Munida specimens were frozen on board inmediately after caught by the scientific research vessel “Humboldt” belonging to Instituto de Mar del Perú (IMARPE) in June 2019 at the area 18°6’ 20.401” S & 70°48’ 14.4” W, in front of Caleta Vila Vila (Tacna), 3.5 nm off the southern coast of Peru. Samples were placed in thermal boxes to maintain cooling until arrival to the Bioactive Compounds Laboratory of Instituto Tecnológico de la Producción (ITP). The size distribution of the specimens ranged from 10 to 16 mm cephalothorax length, mean of 12.9 ± 1.2 mm, mode of 13 mm; female specimens represented 54.4% of total samples with a mode of 13 mm, while males registered 14 mm.

50 kg of “munida” were placed in a cold air dryer (CV-20AN, ASAHI, Japan) at 21 °C for 27 h, then crushed in an analytical mill (A11 basic, IKA, USA) and sieved to obtain a homogeneous material between 0.50 and 0.85 mm particle size. The munida meal (MM) was packed in vacuum-sealed bags and kept at -18 °C until analysis.

Moisture, fat, ash and protein contents were determined by duplicate determinations of fresh munida and MM according to FAO (1986) methodologies.

Four different lipid extraction procedures were performed using a mixture of hexane + isopropyl alcohol 60:40 (v/v) (He-I), acetone (Ac), absolute ethanol (Et) and Supercritical CO₂ + ethanol as cosolvent (SC-CO₂-Et).

He-I: The technique described by Sachindra et al. (2006) was followed. A mixture 60:40 hexane ACS (Fermont, Mexico) with HPLC grade isopropyl alcohol (Fisher Scientific, Spain) was used to dissolve 8 g MM sample in 50 mL tubes with 40 mL, vortexed for 2 min, sonicated at 25 °C for 10 min and centrifuged (Centrifuge 5804 R, Eppendorf, Brazil) for 40 min at 3200 g at 4 °C after 5 min resting time. The extract was filtered through Whatman Nº 42 filter paper and the residue was subjected to further extraction following the same procedure.

Ac: 8 g MM were placed in 50 mL tubes with 40 mL of Ac (ACS, Merck, 99.5% purity), vortexed for 2 min, sonicated at 25 °C for 10 min and centrifuged for 40 min at 3200 g at 4 °C after 5 min resting time. The extract was filtered using Whatman Nº 42 filter paper and the residue was treated with two additional extractions.

Et: According to the Dalei and Sahoo (2015) methodology 10 g MM were thoroughly homogenized with 100 mL Et (ACS Sharlau, Spain) for 1 hour using a magnetic stirrer. The extract was filtered through Whatman Nº 42 filter paper. Solid recovery was performed on the residue by 4 extractions until the filtrate was colorless.

SC-CO ₂ -Et: A multi-solvent extractor Model 2802.000 (Top Industrie, France) equipped with a CO₂ pump (HPFlow Pump 50 - 100), co-solvent pump (90-2491 REV L, SSI), chiller (PCPR 13.02-NED, National Lab), reactor (ø 163 x 353 mm) and a stainless-steel separator (ø 78 x 278 mm) to receive the lipid were used (Barriga-Sánchez et al., 2022). Pressure was manually controlled by a back pressure regulator.

The extraction of 35 g MM by SC-CO₂-Et was performed following the reference parameters reported by Sánchez-Camargo et al. (2012), 200 bar, temperature 50 °C and a solvent ratio 85/15 (CO₂/ethanol) for 2 h. Evaporation of the solvent was carried out using a rotary evaporator (Laborota 4003, Heidolph) at 40 °C and the residue was stored in Ultrapure nitrogen atmosphere (Linde Peru) at -19 °C, until further analysis.

Analyses were conducted in three replicates.

The MM lipid yield was obtained by calculations according to Equation 1.

The methodology of Núñez-Gastélum et al. (2016) was applied in all samples. 1 g lipid was dissolved in 1 mL ACS petroleum ether (Tedia, USA) vortexing for 1 min. 5 µL of each sample were placed on a silica gel 60 F₂₅₄ plate (Merck, Germany) pre-dried at 110 °C for 2 h. The plate was placed in a chamber saturated with 50 mL acetone: hexane (25:75, v/v) as the mobile phase.

Bands were visualized under a 254 nm UV TLC lamp (Merck) and identified by comparing the Retention Factor (Rf) value with the standard ASTX by applying Equation 2. Tests were conducted in triplicate.

Total carotenoids expressed as ASTX were determined in all samples according to the methodology of Sánchez-Camargo et al. (2011). A standard solution of ASTX (98.6%, Dr Ehrenstorfer) was prepared by diluting 1 to 5 µg/mL of ASTX standard in hexane. 50 mg lipid sample were diluted to 10 mL in hexane. The absorbance value of each solution and the sample were measured at 472 nm (highest absorbance observed) using a UV-200 Spectrophotometer (Shimadzu, Japan) with hexane as the calibration blank. Carotenoids were expressed as µg ASTX/g lipid and µg ASTX/g MM.

Fatty acids were determined as described by Prevot and Mordret (1976). A gas chromatograph with a FID detector (Autosystem XL, Perkin Elmer, USA) equipped with a Supelcowax 10 column (Merck, Germany) (30 m × 0.25 mm id; film thickness: 0.25 μm) was used. Peak areas were calculated using Total Chrom Navigator software (Version: 6.2.0.0.0:B27, 2001, USA), and each fatty acid percentage was calculated by comparing the individual peak area with the fatty acid total area. The fatty acid peaks were identified by comparison with the retention times of the standard F.A.M.E. Mix C4-C24 (Supelco, Sigma-Aldrich Inc, USA).

The fatty acid profile of the MM lipid was used to determine its functional quality by means of the AI and TI according to equations 3 and 4, respectively (Ulbricht and Southgate, 1991). The H:H was evaluated in accordance with equation 5 as defined by (Santos-Silva et al., 2002).

Where: C12:0 (lauric acid); C14:0 (myristic acid); C16:0 (palmitic acid); C18:0 (stearic acid); C18:1ω -9 (oleic acid); C18:2ω -6 (linoleic acid); C18:3ω -3 (linolenic acid); C20:4 ω -6 (arachidonic acid); C20:5ω -3 (eicosapentaenoic acid); C22:5ω -3 (docosapentaenoic acid); C22:6 -3 (docosahexaenoic acid); MUFA (monounsaturated Fatty Acids).

Minitab version 17 was used for analysis of variance and Tuckey’s comparison test for lipid yield data, ASTX contents and fatty acid profile obtained for each extraction procedure.

Table 1 shows similar fat content in munida fresh samples to the data obtained by Albrecht-Ruiz and Cueto (2006) and P. planipes fresh samples (Fonseca-Rodríguez and Chavarría-Solera, 2017) and meal data (Civera et al., 2000) showing the higher yield oil in munida as an advantage.

Fat is one of the most variable components in marine animals and is influenced by biotic and abiotic factors (age, catching area, time and depth of capture). Bascur et al. (2017) investigated the effect of seasonal variations and food availability to which P. monodon ovigerous females were exposed during their reproductive period (February to December) and during winter. The results indicated that these organisms adjusted their biochemical processes to ensure their survival and that of their embryos.

Table 2 shows that ASTX (Rf 0.53) was identified in all P. monodon oil extracts. The values obtained were 0.63 and 0.81, which would evidence the esterified form of this molecule (monoesters and diesters) as typical forms which are characteristic in crustaceans (Hornero-Méndez, 2019). The Rf values obtained also agree with the results obtained by Dalei and Sahoo (2015) in crustacean shell residues. These authors also refer to the fact that Rf 0.99 evidences the presence of β-carotene, a molecule that would also be present in the munida lipids.

The highest efficiency in lipid extraction from the MM sample was obtained using Et as solvent. Likewise, Xie et al. (2018) reported the advantage of Et in SC-CO₂ extraction compared to 3 solvents per step working with krill meal. This result could be explained by the SC-CO₂ increased polarity and its ability to dissociate protein-phospholipid complexes (Hardardottir and Kinsella, 1988). However, Ali-Nehari et al. (2012) reported higher efficiency in krill meal lipid extraction using hexane (16.2% lipids) compared to SC-CO₂-Et (12.2%), although Xie et al. (2017) reported higher yields with Et (16.33%) in comparison to hexane yields (12.18%), explaining that alcoholic solvents are more efficient for krill meal lipid extraction.

Carotenoid values expressed as ASTX and extracted under the conditions established in the present study ranged from 2998.01 to 4238.65 µg /g lipid. The highest values were obtained using Ac and SC-CO₂-Et (Table 3). The higher efficiency of Ac compared to Et for extracting ASTX from crustacean lipids has been demonstrated by Dalei and Sahoo (2015) and Xie et al. (2018). The lower polarity of Ac facilitates its penetration through the hydrophobic mass surrounding the pigment and favors its miscibility (Dalei and Sahoo, 2015).

All ASTX values obtained in MM by means of the proposed treatments exceeded those reported in residues of Farfantepenaeus paulensis, a species belonging to the genus Penaeus, (1074 µg ASTX/g lipid) (Sánchez-Camargo et al., 2011) and those obtained in krill oil extracted with SC-CO₂-Et (86.2 µg ASTX/g lipid) and hexane (103.2 µg ASTX/g lipid) by Ali-Nehari et al. (2012). These results suggest not only the affinities of the solvent and extraction conditions but that munida would represent a source of higher contents of ASTX compounds compared to similar species.

Typically, the choice of solvent is made according to the polarity of the target compound. Routray et al. (2019) reported improved ASTX extraction efficiency when using hexane combined with other solvents, although in the present work the use of He-I mixture extracted the lowest ASTX values indicating that isopropyl alcohol did not improve ASTX extraction efficiency (Table 3).

The results of SC-CO₂-Et extraction are in agreement with Routray et al. (2019) research data on the significant improvement in this extraction technology to recover ASTX using Et as cosolvent. Also, Sánchez-Camargo et al. (2011) reported 15% Et as cosolvent to substantially improve ASTX extraction, and highlighted the advantage of the solubilization of polar compounds such as phospholipids and glycolipids.

According to Capelli (2018) the recommended daily intake of ASTX (4 mg) can be provided by one gram of munida lipid obtained by Ac and SC-CO₂-Et extractions (4238.65 and 4086.71 µg/g lipid, respectively); nevertheless, considering Ac toxicity, the use of SC-CO₂-Et is considered the best extraction option for safety concerns among the extraction methods evaluated.

Table 4 shows the fatty acid profile of munida lipids obtained by the different lipid extraction methods. Among the saturated fatty acids (SFA) C16:0 represented the highest percentage in all the extracts while C18:1 ω-9 was the most abundant among monounsaturated fatty acids (MUFA) and its quantity was not affected by the extraction method applied.

A predominant presence of long-chain PUFA acids was observed, and ranged from 40.79 to 41.51% with high values of EPA and DHA. No significant differences (p > 0.05) regarding the extraction methods were observed. The values reported in the present study exceeded those obtained by Xie et al. (2017) using Ac as solvent in three krill species as well as by Ali-Nehari et al. (2012) in krill oil. The later one found higher efficiency by using SC-CO₂-Et compared to hexane. These high long-chain PUFA contents are considered essential for membrane fluidity and inflammatory mediator functionality and show potential benefits in neuronal development and cardiovascular health (Janssen and Kiliaan, 2014).

The high C16:0 and C20:5 ω-3 values in munida lipid samples were similar to those reported by Ali-Nehari et al. (2012) in oily extracts of krill, in residues of Farfantepenaeus paulensis (Sánchez-Camargo et al., 2011) and in Litopenaeus vannamei; while the C22:6 ω-3 contents were higher in the samples of the present study.

The EPA+DHA contents in P. monodon showed no significant differences, and the values obtained by the extraction techniques ranged from 31.15 to 31.85%. ω-6 fatty acids were not detectable in some cases, as in the case of C20:3 ω-6 and C20:4 ω-6. C18:2 ω-6 was the omega-6 fatty acid with the highest content. The results obtained in the present study show similarity to those reported by Ali-Nehari et al. (2012) and Xie et al. (2017) for krill oil.

No differences were observed among the AI values of the MM lipids obtained with all the solvents used in the present study (Table 5), although these values were lower than those calculated from studies carried out by Xie et al. (2017) and Sánchez-Camargo et al. (2012) for Euphausia superba and Penaeus paulensis respectively. This would be advantageous in the case of munida, considering that Turan et al. (2007) refer to AI values and also thrombogenicity indexes (TI) close to zero, which are considered favorable for preventing coronary heart disease.

The AI and TI values obtained in the present study are slightly higher than those obtained by Lopes et al. (2014) for grape pomace oil (AI= 0.18-0.32, TI= 0.06-0.17). Studies by Pinto et al. (2020) on Endopleura uchi oil reported AI values similar to those obtained in our work, although their TI values were higher (AI=0.44, TI=1.32) than ours.

Regarding the use of solvents, the highest H:H ratio was observed in the extracted munida lipids using the He-I mixture (1.81). This value exceeded those obtained by Xie et al. (2017) and Xie et al. (2018), although it was lower than that obtained by Sánchez-Camargo et al. (2012) for Penaeus paulensis when using SC-CO₂-Et. Low H:H values are considered unfavorable as they may induce an increase in cholesterolemia (Santos-Silva et al., 2002); while high values like 2.66 in uxi (Endopleura uchi) are recommended by Pinto et al. (2020).

Our results suggest that the SC-CO₂-Et lipid extraction method applied in Peruvian marine species of commercial importance contributes to quantifying carotenoid pigments and essential fatty acids (EPA and DHA) among other bioactive comounds. Argopecten purpuratus (scallops), Romaleon setosum, Cancer porteri, Platymera gaudichaudii, Paralomis longipes (crabs) Loxechimus albus (sea urchin) and many other fish species represent a promising dietary source.