(Received for publication, January 10, 1997, and in revised form, May 23, 1997)
From the Connective Tissue Research Institute,
Department of Medicine, University of Pennsylvania, and the University
City Science Center, Philadelphia, Pennsylvania 19104, and the
§ Laboratory of Biochemistry, CNRS EP89, University of
Reims, Champagne-Ardenne, F-51095 Reims, France
Our previous studies have shown that a peptide
corresponding to the residue sequence 185-203 of the NC1 domain of the
3 chain of basement membrane collagen (type IV) inhibits the
activation of polymorphonuclear leukocytes. Peptides from the same
region of the
1,
2,
4, and
5(IV) chains did not exhibit
this property. Because of the intimate relationship between
metastasizing neoplastic cells and vascular as well as epithelial
basement membranes, we measured the cell adhesion-promoting activity of
peptides from the NC1 domain of type IV collagen and their effect on
proliferation of human melanoma cells. We found that peptide
3(IV)185-203 (CNYYSNSYSFWLASLNPER) not only promotes adhesion of
human melanoma cells but also inhibits their proliferation. Adhesion
increased by 50-60% over control. Melanoma cell proliferation was
inhibited by 40% when cells were grown in a medium containing 5 µg/ml peptide for 5 days. Studies showed that replacement of serine
in position 189 or 191 by alanine resulted in significantly reduced
adhesion. Similarly, serine replacement resulted in reduced ability to
inhibit proliferation. Our data suggest that a region of the NC1 domain
of the
3(IV) chain, contained within the sequence 185-203, not only
specifically promotes adhesion but also inhibits proliferation of
melanoma cells. These properties appear to be dependent on the presence of the triplet sequence -SNS- (residues 189-191), which is unique to
the
3 chain and may represent an important functional
epitope.
Type IV collagen is a major component of basement membranes. The
predominant molecular species is a heterotrimer composed of two 1
and one
2 chain. The presence of additional type IV collagen chains,
3(IV),
4(IV),
5(IV), and
6(IV), has been reported (1-6).
There is evidence that the latter are distributed in most basement
membranes (4, 5). Type IV collagen not only forms the main structural
framework of all basement membranes, but also serves as scaffolding for
the binding of other basement membrane components (7, 8). One important
function of type IV collagen is its ability to promote the adhesion and
motility of various normal and transformed cell types (9).
The chain of type IV collagen has a long collagenous domain of
about 1,400 amino acid residues and a non-collagenous domain of about
230 residues at the carboxyl terminus, called the NC1 domain. The NC1
domain is thought to play a key role in the alignment and selection of
three
chains forming a triple helical molecule (10, 11). Several
studies have focused on the biological activity of the NC1 domain of
type IV collagen. One synthetic heparin-binding peptide, Hep-I,
originating from NC1 domain of the
1 chain of type IV collagen and
containing 12 amino acids, has been reported to promote the adhesion
and spreading of bovine aortic endothelial cells (12, 13). A series of
synthetic peptides from the NC1 domain of several
chains has been
used to map antigenic epitopes on type IV collagen (14). One of these
peptides, comprising residues 185-203 of the NC1 domain of the
3(IV) chain, has been shown to inhibit the activation of
PMN1 as measured by a
reduction in O2
production
and proteolytic enzyme release (15).
Neoplastic cells have the ability to invade and metastasize. Because of
the intimate relationship between epithelial cells and basement
membranes as well as metastatic cells and vascular basement membranes
we decided to examine the ability of the NC1 domain of type IV collagen
and synthetic peptides from this domain to influence adhesion and
proliferation of melanoma cells. Our studies show that a synthetic
peptide comprising residues 185-203 of the NC1 domain of the 3(IV)
chain promotes adhesion and inhibits proliferation of melanoma cells.
Using monoclonal antibodies that recognize the above peptide, we have
shown that there is a multifunctional domain within the first 12 amino
acids of the residue sequence 185-203 capable of promoting the cell
adhesion and inhibition of proliferation. Because synthetic peptides of
the same region from the other
(IV) chains that lack the triplet
-SNS- (residues 189-191) fail to inhibit melanoma cell adhesion and
proliferation and because the sequence -SNS- is unique to the
3(IV)
chain we can assume that this triplet represents the functional epitope of this peptide.
The human melanoma cell lines used in these studies, WM9, WM164, WM136 1A were kindly provided by Dr. Meenhard Herlyn (Wistar Institute, Philadelphia). Metastatic cell lines WM9 and WM164 were derived from an intermediate stage metastatic lesion. WM136 1A was obtained from a late lesion with concomitant distant metastases at the time of tumor removal (16, 17). The above cell cultures were grown in tumor medium (MCDE 153/L-15, 4:1, Sigma) supplemented with 2% fetal bovine serum and insulin (5 µg/ml, Sigma). In addition, other tumor cell lines including two melanoma cell lines (HT-144 and UACC-903), a fibrosarcoma cell line (HT-1080), and an osteosarcoma cell line (MG-63) were used. Normal fibroblasts were used as controls. The latter cultures were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. All cultures were maintained at 37 °C in a humidified atmosphere containing 95% air and 5% CO2.
Preparation of Synthetic PeptidesPeptides corresponding to
several specific sequences of the human NC1 domains of the 1(IV),
2(IV), and
3(IV) chains, comprising the residues 185-203, and
4(IV) (bovine) and
5(IV) chains, comprising the residues
182-200, were synthesized according to the method of Barany and
Merrifield (18) at the Protein Chemistry Facility of the University of
Pennsylvania. Additional peptides, representing partial sequences of
the
3(IV) peptide, i.e. residues 185-196, and 194-203,
as well as the
3(IV) peptides, where the serines in positions 189 or
191 were replaced by alanine, were prepared (see Table I).
|
Monoclonal antibodies
to the synthetic peptide of the human NC1 domain of the 3(IV) chain
were prepared by the Cell Center of the University of Pennsylvania,
using the purified synthetic human
3(IV)179-208 peptide as
immunogen. The purified
3(IV)179-208 peptide was conjugated to
hemocyanin and injected into mice (19). Monoclonal antibodies were
produced by fusing the mouse lymphocytes and myeloma cells (SP2-0AG4
from ATCC and developed at University of Pennsylvania). Supernatants
from the subclones were screened, and specific subclones were selected
using the ELISA method described by Engvall and Perlmann (20).
Production of a rabbit polyclonal antibody to a synthetic peptide
corresponding to residues 179-208 of the NC1 domain of the 3(IV)
chain was described elsewhere (15).
The
assay was performed essentially according to previously described
methods of Engvall and Perlmann (20), using 96-well plastic plates
(Dynatech Labs Inc., Chantilly, VA). All synthetic peptides were
dissolved in 6 M GuHCl, pH 7.5, 4 °C, 0.5 mg/ml. The
peptide solutions were diluted further in 50 mM sodium
carbonate/bicarbonate buffer, pH 9.6. Various synthetic peptides were
coated at concentrations of 250 ng/well and incubated overnight at
4 °C. Milk (5%) in PBS/Tween 20 was used as the blocking solution
for 1 h. Supernatants from the selected subclones were added to
wells and incubated for 1 h. Secondary antibody (anti-mouse
IgG(H+L)-peroxidase, Boehringer Mannheim) was added to each well and
incubated for 1 h. Finally, ABTS solution (Boehringer
Mannheim) was added to each well. The plates were read at 405 nm
in a Microplate Reader, EL340 (Bio-Tek Instruments). Monoclonal
antibodies against specific residue sequences of 3(IV)179-208 were
selected.
Cell attachment was determined according to the method of Rao and Kefalides (21), with modification. The 24-well plastic plates were coated with the appropriate peptides at various amounts in 1.0 ml of coating buffer (0.1 M Tris-HCl buffer, pH 7.5). The peptides, dissolved in conditioned medium, were added to each well with coating buffer. The coating was terminated after incubating the plates for about 24 h at 4 °C, followed by washing twice with 1 ml of PBS. Wells were blocked by addition of 1.0 ml of attachment medium (Hanks' balanced saline solution containing 2% bovine serum albumin and 10 mM HEPES, pH 7.4) at room temperature for 1 h. Human melanoma cells near confluence were labeled with [35S]methionine (5 µCi/ml) in growth medium overnight. Cells, detached with 0.25% trypsin, EDTA solution and resuspended in attachment medium, were added to each well (2.5 × 105 cells/well) and incubated at 37 °C for 75 min in 5% CO2. The unattached cells were removed with three 1.0-ml washes in PBS. Attached cells were then lysed for 30 min by adding 0.5 ml of 1% Triton X-100 in PBS/well. The extracts were transferred to scintillation vials for counting, and the percent attachment was defined as (radioactivity extracted from attached cells)/(radioactivity in cells added to assay) × 100%. Results are expressed as percent of cells attached/well. Each attachment assay was run in triplicate.
Competition of Cell Binding AssayFor competition of the cell binding assay, 24-well plastic plates were coated with synthetic peptides as described above. The attachment assay was then performed as follows. Both monoclonal and polyclonal antibodies to peptide were tested for their ability to inhibit cell adhesion to the peptide. The antibodies were added to the wells at various dilutions and incubated for 2 h at room temperature before adding the [35S]methionine (5 µCi/ml)-labeled cell suspension (2.5 × 105 cells/well). Since the monoclonal antibody D12H5 did not react with any of the peptides tested, it was used as a negative control. The cells were incubated for 75 min at 37 °C. At end of the incubation, the plates were washed with PBS, the attached cells were solubilized with Triton X-100, and bound radioactivity was measured in a scintillation counter. The percent attachment was calculated as described above.
Cell Proliferation AssayThe determination of cell proliferation was done by counting cells in a coulter counter or by determining the [3H]thymidine incorporation as a measure of DNA synthesis, or by the MTT method. By using the MTT method, the activity of living cells via mitochondrial dehydrogenase activity was measured (MTT cell growth determination kit, Sigma). For cell counting and [3H]thymidine incorporation measurements, confluent cultures were dissociated and suspended in fresh tumor medium. 1 × 105 cells were added to wells of a 12-well plate and incubated for 3-4 h. Then, fresh medium containing one of the peptides was added to the wells. The medium was changed every 2 days. After a 5-day incubation with peptides, cells were either dissociated with 0.25% trypsin, EDTA and cell numbers counted in a coulter counter, or were labeled with [3H]thymidine (3 µCi/ml) overnight. Labeled cells were lysed with 1% Triton X-100 in PBS and radioactivity measured in a scintillation counter.
To test whether it is the peptide binding to the cells in the medium
which induces the inhibitory effect on cell proliferation or whether it
is the peptide adsorbed to the well surface causing strong cell
adhesion which inhibits cell proliferation, cell-peptide interaction
experiments were carried out. In one experiment, peptides were coated
at varying amounts onto the wells of a 96-well plate, and 5 × 104 cells were added to each well. After a 5-day
incubation, the MTT solution in an amount equal to 10% of the culture
volume was added to each well, according to the kit procedure (Sigma).
After a 4-h incubation at 37 °C, the culture medium was removed, and MTT solvent in an amount equal to the original culture volume was added
to each well. The plates were read at 570 nm in a Microplate reader
EL340 (Bio-Tek Instruments), and cell proliferation was determined. The
cells in uncoated wells incubated with the medium containing the
peptide were used as control. In another experiment, we tested whether
the peptide from the medium adsorbed onto the well surface. Cells were
removed by 0.1% sodium dodecyl sulfate after a 5-day incubation with
the medium containing varying amounts of the 3(IV)185-203 peptide.
The presence of peptide attached to the well was detected by adding to
the wells a polyclonal antibody to the
3(IV) peptide at 1:250
dilution and incubating for 1 h at room temperature. The
immunoreactivity of the
3(IV) peptide was detected by ELISA as
described above.
To characterize
the monoclonal antibodies that recognize a functionally important
domain within the peptides of the NC1 domain of the 3(IV) chain of
type IV collagen, we determined the ability of monoclonal antibodies to
bind selectively to the synthetic peptides from the partial sequences
of the
3(IV) chain and from the comparable region of the different
chains. As shown in Table I, the
polyclonal antibody and monoclonal antibodies A5B7 and A5D7 reacted
strongly only with the synthetic peptides from the
3(IV) chain,
i.e.
3(IV)179-208 and
3(IV)185-203, but did not react with the peptides from comparable regions of
1(IV),
2(IV), and
5(IV) chains, even though they share considerable homology. To
determine the critical sequence recognized by the monoclonal antibodies, the partial sequences of the
3(IV) peptide,
i.e. peptides comprising residues 185-196 and 194-203,
were tested. Both the polyclonal antibody and the monoclonal antibodies
A5B7 and A5D7 reacted strongly with the peptide
3(IV)185-196, but they reacted very weakly with the peptide
3(IV)194-203. The data indicate that the epitope recognized by the antibodies is contained within the region of residues 185-196 of
3(IV) chain. It should be
noted that the major difference between the
3(IV)185-196 peptide and comparable peptides from other
chains in this region is the
sequence -SNS- (residues 189-191) within the
3(IV) peptide which is
not present in the others (Table I). It appears that this unique -SNS-
sequence within the
3(IV) peptide is very critical. Peptides where
one or both serines are substituted (
1,
2, or
5) failed to
react with the antibodies (Table I).
It is interesting to note that the monoclonal antibody D12H5 reacted very weakly with all of the peptides, and it was therefore used as the control antibody in the competition of cell binding assay.
Cell Adhesion on Peptide SubstrataSynthetic peptides from
the NC1 domain of 3(IV) chain of the human type IV collagen were
tested for their attachment-promoting activity on human melanoma cells.
Fig. 1 shows that only
3(IV) peptide
promoted from 55-65% attachment of melanoma cells, whereas peptides
from comparable regions of the
1(IV),
2(IV), and
5(IV) chains
did not. To determine whether a specific sequence was the active cell
binding domain, two peptides of the
3(IV) chain, comprising residues
185-196 (CNYYSNSYSFWL) and 194-203 (FWLASLNPER), were tested. Peptide
3(IV)185-196 promoted adhesion of melanoma cells from 45 to 50%.
In contrast, peptide
3(IV)194-203 did not mediate cell adhesion
beyond that of control. The property of cell adhesion is, therefore,
dependent on the sequence of residues 185-196 within the
3(IV)185-203 peptide.
Effect of Peptides on Cell Proliferation
Table
II shows the effects of synthetic
peptides on tumor cell proliferation. Proliferation of melanoma cells
was inhibited from 20 to 42% when cells were grown in medium
containing peptide 3(IV) 185-203 (5 µg/ml). This inhibition was
reproducible in different melanoma cell lines, including WM9, WM164,
WM136 1A, HT-144, and UACC-903, and in two non-melanoma tumor cell
lines (HT-1080 and MG-63). Replication of normal fibroblasts was
unaffected by the
3(IV) peptides. On the other hand, synthetic
peptides from comparable regions of the
1(IV),
2(IV), and
4(IV) chains did not show significant inhibition of replication
(Table II). The effect of peptides representing partial sequences of
3(IV)185-203 was also assessed in these experiments. Fig.
2a shows that the inhibitory
effect was significant only in the cells treated with peptides
3(IV)185-203 and
3(IV)185-196 (about 44% inhibition). The
proliferation of melanoma cells treated with peptide
3(IV)194-203, which lacks the -SNS- triplet, was not significantly affected.
|
There is the possibility that significant amounts of peptide moved from
the medium and became adsorbed to the well surface, causing firm cell
adhesion and thereby decreasing cell proliferation. To exclude this
possibility, melanoma cells were plated onto wells, coated with the
3(IV)185-203 peptide at various concentrations, for 5 days.
Although the
3(IV) peptide induced melanoma cell adhesion, there was
only a minimal decrease in cell proliferation (10-15%) even at
concentrations 5 × (25 µg) that which gives maximum inhibition
of cell proliferation (Fig. 2b). However, when the
3(IV)
peptide was placed in the medium, the decrease in proliferation was
significantly greater (40%) (Fig. 2a). Additional
experiments also showed that after a 5-day peptide treatment, the
majority of the peptide did not make its way into the substrate and was bound to the cell surface (Fig. 2c). The slight increase in
peptide content on the well surface can be accounted for by the type IV collagen synthesized by the cells during incubation. These data suggest
that the cell-peptide contact in the medium is responsible for the
observed inhibition of cell proliferation. To assess whether cell
proliferation resumed after removal of the peptide in the medium, at
the end of the 5-day period the peptide-containing medium was removed,
and cells were incubated in peptide-free medium for an additional 6 days. At the end of this period, cell proliferation was back to control
levels (Fig. 2d).
To test the specificity of the cell binding domain of
the peptide 3(IV)185-203, both monoclonal and polyclonal antibodies against this sequence were used to block peptide-mediated cell adhesion
and inhibition of cell proliferation. WM9 melanoma cell adhesion on
surfaces coated with peptide
3(IV)185-203 (2.5 µg/well) was
monitored in the presence of anti-peptide antibodies. Melanoma cell
adhesion was inhibited to an extent of 53-60% by the polyclonal antibody and by the monoclonal antibody A5D7. No significant inhibition of cell adhesion was observed in the presence of a control nonreactive monoclonal antibody D12H5 (Fig.
3a). The effect of the
3(IV) peptide on cell proliferation, on the other hand, was also
prevented by the antibodies. When the
3(IV) peptide was treated with
antibody before adding to the medium, its inhibitory effect on cell
proliferation was decreased significantly (Fig. 3b). It is
suggested that a functional domain in the
3(IV)185-203 peptide,
recognized by the monoclonal antibody A5D7, is responsible for the
activities of promotion of cell adhesion and inhibition of cell
proliferation.
Identification of the Functional Domain in
The current studies indicate that the sequence of
residues 185-196 of the NC1 domain of the 3(IV) chain contains the
functional domain that is responsible for the cell adhesion and the
inhibitory effect on cell proliferation. The unique triplet -SNS-
within the above sequence appears to represent such a functional
domain. To confirm this, we replaced serine with alanine in either
position 189 or 191 of the
3(IV) peptide (Table I, modified
synthetic peptides) and tested the cell adhesion-promoting activity of
the modified
3(IV) peptides on two melanoma cell lines. Fig.
4a shows the results of these
experiments. Serine substitution resulted in significantly reduced
adhesion. Adhesion fell from 72% with the peptide containing the -SNS-
sequence to 44 and 28% with peptides containing the -ANS- and -SNA-
sequence, respectively, for W9 cells. With W164 cells, the values fell
from 60 to 24 and 20%, respectively. The same modified
3(IV)185-203 peptides were also used to test the effect on melanoma
cell proliferation. When melanoma cells were incubated with the
modified
3(IV) peptides, in both instances, i.e. with
-SNA- or -ANS- sequence, cell proliferation was not inhibited to any
appreciable extent compared with the -SNS- sequence (94 and 88.5% of
control, respectively) (Fig. 4b). It is suggested that the
active domain, which mediates melanoma cell adhesion and inhibits cell
proliferation, is the triplet sequence of -SNS-. Replacement of serine
with alanine in either position 189 or 191 abolishes both activities of
the peptide.
Our previous studies have shown that the peptide a3(IV)185-203
prevents activation of human PMN, and the unique sequence -SNS- is the
domain required for the inhibition of
O2 production and enzyme
release by these cells (15). Because melanoma cells are in direct
contact with basement membranes during tumor cell invasion and
metastasis, we elected to assess the in vitro influence of
type IV collagen peptides on the attachment and the proliferation of
human melanoma cells. In this study, several synthetic peptides from
the NC1 domain of the
chains of type IV collagen were tested.
Residues 185-203 of the
3(IV) chain comprise the only sequence that
significantly increases adhesion of melanoma cells and inhibits their
proliferation. The peptide-mediated melanoma cell adhesion can be
prevented by specific polyclonal and monoclonal antibodies. It is of
interest to note that the monoclonal antibody that recognized the
sequence 185-196 of the
3(IV) peptide also prevented the inhibition
of melanoma cell proliferation. It would appear that the cell
adhesion-promoting activity and the inhibitory effect of the peptide on
cell proliferation may be two functions that are intimately related. To
rule out the possibility that strong attachment of melanoma cells to
the peptide in the substrata was responsible for the effect of the
3(IV) peptide on melanoma cell proliferation, we carried out cell-peptide interaction experiments. The data clearly showed that it
was the binding of the peptide to the apical region of the cells rather
than to the basal region which was responsible for inhibition of cell
proliferation (Fig. 2, b and c). We also observed
that exposure of the cells to the peptide caused aggregation and
clumping (data not shown). It would appear that it is the interaction
of the cell surface with the soluble peptide which is necessary for
inhibition of cell proliferation to occur.
In the current studies, it appears that the sequence -SNS- of the
3(IV)185-203 peptide is an absolute requirement for both cell
adhesion and inhibition of cell proliferation. Although this region is
highly conserved in the NC1 domain of all
chains of type IV
collagen, the sequence -SNS- (residues 189-191) is unique to the
3(IV) chain. Both monoclonal and polyclonal antibodies specifically
recognized the region of the peptide
3(IV)185-196 which contains
the sequence -SNS-. The antibodies efficiently inhibited the cell
binding to the substrate, confirming the specificity of this
interaction. Substitution of either serine in the sequence -SNS-
abolished the cell adhesion activity of the peptide, suggesting that
the sequence -SNS- is a multifunctional cell binding domain that
mediates cell adhesion and inhibits cell proliferation.
Unpublished studies in our laboratory demonstrate that this peptide
inhibits proliferation not only of melanoma cells but of other
epithelial tumors cell and concomitantly increases intracellular cAMP.
It was strongly suggested that a mechanism involving a signal transduction pathway may play a role in the cell adhesion-promoting activity and in inhibition of cell proliferation by the 3(IV) peptide. The same peptide causes an inhibition of PMN activation through a rise in intracellular cAMP (15). An increase in cAMP is
associated with an inhibition of
O2
production as well as
secretion of proteinases and lactoferrin (15, 22). The increase of cAMP
in PMN (15) and in melanoma cells induced by the peptide can be
inhibited by pertussis toxin, suggesting involvement of G proteins
(23). A detailed study of the events in the signal transduction pathway
triggered by the
3(IV) peptide is the subject of a manuscript under
consideration. The direct effect of the
3(IV) peptide on melanoma
cell adhesion and proliferation raises the question of whether a
specific receptor for this peptide is involved. Studies are under way
to isolate and characterize the putative receptor.
Tumor cell proliferation and metastasis are complex processes involving
numerous tumor cell-extracellular matrix interactions that at present
are incompletely understood. The interaction of tumor cells with
basement membrane is the first step in a multifunctional process (24).
It is also unknown what specific role the NC1 domain of the 3(IV)
chain may play in vivo in the maintenance of the normal
phenotype of overlying epithelial cells or of transmigrating tumor
cells. Secondary structure studies have shown that the sequence, which
contains the triplet -SNS- of the
3(IV) chain, occurs within one of
the two
-turns occupied by the
3(IV)185-203 peptide sequence (14). It is hypothesized that the NC1 domain of the
3(IV) chain must
be exposed within the basement membrane in a way that this region of
the chain is on the outside and promotes contact with the
transmigrating PMN or tumor cells or with overlying epithelial cells in
a given tissue.
In this report we have demonstrated that a peptide from the NC1 domain
of the 3(IV) chain of type IV collagen, comprising residues
185-203, contains a multifunctional domain that promotes cell adhesion
and inhibits proliferation of melanoma cells. An absolute requirement
for these biological activities is the presence of a triplet -SNS-
within this peptide. Studies are currently under way to elucidate the
steps in the transduction pathway involved in these biological
activities and to isolate and characterize the putative receptor that
recognizes the above peptide.