Synthesis of an antibacterial compound containing a 1,4-substituted 10-1,2,3-triazole: A scaleable alternative to the click reaction

Article abstract of DOI:10.1021/op900252a

The copper-catalyzed click reaction of an azide with an alkyne has become a popular method to build up 1,4-substituted 1H-1,2,3- triazoles in medicinal chemistry and this approach was used on a laboratory scale during the preparation of novel macrolide 1. However, the manufacture of the key azide component, as well as its subsequent use in the presence of a copper catalyst on a large scale, was associated with potential safety concerns. Therefore, a sequence was developed in which construction of the 1,4- substituted 1H-1,2,3-triazole in 1 was accomplished via cyclocon - densation of an a,a-dichloro tosyl hydrazone with an amine.

Full text of DOI:10.1021/op900252a

Organic Process Research & Development 2010, 14, 152–158
Synthesis of an Antibacterial Compound Containing a 1,4-Substituted
1H-1,2,3-Triazole: A Scaleable Alternative to the “Click” Reaction
Roger Hanselmann,* Gabriel E. Job, Graham Johnson, Rongliang Lou, Jacek G. Martynow, and Maxwell M. Reeve
Rib-X Pharmaceuticals, Inc., 300 George Street, Suite 301, New HaVen, Connecticut 06511, U.S.A.
Abstract:
The copper-catalyzed “click” reaction of an azide with an alkyne
has become a popular method to build up 1,4-substituted 1H-1,2,3-
triazoles in medicinal chemistry and this approach was used on a
laboratory scale during the preparation of novel macrolide 1.
However, the manufacture of the key azide component, as well as
its subsequent use in the presence of a copper catalyst on a large
scale, was associated with potential safety concerns. Therefore, a
sequence was developed in which construction of the 1,4-
substituted 1H-1,2,3-triazole in 1 was accomplished Wia cyclocon-
densation of an r,r-dichloro tosyl hydrazone with an amine.
The discovery route utilized the copper-catalyzed Huisgen
cycloaddition reaction, popularized by Sharpless4 and Meldal,5
as the key transformation to build up the triazole ring in the
synthesis of 1 (Scheme 1). This high-yielding and mild reaction
was well suited for the rapid exploration of a range of novel
substituted macrolides through the attachment of different side
chains to an N-alkynylated macrolide core such as 5.6 However,
there exists the potential for formation of highly explosive
copper azide7 during the course of the catalyzed cycloaddition
reaction.8 Additionally the required azide 3 was synthesized
from 2 by a diazo transfer reaction using a large excess of triflic
azide, a highly energetic and hazardous chemical.9 In order to
avoid these concerns we elected to investigate formation of the
triazole by alternative means.10 Attempting to perform the
triazole formation without any copper catalyst gave a slow
reaction, yielding a 2:1 ratio of the desired 1,4-regioisomer to
its 1,5-regioisomer.11
Introduction
Antimicrobial resistance in the community and hospital
settings has been a growing public health concern due to the
continuing emergence of multidrug-resistant bacterial strains.1
A wide range of pathogens has been seen in hospital and
community settings, including those associated with nosocomial
Gram-positive infections, community respiratory-tract infections,
and uncomplicated skin and soft tissue infections. Macrolides
are a class of very effective antibiotics in those settings and
they have traditionally shown excellent safety profiles.2 Unfor-
tunately, resistance is an emerging problem in this class of
antibiotics. In order to combat this, structural enhancement of
naturally occurring macrolides has proven to be an effective
strategy.3
We recently required an efficient scaleable synthesis of a
novel macrolide 1, which displays promising antibacterial
activity. As a key structural element, 1 contains a 1H-1,2,3-
triazole fragment connected Via a two-carbon linker to the
desosamine nitrogen of clarithromycin.
In searching for safer alternative methods to scale-up the
preparation of 1 reliably, we became interested in synthesizing
the triazole by a method developed by Sakai more than 20 years
(4) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew.
Chem., Int. Ed. 2002, 41, 2596.
* To whom correspondence should be addressed. E-mail: rhanselmann@
Rib-x.com.
(5) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67,
3057.
(1) (a) Cosgrove, S. E.; Carmeli, Y. Clin. Infect. Dis. 2003, 36, 1433. (b)
Seybold, U.; Kourbatova, E. V.; Johnson, J. G.; Halvosa, S. J.; Wang,
Y. F.; King, M. D.; Ray, S. M.; Blumberg, H. M. Clin. Infect. Dis.
2006, 42, 647. (c) Tenover, F. C.; McDougal, L. K.; Goering, R. V.;
Killgore, G.; Projan, S. J.; Patel, J. B.; Dunman, P. M. J. Clin.
Microbiol. 2006, 44, 108.
(6) Bhattacharjee, A.; Du, Y.; Duffy, E. M.; Job, G. E.; Lou, R.; Luo, Z.;
O’Dowd, H.; Tang, Y.; Wu, Y. PCT Int. Appl.WO 2008106224, 2008,
CAN: 149,332567.
(7) Patnaik P. A ComprehensiVe Guide to the Hazardous Properties of
Chemicals, 3rd ed.; John Wiley: Chichester, 2007; p 620.
(8) For example CuN3 could be formed by trace decomposition of 4.
(9) (a) Cavender, C. J.; Shiner, V. J., Jr. J. Org. Chem. 1972, 37, 3567.
(b) Nyffeler, P. T.; Liang, C.-H.; Koeller, K. M.; Wong, C. H. J. Am.
Chem. Soc. 2002, 124, 10773.
(2) Omura, S. Macrolide Antibiotics. Chemistry, Biology, and Practice,
2nd ed.; Academic Press Inc.: New York, 2002.
(3) Kanyo, Z.; Bhattacharjee, A.; Chen, S.; Chen, Y.; Dalton, J.; Devito,
J.; Farmer, J.; Franceschi, F.; Goldberg, J.; Hanselmann, R.; Ippolito,
J.; Johnson, G.; Lawrence, L.; Lou, R.; McConnell, T.; Orbin, A.;
Oyelere, A.; Park, M.; Salvino, J.; Sherer, E.; Sutcliffe, J.; Tang, Y.;
Wang, D.; Wu, Y.; Duffy, E. Enhanced Macrolides: Overcoming
Resistance by Improving Target Affinity, from 49th Interscience
Conference on Antimicrobial Agents and Chemotherapy, San Fran-
cisco, CA, September 12-15, 2009, abstract F1-2050.
(10) Applying alternative technologies such as microreactor technology to
the hazardous click reaction is a viable alternative which was excluded
due to time constrains. For a special issue about microreactor
technology see: Special Feature Section: Organic Synthesis by
Microprocess Technology, Hessel, V., Besser, R., Guest Editors Org.
Process Res. DeV. 2009, 13, 940-1031.
(11) A 1:1 mixture of 4 and 5 was refluxed in toluene for 5 days.
152
•
Vol. 14, No. 1, 2010 / Organic Process Research & Development
10.1021/op900252a  2010 American Chemical Society
Published on Web 11/23/2009

214 (Scheme 4). In the discovery route, there were no concerns
with the selectivity in the O-methylation step because the amino
group was masked as an azide. In the absence of protection of
the amino group we were concerned with concomitant N-
methylation of amino alcohol 2. However, an initial attempt at
methylating the amino alcohol 2 using KHMDS (1.2 equiv) in
THF and methyl iodide gave less than 10% of N-methylated
impurities.15 The reaction proceeded cleanly up to ∼90%
conversion with only small amounts of N-methylation. After-
wards, the remaining starting material was slowly converted
almost entirely to the N-methylated impurities. This observation
implied the importance of an excess of base. Thus, increasing
the amount of KHMDS to 1.35 equiv suppressed the N-
methylation below 1%, and 12 was reproducibly synthesized
in 98% yield. On initial scale up (2.35 kg of 2) the reaction
stalled after 95% conversion. Fortunately this tendency towards
incomplete conversion was eliminated in subsequent batches
of similar sizes by increasing the amount of methyl iodide to
1.36 equiv and KHMDS to 1.49 equiv and optionally adding
10% more of these reagents when the reaction stalled.
The Suzuki reaction to generate 11 was carried out without
protection of the amine group in 12 and low catalyst loading
using K3PO4 as base. In initial runs, K2CO3 was used as the
base. However, during a laboratory scale-up run a new impurity
13 was detected at up to 3% based on HPLC area %. This
impurity was most likely formed by carbamic acid formation
of the liberated CO2 with 11/12, followed by ring closure.
Switching the base to K3PO4 eliminated this impurity entirely,
and two batches (2 × 2.46 kg of 12) were converted quanti-
tatively to 11.
Scheme 1. Discovery Synthesis of 1
Scheme 2. Sakai’s Tosyl Hydrazone Cyclization
ago.12 In this approach, an amine is condensed with R,R-dichloro
tosyl hydrazone to form a 1,4-substituted triazole under mild
reaction conditions (Scheme 2).
However, this methodology has been used infrequently13 and
then only with unbranched amines and simple R,R-dichloro
tosyl hydrazones. The advantage of avoiding the copper-
catalyzed click reaction and using a tosyl hydrazone and an
amine instead of an energetic azide to form the triazole ring in
1 was very compelling, and we decided to investigate the
feasibility of this approach further.
Synthesis of Triazole 17. The ꢀ-keto ester 15 was synthe-
sized from dichloroacetyl chloride 14 by adapting a method
developed by workers at Parke-Davis16 (Scheme 5). This gave
15 reproducibly in 81% yield on a 1 kg scale of 14.
Next ꢀ-keto ester 15 was converted to tosyl hydrazone 10
by treatment with tosyl hydrazine. Propionic acid is the
commonly used solvent for this transformation.17 However, we
wanted to avoid this solvent due to the desire to telescope 10
into the triazole formation, which is performed under basic
conditions. Several reaction solvents were investigated (Table
1), but in all of them various amounts of the byproduct 16 were
formed. Fortunately, in MeCN this impurity precipitated out
and was easily purged by a simple filtration after the reaction
was complete, yielding 10 in 80% yield.
Results and Discussion
In initial laboratory scale reactions for the triazole formation,
the supernatant containing 10 was telescoped into a solution of
We envisioned synthesizing the required triazole 9 through
a cyclocondensation of R,R-dichloro tosyl hydrazone 10 with
amine 11. The active pharmaceutical ingredient (API) 1 could
then be obtained through a late-stage alkylation of the side-
chain 9 with N-desmethylclarithromycin 8 (Scheme 3).
Synthesis of 2-Aminopyrimidine 11. The synthesis of 11
commenced with the selective methylation of amino alcohol
(14) Compound 2 was synthezised by a vendor according to the following
references: (a) Bhattacharjee, A.; Du, Y.; Duffy, E. M.; Job, G. E.;
Lou, R.; Luo, Z.; O’Dowd, H.; Tang, Y.; Wu, Y. PCT Int. Appl. WO
2008106224, 2008; CAN: 149, 332567. (b) Schumacher, D. P.; Clark,
J. E.; Murphy, B. L. Eur. Pat. Appl. EP 359516, 1990, CAN: 113,
115291. (c) Kim, D. W.; Ahn, D.-S.; Oh, Y.-H.; Lee, S.; Kil, H. S.;
Oh, S. J.; Lee, S. J.; Kim, J. S.; Ryu, J. S.; Moon, D. H.; Chi, D. Y.
J. Am. Chem. Soc. 2006, 128, 16394.
(15) For related methylations see: (a) Meyers, A. I.; Poindexter, G. S.; Brich,
Z. J. Org. Chem. 1978, 43, 892. (b) Pyne, S. G. J. Org. Chem. 1986,
51, 81. (c) Glennon, R. A.; Bondarev, M. L.; Khorana, N.; Young,
R.; May, J. A.; Hellberg, M. R.; McLaughlin, M. A.; Sharif, N. A.
J. Med. Chem. 2004, 47, 6034.
(12) (a) Sakai, K.; Tsunemoto, D.; Kobori, T.; Kondo, K.; Hida, N. Eur.
Pat. Appl. EP 103840, 1984. CAN: 101, 72730. (b) Sakai, K.; Hida,
N.; Kondo, K. Bull. Chem. Soc. Jpn. 1986, 59, 179.
(13) (a) Harada, K.; Oda, M.; Matsushita, A.; Shirai, M. Heterocycles 1998,
48, 695. (b) Reck, F.; Zhou, F.; Girardot, M.; Kern, G.; Eyermann,
C. J.; Hales, N. J.; Ramsay, R. R.; Gravestock, M. B. J. Med. Chem.
2005, 48, 499.
(16) Clay, R. J.; Collom, T. A.; Karrick, G. L.; Wemple, J. Synthesis 1993,
290.
(17) Bott, K. Chem. Ber. 1975, 108, 402.
Vol. 14, No. 1, 2010 / Organic Process Research & Development
•
153

Scheme 3. Retrosynthesis of 1
Scheme 4. Synthesis of 11
reaction which is visible as a tarry residue in 17 after aqueous
workup. Recrystallization of crude 17 gave pure material, albeit
in low yield. Fortunately, the tarry residue can rather be easily
purged by incorporating an acid-base extraction during the
workup of 17, affording material in 80% yield and of ∼93%
purity by HPLC.
While scaling up this sequence, the impurity profile of 10
and 17 became progressively worse on a 3.9 kg scale of 15.
Preliminary safety assessment of 10 by DSC showed a
decomposition at 116-145 °C with a decomposition energy
of 135 J/g. Thus, 10 was deemed to be relatively safe to be
handled as a solid.18 Instead of telescoping the MeCN solution
of 10, it was decided to isolate 10 by precipitation from the
partially concentrated crude solution at -15 °C. This approach
efficiently purged most impurities and 10 was obtained in a
somewhat diminished yield of 62%. The damp crystals were
redissolved in MeCN and used for the triazole formation,
yielding 3.1 kg of 17 in 80% yield.
Scheme 5. Synthesis of 17
Sakai proposed a general mechanism for the triazole forma-
tion (Scheme 6, 19-21).19 In this, R,R-dichlorotosylhydrazone
18 is deprotonated to vinyldiazine 19. Subsequent amine
addition and toluenesulfinic acid elimination forms 20, which
cyclizes after deprotonation to the triazole 21. During optimiza-
tion of this reaction, we observed two transient compounds by
HPLC. Analysis of those intermediates by LC-MS gave
molecular weights consistent with those of 22 and 23. This result
suggests a slightly different mechanistic picture in which the
amine adds to the vinyldiazine 19, but the tosyl group is retained
until later in the reaction sequence (Scheme 6, 22-21).
Reduction of Ester 17. Several reducing reagents were
screened for the reduction of ester 17 to alcohol 24 (Table 2).
However, all resulted in various amounts of the over-reduced
dihydroaminopyrimidine 25 (Scheme 7). Preliminary attempts
at oxidizing the over-reduced impurity back to 24 were
successful, but unfortunately the isolated alcohol 24 contained
new unidentified impurities which were difficult to purge.20 Of
the reducing reagents investigated, the combination of
CaCl2-NaBH421 or MgCl2-NaBH422 gave the lowest amount
of over-reduction.
Table 1. Solvent effect on the ratio of 10 vs 16
ratio 10:16
solvent
crude
filtrate
yield (%)b
propionic acid
EtOAc
9:1
5:1
6:1
5:1
1:1
N/Aa
6:1
>99:1
>99:1
3:1
N/A
90
toluene
56
MeCN
MeOH
80
(18) Rowe, S. M. Org. Process Res. DeV 2002, 6, 877.
(19) Sakai, K.; Hida, N.; Kondo, K. Bull. Chem. Soc. Jpn. 1986, 59, 179.
(20) Chloramine-T, NCS, and H2O2 were successfully used for the oxidation
of 25.
59
a No precipitation was observed. b Solution yields of filtrate determined by
HPLC.
(21) (a) Brown, H. C.; Narasimhan, S.; Choi, Y. M. J. Org. Chem. 1982,
47, 4702. (b) Clark, J. E.; Fischer, P. A.; Schumacher, D. P. Synthesis
1991, 891.
11 in EtOH at 10-25 °C. An excess of 0.6 equiv of 10 is
required for this transformation to go to completion. The excess
of 10 is converted to a soluble polymeric residue during the
(22) (a) Brown, H. C.; Mead, E. J.; Subba Rao, B. C. J. Am. Chem. Soc.
1955, 77, 6209. For an example with KBH4-MgCl2 see: (b) Qiu, Y.-
C.; Zhang, F.-L.; Zhang, C.-N. Tetrahedron Lett. 2007, 48, 7595.
154
•
Vol. 14, No. 1, 2010 / Organic Process Research & Development

Scheme 6. Proposed mechanism of triazole formation
Tanabe’s procedure24 in which the tosyl chloride is activated
with trimethyl amine25 as the catalyst worked especially well,
yielding 9 quantitatively on a 0.9 kg scale of 24 (Scheme 8).
N-Desmethylation of Clarithromycin 27. The N-des-
methylation of the commercially available clarithromycin 27
was accomplished by a modification of a procedure developed
by Freiberg.26 Iodine is added in portion to a solution of 27
and NaOAc in refluxing aqueous MeOH at pH 8. Of note is
the importance of slowly distilling off some of the solvent to
remove volatiles from the reaction mixture. If this is omitted,
the reaction stalls at ∼50% conversion. Several 2 kg batches
of 27 were converted to 8 in this way with an average yield of
62%.
Table 2. Effect of reducing reagents
reagent
solvent
ratio 24:25
LiBH4
THF/MeOH, 1:1
THF
7:3
7:3
KBH4
DIBAL
CH2Cl2
>1:99
NaBH4-CaCl2
NaBH4-MgCl2
NaBH4-CeCl3
MeOH
9:1
MeOH
9:1
MeOH
6:4
Scheme 7. Reduction of 17
Alkylation of Macrolide 8. The alkylation of macrolide 8
with tosylate 9 was conducted in MeCN with K2CO3 as base.
Self-polymerization of 9 was a justified concern and was
observed to an extent of ∼10%. As a result, unreacted macrolide
8 remained after the reaction was complete. Fortunately, a short-
path silica gel plug followed by a recrystallization from MeCN
removed the residual unalkylated macrolide 8. When 1.2 kg of
9 was converted, the final product 1 was obtained in 61% yield
and 98.2% purity by HPLC.
In conclusion, safety concerns prevented the use of the
popular Sharpless and Meldal copper-catalyzed azide-alkyne
cycloaddition to synthesize 1. This limitation was successfully
overcome by applying Sakai’s R,R-dichloro tosyl hydrazone
approach to synthesize 3 kg of 1. The main challenge to further
scale-up consists of eliminating several concentrations of
product solutions to oils. We anticipate that simple solvent
exchanges and telescoping the resulting solutions into the
subsequent step will allow a straightforward entry into the plant.
The combination of MgCl2 and NaBH4 was chosen for scale-
up due to its faster and cleaner reaction profile. Further
optimization revealed the importance of the workup. When the
reaction is quenched directly with an aqueous citric acid
solution, the impurity 25 rapidly increases, and up to an
additional 20% of this impurity was seen. Thus, the aqueous
acidic quench did not effectively destroy any excess of
borohydride species but rather accelerated the over-reduction.
To avoid this over-reduction during workup, an initial acetone
quench was incorporated, which completely consumed any
active borohydride species before the aqueous acidic workup.
After further optimization, the reduction was cleanly ac-
complished with a maximum over-reduction of 5%. The mild
citric acid wash removed small amounts of byproduct 25, and
alcohol 24 was isolated in 76% yield and >98% purity.
Synthesis of Tosylate 9. Activation of 24 was best ac-
complished using a tosyl group. Other leaving groups that were
investigated23 either gave no improvement or resulted in
significant amounts of the vinyl elimination byproduct 26.
Experimental Section
General. HPLC analysis was performed using a Waters
Sunfire C18 column (150 mm × 4.6 mm, 3.5 µm silica). Mobile
phase A: 0.3% H3PO4 in water. Mobile phase B: 0.3% H3PO4
in MeCN. Flow rate: 1 mL/min. Gradient: 0 min 5% B, 1 min
5% B, 10 min 100% B, 12 min 100% B, 17 min 5% B. UV
detection: 220 nm. Retention times (min): 1 (8.35), 2 (7.36), 9
(9.44), 11 (2.12), 12 (7.72), 17 (8.33), 24 (7.34), 25 (7.13).
(24) Yoshida, Y.; Sakakura, Y.; Aso, N.; Okada, S.; Tanabe, Y. Tetrahedron
1999, 55, 2183.
(25) Only 20% conversion after 24 h was seen without trimethyl amine as
catalyst.
(23) Mesylate, brosylate, nosylate, and bromide were investigated.
(26) Freiberg, L. A. U.S. Patent 3,725,385, 1973; CAN: 77, 19965.
Vol. 14, No. 1, 2010 / Organic Process Research & Development
•
155

Scheme 8. Synthesis of tosylate 9
(1S,2R)-1-Fluoromethyl-2-(4-iodophenyl)-2-methoxy-eth-
ylamine 12. To a solution of amino alcohol 2 (2.35 kg, 8.0
mol) in THF (7.1 L) was added methyl iodide (1.5 kg, 10.9
mol) at -25 °C. KHMDS (20% in THF, 11.8 kg, 11.9 mol)
was added over 1 h 40 min between -25 and -20 °C. The
reaction was stirred for 1 h at -25 to -20 °C. HPLC indicated
98% conversion to 12. An additional amount of methyl iodide
(168 g, 1.2 mol) and KHMDS (20% in THF, 1.2 kg, 1.2 mol)
was added, and the mixture was stirred for 1 h between -25
to -20 °C. The reaction was quenched with 3 M HCl (10.0 L)
at below -10 °C. The reaction mixture was diluted with water
(3.9 L) and heptane (5.6 L) at 0 °C, and the phases were
separated. The organic solution was extracted with water (1.9
L), and the combined aqueous solutions were washed with
heptane (3.5 L). The resulting acidic aqueous solution was
adjusted to pH ) 11 with 50% aqueous NaOH (2 kg) at below
10 °C and extracted with EtOAc (11.8 L). The organic extract
was concentrated, and the residual oil was azeotropically dried
with toluene (4.7 L) to give 12 (2.8 kg containing 0.36 kg
toluene, 98%) as a light-yellow oil. An analytical sample was
obtained by drying to constant weight. [R]25D ) -58.9 (c 2.39,
pinacol ester (1.8 kg, 8.12 mol), 12 (2.82 kg containing 0.37
kg of toluene, 7.96 mol), toluene (10.3 L), ethanol (3.5 L), and
water (3.5 L) was stirred and degassed with nitrogen for 30
min before the addition of Pd(PPh3)4 (23 g, 20 mmol).
Degassing was continued for another 30 min, and the reaction
was refluxed for 3.5 h. The mixture was cooled to 35 °C, and
water (5.7 L) was added followed by ethyl acetate (9.6 L). The
lower aqueous layer was removed and extracted with ethyl
acetate (4.9 L). The organic layers were combined, and the
solvent was evaporated. The resulting oil was dissolved in ethyl
acetate (12 L) and 2-propanol (7 L) and held below 20 °C while
sparging with HCl gas (800 g, 27.9 mol). The resulting slurry
was stirred for 30 min at 15 °C and filtered. The filter cake
was washed with ethyl acetate (3 × 4.9 L) and dried at 40 °C/
vacuum, yielding 11 (2.78 kg, 100%) as light-yellow crystals.
Mp >235 °C (dec); [R]25D ) -58.2 (c 1.86, MeOH); 1H NMR
(d6-DMSO, 300 MHz): δ 3.16 (s, 3H), 3.70 (m, 1H), 4.12 (ddd,
J ) 3.1, 10.9 Hz, JF ) 46.5 Hz, 1H), 4.44 (d, J ) 9.7, 1H),
4.59 (ddd, J ) 1.5, 10.9 Hz, JF ) 49.0 Hz, 1H), 7.48 (d, J )
8.3 Hz, 2H), 7.80 (d, J ) 8.3 Hz, 2H), 8.2 (broad s, 3H), 8.62
(d, J ) 4.1 Hz, 3H), 8.93 (s, 2H); 13C NMR (d6-DMSO, 75
MHz): δ 54.8 (d, JF ) 18.4 Hz), 56.3, 79.6 (d, JF ) 4.4 Hz),
80.6 (d, JF ) 169.5 Hz), 121.4, 126.2, 128.5, 133.6, 136.0,
155.0, 157.4; IR of free base (thin film): 1473, 1635, 2985,
3186, 3317 cm-1; ES-HRMS m/z: [M+ + 1H] calcd for
C14H18FN4O 277.1459, found 277.1457.
4,4-Dichloro-3-oxobutyric Acid Ethyl Ester, 15. To a
suspension of potassium ethyl malonate (2.37 kg, 13.9 mol) in
MeCN (20.7 L) was added triethylamine (2.9 L, 20.8 mol) at
0-5 °C followed by four portions of MgCl2 (1.62 kg, 16.7
mmol) over 1 h at below 15 °C. The thick slurry was warmed
to 20-25 °C and held for 2 h before cooling again to 0-5 °C.
Dichloroacetyl chloride (1 kg, 10.4 mol) was added over 1 h at
below 15 °C. The reaction mixture was warmed to 20-25 °C
and held for 18 h. The resulting thick paste was thinned by the
addition of ethyl acetate (5.9 L), and the slurry was cooled to
0-5 °C. The mixture was acidified with 1 M HCl (13.6 L) at
0-15 °C, and the aqueous layer was removed. Note: CO2 will
be released during the acidification. The organic layer was
washed twice with 1 M HCl (7 L), and the solvent was
evaporated, yielding 15 (1.64 kg containing 0.55 kg of ethyl
acetate, 81%) as an oil. An analytical sample was obtained by
drying to constant weight. 1H NMR (CDCl3, 300 MHz): Keto
isomer δ 1.27 (t, J ) 7.0 Hz, 3H), 3.83 (s, 2H), 4.21 (q, J )
7.0 Hz, 2H), 6.05 (s, 1H). Enol isomer δ 1.29 (t, J ) 7.1 Hz,
3H), 4.24 (q, J ) 7.1 Hz, 2H), 5.48 (s, 1H), 6.01 (s, 1H), 12.1
(s, 1H); 13C NMR (CDCl3, 75 MHz): Keto isomer δ 13.95,
42.4, 61.9, 69.7, 165.9, 189.2. Enol isomer δ 14.05, 61.2, 66.8,
90.6, 168.5, 171.9; IR (thin film): 1733, 1753, 2941, 2986 cm-1;
ES-HRMS m/z: (M+ + 1Na) calcd for C6H8Cl2O3 220.9743,
found 220.9739.
1
MeOH); H NMR (CDCl3, 300 MHz): δ 1.63 (s, 2H), 3.02
(dddd, J ) 4.8, 4.8, 6.5 Hz, JF ) 21.2 Hz, 1H), 3.25 (s, 3H),
4.10 (d, J ) 6.2 Hz, 1H), 4.16 (ddd, J ) 5.1, 9.2 Hz, JF ) 47.1
Hz, 1H), 4.34 (ddd, J ) 4.5, 9.1 Hz, JF ) 47.2, 1H), 7.07 (d,
J ) 8.3 Hz, 2H), 7.72 (d, J ) 8.3 Hz, 2H); 13C NMR (CDCl3,
75 MHz): δ 57.3 (d, JF ) 18.8 Hz), 57.4, 83.7 (d, JF ) 5.5
Hz), 84.7 (d, JF ) 168.9 Hz), 93.8, 129.3, 137.9, 138.9; IR
(thin film) 1588, 2934, 3317, 3386 cm-1; ES-HRMS m/z: [M+
- OMe + 1H] calcd for C9H11FIN 277.9837, found 277.9835.
5-[4-((1R,2S)-(2-Amino-3-fluoro-1-methoxypropyl))phe-
nyl]pyrimidin-2-ylamine Dihydrochloride 11. A mixture of
K3PO4 (5.18 kg, 24.4 mol), 2-aminopyrimidine-5-boronic acid
Scheme 9. N-Desmethylation of clarithromycin 27
Scheme 10. Alkylation of macrolide 8
156
•
Vol. 14, No. 1, 2010 / Organic Process Research & Development

4,4-Dichloro-3-(4-methyl-phenyl-sulfonyl-hydrazono)bu-
tyric Acid Ethyl Ester, 10. To a solution of 15 (3.89 kg, 19.5
mol) in MeCN (11 L) was added tosyl hydrazide (3.93 kg, 21.1
mol) at 15-25 °C in five portions over 1 h. The thickening
suspension was stirred at 20-25 °C for 18 h. Afterwards, the
precipitate was removed by vacuum filtration, and the filter cake
was rinsed with MeCN (2.2 L). The filtrate was concentrated
by removing MeCN (6 L) on a rotary evaporator. The
concentrated hydrazone solution was held at -15 °C for 48 h
and filtered, yielding 10 as damp, beige crystals (4.44 kg, 62%).
An analytical sample was obtained by drying to constant weight.
Mp 99-101 °C; 1H NMR (CDCl3, 300 MHz): δ 1.26 (t, J )
7.2 Hz, 3H), 2.44 (s, 3H), 3.62 (s, 2H), 4.19 (q, J ) 7.1 Hz,
2H), 6.19 (s, 1H), 7.33 (d, J ) 8.1 Hz, 2H), 7.83 (d, J ) 8.4
Hz, 2H), 9.64 (broad s, 1H); 13C NMR (CDCl3, 75 MHz): δ
14.1, 21.8, 32.6, 63.0, 72.0, 128.2, 129.9, 135.3, 144.7, 146.7,
169.5; IR (thin film): 1716, 1739, 2982, 3231 cm-1; Anal. Calcd
for C13H16N2O4SCl2: C, 42.72; H, 4.28; N, 7.87. Found: C,
42.52; H, 4.39; N, 7.63.
(1-{(1S,2R)-2-[4-(2-Aminopyrimidin-5-yl)-phenyl]-1-fluo-
romethyl-2-methoxy-ethyl}-1H-[1,2,3]triazol-4-yl)-acetic Acid
Ethyl Ester, 17. To a suspension of 11 (3.2 kg, 9.16 mol) in
ethanol (11.2 L) was added N,N-diisopropylethylamine (9.9 L,
56.8 mol) within 30 min at 0-15 °C. A solution of 10 (4.42
kg, 12.1 mol) in MeCN (4.8 kg) was added over 1.5 h at below
25 °C. The reaction was held at 20-25 °C for 1 h and then
warmed to 35 °C and held for 8 h. The reaction mixture was
split into three parts, and each was diluted with ethyl acetate
(9.5 L). Each portion was cooled to 0-5 °C and extracted with
1 M HCl (3 × 7.8 L). The combined aqueous extracts were
basified to pH 11 with 50% aqueous NaOH solution (2 kg) at
0-5 °C. The cold aqueous solution was extracted with ethyl
acetate (2 × 9.5 L), and the combined organic extracts were
evaporated. The oil from all three workup portions were
combined, taken up in ethanol (16 L), and concentrated on a
rotary evaporator, yielding 17 (5.88 kg containing 2.84 kg
ethanol, 80%) as an oil. An analytical sample was obtained by
crystallizing the dry residue from ethyl acetate. Mp 94-99 °C;
[R]25D -151.1 (c 1.57, MeOH); 1H NMR (CDCl3, 300 MHz):
δ 1.30 (t, J ) 6.9 Hz, 3H), 3.22 (s, 3H), 3.87 (s, 2H), 4.21 (q,
J ) 7.2 Hz, 2H), 4.65 (ddd, J ) 5.7, 9.9 Hz, JF ) 46.6 Hz,
1H), 4.71 (ddd, J ) 4.4, 9.8 Hz, JF ) 46.3 Hz, 1H), 4.78 (d, J
) 6.9 Hz, 1H), 5.00 (dddd, J ) 4.4, 5.6, 6.9, JF ) 17.0 Hz,
1H), 7.32 (d, J ) 8.2 Hz, 2H), 7.49 (d, J ) 8.3 Hz, 2H), 7.85
(s, 1H), 8.53 (s, 2H); 13C NMR (d6-DMSO, 75 MHz): δ 14.0,
31.3, 56.5, 60.4, 64.9 (d, JF ) 18.3 Hz), 80.4 (d, JF ) 7.0 Hz),
82.4 (d, JF ) 171.2), 121.5, 123.6, 125.3, 127.8, 2 × 135.5,
139.7, 155.9, 163.0, 170.0; IR (thin film): 1635, 1734, 3191,
3323 cm-1; ES-HRMS m/z: (M+ + 1H) calcd for C20H24FN6O3
415.1888, found 415.1886.
solution (160 mL) was added, and the mixture was stirred for
30 min followed by addition of 1 M citric acid (50 mL) and
CH2Cl2 (150 mL). The layers were separated, and the aqueous
phase was extracted with CH2Cl2 (3 × 150 mL). The combined
organic extracts were concentrated, and the residue was dis-
solved in ethyl acetate (40 mL) and stirred for 3 h. Heptane
(10 mL) was slowly added to the suspension and stirred for 30
min. The suspension was filtered and dried at 60 °C/vacuum,
yielding 24 (10.98 g, 76%) as a tan solid. Mp 135-137 °C;
1
[R]25 ) -140.8 (c 1.28, MeOH); H NMR (d6-DMSO, 300
D
MHz): δ 2.79 (t, J ) 6.9 Hz, 2H), 3.06 (s, 3H), 3.64 (dt, J )
5.3, 6.9 Hz, 2H), 4.59 (ddd, J ) 3.5, 10.1 Hz, JF ) 45.8 Hz,
1H), 4.68 (ddd, J ) 7.2, 10.1 Hz, JF ) 47.4 Hz, 1H), 4.73 (t,
J ) 5.3 Hz, 1H), 4.84 (d, J ) 7.1 Hz, 1H), 5.17 (dddd, J )
3.6, 7.0, 7.3 Hz, JF ) 19.6 Hz, 1H), 6.81 (s, 2H), 7.31 (d, J )
8.3 Hz, 2H), 7.63 (d, J ) 8.2 Hz, 2H), 7.97 (s, 1H), 8.60 (s,
2H); 13C NMR (d6-DMSO, 75 MHz): δ 29.2, 56.5, 60.4, 64.7
(d, JF ) 18.3 Hz), 80.5 (d, JF ) 6.9 Hz), 82.5 (d, JF ) 171.5
Hz), 121.5, 122.3, 125.3, 127.8, 135.4, 135.5, 144.2, 155.9,
162.9; IR (thin film): 2938, 3199, 3330 cm-1; ES-HRMS m/z:
[M+ + 1H] calcd for C18H22FN6O2 373.1783; found 373.1783.
N-Desmethylclarithromycin 8. To a mixture of clarithro-
mycin 27 (2.2 kg, 2.94 mol), water (3.08 L), NaOAc (3.2 kg,
39.5 mol), and methanol (30.8 L) was added iodine (792 g,
3.12 mol) at 50 °C. This was followed by heating the mixture
to gentle reflux. During the course of the reaction volatiles were
slowly distilled off, and a 50% NaOH solution (∼100 mL) was
added every 15 min to adjust the pH to 8. Additional charges
of iodine (106 g, 0.42 mol) were added after 1 and 1.5 h. Upon
completion of the reaction, the reaction was cooled to room
temperature, and the pH of the mixture was adjusted to pH 10
by addition of 50% aqueous NaOH solution (200 mL). Ethyl
acetate (8.25 L) was added to the reaction mixture and the
solvent was stripped to a volume of ∼12 L with an additional
volume of ethyl acetate (8.25 L) being added during the solvent
strip. To the biphasic mixture was added ethyl acetate (17.6 L)
and 5% aqueous Na2SO3 solution (8.4 L). The mixture was
heated to 32 °C until all solids dissolved, and the lower aqueous
layer was removed. The organic layer was washed with water
(3 × 1 L) and concentrated to ∼7 L. The precipitate was filtered,
washed with ethyl acetate (1 L), and dried on the filter. Note:
Compound 8 forms a stable 1:1 solvate with ethyl acetate. The
solid was then suspended in MeCN (8.8 L), and the residual
ethyl acetate was removed by azeotropic distillation (∼4.5 L
solvent removed). The suspension was cooled to 0-5 °C, stirred
for 1 h, and filtered. The cake was washed with MeCN (200
mL) and dried at 50 °C/vacuum, yielding 8 (1.34 kg, 62%) as
white crystals. Mp 231-233 °C; [R]25 ) -69.1 (c 0.73,
D
MeOH); 1H NMR (CDCl3, 300 MHz): δ 0.84 (t, J ) 7.4 Hz,
3H), 1.05 (d, J ) 7.5 Hz, 3H), 1.12 (d, J ) 6.4 Hz, 3H), 1.12
(s, 3H), 1.14 (d, J ) 6.4 Hz, 3H), 1.21 (d, J ) 7.5 Hz, 3H),
1.22 (d, J ) 7.0 Hz, 3H), 1.26 (s, 3H), 1.30 (d, J ) 6.2 Hz,
3H), 1.42 (s, 3H), 1.48 (dq, J ) 11.2, 7.2 Hz, 1H), 1.58 (dd, J
) 15.1, 4.9 Hz, 1H), 1.62-2.0 (m, 7H), 2.30 (d, J ) 9.6 Hz,
1H), 2.35 (d, J ) 15.4 Hz, 1H), 2.41 (s, 3H), 2.47 (ddd, J )
11.5, 4.3, 2.5 Hz, 1H), 2.60 (m, 1H), 2.67 (broad s, 1H), 2.87
(m, 1H), 3.00 (m, 2H), 3.04 (s, 3H), 3.13 (dd, J ) 9.6, 7.5 Hz,
1H), 3.22 (broad s, 1H), 3.32 (s, 3H), 3.53 (ddq, J ) 10.6, 4.3,
2-(1-{(1S,2R)-2-[4-(2-Aminopyrimidin-5-yl)-phenyl]-1-
fluoromethyl-2-methoxy-ethyl}-1H-[1,2,3]triazol-4-yl)etha-
nol 24. To a solution of 17 (16.0 g, 38.6 mmol) in methanol
(160 mL) was added MgCl2 (4.78 g, 50.2 mmol) at 20 °C. After
stirring for 30 min, granular NaBH4 (2.33 g, 61.8 mmol) was
added in three portions over 90 min (1.0 g, 26.4 mmol) at below
5 °C. The mixture was stirred further for 30 min and quenched
by adding acetone (14.2 mL, 193 mmol). A 15% aqueous NaCl
Vol. 14, No. 1, 2010 / Organic Process Research & Development
•
157

1.7 Hz, 1H), 3.67 (d, J ) 7.0, 1H), 3.75 (d, J ) 4.5 Hz, 1H),
3.77 (d, J ) 4.3 Hz, 1H), 3.99 (s, 1H), 4.00 (dq, J ) 6.0, 3.2
Hz, 1H), 4.42 (d, J ) 7.5 Hz, 1H), 4.92 (d, J ) 4.5 Hz, 1H),
5.06 (dd, J ) 11.1, 2.1 Hz, 1H); 13C NMR (CDCl3, 75 MHz):
δ 9.7, 10.8, 12.5, 16.1, 16.2, 18.2, 18.9, 19.9, 21.2, 21.4, 21.6,
33.4, 35.1, 37.5, 37.6, 39.2, 39.5, 45.2, 45.3, 49.6, 50.8, 60.5,
65.9, 68.6, 69.3, 72.9, 74.3, 75.2, 76.9, 77.4, 78.1, 78.5, 81.3,
96.3, 102.5, 175.8, 221.0; IR (thin film): 1684, 1723, 2977,
2500-3400, 3379, 3524, 3566 cm-1; Anal. Calcd for
C37H67NO13: C, 60.55; H, 9.20; N, 1.91. Found: C, 60.60; H,
9.32; N, 1.75.
mol) in MeCN (12.1 L) was stirred at 70 °C for 56 h and then
cooled to 50 °C, and water (1.2 L) was added. The mixture
was concentrated to ∼3 L, and the resulting slurry was
partitioned between ethyl acetate (12.1 L) and water (6 L). The
organic layer was concentrated to yield an oil (4.22 kg), which
was passed through a plug of silica gel (17 kg, 70-230 mesh),
eluting with 2 wt % NH3 in methanol/ethyl acetate (1:9). The
initial 162 L were combined and stripped. MeCN (16 L) was
added to the residual oil and heated to reflux. The clear solution
was diluted with MeCN (6 L) and seeded at 50 °C. The mixture
was slowly cooled to ambient temperature and stirred for 4 h.
The suspension was filtered, washed with MeCN (2 × 2 L),
and dried at 55 °C/vacuum, yielding 1 (1.54 kg, 61%) as beige
Toluene-4-sulfonic Acid 2-(1-{(1S,2R)-2-[4-(2-Aminopy-
rimidin-5-yl)-phenyl]-1-fluoromethyl-2-methoxy-ethyl}-1H-
[1,2,3]triazol-4-yl)ethyl Ester 9. To a mixture of 24 (855 g,
2.3 mol) and Me3NHCl (10.9 g, 0.11 mol) in CH2Cl2 (17.1 L)
was added triethylamine (800 mL, 5.74 mol). Tosyl chloride
(657 g, 3.45 mol) was added over 1 h at 0-5 °C, and the
mixture was stirred for 6 h at 0-5 °C. Excess tosyl chloride
was quenched with N,N-dimethylaminopropylamine (128 g,
1.38 mol) at 0-5 °C. The mixture was stirred for 30 min, and
then 1 M citric acid solution (5.7 L) was added. The layers
were separated, and the organic layer was washed with 1 M
citric acid solution (2 × 5.7 L) diluted with a 26% aqueous
NaCl solution (2 L), 7% aqueous NaHCO3 (5.7 L) diluted with
a 26% aqueous NaCl solution (2 L), and 26% aqueous NaCl
solution (5.7 L). The solvent was evaporated, and the crude oil
was solvent chased with MeCN (4.3 L) to ca. half volume to
yield 9 as a solution in MeCN (2.33 kg containing 1.21 kg of
9, 100%). An analytical sample was obtained by evaporating a
sample to constant weight. 1H NMR (CDCl3, 300 MHz): δ 2.39
(s, 3H), 3.02 (t, J ) 6.3 Hz, 2H), 3.07 (s, 3H), 4.26 (t, J ) 6.3
Hz, 2H), 4.59 (ddd, J ) 3.5, 10.1 Hz, JF ) 46.0 Hz, 1H), 4.74
(ddd, J ) 7.2, 10.2 Hz, JF ) 47.7 Hz, 1H), 4.84 (d, J ) 6.8
Hz, 1H), 5.22 (dddd, J ) 3.5, 6.9, 7.0 Hz, JF ) 19.0 Hz, 1H),
6.84 (s, 2H), 7.29 (d, J ) 8.3 Hz, 2H), 7.44 (d, J ) 8.1 Hz,
2H), 7.61 (d, J ) 8.23 Hz, 2H), 7.74 (d, J ) 8.2 Hz, 2H), 8.02
(s, 1H), 8.59 (s, 2H); 13C NMR (CDCl3, 75 MHz): δ 21.7, 26.0,
57.4, 65.8 (d, JF ) 19.7 Hz), 69.1, 81.1 (d, JF ) 5.1 Hz), 81.8
(d, JF ) 174.9 Hz), 122.6, 123.9, 126.5, 127.8, 127.9, 130.0,
132.8, 135.7, 136.1, 142.7, 145.0, 156.5, 162.6; IR (thin film):
2958, 3188, 3320, 3485 cm-1; ES-HRMS m/z: [M+ + 1H] calcd
for C25H28FN6O4S 527.1871; found 527.1871.
crystals. Mp 231-233 °C; [R]25 ) -96.7 (c 0.56, MeOH);
D
1H NMR (CDCl3, 300 MHz): δ 0.84 (t, J ) 7.4 Hz, 3H), 1.08
(d, J ) 7.5 Hz, 3H), 1.10-1.15 (m, 9H), 1.19 (d, J ) 6.6 Hz,
3H), 1.21 (d, J ) 5.9 Hz, 3H), 1.25 (s, 3H), 1.30 (d, J ) 6.2
Hz, 3H), 1.40 (s, 3H), 1.45 (dq, J ) 11.2, 7.1 Hz, 1H), 1.58
(dd, J ) 15.2, 5.0 Hz, 1H), 1.62-1.75 (m, 2H), 1.78-1.96
(m, 3H), 2.32 (s, 3H), 2.32-2.40 (m, 2H), 2.55 (m, 2H), 2.71
(m, 1H), 2.80-3.08 (m, 7H), 3.04 (s, 3H), 3.21 (s, 3H), 3.21
(m, 2H), 3.32 (s, 3H), 3.45 (s, 1H), 3.48 (m, 1H), 3.66 (d, J )
7.4 Hz, 1H), 3.75 (d, J ) 6.7 Hz, 1H), 3.77 (d, J ) 3.9 Hz,
1H), 4.00 (s, 1H), 4.01 (m, 1H), 4.44 (d, J ) 7.2 Hz, 1H), 4.57
(m, 1H), 4.73 (m, 1H), 4.77 (d, J ) 7.4 Hz, 1H), 4.92 (d, J )
4.2 Hz, 1H), 4.97 (m, 1H), 5.05 (dd, J ) 10.9, 2.0 Hz, 1H),
5.23 (s, 2H), 7.33 (d, J ) 8.2 Hz, 2H), 7.49 (d, J ) 8.2 Hz,
2H), 7.56 (s, 1H), 8.54 (s, 2H); 13C NMR (CDCl3, 75 MHz): δ
9.2, 10.8, 12.5, 16.15, 16.2, 18.2, 18.9, 20.0, 21.2, 21.6, 21.7,
25.3, 30.1, 35.1, 36.8, 37.4, 39.4, 39.5, 45.3, 45.4, 49.7, 50.8,
53.4, 57.6, 65.7, 65.8 (d, JF ) 13.4 Hz), 65.9, 68.9, 69.3, 71.0,
72.9, 74.5, 78.2, 78.6, 78.7, 81.0, 81.3 (d, JF ) 5.1 Hz), 81.9
(d, JF ) 174.8 Hz), 96.4, 103.0, 121.8, 124.3, 126.7, 128.1,
136.1, 136.3, 145.8, 156.6, 162.6, 176.1, 221.0; IR (thin film):
1728, 2972, 3207, 3348 cm-1; ES-HRMS m/z: [M+ + 1H] calcd
for C55H86FN7O14 1088.6290; found 1088.6277.
Acknowledgment
We thank Ricerca Bioscience, LLC (Concord, OH) for
demonstrating this chemistry in their scale-up facility and the
FT-ICR Mass Spectrometry Resource at the W.M. Keck
Foundation Biotechnology Resource Laboratory (New Haven,
CT) for obtaining the high-resolution mass spectra.
5-(4-{2-[4-(2-Amino-ethyl)-[1,2,3]triazol-1-yl]-3-fluoro-1-
methoxypropyl}phenyl)pyrimidin-2-yl-N-desmethylclarithro-
mycin, 1. A suspension of 9 (2.33 kg, 52 wt % in MeCN, 2.3
mol), powdered K2CO3 (475 g, 3.44 mol), and 8 (1.69 kg, 2.3
Received for review September 24, 2009.
OP900252A
158
•
Vol. 14, No. 1, 2010 / Organic Process Research & Development