Synthesis and immunostimulating properties of novel adamant-1-yl tripeptides

Article abstract of DOI:10.1002/cbdv.201100232

The aim of this work was to prepare L- and D-(adamant-1-yl)-Gly-L-Ala-D- isoGln peptides in order to study their adjuvant (immunostimulating) activities. Adjuvant activity of adamant-1-yl tripeptides was tested in the mouse model using ovalbumin as an ant

Full text of DOI:10.1002/cbdv.201100232

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
777
Synthesis and Immunostimulating Properties of Novel Adamant-1-yl
Tripeptides
a
b
b
a
-
by Rosana Ribic´ ), Lidija Habjanec ), Branka Vranesˇic´*, Ruza Frkanec ), and Srdanka Tomic´ )
ˇ
a) Department of Chemistry, Faculty of Science, University of Zagreb, Horvatovac 102a,
HR-10000 Zagreb, Croatia
b) Institute of Immunology, Rockfellerova 2, HR-10000, Zagreb, Croatia (phone: þ38516414380;
fax: þ38516414103; e-mail: bvranesic@imz.hr)
The aim of this work was to prepare l- and d-(adamant-1-yl)-Gly-l-Ala-d-isoGln peptides in order
to study their adjuvant (immunostimulating) activities. Adjuvant activity of adamant-1-yl tripeptides was
tested in the mouse model using ovalbumin as an antigen and in comparison to the peptidoglycan
monomer (PGM; b-d-GlcNAc-(1!4)-d-MurNAc-l-Ala-d-isoGln-mesoDAP(eNH2)-d-Ala-d-Ala) and
structurally related adamant-2-yl tripeptides.
Introduction. – Peptidoglycans are the ubiquitous constituents of bacterial cell
walls, built of large polysaccharide chains and short peptide units. Natural and synthetic
peptidoglycans, as well as their high- or low-molecular-mass fragments were widely
tested due to their remarkable biological activites, particularly as potent immunomo-
dulators [1][2]. Smaller-size peptidoglycans, as well as some peptidoglycan fragments
of which the most widely known are muramyl peptides, had been extensively studied,
since such compounds could be considered for use as adjuvants for human and animal
vaccines [3–5]. Muramyl peptides represent structural analogs of the peptidoglycan
monomer (PGM; 1; Fig. 1) of bacterial cell walls, and MDP (muramyl dipeptide; N-
acetyl-muramyl-l-alanyl-d-isoglutamine) has been known as the smallest synthetic
adjuvant-active molecule capable of replacing whole Mycobacteria in Freundꢀs
adjuvant [5].
On the other hand, compounds containing adamantyl (¼tricyclo[3.3.1.13,7]decyl)
residues are known to exhibit various biological activities, they are used as antiviral
agents, drugs in the treatment of Parkinsonꢀs disease, and depression [6–8]. Recently,
the role of adamantane conjugates with mannose in the process of preventing microbial
adhesion to host cells was discussed as well [9]. Furthermore, it is known that
introducing an adamantyl moiety into substances with known biological activity
improves their pharmacological properties and enhances their activity [10].
Several hundred chemically defined MDP analogs and derivatives were synthe-
sized, in order to modulate, preferably improve, the properties of the parent molecule.
Among those was a new class of compounds where l-Ala-d-isoGln was linked to
adamantanamine [11–15] and a group of desmuramyl peptides comprising a
phthalimido group instead of N-acetylmuramic acid [16–18]. Our research in the
field of potential adjuvants aims at the synthesis of desmuramyl peptides that comprise
an adamantyl moiety as a liphophylic part of the newly synthesized substances, and we
ꢁ 2012 Verlag Helvetica Chimica Acta AG, Zꢂrich

778
CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
Fig. 1. Structures of the PGM (1) and adamant-2-yl tripeptides 2a and 2b [19]
previously reported the synthesis of derivatives in which the N-acetylmuramic portion
of MDP was replaced with adamantylglycine. Previously synthesized admantyl
tripeptides, 2a and 2b (Fig. 1), had the peptide portion linked at C(2) of the adamantyl
moiety [19]. These diastereoisomers were tested separately and were shown to differ in
biological activity in several model systems [19–22].
In this work, adjuvant (immunostimulating) activity of synthetic adamantyl
tripeptides (AdTP), diastereoisomers of d,l-(adamant-2-yl)-Gly-l-Ala-d-isoGln
(Ad2TP1 (2a) and Ad2TP2 (2b)) and the newly synthesized diastereoisomers of d,l-
(adamant-1-yl)-Gly-l-Ala-d-isoGln (Ad1TP1 (3a) and Ad1TP2 (3b); Fig. 2), which
comprise in their structure the adamantylglycyl moiety linked to the dipeptide l-Ala-d-
isoGln, characteristic of the peptide portion of natural peptidoglycans and synthetic
muramyl dipeptides, were studied. The adamantyl moiety was used in order to increase
the lipophilic properties of this dipeptide, but it was also assumed that it might enhance
its adjuvant activity. The adjuvant activities of adamantyl tripeptides and PGM (1)
(peptidoglycan monomer, GlcNAc-MurNac-l-Ala-d-isoGln-mesoDAP(eNH2)-d-Ala-
Fig. 2. Structures of d- and l-(adamant-1-yl)-Gly-l-Ala-d-isoGln diastereoisomers 3a and 3b

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
779
d-Ala) were compared. PGM (1) was established as a reference for adjuvant activity in
a previously well-defined experimental model in vivo, established as suitable for
comparison of the observed effects [23].
Our current study should provide information on structureꢀactivity relationship
concerning the possible difference in the activity of the diastereoisomers at C(1)
(Ad1TP1 (3a) and Ad1TP2 (3b)) and C(2) (Ad2TP1 (2a) and Ad2TP2 (2b)) of the
adamantyl moiety. Comparative study of adjuvant effects of adamantyl tripeptides on
humoral IgG (immunoglobulin G) immune response was estimated and carried out in
our well-defined experimental mouse model in vivo [23][24]. Ovalbumin was used as a
customary antigen for studying the adjuvant effect. The comparison of induced anti-
OVA IgG levels was carried out quantitatively, and the subclasses of IgG, IgG1, and
IgG2a, as indicator of Th1 or Th2 type of immune response, were also determined.
Only a few adjuvants have been approved for prophylactic vaccination of humans,
in spite of decades of research and a few hundreds of pre-clinical candidates [25][26].
Obviously, there is a problem in designing a potent and well tolerated adjuvant that is
needed today with the new generation of vaccines. These highly purified recombinant
proteins (as antigens) are mostly poorly immunogenic and require adjuvants to become
effective vaccines.
Immunomodulating characteristics of natural peptidoglycan molecules are well-
known [2][27–29], and some of these compounds were considered for use as adjuvants
for human and animal vaccines. Furthermore, several successful attempts to syntheti-
cally modify PGM (1) as a parent compound were undertaken in spite of its complex
structure. In previous reports, we described the synthesis of 1 modified with Boc-
tyrosine [30], adamant-1-yl [31], as well as with mannopyranosyl [32] residues. Smaller
synthetic molecules such as muramyl dipeptides and tripeptides were investigated as
well, and a number of structureꢀactivity studies were reported. Thus, several
synthetically modified MDPs, syntheses of which were patented, have already been
included in clinical trials [33]. Furthermore, formulations in liposomes of these
compounds were found to be of interest. Therefore, modifications with lipophilic
substituents are of special relevance, since lipophilic parts in parent molecules render
them better constituents for incorporation in liposomes [34][35]. In view of these
reports including some of our previous work [30][32], it seemed justified to further
explore the influence of the adamantyl moiety with its lipophilic character on biological
properties of the dipeptide building block of MDP, with emphasis on its adjuvant
activity.
Results and Discussion. – Chemistry. In this work, the adjuvant (immunostimulat-
ing) activities of synthetic adamantyltripeptides Ad2TP1 (2a) and Ad2TP2 (2b) [19],
and newly synthesized Ad1TP1 (3a) and Ad1TP2 (3b) were studied.
d- and l-(adamant-1-yl)glycyl-l-alanyl-d-isoglutamine (3a and 3b, resp.) were
prepared by a sequence of steps as described in the Scheme. (tert-Butoxy)carbonyl-d-
glutamic acid 5-tert-butyl ester (Boc-d-Glu-OtBu; 4) was converted to an amide (via
the carbamate intermediate) to give the isoglutamine derivative 5 in excellent yields
[36]. d-Isoglutamine tert-butyl ester hydrochloride (6) was then successfully prepared
by removing the N-Boc group in the presence of the tert-butyl ester, another acid-
sensitive group, by treatment with HCl gas in AcOEt [37][38]. The isolation of

780
CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
Scheme. Synthesis of Adamant-1-yl tripeptides 3a and 3b.
a) ClCOOEt, Et3N, NH3(g), CHCl3. b) HCl(g) in AcOEt. c) 1-Ethyl-3-[3-(dimethylamino)propyl]car-
bodiimide (EDC)·HCl, 1-hydroxybenzotriazole (HOBt)·H2O, N-ethylmorpholine, Fmoc-l-Ala
(Fmoc¼(9H-fluoren-9-yl)methoxycarbonyl), dioxane/CH2Cl2 1:1. d) 1,8-Diazabicyclo[5.4.0]undec-7-
ene (DBU), octane-1-thiol, THF. e) EDC·HCl, HOBt·H2O, Et3N, Boc-(1-Ad)Gly (Boc¼(tert-
butoxy)carbonyl), dioxane/CH2Cl2 1:1. f) CF3COOH (TFA)/CH2Cl2 1:1.
compound 6 was facile, although we have shown that other procedures may be used, as
well. Thus, concentrated H2SO4 in AcOtBu or methanesulfonic acid (MsOH) in
AcOtBu/CH2Cl2 were shown to be efficient reagents for the selective Boc deprotection
of compound 5. The main disadvantage of these methods was the partial solubility of
the starting amino acid 5 in AcOtBu, causing some difficulties in the separation of the
unreacted 5 and the product 6 which precipitates during the reaction progress.
Dipeptide 7 was prepared next by condensation of 6 with [(9H-fluoren-9-yl)methox-
y]carbonyl-l-alanine (Fmoc-l-Ala), followed by removing the N-Fmoc protection [39]
to give l-alanyl-d-isoglutamine tert-butyl ester (8). This method was much more
efficient than the one using piperidine/THF due to the low volatility of the solvent. rac-
(Adamant-1-yl)glycine hydrochloride was prepared by the Co-mediated b-keto ester
alkylation, followed by Schmidt rearrangement and acid hydrolysis [40]. Subsequently,
Boc-l,d-(adamant-1-yl)glycine (Boc-(1-Ad)Gly) was synthesized by N-Boc protection
of rac-(adamant-1-yl)glycine hydrochloride. The racemic amino acid was used, even
though the optically pure (adamant-1-yl)glycines and their derivatives are known
[40][41]. The diastereoisomeric mixture Boc-d/l-(adamant-1-yl)glycyl-l-alanyl-d-iso-
glutamine tert-butyl ester (9a/9b, resp.) was prepared by the condensation of dipeptide
8 with rac-Boc-(1-Ad)Gly [9][32]. The tripeptide was deprotected in the next step
t
without previous separation of the diastereoisomers. The protecting N-Boc and BuO
groups were then removed in one step [42] to yield the mixture d-/l-(adamant-1-
yl)glycyl-l-alanyl-d-isoglutamine (3a/3b, resp.) in good yields. Pure 3a and 3b were
obtained after separation by column chromatography on silica gel, and were

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
781
unequivocally identified by NMR spectroscopy and enzymatic oxidations. Special
batches of endotoxin-free 3a and 3b were prepared for testing their biological activities.
The absolute configurations of the diastereoisomers 3a and 3b were established by
enzymatic oxidations of their hydrolysates using l-amino acid oxidase [19][43] at pH 7.
Total acid hydrolysis of 3a and 3b with 6m HCl gave hydrolysates, which were
incubated, after the removal of HCl, with l-amino acid oxidase. The resulting
incubation mixtures were compared with the starting hydrolysates by TLC. Thus, no l-
Ala was detected in the mixture resulting from the incubation of the hydrolysate of 3a,
since l-Ala was oxidized. On the other hand, only d-Glu (from d-isoGln) was detected
in the oxidation mixture of the hydrolysate of 3b, since l-(Ad-1-yl)Gly and l-Ala were
oxidized.
The described amino acid oxidase experiments not only established the absolute
configurations of 3a and 3b, but also documented the preservation of the l-Ala
configuration during the peptide synthesis.
Testing the Immunostimulating Activity. Adjuvant activities of adamantyl tripep-
tides 2a, 2b, 3a, and 3b were evaluated by their immunostimulatory effects on
secondary humoral response to ovalbumin (antigen) in mice of two genetically
different strains, CBA (H-2k) and NIH/OlaHsd (H-2q), according to our previously
described mouse model in vivo [23][24]. Anti-OVA IgG, anti-OVA IgG1, and anti-
OVA IgG2a were determined in the mice sera after the second booster (Figs. 3 and 4).
Fig. 3. The effect of Ad1TP1 (3a), Ad1TP2 (3b), Ad2TP1 (2a), and Ad2TP2 (2b) on the production of
total anti-OVA IgG (a), and its subtypes anti-OVA IgG1 (b) and anti-OVA IgG2a (c), respectively, in CBA
(H-2k) mice. Mice were immunized with OVA as an antigene and sera were analyzed after second
booster. Experimental groups: 1, OVA alone; 2, OVAþPGM (1); 3, OVAþ3a; 4, OVAþ3b; 5, OVAþ
*
*
2a; 6, OVAþ2b. , roup mean value; , each serum separately. : p<0.05 in comparison to the group
*
connected with dashed line.
In general, in both mice strains, no significant enhancement in total anti-OVA IgG
antibody production was observed, when both adamant-1-yl and adamant-2-yl
tripeptides were applied in comparison to no adjuvant (OVA alone) treated group.
In comparison to the control, PGM-injected group, no better immunostimulating effect

782
CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
Fig. 4. The effect of Ad1TP1 (3a), Ad1TP2 (3b), Ad2TP1 (2a), and Ad2TP2 (2b) on the production of
anti-OVA IgG (a), and its subtypes anti-OVA IgG1 (b) and anti-OVA IgG2a (c), respectively, in NIH/
OlaHsd (H-2q) mice. Mice were immunized with OVA as an antigene, and sera were analyzed after
second booster. Experimental groups: 1, OVA alone; 2, OVAþPGM (1); 3, OVAþ3a; 4, OVAþ3b; 5,
*
*
OVA þ2a; 6, OVAþ2b. , Group mean value; , each serum separately. : p<0.05 in comparison to the
*
group connected with dashed line.
of tested adjuvants was observed either. However, comparing the production of anti-
OVA IgG antibodies in mice of two different strains, a difference was observed. On the
whole, the measured quantity of anti-OVA IgG antibodies was higher in NIH/OlaHsd
(H-2q) mice than in CBA (H-2k) mice. The same trend was detected in both mice
strains, when Ad1TP1 (3a) and Ad1TP2 (3b) were applied; a weak stimulation of
antibody production was achieved with Ad1TP1 (3a), whereas Ad1TP2 (3b) was devoid
of any stimulation. But, when adamant-2-yl tripeptides were applied, a different trend
in antibody production was detected. In CBA (H-2k) mice, the Ad2TP2 (2b) isomer
showed a higher stimulation of anti-OVA IgG antibodies production when compared to
Ad2TP1 (2a). When NIH/OlaHsd (H-2q) mice were used in the experiment, the
reversed situation was observed: Ad2TP1 (2a) elicited the antibody production better
than Ad2TP2 (2b).
Since it is well-known that vaccine adjuvants can enhance or modulate the Th1/Th2
bias of induced immune response [44][45], we were also interested in the effect of
tested adjuvants on the type of immune response. In this study, the type of generated
immune response was indirectly estimated by quantification of OVA (antigen)-specific
IgG1 (for activation of Th2 type) and IgG2a (for activation of Th1 type), and
calculation of the respective IgG1/IgG2a ratio (Table).
When the amount of anti-OVA IgG1 antibodies was measured in two mice strains
injected with adamantyl tripeptides, it was observed that the response in the NIH/
OlaHsd (H-2q) mice was higher than in CBA (H-2k) mice. However, the obtained
values could be considered as a slight suppression in the production of anti-OVA IgG1
antibodies in comparison with the results obtained when OVA alone or OVA with
PGM (1) were administered. For Ad1TP2 (3b), the response was a little lower than for

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
783
Ad1TP1 (3a), and, for Ad2TP2 (2b), it was significantly higher than for Ad2TP1 (2a) in
CBA (H-2k) mice. Basically, it was the same pattern as for anti-OVA IgG antibodies. In
NIH/OlaHsd (H-2q) mice, there was essentially no difference in anti-OVA IgG1
antibody amount observed, neither when Ad1-tripeptides nor Ad2-tripeptides were
applied.
When anti-OVA IgG2a antibodies were measured, the same trend as for anti-OVA
IgG1 antibodies and anti-OVA IgG antibodies was observed in both mice strains.
Hence, in comparison to the experimental group injected with OVA alone, without
tested adjuvants, the amount of anti-OVA IgG2a in CBA(H-2k) mice was higher, when
Ad1TP1 (3a) and Ad2TP2 (2b) were applied, and, in NIH/OlaHsd (H-2q) mice, when
Ad1TP1 (3a) and Ad2TP1 (2a) were administered.
From IgG1/IgG2a ratio calculated for each serum (obtained after the second
booster) and each experimental group of two different mice strains, it could be
observed that neither of tested adamantyl tripeptides significantly shifted the IgG1/
IgG2a ratio in comparison to no adjuvant-treated group of animals (Table).
Accordingly, the tested adamantyl tripeptides did not stimulate immunomodulatory
activity towards more pronounced Th1 or Th2 type of immune response in comparison
to antigen alone.
Table. The Ratio of anti-OVA IgG1 and anti-OVA IgG2a (IgG1/IgG2a) in NIH/OlaHsd (H-2q) and CBA
(H-2k) Mice. For each mouse serum, obtained after the second booster, log10 IgG1/IgG2a was calculated,
and the result for each experimental group (n¼5) is presented as averageꢁstandard deviation (S.D.).
Exsperimental groups
NIH/OlaHsd (H-2q) Mice
CBA (H-2k) Mice
No adjuvant
PGM (1)
Ad1TP1 (3a)
Ad1TP2 (3b)
Ad2TP1 (2a)
Ad2TP2 (2b)
2.07ꢁ0.1
2.01ꢁ0.5
1.65ꢁ0.1
2.11ꢁ0.3
1.83ꢁ0.6
2.03ꢁ0.5
2.55ꢁ0.2
2.23ꢁ0.4
1.87ꢁ1.0
2.33ꢁ0.2
2.12ꢁ0.2
2.59ꢁ0.7
Conclusions. – l- and d-(adamant-1-yl)-Gly-l-Ala-d-isoGln peptides were pre-
pared in six steps. All intermediate products and the final adamantyl peptides were
fully characterized. The desired diastereoisomers, d-(adamant-1-yl)-Gly-l-Ala-d-
isoGln (Ad1TP1; 3a) and l-(adamant-1-yl)-Gly-l-Ala-d-isoGln (Ad1TP2, 3b), were
separated, and their absolute configurations were unequivocally determined.
Adjuvant activities of adamant-1-yl and adamant-2-yl tripeptides were tested in
mice of two genetically different strains, CBA (H-2k) and NIH/OlaHsd (H-2q). All
tested diastereoisomers elicited immune response but lower than that of PGM (1) in
both mouse strains.
Comparing the diastereoisomers of adamant-1-yl tripeptides revealed that the
induction of all three antibody types was better, when Ad1TP1 (3a) was applied in CBA
(H-2k) mouse strain. On the contrary, when administered in CBA (H-2k) mice,
adamant-2-yl tripeptides revealed the opposite behavior: Ad2TP2 (2b) elicited better
immune response with respect to IgG, IgG1, and IgG2a induction. In NIH/OlaHsd (H-
2q) mouse strain, the pattern of IgG, IgG1, and IgG2a induction was basically the same

784
CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
for both adamant-1-yl and adamant-2-yl tripeptides; Ad1TP1 (3a) and Ad2TP1 (2a)
were always slightly better than Ad1TP2 (3b) and Ad2TP2 (2b), respectively, in
inducing immune response.
In both CBA (H-2k) and NIH/OlaHsd (H-2q) mice strains, it was demonstrated that
the immune response specific for OVA alone was Th2-biased due to the predominant
appearance of IgG1 antibodies. PGM (1), as well, dominantly induced IgG1 antibody
production and stimulated Th2-biased immune response.
In comparison to antigen alone, all tested adamantyl tripeptides did not show a
tendency to modulate the OVA-specific response towards more pronounced Th1 or
Th2 type of immune response. But a pronounced difference in IgG1 response was
observed, when different mouse strains were used in the study. Regarding the IgG1
antibody production, the immune response in NIH/OlaHsd (H-2q) mouse strain was
approximately three times higher than in CBA (H-2k) mouse strain.
We thank the Ministry of Science, Education and Sports of the Republic of Croatia for support of this
work (projects 119-1191344-3121 and 021-0212432-2431).
Experimental Part
General. Bovine serum albumin (BSA), Tween 20, monoclonal anti-chicken egg albumin (clone
OVA-14, mouse IgG1 isotype), avidin peroxidase, and ortho-phenylenediamine dihydrochloride (OPD)
were from Sigma (USA). Ovalbumin (OVA) was purchased from Serva, Germany. Horseradish
peroxidase conjugated goat anti-mouse IgG (HRP-anti-mouse IgG) was from Bio-Rad Laboratories,
USA. Biotin-conjugated rat anti-mouse IgG1 and anti-mouse IgG2a monoclonal antibodies and
streptavidin-peroxidase were purchased from PharMingen, Becton Dickinson (USA). l-Amino acid
oxidase (EC 1.4.3.2.) from Crotalus adamanteus, type IV, was purchased from Sigma (USA). Chemicals
for buffers and solns. were from Kemika, Croatia, unless stated otherwise. Peptidoglycan monomer was
prepared in PLIVA (Chemical and Pharmaceutical Works, HR-Zagreb), according to the procedure
described in [46]. (Adamant-2-yl)glycine [47], and adamant-2-yl tripeptides, Ad2TP1 (2a) and Ad2TP2
(2b) [19] were prepared at the Institute of Immunology, as described earlier. Racemic (adamant-1-
yl)glycine hydrochloride was prepared as described in [40]. Chemical reagents used in syntheses were
obtained from Fluka and Aldrich Corp. All org. solvents were purified using standard procedures.
Column chromatography (CC, solvents and proportions are given in the text) was performed on
Merck silica gel 60 (70–230 mesh ASTM). For TLC monitoring, Fluka silica gel (60 F254) plates
(0.25 mm) were used. Visualization was effected with UV light, I2, and ninhydrin. M.p.: in open
capillaries with a Bꢀchi B-540 apparatus. Optical rotations [a]D: at r.t. with the SchmidtþHaensch
Polartronic NH8. NMR Spectra: Bruker Avance (300 MHz) spectrometer; d in ppm rel. to TMS, J in Hz.
MS: Waters MS-Quattro micro instrument; in m/z. C, H, and N analyses were provided by the Analytical
ˇ
´
Services Laboratory of the Ruꢃer Boskovic Institute, Zagreb.
HPLC Analyses (Thermo Separation Products, Spectra SYSTEM 2000) were performed on an RP-18
column (Vydac, 5 mm, 218ꢂ4 mm) at r.t. at a flow rate of 1.0 ml/min, and the eluate was monitored at
214 nm. The gradient solvent system used consisted of MeCN containing 0.035% TFA, and H2O
containing 0.05% TFA. The percentage of MeCN at 0, 20, and 21 min was 3, 17 and 3, resp., and the
running time was 25 min.
[(tert-Butoxy)carbonyl]-d-isoglutamine tert-Butyl Ester (5). To a soln. of Boc-d-Glu-OtBu (4; 2.5 g,
8.24 mmol) in dry CHCl3 (30 ml) at ꢀ158, Et3N (1.26 ml, 9.06 mmol, 1.1 equiv.) and ClCOOEt (866 ml,
9.06 mmol, 1.1 equiv.) were added. The mixture was stirred for 0.5 h at ꢀ158. Dry NH3(g) was then
passed through the stirred mixture for 2 h at 08. The soln. was diluted with CHCl3, extracted twice with
7% K2CO3, and washed twice with H2O. The org. layer was dried (MgSO4) and, after filtration,
evaporated to yield 5 (2.38 g, 95%). White solid. Rf (iPrOH/petroleum ether (PE)/H2O, 5 :3 :1) 0.82. M.p.
122.7–124.58. [a]D ¼ ꢀ4.6 (c ¼ 0.5, MeOH). 1H-NMR (CDCl3): 6.46 (s, NH); 5.81 (s, NH); 5.41 (d, J ¼

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
785
7.3, NH); 4.21–4.13 (m, HꢀC(a)); 2.49–2.27 (m, CH2(g)); 2.16–2.04 (m, 1 H, CH2(b)); 1.96–1.84 (m,
1 H, CH2(b)); 1.45 (s, Boc); 1.44 (s, tBu). 13C-NMR (CDCl3): 174.10 (CONH2); 172.74 (COOtBu); 155.75
(CO of Boc); 80.93 (Me3C); 53.54 (CH(a)); 31.79 (CH2(g)); 28.30 (Me3C of Boc); 28.06 (Me3C); 27.73
(CH2(b).
d-Isoglutamine tert-Butyl Ester Hydrochloride (6). A freshly prepared sat. soln. of dry HCl in dry
AcOEt (40 ml), prepared by bubbling HCl into AcOEt for 1 h, was added to 5 (2 g, 6.6 mmol). The
mixture was stirred for 3 h at r.t. During the reaction, a white solid precipitated. The mixture was
evaporated, and Et2O was added to the residue. Product 6 (0.97 g, 62%) was filtered off as a white solid.
Rf (iPrOH/PE/H2O, 5 :3 :1) 0.51. M.p. 212.2–214.48. [a]D ¼ ꢀ10 (c ¼ 0.5, MeOH). 1H-NMR (D2O): 3.96
t
(t, J¼6.6, HꢀC(a)); 2.39 (t, J¼7.5, CH2(g)); 2.10ꢀ2.03 (m, CH2(b)); 1.36 (s, Bu). 13C-NMR (D2O,
dioxane): 173.39 (CONH2); 171.33 (COOtBu); 83.05 (Me3C); 52.18 (CH(a)); 30.43 (CH2(g)); 27.11
(Me3C); 25.70 (CH2(b)). ESI-MS: 203.20 (Mþ, C9H19N2O3þ ; calc. 203.26).
{[(9H-Fluoren-9-yl)methoxy]carbonyl}-l-alanyl-d-isoglutamine tert-Butyl Ester (7). To a soln. of
Fmoc-Ala (1 g, 3.2 mmol) in dry CH2Cl2 (15 ml) at 08, 3-[3-(dimethylamino)propyl]-1-ethylcarbodiimide
(EDC·HCl; 738 mg, 3.85 mmol, 1.2 equiv.) and N-hydroxybenzotriazole (HOBt; 521 mg, 3.85 mmol, 1.2
equiv.) were added. The mixture was then stirred for 0.5 h at the same temp., and a soln. of 6 (789 mg,
3.85 mmol, 1.2 equiv.) in dioxane (30 ml) and N-ethylmorpholine (813 ml, 6.42 mmol, 2 equiv.) were then
added. The mixture was stirred for 1 h at 08, and for 48 h at r.t. The soln. was diluted with CH2Cl2,
extracted with 0.5m HCl, and washed with sat. aq. NaHCO3. The org. layer was dried (Na2SO4) and, after
filtration, evaporated. The residue was purified by CC (silica gel; MeCN/H2O 5 :1) to afford 7 (1.21 g,
76%). White solid. Rf (MeCN/H2O 5 :1) 0.78. M.p. 190.3–1928. [a]D ¼ þ16.4 (c ¼ 0.5, DMF). 1H-NMR
(CDCl3): 7.74 (d, J¼7.4, HꢀC(4) and HꢀC(5) of Fmoc); 7.56 (d, J¼6.8, HꢀC(1) and HꢀC(8) of Fmoc);
7.41 (d, J ¼ 7.1, NH); 7.39–7.28 (m, HꢀC(2), HꢀC(7), HꢀC(3), and HꢀC(6) of Fmoc); 4.46–4.43 (m,
HꢀC(a) of Ala); 4.37 (d, J ¼ 6.9, CH2 of Fmoc); 4.23 (t, J¼6.5, HꢀC(9) of Fmoc); 4.18 (t, J¼6.1,
HꢀC(a) of isoGln); 2.45–2.42 (m, 1 H, CH2(g)); 2.34–2.31 (m, 1 H, CH2(g)); 2.13–2.10 (m, 1 H,
t
CH2(b)); 2.00–1.94 (m, 1 H, CH2(b)); 1.36 (s, Bu); 1.40 (d, J¼5.9, Me of Ala). 13C-NMR (CDCl3):
173.72, 172.93 (CONH and COOtBu); 158.25 (CO of Fmoc); 143.72, 143.65, 141.25 (arom. C of Fmoc);
127.73, 127.09, 127.06, 119.97 (arom. CH of Fmoc); 81.23 (Me3C); 67.15 (CH2 of Fmoc); 52.59 (CH(a) of
isoGln); 50.90 (CH(a) of Ala); 47.05 (CH of Fmoc); 31.82 (CH2(g)); 28.01 (Me3C); 27.10 (CH2(b)); 18.32
(Me of Ala). ESI-MS: 496.30 ([MþH]þ , C27H34N3O6þ ; calc. 496.24). Anal. calc. for C27H33N3O6: C 65.44,
H 6.71, N 8.48; found: C 65.76, H 7.11, N 8.12.
l-Alanyl-d-isoglutamine tert-Butyl Ester (8). To a soln. of 7 (1000 mg, 2.02 mmol) in dry THF
(20 ml) under N2 , octane-1-thiol (4.2 ml, 20.2 mmol, 10 equiv.) and DBU (99.5 ml, 0.67 mmol,
0.33 equiv.) were added. The mixture was stirred at r.t. for 16 h and monitored by TLC (MeCN/H2O
5 :1). The mixture was then triturated with Et2O, and the white solid residue was purified by CC (silica
gel, CHCl3/MeOH 1:1) to afford 8 (304 mg, 55%). Colorless solid foam. Rf (MeCN/H2O 5 :1) 0.28.
[a]D ¼8.0 (c ¼ 0.5, MeOH). 1H-NMR (CD3OD): 4.36 (dd, J¼5.0, 6.7, HꢀC(a) of isoGln); 3.56 (q, J¼6.9,
HꢀC(a) of Ala); 2.34–2.31 (m, CH2(g)); 2.15–2.09 (m, 1 H, CH2(b)); 1.91–1.84 (m, 1 H, CH2(b)); 1.45
(s, tBu); 1.32 (d, J¼6.9, Me of Ala). 13C-NMR (CD3OD): 176.99, 176.30, 173.82 (CONH and COOtBu);
81.93 (Me3C); 53.53 (CH(a) of isoGln); 51.25 (CH(a) of Ala); 32.66 (CH2(g)); 28.59 (CH2(b)); 28.38
(Me3C); 20.71 (Me of Ala).
[(tert-Butoxy)carbonyl]-l,d-(adamant-1-yl)glycine (Boc-(1-Ad)Gly). l,d-(Adamant-1-yl)glycine
hydrochloride [40] (350 mg, 1.42 mmol) was suspended in dioxane/H2O 2 :1 (12 ml). Di(tert-butyl)
dicarbonate (467 mg, 2.84 mmol, 2 equiv.) and 1m NaOH (6.25 ml, 4.4 equiv.) were added. The mixture
was stirred at r.t. for 24 h and than acidified (pH 3) with sat. aq. KHSO4. The soln. was extracted twice
with AcOEt, the org. layer washed with H2O and dried (Na2SO4). The product Boc-(1-Ad)Gly (392 mg,
1
89%) was isolated as a white solid. Rf (AcOEt/AcOH/H2O 3 :1:1) 0.80. M.p. 187.6–189.98. H-NMR
(CDCl3): 7.27 (s, NH); 4.01 (s, HꢀC(a)); 2.01 (br. s, 3 H of Ad); 1.73–1.55 (m, 12 H of Ad); 1.45 (s, Boc).
13C-NMR (CDCl3): 175.45 (COOH); 152.20 (CO of Boc); 62.40 (CH of Gly); 42.20, 40.24, 38.55, 36.64,
36.42, 36.35 (C and CH2 of Ad); 28.36, 28.32, 28.26, 27.94 (CH of Ad, Me3C).
[(tert-Butoxy)carbonyl]-d/l-(adamant-1-yl)glycyl-l-alanyl-d-isoglutamine tert-Butyl Ester (9a/9b).
To a soln. of Boc-(1-Ad)Gly (257 mg, 0.83 mmol) in dry CH2Cl2 (5 ml) at 08, EDC·HCl (191 mg, 1 mmol,
1.2 equiv.) and HOBt (135 mg, 1 mmol, 1.2 equiv.) were added. The mixture was stirred for 0.5 h at the

786
CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
same temp., and a soln. of 8 (272 mg, 1 mmol, 1.2 equiv.) in dioxane (5 ml) and Et3N (813 ml, 6.42 mmol,
2 equiv.) were then added. The mixture was stirred for 1 h at 08 and than for 48 h at r.t. The soln. was
diluted with CH2Cl2, extracted with 0.5m HCl, and washed with sat. aq. NaHCO3. The org. layer was
dried (Na2SO4) and, after filtration, evaporated. The residue was purified by CC (silica gel, AcOEt) to
afford 9a/9b (374 mg, 80%). Colorless solid foam. Rf (AcOEt) 0.31. 1H-NMR (mixture of diaster-
eoisomers; CDCl3): 7.43 (d, J¼7.6, NH); 7.35 (pt, J ¼ 9.4, 8.6, 2 NH); 6.98 (s, NH); 6.93 (s, NH); 6.07 (s,
NH); 5.99 (s, NH); 5.40 (d, J¼8.0, NH); 5.36 (d, J¼9.0, NH); 4.59–4.56 (m, 2 HꢀC(a) of Ala); 4.53–
4.48 (m, 2 HꢀC(a) of isoGln); 3.85 (d, J ¼ 9.1, 1 HꢀC(a) of 1-AdGly); 3.81 (d, J¼7.9, 1 HꢀC(a) of 1-
AdGly); 2.41–2.28 (m, 2 CH2(g)); 2.17–1.92 (m, 2 CH2(b)); 1.99 (br. s, 3 CH of Ad); 1.97 (br. s, 3 CH of
Ad); 1.73–1.52 (m, 12 CH2 of Ad); 1.44 (s, 2 Boc); 1.43 (s, tBu); 1.42 (br. s, tBu, Me of Ala); 1.37 (d, J¼
6.9, Me of Ala). 13C-NMR (CDCl3): 173.73, 173.59, 172.59, 172.53, 172.41, 172.34, 170.53, 170.49 (CONH
and COOtBu); 156.11, 156.09 (CO of Boc); 81.86 (Me3C); 63.28, 63.09 (CH of Gly); 52.55, 52.14 (CH(a)
of isoGln); 49.17, 48.98 (CH(a) of Ala); 38.77, 38.63, 36.71, 36.69, 31.84, 31.79, 27.76, 27.35 (CH2 of Ad,
CH2(g), C of Ad, CH2(b)); 28.38, 28.37, 28.24, 28.20, 28.04 (CH of Ad, Me3C of Boc); 18.30, 18.23 (Me of
Ala). ESI-MS: 565.40 ([MþH]þ , C29H49N4Oþ7 ; calc. 565.36).
d/l-(Adamant-1-yl)glycyl-l-alanyl-d-isoglutamine (3a/3b). The mixture 9a/9b (324 mg; 0.57 mmol)
was suspended in CH2Cl2 (4.3 ml), and TFA (4.3 ml, 5.7 mmol, 10 equiv.) was added. The mixture was
stirred at r.t. for 6 h and then concentrated in vacuo. The residue was purified by flash chromatography
(FC; silica gel, CHCl3/MeOH 1:1) to give 3a/3b (206 mg, 88%). Colorless solid.
Pure diastereoisomers 3a (88 mg) and 3b (113 mg) were separated from the mixture by CC (silica
gel, CHCl3/MeOH 1:1). The retention times obtained by RP-HPLC separation were 11.96 min for 3a
and 7.72 min for 3b.
d-(Adamant-1-yl)glycyl-l-alanyl-d-isoglutamine (3a). Rf (CHCl3/MeOH 1:1) 0.48. [a]D ¼ ꢀ76.3
1
(c ¼ 0.5, H2O). H-NMR (CD3OD): 4.39 (q, J¼7.3, HꢀC(a) of Ala); 4.08 (dd, J ¼ 3.4, 7.7, HꢀC(a) of
isoGln); 3.48 (s, HꢀC(a) of 1-AdGly); 2.49–2.41 (m, 1 H, CH2(g)); 2.28–1.91 (m, 1 H of CH2(g),
CH2(b)); 2.05 (br. s, 3 CH of Ad); 1.80–1.61 (m, 6 CH2 of Ad); 1.39 (d, J¼7.4, Me of Ala). 13C-NMR
(CD3OD): 182.35 (CONH2); 177.20, 174.74 (CONH); 169.05 (COOH); 64.13 (CH of Gly); 56.73 (CH(a)
of isoGln); 50.25 (CH(a) of Ala); 39.54, 37.53 (CH2 of Ad); 35.78 (C of Ad); 35.16 (CH2(g)); 29.62 (CH
of Ad); 27.86 (CH2(b)); 17.77 (Me of Ala). ESI-MS: 409.30 ([MþH]þ , C20H33N4O5þ ; calc. 409.25). Anal.
calc. for C20H32N4O5: C 58.81, H 7.90, N 13.72; found: C 58.47, H 8.20, N 13.69.
l-(Adamant-1-yl)glycyl-l-alanyl-d-isoglutamine (3b). Rf (CHCl3/MeOH, 1:1) 0.30. [a]D ¼ þ13 (c ¼
1
0.5, H2O). H-NMR (CD3OD): 4.24 (dd, J ¼ 3.7, 6.3, HꢀC(a) of isoGln); 4.16 (q, J¼7.1, HꢀC(a) of
Ala); 3.36 (s, HꢀC(a) of 1-AdGly); 2.32 (t-like, J¼6.3, 6.9, CH2(g)); 2.16–1.92 (m, CH2(b)); 2.03 (br. s,
3 CH of Ad); 1.79–1.64 (m, 6 CH2 of Ad); 1.45 (d, J¼7.1, Me of Ala). 13C-NMR (CD3OD): 181.34
(CONH2); 176.93, 173.98 (CONH); 170.22 (COOH); 63.95 (CH of Gly); 55.38 (CH(a) of isoGln); 51.76
(CH(a) of Ala); 39.41, 37.56 (CH2 of Ad); 36.26 (CH2(g)); 34.49 (C of Ad); 29.66 (CH of Ad); 28.17
(CH2(b)); 16.88 (Me of Ala). ESI-MS: 409.30 ([MþH]þ , C20H33N4O5þ ; calc. 409.25). Anal. calc. for
C20H32N4O5: C 58.81, H 7.90, N 13.72; found: C 59.17, H 8.15, N 13.78.
For testing the biological activities, special batches of endotoxin-free 3a and 3b were prepared by
passing the aq. soln. of 3a and 3b through a Detoxy gel column (Pierce, The Netherlands), followed by
lyophilization.
Enzymatic Oxidations with l-Amino Acid Oxidase. Hydrolysis and Enzymatic Oxidation of 3a. After
total acid hydrolysis of 3a (1 mg) in 6m HCl at 1008 for 16 h, HCl was evaporated, and the residue was
dissolved in phosphate buffer (pH 6.5). The pH of the soln. was adjusted to 7 by the addition of 1m
NaOH. l-Amino acid oxidase from Crotalus adamanteus, type IV (80 ml, 1.0 mg protein/ml, 5.0 units/
mg protein) was added to the hydrolysate, and the mixture was incubated for 24 h at 378. TLC with
i
BuOH/AcOH/H2O 6 :1.5 :2.5 and PrOH/NH3 7:3 revealed the expected amino acid composition: d-
glutaminic acid, d-(adamant-1-yl)glycine, and no l-alanine.
Hydrolysis and Enzymatic Oxidation of 3b. The same procedure (hydrolysis and enzymatic oxidation
with l-amino acid oxidase) was repeated for 3b (1 mg). TLC in the same solvent systems revealed only
d-Glu. No l-(Ad-1-yl)Gly and l-Ala were detected.
Experiments in vivo. Experiments in vivo were performed on NIH/OlaHsd (H-2q) and CBA (H-2k)
inbread mice strains. All mice used were females 2 to 2.5 months old. Commercial food and water were

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
787
provided ad libitum. During the experiments, animals were kept at the animal facility at the Institute of
Immunology and all experiments were performed according to the Croatian Law on Animal Welfare
(The Official Gazette ꢄNNꢀ 135/06).
Exper. groups of five mice were immunized and boosted two times subcutaneously (s.c.) into the tail
base at 21-day intervals. Mice were anesthetized prior to blood collection on 7th day after the second
booster. Sera were collected, decomplemented at 568 for 30 min, and stored at ꢀ208 until tested.
The dose of OVA (antigen) was 10 mg per mouse. The dose of PGM (1) and adamantyl tripeptides
was 200 mg per mouse. OVA and tested substances were dissolved in saline, and the injection volume in
all experimental groups was 0.1 ml per mouse.
Enzyme Immunoassays for Qual. and Quant. Determination of OVA-Specific IgG (anti-OVA IgG) in
Mice Sera. Enzyme immunoassays (ELISA) were performed on flat-bottomed high-binding microtiter
plates (Costar, USA) according to the procedures described in [32][48]. The relative quantities of anti-
OVA IgG were determined by parallel line assay comparing each serum to monoclonal anti-chicken egg
albumin, declared as standard preparation, to which 20,000 arbitrary units per ml (AU/ml) were
voluntarily assigned [49].
Enzyme Immunoassays for Qual. and Quant. Determination of OVA-Specific IgG Subtype Anti-OVA
IgG1 and Anti-OVA IgG2a in Mice Sera. ELISA for determination of anti-OVA IgG1 and anti-OVA
IgG2a were performed as described in [24]. The relative quantities of antibody subtypes were
determined by parallel line assay using appropriate standard preparation. The monoclonal anti-OVA
IgG1 was a standard for relative quantification of anti-OVA IgG1 to which 400,000 AU/ml was assigned,
while polyclonal mouse serum containing high levels of anti-OVA IgG2a was used as a standard for
relative quantification of IgG2a-specific antibodies with 20,000 AU/ml. The ratio IgG1/IgG2a was used
as indication of the Th1/Th2 bias of induced immune response.
Statistics. Statistical analyses were performed with Statistica 6.0 for Windows, StatSoft Inc. The
significant difference between exper. groups was evaluated by KruskalꢀWallis ANOVA, followed by
multiple MannꢀWhitney U-nonparametric tests. A probability values less than 0.05 (p<0.05) were
considered significant.
REFERENCES
[1] A. Adam, J. F. Petit, P. Lefrancier, E. Lederer, Mol. Cell. Biochem. 1981, 41, 7.
[2] D. E. S. Stewart-Tull, Prog. Drug Res. 1988, 32, 305.
[3] I. Azuma, Vaccine 1992, 10, 1000.
[4] Y. C. Yoo, K. Yoshimatsu, Y. Koike, R. Hatsuse, K. Yamanishi, O. Tanishita, J. Arikawa, I. Azuma,
Vaccine 1998, 16, 216.
[5] A. Adam, E. Lederer, ISI Atlas Sci.: Immunol. 1988, 1, 205.
[6] E. Lederer, in ꢄProceedings of the VIIIth International Symposium on Medicinal Chemistryꢀ, Eds.
R. Dahlbom, J. L. G. Nilsson, Swedish Pharmaceutical Press, Stockholm, 1984, Vol. 1, pp. 13–26.
[7] A. A. Spasov, T. V. Khamidova, L. I. Bugaeva, I. S. Morozov, Pharm. Chem. J. 2000, 34, 1.
[8] T. H. Maugh, Science 1976, 192, 130.
´
ˇ
´
´
´
´
´
´
[9] R. Ribic, M. Kovacevic, V. Petrovic-Perokovic, I. Gruic-Sovulj, V. Rapic, S. Tomic, Croat. Chem.
Acta 2010, 83, 421.
[10] K. Gerzon, E. V. Krumkalns, R. L. Brindle, F. J. Marshall, M. A. Root, J. Med. Chem. 1963, 6, 760.
[11] K. Masek, J. Seifert, M. Flegel, M. Krojidlo, J. Kolinsky, Methods Find. Exp. Clin. Pharmacol. 1984,
6, 667.
[12] K. N. Masihi, W. Lange, B. Rohde-Schulz, K. Masek, Int. J. Immunother. 1987, 3, 89.
ˇ
[13] K. N. Masihi, W. Lange, S. Schwenke, G. Gast, P. Huchshorn, A. Palache, K. Masek, Vaccine 1990, 8,
159.
´
[14] P. D. Becker, R. S. Corral, C. A. Guzman, S. Grinstein, Vaccine 2001, 19, 4603.
ˇ ˇ
[15] S. Gobec, U. Urleb, S. Simcic, B. Wraber, Pharmazie 2001, 56, 523.
ˇ ˇ
ˇ
[16] U. Urleb, M. Krbavcic, M. Sollner, D. Kikelj, S. Pecar, Arch. Pharm. (Weinheim) 1995, 328,
113.

788
CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)
ˇ
[17] C. Ochi, N. Norisada, M. Moriguchi, A. Stalc, U. Urleb, S. Muraoka, Arzneim. Forsch./Drug Res.
1999, 49, 72.
ˇ ˇ
ˇ
[18] S. Gobec, M. Sollner-Dolenc, U. Urleb, B. Wraber, S. Simcic, M. Filipic, Farmaco 2004, 59, 345.
ˇ ´ ˇ ´
[19] B. Vranesic, J. Tomasic, S. Smerdel, D, Kantoci, F. Benedetti, Helv. Chim. Acta 1993, 76, 1752.
ˇ ´
ˇ ´
´
ˇ
[20] B. Vranesic, J. Tomasic, ꢅ. Ljevakovic, I. Hrsak, in ꢄImmunotherapy of Infectionꢀ, Ed. N. Masihi,
Marcel Dekker, New York, Basel, Hong Kong, 1994, 241.
ˇ ´
´
ˇ ´
´
ˇ ´
[21] G. Dasic, S. Rabatic, B. Vranesic, J. Tomasic, Period. Biol. 1996, 98, 319.
ˇ
ˇ
ˇ ´
´
ˇ ´
[22] R. Mazuran, B. Vranesic, M. Ikic-Sutlic, D. Simrak, J. Tomasic, Period. Biol. 1996, 98, 305.
ˇ
ˇ
ˇ ´
´
ˇ ´
ˇ ´
[23] J. Tomasic, I. Hanzl-Dujmovic, B. Spoljar, B. Vranesic, M. Santak, A. Jovicic, Vaccine 2000, 18, 1236.
ˇ ´
[24] L. Habjanec, B. Halassy, J. Tomasic, Int. Immunopharmacol. 2010, 10, 751.
[25] D. T. Oꢀ Hagan, E. De Gregorio, Drug Discovery Today 2009, 13, 541.
[26] A. M. Harandi, D. Medaglini, R. J. Shattock, Vaccine 2010, 28, 2363.
[27] P. H. Seidl, K. H Schleifer, ꢄBiological Properties of Peptidoglycansꢀ, Walter de Gruyter, Berlin,
1986.
[28] F. Ellouz, A. Adam, R. Ciorbaru, E. Lederer, Biochem. Biophys. Res. Commun. 1974, 59, 1317.
[29] G. M. Bahr, L. Chedid, FASEB J. 1986, 45, 2541.
ˇ ´
´
ˇ ´
ˇ ´
[30] B. Vranesic, ꢅ. Ljevakovic, J. Tomasic, B. Ladesic, Clin. Chim. Acta 1991, 202, 23.
ˇ
ˇ
´
ˇ ´
´
[31] ꢅ. Ljevakovic, J. Tomasic, V. Sporec, B. Halassy Spoljar, I. Hanzl-Dujmovic, Bioorg. Med. Chem.
2000, 8, 2441.
´
´
ˇ ´
[32] R. Ribic, L. Habjanec, M. Brgles, S. Tomic, J. Tomasic, Bioorg. Med. Chem. 2009, 17, 6096.
[33] M. A. V. Gianan, E. S. Kleinerman, Cancer Biother. Radiopharm. 1998, 13, 363.
ˇ ´
´
ˇ ´
[34] R. Frkanec, V. Noethig-Laslo, B. Vranesic, K. Mirosavljevic, J. Tomasic, Biochim. Biophys. Acta
2003, 1611, 187.
ˇ
ˇ
´
´
ˇ ´
[35] R. Frkanec, D. Travas, M. Krstanovic, B. Halassy Spoljar, ꢅ. Ljevakovic, B. Vranesic, L. Frkanec, J.
ˇ ´
Tomasic, J. Liposome Res. 2003, 13, 279.
[36] P. Lefrancier, E. Bricas, Bull. Soc. Chem. Biol. 1967, 49, 1257.
[37] G. Han, M. Tamaki, V. J. Hruby, J. Pept. Res. 2001, 58, 338.
[38] F. S. Gibson, S. C. Bergmeier, H. Rapoport, J. Org. Chem. 1994, 59, 3216.
[39] J. E. Sheppeck II, H. Kar, H. Hong, Tetrahedron Lett. 2000, 41, 5329.
´
´
˜
[40] J. Clariana, S. Garcꢆa-Granda, V. Gotor, A. Gutierrez-Fernandez, A. Luna, M. Moreno-Manas, A.
Vallribera, Tetrahedron: Asymmetry 2000, 11, 4549.
[41] A. Baskakova, W. Frey, V. Jꢇger, Synthesis 2010, 21, 3693.
[42] K. Shreder, L. Zhang, M. Goodman, Tetrahedron Lett. 1998, 39, 221.
[43] T. Barman, in ꢄEnzyme Handbookꢀ, Springer-Verlag, Berlin, Heidelberg, New York, 1969, Vol. 1,
p. 176.
[44] H. C. Yip, A. Y. Karulin, M. Tary-Lehman, M. D. Hesse, H. Radeke, P. S. Heeger, J. Immunol. 1999,
162, 3942.
[45] X. P. Ioannou, S. M. Gomis, B. Karvonen, R. Hecker, L. A. Babiuk, S. van Drunen Little-van den
Hurk, Vaccine 2002, 21, 127.
´
ˇ ´
ˇ ´
[46] D. Keglevic, B. Ladesic, J. Tomasic, Z. Valinger, R. Naumski, Biochim. Biophys. Acta 1979, 585, 273.
ˇ
ˇ ´
[47] B. Gaspert, S. Hromadko, B. Vranesic, Croat. Chem. Acta 1976, 48, 169.
ˇ ´
[48] L. Habjanec, R. Frkanec, B. Halassy, J. Tomasic, J. Liposome Res. 2006, 16, 1.
´
ˇ ´
[49] B. Halassy, M. Krstanovic, R. Frkanec, J. Tomasic, Vaccine 2003, 21, 971.
Received April 8, 2011