Articles in PresS. Am J Physiol Renal Physiol (January 28, 2015). doi:10.1152/ajprenal.00376.2014
MS#: F-00376-2014R1, Revised Manuscript 1
MS#: F-00376-2014R1, Revised Manuscript
2 3
Am J Physiol Renal Physiol
20150111
4 5 6 7
Extracellular pH affects phosphorylation and intracellular trafficking of AQP2 in inner medullary collecting duct cells
8 9 10
Hyo-Jung Choi1,2, Hyun Jun Jung1, and Tae-Hwan Kwon1,2
11 12 13 14 15
1
Department of Biochemistry and Cell Biology, 2BK21 Plus KNU Biomedical Convergence Program, Department of Biomedical Science, School of Medicine, Kyungpook National University, Taegu, Korea
16 17 18 19 20
Running title: Extracellular pH affects dDAVP-induced AQP2 trafficking
21 22 23 24 25 26 27 28 29
Correspondence: Tae-Hwan Kwon, MD, PhD, Department of Biochemistry and Cell Biology, School of Medicine, Kyungpook National University, Dongin-dong 101, Taegu 700-422, South Korea. Tel: +82 53 420 4825; Fax: +82 53 422 1466. E-mail: [email protected]
30 31 32 33 34 35 36 37 38 1
Copyright © 2015 by the American Physiological Society.
MS#: F-00376-2014R1, Revised Manuscript
39
Abstract
40 41 42
Kidney collecting duct cells are continuously exposed to the changes of extracellular pH
43
(pHe). We aimed to study the effects of altered pHe on dDAVP-induced phosphorylation
44
(S256, S261, S264, and S269) and apical targeting of AQP2 in rat kidney inner medullary
45
collecting duct (IMCD) cells. When freshly prepared IMCD tubule suspensions exposed to
46
HEPES buffer with pH 5.4, 6.4, 7.4, or 8.4 for 1 h were stimulated with dDAVP (10-10 M, 3
47
min), AQP2 phosphorylation at S256, S264, and S269 was significantly attenuated in acidic
48
conditions. Next, IMCD cells primary cultured in the transwell chambers were exposed to
49
transepithelial pH gradient for 1 h (apical pH 6.4, 7.4 or 8.4 vs. basolateral pH 7.4 and vice
50
versa). Immunocytochemistry and cell surface biotinylation assay revealed that exposure to
51
either apical pH 6.4 or basolateral pH 6.4 for 1 h was associated with decreased dDAVP (10-9
52
M, 15 min, basolateral)-induced apical targeting of AQP2 and surface expression of AQP2.
53
Fluorescence resonance energy transfer analysis revealed that dDAVP (10-9 M)-induced
54
increase of PKA activity was significantly attenuated when LLC-PK1 cells were exposed to
55
pHe 6.4, compared to pHe 7.4 and 8.4. In contrast, forskolin (10-7 M)-induced PKA activation
56
and dDAVP (10-9 M)-induced increase of intracellular Ca2+ were not affected. Taken together,
57
dDAVP-induced phosphorylation and apical targeting of AQP2 are attenuated in IMCD cells
58
under acidic pHe, likely via an inhibition of V2R-G-protein-cAMP-PKA action.
59 60
Key words: Acidosis, Collecting duct, FRET, Vasopressin, V2R receptor.
61 62 63 64 2
MS#: F-00376-2014R1, Revised Manuscript
65
Introduction
66 67
Renal tubular epithelial cells, including the kidney collecting duct cells, are
68
continuously exposed to the changes of microenvironment, e.g., luminal fluid shear stress,
69
transepithelial osmotic gradient, and changes of luminal or interstitial pH. We previously
70
demonstrated the effects of altered luminal fluid shear stress (FSS) on the translocation of
71
aquaporin-2 (AQP2) and the reorganization of actin cytoskeleton (F-actin) in the kidney inner
72
medullary collecting duct (IMCD) cells (14). Immunocytochemistry demonstrated that
73
exposure to luminal FSS per se in the microfluidic chamber induced depolymerization of F-
74
actin associated with increased apical targeting of AQP2, even in the absence of vasopressin
75
stimulation. This finding suggests that changes of microenvironment, including extracellular
76
pH, could be importantly involved in the apical targeting of AQP2.
77
Aquaporins (AQPs) are a family of membrane proteins that transport water molecules
78
through biomembranes (18, 21, 31). AQP2 is the water-selective channel protein that plays a
79
key role in arginine vasopressin (AVP)-mediated water reabsorption in the collecting ducts (2,
80
20, 21, 27, 30, 31, 41). Vasopressin V2-receptor (V2R)-mediated cAMP/PKA signaling has
81
been shown to be a principal signaling pathway for both AQP2 trafficking and protein
82
expression via activation of Gsα-mediated adenylyl cyclase activity (7, 16, 30, 41, 42).
83
However, despite the well-known effects of AVP stimulation, the concurrent effects of
84
changes in luminal or interstitial microenvironment in vivo, e.g., pH, on the phosphorylation
85
and apical targeting of AQP2 in the collecting duct cells, are not well understood. Urine pH
86
can vary between 4.6 and 8.0, depending on the amount of urinary net acid excretion. A
87
recent clinical study demonstrated that almost 60% of patients with stones had a urine pH less
88
than 6, a figure consistent with that found in the general population (3). It has been
89
demonstrated that the affinity of vasopressin for V2R was significantly lower under the 3
MS#: F-00376-2014R1, Revised Manuscript 90
condition of acidic pH, compared to that at neutral pH in LLC-FLAG-V2R cells (44).
91
Moreover, rats with metabolic acidosis were associated with significantly decreased urinary
92
excretion of AQP2, whereas renal AQP2 mRNA and protein expressions were increased (28).
93
In addition, it has been demonstrated that urine alkalinization promoted the excretion of
94
urinary exosomal AQP2 independent of vasopressin stimulation in rats (9).
95
Therefore, we hypothesized that changes in the extracellular pH (pHe) could affect the
96
intracellular pH (pHi) of the collecting duct cells, resulting in altered response of AQP2
97
phosphorylation (S256, S261, S264, S269) and apical targeting of AQP2 to vasopressin
98
stimulation. Firstly, dDAVP-induced AQP2 phosphorylation (S256) was examined in IMCD
99
tubule suspensions exposed to pH 5.4, pH 6.4, pH 7.4 and pH 8.4, respectively. To study the
100
selective effects of apical or basolateral pH on the dDAVP-induced intracellular trafficking of
101
AQP2 in the collecting duct cells, IMCD cells of rat kidneys were primary cultured in the
102
transwell chamber system. The effects of acidic (pH 6.4), neutral (pH 7.4), or alkaline (pH
103
8.4) conditions in the apical or basolateral side of the cells were examined on the dDAVP-
104
induced intracellular trafficking and cell surface expression of AQP2. Moreover, for further
105
elucidating the underlying signaling mechanisms, fluorescence resonance energy transfer
106
(FRET) analysis was exploited and time-lapse changes of AVP- or forskolin-induced PKA
107
activity and intracellular Ca2+ dynamics were monitored in LLC-PK1 cells exposed to acidic
108
(pH 6.4), neutral (pH 7.4), or alkaline (pH 8.4) microenvironment.
109 110 111 112 113 114 4
MS#: F-00376-2014R1, Revised Manuscript
115
Materials and Methods
116 117
Primary culture of IMCD cells of rat kidney. The animal protocols were approved by the
118
Animal Care and Use Committee of the Kyungpook National University, Korea (KNU 2012-
119
10). Primary cultures enriched in IMCD cells were prepared from pathogen-free male
120
Sprague-Dawley rats (200 - 250 g, Charles River, Seongnam, Korea), as previously described
121
in detail (6, 23, 24). Briefly, rats were anesthetized under enflurane inhalation, and kidneys
122
were rapidly removed. After isolating IMCD cell suspension, cells were seeded either on the
123
collagen-coated 35-mm glass-base dish (Asahi Techno Glass, Tokyo, Japan) for intracellular
124
pH measurement or on the semi-permeable filters of transwell system (0.4 µm pore size,
125
Transwell® Permeable Supports, catalog no. 3450, Corning) for cell surface biotinylation
126
assay. IMCD cells were fed every 48 h and were grown in hypertonic culture medium (640
127
mOsm/KgH2O) supplemented with 10% fetal bovine serum at 37C in 5% CO2, 95% air
128
atmosphere for 3 days, and then in fetal bovine serum-free culture medium for additional 1
129
day before the experiment at day 5. The culture medium was Dulbecco's modified Eagle's
130
medium/F12 without phenol red, containing 80 mM urea, 130 mM NaCl, 10 mM HEPES, 2
131
mM L-glutamine, penicillin/streptomycin 10,000 units/ml, 50 nM hydrocortisone, 5 pM
132
3,3,5-triiodo-thyronine, 1 nM sodium selenate, 5 mg/L transferrin, and 10% fetal bovine
133
serum (pH 7.4, 640 mOsm/KgH2O).
134 135
IMCD tubule suspensions. Fresh inner medullary collecting duct (IMCD) tubules were
136
prepared from rat kidneys, as previously described (5, 35). Rats were anesthetized under
137
enflurane inhalation. Kidney inner medulla was dissected, minced, and digested in IMCD
138
suspension buffer (250 mM sucrose, 10 mM triethanolamine, pH 7.4) containing collagenase
139
B (3 mg/ml) and hyaluronidase (2 mg/ml). Then, IMCD suspension was continuously 5
MS#: F-00376-2014R1, Revised Manuscript 140
agitated in a warm water bath at 37°C for 90 min. Thereafter, IMCD tubule suspension was
141
centrifuged at 60 g for 30 sec to separate the IMCD-enriched fraction (pellet) and the non-
142
IMCD fraction (supernatant). IMCD-enriched fraction was collected and washed twice with
143
ice-cold IMCD suspension buffer. For the experiments, IMCD tubule suspensions were
144
exposed to the HEPES-buffered saline solution (162.5 mM NaCl, 2.5 mM Na2HPO4, 4 mM
145
KCl, 25 mM HEPES, 2 mM CaCl2, 1.2 mM MgSO4, and 5.5 mM glucose) with different pH
146
(pH 5.4, 6.4, 7.4 or 8.4) for 1 h, which was established by adding HCl (for pH 5.4) or NaOH
147
(for pH 6.4, 7.4, and 8.4) (43). dDAVP stimulation (10-10 M) was done during the last 3 min
148
(Fig. 1A).
149 150
Semiquantitative immunoblotting. SDS-PAGE was performed on 12% polyacrylamide gels,
151
as previously described (34). Primary antibodies used were anti-AQP2 (H7661AP, 1:2,000)
152
(15, 23), anti-phosphorylated AQP2 at S256 (K0307AP, 1:3,000) (34), anti-phosphorylated
153
AQP2 at S261, S264 and S269 (Phosphosolutions, 1:1,000), and anti-β-actin (Sigma,
154
1:100,000).
155 156
Immunofluorescence microscopy of primary cultured IMCD cells. IMCD cells were grown
157
to confluence in each transwell chamber (0.4 µm pore size, Transwell® Permeable Supports,
158
catalog no. 3460, Corning) for 4 days. At day 5, IMCD cells were subjected to transepithelial
159
pH gradient for 1 h (pH 6.4, 7.4 or 8.4 at the apical side vs. pH 7.4 at the basolateral side and
160
vise versa) at 37°C. dDAVP stimulation (10-9 M, 15 min, basolateral side only) was done
161
during the last 15 min and then fixed with 3% paraformaldehyde in PBS, pH 7.4 for 20 min at
162
room temperature. After fixation, cells were washed twice in PBS and permeabilized with
163
0.3% Triton X-100 in PBS at room temperature for 15 min. Cells were washed, the labeled
164
with anti-AQP2 (H7661AP, 1:200) antibody and nucleus was stained with DAPI. The AQP2 6
MS#: F-00376-2014R1, Revised Manuscript 165
localization was determined using laser confocal microscopy (Zeiss LSM 5 EXCITER, Jena,
166
Germany), and an X63 (NA 1.4) objective lens (Zeiss) was used. Digital images were
167
collected and analyzed using the Zeiss Aim Image Examiner program.
168 169
Cell surface biotinylation assays. IMCD cells were primary cultured for 4 days on semi-
170
permeable filters of transwell system (0.4 µm pore size, catalog no. 3450, Transwell®
171
Permeable Supports, Corning Incorporated, MA). At day 5, they were exposed to
172
transepithelial pH gradient for 1 h (apical pH 6.4, 7.4 or 8.4 vs. basolateral pH 7.4 and vice
173
versa) at 37°C. Then, the cells were washed three times with ice-cold PBS-CM (10 mM PBS
174
containing 1 mM CaCl2 and 0.1 mM MgCl2, pH 7.5) and incubated at 4˚C for 45 min in ice-
175
cold biotinylation buffer (10 mM triethanolamine, 2 mM CaCl2, 125 mM NaCl, pH 8.9)
176
containing 1 mg/ml sulfosuccinimidyl 2-(biotinamido)-ethyl-1,3-dithiopropionate (Sulfo-
177
NHS-SS-biotin, Thermo Scientific, Rockford, IL) on the apical side. Cells were washed once
178
with quenching buffer (50 mM Tris-HCl in PBS-CM, pH 8.0) followed by washing twice
179
with PBS-CM. Lysis buffer (1% triton X-100, 150 mM NaCl, 5 mM EDTA, 50 mM Tris-HCl,
180
pH 7.5, 1.15 mM PMSF, 0.4 μg/ml leupeptin, 0.1 mg/ml pefabloc, 0.1 μM okadaic acid, 1
181
mM Na3VO4, 25 mM NaF) was added and the lysates were sonicated 2 × 6 pluses at 20% of
182
amplitude. The lysates were centrifuged at 10,000 g for 5 min at 4˚C. The supernatant was
183
transferred to columns where Neutravidin agarose resin (200 μl) was previously loaded
184
(Thermo Scientific, Rockford, IL) and incubated for 60 min at RT with end-over-end mixing.
185
After 4 times washing with PBS containing protease inhibitors, 1 × sample buffer containing
186
DTT was added to the column and incubated for 60 min at RT. Samples were heated at 65˚C
187
for 10 min.
188 189
Intracellular pH measurement. Intracellular pH (pHi) was measured using the pH-sensitive 7
MS#: F-00376-2014R1, Revised Manuscript 190
fluorescent
191
(BCECF-AM, Molecular Probes). BCECF-AM is converted to the membrane-impermeant
192
fluorescent dye BCECF by intracellular esterases. Primary cultured IMCD cells of rat kidney
193
were seeded and cultured on the collagen-coated 35-mm glass-base dish (Asahi Techno Glass,
194
Tokyo, Japan) in hypertonic culture media (pH 7.4 and 640 mOsm/kgH2O) for 4 days. On
195
day 5, cells were incubated with non-serum IMCD culture media (Dulbecco's modified
196
Eagle's medium/F12 without phenol red, 80 mM urea, 130 mM NaCl, penicillin/streptomycin
197
10,000 units/ml) containing 2 µM BCECF-AM at 37C in 5% CO2-95% air for 30 min. The
198
cells were washed twice in PBS and changed to non-serum IMCD culture media (pH 7.4) for
199
10 min at 37C incubator before the experiments. IMCD cells were mounted in the chamber
200
on the temperature-controlled stage of Nikon Ti-E inverted microscope (Nikon, Tokyo,
201
Japan). The cells were continuously perfused with a Na+-free solution with different pH
202
values (pH 6.4, 7.4, and 8.4, see below) for 60 min by six-channel system (VC-2, Warner
203
Instruments, Hamden, CT) and perfusate was collected by the peristaltic pump (EYELA MP3,
204
Tokyo, Japan).
dye,
(2,7)-biscarboxyethyl-5(6)-carboxyfluorescein
acetoxymethyl
ester
205
The pHi was measured using the dual-excitation fluorescence ratio method. The
206
fluorescence was monitored at 530 nm with excitation wavelengths of 440 nm (a wavelength
207
at which the fluorescence is relatively pH insensitive) and 490 nm (at which the fluorescence
208
is very pH sensitive). The pHi was time-lapse obtained by the Nikon Ti-E inverted
209
microscope (Nikon, Tokyo, Japan) equipped with Cascade 512B (EMCCD) camera (Roper
210
Scientific, Trenton, NJ) and excitation and emission filter wheels (MAC5000, Ludl Electronic
211
Products, Hawthorne, NY). All systems were controlled by MetaMorph software (Universal
212
Imaging, Downingtown, PA). Fluorescent images were acquired sequentially through CFP
213
filter channels. Images were acquired by using the 4 x 4 binning mode and 200-ms exposure
214
time. A X60 (NA 1.4, Nikon, Tokyo, Japan) objective lens was used for the measurements. 8
MS#: F-00376-2014R1, Revised Manuscript 215
The 490/440 nm ratio was created by MetaMorph software. The data were obtained from five
216
independent experiments in both groups, respectively.
217
The pHi was calculated by the nigericin calibration technique (3). IMCD cells were
218
exposed to a Na+-free solution (0 mM Na+, 105 mM K+, 0 mM NH4+, 49 mM NMDG+, 5 mM
219
Mg2+, 0 mM Ca2+, 6.3 mM Cl-, 106.4 mM Aspartate-, 0.4 mM H2PO4-, 1.6 mM HPO42-, 17.8
220
mM HEPES-, 0 mM HCO3-, 5 mM SO42-, 10 mM EGTA2-, 5.5 mM glucose, 5 mM alanine,
221
14.4 mM HEPES) with different pH values (6.4, 7.4, and 8.4) by adding KOH, containing 10
222
µM of the K+-H+ exchanger nigericin. When the [K+] was equal between the intracellular and
223
extracellular compartments, pHi and pHe should be the same with nigericin (490/440 nm
224
ratio = 1.0, pH = 7.0) (3). Therefore the data were normalized to make the ratio at pH 7.0
225
equal to unity. Raw data of intensity at each excitation wavelength were corrected for
226
background-corrected fluorescence intensity prior to calculation of the ratios. The ratios of
227
background-corrected emission intensities (490/440 nm ratio) were transformed into pH
228
values as pH = log [(Rn-Rn(min))/(Rn(max)-Rn)] + pKa, where Rn was the background-
229
substracted ratio normalized to pH 7.0, Rn(min) and Rn(max) are the minimum and maximum
230
obtainable values for the normalized ratio and pKa is the –log of the dissociation constant of
231
the fluorophore (1, 3).
232 233
Fluorescence resonance energy transfer imaging for PKA activity. LLC-PK1 cells stably
234
transfected with AKAR3EV (PKA fluorescence resonance energy transfer (FRET)-based
235
indicator) were cultured on a collagen-coated 35-mm glass-base dish (Asahi Techno Glass,
236
Tokyo, Japan). The plasmid for PKA FRET-based indicator (AKAR3EV) was kindly
237
provided by Prof. M. Matsuda (Kyoto University, Kyoto, Japan). AKAR3EV was composed
238
of the FHA1 (ligand domain), yellow fluorescent protein (YFP), sensor domain of PKA
239
(LRRATLVD), and cyan fluorescent protein (CFP) (17). In this probe design, the sensor 9
MS#: F-00376-2014R1, Revised Manuscript 240
domain changes its conformation upon perception of the PKA signal (17). This sensitized
241
sensor domain interacts with the ligand domain, thereby inducing a global change of the
242
biosensor conformation and concomitant increase of FRET efficiency from the donor to the
243
acceptor. With this AKAR3EV probe, the activity of PKA was time-lapse obtained by Nikon
244
Ti-E inverted microscope (Nikon, Tokyo, Japan) equipped with a Cascade 512B (EMCCD)
245
camera (Roper Scientific, Trenton, NJ) and excitation and emission filter wheels (MAC5000,
246
Ludl Electronic Products, Hawthorne, NY). All systems were controlled by MetaMorph
247
software (Universal Imaging, Downingtown, PA). Fluorescent images were acquired every 15
248
sec sequentially through CFP and FRET filter channels. Images were acquired by using the 4
249
x 4 binning mode and 200-ms exposure time. The FRET/CFP ratio was created with
250
MetaMorph software, and the value was normalized to 1 at the starting time point in both
251
groups. The data were obtained from five independent experiments in both groups,
252
respectively.
253 254
Fluorescence resonance energy transfer imaging for intracellular calcium dynamics. Two
255
days after the transient transfection of YC3.60 (calcium FRET-based indicator) (29), LLC-
256
PK1 cells in Dulbecco’s Modified Eagle Medium (phenol red-free) were subjected to time-
257
lapse imaging. LLC-PK1 cells transiently transfected with YC3.60 were cultured on a
258
collagen-coated 35-mm glass-base dish (Asahi Techno Glass, Tokyo, Japan). The plasmid for
259
calcium FRET-based indicator (YC3.60) was also provided by Prof. M. Matsuda (Kyoto
260
University, Kyoto, Japan). YC3.60 was composed of the calmodulin (CaM), enhanced yellow
261
fluorescent protein (EYFP), a glycylglycine linker, the CaM-binding peptide of myosin light-
262
chain kinase (M13), and an enhanced cyan fluorescent protein (ECFP) (29). Ca2+ ion binding
263
to CaM moiety initiates an intramolecular interaction between the CaM and the M13 domains,
264
which causing the chimeric protein to shift from a spread form to a more compact form, 10
MS#: F-00376-2014R1, Revised Manuscript 265
therefore increasing the efficiency of FRET from the CFP to the YFP (29). With this YC3.60
266
indicator, the changes of intracellular Ca2+ levels were time-lapse obtained by Nikon Ti-E
267
inverted microscope (Nikon, Tokyo, Japan). Images were acquired every 15 sec by using the
268
4 x 4 binning mode and 200-ms exposure time. The FRET/CFP ratio was created with
269
MetaMorph software, and the value was normalized to 1 at the starting time point in all
270
groups. The data were obtained from eight independent experiments in both groups,
271
respectively.
272 273
Statistical analyses. Values were presented as means ± standard errors. Data were analyzed
274
by one-way analysis of variance (ANOVA) followed by Bonferroni’s multiple comparison
275
test. Multiple comparisons tests were only applied when a significant difference was
276
determined in the ANOVA, P
MS#: F-00376-2014R1, Revised Manuscript 1
MS#: F-00376-2014R1, Revised Manuscript
2 3
Am J Physiol Renal Physiol
20150111
4 5 6 7
Extracellular pH affects phosphorylation and intracellular trafficking of AQP2 in inner medullary collecting duct cells
8 9 10
Hyo-Jung Choi1,2, Hyun Jun Jung1, and Tae-Hwan Kwon1,2
11 12 13 14 15
1
Department of Biochemistry and Cell Biology, 2BK21 Plus KNU Biomedical Convergence Program, Department of Biomedical Science, School of Medicine, Kyungpook National University, Taegu, Korea
16 17 18 19 20
Running title: Extracellular pH affects dDAVP-induced AQP2 trafficking
21 22 23 24 25 26 27 28 29
Correspondence: Tae-Hwan Kwon, MD, PhD, Department of Biochemistry and Cell Biology, School of Medicine, Kyungpook National University, Dongin-dong 101, Taegu 700-422, South Korea. Tel: +82 53 420 4825; Fax: +82 53 422 1466. E-mail: [email protected]
30 31 32 33 34 35 36 37 38 1
Copyright © 2015 by the American Physiological Society.
MS#: F-00376-2014R1, Revised Manuscript
39
Abstract
40 41 42
Kidney collecting duct cells are continuously exposed to the changes of extracellular pH
43
(pHe). We aimed to study the effects of altered pHe on dDAVP-induced phosphorylation
44
(S256, S261, S264, and S269) and apical targeting of AQP2 in rat kidney inner medullary
45
collecting duct (IMCD) cells. When freshly prepared IMCD tubule suspensions exposed to
46
HEPES buffer with pH 5.4, 6.4, 7.4, or 8.4 for 1 h were stimulated with dDAVP (10-10 M, 3
47
min), AQP2 phosphorylation at S256, S264, and S269 was significantly attenuated in acidic
48
conditions. Next, IMCD cells primary cultured in the transwell chambers were exposed to
49
transepithelial pH gradient for 1 h (apical pH 6.4, 7.4 or 8.4 vs. basolateral pH 7.4 and vice
50
versa). Immunocytochemistry and cell surface biotinylation assay revealed that exposure to
51
either apical pH 6.4 or basolateral pH 6.4 for 1 h was associated with decreased dDAVP (10-9
52
M, 15 min, basolateral)-induced apical targeting of AQP2 and surface expression of AQP2.
53
Fluorescence resonance energy transfer analysis revealed that dDAVP (10-9 M)-induced
54
increase of PKA activity was significantly attenuated when LLC-PK1 cells were exposed to
55
pHe 6.4, compared to pHe 7.4 and 8.4. In contrast, forskolin (10-7 M)-induced PKA activation
56
and dDAVP (10-9 M)-induced increase of intracellular Ca2+ were not affected. Taken together,
57
dDAVP-induced phosphorylation and apical targeting of AQP2 are attenuated in IMCD cells
58
under acidic pHe, likely via an inhibition of V2R-G-protein-cAMP-PKA action.
59 60
Key words: Acidosis, Collecting duct, FRET, Vasopressin, V2R receptor.
61 62 63 64 2
MS#: F-00376-2014R1, Revised Manuscript
65
Introduction
66 67
Renal tubular epithelial cells, including the kidney collecting duct cells, are
68
continuously exposed to the changes of microenvironment, e.g., luminal fluid shear stress,
69
transepithelial osmotic gradient, and changes of luminal or interstitial pH. We previously
70
demonstrated the effects of altered luminal fluid shear stress (FSS) on the translocation of
71
aquaporin-2 (AQP2) and the reorganization of actin cytoskeleton (F-actin) in the kidney inner
72
medullary collecting duct (IMCD) cells (14). Immunocytochemistry demonstrated that
73
exposure to luminal FSS per se in the microfluidic chamber induced depolymerization of F-
74
actin associated with increased apical targeting of AQP2, even in the absence of vasopressin
75
stimulation. This finding suggests that changes of microenvironment, including extracellular
76
pH, could be importantly involved in the apical targeting of AQP2.
77
Aquaporins (AQPs) are a family of membrane proteins that transport water molecules
78
through biomembranes (18, 21, 31). AQP2 is the water-selective channel protein that plays a
79
key role in arginine vasopressin (AVP)-mediated water reabsorption in the collecting ducts (2,
80
20, 21, 27, 30, 31, 41). Vasopressin V2-receptor (V2R)-mediated cAMP/PKA signaling has
81
been shown to be a principal signaling pathway for both AQP2 trafficking and protein
82
expression via activation of Gsα-mediated adenylyl cyclase activity (7, 16, 30, 41, 42).
83
However, despite the well-known effects of AVP stimulation, the concurrent effects of
84
changes in luminal or interstitial microenvironment in vivo, e.g., pH, on the phosphorylation
85
and apical targeting of AQP2 in the collecting duct cells, are not well understood. Urine pH
86
can vary between 4.6 and 8.0, depending on the amount of urinary net acid excretion. A
87
recent clinical study demonstrated that almost 60% of patients with stones had a urine pH less
88
than 6, a figure consistent with that found in the general population (3). It has been
89
demonstrated that the affinity of vasopressin for V2R was significantly lower under the 3
MS#: F-00376-2014R1, Revised Manuscript 90
condition of acidic pH, compared to that at neutral pH in LLC-FLAG-V2R cells (44).
91
Moreover, rats with metabolic acidosis were associated with significantly decreased urinary
92
excretion of AQP2, whereas renal AQP2 mRNA and protein expressions were increased (28).
93
In addition, it has been demonstrated that urine alkalinization promoted the excretion of
94
urinary exosomal AQP2 independent of vasopressin stimulation in rats (9).
95
Therefore, we hypothesized that changes in the extracellular pH (pHe) could affect the
96
intracellular pH (pHi) of the collecting duct cells, resulting in altered response of AQP2
97
phosphorylation (S256, S261, S264, S269) and apical targeting of AQP2 to vasopressin
98
stimulation. Firstly, dDAVP-induced AQP2 phosphorylation (S256) was examined in IMCD
99
tubule suspensions exposed to pH 5.4, pH 6.4, pH 7.4 and pH 8.4, respectively. To study the
100
selective effects of apical or basolateral pH on the dDAVP-induced intracellular trafficking of
101
AQP2 in the collecting duct cells, IMCD cells of rat kidneys were primary cultured in the
102
transwell chamber system. The effects of acidic (pH 6.4), neutral (pH 7.4), or alkaline (pH
103
8.4) conditions in the apical or basolateral side of the cells were examined on the dDAVP-
104
induced intracellular trafficking and cell surface expression of AQP2. Moreover, for further
105
elucidating the underlying signaling mechanisms, fluorescence resonance energy transfer
106
(FRET) analysis was exploited and time-lapse changes of AVP- or forskolin-induced PKA
107
activity and intracellular Ca2+ dynamics were monitored in LLC-PK1 cells exposed to acidic
108
(pH 6.4), neutral (pH 7.4), or alkaline (pH 8.4) microenvironment.
109 110 111 112 113 114 4
MS#: F-00376-2014R1, Revised Manuscript
115
Materials and Methods
116 117
Primary culture of IMCD cells of rat kidney. The animal protocols were approved by the
118
Animal Care and Use Committee of the Kyungpook National University, Korea (KNU 2012-
119
10). Primary cultures enriched in IMCD cells were prepared from pathogen-free male
120
Sprague-Dawley rats (200 - 250 g, Charles River, Seongnam, Korea), as previously described
121
in detail (6, 23, 24). Briefly, rats were anesthetized under enflurane inhalation, and kidneys
122
were rapidly removed. After isolating IMCD cell suspension, cells were seeded either on the
123
collagen-coated 35-mm glass-base dish (Asahi Techno Glass, Tokyo, Japan) for intracellular
124
pH measurement or on the semi-permeable filters of transwell system (0.4 µm pore size,
125
Transwell® Permeable Supports, catalog no. 3450, Corning) for cell surface biotinylation
126
assay. IMCD cells were fed every 48 h and were grown in hypertonic culture medium (640
127
mOsm/KgH2O) supplemented with 10% fetal bovine serum at 37C in 5% CO2, 95% air
128
atmosphere for 3 days, and then in fetal bovine serum-free culture medium for additional 1
129
day before the experiment at day 5. The culture medium was Dulbecco's modified Eagle's
130
medium/F12 without phenol red, containing 80 mM urea, 130 mM NaCl, 10 mM HEPES, 2
131
mM L-glutamine, penicillin/streptomycin 10,000 units/ml, 50 nM hydrocortisone, 5 pM
132
3,3,5-triiodo-thyronine, 1 nM sodium selenate, 5 mg/L transferrin, and 10% fetal bovine
133
serum (pH 7.4, 640 mOsm/KgH2O).
134 135
IMCD tubule suspensions. Fresh inner medullary collecting duct (IMCD) tubules were
136
prepared from rat kidneys, as previously described (5, 35). Rats were anesthetized under
137
enflurane inhalation. Kidney inner medulla was dissected, minced, and digested in IMCD
138
suspension buffer (250 mM sucrose, 10 mM triethanolamine, pH 7.4) containing collagenase
139
B (3 mg/ml) and hyaluronidase (2 mg/ml). Then, IMCD suspension was continuously 5
MS#: F-00376-2014R1, Revised Manuscript 140
agitated in a warm water bath at 37°C for 90 min. Thereafter, IMCD tubule suspension was
141
centrifuged at 60 g for 30 sec to separate the IMCD-enriched fraction (pellet) and the non-
142
IMCD fraction (supernatant). IMCD-enriched fraction was collected and washed twice with
143
ice-cold IMCD suspension buffer. For the experiments, IMCD tubule suspensions were
144
exposed to the HEPES-buffered saline solution (162.5 mM NaCl, 2.5 mM Na2HPO4, 4 mM
145
KCl, 25 mM HEPES, 2 mM CaCl2, 1.2 mM MgSO4, and 5.5 mM glucose) with different pH
146
(pH 5.4, 6.4, 7.4 or 8.4) for 1 h, which was established by adding HCl (for pH 5.4) or NaOH
147
(for pH 6.4, 7.4, and 8.4) (43). dDAVP stimulation (10-10 M) was done during the last 3 min
148
(Fig. 1A).
149 150
Semiquantitative immunoblotting. SDS-PAGE was performed on 12% polyacrylamide gels,
151
as previously described (34). Primary antibodies used were anti-AQP2 (H7661AP, 1:2,000)
152
(15, 23), anti-phosphorylated AQP2 at S256 (K0307AP, 1:3,000) (34), anti-phosphorylated
153
AQP2 at S261, S264 and S269 (Phosphosolutions, 1:1,000), and anti-β-actin (Sigma,
154
1:100,000).
155 156
Immunofluorescence microscopy of primary cultured IMCD cells. IMCD cells were grown
157
to confluence in each transwell chamber (0.4 µm pore size, Transwell® Permeable Supports,
158
catalog no. 3460, Corning) for 4 days. At day 5, IMCD cells were subjected to transepithelial
159
pH gradient for 1 h (pH 6.4, 7.4 or 8.4 at the apical side vs. pH 7.4 at the basolateral side and
160
vise versa) at 37°C. dDAVP stimulation (10-9 M, 15 min, basolateral side only) was done
161
during the last 15 min and then fixed with 3% paraformaldehyde in PBS, pH 7.4 for 20 min at
162
room temperature. After fixation, cells were washed twice in PBS and permeabilized with
163
0.3% Triton X-100 in PBS at room temperature for 15 min. Cells were washed, the labeled
164
with anti-AQP2 (H7661AP, 1:200) antibody and nucleus was stained with DAPI. The AQP2 6
MS#: F-00376-2014R1, Revised Manuscript 165
localization was determined using laser confocal microscopy (Zeiss LSM 5 EXCITER, Jena,
166
Germany), and an X63 (NA 1.4) objective lens (Zeiss) was used. Digital images were
167
collected and analyzed using the Zeiss Aim Image Examiner program.
168 169
Cell surface biotinylation assays. IMCD cells were primary cultured for 4 days on semi-
170
permeable filters of transwell system (0.4 µm pore size, catalog no. 3450, Transwell®
171
Permeable Supports, Corning Incorporated, MA). At day 5, they were exposed to
172
transepithelial pH gradient for 1 h (apical pH 6.4, 7.4 or 8.4 vs. basolateral pH 7.4 and vice
173
versa) at 37°C. Then, the cells were washed three times with ice-cold PBS-CM (10 mM PBS
174
containing 1 mM CaCl2 and 0.1 mM MgCl2, pH 7.5) and incubated at 4˚C for 45 min in ice-
175
cold biotinylation buffer (10 mM triethanolamine, 2 mM CaCl2, 125 mM NaCl, pH 8.9)
176
containing 1 mg/ml sulfosuccinimidyl 2-(biotinamido)-ethyl-1,3-dithiopropionate (Sulfo-
177
NHS-SS-biotin, Thermo Scientific, Rockford, IL) on the apical side. Cells were washed once
178
with quenching buffer (50 mM Tris-HCl in PBS-CM, pH 8.0) followed by washing twice
179
with PBS-CM. Lysis buffer (1% triton X-100, 150 mM NaCl, 5 mM EDTA, 50 mM Tris-HCl,
180
pH 7.5, 1.15 mM PMSF, 0.4 μg/ml leupeptin, 0.1 mg/ml pefabloc, 0.1 μM okadaic acid, 1
181
mM Na3VO4, 25 mM NaF) was added and the lysates were sonicated 2 × 6 pluses at 20% of
182
amplitude. The lysates were centrifuged at 10,000 g for 5 min at 4˚C. The supernatant was
183
transferred to columns where Neutravidin agarose resin (200 μl) was previously loaded
184
(Thermo Scientific, Rockford, IL) and incubated for 60 min at RT with end-over-end mixing.
185
After 4 times washing with PBS containing protease inhibitors, 1 × sample buffer containing
186
DTT was added to the column and incubated for 60 min at RT. Samples were heated at 65˚C
187
for 10 min.
188 189
Intracellular pH measurement. Intracellular pH (pHi) was measured using the pH-sensitive 7
MS#: F-00376-2014R1, Revised Manuscript 190
fluorescent
191
(BCECF-AM, Molecular Probes). BCECF-AM is converted to the membrane-impermeant
192
fluorescent dye BCECF by intracellular esterases. Primary cultured IMCD cells of rat kidney
193
were seeded and cultured on the collagen-coated 35-mm glass-base dish (Asahi Techno Glass,
194
Tokyo, Japan) in hypertonic culture media (pH 7.4 and 640 mOsm/kgH2O) for 4 days. On
195
day 5, cells were incubated with non-serum IMCD culture media (Dulbecco's modified
196
Eagle's medium/F12 without phenol red, 80 mM urea, 130 mM NaCl, penicillin/streptomycin
197
10,000 units/ml) containing 2 µM BCECF-AM at 37C in 5% CO2-95% air for 30 min. The
198
cells were washed twice in PBS and changed to non-serum IMCD culture media (pH 7.4) for
199
10 min at 37C incubator before the experiments. IMCD cells were mounted in the chamber
200
on the temperature-controlled stage of Nikon Ti-E inverted microscope (Nikon, Tokyo,
201
Japan). The cells were continuously perfused with a Na+-free solution with different pH
202
values (pH 6.4, 7.4, and 8.4, see below) for 60 min by six-channel system (VC-2, Warner
203
Instruments, Hamden, CT) and perfusate was collected by the peristaltic pump (EYELA MP3,
204
Tokyo, Japan).
dye,
(2,7)-biscarboxyethyl-5(6)-carboxyfluorescein
acetoxymethyl
ester
205
The pHi was measured using the dual-excitation fluorescence ratio method. The
206
fluorescence was monitored at 530 nm with excitation wavelengths of 440 nm (a wavelength
207
at which the fluorescence is relatively pH insensitive) and 490 nm (at which the fluorescence
208
is very pH sensitive). The pHi was time-lapse obtained by the Nikon Ti-E inverted
209
microscope (Nikon, Tokyo, Japan) equipped with Cascade 512B (EMCCD) camera (Roper
210
Scientific, Trenton, NJ) and excitation and emission filter wheels (MAC5000, Ludl Electronic
211
Products, Hawthorne, NY). All systems were controlled by MetaMorph software (Universal
212
Imaging, Downingtown, PA). Fluorescent images were acquired sequentially through CFP
213
filter channels. Images were acquired by using the 4 x 4 binning mode and 200-ms exposure
214
time. A X60 (NA 1.4, Nikon, Tokyo, Japan) objective lens was used for the measurements. 8
MS#: F-00376-2014R1, Revised Manuscript 215
The 490/440 nm ratio was created by MetaMorph software. The data were obtained from five
216
independent experiments in both groups, respectively.
217
The pHi was calculated by the nigericin calibration technique (3). IMCD cells were
218
exposed to a Na+-free solution (0 mM Na+, 105 mM K+, 0 mM NH4+, 49 mM NMDG+, 5 mM
219
Mg2+, 0 mM Ca2+, 6.3 mM Cl-, 106.4 mM Aspartate-, 0.4 mM H2PO4-, 1.6 mM HPO42-, 17.8
220
mM HEPES-, 0 mM HCO3-, 5 mM SO42-, 10 mM EGTA2-, 5.5 mM glucose, 5 mM alanine,
221
14.4 mM HEPES) with different pH values (6.4, 7.4, and 8.4) by adding KOH, containing 10
222
µM of the K+-H+ exchanger nigericin. When the [K+] was equal between the intracellular and
223
extracellular compartments, pHi and pHe should be the same with nigericin (490/440 nm
224
ratio = 1.0, pH = 7.0) (3). Therefore the data were normalized to make the ratio at pH 7.0
225
equal to unity. Raw data of intensity at each excitation wavelength were corrected for
226
background-corrected fluorescence intensity prior to calculation of the ratios. The ratios of
227
background-corrected emission intensities (490/440 nm ratio) were transformed into pH
228
values as pH = log [(Rn-Rn(min))/(Rn(max)-Rn)] + pKa, where Rn was the background-
229
substracted ratio normalized to pH 7.0, Rn(min) and Rn(max) are the minimum and maximum
230
obtainable values for the normalized ratio and pKa is the –log of the dissociation constant of
231
the fluorophore (1, 3).
232 233
Fluorescence resonance energy transfer imaging for PKA activity. LLC-PK1 cells stably
234
transfected with AKAR3EV (PKA fluorescence resonance energy transfer (FRET)-based
235
indicator) were cultured on a collagen-coated 35-mm glass-base dish (Asahi Techno Glass,
236
Tokyo, Japan). The plasmid for PKA FRET-based indicator (AKAR3EV) was kindly
237
provided by Prof. M. Matsuda (Kyoto University, Kyoto, Japan). AKAR3EV was composed
238
of the FHA1 (ligand domain), yellow fluorescent protein (YFP), sensor domain of PKA
239
(LRRATLVD), and cyan fluorescent protein (CFP) (17). In this probe design, the sensor 9
MS#: F-00376-2014R1, Revised Manuscript 240
domain changes its conformation upon perception of the PKA signal (17). This sensitized
241
sensor domain interacts with the ligand domain, thereby inducing a global change of the
242
biosensor conformation and concomitant increase of FRET efficiency from the donor to the
243
acceptor. With this AKAR3EV probe, the activity of PKA was time-lapse obtained by Nikon
244
Ti-E inverted microscope (Nikon, Tokyo, Japan) equipped with a Cascade 512B (EMCCD)
245
camera (Roper Scientific, Trenton, NJ) and excitation and emission filter wheels (MAC5000,
246
Ludl Electronic Products, Hawthorne, NY). All systems were controlled by MetaMorph
247
software (Universal Imaging, Downingtown, PA). Fluorescent images were acquired every 15
248
sec sequentially through CFP and FRET filter channels. Images were acquired by using the 4
249
x 4 binning mode and 200-ms exposure time. The FRET/CFP ratio was created with
250
MetaMorph software, and the value was normalized to 1 at the starting time point in both
251
groups. The data were obtained from five independent experiments in both groups,
252
respectively.
253 254
Fluorescence resonance energy transfer imaging for intracellular calcium dynamics. Two
255
days after the transient transfection of YC3.60 (calcium FRET-based indicator) (29), LLC-
256
PK1 cells in Dulbecco’s Modified Eagle Medium (phenol red-free) were subjected to time-
257
lapse imaging. LLC-PK1 cells transiently transfected with YC3.60 were cultured on a
258
collagen-coated 35-mm glass-base dish (Asahi Techno Glass, Tokyo, Japan). The plasmid for
259
calcium FRET-based indicator (YC3.60) was also provided by Prof. M. Matsuda (Kyoto
260
University, Kyoto, Japan). YC3.60 was composed of the calmodulin (CaM), enhanced yellow
261
fluorescent protein (EYFP), a glycylglycine linker, the CaM-binding peptide of myosin light-
262
chain kinase (M13), and an enhanced cyan fluorescent protein (ECFP) (29). Ca2+ ion binding
263
to CaM moiety initiates an intramolecular interaction between the CaM and the M13 domains,
264
which causing the chimeric protein to shift from a spread form to a more compact form, 10
MS#: F-00376-2014R1, Revised Manuscript 265
therefore increasing the efficiency of FRET from the CFP to the YFP (29). With this YC3.60
266
indicator, the changes of intracellular Ca2+ levels were time-lapse obtained by Nikon Ti-E
267
inverted microscope (Nikon, Tokyo, Japan). Images were acquired every 15 sec by using the
268
4 x 4 binning mode and 200-ms exposure time. The FRET/CFP ratio was created with
269
MetaMorph software, and the value was normalized to 1 at the starting time point in all
270
groups. The data were obtained from eight independent experiments in both groups,
271
respectively.
272 273
Statistical analyses. Values were presented as means ± standard errors. Data were analyzed
274
by one-way analysis of variance (ANOVA) followed by Bonferroni’s multiple comparison
275
test. Multiple comparisons tests were only applied when a significant difference was
276
determined in the ANOVA, P
