BC039157
NMDA1 (zeta 1)
NM_008170
NMDA2A (epsilon)
NM_010350
NMDA2C (epsilon)
Progranulin
NM 008175
Histone deacetylase 6
NM 010413
Huntingtin
NM 010414
Immunoglobulin mu binding
NM_009212
protein 2
Calcium-activated potassium
channel alpha subunit 1
U09383
(Calcium-activated potassium
channel, subfamily M, alpha
subunit 1)
Total
in 5'UTR in introns in exons
RNA-seq
in 3'UTR
Ratio
RPKM in RPKM in
TDP-43 KD control KD/control
Splicing changes
Z score
Change upon TDP-43
depletion (exon coordinates)
14
0
14
0
0
21.21
27.16
0.78
-1.24
No
8
1
0
34
0
64
9
18
4
0
0
0
0
0
0
0
0
0
8
1
0
29
0
64
9
18
3
0
0
0
3
0
0
0
0
0
0
0
0
2
0
0
0
0
1
6.67
13.72
21.94
219.61
2.90
16.02
43.25
18.71
6.55
6.94
12.97
9.78
263.72
2.73
19.97
40.78
15.96
6.70
0.96
1.06
2.24
0.83
1.06
0.80
1.06
1.17
0.98
-0.55
0.04
3.79
-0.77
-0.33
-1.31
0.30
0.58
-0.45
No
No
No
No
No
No
No
No
No
77
0
77
0
0
5.57
9.56
0.58
-2.65
No
1
8
3
5
0
0
0
0
1
8
3
5
0
0
0
0
0
0
0
0
16.51
40.37
15.61
9.99
12.82
50.11
17.54
7.35
1.29
0.81
0.89
1.36
0.96
-0.93
-0.81
0.95
No
No
No
No
85
0
85
0
0
6.99
14.99
0.47
-3.76
No
3
0
3
0
0
10.36
8.58
1.21
0.50
No
0
0
0
0
0
17.03
17.38
0.98
-0.31
No
4
0
0
0
0
0
3
0
1
0
92.88
1.57
129.56
1.10
0.72
1.43
-1.46
0.41
No
No
18
1
14
0
3
105.96
110.65
0.96
-0.12
No
More exclusion
(chrX:37898880-37914127)
18
0
17
0
1
33.82
45.41
0.74
-1.35
52
0
52
0
0
17.00
27.85
0.61
-2.46
No
3
0
3
0
0
62.95
83.99
0.75
-1.25
More exclusion
(chr2:25133351-25133414)
83
0
83
0
0
2.19
4.23
0.52
-2.56
No
3
0
3
0
0
7.38
11.76
0.63
-2.45
No
1
3
10
0
0
0
0
3
9
0
0
1
1
0
0
70.17
8.98
8.52
23.12
10.54
16.55
3.04
0.85
0.51
5.22
-1.05
-3.29
no
No
No
1
0
1
0
0
3.26
2.98
1.09
-0.19
No
96
0
96
0
0
8.34
14.36
0.58
-2.66
No
Kif5a
Kinesin family member 5A
NM_001039000
5
0
2
0
3
261.61
364.81
0.72
-1.45
No
Kifap3
Kinesin-associated protein 3
NM 010629
5
0
5
0
0
43.38
50.66
0.86
-0.65
No
doi:10.1038/nn.2779
nature neuroscience
27
Polymenidou et al
Lrrk2
Mapt
Mecp2
Mjd/Atxn3
Nefl
Nlgn1
Nrxn1
Leucine-rich repeat kinase 2
NM 025730
Tau protein
NM 001038609
Methyl CpG binding protein 2
AK158664
Mjd/Ataxin-3 protein
BC087880
Neurofilament, light polypeptide
NM 010910
Neuroligin 1
NM 138666
Neurexin-I-alpha precursor
AB093249
(Neurexin I-alpha)
Nrxn2
Hypothetical Laminin G
Nrxn3
Neurexin III
Optn
Optineurin
Park2
Parkin
Park7
DJ-1 protein
Pink1
PTEN induced putative kinase 1
Pla2g6
Phospholipase A2, group VI
Prnp
Prion protein
Psen1
Presenilin 1
Psen2
Presenilin 2
Slc1a2/Glt1 (EAAT2)*
Slc1a3 (EAAT1)*
Slc1a6 (EAAT4)*
Smn1
Survival motor neuron protein
Snca
Synuclein, alpha
Sncb
Synuclein, beta
Sod1
Superoxyde dismutase 1
Sort1
Sortilin 1
Spg11
Tbp
Tnrc15/Gigyf2
4
10
4
3
1
102
0
0
0
0
0
0
4
10
4
3
0
101
0
0
0
0
0
1
0
0
0
0
1
0
11.28
80.18
29.52
9.65
40.04
8.55
14.49
97.03
30.32
8.49
61.29
16.28
0.78
0.83
0.97
1.14
0.65
0.53
-1.41
-0.80
-0.15
0.21
-1.88
-3.19
No
No
No
No
No
No
162
0
162
0
0
18.66
33.28
0.56
-2.76
No
AK163904
28
0
27
0
1
41.94
45.65
0.92
-0.34
BC060719
178
0
178
0
0
16.25
37.23
0.44
-3.93
NM 181848
AF250294
NM 020569
NM 026880
NM 016915
NM 011170
NM 008943
AF038935
NM 011393
NM 148938
NM 009200
BC045158
AK014472
NM 033610
BC057074
1
39
0
0
0
4
4
1
28
2
3
0
3
3
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
1
39
0
0
0
2
3
1
26
2
3
0
2
1
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
0
0
0
1
1
0
6.07
0.63
15.89
81.90
2.70
179.48
18.45
4.09
139.06
134.38
1.27
6.40
37.26
54.68
49.95
5.06
2.18
14.92
82.18
3.00
153.18
16.55
3.29
156.13
97.73
1.32
5.76
42.97
68.33
45.48
1.20
0.29
1.06
1.00
0.90
1.17
1.11
1.24
0.89
1.37
0.96
1.11
0.87
0.80
1.10
0.29
-3.78
0.07
0.06
-0.81
0.81
0.34
0.23
-0.46
1.55
-0.63
0.01
-0.63
-0.95
0.50
NM_019972
8
0
8
0
0
75.67
86.65
0.87
-0.55
4.75
7.33
14.87
4.26
6.06
13.80
1.12
1.21
1.08
-0.04
0.35
0.12
More exclusion
(chr19:6513715-6513805)
More inclusion
(chr12:90604171-90604261)
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
More inclusion
(chr3:108483527-108483626)
No
No
No
23.19
19.27
1.20
0.80
No
93.43
90.90
1.03
0.20
No
23.29
26.90
0.87
-0.73
No
11.57
9.91
1.17
0.43
No
Spatacsin
BC019404
4
0
4
0
0
TATA binding protein
BC016476
0
0
0
0
0
Tnrc15 protein
BC027137
15
1
14
0
1
Ubiquitin protein ligase E3A
Ube3a
NM_173010
3
0
3
0
0
isoform 1
Ubiquitin carboxy-terminal
Uchl1
NM_011670
0
0
0
0
0
hydrolase L1
Vesicle-associated membrane
Vapb
NM_019806
7
2
5
0
2
protein, associated
Valosin containing protein
NM_009503
1
0
1
0
0
Vcp
Z-scores between -1.96 and 1.96 represent p-values greater than 0.05. Coordinates on mm8 mouse genome. *Human orthologue gene.
doi:10.1038/nn.2779
nature neuroscience
28

Long pre-mRNA depletion and RNA missplicing contribute to

Transcription

Similar documents