Preference and Lysosomal Localization Acid Oxidase with An

IL-4-Induced Gene-1 Is a Leukocyte l-Amino Acid Oxidase with an Unusual Acidic pH Preference and Lysosomal Localization

This information is current as James M. Mason, Mamta D. Naidu, Michele Barcia, Debra of September 28, 2021. Porti, Sangeeta S. Chavan and Charles C. Chu J Immunol 2004; 173:4561-4567; ; doi: 10.4049/jimmunol.173.7.4561 http://www.jimmunol.org/content/173/7/4561 Downloaded from

References This article cites 47 articles, 21 of which you can access for free at: http://www.jimmunol.org/content/173/7/4561.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 28, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

IL-4-Induced Gene-1 Is a Leukocyte L-Amino Acid Oxidase with an Unusual Acidic pH Preference and Lysosomal Localization1

James M. Mason,* Mamta D. Naidu,† Michele Barcia,* Debra Porti,* Sangeeta S. Chavan,† and Charles C. Chu2†

IL-4-induced gene-1 (Il4i1 or Fig1) initially isolated as a gene of unknown function from mouse B lymphocytes, is limited in expression to primarily immune tissues and genetically maps to a region of susceptibility to autoimmune disease. The predicted Il4i1 protein (IL4I1) sequence is most similar to apoptosis-inducing protein and Apoxin I, both L-amino acid oxidases (LAAO; Enzyme Commission 1.4.3.2). We demonstrate that IL4I1 has unique LAAO properties. IL4I1 has preference for aromatic amino acid substrates, having highest specific activity with phenylalanine. In support of this selectivity, IL4I1 is inhibited by aromatic competitors (benzoic acid and para-aminobenzoic acid), but not by nonaromatic LAAO inhibitors. Il4i1 protein and enzyme activity is found in the insoluble fraction of transient transfections, implying an association with cell membrane and possibly Downloaded from intracellular organelles. Indeed, IL4I1 has the unique property of being most active at acidic pH (pH 4), suggesting it may reside preferentially in lysosomes. IL4I1 is N-linked glycosylated, a requirement for lysosomal localization. Confocal microscopy of cells expressing IL4I1 translationally fused to red fluorescent protein demonstrated that IL4I1 colocalized with GFP targeted to lysosomes and with acriflavine, a green fluorescent dye that is taken up into lysosomes. Thus, IL4I1 is a unique mammalian LAAO targeted to lysosomes, an important subcellular compartment involved in Ag processing. The Journal of Immunology, 2004, 173:

4561–4567. http://www.jimmunol.org/

nterleukin-4-induced gene-1 (Il4i13 or Fig1) was initially iso- cated by a locus or loci derived from the NZW mouse strain (8– lated in a screen for IL-4-induced genes from mouse B cells 10). Interestingly, the NZW allele for Il4i1 contains three amino I using cDNA representational difference analysis (1, 2). Re- acid substitutions (11), which could alter a function that results in cently, Fig1 was renamed Il4i1 in keeping with recommendations SLE susceptibility. Furthermore, the Sle3 susceptibility locus from from the Human Gene Nomenclature Committee (3). Il4i1 may this region appears to be expressed primarily in APCs (12), where play an important role in the immune system for several reasons. we have also detected Il4i1 expression. Finally, the human Il4i1 It is induced in an immediate-early fashion by IL-4, a key regulator ortholog is located on chromosome 19q13.3–19q13.4 (11) in a of the immune response (4); its expression is strikingly limited to region that is a hot spot for autoimmune disease susceptibility in by guest on September 28, 2021 immune tissues, with clear expression in B cells (1, 5); and it is general, including rheumatoid arthritis, multiple sclerosis, insulin- also found in professional APCs (dendritic cells (DC) and macro- dependent diabetes mellitus, and SLE (13–15). Thus, an alteration phages) (6, 48). in Il4i1 could contribute to autoimmune disease in both mouse Mouse Il4i1 genetically maps to chromosome 7 between the and man. Klk6 and Fut1 genes (23.1 cM; Mouse Genome Database, Mouse One clue to the function of Il4i1 is its sequence similarity to Genome Informatics; The Jackson Laboratory, Bar Harbor, ME; L-amino acid oxidases (LAAO; Enzyme Commission (EC) www.informatics.jax.org (July 2004)) (1, 7), a region where sus- 1.4.3.2) (Fig. 1). The cDNA sequences of both mouse and hu- ceptibility to systemic lupus erythematosus (SLE) has been impli- man Il4i1 predict a protein containing a putative signal peptide sequence for entry into the endoplasmic reticulum (ER), a re- *Gene Therapy Vector Laboratory, and †Laboratory of Gene Activation, North Shore- gion of similarity to LAAO, and a C-terminal region with no Long Island Jewish Research Institute, and Departments of Medicine, North Shore University Hospital and New York University School of Medicine, Manhasset, NY homology to known proteins (1, 11). The LAAO region of sim- 11030 ilarity most closely resembles apoptosis-inducing protein found Received for publication June 18, 2003. Accepted for publication July 23, 2004. in fish (43% identical over 485 aa) (16) and snake venom The costs of publication of this article were defrayed in part by the payment of page LAAO (37% identical over 484 aa) (17), also known as Apoxin charges. This article must therefore be hereby marked advertisement in accordance I (18). The LAAO-similar portion conserves key domains and with 18 U.S.C. Section 1734 solely to indicate this fact. residues that bind the flavin adenine dinucleotide (FAD) cofac- 1 This work was supported in part by the North Shore-Long Island Jewish Research Institute, the Muriel Fusfeld Foundation, the Leonard Wagner Autoimmunity Fund, tor required for its enzymatic activity, as well as residues in the the SLE Foundation, and U.S. Public Health Service Grant AI44837 awarded by the active site of the LAAO crystal structure (11, 19). National Institutes of Health. Because of the similarity of IL4I1 to LAAO, we examined the 2 Address correspondence and reprint requests to Dr. Charles C. Chu, Laboratory of LAAO enzyme reaction catalyzed by IL4I1. LAAO catalyzes the Gene Activation, North Shore-Long Island Jewish Research Institute, and Depart- ments of Medicine, North Shore University Hospital and New York University oxidation of L-amino acid to 2-oxo acid, via an L-imino acid in- School of Medicine, 350 Community Drive, Manhasset, NY 11030. E-mail address: termediate (20) (Fig. 2). Water and oxygen are required in this [email protected] reaction resulting in hydrogen peroxide (H O ) and ammonia by- 3 Abbreviations used in this paper: Il4i1, IL-4-induced gene-1; IL4I1, Il4i1 protein; 2 2 DC, dendritic cell; SLE, systemic lupus erythematosus; LAAO, L-amino acid oxidase; products. We developed an enzyme assay demonstrating that IL4I1 ER, endoplasmic reticulum; FAD, flavin adenine dinucleotide; EGFP, enhanced GFP is the first LAAO found in immune cells. Although IL4I1 has gene; IRES, internal ribosome entry site; dsRED, Discosom red fluorescent protein; BA, benzoic acid; PABA, para-aminobenzoic acid; diacetyl, 2,3-butanedione; 2H3B, similar enzyme kinetics to other LAAO from other cells and or- 2-hydroxy-3-butynoic acid. ganisms, it is most reactive with aromatic amino acids at acidic

FIGURE 1. IL4I1 is most similar to LAAO. Schematic of mouse IL4I1 is shown to scale with N-linked glycosylation sites (N-Gly) (f) and po- tential tyrosine (Y) phosphorylation sites (E). Predicted FAD cofactor binding domains are indicated by white boxes in the schematic corresponding to the three black lines underneath. The signal peptide and short regions of similarity to many known FAD-binding proteins (MANY) and to spe- FIGURE 2. LAAO reaction. The chemical structure of L-phenylalanine cific proteins (phytoene desaturase (PDS), monoamine oxidase (MAO), is shown as a typical L-amino acid oxidized by LAAO. LAAO is indicated and tryptophan 2-monooxygenase (TMO)) are shown by the next row of with the FAD cofactor and its oxidation state. The reaction is shown in two black lines underneath the schematic, corresponding to the gray-spotted steps, with an L-imino acid intermediate. boxes. Regions of similarity to bacterial (Bacillus cereus) and eukaryotic Downloaded from (fish (Scomber japonicus) and snake (Crotalus atrox, C. adamanteus)) LAAO are shown by the remaining black lines underneath the schematic of solution, extracted with phenol/chloroform mixture, and then precipitated IL4I1. with isopropanol. The resulting RNA was resuspended in 20–50 ␮lof diethyl-pyrocarbonate-treated H2O and quantitated using UV absorption pH. Because acidic pH preference is a common feature of lysoso- measurements at 260/280 nm. For first-strand cDNA synthesis of RT-PCR, 7 ␮l of RNA (200–1000 mal enzymes, we investigated the subcellular localization of IL4I1 ng) plus 2 ␮lof10␮M oligo(dT) (5Ј-TTT TTT TTT TTT TTT V) was http://www.jimmunol.org/ and found that it is the first known LAAO that is targeted to ly- heated at 65°C for 5 min, cooled at room temperature, and then kept on ice. sosomes. Thus, IL4I1 may have a fundamental role in lysosomal To this mixture, 4 ␮l of 2.5 mM dNTP, 1 ␮l of 100 mM DTT, 1 ␮lof200 Ag processing and presentation. U/␮l Moloney murine leukemia virus reverse transcriptase, 4 ␮lof15mM MgCl2, 375 mM KCl, 250 mM Tris (pH 8.3) buffer (Invitrogen Life Tech- nologies), and 1 ␮lof1U/␮l RNase inhibitor (5 Prime33 Prime, Boulder, Materials and Methods CO) were added and incubated at 42°C for 60 min, heated to 99°C for 5 Expression plasmid construction min, and stored at 4°C. This cDNA (1 ␮l) was PCR amplified in a 20-␮l Plasmid pIl4i1-iGFP was constructed by digesting vector pIRES2-EGFP reaction containing 10 mM Tris (pH 8.3), 2.0 mM MgCl2, 50 mM KCl, 0.2 mM each dNTP, 1.5 U of AmpliTaq Gold (Roche Molecular Biochemicals, (BD Clontech, Palo Alto, CA) with restriction enzymes Ecl136II and Ј EcoRI, isolating the 5.3-kb linearized vector, and then ligating a 1.9-kb Indianapolis, IN), and 0.5 mM each of Il4i1-specific primers (CCC52 (5 - CGG AAT TCG GTA CCG GAG AAG ATG CCA GAA AAG) and by guest on September 28, 2021 PvuII/EcoRI fragment from plasmid pIl4i1-IRES-dsRED (data not shown) Ј containing the full-length mouse Il4i1 cDNA into the vector (Fig. 3). In CCC54 (5 -CGG AAT TCG GTA CCG TAA GGC TTC TGC AAA)) or ␤-actin-specific primers (M309-11 (5Ј-GAT GAC GAT ATC GCT GCG pIl4i1-iGFP, both Il4i1 and enhanced GFP gene (EGFP) are expressed on Ј a bicistronic mRNA with an internal ribosome entry site (IRES) between CTG) and M309-12 (5 -GTA CGA CCA GAG GCA TAC AGG)) under Il4i1 and EGFP that permits translation of both proteins. Plasmid pIl4i1- the following PCR conditions: 95°C for 5 min; (94°Cfor5s,55°C for 15 s, dsRED was constructed by restriction enzyme digesting vector pDs- and 72°C for 30 s) times 25–35 cycles; 72°C for 30 s and soak at 4°C. PCR Red1-N1 (BD Clontech) with SmaI and AgeI, isolating the 4.7-kb linear- products were electrophoresed on 1% agarose gels and visualized after ized vector, and ligating a similarly digested 1.9-kb DNA fragment ethidium bromide staining under UV illumination. containing the complete mouse Il4i1 coding sequence and optimized Immunoblot analysis Kozak translational start site (21), but lacking a stop codon (Fig. 3). This DNA fragment was generated by PCR using primers NS217 (5Ј-AAA CCC Supernatants were collected from transfected cell cultures 2 days postwash GGG GCC GCC ACC ATG GCT GGG CTG GCC CTG CGT-3Ј) and and concentrated ϳ100-fold to 100–150 ␮lat4°C in 20-ml Centricon NS218 (5Ј-GGC GAC CGG TGA GTG GTC CCC CAC TCG GTG CAT- Plus-20 tubes (30-kDa molecular mass cutoff; Millipore, Bedford, MA). 3Ј) and plasmid template pLXSN-Il4i1 (data not shown) using a GeneAmp Cell fractions were prepared from ϳ4 ϫ 106 transfected cells resuspended kit (PerkinElmer, Branchburg, NJ). After DNA sequencing of this PCR fragment cloned into pT7Blue-3 (Novagen, Madison, WI) confirmed that no errors were introduced, the 1.9-kb SmaI/AgeI DNA fragment was in- serted into pDsRed1-N1. The resulting pIl4i1-dsRED plasmid expresses the complete Il4i1 coding sequence translationally fused to the Discosom red fluorescent protein (dsRED). Cells and transfections NIH3T3 cells were cultured in D10 medium (high-glucose DMEM sup- plemented with 10% heat-inactivated FBS (Invitrogen Life Technologies, Carlsbad, CA)) to ϳ50% confluence (ϳ1.5 ϫ 106 cells) in 10-cm-diameter tissue culture dishes (BD Falcon, Bedford, MA) at 5% CO2 and 37°C. After incubating cells in fresh medium for1hat37°C, cells were transfected with 40 ␮g of plasmid DNA using the Calcium Phosphate Trans- fection System (Invitrogen Life Technologies). The next day, cells were washed with PBS (pH 7.2; Invitrogen Life Technologies) to remove the

Ca3(PO4)2/DNA precipitate, and fresh D10 medium was added. Transfection efficiency was determined by fluorescent light microscopy 1 or 2 days post- FIGURE 3. Il4i1 expression plasmid constructs. Schematics of pIl4i1- wash. Transfections with Ͼ30% fluorescent cells were further analyzed. iGFP and pIl4i1-dsRED plasmids are shown to scale. Restriction enzyme sites used in construction are indicated. Coding regions of Il4i1, neomycin Total RNA isolation and RT-PCR resistance gene (neo), EGFP, and Il4i1-dsRED are indicated with solid Total RNA was isolated using the guanidine isothiocyanate protocol (22). arrows (except dsRED portion is gray spotted). The CMV immediate-early In brief, 1 ϫ 106 transfected cells were lysed in guanidine isothiocyanate promoter (Pcmv), neo promoters, and IRES are shown in white. The Journal of Immunology 4563

in 250 ␮l of PBS (pH 7.4) containing protease inhibitors (Sigma-Aldrich, 50 mM final concentration in the assay. Benzoic acid (BA), para-amino- St. Louis, MO): 10 ␮g/ml leupeptin, 10 ␮g/ml aprotinin, 1 mM PMSF, 10 benzoic acid (PABA), 2,3-butanedione (diacetyl; Sigma-Aldrich), and ␮g/ml pepstatin A, and 10 ␮g/ml trypsin inhibitor. Cells were sonicated on 2-hydroxy-3-butynoic acid (2H3B; TCI America, Portland, OR) were ice for 5–10 s at 5 Hz in an Ultrasonic Processor, model XL2010 (Heat added at the indicated inhibitory concentrations. Systems, Farmingdale, NY). The sonicate was centrifuged at 13,000 ϫ g for 10 min at 4°C. The soluble fraction was removed and the insoluble cell Subcellular colocalization pellet was resuspended in 250 ␮l of PBS (pH 7.4) with protease inhibitors. Fractions were stored as 50-␮l aliquots at Ϫ80°C. For deglycosylation, 50 In 35-mm glass-bottom, coated microwell dishes (MatTek Corporation, ␮g of protein was treated with 2 ␮l (1000 U) of PNGase F (New England Ashland, MA), NIH3T3 cells were cotransfected with pIl4i1-dsRED and a Biolabs, Beverly, MA) in manufacturer’sbufferfor1hat37°C. vector expressing green fluorescent fusion protein targeted to a particular Supernatants and soluble or insoluble cell fractions (15 ␮l) were mixed subcellular localization (pEGFP-Peroxi, pEYFP-ER, pEGFP-Endo (BD with an equal volume of 0.001% bromophenol blue (Bio-Rad, Hercules, Clontech), pHYAL2-EGFP (23), PTS2-EGFP (24), or YFP-GL-GPI (25)). CA), 4% SDS, 10% 2-ME, 20% glycerol, and 125 mM Tris (pH 6.8; Sigma-Aldrich), and denatured at 95°C for 5 min. This mixture was separated by SDS-PAGE (4% stacking, 7.5% resolving gel) in duplicate for 2–3 h at 40 mA on a MiniPROTEAN II Electrophoresis Cell (Bio-Rad). To detect proteins, one gel was stained with 0.05% Coomassie brilliant blue R-250 (Bio-Rad) in 40% methanol and 10% glacial acetic acid (Fisher Scientific, Fair Lawn, NJ) overnight and destained by washes in 40% methanol and 10% acetic acid. Proteins on the duplicate gel were transferred onto a Trans-Blot nitro- cellulose membrane (0.45 ␮m; Bio-Rad) overnight at 4°Cat15mAin

transfer buffer (20% methanol, 20 mM Tris (pH 8.3), and 153.6 mM gly- Downloaded from cine (Sigma-Aldrich)) using a Mini Trans-Blot Electrophoretic Transfer Cell (Bio-Rad). Membranes were probed using Living Colors Ds Peptide Ab (anti-dsRED; BD Clontech) at 1/1000 dilution or affinity-purified polyclonal rabbit anti-IL4I1 peptide Ab at 1/2000 dilution (generous gift from Dr. S.-K. Jung (Institute for Virus Research, Kyoto University, Kyoto, Japan)) in TTBS (0.1% Tween 20 (Fisher Scientific) in TBS (pH 7.5)). After washing with TTBS, the primary Ab was detected using the ECL http://www.jimmunol.org/ Western blotting analysis system (Amersham Biosciences, Piscataway, NJ). Briefly, membranes were incubated with HRP-conjugated goat anti- rabbit polyclonal Ab (1/50,000 dilution), washed, and incubated with lu-

minol plus H2O2. Chemiluminescence was detected with Hyperﬁlm ECL (Amersham Biosciences) after 1- to 2-min exposure.

Il4i1 enzyme assay One day postwash, supernatants were collected and the cells were har- vested by trypsinization (one 10-cm-diameter dish yields ϳ2 ϫ 106 cells).

Collected supernatants were filtered through a 0.2-␮m filter before assay. by guest on September 28, 2021 Cell pellets were freeze-thawed three times in microfuge tubes alternately between Ϫ70 and 37°C. After centrifugation at 13,000 ϫ g for 10 min at 4°C, the soluble fraction was removed to a separate tube. The remaining insoluble fraction was resuspended in PBS (pH 7.2; 120 ␮l per dish). All fractions were kept on ice before enzyme assay. In a 96-well flat-bottom Nunc-Immuno Maxisorp plate (Nunc, Roskilde, Denmark), 100 ␮lofL-amino acid or amine substrate (all from Sigma- Aldrich, except for dopamine (American Reagent Labs, Shirley, NY)) at

indicated concentrations or dilution of known concentrations of H2O2 (from 1.47 to 1470 ␮M; Sigma-Aldrich) was added to individual wells in FIGURE 4. IL4I1 is expressed in transient transfectants and glycosy- ␮ ␮ duplicate. A fresh premix of 10 l of 200 U/ml HRP type VI-A, 10 lof lated. A, RNA prepared from NIH3T3 cells transiently transfected (2 days ␮ 10 mg/ml o-phenylenediamine, and 20 l of 500 mM phosphate citrate postwash) with pIl4i1-iGFP, pIl4i1-dsRED, or vector alone (either buffer (pH 5.0) (Sigma-Aldrich) was prepared per well. To the premix was pIRES2-EGFP (iGFP) or pDsRed1-N1 (dsRED)) was tested for Il4i1 and added 60 ␮l of supernatant, cell extract, or H O control. The complete 2 2 ␤ 100-␮l premix was added to the wells, and the 200-␮l reaction mix was -actin expression by RT-PCR. B, Protein from supernatants (Sup.) or soluble (Sol.) and insoluble (Insol.) cell fractions of these transfectants incubated for 2 h at 37°C with atmospheric oxygen in a humidiﬁed 5% CO2 ␮ incubator. The reaction was stopped by addition of 11 lof36NH2SO4. were analyzed by immunoblot using anti-IL4I1 peptide antisera. Molecular Samples were centrifuged at 13,000 ϫ g for 10 min to pellet any insoluble masses (in kilodaltons) of protein standards run in parallel are indicated.

material. Soluble material was transferred to a fresh 96-well plate, and A490 Anti-IL4I1-specific proteins are observed in the Il4i1 expression plasmid was measured on a SPECTRAmax 340 microplate spectrophotometer (Mo- transfectants. Samples from dsRED transfectants had similar background lecular Devices, Sunnyvale, CA). The OD490 background reading from as iGFP transfectants (data not shown). Transfection efficiency assessed by pIRES2-EGFP transfectants was subtracted from that of pIl4i1-iGFP trans- fluorescent microscopy was 40–60% for all samples in this representative fectants. From the H O standard curve, the OD was converted to mi- 2 2 490 experiment. Fluorescent microscopy inspection generally agrees with effi- cromolar concentration of H2O2 generated, equivalent to micromolar concentration of amino acid oxidized based on the enzyme reaction (Fig. 2), ciencies observed by flow cytometry in subsequent experiments. From and used to calculate Il4i1 or LAAO enzyme units (nanomoles of amino many experiments, pIl4i1-iGFP typically produces less anti-IL4I1-specific acid oxidized per minute at 37°C). As a positive control, 10 ␮l of 0.05 protein than pIl4i1-dsRED. This may be due to variation in transfection mg/ml crude snake venom (Crotalus adamanteus) LAAO extract (Sigma- efficiency or protein stability. Small amount of observed soluble IL4I1 may Aldrich) was also tested. To determine specific LAAO activity (units per be genuine or may be due to incomplete fractionation. Cell viability is milligram of total protein), total protein was measured by the method of difficult to assess in transient transfections, but does not appear to be af- Bradford using the Bio-Rad Protein Assay. On average, resuspended infected by Il4i1 expression plasmid transfectants in measurements from sub- Ϯ soluble extracts from pIl4i1-iGFP transfectants had 234 89 mg/ml total sequent experiments (not shown). C, Protein from insoluble cell fraction of protein. To vary the pH, the o-phenylenediamine and premix solutions were pIl4i1-dsRED transfected were untreated or treated with PNGase F buffer prepared with alternate buffers (acetic acid/sodium acetate (pH 4.0); 2- with or without PNGase F enzyme and analyzed by immunoblot with anti- morpholinoethanesulfonic acid/NaOH (pH 6.0); or 3-morpholino-2-hy- dsRED Ab. N-Linked glycosylation sites are not found in dsRED. Thus, all droxypropanesulfonic acid/NaOH (pH 7.0) (Sigma-Aldrich)) to result in a PNGase F deglycosylation is found in IL4I1. 4564 IL4I1 IS A LEUKOCYTE LAAO WITH LYSOSOMAL LOCALIZATION

After 24 h, the cells were washed and cultured for another 48 h. Alterna- of cell lysates (Fig. 4B). Similarly, in pIl4i1-dsRED fusion-trans- tively, 48 h postwash pIl4i1-dsRED-transfected cells were incubated with fected cells, the Il4i1-dsRED fusion protein (ϳ120 kDa) was lo- ␮ 0.5 g/ml acriflavine (Sigma-Aldrich) for 20 min, washed, and cultured in calized mainly to the insoluble fraction as detected by both anti- fresh medium (26). Cells were examined by confocal laser-scanning microscopy (Fluoview 300-IX; Olympus, Melville, NY) to determine subcel- IL4I1 (Fig. 4B) and anti-dsRED Ab (data not shown). lular localization by overlapping fluorescence. The expected sizes for IL4I1 and IL4I1-dsRED are 70–68 and 97–95 kDa based on the predicted amino acid sequence and cleav- Results age of the predicted signal peptide. The larger observed protein High-efficiency transient transfection overexpresses detectable sizes may be due to posttranslational modifications, such as N- Il4i1 protein linked glycosylation at three putative sites conserved between hu- To test whether Il4i1 protein had the predicted LAAO activity, we man and mouse IL4I1 (11). Indeed, like snake venom LAAO (27), attempted to generate stable cell lines that overexpressed Il4i1 pro- IL4I1 is N-linked glycosylated, because PNGase F deglycosylation tein. However, stable lines expressing Il4i1 could not be easily yielded a protein of smaller size (Fig. 4C). Furthermore, transfec- obtained, due to toxic effects of the protein (data not shown). To tants treated with tunicamycin, which prevents glycosylation, examine Il4i1 enzyme activity despite its toxicity, we overex- yielded protein of similar small size (data not shown). pressed Il4i1 protein in high-efficiency transient transfections of Il4i1 enzyme preferentially oxidizes aromatic amino acids NIH3T3 cells. To monitor transfection efficiency, Il4i1 was expressed in the pIRES2-EGFP vector (pIl4i1-iGFP) or as an Il4i1- Supernatant and cell fractions from transfections overexpressing dsRED fusion protein in the pDsRed1-N1 vector (pIl4i1-dsRED). Il4i1 were tested for LAAO enzyme activity using a panel of L- Transfection efficiency was monitored by observing the percentage amino acid and amine substrates (Fig. 5). The supernatant and Downloaded from of fluorescent cells. Cells transiently transfected at Ͼ30% effi- soluble cell fraction had little or no LAAO activity, consistent with ciency expressed Il4i1 mRNA (Fig. 4A). Cells transfected with the amount of Il4i1 protein. The insoluble cell fraction, which pIl4i1-iGFP or pIl4i1-dsRED expressed the expected 267-bp Il4i1 contained the most Il4i1 protein, exhibited easily detectable LAAO RT-PCR fragment, whereas cells transiently transfected with the activity, with highest enzymatic activity using common aromatic parental control plasmids pIRES2-EGFP and pDsRed1-N1 did not. L-amino acid substrates: phenylalanine, tyrosine, and tryptophan. A 440-bp ␤-actin RT-PCR fragment demonstrated that RNA from The remaining common protein encoding L-amino acids, as well as http://www.jimmunol.org/ cells transiently transfected with the plasmids is functional for suc- some uncommon L-amino acids (cystine, L-DOPA, trans-4-hy- cessful reverse transcription. droxy-proline) and amines (dopamine, putrescine) did not exhibit The predicted Il4i1 protein sequence has a typical N-terminal significant activity. Thus, IL4I1 is a newly reported LAAO (EC signal peptide and no hydrophobic transmembrane domain (Fig. 1.4.3.2) with a preference for common aromatic L-amino acids. 1). This suggests that Il4i1 traffics into and through the ER, ulti- mately targeted for secretion or into the lumen of an intracellular organelle. Supernatants, and the soluble and insoluble cell fractions were collected from transfected cells and separated by SDS- by guest on September 28, 2021 PAGE. Staining by Coomassie blue showed that equal amounts of protein were loaded for each fraction (data not shown). Immuno- blot analysis of pIl4i1-iGFP-transfected cells demonstrated that Il4i1 protein (ϳ90 kDa) is located mainly in the insoluble fraction

FIGURE 5. IL4I1 LAAO activity prefers aromatic L-amino acids. In- soluble cell extracts from NIH3T3 cells transiently transfected with pIl4i1- iGFP or pIRES2-EGFP vector were assayed for speciﬁc LAAO activity (units per milligram) at pH 5 with various L-amino acid or amine substrates (common amino acids (standard single letter code), cystine (CC), trans-4- hydroxy-proline (H-Pro), dopamine, L-DOPA, putrescine) at 10 mM ﬁnal concentration under atmospheric oxygen. Background activity of extracts FIGURE 6. IL4I1 is inhibited by aromatic competitors. A, Chemical from pIRES2-EGFP transfections was subtracted. The mean of duplicate structures of amino acid substrates (Phe and Leu) and LAAO inhibitors measurements of a representative of three experiments is shown. Super- (BA, PABA, 2H3B, and diacetyl). B, Il4i1 enzyme activity was measured natants from these cell cultures had no detectable activity (Ͻ1 ϫ 10Ϫ5 as in Fig. 5, except inhibitors were added at the indicated concentration, U/mg). With the common aromatic L-amino acid substrates, soluble cell and 25 mM phenylalanine was used. Percent inhibition of Il4i1 enzyme extracts containing IL4I1 occasionally had barely detectable activity that activity was calculated from samples with no inhibitor added. Represen- was at least 10-fold less than the insoluble cell extracts. tative of two experiments is shown. The Journal of Immunology 4565

Il4i1 is inhibited by aromatic competitors LAAO inhibitors have very different chemical structures. PABA and BA, both with aromatic chemical structures (Fig. 6A), inhibit LAAO by substrate competition (20, 28). Diacetyl and 2H3B, both with nonaromatic chemical structures (Fig. 6A), bind the active site and inactivate arginine residues in the active site (29) or modify the FAD cofactor of LAAO (30, 31), respectively. In contrast to other LAAO, IL4I1 prefers aromatic L-amino acid substrates, suggesting that PABA and BA would inhibit Il4i1 enzyme activity better than diacetyl and 2H3B. At concentrations known to inhibit other LAAO, PABA and BA inhibit Il4i1 enzyme activity in a concentration-dependent manner, with Ͼ50% inhibition at the highest inhibitor concentration (Fig. 6B). For comparison, at 1.0, 5.0, and 10.0 mM PABA and BA, snake venom LAAO activity FIGURE 8. Il4i1 enzyme activity is highest at acidic pH. Insoluble cell was inhibited 26, 80, 100%, and 19, 65, and 100%, respectively extracts were measured as in Fig. 7, except that the highest possible con- (28). In comparison, diacetyl (data not shown) or 2H3B did not centration of phenylalanine (50 mM) was used to detect Il4i1 or snake inhibit Il4i1 enzyme activity at any concentration, including that venom (C. adamanteus) LAAO activity at varying pH. Phenylalanine was insoluble at higher concentrations. The mean of duplicate measurements known to inhibit other LAAO (29–31). Thus, these inhibitor stud-

from a representative of eight experiments is shown. Different pH buffers Downloaded from ies confirm the preference of IL4I1 for aromatic substrates. did not significantly affect the H2O2 standard curves (data not shown), which were used to calculate LAAO activity. Il4i1 enzyme kinetics Under atmospheric oxygen conditions, saturation of the enzyme reaction was obtained with increasing substrate (phenylalanine) the range observed for all substrates of LAAO (0.011–100 mM) concentrations, a behavior typical of an apparent first-order en-

(32, 33). Using phenylalanine as a substrate under atmospheric http://www.jimmunol.org/ zyme reaction that obeys Michaelis-Menten kinetics (Fig. 7A). Using an Eadie-Hofstee plot (Fig. 7B), the apparent maximum velocity (Vmax) of the overall enzyme reaction and the apparent Michaelis-Menten constant Km were estimated to be 0.0099 U/mg and 6.5 mM, respectively. This apparent Km falls within by guest on September 28, 2021

FIGURE 9. IL4I1 localized to lysosomes. Representative NIH3T3 cells transfected with pIl4i1-dsRED and cotransfected with a vector expressing a FIGURE 7. Il4i1 enzyme kinetics. A, Insoluble cell extracts from NIH3T3 green fluorescent fusion protein or costained with acriflavine are shown as cells transiently transfected with pIl4i1-iGFP were assayed for specific LAAO separate red, green, and merged fluorescence images. IL4I1-dsRED colocal- activity at various concentrations of phenylalanine. These values are minus izes with ER (pEYFP-ER) and lysosomal markers (pHYAL2-EGFP, acrifla- background activity from insoluble extracts from vector (pIRES2-EGFP) vine). IL4I1-dsRED does not colocalize with endosome (pEGFP-Endo), per- transfectants. The means and SDs are shown. Logarithmic trend line fitted by oxisome (PTS2-EGFP, pEGFP-Peroxi (not shown)), and plasma membrane least-squares calculation is shown. B, An Eadie-Hofstee plot was used to cal- (YFP-GL-GPI (not shown)) fluorescent fusion proteins. pDsRED1-N1 vector culate the apparent Km (in millimolar concentration) and apparent Vmax (in transfectants had broad red fluorescence throughout the cell that did not exhibit units per milligram) from the best-fit line calculated by linear regression. this punctate pattern and also was not excluded from the nucleus (not shown). 4566 IL4I1 IS A LEUKOCYTE LAAO WITH LYSOSOMAL LOCALIZATION

oxygen conditions, the apparent Km has been reported to vary IL4I1 is the first characterized mammalian LAAO enzyme that from 0.011 to 3.5 mM in LAAO from microbial, reptilian, and is similar to the snake venom LAAO family. Previous studies have avian organisms (33–36). identified mammalian LAAO activity in peroxisomes and mito- chondria of rat kidney or liver (42, 43). This LAAO activity was IL4I1 prefers acidic pH due to the B form of L-␣-hydroxy acid oxidase (EC 1.1.3.15) (30), The pH preference of Il4i1 enzyme activity was evaluated to help which has no sequence similarity with IL4I1 or LAAO from snake provide insight into its cellular localization (Fig. 8). IL4I1 activity venom or fish (44, 45). Thus, we have characterized IL4I1 as a was highest at pH 4.0, suggesting that IL4I1 is most active in an novel mammalian LAAO with unique properties. acidic cellular compartment (lysosome, late endosome) and less Recently, the murine minor histocomptibility locus H46 on active at neutral pH normally found in the cytoplasm (37). For chromosome 7 was identified as a polymorphic peptide derived comparison, we confirmed that LAAO from C. adamanteus, which from endogenous Il4i1 protein prepared by the MHC class II pro- has an apparent pH optimum of 7.2–7.5 (38, 39), is less active at cessing pathway within APC and subsequently presented by MHC acidic pH (40) (Fig. 8). class II (6). Our demonstration that IL4I1 localizes to lysosomes is consistent with the notion that endogenous IL4I1 may be naturally IL4I1 is targeted to lysosomes presented by MHC class II by virtue of its targeting to this processing compartment. In addition, processing of endogenous IL4I1 To determine whether IL4I1 is targeted to an acidic subcellular required the N terminus (6), which contains the N-linked glyco- compartment, we analyzed cells cotransfected with pIl4i1-dsRED sylation sites essential for lysosomal targeting. Interestingly, the and a vector expressing a green fluorescent fusion protein targeted IL4I1 polymorphism that is distinguished as minor H46 is identical Downloaded from to a known subcellular compartment. Visualization by laser-scan- with a polymorphism that distinguishes wild type (BALB/c, ning confocal microscopy demonstrated that IL4I1-dsRED colo- C57BL/6, and NZB) from the NZW allele. calized with protein expressed in the ER as expected by the pres- The Sle3 susceptibility locus, which is derived from the NZW ence of a signal peptide sequence in IL4I1 (Fig. 9). Furthermore, strain, maps to the same region of mouse chromosome 7 as Il4i1 IL4I1-dsRED colocalized with protein or dye targeted to lyso- (1, 9). The role of Sle3 in SLE susceptibility appears to involve somes. However, it did not colocalize with protein targeted to en- APC (12). Il4i1 expression is limited to APC, such as B cells http://www.jimmunol.org/ dosomes, peroxisomes, and plasma membrane. Thus, IL4I1 is a stimulated with IL-4, macrophages, and DC (6, 48). IL-4 may also LAAO targeted to lysosomes. regulate expression of Il4i1 in DC, because human DC were prepared from peripheral blood by culture with GM-CSF and IL-4 Discussion (48). The Sle3 susceptibility locus could involve an Il4i1-mediated To characterize Il4i1 enzyme activity despite its cellular toxicity, function in Ag processing. Indeed, a transfected B cell line over- the gene was transiently overexpressed in NIH3T3 cells at high expressing Il4i1 was more efficient in processing and presenting efficiency. IL4I1 has a prototypic signal peptide sequence directing protein Ag (48). Variations in the amount of Il4i1 protein may it to the lumen of the ER and lacks a hydrophobic transmembrane affect Ag presentation by occupation of MHC class II peptide bind- domain for retention in the cell membrane. Mouse IL4I1 contains ing sites with Il4i1 peptides. Alternatively, changes in Il4i1 LAAO by guest on September 28, 2021 three putative N-linked glycosylation sites that are conserved in activity could affect Ag presentation by several possible mecha- human (11). These sites may contain mannose-6-phosphate, which nisms: 1) directly modifying antigenic peptides by oxidation, 2) targets proteins to lysosomes (41). Indeed, IL4I1 behaves like a additionally cleaving peptides affecting the repertoire of antigenic typical lysosomal enzyme with a preference for acidic pH (Fig. 8). peptides, and/or 3) controlling the rate of amino acid removal, This is unique for a LAAO. Previously reported pH optima of which would indirectly control the rate of upstream peptidases. other LAAO vary between pH 6.6 and 9.5 (32). Decreased IL4I1 Abnormal lysosomal peptidase activity and modified amino acids activity at normal physiological pH may serve as a safety feature, are known to alter immunogenicity and possibly autoimmunity ensuring that IL4I1 does not generate highly toxic H2O2 until the (46, 47). enzyme is sequestered into an acidic intracellular compartment (lysosomes/endosomes). Indeed, IL4I1 was found in lysosomes as Acknowledgments determined by colocalization studies (Fig. 9). This article is dedicated to the memory of Dr. Vincent Massey. We thank Substrate specificities of LAAO measured at atmospheric oxygen Dr. Vincent Massey, Dr. Sandro Ghisla, Dr. Peter Macheroux, Diana vary from species to species. Snake venom LAAO from Calloselasma Chang, Dr. Yih-Pai Chu, Dr. Hsin Chu, Dr. Raul Wapnir, Dr. Saul Teich- rhodostoma, Naja naja kaouthia, and Ophiophagus hannah, like berg, and Dr. Nicholas Chiorazzi for critical review, advice, support, and IL4I1, had highest activity with phenylalanine and tryptophan (33). encouragement; and Dr. Leslie Godwin, Dr. Dorothy Guzowski, and Craig However, unlike IL4I1, these LAAO had high activity with additional Gawel for help with confocal microscopy, oligonucleotide synthesis, and DNA sequencing. L-amino acids (methionine and leucine for C. rhodostoma and N. kaouthia, or lysine, leucine, arginine, and methionine for O. hannah). Snake venom LAAO from Trimeresurus mucrosquamatus had high- References 1. Chu, C. C., and W. E. Paul. 1997. Fig1, an interleukin-4-induced mouse B cell est activity with a nonaromatic L-amino acid, leucine, but also had gene isolated by cDNA representational difference analysis. Proc. Natl. Acad. activity with phenylalanine, methionine, and tyrosine as substrates Sci. USA 94:2507. (28). Finally, apoptosis-inducing protein from fish had LAAO activity 2. Chu, C. C., and W. E. Paul. 1998. Expressed genes in interleukin-4 treated B cells only with lysine as a substrate (16). In comparison, IL4I1 efficiently identified by cDNA representational difference analysis. Mol. Immunol. 35:487. 3. Wain, H. M., E. A. Bruford, R. C. Lovering, M. J. Lush, M. W. Wright, and oxidized only the aromatic L-amino acid substrates: phenylalanine, S. Povey. 2002. Guidelines for human gene nomenclature. Genomics 79:464. tyrosine, and tryptophan (Fig. 5). This preference for aromatic sub- 4. Paul, W. E. 1991. Interleukin-4: a prototypic immunoregulatory lymphokine. Blood 77:1859. strates was supported by the observation that only LAAO inhibitors 5. Schroder, A. J., P. Pavlidis, A. Arimura, D. Capece, and P. B. Rothman. 2002. containing aromatic rings were able to inhibit IL4I1 (Fig. 6). Although Cutting edge: STAT6 serves as a positive and negative regulator of gene expres- the biological significance of these substrate preferences is not known, sion in IL-4-stimulated B lymphocytes. J. Immunol. 168:996. 6. Sahara, H., and N. Shastri. 2003. Second class minors: molecular identification of they may be accounted for by sequence differences in the enzymatic the autosomal H46 histocompatibility locus as a peptide presented by major his- active site (11, 33). tocompatibility complex class II molecules. J. Exp. Med. 197:375. The Journal of Immunology 4567

7. Blake, J. A., J. T. Eppig, J. E. Richardson, and M. T. Davisson. 2000. The Mouse 27. Geyer, A., T. B. Fitzpatrick, P. D. Pawelek, K. Kitzing, A. Vrielink, S. Ghisla, Genome Database (MGD): expanding genetic and genomic resources for the and P. Macheroux. 2001. Structure and characterization of the glycan moiety of laboratory mouse: The Mouse Genome Database Group. Nucleic Acids Res. L-amino-acid oxidase from the Malayan pit viper Calloselasma rhodostoma. Eur. 28:108. J. Biochem. 268:4044. 8. Kono, D. H., R. W. Burlingame, D. G. Owens, A. Kuramochi, R. S. Balderas, 28. Ueda, M., C. C. Chang, and M. Ohno. 1988. Purification and characterization of D. Balomenos, and A. N. Theofilopoulos. 1994. Lupus susceptibility loci in New L-amino acid oxidase from the venom of Trimeresurus mucrosquamatus (Taiwan Zealand mice. Proc. Natl. Acad. Sci. USA 91:10168. habu snake). Toxicon. 26:695. 9. Morel, L., U. H. Rudofsky, J. A. Longmate, J. Schiffenbauer, and 29. Christman, M. F., and J. M. Cardenas. 1982. Essential arginine residues occur in E. K. Wakeland. 1994. Polygenic control of susceptibility to murine systemic or near the catalytic site of L-amino acid oxidase. Experientia 38:537. lupus erythematosus. Immunity 1:219. 30. Cromartie, T. H., and C. T. Walsh. 1975. Rat kidney L-␣-hydroxy acid oxidase: 10. Santiago, M. L., C. Mary, D. Parzy, C. Jacquet, X. Montagutelli, isolation of enzyme with one flavine coenzyme per two subunits. Biochemistry R. M. Parkhouse, R. Lemoine, S. Izui, and L. Reininger. 1998. Linkage of a major 14:2588. quantitative trait locus to Yaa gene-induced lupus-like nephritis in (NZW ϫ 31. Stevens, J. L., P. B. Hatzinger, and P. J. Hayden. 1989. Quantitation of multiple C57BL/6)F1 mice. Eur. J. Immunol. 28:4257. pathways for the metabolism of nephrotoxic cysteine conjugates using selective 11. Chavan, S. S., W. Tian, K. Hsueh, D. Jawaheer, P. K. Gregersen, and C. C. Chu. inhibitors of L-␣-hydroxy acid oxidase (L-amino acid oxidase) and cysteine con- 2002. Characterization of the human homolog of the IL-4 induced gene-1 (Fig1). jugate ␤-lyase. Drug Metab. Dispos. 17:297. Biochim. Biophys. Acta 1576:70. 32. Schomburg, I., A. Chang, and D. Schomburg. 2002. BRENDA, enzyme data and 12. Sobel, E. S., L. Morel, R. Baert, C. Mohan, J. Schiffenbauer, and E. K. Wakeland. metabolic information. Nucleic Acids Res. 30:47. 2002. Genetic dissection of systemic lupus erythematosus pathogenesis: evidence 33. Ponnudurai, G., M. C. Chung, and N. H. Tan. 1994. Purification and properties for functional expression of Sle3/5 by non-T cells. J. Immunol. 169:4025. of the L-amino acid oxidase from Malayan pit viper (Calloselasma rhodostoma) 13. Vyse, T. J., and J. A. Todd. 1996. Genetic analysis of autoimmune disease. Cell venom. Arch. Biochem. Biophys. 313:373. 85:311. 34. Coudert, M. 1975. Characterization and physiological function of a soluble L- 14. Becker, K. G., R. M. Simon, J. E. Bailey-Wilson, B. Freidlin, W. E. Biddison, amino acid oxidase in Corynebacterium. Arch. Microbiol. 102:151. H. F. McFarland, and J. M. Trent. 1998. Clustering of non-major histocompati- 35. Luppa, D., and H. Aurich. 1971. Kinetic studies on the reaction mechanism of bility complex susceptibility candidate loci in human autoimmune diseases. Proc. L -amino acid oxidase from Neurospora crassa. Acta Biol. Med. Ger. 27:839. Downloaded from Natl. Acad. Sci. USA 95:9979. 36. Mizon, J., G. Biserte, and P. Boulanger. 1970. Properties of turkey (Meleagris 15. Moser, K. L., B. R. Neas, J. E. Salmon, H. Yu, C. Gray-McGuire, N. Asundi, gallopavo L.) liver L-amino acid oxidase. Biochim. Biophys. Acta 212:33. G. R. Bruner, J. Fox, J. Kelly, S. Henshall, et al. 1998. Genome scan of human 37. Lodish, H., A. Berk, S. L. Zipursky, P. Matsudaira, D. Baltimore, and J. Darnell. systemic lupus erythematosus: evidence for linkage on chromosome 1q in Afri- 2000. Molecular Cell Biology. Freeman, New York. can-American pedigrees. Proc. Natl. Acad. Sci. USA 95:14869. 38. Wellner, D., and A. Meister. 1960. Crystalline L-amino acid oxidase of Crotalus 16. Jung, S. K., A. Mai, M. Iwamoto, N. Arizono, D. Fujimoto, K. Sakamaki, and adamanteus. J. Biol. Chem. 235:2013. S. Yonehara. 2000. Purification and cloning of an apoptosis-inducing protein derived from fish infected with Anisakis simplex, a causative nematode of human 39. Meister, A., and D. Wellner. 1963. Flavoprotein amino acid oxidase. In The Enzymes, 2nd Ed., Vol. 7. P. D. Boyer, H. Lardy, and K. Myrback, eds. Aca-

anisakiasis. J. Immunol. 165:1491. http://www.jimmunol.org/ 17. Raibekas, A. A., and V. Massey. 1998. Primary structure of the snake venom demic, New York, p. 609. 40. Porter, D. J., and H. J. Bright. 1980. Interpretation of the pH dependence of flavin L-amino acid oxidase shows high homology with the mouse B cell interleukin 4-induced Fig1 protein. Biochem. Biophys. Res. Commun. 248:476. reduction in the L-amino acid oxidase reaction. J. Biol. Chem. 255:2969. 18. Torii, S., M. Naito, and T. Tsuruo. 1997. Apoxin I, a novel apoptosis-inducing 41. Puertollano, R., R. C. Aguilar, I. Gorshkova, R. J. Crouch, and J. S. Bonifacino. factor with L-amino acid oxidase activity purified from Western diamondback 2001. Sorting of mannose 6-phosphate receptors mediated by the GGAs. Science rattlesnake venom. J. Biol. Chem. 272:9539. 292:1712. 19. Pawelek, P. D., J. Cheah, R. Coulombe, P. Macheroux, S. Ghisla, and 42. Nakano, M., M. Saga, and Y. Tsutsumi. 1969. Distribution and immunochemical A. Vrielink. 2000. The structure of L-amino acid oxidase reveals the substrate properties of rat kidney L-amino-acid oxidase, with a note on peroxisomes. Bio- trajectory into an enantiomerically conserved active site. EMBO J. 19:4204. chim. Biophys. Acta 185:19. 20. Curti, B., S. Ronchi, and M. P. Simonetta. 1992. D- and L-Amino acid oxidases. 43. Miller, C. H., and J. A. Duerre. 1969. Oxidative deamination of S-adenosyl-L- In Chemistry and Biochemistry of Flavoenzymes, Vol. III. F. Mu¨ller, ed. CRC, homocysteine by rat kidney L-amino acid oxidase. J. Biol. Chem. 244:4273. ␣ Boca Raton, FL, p. 70. 44. Diep Le, K. H., and F. Lederer. 1991. Amino acid sequence of long chain -hy- by guest on September 28, 2021 21. Kozak, M. 1991. An analysis of vertebrate mRNA sequences: intimations of droxy acid oxidase from rat kidney, a member of the family of FMN-dependent translational control. J. Cell Biol. 115:887. ␣-hydroxy acid-oxidizing enzymes. J. Biol. Chem. 266:20877. 22. Chomczynski, P., and N. Sacchi. 1987. Single-step method of RNA isolation by 45. Belmouden, A., K. H. Le, F. Lederer, and H. J. Garchon. 1993. Molecular cloning acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. and nucleotide sequence of cDNA encoding rat kidney long-chain L-2-hydroxy 162:156. acid oxidase: expression of the catalytically active recombinant protein as a chi- 23. Lepperdinger, G., B. Strobl, and G. Kreil. 1998. HYAL2, a human gene expressed maera. Eur. J. Biochem. 214:17. in many cells, encodes a lysosomal hyaluronidase with a novel type of specificity. 46. Manoury, B., D. Mazzeo, L. Fugger, N. Viner, M. Ponsford, H. Streeter, J. Biol. Chem. 273:22466. G. Mazza, D. C. Wraith, and C. Watts. 2002. Destructive processing by aspar- 24. Ghaedi, K., S. Tamura, K. Okumoto, Y. Matsuzono, and Y. Fujiki. 2000. The agine endopeptidase limits presentation of a dominant T cell epitope in MBP. peroxin Pex3p initiates membrane assembly in peroxisome biogenesis. Mol. Biol. Nat. Immunol. 3:169. Cell 11:2085. 47. Doyle, H. A., and M. J. Mamula. 2001. Post-translational protein modifications 25. Keller, P., D. Toomre, E. Diaz, J. White, and K. Simons. 2001. Multicolour in antigen recognition and autoimmunity. Trends Immunol. 22:443. imaging of post-Golgi sorting and trafficking in live cells. Nat. Cell Biol. 3:140. 48. Chavan, S. S., J. M. Mason, D. Porti, M. Barcia, and C. C. Chu. 2004. Interleu- 26. Davies, J. P., F. W. Chen, and Y. A. Ioannou. 2000. Transmembrane molecular kin-4 induced gene-1. (IL4I1) may affect antigen presenting cell function. In pump activity of Niemann-Pick C1 protein. Science 290:2295. Immunology 2004, Vol. 1. Monduzzi Editore, Montreal, p. 59.