¹³C NMR spectroscopy for determining the acylglycerol positional composition of lampante olive oils. Chemical shift assignments and their dependence on sample concentration

Abstract

Two Lampante olive oils with free acidity expressed as oleic acid of 25.9 and 8.4 g per 100 g were used to carry out the ¹³C NMR study for the assignment of all resonances of free fatty acid fraction and of the acylglycerol fraction comprising 1(3)-mono- (1(3)-MAG), 1,3-di- (1,3-DAG), 1,2-di- (1,2-DAG) and triacylglycerols (TAG). The glycerol carbon resonances were found to be resolved according to the esterification degree of the glycerol molecule but no data were available on the length and the unsaturation pattern of the acyl chains. It was the spectral region where the C-1 carboxy carbons resonate that enabled the determination of free fatty acids and of different acylglycerol species. Moreover, within each acylglycerol class, the acyl chain distribution between the 1,3- and 2- glycerol positions was also calculated where the saturated, vaccenate, oleate and linoleate chains were detected as baseline resolved resonances in 1(3)-MAG, 1,3-DAG, 1,2-DAG and TAG. The variability of chemical shifts to sample concentration changes was also evidenced in all the frequency ranges being studied where the methylene C-2 and C-3 carbons underwent the most dramatic changes in their resonance patterns which prevented any resonance assignment.

---

Introduction

Lampante Olive Oil, it is designated “Lampante” because of its ancient use as fuel in oil burning lamps, is a Virgin Olive Oil which is not fit for human consumption without further processing. According to the International Olive Oil Council and the current European Union legislation, Lampante Olive Oil is the lowest grade oil within the group of Virgin Olive Oils. They are all obtained from the fruit of olive tree solely by mechanical or other physical means under conditions that do not lead to any chemical change.^1,2 However, considering that Lampante Olive Oil has a free acidity expressed as oleic acid of more than 3.3 g per 100 g and/or an unpleasant flavor,¹ it has to be refined to become edible.

So Lampante Olive Oils unlike Virgin olive oils, are characterized from both, high free acidity and total amount of diacylglycerols (they originate from triacylglycerols hydrolysis) which depend on raw material quality and product storage conditions. Total diacylglycerols in virgin olive oils freshly made from healthy olive fruits do not exceed 1–3%, but their content is higher (4–5%) in olive oils obtained from damaged olive fruits or from olives stored under unsuitable conditions.³

In particular, sn-1,2-diacyglycerols (they are stereospecifically numbered) are the biosynthetic intermediate of triacylglycerol formation,⁴ whereas sn-2,3- and sn-1,3-diacylglycerols arise from enzymatic hydrolysis of triacylglycerols and from isomerization processes^5,6 by transposition of an acyl group,⁷ respectively. As a result, diacylglycerol fraction can be a useful analytical parameter to determine olive oil grades and detect extra virgin olive oil adulteration with Lampante Olive Oil.^5–7

Quantitative determination of olive oil diacylglycerols has been carried out by using gas chromatography,^8,9 high performance size exclusion chromatography,^10,11 and high performance liquid chromatography.¹²

However, a new trend emerged in the research field of olive oil chemistry which encouraged the use of spectroscopic techniques such as vibrational spectroscopy,^13–15 synchronous fluorescence and total luminescence spectroscopy,¹⁶ along with nuclear magnetic resonance (NMR) spectroscopy.^17–20 Spectroscopic techniques reduce up to zeroing sample handling and improve the repeatability and reproducibility of quantitative data as compared to those obtained by classical analytical methods.

In particular, ¹³C NMR spectroscopy was applied to detect partial acylglycerols and free fatty acids in palm oil by using carbonyl carbon (C-1), C-2 and C-3 carbons of acyl chains and glycerol carbons.²¹

The characterization of partial acylglycerols of olive oil samples has been carried out by using ¹H, ¹³C and ¹P NMR spectroscopy.^22–27

This paper aims to demonstrate that ¹³C NMR succeeds in determining the full pattern of acylglycerol species of lampante olive oil samples along with the positional composition of fatty acids at the 1,3- and 2- glycerol backbone positions.

Experimental

Conventional ¹³C 1D-NMR (one dimensional) spectra were run at 25 °C on a Unity Inova Narrow Bore 500 MHz spectrometer equipped with a UNIX-based Sun Microsystems workstation (Varian NMR Instruments, Palo Alto, CA), a pulse field gradient unit and a 5 mm probe. Proton-decoupled ¹³C spectra were obtained at 125 MHz, and chemical shifts in parts per million (ppm) were measured relative to tetramethylsilane by using the residual ¹H and ¹³C peaks of the solvent as internal standards. One-dimensional ¹³CNMR spectra were obtained using a spectral window of 26 kHz with 128 K data point.

Considering that the compositional profiles of acylglycerols were calculated on the basis of carboxy carbon resonances which are characterized by long T₁ longitudinal relaxation times, intensity distortions of ¹³C resonances due to signal saturation which occurs when insufficient relaxation delay d₁ between pulses is allowed, were overcome by reducing the flip angle α of the pulses down to 45° and applying a delay d₁ of 20 s. The d₁ value was five times longer than the longest T₁ which was measured for C-1 carboxy carbon of glycerol 1(3)-oleate.²⁸

The NOE enhancement factors demonstrated that the carboxy carbons of different chains in different acylglycerols were affected by proton decoupling to the same extent. As a result, the ¹³C spectra can be acquired under full NOE conditions with the benefit of working with higher signal-to-noise ratios in shorter experiment times.²⁸

The spectra were processed by applying a resolution enhancement function both, to improve resolution and optimize signal-to-noise ratio. Time domain signal was zerofilled prior to Fourier transformation. Zeros were added to the end of the time domain that makes the effective zerofilled acquisition time AT_z longer with the consequent decrease of the spacing in the discrete frequency spectrum from 1/AT to 1/AT_z. Zerofilling to the power of 2^2 was applied thus giving a digital resolution of 0.10 Hz.

Results and discussion

Lampante olive oil samples with free acidity expressed as oleic acid of 25.9 and 8.4 g per 100 g (% w/w), were used to carry out the ¹³C NMR study for the assignment of all resonances.

Considering that lampante olive oil samples exhibited both a free fatty acid fraction and an acylglycerol fraction comprising triacylglycerols and partial acylglycerols, i.e. 1(3)- and 2- mono-, 1,2-di- and 1,3- diacylglycerols, five frequency regions were selected to determine the fatty acid composition also on the basis of their position on the glycerol backbone in different acylglycerol species.

In particular, the frequency range from 60 to 74 ppm where the carbons of glycerol backbone moiety resonate, and the ranges where the carboxy carbons C-1 (from 172.0 to 180.0 ppm) and the methylene carbons C-2 (from 33.8 to 34.3 ppm), C-3 (from 24.7 to 25.0 ppm) and C-16 (from 31.40 to 31.95 ppm) resonate, were selected.

¹³C chemical shifts of carbons of glycerol backbone moiety (60–74 ppm)

The structures of acylglycerols as a function of their esterification degree were successfully studied by means of ¹³C nuclei of glycerol backbone which resonate in the spectral region from 60 to 74 ppm. The carbons of acylglycerol backbone moiety were fully assigned²⁹and the chemical shifts are reported in Table 1.

Table 1 ¹³C NMR assignments of free fatty acids, 1(3)-mono-, 1,3-di-, 1,2-di- and triacylglycerols of lampante olive oil samples with 8.4 and 25.9% free acidity

Carbon Resonances	Acylglycerols
	Free Fatty Acids	1(3)-MAG	1,3-DAG	1,2-DAG	TAG
Glycerol moiety
CH2OCOR 1,3-pos	—	63.27	64.95	61.31	62.08
CH2OCOR 2-pos	—	—	—	72.07	68.90
CH2OH 1,3-pos	—	65.00	—	62.14	—
CH2OH 2-pos	—	70.24	68.19	—	—
Acyl chain moiety
Carboxy C-1
Saturated acid, 1,3-pos	179.60	174.21	173.86	173.70	173.24
Vaccenate 1,3-pos	—	—	173.84	—	173.23
Oleic acid, Oleate 1,3-pos	179.60	174.18	173.82	173.66	173.20
Linoleate 1,3-pos	—	—	173.81	—	173.19
Oleate 2-pos	—	—	—	173.38	172.80
Linoleate 2-pos	—	—	—	—	172.79
Methylene C-2
Saturated	—	—	—	—	33.97
Oleate 1,3-pos	—	—	—	—	—
Oleate,Linoleate 1,3-pos	—	—	—	—	33.95
Oleate, Linoleate 2-pos	—	—	—	—	34.11
Methylene C-3
Saturated	24.69	—	—	—	—
Oleic acid	24.67	—	—	—	—
Oleate, Linoleate 1,3-pos	—	—	—	—	24.77
Oleate, Linoleate 2-pos	—	—	—	—	24.81
Methylene C-16
Saturated	—	—	—	—	31.86
Oleate	—	—	—	—	31.84
Vaccenate	—	—	—	—	31.72
Linoleate	—	—	—	—	31.45

---

The symmetrical 1,3-DAG (diacylglycerol) and TAG (triacylglycerol) give 2 signals for the glycerol moiety with an intensity ratio 1 : 2, the asymmetrical 1(3)-MAG (monoacylglycerol) and 1,2-DAG(diacylglycerol) give 3 signals with intensities in a 1 : 1 : 1 ratio. The C-2 signals of glycerol resonate always downfield compared to C-1 and C-3 of glycerol backbone. No data were available on the length and unsaturation pattern of glyceride acyl chains.

The lampante olive oil samples with 25.9 and 8.4% free acidity confirmed the presence of 1(3)- mono-, 1,2- and 1,3- di- and triacylglycerols (Fig. 1).

Fig. 1 The glycerol backbone carbon region 60–74 ppm of the 500 MHz ¹³C spectrum of the lampante olive oil sample with 25.9% free acidity. The resonances of 1(3)-MAG, 1,3-DAG, 1,2-DAG and TAG were assigned.

The integration of glycerol backbone resonances enabled the calculation of full glyceride composition of lampante olive oil samples. In particular, the C-2 glycerol signals of different acylglycerol species were integrated because they were baseline resolved. The results were reported in Table 2.

Table 2 ¹³C NMR to determine fatty acid positional composition of acylglycerols of lampante olive oil samples with 8.4 and 25.9% free acidity

Carbon Resonance	Acylglycerols
	1(3)-MAG (%)		1,3-DAG (%)		1,2-DAG (%)		TAG (%)
	8.4%	25.9%	8.4%	25.9%	8.4%	25.9%	8.4%	25.9%
Glycerol backbone	1.1	4.3	10.8	17.1	3.9	6.7	84.2	71.9
Carboxy C-1
Total Composition
Saturated	—	28	13.8	14.4	6.5	7.9	15.5	17.2
Vaccenate	—	—	3.8	4.5	—	—	3.2	3.6
Oleate	100	72	68.8	73.0	74.2	81.6	69.1	71.4
Linoleate	—	—	13.8	8.1	19.4	10.5	12.2	7.7
1,3-pos Composition
Saturated	—	28.0	13.8	14.4	11.8	14.3	22.9	25.2
Vaccenate	—	—	3.8	4.5	—	—	4.7	5.3
Oleate	—	72.0	68.8	73.0	70.6	76.2	63.3	63.7
Linoleate	—	—	13.8	8.1	17.6	9.5	8.9	5.8
2-pos Composition
Oleate	—	—	—	—	78.9	88.2	81.0	88.0
Linoleate	—	—	—	—	21.4	11.8	19.0	12.0
2-pos Specificity
Oleate	—	—	—	—	47.8	48.4	38.1	39.0
Linoleate	—	—	—	—	50.0	50.0	50.4	48.9
Methylene C-16
Composition
Saturated	—	—	—	—	—	—	17.6	18.8
Vaccenate	—	—	—	—	—	—	4.6	4.3
Oleate	—	—	—	—	—	—	68.9	70.1
Linoleate	—	—	—	—	—	—	8.8	6.8

---

The ratios between 1,2- and 1,3-diacylglycerols were also calculated. Both lampante olive oil samples from Leccino and Dritta cultivars exhibited the same ratio value of 0.4. Considering that according to the Kennedy pathway, the plant tissues sinthesize TAG by acylation of DAG at the sn-3-position, the sole presence of 1,2-DAG is expected to occur in the glyceride fraction of olive oil. So, the increase of 1,3-DAG and the correspondent decrease of 1,2-DAG/1,3-DAG ratio can be explained in term of isomerization of 1,2-DAG into the most stable 1,3-DAG due to a number of interacting factors.^29,30

¹³C chemical shifts of C-1 carboxy carbons of glyceride acyl chain moieties (172–180 ppm)

The partial results obtained by the glycerol backbone resonances where no information can be achieved on the glyceride acyl chains, found their completeness in the frequency region from 172 to 180 ppm where the carboxy carbons resonate.

The spectrum of the lampante olive oil sample 25.9% free acidity showed two well defined sets of resonances. The high frequency set comprised the carboxy carbon resonances of free fatty acids which, however, were not resolved according to the chain unsaturation degree. The low frequency set spread over a shift range of about 2 ppm from 172.6 to 174.6 ppm and grouped into four clusters corresponding to 1(3)- mono-,1,3- di-, 1,2-di- and triacylglycerols from a higher to a lower frequency in that order (Fig. 2).²⁸

Fig. 2 The carboxy carbon region 172.7–174.3 ppm of the 500 MHz ¹³C spectrum of the lampante olive oil sample with 25.9% free acidity. The resonances of Saturated (S), Vaccenate (V), Oleate (O), and Linoleate (L) chains were assigned within each glyceride species comprising 1(3)-MAG, 1,3-DAG, 1,2-DAG and TAG.

The 1(3)-monoacylglycerol cluster comprised the resonances of saturated (174.21 ppm) and oleate chains (174.18 ppm), whereas the 1,3-diacylglycerol cluster exhibited the saturated (173.86 ppm), vaccenate (173.84 ppm, vaccenate and eicosenate chain overlapped in this frequency region) oleate (173.82 ppm) and linoleate (173.81 ppm) chains.

Saturated (173.70 ppm) and oleate (173.66 ppm) chains at 1,3-positions of 1,2-DAG resonated at a higher frequency from the chains at 2-position where the sole oleate chain (173.38) was detected thus confirming that saturated chain esterify only the 1,3-positions of glycerol backbone in triglycerides.

The triacylglycerol cluster showed the acyl chain resonances at 1,3-positions which were higher frequency shifted from the chains at 2-positions. A constant chemical shift difference of 0.41 ppm separated saturated (173.24 ppm), vaccenate (173.23 ppm), oleate (173.20 ppm) and linoleate (173.19 ppm) chains at 1,3-positions from the same chains detected at 2-position, in particular oleate (172.80 ppm) and linoleate (172.79 ppm) chains.These results confirmed that saturated chains were constantly absent in the acylglycerol 2-position in agreement with the most widely accepted theory of distribution of fatty acids among the glycerol positions, i.e. the 1,3-random-2-random theory. The theory predicts the characteristic placement of saturated acids in the 1,3-positions and of unsaturated acids in the glycerol 2-positions. Whenever the saturated acids in the 2-positions exceed the 1.5% level, the olive oil is likely to be adulterated by mixture with esterified olive oil.²⁸

The chemical shift pattern was confirmed according to which the introduction of an increasing double bond number in the chains made them shift at a lower frequency, namely saturated chains resonated at a higher frequency from oleate and linoleate chains, in that order. The pattern was explained in terms of the σ-inductive theory for the transmission of the effect of a double bond dipole through the C–C bonds up to the polarisation of the π electrons of the C=O group.³¹

The chemical shift difference of 0.41 ppm between the chains at 1,3- and 2-positions was likely to be due to two γ-gauche interactions which affected the carboxy carbons of a chain at the 2-glycerol positions making them be shifted at a higher frequency, as compared to one γ-gauche interaction for the chains at 1,3-glycerol positions.³¹

Based on the high resolution of carboxy carbons according to glyceride esterification degree, to chain unsaturation degree and their position on glycerol backbone, the full chain composition of the glyceride species, the chain compositions of the two different pools of fatty acids esterifying the 1,3- and 2- glycerol positions, and the chain specificity for 2-position were calculated. The data are reported in Table 2.

¹³C chemical shifts of C-16 methylene of glyceride acyl chain moieties (31.4–31.9 ppm)

The spectral region from 31.4 to 31.9 ppm detected as baseline resolved signals, the C-16 of saturated (31.86 ppm), oleate (31.84 ppm), vaccenate (31.72 ppm) and linoleate (31.45 ppm) chains. This region was the only spectral range which enabled the unambiguous detection of vaccenate chain which cannot be resolved from eicosenate chain in the carboxy carbon region.³²

The chain composition was calculated by integrating C-16 carbon resonances. They confirmed the values obtained by carboxy carbon resonances (Table 2).

¹³C chemical shifts of C-2 methylene of glyceride acyl chain moieties (33.8–34.3 ppm)

The natural line widths of methylene C-2 of triacylglycerol acyl chain moieties were found to broaden (1.1 Hz) in correspondence of shorter effective transverse relaxation times T₂* thus giving poorly resolved ¹³C NMR resonances in this frequency region.³³

Methylene C-2 carbons of triglycerides of the Lampante olive oil sample 25.9% free acidity (Fig.3, lower trace) were assigned on the basis of the chemical shift difference of 0.16 ppm which was detected in common C-18 acyl chains between the overlapping oleate and linoleate chains at 2-position (34.11 ppm) and at 1,3-positions (33.95 ppm), the saturated chains resonated at 33.97 ppm. Two major resonances which were detected at 34.08 ppm and at 34.00 ppm were only tentatively assigned to 1(3)-MAG and 1,3- DAG, respectively, also on the basis of quantitative ratios as determined in the C-1 frequency region.

Fig. 3 The methylene C-2 carbon region 33.8–34.3 ppm of the 500 MHz ¹³C spectrum of the lampante olive oil sample with 25.9% free acidity at a concentration of 200 mg/0.7 ml CDCL₃ (lower trace) and of 50 mg/0.7 ml CDCL₃ (upper trace).

¹³C chemical shifts of C-3 methylene of glyceride acyl chain moieties (24.6–24.9 ppm)

The spectral region from 24.6 to 25.0 ppm where the methylene C-3 carbons resonate, detected as baseline resolved resonances free saturated and oleic acids at 24.69 and 24.67 ppm, respectively (Fig. 4, lower trace). However, an extensive overlapping prevents any determination of the differentiated acylglycerol resonances. When measuring the ¹³C spectrum of an extra virgin olive oil sample, the oleate and linoleate chains at 2-position of triglycerides overlapped at a higher frequency (24.81 ppm, tentative assignment, Fig.4, lower trace) from the overlapping oleate and linoleate chains at 1,3-positions of triacylglycerols (24.77 ppm, tentative assignment, Fig.4, lower trace).

Fig. 4 The methylene C-3 carbon region 24.6–24.9 ppm of the 500 MHz ¹³C spectrum of the lampante olive oil sample with 25.9% free acidity at a concentration of 200 mg/0.7 ml CDCL₃ (lower trace) and of 50 mg/0.7 ml CDCL₃ (upper trace).

The integration of free fatty acids resonances enabled the calculation of free fatty acid composition which could not be measured in the carboxy carbon region where saturated and oleate chains overlapped.

¹³C chemical shift sensitivity to oil sample concentration

Chemical shift assignments of C-2 and C-3 methylenes in lampante olive oil spectra were made still more difficult at the high sample concentration used to improve spectrum sensitivity and assure accurate integration of C-1 resonances. However, the high sample concentration did not compromise the resolution of C-1 carboxy carbons²⁹ in agreement with the electronic excitation energy of carboxy carbons which is lower than the energy associated with C-2 and C-3 methylenes thus increasing the paramagnetic shielding term σ^para.³⁴

The concentration effect on ¹³C chemical shifts of glycerol and acyl chain moieties of glycerides was checked by measuring the chemical shift differences among two Lampante olive oil (25.9% free acidity) samples containing 50 and 200 mg of oil diluted with 0.7ml CDCl₃. The results are reported in Table 3 whereas Fig. 3 and 4 showed the effect of sample dilution from 200 (lower trace) down to 50 mg/0.7 ml CDCl₃ (upper trace) on the resonance patterns of C-2 and C-3 carbons, respectively.

Table 3 ¹³C NMR chemical shift dependence of free fatty acids, 1(3)-mono-, 1,3-di-, 1,2 -di- and triacylglycerols on lampante olive oil concentration

Carbon Resonances	Chemical Shift	Chemical Shift	Difference
	50 mg	200 mg	Δδ50–200
Glycerol moiety
1,2-DAG 2-pos CH2OCOR	72.093	72.026	0.067
1(3)-MAG 2-pos CH2OH	70.267	70.196	0.071
TAG 2-pos	68.864	68.845	0.019
1,3-DAG 2-pos CH2OH	68.361	68.174	0.187
1(3)-MAG 1(3)-pos CH2OH	65.112	64.969	0.143
1,3-DAG 1,3-pos CH2OCOR	65.013	64.913	0.100
1(3)-MAG 1(3)-pos CH2OCOR	63.281	63.222	0.059
TAG 1,3-pos	62.085	62.035	0.050
1,2-DAG 1(3)-pos CH2OH	62.085	62.035	0.050
1,2-DAG 1(3)-pos CH2OCOR	61.493	61.289	0.204
Acyl chain moiety
Carboxy C-1
Free Fatty Acids	179.553	179.604	−0.051
1(3)-MAG Saturated	174.345	174.198	0.147
1(3)-MAG Oleate	174.309	174.165	0.144
1,3-DAG Saturated	173.915	173.831	0.084
1,3-DAG Oleate	173.880	173.799	0.081
1,3-DAG Linoleate	173.868	173.787	0.081
1,2-DAG 1,3-pos Oleate	173.741	173.641	0.100
1,2-DAG 2-pos Oleate	173.402	173.347	0.055
TAG 1,3-pos Saturated	173.286	173.214	0.072
TAG 1,3-pos Oleate	173.252	173.18	0.072
TAG 1,3-pos Linoleate	173.241	173.170	0.071
TAG 2-pos Oleate	172.834	172.766	0.068
TAG 2-pos Linoleate	172.822	172.754	0.068
Methylene C-16
TAG Saturated	31.909	31.873	0.036
TAG Oleate	31.889	31.852	0.037
TAG Vaccenate	31.768	31.731	0.037
TAG Linoleate	31.508	31.465	0.043
Methylene C-3
Free Fatty Acid Saturated	24.723	24.689	0.034
Free Fatty Acid Oleic	24.706	24.671	0.035

---

C-1 carbon resonances of different acylglycerols were shifted to higher frequencies upon decreasing the oil concentration from 200 to 50 mg/0.7 ml CDCl₃ in agreement with those detected with standard glycerides but in disagreement with the shifts of methylene C-16 and C-1 (3) and C-2 carbons of glycerol.³⁵

The highest shift difference was detected in 1(3)-MAG where the chemical shift of saturated chain increased from 174.198 to 174.345 ppm in correspondence of 200 mg and 50, respectively, where the shift difference was 0.147 ppm.

The highest downfield shift of C-1 carbons of free fatty acids in the most concentrated oil sample was explained with the strong intermolecular hydrogen bonding which predominates at a higher concentration forming carboxylic acid dimers whereas acid monomers predominates at lower concentrations.²⁸

¹³C NMR to determine fatty acid composition, their 2-position specificity,and composition of two fatty acid pools at 1,3- and 2-glycerol positions of glyceride species

Two lampante olive oil samples based on Leccino and Dritta cultivars with 25.9 and 8.4% free acidity were studied by ¹³CNMR and the results were reported in Table 2.

Fatty acid composition of different acylglycerol species were calculated on the basis of integral values measured in correspondence of C-1 resonances and confirmed by C-16 integrals.

C-1 resonances enabled for the first time unlike the chromatographic techniques, the experimental determination of composition of two fatty acid pools esterifying the 1,3- and 2- glycerol positions in all glyceride species.

As the fatty acid pool at 1,3-positions is concerned, both in Leccino (25.9%) and Dritta (8.4%) lampante olive oil samples, the percentages of saturated chains which are esterified at the 1,3-glycerol positions, decreased from TAG (25.2, 22.9) to 1,2-DAG (14.3, 11.8) and to 1,3-DAG (14.4, 13.8) indicating that an enzymatic hydrolysis of triacylglycerols regiospecific for 1(3)- positions operated.

In correspondence, the percentages of oleate and linoleate chains increased from TAG (Oleate: 63.7, 63.3; Linoleate: 5.8, 8.9) to 1,2-DAG (Oleate: 76.2, 70.6; Linoleate: 9.5, 17.6) and slightly decreased in 1,3-DAG (Oleate: 73.0, 68.8; Linoleate: 8.1, 13.8). The patterns evidenced that the largest variation of 1,3-distribution values occurred for saturated chains which halved from TAG to 1,2-DAG, and that saturated chains were replaced by oleate and linoleate chains at 1,3-positions where linoleate almost doubled from TAG to 1,2-DAG.

The specificities of oleate and linoleate chains for 2-positions were calculated by normalizing the 2-position resonance value to both 1,3- and 2-position values for each chain and the data are reported in Table 2.

The results confirmed that saturated chains are esterified only at 1,3-positions and unsaturated chains preferentially entered glycerol 2-position.³⁶

In particular, oleate chain with a specificity for the 2-position of triglycerides of 39.0 and 38.1% moved away from a pure random distribution model (33.3%) less than linoleate chain with a 2-specificity of 48.9 and 50.4% in lampante olive oil samples with 25.9 and 8.4% free acidity, respectively.

The 2-specificity data of two lampante olive oil samples also evidenced that hydrolysis of triglycerides at the 1,3-positions where more oleate than linoleate was present, made the oleate at 1,3-positions lower and consequently the 2-specificity of oleate increased from 39.0 to 48.4 and from 38.1 to 47.8 in 1,2-diglycerides detected in the lampante olive oil samples with 25.9% and 8.4% free acidity, respectively.

Conclusions

¹³C NMR confirmed to be the only technique to be able to determine the full pattern of different acylglycerol species in Lampante olive oil samples.

The acylglycerols were determined according to the esterification degree in the shift range from 60 to 74 ppm where the carbons of glycerol backbone resonate. The quantitative ratio of 1,2- and 1,3- diacylglycerol positional isomers was calculated thus accounting for the isomerization of 1,2-diacylglycerols into the most stable form 1,3-diacylglycerols that is indicative of long storage time of the oil and of the processing conditions as high temperatures and three phase extraction with water dilution.

The compositions of the two different pools of fatty acids esterifying the 1,3- and 2-glycerol positions were also determined in the shift range from 172 to 180 ppm where the carboxy carbons resonate. The positional distribution of fatty acids between the 1,3- and 2- positions can account for the enzyme regiospecificity during the enzymatic hydrolysis of lampante olive oil samples thus opening new application fields for ¹³C NMR.

As an example, a reply should be given to the question whether the 1,3-DAG found in commercial oil samples may not be the results of artefacts but that a possibility exists that part of 1,3-DAG may form in the fruit during the maturation.

Furthermore, a reply should be given to the demand for stereospecific analysis of triacylglycerols where the fatty acid positional analysis of acylglycerols by applying ¹³C NMR directly on the oil sample without any further chemical treatment is a fundamental starting point.

References

1. IOOC, Trade Standard Applying to Olive Oil and Olive-Pomace Oil, COI/T.15/NC no. 3, 1.
2. Characteristics of Olive Oil in Commission Regulation EEC no. 2568/91, J. Eur. Commun., no L248, 5 September 1991, 1.
3. M. C. Perez-Camino, W. Modera and A. Cert, J. Agric. Food Chem., 2001, 49, 699.
4. C. Kitchcock and B. W. Nichols, Plant Lipid Biochemistry, Academic Press, New York, 1971, 176.
5. L. Cossignani, R. Luneia, P. Damiani, M. S. Simonetti, R. Riccieri and E. Tiscornia, Eur. Food Res. Technol., 2007, 224, 379.
6. O. Rosati, S. Albrizio, D. Montesano, R. Riccieri, L. Cossignani, M. Curini, M. S. Simonetti, L. Rastrelli and P. Damiani, J. Agric. Food Chem., 2005, 55, 191.
7. G. Fragaki, A. Spyros, G. Siragakis, E. Salivaras and P. Dais, J. Agric. Food Chem., 2005, 53, 2810.
8. D. Firestone, J. AOAC Int., 1994, 776, 677.
9. C. Planck and E. Lorbeer, J. Chromatogr., A, 1995, 697, 461.
10. T. Gomes, J. Am. Oil Chem. Soc., 1992, 69, 1219.
11. G. Marquez Ruiz, N. Jorge, M. Martin Polvillo and M. C. Dobarganes, J. Chromatogr., A, 1996, 749, 55.
12. J. Liu, T. Lee, E. Jr. Bobik, M. Guzman-Harty and C. Hastilow, J. Am. Oil Chem. Soc., 1993, 70, 343.
13. G. Downey, P. McIntyre and A. N. Davies, J. Agric. Food Chem., 2002, 50, 5520.
14. V. Baeten, M. Meurens, M. Morales and R. Aparicio, J. Agric. Food Chem., 1996, 44, 2225.
15. N. A. Marigheto, E. Kemsley, M. Defernez and R. H. Wilson, J. Am. Oil Chem. Soc., 1998, 75, 987.
16. K. I. Poulli, G. A. Mousdis and C. A. Georgiou, Anal. Chim. Acta, 2005, 542, 151.
17. F. D. Gunstone, Advances in lipid methodology-two In W. W. Christie (Ed.), The Oily Press, UK, 1993.
18. K. F. Wollenberg, J. Am. Oil Chem. Soc., 1990, 67, 487.
19. L. Mannina, M. Patumi, P. Fiordiponti, M. C. Emanuele and A. L. Segre, Italian J. Food Sci., 1999, 2, 139.
20. G. Vlahov, A. D. Shaw and D. B. Kell, J. Am. Oil Chem. Soc., 1999, 76, 1223.
21. Soon Ng, J. Am. Oil Chem. Soc., 2000, 77, 749.
22. F. D. Gunstone, Chem. Phys. Lipids, 1991, 58, 159.
23. P. Sacchi, F. Addeo, I. Giudicianni and L. Paolillo, La Rivista Italiana delle Sostanze Grasse, 1990, 67, 245.
24. F. D. Gunstone, Chem. Phys. Lipids, 1991, 58, 219.
25. O. Rosati and S. Albrizio, J. Agric. Food Chem., 2007, 55, 191.
26. A. Spyros and P. Dais, J. Agric. Food Chem., 2000, 48, 802.
27. P. Fronimaki, A. Spyros, S. Christophoridou and P. Dais, J. Agric. Food Chem., 2002, 50, 2207.
28. G. Vlahov, Triglycerides and Cholesterol Research, Nova Science Publishers Inc., 2006.
29. G. Vlahov, J. Am. Oil Chem. Soc., 1996, 73, 1201.
30. H. R. Barton and C. J. O'Connor, Aust. J. Chem., 1997, 50, 355.
31. O. W. Howarth, C. J. Samuel and G. Vlahov, J. Chem. Soc., Perkin Trans. 2, 1995, 2307.
32. G. Vlahov, P. K. Chepkwony and P. K. Ndalut, J. Agric. Food Chem., 2002, 50, 970.
33. G. Vlahov, C. Schiavone and N. Simone, Magn. Reson. Chem., 2001, 39, 689.
34. E. Breitmaier and W. Voelter, Carbon-13 NMR Spectroscopy, VCH, Weinheim, 1990.
35. L. Mannina, C. Luchinat, M. Patumi, M. C. Emanuela, E. Rossi and A. Segre, Magn. Reson. Chem., 2000, 38, 886.
36. F. H. Mattson and R. A. Volpenheim, J. Lipid Res., 1963, 4, 392.

---