7/30/2019 Sntesis y actividad antiprotozoaria de 2,5-bis (4-guanylphenyl) tiofenos y-pirroles

1/3

Notes Journa l of Medicinal Chemistry, 1977, Vol. 20, N o. 9 1219Synthesis and Antiprotozoal Activity of2,5-Bis(4-guanylphenyl)thiophenes nd -pyrrolesBijan P. Das and David W. Boykin*D e p a r t m e n t of Chemistry, Georgia State University, Atlanta, Georgia 30303. Received February 14, 1977

2,5-Bis(4-guanylphenyl)thiophenend 2,5-bis(4-guanylphenyl)-N-methylpyrrolesnd several of their cyclic amidineanalogues have been synthesized and their an timalarial and antitrypanosom al activity has been assessed. Noneof these com poun ds showed significant a ntim alaria l activity; however, all displayed good levels of activ ity againstTrypan osoma rhodesiense in mice. 2,5-Bis(4-guanylphenyl)thiophene nd 2,5-bis(4-guanylphenyl)-N-methylpyrroleproduced cures in mice a t the - mg/kg dosage level. These two compound s are of comparable activity to stilbamidine,hydroxystilbamidine, and pentamidine in this test. Th e cyclic amidines generally exhibited lower antitrypanosoma lactivity tha n their guanyl counterparts. Th e thiophenes and pyrroles were synthesized by treatm ent of 1,4-bis(p-bromophenyl)-1,4-butanedione ith HzS-HC1 or CH3NHz-HOAc, respectively, to prod uce th e corresponding2,5-bis(4-bromophenyl)thiophene nd 4-m ethylpyrr ole . The dibromophenyl compounds were converted into thecorresponding bis-nitriles by reaction with CU ~ ( C N ) ~ .he latter compounds were converted by way of imidate estersto th e guanyl an d cyclic guanyl targets.

In a recent publication we reported potent antitrypa-nosomal activity for a series of substituted 2,5-bis(4-guanylpheny1)furans and related cyclic amidines. Thesefura n com pounds were envisioned as potential antipro-tozoan agents based upon structural analogy with bio-logica lly act ive ar y l d i am id in e~ ~ ~ ~nd upon their potentialto inte ract with DNA as their bioreceptdr. In additionto expected antitrypanosomal activity, it was hoped thatthese com pounds m ight also exhibit antimalarial activityin view of th e reported activity of diminaze ne and pen-tamidine against P l a s m o d i u m ~ i n c k e i . ~n view of thepotent activity observed for the 2,5-bis(4-guanylphe-nyl)furans, we have prepared and evaluated the analogousthiophenes an d N-methylpyrroles and the results of theseefforts constitute this report.Chemistry. Th e syn thesis of guanyl and cyclic guanylcompounds shown in Table I was prepared by a methodvery similar to that which has been reported for theprepa ration of th e related furans.l Our appro ach isoutlined in Scheme I.Biological Activity. Compounds 6-13 were screenedfor antimalarial activity by testing against Plasmodiumberghei in mice6and n one of the m exhibited any significantactivity. The se same compounds were tested againstTrypa noso ma rhodesiense in mice by the method of h e 7and t he results are shown in Table 11. Included in TableI1 are test results for stilbamidine (14), hydroxystilb-amidine (15), pentamidine (16), and 2,5-bis(4-guanyl-pheny1)furan (171, a very active compound from ourpreviously described fu ran series. In light of th e resultsreported here and our previous results, we conclude tha tcyclic amidines exhibit lower ord ers of antitrypanosom alactivity tha n their tr ue guanyl analogues and th at acutetoxicity is generally encountered with th e cyclic guanylcompounds, particularly at higher dosage levels. Th e twomost active compounds, 6 and 10 , exhibit comparablelevels of activity to the st an dar d diamidines, 14-16, andthe y are app roximately as effective trypanocides as theirfuran counterparts. Th e fact that the thiophene, N -methylpyrrole, a nd f uran systems show comparable ac-tivity suggests that the role of the five-numbered ringheterocycle in these compounds could be nothing morethan a relatively inert spacer for the guanylphenylfunctions. However, our earlier work on th e furan systemshows tha t enhanced activity is found by substitution onthe fu ran ring. These observations were attribu ted todistribution differences and/or differences in binding toth e bioreceptor. A s tudy to determine if binding to DN A(the possible bioreceptor) can be related to th e structuralvariations in these substituted furan, thiophene, and

Scheme I

CH3 C H 34 5

N-methylpyrrole systems is underway. Because th e ac-tivity of the thiophene and pyrrole derivatives reportedhere did not surpass ha t of their furan analogues, a stud yof the effect of substitution on these heterocyclic ringsystems was not undertaken.Experimental SectionMelt ing points reported under 300 C were taken on aThomas-Hoover melting point apparatus; the melting points ofcompo unds melting above 300 C were obtained on a Mel-Tempapparatus and all melting points are uncorrected. IR spectra wererecorded on all new compou nds with a Perkin-E lmer Model 337spectrometer, H NMR spectra were recorded on selectedcompounds with a Varian A-60A instrument, and 13C NM Rspectra on selected compoun ds were obtained with a J E O L FX-60instrument. All spectra were in accord with the structures as-signed. Elemental analyses were performed by Atlantic Microlab,Atlanta, Ga.Compounds 6 an d 10 were prepared from the appropriatebis-nitrile by th e method outlined for preparation of 10. T h ecyclic amidin es 7-9 an d 11-13 were prepared by th e methodgiven for 11.2,5-Bis(4-cyanophenyl)thiophene 3). Dry HC1 gas waspassed into a suspension of 4.0 g (0.01 mol) of 1,2-bis(4-bromobenzoy1)ethane(1) in 160mL f CHC13 an d 6.0 g (0.02 mol)of SnC14 for 1min. The n HCl and H2Swere bubbled throu ghthe mixture a t approximate ra tes of 1 and 2 mL per second,respectively, for 1.5 h. Th e mixture was filtered and th e filtrateon evaporation gave (2.0 g, 50%) 2,5-bis(4-bromophenyl)thiophene(2); recrystallization from etha nol gave m p 198-199 C (lit.* m pA mixture of 2.0 g (0.005mol) of 2,5.6 g (0.03 ol) of CU*( CN )~,an d 15mL of quinoline was refluxed for 2 h. Th e reaction mixturewas cooled an d extra cted w ith ho t CHC13, th e solvent layer wasremoved, washed with dilute HC1 and with water, an d dried, an dth e solvent was evaporated. The residue was dissolved in a smallvolume of acetone and passed throu gh a sho rt alumina columnto remove traces of copper salts. Evaporation of the eluent and

198-199 OC).

7/30/2019 Sntesis y actividad antiprotozoaria de 2,5-bis (4-guanylphenyl) tiofenos y-pirroles

2/3

1220 Journal of Medicinal Chemis try , 1977, Vol.20, N o. 9Table I.

Notes2 , 5-Bis (4-subs t i tu tedpheny1)thiophenes and -N-methylpyrroles

A AC o m p d n o . X A Mp, "Ce % yieldf Mol formula

3'1 S CN 2 8 8 - 2 9 0 8 3 C 1 E H 1 0 N 2 S5 b NCH, CN 1 9 5 - 1 9 8 7 7 C19H13N36' S -c (NH2),C1. 3 9 8 - 4 0 0 4 0 C 1 E H 1 E C 1 2 N 4 S7a S -C , (CHz!)pC 4 3 8 - 4 4 0 5 5 C 2 2 H 2 2 C 1 2 N 4 S+,YH\ luH

NH8 d S -y H z 1 3 C I - 440-4 42 6 0 C,4H2,C12N4S. H20"

9 d S .C\"J H2 c' - 39 4-39 6 75 C24H26C12N4S+,hH-CHCHs+1 d NCH, -c(NH2),c:l- 37 0-3 72 6 2 5 ' H 2 0

1 d NCH, - >CH2!2C 4 0 5 - 4 0 7 6 0 C 2 3 H 2 5 C 1 2 N 512 d NCH, -L ,~~)CH~)~CI-"\ 4 2 6 - 4 2 8 7 0 C 2 5 H 2 9 C 1 2 N 51 d NCH, -c I CI - 3 9 3 - 3 9 5 5 0 C 2 5 H 2 9 C 1 2 N 5

+/%H\ .NH

+UH-CHCH,'luH-CHz

Recrys ta l lized from d ioxane ; analyzed for C and H and t he resul ts were within * 0.4% of th e calculated values.Recrys ta l lized from EtOH-CH,Cl, ; analyzed for C , H , and N and the re su l ts were wi th in +0 . 4% of the c alculated values .Recrys ta l lized from ab solute e thano l; analyzed for C , H , and S and the re su l ts were wi th in i0 . 4 % of the calculated values.Recrys ta l lized from ab solute e than ol; analyzed for C, H, and N an d th e resu l ts were wi th in *0 . 4% of the calculated values .f The y ie lds fo r 6-13 are based uponAll m e l t ing p o in t s a re uncor rec ted ; com poun ds 6-13 m el ted wi th decom pos i t ion .the im ida te e s te r hydroch lor ide .

Table 11. 2,5-Bis(4-guanylpheny1)thiophenes nd -N-m ethy lpyr ro le s "C ur es b o r A M S F at dosaged (m gik g)

N o. 1 . 25 2 . 5 5 10 2 0 4 0 8 0 1 6 0 3 2 0 6 4 06 2 3 4 5 5 5 5 57 1 .6 D 2 38 2 3 1 4 4 5 59 1.0 D 1 . 2 D 3.0 D 2 4 4

---___

10 1 2 4 5 5 5 5 5 511 2 3 4 4 512 2 3 5 5 3 Te T13 3 4 5 5 5 55 5 5 5 5 53 5 5 5 5 51 f 2 5 5 515g 2 5 5 516h 1 4 5 5 5 5 5 5 5 517' 4 4 5 5 5 5 5 5 5 5

a See ref 7 . A cure is defined as a 30-day increase in survival t ime of the t reated animals over the controls . Five miceA M S T is the increase in mea n survival t im e ofDosage is inH y d r o x y -were used per dosage level ; hence, f ive is the max imu m num ber of cures.test animals vs. cont rol s in days.milligrams of com poun d pe r k i logram of body weight of the tes t animal .s t i lbamidine. Pentamidine. l 2,5-Bis(guanylphenyl)furan.

A MST is di f fe ren t i a ted f rom cures by the use o f D, Le., 1.6 D =1 . 6 d a ys .e T = t o x i c d e a t h . f Sti lbamidine.

recrystallization of th e solid from dioxane gave 1.2 g (83% ) meltinga t 288-290 "C .l-Methyl-2,5-bis(4-cyanophenyl)pyrrole5 ) . A 36% me-thylamine solution (70 mL) was extracted with e ther; the e therlayer was added to 200 mL of acetic acid an d th e mixture waswarmed on a s team bath to remove the e ther. 1 ,2-Bis(4-bromobenzoy1)ethane' (12.6 g, 0.032 mol) was added t o theCH3NH2-HOAc solution and the m ixture was refluxed 1 2 h. Oncooling to room tem peratur e a solid appeared which was filtered,washed with HzO, and dried to yield 9.6 g (62% ), mp 197-198"C . This material, whose 'H NMR was consistent with 1-methyl-2,5-bis(4-bromophenyl)pyrrole4), as used directly inth e next step. A mixture of 7.8 g (0.02 mol) of 4 ,4.0 g (0.022 mol)of Cu*(CN)2, an d 25 mL of quinoline was refluxed for 2 h. Th ereaction mixtur e was worked u p as described above for 3, an d4.4 g (77 %) of 5, mp of 195-198 "C after re crystalliza tion fromEtOH-CH2C12, was obtained.l-Methyl-2,5-bis(4-guanylphenyl)pyrrole10). A solutionof 5.0 g (0.018 mol) of 5,10 0 mL of dioxane, and 25 mL of absolute

EtO H was saturated with dry H Cl gas at 5 "C . Th e solution wasshaken in a pressure bottle a t room temperature for 3 days. Theimidate ester hydrochloride (7.5 g, 93% ) was obtained by reducingth e volume of solvent and allowing th e solid to crystallize. Th atthe nitrile group had reacted was confirmed by the IR spectraand the imidate es ter was dried in vacuo a t room te mpera tureovernight. The dry midate ester hyd rochloride (2.0 g, 0.0044 mol)was suspended in 100 mL of absolute EtO H in a pressure bottleand th e mixture was saturated with anhydrous NH s gas at 5 "C.Th e ammonical mixture was shaken for 3 days at room tem -perature an d the solvent volume was reduced under vacuum. Th esolid obtained by filtration was washed w ith ether an d dried invacuo to yield 1.3 g (62%). Recrystallization from absolute ethanolacidified with anhydr ous HC1 gas gave yellow crystals, mp 370-372"C dec.2,2'-[ l-Methyl-2,5-pyrrolediyl)-pp h e n y l e n e ] d i - 2 -im idazo l ine (11). A mixtur e of th e imidate ester hydrochloride(2.2 g, 0.049 mol), ethy lene diam ine (0.6 g, 0.01 mol), an d 25m Lof absolute ethanol was refluxed for 18h. Th e solid which formed

7/30/2019 Sntesis y actividad antiprotozoaria de 2,5-bis (4-guanylphenyl) tiofenos y-pirroles

3/3

N o t e s Journal o f Medicinal Chemistry , 1977, Vol. 20, No. 9 1221was filtered and recrystallized from absolute ethanol acidified with (2)anhy drou s HCl gas to give 1.3 g (60 %), mp 405-407 C dec.

Acknowledgment. Su ppo rt of this work by the US. (3)Army Medical Research and Development Command (4 )under Contract No. DADA17-68-C-8035 s gratefully ac-(5 )Research Program on Malaria. We tha nk Dr. Edgar A. (6 )Steck for calling our attention to the diamidine area and(7 )

References a n d Notes (8)

knowledged. Th is is Contribution No. 1449 from the Arm yfor helpful discussions and advice.(1) B. P. Das and D. W. Boykin, J.Med. Chem., 20,53 1 (1977).

E. A. Steck, Chemotherapy of Protozoan Diseases, Vol.11,U.S. Gove rnmen t Print ing Oftice, Wash ington, D.C., 1972,Chapters 7, 11.E. Hieke, Pharmazie , 18, 653 (1963).A. Kruetzberger, Fortschr. Arzneimittelforsch., 11, 356(1968).R. Heischkeil, 2. rope nmed . Parasitol., 22, 243 (1971).T. S. Osdene, P. B. Russell, and L. Rane , J. Med . Chem. ,10, 431 (1967).L. Rane, D. S. Rane, and K. E. Kinnamon, Am. J . T r o p.Med. Hyg., 25, 395 (1976).E. Campaigne and W. 0. Foye, J . Org. Chem., 17, 1405(1952).

Effect of the Solvent-Dependent Conformational System of Hydroxyureas onPredicted vs. Observed Log PGeorge R. Parker, Tho mas L. Lem ke,D epar tm ent of Medici nal C hemistry, College o f Pharmacy, l iniuers i ty of Houston, Houston, Texas 77004and E. Colleen MooreD epar tm ent of Biochemistry , M.D. Anderson Hospital and Tum or Ins t i tute , Houston, Texas 77025. Received March 4,

Calculated and observed log P values are reported and compared w ith in vivo and in vitro biological action (L1210leukemia ILS % and ribonucleotide reductase IDm) for hydroxyurea, the 1-N methyl and ethyl, and th e 3-N ethyl,n-propyl, isopropyl, n-butyl, tert-butyl , phenyl, and p-chlorophenyl analogues. Th e log P values were calculatedvia the method of Hansch an d Leo from literature values and the observed log P values were obtained by directdeter mina tion after equilibration between octanol and water. Calculations of log P for hydroxyurea were foundto be apprec iably more hydrophilic tha n the values obtained experimentally. Differences in calculated and observedlog P (A log P ) fo r the substituted analogues were lowest with the 1-N and the bulky 3-N substitu ents and greatestwith the 3 -N-substituted straight-chain analogues (A log P =0.70). Different stru ctura l species were observed byinfrared spectroscopy in dry octanol vs. octanol after water equilibration an d drying, and this is proposed as d ueto changes in conformational equilibrium in th e hydroxyurea systems. Differences between calculated and observedlog P are explained via the stabilization of internally hydrogen -bonded con formers in the case of 1- N or bulky 3-Nanalogues or destabilization of various conformers allowing maximal interactions with solvent or water which is the

1977

case Gith straigh t chain 3-N analogues.

Hydrox yurea (I ), a clinically effective antileukemiaagent,2 is a unique drug since molecular modification hasno t produced an analogue with superior biological actionas evidenced by t he in vivo L1210 activity of s ubstitute danalogues (summarized in Table I).3H ,N -C-NH -O H

ITh e activity of hydroxyurea has been attribu ted to itsability to inhibit th e enzyme ribonucleoside diphosphatereductase4* (RDR ), and comparison of the in vitro in-hibition of this enzyme from Novikoff hepatoma by se-lected substituted hydroxyurea4b Table I) with the in vivodata indicates inhibitory ability in some cases at con-

centrations in th e general range observed with hydroxyureaby all of th e compounds tested (IDrnvalues). Th e relativeimportance of drug transportability, metabolizability,availability, and dynamics at t he site of action for this drugclass a t present is not known. In this paper we wish torepor t discrepancies between calculated and observed logP values for hydroxyurea and some of its substitutedanalogues and a n explanation for these differences due tosolvent-depend ent conform ational preferences of variousanalogueswhich may have a n influence on biological actionin vivo.T he transp ort of hydroxyurea molecules to their site ofaction involves passage through m embranes and th e ad-sorption an d desorption t o macromolecules in vivo. Th ein vivo transp ortab ility of drug m oleculescan be evaluated

b

via the pa rtition coefficient in solvents such as octanol-water5 (log P ) which measures a drugs relative affinitytoward lipophilic and hydrophilic phases. Log P valueshave been measured for many drugs6 and methods havebeen developed by Hansch e t al.? whereby the log P valuefor a particular drug can be calculated since log P has beenshown to be an additive-constitutive property of organiccompounds.8 Log P can be calculated by adding thefragmen t values (f) of th e component fun ctional groupsaccording to e q l.9lo g P =s a v q (1)

In calculating a log P for hydroxyurea, the proximityeffect of groups which can hydrogen bond m us t be takeninto consideration and generally th e f value of the mostnegative fragm ent is used. Th us calculated log P is -2.71or -2.18 depending on the fragments used to make hy-droxyurea (eq 2 and 3).f(-CONH) + f ( -N H , ) + f ( - O H ) = log P =-2.71 (2)- 2.71 -1.54 -1.64f(-CONH,) + f(-NH-) + f ( - O H ) = log P = -2 .18-2.18 -2.11 -1.64 (3 )

When these calculated log P values are compared to th eexperimentally obtained log P for hydroxyurea (Table I)it is apparent th at th e actual log P is more lipophilic tha nth at obtained by calculation from t he com ponent func-tional groups. In order to calculate log P values for