Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Budesonide is an antiinflammatory glucocorticoid used against asthma as a dry powder inhalation formulation. Its clinical properties (1, 2) and its pharmacokinetics and metabolism (3, 4) have been extensively investigated. Recently, in a study using a rat trachea superfusion model (5), a new route of metabolism for budesonide was discovered. It was noticed that a small fraction of a budesonide dose was retained by the tracheal tissue and only slowly released on extended superfusion with blank buffer. Surprisingly, liquid chromatography (LC)¹ of ethanol extracts from the tracheas showed that unchanged budesonide comprised only a minor part of the radioactivity retained in the tissue. Instead, some very lipophilic metabolite(s) dominated the chromatograms. Subsequently, formation of metabolites with similar LC-migration properties was observed when budesonide was added to a number of biological systems, including human lung and liver microsomes.

The present study was performed to study the formation of lipophilic metabolites of budesonide in human liver and lung microsomes, to define partly the biochemistry of the process, and to identify the metabolite(s).

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. Budesonide ((22RS)-16,17-butylidenedioxy-11,21-dihydroxypregna-1,4-diene-3,20-dione) (fig. 1A) was obtained from Astra Pharmaceutical Productions, Södertälje, Sweden. [³H]Budesonide (fig. 1B) (radiochemical purity 98.7%, chemical purity 99.0%, specific activity 0.62 TBq/mmol, stock solution 40.2 MBq/ml ethanol), [²H₈]budesonide (fig. 1C), budesonide 21-oleate, budesonide 21-palmitoleate, and budesonide 21-palmitate were synthesized at the Department of Medicinal Chemistry, Astra Draco AB, Lund, Sweden.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Structural formulas of budesonide (A, with some of the numbering indicated), 3H]-budesonide (B), and 2H8-budesonide (C).

Microsomes. Human liver microsomes were purchased from Human Biologics Inc. Human lung parenchymal tissue was obtained from the Department of Pathology, University Hospital, Lund, Sweden. Lung microsomes were prepared as follows: Pieces of lung tissue were pulverized in liquid nitrogen in a mortar and then further homogenized in buffer in a Potter Elvehjelm homogenizer with a Teflon (Dupont, Wilmington, DE) pestle. The buffer used was 50 mM K₂HPO₄, pH 7.5, with 0.25 M sucrose (4 ml/g pulverized tissue). After a centrifugation step for 15 min at 9000g, the supernatant was centrifuged in an Ultracentrifuge (Beckman model L7, rotor type 50.2 Ti) at 100,000g for 80 min. The resulting microsome pellet was resuspended in 5 ml of 50 mM K₂HPO₄, pH 7.5, with 0.15 M KCl, homogenized in a Potter Elvehjelm homogenizer equipped with a glass pestle, and centrifuged at 100,000g once more. This last step was repeated once. The twice-washed microsome pellet was resuspended in the sucrose containing buffer (1 ml/g tissue) and kept in 0.1-0.5-ml portions at 70°C. All steps in the preparation were done with ice-chilled equipment and buffers, and the centrifugations were performed at 4°C.

Incubations and Sample Work-Up. Formation of conjugates. Microsomes from human liver (20 mg protein/ml) and human lung (3.6 mg protein/ml) were diluted with buffer (0.1 M K₂HPO₄, pH 7.5, with 5 mM MgCl₂) to a protein concentration of 1 mg/ml. ATP and CoA were added to final concentrations of 5 mM and 1 mM, respectively. In experiments where the influence of externally added fatty acids was to be studied, these were added as ethanol solutions at this point (100 µM final concentration). After equilibration for 5 min with gentle shaking in a water bath at 37°C, budesonide plus [³H]budesonide was added as an ethanol solution (time zero). The initial concentration of budesonide was 100 µM, containing 1 µM [³H]budesonide. The amount of ethanol present during the incubations varied between 2 and 4.5%. Samples (24 µl) were withdrawn at 5, 10, 15, 30, and 60 min and immediately added to 276 µl acetonitrile/methanol (50/50). After centrifugation at 15,000g for 4 min, 50 µl of the supernatant was injected into LC system 1.

Preparation of samples for liquid chromatography mass spectrometry (LC-MS). The incubation conditions for the formation of conjugates for identification by LC-MS were the same as described above, only differing in the type of budesonide added. Half of the budesonide and [³H]budesonide was replaced by [²H₈]budesonide, keeping the same initial concentration (100 µM) of total glucocorticoid. No external fatty acids were added to these incubations. After a 60-min incubation at 37°C, the reaction was stopped by adding ethanol to a final concentration of 70%. The sample was kept on ice for a few minutes before being centrifuged (4 min at 15,000g). The lipophilic metabolites (4-8% of the total budesonide content) and the remaining budesonide were separated on a BondElut solid phase extraction column (C18, 1 ml, Varian). The BondElut column was first activated with 2 ml of ethanol and then equilibrated with 2 ml of 70% ethanol before the sample was applied. A washing step with 2 ml of 70% ethanol followed. The fatty acid conjugates were then eluted with 1 ml of ethanol. The ethanol eluate was evaporated under a stream of N₂ at room temperature. The residue was dissolved in 50 µl of 90% methanol and centrifuged prior to analysis by LC-MS (10-15 µl injected).

Hydrolysis experiments. The porcine pancreas lipase and the bovine pancreas cholesterol esterase were dissolved in 0.1 M K₂HPO₄ buffer, pH 7.5, at a concentration of 0.25 mg/ml and 0.0125 mg/ml, respectively. In the case of cholesterol esterase, the incubation buffer also contained 10 mM taurocholate. The incubation mixtures were temperated for 5 min with gentle shaking in a water bath at 37°C. The incubation was started by adding the substrate (budesonide oleate or budesonide palmitoleate dissolved in ethanol) at an initial concentration of 5 µM. The concentration of ethanol in the incubations was 0.5%. Aliquots of 150 µl were withdrawn at 0, 1, 3, 5, and 10 min and immediately mixed with 100 µl of ethanol. The samples were kept on ice for a few minutes before being centrifuged at 15,000g for 4 min. 200 µl of the supernatant was injected into LC system 3, and budesonide was quantified.

Chromatography. LC system used for conjugate formation experiments (LC system 1). The chromatography equipment used was from Varian (pump 9012, injector 9100) and controlled by Varian LC Star System. The column was a Li Chro CART 125-4 (RP-18), 125 × 4 mm, 5 µm. The samples were chromatographed isocratically with acetonitrile/methanol/25 mM phosphoric acid (46/46/8). The flow rate was 0.8 ml/min. Fractions were collected (1 min/fraction during 50 min), and radioactivity was determined by liquid scintillation counting.

LC system used for LC-MS (LC system 2). Two Pharmacia LKB 2150 pumps, controlled by Autochrom software via a CIM114 interface, and a Valco manual injector (50-µl loop) were used on line with the mass spectrometer. Analytical column: Applied Biosystems Brownlee Spheri-5 RP-8, 30 × 2.1 mm, 5 µm. For the mobile phase gradient, an isocratic phase (1 min) of 82% methanol (aq., v/v) with 50 mM sodium acetate was followed by 9 min of linear gradient to 91% methanol with 50 mM sodium acetate. After a 2-min isocratic period at 91% methanol, the system was returned to initial values. The flow rate was 0.20 ml/min.

View larger version (30K): [in this window] [in a new window]   Fig. 6.   Reconstructed ion current profiles of MAc ions of metabolites and total ions (bottom profile) from LC-ESI-MS of an extract from an incubation of budesonide + 2H8-budesonide with human lung microsomes are shown. Retention time (min:sec) is indicated on the peak apices. Signal intensity is seen on the right y axis.

LC system used for hydrolysis experiments (LC system 3). The chromatography equipment used was as follows: Waters M510 pumps, WISP 712 (autoinjector), Waters automated gradient controller M680, Waters data module 740 (integrator), and a Metric diode array detector model 1000s set at 245 nm. The column was a Li Chro CART 125-4 (RP-18), 125 × 4 mm, 5 µm, and the system operated at a flow of 0.8 ml/min. The mobile phases were 40% ethanol (A) and 100% ethanol (B). Isocratical separation at 95% A and 5% B was initially maintained for 13 min, then there was a change to 5% A and 95% B, and at 17 min, the system returned back to initial values.

Mass Spectrometry (MS). A Finnigan TSQ700 mass spectrometer was used with a Finnigan electrospray ionization (ESI) interface for the LC system (LC-system 2). Instrument control and data processing were performed on a DECstation 5000/125 with Finnigan ICL and ICIS II software (versions 7.0 and 7.4). Interface variables were as follows: spray voltage, 4 kV; sheath gas (N₂), 80 psi; auxillary gas (N₂), 8 units; capillary temperature, 220°C.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Conjugation: Dependence on CoA and ATP. Budesonide (100 µM) was incubated for 60 min with human lung microsomes with and without CoA and ATP. The radiochromatograms are shown in fig. 2. Lipophilic metabolites were formed, eluting in at least one major and two minor peaks in LC system 1. The formation was dependent on CoA and ATP. Of the reference compounds available, budesonide-21-oleate and budesonide-21-palmitate would comigrate with the major peak (fractions 31-38), while budesonide-21-palmitoleate would migrate among the minor peaks (fractions 18-23). Unchanged budesonide elutes with the first five to seven fractions in this chromatography system. When using human liver microsomes, the chromatograms (cf. fig. 4) qualitatively looked very similar to those obtained with lung microsomes. Cytosolic fractions did not produce these lipophilic metabolites.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Radiochromatograms obtained after incubation of tritiated budesonide (100 µM) with human lung microsomes (1 mg protein/ml) for 1 hr. The cofactors CoA and ATP were present at concentrations of 1 and 5 mM, respectively. LC system 1 was used. Unchanged budesonide elutes during the first few minutes and is thus not included in the figure. Microsomes from one single lung tissue sample were used to collect the data.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   The influence of externally added fatty acids can be seen in the formation of lipophilic metabolites from budesonide in human lung microsomes (top) and human liver microsomes (bottom). CoA and ATP were present in the incubations. For further details, see legend to fig. 2.

Conjugation: Time Courses. Budesonide (100 µM) was incubated for up to 60 min with human liver and lung microsomes in the presence of CoA and ATP. The main fractions of lipophilic metabolites (sum of fractions 31-38, cf. fig. 2) were taken as measure of conjugate formation. The results are shown in fig. 3. Both lung and liver catalyzed formation of the conjugates. The results presented in the figure are derived from only one microsome preparation from each organ. Thus, no conclusion about the relative capacity of the two organs should be made.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Formation of lipophilic metabolites from budesonide in human liver and lung microsomes (1 mg protein/ml) in the presence of CoA and ATP. Concentrations are as in fig. 2. The main fractions of lipophilic metabolites (fractions 31-38; cf. fig. 2) were taken as measure of metabolite formation. One sample of human liver microsomes and microsomes from one single human lung tissue sample were used to collect the data.

Conjugation: Effects of Fatty Acids. Budesonide (100 µM) was incubated with human lung and liver microsomes in the presence of ATP and CoA, and the incubations were also supplemented with oleic acid or palmitoleic acid. The results are shown in fig. 4. The different microsomes responded somewhat differently to the fatty acids. In lung microsomes, oleic acid almost abolished formation of metabolites eluting in fractions 18-23 (fractions covering retention time for budesonide-21-palmitoleate) and also reduced by around 50% the amount of metabolites eluting in fractions 31-38 (fractions covering retention times for budesonide-21-oleate and budesonide-21-palmitate). Palmitoleic acid added to lung microsomes prevented formation of metabolites eluting in fractions 31-38 and increased the amount of metabolites with its maximum in fractions 18-23. In liver microsomes, oleic acid increased formation of metabolites eluting in fractions 31-38 almost 3-fold and reduced amounts of metabolites in fractions 18-23. Palmitoleic acid added to liver microsomes inhibited formation of metabolites eluting in fractions 31-38 and increased the amounts of metabolites in fractions 18-23.

Identification of Lipophilic Conjugates. Electrospray ionization negative ion mass spectra of the reference compounds in the sodium acetate-fortified mobile phase (LC system 2), with the chosen interface conditions, showed mainly the acetate adduct ([M+CH₃COO ], MAc) (fig. 5).

View larger version (9K): [in this window] [in a new window]   Fig. 5.   Electrospray negative ion spectrum of reference budesonide-21-oleate from LC-ESI-MS in LC system 2.

View larger version (11K): [in this window] [in a new window]   Fig. 7.   Electrospray negative ion spectrum of oleic acid conjugates of budesonide + 2H8-budesonide from the LC-ESI-MS experiment in fig. 6.

View larger version (11K): [in this window] [in a new window]   Fig. 8.   Product ion spectra of MAc ions (m/z 753.5) from a human lung microsomal metabolite (top) and reference budesonide-21-oleate (bottom).

View larger version (23K): [in this window] [in a new window]   Fig. 9.   Proposed structures of fragments recorded in product ion spectra of MAc ions from metabolic fatty acid conjugates and reference esters of budesonide. Available references are indicated by a + symbol.

Hydrolysis by Lipase. Budesonide oleate and budesonide palmitoleate were incubated with porcine pancreatic lipase and bovine pancreatic cholesterol esterase. Budesonide was formed during these incubations in a time-dependent manner (fig. 10). The budesonide fatty acid conjugates are thus lipase substrates. Similar experiments showed that porcine liver carboxyl esterase, on the other hand, did not degrade the steroid fatty acid esters tested (not shown).

View larger version (19K): [in this window] [in a new window]   Fig. 10.   Formation of budesonide in incubations of budesonide-21-oleate (squares) or budesonide-21-palmitoleate (triangles) with porcine pancreas lipase (0.25 mg/ml, broken lines) or bovine pancreas cholesterol esterase (0.0125 mg/ml in the presence of 10 mM taurocholate, full lines).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

In the present study, we have shown that microsomes from human liver and lung are able to conjugate the antiinflammatory glucocorticoid budesonide with fatty acids. Moreover, the budesonide fatty acid esters were shown to be lipase substrates; thus, budesonide is regainable from the conjugates (summarized in fig. 11). The conjugation reaction was dependent on ATP and CoA, and various fatty acids (oleic acid, palmitic acid, palmitoleic acid, linoleic acid, and arachidonic acid) were utilized for the conjugation. Theoretically, budesonide contains two possible conjugation sites, the 11- and 21-hydroxyl groups. The conjugation apparently occurs on the 21-hydroxyl group because the metabolites match the synthetic reference 21-esters. We also found that fluticasone propionate does not undergo this metabolism (not shown). Fluticasone propionate lacks the hydroxyl at the 21-position but possesses the 11-hydroxyl. The 21-position is also known to be more accessible to chemical derivatization. For example, cortisol (7) and budesonide can be acetylated chemically on the 21-hydroxyl without affecting the more sterically hindered 11-hydroxyl group.

View larger version (10K): [in this window] [in a new window]   Fig. 11.   Formation and hydrolysis of budesonide-21-fatty acid esters as shown by the present study.

Fatty acid conjugation is well known to occur with cholesterol (8) and with several endogenous steroid hormones (9). Most well studied among the steroid hormones are the estrogens. Fatty acid ester formation has also been shown for xenobiotics like cannabinol (10), haloethanols (11), chlorinated phenols (12), and some others (reviewed in ref. 13). To our knowledge, this type of metabolism has hitherto not been demonstrated for steroid xenobiotics used in medical therapy.

The esterification of cholesterol is catalyzed by the ATP- and CoA-dependent microsomal acyl CoA:cholesterol acyl transferase (ACAT) (8) and possibly (at low taurocholate levels) by the pancreatic cholesterol esterase (14). Estrogen fatty acid conjugates are formed by enzyme(s) that some authors have termed acyl CoA:estradiol-17 acyltransferase (AEAT) (15). It has further been shown that several endogenous steroid hormones are esterified in rat plasma, probably by lecithin:cholesterol acyltransferase (16). Esterification of the same steroids in rat liver microsomes was dependent on CoA and ATP, but differences in the fatty acid composition of these esters compared with cholesterol esters, and lack of sensitivity to an ACAT inhibitor, suggested that perhaps some other enzyme than ACAT was involved (16). The enzymology of the esterification of the xenobiotics is not as well understood, but the reaction seems sometimesbut not alwaysdependent on ATP and CoA, suggesting other pathways beside ACAT or related enzymes (as discussed in ref. 13). The fact that the esterification of budesonide required ATP and CoA indicates that this type of steroid is conjugated by the ACAT/AEAT-type enzymes, but the exact identity of the enzyme responsible remains to be elucidated.

The purpose of the present paper was to identify the metabolites and to examine the biochemistry of the process. To obtain this, we chose to work with microsomes from human lung and liver because microsomes was the simplest system we could show that would form these metabolites (e.g. the cytosolic fractions were virtually devoid of activity). Other data we have collected suggest the formation of these conjugates in a wide variety of biological systems such as liver and lung microsomes from the rat, rat lung and trachea following inhalation or instillation in vivo, rat trachea following superfusion in situ, rabbit trachea in vitro, human blood monocytes, CACO-2 cells (human colon carcinoma cell line), COS cells (monkey kidney cells), F2002 cells (human embryonic lung fibroblasts), J774 mouse macrophages, R1 cells (rat fibroblasts), primary cultures of human bronchial epithelial cells, and human lung tissue following inhalation of budesonide (parts of these data are presented in preliminary form in refs. 5 and 17; further reports are presently being prepared). In most of these cases, the conclusions are based on LC-migration similarities, and no MS identification of budesonide fatty acid esters has yet been obtained.

The results shown in fig. 4 show that microsomes from human lung and human liver responded somewhat differently to the addition of exogenous free fatty acids to the incubations. The reason for this is not clear. However, fatty acid conjugation is a multistep process involving e.g. activation of fatty acids, and perhaps the rate determining step for budesonide ester formation may be different in microsomes from the two organs. Another reason may be that the peaks in the radiochromatograms contain more than one metabolite; the major peak contains at least the oleate and the palmitate ester, while the minor peak(s) contains the palmitoleate ester but probably other components as well. This finding further suggests that in an intact cell with regulated supply of activated fatty acids to the acylating enzymes, the pattern of fatty acid esters may be different than in microsomes.

The fatty acid conjugation of budesonide will most likely attenuate budesonide's acute pharmacological effects, since fatty acid conjugates of this type have negligible glucocorticoid activity per se (17, 18). However, as shown in this study, budesonide is regainable from the conjugates through the action of lipase. The fatty acid conjugates will most likely be retained intracellularly for a longer time than unchanged budesonide. Thus, it may be that the lipoidal conjugates will form an intracellular depot slowly releasing active steroid. The duration of tissue exposure to budesonide will depend partly on the rate of lipase-catalyzed hydrolysis of the conjugates. The conjugation will therefore prolong the duration of the pharmacological effects of budesonide. Investigations into the impact of the conjugation on duration of antiinflammatory effects in cell culture systems (17) and in rat airways in vivo (19) support these assumptions. The duration aspects are of particular interest because not all therapeutically used glucocorticoids can be conjugated in this way. As an example, fluticasone propionate lacks the 21-hydroxyl group and is not involved in this type of metabolism. It is tempting to speculate that the metabolic route of budesonide described in the present study explains why budesonide so efficiently can control mild asthma, even when dosed once daily (20).