Evaluation of the Composition of Infant Foods: Determination of Lactoferrin by HPLC-DAD and Fatty Acids by GC-FID in Human Milk and Infant Formulas

Abstract

Nutrition in early childhood plays a decisive role in growth and lifelong health, with human milk considered the gold standard, while infant formulas (IFs) serve as an alternative when breastfeeding is not possible. This study aimed to quantify and compare the concentrations of human lactoferrin and the fatty acid profile in human milk (colostrum, transitional, and mature) and in IF. Human milk samples were obtained from donors, and the IF were reconstituted according to label instructions. Lactoferrin was quantified by high-performance liquid chromatography; lipids were analyzed by gas chromatography. Data were evaluated using analysis of variance, Tukey’s test, and principal component analysis (PCA). The results showed higher concentrations of lactoferrin in human milk, especially in colostrum (1.173 g L^-1), while IF presented absent or lower levels. Human milk exhibited greater lipid diversity, whereas IF showed a higher proportion of monounsaturated fatty acids and the absence of docosahexaenoic acid (DHA) in some formulations. PCA revealed differences in composition between human milk and IF, concluding that IF does not reproduce the complexity of human milk, highlighting the need for technological advances and experimental and clinical studies to support industrial improvements and guide safer nutritional practices.

Keywords:
lactoferrin; chromatography; infant formulas; infant nutrition; human milk

Introduction

Childhood is a critical period of development, during which nutrition plays a central and determinant role in modulating growth and long-term health outcomes.¹,² In this context, human milk is recognized as the primary nutritional source, recommended exclusively until six months of age and supplemented until two years or more, due to its dynamic composition, capable of adapting to the physiological needs of the infants.³-⁵

Lactation with human milk comprises three phases: colostrum (C), secreted in the first days postpartum (0 6 days), characterized as a thick, yellowish fluid rich in proteins, immunoglobulins, and antimicrobial factors; transitional (T), which presents a gradual evolution of macroand micronutrients between 7 and 14 days; and mature (M ≥ 15 days), whose primary function is to meet the energy requirements of the infant.⁶,⁷ Exclusive breastfeeding is therefore consolidated as the most effective and protective intervention in early childhood, capable of reducing infant mortality from preventable causes by up to 13%.⁸

Despite advances in breastfeeding rates, many women still face difficulties maintaining it, making the use of nutritional alternatives necessary. In these cases, donated human milk from human milk banks is used to meet the needs of premature or low birth weight newborns,⁹,¹⁰ while infant formulas (IFs) are recommended in specific situations of nutritional risk under professional guidance.¹¹,¹² According to Brazilian regulatory authorities, the Agência Nacional de Vigilância Sanitária (ANVISA)¹² classify IFs marketed in Brazil as starter (0-6 months), follow-on (6 to 12 months), and early childhood formulas (1-3 years), categories that reflect distinct nutritional requirements throughout development.

The composition of human milk highlights natural and functional nutrients such as lactoferrin, a multifunctional protein present throughout all stages of lactation, with antimicrobial, immunomodulatory, and intestinal integrity promoting properties.¹³-¹⁵ However, IFs containing lactoferrin use the bovine form, added as an isolated functional ingredient obtained from bovine whey extraction or recombinant technology during industrial processing.¹⁶ The concentration of this protein in bovine milk is significantly lower, resulting in reduced functional bioavailability and providing only partial protective effects.¹⁶,¹⁷

In addition, lipids represent the main energy source in human milk, accounting for approximately 50% of its total energy content.⁴,⁵ Similarly, IFs are regulated to provide between 40 and 55% of total energy from lipids, aiming to replicate the energetic profile of human milk and meet the high demands of infant development.¹⁸ In human milk, fat is organized into globules rich in triacylglycerols, with emphasis on polyunsaturated fatty acids (PUFAs) such as arachidonic acid (ARA), eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA), which are essential for brain and immune system development.⁷,¹⁹ Although formulas use lipid blends from different sources to approximate this profile, they still exhibit structural and bioavailability differences compared to human milk due to the organization of fatty acids within the triacylglycerol molecule.⁷,²⁰

Given these particularities, it becomes fundamental to compare donated human milk and commercial IFs to understand the nutritional quality of the foods offered to infants and how closely these formulas approximate the composition of human milk. Thus, the present study aimed to quantify and compare lactoferrin concentrations and fatty acid profiles in donated human milk and commercial IFs, using high-performance liquid chromatography (HPLC) and gas chromatography (GC) to evaluate their nutritional characteristics.

Experimental

Reagents and materials

The reagents used for lactoferrin quantification included dibasic sodium phosphate (Vetec Química Fina, Brazil), ultrapure water obtained from a Marte Científica^® purification system (Brazil), phosphoric acid (Casa da Química, Brazil) for preparation of the extraction buffer, trifluoroacetic acid (ACS Científica, Brazil), and HPLC grade acetonitrile (J.T. Baker, Mexico) as mobile phases for liquid chromatography. A lactoferrin from human milk standard (purity ≥ 85%, Sigma-Aldrich, St. Louis, USA) was used to construct the calibration curve for lactoferrin analysis. All reagents were of analytical or HPLC grade, ensuring reproducibility and compatibility with chromatographic analyses.

Reagents used for lipid extraction included methanol (purity ≥ 99.8%, Synth, Brazil), chloroform (purity ≥ 99.8%, ACS Científica, Brazil), and ultrapure water. For the esterification of fatty acid methyl esters (FAMEs), n-heptane (≥ 99.0%, Mallinckrodt Chemicals, USA), potassium hydroxide (purity ≥ 85%, Fmaia, Brazil), and methanol (purity ≥ 99.8%, Synth, Brazil) were used. The FAME Mix C4-C24 unsaturated analytical standard (purity ≥ 97%, Sigma-Aldrich, USA) was employed as an external standard for fatty acids analysis.

Sampling

Representative samples of dietary intake for exclusively breastfed newborns (0 6 months) were selected. Human milk at the three stages of lactation: C (0-6 days), T (7 14 days), and M (≥ 15 days); was provided by the Human Milk Bank of the Hospital Universitário Regional de Maringá (Brazil), under approval by the Research Ethics Committee (approval No. 6.233.963/2025). The classification of lactation stages followed the criteria adopted by human milk banks in Brazil.²¹ Sample selection included raw human milk that had been discarded due to contamination, identified by the presence of foreign matter such as dirt, hair strands, or textile particles, and therefore considered noncompliant with consumption standards. Human milk samples were transported at a refrigerated temperature of 4 °C. The collected volumes were approximately 800 mL of C (10 donors), 1500 mL of T (12 donors), and 1800 mL of M (19 donors). Each lactation stage was homogenized separately, forming three individual pools, which were stored in glass containers at 18 °C until analysis. It is important to note that no information was available regarding maternal age or dietary intake of the donors whose human milk samples were used in this study. These factors are known to influence both lactoferrin levels and fatty acid composition in human milk. Therefore, this lack of donor-specific information represents a limitation of the present study and was considered on interpreting the results.

In addition, seven commercially available IFs (IF1-IF7) (400 g each) were donated by the Hospital Universitário Regional de Maringá and were included in this study. The formulas were reconstituted using distilled water according to the proportions specified on the labels of the manufacturers. Detailed product information is presented in Table 1. It should be noted that IF labels state that “this product should only be used in the feeding of children under 1 year of age with the explicit recommendation of a physician or nutritionist”, in accordance with Decree No. 9579,²² which regulates the commercialization of foods for infants and young children.

Table 1
Description of the label of commercial IFs

Sample	Recommended age for consumption	Reconstitution indicated / (g of powder)	Water / mL	Portion / (g mL-1)	Carbohydratesa / g	Lipidsa / g	Proteinsa / g
IF1	0-6 months	13.1	90	0.146	60	26	9.5
IF2	0-6 months	13.5	90	0.15	55	26	11
IF3	0-6 months	13.3	90	0.148	55	28	9.3
IF4	0-6 months	13.8	90	0.153	55	24	8.5
IF5	0-6 months	13.8	90	0.153	52	26	9.8
IF6	0-3 years	13.8	100	0.138	54	25	14
IF7	0-3 years	12.9	90	0.143	53	27	14

• IF: infant formula;
• ^a
  nutritional data of IFs refer to 100 g of powdered product.

Human milk samples were organized into pooled samples, whereas IFs were analyzed individually. For both groups, analyses were performed in analytical triplicate. This experimental design was applied to the determination of lactoferrin and fatty acid profiling.

Lactoferrin quantification by HPLC

Lactoferrin was quantified using a HPLC (Alliance e2695, Waters, USA) coupled to a diode array detector (DAD) (2998, Waters, USA), following a method adapted from Frueh et al.²³

Samples, organized according to the previously described sampling design, were prepared in analytical triplicate. For sample preparation, approximately 4.0 g of liquid samples and 1.5 g of powdered samples were weighed on a gravimetric basis, considering concentration differences among matrices. Samples were homogenized in 10 mL of dibasic sodium phosphate buffer under agitation at 400 rpm (Fisatom 761-5, Brazil) for 10 min. Subsequently, samples were centrifuged at 6000 rpm at 4 °C for 15 min in a refrigerated centrifuge (Harrier 18/80, Sanyo MSE, Kent, UK). The fat layer was manually removed using a spatula, and samples were centrifuged again under the same conditions to ensure complete separation of the fat, supernatant, and precipitate layers. Thereafter, 5 mL of the supernatant was transferred to a 25 mL volumetric flask and brought to volume with dibasic sodium phosphate buffer. An aliquot of 1.5 mL was filtered through a polytetrafluoroethylene (PTFE) membrane (0.25 µm, Filtrilo, Brazil) using a syringe filter, transferred to vials, and subjected to chromatographic analysis.

Separation was performed on a Nucleodur Macherey-Nagel C8 column (4.6 × 150 mm, 5 µm, 110 Å) using an injection volume of 30 µL. Both column and sample temperatures were maintained at 25 °C. Before analysis, the chromatographic system was equilibrated for 5 min until baseline stabilization, defined as a delta variation below 30. The mobile phase consisted of ultrapure water (A) and acetonitrile (B), both acidified with 0.1% (v/v) trifluoroacetic acid. Elution was carried out using a linear gradient as follows: 70% A:30% B (0-5 min), 55% A:45% B (5-10 min), 40% A:60% B (10-12 min), 30% A:70% B (12-14 min), followed by re-equilibration to the initial conditions of 70% A:30% B (14-16 min). The mobile phase flow rate was set at 0.5 mL min^-1, and detection was performed using a DAD at 280 nm. A representative chromatogram of the human lactoferrin standard at a concentration of 50 mg L^-1 is shown in Figure 1. Lactoferrin quantification was performed by linear regression of peak area as a function of concentration, using an analytical curve constructed with lactoferrin standards ranging from 5 to 500 mg L^-1 (y = 3601.2x - 64,677; R² = 0.9919). Results were expressed in g L^-1.

Figure 1
Representative chromatogram of the HPLC-DAD chromatographic run detecting human lactoferrin.

Fatty acid quantification

Samples, organized according to the previously described sampling design, were subjected to lipid extraction following the method of Folch et al.²⁴ This procedure involved mixing the sample with a chloroform:methanol solution (2:1, v/v) under magnetic stirring (Fisatom, model 761-5, São Paulo, Brazil) at 400 rpm for 3 min, followed by the addition of distilled water and further agitation, resulting in a biphasic system. The upper phase, corresponding to the aqueous fraction, was discarded, while the lower phase containing the lipids was collected. The organic solvent was removed by evaporation under vacuum using a rotary evaporator. FAMEs were then prepared from the extracted lipids according to the International Organization for Standardization (ISO) No. 12966:2017.²⁶

Subsequently, the FAMEs were analyzed using a GC (Thermo Scientific, Trace GC Ultra model, Waltham, USA), equipped with a flame ionization detector (FID). Separation was performed using a CP-7420 capillary column (Agilent, Santa Clara, USA). The sample injection volume was 2.0 µL, with the injector set to split mode with a split ratio of 1:40. The flow rates of the gases used were fixed at 1.4 mL min^-1 for the carrier gas (hydrogen, H₂), 30 mL min^-1 for the replacement gas (nitrogen, N₂), and 30- and 300-mL min^-1 for the flame gases (H₂ and synthetic air), respectively. The injector and detector temperatures were set at 230 and 250 °C, respectively. The remaining heating conditions were set as follows: an initial temperature of 65 °C, maintained for 4 min, followed by heating to 185 °C at a rate of 16 °C min^-1, maintaining this temperature for 12 min. Subsequently, the temperature was increased to 235 °C at 20 °C min^-1 and maintained for 9 min, totaling an analysis time of 35 min. The identification of FAMEs was performed by comparing the retention times (t_R) of sample components with those of the FAME Mix C4-C24 (purity ≥ 97%) analytical standard, which was analyzed under the same chromatographic conditions as the samples (Figure 2). Peak areas were integrated using ChromQuest 5.0 software, and fatty acid composition was expressed as a percentage of relative area, with all samples analyzed in triplicate.

Figure 2
Chromatogram representing the identification of fatty acids by GC-FID in human milk and IFs. 1: butyric acid (4:0); 2: caproic acid (6:0); 3: caprylic acid (8:0); 4: capric acid (10:0) 5: lauric acid (12:0); 6: myristic acid (14:0); 7: myristoleic acid (14:1 n-9); 8: palmitic acid (16:0); 9: palmitoleic acid (16:1 n-7); 10: palmitoleic acid (16:1 n-9); 11: stearic acid (18:0); 12: oleic acid (18:1 n-9); 13: vaccenic acid (18:1 n-7); 14: linoleic acid (18:2 n-6); 15: linoleic acid conjugated (18:2 9c, 11t); 16: linoleic acid conjugated (18:2 10t, 12c); 17: α-linolenic acid (18:3 n-3); 18: γ-linolenic acid (18:3 n-6); 19: arachidic acid (20:0); 20: eicosenoic acid (20:1n-9); 21: heneicosanoic acid (21:0); 22: eicosadienoic acid (20:2 n-6); 23: dihomo-gamma-linoleic acid (20:3 n-6); 24: arachidonic acid (20:4n-6 ARA); 25: behenic acid (22:0); 26: tricosanoic acid (23:0); 27: erucic acid (22:1 n-9); 28: lignoceric acid (24:0); 29: nervonic acid (24:1 n-9); 30: eicosapentaecnoic acid (20:5n-3 EPA); 31: docosahexaenoic acid (22:6n-3 DHA).

Data analysis

The data obtained were submitted to the R software (v 4.1.1, Factoextra and FactoMineR packages),²⁶ for analysis of variance (ANOVA) and Tukey’s mean comparison test (p < 0.05), as well as PCA applied to the data obtained by the HPLC-DAD and GC-FID analytical techniques.

Results and discussions

Lactoferrin content and fatty acid profile

Table 2 presents the lactoferrin content determined by HPLC-DAD and the fatty acid profile assessed by GC-FID.

Table 2
Lactoferrin content was determined by HPLC-DAD, and fatty acid composition analyzed by GC-FID (relative area, %) of all analyzed samples

Compound	Sample
	C	T	M	IF1	IF2	IF3	IF4	IF5	IF6	IF7
Lactoferrin / (g L-1)	1.173 ± 0.102a	0.912 ± 0.02 b	0.841 ± 0.023b	0.407 ± 0.043d	nd	nd	nd	nd	nd	0.618 ± 0.013c
4:0 / %	0.008 ± 0.004b	0.032 ± 0.029b	0.016 ± 0.002b	2.608 ± 0.014a	0.007 ± 0.001b	0.040 ± 0.002b	0.009 ± 0.001b	0.011 ± 0.006b	0.024 ± 0.004b	0.032 ± 0.005b
6:0 / %	0.462 ± 0.041a,b	1.260 ± 0.968a,b	0.032 ± 0.001a	0.008 ± 0.001a	1.653 ± 0.167b	1.106 ± 0.057a,b	0.807 ± 0.323a,b	0.532 ± 0.122a,b	1.593 ± 0.269b	0.515 ± 0.092a,b
8:0 / %	0.096 ± 0.032a	0.226 ± 0.063a	0.151 ± 0.028a	1.132 ± 1.516a	0.904 ± 0.015 a	0.474 ± 0.117a	0.129 ± 0.057a	nd	0.534 ± 0.025a	0.190 ± 0.021a
10:0 / %	0.920 ± 0.242a	1.676 ± 0.103a	1.645 ± 0.150a	1.950 ± 0.001a	0.983 ± 0.012a	0.633 ± 0.051a	nd	0.560 ± 0.294a	14.583 ± 2.683b	28.489 ± 4.314b
12:0 / %	5.818 ± 0.896d	11.776 ± 1.581c	9.049 ± 0.411b,c	15.640 ± 0.106a	13.589 ± 0.405a,b	9.823 ± 0.238c	nd	nd	nd	nd
14:0 / %	7.862 ± 0.422e	10.932 ± 1.126d	10.229 ± 0.200d	5.549 ± 0.061a	4.668 ± 0.115a	4.121 ± 0.040a	0.621 ± 0.054b	2.540 ± 0.011c	0.419 ± 0.024b	0.404 ± 0.023b
16:0 / %	24.438 ± 0.037g	22.302 ± 0.413f	27.330 ± 0.049e	6.118 ± 0.120a	20.551 ± 0.118b	20.787 ± 0.322b	19.382 ± 0.581c	20.797 ± 0.093b	6.670 ± 0.005a	10.078 ± 0.086d
18:0 / %	6.538 ± 0.139e	5.470 ± 0.619c	7.785 ± 0.147d	2.724 ± 0.031a	2.768 ± 0.025a	3.310 ± 0.025a,b	3.124 ± 0.070a,b	5.281 ± 0.028c	3.857 ± 0.130b	2.952 ± 0.433a,b
20:0 / %	0.575 ± 0.022c	0.324 ± 0.139b	0.293 ± 0.007b	0.352 ± 0.009a,b	0.334 ± 0.004a,b	0.350 ± 0.044a,b	0.527 ± 0.002a,c	0.426 ± 0.028a,b,c	0.363 ± 0.018a,b	0.424 ± 0.045a,b,c
21:0 / %	0.956 ± 0.029d	0.593 ± 0.138c	0.355 ± 0.001b	0.015 ± 0.001a	0.009 ± 0.001a	0.019 ± 0.005a	0.023 ± 0.001a	0.024 ± 0.006a	0.013 ± 0.001a	0.022 ± 0.002a
22:0 / %	0.165 ± 0.010c	0.0076 ± 0.052a,b,c	0.077 ± 0.003a,b,c	0.062 ± 0.001a,b	0.017 ± 0.001a	0.043 ± 0.021a,b	0.118 ± 0.003b,c	0.079 ± 0.042a,b,c	0.047 ± 0.003ab	0.122 ± 0.018b,c
24:0 / %	0.402 ± 0.007c	0.169 ± 0.063b	0.066 ± 0.009a,b	nd	nd	nd	0.017 ± 0.002a	0.017 ± 0.010a	ND	nd
∑SFA / %	48.239 ± 1.390d,g	54.836 ± 3.271f,g	57.028 ± 0.577f	36.157 ± 1.202a,b	45.484 ± 0.439c,d	40.706 ± 0.034a,c	24.757 ± 0.951e	30.267 ± 0.516be	28.102 ± 2.310e	43.227 ± 3.644a,c,d
14:1n-9 / %	0.133 ± 0.073a	0.151 ± 0.132a	0.235 ± 0.002a	nd	nd	0.052 ± 0.001a	nd	0.181 ± 0.013a	nd	nd
16:1n-7 / %	0.465 ± 0.004a	1.171 ± 1.055a,b	1.852 ± 0.003b	0.037 ± 0.001a	0.175 ± 0.003a	0.215 ± 0.007a	0.239 ± 0.008a	0.479 ± 0.010a	0.033 ± 0.002a	0.179 ± 0.001a
16:1n-9 / %	1.954 ± 0.003b	1.052 ± 1.329a,b	0.108 ± 0.004a	0.139 ± 0.004a	0.026 ± 0.002a	0.032 ± 0.005a	0.029 ± 0.001a	0.040 ± 0.016a	0.137 ± 0.003a	0.014 ± 0.001a
18:1n-9 / %	29.158 ± 0.884g	24.471 ± 2.009f	24.371 ± 0.310f	45.280 ± 0.757a	36.206 ± 0.284b,c	38.588 ± 0.073b	56.935 ± 0.845d	51.930 ± 0.479e	46.456 ± 1.059a	33.387 ± 1.863c
18:1n-7 / %	2.067 ± 0.033b	1.828 ± 0.119a,b	1.241 ± 0.121a,b	1.268 ± 0.071a,b	1.179 ± 0.016a	1.200 ± 0.040a,b	1.535 ± 0.070a,b	1.531 ± 0.218a,b	1.546 ± 0.495a,b	1.436 ± 0.389a,b
20:1n-9 / %	0.083 ± 0.002b	0.037 ± 0.034a,b	0.047 ± 0.001a,b	0.002 ± 0.001a	0.002 ± 0.001a	0.028 ± 0.009a,b	nd	nd	nd	nd
22:1n-9 / %	0.013 ± 0.002a	0.017 ± 0.010a	0.016 ± 0.001a	nd	0.040 ± 0.003a	0.025 ± 0.001a	0.069 ± 0.017b	0.027 ± 0.014a	nd	nd
24:1n-9 / %	0.253 ± 0.017a,c	0.160 ± 0.061a,b	0.136 ± 0.007b	nd	nd	nd	0.164 ± 0.014a,b	0.265 ± 0.018a,c	0.308 ± 0.018c	nd
∑MUFA / %	34.127 ± 0.873c	28.886 ± 1.806f	28.006 ± 0.443f	46.727 ± 0.833a	37.676 ± 0.302b,c	40.140 ± 0.055b	58.999 ± 0.751d	54.454 ± 0.280e	48.481 ± 1.567a	35.015 ± 2.225c
18:2n-6 / %	15.323 ± 0.449a,b	14.364 ± 1.201a,b,c,d	13.419 ± 0.112a,b,d	14.199 ± 0.297a,b,c,d	14.635 ± 0.112a,b,c	16.289 ± 0.009c,e	12.916 ± 0.180b,d	12.210 ± 0.162d	20.338 ± 0.613f	17.966 ± 1.127e
18:2 9c, 11t / %	0.040 ± 0.003a	0.042 ± 0.009a	0.061 ± 0.001a	nd	nd	nd	nd	nd	nd	nd
18:2 10t, 12c / %	0.128 ± 0.005a	0.171 ± 0.093a	0.071 ± 0.003a	nd	nd	nd	nd	nd	nd	nd
18:3 n-3 / %	0.959 ± 0.029d	0.824 ± 0.120d	0.719 ± 0.009d	1.748 ± 0.039a	1.760 ± 0.019a	1.989 ± 0.002a,b	2.265 ± 0.033b	1.879 ± 0.051a	1.861 ± 0.055a	2.888 ± 0.184c
18:3 n-6 / %	0.195 ± 0.018a,c	0.158 ± 0.020c	0.157 ± 0.004c	0.248 ± 0.007a	0.312 ± 0.003b	0.341 ± 0.006b	0.372 ± 0.015b	0.341 ± 0.012b	0.239 ± 0.012a	0.231 ± 0.033a
20:2n-6 / %	0.569 ± 0.016c	0.367 ± 0.110b	0.289 ± 0.005b	0.019 ± 0.001a	nd	0.017 ± 0.003a	0.030 ± 0.018a	0.053 ± 0.001a	0.032 ± 0.001a	0.016 ± 0.001a
20:3n-6 / %	0.180 ± 0.010b,c	0.135 ± 0.019c,f	0.087 ± 0.005f	0.232 ± 0.010a,b	nd	0.220 ± 0.011a,b	0.188 ± 0.003a,b,c	0.359 ± 0.004d	0.438 ± 0.028e	0.245 ± 0.028a
20:4n-6 (ARA) / %	0.143 ± 0.007b,c	0.099 ± 0.071c	0.108 ± 0.003c	0.472 ± 0.007a	0.117 ± 0.002b,c	0.139 ± 0.002b,c	0.422 ± 0.002a	0.395 ± 0.015a	0.488 ± 0.035a	0.225 ± 0.036b
20:5n-3 (EPA) / %	0.015 ± 0.002a	0.085 ± 0.026b	0.039 ± 0.001a	0.018 ± 0.002a	0.016 ± 0.002a	0.010 ± 0.010a	0.027 ± 0.001a	0.023 ± 0.001a	nd	0.014 ± 0.015a
22:6n-3 (DHA) / %	0.084 ± 0.002c	0.032 ± 0.023b,c	0.015 ± 0.001b	0.180 ± 0.005a	nd	0.151 ± 0.009a	0.023 ± 0.012b	0.021 ± 0.010b	nd	0.173 ± 0.024a
∑PUFA / %	17.635 ± 0.518a,b	16.278 ± 1.465b	14.965 ± 0.134b	17.116 ± 0.369a,b	16.840 ± 0.137a,b	19.154 ± 0.021a,c	16.244 ± 0.200b	15.280 ± 0.236b	23.397 ± 0.743d	21.758 ± 1.418c,d

Lactoferrin exhibited the highest concentrations in human milk, particularly in the C (1.173 g L^-1), which differed significantly from the other samples (p < 0.05). T (0.912 g L^-1) and M (0.841 g L^-1) showed the expected reductions, with no significant differences between them (p > 0.05), reflecting the physiological decline of lactoferrin throughout lactation. These findings are consistent with the literature, which reports particularly high lactoferrin concentrations in colostrum due to the increased need for immunological protection during the neonatal period, characterized by adaptation to the extrauterine environment and immune system immaturity in newborns.²⁷ In contrast, IFs showed absent or significantly lower lactoferrin levels, such as IF1 (0.407 g L^-1) and IF7 (0.618 g L^-1). Part of this discrepancy can be attributed to the fact that the present study employed analytical methods using a human lactoferrin standard, whereas IFs typically contain bovine lactoferrin (bLF) or recombinant human lactoferrin (rhLF) produced through genetic engineering to mimic the bioactive functions of human milk, including immune modulation, antimicrobial activity, and support of intestinal integrity.²³,²⁸,²⁹ Nevertheless, Brazilian legislation allows the inclusion of these ingredients provided as they are demonstrated to be safe and suitable for use in IFs.³⁰ In this context, further studies are required to comprehensively compare different lactoferrin sources such human lactoferrin (hLF), bLF, and rhLF, considering their structural similarities, analytical differences, and functional and nutritional implications.

Additionally, GC-FID analysis revealed marked structural differences between human milk and IFs. Human milk exhibited greater lipid diversity, with 30 fatty acids identified in C, T, and M samples, whereas IFs ranged from 21 to 26 components depending on the manufacturer. The complexity observed in human milk reflects adaptive physiological mechanisms involving endogenous biosynthesis in the mammary gland, mobilization of fatty acids stored during pregnancy, and, most importantly, the direct influence of maternal diet.³¹ In contrast, IFs present a rigidly controlled lipid profile resulting from blends of vegetable oils, animal-derived fats (bovine and fish), and, in some cases, structured lipids such as mediumand long-chain triacylglycerols.³²,³³ Although these blends are formulated to optimize lipid absorption by infants, they differ substantially from human milk, particularly due to the absence or lower proportions of PUFAs and the distinct distribution of fatty acids within the triacylglycerol molecule at the sn-1, sn-2, and sn-3 positions,³⁴ which explains the absence of certain fatty acids saturated fatty acids (SFAs) stood out as the predominant fraction in human milk, with total sums of 48.239% in C, 54.836% in T, and 57.028% in M. In contrast, some formulas, such as IF1, IF4, IF5, and IF6, presented values that corresponded to approximately half of these levels, evidencing substantial differences in the structural composition of triglycerides. It is widely recognized that SFAs are the predominant fatty acids in human milk, followed by monounsaturated fatty acids (MUFAs) and PUFAs.³⁵,³⁶ Palmitic acid (16:0) was shown to be the predominant SFA in human milk, which presented values of 22.302% in T, 24.438% in C and 27.330% in M, while all IFs exhibited significantly lower fractions, especially IF1 (6.118%), IF6 (6.670%) and IF7 (10.078%), in agreement with Wu et al.³⁷ Although IFs use vegetable oils as sources of SFA (8:0-18:0), they cannot reproduce the pattern found in human milk.²⁰ This occurs because in human milk, long-chain SFAs (≥ 16:0) and unsaturated fatty acids are taken up via plasma, originating from the diet and lipid reserves, as well as shorter-chain SFAs, which are attributed to de novo synthesis in the mammary gland, limited by an enzyme that interrupts lipogenesis at 8, 10, 12, and 14 carbon.³⁸,³⁹

Conversely, IFs exhibited a predominance of MUFAs, particularly oleic acid (18:1n-9), which ranged from 33.387 to 56.935%, representing levels 1.3- to 2.2-fold higher than those observed in human milk samples (24.371-29.158%). These findings are consistent with the literature for both sample types.⁴⁰ Wu et al.³⁷ also reported that 18:1n-9 levels were approximately 22% higher in IFs than in human milk. These results reflect the extensive use of vegetable oils in IFs, which are major sources of 18:1n-9, linoleic acid (18:2n-6), and α-linolenic acid (18:3n-3). Such oils are preferred due to their oxidative stability, low cost, availability, and suitability for standardizing omega fatty acid classes in accordance with regulatory requirements.⁷,²⁰,⁴⁰,⁴¹ 18:1n-9 serves as an energy source, supports brain structure, modulates medium-chain fatty acid synthesis, lowers triacylglycerol melting point, and maintains lipid globule fluidity essential for metabolism.⁴⁰,⁴²

Regarding PUFAs, samples IF3, IF6, and IF7 showed levels 2-5% higher than the other IFs and human milk, a difference directly associated with the high content of 18:2n-6, which reached 16.289, 20.338 and 17.966%, respectively. The high value observed results from the industrial strategy of using vegetable oils as ingredients, especially soybean, corn, and sunflower, because they are economically accessible and naturally rich in omega-6.⁴³ The inclusion of these oils increases energy density, enhances the absorption of fat-soluble vitamins,⁴⁴ and ensures compliance with regulatory recommendations of 7-20% of 18:2n-6 relative to total fatty acids.³⁵ However, such high levels are not always physiologically necessary, as the minimum amount of 18:2n-6 required to prevent clinical deficiency in infants is less than 1% of energy intake, whereas vegetable oils often supply values more than ten times higher.⁴⁵ Furthermore, conjugated linoleic acid isomers (18:2 9c,11t and 18:2 10t, 12c) were detected in all stages of human milk (0.040-0.061%) but were not detected (nd) in any IFs. This exclusive presence contributes to the greater lipid diversity of human milk (30 fatty acids) compared to IFs (21-26), reflecting adaptive biosynthesis and maternal dietary influence. Although IFs use blends of vegetable oils to meet fatty acid and energy requirements, these formulations fail to replicate bioactive compounds with immunomodulatory and anti inflammatory functions and, despite industrial fortification, do not match the functional complexity or the natural lipid profile of human milk.⁷,²⁰

From a metabolic perspective, both 18:2n-6 and 18:3n-3 are essential fatty acids obtained exclusively through the diet. They compete for the same desaturation and elongation enzymes and serve as precursors of long-chain PUFAs. 18:2n-6 gives rise to ARA (20:4n-6), while 18:3n-3 is converted into EPA (20:5n-3) and DHA (22:6n-3), which are essential for growth, neurological development, and immune function.³⁵,³⁶ Despite technological advances, significant differences persist between the lipid profile of IFs and that of human milk, since the vegetable oils that constitute the lipid base of IFs mainly provide 18:1n-9 and 18:2n-6, but do not guarantee an adequate supply of 18:3n-3 or long-chain PUFAs.⁴¹ In the present study, the adoption of enrichment strategies with 18:3n-3 was observed, with values ranging from 1.754 (IF1 and IF2) to 2.88% (IF7), representing levels 2 to 4 times higher than those of human milk, and can be associated with sources such as flaxseed, canola and soybean oils, chia seeds and eggs.⁴⁶

However, the literature⁴⁷ indicates that IFs often contain high levels of 18:2n-6 without proportional supplementation of DHA and ARA, potentially leading to significant metabolic imbalances, including a threefold increase in plasma 18:2n-6 and up to a threefold reduction in ARA during the first weeks of life. Consistently, the present study observed the absence of DHA in IF2 and IF6. To mitigate such imbalances, long-chain PUFAs are added to IFs through fish oil, DHA-rich algal oil derived from Crypthecodinium cohnii, ARA-rich fungal oil from Mortierella alpina, or egg yolk phospholipids.²⁰,³²,⁴³ This supplementation explains the relatively high ARA levels observed in IF1 (0.472%), IF4 (0.422%), IF5 (0.395%), IF6 (0.488%), and IF7 (0.225%), as well as DHA levels detected in IF1 (0.180%), IF3 (0.151%), and IF7 (0.173%). The elevated ARA content observed in the present study aligns with its increasing incorporation into health-oriented food products, particularly IFs and products related to early-life nutrition.⁴⁸

The combination of lactoferrin and fatty acids represents an innovative advance in the field of infant nutrition, as it enables the formation of protein-lipid complexes with enhanced biological activity, promoting antimicrobial and immunomodulatory effects as well as supporting gastrointestinal development. This interaction opens new avenues for the development of functional foods and more sophisticated IFs, incorporating models already applied using bovine lactoferrin and milk fat globule membrane enriched with fatty acids to potentiate nutritional and neurodevelopmental benefits.⁴⁸,⁴⁹

Chemometric data analysis

Fatty acid composition was also evaluated through multivariate exploration using principal component analysis (PCA), following data pre-processing by autoscaling. PCA was applied to investigate similarities among samples based on lactoferrin content and fatty acid profiles, serving as an exploratory tool to identify patterns and group samples with similar characteristics (Figure 3).

Figure 3
PCA biplot of lactoferrin concentration and fatty acid composition for human milk (C, T, and M) and IFs (IF1 to IF7): (a) PC1 vs. PC2 and (b) PC1 vs. PC3 (c) correlation matrix showing the contribution of variables to the first three principal components (PC1, PC2, and PC3).

The correlation matrix between variables and principal components (Figure 3c) provides an integrated overview of how each fatty acid and lactoferrin contributed to the formation of the PCA axes. In this representation, red indicates positive loadings and purple indicates negative loadings, allowing interpretation of the contributions observed in Figures 3a and 3b. Overall, PC1 was strongly explained by variables that clearly differentiate human milk from IFs, whereas PC2 and PC3 served as secondary axes, refining the separation within these groups and revealing differences in the lipid composition of the evaluated infant foods. Individual data from the PCA load score charts (PC1 to PC4) are available in Supplementary Information (SI) section.

Figures 3a and 3b depict the PCA performed to assess similarities among samples, accounting for 69.1% of the total variance. PC1 explained the largest proportion of the total variance (40.3%), followed by PC2 (17.3%) and PC3 (11.5%), which together accounted for 28.8%. Overall, PC1 was decisive in clearly separating human milk samples (C, T, and M), positioned on the positive side of the axis, from IFs samples (IF1-IF7), located on the negative side.

Regarding the positive side of PC1, variables with strong positive loadings included lactoferrin, 20:2n-6, 21:0, 18:0, 16:1n-9, 16:1n-7, 16:0, 20:5n-3 (EPA), and 14:0, which were associated with the clustering of human milk samples (C, T, and M). These components consistently exhibited higher contributions in human milk compared with IFs. In contrast, the negative side of PC1 was more strongly influenced by 18:1n-9, 18:3n-3, 18:3n-6, 20:3n-6, DHA, and ARA, indicating a greater contribution of these fatty acids to the similarity observed among IFs. Palmitic acid (16:0) was the most predominant saturated fatty acid in human milk, whereas IFs showed higher concentrations of shortand medium-chain fatty acids ranging from 4:0 to 12:0, as observed in Figure 3c, since these fatty acids are commonly incorporated into IFs through the use of vegetable oils.²⁰

Additionally, 20:5n-3 (EPA) was present at higher concentrations in human milk, particularly in transition (0.085%) and mature milk (0.039%), corroborating the findings of Chamorro et al.⁵⁰ who reported that certain classes of fatty acids decrease throughout lactation. In contrast, variables with strong negative loadings on PC1, such as DHA and ARA, were present at higher proportions in IFs, reaching values of 0.180 in IF1 and 0.488% in IF6, respectively. These fatty acids are associated with ingredients used as sources of long-chain PUFAs for formula enrichment, including Mortierella alpina algae and fish oil as sources of ARA and DHA.²⁰

The differences observed along PC1 visually distinguish human milk samples on the positive side from IFs on the negative side, reflecting the typical lipid and protein signature of human milk. In contrast, negative loadings represent fatty acids more closely associated with IFs, particularly those added to meet nutritional requirements, including ARA, DHA, and precursors of the n-3 and n-6 series.⁵¹

Regarding PC2, variables with strong positive loadings included 20:0 and 22:0. Although 18:1n-7 showed a positive contribution to PC1, its strongest contribution was observed in PC2. These variables were associated with the clustering of IF4, IF5, IF6, IF7, and C on the positive side of PC2, whereas the negative side of PC2 was influenced by 16:1n 7, 14:0, 12:0, and 8:0, leading to the grouping of IF1, IF2, IF3, T, and M.

Although Figure 3c shows that PC2 received strong contributions from long-chain saturated fatty acids, such as arachidic acid (20:0) and behenic acid (22:0), Table 2 indicates that these components occur at low concentrations in both human milk and IFs. Vaccenic acid (18:1n-7) showed a relevant contribution to the clustering of human milk in the positive region of PC1 and exerted an even more pronounced influence in the positive region of PC2. As shown in Table 2, this contribution is associated with higher concentrations of this fatty acid in colostrum (2.067%) and in IF4-IF7, with values ranging from 1.436 to 1.535%.

Palmitoleic acid (16:1n-7) already strongly contributed to the clustering of human milk samples in PC1, also showed a strong contribution to the negative side of PC2, likely due to its higher concentrations in transition and mature milk (1.171 and 1.852%, respectively), whereas in IFs these concentrations ranged from 0.037 in IF1 to 0.215% in IF3. This fatty acid is synthesized in the endoplasmic reticulum through desaturation of palmitic acid (16:0) at carbon 9 via stearoyl-CoA desaturase-1 (SCD-1, ∆9 desaturase).⁵² Studies indicate that palmitoleic acid acts as a lipokine, with the ability to modulate metabolic pathways, enhance insulin sensitivity, and regulate lipolytic and lipogenic activity.⁵³ As shown in Table 2, its concentrations are higher in transition and mature human milk, while remaining low in IFs and colostrum. This pattern reflects intensified maternal lipogenesis during the postpartum period, a physiological process that IFs cannot replicate, since their lipid matrix is predominantly derived from vegetable oils with naturally low levels of this fatty acid.²⁰,⁵⁴

PC3, although explaining a smaller fraction of the total variability, highlights subtle differences in fatty acid profiles, mainly influenced by 18:2n-6, 24:0, and DHA, which showed the highest contributions to this component. According to Table 2, the highest concentrations of 18:2n-6 were observed in IF1, IF2, IF3, IF6, IF7, transition, and colostrum samples (positive side of PC3), ranging from 14.199 to 20.338%. In contrast, mature milk and formulas IF4 and IF5 (negative side of PC3) showed slightly lower values, ranging from 12.210 to 13.419%. Previous studies have also reported higher levels of 18:2n-6 in certain IFs compared with mature human milk,⁴¹ corroborating our findings and highlighting this fatty acid as an important variable contributing to differences among IFs along PC3.

Although DHA had already contributed to the similarity observed among IFs in PC1, its most pronounced contribution occurred in PC3. In this component, DHA concentrations were directly associated with IF1, IF3 and IF7, transition, and colostrum samples (0.032-0.180%), whereas mature milk and formulas IF4 and IF5 exhibited lower values (0.015-0.023%). DHA highlights the contrast between the naturally dynamic profile of human milk and the industrially adjusted composition of IFs, explaining its strong contribution to sample segregation in PC1 and PC3. In contrast, 24:0 was not detected in IF1, IF2, IF3, IF6, and IF7, but was present at higher concentrations in C (0.402%) and T (0.169%) compared with M (0.066%), IF4 and IF5 (0.017%).

Conclusions

Human milk exhibits unique protein and lipid characteristics, with high levels of lactoferrin, particularly in colostrum, and greater fatty acid diversity, predominantly long-chain saturated fatty acids such as palmitic acid (16:0). An adaptive pattern throughout lactation was also observed, characterized by a gradual reduction in lactoferrin levels and changes in fatty acid composition, reflecting physiological processes, mobilization of maternal reserves, and dietary influences. In contrast, IFs showed lower or absent lactoferrin levels, standardized lipid profiles, and a predominance of MUFAs, such as 18:1n-9, resulting from the use of vegetable oils and industrial fortification strategies that do not fully reproduce the bioactive complexity of human milk.

PCA reinforced these differences by clearly separating human milk samples from IFs and highlighting, through the correlation matrix, the specific contribution of certain fatty acids and lactoferrin to group discrimination. The findings emphasize that, although modern IFs aim to meet specific nutritional requirements through the addition of selected components, such as DHA, gaps in bioavailability and biological functionality remain. In this context, the results reinforce the importance of breastfeeding as the nutritional gold standard or, when not possible, the use of donor human milk from human milk banks. Moreover, these findings provide a scientific basis for improving IFs, encouraging approaches that consider not only quantitative composition but also the biological effects of these components. Finally, future studies should prioritize the investigation of the bioavailability and functional behavior of IFs components, aiming to elucidate how compositional differences between human milk and IFs translate into distinct biological outcomes.

Supplementary Information

Supplementary information (score and loading plots of PCA, PC1-PC4) is available free of charge at http://jbcs.sbq.org.br as a PDF file.

Supplementary PDF

Supplementary PDF

Acknowledgments

The authors thank CAPES, CNPq and PPSUS for financial support. They also thank the Human Milk Bank of the Hospital Universitário Regional de Maringá for the partnership and donation of human milk samples.

Data Availability Statement

Data will be made available on request.

References

• ¹ Hoffman, D. J.; J. Nutr. 2022, 152, 1803. [Crossref]
  » Crossref
• ² Li, S.; Nor, N. M.; Kaliappan, S. R.; SSM Popul. Health 2023, 24, 101533. [Crossref]
  » Crossref
• ³ Cândido, F. G.; Freitas, B. A. C.; Soares, R. C. S.; Bittencourt, J. M.; Ribeiro, D. N.; Morais, D. C.; Niquine, C. F.; Ribeiro, S. A. V.; Araújo, R. M. A.; Zucchetto, B. R.; Carvalho, T. C.; Rezende, I. C.; Einstein 2021, 19, eAO6148. [Crossref]
  » Crossref
• ⁴ Almeida, C. C.; Pereira, B. F. M.; Leandro, K. C.; Costa, M. P.; Spisso, B. F.; Conte Jr., C. A.; Int. J. Food Sci. 2021, 2021, 6657593. [Crossref]
  » Crossref
• ⁵ Yi, D.; Kim, S.; Nutrients 2021, 13, 3094. [Crossref]
  » Crossref
• ⁶ Ientz, G.; Alves, E.; Castro, M.; Frigo, G.; Tavares, C.; Visentainer, J. E. L.; Santos, O.; Visentainer, J. V.; J. Braz. Chem. Soc. 2023, 34, 1887. [Crossref]
  » Crossref
• ⁷ Muxfeldt, L. C.; Alves, E. S.; Zangirolami, M. S.; Silva, C. R.; Santos, P. D. S.; Santos, O. O.; Visentainer, J. V.; J. Food Compos. Anal. 2025, 148, 108216. [Crossref]
  » Crossref
• ⁸ Ministério da Saúde; Ministério da Saúde Lança Campanha de Amamentação com Foco na Redução de Desigualdades; Ministério da Saúde, 2024. [Link] accessed in July 2026
  » Link
• ⁹ Alves, E.; Castro, M.; Saqueti, B.; Manin, L.; Alencar, J.; Zacarias, J.; Bruni, A.; Madrona, G.; Visentainer, J. E. L.; Cristianini, M.; Santos, O.; Visentainer, J. V.; J. Braz. Chem. Soc. 2024, 35, e20230102. [Crossref]
  » Crossref
• ¹⁰ Zheng, Y.; Fowler, J.; Rector, C.; Tumin, D.; Herco, M.; J. Perinatol. 2025, 45, 665. [Crossref]
  » Crossref
• ¹¹ Moraes, I.; Sena, N.; Oliveira, H.; Albuquerque, F.; Rolim, K.; Fernandes, H.; Silva, N.; Rev. Enferm. Ref. 2020, 5, e19065. [Crossref]
  » Crossref
• ¹² Agência Nacional de Vigilância Sanitária (Anvisa); Anvisa Orienta sobre Uso Seguro de Fórmulas Infantis; Anvisa: Brasília, 2024. [Link] accessed in July 2026
  » Link
• ¹³ Zhao, C.; Chen, N.; Ashaolu, T. J.; J. Funct. Foods 2023, 108, 105741. [Crossref]
  » Crossref
• ¹⁴ Dierick, M.; Vanrompay, D.; Devriendt, B.; Cox, E.; Biochem. Cell Biol. 2021, 99, 61. [Crossref]
  » Crossref
• ¹⁵ Liu, N.; Feng, G.; Zhang, X.; Hu, Q.; Sun, S.; Sun, J.; Sun, Y.; Wang, R.; Zhang, Y.; Wang, P.; Li, Y.; Front. Nutr. 2021, 8, 701467. [Crossref]
  » Crossref
• ¹⁶ Korhonen, H.; Marnila, P. In Encyclopedia of Dairy Sciences; Fuquay, J. W.; Fox, P. F.; McSweeney, P. L. H., eds.; Elsevier: London, UK, 2011.
• ¹⁷ Lönnerdal, B.; Am. J. Clin. Nutr. 2014, 99, 712S. [Crossref]
  » Crossref
• ¹⁸ Agência Nacional de Vigilância Sanitária (Anvisa); Resolução da Diretoria Colegiada (RDC) No. 43, de 19 de setembro de 2011, Estabelece o Regulamento Técnico para Fórmulas Infantis para Lactentes; Diário Oficial da União (DOU), Brasília, 2011. [Link] accessed in July 2026
  » Link
• ¹⁹ Mititelu, M.; Lupuliasa, D.; Neacşu, S. M.; Olteanu, G.; Busnatu, S. Ş.; Mihai, A.; Popovici, V.; Măru, N.; Boroghină, S. C.; Mihai, S.; Ioniţă-Mîndrican, C.-B.; Scafa-Udrişte, A.; Foods 2024, 14, 46. [Crossref]
  » Crossref
• ²⁰ Bolognese, M. A.; Ferreira, C. S. R.; Alves, E. S.; Santos, P. D. S.; Figueiredo, A. L.; Zangirolami, M. S.; Santos Jr., O. O.; Visentainer, J. V.; J. Food Compos. Anal. 2024, 135, 106683. [Crossref]
  » Crossref
• ²¹ Agência Nacional de Vigilância Sanitária (Anvisa); Banco de Leite Humano: Funcionamento, Prevenção e Controle de Riscos; Anvisa, Brasília, 2008. [Link] accessed in July 2026
  » Link
• ²² Presidência da República; Decreto No. 9.579, de 22 de novembro de 2018; Diário Oficial da União (DOU), Brasília, Brazil, 2018. [Link] accessed in July 2026
  » Link
• ²³ Frueh, J. L.; Shu, P.; Vennard, T. R.; Gray, M. A.; Phillips, S. C.; J. AOAC Int. 2024, 107, 693. [Crossref]
  » Crossref
• ²⁴ Folch, J.; Lees, M.; Stanley, G. H. S.; J. Biol. Chem. 1957, 226, 497. [Crossref]
  » Crossref
• ²⁵ ISO 12966-2:2017: Animal and Vegetable Fats and Oils Gas Chromatography of Fatty Acid Methyl Esters Part 2: Preparation of Methyl Esters of Fatty Acids; ISO: Geneva, 2017. [Link] accessed in July 2026
  » Link
• ²⁶ RStudio; RStudio: Integrated Development for R, version 2024.04.2+764; RStudio; Boston, MA, USA, 2024.
• ²⁷ Zhang, Y.; Zhang, X.; Mi, L.; Li, C.; Zhang, Y.; Bi, R.; Pang, J.; Li, Y.; Nutrients 2022, 14, 3648. [Crossref]
  » Crossref
• ²⁸ Aly, E.; Ros, G.; Frontela, C.; J. Food Res. 2013, 2, 25. [Crossref]
  » Crossre
• ²⁹ Nandi, S.; Suzuki, Y. A.; Huang, J.; Yalda, D.; Pham, P.; Wu, L.; Bartley, G.; Huang, N.; Lönnerdal, B.; Plant Sci. 2002, 163, 713. [Crossref]
  » Crossref
• ³⁰ Agência Nacional de Vigilância Sanitária (Anvisa); Resolução da Diretoria Colegiada (RDC) No. 44, de 19 de setembro de 2011, Estabele o Regulamento Técnico para Fórmulas Infantis de Seguimento para Lactentes e Crianças de Primeira Infância; Diário Oficial da União (DOU), Brasília, Brazil, 2011. [Link] accessed in July 2026
  » Link
• ³¹ Bobiński, R.; Bobińska, J.; Int. J. Vitam. Nutr. Res. 2022, 92, 280. [Crossref]
  » Crossref
• ³² Mazzocchi, A.; D’Oria, V.; Cosmi, V. D.; Bettocchi, S.; Milani, G. P.; Silano, M.; Agostoni, C.; Nutrients 2018, 10, 567. [Crossref]
  » Crossref
• ³³ Korma, S. A.; Zou, X.; Ali, A. H.; Abed, S. M.; Jin, Q.; Wang, X.; Food Bioprod. Process. 2018, 107, 121. [Crossref]
  » Crossref
• ³⁴ Souza, C. O.; Leite, M. E. Q.; Lasekan, J.; Baggs, G.; Pinho, L. S.; Druzian, J. I.; Ribeiro, T. C. M.; Mattos, A. P.; Menezes-Filho, J. A.; Costa-Ribeiro, H.; Lipids Health Dis. 2017, 16, 78. [Crossref]
  » Crossref
• ³⁵ Purkiewicz, A.; Pietrzak-Fiećko, R.; Nutrients 2024, 16, 2411. [Crossref]
  » Crossref
• ³⁶ Alves, E. S.; Bruni, A. R. S.; Teixeira, L. G.; Plasencia, P.; Santamaria-Echart, A.; Santos, O. O.; Madrona, G. S.; Visentainer, J. V.; Barreiro, M. F.; ACS Food Sci. Technol. 2025, 5, 4310. [Crossref]
  » Crossref
• ³⁷ Wu, D.; Zhang, L.; Zhang, Y.; Shi, J.; Tan, C. P.; Zheng, Z.; Liu, Y.; Foods 2023, 12, 600. [Crossref]
  » Crossref
• ³⁸ Giuffrida, F.; Fleith, M.; Goyer, A.; Samuel, T. M.; Elmelegy Masserey, I.; Fontannaz, P.; Cruz-Hernandez, C.; Thakkar, S. K.; Monnard, C.; de Castro, C. A.; Lavalle, L.; Rakza, T.; Agosti, M.; Al-Jashi, I.; Pereira, A. B.; Costeira, M. J.; Marchini, G.; Vanpee, M.; Stiris, T.; Stoicescu, S.; Silva, M. G.; Picaud, J.-C.; Martinez-Costa, C.; Domellöf, M.; Billeaud, C.; Eur. J. Nutr. 2022, 61, 2167. [Crossref]
  » Crossref
• ³⁹ Koletzko, B.; Rodriguez-Palmero, M.; Demmelmair, H.; Fidler, N.; Jensen, R.; Sauerwald, T.; Early Hum. Dev. 2001, 65, S3. [Crossref]
  » Crossref
• ⁴⁰ Mendonça, M. A.; Araújo, W. M. C.; Borgo, L. A.; Alencar, E. R.; PLoS One 2017, 12, e0177812. [Crossref]
  » Crossref
• ⁴¹ Purkiewicz, A.; Pietrzak-Fiećko, R.; Molecules 2024, 29, 2044. [Crossref]
  » Crossref
• ⁴² Devaraj, S.; Giuffrida, F.; Hartweg, M.; Estorninos, E. M.; Buluran, K. B.; Lawenko, R. B.; Thakkar, S. K.; Samuel, T. M.; Prostaglandins, Leukotrienes Essent. Fatty Acids 2023, 190, 102543. [Crossref]
  » Crossref
• ⁴³ Al-Beltagi, M.; Saeed, N. K.; Bediwy, A. S.; Elbeltagi, R.; Hamza‎, M. B.; World J. Clin. Pediatr. 2024, 13, 88783. [Crossref]
  » Crossref
• ⁴⁴ Suwitri, N. P. E.; Int. J. Health Sci. 2018, 2, 21. [Crossref]
  » Crossref
• ⁴⁵ Brenna, J. T.; Akomo, P.; Bahwere, P.; Berkley, J. A.; Calder, P. C.; Jones, K. D.; Liu, L.; Manary, M.; Trehan, I.; Briend, A.; BMC Med. 2015, 13, 117. [Crossref]
  » Crossref
• ⁴⁶ Patel, A.; Desai, S. S.; Mane, V. K.; Enman, J.; Rova, U.; Christakopoulos, P.; Matsakas, L.; Trends Food Sci. Technol. 2022, 120, 140. [Crossref]
  » Crossref
• ⁴⁷ Crawford, M. A.; Wang, Y.; Forsyth, S.; Brenna, J. T.; Prostaglandins, Leukotrienes Essent. Fatty Acids 2015, 102, 1. [Crossref]
  » Crossref
• ⁴⁸ Wang, A.; Yuan, C.; Wu, L.; Int. J. Food Sci. Technol. 2025, 60, vvaf001. [Crossref]
  » Crossref
• ⁴⁹ Chichlowski, M.; Bokulich, N.; Harris, C. L.; Wampler, J. L.; Li, F.; Berseth, C. L.; Rudolph, C.; Wu, S. S.; Curr. Dev. Nutr. 2021, 5, nzab027. [Crossref]
  » Crossref
• ⁵⁰ Chamorro, R.; Bascuñán, K. A.; Barrera, C.; Sandoval, J.; Puigrredon, C.; Valenzuela, R.; Int. J. Environ. Res. Public Health 2022, 19, 1930. [Crossref]
  » Crossref
• ⁵¹ Falcão, M. C.; Braspen J. 2020, 35, 294. [Crossref]
  » Crossref
• ⁵² Bermúdez, M. A.; Pereira, L.; Fraile, C.; Valerio, L.; Balboa, M. A.; Balsinde, J.; Cells 2022, 11, 2146. [Crossref]
  » Crossref
• ⁵³ Cruz, M. M.; Simão, J. J.; Sá, R. D. C. C.; Farias, T. S. M.; Silva, V. S.; Abdala, F.; Antraco, V. J.; Armelin-Correa, L.; Alonso-Vale, M. I. C.; Front. Endocrinol. 2020, 11, 537061. [Crossref]
  » Crossref
• ⁵⁴ McNamara, J. P.; J. Nutr. Biochem. 1995, 6, 120. [Crossref]
  » Crossref

Edited by

• Editor handled this article:
  César R. T. Tarley (Associate)

Publication Dates

• Publication in this collection
  10 Aug 2026
• Date of issue
  2026

History

• Received
  13 Feb 2026
• Accepted
  24 July 2026