Metabolic Fate of Phenyltrimethylsilane and Phenyldiinethylsilancl
T h e iirinary metabolic ciid prodiicts of' :iroiii:itic compounds have been studied by :t uunibc~rof irivestig:ttoi,S I n general, these compound3 undergo hydrouylut ion either on the aromatic ring or on the alkyl substitiwilt (krmsne to this report is the ~ v o r kof ITilliam f:ite of t-butylbenzene4 and isopropylbenzenc. h i n g given orally to a r:rbbit, each of thesc compoundis hydroxylated on tlic alliyl portion of thc' iiiolrciilc and eliminated :is i~ gluciironide con,iiigntc or ns :I higher oxidized product. Wil1i:tms tlirl not I ch1)oi.t : I I I ~ :ironmtic riiig-hy(~rox\rI:~tc.d iwtLh)litt~-.6
CH
("
j
('H
("
CH
0
= glucuronide
Tlie title compounds n crc rhovri for itiitly becausv they contain two key structur:d feiLturex coinmoii t o many organosilicon CornpouiidS- t Iic silicori--phciiyl bond and the silicon-hydride borid. Iri iidditioii, hinw little work has been reported in the :trt':L of organohilicori metnbolism,7 these t n ci striictweb provide :I suitable starting point for future norl; 111 thc ire^ tho identification of t1i.I T h e g o d of the project major urinary metabolites of the title c ~ m p o u i i dI ~: i r d 11, n s well as the obtaining of d:tt:t concerning the d)sorption and eliniination of t h ( v c o i q m i i i d i :ifter or:il
q r
CH
tloiing. (1) This research was s u i . l p o r t d 1'4. l'iil,lic Healtli lteaearcli (:rant ( ; > I 15860 from t h e National Institute of General AIPdical Sciences. (2) Alfred P. Sloan Fellow, 1965-1969. 13) For a general reference see \V, 11. Fisliman. "Chemistry of U r t ~ i . hIetabolism," Charles C Thomas, Publisher, Yiiringfield, Ill., 19151. 14) D. Robinson and R . T. FVilIiams, Btochem. J . , 59, 159 (19%). ( 5 ) D. Robinson, .J. S . Smith, and 11. T. Williams, i b z d . , 59, 153 (19551. (6) Preliminary evidence f r o m this l a l ~ o r a t o r ~indiratet l i i i t ring I t \ droxylation also occurs with f-l,utylbenzene ( r a t ) . (7) It. J. Fesaeriden and C . hhlfurs, .I. .\Ted. C h v ~ n . 10, , X I 0 11!~1i71.
0= unknown conjugate I n another ruii, using an isotopic dilution techniyiw, ,'3'< of the total urinary activity was shown to bc from
liesaniethyldi3ilox:iri~~. One of the metabolites, 11tritnethplsilylphcnol (IT-), is k i i o \ ~ r i to undergo rv-
arrangement to trimethylsiloxybenzene (VI).x This compound, in turn, reacts with water to form phenol and trimethylsilanol, which undergoes dehydration to hexamethyldisiloxane. It is felt that the small amount of hexamethyldisiloxane found was due to the chemical rearrangement and cleavage rather than to a biological process. 1-110
A
p-HOC6H~Si(C&)3-+- C6HBOSi(CH~)3 +CSH,OI%f IV VI - Ii>O (CH&SiOH + (CH3)3SiOSi(CII3)3
Although the tlc isolation procedure did not yield analytically pure materials, the procedure did concentrate the radioactive material to a point where a conibination of ir, nmr, and mass spectral data provided sufficient evidence for structural assignment. As further evidence for the assignment of structures to the radioactive metabolites, the phenol IV and the hydroxymethyl compound I11 were synthesized by independent routes and their spectra were compared to those of the isolated metabolites. In the case of the unknow~nconjugate, the structure of the conjugate was not determined; the structure of the silicon part of the compound is based upon spectral evidence. This metabolite clearly is not I11 or IV as determined by tlc R f values. The nmr spectrum indicated the presence of silicon-phenyl and siliconmethyl groups. The ratio of the silicon-phenyl to silicon-methyl protons in the nmr was approximately 5 to 6. The mass spectrum of this material (direct probe) showed a base peak a t m/e 135 vhich can be assigned to the ion, COHSSi(CH&+. These data indicate that the metabolite contains a conjugated hydroxymethylsilicon group rather than a conjugated hydroxylated ring. Phenyldimethylsilane (II).-After oral dosing of 11, SS% of the radioactivity appeared in the urine within 36 hr. After the pooled urine was made slightly acidic, the activity was easily extracted (EtlO). A mass spectrum of the crude concentrated extract shoned a base peak a t m / e 137 (36y0of the total ionizing current). The synthetic silanol VI1 also shows a base peak at n z l e 137 (95yoof the total ionizing current), but the corresponding disiloxane VI11 does not. The crude extract was purified by the tlc procedure. Two active spots were observed, 90% on the plate and 10% at the origin. Collection and spectral analysis of the 90% spot showed this material to be diphenyltetramethyldisiloxane (VIII).
,
CH3 VII, -907,
VI11
From the mass spectral data before and after purification, it is clear that the metabolite was phenyldimethylsilanol (VII) (or a conjugate of the silanol), which underwent condensation to the disiloxane VI11 during purification. Unlike the cases of isopropylbenzene, t-butylbenzene, and trimethylphenylsilane, other hydroxylated products were not observed. We conclude that the silicon(8) J. L. Speier. J. Am. Chem. Soc.. 74, 1003 (1952).
hydride is iiot st':rblc ifi o i w and that this bond is the princiru1 s i t e of nictabolic attack on this compound.
Experimental Section P3:eriyIdimethyI[~aC]methylsilane (I).-Phenyldiniethylchloroqilaiie, bp 4S-52' ( 3 mm), was prepared in yield by the reactioii of 1.0 niole of PhlLgBr with 4.0 moles of dimethyldic.hlorcrsilarie. To 0.2 g (8.3 mmoles) of 1 I g in 8 ml of dry Et20 was added 1.0 g (6.!1 mmoles) (ca. 50 pCi) of [14C]hIeI. The mixture was heated at reflLis for 30 min. To this Grignard reagent was addcd 2.4 g (14 mmoles) of PhSiClMez and the mixture was heated at t,eflux for 2 hr. Then 10 ml of 0.5 X HCl was added slowly. The organic material was extracted (EhO) and dried (Xa2W4). To the Et& solution was added 1.0 g of nonradioacf ive I. Fractional distillation through a 15.2-cm glass helices packed column yielded 1 . 3 g (497c) of t'he product, bp 164-165', with an activity of 15.5 pCi/g. In repeat runs the yields were 47-69% with levels of activity of 23.6-34.5 pCi/g. Prior to dosing, the material was further diluted with nonradioactive I. Phenylmethyl[14C]rnethyl~ilane (11)-Phenylmethylchlorosilaiie, bp 50" ( 3 mm), was obtained in 52% yield from PhhIgBr and methyldichlorosilane. Using the procedure described above, the chlorosilane was added to [14C]11eMgIto yield 1.38 g (41%) of the active product (11),bp 151-153', 20.5 pCi/g. Comparison Compounds.-(Hydroxymet~hg1)dimethylphenylsilane1° and p-trimethylsilglphenol11 have been prepared previously in our laboratory. Diniethylphenylsilanol was obtained from the hydrolysis of an Et20 solution of phenyldimethylchlorosilane. Et,O n-as removed using a rotary evaporator and the residue was used directly for spectral analysis. This silanol was heated to effect condensation to the diphenyltetramethyldisiloxane, bp 190" (10 mm).12 Dosing, Elimination, and Extraction. A. Phenyltrimethylsilane (1).-A4 male Long Evans rat (300-400 g) was given a single oral dose of 0.4 ml (5.3 pCi) of I by means of a stomach tube. The rat was placed in a metabolism cage and allowed food and water ad libitum. The urine and the fecal material were collected separat,ely for each 24-hr period. Only the urinary metabolites were investigated. Within the first 24 hr, 1.8 pCi (34%) appeared in the urine, and, in the second 24 hr, 0.85 pCi (167,). On the third and fourth days only trace amounts of activity could be detected. A total recovery of 50Y0 (2.68 pCi) was effected. For the isolation work, the first 48 hr of urine of 43 rats was collected. The per cent of dosed activity eliminated by individuals in this group varied from 32 to 40% on the first day and 8 to 207, on the second. In one experiment, a dosed rat was placed in a cage equipped for the collection of COZ by passing the respired air through NaOH soluion. KO 14C02could be detected. A number of extraction procedures were tried with varying degrees of success. Although BuOH was an effective solvent for the extraction, large amounts of nonactive organic material, which interfered with the final purification, were also extracted. EtsO was not effective. The use of continuous extraction was explored but was discarded because of emulsion formation. The pooled urine from six to ten rats (50 ml) was filtered through glass wool then placed in a separatory funnel and extracted with ten 200-ml portions of EtOAc. By this procedure, 85-9070 of the activity was extracted from the urine. EtOAc was removed by a rotoevaporator a t room temperature. N o activity was lost in this concentration. The residue was taken directly into the tlc separation procedure. B. Phenyldimethylsilane (II).-The procedure described for I was also used for 11. The rats were given 0.2 ml (3.67 pCi) of I1 and eliminated 3.24 pCi (88%) of the dosed activity in the urine within 36 hr. Oral dose levels of 0.3-0.5 ml of I1 were toxic. No attempt was made to assay for respired l4CO2. (9) A Packard EX-314 scintillation counter mas used for all radioactive measurements. The scintillation solution contained tolueneethanol ( 8 0 : 2 0 ) and 4.4 g/l. of Packard Pre-Mix M . The aliquot counted was either 50 or 100 p l . The cpm were converted to dpm and expressed in this report as pCi. The ir spectra were obtained using a Beckman IR7 equipped with a beam condenser and micro cell, the nmr spectra using a Varian H A 4 0 IL instrument, and the mass spectra using a modified CEC-103. (IO) R . J. Fessenden and M. D . Coon, J . M e d . Chem.. 9, 262 (1966). (11) R . J. Fessenden, K. Seeler, and AI. Dagani, J. O w . Chem.. 31, 2483 (1966). (12) W. H.Daudt and J. F . Hyde. J. A m . Chem. Soc., 74, 386 (1952).
