Synthesis, In vitro Antifungal and Antitumour Activity of Some Triorganotin(IV) N,C,N-Chelates

The in vitro antifungal activity of compounds 1-3 ( { [ ( CH 3 ) 2 NCH 2 ] 2 C 6 H 3 } R 2 SnX ; (where X=Cl, R=n-Bu for 1, X=Br, R=n-Bu for 2 and x= PF 6 , R=n=Bu for 3)) was estimated with the help of a modified microdilution format of the M27-A guidelines and was compared with in vitro activity of their diphenyltin(IV) analogues 4 and 5 (where X=Br, R=Ph for 4 and X= PF 6 , R=Ph for 5), and of drugs currently in clinical use (ketoconazole, fluconazole and amphotericin B). It was found that in coordinating solvents the more soluble derivative 2 is less active than the phenyl one (4), and compounds 1 and 3 are even inactive. In this paper, the in vitro antitumour activity of ionic diphenyltin(IV) complexes 4 and 5 against seven tumoural cell lines of human origin is also reported. The preparation and characterization ( H 1 , C 13 and Sn 119 NMR spectroscopy and electrospray ionization mass spectrometry) of the novel compound 3 is mentioned too.


INTRODUCTION
Organotin(IV) compounds, especially triorganotin derivatives, have been extensively studied due to their potential in vitro antifungal activities against some medically important fungi/1/. Many compounds of this type display some in vitro antitumour activities against tumour cell lines of human origin/2,3/. Moreover, recently, Susperregui et al. have reported on in vivo trypanocidal activities of organotin(IV) compounds/4/.
In 1985 Atassi suggested that such in vivo antitumour activity and further clinical use of these compounds in medicine could be hampered by low water solubility/5/. A possibility to increase the water solubility of organotin compounds is to prepare compounds containing some heteroatoms in their structure/6/, or even ionic compounds. We have previously reported on such ionic compounds (water solubility ca. 200 rag/100 ml/room temperature) and their in vitro antifungal activities. Some of these organotin derivatives display MIC values (minimum inhibitory concentration tmol.I ") comparable to those for currently used drugs/7/. I'oL 9, Nos. 1-2. 2002 Synthesis, in vitro Anti./imgal and Antitmnom* ActiviO, ?['Some Triorganotin(l l')N, C,N-Chelates We now present the results of in vitro antitumour screening for two selected diphenyltin(IV) complexes 4 and 5 and also the in vitro antifungal activity of their dibutyltin(IV) analogues 1-3.

EXPERIMENTAL General comments and synthesis
All solvents were obtained frown commercial sources, dried by standard procedures and distilled prior to use, Compounds l, 2, 4 and 5 (see Fig. l

Spectra
The solution state llgsn, 3lp, 13 C and tH NMR spectra were acquired at 134.28, 145.79, 90.56 and 360.13 MHz respectively, on a Bruker AMX 360 NMR spectrometer, using a 5 mm tuneable broad band probe at 300 K. Appropriate chemical shifts were calibrated on: tH-residual peak of CHCI3 ( 7.25 ppm), t3C.-signal of CDCla (5 77.0 ppm), atP-external 85% H3PO4 (t5 0.0 ppm) and t9Sn-external tetramethylstannane (t5 Electrospray ionisation (ESI) mass spectra were measured on an ion trap analyser (Esquire 3000, Bruker Daltonics) and on a quadrupole analyser (Platform, Micromass). The samples were dissolved in acetonitrile and analysed by direct infusion at a flow rate of tl.min "t. Mass spectra weree recorded in the range m/z 15-800, in both negative-ion and positive-ion mode.
In vitro antifungai screening The in vitro testing was carried out using a modified microdilution broth of the M27-A guidelines (National Committee for Clinical Laboratory Standards. Reference method for broth dilution antifungal L. Dosla! el Metal-Based Drugs susceptibility testing of yeasts. Approved standard. Document M27-A, Wayne, PA: National Committee tbr Clinical Laboratory Standards, 1997). Quality control strains (Candida albicans ATCC 90028, Candida parapsilosis ATCC 22019, Candida krusei ATCC 6258) and amphotericin B, fluconazole (Pfizer), ketoconazole (Janssen-Cilag, Beerse) as a reference drug were involved. All fungal strains were passaged on Sabouraud dextrose agar at 35C prior to being tested.
The minimum inhibitory concentration (MIC) was determined by the following method: DMSO (dimethylsulfoxide) served as a diluent for all compounds tested. DMSO did not exceed a final concentration of 2%. RPMI 1640 (Sevapharma, Prague) medium, supplemented with L-glutamine and buffered with 0.165 M morpholinepropanesulfonic acid (Serva) to pH 7.0 (using l0 M NaOH), was used as a test medium. Each well of the microdilution tray was filled with 200 gl of the RPMI 1640 medium with a diluted compound, tested and then inoculated with 10 gl of suspension of a given fungal strain in sterile water. Fungal inoculum was prepared to give a final size of 5 x 103 +/-0.2 CFU ml"1. The trays were incubated at 35 C and MICs read after 24 and 48h. Owing to slow growth, the Trichophyton mentagrophytes strain was read at 72 and 120 h. The MICs were determined visually and defined as 80% inhibition of the growth of control.

In vitro antitumour screening
The protocol followed for the in vitro antitumour screenings has already been reported/9/.

RESULTS AND DISCUSSION
Characterization and structure Compounds 1, 2, 4 and 5 were previously investigated/7,8,10,1 l/by multinuclear NMR spectroscopy, electrospray ionisation mass spectrometry and X-ray diffraction techniques. The identity and purity of all. derivatives were verified by elemental analysis too.
R=nBu, X=CI 2 R=nBu, X=Br 3 R=nBu, X=PF 6 4 R=Ph, X=Br 5 R=Ph, X=PF 6 The compound 3 has been characterized with the help of NMR spectra parameters which correspond to appropriate ones in analogous above-mentioned compounds. One set of sharp signals was observed in the proton spectrum measured in CDCI3 at room temperature, indicating equivalency of both amine donor and butyl groups, respectively. The 9Sn chemical shift value (53.6 ppm) is comparable with previous results/8/ for ionic compounds (for example I and 2), from the range for five-coordinated (or better [3+2] tin atom in ionic butyltin compounds. The extent of interatomic C(Bu)-Sn-C(Bu) angle can be calculated/12/from the values of J(9Sn, 13C) coupling constant (413.4 Hz, 116.1). The symmetrical septet with lj(3p, 9F) 713.4 Hz in the 31p NMR spectrum shows that the PF6 group which does not interact with the tin atom is present.
On the basis of these findings, we can conclude that the vicinity of the tin central atom is slightly distorted trigonal bipyramidal, with donor amino groups in axial and carbon atoms in equatorial positions, forming the cationic unit. The hexafluorophosphate anion is out of the primary coordination sphere of the tin atom.

In vitro antifungal activity
The results of in vitro antifungal screening for all compounds 1-5 are summarized in Table together with MIC values (minimum inhibitory concentration tmol "l) for conventional antimycotic drugs. The MICs for derivatives 4,5 were reported earlier elsewhere/7/and are only used here for comparison of antimycotic activity of their di-n-butyl analogues 1-3. The surprising discovery is that the bromobutyl derivative (2), more soluble in coordinating solvents, is less active than the phenyl one, and compounds 1 and 3 are even inactive. In vitro antitumour activity Compounds 4 and 5 were chosen for in vitro antitumour screening against seven tumoural cell lines of human origin: MFC-7 and EVSA-T: two breast cancers; WiDr: a colon carcinoma; IGROV: an ovarian cancer; M19MEL: a melanoma; A 498: a renal cancer and H 226: a non small cell lung cancer. The results of in vitro tests, given in the form of IDs0 (the inhibition doses ng ml") for these two derivatives are shown in Table 2 and compared with IDso values of some drugs with clinical applications. In spite of the fact that both derivatives 4 and 5 display increased water solubility and promising in vitro antifungal activity/7/, Table 2 shows that both tested compounds are entirely inactive on all tumour cell lines. Table 2 in vitro inhibition doses IDs0 in ng. ml " for compounds 4  See Figure 1. IDa0 values taken from the literature, see/3/.