Biodegradability of Para-aramid Respirable-Sized Fiber-Shaped Particul
http://www.100md.com
《毒物学科学杂志》
DuPont Haskell Laboratory for Health and Environmental Sciences, Newark, Delaware
National Health and Environmental Effects Research Laboratory, US Environmental Protection Agency, Chapel Hill, North Carolina
2 To whom correspondence should be addressed at DuPont Haskell Lab, 1090 Elkton Road, Newark, DE 19714-0050. Fax: (302) 366-5207. E-mail: david.b.warheit@usa.dupont.com.
ABSTRACT
Using both in vivo (inhalation) and in vitro (cell culture) studies, we previously reported that p-aramid respirable fibers (RFP—defined as respirable-sized fiber-shaped particulates) are biodegraded in lungs and lung cells of rats following exposures. The current studies were undertaken to determine whether shortening mechanisms of p-aramid RFP biodegradability are also operative in human lung cells. Cultures of human A549 lung epithelial cells (A549), primary alveolar macrophages (HBAL) (collected via bronchoalveolar lavage [BAL]) from volunteers), and co-cultures (Co) of the A549 and HBAL were incubated with p-aramid RFP for either 1 h, 1 day, or 1 week to assess RFP shortening. Lengths of RFP were measured using scanning electron microscopy (SEM) following fixation, digestion of culture tissue components, and processing. Similar to findings using rat lung cells, only slight RFP shortening was measured in A549 cultures at 1-day and 1-week post-incubation. More importantly, in HBAL and Co groups, greater transverse cleavage of p-aramid RFP was measured at 1-day and 1-week postexposure compared to 1-h HBAL or Co groups, or in any A549 groups. In contrast, cellulose RFP, a biopersistent reference control fiber, were not measurably shortened under similar circumstances. Second, p-aramid RFP were incubated either with phosphate-buffered saline (PBS), or acellular BAL fluids from human volunteers or rats and processed for SEM analysis of RFP lengths. Mean lengths of p-aramid RFP incubated with human or rat BAL fluids were substantially decreased compared to PBS. Similar to our findings with rat lung cells, components of human lung fluids coat the p-aramid RFP as a prerequisite for subsequent enzymatic cleavage by human phagocytic lung cells and this finding reinforces the concept that inhaled p-aramid RFP are likely to be biodegradable in the lungs of humans.
Key Words: p-aramid respirable fibers; human lung cells; fiber shortening mechanisms.
INTRODUCTION
Para-aramid fibers have diameter dimensions which are considered to be nonrespirable (i.e., in the range of 12–15 μm), but fibrils (i.e., p-aramid RFP) with diameter dimensions of 0.3–0.7 μm, and lengths up to 100 μm, can be produced under conditions of abrasion (Fig. 1). Para-aramid fibers in the form of pulp contain RFP as an integral component (ECETOC, 1996). Pulp is utilized commercially in gaskets and in friction products such as brake linings.
Unlike most other organic fiber-types, the pulmonary toxicity effects of p-aramid RFP inhalation have been extensively tested. In this regard, the results of inhalation toxicity studies in rats and hamsters from several different laboratories throughout the world consistently have demonstrated that after an initial buildup of inhaled RFP in the lung, there is a rapid clearance of the longest fibrils, concomitant with an initial increase in the numbers of shorter fibrils (Bellmann et al., 2000; Kelly et al., 1993; Searl, 1997; Warheit et al., 1992, 1994). The cardinal feature of this RFP clearance pattern in the lungs of exposed animals is a progressive decrease in the average lengths of retained RFP with increasing postexposure time period intervals (Warheit et al., 1992). This mechanism has been reported to occur via transverse cleavage, i.e., shortening of the retained RFP, which occurs with increasing residence time in the lung.
In earlier methods-development studies, Searl and Cullen (1997) proposed that p-aramid RFP cleavage occurred as a result of the synergistic effects of coating of RFP with lung fluids along with enzymatic digestion of the fibril. In previously reported studies, we developed an in vitro culture system with rat lung cells to (1) test the validity of this hypothesis; and (2) develop an in vitro system that might be utilized as a bridging technique to assess the biodegradability of p-aramid RFP for human lung cells.
Several studies have been undertaken to elucidate mechanisms of p-aramid biodegradability. Based on preliminary findings by Searl (1997), who reported that the enzymatic digestion method artificially cleaved p-aramid RFP after instillation into the lungs, we sought to test the hypothesis that (1) inhaled p-aramid RFP are biodegraded in the lungs of exposed animals via an enzymatic mechanism; and (2) this process of biodegradation is facilitated by coating of the RFP with lung fluids.
The results of our earlier study with rat cells in vitro demonstrated that incubation of p-aramid RFP with rat macrophages and co-cultures containing rat epithelial cells and macrophages produced shortening at 1 day and 1 week post incubation/exposure. In contrast, under identical culture conditions, incubation with a biopersistent RFP, namely cellulose, did not result in fiber shortening. Thus it was concluded that inhaled p-aramid RFP are biodegraded in the lungs, likely through an enzymatic mechanism and that this is a specific response to p-aramid RFP. A prerequisite for this activation process appears to be the interactions of RFP with acellular components of lung fluids (Warheit et al., 2001a).
The current studies were designed to test this hypothesis of p-aramid biodegradability using human lung cells in vitro. It would be inappropriate to expose humans directly to the p-aramid RFP aerosols. Therefore, it was postulated that, given the firm relationship between in vivo and in vitro studies in rats, the finding of p-aramid biodegradability in human lung cells in vitro would provide further validation for the likelihood of in vivo fiber biodegradability in humans. This would lend additional support to the safety of aerosol exposure to p-aramid RFP at the current (low) occupational exposure levels.
MATERIALS AND METHODS
Fiber preparations.
Preparations of p-aramid RFP were obtained from the DuPont Company. p-aramid RFP were prepared as respirable aerosols to ensure the relevance of the in vitro exposures relative to the effects measured in the inhalation studies. Mean RFP lengths and diameters were determined to be 16.8 μm and 0.4 μm, respectively, with a range in length dimensions of 5.2–47.0 μm. Fiber toxicology studies with this material have previously been reported (Bellmann et al., 2000; Kelly et al., 1993; Searl, 1997; Warheit et al., 1992, 1994, 2001).
Samples of cellulose RFP consisted of Thermocell mechanical wood pulp and were purchased from Laxa Bruks AB, Rofors, Sweden. This material is commonly utilized as a bitumen stabilizer in road construction and was supplied as a high-purity cellulose material. Cellulose RFP were prepared as respirable aerosols to compare to the effects measured in inhalation studies. Mean cellulose RFP lengths and diameters were determined to be 21.0 μm and 0.4 μm, respectively, with a range in length dimensions of 6.0–44.0 μm. Fiber toxicology studies with this material have previously been reported (Warheit et al., 1998, 2001a).
Generation of respirable p-aramid and cellulose RFP for in vitro studies.
Fiber materials, either p-aramid or cellulose containing respirable-sized fiber shaped particulate (RFP), were placed into the hopper of a volumetric feeder. The feeder was used to meter the fibrous material into a jet mill. The jet mill opens up the fiber material by the use of high-pressure air. The airflow carries the open fibers through the jet mill into a cyclone. The cyclone was used to size separate the fibers. In the cyclone, the large fibers fall out of the air and into a collection container at the bottom. The thinner, respirable fibers (RFP) remained in the air stream, flowed into a sampling chamber and were collected.
General experimental design.
Three types of studies were conducted to assess mechanisms of biodegradability. First, cell cultures of (1) human lung epithelial cells, (2) human primary alveolar macrophages, or (3) co-cultures of human lung epithelial cells and alveolar macrophages were treated with p-aramid RFP or cellulose RFP for periods of 1 h, 1 day, or 1 week. Subsequently, the cultures were digested with a validated hypochlorite solution (Warheit et al., 2001a), thus leaving behind the RFP. For each of these experiments, the measured endpoints were RFP lengths, as assessed using scanning electron microscopy evaluations (after the first set of experiments it was determined that the diameter dimensions were not significantly altered at any of the time points, for any of the cell culture-types or fiber-types evaluated). Therefore, only the length dimensions were measured and recorded. In addition, each experiment was repeated at least 4 (p-aramid) or 3 times (cellulose) for each cell culture-type and each time period.
A second preliminary "double-dose" experiment was conducted to determine whether the addition of additional macrophages to the culture could accelerate the cleavage kinetics following incubation of cells with p-aramid RFP. In these studies a two-fold increase in the numbers of human alveolar macrophages (plated on days 1 and 3) were utilized to determine whether the cleavage kinetics of p-aramid RFP could be accelerated.
Third, preparations of p-aramid RFP were incubated in vitro with either phosphate buffered saline or acellular (i.e., cell-free) components of human BAL fluids or rat BAL fluids and then processed through a simulated digestion method (i.e., no cells or tissues were digested). For these experiments, the measured endpoint was RFP length, as assessed using scanning electron microscopy evaluations. These experiments were repeated at least two additional times.
Human pulmonary lavage.
The method used in the recovery of alveolar macrophages from human subjects has been previously described (Ghio et al., 1998). Briefly, volunteers were recruited through pamphlets posted and displayed around the campus of University of North Carolina (UNC) at Chapel Hill and advertisements in the local and University newspapers. To qualify as a normal (healthy) research subject for bronchoscopy, a respondent must have been between 18 and 40 years of age, have had no active history of allergies or respiratory disease, must have been a nonsmoker for at least 5 years, and was not presently on any medication prescribed by a physician (except birth control).
Subjects received a physical examination in the Human Studies Division facilities on the UNC-CH Campus. A blood sample (15cc) was obtained for an SMA-20 serum chemistry screen and a complete blood count was evaluated via differential analyses. A urine pregnancy test was administered to all female subjects and a positive result excluded subjects from further participation. Subjects were also given a brief physical exam to assess their suitability to undergo fiberoptic bronchoscopy.
Prior to participation in the study, subjects were informed of the procedures and potential risks, and each signed a statement of informed consent. The protocol and consent procedures were approved by the University of North Carolina School of Medicine Committee on the Protection of the Rights of Human Subjects. To eliminate complications from anesthesia, the total dose of lidocaine instilled through the bronchoscope was limited during the procedure and no sedatives or narcotics were used. The bronchoscopy and corresponding lavage occurred through a transnasal procedure.
Ten million lavaged human macrophages in RPMI-1640 culture media (Roswell Park Memorial Institute), containing gentamicin and 5% fetal bovine serum, as well as human bronchoalveolar lavage fluid were sent overnight on ice from the USEPA site in Chapel Hill, NC, to DuPont Haskell Laboratory in Newark, DE, by Federal Express Corp. and arrived at Haskell the next day. The cells were counted and evaluated for viability and cell differential analyses according to procedures previously reported (Warheit et al., 1991). Cells were usually available on a twice per month basis.
In vitro cellular studies.
The human lung epithelial cell line A549 was purchased from American Type Culture Collection (Rockville, MD) and grown in F-12K (Kaighn's modification of Ham's F-12) (10% FBS + Pen-Strep solution) at 37°C and 5% CO2. Following passage from culture flasks, approximately 570,000 A549 cells were added to Thermanox plastic disks contained within 10 x 35 mm petri dishes. The cells were incubated for four days to allow the formation of monolayers on the Thermanox disks. On day four, 500 μl of a standardized preparation of p-aramid (2100 RFP/cc) or cellulose (1800 RFP/cc) was added to the culture for either 1 h, 1 day, or 1 week. At the end of the incubation period, the cells were fixed with Karnovsky's solution and either underwent SEM evaluation or a tissue digestion procedure with a 1.25% hypochlorite solution (i.e., 25% Clorox bleach) at 60°C for 9 min. This time course of this digestion procedure had been validated in earlier methods development-type experiments, either with p-aramid exposed lung tissues or in vitro co-cultures exposed to p-aramid RFP or cellulose RFP, as well as lung tissue recovered from Nylon-exposed rats. The tissue digestion method for 9 min is critical and selectively digested the cells but not the p-aramid or cellulose RFP, which were filtered and processed for scanning electron microscopy evaluation of length dimensions (Fig. 2). Similar numbers of A549 epithelial cells were plated in the "A549" and the Co-culture groups. Equivalent numbers (to the A549 cells) of human alveolar macrophages were plated in the HBAL and the Co-culture groups. As a consequence, there were more total cells plated in the Co-culture groups.
Rat pulmonary lavage.
BAL procedures were conducted according to methods previously described (Warheit et al., 1984, 1992). The alveolar macrophages recovered from the animals were utilized for other in vitro experiments ongoing in the laboratory. Lavaged fluids from normal CD rats were centrifuged at 250 x g, and the supernatant was removed and the cells were resuspended in Ham's F-12K, supplemented with penicillin and streptomycin. Cell numbers were quantified and the acellular BAL fluid component was utilized for studies described below.
Acellular in vitro biodegradability studies.
The p-aramid RFP were incubated with acellular components of human or rat bronchoalveolar lavage fluids or a PBS control for 3 h. At the end of the incubation period, the "acellular cultures" were processed through a "simulated" tissue digestion procedure with a 1.25% hypochlorite solution (i.e., 25% Clorox bleach) at 60°C for 9 min. This step was followed by processing for SEM and corresponding quantification of RFP lengths and compared with control (PBS-incubated) samples.
Digestion procedures.
The lung digestion techniques used in this study have previously been reported (Searl, 1997; Warheit et al., 1992, 1997a). Following incubation with p-aramid or cellulose RFP, the cells were fixed with Karnovsky's solution and then underwent a tissue digestion procedure with a 1.25% hypochlorite solution (i.e., 25% Clorox bleach) at 60°C for 9 min.
Scanning electron microscopy assessment of RFP lengths.
Filters containing p-aramid or cellulose RFP were prepared for scanning electron microscopy. Evaluations of RFP lengths were conducted on a calibrated JEOL 840 scanning electron microscope. The assessments were made at magnifications of 800x–5000x. A minimum of 100 RFP were measured and recorded for each cell culture group at each postexposure time period. For the p-aramid experiments, there were 4 groups each for each cell culture-type/each time period. Digital micrographs were made for each field of vision. Subsequently, the micrographs were printed and lengths of each individual RFP were measured and recorded.
Statistics.
For the in vitro cellular studies with p-aramid RFP, A549 human lung epithelial cells, primary human macrophages (HBAL), as well as macrophage-epithelial cell co-cultures of (Co) were treated with p-aramid RFP or cellulose RFP. The lengths of the individual RFP were measured after 1 h, 24 h, and 1 week. The hypothesis was tested that there were differences in the mean RFP lengths of each treatment group at the end of each time interval. The objective of this experiment was to compare the mean RFP lengths of each cell-type across the three time points, and within a time point, to compare the mean RFP length of the three cell culture-types (i.e., A549, HBAL, co-culture). A full factorial two-factor analysis of variance (SAS, Cary, NC) was used to make these comparisons. For several cell-type by time combinations, there were two or more groups of data available, so a nested variance structure was used. The data for this was provided in an Excel file. The criterion for statistical significance for the nonadjusted differences was set at p < 0.05. The p values for these differences were then adjusted using the Tukey-Kramer multiple comparison adjustment.
For the in vitro acellular studies, preparations of p-aramid RFP were incubated in vitro either with PBS, human BAL fluid, or rat BAL fluid and processed through a simulated digestion method. The objective here was to compare the three processing types on p-aramid RFP lengths. The hypothesis tested was that the mean RFP lengths of the groups incubated with human or rat BAL fluids were different from the mean RFP lengths of the groups incubated with PBS. A one-way analysis of variance (SAS Institute, Cary, NC) was used for this with a nested variance structure reflecting three physical experiments and cell cultures within experiment. Data for this analysis was given in an Excel file. The p values for the nonadjusted differences were set 0.05. The p values for these differences were then adjusted using the Tukey-Kramer multiple comparison adjustment.
RESULTS
Cellular differential analyses of cytocentrifuge preparations of human and rat BAL cells are presented in Figure 3. Human BAL cells consisted of 90% macrophages and 10% lymphocytes. Greater than 98% of rat BAL cells consisted of macrophages. Although rat BAL cells were not used in these experiments, it was important to determine that the rat BAL fluids were free of inflammatory cells.
Data from the in vitro human cellular studies with p-aramid RFP demonstrated that after 1 h there was no statistical evidence of differences in the mean RFP lengths among the groups. However, as demonstrated in Figures 6a and 6b, at 1-day and 1-week post-incubation, the mean p-aramid RFP lengths in the co-cultures or HBAL cells culture groups were significantly smaller than that of the A549 cell culture, but the co-culture and HBAL cell culture groups were indistinguishable from one another at each time point. Furthermore, the mean lengths at the 1-day or 1-week time points were significantly reduced compared to the 1-h time point for each cell-type. There were no significant differences between the 1-day and 1-week values for any cell-type (Fig. 4).
Data from the in vitro human cellular studies with cellulose RFP demonstrated no differences in the mean RFP lengths among the three groups at any of the three time points (i.e., 1-h, 1-day, or 1-week post-incubation) (Fig. 5).
Data from the in vitro "double dose" cellular studies with p-aramid RFP demonstrated that the exposures of p-aramid RFP to double doses of human macrophages (plated on days 1 and 3) did not accelerate p-aramid RFP shortening (Fig. 6).
Data from the in vitro acellular studies with p-aramid RFP demonstrated that the combination of incubation in human or rat BAL fluids along with simulated digestion resulted in a significant reduction in mean lengths when compared to the p-aramid RFP incubated with PBS and undergoing simulated digestion (Fig. 7).
Scanning electron micrographs of representative epithelial, and epithelial-macrophage co-cultures are presented in Figures 8–9 , which illustrate cellular-fibril interactions prior to digestion and counting.
DISCUSSION
Biopersistence of fibers is broadly defined as the extended time period of fiber retention in the respiratory tract or other tissues following deposition. A variety of inhalation studies in rats with inorganic synthetic vitreous fiber-types have demonstrated a strong, direct relationship between the biopersistence of a fiber in the lung and its potential to cause adverse effects (Hesterberg et al., 1996). Some evidence suggests that this relationship may also be operative with man-made organic fiber-types (Warheit et al., 2001b). Durability and fiber length are the two most critical determinants of lung biopersistence. The degree of biopersistence may be influenced by a variety of factors including the inhaled fiber burden (i.e., dose or number of respirable fibers present in the respiratory tract), fiber dimensions, surface characteristics, chemical composition, and other factors. Alterations in any of these parameters could influence the development of fiber-related pulmonary toxicity, as the enhanced length of contact between retained fibers and lung tissues increases the potential for adverse effects. Biodegradability refers to the breakdown and/or dissolution of fibers in the respiratory tract. Mechanical, chemical, and/or enzymatic factors may be operative in the clearance process. Fibers may be cleared from the respiratory tract via macrophage uptake and transport via the airways, transport via fluid fluxes on the mucociliary escalator or by dissolution processes (ECETOC, 1996). It generally is regarded that short fibers are more readily cleared from the lungs than longer fibers, primarily via alveolar macrophage phagocytosis and airway mucociliary clearance effects (Morgan and Holmes, 1984). Therefore, the demonstrated transverse cleavage and corresponding shortening of long p-aramid RFP in the lungs of exposed rats and hamsters (1) bodes well to facilitate a more rapid clearance of these RFP and (2) reduces the likelihood of developing fiber-related adverse pulmonary effects.
The results of in vitro mechanistic studies with human lung cells demonstrated that incubation of p-aramid RFP with primary macrophages or macrophage-epithelial cell co-cultures, but not epithelial cells alone, produced shortening of RFP at 1 day and 1 week postexposure, but not after a single hour of incubation. This is an interesting finding because the co-culture groups contained more plated cells when compared to the HBAL cultures. But this result is not surprising since it would appear that the alveolar macrophage component was likely responsible for the shortening of p-aramid RFP. In addition, utilizing a double dose of macrophages did not accelerate the rate of p-aramid RFP shortening. In contrast to the data with p-aramid RFP, incubation of cell cultures with biopersistent cellulose RFP produced no shortening in any of the cell-types at any time postexposure.
In another set of experiments, p-aramid RFP were incubated with PBS, acellular lavaged fluids from human subjects or rats, followed by a simulated tissue digestion and then assessed for changes in p-aramid length dimensions. Incubation with either human or rat BAL fluids produced substantial shortening of the p-aramid RFP compared to phosphate-buffered saline.
Taken together, these results support the hypothesis that components of human lung cells and fluids facilitate the biodegradability of p-aramid RFP.
Previously, we have reported that p-aramid RFP are biodegraded in the lungs of exposed rats and hamsters following inhalation exposures (Warheit et al., 1992, 1997a). In addition, several other investigators have demonstrated shortening of p-aramid RFP leading to more rapid clearance of the longest p-aramid fibrils during the months following exposure. It is interesting to note that this clearance pattern is associated with an initial increase in the numbers of retained shorter fibrils, i.e., an indication of cleavage of long RFP into shorter ones, and subsequent rapid RFP clearance thereafter (Bellmann et al., 2000; Kelly et al., 1993; Searl, 1997; Warheit et al., 1992, 1994).
A variety of inhalation toxicity studies have been conducted in rats with p-aramid RFP demonstrating biodegradability of RFP. In evaluating deposition and clearance patterns, Warheit et al. (1992) exposed rats to aerosols of p-aramid RFP for 2 weeks and followed RFP clearance patterns in the lung for a postexposure period of 1 year. A transient increase in the numbers of retained fibrils was noted at one week after exposure, with rapid clearance of RFP thereafter, and a retention half-time of 30 days. Following a six-month postexposure period, there was a measured progressive decrease in the mean lengths of lung-retained RFP from 12.5 (immediately after exposure) to 7.5 μm and a slight decrease in mean diameter from 0.3 to 0.2 μm. The percentages of retained RFP > 15 μm in length were reduced from 30% immediately after exposure to 5% after 6 months; the corresponding percentages of retained p-aramid RFP in the 4–7 μm length range was increased from 25 to 55% during the same period (Warheit et al., 1992).
Kelly et al. (1993) assessed the clearance kinetics in rats of lung-retained p-aramid RFP during a two-year inhalation study (Lee et al., 1988) using PCOM (phase contrast microscopy) methods. Mean lengths of inhaled p-aramid RFP in the aerosol chamber were determined to be 12 μm. Following 2 years of exposure at concentrations of 2.5, 25, or 100 f/ml, or 1 year of exposure at 400 f/ml plus 1 year recovery, mean lengths were determined to be 4 μm in RFP recovered from the lungs of rats. In addition, the time required for retained RFP to be reduced to <5 μm in the lung was greatest at the higher concentration and shortest at the lowest aerosol concentration.
Searl (1997) exposed rats to aerosols of three different fiber-types, namely respirable (1) p-aramid RFP, (2) chrysotile asbestos fibers, and (3) Code 100/475 glass fibers, and assessed the relative biopersistence/retention characteristics after 10-day inhalation exposures at concentrations of 700 f/ml. The most significant reduction in lung fiber burden was measured during the first 3 months after exposure, but the pattern of clearance of different size classes varied with each fiber-type. For example, inhalation of p-aramid RFP resulted in rapid clearance of the longest RFP during the first month postexposure, and this was associated with an initial enhancement of the numbers of shorter RFP. Searl concluded that this clearance pattern was consistent with cleavage of p-aramid RFP into successively shorter fragments which subsequently facilitated rapid clearance by alveolar macrophages. Similar clearance patterns were measured with the inhaled glass fibers. In contrast, exposures to chrysotile asbestos fibers initially induced a rapid clearance in the numbers of inhaled short fibers relative to long fibers, and this is consistent with preferential clearance of short fibers by macrophages. However, the longer chrysotile fibers were retained for prolonged or indefinite periods in the lung. Accordingly, the author concluded that biopersistence of long (>15 μm) chrysotile asbestos fibers was substantially greater than that of long RFP of p-aramid or glass fibers, indicating that the subpopulation of long, chrysotile respirable fibers was selectively retained in the lung without any apparent clearance.
Bellmann et al. (2000) also evaluated clearance kinetics of p-aramid RFP in the lungs of exposed rats. Male Wistar rats inhaled aerosol concentrations of 50, 200, or 800 p-aramid RFP/ml for 3 months. The average lengths of retained p-aramid RFP were progressively shortened during the postexposure period of 3 days to 3 months. Moreover, the numbers of RFP < 10 μm in length actually were increased at 1 month postexposure when compared to the 3-day recovery, whereas for the RFP fraction with lengths above 10 μm the number of RFP per lung decreased. This pulmonary clearance effect is likely explained by cleavage of retained, longer RFP (length > 10 μm) in the lungs of exposed animals. The investigators concluded that the clearance kinetics of RFP > 20 μm in length was associated with the fastest elimination. Therefore, breakage of long RFP best explained the experimental observations. These p-aramid RFP clearance patterns are consistent with the findings reported by Warheit et al. (1992, 1994), Kelly et al. (1993), and Searl (1997).
More recently we have carried out both in vivo and in vitro cellular and acellular investigations with p-aramid and cellulose RFP in rats. First, p-aramid or cellulose RFP were instilled into the lungs of rats and the lungs digested 24 h postexposure using two different digestion techniques: (1) a conventional ethanolic KOH method, and (2) an enzymatic method which simulates the action of lung enzymes. Cellulose RFP were utilized as a control organic fiber-type which is known to be biopersistent. The enzyme method demonstrated shortening of the p-aramid RFP, but the KOH method did not. In addition, neither method was successful in shortening the cellulose RFP. In addition, standardized preparations of p-aramid RFP or cellulose RFP were incubated with saline or lung fluids and processed by one of 2 tissue digestion techniques. Mean lengths of p-aramid RFP incubated with saline and processed with KOH or the enzyme method were not altered. The preparation of p-aramid RFP that had been incubated with BAL fluids and processed with the enzyme solution resulted in cleavage of p-aramid RFP. In contrast to the in vitro acellular studies with p-aramid RFP, the combination of BAL fluid incubation and enzyme digestion method had no measurable effect on shortening of cellulose RFP, indicating that the results with p-aramid RFP were specific. In a final set of in vitro cellular studies, cultures of rat lung epithelial cells, alveolar macrophages, or co-cultures of epithelial cells and macrophages were treated with p-aramid RFP for 1 h, 1 day, or 1 week to determine whether RFP shortening occurs directly in the phagocytic cells. The results demonstrated that (1) no significant degree of shortening occurred in the epithelial cell cultures at any time point; however, in the macrophage and co-cultures, cleavage of p-aramid RFP was observed at 1 day and 1 week postexposure. We concluded from this study that components of lung fluids coat the p-aramid RFP as a prerequisite for enzymatic cleavage. (Warheit et al., 2001a).
The data from the in vitro "double dose" cellular studies with p-aramid RFP demonstrated that the exposures of p-aramid RFP to double doses of human macrophages (plated on test days 1 and 3) did not accelerate p-aramid RFP shortening. This preliminary finding, although based on only two experiments, was a minor surprise. An alternative interpretation is that most of the RFP shortening occurred after the first 24 hours of incubation, with very little additional cleavage occurring at the 1 week postexposure time period. Thus, perhaps it is not surprising that additional shortening did not occur at 3 days postexposure. Unfortunately, sufficient numbers of human macrophages were not available at the 1 day postexposure time period to test whether a true double dose of macrophages would have accelerated fiber shortening.
To summarize our findings, the studies reported herein were developed to test a hypothesis of p-aramid biodegradability using human lung cells in vitro, and to determine whether the same mechanisms, previously demonstrated in rat lung cells might be operative in human lung cells. Absent in vivo human data with p-aramid RFP, it was postulated that p-aramid biodegradability in human lung cells in vitro would provide further validation for the likelihood of in vivo fiber biodegradability in humans, and lend additional information on the safety of aerosol exposures to p-aramid RFP. As indicated in Figure 13, using a Hazard Assessment Bridging Paradigm—the demonstration of p-aramid biodegradability in the lungs of rats (in vivo) and the corresponding mechanism in rat lung cells (in vitro) represents a defined relationship. Confirmation of these measured effects of p-aramid RFP shortening in human cultured lung cells supports the conclusions previously measured in lungs cells and lungs of exposed rats. These measured effects are selective for p-aramid RFP and were not measured in similar studies with cellulose RFP, a biopersistent fiber-type. As a consequence, these data provide strength, via an extrapolated relationship, to the notion that inhaled p-aramid RFP appear to be biodegradable in the lungs of humans (Fig. 10).
The findings presented here reinforce the logic derived from animal lung exposure studies which suggest that inhaled p-aramid RFP biodegrade in the lungs of humans. As a contribution towards improved hazard and risk assessments, the methodology introduced in these studies has (1) demonstrated the practical viability of in vitro exposure testing with human lung cells, and (2) "bridged" in vitro animal and human cell exposure studies through systematic design and execution of experiments, in comparison to in vivo animal studies under equivalent conditions.
NOTES
1 The authors ceritify that all research involving human subjects was done under full compliance with all government policies and the Helsinki Declaration.
ACKNOWLEDGMENTS
This study was supported by the DuPont Advanced Fiber Systems Strategic Business Unit and by the Teijin Twaron Company. The authors wish to thank Drs. Odorich von Susani, Chris Braun, Jan Roos, Fred Pfister, Gerald Kennedy, and Ted Merriman for helpful insights into the preparation of the manuscript. We also thank Dr. John W. Green for providing statistical analyses of the data, and Dr. Liang Liang for technical assistance with the scanning electron microscope.
REFERENCES
Bellmann, B., Creutzenberg, O., Dasenbrock, C., Ernst, H., Pohlmann, G., and Muhle, H. (2000). Inhalation tolerance study of p-aramid respirable fiber-shaped particulates (RFP) in rats. Toxicol. Sci. 54, 237–250.
ECETOC. (1996). Toxicology of Man-Made Organic Fibres. Technical Report No. 69. European Centre for Ecotoxicology and Toxicology of Chemicals, Brussels, Belgium. p. 41.
Ghio, A. J., Bassett, M., Chall, A. N., Levin, D. G., and Bromberg, P. A. (1998). Bronchoscopy in healthy volunteers. J. Bronchology 5, 185–194.
Hesterberg, T. W., Miiller, W. C., Musselman, R. P., Kamstrup, O., Hamilton, R. D., and Thevenaz, P. (1996). Biopersistence of man-made vitreous fibers and crocidolite in rat lung following inhalation. Fundam. Appl. Toxicol. 29, 267–279.
Kelly, D. P., Merriman, E. A., Kennedy, G. L. Jr., and Lee, K. P. (1993). Deposition, clearance, and shortening of Kevlar para-aramid fibrils in acute, subchronic, and chronic inhalation studies in rats. Fundam. Appl. Toxicol. 21, 345–354.
Lee, K. P., Kelly, D. P., O'Neal, F. O., Stadler, J. C., and Kennedy, G. L., Jr. (1988). Lung response to ultrafine Kevlar aramid synthetic fibrils following 2-year inhalation exposure in rats. Fundam. Appl. Toxicol. 11, 1–20.
Morgan, A., and Holmes, A. (1984). The deposition of MMMF in the respiratory tract of the rat, their subsequent clearance, solubility in vivo and protein coating. In Biological Effects of Man-Made Fibres, Vol. 2 (T. Guthe, Ed.), pp. 1–17. WHO, Copenhagen, Denmark.
Searl, A. (1997). A comparative study of the clearance of respirable para-aramid, chrysotile and glass fibres from rat lungs. Ann. Occup. Hyg. 41, 217–233.
Searl, A., and Cullen, R. T. (1997). An enzymatic tissue digestion method for fibre biopersistence studies. Ann. Occup. Hyg. 41, 721–727.
Warheit, D. B., Hill, L. H., and Brody, A. R. (1984). Surface morphology and correlated phagocytic capacity of pulmonary macrophages lavaged from the lungs of rats. Exp. Lung Res. 6, 71–82.
Warheit, D. B., Carakostas, M. C., Hartsky, M. A., and Hansen, J. F. (1991). Development of a short-term inhalation bioassay to assess pulmonary toxicity of inhaled particles: Comparisons of pulmonary responses to carbonyl iron and silica. Toxicol. Appl. Pharmacol. 107, 350–368.
Warheit, D. B., Kellar, K. A., and Hartsky, M. A. (1992). Pulmonary cellular effects in rats following aerosol exposures to ultrafine Kevlar aramid fibrils: Evidence for biodegradability of inhaled fibrils. Toxicol. Appl. Pharmacol. 116, 225–239.
Warheit, D. B., Hartsky, M. A., McHugh, T. A., and Kellar, K. A. (1994). Biopersistence of inhaled organic and inorganic fibers in the lungs of rats. Environ. Health Perspect. 102(Suppl. 5), 151–157.
Warheit, D. B., Snajdr, S. I., Hartsky, M. A., and Frame, S. R. (1997a). Pulmonary responses to inhaled para-aramid fibrils in hamsters: Evidence of biodegradability in the lungs of a second rodent species. Inhal. Toxicol. 9, 181–187.
Warheit, D. B., Snajdr, S. I., Hartsky, M. A., and Yuen, I. S. (1997b). Assessments of in vitro effects of dusts on pulmonary cells using macrophage-epithelial cells in co-culture. In Correlations between in Vitro and in Vivo Investigations in Inhalation Toxicology (U. Mohr, D. L. Dungworth, K. B. Adler, C. C. Harris, and C. G. Plopper, Eds.), pp. 432–446. ILSI Press, Washington, DC.
Warheit, D. B., Snajdr, S. I., Hartsky, M. A., and Frame, S. R. (1998). Two-week inhalation study in rats with cellulose fibers. In Advances in the Prevention of Occupational Respiratory Disease (K. Chiyotani, Y. Hosoda, and Y. Aizawa, Eds.), pp. 579–582. Elsevier Science BV, Amsterdam.
Warheit, D. B., Hartsky, M. A., Reed, K. L., and Webb, T. R. (2001a). Biodegradability of inhaled para-aramid RFP: Mechanistic in vivo and in vitro studies. Toxicol. Appl. Pharmacol. 174, 78–88.
Warheit, D. B., Hart, G. A., Hesterberg, T. W., Collins, J. J., Dyer, W. M., Swaen, G. M. H., Castranova, V., Soiefer, A. I., and Kennedy, Jr., G. L. (2001b). Potential pulmonary effects of man-made organic fiber (MMOF) dusts. Crit. Rev. Toxicol. 31, 697–736.(D. B. Warheit, K. L. Reed, J. D. Stonehu)
National Health and Environmental Effects Research Laboratory, US Environmental Protection Agency, Chapel Hill, North Carolina
2 To whom correspondence should be addressed at DuPont Haskell Lab, 1090 Elkton Road, Newark, DE 19714-0050. Fax: (302) 366-5207. E-mail: david.b.warheit@usa.dupont.com.
ABSTRACT
Using both in vivo (inhalation) and in vitro (cell culture) studies, we previously reported that p-aramid respirable fibers (RFP—defined as respirable-sized fiber-shaped particulates) are biodegraded in lungs and lung cells of rats following exposures. The current studies were undertaken to determine whether shortening mechanisms of p-aramid RFP biodegradability are also operative in human lung cells. Cultures of human A549 lung epithelial cells (A549), primary alveolar macrophages (HBAL) (collected via bronchoalveolar lavage [BAL]) from volunteers), and co-cultures (Co) of the A549 and HBAL were incubated with p-aramid RFP for either 1 h, 1 day, or 1 week to assess RFP shortening. Lengths of RFP were measured using scanning electron microscopy (SEM) following fixation, digestion of culture tissue components, and processing. Similar to findings using rat lung cells, only slight RFP shortening was measured in A549 cultures at 1-day and 1-week post-incubation. More importantly, in HBAL and Co groups, greater transverse cleavage of p-aramid RFP was measured at 1-day and 1-week postexposure compared to 1-h HBAL or Co groups, or in any A549 groups. In contrast, cellulose RFP, a biopersistent reference control fiber, were not measurably shortened under similar circumstances. Second, p-aramid RFP were incubated either with phosphate-buffered saline (PBS), or acellular BAL fluids from human volunteers or rats and processed for SEM analysis of RFP lengths. Mean lengths of p-aramid RFP incubated with human or rat BAL fluids were substantially decreased compared to PBS. Similar to our findings with rat lung cells, components of human lung fluids coat the p-aramid RFP as a prerequisite for subsequent enzymatic cleavage by human phagocytic lung cells and this finding reinforces the concept that inhaled p-aramid RFP are likely to be biodegradable in the lungs of humans.
Key Words: p-aramid respirable fibers; human lung cells; fiber shortening mechanisms.
INTRODUCTION
Para-aramid fibers have diameter dimensions which are considered to be nonrespirable (i.e., in the range of 12–15 μm), but fibrils (i.e., p-aramid RFP) with diameter dimensions of 0.3–0.7 μm, and lengths up to 100 μm, can be produced under conditions of abrasion (Fig. 1). Para-aramid fibers in the form of pulp contain RFP as an integral component (ECETOC, 1996). Pulp is utilized commercially in gaskets and in friction products such as brake linings.
Unlike most other organic fiber-types, the pulmonary toxicity effects of p-aramid RFP inhalation have been extensively tested. In this regard, the results of inhalation toxicity studies in rats and hamsters from several different laboratories throughout the world consistently have demonstrated that after an initial buildup of inhaled RFP in the lung, there is a rapid clearance of the longest fibrils, concomitant with an initial increase in the numbers of shorter fibrils (Bellmann et al., 2000; Kelly et al., 1993; Searl, 1997; Warheit et al., 1992, 1994). The cardinal feature of this RFP clearance pattern in the lungs of exposed animals is a progressive decrease in the average lengths of retained RFP with increasing postexposure time period intervals (Warheit et al., 1992). This mechanism has been reported to occur via transverse cleavage, i.e., shortening of the retained RFP, which occurs with increasing residence time in the lung.
In earlier methods-development studies, Searl and Cullen (1997) proposed that p-aramid RFP cleavage occurred as a result of the synergistic effects of coating of RFP with lung fluids along with enzymatic digestion of the fibril. In previously reported studies, we developed an in vitro culture system with rat lung cells to (1) test the validity of this hypothesis; and (2) develop an in vitro system that might be utilized as a bridging technique to assess the biodegradability of p-aramid RFP for human lung cells.
Several studies have been undertaken to elucidate mechanisms of p-aramid biodegradability. Based on preliminary findings by Searl (1997), who reported that the enzymatic digestion method artificially cleaved p-aramid RFP after instillation into the lungs, we sought to test the hypothesis that (1) inhaled p-aramid RFP are biodegraded in the lungs of exposed animals via an enzymatic mechanism; and (2) this process of biodegradation is facilitated by coating of the RFP with lung fluids.
The results of our earlier study with rat cells in vitro demonstrated that incubation of p-aramid RFP with rat macrophages and co-cultures containing rat epithelial cells and macrophages produced shortening at 1 day and 1 week post incubation/exposure. In contrast, under identical culture conditions, incubation with a biopersistent RFP, namely cellulose, did not result in fiber shortening. Thus it was concluded that inhaled p-aramid RFP are biodegraded in the lungs, likely through an enzymatic mechanism and that this is a specific response to p-aramid RFP. A prerequisite for this activation process appears to be the interactions of RFP with acellular components of lung fluids (Warheit et al., 2001a).
The current studies were designed to test this hypothesis of p-aramid biodegradability using human lung cells in vitro. It would be inappropriate to expose humans directly to the p-aramid RFP aerosols. Therefore, it was postulated that, given the firm relationship between in vivo and in vitro studies in rats, the finding of p-aramid biodegradability in human lung cells in vitro would provide further validation for the likelihood of in vivo fiber biodegradability in humans. This would lend additional support to the safety of aerosol exposure to p-aramid RFP at the current (low) occupational exposure levels.
MATERIALS AND METHODS
Fiber preparations.
Preparations of p-aramid RFP were obtained from the DuPont Company. p-aramid RFP were prepared as respirable aerosols to ensure the relevance of the in vitro exposures relative to the effects measured in the inhalation studies. Mean RFP lengths and diameters were determined to be 16.8 μm and 0.4 μm, respectively, with a range in length dimensions of 5.2–47.0 μm. Fiber toxicology studies with this material have previously been reported (Bellmann et al., 2000; Kelly et al., 1993; Searl, 1997; Warheit et al., 1992, 1994, 2001).
Samples of cellulose RFP consisted of Thermocell mechanical wood pulp and were purchased from Laxa Bruks AB, Rofors, Sweden. This material is commonly utilized as a bitumen stabilizer in road construction and was supplied as a high-purity cellulose material. Cellulose RFP were prepared as respirable aerosols to compare to the effects measured in inhalation studies. Mean cellulose RFP lengths and diameters were determined to be 21.0 μm and 0.4 μm, respectively, with a range in length dimensions of 6.0–44.0 μm. Fiber toxicology studies with this material have previously been reported (Warheit et al., 1998, 2001a).
Generation of respirable p-aramid and cellulose RFP for in vitro studies.
Fiber materials, either p-aramid or cellulose containing respirable-sized fiber shaped particulate (RFP), were placed into the hopper of a volumetric feeder. The feeder was used to meter the fibrous material into a jet mill. The jet mill opens up the fiber material by the use of high-pressure air. The airflow carries the open fibers through the jet mill into a cyclone. The cyclone was used to size separate the fibers. In the cyclone, the large fibers fall out of the air and into a collection container at the bottom. The thinner, respirable fibers (RFP) remained in the air stream, flowed into a sampling chamber and were collected.
General experimental design.
Three types of studies were conducted to assess mechanisms of biodegradability. First, cell cultures of (1) human lung epithelial cells, (2) human primary alveolar macrophages, or (3) co-cultures of human lung epithelial cells and alveolar macrophages were treated with p-aramid RFP or cellulose RFP for periods of 1 h, 1 day, or 1 week. Subsequently, the cultures were digested with a validated hypochlorite solution (Warheit et al., 2001a), thus leaving behind the RFP. For each of these experiments, the measured endpoints were RFP lengths, as assessed using scanning electron microscopy evaluations (after the first set of experiments it was determined that the diameter dimensions were not significantly altered at any of the time points, for any of the cell culture-types or fiber-types evaluated). Therefore, only the length dimensions were measured and recorded. In addition, each experiment was repeated at least 4 (p-aramid) or 3 times (cellulose) for each cell culture-type and each time period.
A second preliminary "double-dose" experiment was conducted to determine whether the addition of additional macrophages to the culture could accelerate the cleavage kinetics following incubation of cells with p-aramid RFP. In these studies a two-fold increase in the numbers of human alveolar macrophages (plated on days 1 and 3) were utilized to determine whether the cleavage kinetics of p-aramid RFP could be accelerated.
Third, preparations of p-aramid RFP were incubated in vitro with either phosphate buffered saline or acellular (i.e., cell-free) components of human BAL fluids or rat BAL fluids and then processed through a simulated digestion method (i.e., no cells or tissues were digested). For these experiments, the measured endpoint was RFP length, as assessed using scanning electron microscopy evaluations. These experiments were repeated at least two additional times.
Human pulmonary lavage.
The method used in the recovery of alveolar macrophages from human subjects has been previously described (Ghio et al., 1998). Briefly, volunteers were recruited through pamphlets posted and displayed around the campus of University of North Carolina (UNC) at Chapel Hill and advertisements in the local and University newspapers. To qualify as a normal (healthy) research subject for bronchoscopy, a respondent must have been between 18 and 40 years of age, have had no active history of allergies or respiratory disease, must have been a nonsmoker for at least 5 years, and was not presently on any medication prescribed by a physician (except birth control).
Subjects received a physical examination in the Human Studies Division facilities on the UNC-CH Campus. A blood sample (15cc) was obtained for an SMA-20 serum chemistry screen and a complete blood count was evaluated via differential analyses. A urine pregnancy test was administered to all female subjects and a positive result excluded subjects from further participation. Subjects were also given a brief physical exam to assess their suitability to undergo fiberoptic bronchoscopy.
Prior to participation in the study, subjects were informed of the procedures and potential risks, and each signed a statement of informed consent. The protocol and consent procedures were approved by the University of North Carolina School of Medicine Committee on the Protection of the Rights of Human Subjects. To eliminate complications from anesthesia, the total dose of lidocaine instilled through the bronchoscope was limited during the procedure and no sedatives or narcotics were used. The bronchoscopy and corresponding lavage occurred through a transnasal procedure.
Ten million lavaged human macrophages in RPMI-1640 culture media (Roswell Park Memorial Institute), containing gentamicin and 5% fetal bovine serum, as well as human bronchoalveolar lavage fluid were sent overnight on ice from the USEPA site in Chapel Hill, NC, to DuPont Haskell Laboratory in Newark, DE, by Federal Express Corp. and arrived at Haskell the next day. The cells were counted and evaluated for viability and cell differential analyses according to procedures previously reported (Warheit et al., 1991). Cells were usually available on a twice per month basis.
In vitro cellular studies.
The human lung epithelial cell line A549 was purchased from American Type Culture Collection (Rockville, MD) and grown in F-12K (Kaighn's modification of Ham's F-12) (10% FBS + Pen-Strep solution) at 37°C and 5% CO2. Following passage from culture flasks, approximately 570,000 A549 cells were added to Thermanox plastic disks contained within 10 x 35 mm petri dishes. The cells were incubated for four days to allow the formation of monolayers on the Thermanox disks. On day four, 500 μl of a standardized preparation of p-aramid (2100 RFP/cc) or cellulose (1800 RFP/cc) was added to the culture for either 1 h, 1 day, or 1 week. At the end of the incubation period, the cells were fixed with Karnovsky's solution and either underwent SEM evaluation or a tissue digestion procedure with a 1.25% hypochlorite solution (i.e., 25% Clorox bleach) at 60°C for 9 min. This time course of this digestion procedure had been validated in earlier methods development-type experiments, either with p-aramid exposed lung tissues or in vitro co-cultures exposed to p-aramid RFP or cellulose RFP, as well as lung tissue recovered from Nylon-exposed rats. The tissue digestion method for 9 min is critical and selectively digested the cells but not the p-aramid or cellulose RFP, which were filtered and processed for scanning electron microscopy evaluation of length dimensions (Fig. 2). Similar numbers of A549 epithelial cells were plated in the "A549" and the Co-culture groups. Equivalent numbers (to the A549 cells) of human alveolar macrophages were plated in the HBAL and the Co-culture groups. As a consequence, there were more total cells plated in the Co-culture groups.
Rat pulmonary lavage.
BAL procedures were conducted according to methods previously described (Warheit et al., 1984, 1992). The alveolar macrophages recovered from the animals were utilized for other in vitro experiments ongoing in the laboratory. Lavaged fluids from normal CD rats were centrifuged at 250 x g, and the supernatant was removed and the cells were resuspended in Ham's F-12K, supplemented with penicillin and streptomycin. Cell numbers were quantified and the acellular BAL fluid component was utilized for studies described below.
Acellular in vitro biodegradability studies.
The p-aramid RFP were incubated with acellular components of human or rat bronchoalveolar lavage fluids or a PBS control for 3 h. At the end of the incubation period, the "acellular cultures" were processed through a "simulated" tissue digestion procedure with a 1.25% hypochlorite solution (i.e., 25% Clorox bleach) at 60°C for 9 min. This step was followed by processing for SEM and corresponding quantification of RFP lengths and compared with control (PBS-incubated) samples.
Digestion procedures.
The lung digestion techniques used in this study have previously been reported (Searl, 1997; Warheit et al., 1992, 1997a). Following incubation with p-aramid or cellulose RFP, the cells were fixed with Karnovsky's solution and then underwent a tissue digestion procedure with a 1.25% hypochlorite solution (i.e., 25% Clorox bleach) at 60°C for 9 min.
Scanning electron microscopy assessment of RFP lengths.
Filters containing p-aramid or cellulose RFP were prepared for scanning electron microscopy. Evaluations of RFP lengths were conducted on a calibrated JEOL 840 scanning electron microscope. The assessments were made at magnifications of 800x–5000x. A minimum of 100 RFP were measured and recorded for each cell culture group at each postexposure time period. For the p-aramid experiments, there were 4 groups each for each cell culture-type/each time period. Digital micrographs were made for each field of vision. Subsequently, the micrographs were printed and lengths of each individual RFP were measured and recorded.
Statistics.
For the in vitro cellular studies with p-aramid RFP, A549 human lung epithelial cells, primary human macrophages (HBAL), as well as macrophage-epithelial cell co-cultures of (Co) were treated with p-aramid RFP or cellulose RFP. The lengths of the individual RFP were measured after 1 h, 24 h, and 1 week. The hypothesis was tested that there were differences in the mean RFP lengths of each treatment group at the end of each time interval. The objective of this experiment was to compare the mean RFP lengths of each cell-type across the three time points, and within a time point, to compare the mean RFP length of the three cell culture-types (i.e., A549, HBAL, co-culture). A full factorial two-factor analysis of variance (SAS, Cary, NC) was used to make these comparisons. For several cell-type by time combinations, there were two or more groups of data available, so a nested variance structure was used. The data for this was provided in an Excel file. The criterion for statistical significance for the nonadjusted differences was set at p < 0.05. The p values for these differences were then adjusted using the Tukey-Kramer multiple comparison adjustment.
For the in vitro acellular studies, preparations of p-aramid RFP were incubated in vitro either with PBS, human BAL fluid, or rat BAL fluid and processed through a simulated digestion method. The objective here was to compare the three processing types on p-aramid RFP lengths. The hypothesis tested was that the mean RFP lengths of the groups incubated with human or rat BAL fluids were different from the mean RFP lengths of the groups incubated with PBS. A one-way analysis of variance (SAS Institute, Cary, NC) was used for this with a nested variance structure reflecting three physical experiments and cell cultures within experiment. Data for this analysis was given in an Excel file. The p values for the nonadjusted differences were set 0.05. The p values for these differences were then adjusted using the Tukey-Kramer multiple comparison adjustment.
RESULTS
Cellular differential analyses of cytocentrifuge preparations of human and rat BAL cells are presented in Figure 3. Human BAL cells consisted of 90% macrophages and 10% lymphocytes. Greater than 98% of rat BAL cells consisted of macrophages. Although rat BAL cells were not used in these experiments, it was important to determine that the rat BAL fluids were free of inflammatory cells.
Data from the in vitro human cellular studies with p-aramid RFP demonstrated that after 1 h there was no statistical evidence of differences in the mean RFP lengths among the groups. However, as demonstrated in Figures 6a and 6b, at 1-day and 1-week post-incubation, the mean p-aramid RFP lengths in the co-cultures or HBAL cells culture groups were significantly smaller than that of the A549 cell culture, but the co-culture and HBAL cell culture groups were indistinguishable from one another at each time point. Furthermore, the mean lengths at the 1-day or 1-week time points were significantly reduced compared to the 1-h time point for each cell-type. There were no significant differences between the 1-day and 1-week values for any cell-type (Fig. 4).
Data from the in vitro human cellular studies with cellulose RFP demonstrated no differences in the mean RFP lengths among the three groups at any of the three time points (i.e., 1-h, 1-day, or 1-week post-incubation) (Fig. 5).
Data from the in vitro "double dose" cellular studies with p-aramid RFP demonstrated that the exposures of p-aramid RFP to double doses of human macrophages (plated on days 1 and 3) did not accelerate p-aramid RFP shortening (Fig. 6).
Data from the in vitro acellular studies with p-aramid RFP demonstrated that the combination of incubation in human or rat BAL fluids along with simulated digestion resulted in a significant reduction in mean lengths when compared to the p-aramid RFP incubated with PBS and undergoing simulated digestion (Fig. 7).
Scanning electron micrographs of representative epithelial, and epithelial-macrophage co-cultures are presented in Figures 8–9 , which illustrate cellular-fibril interactions prior to digestion and counting.
DISCUSSION
Biopersistence of fibers is broadly defined as the extended time period of fiber retention in the respiratory tract or other tissues following deposition. A variety of inhalation studies in rats with inorganic synthetic vitreous fiber-types have demonstrated a strong, direct relationship between the biopersistence of a fiber in the lung and its potential to cause adverse effects (Hesterberg et al., 1996). Some evidence suggests that this relationship may also be operative with man-made organic fiber-types (Warheit et al., 2001b). Durability and fiber length are the two most critical determinants of lung biopersistence. The degree of biopersistence may be influenced by a variety of factors including the inhaled fiber burden (i.e., dose or number of respirable fibers present in the respiratory tract), fiber dimensions, surface characteristics, chemical composition, and other factors. Alterations in any of these parameters could influence the development of fiber-related pulmonary toxicity, as the enhanced length of contact between retained fibers and lung tissues increases the potential for adverse effects. Biodegradability refers to the breakdown and/or dissolution of fibers in the respiratory tract. Mechanical, chemical, and/or enzymatic factors may be operative in the clearance process. Fibers may be cleared from the respiratory tract via macrophage uptake and transport via the airways, transport via fluid fluxes on the mucociliary escalator or by dissolution processes (ECETOC, 1996). It generally is regarded that short fibers are more readily cleared from the lungs than longer fibers, primarily via alveolar macrophage phagocytosis and airway mucociliary clearance effects (Morgan and Holmes, 1984). Therefore, the demonstrated transverse cleavage and corresponding shortening of long p-aramid RFP in the lungs of exposed rats and hamsters (1) bodes well to facilitate a more rapid clearance of these RFP and (2) reduces the likelihood of developing fiber-related adverse pulmonary effects.
The results of in vitro mechanistic studies with human lung cells demonstrated that incubation of p-aramid RFP with primary macrophages or macrophage-epithelial cell co-cultures, but not epithelial cells alone, produced shortening of RFP at 1 day and 1 week postexposure, but not after a single hour of incubation. This is an interesting finding because the co-culture groups contained more plated cells when compared to the HBAL cultures. But this result is not surprising since it would appear that the alveolar macrophage component was likely responsible for the shortening of p-aramid RFP. In addition, utilizing a double dose of macrophages did not accelerate the rate of p-aramid RFP shortening. In contrast to the data with p-aramid RFP, incubation of cell cultures with biopersistent cellulose RFP produced no shortening in any of the cell-types at any time postexposure.
In another set of experiments, p-aramid RFP were incubated with PBS, acellular lavaged fluids from human subjects or rats, followed by a simulated tissue digestion and then assessed for changes in p-aramid length dimensions. Incubation with either human or rat BAL fluids produced substantial shortening of the p-aramid RFP compared to phosphate-buffered saline.
Taken together, these results support the hypothesis that components of human lung cells and fluids facilitate the biodegradability of p-aramid RFP.
Previously, we have reported that p-aramid RFP are biodegraded in the lungs of exposed rats and hamsters following inhalation exposures (Warheit et al., 1992, 1997a). In addition, several other investigators have demonstrated shortening of p-aramid RFP leading to more rapid clearance of the longest p-aramid fibrils during the months following exposure. It is interesting to note that this clearance pattern is associated with an initial increase in the numbers of retained shorter fibrils, i.e., an indication of cleavage of long RFP into shorter ones, and subsequent rapid RFP clearance thereafter (Bellmann et al., 2000; Kelly et al., 1993; Searl, 1997; Warheit et al., 1992, 1994).
A variety of inhalation toxicity studies have been conducted in rats with p-aramid RFP demonstrating biodegradability of RFP. In evaluating deposition and clearance patterns, Warheit et al. (1992) exposed rats to aerosols of p-aramid RFP for 2 weeks and followed RFP clearance patterns in the lung for a postexposure period of 1 year. A transient increase in the numbers of retained fibrils was noted at one week after exposure, with rapid clearance of RFP thereafter, and a retention half-time of 30 days. Following a six-month postexposure period, there was a measured progressive decrease in the mean lengths of lung-retained RFP from 12.5 (immediately after exposure) to 7.5 μm and a slight decrease in mean diameter from 0.3 to 0.2 μm. The percentages of retained RFP > 15 μm in length were reduced from 30% immediately after exposure to 5% after 6 months; the corresponding percentages of retained p-aramid RFP in the 4–7 μm length range was increased from 25 to 55% during the same period (Warheit et al., 1992).
Kelly et al. (1993) assessed the clearance kinetics in rats of lung-retained p-aramid RFP during a two-year inhalation study (Lee et al., 1988) using PCOM (phase contrast microscopy) methods. Mean lengths of inhaled p-aramid RFP in the aerosol chamber were determined to be 12 μm. Following 2 years of exposure at concentrations of 2.5, 25, or 100 f/ml, or 1 year of exposure at 400 f/ml plus 1 year recovery, mean lengths were determined to be 4 μm in RFP recovered from the lungs of rats. In addition, the time required for retained RFP to be reduced to <5 μm in the lung was greatest at the higher concentration and shortest at the lowest aerosol concentration.
Searl (1997) exposed rats to aerosols of three different fiber-types, namely respirable (1) p-aramid RFP, (2) chrysotile asbestos fibers, and (3) Code 100/475 glass fibers, and assessed the relative biopersistence/retention characteristics after 10-day inhalation exposures at concentrations of 700 f/ml. The most significant reduction in lung fiber burden was measured during the first 3 months after exposure, but the pattern of clearance of different size classes varied with each fiber-type. For example, inhalation of p-aramid RFP resulted in rapid clearance of the longest RFP during the first month postexposure, and this was associated with an initial enhancement of the numbers of shorter RFP. Searl concluded that this clearance pattern was consistent with cleavage of p-aramid RFP into successively shorter fragments which subsequently facilitated rapid clearance by alveolar macrophages. Similar clearance patterns were measured with the inhaled glass fibers. In contrast, exposures to chrysotile asbestos fibers initially induced a rapid clearance in the numbers of inhaled short fibers relative to long fibers, and this is consistent with preferential clearance of short fibers by macrophages. However, the longer chrysotile fibers were retained for prolonged or indefinite periods in the lung. Accordingly, the author concluded that biopersistence of long (>15 μm) chrysotile asbestos fibers was substantially greater than that of long RFP of p-aramid or glass fibers, indicating that the subpopulation of long, chrysotile respirable fibers was selectively retained in the lung without any apparent clearance.
Bellmann et al. (2000) also evaluated clearance kinetics of p-aramid RFP in the lungs of exposed rats. Male Wistar rats inhaled aerosol concentrations of 50, 200, or 800 p-aramid RFP/ml for 3 months. The average lengths of retained p-aramid RFP were progressively shortened during the postexposure period of 3 days to 3 months. Moreover, the numbers of RFP < 10 μm in length actually were increased at 1 month postexposure when compared to the 3-day recovery, whereas for the RFP fraction with lengths above 10 μm the number of RFP per lung decreased. This pulmonary clearance effect is likely explained by cleavage of retained, longer RFP (length > 10 μm) in the lungs of exposed animals. The investigators concluded that the clearance kinetics of RFP > 20 μm in length was associated with the fastest elimination. Therefore, breakage of long RFP best explained the experimental observations. These p-aramid RFP clearance patterns are consistent with the findings reported by Warheit et al. (1992, 1994), Kelly et al. (1993), and Searl (1997).
More recently we have carried out both in vivo and in vitro cellular and acellular investigations with p-aramid and cellulose RFP in rats. First, p-aramid or cellulose RFP were instilled into the lungs of rats and the lungs digested 24 h postexposure using two different digestion techniques: (1) a conventional ethanolic KOH method, and (2) an enzymatic method which simulates the action of lung enzymes. Cellulose RFP were utilized as a control organic fiber-type which is known to be biopersistent. The enzyme method demonstrated shortening of the p-aramid RFP, but the KOH method did not. In addition, neither method was successful in shortening the cellulose RFP. In addition, standardized preparations of p-aramid RFP or cellulose RFP were incubated with saline or lung fluids and processed by one of 2 tissue digestion techniques. Mean lengths of p-aramid RFP incubated with saline and processed with KOH or the enzyme method were not altered. The preparation of p-aramid RFP that had been incubated with BAL fluids and processed with the enzyme solution resulted in cleavage of p-aramid RFP. In contrast to the in vitro acellular studies with p-aramid RFP, the combination of BAL fluid incubation and enzyme digestion method had no measurable effect on shortening of cellulose RFP, indicating that the results with p-aramid RFP were specific. In a final set of in vitro cellular studies, cultures of rat lung epithelial cells, alveolar macrophages, or co-cultures of epithelial cells and macrophages were treated with p-aramid RFP for 1 h, 1 day, or 1 week to determine whether RFP shortening occurs directly in the phagocytic cells. The results demonstrated that (1) no significant degree of shortening occurred in the epithelial cell cultures at any time point; however, in the macrophage and co-cultures, cleavage of p-aramid RFP was observed at 1 day and 1 week postexposure. We concluded from this study that components of lung fluids coat the p-aramid RFP as a prerequisite for enzymatic cleavage. (Warheit et al., 2001a).
The data from the in vitro "double dose" cellular studies with p-aramid RFP demonstrated that the exposures of p-aramid RFP to double doses of human macrophages (plated on test days 1 and 3) did not accelerate p-aramid RFP shortening. This preliminary finding, although based on only two experiments, was a minor surprise. An alternative interpretation is that most of the RFP shortening occurred after the first 24 hours of incubation, with very little additional cleavage occurring at the 1 week postexposure time period. Thus, perhaps it is not surprising that additional shortening did not occur at 3 days postexposure. Unfortunately, sufficient numbers of human macrophages were not available at the 1 day postexposure time period to test whether a true double dose of macrophages would have accelerated fiber shortening.
To summarize our findings, the studies reported herein were developed to test a hypothesis of p-aramid biodegradability using human lung cells in vitro, and to determine whether the same mechanisms, previously demonstrated in rat lung cells might be operative in human lung cells. Absent in vivo human data with p-aramid RFP, it was postulated that p-aramid biodegradability in human lung cells in vitro would provide further validation for the likelihood of in vivo fiber biodegradability in humans, and lend additional information on the safety of aerosol exposures to p-aramid RFP. As indicated in Figure 13, using a Hazard Assessment Bridging Paradigm—the demonstration of p-aramid biodegradability in the lungs of rats (in vivo) and the corresponding mechanism in rat lung cells (in vitro) represents a defined relationship. Confirmation of these measured effects of p-aramid RFP shortening in human cultured lung cells supports the conclusions previously measured in lungs cells and lungs of exposed rats. These measured effects are selective for p-aramid RFP and were not measured in similar studies with cellulose RFP, a biopersistent fiber-type. As a consequence, these data provide strength, via an extrapolated relationship, to the notion that inhaled p-aramid RFP appear to be biodegradable in the lungs of humans (Fig. 10).
The findings presented here reinforce the logic derived from animal lung exposure studies which suggest that inhaled p-aramid RFP biodegrade in the lungs of humans. As a contribution towards improved hazard and risk assessments, the methodology introduced in these studies has (1) demonstrated the practical viability of in vitro exposure testing with human lung cells, and (2) "bridged" in vitro animal and human cell exposure studies through systematic design and execution of experiments, in comparison to in vivo animal studies under equivalent conditions.
NOTES
1 The authors ceritify that all research involving human subjects was done under full compliance with all government policies and the Helsinki Declaration.
ACKNOWLEDGMENTS
This study was supported by the DuPont Advanced Fiber Systems Strategic Business Unit and by the Teijin Twaron Company. The authors wish to thank Drs. Odorich von Susani, Chris Braun, Jan Roos, Fred Pfister, Gerald Kennedy, and Ted Merriman for helpful insights into the preparation of the manuscript. We also thank Dr. John W. Green for providing statistical analyses of the data, and Dr. Liang Liang for technical assistance with the scanning electron microscope.
REFERENCES
Bellmann, B., Creutzenberg, O., Dasenbrock, C., Ernst, H., Pohlmann, G., and Muhle, H. (2000). Inhalation tolerance study of p-aramid respirable fiber-shaped particulates (RFP) in rats. Toxicol. Sci. 54, 237–250.
ECETOC. (1996). Toxicology of Man-Made Organic Fibres. Technical Report No. 69. European Centre for Ecotoxicology and Toxicology of Chemicals, Brussels, Belgium. p. 41.
Ghio, A. J., Bassett, M., Chall, A. N., Levin, D. G., and Bromberg, P. A. (1998). Bronchoscopy in healthy volunteers. J. Bronchology 5, 185–194.
Hesterberg, T. W., Miiller, W. C., Musselman, R. P., Kamstrup, O., Hamilton, R. D., and Thevenaz, P. (1996). Biopersistence of man-made vitreous fibers and crocidolite in rat lung following inhalation. Fundam. Appl. Toxicol. 29, 267–279.
Kelly, D. P., Merriman, E. A., Kennedy, G. L. Jr., and Lee, K. P. (1993). Deposition, clearance, and shortening of Kevlar para-aramid fibrils in acute, subchronic, and chronic inhalation studies in rats. Fundam. Appl. Toxicol. 21, 345–354.
Lee, K. P., Kelly, D. P., O'Neal, F. O., Stadler, J. C., and Kennedy, G. L., Jr. (1988). Lung response to ultrafine Kevlar aramid synthetic fibrils following 2-year inhalation exposure in rats. Fundam. Appl. Toxicol. 11, 1–20.
Morgan, A., and Holmes, A. (1984). The deposition of MMMF in the respiratory tract of the rat, their subsequent clearance, solubility in vivo and protein coating. In Biological Effects of Man-Made Fibres, Vol. 2 (T. Guthe, Ed.), pp. 1–17. WHO, Copenhagen, Denmark.
Searl, A. (1997). A comparative study of the clearance of respirable para-aramid, chrysotile and glass fibres from rat lungs. Ann. Occup. Hyg. 41, 217–233.
Searl, A., and Cullen, R. T. (1997). An enzymatic tissue digestion method for fibre biopersistence studies. Ann. Occup. Hyg. 41, 721–727.
Warheit, D. B., Hill, L. H., and Brody, A. R. (1984). Surface morphology and correlated phagocytic capacity of pulmonary macrophages lavaged from the lungs of rats. Exp. Lung Res. 6, 71–82.
Warheit, D. B., Carakostas, M. C., Hartsky, M. A., and Hansen, J. F. (1991). Development of a short-term inhalation bioassay to assess pulmonary toxicity of inhaled particles: Comparisons of pulmonary responses to carbonyl iron and silica. Toxicol. Appl. Pharmacol. 107, 350–368.
Warheit, D. B., Kellar, K. A., and Hartsky, M. A. (1992). Pulmonary cellular effects in rats following aerosol exposures to ultrafine Kevlar aramid fibrils: Evidence for biodegradability of inhaled fibrils. Toxicol. Appl. Pharmacol. 116, 225–239.
Warheit, D. B., Hartsky, M. A., McHugh, T. A., and Kellar, K. A. (1994). Biopersistence of inhaled organic and inorganic fibers in the lungs of rats. Environ. Health Perspect. 102(Suppl. 5), 151–157.
Warheit, D. B., Snajdr, S. I., Hartsky, M. A., and Frame, S. R. (1997a). Pulmonary responses to inhaled para-aramid fibrils in hamsters: Evidence of biodegradability in the lungs of a second rodent species. Inhal. Toxicol. 9, 181–187.
Warheit, D. B., Snajdr, S. I., Hartsky, M. A., and Yuen, I. S. (1997b). Assessments of in vitro effects of dusts on pulmonary cells using macrophage-epithelial cells in co-culture. In Correlations between in Vitro and in Vivo Investigations in Inhalation Toxicology (U. Mohr, D. L. Dungworth, K. B. Adler, C. C. Harris, and C. G. Plopper, Eds.), pp. 432–446. ILSI Press, Washington, DC.
Warheit, D. B., Snajdr, S. I., Hartsky, M. A., and Frame, S. R. (1998). Two-week inhalation study in rats with cellulose fibers. In Advances in the Prevention of Occupational Respiratory Disease (K. Chiyotani, Y. Hosoda, and Y. Aizawa, Eds.), pp. 579–582. Elsevier Science BV, Amsterdam.
Warheit, D. B., Hartsky, M. A., Reed, K. L., and Webb, T. R. (2001a). Biodegradability of inhaled para-aramid RFP: Mechanistic in vivo and in vitro studies. Toxicol. Appl. Pharmacol. 174, 78–88.
Warheit, D. B., Hart, G. A., Hesterberg, T. W., Collins, J. J., Dyer, W. M., Swaen, G. M. H., Castranova, V., Soiefer, A. I., and Kennedy, Jr., G. L. (2001b). Potential pulmonary effects of man-made organic fiber (MMOF) dusts. Crit. Rev. Toxicol. 31, 697–736.(D. B. Warheit, K. L. Reed, J. D. Stonehu)