A zinc finger HIT domain-containing protein, ZNHIT-1,
interacts with orphan nuclear hormone receptor Rev-erbb
and removes Rev-erbb-induced inhibition of apoCIII
transcription
Jiadong Wang
1
, Yang Li
1
, Min Zhang
1
, Zhongle Liu
1
, Cong Wu
1
, Hanying Yuan
1
, Yu-Yang Li
1
,
Xin Zhao
2
and Hong Lu
1
1 State Key Laboratory of Genetic Engineering, School of Life Sciences, Fudan University, Shanghai, China
2 Department of Animal Science, McGill University, Montreal, Canada
The nuclear receptors (NRs) are a family of transcrip-
tion factors which regulate a wide array of biological
processes, including those involved in diabetes, obesity,
cardiovascular disease and cancer [1]. Many important
medicines, including 12 of the top 100 selling drugs,
target NRs [2]. Generally, NRs regulate the activity of

transcription by binding to specific ligands and coregu-
lators [3]. However, the regulation of transcriptional
activity by orphan nuclear hormone receptors, whose
ligands are not identified or are lost during evolution,
is less clear.
The Rev-erb family is a subgroup of orphan recep-
tors, two members of which have been isolated from
mammalian genotypes: Rev-erba, also known as Ear-1
or NR1D1 (official nomenclature), and Rev-erbb, also
known as RVR, BD73 or NR1D2 (official nomencla-
ture) [4]. Rev-erba and Rev-erbb are unique NRs that
lack the activation function 2 domain required for
Keywords
apoCIII; Rev-erbb; transcriptional regulation;
zinc finger HIT domain-containing protein 1;
ZNHIT-1
Correspondence
H. Lu, State Key Laboratory of Genetic
Engineering, School of Life Sciences, Fudan
University, Shanghai 200433, China
Fax: +86 21 65642505
Tel: +86 21 65642505
E-mail:
(Received 27 June 2007, revised 15 August
2007, accepted 22 August 2007)
doi:10.1111/j.1742-4658.2007.06062.x
The orphan receptors, Rev-erba and Rev-erbb, are members of the
nuclear receptor superfamily and specifically repress apolipoprotein CIII
(apoCIII) gene expression in rats and humans. Moreover, Rev-erba null
mutant mice have elevated very low density lipoprotein triacylglycerol

and apoCIII levels. However, ligands for Rev-erb are unknown and the
regulatory mechanism of Rev-erb is poorly understood. Conceivably, co-
factors for Rev-erb may play an important role in the regulation of lipid
metabolism. In this study, a zinc finger HIT domain-containing protein,
ZNHIT-1, interacted with Rev-erbb. ZNHIT-1 was found to be a con-
served protein in eukaryotes and was highly abundant in human liver.
Furthermore, ZNHIT-1 was identified as a nuclear protein. Serial trun-
cated fragments and substitution mutations established a putative nuclear
localization signal at amino acids 38–47 of ZNHIT-1. A putative ligand-
binding domain of Rev-erbb and the FxxLL motif of ZNHIT-1 were
required for their interaction. Finally, ZNHIT-1 was recruited by Rev-
erbb to the apoCIII promoter and removed the Rev-erbb-induced inhibi-
tion of apoCIII transcription. These findings demonstrate that ZNHIT-1
functions as a cofactor to regulate the activity of Rev-erbb, and may play
a role in lipid metabolism.
Abbreviations
ChIP, chromatin immunoprecipitation; Co-IP, coimmunoprecipitation; GFP, green fluorescent protein; GST, glutathione S-transferase; LBD,
ligand-binding domain; NCBI, National Center for Biotechnology Information; NLS, nuclear localization signal; NR, nuclear receptor; RFP, red
fluorescent protein; SRCAP, SNF-2 related CBP activator protein; VLDL, very low density lipoprotein; ZNHIT-1, zinc finger HIT domain-
containing protein 1.
5370 FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS
ligand-dependent activation of transcription by other
members of the NR superfamily [5]. Therefore, Rev-
erb receptors constitutively behave as unliganded
receptors and repress transcription by binding to core-
pressor molecules [6]. Both Rev-erba and Rev-erbb are
widely expressed [4]. Rev-erb represses transcription of
the Rev-erba gene itself [7], as well as N-myc [8], Bmal1
[9], enoyl-CoA hydratase ⁄ 3-hydroxyacyl-CoA dehydro-
genase bifunctional enzyme [10], a-fetoprotein [11] and

rat apoAI [12] genes. Rev-erba is also involved in cir-
cadian timing in brain and liver tissue, and regulates
Bmal1 which is the master regulator of circadian
rhythm in mammals [13–16]. However, it is not known
whether Rev-erbb is involved in circadian regulation.
Rev-erb also represses the expression of apoCIII
[17]. ApoCIII is a 79-residue glycoprotein. It is synthe-
sized in the liver as part of the very low density lipo-
protein (VLDL). It is well established that the plasma
concentration and synthesis rate of apoCIII are posi-
tively correlated with plasma triacylglycerols in both
normal and hypertriglyceridaemic subjects [18–20].
Large scale clinical trials have indicated that hyper-
triglyceridaemia is an independent risk factor for
coronary artery disease and atherosclerosis [21,22].
Therefore, a better understanding of the regulation of
the expression of the apoCIII gene is of major impor-
tance for the treatment of dyslipidaemia.
In this study, a zinc finger HIT domain-containing
protein, ZNHIT-1 (NM_006349.2), was found to
interact with Rev-erbb by a yeast two-hybrid assay.
The sequence of ZNHIT-1 was first identified by high
throughput genomic sequences in humans. This pro-
tein contains a conserved zinc finger HIT domain
originally found in the yeast HIT-1 protein [23], and
so has been named ZNHIT-1 (‘zinc finger HIT
domain-containing protein 1’). Previous knowledge
about the function of ZNHIT-1 is limited. One report
has described ZNHIT-1 as a subunit of the SNF-2
related CBP activator protein (SRCAP) complex,

which can remodel chromatin by incorporating the
histone variant H2A.Z into nucleosomes [24]. The
interaction between ZNHIT-1 and Rev-erbb was
further confirmed using a glutathione S-transferase
(GST) pull-down assay and a coimmunoprecipitation
(Co-IP) assay. In addition, the homologues and tissue
distribution of ZNHIT-1 were analysed. Furthermore,
a nuclear localization signal (NLS) of ZNHIT-1 was
identified. Using the chromatin immunoprecipitation
(ChIP) assay, we found that ZNHIT-1 was recruited
to the human apoCIII promoter by Rev-erbb, and
subsequently removed Rev-erbb-induced inhibition of
apoCIII transcription without changing the DNA-
binding activity of Rev-erbb.
Results
Yeast two-hybrid screening
In order to identify Rev-erbb binding partners, a human
fetal liver cDNA library was screened in a yeast two-
hybrid assay using Rev-erbb as bait. A screen of
approximately 10
6
yeast transformants revealed that
ZNHIT-1 interacted with Rev-erbb. The zinc finger
HIT domain is a motif found in many proteins, and
plays an important role in gene regulation and chro-
matin remodelling [25]. So far, five human proteins
containing the HIT domain have been identified. A
phylogenetic tree was constructed to show the relation-
ship between ZNHIT-1 and other members of the zinc
finger HIT domain proteins. According to the evolu-

tionary distance between the ZNHIT protein members,
estimated from our phylogenetic analysis, it was
concluded that a close relationship existed between
ZNHIT-1 and ZNHIT-4, as well as between ZNHIT-3
and ZNHIT-5 (see Supplementary material Fig. S1).
A liquid assay for b-galactosidase activity was further
performed to establish the interaction between Rev-
erbb and ZNHIT-1. As shown in Fig. 1, the Rev-erbb
homodimer was used as a positive control. ZNHIT-1
and Rev-erbb cotransformants showed a stronger
b-galactosidase activity than the Rev-erbb homodimer,
whereas almost no b-galactosidase activity was detected
in ‘no insert’ or lamin-negative control vectors.
Homologues and tissue distribution analysis of
human ZNHIT-1
blastp of human ZNHIT-1 was performed at National
Center for Biotechnology Information (NCBI) sites,
Fig. 1. Evaluation of the interaction strength by a relative b-galacto-
sidase activity assay. The homodimer of Rev-erbb was used as a
positive control. No insert or lamin was used as a negative control.
Results are the means ± SD of three independent experiments
performed in triplicate. ‘BD’ and ‘AD’ represent the DNA-binding
domain in pGBKT-7 and the DNA activation domain in pADKT-7,
respectively. These domains were fused with target proteins.
J. Wang et al. ZNHIT-1 as a cofactor of Rev-erbb
FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS 5371
and homologues were identified in dog, mouse, frog,
zebra fish, fruit fly and yeast. As shown in Fig. 2A,
human ZNHIT-1 encodes a full-length protein of 154
amino acids and shares high identity with its homo-

logues (100% with Canis familiaris, 97% with Mus
musculus, 87% with Xenopus tropicalis, 87% with
Danio rerio, 68% with Drosophila melanogaster and
33% with Schizosaccharomyces pombe), indicating that
ZNHIT-1 is a conservative protein in eukaryotes.
To investigate the tissue distribution of ZNHIT-1, a
real-time quantitative PCR was performed with human
multiple tissue cDNA panels 1 and 2 from 16 human
tissues (Catalogue # 636742, 636752; Clontech, Tokyo,
Japan) as templates. ZNHIT-1 was expressed in all
analysed tissues, abundantly in human liver, but weakly
in skeletal muscle, ovary and small intestine (Fig. 2B).
Subcellular localization analysis of ZNHIT-1
Bioinformatic analyses by PSORT and other databases
failed to predict the potential localization of ZNHIT-1.
To examine the subcellular localization of ZNHIT-1
proteins, Hela and HepG2 cells were transfected with
expression vectors for pEGFPC1-ZNHIT-1 or pDS-
RED1C1-ZNHIT-1. Their subcellular localization was
visualized by laser microscopy. Similar results were
obtained from both types of cell. As shown in
Fig. 3A,B for Hela cells, both red fluorescent protein
(RFP)-ZNHIT-1 and green fluorescent protein (GFP)-
ZNHIT-1 accumulated in the nucleus and exhibited a
punctate distribution.
To map the ZNHIT-1 NLS, serial deletion mutants
of ZNHIT-1 were constructed (Fig. 3E). Like the full-
sized protein, the ZNHIT-1 (1–54) mutant was local-
ized exclusively in the nucleus of Hela cells. In con-
trast, the ZNHIT-1 (54–154) mutant, without the

NH
2
-terminus, was localized in both the nucleus and
cytoplasm, similar to the empty GFP vector. Further
deletions in ZNHIT-1 (1–54) showed that amino acids
38–47 were sufficient for the nuclear localization
of ZNHIT-1 (Fig. 3C). Alignment, even with yeast
ZNHIT-1, revealed that four amino acids
A
B
Fig. 2. Homologue and expression analysis of human ZNHIT-1. (A) Sequence comparison of human ZNHIT-1 and its homologues in Canis
familiaris, Mus musculus, Xenopus tropicalis, Danio rerio, Drosophila melanogaster and Schizosaccharomyces pombe (GenBank accession
nos. XP_536855.2, Q8R331, NP_001017056.2, NP_001017401.1, NP_608895.1 and NP_595833.1). Residues identical in all compared
sequences are presented on a black background, whereas residues similar in four or more compared sequences are presented on a grey
background. (B) The relative expression levels of ZNHIT-1 mRNA in 16 human tissues (human multiple tissue cDNA panels 1 and 2 from
Clontech) determined by real-time quantitative PCR using b
2
-microglobulin as a reference standard.
ZNHIT-1 as a cofactor of Rev-erbb J. Wang et al.
5372 FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS
(38DNFxD42) of ZNHIT-1 were highly conserved.
Substitution mutations of these conserved amino acids
decreased the nuclear localization of ZNHIT-1
(Fig. 3D). Collectively, the data in Fig. 3 established
that the fragment 38DNFQDDPHAG47 of ZNHIT-1
contains a putative NLS, which is sufficient to target
ZNHIT-1 into the nucleus.
Characterization of the interaction between
ZNHIT-1 and Rev-erb
The interaction between Rev-erbb and ZNHIT-1 was

further verified by a GST pull-down assay and a
Co-IP assay. As shown in Fig. 4A, bacterially
expressed GST-ZNHIT-1, when conjugated to gluta-
thione–Sepharose beads, efficiently and specifically
pulled down 6His-tagged Rev-erbb, which was about
66 kDa. Conversely, GST alone did not. In the Co-IP
assay, anti-Myc serum and protein A⁄ G agarose
were added to the HepG2 cell lysates containing
overexpressed Myc-ZNHIT-1 so as to precipitate the
Myc-ZNHIT-1 ⁄ Rev-erbb complex, and anti-Rev-erbb
serum was used in western blotting to detect Rev-erbb
in the precipitate. As shown in Fig. 4B, endogenous
Rev-erbb interacted with ZNHIT-1 in mammalian
cells.
It is well known that Rev-erba is closely related to
Rev-erbb, especially in the DNA-binding domain and
putative ligand-binding domain (LBD). In order to
A
B
C
D
E
Fig. 3. Subcellular localization analysis of ZNHIT-1 in Hela cells.
(A) Nuclear localization of RFP-ZNHIT-1. (B) Nuclear localization of
GFP-ZNHIT-1. (C) Nuclear localization of GFP-ZNHIT-1 (38–47).
(D) Cytoplasmic and nuclear localization of GFP-ZNHIT-1 NLS
mutant. (E) Subcellular localization of various ZNHIT-1 deletion
mutants.
A
B

C
Fig. 4. Validation of the interaction between ZNHIT-1 and Rev-
erbb or Rev-erba. (A) The GST pull-down assay in vitro.
GST-ZNHIT-1 or GST immobilized on the beads was incubated
with 6His-Rev-erbb. Interacting proteins were immunoblotted with
anti-6His serum. Lane 2 is the input, which acts as a positive
control. (B) The interaction between ZNHIT-1 and Rev-erbb vali-
dated by a Co-IP assay. HepG2 cells were transfected with
pCMV-Myc-ZNHIT-1. Expression of Myc-ZNHIT-1 in transfected
cells was analysed by western blotting. Cell lysates were then
precipitated by anti-Myc serum. The precipitated proteins were
eluted from the protein A ⁄ G PLUS agarose and analysed by
western blotting using anti-Rev-erbb serum. HC represents the
heavy chain of mouse IgG. (C) The interaction between ZNHIT-1
and Rev-erba validated by a Co-IP assay. HepG2 cells were
cotransfected with Myc-ZNHIT-1 and pCMV-HA-Rev-erba.LC
represents the light chain of mouse IgG.
J. Wang et al. ZNHIT-1 as a cofactor of Rev-erbb
FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS 5373
determine whether Rev-erba could also interact with
ZNHIT-1, Rev-erba was amplified from a human fetal
liver cDNA library, and a Co-IP assay between Rev-
erba and ZNHIT-1 was performed. As shown in
Fig. 4C, Rev-erba was also capable of binding
ZNHIT-1.
Mapping the regions of interaction between
ZNHIT-1 and Rev-erbb
Rev-erbb consists of five domains: A ⁄ B, C, D and E
[5]. The highly conserved region C is responsible for
DNA binding, and region E contains a putative LBD

and mediates the recruitment of cofactors [5]. To iden-
tify the region of Rev-erbb essential for the interaction
with ZNHIT-1, serial deletion assays were performed
using the yeast two-hybrid method. N- and C-terminal
deletion constructs of Rev-erbb (Fig. 5A) were fused
in-frame to the Gal4-binding domain and tested for
their abilities to bind ZNHIT-1 by checking the activa-
tion of the reporter genes Ade2, His3 and LacZ. It
was shown that ZNHIT-1 binds specifically to the
LBD of Rev-erbb. The interaction between ZNHIT-1
and all the serial truncated fragments of Rev-erbb was
further confirmed by the GST pull-down in vitro assay
(Fig. 5B). Similarly, the serial deletion constructs of
ZNHIT-1 were fused to the Gal4 activation domain
and tested for their abilities to bind Rev-erbb.As
shown in Fig. 5C, ZNHIT-1D2 (amino acids 72–110)
was required for the interaction with Rev-erbb. This
fragment contains an FxxLL motif (x denotes any
amino acid), which has been reported to function in
the same manner as an LxxLL motif, and mediates
transcriptional coactivator binding to NRs [26–28]. To
further investigate whether the ZNHIT-1 FxxLL motif
was required for the interaction with Rev-erbb, a dou-
ble L87A ⁄ L88A mutation in the context of full-length
ZNHIT-1 was introduced. As a result, the mutant of
the FxxLL motif failed to interact with Rev-erbb. The
results from the GST pull-down assay (Fig. 5D) were
consistent with those from the yeast two-hybrid assay.
A
B

C
D
E
Fig. 5. Mapping the regions for the interaction between Rev-erbb
and ZNHIT-1. (A) The Rev-erbb region required for interaction with
ZNHIT-1 was revealed by a yeast two-hybrid assay. The serial trun-
cated fragments of Rev-erbb were separately coexpressed with
full-length ZNHIT-1 in AH109. (B) Serial truncated fragments of
Rev-erbb interacting with ZNHIT-1 were confirmed by a pull-down
assay. (C) The ZNHIT-1 region required for the interaction with Rev-
erbb was mapped by a yeast two-hybrid assay. The serial truncated
fragments of ZNHIT-1 were individually coexpressed with full-length
Rev-erbb in AH109. The activation of the reporters was analysed.
(D) All serial truncated fragments of ZNHIT-1 interacting with Rev-
erbb were confirmed by a pull-down assay. (E) The interaction
between Rev-erbb, ZNHIT-1 and N-CoR. GST-ZNHIT-1 was first
immobilized on glutathione–Sepharose and was then incubated
with 6His-Rev-erbb together with a myc-tagged N-CoR fragment
(amino acids 2053–2453 containing IDI + IDII domains). The mix-
ture was incubated at 4 °C for 6 h. After three washes with
NaCl ⁄ P
i
buffer, the sample was boiled and analysed by western
blotting using anti-6His or anti-myc.
ZNHIT-1 as a cofactor of Rev-erbb J. Wang et al.
5374 FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS
The transcriptional repressor, N-CoR, has been
reported to interact with Rev-erbb through the E
region [5,29]. Deletion of one of two fragments (amino
acids 394–416 and 561–576) in the E region of Rev-

erbb ablates N-CoR ⁄ Rev-erbb interaction [29]. In
order to identify whether the region mediating the
Rev-erbb ⁄ N-CoR interaction is the same as that medi-
ating the Rev-erbb ⁄ ZNHIT-1 interaction, the amino
acid fragment 416–561 (Rev-erbbD6) was tested for its
ability to bind to ZNHIT-1. It was shown that Rev-
erbbD6 interacted with ZNHIT-1 (lane 2 in the right
panel of Fig. 5E), suggesting that the regions mediat-
ing the Rev-erbb ⁄ N-CoR and Rev-erbb ⁄ ZNHIT-1
interactions are different. As ZNHIT-1 and N-CoR
bind to different regions of the E domain of Rev-erbb,
the possibility that these three proteins could form a
ternary complex exists. To answer this question, GST-
ZNHIT-1 was affinity immobilized on glutathione–
Sepharose and incubated with 6His-Rev-erbb and a
myc-tagged N-CoR fragment (amino acids 2053–2453
containing IDI + IDII domains). As shown in
Fig. 5E, ZNHIT-1 could pull-down 6His-Rev-erbb
directly, but not the myc-tagged N-CoR fragment, in
the presence of Rev-erbb. These results suggest that
these three proteins could not form a ternary complex
in vitro.
Colocalization of ZNHIT-1 and Rev-erbb in the
nucleus
To determine whether the interaction between ZNHIT-1
and Rev-erbb might be of physiological relevance in
mammalian cells, the subcellular localization and dis-
tribution of ZNHIT-1 and Rev-erbb in Hela and
HepG2 cells were assessed using laser microscopy.
Similar results were obtained from both types of cell.

GFP-Rev-erbb accumulated in the nucleus and exhib-
ited a diffuse distribution (Fig. 6A). As shown in
Fig. 6B, confocal images of cells expressing wild-type
RFP-ZNHIT-1 and GFP-Rev-erbb indicated that both
proteins were nuclear and that the majority of these
expressed proteins were colocalized in Hela cells. To
extend this observation, coexpression studies with Rev-
erbb and the ZNHIT-1 NLS mutant were performed.
As illustrated in Fig. 3D, the ZNHIT-1 NLS mutant
was expressed in both the cytoplasm and nucleus in
Hela cells without transfection of Rev-erbb. Surpris-
ingly, in cells with coexpressed ZNHIT-1 mutant and
RFP-Rev-erbb, the ZNHIT-1 NLS mutant adopted a
predominantly nuclear expression profile and colocal-
ized with Rev-erbb (Fig. 6C). Thus, these data provide
additional evidence that Rev-erbb interacts with
ZNHIT-1 in the nucleus, and implies that this interac-
tion with Rev-erbb recruits the cytoplasmic ZNHIT-1
NLS mutant into the nuclear compartment.
Recruitment of ZNHIT-1 by Rev-erbb to the
apoCIII promoter and its effect on the apoCIII
promoter
Rev-erbb has been reported to repress the expression
of apoCIII [17]. To extend this observation, Rev-erb
b
was found to bind to the promoter of apoCIII in vivo
(Fig. 7A) and to function as a transcriptional silencer
(Fig. 7B). To study the effect of the interaction
between ZNHIT-1 and Rev-erbb on the Rev-erbb-
mediated transcription of apoCIII, recruitment of

ZNHIT-1 to the apoCIII promoter was first studied.
No ZNHIT-1 was detectable in association with the
apoCIII promoter in the absence of transfected Rev-
erbb. However, a relatively strong association of
ZNHIT-1 with the apoCIII promoter was detected
after transfection of Rev-erbb. These results suggest
that Rev-erbb is essential to recruit ZNHIT-1 to the
apoCIII promoter (Fig. 7A). In order to further
address the functional significance of ZNHIT-1-medi-
ated regulation on Rev-erbb, HepG2 cells expressing
endogenous ApoCIII were transiently transfected with
HA-Rev-erbb and either Myc-ZNHIT-1 or empty
vector, together with the pGL3-apoCIII-pro reporter,
which contains the promoter of the human apoCIII
gene (nucleotides )1408 to +24) and pRL. As illus-
A
B
C
Fig. 6. Colocalization of ZNHIT-1 and Rev-erbb in Hela cells. (A)
Nuclear localization of GFP-Rev-erbb. (B) Colocalization of RFP-
ZNHIT-1 and GFP-Rev-erbb. (C) Effect of Rev-erbb on the localiza-
tion of the ZNHIT-1 NLS mutant. Hela cells were cotransfected
with RFP-Rev-erbb and GFP-ZNHIT-1 NLS mutant.
J. Wang et al. ZNHIT-1 as a cofactor of Rev-erbb
FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS 5375
trated in Fig. 7B, overexpression of Rev-erbb repressed
the expression of apoCIII by approximately 60%, and
the repression was totally removed by coexpressed
ZNHIT-1. In contrast, the L87A ⁄ L88A mutant of
ZNHIT-1 did not relieve the repression, suggesting

that ZNHIT-1-mediated regulation is dependent on
the binding of Rev-erbb. Furthermore, ZNHIT-1 D4,
without the zinc finger HIT domain, did not remove
the inhibition of apoCIII (Fig. 7B), indicative of the
requirement of the HIT domain in ZNHIT-1-mediated
coregulation.
To further evaluate the effect of ZNHIT-1 on apoC-
III expression in vivo, HepG2 cell lines with stable
ZNHIT-1 siRNA expression were established. These
HepG2 RNAi cell lines showed a significant decrease
in mRNA expression of the endogenous ZNHIT-1
genes, as measured by quantitative real-time PCR. The
apoCIII level was also measured by quantitative real-
time PCR. As shown in Fig. 7C, the endogenous
apoCIII level was reduced to about 70% in the HepG2
cell line with stable ZNHIT-1 siRNA expression, sug-
gesting that endogenous ZNHIT-1 affects apoCIII
transcription.
No change in Rev-erbb DNA-binding activity to
the apoCIII promoter by ZNHIT-1
Next, the possible effect of ZNHIT-1 on the binding
capacity of Rev-erbb to the apoCIII promoter was
investigated. ChIP analysis using anti-Rev-erbb serum
was performed in HepG2 cells transfected with
Rev-erbb alone or cotransfected with Rev-erbb and
ZNHIT-1. The binding capacity of Rev-erbb to the
apoCIII promoter was assessed by quantitative PCR
for ChIP DNA. As shown in Fig. 8, ZNHIT-1 did not
affect Rev-erbb binding to the apoCIII promoter.
These results indicate that Rev-erbb remains on the

apoCIII promoter after recruitment of ZNHIT-1.
Discussion
It is well known that Rev-erb is involved in lipid
metabolism by the regulation of apoCIII transcription.
A
B
C
Fig. 7. Recruitment of ZNHIT-1 by Rev-erbb to the apoCIII promoter
and its effect on the apoCIII promoter. (A) Recruitment of ZNHIT-1
by Rev-erbb to the apoCIII promoter. A ChIP assay of the apoCIII
promoter was performed in human HepG2 cells with the indicated
antibodies. Promoter-specific primers are described in ‘Experimen-
tal procedures’. (B) The effect of ZNHIT-1 on Rev-erbb activity was
assessed by a luciferase assay. HepG2 cells were cotransfected
with pGL3-apoCIII-pro and pRL, as well as the plasmid or plasmids
as indicated. Plasmid pRL was used to normalize the transfection
efficiencies, and the empty vector pCMV-HA was used as a nega-
tive control. All luciferase activities are the means ± SD of three
independent experiments performed in triplicate. (C) The effect of
ZNHIT-1 on endogenous apoCIII expression was analysed by real-
time PCR with total RNA from the stable knock-down of ZNHIT-1
cells. The mRNA levels of ZNHIT-1 and apoCIII were normalized to
the endogenous b
2
-microglobulin mRNA level.
ZNHIT-1 as a cofactor of Rev-erbb J. Wang et al.
5376 FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS
The ApoCIII protein is a major component of VLDL
and plays a key role in hypertriglyceridaemia [18–20].
Human Rev-erba and Rev-erbb specifically repress

apoCIII gene expression in rats and humans [17,30].
Rev-erba null mutant mice have elevated VLDL tri-
acylglycerol and ApoCIII [31]. Cofactors that interact
with Rev-erb and modify its effect on transcription will
no doubt be involved in the regulation of its target
genes. One corepressor, N-CoR, has been reported to
interact with Rev-erba and Rev-erbb and, conse-
quently, to intensify the transcriptional repression
[6,9]. However, no study has reported on how the tran-
scriptional repression mediated by Rev-erba or Rev-
erbb can be removed. In this study, it has been shown
that ZNHIT-1 can interact with Rev-erbb and relieve
its inhibitory effect on the transcription of apoCIII.
Interestingly, ZNHIT-1 is highly abundant in the liver
(Fig. 2B) where ApoCIII is synthesized. Thus, it is
plausible that ZNHIT-1 can affect lipid metabolism.
However, further experimentation will be needed to
confirm this assumption.
ZNHIT-1 was identified as a nuclear protein. Serial
truncated fragments and substitution mutations estab-
lished a putative NLS in amino acids 38–47 of
ZNHIT-1. Furthermore, the ZNHIT-1 NLS mutant
was expressed in both the cytoplasm and nucleus.
Interestingly, in cells that coexpressed the ZNHIT-1
NLS mutant together with Rev-erbb, the ZNHIT-1
NLS mutant adopted a predominantly nuclear expres-
sion profile. This shuttling of the ZNHIT-1 NLS
mutant to the nucleus was selective for Rev-erbb
expression, as the expression of another nuclear pro-
tein, such as p53, did not alter the cytoplasmic locali-

zation of the ZNHIT-1 NLS mutant (data not shown).
Thus, these data provide evidence that the interaction
with Rev-erbb recruits the cytoplasmic ZNHIT-1 NLS
mutant to the nuclear compartment.
Many cofactors bind to a common LBD region of
NRs to regulate transcription. These cofactors usually
contain a short conserved LxxLL motif, which was
originally identified in the NR-interacting domain of
transcriptional intermediary factor 1a [26], and subse-
quently found in other putative NR coactivators [27,28].
Furthermore, LxxLL appears to be both necessary and
sufficient for such interactions [26]. However, some
NR-binding proteins, such as NR-binding SET domain-
containing protein, also contain a variant motif FxxLL
that is responsible for ligand-dependent binding of NRs
[32]. In ZNHIT-1, a segment extending from residue 72
to residue 110 is required for interaction with Rev-erbb,
and is predicted to form an a-helix. This segment con-
tains the FxxLL motif. Our data show that the FxxLL
motif of ZNHIT-1 is required for the binding of
ZNHIT-1 to Rev-erbb.The analysis of homologues
showed that ZNHIT-1 is conserved in eukaryotes. In
particular, its six cysteines are absolutely conserved
from humans to yeast. Interestingly, ZNHIT-1 D4, with-
out the zinc finger HIT domain, did not remove the
inhibitory effect of Rev-erbb on the transcription of
apoCIII (Fig. 7B). This suggests that the HIT domain is
a functional activity domain for ZNHIT-1.
Rev-erbb is a transcriptional silencer and does not
possess an activation domain; however, it does contain

a transcriptional silencing domain in the C-terminal
putative LBD [5]. ChIP assay showed that Rev-erbb
could recruit ZNHIT-1 to the apoCIII promoter
(Fig. 7A), and this recruitment did not modulate Rev-
erbb DNA-binding activity (Fig. 8). Furthermore,
ZNHIT-1 removed the repression of apoCIII induced
by Rev-erbb (Fig. 7B). The function of ZNHIT-1 was
dependent on the presence of the FxxLL motif for
binding to Rev-erbb through its interaction with LBD
(Fig. 5), and the HIT domain for its activity (Fig. 7B).
Like Rev-erbb, Rev-erba was also capable of bind-
ing ZNHIT-1 (Fig. 4C). It is well known that Rev-erba
is a major player in the circadian rhythm. However,
ZNHIT-1 expression in the liver and heart was not cir-
cadian in an extensive and divergent circadian gene
Fig. 8. No effect of ZNHIT-1 on Rev-erbb DNA-binding activity to
the apoCIII promoter. Quantitative PCR for ChIP products precipi-
tated with anti-Rev-erbb serum was performed using promoter-
specific primers. The relative ChIP DNA level was normalized
against the input DNA.
J. Wang et al. ZNHIT-1 as a cofactor of Rev-erbb
FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS 5377
expression study [33]. In order to further verify this,
mRNA of mouse liver was extracted every 6 h for
30 h, and real-time quantitative PCR showed that
ZNHIT-1 expression was not circadian (data not
shown).
In addition to ZNHIT-1, we recently found that
Rev-erbb could interact with different cofactors, includ-
ing coactivators such as histone acetyl-transferase and

corepressors such as histone deacetylase (J. Wang et al.,
School of Life Sciences, Fundan University, Shanghai,
China, unpublished results). In the case of ‘classical’
NRs, NRs recruit corepressors to repress transcription
in the absence of ligands. In contrast, ligand binding
induces a conformational change of the receptor,
excludes corepressors from the complex and leads to
the binding of coactivators instead. However, Rev-erbb
is an orphan NR, whose ligands, if any, have not been
identified. Reinking et al. [34] have shown that the
ligand-binding pocket of E75, a Drosophila orthologue
of human Rev-erb, is tightly bound by haem. Haem
binding of E75 can be disrupted by mutating the two
most highly conserved histidine residues in the LBD,
and these two residues are also present in the vertebrate
orthologue of E75, Rev-erba and Rev-erbb. Interest-
ingly, they also identified E75 as a potential gas sensor,
and showed that CO or NO binds to E75 to interfere
with E75-mediated repression. All of the above findings
suggest that there may be certain unidentified ligands
which may regulate the activity of Rev-erbb and con-
trol the recruitment of different cofactors.
The mechanism by which ZNHIT-1 removes the
Rev-erbb-induced inhibition of transcription remains
unclear. Most cofactors affect transcription through
direct regulation of chromatin remodelling or recruit-
ment of other chromatin remodelling proteins. Human
ZNHIT-1 shares 30% identity with Vps71 (P46973)
from Saccharomyces cerevisiae and 67% identity with
its HIT domains. Vps71 is a subunit of the SWR1

chromatin remodelling complex that incorporates the
histone variant H2A.Z into nucleosomes [35]. The
histone variant H2A.Z is implicated in transcription
activation and the prevention of ectopic spread of
heterochromatin [36]. Cai et al. [24] reported that the
ZNHIT-1 protein was a subunit of the SRCAP com-
plex, which is the closest mammalian homologue of
SWR1. Recent research has found that the human
SRCAP complex can remodel chromatin and activate
gene transcription [37]. The above findings imply that
ZNHIT-1 may affect transcriptional regulation
through the recruitment of a chromatin remodelling
complex. Further studies are needed to clarify the
molecular mechanism underlying the activation of
apoCIII by ZNHIT-1.
In summary, a zinc finger HIT domain-containing
protein, ZNHIT-1, interacted with Rev-erbb. The
FxxLL motif of ZNHIT-1 and the putative LBD of
Rev-erbb were required for this interaction. Further-
more, ZNHIT-1 was recruited to the apoCIII promoter
by Rev-erbb and relieved the transcriptional repression
of Rev-erbb. These findings indicate that ZNHIT-1
functions as a cofactor to regulate the activity of
Rev-erbb, possibly by chromatin remodelling.
Experimental procedures
Plasmid construction and protein expression
The coding sequence (CDS) of human Rev-erbb was ampli-
fied from the human marathon liver cDNA library (Clon-
tech, Tokyo, Japan) by PCR and subcloned into pET28a
(catalogue # 69864-3; Novagen, Darmstadt, Germany) and

pCMV-HA (Clontech) vectors. The CDS of human
ZNHIT-1 was amplified from the human liver cDNA
library (Clontech) by PCR and subcloned into pGEX4T-3
(catalogue # U13855; Pharmacia Biotech, Piscataway, NJ,
USA) and pCMV-Myc (Clontech) vectors. ZNHIT-1
L87A ⁄ L88A was generated by PCR, and subsequently
cloned into pEGX4T-3 and pCMV-Myc vectors. pET28a-
Rev-erbb, pGEX4T-3-ZNHIT-1 and plasmids containing
serial truncated fragments were expressed in Escherichia coli
strain BL21 DE3. The purification of protein was performed
according to the manufacturer’s instructions (Novagen).
Yeast two-hybrid screening
Yeast two-hybrid screening was performed as described
previously [37]. Briefly, the yeast two-hybrid screen was car-
ried out using the Matchmaker Two-hybrid system 3 (Clon-
tech). Rev-erbb was used as bait to screen the human fetal
liver cDNA library (Clontech). The mating test was used to
pick out the specific interaction between bait and prey.
GST pull-down assay
GST-ZNHIT-1 and GST-ZNHIT-1 fragment fusion proteins
were expressed in bacteria and purified on glutathione–
Sepharose (catalogue # 17-0757-01; Pharmacia) as specified
by the manufacturer. The 6His-tagged Rev-erbb and Rev-
erbb fragments were expressed in bacteria and purified on
nickel nitrilotriacetic acid agarose (catalogue # 30210;
Qiagen, Venlo, the Netherlands). GST pull-down assays were
performed as described previously [37].
Co-IP assay
For the determination of the interaction between ZNHIT-1
and Rev-erbb, HepG2 cells were transfected with Myc-

ZNHIT-1 as a cofactor of Rev-erbb J. Wang et al.
5378 FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS
ZNHIT-1. The lysate was first precipitated with anti-Myc
and then detected with anti-Rev-erbb. The interaction
between ZNHIT-1 and Rev-erba was assessed using HepG2
cells cotransfected with Myc-ZNHIT-1 and pCMV-
HA-Rev-erba. Anti-HA and anti-Myc were used for precip-
itation and detection, respectively. The Co-IP assay was
carried out as described previously [38].
Subcellular localization analysis
The complete open reading frames of Rev-erbb and
ZNHIT-1 were constructed in-frame in the plasmids pEG-
FP and pDsRed, respectively. Hela and HepG2 cells grown
on coverslips were transiently transfected with plasmids
pEGFP-Rev-erbb and pDsRed-ZNHIT-1, or cotransfected
with both plasmids. After 36 h of transfection, the cells
were fixed with 4% formaldehyde and stained with
1 lgÆmL
)1
of 4¢,6¢-diamidino-2-phenylindole to visualize the
nuclei with an Olympus IX71 laser microscope.
Luciferase reporter gene assay
The promoter of the human apoCIII gene (nucleotides
)1408 to +24) was obtained from human genomic DNA
by PCR and subcloned into the promoter-less luciferase
reporter plasmid pGL3-basic (catalogue # E1751; Promega,
Madison, WI, USA), generating pGL3-apoCIII-pro. HepG2
cells were cotransfected using FuGENE 6 reagent (catalo-
gue # 11814443001; Roche, Basel, Switzerland) with 100 ng
of apoCIII promoter-driven luciferase expression plasmid

pGL3-apoCIII-pro and the indicated amount of human
Rev-erbb-expressing plasmid pCMV-Rev-erbb, as well as
10 ng of pRL (sea pansy) as an internal control for trans-
fection efficiency. The dosage of transfected plasmids in
one well was kept constant by the addition of appropriate
amounts of the empty vector pCMV. After 36 h, the cells
were lysed by 200 lL of Promega lysis buffer for 10 min at
room temperature. Firefly and Renilla luciferase activities
were measured using a Dual-LuciferaseÒ Reporter Assay
Kit (catalogue # 192445; Promega) on a Lumistar lumi-
nometer (BMG Laboratory Technologies, Offenburg,
Germany). Firefly luciferase activity values were divided by
Renilla luciferase activity values to obtain normalized lucif-
erase activities. To facilitate comparisons within a given
experiment, the activity data were presented as relative
luciferase activities. The final relative activity was calculated
from three independent experiments performed in triplicate.
ChIP assay
HepG2 cells were cotransfected with 1 lg of pCMV-Rev-
erbb and either pCMV-ZNHIT-1 or pCMV vector. After
incubation, ChIP assay and PCR were performed as
described previously [39]. The primers for PCR were
designed to ensure specific amplification of a 230-bp frag-
ment of the apoCIII promoter, with the forward primer
(TCTCCTAGGGATTTCCCAACTCTCC) and the reverse
primer (CTGCCTCTAGGGATGAACTGAGCAG). Quan-
titative real-time PCR was performed as described previ-
ously [40]. Quantitative PCR for ChIP products
precipitated with anti-Rev-erbb serum was performed using
promoter-specific primers. The relative ChIP DNA level

was normalized against the input DNA.
Stable knock-down of ZNHIT-1
The oligonucleotides encoding the ZNHIT-1 siRNA were 5¢-
GATCCGAGACTGCCTCAGTTTGATTCAAGAGATCA
AACTGAGGCAGTCTCTTTTTT-3¢ and 5¢-AGCTTAAA
AAAGAGACTGCCTCAGTTTGATCTCTTGAATCAAA
CTGAGGCAGTCTCG-3¢. These oligonucleotides were
annealed and subcloned to downstream of the U6 promoter
in pGCsi-U6 ⁄ Neo ⁄ GFP (Genechem, Shanghai, China) using
HindIII and BamHI. The empty plasmid or RNAi plasmid
was transfected into HepG2 cells using Fugene-HD transfect
reagent (catalogue # 93539521; Roche). After 1 day of incu-
bation, media from all plates were replaced with selective
medium containing 300 lgÆmL
)1
of Geneticin (Gibco ⁄ BRL,
Grand Island, NY, USA). Cells were grown in selective
media for 2 weeks, and G418-resistant colonies were estab-
lished in six-well plates, expanded and cloned independently.
Acknowledgements
This work was supported by grants from the National
Nature Science Foundation of China (NSFC 30671175
and 30370752) and from the Specialized Research
Fund for the Doctoral Program of High Education
(SRFDP 20060246017).
References
1 Robinson-Rechavi M, Escriva Garcia H & Laudet V
(2003) The nuclear receptor superfamily. J Cell 116,
585–586.
2 Rosen J, Marschke K & Rungta D (2003) Nuclear hor-

mone receptor assays for drug discovery. Curr Opin
Drug Discov Devel 6, 224–230.
3 Smith CL & O’Malley BW (2004) Coregulator function:
a key to understanding tissue specificity of selective
receptor modulators. Endocr Rev 25 , 45–71.
4 Lazar MA, Hodin RA, Darling DS & Chin WW (1989) A
novel member of the thyroid ⁄ steroid hormone receptor
family is encoded by the opposite strand of the rat c-erbA
alpha transcriptional unit. Mol Cell Biol 9, 1128–1136.
5 Burke L, Downes M, Carozzi A, Gigue’re V & Mus-
cat GE (1996) Transcriptional repression by the
J. Wang et al. ZNHIT-1 as a cofactor of Rev-erbb
FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS 5379
orphan steroid receptor Rev-erbb ⁄ Rev-erb beta is
dependent on the signature motif and helix 5 in the E
region: functional evidence for a biological role of
Rev-erbb in myogenesis. Nucleic Acids Res 24, 3481–
3489.
6 Downes M, Burke LJ, Bailey PJ & Muscat GE (1996)
Two receptor interaction domains in the corepressor,
N-CoR ⁄ RIP13, are required for an efficient interaction
with Rev-erbA alpha and Rev-erbb: physical association
is dependent on the E region of the orphan receptors.
Nucleic Acids Res 24, 4379–4386.
7 Adelmant G, Be’gue A, Stehelin D & Laudet V (1996)
A functional Rev-erb responsive element located in the
human Rev-erb promoter mediates a repressing activity.
Proc Natl Acad Sci USA 93, 3553–3558.
8 Dussault I & Giguere V (1997) Differential regulation
of the N-myc proto-oncogene by ROR alpha and Rev-

erbb, two orphan members of the superfamily of
nuclear hormone receptors. Mol Cell Biol 17, 1860–
1867.
9 Yin L & Lazar MA (2005) The orphan nuclear receptor
Rev-erbalpha recruits the N-CoR ⁄ histone deacetylase 3
corepressor to regulate the circadian Bmal1 gene. Mol
Endocrinol 19, 1452–1459.
10 Kassam A, Capone JP & Rachubinski RA (1999)
Orphan nuclear hormone receptor RevErb modulates
expression from the promoter of the hydratase-dehydro-
genase gene by inhibiting peroxisome proliferator-acti-
vated receptor a-dependent transactivation. J Biol Chem
274, 22 895–22 900.
11 Bois-Joyeux B, Chauvet C, Nacer-Che’rif H, Bergeret
W, Mazure N, Gigue’re V, Laudet V & Danan JL
(2000) Modulation of the far upstream enhancer of the
rat a-fetoprotein gene by members of the RORa, Rev-
erba and Rev-erbb groups of monomeric orphan
nuclear receptors. DNA Cell Biol 19, 589–599.
12 Vu-Dac N, Chopin-Delannoy S, Gervois P, Bonnelye E,
Martin G, Fruchart JC, Laudet V & Staels B (1998)
The nuclear receptors peroxisome proliferator-activated
receptor alpha and Rev-erbalpha mediate the species-
specific regulation of apolipoprotein AI expression by
fibrates. J Biol Chem 273, 25 713–25 720.
13 Balsalobre A, Damiola F & Schibler U (1998) A serum
shock induces circadian gene expression in mammalian
tissue culture cells. Cell 93, 929–937.
14 Delaunay F, Thisse C, Marchand O, Laudet V & Thisse
B (2000) An inherited functional circadian clock in

zebrafish embryos. Science 289, 297–300.
15 Triqueneaux G, Thenot S, Kakizawa T, Antoch MP,
Safi R, Takahashi JS, Delaunay F & Laudet V (2004)
The orphan receptor Rev-erbalpha gene is a target of
the circadian clock pacemaker. J Mol Endocrinol 33,
585–608.
16 Preitner N, Damiola F, Lopez-Molina L, Zakany J,
Duboule D, Albrecht U & Schibler U (2002) The
orphan nuclear receptor REV-ERBalpha controls
circadian transcription within the positive limb of the
mammalian circadian oscillator. Cell 110, 251–260.
17 Coste H & Rodriguez JC (2002) Orphan nuclear hor-
mone receptor Rev-erbalpha regulates the human apoli-
poprotein CIII promoter. J Biol Chem
277, 27 120–27
129.
18 Shachter NS (2001) Apolipoproteins C-I and C-III as
important modulators of lipoprotein metabolism. Curr
Opin Lipidol 12, 297–304.
19 Assmann G, Schulte H & von Eckardstein A (1996)
Hypertriglyceridemia and elevated lipoprotein (a) are
risk factors for major coronary events in middle-aged
men. Am J Cardiol 77, 1179–1184.
20 Krauss RM (1998) Atherogenicity of triglyceride-rich
lipoproteins. Am J Cardiol 81, 13B–17B.
21 Hasty AH, Shimano H, Osuga J & Yamada N (2001)
Severe hypercholesterolemia, hypertriglyceridemia, and
atherosclerosis in mice lacking both leptin and the low
density lipoprotein receptor. J Biol Chem 276, 37 402–
37 408.

22 Tybjaerg-Hansen A, Nordestgaard BG, Gerdes LU,
Faergeman O & Humphries SE (1993) Genetic markers
in the apo AI-CIII-AIV gene cluster for combined
hyperlipidemia, hypertriglyceridemia, and predisposition
to atherosclerosis. Atherosclerosis 100, 157–169.
23 Galibert F, Alexandraki D, Baur A, Boles E, Chalwat-
zis N, Chuat J-C, Coster F, Cziepluch C & Karpfinger-
Hartl L (1996) Complete nucleotide sequence of Saccha-
romyces cerevisiae chromosome X. EMBO J 15, 2031–
2049.
24 Cai Y, Jin J, Florens L, Swanson SK, Kusch T, Li B,
Workman JL, Washburn MP, Conaway RC & Con-
away JW (2005) The mammalian YL1 protein is a
shared subunit of the TRRAP ⁄ TIP60 histone acetyl-
transferase and SRCAP complexes. J Biol Chem 280,13
665–13 670.
25 He F, Umehara T, Tsuda K, Inoue M, Kigawa T,
Matsuda T, Yabuki T, Aoki M, Seki E, Terada T et al.
(2007) Solution structure of the zinc finger HIT domain
in protein FON. Protein Sci 16, 1577–1587.
26 Le Douarin B, Lade Nielsen A, Garnier J-M, Ichinose
H, Jeanmougin F, Losson R & Chambon P (1996) A
possible involvement of TIF1a and TIF1b in the epige-
netic control of transcription by nuclear receptors.
EMBO J 15, 6701–6715.
27 Heery DM, Kalkhoven E, Hoare S & Parker MG
(1997) A signature motif in transcriptional co-activators
mediates binding to nuclear receptors. Nature 387, 733–
736.
28 Heery DM, Hoare S, Hussain S, Parker MG & Shepp-

ard H (2001) Core LXXLL motif sequences in CREB-
binding protein, SRC1, and RIP140 define affinity and
selectivity for steroid and retinoid receptors. J Biol
Chem 276, 6695–6702.
ZNHIT-1 as a cofactor of Rev-erbb J. Wang et al.
5380 FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS
29 Burke LJ, Downes M, Laudet V & Muscat GE (1998)
Identification and characterization of a novel corepres-
sor interaction region in RVR and Rev-erbA alpha.
Mol Endocrinol 12, 248–262.
30 Raspe E, Duez H, Mansen A, Fontaine C, Fievet C,
Fruchart JC, Vennstrom B & Staels B (2002) Identifica-
tion of Rev-erbalpha as a physiological repressor of
apoC-III gene transcription. J Lipid Res 43, 2172–2181.
31 Chomez P, Neveu I, Mansen A, Kiesler E, Larsson L,
Vennstrom B & Arenas E (2000) Increased cell death
and delayed development in the cerebellum of mice
lacking the Rev-erbA (a) orphan receptor. Development
127, 1489–1498.
32 Huang N, vom Baur E, Garnier JM, Lerouge T,
Vonesch JL, Lutz Y, Chambon P & Losson R (1998)
Two distinct nuclear receptor interaction domains in
NSD1, a novel SET protein that exhibits characteristics
of both corepressors and coactivators. EMBO J 17,
3398–3412.
33 Storch KF, Lipan O, Leykin I, Viswanathan N, Davis
FC, Wong WH & Weitz CJ (2002) Extensive and diver-
gent circadian gene expression in liver and heart. Nature
417, 78–83.
34 Reinking J, Lam MM, Pardee K, Sampson HM, Liu S,

Yang P, Williams S, White W, Lajoie G, Edwards A
et al. (2005) The Drosophila nuclear receptor e75 con-
tains heme and is gas responsive. Cell 122, 195–207.
35 Krogan NJ, Baetz K, Keogh M-C, Datta N, Sawa C,
Hieter TP & Greenblatt JF (2004) Regulation of chro-
mosome stability by the histone H2A variant Htz1, the
Swr1 chromatin remodeling complex, and the histone
acetyltransferase NuA4. Proc Natl Acad Sci USA 101,
13 513–13 518.
36 Li B, Pattenden SG, Lee D, Gutie
´
rrez J, Chen J, Seidel C,
Gerton J & Workman JL (2005) Preferential occupancy
of histone variant H2AZ at inactive promoters influences
local histone modifications and chromatin remodeling.
Proc Natl Acad Sci USA 102, 18 385–18 390.
37 Ruhl DD, Jin J, Cai Y, Swanson S, Florens L, Wash-
burn MP, Conaway RC, Conaway JW & Chrivia JC
(2006) Purification of a human SRCAP complex that
remodels chromatin by incorporating the histone variant
H2A.Z into nucleosomes. Biochemistry 45, 5671–5677.
38 Shen L, Hu J, Lu H, Wu M, Qin W, Wan DF, Li YY
& Gu JR (2003) The apoptosis-associated protein
BNIPL interacts with two cell proliferation-related pro-
teins, MIF and GFER. FEBS Lett 540, 86–90.
39 Reid G, Metivier R, Lin CY, Denger S, Ibberson D,
Ivacevic T, Brand H, Benes V, Liu ET & Gannon F
(2005) Multiple mechanisms induce transcriptional
silencing of a subset of genes, including oestrogen
receptor alpha, in response to deacetylase inhibition by

valproic acid and trichostatin A. Oncogene 24,
4894–4907.
40 Hearnes JM, Mays DJ, Schavolt KL, Tang L, Jiang X
& Pietenpol JA (2005) Chromatin immunoprecipitation-
based screen to identify functional genomic binding sites
for sequence-specific transactivators. Mol Cell Biol 25,
10 148–10 158.
Supplementary material
The following supplementary material is available
online:
Fig. S1. A phylogenetic tree showing the relationship
between ZNHIT-1 and other zinc finger HIT
domain proteins was constructed by neighbour-joining
analysis.
This material is available as part of the online article
from
Please note: Blackwell Publishing is not responsible
for the content or functionality of any supplementary
materials supplied by the authors. Any queries (other
than missing material) should be directed to the corre-
sponding author for the article.
J. Wang et al. ZNHIT-1 as a cofactor of Rev-erbb
FEBS Journal 274 (2007) 5370–5381 ª 2007 The Authors Journal compilation ª 2007 FEBS 5381