Enhanced Pulmonary Epithelial Replication and Axial Airway Mucosubstance Changes in F344 Rats Exposed Short-Term to Mainstream Cigarette Smoke

Toxicology and Applied Pharmacology 161, 171–179 (1999) Article ID taap.1999.8798, available online at http://www.idealibrary.com on

Enhanced Pulmonary Epithelial Replication and Axial Airway Mucosubstance Changes in F344 Rats Exposed Short-Term to Mainstream Cigarette Smoke T. H. March, L. M. Kolar, 1 E. B. Barr, G. L. Finch, M. G. Me´nache, and K. J. Nikula Inhalation Toxicology Laboratory, Lovelace Respiratory Research Institute, P.O. Box 5890, Albuquerque, New Mexico 87185-5890 Received April 19, 1999; accepted September 14, 1999

Enhanced Pulmonary Epithelial Replication and Axial Airway Mucosubstance Changes in F344 Rats Exposed Short-Term to Mainstream Cigarette Smoke. March, T. H., Kolar, L. M., Barr, E. B., Finch, G. L., Me´nache, M. G., and Nikula, K. J. (1999). Toxicol. Appl. Pharmacol. 161, 171–179. Cigarette smoking is associated with respiratory diseases that may be caused by injury to specific pulmonary cells. The injury may manifest itself as site-specific enhanced cellular replication. In this study, rats were exposed either to mainstream cigarette smoke (CS; 250 mg total particulate matter/m 3) or to filtered air (FA) for 6 h/day, 5 days/week, for 2 weeks. In one group, cells in S-phase were labeled over 7 days by bromodeoxyuridine (BrdU) released from implanted osmotic pumps (pump labeled), while another group received BrdU by injection 2 h prior to necropsy (pulse labeled). Morphometry showed that the type II epithelial BrdU labeling index (LI) was significantly elevated in the CSexposed animals of both labeling groups. The axial airway and terminal bronchiolar LIs were enhanced by CS only in the pumplabeled group. In a third group (pulse labeled), 2 weeks of recovery following exposure to CS allowed a normalization in the type II LI. In the pump-labeled rats, the CS-induced elevation of the type II LI was greater than the LI elevation in conducting airways, suggesting that the parenchyma may have been injured more than the conducting airways. The terminal bronchiolar LI in the pumplabeled group, regardless of exposure, was significantly greater than the axial airway LI. Pump labeling, in contrast to pulse labeling, could therefore discern differences among replication rates of conducting airway epithelium in different regions of the lung. Mucosubstance (MS) within the axial airway epithelium was quantified by morphometry. The CS exposure did not increase the total number of MS-containing cells or the total number of axial airway epithelial cells, but there was a phenotype change in the MS cells. Neutral MS cells (periodic acid–Schiff-positive) were significantly decreased, while acid MS cells (alcian blue-positive) were slightly increased by CS exposure. Either cell replication and differentiation or differentiation alone may have changed the phenotype in the MS cell population. © 1999 Academic Press Key Words: cigarette smoke; rats; lung; cell injury; cell replication; BrdU; mucosubstance.

1

Current address: 2825 NE Expressway NE #M2, Atlanta, GA 30345.

Cell replication in toxicant-exposed respiratory tissue, determined by identifying cells labeled with bromodeoxyuridine (BrdU) or tritiated thymidine during division, is a site-specific indicator of injury (Monticello et al., 1991; Henderson et al., 1993; Rajini and Witschi, 1994; Witschi and Rajini, 1994; Lee et al., 1995). Cell replication in short-term studies, even in the absence of histologic evidence of injury, may identify the location and types of cells at risk for carcinogenesis during prolonged exposures (Butterworth, 1990; Cohen and Ellwein, 1991; Klaunig et al., 1991). Cigarette smoke (CS) exposure increases cell replication in rat nasal and extrapulmonary airway epithelium (Hotchkiss et al., 1995; Jeffery et al., 1982), but few studies have assessed the effect of short-term, mainstream CS exposure on pulmonary cell replication. Rajini and Witschi (1994) evaluated BrdU-labeling indices (LIs) in the pulmonary conducting airway epithelium and parenchymal cells of mice exposed to sidestream CS (1 mg total particulate material [TPM]/m 3) for 1–5 days and found that the lung conducting airway epithelium, but not the parenchyma, of the A/J strain was susceptible to injury, while the airway epithelium of the C57BL/6 strain was resistant. LIs in the respiratory tracts of hamsters exposed to the same level of sidestream smoke for 1–3 weeks differed very little from control values (Witschi and Rajini, 1994). Sekhon et al. (1994) found elevated LIs in the terminal bronchioles and associated blood vessels in lungs of rats exposed to mainstream CS for 1–7 days (daily nose-only exposures to CS from seven cigarettes). Jones and Reid (1978) reported on previous work that demonstrated increased mitoses in extrapulmonary airways of rats after as little as several hours of exposure to mainstream CS, but mitoses in pulmonary conducting airways were not increased by exposure to CS for as much as 2 weeks (daily whole-body exposures to CS from 25 cigarettes). Other work has shown that replication increases in pulmonary conducting airways following 1 day of CS exposure, but further exposure for up to 2 weeks is associated with a normalization of the number of replicating airway cells (Jeffery et al., 1985; Bolduc et al., 1981). Variation in the methods of labeling and counting replicating cells may cause discrepancies in the reported LIs of pulmonary tissues. Rajini and Witschi (1994) found that pulmonary pa-

171

0041-008X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.

172

MARCH ET AL.

renchymal cells of A/J mice labeled with BrdU from implanted osmotic pumps (continuous labeling) had a lower basal LI than parenchymal cells in C57BL/6 mice. These data contrasted with earlier results in A/J and C57BL/6 mice where replicating cells were labeled by a single “pulse” injection of tritiated thymidine or BrdU, and cells in the alveolar region of A/J mice, a strain susceptible to spontaneous pulmonary neoplasia, had a higher basal LI (Thaete et al., 1986; Koizumi et al., 1993). Although numerous reasons could account for the different results between these laboratories (e.g., sources of animals), differences in labeling methods (continuous vs. pulse) as well as the methods of determining LI may have contributed to disparity in results (Rajini and Witschi, 1994). Cell replication and hyperplasia may be adaptive responses to injurious substances. One such adaptation in the respiratory tract is the increase in number of sulfated glycoprotein- or MS-secreting cells (Basbaum and Jany, 1990). Several studies have shown hyperplasia of secretory cells, particularly those containing MS, within pulmonary and extrapulmonary conducting airways of rats during short-term CS exposure (Jones and Reid, 1978; Smith et al., 1978; Jeffery and Reid, 1981; Rogers and Jeffery, 1986). In one study of short-term CS exposure (Jones and Reid, 1978), a shift from low to high prevalence of acid MS-containing secretory cells accompanied the extrapulmonary airway hyperplasia, while an increase in the number of individual cells containing both acid and neutral MS characterized the MS cell hyperplasia in the pulmonary airways. Further work (Jeffery, 1986; Jeffery et al., 1982; Jeffery and Reid, 1981) suggested that the increase in the number of MS cells in extrapulmonary airways in response to CS exposure results from a combination of replication among MS cells themselves and the division and differentiation to MS cells among basal cells and other, non-MS cells. The purpose of this study was to evaluate the BrdU LIs in the lungs of rats in order to determine the histologic regions of greatest pulmonary epithelial injury after short-term exposure to a high level of mainstream CS. The characteristics of the CS exposure in this study have been thoroughly described and equated to a heavy human smoking pattern (Finch et al., 1998). LIs were compared from the lungs of animals continuously labeled over 7 days with BrdU or pulse-labeled with a single injection of BrdU. The main axial airway from the left lungs of CS-exposed and control animals was assessed for differences in the number of MS cells and in the intracellular volume of MS. MS cell parameters and cell replication parameters provided an indication of the adaptive response in the axial airways of rats to short-term CS exposure. MATERIALS AND METHODS Animals and exposures. Forty-four, 12–13 week-old Fischer 344/N rats (28 male, 16 female) from the Institute’s barrier-maintained colony were used for the study (Table 1). The rats were exposed to mainstream CS (Finch et al., 1995; Chen et al., 1989, 1992) for 6 h/day, 5 days/week, during a 12-day period. Briefly, two 70-cm 3 puffs per minute from Type 1R3 research cigarettes (Tobacco Health Research Institute; Lexington, KY) were generated by

TABLE 1 Experimental Design Number of animals (male/female) Treatment group 2-Week exposure b Cigarette smoke Filtered air 2-Week exposure plus 2-week room air Cigarette smoke Filtered air

Pulse-labeled a

Pump-labeled

4/4 4/4

4/4 4/4

8/0 4/0

a S-phase cells in rat tissues were either labeled with BrdU by an intraperitoneal injection 2 h prior to killing and necropsy (pulse) or were continuously labeled over 7 days by a subcutaneously implanted osmotic pump. b Rats were exposed to mainstream CS for 10 days out of a 12-day (approximately 2 week) exposure period (see Materials and Methods).

smoking machines (Type 1300; AMESA Electronics; Geneva, Switzerland), diluted with filtered air (FA), and delivered to H-2000 whole-body exposure chambers (Hazleton Systems, Inc.; Aberdeen, MD). The nominal amount of TPM produced per cigarette was 27.1 mg, and six to seven puffs per cigarette were generated (Finch et al., 1998). The mass concentration (TPM/m 3) of the CS in the chamber atmosphere was determined by gravimetric analysis of filter samples taken during the exposure periods. The rats were acclimated to a concentration of 149 mg TPM/m 3 during the first 2 exposure days and thereafter were exposed to 249 mg TPM/m 3. Rats were not exposed to CS during an intervening weekend; exposures were completed during the following week. Control animals were sham-exposed to FA during the time period. On day 6 of the exposure period, osmotic pumps (Alzet 2ML1; 11.8 ml infusion/h; Alza Corp.; Palo Alto, CA) containing 2 ml of BrdU (20 mg/ml; Sigma Chemical Co.; St. Louis, MO) were subcutaneously implanted in one group of rats in order to label dividing cells continuously (pump-labeled group). On day 13, the pump-labeled rats were killed and necropsied. On day 12, BrdU (50 mg/kg body wt) in saline was administered by intraperitoneal injection to a second group (pulse-labeled group). Rats in this group were killed 2 h later, and then necropsied. After day 12, a third group of CS- and FA-exposed animals were kept for an additional 2 weeks in room air (exposure 1 recovery group). Rats in the third group were pulse-labeled with BrdU, killed, and necropsied on day 26 after the beginning of the exposure period. Necropsy and histology. All animals were killed with an overdose of pentobarbital and exsanguination via an abdominal aortic incision. The lungs were perfused via the trachea with 10% buffered formalin at 25 cm of water constant pressure for 2 h. The tracheas were tied off, and the lungs were further fixed for 48 h or more by immersion in formalin. The left lungs were systematically sectioned in a dorsoventral-transverse direction at 4-mm intervals with the first slice randomly positioned within the cranial 4 mm of tissue (Weibel, 1979; Bolender et al., 1993). This stratified random sampling generally produced seven blocks of tissue, numbered from cranial to caudal, with the mainstem bronchus entering the left lung of most rats at the level of block 3. Tissues were routinely processed and embedded in paraffin. Sections of 5-mm thickness from the caudal faces of the blocks were used for immunohistochemistry and histochemistry. Sections were stained for BrdU-positive nuclei with a monoclonal anti-BrdU antibody (Becton-Dickinson; Mountain View, CA), an avidin– biotin–peroxidase complex system (Vectastain Elite kit; Vector Laboratories; Burlingame, CA), diaminobenzidine chromogen, and a counterstain of hematoxylin and eosin (H & E). Parallel sections from the pump-labeled animals were stained with alcian blue (pH 2.5) and periodic acid–Schiff (AB/PAS) for the measurement of airway epithelial acid and neutral MS, respectively.

EPITHELIAL CHANGES IN CIGARETTE SMOKE-EXPOSED RATS Replicating cell counts. The BrdU–H & E-stained lung sections were evaluated using light microscopy for descriptive histopathology. In random fields of the parenchyma in left lung sections 2, 4, and 6, BrdU-labeled type II epithelial cell nuclei were counted using light microscopy with an Olympus BH-2 microscope (603 objective, 103 ocular), and the total number of type II cells and other alveolar cells were estimated as described by Shami et al. (1985). BrdU-labeled and total nuclei were counted in the terminal bronchiolar epithelium of sections 2, 4, and 6 and in the axial airway epithelium of sections 3, 4, 5, and 6. Terminal bronchioles, lined by simple cuboidal to columnar epithelium, were identified as small airways with open ends connected to alveolar ducts. The axial airways, lined by pseudostratified or simple columnar epithelium and running the length of the caudal part of the left lung, were defined by large circular profiles surrounded by interstitium continuous with invaginations of the dorsomedial pleura (one per section). Axial airway and terminal bronchiole cells were not counted for the recovery group. The conducting airway epithelial cells were not categorized by cell type. The LI for type II epithelium was calculated as the number of BrdU-labeled cells divided by the estimated number of type II cells. The LI for the terminal bronchiolar epithelium was determined by the number of BrdU-labeled cells divided by the total number of epithelial cells counted in the terminal bronchioles of left lung sections 2, 4, and 6. For the whole left lung, the LI for the axial airway epithelium was determined by the number of BrdU-labeled cells divided by the total number of epithelial cells counted in the axial airways of left lung sections 3, 4, 5, and 6. The LI for axial airway epithelium was also calculated for the individual sections to assess the effect of airway path length. Mucosubstance measurements. The axial airway epithelial cells containing AB- and/or PAS-positive MS in left lung sections 3, 4, 5, and 6 from the pump-labeled group were counted using a BX60 light microscope with 603 PlanApo objective and 103 ocular lenses (Olympus Corp.; Lake Success, NY). The MS cells were subcategorized into those containing primarily acid, AB-positive MS granules (dark blue to dark violet) and those containing primarily neutral, PAS-positive MS granules (deep pink). In many of the MS cells, dense aggregates of stained granules were located within cell apices protruding above the surrounding epithelial surface (goblet cells). In some of the MS cells, particularly those with neutral MS, the stain extended deeply and encompassed nuclei. In sparsely granulated AB-positive cells, the MS often was limited to a thin apical rim, while, in sparsely granulated PAS-positive cells, the MS was often dispersed unevenly in perinuclear and basilar compartments. The number of MS-containing cells in an airway was divided by the basal lamina length (measurement method outlined below) to yield the number of MS cells per millimeter. The length included epithelium overlying submucosal lymphoid tissue, although usually few MS-containing cells were located in these regions. The volume density (V S ) of the MS was also measured in the axial airways of left lung sections 3, 4, 5, and 6 from the pump-labeled group. Video images were made with a computer-interfaced black-and-white CCD camera (TM480; Pulnix America; Sunnyvale, CA) connected to an Olympus BX60 light microscope with 203 and 403 PlanApo objective lenses. The video images were measured with morphometric software (ImageMeasure; Microscience, Inc.; Federal Way, WA). The length of the basal lamina along the perimeter of the axial airway was measured (20603 image), and computer-generated and operator-adjusted intensity thresholds allowed detection and measurement of the cross-sectional areas of AB- and PAS-stained MS within the airway epithelium (41203 image). The V S of the MS within a surface area unit of epithelium was defined by the following equation (Weibel, 1979; Harkema et al., 1987):

V S ~pl/mm 2 ! 5

p 3 ~MS area @in mm 2 #! 3 10 6 . 4 3 ~basal lamina length @in mm#!

Not all of the MS in sparsely granulated cells, particularly those cells that were primarily PAS-positive, was detected during area measurements, although all cells containing visible amounts of MS were included in the cell counts. For an indicator of hyperplasia, the number of epithelial cells per millimeter of axial airway basal lamina for each airway was calculated from the pump-

173

labeled group. In the cell replication phase of the study, the axial airway epithelial cells were counted for the LI, but the length of the basal lamina was not measured. Subsequently, the axial airway epithelial cells (those not overlying bronchiole-associated lymphoid tissue) in the anti-BrdU/H & E-stained sections were counted, and corresponding lengths of the basal lamina were measured as outlined above. The LI of the axial airways was multiplied by the number of cells per millimeter of basal lamina to produce the number of BrdU-labeled cells per millimeter basal lamina. Statistical analysis. The numbers of epithelial cells counted or estimated in each pulmonary region (axial airway, terminal bronchioles, and parenchyma) were compared among the exposure and labeling method groups by one-way analysis of variance (ANOVA; SigmaStat for Windows; Jandel Scientific; Corte Madera, CA). Where significant differences were indicated ( p , 0.05), the Student–Neuman–Keuls (SNK) pairwise comparison test was used to isolate which group significantly differed from the others ( p , 0.05). For differences in epithelial LI at different locations along the length of the axial airway, a repeated-measures ANOVA was used with exposure as the independent variable and lung section as the repeated factor. Data were transformed by ranking prior to ANOVA when they were not normally distributed (Kolmogorov–Smirnov test, p , 0.05) or had unequal variances (Levene’s median test, p , 0.05). In unbalanced sets of data (e.g., in some animals, left lung sections 3 or 6 did not contain axial airway profiles), a general linear model of regression was used to estimate missing values and degrees of freedom. Where significant differences were indicated ( p , 0.05), the SNK test was used to isolate which group (exposure, lung section, or exposure 3 lung section) significantly differed from the others. To analyze differences in injury among epithelial cells in different pulmonary regions, the mean LIs from whole-left lung measurements within each labeling group (pump-labeled or pulse-labeled) were compared by a repeatedmeasures ANOVA with exposure (CS or FA) as the independent variable and regional epithelial cell type (type II, terminal bronchiolar, or axial airway) as the repeated factor. Where a statistically significant difference was indicated ( p , 0.05), the SNK test was used to isolate which group significantly differed from the others ( p , 0.05). One-way ANOVA was used to compare mean type II epithelial LI between the pulse-labeled, 2-week-exposed and the recovery groups. Data for the MS and cell hyperplasia measurements were compared by repeated-measures ANOVA, in a manner similar to the analysis of axial airway LIs, in order to assess differences at different locations along the length of the axial airway. Means from the CS- and FA-exposed animals for each of the following parameters were compared: total epithelial cells per millimeter basal lamina, total MS cells per millimeter, PAS- and AB-positive cells per millimeter, V S , and the BrdU-labeled cells per millimeter. Where a significant difference was found ( p , 0.05), the SNK test was used to isolate which means differed significantly from the others.

RESULTS

No inflammation or morphologic evidence of cell damage was found in the pulmonary conducting airways of CS- or FA-exposed animals or in the recovery group. The lung parenchyma of rats exposed to CS for 2 weeks had minimal to mild, disseminated, alveolar macrophage hyperplasia. In the CSexposed rats allowed a 2-week room air recovery period, intraalveolar macrophages were aggregated three or four to a group in a few widely scattered foci, but the disseminated macrophage hyperplasia had resolved. Cigarette smoke exposure was not associated with statistically significant differences in the total number of cells in any histologic region of the lung except for the estimated number of type II epithelial cells (Table 2). These were significantly more numerous in the 2-week CS-exposed animals regardless

174

MARCH ET AL.

TABLE 2 Total Number of Cells Counted or Estimated in Different Regions of the Rat Lung a Axial airway Parenchymal cells By left lung section Exposure group 2-Week exposure, pump-labeled Cigarette smoke Filtered air 2-Week exposure, pulse-labeled Cigarette smoke Filtered air 2-Week exposure plus 2-week room air Cigarette smoke Filtered air

LL 5

LL 6

By whole left lung

587 6 82 (7) 576 6 47 832 6 104 605 6 85

618 6 68 527 6 50

517 6 67 470 6 14 (7)

2223 6 157 2374 6 177

1844 6 144 4040 6 8 1913 6 122 4044 6 8

445 6 7* 393 6 13†

673 6 85 572 6 58

600 6 50 600 6 55

601 6 51 516 6 52

457 6 26 515 6 69

2331 6 124 2203 6 152

1764 6 124 4027 6 9 1540 6 93 4055 6 11

428 6 6* 394 6 8†

NC NC

NC NC

LL 3

NC b NC

LL 4

NC NC

NC NC

Terminal bronchioles

NC NC

All cells

4034 6 8 4036 6 4 (4)

Type II epithelium

371 6 7† ,‡ 346 6 6‡ (4)

a See Materials and Methods for counting and estimating procedures. Values are means 6 SE of total cells (BrdU-labeled and unlabeled) in different regions of the lung. Numbers in parentheses, where present, indicate number of animals evaluated. Otherwise, n 5 8 per group. Values in each column among all exposure groups were compared by one-way ANOVA. Significant differences among the exposure groups were apparent only for the estimated number of type II epithelial cells. Results of SNK pairwise comparisons for this column are indicated by symbols (*, †, or ‡). Those values not sharing like symbols are significantly different from each other ( p , 0.05). b NC, not counted.

of the BrdU-labeling group to which they belonged. Two weeks of room air recovery, however, was associated with a normalization in the number of type II cells. The animals allowed this recovery time also had lower numbers of type II cells, regardless of exposure, than the animals exposed for 2 weeks only. This may have been attributable to decreasing numbers of type II cells with increasing age. In order to assess an effect of approximate path length on the LI, data for the axial airways were first analyzed at different

locations by left lung section (Fig. 1). The repeated-measures analysis showed no statistically significant effect of an interaction between exposure and lung section. A statistically significant effect of exposure on the axial airway LI was observed ( p , 0.01) but only in the pump-labeled group. There was no effect of lung section, suggesting that average cell replication in the pump-labeled group (for both CS- and FA-exposed animals) as well as the CS-associated elevation in LI were similar along the entire length of the axial airway.

FIG. 1. Comparison of the axial airway epithelial labeling indices (LIs) in response to 2-week cigarette smoke exposure at different locations along the length of the pathway (by left lung section). Circles (6SE) are the mean values. In both labeling groups (pump or pulse), the data failed normality tests, and differences between the mean rank-transformed values were evaluated by repeated-measures analysis of variance. The analysis showed a statistically significant effect of exposure (CS or FA) only in the pump-labeled group ( p , 0.05); there was no effect of lung section or due to an interaction between exposure and section.

EPITHELIAL CHANGES IN CIGARETTE SMOKE-EXPOSED RATS

175

FIG. 2. Comparison of pulmonary epithelial labeling indices (LIs) among different regions after exposure to cigarette smoke for 2 weeks. Pump labeling resulted in LIs approximately 10-fold higher than LIs in pulse-labeled animals. Data for the pump-labeled animals were not normally distributed and were transformed by ranking prior to ANOVA. Repeated-measures ANOVA within each labeling group (pump or pulse) showed a significant interaction between exposure (cigarette smoke or filtered air) and regional cell type (type II, terminal bronchiolar [T. Bron.], or axial airway [Ax. Airway] epithelium). Bars (mean LI plus SE) within each labeling group with differing letter labels are significantly different from each other by Student–Neuman–Keuls pairwise comparison test ( p , 0.05).

Because the level of lung section did not affect the axial airway LI, data from all lung sections were summed to yield whole-lung measurements and then compared with the LIs of other regional cell types in the lung (Fig. 2). Pump labeling resulted in LIs that were approximately 10-fold higher than corresponding LIs in pulse-labeled animals. The repeated-measures ANOVA showed a statistically significant effect of an interaction between exposure (CS or FA) and cell type (type II, terminal bronchiolar, or axial airway epithelium) in both the pump-labeled ( p , 0.0001) and pulse-labeled ( p , 0.005) groups. Regardless of the labeling method, the LI for type II cells in CS-exposed animals by SNK comparison was significantly greater than that in the FA-exposed animals. For the terminal bronchiolar and axial airway epithelium, pairwise comparison showed significantly elevated LIs in the pumplabeled group only. The type II epithelial cell LI in CSexposed, pulse-labeled animals was 1.7-fold greater than that of FA-exposed animals. The elevation of type II epithelial cell LI in CS-exposed, pump-labeled animals was 2.3-fold, while terminal bronchiolar and axial airway epithelial LIs were elevated 1.3- and 2.1-fold, respectively, over control values. Further, the greater LI of the terminal bronchioles vs. the axial airways, regardless of exposure, was statistically significant only in the pump-labeled group. This suggested that, for replication rates among conducting airway epithelial cells of different pulmonary regions, BrdU labeling via implanted osmotic pumps provided better distinction than pulse labeling. Axial airway and terminal bronchiole LIs were not determined in the exposure plus recovery group, because these animals were pulse-labeled with BrdU, and LI results of the pulse-labeled, 2-week exposure group showed no difference between the conducting airway LIs of the CS- and FA-exposed

animals. The mean LIs of type II cells among the pulse-labeled, 2-week exposure group and the group allowed 2 weeks of recovery following CS exposure were compared by one-way ANOVA and SNK tests (Fig. 3), and only the mean LI from the CS-exposed, nonrecovered animals was significantly elevated. There was no difference between the CS-exposed, recovered animals and either group of control animals, suggesting that 2 weeks of recovery allowed normalization of the type II cell LI in CS-exposed animals. Hyperplasia and MS were measured at different locations by lung section along the length of the axial airway of left lungs from the pump-labeled animals (Fig. 4). There was no significant effect of an interaction between exposure and lung section on any parameter ( p $ 0.05). The CS exposure did not cause a statistically significant increase in the total number of epithelial cells or total number of MS cells per millimeter basal lamina. There was, however, a significant effect of lung section ( p , 0.05 and p , 0.005, respectively), suggesting that these parameters, while unaffected by CS exposure, are dependent on the location within the length of the axial airway. The total cells and MS cell numbers were greater, generally, in the cranial aspects of the axial airway. There was a statistically significant effect of exposure and lung section ( p , 0.001 and p , 0.01, respectively) on BrdU-labeled cells per millimeter basal lamina. The CS-associated enhancement of this parameter was similar to that for the axial airway epithelial LI in pump-labeled animals (Fig. 1). Like the total number of MS cells per millimeter, the V S of the MS was not significantly affected by CS exposure. The location in the axial airway (lung section) also had no significant effect on V S , but the parameter varied greatly from cranial to caudal portions of the airway in the FA-exposed animals. Along the length of the axial airway,

176

MARCH ET AL.

FIG. 3. Comparison of the type II epithelial labeling indices (LIs) between the pulse-labeled, 2-week cigarette smoke exposure group and the 2-week exposure plus recovery group. Bars indicate the mean values (plus SE). One-way ANOVA revealed a significant effect of exposure ( p , 0.05), and bars with differing letter labels are significantly different from each other by the Student–Neuman–Keuls test ( p , 0.05).

V S generally followed the number of MS cells per millimeter, but being slightly more variable than the cell numbers, the trend was not statistically significant. While the total number of MS cells per millimeter was unaffected, the changes in the numbers of AB- and PAS-positive cells were consistent with a CS-induced phenotypic change in MS production. The number of PAS-positive, neutral MS cells per millimeter of basal lamina was significantly decreased by the CS exposure ( p , 0.05). The number of AB-positive, acid MS cells per millimeter was apparently enhanced by the CS exposure, although the repeated-measures ANOVA showed the elevation was not statistically significant ( p 5 0.09). The interaction, however, between exposure and lung section for the AB-positive cells per millimeter approached the significance level ( p 5 0.053). For the two types of MS cells, the effect of lung section was significant only for the number of PAS-positive cells per millimeter ( p , 0.001). DISCUSSION

In the present study, the pulmonary parenchymal epithelial cells of rats exposed to CS at 250 mg TPM/m 3 for 6 h per day for 10 days (with an intervening weekend of rest) were injured

based on the significantly increased type II epithelial LI (Figs. 2 and 3). The results were qualitatively similar whether the dividing cells were labeled with BrdU by continuous infusion via a subcutaneous osmotic pump (pump labeling) or by a single intraperitoneal injection (pulse labeling). Therefore, the increased type II epithelial LI in response to CS exposure probably was associated with ongoing and cumulative injury rather than cumulative injury alone. The LIs for epithelial cells in the pulmonary conducting airways of CS-exposed rats were significantly elevated above those of the FA-exposed animals but only in the pump-labeled group (Fig. 2). The enhancement of type II epithelial LI by CS exposure was greater than that in the axial airway and terminal bronchiolar regions (2.3-, 2.1-, and 1.3-fold, respectively), and the elevation in type II cell replication may be considerably more given that the total number of type II cells was also elevated in association with CS exposure (Table 2). This sustained increase in replication suggests that the incidence for diseases associated with cell damage and replication (e.g., neoplasia) may be greater in the pulmonary parenchyma than in the airways of rats chronically exposed to CS. Airway LIs were not measured during the first 5 days of the exposure regimen. Jeffery et al. (1985) and Bolduc et al. (1981) have demonstrated that replication in rat pulmonary airway epithelium rapidly increases following 1 day of CS exposure (4 h/day, CS from 25 cigarettes), and then returns to control levels by day 3 of exposure. The previous models of short-term CS exposure (2– 6 weeks) employed periods of daily CS exposure with intervening rest days resulting in 6 exposure days per week (Jeffery et al., 1985; Bolduc et al., 1981). Temporal analysis showed that a second peak in pulmonary airway “mitotic activity” occurs on the exposure day following the first rest day (Bolduc et al., 1981). A similar early enhancement and adaptation phenomenon in the present study may explain the conducting airway LI findings in the pump-labeled animals. An intervening rest period was also used in the present study, and the elevation in axial airway and terminal bronchiole LIs in the CS-exposed, pump-labeled animals may have resulted from labeling the dividing cells in response to the first day of CS exposure after the rest period. This was not evident in the pulse-labeled group, where the number of dividing cells in the airways at the time of intraperitoneal BrdU injection would be expected to be similar to that of control animals. The type II epithelial LI did not differ between the CS- and FA-exposed animals allowed 2 weeks of recovery in room air (Fig. 3). The results suggested that the injury sustained by the parenchymal epithelium resolves after 2 weeks of recovery. In fact, a small but nonsignificant reduction of the LI in the CS-exposed animals was noted in comparison to controls. Witschi and Rajini (1994) found a similar reduction in the nasal mucosa LI of hamsters exposed to sidestream CS for 1, 2, or 3 weeks followed by 1 week of recovery. The reduction (significant in the maxillary turbinates) may have been a persistent toxic response to sidestream CS or a peculiar response

EPITHELIAL CHANGES IN CIGARETTE SMOKE-EXPOSED RATS

177

FIG. 4. Comparison of axial airway hyperplasia and mucosubstance (MS) parameters in response to cigarette smoke exposure at different locations along the length of the pathway (by left lung section). See Materials and Methods for definitions. The circles (6SE) indicate the mean values. Data for each parameter were either not normally distributed or had unequal variances. Differences between the mean rank-transformed values were evaluated by repeated-measures ANOVA. The analysis showed a statistically significant effect of exposure (E) and lung section (S) for many of the parameters.

of the nasal mucosa, given that LI reductions after a recovery period were not seen in the pulmonary epithelium of the exposed hamsters (Witschi and Rajini, 1994). In pulmonary airways of rats recovered from short-term exposures to mainstream CS (Smith et al., 1978), reduction of epithelial replication rates has not been reported. Elevation of the type II epithelial LI with a lesser elevation of pulmonary conducting airway LIs in response to CS exposure in this study contrasts with previous results (Rajini and

Witschi, 1994). There, A/J mice were exposed for 5 days at 6 h/day to 1 mg TPM/m 3 sidestream CS and continuously labeled with BrdU osmotic pumps. Those mice had significantly elevated LIs in the airway epithelium but no significant elevation of LIs in the parenchyma. Additionally, A/J mice exposed to sidestream CS at a level of 4 mg TPM/m 3 for 1 week had elevated LIs in the pulmonary airway epithelium but not in the trachea or parenchyma; no pulmonary or extrapulmonary LIs were elevated after further exposure of up to 16 weeks (Witschi

178

MARCH ET AL.

et al., 1995). The level of CS exposure, the type of CS (e.g., mainstream vs. sidestream), and species differences may have contributed to the contrasting results of the present study with those using A/J mice. Based on the deposition of particulate material per gram of lung, the exposure level in the present study simulated that of a heavy, three-pack-per-day, human smoker (Finch et al., 1998). This high level of exposure probably caused much different patterns of deposition, clearance, and/or metabolism of CS components than those obtained with low-level, sidestream CS. Such differences would likely affect the specific site of cell injury. Exposure to CS in a number of species has enhanced the number of MS cells and/or MS production in the extrapulmonary and pulmonary conducting airways (Wanner, 1985). In some studies, the hyperplastic response occurred within hours following exposure (Jones and Reid, 1978). In contrast, CS exposure in the present study did not significantly enhance the number of MS cells or the V S in the axial airway epithelium (Fig. 4). Many studies of CS-induced changes in rat pulmonary mucous cells have demonstrated increases in secretory cell number in both cranial and caudal aspects of the axial airway as well as in smaller “lateral” airways (Jones and Reid, 1978, 1979; Jeffery et al., 1982, 1985; Rogers and Jeffery, 1986). Other studies have focused on the pulmonary axial airway epithelium at only one level of the left lung (Smith et al., 1978) or have ignored changes in the axial airway altogether (Jeffery and Reid, 1981). The results here show an increase in MS cells per millimeter of basal lamina and an increase in the V S of MS in the cranial axial airway in the left lung (section 3, Fig. 4). When analyzed with data from the other lung levels, these increases were not statistically significant, but by themselves they may be similar to previous results. The contrasting results to previous work for MS values presented here may have been due to a “discharge phenomena” as previously described (Jones and Reid, 1978; Jeffery and Reid, 1981), where MS cells rapidly discharge their contents in response to CS irritation, and MS cells that are detectable by light microscopy effectively are reduced in number. The high concentration of CS in exposure atmospheres of the present study may have produced a more severe discharge effect than the CS atmospheres of previous studies. Exposure of rats for 8 weeks to CS at 250 mg/m 3 significantly increases the number of axial airway MS cells and the intracellular V S of MS (T. H. March, unpublished observations), thus longer exposures under the same conditions as the present work are associated with a compensatory increase in MS production as described by others (Smith et al., 1978). Volume densities of MS in pulmonary airways of CS-exposed rats have not been reported. Airway secretory cells in normal rats or in rats exposed to 2 weeks of CS have been subjectively described as containing little MS (Jones and Reid, 1978), while secretory cells of rat extrapulmonary airways after 6 weeks of CS exposure have been described as “distended with granules” (Jones et al., 1973). Mucosubstance-containing cells were differentiated as containing primarily acid, AB-positive material or neutral, PAS-

positive material. Individual cells containing relatively equal amounts of both types of MS were not readily discernible. Exposure to CS increased the number of AB-positive, acidic MS cells and significantly decreased the number of PASpositive, neutral MS cells per millimeter of basal lamina in the axial airways (Fig. 4). This phenotypic change in the MS cell population is similar to results reported by some (Jeffery et al., 1985; Rogers and Jeffery, 1986), but contrasts with other work (Jones and Reid, 1978) where CS exposure was associated with a decreased prevalence of AB-positive cells in the axial airway. Exposure to CS alters the normal pattern of secretory cell maturation in pulmonary and extrapulmonary airways (Jeffery et al., 1982). For hyperplastic responses of MS cells to CS exposure, some have argued that CS induces MS cell replication (Jeffery et al., 1982). Others have purported an increase in basal or other cell replication with subsequent differentiation to MS cells (Samet and Cheng, 1994; Jany and Basbaum, 1991). In the present study, the number of MS cells and the total number of epithelial cells in the axial airways were not enhanced by CS exposure, but the axial airway epithelial LI (based on the pump-labeled animals) and BrdU-labeled cells per millimeter of basal lamina indicated a CS-induced increase in cell replication (Figs. 1 and 4). Enhanced replication without hyperplasia is consistent with an equivalence between the rates of cell replication and cell loss. The cells that were lost may have included those containing neutral, PAS-positive MS; however, the CS-induced cell replication may have also resulted in the production of greater (albeit statistically insignificant) numbers of AB-positive cells. Whether the changes presented here represent CS-induced enhanced replication and differentiation or merely differentiation of existing cell types into AB-positive cells requires evaluation of cells at different time points during a short-term exposure. This study has shown that short-term CS exposure of rats, on the basis of enhanced replicating cell LIs, is associated with continuous damage to the pulmonary epithelial cells. When compared to type II epithelial LIs, the terminal bronchiolar and axillary airway epithelial LIs were elevated to a lesser extent by CS exposure. This may have been due to a lesser impact of CS on these sites or to an adaptive response of the conducting airway epithelium during the short-term exposure. Unlike past studies, CS exposure was not associated with significant elevation in the total number of MS-containing cells nor in the V S of MS within the pulmonary axial airway mucosa. However, CS exposure significantly decreased the numbers of neutral MS cells (PAS-positive) and increased the numbers of acidic MS cells (AB-positive) within the axial airway. Whether this involved phenotypic changes induced in the resident cell population or involved MS cell death followed by production of new MS cell types requires further study. ACKNOWLEDGMENTS This research was supported by the Assistant Secretary for Defense Programs and Office of Biological and Environmental Research, U.S. Department

EPITHELIAL CHANGES IN CIGARETTE SMOKE-EXPOSED RATS of Energy, under Cooperative Agreement No. DE-FC04-96AL76406, in facilities fully accredited by the Association of Assessment and Accreditation of Laboratory Animal Care International. Linda Kolar was an Associated Western Universities Summer Student Research Participant. The authors thank the technical staff in the Applied Toxicology Program, Exposure Operations, Small Animal Care, Necropsy, and Histopathology Sections at LRRI for assistance with the project and Paula Bradley and the Technical Communications Unit for help preparing this manuscript.

REFERENCES Basbaum, C., and Jany, B. (1990). Plasticity in the airway epithelium. Am. J. Physiol. 259(Lung Cell. Mol. Physiol. 3), L38 –L46. Bolduc, P., Jones, R., and Reid, L. (1981). Mitotic activity of airway epithelium after short exposure to tobacco smoke and the effect of the antiinflammatory agent phenylmethyloxadiazole. Br. J. Exp. Pathol. 62, 461– 468. Bolender, R. P., Hyde, D. M., and Dehoff, R. T. (1993). Lung morphometry: A new generation of tools and experiments for organ, tissue, cell and molecular biology. Am. J. Physiol. 265(Lung Cell. Mol. Physiol. 9), L521– L548. Butterworth, B. E. (1990). Consideration of both genotoxic and nongenotoxic mechanisms in predicting carcinogenic potential. Mutat. Res. 239, 117–132. Chen, B. T., Bechtold, W. E., Barr, E. B., Cheng, Y.-S., Mauderly, J. L., and Cuddihy, R. G. (1989). Comparison of cigarette smoke exposure atmospheres in different exposure and puffing modes. Inhal. Toxicol. 1, 331–347. Chen, B. T., Bechtold, W. E., and Mauderly, J. L. (1992). Description and evaluation of a cigarette smoke generation system for inhalation studies. J. Aerosol. Med. 5, 19 –30. Cohen, S. M., and Ellwein, L. B. (1991). Genetic errors, cell proliferation, and carcinogenesis. Cancer Res. 51, 6493– 6505. Finch, G. L., Lundgren, D. L., Barr, E. B., Chen, B. T., Griffith, W. C., Hobbs, C. H., Hoover, M. D., Nikula, K. J., and Mauderly, J. L. (1998). Chronic cigarette smoke exposure increases the pulmonary retention and radiation dose of 239Pu inhaled as 239PuO 2 by F344 rats. Health Phys. 75(6), 597– 609. Finch, G. L., Nikula, K. J., Chen, B. T., Barr, E. B., Chang, I.-Y., and Hobbs, C. H. (1995). Effect of chronic cigarette smoke exposure on lung clearance of tracer particles inhaled by rats. Fundam. Appl. Toxicol. 24, 76 – 85. Harkema, J. R., Plopper, C. G., Hyde, D. M., St. George, J. A., and Dungworth, D. L. (1987). Effects of an ambient level of ozone on primate nasal epithelial mucosubstances. Am. J. Pathol. 127, 90 –96. Henderson, R. F., Hotchkiss, J. A., Chang, I., Scott, B. R., and Harkema, J. R. (1993). Effect of cumulative exposure on nasal response to ozone. Toxicol. Appl. Pharmacol. 19, 59 – 65. Hotchkiss, J. A., Evans, W. A., Chen, B. T., Finch, G. L., and Harkema, J. R. (1995). Regional differences in the effects of mainstream cigarette smoke on stored mucosubstances and DNA synthesis in F344 rat nasal respiratory epithelium. Toxicol. Appl. Pharmacol. 131, 316 –324. Jany, B., and Basbaum, C. B. (1991). Mucin in disease. Modification of mucin gene expression in airway disease. Am. Rev. Respir. Dis. 144, S38 –S41. Jeffery, P. K. (1986). Anti-inflammatory drugs and experimental bronchitis. Eur. J. Respir. Dis. 69(Suppl. 146), 245–257. Jeffery, P. K., Ayers, M., and Rogers, D. F. (1982). The mechanisms and control of bronchial mucous cell hyperplasia. Chest 81(Suppl.), 27S–29S. Jeffery, P. K., and Reid, L. M. (1981). The effect of tobacco smoke, with or without phenylmethyloxadiazole (PMO), on rat bronchial epithelium: A light and electron microscopic study. J. Pathol. 133, 341–359. Jeffery, P. K., Rogers, D. F., and Ayers, M. M. (1985). Effect of oral

179

acetylcysteine on tobacco smoke-induced secretory cell hyperplasia. Eur. J. Respir. Dis. 66(Suppl. 139), 117–122. Jones, R., Bolduc, P., and Reid, L. (1973). Goblet cell glycoprotein and tracheal gland hypertrophy in rat airways: The effect of tobacco smoke with or without the anti-inflammatory agent phenylmethyloxadiazole. Br. J. Exp. Pathol. 54, 229. Jones, R., and Reid, L. (1978). Secretory cell hyperplasia and modification of intracellular glycoprotein in rat airways induced by short periods of exposure to tobacco smoke, and the effect of the antiinflammatory agent phenylmethyloxadiazole. Lab. Invest. 39(1), 41– 49. Jones, R., and Reid, L. (1979). b-Agonists and secretory cell number and intracellular glycoprotein in airway epithelium. Am. J. Pathol. 95, 407– 422. Klaunig, J. E., Siglin, J. C., Schafer, L. D., Hartnett, J. A., Weghorst, C. M., Olson, M. J., and Hampton, J. A. (1991). Correlation between species and tissue sensitivity to chemical carcinogenesis in rodents and the induction of DNA synthesis. In Chemically Induced Cell Proliferation: Implication for Risk Assessment (B. E. Butterworth, T. J. Slaga, W. Farland, and M. McClain, Eds.), pp. 185–194. Wiley-Liss, New York. Koizumi, A., Tsukada, M., Hirano, S., Kamiyama, S., Masuda, H., and Suzuki, K. Y. (1993). Energy restriction that inhibits cellular proliferation by torpor can decrease susceptibility to spontaneous and asbestos-induced lung tumors in A/J mice. Lab. Invest. 68, 728 –739. Lee, C., Rajini, P., Pinkerton, K. E., and Witschi, H. (1995). Correlation between cumulative labeling indices measured in the terminal bronchioles and in the centriacinar region in the lungs of rats exposed to oxidant air pollutants. Toxicol. Lett. 80, 25–28. Monticello, T. M., Miller, F. J., and Morgan, K. T. (1991). Regional increases in rat nasal epithelial cell proliferation following acute and subchronic inhalation of formaldehyde. Toxicol. Appl. Pharmacol. 111, 409 – 421. Rajini, P., and Witschi, H. (1994). Short-term effects of sidestream smoke on respiratory epithelium in mice: Cell kinetics. Fundam. Appl. Toxicol. 22, 405– 410. Rogers, D. F., and Jeffery, P. K. (1986). Inhibition by oral N-acetylcysteine of cigarette smoke-induced “bronchitis” in the rat. Exp. Lung Res. 10(3), 267–283. Samet, J. M., and Cheng, P.-W. (1994). The role of airway mucus in pulmonary toxicology. Environ. Health Perspect. 102(Suppl. 2), 89 –103. Sekhon, H. S., Wright, J. L., and Churg, A. (1994). Cigarette smoke causes rapid cell proliferation in small airways and associated pulmonary arteries. Am. J. Physiol. 267(Lung Cell. Mol. Physiol. 11), L557–L563. Shami, S. G., Caton, T. M., Griffith, W. C., and Evans, M. J. (1985). An improved method for calculating labeling indices of lung epithelial and interstitial cells. J. Toxicol. Environ. Health 15, 229 –236. Smith, G., Wilton, L. V., and Binns, R. (1978). Sequential changes in the structure of the rat respiratory system during and after exposure to cigarette smoke. Toxicol. Appl. Pharmacol. 46, 579 –591. Thaete, L. G., Beer, D. G., and Malkinson, A. M. (1986). Genetic variation in the proliferation of murine pulmonary type II cells: Basal rates and alterations following urethan treatment. Cancer Res. 46, 5335–5338. Wanner, A. (1985). A review of the effect of cigarette smoke on airway mucosal function. Eur. J. Respir. Dis. 66(Suppl. 139), 49 –53. Weibel, E. R. (1979). Stereologic Methods. Vol. 1. Practical Methods for Biological Morphometry. Academic Press, New York. Witschi, H., and Rajini, P. (1994). Cell kinetics in the respiratory tract of hamsters exposed to cigarette sidestream smoke. Inhal. Toxicol. 6, 321–333. Witschi, H., Oreffo, V. I. C., and Pinkerton, K. E. (1995). Six-month exposure of strain A/J mice to cigarette sidestream smoke: Cell kinetics and lung tumor data. Fundam. Appl. Toxicol. 26, 32– 40.