Citation
Synthesis and characterization of 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-1-carboxylic acid

Material Information

Title:
Synthesis and characterization of 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-1-carboxylic acid
Creator:
Peters, William Kenneth
Publication Date:
Language:
English
Physical Description:
viii, 43 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Carboxylic acids ( lcsh )
Peptides ( lcsh )
Carboxylic acids ( fast )
Peptides ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 42-43).
Thesis:
Department of Chemistry
Statement of Responsibility:
by William Kenneth Peters.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
259702676 ( OCLC )
ocn259702676
Classification:
LD1193.L46 2008m P47 ( lcc )

Full Text
SYNTHESIS AND CHARACTERIZATION OF 4-AMINO-3-
/
OXO-2-AZAB ICY CL0[2.2.2]0CTANE-1 -CARBOXYLIC ACID
by
William Kenneth Peters
B.A., University of Colorado at Boulder, 2004
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Chemistry
2008


This thesis for the Master of Science
degree by
William Kenneth Peters
has been approved
by
Hai Lin

'2oor
Date


Peters William K (M.S., Chemistry)
Synthesis and Characterization of 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-l-
carboxylic acid
Thesis directed by Professor Douglas F. Dyckes
ABSTRACT
The dipeptide analog 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-l-
carboxylic acid has been synthesized and characterized by 'H-NMR and FT-IR
spectroscopy. Protected forms of this compound were desired in order to
incorporate the core structure into polypeptide chains. To that end the methyl ester,
the N-butyloxycarbonyl derivative, and the doubly-protected version have all been
synthesized.
This dipeptide is potentially useful in peptidomimetic drug compounds,
where it can induce rigid, linear structures not commonly available with alpha-
amino acid backbones. It may also find use in the emerging field of
nanotechnology, where non-peptide versions of the bicyclo[2.2.2]octane structure
have already been used. Substituting this novel structure into those applications
would increase biocompatibility, solubility, and chemical availability.
Early attempts at dimerizing the monomer have not been successful. Future
work will likely include exploring dimerization and polymerization strategies as
well as coupling this compound to natural amino acids.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed
Douglas F. Dyckes


TABLE OF CONTENTS
Figures...................................................................vii
Schemes...................................................................viii
Chapter
1. Introduction..............................................................1
1.1 Motivation.............................................................. 2
1.2 Background..............................................................7
2. Results and Discussion..................................................10
2.1 Synthesis of l,4-diaminocyclohexane-l,4-dicarbonitrile (compound I)
and l,4-diaminocyclohexane-l,4-dicarboxylic acid (compound II)..........11
2.2 Synthesis of dimethyl l,4-diaminocyclohexane-l,4-dicarboxylate
(compound III)..........................................................12
2.3 Synthesis of methyl 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-l-
carboxylate (compound IV)...............................................17
2.4 Synthesis of 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-l-carboxylic
acid (compound V).......................................................19
2.5 Synthesis of 4-[(ter/-butoxycarbonyl)amino]-3-oxo-2-
azabicyclo[2.2.2]octane-l -carboxylic acid (compound VI)................23
2.6 Synthesis of methyl 4-[(tert-butoxycarbonyl)amino]-3-oxo-2-
IV


azabicyclo[2.2.2]octane-l-carboxylate (compound VII)..................26
2.7 Attempted synthesis of 4-[(4-Amino-3-oxo-2-aza-
bicyclo[2.2.2]octane-l-carbonyl)-amino]-3-oxo-2-aza-
bicyclo[2.2.2]octane-l-carboxylic acid methyl ester (compound VIII)...28
3. Experimental..........................................................30
3.1 Reagents..............................................................30
3.2 General Methods......................................................30
3.2.1 Thin Layer Chromatography...........................................30
3.2.2 Nuclear Magnetic Resonance..........................................31
3.2.3 Infrared spectroscopy...............................................31
3.3 Synthetic Procedure.................................................33
3.3.1 Synthesis of l,4-diaminocyclohexane-l,4-dicarbonitrile
(compound I)........................................................33
3.3.2 Synthesis of l,4-diaminocyclohexane-l,4-dicarboxylic acid
(compound II).......................................................34
3.3.3 Synthesis of dimethyl l,4-diaminocyclohexane-l,4-dicarboxylate
(compound III)......................................................35
3.3.4 Synthesis of methyl 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-l-
carboxylate (compound IV)...........................................37


3.3.5 Synthesis of 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-l-carboxylic
acid (compound V)..................................................38
3.3.6 Synthesis of 4-[(ter/-butoxycarbonyl)amino]-3-oxo-2-
azabicyclo[2.2.2]octane-l-carboxylic acid (compound VI)............39
3.3.7 Synthesis of methyl 4-[(tert-butoxycarbonyl)amino]-3-oxo-2-
azabicyclo[2.2.2]octane-l-carboxylate (compound VII).............40
3.3.8 Attempted synthesis of 4-[(4-Amino-3-oxo-2-aza-
bicyclo[2.2.2]octane-l-carbonyl)-amino]-3-oxo-2-aza-
bicyclo[2.2.2]octane-l-carboxylic acid methyl ester
(compound VIII).................................................. 41
References..............................................................42
vi


LIST OF FIGURES
Figure
1.1. NANORODS OF COMPOUND V..........................6
1.2. SELF ASSEMBLY OF COMPOUND V POLYMERS............7
1.3. PREVIOUSLY SYNTHESIZED COMPOUNDS................7
2.2.1 - H-NMR SPECTRUM OF COMPOUND III in CDC13......14
2.2.2 - FT-IR SPECTRUM OF COMPOUND III...............15
2.3.1 - H-NMR SPECTRUM OF COMPOUND IV IN MeOD........18
2.4.1 H-NMR SPRECTRUM OF COMPOUND V IN D20, NaOD...20
2.4.2 H-NMR SPECTRUM OF COMPOUND V IN D20, DC1.....21
2.4.3 - FT-IR SPECTRUM OF COMPOUND V.................22
2.5.1 - H-NMR SPECTRUM OF COMPOUND VI IN MeOD........24
2.5.2 - FT-IR SPECTRUM OF COMPOUND VI................25
2.6.1 - H-NMR SPECTRUM OF COMPOUND VII IN CDCI3......27
2.7.1 - H-NMR SPECTRUM OF PRODUCTS FROM EDC COUPLING.29
vii


LIST OF SCHEMES
Scheme
1....................................................................1
2....................................................................10
viii


1. Introduction
The goal of this research was to synthesize 4-amino-3-oxo-2-
azabicyclo[2.2.2]octane-l-carboxylic acid (compound V in scheme 1) and some of
its potentially useful protected forms. Long-term goals include oligomerization of
compound V to explore its use as a nanomaterial and incorporation of compound V
into mixed peptide chains to explore its use as a peptide mimic. To that end the
protecting groups chosen were taken from the peptide synthesis community, as the
use of these would allow future studies to be done using the existing technologies
for peptide polymerization.
compound V
compound IV
o H
o H
compound VI
compound VII
SCHEME 1
1


1.1 Motivation
Our interest in compound V is based on it being both a bis-amino acid and a
bicyclo[2.2.2]octane. This combination is interesting in medical chemistry, as it
provides a method to restrict peptide analogs to specific conformations. It is
common when designing such a drug compound to restrict its conformational
degrees of freedom, often by incorporating a cyclic structure annealed to the
backbone1'5. Doing so helps both to elucidate the active conformation, as well as to
force the molecule to spend more of its time near that conformation. This
combination is also interesting in materials chemistry, as a plastic could benefit
from the stiffness of the bicyclic core while also being biodegradable due to its
similarity to natural polymers67. Such a stiff polymer may even find use as
nanorods or other structures of nanotechnology.
A simple example of using cyclic peptide analogs as drug compounds is
seen in the penicillin class of antibiotics Penicillin acts as a suicide inhibitor of
glycopeptide transpeptidase, which links the C-terminus of a cell wall protein
(ending in D-alanine-D-alanine) with the N-terminal glycine of another protein.
Briefly, the enzyme breaks the last peptide bond of the one chain, allows the
terminal alanine to diffuse away, then allows the glycine end of the other chain to
approach and make a new peptide bond. The (3-lactam ring of penicillin has two
functions in inhibiting the transpeptidase. First, it acts to lock the rest of the
2


molecule in a structure resembling the active conformation of the natural substrate.
Second, after the enzyme breaks the amide bond, the fact that it was part of a ring
means that no part of the molecule is free to diffuse away. Thus no other peptide
can approach and form a new peptide bond, and the enzyme is stuck with its now-
permanent substrate.
These two functions of the P-lactam ring are both possible uses of
compound V. The bis-amino acid should be easily introducible into compounds
mimicking peptide chains. The inflexibility of the bicycle[2.2.2]octane core will
force the peptide mimic into a smaller range of conformations than would be
sampled by natural amino acid chains; for some applications this could enhance
drug binding. Its internal peptide bond, like that of penicillin, forms part of a ring;
thus it is possible that this compound could paralyze a protease by refusing to allow
another substrate to enter the active site and free the enzyme.
Continuing to consider the amide bond of compound V, it is seen to be a cis
bond. This configuration is higher energy for typical amino acid dimers, and thus it
can be more difficult to engineer into a peptidomimetic compound. Another
important feature of this dimer analog is that the terminal RNH2 and terminal
RCOOH bonds are nearly collinear. This provides the opportunity to engineer a
roughly straight section into a peptidomimetic compound. There are a number of
bridged bicyclic amino acid analogs in the literature9'11 that stress the usefulness of
3


such a conformationally restricted molecule. However, few if any force either a cis
peptide bond or a linear structure.
In materials science, bicyclooctane structures have for a long time seen use
for their rigidity and for their steric bulk. There are a couple recent uses in
developing technology that could benefit from adding peptide chemistry to these
rings.
S. Ito et al. report stabilizing the conformations of polypyrrole conductive
polymers by fusing bicyclo[2.2.2]octane rings onto the pyrrole subunits12. The
properties of these materials are strongly dependent on the morphology of the
pyrrole chains. The rigidity of the group turns out to be more effective than alkyl
chains in forcing the pyrrole groups into long conjugation lengths. Using structures
such as compound V would be a viable method to accomplish the same goals, but
with different polarity and solubility effects, as well as adding the capability to
further functionalize the subunits with peptide chemistry.
These rings have also seen some use in self-assembled monolayers (SAMs).
It is well known that typical, long-alkane chain SAMs form structures where the
chains are more or less elongated but tilted at an angle from the normal. This limits
the area that can be uniformly covered, as different regions on a surface will be
tilted different directions. S. Fujii et al. report using bicyclo[2.2.2]octanes to force
monolayers with the subunits oriented perpendicular to the surface13. This increases
4


the long-range order of the thin film. The rigidity of these structures also helps keep
thermal motion to a minimum, which is known to be an important disturbance in
alkane films. Molecules such as compound V may be a way to further orient similar
thin films, and potentially even create long-range dipoles parallel to the surface.
Bicyclo[2.2.2]octane rings have also seen use in nanotechnology, as rigid
spacers and wires that are called nanorods. An authoritative review6 several years
ago by Josef Michl and two of his former students identified a number of problems
facing the development of nanorods, but some of them might be easily dealt with
by oligomers of compound V. One of the issues facing nanorods in general is
solubility, which is not expected to be a major issue for short oligomers of
compound V, as it should have solubility properties similar to any other short
peptide chain. Two of the other major hurdles are the synthesis of a controlled size
distribution, and purification. Both of these may also turn out to be relatively easy,
as again the existing peptide technologies (including solid-phase synthesis) are
likely to be useful.
Other groups are also developing peptide-based answers to the difficulties
outlined in the previously mentioned review on nanorods. The Nowick group is
using a common technique from medicinal chemistry to introduce a stiff linear
region in a peptide analog attach the amino and acid functions to opposite sides of
an aromatic moiety7. This, of course, breaks the peptide backbone, and gives long
5


conjugation lengths. It would be useful to have another linear peptide nanorod in
the chemistry toolkit. The Schafmeister group is forming very well defined and stiff
nanorods out of bis-amino acids, but is completely removing freely rotating bonds
from their scaffolds14. Short oligomers of compound V would fill a need for stiff
but freely rotating nanorods built entirely from close peptide analogs. Such a
nanorods would likely form one of the two structures shown in Figure 1.
FIGURE 1.1. NANORODS OF COMPOUND V
Short oligomers of compound V may also show P-sheet or thin film
formation. Simple examination of the monomer reveals a polar face capable of
forming two hydrogen bonds (one donor and one acceptor) and a greasy face. It is
not hard to imagine then either an anti-parallel P-sheet-like structure or a self-
assembled monolayer at an oil-water interface might form, as in Figure 2.
6


a)
o
//
o
FIGURE 1.2. SELF ASSEMBLY OF COMPOUND V POLYMERS
1.2 Background
Our interest in this molecule has grown from earlier work on the precursor
molecules, compounds II and III. The first student to work on this series of
potentially useful protected forms, including compound III. She was able to
optimize a standard Strecker synthesis of compound II, but was able to make only a
couple derivatives. Compound II is almost totally insoluble under most useful
reaction conditions. The N-formylated and N-acylated compounds were, however,
successfully made.
compounds was Carolina Wilson15. She was pursuing compound II and its
compound II
corrpound III
compound IV
FIGURE 1.3. PREVIOUSLY SYNTHESIZED COMPOUNDS
7


The next students to work with these were able to make a small amount of
progress. Samuel Allen was able to obtain high-quality crystals of the copper
sulfate salt of compound II and verify its structure by X-ray crystallography16. This
confirmed that the compounds I and II were collected exclusively in the trans
conformation.
Rentsenmyadag Dashzeveg continued work on trying to.derivatize
i -T
compound II She tried several methods of esterification on both compound II and
its formylated derivative. Although most of these met with little success, there was
one attempt that formed a small amount of compounds III and IV. Unfortunately
there was not enough product to thoroughly characterize. She also tried to obtain
compound III by direct methanolysis of compound I, but with no success.
About this time we came across an old report of the synthesis for compound
III18. It followed just the path that both Wilson and Dashzeveg had attempted an
HCl/MeOH Fischer esterification. This motivated us to revisit these conditions.
Wilson and Dashzeveg both report that compound II only dissolves in acidic
methanol in a narrow range of acid concentration, and precipitates if more HC1 is
added. Overnight refluxes under solution conditions, as well as under more acidic
conditions when the starting material was in suspension, had failed to turn up any
results. Piper et al. reported saturating with HC1 and refluxing for several days until
forming a complete solution. This was partially reproduced by Dashzeveg, who
8


found that this method gave poor yields and also produced compound IV. In fact,
Dashzeveg succeeded in completely converting the sample to compound IV by
heating in methanol.
It is starting at this point, when we had a seen trace amount of compounds
III and IV in an NMR spectrum and then found reports that the esterification
proceeded well if left to run for several days, that the currently reported research
began. Since the completion of this work, another preparation of compound III has
been reported19, where compound II is N-protected, then esterified by methyl
iodide and the amines deprotected.
9


2. Results and Discussion
The monomer 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-l -carboxylic acid
(compound V) and potentially useful protected forms of it have been prepared
according to scheme 2. Early attempts at dimerizing the monomer have not been
successful.
H2N
-------- NC
6^ HjN
1,4-cyclohexadione compound I
NHj
compound II
MeOOC
NHj
^^COOMa
compound III
compound VIII compound VI compound VII
SCHEME 2
10


2.1 Synthesis of l,4-diaminocyclohexane-l,4-dicarbonitrile
(compound I) and l,4-diaminocycIohexane-l,4-
dicarboxylic acid (compound II)
Scheme II proceeded easily through the first two steps, following a
procedure by Wilson15. This was a Strecker method where a ketone was mixed with
ammonium chloride and a cyanide salt to make a cyanoamine, which was then
refluxed in aqueous HC1 to hydrolyze the cyano group. The procedure reliably
forms compound II in 50-60% overall yield from 1,4-cyclohexadione.
While compound I could be recrystallized from hot water, it was simpler to
use the crude to make compound II and purify at that time. The purification of
compound II was simple, since it is not soluble in organics and only soluble in
water at high or low pH (when it carries a net charge). Filtering a basic solution of
compound II removed any unreacted compound I. The solution was then
neutralized, which precipitated compound II. The solids were collected and washed
with methanol, then dried in a vacuum dessicator to yield pure compound II.
Compounds I and II were identified by comparing both 'H-NMR and IR
spectra with known samples. The structure of compound II has previously been
verified by x-ray crystallography16, and is only found in the trans isomer. The
spectra of these compounds have been described and discussed by Wilson.
11


2.2 Synthesis of dimethyl l,4-diaminocyclohexane-l,4-
dicarboxylate (compound III)
Compound III was made with a Fischer esterification, using dry methanol
and either hydrogen chloride gas or concentrated sulfuric acid. The two acids give
remarkably different reaction conditions and impose different work-up
requirements that are worth discussing.
Piper et al. report the synthesis of this compound in a Fischer esterification
using HC1 gas We find that, if hydrochloride gas was passed through a
methanolic suspension of compound II, the suspension dissolved and then
precipitated as more acid was introduced. Any attempts made while the starting
material was dissolved were unsuccessful (reflux times of up to two weeks have
been tried). It appears that there was only enough acid to catalyze the esterification
if there was also enough acid to minimize the solubility of the starting material.
This then forced week-long refluxes, during which the gaseous acid could escape
through standard condensers. Sometimes it was necessary to re-saturate the mixture
with HC1 gas every few days. It should also be mentioned than no product was ever
observed unless the solvents used had all been rigorously dried. For example,
methanol was stirred over sodium metal overnight before distillation. It was not
necessary to run the reactions under inert atmosphere, however.
12


After several days (typically about a week) of refluxing the suspension, 'H-
NMR spectra of the reaction mix showed 85-90% product and 10-15% starting
material (both as the HC1 salt). These numbers were obtained by comparing the
integral of the methyl ester peak to the total signal for ring protons, which overlap
for the two compounds. The calculation assumed that these were the only two
cyclohexane derivatives present, although compound IV could sometimes be found
in small amounts.
Aqueous workups of these reactions always yielded starting material. It
appears that this compound, as well as compound IV, is particularly susceptible to
base hydrolysis. Any aqueous environment basic enough to get the free base of the
product was basic enough to also hydrolyze the product. Therefore, the reaction
was worked up with organic solvents. TEA/Et20 gave poor results. NH3/Et20 (as
reported by Piper et al.) unpredictably yielded either compound III or compound
IV, or a mixture, but in good yields. NaH/DMF/DCM gave poor yields but pure
compound III, and is the workup reported in the experimental section.
The compound is identified by its 'H-NMR spectrum (Figure 2.2.1), which
looks similar to those for compounds I and II, but with the addition of a methyl
ester signal at 3.73 ppm. The IR spectrum (figure 2.2.2) confirms that the acid has
been converted to an ester.
13


FIGURE 2.2.1 H-NMR SPECTRUM OF COMPOUND III in CDC13
file: C:\Documentsand Settinas\scistudent\Desktop\SpectralData\NMRFs\Bp103a cdc!3.fid\fid block# 1 expt: ftfeijro ppm: 199 975067 MHz
Hz processed size: 65538 complex points
time domain size: 9998 points
width: 1999.75 Hz 9.999955 ppm 0.200015 Hz/pt
number of scans: 8
LB: 0.000 GB: 0.0000


FIGURE 2.2.2 FT-IR SPECTRUM OF COMPOUND III
35^ .............
4000 3500 3000 2500 2000 1500 1000........... 500
Wavenumbers (cm-1)


The long refluxes and hassles of working with HC1 gas lead to an
investigation of sulfuric acid as a catalyst. Dissolving the suspension of compound
II required relatively large amounts of sulfuric acid (about 1.25 mL per gram of
starting material), but allows the reaction to proceed to completion during an
overnight reflux. During the workup it became difficult to remove the product from
the sulfate salts; in the future it might be beneficial to investigate the use of ion-
exchange chromatography. It was also difficult to prevent the product from
cyclizing into compound IV.
16


2.3 Synthesis of methyl 4-amino-3-oxo-2-
azabicyclo[2.2.2]octane-l-carboxylate (compound IV)
As mentioned in the previous section, f^SCVcatalyzed Fischer
esterifications often caused the diester III to cyclize into compound IV. This could
be obtained in high yields, but it was difficult to remove uncyclized diester and
sulfate salts without hydrolyzing the ester.
The complex patterns found in !H-NMR spectra of this substituted
bicyclo[2.2.2]octane hinder detailed spectral analysis of these compounds; this is a
result of all the ring protons experiencing both vicinal and geminal coupling, as
well as losing half of the symmetry of the compounds described earlier. We can
infer a cyclized product, however, on several lines of evidence. First, the
complexity itself suggests a different situation than seen in the previous three
compounds, which all had simple spectra. An amide peak (6.92 ppm) has appeared
in the 'H-NMR, and two carbonyl peaks now appear in the IR: one at a
characteristic frequency for esters (1736 cm*1) and one at a characteristic frequency
for amides (1669 cm"1). Also, the methyl ester NMR peak has moved from 3.73
ppm to 3.82 ppm, a consequence of the greater electron-withdrawing power of an
amide nitrogen compared to an amino nitrogen. Finally, this methyl peak now
integrates to 3 if the region from 1.5-2.5 ppm is set equal to 10 (ring plus amino
protons) rather than integrating to 6 if the downfield region is set to 12.
17


FIGURE 2.3.1 H-NMR SPECTRUM OF COMPOUND IV IN MeOD
(NtM
II \\f I
I
3.2
I
2.4
I
2.0
r~
0.8
r~
0.4
file: C:\Documents and Settings\sdstudent\Desktop\Spectral Data\NMR FID$\Bp103f meod.fkftfid block# 1 expt: HouCT 0 ppm: 1
transmitter freq.: 199.976757 MHz processed size-
time domain size: 9998 points LB: 0.200
width: 1999.76 Hz> 9.999955 ppm > 0.200016 Hz/pt
number of scans: 8
199.975861 MHz
65536 complex points
GB: 0.0000


2.4 Synthesis of 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-
1-carboxylic acid (compound V)
It was mentioned earlier, in the discussion of compound III, that any
aqueous solution basic enough to yield the free base would also hydrolyze the ester
functions. It turns out that, if the pH was kept around 11 or 12 (judged by pH
paper), the cyclization occured faster than hydrolysis, and the hydrolyzed product
obtained was compound V rather than compound II.
Compound V was easily recrystallized from water kept at a pH near 5. !H-
NMR in basic water shows a ring-proton region with the same pattern as for
compound IV, and no ester peak.
19


0000 0
S169 l
909M
CS19 1
6068 1
6006'l
Z196 1-
moz
£960 2 -7=
26012 J
3=
i!
2*5.
£gg
ss§
oxo m
ro
e.m
=S8
o o
*5*
CM
X
9
2
£
I
at
0*.
5s
o3
Cm C
fSi
(0
iriis
Biggs
slfil
FIGURE 2.4.1 H-NMR SPRECTRUM OF COMPOUND V IN D20, NaOD
20


FIGURE 2.4.2 H-NMR SPECTRUM OF COMPOUND V IN D20, DC1
21
file: C:\Documents and Setting5\scisUident\Desktop'Spectral DataWMR FIDs\Bp42c!phDSS.fid\fid block# 1 expt: 'W&utff 0 ppm: 199.975544 MHz
transmitterfreg.: 199.976563 MHz processed size: 65536 complex^
time domain size: 11998 points LB: 0.000 GB:0.0000
width: 2399.71 Hz 11.999940 ppm 0.200009 Hz/pt
number of scans: 8


FIGURE 2.4.3 FT-IR SPECTRUM OF COMPOUND V
Wavenumbers (cm-1)


2.5 Synthesis of 4-[(te/*-butoxycarbonyI)amino]-3-oxo-2-
azabicyclo[2.2.2]octane-l-carboxylic acid (compound VI)
A ter/-butoxycarbonyl (BOC) group was introduced onto compound V
following a method by Khalil et al. Briefly, they found that standard methods of
introducing BOC groups onto a-amino acids (such as given by Bodanszky and
Bodanszky21: dioxane, water, NaOH, and BOC anhydride) were insufficient for
amines attached to tertiary carbons. These sterically hindered amines were slowed
down to the point that the hydroxide ion could compete as a nucleophile in
attacking the BOC anhydride. They solved the problem by introducing the phase-
transfer catalyst tetramethylammonium hydroxide, and performing the reaction
without an aqueous phase.
Although the methods of Khalil et al. did give better yields that earlier
attempts using conditions from Bodanszky and Bodanszky, the reaction still only
proceeded in -50% yield. The reaction is hindered by poor solubility of the starting
material in organic solvents. The ]H-NMR spectrum shows a tall singlet at 1.44
ppm that integrates to 9, relative to ring protons that sum to 8.
23


FIGURE 2.5.1 H-NMR SPECTRUM OF COMPOUND VI IN MeOD
r~
7.2
I
6.0
I
5.2
I
4.4
3.6
I
3 2
24
~r~
2.0
file: C:\Documents and Settings\sdstudent\Desktop\Spectral DataVNMR FIDs\6pl07a meod.fid\fid block# 1 expt: fafipuJTO ppm: 199 975862 MHz
transmitter freq.: 199.976757 MHz ~ processed size: 65536 complex points
time domain size: 9996 points [n- 0 000 GB 0 0000
width: 1999.76 Hz* 9.999955 ppm 0.200016 Hz/pt
number of scans: 6


FIGURE 2.5.2 FT-IR SPECTRUM OF COMPOUND VI


2.6 Synthesis of methyl 4-[(terf-butoxycarbonyl)amino]-3-
oxo-2-azabicyclo[2.2.2] octane-1 -carboxylate
(compound VII)
The low yields encountered in the protection of compound V lead to an
investigation of introducing the protecting group earlier in the synthesis. It was
hoped that the better solubility characteristics of compound IV would make it
easier to protect. The reaction was carried out under conditions similar to those of
Khalil et al.20, but without the phase transfer catalyst. This gave -50% yield on the
first attempt, and it is likely that this reaction can be optimized to a much higher
yield.
The 'H-NMR spectrum looks similar to that of compound VI, with an
additional peak at 3.82 ppm for the methyl ester. Compared to compound VI, a pair
of ring protons are shifted downfield to 3.0 ppm and another pair shifted upfield to
about 1.5 ppm. The IR spectrum clearly shows three carbonyls: one ester at 1736
cm'1, and two amides at 1687 and 1648 cm'1.
26


FIGURE 2.6.1 H-NMR SPECTRUM OF COMPOUND VII IN CDC13
file: C:\Documents and Settings\scistudent\Desktop\Spectral Data\NMR RDs\Bp111d cdcl3.fidVid block# 1 exert: fttautfTOppm: 199.975075 MHz
transmitterfreg.: 199.975969 MHz processed size: 65536 complex points
time domain size: 9998 points LB- 0 000 GB' 0 0000
width: 1999.75 Hz = 9.999955 ppm = 0.200015 Hz/pt
number of scans: 32
2.544


2.7 Attempted synthesis of 4-[(4-Amino-3-oxo-2-aza-
bicyclo[2.2.2]octane-l-carbonyl)-amino]-3-oxo-2-aza-
bicyc!o[2.2.2]octane-l-carboxylic acid methyl ester
(compound VIII)
Early attempts at coupling compound VI with compound IV have not been
successful. Following small variations on a procedure by Bodanszky and
Bodanszky21, compound VI was mixed with l-(3-dimethylaminopropyl)-3-
ethylcarbodiimide hydrochloride (EDC), a water-soluble analog of DCC.
Compound IV was then introduced. This was stirred at room temperature for two
days before quenching and workup. The crude product showed 2 spots on a TLC
plate which were separated on a column. These two fractions were then shown to
each be composed of two very closely spaced spots on a TLC plate and were not
further purified.
The NMR of the fast-moving fraction suggests the coupling may have
worked, although no conclusion can be drawn. The signal at 3.0 ppm is similar to
what is seen in the same place for compound VII, the doubly-protected monomer.
The signal at 2.5 ppm resembles a peak in the same location in compound VI,
although unreacted acid should have been lost in the workup. It may be that the
coupling was successful, but the workup conditions hydrolyzed the ester. Repeating
the procedure with milder workup conditions, or with a workup designed to isolate
the acid product may give further insight.
28


FIGURE 2.7.1 H-NMR SPECTRUM OF PRODUCTS FROM EDC COUPLING
29


3. Experimental
3.1 Reagents
Sodium cyanide was purchased from EM Sciences. 1,4-cyclohexadione and
ammonium chloride were purchased from Acros organics. Potassium cyanide was
purchased from Mallinkrodt Chemical Works. Ninhydrin was purchased from
Pierce Chemical. BOC2O was purchased from Aldrich. All acids and solvents were
purchased from either Aldrich or Fischer.
3.2 General Methods
3.2.1 Thin Layer Chromatography
Thin layer chromatography (TLC) was run on aluminum sheets coated with
a 0.25 mm of silica gel with a UV254 dopant (Whatman, Ltd, AL SIL G/UV).
Samples were spotted with 1 pL micropipets that had been flame-drawn to a point.
Chromatograms were developed in either 9:1 chloroform/methanol or 9:1
dichloromethane/methanol systems, specifically noted for each compound in the
synthetic procedures. Developed plates were visualized with one or more of the
following methods, also specifically noted for each compound in the synthetic
procedures.
30


Method A: The plate was observed under ultraviolet light at 254 nm, after allowing
solvent to evaporate. The silica gel on the plates contains a fluorescent
material which is quenched by some analytes.
Method B: The plate is immersed in silica gel saturated with iodine for about a
minute.
Method C: The plate is sprayed with a ninhydrin solution (1 g ninhydrin in 100 mL
acetone), then placed in a 110C oven for a few minutes.
Method D: The plate is briefly dipped in a basic potassium permanganate solution
(1.5 g potassium permanganate, 10 g potassium carbonate, and 1.25 mL
10% aqueous sodium hydroxide in 200 mL water), then placed in a
110C oven for a few minutes.
3.2.2 Nuclear Magnetic Resonance
Nuclear magnetic resonance (NMR) spectra were recorded on a Varian 200
MHz spectrometer. Samples were dissolved in either D2O, D2O/DCI, D20/NaOD,
CDCI3 or CD3OD. Organic solvents contain tetramethyl silane (TMS) as a standard
and aqueous solvents contain sodium 3-(trimethylsilyl)-l-propanesulfonate (DSS)
as a standard. Chemical shifts are reported as ppm.
3.2.3 Infrared spectroscopy
Infrared (IR) spectra were recorded with a Thermo-Nicolet Avatar 360
E.S.P. FTIR spectrometer running OMNIC software. Samples were ground with a
31


mortar and pestle before being applied directly onto the instrument lens for
analysis.
32


3.3 Synthetic Procedure
3.3.1 Synthesis of l,4-diaminocyclohexane-l,4-
dicarbonitrile (compound I)
1,4-cyclohexadione (18.6 g, 166 mmol) was added to 200 ml of 50%
aqueous methanol with stirring. Ammonium chloride (18.4 g, 345 mmol) and then
sodium cyanide (16.3 g, 333 mmol) were added. The mixture was stirred at room
temperature for 2 days and then filtered. Methanol was removed from the filtrate
under reduced pressure and remaining water filtered again. The two crops of
precipitate were combined to yield 26.4 g (97%) of compound I, used without
purification.
TLC analysis (9:1 methylene chloride-methanol): a sample dissolved in methanol
was a single spot with an Rf of 0.7, visualized with method D.
'H-NMR (CD3OD), 5 (ppm): 2.20 (4H, doublet J=9.4); 1.80 (4H, doublet J=9.8).
IR frequency (cm1): 3339 (NH), 3280 (NH), 2954 (CH), 2233 (CN).
33


3.3.2 Synthesis of l,4-diaminocyclohexane-l,4-
dicarboxylic acid (compound II)
On an ice bath, 26.4 g (161 mmol) of compound I was added to 500 mL of
12N HC1, which was then stirred at room temperature for 2 days. The mixture was
then diluted with 400 mL of water and refluxed for 3 days. This mixture was
concentrated under reduced pressure to about 100 mL and placed in an ice bath.
220 mL 6N NaOH was added, the mixture filtered and the solids discarded. The
filtrate was neutralized with 2N HC1 and the precipitate collected by filtration. The
solids were washed five times with 100 mL methanol and dried under reduced
pressure to yield 18.9 g (58%) of compound II.
NMR: 'H-NMR (D20, NaOD), 5 (ppm): 1.82 (4H, doublet J=9.4); 1.28 (4H,
J=10). 'H-NMR (D20, DC1), 5 (ppm): 2.01 (4H, triplet J=9.7); 1.94 (4H, triplet
J=9.7).
IR: IR frequency (cm'1): 3450 (NH), 3215 (NH), 3100-2500 (OH, CH), 1578
(C=0).
34


3.3.3 Synthesis of dimethyl l,4-diaminocyclohexane-l,4-
dicarboxylate (compound III)
250 mL of methanol (freshly distilled from sodium metal) was saturated
with HC1 gas. Compound II (10.0 g 49.5 mmol) was then added to the solution
and HC1 gas passed through for an additional 10 minutes. This mixture was
refluxed for 6 days, during which 100 mL methanol was added to maintain volume.
The mixture was brought to dryness under reduced pressure to yield 13.6 g of a
crude product (NMR showed 10% HC1 salt of compound II and 90% HC1 salt of
compound III).
11.0 g of the crude was suspended in 100 mL DMF and placed under nitrogen gas.
The suspension was added via cannula to 11.0 g NaH (60% in mineral oil, under
nitrogen and on an ice bath) and stirred for 30min. 1.35 L methylene chloride was
added and the suspension stirred another 30 minutes before filtering. The filtrate
was concentrated under vacuum to yield 2.5 g (27%) of compound III. The solids
were stirred for 20 minutes in 200 ml 1:1 methylene chloride / methanol and
refiltered. This filtrate yielded 3.75 g (51%) of compound V.
35


NMR: 'H-NMR (CDCI3), 5 (ppm): 3.73 (6H, singlet, methyl); 2.23 (4H, doublet
J=9.8, ring protons); 1.57 (5H, singlet, amino and water); 1.46 (4H, doublet J=9.4,
ring protons).
IR: IR frequency (cm'1): 3359 (NH), 3299 (NH), 2920 (CH), 2853 (CH), 1716
(C=0), 1223 (C-O), 1046 (C-0).
36


3.3.4 Synthesis of methyl 4-amino-3-oxo-2-
azabicyclo[2.2.2]octane-l-carboxylate (compound IV)
To 25 mL methanol was added 1.00 g (4.9 mmol) of compound II and 1.25
mL concentrated sulfuric acid. This mixture was refluxed for 2 days then poured
into 100 mL chloroform on an ice bath. Ammonia gas was passed through this
solution for 10 minutes. The mixture was then refluxed for 1 hour and filtered. The
filtrate was concentrated to yield crude compound IV (1.28 g, 112%). The mixture
hydrolyzed into compound V during purification attempts.
NMR: 'H-NMR (CDCI3), 5 (ppm): 6.92 (1H, singlet, amide); 3.82 (3H, singlet,
methyl); 2.22-1.66 (11H, unresolved amino, water and ring protons).
IR: IR frequency (cm*1): 3352 (NH), 3292 (NH), 3225 (NH), 2919 (CH), 2866
(CH), 1736 (C=0 ester), 1669 (C=0 amide), 1271 (C-O), 1081 (C-O).
37


3.3.5 Synthesis of 4-amino-3-oxo-2-azabicyclo[2.2.2]octane-
1-carboxylic acid (compound V)
A stock solution of 0.5 mL concentrated sulfuric acid solution and 9.5 mL
methanol was made. To 2.5 mL of this solution was added 117 mg (0.58 mmol) of
compound II, the mixture placed in a pressure tube and put on a 65 C oil bath. The
tube was transferred to an ice bath after 2 days, and 0.83 mL 6N NaOH was added.
The tube was allowed to sit at room temperature for 2hr before being concentrated
under reduced pressure to about 1.5 mL. This was neutralized with 2M HC1 and let
sit overnight before filtering to yield 25 mg (23%) of compound V.
NMR: 'H-NMR (D2O, NaOD), 8 (ppm): 2.16-1.56 (8H, unresolved ring protons).
:H-NMR (D2O, DC1), 8 (ppm): 2.30 (4H, unresolved ring protons); 2.01 (4H,
unresolved ring protons).
IR: IR frequency (cm'1): 3512 (NH), 3280 (NH), 3113-2469 (NH3+ and CH), 1691
(C=0 amide), 1540 (OO acid).
38


3.3.6 Synthesis of 4-[(tert-butoxycarbonyI)amino]-3-oxo-2-
azabicyclo[2.2.2]octane-l-carboxy!ic acid (compound VI)
To 20mL of acetonitrile (freshly distilled from sodium metal) was added
329 mg (1.8 mmol) of compound V and 373 mg (4.1 mmol) of
tetramethylammonium hydroxide pentahydrate and the suspension allowed to stir
for 1 hour. Di-tert-butyl dicarbonate (673 mg, 3.08 mmol) was then added and the
suspension stirred for 4 days. The mixture was concentrated under reduced pressure
and the residue partitioned between water and ethyl acetate. The aqueous layer was
washed once with fresh ethyl acetate and the organic layers discarded. The aqueous
layer was acidified to a pH of 2 with 1 M NaHSC>4 and extracted 3 times with ethyl
acetate. The organics were combined, dried over MgSCU, and concentrated to yield
256 mg (52%) of compound VI.
NMR: 'H-NMR (CD3OD), 5 (ppm): 2.56 (2H, doubled triplet J=10.6, 3, ring
protons); 2.20 (2H, doubled triplet J=12, 3.6, ring protons); 1.80 (4H, doubled
pentet, J= 12, 3.4, ring protons); 1.44 (9H, singlet, tert-butyl protons).
IR: IR frequency (cm1): 3441 (NH), 3207 (NH), 3100-2555 (OH and CH), 1712
(C=0 acid), 1667 (OO amide), 1152 (C-O).
39


3.3.7 Synthesis of methyl 4-[(te/*-butoxycarbonyl)amino]-
3-oxo-2-azabicyclo[2.2.2]octane-l-carboxylate
(compound VII)
251 mg (1.27 mmol) of compound IV is suspended in 10 mL of acetonitrile
(freshly distilled from Caty. Di-tert-butyl dicarbonate (450 mg, 2.06 mmol) was
added and the mixture stirred at room temperature for 2 days. The mixture was
evaporated under reduced pressure and the residue stirred in 10 mL 50% methylene
chloride acetic acid for 30 minutes. This mixture was then evaporated under
reduced pressure and the residue partitioned between 10 mL water and 10 mL
methylene chloride. The aqueous layer was adjusted to a pH of 2 with 1 M
NaHSC>4. The layers were separated, the aquoues extracted twice with methylene
chloride, and the organics combined. These were washed once with brine, dried
over MgSC>4, and concentrated. This residue was purified by column
chromatography (eluted with chloroform) to yield 200 mg (53%) of compound VII.
NMR: *H-NMR (CDCI3), 8 (ppm): 7.10 (1H, singlet, amide); 6.06 (1H, singlet,
amide); 3.82 (3H, singlet, methyl); 2.96 (2H, triplet J=9.8, ring protons), 2.18 (2H,
triplet, J=7.6, ring protons); 1.86 (2H, triplet, J=12.8, ring protons), 1.57 (4H,
triplet, J=11.6, ring protons and water); 1.43 (9H, singlet, tert-butyl).
IR: IR frequency (cm1): 3334 (NH), 2951 (CH), 1736 (C=0 ester), 1687 (C=0
amide), 1648 (C=0 amide), 1257 (C-O).
40


3.3.8 Attempted synthesis of 4-[(4-Amino-3-oxo-2-aza-
bicycIo[2.2.2]octane-l-carbonyI)-amino]-3-oxo-2-
aza-bicyclo[2.2.2)octane-l-carboxylic acid methyl
ester (compound VIII)
To 25 mL of freshly distilled methylene chloride was added 162 mg (0.57
mmol) of compound VI, then 126mg (0.66 mmol) of l-(3-dimethylaminopropyl)-3-
ethylcarbodiimide hydrochloride, then 112 mg (0.57 mmol) of compound IV.
Precipitate began forming immediately. The mixture was stirred at room
temperature for 2 days.
TLC (9:1 methylene chloride-methanol. Visualized with method D) at this point
showed no starting material and 3 new spots (0.4, 0.6, and 0.9).
The reaction was quenched by the addition of 10 mL 50% acetic acid-water and
stirred for 4 hours. The organic layer was washed with brine, dried over MgSCL,
and concentrated, yielding 114 mg of crude (TLC shows spots at 0.4 and 0.5). This
mixture is separated on a silica gel column, eluting with 19:1 chloroform-methanol
to yield two samples. One showed two closely spaced spots on a TLC with Rfs
about 0.4 and the other showed two closely spaced spots on a TLC with Rfs about
0.5. Neither could be clearly identified.
41


REFERENCES
1. Cowell, SM; Lee, YS; Cain, JP; Hruby, VJ. Current Medicinal Chemistry (2004)
11:2785-2798
2. Freidinger, RM; Veber, DF; Perlow, DS; Brooks, JR; Saperstein, R. Science
(1980)210:656-658
3. Trabocchi, A; Scarpi, D; Guama, A. Amino Acids (2008) 34:1-24
4. Gillespie, P; Cicariello, J; Olson, GL. Biopolymers (1997) 43:191-217
5. Hanessian, S; McNaughton-Smith, G; Lombart, HG; Lubell, WD. Tetrahedron
(1997) 53(38): 12789-12854
6. Schwab, PFH; Levin, MD; Michl, J. Chemical Reviews (1999) 99:1863-1933
7. Gothard, CM; Rao, NA; Nowick, JS. Journal of the American Chemical Society
(2007) 129:7272-7273
8. Berg, JM; Tymoczko, JL; Stryer, L. Biochemistry, 5th ed. New York. W. H.
Freeman and Company, 2002. p. 214-216
9. Hart, BP; Rapoport, H. Journal of Organic Chemistry (1999) 64:2050-2056
10. Ranganathan, D; Haridas, V; Kurur, S; Thomas, A; Madhusudanan, KP;
Nagaraj, R; Kunwar, AC; Sarma, AVS; Karle, IL. Journal of the American
Chemical Society (1998) 120:8448-8460
11. Dinsmore, CJ; Bergman, JM; Bogusky, MJ; Culberson, JC; Hamilton, KA;
Graham, SL. Organic Letters (2001), 3(6):865-868
12. Ito, S; Watanabe, H; Uno, H; Murashima, T; Ono, N; Tsai, YC; Compton, RG.
Tetrahedron Letters (2001) 42:707-710
13. Fujii, S; Akiba, U; Fujihira, M. Chemical Communications (2001) 1688-1689
42


14. Levins, CG; Schafineister, CE. Journal of the American Chemical Society
(2003) 125:4702-4703
15. Wilson, C. Synthesis of 1.4-diamino-1.4-cvclohexanedicarboxylic acid, a New
Polymerization Monomer, and Its Derivatives, University of Colorado at Denver,
1999
16. Dashzeveg, R; Wilson, CR; Allen, S; Dyckes, DF. Chemistry and Chemical
Technology, National University of Mongolia (2006) 6:82-89
17. Dashzeveg, R. Exploration of Pathways for the Synthesis of dimethyl- 1.4-
diamino- 1.4-cvclohexane dicarbozvlate, a Novel Bis-Amino Acid Ester. University
of Colorado at Denver, 2003
18. Piper, JR; Stringfellow, Jr., CR; Johnston, TP. Journal of Medicinal Chemistry
(1966) 9(6):911-920
19. Chu, Y; Lynch, V; Iverson, BL; Tetrahedron (2006) 62:5536-5548
20. Khalil, EM; Subasinghe, NL; Johnson, RL; Tetrahedron Letters (1996)
37(20):3441-3444
21. Bodanszky, M; Bodanszky, A; The Practice of Peptide Synthesis. New York.
Springer-Verlag. 1984
43