toxins Article

A Simple and Fast Procedure to Determine 3-Nitropropanoic Acid and 3-Nitropropanol in Freeze Dried Canadian Milkvetch (Astragalus canadensis) Huaizhi Liu 1 , Suqin Shao 1, * and Mike Schellenberg 2 1 2

*

Guelph Research and Development Centre, Agriculture and Agri-Food Canada, Guelph, ON N1G 5C9, Canada; [email protected] Swift Current Research and Development Centre, Agriculture and Agri-Food Canada, Swift Current, SK S9H 3X2, Canada; [email protected] Correspondence: [email protected]; Tel.: +1-226-217-8142

Received: 13 April 2017; Accepted: 27 June 2017; Published: 29 June 2017

Abstract: Canadian milkvetch (Astragalus canadensis) is a North American plant species in the legume family and some of this plant is fatally poisonous to livestock. The poisoning is attributed to the natural occurrence of notrotoxins, i.e., 3-nitropropanoic acid and 3-nitropropanol, present as aglycones and conjugated forms in the plant. Those compounds cause nitrite oxidization of hemoglobin and inhibition of cellular metabolism. To determine the toxicity of the plant, it is very important to develop an analytical method for the contents of the compounds in the plant. In this study, we have successfully developed an extraction procedure followed by HPLC-UV analysis to simultaneously analyze notrotoxins. The aglycones could be released from its conjugated forms in the freeze dried plant and extracted by water at room temperature. An HPLC-UV method using a Phenomenex Kinetex 2.6 µ F5 100 Å 100 × 4.6 mm column with pH 3.5 phosphonate buffer as mobile phase have been developed and validated for the detection of the two compounds at 210 nm. This developed procedure for the analysis of 3-nitropropanoic acid and 3-nitropropanol has proven simple and efficient and it has been successfully applied for batch sample analysis. Keywords: Canadian milkvetch; 3-nitropropanoic acid; 3-nitropropanol; plant toxin

1. Introduction Canadian milkvetch (Astragalus canadensis) is a native North American plant species in the legume family and can be found throughout many parts of Canada and the United States. Although this plant could be potentially used as forage, Canadian milkvetch could be fatally poisonous to livestock [1]. The poisoning is attributed to the natural occurrence of toxic aliphatic nitro compounds, glycosides of 3-nitropropanol (3-NPOH) and glucose esters of 3-nitropropanoic acid (3-NPA) [2], including miserotoxin, cibarian, karakin, hiptagin, and free 3-NPOH and 3-NPA. Additionally, Astragalus canadensis also contains oxotetrahydrofuranyl and isoxazolinone esters of 3-NPA [3]. Upon hydrolysis, miserotoxin will release 3-NPOH, Cibarian, Karakin, and hiptagin and other esters of 3-NPA results in the formation of 3-NPA. Both 3-NPOH and 3-NPA have been found toxic to chicks, sheep, horses, and cattle [4]. Although the toxic property of A. canadensis has been well known, limited research has been conducted on this plant. Pioneer work done in the 70s and early 80s focused on the identification of those toxic compounds and limited work on the quantification of those compounds was done through spectrophotometric method [2,5,6]. For example, miserotoxin, 3-nitro-1-propyl β-D-glucopyranoside was firstly identified in Astragalus miser by Stermitz and Lowry (1972) [7] and later semi-quantified in A. canadenis by Williams at al. (1975) using TLC method [2]. Spectrophotometric method, which was Toxins 2017, 9, 204; doi:10.3390/toxins9070204

www.mdpi.com/journal/toxins

Toxins 2017, 9, 204

2 of 10

used for quantification, was based on the nitrite ion released from nitro compounds with sulfanilic acid and 1-naphthylamine hydrochloride. However, the method was not specific and could interact with the pigments co-extracted from plants, and the reaction with 3-NPA and 3-NPOH was also found to be incomplete [8]. Ever since the mid-80s, not much work has occurred on the analysis of those compounds in the plant although some work was found on the analysis of 3-NPA and 3-NPOH in other matrices. For examples, previously, gas chromatography (GC) had been used for the analysis of 3-NPA in cheese after liquid-liquid extraction and derivatization with pentafluorobenzyl [9]. High performance liquid chromatography (HPLC) was used to quantify 3-NPA and 3-NPOH in bovine plasma after complicated sample preparation, including deproteinization, liquid-liquid extraction, and solid phase extraction [10]. Since the toxicity of A. canadensis varies a lot between different genomic backgrounds and/or growing conditions and there is increasing interest in using A. canadensis as forage, it is essential to develop a reliable analytical method to determine the contents of 3-NPA and 3-NPOH in the plant. Both 3-NPA and 3-NPOH are soluble in a variety of polar and non-polar solvents, including water, ethanol and diethyl ether. However, considering the complexity of the form of both compounds in the plant material, and the complexity of the material itself, the optimal solvent that can be used for efficient extraction of the compounds in the sample needs to be identified through systematic study. Additionally, due to the high solubility of both compounds in water, they are hard to be retained and separated in commonly used reverse phase C18 HPLC column. An HPLC method needs to be developed for the analysis of the two compounds. 2. Results and Discussion 2.1. Extraction of 3-NPA and 3-NPOH from Canadian Milkvetch Pure organic solvent ethanol, acetone or acetonitrile was found not able to extract neither 3-NPA or 3-NPOH from Canadian milkvetch samples when 10 mL solvent was used for 0.50 g freeze dried samples, even for a long time (48 h). However, when water or HCl was mixed with organic solvent, some 3-NPA was detected in the extract, indicating the role of water or HCl in extraction of the two compounds from the sample. Since pure 3-NPA or 3-NPOH are soluble in the organic solvent listed above, this result implied that 3-NPA or 3-NPOH exists in the plant mainly as conjugated forms, water or acid in the solvent could release them. This was proved by sequential extraction with water or acid followed by organic solvents. As Table 1 showed the amount determined in the same sample was significantly higher in the sequential extraction, suggesting pure water or acid is more efficient in releasing 3-NPA from the conjugates, though only a trace amount of 3-NPOH was found in all the extracts. In literature, inorganic acid (from 0.6 mol/L HCl to 2 mol/L of H2 SO4 ) had been used to hydrolysis the conjugates of 3-NPA and 3-NPOH in species of Astragalus from North American and European herbaria [5,7,11]. There have been reports that enzymes could release 3-NPOH from its conjugate forms in literature. For example, the microbial β-glucosidase and esterase in the rumen could rapidly hydrolyze and liberate free 3-NPA and 3-NPOH from their conjugates in the in vivo experiment [12]. 3-NPOH was found to be released very quickly from miserotoxin when mixed with rumen fluid in in vitro experiments [13]. One report believed that enzymatic hydrolysis of glycosides of 3-NPOH from miserotoxin was probably the reason why its yield was extremely low while using aqueous extraction [5]. Our experiment indicated that endogenous enzymes in A. canadensis could also release 3-NPA or 3-NPOH from the conjugates. Based on this, we hypothesized that by adding water to the freeze dried sample the autohydrolysis process by the enzymes in the plant material itself was induced. This observation led us to experiment with pure water extraction for the determination of 3-NPA and 3-NPOH in freeze dried Canadian milkvetch samples. Time course of the concentration of 3-NPA and 3-NPOH extracted by water from Canadian milkvetch is shown in Figure 1. It showed that after water was added to the sample, the concentration of 3-NPA and 3-NPOH in the extract kept increasing during the first 24 h and then it reached a plateau.

Toxins 2017, 9, 204

3 of 10

Toxins 2017, 9, 204 3 offorms 10 Possible reasons for this could be that time is needed to release 3-NPA or 3-NPOH from other of the compounds in the plant material and/or it is a slow distribution equilibrium (24 h) of 3-NPA or from other forms of the compounds in the plant material and/or it is a slow distribution equilibrium 3-NPOH between extraction solvents and the plant material. To determine theTo most possible (24 h) of 3-NPA or 3-NPOH between extraction solvents and the plant material. determine thereason for this, we compared hot water extraction and room temperature extraction. most possible reason for this, we compared hot water extraction and room temperature extraction.

Table Table 1. Measured amount of 3-NPA (mg/g) driedCanadian Canadian milkvetch sample by different 1. Measured amount of 3-NPA (mg/g) in in freeze freeze dried milkvetch sample by different extraction. extraction.

Extract

H2O:Acet H2O:Acetonit H2O:Etha H2 O:Acetone H2 O:Acetonitrile H2 O:Ethanol Extract nol (2:8) (2:8) one (2:8) (2:8)rile (2:8) (2:8) 1.4

2.3

2.3

2.3 2.3

2.9 2.9

3.2 3.2

2.0 2.0

6.2

6.8

6.8

6.2 6.2

9.8 9.8

10.3 10.3

4.2 4.2

Concentration of 3-NPA and 3-NPOH (mg/L)

One-TimeOne-Time 1.4 ExtractionExtraction Sequential Sequential 6.2 Extraction Extraction

1 mol/L 11 mol/L 1 1mol/L 1 mol/L1 mol/L mol/L mol/L HCL:Et 1 mol/L HCL:Acet HCL:Acetonitril H2 O HCL:Acetonitrile hanol HCL:Ethanol HCL:Acetone HCL HCL one (2:8) e (2:8) (2:8) (2:8) (2:8) (2:8) 4.6

H2 O

4.69.3

9.3

600 500

3-NPA 3-NPOH

400 300 200 100 0 0

12

24

36

48

60

72

Time (Hour) Figure 1. Time course of concentration of 3-NPA and 3-NPOH in the water extract of Canadian

Figure 1. Time course of concentration of 3-NPA and 3-NPOH in the water extract of Canadian milkvetch sample. milkvetch sample. Table 2 shows that heated water only extracted 22% of 3-NPA and 65% of 3-NPOH from the Canadian milkvetch samples water compared the water 22% at room temperature, and longer extraction Table 2 shows that heated onlytoextracted of 3-NPA and 65% of 3-NPOH from the time did not make any difference. More than three times the amount of 3-NPA was extracted by time Canadian milkvetch samples compared to the water at room temperature, and longer extraction water at room temperature than heated water with a 4 h extraction. This indicated the time course did not make any difference. More than three times the amount of 3-NPA was extracted by water at results of the room temperature water extraction were not due to a slow equilibrium, but rather slow room temperature than heated water with a 4 h extraction. This indicated the time course results of the release of free 3-NPA and 3-NPOH from the plant material. Since hot water resulted in a significant room temperature water extraction were not due to a slow equilibrium, but rather slow release of free reduction of the amount of the extracted compounds, it suggested that the release of 3-NPA or 3-NPA3-NPOH and 3-NPOH from the plantofmaterial. Sinceforms hot water resulted in aexisting significant was done by autolysis the conjugated by certain enzyme in the reduction plant itself of the amount the extracted compounds, it suggested the release of 3-NPA 3-NPOH was done by andoftime is needed for the enzymatic hydrolysis.that Heated water stopped the or activity of enzymes, autolysis of the extraction conjugated forms bynot certain enzyme existing inofthe plantoritself and the time is needed for extending time would increase the concentration 3-NPA 3-NPOH, measured 3-NPA orhydrolysis. 3-NPOH from heatedwater waterstopped most probably came of from some non-denatured enzyme time the enzymatic Heated the activity enzymes, extending extraction activity during the temperature development in the sample. The extractions of Canadian milkvetch would not increase the concentration of 3-NPA or 3-NPOH, the measured 3-NPA or 3-NPOH from by pure organic solvents (ethanol acetone and acetonitrile) or 80% organic solvents were not heatedsamples water most probably came from some non-denatured enzyme activity during the temperature efficient, since the high amount of organic solvents denatured the enzyme. development in the sample. The extractions of Canadian milkvetch samples by pure organic solvents (ethanol acetone acetonitrile) or 80% organic solvents wereofnot efficient, since the highin amount of Table 2.and Effect of water temperature on the measured content 3-NPA and 3-NPOH (mg/g) organic solvents denatured enzyme. freeze dried Canadianthe milkvetch sample. Temperature Time (H) 3-NPA 3-NPOH Table 2. Effect of water temperature on the measured content of 3-NPA and 3-NPOH (mg/g) in freeze 95 °C 1.5 2.02 0.15 dried Canadian milkvetch sample. 95 °C 4 2.49 0.15 95 °C 48 2.31 0.14 Temperature Time (H) 3-NPA 3-NPOH 25 °C 4 8.48 0.20 ◦ 2.02 0.15 25 °C 95 ◦ C 48 1.5 9.27 0.23 95 C 95 ◦ C 25 ◦ C 25 ◦ C

4 48 4 48

2.49 2.31 8.48 9.27

0.15 0.14 0.20 0.23

Toxins 2017, 9, 9, 204 204 Toxins 2017,

44 of of 10 10

As shown in Table 1 the amount of 3-NPA released by 1 mol/L HCl was significantly smaller in Table 1 the possibly amount of 3-NPA by 1 mol/L HCland was3-NPOH significantly than As thatshown of water extraction, due to thereleased esterification of 3-NPA undersmaller acidic than that of water extraction, possibly due to the esterification of 3-NPA and 3-NPOH under acidic condition. The detected amount of 3-NPA by sequential extraction of 1 mol/L HCl followed by condition. The detected amount of 3-NPA by sequential extraction 1 mol/L HCl followed by acetone or acetonitrile showed no significant difference with the waterofextraction. Besides the more acetone or acetonitrile showed no significant difference with the Besides the more complicated sample preparation by sequential extraction, thewater peakextraction. shape, especially that of complicated sample preparation by sequential extraction, the peak shape, especially that of 3-NPOH, 3-NPOH, showed severe band broadening when the sequential extract was injected into HPLC showed band broadening whenas thediscussed sequential into HPLC underthat the under thesevere developed HPLC condition, in extract Section was 2.2. injected All the results supported developed HPLC condition, as discussed in Section 2.2. All the results supported that water extraction water extraction at room temperature is the best procedure for the analysis of 3-NPA and 3-NPOH in at room temperature the best procedure for the analysis supports of 3-NPAthat andwater 3-NPOH in freeze dried freeze dried Canadianismilkvetch samples. Our experiment extraction at room Canadian milkvetch Our experiment that water extraction at room temperature temperature for 24 hsamples. is the optimal proceduresupports for extraction of 3-NPA and 3-NPOH from freeze for 24 h is the optimal procedure for extraction 3-NPA and 3-NPOH freeze Canadian dried Canadian milkvetch samples. Therefore,of this procedure was from adopted fordried batch sample milkvetch samples. Therefore, this procedure adopted for batch sample treatment However, for 3-NPA and treatment for 3-NPA and 3-NPOH analysiswas in freeze dried Canadian milkvetch. one 3-NPOH analysis freeze dried Canadian milkvetch. one special note thisdried extraction special note for thisinextraction procedure is that it couldHowever, only be applied to fresh or for freeze plant procedure as is that it could only be applied to fresh or freezeExtended dried plant materials it depends on the materials it depends on the endogenous enzymes. storage or as drying at elevated endogenous enzymes. storageoforthe drying at elevated temperature could leadwater to deactivation temperature could leadExtended to deactivation enzymes, and therefore the proposed extraction of the enzymes, and therefore the proposed water extraction will not lead to the total release of free will not lead to the total release of free 3-NPA or 3-NPOH. 3-NPA or 3-NPOH. 2.2. HPLC Conditions for the Analysis of 3-NPA and 3-NPOH 2.2. HPLC Conditions for the Analysis of 3-NPA and 3-NPOH Both 3-NPA and 3-NPOH are polar small molecules and hardly retained on the reverse phase, and 3-NPOH are polar small molecules and ahardly retainedKinetex on the reverse such Both as the3-NPA most commonly used C18 columns. Here we chose Phenomenex 2.6 μ F5 phase, 100 Å such as the most commonly used C18 columns. Here we chose a Phenomenex Kinetex 2.6 µ F5 100 Å 100 × 4.6 mm column to study the retention and separation of 3-NPA and 3-NPOH. Combining 100 × 4.6 mm column to study the retention and separation of 3-NPA and 3-NPOH. Combining core-shell technology and five interaction mechanisms with Pentafluorophenyl with TMS core-shell technology and five interaction mechanisms with Pentafluorophenyl with TMS endcapping endcapping as stationary phase, this column has been proven for polar compound retention with as stationary phase, this column has been proven for polar compound retention with polar mobile polar mobile phase. The UV spectra of 3-NPA and 3-NPOH (Figure 2) showed that the optimal UV phase. Theof UV spectra 3-NPA and 3-NPOH 2) wavelength, showed thatthe theabsorbance optimal UV absorption absorption 3-NPA andof3-NPOH is 210 nm and(Figure at higher dramatically of 3-NPA Therefore and 3-NPOH is 210 nmdetection and at higher wavelength, the nm) absorbance dramatically dropped. dropped. a very low wavelength (e.g., 210 was chosen to measure both Therefore a very low detection wavelength (e.g., 210 nm) was chosen to measure both compounds. compounds. This has limited the choice of mobile phase and modifiers too. UV transparent solvents Thiswater has limited the choice of phase modifiers too. UV solvents like and like and acetonitrile aremobile preferred toand methanol, acetone, etc.transparent and isocratic elution is water preferred acetonitrile are preferred to methanol, acetone, etc. and isocratic elution is preferred to gradient elution. to gradient elution.

Figure 2. UV spectra of 3-nitropropanoic acid (3-NPA) and 3-nitropropanol (3-NPOH). Figure 2. UV spectra of 3-nitropropanoic acid (3-NPA) and 3-nitropropanol (3-NPOH).

Due to the acidic nature of the compounds, acidic buffers have been tried for the separation of Due to the acidic nature of the compounds, acidic buffers have been tried for the separation of the the two compounds. With mobile phase of 10 mM ammonium acetate (pH 6.58) or 10 mM two compounds. With mobile phase of 10 mM ammonium acetate (pH 6.58) or 10 mM ammonium ammonium acetate with 0.02% acetic acid (pH 4.85), 3-NPOH was retained, but 3-NPA was not. acetate with 0.02% acetic acid (pH 4.85), 3-NPOH was retained, but 3-NPA was not. When 0.1% When 0.1% acetic acid (pH 2.93) was used as mobile phase, both 3-NPA and 3-NPOH were retained acetic acid (pH 2.93) was used as mobile phase, both 3-NPA and 3-NPOH were retained and they had and they had baseline separation, resolution between two peaks was 2.74, the capacity factor of baseline separation, resolution between two peaks was 2.74, the capacity factor of 3-NPA and 3-MPOH 3-NPA and 3-MPOH were 1.729 and 1.781, respectively. However, a negative peak was present in were 1.729 and 1.781, respectively. However, a negative peak was present in the chromatograph with the chromatograph with this mobile phase at this wavelength, which made it impractical to analyze this mobile phase at this wavelength, which made it impractical to analyze real samples. Additionally, real samples. Additionally, it only worked for the 3-NPA and 3-NPOH in water or in 0.1% acetic it only worked for the 3-NPA and 3-NPOH in water or in 0.1% acetic acid, whereas huge solvent acid, whereas huge solvent peaks appeared and disturbed the separation when 10% acetonitrile, peaks appeared and disturbed the separation when 10% acetonitrile, acetone, or ethanol was used acetone, or ethanol was used as reconstitution solvent. With 25 mM potassium dihydrogen as reconstitution solvent. With 25 mM potassium dihydrogen phosphate in water (pH 4.32) used as phosphate in water (pH 4.32) used as mobile phase, 3-NPOH were retained, but 3-NPA was still not mobile phase, 3-NPOH were retained, but 3-NPA was still not reasonably retained and the peak of reasonably retained and the peak of 3-NPOH showed front tailing and peak broadening. However, 3-NPOH showed front tailing and peak broadening. However, when 25 mM potassium dihydrogen when 25 mM potassium dihydrogen phosphate in water was adjusted to pH 3.0 with phosphoric

Toxins 2017, 9, 204 Toxins 2017, 9, 204

5 of 10 5 of 10

acid and used as mobile phase, to allpH the3.0 problems were solved. Further showed the phosphate in water was adjusted with phosphoric acid and usedstudies as mobile phase,that all the blockage of thesolved. flow path of 1290 HPLC system occurred sometimes thepath salt of if not problems were Further studies showed that the blockage of theby flow 1290washed HPLC immediately with water after use. To solve this problem, potassium dihydrogen phosphate system occurred sometimes by the salt if not washed immediately with water after use. To solve was this replaced potassium by ammonium dihydrogen phosphate, and theby concentration was halved phosphate, to 12.5 mM. A problem, dihydrogen phosphate was replaced ammonium dihydrogen and typical chromatogram of the to standard solution of 20 mg/L 3-NPAof and shown of in 20 Figure the concentration was halved 12.5 mM. A typical chromatogram the3-NPOH standardissolution mg/L3 with pH 3.0 12.5 mM ammonium phosphoric buffer on a Phenomenex Kinetex 2.6 μ F5 100 Å 3-NPA and 3-NPOH is shown in Figure 3 with pH 3.0 12.5 mM ammonium phosphoric buffer100 on a× 4.6 mm at 210Kinetex nm. 2.6 µ F5 100 Å 100 × 4.6 mm at 210 nm. Phenomenex

Figure A typical typical HPLC HPLCchromatogram chromatogramofofmixture mixturestandards standardsofof3-NPA 3-NPAand and3-NPOH 3-NPOHatat2020 mg/L Figure 3. 3. A mg/L in in water with 12.5 mM phosphoric buffer as mobile phase a Phenomenex Kinetex F5 water with pHpH 3.03.0 12.5 mM phosphoric buffer as mobile phase on aon Phenomenex Kinetex 2.6 μ2.6 F5µ100 100 Å column. Å column.

The capacity capacity factor factor (k’) (k’) of of 3-NPA 3-NPA and and 3-NPOH 3-NPOH were were 2.06 2.06 and and 2.49, 2.49, respectively respectively with with pH pH 3.0 3.0 The ammonium phosphate buffer as mobile phase. However, as water extract of a freeze-dried sample ammonium phosphate buffer as mobile phase. However, as water extract of a freeze-dried sample is is a a very complex solution, need to verify this condition is specific the analysis of the plant very complex solution, we we need to verify if thisifcondition is specific for the for analysis of the plant extracts. extracts. done through pHmobile of the mobile gradually (theofpH the sample This wasThis donewas through altering altering the pH the of the phase phase gradually (the pH theofsample was was adjusted accordingly) and comparing the UV spectra of the peaks of authentic 3-NPA adjusted accordingly) and comparing the UV spectra of the peaks of authentic 3-NPA and 3-NPOH and and 3-NPOH that The of the samples. Theand chromatograms thepeaks UV of spectra of the peaks of the that of theand samples. chromatograms the UV spectraand of the the standards, the samples standards, the samples sampleswith and elution the spiked samples at different pH were presented in the and the spiked at different pHwith wereelution presented in the supplement (Figures S1–S17). supplement (Figures S1–S17). Figure S1 showed that the peaks of 3-NPA and 3-NPOH were Figure S1 showed that the peaks of 3-NPA and 3-NPOH were not separated at pH 2.0, and the k’not of separated pH 2.0,was andthe thesame k’ of2.58. 3-NPA and was the same 2.58. pH 2.5separated the two 3-NPA andat3-NPOH At pH 2.53-NPOH the two compounds were notAt baseline compounds were separated (Figure S2) and 3-NPA and was 2.46 and (Figure S2) and thenot k’ofbaseline 3-NPA and 3-NPOH was 2.46 and the 2.56k’of respectively. At3-NPOH pH 3.0, the two peaks 2.56 respectively. At pH 3.0, the two peaks of the standards were baseline separated with k’of 3-NPA of the standards were baseline separated with k’of 3-NPA and 3-NPOH 2.06 and 2.49 respectively and 3-NPOH andboth 2.49 of respectively S3). 3-NPOH At this pH, both of the peaks of 3-NPA (Figure S3). At 2.06 this pH, the peaks of(Figure 3-NPA and in the sample were symmetric, theand UV 3-NPOH in the sample were symmetric, the UV spectra of 3-NPA of the standard and the sample spectra of 3-NPA of the standard and the sample were exactly the same (Figure S4), but the spectrum were same (Figure S4), butwas the spectrum thesame peak as of that 3-NPOH the sample was S5), not of theexactly peak ofthe 3-NPOH of the sample not exactlyofthe of theofstandard (Figure exactly the same as that of the standard (Figure S5), indicating some kind of interference for the indicating some kind of interference for the analysis of 3-NPOH in the sample under this pH. When the analysis of 3-NPOH in the sample under this pH. When the pH was adjusted to 3.5, k’ of 3-NPA and pH was adjusted to 3.5, k’ of 3-NPA and 3-NPOH were 1.48 and 2.50 respectively with peaks baseline 3-NPOH were 1.48S6). and 2.50ofrespectively withofpeaks separated (Figure Both of thewere UV separated (Figure Both the UV spectra 3-NPAbaseline and 3-NPOH of the peaksS6). of the sample spectra of 3-NPA and 3-NPOH of the peaks of the sample were exactly the same as that of the exactly the same as that of the standards (Figures S7 and S8), implying the specificity for the analysis standards (Figures S7 and S8), implying the specificity for the analysis of the two compounds under of the two compounds under this condition. The side peak eluted just behind the peak of 3-NPOH was this condition. The sideoverlapped peak elutedwith just3-NPOH behind the peak wask’possibly compound possibly the compound at pH 3.0.ofAt3-NPOH pH 4.0, the of 3-NPAthe and 3-NPOH overlapped with 3-NPOH at pH 3.0. At pH 4.0, the k’ of 3-NPA and 3-NPOH were 0.85 and 2.50 were 0.85 and 2.50 respectively with peaks baseline separated (Figure S9), and the UV spectra of 3-NPA respectively peaks baseline and the UV spectra 3-NPA and 3-NPOH of and 3-NPOHwith of the sample wereseparated the same(Figure as that S9), of the standard (FiguresofS10 and S11). At pH 5.0 the sample were the same as that of the standard (Figure S10 and S11. At pH 5.0 and 6.0 3-NPA and and 6.0 3-NPA and 3-NPOH were baseline separated (Figures S12 and S15), and the k’ of 3-NPA and 3-NPOH were were 0.40 baseline (FigureatS12 and and the k’ of respectively 3-NPA and 3-NPOH 3-NPOH and separated 2.50 respectively pHand 5.0 S15), and 0.32 2.53 at pH.6.0.were The0.40 UV and 2.50ofrespectively at 3-NPA pH 5.0 and and 2.53 respectively at pH.6.0. Theas UVthat spectra ofstandards the peaks spectra the peaks of and 0.32 3-NPOH of the sample were the same of the of 3-NPAS13, andS14, 3-NPOH of S17) the sample the conditions. same as thatHowever, of the standards (Figure S13, S14, S16 (Figures S16 and at thosewere two pH there was peak tailing under and S17) at those two pH conditions. However, there was peak tailing under those two pH those two pH conditions for 3-NPA and it was also very close to void volume (Figures S12 and S15). conditions for of 3-NPA it was alsoasvery close to void (Figures S12 and S15).result, The The correlation k’ andand pH was shown Figure 4. Taking intovolume the consideration of the above correlation of k’ and pH was shown as Figure 4. Taking into the consideration of the above result, pH pH 3.5–4.0 was demonstrated the optimal pH value for the analysis of 3-NPA and 3-NPOH in real 3.5–4.0 was demonstrated the optimal pH value for the analysis of 3-NPA and 3-NPOH in real plant plant samples. samples.

Toxins 2017, 9, 204

6 of 10

Toxins 2017, 9, 204

6 of 10

Capacity Factor ( K')

3

2

3-NPOH 3-NPA

1

0 2

3

4

5

6

pH

Figure 4. Effect of pH capacityfactor factor of of 3-NPA 3-NPA and separated on aonPhenomenex Figure 4. Effect of pH onon thethe capacity and3-NPOH 3-NPOH separated a Phenomenex Kinetex 2.6 μ F5 100 Å 100 × 4.6 mm column with ammonium phosphate buffer. Kinetex 2.6 µ F5 100 Å 100 × 4.6 mm column with ammonium phosphate buffer.

2.3. Method Validation

2.3. Method Validation

The linearity and LODs and LOQs of the HPLC analysis was done using 3-NPA and 3-NPOH

The linearity and due LODs andfact LOQs the matrix HPLCthat analysis was done 3-NPA using and 3-NPA and 3-NPOH standard solutions, to the that aofplant does not contain 3-NPOH was standard solutions,Therefore, due to thethe fact that a plant matrix that doesofnot contain 3-NPOH not available. linearity and LODs and LOQs this study 3-NPA are the and linearity and was not available. the analysis, linearityand andcould LODsonly and be LOQs study are and sensitivity sensitivityTherefore, of the HPLC usedofasthis reference for the real linearity plant samples. The HPLC method demonstrated an as excellent linear between the area and of thedeveloped HPLC analysis, and could only be used reference forrelationship real plant samples. Thepeak developed HPLC 2) ≥ 0.9997 (Table 3). the concentration for both 3-NPA and 3-NPOH with correlation coefficients (R method demonstrated an excellent linear relationship between the peak area and the concentration for 2 ) ≥for linearity ranges covered orders from 1 to 600(R mg/L both compounds, thelinearity LODs were both The 3-NPA and 3-NPOH with three correlation coefficients 0.9997 (Table 3). The ranges ≤0.038 ng/mL and the LOQs were ≤0.123 ng/mL, which is satisfactory for actual samples analysis. covered three orders from 1 to 600 mg/L for both compounds, the LODs were ≤0.038 ng/mL and the The recovery of each of the compounds in the freeze dried samples was studied by adding a certain LOQs were ≤0.123 ng/mL, which is satisfactory for actual samples analysis. The recovery of each of amount of 3-NPA and 3-NPOH water stock solution to the freeze dried samples and the spiked the compounds in the freeze dried samples was studied by adding a certain amount of 3-NPA and samples went through the analytical procedure including sample preparation and HPLC analysis as 3-NPOH waterinstock solution the freeze samples and spiked went through developed this study. Theto spiking level dried included 100, 400, andthe 1000 ng/mgsamples dry weight (DW). The the analytical procedure including sample preparation and HPLC analysis as developed in this results are listed in Table 4. Both compounds at the different spiking levels showed good recoveries,study. The spiking 100, 400, and 1000 ng/mg dry weight (DW). results arevalues listedof inthe Table 4. ranginglevel from included 97.4 to 116.8%, indicating the accuracy of the method. TheThe obtained RSD were lower than 3.28% for levels both compounds in the real samples, indicating the to high Both procedure compounds at the different spiking showed good recoveries, ranging from 97.4 116.8%, precision the method 5). These data indicate the of method is highly accurate and than indicating the of accuracy of the(Table method. The obtained RSDthat values the procedure were lower Moreover,inthe method is much and has of a lower cost when 3.28%reproducible. for both compounds thedeveloped real samples, indicating thesimpler high precision the method (Table 5). compared to the traditionally used spectrometric method or TLC method. The extract compounds These data indicate that the method is highly accurate and reproducible. Moreover, the developed can be baseline separated on the Phenomenex Kinetex 2.6 μ F5 100 Å 100 × 4.6 mm column in 6 min. method is much simpler and has a lower cost when compared to the traditionally used spectrometric All these suggest that the method is valid and highly efficient, therefore can be used to analyze method or TLC method. The extract compounds can be baseline separated on the Phenomenex Kinetex 3-NPA and 3-NPOH in freeze dried Canadian milkvetch. 2.6 µ F5 100 Å 100 × 4.6 mm column in 6 min. All these suggest that the method is valid and highly efficient, therefore canofbe usedvalidation to analyze 3-NPA andmethod 3-NPOH in analysis freeze dried Canadian milkvetch. Table 3. Data method of the proposed for the of 3-NPA and 3-NPOH in Canadian milkvetch 1.

Table 3. Data of method validation of the proposed method for the analysis of 3-NPA and 3-NPOH in Compounds Linear Range (μg/mL) R2 LOD (μg/mL) LOQ(μg/mL) 1. Canadian milkvetch 3-NPA 1–600 0.9997 0.038 0.123 3-NPOH 1–600 0.9998 0.036 0.116 2 Compounds Linear coefficient; Range (µg/mL) LOD (µg/mL) LOQ(µg/mL) R 1 R2: determination LOD: limit of detection; LOQ: limit of quantification. 3-NPA 1–600 0.9997 0.038 0.123 0.9998 0.036 0.116 addition Table3-NPOH 4. Recoveries of 3-NPA 1–600 and 3-NPOH following the developed procedure by standard 1 R=2 :3)determination coefficient; LOD: limit of detection; LOQ: limit of quantification. method (n Percent Recoveries (%) in in Freeze Dried Canadian Milkvetch Samples at Different Spiking (ng/mg DW) Table Compounds 4. Recoveries of 3-NPA and 3-NPOH following theLevels developed procedure by standard addition 100 400 1000 method (n = 3). 3-NPA 3-NPOH

100.09 97.4

Compounds

3-NPA 3-NPOH

102.7 106.8 116.8 Percent Recoveries (%) in in Freeze Dried103.5 Canadian Milkvetch Samples at Different Spiking Levels (ng/mg DW) 100

400

1000

100.09 97.4

102.7 116.8

106.8 103.5

Toxins 2017, 9, 204

7 of 10

Table 5. Relative standard deviation (%) of the concentration of 3-NPA and 3-NPOH in the Canadian milkvetch sample analyzed by the proposed procedure (n = 5). Compounds

Measured Amount (mg/g)

RSD (%)

3-NPA 3-NPOH

16.65 1.20

1.91 3.28

3. Conclusions In conclusion, a simple and efficient procedure for the determination of 3-nitropropanoic acid and 3-nitropropanol in freeze dried Canadian milkvetch has been developed through this study. Since those two compounds exist as both aglycones and conjugated forms, water extraction for 24 h with shaking has been proven to be sufficient to release and extract them from the freeze-dried plant material. The extracted compounds could be based line separated without any interference on a Phenomenex Kinetex 2.6 µ F5 100 Å 100 × 4.6 mm column with pH 3.5 phosphonate buffer as mobile phase, and can be detected at 210 nm. The developed method has been validated for its accuracy, repeatability, and robustness. It has been successfully applied for batch sample analysis. 4. Materials and Methods 4.1. Chemicals and Reagents 3-nitro-1-propanoic acid was purchased from Sigma-Aldrich (St. Louis, MO, USA), and 3-nitro-1-propanol was from Merck Millipore (Darmstadt, Germany). Stock solutions of each standard were prepared in methanol or water at 2.0 mg/mL level and were stored at −20 ◦ C. Stock solution of mixed standards was prepared by mixing each individual standard stock solution in equal volume with a final concentration at 1.0 mg/L level. HPLC grade acetonitrile, methanol, and dichloromethane were supplied by Caledon Laboratories Chemicals (Georgetown, DC, Canada). High purity water was made by a Barnstead Nanopure purification system (Thermo Scientific, Dubuque, IA, USA). A. canadensis was harvested from the experimental field of the Swift Current Research and Development Centre of Agriculutre-Agri-Food Canada in Saskatchewan, SK, Canada. The sample was then freeze dried and sheared with a SmartGrind (Black & Decker Inc., Mississauga, ON, USA) for about 2 min, sieved through an 80 mesh sieve, and then stored at −20 ◦ C for further analysis. 4.2. Extraction of 3-NPA and 3-NPOH from A. canadensis Extraction procedures used in this study include one-step solvent extraction, two-step sequential extraction and pure water extraction. The solvent systems used in one-step extraction include (Table 1), 100% acetone, 100% acetonitrile, 100% ethanol, mixture of water and acetone (2:8), mixture of water and acetonitrile (2:8), mixture of water and ethanol (2:8), 1 mol/L HCl and acetone (2:8), 1 mol/L HCl and acetonitrile (2:8), and 1 mol/L HCl and ethanol (2:8). The procedure of one step extraction is that 0.5 gram of dried A. canadensis was vortexed with 10 mL each of the above solvent system at room temperature for 1 min and then shaken for 4 h at Heidolph Titramax 1000 platform shaker (Heidolph North America, Elk Grove Village, IL, USA). The suspension was centrifuged at 3000 g for 15 min at room temperature, the supernatant was then filtered through a 0.45 µm PVDF filter (Whatman, Sanford, ME, USA) and stored at 4◦ before HPLC analysis or further processing within one week. The procedure for two-step sequential extraction is that, to 0.50 gram of freeze dried A. canadensis, a certain amount of first solvent was added and vortexed for one minute and the mixture was shaken at room temperature for 4.0 h and then the second solvent was added. The second extraction was also conducted on a Heidolph Titramax 1000 platform shaker (Heidolph North America, Elk Grove Village, IL, USA) at room temperature for 2.0 h. After centrifugation at 3000 g for 15 min at room temperature, the supernatant was collected and filtered through a 0.45 µm PVDF filter (Whatman, Sanford, ME, USA) and stored at 4 ◦ C before HPLC analysis or further processing within one week. The solvent

Toxins 2017, 9, 204

8 of 10

system used in the two step sequential extraction includes (Table 1): 2 mL water followed by 8 mL acetone, 2 mL of water followed by 8 mL acetonitrile, 2 mL of water followed by 8 mL ethanol, 2 mL of 1 mol/L HCl followed by 8 mL acetone, 2 mL of 1 mol/L HCl followed by 8 mL acetonitrile, 2 mL of 1 mol/L HCl followed by 8 mL ethanol. Water extract: To 0.50 g dried A. canadensis sample in a 15 mL centrifuge tube 10.0 mL Milli-Q water was added. The mixture was then vortexed and kept on shaking on a Heidolph MultiReax (Heidolph North America, Elk Grove Village, IL, USA) at a speed of 1500 rpm at room temperature. At a different time, a 200 µL of sample was quickly pipetted from the middle of the centrifuge tube into a Ultrafree-MC microcentrifuge filter (Sigma-Aldrich, Mississauga, ON, Canada, Capacity 500 µL durapore PVDF membrane, pore size 0.22 µm), the microcentrifuge filter was centrifuged at 11,000 g at an Eppendorf centrifuge 5417R (Eppendorf Canada, Mississauga, ON, Canada) with temperature set at 4 ◦ C for 10 min. The filtrate was then diluted 10 times and transferred to an autosampler vial with inserts for HPLC analysis. Hot water extract: To evaluate the effect of heated water on the extraction of 3-NPA and 3-NPOH from Canadian milkvetch, 10 mL of boiling water was added to 0.50 g of dry Canadian milkvetch samples in a Supelco 15 mL of clear glass vial, 10.0 mL of boiling water was added and capped with PTFE/Silicone lined cap, the mixture were then put into a heating block (Boekel Industries Inc. Feasterville-Trevose, PA, USA,), and the temperature was kept at 95 ◦ C. The mixture was manually shaken frequently. After pre-set time, the mixture were removed from the block, cooled down and centrifuged at 3000 rpm for 15 min, each supernatant was diluted 10 times and transferred to an HPLC autosampler vial and concentration of 3-NPA and 3-NPOH was determined by HPLC. To compare, water at room temperature was used to extract Canadian milkvetch. 4.3. HPLC Conditions HPLC conditions: An Agilent 1290 series HPLC (Agilent Technology, Palo Alto, CA, USA) equipped with a binary pump, an inline degasser, a thermostatic autosampler, and a DAD was used for the identification and quantification of 3-NPA and 3-NPOH in the samples. A Phenomenex Kinetex 2.6 µ F5 100 Å 100 × 4.6 mm column (Torrance, CA, USA) was used for the separation of 3-NPA and 3-NPOH. The isocratic mobile phase was ammonium phosphate buffer at different pH at a flow rate of 1.0 mL/min for a run time of 6.5 min. The injection volume was 1 or 2 µL for all standards and samples. The DAD collected data from 190–600 nm, and absorbance at 210 nm only was used to monitor and quantify 3-NPA and 3-NPOH. 3-NPA and 3-NPOH were identified by a combination of the retention time in HPLC chromatograms, and UV spectra pattern of pure standard compounds. 4.4. Quantification All calibration solutions (1.00, 2.00, 5.0, 10.0, 20.0, 50.0, 100.0, and 200.0 mg/L) were prepared by serial dilution of the individual stock solutions with water. 2 µL of each solution were injected into the HPLC system, all in duplicate, and a calibration curve was generated between the HPLC peak area of the compound and the concentration. Peak areas were integrated automatically and the correlation coefficients were obtained by ChemStation® (Rev. C. 01.05 [35], Agilent Technology, Palo Alto, CA, USA). The contents of 3-NPA and 3-NPOH were expressed in mg/g DW. 4.5. Method Validation The instrumental linearity was determined with known concentrations of mixed standard solutions (1.00, 2.00, 5.0, 10.0, 20.0, 50.0, 100.0, 200.0, 500.0, and 600.0 mg/L). 1 µL of each solution were directly injected into the HPLC system, all in triplicate, and a correlation was generated between the HPLC peak area of the compound and the concentration. The precision of the method was evaluated by analyzing the same sample (n = 6) six times and the concentration of each compound in the sample was determined and the relative standard deviations were calculated. The recovery rate was conducted according to the standard addition method [14]. Water stock solutions of the standards were added to

Toxins 2017, 9, 204

9 of 10

the freeze dried A. canadensis samples at different concentrations. The spiked samples went through the sample preparation and HPLC analysis and the recovery percentage was determined. The recovery experiment was done in triplicate. To determine the limit of detection (LOD) and limit of quantification (LOQ) of the HPLC analysis, low concentrations of working standard solutions of 3-NPA and 3-NPOH were prepared by series dilution of 1 mg/mL of stock solution. Six replicates of 2 µL of working standards and water blank were injected into the HPLC. Peak area and peak height were recorded. Limit of quantitation (LOQ) were calculated with 10 times the signal to noise of peak height and limit of detection LOD with 3 times the signal to noise of peak height. Supplementary Materials: The following are available online at www.mdpi.com/2072-6651/9/7/204/s1, Figure S1: Overlaid HPLC chromatograms of mixture standards of 3-NPA and 3-NPOH at 10 ppm in water and spiked Canadian milkvetch sample, with pH 2.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S2: Overlaid HPLC chromatograms of mixture standards of 3-NPA and 3-NPOH at 10 ppm in water and spiked Canadian milkvetch sample, with pH 2.5 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S3: Overlaid HPLC chromatograms of mixture standards of 3-NPA and 3-NPOH at 10 ppm in water, Canadian milkvetch sample and spiked Canadian milkvetch samples, with pH 3.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S4: UV spectra of the peak of 3-NPA in Canadian milkvetch under the HPLC condition with pH 3.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S5: UV spectra of the peak of 3-NPOH in Canadian milkvetch under the HPLC condition with pH 3.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S6: Overlaid HPLC chromatograms of mixture standards of 3-NPA and 3-NPOH at 10 ppm in water, Canadian milkvetch sample and spiked Canadian milkvetch samples, with pH 3.5 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S7: UV spectra of the peak of 3-NPA in Canadian milkvetch under the HPLC condition with pH 3.5 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S8: UV spectra of the peak of 3-NPOH in Canadian milkvetch under the HPLC condition with pH 3.5 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S9: Overlaid HPLC chromatograms of mixture standards of 3-NPA and 3-NPOH at 10 ppm in water, Canadian milkvetch sample and spiked Canadian milkvetch samples, with pH 4.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S10: UV spectra of the peak of 3-NPA in Canadian milkvetch under the HPLC condition with pH 4.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S11: UV spectra of the peak of 3-NPOH in Canadian milkvetch under the HPLC condition with pH 4.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S12: Overlaid HPLC chromatograms of mixture standards of 3-NPA and 3-NPOH at 10 ppm in water, Canadian milkvetch sample and spiked Canadian milkvetch samples, with pH 5.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S13: UV spectra of the peak of 3-NPA in Canadian milkvetch under the HPLC condition with pH 5.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S14: UV spectra of the peak of 3-NPOH in Canadian milkvetch under the HPLC condition with pH 5.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S15: Overlaid HPLC chromatograms of mixture standards of 3-NPA and 3-NPOH at 10 ppm in water, Canadian milkvetch sample and spiked Canadian milkvetch samples, with pH 6.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S16: UV spectra of the peak of 3-NPA in Canadian milkvetch under the HPLC condition with pH 6.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column, Figure S17: UV spectra of the peak of 3-NPOH in Canadian milkvetch under the HPLC condition with pH 6.0 12.5 mM ammonium phosphate buffer as moble phase on a Phenomenex Kinetex 2.6 µ F5 100 Å column. Acknowledgments: This project was supported by Agriculture and Agri-Food Canada. Author Contributions: H.L. and S.S. conceived and designed the experiments; H.L. performed the experiments; H.L. and S.S. analyzed the data; M.K. contributed reagents, materials and analysis tools; H.L. and S.S. wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

References 1.

Williams, M.C.; James, L.F. Toxicity of nitro-containing astragalus to sheep and chicks. J. Range Manag. 1975, 28, 260–263. [CrossRef]

Toxins 2017, 9, 204

2. 3. 4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14.

10 of 10

Williams, M.C.; Stermitz, F.R.; Thomas, R.D. Nitro compounds in astragalus species. Phytochemistry 1975, 14, 2306–2308. [CrossRef] Benn, M.; Bai, Y.; Majak, W. Aliphatic nitro-compounds in astragalus canadensis. Phytochemistry 1995, 40, 1629–1631. [CrossRef] Majak, W.; Pass, M.A. Aliphatic nitrocompounds. In Toxicants of Plant Origin; Cheeke, P.R., Ed.; CRC Press: Boca Raton, FL, USA, 1989; pp. 143–159. Majak, W.; Bose, R.J. Chromatographic methods for the isolation of miserotoxin and detection of aliphatic nitro compounds. Phytochemistry 1974, 13, 1005–1010. [CrossRef] Matsumoto, H.; Unrau, A.M.; Hylin, J.W.; Temple, B. Spectrophotometric determination of 3-nitropropanoic acid in biological extracts. Anal. Chem. 1961, 33, 1442–1444. [CrossRef] Stermitz, F.R.; Lowry, W.T.; Norris, F.A.; Buckeridge, F.A.; William, M.C. Aliphatic nitro compounds from astragalus species. Phytochemistry 1972, 11, 1117–1124. [CrossRef] Greenwood, D.R. Determination of aliphatic nitro compounds in roots of lotus pedunculatus-the effect of maceration on leves of components. J. Sci. Food Agric. 1990, 52, 499–508. [CrossRef] Gilbert, M.; Penel, A.; Kosikowski, F.V.; Henion, J.D.; Maylin, G.A.; Lisk, D.J. Electron affinity gas chromatographic determination of beta nitropropionic acid as its pentafluorobenzyl derivative in cheeses and mold filtrates. J. Food Sci. 1977, 46, 1650–1653. [CrossRef] Muir, A.D.; Majak, W. Quantitative determination of 3-nitropropionic acid and 3-nitropropanol in plasma by hplc. Toxicol. Lett. 1984, 20, 133–136. [CrossRef] William, M.C. 3-Nitropropionic acid and 3-nitro-1-propanol in species of astragalus. Can. J. Bot. 1982, 60, 1956–1963. [CrossRef] Anderson, R.C.; Rasmussen, M.A.; Allison, M.J. Metabolism of the plant toxins nitropropionic acid and nitropropanol by ruminal microorganisms. Appl. Environ. Microbiol. 1993, 59, 3056–3061. [PubMed] Majak, W.; Clark, L.J. Metabolism of aliphatic nitro compounds in bovine rumen fluid. Can. J. Anim. Sci. 1980, 60, 319–325. [CrossRef] Burns, D.T.; Danzer, K.; Townshend, A. Use of the terms “recovery” and “apparent recovery” in analytical procedures (iupac recommendations 2002). Pure Appl. Chem. 2002, 74, 2201–2205. [CrossRef] © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

A Simple and Fast Procedure to Determine 3-Nitropropanoic Acid and 3-Nitropropanol in Freeze Dried Canadian Milkvetch (Astragalus canadensis).

Canadian milkvetch (Astragalus canadensis) is a North American plant species in the legume family and some of this plant is fatally poisonous to lives...
931KB Sizes 0 Downloads 7 Views