1682
A.C. Wood et al. / Tetrahedron 71 (2015) 1679e1683
Firstly, the reaction between deprotected ribitylated diamine 3
100%]; no di-protected product was observed (C18H28N2O4: m/z
336).
and diethyl 2-bromo-3-oxogluarate 4 was found to be largely
influenced by the solvent mixture chosen for the reaction, with
poor miscibility between reactant solutions negatively affecting
product yields. Strict control in the formation of diethyl 2-bromo-3-
oxogluarate 4 was used to ensure the production of a single mono-
brominated product (with spectroscopic evidence suggesting sub-
stantial di-bromination of this compound when prepared accord-
ing to the original method1). While the use of a phase-transfer
catalyst (DMAP) was found to be ineffective, a more efficient sol-
vent mixture consisting of 1:1 dichloromethane/methanol was
found, improving the yield for this step of the reaction from 55% to
77%.
Secondly, efficient ring-closure of the bicyclic intermediate
compound 5 to furnish 1-deazariboflavin 2 was performed using
a solution of ammonia in methanol. The concentration of ammonia
played a key role, with low concentrations furnishing corre-
spondingly low yields of 1-deazariboflavin 2. Dissolution of am-
monia in methanol cooled to ꢁ78 ꢀC was found to produce a highly
concentrated solution with sufficient stability for later use at am-
bient temperature and which allowed yields of 92% to be obtained;
a substantial improvement over the 33% reported previously.1
4.3. 1-N-(Ribityl),2-N(Boc)-1,2-diamino-4,5-dimethylbenzene
10
tert-Butyl (2-amino-4,5-dimethylphenyl)carbamate 9 (3.30 g,
14.0 mmol), D-ribose (5.25 g, 35 mmol, 2.5 equiv) and sodium
cyanoborohydride (1.57 g, 25 mmol, 1.8 equiv) were powdered and
pre-dried under vacuum in flame-dried glassware, before dissolu-
tion in anhydrous MeOH (150 mL) under an Ar/N2 atmosphere. The
mixture was heated at reflux for 48 h before cooling to room
temperature, whereupon the solvent was removed by evaporation.
The brown residue was dissolved in 1 M HCl(aq) (50 mL) and swirled
for 30 s, before swift neutralisation using a saturated aqueous so-
lution of sodium hydrogen carbonate (75 mL). The resulting solu-
tion was extracted into ethyl acetate (5ꢂ75 mL), with the combined
organic layers washed using brine (2ꢂ100 mL) and dried over
Na2SO4. Solvent was removed under reduced pressure and the re-
sultant solid further dried under high vacuum for 8 h, to give the
ribitylated product 10 as an orange crystalline solid (4.89 g,
13.20 mmol, 94%), mp 110e112 ꢀC; 1H NMR (500 MHz, MeOD-d3)
d
¼6.90 (s, 1H), 6.64 (s, 1H), 3.98e3.95 (m, 1H), 3.81 (dd, 1H,
4. Experimental detail
JAM¼3.5 Hz, JAX¼11.0 Hz), 3.79e3.76 (m, 1H), 3.67 (q, 2H, J¼6.5 Hz),
3.46 (dd, 1H, JAM¼3.0 Hz, JAX¼13.0 Hz), 3.17 (dd, 1H, JAM¼3.1 Hz,
JAX¼8.0 Hz), 2.22 (s, 3H), 2.16 (s, 3H), 1.53 (s, 9H); 13C NMR
4.1. General experimental techniques
(100 MHz)
d
¼153.1, 132.9, 132.3, 131.8, 126.4, 122.0, 118.6, 74.4, 73.1,
All reagents were purchased from SigmaeAldrich or Fisher
Scientific, and were of 98% (or higher) purity. All solvents were
purchased from Fisher Scientific. Anhydrous solvents were ob-
tained from an MBraun MB SPS-800 solvent purification system,
or (in the case of methanol) distilled over CaH2 before being fur-
ther dried over molecular sieves. ‘High vacuum’ refers to ultra-low
vacuum pressure (<10ꢁ2 Torr), provided by an Oerlikon TRIVAC
mechanical oil-pump. Melting points were measured in a Gallen-
kamp Variable heater apparatus fitted with an internal tempera-
ture probe and confirmed using a mercury thermometer and are
uncorrected; literature values (where available) are given along-
side observed results. NMR spectroscopy was performed using
Bruker AVANCE DPX 400 MHz or 500 MHz (Ultrashielded) spec-
trometers (1H, 13C) or a Bruker AVANCE DPX 250 MHz spectrom-
eter (13C). All spectra were recorded in fully deuterated solvents at
20 ꢀC. Samples for mass spectrometry were analysed via the in-
ternal MS Service at Cardiff University, obtained using a Waters
LCT Premier XE TOF Spectrometer. Ionisation was performed by EI,
ESI or APCI, with detection in positive or negative mode as re-
quired for each product.
70.7, 63.4, 46.5, 28.2, 19.3, 18.9; m/z (EIþ) 394 [(MþHþþNaþ), 20%],
370 [(MþHþ), 100%]; (ESþ) 370 [(MþHþ), 100%].
4.4. 1-N-(Ribityl),2-diamino-4,5-dimethylbenzene 3
1-N-(Ribityl),2-N(Boc)-diamino-4,5-dimethylbenzene
(10)
(1.20 g, 3.24 mmol) was dissolved in 2 M aqueous HCl (25 mL) and
stirred at ambient temperature for 3 h. The solution was lyophilised
to furnish deprotected intermediate 3 as a green crystalline solid
(876 mg, 3.24 mmol, 100%), 1H NMR (400 MHz, D2O)
d
¼7.07 (s, 1H),
6.76 (s, 1H), 3.92 (m, 1H), 3.83 (dd, 1H, JAM¼3.5 Hz, JAX¼11.2 Hz),
3.66 (m, 1H), 3.62 (dd, 2H, JAM¼6.5 Hz, JAX¼12.8 Hz), 3.47 (dd, 1H,
JAM¼3.0 Hz, JAX¼12.8 Hz), 3.31 (m, 1H), 3.16 (dd, 1H, JAM¼3.1 Hz,
JAX¼8.0 Hz), 2.14 (s, 3H), 2.08 (s, 3H); 13C NMR (100 MHz, D2O)
d
¼131.1, 130.4, 125.8, 124.0, 115.8, 113.2, 73.8, 72.9, 72.0, 64.7, 47.2,
19.0, 18.5; m/z (EIþ) 294 [(MþHþþNaþ), 20%], 270 [Mþ, 100%].
4.5. 5,6-Dimethylbenzimidazole 11
1-(N-(Ribityl)),2-diamino-4,5-dimethylbenzene
3 (200 mg,
0.74 mmol) was dissolved in water (100 mL), and the resulting
solution stirred vigorously at ambient temperature. Compressed air
was continuously bubbled through the solution. After 96 h, the
solution was extracted with EtOAc (3ꢂ30 mL), and the aqueous
4.2. tert-Butyl (2-amino-4,5-dimethylphenyl)carbamate 9
4,5-Dimethylbenzene-1,2-diamine 8 (1.00 g, 7.30 mmol), di-tert-
butyl dicarbonate (1.60 g, 7.30 mmol, 1.0 equiv) and sodium hy-
drogen carbonate (0.62 g, 8.34 mmol, 1.15 equiv) were dissolved in
a mixture of dioxane (50 mL) and water (50 mL). The mixture was
stirred for 3 h at room temperature, with the original orange colour
of the solution changing to brick red. The mixture was diluted with
water (150 mL) and extracted into dichloromethane (3ꢂ50 mL); the
combined organic extracts were washed with saturated aqueous
sodium hydrogen carbonate (50 mL), brine (50 mL), and further
dried over MgSO4. The solvent was removed under reduced pres-
sure, and the resulting solid was dried under high vacuum for 4 h,
yielding carbamate 9 as a red-orange solid (1.59 g, 6.73 mmol, 92%),
fraction lyophilised to furnish starting material
3 (174 mg,
0.64 mmol, 87%). Combined organic extracts were washed with
saturated aqueous sodium hydrogen carbonate (50 mL) and brine
(2ꢂ50 mL), before drying over MgSO4. The solvent was evaporated
removed and the crude brown solid recrystallised from diethyl
ether to furnish 5,6-dimethylbenzimidazole 11 (9 mg, 0.06 mmol,
8%) as a pale yellow crystalline solid, mp 203e205 ꢀC [lit.17 mp
201e205 ꢀC; lit.18 mp 205e206 ꢀC]; 1H NMR (400 MHz, CDCl3)
d
¼8.00 (s, 1H), 7.44 (s, 2H), 2.38 (s, 6H); 13C NMR (100 MHz),
d
¼139.9, 136.3, 132.0, 115.6, 20.4; m/z (EIþ) 146 [Mþ, 100%].
mp 153e155 ꢀC [lit.16 mp 146 ꢀC]; 1H NMR (400 MHz, CDCl3)
(s,1H), 6.57 (s, 1H), 6.19 (br s,1H), 3.53 (br s, 2H), 2.15 (s, 3H), 2.14 (s,
3H), 1.50 (s, 9H); 13C NMR (62.5 MHz, CDCl3)
¼154.1, 137.4, 134.3,
132.3, 127.9, 122.3, 118.6, 80.3, 28.4, 19.3, 18.9; m/z (EIþ) 236 [Mþ,
d¼7.02
4.6. Diethyl 2-bromo-3-oxoglutarate 411
d
In a flame-dried three-neck-round bottom flask, diethyl 3-
oxoglutarate 12 (2.90 mL, 3.50 g, 20 mmol) was dissolved under