Ursodeoxycholic acid

ABSTRACT

We previously reported that the efficacy of photodynamic therapy (PDT) in cell culture was enhanced by ursodeoxycholic acid (UDCA), a nontoxic bile acid. In this study, we examined the ability of UDCA to promote tumor control by PDT in the mouse, using the radiation-induced fibrosarcoma tumor and the photosensitizing agent tin etiopurpurin (SnET2). These experiments revealed that the addition of UDCA to a PDT protocol promoted inhibition of tumor growth, a phenomenon unrelated to either altered SnET2 biodistribution or the level of vascular shutdown during irradiation. These results indicate that UDCA acts solely by promoting direct tumor cell kill by PDT.

Abbreviations: PDT, photodynamic therapy; RIF, radiation-induced fibrosarcoma; SnET2, tin etiopurpurin; UDCA, ursodeoxycholic acid.

INTRODUCTION

We have reported that ursodeoxycholic acid (UDCA), a relatively nontoxic member of the bile acid family, can enhance the efficacy of photodynamic therapy (PDT) in cell culture (1,2). Although initial studies were carried out with tin etiopurpurin (SnET2), other sensitizers were later examined (3). Experimental evidence indicates that UDCA acts by sensitizing the antiapoptotic protein Bcl-2 to photodamage and is only effective where Bcl-2 is among the PDT targets (3). To extend these findings we examined the effect of UDCA on PDT efficacy in a murine tumor model, the mouse radiation-induced fibrosarcoma (RIF), using SnET2 as the photosensitizing agent. Two factors pertinent to photodamage in animal models are biodistribution and vascular shutdown; the latter is often a major factor in cancer control (4). In this study, we report on the effects of UDCA on these factors and on the photodynamic efficacy of SnET2.

MATERIALS AND METHODS

Photosensitizer. SnET2 was prepared in the Department of Chemistry, Louisiana State University. This generally followed a published procedure (5), with modifications that will be reported in detail elsewhere. Absorbance spectra were acquired with a Genesis 5 spectrophotometer (Spectronic Instruments, Rochester, NY). Extinction coefficients were calculated and compared with published parameters (6,7). The sensitizer was formulated as described previously (8).

Animal model. Male C3H/HeJ virus-free mice were provided by the Charles River Laboratories and housed in a temperature-controlled facility. Animals were provided with fresh bedding and water dispensers daily and were fed ad libitum with standard rat chow. All experiments conformed to the guidelines and protocols established by the University of Louisville and Wayne State University.

Tumor system. The RIF tumor was maintained according to established protocols in C3H/HeJ mice (9). Before inoculation of the tumor, hair was removed from the right Hank of the mouse by shaving and depilation. Tumors were used for experimentation when they reached a surface diameter of 4-5 mm and a thickness of 2-3 mm and were free of evident necrosis.

PDT protocols. Male mice bearing the RIF tumor were treated with 2 mg/kg of SnET2 by tail-vein injection 24 h before irradiation. In some cases, UDCA (10 mg/kg) was administered intraperitoneally 15 min before irradiation. Tumor control was assessed after irradiation of tumor-bearing animals 24 h after SnET2, using 665 nm light (100 J/cm^sup 2^).

Drug biodistribution. Animals were treated with 2 mg/kg SnET2 by tail-vein injection 24 h before tissue collection. Where specified, UDCA (10 mg/kg) was administered by intraperitoneal injection 15 min before tissue collection.

Tissue samples (

Blood was collected in chilled tubes containing a few crystals of ethylenediaminetetraacetic acid. The red cells were removed by centrifugation (300 g, 5 min, 10[degrees]C) and the supernatant plasma was collected. A measured aliquot of the plasma (100-200 [mu]L) was diluted in 3 mL of a 10 mM solution of the nonionic detergent Triton X-100, and fluorescence of the mixture was measured as described above. SnET2 standards were prepared similarly. All fluorescence studies were carried out using a charge-coupled device array (Instaspec IV, Oriel Co., Stratford, CT) so that a complete emission spectra could be obtained.

Light sources. For photosensitizer activation, the light source was an LDX Optronics (New Brunswick, NJ) fiber-coupled laser (665 + or - 10 nm) with a Portable Power System thermoelectric cooler and Laser Driver from Portable Power Systems Inc. (St. Louis, MO). The output of this laser was coupled to a 400 [mu]m diameter glass fiber-optic fitted with a microlens at its end. The power density was adjusted to 75 mW/cm^sup 2^ as measured by a Photomedica Laserguide power meter. Mice were treated with 100 J/cm^sup 2^. The treatment area was a circle of 1.5 cm diameter; this included both tumor and tumor-free regions.

Tumor response. Groups of 20 C3H/HEJ mice were intradermally implanted with the RIF tumor on the right thigh and were observed until the tumor reached a diameter of 4-5 mm. Animals were then treated with 1 or 2 mg/kg SnET2 intravenously. Some animals also received an intraperitoneal injection of 10 mg/kg UDCA 15 min before irradiation. Twenty-four hours after injection of the sensitizer, a 1.5 cm diameter spot containing tumor and normal skin was exposed to 665 nm light at a power density of 75 mW/cm^sup 2^, using a light dose of 100 J/cm^sup 2^. Mice were examined for tumor regrowth daily for the first 14 days after treatment and weekly thereafter for a total of 42 days. Any photodamage to normal tissues became evident 1 day after irradiation.

Window chambers. Male C3H/HEJ mice were depilated and implanted with RIF tumor subcutaneously to the right of the midline at the approximate level of the 12th thoracic vertebrae and 8-10 mm from the spinous processes. Five days later, the animals were anesthetized with 60 mg/kg sodium pentobarbital and placed in a sterile environment, and a 1.5 cm oval of skin was removed from the left side of the midline on the dorsum of the animal. The animals were then fitted with a skin chamber (JPS Inc., Saugus, CA), using an Ethicon 4-0 silk suture with a C-4 cutting needle. Animals were then allowed to equilibrate 30 min after surgery on a heating pad kept at 37[degrees]C and then transferred to Plexiglas holders, and a digital picture was taken using a Sony CyberShot digital camera. A cropped version of the digital picture was used as a vessel map. The animals were then transferred to the intravital microscopy station where vessel measurements were carried out.

To better observe the tumor vasculature we also followed a previously published method for skin chamber studies in mice (10). The two symmetrical titanium frames were mounted in the back of the animal to sandwich the extended double layer of skin. The RIF tumor cells were implanted intradermally on the left side of the double layer. On the next day, the opposite side of the skin (a 15 mm circle) was removed aseptically, and the glass coverslips were incorporated on both sides of the frame. After 6-7 days, the tumor reached a diameter of 3 mm; at this time, sensitizer or UDCA (or both) was administered and the tumor irradiated as described above.

Intravital microscopy. The entire area of the chamber was monitored and recorded after a 30 min equilibration period following surgical intervention. Subsequent measurements and recordings were made immediately after PDT and at 24 and 48 h after PDT. Vessel measurements and recordings were performed using a Zeiss standing microscope with an MTI DC330 Hamamatsu color camera (Dage-MTI, Michigan City, IN) connected via S-video to a 20 A/V controller (B&K Components, Ltd., Buffalo, NY). The signal was then passed through an S-VHS recorder to a Time-Date generator (TitleMaker 3000, Videonics Inc., Campbell, CA) and into a Videum(R) Capture card and displayed on a computer monitor. A second signal was sent from the B&K A/V controller to an S-VHS recorder and displayed on a monitor. Observations were performed using 4x and 10x objectives. Vascular measurements were calibrated using 10 and 100 [mu]m reference slides.

Data analysis for intravital microscopy. Observations were recorded for off-line analysis. Tumor was divided in eight quadrants, and vessel numbers and size were measured in each area at each observation point (an average of eight of 12 vessels per quadrant were counted). An image of each quadrant at each lime point was captured as a still image with the Video Capture Program (Winnov USA, Sunnyvale, CA), and vessel number and size were examined after transferring the files in Adobe Photoshop 5.0. Nonfunctioning vessels were defined as those that disappear from the field of view, along with vessels that were visible but contained nonflowing red cells. The percentage of functioning vessels in each quadrant was calculated by dividing the number obtained at each observation point by the initial value (before treatment). For each animal the measurements from the eight quadrants were averaged at each observation time. Three animals were used in each experimental group. The average of the three animals' analyses was plotted as the number of tumor functioning vessels (percentage of the total number before PDT) vs the observation time.

Arteriole and venule pairs were also chosen for study based on their diameter (20-30 [mu]m), as described previously (11,12). The red blood cell column diameter is the measurement of the inner lumen of the vessels.

Statistical analysis of tumor growth studies. This study was carried out by the Statistical Consulting Center, Department of Biostatistics and Bioinformatics, School of Public Health/Information Science, University of Louisville. This survival analysis began with the fitting of a Cox proportional hazards model (13) that included the photosensitizing agent (SnET2) and the inhibitor (UDCA) as covariates.

RESULTS AND DISCUSSION

The bile acid UDCA was previously shown to enhance PDT-induced phototoxicity in cell culture when the protein Bcl-2 was among the PDT targets. This study was designed to determine whether UDCA would also promote tumor control in an animal tumor model. The data obtained are consistent with the proposal that only direct tumor cell kill and not vascular shutdown is promoted by the addition of UDCA to a PDT protocol.

Biodistribution studies

The biodistribution of SnET2 (2 mg/kg) 24 h after administration of the drug by tail-vein injection to tumor-bearing mice is shown in Table 1. The intraperitoneal injection of UDCA (10 mg/kg) 15 min before sacrifice of animals did not affect the levels of SnET2 in any tissue or in plasma. Therefore, we cannot attribute any effect of UDCA on alterations in drug biodistribution.

PDT: tumor regression

Coadministration of UDCA 15 min before irradiation increased tumor responses at both 1 and 2 mg/kg doses of SnET2. Irradiation (100 J/cm^sup 2^ at 665 nm) was carried out 24 h after injection of the sensitizer and 15 min after injection of UDCA. The fluence rate was 75 mW/cm^sup 2^ (n = 20 mice). Figure 1 shows results obtained with two different SnET2 doses. Animals were classified as to whether they had tumor with a diameter of 12 mm or less. The number of animals with tumor

At either SnET2 dose, the tumor regrowth data (Fig. 1) indicate a detectable effect of UDCA on the outcome. Statistical analysis indicated that addition of UDCA provided a decrease in instantaneous risk by a factor of 0.497 (P = 0.006), whereas use of SnET2 alone provided a decrease in instantaneous risk by a factor of 0.2707 (P

PDT: vascular shutdown

Figure 2 shows the number of functional tumor vessels after PDT, compared with the initial number. Administration of UDCA without irradiation or irradiation without drugs did not affect either normal or tumor vasculature. In the PDT studies, irradiation (100 J/cm^sup 2^ at 665 nm) was carried out 24 h after administration of SnET2 and 15 min after injection of UDCA. Hemorrhage with venule leakage was observed immediately after PDT in the tumor at both SnET2 dosages, but 92% and 63% of the tumor vessels were still open and flowing with doses of 1 and 2 mg/kg SnET2, respectively.

Observations carried out 24 h after PDT indicated that the leakage was more extensive with no blood flow detected with a 2 mg/kg SnET2 dose, whereas a few of the larger vessels (including some of the feeding vessels) were still open and flowing after a 1 mg/kg drug dose. After 48 h, tumors in animals that had received the 2 mg/kg SnET2 dose had no open or flowing vessels, whereas tumors treated with the 1 mg/kg drug dose showed some blood flow in veins at the periphery of the tumor.

We found that addition of UDCA did not promote or retard the vascular shutdown associated with PDT at either SnET2 dose. Because vascular effects form an important element of the cancer control evoked by PDT (4), these results suggest that promotion of direct phototoxicity, previously identified in cell culture (1), likely represents the mechanism whereby tumor control is enhanced by UDCA in vivo.

Figure 3 shows changes in the arteriole vessel diameter before and immediately after irradiation, as well as at 24 and 48 h later. As in previous protocols, tumor sites were irradiated (100 J/cm^sup 2^ at 665 nm) 24 h after injection of SnET2 and 15 min after administration of UDCA. Under these conditions, the normal vessels showed only a minimal effect after PDT, whereas the mean diameter of tumor vessels was substantially decreased by PDT. There was no detectable effect of UDCA on these results.

In a previous study using SnET2 formulated with Cremophor EL and at a higher light dose (360 J/cm^sup 2^), normal venule leakage and arteriole dilatation was reported, and macromolecule leakage was also evident (14). The latter study involved the cremaster intravital model, and tumor blood flow was still evident 2 h after PDT treatment. These results agree with data obtained in this study, obtained immediately after tumor irradiation.

CONCLUSIONS

We previously reported that UDCA could enhance the apoptotic response to SnET2 (1) via the promotion of photodamage to the antiapoptotic protein Bcl-2 (3). Bcl-2 is known to be the target for a variety of photosensitizing agents (15,16). In this report, we show that the ability of UDCA to promote the photodynamic efficacy of SnET2 can be demonstrated with statistical significance in a murine tumor model in vivo. This effect of UDCA was not related to alterations in SnET2 biodistribution; tumor and plasma levels of the sensitizer, at the time of irradiation, were not affected by UDCA. We found that the addition of UDCA did not enhance vascular shutdown associated with PDT. Because the UDCA can affect tumor oxygenation and limit the phototoxic action of PDT (17,18), it may be generally useful for the promotion of PDT efficacy involving the use of photosensitizing agents that preferentially target Bcl-2. Because of the lack of a vascular effect of UDCA, we conclude that the enhanced PDT efficacy afforded by addition of this bile acid to a PDT protocol derives solely from an enhanced direct tumor cell kill. Studies are now underway to define the precise mechanism of this proapoptotic effect of UDCA.

Acknowledgements-We thank Linda T. Harrison, Ann Marie Santiago and Brendan Leeson for excellent technical assistance. This work was supported by grants CA 23378 and CA92618 from the National Cancer Institute, National Institutes of Health.

[para] Posted on the website on 30 July 2003

REFERENCES

1. Kessel, D., J. A. Caruso and J. J. Reiners Jr. (2000) Potentiation of photodynamic therapy by ursodeoxycholic acid. Cancer Res. 60, 6985-6988.

2. Kessel, D., M. Castelli and J. J. Reiners Jr. (2002) Apoptotic response to photodynamic therapy versus the Bcl-2 antagonist HA14-1. Photochem. Photobiol. 76, 314-319.

3. Castelli, M., J. J. Reiners Jr. and D. Kessel (2003) Promotion of PDT efficacy by bile acids. Proc. SPIE 4952, 10-16.

4. Dougherty, T. J., C. J. Gomer, B. W. Henderson, G. Jori, D. Kessel, M. Korbelik, J. Moan and Q. Peng (1998) Photodynamic therapy. J. Natl. Cancer Inst. 90, 889-905.

5. Morgan, A. R. and N. C. Tertel (1986) Observations on the synthesis and spectroscopic characteristics of purpurins. J. Org. Chem. 51, 1347-1350.

6. Garbo, G. M. (1996). Purpurins and benzochlorins as sensitizers for photodynamic therapy. J. Photochem. Photobiol. B: Biol. 34, 109-116.

7. Pogue, B. W., R. R. Redmond, N. Trivedi and T. Hasan (1998) Photophysical properties of tin ethyl etiopurpurin I (SnET2) and tin octaethylbenzochlorin (SnOEBC) in solution and bound to albumin. Photochem. Photobiol. 68, 809-815.

8. Garbo, G. M., V. H. Fingar, T. J. Wieman, E. B. Noakes III, P. S. Haydon, P. B. Cerrito, D. Kessel and A. R. Morgan (1998) In vivo and in vitro photodynamic studies with benzochlorin iminium salts delivered by a lipid emulsion. Photochem. Photobiol. 68, 561-568.

9. Fingar, V.H. and B. W. Henderson (1987) Drug and light dose dependence of photodynamic therapy: a study of tumor and normal tissue response. Photochem. Photobiol. 46, 837-841.

10. Leuning, M., F. Yuan, M. D. Menger, Y. Boucher, A. E. Goetz, K. Messmer and R. K. Jain (1992) Angiogenesis, microvascular architecture, microhemodynamics, and interstitial fluid pressure during early growth of human adenocarcinoma LS174T in SCID Mice. Cancer Res. 52, 6553-6560.

11. Fingar, V. H., P. K. Kik, P. Hayden, P. B. Cerrito, M. Tseng, E. Abang and T. J. Wieman (1999) Analysis of acute vascular damage after photodynamic therapy using benzoporphyrin derivative (BPD). Br. J. Cancer 79, 1702-1708.

12. McMahon, K. S., T. J. Wieman, P. H. Moore and V. H. Fingar (1994) Effects of photodynamic therapy using mono-l-aspartyl chlorin e6 on vessel constriction, vessel leakage, and tumor response. Cancer Res. 54, 5374-5379.

13. Cox, D. R. and D. Oakes (1984) Analysis of survival data. In Monographs on Statistics and Applied Probability 21. Chapman & Hall, New York.

14. Fingar, V. H., T. J. Wieman, S. A. Wiehle and K. A. Siegel (1992) Effect of PDT using tin etiopurpurin on tissue microvasculature. Photochem. Photobiol. 55S, 55s-56s.

15. Kessel, D. and M. Castelli (2001) Evidence that bcl-2 is the target of three photosensitizers that induce a rapid apoptotic response. Photochem. Photobiol. 74, 318-322.

16. Oleinick, N. L. and H. H. Evans (1998) The photobiology of photodynamic therapy: cellular targets and mechanisms. Radiat. Res. 150(5 Suppl.), S146-S156.

17. Fingar, V. H. (1996) Vascular effects of photodynamic therapy. J. Clin. Laser Med. Surg. 14, 323-328.

18. Busch, T. M., E. P. Wileyto, M. J. Emanuele, F. Del Piero, L. Marconato, E. Glatstein and C. J. Koch (2002) Photodynamic therapy creates fluence rate-dependent gradients in the intratumoral spatial distribution of oxygen. Cancer Res. 62, 7273-7279.

Greta M. Garbo1, M. Graca H. Vicente2, Victor Fingar1 and David Kessel*3

1 Department of Surgery, University of Louisville, Louisville, KY;

2 Department of Chemistry, Louisiana State University, Baton Rouge, LA and

3 Departments of Pharmacology and Medicine, Wayne State University School of Medicine, Detroit, MI

Received 13 May 2003; accepted 18 July 2003

* To whom correspondence should be addressed at: Department of Pharmacology, Wayne State University School of Medicine, Detroit, MI 48201, USA. Fax: 313-577-6739; e-mail: dhkessel@med.wayne.edu

Copyright American Society of Photobiology Oct 2003
Provided by ProQuest Information and Learning Company. All rights Reserved