Effects of Two Amphiphilic Diesters of L-Ascorbic Acid on the Oxidative Stability of Rabbit Meatballs

Abstract

Lipid oxidation involves a cascade of phenomena leading to serious impairments of meat quality during storage. Novel strategies for lipid protection are therefore highly desirable. Herein, two amphiphilic diesters of L-ascorbic acid with myristic (DA) and stearic (DB) acids were synthesised and added at a 0.1% (w/w) to minced rabbit meat before preparing meatballs. Then, pH, colour indexes, weight loss, fatty acid profile and primary and secondary lipid oxidation products were analysed for meatballs treated with DA (n = 16), DB (n = 16), or not treated (C, n = 16), and stored for 80 days at −10 °C. Results showed that DA and DB did not specifically prevent weight loss and lipid oxidation. Nevertheless, the addition of DA on stored rabbit meatballs seemed to prevent colour modification and reduced (p = 0.0613) TBARS levels in the treated stored meat. For these reasons, further investigations on the properties of L-ascorbyl diesters on the oxidative stability of meat will likely be performed.

1. Introduction

Rabbit is, after poultry, the most efficient converter of vegetal proteins from cellulose-rich plants into high-value animal proteins, showing a low-fat content in the meat cuts (loins, ~2 g/100 g meat; hind legs, ~3 g/100 g meat; whole carcass, ~8 g/100 g meat) and an interesting fatty acid (FA) profile. Indeed, differently from other livestock, polyunsaturated fatty acids (PUFA) in rabbit meat are generally high, amounting from 27 to 33% of total FAs [1]. Owing to its nutritional composition, rabbit meat demand is steadily expanding, and its production is expected to reach 1.8 million tonnes of fresh meat by 2025 [2]. Rabbit meat is frequently sold as whole fresh or frozen carcass, but processed rabbit meat as patties, meatballs, nuggets and other ready-to-cook have been proposed as a valuable strategy to fit the needs of specific consumer groups [3,4]. However, chemical and physical characteristics of meat are modified during the grinding process and storage due to a cascade of events led by spoilage, as well as lipid and protein oxidation, depending on the storage temperature [5,6].

In this context, the application of effective antioxidants, capable of preventing oxidative stress-related phenomena and reducing lipid oxidation processes, is highly desirable, as confirmed by the variety of natural and synthetic antioxidants studied over recent years [7,8,9,10,11,12,13,14]. Among others, L-ascorbic acid arguably represents the most used and widely studied naturally-occurring chain-breaking antioxidant [15], and such a property is due to the presence of a five-membered lactone ring bearing an ene–diol moiety. As recently reviewed [16], ascorbic acid added at 0.02–0.1% (w/w) improves meat quality and extends the shelf life of pork and beef meat. Nevertheless, ascorbic acid addition at 0.1% (w/w) in rabbit meat did not effectively hurdle the degradation processes during a refrigerated storage [12,13]. Even though the authors found that the antioxidant capacity significantly increased in fresh patties, lipid oxidation and physical changes at the end of storage were not prevented in rabbit meat mixed with ascorbic acid. This ineffective protection could be caused by the poor solubility of ascorbic acid in lipid-rich matrixes, which represents a significant hurdle for the use of L-ascorbic acid to effectively protect lipids from oxidative phenomena in food. In this regard, the functionalization of the hydroxyl groups at the C-5 and/or C-6 represents an attractive tool in order to obtain L-ascorbic acid derivatives with improved lipophilicity and unaltered antioxidant properties. The esterification of the OH functionalities at the C-5 and C-6 with long FAs enables the synthesis of non-toxic L-ascorbyl esters characterized by interesting amphiphilic properties [17,18]. For example, 6-O-palmitoyl-L-ascorbic acid is an amphipathic ester that, due to its antioxidant properties, is widely used in the food industry [19]. Based on currently available safety data, ascorbyl palmitate and ascorbyl stearate are considered safe at the reported uses and use levels [19]. However, 5-O,6-O-diesters of L-ascorbic acid have received less attention with respect to related 6-O-esters, probably because of the multi-step procedures required for their synthesis [20]. The physico-chemical properties of L-ascorbic acid derivatives are strongly influenced by the functionalization of the hydroxyl group in position 5. While the antioxidant activity of L-ascorbyl diesters is substantially retained, their lipophilicity is significantly increased. Their water solubility is very poor, whereas they are able to dissolve and aggregate in organic solvents and lipid-rich matrixes. The ─OH group in position 5 of ascorbic acid has a key role in imparting water solubility to ascorbyl amphiphiles. In the single-chain ascorbyl alkanoate-based surfactants, the hydroxyl group in position 5 is free and can be involved in intermolecular hydrogen bonds. These features potentially render L-ascorbyl dialkanoates interesting candidates for the protection of lipid-rich matrixes from antioxidant damage. In addition, the safety assessment of ascorbyl dipalmitate in cosmetics has been reported [21], and the non-toxicity of fatty acid-derived ascorbyl diesters can be extrapolated from data describing the toxicity of ascorbic acid and its related fatty acids [19].

Based on these considerations, this work represents a first attempt to evaluate whether L-ascorbyl-5-O,6-O-dialkanoates from myristic (DA) and stearic (DB) acids, specifically synthetized for the trial, could enhance the storability of rabbit meatballs, with the overarching aim of satisfying the increasing consumers’ demand for easy-to-use products. Based on previous evidences about the inefficacy of 0.1% addition of ascorbic acid to preserve rabbit meat [12,13] and considering the present work a preliminary trial, we decided to compare the fatty acid-derived ascorbyl diesters with untreated samples.

2. Materials and Methods

2.1. Synthesis of L-Ascorbyl Derivatives

2.1.1. General

NMR spectra were recorded in CDCl₃ with Varian Gemini 200, Mercury 400, Inova 400 and Bruker 400 spectrometers operating at 200 and 400 MHz (¹H), and 50 and 100 MHz (¹³C). NMR signals were referenced to non-deuterated residual solvent signal 7.26 ppm for ¹H, and 77.0 ppm, central line of CDCl₃, for ¹³C. Mass spectra (MS) were obtained by ESI. Solvents were dried using a solvent purification system (Pure-Solv^TM). Flash column chromatography was performed using silica gel (230–400 mesh). Where not specified, products were commercially available and used as received without further purification.

2.1.2. Synthesis of 2-O,3-O-dibenzyl-L-ascorbic Acid 1

Benzyl bromide (4.10 g, 24 mmol) was added to a suspension of L-ascorbic acid (1.76 g, 10 mmol) and K₂CO₃ (4.14 g, 30 mmol) in dimethyl sulfoxide/etrahydrofuran (2:1 vol) and the reaction was stirred at 50 °C for 3 h. Afterwards the mixture was filtered through a celite pad and the organic phase was extracted with EtOAc and washed with brine and H₂O. The organic layer was dried on Na₂SO₄ and filtered, and the solvent was removed under reduced pressure. The crude material was purified by flash chromatography (petroleum ether/ethyl acetate) to yield 1 as a white solid (1.53 g, 43%). ¹H-NMR (400 MHz, CDCl₃): δ (ppm) 2.09 (2H, bs, OH), 3.75 (1H, dd, J 5.6, 8.0 Hz) 3.80 (1H, dd, J 5.2, 8.0 Hz), 3.89–3.92 (1H, m), 4.53 (1H, d, J 2.8 Hz), 4.69 (2H, ap d, ls 2.8 Hz), 4.70 (2H, s), 7.18–7.28 (2H, m), 7.34–7.39 (8H, m). MS (ESI positive) m/z (%): 379.1 [M + Na]⁺, (100) [20,22].

2.1.3. Synthesis of (S)-1-((R)-3,4-bis(benzyloxy)-5-oxo-2,5-dihydrofuran-2-yl)ethane-1,2-diyl dimyristate 2a

4-dimethylaminopyridine (4-DMAP, 0.73 g, 6 mmol, 3.0 eq.) and N,N’-dicyclohexylcarbodiimide (DCC, 1.24 g, 6 mmol, 3.0 eq.) were added to a stirred solution of 1 (0.71 g, 2 mmol, 1.0 eq.) in MeCN (20 mL) under inert atmosphere at room temperature. Then, myristoyl chloride (1.14 g, 4.6 mmol, 2.3 eq.) was slowly added and the mixture was stirred for 12 h. Afterwards, the solvent was removed under reduced pressure and the crude material purified by flash chromatography (petroleum ether/EtOAc 12:1) to yield 2a as a colourless oil (1.21 g, 78%). ¹H-NMR (400 MHz, CDCl₃): δ (ppm) 0.88 (ap t, J 6.5 Hz, 6H), 1.19–1.32 (m, 40H), 1.50–1.62 (m, 4H), 2.17–2.23 (m, 2H), 2.25–2.29 (m, 2H), 4.23 (dd, J 7.2, 11.6 Hz, 1H, CH_aH_bO), 4.32 (dd, J 5.4, 11.6 Hz, 1H, CH_aH_bO), 4.80 (bs, 1H, OCHCH), 5.11 (ap s, 2H, CH₂Ph), 5.14 (d, J 11.5 Hz, 1H, CH_aH_bPh), 5.17 (d, J 11.5 Hz, 1H, CH_aH_bPh), 5.35–5.40 (m, 1H, CHCH₂O), 7.21–7.26 (m, 2H), 7.33–7.39 (m, 8H). ¹³C-NMR (100 MHz, CDCl₃): δ (ppm) 14.1, 22.7, 24.7, 25.0, 29.1, 29.2, 29.3, 29.4, 29.5, 29.6, 29.65, 29.7, 32.0, 33.9, 34.0, 61.9, 67.3, 73.7, 73.73, 77.2, 121.4, 128.0, 128.66, 128.67, 128.7, 128.8, 128.9, 135.0, 135.9, 155.1, 168.6, 172.2, 173.0. MS (ESI positive) m/z (%): 799.3 [M + Na]⁺, (100).

2.1.4. Synthesis of (S)-1-((R)-3,4-bis(benzyloxy)-5-oxo-2,5-dihydrofuran-2-yl)ethane-1,2-diyl distearate 2b

4-DMAP (0.73 g, 6 mmol, 3.0 eq.) and DCC (1.24 g, 6 mmol, 3.0 eq.) were added to a stirred solution of 1 (0.71 g, 2 mmol, 1.0 eq.) in MeCN (20 mL) under inert atmosphere at room temperature. Then, stearoyl chloride (1.49 g, 4.6 mmol, 2.3 eq.) was slowly added and the mixture was stirred for 12 h. Afterwards the solvent was removed under reduced pressure and the crude material purified by flash chromatography (petroleum ether/EtOAc 12:1) to yield 2b as a colourless oil (1.58 g, 89%). ¹H-NMR (400 MHz, CDCl₃): δ (ppm) 0.86–0.90 (m, 6H), 1.22–1.32 (m, 56H), 1.50–1.64 (m, 4H), 2.15–2.23 (m, 2H), 2.25–2.29 (m, 2H), 4.23 (dd, J 7.8, 10.9 Hz, 1H, CH_aH_bO), 4.33 (dd, J 5.6, 10.9 Hz, 1H, CH_aH_bO), 4.80 (d, J 1.9 Hz, 1H, OCHOH), 5.11 (ap s, 2H, CH₂Ph), 5.14 (d, J 12.1 Hz, 1H, CH_aH_bPh), 5.18 (d, J 12.1 Hz, 1H, CH_aH_bPh), 5.35–5.40 (m, 1H, CHCH₂O), 7.22–7.25 (m, 2H), 7.33–7.40 (m, 8H). ¹³C-NMR (100 MHz, CDCl₃): δ (ppm) 14.1, 22.7, 24.8, 29.0, 29.1, 29.25, 29.28, 29.4, 29.5, 29.66, 29.7, 31.9, 33.9, 34.0, 61.9, 67.3, 73.68, 73.7, 77.2, 121.4, 128.0, 128.66, 128.67, 128.7, 128.8, 128.9, 135.1, 135.9, 155.1, 168.6, 172.2, 173.1. MS (ESI positive) m/z (%): 911.7 [M + Na]⁺, (100).

2.1.5. Synthesis of (S)-1-((R)-3,4-dihydroxy-5-oxo-2,5-dihydrofuran-2-yl)ethane-1,2-diyl dimyristate DA

To a solution of ascorbyl 5-O-,6-O-dimyristate-2,3-dibenzyl ethers 2a (1.0 g, 1.29 mmol) in ethyl acetate (20 mL), Pd/C (10%) was added. Then, a balloon filled with H₂ was attached to the flask and the reaction was stirred for 2 h. The reaction progress was monitored by TLC. After the complete consumption of the starting product, the mixture was filtered through celite and the solvent was removed under reduced pressure. Crystallization from diethyl ether/petroleum ether gave ascorbyl 5-O-,6-O-dimyristate DA as a white solid (0.75 g, 97%). ¹H-NMR (400 MHz, CDCl₃): δ (ppm) 0.88 (t, 6H, J 6.4 Hz), 1.19–1.35 (m, 40H), 1.54–1.66 (m, 4H), 2.27–2.36 (m, 4H), 4.30 (dd, 1H, J 7.2, 11.4 Hz, CH_aH_bO), 4.42 (dd, 1H, J 4.6, 11.4 Hz, CH_aH_bO), 4.90 (d, 1H, J 2.8 Hz, CHCOH), 5.40–5.45 (m, 1H, CHCHOH). ¹³C-NMR (100 MHz, CDCl₃): δ (ppm) 14.1, 22.7, 24.9, 29.0, 29.1, 29.3, 29.4, 29.5, 29.6, 29.7, 31.9, 34.1, 62.0, 68.1, 74.5, 119.4, 149.0, 170.3, 173.0, 173.7. MS (ESI negative) m/z (%): 595.2 [M − H]^─, (100).

2.1.6. Synthesis of (S)-1-((R)-3,4-dihydroxy-5-oxo-2,5-dihydrofuran-2-yl)ethane-1,2-diyl distearate DB

To a solution of ascorbyl 5-O-,6-O-distearate-2,3-dibenzyl ethers 2b (1 g, 1.13 mmol) in ethyl acetate (20 mL), Pd/C (10%) was added. Then, a balloon filled with H₂ was attached to the flask and the reaction was stirred for 2 h. The reaction progress was monitored by TLC. After the complete consumption of the starting product, the mixture was filtered through celite and the solvent was removed under reduced pressure. Crystallization from diethyl ether/petroleum ether gave ascorbyl 5-O-,6-O-distearate DB as a white solid (0.74 g, 93%). ¹H-NMR (200 MHz, CDCl3): δ (ppm) 0.86 (t, 6H, J 6.4 Hz), 1.15–1.38 (m, 56H), 1.49–1.71 (m, 4H), 2.24–2.39 (m, 4H), 4.30 (dd, 1H, J 7.2, 11.4 Hz, CH_aH_bO), 4.42 (dd, 1H, J 4.6, 11.4 Hz, CH_aH_bO), 4.89 (d, 1H, J 3.2 Hz CHCOH), 5.38–5.64 (m, 1H, CHCHOH). ¹³C-NMR (50 MHz, CDCl3): δ (ppm) 14.1, 22.7, 24.7, 24.8, 29.0, 29.1, 29.3, 29.4, 29.5, 29.7, 31.9, 33.8, 34.0, 62.1, 68.0, 74.4, 119.4, 149.1, 170.3, 172.9, 173.6. MS (ESI negative) m/z (%): 707.6 [M − H]^─, (100).

2.2. Determination of the in vitro Chain-Breaking Antioxidant Activity of Synthesised Compounds

The chain-breaking antioxidant activity was measured according to a literature reported procedure [23]. A volume of 20 μL of 10 mM ethanolic solution of compounds DA or DB were added to 1.98 mL of 200 μM 2,2-difenil-1-picrylhydrazyl (DPPH) in ethanol. The reaction was followed by spectrophotometric analysis measuring the absorbance at 515 nm. In order to determine the stoichiometry of the reaction (n, number of radical trapped), the assay was also performed by using an excess of DPPH. Thus, 20 μL of 5 mM ethanolic solution of compounds DA or DB were added to 1.98 mL of 200 μM DPPH in ethanol and rapidly mixed. The progress of the reaction was monitored by spectrophotometric analysis measuring the absorbance at 515 nm every 20 s for 10 min.

2.3. Meatball Preparation and Storage Trial

Overall, 9 carcasses of male rabbits were purchased from a private farm (Italy). Rabbits were reared under the same conditions and fed ad libitum the same commercial diet until they were 94-days-old, when animals were slaughtered in a specialised slaughterhouse. Only carcass meat (CM) was here utilised while the hind legs, the Longissimus thoracis et lumborum muscle, bone, and fat (inguinal, abdominal, and interscapular depots) were dissected. Overall, approximately 1400 g of all 9 CMs were manually cut, then minced by a 0.5-hole diameter common cooking grinder (Westmark, Elspe, Germany). The meat was divided into 3 groups; one group (control, C) was utilized without any addition, and the other two were added with 0.1% (w/w) of myristoyl L-ascorbic acid derivative (DA) or stearoyl L-ascorbic acid derivative (DB). The meat was mixed for 30 s, then meatballs were prepared by inserting around 30 g of meat inside a Meatball Maker (Snips, LO, Italy) equipped with 16 shapes. After this, every meatball (n = 16 for each group) was weighed. Samples from each group were analysed immediately (0 day) and during a frozen storage experiment (−10 °C; sampling points: 20, 40, and 80 days). At each sampling point, four meatballs for each group were thawed overnight at 4 °C, dried with a common paper towel and weighed before being analysed. The unconventional storage temperature and the length of the storage were chosen based on existing literature on rabbit meat quality, which reported increased lipid oxidation after 40 days at −12 °C while slowing down bacterial proliferation [24].

2.4. Physical and Chemical Characterization, Statistics

The pH value was monitored using a pH meter (Mettler-Toledo, Schwerzenbach, Switzerland) in three different points of the meat balls; similarly, the colour parameter values were measured with a Konica Minolta colourimeter (Chiyoda-ku, Tokyo, Japan) and expressed as lightness (L*), redness (a*) and yellowness (b*) indexes [25].

The total lipid content of the samples was extracted [26], gravimetrically quantified, then the FAs were esterified to methyl esters (FAME) [27]. The FA composition was determined by gas chromatography using a Varian GC 430 gas chromatograph (GC; Varian Inc., Palo Alto, CA, USA), equipped with a flame ionization detector (FID) and a Supelco Omegawax™ 320 capillary column (30 m × 0.32 mm i.d., 0.25-μm film and polyethylene glycol-bonded phase; Supelco, Bellefonte, PA, USA). The GC conditions were: injection 1:20 split ratio at 220 °C; the oven temperature was programmed to rise from 130 to 230 °C in 40 min; the detector was set at 300 °C; helium was the carrier gas that flowed at 1.5 mL/min. Chromatograms were recorded with the Galaxie Chromatography Data System 1.9.302.952 computing integrator software (Varian Inc., Palo Alto, CA, USA). FAs were identified by comparing the FA methyl ester (FAME) retention time with the Supelco 37 component FAME mix (Supelco, Bellefonte, PA, USA).

Primary and secondary lipid oxidation products were determined as conjugated dienes (primary) [28] and 2-thiobarbithuric acid reactive substances (TBARS) [29].

A GLM PROC of SAS [30] was applied to the data using Treatment (T: C, DA, DB), Storage (S: 0, 20, 40, 80 days) and their interaction (T × S) as fixed effects. A Tukey–Kramer post hoc test was utilised to define the differences among the experimental groups, considered significant at p ≤ 0.05. When the interaction between the main factors was not significant, only the mean of T (i.e., C, DA, and DB) and S (i.e., 0, 20, 40 and 80 days) were reported in the tables.

3. Results and Discussion

3.1. Synthesis and in vitro Antioxidant Activity of L-Ascorbyl Diesters

L-ascorbyl diesters DA and DB were prepared following the synthetic strategy reported in the Figure 1.

As the physico-chemical properties of amphiphilic fatty acid esters depend on the length of the hydrocarbon chain [17,18,31], we focused our attention on the synthesis of myristoyl and stearoyl L-ascorbic acid derivatives DA and DB. Typically, a three-step strategy is employed to convert L-ascorbic acid into the corresponding 2,3-dibenzyl ether 1. A reaction sequence involving the protection of OH on C-6 and C-5 as acetonide, followed by treatment of enol hydroxyls with benzyl bromide and a final cleavage of the acetonide is the most commonly used route. However, an alternative recently developed procedure enables straightforward access to 1 upon reaction of L-ascorbic acid with benzyl bromide and K₂CO₃ (Figure 1) [20,22]. Due to the high reactivity of enol hydroxyls under basic conditions (OH at C-2, pK_a = 11.6; OH at C-3, pK_a = 4.2), their preliminary protection, generally as benzyl ethers, is required. Treatment of 1 with myristoyl or stearoyl chloride in the presence of 4-DMAP and DCC enables the synthesis of diesters 2a and 2b, protected as dibenzyl ethers. Finally, a Pd/C catalysed hydrogenation reaction, employed to cleave the protecting groups, afforded fatty acid-derived L-ascorbyl diesters DA and DB bearing the free ene–diol moiety, required in order to maintain the antioxidant properties of L-ascorbic acid. The identity and purity of the synthesised compounds were confirmed by NMR and MS analysis. Regarding ¹H NMR spectroscopic data, as expected, the esterification of hydroxyl groups at C-5 and C-6 shifts downfield the resonance frequency of the CH(5) and CH₂(6) signals. Indeed, while the chemical shift of the CH(5) of the diol 1 is 3.89–3.92 ppm, the CH(5) signal of 2a is shifted to 5.35–5.40 ppm. Similarly, the diastereotopic CH₂(6) protons of 2a (4.23; 4.32 ppm) are significantly deshielded with respect to the CH₂(6) protons of 1 (3.75; 3.80 ppm). Related spectroscopic data were observed for the compound 2b.

The cleavage of the protecting groups of the ene–diol moiety of 2a,b is easily confirmed by the disappearance of signals of aromatic and benzylic protons in ¹H NMR spectra of DA and DB.

Diagnostic ¹H NMR signals of the above-mentioned compounds are summarised in

Table 1. Diagnostic ¹H NMR data of L-ascorbyl diesters DA and DB and of intermediates 1 and 2a,b.

Compound	CH (4) a	CH (5) a	CHaHb (6) a
	4.53 (d, J = 2.8 Hz)	3.89–3.92 (m)	3.75 (dd, J = 5.6, 8.0 Hz); 3.80 (1H, dd, J = 5.2, 8.0 Hz)
	2a: 4.80 (bs)	2a: 5.35–5.40 (m)	2a: 4.23 (dd, J = 7.2, 11.6 Hz); 4.32 (dd, J = 5.4, 11.6 Hz)
	2b: 4.80 (d, J = 1.9 Hz)	2b: 5.35–5.40 (m)	2b: 4.23 (dd, J = 7.8, 10.9 Hz); 4.33 (dd, J = 5.6, 10.9 Hz)
	A: 4.90 (d, J = 2.8 Hz)	A: 5.40–5.45 (m)	A: 4.30 (dd, J = 7.2, 11.4 Hz); 4.42 (dd, J = 4.6, 11.4 Hz)
	B: 4.89 (d, J = 3.2 Hz)	B: 5.38–5.64 (m)	B: 4.30 (dd, J = 7.2, 11.4 Hz); 4.42 (dd, J = 4.6, 11.4 Hz)

^a The chemical shift (δ, ppm) along with the multiplicity and the coupling constant (Hz) is reported for each proton.

. Complete ¹H NMR and ¹³C NMR spectroscopic characterisation of all compounds is reported in the Section 2.1.

Ascorbic acid contains a five-membered lactone ring bearing an ene–diol moiety capable of reacting with free radicals, thus behaving as a scavenger and preventing the oxidation of other molecules [32,33,34]. In order to preliminarily evaluate their antioxidant capacity, L-ascorbyl diesters DA and DB were tested through the DPPH assay. Complete decolouration of the ethanolic DPPH radical solution, observed within 5 s from the addition of DA or DB, indicated that the radical quenching rapidly occurred. These results highlighted the remarkable chain-breaking activity of the synthesised L-ascorbyl diesters DA and DB, which proved to be comparable to that of L-ascorbic acid. As expected for L-ascorbyl derivatives bearing the free ene–diol moiety, the stoichiometry of the reaction, referring to the number of radicals trapped by the antioxidant scavenger within 10 min (commonly indicated as n), was determined to be n = 2.

3.2. Physical Characterization, Fatty Acid Profile and Lipid Oxidation of Meatballs

To the authors’ knowledge, this is the first study determining the effects of ascorbic acid diesters on the oxidative stability of rabbit meatballs, even if the obtained results lead to more questions than proposed solutions. No significant effects of the two diesters on meat physical characteristics were highlighted (

Table 2. The pH, colour parameter values and weight loss (%) of rabbit meatballs with (DA, DB) or without (C) antioxidants sampled at 0, 20, 40 and 80 days of storage (−10 °C). Data are expressed as mean value ± standard error.

	Treatment, T			Storage, S				T	S	T × S
	C	DA	DB	0	20	40	80
pH	6.07 ± 0.01	6.07 ± 0.01	6.04 ± 0.01	6.00 c ± 0.01	6.03 b ± 0.01	6.09 a ± 0.02	6.12 a ± 0.01	0.056	0.006	0.487
L*	48.99 ± 0.44	47.94 ± 0.50	49.03 ± 0.50	50.99 a ± 0.35	46.79 b ± 0.61	47.74 b ± 0.60	47.76 b ± 0.61	0.280	<0.0001	0.085
a*	15.25 ± 0.30	14.91 ± 0.34	14.94 ± 0.34	17.24 a ± 0.24	16.01 ab ± 0.40	15.71 b ± 0.40	11.17 c ± 0.41	0.999	<0.0001	0.607
b*	5.83 ± 0.17	5.84 ± 0.20	5.98 ± 0.21	6.07 a ± 0.14	5.04 b ± 0.24	5.85 ab ± 0.24	6.57 a ± 0.24	0.881	0.0002	0.028
Weight loss	2.15 ± 0.13	2.46 ± 0.15	2.59 ± 0.15	-	1.83 b ± 0.14	2.59 a ± 0.14	2.76 a ± 0.14	0.076	0.0003	0.020

^a,b,c: Within criterion, means with different letters in the same row are significantly different (p < 0.05). Number of replicates per group: n = 4.

). Only a trend was observed for pH (p = 0.056) and weight loss (p = 0.076), with the DB group having the lowest pH and the highest weight loss value. Contrariwise, pH and weight loss steadily increased throughout the storage and reached a maximum value on day 80. In line with other studies, the addition of 0.1% of ascorbic acid did not play a significant role in preserving the meat from discolouration [12], which in turn was highly affected by the storage (p < 0.0001). Specifically, the gradual decrease in L* and a* values and the increase in b* have been previously reported for rabbit meat stored over 7 days at 4 °C [12] or 40 days at −12.5 °C [24].

Recently, some authors have suggested that microbial spoilage, lipid oxidation, myoglobin autoxidation and protein oxidation jointly induce meat colour changes, which slowe down at temperatures below zero [6]. This fact explains why changes in physical characteristics were evident only after 40 days of frozen storage (−12.5 °C) [24], in line with the present findings. Furthermore, the same researchers found that the protein oxidation of rabbit meat started from the very early beginning of the storage at −12.5 °C, inducing the loss of sulfhydryl groups, which was more pronounced after 20 days when even the carbonyl content started to sharply increase [6]. Such modification of the protein structure could cause the loss of water holding capacity [12,35,36] here observed through calculating the weight loss. The highest liquid loss was obtained in the DA group after 40 days (

Table 3. Yellowness index (b*) and weight loss (%) obtained considering the significant interaction (T × S). Data are expressed as mean value of each group ± standard error.

	C				DA				DB
	0	20	40	80	0	20	40	80	0	20	40	80
b*	6.07 ± 0.24 ab	4.81 ± 0.38 b	5.62 ± 0.40 ab	6.83 ± 0.39 a	6.07 ± 0.24 ab	5.73 ± 0.43 ab	4.97 ± 0.44 ab	6.59 ± 0.42 ab	6.07 ± 0.24 ab	4.58 ± 0.38 b	6.98 ± 0.44 a	6.29 ± 0.43 ab
Weight loss	-	1.85 ± 0.22 b	1.78 ± 0.22 b	2.81 ± 0.22 ab	-	1.60 ± 0.25 b	3.09 ± 0.22 a	2.64 ± 0.21 ab	-	2.02 ± 0.21 ab	2.91 ± 0.22 ab	2.84 ± 0.25 ab

^a,b: Different letters in the same row indicate significant differences for T × S interaction (p < 0.05).

), thus suggesting that the diester, at the concentration tested, did not prevent protein oxidation.

The FA profile of the muscle of rabbits is affected by a variety of factors such as feed, lipid content, and FA profile [37], level of exercise (hence farming condition and available space per head) [38] during rearing, and the meat cut. As shown in

Table 4. Total lipids (g/100 g meat) content and fatty acid profile (g/100 g fatty acid methyl esters) of the meatballs with (DA, DB) or without (C) antioxidants sampled at 0, 20, 40 and 80 days of storage (−10 °C). Data are expressed as mean value ± standard error.

	Treatment, T			Storage, S				T	S	T × S
	C	DA	DB	0	20	40	80
Total lipids	3.96 ± 0.36	3.39 ± 0.40	4.41 ± 0.41	2.69 ± 0.35	4.21 ± 0.48	4.52 ± 0.47	4.27 ± 0.49	0.218	0.183	0.913
C14:0	1.96 b ± 0.17	2.82 a ± 0.19	1.92 b ± 0.18	1.98 ± 0.18	2.66 ± 0.22	2.17 ± 0.22	2.11 ± 0.22	0.008	0.203	0.115
C15:0	0.52 ± 0.01	0.51 ± 0.01	0.51 ± 0.01	0.52 a ± 0.01	0.49 b ± 0.01	0.49 b ± 0.01	0.54 a ± 0.01	0.655	<0.0001	0.065
C16:0	24.65 ± 0.22	24.37 ± 0.25	24.68 ± 0.23	24.12 ± 0.23	24.43 ± 0.29	24.80 ± 0.28	24.91 ± 0.28	0.615	0.144	0.239
C16:1n-9	0.38 ± 0.01	0.38 ± 0.01	0.38 ± 0.01	0.39 a ± 0.01	0.37 b ± 0.01	0.37 b ± 0.01	0.39 a ± 0.01	0.948	0.004	0.420
C16:1n-7	0.91 ± 0.03	0.96 ± 0.03	0.96 ± 0.03	1.01 ± 0.03	0.89 ± 0.04	0.92 ± 0.03	0.94 ± 0.04	0.479	0.086	0.563
C17:0	0.83 ± 0.01	0.81 ± 0.01	0.82 ± 0.01	0.84 ab ± 0.01	0.79 b ± 0.01	0.80 b ± 0.01	0.85 a ± 0.01	0.441	0.001	0.499
C18:0	7.35 b ± 0.15	7.18 b ± 0.16	7.76 a ± 0.16	7.36 ± 0.16	7.25 ± 0.19	7.39 ± 0.19	7.70 ± 0.19	0.049	0.388	0.654
C18:1n-9	24.78 ± 0.21	24.57 ± 0.23	24.83 ± 0.24	24.13 b ± 0.22	25.20 a ± 0.27	25.14 a ± 0.27	24.44 ab ± 0.27	0.713	0.009	0.980
C18:1n-7	1.06 ± 0.04	1.05 ± 0.04	0.96 ± 0.04	1.10 ± 0.04	1.02 ± 0.04	0.93 ± 0.05	1.05 ± 0.05	0.178	0.068	0.148
C18:2n-6	29.74 ± 0.20	29.42 ± 0.22	29.48 ± 0.23	29.38 ± 0.22	29.95 ± 0.22	29.80 ± 0.27	29.04 ± 0.26	0.536	0.081	0.365
C18:3n-3	2.57 ± 0.04	2.50 ± 0.04	2.56 ± 0.04	2.50 ± 0.05	2.63 ± 0.05	2.57 ± 0.05	2.49 ± 0.05	0.519	0.279	0.711
C20:3n-6	0.30 ± 0.02	0.31 ± 0.02	0.29 ± 0.02	0.39 a ± 0.02	0.23 b ± 0.03	0.26 b ± 0.03	0.33 ab ± 0.03	0.793	0.001	0.927
C20:4n-6	2.72 ± 0.22	2.81 ± 0.24	2.62 ± 0.22	3.69 a ± 0.23	2.07 b ± 0.28	2.30 b ± 0.29	2.81 ab ± 0.28	0.850	0.000	0.928
C22:4n-6	0.43 ± 0.03	0.45 ± 0.03	0.43 ± 0.03	0.55 a ± 0.03	0.37 b ± 0.04	0.38 b ± 0.04	0.45 ab ± 0.04	0.806	0.001	0.916
C22:5n-3	0.27 ± 0.02	0.27 ± 0.02	0.25 ± 0.02	0.35 a ± 0.01	0.20 b ± 0.01	0.23 b ± 0.01	0.27 ab ± 0.02	0.801	0.000	0.951
SFA	35.93 ± 0.28	36.32 ± 0.31	36.30 ± 0.32	35.47 ± 0.20	36.21 ± 0.36	36.25 ± 0.37	36.80 ± 0.36	0.573	0.056	0.037
MUFA	27.45 ± 0.21	27.31 ± 0.23	27.47 ± 0.23	26.98 ± 0.22	27.81 ± 0.27	27.67 ± 0.27	27.17 ± 0.27	0.861	0.079	0.976
n-6 PUFA	33.65 ± 0.25	33.47 ± 0.27	33.28 ± 0.28	34.54 a ± 0.27	33.03 b ± 0.33	33.17 b ± 0.32	33.12 b ± 0.32	0.619	0.002	0.104
n-3 PUFA	2.94 ± 0.03	2.87 ± 0.04	2.92 ± 0.03	2.97 ± 0.03	2.91 ± 0.04	2.89 ± 0.04	2.87 ± 0.04	0.543	0.201	0.252

^a,b Within criterion, means with different letters in the same row are significantly different (p < 0.05). Number of replicates per group: n = 4.

, L-ascorbyl-5-O,6-O-dialkanoates from myristic and stearic acids increased the C14:0 and C18:0 content in DA and DB groups, respectively (p < 0.05). After 20 days of frozen storage, several FAs declined as in the case of C15:0, C16:1n-9, C17:0, C20:3n-6, C20:4n-6, C22:4n-6, C22:5n-3 and the sum of n-6 PUFA (p < 0.05), in agreement with some authors [12]. Such a decrease may be reasonably due to the higher content of n-6 than n-3 PUFA, which undergo concentration-dependent enzymatic degradation at a greater extent.

Changes in FA relative proportions anticipated data regarding lipid oxidation as highlighted by conjugated diene (Figure 2A) level, which was significantly augmented at the 20th day of storage regardless the antioxidant addition. Analogously, according to recent findings [5], the presence of primary products of lipid oxidation indicated that this reaction also initiates early during storage, which was the only variable affecting the conjugated dienes (p < 0.001). However, both hydroperoxides and conjugated dienes were not stable and quickly decomposed to other molecules, whose content increased during storage (p < 0.001) and was only marginally slowed down by the antioxidant addition, as highlighted by the TBARS assay. In this regard, the formation of lipid peroxidation products were lower (p = 0.0613) in the DA group (0.51 mg MDAeq./kg) than in the C (0.56 mg MDAeq./kg) and DB (0.82 mg MDAeq./kg). According to the TBARS assay, the maximum extent of lipid oxidation was found at day 80 (Figure 2B), when the levels of lipid peroxidation-derived products reached 1.01, 1.08, and 1.55 mg MDAeq./kg in DA, C, and DB, respectively (p > 0.05). Although the TBARS assay is a common method to assess lipid oxidation, no limit values for MDA concentrations in meat and meat products are officially accepted; this might be caused by the fact that the relationship between MDA level and perceived rancidity depends on the type of product (for example, raw, mechanically-deboned meat and manufactured frankfurters [39]), the type of meat, i.e., bovine, fish, or chicken, and on the method of analysis [40]. The results obtained at the end of the present storage trial showed that rabbit meatballs did not exceed the suggested threshold for fresh chicken sausages, namely 3 mg/kg sample [41]. Nevertheless, the MDA eq. content doubled the value obtained by other studies in samples stored at −12 °C for 45 days [24]. This could be caused by the different processing; indeed, while the other authors [24] stored and analysed the whole cut of meat (from Longissimus thoracis et lumborum), in the present trial rabbit carcass meat was minced. Mincing causes the disruption of cellular integrity, increasing the surface of exposure of proteins and lipids to molecular oxygen, which is incorporated during the mixing step [42]. The disruption of cellular integrity also leads to the release of enzymes, including those involved in oxidative phenomena. These structural damages could be summed to those induced by the freeze/thaw process, such as the fast release of catalytic iron [5], which can directly enter the oxidative processes of proteins [43] and lipids [5]. Besides, ascorbic acid can also react with Fe³⁺, providing ascorbyl radicals and Fe²⁺, which catalyses a range of peroxide-mediated oxidation processes (Fenton reaction) [44,45]. Hence, this complex series of events, also including the multifaceted nature of ascorbic acid and ascorbic acid derivatives, could help us understand the reason why the ascorbic diesters tested here for the first time seemed to be scarcely effective in preventing lipid oxidation despite their increased lipophilicity.

4. Conclusions

The effect of two L-ascorbyl diesters on the oxidative stability of rabbit meatballs has been investigated. According to this preliminary evidence, the two diesters of ascorbic acid did not show a striking effect in preventing meat discoloration, weight loss, and lipid oxidation. However, the tendency of the meat treated with the myristic-derived diester (DA) to have a reduced TBARS content might suggest new studies for the development of nature-inspired semisynthetic antioxidants for food preservation. Researches focusing on the investigation of the properties of fatty-acid derived L-ascorbyl diesters of different chain length, or on the effects of such antioxidants on different matrixes (as fish) and at different concentrations are also encouraged.