New 2,3-diaminopropionic acid inhibitors of AGE and ALE formation

Article abstract of DOI:10.1039/c2ob27084f

Novel 2,3-diaminopropionic acid-based molecules were synthesised and tested successfully as glyoxal/methylglyoxal scavengers and as AGE inhibitors. Addition of an 8-hydroxyquinoline moiety led to an increase of the overall activity. The compounds tested in this study were also proved to be efficient in trapping ALE precursor malondialdehyde. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c2ob27084f

Organic &
Biomolecular
Chemistry
View Article Online
View Journal
PAPER
New 2,3-diaminopropionic acid inhibitors of AGE and
ALE formation
Cite this: DOI: 10.1039/c2ob27084f
Nicolas Audic,a Guy Potiera and N. André Sasaki*a,b
Received 25th October 2012,
Accepted 30th November 2012
Novel 2,3-diaminopropionic acid-based molecules were synthesised and tested successfully as glyoxal/
methylglyoxal scavengers and as AGE inhibitors. Addition of an 8-hydroxyquinoline moiety led to an
increase of the overall activity. The compounds tested in this study were also proved to be efficient in
trapping ALE precursor malondialdehyde.
DOI: 10.1039/c2ob27084f
www.rsc.org/obc
potent scavengers of the reactive α-dicarbonyls or inhibitors of
AGE formation. However, among these compounds, only amino-
Introduction
Advanced glycation endproducts (AGE) are cross-linking and guanidine,6 pyridoxamine,7 carnosine8 are reported to be
non-cross-linking modifications of proteins, caused by reactive efficient inhibitors of ALE formation. Therefore, it would be of
α-dicarbonyl species, such as glyoxal (GO), methylglyoxal (MG) great therapeutic interest to develop an agent that might
and 3-deoxyglucosone (3-DG). Under physiological conditions, prevent both AGE and ALE formation.
these dicarbonyl compounds are primarily derived from the
oxidation of Amadori products that are formed by rearrange-
ment of Schiff bases between glucose and amine functions of
proteins.
Results and discussion
Synthesis
A large amount of evidence indicates an important role of
In previous studies,9 we reported dipeptides that contain
N-terminal 2,3-diaminopropionic acid (Dap) as efficient scaven-
gers of MG. Although these peptides (e.g. D-Dap-L-Leu) were
proved to be efficient in trapping MG and GO, they presented
drawbacks such as poor pharmacokinetic properties and low
stability upon prolonged aqueous storage (diketopiperazine
formation). In order to overcome these problems as well as to
improve the α-oxoaldehydes trapping activity, we prepared a
number of new Dap-containing derivatives 2a–j (Scheme 1).
These compounds 2a–j were synthesised in 3–4 steps, either
starting from commercially available D-Dap or using the more
cost-effective pathway involving the Hofmann rearrangement
of Boc-D-Asn.10 The following two steps, namely amide bond
formation and N-Boc cleavage, were respectively carried out
with COMU and hydrochloric acid in diethyl ether to give the
final molecules 2a–j with yields in the range of 80–97% (over
the 2 steps). No chromatographic purification was required
during the syntheses.
Besides, some chemical reactions involved in the degra-
dation of glucose into α-oxoaldehydes (e.g. autoxidation of
glycolaldehyde to GO) are strongly suspected to be catalysed by
redox active metal ions (Cu2+ and Fe3+).11 Therefore, we were
interested to know whether the grafting of a well-chosen metal
ion chelator moiety on the molecules described above could
provide some advantages in terms of anti-AGE activity.
The strong iron chelator and antioxidant 8-hydroxyquinoline
reactive α-dicarbonyl species in the development of diabetes
mellitus (DM) and diabetic complications1 that lead to retino-
pathy, neuropathy, nephropathy, macro- and micro-vascular
injuries, and neurodegenerative diseases, such as Alzheimer’s
disease and Parkinson’s disease.2
Another aspect of protein modification is concerned with
advanced lipid peroxidation endproducts (ALE) formation.
Endogenous peroxidation of polyunsaturated fatty acids
induced by oxidants and oxidative stress generates highly reac-
tive lipid α,β-unsaturated aldehydes, such as acrolein, malon-
dialdehyde (MDA), and 4-hydroxy-2-nonenal (4-HNE),3
together with formaldehyde, acetaldehyde, GO and MG. These
reactive aldehydes also react with cellular and tissue proteins
to form adducts and cross-links on proteins. Accumulating
studies on the causes of neurodegeneration indicate that
the ALE precursors, such as acrolein, MDA and 4-HNE, are
involved in the development of Alzheimer’s disease.4 It is
also to be noted that DNA damage is induced by the ALE
precursors.5
In view of the totally deleterious roles of these reactive car-
bonyl compounds, a number of agents have been developed as
aPharmamens, 9 avenue Théophile Gautier, 75016 Paris, France
bLaboratoire des Glucides, CNRS FRE 3517, UFR de Pharmacie, Université de
Picardie Jules Verne, 1 rue des Louvels, 80037 Amiens Cedex 1, France.
E-mail: andre.sasaki@u-picardie.fr; Tel: +33 (0)3 22 82 74 98
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
(8-HQ)12 was selected for this purpose and 8-HQ containing
Dap derivatives 8 and 10 were synthesized (Scheme 2).
In the absence of any tested compound, concentration of
GO and MG increased linearly in the course of incubation, as
Reaction between 5-chloromethyl-8-hydroxyquinoline 313 already reported.11 Addition, on the 7th day of incubation,
and the piperazine derivative 6 afforded, after acidic treatment, of a single dose of a compound 2a–j (final concentration of
the expected molecule 8. On the other hand, peptide coupling 100 μM) caused the GO and MG levels to drop rapidly.
of the same compound 6 with carboxylic acid 5, followed by
Among the screened molecules, Dap-derivatives 2e and 2j
N-Boc deprotection, allowed us to obtain the molecule 10.
showed the best results. On the other hand, 8-HQ-based
GO and MG trapping activity
Activity of the newly synthesised compounds was assessed by
measuring their effect upon the levels of GO and MG gener-
ated in situ in glucose solutions (50 mM in 100 mM phosphate
buffer solution, pH 7.4) incubated at 37 °C, in the dark, during
14 days. Aminoguanidine (AG), as well as D-Dap-L-Leu and
8-HQ, were used as reference compounds. A selection of repre-
sentative results is presented (Fig. 1).
Fig. 1 Evolution of GO and MG concentrations in glucose solutions after
Scheme 1 Synthesis of compounds 2a–j.
addition of a tested compound on the 7th day.
Scheme 2 Synthesis of 8-hydroxyquinoline derivatives 8 and 10.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Organic & Biomolecular Chemistry
Paper
compounds 8 and 10 presented even more remarkable results. MDA trapping activity
Indeed, in addition to be efficient GO and MG scavengers, they
Nucleophilic functional groups of lysine, histidine and
cysteine residues of proteins react with α,β-unsaturated alde-
hydes derived from lipid peroxidation leading to the formation
of Schiff bases and Michael addition products.14 We assumed
that the two vicinal amine groups of Dap derivatives could also
undergo nucleophilic attacks on α,β-unsaturated aldehydes
either by forming the Schiff base or the 1,4-Michael addition
product or, by the combination of both mechanisms providing
1,4-diazepine derivatives. However, investigation on the poten-
tial of an alkyl vicinal diamine as Dap derivatives for scaven-
ging α,β-unsaturated aldehydes has not been reported, to the
best of our knowledge.
We have thus investigated the potential of the Dap deriva-
tives to irreversibly trap the α,β-unsaturated aldehydes. To our
delight, compound 2a reacted rapidly with MDA (10 mM in
water, 37 °C), to form a single major adduct after 24 h. Accord-
ing to LC-MS and NMR analysis, we propose the structure of
this adduct as 2,3-dihydro-1H-1,4-diazepinium 11a as shown
in Scheme 3. Molecules 2b–j, 8 and 10 were observed to react
in the same fashion.
showed themselves able to strongly inhibit the production of
these two α-oxoaldehydes.
When observing the activity of parent molecule 2e and
8-HQ tested separately, this dual behaviour clearly appeared to
be the result of the combined properties of the vicinal diamine
fragment and the 8-HQ moiety.
Inhibition of AGE formation
Inhibitory activity of the new products was further investigated
by testing their ability to attenuate the formation of AGE in
human serum albumin (HSA) incubated with glucose. Devel-
opment of specific fluorescence, at 440 nm after excitation at
370 nm, of HSA solutions (50 g L−1 in 100 mM PBS, pH 7.4)
was examined after incubation with glucose (500 mM) at 37 °C
for 20 days in the presence or absence of a test compound
(50 mM). The results are expressed in Fig. 2 in percentage with
respect to the control (20 days in the absence of a scavenger).
After 20 days, the control sample showed an expected
increase (about 7-fold) of fluorescence (relative fluorescence =
100%). According to the present study, AG is a highly efficient
anti-AGE agent (3% relative fluorescence) compared with car-
nosine (60%) and D-Dap-L-Leu (43%). Compounds 2a–j likewise
inhibited AGE formation (relative fluorescence 11–23%).
A large difference in relative fluorescence between compound
2a (D-Dap-morpholine) and analogue D-Ala-morpholine (pos-
sessing only one amine group) unambiguously indicates the
1,2-diamine structure as essential for the anti-AGE activity of
these molecules. Yet, once again, 8-HQ derivatives 8 and 10
(relative fluorescence of 3% and 2%, respectively) were found
to be the most efficient molecules among the tested Dap con-
taining compounds and shown to be as powerful as reference
molecule aminoguanidine.
Among the reactive α,β-unsaturated aldehydes, MDA may be
the most frequently studied.15 Due to its high nucleophilicity
towards MDA, N-acetyl lysine is the model compound exten-
sively used in the studies on the protein modifications by ALE
precursors, such as MDA.16
Scheme 3 Trapping of MDA by Dap derivative 2a.
Fig. 2 Inhibition of AGE formation by measurement of specific glucose-mediated fluorescence of HSA.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 3 Trapping of MDA by N-Ac-Lys versus tested compounds with (a) control, (b) compound 2e, (c) hydralazine and (d) carnosine.
When N-Ac-Lys was incubated, for 2 h at 37 °C, with MDA ALE formation. Results of current in vivo testing of these new
(each 10 mM) in phosphate buffer solution (100 mM, pH 7.4), molecules against diabetic nephropathy will be reported in
Nα-acetyl-Nε-(3-propenal)-Lys was detected as the major MDA due course.
adduct (Fig. 3a).
In order to determine if Dap derivatives 2, 8 and 10 could
prevent or limit this kind of modification, they were co-incu-
bated along with N-Ac-Lys and MDA under the same con-
Experimental
Instruments and materials
ditions and the resulting mixtures were analysed by LC-MS.
All starting materials and solvents were commercially available
The same experiments were conducted with other known sca-
vengers17 of ALE precursors such as aminoguanidine, carno-
(reagent grade) and were used as purchased from Sigma-
1
Aldrich. H and 13C NMR were recorded on a Bruker DPX-300
sine, pyridoxamine, taurine and hydralazine to establish a
comparison of activity.
spectrometer. Chemical shifts (δ) are given in parts per million
(ppm) relative to the residual solvent peak. Coupling constants
Compounds 2, 8 and 10 were all found to efficiently protect
(J) are given in Hertz. The abbreviations s, d, t, m, dd stand for
N-Ac-Lys against MDA. Indeed, formation of Nα-acetyl-Nε-
singlet, doublet, triplet, multiplet, doublet of doublets,
(3-propenal)-Lys was almost negligible compared with the
respectively. High-resolution mass spectra were obtained on a
Micromass-Waters Q-TOF Ultima spectrometer operating in
positive electron ionization mode. LC-MS analyses were carried
out using an integrated Shimadzu LCMS-2020 system
control experiment, as seen in Fig. 3b in the representative
case of compound 2e. The Dap derivatives compare well with
previously reported MDA scavengers: only aminoguanidine and
hydralazine (Fig. 3c) showed the same high level of efficiency
equipped with an autosampler, a temperature-controlled com-
whereas pyridoxamine and especially taurine and carnosine
partment for column and a single quadrupole mass spec-
(Fig. 3d) were found to have a low activity in this test.
trometer. Preparation of experiments involving incubation of
glucose solutions was carried out using sterile techniques
(autoclaved glassware, 0.22 μm filter sterilized solutions, hand-
ling under vertical laminar flow hood).
Conclusions
The new Dap derivatives are proved to be excellent scavengers
of α-oxoaldehydes GO and MG as well as α,β-unsaturated alde-
Synthetic procedures
hyde MDA. In particular, molecules 8 and 10 are among the
General procedure for the synthesis of compounds 2. To a
best compounds able to inhibit AGE formation. It is also to be stirred solution of Boc-D-Dap(Boc)-OH (12.2 g, 40 mmol) in
noted that compounds 2a–j and 10 show good plasma stability CH2Cl2 (400 mL) was added the desired mono-substituted
and good stability upon prolonged aqueous storage (less than piperazine (40 mmol) followed by COMU (16.8 g, 40 mmol).
1% degradation on standing for 28 days at 37 °C, at 5 wt% The heterogeneous reaction mixture was stirred at 20 °C for
in water). 12 h during which it progressively became homogeneous. The
These results indicate that the new Dap derivatives are resulting solution was then washed with a saturated Na2CO3
therapeutically interesting molecules to attenuate AGE and solution (4
×
80 mL), dried over Na2SO4, filtered and
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Organic & Biomolecular Chemistry
Paper
concentrated under vacuum to give a viscous residue. This 209 (72%, M + Na), 395 (25%, 2M + Na). HRMS (ESI+): calcu-
residue was dissolved in Et2O (100 mL) and 4 M HCl in lated for C8H19N4O (M + H)+ 187.1559, found 187.1568.
dioxane (100 mL, 400 mmol) was added dropwise, under very
vigorous stirring. After 18 h, the heterogeneous off-white
mixture was diluted with Et2O (400 mL), stirred vigorously for
30 min and sonicated for 30 min in order to obtain a fine sus-
pension. The mixture was then filtered on a sintered glass
funnel and the residual solid was washed thoroughly with
Et2O (4 × 200 mL) before being dried under vacuum. Expected
compounds were obtained as solids (80–97%).
(2R)-2,3-Diamino-1-(4-(3-hydroxypropyl)piperazin-1-yl)propan-
1-one trihydrochloride (2g). 1H NMR (D2O, 300 MHz): δ (ppm)
5.00 (m, 1H), 4.26 (m, 1H), 3.87–3.68 (m, 5H), 3.67–3.51
(m, 4H), 3.46–3.12 (m, 4H), 2.02 (m, 2H). 13C NMR (D2O,
75 MHz): δ (ppm) 164.0, 58.0, 54.2, 50.5, 47.6, 42.1, 39.2, 38.4,
25.5. MS (ESI+): m/z 231 (65%, M + H), 253 (100%, M + Na),
294 (33%, M + MeCN + Na). HRMS (ESI+): calculated for
C10H23N4O2 (M + H)+ 231.1821, found 231.1808.
(2R)-2,3-Diamino-1-(4-cyclohexylpiperazin-1-yl)propan-1-one
trihydrochloride (2h). 1H NMR (D2O, 300 MHz): δ (ppm) 4.97
(m, 1H), 4.68 (m, 1H), 4.28 (m, 1H), 3.86–3.51 (m, 5H),
3.46–3.16 (m, 4H), 2.11 (m, 2H), 1.92 (m, 2H), 1.68 (m, 1H),
1.58–1.08 (m, 5H). 13C NMR (D2O, 75 MHz): δ (ppm) 163.9,
65.9, 65.7, 47.6, 47.2, 46.8, 42.4, 39.5, 38.7, 38.3, 26.1, 24.0.
MS (ESI+): m/z 255 (100%, M + H), 277 (75%, M + Na). HRMS
(ESI+): calculated for C13H27N4O (M + H)+ 255.2185, found
255.2184.
(2R)-2,3-Diamino-1-(4-phenylpiperazin-1-yl)propan-1-one trihy-
drochloride (2i). 1H NMR (D2O, 300 MHz): δ (ppm) 7.60
(m, 5H), 5.07 (dd, J = 3.8, 6.2 Hz, 1H), 4.32–4.22 (m, 1H),
4.20–4.07 (m, 2H), 4.04–3.93 (m, 1H), 3.91–3.78 (m, 4H),
3.73–3.57 (m, 2H). 13C NMR (D2O, 75 MHz): δ (ppm) 164.1,
140.7, 130.1, 129.5, 120.0, 53.3, 53.2, 47.6, 42.9, 40.0, 38.5. MS
(ESI+): m/z 249 (100%, M + H), 271 (71%, M + Na). HRMS
(ESI+): calculated for C13H21N4O (M + H)+ 249.1715, found
249.1719.
(2R)-2,3-Diamino-1-morpholinopropan-1-one dihydrochloride
(2a). 1H NMR (D2O, 300 MHz): δ (ppm) 4.95 (m, 1H),
3.92–3.46 (m, 10H). 13C NMR (D2O, 75 MHz): δ (ppm) 164.3,
66.0, 65.9, 48.0, 46.0, 43.0, 39.0. MS (ESI+): m/z 174 (100%,
M + H), 215 (61%, M + MeCN + H), 237 (31%, M + MeCN + Na).
HRMS (ESI+): calculated for C7H16N3O2 (M + H)+ 174.1243,
found 174.1259.
(2R)-2,3-Diamino-1-thiomorpholinopropan-1-one dihydrochlor-
ide (2b). 1H NMR (D2O, 300 MHz): δ (ppm) 4.96 (dd, J = 4.1,
6.1 Hz, 1H), 4.21–4.09 (m, 1H), 4.05–3.94 (m, 1H), 3.89–3.80
(m, 1H), 3.72–3.61 (m, 1H), 3.57 (m, 2H), 2.96–2.66 (m, 4H).
13C NMR (D2O, 75 MHz): δ (ppm) 163.7, 48.1, 47.7, 45.2, 38.5,
26.6, 25.9. MS (ESI+): m/z 190 (100%, M + H), 212 (59%,
M + Na). HRMS (ESI+): calculated for C7H16N3OS (M + H)+
190.1014, found 190.1023.
(2R)-2,3-Diamino-1-(piperidin-1-yl)propan-1-one dihydrochlor-
ide (2c). 1H NMR (D2O, 300 MHz): δ (ppm) 4.93 (m, 1H),
3.84–3.34 (m, 6H), 1.68 (s, 6H). 13C NMR (D2O, 75 MHz):
(2R,2′R)-1,1′-(Piperazine-1,4-diyl)bis(2,3-diaminopropan-1-one)
tetrahydrochloride (2j). Compound 2j was prepared using the
general synthetic procedure with 0.5 equiv. of piperazine
δ
(ppm) 163.5, 48.2, 47.2, 44.4, 39.2, 25.9, 24.9, 23.4.
MS (ESI+): m/z 172 (100%, M + H), 213 (56%, M + MeCN + H),
235 (32%, M + MeCN + Na). HRMS (ESI+): calculated for
C8H18N3O (M + H)+ 172.1450, found 172.1465.
1
instead of 1.0 equiv. of mono-substituted piperazine. H NMR
(D2O, 300 MHz): δ (ppm) 4.99 (m, 2H), 3.98 (m, 2H), 3.88–3.44
(m, 10H). 13C NMR (D2O, 75 MHz): δ (ppm) 164.2, 164.0, 47.7,
47.6, 44.5, 43.9, 41.9, 41.4, 38.5, 38.4.
(2R)-2,3-Diamino-1-(4-methylpiperidin-1-yl)propan-1-one dihydro-
chloride (2d). 1H NMR (D2O, 300 MHz): δ (ppm) 4.95 (m, 1H),
4.32 (m, 1H), 3.87 (m, 1H), 3.55 (m, 2H), 3.29 (m, 1H), 2.85 (m,
1H), 1.79 (m, 3H), 1.19 (m, 2H), 0.96 (m, 3H). 13C NMR (D2O,
75 MHz): δ (ppm) 163.1, 162.9, 47.7, 46.1, 45.7, 43.4, 42.9,
38.8, 38.5, 33.4, 33.1, 32.5, 32.1, 29.6, 29.2, 20.3, 19.9. MS
(ESI+): m/z 186 (100%, M + H), 208 (39%, M + Na). HRMS
(ESI+): calculated for C9H20N3O (M + H)+ 186.1606, found
186.1619.
5-(Chloromethyl)quinolin-8-ol hydrochloride (3). To a vigor-
ously stirred suspension of 8-hydroxyquinoline (2.9 g, 20 mmol)
in formaldehyde (4 mL) was added dropwise (CAUTION:
exothermic reaction) concentrated hydrochloric acid (10 mL).
A stream of hydrogen chloride gas was bubbled for 1 h into
the yellow solution which was stirred for 5 h. The reaction
mixture was then filtered on a sintered glass funnel and the
residual solid was washed with concentrated hydrochloric
acid (4 × 5 mL) and dried under vacuum. Compound 3 was
obtained as a bright yellow hygroscopic solid (3.15 g, 68%)
which was immediately stored at 4 °C under an Ar atmosphere.
1H NMR (DMSO-d6, 300 MHz): δ (ppm) 9.24 (dd, J = 1.2,
8.7 Hz, 1H), 9.13 (dd, J = 1.2, 5.2 Hz, 1H), 8.13 (dd, J = 5.2,
8.7 Hz, 1H), 7.87 (d, J = 8.1 Hz, 1H), 7.54 (d, J = 8.1 Hz, 1H),
5.32 (s, 2H). 13C NMR (DMSO-d6, 75 MHz): δ (ppm) 149.5,
144.4, 142.9, 132.3, 129.6, 127.8, 124.6, 122.8, 115.2, 43.1.
2-(8-Hydroxyquinolin-5-yl)acetonitrile (4). To a stirred solu-
(2R)-2,3-Diamino-1-(piperazin-1-yl)propan-1-one trihydrochloride
(2e). Compound 2e was prepared using the general synthetic
procedure with N-Boc-piperazine as mono-substituted pipera-
zine. 1H NMR (D2O, 300 MHz): δ (ppm) 5.00 (dd, J = 3.9,
6.3 Hz, 1H), 4.17–3.28 (m, 10H). 13C NMR (D2O, 75 MHz):
δ (ppm) 164.7, 48.1, 42.8, 42.6, 42.5, 39.5, 39.0. MS (ESI+):
m/z 173 (100%, M + H). HRMS (ESI+): calculated for C7H17N4O
(M + H)+ 173.1403, found 173.1416.
(2R)-2,3-Diamino-1-(4-methylpiperazin-1-yl)propan-1-one trihy-
drochloride (2f). 1H NMR (D2O, 300 MHz): δ (ppm) 5.01
(m, 1H), 4.62 (m, 1H), 4.27 (m, 1H), 3.85–3.53 (m, 5H), tion of potassium cyanide (1.98 g, 30 mmol) in anhydrous
3.41–3.14 (m, 3H), 3.00 (m, 3H). 13C NMR (D2O, 75 MHz): DMF (30 mL), under an argon atmosphere, were added
δ (ppm) 164.0, 52.0, 51.9, 51.7, 47.7, 47.5, 42.6, 42.4, 42.3, activated 4 Å molecular sieve (6 g) and compound 3 (1.38 g,
42.2, 39.4, 39.2, 38.7, 38.2. MS (ESI+): m/z 187 (100%, M + H), 6 mmol). After 19 h stirring at 20 °C, the greenish reaction
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
mixture was filtered and the filtrate was concentrated under the following conditions; column: Chromabond Flash RS15
vacuum. The residue was diluted with water (20 mL) and the C18ec, flow rate: 15 mL min−1, gradient elution: 10–25% B
resulting solution was carefully (under a fume hood) neutral- (0–40 min), 25–80% B (40–45 min) then 80% B (45–50 min)
ized with 1 M HCl and extracted with CH2Cl2 (4 × 60 mL). The with A: water and B: acetonitrile, detection: UV at 200 nm. The
combined organic layers were washed with brine (20 mL), fractions of interest (retention time: 41 min) were lyophilized
1
dried over Na2SO4, filtered and concentrated under vacuum to to give 7 as a greenish solid (390 mg, 74%). H NMR (CDCl3,
give a waxy black solid. After careful trituration in Et2O, fil- 300 MHz): δ (ppm) 8.79 (dd, J = 1.5, 4.2 Hz, 1H), 8.63 (dd, J =
tration and drying in vacuo, the compound 4 was obtained as a 1.5, 8.6 Hz, 1H), 7.46 (dd, J = 4.2, 8.6 Hz, 1H), 7.31 (d, J =
light brown solid (974 mg, 88%). 1H NMR (DMSO-d6, 7.8 Hz, 1H), 7.07 (d, J = 7.8 Hz, 1H), 5.56 (d, J = 7.7 Hz, 1H),
300 MHz): δ (ppm) 9.97 (s, 1H), 8.92 (dd, J = 1.5, 4.1 Hz, 1H), 5.01 (m, 1H), 4.72 (m, 1H), 3.81 (s, 2H), 3.68–3.46 (m, 4H), 3.41
8.45 (dd, J = 1.5, 8.6 Hz, 1H), 7.67 (dd, J = 4.1, 8.6 Hz, 1H), 7.51 (m, 1H), 3.18 (m, 1H), 2.60–2.34 (m, 4H), 1.42 (s, 18H). 13C
(d, J = 7.9 Hz, 1H), 7.09 (d, J = 7.9 Hz, 1H), 4.37 (s, 2H). NMR (CDCl3, 75 MHz): δ (ppm) 169.6, 156.0, 155.5, 152.0,
13C NMR (DMSO-d6, 75 MHz): δ (ppm) 153.4, 148.2, 138.7, 147.6, 138.8, 133.9, 129.1, 126.0, 123.7, 121.5, 108.6, 80.0, 79.2,
132.2, 127.9, 126.4, 122.1, 119.1, 117.0, 110.6, 19.3. MS (ESI+): 60.5, 52.9, 52.4, 50.3, 45.6, 43.3, 42.2, 28.3. MS (ESI+): m/z 530
m/z 185 (100%, M + H).
(100%, M + H), 552 (26%, M + Na).
2-(8-Hydroxyquinolin-5-yl)acetic acid hydrochloride (5). A
stirred suspension of compound 4 (184 mg, 1 mmol) in con-
centrated hydrochloric acid (5 mL) was refluxed for 3 h. The
reaction mixture was then concentrated under high vacuum
yielding compound 5 as a pale yellow solid (240 mg, 100%).
1H NMR (DMSO-d6, 300 MHz): δ (ppm) 9.11 (m, 2H), 8.07
(dd, J = 5.2, 8.6 Hz, 1H), 7.66 (s, 1H), 7.65 (d, J = 7.9 Hz, 1H),
7.56 (d, J = 7.9 Hz, 1H), 7.49 (s, 1H), 7.32 (s, 1H), 4.12 (s, 2H).
13C NMR (DMSO-d6, 75 MHz): δ (ppm) 172.2, 147.9, 144.0,
143.3, 131.6, 129.6, 128.7, 123.0, 122.0, 115.3, 36.8. MS (ESI+):
m/z 204 (M + H, 100%).
(2R)-Di-tert-butyl(3-oxo-3-(piperazin-1-yl)propane-1,2-diyl)-
dicarbamate (6). To a stirred suspension of Boc-D-Dap(Boc)-OH
(1.82 g, 6 mmol) and HOBt·H2O (1.01 g, 6.6 mmol) in aceto-
nitrile (60 mL) at 0 °C was added EDCI (1.27 g, 6.6 mmol).
After 15 min, the resulting solution was added dropwise (1 mL
min−1) to a vigorously stirred solution of piperazine (5.17 g,
60 mmol) in acetonitrile (540 mL) at 0 °C. After the addition,
the reaction mixture was stirred for 15 h and allowed to warm
to 20 °C. The obtained suspension was filtered on a sintered
glass funnel and the filtrate was concentrated under vacuum
to give a solid residue which was purified by reverse phase
chromatography, in 3 batches, using the following condi-
tions; column: Chromabond Flash RS40 C18ec, flow rate:
40 mL min−1, gradient elution: 10% B (0–5 min) then 10–80%
B (5–35 min) with A: water and B: acetonitrile, detection: UV at
200 nm. The fractions of interest (retention time: 15 min) were
lyophilized to give 6 as a white solid (1.68 g, 75%). 1H NMR
(CDCl3, 300 MHz): δ (ppm) 5.66 (s, 1H), 5.15 (s, 1H), 4.70
(s, 1H), 3.60–3.44 (m, 4H), 3.34 (m, 1H), 3.18 (m, 1H),
2.90–2.76 (m, 4H), 1.40 (s, 18H). 13C NMR (CDCl3, 75 MHz):
δ (ppm) 168.4, 155.9, 155.4, 79.8, 79.4, 50.0, 46.7, 46.1, 45.6,
43.2, 43.1, 28.2. MS (ESI+): m/z 373 (100%, M + H), 395
(75%, M + Na).
(2R)-2,3-Diamino-1-(4-((8-hydroxyquinolin-5-yl)methyl)pipera-
zin-1-yl)propan-1-one tetrahydrochloride (8). To a stirred solu-
tion of compound 7 (390 mg, 0.74 mmol) in Et2O (2 mL) was
added 4 M HCl in dioxane (2 mL, 8 mmol). The mixture was
stirred for 12 h before being filtered. The recovered yellow
solid was washed thoroughly with Et2O (5 × 4 mL) before being
dried under vacuum to yield expected compound 8 (270 mg,
77%). 1H NMR (D2O, 300 MHz): δ (ppm) 9.39 (d, J = 8.8 Hz,
1H), 9.13 (d, J = 5.3 Hz, 1H), 8.22 (dd, J = 5.3, 8.8 Hz, 1H), 8.04
(d, J = 8.1 Hz, 1H), 7.56 (d, J = 8.1 Hz, 1H), 4.99 (m, 3H),
4.18–3.78 (m, 10H). 13C NMR (D2O, 75 MHz): δ (ppm) 164.7,
149.9, 143.4, 142.8, 142.7, 136.6, 129.4, 123.0, 115.8, 115.6,
55.7, 50.9, 50.7, 48.1, 42.6, 39.7, 38.9. MS (ESI+): m/z 330
(100%, M + H), 371 (16%, M + MeCN + H).
(2R)-Di-tert-butyl(3-(4-(2-(8-hydroxyquinolin-5-yl)acetyl)pipera-
zin-1-yl)-3-oxopropane-1,2-diyl)dicarbamate (9). To a stirred sus-
pension of compound 5 (240 mg, 1 mmol) in acetonitrile
(20 mL) were successively added Et3N (404 μL, 3.6 mmol), com-
pound 6 (372 mg, 1 mmol), EDCI (230 mg, 1.2 mmol) and
HOBt·H2O (184 mg, 1.2 mmol). After 24 h of stirring, the
mixture was filtered and concentrated under vacuum. The
residue was purified by reverse phase chromatography using
the following conditions; column: Chromabond Flash RS40
C18ec, flow rate: 40 mL min−1, gradient elution: 10% B
(0–5 min) then 10–50% B (5–35 min) with A: water and B:
acetonitrile, detection: UV at 200 nm. The fractions of interest
(retention time: 30 min) were lyophilized to give 9 as a brown-
ish solid (315 mg, 56%). 1H NMR (DMSO-d6, 300 MHz):
δ (ppm) 8.79 (d, J = 4.1 Hz, 1H), 8.38 (d, J = 8.5 Hz, 1H), 7.48
(dd, J = 4.1, 8.5 Hz, 1H), 7.27 (d, J = 7.8 Hz, 1H), 7.09 (d, J =
7.8 Hz, 1H), 5.57 (d, J = 7.3 Hz, 1H), 5.04 (m, 1H), 4.69 (m, 1H),
4.04 (m, 2H), 3.56 (m, 8H), 3.32 (m, 2H), 1.39 (m, 18H).
13C NMR (DMSO-d6, 75 MHz): δ (ppm) 169.5, 168.6, 155.9,
155.3, 151.7, 147.7, 138.6, 132.9, 128.2, 127.4, 121.9, 121.1,
109.2, 79.9, 79.5, 50.4, 45.9, 45.3, 42.9, 42.0, 41.3, 37.5, 28.3.
(2R)-Di-tert-butyl(3-(4-((8-hydroxyquinolin-5-yl)methyl)pipera-
zin-1-yl)-3-oxopropane-1,2-diyl)dicarbamate (7). To a stirred sus- MS (ESI+): m/z 558 (100%, M + H), 580 (25%, M + Na).
pension of compound 3 (230 mg, 1 mmol) in acetonitrile
(2R)-2,3-Diamino-1-(4-(2-(8-hydroxyquinolin-5-yl)acetyl) piper-
(10 mL) were successively added K2CO3 (276 mg, 2 mmol) and azin-1-yl)propan-1-one (10). To a stirred solution of compound 8
compound 6 (373 mg, 1 mmol). After 24 h of vigorous stirring, (315 mg, 0.56 mmol) in Et2O (2.8 mL) was added 4 M HCl in
the mixture was filtered and concentrated under vacuum. The dioxane (2.8 mL, 11.2 mmol). The mixture was stirred for 24 h
residue was purified by reverse phase chromatography using before being concentrated under vacuum. The recovered
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Organic & Biomolecular Chemistry
Paper
residue was dissolved in water (10 mL) and the resulting solu- interface voltage: 4.5 kV, DL voltage: 10 V, Q-array DC: 0 V, Q-
tion was washed with Et2O (3 × 10 mL), filtered through a array RF: 40 V (for m/z = 181) or 10 V (for m/z = 195). GO–DAN
0.45 μm membrane and lyophilized. Compound 10 was and MGO–DAN were respectively detected at 3.9 min and
obtained as a yellowish solid (240 mg, 92%). 1H NMR (D2O, 5.4 min.
300 MHz): δ (ppm) 8.98 (m, 2H), 8.04 (dd, J = 5.9, 8.1 Hz, 1H),
7.55 (d, J = 7.5 Hz, 1H), 7.35 (d, J = 7.5 Hz, 1H), 5.01 (m, 1H),
AGE assay
4.36 (s, 2H), 4.10–3.50 (m, 10H). 13C NMR (D2O, 75 MHz)
δ (ppm): 171.1, 164.1, 145.9, 143.4, 141.9, 131.6, 128.8, 128.7,
123.1, 121.3, 115.5, 47.7, 44.7, 44.6, 44.4, 44.1, 42.1, 41.9, 41.2,
40.7, 38.5, 35.1. MS (ESI+): m/z 358 (100%, M + H), 380 (34%,
M + Na). HRMS (ESI+): calculated for C18H24N5O3 (M + H)+
358.1879, found 358.1877.
A solution containing 50 g L−1 of human serum albumin
(HSA) and 500 mM of glucose in 100 mM PBS (pH 7.4) was
prepared. Tubes were charged with the test compounds
(50 μmol) and the previous solution (1000 μL). After homogen-
isation, the samples were filtered through a 0.22 μm filter into
sterile capped vials which were incubated at 37 °C, in the dark,
for 20 days. The reaction mixtures were then cooled to 20 °C
and the intensity of fluorescence was measured on a 200 μL
aliquot of each sample, at an excitation wavelength of 370 nm
and an emission wavelength of 440 nm.
(2,3-Dihydro-1H-1,4-diazepin-2-yl)(morpholino)methanone
hydrochloride (11a). A solution of 1,1,3,3-tetraethoxypropane
(240 μL, 1 mmol) in 0.1 M HCl (10 mL) was stirred for 2 h at
20 °C, then a solution of compound 2a (123 mg, 0.5 mmol) in
water (40 mL) was added. The mixture was incubated at 37 °C
for 24 h before being lyophilized. The solid residue was puri-
fied by reverse phase chromatography on a Chromabond Flash
RS40 C18ec column, with isocratic elution (100% water) at a
flow rate of 40 mL min−1. The fractions of interest (retention
time: 2.2 min) were lyophilized to give 11a as an off-white solid
MDA trapping experiments
MDA solution (100 mM) was freshly prepared by hydrolysing
1,1,3,3-tetraethoxypropane (100 mM in 0.1 M HCl) at 20 °C, for
2 h, under vigorous stirring. Screw-capped test tubes were
charged with test compounds (50 μmol), 100 mM N-Ac-Lys
solution (500 μL), 100 mM MDA solution (500 μL) and 100 mM
phosphate buffer solution (4.0 mL). The samples were incu-
bated for 2 h at 37 °C before being submitted to LC-MS analy-
sis. Analyses were conducted at 30 °C on a Waters Atlantic
HILIC Silica 5 μm column (250 × 4.6 mm, 100 Å) with gradient
elution (0–3 min: 50% B, 3–13 min: 50–0% B, 13–20 min:
0% B, with A: water/acetonitrile 50/50 + 25 mM formic acid +
25 mM ammonium formate and B: water/acetonitrile 10/90 +
25 mM formic acid + 25 mM ammonium formate) at a flow
rate of 1 mL min−1. Injection volume was 2 μL. Following mass
spectrometric conditions were applied; ionization interface:
ESI+, detection mode: SIM (m/z = M + H), interface voltage:
4.5 kV, DL voltage: 10 V, Q-array DC: 0 V, Q-array RF: 40 V.
N-Ac-Lys and Nα-acetyl-Nε-(3-propenal)-Lys were respectively
detected at 8.5 min and 5.6 min.
1
(89 mg, 72%). H NMR (D2O, 300 MHz): δ (ppm) 7.79 (d, J =
7.7 Hz, 1H), 7.75 (d, J = 7.7 Hz, 1H), 5.36 (t, J = 7.7 Hz, 1H),
4.97 (dd, J = 3.7, 6.6 Hz, 1H), 3.89–3.58 (m, 10H). 13C NMR
(D2O, 75 MHz) δ (ppm): 165.3, 158.3, 156.1, 89.1, 65.7, 65.5,
45.6, 42.4, 38.5. MS (ESI+): m/z 210 (100%, M + H).
GO and MG assay
Eppendorf micro-test tubes with locking caps were charged (14 ×
1000 μL) with a glucose solution (50 mM in 100 mM phos-
phate buffer solution, pH 7.4), closed and incubated at 37 °C,
in the dark. Every 24 h, until the 14th day, a tube was removed
from the temperature controlled oven and immediately stored
at −20 °C (control experiment). Another part of the same
glucose solution was incubated at the same time, under the
same conditions, during 7 days. On the 7th day, batches of
this glucose solution (6860 μL) were treated with 5 mM sol-
utions of the test compounds (140 μL). Each mixture was then
split into Eppendorf test tubes (6 × 1000 μL) which were closed
and incubated at 37 °C, in the dark. Every 24 h, a tube from
each series was removed from the oven and stored at −20 °C
(test experiments).
On the 14th day, samples from the control and the test
experiments were defrosted and treated each with a 10 mM sol-
ution of 2,3-diaminonaphthalene (DAN) in acetonitrile
(100 μL) in order to derivatize GO and MG.18 After homogenis-
ation, the samples were left for 24 h, at 20 °C, in the dark and
filtered through a 0.22 μm filter. GO–DAN and MG–DAN levels
were then directly measured by LC-MS using an external cali-
bration method. Analyses were conducted at 40 °C on a Shim-
pack XR-ODS II UFLC column (75 × 2 mm, 80 Å) with isocratic
elution (50% water, 50% methanol, 0.1% formic acid) at a flow
rate of 300 μL min−1. Injection volume was 1 μL. Follow-
ing mass spectrometric conditions were applied; ionization
interface: ESI+, detection mode: SIM (m/z = 181 and 195),
References
1 N. Ahmed, Diabetes Res. Clin. Pract., 2005, 67, 3–21.
2 V. Srikanth, A. Maczurek, T. Phan, M. Steele, B. Westcott,
D. Juskiw and G. Münch, Neurobiol. Aging, 2011, 32,
763–777.
3 H. Esterbauer, R. J. Schauer and H. Zollner, Free Radical
Biol. Med., 1991, 11, 81–128.
4 (a) M. Singh, D. T. Nam, M. Arseneault and C. Ramassamy,
J. Alzheimer’s Dis., 2010, 21, 741–756; (b) D. A. Butterfield,
L. Bader, L. Miranda and R. Sultana, Biochim. Biophys. Acta,
Mol. Cell Biol. Lipids, 2010, 1801, 924–929.
5 I. G. Minko, I. D. Kozekov, T. M. Harris, C. J. Rizzo,
R. S. Lloyd and M. P. Stone, Chem. Res. Toxicol., 2009, 22,
759–778.
6 Y. Al-Abed and R. Bucala, Chem. Res. Toxicol., 1997, 10,
875–879.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
7 J. M. Onorato, A. J. Jenkins, S. R. Thorpe and J. W. Baynes, 14 (a) K. Uchida, M. Kanematsu, Y. Morimitsu, T. Osawa,
J. Biol. Chem., 2000, 275, 21177–21184.
N. Noguchi and E. Niki, J. Biol. Chem., 1998, 273,
16058–16066; (b) H. Esterbauer, Am. J. Clin. Nutr., 1993, 57,
779S–785S; (c) R. J. Schauer, Mol. Aspects Med., 2003, 24,
149–159.
8 A. R. Hipkiss, Exp. Gerontol., 2006, 41, 464–473.
9 N. A. Sasaki, M. C. Garcia-Alvarez, Q. Wang, L. Ermolenko,
G. Franck, N. Nhiri, M.-T. Martin, N. Audic and P. Potier,
Bioorg. Med. Chem., 2009, 17, 2310–2320.
10 L. Zhang, G. S. Kauffman, J. A. Pesti and J. Yin, J. Org.
Chem., 1997, 62, 6918–6920.
11 P. J. Thornalley, A. Langborg and H. S. Minhas, Biochem. J.,
1999, 344, 109–116.
12 R. Kayyali, A. S. Pannala, H. Khodr and R. C. Hider,
Biochem. Pharmacol., 1998, 55, 1327–1332.
15 D. Del Rio, A. J. Stewart and N. Pellegrini, Nutr. Metab.
Cardiovasc. Dis., 2005, 15, 316–328.
16 K. Uchida, K. Sakai, K. Itakura, T. Osawa and S. Toyokuni,
Arch. Biochem. Biophys., 1997, 346, 45–52.
17 P. C. Burcham, L. M. Kaminskas, F. R. Fontaine,
D. R. Petersen and S. M. Pyke, Toxicology, 2002, 181–182,
229–236.
13 F. Wang, R. Peng and Y. Sha, Molecules, 2008, 13, 18 H. Odani, T. Shinzato, Y. Matsumoto, J. Usami and
922–930.
K. Maeda, Biochem. Biophys. Res. Commun., 1999, 256, 89–93.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012