Respiratory Virus Infection of Monolayer Cultures of Human Nasal Epithelial Cells1-3
BIRGIT WINTHER,4 JACK M. GWALTNEY, and J. OWEN HENDLEY Introduction SUMMARY The effect on nasal epithelial cells in monolayer cultures of infection with rhinovirus, Human infections with respiratory vicoronavirus 229E, influenza type A, and adenovirus was studied. Fragments (1 )( 2 mm) of epitheliruses are common. Information on the um with adherent submucosa from nasal polyps, adenoids, or nasal turbinates were cultured in effect of viral replication on respiratory plastic dishes with media containing a serum supplement. Within a week a monolayer of epithelial epithelium in vivo is sparse. Biopsies of cells with interspersed ciliated cells surrounded each fragment. Epithelial monolayers were exposed the nasal mucosa of patients reporting to 10' to 10' TCIDso/ml of each virus; unattached virus was removed after 3 h. Replication of virus symptoms of a cold have been examined was examined by doing viral t1trations of the media daily, and ciliated and nonciliated epithelial in two studies. In the 1930s, Hilding (1) cells In the monolayer were examined morphologically with an inverted microscope. showed marked destruction of the nasal The peak titer of rhinovirus and coronavirus occurred 24 to 48 h after Inoculation, indicating that viral replication was occurring. However, no detectable damage or cytopathic effect (CPE) develepithelium of people with colds, whereoped in the monolayer of epithelial cells when compared with that in unInfected control cells. In as Winther and coworkers (2) found that contrast, infection of epithelial monolayers with Influenza or adenovirus resulted in CPE and dethe nasal epithelium appeared normal struction of the monolayer. The marked difference In the effect of these respiratory viruses on nasal in biopsies examined with the scanning epithelial cells may reflect similar differences in their effect on respiratory tract mucosa In the intact electron microscope. Both studies used pahost. If so, the pathogenesis of symptom production during rhinovirus and coronavlrus infections tients with naturally acquired colds in must be by mechanisms other than destruction of the nasal lining. whom the viral etiology was not deterAM REV RESPIR DIS 1990; 141:839-845 mined. Turner and colleagues (3) employed volunteers with experimental rhinovirus colds in order to examinethe types of cells in nasal secretions during illness. adenovirus produced destruction of the maximize attachment of the fragments to the They found that 1 to 2070 of the cells in monolayer of epithelial cells, whereas plastic. Fragments in each dish were then covnasal secretions from rhinovirus-infected rhinovirus and coronavirus had no dis- ered with 1.5ml of medium consisting of Easubjects were sloughed ciliated cells and cernible destructive effect of the epithe- gle's minimum essential medium with Earle's salts and o-valine (Gibco Laboratories, Grand that there were none in uninfected con- lial monolayers during viral growth. Island, NY), 10010 medium supplement (Nutrol subjects. Serums; Collaborative Research, Lexington, In vitro study of respiratory virus inMethods MA), and antibiotics (as above). Medium confection requires the use of human tissue taining o-valine instead of L-valine was used Nasal Mucosal Specimens since rhinovirus replicates only in higher Specimens of nasal mucosa from turbinates, in order to inhibit growth of fibroblasts (7). primates (4). Hoorn and Tyrrell (5) cul- polyps, or adenoids were obtained from 74 In preliminary studies, the medium suppletured fragments of human fetal trachea patients who were undergoing elective surgery ment, which contained several hormones and and found ciliated activity on the edge for benign conditions such as nasal obstruc- growth factors, was found to be as good for of the fragments despite rhinovirus repli- tion, allergy, or secretory otitis. Use of these cation in the cultures. However, evalua- tissues, which were to be discarded, was aption of the ciliary activity on the edge proved by the Human Investigation Commit(Received in original form April 11, 1989 and in of fragments did not allow a distinct view tee at the University of Virginia. Each0 speci- revised form August 9, 1989) men of nasal mucosa was chilled to 4 C imof single ciliated cells. Hamory and co- mediately after excision in Hank's balanced workers (6)inoculated cultured fragments salt solution (HBSS) containing penicillin (100 From the Department of Otorhinolaryngoloof human nasal polyps with rhinovirus ug/rnl), streptomycin (100 ug/ml), and am- gy, Rigshospitalet, University of Copenhagen, Coand studied the epithelium after fixation photericin B (1 ug/rnl), and was prepared penhagen, Denmark; the Departments of OtorHead and Neck Surgery, and and staining. No histologic changes were for culture within 6 h after removal from the hinolaryngology, Internal Medicine and Pediatrics, University of Virdemonstrated in rhinovirus-infected pol- patient. ginia School of Medicine, Charlottesville, Virginia. yp fragments when compared with non2 Supported by a grant from the Vicks Health Cultures of Mucosal Fragments Care Division, Richardson-Vicks, Inc., Shelton, infected polyps. In the current study, we developed a Areas of intact nasal mucosal surface identi- Connecticut, and by the Pendleton Laboratory Endowment Fund. modified organ culture system for pro- fied microscopically were dissected free from 3 Correspondence and requests for reprints the underlying tissue. The epithelium with adduction of a monolayer of differentiat- herent submucosa was cut into fragments should be addressed to Dr. J. Owen Hendley, ed respiratory epithelial cells from frag- 1 x 2 x 3 mm. Fragments, seven to a dish, Department of Pediatrics, Box 386, University of ments of nasal mucosa. This allowed di- were placed on the dry surfaces of plastic Virginia School of Medicine, Charlottesville, VA rect observation of single epithelial cells tissue culture dishes 60 x 15 mm (Falcon 22908. • Recipient of a Research Fellowship at the Univerin the monolayer during infection with Primaria'"; Becton Dickinson, Oxnard, CA) sity of Copenhagen. Travel expenses were granted respiratory viruses. Influenza virus and and left for 20 min at room temperature to by the Danish Pasteur Society. 1
839
840
maintaining epithelial outgrowth as 10or 2070 fetal calf serum. Dishes containing the nasal fragments were incubated at 35° C in an atmosphere of 5% CO 2 and 100% humidity in a sealed tissue culture box (CBS Scientific Co., Del Mar, CA). The medium in the dishes was changed every 3 to 5 days. The dishes in the tissue culture box wereincubated in a stationary position since side-to-side rocking of the box (and dishes) to an angle of 30 degrees at a frequency of 3 cycles/min did not appear to enhance viral growth or development of viral cytopathic effect (CPE).
Viral Inoculum Pools Pools of rhinovirus, coronavirus, influenza A, and adenovirus wereused to inoculate nasal epithelial fragment cultures. Laboratory strains of rhinovirus (types 10, 14,and 39 and strain HH, which is presently untyped) and of coronavirus 229E wereharvested after three to five passages in human embryonic lung fibroblast cells (MRC-5 and/or WI-38; M.A. Bioproducts, Walkersville, MD). Rhinovirus type 39 in nasal mucus was obtained from two volunteers with experimental rhinovirus colds. High titered pools of rhinovirus type 14 and type 39 passaged in Hela cells were kindly donated by Dr. Roland Rueckert (University of Wisconsin). Clinical isolates of influenza type A and adenovirus were provided by Dr. Frederick G. Hayden (Viral Diagnostic Laboratory, University of Virginia). The pool of influenza virus was harvested from rhesus monkey kidney cells (MK; M.A. Bioproducts) after three passages, and the adenovirus was harvested from human embryonic kidneycells(HEK; M.A. Bioproducts) after one passage. Inoculation of Nasal Cultures with Virus A O.lO-ml suspension containing 104 to 105 tissue culture infectious doses 50 (TCID50/ml) of virus was inoculated into each dish so that the nasal fragments were exposed to UP or 104 TCID 50/ml. After 3 h of exposure, unattached virus was removed by washing each dish four times with 1.5 ml of HBSS. Control dishes, which were not exposed to virus, were washed in a similar fashion. After washing, dishes were refed with 1.5 ml of medium and incubated. The medium in the dishes was changed again at 48 h after exposure to virus. In each experiment, two dishes with nasal fragments wereexposed to virus and one dish with fragments from the same patient served as the control. In order to quantitate growth of virus a O.l5-ml aliquot of medium was removed from each dish just before washing, at the time of the fourth wash, and daily after virus inoculation. The samples from the virus-exposed dishes were pooled for virus titration. Infectivity titrations of medium from culture dishes exposed to rhinovirus or coronavirus were performed in MRC-5 or WI-38 fibroblast cultures in screw-capped tubes. Infectivity titrations of adenovirus were done
WINTHER, GWALTNEY, AND HENDLEY
in HEK cells and of influenza type A in MK cells with hem absorption after 1 and 2 wk.
Evaluation of the Effect of Viral Infection on Fragments and Epithelial Monolayers Each nasal fragment and the surrounding monolayer of epithelial cells that had grown out from the fragment were examined at a magnification x 100 with an inverted microscope. Before viral inoculation, a detailed map of the topographic characteristics of each fragment and of the surrounding monolayer was drawn, which made it possible to identify the same fragment and cellsin the monolayer every day. Prior to and daily after inoculation the morphologic features of the adjacent epithelial outgrowth were described (1) by grading the extent to which the fragment was surrounded by monolayer extending 3 mm from the fragment (as +, + +, or + + + ), (2) by noting the proportion of the monolayer in which there were cells with beating cilia, and (3) by noting morphologic changes (cytopathic effect) of the epithelial cells. In addition, the ciliary activity of each fragment was estimated by recording in percent the proportion of the edge of the fragment with ciliary motion. Ciliary activity on the fragments and morphologic features of the monolayer of epithelial cells were evaluated without knowledge of whether the dish had been exposed to virus or not. Histologic Examination of Nasal Fragments Nasal fragments from infected and corresponding noninfected control cultures were fixed with 10% formalin at 96, 120, or 168h after virus inoculation. The fragments were embedded in paraffin, cut, and stained with hematoxylin-eosin. The histologic appearance of the epithelium on fragments was described on coded slides. Inoculation of Lymphocyte and Monocyte Suspensions with Rhinovirus Suspensions of lymphocytes and monoeytes in culture (donated by Dr. Richard Pearson, Department of Internal Medicine, University of Virginia) were prepared from peripheral blood of healthy adult donors. Mononuclear cells separated with Ficoll-sodium diatrizoate (Sigma Chemical Co., St. Louis, MO) sedimentation were suspended at a concentration of 10' cells/ml in Medium 199(Gibco Laboratories) and plated onto sterile plastic petri dishes. Nonadherent cells,which were 90% lymphocytes, were removed after 2 h. The dishes were washed, and after 3 h the adhering mononuclear cells were released with a rubber policeman. The cells were suspended in Medium 199with 10% fetal calf serum without antibiotics. The lymphocyte suspension contained an averageof 8 x 10" cells/ml, and the monocyte suspension contained 5 x 105 monocytes/ml. The cell suspensions were maintained in 36 x 27 mm Teflon'"vials (Scientific SpecialtiesServices, Inc., Randallstown,
MD) at 35° C in a 5% CO2 atmosphere,Rhinovirus type 10 (105 TCID 50/ml) was inoculated into each cell suspension. An aliquot of each inoculated cell suspension was removed at 24 and 72 h for virus titration in MRC-5 cell culture tubes. Each aliquot was titered both unfrozen and after 3 freeze-thaw cycles to release intracellular virus.
Results
Establishment of Monolayers of Nasal Epithelial Cells Nasal polyp specimens from 40 patients, 30 specimens of adenoid tissue, and four nasal turbinates were prepared for culture. Fragment cultures from 36 (900/0) of the nasal polyps, 14(47%) of the adenoids, and three (75%) of the nasal turbinates had surrounding outgrowth of a confluent monolayer of cuboidal cells within a week after initiation (figure 1A). The leading edge of the outgrowth extended approximately 2 to 3 mm from the fragment. Ciliated cells, which constituted 2 to 20% of the monolayer, were either concentrated in localized areas or scattered throughout the monolayer. Ciliated activity in the monolayer continued in cultures maintained for a month. Beginning approximately 3 wk or more after initiation, groups of spindle-shaped cells oriented in parallel to each other wereseen in some cultures. These spindleshaped cells were considered to be fibroblasts. The fibroblasts could be easily distinguished from the ciliated epithelial cells and the nonciliated cuboidal cells. A definite interface was usually present between the epithelial cells and fibroblast types (figure 1B). Two preliminary experimentsweredone to determine whether outgrowth of the ciliated and cuboidal cells was enhanced by culturing mucosal fragments on a layer of collagen matrix (Vitrogens 100,Collagen Corporation, Palo Alto, CA) or on an extracellular matrix elaborated by cultured endothelial cells from human umbilical cord vein (8). Neither matrix was superior to the plastic (Primarias; Falcon) with regard to rapidity of appearance of the outgrowth, extent of monolayer, or maintenance of ciliary activity by cells. As a consequence, fragments attached to plastic dishes were used in all subsequent experiments. Growth of Virus and Detection of CPE in Nasal Epithelial Monolayers The growth 0 f respiratory viruses in fragment/monolayer cultures and production of viral CPE in the monolayers was examined in 29 experiments with fragment/
841
RESPIRATORY VIRAL INFECTION OF HUMAN NASAL EPITHELIAL CELLS
Fig. 1. Outgrowth of monolayer of epithelial cells from fragments of nasal polyp mucosa (phase contrast; magnification: x100). Beating cilia, which are seen in the fresh state, are not visible in photomicrographs. A. Seven days in culture; the fragment is seen to the left of the epithelial cell monolayer (EC). B. Three weeks in culture. Interface between epithelial cell monolayer (EC) and spindle-shaped fibroblasts (FB) is shown.
Fig. 2. Viral growth curves, assessed by titers of viable virus in media, in culture dishes inoculated with rhinovirus type 10. Dishes were washed to remove unattached virus 3 h after inoculation. Arrows denote virus inoculation of dishes. A. Two dishes containing mucosal fragments and epithelial cell monolayer from adenoid. B. Two dishes containing epithelial cell monolayer from adenoid without fragments. Fragments were removed 24 h prior to inoculation. C. Two dishes containing media without fragments or outgrowth. Dishes were not washed 3 h after inoculation.
4
0 0
/_0
E
BIRGIT WINTHER,4 JACK M. GWALTNEY, and J. OWEN HENDLEY Introduction SUMMARY The effect on nasal epithelial cells in monolayer cultures of infection with rhinovirus, Human infections with respiratory vicoronavirus 229E, influenza type A, and adenovirus was studied. Fragments (1 )( 2 mm) of epitheliruses are common. Information on the um with adherent submucosa from nasal polyps, adenoids, or nasal turbinates were cultured in effect of viral replication on respiratory plastic dishes with media containing a serum supplement. Within a week a monolayer of epithelial epithelium in vivo is sparse. Biopsies of cells with interspersed ciliated cells surrounded each fragment. Epithelial monolayers were exposed the nasal mucosa of patients reporting to 10' to 10' TCIDso/ml of each virus; unattached virus was removed after 3 h. Replication of virus symptoms of a cold have been examined was examined by doing viral t1trations of the media daily, and ciliated and nonciliated epithelial in two studies. In the 1930s, Hilding (1) cells In the monolayer were examined morphologically with an inverted microscope. showed marked destruction of the nasal The peak titer of rhinovirus and coronavirus occurred 24 to 48 h after Inoculation, indicating that viral replication was occurring. However, no detectable damage or cytopathic effect (CPE) develepithelium of people with colds, whereoped in the monolayer of epithelial cells when compared with that in unInfected control cells. In as Winther and coworkers (2) found that contrast, infection of epithelial monolayers with Influenza or adenovirus resulted in CPE and dethe nasal epithelium appeared normal struction of the monolayer. The marked difference In the effect of these respiratory viruses on nasal in biopsies examined with the scanning epithelial cells may reflect similar differences in their effect on respiratory tract mucosa In the intact electron microscope. Both studies used pahost. If so, the pathogenesis of symptom production during rhinovirus and coronavlrus infections tients with naturally acquired colds in must be by mechanisms other than destruction of the nasal lining. whom the viral etiology was not deterAM REV RESPIR DIS 1990; 141:839-845 mined. Turner and colleagues (3) employed volunteers with experimental rhinovirus colds in order to examinethe types of cells in nasal secretions during illness. adenovirus produced destruction of the maximize attachment of the fragments to the They found that 1 to 2070 of the cells in monolayer of epithelial cells, whereas plastic. Fragments in each dish were then covnasal secretions from rhinovirus-infected rhinovirus and coronavirus had no dis- ered with 1.5ml of medium consisting of Easubjects were sloughed ciliated cells and cernible destructive effect of the epithe- gle's minimum essential medium with Earle's salts and o-valine (Gibco Laboratories, Grand that there were none in uninfected con- lial monolayers during viral growth. Island, NY), 10010 medium supplement (Nutrol subjects. Serums; Collaborative Research, Lexington, In vitro study of respiratory virus inMethods MA), and antibiotics (as above). Medium confection requires the use of human tissue taining o-valine instead of L-valine was used Nasal Mucosal Specimens since rhinovirus replicates only in higher Specimens of nasal mucosa from turbinates, in order to inhibit growth of fibroblasts (7). primates (4). Hoorn and Tyrrell (5) cul- polyps, or adenoids were obtained from 74 In preliminary studies, the medium suppletured fragments of human fetal trachea patients who were undergoing elective surgery ment, which contained several hormones and and found ciliated activity on the edge for benign conditions such as nasal obstruc- growth factors, was found to be as good for of the fragments despite rhinovirus repli- tion, allergy, or secretory otitis. Use of these cation in the cultures. However, evalua- tissues, which were to be discarded, was aption of the ciliary activity on the edge proved by the Human Investigation Commit(Received in original form April 11, 1989 and in of fragments did not allow a distinct view tee at the University of Virginia. Each0 speci- revised form August 9, 1989) men of nasal mucosa was chilled to 4 C imof single ciliated cells. Hamory and co- mediately after excision in Hank's balanced workers (6)inoculated cultured fragments salt solution (HBSS) containing penicillin (100 From the Department of Otorhinolaryngoloof human nasal polyps with rhinovirus ug/rnl), streptomycin (100 ug/ml), and am- gy, Rigshospitalet, University of Copenhagen, Coand studied the epithelium after fixation photericin B (1 ug/rnl), and was prepared penhagen, Denmark; the Departments of OtorHead and Neck Surgery, and and staining. No histologic changes were for culture within 6 h after removal from the hinolaryngology, Internal Medicine and Pediatrics, University of Virdemonstrated in rhinovirus-infected pol- patient. ginia School of Medicine, Charlottesville, Virginia. yp fragments when compared with non2 Supported by a grant from the Vicks Health Cultures of Mucosal Fragments Care Division, Richardson-Vicks, Inc., Shelton, infected polyps. In the current study, we developed a Areas of intact nasal mucosal surface identi- Connecticut, and by the Pendleton Laboratory Endowment Fund. modified organ culture system for pro- fied microscopically were dissected free from 3 Correspondence and requests for reprints the underlying tissue. The epithelium with adduction of a monolayer of differentiat- herent submucosa was cut into fragments should be addressed to Dr. J. Owen Hendley, ed respiratory epithelial cells from frag- 1 x 2 x 3 mm. Fragments, seven to a dish, Department of Pediatrics, Box 386, University of ments of nasal mucosa. This allowed di- were placed on the dry surfaces of plastic Virginia School of Medicine, Charlottesville, VA rect observation of single epithelial cells tissue culture dishes 60 x 15 mm (Falcon 22908. • Recipient of a Research Fellowship at the Univerin the monolayer during infection with Primaria'"; Becton Dickinson, Oxnard, CA) sity of Copenhagen. Travel expenses were granted respiratory viruses. Influenza virus and and left for 20 min at room temperature to by the Danish Pasteur Society. 1
839
840
maintaining epithelial outgrowth as 10or 2070 fetal calf serum. Dishes containing the nasal fragments were incubated at 35° C in an atmosphere of 5% CO 2 and 100% humidity in a sealed tissue culture box (CBS Scientific Co., Del Mar, CA). The medium in the dishes was changed every 3 to 5 days. The dishes in the tissue culture box wereincubated in a stationary position since side-to-side rocking of the box (and dishes) to an angle of 30 degrees at a frequency of 3 cycles/min did not appear to enhance viral growth or development of viral cytopathic effect (CPE).
Viral Inoculum Pools Pools of rhinovirus, coronavirus, influenza A, and adenovirus wereused to inoculate nasal epithelial fragment cultures. Laboratory strains of rhinovirus (types 10, 14,and 39 and strain HH, which is presently untyped) and of coronavirus 229E wereharvested after three to five passages in human embryonic lung fibroblast cells (MRC-5 and/or WI-38; M.A. Bioproducts, Walkersville, MD). Rhinovirus type 39 in nasal mucus was obtained from two volunteers with experimental rhinovirus colds. High titered pools of rhinovirus type 14 and type 39 passaged in Hela cells were kindly donated by Dr. Roland Rueckert (University of Wisconsin). Clinical isolates of influenza type A and adenovirus were provided by Dr. Frederick G. Hayden (Viral Diagnostic Laboratory, University of Virginia). The pool of influenza virus was harvested from rhesus monkey kidney cells (MK; M.A. Bioproducts) after three passages, and the adenovirus was harvested from human embryonic kidneycells(HEK; M.A. Bioproducts) after one passage. Inoculation of Nasal Cultures with Virus A O.lO-ml suspension containing 104 to 105 tissue culture infectious doses 50 (TCID50/ml) of virus was inoculated into each dish so that the nasal fragments were exposed to UP or 104 TCID 50/ml. After 3 h of exposure, unattached virus was removed by washing each dish four times with 1.5 ml of HBSS. Control dishes, which were not exposed to virus, were washed in a similar fashion. After washing, dishes were refed with 1.5 ml of medium and incubated. The medium in the dishes was changed again at 48 h after exposure to virus. In each experiment, two dishes with nasal fragments wereexposed to virus and one dish with fragments from the same patient served as the control. In order to quantitate growth of virus a O.l5-ml aliquot of medium was removed from each dish just before washing, at the time of the fourth wash, and daily after virus inoculation. The samples from the virus-exposed dishes were pooled for virus titration. Infectivity titrations of medium from culture dishes exposed to rhinovirus or coronavirus were performed in MRC-5 or WI-38 fibroblast cultures in screw-capped tubes. Infectivity titrations of adenovirus were done
WINTHER, GWALTNEY, AND HENDLEY
in HEK cells and of influenza type A in MK cells with hem absorption after 1 and 2 wk.
Evaluation of the Effect of Viral Infection on Fragments and Epithelial Monolayers Each nasal fragment and the surrounding monolayer of epithelial cells that had grown out from the fragment were examined at a magnification x 100 with an inverted microscope. Before viral inoculation, a detailed map of the topographic characteristics of each fragment and of the surrounding monolayer was drawn, which made it possible to identify the same fragment and cellsin the monolayer every day. Prior to and daily after inoculation the morphologic features of the adjacent epithelial outgrowth were described (1) by grading the extent to which the fragment was surrounded by monolayer extending 3 mm from the fragment (as +, + +, or + + + ), (2) by noting the proportion of the monolayer in which there were cells with beating cilia, and (3) by noting morphologic changes (cytopathic effect) of the epithelial cells. In addition, the ciliary activity of each fragment was estimated by recording in percent the proportion of the edge of the fragment with ciliary motion. Ciliary activity on the fragments and morphologic features of the monolayer of epithelial cells were evaluated without knowledge of whether the dish had been exposed to virus or not. Histologic Examination of Nasal Fragments Nasal fragments from infected and corresponding noninfected control cultures were fixed with 10% formalin at 96, 120, or 168h after virus inoculation. The fragments were embedded in paraffin, cut, and stained with hematoxylin-eosin. The histologic appearance of the epithelium on fragments was described on coded slides. Inoculation of Lymphocyte and Monocyte Suspensions with Rhinovirus Suspensions of lymphocytes and monoeytes in culture (donated by Dr. Richard Pearson, Department of Internal Medicine, University of Virginia) were prepared from peripheral blood of healthy adult donors. Mononuclear cells separated with Ficoll-sodium diatrizoate (Sigma Chemical Co., St. Louis, MO) sedimentation were suspended at a concentration of 10' cells/ml in Medium 199(Gibco Laboratories) and plated onto sterile plastic petri dishes. Nonadherent cells,which were 90% lymphocytes, were removed after 2 h. The dishes were washed, and after 3 h the adhering mononuclear cells were released with a rubber policeman. The cells were suspended in Medium 199with 10% fetal calf serum without antibiotics. The lymphocyte suspension contained an averageof 8 x 10" cells/ml, and the monocyte suspension contained 5 x 105 monocytes/ml. The cell suspensions were maintained in 36 x 27 mm Teflon'"vials (Scientific SpecialtiesServices, Inc., Randallstown,
MD) at 35° C in a 5% CO2 atmosphere,Rhinovirus type 10 (105 TCID 50/ml) was inoculated into each cell suspension. An aliquot of each inoculated cell suspension was removed at 24 and 72 h for virus titration in MRC-5 cell culture tubes. Each aliquot was titered both unfrozen and after 3 freeze-thaw cycles to release intracellular virus.
Results
Establishment of Monolayers of Nasal Epithelial Cells Nasal polyp specimens from 40 patients, 30 specimens of adenoid tissue, and four nasal turbinates were prepared for culture. Fragment cultures from 36 (900/0) of the nasal polyps, 14(47%) of the adenoids, and three (75%) of the nasal turbinates had surrounding outgrowth of a confluent monolayer of cuboidal cells within a week after initiation (figure 1A). The leading edge of the outgrowth extended approximately 2 to 3 mm from the fragment. Ciliated cells, which constituted 2 to 20% of the monolayer, were either concentrated in localized areas or scattered throughout the monolayer. Ciliated activity in the monolayer continued in cultures maintained for a month. Beginning approximately 3 wk or more after initiation, groups of spindle-shaped cells oriented in parallel to each other wereseen in some cultures. These spindleshaped cells were considered to be fibroblasts. The fibroblasts could be easily distinguished from the ciliated epithelial cells and the nonciliated cuboidal cells. A definite interface was usually present between the epithelial cells and fibroblast types (figure 1B). Two preliminary experimentsweredone to determine whether outgrowth of the ciliated and cuboidal cells was enhanced by culturing mucosal fragments on a layer of collagen matrix (Vitrogens 100,Collagen Corporation, Palo Alto, CA) or on an extracellular matrix elaborated by cultured endothelial cells from human umbilical cord vein (8). Neither matrix was superior to the plastic (Primarias; Falcon) with regard to rapidity of appearance of the outgrowth, extent of monolayer, or maintenance of ciliary activity by cells. As a consequence, fragments attached to plastic dishes were used in all subsequent experiments. Growth of Virus and Detection of CPE in Nasal Epithelial Monolayers The growth 0 f respiratory viruses in fragment/monolayer cultures and production of viral CPE in the monolayers was examined in 29 experiments with fragment/
841
RESPIRATORY VIRAL INFECTION OF HUMAN NASAL EPITHELIAL CELLS
Fig. 1. Outgrowth of monolayer of epithelial cells from fragments of nasal polyp mucosa (phase contrast; magnification: x100). Beating cilia, which are seen in the fresh state, are not visible in photomicrographs. A. Seven days in culture; the fragment is seen to the left of the epithelial cell monolayer (EC). B. Three weeks in culture. Interface between epithelial cell monolayer (EC) and spindle-shaped fibroblasts (FB) is shown.
Fig. 2. Viral growth curves, assessed by titers of viable virus in media, in culture dishes inoculated with rhinovirus type 10. Dishes were washed to remove unattached virus 3 h after inoculation. Arrows denote virus inoculation of dishes. A. Two dishes containing mucosal fragments and epithelial cell monolayer from adenoid. B. Two dishes containing epithelial cell monolayer from adenoid without fragments. Fragments were removed 24 h prior to inoculation. C. Two dishes containing media without fragments or outgrowth. Dishes were not washed 3 h after inoculation.
4
0 0
/_0
E
