A bromoarene based approach to phenylalanine analogues Hic and Nic

Article abstract of DOI:10.1055/s-1999-3088

A Heck reaction system incorporating Ph₄PCl and using bromoarenes as substrates has enabled the syntheses of the conformationally constrained phenylalanine analogues Hic and Nic to be dramatically shortened thus facilitating the production of significant quantities of these amino acids.

Full text of DOI:10.1055/s-1999-3088

954
LETTER
A Bromoarene Based Approach to Phenylalanine Analogues Hic and Nic
Susan E. Gibson (née Thomas),*a1 Jerome O. Jones,a Ray McCague,b Matthew J. Tozerc and Nicole J. Whitcombea
a Department of Chemistry, Imperial College of Science, Technology and Medicine, South Kensington, London SW7 2AY, UK
b Chirotech Technology Ltd, Cambridge Science Park, Milton Road, Cambridge CB4 4WE, UK
c James Black Foundation, 68 Half Moon Lane, London SE24 9JE, UK
E-mail: s.gibson@kcl.ac.uk
Received 18 January 1999
for the biological activity of compounds containing the
Abstract: A Heck reaction system incorporating Ph4PCl and using
amino acid residues.4
bromoarenes as substrates has enabled the syntheses of the
conformationally constrained phenylalanine analogues Hic and Nic
to be dramatically shortened thus facilitating the production of
significant quantities of these amino acids.
In order to fully exploit the potential of Sic, Hic and Nic
in a biological context, significant quantities of the amino
acids were required. We found that syntheses of multi-
gram quantities of Hic and Nic in particular were ham-
pered, however, by the number of steps in the original
approach, (nine and eleven respectively). We wish to re-
port herein that both Hic and Nic may be readily synthe-
sised in multigram quantities using six-step procedures
which rely on a versatile intermolecular Heck reaction and
a challenging bromoarene-based intramolecular Heck re-
action, successfully executed using a recently reported re-
action system.
Key words: amino acid synthesis, Heck reaction, phenylalanine
analogues, cyclisation, palladium catalysis
Conformationally constrained analogues of amino acids
are useful tools in the design of more selective and/or
more potent peptides as enzyme inhibitors, and agonists
or antagonists at receptors.2 We recently reported the first
syntheses of the conformationally constrained phenylala-
nine analogues 2,3,4,5-tetrahydro-1H-3-benzazepine-2-
carboxylic acid (Sic), 1,2,3,4,5,6-hexahydro-3-benzazo-
cine 2-carboxylic acid (Hic) and 2,3,4,5,6,7-hexahydro-
1H-3-benzazonine-2-carboxylic acid (Nic).3 We have
now started to use the amino acids [which form a series of
analogues with the well-established phenylalanine substi-
tute 1,2,3,4-tetrahydroisoquinoline-3-carboxylic acid
(Tic)] in biological studies, and have demonstrated that
the size of the restricting ring has important consequences
Our original approach to Hic and Nic started with four-
and six-step syntheses of iodoaldehydes 1 and 2. Reduc-
tive amination of 1 and 2 followed by nitrogen protection
and introduction of a carbon-carbon double bond gave io-
doarene substrates for Heck cyclisation. After some ex-
perimentation, conditions were found which gave good
yields of the eight- and nine-membered rings 3 and 4. Fi-
nally, hydrogenation and acid hydrolysis of 3 and 4 pro-
ceeded in essentially quantitative yield to give Hic.HCl
and Nic.HCl.
Synlett 1999, S1, 954–956 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
A Bromoarene Based Approach to Phenylalanine Analogues Hic and Nic
955
In order to shorten the original syntheses, we proposed to uct, were inconsistent with the transfer of aryl groups to
replace their early stages with palladium-catalysed cou- the alkene via Ph4PCl as proposed by Reetz and supported
plings between 1-bromo-2-iodobenzene and the three and by strong evidence from experiments with chloroarenes,8
four carbon units allyl alcohol and 3-buten-1-ol, reactions we decided to perform a control experiment without
based on sound literature precedent.5 Although these reac- Ph4PCl (Table, Entry 5). This gave essentially no cyclisa-
tions, if successful, would give rapid access to bromoalde- tion, indicating that there is a role for Ph4PCl in this reac-
hydes 5 and 6, this approach would introduce the non- tion albeit somewhat different from its role in the catalysis
trivial challenge of using bromoarene based intramolecu- of chloroarene-based Heck reactions.
lar Heck reactions to make the medium-sized rings.
Later in 1998, Reetz reported that Pd(OAc)2 in the pres-
In practise, the conversions of 1-bromo-2-iodobenzene to ence of N,N-dimethylglycine (DMG) is an effective cata-
the previously reported bromoaldehyde 56 and the novel lyst of Heck reactions of bromoarenes.10 Reaction of 8
bromoaldehyde 6 proceeded smoothly and were readily using a higher loading of Pd(OAc)2 (5 mol% rather than
performed using 20 g of the arene.7 (Interestingly, at- 1.5 mol%) but with otherwise identical conditions to those
tempts to react 4-penten-1-ol with 1,2-diiodobenzene reported to give a quantitative conversion of bromoben-
gave only returned starting material, an observation tenta- zene and styrene to Heck products, however, gave only re-
tively attributed to steric resistance to the initial palladium covered starting material (Table, Entry 6), thus
insertion step.) The bromoaldehydes 5 and 6 were then underlining the challenge associated with cyclising bro-
converted into the Heck substrates 7 and 8 using the same moarene 8 to the nine-membered ring 4. Finally, applica-
methodology as in the original synthesis and in very sim- tion of the Ph4PCl system to bromoarene 7 gave the
ilar yield.
desired cyclised product 3 in 58% yield, thus facilitating
the synthesis of Hic.
Heck reactions employing bromoarenes are generally
more difficult to effect than those using the corresponding In summary, the Ph4PCl based Heck system has allowed
iodides.8 It was thus not surprising to find that the condi- us to use bromoarenes in the synthesis of Hic and Nic,
tions used for the formation of 4 (in 86% yield) from its thus shortening the existing syntheses by three and five
iodoarene precursor gave a very low yield (4%) of product steps respectively. The role of Ph4PCl in these systems,
when applied to the bromoarene 8 (Table, Entry 1). Rais- which appears to be different to its role in chloroarene
ing the reaction temperature from 110 °C to 150 °C, in an based reactions, is under investigation.
attempt to force insertion of palladium into the carbon-
bromine bond only led to a lower recovery of starting ma-
terial (Table, Entry 2).
Acknowledgement
We thank Chirotech Technology (Studentship to N.J.W.), The
James Black Foundation and The Royal Society of Chemistry (Stu-
dentship to J.O.J.) for their support of this work.
In early 1998, Reetz reported that the use of palladium ac-
etate in the presence of Ph4PCl gives high catalytic activ-
ity with chloro- and bromoarene substrates.8 On applying
this approach to our substrate, we were delighted to find
References and Notes
that bromoarene 8 cyclised to the Nic precursor 4 in 57%
yield on a trial scale (Table, Entry 3) and in 69% yield for
a typical preparative reaction (Table, Entry 4).9 As our use
of 0.3 equiv. of Ph4PCl, the 69% yield of 4, and our failure
to detect phenyl-substituted 8 i.e. an aryl scrambling prod-
(1) New address: S.E. Gibson (née Thomas), Department of
Chemistry, King's College London, Strand, London WC2R
2LS.
(2) Gibson (née Thomas), S.E.; Guillo, N.; Tozer, M.J.
Tetrahedron 1999, 585.
Synlett 1999, S1, 954–956 ISSN 0936-5214 © Thieme Stuttgart · New York

956
S. E. Gibson et al.
LETTER
(3) Gibson (née Thomas), S.E.; Guillo, N.; Middleton, R.J.;
Thuilliez, A.; Tozer, M.J. J. Chem. Soc., Perkin Trans. 1 1997,
447.
(4) Gibson (née Thomas), S.E.; Guillo, N.; Kalindjian, S.B.;
Tozer, M.J. Bioorg. Med. Chem. Lett. 1997, 7, 1289.
(5) Larock, R.C.; Leung, W-Y.; Stolz-Dunn, S. Tetrahedron Lett.
1989, 30, 6629.
(6) a) Cooke, M.P.; Widener, R.K. J. Org. Chem. 1987, 52, 1381;
b) Wolfe, J.P.; Rennels, R.A.; Buchwald, S.L. Tetrahedron
1996, 52, 7525.
(7) In a typical procedure for the synthesis of 6, a Schlenk tube
was charged with 2-bromoiodobenzene (20.000 g, 70.69
mmol), 3-buten-1-ol (7.647 g, 106.04 mmol), sodium hydro-
gencarbonate (14.855 g, 176.83 mmol), tetra-n-butylammoni-
um chloride (19.696 g, 70.87 mmol) and palladium acetate
(0.317 g, 1.41 mmol, 2 mol%). Anhydrous DMF (70 cm3) was
added and the black reaction mixture saturated with argon.
The Schlenk tube was lowered into a preheated oil-bath at
40 °C and heated at this temperature, under an atmosphere of
argon, for 20.5 h. After this time, the solution was allowed to
cool to room temperature, diluted with diethyl ether (150 cm3)
and filtered. After evaporation of the solvents under reduced
pressure and preabsorption of the red/black oil on SiO2, co-
lumn chromatography (SiO2; hexane-ethyl acetate, 5:1) gave
aldehyde 6 as a yellow oil (13.944 g, 61.40 mmol, 87%);
(Found: C, 52.8; H, 5.0. C10H11BrO requires C, 52.88; H,
4.88%); nmax(neat)/cm-1 1722vs (C=O); dH (300 MHz)
(CDCl3) 1.90-2.00 (2 H, m, ArCH2CH2), 2.44 (2 H, t, J 7,
CH2CHO), 2.73 (2 H, t, J 8, ArCH2), 7.00-7.05 (1H, m, H-5),
7.16-7.23 (2 H, m, H-3, H-4), 7.50 (1 H, d, J 8, H-6), 9.72 (1
H, s, CHO); dC{1H} (75 MHz) (CDCl3) 21.7 (ArCH2CH2),
34.9 (ArCH2), 42.8 (CH2CHO), 124.4 (C-1), 127.3, 128.2,
130.2, (C-3, C-4, C-5), 132.6 (C-6) 140.3 (C-2) and 201.9
(CHO); m/z (CI, NH3) 262, 260 (MNH4+ + H2O - 2H, 100%),
246, 244 (MNH4, 40) and 184, 182 (M-CH2CHO-H, 41).
(8) Reetz, M.T.; Lohmer, G.; Schwickardi, R. Angew. Chem., Int.
Ed. Engl. 1998, 37, 481.
(9) In a typical reaction, a two-neck r.b. flask fitted with a conden-
ser was charged with thoroughly mixed palladium acetate
(0.0871 g, 0.388 mmol, 5 mol%) and tetraphenylphosphonium
chloride (0.819 g, 2.18 mmol, 30 mol%) followed by anhy-
drous sodium acetate (1.197 g, 14.60 mmol). The apparatus
was evacuated and purged with argon four times after which a
solution of bromoarene 8 (3.002 g, 7.28 mmol) in anhydrous
DMF (14 cm3) was pipetted into the reaction mixture followed
by a further 200 cm3 of anhydrous DMF. After saturating the
solution with argon, the flask was lowered into a silicone oil
bath preheated to 120 °C and the reaction mixture heated at
this temperature for 30 min under an atmosphere of argon be-
fore raising the temperature of the bath to 150 °C and heating
for a further 4.25 h when the reaction mixture turned black.
The reaction mixture was allowed to cool to room tempera-
ture, diluted with diethyl ether (150 cm3) and filtered through
a short pad of Kieselguhr. After evaporation of the solvents
under reduced pressure and preabsorption of the red/brown oil
on SiO2, column chromatography (SiO2; 40-60 petroleum
ether - diethyl ether, 2:1) gave the heterocycle 4 as a colourless
oil (1.653 g, 4.99 mmol, 69%). The physical data for 4 were
identical to previously reported data.2
(10) Reetz, M.T.; Westermann, E.; Lohmer, R.; Lohmer, G.
Tetrahedron Lett. 1998, 39, 8449.
Article Identifier:
1437-2096,E;1999,0,S1,0954,0956,ftx,en;W02399ST.pdf
Synlett 1999, S1, 954–956 ISSN 0936-5214 © Thieme Stuttgart · New York