Nov. 20, 1962
COMMUNICATIONS TO THE EDITOR
basis of the recent report on the enzymatic conversion of phenylalanine t o phenylserine followed by cleavage with threonine aldolase to give benzaldehyde and g1y~ine.I~If a similar enzyme were functioning in N . incomparabilis, i t may be possible for phenylalanine + phenylserine + protocatechuic aldehyde without going through tyrosine. The protocatechuic aldehyde then would be the c&I fragment. To test this possibility, Ha-protocatechuic aldehyde was injected into four plants, and 0.23y0 of the tritium was incorporated into the lycorine (recrystallized four times) (Table I). The exact location of the tritium is currently being determined. Protocatechuic acid-C14is not incorporated.16 On the basis of the above results, it appears that the two aromatic amino acids, phenylalanine and tyrosine, give rise to protocatechuic aldehyde and tyramine which serve as the Cc-c~ and C&2 units, respectively, in the biosynthesis of lycorine. Tyrosine or tyramine can not replace phenylalanine as the c&1 fragment and phenylalanine can not replace tyrosine or tyramine as the c&2 unit. (15) A. L. Jordan, W. J. H a r t m a n , Fed. Proc., '21, 51 (1962).
4349
was synthesized and found also to be photochromic, white crystals becoming yellow on irradiation. Other salts of type I, including triH
F(C&)3
I
phenylphosphonium methylenetriphenylphosphorane tetrafluoroborate, Ib, and iodide, IC,change color more slowly than Ia. Colors resulting from ultraviolet irradiation are listed in Table I, together with analytical data. Color changes were observed in air or nitrogen a t room temperature, and were not reversed by heating to looo. When the source was removed, the intensity of the color slowly diminished to a pale yellow. Infrared and ultraviolet spectra of the compounds were unchanged by irradiation. Characteristic infrared bands (KBr pellet) attributed to the mesomeric cation occur as a triplet of increasing intensity a t 1027, 1010, and 988 cm.-'. (16) A. R. Battersby, Private Communication. Ultraviolet A,, (acetonitrile) occur a t 267 mp RESEARCH LABORATORIES ( E = 7200-10,800). For Ia, continuous absorption DEPARTMENT OF BIOCHEMISTRY R. J. SUHADOLNIKwith no peaks occurs in acetonitrile in the visible A. G. FISCHER ALBERTEINSTEINMEDICALCENTER PHILADELPHIA 41,PENNSYLVANIA J. ZULALIAN region ( E = 1, 7, and 51 a t 600 ,500 and 400 mp, respectively). RECEIVED SEPTEMBER 20, 1962 The mesomeric compounds were prepared from triphenylphosphonium methylenetriphenylphosPHOTOCHROMISM I N MESOMERIC PHOSPHONIUM phorane bromide,2 Id, which is not photochromic. SALTS' Id was obtained by treating methylenebis-(triSir : phenylphosphonium bromide) with aqueous sodium We wish to report preliminary results on the carbonate2 or lithium butyl in ether. Metathesis photochromism of mesomeric phosphonium salts. reactions of Id, in methanol or water, yielded triWhen white crystals of triphenylphosphonium phenylphosphonium methylenetriphenylphosphomethylenetriphenylphosphorane tetraphenylboron, rane tetraphenylboron, Ia, tetrafluoborate, Ib, and Ia, are irradiated by sunlight or an ultraviolet iodide, IC. The inner salt I1 was synthesized by lamp, an orange-red color rapidly develops. the addition of triphenylboron to hexaphenylTABLE I MESOMERIC PHOSPHONIUM SALTS Compound
Ia
Empirical formula
CsiHmPd3
M.p.,O o c .
205-208
Irradiated color
Orange-red
C
H
Elemental analyses, %* B P
85.5 6.0 7.2 1.3 85.3 6.1 7.4 1. 0 5.0 9.95 1.7 Ib C37HaiP2BF4 259.5-260.5 Yellow 71.2 5.55 10.0 1.5 71.5 4.7 9.3 .. IC C~HNP~I 253-254 Yellow 66.9 4.7 9.4 .. 67.2 5.0 10.0 .. Id C37H31PzBr 2i4-275 White 72.0 5.2 10.1 .. i1.9 I1 CssHaaPrB 253-255 Yellow 84.8 5.8 7.95 1.4 85.1 5.9 7.7 1.5 a Melting points are corrected. * Calculated values are listed on the upper line; found values on the lower. are for halogen as indicated in empirical formulas.
X C .
I
.. 12.2 12.0 19.1 19,o 13.0 13.2
..
.. X values
With a Hanovia Lamp (Model 30600), a t a distance of four centimeters, the change occurs in four seconds. E.p.r. measurements show that unpaired electrons are present in the orange-red material, but not in the white crystals. The corresponding inner salt, triphenylphosphonium(triphenylboronyl)methylenetriphenylphosphorane,11,
carbodiphosphorane,2 obtainable from Id by reactions with potassium,2sodium hydride or lithium butyl. Proton and P31 n.m.r. spectra are consistent with the assigned mesomeric structures. For Ia, BII resonance was observed a t +25 p.p.m. relative to trimethyl borate in methylene chloride. For the inner salt I1 it was necessary to utilize
( 1 ) This work was supported in part by the O 5 c e of Naval Research.
(2) F. Ramirez, N. B. Desai, B. Hansen and N. McKelvie, J . A m . Chem. SOC.,8S, 3539 (1961).
the rapid passage technique in order to observe the BI1 resonance a t +20 p.p.m. in the same medium. The study of photochromism in mesomeric phosphonium salts and other organophosphorus compounds is continuing.
chromatographic analysis indicated a ratio of insertion into primary vs. secondary carbonhydrogen bonds of 0.05 for both photolysis and pyrolysis. The remarkable reactivity differences of chlorocarbene derived from the diazocompound and of CLIFFORD S.ATTH THEWS the formally identical intermediate in a-elimination MONSASTO RESEARCH CORPORATION JOHS S. DRISCOLL from methylene chloride4 is noteworthy. The BOSTOSLABORATORIES J O H N E HARRIS failure of the latter intermediate to insert into 49, MASSACIWSETTS KORERTJ WIVEMAN carbon-hydrogen bonds and its relatively high EVERETT RECEIVED OCTOBER 8, 196:' steric discrimination in additions to olefins6 (e.g. ratio of the two I -chloro-cis-2,3-dimethylcycloTHE PREPARATION, PYROLYSIS AND PHOTOLYSIS propanes resulting from the addition to cis-2butene is 5.5)6suggests that in the a-eliminations OF CHLORODIAZOMETHANE1 a truly free carbene might be bypassed.' Sir : ( 6 ) G. L. Closs and G. >I. Schwartz, ibid., 82, 5729 (1960). Halogen derivatives of diazomethane are of the similar observations for diphenylcarbene; G. I,. Closs interest as potential precursors of divalent carbon and(7)L.Cf. E. Closs, Angew. Chem., 74, 431 (1962). intermediates. Particularly, comparison of re(8) A. P. Sloan Foundation Fellow, 1962-190.1. activities of halomethylenes derived from such di- DEPARTYEST OF CHEMISTRY G. L. CLOSS* azocompounds with those of the formally identical THEUNIVERSITY OF CHICAGO carbenes generated by a-eliminations promises to CHICAGO 37, ILLINOIS J. J. COYLE give new insights into some problems of carbene RECEIVED SEPTEMBER 21, 1962 chemistry.2 We wish to report the preparation of solutions of chlorodiazomethane and some of their ON THE MECHANISM OF ERGOT ALKALOID physical and chemical properties. BIOGENESIS' Reactions of t-butyl hypochlorite with dilute Sir : solutions of diazomethane in ether or hydrocarbons Although the biogenetic origin of the C-8 suba t -100' give solutions of a red compound which we believe to be chlorodiazomethane on the basis of stituted N-methylergoline skeleton of the ergot the following observations : The electronic absorp- dlkaloids has been established as arising from tion spectrum in pentane exhibits a bathochromic tryptophan,2 mevalonate2 and the S-methyl of shift relative to diazomethane with three maxima methionine, the mechanism of the biosynthesis of about equal intensities a t 485, 518 and ,745 mp is still unknown. The association of mevalonic acid (I) with tryptoi E about 15),3 respectively. The infrared spectrum shows a strong band a t 2066 ern.-' attributed phan (111) involves the bond formation of C-5 of to the stretching vibration of a diazo group (z'P. mevalonate (I) with the C-4 and a-carbon of tryp2097 cni.-l for diazomethane). The solutions tophan (111), and the C-3' of mevalonate (I) with slowly decolorize with loss of nitrogen a t tempera- the a-amino group of tryptophan (111),respectively tures above -40'. Reactions with carboxylic In order to obtain information on the mechanism acids lead to rapid nitrogen evolution and de- of the formation of the new C-C linkages and of the colorization. Chloromethyl acetate and chloro- biosynthetic route by which the different substituents appear a t C-8 of the ergot alkaloids, the methyl propionate have been prepared in this way. Confirming evidence for the structure of the chlo- changes in the oxidation state a t C-5 and C-2 of rination product was obtained from the pyrolytic mevalonate were examined during its conversion i- 20') and photolytic (-80') decompositions of into ergot alkaloids. During the over-all biosynthetic reaction the these solutions. Using olefins as solvents, satisfactory (30-40yc) yields of chlorocyclopropanes T,Cl4 ratio of DL-mevalonic acid-2-C14-5-T comwere isolated. Thus, trans-2-butene reacted stereo- pared to the same ratio of the ergot alkaloid (festuspecifically to give 1-chloro-fr~ns-2.~-dimethylcpclo-clavine and pyroclavine) has been changed to a propane while ris-2-butcnc gave the two epimeric degree which indicates the net loss of 1 atom of hy1 - chloro - c i s - 2.3 - dimethplcyclo~iropallesin cqu?l drogen. The T,Cl4 ratio of rx-mevalonic acidcluantities. Similarly. with cyclohexene both T- 2-C14-2-T did n o t change during its conversion chloronorcareries were formed in a 1 : 1 ratio. ' into agroclavine, indicating that there is no change Froin reactions in n-pentane the three possible in the oxidation level a t C-2 of mevalonate during carbon--hydrogen insertion products, I -chloro- the biosynthesis (Table I). The loss of one atom of hydrogen from DLhexane. 1-chloro-2-niethylpentane and l-chloro-2ethylbutane, have been identified.j Vapor phase mevalonic acid-2-C14-5-T may be associated with the oxidation of the primary alcoholic group of (1) Supported by a grant from t h e Petroleum Research Fund, admevalonate (I) into an aldehyde group before its ministered by t h e American Chemical Society. bond formation with tryptophan. Another pos(2) For a recent review see: W. Rirmse, Angcw. Clienz., 73, 161 sibility which may exist is that the first bond for(1961). ( 3 ) Since a satisfactury method for assaying t h e chlorodiazomethane mation occurs without any change in the oxidation ~
wlutionb has not yet been found, the c value is based on a n estimated yield of ,505; and undoubtedly will he subject t o revision. ( 4 ) Ci.G . I.. Closb and I.. 12. Closs, .I. A m C h ~ m .S o ( . , 82, 3 7 2 3 IlOIiO).
12) These compounds were isolated by v.p.c. and their ini'rarrd s l x c Ira wi'rr comi>art.,l w i t h t h u w
