Reversed anion selectivity in cultured cystic fibrosis sweat duct cells C. L. BELL, Division
M. M. REDDY,
AND
P. M. QUINTON
of Biomedical Sciences, University of California, Riverside, California 92521-0121
Bell, C. L., M. M. Reddy, and P. M. Quinton. Reversed anion selectivity in cultured cystic fibrosis sweat duct cells. Am. J. Physiol. 262 (Cell Physiol. 31): C32-C38, 1992.-The human genetic disease cystic fibrosis (CF) is characterized by defective epithelial Cl- conductance (GcJ. To distinguish the CF-affected Gel from other Cl- channels, we have studied the properties of Gel in normal and CF cells grown from explanted reabsorptive sweat ducts (RD). The cultured cells from normal subjects retained some of the typical duct cell properties. The Na’ conductance inhibitor amiloride hyperpolarized intracellular potentials (I&-,) by 10.4 t 1.6 mV (n = 12). Substitution of gluconate for Cl- depolarized Vm by 15.5t 1.1 mV (n = 33). The apparent Gel (G&J of normal cells was sensitive to adenosine 3’,5’-cyclic monophosphate (forskolin, 10v6 M), as evidenced by a significant increase (63%, n = 9) in the Cl- gradient induced depolarization, and more selective for Cl- than I(substitution of Cl- by I- depolarized Vm by 6.3 t 0.3 mV, n = 49). Although the cells from CF subjects were statistically indistinguishable from normal cells based on Vm (-18.5 t 1.2 mV, n = 49 vs. -20.1 t 1.8 mV, n = 28), CF cells expressed differences in G&, responses to forskolin, and anion selectivity. CF cells had a significantly reduced G& as indicated by blunted responses to imposed Cl- gradients (26% of normal, n = 28). In contrast to our observations in normal cells, the G& of CF cells was insensitive to forskolin. Further, the residual G& of CF cells was clearly distinguishable from the predominant G& of normal cells because the anion selectivity of CF cells was reversed compared with normal, I- > Cl- ( Vm hyperpolarized by 3.9 & 0.4 mV, n = 15, when I- was substituted for Cl-). Although CF cells may be distinguished from normals by several properties, we found that this reversed anion selectivity was the most consistent and could be a new, highly reliable diagnostic marker for expression of the CF gene in vitro.
electrophysiological properties expressed by the native tissue (3, 10, 23, 24). However, to use cultured cells to study abnormal ion transport, it is essential to be able to consistently demonstrate a detectable functional expression of the CF defect. Although statistical differences in ,8-adrenergic responsiveness and in the relative size of the apparent Gel (G&l) have been demonstrated in cultured RD cells from CF and normal subjects, these differences have not been able to serve as reliable diagnostic tools for expression of the CF defect because 1) there is a significant overlap in the values determined for relative G& in normal and CF cells (23, 24) and 2) normal cells with intracellular potentials ( Vm) close to the apparent Cl- equilibrium potentials do not respond to ,8-adrenergic stimulation (23, 24). However, because physiologically distinct channels may express distinct anion selectivity sequences (2,5,9,28), we proposed that characterization of different anion selectivity properties in normal and CF cells might provide a simple and consistent diagnostic tool for the expression of the CF defect. The present study confirms that normal and CF-cultured RD cells can be tentatively identified based on the responses to forskolin and the magnitude of relative G&. However, we demonstrate that the most consistent diagnostic marker to distinguish between CF and normal cultured RD cells is their anion selectivity. The predominant anion conductance of normal cells is more selective for Cl- than I-, whereas the sequence is reversed in CF cells where I- is preferred to Cl-.
regulation; forskolin; adenosine 3’,5’-cyclic monophosphate; intracellular potentials; iodide; anion permeability; membrane diffusion potentials
MATERIALS
cystic fibrosis (CF) is characterized by abnormal Cl- conductance (GcJ in the exocrine glands (18). However, characterization of the CF-affected Gel (CF-GcJ has been impeded by the presence of more than one physiologically distinct Gel in CF-affected tissues (2, 5, 7, 12, 26). The conductance that is accessible or visible for analysis probably depends upon the physiological status of the tissue or the environmental conditions, as in patch clamp. The reabsorptive sweat duct (RD) is an extremely useful model to study the physiological properties of the CF-Gel. The predominant anion conductance (>80%) of the normal RD is composed of the CF-Gel (4, 16, 17) so that properties of the whole cell Gel must reflect the physiological properties of the CF-Gel. Unfortunately, small size and quantity of intact tissue are significant limitations in characterizing the Gel in the RD, but RD cells in primary culture can be used as an alternative model for electrophysiological investigation of ion transport (3,6,10,13,14,23,24). We have shown that cultured RD cells from normal and CF subjects retain certain
THE GENETIC
c32
DISEASE
0363-6143/92
$2.00
Copyright
AND
METHODS
Cell Culture Initiation of primary cultures. Skin specimens used for isolation of eccrine sweat glands were obtained from abdominal incisions at surgery or by dermatologic punch biopsy over the scapula after informed consent. Primary cultures were initiated as described previously (10). Briefly, after digestion in collagenase (85 U/ml) for 20-24 h, the tissue was transferred to culture medium and the epidermis was pulled from the underlying connective tissue, revealing the RDs that remained attached to the embedded secretory coils (SCs). Further pulling resulted in the separation of the RDs from the SCs at a region near the junction between the two. This region was then removed and discarded together with a short length of adjacent RD and attached SC. Individual ducts were transferred in a small drop of culture medium to 35-mm plastic tissue culture dishes (10). The explants were left undisturbed for at least 12 h and maintained at 37°C in a humidified 5% COZ-95% air incubator. On the following day, 1.5 ml of fresh medium were added, and the medium was subsequently changed every 2 days. Culture medium. Culture medium was MCDB 170 (8) supplemented with whole bovine pituitary extract (70 pg/ml), epidermal growth factor (EGF; 25 rig/ml), hydrocortisone (0.5 pg/ml), insulin (5 pg/ml), transferrin (5 pg/ml), ethanolamine (lo-* M), phosphoethanolamine (lo-* M), penicillin (50 U/ml), and streptomycin (50 pg/ml).
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.
CYSTIC FIBROSIS
AFFECTS
ANION
SELECTIVITY
c33
Fig. 1. Reabsoxptive duct cultures showing initial outgrowth along entire length of explanted duct 24 h after explantation (A; x150) and after 21 days in culture (B; X300).
the cells could be observed with an inverted microscope.The cultures were bathed with Ringer salt solution containing the following (in mM): 150 NaCl, 0.38 KH2P04, 2.13 K2HP04, 1 MgSO,. 7Hz0, 1 Ca(CH&OO),, and 10 glucose,at pH 7.4 and Control Amiloride AV, n 37°C. In experiments to investigate anion permeability, I- and -32.7+3.0* -10.4e1.6 Normal 12 -22.3k3.3 the impermeant anion gluconatewere substituted for Cl-. Amil-22.2k1.7 -32.2+2.3* -1l.Ok1.2 CF 24 oride (10m5M) and forskolin (10e6 M) were added to the Values given in mV are means f SE; n, number of cells impaled in perfusing solution as needed. 3 cultures from 3 normal subjects and in 3 cultures from 3 CF subjects. Intracellular impalements.Microelectrodes were fabricated * Significantly different from the control (P < 0.05). V,, intracellular from l-mm (OD) filamented glass capillaries on a Flaming potential; CF, cystic fibrosis. Brown Micropipette Puller @utter Instrument, San Francisco, CA) and filled with 4 M K+ acetate. Electrodes were usedat a Electrophysiology resistanceof 40-100 Ma. V,,, levels were measuredwith referSolutions. For microelectrode impalements, a rectangle of ence to the bath ground made of 4% agar in 3 M KC1 solution plastic bearing the explant culture was cut from the 35-mm using a singleelectrode voltage clamp amplifier (Dagan, model culture dish and transferred to an experimental chamberwhere 8100-l). A hybrid mechanopiezoelectric microstepper (Fran-
Table
1. Effect of amiloride on V,,, of normal and CF cells
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.
c34
CYSTIC
FIBROSIS
AFFECTS
SELECTIVITY
Data Analysis
‘sis +2
V
ANION
Results are expressed as the mean and standard error of the mean; n refers to the number of cells examined. As electrophysiological responses were the same for day 1 to day 21, data were analyzed irrespective of the age of the culture. Statistical significance was determined at the 5% confidence level using the Student’s t test for paired values. We did not attempt to calculate the permeability coefficients for the anions based on the Goldman-Hodgkin-Katz equation due to uncertainties about uncharacterized shunt pathways and cation conductances in RD cells grown on plastic substrate. Because we were mainly interested in the qualitative differences, apparent anion selectivity was evaluated on the basis of relative shifts in Vm during complete substitution of Cl-.
m-9
WI
-10
-22 2
min
RESULTS
Fig. 2. Effect of complete Cl- substitution by either the impermeable anion gluconate or I- on the intracellular potential (V,) of a normal cell. Note that I-, like gluconate, depolarized V,, although to a lesser degree, indicating that normal duct cells are less permeable to I- than Cl-. Gluconate
I
2 min
Fig. 3. A typical response of V, of a cystic fibrosis (CF) duct cell to complete Cl- substitution either by gluconate or I-. Note that, compared with the normal cell responses (Fig. 2), the CF cell’s depolarization with gluconate is significantly blunted and substitution of Cl- by I- resulted in a hyperpolarization of V, compared with the depolarization of the normal cell (Fig. 2).
Table 2. Effect of Cl- substitution by gluconate on Vm of normal and CF cells
Normal CF
n
Control (NaCl)
Experimental (NaGlu)
33 28
-22.0t1.5 -20.11tl.B
-15.5tl.B*
-5.ot1.4*
15.5t1.1
AND
DISCUSSION
We have previously demonstrated that cultured RD cells from the normal eccrine sweat gland retain at least some of the characteristic electrophysiological properties expressed by the parental tissue (3, 10, 23, 24). We have also indicated that the RD cells cultured from CF subjects exhibited abnormal Cl- permeability and a lack of responsiveness to ,&adrenergic stimulation (23). However, none of the electrophysiological differences between normal and CF cells described thus far could serve as a definitive marker for the expression of the CF gene product in the cell membranes in vitro due to inconsistent responses or overlaps in values representing these properties. Lack of a consistent diagnostic tool in vitro for the identification of the functional CF phenotype has impeded progress in developing cell cultures and cell lines, even in cells that express the CF gene product. We have, therefore, studied the electrophysiological properties of normal and CF cells in primary culture with two basic objectives in view: 1) elucidation of the characteristic properties of the CF-affected Gel in a tissue whose predominant anion conductance is affected by CF (16, 17, 20, 21), and 2) identification of diagnostic electrophysiological properties that can serve to reliably confirm the expression of the CF gene product in vitro.
4.OkO.B"f
Values given in mV are means t SE; n, number of cells impaled in 14 cultures from 7 normal subjects and in 3 cultures from 3 CF subjects. * Significantly different from NaCl (P c 0.05). t Significantly different from normal (P < 0.05).
kenberger PM 500-20, FRG) was used to obtain stable impalements. Intracellular potentials were measured according to criteria described by Reddy et al. (24). Microelectrode impalements of single cells were performed on cultures on solid substrates without passage from 1 to 21 days following explantation. We observed no correlation between the age of the culture and forskolin responsiveness or anion selectivity. Materials Collagenase (CLS II) was purchased from Worthington Biochemical (Freehold, NJ). MCDB 170 was purchased from UCSF Cell Culture Facility (San Francisco, CA). Whole bovine pituitary extract was supplied by Hammond Cell Tech (Alameda, CA). EGF was obtained from Amgen (Thousand Oaks, CA). Amiloride was a gift from Merck Sharp & Dohme Laboratories (West Point, PA). All other biochemicals were obtained from Sigma Chemical (St. Louis, MO).
Cell Growth Tightly packed polygonal cells grew out along the entire length of the explant within 24 h of explantation (Fig. 1A ). However, as reported previously for cells grown in MCDB 170 medium (3), cells began to senesceas early as 2 days in culture, with cells losing their characteristic shape and becoming granular. This was followed between day 5 and day 8 with a new outgrowth of viable, polygonal cells, which continued to proliferate for up to 3 to 4 wk (Fig. 1B). No differences in growth parameters were observed between cultures initiated from CF and normal individuals. This is in agreement with the observations of others (13, 14). Normal and CF Cells Have Similar Resting Potentials In an earlier report (20), we showed that the apical membrane of intact sweat ducts from CF subjects was reversed in polarity compared with normal duct cells. We have, therefore, compared the Vm values of normal and CF cultured RD cells for possible use as a diagnostic
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.
CYSTIC
FIBROSIS
AFFECTS
2
ANION
SELECTIVITY
c35
min
Fig. 4. A rare example of a CF cell showing near normal responses to Cl- substitution by gluconate. Unlike the majority of CF cells, we occasionally (
M. M. REDDY,
AND
P. M. QUINTON
of Biomedical Sciences, University of California, Riverside, California 92521-0121
Bell, C. L., M. M. Reddy, and P. M. Quinton. Reversed anion selectivity in cultured cystic fibrosis sweat duct cells. Am. J. Physiol. 262 (Cell Physiol. 31): C32-C38, 1992.-The human genetic disease cystic fibrosis (CF) is characterized by defective epithelial Cl- conductance (GcJ. To distinguish the CF-affected Gel from other Cl- channels, we have studied the properties of Gel in normal and CF cells grown from explanted reabsorptive sweat ducts (RD). The cultured cells from normal subjects retained some of the typical duct cell properties. The Na’ conductance inhibitor amiloride hyperpolarized intracellular potentials (I&-,) by 10.4 t 1.6 mV (n = 12). Substitution of gluconate for Cl- depolarized Vm by 15.5t 1.1 mV (n = 33). The apparent Gel (G&J of normal cells was sensitive to adenosine 3’,5’-cyclic monophosphate (forskolin, 10v6 M), as evidenced by a significant increase (63%, n = 9) in the Cl- gradient induced depolarization, and more selective for Cl- than I(substitution of Cl- by I- depolarized Vm by 6.3 t 0.3 mV, n = 49). Although the cells from CF subjects were statistically indistinguishable from normal cells based on Vm (-18.5 t 1.2 mV, n = 49 vs. -20.1 t 1.8 mV, n = 28), CF cells expressed differences in G&, responses to forskolin, and anion selectivity. CF cells had a significantly reduced G& as indicated by blunted responses to imposed Cl- gradients (26% of normal, n = 28). In contrast to our observations in normal cells, the G& of CF cells was insensitive to forskolin. Further, the residual G& of CF cells was clearly distinguishable from the predominant G& of normal cells because the anion selectivity of CF cells was reversed compared with normal, I- > Cl- ( Vm hyperpolarized by 3.9 & 0.4 mV, n = 15, when I- was substituted for Cl-). Although CF cells may be distinguished from normals by several properties, we found that this reversed anion selectivity was the most consistent and could be a new, highly reliable diagnostic marker for expression of the CF gene in vitro.
electrophysiological properties expressed by the native tissue (3, 10, 23, 24). However, to use cultured cells to study abnormal ion transport, it is essential to be able to consistently demonstrate a detectable functional expression of the CF defect. Although statistical differences in ,8-adrenergic responsiveness and in the relative size of the apparent Gel (G&l) have been demonstrated in cultured RD cells from CF and normal subjects, these differences have not been able to serve as reliable diagnostic tools for expression of the CF defect because 1) there is a significant overlap in the values determined for relative G& in normal and CF cells (23, 24) and 2) normal cells with intracellular potentials ( Vm) close to the apparent Cl- equilibrium potentials do not respond to ,8-adrenergic stimulation (23, 24). However, because physiologically distinct channels may express distinct anion selectivity sequences (2,5,9,28), we proposed that characterization of different anion selectivity properties in normal and CF cells might provide a simple and consistent diagnostic tool for the expression of the CF defect. The present study confirms that normal and CF-cultured RD cells can be tentatively identified based on the responses to forskolin and the magnitude of relative G&. However, we demonstrate that the most consistent diagnostic marker to distinguish between CF and normal cultured RD cells is their anion selectivity. The predominant anion conductance of normal cells is more selective for Cl- than I-, whereas the sequence is reversed in CF cells where I- is preferred to Cl-.
regulation; forskolin; adenosine 3’,5’-cyclic monophosphate; intracellular potentials; iodide; anion permeability; membrane diffusion potentials
MATERIALS
cystic fibrosis (CF) is characterized by abnormal Cl- conductance (GcJ in the exocrine glands (18). However, characterization of the CF-affected Gel (CF-GcJ has been impeded by the presence of more than one physiologically distinct Gel in CF-affected tissues (2, 5, 7, 12, 26). The conductance that is accessible or visible for analysis probably depends upon the physiological status of the tissue or the environmental conditions, as in patch clamp. The reabsorptive sweat duct (RD) is an extremely useful model to study the physiological properties of the CF-Gel. The predominant anion conductance (>80%) of the normal RD is composed of the CF-Gel (4, 16, 17) so that properties of the whole cell Gel must reflect the physiological properties of the CF-Gel. Unfortunately, small size and quantity of intact tissue are significant limitations in characterizing the Gel in the RD, but RD cells in primary culture can be used as an alternative model for electrophysiological investigation of ion transport (3,6,10,13,14,23,24). We have shown that cultured RD cells from normal and CF subjects retain certain
THE GENETIC
c32
DISEASE
0363-6143/92
$2.00
Copyright
AND
METHODS
Cell Culture Initiation of primary cultures. Skin specimens used for isolation of eccrine sweat glands were obtained from abdominal incisions at surgery or by dermatologic punch biopsy over the scapula after informed consent. Primary cultures were initiated as described previously (10). Briefly, after digestion in collagenase (85 U/ml) for 20-24 h, the tissue was transferred to culture medium and the epidermis was pulled from the underlying connective tissue, revealing the RDs that remained attached to the embedded secretory coils (SCs). Further pulling resulted in the separation of the RDs from the SCs at a region near the junction between the two. This region was then removed and discarded together with a short length of adjacent RD and attached SC. Individual ducts were transferred in a small drop of culture medium to 35-mm plastic tissue culture dishes (10). The explants were left undisturbed for at least 12 h and maintained at 37°C in a humidified 5% COZ-95% air incubator. On the following day, 1.5 ml of fresh medium were added, and the medium was subsequently changed every 2 days. Culture medium. Culture medium was MCDB 170 (8) supplemented with whole bovine pituitary extract (70 pg/ml), epidermal growth factor (EGF; 25 rig/ml), hydrocortisone (0.5 pg/ml), insulin (5 pg/ml), transferrin (5 pg/ml), ethanolamine (lo-* M), phosphoethanolamine (lo-* M), penicillin (50 U/ml), and streptomycin (50 pg/ml).
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.
CYSTIC FIBROSIS
AFFECTS
ANION
SELECTIVITY
c33
Fig. 1. Reabsoxptive duct cultures showing initial outgrowth along entire length of explanted duct 24 h after explantation (A; x150) and after 21 days in culture (B; X300).
the cells could be observed with an inverted microscope.The cultures were bathed with Ringer salt solution containing the following (in mM): 150 NaCl, 0.38 KH2P04, 2.13 K2HP04, 1 MgSO,. 7Hz0, 1 Ca(CH&OO),, and 10 glucose,at pH 7.4 and Control Amiloride AV, n 37°C. In experiments to investigate anion permeability, I- and -32.7+3.0* -10.4e1.6 Normal 12 -22.3k3.3 the impermeant anion gluconatewere substituted for Cl-. Amil-22.2k1.7 -32.2+2.3* -1l.Ok1.2 CF 24 oride (10m5M) and forskolin (10e6 M) were added to the Values given in mV are means f SE; n, number of cells impaled in perfusing solution as needed. 3 cultures from 3 normal subjects and in 3 cultures from 3 CF subjects. Intracellular impalements.Microelectrodes were fabricated * Significantly different from the control (P < 0.05). V,, intracellular from l-mm (OD) filamented glass capillaries on a Flaming potential; CF, cystic fibrosis. Brown Micropipette Puller @utter Instrument, San Francisco, CA) and filled with 4 M K+ acetate. Electrodes were usedat a Electrophysiology resistanceof 40-100 Ma. V,,, levels were measuredwith referSolutions. For microelectrode impalements, a rectangle of ence to the bath ground made of 4% agar in 3 M KC1 solution plastic bearing the explant culture was cut from the 35-mm using a singleelectrode voltage clamp amplifier (Dagan, model culture dish and transferred to an experimental chamberwhere 8100-l). A hybrid mechanopiezoelectric microstepper (Fran-
Table
1. Effect of amiloride on V,,, of normal and CF cells
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.
c34
CYSTIC
FIBROSIS
AFFECTS
SELECTIVITY
Data Analysis
‘sis +2
V
ANION
Results are expressed as the mean and standard error of the mean; n refers to the number of cells examined. As electrophysiological responses were the same for day 1 to day 21, data were analyzed irrespective of the age of the culture. Statistical significance was determined at the 5% confidence level using the Student’s t test for paired values. We did not attempt to calculate the permeability coefficients for the anions based on the Goldman-Hodgkin-Katz equation due to uncertainties about uncharacterized shunt pathways and cation conductances in RD cells grown on plastic substrate. Because we were mainly interested in the qualitative differences, apparent anion selectivity was evaluated on the basis of relative shifts in Vm during complete substitution of Cl-.
m-9
WI
-10
-22 2
min
RESULTS
Fig. 2. Effect of complete Cl- substitution by either the impermeable anion gluconate or I- on the intracellular potential (V,) of a normal cell. Note that I-, like gluconate, depolarized V,, although to a lesser degree, indicating that normal duct cells are less permeable to I- than Cl-. Gluconate
I
2 min
Fig. 3. A typical response of V, of a cystic fibrosis (CF) duct cell to complete Cl- substitution either by gluconate or I-. Note that, compared with the normal cell responses (Fig. 2), the CF cell’s depolarization with gluconate is significantly blunted and substitution of Cl- by I- resulted in a hyperpolarization of V, compared with the depolarization of the normal cell (Fig. 2).
Table 2. Effect of Cl- substitution by gluconate on Vm of normal and CF cells
Normal CF
n
Control (NaCl)
Experimental (NaGlu)
33 28
-22.0t1.5 -20.11tl.B
-15.5tl.B*
-5.ot1.4*
15.5t1.1
AND
DISCUSSION
We have previously demonstrated that cultured RD cells from the normal eccrine sweat gland retain at least some of the characteristic electrophysiological properties expressed by the parental tissue (3, 10, 23, 24). We have also indicated that the RD cells cultured from CF subjects exhibited abnormal Cl- permeability and a lack of responsiveness to ,&adrenergic stimulation (23). However, none of the electrophysiological differences between normal and CF cells described thus far could serve as a definitive marker for the expression of the CF gene product in the cell membranes in vitro due to inconsistent responses or overlaps in values representing these properties. Lack of a consistent diagnostic tool in vitro for the identification of the functional CF phenotype has impeded progress in developing cell cultures and cell lines, even in cells that express the CF gene product. We have, therefore, studied the electrophysiological properties of normal and CF cells in primary culture with two basic objectives in view: 1) elucidation of the characteristic properties of the CF-affected Gel in a tissue whose predominant anion conductance is affected by CF (16, 17, 20, 21), and 2) identification of diagnostic electrophysiological properties that can serve to reliably confirm the expression of the CF gene product in vitro.
4.OkO.B"f
Values given in mV are means t SE; n, number of cells impaled in 14 cultures from 7 normal subjects and in 3 cultures from 3 CF subjects. * Significantly different from NaCl (P c 0.05). t Significantly different from normal (P < 0.05).
kenberger PM 500-20, FRG) was used to obtain stable impalements. Intracellular potentials were measured according to criteria described by Reddy et al. (24). Microelectrode impalements of single cells were performed on cultures on solid substrates without passage from 1 to 21 days following explantation. We observed no correlation between the age of the culture and forskolin responsiveness or anion selectivity. Materials Collagenase (CLS II) was purchased from Worthington Biochemical (Freehold, NJ). MCDB 170 was purchased from UCSF Cell Culture Facility (San Francisco, CA). Whole bovine pituitary extract was supplied by Hammond Cell Tech (Alameda, CA). EGF was obtained from Amgen (Thousand Oaks, CA). Amiloride was a gift from Merck Sharp & Dohme Laboratories (West Point, PA). All other biochemicals were obtained from Sigma Chemical (St. Louis, MO).
Cell Growth Tightly packed polygonal cells grew out along the entire length of the explant within 24 h of explantation (Fig. 1A ). However, as reported previously for cells grown in MCDB 170 medium (3), cells began to senesceas early as 2 days in culture, with cells losing their characteristic shape and becoming granular. This was followed between day 5 and day 8 with a new outgrowth of viable, polygonal cells, which continued to proliferate for up to 3 to 4 wk (Fig. 1B). No differences in growth parameters were observed between cultures initiated from CF and normal individuals. This is in agreement with the observations of others (13, 14). Normal and CF Cells Have Similar Resting Potentials In an earlier report (20), we showed that the apical membrane of intact sweat ducts from CF subjects was reversed in polarity compared with normal duct cells. We have, therefore, compared the Vm values of normal and CF cultured RD cells for possible use as a diagnostic
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.
CYSTIC
FIBROSIS
AFFECTS
2
ANION
SELECTIVITY
c35
min
Fig. 4. A rare example of a CF cell showing near normal responses to Cl- substitution by gluconate. Unlike the majority of CF cells, we occasionally (
