| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, May 16, 1994; and in revised form, October 20, 1994) From the
Several receptor tyrosine kinases generate soluble ligand
binding domains either by differential splicing resulting in a
truncated RNA transcript, or by proteolytic cleavage. Although the
exact role in vivo of these soluble extracellular domains is
unclear, proteolysis may function to down-regulate the receptor, and
soluble extracellular domains (ECD) may compete with the intact
receptor binding to ligand. Axl is a member of a new class of receptor
tyrosine kinases characterized by an ECD resembling cell adhesion
molecules and unique sequences in the kinase domain. In addition, Axl
is transforming in both fibroblast and hematopoietic cells, and appears
to be involved in mesenchymal development. We now find that Axl is
post-translationally processed by cleavage in a 14 amino acid region
immediately NH The ECD ( In
addition to these cis regulatory functions, processed and soluble ECDs
of various receptors also appear to modulate receptor function. The
most notable examples of biologically functional secreted receptors are
the forms of the interleukin-6 and ciliary neurotrophic factor
receptors which are capable of interacting with their respective
ligands and gp130 to trigger transmembrane
signaling(10, 11) . In RTKs, a secreted form of the
EGFR ECD is produced as the result of alternative splicing of the
message(12, 13, 14) . The function of this
soluble ECD is unclear but may serve to regulate the activity of the
full-length receptor(14, 15) . Additionally, soluble
forms of CSF-1R, MET, and HER-2/neu have also been detected in the CM
of cells expressing the
receptor(16, 17, 18, 19) . In
contrast to the soluble EGFR, the soluble CSF-1R and MET ECDs arise
through proteolytic cleavage of the intact receptor by an, as yet,
unknown protease. This proteolytic process is regulated in part by
protein kinase C and is thought to down-regulate the receptor. The
mechanism which generates the soluble neu/erbB2 protein is not
clear but may also involve proteolytic processing(18) . Thus,
proteolytic cleavage may represent a general mechanism to modulate RTK
function. In this study, we describe the processing of the Axl RTK
at the cell surface. This receptor was first identified in the DNA of
affected cells from leukemia patients employing
transfection/tumorigenicity assays and is capable of transforming
murine fibroblast and a myeloid cell line through
overexpression(20, 21, 22) . We now
demonstrate that cells expressing Axl produce a soluble protein
corresponding to the ECD of Axl (Axl
The Axl-expressing cell lines AF6295,
TF14B12, and TF17B were generated as described previously(20) .
AF6295 cells express genomic sequences, whereas TF14B12 and TF17B are
independent clones derived from NIH-3T3 cells transfected with axl cDNAs. TF14B12 overexpresses axl cDNA1-4 which
encodes the full-length 894 amino acid Axl protein
(Axl
For
analysis of secreted Axl proteins, cells were grown in 10-cm plates to
approximately 75% confluence and then placed in 7 ml of serum-free
media. After 24 h, CM from cell lines were removed and a portion either
concentrated on Centricon-10 ultrafilters as per the
manufacture's instructions or lyophilized and resuspended in 1
Figure 2:
[
Figure 5:
Production of Axl
Figure 1:
An antibody directed against the
amino-terminal immunoglobulin domain of Axl recognizes Axl-specific
proteins in the lysates and conditioned media (CM) from Axl-expressing
cells. For analysis of CM, 4 ml of CM from AF6295 or NIH/3T3 cells were
concentrated to 1 ml. Five µg of affinity-purified anti-Igl were
added to 1 ml of concentrated CM and 1 ml of cell lysate.
Immunoprecipitations were performed as described under
``Experimental Procedures.'' Immune complexes were
resuspended in 50 µl of 4
Since AF6295 cells were transformed by the
human Axl genomic DNA, it is possible that Axl
Figure 4:
Axl protein expression in various cell
lines. A, cell lysates were analyzed as in Fig. 1using
the anti-IgL antibody except one-fifth of the immunoprecipitate from
each lysate was separated by SDS-PAGE on a 10% polyacrylamide gel. B, for analysis of CM, 500 µl of medium from each cell
line was concentrated on a Centricon-30 ultrafilter, resuspended in a
final volume of 80 µl 1
Axl
Figure 3:
N-Glycanase treatment of
Axl
We reported previously the
isolation of two alternatively spliced forms of the Axl message. These
transcripts encode proteins that either contain or lack 9 amino acids
carboxyl-terminal to the fibronectin III domains in the ECD of Axl (Fig. 6)(20) . These two forms are designated
Axl
Figure 6:
Schematic representation of the predicted
Axl protein structure. The immunoglobulin (IgL) and
fibronectin III (FNIII) domains are indicated with arrows. The amino acid sequence of Axl between the final FNIII
domain and the transmembrane domain is shown to the right. The boxed 9 amino acids correspond to the differentially spliced
Axl-box region(20) . The remaining 14 amino acids represent the
putative protease cleavage site.
Figure 7:
The proteolytic production of Axl
Figure 8:
Proteolytic cleavage of Axl results in the
production of a COOH-terminal phosphorylated protein. AF6295 cells were
grown as in Fig. 1. Medium was removed and cells washed once
with PBS and then placed in serum-free medium ± TPA (100
nM) for 0, 1, 2, or 4 h. Cells were then lysed and equivalent
amounts of protein from each sample fractionated by SDS-PAGE in a 7.5%
polyacrylamide gel. Western blots of gels were probed with antibodies
directed against phosphotyrosine (
Figure 9:
The COOH-terminal Axl cleavage product,
Axl
Figure 10:
Proteolytic cleavage of Axl occurs at the
plasma membrane. AF6295 cells were placed on ice for 15 min. Growth
medium was supplemented with 20 mM HEPES, pH 7.3, to prevent
media from becoming basic during the course of the experiment. Medium
was removed and cells were washed once with ice-cold PBS then placed in
7 ml of cold serum-free medium with (+) or without(-) TPA
(100 nM) at 4 °C. CM from each sample was analyzed by
Western blot using anti-IgL as described under ``Experimental
Procedures.'' The secreted Axl ECD is designated with an arrow.
Axl is a novel RTK isolated from patients with myeloid
leukemic disorders(20, 21) . The structure of Axl,
with its ECD resembling cell adhesion molecules and the intracellular
domain bearing the unique consensus kinase sequence KW(I/L)A(I/L)E,
defines a new class of receptor tyrosine kinases which include
v-ryk, sky/tif/rse, eyk, and the neural
specific kinases tyro3/brt, and
tyro12(28, 29, 30, 31, 32, 33) .
We have demonstrated previously that Axl is transforming due to
overexpression as opposed to genetic alteration (20) and have
shown (foster) ligand-dependent transforming activity in the murine
interleukin-3-dependent 32D myeloid cell line(22) . During
mouse development, axl is expressed in the developing
mesenchyme and may be involved in organogenesis(34) . In the
adult, axl message is detected in a variety of tissues and at
especially high levels in the heart, skeletal muscle, and ovarian
follicles, as well as in myeloid precursors in the bone
marrow(27, 34) . In this study, we have examined the
biochemical regulation of the receptor at the cell surface. Using
antibodies directed against different epitopes in the ECD, we
demonstrate that a truncated form of the receptor, Axl Processing of Axl is regulated in part
by protein kinase C as demonstrated by the increase in cleavage of the
intact receptor upon phorbol ester treatment of Axl-expressing cells.
This proteolytic cleavage is independent of protein synthesis as
treatment of AF6295 cells with cycloheximide does not affect the
TPA-stimulated release of Axl Other RTKs have been reported to undergo similar
processing events as seen with Axl(16, 19) . Both MET
and CSF-1R are cleaved by an as yet unknown protease to release soluble
ligand binding domains. In addition, secreted isoforms of other RTKs,
such as EGFR, are produced as a result of alternative
splicing(13, 35) . In both cases, the presence of a
soluble ligand binding domain suggests that this truncated protein may
function in signal transduction as a possible inhibitor of the
membrane-bound receptor. Consistent with this possibility, Flickinger et al.(35) have shown that a secreted form of the
chicken EGFR is able to block the transforming growth
factor- Using an
antibody directed against the intracellular portion of Axl, we detected
a membrane-bound, truncated kinase fragment, Axl In some cases,
removal of the ECD of RTKs has also been shown to activate the kinase
domain. For example, removal of the ECD of the sevenless receptor, a protein involved in the development of the compound
eye in Drosophila, results in supernumary photoreceptor cells,
a phenotype characteristic of constitutive signaling through the sevenless pathway(9) . Furthermore, truncation of the
EGFR ECD as a result of retroviral integration by avian leukosis virus
or artificially by recombinant approaches constitutively activates the
receptor(37, 7) . In the case of Axl, Western blot
analysis of cell lysates from AF6295 cells treated with TPA revealed
tyrosine phosphorylation of the 55-kDa reminant COOH-terminal kinase
domain of Axl (Fig. 9). Since tyrosine phosphorylation of the
enzymatic portion of tyrosine kinases is a marker of enzymatic
activation, we suspect that the proteolytically generated Axl We have shown that Axl is
proteolytically cleaved to release a soluble extracellular domain as
well as a membrane-bound, truncated kinase domain. This finding adds to
the growing list of RTKs that are naturally modulated by specific
proteolysis. Although it is unclear whether this event is up-regulatory
or down-regulatory, the specificity of the cleavage and its modulation
by such agents as TPA suggest that this post-translational process is
important in the functional regulation of RTKs such as CSF-1R, MET, and
Axl and that proteases regulate these processes.
Volume 270,
Number 2,
Issue of January 13, 1995 pp. 551-557
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
-terminal to the transmembrane domain
resulting in a soluble ECD and a membrane bound kinase domain. The
sequence of this putative cleavage site shares no homology with
recognition sites of known proteases. Characterization of this
proteolytic processing shows that it does not require protein synthesis
or transport but is augmented by phorbol ester treatment. Since the
cleavage of Axl enhances turnover of the kinase on the cell surface, we
suggest that proteolytic processing down-regulates Axl kinase activity.
)of RTKs plays a regulatory role in
controlling receptor kinase activity. Studies of several retrovirally
transduced oncogenes including v-kit, v-ros, and
v-erbB suggest that truncation of the ECD of RTKs renders
these proteins
oncogenic(1, 2, 3, 4) . Indeed, in vitro deletion of the ECD of a number of RTKs, including
the EGFR, the insulin receptor, and neu/erbB2 renders these
receptors transforming in the absence of
ligand(5, 6, 7) . Studies of the sevenless (sev) RTK of Drosophila provide
compelling evidence for the importance of the ECD in regulation of the
activity of RTKs. sev is required for the proper development
of the R7 photoreceptor of the compound eye(8) .
Loss-of-function sev alleles result in the absence of R7
photoreceptor cells. Conversely, mutations which activate the sev pathway, such as removal of the ECD of sev, result in
supernumerary R7 cells(9) . Flies that contain this truncated sev gene do not require the presence of the sev ligand encoded by the bride-of-sevenless gene (boss). These results are evidence that deletion of the
extracellular ligand binding domain relieves negative constraints that
regulate the kinase activity of the receptor at the cell surface.) which arises in
vivo through post-translational proteolytic processing of the
full-length receptor and may play a role in Axl-dependent signaling.
Materials
Tissue culture plasticware was
purchased from Corning. Cell culture media, FBS, PBS, and antibiotics
were purchased from the Lineberger Comprehensive Cancer Center Tissue
Culture Facility. N-Glycanase was obtained from Genzyme.
Immobilon-P membranes were purchased from Millipore. Amicon was the
source of Centricon ultrafiltration devices. The Tissue-Tearor
homogenizer was obtained from Fisher Scientific. Horseradish peroxidase
conjugates of anti-rabbit IgG and anti-mouse IgG, ECL chemiluminescence
reagent kits and Hyperfilm-ECL were purchased from Amersham Corp.
Isotopically labeled compounds and EN
HANCE were obtained
fron DuPont NEN. X-Omat AR x-ray film was purchased from Eastman Kodak
Co. Protein A-Sepharose, anti-phosphotyrosine monoclonal antibody
(PT-66), and other reagents were purchased from Sigma.Cell Culture and Immunoprecipitation
All cells
were grown at 37 °C in the presence of 5% CO, 95% air.
A549, a human lung carcinoma cell line, T24, a human bladder carcinoma
cell line which expresses a mutated H-ras allele, NIH/3T3
cells, and axl-transfected NIH/3T3 cells were grown in
Dulbecco's modified Eagle's medium containing high glucose
(4500 mg/liter) and supplemented with 10% FBS. The human
proerythroblastic leukemia cell line K562 was grown in RPMI 1640 media
supplemented with 10% FBS. K562 cells were differentiated along the
megakaryocytic pathway by the addition of TPA as described
previously(23) ., (20) ). TF17B overexpresses axl cDNA6-2 which encodes an Axl isoform lacking 9 amino acids
amino-terminal to the transmembrane domain (Axl
, (20) ).
Production of Axl-specific Antibodies
A fragment
of Axl corresponding to amino acids 33-136 was generated by
polymerase chain reaction amplification. This fragment was cloned into
the bacterial expression vector pCFM1656(24) . Recombinant
protein was expressed as described previously(25) . Inclusion
bodies were solubilized in 5 M urea, 50 mM Tris-HCl,
pH 8.0, 10 mM DTT and protease inhibitors. The soluble protein
was fractionated on a Q-Sepharose column. Axl protein was pooled, and
the urea was removed by sequential dialysis. The dialyzed pool was
reloaded on a Q-Sepharose column and step eluted with 0.25 M NaCl. The eluted protein was then fractionated by gel filtration
in the presence of PBS. This material was injected into rabbits for the
generation of polyclonal antibodies by standard methods.Analysis of Cell Lysates and CM for Axl
Proteins
Adherent cells were grown in 10-cm plates to
approximately 75% confluence, washed with cold PBS, and lysed in 1 ml
of cold Nonidet P-40 lysis buffer (1% Nonidet P-40, 20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 10%
glycerol, 0.1 mg/ml PMSF, 100 µM NaVO
). All immunoprecipitations were
performed at 4 °C. Samples were gently rotated for 30 min and
centrifuged to pellet insoluble material. Affinity-purified anti-IgL (5
µg/ml lysate) was added to a fraction of the samples and then
incubated with gentle rotation for 1 h. Protein A-Sepharose beads (50
µl/ml lysate; Sigma) were added to samples and incubation continued
for an additional h. Immune complexes were recovered by centrifugation
for 3 min followed by a brief wash with cold lysis buffer. Samples were
resuspended in 1 
Laemmli gel sample loading buffer (10 mM Tris-HCl, pH 7.8, 3% SDS, 5% glycerol, 2% 2-mercaptoethanol, and
0.05% bromphenol blue), boiled, resolved by SDS-PAGE, transferred to
Immobilon-P filters, and analyzed by Western blotting techniques as
described previously(20) . Filters were blocked for 1 h in TBST
(50 mM NaCl, 10 mM Tris-HCl, pH 7.5, 0.05% Tween-20)
containing 1% bovine serum albumin. Filters were then incubated with
antibodies as described previously (20) . After incubation with
secondary antibody, filters were extensively washed with TBST,
incubated with ECL reagents and then placed on Hyperfilm-ECL. Laemmli gel sample loading buffer. Samples were boiled,
resolved by SDS-PAGE, and then transferred to Immobilon-P filters and
analyzed by Western blotting as above.[
AF6295 cells (5 S]Methionine Labeling of AF6295
Cells
10
) were seeded in a
3.5-cm plate. On the following day, cells were washed once with PBS and
placed in methionine free-Dulbecco's modified Eagle's
medium, 10% FBS. After growth for 1 h at 37 °C, the medium was
supplemented with 50 µCi L-[S]methionine and cells were grown
for an additional 1 (Fig. 2) or 4 h (Fig. 5B).
Cultures were then washed once in PBS, placed in unlabeled complete
medium (time 0), and lysed in Nonidet P-40 lysis buffer at the
designated time points. For phorbol ester treatment, TPA was added
following the 4 h labeling and 1-h incubation in unlabeled medium (time
0). Cells were then lysed after the designated time points following
TPA addition. Axl proteins were immunoprecipitated from half of the
lysate with anti-IgL antibody as described above. Half of the
immunoprecipitate was fractionated by SDS-PAGE with a 7.5%
polyacrylamide gel. Gels were fixed for 1 h in methanol/acetic
acid/water (1:3:6) and then placed in EN
HANCE reagent for
30 min followed by a 30-min incubation in cold water. Gels were dried
and exposed to Kodak XAR film for 3 days at -70 °C with an
intensifying screen.
S]Methionine pulse
labeling of AF6295 cells reveal that Axl
is the fully
processed form of the holoenzyme. Cells (5
10) were
grown in a 3.5-cm plate in the presence of
[S]methionine followed by a 1-h chase with cold
methionine for the designated number of hours (0-4 h). Cells were
then lysed and Axl proteins immunoprecipitated as described under
``Experimental Procedures.'' The immunoprecipitate was
fractionated by SDS-PAGE on a 7.5% polyacrylamide gel. Thereafter, the
gel was fixed for 1 h in methanol/acetic acid/water (1:3:6). Signal was
amplified by the addition of EN
HANCE reagent for 30 min
followed by a 30-min incubation in cold water. The gel was then dried
and placed on XAR film for 3 days at -70 °C with an
intensifying screen. The positions of the fully processed Axl receptor
(Axl) and the partially processed forms (Axl
and Axl
) are indicated with arrows. The
positions of molecular mass markers are indicated with dashes.
is
augmented by phorbol esters. A, AF6295 cells were grown in 7
ml of complete medium in the presence or absence of 100 nM TPA. At 0, 1, 2, 4, 8, 12, and 24 h, 1-ml samples were removed,
filtered, and frozen. The equivalent of 70 µl of CM from each
sample was fractionated by SDS-PAGE in a 10% polyacrylamide gel,
transferred to Immobilon-P filters, and the Western blot incubated with
either the anti-IgL or anti-Axl-box antibody (indicated above each gel)
as described under ``Experimental Procedures.'' Dashes indicate the position of molecular mass markers of 97, 68, and 46
kDa. -, TPA not added; +, TPA added. B, AF6296
cells were grown in media containing
[
S]methionine for 4 h, washed with PBS, and then
placed in fresh medium to chase label. TPA (100 nM) was added
to fresh medium after 1 h (time 0). Axl proteins were
immunoprecipitated from cell lysates (B) and fractionated by
SDS-PAGE in a 7.5% polyacrylamide gel and processed as described under
``Experimental Procedures.'' Exposure time was 28 h at
-70 °C with an intensifying screen. -, TPA not added;
+, TPA added.
N-Glycanase Treatment of Secreted Axl
Proteins
Samples of 24-h CM from AF6295 cells (50 µl from a
total of 7 ml) and A549 cells (400 µl from a total of 7 ml) were
lyophilized and then resuspended in 10 ml of a solution containing 0.4 M NaP
, pH 8.6, 0.5% SDS, 53.4 mM 2-mercaptoethanol. Samples were boiled for 5 min followed by the
addition of 5 µl of 7.5% Nonidet P-40, 13 µl of HO,
1 µl of PMSF (10 mg/ml), and 1 unit of N-glycanase and
overnight incubation at 37 °C. Reactions were stopped by the
addition of 15 µl of 4 Laemmli gel-loading buffer. Samples
were boiled, fractionated by SDS-PAGE with a 10% polyacrylamide gel,
transferred to Immobilon-P filters, and analyzed by Western blotting
with anti-IgL as described above.Cycloheximide Treatment of Axl-expressing
Cells
10-cm plates containing equivalent numbers of AF6295 cells
were grown in complete medium in the presence or absence of
cycloheximide (10 µg/ml) for 30 min. Cells were then washed once
with PBS and placed in serum-free medium in the presence or absence of
both TPA and cycloheximide. Samples were analyzed by Western blot as
described above.Membrane Fractionation
Ten plates (10 cm) of
AF6295 cells were grown to approximately 90% confluence. Cells were
washed once with PBS and then placed in media supplemented with 100
nM TPA but lacking FBS and antibiotics. Cells were grown for
an additional hour at 37 °C, washed once with cold PBS, and then
scraped into PBS and transferred to a 50-ml conical tube. Plates were
washed a final time with cold PBS, and this wash was combined with the
resuspended cells. Cells were pelleted by gentle centrifugation,
resuspended in 2 ml of ice-cold TSA (2 mM Tris, pH 8.0, 0.14 M NaCl), and stirred gently on ice for 1 h. Cells were
homogenized with a Tissue Tearor two to three times on ice and brought
to a final volume of 2.5 ml of TSA containing 0.25 M sucrose,
1 mM EDTA, 0.1 mg/ml PMSF, 25 µg/ml leupeptin, 100
µM NaVO and then centrifuged 10
min at 125,000 g at 4 °C to remove nuclei. The
supernatant was saved, and the pellet was resuspended in 2.5 ml of the
TSA/sucrose solution and centrifuged again. The supernatants were
combined and centrifuged at 125,000 g for 2 h at 4
°C to pellet the membrane fraction. The supernatant was removed and
saved. The pellet was resuspended in 1 ml of TSA containing 0.5% Triton
X-100 and 0.5% sodium deoxycholate and mixed for 30 min at 4 °C to
solubilize membranes. Equivalent amounts of total lysate, cytosol and
membrane fractions were fractionated by SDS-PAGE on 7.5% polyacrylamide
gels and analyzed by Western blot as described above.
Processing of the Axl Protein
In order to
examine the processing of the intact Axl receptor, we analyzed cell
lysates of a secondary nude mouse tumor cell line expressing the
genomic axl (AF6295, 20). Cell lysates were immunoprecipitated
with a polyclonal antibody directed against a bacterially expressed
first immunoglobulin loop of Axl (see ``Experimental
Procedures'') and then analyzed by Western blot. A 140-kDa protein
corresponding to the full-length Axl receptor as well as several
smaller immunoreactive species of 120 and 104 kDa were detected (Fig. 1). In cells pulse labeled with
[S]methionine, the 104-kDa Axl immunoreactive
protein disappears first followed by loss of the 120-kDa protein and
finally by the disappearance of the 140-kDa band (Fig. 2). N-Glycanase treatment of Axl protein immunoprecitated with
anti-IgL reduces the apparent molecular mass of these species to 104
kDa (data not shown). These results suggest that the smaller 120- and
104-kDa proteins, respectively, are partially glycosylated and
unglycosylated precursors for the full-length protein which is
consistent with the predicted molecular mass of the unprocessed Axl
protein(20) . In addition, the disappearance of the
S-labeled Axl proteins seen in Fig. 2suggests that
the half-life of the receptor in AF6295 cells is on the order of 2 h.
SDS-sample buffer and boiled to
disrupt complexes. Half of each sample was fractionated by SDS-PAGE on
a 10% polyacrylamide gel and transferred to Immobilon-P filters.
Filters were hybridized with affinity-purified anti-Axl IgL (1
µg/ml). Immunoreactive bands were visualized by ECL kit reagents.
Axl-specific proteins are designated on the right of the
figure. The position of molecular mass markers is shown to the right. CM, conditioned media; LYS, cell
lysate. The dark band at the bottom of the figure corresponds
to heavy chain IgG.
A Soluble ECD of the Axl Receptor Results from
Proteolytic Cleavage of the Holoenzyme
As mentioned previously,
a number of isoforms of RTKs exist, including that of a soluble ECD
without the catalytic kinase domain. To test for the presence of a
soluble form of Axl, we analyzed CM from AF6295 cells. Using anti-IgL
antibodies, we detect an immunoreactive band of 80 kDa in CM (Fig. 1). Inclusion of the Axl immunoglobulin peptide antigen
along with the primary antibody blocks immunoreactivity (data not
shown). Furthermore, this 80-kDa soluble protein does not react with an
antibody directed against the carboxyl-tail of the receptor. These
results demonstrate that the soluble Axl-related protein, which we will
hereafter refer to as Axl, corresponds to the ECD of the
full-length receptor.
is
translated from a differentially spliced transcript encoding a
truncated Axl protein. However, we have not detected expression of axl transcripts encoding a soluble Axl ECD in AF6295 cells
making this mechanism unlikely. More importantly, using the anti-IgL
antibody, we also detect Axl
in the CM of TF14B12, an
NIH/3T3 derived cell line which overexpresses a full-length axl cDNA (Fig. 4B)(20) . However, CM from
untransfected NIH/3T3 cells shows no detectable Axl indicating that
Axl
arises through proteolytic processing of the
full-length receptor and not by alternative splicing of the RNA ( Fig. 1and Fig. 4B).
SDS sample buffer, and one-tenth of
each sample analyzed by Western blot using the anti-IgL antibody as in Fig. 1.
is not
restricted to expression in mouse fibroblasts. A similar Axl
protein is seen in the CM from T24, a human bladder carcinoma
cell line possessing an activated H-ras oncogene (Fig. 4B). A slightly smaller soluble protein is
detected in the CM from A549, a human lung carcinoma cell line. This
difference in size in the soluble ECD is due to differential
glycosylation of the protein between mouse and human cells as treatment
of CM from A549 and AF6295 cells with N-glycanase reduces the
apparent molecular mass of the two secreted proteins to identical sizes
(58 kDa) consistent with the predicted molecular mass of the
unprocessed Axl ECD (Fig. 3). Both T24 and A549 express
significant amounts of the Axl holoenzyme as assessed by Western
blotting (Fig. 4A).
. Samples of 24 h CM from AF6295 cells and A549 cells
were lyophilized and then digested overnight with N-glycanase
as described under ``Experimental Procedures.'' Half of each
sample was fractionated by SDS-PAGE on a 10% polyacrylamide gel,
transferred to Immobilon-P, and the Western blot incubated with
anti-IgL (1 µg/ml). -, samples without the glycanase; +,
samples with the glycanase. The secreted and glycosylated Axl proteins
are designated Axl
, whereas the deglycosylated Axl
proteins are designated Axl.
and Axl
to refer to the presence
or absence, respectively, of these 9 amino acids. Although these two
forms of the receptor are variably expressed in a number of cell lines,
both are equivalent in their transforming potential(20) . Using
an affinity purified, polyclonal antibody directed against the 9 amino
acid stretch (hereafter referred to as the Axl-box), we analyzed the CM
from various Axl expressing cell lines. As shown in Fig. 5A,
Axl
in AF6295 is recognized by the anti-IgL and
anti-Axl-box antibodies. Similar results were obtained with Axl
from TF14B12, an NIH/3T3 cell line expressing the axl
cDNA(20) . In TF17B, an NIH/3T3
cell line expressing the axl
cDNA, we detect
a soluble Axl
protein in the CM immunoreactive only with
the anti-IgL antibody (data not shown). These results map the
proteolytic cleavage site to a 14-amino acid region between the Axl-box
and the transmembrane domain harboring the amino acid sequence
VKEPSTPAFSWPWW (between amino acids 438 and 451, Fig. 6).
Comparison of this region of Axl with known proteins does not reveal
homology to any known protease cleavage sites.
Production of Axl
Proteolytic cleavage of both the CSF-1R and MET
receptors is augmented by protein kinase C(16, 19) .
To test for the involvement of protein kinase C in the production of
Axl Is Regulated by Protein
Kinase C
, we treated both Axl genomic and cDNA transfected
NIH/3T3 cell lines with TPA and analyzed the CM. Treatment of AF6295
cells with 100 nM TPA dramatically increases the level of
Axl
released into the culture medium as compared with
untreated cells (Fig. 5A). This induction is seen as
early as 15 min of treatment (data not shown) and is still evident
after 12 h. By 24 h, however, treated and untreated cells produce
equivalent amounts of Axl
, suggesting that the effects of
TPA are transient. Similar results are seen with an axl cDNA-transfected cell line, TF14B12 (data not shown). Furthermore,
production of both forms of Axl
(the Axl
and the Axl
forms) is induced by addition of
TPA suggesting that there is no difference in the TPA-induced
processing of either Axl forms (Fig. 5A). Analysis of
the cell lysates of TPA-treated cells indicates there is a concomitant
decrease in the amount of Axl
as Axl
is
released into the media (data not shown). Extending our methionine
labeling experiments in AF6295 cells, we found that treatment of
[
S]methionine-labeled cells with TPA results in
a rapid decrease in the amount of Axl
(Fig. 5B). Concomitant with the loss of
Axl
in cell lysates is the appearance of radiolabeled
Axl
in the CM of AF6295 cells (data not shown). These
data support the premise that Axl
is generated by the
proteolytic processing of the full-length Axl receptor and that TPA
regulates this protease activity.
Production of Axl
The finding that TPA dramatically induces the amount
of Axl Is Independent of Protein
Synthesis
suggests that production of Axl
involves a protein kinase C-regulated protease. To rule out the
possibility that TPA is inducing transcription of Axl message in 3T3
cells, an effect seen in other cell lines of myeloid
origin(26, 27) , we tested the effect of cycloheximide
on production of Axl
in Axl-transfected 3T3 cells.
Cycloheximide treatment does not affect the basal production of
Axl
(Fig. 7). In addition, there does not appear
to be any effect of cycloheximide on TPA-stimulated proteolysis of the
intact receptor (data not shown). These data suggest that the effects
of TPA on the production of Axl
are independent of
protein synthesis and further support the notion that Axl
is derived from proteolytic cleavage and not translation of an
alternatively spliced transcript. In addition, the level of
Axl
decreases over time with cycloheximide treatment in
comparison with Axl
(data not shown). This finding is in
agreement with the previous pulse-chase experiments that point to
Axl
as a precursor for Axl
.
is independent of protein synthesis. 10-cm plates containing
equivalent numbers of AF6295 cells were grown in the presence (+)
or absence(-) of cycloheximide (10 µg/ml) for 30 min. Cells
were then washed with PBS and placed in serum-free medium ± TPA
and ± cycloheximide for 1 or 2 h. Equivalent amounts of CM (70
µl) from each sample were analyzed by SDS-PAGE in a 7.5%
polyacrylamide gel, transferred to Immobilon-P, and the Western blot
incubated with anti-IgL as described under ``Experimental
Procedures.'' The position of the secreted Axl extracellular
domain (Axl
) is indicated by an arrow.
Cleavage of Full-length Axl Results in the Production of
a Membrane-bound Phosphorylated Carboxyl-terminal Peptide
Since
Axl derives from proteolytic cleavage of the full-length
receptor at or near the transmembrane domain, a predicted 55-kDa
carboxyl-terminal cleavage product containing the kinase domain should
also be produced. Using an affinity-purified polyclonal antibody
directed against the carboxyl terminus of Axl (see ``Experimental
Procedures''), we are able to detect a 55-kDa
tyrosine-phosphorylated protein, Axl
, suggesting that
Axl
corresponds to the kinase domain of Axl (Fig. 8, A and B). Separation of the membrane
and cytosol fractions of AF6295 cells treated for 1 h with TPA
demonstrated that this carboxyl-terminal fragment of Axl partitioned
with the membrane fraction of cells further supporting the model that
cleavage of the holoenzyme occurs between the Axl-box region and the
transmembrane spanning region (Fig. 9). Furthermore, secretion
of the soluble Axl ECD occurs when these experiments were carried out
under conditions which block protein trafficking (Fig. 10).
These data suggest that cleavage of the receptor occurs on the surface
of the plasma membrane as opposed to within intracellular vesicles.
Whereas the half-life of the holoenzyme is in the range of 2 h, the
half-life of the truncated kinase domain is considerably less in that
no Axl
is detectable after 2 h of TPA exposure.
-PTYR) (A) or
the carboxyl-tail of Axl (-COOH) (B). The fully
and partially processed Axl proteins are designated Axl and Axl
, respectively. The COOH-terminal Axl
protein is marked Axl
. The positions of molecular mass
markers are designated with dashes.
, partitions with the membrane fraction of cells.
AF6295 cells were grown for 1 h in the presence of TPA (100
nM), lysed, and fractionated into membrane (M) and
cytosol (C) fractions. Equivalent amounts of protein from each
sample, including total lysate (T), were fractionated by
SDS-PAGE on a 7.5% polyacrylamide gel and analyzed by Western blot
using antibodies directed against the carboxyl-tail of Axl (a-COOH) and anti-phosphotyrosine (a-PTYR). The
carboxyl-terminal cleavage product corresponding to the kinase domain
of Axl is designated Axl
. The positions of molecular mass
markers are designated with dashes.
,
is released into the CM of Axl expressing cells. Although Axl
may arise through two mechanisms, alternative splicing of the
message or proteolysis of the full-length receptor, our data support
the latter scenario. Axl
is present in cells transfected
with the genomic Axl as well as full-length axl cDNA clones
but not in untransfected NIH/3T3 cells. Thus, Axl
is
derived from the full-length Axl protein and not an alternative
transcript. Although alternatively spliced transcripts encode truncated
forms of the trkB and EGF receptor
kinases(3, 12, 13, 35, 36) ,
we have not, as yet, found evidence for axl transcripts
encoding a truncated protein.
. Moreover, TPA treatment
appeared to decrease the levels of the full-length receptor due to
cleavage of the holoenzyme and the subsequent release of soluble
Axl
.
-dependent soft agar colony growth of chicken embryo
fibroblasts through binding of the growth factor. In addition, Basu et al.(15) have shown that a secreted human EGFR
protein is able to inhibit the activity of the full-length receptor
presumably through intermolecular interactions. In contrast to the
secreted chicken EGFR, the human form appears to inhibit the activity
of the intact membrane bound receptor through a mechanism other than
simple competition for ligand since excess EGF is not sufficient to
suppress the inhibitory effects of the truncated receptor(15) .
Although the possibility exists that Axl may bind ligand,
preliminary data suggest that Axl
does not inhibit the
transforming ability of the intact Axl protein. AF6295 cells produce
large amounts of Axl
yet are still highly transformed
with high levels of phosphorylated Axl
and
Axl
. Furthermore, addition of a baculovirus-expressed
Axl ECD to the media of AF6295 cells does not appear to affect the
growth of cells or phosphorylation of intact Axl receptor. (
)These results suggest that Axl is not a
dominant-negative inhibitor of the full-length receptor.
. The
appearance of this fragment occurs with the same kinetics as the
release of Axl
into the CM of cells, indicating that
Axl
is derived from the cleavage of the full-length
protein. Although a carboxyl-terminal MET fragment was not detected
upon TPA-stimulated proteolysis of the full-length MET
receptor(19) , cleavage of the CSF-1R also generates a similar
carboxyl-terminal kinase fragment(16) . The exact role of
proteolytic cleavage in RTK function is unclear.
is at least a partially activated kinase. However, RTKs activated
by engineered ECD truncations are transforming in large part due to the
retention of the deregulated kinase on the cell surface. By contrast,
the considerably shortened half-life (<1 h) of the membrane bound,
truncated Axl kinase after TPA exposure suggests that proteolytic
cleavage results in receptor down-regulation. Further work is in
progress to confirm this possibility.
))
We wish to thank Edward W. Baptist for his technical
assistance.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Sather, K. D. Kenyon, J. B. Lefkowitz, X. Liang, B. C. Varnum, P. M. Henson, and D. K. Graham A soluble form of the Mer receptor tyrosine kinase inhibits macrophage clearance of apoptotic cells and platelet aggregation Blood, February 1, 2007; 109(3): 1026 - 1033. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. D'Arcangelo, V. Ambrosino, M. Giannuzzo, C. Gaetano, and M. C. Capogrossi Axl receptor activation mediates laminar shear stress anti-apoptotic effects in human endothelial cells Cardiovasc Res, September 1, 2006; 71(4): 754 - 763. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Vajkoczy, P. Knyazev, A. Kunkel, H.-H. Capelle, S. Behrndt, H. von Tengg-Kobligk, F. Kiessling, U. Eichelsbacher, M. Essig, T.-A. Read, et al. Dominant-negative inhibition of the Axl receptor tyrosine kinase suppresses brain tumor cell growth and invasion and prolongs survival PNAS, April 11, 2006; 103(15): 5799 - 5804. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Budagian, E. Bulanova, Z. Orinska, E. Duitman, K. Brandt, A. Ludwig, D. Hartmann, G. Lemke, P. Saftig, and S. Bulfone-Paus Soluble Axl Is Generated by ADAM10-Dependent Cleavage and Associates with Gas6 in Mouse Serum Mol. Cell. Biol., November 1, 2005; 25(21): 9324 - 9339. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Singh, R. D. Meyer, H. Band, and N. Rahimi The Carboxyl Terminus of VEGFR-2 Is Required for PKC-mediated Down-Regulation Mol. Biol. Cell, April 1, 2005; 16(4): 2106 - 2118. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Konishi, T. Aizawa, A. Mohan, V. A. Korshunov, and B. C. Berk Hydrogen Peroxide Activates the Gas6-Axl Pathway in Vascular Smooth Muscle Cells J. Biol. Chem., July 2, 2004; 279(27): 28766 - 28770. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. R. van Ginkel, R. L. Gee, R. L. Shearer, L. Subramanian, T. M. Walker, D. M. Albert, L. F. Meisner, B. C. Varnum, and A. S. Polans Expression of the Receptor Tyrosine Kinase Axl Promotes Ocular Melanoma Cell Survival Cancer Res., January 1, 2004; 64(1): 128 - 134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Collett, A. Wood, M. Y. Alexander, B. C. Varnum, R. P. Boot-Handford, V. Ohanian, J. Ohanian, Y.-W. Fridell, and A. E. Canfield Receptor Tyrosine Kinase Axl Modulates the Osteogenic Differentiation of Pericytes Circ. Res., May 30, 2003; 92(10): 1123 - 1129. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Trivier and T. S. Ganesan RYK, a Catalytically Inactive Receptor Tyrosine Kinase, Associates with EphB2 and EphB3 but Does Not Interact with AF-6 J. Biol. Chem., June 14, 2002; 277(25): 23037 - 23043. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Healy, J. J. Schwartz, X. Zhu, B. E. Herrick, B. Varnum, and H. W. Farber Gas 6 promotes Axl-mediated survival in pulmonary endothelial cells Am J Physiol Lung Cell Mol Physiol, June 1, 2001; 280(6): L1273 - L1281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. YANAGITA, K. ISHII, H. OZAKI, H. ARAI, T. NAKANO, K. OHASHI, K. MIZUNO, T. KITA, and T. DOI Mechanism of Inhibitory Effect of Warfarin on Mesangial Cell Proliferation J. Am. Soc. Nephrol., December 1, 1999; 10(12): 2503 - 2509. [Abstract] [Full Text]
|
|
|
|
 |
|
 K. O'Donnell, I. C. Harkes, L. Dougherty, and I. P. Wicks Expression of Receptor Tyrosine Kinase Axl and its Ligand Gas6 in Rheumatoid Arthritis : Evidence for a Novel Endothelial Cell Survival Pathway Am. J. Pathol., April 1, 1999; 154(4): 1171 - 1180. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. G. Melaragno, D. A. Wuthrich, V. Poppa, D. Gill, V. Lindner, B. C. Berk, and M. A. |
