CA1172644A - Process for the preparation of trans-3-formylbut-2- enenitrile - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE
A process is provided for the preparation of trans-3-formylbut-2-enenitrile (V), a key intermediate in the synthetic pathway leading to trans-zeatin (I) and dihydrozeatin (II), both of which are naturally occurring cytokinins. The process involves a base catalyzed condensation of a dialkyl or cyclic acetal of pyruvaldehyde (III) with acetonitrile to yield the corresponding dialkyl or cyclic acetal of 3-formylbut-2-enenitrile (IV). The reaction proceeds regioselectively to form the favoured trans-isomer in good yield. The .alpha.,.beta.-unsaturated nitrile thus formed is hydrolyzed under acidic conditions to yield trans-3-formylbut-2-enenitrile (V), which can be easily distilled without contamination of the cis-isomer. The trans-3-formylbut-2-enenitrile can be selectively or exhaustively reduced to form either trans-3-hydroxymethylbut-2-enylamine (VI) or 3-hydroxymethylbutylamine (VII), which compounds can be condensed with 6-chloropurine (IX) by known methods to form trans-zeatin or dihydrozeatin respectively.

Description

6 ~ 4 BACKGROUND OF THE INVENTION
The present invention relates. to proces.ses for the production of key intermediates in the syntheti.c pathway leadi.ng to naturally occurring cytokini.ns, particularly ~an~-zeati:n and dihydrozeatin.
Cytokinins are naturally occurring 6-s.ubs.ti.tuted aminopurines, which are plant hormones known for th.eir biologi:cal acti:vity on plant growth. The more important of these effects are the ab.ilities to induce ce11 division and regulate di.fferentiation i.n exci.s.ed plant tissue. Zeatin is one of the most acti.ve forms of the cytok.i:nins.
: 10 Owing to the diffi:culty in i.solati:ng even mi;nute. quantiti.es of the naturally occurring cytoki.nins, researchers have turned their efforts to investigating synthetic path~ays to form bi:ologically acti.ve 6-substituted ami.nopuri.nes. For ~an~.-zeatin, much. of the research has been focus~ed on a suitable synthesi.s of ~a~-3-h~droxymethylbut-2-: 15 enylami.ne (VI.), th.e h.ighly functi;onali:zed unsaturated side chain ~hi.ch.can be condensed, by kno~n methods, wi.th 6-ch.lorQpuri`ne to fcrm ~ai~-: ~ zeati.n. In the case of dihydrozeati:n. (II~, th~ des-i:red si:de. chai:n i5 3-hydraxymethyl butyl ami.ne ( VI I ) .
; ~ Thu~ far, kna~n methods. for the synthesi`s of ~an~-3-hydroxymethylbut-2-enylamine fall into one of the following th.ree : categori.es:
1. A multi.step synthe~.is involvi.ng an allyli`c 6romi.natïon and starti.ng with tiglic acid, see. for example G. Shaw et alr,J. Ch.em. SQG. (C~, q21~ 1966, and D. S. Letham et al~,Phytochemi:stry, 1Q, 2n77, lq71;

2. A multi.~tep synthes:is. starting from acetone and cyanoaceti.c aci:d and i.nvolvi.ng allylic 6rominati`on, see. for example D. S. Letham et al., Aust. J. Chem., 22, 205, 1969; and

3. A Ga~riel ph.thali.mi:de synthesi:s far the selecti:ve allyli`c 3n ami.nation, s.tarti:ng from i:soprene or an isoprenoid hali.de, see for example M. Ohsugi: et al.,Agr. Biol. Chem., 38, 1925, 1974, R. Mornet et al.,Tetrahedron Lett., 167, 1~77, J. Corse et al.,Synthesi:s, 618, 1972, and G. Desvages et I 1 ~26~4 al., Bull. Soc. Chim. Fr., 3329, 1~6~.
Generally the methods of the fi:rst tWQ categori.es involve many s.teps:, provi`de low yields and requi:re the diffi:cult separati'on of geometri:c i.samers of the ~ unsaturated nitri:le necessarily produced by an allyli:c ~rami:nati:on. The thi'rd method involves an unstable dibromide intermediate and requires an undesirable recrystallizati:on step in order to separate an i.ntermedi`ate mi:xture.
By th.e method descri:'bed by D.S. Letham et al., Aust. J.
; ChRm., 22, 205, 196~, it i`s possi:b.le to selecti:vely reduce ~an~-3-h~droxymethylbut-2-enenitrile (VI~ ) to ~han~-3-hydroxymeth~lbut-2-enylami.ne (VI) using 2-tetrahydropyranyl as: a protect;ng group for the hydroxyl functi:on. It mus.t be poi.nted out that, on the basis of the pres.ent i.nventars expe.ri:ences, li.thium alumi:num hydride reduction of ; the 2-tetrahydropyranyl ether of hydroxyni:trile (VIII), after the pro-cedure of Leth.am et al., invari:ably leads- to a complex mixture of products, ;.n ~hi.ch saturated ami:nes ~re ~found to be main const;tuents.
Othe~wi.se the hydroxynitri.le (VII:I~ can be exhaustively reduced to form 3-hydroxymethylbutylami.ne (VII~. Condensat;on of the unsaturated or saturated ami.ne w;th 6-chloropuri.ne, by for example the method of Leth.am et al., i.s kno~n to yield the ~n~-zeati.n or dihydrozeatin.
I:t is therefore an object of the present invention to provi.de processes for the preparati:on of the unsaturated amino alcohol (VI) and the saturated ami.no alcoh.ol (VII:~, whi:ch can th.en be transformed by known methods to ~n~.-zeatin and dihydrozeatin respecti'vely.
Tn a process descri:bed i:n Uni:ted States Patent No. 3,960,923 is.sued to De5imone, ~ unsaturated nitri:les can be prepared by reaction of a ketone and acetonitri:le i:n the presence of a base. However, as reported i.n a publi.cati:on b.y S. A. Di:Bi:ase et al., (Synthesis, 629, 1977) and as s:upported by th.e poor yield data presented by De5imone, this base catalyzed condensati:an fails or gives ver~ poar yi:elds wi:th methyl ketones.
Si:nce th.e starti:ng materi:al for the synthesis of the desired unsaturated ami.no alcohol ('VI) or saturated am;:no alcohol (~ , using the reactian scheme of DeSi.mone, i.s necessarily a methyl ketone, this process ~ 1 ~2~

is not feasi~le for an unact;vated methyl ketone.
SUMMARY'OF'TH NVENTI.ON
The inventor has di:scovered that whereas a methyl ketone such as acetone cannot be sati:sfactorily condensed wi:th acetoni.~ri:le, an acetal of pyruvaldehyde, whi.ch i.s: an a~a-di:alkoxymethyl s.ubstituted methyl ketone, can be condensed with acètoni.tr;:le with surprisi.ngly good results. The pres:ence of the acètal group h.as a number of unexpected effects on the condensati`on reacti`on.
Fi:rstly, the fact that tfie reaction proceeds at all suggests that the acetal group has a stabili:z;ng effect on the polari:zing carbonyl group of the pyruvaldefiyde acetal. Secondly, the acetal group, being a hulky s.uhs.t;:tuent, i:nfluences: the stereocfiemistry of the condensation react;on to gi:ve the preferred ~an~- configurati:on in good yield. Thirdly, th.e acetal group can be hydrolyzed and sufisequently reduced to yield the 3-hydroxymethyl functi:anality i.n the target intermediates for the synthesis of ~an~.-zeati.n CI:) and dihydrozeatin (II)~ namely t~a~-3-hydroxymethylbut-2-enylamine ~VI.~ and 3-hydroxymethylbutylamine (:VII) respectively.
Thus, in accordance wi:th the present invention a pyruvaldehyde acetal i.s condensed with acetonitrile in the presence of a strong base.
Preferably th.e hase i.s s;elected from tfie group consi:sti'ng of an alkali alkoxide, sodi:um hydroxi.de and potassi'um hydroxide. The reaction proceeds at elevated temperatures;, preferably at reflux temperature. Excess acetoni:tr;:le is preferably i.ncluded as the solvent for the reaction.
The base catalyzed condensati:on product formed is a c~d, ~han~
mi.xture of th.e correspond;.ng acetal of 3-formylbut-2-eneni'trile (IV), a novel product. As mentioned previ:ous1y, the reaction proceeds regio-selecti.vely favouring the ~han~-i:somer.
Broadly stated, the present invention provides a process compri.s.ing: reacti.ng a pyruvaldehyde acetal having the general formula ~ ~72~

R~ /R
Q \ /0 CH
Cl w o where. R ;s an alkyl, cycloalkyl, s~ub.sti`tuted al:kyl~ or alkenyl group havi.ng from ab.out l - lO carhon atoms~ or part of a methylene chain in a cyclic acetal of a 5 - lO membered r~ng structure, with acetonitrile i:n the presence of a strong 6ase at an e1evated temperature for a time su~fi;ci.ent to form an acetal af 3-formylbut-2-enenitrile.
I:n a further aspect of tfie invention~ the acetal of 3-formylbut-2;e.nenitri.1e ~IV) i.s acid hydrolyzed, pre~e.rably i:n an aqueous solution ; of a mine.ral acid or in an aqueous solu~ion of a m;neral acid with a water mi.sclble organi:c solvent sucn. as me~hanol, ethanol, or tetrahydrofuran, to 1.0 form ~an~-3-formylbut-2-eneni.tri:1e ~V~, also a novel product. Dis-:

- 4a -~ ~ 7264~

tillation of the crude product yi.elds~ exclus;:vely the ~a~-i.somer, eli.mi.nati.ng the need for separating geometric i.somers at any poi:nt ;n the overall syntheti:c path~ay. Pres.umabl~ the c,~.-is.omer, i.f formed, ~urth.er reacts i:ntra- or i:ntermole.cularly to form a h~:gher boili.ng fraction.
The aldehyde group of ~an~-3-form~lbut-2-enenitri:le can be selectively reduced i:n the presence of sodi.um ~orohy-dri.de in the presence of a suita~le s:olvent s:uch as: met~lanol, eth.anol, or tetrahydrofuran to yi:eld ~on~-.-3-hydroxymethylbut-2-enenitrile (VII:I~ wi.thout csntamination of th.e saturated alcohol. By mask.i:ng the allyl;:c hydroxy function ~ith a bulk~ s.ubsti:tuent~ th.e ~ unsaturated n:i:tri.le (VIII~ can be further selectively reduced with lithi.um alumi:num hydride to form ~on~-3-hydroxymethylbut-2-enylam;ne (VI). A number of bulky sily groups, such as ~Qh~-butyldi.methylsilyl group, see for example E.J. Corey et al., J. Am.
Chem. Soc., 94, 6190, 1972, have been found to be effective for this pur-pose. The uns.aturated ami.no alcohol ~VI:) thus formed can be subsequently : condens.ed wi.th 6-chloropurine (.IX~ to form ~La~-zeati.n by known methods, s.ee for example G. ~haw et al.
Other~i.se, the ~on~-3-formyl~ut-2-enenitrile (V~ can be 2a exh.austively reduced wi:th a sui.table reducing agent to form 3-h~droxymethylbutylami:ne (VI:I.). The saturated amino alcoh.ol (VII~ can be condensed wi.th 6~chloropuri:ne, for example, by the method of Letham et al.
to form dihydrozeati.n (II). Preferably the reduci:ng agent is a metal hydri.de-transi~tion metal s.alt system such as NaBH4-CoC12 6H20, although a metal hydri.de such as Li.AlH4, or a s.ui:table hydrogenation catalyst in the presence of hydrogen can also be employed.
DESCRI.PTION OF THE DRAWI:NGS
Fi.gure 1 i.s a formula s.heet showi:ng structural formulas and names for the compounds referred to in the speci:fication, and Fi:gure 2 i.s a reacti.on sheet s.h.owi:ng an example of the reac~ions descri~ed i:n the s.pec;:fi:cati:on.

~ 77~6~

DESCRIPTION OF THE PREFERRED EMBODIMENT
The present invention involves a base catalyzed condensation of a pyruvaldehyde acetal with acetonitrile to form the corresponding acetal of 3-formylbut-2-enen;trile, which is subsequently hydrolyzed in acidic conditions to form ~a~-3-formylbut-2-enenitrile. Figure 2 shows the reactions involved in the present invention starting from pyruvaldehyde dimethyl acetal.
A large number of pyruvaldehyde acetals are s.uitable as a starting material for the process. Generally the structural formula for the pyruvaldehyde acetal i5 R R

\C/
C = Q

~here R i.s. an alk~l, cycloalkyl, su~stituted alkyl or alken~l group, having from a~out 1 - 10 carhon atoms, or R i5 a methylene group in a polymethylene chai.n -(:CH2~n- i:n a cycli`c acetal having a 5 - 1~ memb.ered ri.ng structure. Thus the general formula as given i:s meant to represent both. substituted acetals, such as~ pyruvaldehyde dimethyl acetal, and cyclic acetals such as 2-acet~yl-1,3-di.oxan.
A number of pyruva'ldehyde acetals- are commerci:ally avai.lable, however they can also be prepared by reacti.'ng 1,3-dihydroxy-2-propanone wi.th monohydri:c or di:hydri.c alkanols 6y, ~or example, the methqd of S. K.. Gupta, J. Qrg. Chem., 41, 2642, 1~76'.
The condensation reacti:on proceeds wi:th stoi:chi:ometri.c quanti.ties. of acetoni.tri:le and the pyruvaldehyde acetall however i:t i.s usually desi.rab.le to use a large excess of acetoni:tri'le to act as- both a reagent and a solvent. The excess- acetonitrile can be recovered at the end of the reacti.on and recycted. I'ncreas:ed ~yi'elds are generally noted \
' 1 726~4 when acetonitrile is. used in large exces.s..
The base us:ed ta catalyze the reacti.on i;s. ~ell knawn i.n con-densation reactions. Preferably the base is an alkali: metal alkoxide such as sodium or potassium methoxi.de or s.odium t-butoxi:de. Mos.t preferably the alkali. alkoxi:de carri:es the same alk~l group as the R group in the pyruvaldehyde acetal. For i`nstance, sodi.um or pokassi:um methaxi.de are preferred bases for the condensation of acetonitri:le ~ith pyruvaldehyde dimethyl acetal.
Sodium or potassi.um hydroxi.de ma~ be us:ed i:n lieu of the lO alkoxide, however the yi.eld of th.e hydro~ide catalyzed reacti.on i:s aften inferior.
The amount of base us.ed i.s. prefe.rab.ly i:n th.e range of 0.2 to 1.2 moles. per mole of pyruvaldehyde acetal. Th.e amount of bas-e. used has no signi.ficant effect on the final ~i:eld of the a,~-uns:aturated nitrile, 15 however the rate of addi.ti.on of th.e cyanometh.ylene anion to the p~ruvaldeh~de acetal and th.e suhsequent dehydrati:on to gi:ve. the ~ unsaturated nitrile i.ncreases wi.th i.ncreasi.ng quanti:ti.es! of base. For example, ~hen pyruvaldehy-de dimethyl acetal was allowed to react ~ith. acetoni:trile i.n th.e presence of one equivalent of sodium meth.oxi:de at reflux tempera.ture ~or about 2 hours 20 the product mi.xture yi.elded ahout 25~ of the te.rti~ary alcohol and 4~% ofunsaturated nitri.le. The yi:eld af the alcohol ~as found to decrease with both an i.ncreas.e i.n the amount of base, as- ~ell as ~ith an increase in the reactian ti.me.
: The base may be fi.rst i:ntroduced i:n the acetonitri:le and the 25 mi.xture subsequentl~ brought up ta reflux temperature. A solution Qf the pyruvaldeh~de acetal i.n acètonitrile i.s then added dropwise over about 2 - 3 hours and the reacti.on mi.xture refluxed for ahout 3 - 5 hours longer. Alternati:vely, the pyruvaldehyde acetal, base and acetanitri.le may be charged i.n one reacti:on vessel and whole refluxed until the completi.on of the reacti.on.

! 1 7~6~1 The reaction time has been found to be largely variable with the R-substituent of the pyruvaldehyde acetal. The addition oF
the cyanomethylene anion, which is generated in situ, to the pyruvaldehyde acetal may be completed within about one hour, however an extended reaction time of as much as 10 hours may be required to ensure the dehydration of the primary adduct to give the desi:red ~,~-unsaturated nitrile. Generally, a pyruvaldehyde acetal which carries a bulky acetal substituent, such as diisopropylacetal or di.oxan, requires a longer time for the dehydration of the primary adduct.
At the completion of the reacti:an, the res:ultant m;.xture i.s washed with water to remove the major;:ty of base and acetami:de, a byproduct resulting from the hydrati:on of acetoni:tri.le i:n the presence of base. Usually, the amount oF water used for ~as.hi.ng should be less than one tenth of the volume of acetoni:tri:le us.ed for the reacti:on sa that the organi.c phase remai.ns immiscible wi:th the aqueous pKase.. Th.e organic phase may b.e washed agai:n ~i.th wate~. The combi;ned e.xtract is back-~ashed wi.th lo~.-boi:li:ng organi:c solvent tQ gi`ve addi`ti.onal crop of unsaturated ni.trile. Evaporati:on of tKe s.olvent under reduced press.ure gi.ves a crude li.qui:d product. Th.e comb.ined crude product i.5 distilled at reduced pressure to gi:ve a c4~,~nan~ miXture of 3-formylbut-2-enenitrile acetal (.I.V). The se~aration of th.e i:someri:c mi`xture, which contains about 10 - 15% of C4~. form is nat necess:ary since. the subs.equent transFormati:on of the acetal to the aldehy-de produces ~nan~-3-formylbut-2 eneni.trile C~) as the mai;n product,wh.ich can he eas~ily di.stilled wi:thout contami:nati:on of i:ts; c~-isomer.
The above formed i.someri:c mixture of the acetal of 3-farmylbut-2-eneni.tri:le (I:V) can be hydrolyzed under acidi`c conditions to remove the acetal group. The aci.di:c sQlution us.ed i`s: generally an aqueous s.olut~on of a mineral or strong organi:c acid. Alternately the acidi:c solution can be an aqueous solut;on of a mineral or suitable strong organi.c acid together ~ith a water mi:scible organic solvent to solubilize I 1 726 ~ ~
the start~ng material~ A 5% perchloric'aci:d solub~on has been found to be sui:table for the hydrolys.is reaction. Other suitable acids include hydrochl.oric~ sulphuri:c and oxalic acid Acid hydrolysis will proceed at ambient temperatures, however the reaction time may be reduced from about 10 hours to about one hour by hydrolyzing at a reflux temperature.
' The product from the acid hydrolysis is extracted with low-boiling organic solvent, such as methylene chloride, to give crude hydrolyzed products. The organic extract ;s dried with an appropri.ate drying agent 10- and then concentrated to give crude aldehyde which can be distilled either at atmospheric pressure (182 - 187C at Edmonton, Alberta, Canada~
or at reduced pressure and a low temperature.
The novel aldehyde of thi.s i.nvention, namely ~an~-3-formylbut-2-enenitrile (V), has been prepared and shown to have the following phys.ical parameters:
b.p. 71/11 mm Hg; NMR (60 Mi-lz TM~ as internal standard i.n CDC13 solventl ~ 9.73 (singlet, lil), ~ 6.40 (quartet, J =
1.5 Hz, lH) and ~ 2.10 (doublet, J = 1.5 Hz, 3Hl, IR (liqui'd fi;lm) 2843, 2730, 2220, 17Q5, 135a, 1195, 1027 and 830 cm 1.
20. 2,4-dini.trophenylhydrazone deri.vative, m.p. 276 (decomp.~;
bis(~hon~-3-formylb.ut-2-enenitri.le)hydrazoneg m.p. 160 -161C. The li.qui.d aldehyde and the two crystalli'ne hydrazone derivati.ves gave correct elemental analys.i`s and mass. spectral data.
The aldehyde of this i.nventi:on can be selectively reduced with sodi.um borohydri.de i.n the presence of'a low-boi'li:ng alcohol as solvent, such as methanol or ethanol, to gi:ve ~a~-3-hydroxymethylbut-2-eneni.tri.le (VIII~ wi'thout contaminati'on of a saturated alcohol resulting from reducti:on of the conjugated carbon-carbon double bond. The fact 30- that the conjugated carbon-carbon bond remai'ns. i.'ntact with Na~H4 reducti:on sugges.ts that keto-enol tautomeris.m i.n tiie c~,~-unsaturatecl aldehyde is i.nsi:gni:ficant. Th.ïs i.s clearly due to its extended conjugatlon ~i.th the ~ ~ 72~4~

nitrile group.
The ~nan~-3-hydroxymethylbut-2-enenitrile (VIII~ formed by this selecti:Ye reduction ;s a natural product Found as the alcohol moiety of many plant li:pids-.
The hydroxyn;`trile (VIrr) thus ~ormed can be further trans-formed to ~a~--3-hydroxyme~hylbut-2-enrlami`ne (VI) in very low yield by the method of Letham et al. Alternati'vel~, it has been found that a number of s;lyl ethers of ~an~-3-hydroxymetnylbut-2-enenitr;le (VIII) can be smoothly transformed to the unsaturated amine (VI) ~ith a LiAlH4 reduction in the presence of diethyl ether solvent. The bulky silyl groups are able to mask the allylîc fiydroxyl function i'n the unsaturated nitrile (VIII) from the metal hydride complexati'on and thus prevent hydride reduction of the carbon-carbon double bond. Suitable silyl reagents include t-butyld;'methyls;lyl hali'des and a range of tri-alkylated silyl halides ~15 which can protect the hydrox~l function from metal chelation and resist hydride reduct;on. The method for their use is disclosed by E. J. Corey et al., J. Am. Chem. Soc., '~4, 6190, 1972. The protective silyl group of the protected nitrile (X~ is removed by ac;d hydrolysis, preferably in a solut;on consist;ng of 2 normal equivalents of an aqueous mineral acid and 20~ water misci~le organi'c solvent, such as methanol or tetrahydrofuran. The hydrolysate is ~ashed ~i'th ether or a low-boiling water immiscible organic ` ; solvent. The product unsaturated am;ne ~VI) ~s recovered from the aqueous solut;on as a salt of an oxo acid or hydrohalide, such as a sulfate or chloride.
t is also possible in accordance ~;th the invent;on to reduce the ~ unsaturated aldehyde (V) to (+~ -3-hydroxymethylbutylamine (VII~ w;th a metal hydride-trans;tion metal salt system or with an appropr;a~e metal hydride, or ~ith the uptake of four moles of hydrogen by means of catalyt;c hydrogenat;on. The reduct;on method used in the invention ;.s prefera~ly a m;xed sodium borohydr;de-metal salt system, such as NaBH4 CoC12-6H2Q, ;n the presence of a low boiling alcohol.

- lû -~ ~ ~2~4~1 The reduction can be performed by adding five to ten moles excess of sodium borohydride to a 1:1 mole equivalent mixture of cobaltous chloride hexahydrate and the ~ unsaturated aldehyde. Preferably however, the aldehyde is first reduced with one mole equivalent of the metal hydride and then one mole equivalent of CoC12-6H20 is introduced, followed by adding progressively a large excess of sodium borohydride.
The use of this metal hydride-transition metal salt reducing agent has been documented by T.Satoh et al., Tetrahedron Lett., 52, 4555, 1969.
The ~na~-3-hydroxymethylbut-2-enylamine (VI) or its salt (XI) and the 3-hydroxymethylbutyl amine (VII) can be condensed with 6-chloropurine (IX) to form ~o~-zeatin (I) and dihydrozeatin CII~
respectively. Experimental procedures and conditions have been ~ell documented in the literature and thus wi;ll not be disclosed herein. See for example Letham et al., Aust. J. Chem. 22, 205, 1~69.
It should be pointed out that the 3-hydroxymethylhutylamine (VII) formed by the above described proce~s has an asymmetric centre.
Thus the dihydrazeatin formed therefrom is a racemic mixture. The production of racemic dihydrozeatin and opti:cally pure enantiamers has been reported by T. Fujii et al., Tetrahedron Lett., 30, 3a75, 1~72.
The following examples further illustrate the nature of the inventian and the manner of practicing the ~ame:
Example 1. 3-~Formyl~ut-2-enenitri:le dimethylacetal A mixture of 6~0 ml of acetonitrile and 23 g of s-odi`um methoxide was~ heated to reflux temperature under nitrogen purge and mechanical stirring. To the a~ove mixture was- added drap~i~se, through a pressure equalizing addition funnel9 50 g af pyruvaldehyde di`methylacetal in 200 ml of acetonltrile aver a peri:od of tfiree hours. At the campletion of addition, the whole mixture ~as heated to re~lux temperature for a further five hours and then allo~ed ta caol to room temperature. The resultant mixture was then shaken t~ice with a ~0 ml porti`on of ~ater 1 1 _ ~ ~2~44 to remove most of the base. The acetonitrile solution was separated and concentrated at 40C under reduced pressure of about 15 mm Hg to give crude ~ unsaturated nitrile. The combined aqueous extract was backwashed twice with a 50 ml portion of ether to recover an additional crop of the product. The combined crude product was distilled under reduced pressure of 0.05 to 0.5 mm Hg at a distilling head temperature from 32 to 70C yielding 41.8 g of ~,~-unsaturated ni.tri.le (7Q% yi.eld).
This product was a colorless liqui.d which consists of about 11% of c~-3-formylbut-2-enenitrile dimeth~'lacetal and 89% of ~hàK~-isomer, as analyzed~hy gas chromatography.

; Example 2. 3-(1,3-Di.oxan-2-yl)but-2-eneni`trile.
A mi.xture of 4.2 g of sodium me.thoxi:de and 120. ml of aceton1.trile was stirred mechani.call~ and heated ta reflux under nitrogen purge. Tv. thi.s was then.added dropwi;s-e 10 ~ of 2-acetyl-1,3-dioxan in 50 ml of acetoni.tri.le aver a periPd of 2 hours and followed hy a 10 hour reflux peri.od. After cooling, the reacti:on mixture was~ washed twice with a 10. ml portion of water. The organic phase was then concentrated under reduced pressure at about 4QC to give. the crude ~ unsaturated nitri.le.
The aqueous phase was washed twice ~i.th. 20 ml pQrtiOnS af ether and the extracts. combined, dried, and evaporated to gi've an addi:tional crop : of the product. The comb.i.ned crude prQduct ~a.s disti`lled under a re-duced pressure of G.2 mm Hg and collected i`:n a ~oi~ ng range of 72 to 78C, gi.vi.ng 7.~4 g (65% yi.eld) of 3-(1,3-diPxan-2-yl~but-2-enenitrile : as a colorless li:qui.d. Both NMR and GLC analyses of the above disti.llate s.howed that the liqui'd consi.sted o~ 88~ of tha~ and 12,~
of the c~s-~isomer of 3-C1,3-di.axan-2-yl)but 2-eneni:tri.le.
I:n a separate experiment in whi:ch the reflux period ~as s.hortened to 5 hours after addi.tion of the acetylacetal, the primary adduct , 3-(~1,3-di:oxan-2-yl~-3-hydroxyhutyronitri:le, b.p. 92-lUQC/0.2 3Q mm Hg, was i:s.olated i:n about 25% yield.

g l 7264`~1 Example 3. 3-Formylbut-2-enenitrile dimethylacetal As in example 1, acetals of 3-formylbut-2-enenitrile can also be prepared by directly heating a mixture of an acetal of pyruvaldehyde and sodium methoxide in a large excess of acetonitrile. Thus a mixture of pyruvaldehyde dimethylacetal (50 g), sodium methoxide (11.5 g, 0.5 equivalent) and acetonitrile (600 ml) was heated to reflux temperature for 12 hours. After cooling, the resultant mixture was worked up in a manner similar to that described in example 19 giving 40 g of 3-formylbut-2-enenitrile dimethylacetal.
The results obtained from the condensation of other pyruvaldehyde acetals ~ith acetonitrile are summarized in Table I.

~ - -\

l 1~2~

E r .
oo ~ O c~l O O
I -- LS~ C~
E ~ D r~
Q~ _-o N C~ O

.
~ ~ .,~
a) ~o co Cff ~ U~ 00 ~ ~ ~ $
~S . . r-- N ~ 00 C~J
`

a) ~ z~
CC t_) :iZ V V
:a.. ) ~ c~ T ~ t_~
,.~ ~ 11 ~7 ~ r) V
~ I/ \~ ~~) ~ ~ I A
: ~> ~ N 6.~ \
.~ ,_ O O ~3 0 ~ O L~

'. ,,,1a~ ~ V
C~ ~, _~
CC ~< U) l_ .f O
r- 00 U~ ~D o o~
~ ~ ,~
~:IJ ~ E
a:-~

~ U~
~ a~

t_ ~ IA
a~ ~ ~
~ > .~
r_ r ~ r~ O r~ r_ r c~ ~
L~l o a)z o a~ I I I I
~ ~ O O O
~) ~0~ V
~ -- ~ N ~1 \~
-- ~ N
~_ o ~o O~O
3 CC ~r) I Cl. I I
~_ T C~l I i I
~1 ~ C~ ~ ~ I
C~ ~
V

~ 1 726~`~

Example 4. ~a~-3-Formylhut~2-enenitri~le (V) FQrt~-three grams of 3-formylhut-2-enen;tr;le dimethylaceta'l prepared according to Example 1 or 3 (`described above) was d;ssolved in a methanolic hydrochloric acid solut;on, prepared by mixing 80 ml of methanol, 500 ml water, and 26 ml of 12N nydrochloric acid. The mixture ~as heated to reflux for 1 hour. After cool;'ng, the mïx~ure was extracted w;th three 300 ml portions of methylene cfilor;~de. The organ;c layer ~as separated and ~ashed ~itfi br;ne, dr;`ed over Na2S04 and evaporated to remove the solvent. The residue was disti`lled at 70 - 73C at a re-0 duced pre~sure of 11 mm Hy to give 23.5 g (81% y;eld) of ~an~-3-formylbut-2-enenitr;le. The ~an~-aldehyde was free from the c~-isomer as evidenced from an NMR analysi's. The isolated y;eld of pure ~an~-a,~-unsaturated aldehyde~ based on its precursor ~n~-dimethoxymethyl~ut-2-enen;trile, was 91%.

Example 5. ~an~-3-Hydroxymethylbut--2-enen;trile ~VIII) To a solution of 3.2 g of ~an~-3-formyl-2-enenitrile in 80 ml of ethanol, cooled în an ice ba~h and sti'rred wi`th a magnetic stirrer, wa~ added 1.28 9 of NaBH4 in port;ons over a period of 5 m;nutes. The m~xture was then kept st;rred in the ;ce bath for an add;tional 10 minutes and subsequently decomposed w;th 100 ml water. The aqueous ethanol solution was extracted four times wi't~ 100 ml portions of ether. The combined ethereal extract ~as ~ashed with 200 ml of ~ater to remove most of the ethanol. The orsanic phase ~as then dried and evaporated to re-move the solvent to give practically pure ~an~-3-hydroxymethylbut-2-enen;tr;le ~VIII~ (2.~4 9, ~0~ yield). An analytical sample purified by dist;llat;on (72 - 73C/0.15 mm Hg, 68 - 71C/0.25 mm Hg) was identical to that reported in the literature (Letham et al.).

Example 6. 3-Hydroxymethylbutylamine ~VII) To a solution of 5 9 of ~han~-3-formyl-2-enell;trile in 100 ml of methanol coaled in an ice ~ath and stirred with a magnetic stirrer, was I ~726A4 added 2 g of NaBH4 over 5 minutes. After stirring for an additional 10 minutes in the ice bath, 12.6 9 of CoC12 6tI20 was added and left to stir for 5 minutes. To the above mixture was added slowly 8.1 g of NaBH4 over a period of 15 minutes. When the addition was complete, the reaction mixture was left to stir for 2 hours before decomposing ~ith 60 ml of 6 N
hydrochloric acid. The resultant mixture was then basified with sodium hydroxide pellets to pH 12 and extracted four times with 150 ml portions of methylene chloride. The combined extracts ~ere dried over K2C03 and evaporated to remove the solvent, giving 4.Q g of crude 3-hydroxy-methylbutylamine (74% yield). The crude amino alcohol had an 60 HzNMR spectrum in CDC13 of ~O.g (doublet, J=6 Hz, 3H), ~1.5 (multiplet, 3H), ~2.7 (multiplet, 2H), ~3.4 (doublet, J=5.5 Hz, 2Hl and a D20 exchangeable singlet at around ~3.5 (3H). The product was sufficiently pure for the synthesis of racemic dihydrozeatin by the method of Fujii et al. An additional amount (430 mg~ of the crude amino alcohol product, which was less pure than that isolated above, ~as o6tained by continuous extraction of the aqueous solution with ether or methylene chloride.

Example 7. ~a~-3-Hydroxymethyl-2-enenylami;ne sulfate (XI:~
The compound ~a~6-3-h~droxymethylbut-2-enenitrile ~-butyldimethylsilyl ether (X) was; prepared in quanti:tative yield from ~an~-3-hydroxymethyl~ut-2-eneni~rile and ~h~-but~ldi:methylsilyl chloride by the method oF E. J. Corey and A. Venkateswarlu, J. Amer. Chem.
Soc. 94, 6190, 1972. The compound X had the following physical parameters:
b.p. 72 - 74C/0005 mm Hg, NMR (80 MHz, TMS as an internal standard in CDC13solvent) ~0.07 (singlet, 6H), ~Q.~2 (singlet, 9H), ~1.96 (doublet of triplet, J = 1.1 Hz, J = 0.85 Hz, 3H)~ ~4.15 (doublet of quartets, J =
2 Hz, J = 0.85 Hz, 2H), ~5.53 (eight-line multiplet, J = 2 Hz, J = 1.1 Hz), IR (liquid film) 2955, 2930, 2220, 16809 1255, 1120, 840~ and 780 cm 1.
To a solution of 3 g oF the above nitrile (X) in 100 ml of `~ 17~6~4 anhydrous ether, cooled and stirred in an ice bath, was added 810 mg of LiAlH4 over 5 minutes. The mixture was stirred 1 hour in an ice bath, 1 hour at room temperature and 1 hour at reflux temperature. The resultant complex was allowed to cool, was decomposed with 2.5 ml of 5 water and then treated with MgS04 to aid coagulation of the solid. The solid was removed by filtration and washed thoroughly with 100 ml of ether. After evaporating the solvent, 2.6 9 oF a light yellow liquid product were obtained. Gas chromatographic analysis of the crude product showed it to contain 70% of the desired unsaturated amino alcohol silyl 0 ether. A pure colorless sample obtained by distillation (64 - 67C/0.05 mm Hg) had an NMR spectrum (80 Hz in CDC13 solvent) of ~0.06 (~singlet, 6H~, ~0.93 (singlet, 9H), ~1.62 (singlet with fine splitting J = 1.4 Hz, 3H~, ôl.84 (broad singlet, D20 exchangeable, 2tl), ~3.20. (doublet, J = 6.8 Hz with fine splitting), ~4.05 (singlet with fine splitting, 2H~, ~5.56 15 (triplet of quartet, J = 6.8 Hz, J = 1.4 Hz, lH)i an IR (liqui:d fi:lml of 2955, 2q30, 2860, 1670, 1550, 1473, 1465, 134a, 1363, 1255,1a80, lOn5, 940, 835, and 773 cm 1.
A quantity of 1.8 9 of tfie above p~epared crude amino alcohol was dissolved in 20 ml of methanolic s;ulfuri`c acid (5~ v~v lN H2S04 in 20 methanol) and stirred at room temperature ~or 4 hours. The resultant mixture was washed twice with 30 ml porti:on~ o~ ether. The aqueous layer was separated and evaporated under reduced pressure at 50C to yield 1.3 9 of a light bro~n syrup. The liquid product had an NMR spectru~
(80 Hz in D20) af ~1.74 (;singlet, 3H~, ~3.7 (doublet, J = 7.5 Hz, 2H~, 25 ~4.0 (singlet, 2H),ô5.5 (triplet, 7.5 Hz).
The sulfate (XI) could not be easily obtained in a crystalline form, however, it can be used directly for the synthesis of zeatin according to the method of G.Shaw et al.,J. Chem. Soc. (c), ~21, lq66.

26~

It should be realized that although the present invention has been disclosed with an objective of providing a synthetic pathway to zeatin, Gne skilled in the art will recogn;ze that the *~a~-3-formylbut-2-enenitrile (V) product formed herein is a isoprenoid which may have further utility in the synthesis of other natural products or useful chemical building blocks.
Furthermore~ while the present invention has been disclosed in connection with the preferred embodiments. thereof, it should be under-stood that there may be other embodiments whi:ch Fall within the spirit and s.cope of the present i.nvention as. defi:ned by the follo~ing claims.

~ '

Claims (22)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. The process comprising:
reacting a pyruvaldehyde acetal having the general formula where R is an alkyl, cycloalkyl, substituted alkyl, or alkenyl group having from about 1 - 10 carbon atoms, or part of a methylene chain in a cyclic acetal of a 5 - 10 membered ring structure, with acetonitrile in the presence of a strong base at an elevated temperature for a time sufficient to form an acetal of 3-formylbut-2-enenitrile.

2. The process as set forth in claim 1 wherein:
the pyruvaldehyde acetal is a pyruvaldehyde dialkylacetal.

3. The process as set forth in claim 1 wherein:
the pyruvaldehyde acetal is pyruvaldehyde dimethylacetal.

4. The process as set forth in claim 1 wherein:
the pyruvaldehyde acetal is pyruvaldehyde diethylacetal.

5. The process as set forth in claim 1 wherein:
the pyruvaldehyde acetal is pyruvaldehyde diisopropylacetal.

6. The process as set forth in claim 1 wherein:
the pyruvaldehyde acetal is 2-acetyl-1,3-dioxacycloalkane.

7. The process as set forth in claim 1 wherein:
the pyruvaldehyde acetal is 2-acetyl-1,3-dioxan.

8. The process as set forth in claim 1 wherein the base is selected from the group consisting of an alkali alkoxide, sodium hydroxide, potassium hydroxide; and the reaction is carried out in excess acetonitrile at reflux temperature.

9. The process as set forth in claim 1, further comprising:
hydrolyzing the acetal of 3-formylbut-2-enenitrile in an acidic solution selected from the group consisting of ar aqueous mineral acid, an aqueous strong organic acid, an aqueous mineral acid and a water miscible organic solvent9 and an aqueous strong organic acid and a water miscible organic solvent to form a product mixture containing trans-3-formylbut-2-enenitrile.

10. The process as set forth in claim 9 wherein:
the acetal of 3-formylbut-2-enenitrile is a dialkyl acetal of 3-formylbut-2-enenitrile.

11. The process as set forth in claim 9 wherein:
the acetal of 3-formylbut-2-enenitrile is the dimethylacetal of 3-formylbut-2-enenitrile.

12. The process as set forth in claim 9 wherein:
the acetal of 3-formylbut-2-enenitrile is a cyclic acetal of 3-formylbut-2-enenitrile.

13. The process as set forth in claim 9 or 11 wherein the product mixture is distilled to separate trans-3-formylbut-2-enenitrile.

14. trans-3-Formylbut-2-enenitrile having the structure:

15. The process as set forth in claim 9 , further comprising:
selectively reducing the aldehyde group of trans-3-formylbut-2-enenitrile to yield trans-3-hydroxymethylbut-2-enenitrile.

16. The process as set forth in claim 9 , further comprising:
selectively reducing the aldehyde group of trans-3-formylbut-2-enenitrile with a metal hydride in a suitable solvent, to yield trans-3-hydroxymethylbut-2-enenitrile.

17. The process as set forth in claim 16 wherein the metal hydride is sodium borohydride included in about one mole equivalent and the solvent is selected from the group consisting of a low boiling alcohol and a water miscible ethereal solvent.

18. The process as set forth in claim 9 , further comprising:
exhaustively reducing trans-3-formylbut-2-enenitrile to form 3-hydroxymethylbutylamine by reacting the nitri1e in a suitable solvent with a reducing agent selected from the group consisting of a metal hydride-transition metal salt mixture, a metal hydride, and a hydrogenation catalyst in the presence of hydrogen.

19. The process as set forth in claim 18 wherein about 5 - 10 moles excess of sodium borohydride is added to a solution containing about 1:1 mole equivalent of cobaltous chloride hexahydrate and the trans-3-formylbut-2-enenitrile.

20. The process as set forth in claim 9, further comprising:
reducing the aldehyde group of trans-3-formylbut-2-enenitrile with about one mole equivalent of a metal hydride; and further reacting the resulting mixture with about one mole equivalent of cobaltous chloride hexahydrate in the presence of excess sodium borohydride to form 3-hydroxymethylbutylamine.

21. The process as set forth in claim 15 or 16, which further comprises:

reacting trans-3-hydroxymethylbut-2-enenitrile with a bulky tri-alkylated silyl halide to form a corresponding silyl ether of trans-3-hydroxymethylbut-2-enenitrile;
selectively reducing the nitrile group of the silyl ether thus formed with a metal hydride in a suitable solvent; and hydrolyzing the reduced silyl ether in a solution of a mineral acid and a water miscible organic solvent to form a salt of trans-3-hydroxymethylbut-2-enylamine.

22. The process as set forth in claim 15 or 16, which further comprises:
reacting trans-3-hydroxymethylbut-2-enenitrile with tert-butyldimethylsilyl chloride to form the corresponding silyl ether of trans-3-hydroxymethylbut-2-enenitrile, selectively reducing the nitrile group of the silyl ether thus formed with a metal hydride in a suitable solvent;
hydrolyzing the reduced silyl ether in a solution of a mineral acid and a water miscible organic solvent to form a salt of trans-3-hydroxymethylbut-2-enylamine.