Introduction

Noodles are an important part of the diet in many countries of eastern and south-eastern Asia, accounting for approximately one third of Australia’s bread wheat exports and about 40% of wheat consumed in Asian countries (Hou and Kruk, 1998). Asian noodles made from wheat may be divided into two general classes based on the ingredients used: white salted noodles (WSN, made from flour, sodium chloride and water) and yellow alkaline noodles (YAN, made from flour, alkaline salts such as sodium and potassium carbonate and water). Asian customers prefer bright yellow alkaline noodles that retain a stable colour for 24-48 hours after preparation and perceive red or dull grey colours as undesirable. A preliminary study by Mares et al. (1997) indicated that the colour of alkaline noodles was due partly to xanthophylls and partly to compounds that could be extracted with aqueous solvents. These latter compounds, in contrast to the xanthophylls, were colourless at neutral or acid pH but turned yellow at higher pH.

Wang (2001) developed an efficient method for the quantification of important pigments in yellow alkaline noodles, using hydroxylamine extraction of wheat flour. Alkaline noodles prepared from flour that had been extracted with hydroxylamine solution, failed to turn yellow and had reflectance curves that were similar to control noodles made from flour and 2% NaCl solution. While flavone-C-glycosides constituted an important group of pigments (Asenstorfer et al., 2005), another pigment was identified as a possible contributor to the colour of YAN. This pigment was relatively unstable, colourless at low pH (λ_max 340 nm) and yellow at high pH (λ_max 400 nm). Electrospray mass spectrum yielded a major ion m/z 184.0 and a possible parent ion m/z 367.2 (MH⁺) (Figure 1). A neutral mass of 183 suggested the presence of nitrogen.

Figure 1. Mass spectrum of the unidentified pigment

2,6-Dimethoxy-p-quinone (1), a possible contributor to YAN colour, can be isolated from wheat germ fermented with baker's yeast (Vuataz, 1950; Cosgrove et al., 1952) This compound can form two oxime isomers, 2,6-dimethoxy-1,4-benzoquinone-4-oxime (2) and 3,5-dimethoxy-1,4-benzoquinone-4-oxime (3) with mass 183. These isomers were synthesised and their properties compared with the natural product

Methods

2,6-Dimethoxy-1,4-benzoquinone-4-oxime (2,6-dimethoxy-4-oximino-2,5-cyclohexadienone-1) was synthesised from 2,6-dimethoxy-1,4-benzoquinone and hydroxylamine according to Bolker and Kung (1969). 3,5-Dimethoxy-1,4-benzoquinone-4-oxime was prepared by nitrosodemethylation of 1,3,5-trimethoxybenzene according to Shpinel et al. (1991). These compounds were compared with the natural product by HPLC, absorbance and mass spectrometry.

Approximately 10 mg of 2,6-dimethoxy-1,4-hydroxybenzene and 20 mg hydroxylamine hydrochloride were added to 2 mL of water and adjusted to pH 7. This was allowed to react for 16 hours at 45oC. The solution was extracted with dichloromethane. KCl was added and the dichloromethane was carefully evaporated under reduced pressure. This product was compared with the natural product by HPLC, absorbance and mass spectrometry and with the 2,6-dimethoxy-1,4-benzoquinone oxime products by NMR and infrared spectroscopy.

Results

Natural Product

HPLC retention time = 21.4 min. ESI-MS, m/z: 367.2 (MH⁺). Absorbances (aq.): λ_max 340 nm at pH 2.5 and λ_max 400 nm at pH 10.

2,6-Dimethoxy-1,4-benzoquinone-4-oxime

HPLC retention time = 19.1 min. ESI-MS, m/z: 184.0 (MH⁺). Absorbance (aq.): λ_max 300 nm (ε = 16 000), 395 nm (ε =1 200) at pH 2.5 and λ_max 353 nm (ε = 25 300) at pH 10. ¹H NMR (DMSO-d6): d 3.71 (s, 6H, OCH3), 5.61 (s, 2H); ¹³C NMR (DMSO-d6): d 56.1 (OCH3), 102.4 broad (C3 or C5), 137.7 (C4), 185.6 (C1). Quaternary carbons (C2 and C6) were not observed. IR (cm^-1): 3064 (broad), 2750s (broad), 1626vs (C=O), 1571vs (C=N), 1452s (C-O stretch), 1246s (C-O stretch), 1228s, 1122vs (aryl C-N stretch), 1051vs (N-OH), 702s.

3,5-Dimethoxy-1,4-benzoquinone-4-oxime

HPLC retention time = 25.4 min. ESI-MS, m/z: 184.0 (MH⁺). Absorbance (aq.): λ_max 298 nm (ε = 18 000), 395 nm (ε = 1 400) at pH 2.5 and λ_max 353 nm (ε = 29 500) at pH 10. ¹H NMR (DMSO-d6): d 3.72 (s, 3H, OCH3), 3.74 (s, 3H, OCH3), 5.60 (s, 1H), 5.63 (s, 1H); ¹³C NMR (DMSO-d6): d 56.1 (OCH3), 56.2 (OCH3), 102.1 (C2 or C6), 103.3 (C2 or C6), 137.5 (C4), 157.6 (C3 or C5), 161.1 (C3 or C5), 185.6 (C1). IR (cm^-1): 2800s(broad), 1630vs (C=O), 1575vs (C=N), 1454vs (C-O stretch), 1409s, 1249vs (C-O stretch), 1125vs (aryl C-N stretch), 1052vs (N-OH), 704s.

Figure 2. Comparison of the spectrum of 2,6-dimethoxy-1,4-benzoquinone-4-oxime with the unidentified pigment.

Figure 3. The m/z 184 MS/MS spectra of; (A) 2,6-dimethoxy-1,4-benzoquinone-4-oxime and (B), the unidentified pigment.

Reaction of 2,6-dimethoxy-1,4-hydroquinone with hydroxylamine (4,4'-dihydroxy-3,3',5,5'-tetramethoxy-azodioxybenzene (3))

HPLC retention time = 21.4 min. ESI-MS, m/z: 367.2 (MH⁺). Absorbance (aq.): λ_max 340 nm (ε = 16,000) pH 2.5 and λ_max 400 nm (ε = 17 000) at pH 10. ¹H NMR (DMSO-d6): d 3.711 (s, 6H, OCH3), 3.739 (s, 6H, OCH3), 6.510 (d, 2 H, J=2.17 Hz), 6.803 (d, 2 H, J=2.00 Hz); ¹³C NMR (DMSO-d6): d 55.5 (OCH3), 55.6 (OCH3), 94.5 (C2, C2’), 108.6 (C6, C6’), 148.0 (C1, C1’), 151.9 (C3, C3’), 153.6 (C5, C5’), 175.0 (C4, C4’). IR (cm-1): 3268vs (broad, OH), 1644vs (C=O), 1581vs (C=N), 1455m (C-O stretch), 1316vs (N⁺-O^-), 1249vs (C-O stretch), 1219s, 1113vs (aryl C-N stretch), 952s (N⁺-O^-).

Figure 4. Comparison of the spectra of the hyroxylamine/2,6-dimethoxy-p-hydroquinone reaction product and the unidentified pigment

Figure 5. The m/z 184 MS/MS spectrum of the hyroxylamine/2,6-dimethoxy-p-hydroquinone reaction product

Discussion

The absorbance and mass spectral data of the two oximes were similar, only the retention times were different. However, discrepancies in the spectra (Figure 2), HPLC retention times and mass spectra fragmentation patterns (Figure 3) clearly indicate that the oximes and natural product were different. Fortuitously, a compound with similar absorbances (Figure 4), HPLC retention times and mass spectra (Figure 5) as the unknown pigment was detected amongst the reaction products of hydroxylamine and 2,6-dimethoxy-p-hydroquinone (1,4-dihydroxy-2,6-dimethoxybenzene). This compound was subsequently purified and identified using infrared, ¹³C and ¹H NMR spectroscopy as 4,4'-dihydroxy-3,3',5,5'-tetramethoxy-azodioxybenzene, a dimer of 2,6-dimethoxy-1,4-benzoquinone-4-oxime (2).

It is important to note that during the extraction process, hydroxylamine reacts with the hydroquinone rather than the quinone. Moreover while 2,6-dimethoxyquinone was isolated from fermented wheat germ (Vuataz, 1950; Cosgrove et al., 1952) and it has been suggested that the precursor is a hydroquinone glycoside (Bouvier and Horváth, 1987), the precursor has, however, never been identified. 2,6-Dimethoxy-p-hydroquinone, itself, is short lived in the noodle (t_1/2=15 hrs) and is probably oxidised to the quinone. Differences in the extinction coefficients of the 2,6-dimethoxy-p-quinone and 4,4'-dihydroxy-3,3',5,5'-tetramethoxy-azodioxybenzene (>40 fold) suggest the quinone is only a minor contributor to the yellow colour of YAN. 2,6-Dimethoxy-p-hydroquinone and its quinone may have a role in the browning of alkaline noodles. This is currently under investigation.

References

Asenstorfer, R.E., Wang and Y. Mares, D.J. (2005) J. Cereal Sci. Submitted

Bolker, H.I. and Kung, F.L. (1969) Can. J. Chem. 47: 2109-2115.

Bouvier, E. and Horváth, C. (1987) Acta Biochim. Biophys. Hung. 22(2-3): 215-228.

Cosgrove, D.J., Daniels, D.G.H., Whitehead, J.K., Goulden, J.D.S. (1952) J. Chem. Soc. 4821.

Hou, G. and Kruk, M. (1998) American Institute of Baking Technical Bulletin 20: 1-10.

Mares, D.J., Wang, Y. and Cassidy, C.A. (1997) Cereals 97. Proceedings of the 47th Cereal Chemistry Conference, RACI, Melbourne, Australia. 114-117.

Shpinel, Y.I., Klimova, I.V. and Belyev, E.Y. (1991) Zh. Obshch. Khim. 27: 1493-1497.

Vuataz, L. (1950) Helv. Chim. Acta 3: 433-43.

Wang, Y. (2001) Characterization of grain constituents responsible for the yellow colour of Asian alkaline noodles. University of Sydney. PhD thesis.