INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

LOOP DIURETICS HAVE LONG BEEN an integral part of the therapy for chronic lung disease in the premature infant. In addition to their diuretic effects, initial clinical observations show decreased airway resistance in response to loop diuretics (12, 16). Furthermore, aerosolized delivery of furosemide also results in bronchodilatation in both adult (2, 17) and pediatric (3, 21) patients with asthma. This is supported by in vitro studies that demonstrate relaxation of isolated airways from various animal models (1, 10, 23). Recently these loop diuretic-mediated relaxation responses were confirmed in vitro in human fetal airway tissue (11).

The renal mechanism for diuresis by these sulfonamide compounds is through inhibition of a Na-K-2Cl (NKCC) cotransporter. Work by Lavallee and coworkers (13) demonstrates that, except at very low chloride substrate concentrations, there is a correlation between airway relaxation to furosemide and functional cotransporter inhibition in whole airway tissues. In addition, the relative potencies for cotransporter inhibition and airway relaxation between furosemide and bumetanide are similar (13). Taken together, these findings suggest that the mechanism for loop diuretic-mediated airway relaxation may also involve a NKCC cotransporter.

To date, the expression of a NKCC cotransporter in whole airway tissue has only been inferred through loop diuretic-mediated ion exchange inhibition (10, 13, 22). The first step in confirming the link between airway relaxation and NKCC cotransporter function is to demonstrate that this protein is expressed in airway smooth muscle.

Thus we hypothesized that a functional NKCC cotransporter protein can be demonstrated in airway smooth muscle cells. In this study we used Western blot analysis to demonstrate that an NKCC cotransporter protein was expressed in isolated human fetal trachea and normal human bronchial smooth muscle cells (BSMC). In addition, we used immunohistochemistry and indirect immunofluorescence with laser confocal microscopy to localize the NKCC cotransporter to the cell membrane in tissue from human fetal airways and in BSMC. Cotransporter function in BSMC was confirmed in vitro by bumetanide-mediated inhibition of rubidium uptake.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study was approved by the Human Use Committee at Kapiolani Medical Center for Women and Children.

Tissue and cell preparation. Human fetal airway and kidney tissues (11-16 wk gestation) were obtained from Central Laboratory for Human Embryology, University of Washington, Seattle. Informed consent was obtained for fetal tissue donation and use in the research of birth defects and other diseases. Fetuses with known congenital or chromosomal abnormalities were excluded. Samples were placed in cold Waymouth medium supplemented with 10% fetal bovine serum, 0.25 µg/ml fungizone, and 50 µg/ml gentamicin (Life Technologies, GIBCO, Gaithersburg, MD) and placed on ice or fixed in 10% formalin and express mailed to arrive within 36 h postmortem. On receipt of tissues, the lung, airways, and kidneys were frozen in liquid nitrogen. Formalin-fixed tissues were paraffin embedded and sectioned.

NKCC cotransporter protein identification. The monoclonal (T4) antibody to the human colonic T84 epithelial cell NKCC cotransporter protein (Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, IA) was used for Western blot analysis of human fetal tracheal homogenates and human BSMC lysates (14). The epithelial cells from the tracheal segments were removed by rubbing before freezing.

Immunohistochemistry. Formalin-fixed tissues were paraffin embedded and sectioned to 4-µm thickness. BSMC were grown to near confluence on chamber slides and fixed in a paraformaldehyde-lysine-sodium periodate fixative.

Indirect immunofluorescence with laser confocal microscopy. BSMC were grown on chamber slides, fixed in 4% paraformaldehyde, and permeabilized in an Igepal solution with normal goat serum and bovine serum albumin to block. To examine the cells without permeabilizing the membrane, the normal growth medium was replaced with a HEPES-buffered medium and the slides were chilled. Slides were then blocked with a normal goat serum with bovine serum albumin solution.

NKCC cotransporter activity. BSMC were grown to subconfluence in 75-cm² flasks. Cell cycle synchronization was achieved by serum deprivation (<0.1% FBS) for 48 h, with subsequent resumption of the complete media overnight. The cells were then changed to a HEPES-buffered MCDB 131 media and allowed to equilibrate for 4 h at 37°C. Cultures were incubated with and without 10 µM bumetanide or 30 µM furosemide (Roche Laboratories, Nutley, NJ) for 30 min, in the presence of 10 µM ouabain (Sigma Chemical) to inhibit the Na-K-ATPase channel. ⁸⁶Rb (1 µCi/ml; Perkin-Elmer New England Nuclear, Boston, MA) was then added. After 30 min, ⁸⁶Rb influx was terminated by removing the medium and washing with ice-cold isotonic MgCl₂. Cell membranes were solubilized with 0.1% SDS. Lysates were placed in liquid scintillation cocktail, and  emission was measured (Tri-Carb, Packard Instruments, Meriden, CT). Total protein of the SDS extract was measured by Bradford assay (Bio-Rad Laboratories, Hercules, CA). ⁸⁶Rb uptake by control and loop diuretic-blocked cells were compared by use of paired t-test (Sigmastat, Jandel Scientific, San Rafael, CA), with significance at P < 0.05. Normality of distribution was determined by the Kolmogorov-Smirnov test. Data are expressed as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NKCC cotransporter protein identification. NKCC protein was identified using the T4 monoclonal antibody in the cell lysate of human BSMC and in tissue homogenates of human fetal trachea. A single band was seen at ~130 kDa in each of these samples (Fig. 1A). A similar band at 130 kDa was also obtained when the rabbit polyclonal antibody (Fig. 1C) and the T4 monoclonal antibody (Fig. 1B) were used to probe the normal human BSMC lysates. Human fetal kidney homogenates, examined for comparison, gave identical results (not shown).

View larger version (71K): [in this window] [in a new window]   Fig. 1.   Na-K-2Cl (NKCC) cotransporter protein detection. NKCC cotransporter protein was detected as a single band at 130 kDa in tracheal tissue and in bronchial smooth muscle cells (BSMC). A: T4 monoclonal antibody to NKCC cotransporter protein was used with a chemiluminescence detection system. Lane 1, molecular weight (MW) marker; lanes 2-4, human fetal tracheal lysates 10, 20, and 40 µg protein, respectively. B and C: BSMC lysates (40 µg) probed with the rabbit polyclonal antibody (Ab; C) and with the T4 monoclonal Ab (B), with respective MW marker lanes.

NKCC localization to airway smooth muscle. Immunostaining for the NKCC cotransporter by using the polyclonal antibody was positive in bronchial epithelium. Positive staining was also seen in airway smooth muscle (Fig. 2, A and B). The NKCC cotransporter protein appears to be present throughout the airway smooth muscle cell membrane, because a specific intracellular localization of staining is not apparent. Vascular endothelium and smooth muscle also stained positively (not shown). No DAB-positive areas were demonstrated when tissue sections were stained with preimmune serum (Fig. 2C).

View larger version (108K): [in this window] [in a new window]   Fig. 2.   Immunolocalization of the NKCC cotransporter. A-C: human fetal trachea. D-E: normal human BSMC. A rabbit polyclonal antibody to the NKCC cotransporter was used with 3,3'-diaminobenzidine (DAB) detection and hematoxylin counter stain. A: human fetal trachea, ×50. B: ×200 magnification. Positive staining is seen in airway smooth muscle cells as well as in the bronchial epithelium. C: no positive staining with preimmune serum, ×200. D: positively stained BSMC, ×200 magnification. E: no positive staining is seen in controls probed with preimmune serum, ×200.

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Indirect immunofluorescence with laser confocal microscopy. BSMC probed with rabbit polyclonal NKCC cotransporter antibody (A-C) and with preimmune serum (B-D), ×40 oil immersion lens. In permeabilized cells, diffuse staining is noted with the NKCC antibody (A), compared with the preimmune serum (B). In nonpermeabilized cells, there is very distinct staining of the plasma membrane with the NKCC antibody (C), with no staining with the preimmune serum (D).

NKCC cotransporter activity. NKCC cotransporter activity was measured as a function of potassium uptake by using ⁸⁶Rb, an analog for potassium ion. BSMC exposed to either loop diuretic, bumetanide (Fig. 4A), or furosemide (Fig. 4B), demonstrated significantly less ⁸⁶Rb uptake compared with controls.

View larger version (11K): [in this window] [in a new window]   Fig. 4.   NKCC activity. 86Rb uptake was significantly inhibited in BSMC incubated with 10 µM bumetanide (n = 3; P < 0.03; A) and with 30 µM furosemide (n = 8; P < 0.03; B) compared with controls. 86Rb uptake was measured by liquid scintillation  emission, corrected for total protein.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Until now the physical presence of the NKCC cotransporter protein in airway smooth muscle has always been inferred. To identify the NKCC cotransporter protein in the airway smooth muscle, we used the T4 monoclonal antibody to probe human fetal tracheal homogenates and BSMC lysates by Western blotting. The expression of a NKCC cotransporter in airway smooth muscle was confirmed in isolated normal human BSMC. Thus the demonstration of this protein in whole tracheal homogenates as well as in cultured airway smooth muscle cells excludes the possibility that NKCC cotransporter expression was induced in culture, as reported for other cell types (20).

Once NKCC cotransporter protein was positively identified by Western blotting, immunohistochemistry was used for in situ localization. Crouch and coworkers (6) used the T4 antibody with no counterstaining to identify NKCC cotransporter in the gerbil inner ear, and Dunn et al. (8) used this monoclonal antibody with immunofluorescence to visualize the protein in bovine ciliary epithelium. However, our efforts to use the T4 monoclonal antibody with DAB chromogen for immunolocalization of NKCC cotransporter protein in lung tissue resulted in very faint staining, which was difficult to see in conjunction with counterstain. Thus we developed a polyclonal antibody that allowed us to directly visualize NKCC cotransporter protein in the BSMC. The specificity of this rabbit polyclonal antibody was confirmed in BSMC lysates by Western blotting.

In the stained lung tissue sections, cotransporter protein was shown to be present at the basolateral membrane of the bronchial epithelium. This is an expected finding, given the secretory function of these airway lining cells (8). The diffuse immunostaining of the airway smooth muscle cells and the specific staining of the plasma membrane confirm the widespread distribution of the NKCC cotransporter. These findings suggest that this cotransporter protein may have an important role in the functioning of these nonsecretory cells.

⁸⁶Rb uptake was used to demonstrate functional presence of a NKCC cotransporter protein by O'Donnell et al. (18) and Owen (19) in bovine and rat vascular smooth muscle cells in culture, respectively; in rat and rabbit aorta by Deth and coworkers (7); and in isolated guinea pig airways (10, 13, 22). In this study, ⁸⁶Rb uptake was used to correlate NKCC cotransporter activity with the expression of the protein in normal human BSMC.

A question still remains as to whether the NKCC cotransporter may be involved in the regulation of airway smooth muscle tone. Lavallee et al. (13) correlated NKCC activity with airway relaxation responses to loop diuretic-mediated inhibition, suggesting cotransporter involvement in the change in airway muscle tone. Thus the confirmation of NKCC cotransporter expression in airway smooth muscle suggests that this protein is directly involved in the alteration of airway smooth muscle tone. This is also supported by the demonstration of relaxation responses in isolated human airways (11). However, because bumetanide binding sites on the NKCC molecule are not the same as the regions involving ion transport (9), ion transport inhibition and smooth muscle relaxation may be mediated by separate mechanisms that are both initiated by loop diuretic binding. This would also support the observations by Corboz and co-workers (4, 5) that the airway relaxation effects due to loop diuretics may be secondary to vascular changes.

The potential clinical relevance of defining the role of the NKCC cotransporter in airway smooth muscle function is highlighted by in vitro evidence that cotransporter function can be pharmacologically downregulated by prior exposure to loop diuretics (10) and by studies demonstrating that cotransporter expression can be altered by specific infections and inflammation. For example, inflammatory cytokines interleukin-1 and tumor necrosis factor- upregulate NKCC cotransporter mRNA and protein expression in human umbilical vein endothelial cells (24). In addition, CMV infection abolishes NKCC cotransporter function and protein expression in human fibroblasts (15).

In summary, we have definitively shown that functional NKCC cotransporter protein is expressed and functional in human airway smooth muscle. This protein was identified and localized to the cell membrane in isolated human fetal trachea and normal human BSMCs. Our present findings lay the foundation for additional studies on the role of a NKCC cotransporter in the regulation of airway smooth muscle tone.