Báo cáo khoa học: Regulation of the human leukocyte-derived arginine aminopeptidase/endoplasmic reticulum-aminopeptidase 2 gene by interferon-c pot

Regulation of the human leukocyte-derived arginine
aminopeptidase/endoplasmic reticulum-aminopeptidase 2
gene by interferon-c
Toshihiro Tanioka
1,
*, Akira Hattori
1
, Shigehiko Mizutani
2
and Masafumi Tsujimoto
1
1 Laboratory of Cellular Biochemistry, RIKEN, Wako, Saitama, Japan
2 Department of Obstetrics and Gynecology, Nagoya University School of Medicine, Showa, Nagoya, Japan
Keywords
aminopeptidase; antigen-presentation;
interferon regulatory factor; interferon-c;
PU.1
Correspondence
M. Tsujimoto, Laboratory of Cellular
Biochemistry RIKEN (The Institute of
Physical and Chemical Research), 2-1
Hirosawa, Wako-shi, Saitama 351-0198
Japan
Fax: +81 48 462 4670
Tel: +81 48 467 9370
E-mail:
*Present address
Laboratory of Medical Information, Showa
University School of Pharmaceutical
Sciences, Shinagawa, Tokyo 142-8555,
Japan

Note
The nucleotide sequence of human
L-RAP(s)1 and L-RAP(s)2 reported in this
paper have been submitted to the
GenBank
TM
⁄ EMBL ⁄ DDBJ Data Bank with
accession numbers AY028805 and
AB163917, respectively.
(Received 12 October 2004, revised 26
November 2004, accepted 8 December
2004)
doi:10.1111/j.1742-4658.2004.04521.x
The leukocyte-derived arginine aminopeptidase (L-RAP) is the second ami-
nopeptidase localized in the endoplasmic reticulum (ER) processing anti-
genic peptides presented to major histocompatibility complex (MHC) class
I molecules. In this study, the genomic organization of the gene encoding
human L-RAP was determined and the regulatory mechanism of its expres-
sion was elucidated. The entire genomic structure of the L-RAP gene is
similar to both placental leucine aminopeptidase (P-LAP) and adipocyte-
derived leucine aminopeptidase (A-LAP) genes, confirming the close relation-
ship of these three enzymes. Interferon (IFN)-c up-regulates the expression
of the L-RAP gene. Deletion and site-directed mutagenic analyses of the
5¢-flanking region of the L-RAP gene and electrophoretic mobility shift
assay indicated that while interferon regulatory factor (IRF)-2 is important
in the basal condition, IRF-1 is the primary regulator of IFN-c-mediated
augmentation of the gene expression. In addition, PU.1, a member of the
E26 transformation-specific family of transcription factors, also plays a role
in the regulation of gene expression. The maximum expression of the gene
was achieved by coexpression of IRF-1 and PU.1 in HEK293 cells and

IRF-2 suppressed the IRF-1-mediated enhancement of gene expression,
suggesting that IFN-c-induced L-RAP gene expression is cooperatively
regulated by IRFs and PU.1 transcription factors.
Abbreviations
A-LAP, adipocyte-derived leucine aminopeptidase; BAC, bacterial artificial chromosome; CHX, cycloheximide; DTT, dithiothreitol; EMSA,
electrophoretic mobility shift assay; ER, endoplasmic reticulum; ERAP, endoplasmic reticulum aminopeptidase; Ets, E26 transformation-
specific; IFN, interferon; IL, interleukin; IRF, interferon regulatory factor; L-RAP, leukocyte-derived arginine aminopeptidase; MHC, major
histocompatibility complex; P-LAP, placental leucine aminopeptidase; PMSF, phenylmethanesulfonyl fluoride; TAP, transporter associated
with antigen processing.
916 FEBS Journal 272 (2005) 916–928 ª 2005 FEBS
Aminopeptidases hydrolyze N-terminal amino acids of
proteins or peptide substrates. They are distributed
widely in animal and plant tissues as well as in bacteria
and fungi, suggesting that they play important roles in
various biological processes [1]. Among them, M1
zinc-metallopeptidases (gluzincins) share the consensus
GAMEN and HEXXH(X)
18
E zinc-binding motifs
essential for enzymatic activity [2,3]. So far, nine
enzymes belonging to the family were identified and
laeverin was recently reported as the tenth member
although its enzymatic activity was not reported [4,5].
In our previous work, we cloned a cDNA for the pla-
cental leucine aminopeptidase (P-LAP) ⁄ oxytocinase, a
type II membrane-spanning protein which belongs to
the M1 family of aminopeptidases [6]. Subsequently we
cloned cDNAs encoding adipocyte-derived leucine
aminopeptidase (A-LAP) [7], which is also designated as
puromycin-insensitive leucine-specific aminopeptidase

(PILS-AP) or ER-aminopeptidase (ERAP)-1 [8,9], and
leukocyte-derived arginine aminopeptidase (L-RAP) ⁄
ERAP2 [4] as highly homologous proteins to P-LAP.
Structural and phylogenetic analyses indicated the close
relationship between these three enzymes. Therefore we
proposed that they should be classified into the oxyto-
cinase subfamily of M1 aminopeptidases [10].
Recent evidence facilitates new insights into the
biological significance of the oxytocinase subfamily
of aminopeptidases. P-LAP ⁄ oxytocinase, which is
also designated as insulin-regulated aminopeptidase
(IRAP), was shown to translocate from the intracellu-
lar compartment to the plasma membrane in a stimu-
lus-dependent manner and may regulate concentration
of substrate peptide hormones on the cell surface
[11–14]. This enzyme was recently shown to be the
angiotensin IV receptor and may play a role in
memory retention and retrieval [15,16]. On the other
hand, we and others reported that A-LAP ⁄ ERAP1 is a
final processing enzyme of the precursors of the major
histocompatibility complex (MHC) class I-presented
antigenic peptides [9,17]. A-LAP ⁄ ERAP1 was also
shown to play roles in blood pressure regulation and
angiogenesis [18–20]. Moreover, the enzyme was shown
to bind to cytokine receptors such as tumour necrosis
factor type I receptor, interleukin (IL)-6 a-receptor
and IL-1 type II receptor, and facilitate ectodomain
shedding of these receptors [21–23]. As for
L-RAP ⁄ ERAP2, we have shown that the enzyme is
retained in the endoplasmic reticulum (ER) and can

trim some precursor peptides to MHC class I ligands
[4]. These results indicate that the mammalian amino-
peptidases belonging to the oxytocinase subfamily play
important roles in the regulation of several biological
processes.
Precursors of MHC class I-presented peptides with
extra N-terminal residues are trimmed to mature epi-
topes in the ER [24,25]. The peptides are first cleaved
from endogenously synthesized proteins by proteasome
or tripeptidyl peptidase II in the cytoplasm, transpor-
ted into ER-lumen and then trimmed by certain
aminopeptidases. Until now, only two aminopeptid-
ases (A-LAP ⁄ ERAP1 and L-RAP ⁄ ERAP2) have been
identified in the ER-lumen and have shown their
ability to trim antigenic peptides [4,9,17]. Characteris-
tically, as in the case of other components included in
the presentation of MHC class I ligands such as MHC
class I molecules, transporter associated with antigen
processing (TAP) and proteasome b-subunits, inter-
feron (IFN)-c enhances the expression of these
enzymes [24]. Although IFN-c-inducible aminopeptid-
ases including lens LAP, A-LAP, and L-RAP play
roles in the processing or degradation of antigenic
peptides [4,9,17,26], the mechanisms of IFN-c-medi-
ated regulation of aminopeptidase gene expression
have never been examined.
In the current study, we have elucidated the genomic
structure of human L-RAP gene for the first time.
Characterization of the promoter region of the gene
indicates that while interferon regulatory factor (IRF)-

2 is important in the basal condition, IRF-1 is the
primary regulator of IFN-c-mediated augmentation of
the gene expression. It is also shown that PU.1, a
member of the E26 transformation-specific (Ets) family
of transcription factors, plays a role in the regulation
of gene expression. Our data provide the molecular
basis of regulatory mechanisms of enzymatic activity
of L-RAP, which may play an important role in the
processing of antigenic peptides presented to MHC
class I molecules in the ER.
Results
Genomic organization of human L-RAP gene
To elucidate the genomic organization of the L-RAP
gene, we screened a human genome database prepared
in a bacterial artificial chromosome (BAC). This led to
the identification of BAC clone RPCI-11-496M2
(accession number AC009126), which contains all
exons of the L-RAP gene. All exons and intron–exon
junctions were determined from the database. The
sequences of splice junctions obey the GT-AG rule
[27]. As shown in Fig. 1, the gene spans 45 kb and
contains 19 exons ranging in size from 28 bp (exon 1)
to 697 bp (exon 2) and the overall structure of the
gene is quite similar to both P-LAP and A-LAP genes.
The first exon includes only the 5¢-untranslated region.
T. Tanioka et al. Regulation of L-RAP ⁄ ERAP2 gene expression
FEBS Journal 272 (2005) 916–928 ª 2005 FEBS 917
Exon 2 includes the remaining 5¢-untranslated region
and coding sequence for the first 192 amino acids. The
GAMEN and HEXXH motifs are encoded in exon 6

and an essential glutamic acid residue located 19
amino acids downstream from the HEXXH motif is
encoded in exon 7. Exon 19 contains the coding
sequence for the last 47 amino acids, stop codon
(TAA) and the 3¢-untranslated region.
In our previous work, we obtained a cDNA enco-
ding the truncated form of L-RAP [4]. Exon 10 that is
deleted in the P-LAP gene encodes a sequence that
may function as a hinge region [28]. As shown in
Fig. 1, a transcript encoding this form, termed L-
RAP(s)1 (accession number AY028805), is generated
by differential usage of exon 10. When an intermediate
nucleotide sequence (AACATGgtaag) matching the
GT-AG rule in the exon functions as a splicing donor,
full length L-RAP transcript is generated. If the
sequence does not act as a splicing donor, the stop
codon (TGA) in the exon causes the generation of
a truncated form. Thereafter, another transcript,
L-RAP(s)2 (accession number AB163917), encoding
the same truncated form was also cloned. Differential
usage of exon 15 causes the generation of two trun-
cated forms of the transcripts. These results indicate
that at least three mRNAs encoding either the full
length or truncated form of L-RAP protein are gener-
ated from a single gene.
Isolation and characterization of human L-RAP
gene promoter
Figure 2 shows the nucleotide sequence of the 5¢-flank-
ing region of the human L-RAP gene. The transcrip-
tional initiation site of the gene was determined by

5¢ RACE and is shown as nucleotide position +1
in the figure. The site (TCAGTC) matches well with
the pyrimidine-rich initiator consensus sequence,
PyPy(A+1)N(T ⁄ A)PyPy, in which Py represents a
pyrimidine residue [29]. Computer analysis of the
sequence revealed no canonical TATA- or CCAAT-
box, suggesting a housekeeping nature for the gene. On
the other hand, several potential transcription factor
binding motifs, such as IRF, GATA-1, Sp-1 and AP-1
were identified in the promoter region of the gene [30].
To characterize the regions regulating transcriptional
activity of the L-RAP gene, chimeric reporter plasmids
encoding the luciferase gene and different lengths of
L-RAP gene were constructed. The resultant chimeric
constructs were then transfected into HEK293 cells to
analyze the promoter activity. As a negative control,
the promoter-less pGL3 basic plasmid was transfected
into the cells. As shown in Fig. 3A, substantial promo-
ter activity was detected when chimeric constructs con-
taining the 5¢-flanking sequence upstream from the A
at position )10 were transfected, confirming that the
5¢-flanking sequence of the L-RAP gene is indeed able
to support transcriptional initiation. The maximum
Fig. 1. Genomic structure of the human
L-RAP gene. Schematic exon–intron
structure of the gene is shown at the center.
Exons are numbered and depicted as boxes.
Asterisks indicate the sites used for the
generation of different mRNAs described in
the text. Generation scheme of L-RAP and

L-RAP(s) proteins is also shown in the figure.
Numbers shown in each protein molecule are
number of amino acids derived from the
respective exons.
Fig. 2. Nucleotide sequence of the 5¢-flanking region of the human
L-RAP gene. The 5¢-flanking region was searched for transcription
factor binding sites by
TF-SEARCH. The exon sequence is shown in
uppercase letters, while that of the untranscribed region is given in
lowercase letters. The transcriptional initiation site shown as +1 was
determined by 5¢-RACE. For the measurement of promoter activity,
the start point of each construct is indicated by an arrowhead.
Regulation of L-RAP ⁄ ERAP2 gene expression T. Tanioka et al.
918 FEBS Journal 272 (2005) 916–928 ª 2005 FEBS
promoter activity was obtained with the constructs
containing the sequence from )33 to +5. Deletion of
the sequence from )33 to )17 caused a decrease in the
promoter activity by nearly 50%. Further deletion to
T at position )9 caused almost complete loss of activ-
ity. Employing Jurkat-T cells, we obtained nearly the
same results.
The promoter sequence from )33 to )17 contains a
sequence (AGAAAGTGAAAGC) with resemblance to
the IRF-E consensus sequence [G(A)AAASYGAA
ASY]. We next examined the role of this site on the
basal promoter activity by constructing mutant
plasmid [phLP5(MuI)] from phLP5 plasmid. We
describe this sequence as the IRF-E site hereafter [31].
Fig. 3. Role of the IRF-E site in IFN-c-induced enhancement of L-RAP gene expression in HEK293 cells. (A) Luciferase expression plasmids
(1 lg) containing sequentially deleted fragments of the L-RAP chimera were transfected into HEK293 cells. Luciferase activity was measured

as described in Experimental procedures and normalized to the b-galactosidase activity of a cotransfected internal control plasmid. The luci-
ferase activity obtained in the basal condition from cells transfected with phLP1 was taken as 100%. (B) Enhancing effect of IFN-c is shown
by fold increase in the figure. (C) Functional analysis of the IRF-E site of the L-RAP gene promoter in HEK293 cells. The phLP5 plasmid hav-
ing either an intact or mutated IRF-E site was transfected into HEK293 cells and the luciferase activity was measured as described above.
Open bars indicate luciferase activity in the basal condition and closed bars in the IFN-c-stimulated condition.
T. Tanioka et al. Regulation of L-RAP ⁄ ERAP2 gene expression
FEBS Journal 272 (2005) 916–928 ª 2005 FEBS 919
Although substantial activity was retained, mutation
of the IRF-E site caused a significant decrease in the
promoter activity (Fig. 3C). These results suggest that
the IRF-E site is crucial for the maximum promoter
activity of the L-RAP gene in the basal condition.
Mechanism of IFN-c-mediated regulation
of L-RAP gene expression
Because L-RAP gene expression is enhanced by IFN-c,
we next examined its regulatory mechanism. As an ini-
tial experiment, decay rates of L-RAP mRNAs pre-
pared from Jurkat-T cells treated with or without
IFN-c were compared in the presence of actinomycin
D. As shown in Fig. 4A, there was little difference
between decay rates of L-RAP mRNAs in the cells
treated with or without IFN-c, indicating that the
cytokine-induced enhancement of mRNA accumula-
tion could not be attributed to the change of stability
of L-RAP mRNA. In addition, it was found that the
increase in mRNA accumulation was not observed in
the presence of cycloheximide (CHX), indicating that
de novo protein synthesis was required for the action
of IFN-c (Fig. 4B).
To elucidate the role of IFN- c in the regulation of

L-RAP gene expression, the luciferase-reporter assay
was conducted to identify cytokine responsive elements
in the gene. As shown in Fig. 3A,B, IFN-c induced
about a fourfold increase in the expression of luci-
ferase activity in HEK293 cells transfected with con-
structs containing the sequence from )33 to )17. After
deletion of this sequence, IFN-c had no enhancing
activity. These results indicate that the sequence is
essential for the cytokine-induced increase in L-RAP
gene expression.
Because the sequence from )33 to )17 contains the
IRF-E site, we next examined the role of this site in
L-RAP gene regulation. As shown in Fig. 3C, muta-
tion in this site caused complete loss of IFN-c-induced
increase in the luciferase activity, confirming the role
of the IRF-E site in the cytokine-mediated L-RAP
gene regulation. These results indicate that the IRF-E
site plays an important role both in the basal and
IFN-c-mediated L-RAP gene expression.
On the other hand, it was found that in addition to
the IRF-E site, the Ets site in the promoter also plays
a role in gene expression in Jurkat-T cells. As in the
case of HEK293 cells, the IRF-E site was required for
the maximum expression of the gene. IFN-c induced
an eightfold increase in the gene expression in cells
transfectied with phLP5 plasmid. When Jurkat-T cells
were transfected with plasmids containing the sequence
from )67 to )33 that contains the Ets site, a further
increase (about 13- to 15-fold) was observed
(Fig. 5A,B). Mutation of either the IRF-E

[phLP4(MuI)] or Ets site [phLP4(MuE)] caused a
substantial decrease in the IFN-c-mediated gene
expression. Plasmid having mutations in both sites
[phLP4(MuE ⁄ MuI)] had little activity, indicating that
in Jurkat-T cells both the IRF and Ets sites are
required for maximum enhancement of IFN-c-induced
gene expression (Fig. 5C). In contrast, it was found
that the Ets site had no effect on the promoter activity
in HEK293 (data not shown). We found by RT-PCR
A
B
Fig. 4. Requirement of de novo protein synthesis for IFN-c -medi-
ated enhancement of L-RAP gene expression. (A) Effect of IFN-c
on the stability of L-RAP mRNA. Jurkat-T cells were treated with or
without 30 ngÆmL
)1
IFN-c for 12 h at 37 °C and further incubated in
the presence of 1 l
M actinomycin D (ActD) for indicated times.
After incubation, expression levels of L-RAP mRNAs were meas-
ured by Northern blot analysis. Densitometric data showing the
decay rates of mRNAs are also shown. (B) Effect of CHX on the
IFN-c-induced enhancement of L-RAP gene expression. Jurkat-T
cells were preincubated for 3 h in the presence or absence of
1 lgÆmL
)1
cycloheximide (CHX) and further incubated with
30 ngÆmL
)1
IFN-c for 12 h at 37 °C. L-RAP mRNAs were detected

by Northern blot analysis.
Regulation of L-RAP ⁄ ERAP2 gene expression T. Tanioka et al.
920 FEBS Journal 272 (2005) 916–928 ª 2005 FEBS
that expression of PU.1 was detectable in Jurkat-T
cells but not in HEK293 cells (data not shown).
Although the Oct-1 site is located proximal to the
Ets site, mutational analysis indicated that this site
had little effect on the gene expression (data not
shown).
Role of IRFs in the regulation of L-RAP gene
expression
Transcription factors, IRF-1 and -2 are known to bind
to the IRF-E site and regulate IFN-c action [31]. To
examine whether IFN-c affects the expression levels of
Fig. 5. Role of the IRF-E and Ets sites in IFN-c-induced enhancement of L-RAP gene expression in Jurkat-T cells. (A) Luciferase expression
plasmids (10 lg) containing sequentially deleted fragments of the L-RAP chimera were transfected into Jurkat-T cells. Cells were then trea-
ted with or without 30 ngÆmL
)1
IFN-c for 15 h at 37 °C. Luciferase activity was measured as described in Experimental procedures and nor-
malized to the b-galactosidase activity of a cotransfected internal control plasmid. The luciferase activity obtained in the basal condition from
cells transfected with phLP1 was taken as 100%. (B) Enhancing effect of IFN-c is shown by fold increase in the figure. (C) Mutational analy-
sis of the IRF-E and Ets sites of the L-RAP gene promoter in Jurkat-T cells. The phLP4 plasmid having either intact or mutated sites shown
in the left panel of the figure was transfected into Jurkat-T cells and luciferase activity was measured. Open bars indicate luciferase activity
in the basal condition and closed bars in the IFN-c-stimulated condition.
T. Tanioka et al. Regulation of L-RAP ⁄ ERAP2 gene expression
FEBS Journal 272 (2005) 916–928 ª 2005 FEBS 921
IRF-1 and -2 proteins, HEK293 cells were cultured in
the presence or absence of the cytokine (Fig. 6A).
When compared with untreated cells, an increase in the
expression level of IRF-1 was observed in cells treated

with IFN-c, indicating that IRF-1 is a candidate for the
mediator of IFN-c action. In contrast, no effect was
observed on the expression of IRF-2 protein.
To determine whether IRF-1 and -2 bind to the
IRF-E site in the L-RAP promoter sequence, we per-
formed an electrophoretic mobility shift assay (EMSA)
using nuclear extracts from HEK293 cells treated with
or without IFN-c (Fig. 6B). An oligonucleotide probe
corresponding to the IRF-E site bound to nuclear
extracts prepared from both IFN-c-treated and
untreated cells. In spite of the presence of an unex-
pected nonspecific large band, bands shown as IRF-1
and -2 in Fig. 6 were replaced by unlabeled oligo-
nucleotide competitor, indicating their specificity. The
specificity of these bands was also confirmed by oligo-
nucleotide with mutations at the IRF-E site. Although
repeated trials were made, we could not remove the
nonspecific bands.
To identify transcription factors bound to the IRF-
E site, supershift assay was performed using antibodies
raised against IRF-1 and IRF-2. When anti-(IRF-1)
IgG was employed, supershift was observed only in the
IFN-c-treated cells. On the other hand, anti-(IRF-2)
IgG caused supershift in both treated and untreated
cells. Moreover, appearance of two supershift bands
and complete disappearance of the oligonucleotide spe-
cific bands was observed in the presence of both anti-
bodies. These results suggest that while in untreated
cells IRF-2 but not IRF-1 was bound to the IRF-E
site, IRF-1 and IRF-2 co-occupied the site after treat-

ment with IFN-c. We obtained the same results in Jur-
kat-T cells. However, it should be noted here that an
IFN-c-inducible band other than IRF-1 and -2 was
also observed in this experiment, suggesting that tran-
scription factors other than IRF-1 and -2 also partici-
pate in the regulation of L-RAP gene expression.
A
B
CD
Fig. 6. Role of IRF-1 and PU.1 in the expression of the L-RAP gene.
(A) IFN-c-induced increase in the expression of IRF-1 protein in
HEK293 cells. HEK293 cells were treated with or without
30 ngÆmL
)1
IFN-c for 5 h at 37 °C. IRF-1 and IRF-2 in the nuclear
extracts were measured by Western blot analysis. (B) Binding of
IRF-1 and IRF-2 to the IRF-E site of the L-RAP gene. HEK293 cells
were treated with or without 30 ngÆmL
)1
IFN-c for 5 h at 37 °C.
EMSAs were then performed in nuclear extracts using a probe con-
taining the IRF-E site from the L-RAP gene promoter. Supershifts
were conducted using antibodies against indicated factors. Arrow-
heads indicate the positions of IRF-1 and IRF-2. Arrow indicates the
position of unidentified IFN-c-inducible IRF-E binding protein. Lanes
shown as probe are loaded only with labeled probe. (C) Enhance-
ment of promoter activity of the L-RAP gene by transcription fac-
tors. HEK293 cells were cotransfected with phLP4 (0.5 lg) and
various expression plasmids (0.5 lg) shown in the figure. After
24 h, luciferase activity was measured as described in Experimental

procedures. Enhancing effects of the transcription factors are
shown by fold increase using cells transfected only with phLP4 or
as a control. (D) Induction of endogenous L-RAP gene expression
by transcription factors. HEK293 cells were transfected with
expression plasmids (0.5 lg) shown in the figure. For RT-PCR ana-
lysis, total RNA was extracted after 24 h-incubation. Relative
expression level is shown by fold increase in the figure using
mock-transfected cell as a control.
Regulation of L-RAP ⁄ ERAP2 gene expression T. Tanioka et al.
922 FEBS Journal 272 (2005) 916–928 ª 2005 FEBS
To further elucidate the role of IRF-1 and -2 in the
regulation of L-RAP gene expression, phLP4 plasmid
was cotransfected either with pTARGET-IRF-1 or
pTARGET-IRF-2 plasmid into HEK293 cells
(Fig. 6C). IRF-1 overexpressed in HEK293 cells
induced 8.6-fold increase in the luciferase activity, indi-
cating that IRF-1 is indeed able to augment L-RAP
gene expression. IRF-2 and PU.1, a transcription fac-
tor which is known to bind to the Ets site [32], also
had enhancing effects on the luciferase activity, and a
4.7- and 2.1-fold increase in the gene expression was
observed, respectively. Furthermore, maximum gene
expression (18.3-fold increase) was achieved in a syner-
gistic manner when IRF-1 and PU.1 were coexpressed.
On the other hand, coexpression of IRF-1 and IRF-2
caused lower expression of luciferase activity than
expected, suggesting that IRF-2 suppressed IRF-1-
mediated augmentation of the gene expression. We
also examined other Ets site-binding factors such as
Ets-1 and Ets-2 and found that they had little effect

on the gene expression (data not shown). When the
same experiments were performed using either
phLP4(MuE), phLP4(MuI) or phLP4(MuE ⁄ MuI), no
synergistic effect between IRF-1 and PU.1 was
observed (data not shown), confirming that native sites
of both IRF-E and PU.1 are required for the synergis-
tic action. Taken together these results indicate that
while IRF-2 plays a role in the basal condition, IRF-1
can mediate IFN-c-stimulated L-RAP gene expression
synergistically with PU.1.
To determine whether IRF-1 indeed mediates
L-RAP gene expression in vivo, we next examined the
effect of transcription factors on the expression of
endogenous L-RAP gene. Several combinations of
plasmids were transfected into HEK293 cells and their
ability to enhance gene expression was examined by
RT-PCR (Fig. 6D). In mock-transfected cells, L-RAP
mRNA was barely detectable. Among trancription fac-
tors tested, the highest expression was achieved when
IRF-1 was overexpressed. Further increase in the gene
expression was observed when IRF-1 and PU.1 were
coexpressed. In contrast, IRF-2 suppressed IRF-1-
mediated increase in the gene expression, although
IRF-2 alone had some enhancing activity. These
results further confirm the roles of transcription factors
in the expression of the L-RAP gene.
Discussion
L-RAP ⁄ ERAP2 is a newly identified ER aminopepti-
dase and the third member of the oxytocinase sub-
family of M1 family of aminopeptidases [4]. It was

suggested by its ability to generate certain antigenic
peptides that the enzyme plays an important role in
the processing of antigenic peptides presented to MHC
class I molecules in the ER. In this study, we deter-
mined the genomic structure of the human L-RAP
gene and characterized the regulatory mechanisms of
the gene expression. It was found that two forms of
the L-RAP proteins are generated by alternative spli-
cing. While the full length form exerts distinct amino-
peptidase activity, the truncated form has no
enzymatic activity [4]. It is necessary to elucidate the
relationship between these two forms.
The entire genomic structure of the L-RAP gene is
quite similar to the P-LAP and A-LAP genes [28,33].
The GAMEN and HEXXH(X)
18
E motifs essential for
the enzymatic activity are encoded by exons 6 and 7
of the respective genes. Becasue these three genes are
located contiguously around human chromosome
5q15, between versican and calpastatin [28], these data
further confirm the latest divergence of the genes from
a common ancestral gene.
Expression of L-RAP is up-regulated by IFN-c [4].
IFN-c also stimulates the induction of components
included in the processing of MHC class-I ligands [24].
Considering the evidence that suggest a role of L-RAP
as an antigen-trimming enzyme in the ER, it is import-
ant to elucidate the mechanism of the IFN-c-induced
increase in L-RAP gene expression. The transient

expression of the 5¢-flanking region of the L-RAP gene
fused to the luciferase gene in either HEK293 or
Jurkat-T cells allowed the analysis of basal and IFN-c-
mediated regulation of promoter activity. We demon-
strated by deletion analysis that the sequence ranging
from )16 to )10 carries the minimum promoter activ-
ity of the gene. In addition, mutational analysis sug-
gested that the IRF-E site is required for the
maximum enhancement of gene expression in the basal
condition. Because EMSA indicated that IRF-2 is
associated with the IRF-E site in unstimulated
HEK293 cells, our data suggest that IRF-2 can aug-
ment the transcription of the L-RAP gene in the basal
condition. In fact, IRF-2 had some enhancing effect
on the gene expression when overexpressed in HEK293
cells. Although it is generally considered that IRF-2 is
a negative regulator of gene expression [31], it has also
been shown to up-regulate the expression of several
genes such as histone 4 and VCAM-1 [34,35]. However,
we could not completely rule out the possible contribu-
tion of other IRF-E site-binding proteins [31], because
an IFN-c-inducible IRF-E binding protein other than
IRF-1 and IRF-2 was detectable by EMSA, even in
the basal condition. Identification and elucidation of
the role of this protein in L-RAP gene expression is
required.
T. Tanioka et al. Regulation of L-RAP ⁄ ERAP2 gene expression
FEBS Journal 272 (2005) 916–928 ª 2005 FEBS 923
It is worthy of noting here that both P-LAP and
A-LAP promoters also contain IRF-E sites [28,33]. It

was reported that IFN-c had no enhancing effect on
P-LAP gene expression [33]. In our preliminary results,
the IRF-E site in the A-LAP promoter had little effect
on IFN-c-induced gene expression, raising the possib-
lity that IFN-c affects the expression of the A-LAP
gene differently from that of the L-RAP gene.
IFN-c-induced increase in L-RAP gene expression
required de novo protein synthesis and it was found
that IRF-1 was induced by IFN-c. Deletion and muta-
tional analyses indicated that the IRF-E site is res-
ponsible for the cytokine-mediated increase in gene
expression in HEK293 cells. Mutation of the site
caused loss of the cytokine action. EMSA indicated
that while IRF-2 but not IRF-1 bound to the IRF-E
site in the basal condition, binding of both IRF-1 and
-2 were detectable after treatment with IFN-c. More-
over, transfection of the IRF-1 expression plasmid
into HEK293 cells caused an increase in the gene
expression. These results strongly suggest that IRF-1 is
a primary mediator of IFN-c-mediated enhancement
of L-RAP gene expression in HEK293 cells. Because
coexpression of IRF-1 and IRF-2 caused a decrease in
L-RAP gene expression, it is plausible that IRF-2 acts
as a negative regulator of IRF-1-mediated enhance-
ment of gene expression, by co-occupation of the
IRF-E site with IRF-1 in IFN-c-treated cells.
On the other hand, the mechanism of IFN-c-induced
L-RAP gene regulation in Jurkat-T cells is rather com-
plex. It is obvious that in addition to the IRF-E site,
the Ets site located between )63 and )56 also plays a

role in the maximum enhancement of IFN-c-induced
gene expression. It is unlikely that another Ets site
located between )352 and )345 plays a role in the
gene expression, because deletion of this sequence had
little effect on the enhancement of the gene expression.
Among the Ets family transcription factors tested, only
PU.1 could mediate the cytokine action. The expres-
sion of PU.1 is limited to hematopoietic lineages and
is necessary for the differentiation of myeloid cell line-
ages such as macrophages and osteoclasts [36,37]. Our
analysis by RT-PCR indicated the expression of PU.1
in Jurkat-T cells but not HEK293 cells. Therefore, it is
plausible that the lineage-dependent expression of
PU.1 may determine the expression level of the L-RAP
gene. Consistent with this notion, IRF-1 and PU.1
mediated the maximum expression of both reporter
and endogenous genes when overexpressed, even in
HEK293 cells.
The molecular basis of the cooperative action of
transcription factors is considered to be the physical
interaction of the transcription factors involved [37].
For instance, it was reported that IRFs and PU.1
synergistically mediate the transcriptional enhancement
of human interleukin-1 (IL-1)b gene expression. It was
suggested by physical interaction analysis that the fac-
tors might function together as an enhanceosome [38].
As shown in other genes [39,40], it is plausible that
IRF-1 mediates the IFN-c-stimulated L-RAP gene
expression through interaction with PU.1. On the other
hand, it is possible that IRF-2 acts independently from

PU.1 to enhance the gene expression in the basal con-
dition. Because binding of IRF-2 to the IRF-E site was
still observed after IFN-c treatment, it is possible to
speculate that IRF-2 modulates the gene expression by
interacting with IRF-1 in the IRF-E site of the L-RAP
gene. As with the L-RAP gene, Gobin et al. reported
that binding of IRF-2 to the MHC class I molecule
promoter was observed after IFN-c treatment [41]. It
was also reported that IRF-2 co-occupies the IRF-E
site of the class II transactivator type IV promoter with
IRF-1 and synergistically activates the promoter [42].
L-RAP ⁄ ERAP2 is the second ER-lumenal amino-
peptidase to be determined, and can trim certain pre-
cursors of antigenic peptides presented to MHC class I
molecules [4], suggesting the potential significance of
this enzyme in antigen processing. In this study, we
characterized the L-RAP gene for the first time. We
have shown that the IRF-E site located in the proximal
region of the transcription initiation site plays a pivotal
role in the regulation of human L-RAP gene expres-
sion both in the basal and IFN-c -stimulated condi-
tions. As shown in several genes [43–45], it was found
that while IRF-2 plays a role in the basal condition,
IRF-1 is the primary regulator of the gene expression
induced by IFN-c. However, transcription factor(s)
other than IRF-1 and -2 may also regulate L-RAP
gene expression. Further works are required to exam-
ine the involvement of other transcription factors that
bind to the IRF-E site. Considering the significance of
the processing of antigen presented to MHC class I

molecules in several pathophysiological conditions such
as virus infection, tumor generation and self-antigen
generation, it is important to elucidate the total aspects
of the antigen presentation process including the regu-
latory mechanisms of L-RAP gene expression.
Experimental procedures
Identification of the human L-RAP gene
A genomic sequence of  178 kb from the Gen-
Bank
TM
⁄ EMBL ⁄ DDBJ Data Bank was obtained. This
sequence encompasses a region on human chromosome
5q15 where the known P-LAP gene is located. The loca-
Regulation of L-RAP ⁄ ERAP2 gene expression T. Tanioka et al.
924 FEBS Journal 272 (2005) 916–928 ª 2005 FEBS
tions of the exons of the L-RAP gene were determined
using the blast program.
Cell culture
Jurkat-T cells were obtained from the RIKEN Cell Bank
(Tsukuba, Japan) and maintained in RPMI 1640. Human
embryonic kidney (HEK) 293 cells were purchased from
ATCC (Manassas, VA, USA) and maintained in RPMI
1640. All media used in this study was from Sigma (St.
Louis, MO, USA) and supplemented with 10% fetal
bovine serum (JRH Biosciences, Lenexa, KS, USA), peni-
cillin G, and streptomycin (Meiji Seika Co., Kanagawa,
Japan). The cells were cultured in a humidified 5% CO
2
and 95% air incubator at 37 °C.
Analysis of promoter activity

The human L-RAP 5¢-flanking region and its fragments,
which include the transcriptional initiation site, were
amplified by PCR from a BAC clone. PCR products
were inserted into the promoter-less plasmid, pGL3 basic
(Promega, Madison, WI, USA). The nucleotide sequence
of the primers used for PCR amplification were as
follows (5¢-3¢): CGGGTACCTGAACCAGCTAGTACT
TACTG (sense strand of the sequence from )88 to )68)
for phLP3; CGGGTACCTACTCAGGAAGCATGC
AAGT (sense strand of the sequence from )67 to )47)
for phLP4; CGGGTACCACAGAAAGTGAAAGCA
(sense strand of the sequence from )33 to )18) for
phLP5; CGACGCGTTGACTGAAGGGGAATTTACTTT
(antisense strand of the sequence from )17 to +5) for
all constructs. For phLP6 and 7, plasmids were construc-
ted by using synthetic oligonucleotides corresponding to
the 5¢-flanking region. We also constructed phLP1 and
2 plasmids by sequetial deletion of SmaI and Van91I
fragment from phLP0 covering the sequence from )1042
to +5.
Mutagenesis of the IRF and Ets sites was performed
by using mutated oligonucleotides when constructing
the respective plasmids. For making mutations in IRF,
5¢-CACAGA
GGGTGAGGGCAAAAGTAAATTCCCCTT
CAGTCAA-3¢ (sense sequence of mutant IRF-E site) and
5¢-CGCGTTGACTGAAGGGGAATTTACTTTTGC
CCT
CAC
CCTCTGTGGTAC-3¢ (antisense sequence of mutant

IRF-E site) were used. For the Ets mutant, 5¢-GGGGTA
CCTACTCA
AAGAGCATGCAAAGT-3¢ (sense sequence
of mutant Ets site) and 5 ¢-CGACGCGTTGACTGAAGG
GGAATTTACTTT-3¢ (antisense sequence of mutant Ets
site) were used. For the construction of the double mutant
plasmid, 5¢-GGGGTACCTACTCA
AAGAGCATGCAAA
GT-3¢ and 5¢-CGACGCGTTGACTGAAGGGGAATTTA
CTTTTGC
CCTCACCCTCTGTTCTAA-3¢ (antisense sequ-
ence of mutant IRF-E site) were used. The mutated nucleo-
tide sequences are underlined.
The PU.1 expression plasmid pcDNA3-PU.1 which was
constructed by using pcDNA3 (Invitrogen, Carlsbad, CA,
USA) was obtained from M Matsumoto of Saitama Med-
ical School (Saitama, Japan) [46].
Transfection and luciferase assay
For transfection of the reporter plasmid, HEK293 cells
were plated on 24 well plates at a density of 1 · 10
5
cells per well on the day before transfection, while Jurkat-T
cells (1 · 10
6
cells) were plated in a 100 mm dish. Plasmid
DNA was mixed with LipofectAmine (Invitorogen) and
transfected into HEK293 cells following the manufacturer’s
protocol. A pCMVb (0.5 lg) plasmid was employed as an
internal control of transfection efficiency. To transfect into
Jurkat-T cells, the electroporation (225 V, 350 lF) method

was employed, using Electro Cell Manipulator (BTX, San
Diego, CA, USA). After 24 h of transfection, the cells were
washed three times with NaCl ⁄ P
i
and then lysed in reporter
lysis buffer (Promega). The luciferase activity was then
measured with a luciferase assay system (Promega) accord-
ing to the manufacturer’s instruction. Luciferase activity
was measured in triplicate, averaged, and then normalized
to b-galactosidase activity to correct for transfection effi-
ciency. b-Galactosidase activity was measured using o-nitro-
phenyl-b-d-galactopyranoside as a substrate.
Electrophoretic mobility shift assay
Double-stranded oligonucleotides containing the consensus
sequence IRF-E site were radiolabeled with [
32
P]dCTP[aP]
at the 3¢ end with a Klenow fragment and then purified
with a MicroSpin column (Amersham Biosciences, Piscata-
way, NJ, USA). The sense sequences of the synthesized
oligonucleotides used were as follows: for the native IRF-E
site (GAACAGAAAGTGAAAG) in the human L-RAP
5¢-flanking region, and for mutation in the IRF-E site
(GAACAGA
GGGTGAGGG). and the antisense sequences
of synthesized oligonucleotides used were as follows: for
native IRF-E site (TTTGCTTTCACTTTCT) in the human
L-RAP 5¢-flanking region, for mutation in IRF-E site
(TTTGC
CCTCACCCTCT). The mutated nucleotide sequ-

ences are underlined.
Nuclear extracts of the cells were prepared as follows: the
cells suspended in 500 lL of ice-cold buffer A [10 mm He-
pes, 10 mm KCl, 1.5 mm MgCl
2
, 0.5 mm dithiothreitol
(DTT), 0.2 mm phenylmethanesulfonyl fluoride (PMSF):
pH 7.9] were lysed in a Dounce-homogenizer and then the
nuclei were pelleted by centrifugation at 3300 g for 15 min.
The pellets were then suspended in an equal volume of low
salt buffer (20 mm Hepes, 20 mm KCl, 1.5 mm MgCl
2
,
0.2 mm EDTA, 0.5 mm DTT, 0.2 mm PMSF, 25% glycerol:
pH 7.9). Thereafter, half volume of high salt buffer (20 mm
Hepes, 1.4 m KCl, 1.5 mm MgCl
2
, 0.2 mm EDTA, 0.5 mm
DTT, 0.2 mm PMSF, 25% glycerol: pH 7.9) was added
T. Tanioka et al. Regulation of L-RAP ⁄ ERAP2 gene expression
FEBS Journal 272 (2005) 916–928 ª 2005 FEBS 925
stepwise, incubated at 4 °C for 30 min and centrifuged at
25 000 g for 45 min. Supernatant was then collected.
The nuclear extracts were incubated with 1 ng of radio-
labeled oligonucleotide corresponding to the IRF-E site in
binding buffer [10 mm Hepes, 50 mm KCl, 5 mm MgCl
2
,
0.5 mm EDTA, 5 mm DTT, 0.7 mm PMSF, 2 lgÆmL
)1

leu-
peptin, 2 lgÆmL
)1
pepstatin, 2 lgÆmL
)1
aprotinin, 10% gly-
cerol, 1 lgÆmL
)1
poly(dI-dC): pH 7.9] for 30 min at 25 °C.
For competition assay, 100-fold excess of unlabeled oligo-
nucleotide was added to the binding reaction prior to the
addition of the radiolabeled probe. For supershift analysis,
2 lg of the required antibody was added following the bind-
ing reaction for 1 h at 4 °C. These incubation mixtures were
then electrophoresed in 8% native polyacrylamide gel with
Tris ⁄ borate ⁄ EDTA buffer. The gel was dried and analyzed
with a BAS2000 Fuji photo film (Fuji, Kanagawa, Japan).
Western blot analysis
Test samples were separated by SDS ⁄ PAGE on an 8%
separating gel and transferred to poly(vinylidene difluo-
ride) membranes (Pall Corp, East Hills, NY, USA).
The membranes were blocked with Tris ⁄ HCl-buffered
saline (NaCl ⁄ Tris) (pH 7.4) containing 0.1% Tween-20
(NaCl ⁄ Tris ⁄ Tween) with 5% skimmed milk for 1 h at
room temperature, then incubated in NaCl ⁄ Tris ⁄ Tween,
5% skimmed milk, and 2.5 lgÆmL
)1
rabbit either anti-
(human IRF-1) IgG or anti-(human IRF-2) IgG for 2 h at
room temperature. The filter was washed three times with

NaCl ⁄ Tris ⁄ Tween and incubated for 1 h with horseradish
peroxidase-conjugated goat anti(rabbit IgG) IgG (Pro-
mega), diluted to 1 : 20 000 in NaCl ⁄ Tris ⁄ Tween contain-
ing 5% skimmed milk. After washing the filter three times
with NaCl ⁄ Tris ⁄ Tween, the blots were detected by an
enhanced chemiluminescence method using an ECL plus
Western blotting kit obtained from Amersham Biosciences.
The results were visualized by fluorography using RX-U
Fuji medical X-ray film.
Detection of L-RAP mRNA by RT-PCR
RT-PCR was preformed as described previously [46] using
the primer pair of 5¢-ATGACAAGTAACATGCTCGC-3¢
(sense) and 5¢-AATGAGTTGGTCCCATCCAT-3¢ (anti-
sense). PCR was carried out for 1 cycle at 95 °C for 9 min,
followed by 30 cycles of 94 °C for 30 s, 55 °C for 30 s, and
72 °C for 30 s.
Materials
Recombinant human IFN-c was obtained from PeproTech
(Rocky Hill, NJ, USA). Antibodies against human IRF-1
and IRF-2 were from Santa Cruz Biotechnology (Santa
Cruz, CA, USA).
Acknowledgement
This work was supported in part by grants-in-aid
(Nos. 15390032, 15790061 and 14771302) from the
Ministry of Education, Science, Sports and Culture of
Japan and a grant for ‘Chemical Biology Research
Program’ from RIKEN.
References
1 Taylor A (1993) Aminopeptidases: structure and func-
tion. FASEB J 7, 290–298.

2 Hooper NM (1994) Families of zinc metalloproteases.
FEBS Lett 354, 1–6.
3 Laustsen PG, Vang S & Kristensen T (2001) Mutational
analysis of the active site of human insulin-regulated
aminopeptidase. Eur J Biochem 268, 98–104.
4 Tanioka T, Hattori A, Masuda S, Nomura Y, Nakay-
ama H, Mizutani S & Tsujimoto M (2003) Human leu-
kocyte-derived arginine aminopeptidase: the third
member of the oxytocinase subfamily of aminopepti-
dases. J Biol Chem 278, 32275–32283.
5 Fujiwara H, Higuchi T, Yamada S, Hirano T, Sato Y,
Nishioka Y, Yoshioka S, Tatsumi K, Ueda M, Maeda
M & Fujii S (2004) Human extravillous trophoblasts
express laeverin, a novel protein that belongs to mem-
brane-bound gluzincin metallopeptidases. Biochem Bio-
phys Res Commun 313, 962–968.
6 Rogi T, Tsujimoto M, Nakazato H, Mizutani S &
Tomoda Y (1996) Human placental leucine aminopepti-
dase: a new member of type II membrane-spanning zinc
metallopeptidase family. J Biol Chem 271, 56–61.
7 Hattori A, Matsumoto H, Mizutani S & Tsujimoto M
(1999) Molecular cloning of adipocyte-derived leucine
aminopeptidase highly related to placental leucine
aminopeptidase ⁄ oxytocinase. J Biochem (Tokyo) 125,
931–938.
8 Schomburg L, Kollmus H, Friedrichsen S & Bauer K
(2000) Molecular characterization of a puromycin-insen-
sitive leucyl-specific aminopeptidase, PILS-AP. Eur J
Biochem 267, 3198–3207.
9 Saric T, Chang S-C, Hattori A, York IA, Markant S,

Rock KL, Tsujimoto M & Goldberg AL (2002) An
IFN-c-induced aminopeptidase in the ER, ERAPI,
trims precursors to MHC class I-presented peptides.
Nat Immunol 3, 1169–1176.
10 Tsujimoto M & Hattori A (2004) The oxytocinase sub-
family of M1 aminopeptidases. Biochim Biophys Acta
doi: 10.1016/j.bbapap.2004.09.011.
11 Keller SR (2003) The insulin-regulated aminopeptidase:
a companion and regulator of GLUT4. Front Biosci 8,
S410–S420.
12 Keller SR, Scott HM, Mastick CC, Aebersold R &
Lienhard GE (1995) Cloning and characterization of a
Regulation of L-RAP ⁄ ERAP2 gene expression T. Tanioka et al.
926 FEBS Journal 272 (2005) 916–928 ª 2005 FEBS
novel insulin-regulated membrane aminopeptidase from
Glut4 vesicles. J Biol Chem 270 , 23612–23618.
13 Nakamura H, Itakura A, Okamura M, Ito M, Iwase
A, Nakanishi Y, Okada M, Nagasaka T & Mizutani
S (2000) Oxytocin stimulates the translocation of
oxytocinase of human vascular endothelial cells via
activation of oxytocin receptors. Endocrinology 141,
4481–4485.
14 Masuda S, Hattori A, Matsumoto H, Miyazawa S,
Natori Y, Mizutani S & Tsujimoto M (2003) Involve-
ment of the V
2
receptor in vasopressin-stimulated
translocation of placental leucine aminopeptidase ⁄
oxytocinase in renal cells. Eur J Biochem 270, 1988–
1994.

15 Albiston AL, McDowall SG, Matsacos D, Sim P, Clune
E, Mustafa T, Lee J, Mendelsohn FAO, Simpson RJ,
Connolly LM & Chai SY (2001) Evidence that the
angiotensin IV (AT
4
) receptor is the enzyme insulin-
regulated aminopeptidase. J Biol Chem 276, 48623–
48626.
16 Albiston AL, Mustafa T, McDowall SG, Mendelsohn
FAO, Lee J & Chai SY (2003) AT
4
receptor is insulin-
regulated membrane aminopeptidase: potential mechan-
ism of memory enhancement. Trends Endocrinol Metab
14, 72–77.
17 Serwold T, Gonzalez F, Kim J, Jacob R & Shastri N
(2002) ERAAP customizes peptides for MHC class I
molecules in the endoplasmic reticulum. Nature 419,
480–483.
18 Yamamoto N, Nakayama J, Yamakawa-Kobayashi K,
Hamaguchi H, Miyazaki R & Arinami T (2002) Identi-
fication of 33 polymorphisms in the adipocyte-derived
leucine aminopeptidase (ALAP) gene and possible
association with hypertension. Hum Mutat 19, 251–
257.
19 Miyashita H, Yamazaki T, Akada T, Niizeki O, Ogawa
M, Nishikawa S & Sato Y (2002) A mouse orthologue
of puromycin-insensitive leucyl-specific aminopeptidase
is expressed in endothelial cells and plays am important
role in angiogenesis. Blood 99, 3241–3249.

20 Akada T, Yamazaki T, Miyashita H, Niizeki O, Abe M,
Sato A, Satomi S & Sato Y (2002) Puromycin insensitive
leucyl-specific aminopeptidase (PILSAP) is involved in
the activation of endothelial integrins. J Cell Physiol
193, 253–262.
21 Cui X, Hawari F, Alsaary S, Lawrence M, Combs CA,
Geng W, Rouhani FN, Miskinis D & Levine SJ (2002)
Identification of ARTS-1 as a novel TNFR1-binding
protein that promotes TNFR1 ectodomain shedding.
J Clin Invest 11, 515–526.
22 Cui X, Rouhani FN, Hawari F & Levine SJ (2003)
ARTS-1, is required for interleukin-6 receptor shedding.
J Biol Chem 278, 28677–28685.
23 Cui X, Rouhani FN, Hawari F & Levine SJ (2003)
Shedding of the type II IL-1 decoy receptors requires a
multifunctional aminopeptidase, aminopeptidase regula-
tor of TNF receptor type 1 shedding. J Immunol 171,
6814–6819.
24 Goldberg AL, Cascio P, Saric T & Rock KL (2002)
The importance of the proteasome and subsequent
proteolytic steps in the generation of antigenic peptide.
Mol Immunol 39, 147–164.
25 Hattori A & Tsujimoto M (2004) Processing of anti-
genic peptides by aminopeptidases. Biol Pharm Bull 27,
777–780.
26 Beninga J, Rock KL & Goldberg AL (1998) Inter-
feron-c can stimulate post-proteasomal trimming of
the N-terminus of an antigenic peptide by inducing
leucine aminopeptidase. J Biol Chem 273, 18734–
18742.

27 Shapiro MB & Senapathy P (1987) RNA splice junc-
tions of different classes of eukaryotes: sequence statis-
tics and functional implications in gene expression.
Nucleic Acids Res 15, 7155–7174.
28 Hattori A, Matsumoto K, Mizutani S & Tsujimoto M
(2001) Genomic organization of the human adipocyte-
derived leucine aminopeptidase gene and its relationship
to the placental leucine aminopeptidase ⁄ oxytocinase
gene. J Biochem (Tokyo) 130, 235–241.
29 Javahery R, Khachi A, Lo K, Zenzie-Gregory B &
Smale ST (1994) DNA sequence requirements for tran-
scriptional initiator activity in mammalian cells. Mol
Cell Biol 14, 116–127.
30 Faisst S & Meyer S (1992) Compilation of vertebrate-
encoded transcription factors. Nucleic Acids Res 20,
3–26.
31 Taniguchi T, Ogasawara K, Takaoka A & Tanaka N
(2001) IRF family of transcription factors as regulators
of host defense. Annu Rev Immunol 19, 623–650.
32 Oikawa T & Yamada T (2003) Molecular biology of the
Ets family of transcription factors. Gene 303, 11–34.
33 Rasmussen TE, Pedraza-Diaz S, Hardre R, Laustsen
PG, Carrion AG & Kristensen T (2000) Structure of the
human oxytocinase ⁄ insulin-regulated aminopeptidase
gene and localization to chromosome 5q21. Eur J Bio-
chem 267, 2297–2306.
34 Vaughan PS, Aziz F, van Wijnen AJ, Wu S, Harada H,
Taniguchi T, Soprano KJ, Stein JL & Stein GS (1995)
Activation of a cell-cycle-regulated histone gene by the
oncogenic transcription factor IRF-2. Nature 377, 362–

365.
35 Jesse TL, LaChance R & Iademarco MF & Dean DC
(1998) Interferon regulatory factor-2 is a transcriptional
activator in muscle where it regulates expression of vas-
cular cell adhesion molecule-1. J Cell Biol 140, 1265–
1276.
36 Sharrocks AD (2001) The ETS-domain transcription
factor family. Nat Rev 2, 827–837.
37 Tondravi MM, McKercher SR, Anderson K, Erdmann
JM, Quiroz M, Maki R & Teitelbaum SL (1997)
T. Tanioka et al. Regulation of L-RAP ⁄ ERAP2 gene expression
FEBS Journal 272 (2005) 916–928 ª 2005 FEBS 927
Osteopetrosis in mice lacking haematopoietic transcrip-
tion factor PU.1. Nature 386, 81–84.
38 Marecki S, Riendeau CJ, Liang MD & Fenton MJ
(2001) PU.1 and multiple IFN regulatory factor proteins
synergize to mediate transcriptional activation of the
human IL-1 b gene. J Immunol 166, 6829–6838.
39 Eklund EA, Jalava A & Kakar R (1998) PU.1, inter-
feron regulatory factor 1, and interferon consensus
sequence-binding protein cooperate to increase gp91
(phox) expression. J Biol Chem 273, 13957–13965.
40 Eklund EA & Kakar R (1999) Recruitment of CREB-
binding protein by PU.1, IFN-regulatory factor-1, and
the IFN consensus sequence-binding protein is necessary
for IFN-c-induced p67phox and gp91phox expression.
J Immunol 163, 6095–6105.
41 Gobin SJP, van Zutphen M, Woltman AM & van den
Elsen PJ (1999) Transactivation of classical and nonclas-
sical HLA class I genes through the IFN-stimulated

response element. J Immunol 163, 1428–1434.
42 Xi H, Eason DD, Ghosh D, Dovhey S, Wright KL
& Blanck G (1999) Co-occupancy of the interferon
regulatory element of the class II transactivator
(CIITA) type IV promoter by interferon regulatory
factors 1 and 2. Oncogene 18, 5889–5903.
43 van Gravesande KS, Layne MD, Ye Q, Le L, Baron RM,
Perrella MA, Santambrogio L, Silverman ES & Riese RJ
(2002) IFN regulatory factor-1 regulates IFN-c-depend-
ent cathepsin S expression. J Immunol 168, 4488–4494.
44 Hisamatsu T, Suzuki M & Podolsky DK (2003) Inter-
feron-c augments CARD4 ⁄ NOD1 gene and protein
expression through interferon regulatory factor-1 in
intestinal epithelial cells. J Biol Chem 278, 32962–32968.
45 Liu J, Cao S, Herman LM & Ma X (2003) Differential
regulation of interleukin (IL)-12 p35 and p40 gene expres-
sion and interferon (IFN)-c-primed IL-12 production by
IFN regulatory factor 1. J Exp Med 198, 1265–1276.
46 Matsumoto M, Hisatake K, Nogi Y & Tsujimoto M
(2001) Regulation of receptor activator of NF-jB lig-
and-induced tartrate-resistant acid phosphatase gene
expression by PU.1-interacting protein ⁄ interferon regu-
latory factor-4. Synergism with microphthalmia tran-
scription factor. J Biol Chem 276, 33086–33092.
928 FEBS Journal 272 (2005) 916–928 ª 2005 FEBS
Regulation of L-RAP ⁄ ERAP2 gene expression T. Tanioka et al.