Cloning and characterization of SARI (suppressor of AP-1, regulated by IFN).
ABSTRACT We describe a novel basic leucine zipper containing type I IFN-inducible early response gene SARI (Suppressor of AP-1, Regulated by IFN). Steady-state SARI mRNA expression was detected in multiple lineage-specific normal cells, but not in their transformed/tumorigenic counterparts. In normal and cancer cells, SARI expression was induced 2 h after fibroblast IFN (IFN-beta) treatment with 1 U/ml of IFN-beta. Antisense inhibition of SARI protected HeLa cells from IFN-beta-mediated growth inhibition. As a corollary, overexpression of SARI inhibited growth and induced apoptosis in cancer cells, but not in normal cells. SARI interacted with c-Jun via its leucine zipper, resulting in inhibition of DNA binding of activator protein (AP-1) complex and consequently AP-1-dependent gene expression. Transformed cells relying on AP-1 activity for proliferative advantage demonstrated increased susceptibility to SARI-mediated growth inhibition. These findings uncover a novel mode of IFN-induced anti-tumor growth suppression and suggest potential gene therapy applications for SARI.
- SourceAvailable from: uah.es
Article: How cells respond to interferons.[show abstract] [hide abstract]
ABSTRACT: Interferons play key roles in mediating antiviral and antigrowth responses and in modulating immune response. The main signaling pathways are rapid and direct. They involve tyrosine phosphorylation and activation of signal transducers and activators of transcription factors by Janus tyrosine kinases at the cell membrane, followed by release of signal transducers and activators of transcription and their migration to the nucleus, where they induce the expression of the many gene products that determine the responses. Ancillary pathways are also activated by the interferons, but their effects on cell physiology are less clear. The Janus kinases and signal transducers and activators of transcription, and many of the interferon-induced proteins, play important alternative roles in cells, raising interesting questions as to how the responses to the interferons intersect with more general aspects of cellular physiology and how the specificity of cytokine responses is maintained.Annual Review of Biochemistry 02/1998; 67:227-64. · 27.68 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The innate immune system plays critical roles in recognizing viral infections and evoking initial anti-viral responses. Nucleotides from RNA viruses are recognized by retinoic acid-inducible gene I (RIG-I)-like helicases (RLHs) and Toll-like receptors (TLRs), and the recognition triggers signaling cascades that induce anti-viral mediators such as type I interferons (IFNs) and pro-inflammatory cytokines. The RLH signaling pathways play essential roles in the recognition of RNA viruses in various cells, with the exception of plasmacytoid dendritic cells (pDCs). However, TLRs are important for the production of type I IFNs in pDCs but not in other cell types. The contributions of RLHs and TLRs to the production of type I IFNs in response to RNA viruses vary depending on the route of infection. Specifically, local infections induce IFNs through RLHs but not TLRs, whereas systemic infections strongly stimulate TLRs in pDCs. In this review, we discuss recent advances toward clarifying the signaling pathways activated by RLHs and TLRs.Immunological Reviews 01/2008; 220:214-24. · 12.16 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The interferons (IFNs) play an integral role in cellular host defense against virus infection and conceivably tumorigenesis. Despite over 50 years of research, however, the molecular mechanisms underlining IFN action remain to be fully elucidated, in part because of the large number of genes, with an uncharacterized function that appears to be induced by these cytokines. Although the majority of in vitro studies indicate that IFNs antiviral properties involve inhibiting viral replication while maintaining the integrity of the cell, numerous reports have now implicated that a number of IFN-induced genes, IFN transcriptional regulatory factors and IFN signaling molecules can also mediate apoptosis. Here, we review some of what is known about IFN's ability to invoke programmed cell death as part of an intricate arsenal intended to prevent viral infection and malignant disease.Seminars in Cancer Biology 05/2000; 10(2):103-11. · 7.44 Impact Factor
Cloning and characterization of SARI (suppressor
of AP-1, regulated by IFN)
Zao-zhong Sua,1,2, Seok-Geun Leea,b,1, Luni Emdada, Irina V. Lebdevaa, Pankaj Guptaa, Kristoffer Valeriec,d,
Devanand Sarkara,b,c,1, and Paul B. Fishera,b,c,2
Departments ofaHuman and Molecular Genetics andcRadiation Oncology,dVCU Institute of Molecular Medicine, andbMassey Cancer Center,
Virginia Commonwealth University School of Medicine, Richmond, VA 23298
Edited by George R. Stark, Cleveland Clinic Foundation, Cleveland, OH, and approved November 1, 2008 (received for review August 12, 2008)
We describe a novel basic leucine zipper containing type I
IFN-inducible early response gene SARI (Suppressor of AP-1,
Regulated by IFN). Steady-state SARI mRNA expression was
detected in multiple lineage-specific normal cells, but not in their
transformed/tumorigenic counterparts. In normal and cancer
cells, SARI expression was induced 2 h after fibroblast IFN (IFN-?)
treatment with 1 U/ml of IFN-?. Antisense inhibition of SARI
protected HeLa cells from IFN-?-mediated growth inhibition. As
a corollary, overexpression of SARI inhibited growth and in-
duced apoptosis in cancer cells, but not in normal cells. SARI
interacted with c-Jun via its leucine zipper, resulting in inhibition
of DNA binding of activator protein (AP-1) complex and conse-
quently AP-1-dependent gene expression. Transformed cells
relying on AP-1 activity for proliferative advantage demon-
strated increased susceptibility to SARI-mediated growth inhi-
bition. These findings uncover a novel mode of IFN-induced
anti-tumor growth suppression and suggest potential gene
therapy applications for SARI.
IFN-inducible gene ? Jun-interacting protein ? cancer growth
effects have been attributed to these molecules, such as inhibi-
tion of viral, bacterial, and parasitic pathogenesis; inhibition of
cell growth; induction of apoptosis; inflammation; immuno-
modulation; and anti-angiogenesis (1–6). IFNs exert their di-
verse effects by stimulating the expression of a plethora of
IFN-stimulated genes in target cells (2). Both type I (IFN-?,
IFN-?, IFN-?, and IFN-?) and type II (IFN-?) IFNs exert potent
anti-tumor effects and are used clinically either as a mono-
therapy or as an adjuvant to chemotherapy or radiotherapy for
a number of solid tumors and hematological malignancies that
include melanoma, renal cell carcinoma, Kaposi sarcoma, ma-
lignant glioma, lymphomas, and leukemias (1). The anti-tumor
effects of IFNs are mediated by direct inhibition of proliferation
of cancer cells as well as by indirect effects such as inhibition of
tumor angiogenesis and up-regulation of tumor-specific antigens
and adhesion molecules.
In an in vitro system, treatment of human melanoma cells with
type I IFN (IFN-?) and the protein kinase C activator mezerein
(MEZ) induces irreversible growth arrest and ‘‘terminal differ-
entiation’’ (7, 8). Differential gene expression analysis between
IFN-?/MEZ-treated terminally differentiated versus control
melanoma cells identified a variety of novel ‘‘melanoma differ-
entiation-associated’’ (mda) genes, the functional characteriza-
tion of which has revealed their central role in various IFN-
regulated events, such as viral interference, growth inhibition,
and induction of apoptosis (9–16).
Activator protein-1 (AP-1), one of the first identified mam-
malian transcription factors, plays an essential role in regulating
cell proliferation and oncogenic transformation (17). AP-1
proteins that belong to the Jun (c-Jun, JunB, JunD), Fos (c-Fos,
lthough IFNs were originally identified as proteins confer-
ring viral interference, in the past 50 years a variety of
FosB, Fra-1, and Fra-2), Maf, and ATF subfamilies that recog-
nize either the 12-O-tetradecanoylphorbol-13-acetate (TPA)
response element (5?-TGAG/CTCA-3?) or the cAMP response
element (5?-TGACGTCA-3?) (18). Among the Jun proteins,
c-Jun is the most potent transcriptional activator, and the
c-Fos-c-Jun heterodimer positively regulates cell proliferation
and transformation (17, 19, 20). c-Jun co-operates with Ha-ras
in rat embryonic fibroblast transformation and c-jun?/? cells
are refractory to the transforming activity of oncogenic Ras (21,
22). Additionally, overexpression of c-Jun itself can immortalize
rodent fibroblasts in culture (21). Augmented AP-1 activity has
been observed in more than 90% of human cancers, indicating
its seminal importance in human carcinogenesis (23).
The present manuscript describes the cloning and character-
ization of a novel type I IFN-inducible early response gene. The
gene product is a bZIP containing protein that inhibits cell
growth by interacting with c-Jun and inhibiting AP-1 activity.
The gene was named SARI (Suppressor of AP-1, Regulated by
IFN) and its functional analysis reveals a perviously unrecog-
nized molecular pathway underlying the anti-tumor action of
Cloning of SARI. In an effort to unravel the molecular basis of
terminal differentiation, we performed differential gene ex-
pression analysis by rapid subtraction hybridization employing
temporal RNAs isolated from solvent-treated (i.e., control)
and IFN-?/MEZ-treated (i.e., terminally differentiated) HO-1
human melanoma cells (9). This allowed us to clone a broad
spectrum of mda genes. One of the mda genes, later named
SARI based on its functional properties, was originally iden-
tified as a novel gene, mda-D-74. The cDNA consists of 2,140
bp and codes for a putative protein of 274 aa residues with a
predicted molecular mass of 29.4 kDa and pI 7.2 (Fig. 1A).
Genomic BLAST search indicated that the SARI gene is
located in the long arm of chromosome 11 between 11q12 and
11q13. The gene consists of three exons. Translation starts in
exon 1 and the stop codon is located in exon 3 (Fig. 1B).
Sequence analysis using InterPro Scan identified an L-X6-L-
X6-L-X6-L motif between 45 and 66 aa residues preceded by a
highly basic region, suggesting that SARI might be a basic-
region leucine zipper containing transcription factor (Fig. 1A).
Author contributions: Z.-z.S., S.-G.L., L.E., I.V.L., D.S., and P.B.F. designed research; Z.-z.S.,
S.-G.L., L.E., I.V.L., and D.S. performed research; Z.-z.S., S.-G.L., L.E., P.G., and K.V. contrib-
uted new reagents/analytic tools; Z.-z.S., S.-G.L., L.E., I.V.L., D.S., and P.B.F. analyzed data;
and D.S. and P.B.F. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1Z.-z.S., S.-G.L., and D.S. contributed equally to this study.
2To whom correspondence may be addressed. E-mail: firstname.lastname@example.org or email@example.com.
This article contains supporting information online at www.pnas.org/cgi/content/full/
© 2008 by The National Academy of Sciences of the USA
December 30, 2008 ?
vol. 105 ?
no. 52 www.pnas.org?cgi?doi?10.1073?pnas.0807975106
Gene expression analysis identified SARI as a type I IFN
(IFN-?/?)-inducible early response gene. In HO-1 human mel-
anoma cells and SV40 T/tAg-immortalized primary human
melanocytes (FM516-SV), SARI mRNA expression could be
detected as early as 2 h after IFN-? treatment (100 U/ml) with
maximum induction at 12 h post-treatment with a gradual
decrease thereafter and a return by 48 h to the basal level of
expression (Fig. 1C). The induction could be detected with as
little as 1 U/ml of IFN-?. Interestingly, in FM516-SV cells, but
not in HO-1 melanoma cells, SARI expression could be detected
under de novo conditions (Fig. 1C). A similar expression and
induction profile could be detected in a series of normal and
cancer cells from diverse tissue origins [supporting information
(SI) Fig. S1]. More importantly, under basal conditions, SARI
mRNA expression could be detected in multiple normal cells,
including primary human fetal astrocytes (PHFAs), immortal
normal mammary epithelial cells (HBL100), immortal prostate
epithelial cells (P69), and immortal pancreatic mesothelial cells
(LT2), but not in their malignant counterparts (Fig. S1). Quan-
titative RT-PCR confirmed that SARI mRNA expression was
significantly higher in PHFA versus T98G malignant glioma
cells, P69 versus DU145 prostate cancer cells, as well as
FM516-SV versus HO-1 melanoma cells (Fig. 1D). Treatment
with cycloheximide, a protein translation inhibitor, did not by
itself induce SARI mRNA expression and moderately increased
IFN-?-mediated induction, suggesting that IFN-? predomi-
nantly regulates SARI expression on a transcriptional level
(Fig. 1E). The half-life of the IFN-?-inducible transcript was ?2
h as revealed by actinomycin D treatment, which prevents new
RNA transcription (Fig. 1F). Multiple tissue Northern blots
revealed SARI expression in diverse tissues, with highest expres-
sion in pancreas and spleen and moderate expression in colon,
heart, kidney, liver, lung, and prostate (Fig. 1G). Low-level
expression could be detected in placenta, stomach, small intes-
tine, and salivary gland. Brain, muscle, and testis were conspic-
uous by the absence of SARI expression.
SARI Inhibits Growth of Cancer Cells but not Normal Cells. The
observation that basal SARI mRNA expression was detected in
normal cells but not in their malignant counterparts suggested
that SARI might display tumor suppressor functions that are lost
during tumor progression. Based on this consideration, we
analyzed the biological consequences of overexpression of this
molecule in normal and tumor cells. We constructed a replica-
tion-incompetent adenovirus expressing SARI (Ad.SARI), in-
fected PHFA, H4 (malignant glioma), P69, DU-145, FM516-SV,
and MeWo (i.e., melanoma) cells with Ad.SARI (100 pfu/cell)
and monitored cell growth by viable cell counting. Empty
adenovirus (Ad.vec) served as a control. Very interestingly,
Ad.SARI profoundly inhibited cancer cell growth (H4, DU145,
and MeWo) with little to no effect on normal cells (PHFA, P69,
and FM516-SV; Fig. 2A). This finding was confirmed in trans-
independent growth assays in soft agar (data not shown). It
infection was equivalent in all cell lines, as confirmed by Western
blot analysis using anti-SARI antibody (data not shown). To
determine if this inhibition of growth was a result of induction
of apoptosis, flow cytometry studies were performed to follow
Annexin V staining or propidium iodide staining to identify
subG1(A0) cell populations. Ad.SARI, but not Ad.vec, resulted
in a significant induction of apoptosis in cancer cells (H4,
DU145, and HeLa), but not in PHFA (Fig. 2B).
The growth inhibitory effect of SARI was confirmed by in vivo
studies. Subcutaneous xenografts from DU-145 prostate cancer
cells were established in the left flank of athymic nude mice.
zipper are underlined. (B) Genomic structure of the SARI gene. The white boxes represent the exons and the black lines represent the introns. The numbers at
the bottom of the white boxes represent the size of the exons and those on top of the black lines represent the size of the introns in bp. The arrow indicates
the translation start site and the octagon indicates the stop codon. (C) Dose- (Left) and time-dependent (Right) induction of SARI mRNA by IFN-? in HO-1 and
FM516-SV cells analyzed by Northern blotting. GAPDH mRNA expression was used as loading control. (D) Quantitative SARI mRNA expression analysis in cancer
(T98G, DU145, and HO-1) and normal cells (PHFA, P69, and FM516-SV; in arbitrary units [a.u.]). Data represent mean ? SD of three independent experiments.
(E) IFN-? induction of SARI mRNA occurs at a transcriptional level. HeLa cells were treated with cycloheximide (10 ?M) followed by IFN-? (100 U/ml). SARI and
for 0.5 to 6 h. SARI and GAPDH mRNA expression was analyzed by Northern blotting. (D) Analysis of SARI and GAPDH mRNA expression in different tissues using
multiple tissue Northern blot (Clontech).
Su et al.
December 30, 2008 ?
vol. 105 ?
no. 52 ?
When the tumors reached ?100 mm3in size, requiring ?7 days,
intratumoral injections of PBS solution, Ad.vec, or Ad.SARI
(108pfu/injection) were administered three times during the first
week and then twice weekly for two more weeks for a total of
seven injection. Tumor growth was significantly inhibited by
Ad.SARI treatment compared with PBS or Ad.vec treatment
(data not shown). Monitoring animal survival demonstrated that
Ad.SARI significantly prolonged survival compared with PBS-
treated or Ad.vec-infected animals, confirming the tumor-
suppressor properties of SARI in vivo (Fig. 2C).
To interrogate the molecular pathway(s) involved in Ad.
SARI-mediated growth inhibition, we used normal immortal rat
embryonic fibroblasts (CREF) and a clone of CREF trans-
formed by H-ras, v-src, HPV-18, or a specific temperature (i.e.,
cold)-sensitive mutant of type 5 adenovirus H5hrl (CREF-ras,
CREF-src, CREF-HPV, and CREF-H5hrl, respectively) (24,
25). The parental CREF do not grow in soft agar and are
non-tumorigenic, whereas transformed clones all grow in soft
agar and are tumorigenic. Ad.SARI had no discernible effect on
the growth of CREF. Among the tumorigenic clones, a prefer-
ential inhibition of growth was observed in CREF-ras and
CREF-src clones with a significantly reduced effect on CREF-
HPV and CREF-H5hrl clones (Fig. 3A). These findings were
bolstered by clonogenic assays as well as by soft agar assays (Fig.
3B and data not shown). These results indicate that signaling
pathways, activated by ras and src, but not by nuclear-acting
oncogenes such as HPV-18 and H5hrl, make CREF susceptible
to Ad.SARI-mediated growth inhibition.
SARI Mediates IFN-?-Induced Growth Inhibition. Bioinformatics
analysis of SARI predicted a putative basic leucine zipper motif
that might contain a c-JUN-dimerization motif. As c-JUN, a
basic leucine zipper containing protein, is a component of the
AP-1 transcription factor, and because both ras- and src-
mediated signaling result in AP-1 activation, we hypothesized
that SARI might interact with c-JUN, thereby interfering with
AP-1 function and inhibiting growth of cells dependent on AP-1
for proliferation and survival (e.g., CREF-ras and CREF-src).
These assumptions were validated through a series of experi-
ments. We confirmed that CREF were resistant to growth
suppression by Ad.SARI, whereas CREF stably overexpressing
c-jun became susceptible to Ad.SARI-mediated growth inhibi-
tion (Fig. 4A). Stable HeLa cells expressing antisense SARI
(HeLa-SARIAs) were established, and we analyzed expression
and IFN-? induction of SARI protein in these clones. Cytoplas-
mic and nuclear extracts were prepared from parental HeLa
cells, a control HeLa clone not expressing SARIAs, and HeLa-
SARIAs clones untreated or treated with IFN-?. SARI protein
was detected exclusively in the nucleus and induction of SARI
protein by IFN-? was observed in HeLa cells and in the control
clone, but not in HeLa-SARIAs clones, confirming the authen-
ticity of these clones (Fig. 4B). Growth of HeLa and HeLa-
and in vivo. (A) Cancer (H4, DU145, and MeWo) and
with either Ad.vec or Ad.SARI at a multiplicity of
infection of 100 pfu/cell and cell number was counted
at the indicated time points. The data represent
mean ? SD of three independent experiments. (B)
were infected as in A and apoptosis was analyzed by
Annexin V binding assay (Upper) and propidium io-
dide staining assay (Lower), followed by flow cytom-
etry 48 h after infection. (C) Subcutaneous xenografts
were established in the flanks of athymic nude mice
with DU145 human prostate cancer cells. Established
tumors were injected with PBS, Ad.vec, or Ad.SARI as
described in Materials and Methods and survival of
animals was analyzed by the Kaplan-Meier test.
manner. (A) Normal immortal rat embryonic fibroblasts (CREF) and clones of
CREF transformed by H-ras, v-src, HPV-18, or a specific temperature-sensitive
mutant of type 5 adenovirus H5hrl (CREF-ras, CREF-src, CREF-HPV, and CREF-
H5hrl, respectively) were infected with either Ad.vec or Ad.SARI at a multi-
plicity of infection of 100 pfu/cell and cell number was counted at the
indicated time points. (B) Soft agar colony formation assay of CREF, CREF-ras,
CREF-src, CREF-HPV, and CREF-H5hrl cells infected as in A. The data represent
mean ? SD of three independent experiments.
www.pnas.org?cgi?doi?10.1073?pnas.0807975106Su et al.
SARIAs cells were evaluated upon IFN-? treatment. HeLa-
SARIAs cells were more resistant to growth inhibition by IFN-?
compared with parental HeLa cells, indicating that SARI plays
an important role in mediating IFN-? action (Fig. 4C).
Parental HeLa and HeLa-SARIAs cells were transfected with
an AP-1 reporter plasmid, in which the luciferase gene is
regulated by consensus AP-1-binding sites (AP-1-luc), and then
treated with TPA, IFN-?, or both. HeLa and HeLa-SARIAs
cells responded equally to TPA for AP-1-luc induction (Fig. 4D).
However, IFN-? significantly reduced both basal and TPA-
induced activation of AP-1-luc only in HeLa cells. HeLa-
SARIAs cells displayed significant resistance to the inhibitory
effect of IFN-? (Fig. 4D). These observations suggest that under
basal conditions TPA efficiently activates AP-1-luc activity.
However, SARI, upon induction by IFN-?, interfered with
TPA-mediated AP-1-luc activation that could be rescued by
antisense inhibition of SARI induction. These findings were
further confirmed by electrophoretic mobility shift assay
(EMSA) using nuclear extracts from TPA- or TPA/IFN-?-
treated HeLa and HeLa-SARIAs cells and a radiolabeled
consensus AP-1 probe. In HeLa cells, TPA-induced augmenta-
tion of AP-1-binding activity was markedly reduced by IFN-?
treatment (Fig. 4E). This inhibition by IFN-? was profoundly
abrogated in HeLa-SARIAs cells. These findings indicate that
SARI plays a key role in mediating IFN-?-mediated down-
regulation of AP-1 activity. The inhibition of AP-1 activity by
SARI was reflected in the expression level of an AP-1 down-
stream gene, IL-8. TPA treatment resulted in IL-8 mRNA
induction in control and Ad.vec-infected HeLa cells. However,
upon infection with Ad.SARI, TPA-mediated induction of IL-8
mRNA was markedly reduced (Fig. 4F). The steady-state ex-
pressions of cell cycle regulatory proteins such as cyclin D1 and
cyclin E, which are downstream of AP-1, were also significantly
down-regulated upon Ad.SARI infection (Fig. 4G).
SARI Interacts with c-JUN. Potential interactions between c-JUN
and SARI were evaluated using multiple approaches. We used
a mammalian two-hybrid assay in which plasmid pCMV-AD was
used for cloning the c-jun cDNA to be expressed as a fusion with
the activation domain of VP16 (pAD-c-jun) and plasmid
pCMV-BD was used to clone the SARI cDNA to be expressed
as a fusion with the DNA-binding domain of GAL4 (pBD-
SARI). These constructs were transfected along with a reporter
vector that contains five GAL4 binding sites (GAL4UAS)
upstream of a minimal TATA box promoter driving the expres-
sion of a luciferase reporter gene in HeLa, DU145, FM516-SV,
and MeWo cells. Interaction between SARI and c-JUN resulted
in the association of the GAL4 DNA binding domain with the
VP16 transcription activation domain and a significant induction
of luciferase activity was observed only when pAD-c-jun and
pBD-SARI were transfected together (Fig. 5A). HeLa cells were
infected with Ad.SARI and the cells were processed for dual
immunofluorescence analysis using anti-HA antibody to detect
HA-tagged SARI and anti-c-Jun antibody to detect c-Jun. Both
SARI (green) and c-JUN (red) were detected in the nucleus and
the merged images showed overlapping distribution of green,
red, and blue (DAPI detecting nucleus), producing a white color
(Fig. 5B). Co-immunoprecipitation analysis using SARI-
overexpressing HeLa cell lysates and anti-HA and anti-c-JUN
antibodies confirmed their interaction (Fig. 5C). Anti-HA and
anti-c-JUN antibodies could effectively pull down c-JUN and
SARI, respectively. It should be noted that similar assays did not
find any interaction between SARI and either c-Fos or ATF-2,
indicating that the interaction of SARI with c-Jun is specific
(data not shown). Mutation of leucine59in the bZIP domain of
the interaction between c-JUN and SARI. Whereas WT SARI
could significantly inhibit colony formation of CREF-c-Jun-13
cells, the mutant SARI almost completely lost this activity,
indicating that growth inhibition by SARI is mediated by its
interaction with c-JUN (Fig. 5E).
IFNs induce growth inhibition by multiple pathways that involve
many IFN-stimulated genes (2). SARI is one of these genes that,
based on a number of unique properties, highlights its signifi-
cance in IFN-signaling. SARI expression is induced as early as 2 h
after IFN-? treatment with as little as 1 U/ml of IFN-?, pointing
to a role as an early mediator of IFN-? action. Inhibition of SARI
by an antisense approach in HeLa cells significantly abrogated
IFN-?-induced growth inhibition, thus establishing SARI in
mediating IFN anti-proliferative action. Whereas steady-state
expression of SARI mRNA was detected in normal cells of
diverse lineages, such as melanocytes, astrocytes, breast and
prostate epithelial cells, and pancreatic mesothelial cells, expres-
sion was not detected in multiple cancer cell lines of the same
expressing c-jun (CREF-c-jun 13, CREF-c-jun 16, and CREF-c-jun 53) were
treated and analyzed as in Fig. 3A. (B) Cytoplasmic and nuclear extracts were
HeLa clones stably expressing antisense SARI (As) treated with IFN-? or un-
treated, and the expression of SARI was detected using rabbit anti-SARI
antibody. The purity of the extracts was confirmed by analyzing Actin (for
cytoplasm) and Sp1 (nucleus) expression. (C) HeLa cells and HeLa cells stably
expressing antisense SARI (HeLa-SARIAs) were treated with IFN-? (100, 500,
and 2,000 U/ml) and cell numbers were counted at the indicated time points.
(D) HeLa and HeLa-SARIAs cells were transfected with AP-1-luc, treated with
TPA or IFN-? or a combination, and luciferase activity was monitored by a
luminometer. Luciferase activity was normalized by ?-galactosidase activity.
For A–C, data represent mean ? SD of three independent experiments. (E)
HeLa and HeLa-SARIAs cells were treated with TPA or IFN-? and nuclear
extracts were used for EMSA using a radiolabeled consensus AP-1 probe. (F)
HeLa cells were infected with Ad.vec or Ad.SARI at 100 pfu/cell and then
treated with TPA. Expression of IL-8 and GAPDH was analyzed by RT-PCR. (G)
HeLa cells were infected with Ad.vec or Ad.SARI at 50 pfu/cell for 16 h and the
expression of the indicated proteins in total cell lysates were analyzed by
Western blot analysis.
SARI inhibits AP-1 activation. (A) CREF and CREF clones stably over-
Su et al.
December 30, 2008 ?
vol. 105 ?
no. 52 ?
tissue of origin. These findings indicate that, although normal
cell survival is compatible with SARI expression, continued
proliferation of cancer cells requires that SARI expression be
suppressed, a phenomenon characteristic of many tumor sup-
pressor proteins. The most intriguing finding is that, even when
using an adenovirus-mediated delivery approach that ensures
high-level gene expression in ?90% of cells, SARI overexpres-
sion did not adversely affect normal cell survival while inducing
profound growth inhibition and apoptosis in cancer cells. Ac-
cordingly, targeted overexpression of SARI might provide an
effective gene for cancer therapy. The molecular mechanism in
which steady-state SARI expression is suppressed in cancer cells
remains to be determined. However, IFN-induction of SARI
expression is retained by cancer cells, indicating a potential
separation of regulatory control between steady-state and in-
ducible expression of SARI. Experimental evidences indicate
that IFN-? regulates SARI expression at the transcriptional level
and characterization of its promoter region would provide
further insights into the regulation of this interesting gene.
We demonstrate that, whereas CREF are resistant to SARI-
induced growth inhibition, ras, src, and c-jun transformed CREF
clones acquire sensitivity to SARI. Both ras and src signaling
induce c-jun activation indicating that cells that employ AP-1 for
proliferative advantage become susceptible to growth inhibition
by SARI. This is a significant observation for using SARI for
potential cancer gene therapy purposes. Ras, src, and AP-1
activations are extremely common events in the process of
carcinogenesis and SARI would therefore have significant po-
tential to exert broad-spectrum anti-tumor activity (23, 26, 27).
SARI contains a bZIP domain, and our mutational studies
confirm that the leucine zipper in SARI mediates its interaction
with c-Jun. EMSA using a consensus AP-1 probe and in vitro
translated SARI or c-Jun proteins demonstrated that SARI did
not bind to the consensus AP-1 site or the AP-1 site in the IL-8
promoter either alone or as a heterodimer with c-Jun (data not
AP-1 binding site in the IL-8 promoter and anti-SARI antibody
also failed to demonstrate DNA binding by SARI (data not
shown). SARI is a nuclear protein and, in the nucleus, SARI
binds with c-Jun in solution and squelches the DNA-binding
activity of AP-1 complexes, thus inhibiting AP-1-mediated gene
transcription. A relevant question is why SARI, which contains
a putative bZIP basic region domain, fails to bind DNA. One
possible explanation could be the lack of stretches of three or
more basic amino acid residues in the putative basic region of
SARI that is found in other bZIP domain-containing proteins
that do bind DNA (28).
SARI is comparable to the IFN-inducible p200 family of
proteins (IFI-200) (29, 30). IFI-200 family proteins are induced
by IFN, inhibit cell proliferation, and interact with a plethora of
proteins such as retinoblastoma protein, p50/p65 nuclear factor-
?B, c-Fos, c-Jun, c-Myc, and p53. Thus, these proteins display
diverse and pleiotropic effects in target cells. The presence of the
bZIP structural motif in SARI indicates that its interaction
activities might be more focused on bZIP domain containing
proteins. Current studies are focused on identifying other inter-
acting partners of SARI as well as defining its involvement in
regulation of relevant transcription factor functions.
In summary, we identified a novel IFN-inducible nuclear
protein that, by direct protein-protein interaction, interferes
with the function of the AP-1 transcription factor. Apart from
mediating IFN-induced growth inhibition, it would be intriguing
to identify whether SARI plays any role in other IFN actions,
in normal cells, does SARI play any role in regulation of cell cycle
activity? Is it part of the normal cell cycle checkpoint control?
Inhibition of SARI in normal cells as well as generation of a
SARI-KO mouse would provide insights into these questions.
molecular function of this important molecule, which will pro-
vide valuable insights into the mechanism of action of IFNs and
their role as cancer-suppressing molecules.
Materials and Methods
Cell Lines and Culture Conditions. A detailed description of the cell lines used
in this study is provided in SI Text.
Cell Growth and Apoptosis Studies. Cell growth was monitored by trypan blue
dye exclusion assay. Apoptosis was monitored by Annexin V binding assays
and propidium staining followed by flow cytometry as described (31, 32).
Transient Transfections and Luciferase Assays. Transient transfections and
luciferase assays were performed as described (33). Luciferase activity was
normalized by ?-galactosidase activity.
malian two-hybrid assay was performed using pAD-c-jun
HeLa cells were infected with Ad.vec or Ad.SARI at 50
pfu/cell. The cells were fixed 24 h later and subjected to
immunofluorescence analysis using anti-HA and anti-c-
JUN primary antibodies and FITC- and rhodamine-
medium containing DAPI was used for staining the nu-
cleus. Confocal laser scanning microscopy was used to
analyze the images. (C) HeLa cells were infected with
Ad.vec or Ad.SARI at 50 pfu/cell. Total cell lysates were
used for co-immunoprecipitation analysis using anti-HA
antibody for immunoprecipitation and anti-c-JUN anti-
body for immunoblotting and vice versa. (D) Expression
plasmid in which Leucine59of SARI was mutated to pro-
with empty vector (pcDNA) or expression plasmids ex-
pressing WT SARI or SARI-mt. The cell lysates were sub-
C. (E) CREF-c-jun-13 cells were transfected with empty
vector (pcDNA) or expression plasmids expressing WT
SARI or SARI-mt and subjected to clonogenic assay for 2
weeks upon selection with hygromycin. Data represent
mean ? SD of three independent experiments.
SARI directly interacts with c-JUN. (A) Mam-
www.pnas.org?cgi?doi?10.1073?pnas.0807975106 Su et al.
5? and 3? Rapid Amplification of cDNA Ends. 5? and 3? rapid amplification of
cDNA ends was performed by using a GeneRacer kit (Invitrogen) according to
the manufacturer’s instructions.
RNA Extraction, Northern Blot Analysis, RT-PCR, Quantitative RT-PCR, and
Multiple Tissue Northern Blot. Total RNA extraction and Northern blotting
were performed as described (33). The primers used for RT-PCR were as
follows: IL-8 sense, 5? GGTGCAGAGGGTTGTGGAGAA 3?; IL-8 antisense, 5?
GCAGACTAGGGTTGCCAGATT 3?; GAPDH sense, 5? ATGGGGAAGGTGAAG-
GTCGGAGTC 3?; GAPDH antisense, 5? GCTGATGATCTTGAGGCTGTTGTC 3?.
Quantitative RT-PCR was performed using a kit from Stratagene based on
SYBR Green I DNA-binding dye. Multiple tissue Northern blot was obtained
from Clontech and hybridization was performed according to the manufac-
Preparation of Whole-Cell Lysates, Western Blotting, and Co-Immunoprecipita-
tion and Immunofluorescence Analysis. Western blotting was performed as
previously described (33). The primary antibodies used were anti-HA (Co-
vance), anti-Cyclin A, anti-Cyclin D1, anti-Cyclin E and anti-EF1? (Upstate).
Co-immunoprecipitation and immunofluorescence analyses were performed
as described (34).
EMSA. Nuclear extracts were prepared from 2 to 5 ? 108cells and EMSA was
performed using consensus AP-1 probe (Santa Cruz Biotechnology) as de-
ChIP Assays. ChIP assays were performed using a commercially available kit
from Active Motif (35). PCR was performed using IL-8 promoter-specific
primers: sense, 5? ATGTCAGCTCTCGACGAAAATAGA 3?; and antisense, 5?
Generation of Anti-SARI Antibody. His-tagged recombinant SARI protein was
generated in Escherichia coli and purified using Ni-NTA metal chelate affinity
in the rabbits according to the manufacturer’s protocol (Invitrogen).
Mammalian Two-Hybrid Assay. Plasmid pCMV-AD was used for cloning the
c-jun cDNA to be expressed as a fusion with the activation domain of VP16
as a fusion with DNA-binding domain of GAL4. These constructs were trans-
fected along with a reporter vector that contains five GAL4 binding sites
(GAL4UAS) upstream of a minimal TATA box promoter driving the expression
of a luciferase reporter gene. Luciferase activity was monitored by a lumi-
nometer and normalized by ?-galactosidase activity.
Statistical Analysis. All of the experiments were performed at least three
times. The results are expressed as mean ? SD. Statistical comparisons were
made using an unpaired two-tailed Student t test. A P value ?0.05 was
considered as significant.
ACKNOWLEDGMENTS. We thank Dr. Tina Cirman for initial IFN studies with
SARI and Nichollaq Vozhilla for technical assistance with the animal studies.
The present study was supported in part by National Institutes of Health
Grants CA035675 and CA097318; the Samuel Waxman Cancer Research Foun-
dation (SWCRF); and by the National Foundation for Cancer Research. D.S. is
the Harrison Endowed Scholar in Cancer Research. P.B.F. holds the Thelma
Newmeyer Corman Chair in Cancer Research and is an SWCRF Investigator.
1. Borden EC, et al. (2007) Interferons at age 50: past, current and future impact on
biomedicine. Nat Rev Drug Discov 6:975–990.
interferons. Annu Rev Biochem 67:227–264.
3. Zhang SY, et al. (2007) Human Toll-like receptor-dependent induction of interferons
in protective immunity to viruses. Immunol Rev 220:225–236.
4. Takeuchi O, Akira S (2007) Recognition of viruses by innate immunity. Immunol Rev
5. Barber GN (2000) The interferons and cell death: guardians of the cell or accomplices
of apoptosis? Semin Cancer Biol 10:103–111.
6. Fisher PB, Grant S (1985) Effects of interferon on differentiation of normal and tumor
cells. Pharmacol Ther 27:143–166.
7. Fisher PB, et al. (1986) Effect of recombinant human fibroblast interferon and
mezerein on growth, differentiation, immune interferon binding and tumor associ-
ated antigen expression in human melanoma cells. Anticancer Res 6:765–774.
8. Fisher PB, Prignoli DR, Hermo H Jr, Weinstein IB, Pestka S (1985) Effects of combined
treatment with interferon and mezerein on melanogenesis and growth in human
melanoma cells. J Interferon Res 5:11–22.
human melanoma cells. Mol Cell Differ 2:221–239.
11. Jiang H, et al. (1995) The melanoma differentiation-associated gene mda-6, which
encodes the cyclin-dependent kinase inhibitor p21, is differentially expressed during
growth, differentiation and progression in human melanoma cells. Oncogene
12. Jiang H, et al. (1995) Cell cycle gene expression and E2F transcription factor complexes
in human melanoma cells induced to terminally differentiate. Oncogene 11:1179–
a novel melanoma differentiation associated gene, mda-7, modulated during human
melanoma differentiation, growth and progression Oncogene 11:2477–2486.
14. Jiang H, Su ZZ, Boyd J, Fisher PB (1993) Gene expression changes associated with
reversible growth suppression and the induction of terminal differentiation in human
melanoma cells. Mol Cell Differ 1:41–66.
15. Kang DC, et al. (2002) mda-5: An interferon-inducible putative RNA helicase with
double-stranded RNA-dependent ATPase activity and melanoma growth-suppressive
properties. Proc Natl Acad Sci USA 99:637–642.
16. Kang DC, LaFrance R, Su ZZ, Fisher PB (1998) Reciprocal subtraction differential RNA
display: an efficient and rapid procedure for isolating differentially expressed gene
sequences. Proc Natl Acad Sci USA 95:13788–13793.
17. Angel P, Karin M (1991) The role of Jun, Fos and the AP-1 complex in cell-proliferation
and transformation. Biochim Biophys Acta 1072:129–157.
18. Chinenov Y, Kerppola TK (2001) Close encounters of many kinds: Fos-Jun interactions
that mediate transcription regulatory specificity. Oncogene 20:2438–2452.
19. Maki Y, Bos TJ, Davis C, Starbuck M, Vogt PK (1987) Avian sarcoma virus 17 carries the
jun oncogene. Proc Natl Acad Sci USA 84:2848–2852.
20. Ryseck RP, Bravo R (1991) c-JUN, JUN B, and JUN D differ in their binding affinities to
AP-1 and CRE consensus sequences: effect of FOS proteins. Oncogene 6:533–542.
21. Vandel L, et al. (1996) Stepwise transformation of rat embryo fibroblasts: c-Jun, JunB,
or JunD can cooperate with Ras for focus formation, but a c-Jun-containing het-
erodimer is required for immortalization. Mol Cell Biol 16:1881–1888.
22. Johnson R, Spiegelman B, Hanahan D, Wisdom R (1996) Cellular transformation and
malignancy induced by ras require c-jun. Mol Cell Biol 16:4504–4511.
23. Eferl R, Wagner EF (2003) AP-1: a double-edged sword in tumorigenesis. Nat Rev
24. Boylan JF, Shih TY, Fisher PB, Zimmer SG (1992) Induction and progression of the
transformed phenotype in cloned rat embryo fibroblast cells: studies employing
type 5 adenovirus and wild-type and mutant Ha-ras oncogenes. Mol Carcinog
25. Lin L, et al. (2006) Activation of Ras/Raf protects cells from melanoma differentiation-
associated gene-5-induced apoptosis. Cell Death Differ 13:1982–1993.
26. Schubbert S, Shannon K, Bollag G (2007) Hyperactive Ras in developmental disorders
and cancer. Nat Rev Cancer 7:295–308.
27. Yeatman TJ (2004) A renaissance for SRC. Nat Rev Cancer 4:470–480.
28. Vinson C, Acharya A, Taparowsky EJ (2006) Deciphering B-ZIP transcription factor
interactions in vitro and in vivo. Biochim Biophys Acta 1759:4–12.
their roles in cell cycle regulation and differentiation. Blood Cells Mol Dis 32:155–167.
30. Landolfo S, Gariglio M, Gribaudo G, Lembo D (1998) The Ifi 200 genes: an emerging
family of IFN-inducible genes. Biochimie 80:721–728.
31. Lebedeva IV, et al. (2003) Bcl-2 and Bcl-x(L) differentially protect human prostate
cancer cells from induction of apoptosis by melanoma differentiation associated
gene-7, mda-7/IL-24 Oncogene 22:8758–8773.
32. Lebedeva IV, et al. (2002) The cancer growth suppressing gene mda-7 induces apo-
ptosis selectively in human melanoma cells. Oncogene 21:708–718.
33. Su Z, et al. (2001) PEA3 sites within the progression elevated gene-3 (PEG-3) promoter
and mitogen-activated protein kinase contribute to differential PEG-3 expression in
Ha-ras and v-raf oncogene transformed rat embryo cells. Nucleic Acids Res 29:1661–
34. Sarkar D, et al. (2008) Molecular basis of NF-kappaB activation by astrocyte elevated
gene-1 (AEG-1). Cancer Res 68:1478–1484.
35. Su ZZ, Sarkar D, Emdad L, Barral PM, Fisher PB (2007) Central role of interferon
regulatory factor-1 (IRF-1) in controlling retinoic acid inducible gene-I (RIG-I) expres-
sion. J Cell Physiol 213:502–510.
Su et al.
December 30, 2008 ?
vol. 105 ?
no. 52 ?