Journal of Medicinal Chemistry, 1971, Vol. 14, No. 6 475
SYNTHESIS OF TRH
Synthesis of the Thyrotropin-Releasing Hormonet JANB$LER,JAW-KANG CHANG,FRANZ ENZMANN, AND KARLFOLKERS* Institute for Biomedical Research, The University of Texas at Austin, Austin, Texas 78713 Received December 29, 1970 Two syntheses of the thyrotropin-releasing hormone (TIIH), pGlu-His-Pro-NHy, were an integral phase of the structliral elucidation of this hormone. The first syntheliis consisted in the conversion of Glli-His-Pro to TIIH by reaction with HCl in MeOH and then "3. The second synthesis was based on combining pyroglutaniic acid and histidine without the advantage of any N-protecting group followed by reaction of the dipeptide with Pro-NH? t o give TIIH. This second synthesis was subsequently modified by using S-carbobenzoxypyroglutamic acid in the form of a mixed anhydride to result in animproved total yield of T R H .
The porcine hypothalamic releasing hormone was isolated by Schally, et aZ.,1 from a total of about 265,000 fragments. The several fractions of the hormone from the isolation which amounted to about 5-7 mg presumably ranged in purity from about 30 to 50%. Some techniques of classical structural elucidation were impossible for use on porcine T R H and other procedures were severely restricted because of the paucity of hormone, the impurities, and the cost and time involved. Nevertheless, useful spectral data on T R H were obtained; particularly, the probable relevance of 3 amino acids, which were in the sequence of Glu-His-Pro, to the structure of the hormone had been proposed by Schally, et aLz However, the peptide concept of T R H , as based on studies on the ovine hormone by Guillemin, et had been doubted in early efforts to gain clues on the chemical nature of the hormone. Also, the behavior of the porcine hormone in gel filtration led to consideration of a molecular weight which would be compatible with a yield of about 30% of the 3 amino acids, Glu, His, and Pro. On this basis, up to 70% of the structure of the hormone remained unaccounted and this concept was reasonable if the isolated porcine hormone were 90% or higher in purity and had the appropriate molecular weight. Preference was given to the interpretation that the porcine hormone was closer to about 2 5 4 0 % purity rather than 90% or so on the basis of a report by Guillemin, et u Z . , ~ that the ovine T R H yielded 8 0 4 5 % of the 3 amino acids in comparison with the 30% yield of these amino acids from the porcine T R H . Direct comparison of the yields could be based on the probability that the porcine and ovine hormones would be proven to be identical. The discovery: that synthetic pGlu-His-Pro-SHz (8) from the methylation and ammonolysis of GluHis-Pro exhibited the hormonal activities of the porcine T R H in the nanogram dosage range unexpectedly reoriented conceptional and experimental efforts on the structural elucidation of the porcine TRH. Since only a few milligrams of the free tripeptide, Glu-His-Pro, was available at the critical time of discovery of the hormonal activity of pGlu-His-Pro-NHz, a second synt
Hlpothalamic Hormones 15 (1) 4 V Schalls, C Y Bovers, T T\ Redding and J F Barrett, Btochem Baophys Res Commun , 9 6 , 165 (1966) (2) A V Scha114, T \\ Redding, C P Boners and J F Barrett, J Bad Chem ,244,4077 (1969) (3) R Guillemin, E Sakiz, and D l i Ward, Proc Soc E z p Btol N e d , 118, 1132 (1965) (4) R Burgus, T F D u n n D R I Desiderio, D N Ward, \\ Vale, R Guillemin, R I Felix, D Gillessen, and R 0 Studer, E n d o c r m o l o g y , 86, 573 (1970) (5) K Folkers F Enzmann, J Bdler, C Y Boners, and A V Schalls, Btochem Baophys Res Commun SI, 123 (1969)
thesis of pGlu-His-Pro-NHz was urgently needed. The second synthesis was promptly achieved by using the 3 amino acids without any protecting group for any N atom. I t was expedient to treat directly the readi!y available pyroglutamic acid (1) with histidine Me ester (2) to give the dipeptide pGlu-His-OCH3 (3). Conversion of the Me ester to the free acid, pGlu-HisOH (6), and coupling with prolinamide (7) gave pGluHis-Pro-NHz (8). For the objective of the communication,6the nmr spectral data of the pGlu-His-Pro-NHz (8) from this second synthesis was reported. The details of our second synthesis of pGlu-His-Pro-SH2 (8) are described herein. The synthetic pGlu-HisPro-SH2 (8) from both of the syntheses possessed identical R f values and these Rf values in turn n-ere identical with those of the porcine T R H in 17 diversified systems as described.6 pGlu-OH 1
--
+ His-OCH,
DCI
2
pGlu-His-OCH, 3 tPd/C
pGlu-OH
I Cbz
+ His-OCH,
mixed anhydrlder
H
pGlu-His-OCH3
I
Cbz 4
5
NaOH
3 MeOH
pGlu-Hls-OH 6 Pro-NH,
I
T
7 J
JTLONHCHCON H
I
N v "
Co"L
TRH 8
The initial procedure of the second synthesis which involved the direct reaction bet\veen L-pyroglutamic acid (1) and L-histidine ;\le ester (2) gave pyroglutamylhistidine AIe ester (3), in a yield of approximately 55%, and employed the N,N'-dicyclohexylcarbodiimide (DCI) m e t h ~ d . ~ (6) J. B d e r , F. Enimann. K . Folkers, C. Y. Bowers, a n d A . V. Sclially, i b i d . , S I , 705 (1969). (7) K. Sturm, R. Geiger, and W ,Siedel, Ber., 96, 609 (1963).
476
Journal of Mtdicznal Chenzzstry, 1971, Vol. 1!+, S o . 6
Experimental Section 1Ieltiiiy poiiits were obtaiiied w i t h a Thomas-Hoover capilliir! ineltirig point app:iratus and are iiiicorrected. IIicroanalyhewere performed by the Galbraith Labor:itories:, Iiic., Knoxville, 'retin,> aiid by the hIikroanalytisches Laboratorium, Bonn, \Te,i (lerni:iny, on sntnples which were dried at GO" for 24 hr at 0.5 iniii. \Vhere aiialyqes :ire iiidicated orily by synibols of the eleiiieiity, aiialytical resiilts obtained for those elements were withiii 10.4(,',of the theoretical valrie,?. 011t l c (bilica gel G ) )IZr' iiiid 1 2 1 2 valties refer i o !he ,t em of n-BuOIf -glacial AcOIl-li:tOh(~1 I 1 0 ( I : 1 : 1 : 1 ) :itid C I I C I 3 - M e O I I - 3 0 ~ ~KIlrOII (60:45:20), res]). The lowtioii of the peptide sput was revealed k)>- t h e iiitihydiiii reagent, the I'ady reageill, or the CI method.12 The timr bpectra were obtained at G O H z o i i a 1':jrian Aswciates .specir~iireter:iiitl IIA-1 00 i i m r hpect,roniet,er with AIeaSi a- interlid refeiwice, the chemical ahifts are expressed i n 7- value*. The iii:x+ +pectrirni wa1-' takeii oti ii Bell and IIowell 21-491 niass spec*troiiieler. The specific rotations were ohtaiiied with a PerkiiiI
