Vol. 277, Issue 6, C1160-C1169, December 1999
EDITORIAL FOCUS
Actin filament organization is required for proper
cAMP-dependent activation of CFTR
Adriana G.
Prat1,2,
C. Casey
Cunningham2,3,
G. Robert
Jackson Jr.1,
Steven C.
Borkan4,
Yihan
Wang4,
Dennis A.
Ausiello1,2, and
Horacio F.
Cantiello1,2
1 Renal Unit, Massachusetts General Hospital
East, Charlestown 02129; 2 Department of Medicine,
Harvard Medical School, and 3 Experimental Medicine,
Brigham and Women's Hospital, Boston 02115; and
4 Renal Section, Boston Medical Center, Boston,
Massachusetts 02118
 |
ABSTRACT |
Previous studies
have indicated a role of the actin cytoskeleton in the regulation of
the cystic fibrosis transmembrane conductance regulator (CFTR) ion
channel. However, the exact molecular nature of this regulation is
still largely unknown. In this report human epithelial CFTR was
expressed in human melanoma cells genetically devoid of the filamin
homologue actin-cross-linking protein ABP-280 [ABP(
)]. cAMP stimulation of ABP() cells or
cells genetically rescued with ABP-280 cDNA [ABP(+)] was
without effect on whole cell Cl
currents. In
ABP() cells expressing CFTR, cAMP was also without effect on
Cl conductance. In contrast, cAMP induced a 10-fold
increase in the diphenylamine-2-carboxylate (DPC)-sensitive whole cell
Cl currents of ABP(+)/CFTR(+) cells. Further, in
cells expressing both CFTR and a truncated form of ABP-280 unable to
cross-link actin filaments, cAMP was also without effect on CFTR
activation. Dialysis of ABP-280 or filamin through the patch pipette,
however, resulted in a DPC-inhibitable increase in the whole cell
currents of ABP()/CFTR(+) cells. At the single-channel level,
protein kinase A plus ATP activated single Cl
channels only in excised patches from ABP(+)/CFTR(+) cells.
Furthermore, filamin alone also induced Cl channel
activity in excised patches of ABP()/CFTR(+) cells. The present
data indicate that an organized actin cytoskeleton is required for
cAMP-dependent activation of CFTR.
ABP-280; cystic fibrosis transmembrane conductance regulator; actin
cytoskeleton; adenosine 3',5'-cyclic monophosphate
| |
INTRODUCTION |
THE CYSTIC FIBROSIS transmembrane conductance regulator
(CFTR) is an anion-selective channel whose dysfunction leads to
the onset of cystic fibrosis (1, 2, 29). CFTR activation is normally
elicited by stimulation of the cAMP pathway and protein kinase A (PKA)
activation. This is thought to be consistent with the fact that several
PKA-dependent phosphorylation sites have been found in CFTR. Other
regulatory mechanisms, however, have also been implicated in CFTR
regulation (8, 11, 30). Recent studies (4, 27) have determined a
regulatory role of actin in the activation of CFTR by the cAMP pathway
targeting PKA activation. In those studies, partial disruption of the
actin cytoskeleton with cytochalasin D induced activation of CFTR
Cl channel activity in the absence of PKA activation
(27). Furthermore, extended treatment with cytochalasin D (6-9 h)
to collapse the actin cytoskeleton completely prevented CFTR activation
by direct addition of PKA (27). However, PKA-insensitive CFTR function was readily restored by addition of exogenous actin, which is consistent with the presence of potential actin-binding domains in CFTR
(27). This raises the possibility for the actin cytoskeleton to
directly interact with and regulate CFTR (27). The ubiquitous and
abundant distribution of actin, however, may contribute against the
idea of this molecule behaving as a conventional "ligand" in the
regulation of CFTR, since actin filaments can take several conformations within the cytoplasm. We have previously demonstrated, for example, that a distinct form of "short" actin filaments may be responsible for CFTR activation, similar to that reported in previous studies with epithelial Na+ channels (26) and the
Na+-K+-ATPase (3).
The actin-binding protein ABP-280 and its muscle isoform filamin induce
the orthogonal cross-linking of actin filaments into three-dimensional
networks (20). Although deriving from different genes, both
actin-cross-linking protein isoforms are >70% identical and are thus
expected to have functional similarities (14, 15). Human melanoma cells
devoid of the actin-binding protein ABP-280 [ABP()]
display an impaired motility and a dysfunctional actin organization but
recover both a normal cytoskeletal phenotype and functional properties
by transfection of the full-length ABP-280 cDNA [ABP(+)
cells] (10). Furthermore, ABP() cells are unable to elicit
a normal cell volume regulatory response due to their inability to
modulate ion channel activity (7). Genetically rescued ABP(+) cells,
however, recover both the ability to regulate cell volume and the
ability to modulate ion channel function. Moreover, the dynamics of
actin filament organization are also relevant for CFTR regulatory
mechanisms, because further addition of filamin had an inhibitory
effect on its ion channel function (27). Therefore, in the present
study it was hypothesized that ABP() and ABP(+) cells were
excellent models to further assess the regulatory role of actin
filament organization in CFTR regulation. Our studies indicate that
cross-linked actin networks organized by the interaction between
ABP-280 and actin are essential for CFTR to function as an anion channel.
| |
MATERIALS AND METHODS |
Human Melanoma Cell Lines
Human melanoma cells, grown as previously described (10), were
originally derived from the transfection of a parent ABP() cell
line (M2) with an LK444 expression vector encoding for resistance to
G418, which either did or did not contain the cDNA for full-length ABP-280 (10). ABP() and ABP(+) cells were further transfected with the pREP vector encoding for resistance to hygromycin, which either did or did not contain the full-length wild-type CFTR cDNA. After previously described standard transfection procedures (10) were
performed, CFTR-expressing clones were selected and individually cultured with hygromycin (200 µg/ml)-containing medium. Four cell lines were thus obtained with all possible combinations of presence or
absence of ABP-280 and CFTR, namely, ABP()/CFTR(),
ABP(+)/CFTR(), ABP()/CFTR(+), and ABP(+)/CFTR(+) cells.
CFTR was also expressed in human melanoma cells expressing a mutated
form of ABP-280 that binds to membrane proteins but is specifically
truncated at the self-association domain of the molecule, thus
rendering the protein unable to cross-link actin filaments into gel
networks. Melanoma cells transfected with this mutated form of ABP-280
[ABP-Trunc(+)/CFTR(+)] expressed a protein missing up to
250 amino acids from the COOH-terminal end of ABP-280.
Briefly, truncated ABP-280 cDNA was obtained by using controlled
Exo III endonuclease digestion of the full-length ABP-280 cDNA truncated from 80 to 1,100 bp deleted from the 3'-end.
Patch-Clamp Techniques
Whole cell currents. Patch pipettes were made of KG-33 glass
capillaries (Garner Glass, Claremont, CA). Actual currents and command
voltages were obtained and driven with a Dagan 3900 amplifier (Dagan,
Minneapolis, MN) using a 1-G
feedback resistor in the head stage.
Data were acquired, digitized, and stored as indicated below. Holding
potentials refer to the patch pipette. The patch pipette and bathing
solution was (in mM) 140 NaCl, 5.0 KCl, 0.8 MgCl2, 1.2 CaCl2 and 10 (HEPES), pH 7.4. In some experiments, NaCl in
the pipette (140 mM) was replaced by MgCl2 (70 mM), all other solutes remaining the same. To determine anion selectivity of the
cAMP-activated whole cell currents, the bathing NaCl solution was
replaced with solutions containing HEPES (10 mM) and equimolar concentrations of either NaBr, NaI, or sodium gluconate, pH 7.4. Whenever indicated, the patch pipette was filled up to one-third of its
height with MgATP (100 mM, pH 7.4, adjusted with
N-methylglucamine) as previously described (8, 27, 28). The
rest of the pipette was backfilled with the NaCl-containing solution as
described above.
Single-channel currents. Actual currents and command voltages
were obtained and driven with a PC-501 patch-clamp amplifier (Warner
Instruments, Hamden, CT) using a 10-G feedback resistance in the
head stage as previously reported (9, 25). Signals were filtered at
1,000 Hz with an eight-pole Bessel filter (Frequency Devices,
Haverhill, MA). Data were acquired, digitized, and stored in a hard
disk of a personal computer through a TLL interface (Tecmar) until
further analysis with pCLAMP 6.0.3 (Axon Instruments, Foster City, CA).
Patch pipette and bathing solutions were as indicated for the whole
cell current experiments, containing either MgCl2 (70 mM)
or N-methylglucamine chloride (140 mM). Following our previous
studies on the role of the actin cytoskeleton in CFTR function (27),
whole cell and single-channel experiments were conducted at room
temperature (22 °C).
Detection of ABP-280 and CFTR in Melanoma Cells by Western Blot
Analysis
The presence or absence of CFTR and ABP-280 were determined in
control and transfected human melanoma cells using immunoblot analysis. Cells were harvested by washing with ice-cold
Ca2+-free PBS and were scraped, centrifuged, and
resuspended in ice-cold lysis buffer [1% Triton X-100, 0.5%
Nonidet P-40, 150 mM NaCl, 10 mM Tris · HCl (pH 7.5),
1 mM EGTA, 0.25 mM sodium vanadate, 10 µg/ml phenylmethylsulfonyl
fluoride, and 10 µg/ml aprotinin], scraped, and then frozen at
80°C. Samples were thawed on ice, and a whole cell lysate
was obtained by passing each sample through a 26-gauge needle
(×15 passes). Protein content was determined with the BCA
protein assay (Pierce, Rockford, IL). CFTR and ABP-280 samples were
separated by a 7.5% SDS-polyacrylamide gel and a 4-20%
SDS-polyacrylamide gradient gel, respectively. Samples were then
transferred to polyvinylidene difluoride membranes and blocked for 1 h
with 5% dried milk also containing 0.5% nonimmune goat serum in
50 mM Tris · Hcl, pH 7.6, 141 mM NaCl, and 0.2% Tween 20 (TBST). Blots were probed with one of two monoclonal antibodies directed against CFTR (MAb 13-1 or MAb 24-1 from Genzyme,
Cambridge, MA; 0.5-1 µg/ml) or ABP-280 (1 µg/ml), diluted in
TBST containing 1% BSA, for 24-48 h at 4°C. Binding of
primary antibody was detected with a horseradish peroxidase-based
enzyme-linked chemiluminescence system (Kirkegaard & Perry,
Gaithersburg, MD).
Actin-Binding Proteins
Muscle filamin from chicken gizzard (Sigma), ~1 mg/ml stock solution
in water, was diluted 200-fold into either the patch pipette or the
chamber. Nonmuscle filamin, ABP-280, purified from rabbit alveolar
macrophages as previously described (17), was a kind gift from Dr. John
H. Hartwig (Brigham and Women's Hospital, Boston, MA).
Other Reagents
The cAMP stimulatory cocktail contained 8-bromoadenosine
3',5'-cyclic monophosphate (500 µM), IBMX (200 µM), and forskolin (10 µM). The Cl channel blocker
diphenylamine-2-carboxylate (DPC; Fluka Chemical, Ronkonkoma, NY) was
kept in a 20 mM stock solution in ethanol. DIDS (Sigma) was kept in a
10 mM stock solution in distilled water and used at a final
concentration of 500 µM. The catalytic subunit of the cAMP-dependent
protein kinase (PKA; Sigma) was used at a final concentration of 10 µg/ml. The monoclonal antibody MAb 13-1 (Genzyme) raised against
the R-domain of CFTR was used at a final concentration of 2.92 µg/ml.
Whenever indicated, a heat-inactivated antibody was prepared by
incubating the antibody at 100 °C for 30 min.
Permeability-to-Selectivity Ratio Calculations
The ATP over Cl
permeability-to-selectivity ratio
(PATP/PCl) was calculated with
a derivation of the Goldman-Hodgkin-Katz equation (18) from the
cAMP-stimulated Cl and ATP currents obtained under
asymmetrical conditions, such that
where
a = F/RT, zA is the valence
of ATP, Ai is the concentration of ATP (100 mM), and
Vr is the reversal potential of the whole cell
currents under asymmetrical conditions.
Similarly, the permeability-to-selectivity ratio of different anions
(PCl/PY) was calculated
using a modified equation such that
where
a = F/RT, zCl is the
valence of Cl (1), Cli is
the concentration of intracellular Cl (149 mM),
zY is the valence of the substituted anion
(1 for either Br, I, or
gluconate), and Yi is the concentration of extracellular anion (140 mM). U is the difference between the calculated
reversal potential (EY, 1.59 mV) and the
observed reversal potential (Vr) under asymmetrical conditions.
| |
RESULTS |
Effect of ABP-280 Expression on the Whole cell Currents of
CFTR-Expressing Human Melanoma Cells
To determine the role of ABP-280 in CFTR activation, the effect of a
cAMP-stimulatory cocktail was assessed on ABP() and ABP(+)
melanoma cells also expressing CFTR. Addition of cAMP cocktail to
either the ABP() cells also lacking CFTR
[ABP()/CFTR()] or cells transfected with
ABP-280 alone [ABP(+)/CFTR()] was without effect on
the whole cell Cl currents [0.67 ± 0.19 nS/cell (n = 7) vs. 0.31 ± 0.11 nS/cell (n = 6)
and 1.50 ± 0.72 nS/cell (n = 3) vs. 0.81 ± 0.50 nS/cell (n = 3) for control and cAMP conditions, respectively; Fig.
1, A and B]. Whole cell currents
from CFTR-expressing ABP() cells [ABP()/CFTR(+)] were also insensitive to the cAMP
stimulatory cocktail [0.91 ± 0.31 nS/cell (n = 9) vs.
1.68 ± 0.67 nS/cell (n = 9); Figs. 1C and
2A]. Addition
of cAMP stimulatory cocktail to CFTR-expressing, ABP(+) cells,
[ABP(+)/CFTR(+)], in contrast, induced a 1,280% increase
in the whole cell Cl currents [0.85 ± 0.36 nS/cell (n = 11) vs. 10.4 ± 2.25 nS/cell (n = 8),
P < 0.001; Figs. 1D and 2B]. However,
cells expressing both CFTR and a truncated form of ABP-280 unable to
cross-link actin filaments [ABP-Trunc(+)/CFTR(+) cells]
were also insensitive to cAMP stimulation [2.56 ± 0.65 nS/cell
(n = 10) vs. 2.59 ± 1.00 nS/cell (n = 6); Fig.
2C]. One experiment out of seven tested in
ABP-Trunc(+)/CFTR(+) cells showed cAMP activated whole cell Cl currents (from 2.9 to 11.2 nS/cell), indicating a
maximal response similar to that of ABP(+)/CFTR(+) cells.
View larger version (12142015K):
[in this window]
[in a new window]
|
Fig. 1.
Effect of cAMP on basal Cl currents of cystic
fibrosis transmembrane conductance regulator (CFTR) transfected human
melanoma cells. Representative whole cell currents obtained between
+100 and 100 mV in symmetrical Cl (140 mM) in
ABP()/CFTR() (A), ABP(+)/CFTR()
(B), ABP()/CFTR(+) (C), and ABP(+)/CFTR(+)
(D) cells, before (top trace) and after (middle
trace) addition of a cAMP-stimulatory cocktail. cAMP-activated
whole cell currents from ABP(+)/CFTR(+) cells were readily inhibited by
diphenylamine-2-carboxylate (DPC; 0.5 mM; bottom traces). Data
are representative of 3-9 experiments for the various cell
types.
|
|
View larger version (141713K):
[in this window]
[in a new window]
|
Fig. 2.
Effect of cAMP on Cl whole cell currents of
CFTR-transfected human melanoma cells. Current-voltage relationships
from the whole cell currents of ABP()/CFTR(+) (A),
ABP(+)/CFTR(+) (B), and ABP-Trunc(+)/CFTR(+) (C) cells
were obtained before ( ) and after ( ) addition of a
cAMP-stimulatory cocktail in the presence of symmetrical
Cl (140 mM). Data are from 4-6 experiments.
Whole cell currents (I, nA) were obtained between +100 and
100 mV. Vh, holding potential applied to the
pipette electrode.
|
|
The cAMP-activated Cl currents of ABP(+)/CFTR(+)
cells were readily inhibited by DPC [0.5 mM; 10.4 ± 2.25 nS/cell (n = 8) vs. 2.31 ± 0.78 nS/cell (n = 4) for
cAMP-activated and after DPC, respectively, P < 0.01; Fig.
3], indicative of the presence of CFTR-associated Cl channel activity as previously
reported (21, 28, 29). In two of seven experiments, the whole cell
Cl currents of ABP(+)/CFTR(+) cells were
spontaneously activated (6.10 ± 0.80 nS/cell), although they were
further activated by cAMP addition (12.8 ± 4.20 nS/cell).
Spontaneously active Cl currents were inhibitable by
DPC (data not shown).
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 3.
Effect of cAMP and DPC on Cl and ATP whole cell
currents of ABP(+)/CFTR(+) cells. Whole cell currents were obtained in
symmetrical Cl (140 mM) or in the presence of MgATP
(100 mM) in the patch pipette and bathing NaCl (140 mM). The whole cell
conductance was calculated before (left bar) and after
(middle bar) addition of a cAMP-stimulatory cocktail and after
further addition of DPC (500 µM) to the bathing solution (right
bar). Data are from 11, 8, and 4 experiments for control, cAMP, and
cAMP + DPC, respectively. * P < 0.001 and
** P < 0.01.
|
|
Anion Selectivity of the cAMP-Activated Pathway
The anion permeability-to-selectivity ratio of the cAMP-activated whole
cell conductance of ABP(+)/CFTR(+) cells was Cl
(1.0) > Br (0.88) 
I (0.81)
gluconate (0.53), determined by the shift in reversal potential
(n = 4-5, P < 0.05, for all ratios) when
extracellular Cl was replaced with other anions. This is in agreement
with previous data of wild-type CFTR, which follow almost the same
permeability-to-selectivity ratio series (33). Another indication for
CFTR to be activated in the CFTR(+)/ABP(+) cells comes from the
permeability to cellular ATP observed, in agreement with previous
studies (27, 28). Under asymmetrical ATP/Cl conditions (100 mM MgATP in
pipette, 140 mM NaCl in bath), ATP-permeable whole cell currents were
also observed after activation with the cAMP cocktail [1.22 ± 0.37 nS/cell (n = 6) vs. 5.70 ± 1.30 nS/cell (n = 3), P < 0.02, for control and cAMP activated,
respectively]. The Cl and ATP currents were
simultaneously inhibited by bathing DPC [5.70 ± 1.30 nS/cell
(n = 3) vs. 1.80 ± 1.30 nS/cell (n = 2)]. Further, the reversal potential of the cAMP-activated
ATP/Cl currents indicated an
ATP/Cl permeability-to-selectivity ratio of 0.40, in
agreement with values previously reported for CFTR-expressing cells
(28) and purified CFTR (6). In summary, the cAMP stimulatory cocktail only activated CFTR-mediated whole cell currents in the
ABP-280-expressing [ABP(+)/CFTR(+)] cells.
Effect of Intracellular ABP-280 Dyalisis on the Whole cell Currents
of CFTR-Expressing ABP() Cells
To further assess the regulatory nature of actin filament organization
on the cAMP-dependent activation of CFTR in human melanoma cells, the
effect of the actin-cross-linking protein ABP-280 was assessed in
ABP()/CFTR(+) cells by dialysis of either filamin (0.1-20
nM) or the nonmuscle isoform ABP-280 (2 nM) from the patch pipette
(Fig. 4, A and
B). In the presence of intracellular ABP-280, addition of cAMP
stimulatory cocktail induced CFTR activation similar to that observed
in ABP(+)/CFTR(+) cells in five of six experiments (Fig. 4A).
When filamin was intracellularly dialyzed from the patch pipette, a
dose-response effect was observed. cAMP-dependent activation of the
whole cell currents of ABP()/CFTR(+) cells was observed in five
of five experiments between 0.5 and 20 nM intracellular filamin,
whereas no effect was detected when using 0.1 nM filamin (n = 3, Fig. 4B). Taken together, these data further suggest that
CFTR is expressed in the plasma membrane of the cells but requires the
presence of an actin-cross-linking protein to be sensitive to the cAMP
pathway. The presence of CFTR in ABP()/CFTR(+) cells was further
determined by Western blotting, indicating that the level of CFTR
expression was similar to that of ABP(+)/CFTR(+) cells (Fig.
4C). CFTR() cells displayed no specific labeling for
this channel protein (Fig. 4C).
View larger version (181669K):
[in this window]
[in a new window]
|
Fig. 4.
Effect of intracellular ABP-280 on basal Cl currents
of ABP()/CFTR(+) cells. A: whole cell currents at +100
mV were obtained in symmetrical NaCl (140 mM) in the presence () or
the absence () of ABP-280 (2 nM) in the patch pipette and followed
as a function of time. Arrows indicate addition of cAMP stimulatory
cocktail and DPC. Data are from a representative experiment for each
condition from 5 and 4 experiments for ABP-280 and control,
respectively. B: cAMP-dependent whole cell
Cl currents at +100 mV were also obtained in the
presence of intracellular filamin (0.1-20 nM;  ). Results were
compared with those obtained in the presence of ABP-280 (2 nM) in the
patch pipette ( ). Data are from 8 and 5 experiments for filamin and
ABP-280, respectively. C: detection of ABP-280 and CFTR in
human melanoma cells. Presence or absence of CFTR and ABP-280 were
determined in control and transfected human melanoma cells as described
in MATERIALS AND METHODS. The CFTR and ABP-280 studies were
each performed on a single gel but are shown separated to improve
clarity. Immunolabeling of CFTR (top) was conducted in
membranes isolated from ABP(+)/CFTR(+) (left lane),
ABP()/CFTR(+) (middle lane), and
ABP()/CFTR() (right lane) cells. Data are
representative of 3 experiments.
|
|
Single-Channel Currents of CFTR-Expressing Human Melanoma Cells
The effect of ABP-280 on the cAMP-dependent activation of CFTR was also
assessed at the single-channel level. Addition of PKA (10 µg/ml), in
the presence of ATP (1 mM), to the cytoplasmic side of excised,
inside-out patches of either ABP()/CFTR() (Fig. 5A) or ABP(+)/CFTR() (Fig.
5B) cells had no effect on single ion channel activity
(n = 3 for each condition). Furthermore, PKA plus ATP was also
without effect on excised patches from ABP()/CFTR(+) cells
(n = 3, Fig. 5C). However, addition of PKA plus ATP
readily activated single Cl channel currents in
excised patches from ABP(+)/CFTR(+) cells (n = 6, Fig.
5D). The PKA-activated ion channels of ABP(+)/CFTR(+) cells had
a conductance of 9.57 ± 0.58 pS (n = 12, Fig.
6A) in symmetrical
Cl (150 mM). The PKA-activated channels were
inhibited by anti-CFTR antibodies (2.92 µg/ml, Fig. 6B) and
by the Cl channel blocker DPC (0.5 mM, Fig.
6C) but were insensitive to both a heat-inactivated antibody
(2.92 µg/ml, Fig. 6B) and to DIDS (0.4 mM, n = 3, data not shown). The current-voltage relationship and the inhibitory
effects of DPC and anti-CFTR antibodies, but not DIDS, were consistent
with the functional fingerprinting of Cl-permeable
CFTR-mediated single-channel currents. Addition of actin (1 mg/ml) to
the cytoplasmic side of excised patches from ABP(+)/CFTR(+) cells also
induced ion channel activity similar to PKA (data not shown).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 5.
Effect of protein kinase A (PKA) on CFTR activation in
ABP-280-expressing cells. Addition of PKA (10 µg/ml) + ATP (1 mM) to the cytoplasmic side of quiescent excised, inside-out patches
from ABP()/CFTR() (A), ABP(+)/CFTR()
(B), ABP()/CFTR(+) (C), and ABP(+)/CFTR(+)
(D) cells induced and/or increased Cl
channel activity only in cells transfected with both CFTR and ABP-280.
Holding potential was 80 mV. Channels in ABP(+)/CFTR(+) cells
activated in clusters (see also Fig. 6B). Dashed line indicates
the closed state of the ion channels. Traces obtained in symmetrical
Cl conditions (70 mM MgCl2/140 mM NaCl)
are representative of 3 (A-C) and 6 (D) experiments.
|
|
View larger version (163521K):
[in this window]
[in a new window]
|
Fig. 6.
Single-channel currents of CFTR-mediated Cl currents
in human melanoma cells. A: single-channel currents
(left) were obtained in symmetrical Cl (140 mM N-methylglucamine chloride) from excised inside-out patches
of ABP(+)/CFTR(+) cells. Current-voltage relationship (right)
had a slope of 9.57 ± 0.58 pS (n = 12). B: addition
of the anti-CFTR monoclonal antibody MAb 13-1 (2.92 µg/ml) to
excised patches from ABP(+)/CFTR(+) cells (bottom trace)
inhibited the PKA-stimulated ion channel activity (top trace).
Heat-inactivated MAb 13-1 (2.92 µg/ml), however, was without
effect on inhibiting the PKA-stimulated channels (middle
trace). Data are representative of 6 experiments. C:
PKA-activated single-channel Cl currents (top
trace) were inhibited by DPC (500 µM, bottom trace). Data
are representative of 6 experiments.
|
|
Effect of Exogenous ABP-280 on CFTR Activation in ABP()
Cells
Addition of filamin (30 nM) to the cytoplasmic side of excised patches
from ABP()/CFTR(+) cells in the absence of PKA and ATP induced
ion channel activity similar to that observed in ABP(+)/CFTR(+) cells
in four of five experiments (Fig. 7). These
data are in agreement with the whole cell currents and further support
the requirement of cross-linked actin networks for proper activation of
CFTR.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 7.
Effect of filamin on CFTR activation in ABP()/CFTR(+) cells.
Addition of filamin (30 nM) alone to excised patches from
ABP()/CFTR(+) cells stimulated Cl channel
activity (compare top and bottom traces) in a manner
similar to the PKA-stimulated ion channel activity of ABP(+)/CFTR(+)
cells and previously reported Cl channel activity
(27). Channels also activated in clusters as in Fig. 5D. Traces
are representative of 4 experiments.
|
|
| |
DISCUSSION |
The actin cytoskeleton plays a functional role in the cAMP response of
epithelial cells (12, 26). The vasopressin-mediated increase in
Na+ channel activity of A6 epithelial cells, for example,
is mediated by the cAMP activation of PKA, which in turn modulates
Na+ channel activity (25). This response was prevented by
disruption of actin filament organization but was reestablished by
addition of exogenous actin (26).
cAMP stimulation is also a paradigm for the activation of CFTR in
epithelial tissues. However, despite the fact that CFTR contains
multiple sites for phosphorylation by PKA, little is known about the
role of PKA-mediated phosphorylation on the CFTR activation process
itself. Recent studies from our laboratory indicated that a proper
cAMP-dependent activation of CFTR requires an organized actin
cytoskeleton (27). In those studies, it was demonstrated that
cytoskeletal disruption with cytochalasin D blunted completely the
cAMP-mediated activation of CFTR, and, only in the presence of an
organized actin cytoskeleton does PKA induce CFTR-associated ion
channel activity. However, actin-cytoskeleton-modifying agents,
including cytochalasin D and phalloidin, do not diminish the total pool
of actin but instead modify the structural arrangement of
three-dimensional actin networks. Therefore, in this report it was
postulated that specific changes in actin network conformations may
also modify CFTR activation. The present study was an attempt to
initiate a characterization of the role of three-dimensional actin
structures on the functional interaction between the cAMP/PKA pathway
and the CFTR activation process. The data in this report indicate that
not only actin is necessary for a proper cAMP-mediated activation of
CFTR but that proteins conveying specific three-dimensional actin
network conformations are also necessary for this response to be
properly accomplished. Although the molecular mechanism for this
functional interaction between CFTR and the actin cytoskeleton is still
largely unknown, it is clear that actin and actin-binding proteins are
required in this process. Thus the actin filamental activation of CFTR
may require the presence of vicinal scafolding proteins to elicit the
activation process. One possibility may entail the role of actin
networks in helping either PKA and/or other adjacent proteins interact
with the ion channel. Recent studies have suggested, for example, that
the actin-binding protein ezrin may interact with CFTR via the
postsynaptic density disc-large ZO-1 domains of
ERM-binding phosphoprotein 50 (EBP50), through the CFTR-conserved
sequence DTRL in the COOH terminus of the channel protein (32). This is
in agreement with previous studies indicating that another ion
transport protein, the Na+/H+ exchanger, is
associated with the EBP50 homologue NHE-RF, which confers cAMP
sensitivity to the transporter (19, 35). In this context, it is
possible that actin-binding proteins may actually help dissociate the
PKA catalytic subunits from the anchored regulartory units of the
inactive complex, thus conveying a compartmentalized dimension to the
PKA activation process. Several reports have already established a
regulatory role of the actin cytoskeleton in the dissociation of
regulatory and catalytic subunits of PKA I (23) and II isoforms (13).
Nevertheless, the cytoskeletal regulation of CFTR may be also
independent of PKA activation. Previous evidence on the activation of
cardiac CFTR by the anti-COOH antibody raised against the CTRL sequence
of the channel would suggest that this process actually requires actin
cytoskeletal integrity but is elicited in the absence of PKA activation
(4). However, several CFTR reconstitution studies have previously
determined that, in the likely absence of actin, PKA is still capable
of inducing conformational changes to elicit a functional CFTR. Future studies are required to assess the nature of the modulation by both PKA
and the actin cytoskeleton, which may work in concert to elicit a
functional CFTR.
The present study focused on the ability of cross-linked actin networks
to elicit a proper cAMP response driving a functional CFTR. Filamin and
its homologue actin-binding protein (ABP-280) are homodimeric proteins
with a molecular mass of ~540 kDa and are known to cross-link actin
filaments (16, 31), generating three-dimensional F-actin networks that
behave as intracellular gels (34). ABP-280 also links the actin
cytoskeleton to the plasma membrane, thus enabling a functional
interaction with plasma membrane structures including receptors (22,
24). Although the ability of these proteins to cross-link actin relies
on dimeric binding to more than one actin filament, it is the angle
between the actin filaments that may be relevant for their final
conformation. The 
-actinin homodimers, for example, tightly bind
actin into bundles instead of orthogonal cross-linked actin filaments
formed by filamin. Interestingly, previous studies from our laboratory have shown that, while filamin inhibits (9), -actinin activates (5)
epithelial Na+ channel activity, thus suggesting that the
spatial arrangement of cross-linked filaments is also relevant in the
regulation of a particular ion channel response.
The presence of apically added filamin/ABP-280 has been observed to be
a tonic inhibitor of epithelial ion channel activity. Filamin was
previously shown to inhibit spontaneous (9) as well as PKA-activated
(26) and actin-activated (9) Na+ channels in epithelial
cells. Further, ABP-280 and filamin inhibited spontaneous
K+ channel activity in human melanoma cells (7) and the
PKA-mediated activation of CFTR (27). In agreement with these findings,
cells lacking a functional ABP-280 are unable to volume regulate, due to a dysfunctional and constitutive K+ channel activation
(7). Genetically rescued melanoma cells transfected with the ABP-280
cDNA, in contrast, have a lower basal K+ permeability and
recover the ability to elicit cell volume regulation (7). Therefore, in
the present study the effect of ABP-280 on CFTR function was evaluated
in ABP() and ABP(+) melanoma cell lines transfected with CFTR.
cAMP only activated CFTR in the presence of a functional ABP-280. In
close agreement with previous studies on the effect of cytochalasin D,
however, we found that cAMP and PKA did not induce CFTR ion channel
activity in the ABP()/CFTR(+) cells. This in itself confirmed
that the three-dimensional nature of the actin cytoskeleton is
essential for proper regulation of ion channel activity. This
phenomenon was further confirmed both by transfection of ABP-280 in
CFTR-expressing ABP() cells and by intracellular dialysis of
ABP-280 in ABP()/CFTR(+) cells. The data in this report
strengthen our previous studies, indicating that actin filament
organization is a key component for a proper PKA response in vivo,
since cells whose actin cytoskeleton was collapsed by a 6- to 9-h
exposure to cytochalasin D (27) or by the lack of ABP-280 (this report)
were insensitive to cAMP stimulation under whole cell conditions and
most clearly to direct addition of PKA under excised conditions.
The present data are thus most consistent with a particular structural
conformation of actin in the proper PKA-dependent regulatory mechanism
of CFTR. It is important, however, to indicate that all three relevant
proteins involved in this interface, namely, CFTR, actin, and ABP-280,
are substrates for phosphorylation. Further studies, such as specific
mutations of the proteins involved, will be required, therefore, to
assess the specific molecular steps of this functional interface.
| |
ACKNOWLEDGEMENTS |
We thank Dr. Thomas P. Stossel (Experimental Medicine,
Hematology-Oncology Division, Brigham and Women's Hospital, Boston, MA) for his thorough review of the original manuscript. We are also
grateful to Dr. John Hartwig (Experimental Medicine,
Hematology-Oncology Division, Brigham and Women's Hospital, Boston,
MA) for providing ABP-280.
| |
FOOTNOTES |
These studies were supported in part by National Institute of Diabetes
and Digestive and Kidney Diseases Grant DK-48040 (H. F. Cantiello).
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests and other correspondence: H. F. Cantiello, Renal Unit, Massachusetts General Hospital East, 149 13th
St., Charlestown, MA 02129 (E-mail: cantiello{at}helix.mgh.harvard.edu).
Received 13 August 1998; accepted in final form 22 July 1999.
| |
REFERENCES |
1.
Anderson, M. P.,
D. P. Rich,
R. J. Gregory,
A. E. Smith,
and
M. J. Welsh.
Generation of cAMP-activated chloride current by expression of CFTR.
Science
251:
679-682,
1991[Abstract/Free Full Text].
2.
Berger, H. A.,
M. P. Anderson,
R. J. Gregory,
S. Thompson,
P. W. Howard,
R. A. Maurer,
R. Mulligan,
A. E. Smith,
and
M. J. Welsh.
Identification and regulation of the cystic fibrosis transmembrane conductance regulator-generated chloride channel.
J. Clin. Invest.
88:
1422-1431,
1991.
3.
Cantiello, H. F.
Actin filaments stimulate Na+-K+-ATPase.
Am. J. Physiol.
269 (Renal Fluid Electrolyte Physiol. 38):
F637-F643,
1995[Abstract/Free Full Text].
4.
Cantiello, H. F.
Role of the actin cytoskeleton in the regulation of the cystic fibrosis transmembrane conductance regulator.
Exp. Physiol.
83:
505-514,
1996.
5.
Cantiello, H. F.
Role of the actin cytoskeleton on epithelial Na+ channel regulation.
Kidney Int.
48:
970-984,
1995[Medline].
6.
Cantiello, H. F.,
G. R. Jackson, Jr.,
C. F. Grosman,
A. G. Prat,
S. C. Borkan,
Y.-H. Wang,
I. L. Reisin,
C. R. O'Riordan,
and
D. A. Ausiello.
Electrodiffusional ATP movement through the cystic fibrosis transmembrane conductance regulator.
Am. J. Physiol.
274 (Cell Physiol. 43):
C799-C809,
1998[Abstract/Free Full Text].
7.
Cantiello, H. F.,
A. G. Prat,
J. V. Bonventre,
C. C. Cunningham,
J. Hartwig,
and
D. A. Ausiello.
Actin-binding protein contributes to cell volume regulatory ion channel activation in melanoma cells.
J. Biol. Chem.
268:
4596-4599,
1993[Abstract/Free Full Text].
8.
Cantiello, H. F.,
A. G. Prat,
I. L. Reisin,
E. H. Abraham,
L. B. Ercole,
J. F. Amara,
R. J. Gregory,
and
D. A. Ausiello.
External ATP activates the cystic fibrosis transmembrane conductance regulator.
J. Biol. Chem.
269:
11224-11232,
1994[Abstract/Free Full Text].
9.
Cantiello, H. F.,
J. Stow,
A. G. Prat,
and
D. A. Ausiello.
Actin filaments control epithelial Na+ channel activity.
Am. J. Physiol.
261 (Cell Physiol. 30):
C882-C888,
1991[Abstract/Free Full Text].
10.
Cunningham, C. C.,
J. B. Gorlin,
D. J. Kwiatkowski,
J. H. Hartwig,
P. A. Janmey,
H. R. Byers,
and
T. P. Stossel.
Actin-binding protein requirement for cortical stability and efficient locomotion.
Science
255:
325-327,
1992[Abstract/Free Full Text].
11.
Dechecchi, M.,
A. Tamanini,
G. Berton,
and
G. Cabrini.
Protein kinase C activates chloride conductance in C127 cells stably expressing the cystic fibrosis gene.
J. Biol. Chem.
268:
11321-11325,
1993[Abstract/Free Full Text].
12.
Ding, G.,
N. Franki,
J. Condeelis,
and
R. M. Hays.
Vasopressin depolymerizes F-actin in toad bladder epithelial cells.
Am. J. Physiol.
260 (Cell Physiol. 29):
C9-C16,
1991[Abstract/Free Full Text].
13.
Feliciello, A.,
Y. Li,
E. Avvedimento,
M. Gottesman,
and
C. Rubin.
A-kinase anchor protein 75 increases the rate and magnitude of cAMP signaling to the nucleus.
Curr. Biol.
7:
1011-1014,
1997[Medline].
14.
Gariboldi, M.,
E. Maestrini,
F. Canzian,
G. Manenti,
L. De Gregorio,
S. Rivella,
A. Chatterjee,
G. Herman,
N. Archidiacono,
R. Antonacci,
M. Pierotti,
T. Dragani,
and
D. Toniolo.
Comparative mapping of the actin-binding protein 280 genes in human and mouse.
Genomics
21:
428-430,
1994[Medline].
15.
Gorlin, J.,
E. Henske,
S. Warren,
C. Kunst,
M. D'Urso,
G. Palmieri,
J. Hartwig,
G. Bruns,
and
D. Kwiatkowski.
Actin-binding protein (ABP-280) filamin gene (FNL) maps telomeric to the color vision locus (R/GCP) and centromeric to G6PD in Xq28.
Genomics
17:
496-498,
1993[Medline].
16.
Hartwig, J.,
and
D. Kwiatkowski.
Actin-binding proteins.
Curr. Opin. Cell Biol.
3:
87-97,
1991[Medline].
17.
Hartwig, J.,
and
T. P. Stossel.
Isolation and properties of actin, myosin, and a new actin binding protein in rabbit alveolar macrophages.
J. Biol. Chem.
250:
5696-5705,
1975[Abstract/Free Full Text].
18.
Hille, B.
Ionic Channels of Excitable Membranes. Sunderland, MA: Sinauer, 1992.
19.
Lamprecht, G.,
E. Weinman,
and
C. Yun.
The role of NHERF and E3KARP in the cAMP-mediated inhibition of NHE3.
J. Biol. Chem.
273:
29972-29978,
1998[Abstract/Free Full Text].
20.
Matsudaira, P.
Modular organization of actin crosslinking proteins.
Trends Biol. Sci.
16:
87-92,
1991.
21.
McCarty, N. A.,
S. McDonough,
B. N. Cohen,
J. R. Riordan,
N. Davidson,
and
H. A. Lester.
Voltage-dependent block of the cystic fibrosis transmembrane conductance regulator Cl channel by two closely related arylaminobenzoates.
J. Gen. Physiol.
102:
1-23,
1993[Abstract/Free Full Text].
22.
Meyer, S.,
S. Zuerbig,
C. Cunningham,
J. Hartwig,
T. Bissell,
K. Gardner,
and
J. Fox.
Identification of the region in actin-binding protein that binds to the cytoplasmic domain of glycoprotein Ib
.
J. Biol. Chem.
272:
2914-2919,
1997[Abstract/Free Full Text].
23.
Moos, J.,
J. Peknicova,
G. Geussova,
V. Philimonenko,
and
P. Hozak.
Association of protein kinase A type I with detergent-resistant structures of mammalian sperm cells.
Mol. Reprod. Dev.
50:
79-85,
1998[Medline].
24.
Ohta, Y.,
T. P. Stossel,
and
J. H. Hartwig.
Ligand-sensitive binding of actin-binding protein to immunoglobulin G Fc receptor I (Fc
RI).
Cell
67:
1-20,
1991[Medline].
25.
Prat, A. G.,
D. A. Ausiello,
and
H. F. Cantiello.
Vasopressin and protein kinase A activate G protein-sensitive Na channels.
Am. J. Physiol.
265 (Cell Physiol. 34):
C218-C223,
1993[Abstract/Free Full Text].
26.
Prat, A. G.,
A. M. Bertorello,
D. A. Ausiello,
and
H. F. Cantiello.
Activation of epithelial Na+ channels by protein kinase A requires actin filaments.
Am. J. Physiol.
265 (Cell Physiol. 34):
C224-C233,
1993[Abstract/Free Full Text].
27.
Prat, A. G.,
Y.-F. Xiao,
D. A. Ausiello,
and
H. F. Cantiello.
cAMP-independent regulation of CFTR by the actin cytoskeleton.
Am. J. Physiol.
268 (Cell Physiol. 37):
C1552-C1561,
1995[Abstract/Free Full Text].
28.
Reisin, I. L.,
A. Prat,
E. H. Abraham,
J. F. Amara,
R. J. Gregory,
D. A. Ausiello,
and
H. F. Cantiello.
The cystic fibrosis transmembrane conductance regulator (CFTR) is a dual ATP and chloride channel.
J. Biol. Chem.
269:
20584-20591,
1994[Abstract/Free Full Text].
29.
Rich, D. P.,
M. P. Anderson,
R. I. Gregory,
S. H. Cheng,
S. Paul,
D. N. Jefferson,
I. D. McCann,
K. W. Klinger,
A. E. Smith,
and
M. J. Welsh.
Expression of cystic fibrosis transmembrane conductance regulator corrects defective chloride channel regulation in cystic fibrosis airway epithelial cells.
Nature
347:
358-363,
1990[Medline].
30.
Schwiebert, E.,
N. Kizer,
D. Gruenert,
and
B. Stanton.
GTP-binding proteins inhibit cAMP activation of chloride channels in cystic fibrosis airway epithelial cells.
Proc. Natl. Acad. Sci. USA
89:
10623-10627,
1992[Abstract/Free Full Text].
31.
Shizuta, Y.,
H. Shizuta,
M. Gallo,
P. Davies,
and
I. Pastan.
Purification and properties of filamin, an actin binding protein from chicken gizzard.
J. Biol. Chem.
251:
6562-6567,
1976[Abstract/Free Full Text].
32.
Short, D.,
K. Trotter,
D. Reczek,
S. Kreda,
A. Bretscher,
R. Boucher,
M. Jackson Stutts,
and
S. Milgram.
An apical PDZ protein anchors the cystic fibrosis transmembrane conductance regulator to the cytoskeleton.
J. Biol. Chem.
273:
19797-19801,
1998[Abstract/Free Full Text].
33.
Tabcharani, J.,
P. Lindsell,
and
J. Hanrahan.
Halide permeation in wild-type and mutant cystic fibrosis transmembrane conductance regulator chloride channels.
J. Gen. Physiol.
110:
341-354,
1997[Abstract/Free Full Text].
34.
Weihing, R. R.
The filamins: properties and functions.
Can. J. Biochem. Cell Biol.
63:
397-413,
1985[Medline].
35.
Yun, C.,
G. Lamprecht,
D. Forster,
and
A. Sidor.
NHE3 kinase A regulatory protein E3KARP binds the epithelial brush border Na+/H+ exchanger NHE3 and the cytoskeletal protein ezrin.
J. Biol. Chem.
273:
25856-25863,
1998[Abstract/Free Full Text].
Am J Physiol Cell Physiol 277(6):C1160-C1169
0002-9513/99 $5.00
Copyright © 1999 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
![<IT>P</IT><SUB>ATP</SUB>/<IT>P</IT><SUB>Cl</SUB> = {<IT>z</IT><SUP>2</SUP><SUB>Cl</SUB> × Cl<SUB>o</SUB> × [1 − exp(<IT>az</IT><SUB>A</SUB><IT>V</IT><SUB>r</SUB>)]}](/content/vol277/issue6/fulltext/C1160/img001.gif)
![/{<IT>z</IT><SUP>2</SUP><SUB>A</SUB> × A<SUB>i</SUB> × exp(<IT>az</IT><SUB>A</SUB><IT>V</IT><SUB>r</SUB>) × [1 − exp(<IT>az</IT><SUB>Cl</SUB><IT>V</IT><SUB>r</SUB>)]}](/content/vol277/issue6/fulltext/C1160/img002.gif)
![<IT>P</IT><SUB>Cl</SUB>/<IT>P</IT><SUB>Y</SUB> = {<IT>z</IT><SUP>2</SUP><SUB>Y</SUB> × Y<SUB>o</SUB> × [1 − exp(<IT>az</IT><SUB>Cl</SUB><IT>U</IT>)]}](/content/vol277/issue6/fulltext/C1160/img003.gif)
![/{<IT>z</IT><SUP>2</SUP><SUB>Cl</SUB> × Cl<SUB>i</SUB> × exp(<IT>az</IT><SUB>Cl</SUB><IT>U</IT>) × [1 − exp(<IT>az</IT><SUB>Y</SUB><IT>U</IT>)]}](/content/vol277/issue6/fulltext/C1160/img004.gif)
