Cloning and Characterization of Mouse Vascular
Adhesion Protein-1 Reveals a Novel Molecule
with Enzymatic Activity
This information is current as
of July 31, 2017.
Subscription
Permissions
Email Alerts
J Immunol 1998; 160:5563-5571; ;
http://www.jimmunol.org/content/160/11/5563
This article cites 32 articles, 12 of which you can access for free at:
http://www.jimmunol.org/content/160/11/5563.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
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 © 1998 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
References
Petri Bono, Marko Salmi, David J. Smith and Sirpa Jalkanen
Cloning and Characterization of Mouse Vascular Adhesion
Protein-1 Reveals a Novel Molecule with Enzymatic Activity1,2
Petri Bono,3 Marko Salmi, David J. Smith, and Sirpa Jalkanen
C
ontinuous recirculation of lymphocytes between blood
and lymphoid organs is fundamental for the maintenance
of an adequate immune response against antigenic challenge caused by microbial and inflammatory agents (1, 2). In lymphatic tissues, lymphocytes leave the blood by selectively binding
to specialized venules called high endothelial venules (HEV)4 and
then extravasate between the endothelial cells into the surrounding
lymphatic tissue (3). At the molecular level, lymphocyte binding to
HEV is mediated by several lymphocyte surface receptors that
bind to their specific ligands on the endothelial cell surface during
a successful multistep emigration cascade (1, 2, 4, 5).
Human vascular adhesion protein-1 (VAP-1) is an important
endothelial cell adhesion molecule involved in the binding of lymphocytes to endothelium (6). In vivo human VAP-1 (hVAP-1) is a
170- to 180-kDa dimer consisting of two identical subunits of 90
kDa (7). It is expressed mainly in the HEV of peripheral lymph
node (PLN)-type lymphatic tissues (6), and the surface expression
of VAP-1 is up-regulated after prolonged chronic inflammation in
different tissues, such as gut, tonsil, skin, and synovium (8, 9).
Based on the reactivity with the original anti-VAP-1 mAb 1B2,
MediCity Research Laboratory, University of Turku, and National Public Health Institute, Turku, Finland
Received for publication November 10, 1997. Accepted for publication January
30, 1998.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
This work was supported by the Finnish Academy, the Finnish Cancer Union, the
Finnish Cultural Foundation, and the Sigrid Juselius Foundation.
2
Sequence data have been submitted to the EMBL/GenBank databases under accession number AF054831.
3
Address correspondence and reprint requests to Dr. Petri Bono, MediCity Research
Laboratory, University of Turku, Tykistökatu 6A, FIN-20520 Turku, Finland. E-mail
address: [email protected]
4
Abbreviations used in this paper: HEV, high endothelial venule; VAP-1, vascular
adhesion protein-1; hVAP-1, human vascular adhesion protein-1; PLN, peripheral
lymph node; PNAd, peripheral lymph node addressin; mVAP-1, mouse vascular adhesion protein-1; GST, glutathione S-transferase; FAD, flavin adenine dinucleotide;
BSAO, bovine serum amine oxidase; SSAO, semicarbazide-sensitive amine oxidase.
Copyright © 1998 by The American Association of Immunologists
VAP-1, with a slightly different molecular mass, can also be found
in smooth muscle cells and follicular dendritic cells, although its
function on these cells is unknown.
VAP-1 is distinct from the peripheral lymph node addressin
(PNAd), defined by mAb MECA-79 (10), although both VAP-1
and PNAd can mediate lymphocyte binding to PLN under shear
conditions. However, in contrast to PNAd, VAP-1 can operate in
an L-selectin-independent manner and support the binding of Lselectin-negative lymphocytes to the endothelium (11). Studies of
tonsillar VAP-1 using digestion with specific glycosidases have
shown that VAP-1 is a sialoglycoprotein, probably containing both
N- and O-linked sugars with abundant sialic acid residues. It has
also been shown that sialic acids are required for the adhesive
properties of VAP-1, since the desialylated molecule can no longer
support lymphocyte binding in the frozen section assay (7).
We have recently isolated the cDNA encoding the human
VAP-1 and shown that it encodes a novel functional adhesion molecule (38). The hVAP-1 cDNA encodes a type II transmembrane
protein of 84.6 kDa with six potential N-glycosylation sites and
three potential O-glycosylation sites in the long extracellular domain. The hVAP-1 cDNA transfected Ax endothelial cells that
express VAP-1 on their surface bind lymphocytes, and the binding
can be inhibited with anti-hVAP-1 mAbs. The hVAP-1 protein
sequence has no similarity to any known adhesion molecules, but
it has significant identity to a family of enzymes called the coppercontaining amine oxidases.
In this paper we describe the isolation of mouse VAP-1 cDNA
and production of a mAb against this mouse molecule. The isolated cDNA encodes a previously unknown mouse molecule with
highly significant identity to human VAP-1 and also to a family of
enzymes called copper-containing amine oxidases. We also show
that this mouse molecule possesses enzyme activity against monoamines and characterize its m.w. and tissue distribution with the
new anti-mVAP-1 mAb. Taken together, these results suggest that
we have cloned mouse VAP-1, which will allow us to study in
vivo the function and significance of VAP-1 in lymphocyte
recirculation.
0022-1767/98/$02.00
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
Human vascular adhesion protein-1 (VAP-1) is a sialylated endothelial cell adhesion molecule mediating the initial L-selectinindependent interactions between lymphocytes and endothelial cells in man. In this work we cloned and characterized mouse
VAP-1 (mVAP-1) and produced an anti-mVAP-1 mAb against a recombinant mVAP-1 fusion protein. The isolated cDNA encodes
a novel 84.5-kDa mouse molecule. The anti-mVAP-1 mAb stained high endothelial venules in peripheral lymph nodes, and smooth
muscle cells and lamina propria vessels in gut. During immunoblotting, this anti-mVAP-1 mAb recognized a 110/220-kDa Ag,
suggesting that mVAP-1 is a dimer. Since mVAP-1 has significant sequence identity to members of a family of enzymes called the
copper-containing amine oxidases, we showed that mVAP-1 possesses monoamine oxidase activity. Thus, mVAP-1 is the first
mouse membrane-bound amine oxidase identified at the molecular level. Based on the 83% identity between the isolated cDNA
and human VAP-1 cDNA, the expression pattern, the molecular mass, and the enzyme activity against monoamines, the cloned
molecule represents a mouse homologue of human VAP-1. Cloning of mVAP-1 provides a valuable tool for in vivo studies of the
significance of VAP-1 for lymphocyte-endothelial cell interactions and of the possible relationship between leukocyte adhesion and
amine oxidase activity. The Journal of Immunology, 1998, 160: 5563–5571.
5564
Materials and Methods
Isolation of mVAP-1 cDNA
Northern blot analysis
A mouse multiple tissue Northern blot was obtained from Clontech Laboratories (Palo Alto, CA) and was hybridized according to the manufacturer’s instructions at 65°C using the PCR-amplified 800-bp mVAP-1 PCR
fragment as a probe.
oma supernatant (TK10 –79) positively stained the sections and was subcloned twice by limiting dilution before use in additional experiments.
Transfections and cell culture
For transfections, the mVAP-1 cDNA fragment covering 1 to 3101 bp was
isolated from pUC18 after EcoRI and XbaI digestions and subcloned into
the pcDNA3 expression vector (Invitrogen, San Diego, CA). CHO cells
were used as hosts in transfections to express the mVAP-1 protein, and
they were grown in a-MEM (Life Technologies, Paisley, U.K.) plus CHO
nucleosides supplemented with 20% FCS, 2 mM glutamine (Biological
Industries, Beit Haemek, Israel), 128 U/ml penicillin, and 128 mg/ml streptomycin. Electroporation was performed using a Bio-Rad Gene Pulser apparatus (Richmond, CA; 0.3 kV, 960 mF, 0.4-cm cuvette in RPMI plus 1
mM sodium pyruvate, 2 mM L-glutamine without serum, and 20 mg of
expression plasmid). These stably transfected cells were selected by culturing them in the presence of 0.5 mg/ml geneticin (Life Technologies,
Grand Island, NY).
Immunoblotting
mVAP-1 CHO transfectants were scraped from cell culture flasks and lysed
with lysis buffer (150 mM NaCl, 10 mM Tris base (pH 7.2), 1.5 mM
MgCl2, 1% Nonidet P-40, 1% aprotinin, and 1 mM PMSF). After 2 h at
7°C, the lysates were centrifuged (13,000 rpm, 30 min, 7°C), and the supernatants were collected. Noninflamed mouse gut samples from BALB/c
mice were dissected free from the lamina propria. The smooth muscle wall
was minced into small pieces, lysed 2 h in the lysis buffer at 7°C, and
centrifuged, and the supernatants were collected. Twenty-five microliters
of the gut smooth muscle lysate supernatant or CHO cell lysate supernatant
was mixed with an equal volume of Laemmli’s sample buffer for electrophoresis under nonreducing conditions. Before loading on a 5 to 12.5%
SDS-PAGE gel, the samples were boiled for 5 min at 95°C. The resolved
proteins were transferred onto nitrocellulose sheets (Hybond-ECL, Amersham) by a Hoefer electroblotter (San Francisco, CA). Nitrocellulose strips
were stained using an enhanced chemiluminescence detection kit (Amersham) according to the manufacturer’s recommendations. Briefly, blocking
was performed with PBS containing 10% nonfat milk powder and 0.3%
Tween-20 for 1 h, and a 1/20 dilution of the hybridoma TK10 –79 supernatant was used as a primary Ab. The mAb Hermes-1 was used as a negative control Ab. Peroxidase-conjugated goat anti-rat Ig containing 5%
normal mouse serum was used in the second stage of staining.
Abs and immunohistochemistry
Amine oxidase assays
An anti-mVAP-1 mAb was produced by immunizing rats with recombinant
mVAP-1. In brief, soluble mVAP-1 fusion protein was produced by amplifying a 117-bp long C-terminal part of the mVAP-1 cDNA with PCR
primers MVAPGST8; GTC AGG ATC CGG GAG GGC CAG GAT G
(designed from the mVAP-1 protein sequence REGQDA; residues 726 –
731 of the predicted protein sequence) and MVAPGST9; TCC CGA ATT
CTG TCT CTG TAA GCA AAG C (designed from the protein sequence
GFAYRDN; residues 759 –765, respectively). Reaction conditions in the
amplification were 94°C for 1 min, 60°C for 1 min, and 72°C for 1 min for
30 cycles. The amplified fragment was digested with BamHI and EcoRI
and ligated into BamHI- and EcoRI-digested pGEX3X vector (glutathioneS-transferase (GST) gene fusion vector, Pharmacia). A portion of the fusion protein was run on SDS-PAGE gel and stained with Coomassie blue
to confirm the purity and solubility of the resulting GST-mVAP-1 fusion
protein. Finally, the fusion protein was mixed with IFA and used to immunize specific pathogen-free Sprague-Dawley rats. After three s.c. injections into the foodpads at 1-wk intervals, the rats were sacrificed, and the
popliteal lymph nodes were collected. The isolated lymphocytes were then
fused with nonsecreting Sp2/0 myeloma cells using standard procedures.
The resulting hybridoma supernatants were tested by immunoperoxidase
staining of acetone-fixed cryostat sections obtained from specific pathogenfree BALB/c mice gut. The sections were incubated with primary Abs (100
ml of hybridoma supernatant from mVAP-1 fusions or 100 ml of supernatant from a hybridoma producing a negative control mAb, Hermes-1,
which is a rat mAb against human CD44) (13) for 120 min, washed twice
in PBS, and incubated for 30 min with peroxidase-conjugated goat anti-rat
Ig (Dako, Glostrup, Denmark) containing 5% normal mouse serum. The
reaction was developed for 5 min using 39,39-diaminobenzidine hydrochloride in PBS containing 0.03% hydrogen peroxide as a chromogen. After the
stainings, the sections were counterstained with hematoxylin. One hybrid-
The mVAP-1-transfected and mock (pcDNA3 vector) transfected CHO
cells (typically 10 –15 3 106) were scraped into 10 ml of PBS and centrifuged at 4°C. The cell pellet was washed twice with PBS and finally resuspended in 1 ml of PBS before sonication on ice at medium power
(Braun sonicator, Melsungen, Germany). Sonicated lysates were used in
the assay immediately or after storage at 220°C. All enzyme reactions
were performed with or without heating the cell lysates for 5 min at 95°C
before addition of the substrate. The enzyme activity assay protocol was
modified from the method described by D’Agostino (14). Briefly, in benzylamine assay, 100 ml of the cell lysate was mixed with 300 ml of 100 mM
NaPO4 and 100 ml of the substrate, and in the putrescine assay, 100 ml of
the cell lysate was mixed with 1.7 ml of 100 mM NaPO4, 200 ml of 400
mM acetaldehyde, and 100 ml of the substrate. [14C]benzylamine hydrochloride (0.2 mCi; Amersham) in 100 mM NaPO4 containing 0.8 mM
unlabeled benzylamine (Sigma, St. Louis, MO) or 0.1 mCi of [14C]putrescine dihydrochloride (Amersham) plus 0.04 mM unlabeled putrescine
(Sigma) were used as substrate in these reactions. One milligram of porcine
kidney diamine oxidase (Sigma) was used as a positive control for diamine
oxidase activity, and 0.5 mg of bovine plasma monoamine oxidase was
used as a control source for monoamine oxidase activity. The reactions
were incubated at 37°C for 60 min before double extraction of the labeled
aldehyde reaction products in 700 ml of toluene-based scintillation mixture
containing toluene and 0.35 g/l diphenyloxazole. The extractions were
measured for 14C by a liquid scintillation counter (15). Hydroxylamine
hydrochloride (5 mM; Sigma) and semicarbazide hydrochloride (0.1 mM;
Sigma) were used as specific inhibitors (16) and incubated for 10 min with
100 ml of the cell lysate before addition of the substrate. Total protein
concentration measurements were performed using the Bradford method
with bovine g-globulin as a standard. The results are presented as picomoles of substrate used per minute per milligram of protein.
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
Total RNA from BALB/c mouse kidney was prepared using an Ultraspec
kit (Biotecx, Houston, TX), and a mVAP-1 PCR fragment was amplified
using partially degenerate oligonucleotide primers PB3 and PB5 (KEBO
Laboratory, Espoo, Finland): PB3 TTTTGTGTGTTYGARCARAAY (designed from the hVAP-1 protein sequence FCVFEQN; residues 429 – 435
of the predicted hVAP-1 protein sequence) and PB5 GTTRGGAATRT
CYTCWGCRTG (designed from the hVAP-1 protein sequence HAEDIPN; residues 687– 693, respectively). The reaction conditions in PCR
were 94°C for 1 min, 50°C for 1 min, and 72°C for 1 min, repeated for 30
cycles. The amplified 800-bp fragment was then blunt end cloned into
pUC18 vector using a SureClone kit (Pharmacia, Uppsala, Sweden). The
cloned DNA fragment was sequenced using a Sequenase version 2.0 kit
(U.S. Biochemical Corp., Cleveland, OH) and the sequencing service facilities of the Department of Medical Genetics, University of Turku
(Turku, Finland). Sequence analysis and database comparisons were performed using the Wisconsin Package version 8.1-UNIX of the Computer
Group (Madison, WI).
Approximately 1 3 106 recombinant l phage from a mouse heart 59
STRETCH PLUS cDNA library in the lgt10 cloning vector (Clontech
Laboratories, Palo Alto, CA) were screened with the reverse transcriptasePCR-amplified mVAP-1 PCR fragment from mouse kidney mRNA. The
fragment was labeled using an Amersham Multiprime DNA labeling kit
and [a-32P]dCTP at about 3000 Ci/mmol (Amersham International, Aylesbury, U.K.). Approximately 100,000 plaques/plate were transferred to
Hybond-N nylon filters (Amersham), and duplicate filters were hybridized
at 65°C in 53 SSC, 53 Denhardt’s reagent, 0.5% SDS, and 0.5 mg/ml
denatured, sheared salmon sperm DNA. The filters were washed twice for
30 min each time at 65°C in 0.1 3 SSC and 0.1% SDS and autoradiographed with intensifying screens (Biomax MS film, Eastman Kodak,
Rochester, NY). One positive clone was identified, which was subjected to
secondary screening, from which a single plaque was purified, and the
insert of this clone was subcloned into EcoRI-digested pUC18 for sequencing. Plaque hybridization, l phage purification, Escherichia coli transformation, and plasmid purifications were all performed according to standard
molecular biology protocols (12).
CLONING OF MOUSE VAP-1
The Journal of Immunology
Results
Isolation of a mVAP-1 cDNA
The mouse VAP-1 protein and its homology studies
The open reading frame encoded a 765-amino acid protein with a
predicted molecular size of 84.5 kDa (Fig. 1). The mVAP-1 protein has six potential N-glycosylation sites and six potential Oglycosylation sites (according to the O-glycosylation site prediction server, [email protected] (17)). The predicted amino acid
sequence revealed that the identity of mVAP-1 to hVAP-1 is 83%,
distributed evenly throughout the molecules. A search of the most
recent releases from SwissProt and GenEMBL sequence databanks
revealed that mVAP-1 has no significant identity to any other adhesion molecule or mouse protein, but it has, like hVAP-1, significant identity to a family of enzymes called copper-containing
amine oxidases.
Amine oxidases are enzymes classified into two main classes on
the basis of the chemical nature of their cofactors (16, 18): the
flavin adenine dinucleotide (FAD)-containing amine oxidases and
the copper-containing amine oxidases. The FAD-containing amine
oxidases include both the A and B forms of the widely studied
outer mitochondrial membrane monoamine oxidases, which possess activity not only against primary amines but also against secondary and tertiary amines (19). In addition to the differences in
their cofactors, the copper-containing amine oxidases are distinct
from the FAD-containing amine oxidases with regard to their location inside the cell and their substrate specificity. A multiple
alignment of mVAP-1, hVAP-1, and five other cloned mammalian
members of the copper-containing amine oxidases is shown in Figure 2. The identity of mVAP-1 to these molecules varied between
42 and 95%. The highest identity (95%) was found to the recently
published partial cDNA sequence of a rat membrane-bound amine
oxidase (20), which together with hVAP-1 comprises the first identified transmembrane protein members of this family. Figure 2
shows also that the hydrophobic amino acids present in the rat
(residues 6 –26) and human (residues 5–27) membrane-spanning
regions are 100% identical between mouse and rat molecules and
highly conserved between mouse and human VAP-1. In contrast,
the putative transmembrane sequence in the mouse has no homology to other members of the copper-containing amine oxidase
family (for example bovine serum amine oxidase (BSAO)), which
are secreted proteins. This indicates that mVAP-1 probably has a
transmembrane domain close to the N-terminus of the molecule
and that it belongs to the membrane-bound glycoprotein group of
the copper-containing amine oxidase family.
Expression of mVAP-1
In Northern blot analysis of mRNA from BALB/c mice using the
800-bp mVAP-1 PCR fragment as a probe, a single 4.4-kb band
was found to hybridize from different tissues (Fig. 3). No other
mRNA species were detected even after prolonged autoradiography (data not shown). The expression levels varied; they were
strong in heart, lung, skeletal muscle, kidney, and testis. In liver,
brain, and spleen only low amounts of mVAP-1 mRNA were detected in longer exposures. The expression pattern of mVAP-1
mRNA is very similar to that of hVAP-1, except that in mouse
brain a weak mRNA message was detected during the prolonged
exposure, whereas human brain did not express any detectable
hVAP-1 mRNA in Northern blot analysis (data not shown).
The mVAP-1 possesses enzyme activity against monoamines
The fact that mVAP-1 is the first cloned mouse molecule with
significant identity to proteins belonging to the amine oxidase family (16) led us to test whether mVAP-1 possesses enzyme activity
against different amines. In enzyme assays, mVAP-1-transfected
CHO cell lysates were assayed with benzylamine (a monoamine)
or putrescine (a diamine) as a substrate. In these experiments
mVAP-1 showed a clear activity against benzylamine but not
against putrescine (Table I), and the activity could be destroyed by
heating the cell lysates for 5 min at 95°C before addition of the
substrate. Mock transfected CHO cell lysates did not show activity
against any substrates. Porcine kidney diamine oxidase (for diamine oxidase activity) and bovine plasma monoamine oxidase (for
monoamine oxidase activity) were used as positive controls (18,
21) against putrescine and benzylamine, respectively, and the activities of these controls were abolished by heating. To test
whether mVAP-1 activity against benzylamine could be inhibited
by semicarbazide or hydroxylamine, two known inhibitors of copper-containing amine oxidases (16), these inhibitors were incubated separately with the mVAP-1-transfected CHO cell lysates
before addition of the substrate. In the presence of 0.1 mM semicarbazide or 5 mM hydroxylamine, mVAP-1 showed no activity
against benzylamine (Table II). In conclusion, mVAP-1 has a clear
activity against benzylamine but not against putrescine (a diamine), and the activity is inhibitable with the known copper-containing amine oxidase inhibitors.
Production of an anti-mVAP-1 mAb
To study the expression of mVAP-1 and the molecular size of
mVAP-1, an anti-mVAP-1 mAb was produced. First, a C-terminal
117-bp fragment of the mVAP-1 cDNA (coding for amino acids
726 –765 in the mVAP-1 protein sequence) was amplified with
PCR and cloned into the pGEX3X vector to produce a GSTmVAP-1 fusion protein. After isopropylthiogalactoside induction
and purification of the resulting recombinant protein from E. coli,
part of the sample was run on an SDS-PAGE gel to test its purity
and solubility. Because the resulting recombinant protein was soluble and of the expected size, the rest of this material was used to
immunize rats from which popliteal lymph node lymphocytes were
collected after 3 wk and fused with Sp2/0 mouse myeloma cells.
The resulting hybridoma supernatants were screened by immunohistologic staining of frozen tissue sections. Based on the staining
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
Based on the hVAP-1 sequence, several degenerate oligonucleotide primers were designed to be used in reverse transcriptase-PCR
experiments with mRNA isolated from different mouse tissues. A
single 800-bp fragment of the expected size was amplified from
mouse kidney mRNA with the degenerate oligonucleotides PB3
and PB5 as primers, and this specific fragment was cloned for
sequencing. The sequence revealed that the amplified 800-bp fragment had 80% amino acid identity to hVAP-1, indicating that the
amplified fragment could represent mVAP-1. PCR screening of
different mouse libraries showed that a mouse heart cDNA library
was positive with the same primer pair, so this library was selected
for screening with the PCR-amplified mVAP-1 fragment as a
probe to search for the full-length mVAP-1 cDNA. One phage
with a large 4.4-kb insert was isolated, cloned, and sequenced, and
the preliminary sequence revealed that the isolated cDNA contained the whole 800-bp PCR fragment sequence. Thus, the whole
isolated insert was sequenced in two directions. The insert had an
ATG starting codon 350 bp down from the 59 end of the cDNA and
was followed by one continuous open reading frame of 2298 bp
(Fig. 1). A 59 untranslated region of 349 bp and a 39 untranslated
region of 1830 bp following the TGA stop codon were present in
the isolated clone. At the 39 end of the clone, a polyadenylation
signal and a poly(A) tail were found, suggesting that the isolated
clone also included the whole 39 end of the cDNA encoding
mVAP-1.
5565
5566
CLONING OF MOUSE VAP-1
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
FIGURE 1. The complete nucleotide sequence of the mVAP-1 cDNA and the predicted amino acid sequence. Potential N-glycosylation sites (asparagines) are circled, and potential O-glycosylation sites are boxed. The putative transmembrane part of mVAP-1 in residues 6 to 26 is boxed and shaded.
The important tyrosine for the enzyme function in residue 471 (see Discussion) is in bold and underlined together with the consensus sequence NYD in
residues 470 to 472. The three conserved histidines that act as a ligand for copper are also shown in italics and underlined in residues 520, 522, and 684.
The 39 polyadenylation signal is in bold and underlined. The sequence data have been submitted to the EMBL/GenBank databases under accession number
AF054831.
The Journal of Immunology
5567
reactivity of all anti-hVAP-1 mAbs with smooth muscle in addition to endothelial cells, we decided to test the hybridomas on
mouse gut sections. One of the hybridoma supernatants positively
stained these gut sections, while the negative controls showed no
reactivity, and the staining pattern resembled the that of antihVAP-1 mAbs on human gut sections.
To test the specificity of the new mAb, designated TK10-79,
lysates from CHO mVAP-1 transfectants were tested in immunoblotting. Staining of the filter with mAb TK10-79 revealed specific
bands with apparent molecular masses of 110 and 220 kDa from
CHO mVAP-1 transfectants under nonreducing conditions (Fig. 4,
lane 1), while the mock transfectants (Fig. 4, lanes 2 and 4) and
control Ab stainings were negative. These results confirmed that
mAb TK10-79 is specific for mVAP-1.
Expression of mVAP-1 in gut and PLN
The resulting new anti-mVAP-1 mAb TK10-79 was used in immunoperoxidase stainings to characterize the expression of
FIGURE 3. Northern blot analysis of mVAP-1 mRNA. Poly(A)1 RNA
isolated from the indicated tissues (a mouse multiple tissue Northern blot
from Clontech) was probed with an 800-bp mVAP-1 fragment. No other
mRNA species than the 4.4-kb form were detected even after prolonged
exposure. Each lane contains approximately 2 mg of poly(A)1 RNA.
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
FIGURE 2. A multiple alignment of the predicted mVAP-1 protein sequence with hVAP-1 and other cloned members of the copper-containing amine
oxidase family. The numbers on the left of each row correspond to the first amino acid of the lane. The putative transmembrane part of mVAP-1 (residues
6 –26) is 100% identical with the transmembrane residues of the rat membrane-bound amine oxidase (residues 6 –26). PDAO1, human placental diamine
oxidase 1 (35); PDAO2, human placental diamine oxidase 2 (35); BSAO (36); RatDAO, rat diamine oxidase (37); RatAO, rat membrane amine oxidase
(partial sequence) (20). The sequence alignments were performed with GCG Pileup, and residues with identity to mouse VAP-1 were identified using GCG
Boxshade.
5568
CLONING OF MOUSE VAP-1
Table I. mVAP-1 is a monoamine oxidase
Enzyme Activity with Different Substratesb
pmol min21 mg prot21 (mean 6 SD)
Samplea
Benzylamine
Putrescine
CHO-mVAP
CHO-mock
DAO
MAO
1449 6 73.1
20.80 6 3.75
2.28 6 0.1
20.03 6 0.04
18.3 6 1.2
528 6 22.3
CHO-mVAP 5 CHO cells transfected with pcDNA3 containing mVAP-1
cDNA; CHO-mock 5 CHO cells transfected with pcDNA3; DAO 5 porcine kidney
diamine oxidase, a positive control for diamine oxidase activity; MAO 5 bovine
plasma monoamine oxidase, a positive control for monoamine oxidase activity.
b
The values in the table are presented as the means of five independent experiments (performed as triplicates).
a
The mVAP-1 is a 110/220-kDa molecule
To determine the molecular size of the Ag that the mAb TK10-79
recognizes in tissue sections and to compare the size of this Ag to
that of mVAP-1 CHO transfectants, lysates from mouse gut
smooth muscle and from CHO mVAP-1 transfectants were run in
parallel on SDS-PAGE and used for immunoblotting with the mAb
TK10-79. From mouse gut smooth muscle cell lysate, mAb
TK10-79 revealed, under nonreducing conditions, a band of approximately 110 kDa and a smeared band of around 220 kDa (Fig.
4, lane 5), indicating that mAb TK10-79 is detecting in vivo the
same size molecule as that from transfectants. The negative control
Ab staining of the gut smooth muscle lysate was negative. The
220-kDa form of the molecule from both mVAP-1 CHO transfectants and mouse gut smooth muscle lysates probably represents a
dimeric form, which is a typical feature for hVAP-1 (see Footnote
5) and a subgroup of membrane-bound amine oxidases (semicarbazide-sensitive amine oxidases (SSAO), a subgroup of coppercontaining amine oxidases (16, 22)).
Discussion
In this work we have cloned and characterized a previously unknown mouse molecule and used the cloned cDNA to produce a
Table II. SSAO inhibitors block the mVAP-1 enzyme activity
Enzyme Activity with Different Substratesa
pmol min21 mg prot21 (mean 6 SD)
Sample
Bzb
Bz 1
semicarbazideb
Bz 1
hydroxylamineb
CHO-mVAP
CHO-mock
1449 6 73.1
20.80 6 3.75
43.0 6 11.5
6.16 6 0.21
18.2 6 1.49
2.26 6 4.59
a
The values in the table are presented as the means of five independent experiments (performed as triplicates).
b
All of the assays were performed with benzylamine (Bz) as a substrate (as
described in Table I) with or without SSAO inhibitors (semicarbazide or hydroxylamine).
FIGURE 4. mAb TK10-79 detects a 110- and a 220-kDa molecule in
immunoblotting. Mouse gut smooth muscle lysate (gut sm) and CHO cell
lysates were analyzed under nonreducing conditions using SDS-PAGE and
immunoblotting. In CHO-mVAP transfectants, mAb TK10-79 reacted with
110- and a 220-kDa molecule (lane 1), and in gut smooth muscle lysate,
mAb TK10-79 detected a 110-kDa band and a smeared 220-kDa band
(lane 5). In control reactions with an irrelevant mAb or with mock transfectants (lanes 2– 4 and 6), no specific bands were seen. Molecular mass
standards are indicated on the left in kilodaltons.
specific mAb against this novel protein. The protein is named
mVAP-1 because it represents the mouse homologue of hVAP-1
based on the following criteria. 1) The predicted protein sequence
from the isolated cDNA has significant identity (83%) to hVAP-1
protein. 2) The anti-mVAP-1 mAb TK10-79 that was raised
against the C-terminal end of the molecule detects in immunoblotting 110- and 220-kDa proteins, indicating that mVAP-1 is probably, like human VAP-1, a dimer and composed of two identical
subunits. 3) In immunoperoxidase stainings, the anti-mVAP-1
mAb stained cells in mouse gut smooth muscle and mouse PLN
tissue sections in a manner similar to anti-hVAP-1 mAb staining
on corresponding human tissues. 4) In Northern blot experiments,
the tissue distributions of mVAP-1 and hVAP-1 mRNA are very
similar.
The hVAP-1 has been shown to be a transmembrane molecule
with a short four-amino acid intracellular part and a long extracellular domain (see Footnote 5). Very recently, a new rat transmembrane protein belonging to the same copper-containing amine
oxidase family as hVAP-1 has been partly cloned, and the predicted protein sequence of this cloned fragment of the molecule
has 95% identity to mVAP-1 and 83% identity to hVAP-1, indicating that the published partial cDNA is probably encoding the rat
homologue of mouse and human VAP-1 (20). Because the transmembrane-spanning amino acids among these three molecules
(Fig. 2, residues 5–26 in hVAP-1 and 6 –26 in rat AO) are highly
conserved, mVAP-1 is probably also a type II transmembrane protein with the N-terminus of the molecule located intracellularly
and the C-terminus located extracellularly (23). The predicted
mVAP-1 amino acid sequence revealed that mVAP-1 has six potential N-glycosylation and six putative O-glycosylation sites, and
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
mVAP-1 in mouse PLN and gut. In mouse gut sections, mAb
TK10-79 stained smooth muscle cells (Fig. 5A) and faintly stained
vessels of the lamina propria (Fig. 5B). The staining pattern of
mVAP-1 in noninflamed gut resembles the staining pattern of
hVAP-1 in human gut, with hardly any staining of mucosal HEV
in Peyer’s patches. In mouse PLN, anti-mVAP-1 mAb reactivity
was mainly seen on HEV (Fig. 5D) where the initial lymphocyte
binding to endothelium and the transmigration take place. mAb
TK10-79 did not stain any lymphocytes present in PLN. Thus,
mVAP-1, like hVAP-1 (8), is present in both endothelial cells and
smooth muscle cells.
The Journal of Immunology
5569
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
FIGURE 5. Distribution of mVAP-1 in mouse gut and PLN. During immunoperoxidase staining of mouse gut (A) with anti-mVAP-1 mAb TK10-79,
mVAP-1 is expressed in smooth muscle cells of the bowel wall (Bw) and muscularis mucosae (Mm) and also in the lamina propria vessels (B, thin arrows).
The negative control staining of the gut (C) lacks any specific signals. In PLN (D), mVAP-1 is faintly expressed in HEV (thick arrows), whereas
lymphocytes are negative. In the control staining of PLN (E), a negative HEV is indicated by arrowheads. The original magnifications were 3100 (A),
3400 (B, D, and E), and 3200 (C).
all the potential glycosylation sites are located in the putative extracellular part of the molecule. Thus, different oligosaccharides
may also play a role in the function of mVAP-1 as an adhesion
molecule, as has been shown for hVAP-1 (7). The functional importance of these possible oligosaccharide modifications of
mVAP-1 and the adhesion properties of CHO-mVAP-1 transfectants remain to be determined, since the anti-mVAP-1 mAb
TK10-79 is apparently not a function-blocking mAb in adhesion
assays. This is not surprising, since the mAb was raised against a
short peptide sequence of mVAP-1 that was expressed in a
bacterial host.
In immunoblotting experiments the anti-mVAP-1 mAb detected
110- and 220-kDa Ag from mouse gut smooth muscle lysate and
from CHO mVAP-1 transfectants. Both 110- and 220-kDa forms
of the detected molecule are larger than the corresponding hVAP-1
bands (;90/170 kDa) detected by immunoblotting (7). This may
be due to the species-specific glycosylation differences, since the
core proteins of the molecules are almost equal in size, whereas the
number of potential glycosylation sites is even higher in mVAP-1
than in the hVAP-1 protein core.
The ligands of both hVAP-1 and mVAP-1 are unknown. The
predicted protein sequence of hVAP-1 (see Footnote 5) revealed at
residues 726 to 728 an RGD motif that is an important integrin
binding site (24) present in a variety of integrin ligands. At present,
the significance of this RGD motif for the function of hVAP-1 is
unknown. Although this motif is not completely conserved between hVAP-1 and mVAP-1, residues 726 to 728 in mVAP-1
(REG) may also represent a possible integrin binding site, since a
5570
specificity and enzyme kinetics of these enzymes is available for
BSAO and rat SSAO. We have cloned a novel mouse molecule
with activity against monoamines that is inhibitable by carbonyl
reagents such as semicarbazide and hydroxylamine. Thus, the
cloned mouse molecule most likely belongs to the same SSAO
subgroup of amine oxidases as hVAP-1 and represents the first
cloned mouse SSAO.
Relatively little is known about the biologic role of SSAO, especially in higher eukaryotes. One problem has been the different
substrate specificities, which can limit the value of studies with
different laboratory animals in attempts to define the physiologic
role of primary amine oxidases in man. The identification and
cloning of a mouse molecule with similar enzyme activity to the
human homologue will be a valuable tool in future studies on the
biologic role of these enzymes. Now it will be possible to investigate in a mouse model whether enzyme activity and lymphocyte
adhesion are linked or independent functions.
Although the significance of hVAP-1 as a functionally important endothelial cell adhesion molecule that can mediate early interactions between endothelial cells and lymphocytes in the normal
recirculation of lymphocytes as well as in the extravasation of
lymphocytes into sites of inflammation has been well established
in vitro (6 –9, 11), only limited information on the functional significance of this endothelial cell adhesion molecule in vivo is
available. With intravital microscopy, VAP-1 has been shown to
be involved in the initial interaction of lymphocytes and endothelial cells in inflamed rabbit mesenterial venules (11). However, a
mouse model would be extremely useful for further analyses of the
site of VAP-1 in the multistep adhesion cascade and of the significance of VAP-1 in vivo. In this work we have isolated a cDNA
encoding mVAP-1, produced an Ab against it, and characterized
with this mAb the mouse homologue of hVAP-1. The cloning of
the mVAP-1 gene is now in progress, and the targeted disruption
of the gene will give valuable information about the role of this
adhesion molecule in vivo.
Acknowledgments
We thank M. Pohjansalo, T. Kanasuo, and M. Iljamo for excellent technical
assistance, and A. Sovikoski-Georgieva for secretarial help.
References
1. Springer, T. A. 1994. Traffic signals for lymphocyte recirculation and leukocyte
emigration: the multistep paradigm. Cell 76:301.
2. Butcher, E. C., and L. J. Picker. 1996. Lymphocyte homing and homeostasis.
Science 272:60.
3. Girard, J.-P., and T. A. Springer. 1995. High endothelial venules (HEVs): specialized endothelium for lymphocyte migration. Immunol. Today 16:449.
4. Butcher, E. C. 1991. Leukocyte-endothelial cell recognition: three (or more) steps
to specificity and diversity. Cell 67:1033.
5. Salmi, M., and S. Jalkanen. 1997. How do lymphocytes know where to go:
current concepts and enigmas of lymphocyte homing. Adv. Immunol. 64:139.
6. Salmi, M., and S. Jalkanen. 1992. A 90-kilodalton endothelial cell molecule
mediating lymphocyte binding in humans. Science 257:1407.
7. Salmi, M., and S. Jalkanen. 1996. Human vascular adhesion protein-1 (VAP-1)
is a unique sialoglycoprotein that mediates carbohydrate-dependent binding of
lymphocytes to endothelial cells. J. Exp. Med. 183:569.
8. Salmi, M., K. Kalimo, and S. Jalkanen. 1993. Induction and function of vascular
adhesion protein-1 at sites of inflammation. J. Exp. Med. 178:2255.
9. Arvilommi, A.-M., M. Salmi, K. Kalimo, and S. Jalkanen. 1996. Lymphocyte
binding to vascular endothelium in inflamed skin revisited: a central role for
vascular adhesion protein-1 (VAP-1). Eur. J. Immunol. 26:825.
10. Streeter, P. R., B. T. N. Rouse, and E. C. Butcher. 1988. Immunohistologic and
functional characterization of a vascular addressin involved in lymphocyte homing into peripheral lymph nodes. J. Cell Biol. 107:1853.
11. Salmi, M., S. Tohka, E. L. Berg, E. C. Butcher, and S. Jalkanen. 1997. Vascular
adhesion protein 1 (VAP-1) mediates lymphocyte subtype-specific, selectin-independent recognition of vascular endothelium in human lymph nodes. J. Exp.
Med. 186:589.
12. Sambrook, J. F., T. Maniatis, and E. F. Fritsch. 1989. Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring
Harbor, NY.
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
cyclic peptide containing the sequence RE has been shown to be
able to interact with the same or an overlapping binding site in the
a5b1 integrin as RGD (25). In principle, in RE, the basic group of
arginine is one amino acid closer to the acidic group of the aspartate than in RGD, making the sequences REG and RGD closely
related in spatial terms (E. Ruoslahti, unpublished observations).
In addition to the high homology to hVAP-1, the isolated
mVAP-1 cDNA showed significant identity in database comparisons to enzymes belonging to the family of copper-containing
amine oxidases, but no similarity to FAD-containing amine oxidases, which include monoamine oxidases A and B. The coppercontaining amine oxidases are enzymes that catalyze the oxidative
deamination of various biogenic amines, and they have widely
differing substrate specificities (21). They are composed of two
identical subunits, each containing one copper atom and one covalently bound carbonyl-containing cofactor, the chemical identity
of which is still controversial. It is not clear whether the cofactor
is the same in each member of the family or whether the function
of all these enzymes is actually copper dependent (26). The cofactor at the active site of BSAO has been identified as topa-quinone (6-hydroxydopa) (27). The topa-quinone is formed from a
precursor tyrosine that has to be part of a consensus sequence
Asn-Tyr-Asp/Glu to become post-translationally modified to topaquinone (28). This active site consensus sequence is also found in
the mVAP-1 sequence in residues 470 to 472. The crystal structure
of one copper-containing amine oxidase (E. coli copper amine
oxidase) has been published (29), and it confirmed the importance
of the modified tyrosine residue in the active site and the existence
of three histidines that also have to be conserved for enzyme activity (27, 30). These three histidines that coordinate copper binding in the protein core can also be found in mVAP-1, and they are
located in positions 520, 522, and 684 in mVAP-1 cDNA sequence
(Fig. 1).
In addition to identification of the structural requirements for the
enzyme activity in mVAP-1, we have directly shown that mVAP-1
is enzymatically active. The mVAP-1 cDNA-transfected CHO cell
lysates had activity against benzylamine, but not against putrescine, indicating that mVAP-1 has activity at least against
monoamines but probably not against diamines. Our results differ
from previous reports in which tissue-bound amine oxidase from
mouse white adipose tissue homogenate has been shown to possess
activity not only against benzylamine but also against acetylputrescine (31). Tissue-specific substrate differences may be one explanation for this detected difference, whereas the most likely explanation is that crude homogenate of the white adipose tissue
included more than one amine oxidase that were not separated
before the enzyme activity measurements. However, our findings
show that the substrate specificity of mVAP-1 closely resembles
the specificity of hVAP-1, suggesting that the cloned mouse and
human VAP-1 may be close homologues with regard to their substrates in addition to their structural identity. This finding is of
great importance, as much species-dependent variation in substrate
specificity of amine oxidases has been reported. For example,
some amines, such as tyramine and tryptamine, have been shown
to be good substrates for the rat enzyme but poor substrates for the
human enzyme (32, 33).
The hVAP-1 has been shown to be an ecto-enzyme and to include a quinone cofactor (see Footnote 5). The enzyme activity of
hVAP-1 is inhibitable by semicarbazide and hydroxylamine; based
on these criteria, hVAP-1 belongs to a subgroup of amine oxidases
commonly referred to as SSAO, which are mammalian coppercontaining monoamine oxidases found in various mammalian tissues and organs both in the cell membrane and in a soluble form
in serum (34). The most detailed information about the substrate
CLONING OF MOUSE VAP-1
The Journal of Immunology
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
enzymes: 6-hydroxydopa at the active site of bovine serum amine oxidase. Science 248:981.
Janes, S. M., M. M. Palcic, C. H. Scaman, A. J. Smith, D. E. Brown,
D. M. Dooley, M. Mure, and J. P. Klinman. 1992. Identification of topaquinone
and its consensus sequence in copper amine oxidases. Biochemistry 31:12147.
Parsons, M. R., M. A. Convery, C. M. Wilmot, K. D. S. Yadav, V. Blakeley,
A. S. Corner, S. E. V. Phillips, M. J. McPherson, and P. F. Knowles. 1995.
Crystal structure of a quinoenzyme: copper amine oxidase of Escherichia coli at
2 Å resolution. Structure 3:1171.
Fontecave, M., and H. Eklund. 1995. Copper amine oxidase: a novel use for a
tyrosine. Structure 3:1127.
Raimondi, L., R. Pirisino, G. Banchelli, G. Ignesti, L. Conforti, E. Romanelli, and
F. Buffoni. 1992. Further studies on semicarbazide-sensitive amine oxidase activities (SSAO) of white adipose tissue. Comp. Biochem. Physiol. B Biochem.
Mol. Biol. 102:953.
Barrand, M. A., and B. A. Callingham. 1982. Monoamine oxidase activities in
brown adipose tissue of the rat: some properties and subcellular distribution.
Biochem. Pharmacol. 31:2177.
Precious, E., and G. A. Lyles. 1988. Properties of a semicarbazide-sensitive
amine oxidase in human umbilical artery. J. Pharm. Pharmacol. 40:627.
Lyles, G. A., A. Holt, and C. M. S. Marshall. 1990. Further studies on the metabolism of methylamine by semicarbazide-sensitive amine oxidase activities in
human plasma, umbilical artery and rat aorta. J. Pharm. Pharmacol. 42:332.
Zhang, X., J. Kim, and W. S. McIntire. 1995. cDNA sequences of variant forms
of human placenta diamine oxidase. Biochem. Genet. 33:261.
Mu, D., K. F. Medzihradszky, G. W. Adams, P. Mayer, W. M. Hines,
A. L. Burlingame, A. J. Smith, D. Cai, and J. P. Klinman. 1994. Primary structures for a mammalian cellular and serum copper amine oxidase. J. Biol. Chem.
269:9926.
Lingueglia, E., S. Renard, N. Voilley, R. Waldmann, O. Chassande,
M. Lazdunski, and P. Barbry. 1993. Molecular cloning and functional expression
of different molecular forms of rat amiloride-binding proteins. Eur. J. Biochem.
216:679.
Smith, D. J., M. Salmi, P. Bono, J. Hellman, T. Leu, and S. Jalkanen. 1998.
Cloning of vascular adhesion protein-1 reveals a novel multifunctional adhesion
molecule. J. Exp. Med. In press.
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
13. Jalkanen, S. T., R. F. Bargatze, L. R. Herron, and E. C. Butcher. 1986. A lymphoid cell surface glycoprotein involved in endothelial cell recognition and lymphocyte homing in man. Eur. J. Immunol. 16:1195.
14. D’Agostino, L., B. Daniele, S. Pignata, G. D’Argenio, and G. Mazzacca. 1989.
Modifications in enterocyte diamine oxidase distribution induced by heparin in
the rat. Biochem. Pharmacol. 38:47.
15. Okuyama, T., and Y. Kobayashi. 1961. Determination of diamine oxidase activity
by liquid scintillation counting. Arch. Biochem. Biophys. 95:242.
16. Lyles, G. A. 1996. Mammalian plasma and tissue-bound semicarbazide-sensitive
amine oxidases: biochemical, pharmacological and toxicological aspects. Int.
J. Biochem. Cell Biol. 28:259.
17. Hansen, J. E., O. Lund, J. Engelbrecht, H. Bohr, J. O. Nielsen, J.-E. S. Hansen,
and S. Brunak. 1995. Prediction of O-glycosylation of mammalian proteins: specificity patterns of UDP-GaINAc: polypeptide N-acetylgalactosaminyltransferase.
Biochem. J. 308:801.
18. Lewinsohn, R. 1984. Mammalian monoamine-oxidizing enzymes, with special
reference to benzylamine oxidase in human tissues. Braz. J. Med. Biol. Res.
17:223.
19. Lewinsohn, R., K.-H. Böhm, V. Golver, and M. Sandler. 1978. A benzylamine
oxidase distinct from monoamine oxidase B: widespread distribution in man and
rat. Biochem. Pharmacol. 27:1857.
20. Morris, N. J., A. Ducret, R. Aebersold, S. A. Ross, S. R. Keller, and
G. E. Lienhard. 1997. Membrane amine oxidase cloning and identification as a
major protein in the adipocyte plasma membrane. J. Biol. Chem. 272:9388.
21. Gorkin, V. Z. 1983. Amine oxidases in clinical research. Pergamon Press, Oxford, p. 282.
22. Barrand, M. A., and B. A. Callingham. 1985. The interaction of hydralazine with
a semicarbazide-sensitive amine oxidase in brown adipose tissue of the rat: its use
as a radioactive ligand for the enzyme. Biochem. J. 232:415.
23. Hartmann, E., T. A. Rapoport, and H. F. Lodish. 1989. Predicting the orientation
of eukaryotic membrane-spanning proteins. Proc. Natl. Acad. Sci. USA 86:5786.
24. Ruoslahti, E. 1991. Integrins. J. Clin. Invest. 87:1.
25. Koivunen, E., B. Wang, and E. Ruoslahti. 1994. Isolation of a highly specific
ligand for the a5b1 integrin from a phage library. J. Cell Biol. 124:373.
26. Lyles, G. A. 1995. Substrate-specificity of mammalian tissue-bound semicarbazide-sensitive amine oxidase. Prog. Brain Res. 106:293.
27. Janes, S. M., D. Mu, D. Wemmer, A. J. Smith, S. Kaur, D. Maltby,
A. L. Burlingame, and J. P. Klinman. 1990. A new redox cofactor in eukaryotic
5571