An efficient kilogram-scale synthesis of N,N2 -bis(4,6-disubstituted 1,3,5-triazin-2-yl)-4-aminophenetylamine

Article abstract of DOI:10.1021/op900202b

A practical large-scale synthesis was developed for the preparation of N,N2 -bis(4,6-disubstituted 1,3,5-triazin-2-yl)-4-aminophenethylamine derivatives exemplified by compound 2. This route is a sixstep procedure starting from commercially available cyan

Full text of DOI:10.1021/op900202b

Organic Process Research & Development 2009, 13, 1156–1160
An Efficient Kilogram-Scale Synthesis of N,N′-Bis(4,6-disubstituted
1,3,5-triazin-2-yl)-4-aminophenetylamine
Jean-Simon Duceppe, Vale´rie Perron, Jose´e Cloutier, Christopher Penney, and Boulos Zacharie*
ProMetic BioSciences Inc., 500 bouleVard Cartier Ouest, Bureau 150, LaVal, Que´bec, Canada H7V 5B7
Abstract:
A practical large-scale synthesis was developed for the preparation
of N,N′-bis(4,6-disubstituted 1,3,5-triazin-2-yl)-4-aminophenethyl-
amine derivatives exemplified by compound 2. This route is a six-
step procedure starting from commercially available cyanuric
chloride and 3-(tert-butoxycarbonylamino) aniline based on the
reactivity differences of the three chlorine atoms on the triazine
ring. The optimized procedure was scaled up to give the final
product 2 in an overall yield of 61% and 99.9% purity. These
low-molecular weight compounds mimic the ability of protein A
to bind to IgG antibody.
Introduction
Protein A (MW ) 42,000) is found on the surface of the
bacteria Staphylococcus aureus and binds with high affinity to
the tail portion of human and mouse antibodies.1 These binding
Process Research
properties make protein A commercially important for the
purification of monoclonal antibodies. However, purification
with protein A affinity absorbents is costly, and these columns
can leak protein A and contaminate purified antibodies. Few
synthetic compounds have been reported in the literature which
can replace protein A and then only when these molecules are
attached to solid-phase matrices.2 Protein A also has therapeutic
utility,3 but its toxicity and cost limit this use.4 As such, there
is a need for a novel, safe, nontoxic small-molecule mimetic
of protein A which is economical, stable, and can be admin-
istrated as a drug.5 We have identified a series of low-molecular
weight triazine compounds that mimic protein A which are
exemplified by the general structure 1.
Large amounts of various high-purity triazines 1 were
required for our therapeutic program. Consequently, there was
a need to develop a simple, robust general process for the
preparation of these compounds. Numerous routes were evalu-
ated for their synthesis on the basis of the reactivity differences
of the three leaving groups on the triazine ring. In the case of
cyanuric chloride, the substitution of chlorine can be controlled
by temperature to run in a stepwise manner. An empirical rule
is that monosubstitution of chlorine occurs below or at 0 °C,
disubstitution at room temperature, and trisubstitution above 60
°C.6 Thus, an efficient formation of compound 1 could be
considered using the order of substitution of chlorine atoms with
different amines. Our efforts resulted in the general synthesis
illustrated in Scheme 2 where compound 2 was selected as a
representative example of N,N′-bis(4,6-disubstituted 1,3,5-
triazin-2-yl)-4-aminophenethylamine derivatives. More than 50
analogues of this compound were synthesized using this
procedure. A retrosynthetic analysis of protected 2 (compound
7, Scheme 1) reveals a disconnection of compound 6 to give a
monosubstituted dichloro-triazine moiety derived from the
displacement of one chlorine atom in cyanuric chloride by
monoprotected aminoaniline. The other entity, alcohol 5, can
be seen as derived from substitution of the three chlorines in
cyanuric chloride by different nucleophiles in a stepwise
manner.
* Author for correspondence. E-mail: b.zacharie@prometic.com.
(1) (a) Langone, J. J. AdVances in Immunology; Dixon, F. J., Kunkel, H. G.,
Eds.; Academic Press: New York, 1982; Vol. 32, pp 157-252. (b)
Nilsson, B.; Abrahmse´n, L. Methods Enzymol. 1990, 185, 144–161.
(c) Moks, T.; Abrahmse´n, L.; Nilsson, B.; Hellman, U.; Sjoquist, J.;
Uhle´n, M. Eur. J. Biochem. 1986, 156, 637–643.
(2) (a) Li, R.; Dowd, V.; Stewart, D. J.; Burton, J.; Lowe, C. R. Nat.
Biotechnol. 1998, 16, 190–195. (b) Fern Teng, S.; Sproule, K.; Hussain,
A.; Lowe, C. R. J. Mol. Recognit. 1999, 12, 67–75. (c) Delano, W. L.;
Ultsch, M. H.; de Vos, A. M.; Wells, J. A. Science 2000, 287, 1279–
1283. (d) Ehrlich, G. K.; Bailon, P. J. Biochem. Biophys. Methods 2001,
49, 443–454. (e) Fassina, G.; Verdoliva, A.; Oderna, M. R.; Ruvo, M.;
Cassini, G. J. Mol. Recognit. 1996, 9, 564–569. (f) Johnson, C. P.;
Jensen, I. E.; Prakasam, A.; Vijayendran, R.; Leckband, D. Bioconjugate
Chem. 2003, 14, 974–978.
(3) Sasso, E. H.; Merrill, C.; Furst, D. E. Ther. Apheresis 2001, 5, 84–91.
(b) Brunner, J.; Kern, P. M.; Gaipl, U. S.; Voll, R. E.; Kalden, J. R.;
Wiesenhutter, C. W.; Hermann, M. Mod. Rheumatol. 2005, 15, 9–18.
(4) Silverman, G.; Goodyear, C. S.; Siegel, D. L. Transfusion 2005, 45,
274–280.
In order to supply interim amounts of 2 for biological testing,
the first batches were prepared on a gram scale. This work
(5) Zacharie, B.; Fortin, D.; Wilb, N.; Bienvenue, J.-F.; Asselin, M.; Grouix,
B.; Penney, C. Bioorg. Med. Chem. Lett. 2009, 19, 242–246.
(6) Blotny, G. Tetrahedron 2006, 62, 9507–9522.
1156
•
Vol. 13, No. 6, 2009 / Organic Process Research & Development
10.1021/op900202b CCC: $40.75  2009 American Chemical Society
Published on Web 10/01/2009

Scheme 1. Retrosynthetic analysis of compound 7
Scheme 2. Gram-scale synthesis of 2
revealed several limitations for the subsequent large-scale
preparation of this compound using the chemistry detailed in
Scheme 2. For example, selective monoprotection of the amine
of 1,3-phenylenediamine using di-tert-butyldicarbonate (Boc2O)
required an excess of the diamine (6 equiv), the product yield
was 15% mixed with unreacted material, and purification
necessitated the use of large amounts of solvent on silica gel.
The next step gave a (2:1) mixture of mono- and diaminoaniline
triazine products. After purification, the yield of monosubstituted
product 3 was only 45%. The third step was the introduction
of the ethanolamine substituent on the triazine ring in the
presence of base. This gave the expected chloro-triazine 4 in
55% yield contaminated with the corresponding diethanolamine
triazine byproduct. The formation of alcohol 5 was achieved
by reacting 4-aminophenethylamine with 4 in the presence of
organic base. This reaction gave product that required tedious
purification due to contaminants (1-5%) present in the ami-
nophenethylamine, including oxidized amines. This resulted in
Vol. 13, No. 6, 2009 / Organic Process Research & Development
•
1157

inconsistent product yields related to these impurities which
were carried downstream and complicated the subsequent steps.
The next step was reaction of triazine 3 and the alcohol
derivative 5 in the presence of base to form the dimer 6.
Although the desired product was formed in 75% yield, the
purity of 6 was only <95 wt % due to contaminants from the
former step. All attempts to crystallize the product did not
remove the impurities. The chloro-triazine 6 was then reacted
with 4-aminophenethylamine and triethylamine to afford the
penultimate 7. Again, the purification of this product was
difficult, and the yield (58%) and purity (<85% weight %) were
not acceptable. The last step was deprotection of the Boc groups
using HCl/dioxane to give the salt 2. In order to remove
remaining impurities, particularly the residual oxidized amine
sideproducts, 2 was suspended in methanol and heated at reflux
for several hours. This gave the final product 2 in 69% yield
and <97 wt % % purity.
However, this procedure was sufficient to produce six
batches of 0.04-0.5 g of product in overall yields of 10-15%
and 85-95 area % purity. The following safety, technical, and
operational issues were identified during this phase. The process
(1) gave low yields of moderate-purity product and high waste
loads, (2) entailed long cycle times, (3) required column
chromatography, (4) involved excessive handling operations,
(5) contained several operations of concentrations to dryness,
and (6) required the use of high-quality aminophenethylamine,
a relatively expensive reagent.
nucleophilic substitution by the amine and base-catalyzed
hydrolyses of the chlorine atom. This gave high yields (>97%)
of triazine 3. Better results were also obtained by lowering the
temperature of the reaction from 0-5 to -10 °C. These two
changes avoided the formation of the di(Boc-aminoaniline) side
product and exclusively gave the triazine 3 in high yield (99%)
and high purity (>99.7%).
Further improvements were achieved in the third step by
slowing the addition rate of ethanolamine to triazine 3 to a total
of 30 min at 0-5 °C. Again, the rate, the order of addition of
the reactants, as well as the control of the internal reaction
temperature played important roles for the achievement of high
yields of triazine 4. The internal reaction temperature was
maintained e10 °C to avoid the formation of the diethanolamine
triazine byproduct. In order to increase the yield, saturated
sodium bicarbonate was added last to the reaction mixture.
Chloro-triazine 4 was subsequently isolated in 90% yield after
workup by precipitation from TBME and heptane.
The third substitution on the triazine ring required the
addition of 4-aminophenethylamine. Immediate improvement
was achieved by the use of high purity (99.9%) 4-aminophen-
ethylamine hydrochloride in place of the free base which
contained numerous impurities. The hydrochloride salt is a
stable compound and easy to handle compared to the free base.
The salt was first added portionwise over 15 min, followed by
DIPEA, to a solution of chloro-triazine 4 in aqueous acetone.
A 2-fold excess of the salt was necessary to avoid formation
of the corresponding cross-linked side product formed between
chloro-triazine 4 and product 5. This modified procedure
produced the alcohol 5, isolated by precipitation from methylene
chloride in 83% yield.
In order to avoid the displacement of both chlorine atoms
in 3 by the 4-aminophenethylamine derivative 5, various
conditions were examined including varying the addition rate
of 3, the reaction temperature, and stoichiometry. The best result
for the formation of triazine 6 was achieved when both 3 and
5 were used in equimolar ratio. Thus, to a solution of alcohol
5 in tetrahydrofuran was added 3 in acetone/water over 10 min
followed by the addition of saturated sodium bicarbonate. After
workup and crystallization from ethyl acetate/heptane, com-
pound 6 was obtained in 94% yield and 99.9% area % purity.
The results were reasonably consistent from batch to batch on
a 0.5 kg scale.
Process Development
To address these aforementioned issues, process develop-
ment was performed prior to further triazine 2 scale-up.
Although 3-(tert-butoxycarbonylamino)aniline is a commercial
product, it is not readily available, and is relatively expensive.
We hoped we could prepare this from 1,3-phenylenediamine
and avoid an expensive starting material. For the scale-up of
compound 2, the best result was obtained when Boc2O was
dissolved in tetrahydrofuran and was added slowly to a solution
of phenylenediamine in methanol over one hour at -10 °C.
Under these conditions, the excess of 1,3-phenylenediamine
could be reduced from six to one equivalent which permitted
facile purification of the final product. The workup consisted
of solvent exchange to first tetrahydrofuran and then dichlo-
romethane, a simple filtration of the crude mixture (consisting
of product, traces of unreacted 1,3-phenylenediamine and
oxidized amines) on a silica gel pad, and recrystallization to
produce the protected 1,3-phenylenediamine in 64-65% yield.
The product yield of this step was increased over 4-fold, and
the volume of solvents used was reduced by 80%.
The next step was preparation of monosubstituted triazine
3. It has been reported2a that aniline reacted with cyanuric
chloride in the presence of bicarbonate at 0-5 °C (pH ) 5-7)
and gave 2-anilino-4,6-dichlorotriazine. In our hands, this
reaction tended to be problematic with regards to control of
the temperature, pH, and product yield (<50%) and gave a (2:
1) mixture of mono- and dianilino-triazine products. To address
these issues, the addition rate of aniline was reduced to ∼2 h,
and the base (saturated sodium bicarbonate) was added only at
the end of the reaction to minimize competition between
In the penultimate step, initial attempts using equimolar
amounts of 6 and 4-aminophenethylamine hydrochloride in the
presence of sodium bicarbonate failed to afford a good yield
of 7. On the other hand, using portionwise addition of a 2-fold
excess (to accelerate the reaction) of 4-aminophenethylamine
hydrochloride followed by triethylamine, triazine 7 was pro-
duced in 97% yield and 99.9 area % purity after refluxing for
one day. The excess of 4-aminophenethylamine was easily
removed by acidic extraction during workup. On a 0.7 kg scale,
a quantitative yield of product 7 was obtained. Deprotection of
the Boc group using 4 N HCl/dioxane produced the API, which
after trituration gave salt 2 in high purity (>99.9%) and high
yield (90%).
This improved process produced 15 < 30 g batches, 5 ×
250 g batches, and one 450 g batch of product 2 in an overall
1158
•
Vol. 13, No. 6, 2009 / Organic Process Research & Development

yield of 62% starting from protected 3-aminoaniline. Develop-
ment is in progress, and we anticipate to incorporate further
improvements in the upcoming larger scale campaign.
was obtained. This material was stirred at room temperature for
60 min (pH ) 2) and then basified to pH 8 by pouring onto a
saturated solution of sodium bicarbonate (10 L). The mixture was
stirred for 30 min. The precipitate was filtered, washed with water
(5 × 4 L), and dried under vacuum for 2 d to yield an amorphous
Conclusions
1
white solid (850.3 g, 99.6%). Mp ) 172 - 173 °C; H NMR
A practical large-scale procedure has been developed for the
preparation of N,N′-bis(4,6-disubstituted 1,3,5-triazin-2-yl)-4-
aminophenethylamine 2. The optimized seven-step procedure
uses inexpensive starting materials and eliminates several
nonscalable operations, and nonprocess-friendly solvents such
as ether and hexanes are no longer required. The process is
now well suited for further scale-up. As a result of this improved
synthesis, the overall product yield increased from 7% to 62%
(starting from protected 3-aminoaniline). Also, the total solvent
volumes were significantly reduced: organics by 70%, and
aqueous by 20%, although there is still significant latitude for
further solvent reduction. The compound 2 is currently under
evaluation by Laboratorios Dermatologicos, Mexico, as part of
a development program that targets novel drugs for the treatment
of dermatological diseases.
(CD3OD, 400 MHz) δ: 7.69-7.73 (m, 1H); 7.14-7.30 (m, 3H);
1.52 (s, 9H); 13C NMR (CD3OD, 100 MHz) δ: 170.32; 169.42;
164.52; 153.39; 140.72; 137.71; 129.53; 116.47; 115.78; 112.55;
79.78; 28.78 (3 carbons overlapped); LRMS (ESI): m/z 358 (MH+);
HPLC: tR ) 4.5 min (purity >99.2% at 210 nm and >99.3% at
254 nm).
2-(4-(3-(tert-Butoxycarbonylamino)phenylamino)-6-chloro-
1,3,5-triazin-2-ylamino)ethanol (4). The triazine 3 (835.0 g,
2.35 mol) was dissolved in THF (11.7 L) and cooled to 0-5
°C. A solution of ethanolamine (157.5 g, 2.59 mol) in acetone
(3.3 L) and water (3.3 L) was added portion-wise over 30 min
while maintaining internal temperature at 1-8 °C. A saturated
solution of sodium bicarbonate (668.0 mL) was then added in
one portion to the light-yellow mixture. The solution was stirred
at room temperature for 20 h, concentrated under vacuum to 3
L, and coevaporated with ethyl acetate (3 × 2 L). The solution
was diluted with ethyl acetate (6.7 L) and washed with water
and brine, dried over magnesium sulfate-charcoal-silica gel
for removal of unreacted starting material and color, filtered,
and concentrated to 1.3 L. The yellow solution was diluted with
TBME to 3.3 L and added dropwise (1 h) under nitrogen to
heptane (22.5 L). This produced a white solid which was
filtered, washed with heptane (2 × 800 mL), and dried. It was
then triturated with TBME/heptane (2:1 solution, 1.3 L), filtered,
and dried under vacuum to yield a white, amorphous solid
Experimental Section
All HPLC and mass spectra were recorded on a HP 1100
LC-MS Agilent instrument using a diode array detector at 210
and 254 nm. An analytical C18 column (75 mm × 4.6 mm, 5
µm) with a gradient of 15-99% acetonitrile/water containing
0.01% TFA and a flow of 2 mL/min (temperature 30 °C) was
used.
3-(tert-Butoxycarbonylamino)aniline. 1,3-Phenylenedi-
amine (400.0 g, 3.7 mol) was dissolved in methanol (16 L) at
-10 °C. A solution of Boc2O (768.0 g, 3.52 mol) in THF (4
L) was added over 24 h at -10 °C ( 4 °C. The mixture was
then stirred at room temperature for 20 h, diluted with THF
(500 mL), and concentrated under vacuum to 2 L. The dark-
brown solution was coevaporated with THF (5 × 1 L) and with
dichloromethane (5 × 1 L) and then diluted to 3.5 L with
dichloromethane. The solution was filtered through a silica gel
pad (20 × 30 cm2) and eluted with 35% ethyl acetate/heptane
in order to remove a black contaminant. The filtrate was
concentrated in vacuo to 1.0 L. The yellow solution was
coevaporated with TBME (5 × 1 L), diluted to 1.6 L with
TBME, and crystallized by the addition of heptane (6.6 L). The
material was collected by filtration, washed with heptane (2 ×
1 L), and dried under vacuum to yield a white solid (473.26 g,
64%). Mp ) 94-95 °C; 1H NMR (CDCl3, 400 MHz) δ: 7.12
(s, 1H); 7.06 (t, J ) 8.0 Hz, 1H); 6.91 (d, J ) 8.2 Hz, 1H);
6.80 (s, 1H); 6.63 (d, J ) 7.8 Hz, 1H); 5.96 (broad, 2H); 1.47
(s, 9H); 13C NMR (CDCl3, 100 MHz) δ:153.02; 146.28; 139.64;
129.96; 110.67; 109.61; 105.99; 80.66; 28.58 (3 carbons
overlapped); LRMS (ESI): m/z 209 (MH+), 231 (MNa+);
HPLC: tR ) 1.7 min (purity 100% at 210 and 254 nm).
2-[3-(tert-Butoxycarbonylamino)phenylamino]-4,6-dichloro-
1,3,5-triazine (3). Acetone (3.5 L) and water (3.5 L) were stirred
vigorously and cooled to -10 °C. Cyanuric chloride (442.1 g, 2.40
mol) was added portionwise over 15 min to obtain a homogeneous
suspension. A solution of mono-Boc-1,3-phenylenediamine (499.3
g, 2.4 mol) in acetone (3.5 L) was added dropwise over 2 h,
keeping the temperature between -5 to -10 °C. A white slurry
1
(798.33 g, 89.5%). Mp ) 157-159 °C; H NMR (CD3OD,
400 MHz) δ: 8.2 (broad, 1H); 7.76 (broad, 1H); 7.24-7.30
(m, 1H); 7.08-7.22 (m, 3H); 6.96 (broad, 1H); 3.72 (t, J )
5.6 Hz, 1H); 3.67 (t, J ) 5.6 Hz, 1H); 3.61 (broad, 1H);
3.46-3.52 (m, 1H); 1.51 (s, 9H); 13C NMR (CD3OD, 100 MHz)
δ: 169.37; 168.56; 166.36; 166.19; 164.12; 163.93; 154.06;
153.98; 139.85; 139.76; 139.62; 139.24; 139.15; 128.92; 128.68;
128.62; 115.23; 114.76; 113.88; 111.36; 111.03; 79.77; 79.64;
65.75; 60.51; 60.38; 43.28; 43.04; 27.79; 27.60 (3 carbons
overlapped); (mixture of rotamers); LRMS (ESI): m/z 381
(MH+), 403 (MNa+); HPLC: tR ) 3.3 min (purity 97% at 210
and 254 nm).
Diethanolamine triazine byproduct (2-(4-(3-(tert-butoxy-
carbonylamino)phenylamino)-6-(3-(tert-butoxycarbonylami-
no)phenylamino-1,3,5-triazin-2-ylamino)ethanol): 1H NMR
(DMSO-d6, 400 MHz) δ: 10.12 (broad, 2H); 9.21 (broad, 2H);
7.40-7.55 (m, 2H); 7.22-7.40 (m, 1H) 7.04-7.25 (m, 5H); 1.52
(s, 18H); LRMS (ESI): m/z 529 (MH+), 551 (MNa+); HPLC: tR
) 5.0 min.
2-(4-(4-Aminophenethylamino)-6-(3-(tert-butoxycarbony-
lamino)phenylamino)-1,3,5-triazin-2-ylamino)ethanol (5).
The chloro-triazine 4 (790.0 g, 2.07 mol) was dissolved in a
mixture of acetonitrile (15.8 L) and water (316 mL). 4-Ami-
nophenethylamine hydrochloride (862.82 g, 4.15 mol) was
added portionwise, followed by DIPEA (1.07 kg, 8.29 mol).
The mixture was then heated at 60 °C under nitrogen for 48 h.
The mixture was cooled and filtered on fiberglass. The filtrate
Vol. 13, No. 6, 2009 / Organic Process Research & Development
•
1159

was concentrated to 3 L, diluted with water (3 L), and
concentrated again to 3 L to remove acetonitrile. This solution
was diluted to 5 L with water and extracted with ethyl acetate
(7.4 L). The organic layer was separated and washed with brine
and then extracted with 2 N hydrochloric acid (5.5 L). This
acidic phase was washed with ethyl acetate and basified to pH
8 with solid sodium bicarbonate. A residue was obtained which
was redissolved in ethyl acetate (7.3 L), washed with brine,
treated with magnesium sulfate-charcoal, filtered on Celite, and
concentrated to 2.8 L. This mixture was added dropwise to
heptane (18 L) under nitrogen. A gummy, brown solid was
obtained and separated by decantation. Methylene chloride (3
L) was added, and the resulting mixture was slowly stirred
overnight at 20-25 °C. The fluffy solid was filtered off, washed
with methylene chloride (1.5 L), and dried under vacuum to
yield a beige amorphous solid (826.5 g, 83%). Mp ) 154-155
°C; 1H NMR (CD3OD, 400 MHz) δ: 7.26 (broad, 1H); 7.13 (t,
J ) 7.8 Hz, 1 H); 6.98 (d, J ) 8.2 Hz, 4H); 6.68 (d, J ) 2.5
Hz, 2H); 3.68 (broad, 2 H); 3.52 (broad, 4H); 2.72 (t, J ) 7.4
Hz, 2H); 1.49 (s, 9H); 13C NMR (CD3OD, 100 MHz) δ: 154.06;
145.36; 139.61; 129.47; 129.31 (3 carbons overlapped); 128.52;
115.85 (4 carbons overlapped); 115.21; 114.57; 112.81; 110.90;
79.54; 61.32; 42.83; 42.43; 35.11; 27.55 (3 carbons overlapped);
LRMS (ESI): m/z 481 (MH+), 503 (MNa+); HPLC: tR ) 1.8
min (purity 100% at 210 and 254 nm).
2-(4-(3-(tert-Butoxycarbonylamino)phenylamino)-6-(4-(4-
(3-(tert-butoxycarbonylamino)phenylamino)-6-chloro-1,3,5-
triazin-2-ylamino)phenethylamino)-1,3,5-triazin-2-ylamino)-
ethanol (6). The alcohol 5 (820.0 g, 1.71 mol) was dissolved in
THF (10.9 L), and a solution of triazine 3 (607.7 g, 1.71 mol) in
acetone (2.9 L) and water (2.9 L) was added over 10 min followed
by a saturated solution of sodium bicarbonate (2.9 L). The mixture
was stirred at 20-25 °C for 24 h and then concentrated to 5 L to
remove volatiles. The solution was diluted with water (2 L) and
concentrated to 4 L. The brown solution was then extracted with
ethyl acetate (9.0 L), washed with water and brine, and treated
with magnesium sulfate-charcoal. The solution was then filtered
and concentrated to 3.0 L. This mixture was added dropwise under
nitrogen to heptane (18.0 L) with stirring. The precipitated beige
solid was filtered, washed with heptane (2 × 1 L), and dried under
vacuum to yield a beige amorphous solid (1.29 kg; 94.7%). Mp
) 146-149 °C; 1H NMR (CD3OD, 400 MHz) δ: 7.75-7.72 (m,
3H); 6.88-7.32 (m, 9H); 3.68 (broad, 2H); 3.60 (broad, 2H); 3.54
(broad, 2H); 2.84 (t, J ) 7.2 Hz, 2H); 1.49 (s, 18H); 13C NMR
(CD3OD, 100 MHz) δ: 171.83; 165.94; 164.52 (2 carbons
overlapped); 154.02; 140.50; 139.55 (2 carbons overlapped);
136.57; 135.22; 128.96 (3 carbons overlapped); 128.71 (3 carbons
overlapped); 128.50; 120.94 (3 carbons overlapped); 114.54; 112.85
(2 carbons overlapped); 110.90; 79.85; 79.57; 61.35; 60.38; 42.85;
42.11; 35.35; 27.59 (3 carbons overlapped); 27.58 (3 carbons
overlapped);19.71; 13.30; LRMS (ESI): m/z 800 (MH+), 822
(MNa+); HPLC: tR ) 3.3 min (purity 100% at 210 and 254 nm).
2-(4-(4-(4-(4-Aminophenethylamino)-6-(3-(tert-butoxycar-
bonylamino)phenylamino)-1,3,5-triazin-2-ylamino)phenyl-
ethylamino)-6-(3-(tert-butoxycarbonylamino)phenylamino)-
1,3,5-triazin-2-ylamino)ethanol (7). The triazine 6 (1.1 kg, 1.39
mol) in acetonitrile (15.5 L) and water (465 mL) was treated with
4-aminophenethylamine hydrochloride (0.8 kg, 3.84 mol) portion-
wise, followed by the addition of triethylamine (0.7 kg, 6.82 mol).
The mixture was then heated at 60 °C for 24 h. The solvent was
concentrated under vacuum to 3.0 L, coevaporated with ethyl
acetate (3 × 2 L), and diluted with ethyl acetate to 8.0 L. The
solution was washed with water and brine, treated with magnesium
sulfate-charcoal, and filtered. The filtrate was acidified with 2 N
hydrochloric acid (1.5 L) to afford a brown, oily residue which
was separated with the aqueous phase from the organic layer. The
aqueous mixture was basified to pH 8 with solid sodium bicarbon-
ate. The solid was separated, dissolved in ethyl acetate (9 L), treated
with magnesium sulfate-charcoal, filtered, and concentrated to 3.5
L under vacuum. This solution was added dropwise to heptane
(18.6 L) to form a solid. The precipitate was filtered, washed with
heptane (2 × 750 mL), and dried under vacuum to yield a beige
amorphous solid (1.24 kg, 97%). Mp ) 104-107 °C; 1H NMR
(CD3OD, 400 MHz) δ: 7.48-8.14 (m, 4H); 7.06-7.34 (m, 7H);
6.90-7.06 (m, 3H); 6.66 (d, J ) 8.2 Hz, 2H); 3.68 (broad, 2H);
3.53 (broad, 6H); 2.82 (broad, 2H); 2.74 (t, J ) 7.6 Hz, 2H); 1.48
(s,18H); 13C NMR (CD3OD, 100 MHz) δ: 171.13; 166.07; 165.99;
165.63; 165.00; 164.66; 164.33; 154.08; 154.02; 145.14; 140.43;
140.19; 139.60; 139.51; 138.11; 137.73; 133.91; 133.53; 129.60;
129.47; 129.35 (2 carbons overlapped); 129.23; 128.82 (2 carbons
overlapped); 128.58; 120.89; 120.27; 115.94 (2 carbons over-
lapped); 115.05; 114.67; 112.95; 111.31; 110.93; 79.61; 61.30;
60.38; 42.91; 42.56; 42.20; 36.25; 35.15; 34.67; 34.60; 34.56;
34.51; 31.58; 27.60 (6 carbons overlapped); 27.59 (3 carbons
overlapped); 25.13; 25.03; 22.53; 21.83; 19.86; 19.76; 19.71; 17.98;
13.30; 13.29; 10.62. (mixture of rotamers); LRMS (ESI): m/z 901
(MH+) 923 (MNa+); HPLC: tR ) 2.7 min (purity 100% at 210
and 254 nm).
2-(4-(4-(4-(4-Aminophenethylamino)-6-(3-aminopheny-
lamino)-1,3,5-triazin-2-ylamino)phenylethylamino)-6-(3-ami-
nophenylamino)-1,3,5-triazin-2-ylamino)ethanol pentahy-
drochloride salt (2). The protected triazine 7 (1.24 kg, 0.7 mol)
was dissolved in a mixture of 4 N hydrochloric acid/dioxane (7.9
L) and water (0.73 L). This mixture was stirred at 20-25 °C for
60 min, diluted with acetonitrile (23 L), and stirred 15 min to afford
a beige solid. The material was filtered and washed with methanol
(12 L), tetrahydrofuran (6 L), and TBME (6 L) and then dried for
24 h. The solid was dissolved in water (58 L). The clear, pale-
yellow solution was filtered on a 0.22 µm membrane, frozen and
lyophilized. A light beige fluffy solid was obtained (1.11 kg, 90%).
1
Mp ) 220-222 °C; H NMR (D2O, 400 MHz) δ: 6.80-7.80
(m, 16H); 3.00-3.70 (m, 8H); 2.60 (broad, 4H); 13C NMR
(CD3OD, 100 MHz) δ: 161.27; 155.04; 154.22; 151.99; 139.28;
137.88; 137.48; 135.47; 134.27; 130.26 (8 carbons overlapped);
128.92 (3 carbons overlapped); 128.36; 123.14 (3 carbons over-
lapped); 120.17; 118.91; 114.64 (2 carbons overlapped); 59.53;
42.95; 42.43 (2 carbons overlapped); 34.28 (2 carbons overlapped);
LRMS (ESI): m/z 700 (MH+) 350 ((M + 2 H+)/2); HPLC: tR )
3.6 min (purity 100% at 210 and 254 nm).
Acknowledgment
We thank Shaun D. Abbott for reading the manuscript and
Joe¨l Lechasseur for technical assistance.
Received for review August 3, 2009.
OP900202B
1160
•
Vol. 13, No. 6, 2009 / Organic Process Research & Development