Synthesis and characterization of CIS methyl 4-amino-5-oxopyrrolidine-2-carboxylate

Material Information

Synthesis and characterization of CIS methyl 4-amino-5-oxopyrrolidine-2-carboxylate
Axen, Melissa Anne
Publication Date:
Physical Description:
xi, 51 leaves : ; 28 cm


Subjects / Keywords:
Racemization ( lcsh )
Amino acids -- Synthesis ( lcsh )
Amino acids -- Synthesis ( fast )
Racemization ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 49-51).
Department of Chemistry
Statement of Responsibility:
by Melissa Anne Axen.

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:
318536485 ( OCLC )
LD1193.L46 2008m A83 ( lcc )

Full Text
Melissa Anne Axen
B.S., University of Colorado Denver, 2007
A thesis submitted to the
University of Colorado Denver
In partial fulfillment
of the requirements for the degree of
Master of Science

This thesis for the Master of Science
Degree by
Melissa Anne Axen
has been approved


Axen Melissa Anne (M.S., Chemistry)
Preparation of Syn methyl 4-amino-5-oxopyrrolidine-2-carboxylate, a y-amino acid
Thesis directed by Professor Douglas F. Dyckes
A racemic mixture of a synthetic y-amino acid, (R,R) and (S,S) methyl-4-
amino-5-oxopyrrolidine-2-carboxylate, has been produced from glutaric acid via a
five step synthesis. The procedure utilizes common, inexpensive starting materials
and proceeds under relatively gentle conditions. The production of a racemic mixture
of (2R,4S)- and (2S,4R)-diaminopentanedioic acid as an intermediate is also of
interest as it is generated through a simplified method compared to previously
published routes. Compound identities were confirmed with ER and 'H-NMR
This abstract accurately represents the content of the candidatess thesis. I recommend its
Douglas F. Dyckes

I dedicate this thesis to my family, who taught me the value of education of
provided unfaltering support during my education. I would especially like to thank
my father and Dave Baughman, who pointed me in the direction of chemistry at a
young age. This work would also not have been complete without my mother, who
managed to keep the family, and more importantly her sanity, intact even as the only
island of reality in an ocean of engineers, chemists and geneticists.

I would like to thank Dr. J. Hoyt Meyer for his continued patience, commitment to
teaching and willingness to answer even the most random questions. I would also
like to acknowledge my advisor, Dr. D.F. Dyckes for changing the direction of my
career and teaching me to trust myself and my data.

1. Synthetic amino acids in peptidomimetics and peptide nanotubes............1
1.1. Research Goal Cis methyl 4-amino-5-oxopyrrolidine-2-carboxylate.........4
1.1.2 Previous Synthetic Strategies..........................................6
2. Results and Discussion...................................................12
2.1 Results for the synthesis of 2,4-dibromopentanedioyl dibromide (1) and 2,4-
dibromopentanedioic acid (2)........................................14
2.2 Results for the synthesis of 2,4-diaminopentanedioic acid (3)..........22
2.3 Results for the synthesis of 2,4-diaminopentanedioic acid dimethyl ester (4)....25
2.4 Results for the synthesis of
methyl 4-amino-5-oxopyrrolidine-2-carboxylate (5)..................29
2.4.1 Results for the Synthesis of 5 by deprotonation and salt precipitation.29
2.4.2 Results for the Synthesis of 5 by amine protection...................35
3 Conclusions and Future Studies..........................................37
4 Experimental............................................................38
4.1 Reagents...............................................................38
4.2 General Methods.......................................................38
4.2.1 Nuclear Magnetic Resonance Spectroscopy..............................38
4.2.2 Infrared Spectroscopy................................................38
4.2.3 Thin Layer Chromatography............................................39
4.3 Synthetic procedures.................................................39

4.3.1 Procedure for drying bromine...........................................39
4.3.2 Procedure for drying methanol..........................................39
4.3.3 Procedure for saturating methanol with HCl(g)..........................39
4.3.4 Procedure for saturating methanol with NH3(g).........................40
4.3.5 Procedure for drying Methylene chloride...............................40
4.3.6 Procedure for drying Triethylamine....................................40
4.3.7 Synthesis of 2,4-dibromopentanedioyl dibromide (1) (PBr3).............40
4.3.8 Synthesis of 2,4-dibromopentanedioyl dibromide (1) (PCI3).............41
4.3.9 Attempted synthesis of 2,4-diiodopentanedioyl diiodide (PBr3).........42
4.3.10 Attempted synthesis of 2,4- diiodopentanedioyl diiodide with Cu salts.42
4.3.11 Synthesis of 2,4-dibromopentanedioic acid (2)........................43
4.3.12 Synthesis of 2,4-diaminopentanedioic acid (3).........................43
4.3.13 Synthesis of 2,4-diaminopentanedioic acid dimethyl ester sulfate salt.44
4.3.14 Attempted preparation of 5 from sulfate salt of 4
by precipitation of BaSC>4............................................45
4.3.15 Attempted preparation of 5 from sulfate salt of 4....................45
4.3.16 Preparation of methyl 4-amino-5-oxopyrrolidine-2-carboxylate (5)
with basic methanol...................................................46
4.3. 17 Preparation of methyl 4-amino-5-oxopyrrolidine-2-carboxylate (5)
with basic chloroform.................................................46
4.3.18 Preparation of 2,4-Bis-(9H-fluoren-9-ylmethoxycarbonylamino)-pentanedioic
acid (finoc protected 3)..............................................47
4.3.19 Attempted Preparation of 2,4-Bis-benzyloxycarbonylamino-pentanedioic
acid (Cbz protected 3)................................................48

1. Wire model of a six member hypercycle.................................3
2. Space filling model of a six member hypercycle........................3
3. Proposed route to a peptide nanotube..................................4
4. Cis methyl 4-amino-5-oxopyrrolidine-2-carboxylate.....................5
5. Cis amino acid (R,R) or (S,S).........................................13
6. Trans amino acid (R,S) or (S,R).......................................13
7. Meso and chiral isomers of 2..........................................16
8. 'H NMR data for product 3, chiral and meso isomers....................17
9. Coupling tree for Ha and Ha-. ........................................18
10. Coupling tree for Hc and He.........................................19
11. Coupling tree for Ha and Hd..........L..............................19
12. The NMR spectrum of product 2, chiral isomer.........................20
13. The IR spectrum of 2.................................................21
14. The *H NMR spectrum of Product 3.....................................23
15. The IR spectrum of 3.................................................24
16. The 'H NMR spectrum of Product 4.....................................27
17. The IR spectrum of 4 as a sulfate salt...............................28
18. The *H NMR spectrum of Product 5.....................................33
19. Coupling tree for Hl-................................................34
20. Coupling tree for Hl.................................................34
21. Coupling tree for H*.................................................34
22. Coupling tree for Hm.................................................34

1. Englishs Synthetic approach to compound 516..............................6
2. Avenozas synthetic route to (R,R) enantiomers of 3.......................8
3. Synthesis of Gamers Aldehyde from (L)-serine..............................9
4. Tanakas synthetic route to (S,S) enantiomers of 3.........................9
5. Rabats synthetic route to (R,R) enantiomers of 3.........................10
6. Current route to cis y-amino acid.........................................13
7. Synthesis of 2,4-dibromopentanedioic acid (2).............................14
8. HVZ synthetic pathway.....................................................14
9. Synthesis of 2,4-diamionpentanedioicacid (3)..............................22
10. Synthesis of 2,4-diaminopentanedioic acid dimethyl ester (4).............25
10. Synthesis of methyl 4-amino-5-oxopyrrolidine-2-carboxylate (5)...........29
11. Purposed alternative route to 5, fmoc protection. .......................36
12. Purposed alternative route to 5, Cbz protection..........................36
1. Comparison of synthetic routes to enantiomerically pure (R,R) or (S,S) 3..11
2. Comparison of cyclization techniques......................................31

1. Synthetic amino acids in peptidomimetics and peptide nanotubes
Synthetic amino acids are molecules which contain the amine and carboxylic acid
functional groups found in naturally occurring amino acids. They differ from natural
amino acids in terms of type of side chain and distance of the amino group from the
carboxyl group. Synthetic amino acids can be incorporated into a chain of naturally
occurring amino acids in a protein and this synthetic analogue may affect the function
of the protein by altering its structure. One such alteration is an induced P-tum.
Synthetic amino acids can also be used to form novel structures such as peptide
Paul et al. define a P-tum as a .. .stable secondary structural feature found in proteins
and peptides defined over a length of four amino-acid residues, encompassing a
tripeptide unit, which cause a reversal in the direction of the peptide chain. 1,2 P-
tums affect the structure of the peptides in which they are found; thus, they are central
to the biological activity of their peptides. Recent research has focused on
developing compounds that can induce a P-tum in a peptide. To this end, two main
classes of compounds have been developed, peptidomimetics and molecular
scaffolds. Peptidomimetics are similar to naturally occurring P-tum inducing
sequences of amino acids, while molecular scaffolds utilize rigid structures to mimic
the angles in a beta-turn.3 In 1991 Paul et al. modeled a pyro-glutamic (Glp) acid

residue and discovered that the dihedral angles of the constrained cis Glp residue
approximately match that of a type VI p-tum or its inverse.1 When further modeling
was performed, they discovered that they could extend a chain of amino acid residues
from both the carboxylate and amino groups on either side of the ring. The result was
an induced p-tum.
Synthetic amino acids also have applications in nanostructures such as peptide
nanotubes (PNTs). As early as 19744, it was postulated that biopolymer-like
structures, PNTs, could form from repeating peptide units. In 1993, Ghadiri et al.
produced crystallographic structures of PNTs, constructed from self-assembling
macrocycles composed of alternating d- and /-amino acids. The macrocycles stacked
vertically and were held together by hydrogen bonding.5 Since that time, numerous
variations of amino acids and peptides have been used to produce peptide nanotubes.
The first variations were based on pseudopeptide units or synthetic peptides. The
next generation of PNTs was based on P-amino acids.5 These amino acids have been
heavily studied because they allow more rotational freedom as well as greater
opportunities for producing derivatives than a-amino acids. However, because of the
high degree of rotational flexibility these monomers form floppy amino acid chains
which may not self assemble. Peptides made of rigid cyclic y-amino acids have a
greater propensity to self assemble in a regular manner because the high degree of

rotational freedom in the monomers is minimized. These structures allow greater
structural variability than a-amino acids, but in a more conformationally restricted
manner than linear P-amino acids.6,7,8 Hypercycles are formed by first linking y-
amino acids together with amide bonds, forming an amino acid chain; the ends of this
chain are then joined to produce the hypercycle.6 Two views of a potential 6 member
hypercycle are shown in Figures 1 and 2 (modeled by D. F. Dyckes) ,9
Figure 1. Wire structure. Figure 2. Space filling model.
When exposed to the correct conditions, hypercycles will self assemble into peptide
nanotubes.7 A diagram of this process is shown in Figure 3.


two self assembled
six membered hypercycle
Figure 3. Proposed route to a peptide nanotube.
Thus far, pre-determining the internal diameter of PNTs has been a difficult
process.7,9 However, with this system it may be possible to manipulate the diameter
and size of hypercycles by controlling the number of monomeric units in the peptide
chain before the formation of the hypercycle. This could allow a greater degree of
selectivity for specific guest molecules.
1.1 Research Goal: Cis methyl 4-amino-5-oxopyrrolidine-2-carboxylate
The immediate goal of this research has been the synthesis of the y-amino acid, cis
methyl 4-amino-5-oxopyrrolidine-2-carboxylate. A five membered lactam rigidities
the molecules is present within the ring, and carboxylic acid and amino groups are
attached to the ring.

Figure 4. Cis methyl 4-amino-5-oxopyrrolidine-2-carboxylate.
Because of its structural similarity to Glp, methyl-4-amino-5-oxopyrrolidine-2-
carboxylate may function as a molecular scaffold P-tum inducer. Like Glp, the target
also contains a five membered ring, which locks the amino acid structure into a
single conformation. Like Glp, the target molecule also contains cis functional
groups, located in similar positions on the ring. The similarity between the two
molecules suggests that the computational model of Glp may also approximate the
desired compound. If this is true, then the dihedral angles of the molecules will be
similar and the target compound will also induce P-tums. Furthermore, it is likely
that a chain of amino acid residues could also be extended from both sides of the
target compounds ring.
Previous research has shown that hypercycles made from similar amino acids adopt a
quasi-planar conformation and are able to self-assemble into peptide nanotubes.10
These data give good reason to believe that the target compound will also form

nanotubes. In addition, initial computer models show that such a hypercycle would
have a lipophilic exterior and a hydrophilic interior, shown in Figures 1 and 2.
Therefore, it is believed that a PNT formed from such a structure may have
applications as a synthetic ionophore, similar to naturally occurring zervamicin and
antimoebin11, which transport K+. It has been shown that similar types of PNTs can
be successfully incorporated into a lipid bilayer. After insertion into the lipid bilayer,
the PNT can function as a transmembrane channel for ion transport.6
1.1.2 Previous Synthetic Strategies
The synthesis of the target molecule 5 expanded on research performed by previous
members of Dr. Dyckes group,1415 most recently Michelle English.16 The overall
reaction scheme employed by English is shown in Scheme 1. This work successfully
produced compounds 1,2, and 3 in good yield as well as compound 4 in uncalculated
Scheme 1. Englishs Synthetic approach to compound 5.16

In order to produce the cis cyclic amino acid, the precursor linear compound
possesses either (2S,4S) or (2R,4R) stereochemistry. The linear meso molecule,
which is (R,S), will produce the trans cyclic amino acid. The stereocenters are set
during the first reaction, which produces both the meso and chiral forms of 1.
However, these compounds were not isolated; rather the mixture was washed with
water and extracted into benzene. This process replaced the acyl bromine with the
hydroxyl groups, producing compound 2 (both meso and chiral) with a 98% yield.
English discovered that at this point it was possible to selectively remove the meso
isomer by washing with chloroform and acetone. A mixture of the chiral
enantiomers, (R.R) and (S,S) 2,4-dibromopentanedioyl acid, were obtained with only
10% impurity from the meso isomer using this method. Once English had the chiral
isomers in hand, a Sn2 reaction was employed to replace the bromine with amine
groups. This reaction produced compound 3 in 44% yield. Finally, a Fisher
esterification was used to produce compound 4 by refluxing compound 3 with
methanol and HC1 overnight. However, compound 4 was collected as a wet, chloride
salt from which yield could not be calculated. Compound 5 was not produced.
The pathway to 3 described above is of interest as previous research teams have also
attempted to synthesize pure (R,R) or (S,S) enantiomers of 3. However, these

pathways required uncommon or expensive chiral starting materials and proceed
under harsh conditions.
Scheme 2. Avenozas synthetic route to the (R,R) enantiomers of 317.
In 1997, Avenoza et al. produced the (R,R) enantiomer of 3 via the pathway shown in
Scheme 2. In this synthesis, Gamers aldehyde provided an initial stereocenter. The
site for the second stereocenter was added by reacting Gamers aldehyde with methyl
2-benzamido-2-(diethoxyphosphoryl)-acetate and potassium tert-butoxide under
Wadsworth-Homer-Emmons olefination conditions. This was intended to produced
only (E)-{R)-3-ter/-butoxycarbonyl-2, 2-dimethyl-4-[2-(benzamido)-2-
(methyoxycarbonyl)ethenyl]-l,3-oxazolidine. However, during this reaction it was
observed that approximately 18% of the (R)-(E)-olefin racemized into the (S)-(E)-
olefin, a process not uncommon with Gamers aldehyde under these conditions. The
key step to producing the second stereocenter was the asymmetric hydrogenation of

the (E)-(R)-olefin under heterogeneous conditions. This was followed by oxidation to
open the ring and add the second methyl ester. Finally, the amine protecting groups
1 -7
were removed by hydrolysis.
MejqOMe)! Q'
CO, Me
1 ft
Scheme 3. Synthesis of Gamers Aldehyde from (/)-serine .
The overall yield for this reaction was 16% from Gamers aldehyde. However this
aldehyde is not a common, inexpensive material; rather, it is synthesized via the route
shown in Scheme 3. The yield for this synthesis is roughly 50%. Therefore, the
overall yield for Avenozas pathway is only 8%.
HO. Ob, ", 1. Boc^Q E%N / ^ 2. O4-hrtyt4iNsStMfvq7ylB0uei / \ hhNkDMF N \ I. H2, Pd-C ' j RuQ2.*H2QNaJO,
/ \ 3. NfaQ, pyridine H QCr Doc 1 \ Z BocjQ AcOEt ^N'^N00,'Bu Boc ^N/^VOp(lBu Boc
78% reaction yield 86% reaction yield 90% reaction yield / 88% reaction yield
82% reaction yidd
2. Otbitfyl4^N -TV.
60% reaction yield {>95% ce)
Scheme 4. Tanakas synthetic route to the (S,S) enantiomer of 3.

The next year, Tanaka et al. reported another enantioselective procedure to create the
(S,S) enantiomers of 3. The starting material, trans-4-hydroxy-Z,-proline, provided
the first stereocenter. After protection of the carboxylic acid function and the amide,
the alcohol is mesylated to form a good leaving group. The stereochemistry of this
center was inverted by a Sn2 reaction with NaN3 to from (R) to (S). A combined
application of ruthenium tetroxide (RUO4) oxidation and regioselective ring opening
provided the desired stereochemistry.19 Finally, the protecting groups were cleaved
with acid. The yield for this reaction, 33%, was significantly better than that of
Scheme 5. Rabats synthetic route to the (R,R) enantiomers of 3.20
Finally, Rabat et al. reported an adaptation of a peptidomimetic route in 2001.20 A
three step synthesis was used to convert (2S,3R)-(+)-6-oxo-2,3-diphenyl-4-
morpholinecarboxylate into (R,R) 3. While the starting material, which itself is
94% reactiw yield

several synthetic steps removed from glycine, contained two stereocenters, neither of
these appear in the final product. Instead, they were used to set the first permanent
stereocenter during an alkylation reaction with PhSeBr. The large phenyl groups on
the starting material force the new group to add anti. This was followed by a second
radical reaction between the phenyl selenide and methyl 2-acetamidoacrylate, adding
half of the carbon backbone for the final product. Although this step of the synthesis
proceeded with a 74% yield, approximately 40% of the product was the undesired
(R,S) isomer. Finally, the morpholinecarboxylate was removed by hydrogenation and
the protecting groups by acid hydrolysis. The yield for this reaction was 31% from
Group First stereocenter Second stereocenter % yield (R,R) (S,S)
Avenoza17 Gamers aldehyde asymmetric hydrogenation 7-9%
Rabat19 alkylate (2S,3R)-(+)-6- oxo-2,3-diphenyl-4- morphol inecarboxylate with PhSeBr radical reaction, hydrogenation, acid hydrolysis of protecting group 33%
Tanaka20 trans-4-hydroxy-Z,-proline Ru04 oxidation, regioselective ring opening 31%
Dyckes16 Bromination of glutaric acid stereocenters already set 22%
Table 1. Comparison of synthetic routes to enantiomerically pure (R,R) or (S,S) 3.

Although it initially appears as though Dyckes group, with a percent yield of 44%,
developed the most efficient synthesis, this is not the case since the product of this
pathway is a racemic mixture of both enantiomers, which must still be resolved.
Therefore, under the best of conditions, the percent yield of each enantiomer (R,R) or
(S,S) could only be 22%. Both Tanaka and Kabat developed routes with higher
yields, 31% and 33% respectively. Although Dyckes route may have a lower overall
yield than that of either Tanaka or Kabat, it is not without merit since the route uses
less expensive starting material, proceeds under more gentle conditions, and produces
all of the potential isomers. Therefore, while the route may not be as efficient as these
two methods, it may prove to be more useful because of its economy.
2. Results and Discussion
The stereochemistry of the final amino acids must be considered throughout the
synthesis because the conformation of each isomer will affect its behavior as a
structural element. In this paper the (R,R) amino acid is often the only isomer shown,
particularly in 'HNMR spectra. However, it represents both cis isomers; resolved
enantiomers will be denoted.

Figure 5. Cis amino acid (R,R) or (S,S). Figure 6. Trans amino acid (R,S) or (S,R).
The trans isomers may form stair step types of structures, whereas the syn analogs
may act as turning elements. The cis analogs are of more interest to this research for
reasons previously discussed. Therefore, the synthesis focused on selectively
producing these isomers. The overall reaction scheme employed is shown below.
The synthesis follows that of English exactly through the synthesis of compound 3.
After this point several new approaches were explored.

2.1 Results for the synthesis of 2,4-dibromopentanedioyl dibromide (1) and 2,4-
dibromopentanedioic acid (2)
2 (R.R) and (S.S)
Scheme 7. Synthesis of 2,4-dibromopentanedioic acid (2).
In the first step of the reaction, glutaric acid was brominated using a Hell-Volhard-
Zelinski reaction, which produced compound 1. The procedure was modeled on the
procedure explained by a reaction mechanism described by Allen and Kalm in
192i2| xhe mechanism is shown in Scheme 8.
+ Br'
Scheme 8. HVZ synthetic pathway.
It is believed that the first step of the reaction is the replacement of the hydroxyl
groups by bromine, via a reaction with phosphorus tribromide. The resulting acyl
bromide is able to exist in the enol form, which is rapidly brominated at the a-carbons
in the presence of Br2 via an alkene addition. Bromination is accomplished by adding
liquid bromine to a mixture of glutaric acid and PBrj. The reaction proceeds in a

similar manner when PCI3 is substituted for PBr3. As the bromine is slowly added, a
bright yellow color is observed. This color corresponds to the formation of
phosphorous tribromide. A molar excess of Br2 is used to facilitate the bromination
of both a-carbon positions. Once bromine is present in excess, the solution retains
the cherry red color of bromine. This reaction produces compound 1, which is not
isolated. Because the a-carbons of the enol are sp2 hybridized, the first bromine is
able to add to either side of the double bond. The second step of the sequence
hydrolyzes the acyl bromides (or chlorides), reforming the dicarboxylic acid,
compound 2. This addition/elimination reaction is accomplished by vigorously
shaking compound 1, deionized water, ice and benzene in a separatory funnel, and
recovering the brominated compound from the organic layer. The organic solution is
dried, concentrated on a rotary evaporator and allowed to stand in a hood overnight at
room temperature. After 12-24 hours, the orange liquid become a dark brown or
black color and a white solid is also observed. Washing the mixture in a Hirsch
funnel with small aliquots of a chloroform/acetone solution removes not only the
black tinted benzene but also the meso compound. The chiral compound is collected
as a white solid. Figure 8 shows a *H NMR of 2 before washing. Because the Hell-
Volhard-Zelinski reaction involves the addition of Br to a sp2 hybridized carbon it is
not enantioselective. However, preliminary evidence suggests that the reaction is
somewhat stereoselective, producing 60% chiral products and 40% meso. Yield for

the crude product, which includes both the chiral and meso products, calculated for
both the PBr3 and PCI3 bromination were the same at approximately 94%. Yield for
the purified chiral enantiomers was also comparable.
Figure 7. Meso and chiral isomers of 2.
It was thought that this stereoselectivity could be enhanced by using a bulkier group,
such as iodine, as the nucleophile. Although the first iodine will still add equally to
either face of the sp2 hybridized enolic carbon, the stereochemistry of the second
chiral center may favor addition anti to the first iodine because of steric hindrance.
To this end, two synthetic routes were attempted. The first attempt was the
replacement of Br2 with I2 in a standard Hell-Volhard-Zelinski reaction. The second
procedure was a copper catalyzed reaction modeled from the work of Hirouchi et al.22
Thus far, neither route has produced the desired molecule.

AMn, i* M l < m
Figure 8. The 'H NMR spectrum of product 2, before washing.
Although the molecules are stereoisomers, the coupling patterns of the protons are
very different. The chiral molecule contains two different groups of protons. The
protons attached to the P-carbon (Ha and Ha) are equivalent. These protons appear at
a chemical shift of 2.72 ppm. However, in the NMR spectrum they do not appear
as a simple triplet, but rather as a doublet of doublets. This is because Ha interacts
differently with Hb than Hb. Although the coupling constant of Ha> and Hy is the

same as that between Ha and Hb, the coupling between Ha- and Hb is not identical.
Similarly, the coupling between Ha and Hb- is not the same as that of Ha- and Hb-, but
is identical to that of Ha and Hb. The coupling constants for the doublet of doublets
are 7.6 and 7.8 Hz. An identical pattern is seen at 4.59 ppm. It is due to Hb and Hb-.
Figure 9. Coupling tree for Ha and Ha-
The meso isomer produces a more complicated 'H NMR spectrum than the chiral
isomer pair because the central protons are not equivalent. The protons attached to
the a-carbons, Ha and Ha-, are both split equally by the protons on the central carbon,
He and HC. The result of this is a triplet at 4.56 ppm. The coupling constant is 7.3
Hz. The central pair of hydrogen atoms displays a more complex pattern. The proton
He- is split equivalently by both Ha and Ha-. The resulting triplet is further split by the
geminal neighbor, He. This changes the triplet to a doublet of triplets. However, the
peaks do not appear as a doublet of triplets because the center peaks in each triplet
overlap. As a result, only five peaks are seen in the set. The area of the central peak
is equal to the sum of the extreme end peaks in the set. The proton Hc shows a similar
I 7.8 Hz
J 7.6 Hz

pattern, producing a set of peaks which are the mirror image HC. The chemical shifts
of these protons cannot be determined. They appear as two distinct sets of peaks at
2.59 and 2.93 ppm.
Figure 10. Coupling tree for Hc and He-. Figure 11. Coupling tree for Hd and Ha.
Although protons Hc and HC are different, they produce the same coupling constant
with the protons Ha and Ha. This is an accidental coupling. The coupling constant for
each is Jca~ JCd=7.3 Hz. The geminal coupling constant for Hc and He- is 14.4 Hz.

Hamul IIa
Hb and Hb
' I
. A
4.5 4.4 4.2 4.1 3.8 3.6 3.4 3.2 3.1 2-8 ppt
5 <1
Figure 12. The 'H NMR spectrum of product 2, the chiral isomer.
Using the washing technique described, the chiral isomers of 2 were obtained with
only minor contamination from the meso compound. Figure 12 shows the *H NMR
spectrum after washing. The meso isomer is visibly absent from the regions of 2.59,
2.93 and 4.56 ppm. Only the doublets of doublets corresponding to the chiral isomers
are seen at 2.74 and 4.6 ppm.

Wavenumbers (cm-1)
Figure 13. The IR spectrum of 2.
The IR spectrum of compound 2 contains a large C-0 absorption band at 1707.6 cm'1.
In a carboxylic acid containing an a-halogen, this band characteristically appears
between 1720 and 1690 cm"1. In compound 2, the peak appears in the middle of this
range. The large, broad peak between 3200 and 2600 cm'1 is due to O-H stretching of
the carboxylic acid. The aliphatic C-H stretches are seen at 2978.4 and 2896.7 cm'1,
superimposed on the O-H band. In the fingerprint region there is a peak at 878.1 cm'1
that is most likely due to the carbon-bromine bond.

2.2 Results for the synthesis of 2,4-diaminopentanedioic acid (3)
2 (R,R) and (S,S) 3 (S,S) and R,R)
Scheme 8. Synthesis of 2,4-diamionpentanedioic acid, 3.
Compound 3 was prepared by animating the a-carbons of the chiral enantiomers of
compound 2. This was accomplished by stirring compound 2 in a flask with
ammonium hydroxide for five days at room temperature. The solution was
concentrated by rotary evaporation, methanol was added, and then it was placed in
the refrigerator for two days. On the second day a white precipitate, compound 3,
was collected. No further purification was necessary.16 The yield for this reaction is
77%. The overall yield for the synthesis up to this point is 44% for the chiral

S*xnVrtjrVs 2 5
Figure 14. The *H NMR spectrum of Product 3.
The H NMR spectrum of compound 3 is shown in Figure 14. The upfield and
downfield doublet of doublets now appear at 1.92 and 4.05 ppm, respectively. The
shift of these protons to a more shielded region is due not only to the replacement of
the a-bromine groups with positively charged amine groups, but also the negatively

charged carboxylate groups.
4000 3500 3000 2500 2000 1 500 1000 500
Wawtumbms (cm-1)
Figure 15. The IR spectrum of 3.
Examination of the O-H and carbonyl stretches in the IR of compound 3 suggest that
the compound is the bis-zwitterion. The carbonyl stretch appears at 1646.3 cm'1
instead of 1700 cm'1, which is characteristic of a carboxylate.23 In addition the large
O-H stretch is absent in this spectrum. At 1086.5 cm'1 there is a new C-N vibrational
band. There is also a new N-H stretching band between 3400 and 3000 cm'1.

2.3 Results for the synthesis of 2,4-diaminopentanedioic acid dimethyl ester (4)
J (S.S) and R,R) 4 SO/
Scheme 9. Synthesis of 2,4-diaminopentanedioic acid dimethyl ester (4)
Compound 4 was prepared via a Fisher esterification of 3 in methanol. It was
isolated as either the sulfate or chloride salt, depending on which acid (sulfuric or
hydrochloric) was used during the esterification. The free base of compound 4 was
not isolated because the positively charged amine groups readily associated with
negatively charged counter ions, forming a salt. The counter ions were difficult to
remove and affected the properties of the molecule. The salts were hygroscopic,
difficult to characterize, and presumably less reactive than the free base. In order to
collect the molecule as a free base, it was thought that the counter ions could be
removed by and acid base reaction and precipitation with either Ba(OH)2 or NH3 in
an organic solvent to form BaSC>4 or NH4CI. However, it was discovered that the free
base cyclized spontaneously under these conditions, which was not an unwelcome
discovery. In addition to cyclization, a small to substantial amount of hydrolysis of
the methyl esters was also always observed when water was present in the reaction,

even when in minimal amounts. Inevitably, a mixture of compounds 3,4 and 5 were
collected in varying amounts under these conditions. Although compound 3 can be
readily removed by column chromatography, compounds 4 and 5 show similar Rf
values and are difficult to separate by either thin layer chromatography or column
Esterification with hydrochloric acid and subsequent precipitation of the chloride ions
under anhydrous conditions was viewed as the more promising of the two routes.
This is because water contamination was always introduced during precipitation with
aqueous barium hydroxide, increasing methyl ester hydrolysis and lowering yields. .
Additionally, it is thought that the sulfate counter ion acts as a bidentate ligand,
bridging both of the positively charged amines. It is therefore more difficult to
remove a single sulfate counter ion than two chloride ions. Therefore, esterification
with sulfuric acid and subsequent precipitation with aqueous barium hydroxide was
not considered an optimal route for the production either of the free base of 4 or of
compound 5 and was not extensively studied.

Figure 16. The fH NMR spectrum of compound 4 as a sulfate salt.
The 'll NMR spectrum for compound 4 is shown in Figure 16. The familiar key
signals, the doublets of doublets, are still present but have shifted slightly downfield.
The more downfield set of peaks now appears at 4.4 ppm, while the upfield peaks
appear at 2.0 ppm. Additionally, a new singlet peak has appeared at 3.75 ppm. When
area under this signal was measured, it was found to represent six protons due to the
six ester methyl groups.

4000 3500 3000 2500 2000 1500 1000 50C
Wavanumbere (em-1)
Figure 17. The IR spectrum of compound 4 as a sulfate salt.
The IR spectrum of 4 is very similar to that of 3, as would be expected. The most
notable difference is the presence of an ester bond. This new bond shifts the carbonyl
stretch from its position in the zwitterions, 1646.3 cm'1, to 1719.8 cm'1. Because the
compound is a salt and its amines are positively charged, the region between 3500
and 2700 cm'1 is noticeably more complicated than that of a free amine.

2.4 Results for the synthesis of methyl 4-amino-5-oxopyrrolidine-2-carboxylate
2.4.1 Results for the Synthesis of 5 by deprotonation and salt precipitation
Scheme 10. Synthesis of methyl 4-amino-5-oxopyrrolidine-2-carboxylate (5).
Table 2 summarizes the synthetic strategies for producing compound 5. Initial
attempts to cyclize compound 4 following Englishs route did not produce any
products. Following this method, only starting material, compound 4, was recovered
under the best of conditions. In the worst case scenario this had hydrolyzed back to
compound 3. These reactions involved stirring compound 4 with a hindered base,
such as with triethylamine andl,8-diazabicyclo[5.4.0]undec-7-ene (DBU). It was
thought that cyclization could be accomplished by using a hindered base to
deprotonate one of amine groups, thus promoting a nucleopilic attack on a carbonyl
carbon. Counter ion removal was not considered, and these initial efforts were
unsuccessful. It is now known that this was because the amine groups remained
positively charged and could not act as nucleophiles.

Compound 5 was eventually produced in its most pure form by esterifying 4 under
anhydrous conditions to produce the chloride salt, as previously discussed. The
intermediate chloride salt was not isolated or characterized. The solvent, acidic
methanol, was then removed using a rotary evaporation and replaced with ammonia
saturated methanol. The chloride salt was stirred in the basic methanol, producing
NH4CI. Diethyl ether was added to ensure precipitation and to remove compound 5
from the salt. The yield for this reaction was only 1-2%. Initial attempts to
precipitate NH4C1 and extract compound 5 in one step using basic chloroform in
place of methanol have been nominally successful, with an overall yield of 22%.
However, spectral evidence suggests that approximately 60% of the product was still
the chloride salt of compound 4 and only 40% of the product was compound 5.
Replacing chloroform with basic a diethyl ether/ chloroform solution returned a
comparable overall yield of 17-23%. However, analysis by ]H NMR indicated that
the product of this reaction was primarily (90%) compound 4. Attempts to use only
basic ethyl ether resulted in no reaction.

Route esterification acid (in CH3OH) base precipitate solvent overall yield yield of 5
1 HCl(aq) none none CH2Cl2 7% 70%
2 H2S04,aq, TEA none ch2ci2 48% 10%
3 HaSO^, Ba(OH)2faal BaS04 ch3oh 70% 10%
4 HC1(k) NH3(k) HN4CI CH30H 1-2% 100%
5 HCloo NH300 NH4CI CH2C12 55% 40%
6 HN4CI CHCl3/ethyl ether 22% 10%
Table 2. Comparison of cyclization techniques.
Compound 5 was produced using several techniques, the most important of which are
shown in Table 2. Route 1 was applied before it was understood that compound 4
contained protonated amines and counter ions. In an attempt to discover a suitable
TLC solvent, it was observed that some of the water could be removed from the salt
by repeated washing with methylene chloride. While the yield of compound 5 was
high for this reaction, the overall yield was not. Route 1 was then modified by
refluxing the solution, which produced the epimer of compound 5 as a side product.
The second route was an attempt to perform an acid base reaction with triethylamine
(TEA) to deprotonate the amines. However, it did not account for the strong
attraction between the counter ion and amines. Subsequent attempts with 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU) produced similar results, which are not
shown. The third route was performed in an attempt to simultaneously deprotonate

the amines by raising the pH and remove the sulfate counter ion through precipitation.
However, the presence of water produced a mixture of compounds 3,4 and 5. The
final three techniques utilized an identical esterification procedure which minimized
water contamination by using dry methanol and anhydrous HC1 gas. The second step
of these reactions varied in extraction solvent, however all of the reactions used
ammonia saturated solvent to remove the chloride counter ions as ammonium
chloride while raising the pH to deprotonate the amines. In these reactions, the ester
itself was never characterized. These routes were modifications of the work of Piper
et al.24 The third route most closely follows Pipers work, using dry methanol as the
neutralization solvent and diethyl ether to extract the cyclized product. This is the
only route thus far which has produced compound 5 exclusively, although in nominal
yield. The fourth route used basic methylene chloride instead of methanol because it
was thought that the methanol may reopen the ring to reform compound 4. While this
modification enhanced overall yield, the methylene chloride may have been wet and
did not exclude the chloride salt. The fifth technique gives the highest yield;
however, the chloride salt of 4 is still present as 60% of the product. The *H NMR
spectrum shown below was produced using the fourth technique and contains only
compound 5.


jl "V'i' m*iy4^VjI,^,S4M I

............ 1 1 1 I rn H 'P
3.8 3.6 3.6 3.2 3.1 2.6
2.4 pptt
Figure 18. The !H NMR spectrum of Product 5.
The two methylene protons, Hl and Hf, appear as a doublet of doubled doublets
(ddd) and as a doublet of triplets (dt) at 2.6 ppm and 2.4 ppm, respectively. These
protons are no longer equivalent in the cyclic structure because they are on different
faces of the ring. The proton on the P-face of the ring, Hl appears at a more
downfield position because of its proximity to the amine and carboxylic ester. The
proton on the opposite face of the ring, Hl, is upheld because it is more distant to the
electron withdrawing groups than HL .

Figure 19. Coupling tree for Hl Figure 20. Coupling tree for HL-
Hl appears as a doublet of doubled doublets because its signal is split by both
germinal and viscinal neighbors. The geminal coupling constant from Hl is 13.0 Hz.
Each branch of this doublet of doublets is then further split by Hm and Hk, which do
not split the signal equally. The coupling constant for Jun is 8.0 Hz, while the
coupling constant for JLk is 6.8 Hz. Figure 19 shows this coupling tree. HL' appears
as a doublet of triplets. Unlike Hl, Hl- is split equally by both Hk and Hm, producing
a triplet. The coupling constant for JL'm=JLk= 9.6 Hz. The geminal splitting from Hl
(13.0 Hz) then splits the signal again, generating a doublet of triplets (Figure 20).
Figure 21. Coupling tree for Hk Figure 22. Coupling tree for Hm

Both Hk and Hm appear as doublets of doublets. The coupling constants for JLm and
JLm are not the same. Likewise, the coupling values for JL k are not equal to those of
Ju<- Hm appears farther downfield than Hk because it is adjacent to an acylated amine
2.4.2 Results for the Synthesis of 5 by amine protection
After the salt precipitation data were analyzed, it became apparent that another
approach needed to be investigated. The alternative pathways shown in Schemes 11
and 12 were developed in an attempt to avoid the salt problem entirely by protecting
the amines of the bis-zwitterion. It was thought that this could be accomplished
through the use of an Fmoc group (Scheme 11) which could withstand the acidic
conditions of esterification. Because this group is removed in mildly basic
conditions, perhaps its removal would promote cyclization. Alternatively, it was
thought that the free base of 4 could be obtained by protecting with a
benzylchloroformate (Z) group, esterifying and removing the protecting groups to
yield the free base (Scheme 12). Once the free base of 4 had been characterized, it
could be cyclized using the original reaction with TEA in an organic solvent. Both
schemes were attempted using compound 3 as a starting material. Protection with
Fmoc was attempted using the procedure outlined by Bodanszky,25 while Z group
protection was performed according to Meyers procedure.26 Both reactions are run
under aqueous conditions with either an organic cosolvent or vigorous stirring.

Scheme 11. Proposed alternative route to 5, Fmoc protection.
compound S
Scheme 12. Purposed alternative route to 5, Cbz protection.
Although these reactions have high yields in the literature, often greater than 90%,
this was not observed under these conditions. While the Fmoc protection was
moderately successful, Z group protection returned only starting material. Fmoc
protected 3 was obtained in 25% yield by *H NMR peak integration. The compound
was never isolated, but appeared to be only monoprotected even though an excess of
9-fluorenylmethyl chlorocarbonate (FmocCl) was used. Due to time constraints, this
mixture was not purified or characterized, but rather an esterification in methanol and
12M HC1 was attempted. Because of the significant contamination from the FmocCl,
it was difficult to determine whether the reaction was successful.

3 Conclusion and Future Studies
The immediate goals of the research have been met. Compound 3 was produced
using milder conditions than previous methods. Although the route is less efficient, it
is more economical and produces not only the cis isomers, but also the trans. In
addition to this, the reactivity of compound 4 has been elucidated. Finally, compound
5 was produced using several different methods. The next step of the research will
focus on increasing the percent yield of 5 by developing alternative procedures for
protecting the amines in a non aqueous system.
Once compound 5 is in hand, different esterases will be explored in order to find one
that is able to selectively hydrolyze the methyl groups from one of the enantiomers.
Potential mammalian esterases include horse and pig liver esterase, as well as porcine
pancreatic lipase. Potential bacterial targets include an esterase from Pseudomonas
putida MR-2068. If a suitable esterase cannot be found, the enantiomers may be
resolved at an earlier step as a tartaric acid salt. This resolution could be
accomplished with either compound 2 or compound 4. Once the enantiomers are
separated, the synthesis of linear peptides and hypercycles can be investigated.

4 Experimental
4.1 Reagents
All of the reagent grade solvents were purchased from Fisher (bromine, ammonium
hydroxide, methanol and acetone), Acros (phosphorus tribromide 99%, glutaric acid
99%), Baker (chloroform), Mallinckrodt (methylene chloride) or Alfa Aesar (benzene
99%). Deuterated solvents were purchased from Acros. All reagents were used
without further purification unless otherwise stated.
4.2 General Methods
4.2.1 Nuclear Magnetic Resonance Spectroscopy
Proton *H NMR spectra were obtained on a Varian 200 MHz spectrometer. Internal
references included TMS (when the samples were run in methanol, acetone or
chloroform). When samples were run in D2O no internal reference was used; instead
the water peak was set at 4.7 ppm. The 8 values are given in ppm; calculated J values
are given in Hz.
4.2.2. Infrared Spectroscopy
IR spectra were recorded neat on an Avatar 360 FT-LR ESP (manufactured by
Thermo Nicolet). Absorbances are given in cm*1.

4.2.3 Thin Layer Chromatography
Samples were spotted from an open-ended Kimax-51 capillary tube onto Watman
aluminum backed silica plates. Spots were visualized by UV light followed by
dipping the plate into KMn04 and heating in an oven at 100C for 5-10 min.
4.3 Synthetic Procedures
4.3.1 Procedure for drying bromine
Bromine (50 mL) was dried by adding phosphorus pentoxide (1-2 g) in a sealed bottle
and shaking for several minutes. The mixture was then filtered in a glass funnel.16
4.3.2 Procedure for drying methanol
Methanol (70 mL) was distilled under nitrogen from a 100 mL round bottom flask
containing approximately one gram of Mg metal and an L crystal.
4.3.3 Procedure for saturating methanol with HCl(g)
HCl(g) was bubbled through dry methanol (20 mL) on an ice bath for approximately
30 min.25

4.3.4 Procedure for saturating methanol with NH3(g)
NH3(g)was bubbled through dry methanol (10 mL) on an ice bath for approximately
30 min.25
4.3.5 Procedure for drying methylene chloride
Methylene chloride was dried by adding several grams of anhydrous MgSC>4 to
1 OOmL of methylene chloride. The mixture was then stirred in a stoppered bottle for
several hours and filtered.
4.3.6 Procedure for drying triethylamine
Triethylamine (TEA) was dried by adding -100 mL of triethylamine to a 250 mL
round bottom flask, equipped with a stir bar and 102 g of ninhydrin. The TEA was
distilled from the ninhydrin under an inert atmosphere (nitrogen gas) and stored over
4 A molecular sieves.27
4.3.7 Synthesis of 2,4-dibromopentanedioyl dibromide (1) with PBr3
Approximately 6.11 g of glutaric acid (46 mmol) was added to a 100 mL three neck
flask containing a stir bar and 10 mL phosphorus tribromide (110 mmol). The center
neck of the flask contained a condensing column topped with a drying tube full of
CaCL. One side neck contained a thermometer (not calibrated), the other held an

addition funnel. Approximately 11 mL of dried bromine was added from the addition
funnel in a drop wise manner over a 30 min period, until the solution remained cherry
red. The solution was then refluxed for 13 hours, during which time the temperature
was maintained between 88 C and 95 C. After the 13 hours, an additional 1.5 mL
of dried bromine was added drop wise via the addition funnel over 20 min. The
solution was then heated at 88 C for an additional 4.5 hours, and then stirred at room
temperature for an additional 1.5 hours. The product was not isolated or
characterized at this point. Instead, it was immediately reacted according to
4.3.8 Synthesis of 2,4-dibromopentanedioyl dibromide (1) with PCI3
A stir bar, 6.135 g (46 mmole) of glutaric acid and 15 mL of PCI3 were added to a
100 mL three-neck flask. The center neck of the flask contained a condenser topped
with a drying tube full of CaCh. One side arm contained a thermometer (not
calibrated); the other contained a glass addition funnel. Dried Br2 (10 mL) was added
to the flask via the addition funnel over a period of 40 min. After all of the bromine
had been added and the temperature returned to room temperature, the mixture was
heated until it reached 87C. The reaction was heated at this temperature for 21
hours. The product was not isolated or characterized at this point. Instead, it was
immediately reacted according to procedure 4.3.11.

4.3.9 Attempted synthesis of 2,4-diiodopentanedioyI diiodide with PBr3
Approximately 10 mL anhydrous tetrahydrofuran (THF) was added to a dry 25 mL
round bottom flask containing 0.515 g (3.9 mmol) glutaric acid and 4.0 g (15.6 mmol)
iodine. After dissolution of most of the solids, 1.5 mL (7.8mmol) of anhydrous PBr3
was added. The solution was refluxed with a drying tube full of CaCL for 7 days at
approximately 60C. After this time, the solution was concentrated on a rotary
evaporator and a 'H NMR taken. Yield not calculated. H-NMR (CDC13): 8(ppm)=
1.65 (s, 4H)*; 1.9 (m, 6H)*/**; 2.39 (t, 4H)**; 3.47 (s, 4H)*;
*THF, **glutaric acid
4.3.10 Attempted synthesis of 2,4-diiodopentanedioic acid with copper chloride
Approximately 0.54g (4.1 mmol) of glutaric acid, 2.05 g (8.18mmol) iodine, 0.678 g
(4.1 mmol) cuprous chloride dihydrate, and 0.405 g (4.1 mmol) cupric chloride were
dissolved in 25 mL of anhydrous tetrahydrofuran and refluxed for five days under
nitrogen. Following this, the suspension was vacuum filtered in a Hirsch funnel and
the filtrate reduced to a solid on a rotary evaporator. Only starting materials were
recovered.22 'H-NMR (CDCI3): 8(ppm)= 1.8 (s, 6H)*/**; 2.4 (t, 4H)**; 3.9 (s, 4H)*.
*THF, **glutaric acid

4.3.11 Synthesis of 2,4-dibromopentanedioic acid (2)
Compound 1, produced by reacting PBr3 or PC13, with glutaric acid and Br2 according
to procedures 4.3.7 and 4.3.8, was added to a separatory funnel containing
approximately 200 g of crushed ice, 150 mL of ice cold DI and 150 mL of benzene.
The resulting mixture was shaken vigorously for several minutes. The organic and
aqueous layers were then separated. The organic layer was washed twice with 100
mL of cold water and crushed ice. Aqueous washes were combined and washed twice
with 50 mL aliquots of benzene. The organic washes were combined and dried by
stirring with several grams of MgSCL then concentrated on a rotary evaporator and
left overnight. After 24 hours of standing in the hood, a brown/black solid was
observed. The solid was washed three times with 10 mL aliquots of chloroform,
followed by three 10 mL aliquots of a 1/20 (v/v) solution of acetone and chloroform.
A cream colored white solid was collected. The crude yield was 93%; the percent
yield for the 90% pure chiral isomer was 75%.16 'H NMRmeso and chiral 2 (d-acetone): 8=
2.581 (dt, 1H); 2.72 (dd, 2H); 2.93 (dt, 1H); 4.50 (t, 2H); 4.95 (dd, 2H). *H NMRcWran
(d-acetone): 5= 2.72 (dd, 2H); 4.95 (dd, 2H).
4.3.12 Synthesis of 2,4-diaminopentanedioic acid (3)
Approximately 100 mL of concentrated ammonium hydroxide was added to a 125
mL vacuum flask containing 3.299 g of 2 and a stir bar. A clamped tube sealed the

vacuum arm of the flask, while the top of the flask was stoppered and wired shut.
The reaction mixture was stirred at room temperature for 7 days then transferred to a
250 mL round bottom flask and its volume reduced to approximately 30 mL of liquid.
Following this, 90 mL of methanol was added to the concentrated liquid, and the
mixture was stored at 5 C two days. After two days, a white precipitate was
collected by vacuum filtration and washed with small aliquots of ice cold methanol.16
Yield 77%. 'H-NMR (D20): 8(ppm)= 1.90(dd, 2H); 4.75(dd, 2H).
4.3.13 Synthesis of 2,4-diaminopentanedioic acid dimethyl ester sulfate salt (4)
Concentrated sulfuric acid was added drop wise to a 25 mL round bottom flask
containing methanol (10 mL) and 0.436 g (2.7 mmole) of 3 until the pH of was 1.0.
After five days stirring at room temperature, the solution was vacuum filtered in a
Hirsch funnel. The filtrate was concentrated on a rotary evaporator to a saturated
solution of aqueous barium hydroxide was added drop wise until the pH was 8. The
solvent was then removed with a rotary evaporator, producing a white oil. The oil
was washed three times with 8 mL aliquots of methylene chloride. The solvent was
then removed by rotary evaporation. Yield 72%.
'H NMR (D20): 8(ppm)= 2.05(dd, 2H); 3.65 (s, 6H); 4.35( dd, 2H).

4.3.14 Attempted preparation of methyl 4-amino-5-oxopyrrolidine-2-
carboxylate (5) from sulfate salt of 4 by precipitation of BaSC>4
Five drops of concentrated sulfuric acid were added to a 100 mL round bottom flask
containing 0.271 g (1.7 mmol) compound 3 and 30 mL methanol. The solution was
refluxed for 8 days. Following this, a concentrated solution of aqueous barium
hydroxide was added dropwise until the pH of the solution was 8.0. The mixture was
then vacuum filtered with a Hirsch funnel to remove all solids. The solvent was
removed from the supernatant by rotary evaporation. Yield 48% (90% compound 4,
10% compound 5). 'H NMR (D20): 5(ppm)= 2.05(dd, 2H); 3.65 (s, 6H); 4.35( dd,
4.3.15 Attempted preparation of methyl 4-amino-5-oxopyrrolidine-2-
carboxylate (5) from sulfate salt of 4
A 1:2.5 molar equivalent of dried triethyl amine (TEA) (0.511 mL) was added to
0.422 grams (1.5 moles) of the sulfate salt of 4 in a 50 mL round bottom flask
equipped with a stir bar and 20.0 mL methylene chloride. The solution was refluxed
for 4 days. After this time the mixture was vacuum filtered and the supernatant dried
on a rotary evaporator. After this, residual TEA was removed by dissolving the solid
in a 1:24 methanol: methylene chloride solution (v/v) and eluting from a silica
column. Yield 15% (70% compound 5,30% compound 4). *H NMR (D20): 6(ppm)=

2.05(dd, 2H); 2.15(dt, 1H)*; 2.95(ddd, 1H)*; 3.65(s, 6H); 3.75(s, 3H)*; 4.35( dd,
1H); 4.65(dd, 2H)*; 5.05(dd, 1H)*.
*compound 5
4.3. 16 Preparation of methyl 4-amino-5-oxopyrrolidine-2-carboxylate (5) with
basic methanol
Approximately 7.0 mL of dry methanol saturated with HCl(g) was added to a 25 mL
round bottom flask equipped with a micro-stirbar and 0.322 g (2 mmol) of 3. The
mixture was refluxed until all the solids had dissolved, roughly 24 hours. After 24
hours the acidic methanol removed by rotary evaporation. Approximately 5.0 mL of
methanol saturated with Ntfyg) was added to the resulting solid and the mixture was
stirred for 15 min. After this time, 14.0 mL of ethyl ether were added to the solution.
The resulting precipitate was removed by vacuum filtration in a Hirsch funnel and the
supernatant dried on a rotary evaporator.23 Yield 1%. 'H NMR (CDCI3): 8(ppm)=
2.42(dt, 1H); 2.62(ddd, 1H); 3.82(s, 3H); 4.48(dd, 2H); 4.60(dd, 1H).
4.3.17 Preparation of methyl 4-amino-5-oxopyrrolidine-2-carboxylate (5) with
basic chloroform
Approximatly 12 mL of dry methanol saturated with HCl(g> was added to a 25 mL
round bottom flask equipped with a micro-stirbar and 0.4146 g (2.5 mmol) of 3. The

mixture was refluxed until all the solids had dissolved, roughly 18 hours. After this
time, the methanol was removed by rotary evaporation and approximately 10 mL of
chloroform saturated with NH3(g) was added to the resulting solid and the mixture was
stirred for 15 min. After this time, the resulting precipitate was removed by vacuum
filtration in a Hirsch funnel and the supernatant dried on a rotary evaporator. Yield
55% (60% compound 4, 40% compound 5). *H NMR (CDCR): 5(ppm)= 2.05(dd,
2H); 2.42(dt, 1H); 2.62(ddd, 1H); 3.65 (s, 6H); 3.82(s, 3H); 4.4( dd, 2H); 4.48(dd,
2H); 4.60(dd, 1H).
4.3.18 Preparation of 2,4-Bis-(9H-fluoren-9-ylmethoxycarbonylamino)-
pentanedioic acid (fmoc protected 3)
Dissolved 0.4500 g (2.78 mmol) of compound 3 and 0.7814 (5.56 mmol) of
potassium carbonate in 6.0 mL of DI water and an ice bath was added. A solution of
1.7016 g (2.8 mmol) 9-fluorenylmethyl chlorocarbonate (fmoc Cl) in 4.0 mL dioxane
was added to the aqueous solution over the period of an hour with vigorous stirring.
When all of fmoc Cl solution had been added, the ice bath was removed and the
mixture stirred at room temperature for 20 hours. After this time, the fmoc Cl had
precipitated. It was collected by vacuum filtration, dissolved in 3 mL dioxane, and
added back into the solution. The mixture was stirred for an additional 24 hours at
room temperature. After this time a yellow precipitate was collected by vacuum

filtration and a*H NMR spectrum taken.25 Yield 33%. 'H NMR (d-acetone):
5(ppm)= 2.05(dd, 2H); 3.2(s, 4dH); 3.9(dd, 2H); 4.2 (t, 1H); 4.35(t, 3H); 4.55(d, 2H);
6.25(broad peak, 3H); 7.4(m, 4H); 7.7(m, 2H); 7.9(m, 2H).
4.3.19 Attempted Preparation of 2,4-Bis-benzyloxycarbonylamino-pentanedioic
acid (Cbz protected 3)
Approximately 0.2472 g (1.5 mmol) of compound 3, 0.296 g (mmol) potassium
carbonate, and 0.3081 g (mmol) sodium bicarbonate were dissolved in 3.0 mL of 2N
NaOH in a 25 mL round bottom flask equipped with a stir bar and an ice bath was
added. Once all solids dissolved, 5.4 mL (3.81 mmol) benzylchloroformate was
added in equal portions over three hours. Once all of the benzylchloroformate had
been added, the ice bath was removed and the mixture stirred at room temperature for
22 hours. The mixture was diluted with 5 mL of deionized water, acidified to pH 1.2
with 12M HC1 and extracted with ethyl acetate (10x3 mL). The organic fraction
was rotary evaporated.21 *H NMR (D20): 5(ppm)= 2.0(dd, 2H); 4.3(dd, 3H); 5.0(s,
2H); 7.35(s, 5H).

[1] Paul, P.K.C.; Burney, P.; Campbell, M.; Osguthorpe, D.J. Bioorganic and
Medicinal Chemistry Letters. Vol 2 No 2. 1992: 141-142.
[2] Venkatachalam, C.M., Biopoymers. 6, 1968; 1425.
[3] Danieli, E., Trabocchi, A., Menchi, G., Guama, A. European Journal of Organic
Chemistry, 10, 2007: 1659-1668.
[4] DeSantos, P.; Morosetti, S.; Rizzo, R. Macromolecules, 7, 197: 452-58.
[5] Ghadiri, M.; Granja, M.; Milligan, R.; McRee D.; Khazanovich, N. Nature, 366,
[6] Kim, H.; Hartgerink, J.; Ghadiri, M. J: Am. Chem. Soc., 120, 1998: 4417-4424.
[7] Farrera, J.; Zaccaro, L.; Vidal, D.; Salvatella, X.; Giralt, E.; Pons, M.; Albericio,
F. J. Am. Chem. Soc., 126, 2004: 6048-6057.
[8] Crisma, M.; Toniolo, C.; Royo, S.; Jimenez, A.; Cativiela, C. Organic Letters,
vol 8, number 26,2006: 6091-6094.
[9] Modeling completed by D. Dyckes using Spartan P-Chem 1.0.0
[10] Crisma, M.; Moretto, A.; Toniolo, C.; Kaczmarek, K.; Zabrocki, J.
Macromolecules, 34, 2001: 5048-5042.
[11] Reiriz, C.; Castedo, L.; Granja, J. of Peptide Science, 14, 2007: 241-249.
[12] Ranganathan, D.; Karle, I. J. of Molecular structure, 647, 2003: 85-96.
[13] Ghadiri, M.; Home, W.; Maarseveen, J. Organic Letters, vol 7, number 20, 2005:

[14] Robbins, E. Synthesis of Cyclic Oligomers From the Lactam Amide of 2,4-
diaminoglutaric acid. Unpublished UROP Application, 2004.
[15] Long, J., Attempted Synthesis of2,4-DiaminoglutarateMethyl Ester.
Unpublished independent research report, 2001.
[16] English, Michelle. Study of a Proposed (Simplified) Preparation of Methyl 4-
Amino-5-Oxopyrrolidine-2-Carboxylate, a Cyclic Monomer Designed for use
in Hypercyclic Structures. Undergraduate honors thesis, University of
Colorado Denver, 2006.
[17] Avenoza, A., Cativiela, C., Peregrina, J. M., Zurbano, M. M. Tetrahedron:
Asymmetry, 8, 1997: 863-871.
[18] Liang, X., Andersch, J., Bols, M. J. Chem. Soc., Pekin Trans., 1, 200:
[19] Tanaka, K., Sawanishi, H. Tetrahedron: Asymmetry, 9,1998: 71-77.
[20] Kabat, M. M. Tetrahedron Letters, 42, 2001: 7521-7524.
[21] Allen, C. F., Kalm, M. J. Journal of Organic Syntheses, 4, 1921: 616-620.
[22] Horiuchi, C.A., Satoh, J.Y., Chemistry Letters, 1984:1509-1510.
[23] Silverstein, R.M., Bassler, G.C., Marrill, T.C. Spectrometric identification of
organic compounds. 5th ed. John Wiley & Sons, INC.: New York, 1991.

[24] Piper, J.R.; Stringfellow, Jr., C.R.; Johnston, T.P. Journal of Medicinal
Chemistry. Vol 9, No 6. 1966: 911-920.
[25] Bodanszky, M., Bodanszky, A. The Practice of Peptide Synthesis. Springer-
Verlag, Berlin; 1984.
[26] Meyer, Hoyt. Unpublished Ph.D Dissertation, University of California Berkeley,
[27] Lee, S.S., Jeong, H.M., Jho, J.Y., Oan, T.A. Journal of Polymer Science Part B:
Polymer Physics, vol 38, issue 5, 1999: 803-811.