Identification of NGF-response element in the rat neuropeptide Y gene and induction of the binding proteins

BIOCHEMICAL

Vol. 189, No. 3, 1992 December

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS Pages

30, 1992

1553-l

560

IUENTIFICATION OF NGF-RESPONSE ELIN TBE RAT NEUROPEPTIUEY GENEANUINUUCTIONOFTBEBINUINGPROTEINS Hiroshi

Higuchi,’ Koichi

Nakano and Naomasa Miki

Department of Pharmacology I, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka 565, Japan Received

November

24,

1992

Y (NPY) increases by 100 SUMMARY Gene expression of the rat neuropeptide into sympathetic neuron-like cells times, as the PC12 cells differentiate with NGF treatment and this increase is partly due to transcriptional activation of the NPY gene (Sabol and Higuchi, Mol. Endocrinol. 4, 384, 1990). To identify the NGF-response element, a transient expression assay was carried out by using the CAT reporter genes containing various lengths of The 48-base the 5’ upstream region of the NPY gene in the PC12 cells. element (-80/-33 upstream of the Cap site) was identified as a NGF-response the existence of at least two element (NGFRE). Gel shift assay indicated The binding activity of the protein(s) DNA-binding proteins to NGFRE. (NDFl) to the upper region (-SO/-63) was increased by J-fold with NGF treatment for 24 h. These findings suggest that these nuclear proteins are involved in the enhanced transcription of the NPY gene by NGF. o 1992?,cademl. PTPSL,Inc.

The rat studying ment

PC12 pheochromocytoma

the molecular

with

basis

GTP) is produced tion

of

protein),

of nerve

NGF differentiates

cells through high affinity horylation of Trk A itself both

within

the

tional

factors of

growth

provides factor into

a useful

model for Treat(NGF) actions.

sympathetic

neuron-like

NGF receptor (Trk A), following tyrosine phospwithin 5 min (1). The active form of Ras (Ras-

20 min after

NGF treatment stimulator)

as the result

of activa-

and GAP (GTPase activating

the

various serine/threonine kinases are actiIt has been suggested that the various active Ras (2,3). are transcriptional factors, genes, which are generally

induced by the activation of these late genes such as neuron-specific induction

line

PC12 cells

GDS (GDP-dissociation

and subsequently

vated by the immediate-early

cell

slowly late

protein kinases in 2 h (4,5). Then the genes are induced by these transcrip-

and persistently

genes

results

over

in the

several

differentiation

hours

or days, and the of

PC12 cells

into

1To whom correspondence should be addressed. 0006-291X/92

1553

$4.00

Copvight 0 1992 by Academic Press, Inc. Ali rights of reproduction in any form reserwd.

Vol.

189,

No.

BIOCHEMICAL

3, 1992

sympathetic

neuron-like

However,

cells.

nism of how the neuron-specific Neuropeptide

into

The data To NGF (6).

responsiveness

that

activation

synthesis. by

little

at

that the

is known

of

neuron-like

about

in the cells

is partly

NPY gene, which

level,

mecha-

neurons

by NGF. (by 8 fold)

requires

we analysed

region of the NPY gene by chloramphenicol acetyltransferase and identified the DNA-binding factors to NGFRE.

(6-11).

PC12 cells, Nuclear due to

ongoing

the NPY gene is one of late cis-element (NGFRE) which

transcriptional

the

by NGF.

sympathetic

by loo-fold

increase

of the

indicate identify the

this

RESEARCH COMMUNICATIONS

genes are induced

is increased

the sympathetic

on assay demonstrated transcriptional

late

Y (NPY) is a cotransmitter

The NPY gene expression differentiated

AND BIOPHYSICAL

the

when runthe

protein

genes induced confers NGF5’ upstream (CAT) assay

MATERIALSANDMETEODS Materials. NGF was obtained from Sigma. [a- 32P]dCTP and [14Clchloramphenicol were Restriction and modifying enzymes were from purchased from Amersham. Toyobo and Takara. Cell culture. Rat PC12 pheochromocytoma cells, obtained fro ATCC (American Type Culture Collection) were cultured at 37°C in 75 cm 9 flasks in 85% DMEM, 10% heat-inactivated horse serum, and 5% heat-inactivated fetal calf serum with an atmosphere of 90% air-lot carbon dioxide. Medium was changed every day (6). Plasmid construction. The source of rat NPY genomic DNA was a plasmid subclone (pRNPYKB1) containing 3.5kb KpnI-BglII fragment covering the 5’-upstream region, first exon, and first intron of rat NPY gene (12). The upstream region (3.5 kb KpnI-BglII fragment) was inserted into the promoterless pUCATSV1 CAT reporter plasmid (12). Deletion mutants containing various lengths of 5’-flanking region of the rat NPY gene were obtained by digestion of the longest pUCATNPY plasmid with appropriate restriction enzymes or by PCR. DNA sequencing. Inserted portions of pUCATNPY plasmids were confirmed with the pUC sequencing kit (Boeringer) by double stranded sequencing. DNA transfection. CaP04-DNA precipitates (10-25~ g of test plasmid per plate) were prepared as described (12). After incubation for 6 h with CaP04-DNA precipitates, cells were shocked for 2 min with 15% glycerol in HBS buffer. The transfected cells were incubated with NGF for 48 h and harvested for the measurement of CAT activity. Most experiments included a positive control transfection with pRSVCAT DNA. CAT assay. CAT assays were carried out as previously described (12). Protein (500 fig) from cell extract was incubated at 37°C for 30 min in 180~ 1 of the reaction mixture containing 0.2 M Tris-HCl (PH 7.5), 80 mM acetyl CoA, 3.7 kBq [14Clchloramphenicol. Acetylated and nonacetylated forms of [14C] chloramphenicol were separated on thin layer plate (Whatman LKGDF) and quantitated by scanning the plate with a Fujix Bioimage Analyzer (BAS2000). Gel retardation assay. Nuclear extracts from the PC12 cells were prepared according to the method of Dignam et al (13). Protein assay was carried out by Bradford’s method (14). The two double-stranded oligonucleotides (NGFUds (-80/-51) and NGFDds (-62/-33)), and other oligonucleotides containing the consensus 1554

Vol.

189,

No.

BIOCHEMICAL

3, 1992

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

sequences (APl, Spl, and zif268) were synthesized. Gel retardation assay was performed on 4% polyacrylamide gel for 2 h at 150 V in a solution containing 6.7 mM Tris-glycine (pH 7.9), 3.3 mM sodium acetate and 1 mM EDTA, as previously described (15) and the bands of DNA-protein complexes were quantitated by a Fujix Bioimage Analyzer (BAS2000).

RESULTS

In coincidence

with

AND

the previous

DISCUSSION

finding

that

undifferentiated

express low amount of NPY mRNA (0.16 pg/p g total PY (-2.9kb/+397b) CAT reporter gene exhibited the ity,

which

activity rat

was estimated

was lower

to

than

preproenkephalin

be 0.3 % of

that

that

RNA; Ref.lG), the pUCATNlow transcriptional activof

control

of a pUCATENKHSBL plasmid

A (ENK A) gene promorter

PC12 cells

pRSVCAT. which

from -1.2kb

This

contains

to

the

+49b

in the

markedly

the

same CAT vector. However,

treatment

transcriptional accordance Ref.6).

with

the

As shown determined various

in

not

Fig.

data

this

lA,

regions

near

the

finally

identified

of

inducing

the

bind

element

Cap site

to

the middle unique

conserved

81% homology), this

region

a small

portion.

28 bp

near

(Fig.

through

high

by

(-80/-33 between

in

1A and 1 x

affinity

that

the

structure

primers Cap site) (Fig.

(Fig.

as a in

factors

2).

This

cis

5’-GAGGCCCCTC-3’) in

(-90/-63;

a part

of the

5’-TGGGAGTCACCC-

ester-

similar

and

1B).

transcriptional

constitutes

in phorbol

the

promoter

(39 nucleotides

(-61/-52;

motifs

the

the

conserved

(-80/-63)

using

with

proper

of

was

enzyme sites.

NGF-inducibility

structure

are

by

genes

upstream

lated

There

NPY gene

Cap site

human and rat

suggesting

region

the

PCR using

5-fold

palindromic

(16).

of

at restriction

are evolutionarily

The upper

the

GGGCGTGACTGCCC-J’),which may participate transcription

8.5-fold,

assay

NPY-CAT reporter

palindromic

imperfect

element

ligation

constructed

a 48 bp-element

contains

by

the NPY-CAT gene was about

(-80/+18)

made by the

we made the

This NGFRE is well which

run-on

is mediated

NGF-response

NGFRE of the NPY gene which confers 48 nucleotides;

nuclear

induction

as a 98 bp-element analysis

48 h induced

shown).

NPY-CAT plasmids

further

for

pUCATNPY (-2.9kb/+397b)

of NGF for

that

(data

the

previous

M, indicating

NGF receptor

NGF (50ng/ml)

of

The ED50 value

10-l’

For

with

activity

and CAMP-stimu-

to APl

(-72/-65;

5’-

TGACTGCC-3’), AP2 (-65/-58; 5’-CCCCGAGG-J’), zif268 (NGFIA) (-68/-60; 5’TGCCCCCGA-3’). and Spl (-57/-51; 5’-CCCCTCC-3’) consensus sequences in NGFRE (Fig.

2).

Especially

100% identical

the putative

between

APl and Spl consensus-like

human and rat,

whereas

neither

sequences

AP2 nor

are

NGFIA con-

sensus sequence is conserved in the human gene (17, 18). This NGFRE is not similar to the cis-elements of other genes which confers NGF-stimulated transcription, such as vgf gene (19). 1555

vol.

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

189, No. 3, 1992 A

-3k

-4k CAT fold-increase

b

1 Kpn

I Xba

-2k I I Hint

-1k

lk

Cap 8 I Nhe

IV Xho

--.

Sph

%itc

---I

7.1f0.6

1

8.5fl.6

I C& I CAT

7.0f0.7

4.0~0.8

I

f

CAT

4.2i0.2 4.1a0.7 6.6~1

.O

NGF-RE

B

I-

6

PUCAT NGFl -60/+16

pUCAT NGFP -62/+16

pUCAT NGF3 -32/+16

pUCAT (NGF3)3

pUCATSV1

C-32/+16)3

Fig.1. Deletion analysis of NPY promoter function in PC12 cells. Plasmids containing different fragments of the NPY 5’ upstream region ligated to the CAT gene were introduced into PC12 cells by CaPO -glycerol transfection, and their CAT activities in the presence of NGF 4 50 rig/ml) were compared to those in the absence of NGF. (A) Structures of NPYCAT fusion plasmids and NGF-induced CAT activity. Various pUCATNPY Materials and Methods. plasmids were constructed as described in In each plasmid, the 3’ end of the NPY gene is positioned at +397 (SphI) or +18 (Sac11 from the Cap site. The upstream regions varying from -3.3kb to -8Ob were fused to CAT gene as indicated. The NGF-induction in CAT activity is indicated as fold-increase (mean + SEM) in the left. The NGF response element was narrowed between -80 and +18. (B) NGFinduction of CAT activity of pUCATNGF plasmids. pUCATNGF reporter plasmids with restricted promoter regions of the NPY gene in the pUCATSV1 vector were constructed by PCR. Open and closed columns represent control and NGF-treated groups, respectively. The inserted regions are shown under the columns. pUCAT(NGF313 has three tandem

repeats of the fragment

(-32/+16) in front

of the CAT coding region.

To identify the DNA-binding proteins to NGFRE, we synthesized two oligonucleotides which include the imperfect palindromic structure (-61/-521, designated NGFUds (upstream; -8O/-51) and NGFDds (downstream; -62/-331, and As shown in Fig. 3, performed the gel retardation assay (Fig. 2 and 3). 1556

Vol.

189,

No.

3, 1992

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

-80

CCCGGGCGTGACTGCCCCCGAGGCCCCTCCTGCCGCGACAAGGGCG-~~ . * -80 .

NGFUds

-5 1 * -62 .

NGFDds

-33 *

Fig.2. NGF response element (NGFRE)of the rat NPY gene. The sequence between -80 and -33. the evolutionarily conserved regions (APl- and Spl-like), and the imperfect palindromic structure (t-H ) are shown. The arrows indicate the two oligonucleotides (NGFUds and NGFDds) used for the gel retardation assay.

there

were found

specifically, molar

excess

binding

at

least

in the of

two proteins,

nuclear

extracts

each unlabeled

of nuclear

proteins

complex

was completely

inhibited

stream)

oligonucleotide

(Fig.

the

palindromic

loo-fold

molar

NGFRE (-80/-33)

One hundred-fold completely

the

NGFUds and NGFDds, respec-

using

loo-fold

32P-labeled molar

3B), suggesting

portion

that

of 32P-labeled

NGFDds (down-

excess

of

a nuclear NGFDds.

NGFUds (upprotein

could

The binding

pro-

since the Spl protein can bind the proteins, (5’-CCCCTCC-3’) in the human NPY gene (18) and

teins seemed to be Spl-like Spl consensus-like sequence also

by

to

suppressed

to the 32P-labeled

stream) to

can bind

PC12 cells.

oligonucleotide

tively (Fig. 3A and 3B). The formation of DNA-protein

bind

which

of the

excess of cold

GGCGGGGCGATC-3’) suppressed

Spl consensus

formation

of

sequence

DNA-protein

(5’-GATCGATCGG-

complex

(data

not

shown). The specific abundant

binding

(by lo-20-fold)

to 32P-labeled NGFRE-binding

binding

bind

proteins

that

(Fig.

proteins

portion

3’) and the

than

NGFUds was hardly

NGFDds oligonucleotide the upper

of nuclear

to NGFUds (upstream)

to NGFDds (downstream). inhibited

3A).

This

to the upper proteins

and this

(by 5%) by the addition indicated

that

portion

(-80/-63)

possess APl consensus-like of nuclear

were more

to

sequence this

portion

most

binding of cold

(95%) of

of NGFRE.

(-72/-65;

the Since

5’-TGACTGCC-

was competed

with

loo-fold molar excess of the APl consensus sequence (5’-CTAGTGATGAGTCAGCCGGATC-3’) by 92% (data not shown), the binding proteins to NGFUds seem A single band formation with Spl and APl consento be the APl protein. sus-like

sequences

in NGFUds shown

in Fig. 3A could be separated Thus there are at least two

acrylamide gel (data not shown). proteins which bind to the APl consensus-like consensus-like sequences in NGFRE, respectively. 1557

(more

abundant)

on 10% nuclear and

Spl

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Vol. 189, No. 3, 1992

A

NGFUds

NE.(,,g)

-

10

10

10

Competitor

-

-

u

D

-

10

-

-

Free

10

D

C

NGFComp

Free

NGFUds

”

C

NGF Comp

NGFDds

PC 12 cells

.~

oc

10

0

4

a

24 Time (hr)

Binding to Gel retardation assay with NGFRE oligonucleotides. the 5’ end-labeled 30-bp double-stranded oligonucleotides (NGFUds c-80/51) (A) and NGFDds t-62/-33) (B)) were performed with equal amounts of nuclear extracts (10 ,ug of protein) of PC12 cells. Competitions were performed with a 100x excess of competitor (U:NGFUds, D;NGFDds). The (C) NGF-induction of DNAarrows indicate the DNA-protein complex. protein complexes. PC12 cells were treated with or without NGF (50ng/ml) for 24 h. Ten fig of nuclear protein from treated (NGF) or untreated (C) cells were used for gel retardation assay, Comp means that a loo-fold molar excess of cold oligonucleotide was present in the assay of the NGF-treated sample. (D) Time course of NGF-induced change of binding activity. PC12 cells were cultured with or without NGF (50 rig/ml) for indicated periods and then nuclear extracts were prepared. The specific bands of DNA-protein complexes using NGFUds and NGFDds were quantitated as described in Materials and Methods. Data are mean of 3-4 independent samples. Fig.3.

We designated

binding (-72/-65) Next by NGF sus-like whereas tally

these

nuclear

proteins

as NDFl

(NGF-induced

nuclear

DNA-

protein factor-l) and NDF2. which may bind the APl consensus-like and the Spl consensus-like t-57/-51) elements, respectively. we investigated the change in binding activities of NDFl and NDF2 treatment (Fig. 3C and 3D). The binding of NDF2 to the Spl consenelement did not change (at most 120% 24 h after NGF treatment), that of NDFl to the APl consensus-like sequence increased biphasi-

upto

254% until

24 h in NGF-treataed

1558

PC12 cells

(Fig.

3D).

Vol.

189,

BIOCHEMICAL

N’o. 3, 1992

Although

NDFl seemed to

possibility

of

formation

unidentified

of

AND BIOPHYSICAL

be APl APl-like

DNA-protein

complex

proteins

is

c-Fos not

typical

increased

(data

sequence

(5’-GATCTCGCGGGGGCGAGGGGGATC-3’(loo-fold )

are

the NDFl binding

necessary

to

Interestingly

moderately

characterize

the

cold

(data not binding

and c-Jun),

excluded, APl

only

treatment inhibited

shown).

(e.g.

by using

(5’-CTAGTGATGAGTCAGCCGGATC-3’) was not

proteins

RESEARCH COMMUNICATIONS

since

consensus

within zif268

4h

the

element

after

NGF

(NGFIA) consensus

molar

excess)

shown). Further

proteins

the

to

the

also

experiments NGFRE in

the

NPY gene promoter. The data nuclear

proteins

NPY gene. of

suggested Actually

weaker

this than

the

increase

(NDFl) is essential

TK promoter

However,

that

an inverted

CAT gene of whole

DNA-binding

NGF-stimulated

activity

via

sequence (-80/-631 exhibited an enhancer

the

upper

NGFRE element,

portion

of

transcription

repeat

(20) still

transactivation that

for

in

the of the

ligated in front characteristics.

of NGFRE is relatively

as the CAT reporter

gene with

an

inverted repeat sequence (-80/-63) showed 1.9-fold induction of CAT activity by 48 h-NGF treatment (data not shown). Therefore it is suggested that the increase tant

but

in the binding not

enough

of NDFl to the APl consensus-like for

full

NGF-inducibility

pUCATNGF2 (-62/+16) plasmid

still

and

synergistically

NDF2 probably

The identification like)

proteins

NGF stimulates

exert

transcription

the

on of

to understand of

the

NGF-induced

NDFl (APl-like) the molecular

NPY gene

through

is impor-

NPY gene.

have a weak NGF-inducibility

and characterization are inevitable

of

element

As the (Fig. lB), NDFl

transactivation. and NDF2 (Splmechanism of how high-affinity

NGF

receptor.

REFJXRENCES 1. Kaplan, D. R., Hempstead, B. L., Martin-Zanca, D., Chao, M. V. and Parada, L. F. (19911 Science 252, 554-558. 2. Li, B.-Q., Kaplan, D., Kung, H. and Kamata, T. (1992) Science 256, 1456-1459. 3. Wood, K.W., Sarnecki, C., Roberts, T. M. and Blenis, J. (1992) Cell 68, 10411050. 4. Milbrandt, J. (1987) Science 238, 797-799. 5. Crosby, S. D., Veile, R. A., Donis-Keller, H., Baraban, J. M., Bhat, R. V., Simburger. K. S. and Milbrandt, J. (1992) Proc. Natl. Acad. Sci. USA 89, 4739-4743. 6. Sabol, S. L. and Higuchi, H. (1990) Mol. Endocrinol. 4, 384-392. 7. Higuchi, H., Costa, E. and Yang, H.-Y. T. (1988) J. Pharmacol. Exp. Ther. 244, 468-474. 8. Higuchi, H., Yang, H.-Y. T. and Costa, E. (1988) J. Neurochem. 50, 18791886. 9. Higuchi, H., Iwasa, A., Yoshida, H., and Miki, N. (1990) Mol. Pharmacol. 38, 614-623. 10. Higuchi, H., Yokokawa, K., Iwasa, A., Yoshida. H., and Miki, N. (1991) J. Neurochem. 57, 1840-1847. 11. Allen, J. M. and Koenig, J. I. (1990) Annals New York Acad. Sci. Vol. 611.. 1559

Vol. 189, No. 3, 1992

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMilNlCATlONS

12. M isaki. N.. Higuchi, H., Yamagata. K., and M iki, N. (1992) Neurochem. Int. 21, 185-189. 13. Dignam. J. D., Lebowitz, R. M . and Roeder, R. G. (1983) Nucleic Acids Res. 11, 1475-1489. 14. Bradford, M . M . (1976) Anal. Biochem. 72, 248-254. 15. Osugi, T., Taniura, II.. Ikemoto, M . and M iki, N. (1991) Biochem. Biophys. Res. Commun.174, 25-31. 16. Higuchi, H., Yang, H.-Y. T., and Sabol, S. L. (1988) J. Biol. Chem. 263, 6288-6295. 17. Larhammar, D., Ericsson, A. and Persson, H. (1987) Proc. Natl. Acad. Sci. USA 84, 2068-2072. 18. M inth, C. D., and Dixon, J. E. (1990) J. Biol.Chem. 265, 12933-12939. 19. Possenti, R., Di Rocco, G.. Nasi, S. and Levi, A. (1992) Proc. Natl. Acad. Sci. USA 89. 3815-3819. 20. Luckow, B. and Schitz, G. (1987) Nucleic Acids Res. 15, 5490.

1560