COXMUNICATIONS TO THE EDITOR
1745
change in a0 in going from water to S07Gdioxane is approximately half as large as the difference in a0 between denatured and native “non-helical” proteins, data for two of which are shown in the table for comparison. T h a t these results are not due to the presence of a small number of ionized carboxyl groups in the purely aqueous solution, with suppression of the ionization as dioxane is added, was demonstrated by repeating some of the experiments in the presence of 0.1 M HC1. No significant change was observed. The general trend of the data is also independent of the choice of Xo, within narrow limits. m’ith Xo = 234 mp, for example, bo = 45’ and a. goes from -138 to + 5 3 O . Data similar to those shown here have been obtained with N-acetyl-L-leucine X-benzoyl-L-glutamic acid, and N-benzoyl- L-leucine amide, and with a wide variety of solvents For each substance, the effect of the solvent is to change ar and not bo. In each case the non-polar solvents lie a t one extreme and the more polar ones a t the other. The results will be reported in full a t a later date. The conclusion to be drawn from all the data is t h a t a change in environment of peptide groups of a protein, such as accompanies denaturation, is likely to make a substantial contribution to the over-all change in optical rotation, and that this contribution is likely to appear as a change in anwhen the data are analyzed by use of equation 1 (6) T h e able technical assistance of Mrs knowledged
P
hl
Hudson
15
ac
DEPARTMEXT OF BIOCHEMISTRY DUKEUNIVERSITY CHARLESTAX FORD^ DURHAV, N C RECEIVEDFEBRUARY 15, I962
Sir :
SYNTHESIS OF PODOPHYLLOTOXIN’
Podophyllotoxin (V), containing a strained lactone system, is easily and essentially completely isomerized with alkaline catalysts to the thermodynamically more stable picropodophyllin (I). We now find t h a t protonation of the enolate3 of a picropodophyllin derivative makes possible the regeneration of podophyllotoxin (17) from picropodophyllin (I). Protonation of the enolate of picropodophyllin (as in 111) could lead either to podophyllotoxin or to picropodophyllin. Examination of models suggested t h a t the front of the enolate is somewhat more accessible to a n approaching proton donor than the back. Conceivably, therefore, frontside protonation, which would give podophyllotoxin, could compete effectively with backside protonation, which would give picropodophyllin. (1) This report. describing a portion of t h e Doctoral research of C. D. Gatsonis, represents paper No. XI11 in t h e series on “Compounds Related t o Podophyllotoxin”; the preceding paper is by W. J . Gensler, F. Johnson, and A . D. B. Sloan, J . A m . Chem. Soc., 82, 6071 (1960). ( 2 ) A comprehensive resierv is given by J. I,. Hartwell and A . W. Schrecker, Forfschr. Chem. O r g . .\-afwsfofe, 17, 83 (19.58) (3) T h e rate controlled protonation of enols has been studied carefully by Zimmerman and his associates Pertinent references may be found in a paper by H. E. Zimmerman and T . W. Cutshall, J . A m . Chem. S u c , 81, 1305 ( 1 9 5 9 ) . I n the present note, we have made no distinction between enol and enolate; the same stereochemical arguments apply to hoth.
I-01, s4
OH
1
AI, ’0
OCHj 1
TT
I1
1
Picropodophyllin
* y & 0 6 G o Ar
I11
0-
Ar
0
Iv
‘ Ar
0
v
Podophyllotoxin Test of this possibility required formation of the enolate of picropodophyllin. T o avoid complications arising from the presence of hydroxyl hydrogen, picropodophyllin (I) was converted by combination with dihydropyran t o O-tetrahydropyranylpicropodophyllin (11). One of the two crystalline diastereoisomers of 114 was treated a t room temperature with just over one molar equivalent of triphenylmethylsodium in ether.5 Then to neutralize enolate I11 in the resulting tan mixture, cold acetic acid containing a trace of sulfuric acid was added in one portion. After preliminary fractionation of the protonated material, the tetrahydropyranyl derivative (IV) of podophyllotoxin was treated with hot hydrochloric acid in alcohol to remove the protecting group. Crystallization and chromatography of the product furnished pure podophyllotoxin (V) in 23YGyield.6 Approximately 40% of the protonated product consisted of the tetrahydropyranyl derivative of picropodophyllin (11) mixed with some picropodophyllin. It is expected that proton sources bulkier than acetic acid will increase the ratio of podophyllotoxin to picropodophyllin. Since picropodophyllin has been synthesized,’ the work described here completes a total synthesis (4) Some of the properties of this form are: m p. 203-204‘; lalz6D + l o 3 ( c = 1 in chloroform): ultraviolet absorption maximum a s a 10-4 M solution in alcohol, 201 m e (log c 3 5 1 ) ; no infrared absorption for hydroxyl a t 3700-3126 cm.-’; A m i . Calcd. for CmHsnO?i, C, 65.05, H , 6.07. Found: C , 65.31: H, 6.10. Exploratory work by Dr. S. C. Chakravarti showed t h a t t h e reaction of podophyllotoxin with dihydropyran and hydrochloric acid catalyst gave t h e crude tetrahydropyranylpodophyllotoxin in low yield. Later, R . G . RlcInnes obtained the tetrahydropyranyl derivatives of podophyllotoxin and of picropodophyllin, as mixtures. by using dihydropyran as s d v e n t and a trace of phosphorus oxychloride as catalyst. I n the present work, the tetrahydropyranyl derivatives of picropodophyllin were prepared with dihydropyran in chloroform solvent with fi-toluenesulfonic acid as catalyst. ( 5 ) W. B. Renfreu-, Jr , and C. R. Hauser, “ O Y ~ .Syntheses,” Collective Volume 2 , 607 (1943); C. R . Hauser and B. E. Judson. J r . , O r g . Reacfioizs, 1, 286 (1912). (6) Some of t h e d a t a are: m.p. 160-161°; t h e synthetic podophyllotoxin showed m.p. 156-157‘ when mixed with authentic material ( m p. 156-1Ri0), [a]% - 1 3 2 O ( c = 1 in ch1.iroform); t h e infrared absorption curves of synthetic and authentic podophyllotoxin are the same, treatment of synthetic podophyllotoxin with piperidine causes isomerization t o picropodophyllin. ( 7 ) W .J. Gensler, C . .2.1 Samour. Shih Vi Wang and P J(,hnson, J . A m . ( ‘ h ~ m.?or , 82, li14 fl960).
hfay 5 , 1962
COMMUNICATIONS TO THE EDITOR
1739
tained 9.47, Cr, 9.37@P, 58.57, C, and 4.97, H ; a fraction insoluble in ethanol and benzene but soluble in chloroform S.6yo Cr, 10.87,P, 59.67, C, and 4.8% H ; a fraction insoluble in all three S.G% Cr, 10.3% P, 59.1% C, and 4.9% H . The latter two agree well with S.SS% Cr, 10.5Syo P, 59.497, c, and 4.657, H , the values calculated for [Cr(AcCHAc)(OPPh20)2]n; the former with the values 9.66% Cr, 9.21% P, 5S.46% C, and 433% H calculated for (AcCHAc) [Cr(AcCHAc)(OPPhsO)z]rCrDEPARTMENT OF CHEMISTRY BOSTONUNIVERSITY WALTERJ. GENSLER (AcCHAC)~. BOSTON15, MASSACHUSETTS CHRISTOSD. GATSOXIS Ebullioscopic measurements in benzene have RECEIVED MARCH7, 1962 yielded number-average molecular weights of 1940 to 2633 for benzene-soluble fractions and up to lO,S70 for chloroform-soluble fractions. The benPOLY- [DI-p-DIPHENYLPHOSPHINATOzene-soluble fraction thus contains on the average ACETYLACETONATOCHROMIUM(III)]. A COORDINATION POLYMER WITH AN INORGANIC four to five chromium-containing units per polymer BACKBONE segment assuming either acetylacetonate or diSir: phenylphosphinate end groups, the chloroformMuch of the recent activity directed toward the soluble, about eighteen. We have not been able to synthesis of cocrdination polymers has involved the determine the molecular weight of the insoluble investigation of systems in which metallic ions are fractions yet, but, since some of the soluble as well catenated by bis-chelating agents.l-I1 Such cate- as the insoluble fractions have compositions agreenating groups are, perforce, organic in nature, ing with the infinite polymer, the difference bewith the result that the backbone of the polymer tween them presumably is one of molecular weight, contains organic linkages. At this time we wish the less soluble fractions having higher molecular to report the preparation of a new kind of coordi- weights. I t thus appears t h a t we have succeeded nation polymer with a completely inorganic back- in forming a polymer with degree of polymerization bone. dependent upon reaction conditions, substantially By a substitution-addition polymerization,12 greater amounts of higher molecular weight species we have made a series of products with the basic forming a t higher reaction temperatures. composition Cr(AcCHAc) (OPPhzO)e. Although The structure of these polymers probably inwe have produced such products by several tech- volves double diphenylphosphinate bridges13 beniques, the most satisfactory has been the direct tween chromium atoms in the backbone, i . e . reaction between chromium(II1) acetylacetonate and diphenylphosphinic acid a t temperatures from 175 to 250' under a slow sweep of nitrogen. Heating is continued until acetylacetone can no longer be detected in the exit gas by reaction with ferric ion. The residue then is separated into fractions by extracting successively with ethanol, benzene, and chloroform in a Soxhlet extractor. After these extractions the amount of insoluble residue from the reaction between 10.5 g. of Strong evidence for such an assumption is afforded Cr(AcCHXc)3 and 13.1g. of Ph2P(0)OHwas 2.7 g. by the isolation and characterization of the first a t 175', 5.S g. a t 200°, and 9.4 g. a t 250°, a marked member of the series, the dimer (AcCHAc)&rincrease with increasing temperature. A fraction (OPPh20)2Cr(L4cCHAc)2.It is readily produced insoluble in ethanol but soluble in benzene con- in high yield by the reaction of diphenylphosphinic acid with excess chromium(II1) acetylacetonate a t (1) D. N. Chakravarty and W. C. Drinkard, Jr., J . Indian Chem. 240°, gives proper analysis for the formula given SOC.,37, 517 (1960). (2) W. C. Drinkard, Jr., D. Ross, and J . Wiesner, J . Org. Chem., (found 11.0% Cr, 6.5% P, 56.7Y6 C, and 5.270 H ; 26, 619 (1961). calcd. 11.13Y0Cr, 6.637, P, 56.547, C, and 5.16% (3) N. A . Glukhov, hl. hl. Koton, and Y. V. Mitin, VysokomolekulyH ) , and gives an ebulliometric molecular weight of w n y e Soedineniya, 2 , 791 (1960). 917 in benzene (calcd. 934.8). Increasing the (4) R. N. Hurd, G. DeLaMater, G. C. McElheny, and L. V. Pfeiffer, 1. Am. Chem. Soc., 82, 4464 (1960). relative amount of diphenylphosphinic acid results ( 5 ) C. S . Kenney, Chem. & I n d . , 880 (1960). in the continuation of the chain started in the dimer, (6) R . U'. Kluiher and J. W. Lewis, J . A m . Chem. SOC.,82, 5777 building up a polymer of Cr(AcCHAc)(OPPh,O)Z (1960). units. An alternate technique we have used for ' 1 Korshak, et al., Vysokomolekulyarne Soedineniya, 2, 498 (7) V. . and 662 (1960). the preparation of the polymer is the reaction of (8) A. B. P. Lever, J. Lewis, and R . S. Nyholm, S a t w e , 189, 58 the dimer with diphenylphosphinic acid in 1 : 2 (1961). ratio. (9) L. E. Mattison, hl. S. Phipps, J. Kazan, and L. Alfred, J . An examination of the structure of the polymer Polymev Sci., 54, 117 (1961). (IO) A. S.Sesmeyanov, M. I Ryhinskaya, and G. L. Slonimskii, with Stuart-Briegleb models shows that closure of Vysokomolekulyarne Soedineniya, 2 , 526 (1960). a ring requires several units and is unlikely. of podophyllotoxin. The method also indicates the feasibility of synthesizing base-sensitive podophyllotoxin derivatives, of interest in connection with cancer chemotherapy, via the stereochemically stable picropodophyllin forms. Financial support from the National Cancer Institute, U. S. Public Health Service, in the form of Research Grant CY-2891 is gratefully acknowledged.
(11) H. 1,. Schafer and 0. Kling, Z . anovg. u . allgem. C h e m . , 309,
24.5 (1961).
(12) B . P. Bluck and G. Barth-Wehrenalp, J . I X O V E .& Sucleai C h e n . , in press.
(13) Similar bridges have been postulated t o he present in UOz(OP(OBu)20)2: C. F. Baes, J r , , R . A. Zingaro, and C . F Coleman, J . Phys. Chenz., 62, 129 (1958).
