Lethal photosensitisation of Nelicobacter species

Charles Millsonl, Michael Wilson2, Alexander MacRobertl, Wendy Thurrell3, Peter Mlkvyl, Claire Daviesl, 
Stephen
Bown 1.

1. The National Medical Laser Centre, University College London Medical School, The Rayne Institute, 5 
University Street,
London WC1E 6JJ.
2. Department of Microbiology, Eastman Dental Institute for Oral Health Care Sciences, 256, Gray's Inn 
Road, London
WC1X 8LD.
3. Department of Histopathology, Whittington Hospital, Highgate Hill, London N19 SNF.

ABSTRACT
Helicobacter pylori f H. pylori) is associated with a large number of gastroduodenal disorders. Clearance of 
the bacteria has been shown to benefit patients with duodenal ulcers, gastric ulcers and certain rare types of 
gastric tumours. Broad-spectrum antibiotics are the mainstay of current treatment strategies but side-effects, 
poor compliance and drug resistance limit their usefulness. We sensitised H. pylori with toluidine blue, 
haematoporphyrin derivative, aluminium disulphonated
• phthalocyanine, methylene blue or protoporphyrin IX prior to exposure to low-power laser light from either 
a gallium aluminium arsenide laser or a helium neon gas laser. All 5 sensitisers caused reductions of greater 
than 1000-fold in the number of viable bacteria. Light alone had no effect and only HpD caused a significant 
decrease in bacterial numbers without laser light. Next, we sensitised H. mustelae on explanted ferret gastric 
mucosa (ex vivo )with the same sensitisers and exposed them to light from a copper vapour pumped dye laser 
tuned appropriately. MB caused significant reductions in bacterial counts.
Successful lethal photosensitisation of Helicobacter pylori both in vitro and ex vivo raises the possibility of a 
local method for eradicating the bacteria, especially as the bacteria are only found in those parts of the upper 
gastrointestinal tract that are accessible to the endoscope.

1. INTRODUCTION

The presence of spiral-shaped organisms associated with gastric mucosa was first reponed in 1893 1. Nearly 
100 years later, y Marshall described Campylobacter pylori after prolonged culture of gastric specimens 
revealed the presence of bacterial colonies. A change of name to Helicobacter pylori followed as did a vast 
amount of research into the pathophysiological significance of this fastidious organism 2-4. The discovery of 
several species-specific Helicobacter including one in the ferret called Helicobacter mustelae has aided this 
research effon 5.

Every social and ethnic group studied around the world has been shown to harbour the organism 6. The 
prevalence in European populations rises with age from approximately 20% in 20 year olds to 50% in 60 year 
olds after which it begins to fall. Helicobacter pylori is usually found in the antral mucosa of the stomach. 
Although H. pylori is well adapted to survival in the highly acidic environment of the stomach, it favours the 
brush border, underneath the thick, protective mucus gel layer, where the pH approaches neutral 4.
Interest in H. pylori has arisen because the following findings. Firstly, that H. pylori is the main cause of the 
commonest type of gastritis (type B gastritis). Nearly 100% of patients with duodenal ulcers are colonised 
with H. pylori. Over 80% of patients with duodenal ulcers and associated H. pylori relapse within 12 months 
after healing of their ulcer, however eradication of the organism reduces that relapse rate to under 5% 7. The 
figures for the less common gastric ulcer suggest a similar relationship. A large European study reported in 
1993 suggested that H. pylori infection increased the risk of developing gastric cancer six-fold, and 
Wotherspoon et al described 6 cases of the rare MALT-lymphomas of the stomach which were cured by 
eradication of H. pylori 8,9 Finally, there have been numerous publications linking H. pylori infection with 
the common, but rather ill-defined disorder known as non-ulcer dyspepsia 4.

Treatment for H. pylori must be considered in Ihe light of current knowledge of who will benefit. Because of 
the high prevalence of the organism in the population, the majority of these carriers will be entirely 
asymptomatic, and indeed will never suffer from a gastric/duodenal related disease in their lives. Most 
clinicians agree that patients with duodenal ulcers benefit from eradication of their H. pylori, and similar 
cvidence is accumulating for the less common gastric ulcers (other than those induced by non-steroiW 
anti-inflammatory drugs). Populations at risk of developing gastric cancer cannot, at present, be 
recommended for eradication on the basis of current evidence, and the same is true for non-ulcer dyspepsia

I

Current treatment regimes for eradication of H. pylori rely on antibiotics, usually in combination with some 
sort of acidsuppressing agent. Single agent, double agent and triple antibiotic therapies attest to the 
difficulties of antibiotic penetration
to the relevant sites and the sturdiness of the organism 10 Formerly, triple therapy consisting of 
metronidazole, amoxycillin and bismuth was the combination of choice, but this has now given way to high 
dose amoxycillin and omeprazole, which seems to have similar efficacy.

These combinations achieve eradication rates of over 80%, but there are concerns with these therapies in 
three main areas.; Every antibiotic in use has side-effects which usually include gastrointestinal disturbance. 
Side-effects are often cited by patients as the reason for their poor compliance, but compliance is critical to 
the success of therapy. Finally, the use of broad spectrum antibiotics encourages resistance amongst normal 
flora resident elsewhere in the body, as well as resistance in H. pylori . Estimates from Africa where 
metronidazole use is widespread suggest that over 90% of H. pylori have acquired resistance to this 
antimicrobial agent 11.

The use of light to kill sensitised microbes probably dates back to the initial experiments of Raab with 
malaria paramecia at
the beginning of this century 12. The 1960's heralded the start of a flourish of interest in lethal 
photosensitization of bacteria coincident with the development of lasers and the opportunity to match light to 
the peak absorption wavelength of the sensitiser and to control power output. The development of this method 
of killing microbes could include a wide range of applications such as the eradication of viruses from whole 
blood, the eradication of Staph. agrees from cisterna magna of sheep, the clearance of pathogens associated 
with periodontitis from the oral cavity and control of candidiasis in the immunocompromised 13. The linking 
of sensitisers to monoclonal antibodies has opened up further windows of opportunity both for selectivity and 
efficacy 14.

The efficacy of lethal photosensitization of bacteria was thought to depend upon the presence or absence of 
the outer
membrane that surrounds the peptidoglycan membrane of some bacteria termed Gram-negative 15. 
Gram-positive bacteria (i.e. those without this membrane) appear to be susceptible to the technique, whilst 
Gram-negative bacteria were initially thought to require membrane stripping to render them susceptible to 
lethal photosensitization. This is no longer considered to be necessary fob all Gram-negative bacteria.

There have been two papers already published concerned with the lethal photosensitization of H. pylori . 
Bedwell et al used
a phthalocyanine and a copper vapour pumped dye laser to achieve eradication of the bacteria in vitro 16 
Wolfsen et al sensitised H. pylori attached to a gastric carcinoma cell line (Kato III) and achieved reductions 
of 1000-fold in bacterial
counts using haematoporphyrin derivative as sensitiser 17. Neither of these sensitizers have been further 
evaluated, nor has any eradication been attempted on gastric mucosa.

H. pylori colonises those areas of the upper gastrointestinal tract that are routinely accessed during a normal 
diagnostic endoscopy. This paper consists of an in vitro assessment of five sensitisers followed by an attempt 
to demonstrate lethal photosensitization of the bacteria on ex vivo specimens of gastric mucosa using an 
appropriate animal model.

2. MATERlALS AND METHODS
2.1. Bacteria

[lelicobacter pylori (NCTC 11637-11) were grown on blood agar plates containing a selective supplement 
(Oxoid SR069), for a period of 3 to 5 days in a microaerophilic atmosphere (Oxoid SR 056). On the day of 
the experiments, the bacteria were harvested from the plates and placed in solution of Wilkins-Chalgren 
broth. flelicobacler mastelae were harvested from the stomach of a ferret, and grown up on agar plates as for 
H. pylori.

2.2. Sensitisers

Five sensitisers were prepared in a solution of WC broth; toluidine blue O [TBO, Sigma Itd, Poole, UK] 
Haematoporphyrin derivative [HpD, Paisley Biochemicals, Glasgow], methylene blue [MB, Sigma Ltd], 
protoporphyrin IX [PPIX, sodium salt, Sigma Ltd] and aluminium disulphonated phthalocyanine [S2 from 
Prof. D. Phillips, Imperial College, London]. Each sensitiser was filter-sterilised prior to use. In the ex vivo 
experiment, ALA [Dusa Pharmaceuticals, Inc., Denville, NJ] was used. a-Aminolavulinic acid (ALA) is the 
rate limiting step in the haem biosynthetic pathway, of which PPIX is a precursor. Exogenous ALA causes a 
build up of haem precursors, the most photoactive of which is PPIX.

23 Lasers

Three lasers were used in this study; a gallium-aluminium-arsenide (GaAIAs) laser [Omega fibres & 
Technology] with a beam diameter of 9mm, a wavelength of 660nm and a power density of 17mW/cm2. The 
7.3mW helium-neon (HeNe) gas laser [NEC, Japan] has a wavelength of 632.8nm in a beam of 1.3mm 
diameter and a power density of 500mW/cm2. The copper vapour pumped dye laser can be tuned to the 
appropriate wavelength and has a variable power output.

2.4 . Animals

Adult ferrets (600g) [Mustela patons Pro ] were housed individually, in accordance with Home Office 
regulations.

25. In vitro experiments

One well of a sterile 96-well tissue culture plate was filled with 100>1 of sensitiser in WC broth and 100111 
of bacteria in WC broth. A 4mm mixing bar was added and the plate placed on a magnetic stirrer. Other wells 
were filled with the appropriate controls, as outlined below. The culture plate was covered and mixing 
allowed for a fixed period known as the pre-irradiation time. After the pre-irradiation time was over, the test 
well was exposed to laser light whilst the rest of the plate remained covered. After exposure to laser light, 
1001 of the test well contents was pipetted into a bijoux containing 900^11 of WC broth, and a series of 
dilutions prepared which were then plated out onto agar to allow counting of surviving organisms, usually 
after 3 to 5 days of incubation in a microaerophilic atmosphere. The bacteria which received sensitiser and 
light were designated 1+s+'. The effect of the sensitiser alone was observed by omitting the light exposure 
stage from a control well (I-s+), and the effect of the laser light alone was investigated by mixing the bacteria 
with 100R1 of WC broth and exposing this control well as previously described (l+s-). Finally, the original 
bacterial count was determined from an unsensitised, unirradiated well (lest). All experiments were 
performed with minimal exposure to ambient light, and with identical periods of exposure to ambient 
atmosphere. All experiments were performed with duplicate test and control wells.

The pre-irradiation time, the laser exposure time (energy density) and the sensitiser concentration were all 
varied in subsequent experiments to observe the effect on the efficacy of lethal photosensitization. Three 
concentrations of TBO were investigated as follows: Three test wells containing different sensitiser 
concentrations were incubated with bacteria and then exposed to the same energy dose. Three control wells 
containing identical concentrations of sensitiser were incubated with bacteria but not exposed to laser light. 
The energy density was then varied by exposing 4 wells containing the same TBO concentration and bacteria 
to four different energy densities. Three different pre-irradiation times were tested in three different wells 
with the same TBO concentration, prior to exposure to the same energy density. The three variables, pre-
irradiation time, sensitiser concentration and energy density were altered in the same way for each of the five 
sensitisers.

2.6Exvivo experiments

Ferret stomachs were removed immediately after killing the animal and tied at the duodenum. Two to five 
mls of sensitiser were introduced via the cardia and the cardia tied. After one hour of pre-irradiation time, the 
stomach was emptied, opened, and the antrum divided into four strips of lcm long and 0.5cm wide, these were 
subdivided into eight squares. Four of the squares were exposed to light from the copper vapour pumped dye 
lasa, at powa densities similar to those used in the in vivo experiments, as appropriate for each sensitiser. The 
second square was not exposed to laser light (I-s+). After laser exposure, the squares were placed in separate 
bijoux bottles containing WC broth, homogenised, and diluted serially as before. A control stomach was filled 
with WC broth and exposed to laser light in an identical manner to the sensitised stomachs, to ensure that 
laser light alone did not alter the viability of bacteria on gastric mucosa. Survivors were counted after 3 to 5 
days.

3.1. In vitro experiments

3. RESULTS

The five sensitisers TBO, HpD, MB, S2 and PPIX all successfully sensitised H. pylori to killing by low-power 
laser light. Laser light alone had no effect on bacterial viability in the absence of sensitisers. Figure 2 
compares the five sensitisers in terms of the energy density and sensitiser concentration required to cause at 
least a 3-loglo reduction in the viable count.

Figure 2. Comparison of all five sensitisers by plotting the minimum energy density and sensitiser 
concentration required to cause a 3-log lo reduction in colony forming units of H. pylori.

Only the sensitiser HpD reduced the viability of H. pylori in the absence of laser light and that to a much 
lesser extent than wilh light and sensitiser. The TBO- and PPIX-treated bacteria both showed a marked 
increase in bacterial kill when a more prolonged pre-irradiation time was used, which was not seen with the 
other three sensitisers.

3.2. Ex vivo experiments

The laser light alone did not affect the viability of H. mustelae when no sensidser was present. Lethal 
photosensitisation was unsuccessful with HpD, ALA, S2 and PPIX, up to a maximum energy dose of 
200J/cm2. Higher concentrations of MB were found to be more toxic to H. mustelae in the dark than the 
lower doses used in section 3.1, al;most complete eradication of H. mustelae could be achieved in these 
animals with exposure to laser light of sensitised regions of gastric mucosa harbouring the bacteria.

4. DISCUSSION

This investigation has demonstrated that HpD, MB, S2, PPIX or TBO can sensitise Helicobacter pylori to 
killing by low power laser light in vitro . Furthermore, irradiation with laser light of ferret gastric mucosa 
which has been pre-sensitised with MB, lead to localised near eradication of H. mustelae resident on the 
mucosa. Laser light alone did not affect bacterial viability at the energy densities used in this study.

Lethal photosensitisation using TBO has not previously been described with H. pylori although there have 
been several reports of lethal photosensitisation of other microbes with this sensitiser. Gram-positive (Sarcina 
lutea ) and Gram-negative bacteria (Escheriscia coli ) have been successfully killed with TBO and HeNe laser 
light 13. Also Wilson & Mia achieved
successful lethal photosensitisation of Candida albicans using TBO and 48J/cm2 delivered by a HeNe laser 
18. At a similar concentration of TBO successful killing of the Gram-negative periodontopathogens 
Fusobacterium nucleatum, Porphyromonas gingivalis and Actinobacillus actinomycetemcomitans was 
achieved with an energy dose of only 16J/cm2
which is considerably lower than the 160J/cm2 required to kill H. pylori 19. It is clear from these data that the 
requirement for a high energy dose is not dictated by the Grarn status of the organism, but other factors 
perhaps pertaining to the mechanism of cell death may play a role. This will be discussed further.

HpD has been previously described as a successful sensitiser of H. pylori . Wolfsen et al reported a method of 
attaching H. pylori to a gastric carcinoma cell line and killing the bacteria with PDT 17. They used an 
identical concentration of HpD
and a 200 Watt Xenon arc lamp with a 515nm long pass filter with an energy dose of between 1 and 10 
Jlcm2. They went on to investigate the effect of light on sensitised normal gastric mucosa and found that pig 
gastric mucosa sprayed with HpD and exposed to 100J/cm2 exhibited minor and reversible histological 
changes only in the gastric mucosa.

The endogenous porphyrin, protoporphyrin IX like haematoporphyrin and its derivatives, has been widely 
used in
photosensitising tumours but has not been used with bacteria. Our study showed that an energy dose of over 
300J/cm2 was required to achieve a reduction in bacteria equivalent to that achieved with the other 
sensitisers. The PPIX preparation used in this study was an acid rather than a salt, which suggests that it 
should be quite stable if administered orally, however PPIX is one of the photoactive products produced 
during haem synthesis and will accumulate if the rate determining step (the synthesis of ALA) is bypassed by 
exogenous ALA. Previous studies have shown that ALA administered orally to a
mammal results in extra glandular fluorescence of PPIX in gastric mucosa some 7-9 hours later 20. It is thus 
possible that oral ALA might lead to sensitisation of H. pylori in vivo, in the stomach and duodenum.
I

Methylene blue has been used as a sensitiser in PDT for over 60 years. The micro-organisms which are 
already known to be photo-inactivated by MB include Herpes, HIV and several other viruses, Proteus 
mirabilis, Escherichia coli and Salmonella typhimurium. 21. Karita et al described a nude mouse model 
colonised with H. pylori which were reduced in number when treated with 3mg/ml of MB alone 22. Although 
these reductions were modest the fact that any bacteria were killed suggested that the sensitiser must be stable 
in the gastric environment and must be gaining access to the bacteria, under the mucus layer. We were unable 
to confirm this 'dark toxicity' of MB towards H. pylori at the concentrations used in the in vilro experiments, 
but we did observe dark toxicity at the higher doses used in the ex vivo experiments. However the further 
addition of laser light resulted in near eradication of the bacteria in the small areas tested in our animal 
model, which was one of the aims of this ex vivo experiment. Furthermore, there is evidence to suggest that 
gastric mucosa sensitised with these concentrations of MB is unaffected by energy doses of over 200J/cm2 
(unpublished data). Clinically, MB is already in use for the rare blood disorder methaemaglobinaemia, and 
has been used for delineating regions of dysplasia in the stomach, also. It has also been used for the 
photodynamic therapy of superficial bladder tumours, so clinical trials might be possible in the not too distant 
future.

Only HpD and phthalocyanine have been described previously as effective photoinactivators of H. pylori. 
Bedwell et al used a mainly tri-sulphonated phthalocyanine incubated with H. pylori for 4 hours prior to 
exposure to 300 seconds of laser light supplied by a copper vapour pumped dye laser tuned to 675nm 16. Our 
experiment differed in that we employed a much shorter pre-irradiation time, our phthalocyanine was a pure 
disulphonated compound, and we used for the in vitro experiment the GaAIAs laser whose wavelength is 
660nm.

Overall, the data from the in vitro study were encouraging as it is clear that H pylori are susceptible to 
photoinactivation, but the ex vivo data suggest that only MB will lead to killing of H. pylori in its natural 
habitat, with the following proviso;

MB was only successful ex vivo at much higher doses that used in the in vitro experiments. A further 
evaluation of the other four sensitisers should include treatment at 10 to 100 times the doses we used.

The mechanism of action of lethal photosensitisation of microbes is likely to depend upon the type of 
sensitiser used, and may also vary with the Gram status of the bacteria. Membrane bound sensitisers like 
HpD, S2 and PPIX have been extensively studied in mammalian cells where the type II mechanism, involving 
the generation of singlet oxygen, is thought to be the main mechanism of oxygen-dependent cytotoxicity. 
However, the existence of reducing agents like NAD(P)H and glutathione in prokaryotic and eukaryotic cells 
might favour a type I mechanism with those sensitisers that can gain access to the cytoplasm. The use of free 
radical and superoxide scavengers to protect E. coli from lethal sensitisation with MB suggest both pathways 
are important whereas S. Iutea favours a type II pathway predominantly 21s23. These data are difficult to 
interpret as the scavengers have been shown to be non selective, but they do suggest that bacterial sensitiser 
interactions may not involve the same photodestructive pathway as mammalian cells, and may differ from 
species to species.

Those data show that H. pylori is susceptible to lethal photosensitisation and suggests that this may be a 
possible alternative method of eliminating the bacteria from the upper gastrointestinal tract. Clearly much 
further work is required to define the light and drug doses necessary to eliminate the bacteria, although the 
biggest challenge is likely to be the development of a light delivery system to ensure that a minimum light 
dose reaches all colonised regions. However, if these technical difficulties can be overcome, this study shows 
that an endoscopic therapy for H. pylori could offer a credible alternative to current antibiotic regimes.

ACKNOWLEDGEMENTS

This project was funded by the Sir Jules Thorn Trust. Prof. Bown is supported by the Imperial Cancer 
Research Fund. We should also like to thank Prof. D.Phillips (Imperial College, London), who provided the 
phthalocyanine.

REFERENCES

1. G. Bizzozero, "Uber die Schlauchformigen Drusen des Magendarmkanals und die Bezienhungren ihres 
Epithels zu dem Oberflachenepithel der Schleimhaut", Arch. Mikrobial. Anat., 42, 82-152, 1893. 2. D. Vaira, 
J. Holton, J. Dowsett, B.L.Oderda., "Helicobacter pylori iLS role in gastric disease", Dig. Dis, 8, 322-6, 1990. 
3. S. Moss & J. Calam, "Helicobacter pylori and peptic ulcers: the present position", Gut,33, 289-292, 1992. 
4. C.P.Dooley. "Helicobacter pylori: review of research findings", AlimenL Pharrnacol. Therap. 5(suppl.1), 
129-143, 1991. 5. J.G.Fox, P.Correa, N.S.Taylor, A.Lee, G.Otto, J.C.Murphy, R.Rose, "Helicobacter mustelae 
- associated gastritis in ferrets: an animal model of Helicobacter pylori gastritis in humans", 
Gastroenterology,99, 352-361, 1990. 6. F.Megraud, M.P.Brassens-Rabbe, F.Denis, A.Belbouri, D.Q.Hoa, 
"Seroopidemiology of Campylobactor pylori infection in various populations", J. Clin. Microbiol, 27, 1870-3, 
1989. 7. E.A.Rauws & G.N.J.Tytgat, "Cure of duodenal ulcer associated with erafication of Helicobacter 
pylori," Lancet, 335, 1233-5, 1990. 8. Eurogast study, "An international association between Helicobacter 
pylori infection and gastric cancer", LanceL 341, 1359-1362, 1993. 9. A.C.Wotherspoon, C.Doglioni, 
T.C.Diss, L.Pan, A.Moschini, M.de Boni, P.G.Isaacson, Lancet,342, 575-577, 1993. 10. N.Chiba, R.Babu, 
J.Rademaker, R.Hunt, "Meta-analysis of the efficacy of antibiotic herapy in eradicating Helicobacter pylori ," 
Lancet, 87, 1716-1726, 1992. 11. D.Y.Graham, G.M.Lew, H.M.Malaty, "Factors innuencing the eradication of 
Helicobacter pylori with triple therapy", Gastroenterology, 102, 493496, 1992. 12. O.Raab, "0ber die Wirkung 
fluorescierenden stoffe auf infusioreren", Z.biol, 39, 52446, 1900. 13. M.Wilson, "Photolysis of oral bacteria 
and its potential use in the treatment of caries and periodontal disease", J App Bac,75, 299-306, 1993. 14. 
J.S.Friedberg, R.G.Tompkins, S.L.Rakestraw, S.W.Warrenj A.J.Fischman, M.L.Yarmush, "Antibody-targeted 
photolysis. Bacteriocidal effects of Sn (IV) chlorin e6-dextran-monoclonal antibody conjugates", Annals of 
the New York Academy of Science, 618, 383-393, 1991. 15. Z.Malik, J.Hanania, N.Yehayau, New trends in 
photobiology. "Bacteriocidal effects of photoactivated porphyrrins- an alternative approach to antimicrobial 
drugs", J. Photochem Photobiol, 5, 281-293, 1990.

16.1.Bedwell, J.Holton, D.Vaira, AJ.MacRobert, S.GBown, "In vitro killing of Helicobacter pylori with 
photodynamic therapy", Lancet (letter), 335, 289-292, 1990. 17. H~Wolfsen, K.Wang, D.Alquist, M 
Pittelkow, F.Cockerill, "Light dositnetry in PDT for Hclicobacter pylon ", Photodvnamic theranv and 
biomedical lasers. Editors: P.Spinelli, M.Dal Fante dc R.Marchesini, p281-285, Elsevier Science Publishers 
B.V, Amsterdam, 1992. 18. M.Wilson & N.Mia, "Sensitisation of Candida albicans to killing by low-power 
laser", J Oral Pathol Med (in press). 19. M.Wilson, J.Dobson, S.Sarkar, "Sensitisation of 
periodontopathogenic bacteria to killing by light from a low-power laser", Oral Microbiol Immunol,8, 
182-187, 1993. 20. JRegula, AJ.MacRobert, A.Gorchein, G.Buonaccorsi, S.M.Thorpe, G.M.Spencer, 
AR.W.Hatfield, S.G.Bown, "Photosensitisation and photodynamic therapy of oesophageal, duodenal and 
colorectal tumours using ALA induced protoporphyrin IX- a pilot sXdy", Gut, (in press). 21. E.M.Tuite dc 
J.M.Kelly, "Photochemical interactions of methylene blue and analogues with DNA and other biological 
substrates", J.Photochem Photobiol, 21, 103-124, 1993. 22. M.Karita, QLi, K.Okita, "Evaluation of new 
therapy for eradication of Helicobacter pylori infection in nude mouse model", Am J Gastroenterol, 88, 1366 
1372, 1993. 23. J.Moan, K.Berg, E.Kvam, "Effects of Photodynamic therapy on DNA and DNA-related cell 
functions", ln: Photodvnamic theranv of neoolastic disease, Ed, D.Kessel, 197-209, CRC Press, Boston, 1992.