Cell Host & Microbe
The Double-Stranded RNA-Binding Protein PACT
Functions as a Cellular Activator of RIG-I
to Facilitate Innate Antiviral Response
Kin-Hang Kok,1Pak-Yin Lui,1Ming-Him James Ng,1Kam-Leung Siu,1Shannon Wing Ngor Au,2and Dong-Yan Jin1,*
1Department of Biochemistry and State Key Laboratory for Liver Research, Li Ka Shing Faculty of Medicine, The University of Hong Kong,
Pokfulam, Hong Kong
2Department of Biochemistry, Faculty of Science, Chinese University of Hong Kong, Shatin, New Territories, Hong Kong
RIG-I, a virus sensor that triggers innate antiviral
response, is a DExD/H box RNA helicase bearing
structural similarity with Dicer, an RNase III-type
nuclease that mediates RNA interference. Dicer
requires double-stranded RNA-binding protein part-
ners, such as PACT, for optimal activity. Here we
show that PACT physically binds to the C-terminal
repression domain of RIG-I and potently stimulates
RIG-I-induced type I interferon production. PACT
potentiates the activation of RIG-I by poly(I:C) of
intermediate length. PACT also cooperates with
RIG-I to sustain the activation of antiviral defense.
Depletion of PACT substantially attenuates viral
induction of interferons. The activation of RIG-I by
PACT does not require double-stranded RNA-
dependent protein kinase or Dicer, but is mediated
by a direct interaction that leads to stimulation of
its ATPase activity. Our findings reveal PACT as an
important component in initiating and sustaining
the RIG-I-dependent antiviral response.
Host cells sense invading viruses and mobilize innate immune
response to counteract their infection. Detection of viral nucleic
acids by the cytoplasmic sensor RIG-I generates an activation
signal which leads ultimately to the production of type I inter-
ferons (IFNs) that are important effectors in innate immunity
(Yoneyama et al., 2004). RIG-I discriminates between viral and
cellular RNAs by recognizing 50-triphosphates and base-paired
structures (Hornung et al., 2006; Pichlmair et al., 2006; Schlee
et al., 2009b; Schmidt et al., 2009). On the other hand, RIG-I
senses viral DNA in the cytosol (Choi et al., 2009) through
multiple mechanisms, including the recognition of 50-triphos-
phate RNA generated from DNA template by RNA polymerase
III (Chiu et al., 2009). However, some double-stranded RNAs
without 50-triphosphates can also activate RIG-I (Hausmann
et al., 2008; Schlee et al., 2009a). RIG-I is most efficiently acti-
vated by viruses, but the nature of viral RNA agonists bound to
endogenous RIG-I during the course of infection has not been
unequivocally determined (Baum et al., 2010; Rehwinkel et al.,
2010). RIG-I is activated potently by unattached polyubiquitin
chains (Zeng et al., 2010), but the physiological inducers of this
activation arenot understood.Italso remainstobeseen whether
the action of RIG-I might require additional cellular partners and
RIG-I is a DExD/H box RNA helicase bearing significant struc-
tural similarity with Dicer, an RNase III-type nuclease required for
RNA interference (RNAi). Human Dicer requires dsRNA-binding
protein partners TRBP and PACT for optimal activity in RNAi
(Chendrimada et al., 2005; Haase et al., 2005; Lee et al., 2006;
Kok et al., 2007). Surprisingly, a C. elegans protein named
p110, which is an unrecognized homolog of RIG-I, was found
to interact physically with PACT homolog RDE4 in an early study
(Tabara et al., 2002). In addition, before the identification of
RIG-I, PACT was also known to be capable of stimulating viral
induction of type I IFNs (Iwamura et al., 2001). These findings
prompted us to examine the interaction between human RIG-I
and PACT as well as the implications in innate antiviral response.
PACT Interacts with RIG-I
We employed two different approaches to investigate the inter-
tation experiments with lysates of transfected and untransfected
HEK293 cells. For transfected cells expressing Myc-PACT and
separately with anti-Myc and anti-Flag (Figure 1A, lanes 2 and 3
compared to lane 1), and the formation of a RIG-I-PACT protein
complex inside cells was confirmed. In contrast, Myc-PACT did
not form a complex with Flag-TRAF3 (Figure 1A, lane 3), lending
and RIG-I. Interestingly, only a fraction of RIG-I was seen to be
associated with PACT. Likewise, only a subset of PACT bound
to RIG-I. These results were compatible with the notion that the
two multifunctional proteins might form functional complexes
with other partners (Yoneyama et al., 2004; Kok et al., 2007).
When we repeated the coimmunoprecipitation experiment
using mock-transfected and mock-infected A549 cells, RIG-I
was not detected in either anti-PACT or anti-RIG-I precipitate
(Figure 1B, lanes 2 and 3), indicating that the amount of RIG-I
was too low in these cells. After induction of RIG-I expression
Cell Host & Microbe 9, 299–309, April 21, 2011 ª2011 Elsevier Inc. 299
by Sendai virus (SeV), RIG-I was found in the anti-PACT precipi-
tate (Figure 1B, lane 5), and reciprocally, PACT was detected in
the anti-RIG-I precipitate (Figure 1B, lane 6). Thus, endogenous
PACT forms a complex with endogenous RIG-I in virus-infected
Next, to rule out the possibility that the interaction between
RIG-I and PACT is adapted through another protein, we carried
out GST pull-down assay in vitro with purified recombinant
proteins. His-PACT and GST-RIG-I proteins were expressed in
the baculovirus-insect cell system and purified to homogeneity
(see Figure S1 available online). Retention of purified His-PACT
in the glutathione beads bound to purified GST-RIG-I (Figure 1C,
lane 2) indicated a direct interaction between the two entities.
Further analysis with truncated GST-RIG-I mutants revealed
that the C-terminal repression domain (CTD) of RIG-I was suffi-
cient for binding with PACT (Figure 1D, lane 10 compared to
lanes 7–9 and 11), whereas the CARD and helicase domains
were dispensable for this interaction. Nonspecific binding of
GST to PACT was also excluded in light of the inability of
GST-E to interact with PACT (Figure 1D, lanes 5 and 11).
Together, our results confirmed the interaction between RIG-I
and PACT both in cultured cells and in vitro.
PACT Is a Potent Activator of RIG-I
The interaction of PACT with the CTD of RIG-I might plausibly
affect the activation of RIG-I. With this in mind, we went on to
characterize the influence of PACT on RIG-I-induced activation
of IFN-b production in HEK293 cells in which the expression of
endogenous RIG-I was undetectable. We noted that enforced
expression of PACT alone had no influence on the activity of
IFN-b promoter (Figure 2A, bars 2–4). In contrast, coexpression
of RIG-I and PACT resulted in substantial enhancement of RIG-I-
induced activation of IFN-b promoter (Figure 2A, bars 6–8). The
helicase activity of RIG-I was seemingly required for the activa-
tion of RIG-I by PACT, since the helicase-dead K270A mutant
of RIG-I (Yoneyama et al., 2004) was not activated by PACT (Fig-
ure 2A, bars 10–12). Similar results were also obtained when
another reporter driven by the IRF3-binding enhancer elements
alone was used (Figure 2B), suggesting that the stimulatory
effect was mediated through activation of IRF3. In further
support of this, the formation of IRF3 dimer, which is the active
form of IRF3, was observed in HEK293 cells expressing both
RIG-I and PACT (Figure 2C, lanes 2 and 3 compared to lanes
1, 4, 5, and 8). Interestingly, IRF3 dimer was also detected
when RIG-I was induced by IFN-b in PACT-expressing
HEK293 cells (Figure 2C, lane 7 compared to lane 6). The stimu-
lation of RIG-I-dependent activation of IRF3 by PACT was highly
specific, since PACT did not potentiate RIG-I-induced activation
of NF-kB (Figure S2A, bars 9–12 compared to bars 1–4 and 5–8).
Notably, we also observed a similar stimulatory effect of PACT
on MDA5- mediated activation of IFN-b promoter and IRF3 (Fig-
ure S2B, bars 1–4 and 5–8), but not on MDA5-inudced activation
of NF-kB (bars 9–12). Although further experiments are required
tofully characterize theimpact ofPACT onMDA5,ourresultsdid
suggest that PACT might target the conserved CTD in RIG-I and
MDA5. Consistent with the binding of PACT to the CTD of RIG-I
(Figure 1C), PACT had no effect on RIG-IN containing the
N-terminal CARD domain alone (Figure S2C, bars 5–8 compared
to bars 9–12). Finally, facilitation of RIG-I-activated IFN produc-
tion by PACT was also shown by direct measurement of the anti-
viral activity of IFNs. Only in HEK293 cells expressing both RIG-I
icantinhibition (?104-fold lessvirus) of subsequent infection bya
green fluorescent protein (GFP)-expressing vesicular stomatitis
Figure 1. PACT Directly Interacts with RIG-I
(A) RIG-I and PACT were coimmunoprecipitated from lysates of transfected
HEK293T cells. Cells were transfected with the indicated combinations of
Myc-PACT, Flag-RIG-I, and Flag-TRAF3 expression plasmids. Reciprocal
immunoprecipitation (IP) and western blotting (WB) were performed with the
(B) RIG-I and PACT were coimmunoprecipitated from lysates of untransfected
A459 cells. Cells were mock infected or infected with 80 HA units of SeV for
24 hr. Immunoprecipitation was performed with the indicated antibodies.
Endogenous PACT and RIG-I in the precipitates were probed with anti-PACT
(C) GST pull-down assay. GST-RIG-I and V5/His-PACT proteins purified to
homogeneity were incubated for 1 hr and bound to glutathione beads. Bound
proteins were analyzed with anti-His (upper panel) and anti-GST (lower panel).
(D) PACT binds to CTD of RIG-I. GST pull-down assay was performed with
PACT and truncated mutants of RIG-I. The upper diagram depicts the domain
organization of truncated mutants. Results are representative of three inde-
Cell Host & Microbe
PACT Is an Activator of RIG-I in Antiviral Response
300 Cell Host & Microbe 9, 299–309, April 21, 2011 ª2011 Elsevier Inc.
Satoh, T., Kato, H., Kumagai, Y., Yoneyama, M., Sato, S., Matsushita, K.,
Tsujimura, T., Fujita, T., Akira, S., and Takeuchi, O. (2010). LGP2 is a positive
regulator of RIG-I- and MDA5-mediated antiviral responses. Proc. Natl. Acad.
Sci. USA 107, 1512–1517.
Schlee, M., Hartmann, E., Coch, C., Wimmenauer, V., Janke, M., Barchet, W.,
and Hartmann, G. (2009a). Approaching the RNA ligand for RIG-I? Immunol.
Rev. 227, 66–74.
Schlee, M., Roth, A., Hornung, V., Hagmann, C.A., Wimmenauer, V., Barchet,
W., Coch, C., Janke, M., Mihailovic, A., Wardle, G., et al. (2009b). Recognition
of 50triphosphate by RIG-I helicase requires short blunt double-stranded RNA
as contained in panhandle of negative-strand virus. Immunity 31, 25–34.
Schmidt, A., Schwerd, T., Hamm, W., Hellmuth, J.C., Cui, S., Wenzel, M.,
Hoffmann, F.S., Michallet, M.-C., Besch, R., Hopfner, K.-P., et al. (2009).
50-triphosphate RNA requires base-paired structures to activate antiviral
signaling via RIG-I. Proc. Natl. Acad. Sci. USA 106, 12067–12072.
Schro ¨der, M., Baran, M., and Bowie, A.G. (2008). Viral targeting of DEAD box
protein 3 reveals its role in TBK1/IKK3-mediated IRF activation. EMBO J. 27,
Schulz, O., Pichlmair, A., Rehwinkel, J., Rogers, N.C., Scheuner, D., Kato, H.,
Takeuchi, O., Akira, S., Kaufman, R.J., and Reis e Sousa, C. (2010). Protein
kinase R contributes to immunity against specific viruses by regulating inter-
feron mRNA integrity. Cell Host Microbe 7, 354–361.
Siu, K.L., Kok, K.H., Ng, M.H.J., Poon, V.K.M., Yuen, K.Y., Zheng, B.J., and
Jin, D.Y. (2009). Severe acute respiratory syndrome coronavirus M protein
TRAF3$TANK$TBK1/IKK3 complex. J. Biol. Chem. 284, 16202–16209.
type I interferonproduction by impeding the formation of
Tabara, H., Yigit, E., Siomi, H., and Mello, C.C. (2002). The dsRNA binding
protein RDE-4 interacts with RDE-1, DCR-1, and a DExH-box helicase to
direct RNAi in C. elegans. Cell 109, 861–871.
Takahasi, K., Yoneyama, M., Nishihori, T., Hirai, R., Kumeta, H., Narita, R.,
Gale, M., Jr., Inagaki, F., and Fujita, T. (2008). Nonself RNA-sensing mecha-
nism of RIG-I helicase and activation of antiviral immune responses. Mol.
Cell 29, 428–440.
Yoneyama, M., Kikuchi, M., Natsukawa, T., Shinobu, N., Imaizumi, T.,
Miyagishi, M., Taira, K., Akira, S., and Fujita, T. (2004). The RNA helicase
RIG-I has an essential function in double-stranded RNA-induced innate anti-
viral responses. Nat. Immunol. 5, 730–737.
Zeng, W., Sun, L., Jiang, X., Chen, X., Hou, F., Adhikari, A., Xu, M., and Chen,
Z.J. (2010). Reconstitution of the RIG-I pathway reveals a signaling role of
unanchored polyubiquitin chains in innate immunity. Cell 141, 315–330.
Zhang, B., Li, M., Chen, L., Yang, K., Shan, Y., Zhu, L., Sun, S., Li, L., and
Wang, C. (2009). The TAK1-JNK cascade is required for IRF3 function in the
innate immune response. Cell Res. 19, 412–428.
Zhou, R., Czech, B., Brennecke, J., Sachidanandam, R., Wohlschlegel, J.A.,
Perrimon, N.,and Hannon, G.J.
endo-siRNAs depends on aspecific Loquacious isoform. RNA 15,1886–1895.
(2009). Processingof Drosophila
Cell Host & Microbe
PACT Is an Activator of RIG-I in Antiviral Response
Cell Host & Microbe 9, 299–309, April 21, 2011 ª2011 Elsevier Inc. 309