Gene. 105 (1991) 221-227 0 1991 Elsevier Science Publishers B.V. All rights reserved. 0378-1119/91/$03.50
221
GENE 06000
Cloning and expression of a cDNA encoding mouse indoleamine 2,3-dioxygenase (Recombinant DNA; oxygenase; enzyme induction; tryptophan degradation; microbicidal; allograft rejection; interferon; CAMP)
Akemi Habara-Ohkuho,
Osamu Takikawa
and Ryotaro Yoshida
Department of Cell Biology, Osaka Bioscience Instituted Suita 565 (Japan) Received by A. Nakazawa: 18 March 1991 Revised/Accepted: 8 May/l 1 May 1991 Received at publishers: 18 June 1991
SUMMARY
The depletion of an essential animo acid (aa), tryptophan, caused by interferon-y (IFN-y)-mediated induction of indoleamine 2,3-dioxygenase (IDO) in mouse allografted tumor cells, has been suggested as a reason for the allograft rejection. To elucidate the mechanism of this ID0 induction, attempts were made to isolate cDNA clones encoding mouse IDO. In seven of 25 mouse cell lines, ID0 was induced by IFN-y, and the highest ID0 induction was observed in the case of rectal cancer (CMT-93) cells, which were further stimulated two- to threefold by the simultaneous addition of dibutyryl cyclic AMP (Bt,cAMP). A cDNA library was prepared from poly(A) + RNA isolated from CMT-93 cells treated with IFN-y/Bt,cAMP. The cDNA clones were isolated using the cDNA encoding human ID0 as a probe. The mouse ID0 cDNA encodes a 407-aa protein with an iW=of 45 639. The deduced aa sequence agreed with partial aa sequences derived from endopeptidase digestion of purified mouse ID0 and revealed 6 1y0 homology with that of human IDO. Transient expression of the mouse ID0 cDNA in COS-7 cells yielded a high level of ID0 activity in the cells. Northern hybridization analysis of RNA in CMT-93 cells indicated that IFN-y induced the ID0 mRNA, and that the level of RNA was increased by simultaneous addition of Bt,cAMP, while Bt,cAMP itself had no effect on mRNA induction.
INTRODUCTION
ID0 is a hemoprotein (Shiiizu et al., 1978) that catalyzes the degradation of tryptophan to N-formylkynurenine (Takikawa et al., 1986). This enzyme is widely
C~~espo~e~e to: Dr. R. Yoshida, Department of Cell Biology, Osaka Bioscience Institute, 6-2-4 Furuedai, Suita 565 (Japan) Tel. (81-6)872-4817; Fax(81-6)872-4818. Abbreviations: aa, amino acid(s); bp, base pair(s); BtacAMP, dibutyryl CAMP; CAMP, cyclic AMP; cDNA, DNA complementary to RNA; HPLC, high-performance liquid chromatography; IDO, indoleamine 2,3dioxygenase; ZDO, gene (DNA) encoding IDO; IFN, interferon; i.p., intraperitoneal; kb, kilobase or 1000 bp; nt, nucleotide(s); ORF, open reading frame; SDS, sodium dodecyl sulfate; SSC, 0.15 M NaCI/O.OIS M Na, . citrate pH 7.6.
distributed in various organs of mammals and is induced by IFNs or IFN-inducers (Yoshida and Hayaishi, 1978; Yoshida et al., 1979; 1981; 1986; Pfefferkom, 1984; Yasui et al., 1986; Byrne et al., 1986; Takikawa et al., 1988). In particular, in a variety of cultured human cell lines, ID0 is markedly induced by IFN-7. Recently, Tone et al. (1990) and Dai and Gupta (1990) have isolated a cDNA encoding human IDO. In 1988, we reported that ID0 was induced in tumor cells undergoing rejection but not in infiltrating host cells of mouse after i.p. transplantation of allogeneic tumor cells (Yoshida et al., 1988). Subsequently, IFN-y was shown to be the inducer of this enzyme (Takikawa et al., 1990). To gain further insight into the cellular mechanism of ID0 induction by IFN-7 and the role played by ID0 in allog& rejection, cDNA clones encoding mouse ID0 were pre-
222 pared from poly(A) + RNA of mouse rectal cancer cells using human ID0 cDNA as a probe. The primary structure of the protein was deduced from the nt sequence, compared with that of human ID0 cDNA, and the identity of the protein as mouse ID0 was confirmed by the transient expression
of cDNA
IFN-/I
(data
not
shown).
The
highest
enzyme
activity
(167.6 nmol/h/mg of protein) was obtained with IFN-ytreated CMT-93 cells. Since synergistic effects of phorbol ester and IFN-7 on ID0 induction in human THP-1 cells have been reported by Edelstein et al. (1989) effects of various agents modifying the activities of protein kinases on the ID0 activity in mouse CMT-93 cells were examined. The ID0 activity of
in COS-7 cells.
(a) Induction of ID0 by IFN-y in various mouse cell lines and cDNA cloning from IFN-y-treated CMT-93 cells
IFN-y-treated CMT-93 cells was further enhanced approx. threefold upon simultaneous addition of Bt,cAMP or 8-Br-CAMP (Fig. l), whereas pre- or post-treatment with these agents had little, if any, enhancing effect (data not
To determine the best source for preparing a cDNA library for mouse IDO, 25 mouse cell lines were incubated
shown). Phorbol ester and Na * butyrate had almost no effect on IFN-y-mediated ID0 induction, and phorbol
with IFN-y, IFN-a or IFN-/I for three days. In seven of 25 cell lines, ID0 was induced by IFN-y (Table I), whereas no ID0 activity was detected in these cells either before incubation with IFN-y or after incubation with IFN-a or
ester was inhibitory
RESULTS AND DISCUSSION
at high concentrations.
Therefore,
a
LZap II cDNA library was constructed from poly(A)+ RNA of CMT-93 cells harvested on day 3 after treatment with IFN-y/Bt,cAMP (Chomczynski and Sacchi, 1987).
TABLE I ID0 induction by IFN-1 in various mouse cells Cell line”
Morphology b
Origin
ID0 activity”
CMT-93 Meth A Colon 26 MMT L-929 NMuMG C 127 I Bl6 3LL CL-S1 LA-4 KLN 205 Balb-MC FM-3A Y-1 I-10 F9-41 Ehrlich Balb/3T12-3 Balb/3T3 McCoy C3H/lOT1/2 P388D, Yac-1 L1210
E F E E F E E E E E E E E E E E E E F F F F L L L
Rectum, carcinoma Fibrosarcoma Rectum, carcinoma Mammary gland, carcinoma Connective tissue, normal Mammary gland, normal Mammary tumor Melanoma Lung, carcinoma Mammary alveolar nodules, preneoplastic Lung, adenoma Squamous cell, carcinoma Mammary tumor Mammary tumor Adrenal tumor Leydig cell testicular tumor Teratocarcinoma Ascites tumor Embryo, tumorigenic Embryo, nontumorigenic Connective tissue Embryo, nontumorigenic Lymphoid neoplasm Lymphoma Lymphoma
(nmol of kynurenine/h/mg of protein) 167.6 30.0 26.9 1.4 1.4 0.5 0.4 n.d. nd. n.d. n.d. nd. n.d. n.d. n.d. n.d. n.d. n.d. nd. n.d. n.d. n.d. n.d. n.d. n.d.
a MMT, CLSl, Balb-MC, FM-3A, Y-l, I-10, F9-41, Balb/3T3 and C3H/lOT1/2 cells were obtained from the Japanese Cancer Research Resources Bank (Tokyo). Other cells were purchased from Dainippon Pharmaceutical (Osaka). ’ E, epithelial cell; F, fibroblastic cell; L, lymphoblastic cell. ’ Cells at almost confluence were cultured in a 35-mm diameter plastic dish with 2 ml of medium containing 1000 units/ml of IFN-y under 5% CO1 and 95% air. After 72 h of culturing, cells were harvested by trypsinixation, and the ID0 activity was assayed by HPLC as described (Takiiawa et al., 1988). Protein was determined by the method of Lowry et al. (1951) with bovine serum albumin as a standard. Each value represents the mean of duplicate dishes except for CMT-93 cells (n = 7). n.d., nondetectable (~0.1).
223 P-45Os, hemoglobin, cyclooxygenase and hepatic phan 2,3-dioxygenase, except for human IDO.
Relative specific activity of ID0 (%)
IFN-Y (1000 units/ml)
(c) Alignment human IDOs
IFN-T + Bt$2AMP (0.1 mm) IFN-Y + B&CAMP (1 mM) IFN-Y + &
(3 mM)
IFN-Y + BrcAMP (1 mM) IFN-Y + BrcAMP (3 mM) IFN-Y + TPA (0.1 PM)
1990). Particularly, aa 101-184 of mouse ID0 were highly homologous with the human counterpart (aa 96-181) the identity being 89%. Moreover, since this region
IFN-Y + TPA (1 FM) IFN-T + Nahttylate
(0.1 mM) (1 “IM)
Fig. 1. Effects of various agents modifying the activities of protein kinases on IFN-y-mediated ID0 induction. CMT-93 cells at nearconfluence were cultured in a plastic dish (35 mm) with 2 ml of medium containing 1000 units/ml of IFN-7 (4 x lo6 units/mg of protein) and various agents under 5 y0 CO, and 95 y0 air. After 72 h culture, the cells were harvested by trypsinixation and the ID0 activity was determined as described in Table I, footnote c. The ID0 activity induced by IFN-), (1000 units/ml) alone was 167.6 k 41.1 mnol kynurenine formed/h/mg of protein (mean f SD, n = 7). The value represents 100% activity. Data (relative to IFN-), treatment) are expressed as mean k SEM of seven cultures of three different experiments. BrcAMP = 8-Br-CAMP.
On screening, with human ID0 cDNA 2.9 x 10’ phages yielded 33 positive clones.
of the deduced aa sequences of mouse and
As shown in Fig. 3, the deduced aa sequence of mouse ID0 (407 aa) revealed a 61% overall homology to that of human ID0 (403 aa) (Tone et al., 1990; Dai and Gupta,
IFN-Y + BrcAMP (0.1 mm)
,FN--, + Ndutytate
trypto-
as
a probe,
(b) Nucleotide sequence of mouse ID0 cDNA These positive clones contained cDNA inserts of 1.2-1.5 kb, which were each slightly different in the 5’- or 3’-untranslated regions. The nt sequences of two independent clones, pBS25 and pBS27, were found to be identical in the overlapping regions. The cDNA sequence was 1555 bp long and contained an ORF of 1221 nt (nt 71-1291) flanked by 70 bp and 261 bp of 5’- and 3’untranslated regions, respectively (Fig. 2). The 3’-untranslated region was highly A + T-rich and the putative polyadenylation signal, AATAAA, was present at 244 bp downstream from the stop codon (nt 1292-1294). The aa sequence deduced from the nt sequence contained partial aa sequences of endopeptidase digests of purified mouse ID0 (Fig. 2, underlined). In addition, the sequence, 5’-GCAATGG, near the putative start codon, ATG, agreed with the optimal sequence of Kozak (1986) for eukaryotic mRNAs. Therefore, the ORF encodes a protein of 407 aa with an A4, of 45 639. In the coding region, there was no potential Asn-linked glycosylation site (Asn-XaaThr/Ser), suggesting that, in contrast to the purified enzyme from rabbit intestine (Shimizu et al., 1978), the mouse ID0 may not be a glycoprotein. When the sequence of mouse IDO, a heme-containing enzyme, was compared with the GenBank and NBRF databases, no significant homology was found with the nt or aa sequences of other hemoproteins such as cytochrome
(aa 101-184) was rich in hydrophobic aa residues (Fig. 2) it might function as a catalytic domain. The whole hydropathy pattern of mouse ID0 was very similar to that of human ID0 (data not shown). (d) Expression of mouse ID0 cDNA in COS-7 cells To confirm that the cDNA clone actually encodes functional mouse IDO, the cDNA was inserted with two orientations (sense and antisense) into the expression vector (pdKCR) (Fukunaga et al., 1984), and was transfected into COS-7 cells. As shown in Fig. 4, the COS-7 cells transfected with the pdKCR-MZDOsense yielded a high level (400 nmol/h/mg) of ID0 activity in the cells, while no enzyme activity was detected in the cells transfected with the pdKCR-M ID0 antisense. The expression of functional ID0 in COS-7 cells transfected with pdKCR-MIDOsense was confirmed by a decrease in tryptophan concentration or increase in the amount of kynurenine, an IDO-mediated metabolite of tryptophan, in the culture medium. In the case of pdKCR-M ID0 antisense, however, there was no change in the metabolism of tryptophan to kynurenine. Furthermore, the ID0 expressed by pdKCR-M ID0 sense had the same catalytic properties as those of native IDO; both methylene blue and ascorbic acid were absolutely required for enzyme activity (data not shown). (e) Time course of induction of ZDO mRNA and ID0 activity by IFN-7 in CMT-93 cells To explore the mechanisms of ID0 induction by IFN-y/Bt,cAMP, the amounts of ID0 mRNA and the ID0 activities in CMT-93 cells were monitored at various time intervals after treatment with IFN-y, Bt,cAMP or a combination of both. Fig. 5A shows a Northern-blot analysis of total cellular RNAs using mouse ID0 cDNA as a probe. Although no ID0 mRNA was detected in untreated CMT-93 cells, the level of mRNA began to increase within 24 h after IFN-y treatment and reached a maximum on day 2 (Fig. 5B). The level of mRNA gradually declined and was undetectable on day 5. Similar time-dependent changes in the levels of ID0 mRNA were seen with the IFN-y/Bt,cAMP-treated CMT-93 cells, and the maximum level was 1.5- to twofold higher than that of IFN-y-treated
224 GCThMCCCThThkGMhTGCTGGGhChTTCCTTCAGTGGCChhGTGGGGGG'KAGTGGhGThGhChGCh7G AT0 GCA CTC hGT hhh ATh TCT CCT hCh GM GGT TCT hGh hGG hTC CTT Ghh GhC ChC 127 I PlrtAla Lou -I Lye 118 9er Pro Thr Glu Gly S-r k9 A%i II* h" Gl" AW iii, CACATAGAT GAAGATGTG GGC TTT GCT CTACCACATCCACTC GTG GAGCTG CCC GAC 164 20 ,,I.Zl. A,p Cl" A.p Val Gly P,,.hl9 L.u Pro Hi. Pro L.U Vd 01" LO" PI0 h.p GCh TAC AGC CCC TOG OTC CTT GTG GCT hGh MT 39 Ala Tyr 6.r Pro Trp VSl In" Val Ala &g Am
CTG CCT OTC CT0 ATT GU3 AAI:GGG 241 Lo" Pro '?a1L*" Il. Cl" h.n Gly
~CTTCWLDMOAAGTT~Al*;Cn:CCCh~Cn;wCAcG~~Clr;~u;r299 39 Gin Mu hr9 Cl" Glu Vrl Cl" Ly, Lo" Pro Thr Lo" 6.r Thr Up CAC AGGTTACAO CGC CTGGCACAC 77 tliahrvgIau Gin hrG Le" Ala Hia
01~ LO" -01~
CTt3GCC CTE G.2GTM: ATChCChTGGCG TATGTG 355 L8” Air Is+” Gly Tyr I19 Tbt Mwt Al& Tyr Vd
TGG hhC CGhGGG GhT GhC GhTGTTCGhMGGn; CTG CCC CGCiUT ATT GCTGTTCCC 412 96 Trp hen hrG Gly hap A.p A.p Val &9 Ly, Val L." Pro Ar9 h.8,11a Ala Vd PrO TAC TGC GAG CTC TCA WhhG TTG GGC CTG CCT CCTATT Cn: ET TAT GChGhC TGT 469 115 Tyr Cye Cl" La" 6.r Cl" Lya IA" Gly Leu Pro Pro Ila Lo" SO):Tyr Ala h.p Cy9 OTC CTG GCh hhC TGG MO hU hhG GhC CCC MT 134 VSl La" Ala hen Trp Ly8 Lye Lye AspPro Am
GGG CCC ATG hC& ThC GAO h&C ATG 526 Gly Pro blot Thr Tyr Cl” Aon Ict
GAC hTTCTG TTC TChTTT CCTGGTGGG GhC TGC GhC MOGGC TTC TTC CTC GTC TCT 593 153 Asp Ila Lo" Ph. G.r Phe Pro Gly Gly Aep Cys Amp Lye Gly PlM Pha In" Val Sot
CTA TTG GTG GAA ATC Wh GCT TCT CCT GCA ATC hhh GCh ATC CCC hCT GTA WC AGT 640 172 Leu Lw Val 0111IlS Air Ala S-r Pro Ala Ila Lye hla 11. Pro Thr Val S@r SW GCA GTA GAG COT CAA GAC CT0 AAA GCA TTG OM AAG GCA CTG CM: GhC ATA GCT hCC 697 191 Ala Val Cl" Ax9 Gln ASP La" Lys Ala Lou Glu Lys Ala La" Hi8 hep Il* Ale Thr AGTCW~AMOCCMCGMhTT~~G~hn;CGTG~PTTGTEGllCCCA~C734 210 Ser Ie" Cl" Lye hla Lys Cl" Xl* Phe Lys hrg Mnt hr9 Asp PM
V11 h*p Pro -p
ACG TTT TTC ChC OTT CTC CCC ATA ThT CTG TCT GGC TCG AM MC hGC TCC hhG CT0 811 229 Thr Pha Ph. Hia Val LOU Ar9 Ile Tyr Iru Sex Gly Trp Ly. Cya 8~ Ser Lys -LDU CCA GAA GGT CTG CT0 TAT GhG GGG OTC TOG GhC hCC CCA MA ATG TTT TCh GGG GGC 968 249 Pro Cl" Gly Lm" La" Tyr Cl" Gly Val Trp Alp Thr Pro Lym BlStPho Sar Gly Gly AGT GCh GGC ChG WC 1IM:ATC TTC ChG ROT CTT GhT GTC CTT CTG GGh ATA h&A ChC 923 267 $0~ hla Gly Gln Ser So? Ile Phe Gin 9.r Lo" hap V&l Leu Leu Gly Ile Lya lie GADOCTM;C~OM~TCCT~OAA.m:CTc~~hn:lu;hou;TlrchTcccT992 286 01~ hla Gly Lye Cl" SW Pro Ala Cl" Ph. La" Cln Cl" Met Ar9 Cl" Tyr Mat Pro CCA GCC CAC CGG MC TTC CTT TTC TTC TTA GAO TCh GCT CCC CCh OTC CGT GhG TTT 1039 305 Pro Ala Him hr9 ha" Ph. LW Pha Phe Lau Glu 8-r Ala Pro Pro Val hrq Cl" Ph.
OTC ATT TCAAGACAC MT 324 Val 11. Bar AtQ HI. Am
GhhGhC TTG hCGAMGCT ThTAAC GSR TGTGTG hhT GGT1096 Cl" Asp 1-u Thr Lye Ala Tyr han Glu Cys Vsl &n Gly
CTG OTC TCT GTG MA AAG TTC CAC C'K GCA ATA GTA GAT ACT TAC ATT ATG hRh CCT 1153 343 &WI Val Ser Val Arq Lys Phm Him Inu Ala 11s V&l Aep The Ty? Il. @IatLps Pro TCG 8.AG A&G AAG CCC ACT
362 SW
OAT GGC G&C h?.GTCG CM GhG CCC TCh AAT GTG Ghh hGC 1210 Lye Ly* Lye Pzo Thr Asp Gly ABP Lya Ssr Cl" Cl" Pro SW Am Val Cl" Sar
AGAGGG ACT GGG GGTACGMTCCChTGhCT TTC CTAAGGAGTGTGAAAGATACAACC 1267 381 Arg Gly Thr Gly Gly Thr Am Pro Met Thr Phe Las Ax9 Sar Val Lya Asp Thr Tht GAGAAAGCT CTT CTG ACT TGG CCT ThG 390 Cl" Lyn Ala Le" Ldu 6.r Trp PTO '**
1294
TGThGChhGCTCcACTT%ZTh~ CTTCXKCTGFCATAGCTCATTAMTTCAGATCCACCA 1369 TchlcZ 1444 hGThhGhCThThGhhThMTTGCCTUTTGThT~hWThGkMT~~~;MCT~TTT~~TA~T~~TAT~T~T~T~~ATA~T~T~T~T~ ChhchhTGhGhcCChGTBTThTTGTGhA
1919
1555
Fig. 2. Nucleotide sequence of mouse ID0 cDNA and deduced aa sequence. A cDNA library was ligated with the vector AZap (Stratagene, La Jolla, CA) at the EcoRI site. The recombinant DNA was packaged in vitro (Sambrook et al., 1989) and the phages were grown in E. co/i X.1-Blue. The recombinant phages were screened by plaque hybridization (Benton and Davis, 1977) with the l.l-kb ofEcoRLEcoT141 fragment of human ID0 cDNA. which was labeled with 32P by the random primer method (Feinberg and Vogelstein, 1983). Hyb~dization was carried out by the method of Wahl et al. (1979), except that the hybridization temperature was lowered to 32°C. The pBluescript SK{ - ) plasmid DNAs from the positive clones were recovered using helper phage R408. The nt sequences were determined by the dideoxy chain-termination method (Sanger et al,, 1977). The GenBank accession No. for this sequence is M69109. Numbers on the right margin refer to nt position, and numbers on the left margin refer to aa position, The putative polyadenylation signal (AATAAA) is boxed. Those aa which were confirmed by protein sequencing of purified mouse IDO are underlined. Mouse IDO was purified from the epididymis by the method of K. Nakata, Y. Watanabe and 0. Hayaishi (details to be published elsewhere). The purified IDO was digested with @syl endopeptidase (Tsunasawa et al., 1989) and the fragments were sequenced with an Applied Biosystem model 477A Protein Sequencer (Foster City, CA).
cells. Although the time course of ID0 mRNA showed a pattern similar to that of ID0 activity (Fig. X2), the former preceded the latter by approx. one day. Bt,cAMP itself, however, induced neither 11)O mRNA nor ID0 activity in CMT-93 cells. These observations suggested that the ID0
mRNA was induced by IFN-), in CMT-93 cells, and that mRNA induction by IFN-y was enhanced by the addition of Bt,cAMP. Since Northern-blot analysis only indicates the steady-state level of ID0 mRNA, the precise molecular basis of transcriptional control in CMT-93 cells (increase
225
A No treatment Bt 2CAMP
MID0
-SPAIKAIPT”SSA”---ER*D-LKALEKALHDIATS-LEKAKEIFKRMRDFVDPDTFFH
232
*IDO
b._...“....~K.MQ~Q.._.T.__._...~E.._.C....~Q”.”QIH.H.N.KA..S
*2*
IFN-Y IFN- Y+Bt 2 CAMP 012345
Culture
MID0
KALLSWP
407
HID0
.S,.KEG
403
Fig. 3. Comparison
of aa sequences
sequence
ID0
of mouse
(HIDO). mouse
Dots represent IDO; dashes
of mouse and human
(MIDO)
is aligned
aa residues
represent
of human
IDOs. The aa
with that of human ID0 identical
ID0
to those of
gaps.
A
B
Culture mock
-
pdKCR
_
Days
Days
‘3: 150.
pdKCR-Mm (anlironre) t
loo-
pdKCR-ME (4ense) 0
100 ID0
Fig. 4. Transient
zoo Activity
300
site of a eukaryotic
1984). The intact
region
DNA polymerase using BarnHI
of mouse
and inserted
adapters.
alone were transferred
20
40
ID0
(nt 65-1364)
cDNA,
These constructs into COS-7
(antisense
the
with T4 pdKCR,
and sense) or vector
cells by the DEAE-dextran
method
(Sambrook et al., 1989). (A) ID0 activity (hatched bar) in the supernatant fraction ofcell lysate. The ID0 activity was determined as described in Table I, footnote kynurenine of tryptophan HPLC
c.(B) Concentrations
(blackened
and kynurenine
as described
oftryptophan
boxes) in the culture (Takikawa
(open boxes) and
media. The concentrations
in the culture medium
1
2
were measured
by
et al., 1988).
4
Days
cells (2 x lo5 cells/ml) were cultured
the medium alone (0) or medium containing 3 mM Bt,cAMP analysis.
( x ), or a combination
for one to live days with 1000 units/ml of IFN-y(m),
of both (0).
On day 3 after the administration
phoresis
(Sambrook
GeneScreen
to 1.2% agarose/2.2 et al.,
the blot was incubated NaCl/lO%
1989) After
Plus filter (DuPont-New dextran
at 42°C
sulfate/l
(A) Northern-blot
of IFN-)I and/or
total RNA of an aliquot (S/10) of cell suspension of RNA were subjected
5
of mouse ID0 mRNA and ID0 activity.
Fig. 5. Time course of induction CMT-93
in a mixture containing
in the rate of transcription, stabilization of mRNA, etc.) remains to be elucidated. As shown in Fig. 5, the induction of ID0 in mouse CMT-93 cells was regulated at the ID0 mRNA level, and two days were needed to reach the maximum level of IFN-y-induced ID0 mRNA. Furthermore, preliminary
3
Culture
et al.,
containing plasmid,
0
into the
(Fukunaga
was blunt-ended
into a BamHI-cleaved
(PM)
in CO.%7 cells. The
vector pdKCR
fragment
60
or Kynurenine
SK( - ) plasmid was subcloned
expression
1.3-kb SmaI-BsmI
coding
0
Tryptophan
of mouse ID0 cDNA
expression
mouse cDNA of the pBluescript BarnHI
400
(nmol/h/mg)
M formaldehyde transferring
England
in a mixture
Bt,cAMP,
was isolated, the
Nuclear)
and 20 ng gel electroRNA
of 50% formamide/l
% SDS, for 6 h, and then hybridized
both denatured
to a
with 20 x SSC, M
at 42°C
salmon sperm DNA (100 ng/ml)
and denatured “P-labeled 0.95-kb _EcoRI fragment of mouse ID0 cDNA probe. (B) Relative levels of ID0 mRNA. The autoradiogram of a Northern Tokyo)
blot was scanned to determine
aliquot of cell suspension lysate and the ID0 Table I, footnote ments.
with a BAS2000
relative
amounts
was disrupted
activity
Image Analyzer
of mRNA.
(C) ID0
by sonication
in the cells was determined
c. Each value represents
(Fuji Film, activity.
to prepare
as described
the mean of duplicate
An
the cell in
experi-
226 experiments from our laboratory have demonstrated that cycloheximide, a protein synthesis inhibitor, completely blocked the IFN-y-mediated induction of ID0 mRNA, suggesting that de novo synthesis of protein(s) was essential for mRNA induction by IFN-y. Indeed, Caplen and Gupta (1988) recently suggested that (a) newly synthesized protein(s) is (are) required for the IFN-y-mediated induction of human ID0 mRNA. In addition to B& (Fig. 4), other investigators have shown that various kinds of agents such as tumor necrosis factor-a (in mouse lung; Bianchi et al., 1988) and phorbol ester (in human monocytic leukemia cells; Edelstein et al., 1989) enhance ID0 induction by IFN-y. Therefore, several signaling pathways may be involved in ID0 induction by IFN-y.
Byrne, G.I., Lehmann, catabolism Immun. Caplen,
G.J.: Induction
of tryptophan
for gamma-interferon-mediated
Chlumydia psittaci replication
tion of intracellular
inhibi-
in T24 cells. Infect.
53 (1986) 347-351.
H.S. and Gupta,
by human
S.L.: Differential
interferon-y
regulation
and interferon-a.
of a cellular gene
J. Biol. Chem. 263 (1988)
332-339. Chomczynski, by
P. and Sacchi,
acid
guanidinium
Analyt.
Biochem.
N.: Single-step
S.L.: Molecular
cloning,
of human interferon-y-inducible Biochem. Edelstein,
method
of RNA isolation
thiocyanate-phenol-chloroform
Biophys.
sequencing
indoleamine
Res. Commun.
M.P., Ozaki, Y. and Duch, D.S.: Synergistic
THP-1 monocytic
leukemia
endonuclease
Biochem. Fukunaga,
B.: A technique fragments
2,3-dioxygenase
in
143 (1989) 2969-2973. for radiolabeling
to high specific
DNA
activity.
Anal.
132 (1983) 6-13.
R., Sokawa,
Y. and Nagata,
human
interferons
vector.
Proc. Natl. Acad.
S.: Constitutive
production
by mouse cells with bovine papillomavirus
Kozak, M.: Point mutations codon
cDNA.
effects of phorbol
of indoleamine
cells. J. Immunol.
A.P. and Vogelstein,
and expression
2,3_dioxygenase
168 (1990) l-8.
ester and IFN-)I on the induction Feinberg,
extraction.
162 (1987) 156-159.
Dai, W. and Gupta,
restriction
(f) Conclusions (I) A mouse ID0 cDNA was cloned from IFN-y/ Bt,cAMP-treated CMT-93 cells using human ID0 cDNA as a probe. (2) The mouse ID0 cDNA was 1555 bp long and encoded 407 aa residues with an h4,. of 45639. (3) The deduced aa sequence of mouse ID0 revealed a 61% overall homology with that of human IDO. (4) Transient expression of mouse ID0 cDNA in COS-7 cells yielded a high level of ID0 activity. (5) Northern hybridization analysis indicated that IFN-), induced the ID0 mRNA, and that the level of RNA was further increased by simultaneous addition of Bt,cAMP, while Bt,cAMP itself had no effect on mRNA induction.
L.K. and Landry,
is the mechanism
Sci. USA 81 (1984) 5086-5090. define a sequence
that modulates
of as a
translation
flanking the AUG initiator
by eukaryotic
ribosomes.
Cell 44
(1986) 283-292. Lowry,
O.H., Rosebrough,
measurement
N.J., Farr,
A.L. and Randall,
R.J.: Protein
with the Folin phenol reagent. J. Biol. Chem. 193 (1951)
265-275. Pfefferkorn, human
by inducing
Proc. Natl. Acad. Sambrook,
E.F. and Maniatis,
Manual,
terminating
T.: Molecular
2nd ed. Cold Spring Harbor
Cold Spring Harbor, F., Nicklen,
the host cells to degrade
tryptophan.
Sci. USA 81 (1984) 908-912.
J., Fritsch,
Laboratory Sanger,
y blocks the growth of Toxoplasma gondii in
E.R.: Interferon libroblasts
Cloning.
Laboratory
A
Press,
NY, 1989.
S. and Coulson,
inhibitors.
Proc.
A.R.: DNA sequencing Natl.
Acad.
Sci.
with chain-
USA
74 (1977)
5463-5467. Shimizu,
T., Nomiyama,
2,3-dioxygenase:
ACKNOWLEDGEMENTS
S., Hirata,
purification
F. and Hayaishi,
and some properties.
0.: Indoleamine J. Biol. Chem. 253
(1978) 4700-4706. Takikawa,
The authors are indebted to Dr. 0. Hayaishi for his continued encouragement during this investigation. The authors also thank Shionogi Pharmaceutical (Osaka, Japan) for kind providing IFN-y, Dr. S. Nagata of this Institute for pertinent discussion, and Ms. K. Hirata for aid in the preparation of the manuscript. This work was supported in part by research grants from the Sankyo Foundation of Life Science and Grant-in Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan.
O., Yoshida,
degradation
R., Kido,
in mice initiated
R. and Hayaishi,
by indoleamine
0.:
Tryptophan
2,3-dioxygenase.
J. Biol.
Chem. 261 (1986) 3648-3653. Takikawa,
O., Kuroiwa,
interferon-y
T., Yamazaki,
F. and Kido, R.: Mechanism
action: characterization
in cultured
human cells induced
enzyme-mediated
tryptophan
of indoleamine
by interferon-y degradation
of
2,3dioxygenase
and evaluation
in its anticellular
of the activity.
J. Biol. Chem. 263 (1988) 2041-2048. Takikawa,
O., Habara-Ohkubo,
of indoleamine rejection.
A. and Yoshida,
2,3-dioxygenase
J. Immunol.
in allogralted
R.: IFN-yis
the inducer
tumor cells undergoing
145 (1990) 1246-1250.
Tone, S., Takikawa, O., Habara-Ohkubo, A., Kadoya, A., Yoshida, R. and Kido, R.: Primary structure of human indoleamine 2,3dioxygenase deduced
from the nucleotide
sequence
of its cDNA.
Nucleic
Acids Res. 18 (1990) 367. Tsunasawa,
REFERENCES
S., Ma&i,
The primary Benton,
W.D. and Davis,
hybridization
R.W.: Screening
to single plaques
Igt recombinant
in situ. Science
clones by
196 (1977) 180-182.
Bianchi, M., Bertini, R. and Ghezzi, P.: Induction of indoleamine 2,3dioxygenase by interferon in mice: a study with different recombinant interferons and various cytokines. 152 (1988) 237-242.
Biochem.
Biophys.
Res. Commun.
structure
T., Hirose,
M., Soejima,
lyticus protease I, a lysine-specific (1989) 3832-3839.
M. and Sakiyama,
F.:
characteristics
ofAchromobacter
serine protease.
J. Biol. Chem. 264
and structural
Wahl, G.M., Stem, M. and Stark, G.R.: Efficient transfer of large DNA fragments from agarose gels to diazobenzyloxymethyl-paper and rapid hybridization by using dextran USA 76 (1979) 3683-3687.
sulfate.
Proc. Natl. Acad.
Sci.
227 Yasui, H., Takai, K., Yoshida, R. and Hayaishi, 0.: Interferon enhances tryptophan metabolism by inducing pulmonary indoleamine 2,3dioxygenase: its possible occurrence in cancer patients. Proc. Natl. Acad. Sci. USA 83 (1986) 6622-6626. Yoshida, R. and Hayaishi, 0.: Induction of pulmonary indoleamine 2,3dioxygenase by intraperitoneal injection of bacterial lipopolysaccharide. Proc. Natl. Acad. Sci. USA 75 (1978) 3998-4000. Yoshida, R., Urade, Y., Tokuda, M. and Hayaishi, 0.: Induction of indoleamine 2,3-dioxygenase in mouse lung during virus infection. Proc. Natl. Acad. Sci. USA 76 (1979) 4048-4086.
Yoshida, R., Imanishi, J., Oku, T., Kishida, T. and Hayaishi, 0.: Induction of pulmonary indoleamine 2,3-dioxygenase by interferon. Proc. Natl. Acad. Sci. USA 78 (1981) 129-132. Yoshida, R., Oku, T., Imanishi, J., Kishida, T. and Hayaishi, 0.: Interferon: a mediator of indoleamine 2,3-dioxygenase induction by lipopolysaccharide, poly(1) *poly(C) and pokeweed mitogen in mouse lung. Arch. Biochem. Biophys. 249 (1986) 596-604. Yoshida, R., Park, S.W., Yasui, H. and Takikawa, 0.: Tryptophan degradation in transplanted tumor cells undergoing rejection. J. Immunol. 141 (1988) 2819-2823.
221
GENE 06000
Cloning and expression of a cDNA encoding mouse indoleamine 2,3-dioxygenase (Recombinant DNA; oxygenase; enzyme induction; tryptophan degradation; microbicidal; allograft rejection; interferon; CAMP)
Akemi Habara-Ohkuho,
Osamu Takikawa
and Ryotaro Yoshida
Department of Cell Biology, Osaka Bioscience Instituted Suita 565 (Japan) Received by A. Nakazawa: 18 March 1991 Revised/Accepted: 8 May/l 1 May 1991 Received at publishers: 18 June 1991
SUMMARY
The depletion of an essential animo acid (aa), tryptophan, caused by interferon-y (IFN-y)-mediated induction of indoleamine 2,3-dioxygenase (IDO) in mouse allografted tumor cells, has been suggested as a reason for the allograft rejection. To elucidate the mechanism of this ID0 induction, attempts were made to isolate cDNA clones encoding mouse IDO. In seven of 25 mouse cell lines, ID0 was induced by IFN-y, and the highest ID0 induction was observed in the case of rectal cancer (CMT-93) cells, which were further stimulated two- to threefold by the simultaneous addition of dibutyryl cyclic AMP (Bt,cAMP). A cDNA library was prepared from poly(A) + RNA isolated from CMT-93 cells treated with IFN-y/Bt,cAMP. The cDNA clones were isolated using the cDNA encoding human ID0 as a probe. The mouse ID0 cDNA encodes a 407-aa protein with an iW=of 45 639. The deduced aa sequence agreed with partial aa sequences derived from endopeptidase digestion of purified mouse ID0 and revealed 6 1y0 homology with that of human IDO. Transient expression of the mouse ID0 cDNA in COS-7 cells yielded a high level of ID0 activity in the cells. Northern hybridization analysis of RNA in CMT-93 cells indicated that IFN-y induced the ID0 mRNA, and that the level of RNA was increased by simultaneous addition of Bt,cAMP, while Bt,cAMP itself had no effect on mRNA induction.
INTRODUCTION
ID0 is a hemoprotein (Shiiizu et al., 1978) that catalyzes the degradation of tryptophan to N-formylkynurenine (Takikawa et al., 1986). This enzyme is widely
C~~espo~e~e to: Dr. R. Yoshida, Department of Cell Biology, Osaka Bioscience Institute, 6-2-4 Furuedai, Suita 565 (Japan) Tel. (81-6)872-4817; Fax(81-6)872-4818. Abbreviations: aa, amino acid(s); bp, base pair(s); BtacAMP, dibutyryl CAMP; CAMP, cyclic AMP; cDNA, DNA complementary to RNA; HPLC, high-performance liquid chromatography; IDO, indoleamine 2,3dioxygenase; ZDO, gene (DNA) encoding IDO; IFN, interferon; i.p., intraperitoneal; kb, kilobase or 1000 bp; nt, nucleotide(s); ORF, open reading frame; SDS, sodium dodecyl sulfate; SSC, 0.15 M NaCI/O.OIS M Na, . citrate pH 7.6.
distributed in various organs of mammals and is induced by IFNs or IFN-inducers (Yoshida and Hayaishi, 1978; Yoshida et al., 1979; 1981; 1986; Pfefferkom, 1984; Yasui et al., 1986; Byrne et al., 1986; Takikawa et al., 1988). In particular, in a variety of cultured human cell lines, ID0 is markedly induced by IFN-7. Recently, Tone et al. (1990) and Dai and Gupta (1990) have isolated a cDNA encoding human IDO. In 1988, we reported that ID0 was induced in tumor cells undergoing rejection but not in infiltrating host cells of mouse after i.p. transplantation of allogeneic tumor cells (Yoshida et al., 1988). Subsequently, IFN-y was shown to be the inducer of this enzyme (Takikawa et al., 1990). To gain further insight into the cellular mechanism of ID0 induction by IFN-7 and the role played by ID0 in allog& rejection, cDNA clones encoding mouse ID0 were pre-
222 pared from poly(A) + RNA of mouse rectal cancer cells using human ID0 cDNA as a probe. The primary structure of the protein was deduced from the nt sequence, compared with that of human ID0 cDNA, and the identity of the protein as mouse ID0 was confirmed by the transient expression
of cDNA
IFN-/I
(data
not
shown).
The
highest
enzyme
activity
(167.6 nmol/h/mg of protein) was obtained with IFN-ytreated CMT-93 cells. Since synergistic effects of phorbol ester and IFN-7 on ID0 induction in human THP-1 cells have been reported by Edelstein et al. (1989) effects of various agents modifying the activities of protein kinases on the ID0 activity in mouse CMT-93 cells were examined. The ID0 activity of
in COS-7 cells.
(a) Induction of ID0 by IFN-y in various mouse cell lines and cDNA cloning from IFN-y-treated CMT-93 cells
IFN-y-treated CMT-93 cells was further enhanced approx. threefold upon simultaneous addition of Bt,cAMP or 8-Br-CAMP (Fig. l), whereas pre- or post-treatment with these agents had little, if any, enhancing effect (data not
To determine the best source for preparing a cDNA library for mouse IDO, 25 mouse cell lines were incubated
shown). Phorbol ester and Na * butyrate had almost no effect on IFN-y-mediated ID0 induction, and phorbol
with IFN-y, IFN-a or IFN-/I for three days. In seven of 25 cell lines, ID0 was induced by IFN-y (Table I), whereas no ID0 activity was detected in these cells either before incubation with IFN-y or after incubation with IFN-a or
ester was inhibitory
RESULTS AND DISCUSSION
at high concentrations.
Therefore,
a
LZap II cDNA library was constructed from poly(A)+ RNA of CMT-93 cells harvested on day 3 after treatment with IFN-y/Bt,cAMP (Chomczynski and Sacchi, 1987).
TABLE I ID0 induction by IFN-1 in various mouse cells Cell line”
Morphology b
Origin
ID0 activity”
CMT-93 Meth A Colon 26 MMT L-929 NMuMG C 127 I Bl6 3LL CL-S1 LA-4 KLN 205 Balb-MC FM-3A Y-1 I-10 F9-41 Ehrlich Balb/3T12-3 Balb/3T3 McCoy C3H/lOT1/2 P388D, Yac-1 L1210
E F E E F E E E E E E E E E E E E E F F F F L L L
Rectum, carcinoma Fibrosarcoma Rectum, carcinoma Mammary gland, carcinoma Connective tissue, normal Mammary gland, normal Mammary tumor Melanoma Lung, carcinoma Mammary alveolar nodules, preneoplastic Lung, adenoma Squamous cell, carcinoma Mammary tumor Mammary tumor Adrenal tumor Leydig cell testicular tumor Teratocarcinoma Ascites tumor Embryo, tumorigenic Embryo, nontumorigenic Connective tissue Embryo, nontumorigenic Lymphoid neoplasm Lymphoma Lymphoma
(nmol of kynurenine/h/mg of protein) 167.6 30.0 26.9 1.4 1.4 0.5 0.4 n.d. nd. n.d. n.d. nd. n.d. n.d. n.d. n.d. n.d. n.d. nd. n.d. n.d. n.d. n.d. n.d. n.d.
a MMT, CLSl, Balb-MC, FM-3A, Y-l, I-10, F9-41, Balb/3T3 and C3H/lOT1/2 cells were obtained from the Japanese Cancer Research Resources Bank (Tokyo). Other cells were purchased from Dainippon Pharmaceutical (Osaka). ’ E, epithelial cell; F, fibroblastic cell; L, lymphoblastic cell. ’ Cells at almost confluence were cultured in a 35-mm diameter plastic dish with 2 ml of medium containing 1000 units/ml of IFN-y under 5% CO1 and 95% air. After 72 h of culturing, cells were harvested by trypsinixation, and the ID0 activity was assayed by HPLC as described (Takiiawa et al., 1988). Protein was determined by the method of Lowry et al. (1951) with bovine serum albumin as a standard. Each value represents the mean of duplicate dishes except for CMT-93 cells (n = 7). n.d., nondetectable (~0.1).
223 P-45Os, hemoglobin, cyclooxygenase and hepatic phan 2,3-dioxygenase, except for human IDO.
Relative specific activity of ID0 (%)
IFN-Y (1000 units/ml)
(c) Alignment human IDOs
IFN-T + Bt$2AMP (0.1 mm) IFN-Y + B&CAMP (1 mM) IFN-Y + &
(3 mM)
IFN-Y + BrcAMP (1 mM) IFN-Y + BrcAMP (3 mM) IFN-Y + TPA (0.1 PM)
1990). Particularly, aa 101-184 of mouse ID0 were highly homologous with the human counterpart (aa 96-181) the identity being 89%. Moreover, since this region
IFN-Y + TPA (1 FM) IFN-T + Nahttylate
(0.1 mM) (1 “IM)
Fig. 1. Effects of various agents modifying the activities of protein kinases on IFN-y-mediated ID0 induction. CMT-93 cells at nearconfluence were cultured in a plastic dish (35 mm) with 2 ml of medium containing 1000 units/ml of IFN-7 (4 x lo6 units/mg of protein) and various agents under 5 y0 CO, and 95 y0 air. After 72 h culture, the cells were harvested by trypsinixation and the ID0 activity was determined as described in Table I, footnote c. The ID0 activity induced by IFN-), (1000 units/ml) alone was 167.6 k 41.1 mnol kynurenine formed/h/mg of protein (mean f SD, n = 7). The value represents 100% activity. Data (relative to IFN-), treatment) are expressed as mean k SEM of seven cultures of three different experiments. BrcAMP = 8-Br-CAMP.
On screening, with human ID0 cDNA 2.9 x 10’ phages yielded 33 positive clones.
of the deduced aa sequences of mouse and
As shown in Fig. 3, the deduced aa sequence of mouse ID0 (407 aa) revealed a 61% overall homology to that of human ID0 (403 aa) (Tone et al., 1990; Dai and Gupta,
IFN-Y + BrcAMP (0.1 mm)
,FN--, + Ndutytate
trypto-
as
a probe,
(b) Nucleotide sequence of mouse ID0 cDNA These positive clones contained cDNA inserts of 1.2-1.5 kb, which were each slightly different in the 5’- or 3’-untranslated regions. The nt sequences of two independent clones, pBS25 and pBS27, were found to be identical in the overlapping regions. The cDNA sequence was 1555 bp long and contained an ORF of 1221 nt (nt 71-1291) flanked by 70 bp and 261 bp of 5’- and 3’untranslated regions, respectively (Fig. 2). The 3’-untranslated region was highly A + T-rich and the putative polyadenylation signal, AATAAA, was present at 244 bp downstream from the stop codon (nt 1292-1294). The aa sequence deduced from the nt sequence contained partial aa sequences of endopeptidase digests of purified mouse ID0 (Fig. 2, underlined). In addition, the sequence, 5’-GCAATGG, near the putative start codon, ATG, agreed with the optimal sequence of Kozak (1986) for eukaryotic mRNAs. Therefore, the ORF encodes a protein of 407 aa with an A4, of 45 639. In the coding region, there was no potential Asn-linked glycosylation site (Asn-XaaThr/Ser), suggesting that, in contrast to the purified enzyme from rabbit intestine (Shimizu et al., 1978), the mouse ID0 may not be a glycoprotein. When the sequence of mouse IDO, a heme-containing enzyme, was compared with the GenBank and NBRF databases, no significant homology was found with the nt or aa sequences of other hemoproteins such as cytochrome
(aa 101-184) was rich in hydrophobic aa residues (Fig. 2) it might function as a catalytic domain. The whole hydropathy pattern of mouse ID0 was very similar to that of human ID0 (data not shown). (d) Expression of mouse ID0 cDNA in COS-7 cells To confirm that the cDNA clone actually encodes functional mouse IDO, the cDNA was inserted with two orientations (sense and antisense) into the expression vector (pdKCR) (Fukunaga et al., 1984), and was transfected into COS-7 cells. As shown in Fig. 4, the COS-7 cells transfected with the pdKCR-MZDOsense yielded a high level (400 nmol/h/mg) of ID0 activity in the cells, while no enzyme activity was detected in the cells transfected with the pdKCR-M ID0 antisense. The expression of functional ID0 in COS-7 cells transfected with pdKCR-MIDOsense was confirmed by a decrease in tryptophan concentration or increase in the amount of kynurenine, an IDO-mediated metabolite of tryptophan, in the culture medium. In the case of pdKCR-M ID0 antisense, however, there was no change in the metabolism of tryptophan to kynurenine. Furthermore, the ID0 expressed by pdKCR-M ID0 sense had the same catalytic properties as those of native IDO; both methylene blue and ascorbic acid were absolutely required for enzyme activity (data not shown). (e) Time course of induction of ZDO mRNA and ID0 activity by IFN-7 in CMT-93 cells To explore the mechanisms of ID0 induction by IFN-y/Bt,cAMP, the amounts of ID0 mRNA and the ID0 activities in CMT-93 cells were monitored at various time intervals after treatment with IFN-y, Bt,cAMP or a combination of both. Fig. 5A shows a Northern-blot analysis of total cellular RNAs using mouse ID0 cDNA as a probe. Although no ID0 mRNA was detected in untreated CMT-93 cells, the level of mRNA began to increase within 24 h after IFN-y treatment and reached a maximum on day 2 (Fig. 5B). The level of mRNA gradually declined and was undetectable on day 5. Similar time-dependent changes in the levels of ID0 mRNA were seen with the IFN-y/Bt,cAMP-treated CMT-93 cells, and the maximum level was 1.5- to twofold higher than that of IFN-y-treated
224 GCThMCCCThThkGMhTGCTGGGhChTTCCTTCAGTGGCChhGTGGGGGG'KAGTGGhGThGhChGCh7G AT0 GCA CTC hGT hhh ATh TCT CCT hCh GM GGT TCT hGh hGG hTC CTT Ghh GhC ChC 127 I PlrtAla Lou -I Lye 118 9er Pro Thr Glu Gly S-r k9 A%i II* h" Gl" AW iii, CACATAGAT GAAGATGTG GGC TTT GCT CTACCACATCCACTC GTG GAGCTG CCC GAC 164 20 ,,I.Zl. A,p Cl" A.p Val Gly P,,.hl9 L.u Pro Hi. Pro L.U Vd 01" LO" PI0 h.p GCh TAC AGC CCC TOG OTC CTT GTG GCT hGh MT 39 Ala Tyr 6.r Pro Trp VSl In" Val Ala &g Am
CTG CCT OTC CT0 ATT GU3 AAI:GGG 241 Lo" Pro '?a1L*" Il. Cl" h.n Gly
~CTTCWLDMOAAGTT~Al*;Cn:CCCh~Cn;wCAcG~~Clr;~u;r299 39 Gin Mu hr9 Cl" Glu Vrl Cl" Ly, Lo" Pro Thr Lo" 6.r Thr Up CAC AGGTTACAO CGC CTGGCACAC 77 tliahrvgIau Gin hrG Le" Ala Hia
01~ LO" -01~
CTt3GCC CTE G.2GTM: ATChCChTGGCG TATGTG 355 L8” Air Is+” Gly Tyr I19 Tbt Mwt Al& Tyr Vd
TGG hhC CGhGGG GhT GhC GhTGTTCGhMGGn; CTG CCC CGCiUT ATT GCTGTTCCC 412 96 Trp hen hrG Gly hap A.p A.p Val &9 Ly, Val L." Pro Ar9 h.8,11a Ala Vd PrO TAC TGC GAG CTC TCA WhhG TTG GGC CTG CCT CCTATT Cn: ET TAT GChGhC TGT 469 115 Tyr Cye Cl" La" 6.r Cl" Lya IA" Gly Leu Pro Pro Ila Lo" SO):Tyr Ala h.p Cy9 OTC CTG GCh hhC TGG MO hU hhG GhC CCC MT 134 VSl La" Ala hen Trp Ly8 Lye Lye AspPro Am
GGG CCC ATG hC& ThC GAO h&C ATG 526 Gly Pro blot Thr Tyr Cl” Aon Ict
GAC hTTCTG TTC TChTTT CCTGGTGGG GhC TGC GhC MOGGC TTC TTC CTC GTC TCT 593 153 Asp Ila Lo" Ph. G.r Phe Pro Gly Gly Aep Cys Amp Lye Gly PlM Pha In" Val Sot
CTA TTG GTG GAA ATC Wh GCT TCT CCT GCA ATC hhh GCh ATC CCC hCT GTA WC AGT 640 172 Leu Lw Val 0111IlS Air Ala S-r Pro Ala Ila Lye hla 11. Pro Thr Val S@r SW GCA GTA GAG COT CAA GAC CT0 AAA GCA TTG OM AAG GCA CTG CM: GhC ATA GCT hCC 697 191 Ala Val Cl" Ax9 Gln ASP La" Lys Ala Lou Glu Lys Ala La" Hi8 hep Il* Ale Thr AGTCW~AMOCCMCGMhTT~~G~hn;CGTG~PTTGTEGllCCCA~C734 210 Ser Ie" Cl" Lye hla Lys Cl" Xl* Phe Lys hrg Mnt hr9 Asp PM
V11 h*p Pro -p
ACG TTT TTC ChC OTT CTC CCC ATA ThT CTG TCT GGC TCG AM MC hGC TCC hhG CT0 811 229 Thr Pha Ph. Hia Val LOU Ar9 Ile Tyr Iru Sex Gly Trp Ly. Cya 8~ Ser Lys -LDU CCA GAA GGT CTG CT0 TAT GhG GGG OTC TOG GhC hCC CCA MA ATG TTT TCh GGG GGC 968 249 Pro Cl" Gly Lm" La" Tyr Cl" Gly Val Trp Alp Thr Pro Lym BlStPho Sar Gly Gly AGT GCh GGC ChG WC 1IM:ATC TTC ChG ROT CTT GhT GTC CTT CTG GGh ATA h&A ChC 923 267 $0~ hla Gly Gln Ser So? Ile Phe Gin 9.r Lo" hap V&l Leu Leu Gly Ile Lya lie GADOCTM;C~OM~TCCT~OAA.m:CTc~~hn:lu;hou;TlrchTcccT992 286 01~ hla Gly Lye Cl" SW Pro Ala Cl" Ph. La" Cln Cl" Met Ar9 Cl" Tyr Mat Pro CCA GCC CAC CGG MC TTC CTT TTC TTC TTA GAO TCh GCT CCC CCh OTC CGT GhG TTT 1039 305 Pro Ala Him hr9 ha" Ph. LW Pha Phe Lau Glu 8-r Ala Pro Pro Val hrq Cl" Ph.
OTC ATT TCAAGACAC MT 324 Val 11. Bar AtQ HI. Am
GhhGhC TTG hCGAMGCT ThTAAC GSR TGTGTG hhT GGT1096 Cl" Asp 1-u Thr Lye Ala Tyr han Glu Cys Vsl &n Gly
CTG OTC TCT GTG MA AAG TTC CAC C'K GCA ATA GTA GAT ACT TAC ATT ATG hRh CCT 1153 343 &WI Val Ser Val Arq Lys Phm Him Inu Ala 11s V&l Aep The Ty? Il. @IatLps Pro TCG 8.AG A&G AAG CCC ACT
362 SW
OAT GGC G&C h?.GTCG CM GhG CCC TCh AAT GTG Ghh hGC 1210 Lye Ly* Lye Pzo Thr Asp Gly ABP Lya Ssr Cl" Cl" Pro SW Am Val Cl" Sar
AGAGGG ACT GGG GGTACGMTCCChTGhCT TTC CTAAGGAGTGTGAAAGATACAACC 1267 381 Arg Gly Thr Gly Gly Thr Am Pro Met Thr Phe Las Ax9 Sar Val Lya Asp Thr Tht GAGAAAGCT CTT CTG ACT TGG CCT ThG 390 Cl" Lyn Ala Le" Ldu 6.r Trp PTO '**
1294
TGThGChhGCTCcACTT%ZTh~ CTTCXKCTGFCATAGCTCATTAMTTCAGATCCACCA 1369 TchlcZ 1444 hGThhGhCThThGhhThMTTGCCTUTTGThT~hWThGkMT~~~;MCT~TTT~~TA~T~~TAT~T~T~T~~ATA~T~T~T~T~ ChhchhTGhGhcCChGTBTThTTGTGhA
1919
1555
Fig. 2. Nucleotide sequence of mouse ID0 cDNA and deduced aa sequence. A cDNA library was ligated with the vector AZap (Stratagene, La Jolla, CA) at the EcoRI site. The recombinant DNA was packaged in vitro (Sambrook et al., 1989) and the phages were grown in E. co/i X.1-Blue. The recombinant phages were screened by plaque hybridization (Benton and Davis, 1977) with the l.l-kb ofEcoRLEcoT141 fragment of human ID0 cDNA. which was labeled with 32P by the random primer method (Feinberg and Vogelstein, 1983). Hyb~dization was carried out by the method of Wahl et al. (1979), except that the hybridization temperature was lowered to 32°C. The pBluescript SK{ - ) plasmid DNAs from the positive clones were recovered using helper phage R408. The nt sequences were determined by the dideoxy chain-termination method (Sanger et al,, 1977). The GenBank accession No. for this sequence is M69109. Numbers on the right margin refer to nt position, and numbers on the left margin refer to aa position, The putative polyadenylation signal (AATAAA) is boxed. Those aa which were confirmed by protein sequencing of purified mouse IDO are underlined. Mouse IDO was purified from the epididymis by the method of K. Nakata, Y. Watanabe and 0. Hayaishi (details to be published elsewhere). The purified IDO was digested with @syl endopeptidase (Tsunasawa et al., 1989) and the fragments were sequenced with an Applied Biosystem model 477A Protein Sequencer (Foster City, CA).
cells. Although the time course of ID0 mRNA showed a pattern similar to that of ID0 activity (Fig. X2), the former preceded the latter by approx. one day. Bt,cAMP itself, however, induced neither 11)O mRNA nor ID0 activity in CMT-93 cells. These observations suggested that the ID0
mRNA was induced by IFN-), in CMT-93 cells, and that mRNA induction by IFN-y was enhanced by the addition of Bt,cAMP. Since Northern-blot analysis only indicates the steady-state level of ID0 mRNA, the precise molecular basis of transcriptional control in CMT-93 cells (increase
225
A No treatment Bt 2CAMP
MID0
-SPAIKAIPT”SSA”---ER*D-LKALEKALHDIATS-LEKAKEIFKRMRDFVDPDTFFH
232
*IDO
b._...“....~K.MQ~Q.._.T.__._...~E.._.C....~Q”.”QIH.H.N.KA..S
*2*
IFN-Y IFN- Y+Bt 2 CAMP 012345
Culture
MID0
KALLSWP
407
HID0
.S,.KEG
403
Fig. 3. Comparison
of aa sequences
sequence
ID0
of mouse
(HIDO). mouse
Dots represent IDO; dashes
of mouse and human
(MIDO)
is aligned
aa residues
represent
of human
IDOs. The aa
with that of human ID0 identical
ID0
to those of
gaps.
A
B
Culture mock
-
pdKCR
_
Days
Days
‘3: 150.
pdKCR-Mm (anlironre) t
loo-
pdKCR-ME (4ense) 0
100 ID0
Fig. 4. Transient
zoo Activity
300
site of a eukaryotic
1984). The intact
region
DNA polymerase using BarnHI
of mouse
and inserted
adapters.
alone were transferred
20
40
ID0
(nt 65-1364)
cDNA,
These constructs into COS-7
(antisense
the
with T4 pdKCR,
and sense) or vector
cells by the DEAE-dextran
method
(Sambrook et al., 1989). (A) ID0 activity (hatched bar) in the supernatant fraction ofcell lysate. The ID0 activity was determined as described in Table I, footnote kynurenine of tryptophan HPLC
c.(B) Concentrations
(blackened
and kynurenine
as described
oftryptophan
boxes) in the culture (Takikawa
(open boxes) and
media. The concentrations
in the culture medium
1
2
were measured
by
et al., 1988).
4
Days
cells (2 x lo5 cells/ml) were cultured
the medium alone (0) or medium containing 3 mM Bt,cAMP analysis.
( x ), or a combination
for one to live days with 1000 units/ml of IFN-y(m),
of both (0).
On day 3 after the administration
phoresis
(Sambrook
GeneScreen
to 1.2% agarose/2.2 et al.,
the blot was incubated NaCl/lO%
1989) After
Plus filter (DuPont-New dextran
at 42°C
sulfate/l
(A) Northern-blot
of IFN-)I and/or
total RNA of an aliquot (S/10) of cell suspension of RNA were subjected
5
of mouse ID0 mRNA and ID0 activity.
Fig. 5. Time course of induction CMT-93
in a mixture containing
in the rate of transcription, stabilization of mRNA, etc.) remains to be elucidated. As shown in Fig. 5, the induction of ID0 in mouse CMT-93 cells was regulated at the ID0 mRNA level, and two days were needed to reach the maximum level of IFN-y-induced ID0 mRNA. Furthermore, preliminary
3
Culture
et al.,
containing plasmid,
0
into the
(Fukunaga
was blunt-ended
into a BamHI-cleaved
(PM)
in CO.%7 cells. The
vector pdKCR
fragment
60
or Kynurenine
SK( - ) plasmid was subcloned
expression
1.3-kb SmaI-BsmI
coding
0
Tryptophan
of mouse ID0 cDNA
expression
mouse cDNA of the pBluescript BarnHI
400
(nmol/h/mg)
M formaldehyde transferring
England
in a mixture
Bt,cAMP,
was isolated, the
Nuclear)
and 20 ng gel electroRNA
of 50% formamide/l
% SDS, for 6 h, and then hybridized
both denatured
to a
with 20 x SSC, M
at 42°C
salmon sperm DNA (100 ng/ml)
and denatured “P-labeled 0.95-kb _EcoRI fragment of mouse ID0 cDNA probe. (B) Relative levels of ID0 mRNA. The autoradiogram of a Northern Tokyo)
blot was scanned to determine
aliquot of cell suspension lysate and the ID0 Table I, footnote ments.
with a BAS2000
relative
amounts
was disrupted
activity
Image Analyzer
of mRNA.
(C) ID0
by sonication
in the cells was determined
c. Each value represents
(Fuji Film, activity.
to prepare
as described
the mean of duplicate
An
the cell in
experi-
226 experiments from our laboratory have demonstrated that cycloheximide, a protein synthesis inhibitor, completely blocked the IFN-y-mediated induction of ID0 mRNA, suggesting that de novo synthesis of protein(s) was essential for mRNA induction by IFN-y. Indeed, Caplen and Gupta (1988) recently suggested that (a) newly synthesized protein(s) is (are) required for the IFN-y-mediated induction of human ID0 mRNA. In addition to B& (Fig. 4), other investigators have shown that various kinds of agents such as tumor necrosis factor-a (in mouse lung; Bianchi et al., 1988) and phorbol ester (in human monocytic leukemia cells; Edelstein et al., 1989) enhance ID0 induction by IFN-y. Therefore, several signaling pathways may be involved in ID0 induction by IFN-y.
Byrne, G.I., Lehmann, catabolism Immun. Caplen,
G.J.: Induction
of tryptophan
for gamma-interferon-mediated
Chlumydia psittaci replication
tion of intracellular
inhibi-
in T24 cells. Infect.
53 (1986) 347-351.
H.S. and Gupta,
by human
S.L.: Differential
interferon-y
regulation
and interferon-a.
of a cellular gene
J. Biol. Chem. 263 (1988)
332-339. Chomczynski, by
P. and Sacchi,
acid
guanidinium
Analyt.
Biochem.
N.: Single-step
S.L.: Molecular
cloning,
of human interferon-y-inducible Biochem. Edelstein,
method
of RNA isolation
thiocyanate-phenol-chloroform
Biophys.
sequencing
indoleamine
Res. Commun.
M.P., Ozaki, Y. and Duch, D.S.: Synergistic
THP-1 monocytic
leukemia
endonuclease
Biochem. Fukunaga,
B.: A technique fragments
2,3-dioxygenase
in
143 (1989) 2969-2973. for radiolabeling
to high specific
DNA
activity.
Anal.
132 (1983) 6-13.
R., Sokawa,
Y. and Nagata,
human
interferons
vector.
Proc. Natl. Acad.
S.: Constitutive
production
by mouse cells with bovine papillomavirus
Kozak, M.: Point mutations codon
cDNA.
effects of phorbol
of indoleamine
cells. J. Immunol.
A.P. and Vogelstein,
and expression
2,3_dioxygenase
168 (1990) l-8.
ester and IFN-)I on the induction Feinberg,
extraction.
162 (1987) 156-159.
Dai, W. and Gupta,
restriction
(f) Conclusions (I) A mouse ID0 cDNA was cloned from IFN-y/ Bt,cAMP-treated CMT-93 cells using human ID0 cDNA as a probe. (2) The mouse ID0 cDNA was 1555 bp long and encoded 407 aa residues with an h4,. of 45639. (3) The deduced aa sequence of mouse ID0 revealed a 61% overall homology with that of human IDO. (4) Transient expression of mouse ID0 cDNA in COS-7 cells yielded a high level of ID0 activity. (5) Northern hybridization analysis indicated that IFN-), induced the ID0 mRNA, and that the level of RNA was further increased by simultaneous addition of Bt,cAMP, while Bt,cAMP itself had no effect on mRNA induction.
L.K. and Landry,
is the mechanism
Sci. USA 81 (1984) 5086-5090. define a sequence
that modulates
of as a
translation
flanking the AUG initiator
by eukaryotic
ribosomes.
Cell 44
(1986) 283-292. Lowry,
O.H., Rosebrough,
measurement
N.J., Farr,
A.L. and Randall,
R.J.: Protein
with the Folin phenol reagent. J. Biol. Chem. 193 (1951)
265-275. Pfefferkorn, human
by inducing
Proc. Natl. Acad. Sambrook,
E.F. and Maniatis,
Manual,
terminating
T.: Molecular
2nd ed. Cold Spring Harbor
Cold Spring Harbor, F., Nicklen,
the host cells to degrade
tryptophan.
Sci. USA 81 (1984) 908-912.
J., Fritsch,
Laboratory Sanger,
y blocks the growth of Toxoplasma gondii in
E.R.: Interferon libroblasts
Cloning.
Laboratory
A
Press,
NY, 1989.
S. and Coulson,
inhibitors.
Proc.
A.R.: DNA sequencing Natl.
Acad.
Sci.
with chain-
USA
74 (1977)
5463-5467. Shimizu,
T., Nomiyama,
2,3-dioxygenase:
ACKNOWLEDGEMENTS
S., Hirata,
purification
F. and Hayaishi,
and some properties.
0.: Indoleamine J. Biol. Chem. 253
(1978) 4700-4706. Takikawa,
The authors are indebted to Dr. 0. Hayaishi for his continued encouragement during this investigation. The authors also thank Shionogi Pharmaceutical (Osaka, Japan) for kind providing IFN-y, Dr. S. Nagata of this Institute for pertinent discussion, and Ms. K. Hirata for aid in the preparation of the manuscript. This work was supported in part by research grants from the Sankyo Foundation of Life Science and Grant-in Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan.
O., Yoshida,
degradation
R., Kido,
in mice initiated
R. and Hayaishi,
by indoleamine
0.:
Tryptophan
2,3-dioxygenase.
J. Biol.
Chem. 261 (1986) 3648-3653. Takikawa,
O., Kuroiwa,
interferon-y
T., Yamazaki,
F. and Kido, R.: Mechanism
action: characterization
in cultured
human cells induced
enzyme-mediated
tryptophan
of indoleamine
by interferon-y degradation
of
2,3dioxygenase
and evaluation
in its anticellular
of the activity.
J. Biol. Chem. 263 (1988) 2041-2048. Takikawa,
O., Habara-Ohkubo,
of indoleamine rejection.
A. and Yoshida,
2,3-dioxygenase
J. Immunol.
in allogralted
R.: IFN-yis
the inducer
tumor cells undergoing
145 (1990) 1246-1250.
Tone, S., Takikawa, O., Habara-Ohkubo, A., Kadoya, A., Yoshida, R. and Kido, R.: Primary structure of human indoleamine 2,3dioxygenase deduced
from the nucleotide
sequence
of its cDNA.
Nucleic
Acids Res. 18 (1990) 367. Tsunasawa,
REFERENCES
S., Ma&i,
The primary Benton,
W.D. and Davis,
hybridization
R.W.: Screening
to single plaques
Igt recombinant
in situ. Science
clones by
196 (1977) 180-182.
Bianchi, M., Bertini, R. and Ghezzi, P.: Induction of indoleamine 2,3dioxygenase by interferon in mice: a study with different recombinant interferons and various cytokines. 152 (1988) 237-242.
Biochem.
Biophys.
Res. Commun.
structure
T., Hirose,
M., Soejima,
lyticus protease I, a lysine-specific (1989) 3832-3839.
M. and Sakiyama,
F.:
characteristics
ofAchromobacter
serine protease.
J. Biol. Chem. 264
and structural
Wahl, G.M., Stem, M. and Stark, G.R.: Efficient transfer of large DNA fragments from agarose gels to diazobenzyloxymethyl-paper and rapid hybridization by using dextran USA 76 (1979) 3683-3687.
sulfate.
Proc. Natl. Acad.
Sci.
227 Yasui, H., Takai, K., Yoshida, R. and Hayaishi, 0.: Interferon enhances tryptophan metabolism by inducing pulmonary indoleamine 2,3dioxygenase: its possible occurrence in cancer patients. Proc. Natl. Acad. Sci. USA 83 (1986) 6622-6626. Yoshida, R. and Hayaishi, 0.: Induction of pulmonary indoleamine 2,3dioxygenase by intraperitoneal injection of bacterial lipopolysaccharide. Proc. Natl. Acad. Sci. USA 75 (1978) 3998-4000. Yoshida, R., Urade, Y., Tokuda, M. and Hayaishi, 0.: Induction of indoleamine 2,3-dioxygenase in mouse lung during virus infection. Proc. Natl. Acad. Sci. USA 76 (1979) 4048-4086.
Yoshida, R., Imanishi, J., Oku, T., Kishida, T. and Hayaishi, 0.: Induction of pulmonary indoleamine 2,3-dioxygenase by interferon. Proc. Natl. Acad. Sci. USA 78 (1981) 129-132. Yoshida, R., Oku, T., Imanishi, J., Kishida, T. and Hayaishi, 0.: Interferon: a mediator of indoleamine 2,3-dioxygenase induction by lipopolysaccharide, poly(1) *poly(C) and pokeweed mitogen in mouse lung. Arch. Biochem. Biophys. 249 (1986) 596-604. Yoshida, R., Park, S.W., Yasui, H. and Takikawa, 0.: Tryptophan degradation in transplanted tumor cells undergoing rejection. J. Immunol. 141 (1988) 2819-2823.
