Monodisperse liquid crystalline peptides

Article abstract of DOI:10.1039/a701084b

Solid phase peptide synthesis (SPPS) has been used as a synthetic tool to prepare monodisperse mesogen-oligopeptide conjugates with novel molecular architectures. Building upon previous success in this area, we have extended our methodologies to include a mesogenically substituted L-glutamic acid residue, and incorporated this derivative, as well as the L-lysine derivative previously reported, into a variety of new structures. Some novel liquid crystalline materials have been discovered. Both α-helical and β-sheet secondary structures have been explored as alternative scaffolds for the pendant mesogenic groups. The helical systems show most promise in terms of their ease of synthesis and handling, but at the relatively lower levels of mesogenic substitution studied so far, compared to the previously reported homo-oligopeptide materials, mesophases are not observed. The β-sheet materials prepared are rather insoluble and too high melting to have any practical value as thermotropic materials without further manipulation of their structures.

Full text of DOI:10.1039/a701084b

Monodisperse liquid crystalline peptides
Peter A. G. Cormack, Barry D. Moore*† and David C. Sherrington*
Department of Pure and Applied Chemistry, University of Strathclyde, T homas Graham Building, 295 Cathedral Street,
Glasgow, UK G1 1XL
Solid phase peptide synthesis (SPPS) has been used as a synthetic tool to prepare monodisperse mesogen–oligopeptide conjugates
with novel molecular architectures. Building upon previous success in this area, we have extended our methodologies to include a
mesogenically substituted -glutamic acid residue, and incorporated this derivative, as well as the -lysine derivative previously
reported, into a variety of new structures. Some novel liquid crystalline materials have been discovered. Both a-helical and b-sheet
secondary structures have been explored as alternative scaffolds for the pendant mesogenic groups. The helical systems show most
promise in terms of their ease of synthesis and handling, but at the relatively lower levels of mesogenic substitution studied so far,
compared to the previously reported homo-oligopeptide materials, mesophases are not observed. The b-sheet materials prepared
are rather insoluble and too high melting to have any practical value as thermotropic materials without further manipulation of
their structures.
For a number of years, one of the major goals in polymer
science has been to prepare high polymers with narrow polydis-
persities and pre-determined molecular mass. This synthetic
control would allow tailoring of physical properties and lead
to a widening of polymers’ potential usefulness, and has now
been achieved successfully for a number of polymer classes,
using a variety of different methodologies. This synthetic aim
is no different in the field of liquid crystalline polymers (LCPs),
where it has been shown that the degree of polymerisation can
exert a significant influence upon the mesophase stability, as
well as the nature of the mesophase itself.1 It is also known
that polydisperse LCPs can display multi-component meso-
phases.2 As a result of these effects, many of the methods for
controlled polymerisation which are now well known in the
literature have been applied to the synthesis of LCPs, with
some success. These include cationic,1,3 anionic,4,5 group-
transfer6,7 and ring-opening metathesis8,9 polymerisation
studies.
As outlined in a preliminary communication,10 we became
interested in applying the synthetic control offered by solid
phase methods in the preparation of side-chain liquid crystal-
line oligopeptides and polypeptides. This approach is very
attractive in terms of preparing truly monodisperse materials,
but also, by the very nature of the technique, it offers unpre-
cedented control over the primary and secondary structures of
the peptide backbones, promising entry into structurally inter-
esting and unusual materials. Although to date solid phase
methods have been used extensively for the routine synthesis
of peptides, and have become central to the preparation of
libraries of compounds by combinational methods,11–13 it had
never before been reported, to our knowledge, for the premedi-
tated synthesis of structurally well-defined liquid crystalline
materials prior to our preliminary communication.10
primary sequences, utilising both the -lysine derivative
reported previously, as well as a mesogenically substituted
amino acid based on -glutamic acid. Amino acid sequences
which support either a-helical or b-sheet secondary structures
are also explored as alternative scaffolds for the mesogenic
groups.
Results and Discussion
Synthetic concepts and approaches
At the beginning of this current study, we identified two
possible routes into structurally well-defined, side-chain liquid
crystalline peptides, which were directly analogous in concept
to those utilised in the preparation of side-chain functionalised
polymers. Thus one could modify, by attachment of a mesogen,
a pre-formed peptide which has been prepared by solid phase
means. Alternatively, an amino acid derivative appropriately
functionalised with a mesogen could be utilised in the solid
phase synthesis. Although the former is certainly attractive in
several respects, it was the latter approach which we adopted
because it offered potentially the greatest degree of synthetic
control.
In the design and synthesis of functionalised amino acids
suitable for our purposes, there were a number of aspects
which were considered, to give our systems synthetic and
structural versatility, whilst maximising the likelihood of meso-
phase formation. These included not only their ease of synthesis
and their stability, but also their potential to facilitate the
formation of ordered structure over and above that of the
liquid crystal phase. We concentrated upon the a-amino acids
because of their ready commercial availability in an optically
pure form, their well established chemistry and their interesting
secondary structure forming abilities.14 Bearing in mind the
implications of the ‘spacer concept’15 for side-chain liquid
crystalline polymers, as well as the structures of certain polydis-
perse liquid crystalline polypeptides already reported in the
literature, we felt that the incorporation of a flexible spacing
unit in all our designs was necessary to promote liquid
crystallinity. Mesogenic groups based on 4-cyano-4∞-hydroxy-
biphenyl were employed throughout, because these have been
extensively studied in both low molecular mass materials and
in polymers, and are reasonably stable under a variety of
conditions.
In the latter, we described the successful evolution of a solid
phase approach to the synthesis of -lysine oligomers substi-
tuted in the side-chain with a mesogenic group. We detailed
the preparation of the dimer, the trimer and the tetramer of a
mesogenically substituted -lysine derivative, and showed that
these did indeed display thermotropic behaviour, as antici-
pated. In this full paper we give further details, and will show
how we have extended our methodologies to include other
* Correspondence should be addressed to: Professor D. C.
Sherrington.
†E-mail: b.d.moore@strath.ac.uk
J. Mater. Chem., 1997, 7(10), 1977–1983
1977

Scheme 2 Reagents and conditions: i, Cl(CH ) OH,KOH; ii, Ac O;
2 6
iii, toluene, heat; iv, PhNHNH , Bu N; v, Fmoc-OSu, Na CO
2
2
3
2
3
The efficiency with which the modified amino acid coupled to
the growing peptide chain was particularly notable, and yielded
materials which did not require further purification. This high
purity was borne out by the sharp thermal transitions which
were observed during heating and cooling cycles. Hot-stage
polarised optical microscopy revealed thermotropic phases for
all three materials which are visually of indistinct texture, but
which we believe to be the nematic phase, since the samples
displayed shear birefringence as well as showing no fan texture
typical of some smectic phases.
Scheme 1 Reagents and conditions: i, MeOH, H SO ; ii, K CO ,
NCC H C H OH; iii, KOH; iv, SOCl ; v, CuLys , NaOH;
2
4
2
3
6
4
6
4
2
2
vi, EDTAΩ2Na; vii, Fmoc-Cl, Na CO
2
3
The peptides 16–18 are considered as ‘uncapped’ since the
N-terminus is present as the free amine. We realised that
perhaps an opportunity existed for modifying the mesophases
and/or the transition temperatures through careful control of
the capping group at the N-terminus, and so proceeded to
prepare the N-acetyl derivatives Ac-[Lys(M)] -NH , n=2
Synthesis of amino acid derivatives
The -lysine derivative 8 was prepared by the route shown in
Scheme 1. 6-(4-Cyanobiphenyl-4∞-yloxy)hexanoic acid 4 was
prepared from 6-bromohexanoic acid 1 according to a litera-
ture procedure.16 In the key step of the synthesis, the acid
chloride of 4 was coupled selectively to the e-amino group in
the -lysine copper complex using interfacial reaction con-
ditions, in a fashion similar to that previously reported by
Gue´niffey for the reaction of long-chain acid chlorides with
the -lysine copper complex.17 Thereafter, decomplexation
followed by fluorenylmethoxycarbonyl (Fmoc) protection gave
8 in an overall yield of 14%, in eight steps.
The -glutamic acid derivative 15 was prepared via the route
shown in Scheme 2. 6-(4-Cyanobiphenyl-4∞-yloxy)hexanol 1018
and N-phthaloyl--glutamic anhydride 1219 were prepared
according to literature procedures, and coupled under con-
ditions similar to those employed by Feijen20 for a related
system, to give 13. Phthaloyl deprotection21 followed by Fmoc
protection gave 15 in an overall yield of 10% in five steps.
n
2
(19), 3 (20) and 4 (21) of the three peptides. To our surprise,
acetylating the N-terminus completely destroyed the mesogenic
properties of the peptides. We believe that this may be due to
a hydrogen bonding interaction between the amide group at
the N-terminus and the amide group in the side-chain, but we
are currently carrying out further capping studies to probe
this interesting effect further.
In a similar fashion to the -lysine system we have also
begun to construct the -glutamic acid series of homo-oligopep-
tides. In the first instance we have prepared the dimer, with
the N-terminus either acetylated or present as the free amine
X-[Glu(M) -NH , X=H (22) or Ac (23). Again, the ‘as cleaved’
2
2
materials were of apparently high purity, and this time both
the acetylated and the uncapped peptide displayed a meso-
phase, which was readily identifiable as the nematic phase in
both cases. We were unable to crystallise 22, and indeed this
oily compound was observed to be birefringent at room
temperature. Further extension of this series to include the
higher analogues will be required to establish structure–prop-
erty relationships, and also to ascertain the effect of developing
secondary structure upon the liquid crystalline behaviour as
the main-chains become longer.
Solid phase synthesis of oligopeptides
Homo-oligopeptides. As detailed elsewhere,10 the dimer,
trimer and the tetramer H-[Lys(M)] -NH , n=2 (16), 3 (17)
and 4 (18)‡ of the -lysine derivative were successfully prepared.
n
2
‡Throughout, M=6-(4-cyanobiphenyl-4∞-yloxy)hexanoyl for lysine
derivatives and 6-(4-Cyanobiphenyl-4∞-yloxy)hexyl for glutamic acid
derivatives.
The fact that the acetylated derivative 23 was liquid crystal-
line in this case can be considered as lending some supporting
1978
J. Mater. Chem., 1997, 7(10), 1977–1983

evidence to the reasoning behind why the acetylated lysine
derivatives 19–21 were not liquid crystalline, since 23 has an
ester functionality in the side-chain, and not an amide.
Targetted a-helical peptides. In developing the concept of
monodisperse liquid crystalline peptides with additional sec-
ondary structure imposed by the peptide backbone, we have
become interested in the preparation of peptides where the
main-chain is in the a-helical conformation, since this structural
feature is common to many natural peptides and proteins, and
is present in virtually all polydisperse polypeptides which
display mesomorphic behaviour.22 To succeed in this goal, one
must have an understanding of the structural parameters which
dictate secondary structure formation. These include the pri-
mary sequence of the peptide,23 the length of the peptide
chain,24,25 the presence or otherwise of substituents at the N-
and C-termini26,27 and other factors such as the ability to form
salt bridges or disulfide bridges which can stabilise secondary
structures.
Our first a-helical peptide target was the dodecapeptide
24, Ac-AlaAlaLys(M)AibAlaLys(M)AlaAibAlaLys(M)AlaAla-
NH , which contains seven alanine (Ala) residues, two a-
2
aminoisobutyric acid residues (Aib) and three mesogenically
substituted -lysine residues [Lys(M)]. We considered the
level of mesogenic substitution to be rather lower (25% of
amino acid residues are mesogenic) than would be considered
ideal for the formation of liquid crystalline phases, but nonethe-
less 24 provided an ideal initial target to assess the feasibility
of the approach.
Fig. 1 A molecular model of the dodecapeptide 24, showing the three
mesogenic groups pendant on one side of the a-helix. The main axis
of the a-helix is normal to the plane of the page.
The amino acids in the sequence were chosen on the basis
of their a-helix forming propensity; alanine is well known for
its helix forming abilities,23 as is the non-natural amino acid
a-aminoisobutyric acid.26,28 The conformational parameters of
the non-natural Lys(M) residue were not known, but it was
anticipated that it would either be a weak helix former or
relatively neutral conformationally, and hence quite readily sit
in a helical conformation in the presence of other strong helix
formers. To our advantage, not only do Aib residues show
strong helical conformation preferences, but they also inhibit
the formation of b-sheets. The C-terminus and the N-terminus
of 24 were capped as the acid amide and the acetyl derivative
respectively, to enhance the solubility of the peptide in organic
solvents and also to enhance the likelihood of helix forma-
tion.27,29 Crucially, the positions of the Lys(M) residues within
the sequence were carefully chosen so that all three of the
mesogenic groups would lie on approximately the same side
of the a-helix (Fig. 1). We believed that this arrangement of
side-chains offered the best prospect for mesophase formation
at this level of mesogenic substitution.
After purification by reverse phase HPLC and precipitation
from methanol, which in our hands has often proved to be
helicogenic,22 24 was indeed shown to adopt an a-helical
conformation, as confirmed by the absorption bands present
at 1659 (amide I) and 1541 cm−1 (amide II) in the solid state
FTIR spectrum.30 Upon heating, it melted directly into the
isotropic liquid state at 196 °C, and crystallised directly from
the melt upon cooling. That is, no thermotropic behaviour
was observed, as perhaps anticipated. Nonetheless we at least
demonstrated that this was a feasible synthetic approach
towards these well-defined helical structures.
would lie on one side of the helix, with the remaining three on
the opposite side. This, we felt, was the most favourable
arrangement of side-chains at this level of mesogenic substi-
tution, in terms of potential mesophase formation. This
arrangement is most easily visualised in the molecular model
(Fig. 2).
After purification by reverse phase HPLC and precipitation
from the helicogenic solvent acetonitrile,22 the conformation
of 25 was investigated using infrared spectroscopy. Bands were
observed at 1668 and 1543 cm−1.30 The former is outside the
range for the amide I band of a helix and instead suggests 25
is folded into a series of beta turns or loops. Thermal studies
indicated that the material did not exhibit any thermotropic
phases, with melting from the crystalline to the isotropic liquid
being observed at 75 °C, and crystallisation directly from the
isotropic melt occurring upon cooling. This could be because
the level of mesogenic substitution is simply not
high enough to sustain a mesophase or because the poorly
defined peptide conformation is unfavourable for ordering.
Accordingly, we are currently working towards more robust
helical systems where the mesogenic content is still higher.
Targetted b-sheet peptides. To date, all of the polydisperse
liquid crystalline polypeptides reported in the literature have
been based upon a-helical main-chains. We were very inter-
ested, therefore, in exploiting the opportunities lent to us by
our methodologies to prepare materials with different second-
ary structures. Accordingly, a series of six peptides (26–31)
based upon a simple repeat unit X-[ValLys(M)] -NH ; X=H
As a second a-helical peptide target, we selected the tetra-
decapeptide 25, Ac-AlaAlaGlu(M)AibGlu(M)AlaGlu(M)Glu-
(M)AiBGlu(M)AlaGlu(M)AlaAla-NH , which again contains
n
2
or Ac; n=1, 2 or 3 was prepared, designed to adopt an
extended b-sheet conformation. Valine (Val) has strong b-sheet
preferences. Every second residue in the sequence was meso-
genically substituted.
2
predominately Ala and Aib residues. The level of mesogenic
substitution is significantly higher (ca. 43% of residues are
mesogenic) than for 24 to favour mesophase formation, and
because of this we elected to utilise the -glutamic acid
derivative because the ability of -glutamic esters themselves
to support an a-helix is well known.22 The mesogenic residues
within the primary sequence were arranged such that three
The solid phase synthesis of these peptides proceeded reason-
ably well, but the presence of deletion peptides was apparent
in the mass spectra of the crude samples of the higher (n>2)
analogues. The low solubility of these materials in a range of
solvents meant that satisfactory purification was not possible,
J. Mater. Chem., 1997, 7(10), 1977–1983
1979

Table 1 Summary of the peptides prepared, showing their composition
and their thermal behaviour
no. of amino mesogenic
acids in chain content (%)
peptide
thermal behaviour
homo-oligopeptides--lysine series
16a
17a
18a
19b
20b
21b
2
3
4
2
3
4
100
100
100
100
100
100
K 142 °C LC 173 °C I
K 163 °C LC 190 °C I
K 204 °C I 175 °C LC
mp 116 °C
mp 120 °C
mp 118 °C
homo-oligopeptides--glutamic acid series
22a
23b
2
2
100
100
N 71.6 °C I
K 109 °C I 105 °C N
60 °C K
a-helical peptides
25
24
25
12
14
mp 196 °C
~43 mp 75 °C
b-sheet peptides
26–31
4, 6 or 8
50
mp ~300–350 °C (decomp.)
aUncapped, bCapped.
much longer peptides, containing a number of mesogenic units,
by this approach (up to 14 residues in length), and control
their secondary structures through appropriate design of the
amino acid sequence. Although we have not yet been able to
prepare longer peptides which are thermotropic, either because
the level of mesogenic substitution was too low or the materials
were too high melting, we are nonetheless currently pursuing
a number of approaches which will undoubtedly make this
possible.
Fig. 2 A molecular model of the tetradecapeptide 25, showing three
mesogenic groups pendant on one side of the a-helix, and the
remaining three pendant on the opposing side. The main axis of the
a-helix is normal to the plane of the page.
Experimental
Materials
All materials used in this study were commercial samples, and
were used as supplied unless otherwise stated. For all the solid
phase syntheses, the amino acids and coupling reagents
employed were of peptide synthesis grade, and were sup-
plied by Novabiochem. N,N,-Dimethylformamide (DMF)
(Rathburn, peptide synthesis grade), piperidine (Rathburn,
peptide synthesis grade), N,N-diisopropylethylamine (Aldrich,
99%) and trifluoroacetic acid (TFA) (Rathburn, gas phase
sequencer grade) were also used as supplied. The resin support
used throughout (PR500, 0.36 mmol g−1 nominal loading) was
supplied by Novabiochem. All chromatographic solvents were
distilled prior to use, and other solvents purified, where appro-
priate, using standard procedures.
and as a result the thermal measurements were carried out on
the ‘as cleaved’ materials.
FTIR spectroscopy showed that an extended chain (b-sheet)
conformation had indeed been adopted in every case, as shown
by the amide I and amide II absorption bands present
at ca. 1635 and ca. 1545 cm−1 respectively in each sample.30
However, thermal studies revealed that all these materials
either melted or decomposed at high temperatures
(typically>300 °C) without entering into a liquid crystal phase.
This was deemed too high to be of any practical value, so we
are currently pursuing structural modifications aimed at lower-
ing the melt temperatures and allowing fluid phases to be
accessed at more reasonable temperatures. In principle, these
modifications, which involve disrupting the hydrogen bonding,
should also make these materials more soluble and hence
easier to purify and handle.
Instrumentation
1HNMR spectra were recorded on a 250 MHz Bruker WM-
250 spectrometer. FTIR spectra were recorded either on a
UNICAM Matteson 1000 spectrometer or on a Nicolet Impact
400D spectrometer. Elemental microanalysis was carried out
by the Microanalytical Laboratory at the University of
Strathclyde. Melting points were recorded on a Gallenkamp
melting point apparatus, and are uncorrected. HPLC was
Summary and Prospects
In conclusion, we have demonstrated that the solid phase
approach to the synthesis of monodisperse liquid crystalline
oligopeptides is indeed very powerful, and has allowed us to
prepare a number of interesting mesogen–peptide conjugates
of unique structure; the materials prepared to date are summar-
ised in Table 1. In the first instance we prepared the dimer, the
trimer and the tetramer of a mesogenically substituted -lysine
derivative, and found all of these to be liquid crystalline when
the N-terminus was uncapped. We are currently extending this
series to include the higher analogues, whilst at the same time
developing an alternative series based upon an -glutamic acid
derivative. We have also shown that we can readily prepare
performed on a C column (25×4.6 mm, packed with 5 mm
Spherisorb ODS2, supplied by Anachem) using Gilson 306
8
pumps and a Gilson 117 UV detector. ES–MS spectra were
recorded on a Fisons VG Platform spectrometer. DSC was
carried out on a Perkin-Elmer DSC 7 differential scanning
calorimeter. Molecular modelling was carried out using
MACROMODEL, Version 4.5, on a Silicon Graphics Indigo
work-station.
1980
J. Mater. Chem., 1997, 7(10), 1977–1983

Solid phase synthesis of peptides
6-(4-Cyanobiphenyl-4∞-yloxy)hexanoyl chloride 5. The acid 4
(3.00 g, 9.70 mmol) was stirred with thionyl chloride (10 cm3,
137 mmol) for 3 h at room temp. in a flask fitted with a
calcium chloride guard tube, giving a clear yellow solution.
This solution was then gently heated for 90 min to expel any
remaining gases from solution. The excess thionyl chloride was
then removed under reduced pressure, and the oily residue co-
evaporated several times with dry diethyl ether. The acid
chloride was then used in the next step without further
Oligopeptides were synthesised on a Novasyn Crystal solid
phase peptide synthesiser using Novasyn PR500 resin and
PyBOP coupling chemistry. Typically, each residue was double
coupled using a two-fold excess of amino acid. Cleavage of the
peptide from the resin was with 10% TFA in CH Cl for
2
2
90 min.
Liquid crystal characterisation of peptides
purification; n /cm−1 (selected band) 1805.
max
Hot stage polarised optical microscopy was performed either
on an Olympus CH-2 microscope fitted with a Mettler FP-5
hot-stage, JVC TK-1085E colour video attachment and a Sony
UP-3000P colour video printer, or on an Olympus Vanox
microscope fitted with a Linkam TH600 hot-stage and a
Linkam PR600 thermal controller. Microscope slides were
pretreated with a homogeneous aligning agent.
Copper complex of e-[6-(4-cyanobiphenyl-4∞-yloxy)hexanam-
ido]--lysine 6. To a boiling solution of -lysine monohydro-
chloride (3.54 g, 19.4 mmol) in water (60 cm3), was added basic
copper() carbonate (9.43 g, 42.7 mmol) over a period of
10 min; the addition caused effervescence and the supernatant
solution turned bright blue. The mixture was boiled for a
further 10 min, cooled to room temp. and the excess copper()
carbonate filtered off. A solution of sodium hydroxide (0.78 g,
19.4 mmol) in water (5 cm3) was then added to the bright blue
filtrate and the solution cooled to 0 °C on an ice-bath. A
solution of 5 (9.7 mmol) in dry dichloromethane (DCM)
(30 cm3) was added to the rapidly stirred solution over a
period of 2.5 h, the temperature being maintained at 0 °C, and
the mixture then stirred for a further 4.5 h at the same
temperature. The lilac coloured precipitate which formed was
filtered off, washed with ethanol, water and diethyl ether, and
then air-dried (4.55 g, 100% technical yield). Further purifi-
Syntheses
L
-Lysine derivative 8. Methyl 6-bromohexanoate 2. 6-
Bromohexanoic acid (22.00 g, 113 mmol), concentrated sulfuric
acid (3 cm3) and methanol (50 cm3) were refluxed for 3 h and,
after cooling, excess methanol was removed under reduced
pressure. Portions of deionised water (100 cm3) and chloroform
(100 cm3) were added to the colourless liquid residue, and the
organic layer separated. The aqueous layer was further
extracted with portions of chloroform (3×50 cm3). The organic
layers were combined, washed with 5% aqueous sodium
hydrogen carbonate (100 cm3), then water (50 cm3), dried over
magnesium sulfate and the chloroform removed under reduced
pressure. The crude liquid residue was distilled under reduced
pressure and 2 collected as a colourless liquid (17.16 g,
73%), bp 40 °C at 0.01 mbar. (Found: C, 40.6; H, 6.6; Br, 37.7.
C H BrO requires C, 40.2; H, 6.3; Br, 38.2%); n /cm−1
cation was not carried out; n /cm−1 3430, 3306, 2936, 2865,
max
2225, 1620, 1604, 1541, 1496, 1390, 1291, 1253, 1181 and 822.
e-[6-(4-Cyanobiphenyl-4∞-yloxy)hexanamido]--lysine 7. The
copper complex 6 (94.43 g, 4.73 mmol) was stirred with 10%
aqueous ethylenediaminetetraacetic acid (EDTA) disodium salt
(200 cm3) for 20 h at room temp. The precipitate which formed
was filtered off and again stirred with 10% aqueous EDTA
disodium salt (200 cm3) for a further 20 h at room temp. After
filtering, the solid was washed well with water and dried in
vacuo at 50 °C for 64 h (3.70 g, 89% technical yield), mp 285 °C.
7
13
2
max
2947, 2865, 1740, 1436, 1362, 1255, 1202, 1173, 1122 and 1000;
d (CDCl ) 1.40–1.52 (m, 2H), 1.57–1.76 (m, 2H), 1.78–1.98
(m, 2H), 2.33 (t, 2H), 3.38 (t, 2H) and 3.66 (s, 3H).
H
3
Methyl 6-(4-cyanobiphenyl-4∞-yloxy)hexanoate 3. The ester 2
(8.80 g, 42.1 mmol), 4-cyano-4∞-hydroxybiphenyl (8.21 g,
42.1 mmol) and anhydrous potassium carbonate (5.82 g,
42.1 mmol) were stirred in DMF (20 cm3) for 24 h in a flask
fitted with a calcium chloride guard tube. The viscous yellow
slurry which formed was poured into water (200 cm3) and the
resultant white precipitate filtered off and washed with water.
The crude product was dried over phosphorus pentoxide for
48 h, and then recrystallised from 80% aqueous ethanol
(175 cm3) giving 3 as a white powder (9.90 g, 73%), mp
88–89 °C (lit.,16 80–81 °C). (Found: C, 74.4; H, 6.8; N, 4.3.
C H NO requires C, 74.3; H, 6.6; N, 4.3%); n /cm−1 2947,
Further purification was not carried out; n /cm−1 3310,
max
3071, 2938, 2866, 2229, 1638, 1606, 1582, 1543, 1521, 1494,
1407, 1289, 1253, 1181 and 824.
e-[6-(4-Cyanobiphenyl-4∞-yloxy)hexanamido]-N-fluorenyl-
methoxycarbonyl--lysine 8. Compound 7 (3.00 g, 6.86 mmol)
was suspended in a mixture of 10% aqueous sodium carbonate
(60 cm3) and dioxane (40 cm3), and the mixture stirred for 1 h
at room temp. A solution of fluorenyl chloroformate (1.95 g,
7.54 mmol) in dioxane (20 cm3) was then added to the stirred
mixture; precipitation of a white solid was observed approxi-
mately 5 min after the addition. After stirring for a further
24 h, the mixture was poured into water (75 cm3) giving a
turbid solution. Acidification to pH 3 with 20% aqueous
hydrochloric acid yielded a sticky white material which was
extracted into chloroform (3×150 cm3). The organic layers
were combined, washed with water (150 cm3) and dried over
magnesium sulfate. The solvent was removed under reduced
pressure and the crude product recrystallised from chloroform–
light petroleum, giving 8 as a white powder (2.05 g, 45%), mp
154–156 °C. (Found: C, 72.4; H, 6.4; N, 6.1. C H N O
20 21
3
max
2869, 2222, 1722, 1600, 1497, 1473, 1294, 1272, 1254, 1181,
1031 and 829; d (CDCl ) 1.46–1.92 (m, 6H) 2.37 (t, 2H), 3.68
(s, 3H), 4.02 (t, 2H), 6.98 (m, 2H) and 7.50–7.73 (m, 6H).
H
3
6-(4-Cyanobiphenyl-4∞-yloxy)hexanoic acid 4. The ester 3
9.00 g, 27.8 mmol) and potassium hydroxide (16.00 g,
285 mmol) were stirred in ethanol (150 cm3) for 3 h at room
temp. in a flask fitted with a calcium chloride guard tube. The
resultant yellow slurry was poured into iced-water (450 cm3),
the mixture allowed to warm to room temp. and then neutral-
ised with concentrated sulfuric acid. The white precipitate was
filtered off, washed with water and dried over calcium chloride.
Recrystallisation from ethanol (400 cm3) gave 4 as a white
powder (5.73 g, 67%), K 163.7 °C I 157.1 °C N 134.3 °C (from
DSC) (lit.,16 mp 165 °C) (Found: C, 73.6; H, 6.2; N, 4.1.
C H NO requires C, 73.8; H, 6.2; N, 4.5%); n /cm−1 2942,
40 41
3 6
requires C, 72.8; H, 6.3; N, 6.4%); n /cm−1 3327, 3065, 2942,
max
2862, 2224, 1751, 1694 (sh. at 1721), 1602, 1544, 1494, 1451,
1253, 1181, 821, 760 and 741; d (DMSO) 1.06–1.84 (m, 12H),
2.07 (t, 2H), 3.02 (m, 2H), 3.84–4.06 (m, 3H), 4.18–4.38 (m,
3H) and 6.93–8.35 (m, 18H); M 659.8; found, m/z 682.5
(M+Na+) and 660.7 (M+H+).
H
19 19
3
max
2869, 2229, 1703, 1601, 1445, 1251, 1183, 1076, and 820; d
L
-Glutamic acid derivative 15. 6-(4-Cyanobiphenyl-4∞-yloxy)
H
([2H ]THF) 1.30–1.88 (m, 6H), 2.25 (t, 2H), 4.04 (t, 2H) and
6.98–7.88 (m, 8H).
hexanol 10. 4-Cyano-4∞-hydroxybiphenyl (6.43 g, 32.9
mmol), potassium hydroxide (1.85 g, 32.9 mmol) and a few
8
J. Mater. Chem., 1997, 7(10), 1977–1983
1981

crystals of potassium iodide were dissolved in an ethanol–
water mixture (451, 165 cm3) at room temp. 6-Chlorohexanol
(4.87 cm3, 36.6 mmol) was then added, and the mixture refluxed
for 24 h. After cooling, the ethanol was removed under reduced
pressure, and the resultant yellow slurry filtered off on a glass
sinter. The product was carefully washed with water, dilute
aqueous sodium hydroxide, and then again with water.
Recrystallisation from methanol (50 cm3) gave 10 as a white
powder (4.10 g, 42%), K 92.3 °C N 113.0 °C I (from DSC)
(lit.,18 K 93.5 °C N 110.9 °C I). (Found: C, 76.7; H, 6.7; N, 4.7.
C H NO requires C, 77.2; H, 7.2; N, 4.7%); n /cm−1 3289,
mixture was observed 15 min after the addition. The free-
flowing white coloured mass which formed was poured into
water (500 cm3), and the resulting turbid solution–suspension
acidified with glacial acetic acid. The white precipitate which
formed was filtered off and washed with water and diethyl
ether. Purification by flash column chromatography on silica
gel using ethyl acetate as the eluent gave 15 as a sticky, yellow
solid (1.72 g, 69%). (Found: C, 72.9; H, 6.3; N, 4.1, C H N O
39 38
2 7
requires C, 72.4; H, 5.9 N, 4.3%); n /cm−1 3351, 3064, 3039,
760 and 741; d (CDCl ) 1.32–2.70 (m, 12H), 3.73–4.58 (m,
max
2935, 2862, 2225, 1726, 1602, 1522, 1493, 1251, 1178, 1051, 822,
19 21
2
max
H
3
2943, 2868, 2226, 1602, 1494, 1473, 1291, 1252, 1183, 1072,
8H), 6.82–7.83 (m, 17H); M 646.8; m/z 647.9 (M+H+) and
1012 and 829; d (CDCl ) 1.23–1.96 (m, 9H), 3.68 (t, 2H), 4.02,
(t, 2H), 6.98 (d, 2H), 7.42–7.83 (m, 6H).
311.2 (M−C H NO +H+).
H
3
21 22
3
Resin washing protocol
N-Phthaloyl--glutamic anhydride 12. A mixture of N-
phthaloyl--glutamic acid (5.07 g, 18.3 mmol) and acetic anhy-
dride (10 cm3, 106 mmol) was refluxed for 10 min under a dry
nitrogen atmosphere. Upon cooling to room temp., a white
precipitate was seen for form. Dry diethyl ether (50 cm3) was
added to the mixture, the solids filtered off, washed with a
further volume of dry diethyl ether, and then air-dried.
Recrystallisation from dry ethyl acetate (75 cm3) furnished 12
as colourless crystals, with a second crop being collected after
cooling the filtrate overnight in the refrigerator (3.26 g, 69%),
mp 196–197 °C (decomp.) (lit.,19 195–196 °C). (Found: C, 60.0;
H. 3.5; N, 5.4. C H NO requires C, 60.2; H, 3.5; N, 5.4%);
Upon completion of each peptide synthesis, and prior to
cleavage of the peptide from the resin, the resin–peptide
conjugate was washed according to the following protocol to
remove any residual traces of DMF as well as any other
soluble impurities.
The resin–peptide conjugate was removed from the Novasyn
Crystal reactor column, placed on a glass sinter and then
washed sequentially with DMF, tert-pentyl alcohol, acetic acid,
tert-pentyl alcohol, DCM and finally diethyl ether. It was then
dried in vacuo for 8 h.
13
9
5
n
and 723; d (DMSO) 2.53–2.70 (m, 2H), 2.90–3.20 (m, 2H),
/cm−1 1813, 1772, 1715, 1470, 1389, 1226, 1074, 1030, 974
5.43–5.50 (m, 1H), 7.88–7.97 (m, 4H).
Acetylation of resin-bound peptides
max
H
Where appropriate, peptides were acetylated at the N-terminus
as follows. The dry resin-peptide conjugate (20 mg) was placed
in a round-bottomed flask and distilled DCM (4 cm3) added,
followed by a few crystals of dimethylaminopyridine (DMAP)
and distilled acetic anhydride (1 cm3). The mixture was gently
shaken for 4 h, the resin filtered off and washed well with fresh
DCM. The resin was then dried in vacuo for several hours.
5-[6-(4-Cyanobiphenyl-4∞-yloxy)hexyl] hydrogen N-phthal-
oyl--glutamate 13. Compounds 10 (3.56 g, 12.0 mmol) and 12
(3.12 g, 12.0 mmol) were refluxed in dry toluene (300 cm3)
under a dry nitrogen atmosphere for 24 h. After cooling, the
toluene was removed under reduced pressure and the oily
residue purified by flash column chromatography on silica gel,
using ethyl acetate as the eluent, giving 13 as a sticky yellow
oil (5.34 g, 80%). (Found: C, 69.6; H, 5.4; N, 4.8. C H N O
Cleavage of peptides from the resin support
The dry resin–peptide conjugate (20 mg) was placed in a round
bottomed flask and 10% TFA in distilled DCM (10 cm3)
added; the resin was observed to turn bright red shortly
thereafter. The flask contents were swirled every 10 min or so,
and after 90 min the resin was filtered off on a glass sinter and
washed with 10% TFA in DCM (4×5 cm3) followed by DCM
(4×5 cm3). The filtrate was concentrated under reduced press-
ure, and the residue then co-evaporated several times with
acetonitrile (HPLC grade). Addition of diethyl ether (glass
distilled grade) to the resultant oily residue yielded a mobile,
white precipitate which was collected by centrifugation and
then dried in vacuo for 8 h.
32 30
2 7
requires C, 69.3; H, 5.5; N, 5.1%); n /cm−1 2939, 2863, 2225,
720; d (CDCl ) 1.32–1.90 (m, 8H), 2.34–2.73 (m, 4H) 3.92–4.22
(m, 4H), 4.87–5.05 (m, 1H), 6.98 (d, 2H) and 7.47–7.96 (m, 10H).
max
1776, 1720, 1603, 1494, 1468, 1391, 1251, 1178, 1114, 823 and
H
3
5-[6-(4-Cyanobiphenyl-4∞-yloxy)hexyl] hydrogen -glutamate
14. Compound 13 (5.34 g, 9.63 mmol) was dissolved in 96%
aqueous ethanol (10 cm3). Tributylamine (2.29 cm3,
9.63 mmol) and freshly distilled phenylhydrazine (2.84 cm3,
28.9 mmol) were then added and the mixture refluxed for 2 h;
a solid was observed to precipitate from solution approximately
15 min after reflux was attained. Methyl ethyl ketone (20 cm3)
was added, the mixture refluxed for a further 15 min, and then
cooled to room temp. Glacial acetic acid (0.83 cm3, 14.4 mmol)
was added, the mixture stirred at room temp. for approximately
10 min, and the white solid filtered off. This was washed
carefully with methyl ethyl ketone and then dried in vacuo for
7 h at 40 °C giving 14 as a white powder (1.67 g, 41% technical
yield) mp 159.7 °C (from DSC). Further purification was not
carried out. (Found: C, 67.1; H, 7.0; N, 6.7. C H N O
Homo-oligopeptides
Satisfactory acylation and deprotection traces were obtained
for all residues:
H-[Lys(M)] -NH 16. K 142 °C LC 173 °C I; M 856.2, m/z
2
2
857.2 (M+H+).
24 28
2 5
requires C, 67.9 H, 6.7; N, 6.6%); n /cm−1 3445, 2938, 2864,
max
H-[Lys(M)] -NH 17. K 163 °C LC 190 °C I; M 1275.8, m/z
2228, 1730, 1604, 1580, 1496, 1414, 1329, 1254, 1181 and 823.
3
2
1298.2(M+Na+), 1276.3 (M+H+), 650.0 (M+H++Na+)
and 639.0 (M+2Na+).
5-[6-(4-Cyanobiphenyl-4∞-yloxy)hexyl] hydrogen N-fluorenyl-
methoxycarbonyl--glutamate 15. Compound 14 (1.63 g,
3.84 mmol) was suspended in a mixture of 10% aqueous
sodium carbonate (50 cm3) and dioxane (35 cm3); partial dis-
solution was observed. To the rapidly stirred suspension was
added a solution of fluorenylmethoxycarbonyl-N-hydroxysuc-
cinimide (1.30 g, 3.84 mmol) in dioxane (15 cm3), and the
mixture stirred at room temp. for 23 h; precipitation from the
H-[Lys(M)] -NH 18. K 204 °C I 175 °C LC; M 1695.3, m/z
4
2
870.4
(M+2Na+),
867.5
(M+H++K+),
859.4
(M+H++Na+) and 848.5 (M+2H+).
Ac-[Lys(M)] -NH 19. mp 116 °C; M 898.2, m/z 921.0
2
2
(M+Na+) and 899.0 (M+H+).
1982
J. Mater. Chem., 1997, 7(10), 1977–1983

Ac-[Lys(M)] NH 20. mp 120 °C; M 1317.8, m/z 1318.6
(M+H+), 670.9 (M+H++Na+) and 659.9 (M+2H+).
Ac-[ValLys(M)] -NH 30. mp 330 °C (decomp.); n /cm−1
(selected bands) 1633 and 1547.
3
2
3
2
max
Ac-[ValLys(M)] -NH 31. mp 340 °C (decomp.); n /cm−1
Ac-[Lys(M)] -NH 21. mp 118 °C; M 1737.4, m/z 1738.6
(M+H+), 880.5 (M+H++Na+) and 869.7 (M+2H+).
4
2
(selected bands) 1634 and 1548.
max
4
2
We gratefully acknowledge financial support from both the
EPSRC and Merck Ltd. We would also like to thank Professor
George W. Gray, Dr David Coates and Ian Bonny for their
useful discussions, and Novabiochem for assistance with the
peptide syntheses.
H-[Glu(M)] -NH 22. N 71.6 °C I; M 830.1, m/z 830.9
2
2
(M+H+), 495.6 (M−C H NO +H+) and 477.7
21 22
3
(M−C H NO −NH −).
21 22
3
2
Ac-[Glu(M)] -NH 23. K 109 °C I 105 °C N 60 °C; M 872.1,
2
2
m/z
894.7
(M+Na+),
872.6
(M+H+),
559.1
(M−C H NO +H+) and 537.0 (M−C H NO +H+).
References
21 22
3
21 22
3
1
2
3
4
V. Percec, M. Lee and C. Ackerman, Polymer, 1992, 33(4), 703.
H. Finkelmann, Angew. Chem., Int. Ed. Engl., 1987, 26, 816.
V. Percec and D. Tomazos, Adv. Mater., 1992, 4(9), 548.
N. Koide, T. Kumada and K. Iimura, Abstr., 14th Symp. L iquid
Crystals, Sendai, Sept. 29th, 1988, 164.
a-Helical peptides
Satisfactory acylation and deprotection traces were obtained
for all residues.
5
6
7
8
9
R. Bohnert, H. Finkelmann and P. Lutz, Makromol. Chem. Rapid
Commun., 1993, 14(2), 139.
W. Kreuder, O. W. Webster and H. Ringsdorf, Makromol. Chem.
Rapid Commun., 1986, 7, 5.
M. Hefft and J. Springer, Makromol. Chem. Rapid Commun., 1990,
11(8), 397.
Z. Komiya, C. Pugh and R. R. Schrock, Macromolecules, 1986,
25, 6586.
C. Pugh, Macromol. Symp., 1994, 77, 325.
Ac-AlaAlaLys(M)AibAlaLys(M)AlaAibAlaLys(M)AlaAla-
NH 24. mp 196 °C; n /cm−1 (selected bands) 1659 and 1541;
2
max
M 1958.7, m/z M+H++K+), 1004.6 (M+H++Na+), 993.6
(M+2H+) and 985.2 (M−NH−+H+).
2
Ac-AlaAlaGlu(M)AibGlu(M)AlaGlu(M)Glu(M)AibGlu(M)-
AlaGlu(M)AlaAla-NH 25. mp 75 °C; n /cm−1 (selected
2
max
10 P. A. G. Cormack, D. C. Sherrington and B. D. Moore, Chem.
Commun., 1996, 353.
11 M. A. Gallop, R. W. Barrett, D. J. Dower, S. P. A. Fodor and
E. M. Gordon, J. Med. Chem., 1994, 37, 1233.
12 R. M. Baum, Chem. Eng. News, 1994, 20.
bands) 1668 and 1543; M 3095.0, m/z 796.7 (M+4Na+), 701.8
C H NO ).
(M+2H++2Na+−C H NO ) and 575.6 (M+5Na+−
21 22
3
21 22
3
b-Sheet peptides
13 R. M. J. Liskamp, Angew. Chem., Int. Ed. Engl., 1994, 33, 633.
14 M. Bodanszky, Peptide Chemistry: A Practical T extbook, Springer
Verlag, Berlin, 1988.
15 Side Chain L iquid Crystal Polymers, ed. C. B. McArdle, Blackie,
Glasgow, New York, 1989.
16 J. L. West, US Patent 5,093,471.
17 H. Gue´niffey, R. Garnier and C. Pinazzi, Makromol. Chem., 1977,
178, 1277.
18 V. Percec and M. Lee, Macromolecules, 1991, 24(5), 1017.
19 F. E. King and D. A. A. King, J. Chem. Soc., 1949, 3315.
20 J. Feijen, W. M. Sederel, K. de Groot, A. C. de Visser and
A. Bantjes, Makromol. Chem., 1974, 175, 3193.
The low solubility of these materials in a range of solvents
meant that satisfactory purification was often not possible, and
as a result the thermal measurements were carried out on the
‘as cleaved’ materials. For some peptides it was not possible
to acquire ES–MS data; in these cases the identity of the
peptides were inferred from the satisfactory acylation and
deprotection traces which were obtained during the solid phase
synthesis.
21 I. Schumann and R. A. Boissonnas, Helv. Chim. Acta, 1952, 35,
2235.
22 W. H. Daly, D. S. Poche´ and I. I. Negulescu, Prog. Polym. Sci.,
1994, 19(1), 79.
23 P. Y. Chou and G. D. Fasman, Annu. Rev. Biochem., 1978, 47, 251.
24 M. Narita, Y. Tomotake, S. Isokawa, T. Matsuzawa and
T. Miyauchi, Macromolecules, 1984, 17, 1903.
25 M. Mutter, Angew. Chem., Int. Ed. Engl., 1985, 24, 639.
26 C. Toniolo, G.M. Bonora, E. Benedetti, A. Bavoso, B. D. Blasio,
V. Pavone and C. Pedone, Biopolymers, 1983, 22, 1335.
27 A. Bierzynski, P. S. Kim and R. L. Baldwin, Proc. Natl. Acad. Sci.
USA, 1982, 79, 2470.
H-[ValLys(M)] -NH 26. mp 350 °C (decomp.); n /cm−1
2
2
max
(selected bands) 1634 and 1550; M 1054.5, m/z 1077.3
(M+Na+), 1055.3 (M+H+), 547.4 (M+H++K+) and 528.3
(M+2H+).
H-[ValLys(M)] -NH 27. mp 275 °C (decomp.); n /cm−1
3
2
max
(selected bands) 1633 and 1546; M 1573.2, m/z 1573.4 (M+H+),
1154.3 [M−Lys(M)+H+], 1055.2 [M−Lys(M)−Val+H+ ],
806.8 (M+H++K+), 798.7 (M+H++Na+) and 787.7
(M+2H+).
28 I. L. Karle and P. Balaram, Biochemistry, 1990, 29, 6747.
29 K. R. Shoemaker, P. S. Kim, O. N. Brems, S. Marqusee, E. J. York,
I. M. Chaiken, J. M. Stewart and R. L. Baldwin, Proc. Natl. Acad.
Sci. USA, 1985, 82, 2349.
H-[ValLys(M)] -NH 28. mp 340 °C (decomp.); n /cm−1
4
2
(selected bands) 1634 and 1549.
max
30 S. Krimm and J. Bandekar, Adv. Protein Chem., 1986, 38, 181.
Ac-[ValLys(M)] -NH 29. mp 330 °C (decomp.); n /cm−1
2
2
max
(selected bands) 1633 and 1550; M 1096.5, m/z 1096.3 (M+H+).
Paper 7/01084B; Received 17th February, 1997
J. Mater. Chem., 1997, 7(10), 1977–1983
1983