Supplementary Figure S1. Construction of the AAV library and CRE
-
dependent PCR.
a
,
Randomized 21
-
basepair DNA fragment was inserted into the AAV9 capsid between amino acids
588 and 589, which resides at the exterior of an AAV capsid (inset). AAV capsid was produced
within the AAV genome allowing for recovery of the capsid sequenced fro
m transduced cells. The
capsid coding sequence was followed by a polyA (pA) sequence flanked by a double
-
inverted
floxed open reading frame (DIO).
b
, The DIO sequence can be recombined and inverted in the
presence of Cre enzyme. That sequence inversion can
then be detected using PCR. Therefore,
the DNA from AAVs that transduced cells expressing Cre can be amplified using a PCR reaction.
In our study, we used hSyn1
-
Cre mice which express Cre selectively in neurons, and thus, we
selected for neuron
-
transducin
g AAVs.
Supplementary Figure S2. AAV9 GFP + AAV9 mCherry.
Transduced cell counts
in the brain
comparing
AAV9 carrying GFP and mCherry are highly correlated and (Rsq=0.99).
The AAVs
were administered at 1E10 VGs per gram of body weight to each animal.
The mean numbers of
transduced cells are not significantly different (fold difference between AAV9
-
GFP and AAV9
-
mCherry: 1.07
-
fold, p=0.081 (ns), paired t
-
test, 6 sections tested from 2 mice).
Supplementary Figure S3. Pairwise comparison of AAV.FUS candidates’ transduction of
the brain.
Non
-
significant comparisons not shown for clarity
.
(**** = p<0.0001; *** = p<0.001; **
= p <0.01; * = p<0.05; F(4, 24) = 14.96, P<0.0001, One
-
way ANOVA with Tukey HSD post hoc
test.). Non
-
significant pairwise comparisons not shown for clarity. Error bars are 95% CI. All 10
detailed p
-
values provided in th
e source data appendix.
n=6 mice used for all serotypes except
AAV.FUS.2 which used n=5 mice.
IV injection dos
e, 10
10
vg/g
Supplementary Figure S4.
Pairwise comparison of AAV.FUS candidates’ transduction of
the liver. Non
-
significant comparisons not shown for clarity.
AAV.FUS.3 shows significantly
reduced liver transduction compared to other AAV.FUS candidates. One
-
way ANOVA with Tukey
HSD post
-
hoc test. F (4, 24) = 96.69. P<0.0001; All pairwise comparisons are below p<0.0001,
except AAV.FUS.2 vs AAV.FUS.4 (p=0.0001), A
AV.FUS.2 vs AAV.FUS.5 (p=0.3524),
AAV.FUS.3 vs AAV.FUS.4 (p=0.01), and AAV.FUS.4 vs AAV.FUS.5 (p=0.0099)). (**
** =
p<0.0001; *** = p<0.001; ** = p <0.01; * = p<0.05, ns = non
-
significant). Error bars are 95% CI.
Detailed p
-
values provided in the source data appendix.
n=6 mice used for all serotypes except
AAV.FUS.2 which used n=5 mice.
IV injection dose, 10
10
vg/g
Supplementary Figure S5. Transduction of AAV9 and AAV.FUS.3 in peripheral tissues.
We
observed no substantial transduction observed in the
a)
kidneys and
b)
lungs, consistent with
previous reports
21
.
Representative images were obtained from mice co
-
injected with AAV9 (red,
mCherry) and AAV.FUS.3 (green, GFP).
IV injection dose, 10
10
vg/g
.
Sections were imaged on a
confocal microscope with 10x objective counterstained with a nuclear stain (DAPI, blue). Example
positive cells designated with arrows. Scale bars are 100 microns.
Supplementary Figure S6. Representative images of transduction in brain and liver for all
AAV.FUS (green, EGFP) and corresponding co
-
injected AAV9 control (red, mCherry).
IV
injection dose, 10
10
vg/g
.
Scale bars, 200 microns for the brain, 100 microns for the liver.
Supplementary Figure S7. Transduction of non
-
neuronal brain cells by AAV.FUS.3 and
AAV9.
We observed lower transduction of
microglia/macrophages, astrocytes and
oligodendrocytes
in the brain by either AAV9 or AAV.FUS.3 as compared to neurons.
Representative images were obtained from mice co
-
injected with AAV9 (red, mCherry) and
AAV.FUS.3 (green, GFP
). a
) The transduction of astrocytes (GFAP
+
) was comparable between
the AAV9 and AAV.FUS.3, with 8% and 3.4% average transduction, respectively (
n=3 mice
,
p=0.0552, two
-
tailed paired t
-
test; t=4.076).
b
) Similarly,
microglial/macrophagic
transduction was
also less efficient than neuronal transduction for both serotypes with AAV9 transducing 3.5% and
AAV.FUS.3 transducing 0.7% of microglial cell, which was a statistically significant difference (
n=3
mice
,
p=0.0174, two
-
tailed paired t
-
test, t=7.487).
c
) AAV.FUS.3 transduction of oligodendrocytes
was significantly reduced
relative to AAV9, with average transduction being 3.4% for AAV.FUS.3
and 74.3% for AAV9
(
n
=
6
mice
,
p
<0.0001
, two
-
tailed paired t
-
test, t=
12.32
).
Sections were
imaged on a confocal microscope with 20x objective
counterstained for glial cells (GFAP, blue),
microglia/macrophag
es
(Iba1, blue) and oligodendrocytes (Olig2, blue).
IV injection dose, 10
10
vg/g
.
Scale bars are 50 microns.