| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Protein Chemistry, Genentech, Inc., South San Francisco, California 94080
Address all correspondence and requests for reprints to: Srividya Sundaresan, Ph.D., 1 DNA Way, MS 45, Genentech, Inc., South San Francisco, California 94080. E-mail: vidya{at}gene.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(1) and
amphiregulin (2). The second group consists of ligands that bind to
both EGFR and HER4, such as, betacellulin (BTC) (3, 4), heparin-binding
EGF-like growth factor (5), and epiregulin (6). Neuregulin-3 (NRG3)
forms the third group with specificity for ErbB4 alone (7). The last
group is made up of NRG1, which includes heregulins (HRGs)/neu
differentiation factors such as HRGß1 and sensory and motor
neuron-derived factor (SMDF), and NRG2 (8, 9). These ligands bind to
ErbB3 and ErbB4 (9, 10, 11, 12, 13, 14, 15, 16). ErbB2 is the preferred heterodimerization
partner for all of the other ErbB receptors, enhancing their affinities
for the ligands and amplifying the elicited signals, even though it
does not directly bind any of the above-mentioned ligands (17, 18, 19).
Binding of a specific ligand induces dimerization, activation of the
intracellular kinase domain, and receptor cross-phosphorylation (15, 20, 21). ErbB receptors and ligands are widely expressed in epithelial, mesenchymal, and neuronal tissues and play fundamental roles during development (22, 23, 24, 25, 26). Inactivation of the EGFR gene was found to cause impaired epithelial development in several organs, including lung and gastrointestinal tract (26). Similarly, deletion of HRG, ErbB2, ErbB3, and ErbB4 resulted in several developmental abnormalities, with severe defects in cardiac, neural, and gastrointestinal development (23, 27, 28). In addition to their roles in development and differentiation, aberrant expression of these receptors has been reported in a wide variety of human cancers and in many instances has been correlated with poor patient prognosis (29, 30, 31). To understand the role played by ErbB receptors in neoplastic transformation and cancer progression, it is critical to understand the functioning of these receptors in nontransformed cells that normally express them. With a few exceptions (32, 33), many of the systems that have been employed to address the question of ErbB receptor expression and regulation have consisted of transformed cell lines or transfected cells that overexpress one or more of these receptors. We hypothesized that under normal conditions, it might be possible to predict the responsiveness to ErbB ligands based on the pattern and relative levels of the various ErbB receptor subunits in any given tissue. In this study, we test this hypothesis by examining the relationship between expression of various ErbB receptors and responsiveness to stimulation by different ErbB ligands in a series of nontransformed rodent cells and cell lines derived from lung, heart, nervous system, and pancreas, which are known targets for ErbB function.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. BUD cells (Stephan,
J., P. E. Roberts, L. Bald, J. Lee, G. Qimin, A. Helmrich, D. W.
Barnes, J. P. Mather, manuscript submitted) were grown on dishes coated
with fibronectin (Sigma Chemical Co., Inc., St. Louis, MO)
in DMEM-Ham’s F-12 medium (a 1:1 mixture; obtained in powder form from
Life Technologies, Grand Island, NY), containing
recombinant human insulin (10 µg/ml; Genentech, Inc.,
South San Francisco, CA), aprotinin (10 µg/ml; Boehringer Mannheim,
Indianapolis, IN), recombinant human HRG-ß1 fragment corresponding to
amino acids 177–244 (henceforth referred to as HRG; 10 nM;
Genentech), group A (5 ng/ml EGF, 10 µg/ml bovine
transferrin, 1 µM ethanolamine, 1 µM
phosphatidylethanolamine, 40 nM selenium, and 5
pM T3; recombinant human EGF was
obtained from Collaborative Research (Waltham, MA),
selenious acid from Alfa Aesar Organics (Ward Hill, MA), and the rest
from Sigma Chemical Co.], group C [1 nM
hydrocortisone, 10 nM progesterone, and 1
µM forskolin; forskolin was obtained from
Calbiochem (San Diego, CA) and the rest from Sigma Chemical Co.], and 5 µl/ml bovine pituitary extract [BPE;
prepared as previously described (35)]. ASC cells (36) were grown on
laminin-coated plates in DMEM-Ham’s F-12 medium containing insulin (10
µg/ml), transferrin (10 µg/ml), forskolin (4 µM),
vitamin E (5 µg/ml; Aldrich, Milwaukee, WI),
progesterone (3 nM), HRG (5 nM),
and BPE (2 µl/ml). BR516 cells (35) were grown in DMEM-Ham’s F-12
medium supplemented with group A, group C, insulin (10 µg/ml), and
BPE (8 µl/ml). SFME cells (37) were cultured on fibronectin-coated
plates in DMEM-Ham’s F-12 medium containing selenium (40
nM), transferrin (10 µg/ml), insulin (10 µg/ml), and
EGF (10 µg/ml). Primary cardiac myocyte (CM) cultures were isolated
and prepared as described previously (38). They were cultured in
DMEM-Ham’s F-12 medium supplemented with transferrin (10 µg/ml),
insulin (1 µg/ml), aprotinin (1 µg/ml), glutamine (2 mmol/liter;
Life Technologies), penicillin G (100 U/ml; Life Technologies), and streptomycin (1 µg/ml; Life Technologies).
|
Real-time quantitative RT-PCR analyses for EGFR, ErbB2, ErbB3,
ErbB4, EGF, BTC, and HRG were performed using an ABI PRISM 7700
Sequence Detection System instrument and software (PE Applied Biosystems, Inc., Foster City, CA). The principles and protocols
for RT-PCR analyses using this system have been previously described
(39, 40, 41). Briefly, this method uses an oligonucleotide hybridization
probe that is labeled with a reporter fluorescent dye
(6-carboxy-fluorescein) at the 5'-end and with a quencher fluorescent
dye (6-carboxy-tetramethylrhodamine) at the 3'-end. Before the start of
the PCR reaction, when the probe is intact, the reporter dye emission
is due to the physical proximity of the reporter and quencher
fluorescent dyes. During the extension phase of the PCR cycle, however,
the nucleolytic activity of the Taq DNA polymerase cleaves
the hybridization probe and releases the reporter dye from the probe.
The resulting relative increase in reporter fluorescent dye emission is
monitored in real-time during PCR amplification using the Sequence
Detection System. The ribosomal protein L19 (RPL19) was chosen as an
internal standard to control for variability in amplification due to
differences in starting messenger RNA (mRNA) concentrations. The
expression of RPL19 mRNA itself is relatively uniform across the
different cell lines tested (Fig. 1).
Sequences for all primers and probes used in these analyses are listed
in Table 2. It should be noted that the
probe used to detect HRG mRNA will pick up any NRG1 isoform, including
HRGß1 and SMDF. All primer sets were designed to yield products
150–250 bp in size. All reagents for RT-PCR reactions were purchased
from Perkin Elmer (Norwalk, CT), except for the primers
and probes that were synthesized at the oligonucleotide synthesis core
facility at Genentech, Inc.. Each RT-PCR reaction
consisted of 100 ng cytoplasmic RNA in buffer A (50 mM KCl,
10 mM Tris-HCl, and the internal standard dye, ROX),
appropriate primers (200 nM) and probe (100
nM), 4.5 mM MgCl2, reverse
transcriptase, 2.5 U AmpliTaq DNA polymerase, and 200 µM
of each deoxy-NTP brought up to a final volume of 50 µl with water.
Two sets of control reactions were performed for all RNA samples: one
in which all of the above components were present except for RNA
itself, and the other in which the reverse transcriptase alone was
omitted. All reactions were performed under identical conditions, which
consisted of RT at 48 C for 30 min, followed by 40 cycles of
amplification with denaturation at 95 C for 30 sec, annealing at 55 C
for 30 sec, and elongation at 72 C for 1 min. All RNA extractions and
subsequent RT-PCR reactions were performed in triplicate. The relative
expression level of the gene of interest was computed with respect
to the mRNA expression level of the internal standard, RPL19, using the
formula:
 |
|
|
Growth assays
Cell proliferation in response to stimulation by ErbB ligands
was measured by a colorimetric assay using the Alamar Blue reagent
(Alamar Biosciences, Inc., Sacramento, CA). The reagent consists of an
oxidation-reduction (REDOX) indicator that undergoes a colorimetric
change due to the reduction of dye by living cells. The cells were
plated in 96-well dishes coated with either fibronectin (BUD and SFME)
or laminin (ASC) in the presence or absence (controls) of defined
concentrations of EGF, recombinant human betacellulin (BTC; R & D Systems, Inc., Minneapolis, MN), HRG, or recombinant human
SMDF (Genentech) (12), as indicated in the figure legends,
in a total volume of 100 µl. After incubation at 37 C for 70 h,
Alamar Blue reagent was added according to the manufacturer’s
suggested procedure (Alamar Biosciences) and incubated at 37 C for
2 h. At the end of the incubation period, the fluorescence (which
is proportional to cell number) (42, 43) was read at 
ex
- 530 nm and em - 595 nm in a Biolumin 960
fluorescence plate reader (Molecular Dynamics, Inc.,
Sacramento, CA). All treatments were performed in triplicate and
repeated in at least two separate experiments.
Statistical analyses
For RT-PCR data, a factorial ANOVA for receptor/ligand
expression with gene and cell line as the factors was performed
(software from StatView, Berkeley, CA). For growth assay data, ANOVA
for fold induction of growth with ligand and concentration as the
factors was performed for each cell line. Post-hoc analyses
were performed using the Bonferroni-Dunn correction. P
< 0.01 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The
highest levels of EGF mRNA were observed in the CM cultures, followed
by lower levels in ASC cells, and undetectable levels in the rest. BTC
mRNA was undetectable in all the cell lines examined. Although HRG mRNA
expression was detectable in ASC and CM cells, much higher expression
levels were observed in the BUD and BR516 cell lines (P
< 0.001 with respect to those in all other cell lines; Fig. 2).
|
). Highest levels of EGFR mRNA
were seen in the SFME cell line (P < 0.0001 with
respect to all cell lines), with lower levels in BUD and BR516 cell
lines. There was no detectable expression of EGFR mRNA in the ASC cell
line. Expression of ErbB2 mRNA was less restricted than that of the
other receptor types and was detected in all of the cell lines tested.
The highest expression of ErbB3 mRNA was seen in the Schwann cells
(ASC; P < 0.0001 with respect to all cell lines).
Lower levels were also detectable in the BUD and BR516 cell lines. The
expression of ErbB4 mRNA appeared to be most restricted, with highest
levels in the neonatal CM (P < 0.0005 with respect to
all cell lines, except SFME) and lower levels in SFME cells.
|
).
These three cell lines were chosen because they represent the three
types of receptor combinations found in this series of cell lines. The
strongest immunoreactivity for EGFR was seen in SFME cells, with a much
lower signal in the BUD cells, whereas highest levels of ErbB3 protein
were seen in ASC cells, followed by a weaker signal in the BUD cells.
All three cell lines expressed significant levels of ErbB2,
consistent with the mRNA measurements. Due to the presence of two
closely migrating species of ErbB2 in the BUD cells, the ErbB2 band
appears wider in the BUD cells than in the ASC cells, although the
intensity of the ErbB2 band is higher in the ASC cells. Similar
migration of two ErbB2 species was seen in the SFME cell lysate,
although here the separation between the bands was more distinct due to
lower levels of ErbB2 in the SFME cells. Immunoreactivity for ErbB4 at
the correct molecular mass (185 kDa) was only detected in SFME cells.
Note that the use of different antibodies and exposure times did not
allow direct comparison of protein levels between receptor immunoblots.
The band intensities on each blot are proportional to the differences
in mRNA expression levels, except in the case of ErbB2.
|
.
Significant growth stimulation was observed in BUD and BR516 cells in
response to stimulation with EGF, BTC, and HRG (P <
0.001). Neither of these cell lines responded to SMDF at any of the
doses tested. Both EGF and BTC stimulated cell proliferation in SFME
cells (P < 0.001), although these cells were
unresponsive to HRG and SMDF addition. The ASC cells, on the other
hand, responded robustly to stimulation by HRG and SMDF
(P < 0.001), but were unresponsive to EGF and BTC
addition. The maximal growth response was observed even at the lowest
dose of 2.5 nM for EGF and BTC in the BUD, BR516, and SFME
cells. Likewise, the growth response in ASC cells to HRG and SMDF was
maximal at 2.5 nM. The CM cells were not tested in the
growth assay, because they are terminally differentiated myocytes and
do not proliferate any further in vitro. However, an
important end point of responsiveness to hormonal and cytokine
stimulation in these cells is a hypertropic response (38). Of the four
ligands tested, EGF, BTC, and SMDF did not cause any significant
hypertrophy, whereas HRG induced a slight hypertropic response (data
not shown).
|
), except CM. Treatment with EGF
and BTC elicited responses in BR516 and SFME cells at 5–25
pM, although the magnitude of response was different
between these two cell lines. The ASC cell line, which has high levels
of ErbB3, showed the greatest sensitivity to HRG, with detectable
response at 5–25 pM. The BR516 cell line did not respond
to HRG in this dose range. The BUD cell line did not respond to any of
these ligands in this dose range.
|
). This protein consists of the
extracellular domain of ErbB4 fused to IgG and would be expected to
bind and sequester free extracellular HRG (44). The BUD cells remained
unaffected in the presence of the ErbB4-IgG fusion protein and showed
no change in the basal growth rate. In contrast, the BR516 cells showed
a 40% reduction in the basal growth rate in the presence of the
ErbB4-IgG fusion protein (P < 0.005).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Measurement of ErbB receptor mRNA levels indicated that all the cell lines examined expressed either moderate (BUD, BR516, SFME, and CM) or high (ASC) levels of ErbB2. This result is in agreement with the generally accepted model of the ErbB2 receptor acting as a signaling partner for the other ErbB receptors. The Schwann cells showed high levels of expression of ErbB3 compared with the other cell lines (35-fold higher than BR516 and BUD cell lines), and these cells responded most robustly to stimulation with either HRG or SMDF, another neuregulin isoform. The response to HRG was detected even at 25 pM, whereas the response to SMDF was detectable from 125 pM. Even though the BUD and BR516 cells respond to HRG at nanomolar doses, they did not show any significant increases in growth rates upon treatment with picomolar doses of HRG. One possible explanation for this result would involve the autocrine stimulation of these cells by endogenously synthesized HRG secreted by the cells themselves, thereby making them insensitive to low doses of exogenous ligand. Conditioned medium collected from BUD and BR516 cells did not contain detectable levels of the ligand, as measured by an enzyme-linked immunosorbent assay assay (data not shown) that detects HRG levels as low as 1.25 nM. However, very low concentrations of HRG might be sufficient to produce an autocrine effect, and in addition, membrane-bound forms of HRG and other ErbB ligands have been shown to be capable of signaling (45, 46, 47, 48). Therefore, we used a fusion protein consisting of the extracellular domain of ErbB4 conjugated to IgG to antagonize endogenous HRG that may be membrane bound or secreted into the medium of BUD and BR516 cells. Addition of this fusion protein caused no change in the growth properties of the BUD cells under basal conditions, indicating that these cells are not likely to undergo autocrine stimulation by HRG. However, the reduction in the growth rate of BR516 cells indicates that HRG and/or some other secreted ligand that is capable of binding to ErbB4 is involved in autocrine stimulation of these cells.
In addition to such autocrine effects, there is evidence for a paracrine role for the ErbB ligands as well, with the active ligand secreted from the adjacent mesenchymal compartment or other closely associated cells (23, 25, 49, 50). For instance, CMs are in close contact with fibroblasts in the surrounding heart tissue. When RNA extracted from a purified population of these fibroblasts was assayed for expression of ErbB ligands and receptors, high levels of HRG mRNA were detected, with no detectable levels of ErbB3 and ErbB4 (data not shown). Thus, HRG secreted by cardiac fibroblasts may be available to simulate CMs in a paracrine fashion. Cardiac fibroblasts have been shown to secrete other factors in culture that can stimulate hypertrophy in CMs (38).
The differential sensitivities of BUD, BR516, and ASC cells to HRG suggest that in addition to determining whether a particular cell will respond to a specific ligand, the receptor levels also determine the sensitivity of the cell to that ligand. This effect is also seen with respect to the effect of low doses of EGF on the BUD and SFME cell lines. In this instance, although the BUD cells respond to EGF at higher doses, they fail to do so in the picomolar dose range. In contrast, the SFME cell line that expresses high levels of EGFR (7- and 20-fold greater than BR156 and BUD, respectively) is able to respond to picomolar doses of EGF. As the SFME cells require EGF for survival (51), it is difficult to interpret data from treatments with doses lower than 25 pM, which is the minimum dose required for survival. At doses higher than 25 pM, SFME cells respond more robustly to EGF than any of the other cell lines in this series (10-fold greater response that BUD or BR516 cell lines). This effect may be partly attributed to the death of control cells grown in the absence of EGF, thereby increasing the apparent magnitude of response. Although BUD and BR516 cells appear to have similar ErbB receptor levels, it should be noted that there is a 50% higher expression of EGFR in BR516 cells compared with that in BUD cells. This difference might account for the differential responsiveness of BUD and BR516 cells to picomolar doses of EGF.
Neither BR516 nor BUD cells responded to SMDF even in the higher (nanomolar) dose range. It has been demonstrated in vitro that the binding affinity of SMDF for ErbB2/3 dimer is lower than its affinity for the ErbB2/4 pair, but in either case it is not nearly as high as the affinity of these heterodimers for HRG (Osheroff, P. L., S. P. Tsai, N. Y. Chiang, K. L. King, J. P. Mather, manuscript in preparation). However, the lower affinity of the ligand may be compensated for in the Schwann cells by the high expression levels of ErbB2 and ErbB3, providing a possible explanation for the responsiveness of these cells to SMDF, in contrast to that of the BUD and BR516 cell lines.
The tight correlation observed between the levels of expression of
certain ErbB receptors and the growth responsiveness to ligand
stimulation (summarized in Table 3)
supports the hypothesis that the growth responsiveness of a tissue type
to a specific ErbB ligand is dependent on the ErbB receptor expression
profile in that tissue. Similar results have been obtained in certain
other studies that have used cells transfected with one or more ErbB
receptors (53, 54, 55). Our study further extends these observations to
include nontransformed and untransfected cells derived from normal
tissues. In the current series of experiments, high expression of EGFR
correlated with increased growth stimulation by EGF and BTC, whereas
high levels of ErbB3 correlated with a greater mitogenic response to
HRG and SMDF, with all cell lines expressing appreciable levels of
ErbB2. However, unlike the other ErbB receptors, levels of ErbB2
protein did not correlate well with mRNA levels. It should be noted
that mRNA levels do not always accurately reflect protein levels for
many reasons, including differential rates of synthesis and proteolysis
for different proteins. It should also be noted that we have only
examined one end point of ligand stimulation, namely growth. Whether
other effects of ErbB stimulation, such as differentiation or survival,
are similarly correlated with the expression levels of ErbB receptors
or whether different levels of receptor occupancy might be required for
different responses remains to be determined. Furthermore, it is
conceivable that treatment of these cell lines with ErbB ligands may
induce the expression of growth factors or alter the expression of
different ErbB receptors and ligands. In addition, these growth
experiments can be expected to estimate the upper range of dose
required for biological responsiveness, because in vivo
ligand production is likely to be continuous (at least for short
periods of time) rather than the single dose stimulation given in
vitro in these experiments.
|
|
Acknowledgments |
|---|
Received June 30, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and erbB-3 receptor in
human prostatic adenocarcinoma. Br J Urol 79:212–216[Medline]
-Heregulin: a novel heregulin isoform that is an autocrine growth
factor for the human breast cancer cell line, MDA-MB-175. Oncogene 15:1385–1394[CrossRef][Medline]
precursors
activate EGF/TGF- receptors. Cell 56:691–700[CrossRef][Medline]
precursor expressed
on the cell surface binds to the EGF receptor on adjacent cells,
leading to signal transduction. Cell 56:495–506[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  K. Zscheppang, W. Liu, M. V. Volpe, H. C. Nielsen, and C. E. L. Dammann ErbB4 regulates fetal surfactant phospholipid synthesis in primary fetal rat type II cells Am J Physiol Lung Cell Mol Physiol, August 1, 2007; 293(2): L429 - L435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Gordon, D. Matei, C. Aghajanian, U. A. Matulonis, M. Brewer, G. F. Fleming, J. D. Hainsworth, A. A. Garcia, M. D. Pegram, R. J. Schilder, et al. Clinical Activity of Pertuzumab (rhuMAb 2C4), a HER Dimerization Inhibitor, in Advanced Ovarian Cancer: Potential Predictive Relationship With Tumor HER2 Activation Status J. Clin. Oncol., September 10, 2006; 24(26): 4324 - 4332. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. R. Sussman, T. Vartanian, and R. H. Miller The ErbB4 Neuregulin Receptor Mediates Suppression of Oligodendrocyte Maturation J. Neurosci., June 15, 2005; 25(24): 5757 - 5762. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. B. Agus, M. S. Gordon, C. Taylor, R. B. Natale, B. Karlan, D. S. Mendelson, M. F. Press, D. E. Allison, M. X. Sliwkowski, G. Lieberman, et al. Phase I Clinical Study of Pertuzumab, a Novel HER Dimerization Inhibitor, in Patients With Advanced Cancer J. Clin. Oncol., April 10, 2005; 23(11): 2534 - 2543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Dagnon, E. Pacary, F. Commo, M. Antoine, M. Bernaudin, J.-F. Bernaudin, and P. Callard Expression of Erythropoietin and Erythropoietin Receptor in Non-Small Cell Lung Carcinomas Clin. Cancer Res., February 1, 2005; 11(3): 993 - 999. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Sithanandam, G. T. Smith, A. Masuda, T. Takahashi, L. M. Anderson, and L. W. Fornwald Cell cycle activation in lung adenocarcinoma cells by the ErbB3/phosphatidylinositol 3-kinase/Akt pathway Carcinogenesis, October 1, 2003; 24(10): 1581 - 1592. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. D. Leikauf, M. T. Borchers, D. R. Prows, and L. G. Simpson Mucin Apoprotein Expression in COPD* Chest, May 1, 2002; 121(5_suppl): 166S - 182S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. Lingohr, L. M. Dickson, J. F. McCuaig, S. R. Hugl, D. R. Twardzik, and C. J. Rhodes Activation of IRS-2--Mediated Signal Transduction by IGF-1, but not TGF-{alpha} or EGF, Augments Pancreatic {beta}-Cell Proliferation Diabetes, April 1, 2002; 51(4): 966 - 976. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Pawlowski, F. Revillion, M. Hebbar, L. Hornez, and J.-P. Peyrat Prognostic Value of the Type I Growth Factor Receptors in a Large Series of Human Primary Breast Cancers Quantified with a Real-Time Reverse Transcription-Polymerase Chain Reaction Assay Clin. Cancer Res., November 1, 2000; 6(11): 4217 - 4225. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Feldman, B. H. Lorell, and S. E. Reis Trastuzumab in the Treatment of Metastatic Breast Cancer : Anticancer Therapy Versus Cardiotoxicity Circulation, July 18, 2000; 102(3): 272 - 274. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
