Protoporphyrin IX distribution following local application of 5-aminolevulinic acid and its esterified derivatives in the tissue layers of the normal rat colon

Photodynamic diagnosis and especially therapy after sensitization with 5-aminolevulinic acid (ALA) is hampered by limitations of uptake and distribution of ALA due to its hydrophilic nature. Chemical modification of ALA into its more lipophilic esters seems to be promising to overcome these problems. The aim of the present study was to investigate the comparative kinetics of protoporphyrin IX (PPIX) fluorescence in rat colonic tissue after topical application of ALA and its esterified derivatives, ALA-hexylester (h-ALA), ALA-methylester (m-ALA) and ALA-benzylester (b-ALA). Fluorescence intensity induced by PPIX in normal colonic tissue was quantified using fluorescence microscopy at 1, 2, 4, 6 and 8 h after sensitization. Mucosa exhibited higher fluorescence levels compared to the underlying submucosa or smooth muscle. Peak fluorescence intensities were seen 4 h after local sensitization with 86.0 mol ml−1 ALA (513 ± 0.57 counts per pixel), 6.6 mol ml−1 m-ALA (508 ± 35.50) and 4.8 mol ml−1 h-ALA (532 ± 128.80), and 6 h after sensitization with 4.6 mol ml−1 b-ALA (468 ± 190.27). A 13–18 times lower concentration of ALA esters was required for fluorescence intensities reached with ALA alone. A similar degree of the fluorescence ratio between mucosa and muscularis (5–6:1) was detected for ALA and its derivatives. The time point of the maximum value of this ratio was consistent with peak fluorescence levels for ALA and each ALA-ester. The clinical feasibility and the advantages of topical ALA ester-based fluorescence for detection of malignant and premalignant lesions need further investigations. © 2001 Cancer Research Campaign   http://www.bjcancer.com

Exogenous 5-aminolevulinic acid (ALA) induced protoporphyrin IX (PPIX) fluorescence endoscopy and photodynamic therapy (PDT) are new methods for detection and treatment of premalignant and malignant lesions in the gastrointestinal tract (Gossner et al, 1998;Endlicher et al, 1999Endlicher et al, , 2001Overholt et al, 1999). ALA is a prodrug in haem biosynthesis, which is converted into the photosensitizing agent protoporhyrin IX (PPIX). Due to a low ferrochelatase activity in tumour cells, PPIX accumulates selectively in malignant tissue Krieg et al, 2000). Furthermore PPIX is associated with only limited skin photosensitivity of 1-2 days (Regula et al, 1995).
In urology fluorescence cystoscopy after local instillation of ALA into the bladder has significantly enhanced the diagnosis of malignant and premalignant bladder lesions (Kriegmair et al, 1996;Filbeck et al, 1999;Koenig et al, 1999). The local application of ALA seems to be a promising alternative approach as well compared to the systemic administration in the detection of malignant and premalignant lesions in the gastrointestinal tract (Endlicher et al, 1999).
However, when using ALA topically for the PDT of malignant and premalignant lesions, this modality appears to be limited by the amount of uptake and penetration of ALA due to its hydrophilic nature. Therefore, some lipophilic molecules such as esterified ALA derivatives have been used both in vitro (Gaullier et al, 1997;Kloek et al, 1998;Casas et al, 1999;Eleouet et al, 2000), and in vivo (Lange et al, 1999), and better effects were achieved, compared to the application of unmodified ALA. The biosynthesis of PPIX induced by ALA esters depends on the aliphatic length of the alcohol used for esterification. While the long-chained ALA-esters (C5-C8) induced a higher PPIX accumulation in human tumour cell lines, the short-chained ALAesters (C1-C3) were less effective (Gaullier et al, 1997;Washbrook and Riley, 1997), indicating that the mechanism and rate of uptake of ALA and its esterified derivatives may be different. Berg et al demonstrated, that the transport of ALA into human adenocarcinoma cells is mediated by an active transport mechanism via β-amino acid and gamma aminobutyric acid carriers, whereas ALA esters are taken up by transporters other than those used for uptake of ALA or they may penetrate the plasma membrane by passive diffusion (Rud et al, 2000).
The aim of the present study was to investigate the comparative kinetics of PPIX fluorescence in rat colonic tissue for topical application of ALA and its esterified derivatives. Following our preclinical studies on different carcinoma cell lines (CaCO2, SW480 and HT29) with different ALA derivatives we choose h-ALA and b-ALA because the results indicated that the maximum benefit for the formation of PPIX is obtained with these esters (Hausmann et al, 2001). ALA methylester was used for the comparative study since to our knowledge there are no clinical data for short-chained ALA esters available.

MATERIAL AND METHODS
Female CD rats approximately 200 g in weight were used for this study. ALA was obtained from medac GmbH, Hamburg, Germany. ALA methyl-, hexyl-and benzylester were synthesized at the Institute of Inorganic Chemistry, University of Regensburg. All experiments were carried out under nitrogen atmosphere. ALA esters were prepared by reacting ALA with the corresponding alcohol in the presence of thionyl chloride giving the ALA esters as the hydrochloride salts (Takeya, 1992). An excess of the freshly destilled thionyl chloride was added dropwise. Then ALA was added to the solution and stirred at 60˚C for 5 hours. The excess of alcohol was evaporated in vacuo. After recrystallization from methanol/diethylether at -20˚C pure ALA esters were obtained as white hydrochloride salts. Identification of the synthesized compounds was based on mass spectroscopy, elemental analysis and nuclear magnetic resonance (NMR). The protocol for animal studies was approved by the animal board of Unterfranken, Germany, adhering to standards required to meet UKCCCR guidelines regarding animal ethics. The animals were anaesthetized with Ketanest © /Rompun © and underwent a small median laparotomy with mobilization of the caecum. After ligation of the terminal ileum the caecum was incised, a probe placed into the lumen to wash out faeces by a lavage with about 50 ml of 0.9% NaCl solution until the fluid became clear. Before sensitization with 10 ml of a sterile solution of ALA (86.0 mol ml -1 : 14.4 mg ml -1 ; pH 6.8-7.4), m-ALA (6.6 mol ml -1 ; 1.2 mg ml -1 : pH 8.8-7.4), h-ALA (4.8 mol ml -1 : 1.2 mg ml -1 ; pH 6.8-7.4) or b-ALA (4.6 mol ml -1 : 1.2 mg ml -1 ; pH 6.8-7.4) the anus was ligated with a running loop suture. 1, 2, 4 and 6 h after administration of ALA, m-ALA, h-ALA, b-ALA and additionally 8 h after application of b-ALA animals were killed. The entire colon was removed and 7-9 tissue samples stored immediately in liquid nitrogen. In addition, tissue from liver and kidney were sampled. The pieces were prepared in a darkened room in order to avoid photobleaching of the tissue. Frozen sections of 20 µm thickness were cut and stored at -80˚C.
In parallel conventional 5 µm cryostat sections were cut and stained with haematoxylin and eosin (H&E) for histological diagnosis. For analysis of PPIX fluorescence a fluorescence microscope (Axiovert S 100, Zeiss, Oberkochen, Germany) was used. Specimen were illuminated by a mercury arc lamp (360 +/-20 nm) and emitted fluorescence (630 +/-30 nm) was recorded by a highsensitive (4096 levels of gray) and high-resolution (1317 × 1035 pixel, with a pixel size of 6.8 × 6.8 µm) peltier cooled (-15˚C) CCD-camera (Princeton Instruments NJ, USA), using a 10 × /0.30 PLAN-NEOFLUAR lens. Care was taken to ensure the same experimental setup for each acquisition (exposure time) to allow accurate comparability of fluorescence intensities. Specific tissue layers and whole images (liver, kidney) were analysed using this technique to determine the relative intensities of porphyrin fluorescence, after correction for autofluorescence levels of each respective tissue layer as measured on specimens from unsensitized animals (MetaMorph, Universal Imaging Corp, PA, USA). For confirmation of histology the sections of the microscopic areas of fluorescence imaging were compared with the stained sections.

RESULTS
M-ALA, h-ALA and b-ALA were used in a dose of 6.6 mol ml -1 (1.2 mg ml -1 ), 4.8 mol ml -1 (1.2 mg ml -1 ) and 4.6 mol ml -1 (1.2 mg ml -1 ). 4 h after administration of the 3 ALA-esters nearly the same intensity of porphyrin fluorescence was achieved using a 13-to 18-fold higher concentration of ALA. Consequently a dose of 86.0 mol ml -1 (14.4 mg ml -1 ) ALA was used for the fluorescence kinetic study.
Fluorescence intensity was highest in the mucosa. Figure 1(A-D) shows the fluorescence levels in the mucosa, muscularis mucosa, submucosa and muscularis propria. Following administration of m-ALA and h-ALA fluorescence intensity peaked 4 h after administration. Fluorescence with b-ALA peaked at 6 h and reached a Mean fluorescence intensities (± standard deviation) in the mucosa, muscularis mucosae, submucosa and muscularis propria at various times of incubation with h-ALA (4.8 mol ml -1 : 1.2 mg ml -1 ) plateau after 4 h for ALA, respectively, but no further time points were examined for sensitization with ALA. Peak fluorescence induced by 4.6 mol ml -1 b-ALA (6 h after application) were higher than those 4 h after administration of h-ALA (4.8 mol ml -1 ), m-ALA (6.6 mol ml -1 ) and ALA (86.0 mol ml -1 ).
Different fluorescence intensities of mucosa and muscularis propria as represented by fluorescence ratios were achieved (Figure 2). Following administration of ALA, m-ALA and h-ALA the highest fluorescence ratios between mucosa and muscularis propria occured 4 h after sensitization, following application of b-ALA the maximum fluorescence ratio was seen at 6 h respectively, in accordance to peak mucosal fluorescence levels.
In order to evaluate systemic bioavailability after local application of ALA, h-ALA, m-ALA and b-ALA PPIX fluorescence intensity was also measured in liver and kidney at various time points.
While the esterified derivatives of ALA (h-ALA 4.8 mol ml -1 , b-ALA 4.6 mol ml -1 and m-ALA 6.6 mol ml -1 ) induced no significant increase of PPIX fluorescence intensity in liver and kidney, application of ALA (86.0 mol ml -1 ) resulted in higher values in these organs with a peak or plateau after 4 h (Figures 3  and 4). Figure 5 shows a typical microscopic fluorescence image of a frozen section of the rat colon after sensitization with ALA and h-ALA.

DISCUSSION
We have investigated the microscopic distribution of ALA-induced PPIX fluorescence intensity for the esterified derivatives of ALA in the rat colon as well as in liver and kidney. We quantified the PPIX fluorescence at various time intervals after administration of