Synthesis and reactivity of some 6-substituted-2,4-dimethyl-3-pyridinols, a novel class of chain-breaking antioxidants

Article abstract of DOI:10.1021/jo048842u

The synthesis and study of a series of 6-substituted-2,4-dimethyl-3- pyridinols having interesting antioxidant properties is reported. The general synthetic strategy leading to the compounds involved a low-temperature aryl bromide-to-alcohol conversion as the last step. 2,4-Dimethyl-3-pyridinol (1a), 2,4,6-trimethyl-3-pyridinol (1b), and 2,4-dimethyl-6-(dimethylamino)-3-pyridinol (1d) were thus prepared from the corresponding 3-bromopyridine precursor. The methoxy derivative 2,4-dimethyl-6-(methoxy)-3-pyridinol (1c) was also prepared by an alternate route via a Baeyer-Villiger reaction on the substituted benzaldehyde precursor. Novel bicyclic pyridinols 2 and 3 required prior construction of the ring structure. Thus, 2 was prepared by the use of a 6-step intramolecular Friedel-Crafts strategy, and 3 required an 11-step sequence with a thermolytic intramolecular inverse-demand Diels-Alder reaction between a pyrimidine ring and an alkyne as the key step. Basicities of the pyridinols approached physiological pH with increasing electron density in the ring. Pyridinols 1a-d were found to be indefinitely stable to air oxidation while 2 and 3 decomposed upon extended exposure to the atmosphere. The reactivities of the pyridinols toward chain-carrying peroxyl radicals in homogeneous organic solution were examined by studying the kinetics of radical-initiated styrene autoxidations under controlled conditions. These experiments revealed that some of the newly synthesized pyridinols are the most effective phenolic chain-breaking antioxidants reported to date.

Full text of DOI:10.1021/jo048842u

Synthesis and Reactivity of Some
6-Substituted-2,4-dimethyl-3-pyridinols, a Novel Class of
Chain-Breaking Antioxidants
Maikel Wijtmans,‡,§ Derek A. Pratt,*,§,# Johan Brinkhorst,§ Remigiusz Serwa,§
Luca Valgimigli, Gian Franco Pedulli, and Ned A. Porter*,§
Department of Chemistry, Vanderbilt University, Box 1822 Station B, Nashville, Tennessee 37215, and
Department of Organic Chemistry “A. Mangini”, University of Bologna, Via San Donato 15,
I-40127 Bologna, Italy
dpratt@uiuc.edu; n.porter@vanderbilt.edu
Received July 8, 2004
The synthesis and study of a series of 6-substituted-2,4-dimethyl-3-pyridinols having interesting
antioxidant properties is reported. The general synthetic strategy leading to the compounds involved
a low-temperature aryl bromide-to-alcohol conversion as the last step. 2,4-Dimethyl-3-pyridinol
(1a), 2,4,6-trimethyl-3-pyridinol (1b), and 2,4-dimethyl-6-(dimethylamino)-3-pyridinol (1d) were thus
prepared from the corresponding 3-bromopyridine precursor. The methoxy derivative 2,4-dimethyl-
6-(methoxy)-3-pyridinol (1c) was also prepared by an alternate route via a Baeyer-Villiger reaction
on the substituted benzaldehyde precursor. Novel bicyclic pyridinols 2 and 3 required prior
construction of the ring structure. Thus, 2 was prepared by the use of a 6-step intramolecular
Friedel-Crafts strategy, and 3 required an 11-step sequence with a thermolytic intramolecular
inverse-demand Diels-Alder reaction between a pyrimidine ring and an alkyne as the key step.
Basicities of the pyridinols approached physiological pH with increasing electron density in the
ring. Pyridinols 1a-d were found to be indefinitely stable to air oxidation while 2 and 3 decomposed
upon extended exposure to the atmosphere. The reactivities of the pyridinols toward chain-carrying
peroxyl radicals in homogeneous organic solution were examined by studying the kinetics of radical-
initiated styrene autoxidations under controlled conditions. These experiments revealed that some
of the newly synthesized pyridinols are the most effective phenolic chain-breaking antioxidants
reported to date.
SCHEME 1. Chain Propagation Steps for Free
Radical Autoxidation and Inhibition by Phenolic
Compounds
Introduction
Free radical chain oxidation, commonly referred to as
lipid peroxidation when the reactant is a lipid, has been
implicated as a key factor in a number of degenerative
diseases such as atherosclerosis.1 Polyunsaturated fatty
acids and esters are the principal targets of chain-
carrying peroxyl radicals in this transformation.2 Chain-
breaking antioxidants, most commonly substituted phe-
nols, can effectively quench the propagating chain (eq 1).3
They do so by transferring their phenolic H-atom to the
propagating lipid peroxyl radical (eq 3) at a rate faster
than chain propagation (eq 1). Reaction of the phenols
with L•, rather than with LOO•, can be neglected at low
[ArOH] and normal [O2] since the reaction of L• with O2
(eq 2) dominates as its rate is at or near the diffusion-
controlled limit. The relevant chain propagation and
inhibition steps are presented in Scheme 1.
‡ Taken in part from the Ph.D. dissertation of Maikel Wijtmans,
The University of Nijmegen, Toernooiveld 1, 6525 ED, Nijmegen, The
Netherlands.
§ Vanderbilt University.
University of Bologna.
The most famous example of a phenolic antioxidant is
Nature’s best lipophilic antioxidant, R-tocopherol (R-
TOH), the most potent form of Vitamin E. Typical
propagation rate constants, kp, for polyunsaturated lipid
substrates are on the order of 10 to 100 M-1 s-1 while
the inhibition rate constant for R-tocopherol, kinh, is
# Current Address: Department of Organic Chemistry, University
of Illinois at Urbana-Champaign, Urbana, IL 61801.
(1) For example: (a) Steinburg, D.; Parhsarathy, S.; Carew, T. E.;
Khoo, J. C.; Witztum, J. L. New Engl. J. Med. 1989, 320, 915. (b)
Chisholm, G. M.; Steinburg, D. Free Radical Biol. Med. 2000, 28, 1815.
(2) (a) Porter, N. A. Acc. Chem. Res. 1986, 19, 262. (b) Porter, N. A.;
Caldwell, S. E.; Mills, K. A. Lipids 1995, 30, 277.
greater than 106 M-1 s-1
.
(3) (a) Burton, G. W.; Doba, T.; Gabe, E. J.; Hughes, L.; Lee, F.;
Prasad, L.; Ingold, K. U. J. Am. Chem. Soc. 1985, 107, 7053. (b) Burton,
G. W.; Ingold, K. U. J. Am. Chem. Soc. 1981, 103, 6472. (c) Bowry, V.
R.; Ingold, K. U. Acc. Chem. Res. 1999, 32, 27.
Extensive efforts to design synthetic antioxidants
whose inhibition rate constants are greater than that of
R-tocopherol have been made in recent years because of
10.1021/jo048842u CCC: $27.50 © 2004 American Chemical Society
Published on Web 11/10/2004
J. Org. Chem. 2004, 69, 9215-9223
9215

Wijtmans et al.
the therapeutic and industrial importance of these
compounds, but this search has met with limited success
due to the high reactivity of many promising candidates
to oxygen.3a-7 Recently, some of us proposed8 that sub-
stituted 5-pyrimidinol and 3-pyridinol derivatives would
possess similar reactivities toward chain-carrying peroxyl
radicals as equivalently substituted phenols, but be much
more stable to air oxidation. This was suggested to allow
the introduction of very strongly electron-donating groups
(e.g., N,N-dialkylamino) para to the O-H on the ring to
enhance the reactivity of these compounds beyond that
of R-TOH. Substitution of such groups on phenols leads
to compounds that are unstable in air. For example, 4,6-
dimethyl-2-(dimethylamino)-5-pyrimidinol, shown below,
was predicted by theory to be slightly more reactive to
peroxyls than R-TOH, but much more stable to air
oxidation. Subsequent experimental studies have con-
firmed both of these predictions.8
and its slightly more reactive pentamethylbenzofuranol
(five-membered heterocyclic ring) analogue, further re-
duces the O-H bond dissociation enthalpy (BDE) and
improves antioxidant efficacy.9 This report describes
synthetic efforts undertaken to prepare pyridinols 1a-
d, having different substituents at the 6-position on the
ring, along with details of the preparation of the bicyclic
compounds 2 and 3. We provide here reliable routes for
this interesting class of compounds along with some key
physical properties.
Results and Discussion
Synthesis. (a) General Strategy: The Hydroxyl-
ation Step. Several parameters were important in
devising a general synthetic strategy. For example,
p-aminopyridinols are electron-rich and can be reactive
with molecular oxygen. This suggests that introduction
of the OH group should be performed as late in the
sequence as possible to avoid decomposition of unstable
intermediates. Furthermore, the hydroxylation step should
be performed under very mild conditions given the
expected moderate stabilities of some of the pyridinols.
A sequence starting with construction of the appropriate
mono- or bicyclic aminopyridine precursor was envi-
sioned, followed by a hydroxylation procedure. The Boy-
land-Sims peroxidation, the only general method to
directly para-hydroxylate activated pyridine rings,10 is
known to fail for aminopyridines when the ortho position
is vacant.11 Some direct hydroxylations with aq H2O2
(AcOH, 100 °C) have been reported,12 but in all cases very
low yields were obtained. It is likely that such a reaction
will have very low selectivity for the electron-rich sub-
strates 1d, 2, and 3 due to their expected high oxidiz-
abilities. Baeyer-Villiger approaches call for the appro-
priately substituted aldehydes, which are themselves
difficult to prepare.12b Diazonium precursors have been
shown to give undesired coupling products upon at-
tempted hydrolysis (CuSO4, H2O, 100 °C).13 A procedure
by Van Boeckel et al., which utilized a low-temperature
bromide/lithium exchange, followed by a nitrobenzene
quench14 to obtain a 6-amino-3-pyridinol from the corre-
sponding 6-amino-3-bromopyridine15 piqued our interest,
and we set out to use it in a general way to prepare the
simplest aminopyridinol, 1d, from the readily prepared
3-bromo precursor. This approach is illustrated in Scheme
2.
In a recent communication, we predicted by theory that
6-amino-3-pyridinols should be even more reactive anti-
oxidants than the corresponding pyrimidinols (e.g. above),
while still possessing acceptable air stability.9 We also
showed that these pyridinols are, in fact, some of the best
phenolic chain-breaking antioxidants known, with three
derivatives (1d, 2, and 3) being 5, 28, and 88 times more
active than R-TOH in intercepting peroxyl radicals. It
was shown that including the para heteroatom in an
aliphatic ring to form bicyclic compounds, such as R-TOH
The appropriate starting material leading to 1d,
2-amino-4,6-lutidine, is commercially available and in-
expensive. It was brominated regioselectively para to the
amino group to give pyridine 4,16 which was methylated
under Eschweiler-Clark conditions to afford 5. Hy-
droxylation of arylbromides via aryllithium intermediates
(4) (a) Mukai, K.; Okabe, K.; Hosose, H. J. Org. Chem. 1989, 54,
557. (b) Barclay, L. R. C.; Vinqvist, M. R.; Mukai, K.; Itoh, S.; Morimoto,
H. J. Org. Chem. 1993, 58, 7416.
(5) (a) Brownstein, S.; Burton, G. W.; Hughes, L.; Ingold, K. U. J.
Org. Chem. 1989, 54, 560. (b) Noguchi, N.; Iwaki, Y.; Takahashi, M.;
Komuro, E.; Kato, Y.; Tamura, K.; Cynshi, O.; Kodama, T.; Niki, E.
Arch. Biochem. Biophys. 1997, 342, 236. (c) Tamura, K.; Kato, Y.;
Ishikawa, A.; Kato, Y.; Himoro, M.; Yoshida, M.; Takashima, Y.;
Suzuki, T.; Kawabe, Y.; Cynshi, O.; Kodama, T.; Niki, E.; Shimizu, M.
J. Med. Chem. 2003, 46, 3083.
(10) Behrman, E. J.; Pitt, B. M. J. Am. Chem. Soc. 1958, 80, 3717.
(11) Behrman, E. J. J. Org. Chem. 1992, 57, 2266.
(6) Lucarini, M.; Pedrielli, P.; Pedulli, G. F.; Valgimigli, L.; Gigmes,
D.; Tordo, P. J. Am. Chem. Soc. 1999, 121, 11546.
(12) (a) Grozinger, K. G.; Byrne, D. P.; Nummy, L. J.; Ridges, M.
D.; Salvagno, A. J. Heterocycl. Chem. 2000, 37, 229. (b) Sherlock, M.
H.; Kaminski, J. J.; Tom, W. C.; Lee, J. F.; Wong, S.; Kreutner, W.;
Bryant, R W.; McPhail, A T. J. Med. Chem. 1988, 31, 2108.
(13) Van de Poe¨l, H.; Guillaumet, G.; Viaud-Massuard, M. C.
Heterocycles 2002, 57, 55.
(7) (a) Foti, M. C.; Johnson, E. R.; Vinqvist, M. R.; Wright, J. S.;
Barclay, L. R. C.; Ingold, K. U. J. Org. Chem. 2002, 67, 5190. (b)
Hussain, H. H.; Babic G.; Durst, T.; Wright, J. S.; Flueraru, M.;
Chichirau, A.; Chepelev, L. L. J. Org. Chem. 2003, 68, 7023.
(8) (a) Pratt, D. A.; DiLabio, G. A.; Brigati, G.; Pedulli, G. F.;
Valgimigli, L. J. Am. Chem. Soc. 2001, 123, 4625. (b) Valgimigli, L.;
Brigati, G.; Pedulli, G. F.; DiLabio, G. A.; Mastragostino, M. M.;
Arbizzani, C.; Pratt, D. A. Chem. Eur. J. 2003, 9, 4997.
(14) Buck, P.; Kobrich, G. Tetrahedron Lett. 1967, 8, 1563.
(15) van Boeckel, C. A. A.; Delbressine, L. P. C.; Kaspersen, F. M.
Recl. Trav. Chim. Pays-Bas 1985, 104, 259.
(16) Heitsch, H.; Becker, R. H. A.; Kleemann, H.-W.; Wagner, A.
Bioorg. Med. Chem. 1997, 5, 673.
(9) Wijtmans, M.; Pratt, D. A.; Valgimigli, L.; DiLabio, G. A.; Pedulli
G. F.; Porter, N. A. Angew. Chem., Int. Ed. 2003, 42, 4370.
9216 J. Org. Chem., Vol. 69, No. 26, 2004

Synthesis of 6-Substituted-2,4-dimethyl-3-pyridinols
SCHEME 2. Synthesis of Monocyclic Pyridinol 1da
SCHEME 3. Synthesis of Pyridinol 1ca
a Reagents and conditions: (a) dibromodimethylhydantoin,
DCM, -40 °C, 45 min, 67%; (b) HCOOH, i-PrOH, H2CO (aq),
reflux, 100%; (c) (1) n-BuLi, THF, -78 °C, (2) 2,6-dimethyl-
nitrobenzene, 63%.
a Reagents and conditions: (a) NaONO, H3PO2, H2O, 5 °C, 87%;
(b) Ag2CO3, CH3I, CH2Cl2, 24 h, 90%; (c) (1) n-BuLi, THF, -78
°C, (2) DMF, 87%; (d) (1) m-CPBA, CHCl3, (2) KOH, MeOH, 33%.
SCHEME 4. Synthesis of Pyridinols Containing
an Annulated Aliphatic Six-Membered Ringa
has been reported in the literature. Reagents that have
been used for the oxygenation step include bis(trimeth-
ylsilyl)peroxide,17 lithium tert-butylhydroperoxide,18 or
trimethylborate/CH3CO3H.13 However, all of these re-
agents contain active peroxy bonds and problems in their
compatibility with electron-rich aminopyridines or amino-
pyridinols were expected. Therefore, we followed the
protocol used by Van Boeckel et al.15 Briefly, bromide 5
underwent smooth lithium/bromide exchange with n-
BuLi in THF at -78 °C and the resulting pyridyllithio
reagent was quenched with 2,6-dimethylnitrobenzene
(vide infra) to afford compound 1d as a yellow solid in
63% isolated yield. Purification consisted of a base/acid
extraction, which eliminated the need for purification by
column chromatography. Another advantage of the syn-
thesis outlined in Scheme 2 is that a variety of exocyclic
amine substituents can be introduced by reductive alkyl-
ation of 4. This would provide antioxidants having a
range of lipophilicity, a parameter that has been shown
to be important in determining antioxidant efficacy for
some applications.
a Reagents and conditions: (a) acrylic acid, py, reflux, 24 h, 30%;
(b) PPA, 125 °C, 40 min, 75%; (c) BH3‚THF, THF, reflux, 18h, 83%;
(d) dibromodimethylhydantoin, DCM, -78 °C, 93%; (e) HCOOH,
H2CO(aq), i-PrOH, reflux, 77%; (f) (1) n-BuLi, THF, -78 °C, (2)
2,6-dimethylnitrobenzene, 25%; (g) H2, 10% Pd/C, EtOH, over-
night, 74%.
to the pyridinol by the same lithiation-hydroxylation
sequence. Compound 1c was prepared from the corre-
sponding bromide 8, which in turn was prepared by
methylation (Ag2CO3/CH3I) of 3-bromo-2,4-dimethyl-6-
hydroxypyridine 7. Compound 7 was obtained from
amine 4 by hydrolysis of the corresponding diazonium
salt in the presence of H3PO2. Generally, under these
conditions deaminated products form, but in our case
hydrolysis prevails over reduction of the diazonium salt.
We attribute this to the neighboring pyridine nitrogen.
A Baeyer-Villiger sequence was used to generate 1c from
the corresponding aldehyde precursor 9, as shown in
Scheme 3. The reactions leading to the aldehyde inter-
mediate proceeded in high yield but the Baeyer-Villiger
reaction also gave the product pyridinol in only modest
yields, offering no advantage to this alternate procedure.
(b) Pyridinols with an Annulated Six-Membered
Ring. The construction of the bicyclic pyridinol 2 re-
quired the formation of a tetrahydronaphthyridine struc-
ture prior to bromination and hydroxylation. A Friedel-
Crafts approach was used for the annulation process
(Scheme 4).21 A sluggish Michael reaction of 2-amino-4,6-
lutidine with acrylic acid in refluxing pyridine afforded
amino acid 10. The subsequent ring closure proceeded
in either hot polyphosphoric acid (PPA) or hot methane-
sulfonic acid. The reduction of aryl ketone 11 was
achieved by reacting it with excess BH3-THF complex
at reflux to afford 12 in 83% yield and in high purity.
Alternatively, 12 was obtained in 74% recrystallized yield
Clear advantages of this hydroxylation protocol for our
targets were the low reaction temperature (-78 °C) and
the fact that nitroarenes, which are usually not consid-
ered oxidants, are compatible with electron-rich pyri-
dines. Indeed, as will be shown, this protocol was suc-
cessfully and reproducibly applied to all of our target
substrates. However, the yields of the reaction were
modest to low. Poor chromatography of the pyridinols
could be partially avoided by using Et3N-deactivated
silica gel as the stationary phase. A more significant
problem in this step was the formation of large amounts
of the reduced pyridine (i.e. 6 in the case of 1d) due to
quenching of the pyridyllithio species. Cava et al. have
found that the formation of the reduced product is caused
by abstraction of an ortho proton from PhNO2 by the
aryllithium intermediate.19 Therefore, for the present
study, 2,6-dimethylnitrobenzene was selected as the
electrophile rather than nitrobenzene. This reagent gave
comparable or slightly higher yields than nitrobenzene
and it was easier to remove from the crude product
because of its reduced polarity. It did not, however,
eliminate the formation of the reduced byproduct. Follow-
up studies, aimed at developing a means to avoid this
serious side reaction, were fruitless.
Syntheses of 1a and 1b, which are known compounds,20
proceeded by conversion of the requisite bromopyridine
(17) Taddei, M.; Ricci, A. Synthesis 1986, 8, 633.
(18) Julia, M.; Pfeuty-Saint Jalmes, V.; Ple, K.; Verpeaux, J. Bull.
Soc. Chim. Fr. 1996, 133, 15.
(19) Wiriyachitra, P.; Cava, M. P. J. Org. Chem. 1977, 42, 2274.
(20) (a) Wang, Y.; Liu, M.; Lin, T.; Sartorelli, A. J. Med. Chem. 1992,
35, 3667. (b) Plazek Chem. Ber. 1939, 72, 577.
(21) Ferrarini, P. L.; Mori, C.; Badawneh, M.; Calderone, V.; Greco,
R.; Manera, C.; Martinelli, A.; Nieri, P.; Saccomanni, G. Eur. J. Med.
Chem. 2000, 35, 815.
J. Org. Chem, Vol. 69, No. 26, 2004 9217

Wijtmans et al.
SCHEME 5. Synthesis of Pyridinols Containing an Annulated Aliphatic Five-Membered Ringa
a Reagents and conditions: (a) 15 M KOH, reflux, 4 h, 80%; (b) Ph2PON3, Et3N, t-BuOH, reflux, 79%; (c) HCl, ether, 70%; (d) 4,6-
dimethylpyrimidinyltriflate (25), Et3N, DMF, 1 d, 93%; (e) Ac2O, DMAP, 100 °C, 20 h, 91%; (f) Ph2O, 1 d, reflux, 82%; (g) KOH, MeOH,
reflux, 100%; (h) dibromodimethylhydantoin, DCM, -78 °C, 83%; (i) HCOOH, H2CO(aq), i-PrOH, reflux, 84%; (j) (1) n-BuLi, THF, -78
°C, (2) 2,6-dimethylnitrobenzene, 27%.
from 2,4-dimethyl-[1,8]-naphthyridine (15) by selective
catalytic hydrogenation.
base, which isomerized the triple bond to give acid 16.29
This compound was mixed with diphenylphosphoryl azide
in refluxing t-BuOH, which induced a one-pot acid
azidation, Curtius rearrangement, and alcoholysis.30 The
crude Boc-protected amine 17 was deprotected with 1.0
M ethereal HCl and salt 18 was isolated in good yield
simply by filtration.
The bromination of 12 at -78 °C proceeded well, but
care had to be taken that over-bromination, presumably
on the benzylic positions, did not take place. After the
reaction, product 13 was precipitated by adding aqueous
KOH and Na2S2O3. The methyl derivative 14 was ob-
tained under the usual Eschweiler-Clark conditions. The
bromide was converted to the corresponding pyridinol by
the standard hydroxylation method to obtain 2 in 25%
isolated yield.
(c) Pyridinols with an Annulated Five-Membered
Ring. Bicyclic pyridinols with an annulated five-mem-
bered ring (i.e. with a dihydropyrrolopyridine system)
represented the synthetically most challenging class of
pyridinol reported here. Simple dihydropyrrolopyridine
structures have been prepared from chloride precursors,22
by radical cyclizations23 or by sluggish hydrogenations
of aza-indoles.24 All of these reactions required precursors
that were not readily available with the desired substitu-
tion pattern of the pyridinol product. After some unsuc-
cessful attempts with intramolecular Friedel-Crafts
chemistry, a sequence with Van der Plas’s reaction as
the key step was selected.25 This reaction consists of an
intramolecular inverse electron demand Diels-Alder
reaction between an alkyne and pyrimidine ring followed
by a cycloreversion (vide infra). The whole sequence is
depicted in Scheme 5.
Coupling of this salt with triflate 25, prepared from
4,6-dimethyl-2-pyrimidinol and triflic anhydride,31 was
achieved by a nucleophilic aromatic substitution in
distilled DMF and in the presence of excess Et3N to afford
amine 19. The two methyls on the pyrimidine ring
reduced the coupling rate considerably compared to
nonsubstituted pyrimidines27 and it was for this reason
that the triflate proved superior over other activating
groups such as chloride27 and tosylate. Acetylation of 19
was accomplished by heating in acetic anhydride in the
presence of DMAP, giving acetamide 20. The next step
was the Van der Plas reaction.27 After 3 d of reflux in
nitrobenzene (210 °C), a common solvent for this reaction,
21 was formed with good conversion. It was found,
however, that the reaction proceeded significantly faster
and cleaner in refluxing diphenyl ether (250 °C), allowing
an easy workup consisting of an acid/base extraction to
give light-colored recrystallized product 21 in 82% yield.
The presence of two methyl groups on the pyrimidine
ring, even though one is eliminated as MeCN, avoids a
potential selectivity issue in the cycloreversion step.
Subsequently, acetamide 21 was subjected to basic
methanolysis which cleaved the acetamide group. The
rest of the synthesis was performed by a sequence similar
to that of the annulated six-membered analogue 2. Thus,
22 was brominated to give 23, which was then converted
into the methylated compound 24 by an Eschweiler-
Clark methylation. The bromide 24 was hydroxylated to
give 3 in 27% isolated yield after column chromatogra-
phy.
Compound 18 is a known compound, but the literature
procedures are lengthy26,27 or give very low yields when
repeated.28 The alternative synthesis shown in Scheme
5 was very easy to scale up and provided crystalline 18.
5-Hexynoic acid was treated with concentrated aqueous
(22) Yahkontov, A.; Propokov, T. Usp. Khim. 1980, 49, 419.
(23) Johnston, J. N.; Plotkin, M. A.; Viswanathan, R.; Prabhakaran,
E. N. Org. Lett. 2001, 3, 1009.
(24) Clemo, G. R.; Swan, G. A. J. Chem. Soc. 1945, 603.
(25) Frissen, A. E.; Marcelis, A. T. M.; Van der Plas, H. C.
Tetrahedron 1989, 45, 803.
(29) Jones, E. R. H.; Whitham, G. H.; Whiting, M. C. J. Chem. Soc.
(26) Rahm, A.; Wulff, W. D. J. Am. Chem. Soc. 1996, 118, 1807.
(27) Esch, P. M.; Hiemstra, H.; de Boer, R. F.; Speckamp, W. C.
Tetrahedron 1992, 22, 4659.
1954, 3201.
(30) Ninomiya, K.; Shioiri, T.; Yamada, S. Tetrahedron 1974, 30,
2151.
(28) Doetz, K. H.; Schafer, T. O.; Harms, K. Synthesis 1992, 146.
(31) Sandosham, J.; Undheim, K. Heterocycles 1994, 37, 501.
9218 J. Org. Chem., Vol. 69, No. 26, 2004

Synthesis of 6-Substituted-2,4-dimethyl-3-pyridinols
TABLE 1. Some Relevant Properties of Pyridinols
1a-d, 2, and 3
half-life
(h)a
mp
(°C)
kinh
kinh/
kinh(R-TOH)
c
d
compd
pKa
(M-1s -1
)
1a
1b
1c
1d
2
.24b 102
.24b 129
.24b 77
.24b 132
5.45 (5.8 ( 2.3) × 103
5.95 (2.2 ( 0.3) × 104
6.75 (2.9 ( 0.4) × 105
6.97 (1.6 ( 0.6) × 107
0.0018
0.0069
0.091
5
30
14
126 (dec) >7.0 (8.8 ( 3.2) × 107
28
3
115 (dec) >7.0 (2.8 ( 1.8) × 108
88
1
R-TOH .24b n/a
n/a
3.2 × 106 e
a Estimated from decay of UV absorbance at λmax in aerated tert-
butylbenzene (ca. 0.3 mM) at 37 °C. Values taken from ref 9. b No
significant change in UV signal observed in 24 h. c The pKa value
for the protonated form of the pyridinol (see Supporting Informa-
tion). d Obtained from autoxidations of styrene in chlorobenzene
at 303 K. e Value for R-TOH taken from ref 3a.
Some Relevant Properties. Since the stability of
electron-rich phenols upon exposure to air has been a
major hurdle in the successful design of novel phenolic
antioxidants with enhanced activities relative to R-toco-
pherol, we first sought to examine the stabilities of our
newly synthesized pyridinols in organic solutions. For
this purpose, we simply monitored the absorbance of the
intact (i.e., non-oxidized) chromophore as a function of
time in aerated tert-butylbenzene. The results are pre-
sented in Table 1.
The introduction of a basic nitrogen atom in the
aromatic ring of phenolic compounds can potentially
create a problem for their efficacy as antioxidants in
protic solvents should the basic pyridine nitrogen (or the
p-amine substituent in the case of 1d, 2, and 3) be
protonated to a significant extent at equilibrium. The
basicity of these compounds is also relevant to issues
associated with the isolation and purification of those
pyridinols presented here. The pyridinium ion itself has
a pKa of 5.25 but it was expected that the electron-
donating substituents in the target pyridinols may raise
this value substantially. Titrations of the dibasic pheno-
late salts of 1a-3 with dilute HCl in MeOH/H2O (1/3)
were therefore performed to provide some insight into
their acid/base equilibria. These results are also collected
in Table 1 with additional details available as Supporting
Information.
Last, the rate of the reaction between the pyridinol and
a peroxyl radical is of definitive importance for the overall
antioxidant activity. We found that standard methyl
linoleate autoxidations32 are greatly inhibited in the
presence of compounds 1d, 2, and 3. Styrene autoxida-
tions3a inhibited by 1a-d, 2, and 3 were well-behaved
and provided inhibition rate data (kinh) for each of the
six compounds.33 The stoichiometric factor n (number of
peroxyl radicals trapped by each molecule of antioxidant)
was determined to be 2 for all of the compounds, similar
to phenols and pyrimidinols. These data are collected in
Table 1.
FIGURE 1. Inhibition rate constants plotted as a function of
the σp+ constant of the substituent para to the hydroxyl group
in 2,4-dimethyl-3-pyridinols (closed symbols, solid line) and
2,6-dimethylphenols (open symbols, dashed line).
3, heating results in decomposition indicated by the
formation of dark brown material. Both trends are
indicative of the anticipated decrease in aerobic stability
upon increasing the electron-density in the aromatic ring.
In an analogous manner, basicity increases as electron-
donating substituents are incorporated. Titrations for 2
and 3 yielded somewhat distorted plots, probably due to
decomposition of the very electron-rich pyridinoxide
anions. Yet, it was evident from the curves that the pKa
values for the protonated forms of these compounds were
higher than 7.0 (7.0-7.2) and were thus approaching
physiological pH values. The anticipated substituent
effect was also markedly clear from the kinh values. An
increase of as many as 4.5 orders of magnitude was
observed between the values of 1a and 3.
The bond dissocation enthalpy (BDE) of the phenolic
O-H bond plays a central role34,35 in determining anti-
oxidant efficacy, compounds having lower BDEs generally
being better antioxidants. The introduction of electron-
donating groups at the para position relative to the
phenolic OH leads to compounds with lower BDEs. This
effect is obviously a factor that controls the reactivity of
the pyridinols studied here. It is difficult to obtain BDEs
for 1a-c by the radical equilibration electron paramag-
netic resonance (REqEPR) technique as we have done9
for 1d, 2, and 3 due to the asymmetry and relatively low
persistence of their corresponding 3-pyridinoxyl radicals.
It is well-known, however, that phenolic O-H BDEs
correlate very well with σp+ substituent constants,36 and
as such, kinh for 1a-d should also correlate well with
these substituent constants if the O-H BDE is indeed a
factor that controls the reactivity of the pyridinols studied
+
Inspection of the table shows clear and consistent
trends. In the subseries 1d, 2, and 3, the half-life in
aerated solution decreases. For bicyclic members 2 and
here. A plot of log kinh for 1a-d as a function of σp of
(34) (a) Wayner, D. D. M.; Lusztyk, E.; Ingold, K. U.; Mulder, P. J.
Org. Chem. 1996, 61, 6430. (b) Lucarini, M.; Pedrielli, P.; Pedulli, G.
F.; Cabiddu, S.; Fattuoni, C. J. Org. Chem. 1996, 61, 9263.
(35) Brigati, G.; Lucarini, M.; Mugnaini, V.; Pedulli, G. F. J. Org.
Chem. 2002, 67, 4828.
(36) Pratt, D. A.; DiLabio, G. A.; Mulder, P.; Ingold, K. U. Acc. Chem.
Res. 2004, 37, 334.
(32) Tallman, K. A.; Pratt, D. A.; Porter, N. A. J. Am. Chem. Soc.
2001, 123, 11827.
(33) Amorati, R.; Pedulli, G. F.; Valgimigli, L.; Attanasi, O. A.;
Filippone, P.; Fiorucci, C.; Saladino, R. J. Chem. Soc., Perkin Trans. 2
2001, 2142.
J. Org. Chem, Vol. 69, No. 26, 2004 9219

Wijtmans et al.
stirred for 2 h at -78 °C, the reaction mixture was warmed to
room temperature and quenched with saturated aq Na2CO3.
This mixture was extracted with ether (3×) to remove 6 and
2,6-dimethylnitrobenzene. The aqueous phase was cooled in
an ice bath and carefully neutralized by adding concentrated
HCl dropwise. The yellow mixture was extracted with EtOAc
(3×). The organic extracts were dried (MgSO4), filtered, and
concentrated to give a yellow solid product (400 mg, 63%) that
was a single highly fluorescent spot by TLC. NMR revealed
that the product was g95% pure. The product could be
recrystallized by dissolving it quickly in a minimum amount
of hot ether/hexanes (1/1), cooling, storing at -30 °C, filtering,
washing, and drying. This protocol yielded yellow crystals
which were stable under argon in the freezer for over a year.
Mp 132 °C; λmax (MeOH) 328 nm; 1H NMR (CDCl3, 300 MHz)
δ 6.19 (br s, 1H, Ar-H), 4.75 (br s, OH), 2.96 (br s, 6H, N-Me),
2.29 (br s, 3H, 6-Me), 2.16 (br s, 3H, 4-Me); 1H NMR (CDCl3 +
drop of D2O, 300 MHz) same shifts as without D2O, but sharp
signals and no -OH signal; 13C NMR (CDCl3 + drop of D2O,
75 MHz) δ 154.8, 143.0, 141.1, 136.5, 106.3, 39.3, 19.3, 16.8;
2D NOESY-NMR (CDCl3 + drop of D2O, 400 MHz) couplings
Ar-H × N-Me, Ar-H × 4-Me; IR (film) ν (cm-1) 3400, 3000,
2920, 1610, 1480, 1420, 1402, 1260; HRMS (FAB) for C9H14N2O
[M] 166.1106, found 116.1108. Anal. Calcd for C9H14N2O: C
65.03, H 8.49, N 16.85. Found: C 64.82, H 8.71, N 16.59.
the 6-substituent, as shown in Figure 1 (closed symbols
and solid line), yields a straight line that is essentially
parallel to that for the equivalently substituted phenols
(open symbols and dashed line).3a The lower y-intercept
for the 3-pyridinols is expected on the basis of our
theoretical calculations, which indicate that the O-H
BDEs in pyridinols are roughly 1-1.5 kcal/mol higher
than those of equivalently substituted phenols.8a,9
The parallel lines confirm that substituent effects on the
O-H BDEs of 3-pyridinols are essentially equivalent to
those on phenols as was predicted by our theoretical
calculations.8a,9
Summary
A series of 6-substituted-2,4-dimethyl-3-pyridinols was
synthesized by using a common general strategy. The
introduction of the reactive hydroxyl group in the final
step allowed for easy scale-up and storage of all inter-
mediates up to the stable penultimate pyridyl bromides.
The hydroxylation step was a relatively low-yielding step,
but it proved reproducible and generally applicable. Thus,
a general route to this interesting class of compounds has
been established. The stabilities of the pyridinols toward
air oxidation are in qualitative agreement with their
previously calculated ionization potentials. The basicity
of the pyridinols increased as the electron density in the
ring increased and it approached physiological pH values
for the bicyclic compounds 2 and 3. The rate constants
for inhibition of styrene autoxidations were determined
for all of the pyridinols and the more electron-rich
compounds proved to be the best antioxidants. The rate
constants correlate with σp+ for the pyridinol series 1a-
d, a result that is fully consistent with that observed for
the analogous phenols.
5-Bromo-4,6-dimethylpyridin-2(1H)-one (7). Amine 4
(6.00 g, 29.8 mmol) was mixed with a 50% aqueous solution
of hypophosphorous acid (25 mL, 241 mmol) and water (55
mL). The mixture was cooled to about 2 °C and a solution of
sodium nitrite (2.40 g, 34.8 mmol) in water (12 mL) was added
with vigorous stirring keeping the temperature below 5 °C.
The mixture was stirred for 30 min at a lower temperature
and then for an additional 12 h at room temperature. The
mixture was neutralized to pH 6-7 and kept at 5 °C for 6 h.
The resulting precipitate was filtered and washed with water.
After drying under high vacuum, a white solid was obtained
1
(5.25 g, 87%). H NMR (CDCl3, 300 MHz) δ 6.32 (s, 1H, Ar-
H), 2.45 (s, 3H, 6-Me), 2.27 (s, 3H, 4-Me); 13C NMR (d6-DMSO,
100 MHz) δ 161.6, 151.0, 144.5, 116.3, 102.2, 23.0, 20.2; mp
255-257 °C (lit.37 mp 257-259 °C).
3-Bromo-6-methoxy-2,4-dimethylpyridine (8). Silver
carbonate (9.55 g, 34.7 mmol) and iodomethane (16.5 mL, 266
mmol) were added to bromide 7 (5.20 g, 25.3 mmol) dissolved
in CH2Cl2 (260 mL) with protection from light. The reaction
mixture was stirred until all starting material was consumed
(overnight) and the inorganic solids were removed by filtration
and washed with CH2Cl2 (3 × 40 mL). The filtrate was
concentrated in vacuo to yield a rust colored oil (5.02 g, 90%).
1H NMR (CDCl3, 300 MHz) δ 6.43 (s, 1H, Ar-H), 3.85 (s, 3H,
O-Me), 2.54 (s, 3H, 2-Me), 2.30 (s, 3H, 4-Me); 13C NMR (CDCl3,
75 MHz) δ 162.0, 154.4, 149.5, 115.3, 109.4, 53.42, 25.20, 23.28;
HRMS for C8H10NOBr [M + Na] 237.9838, found 237.9840.
2,4-Dimethyl-6-methoxynicotinaldehyde (9). Bromide
8 (4.00 g, 18.4 mmol) was dissolved in THF (53 mL) and the
solution was cooled to -78 °C. To this was added n-BuLi (2.5
M in hexanes, 7.8 mL, 19.4 mmol) and the reaction mixture
was stirred at -78 °C for 1 h (the Br/Li exchange was
monitored by TLC and was quantitative). Then dry DMF (2.8
mL, 36.1 mmol) was added and stirring was continued at -78
°C for 30 min. The cold mixture was poured into a stirred
aqueous solution of sodium bicarbonate (5% NaHCO3). The
mixture was extracted with ether (3×) and the combined
organic extracts were washed with brine (3×) and dried over
K2CO3. Rotary evaporation yielded a white solid (2.92 g, 89%).
1H NMR (CDCl3, 300 MHz) δ 10.42 (s, 1H, aldehyde), 6.37 (s,
1H, Ar-H), 3.91 (s, 3H, O-Me), 2.69 (s, 3H, 2-Me), 2.50 (s, 3H,
4-Me); 13C NMR (CDCl3, 75 MHz) δ 191.4, 165.5, 162.5, 153.1,
123.3, 111.1, 54.07, 23.38, 21.17; HRMS for C9H11NO2 [M +
H] 166.0862, found 166.0856; mp 80.5-81 °C.
Experimental Section
2-Amino-5-bromo-4,6-dimethylpyridine (4). The title
compound was prepared according to the literature16 (with 0.5
equiv of 1,3-dibromo-5,5-dimethylhydantoin) in lower yield but
higher purity, typically 67% yield after recrystallization from
EtOAc: mp 136 °C (lit.16 mp 108-110 °C). The literature yield
and purity are very likely incorrect since 92% of pure 4 was
reported after recrystallization while there was a reported 1:10
ratio of regioisomers in the crude mixture.
5-Bromo-4,6-dimethyl-2-(dimethylamino)pyridine (5).
Bromide 4 (1.2 g, 5.94 mmol) was heated at reflux overnight
in a mixture of formic acid (20 mL) and 37% formaline (20
mL). The reaction mixture was cooled and most volatiles were
removed under reduced pressure. The residue was shaken with
a mixture of 1.0 M aq NaOH and CH2Cl2 (2×). The organic
layers were collected, dried (MgSO4), and filtered. The filtrate
was concentrated and purified by column chromatography (3:1
hexanes:EtOAc) to afford a yellowish low-metling solid product
1
(1.06 g, 78%). H NMR (CDCl3, 300 MHz) δ 6.21 (s, 1H, Ar-
H), 3.02 (s, 6H, N-Me), 2.53 (s, 3H, 6-Me), 2.29 (s, 3H, 4-Me);
13C NMR (CDCl3, 75 MHz) δ 157.9, 155.1, 147.9, 110.6, 105.6,
38.4, 26.0, 24.11; HRMS (FAB) for C9H13BrN2 [M - H]
227.0184, found 227.0174.
2,4-Dimethyl-6-(dimethylamino)-3-hydroxypyridine
(1d). Bromide 5 (890 mg, 3.86 mmol) was dissolved in THF
(10 mL) and the solution was cooled to -78 °C. To this was
added n-BuLi (2.5 M in hexanes, 3.5 mL, 8.8 mmol) and the
yellow reaction mixture was stirred at -78 °C for 15-30 min
(the Br/Li exchange was monitored by TLC and was quantita-
tive). Then dry 2,6-dimethylnitrobenzene (2.6 mL, 19.80 mmol)
was added, resulting in the solution turning brown. After being
6-Methoxy-2,4-dimethylpyridin-3-ol (1c). A solution of
m-CPBA (0.61 g, 3.50 mmol) and p-toluenesulfonic acid (9 mg,
0.05 mmol) in chloroform (3.5 mL) was slowly added to a
9220 J. Org. Chem., Vol. 69, No. 26, 2004

Synthesis of 6-Substituted-2,4-dimethyl-3-pyridinols
solution of nicotinaldehyde 9 (0.49 g, 2.97 mmol) in chloroform
(3.5 mL). After being stirred for 24 h at room temperature the
solution was treated with aqueous NaOH (1M) until a pH of
8.5 was reached. The mixture was extracted with chloroform
(3×) and the organic layer was dried over Na2SO4 and
concentrated to yield a yellow oil (0.23 g, 43%). This crude
product was dissolved in a solution of KOH (0.26 g, 4.62 mmol)
in methanol (5 mL). The solution was heated at reflux for 2 h.
The solvent was then evaporated. The resulting solids were
dissolved by addition of water and ether. The aqueous layer
was washed with additional ether and then brought to pH 8.5
with aqueous hydrochloric acid (1 M). The formed precipitate
was extracted with EtOAc (3×). The ethyl acetate extracts
ware combined, dried over MgSO4, and concentrated to yield
a pale orange solid (0.15 g, 33% over both steps). Mp 77-79
°C; 1H NMR (CDCl3, 300 MHz) δ 6.36 (s, 1Η, Αr-Η), 4.4 (br s,
OH), 3.82 (s, 3H, O-Me), 2.36 (s, 3H, 2-Me), 2.19 (s, 3H, 4-Me);
13C NMR (CDCl3, 75 MHz) δ 157.9, 144.1, 141.3, 137.6, 108.9,
53.95, 18.95, 16.52; HRMS for C8H11NO2 [M + H] 154.0862,
found 154.0865.
3-(4,6-Dimethylpyridin-2-ylamino)propanoic Acid (10).
2-Amino-4,6-lutidine (8.5 g, 70 mmol) and acrylic acid (6 g, 84
mmol) were heated at reflux in pyridine (60 mL) overnight.
The mixture was concentrated and dried thoroughly (to remove
traces of pyridine and acrylic acid). The following protocol gave
good quality product albeit in low yield: the residue was
dissolved in water containing NaOH (3.1 g, 77 mmol) and
filtered through Celite. The filtrate was cooled in an ice bath
and slowly acidified with concentrated H2SO4. Around pH 3-4
a precipitate formed, which was filtered and washed with
water and acetone affording an off-white solid (4 g, 30%)
sufficiently pure for further reactions. An analytical sample
was obtained by recrystallization from water and then from
i-PrOH followed by extensive drying. The product (as an
i-PrOH-solvate) was obtained as white crystals. Mp 84 °C; 1H
NMR (d3-MeOD, 300 MHz) δ 6.87 (s, 1H, Ar-H), 6.77 (s, 1H,
Ar-H), 3.86 (t, 2H, NCH2, J ) 5.0 Hz), 2.82 (t, 2H, CH2CO, J
) 5.0 Hz), 2.72 (s, 3H, 6-Me), 2.41 (s, 3H, 4-Me); 13C NMR
(d3-MeOD, 75 MHz) δ 180.9, 157.7, 156.6, 150.1, 115.4, 109.0,
41.3, 38.5, 22.6, 20.2; HRMS for C10H14N2O2 [M + H] 195.1128,
found 195.1118.
4-Me), 1.94 (p, 2H, CH2, J ) 5.5 Hz); 13C NMR (CDCl3, 75 MHz)
δ 156.1, 153.8, 146.0, 114.7, 111.7, 41.7, 24.0, 23.5, 22.3, 18.8;
2D NOESY-NMR (CDCl3, 400 MHz) couplings Ar-H × 4-Me,
Ar-H × 6-Me, Ar-CH2 × 4-Me, Ar-CH2 × CH2CH2; HRMS for
C10H14N2 [M + H] 163.1230, found 163.1224. Anal. Calcd for
C10H14N2: C 74.03, H 8.70, N 17.27. Found: C 73.94, H 8.71,
N 17.20. NMR data have been reported.38 Alternative pro-
cedure: 2,4-Dimethyl[1,8]naphthyridine 1539 (6.8 g, 43.0
mmol) and 10% Pd/C (0.5 g) were stirred in EtOH (70 mL)
overnight under H2 atmosphere. The catalyst was removed by
filtration over Celite and the solvent removed by concentration
giving an off-white solid (6.96 g, 42.9 mmol) in near quantita-
tive yield. The product was recrystallized from hexanes to
afford 5.2 g of yellow needles (74%).
6-Bromo-5,7-dimethyl-1,2,3,4-tetrahydro[1,8]naph-
thyridine (13). Amine 12 (1.6 g, 9.8 mmol) was dissolved in
CH2Cl2 (30 mL). The solution was cooled to -78 °C and 1,3-
dibromo-5,5-dimethylhydantoin (1.54 g, 5.4 mmol) was added
portionwise. To avoid over-bromination of the electron-rich
starting material, the following protocol worked best: the dry
ice/acetone bath was removed for the duration of 1-1.5 min
after which the reaction mixture was recooled to -78 °C and
checked by TLC. Usually, after a few of these cycles, clean and
full conversion was achieved. The cooling bath was removed,
saturated aq Na2S2O3 (10 mL) was immediately added, and
the contents of the flask was swirled thoroughly to destroy
the remaining 1,3-dibromo-5,5-dimethylhydantoin. After the
solution was warmed to room temperature while stirring,
water and excess solid KOH were added. The CH2Cl2 was
removed under reduced pressure and the basic suspension was
filtered. The precipitate was washed with water and dried.
This gave the product as a yellow solid (2.2 g, 93%). An
analytical sample was obtained by recrystallization from
MeOH/water to give yellow needles (according to NMR, the
crystals retained 4.7% MeOH in the lattice even after extensive
1
drying). Mp 146 °C; H NMR (CDCl3, 300 MHz) δ 4.95 (br s,
1H, NH), 3.32 (m, 2H, NCH2), 2.65 (t, 2H, Ar-CH2, J ) 7.6
Hz), 2.44 (s, 3H, 6-Me), 2.24 (s, 3H, 4-Me), 1.91 (p, 2H, CH2, J
) 5.5 Hz); 13C NMR (CDCl3, 75 MHz) δ 154.6, 152.7, 145.2,
113.7, 111.9, 41.4, 25.7, 25.0, 22.1, 19.3; HRMS for C10H13BrN2
[M + H] 241.0335, found 241.0340.
6-Bromo-1,5,7-trimethyl-1,2,3,4-tetrahydro[1,8]naph-
thyridine (14). Bromide 13 (1.1 g, 4.56 mmol) was heated at
reflux overnight in a mixture of formic acid (10 mL) and 37%
formaline (10 mL). The reaction mixture was cooled and most
volatiles were removed by evaporation. Enough aq 15 M KOH
was added to make the solution basic and the suspension was
then cooled in an ice bath. The precipitate was filtered, washed
with water, and dried to give the product as a brown solid (0.9
g, 77%) of satisfactory purity. An analytical sample was
prepared by running the solid though a plug of silica gel with
3:1 hexanes:EtOAc as the eluent. This gave white needles. Mp
5,7-Dimethyl-2,3-dihydro-1H-[1,8]naphthyridin-4-
one (11). Polyphosphoric acid (25 g) was added to amino acid
10 (3.1 g, 16 mmol). The mixture was heated to 125 °C for 40
min. The mixture was cooled, and the PPA was slowly
decomposed by addition of ice and solid NaOH (exothermic!!).
Subsequently, the basic suspension was stored at 0 °C for
several hours and the precipitate was filtered, washed with
water, and dried. This afforded the ring-closed product as an
orange powder (2.1 g, 75%). An analytical sample was obtained
by column chromatography (5% MeOH in CH2Cl2) to yield a
yellow solid. Mp 142 °C; 1H NMR (CDCl3, 400 MHz) δ 6.24 (s,
1H, Ar-H), 6.21 (br s, 1H, NH), 3.50 (t, 2H, NCH2, J ) 6.5
Hz), 2.61 (t, 2H, CH2CO, J ) 5.0 Hz), 2.49 (s, 3H, 6-Me), 2.27
(s, 3H, 4-Me); 13C NMR (CDCl3, 75 MHz) δ 195.0, 162.6, 162.2,
152.8, 117.6, 110.3, 39.7, 39.0, 24.6, 22.6; HRMS for C10H12N2O-
[M + H] 177.1022, found 177.1013.
1
75 °C; H NMR (CDCl3, 300 MHz) δ 3.24 (t, 2H, NCH2, J )
5.5 Hz), 3.06 (s, 3H, N-Me), 2.66 (t, 2H, Ar-CH2, J ) 6.5 Hz),
2.46 (s, 3H, 6-Me), 2.22 (s, 3H, 4-Me), 1.95 (p, 2H, CH2, J )
5.4 Hz); 13C NMR (CDCl3, 75 MHz) δ 155.0, 152.3, 143.8, 114.9,
111.4, 50.0, 37.2, 26.2, 25.9, 22.2, 19.5; HRMS for C11H15BrN2
[M + H] 255.0491, found 255.0491.
5,7-Dimethyl-1,2,3,4-tetrahydro[1,8]naphthyridine (12).
Ketone 11 (2.1 g, 11.9 mmol) was dissolved in THF (10 mL).
To this BH3‚THF (1.0 M in THF, 32 mL, 32.0 mmol) was added
slowly. The reaction mixture was heated at reflux overnight.
After cooling, concentrated HCl was added to decompose the
amine-borane complexes. After the solution was stirred for
10 min at room temperature, an aq 15 M KOH solution was
added until basic pH was reached. The THF was removed
under reduced pressure and the residue was cooled in ice. The
solid was filtered, washed with water, and dried. The product
was obtained as a yellow solid (1.6 g, 83%). An analytical
sample was obtained by recrystallization from hexanes to give
2,4,8-Trimethyl-5,6,7,8-tetrahydro[1,8]naphthyridin-3-
ol (2). Bromide 14 (800 mg, 3.13 mmol) was dissolved in THF
(20 mL) and the solution was cooled to -78 °C. Then n-BuLi
(2.5 M in hexanes, 2.88 mL, 7.2 mmol) was added and the
yellow reaction mixture was stirred at -78 °C for 15-30 min
(the Br/Li exchange was monitored by TLC and was quantita-
tive). Dry 2,6-dimethylnitrobenzene (2.1 mL, 15.4 mmol) was
then added, resulting in a brown mixture. After being stirred
(37) Pessolane, A.; Witzel, B.; Graham, P.; Clark, P.; Jones, H. J.
Heterocycl. Chem. 1985, 22, 265.
(38) Henry, R. A.; Hammond, P. R. J. Heterocycl. Chem. 1977, 14,
1
white needles. Mp 114 °C; H NMR (CDCl3, 300 MHz) δ 6.25
1109.
(s, 1H, Ar-H), 5.02 (br s, 1H, NH), 3.35 (m, 2H, NCH2), 2.58
(t, 2H, Ar-CH2, J ) 6.4 Hz), 2.27 (s, 3H, 6-Me), 2.09 (s, 3H,
(39) Hamada, Y.; Takeuchi, I.; Sato, M. Yakugaku Zasshi 1974, 94,
1328.
J. Org. Chem, Vol. 69, No. 26, 2004 9221

Wijtmans et al.
for 2 h at -78 °C, the reaction mixture was warmed and
quenched with saturated aq NH4Cl. The THF layer was
separated and the aqueous layer extracted once with EtOAc.
The organic layers were dried (MgSO4), filtered, and concen-
trated. The crude product was purified by column chromatog-
raphy (1:1 hexanes:EtOAc) to afford a yellow solid (150 mg,
25%) of sufficient purity for further experiments. λmax (MeOH)
331 nm; mp 126-132 °C dec; 1H NMR (CDCl3 + drop of D2O,
300 MHz) δ 3.19 (br s, 2H, NCH2), 3.03 (br s, 3H, N-Me), 2.62
(br s, 2H, Ar-CH2), 2.33 (br s, 3H, 6-Me), 2.06 (br s, 3H, 4-Me),
1.96 (s, 2H, CH2); after standing overnight in CDCl3 and D2O
110.2, 77.6, 77.2, 40.9, 24.3, 20.1, 3.9; HRMS for C11H15N3 [M
+ Na] 212.1158, found 212.1177. Anal. Calcd for C11H15N3: C
69.81, H 7.99, N 22.20. Found: C 69.92, H 8.10, N 22.38.
2-[N-Acetyl-N-(1-pent-3-ynyl)amino]-4,6-dimethylpyr-
imidine (20). Amine 19 (3.0 g, 15.87 mmol) and DMAP (387
mg, 3.2 mmol) were heated overnight in Ac2O (30 mL) at ca.
100 °C. The reaction mixture was cooled and the solvent was
removed in vacuo. The residue was dissolved in EtOAc and
washed with saturated aq NaHCO3 solution and brine to
remove the bulk of DMAP. The organic layer was dried
(MgSO4), concentrated in vacuo, and dried under high vacuum
to yield a brown-orange solid (3.33 g, 91%) of sufficient purity
for further experiments. An analytical sample was obtained
by column chromatography (2:1 hexanes:EtOAc) followed by
recrystallization from hexanes to afford white crystals. Mp 63
°C; 1H NMR (CDCl3, 300 MHz) δ 6.78 (s, 1H, ArH), 4.12 (t,
2H, NCH2, J ) 7.6 Hz), 2.44 (m, 2H, CH2CtC), 2.42 (s, 6H,
Ar-CH3), 2.31 (s, 3H, COCH3), 1.65 (t, 3H, CtCCH3, J ) 2.5
Hz); 13C NMR (CDCl3, 75 MHz) δ 172.5, 168.5, 161.5, 116.7,
77.6, 77.2, 46.3, 25.9, 24.6, 19.3, 4.1; HRMS for C13H17N3O [M
+ Na] 254.1264, found 254.1274. Anal. Calcd for C13H17N3O:
C 67.51, H 7.41, N 18.17. Found: C 67.13, H 7.35, N 18.06.
1-Acetyl-4,6-dimethyl-2,3-dihydro-1H-pyrrolo[2,3-b]pyr-
idine (21). Acetamide 20 (4.0 g, 17.3 mmol) was heated at
reflux in diphenyl ether (50 mL) for 10 h. The solution was
cooled and ether (25 mL) was added. The resulting solution
was extracted with aq 1 M HCl (3 × 25 mL). The aqueous
layers were combined and carefully neutralized with solid
NaHCO3 until the pH was slightly basic. The suspension was
extracted 3× with CH2Cl2. MgSO4 and activated charcoal were
added to the combined organic layers and the mixture was
stirred for 10 min. After filtration, the filtrate was concentrated
under reduced pressure and dried under high vacuum. The
residue was recrystallized from hexanes/EtOAc to afford
orange crystals (2.69 g, 82%). Mp 105 °C; 1H NMR (CDCl3,
300 MHz) δ 6.53 (s, 1H, Ar-H), 4.07 (t, 2H, NCH2, J ) 8.7
Hz), 2.86 (t, 2H, CH2, J ) 8.7 Hz), 2.65 (s, 3H, COMe), 2.36 (s,
3H, 6-Me), 2.15 (s, 3H, 4-Me); 13C NMR (CDCl3, 75 MHz) δ
170.7, 155.7 (two overlapping C), 144.5, 121.7, 118.8, 45.9, 25.1,
24.4, 23.1, 18.6. HRMS for C11H14N2O [M + H] 191.1179, found
191.1176. Anal. Calcd for C11H14N2O: C 69.45, H 7.42, N 14.73.
Found: C 69.43, H 7.45, N 14.80.
4,6-Dimethyl-2,3-dihydro-1H-pyrrolo[2,3-b]pyridine (22).
Acetamide 21 (2.69 g, 14.1 mmol) was dissolved in MeOH (100
mL). Solid NaOH (2.27 g, 56.6 mmol) was added and the
mixture was heated at reflux overnight. Solid NH4Cl (3.03 g,
56.5 mmol) was added to neutralize the solution and then
MeOH was removed by evaporation. The resulting solids were
extracted with CH2Cl2 (3×). The combined organic extracts
were concentrated in vacuo and dried under high vacuum. The
product was obtained as an orange solid in quantitative yield
with satisfactory purity. Further purification could be achieved
by recrystallization from EtOAc, although the resulting orange
crystals retained traces of CH3COOH. Mp 150 °C; 1H NMR
(CDCl3, 300 MHz) δ 6.05 (s, 1H, Ar-H), 4.56 (br s, 1H, NH),
3.54 (t, 2H, NCH2, J ) 8.4 Hz), 2.91 (t, 2H, CH2, J ) 8.3 Hz),
2.28 (s, 3H, 6-Me), 2.09 (s, 3H, 4-Me); 13C NMR (CDCl3, 75
MHz) δ 164.4, 154.3, 142.4, 117.2, 114.2, 44.2, 26.2, 23.6, 18.2;
HRMS for C9H12N2 [M + H] 149.1073, found 149.1077. Anal.
Calcd for C9H12N2: C 72.94, H 8.16, N 18.90. Found: C 72.37,
H 8.13, N 18.81.
5-Bromo-4,6-dimethyl-2,3-dihydro-1H-pyrrolo[2,3-b]-
pyridine (23). This compound was prepared from amine 22
(0.92 g, 6.2 mmol), CH2Cl2 (25 mL), and 1,3-dibromo-5,5-
dimethylhydantoin (0.86 g, 3.1 mmol) following the same
procedure as for 13. After being quenched with Na2S2O3 and
KOH, the mixture was extracted with CH2Cl2 (3×). The
organic layers were dried (MgSO4), concentrated in vacuo, and
dried under high vacuum to yield a yellow solid. This solid
consisted of ∼90% product and ∼10% remaining starting
material (by NMR, 83% corrected yield). This compound proved
difficult to purify due to its extremely low solubility in any
1
under argon, the peaks sharpened; H NMR (CDCl3 + D2O,
300 MHz) δ 3.19 (t, 2H, NCH2, J ) 5.6 Hz), 3.03 (s, 3H, N-Me),
2.62 (t, 2H, Ar-CH2, J ) 6.6 Hz), 2.33 (s, 3H, 6-Me), 2.06 (s,
3H, 4-Me), 1.96 (p, 2H, CH2, J ) 6.4 Hz); 13C NMR (CDCl3 +
D2O, 75 MHz) δ 151.3, 140.1, 138.5, 132.4, 114.6, 50.1, 37.4,
24.9, 22.1, 18.9, 11.5; HRMS for C11H16N2O [M + H] 193.1335,
found 193.1345. The material was recrystallized from hot
hexanes to give a light brown solid, although the following
elemental analysis of this material actually suggested that the
purity was diminished after this. Anal. Calcd for C11H16N2O:
C 68.72, H 8.39, N 14.57. Found: C 67.78, H 8.41, N 13.92.
Hex-4-ynoic Acid (16). This compound was prepared
according to the literature from hex-5-ynoic acid, mp 100 °C.
NMR data have been reported.40
tert-Butyl Pent-3-ynylcarbamate (17). Acid 16 (7.7 g, 6.9
mmol) was dissolved in t-BuOH (100 mL). Diphenylphosphoryl
azide (16.3 mL, 7.6 mmol) and Et3N (9.57 mL, 6.9 mmol) were
added and the reaction mixture was heated at reflux for 30 h.
The solvent was evaporated and the residue dissolved in
EtOAc. This solution was extracted with 1 N aq HCl, water,
saturated aq NaHCO3, and brine. The organic layer was dried
(MgSO4) and concentrated in vacuo. This afforded 9.97 g (79%)
of a crude yellow oil that was sufficiently pure for the next
1
step. H NMR (CDCl3, 300 MHz) δ 4.82 (br s, 1H, NH), 3.20
(m, 2H, NCH2), 2.30 (m, 2H, CtCCH2), 1.76 (t, 3H, CtCMe,
J ) 2.5 Hz), 1.45 (s, 9H, t-Bu); 13C NMR (CDCl3, 75 MHz) δ
155.8, 79.3, 77.2, 76.3, 39.7, 28.4, 20.2, 3.5. HRMS for
C10H17NO2 [M + Na] 206.1151, found 206.1154.
Pent-3-ynylamine Hydrochloride (18). Crude BOC-
protected amine 17 (16.34 g, 8.9 mmol) was dissolved in
ethereal 2 M HCl (50 mL). The solution was stirred for 24 h,
during which a precipitate was formed. The crystals were
filtered off, washed with ether, and dried. This afforded the
ammonium salt (7.46 g, 70%) as white crystals. 1H NMR (D2O,
300 MHz) δ 3.07 (t, 2H, NCH2, J ) 6.5 Hz), 2.51 (m, 2H,
CH2≡C), 1.73 (t, 3H, Me, J ) 2.1 Hz); 13C NMR (D2O, 75 MHz)
δ 80.8, 74.1, 38.9, 17.3, 2.8. NMR data were in agreement with
a literature report.26
4,6-Dimethylpyrimidin-2-yl Trifluoromethanesulfonate
(25). This compound was prepared according to the literature
procedure31 from 4,6-dimethylpyrimidin-2-ol. It was obtained
as a brown oil that was >95% pure. NMR data have been
reported.31
4,6-Dimethyl-2-(1-pent-3-ynylamino)pyrimidine (19).
Triflate 25 (19.18 g, 74.9 mmol), hydrochloride 18 (7.46 g, 62.4
mmol), and Et3N (34.7 mL, 250 mmol) were stirred in freshly
distilled DMF (150 mL) at room temperature for 1 d. Water
was added to decompose the remaining triflate and the
mixture was extracted with ether (3×) and water. To the
organic layers were added MgSO4 and activated charcoal and
the mixture was stirred for 10 min. After filtration, the filtrate
was concentrated and dried under high vacuum to afford a
light brown solid of good purity (10.98 g, 93%). The product
was recrystallized from water/acetone to afford white crystals
(6.78 g, 57%). Mp 64 °C; 1H NMR (CDCl3, 300 MHz) δ 6.22 (s,
1H, ArH), 5.35 (br s, 1H, NH), 3.58 (q, 2H, NCH2, J ) 6.5 Hz),
2.42 (m, 2H, CH2CtC), 2.27 (s, 6H, Ar-CH3), 1.77 (t, 3H, Ct
CCH3, J ) 2.6 Hz); 13C NMR (CDCl3, 75 MHz) δ 167.8, 162.5,
(40) Carling, R. W.; Clark, J. S.; Holmes, A. B.; Sartor, D. J. Chem.
Soc., Perkin Trans. 1 1992, 95.
9222 J. Org. Chem., Vol. 69, No. 26, 2004

Synthesis of 6-Substituted-2,4-dimethyl-3-pyridinols
solvent. Therefore, it was used in the next step as the 90%
pure product. 1H NMR (d6-DMSO, 300 MHz) δ 4.60 (br s, 1H,
NH), 3.61 (t, 2H, NCH2, J ) 6.3 Hz), 2.97 (t, 2H, CH2, J ) 6.3
Hz), 2.42 (s, 3H, 6-Me), 2.19 (s, 3H, 4-Me); 13C NMR (d6-DMSO,
75 MHz) δ 164.1, 153.0, 141.9 121.4, 111.0, 45.0, 27.7, 26.3,
20.5; HRMS for C9H11BrN2 [M + H] 227.0178, found 227.0178.
5-Bromo-1,4,6-trimethyl-2,3-dihydro-1H-pyrrolo[2,3-b]-
pyridine (24). This compound was prepared from bromide
23 (1.8 g, 7.9 mmol) and a mixture of formic acid (40 mL), 37%
formaline (40 mL), and i-PrOH (40 mL) following the same
procedure as for 14. The crude product was purified by column
chromatography (CH2Cl2) to afford the product (1.7 g, 84%)
as a light-brown solid. Recrystallization from hexanes gave
calibration of the apparatus, the oxygen consumption in the
sample was measured from the differential pressure recorded
with time between the two channels. This instrumental setting
allowed us to have the N2 production and the oxygen con-
sumption derived from the azo-initiator decomposition already
subtracted from the measured reaction rates. Initiation rates,
Ri, were determined for each condition in preliminary experi-
ments by the inhibitor method with R-tocopherol as reference
antioxidant: Ri ) 2[R-tocopherol]/τ. For pyridinols 1c, 1d, 2,
and 3 giving a neat inhibition period in oxygen consumption
plots, the rate constant for reaction with peroxyl radicals was
obtained from the slope of the inhibition period itself, whose
length provided instead the stoichiometric factor n.3a,8 For
compounds 1a and 1b, the retarded oxygen consumption plots
were analyzed according to literature42 to obtain the value of
kinh, using the stoichiometric factor n ) 2 by analogy with the
faster pyridinols. Values for kinh values were averaged over 3
to 12 measurements with different experimental conditions
(e.g. concentration of the pyridinol).
1
analytically pure product as beige crystals. Mp ) 64 °C; H
NMR (CDCl3, 300 MHz) δ 3.42 (t, 2H, NCH2, J ) 7.9 Hz), 2.86
(s + t, 5H, NCH3 + ArCH2), 2.48 (s, 3H, 6-Me), 2.22 (s, 3H,
4-Me); 13C NMR (CDCl3, 75 MHz) δ 162.0, 153.5, 141.6, 120.9,
111.8, 53.0, 33.6, 25.7, 19.8; HRMS for C10H13BrN2 [M + H]
241.0334, found 241.0336.
1,4,6-Trimethyl-2,3-dihydro-1H-pyrrolo[2,3-b]pyridin-
5-ol (3). This compound was prepared from bromide 24 (1000
mg, 4.15 mmol), THF (30 mL), n-BuLi (2.5 M in hexanes, 3.8
mL, 9.5 mmol), and 2,6-dimethylnitrobenzene (2.26 mL, 16.6
mmol) following the same procedure as for 2. The crude
product was purified by careful column chromatography (5%
MeOH in CH2Cl2) under N2 to afford an orange solid (200 mg,
27%) with a purity >96% as judged from NMR. λmax (MeOH)
330 nm; mp 115 °C dec; 1H NMR (CDCl3, 300 MHz) δ 7.26 (br
s, 1H, OH), 3.30 (t, NCH2-ring, J ) 7.1 Hz), 2.80 (s + t, 5H,
NCH3 + ArCH2), 2.17 (s, 3H, 6-Me), 2.03 (s, 3H, 4-Me), D2O
shake removed the signal at 7.26 ppm but did not sharpen
signals much; 13C NMR (CDCl3, 75 MHz) δ 158.1, 142.1, 140.8,
134.0, 121.1, 54.0, 34.8, 25.3, 18.9, 13.2; HRMS for C10H14N2O
[M + H] 179.1179, found 179.1183. Anal. Calcd for C10H14N2O:
C 67.39, H 7.92, N 15.72. Found: C 66.46, H 7.81, N 15.28.
Autoxidation Measurements. Autoxidation experiments
were performed in a two-channel oxygen uptake apparatus,
based on a Validyne DP 15 differential pressure transducer
that has already been described elsewere.41 The entire ap-
paratus was immersed in a thermostated bath that ensured a
constant temperature within (0.1 °C. In a typical experiment,
an air-saturated chlorobenzene solution of styrene (4.3-8.6 M)
containing the pyridinol (1.1 × 10-6 to 1.0 × 10-4 M) was
equilibrated with the reference solution containing only an
excess of R-tocopherol (1 × 10-3 to 1 × 10-2 M) in the same
solvent at 30 °C. After equilibration, a concentrated chloro-
benzene solution of radical initiator 2,2′-azobis(2,4-dimethyl-
valeronitrile) (AMVN, final concentration 0.5-5 × 10-3 M) was
injected in both the reference and sample flasks and, following
Methyl Linoleate Autoxidation. A benzene solution that
contained methyl linoleate (0.2 M) and the antioxidant in
varying concentrations (0.001-0.1 M) was treated with radical
initiator 2,2′-azobis(4-methoxy-2,4-dimethylvaleronitrile) (MeO-
AMVN, final concentration ) 10 mM). The oxidation was run
at 37 °C for 4 h and then quenched by addition of BHT. The
major oxidation products 9-, 11-, and 13-hydroperoxide were
then quantified by normal-phase HPLC (0.5% i-PrOH in
hexanes) with UV detection (207 and 235 nm). Autoxidations
in the presence of 1d gave about four times less oxidation
products than identical oxidations in the presence of R-toco-
pherol. In the case of reactive pyridinols 2 and 3, no clear peaks
could be detected.
Acknowledgment. We thank Julia Philip and Cori
Bruce for assistance with the preparation of compounds
1a-c. M.W. thanks Prof. Binne Zwanenburg of the
University of Nijmegen for support. D.A.P. thanks
NSERC Canada and Vanderbilt University for their
support. This work was supported by NIH HL17921,
P30 ES00267, and CHE 9996188 and by the University
of Bologna and MIUR
Supporting Information Available: General experimen-
tal methods, procedure for pH titration, and 1H and 13C spectra
of final products and selected intermediates. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO048842U
(41) Amorati, R.; Pedulli, G. F.; Valgimigli, L.; Attanasi, O. A.;
Filippone, P.; Fiorucci, C.; Saladino, R. J. Chem. Soc., Perkin Trans. 2
2001, 2142.
(42) Chatgilialoglu, C.; Timokhin, V. I.; Zaborovskly, A. B.; Lutsyk,
D. S.; Prystansky, R. E. Chem. Commun. 1999, 405.
J. Org. Chem, Vol. 69, No. 26, 2004 9223