Human nasal gland (HNG) cells cultured in collagen gel (C) with serum-free culture medium. The sections of cultured HNG cells form a lumen (L) containing mucus stained with Alcian blue. The HNG cells were cultured for 7 days in the conditioned medium containing 10-ng/mL epidermal growth factor and 100-ng/mL retinoic acid. Sections were stained with Alcian blue (pH 2.5), and nuclei were counterstained with hematoxylin-eosin (original magnification ×200). S indicates secretory products.
Transmission electron micrograph of human nasal gland cells after being cultured for 7 days in collagen gel with serum-free culture medium containing 10-ng/mL epidermal growth factor and 100-ng/mL retinoic acid. The lumen (L) is bordered by polarized secretory epithelial cells exhibiting microvilli (M) and numerous secretory granules (SG) located along the L border (original magnification ×18 960). JC indicates junctional complexes.
Bromodeoxyuridine incorporation in human nasal gland cells. The human nasal gland cells were cultured in conditioned medium containing 100-ng/mL retinoic acid without epidermal growth factor (A) and with 10-ng/mL epidermal growth factor (B). Brown color (arrows) indicates bromodeoxyuridine-labeled nuclei (original magnification ×200).
Percentage of bromodeoxyuridine-labeled nuclei (labeling index) in cultured human nasal gland (HNG) cells. A, The effect of epidermal growth factor (EGF) on HNG proliferation for HNG cells cultured in conditioned medium containing 100-ng/mL retinoic acid (RA) and 0-, 3-, 10, and 30-ng/mL EGF. B, The effect of keratinocyte growth factor (KGF) on HNG proliferation for HNG cells cultured in conditioned medium containing 100-ng/mL RA with 0- and 10-ng/mL EGF and 0- and 15-ng/mL KGF. C, The effect of RA on HNG proliferation for HNG cells cultured in conditioned medium containing 10-ng/mL EGF and 0-, 30-, 100-, and 300-ng/mL RA. Brackets indicate significantly different at P<.05. Error bars indicate SD.
Mucus production of cultured human nasal gland cells. The human nasal gland cells were cultured in conditioned medium containing 10-ng/mL epidermal growth factor without retinoic acid (A) and with 300-ng/mL retinoic acid (B). The blue spots (arrows) indicate secretory granules stained with Alcian blue (pH 2.5) (original magnification ×200).
The ratio of the number of mucus-producing cells to the number of total cells (M/T) of human nasal gland (HNG) cells. A, The effect of epidermal growth factor (EGF) on HNG differentiation for HNG cells cultured in conditioned medium containing 100-ng/mL retinoic acid (RA) and 0-, 3-, 10-, and 30-ng/mL EGF. B, The effect of keratinocyte growth factor (KGF) on HNG differentiation for HNG cells cultured in conditioned medium containing 100-ng/mL RA with 0- and 10-ng/mL EGF and 0- and 15-ng/mL KGF. C, The effect of RA on HNG differentiation for HNG cells cultured in conditioned medium containing 10-ng/mL EGF and 0-, 30-, 100, and 300-ng/mL RA. Bracket indicates significantly different at P<.05. Error bars indicate SD.
Kimura T, Majima Y, Guo Y, Yoshida T. The Effect of Growth Factors on the Proliferation and Differentiation of Human Nasal Gland Cells. Arch Otolaryngol Head Neck Surg. 2002;128(5):578-582. doi:10.1001/archotol.128.5.578
Copyright 2002 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2002
To elucidate a mechanism of proliferation and differentiation of nasal gland cells, we established a serum-free 3-dimensional culture system for human nasal gland (HNG) cells and examined the effects of epidermal growth factor, keratinocyte growth factor, and retinoic acid on proliferation and differentiation of cultured HNG cells.
Materials and Methods
Nasal polyps were obtained from patients undergoing endoscopic endonasal sinus surgery. The HNG cells were cultured under a monolayer culture and transferred to a collagen-embedded culture using RPMI 1640 medium containing transferrin, insulin, hydrocortisone, retinoic acid, epidermal growth factor, and keratinocyte growth factor. Cell growth was measured by bromodeoxyuridine incorporation assays. To measure cell differentiation, the percentage of cells containing secretory granules, which were stained with Alcian blue in cytoplasm, was determined.
In the serum-free 3-dimensional culture, the HNG cells showed ductal structures containing secretory products in a lumen. The addition of epidermal growth factor promoted the proliferation of HNG cells in its optimal concentrations, and keratinocyte growth factor also enhanced the proliferation of HNG cells. Conversely, the differentiation of HNG cells was not dependent on epidermal growth factor and keratinocyte growth factor. Retinoic acid suppressed the proliferation, but promoted the differentiation of HNG cells.
Our culture system could be useful for studying the effects of various growth factors and cytokines on HNG proliferation and differentiation to better understand the mechanisms of growth and morphogenesis of nasal glands.
SUBMUCOSAL GLANDS of respiratory tracts develop during organogenesis and remodels in various pathological conditions. Especially in chronic inflammations, glandular proliferation is remarkable. Mucous gland hyperplasia and hypertrophy was reported in the tracheobronchial mucosa in chronic bronchitis.1 Nasal and sinus gland hyperplasia is a common feature of chronic sinusitis.2,3 Such glandular hyperplasia is suggested to be a main cause of viscous nasal hypersecretion.2 The hyperplasia is speculated to be regulated by some growth factors and cytokines: however, there is little evidence about the factors related to proliferation and differentiation of nasal subglandular epithelia.
Our previous study described a 3-dimensional (3-D) culture system for human nasal gland (HNG) cells. The HNG cells in this culture system promoted glandular structures and secretory activity similar to the in situ conditions.4
To elucidate a mechanism of nasal gland hyperplasia, in the present study we established a serum-free 3-D culture system for HNG cells. Using this system, we examined the effects of growth factors and retinoic acid (RA) on proliferation and differentiation of cultured HNG cells.
Nasal polyps were obtained from patients undergoing endoscopic endonasal sinus surgery. The isolation of HNG cells was performed as described by Furukawa et al.5 Nasal polyps were exposed overnight at 4°C to 0.05% protease XIV (Sigma Chemical Co, St Louis, Mo) dissolved in phosphate-buffered saline solution. The enzyme activity was stopped by the addition of fetal calf serum to a final concentration of 2.5%, and small sheets of the surface epithelial cells were dislodged from the nasal polyps by vigorous agitation. The denuded nasal polyps with rich glands were collected and rinsed 3 times with phosphate-buffered saline solution. The polyps were minced and placed in Hanks solution containing 20mM HEPES, 500-U/mL collagenase type IV, 6-U/mL elastase, 200-U/mL hyaluronidase, and 10-U/mL DNase (Sigma Chemical Co) at room temperature for 4 hours. After the solution settled and supernatant decanted, the isolated glands in the solution were collected by centrifugation and resuspended in a mixture of RPMI 1640 medium (Sigma Chemical Co) supplemented with 20% fetal calf serum. The glandular epithelia were plated into a 25-cm2 tissue culture flask (T25) and incubated at 37°C in 5% carbon monoxide and 95% air. The next morning, the medium was replaced with RPMI 1640 medium containing 1-µg/mL transferrin, 1-µg/mL insulin, 0.5-µg/mL hydrocortisone, 10-ng/mL RA(Sigma Chemical Co), and 10-ng/mL epidermal growth factor (EGF) (Becton, Dickinson and Company, Bedford, Mass) as the basal medium. One week after cell plating, the confluent cells were trypsinized in 0.05% trypsin and 0.02% EDTA solution and then collected by centrifugation to transfer to the 3-D culture.
The conditioned medium used for the 3-D culture was RPMI 1640 medium containing 1-µg/mL transferrin; 1-µg/mL insulin; 0.5-µg/mL hydrocortisone; 0-, 30-, 100-, or 300-ng/mL RA; 0-, 3-,10-, or 30-ng/mL EGF; and 0- or 15-ng/mL keratinocyte growth factor (KGF) (Strathmann Biotech GMBH, Hannover, Germany). The collagen gels were prepared by mixing 8 mL of type I collagen solution (3.0 mg/mL) (Nitta Gelatin Inc, Tokyo, Japan), 1 mL of conditioned medium (concentration ×10), and 1 mL of reconstitution buffer. For the collagen-embedded system, 1 mL of this mixture (kept on ice) was first placed in a 12-well plastic plate and immediately warmed to 37°C for gel formation. One milliliter of the basal medium containing 1 × 105 cells was placed on collagen gel of a well. The cells were allowed to attach to the surface of the gel for 12 hours. The medium and unattached cells were removed, and a second layer of collagen was allowed to polymerize on top of the first gel, thus embedding the cells. The gel was further covered with 2 mL of the medium 30 minutes later. Three days after seeding the cells, the gels were removed from plastic substratum by circling a spatula along the inner edge of the well. While being cultured, the HNG cells were observed under phase contrast microscope.
After 7 days, the cells in the collagen gel were fixed with 10% formalin in 0.1M phosphate buffer (pH 7.4) and routinely embedded in paraffin for histochemical studies. Sections were stained with Alcian blue (pH 2.5), and nuclei were counterstained with nuclear fast red. The percentage of cells containing secretory granules stained with Alcian blue in cytoplasm were obtained to examine the degree of cell differentiation (No. of mucus-producing cells–total No. of cells [M/T ratio]). More than 200 cells were counted, and the given values are the means of determinations from 3 individual experiments.
Cell growth fraction was determined by bromodeoxyuridine (BrdU) incorporation assays.6 After 7 days, the cells in the collagen gel were labeled with 10 µg/mL of BrdU (Sigma Chemical Co) for 4 hours and then fixed with 10% formalin in 0.1M phosphate buffer (pH 7.4). Labeled nuclei were detected with monoclonal anti-BrdU antibody (DAKO A/S, Copenhagen, Denmark) and secondary anti-mouse IgG antibody conjugated with peroxidase (Medical & Biological Laboratories Co Ltd, Nagoya, Japan), followed by color development in diaminobenzidine hydrochloride solution. More than 200 cells were counted, and the percentage of nuclei that were labeled with BrdU (labeling index) was determined. The given values of the indexes are the means of determinations from 3 individual experiments.
For electron microscopic observations, the cells in the collagen gel were fixed with 2.5% glutaraldehyde in 0.1M phosphate-buffered saline solution at 4°C for 4 hours, followed by postfixation in 1% osmium tetroxide for 2 hours. The ultrathin sections were stained with 2% uranyl acetate and lead citrate and observed using a transmission electron microscope (model H-800; Hitachi, Tokyo, Japan).
After 7 days of 3-D culturing, the HNG cells in the conditioned medium containing 10-ng/mL EGF and 100-ng/mL RA developed projected ductal structures forming networks of the tubules. The cross-section shows a lumen containing secretory products (Figure 1).
Findings from electron microscopy revealed that 3-D–cultured HNG cells have characteristics similar to gland cells. They were well polarized and interconnected by junctional complexes and also had extensive development of microvilli along the luminal surface and numerous secretory granules in the cytoplasm (Figure 2).
Figure 3A-B shows BrdU-incorporated HNG cells cultured in the defined medium containing 100-ng/mL RA with and without EGF. The BrdU-incorporated nuclei were more prevalent in the cells cultured in the medium containing 10-ng/mL EGF (Figure 3B) compared with those in EGF-free medium (Figure 3A). Also, the labeling indexes of the HNG cells cultured in the medium containing 3- and 10-ng/mL EGF were significantly higher than those in the EGF-free medium (Figure 4A).
Figure 4B shows the labeling index with conditioned medium containing 100-ng/mL RA. The addition of 15-ng/mL KGF significantly increased the labeling index compared with the medium without either KGF or EGF. There were no significant differences in the labeling index between 15-ng/mL KGF and 10-ng/mL EGF. Synergetic effects were not observed when 15-ng/mL KGF and 10-ng/mL EGF were concomitantly added to the medium.
The labeling index of HNG cells in the medium containing 10-ng/mL EGF and 0-, 30-, 100-, or 300-ng/mL RA is shown in Figure 4C. Retinoic acid significantly decreased the labeling index in a dose-dependent manner.
Figure 5A-B shows cross-sections of HNG cells cultured in the medium containing 10-ng/mL EGF with 0- and 300-ng/mL RA. Quantitative analyses demonstrate that the frequency of Alcian blue–positive cells (M/T ratio) in the culture containing 300-ng/mL RA was significantly higher than that in the RA-deficient medium (Figure 6C). Neither EGF nor KGF showed any significant changes in the M/T ratio under the presence of 100-ng/mL RA (Figure 6A-B).
Our previous study showed that HNG cells cultured in the 3-D collagen gel initiated glandularlike morphogenesis and secretory activity.4 To evaluate intrinsic or extrinsic factors influencing the differentiation and proliferation of HNG cells, we cultured HNG cells in the collagen gel with serum-free–defined medium supplemented with transferrin, insulin, hydrocortisone, RA, and EGF. The cells cultured in this medium also developed ductal structures forming networks of the tubules. The structure showed a lumen containing secretory products in the histological section (Figure 1). The cells were positive for Alcian blue, revealing differentiation to mucous cells. Ultrastructural morphology confirmed that the cells form glands and/or ducts having a polarity and junctional complexes (Figure 2). The secretory glandules were also seen in the cytoplasm.
A well-known growth factor of epithelial cells, EGF is found in various body and secreted fluid and is related to the proliferation of gland cells, such as lacrimal, salivary, mammary, and prostate gland cells. The effects of EGF on the proliferation of nasal gland cells are easily speculated, but have not been analyzed. In the present study, the labeling index of BrdU in HNG cells was increased, according to increased concentrations of EGF up to 10 ng/mL. At 30-ng/mL EGF, however, the labeling index was suppressed in the EGF-free medium (Figure 4A). The results suggest that EGF promotes the proliferation of HNG cells in its optimal concentrations. An immunohistochemical study demonstrates that EGF and EGF receptors on bronchial epithelia and glands were upregulated in the asthmatic airway.7 Epidermal growth factor could play an important role in the proliferation of glandular epithelia in chronic inflammatory diseases of the upper respiratory tracts, such as chronic sinusitis and nasal allergy. In contrast, because EGF did not alter the M/T ratio (Figure 6A), differentiations of HNG cells are dependent on factors other than EGF.
Keratinocyte growth factor is a member of fibroblast growth factor family. It stimulates the proliferation of a broad range of epithelial cells including skin, bronchus, gastrointestinal tract, and mammary gland.8 In the present study, the labeling index of BrdU was significantly increased in the HNG cells cultured in the 15-ng/mL KGF medium compared with those in the medium without growth factors (Figure 4B), indicating that KGF enhanced the proliferation of HNG cells. Enhanced expression of KGF and KGF-receptor mRNA was reported in the nasal mucosa of chronic sinusitis.9 Taken together, KGF may contribute to nasal and sinus gland hyperplasia observed in chronic sinusitis.2 Because KGF has a synergetic effect with mammogenic hormones on the proliferation of mammary gland cells,10 we examined an effect of KGF in combination with EGF on the proliferation of HNG cells. However, no synergetic actions of KGF and EGF was detected (Figure 4B). Keratinocyte growth factor enhances differentiation of alveolar type II cells and/or their progenitors and promotes alveolarization during branching morphogenesis.9 In the present study, however, KGF, as well as combinations of KGF and EGF, had no effect on the M/T ratio of HNG cells (Figure 6B) and may have little relation to the differentiation of HNG cells.
Beta carotene and related compounds, summarily called retinoids, are known to regulate cell proliferation, differentiation, and morphogenesis. Depending on the cell or tissue type, retinoids induce or inhibit such regulation.11 Retinoic acid was reported to inhibit squamous differentiation of epithelial cells and was essential for the maintenance of mucociliary epithelium.12 Also, RA up-regulated mucous gene expressions in the retinoid-deficient cultures of airway epithelial cells.11 In the HNG cells, RA increased the M/T ratio in a dose-dependent manner (Figure 4C and Figure 6C), while it decreased the BrdU-labeling index. This clearly indicates that RA promotes the differentiation but suppresses the proliferation of the cells.
Hyperplasia of submucosal gland cells observed in chronic inflammations of upper and lower respiratory tracts is investigated from an aspect of factors influencing the proliferation and differentiation of the glandular cells. Our 3-D culture system with a defined culture medium could be useful for better understanding mechanisms of gland cell hyperplasia.
In conclusion, we established serum-free 3-D culture system for HNG cells. In this condition, morphogenesis of HNG cells was similar to the in situ condition. The differentiation and proliferation of HNG cells was probably related, at least in part, to EGF, KGF, and RA. Our culture system is useful to study effects of growth factors, cytokines, and various substances on the proliferation and differentiation of HNG cells in morphogenesis during development and remodeling under various pathological conditions.
Accepted for publication October 2, 2001.
This report was supported by a grant-in-aid for general scientific research (grant 10470355) from the Ministry of Education, Science, and Culture of Japan, Tokyo.
Corresponding author and reprints: Yuichi Majima, MD, Department of Otorhinolaryngology, Mie University School of Medicine, 2-174 Edobashi, Tsu, Mie 514-8507, Japan.