Lethal and mutagenic effects of photoinduced in haploid yeast (Saccharomyces cerevisiae) by two new monofunctional pyridopsoralens compared to 3-carbethoxypsoralen and 8-methoxyprosalen

Mutation Research. 148 (1985) 47-57 Elsevier

47

MTR 03964

Lethal and mutagenic effects photoinduced in haploid yeast (Saccharomvces cerevisiae) by two new monofunctional pyridopsoralens compared to 3-carbethoxypsoralen and 8-methoxypsoralen D. Averbeck 1, S. Averbeck 1, E. Bisagni 2 and L. Moron 2 I Insntut Curie, Biologie, 26 rue d'UIm, 75231 Paris Cedex 05, and 2 lnstitut Curie, Biologie. Centre Universitaire, B~t. 110, 91405 Orsay (France)

(Received 22 March 1984) (Revision received 7 August 1984) (Accepted 21 August 1984)

Summary The photobioiogical effects of two monofunctional pyridopsoralens (PPs), pyrido[3,4-c]psoralen and pyrido[3,4-cl-7-methylpsoralen were studied and compared to those of 3-carbethoxypsoralen (3-CPs) and 8-methoxypsoralen (8-MOP) in a haploid wild-type strain of yeast (Saccharomyces c e r e o i s i a e ) . The capacity of PPs to photoinduce lethal effects in the presence of 365-nm radiation was not only higher than that of the monofunctional compound 3-CPs, but also higher than that of the bifunctional compound 8-MOP. This activity was apparently independent of oxygen, and it was found that it was probably due to the induction of monoadducts in DNA. A high effectiveness of PPs on the induction of cytoplasmic 'petite' mutations was observed suggesting a high photoaffinity towards mitochondrial DNA. In contrast to 8-MOP, the strong cell killing activity of PPs was not accompanied by a strong inducing effect on nuclear mutations ( H I S + reversions or can R forward mutations). For these endpoints, PPs were less effective per unit dose of 365-nm radiation and also less efficient per viable cell than 8-MOP. From this, it appears that the lesions photoinduced by the former compounds show a more lethal than (nuclear) mutagenic potential. Furthermore, the fact that PPs were even less mutagenic (nuclear) per viable cell than the monofunctional compound 3-CPs suggests that the activity of these agents may differ in frequency and nature of lesions induced. The photobioiogical activity of PPs in haploid yeast appears to be in line with the recent proposition for their use in photochemotherapy.

Photoreactive furocoumarins (psoralens) (Musajo and Rodighiero, 1972; Pathak et al., 1974; Song and Tapley, 1979; Scott et al., 1976; Parsons, 1980) are employed with success in the photochemotherapy of certain skin diseases such as psoriasis, vitiligo and mycosis fungoides (Anderson and Voorhees, 1980; Pathak et al., 1981). One has, however, become aware of their inherent mutagenic (Bridges, 1979; Scott et al., 1976; Burger and Simons, 1979a, b) and carcinogenic potency not only in rodents (Grekin and Epstein, 1981;

Zajdela and Bisagni, 1981; Young et al., 1983; Cartwright and Walter, 1983) but also in man (Stern et al., 1979). The observation that certain monofunctional furocoumarins, such as angelicin and 3-carbethoxypsoralen forming only monoadditions in DNA are less phototoxic and less mutagenic in many cell systems than the bifunctional, DNA-crosslinking furocoumarins, such as 8methoxypsoralen (8-MOP), 4,5',8-trimethylpsoralen (TMP) and 5-methoxypsoralen (5-MOP) (Averbeck et al., 1975, 1979, 1981, 1984; Grant et

0027-5107/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

4~

al., 1979; Coppey et al., 1979; Averbeck and Moustacchi, 1980; Grossweiner and Smith, 1981; Abel and Schimmer, 1981; Venturini et al., 1980; Pani et al., 1981; Papadopoulo et al., 1983), actually used in photochemotherapy (PUVA) (Anderson and Voorhees, 1980; Pathak et al., 1981), has led to the suggestion that new and more photoreactive monofunctionai furocoumarins should be developed as photochemotherapeutic drugs (Averbeck and Moustacchi, 1979; Bordin et al., 1979). The linear derivative of psoralen, 3-carbethoxypsoralen (3-CPs), was one of the first monofunctional compounds showing both photochemotherapeutic activity on psoriatic lesions in human patients without a concomitant induction of acute skin erythema and hyperpigmentation, and a reduced genotoxicity in yeast (Averbeck and Moustacchi, 1979, 1980), and an absence of carcinogenic effects in mice, in contrast to 8-MOP (Dubertret et al., 1979; Averbeck 1982). Recent clinical studies with 3-CPs indicate, however, that in comparison to 8-MOP this compound is less photostable, less effective and may give rise in some patients to side effects such as delayed erythema and photoallergic reactions (Dubertret et al.. 1981; Kimura and Mizuno, 1981; Kimura et al., 1981). At the same time, angular monofunctional derivatives of psoralen, for example 4,5'-dimethylangelicin and 5'-methylangelicin, were proposed as photochemotherapeutic drugs (Bordin et al., 1981). However, their development and general use were hindered because of reports on their apparent carcinogenic activity in mice (West et ai., 1982). Furthermore, a high mutagenicity of 4,5'-dimethylangelicin was found in human skin fibroblasts (Swart et al., 1983). With the aim of developing new monofunctional furocoumarins as photochemotherapeutic drugs which exhibit a high photoreactivity towards DNA but a reduced mutagenic and carcinogenic activity together with an equal or higher activity on psoriasis than the furocoumarins actually in use (for example, 8-MOP), new linear derivatives of psoralen were synthesized (Moron et al., 1983). As in the case of 3-CPs (Averbeck et ai., 1978), the aim was to obtain pure monofunctional furocoumarins by introducing at the 3,4 reaction site

of the psoralen a bulky chemical group favouring only the production of 4',5'-monoadditions to pyrimidine bases in DNA. This was achieved by introducing a pyrido group at the 3.4 reaction site (Moron et al., 1983). The pyrido group was chosen in order to increase the photoreactivity of the molecule, since in the case of antimitotic ellipticines, this group conferred a higher degree of complexing to DNA (Tourbez-Perrin et al., 1980), and complexing in the dark is an important step for photoreactions of psoralens with DNA (Dall'Acqua, 1977). The present paper investigates in the yeast, Saccharomyces ceret, isiae, the photobiological activity of two new monofunctional pyridopsoralens and compares this to the monofunctional furocoumarin, 3-carbethoxypsoralen, and the wellknown bifunctional furocoumarin, 8-methoxypsoralen. Materials and methods

Strains We used the haploid strain N123 (a, hisl) (Moustacchi, 1969) of the yeast Saccharomyces cerevisiae. Media and culture conditions Yeast cells were grown at 30°C in liquid complete medium containing 1% yeast extract (Difco), 2% bacto peptone (Difco) and 2% glucose in distilled water. For the determination of the colonyforming ability, i.e. survival, cells were plated on complete medium solidified with 2.6% bacto agar (Difco). Respiratory-deficient cytoplasmic 'petite' mutants were identified by the tetrazolium overlay technique (Ogur et al., 1957). Reversions from his- to H I S + in strain N123 were detected on a minimal medium containing 0.67% bacto yeast nitrogen base without amino acids (Difco), 3.2% bacto agar (Difco) and 2% glucose. For the detection of forward mutations (CAN s to canR), a minimal medium supplemented with 20 /xg/ml histidine and 40 ~ g / m l L-canavanine sulphate (Sigma) was used. The resistance to canavanine is due to the inactivation of arginine permease, the activity of which is governed by a single gene Canl (Grenson et al., 1966). The clone of N123 (CAN s) used showed a low background

49

level of spontaneous mutations (can R) (7.4 × 10 -6 + 2.8), and the background level of spontaneous H I S .+ revertants was 1.5 × 10- 8 + 0.5. Chemicals 8-Methoxypsoralen (8-MOP) was a commercial product from Chinoin, Milan (Italy). 3-Carbethoxypsoralen (3-CPs) was obtained as previously described (Queval and Bisagni, 1974). Pyrido[3,4-c]psoralen (compound IA, MW 237) and pyrido[3,4-c]-7-methylpsoralen (compound IB, MW 251) were newly synthesized according to Moron et al., 1983). Since the latter compounds showed a very limited solubility in water (1 #g/ml), comparative studies were made using the furocoumarins at final concentrations of 5 ~tM. The molar extinction coefficients (M --~ • 1 • cm -~) at 365 nm were obtained from ethanolic solutions: 1150 for 8-MOP, 6230 for 3-CPs, 600 for compound IA and 850 for compound IB (Blais et al., 1984; J. Blais, personal communication). Fig. 1 shows the structures of the furocoumarins used.

BI-FUNCTIONAL FUROCOUM,ARIN

8-Methoxyp~rolen

MONO-FUNCTIONAL

FUROCOUMARINS

~o~

3-Cartethoxyl~or~en

Py r~do (2 Z.-C) osoralen IA

~,ridof3J,-c) 7-mettO,tosom/en IB

Fig. 1. Molecular structures of the furocoumarins used.

Treatments with furocoumarins and 365-nm (UVA) irradiation Haploid cells in the stationary phase of growth were treated with the photosensitizing compounds as previously described (Averbeck and Moustacchi, 1980; Averbeck et ai., 1981). The treatment involved incubation of cells for 30 min at room temperature in the presence of the furocoumarin (5 #M) followed by exposures of the cell suspension (106 cells/ml for survival studies and 107-108 cells/ml for mutation studies) to 365-nm (UVA) radiation from a HPW 125 Philips lamp. This lamp emits mainly at 365 nm (91%) and only weakly at 313 nm (2.6%), at 334 nm (5.4%) and at 405/408 nm (1%). Due to a 6.2ram-thick pyrex glass filter, the contribution of wavelengths below 340 nm was negligible, and the total radiance was about 99% at 365 nm. The dose rate was 1.2 k J / m 2 / m i n as determined by a digital radiometer J-260 with the sensor J-260-2A (Ultraviolet Products, Inc., San Gabriel, CA, U.S.A.). After treatment, aliquots of the cell suspension were plated on the different media for the detection of lethal and mutagenic effects. Experiments were performed at least in triplicate. The results include the standard errors of the means. Results

In control experiments the incubation of haploid cells (N123) of Saccharomyces cerevisiae in the dark for 30 rain with 5 #M of either of the two pyrido[3,4-c]psoralens (compounds IA and IB), 3CPs or 8-MOP, did not produce a significant inhibition of survival or an induction of cytoplasmic or nuclear mutations (HIS+ revertants, or can R forward mutations). 365-nm (UVA) irradiation alone was also without significant genetic effect in the dose range used in the photosensitization experiments. Thus, we focus here on the description of experiments using combined treatments of furocoumarin and 365-nm UV radiation (UVA). Effects on survival and cytoplasmic 'petite' induction Fig. 2 illustrates the survival curves obtained for the haploid wild-type strain N123 after treatment with the pyrido{3,4-c]psoralens IA and IB, 3-CPs and 8-MOP and UVA. The new compounds

50

10

TABLE 1

'a,,

SURVIVAL P A R A M E T E R S FOR T H E H A P L O I D S-I'RAIN N 123 OF Saccharomyces cere~'istae

O!

s e x lO -t)

LJ

t~

DOt

Compound (5/,tM)

Dq ( k J / m 2 )

D37 ( k J / m 2 )

Do (kJ/m 2 )

8-MOP 3-CPs IA IB

1.5 0 0 0

3.3 19 I 0.5

1.8 19 1 0.5

a: oool

~"~ I B

Y O-MOP

00001! DOSEOF 365 ~ RADIAr/ON(kJm-2) Fig. 2. Survival curves obtained for the haploid wild-type strain N 123 of the yeast Saccharomyces cereois~ae after treatment with the pyrido[3,4-c]psoralen (IA) (A)0 the pyrido{3,4-c]-7-methylpsoralen (IB) (v), 3-carbethoxypsoralen (3-CPs) (El) and 8methoxypsoralen (8-MOP) ( O ) at 5 ~M and 365-nm irradiation. Standard errors of the mean are indicated.

IA and IB are not only more active than the monofunctional compound 3-CPs but also more effective than the bifunctional furocoumarin 8MOP. In other words, for equal cell killing less UVA is needed with compounds IA and IB than with 8-MOP or 3-CPs. Since the molar extinction coefficients of compounds IA and IB at 365 nm are about 1.3-2 times lower, respectively, than that of 8-MOP and about 10 times lower than that of 3-CPs (Blais et al., 1984), these data suggest that these pyridopsoralens IA and IB are more biologically photoreactive than 8-MOP and 3-CPs. Table 1 gives the survival parameters derived from the survival curves defining the extent of the shoulder (Dq = quasi threshold dose), the dose for 37% survival ( 0 3 7 ) and the slope of the exponential part (Do). Exponential survival curves were obtained for compounds IA, IB and 3-CPs. The survival curve for 8-MOP was shouldered and of the multi-hit type. The resistant tail in the survival curves probably corresponded to the fraction of budding cells (late S o r G 2 phase) in the treated

cell population as shown previously (Henriques et al., 1977). Fig. 3 shows the results obtained on the induction of cytoplasmic 'petite' mutations in the haploid wild-type strain N123. As a function of unit dose of UVA (Fig. 3a), the two pyridopsoralens IA and IB demonstrate clearly a higher inducing capacity than 3-CPs and 8-MOP. In order to photoinduce 25% complete (unsectored) 'petite' mutants in the surviving population, only 0.5 and 1.2 k J/m= of UVA are needed with compounds IA and IB, respectively, whereas with the furocoumarins, 3-CPs and 8-MOP, doses of UVA as high as 10 and about 20 (extrapolated) k J / m 2 are needed. These results not only indicate that the compounds act on intracellular organelles (mitochondria) but also favour the idea, in analogy to previous results obtained with 3-CPs (Juliani et al.,

75[

75 L f.!

/ /I

. " tUosex lO

o. I,.'t 0 ~

I

~jj" .

_,.~

"1

a.~oP_j~ .

0 5 tO r5 rOSE O# 355 NM RADI~TION(kJm-2}

/~ tO

8-Mop

OI 00! ~IRVIVING FRAC~'ION

0(901

Fig. 3, Induction of cytoplasmic "petite' mutations (non-sectored colonies) in the haploid wild-type strain N123 of yeast after treatment with pyridopsoralens IA (,',) and IB (v), 3-CPs ([3) and 8-MOP (©) at 5 ~tM as a function of 365-nm irradiation dose (a) and of survival (b). Standard errors of the mean are indicated,

51 1976; Fukuhara et al., 1978; Averbeck et al., 1981), that the pyridopsoralens may possess a high affinity to mitochondrial DNA. At equal survival level, the two pyridopsoralens IA and IB appear to be less efficient than the monofunctional furocoumarin, 3-CPs, but much more efficient than the bifunctional furocoumarin, 8-MOP. Thus with regard to the capacity to induce cytoplasmic 'petite' mutations as a function of cell survival, compounds IA and IB clearly behave as monofunctional furocoumarins such as 3-CPs, angelicin and 4,5'-dimethylangelicin (Averbeck et al., 1979, 1981). Induction o f nuclear mutations in haploid yeast

Fig. 4 shows the results obtained in the haploid wild-type strain N123 of Saccharomyces cerevisiae on the induction of H I S + revertants after treatment with compounds IA and IB, 8-MOP and 3-CPs and UVA. Since for the assessment of nuclear mutations higher cell concentrations were treated, the corresponding survival curves are included although there is little difference with those presented in Fig. 2. As a function of UVA dose,

.

0

2

3

4

.

5

.

.

6

~&~%,,

7

8

9

the two pyridopsoralens are less effective than the bifunctional furocoumarin 8-MOP but more effective than the monofunctional furocoumarin 3-CPs. In other words, the order of activity for the induction of revertants, i.e. 3-CPs < IA < IB < 8-MOP (Fig. 4) is different from that observed for cell killing, i.e. 3-CPs < 8-MOP < IA < IB (Fig. 2). When the data are plotted as a function of survival (Fig. 4b) the two pyridopsoralens IA and IB are much less efficient than 8-MOP and even less efficient than 3-CPs. This may indicate that in comparison to 3-CPs the lesions induced by compounds IA and IB give rise preferentially to lethality and not to mutations, in other words, the lesions have a more lethal than mutagenic potential. As far as the induction of can R forward mutations in the same haploid wild-type strain is concerned, the photomutagenic capacity of the two pyridopsoralens is very much reduced in comparison to the bifunctional compound 8-MOP and hardly descernible above the spontaneous background level (Fig. 5 and Table 2). The capacity of 3-CPs to photoinduce (Fig. 5a) this type of forward mutation is very much limited but rises significantly above the background level at high doses of UVA (above 18 kJ/m2). As a function of survival (Fig. 5b) the mutagenicity of the monofunctional compound 3-CPs is already markedly reduced in comparison to that of the bifunctional compound 8-MOP, however, the mutagenic activity of the two pyridopsoralens is 2OO

J

~o-I

/'

. +-- ....

~ , ...."

-~-~-

20O

~o 8 MOP

tSO

tSO

f~ OOt DOSE OF 365 N/V} RADIA r/ON ( k J m -~:)

SURVIVING FRACTION

IA

Fig. 4. Induction of nuclear reversions ( h i s - --, H I S ÷ ) in the haploid yeast strain N123 after treatment with pyridopsoralens IA (A) and IB (v), 3-CPs (B), and 8-MOP (e) at 5 # M as a function of 365-nm irradiation (a) and of survival (b). Corresponding survival curves are included (upper part of a). The spontaneous mutation frequency at zero dose is subtracted in treated samples.

.

.

.

.

.

I 2 3 4 5 6 7 8 9 ~905£'0F 3 6 5 N M R~OlATlONtkJm-2)

~18 tO tO Ot SURVIVING FRACTION

OOt

Fig. 5. Induction of c a n R forward mutations in the haploid yeast strain N123 after treatment with pyridopsoralens IA (zx) and IB (v), 3-CPs (O), and 8-MOP ( O ) as a function of 365-nm irradiation (a) and of survival (b). The spontaneous mutation frequency at zero dose is subtracted in treated samples.

52 TABLE 2 INDU(Si-ION OF REVERSE AND FORWARD MUTANTS BY FUROCOUMARINS AND 365-nm IRRADIATION IN THE HAPLOID STRAIN N123 OF Saccharomyces cereoisiae Treat ment

8-MOP (5 ~M) -+ + + + + + +

Frequency of HIS' / 108 survivors a

Frequency of c a n R/ 106 survivors ~

1.13 (3) 2.6 (11) 30.6 (78) 87.7 (174) 234.8 (696) 513.1 (848) 1026.1 (458) 1339.3 (261)

5.5 (141) 8.9 (236) 7.6 (265) 16.7 (340) 31.3 (798) 72.9 (968) 128.1 (964) 162.6 (576)

365-nmradiation (M/m:) 0.6 1.2 1.8 3.6 5.4 7.2

3-CPs(5 ~M) + + + + + + + +

(7) (60) (60) (101) (84) (155) (123) (125) (150)

6.3 (160) 14.4 (199) 12.7 (187) 22.3 (211) 28.5 (206) 25.1 (317) 36.1 (497) 41.1 (459) 37.7 (398)

0.2 0.4 0.6 1.0 1.2

1.68 (7) 0.82 (6) 5.5 (11) 7.0 (10) 14.9 (4) 20.0 (5)

4.8 (142) 2.5 (162) 7.2 (126) 10.6 (234) 10.6 (249) 3.2 (78)

0.3 0.6 0.9 1.2 1.8 2.4 3.6

2.6 1.6 3.2 4.2 9.4 15.0 37.2 28.1

9.9 (264) 6.5 (230) 4.3 (234) 6.9 (202) 10.8 (506) 17.9 (442) 12.9 (381) 7.0 (219)

6 12 18 24 36 72 108 144

1.8 44 43 105 98.3 146.5 146.6 163.2 240

Compound IB (5 ~M) + + + + + +

Compound IA ¢5 ~M) + + + + + + + +

(11) (13) (15) (12) (13) (8) (10) (6)

The spontaneous background has been subtracted for the frequency of induced mutants. The number of actual colonies counted is given in brackets. a The values are the means of 2 Expts.

hardly significant. This again agrees with the notion that the pyrido[3,4-c]psoralens photoinduce l e s i o n s w i t h a r e l a t i v e l y h i g h lethal a n d a low mutagenic potential.

Photosensitization in the presence and in the absence of oxygen A l t h o u g h the p h o t o r e a c t i o n s of f u r o c o u m a r i n s have generally been c o n s i d e r e d to be i n d e p e n d e n t o f o x y g e n ( M u s a j o a n d R o d i g h i e r o , 1972; S c o t t et

53

al., 1976), recent findings on the capacity of furocoumarins to generate singlet oxygen (Song and Tapley, 1979; Cannistraro and Van de Vorst, 1977; De Mol and Beijersbergen van Henegouwen, 1979; Averbeck, 1982; Averbeck et al., 1981, 1983a, b; Ronfard-Haret et al., 1982) indicate that photodynamic (oxygen-dependent) processes may contribute to the photobiological and therapeutic effects of furocoumarins (Averbeck et al., 1981, 1983a, b; Averbeck, 1982; Ronfard-Haret et al., 1982; De Mol and Beijersbergen van Henegouwen, 1979; Granger et al., 1982; Dubertret et al., 1981). Thus, it seemed important to determine the effects of the new pyridopsoralens on survival and cytoplasmic 'petite' induction in the presence and absence of oxygen. Fig. 6 shows that compounds IA and IB do not exhibit a significant oxygen effect on cell killing in the haploid wild-type strain N123 of Saccharomyces cerevisiae. Nor was there any oxygen effect for the induction of cytoplasmic 'petite' mutations (data not shown). The finding that the new pyridopsoralens, as 8-MOP (Averbeck, 1982), do not exert an oxygen-dependent (photodynamic) effect in yeast cells is in favour of a good photostability of the compounds and diminishes the risk of acute phototoxicity during photochemotherapeutic use. One may recall at this point that an oxygen effect was demonstrated for

z

70~

s/,M %

~001

00010

a.. ~

1 2 3 ~ 5 6 7 8 9 DOSE Of 365 NM RADIAYION (Rim-2)

Fig. 6. Typical survival curves obtained for the haploid yeast strain N123 after treatment with pyridopsoralens IA and IB at 5 ~aM and 365-nm irradiation in the presence and absence of oxygen. - - , air (Air liquide, France) bubbled samples (closed symbols); , pure nitrogen (99.999% N2, Air liquide France) bubbled samples (open symbols).

3-CPs (Averbeck, 1982; Ronfard-Haret et al., 1982; Averbeck et al., 1983b). Discussion

In the present work, the biological effects of two new monofunctional pyridopsoralens were compared to those of the known monofunctional furocoumarin, 3-carbethoxypsoralen (3-CPs), and of the bifunctional furocoumarin, 8-methoxypsoralen (8-MOP), using a haploid wild-type strain of yeast (Saccharomyces cerevisiae). As previously shown for 3-CPs and 8-MOP (Averbeck et al., 1981; Averbeck and Moustacchi, 1980), pyrido[3,4-c]psoralen (IA) and the pyrido[3,4-c]-7-methylpsoralen (IB) also did not exhibit a significant activity on survival and mutation induction in the dark. However, it should be noted that 8-MOP and compounds IA and IB are able to induce frameshift mutations in Salmonella typhimurium (Ames test) in the dark (Quinto et al., 1984); whereas for the induction of sister-chromatid exchanges, the dark effects of 8-MOP and compound IA were negligible and only slight effects were seen for 3-CPs and compound IB (Billardon et al., 1984). The results obtained here on the induction of lethal effects and cytoplasmic 'petite' mutations in haploid yeast by furocoumarins and 365-nm irradiation (UVA) demonstrate that the pyridopsoralens are photoreactive substances. Their capacity to photoinduce lethality is not only higher than that of the monofunctional furocoumarin, 3-CPs, but also higher than that of the bifunctional furocoumarin, 8-MOP. The order of activity was: compound IB > compound IA > 8-MOP > 3CPs in yeast and similar to that observed in Chinese hamster cells (Papadopoulo et al., 1984). Since the killing activity of the pyridopsoralens was not oxygen dependent, these compounds (IA and IB) clearly exert their photobiological effects without the involvement of photodynamic reactions which would imply phototoxic and membrane-damaging effects. Thus, compounds IA and IB are likely to react as most of the known furocoumarins (Song and Tapley, 1979) mainly via the photoaddition to pyrimidines in DNA. The results obtained on the induction of cytoplasmic 'petite' mutations are in favour of photo-

54

reactions involving only the production of monoadditions in DNA. This is supported by previous findings showing that in contrast to bifunctional furocoumarins such as psoralen, 8-MOP, 5-MOP and 4,5',8-trimethyipsoralen, monofunctional furocoumarins like angelicin, 4,5'-dimethylangelicin, 3-CPs photoinduce per lethal hit a high level of cytoplasmic 'petite' mutations involving damage to mitochondrial DNA (Averbeck et al., 1981; Averbeck, 1982). Furthermore, the induction of cytoplasmic 'petite' mutations was very pronounced in situations in which the presence ef DNA cross-links is disfavoured, for example, when using bifunctional furocoumarins such as 8-MOP in combination with low-dose-rate 365-nm irradiation (Averbeck and Averbeck, 1978) or near-UV irradiation at wavelengths above 365 nm (Averbeck, in preparation). Knowing that the repair of damage via excision repair is non-existent for lesions induced in mitochondrial DNA (Waters and Moustacchi, 1974; Magana-Schwencke et al., 1982), the accumulation of damage in mitochondrial DNA as a function of UVA dose after treatments with furocoumarins can be related to the photoreactivity of the compounds, as shown previously for 3-CPs and 4,5'-dimethylangelicin (Averbeck et ai., 1981). Therefore, the high capacity of the new pyridopsoralens to photoinduce cytoplasmic 'petite' mutations per unit dose of UVA in comparison to 8-MOP and 3-CPs suggests a high photoaffinity of these compounds towards mitochondrial DNA, Indeed, recent studies in vitro (Blais et al., 1984) and in vivo (Moustacchi et al., 1983: Averbeck and Averbeck, 1984; Nocentini, 1984; Papadopoulo et al., in preparation) have not only demonstrated a monofunctional mode of reaction of the pyridopsoralen IB but also a photoaffinity towards DNA higher than that of 8-MOP and 3-CPs, Cellular repair systems may be overwhelmed by the high number of photoadducts induced by this pyridopsoralen resulting in high cell killing. Thus, the strong photoaffinity of this compound may explain at least in part its strong lethal effect and antiproliferative activity. In addition, the type of lesion, i.e. the particular monoadduct formed, may be of crucial importance. Recent analysis of the photoproducts formed by different psoralens suggests that, depending on their struc-

ture, furocoumarins favor the formation of particular psoralen-pyrimidine isomers and their relative proportions may be modified (DalrAcqua et al., 1983a, b; Cadet et al., 1983; Kanne et al., 1982). Biochemical studies on the fate of the photoadducts formed by the pyridopsoralen IB (Moustacchi et al., 1983) indicate that wild-type cells of yeast are able to remove most of the photoinduced lesions in 4 h. However, during the repair of IB-induced lesions, single-strand breaks are produced which remain unsealed during 6 h of posttreatment incubation. In contrast, at equal doses, those occurring during the repair of 8-MOP- or 3-CPs-induced lesions are repaired within 2 h (Moustacchi et al., 1983). The difficulty in rejoining breaks occurring during the repair of IB-induced monoadducts in DNA may well be due to the particular nature of the adduct formed, as well as to their high proportion. Since the biological activity of the pyridopsoralen IA is in many respects comparable to that of the pyridopsoralen IB, we assume that compound IA also exhibits a high photoreactivity towards DNA. Therefore, the predictions made on structure-activity relationships that lead to the synthesis of the compounds appear to be fulfilled. The example of the pyridopsoralens demonstrates that monofunctional furocoumarins may differ enormously in their photoreactivity and photobiological activity. In contrast to 3-CPs, the new pyridopsoralens are clearly more active than the bifunctional furocoumarin 8-MOP and show a similar range of activity as some of the new methylated angelicin derivatives proposed for photochemotherapeutic use (Dall'Acqua et al., 1983a, b; Recchia et al., 1983). From the fact that the two pyridopsoralens show a relatively low mutagenicity in haploid yeast (nuclear reversions and forward mutations) and a high antiproliferative effect compared to 8-MOP, it appears that the lesions photoinduced by these two compounds are highly lethal but only slightly mutagenic. However, in diploid yeast (Averbeck and Averbeck, 1984), in Chinese hamster cells (Papadopoulo et al., 1984) and in human skin fibroblasts (Billardon et al., 1984) the genotoxicity of the two pyridopsoralens was not negligible. This difference in response between haploid and diploid cells will be discussed elsewhere.

55

Recent findings on the increased activity of pyridopsoralens in the PUVA (psoralen plus UVA) treatment of psoriatic lesions (Averbeck et al., 1982, 1983b; Dubertret et al., 1984) and an apparently reduced carcinogenicity in mice (Zajdela, personal communication; Dubertret et al., 1984) in comparison with 8-MOP make these compounds quite promising for photochemotherapeutic use although further investigations are needed to define in more detail the relationship between their photobiological reactivity, their genotoxicity, their general toxicity and their photochemotherapeutic effectiveness.

Acknowledgements This work was supported by the research contracts PRC M6dicament 121001 and 832001 of the Institut National de la Sant6 et de la Recherche M6dicale, the R.C.P. 080572 of the Centre National de la Recherche Scientifique and by the Commissariat :h l'Energie Atomique (Saclay, France). We thank Drs. E. Moustacchi and R. Latarjet for their interest in our work.

References Abel, G., and O. Schimmer (1981) Mutagenicity and toxicity of furocoumarins: comparative investigations in 2 test systems, Mutation Res., 90, 451-461. Anderson, T.F., and J.J. Voorhees (1980) Psoralen photochemotherapy of cutaneous disorders, Annu. Rev. Pharmacol. Toxicol., 20, 235-257. Averbeck, D. (1982) Photobiology of furocoumarins, in: C. Helene, M. Charlier, Th. Montenay-Garestier and G. Laustriat (Eds.), Trends in Photobiology, Plenum, New york, pp. 295-308. Averbeck, D., and S. Averbeck (1978) Dose-rate effects of 8-methoxypsoralen plus 365-nm irradiation on cell killing in Saccharomyces cereuisiae, Mutation Res., 50, 195-206. Averbeck, D., and S. Averbeck (1984) Relationship between DNA damage and genetic effects induced by photoaddition of furocoumarins in diploid yeast (Saccharomyces cerevisiae), Photochem. Photobiol., 39, Suppl., 37 S, Abstr. TAM-G1. Averbeck', D., and E. Moustacchi (1979) Genetic effect of 3-carbethoxypsoralen, angelicin, psoralen and 8-methoxypsoralen plus 365-nm irradiation in Saccharomyces cerevisiae, Induction of reversions, mitotic crossing-over, gene conversion and cytoplasmic 'petite' mutations, Mutation Res., 68, 133-148.

Averbeck, D., and E. Moustacchi (1980) Decreased photo-induced mutagenicity of monofunctional as opposed to hifunctional furocoumarins in yeast, Photochem. Photobiol., 31,475-478. Averbeck, D., P. Chandra and R.K. Biswas (1975) Structural specificity in the lethal and mutagenic activity of furocoumarins in yeast cells, Radiat. Environ. Biophys., 12, 241-252. Averbeck, D., E. Moustacchi and E. Bisagni (1978) Biological effects and repair of damage photoinduced by a derivative of psoralen substituted at the 3,4 reaction site, Photoreactivity of this compound and lethal effect in yeast, Biochim. Biophys. Acta, 518, 464-481. Averbeck, D., E. Bisagni, J.P. Marquet. P. Vigny and F. Gaboriau (1979) Photobiological activity in yeast of derivatives of psoralen substituted at the 3,4 and/or the 4',5' reaction site, Photochem. Photobiol., 30, 547-555. Averbeck, D., S. Averbeck and F. DalrAcqua (1981) Mutagenic activity of three monofunctional and three bifunctional furocoumarins in yeast (Saccharomyces cerevisiae), Il Farmaco, 36, 492-505. Averbeck, D., L. Dubertret, E. Bisagni and J. Moron (1982) Towards more safety in PUVA-therapy with new monofunctional psoralens, Int. Workshops on Investigative Dermatology, Kyoto, May 31-June 1, Abstr. W2-18. Averbeck, D., L. Dubertret, M. Craw, T.G. Truscott, F. Dall'Acqua, P. Rodighiero, D. Vedaldi and E.J. Land (1983a) Photophysical, photochemical and photobiological studies of 4'-methylangelicins, potential agents for photochemotherapy, I1 Farmaco, Ed. Sci. 39, 57-69. Averbeck, D., U Dubertret, E. Bisagni, J. Moron, D. Papadopoulo, S. Nocentini, J. Blais and F. Zajdela (1983b) Photobiological and phototherapeutic properties of new monofunctional pyridopsoralens, Abstract of Joint Meeting Society Investigative Dermatol. Research, April 27-May 1, Washington, DC, J. Invest. Dermatol., 80, 306. Averbeck, D., D. Papadopoulo and i. Quinto (1984) Mutagenic effects of psoralens in yeast and V79 Chinese hamster cells, J. Natl. Cancer Inst., in press. Billardon, C., S. Levy and E. Moustacchi (1984) Induction in human skin fibroblasts of sister-chromatid exchanges (SCE) by photoaddition of two new monofunctional pyridopsoralens in comparison to 3-carbethoxypsoralen and 8-methoxypsoralen, Mutation Res., 138, 63-70. Blais, J., P. Vigny, J. Moron and E. Bisagni (1984) Spectroscopic properties and photoreactivity with DNA of new monofunctional psoralens, the pyridopsoralens, Photochem. Photobiol., 39, 145-156. Bordin, F., F. Carlassare, F. Baccichetti, A. Guiotto, P. Rodighiero, D. Vedaldi and F. DalrAcqua (1979) 4,5'-Dimetbylangelicin: a new DNA photobinding monofunctional agent, Photochem. Pbotobiol., 29, 1063-1070. Bordin, F., Baccichetti, F. Carlassare, M. Perin, F. DalrAcqua, D. Vedaldi, A. Guiotto and P. Rodighiero (1981) Pre--clinicat evaluation of new anti-proliferative agents for the photochemistry of psoriasis: angelicin derivatives, II Farmaco, 36, 506-518.

56 Bridges, B.A. (1979) An estimate of genetic risk from 8methoxypsoralen photochemotherapy, Hum. Genet., 49, 91-96. Burger, P.M., and J.W.I.M. Simons (1979a) Mutagenicity of 8-methoxypsoralen and long-wave ultraviolet irradiation in V-79 Chinese hamster cells, A first approach to a risk estimate in photochemotherapy, Mutation Res., 60, 381-389. Burger, P.M., and J.W.I.M. Simons (1979b) Mutagenicity of 8-methoxypsoralen and long-wave ultraviolet irradiation in diploid human skin fibroblasts, An improved risk estimate in photochemotherapy, Mutation Res., 63, 371-380. Cadet, J., L. Voituriez, F. Gaboriau. P. Vigny and S. Della Negra (1983) Characterization of photocycloaddition products from reaction between thymine and the monofunctional 3-carbethoxypsoralen, Photochem. Photobiol., 37, 363-371. Cannistraro, S., and A. Van de Vorst (1977) ESR and optical absorption evidence for free radical involvement in the photo~nsitizing action of furocoumarin derivatives and for their singlet oxygen production, Biochim Biophys. Acta, 476, 166-167. Cartwright, L.E., and J.E. Walter (1983) Psoralen-containing s u n , t e e n is tumorigenic in hairless mice, J. Am. Acad. Dermatol., 8, 803-836. Coppey, J., D. Averbeck an G. Moreno (1979) Herpes virus production in monkey kidney and human skin cells treated with angelicin or 8-methoxypsoralen plus 365 nm light, Photochem. Photobiol., 29, 797-801. Dall'Acqua, F. (1977) New chemical aspects of the photoreaction between psoralen and DNA, in: A. Castellani (Ed.), Research in Photobiology, Plenum, New York, pp. 245-255. DalrAcqua, F., S. Caffieri, D. Vedaldi, A. Guiotto and P. Rodighiero (1983a) Monofunctional 3,4- and 4',5'-photocycloadducts between 4,5'-dimethylangelicin and thymine, Photochem. Photobiol., 37, 373-379. Dall'Acqua, F., D. Vedaldi, F. Bordin, F. Baccichetti, F. Carlassare, M. Tamaro, P. Rodighiero, G. Pastorini, A. Guiotto, G. Recchia and M. Cristofolini (1983b) 4'-Methylangelicins: new potential agents for the photochemotherapy of psoriasis, J. Med. Chem., 26, 870-876. De Mol, N.J., and G.M.J. Beijersbergen Van Henegouwen (1979) Formation of singlet molecular oxygen by 8-methoxypsoralen, Photochem. Photobiol., 30, 331-335. Dubertret, L., D. Averbeck, F. Zajdela. E. Bisagni, E. Moustacchi. R. Touraine and R. Latarjet (1979) Photochemotherapy [PUVA) of psoriasis using 3-carbethoxypsoralen, a compound non carcinogenic in mice, Br. J. Dermatol., 101, 379-389. Dubertret, L., D. Averbeck, R. Bensasson, E. Bisagni, F. Gaboriau, E.J. Land, S. Nocentini, M.T. Macedo Sa E Melo, E. Moustacchi, P. Morli6re, J.C. Ronfard-Haret, R. Santus, P. Vigny, F. Zajdela and R. Latarjet (1981) Photophysical photochemical, photobiological and phototherapeutic properties of 3-carbethoxypsoralen, in: G. Cahn, B.P. Forlot, C. Grupper, A.E. Meybeck and F. Urbach (Eds.), Psoralens in Cosmetics and Dermatology, Pergamon, New York, pp. 245-256.

Dubertret, L., D. Averbeck, E. Bi~gni, J. Moron, E. Moustacchi, C. Billardon, P. Papadopoulo, S, Nocentini, J. Blais and F. Zajdela (1984) Photobiological and phototherapeutic properties of new monofunctional pyridopsoralens, Photochem. Photobiol., 39, Suppl., 60 S, Abstr. WAM-D5. Fukuhara, H., E. Moustacchi and M. Wesolowski (1978) Preferential deletion of a specific region of mitochondrial DNA in Saccharomyces cereeisiae by ethidium bromide and 3carbethoxypsoralen, Mol. Gen. Genet., 162, 191-201. Granger, M., F. Toulme and C. Helene (1982) Photodynamic inhibition of Escherichia coil DNA polymerase I by 8methoxypsoralen plus near ultraviolet irradiation, Photochem. Photobiol.. 36. 175-180. Grant, E.L., R.C. Von Borstel and M.J. Ashwood-Smith (1979) Mutagenicity of cross-links and monoadducts of furocoumarins (psoralen and angelicin) induced by 360 nm radiation in excision-repair defective and radiation insensitive strains of Saccharomyces ceremsiae, Environ. Mutagen., 1, 55-63. Grekin, D.A., and J.H. Epstein (1981) Psoralens, UVA (PUVA) and photocarcinogenesis, Photochem. Photobiol., 33, 957-960. Grenson, M., M. Mousset, J.M. Wiame and J. Bechet (1966) Multiplicity of the amino-acid permeases in Saccharomyces ceret,isiae, 1. Evidence for a specific arginine-transporting system, Biochim, Biophys. Acta, 127, 325-338. Grossweiner, L.I., and K.C. Smith (1981) Sensitivity of DNA repair-deficient strains of E.wherichia coli K-12 to various furocoumarins and near-ultraviolet radiation, Photochem Photobiol., 33, 317-323. Henriques, J.A.P., R. Chanet, D. Averbeck and E. Moustacchi (1977) Lethality and "petite' mutation induced by the photoaddition of 8-methoxypsoralen in yeast, Mol. Gen. Genet., 158, 68-73. Juliani, M.H., S. Hixon and E. Moustacchi (1976) Mitochondrial damage induced in yeast by a photoactivated furocoumarin in combination with ethidium bromide or ultraviolet light, Mol. Gen. Genet., 145, 249-254. Kanne, D., K. Straub, M. Rapoport and J.E. Hearst (1982) Psoralen-deoxyribonucleic acid photoreaction, Characterization of the monoaddition products from 8-methoxypsoralen and 4,5',8-trimethylpsoralen, Biochemistry, 21, 861 871. Kimura, S., and N. Mizuno (1981) Topical PUVA with 3carbethoxypsoralen for the treatment of psoriasis, in: Proc. 2nd Japan-Korea, J. Meeting Dermatology. Nov. 21-22 1981, Fukuoka, Japan. Kimura, S., S. Hirano, K. Yoshikawa and N. Mizuno (1981) Photoallergy to 3-carbethoxypsoralen, Photomed. Photobiol., 3, 81-82. Magana-Schwencke, N., J.A.P. Henriques. R. Chanet and E. Moustacchi (1982) The fate of 8-methoxypsoralen photoinduced cross-links in nuclear and mitochondrial yeast DNA: comparison of wild type and repair deficient strains, Proc. Natl. Acad. Sci. (U.S.A.), 79, 1722-1726. Moron, J., C.H. Nguyen and E, Bisagni (1983) Synthesis of 5 H-furo( 3',2' : 6,7)( l )benzopyrano{ 3,4-c]pyridine,5-ones and 8H-pyrano(3',2': 5,6)benzo-furo{3,2-c]pyridin-8-ones (pyridopsoralens), J. Chem. Soc. Perkin Trans., 1,225-229.

57 Moustacchi, E. (1969) Cytoplasmic and nuclear genetic events induced by UV light in strains of Saccharomyces cereuisiae with different UV sensitivities, Mutation Res., 7, 171-185. Moustacchi, E., C. Cassier, R. Chanet, N. Magana-Schwencke, T. Saeki and J.A.P. Henriques (1983) Biological role of photo-induced crosslinks and monoadducts in yeast DNA: genetic control and steps involved in their repair, in: E.C. Friedberg and B.A. Bridges (Eds.), Cellular Responses to DNA Damage, Alan Liss. New York, pp. 87-106. Musajo, L., and G. Rodighiero (1972) Mode of photosensitizing action of furocoumarins, in: A. Giese (Ed.), Photophysiology. Vol. VII, Academic Press, New York, pp. 115-147. Nocentini. S. (1984) DNA photobinding of 7-methyl[3,4c]psoralen and 8-methoxypsoralen, removal of lesions and biological effects in cultured human and simian cells, Photochem. Photobiol., 39, Suppl., 55 S, Abstr. TPM-C16. Ogur, M., S. John and S. Nagai (1957) Tetrazolium overlay technic for population studies of respiration deficiency in yeast, Science, 125. 928-929. Pani, B.N., Babudri, S. Venturini, M. Tamaro, F. Bordin and C. Monti-Bragadin ( 1981 ) Mutation induction and killing of prokaryotic and eukaryotic cells by 8-methoxypsoralen, 4,5-dimethylangelicin, 5-methylangelicin, 4'-hydroxymethyl4,5-(-dimetyl)-angelicin, Teratogen. Carcinogen. Mutagen.. 1,407-415. Papadopoulo, D.. F. Sagliocco and D. Averbeck (1983) Mutagenic effects of 3-carbethoxypsoralen and 8-methoxypsoralen plus 365-nm irradiation in mammalian cells, Mutation Res., 124, 287-297. Papadopoulo, D.. D. Averbeck, J. Moron and E. Bisagni (1984) Genotoxic effects photoinduced in mammalian cells in vitro by two new synthesized monofunctional furocoumarins, Photochem. Photobiol., 39, Suppl., 54 S, Abstr. TPM-CI2. Parsons, R.J. (1980) Psoralen photochemistry, Photochem. Photobiol., 32, 813-821. Pathak, M.A., D.M. Kr~trner and T.B. Fitzpatrick (1974) Photobiology and photochemistry of furocoumarins (psoralens), in: M.A. Pathak. L.C. Harber. M. Seiji, A. Kukita (Eds.), Sunlight and Man, University of Tokyo Press, Tokyo, Ch. 22, pp. 335-368. Pathak. M.A., J.A. Parrish and T.B. Fitzpatrick (1981) Psoralens in photochemotherapy of skin diseases, I1 Farmaco, 36, 479-491. Queval, P., and E. Bisagni (1974) New synthesis of psoralen and related compounds, Eur. J. Med. Chem., 9, 335-340. Quinto, I., D. Averbeck, E. Moustacchi, Z. Hrisoho and J. Moron (1984) Frameshift mutagenicity in Salmonella (vphimurium of furocoumarins in the dark, Mutation Res., 136, 49-54. Recchia, G., M. Cristofolini, F. Bordin, F. DalrAcqua and P. Rodighiero (1983) Methylangelicins in the topical photo-

chemotherapy of psoriasis: preliminary report, in: G.B. Galioto, F. Dall'Acqua and L. Santamaria (Eds.), Medicine, Biology, Environment, Bull. Eur. Inst. Ecology and Cancer (INEC), Vol. 11, pp. 470-473. Ronfard-Haret, J.C.. D. Averbeck, R.V. Bensasson, E. Bisagni and E.J. Land (1982) Some properties of the triplet excited state of the photosensitizing furocoumarins: 3-carbethoxypsoralen, Photochem. Photobiol., 35, 479-489. Scott, B.R., M.A. Pathak and G.R Mohn (1976) Molecular and genetic basis of furocoumarin reactions, Mutation Res., 39. 29-74. Song. P.S.. and K,J. Tapley Jr. (1979) Photochemistry and photobiology of psoralens, Photochem. Photobiol., 29, 1177-1197. Stern, R.S., L.A. Thibodeau, R.A. Kleinerman, J.A. Parrish, T.B. Fitzpatrick and 22 investigators (1979) Risk of cutaneous carcinoma in patients treated with oral methoxsalen photochemotherapy for psoriasis, New Engl. J. Med., 300, 809-813. Swart. R.N.J., M A N . Beckers and A.A. Schothorst (1983) Phototoxicity and mutagenicity of 4,5'-dimethylangelicin and long wave ultraviolet irradiation in Chinese hamster cells and human skin fibroblasts, Mutation Res., 124, 271-279. Tourbez-Perrin, M., F. Pochon, C. Ducrocq, C. Rivalle and . Bisagni (1980) Intercalative binding to DNA of new antitumoral agents: dipyridol4,3-b][3.4-f]indoles, Bull. Cancer (Paris), 67, 9-13. Vedaldi, D., F. Dall'Acqua, A. Gennaro and G. Rodighiero (1983) Photosensitized effects of furocoumarins: the possible role of singlet oxygen, Z. Naturforsch., 38c, 866-869. Venturini, S., M. Tamaro, C. Monti-Bragadin, F. Bordin, F. Bacchichetti and F. Carlassare (1980) Comparative mutagenicity of linear and angular furocoumarins in Escherichia coli strains deficient in known repair functions, Chem.-Biol. Interact., 30, 203-207. Waters, R., and E. Moustacchi (1974) The fate of UV-induced pyrimidine dimers in the mitochondrial DNA of Saccharomyces cerevisiae following various post-irradiation cell treatments, Biochim. Biophys. Acta, 366, 241-250. West, J.D., M.P. Mullen and M.A. Pathak (1982) Photosensitivity and carcinogenic potential of monofunctional and bifunctional psoralens, J. Invest. Dermatol., 78, 361. Young, A.R., I.A. Magnus, A.C. Davies and N.P. Smith (1983) A comparison of the phototumorigenic potential of 8-MOP and 5-MOP in hairless albino-mice exposed to solar simulated radiation, Br. J. Dermatol., 108, 507-518. Zajdela, F., and E. Bisagni (1981) 5-Methoxypsoralen, the melanogenic additive in sun-tan preparations is tumorigenic in mice exposed to 365 nm UV radiation. Carcinogenesis, 2, 121-127.