Abstract
Rice wine, in which γ-aminobutyric acid is present, is beneficial to human health and is one of the three most well-known fermented wines in the world, and is very popular in China. The rapid detection of γ-aminobutyric acid was studied in rice wine using near infrared spectroscopy with an optical fibre probe. Through the selection of detection conditions, including a waveband range of 12500–4000 cm−1, a scanning duration of 16 scans and a resolution of 8 cm−1, the near infrared spectrum of rice wine was acquired three times, for every wine sample, with an optical fibre probe. The resulting average value of the spectrum was obtained and the corresponding data were analysed via normalization. By adopting a multivariate calibration partial least squares method (PLS) and establishing a calibration model, the highest precision for γ-aminobutyric acid in rice wine was predicted when the factor coefficient was 17. The overall results demonstrating the content of γ-aminobutyric acid in rice wine was predicted to be between 157.6696–317.5813 mg/L, with a relative standard deviation of prediction between 0.01–5 %, as well as the fact that the single sample measuring time was less than 20 s, prove that near infrared spectroscopy is a rapid, accurate and effective method to adopt for detecting the content of γ-aminobutyric acid in rice wine.
Keywords: Near infrared spectroscopy, Optical fibre probe, γ-aminobutyric acid (GABA), Rice wine, Partial least squares
Introduction
Rice wine, in which γ-aminobutyric acid (GABA) is present, favourably influences human health and is one of the three most well-known fermented wines in the world, and very popular in China. GABA is catalysed from glutamic acid by glutamic acid decarboxylase and exists widely in nature; its effects include calming nerves, resisting anxiety and detoxification etc. (Feudis De 1983; Haugstad et al. 1997; Lieve et al. 2011; Sergio et al. 1995; Thomas et al. 1984). It’s an important inhibitory neurotransmitter (Niranjala et al. 1995) in the central nervous system of mammals, crustaceans and insects.
In recent years, it has been reported that an amino acid analyser, GC or LC-MS, column chromatography fluorescence assay and high performance liquid chromato- graphy (HPLC) have all been applied to the detection of γ-aminobutyric acid (Constantinos et al. 2004; Zhang and Alan 1997; Zhang and Sun 2005; Mao et al. 2011; Osborne et al. 1999; Silva et al. 2009; Ryutaro et al. 2012); however, the detection of GABA using the near infrared spectroscopy (NIR) method, employing chemometrics, has not previously been published.
The PLS method is widely used in many fields, such as chemistry, economics, medicine, psychology and pharmacy, to establish models. In particular, it has the advantage of setting a random variable if needed. For chemometrics, the PLS method has become a standard established tool of the multivariant model.
The near infrared spectrum includes valuable information such as bond strength, chemical composition, electronegativity and hydrogen bonds; for a solid sample, in addition to the above-mentioned parameters, it also includes information on scattering, diffuse-reflectance, polarization by reflection, sample particle and size etc. The routine methods of spectral analysis cannot be employed to extract the complicated and weak information from the near infrared spectrum; therefore, chemometrics has been introduced to near infrared spectrum analysis to effectively extract the information (Ryutaro et al. 2012).
In this paper, data from the near infrared spectrum of rice wine was analysed using a PLS method (Phil 1987) and quantified to give the γ-aminobutyric acid content in rice wine. It provides better results in terms of precision, stability and efficiency.
Materials and methods
Materials and apparatus
The test samples were classified into two groups, consisting of calibration and prediction. HPLC was employed to detect the content of GABA in rice wine for calibration; Table 1 shows the results of the HPLC determination.
Table 1.
The content of γ-aminobutyric acid in rice wine
| Sample No. | GABA (mg/L) | Sample No. | GABA (mg/L) |
|---|---|---|---|
| 1 | 273.3320 | 16 | 277.8917 |
| 2 | 186.7935 | 17 | 317.5885 |
| 3 | 246.0835 | 18 | 260.3254 |
| 4 | 185.2436 | 19 | 210.2488 |
| 5 | 181.9210 | 20 | 179.5326 |
| 6 | 168.7020 | 21 | 157.6710 |
| 7 | 175.5635 | 22 | 265.4957 |
| 8 | 190.9920 | 23 | 163.5925 |
| 9 | 203.3363 | 24 | 222.2330 |
| 10 | 293.9145 | 25 | 215.6950 |
| 11 | 181.0190 | 26 | 217.9965 |
| 12 | 199.5757 | 27 | 269.0320 |
| 13 | 243.3935 | 28 | 273.7770 |
| 14 | 266.1015 | 29 | 191.8780 |
| 15 | 225.9898 | 30 | 250.6909 |
Acquisition of the near infrared spectrum
The spectrum of rice wine was acquired using MPA near infrared spectroscopy with an optical fibre probe at 25 °C. With a detection waveband between 12500–4000 cm−1, a scanning duration of 16 scans (an automatic collection during each scan of 16 points) and an instrument resolution of 8 cm−1, the sample of rice wine was analysed using an optical fibre probe. Spectra were collected in triplicate for every sample and the average value of the spectra was obtained. Background scanning was applied to each sample to enable background correction; a sample spectrum is shown in Fig. 1.
Fig. 1.
Near infrared spectrum of rice wine
Data processing and analysis
Data processing and analysis were executed using the PLS software of the quantitative analysis MATLAB software. The software operation interface is shown in Fig. 2
Fig. 2.
The operational interface for MATLAB
Results and discussion
Experimental results
30 samples were used for calibration and experimental prediction. Firstly, the above acquired spectral data were treated using centralized normalization. Then, the data were quantitatively analysed using the PLS software of MATLAB. The optimal factor of 17 was determined through cross-validation. The samples were predicted from the calibration set; the prediction value and set is shown in Table 2.
Table 2.
The value of prediction and set Unit: mg/L
| PLS prediction | HPLC detector | PLS prediction | HPLC detector |
|---|---|---|---|
| 273.3223205 | 273.3320 | 226.0029244 | 225.9898 |
| 186.817801 | 186.7935 | 277.8939183 | 277.8917 |
| 246.073031 | 246.0835 | 317.581345 | 317.5885 |
| 185.2646826 | 185.2436 | 260.3029468 | 260.3254 |
| 175.2979692 | 181.9210 | 210.2577379 | 210.2488 |
| 175.2979692 | 168.7020 | 179.4969437 | 179.5326 |
| 183.2682281 | 175.5635 | 157.6696101 | 157.6710 |
| 183.2682281 | 190.9920 | 265.5123383 | 265.4957 |
| 203.3364245 | 203.3363 | 163.634374 | 163.5925 |
| 293.9188751 | 293.9145 | 222.2462783 | 222.2330 |
| 180.9945620 | 181.0190 | 215.6724597 | 215.6950 |
| 199.5881844 | 199.5757 | 218.0171332 | 217.9965 |
| 243.3960585 | 243.3935 | 268.9875852 | 269.0320 |
| 266.1000037 | 266.1015 | 273.8337104 | 273.7770 |
In order to detect GABA in rice wine, a multivariant calibration PLS was used to establish a precise calibration model. The predicted content of GABA in rice wine is between 157.6696–317.5813 mg/L with a relative standard deviation of 0.01–5 %, proving that the result is accurate.
The stability of near infrared spectrum detection
Three groups of rice wine samples were simultaneously tested six times using the above-established detection method. The results are shown in Table 3.
Table 3.
The stability experiment results of GABA detection in rice wine
| Sample | Parallel Sample 1 | Parallel Sample 2 | Parallel Sample 3 | Parallel Sample 4 | Parallel Sample 5 | Parallel Sample 6 | RSD |
|---|---|---|---|---|---|---|---|
| 1 | 306.2406 | 312.0598 | 308.1362 | 308.8122 | 310.2404 | 312.1105 | 0.7480 % |
| 2 | 175.2980 | 177.2660 | 175.2981 | 180.2981 | 168.7020 | 176.2991 | 2.1778 % |
| 3 | 243.3961 | 243.3935 | 243.4451 | 248.3163 | 246.5567 | 244.7845 | 0.8373 % |
| Average RSD | 1.2544 % | ||||||
The relative standard deviation of the three groups of simultaneously tested samples is 1.2544 %, which proves that the method is stable.
The testing results are satisfactory for the market sample; as seen in Table 4.
Table 4.
The testing results of the market sample
| Sample | PLS prediction | HPLC detector | RSD |
|---|---|---|---|
| 1 | 274.1229 | 265.4957 | 1.6247 % |
| 2 | 149.1743 | 163.5925 | 4.4067 % |
| 3 | 222.2330 | 208.3496 | 3.3317 % |
| 4 | 234.9821 | 215.6950 | 4.4709 % |
| 5 | 189.7552 | 191.8780 | 0.5532 % |
| Average RSD | 2.8774 % | ||
The rice wines on the market were tested using the near infrared method and after the results were verified by HPLC, the agreement was demonstrated. The relative standard deviation of the results is less than 5 % and the average relative deviation is 2.8774 %, thus, this NIR method is suitable for the detection of GABA content in rice wine.
Conclusion
An optical fibre near infrared spectroscopy probe is an easy and rapid method for detecting the γ-aminobutyric acid content in rice wine; in addition, it takes less than 20 s to detect per sample. It is more precise as the related standard deviation for sample prediction is less than 5 %. It is more stable as the average relative standard deviation for a single detection is less than 1.5 %. The samples on the market were tested and the average relative standard deviation is 2.8774 %. We used the multivariant calibration PLS calibration method which is an established mature and reliable popular method, and is easy to apply and use.
Acknowledgments
This work was supported by the General Administration of Quality Supervision Inspection and Quarantine of P.R. China (Grant No.2012.5) and Zhejiang Bureau of Quality and Technical Supervision (Grant No.20110236). We are delighted to acknowledge the assistance of Dr Liz and Mr Yuan Liu in revising the English, and the discussions with colleagues in our research group.
References
- Constantinos KZ, Theodoridis GA, Anastasios N, Voulgaropoulos J. On-line coupling of sequential injection with liquid chromatography for the automated derivatization and determination of γ-aminobutyric acid in human biological fluids. J Chromatogr B. 2004;808(2):169–175. doi: 10.1016/j.jchromb.2004.05.003. [DOI] [PubMed] [Google Scholar]
- Feudis De FV. γ-aminobutyric acid and cardiovascular function. Exp J. 1983;39(8):845–849. doi: 10.1007/BF01990401. [DOI] [PubMed] [Google Scholar]
- Haugstad TS, Karlsen HE, Krajtči P, Due-Tønnessen B, Larsen M, Sandberg C, Sand O, Brandtzaeg P, Langmoen IA. Efflux of γ-aminobutyric acid caused by changes in ion concentrations and cell swelling simulating the effect of cerebral ischaemia. Acta Neurochir. 1997;139(5):453–463. doi: 10.1007/BF01808883. [DOI] [PubMed] [Google Scholar]
- Lieve L, Kristof B, Jan AD. In situ production of γ-aminobutyric acid in breakfast cereals. Food Chem. 2011;129(2):395–401. doi: 10.1016/j.foodchem.2011.04.090. [DOI] [PubMed] [Google Scholar]
- Mao XY, Cao YG, Min DY, Guo F, Xie N, Chen TB, Shaw C, Cai JQ. RP-LC with fluorescence detection of amino acids in rat brain synaptosomes. Chromatographia. 2011;73:157–163. doi: 10.1007/s10337-010-1822-4. [DOI] [Google Scholar]
- Niranjala JK, Tillakaratne LMK, Gibson KM. Gamma-aminobutyric acid (GABA) metabolism in mammalian neural and nonneural tissues. Comp Biochem Physiol A Physiol. 1995;112(2):247–263. doi: 10.1016/0300-9629(95)00099-2. [DOI] [PubMed] [Google Scholar]
- Osborne PG, Hu Y, Covey DN, Barnes BN, Katz Z, Drew KL. Determination of striatal extracellular γ-aminobutyric acid in non-hibernating and hibernating Arctic ground squirrels using quantitative microdialysis. Brain Res. 1999;839(1):1–6. doi: 10.1016/S0006-8993(99)01627-3. [DOI] [PubMed] [Google Scholar]
- Phil WKN. Near-infrared technology in the agricultural and food industries [M], Minnesota. USA: St. Paul The American Association of Cereal Chemists Inc.; 1987. [Google Scholar]
- Ryutaro T, Hiroyuki A, Tadaaki S, Kiyomi I, Masaki A, Otsuka M. Nondestructive prediction of oren extract powder, a herbalmedicine, in suppositories by chemometric near-infrared spectroscopy. J Nat Med. 2012;66(2):329–332. doi: 10.1007/s11418-011-0596-9. [DOI] [PubMed] [Google Scholar]
- Sergio T, Miguel PM, Luca F, Jesus MF, Beani L, Rambert FA, Fuxe K. Modafinil and cortical γ-aminobutyric acid outflow. Modulation by 5-hydroxytryptamine neurotoxins. Eur J Pharmacol. 1995;273(1–2):63–71. doi: 10.1016/0014-2999(94)00675-w. [DOI] [PubMed] [Google Scholar]
- Silva DMF, Ferraz VP, Ribeiro AM. Improved high performance liquid chromatographic method for GABA and glutamate determination in regions of the rodent brain. J Neurosci Methods. 2009;177:289–293. doi: 10.1016/j.jneumeth.2008.10.011. [DOI] [PubMed] [Google Scholar]
- Thomas LP, Masatoshi I, Karen J, Shirley H. Elevation of γ-aminobutyric acid in human brain may increase dopaminergic neuronal function. Neurosci Lett. 1984;50(1–3):67–71. doi: 10.1016/0304-3940(84)90464-6. [DOI] [PubMed] [Google Scholar]
- Zhang G, Alan WB. The rapid determination of γ-aminobutyric acid. Phytochemistry. 1997;44(6):1007–1009. doi: 10.1016/S0031-9422(96)00626-7. [DOI] [Google Scholar]
- Zhang LY, Sun MX. Selective determination of γ-aminobutyric acid, glutamate and alanine by mixed micellar electrokinetic chromatography and fluorescence detection. J Chromatogr A. 2005;1095(1–2):185–188. doi: 10.1016/j.chroma.2005.09.050. [DOI] [PubMed] [Google Scholar]