Photodiagnosis and Photodynamic Therapy (2005) 2, 129—134
Formation of protoporphyrin IX from carboxylicand amino-derivatives of 5-aminolevulinic acid Miron Kaliszewski M.Sc. a,b,∗∗, Asta Juzeniene a, Petras Juzenas Ph.D. a,∗, Miroslaw Kwasny b, Jaroslaw Kaminski c, Zbigniew Dabrowski c, Jerzy Golinski c, Johan Moan a a
Department of Radiation Biology, Institute for Cancer Research, 0310 Montebello, Oslo, Norway Institute of Optoelectronics, Military University of Technology, ul. Kaliskiego 2, 00-908 Warsaw, Poland c Industrial Chemistry Research Institute, ul. Rydygiera 8, 01-793 Warsaw, Poland b
Available online 23 March 2005 KEYWORDS Photodynamic therapy; Aminolevulinic acid derivatives; Stability; Carcinoma cell line
Summary Background: Stability of ALA is an important factor for photodynamic therapy (PDT). The dimerization of ALA to pyrazines takes place via the amine group. It is, therefore, to be expected that blocking this group by addition of a formyl group should result in a more stable compound. Methods: The ability of a new N-formyl derivative of ALA (N-f-ALA) to form protoporphyrin IX (PPIX) was compared with that of ALA and three of its ester (methyl, butyl and hexyl) derivatives. Dark toxicity of the compounds was measured using MTT assay. Formation of PPIX was measured by ﬂuorescence spectroscopy. Results and conclusions: N-f-ALA showed an outstanding stability in water solutions even at pH 7. However, it induced no PPIX neither in WiDr cells in vitro, nor in mouse skin in vivo. A probable reason is lack of an enzyme that can cleave the bond between the formyl group and ALA. Thus, steric hindrance may prevent processing of the compound into porphobilinogen. N-f-ALA did not inhibit PpIX formation from ALA and is unable to enter the heme cycle. Generation of ALA from its derivatives, therefore, seems to be an essential step in PPIX synthesis. © 2005 Elsevier B.V. All rights reserved.
Corresponding author. Tel.: +47 22 93 51 13; fax: +47 22 93 42 70. ∗∗ Co-corresponding author. Tel.: +48 22 683 70 17; fax: +48 22 666 89 50. E-mail addresses: [email protected]
(M. Kaliszewski), [email protected]
Cancer treatment and detection with photosensitizing and ﬂuorescing dyes is being widely accepted. Photodynamic therapy (PDT) is selective photosensitization of malignant tissue by drugs followed by exposure to visible light. Excitation of the photosensitizing drug with light of proper wave-
1572-1000/$ — see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/S1572-1000(05)00012-8
130 length results in generation of singlet oxygen, the main cytotoxic agent in PDT [1,2]. Although 5-aminolevulinic acid (ALA) is not a photosensitizing agent itself, it is widely used for PDT and ﬂuorescence diagnosis (FD) of different cancers and cutaneous pathological conditions . Enzymatic conversion of ALA, via the heme cycle, leads to formation of the photosensitizing agent protoporphyrin IX (PPIX) [2,4]. Due to the hydrophilic properties of ALA, its penetration through skin and cell membranes is limited. Thus, only superﬁcial changes can be treated [3,5]. Synthesis of more lipophilic ester derivatives may diminish this drawback. Such a modiﬁcation of ALA has been shown to improve PDT considerably, leading to the synthesis of PPIX in cells, a need of lower concentrations of the esters compared to ALA, and, in some cases, a higher selectivity than a parent drug [3,6]. Another important factor concerning the use of ALA is its instability, especially in buffered solutions at physiological pH, that are administered orally or intravesically into the bladder. Under certain conditions dimerization results in generation of cyclic pyrazin derivatives. Thus, the concentration of active drug decreases upon storage . Esteriﬁcation of ALA seems to improve its bioavailability . On the other hand, stability, which depends on temperature, concentration and pH, is not improved. Since the dimerization of ALA to pyrazines takes place via the amine group , it is to be expected that blocking this group by addition of a formyl group should result in a more stable compound.
Materials and methods Chemicals 5-Aminolevulinic acid hydrochloride (ALA, MW = 167.6 g/mol), methyl aminolevulinate hydrochlo-
M. Kaliszewski et al. ride (m-ALA, MW = 181.7 g/mol), butyl aminolevulinate hydrochloride (b-ALA, MW = 223.7 g/mol), hexyl aminolevulinate hydrochloride (h-ALA, MW = 251.8 g/mol) were synthesised by J. Kaminski M.Sc. and Z. Dabrowski M.Sc. (Chemistry Research Institute, Warsaw, Poland) with the method published previously . N-Formyl-aminolevulinate (N-f-ALA, M = 159.1 g/mol) was synthesised by Dr. J. Golinski by the method described below. Chemical structures of the compounds are shown in Fig. 1. RPMI-1640 medium, penicillin/streptomycin solution, l-glutamine, Trypsin—EDTA, phosphatebuffered saline (PBS) and other chemicals were obtained from Sigma-Aldrich Norway AS (Oslo, Norway). Foetal Calf Serum (FCS) was obtained from PAA Laboratories Gmbh (Linz, Austria). All chemicals were of the highest purity commercially available. Stock solution (3 mM) of ALA and its esters in RPMI-1640 medium was prepared before experiment and then diluted to a ﬁnal concentration of 0.01, 0.05, 0.1, 0.25, 0.5 and 1 mM.
Preparation of N-formyl-ALA To a stirred and cooled in ice bath mixture of formic acid (1.5 ml) and acetic anhydride (3 ml), pyridine (5 ml) was added followed by addition of 5-aminolevulinic acid hydrochloride (1.68 g). The mixture was stirred 30 min in ice bath then 45 min at room temperature. Water (1 ml) was added and the mixture was stirred overnight. Potassium carbonate (680 mg) was added and stirred for 30 min and evaporated in vacuum. To the residue hot ethanol (50 ml) was added, inorganic salts ﬁltered off and supernatant evaporated. Recrystalization with ethanol gave N-formyl-5-aminolevulinic acid, 860 mg (60%), mp: 142—143 ◦ C. 1 H nuclear magnetic resonance (NMR) data: (D2 O, TMS) ı = 8.13 (s, 1H, CHO), 4.22 (s, 2H, CH2 ), 2.84 (t, 2H, J = 6.4 Hz) and 2.63 (t, 2H, J = 6.4 Hz) (CH2 CH2 ) ppm.
Figure 1 Chemical structures of investigated derivatives of ALA: 5-aminolevulinic acid (ALA), 5-aminolevulinic acid methyl ester (m-ALA), 5-aminolevulinic acid butyl ester (b-ALA), 5-aminolevulinic acid hexyl ester (h-ALA), N-formylderivative of 5-aminolevulinic acid (N-f-ALA).
Formation of protoporphyrin IX from carboxylic- and amino-derivatives of 5-aminolevulinic acid
Cell line WiDr cells, derived from a primary adenocarcinoma of the human recto-sigmoid colon , were maintained in exponential growth in RPMI 1640 medium (Sigma) with 10% FCS (PAA Laboratories, Linz, Austria), 100 units/ml penicillin and 100 g/ml streptomycin and 2 mM l-glutamine (Sigma). The cells were grown and incubated in 75 cm2 cell culture ﬂasks (Nunc, Roskilde, Denmark) at 37 ◦ C in a humidiﬁed atmosphere 5% CO2 and subcultured twice a week using 0.01% trypsin in 0.02% EDTA.
Incubation of cells with ALA and ALA esters About 5 × 105 WiDr cells were seeded out in 12well plates (Costar, Corning Inc., Corning, NY) containing 2 ml of culture medium and were incubated for 48 h for proper attachment to the substratum. Subsequently, the cells were washed with serum-free medium and incubated for different time in RPMI 1640 medium containing different concentrations of ALA or ALA derivatives. Serumfree medium has been used in order to avoid porphyrin extraction from the cells . Exposure to light was avoided during incubation with ALA or ALA derivatives.
Dark toxicity of ALA and its derivatives The survival of the cells was measured using a colorimetric MTT assay . The cells were seeded out in 12 well plates 48 h before incubation with ALA and its derivatives. Subsequently they were washed with serum free medium and incubated for 24 h in medium containing different concentrations of 5aminolevulinates. MTT was dissolved in phosphate-buffered saline (PBS; pH 7.4) at 5 mg/ml, ﬁltered to become sterile and stored at 4 ◦ C. Stock solution (100 l) of the was added to each well containing 2 ml of medium, and the well plates were incubated at 37 ◦ C for 4 h. After that time medium was removed and the cells were washed with ice cold PBS. The formazan crystals were dissolved by adding 200 l of isopropanol per well. Samples (25 l) were transferred from each well into a 96-well microplate with 200 l of isopropanol. The optical density was read on a Multiskan MS (type 352, Labsystems, Helsinki, Finland) plate reader using a 570 nm bandpass ﬁlter. The cell viability (cell survival) was expressed as a percentage of viable cells relative to the untreated control cells.
Topical delivery of N-formyl-ALA on mouse skin Experiments using mice were approved by the National Animal Research Authority and were performed according to the European Convention for the Protection of Vertebrates Used for Scientiﬁc Purposes. Female hairless BALB/c mice were used. The mice were around 10 weeks of age weighing approximately 25 g. They were anaesthetised with subcutaneous injection of a mixture of Hypnorm (Janssen Pharmaceutica B.V., Tilburg, The Netherlands) and Dormicum (Hoffmann-La Roche AG, Basel, Switzerland) (1:1 v/v) for a short period (around 15 min) at the beginning of the experiments in order to facilitate proper application of the creams. The animals were normally active during the rest of the experiment. For topical application 1 mmol/g of the drug was prepared in a lipophilic cream (Unguentum, Merck, Darmstadt, Germany).
Fluorescence measurements The ﬂuorescence of PPIX in the cells was measured directly with a luminescence spectrometer (LS50B, Perkin-Elmer, Norwalk, CT). The excitation wavelength was 407 nm, corresponding to the maximum of the Soret band of the PPIX excitation spectrum, and ﬂuorescence emission was measured at 636 nm. The 407 nm excitation light from the luminescence spectrometer was of low intensity (less than 1 mW/cm2 ) and did not induce any signiﬁcant photobleaching of PPIX. The excitation and emission slits were set at 5 and 10 nm, respectively. A 515 nm long-pass ﬁlter was used in the emission optical path of the instrument during detection of the emission spectra. A bifurcated ﬁbre-optic probe (Perkin-Elmer accessory, tip diameter 6 mm), connected to the luminescence spectrometer, was placed on the surface of the mouse skin and the ﬂuorescence was measured.
Results Accumulation of PPIX induced by ALA and its esters in cells in vitro The kinetics of PPIX formation for different concentrations of ALA and its derivatives are shown in Fig. 2. Among the compounds tested, the most effective one was h-ALA, which induced measurable levels of PPIX even at the lowest concentration studied (0.01 mM). At concentrations ranging
M. Kaliszewski et al.
Figure 2 Accumulation of PPIX in WiDr cells incubated with ALA, m-ALA, b-ALA, h-ALA and N-f-ALA as a function of the incubation time for various concentrations of the drugs in the medium.
from 0.01 to 0.5 mM the ﬂuorescence increased until 9 h and then reached a plateau. The decrease of ﬂuorescence after 9 h for 1 mM h-ALA is due to cytotoxicity of the compound (Fig. 3) that caused detachment of the cells from the surface of the cell culture dish and thus gave a distortion of the measurements. For the same reason 3 mM h-ALA induced very low levels of PPIX. b-ALA showed a lower ability to form PPIX than h-ALA did. Nevertheless, at 0.01 mM it induced signiﬁcant levels of PPIX. It also showed saturation
after 9 h. Contrary to h-ALA, at concentrations of 1—3 mM, the ﬂuorescence increased up to 24 h. The lowest effective concentration of ALA and m-ALA was 0.25 mM. m-ALA had a slightly lower ability to form PPIX than ALA. However, at 3 mM they produced similar levels of PPIX. At this concentration the ﬂuorescence readings for the latter esters were slightly higher than that for b-ALA. None of the concentrations of N-f-ALA studied induced PPIX formation.
Formation of protoporphyrin IX from carboxylic- and amino-derivatives of 5-aminolevulinic acid
Figure 3 Evaluation of dark toxicity with MTT assay. WiDr cell were incubated for 24 h with ALA or ist derivatives. Data are the mean value ± S.E. (n = 3).
Dark toxicity N-f-ALA did not show any dark toxicity, even at the highest investigated concentration (3 mM). The toxicity of ALA increased with concentration, and at 3 mM the viability of the cells was reduced by about 40%. m-ALA was less cytotoxic than ALA. h-ALA showed the highest cytotoxicity. At 3 mM nearly all cells were killed. The toxicity of b-ALA was slightly higher than that of ALA, but signiﬁcantly lower the than that of h-ALA.
Application of N-f-ALA on mouse skin in vivo N-f-ALA induced no PPIX in normal mouse skin in vivo. The data are not shown because the ﬂuorescence signal after topical application of N-f-ALA was similar to that of the control untreated mouse skin.
Discussion Penetration through the plasma membrane of cells and through the stratum corneum of skin are rate limiting steps in ALA-PDT . It has been suggested that this process can be improved by estriﬁcation of ALA. Considerably lower doses and shorter application times, are needed for such lipophilic ester derivatives compared to those of the more hydrophilic ALA [6,12]. The results presented in this work, concerning PPIX formation in cells with lipophilic esters of ALA, agree with other published works [6,13]. Production of PPIX strongly depends on the lipophilicity of the
ALA ester: for short chain derivatives, like m-ALA, it is lower than for ALA. This is surprising since mALA is more lipophilic than ALA: the difference may be related to the deesteriﬁcation of m-ALA to ALA, which supposedly has to take place before the heme synthesis can proceed. However, for the esters with longer chains (b-ALA and h-ALA) PPIX formation is more efﬁcient than for the parent drug [1,3,6,14]. ALA, as well as its aliphatic esters tested so far, are unstable in water solutions at pH ≥ 5.5 [7,15]. Therefore, an effort was made to synthesize compounds with improved physicochemical properties. N-f-ALA showed an outstanding stability even at pH 7 and at concentrations up to 30 mM. Under such conditions there was no change in the absorption spectrum during 6 h of incubation at 37 ◦ C (Kaliszewski and Kwasny, unpublished data). This improved stability may be attributed to steric hindrance that prevents the formation of a C N bond between two ALA molecules. This step is unavoidable, and leads to the formation of pyrazines at increased pH in competition with enzymatic biosynthesis of porphobilinogen [3,8,16—18]. The latter fact may also explain the lack of PPIX formation with N-f-ALA in the WiDr cell line (Fig. 2) as well as in animal models (not shown; Kwasny, unpublished data). This observation is in agreement with earlier published results that demonstrate that this group of ALA derivatives is in most cases inefﬁcient in inducing PPIX [1,3,14,19]. Since some ALA analogues, with modiﬁed amino groups, like 5-ﬂuorolevulinic acid or levulinic acid, are inhibitors of porphobilinogen synthase (PBGS) [16,18], we tested the inﬂuence of N-f-ALA on the ability of ALA to produce PPIX. The formation of
134 PPIX in WiDr cells, incubated for different times in medium containing 0.5 mM ALA and different concentrations of N-f-ALA ranging from 0 to 2 mM, was not affected by the presence of N-f-ALA (data not shown). Thus, N-f-ALA is not an inhibitor of PBGS and does not enter the heme cycle at all, at least not in the WiDr cells. In conclusion, the presented results show that N-f-ALA, despite its excellent stability, seems to be an inefﬁcient prodrug for PPIX formation in WiDr cells in vitro and in mouse skin in vivo. The lack of PPIX formation by N-f-ALA in the presented systems, seems to be caused by deﬁciency of enzymes that cleave speciﬁcally C N bound. It is possible that in other model like stomach or liver, where peculiar sets of enzymes are present, this compound could enter the heme cycle as ALA after hydrolysis. Moreover the fact that physicochemical properties of N-f-ALA were not affected by increased pH gives very important information on mechanism of ALA degradation and possible synthesis of new generation derivatives that will be not only stable but also efﬁciently induce PPIX directly in the site of action. The facts that N-f-ALA induced no PPIX and, when applied concurrently with ALA, did not inhibit the formation of PPIX from ALA show the necessity of the de-esteriﬁcation of the ALA-derivatives before entering the heme cycle.
Acknowledgement The present work was ﬁnanced by the Norwegian Cancer Society (DNK), by the Norwegian Radium Hospital Research Foundation (RF) and by Polish State Committee for Scientiﬁc Research (KBN).
References  Kloek J, Beijersbergen von Henegouwen GMJ. Prodrugs of 5aminolevulinic acid for photodynamic therapy. Photochem Photobiol 1996;64(6):994—1000.  Roy BN, Van Vugt DA, Weagle GE, Pottier RH, Reid RL. Effect of continuous and multiple doses of 5-aminolevulinic acid on protoporphyrin IX concentrations in the rat uterus. J Photochem Photobiol B: Biol 1997;41:122—7.  Lopez RFV, Lange N, Guy R, Bentley MV. Photodynamic therapy of skin cancer: controlled drug delivery of 5-ALA and its esters. Adv Drug Deliv Rev 2004;56(1):77—94.
M. Kaliszewski et al.  De Rosa FS, Tedesco AC, Lopez RFV, et al. In vitro skin permeation and retention of 5-aminolevulinic acid ester derivatives for photodynamic therapy. J Controlled Release 2003;89:261—9.  Tunstall RG, Barnett AA, Schoﬁeld J, et al. Porphyrin accumulation induced by 5-aminolevulinc acid esters in tumour cells growing in vitro and in vivo. Br J Cancer 2002;87:246—50.  Gaullier JM, Berg K, Peng Q, et al. Use of 5-aminolevulinic aid esters to improve photodynamic therapy on cells in culture. Cancer Res 1997;57:1481—6.  Kaliszewski M, Kwasny M, Kaminski J, Dabrowski Z, Burdziakowska E. The stability of 5-aminolevulinic acid and its ester derivatives. Acta Pol Pharm Drug Res 2004;61(1):15—9.  Bunke A, Zerbe O, Schmid H, Burmeister G, Merkle FTP, Gander B. Degradation mechanism and stability of 5aminolevulinic acid. J Pharm Sci 2000;89(10):1335—41.  Dabrowski Z, Kwasny M, Kaminski J, et al. The synthesis and application of 5-aminolevulinic acid derivatives in photodynamic therapy and photodiagnosis. Acta Pol Pharm Drug Res 2003;60(3):219—24.  Noguchi P, Wallace R, Johnson J, et al. Characterization of WiDr: a human colon carcinoma cell line. In Vitro 1979;15:401—8.  Iinuma S, Farshi SS, Ortel B, Hasan TA. Mechanistic study of cellular photodestruction with 5-aminolaevulinic acidinduced porphyrin. Br J Cancer 1994;70(1):21—8.  Cosserat-Gerardin I, Bezdetnaya L, Notter D, Vigneron C, Guillemin F. Biosynthesis and photodynamic efﬁcacy of protoporphyrin IX (PPIX) generated by 5-aminolevulinic acid (ALA) or its hexylester (hALA) in rat bladder carcinoma cells. J Photochem Photobiol B: Biol 2000;59:72—9.  Juzeniene A, Ma LW, Juzenas P, Iani V, Lange N, Moan J. Production of Protoporphyrin IX from 5-aminolevulinic acid and two of its esters in cells in vitro and tissue in vivo. Cell Mol Biol 2002;48(8):911—6.  Casas A, Batlle AM, Butler AR, et al. Comparative effect of ALA derivatives on protoporphyrin IX production in human and rat skin organ cultures. Br J Cancer 1999;80(10):1525—32.  Novo M, H¨ uttmann G, Diddens H. Chemical instability of 5-aminolevulinic acid used in the ﬂuorescence diagnosis of bladder tumours. J Photochem Photobiol B: Biol 2000;34:143—8.  Jarret C, Stauffer F, Henz ME, et al. Inhibition of Escherichia coli porphobilinogen synthase using analogs of postulated intermediates. Chem Biol 2000;7(3):185—96.  Luond RM, Neier R. On the formation of the mixed pyrrole catalysed by porphobilinogen synthase from Rhodobacter spheroides. Biochim Biophys Acta 1996;1289(1):83—6.  Frere F, Schubert WD, Stauffer F, et al. Structure of porphobilinogen synthase from Pseudomonas aeruginosa in complex with 5-ﬂuorolevulinic acid suggests a double Schiff base mechanism. J Mol Biol 2002;320(2):237—47.  Ninomiya Y, Itoh Y, Tajima S, Ishibashi A. In vitro and in vivo expression of protoporphyrin IX induced by lipophilic 5-aminolevulinic acid derivatives. J Dermatol Sci 2001;27(2):114—20.