Effects of cells of epithelial rests of Malassez and endothelial cells on synthesis of glycosaminoglycans by periodontal ligament fibroblasts in vitro

Cultures of fibroblast-like cells (PLF) and epithelial rest cells (PLE) prepared from explants of porcine periodontal ligament synthesized and secreted four glycosaminoglycans (GAG) in differing proportions. The PLF produced predominantly chondroitin sulfate (greater than 60%) with smaller amounts of hyaluronic acid (HA) (17%), dermatan sulfate (13%), and heparan sulfate (7%), whereas PLE produced predominantly HA (greater than 80%). In coculture and under conditions of reciprocal transfer of conditioned media neither cell type affected the other's GAG synthesis. Endothelial cells (EC), however, or their conditioned growth media, were able to stimulate increased GAG synthesis, especially HA, in PLF. A similar result was obtained with smooth muscles cells (SMC) cultured in EC growth media but here again PLE were unable to stimulate GAG synthesis by SMC. These findings suggest that the spectrum of GAG found in whole ligament results both from independent production by, and from interaction between, the different cell types within the ligament. The results also provide support for a general hypothesis that loose connective tissues, which are rich in HA, are formed and maintained under the influence of epithelial, including endothelial, cells.     

Changes in nuclear non-histone protein composition during normal differentiation and carcinogenesis of intestinal epithelial cells

Nuclei from colonic epithelial cells were isolated and fractionated by centrifugation in discontinuous sucrose density gradients. Nuclei differing in buoyant density differ in size, non-histone protein to DNA ratio, and capacity for DNA synthesis in vivo. They do not differ in histone content or in proportions of the major histone classes. The distribution of cell nuclei after density gradient centrifugation corresponds functionally to their histological localization in the colonic mucosa, as judged by the nuclear capacity for DNA synthesis in both normal and tumor tissues. The nuclei of colonic epithelial cells contain a heterogeneous set of non-histone proteins which change in total amount and in relative proportions during normal differentiation. The complement of nuclear proteins differs in normal intestinal epithelial cells and in colon tumors induced by the carcinogen, 1,2-dimethylhydrazine. There is a striking increase in the nuclear content of two major protein classes (of mol. wt ca 44 000 and 62 000) during carcinogenesis. The accumulation of these proteins in the nuclei of carcinogen-treated cells follows early, selective increases in their rates of synthesis.     

Bystander cell proliferation is modulated by the number of adjacent cells that were exposed to ionizing radiation.

BACKGROUND: Direct cell-to-cell contact appears to be a prerequisite for the proliferative response of bystander WB-F344 cells co-cultured with irradiated cells; however, neither gap junctional intercellular communication nor long-range factors released into the medium appear to be involved (Cytometry 2003;56A:71-80). The present work investigated whether the proliferative bystander response depends on the number of irradiated cells (cells exposed to external gamma-rays or cells exposed to short-range beta-particles emitted by DNA-incorporated (3)H-thymidine) that are adjacent to unirradiated bystander cells. METHODS: Subconfluent monolayers of rat liver epithelial cells (WB-F344) were incubated in the presence of (methyl-(3)H)thymidine at a concentration of 5.8 kBq/ml for 18 h. Radiolabeled cells containing 0.7 x 10(-3) Bq/cell (absorbed dose: 0.14 Gy) were plated together with unlabeled cells in proportions of 6% and 94%, 12% and 88%, 25% and 75%, 50% and 50%, and 75% and 25%, respectively, keeping constant the total number of plated cells. In a parallel experiment, cells acutely exposed to 5 Gy of (137)Cs gamma-rays were plated with unirradiated cells in the same proportions. In both experiments, cells were co-cultured for 24 h followed by a flow cytometric study of their proliferation. The two cell populations in the co-cultures were distinguished by staining one population with carboxyfluorescein diacetate, succinimidyl ester, which metabolizes intracellularly. RESULTS: Increasing the fraction of irradiated cells relative to unirradiated bystander cells led to an increase in proliferation of bystander cells. Specifically, in co-cultures in which irradiated cells were initially mixed with unirradiated cells in proportions of 50% and 50% and of 75% and 25%, respectively, bystander cells showed a statistically significant increase of their proliferation compared with the controls. CONCLUSIONS: The proliferative response of WB-F344 bystander cells is modulated by the number of adjacent cells that are exposed to ionizing radiation from external gamma-rays or intracellularly emitted (3)H beta-particles.     

Immunohistochemical and histometrical studies of the human thymus with special emphasis on age-related changes in medullary epithelial and dendritic cells

Age-related changes in medullary epithelial and dendritic cells in the human thymus were examined quantitatively using immunohistochemistry and histometry. The material used was thymic biopsy specimens obtained from 105 patients during cardiac surgery, none of whom had immunological abnormalities. By using keratin and tissue polypeptide antigen markers to identify epithelial cells and S-100 protein and LN-2 markers to identify dendritic cells, the numbers of these cells in the medulla were counted morphometrically. The relative proportions of the cortex, medulla, Hassall's bodies, perivascular space, connective tissue and fatty tissue were evaluated using a computer image-analysis system and the respective relative volumes were calculated. Our results indicate that the number of medullary dendritic cells/mm2 and the relative volume of cortical thymocytes decrease steadily up to the age of 40 years, whereas no major change was found in the number of medullary epithelial cells/mm2.     

Immunohistochemical and histometrical studies of the human thymus with special emphasis on age-related changes in medullary epithelial and dendritic cells

Age-related changes in medullary epithelial and dendritic cells in the human thymus were examined quantitatively using immunohistochemistry and histometry. The material used was thymic biopsy specimens obtained from 105 patients during cardiac surgery, none of whom had immunological abnormalities. By using keratin and tissue polypeptide antigen markers to identify epithelial cells and S-100 protein and LN-2 markers to identify dendritic cells, the numbers of these cells in the medulla were counted morphometrically. The relative proportions of the cortex, medulla, Hassall’s bodies, perivascular space, connective tissue and fatty tissue were evaluated using a computer image-analysis system and the respective relative volumes were calculated. Our results indicate that the number of medullary dendritic cells/mm² and the relative volume of cortical thymocytes decrease steadily up to the age of 40 years, whereas no major change was found in the number of medullary epithelial cells/mm². Supported in part by a Grant-in-Aid (No. 61570664) for Scientific Research from the Japanese Ministry of Education     

高原鼢鼠发情周期阴道细胞变化特征及血清促性腺激素含量动态

In order to determine the change characteristics of vaginal cells in the estrus cycle and the duration of the estrus cycle, vaginal smears and HE staining methods were used to observe and count the types, morphological changes, and proportions of vaginal cells in plateau zokor (Eospalax baileyi). The hormone secretion dynamics of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) in the serum of plateau zokor in the estrus cycle were determined by ELISA. The results showed that the main vaginal cells of plateau zokor were leukocytes, nuclear epithelial cells, incompletely cornified epithelial cells, and complete cornified epithelial cells. The proportion of nuclear epithelial cells in the proestrus was significantly higher than that during the other three periods (48. 4 ±3. 09)% (n=12, P < 0. 05). The proportion of complete cornified epithelial cells in the oestrus was the highest and significantly higher than that in the other three periods (59. 73 ±7. 59)% (n=15, P < 0. 05). The proportion of leukocytes in the metestrus and dioestrum was significantly higher than that in the proestrus and oestrus (P < 0. 05). The expression level of LH was the highest in the metestrus (4. 709 5 ±1. 094 0) ng/mL (n=6, P < 0. 05), which showed an increasing trend in proestrus and oestrus, and a decreasing trend in the metestrus and dioestrum. There was no significant difference in the expression level of FSH among the four periods (P > 0. 05); the estrus cycle duration of plateau zokor was 16 to 19 d, the characteristics of vaginal cell types in different stages of the estrus cycle were obvious, and the relative number was statistically significant. The study provided a basis for vaginal smear identification in the estrous cycle and reproductive physiology study of plateau zokor.     

Separation of epithelial cells from suspensions of cells from the hamster parotid gland in an isokinetic density gradient of Ficoll in tissue culture medium.

Epithelial cells were separated from suspensions of hamster parotid cells by velocity sedimentation in an isokinetic gradient and by isopycnic sedimentation. Epithelial cells were 48.1 ± 18.0% of the cells in the starting sample suspensions of cells from the disaggregated hamster parotid glands. The purest gradient fractions following velocity sedimentation in a previously described isokinetic gradient contained 98.8 ± 1.8% epithelial cells. The purest fractions obtained from isopycnic sedimentation contained 99.9 ± 0.2% epithelial cells. Purification of parotid epithelial cells by velocity sedimentation in the isokinetic gradient seems preferable to purification using isopycnic centrifugation because a larger proportion of the epithelial cells are obtained in the zone of the gradient which contains highly purified epithelial cells and because velocity sedimentation requires lower centrifugal forces for a shorter period of time.     

Equimolar proportions of somatic-type,core histones in chromatin of spermatogonia

[³H]Leucine incorporation into histones of seminiferous epithelial cells of hypophysectomized rats was used to calculate the molar proportions of the core histones of spermatogonia. The molar proportions H3:H2B:(H2A + protein A24):H4 are 1:1:1:1, viz. identical with those reported by others for somatic cells. Similar results were obtained when molar proportions of histones of seminiferous epithelial cells from immature rat testis (predominantly populated with spermatogonia) were determined by the dye-binding method. These data are relevant to mechanisms for the replacement of some of the core histones by variants during the primary spermatocyte stages.     

Molar proportions and relative rates of synthesis of histones in rat testis

The molar proportions and relative rates of synthesis of histones in normal and hypophysectomized rat testis seminiferous epithelial cells were determined. After hypophysectomy the molar proportions of histones H1, H2B and (H2A + protein A24) in seminiferous epithelial cells of rat testis increased while their corresponding variants TH1-x, TH2B-x and X₂ decreased, but the molar proportions of major-class histones (i.e., sum of subfractions) remained relatively constant and similar to the proportions in somatic cells. The apparent molar proportions of the labeled histones, determined immediately after 2-h periods of [³H]leucine incorporation, were much higher relative to H4 than the proportions of total histones determined by dye binding. The values, however, approached the molar proportions of total histones when rats were killed 11 days after the [³H]leucine injection. Two-dimensional gel electrophoresis confirmed that the high initial molar proportions relative to H4 by [³H]leucine incorporation were not due to the possible contamination by highly-labeled non-histone proteins. The specific activity of histone H4 relative to the specific activity of DNA, determined immediately after 3-h periods of [³H]leucine and [¹⁴C]thymidine incorporations was similar to the value when rats were killed 13 days after the injections. It is proposed that histones of seminiferous epithelial cells are synthesized disproportionally relative to H4 and in excess of the quantities required for polynucleosome assembly. The excess histones are subsequently displaced or degraded slowly.     

Postfreeze viability of human renal epithelial carcinoma cells in quaternary solutions

The postfreeze survival of human renal epithelial carcinoma cells frozen in suspensions based on (MEM + FCS) (Eagle's Minimum Essential Medium with Hanks' salts added and supplemented with 20 vol% heat-inactivated fetal calf serum) to which additions of NaCl and the cryoprotective compounds, dimethyl sulfoxide and glycerol, were made, have been determined by a vital dye exclusion technique. A significant range of aqueous-rich quaternary solutions were found to be highly effective cryoprotective agents for these cells under optimum freezing and thawing conditions.High survival occurs in a composition series emanating from the region of high postfreeze cell viability in the ternary (MEM + FCS)-NaCl-DMSO system, between 80 and 95% (MEM + FCS), along isopleths that substitute glycerol for NaCl in various proportions. In addition, viability remains sufficiently high as a few percentage of glycerol is substituted for DMSO alone.     

Intrinsic epithelial cells repair the kidney after injury

Understanding the mechanisms of nephron repair is critical for the design of new therapeutic approaches to treat kidney disease. The kidney can repair after even a severe insult, but whether adult stem or progenitor cells contribute to epithelial renewal after injury and the cellular origin of regenerating cells remain controversial. Using genetic fate-mapping techniques, we generated transgenic mice in which 94%-95% of tubular epithelial cells, but no interstitial cells, were labeled with either beta-galactosidase (lacZ) or red fluorescent protein (RFP). Two days after ischemia-reperfusion injury (IRI), 50.5% of outer medullary epithelial cells coexpress Ki67 and RFP, indicating that differentiated epithelial cells that survived injury undergo proliferative expansion. After repair was complete, 66.9% of epithelial cells had incorporated BrdU, compared to only 3.5% of cells in the uninjured kidney. Despite this extensive cell proliferation, no dilution of either cell-fate marker was observed after repair. These results indicate that regeneration by surviving tubular epithelial cells is the predominant mechanism of repair after ischemic tubular injury in the adult mammalian kidney.     

Genotyping accuracy for whole-genome amplification of DNA from buccal epithelial cells.

We compared the accuracy of genotyping for DNA extracted from lymphocytes to that of DNA amplified from buccal epithelial cells. Amplification was via a rolling circle/phi29 DNA polymerase commercial kit. Paired buccal and lymphocyte DNA samples were available from 30 individuals. All samples were genotyped for 12 SNPs, 5 microsatellites and 2 VNTRs. The accuracy of genotyping (no-call proportions, reproducibility, and concordance) was similar for DNA from lymphocytes in comparison to amplified DNA from buccal samples. If used with caution, these data suggest that rolling-circle whole-genome amplification can be used to increase the DNA mass available for large-scale genotyping projects based on DNA from buccal cells.     

Normal human endometrial cells in culture: Characterization and immortalization of epithelial and stromal cells by SV 40 large T antigen

Summary— Thirty endometrial biopsies were cultured in order to separate stromal and epithelial cells. Using epidermal growth factor (EGF), cortisol, cholera toxin, insulin with 5% horse serum for epithelial cells or a medium with 20% fetal calf serum for stromal cells, we could specifically enrich endometrial culture in epithelial or stromal cells and culture them for 1 or 2 months. These cultures retained the phenotypic characteristics of epithelial (cytokeratins, mucin HMFG 1) and stromal (vimentin, smooth muscle actin, desmin) lineage by immunostaining analysis. Epithelial and stromal cultures from one individual were respectively immortalized by the SV 40 large T antigen. The immortalized cell lines kept the phenotype of the normal cells from which they derived.     

Clonogenicity of human endometrial epithelial and stromal cells

The human endometrium regenerates from the lower basalis layer, a germinal compartment that persists after menstruation to give rise to the new upper functionalis layer. Because adult stem cells are present in tissues that undergo regeneration, we hypothesized that human endometrium contains small populations of epithelial and stromal stem cells responsible for cyclical regeneration of endometrial glands and stroma and that these cells would exhibit clonogenicity, a stem-cell property. The aims of this study were to determine 1) the clonogenic activity of human endometrial epithelial and stromal cells, 2) which growth factors support this clonogenic activity, and 3) determine the cellular phenotypes of the clones. Endometrial tissue was obtained from women undergoing hysterectomy. Purified single- cell suspensions of epithelial and stromal cells were cultured at cloning density (300-500/cm(2)) in serum medium or in serum- free medium supplemented with one of eight growth factors. Small numbers of epithelial (0.22%) and stromal cells (1.25%) initiated colonies in serum-containing medium. The majority of colonies were small, containing large, loosely arranged cells, and 37% of epithelial and 1 in 60 of stromal colonies were classified as large, comprising small, densely packed cells. In serum-free medium, transforming growth factor-alpha (TGF alpha), epidermal growth factor (EGF), platelet-derived growth factor-BB (PDGF-BB) strongly supported clonogenicity of epithelial cells, while leukemia-inhibitory factor (LIF), hepatocyte growth factor (HGF), stem-cell factor (SCF), insulin-like growth factor-I (IGF- I) were weakly supportive, and basic fibroblast growth factor (bFGF) was without effect. TGF alpha, EGF, PDGF-BB, and bFGF supported stromal cell clonogenicity, while HGF, SCF, LIF, and IGF- I were without effect. Small epithelial colonies expressed three epithelial markers but not stromal markers; however, large epithelial colonies showed little reactivity for all markers except alpha(6)-integrin. All stromal colonies contained fibroblasts, expressing stromal markers, and in some colonies, myofibroblasts were also identified. This analysis of human endometrium has demonstrated the presence of rare clonogenic epithelial and stromal cells with high proliferative potential, providing the first evidence for the existence of putative endometrial epithelial and stromal stem cells.     

Control cells for image cytometric DNA analysis of esophageal tissue and the influence of preoperative treatment.

The nuclear DNA content of 4,960 normal cells in 30 esophagectomy specimens was assessed by image cytometry. In relation to intermediate squamous epithelial cells, the mean transmission values were 25.6% higher for basal squamous epithelial cells, 10.1% higher for fibroblasts and 16.2% lower for lymphocytes. The interslide coefficient of variation (CV) was similar for intermediate and basal squamous epithelial cells, 20.2% and 18.9%, respectively, while fibroblasts and lymphocytes displayed lower interslide CVs, 10.5% and 8.8%, respectively. In spite of their high interslide CVs, the mean DNA values of intermediate and basal squamous epithelial cells had the highest intraslide correlation. This correlation indicates that the preparative and analytical differences between slides that result in alterations in mean DNA values affect squamous epithelial cells of similar origin, size and shape in an analogous manner. Basal squamous epithelial cells replicate their DNA and contain more than the diploid amount; therefore, normal intermediate squamous epithelial cells are suggested as control cells to define the 2c reference value for further DNA studies on squamous epithelial lesions of the esophagus. Preoperative irradiation, with or without chemotherapy, did not alter the DNA values of the investigated control cells.     

In vivo DNA damage in gastric epithelial cells

A number of risk factors have been linked epidemiologically with gastric cancer, but studies of DNA damage in gastric epithelial cells are limited. The comet assay is a simple technique for determining levels of DNA damage in individual cells. In this study, we have validated the comet assay for use in epithelial cells derived directly from human gastric biopsies, determined optimal conditions for biopsy digestion and investigated the effects of oxidative stress and digestion time on DNA damage. Biopsies taken at endoscopy were digested using combinations of pronase and collagenase, ethylenediaminetetra-acetic acid (EDTA) and vigorous shaking. The resultant cell suspension was assessed for cell concentration and epithelial cell and leukocyte content. A score for DNA damage, the comet %, was derived from the cell suspension, and the effect of various digestion conditions was studied. Cells were incubated with H(2)O(2) and DNA damage was assessed. Pronase and collagenase provided optimum digestion conditions, releasing 1. 12x10(5) cells per biopsy, predominantly epithelial. Of the 23 suspensions examined, all but three had leukocyte concentrations of less than 20%. The comet assay had high inter-observer (6.1%) and inter-assay (4.5%) reproducibility. Overnight storage of the biopsy at 4 degrees C had no significant effect on DNA migration. Comet % increased from a median of 46% in untreated cells to 88% in cells incubated for 45 min in H(2)O(2) (p=0.005). Serial 25-min digestions were performed on biopsies from 13 patients to release cells from successively deeper levels in the crypt. Levels of DNA migration were significantly lower with each digestion (r=-0.94, p<0.001), suggesting that DNA damage is lower in younger cells released from low in the gastric crypt. The comet assay is a reproducible measure of DNA damage in gastric epithelial cells. Damage accumulates in older, more superficial cells, and can be induced by oxidative stress.     

Flow cytometric enrichment for respiratory epithelial cells in sputum.

BACKGROUND: Induced sputum, in contrast to bronchoscopic biopsies and lavages, is an easily obtained source of biological specimens. However, obtaining abnormal exfoliated cells for detailed molecular studies is limited because respiratory epithelial cells comprise only about 1% of sputum cell populations. METHODS: We developed a multiparameter flow sorting strategy to purify epithelial cells from nonepithelial sputum cells, using anti-cytokeratin antibody AE1/AE3 to recognize human epithelial cells and DAPI to stain DNA. We excluded cells with a high degree of side-scatter, which were composed predominantly of squamous cells and contaminating macrophages. The remaining cytokeratin-positive respiratory epithelial cells were then sorted based on anti-cytokeratin (PE) vs DNA (DAPI) parameters. RESULTS: In this proof of principle study, the AE1AE3 cytokeratin/DNA flow sorting strategy enriched rare diploid respiratory epithelial cells from an average of 1.1% of cells in unsorted induced sputum samples to average purities of 42%. Thus, AE1AE3 flow-sorting results in a 38-fold enrichment of these cells. CONCLUSIONS: We report a multiparameter flow cytometric assay to detect and enrich rare respiratory epithelial cells from induced sputum samples to average purities of 42%. With further development, this methodology may be useful as part of a molecular screening approach of populations at high risk for lung cancer.     

Isolation and culture of primary equine tracheal epithelial cells

Culture of airway epithelial cells is a useful model to investigate physiology of airway epithelia and airway disease mechanisms. In vitro models of airway epithelial cells are established for various species. However, earlier published method for isolation and culture of equine tracheal epithelial cells requires significant improvements. In this report, the development of a procedure for efficient isolation, characterization, culture, and passage of primary equine tracheal epithelial cells are described. Epithelial cells were isolated from adult equine trachea by exposing and stripping the mucosal epithelium from the adjacent connective tissue and smooth muscle. The tissue was minced and dissociated enzymatically using 0.25% trypsin-ethylenediaminetetraacetic acid (EDTA) solution for 2 h at 37 degrees C. Cells were collected by sieving and centrifugation, and contaminating fibroblasts were removed by differential adhesion. This procedure resulted in a typical yield of 1 x 10(7) cytokeratin-positive epithelial cells per gram tracheal lining tissue. Viability was 95% by trypan blue exclusion and isolates contained approximately 94% cytokeratin-positive cells of epithelial origin. Cells seeded at a density of 6.9 x 10(4) cells/cm(2) in serum-free airway epithelial cell growth medium formed monolayers near confluency within a week. Confluent cells were dissociated using dispase II and first passages (P1) and second passages (P2) were successfully established in serum-free medium. Collagen coating of tissue culture flask was not required for cell adhesion, and cultures could be maintained at the level of P2 over 30 d. In the present study, we could establish a high-yield protocol for isolation and culture of equine tracheal epithelial cells that can serve for in vitro/ex vivo studies on the (patho-)physiology of equine airway disease as well as pharmacological and toxicological targets relevant to airway diseases.     

Morphological features of principal cells in the ovine epididymis: a quantitative and qualitative study

Functions of the epididymis differ by region, and this may be reflected in epithelial structure. Therefore, tissues from the initial segment (IS), proximal and central caput (PCap, CCap), and proximal and central corpus (PCor, CCor) epididymidis were examined by light and transmission electron microscopy. The proportion of principal cells in the epithelium was highest (p less than 0.05) in the CCap (74%) and lowest in the CCor (68%), whereas proportions of basal cells (25%), apical cells (1.4%), and white blood cells (2%) were similar in all regions. Volume density (VD) of the nucleus was lower (p less than 0.05) in principal cells in the IS (7%) than in other regions (10%). There was no regional difference in VD of the Golgi complex (14%) or endoplasmic reticulum (19%) in principal cells. The VD of mitochondria averaged 4% in the IS through CCap, but only 2.5% in PCor or CCor (p less than 0.05). The VD of clear vesicles + multivesicular bodies (8%) and dense vesicles (6%) were higher (p less than 0.05) in the CCap than in other regions (1% each), while there were more lipid droplets (12%) in the PCor than in other regions (less than or equal to 2%). Most quantitative differences in VD of organelles within principal cells were small even though significant. However, there were profound differences in the morphological features of the Golgi complex, endoplasmic reticulum, and mitochondria among regions.     

Membrane antigens of human bronchial epithelial cells identified by monoclonal antibodies

Summary Subpopulations of normal bronchial epithelial cells were identified using a series of murine monoclonal antibodies. These antibodies were used to stain intact bronchial epithelial cells in culture by indirect immunofluorescence. LAM 2 reacted with 80%, LAM 6 with 75%, LAM 7 with 60%, and LAM 8 with 5% of these cells. Sections of human bronchial epithelium were also stained with these antibodies by immunoperoxidase methods. LAM 2 was found to bind with 80%, LAM 6 with 65%, LAM 7 with 50%, and LAM 8 with less than 1% of bronchial epithelial cells. LAM 2 stained both columnar epithelial cells and basal cells; LAM 6 stained mainly basal cells and only a small proportion of columnar cells; LAM 7 showed specificity for basal cells; LAM 8 distinctly stained single cells in the basal cell layer. These antibodies were previously shown to react with the surface membrane of human lung carcinomas, ranging from the broad reactivity of LAM 2 with small cell and non-small cell lung cancers to the highly restricted reactivity of LAM 8 with small cell carcinomas of the lung. Thus, membrane antigens have been identified in bronchial epithelial cells by monoclonal antibodies which exhibit a similar range of cellular reactivity in vitro as in vivo. Inasmuch as these antibodies recognize subsets of cells which could not be easily distinguished by morphologic characteristics, these reagents may be useful in classifying bronchial epithelial cells.