1

Supplemental Information Supplemental Experimental Procedures and Supplemental Figure Legends

Immunohistochemistry of FGF2

FGF2 immunostaining was performed on free-floating sections under contract by Neuroscience

Associates (Knoxville, TN). Briefly, all incubation solutions from the blocking serum onward

used Tris buffered saline (TBS) with Triton X-100 (TX); all rinses were with TBS. Endogenous

peroxidase activity was blocked by hydrogen peroxide treatment and non-specific binding was

blocked with normal goat serum. Following rinses, the sections were immunostained with rabbit

anti-FGF2 antibody (1:1,500 dilution; LifeSpan Biosciences, Seattle, WA) overnight at room

temperature. The sections were then rinsed and incubated in a biotinylated goat anti-rabbit

antibody. The Vectastain elite ABC kit (Vector Labs, Burlingame, CA) was used to visualize the

reaction. The sections were treated with diaminobenzidine tetrahydrochloride (DAB) and

hydrogen peroxide to create a visible reaction product. The sections were mounted on

gelatinized (subbed) glass slides, air-dried, dehydrated in alcohols, cleared in xylene, and

coverslipped with Permount (Fisher, Pittsburgh, PA). One additional section was stained as a

pre-absorption control. Here, FGF2 peptide was incubated at 10-fold excess with FGF2 antibody

for 1 h at RT. The mixture was incubated with the section overnight as described above.

Immunostaining and quantification of Sox2/GFAP and Ki-67 immunoreactive cells

Sox2 is a marker for neuroprogenitors and Ki-67 is a marker for proliferating cells. Free-floating

sections were double stained for Sox2 and GFAP to quantify the number of neuroprogenitors or

were immunostained for Ki-67 to quantify proliferating cells in the SGZ of WT and NF-1-KO

2

mice. For Sox2 and GFAP immunostaining, polyclonal rabbit anti-Sox2 antibody (1:1,000;

Millipore, Billerica, MA) and monoclonal mouse anti-GFAP antibody (1:3,000; Cell Signaling

Technology, Danvers, MA) were used. For Ki-67 staining, an antigen retrieval step (incubation

in sodium citrate for 45 min at 90°C) was performed, followed by blocking as described above,

and stained with a monoclonal rat anti-Ki67 (1:1,000; Dako, Carpinteria, CA). Images of

Sox2/GFAP (double-label), and Ki-67 immunopositive cells in the SGZ of the dentate gyrus

were taken in three sections from the hippocampus of each mouse. Four mice were analyzed for

each group. The staining of Sox/GFAP or Ki-67 positive cells were observed under confocal

microscope and images were taken. For counting of immunoreactive cells in the dentate gyrus,

images were taken with a 20x objective using a fluorescent microscope (Nikon eclipse 80i) with

a digital camera.

Locomotor activity in open field.

The assessment of activity was conducted as outlined in Fukui et al. (ref 33), WT and NF-1-KO

mice were injected with vehicle or 5 ng/kg FGF2 (s.c.) and, after 30 min, were placed into the

open field for 10 min.

Supplementary Figure Legends

Figure S1. Forced swim test of mice after short- or long-term chronic restraint stress.

(A) Mice were restrained for 1 h/day for 7 days (stCRS) or 6 h/day for 21 days (ltCRS) and then

were subjected to the forced swim test ~18h after the last restraint. Bar graphs show immobility

time. Note that mice after short-term restraint stress did not show increased immobility time

compared to the non-restrained control, whereas mice after long-term restraint stress exhibited

significantly increased immobility time, compared to non-restrained animals, indicative of

3

depressive-like behavior. Interestingly, immobility was lower in long-term controls than in short-

term control mice. Values shown are the mean ±SEM (n = 10 mice/time/restraint-condition;

*p<0.05, non-restraint vs restraint in ltCRS; +p<0.05, control stCRS vs control ltCRS; #p<0.05,

restraint stCRS vs restraint ItCRS. (B) HET mice were subjected to the forced swim test. Bar

graphs show immobility time. Values shown are the mean ±SEM (n = 10 wild-type, WT, n = 14

heterozygotes, HET) ; *p<0.05. Note that HETs have increased immobility time compared to

WT mice, indicative of depressive-like behavior.

Figure S2. Expression of NF-1 and FGF2 in the cortex of mice after 7-days of chronic

restraint stress.

(A) Western blot analysis and (B) quantification of NF-1 and the different isoforms of FGF2

protein (normalized to β-actin) in the cortex from naïve mice and mice after stCRS (1 h/day for 7

days). No significant difference in the levels of these proteins was observed between these two

restraint conditions, n = 6/group; values are the mean ±SEM. Twenty µg of total protein were

loaded per lane and the blots were probed for NF-1, FGF2 isoforms, and actin with the

respective antibodies.

Figure S3. FGF2 immunoreactivity in the hippocampal fasciolarum cinereum in NF-1-

KO and WT mice.

Photomicrograph of FGF2 immunoreactivity showing diminished expression of FGF2 in the

fasciolarum cinereum (FC) of the hippocampus of a 6 wk old NF-1-KO mouse (C, D)

compared to a WT control (A, B). The boxes in the left panels (A and C) indicate the enlarged

views presented to the right in panels B and D. CA1 = CA1 hippocampus, DG = dentate gyrus.

4

Figure S4. FGF2 expression is decreased in hippocampus of 30 wk old NF-1-KO mice.

(A) Photomicrograph of FGF2 immunoreactivity in the hippocampus of WT and NF-1-KO

mice. FGF2 was virtually absent from the hippocampus of the NF-1-KO compared to WT mice

which showed substantial FGF2 expression. Box indicates enlarged areas presented on the right

side of panel A. (B) Quantification of the total area of FGF2 immunoreactivity on 8 sections

through the whole hippocampal formation in one hemisphere in 30-wk old WT and NF-1-KO

mice. n = 4/group; values show mean ± SEM.

Figure S5. Analysis and quantification of neuroprogenitor cells in the dentate gyrus of

vehicle versus FGF2 treated WT and NF-1-KO mice.

(A) Photomicrographs showing that Sox2 (red)/GFAP (green) double immmunostained

neuroprogenitors in the subgranular zone (SGZ) of the dentate gyrus of WT and NF-1-KO mice.

Bar = 50µm. (B) Quantification of Sox2/GFAP double immunostained neuroprogenitor cells in

the SGZ of the dentate gyrus. No difference was observed in the number of double positive cells

between of WT and NF-1-KO mice. Values are the mean ±SEM, n = 4/group.

Figure S6. Analysis of hippocampal DCX mRNA in vehicle- versus FGF2-treated NF-1-

KO mice.

Quantification shows increased expression of DCX mRNA in the hippocampus of NF-1-KO

mice treated with FGF2. Values are the mean ±SEM, n = 4/group; *p<0.05, FGF2 treated

compared to untreated in NF-1-KO mice.

5

Figure S7. Ki-67 positive neuron numbers in the SGZ of NF-1-KO mice increased after

FGF2 administration.

(A) Photomicrographs of Ki-67 immmunostaining in the hippocampus of WT and NF-1-KO

mice injected with vehicle (control) or FGF2. Cells were counted in the SGZ. Bar = 50 µm. (B)

Quantification of Ki-67 positive cells in the SGZ of WT and NF-1-KO mice. FGF2 induced a

significant increase in the numbers of Ki-67 positive cells in FGF2 treated NF-1-KO mice

compared to untreated NF-1-KO mice, indicating proliferation. Values are the mean ±SEM, n =

4/group; *p<0.05.

Figure S8. Analysis of Sp1 in hippocampal neurons treated with NF-1.

(A) Merged immunocytochemical images (pink) taken from confocal microscope (20X) indicate

that Sp1 (red) is translocated into the nucleus (DAPI/blue) of neurons incubated with 400 nM

recombinant NF-1/CPE for 3 h. (B) Quantification of the Sp1 signal intensity in the nuclei

shows a significant increase in NF-1-treated versus untreated neurons. n = 6/group; values are

mean ±SEM; *p<0.05, compared to Ctrl. Bar = 50 µm.

Figure S9. Bioinformatic analysis of PPAR binding sites in the NF-1/cpe promoter.

Bioinformatic analysis of the NF-1/Cpe promoter, using the Genomatix software suite

(Genomatix GmbH, Munich, Germany) revealed the presence of 3 putative PPAR binding sites

(black elipses) in rat and 2 in the mouse and human. Black arrows indicate transcription start

sites.

6

Figure S10. PCR array analysis of genes after treatment of hippocampal neurons with NF-

1.

PCR array (purchased from QIAGEN, Valencia, CA) shows that expression of FGF4, a member

of the FGF family was up-regulated (4-fold) after NF-1 treatment in primary cultured

hippocampal neurons. Subsequent qPCR studies confirmed that NF-1 up-regulated FGF4

expression in hippocampal neurons; however, this increase was only 2-fold. This prompted us to

assay for the other FGF family members by qPCR and we found that FGF2 mRNA was up-

regulated in a much more robust manner by NF-1 as shown in main text.

Figure S11. Analysis of locomotor activity of WT and NF-1-KO mice following FGF2

treatment for 30 consecutive days.

Motor activity was assessed in the open field. WT and NF-1-KO mice were injected for 30

consecutive days with vehicle or 5 ng/g FGF2 (s.c.) and on the final day, 30 min later they were

placed into the open field for 10 min. FGF2 treatment reduced significantly locomotor activity

in both genotypes. Values shown are the mean ±SEM (n = 8-9 mice/genotype/treatment-

condition),

Table S1. Analysis of FGF2 immunoreactivity through the hippocampus of WT and NF-

1-KO mice with age.

Semi-quantitative analysis of areas of FGF2 immunoreactivity (IR) on 8 coronal sections through

the hippocampal formation in one hemisphere of WT and NF-1-KO mice at age 6, 14, and 30

weeks showed similar decreases in FGF2 levels in 6 and 14 week old NF-1-KO versus WT

mice. At 30 weeks there was a pronounced reduction in expression. Strength of the IR was

7

assessed on scale of 0 (-) to 3 (+++). F, female; M, male. Empty space means that the

hippocampus was not present on the section.

Cheng et al. Figure S1

Immob

ility (sec)

HETWT0

20

40

60

80

100

120

140

160

180*

A. B.

stCRS Naïve

0

20

40

60

80

100

120

140

160

FGF2

24 KDa

21 KDa

17 KDa

24 KDa 21 KDa 17 KDa

FGF2

% expression/Actin

 

β‐Actin

NaivestCRS

Cheng et al. Figure S2

NF‐α1

NF‐α1

WT

KO

1 mm 200 m

CA1

FC‐CA1

FC

CA1

FC‐CA1

FC

FC‐DG

FC‐DG

A B

C D

Cheng et al.  Figure S3

0

10000

20000

30000

40000

50000

60000

70000

80000

WT KO

Total immun

oreactivearea

 (pixels) per hem

isph

ere

WT

KO

CA2CA1

A.

B.

Cheng et al. Figure S4

AWT KO

WT+FGF2 KO+FGF2

B

0

100

200

300

WT KO KOWTFGF2

Num

ber o

f Sox2/

GFAP +vecells

Veh

Cheng et al., Figure S5

KO

FGF20

50

100

150

200

DCX

/18S

 (%)

Veh

*

Cheng et al., Figure S6

A.

0

200

400

600

800*

Num

ber o

f Ki67 +ve Ce

lls

WT KO KOWTFGF2

B.

Veh

Cheng et al., Figure S7

Red: Sp1  Blue: DAPI

A. B.

Cheng et al., Figure S8

Control

Ctrl0

50

100

150

200**

Sp1

sign

alin

the

nucl

eus (

%)

NF‐α1

NF‐α1

Mouse

Rat

Human

1

1

1

2925

2501

2842

Cheng et al., Figure S9

Cheng et al., Figure S10

100 431Array intensity

‐ +NF‐α1

Cheng et al., Figure S11

6 wk  old  mice (n=8) 14 wk  old  mice (n=8) 30 wk  old  mice (n=8)

WT (n=4) KO (n=4) WT (n=4) KO (n=4) WT (n=4) KO (n=4)

Slide # F F M M F F M M F F M M F F M M F F M M F F M M

1 +

2 ++ ++ ++ ++ +++ +++ ++ ++ + ++ ‐ ++

3 +++ ++ +++ ++ ++ ++ +++ ++ +++ ++ +++ ++ + ++ ++ ++ +++ + ‐ ‐ ++

4 ++ ++ ++ ++ + ++ ++ + ++ + ++ ++ ++ + +++ ++ ++ ++ + ++ + ‐ ‐ +

5 ++ ‐ + + ‐ ‐ ++ + ‐ + + ++ + ++ ++ + + ‐ + + ‐ ‐ ‐ ‐

6 + ‐ + + ‐ ‐ + + ‐ ‐ ‐ + ‐ + + + ‐ ‐ ‐ + ‐ ‐ ‐ ‐

7 ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐

8 ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐

Total 8 7 7 6 5 6 8 3 6 8 5 8 5 4 6 5 8 6 6 8 4 0 0 5

Scores 28 22 27 20 28 9

‐6 ‐7 ‐19

245 m

1750

 m

Cheng et al., Table S1