Supporting Information
Hsiao et al. 10.1073/pnas.1202556109
SI Methods
Mice.
C57BL/6N mice (Charles River) were housed under speci
fi
c
pathogen-free conditions in the California Institute of Technol-
ogy
’
s Broad animal facility. Mice were mated overnight and the
presence of a vaginal plug on the following morning was noted as
embryonic day (E) 0.5. All experiments were performed under the
approval of the California Institute of Technology Institutional
Animal Care and Use Committee.
Spleen and Mesenteric Lymph Node Suspensions.
Spleens and mes-
enteric lymph nodes (MLNs) were harvested, washed in complete
RPMI media on ice, and forced through 40-
μ
m cell strainers (BD
Biosciences). In some trials, spleens from the same treatment
group were pooled in groups of three to facilitate accommodating
a large number of samples, while preserving cell viability. Single-
cell suspensions were subjected to RBC lysis (Sigma Aldrich).
CD4
+
T cells were isolated by negative selection using a CD4
+
Isolation kit (Miltenyi Biotec). Cell counts were performed on
a hemocytometer, and 10
6
cells were aliquoted for each CD4
+
stimulation reaction.
Flow Cytometry.
For extracellular staining, cells were treated with
anti-mouse CD16/CD32 Fc block (eBioscience) before staining
with subsets of these antibody conjugates: Ter119-PerCP-Cy5.5,
CD4-FITC, CD8-FITC, Gr-1-APC, B220-FITC, CD11b-APC
(eBioscience), NK1.1-PE (Biolegend).
Regulatory T-Cell Suppression Assay.
For regulatory T-cell (Treg)
suspression assay, 10
5
CD4
+
CD25
−
T-responder cells were cul-
tured in complete RPMI with 2
×
10
4
irradiated antigen-pre-
senting cells and 2.5
μ
g/mL anti-CD3, with CD4
+
CD25
+
Tregs
at a ratio to T-responder cells of 1:1, 1:2, and 1:4, according to
methods from (1). After 2 d of culture, suspensions were stained
with CD4-APC antibody and assessed by
fl
ow cytometry.
Fetal Liver Suspensions.
Fetal livers were harvested from E13.5
and E15.5 embryos and washed in PBS. Six to seven fetal livers
from a single litter were pooled, forced through a 40-
μ
mcell
strainer, and washed with Iscove
’
smodi
fi
ed Dulbecco
’
sme-
dium (IMDM) + 2% (vol/vol) FBS
(StemCell Technologies).
Cells were counted using a hemocytometer, and 2
×
10
5
cells
used per methylcellulose solution.
Donor Cell Harvest for Bone Marrow Transplantation.
Four adult
saline offspring (two male, two female) and four adult poly(I:C)
offspring (two male, two female) were selected from an in-
dependent cohort of behaviorally tested maternal immune acti-
vation (MIA) and control mice, and designated as bone marrow
(BM) donors. Donor adult saline and poly(I:C) mice were killed by
CO
2
gas, and BM was harvested by
fl
ushing femurs and tibias with
DMEM (Thermo Scienti
fi
c). BM suspensions from each treat-
ment group were pooled, subjected to RBC lysis (Sigma Aldrich),
fi
ltered through a 40-
μ
m cell strainer, and stored on ice.
Behavioral Testing.
For prepulse inhibition (PPI), mice were ac-
climated to an SR-LAB testing chamber (SD Instruments) for
5 min, presented with six, 120-dB pulses of white noise (startle
stimulus), and then subjected to 14 randomized blocks of either
no startle, startle stimulus only, 5-dB prepulse + startle, or 15-dB
prepulse + startle. The startle response was recorded by a piezo-
electric sensor, and PPI de
fi
ned as: (startle stimulus only
–
5or
15 dB prepulse + startle)/startle stimulus only
×
100.
For open-
fi
eld testing, mice were allowed to explore a 50
×
50-cm white Plexiglas box for 10 min. An overhead video camera
was used to record the session, and Ethovision software (Noldus
Information Technology) used to analyze the distance traveled,
and the number of entries and duration of time spent in the
center arena (central square, 17
×
17 cm).
For marble-burying testing, mice were acclimated for 10 min to
a new testing cage containing Aspen pine bedding (NEPCO),
3-cm deep. The mice were brie
fl
y placed in a fresh holding cage
while 18 navy blue marbles (15-mm diameter, washed with 70%
ethanol and mixed in Aspen pine bedding) were laid in a 6
×
3
pattern in the testing cage. Mice were placed back to the testing
cage, and the number of buried marbles (
>
50% covered with
bedding) was scored after a 10-min trial.
For social interaction testing, mice were
fi
rst habituated for
10 min to a 60
×
40-cm Plexiglas box with three equally-sized
chambers, with each of the two side chambers containing an
empty Plexiglas cylinder. During a second 10-min trial, the
mouse was placed in the center chamber and given the choice to
interact with an unfamiliar, same-sex mouse that was placed in
one of the side chamber cylinders, or with a nonsocial object
(novel green, sticky ball toy), placed in the other side chamber
cylinder. In the third and
fi
nal 10-min trial, social preference was
measured by replacing the nonsocial object with a new, un-
familiar, same-sex mouse. In this trial, the mouse was given the
choice to interact with the now familiar mouse (from trial 2) or
the new unfamiliar mouse. The cylinders holding the stimulus
mice or the toy have small holes (
∼
10 mm diameter) over the
entire surface area to permit direct interaction. Mice to be tested
were prehabituated to the Plexiglas cylinders and three-cham-
bered box for 20 min each day for 3 d before testing. All
equipment was cleaned with 70% ethanol and Process NPD
(STERIS Life Sciences) before and after testing. The side
chamber used for placement of the social object or the nonsocial
object, and the familiar mouse or the unfamiliar mouse was al-
ternated in each trial to prevent bias. The duration of time spent
and number of entries into each chamber were recorded using an
overhead video camera and analyzed with Ethovision software.
Equivalent levels of chamber duration and chamber entries
during the habituation phase were used to ensure that the mice
did not exhibit chamber bias.
“
Social preference, chamber du-
ration
”
is measured by: (duration in the novel mouse chamber
–
duration in the familiar mouse chamber).
“
Social preference,
chamber frequency
”
is measured by: number of entries into the
novel mouse chamber
–
number of entries into the familiar
mouse chamber. For post-BM transplant behavior testing, dif-
ferent stimulus (familiar and unfamiliar) mice were used to
prevent potential confounding effects of social memory.
1. Collison LW, Vignali DA (2011) In vitro Treg suppression assays.
Methods in molecular
biology
707:21
–
37.
Hsiao et al.
www.pnas.org/cgi/content/short/1202556109
1of6
Fig. S1.
Some of the immune changes observed in spleens of MIA offspring are recapitulated in the MLN. (
A
) Consistent with what is observed in the spleen,
MLNs from poly(I:C) offspring exhibit decreased levels of total CD4
+
Foxp3
+
, CD4
+
Foxp3
+
CD25
−
, and CD4
+
Foxp3
+
CD25
+
T cells, but no signi
fi
cant difference in
levels of CD4
+
TCR
β
+
IL-17
+
IFN-
γ
+
Th17
+
Th1 cells or CD4
+
TCR
β
+
IL-17+ IFN-
γ
−
Th17 cells (
n
= 5, where each sample represents a pool of three spleens). (
B
) There
are no signi
fi
cant differences between MLNs from adult poly(I:C) versus saline offspring in the distribution of major leukocyte classes [
n
= 8 saline, 9 poly(I:C)].
(
C
) Consistent with what is observed in the spleen, CD4
+
T cells from the MLNs from 15-wk-old poly(I:C) offspring produce increased levels of IL-6 and IL-17 at 1,
2, and 3 d after phorbol 12-myristate 13-acetate (PMA)/ionomycin stimulation (
n
= 4, where each represents a pool MLNs from three animals). (
D
) This hy-
perresponsive phenotype of MLN CD4
+
T cells is maintained in 1-y-old poly(I:C) versus saline mice (
n
=11
–
16). *
P
<
0.05, **
P
<
0.01, ***
P
<
0.001. All panels
represent one representative experiment of at least two separate trials.
Fig. S2.
There is no signi
fi
cant difference in the suppression of CD4
+
CD25
−
T-cell (effector T cells, Teff) proliferation by splenic Tregs from adult saline versus
poly(I:C) offspring. Similar levels of Teff proliferation are observed after 48 h of coculture of Tregs with Teffs at ratios of 1:1, 1:2, and 1:4 (
n
=4).
Hsiao et al.
www.pnas.org/cgi/content/short/1202556109
2of6
Fig. S3.
Hyperresponsiveness of CD4
+
T cells from MIA offspring is established early in development and maintained through adulthood. CD4
+
T cells from the
spleens of 3-wk-old MIA offspring (
A
) and 1-y-old MIA offspring (
B
) secrete elevated levels of IL-6 (
Left
) and IL-17 (
Right
) in response to PMA/ionomycin
stimulation (
n
=5
–
7). *
P
<
0.05, **
P
<
0.01, ***
P
<
0.001. All panels represent one representative experiment of at least two separate trials.
Fig. S4.
Increases in both neutrophilic and monocytic Gr-1
+
CD11b
+
cells constitute the MIA-associated increase in total Gr-1
+
cells. (
A
) Three populations of
Gr-1
+
cells are resolved based on Gr-1 intensity and granularity (side scatter, SSC). (
B
) Further strati
fi
cation by CD11b and Ly6C intensity demonstrates that all
three populations are CD11b
+
and Ly6C+ (high or mid). (
C
) Assessment for Ly6G intensity reveals monocytic Gr-1
+
cells [Ly6G
−
, population (iii)] and neutrophilic
Gr-1
+
cells [Ly6G high or mid, populations (i) and (ii)]. (
D
) Adult poly(I:C) offspring exhibit mild increases in all three populations of Gr-1
+
cells identi
fi
ed: (i) Gr-
1
hi
CD11b
+
Ly6C
mid
Ly6G
hi
SSC
mid
, (ii) Gr-1
mid
CD11b
+
Ly6C
mid
Ly6G
mid
SSC
mid
, and (iii) Gr-1
mid
CD11b
+
Ly6C
hi
Ly6G
−
SSC
lo
, which result in a statistically signi
fi
cant
difference based on treatment group (
n
=5).**
P
<
0.01.
Hsiao et al.
www.pnas.org/cgi/content/short/1202556109
3of6
Fig. S5.
There is no signi
fi
cant difference between unmanipulated saline mice and BM-transplanted mice in the lineage potential of BM HSCs and progenitors
differentiated in vitro. Transplant of poly(I:C) or saline BM into irradiated poly(I:C) offspring corrects the abnormal lineage differentiation p
henotype asso-
ciated with unmanipulated poly(I:C) offspring. Similarly, transplant of poly(I:C) BM into irradiated saline offspring does not transfer the prefe
rential skewing
toward CFU-G (granulocyte) colonies that is observed in unmanipulated poly(I:C) offspring (
n
= 3). *
P
<
0.05.
Fig. S6.
BM transplant recipients are validated to be behaviorally abnormal in PPI, social preference, repetitive marble burying, and open-
fi
eld exploration.
(
A
) MIA offspring display signi
fi
cantly decreased PPI when prepulses are administered at 5 dB and 15 dB above background (PPI5 and PPI15, respectively). (
B
)
Offspring of poly(I:C)-injected mothers exhibit decreased social preference for a novel mouse over a familiar mouse, as measured by differences in c
hamber
duration (
Left
) and chamber frequency (
Right
). (
C
) MIA offspring bury signi
fi
cantly greater numbers of marbles compared with saline controls, pre-BM
transplant. (
D
) Poly(I:C) offspring exhibit increased anxiety in the open
fi
eld compared with saline controls, as measured by decreased entries into and duration
in the center of the arena, and no signi
fi
cant difference in overall locomotion as denoted by total distance traveled (
n
=20
–
21). *
P
<
0.05, **
P
<
0.01. All panels
represent one representative experiment of at least two separate trials. BM transplant data were acquired from one large experiment.
Hsiao et al.
www.pnas.org/cgi/content/short/1202556109
4of6
Fig. S7.
MIA does not affect levels of activated or total CD11b
+
microglia/macrophages in the brains of adult offspring. (
A
) There is no signi
fi
cant difference in
levels of total CD11b
+
cells or MHCI
+
cells in the cerebellum, cortex, or hippocampus of adult poly(I:C) versus saline mice. (
B
) Representative
fl
ow cytometry
plots of the separation of brain CD11b
+
and MHCI
+
cells. (
C
) There is no signi
fi
cant difference in the percentage of CD11b
+
MHCII
+
CD68
+
cells from the
cerebellum, cortex, or hippocampus of adult poly(I:C) versus saline mice. (
D
) Representative
fl
ow cytometry plots of the separation of brain MHCII
+
and CD68
+
cells. (
n
= 4 for each brain region).
Fig. S8.
MIA offspring retested for anxiety in the open
fi
eld at 9 mo of age retain behavioral de
fi
cits of similar intensity to those seen at 2 mo of age. This
result applies to the number of entries into the center of the open
fi
eld (
Left
) and duration of time spent in the center of the open
fi
eld (
Center
). No signi
fi
cant
difference is observed between saline and poly(I:C) offspring in total distance traveled in the open
fi
eld arenas (
Right
), as is typically observed in the MIA model
(
n
=17
–
19). *
P
<
0.05
Hsiao et al.
www.pnas.org/cgi/content/short/1202556109
5of6
Fig. S9.
The abnormal behaviors of MIA offspring are not maintained after irradiation and BM transplant. (
A
) Saline and poly(I:C) offspring irradiated and
transplanted with poly(I:C) BM exhibit no difference in social preference for a novel mouse over a familiar mouse, as indicated by chamber duration (
Left
)and
chamber frequency (
Right
). (
B
) MIA and control offspring irradiated and transplanted with poly(I:C) BM exhibit levels of repetitive marble burying that are
comparable to those observed in transplanted control and untransplanted saline offspring. (
C
) There is no signi
fi
cant difference in level of anxiety in the open
fi
eld between saline and poly(I:C) offspring irradiated and transplanted with saline or poly(I:C) BM, as measured by duration spent in the center of the
open arena (
Left
) and number of entries into the center arena (
Center
). However, irradiated and BM-transplanted mice exhibit increased overall activity in the
open
fi
eld compared with untreated mice, as seen in elevated total distance traveled (
Right
). [
n
=11
–
16 animals for saline and poly(I:C) groups, 4
–
5 animals for
BM-transplant groups]. *
P
<
0.05, **
P
<
0.01, n.s., not signi
fi
cant. BM-transplant data were acquired from one large experiment.
Hsiao et al.
www.pnas.org/cgi/content/short/1202556109
6of6