https://orcid.org/0000-0001-8172-610X
Figures
Abstract
Genetically improved farmed tilapia (GIFT) is a strain of Nile tilapia (Oreochromis niloticus) developed for improved production and commercial parameters. Skin color, one of the characteristics distinguishing tilapia varieties, is an important phenotypic trait associated with qualitative and productive performance. This study aimed to assess fillet and skin characteristics in GIFT Nile tilapia with black and red skin phenotypes. For this, 24 GIFT Nile tilapia from the same spawning stock were divided into two groups based on skin color, namely a black variety (dark skin) and a red variety (reddish skin). There were no significant differences in biometric parameters between varieties. Fish of the red variety had higher 2 h post-slaughter pH and muscle antioxidant capacity and lower yellow-blue axis value (b*), luminosity (L*), thawing loss, cooking loss, and conjugated diene content (P < 0.05). The skin of black tilapia exhibited higher force in the progressive tear test tear, and higher strength, tensile strength, deformation, and elongation in the traction and stretching test than that of red tilapia. Black tilapia skin had higher collagen and hydroxyproline contents. The skin of the red variety exhibited higher expression levels of alpha 1, 2, and 3 type I collagen genes, lower expression levels of the vimentin gene, and higher antioxidant capacity. The results suggest that skin color phenotype may be linked to important metabolic pathways influencing fish fillet and skin traits. These findings can support future research aimed at identifying optimal varieties of Nile tilapia for specific purposes and optimizing the utilization of filleting waste.
Citation: Campos VJ, Gasparino E, Lima Júnior JWR, Khatlab AdS, Bastos MS, Santana TP, et al. (2025) Characterization of fillets and skins from two varieties of genetically improved farmed Nile tilapia (Oreochromis niloticus). PLoS ONE 20(2): e0314928. https://doi.org/10.1371/journal.pone.0314928
Editor: Amel Mohamed El Asely, Benha University, EGYPT
Received: June 6, 2024; Accepted: November 18, 2024; Published: February 21, 2025
Copyright: © 2025 Campos et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Fish farming represents the primary economic activity of many communities around the world, serving as an accessible source of protein with high biological value for human nutrition. According to data from the United Nations’ Food and Agriculture Organization [1], the per capita consumption of fish increased significantly between 1970 and 2019, rising from 11.4 to 20.5 kg. Nile tilapia (Oreochromis niloticus) is a particularly prominent species of cultivated fish due to its tolerance to high stocking densities and water salinity, good adaptability to rearing environments, and excellent disease resistance [2]. The global production of Nile tilapia reached approximately 6.5 million tonnes in 2022, with 860,355 tonnes produced in Brazil alone [3].
Recent trends suggest that the consumption of tilapia meat is increasing, attributed to its high protein quality, affordable price, and easy preparation [4]. Nuryanto et al. [5] showed that Nile tilapia meat (boneless and headless fillet) had an energy value of 1 kcal/g, with protein being the most abundant macronutrient (18.46 g/100 g). The concentration of amino acids found by the authors was 21.56%, and the major amino acids were glutamic acid (3.45%), aspartic acid (2.16%), arginine (1.88%), lysine (1.84%), and leucine (1.69%). Additionally, tilapia meat contained 30.39% fat, comprising both saturated and unsaturated fatty acids, including oleic (8.13%), palmitic (7.87%), linoleic (3.67%), and stearic (4.30%) acids. Furthermore, tilapia meat contained high levels of calcium (74.58 mg/100 g) [5].
Fillets are the primary form used by the fishing industry to market tilapia. As this cut accounts for only 34.2% of the fish weight, filleting generates large amounts of waste, including carcass, viscera, and skin [6]. Designing utilization strategies for tilapia waste is important for minimizing environmental impacts, generating revenue, and adding value to the industrial process. Studies proposed various uses for tilapia waste, such as utilization in human food products [7,8], biofuel production [9], and biomedical applications [10,11]. Some investigations demonstrated the efficacy of Nile tilapia skin as a low-cost wound dressing for burn treatment [12,13].
Collagen fibers derived from aquatic organisms have attracted the attention of scientists from different fields. The reasons for this interest are many, including the low risk of disease transmission to humans, lack of religious restrictions, high cost-effectiveness, low molecular weight, good biocompatibility, and easy absorption by the human body [14]. Fish skin is an abundant source of gelatin and collagen, which can be hydrolyzed into bioactive peptides with antihypertensive, antioxidant, antimicrobial, neuroprotective, antihyperglycemic, and anti-aging properties. These properties are determined by the composition and amino acid sequence of bioactive peptides [15]. According to Reátegui-Pinedo et al. [16], external factors such as nutrition status and pathogen infection, as well as genetic factors, can change the configuration of amino acids in Nile tilapia skin.
Developed three decades ago for improved production and commercial performance, the genetically improved farmed tilapia (GIFT) strain is currently the most cultivated lineage of Nile tilapia worldwide [17]. In GIFT tilapia, skin color is an important phenotypic trait that can be used to distinguish different varieties [18] and is associated with various adaptive, reproductive, and productive characteristics [19]. In view of these considerations, this study aimed to biometrically characterize GIFT Nile tilapia with black and red skin phenotypes and determine the effect of skin color on fillet and skin characteristics. The findings are expected to contribute to the identification of optimal varieties of Nile tilapia for specific purposes and the appropriate utilization of filleting waste.
Materials and methods
The experiment received approval from the Animal Research Ethics Committee (CEUA protocol No. 2942240423) at the State University of Maringá, Brazil.
Fish specimens and body measurements
A total of 60 fingerlings of GIFT Nile tilapia (O. niloticus) were used in the study. Fish were divided into two experimental groups based on their skin color, which are referred to in this paper as the black variety (dark skin, n = 30) and the red variety (reddish skin, n = 30). The experiment lasted 150 days.
The fish were reared at the Experimental Fish Farming Station (CODAPAR, State University of Maringá, Maringá, Paraná, Brazil) in excavated tanks at a stocking density of 3 fish/m3. Water quality was maintained under standard conditions. Each tank was equipped with a conventional aeration system operated with on/off cycles of 30 min, resulting in 12 h of operation and 12 h of rest over a 24 h period. All fish received the same commercial feed, whose nutritional composition varied according to the rearing phase (Table 1). Feeding was carried out 8 times a day in the fingerling phase, 6 times a day in the pre-juvenile phase, 4 times a day in the juvenile and grower phases, and 2 times a day in the finisher phase.
At the end of the 150-day experimental period, 12 black fish and 12 red fish with similar body weights were captured from the pond and stunned by heat shock in one part ice to one part water [20]. Subsequently, the fish were euthanized by severing the spinal cord [21]. The fishes were weighed and measured for total length, standard length, head length, back height, and back width.
After biometric assessment, the fish were eviscerated, and the skin on both sides was removed using specialized pliers. The procedure consisted of peeling the skin from the dorsal region near the head toward the caudal region. The samples were identified and weighed. Then, fragments of skin (n = 12 fish per variety) were excised from the upper-right part of the body, washed with ice-cold saline, placed in cryogenic tubes, and frozen in liquid nitrogen. Subsequently, skin samples were stored in a freezer at −80°C until analysis of gene expression and antioxidant capacity. For histological quantification of collagen types, skin fragments (without scales) measuring about 5 cm were collected from the upper-right side of the body from 7 individuals per treatment. These samples were washed thoroughly with cold sterile saline solution and then placed in microtubes containing 10% buffered formaldehyde solution for fixation. After samples were collected for gene expression and histological analyses, the remaining portions of skin (right and left sides) were placed in plastic bags and stored in a −20°C freezer. These samples were used for analysis of skin resistance parameters, hydroxyproline content, and collagen content.
The skinless fish were decapitated and their heads were weighed. Fillets were separated from the right and left sides of the body (without skin), weighed, and subjected to analysis of meat quality indicators. For analysis of antioxidant capacity, oxidative metabolites, and gene expression, muscle samples (n = 12 fish per variety) were collected from the upper-right side of the body, placed in cryogenic tubes, and frozen in liquid nitrogen. Subsequently, muscle samples were stored in a freezer at −80°C until analysis. Fish carcasses were weighed after filleting.
Indicators of fillet quality
The meat quality indicators assessed in this study were pH, instrumental color, water activity (aw), thawing loss (ThL), cooking loss (CL), and texture profile. For these analyses, each fish was considered an experimental unit (n = 12 per fish variety).
pH, color, and aw analyses were performed on refrigerated fillets collected from the left side of the body at 2 h post-slaughter. For ThL and CL determination, fillets were collected from the right side of the body and stored in a freezer at −20°C until analysis.
Fillet pH was measured 2 h after slaughter [21] and 12 h after thawing [22] using a digital pH meter (HI 99163, Hanna® Instruments Brasil Ltda). The electrode was inserted into three different points of the fillet (dorsal, ventral, and central regions).
Color parameters were measured after slaughter at six different points on the ventral side of fillets by using a CR-400 colorimeter (Konica Minolta Sensing Americas, Inc.) at a 90° angle and room temperature [23]. Luminosity (L*), red-green axis (a*), and yellow-blue axis (b*) are expressed in the CIELAB color space [23]. Chroma and hue angle were calculated according to Gomes et al. [24].
Water activity was determined on samples at room temperature using an AquaLab 4 TE analyzer (Meter Group, USA).
For ThL assessment, frozen fillets (time 0) were weighed and thawed in a digital biochemical oxygen demand incubator (model SSBOD 342L) at 4°C. After 12 h, the fillets were weighed, and ThL (%) was calculated as follows: ThL = 100 × (Weighttime0 –Weighttime12h)/(Weighttime0) [22].
Thawed fillets were placed on trays and left to stand on the bench for 30 min until they reached room temperature. The fillets were then weighed to obtain their initial weight and roasted on a grill (Cadence) lined with aluminum foil until their internal temperature reached 75°C [23]. No seasonings were added during cooking. The internal temperature of fillets was measured at three different points using a digital culinary thermometer (TP101, CITEX, Arapongas, Paraná, Brazil). After the desired temperature was reached, the fillets were removed from the grill, placed on trays on the bench for 5 min at room temperature, and weighed to determine the final weight. CL (%) was calculated as follows: CL = 100 × (Initial weight − Final weight)/Initial weight [25].
The texture profile of fillets was evaluated by determining hardness, elasticity, cohesiveness, adhesiveness, and chewiness according to Nascimento et al. [26].
Antioxidant capacity and oxidative metabolites in muscle tissues
Antioxidant capacity was investigated by measuring 2,2-diphenyl-1-picrylhydrazyl radical (DPPH•) scavenging activity and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonate) radical (ABTS•+) reducing activity. Extracts were prepared by mixing 100 mg of muscle sample (n = 7 fish per variety) with 1000 μL of methanol. Samples were ground until completely dissociated by using a Fisher Scientific 150 portable homogenizer and centrifuged at 10,000 × g for 10 min at 4°C. The supernatants were used as crude extracts.
DPPH• scavenging activity was analyzed as described by Brand-Wiliams et al. [27], with some modifications. Briefly, 400 μL of crude extract was added to a microtube containing 1600 μL of 0.06 mM DPPH• solution in methanol. The reactions were maintained in the dark for 30 min. After this period, samples were read using an Evolution 300 UV-Vis spectrophotometer (Thermo Fisher Scientific) at 515 nm. DPPH• scavenging activity of muscle tissues was determined using the following equation: DPPH• activity (%) = [1 − (Sample absorbance/DPPH• absorbance)] × 100.
The ABTS•+ assay was performed as described by Rufino et al. [28], with some modifications. Initially, a solution was prepared by mixing 5 mL of 7 mM ABTS•+ with 88 μL of 140 mM potassium persulfate. The reaction system was maintained in the dark for 16 h at room temperature. After this period, 1000 μL of the solution was diluted in ethanol to an absorbance of 0.70 ± 0.05 at 734 nm. Then, 40 μL of crude extract was added to a microtube containing 1960 μL of diluted ABTS•+ solution. After 6 min, the absorbance was read at 734 nm using an Evolution 300 UV-Vis spectrophotometer (Thermo Fisher Scientific). ABTS•+ activity (%) was determined using the following equation: ABTS•+ activity (%) = [(ABTS•+ absorbance − Sample absorbance)/ABTS•+ absorbance] × 100.
The determination of primary (conjugated dienes) and secondary (thiobarbituric acid reactive substances) lipid peroxidation and muscle protein oxidation (carbonylated protein) was performed as described by Khatlab et al. [29] and Pontes et al. [30].
Skin resistance: Progressive tear, tensile, and elongation tests
The skin (n = 12 fish per variety) collected from the left side of the body was used for resistance determination according to the methods described by Matiucci et al. [31].
Antioxidant capacity (DPPH• scavenging activity) of the skin
For assessment of DPPH• scavenging activity in skin tissues, crude extracts were prepared as described for muscle tissues. Then, 100 μL of skin extract was added to a microtube containing 900 μL of 0.06 mM DPPH• solution in methanol. The reactions were maintained in the dark for 30 min. After this period, the samples were read using an Evolution 300 UV-Vis spectrophotometer (Thermo Fisher ScientificTM) at 515 nm. The DPPH• scavenging activity of the skin was calculated as described for muscle tissues.
Hydroxyproline and collagen extraction and quantification
Thawed skins (n = 7 samples per variety) were washed with distilled water. Any residual scales and flesh were removed from the skins using a laboratory spoon. Samples were washed with neutral detergent and peracetic acid for 20 min. The water was changed twice. The washing procedure was performed gently to avoid damaging the samples. After cleaning, the skins were weighed and subjected to a pretreatment with acetic acid at 4.5% of the sample weight. Then, the skins were heated for 3.5 h. The liquid fraction containing hydroxyproline and collagen was filtered and left to cool. The extract was dried in a forced air oven for approximately 12 h, ground, and subjected to hydroxyproline and collagen quantification.
Hydroxyproline content was determined based on the oxidation of free hydroxyproline to pyrrole-2-carboxylic acid by chloramine-T, as described by Bergman and Loxley [32]. The absorbance was measured at 558 nm using a spectrophotometer. Hydroxyproline content was calculated from a standard curve. Then, collagen contents were calculated from hydroxyproline contents using a conversion factor of 7.25 (Collagen content = Hydroxyproline content/100 × 7.25).
Determination of the area occupied by type I and type III collagen in the skin of black and red Nile tilapia
After fixation, the samples (n = 7 per variety) were dehydrated in an increasing series of ethanol, cleared in xylene, and embedded in paraffin for histological sectioning. Semi-serial sections measuring 3 μm in the longitudinal direction were obtained using a manual microtome (Leica Biosystems, RM2125 RTS).
Histological sections were stained with Picrosirius Red to differentiate collagen types (Fig 1). For determination of the area occupied by type I and III collagen fibers, 10 images were captured per fish using a Nikon optical microscope (Eclipse®, Shinjuku, Japan) coupled to a high-resolution camera (Nikon®, DS-Fi1c, Shinjuku, Japan) with a 20× objective and optical polarization. Images were processed using NIS-Elements software (version 4.0, Prague, Czech Republic). Type I and III collagen fibers were quantified using Image-Pro Plus version 4.0 (Media Cybernetics). The results are expressed as a percentage, representing the relative area of each collagen type to the dermis area analyzed.
Histological sections of the skin of (A) black and (B) red Nile tilapia (Oreochromis niloticus) stained with Picrosirius Red and observed under a polarized light microscope (20× magnification). Type I collagen fibers are shown in red-orange (white arrows), and type III collagen fibers are shown in green (yellow arrows). Bar = 100 μm.
Genes expression in skin and muscle tissues
The total RNA of skin and muscle samples (n = 7 per variety) was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. The concentration of total RNA was measured using a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific) at a wavelength of 260 nm. The purity and integrity of total RNA in each sample were determined using the 260/280 nm and 260/230 nm absorbance ratios, which yielded values ranging from 1.9 to 2.1. Total RNA was treated with the DNase I amplification grade kit (Invitrogen, Carlsbad CA, USA) to avoid contamination with genomic DNA, according to the manufacturer’s instructions. Complementary DNA (cDNA) was synthesized using the SuperScript III First-Strand Synthesis Super Mix kit (Invitrogen Corporation), following the manufacturer’s instructions. cDNA samples were stored at −20°C until use. Real-time quantitative polymerase chain reactions (RT-qPCR) were conducted in duplicate using PowerUpTM SYBRTM Green Master Mix (Applied Biosystems, Vilnius, Lithuania) and a StepOne Real-Time PCR System version 2.3 (Applied Biosystems). The thermal cycling parameters for all genes followed the recommendations of the PowerUp SYBR® Green Master Mix protocol (Applied Biosystems, Vilnius, Lithuania).
The genes alpha 1 type I collagen (col1a1, GenBank accession number AB603656.1), alpha 2 type I collagen (col1a2, GenBank accession number AB603657.1), alpha 3 type I collagen (col1a3, GenBank accession number AB603658.1), vimentin (vim, GenBank accession number XM_003438066.5), DEAH box helicase 9 (dhx9, GenBank accession number XM_025900168.1), ryanodine receptor 1 (ryr1, GenBank accession number XM_005474818.1), and ryanodine receptor 3 (ryr3, GenBank accession number XM_005453445.1) (Table 2) specific to O. niloticus were designed based on the sequences available on the NCBI website (www.ncbi.nlm.nih.gov), using the primer design software provided by IDT (www.idtdna.com). The beta-actin gene specific to O. niloticus was used as endogenous control (GenBank accession number EU887951.1) (Table 2).
The 2−ΔCt method [33], where ΔCt = (Cttargetgene − Ctendogenousgene), was used for the relative quantification of gene expression. The results are presented as arbitrary units (AU). The scheme of the experimental design is illustrated in Fig 2.
Statistical analysis
The Shapiro–Wilk test was performed to assess the normality of the data. The data were analyzed using one-way ANOVA. The following statistical model was used: Yij = μ + αi + eij, where Yij is the dependent variable, μ is the general mean, αi is the effect of treatment (i = black or red variety), and eij is the residual error. Means with significant differences were compared using Student’s t-test (P < 0.05) (SAS, 2002, version 9.00, SAS Inst. Inc., Cary, NC) [34].
Results
No significant differences in biometric parameters were observed between tilapia varieties (Table 3).
Table 4 shows the impact of skin color phenotype on fillet quality parameters. Red tilapia had a higher pH value 2 h post-slaughter (P = 0.0362), lower L* value (P = 0.0060), lower b* value (P = 0.0468), lower hue angle (P = 0.0003), lower ThL (P = 0.0131), lower CL (P = 0.0140), lower hardness (P = 0.0019), higher cohesiveness (P = 0.0353), lower adhesiveness, and lower chewability (P = 0.0077) than black tilapia. There were no significant differences in fillet elasticity between varieties (P = 0.6587).
Regarding antioxidant capacity and oxidative metabolites, it was found that black tilapia had a higher content of conjugated dienes (P = 0.0420), lower DPPH• scavenging activity (P = 0.0341), and lower ABTS•+ scavenging activity (P = 0.0424) (Table 5). There were no significant differences in muscle TBARS content (P = 0.5751) or carbonyl protein content (P = 0.8391) between black and red tilapia (Table 5).
No significant differences were observed in the expression of ryr1 (P = 0.7270) or ryr3 (P = 0.2915) in the muscle of black and red tilapia (Fig 3).
Each fish was considered an experimental unit (n = 7 individuals per variety). Results are expressed as arbitrary units (AU) and presented as mean ± standard error.
The results of the skin resistance test are shown in Table 6. In the tensile and elongation test, skin color phenotype influenced tensile force (P = 0.0079), tensile strength (P = 0.0365), deformation (P = 0.0077), and elongation (P = 0.0004). The skin of red tilapia exhibited lower tensile force, tensile strength, deformation, and elongation than that of black tilapia. Furthermore, in the progressive tear test, tearing force (P = 0.0237) was influenced by tilapia variety, with red tilapia exhibiting lower tearing force than black tilapia. No significant effect was observed on tear strength (P = 0.01039).
Fig 4 shows the hydroxyproline and collagen contents of the skin of black and red tilapia. Skin color significantly influenced hydroxyproline (P = 0.0003) and collagen (P = 0.0003) contents. The highest contents were observed in the skin of black tilapia.
Each fish was considered an experimental unit (n = 7 individuals per variety). Results are presented as mean ± standard error. a,b Different letters indicate significant differences by Student’s t-test (P < 0.05).
Fig 5 shows the area of skin occupied by type I and type III collagen. There was no significant effect of fish variety on the area occupied by type I (P = 0.2801) or type III (P = 0.2763) collagen.
Each fish was considered an experimental unit (n = 7 individuals per variety). Results are presented as mean ± standard error.
Red tilapia exhibited a higher expression of col1a1 (P = 0.02000), col1a2 (P = 0.0032), and col1a3 (P = 0.0141) in skin tissues than black tilapia (Fig 6). Regarding genes regulating collagen expression, vim expression was highest in black tilapia (P = 0.0067). The expression of the dhx9 gene, which encodes ATP-dependent RNA helicase A, did not differ between tilapia varieties (P = 0.9259) (Fig 6).
Each fish was considered an experimental unit (n = 7 individuals per variety). Results are expressed as arbitrary units (AU) and are presented as mean ± standard error. a,b Different letters indicate significant differences by Student’s t-test (P < 0.05).
Fig 7 shows the results of the antioxidant capacity of the skin. There was a significant difference (P = 0.0171) between varieties, with red tilapia exhibiting a higher antioxidant capacity than black tilapia.
Each fish was considered an experimental unit (n = 7 individuals per variety). Results are presented as mean ± standard error. a,b Different letters indicate significant differences by Student’s t-test (P < 0.05).
Discussion
This study aimed to biometrically characterize Nile tilapia of the GIFT strain exhibiting black or red skin color phenotypes and determine the effect of skin color on fillet and skin characteristics. No significant differences were observed between tilapia varieties in any of the biometric characteristics evaluated. The fish used in this study belong to the same lineage and therefore share the same genetic makeup. Although phenotypic variation in skin color was observed, biometric standardization eliminates the need for separating fish into different batches within the production system.
Fillet is the primary form in which tilapia meat is marketed, and it is regarded as a source of high-quality protein. Although no differences in fillet weight were observed between black and red tilapia, some fillet quality parameters were influenced by skin color phenotype. Tilapia with black skin exhibited a lower fillet pH value, higher L* value, higher b* value, and higher hue angle. Soares and Gonçalves [35] reported a similar pH value to that observed here. According to the authors, the pH is low compared with that observed in other studies, which suggested that the ideal pH of fish fillets is close to 7 [36]. The post-mortem decrease in pH is associated with the transformation of glycogen into lactic acid. Lactic accumulation leads to a decrease in fillet pH. Thus, fillet pH seems to be related to diet [37] and the genetic lineage of fish.
Post-mortem pH is an important indicator of meat quality and is directly related to the reduction in the muscle’s ability to retain water. A reduction in pH leads to the denaturation of muscle proteins and loss of solubility, affecting the texture and color of meat [38]. This effect was evidenced by the increase in ThL and CL in black tilapia fillets. Goes et al. [39] suggested a relationship between the expression of ryr1 and ryr3 genes during pre-slaughter stress and the water holding capacity of Nile tilapia fillets. The authors explained that the reduction in water retention in fillets from stressed fish may be due to a reduction in the expression of ryr1. The protein controls the liberation of Ca2+ from intracellular stores, leading to excess production of Ca2+ in the cytosol. These effects result in hypermetabolism and muscle contraction, thereby reducing post-mortem pH and causing protein denaturation and water loss in meat [40]. However, in our study, we did not observe a significant effect of fish variety on ryr1 or ryr3 expression. This finding suggests that possibly in conditions where fish are not in a state of stress there are other unknown metabolic mechanisms involved in the reduction of post-mortem pH.
Meat color is an important criterion influencing consumer choice. Color is a qualitative attribute that indicates the physicochemical state of muscle fibers and is directly related to the amount of liquid and myoglobin present in muscles [41,42]. Here, it was observed that fillets of black and red tilapia differed in color parameters. The L* and b* of black tilapia fillets indicated a tendency toward reflection of white, resulting in a lighter color. Additionally, the hue angle, which refers to the perceived color of fillets, was very close to 0°, indicating a reddish color. The color components of fillets are dominated by the pigment myoglobin [43]. In black tilapia, fillets are mainly composed of light-colored muscle, also containing dark-colored muscle, particularly along the lateral line (Fig 8). This dark muscle contains large amounts of myoglobin [44]. In red tilapia, fillets are composed mainly of red-colored muscle, as well as dark muscle along the lateral line (Fig 8).
In vivo staining of fresh fillets of (A) black and (B) red Nile tilapia (Oreochromis niloticus). Source: Personal archive. The yellow arrow indicates the lateral line.
In this tilapia variety, although the hue angle was lower than that of black tilapia, it was close to 0°, indicating a reddish color. No previous study had evaluated the color parameters of red tilapia fillets, and, to date, no research has investigated whether the same genetic factors that control skin color are responsible for fillet color. The color differences between fillets from black and red tilapia seem to be mainly related to the location and quantity of myoglobin.
According to Purslow et al. [43], the color of myoglobin may change according to its biochemical state and degree of oxidation or reduction. In addition to being the main pigment in meat, myoglobin acts as a pro-oxidant factor, initiating the process of lipid oxidation. As reviewed by Domínguez et al. [45] lipid oxidation is the main non-microbial cause of deterioration in meat quality. The presence of polyunsaturated fatty acids (PUFAs) and heme pigments makes fish meat more prone to lipid oxidation than other meats, given that these fatty acids are easily oxidized [46].
The presence of metals, in free form or as heme proteins, is one of the main factors determining the oxidative stability of meat. Metals such as iron and copper act as catalysts for different processes and stages of lipid oxidation [45,47]. According to Domínguez et al. [45], most of the iron in biological tissues occurs in the form of hemoglobin and myoglobin, the most abundant heme proteins in meat. Oxidation of heme proteins negatively affects meat color, thereby influencing consumer acceptability. Previous study demonstrated that the oxidation of lipids and heme proteins in meat occurs simultaneously, and these processes seem to stimulate each other [48].
For example, oxyheme oxidation results in the formation of Met–heme and hydroperoxyl or superoxide radicals, which are converted into hydrogen peroxide [45]. Met–heme can react with hydrogen peroxide or pre-formed lipid hydroperoxide to produce highly reactive ferryl heme radicals [49]. All reactive oxygen species generated during oxyheme oxidation, as well as ferryl radicals, can abstract hydrogen from PUFAs and initiate lipid oxidation [48]. Thus, the oxidative stability and color of meat depend on the balance between anti- and pro-oxidant compounds [46].
The oxidative stability of fillets was assessed by determining parameters of primary (conjugated dienes) and secondary (TBARS) lipid oxidation and the antioxidant capacity of tilapia muscles (DPPH• and ABTS•+). Black tilapia fillets exhibited lower oxidative stability, given the greater primary lipid oxidation and lower antioxidant capacity. Thus, black tilapia fillets have lower oxidative and color stability. The lighter color of black tilapia fillets may be a reflection of the increased production of primary lipid oxidation products. It was expected that the conversion of myoglobin to metmyoglobin would change fillet color from bright red to brown. However, a brown color was not observed, possibly due to the lower amount of myoglobin in black tilapia fillets, which are mostly composed of light muscle [50].
In summary, the lower oxidative and color stability of black tilapia fillets might be related to their lower pH value. Several conditions can initiate the oxidation cycle of heme proteins and lipids [51]. For example, in meat with low pH values (<7), the solubility of iron is increased, favoring heme protein oxidation [51]. Furthermore, the findings may be related to the lipid content of black tilapia muscles. Garduño-Lugo et al. [52] demonstrated that the fillet lipid content of wild-type Nile tilapia was higher than that of a red tilapia hybrid (2.07% vs. 0.33%, respectively).
Oxidative reactions are known to reduce the nutritional value of meat, resulting from a loss of essential fatty acids and vitamins [45]. Furthermore, such reactions affect sensory quality by altering the color, texture, and consistency of meat, thereby influencing consumer acceptance [53]. Here, the texture profile of fillets was assessed to gain insight into their quality and potential acceptance by consumers. The analysis evaluates attributes that influence the perception of meat quality during chewing [26]. The results demonstrated the negative effects of primary oxidation on the texture profile of fillets. Black tilapia samples had greater hardness, stickiness, and chewiness, as well as lower cohesiveness. Thus, these fillets had greater resistance to deformation, requiring more force to be chewed but falling apart more easily because of their reduced capacity to maintain structural integrity. As such, there is a greater tendency to adhere to surfaces (utensils, teeth, and lips). These effects can be attributed to lipid oxidation, as lipids are crucial for maintaining moisture (juiciness) and tenderness [54]. Together, these characteristics indicate that black tilapia fillets are less appealing than red tilapia fillets.
As stated by Domínguez et al. [45], deterioration of the qualitative attributes of fillets begins when fish are slaughtered and continues progressively until the final product is consumed. Therefore, it is necessary to adopt strategies to increase the antioxidant capacity of black tilapia fillets. Furthermore, the handling, processing, and storage of fillets must be strictly controlled to avoid the progression of these undesirable reactions and minimize economic losses [55].
Fish meat is mainly sold in the form of fillet cuts. Such processing generates large amounts of waste, including viscera, head, scales, and skin. These materials could be repurposed, minimizing the environmental impacts caused by the indiscriminate disposal of fish waste [6]. For this, it is essential to understand the parameters related to skin resistance. Oliveira et al. [56], in evaluating the quality parameters of tilapia skin, observed that fish leather had a tensile strength of 13.52 N/mm−2 and a tear strength of 53.85 N/mm−1. These values are similar to those observed in the present study for black tilapia (13.15 N/mm−2 and 51.45 N/mm−1, respectively). Although red tilapia exhibited lower tensile (10.81 N/mm−2) and tear (45.70 N/mm−1) strengths than black tilapia, these values are still higher than those observed for some mammals, such as rabbits (8.98 N/mm−2 and 24.25 N/mm−1, respectively) [56]. Such findings confirm the quality of tilapia leather and its potential use in the manufacture of clothes and accessories.
Tilapia skin has also been applied in biomedicine for the treatment of burns and extraction of bioactive compounds, such as collagen. This fibrous insoluble protein is present in large amounts in tilapia skin [10,11,13]. The collagen yield of red tilapia was lower than that of black tilapia (23.01% vs. 25.76%, respectively); nevertheless, these values are higher than those previously reported for red tilapia strains (22.58% [57] and 19.04% [16]). Medina-Medrano et al. [57] assessed the antioxidant properties of collagen from red tilapia skin and concluded that it shows potential as an active ingredient of nutraceuticals, pharmaceuticals, and functional foods. The results of the current study are in agreement with these observations, as the antioxidant capacity of red tilapia skin was higher than that of black tilapia skin.
Hydroxyproline, an amino acid that composes type I collagen, was observed at higher concentrations in the skin of black tilapia. This result is associated with the higher collagen content of this tilapia variety, contrary to the findings of Reátegui-Pinedo et al. [16]. The authors found no significant differences in the skin hydroxyproline content between different tilapia strains, including the red variety.
Type I collagen is composed of two types of peptides, α1 and α2, which fold into a triple helix at a 2:1 ratio, that is, two α1 chains for every one α2 chain. The translation and folding rates of collagen peptides must be tightly regulated to avoid structural imbalances and the synthesis of α1 homotrimers. DEAH box helicase 9 and vimentin play important roles in processes related to structural RNA changes, directly binding to the post-transcriptional element and promoting its unwinding, thereby increasing translation efficiency [58].
The expression of type I collagen subunits α1, α2, and α3 was increased in the skin of red tilapia. However, the expression of genes encoding proteins that regulate the translation and folding of collagen peptides (vim and dhx9) was higher in black tilapia. This result suggests that, although black tilapia has a lower gene expression of type I collagen subunits, its collagen synthesis may be more efficient. Genes encoding proteins that are crucial for the translation and stability of collagen mRNA are more highly expressed in the skin of these individuals.
In this study, skin color phenotype did not influence the biometric characteristics of GIFT Nile tilapia. However, tilapia varieties differed with regard to meat quality and skin texture. Red tilapia fillets had superior qualitative attributes, aspects that may greatly influence consumer acceptance. For instance, fillet pH was higher and b*, L*, ThL, and CL were lower in red tilapia fillets than in black tilapia fillets. Regarding skin quality parameters, black tilapia tended to have tougher skin than red tilapia. There was also a significant difference in collagen yield and the gene expression of type I collagen subunits and cofactors, indicating greater efficiency in the synthesis of type I collagen in the skin of black tilapia.
Conclusions
The results suggest that the skin color phenotype of Nile tilapia may be associated with important metabolic pathways influencing fillet and skin characteristics. These findings can support future research aimed at identifying optimal uses for Nile tilapia varieties and optimizing the utilization of waste.
Acknowledgments
The authors would like to thank the Graduate Program in Animal Science, Department of Animal Science, State University of Maringá (UEM) and the Federal University of Sergipe (UFS) for their technical support. The authors also thank the Fish Farming Experimental Station (CODAPAR/UEM, Floriano, Paraná, Brazil) and the PeixeGen Research Group, particularly Dr. Carlos Antônio Lopes de Oliveira and Dr. Ricardo Pereira Ribeiro, for providing the fish used in this study.
References
- 1.
FAO. Food and Agriculture Organization of the United States. In Brief to The State of World Fisheries and Aquaculture. Towards Blue Transformation. Rome. 2022. https://doi.org/10.4060/cc0463en
- 2. Prabu E, Rajagopalsamy CBT, Ahilan B, Jeevagan IJ, Renuhadevi M. Tilapia–an excellent candidate species for world aquaculture: A review. Annu Res Rev Biol. 2019; 31: 1–14. https://doi.org/10.9734/arrb/2019/v31i330052
- 3.
Peixe BR. Anuário Peixe BR da Psicultura. Brazil. 2023. Available from: https://www.aen.pr.gov.br/sites/default/arquivos_restritos/files/documento/2023-03/anuariopeixebr2023.pdf
- 4. Planello DRF, Coelho SRC, Nunes R, Gameiro AH. The tilapia market in the state of São Paulo. Informações Econômicas. 2016; 46: 36–51
- 5. Nuryanto N, Afifah DN, Sulchan M, Martosuyono P, Ihsani K, Kurniastuti PL. Potential of Nile tilapia (Oreochromis niloticus) as an alternative complementary food ingredient for stunting children. Maced J Med Sci. 2022; 10: 1170–1177. https://doi.org/10.3889/oamjms.2022.9650
- 6. Yoshida GM, Lhorente JP, Correa K, Soto J, Salas D, Yáñez JM. Genome-wide association study and cost-efficient genomic predictions for growth and fillet yield in Nile tilapia (Oreochromis niloticus). G3 (Bethesda). 2019; 9: 2597–2607. https://doi.org/10.1534/g3.119.400116
- 7. Silva JFX, Ribeir K, Silva JF, Cahú TB, Bezerra RS. Utilization of tilapia processing waste for the production of fish protein hydrolysate. Anim Feed Sci Technol. 2014; 196: 96–106. https://doi.org/10.1016/j.anifeedsci.2014.06.010
- 8. Corrêa SS, Gonçalves GO, Coradini MF, Gasparino E, Feihrmann AC, Siemer S, et al. Quality of Oreochromis niloticus and Cynoscion virescens fillets and their by-products in flours make for inclusion in instant food products. PLoS One. 2023; 18: e0279351. https://doi.org/10.1371/journal.pone.0279351
- 9. Mota FA, Costa Filho JT, Barreto GA. The Nile tilapia viscera oil extraction for biodiesel production in Brazil: An economic analysis. Renew Sustain Energy Rev. 2019; 108: 1–10. https://doi.org/10.1016/j.rser.2019.03.035
- 10. Dermawan SK, Ismail ZMM, Jaffri MZ, Abdullah HZ. Extraction and characterization of natural hydroxyapatite from black tilapia fish bone for biomedical applications. Key Engineering Materials. 2022; 908: 159–164. https://doi.org/10.4028/p-3j8b8t
- 11. Lee TC, Mohd Pu’ad NAS, Alipal J, Muhamad MS, Basri H, Idris MI, et al. Tilapia wastes to valuable materials: A brief review of biomedical, wastewater treatment, and biofuel applications. Mater Today Proc. 2022; 57: 1389–1395. https://doi.org/10.1016/j.matpr.2022.03.174
- 12. Lima Junior EM, Picollo NS, Miranda MJBD, Ribeiro WLC, Alves APNN, Ferreira GE, et al. The use of tilapia skin (Oreochromis niloticus), as an occlusive biological dressing, in the treatment of burn wounds. Rev Bras Queimaduras. 2017; 16: 10–17.
- 13. Lima Júnior EM, Moraes Filho MOD, Forte AJ, Costa BA, Fechine FV, Alves APNN, et al. Pediatric burn treatment using tilapia skin as a xenograft for superficial partial-thickness wounds: a pilot study. J Burn Care Res. 2020; 41: 241–247. pmid:31504615
- 14. Jafari H, Lista A, Siekapen MM, Ghaffari-Bohlouli P, Nie L, Alimoradi H, et al. Fish collagen: Extraction, characterization, and applications for biomaterials engineering. Polymers. 2020; 12: 2230. pmid:32998331
- 15. Abuine R, Rathnayake AU, Byun HG. Biological activity of peptides purified from fish skin hydrolysates. Fish Aquatic Sci. 2019; 22: 1–14. https://doi.org/10.1186/s41240-019-0125-4
- 16. Reátegui-Pinedo N, Salirrosas D, Sánchez-Tuesta L, Quiñones C, Jáuregui-Rosas SR, Barraza G, et al. Characterization of collagen from three genetic lines (gray, red and F1) of Oreochromis niloticus (tilapia) skin in young and old adults. Molecules. 2022; 27: 1123. https://doi.org/10.3390/molecules27031123
- 17. Moses M, Chauka LJ, Koning DJ, Palaiokostas C, Mtolera MS. Growth performance of five different strains of Nile tilapia (Oreochromis niloticus) introduced to Tanzania reared in fresh and brackish waters. Sci Rep. 2021; 11: 11147. https://doi.org/10.1038/s41598-021-90505-y
- 18. Wang L, Zhu W, Dong Z, Song F, Dong J, Fu J. Comparative microRNA-seq analysis depicts candidate miRNAs involved in skin color differentiation in red tilapia. Int J Mol Sci. 2018; 19: 1209. pmid:29659520
- 19. Li BJ, Zhu ZX, Gu XH, Lin HR, Xia JH. QTL mapping for red blotches in Malaysia red tilapia (Oreochromis spp.). Mar Biotechnol (NY). 2019; 21: 384–395. https://doi.org/10.1007/s10126-019-09888-9
- 20. Duarte RF, los Santos JRG. Stunning of Nile tilapia by thermonarcosis and its effect on frozen fillets. Pesq. Agropec. Bras. 2024; 59: e03565. https://doi.org/10.1590/S1678-3921.pab2024.v59.03565
- 21. Goes ESR, Goes MD, Castro PL, Lara JAF, Vital ACP, Ribeiro RP. Imbalance of the redox system and quality of tilapia fillets subjected to pre-slaughter stress. PLoS One. 2019; 14: e0210742. pmid:30645627
- 22. Ali S, Rajput N, Li C, Zhang W, Zhou G. Effect of freeze-thaw cycles on lipid oxidation and myowater in broiler chickens. Braz. J Poult. Sci. 2016; 18: 35–40. https://doi.org/10.1590/1516-635x1801035-040
- 23. Goes ESR, Lara JAF, Gasparino E, Goes MD, Zuanazzi JSG, Lopera-Barrero NM, et al. Effects of transportation stress on quality and sensory profiles of Nile tilapia fillets. Sci agric. 2018; 75: 321–328. https://doi.org/10.1590/1678-992X-2016-0387
- 24. Gomes BO, Oliveira CM, Marins AR, Gomes RG, Feihrmann AC. Application of microencapsulated probiotic Bifidobacterium animalis ssp. lactis BB-12 in Italian salami. J Food Process Preserv. 2021; 45: e15841. https://doi.org/10.1111/jfpp.15841
- 25. Pang B, Bowker B, Zhuang H, Yang Y, Zhang J. Research Note: Comparison of 3 methods used for estimating cook loss in broiler breast meat. Poult Sci. 2020; 99: 6287–6290. pmid:33142547
- 26. Nascimento R, Campagnol PCB, Monteiro ES, Pollonio MAR. Substituição de cloreto de sódio por cloreto de potássio: influência sobre as características físico-químicas e sensoriais de salsichas. Alim. Nutr. 2007; 18: 297–302.
- 27. Brand-Wiliams W, Cuvelier ME, Berset C. Use of a free radical method to evaluate antioxidant activity. LWT-Food Sci Technol. 1995; 28 25–30. https://doi.org/10.1016/S0023-6438(95)80008-5
- 28. Rufino MSM, Alves RE, Brito ES, Morais SM, Sampaio CG, Pérez-Jiménez J, et al. Metodologia científica: determinação da atividade antioxidante total em frutas pela captura do radical livre ABTS.+. Comunicado Técnico, 128, 2007. Available from: https://ainfo.cnptia.embrapa.br/digital/bitstream/CNPAT/10225/1/Cot_128.pdf.
- 29. Khatlab AS, Del Vesco AP, Oliveira Neto AR, Fernandes RPM, Gasparino E. Dietary supplementation with free methionine or methionine dipeptide mitigates intestinal oxidative stress induced by Eimeria spp. challenge in broiler chickens. J Anim Sci Biotechnol. 2019; 10: 58. https://doi.org/10.1186/s40104-019-0353-6
- 30. Pontes KM, Del Vesco AP, Khatlab AS, Lima Júnior JWR, Cangianelli GH, López JCC, et al. Effects of inclusion of the blend of essential oils, organic acids, curcumin, tannins, vitamin E, and zinc in the maternal diet, and of incubation temperature on early and late development of quail. Poult Sci. 2024; 103: 104022. pmid:39068694
- 31. Matiucci MA, Feihrmann AC, Oliveira GG, Corrêa SS, Goes ESR, Souza MLR. Strength quality of tilapia and salmon skins submitted to tanning process with vegetable tannin. Res. Soc. Dev. 2021; 10: e43910817242. https://doi.org/10.33448/rsd-v10i8.17242
- 32. Bergman I, Loxley R. Two improved and simplified methods for the spectrophotometric determination of hydroxyproline. Anal Chem. 1963; 35: 1961–1965. https://doi.org/10.1021/ac60205a053
- 33. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods. 2001; 25: 402–408. pmid:11846609
- 34.
SAS. Statistical Analysis System 9. User’s guide: basic and statistic. Intitute Inc. Cary, NC, USA. 2002.
- 35. Soares KMP, Gonçalves AA. Quality index method (QIM) application on shelf-life estimation of skinned fillets of Nile tilapia (Oreochromis niloticus) kept in ice. Semin Cienc Agrar. 2012; 33: 2289–2300. https://doi.org/10.5433/1679-0359.2012v33n6p2289
- 36. Baygar T, Alparslan Y, Kaplan M. Determination of changes in chemical and sensory quality of sea bass marinades stored at +4 (±1)°C in marinating solution. CYTA J Food. 2012; 10: 196–200. https://doi.org/10.1080/19476337.2011.614016
- 37. Zhang Z-Y, Limbu SM, Zhao S-H, Chen L-Q, Luo Y, Zhang M-L, et al. Dietary l-carnitine supplementation recovers the increased pH and hardness in fillets caused by high-fat diet in Nile tilapia (Oreochromis niloticus). Food Chem. 2022; 382: 132367. https://doi.org/10.1016/j.foodchem.2022.132367
- 38. Bainy EM, Bertan LC, Corazza ML, Lenzi MK. Effect of grilling and bakingon physicochemical and textural properties of tilapia (Oreochromis niloticus) fishburger. J Food Sci Technol. 2015; 52: 5111–5119. https://doi.org/10.1007/s13197-014-1604-3
- 39. Goes ES, Lara JA, Gasparino E, Del Vesco AP, Goes MD, Alexandre Filho L, et al. Pre-slaughter stress affects ryanodine receptor protein gene expression and the water-holding capacity in fillets of the Nile tilapia. PLoS One. 2015; 10: e0129145. pmid:26053858
- 40. Robb DHF, Kestin SC, Warriss PD. Muscle activity at slaughter: I. Changes in flesh colour and gaping in rainbow trout. Aquaculture. 2000; 182: 261–269. https://doi.org/10.1016/S0044-8486(99)00273-2
- 41. Mancini RA, Hunt MC. Current research in meat color. Meat Sci. 2005; 71: 100–121. pmid:22064056
- 42. Yu QP, Feng DY, Xiao J, Wu F, He XJ, Xia MH, et al. Studies on meat color, myoglobin content, enzyme activities, and genes associated with oxidative potential of pigs slaughtered at different growth stages. Asian-Australas J Anim Sci. 2017; 30: 1739–1750. pmid:28728400
- 43. Purslow PP, Warner RD, Clarke FM, Hughes JM. Variations in meat colour due to factors other than myoglobin chemistry; a synthesis of recent findings (invited review). Meat Sci. 2020; 159: 107941. pmid:31513992
- 44. Chow CJ, Wu JC, Lee PF, Ochiai Y. Structural and autooxidation profiles of myoglobins from three species and one hybrid of tilapia (Cichlidae, Perciformes). Comp Biochem Physiol B Biochem Mol Biol. 2009; 154: 274–281. pmid:19602446
- 45. Domínguez R, Pateiro M, Gagaoua M, Barba FJ, Zhang W, Lorenzo JM. A Comprehensive Review on Lipid Oxidation in Meat and Meat Products. Antioxidants (Basel). 2019; 8: 429. pmid:31557858
- 46. Sohn J-H, Taki Y, Ushio H, Kohata T, Shioya I, Oshima T. Lipid oxidations in ordinary and dark muscles of fish: influences on rancid off-odor development and color darkening of yellowtail flesh during ice storage. J Food Sci. 2005; 70: s490–s496. https://doi.org/10.1111/j.1365-2621.2005.tb11497.x
- 47. Guyon C, Meynier A, de Lamballerie M. Protein and lipid oxidation in meat: A review with emphasis on high-pressure treatments. Trends Food Sci. Technol. 2016; 50: 131–143. https://doi.org/0.1016/j.tifs.2016.01.026
- 48. Chaijan M. Review: Lipid and myoglobin oxidations in muscle foods. Songklanakarin J. Sci. Technol. 2008; 30:47–53.
- 49. Svistunenko DA. Reaction of haem containing proteins and enzymes with hydroperoxides: the radical view. Biochim Biophys Acta. 2005; 1707: 127–55. pmid:15721611
- 50. Viana FM, Canto ACVCS Canto, Monteiro MLG, Pereira APAAS, Rodrigues BL, Conte-Junior CA. Instrumental color and oxidative stability of light and dark muscles of Nile tilapia. Cienc. Rural. 202; 50: e20200287. https://doi.org/10.1590/0103-8478cr20200287
- 51. Hoa VB, Cho SH, Seong PN, Kang SM, Kim YS, Moon SS, et al. The significant influences of pH, temperature and fatty acids on meat myoglobin oxidation: a model study. J Food Sci Technol. 2021; 58: 3972–3980. pmid:34471321
- 52. Garduño-Lugo M, Granados-Alvarez I, Olvera-Novoa MA, Muñoz-Córdova G. Comparison of growth, fillet yield and proximate composition between Stirling Nile tilapia (wild type) (Oreochromis niloticus, Linnaeus) and red hybrid tilapia (Florida red tilapia×Stirling red O. niloticus) males. Aquac Res. 2003; 34: 1023–1028. https://doi.org/10.1046/j.1365-2109.2003.00904.x
- 53. Huang X, Ahn DU. Lipid oxidation and its implications to meat quality and human health. Food Sci Biotechnol. 2019; 28: 1275–1285. pmid:31695926
- 54. Amaral AB, Silva MV, Lannes SCS. Lipid oxidation in meat: mechanisms and protective factors–a review. Food Sci Technol. 2018; 38: 1–15. https://doi.org/10.1590/fst.32518
- 55. Králová M. The effect of lipid oxidation on the quality of meat and meat products. Int. J. Food Sci. Technol. 2015; 2: 125–132.
- 56. Oliveira GG, Gasparino E, Castilha LD, Marengoni NG, Goes ESR, Almeida FLA, et al. Characterization and strength quality of the Oryctolagus cuniculus leather compared to Oreochromis niloticus leather. Sci. World J. 2022; 2022: 1–9. https://doi.org/10.1155/2022/4561404
- 57. Medina-Medrano JR, Quiñones-Muñoz TA, Arce-Ortíz A, Torruco-Uco JG, Hernández-Martínez R, Lizardi-Jiménez MA, et al. Antioxidant activity of collagen extracts obtained from the skin and gills of Oreochromis sp. J Med Food. 2019; 22: 722–728. https://doi.org/10.1089/jmf.2019.0013
- 58. Stefanovic B. RNA protein interactions governing expression of the most abundant protein in human body, type I collagen. Wiley Interdiscip Rev RNA. 2013; 45: 535–545. https://doi.org/10.1002/wrna.1177