Effects of Chronic Exposure to Crack Cocaine on the Respiratory Tract of Mice

Abstract

Smoked cocaine (crack cocaine) causes several forms of injury to the respiratory tract, including asthma exacerbations, lung edema and hemorrhage, and nasal mucosal alterations. Few studies, however, have assessed respiratory tract pathology in habitual users of crack cocaine. Here, we describe the histological alterations in the respiratory tract of mice caused by chronic inhalation of crack cocaine. Twenty 2-month-old BALB/c mice were exposed to the smoke of 5 g crack cocaine in an inhalation chamber once a day for two months and compared to controls (n = 10). We then morphometrically analyzed nose and bronchiolar epithelial alterations, bronchiolar and alveolar macrophage cell density, alveolar hemosiderin content, and in addition determined the vasoconstriction index and the wall thickness of pulmonary arteries. The serum cocaine level was 212.5 ng/mL after a single inhalation. The mucus content of the nasal epithelium increased in crack-exposed animals, and the nasal and bronchial epithelium thickness decreased significantly. The alveolar hemosiderin content and the alveolar and bronchiolar macrophage cell density increased in animals exposed to crack. The vasoconstriction index increased in the pulmonary arteries of the exposed group. Chronic crack cocaine inhalation causes extensive histological changes along the entire respiratory tract.

Introduction

The development of crack cocaine in the mid-1980s caused a dramatic increase in cocaine consumption during the following decade in the United States (Haim et al. 1995), with a 3.9% annual prevalence among young individuals in 1987 (Baldwin et al. 2002). The prevalence then declined, but it has gradually risen again, reaching 2.7% in 1999 (Baldwin et al. 2002). According to the U.S. Department of Health and Human Services, in 2003, 34.9 million Americans twelve years of age and older reported having used cocaine at some point during their lifetime, with 7.9 million having used crack (Restrepo et al. 2007). In Europe, cocaine consumption has been increasing since the 1980s, and crack consumption is more frequent among the socially marginalized (Prinzleve et al. 2004).

In Brazil, cocaine consumption is a major health and social problem, and consumption among students has increased in the past decade (Guindalini et al. 2006). In a survey of hospitalized drug users in São Paulo, the largest Brazilian city, 38.4% of the patients were crack cocaine smokers, and 31.8% used both inhaled and smoked cocaine (Ferreira Filho et al. 2003). Crack cocaine consumption is associated with lower education and higher unemployment, urban violence, and mortality rates (Meleca et al. 1997).

Crack cocaine can be prepared by heating a solution of cocaine hydrochloride with baking soda until a rock is formed; this rock is then smoked in special paraphernalia (Meleca et al. 1997). Smoked cocaine causes injuries to the respiratory tract. The heated cocaine vapors, the products of pyrolysis (e.g., methylecgonidine), impurities and contaminants with which cocaine may be cut, and the combustion products of the fuel used to ignite the cocaine have the potential to harm the entire respiratory tract (Wood et al. 1996).

The frequency of pulmonary complications caused by crack is not known, but a wide spectrum of clinical and functional alterations have been described in comprehensive reviews (Albertson et al. 1995; Haim et al. 1995; Restrepo et al. 2007). Acute alterations after crack use include episodes of bronchoconstriction, pneumomediastinum and pneumothorax, noncardiogenic pulmonary edema, hemoptysis, chest pain, pulmonary hemorrhage, and diffuse pulmonary infiltrates (Kleerup et al. 2002; Thadani 1996). Less is known, however, about the effects of the regular, chronic abuse of crack on the respiratory tract. Signs of chronic vascular damage, occult bleeding, and pulmonary hypertension have been described (Albertson et al. 1995; Gallouj et al. 1999; Murray et al. 1989). In the nose, irregularities in the nasal mucosa and oropharynx have been described in crack abusers (Meleca et al. 1997; Nassif Filho et al. 1999; Tierney and Stadelmann 1999).

Most histological studies of this subject have been performed using autopsy tissue from patients who died of cocaine overdoses (Bailey et al. 1994; Murray et al. 1989). The difficulties of understanding the chronic effects of crack abuse on the respiratory tract based on such studies derive from several factors: small patient populations, frequent use of concomitant drugs, and difficulties in controlling the composition, as well as the pattern of smoking and devices used (Fligiel et al. 1997). It has been suggested that animal models may represent an alternative way to better understand lung pathology and mechanisms of chronic lung damage in crack smokers (Barroso-Moguel et al. 1999).

Therefore, in this study we used a model of chronic crack inhalation in mice to study histological alterations on the respiratory tract. We hypothesized that the whole respiratory tract would be adversely affected, with nose and lung vascular bed alterations as well as increased lung inflammation.

Methods and Materials

This study was approved by the institutional research Ethical Board.

Animals

Forty 60-day-old male BALB/c mice (ten controls and thirty exposed) were used in this study. During the study period, animals were kept under laboratory conditions according to the Guide for the Care and Use of Laboratory Animal Resources (National Academy of Sciences, Washington DC, 1996). Animals were fed ad libitum with commercial pelleted food for small rodents from Nuvital (Nuvilab CR-1, Colombo, PR, Brazil). This chow is enriched with minerals and vitamins, and it is produced following recommendations of the National Research Council and National Institutes of Health, USA.

Test Substance

The crack cocaine used for this study was of unknown origin, and was ‘‘in natura’’ (rocks) cocaine obtained from the narcotics division of the Marília city (São Paulo, Brazil) police department. The crack was originally from a sealed portion (n. 0047430, process 1096/2002, Criminalistic Institute, Marília, São Paulo, Brazil) tested for its composition and contained 57.7% of cocaine. No other active substance (adulterants, local anaesthetics) was found via gas chromatography-mass spectrometry (GC-MS).

Crack Smoke Exposure

Ten mice were used to determine serum levels of cocaine after a single crack exposure under the same experimental conditions as the study group. Blood was collected within a few minutes after drug inhalation and euthanasia.

Twenty mice were simultaneously exposed to crack smoke in a specially built whole-body chamber five days per week (excluding weekends) for sixty days.

Each exposure took five minutes, enough time to burn 5 g of crack rocks. Crack rocks were placed in a metallic crucible heated from below by a flame. The resulting crack smoke was suctioned into a hood connected to the exposure chamber. Negative pressure required to suction the smoke was provided by a vacuum pump (20 L/min) connected to the chamber (Figure 1). The mean chamber temperature during the crack cocaine exposure was 24°C ± 2°C. The exposure took place safely under outdoor laminar flow exhaustion. After the exposures, animals were kept in animal housing. Control animals (n = 10) were kept in animal housing during the whole study period.

Blood Cocaine Determination

Cocaine serum levels were determined by the Department of Toxicology of the Faculty College of Pharmaceutical Sciences of São Paulo University using GC-MS. The concentration of cocaine in whole blood was determined by liquid–liquid extraction using an n-hexane/dichloromethane mixture (4:1). After shaking and separation, the organic phase was dried under a gentle N₂ stream at 40°C. The residue was resuspended in an n-hexane/ethanol (9:1) mixture, and 1 μL was injected into GC-MS. Benzoylecgonine isopropyl ester was used as the internal standard (Björk et al. 1990). The lowest limit of quantitation (LLOQ) of the method was 10 ng/mL. Applying this method, only cocaine was detected.

Tissue Processing

After the exposure period, mice were anesthetized with 50 mg/kg sodium pentobarbital and euthanized by exsanguination. The head and the lungs were quickly isolated and excised.

The nasal cavity was flushed retrograde through the naso-pharyngeal orifice with 5 mL of 10% neutral buffered formalin. Excess soft tissue and the lower jaw were removed, and the head was immersed in formalin for one day and then decalcified in 5% EDTA for fourteen days. Sections were taken from three different levels of the nasal cavity: proximal level, taken immediately posterior to the upper incisor teeth; medial level, taken through the level of the incisive papilla anterior to the palatal ridge; and distal level, through the middle of the second molar teeth. These levels permit examination of the major epithelial types in the nasal cavity and accurate interpretation of changes in the epithelium after a toxic insult (Herbert and Leininger 1999).

The lungs were weighed, and the trachea was cannulated. Both lungs were inflated under constant positive pressure at 20 cm water pressure with 10% buffered formalin for twenty-four hours, allowing for homogenous expansion of lung parenchyma (Halbower et al. 1994; Kasahara et al. 2000). After that, the lungs were paraffin-embedded, and 5 μm sections from both lungs were cut for morphometric and immunohisto-chemical analyses.

Immunohistochemistry

Sections were deparaffinized and hydrated. After blocking the endogenous peroxidase, antigen retrieval was performed with high-temperature citrate buffer (pH = 6.0). The anti-mouse macrophage marker Mac-2 (1:25,000, clone M3/38; Cedarlane, ON, Canada) was used in the study. The Vectastain ABC Kit (Vector Laboratories, Burlingame, CA, USA) was used as secondary antibody. 3,3-diaminobenzidine (DAB) (Sigma, St. Louis, MO, USA) was used as chromogen. The sections were counterstained with Harris hematoxylin. For negative controls, the first antibody was omitted from the procedure, and BSA (bovine serum albumin) was used instead.

Morphometry

Morphometric analysis was performed by a single investigator. The specimens were coded, and the measurements were performed without knowledge of the study group. An eyepiece square grid with a known area (62,500 μm² at a magnification of 400X) containing 400 points was used for the measurements.

Nose and lung sections were stained with a combination of Schiff’s periodic acid and Alcian blue (PAS/AB) at a pH of 2.5. With the latter technique, neutral and acidic glycoproteins are stained in red and blue, respectively (Jones and Reid 1978). The volumetric proportion of neutral and acidic mucus present was determined in nasal and bronchial epithelium and calculated as follows (Weibel 1990; Weibel and Cruz-Orive 1997):

Volume proportion of mucosubstance = Pm/Pt

Pm is the number of points hitting mucosubstance acidic or neutral, and Pt is the total number of points hitting the epithelium for a given basement membrane length. Volume proportion was expressed as a percentage. Epithelial thickness was expressed by epithelial area (number of points) divided by basement membrane length. Basement membrane length was assessed using the outer border of the grid, which had a constant length of 250 μm at 400X magnification. The outer line of the grid was laid parallel to the basement membrane of the measured structures.

To assess the pulmonary arterioles, slides were stained with H&E. We selected small arteries that were adjacent to the bronchoalveolar junction and in cross-section with a variation of < 10% between its maximum and minimum diameter (Batalha et al. 2002). We chose this location to guarantee that all vessels studied were in the same size range. Additionally, there is evidence that this is the segment of the respiratory tract where particles deposit more efficiently (Saldiva et al. 2002).

In general, each animal contained eight to ten arteries meeting these criteria. The artery was placed in the center of the grid, and the points overlying the arteriole lumen and the points overlying the muscular wall were counted separately. Additionally, the intercepts crossing the external limit of the muscular wall (elastic externa) were quantified. From these data, we calculated the artery wall thickening index and the vasoconstriction index, adapted from Vieira et al. 2007.

For determination of hemosiderin content in the lungs, we used Perls-stained lung slides (Putt 1972). We assessed the volumetric proportion of Perls-positive structures in twenty 400X fields of lung parenchyma and expressed the results as percentages.

We calculated alveolar macrophage density by counting the number of points contacting alveolar/peribronchial tissue in each field. In the alveolar parenchyma, twenty 400X fields were analyzed per animal. The alveolar tissue area in each field was calculated according to the number of points touching alveolar tissue as a proportion of the total grid area. We then counted the number of positive cells within that alveolar tissue area. Similarly, peribronchial macrophage cell density was analyzed in five bronchioles per animal (twenty 400X fields analyzed per animal). We determined the density of the immunostained cells as the number of positive cells in each field divided by tissue area, and the results were expressed as cells/mm² (De Magalhães Simões et al. 2005).

Statistical Analysis

Results were expressed as the mean ± SD, unless otherwise specified. Comparison between groups was performed using Student’s t tests or Mann-Whitney tests, depending on the data distribution. The statistical program SSPS v13.0 computer package for Windows (SPSS, Inc., Chicago, IL, USA) was used for the analysis. Differences were considered significant when p < .05.

Results

Animals

No deaths occurred during the exposure in either group. There were no macroscopic abnormalities in the organs of the crack-exposed animals. Body weight and water and food intake were not affected by crack exposure.

Serum Cocaine Determination

The serum cocaine level in the blood of ten animals (combined) was 212.5 ng cocaine/mL.

Nasal Cavity

The respiratory epithelium of the nasal cavity in crack-exposed animals showed an increase in the volume proportion of both acidic and neutral mucus, with a decrease in the epithelial thickness (Figures 2A, 2B, 3A, 3B) (p < .001 t test). There was no necrosis. We also analyzed separately the same parameters in the three anatomically defined mice nose regions (proximal, medial, and distal) and found similar results to those described above (data not shown).

No histological alterations such as necrosis or acute inflammation could be detected in the olfactory or in the squamous epithelium.

Lungs

The lung weight in controls was 0.28 ± 0.02 g, and in exposed animals 0.27 ± 0.02g, with no statistical difference between groups (p = .44 t test). In the exposed group, descriptive lung histological analysis showed increased alveolar cellularity owing to an increased number of macrophages. These presented a pale brownish cytoplasm, with black granules occasionally visible. There was a mild, variable, non-eosinophilic peribronchiolar and perivascular inflammatory cell infiltration. There was no acute lung edema, signs of recent hemorrhage or interstitial fibrosis, or signs of pneumonia.

In the airways, there was a significant decrease in the bronchial epithelial thickness (p < .001, t test) without changes in mucus volume proportion (crack = 1.34 ± 2.04, controls = 2.05 ± 2.66, p = .423, Mann-Whitney test) (Figures 2C, 2D, and 4). Macrophage peribronchiolar density was increased in cocaine crack-exposed animals (p = .001, Mann-Whitney test) (Figure 5).

The vasoconstriction index was increased in crack-exposed animals when compared to controls (p = .034, t test) (Figures 2E, 2F, and 6). There was no thickening of the vessel walls (crack = 2.30 ± 0.50, controls = 2.22 ± 0.84, p = .782, t test).

There was an increased macrophage cell density in the alveolar parenchyma of crack-exposed animals (p = .001, t test) (Figures 7A, 7B, and 8). The volume proportion of Perls-positive structures in the lung parenchyma was significantly higher in the exposed group when compared to the control group (p < .001, t test) (Figures 7C, 7D, and 9).

Discussion

In this study we describe histological alterations in the nose and lungs of mice chronically exposed to crack cocaine. Our data show that the entire respiratory tract is affected by crack cocaine inhalation, from the nose to the alveolar compartments of the lungs. Crack cocaine inhalation induced nose and bronchial epithelial atrophy, chronic alveolar hemorrhage, parenchymal and bronchial macrophage inflammation, and vasoconstriction. To the best of our knowledge, this is the first experimental study of the effects of chronic smoked crack cocaine exposure on the entire respiratory tract.

Cocaine acts as a powerful sympathomimetic agent, blocking the presynaptic uptake of norepinephrine and dopamine, producing high levels of these modulators near the postsynaptic receptors (Egred and Davis 2005). Cocaine is a potent vasoconstrictor in some organs, such as the heart and the nasal mucosa. In the lungs, the pulmonary effects of cocaine are less well understood. The presence of innervated α- and β-adrenergic receptors has been described in the pulmonary vascular smooth muscle, and administration of adrenergic agonists such as cocaine increased pulmonary vascular tone (Laposata and Mayo 1993). Alternatively, cocaine may alter the central nervous system’s neuroregulation of the lungs (Laposata and Mayo 1993).

There is strong clinical evidence that cocaine impairs pulmonary microcirculation (Haim et al. 1995; Baldwin et al. 2002). Chronic ‘‘healthy’’ crack smokers present with increased numbers of hemosiderin-laden macrophages (as an indicator of alveolar hemorrhage) and increased levels of endothelin-1, a potent vasoconstrictor, in bronchoalveolar lavage (Baldwin et al. 2002; Janjua et al. 2001). Furthermore, in an autopsy study where the lungs of twenty patients who died of cocaine intoxication were studied, pulmonary artery hypertrophy was observed in 20% of studied cases, and signs of occult hemorrhage were found in 35% (Murray et al. 1989). On the other hand, clinical studies show conflicting results in relation to changes in the diffusion capacity (as a sign of direct damage to the alveolar–capillary barrier) and no increase in the pulmonary artery pressure after acute cocaine exposure (Baldwin et al. 2002). Our results suggest that crack inhalation causes microvascular alterations, as indicated by increased content of hemosiderin in the lungs and increased vasoconstriction index. We were not able to demonstrate arterial thickening in this model. It is possible that the exposure time was insufficient to induce vascular structural changes (Murray et al. 1989). In the heart, cocaine smoking exacerbates arterial vasoconstriction in the coronary arteries (Kuhn et al. 1990). Our data suggest that the same effect may occur in the lungs, an observation with important clinical implications.

Barroso-Moguel and colleagues (1999) studied the chronic effects of systemic administration of cocaine (30 mg/kg per day, i.p.) on the alveolar parenchyma. They examined pulmonary histology at multiple time points to seventy-five days of exposure. In this descriptive histological study, hemorrhagic lesions were observed at earlier time points but were gradually replaced by alveolar fibrosis. In our model, no significant lung fibrosis was observed, but it is possible that fibrosis appears only after sixty days of exposure. It is also possible that the different doses and routes of exposure contributed to the observed differences.

We observed an increase in the number of macrophages in the alveolar spaces of exposed animals, suggesting that crack cocaine exposure also leads to inflammatory changes. It is possible that bursts of inflammatory activity after cocaine use contribute to the various forms of lung injury. On the other hand, despite a known increase of macrophages in the alveolar parenchyma, some studies have shown that alveolar macrophages in cocaine-exposed mice have decreased effector functions, such as impaired nitric oxide production with decreased bactericidal activity (Roth et al. 2004). It has been suggested that the impaired macrophage effector activity could play a role in the increased susceptibility to infections seen in crack cocaine smokers (Baldwin et al. 1997; Roth et al. 2004).

Nasal cocaine inhalation can cause rhinitis, decreased olfaction, epistaxis, and nasal septum perforation secondary to its potent vasoconstrictor and anesthetic effects (Mari et al. 2002). Thermal injury and other irritants present in the drug may further injure the nasal mucosa of crack smokers (Nassif Filho et al. 1999). Our results show that crack inhalation harms the nasal mucosa of mice, causing epithelial atrophy with increased mucus content within the epithelial layer. Nasal irritants may preferentially alter specific nasal cavity regions (Camargo Pires-Neto et al. 2006), but in this study we show that chronic crack cocaine smoke causes nasal epithelial alterations along the entire nasal respiratory mucosa. Since in our model thermal burns were avoided, our observations result most likely from cocaine and other irritants present in the smoked crack rock. Our results help explain the high frequency of nasal symptoms in crack abusers.

Crack smokers can present with exacerbated asthma, and Tashkin et al. (1992) demonstrated that smoked, but not intravenous, cocaine induces bronchoconstriction. Its pyrolysis product, methylecgonidine, can also induce acute bronchoconstriction in guinea pigs (Chen et al. 1995). Our data show signs of airway epithelial atrophy with a macrophagic bronchiolar inflammation but no increase in mucus content in the exposed animals. Whereas we observed an increase in nasal mucus content, we could not detect similar alterations in the bronchial epithelium mucus content. There is a single study in which bronchial biopsies taken from crack smokers showed marked epithelial alterations without goblet cell metaplasia (Fligiel et al. 1997). Atrophy of the lining epithelium in nose and bronchi is probably secondary to cocaine-induced vasoconstriction.

This study has some important limitations. We were not able to determine the content of the crack rock used in the study beyond its cocaine content. Although GC-MS did not reveal any other active substance, we cannot exclude the possibility that part of the abnormalities result from contaminants present in the sample. However, our results are compatible with systemic cocaine administration data previously described in humans and in animal models, suggesting that at least part of the observed effects are from cocaine inhalation. The amount of crack cocaine inhaled by the mice in each exposure leads to a serum concentration of cocaine that, if transposed to humans, would be comparable to a nonlethal (safe for the cardiovascular system) dose of 0.69 to 1.37 mg/kg insufflated cocaine (four ‘‘lines of powder’’), as described by Collins et al. (2007). However, we believe that our data are relevant because they show that even in nonlethal doses and over a limited period of time (sixty days), animals develop significant histological alterations such as lung hemorrhage, nose and bronchial epithelium atrophy, and lung macrophagic inflammation.

In summary, chronic exposure to crack cocaine, even for a limited period of time and in nonlethal doses, leads to extensive histological alterations along the entire respiratory tract. Crack cocaine use is a major health and social problem in several countries. Increasing the knowledge of its harmful effects in the lungs is important for broadening our understanding of the serious consequences of crack abuse.