Effects of black soldier fly replacing fish meal on growth performance, serum parameters, and intestinal microbiota of weaned piglets

Partial or complete replacement of fish meal with black soldier fly in piglet diets had no adverse effects on growth performance or health status, and may improve immune responses and antioxidant functions by modulating the intestinal microbiota composition.

Abstract

This study evaluated the effects of replacing fish meal (FM) with black soldier fly (BSF) at different levels on growth performance, nutrient utilization, serum parameters, intestinal microbiota, and microbial metabolites in weaned piglets. A total of 180 weaned piglets (28 days old) were randomly assigned to one of five dietary treatments (n = 6 pens/treatment; 6 pigs/pen): BSFF0 (basal diet), BSFF25 (25% FM replaced by BSF), BSFF50 (50% FM replacement), BSFF75 (75% FM replacement), and BSFF100 (100% FM replacement). Partial or complete replacement of FM with BSF had no adverse effects on the growth performance or diarrhea incidence of piglets. The apparent total tract digestibility of crude protein and ether extract increased linearly with increasing BSF levels (P < 0.05). Serum concentrations of immunoglobulin A, interleukin-10, high-density lipoprotein, Lys, and Asp increased linearly with increasing BSF levels, while blood urea nitrogen content decreased both linearly and quadratically (P < 0.05). Microbial analysis revealed that BSF inclusion modulated the intestinal microbiota, which was characterized by increased abundances of Lactobacillus and Bifidobacterium and a decreased abundance of Streptococcus (P < 0.05). Furthermore, concentrations of fecal valerate (P = 0.071) and acetate (P = 0.070) tended to increase quadratically with increasing BSF levels. The concentrations of biogenic amines (tryptamine, putrescine, cadaverine, and spermidine) decreased linearly with increasing BSF levels (P < 0.05). In conclusion, these findings demonstrate that replacing 25% to 100% of FM with BSF maintained growth performance while enhancing immune status and regulating microbial metabolites in weaned piglets.

Introduction

The growing global population has increased the demand for feed resources in the swine industry (Kim et al., 2024). Fish meal (FM), a critical protein source in piglet nutrition, faces supply constraints due to overfishing, which drives up feed costs (Zhang et al., 2022). Consequently, identifying cost-effective alternative protein sources to reduce dependence on FM has become imperative. Insects are characterized by their high nutrient density (protein, lipids, vitamins, and minerals), potential for upcycling organic waste, and superior feed conversion efficiency (Li et al., 2023a). Among these, the black soldier fly (BSF) has emerged as a promising candidate (Tansil et al., 2023; Christensen et al., 2023). It offers a protein content comparable to FM, a complementary amino acid (AA) profile, and a higher lipid content, making it a viable substitute for FM in piglet diets (Hender et al., 2021; Lu et al., 2022).

Currently, most studies focus on the effects of replacing FM with BSF in fish. These studies have reported no negative effects, with BSF replacement improving microbial composition and mucosal immune responses in fish (Renna et al., 2017; Li et al., 2019b). Chitin and lauric acid derived from BSF further contribute to host health (van Huis, 2013). In weaned piglets, it has been demonstrated that full-fat BSF as an FM replacement changed immune status, intestinal morphology, and intestinal microbiota composition (Yu et al., 2020a; 2020b; Crosbie et al., 2021). Additionally, replacing FM with defatted or hydrolyzed BSF has been reported to not adversely affect growth performance and nutrient digestibility of piglets (Chang et al., 2024; Lee et al., 2024). Song et al. (2024) found that full-fat BSF or defatted BSF replacing soy protein concentrate and FM did not negatively affect growth performance or intestinal health of nursery pigs. Most studies have not focused on the dose-response relationship and mechanism of BSF as a substitute for FM. However, previous research has shown that high dietary levels of BSF might negatively affect piglet performance (Liu et al., 2023). Although the nutritional composition of BSF varies significantly with the rearing substrate, developmental stage, and processing methods, the characteristics of most BSF products used in animal experiments are often inadequately documented (Hopkins et al., 2021; Son et al., 2023; Oh et al., 2024). Therefore, this study systematically evaluated the effects of varying dietary levels of food waste-reared BSF on growth performance, nutrient utilization, immune responses, antioxidant functions, serum parameters, AA metabolism, and intestinal microbiota in weaned piglets.

Materials and Methods

All experimental procedures involving animals were approved by the Institutional Animal Care and Use Committee of China Agricultural University (Aw52104202-1-5). BSF larvae were supplied by Bennong Agricultural Technology Co., Ltd (Zhengzhou, Henan, China). BSF eggs were incubated for 3 days under controlled conditions. The resulting 3-day-old larvae were immediately transferred to containers pre-loaded with food waste collected from local restaurants. After 15 days of rearing, the larvae were separated from the residue, collected, blanched at 70 °C for 1 min, microwave-dried at 75 °C for 20 min, and finally ground into a meal using a mill grinder. The nutritional profile of the BSF meal is detailed in Supplementary Table S1, as previously described (Liu et al., 2023).

Experimental design, animals, and sampling

A total of 180 crossbred (Duroc × Landrace × Yorkshire) 28-day-old weaned piglets (body weight 8.91 ± 1.58 kg) were randomly assigned to 5 dietary treatments with 6 replicates per treatment and 6 piglets per replicate (3 barrows and 3 gilts). The piglets were housed in pens equipped with plastic slatted flooring, nipple drinkers, and self-feeders. The environment was controlled (temperature: 25 °C-26°C; relative humidity: 60%-70%). The dietary treatments were: BSFF0 (a corn-soybean basal diet with FM), BSFF25, BSFF50, BSFF75, and BSFF100 (where 25%, 50%, 75%, and 100% of the protein provided by FM was replaced by BSF, respectively). The 28-day trial was divided into an early (day 1–14) and a late (day 15–28) phase. All diets were formulated to meet the Nutrient Requirements of Swine in China (Li et al., 2020) (Table 1). Minor differences between the analyzed and calculated nutritional values of the experimental diets were observed but fell within the expected range. These differences might be related to the inherent variability of raw materials, such as the degradation of sensitive AA and protein modifications, as well as discrepancies between the actual composition of the ingredients and the values used in the formulation.

Table 1.

Ingredients and chemical composition of experiment diets (%, as-fed basis)

Item	Percentage of BSF replacing FM, %
	0 (BSFF0)	25 (BSFF25)	50 (BSFF50)	75 (BSFF75)	100 (BSFF100)
Corn	64.02	62.74	59.16	56.73	53.42
Soybean meal, 43% CP	19.40	19.20	18.30	19.10	18.50
Extruded full-fat soybean	5.50	5.50	5.50	4.00	4.00
FM	5.00	3.75	2.50	1.25	–
BSF	0.00	2.16	4.30	6.48	8.60
Wheat bran	0.00	1.10	4.80	7.10	10.10
Whey powder	2.50	2.50	2.50	2.50	2.50
Soybean oil	0.40	–	–	–	–
Dicalcium phosphate	1.10	1.12	1.12	1.10	1.08
Limestone	0.50	0.30	0.12	–	–
NaCl	0.25	0.25	0.25	0.25	0.25
L·Lys·HCl (98%)	0.33	0.36	0.40	0.42	0.45
DL·Met	0.10	0.11	0.13	0.14	0.16
L·Thr	0.11	0.12	0.13	0.14	0.15
L·Trp	0.04	0.04	0.04	0.04	0.04
Chromic oxide	0.25	0.25	0.25	0.25	0.25
Premix1	0.50	0.50	0.50	0.50	0.50
Nutrient levels2
Calculated values
Metabolizable energy, kcal/kg	3310	3310	3310	3310	3310
Available phosphorus	0.37	0.37	0.37	0.37	0.37
SID Lys	1.22	1.22	1.22	1.22	1.22
SID Met	0.42	0.42	0.42	0.42	0.43
SID Thr	0.72	0.72	0.72	0.72	0.72
SID Trp	0.21	0.21	0.21	0.21	0.21
Analyzed values
Gross energy, MJ/kg	16.21	16.33	16.50	16.56	16.76
Crude proteinChitin-corr  3	19.30	18.24	18.33	18.79	19.09
Calcium	0.74	0.74	0.74	0.76	0.82
Ether extract	3.66	4.25	4.55	4.75	5.61
Lys	1.19	1.19	1.19	1.19	1.20
Met	0.35	0.38	0.47	0.41	0.42
Thr	0.80	0.79	0.78	0.78	0.81
Trp	0.23	0.23	0.23	0.24	0.28

¹Premix provided the following per kg of complete diet for growing pigs: Vitamin A, 12,000 IU; Vitamin D₃, 2,500 IU; Vitamin E, 30 IU; Vitamin K₃, 3.0 mg; Vitamin B₁₂, 0.012 mg; riboflavin, 4.0 mg; pantothenic acid, 15.0 mg; niacin, 30.0 mg; choline chloride, 400.0 mg; folacin, 0.7 mg; Vitamin B₁, 1.5 mg; Vitamin B₆, 3 mg; Mn (MnSO₄·H₂O), 40.0 mg; Fe (FeSO₄·H₂O), 75.0 mg; Zn (ZnSO₄·H₂O), 75.0 mg; Cu (CuSO₄·5H₂O), 100.0 mg; I (KI), 0.3 mg; Se (Na₂SeO₃), 0.3 mg.

²The values were determined except metabolizable energy, available phosphorus, standardized ileal digestible (SID) Lys, SID Met, SID Thr, and SID Trp.

³CP_Chitin-corr(%) = [(total N−Chitin N) × 6.25].

From days 26 to 28, fresh fecal samples were collected from each pen. A subsample of approximately 300 g per pen was immediately stored at −20°C. These samples were later pooled by pen, dried at 65 °C for 72 h, and ground to pass through a 0.42 mm sieve for chemical analysis. Additionally, fresh rectal fecal samples were collected directly from piglets, snap-frozen in liquid nitrogen, and stored at −80°C for subsequent analysis of short-chain fatty acids (SCFA) and biogenic amines, and for 16S rRNA sequencing. On day 28, blood samples were drawn from the anterior vena cava of one piglet per pen (selected based on the average body weight of the pen). The samples were centrifuged at 3,000 × g for 15 min to collect the supernatant.

Growth performance, diarrhea incidence, and nutrient digestibility

Growth performance was evaluated by calculating average daily gain (ADG), average daily feed intake (ADFI), and the gain-to-feed ratio based on body weight and feed consumption. Throughout the trial, individual examinations were conducted on the anus of piglets at 09:00 and 17:00 daily, and the number of pigs experiencing diarrhea was recorded to obtain the diarrhea incidence. Fecal consistency was assessed using a standardized scoring system: 1 = firm; 2 = mildly soft; 3 = soft, partially formed; 4 = loose, semi-liquid; 5 = watery, mucus-like (Pierce et al., 2005). Diarrhea was defined as a score of 4 or 5 persisting for two consecutive days. The diarrhea incidence was calculated as follows:

$$\text{Diarrhea} \text{incidence} \left(\%\right)=\text{Total} \text{number} \text{of} \text{diarrheal} \text{piglets}/\left(\text{Number} \text{of} \text{piglets}\times \text{Number} \text{of} \text{days} \text{observed}\right)\times 100$$

The apparent total tract digestibility (ATTD) of nutrients was determined using the external indicator method with chromium trioxide. Dry matter (DM), crude protein (CP), organic matter (OM), and ether extract (EE) in the feed and feces were analyzed according to the AOAC (2007). Gross energy (GE) was measured using a Parr 6300 adiabatic oxygen bomb calorimeter (Moline, IL, USA), and chromium (Cr) concentration was determined by a Z-500 atomic absorption spectrometry (Hitachi, Tokyo, Japan) Furthermore, since a portion of the nitrogen (N) in BSF is bound to indigestible chitin, the CP content of the BSF was calculated using a N-to-protein conversion factor of 4.67, as previously established for BSF (Janssen et al., 2017). For the CP content in BSF-containing diets, the chitin-derived N was first calculated and then subtracted from the total dietary N, after which the CP content was determined using the standard conversion factor of 6.25 (Beller et al., 2024). The ATTD of nutrients was calculated as follows:

$$\text{ATTD} \left(\%\right)=\left[1-\left({\text{Cr}}_{\text{diets}}\times { \text{nutrients}}_{\text{feces}}\right)/\left({\text{Cr}}_{\text{feces}}\times { \text{nutrients}}_{\text{diets}}\right)\right]\times 100\%.$$

Serum AA levels

Serum-free AA concentrations were determined using an AA analyzer (S-433D, Sykam, Germany). Prior to analysis, serum samples were mixed with 1.5 mL of 4% sulfosalicylic acid (Sedgwick et al., 1991). After incubation on ice for 20 min and neutralization with 175 µL of LiOH, the mixture was centrifuged, and the supernatant was filtered. The filtered supernatant was then analyzed as follows: Amino acids were first separated on an ion-exchange chromatography column, then derivatized post-column with ninhydrin reagent, and finally detected by fluorescence (Ferré et al., 2019).

Serum parameters

Commercially available assay kits from Laibo Tairui Technology Development Co., Ltd (Beijing, China) were used to measure the concentrations of the following serum parameters: blood urea nitrogen (BUN), total cholesterol, triacylglycerol, high-density lipoprotein (HDL), low-density lipoprotein, immunoglobulin A (IgA), IgG, IgM, tumor necrosis factor-α (TNF-α), interferon-γ, interleukin-1β (IL-1β), IL-6, IL-10, superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GSH-Px), total antioxidant capacity (T-AOC), and malonaldehyde (MDA).

Fecal microbiome analysis

Fecal microbial genomic DNA was extracted from the samples using the E.Z.N.A.^® bacterial DNA kits (Omega Bio-Tek, Norcross, GA, USA) according to the manufacturer’s protocols. The DNA concentration and purity were determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). The V3-V4 region of the bacterial 16S rRNA gene was amplified with primers 338F (5′-ACTCCTACGGGAGGCAGCAG-3′) and 806R (5′-GGACTA CHVGGGTWTCTAAT-3'). Negative controls were included in the amplification to monitor for potential contamination. The PCR reaction mixture (20 µL) contained 10 μL of 2× Pro Taq master mix, 0.8 μL of each primer (5 μM), 10 ng of template DNA, and ddH2O to a final volume of 20 µL. The amplification protocol was as follows: initial denaturation at 95 °C for 3 min; 30 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s, and extension at 72 °C for 45 s; followed by a final extension at 72 °C for 10 min. The amplification products were purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, USA). Purified amplicons were sequenced on an Illumina MiSeq platform (Illumina, USA) for paired-end sequencing (2 × 300 bp) by Shanghai Majorbio Bio-pharm Technology Co., Ltd After quality control, an average of 56,488 high-quality sequences per sample were obtained for downstream analysis. The raw sequencing reads were demultiplexed and quality-filtered using fastp (v0.19.6) and merged using FLASH (v1.2.11) to generate high-quality clean tags, following the QIIME (v1.17) pipeline. Operational taxonomic units (OTU) were clustered with a 97% similarity cutoff using the UPARSE algorithm (v7.1). Taxonomy was assigned to each 16S rRNA gene sequence using the RDP Classifier (v2.13) against the SILVA 16S rRNA database (v138.2) with a confidence threshold of 0.7.

Fecal SCFA concentrations

The concentrations of SCFA in feces were determined by ion chromatography using a previously described method (Liu et al., 2017). Briefly, approximately 0.5 g of feces was homogenized in 8 mL of ultrapure water. The mixture was treated in an ultrasonic bath for 30 min and then centrifuged at 8,000 × g and 4 °C for 10 min. The supernatant was diluted 50-fold with ultrapure water and filtered through a 0.22-μm membrane. The filtered samples were analyzed using a high-performance ion chromatography system (ICS-3000 Dionex, USA) equipped with an AS11 analytical column (4 × 250 mm) and an AG11-HC guard column (4 × 50 mm). The gradient conditions were as follows: 0–5 min, 0.8–1.5 mM; 5–10 min, 1.5–2.5 mM; and 10–15 min, 2.5 mM. The flow rate was maintained at 1.0 mL/min.

Biogenic amine concentrations

The concentrations of biogenic amines in feces were determined by high-performance liquid chromatography according to the methods of Markku (2002) and Li et al. (2019a) with modifications. Briefly, fecal samples were homogenized with the internal standard solution (40 mg/L heptylamine in 0.4 M perchloric acid) and an additional 2 mL of 0.4 M perchloric acid. After shaking for 10 min and centrifugation, the supernatant was derivatized by adding sodium hydroxide, saturated sodium bicarbonate, and dansyl chloride solution. The mixture was incubated at 40 °C for 45 min in the dark. The reaction was terminated with ammonia, followed by a further 30-min incubation at 40 °C. After the addition of acetonitrile and centrifugation, the supernatant was filtered through a 0.22-μm membrane for high-performance liquid chromatography analysis. Chromatographic separation was achieved on an Agilent 1200 system equipped with a C18 column (4.6 mm × 250 mm; 5 μm; Agilent Technologies, USA) using a gradient elution program with ammonium acetate and acetonitrile. The flow rate, column temperature, and detection wavelength were set at 1.0 mL/min, 30 °C, and 254 nm, respectively.

Statistical analysis

Data were analyzed using SAS 9.4 software. The pen was considered the experimental unit for growth performance and nutrient digestibility data, while the individual pig served as the experimental unit for all other parameters. The normality of data distribution was assessed using the Shapiro-Wilk test. Polynomial contrasts were applied to evaluate the linear and quadratic responses to increasing dietary levels of BSF replacing FM. Diarrhea incidence was analyzed using a chi-square (χ²) test. Differences were considered statistically significant at P < 0.05, and 0.05 ≤ P < 0.10 was considered a trend.

Microbiome data were processed and analyzed in R with the vegan and phyloseq packages. The α-diversity indices were calculated using MOTHUR (v1.30.2). The β-diversity was assessed using principal coordinate analysis (PCoA) based on Bray-Curtis distances, and the significance of clustering among dietary groups was tested using analysis of similarity (ANOSIM). Differentially abundant bacterial taxa were identified using the Linear Discriminant Analysis Effect Size (LEfSe) algorithm, with a linear discriminant analysis (LDA) score threshold of 2.0 and a P-value threshold of 0.05.

Results

Growth performance

No significant differences were observed in ADG, ADFI, the gain-to-feed ratio, or diarrhea incidence among the dietary groups during any phase of the experiment (Table 2; Figure 1). However, from days 15 to 28, the gain-to-feed ratio tended to increase in both a linear (P = 0.080) and quadratic (P = 0.072) manner to increasing dietary BSF levels.

Table 2.

Effects of BSF replacing FM on growth performance of weaned piglets

Item	Percentage of BSF replacing FM, %					SEM	P values
	0	25	50	75	100		Linear	Quadratic
Days 1–14
ADG, g	333	369	346	354	342	14.04	0.961	0.457
ADFI, g	564	586	612	586	569	22.85	0.920	0.268
Gain-to-feed ratio	0.59	0.63	0.57	0.60	0.60	0.03	0.948	0.769
Days 15–28
ADG, g	494	512	535	512	538	19.91	0.457	0.826
ADFI, g	955	1057	1011	1003	962	43.91	0.866	0.403
Gain-to-feed ratio	0.52	0.49	0.53	0.51	0.56	0.02	0.080	0.072
Days 1–28
ADG, g	411	438	437	430	438	12.86	0.601	0.668
ADFI, g	753	813	804	792	763	27.16	0.989	0.327
Gain-to-feed ratio	0.55	0.54	0.54	0.54	0.57	0.02	0.423	0.277

Data are presented as the mean of 6 replicates per treatment (n=6)..

Figure 1.

Effects of BSF replacing FM on the diarrhea incidence of weaned piglets. The results are presented as the mean and standard error of the mean. (A) Day 1–14, (B) day 14–28, and (C) day 1–28 of the experiment.

n = 6. BSFF0, control diet. BSFF25, BSF replacing 25% FM in diets. BSFF50, BSF replacing 50% FM in diets. BSFF75, BSF replacing 75% FM in diets. BSFF100, BSF replacing 100% FM in diets.

The ATTD of nutrients

As shown in Table 3, the ATTD of CP increased linearly (P < 0.01) and tended to increase quadratically (P = 0.064) with increasing dietary BSF levels. Similarly, the ATTD of EE increased linearly (P < 0.01) and tended to increase quadratically (P = 0.068). However, the ATTD of DM, OM, and GE were not affected by the dietary treatments.

Table 3.

Effects of BSF replacing FM on apparent total tract digestibility (%) of weaned piglets (%)

Item	Percentage of BSF replacing FM, %					SEM	P values
	0	25	50	75	100		Linear	Quadratic
Dry matter	85.29	83.59	84.02	84.60	83.72	0.465	0.164	0.311
Organic matter	86.05	84.68	85.28	85.99	85.10	0.421	0.668	0.497
Crude protein	77.42	71.99	77.85	79.95	79.06	0.726	<0.010	0.064
Gross energy	84.95	83.54	83.97	84.23	84.04	0.462	0.444	0.204
Ether extract	61.89	59.36	62.92	70.78	71.48	1.494	<0.010	0.068

Data are presented as the mean of 6 replicates per treatment (n=6).

Serum AA levels

As the dietary proportion of BSF replacing FM increased, serum levels of Lys (P = 0.021) and Asp (P < 0.01) increased linearly, and serum Ser tended to increase linearly (P = 0.062) (Table 4).

Table 4.

Effects of BSF replacing FM on serum amino acid levels (μmol/L) of weaned piglets

Item	Percentage of BSF replacing FM, %					SEM	P values
	0	25	50	75	100		Linear	Quadratic
Indispensable amino acids
Arg	223	275	254	327	217	35	0.702	0.068
His	49	49	55	56	52	6	0.447	0.762
Ile	133	143	136	140	126	14	0.685	0.494
Leu	264	292	272	276	263	22	0.790	0.465
Lys	386	477	458	625	542	39	0.021	0.452
Met	64	111	83	63	83	13	0.807	0.343
Phe	94	108	93	90	87	7	0.132	0.387
Thr	280	322	282	311	255	23	0.479	0.213
Trp	56	71	64	69	71	6	0.121	0.486
Val	262	303	286	349	259	24	0.654	0.102
Dispensable amino acids
Ala	878	747	809	822	877	51	0.786	0.319
Asn	77	81	84	82	73	5	0.772	0.303
Asp	48	56	66	60	67	4	< 0.010	0.191
Cys	2.5	2.6	3.8	2.8	2.1	0.7	0.353	0.522
Gln	761	741	891	836	810	66	0.396	0.418
Glu	223	263	271	277	301	39	0.154	0.804
Gly	1168	1223	1116	1072	1118	64	0.202	0.852
Pro	577	568	526	496	550	73	0.567	0.570
Ser	203	221	191	180	163	18	0.062	0.482
Tyr	160	202	181	186	174	14	0.782	0.113

Data are presented as the mean of 6 replicates per treatment (n=6).

Serum parameters

Serum HDL content increased linearly (P < 0.01) and tended to increase quadratically (P = 0.052) with increasing BSF levels (Table 5). The BUN content decreased in both linear and quadratic manners (P < 0.01). Serum concentrations of IgA and IL-10 increased linearly with the level of FM replacement by BSF (P < 0.01). No significant differences were observed in antioxidant parameters among the treatment groups.

Table 5.

Effects of BSF replacing FM on serum biochemical and antioxidant parameters of weaned piglets

Item	Percentage of BSF replacing FM, %					SEM	P values
	0	25	50	75	100		Linear	Quadratic
Total cholesterol, mmol/L	2.14	2.22	2.22	2.16	2.18	0.19	0.978	0.814
Triacylglycerol, mmol/L	0.78	0.70	0.62	0.59	0.54	0.08	0.052	0.723
HDL, mmol/L	0.81	0.87	0.92	0.97	1.26	0.06	< 0.010	0.052
LDL, mmol/L	1.56	1.56	1.55	1.52	1.56	0.11	0.927	0.886
BUN, mmol/L	3.85	2.38	2.33	2.12	2.08	0.25	< 0.010	< 0.010
Ig A, µg/mL	13.79	13.91	14.66	14.80	14.68	0.23	< 0.010	0.188
Ig G, mg/mL	7.91	8.05	8.34	7.94	8.07	0.16	0.629	0.221
Ig M, µg/mL	7.32	7.19	7.32	7.14	7.52	0.25	0.723	0.542
IL-1β, ng/L	88.90	92.71	89.68	90.93	90.98	2.31	0.729	0.692
IL-6, ng/L	43.06	40.67	42.69	40.44	41.50	1.07	0.353	0.537
IL-10, ng/L	17.94	19.15	19.48	20.46	19.88	0.56	< 0.010	0.172
TNF-α, ng/L	65.49	62.65	67.75	62.92	67.09	1.89	0.548	0.556
Interferon-γ, pg/mL	147.79	146.22	152.36	148.02	149.13	5.13	0.772	0.766
MDA, nmol/mL	1.51	1.50	1.55	1.53	1.52	0.07	0.849	0.741
Total antioxidant capacity	9.50	9.47	9.32	9.25	9.49	0.49	0.873	0.720
Superoxide dismutase, U/mL	41.49	39.63	38.50	38.39	36.71	2.44	0.143	0.865
Glutathione peroxidase, µmol/L	11.42	10.84	11.21	11.23	10.27	0.53	0.244	0.557
Catalase, U/mL	15.28	16.50	15.02	14.89	15.64	1.03	0.791	0.912

Data are presented as the mean of 6 replicates per treatment (n=6).

Fecal microbiome

The effects of replacing FM with BSF on the fecal microbiota composition of piglets were analyzed. No significant differences in α-diversity indices were observed among the groups (Table 6). As shown in Figure 2A, a total of 602 OTUs were shared across all treatment groups, while only 93 OTUs were unique to individual groups. Further analysis of the fecal microbial community was conducted (Figure 2B–E). Firmicutes was the predominant phylum across all groups. At the family level, the predominant taxa were Lachnospiraceae and Ruminococcaceae. The LEfSe analysis revealed that, compared with the BSFF0 group, the BSFF25 group showed lower abundances of Scillospiraceae, Monolobaceae, Streptococcus, Acetylomaculum, Monolobus, and Fournierella, and higher abundances of Eubacterium-hallii-group and Bifidobacterium (P < 0.05). In comparison to the BSFF0 group, the BSFF50 group showed a decrease in the abundances of Streptococcus and Micrococcacea, and an increase in the abundances of Mogibacterium, Butyrivbrio, Rothia, and Intestibacter (P < 0.05). Compared with BSFF0, the BSFF75 group exhibited decreased abundances of Streptococcaceae, Rikenellaceae, Mogibacterium, and Catenisphaera, and increased abundances of Carnobacteriaceae and Trichococcus (P < 0.05). In comparison to the BSFF0 group, the BSFF100 group showed lower abundances of Clostridiaceae, Streptococcaceae, Oscillospiraceae, Christensenellaceae, Streptococcus, and Colidextribacte, and higher abundances of Lactobacillaceae and Lactobacillus (P < 0.05).

Table 6.

Effects of BSF replacing FM on alpha diversity indexes of fecal microbial community in weaned piglets

Item	Percentage of BSF replacing FM, %					SEM	P value
	0	25	50	75	100
Sobs	543.40	483.20	472.60	518.80	455.40	61.97	0.622
Shannon	4.13	4.02	3.94	3.84	3.69	0.44	0.639
Simpson	0.05	0.05	0.07	0.08	0.08	0.03	0.585
Ace	636.90	576.20	563.10	604.70	542.60	56.23	0.724
Chao	641.30	587.80	560.20	622.00	551.60	60.24	0.762
Coverage	0.996	0.997	0.997	0.997	0.997	< 0.01	0.913

Data are presented as the mean of 6 replicates per treatment (n=6).

Figure 2.

Effects of BSF replacing FM on microbial composition of feces in piglets (A) OTU Venn diagram. (B) Principal coordinate analysis (R =  0.10, P  =  0.11). (C, D) Barplot analysis of microbial community composition at phylum and family levels. (E) Heatmap analysis of microbial community composition at genus level. (F–I) The LEfSe analysis of microbial community composition from phylum to genus level.

n = 6. BSFF0, control diet. BSFF25, BSF replacing 25% FM in diets. BSFF50, BSF replacing 50% FM in diets. BSFF75, BSF replacing 75% FM in diets. BSFF100, BSF replacing 100% FM in diets.

Fecal metabolites

The effects of BSF on fecal metabolites are shown in Table 7. The concentrations of fecal valerate (P = 0.071) and acetate (P = 0.070) tended to increase in a quadratic manner with increasing BSF levels. Furthermore, the contents of several fecal biogenic amines, including tryptamine (P < 0.01), putrescine (P = 0.024), cadaverine (P = 0.028), tyramine (P = 0.095), spermidine (P = 0.044), and total biogenic amines (P < 0.01), decreased linearly as the dietary proportion of BSF increased.

Table 7.

Effects of BSF replacing FM on contents of fecal SCFA and biogenic amines (mg/kg) of weaned piglets

Item	Percentage of BSF replacing FM, %					SEM	P values
	0	25	50	75	100		Linear	Quadratic
Lactic acid	276.98	165.22	248.90	168.35	159.05	48.06	0.113	0.812
Acetic acid	3384.00	5572.36	4826.42	5046.83	4989.62	513.77	0.104	0.070
Propionic acid	1653.00	2613.51	2087.03	2193.53	2519.08	327.71	0.208	0.597
Formic acid	99.88	129.15	129.42	213.01	126.33	51.27	0.398	0.442
Isobutyric acid	106.29	170.95	97.07	113.81	116.64	34.78	0.721	0.777
Butyric acid	952.03	1651.13	1211.51	1026.73	1280.36	187.44	0.956	0.337
Isovaleric acid	296.26	601.81	382.48	1003.28	393.39	290.71	0.512	0.353
Valeric acid	180.61	487.88	282.83	270.51	270.17	58.79	0.829	0.071
Total SCFA	6949.05	11391.99	9265.66	10036.06	9854.64	1062.32	0.192	0.120
Tryptamine	8.62	3.91	3.50	2.85	1.36	1.45	< 0.010	0.254
Phenethylamine	2.09	1.61	3.21	1.24	1.67	0.60	0.532	0.448
Putrescine	32.48	15.10	13.70	16.59	8.78	5.48	0.024	0.263
Cadaverine	190.50	79.58	49.07	125.90	47.24	33.43	0.028	0.159
Histamine	19.65	14.58	14.72	12.54	7.59	4.12	0.103	0.902
Tyramine	24.34	18.56	15.66	15.60	11.66	4.78	0.095	0.729
Spermidine	27.22	24.89	20.29	28.69	13.75	3.29	0.044	0.337
Spermine	2.76	1.01	0.89	1.63	0.74	0.60	0.113	0.301
Total biogenic amines	307.66	159.23	121.05	205.03	92.79	41.54	< 0.010	0.202

Data are presented as the mean of 6 replicates per treatment (n=6).

Discussion

As previously reported, the nutritional composition of BSF varies significantly with the rearing substrate, developmental stage, and processing methods (Hopkins et al., 2021; Son et al., 2023; Oh et al., 2024). In the current study, the CP content of the full-fat BSF was 27.41%, which aligns with the reported range of 21.6% to 43.9% (Alafif et al., 2025). As no defatting treatment was applied, the CP level was lower than that of FM (63.38%). Consistent with previous findings, the BSF in this study was rich in AA (Schiavone et al., 2017, 2018). The levels of Trp and Tyr were higher than those in FM, while the contents of other AA were similar to or slightly lower. The EE content of the BSF was 31.34%, consistent with the reported EE range of 29.4% to 51.5% for full-fat BSF and higher than that of FM (4.46%) (Shumo et al., 2019; Rawski et al., 2020; de Souza Vilela et al., 2021). Notably, the calcium content of the BSF (5.29%) was higher than that of FM (1.55%), suggesting its potential as a mineral source in swine diets. The chitin level of the BSF was 3.95%, falling within the previously reported range of 3.87% to 7.21% (Lu et al., 2022). Variations in chitin content may be related to the developmental stage of BSF. Although chitin is considered difficult to digest, it has been reported to benefit animal immune function (Lu et al., 2022). Additionally, piglets can secrete chitin-degrading enzymes, and the hydrolysis products of chitin are beneficial to their health (Hu et al., 2018; Xu et al., 2020; Kawasaki et al., 2021). Overall, the nutritional composition of BSF supports its value as a functional ingredient in animal feed.

In the current study, replacing FM with different levels of BSF did not negatively affect growth performance. Moreover, the feed efficiency increased both linearly and quadratically during the late phase, indicating that complete replacement of FM with BSF (included at 8.60%) is feasible in piglet diets. This finding is consistent with several previous studies. Chia et al. (2019) observed no significant alterations in growth performance when FM was replaced with BSF in diets for growing pigs. Chang et al. (2024) reported no negative effects when defatted or hydrolyzed BSF was used to replace 30 g/kg of FM in weaned piglet diets. Our results are in agreement with earlier studies in which a small amount (about 4%-10%) of BSF as a protein source replacing FM had no adverse effects on piglet growth (Spranghers et al., 2018). Furthermore, Yu et al. (2020b) reported that dietary inclusion of full-fat BSF as an FM replacement did not negatively affect the growth performance of weaned piglets, with 2% BSF supplementation even increasing ADG and reducin feed conversion ratio from days 1 to 14. Lee et al. (2024) demonstrated that defatted BSF supplementation throughout the experimental period improved ADG and decreased feed conversion ratio compared with FM in piglets. It has also been reported that increasing supplemental BSF oil from 0 to 6% linearly improved the feed efficiency of nursery pigs, the benefits that may be related to the high concentration of lauric acid in BSF oil (van Heugten et al., 2022). The discrepancies among previous studies could be attributed to differences in life stages of pigs, rearing substrates, processing methods, and inclusion levels of BSF. The improvement in the feed efficiency observed in the late phase of this study may be attributed to the positive effects of lauric acid and chitin present in BSF. Lauric acid has been reported to exhibit antibacterial and anti-inflammatory properties in the small intestine, and chitin has been demonstrated to improve intestinal nutrient absorption and increase intestinal IgA levels (Heo et al., 2013; Fioretti, 2017; Spranghers et al., 2017). The fact that the improvement was observed only during the late phase might be explained by the reported increase in the secretion of chitin-degrading enzymes as piglets mature (Kawasaki et al., 2021). Consequently, we speculate that as the piglets matured, their digestive systems became more developed and capable of adapting to and utilizing BSF, leading to more efficient nutrient absorption in the late stage.

In this study, we observed that the ATTD of CP and EE increased linearly and quadratically with increasing dietary BSF levels. In contrast to our results, Yu et al. (2020b) observed that adding 4% BSF to replace FM reduced the CP and EE digestibility of piglets. Ao and Kim (2019) found that partially replacing FM with dried mealworm reduced the CP digestibility of weaned piglets. Spranghers et al. (2018) found that BSF did not significantly affect nutrient digestibility of weaned pigs. Consistent with our results, Chang et al. (2023) reported that hydrolyzed BSF showed higher CP digestibility than both FM and defatted BSF in broilers. Yordanova et al. (2025) demonstrated that BSF replacing soybean meal linearly increased fat digestibility in pigs, possibly due to the high-fat content of BSF. Kierończyk et al. (2022) found that fat from BSF resulted in higher EE digestibility than soybean oil, which may be attributed to enhanced lipase activity. The medium-chain fatty acids like lauric acid, which are abundant in BSF, are more readily absorbed and can be directly utilized by enterocytes as an energy source (Aslam et al., 2025). The reasons for these different findings on nutrient digestibility remain unclear, partly due to the variable characteristics of insect proteins (Son et al., 2023; Oh et al., 2024). Oh et al. (2024) reported in an in vitro study for pigs that hot-air drying, microwave drying, and freeze-drying of full-fat BSF resulted in relatively high nutrient digestibility compared to the infrared drying method. Chitin is considered a primary contributor to the observed variations in nutrient digestibility. It has been reported that higher chitin levels may negatively affect nutrient digestibility (De Marco et al., 2015). At the same time, pigs have the ability to secrete chitin-degrading enzymes (Kawasaki et al., 2021). The hydrolysis products of chitin, chito-oligosaccharides, have been shown to improve nutrient digestibility in piglets (Chen et al., 2009; Marono et al., 2015). The results of the current study indicate that with a dietary inclusion of 8.60% BSF (containing 0.34% chitin), pigs might secrete sufficient chitin-degrading enzymes to facilitate its digestion. It should be noted that chitin fundamentally remains an indigestible structural fiber (Finke, 2007). This indigestible residue contributes to fecal output, thereby exerting a diluting effect on the ATTD of OM and GE (Bovera et al., 2018; Coutinho et al., 2021). Kalmendal et al. (2011) reported that the inclusion of high-fiber sunflower cake linearly increased the ileal digestibility of fat and protein while simultaneously decreasing the digestibility of DM and energy in broiler chickens. The authors attributed this to the dual role of insoluble fiber: it stimulates digestive activity to enhance nutrient breakdown, while the fiber itself dilutes the overall digestibility coefficients. Other studies in Muscovy ducks have also observed that BSF inclusion can affect CP and EE digestibility without altering the ATTD of DM and OM (Gariglio et al., 2019). The stable OM and GE digestibility observed in our study likely reflects a balance in which the significant improvements in protein and fat availability are offset by the inherent limitations of the indigestible fiber.

Serum AA levels are crucial indicators for evaluating protein and AA metabolism. Liu et al. (2020) demonstrated that insect protein can improve the intestinal AA transport capacity of piglets, suggesting superior AA utilization compared to traditional raw materials. Metzler-Zebeli et al. (2012) found that AA deficiency in sows, caused by long-term feeding of low-protein diets, decreased serum Lys levels, ultimately impairing fetal development and maternal health. Currently, knowledge regarding the effects of BSF on serum AA levels is limited. Nevertheless, the current results indicate that BSF linearly increased serum Lys and Asp levels. According to previous reports, such an increase in serum AA levels may benefit the growth and health status of pigs (Wang et al., 2021b).

Replacing FM with BSF in the diets linearly increased the levels of serum IgA and IL-10 in piglets. Chitin from BSF has been shown to stimulate the immune system and enhance immune function (Harikrishnan et al., 2012). Xu et al. (2020) found that chitosan nanoparticles increased plasma IgA and IgG concentrations in a dose-dependent manner (0, 100, 200 and 400 mg/kg) and inhibited the increase in IL-6 and IL-1β levels in pigs challenged with Escherichia coli lipopolysaccharide. Hu et al. (2018) reported that dietary supplementation with 50 mg/kg of low-molecular-weight chitosan significantly decreased the expression of IL-1β and TNF-α in the jejunal mucosa of piglets. These findings suggest that even low doses of chitin can enhance immune responses. Consistent with our studies, Yu et al. (2019) demonstrated that BSF at the level of 4% (providing 0.90% chitin) increased IL-10 expression levels in finishing pigs. Yu et al. (2020a) found that replacing FM with 1%, 2%, and 4% BSF (providing 0.25%, 0.56%, and 0.87% chitin, respectively) quadratically decreased ileal TNF-α expression levels, and linearly and quadratically increased IgA contents in weaned piglets. In the current study, the 0.3% dietary chitin from BSF, while a low inclusion, is likely a significant contributor to the improved immune parameters we observed. Its effect is probably part of a synergistic interaction with other bioactive components in whole BSF, such as lauric acid and antimicrobial peptides, which collectively enhance immune function. BSF is rich in lauric acid, which and its metabolites exhibit anti-inflammatory and antibacterial properties in the small intestine (Crosbie et al., 2020). Additionally, BSF can produce various antimicrobial peptides that exert antibacterial effects by opening ion channels on bacterial membranes and disrupting membrane structures (Zhang et al., 2024).

Replacing FM with BSF did not affect serum T-AOC, GSH-Px, SOD or MDA levels in piglets. The absence of significant differences in these antioxidant parameters indicates that BSF has an effect on antioxidant functions comparable to that of FM. The current knowledge about the antioxidant functions of insect proteins in pigs is limited. However, the positive effects of insect proteins on antioxidant functions in poultry and fish have been proven (Cullere et al., 2016; Dabbou et al., 2018; Hender et al., 2021). Adding BSF improved serum and liver oxidoreductase activities in juvenile shrimps and fish (Aragão et al., 2022). Adding BSF instead of soybean meal to the diets of broiler chickens increased serum GSH-Px activity, possibly due to tocopherols present in insect protein (Dabbou et al., 2018). Furthermore, the active ingredients such as medium-chain fatty acids and antimicrobial peptides contained in insect proteins can regulate the antioxidant status of the body (Aragão et al., 2022). These findings suggest that the antioxidant potential of BSF deserves further study in swine.

All measured serum biochemical parameters remained within normal physiological ranges, indicating that dietary BSF inclusion did not adversely affect piglet health. Compared with conventional high-quality protein sources, such as FM, BSF induced specific changes in serum biochemical profiles without compromising animal health, supporting its viability as a sustainable protein alternative (Yu et al., 2020b). In particular, BUN, the main end product of AA metabolism, decreased in piglets fed BSF-based diets. This reduction suggests improved nitrogen utilization efficiency, as lower BUN levels are indicative of more balanced AA absorption and metabolism (Yu et al., 2017). In addition, BSF replacement increased serum HDL levels. It has been reported that chitin may modulate lipid metabolism through electrostatic interactions. Its positively charged amino groups can sequester bile salts and free fatty acids in the intestinal lumen, promoting their fecal excretion. This reduces lipid absorption and depletes the bile acid pool, prompting the liver to synthesize new bile acids from hepatic cholesterol. The resultant increase in cholesterol consumption for bile acid synthesis creates a metabolic demand that can suppress de novo lipogenesis (Koide, 1998; Marono et al., 2017). Moreover, lauric acid, which is abundant in BSF, is known to exert regulatory effects on liver function and lipid metabolism, further contributing to the observed improvements in serum lipid profiles.

In this study, replacing FM with BSF promoted the colonization of beneficial bacteria, specifically by increasing the abundances of Lactobacillus and Bifidobacterium. As previously reported, BSF also increased Lactobacillus and Bifidobacterium abundances in piglets, finishing pigs, and laying hens (Cullere et al., 2016; Yu et al., 2019). Lactobacillus exerts positive effects on intestinal physiological function by inhibiting pathogenic bacteria through the production of hydrogen peroxide and other antibacterial factors (Zhang et al., 2018). Bifidobacterium is also of great significance to intestinal health due to its ability to produce acetate and lactate (Gibson et al., 2004). BSF significantly reduced Streptococcus abundance in the current study. Streptococcus can induce diarrhea and gastrointestinal malignancies in pigs, and Streptococcus abundance in fecal samples from colorectal cancer patients significantly increased compared to healthy controls (Boleij and Tjalsma, 2013). The reduction in Streptococcus observed in this study is consistent with the findings of Yu et al. (2019) in finishing pigs. Although the mechanisms by which BSF modulates the gut microbiota require further elucidation, several bioactive components are likely involved. Studies have shown that moderate amounts of chitin, a key component of BSF, can improve gut microbiota composition in mice (Neyrinck et al., 2012). The derivative of chitin, chitosan, has been shown to increase the abundances of Lactobacillus and Bifidobacterium in mice (Zheng et al., 2018). Therefore, the chitin in BSF could be a key factor contributing to the observed modulatory effects on the gut microbiota, such as the increased abundance of Bifidobacterium in the BSFF25 group. Lauric acid exerts selective inhibitory effects on gram-positive bacteria (Skrivanová et al., 2005). In addition, antimicrobial peptides derived from BSF demonstrate broad-spectrum bactericidal activity against both gram-positive and gram-negative pathogens (Yi et al., 2014).

Microbial metabolites in the gut, such as ammonia, SCFA, and biogenic amines, serve as important indicators of intestinal health. Undigested carbohydrates that reach the large intestine are fermented by gut microbiota to produce SCFA (Nyachoti et al., 2006). In the present study, the replacement of FM with BSF led to a quadratic increase in concentrations of valerate and acetate. Previous studies have shown that BSF as a substitute for soybean meal can increase the concentrations of intestinal SCFA in finishing pigs and laying hens. The increase may be related to the increase in SCFA producing microorganisms such as Bifidobacterium (Borrelli et al., 2017). Furthermore, a major component of BSF, chitin, passes undigested to the large intestine and is fermented by gut microbiota. Therefore, the modulation of fermentable substrates in the large intestine is likely a key factor influencing SCFA production. SCFA could benefit gut health through distinct pathways. Acetate has been reported to inhibit pro-inflammatory cytokine production and to support barrier function by enhancing intestinal epithelial cell survival (Liu et al., 2021). Additionally, valerate has been reported to enhance the expression of tight junction proteins in weaned piglets infected with enterotoxigenic Escherichia coli F18 (Kovanda et al., 2024). The increased concentrations of these SCFA observed in this study suggest that replacing FM with BSF may contribute to improved intestinal health in piglets.

The undigested proteins and other nitrogen-containing compounds (e.g., AA) can be fermented by gut microorganisms to produce biogenic amines (Davila et al., 2013). In this study, replacing FM with BSF linearly decreased the concentrations of tryptamine, putrescine, cadaverine, tyramine, spermidine, and total biogenic amines. This reduction suggests a decline in microbial AA decarboxylation (Davila et al., 2013). We hypothesize that this reduction in biogenic amines may be attributed to the increased supply of fermentable substrates. The chitin from BSF is an indigestible polysaccharide that reaches the large intestine and serves as a fermentable substrate for specific microbiota (Mei et al., 2022). Additionally, the necessary increase in wheat bran increased the levels of dietary non-starch polysaccharides. The increased availability of these indigestible fibers could suppress protein fermentation, thereby reducing the production of biogenic amines (Diether and Willing, 2019). The BSF diet decreased the abundance of Streptococcus. Previous studies have reported a positive association between total amine levels and Streptococcus abundance, indicating that this genus can produce amines (Mu et al., 2016; Yu et al., 2018). It has been reported that high levels of tryptamine and tyramine might be harmful to gut health (Shalaby, 1996; Yang et al., 2016). Tyramine has been reported to disrupt tight junction proteins in vitro. These proteins are essential for maintaining the integrity and regulating the permeability of the intestinal barrier (Dala-Paula et al., 2023). Putrescine and cadaverine have been reported to be associated with the occurrence of diarrhea in weaned piglets (Teti et al., 2002). Spermidine may inhibit intestinal mucosal barrier function (Li et al., 2023b). Certain Streptococcus species and their amine metabolites can induce intestinal inflammation, whereas SCFA and SCFA producing bacteria typically downregulate pro-inflammatory responses (Rist et al., 2013; Li et al., 2018; Wang et al., 2019). Therefore, the decrease in pro-inflammatory cytokine contents may be related to the increase in SCFA producing bacteria and SCFA levels, as well as the decrease in potential pathogens and their related compounds.

Conclusion

In conclusion, the partial or complete replacement of FM with food waste-reared BSF did not negatively impact the growth performance or nutrient digestibility of piglets. Furthermore, BSF as a substitute for FM might increase SCFA levels and decrease biogenic amine levels by modulating microbiota composition. These changes in fermentation products of gut microbiota likely contributed to the improved immune responses in piglets.

Abbreviations:

AA

amino acid

ADFI

average daily feed intake

ADG

average daily gain

ATTD

, apparent total tract digestibility

BSF

, black soldier fly

BUN

, blood urea nitrogen

CP

, crude protein

Cr

, chromium

DM

, dry matter

EE

, ether extract

FM

, fish meal

GE

, gross energy

GSH-Px

, glutathione peroxidase

HDL

, high-density lipoprotein

IgA

, immunoglobulin A

IgG

, immunoglobulin G

IgM

, immunoglobulin M

IL-1β

, interleukin-1β

IL-10

, interleukin-10

IL-6

, interleukin-6

MDA

, malonaldehyde

OM

, organic matter

OTU

, operational taxonomic units

SCFA

, short-chain fatty acid

SOD

, superoxide dismutase

T-AOC

, total antioxidant capacity

TNF-α

, tumor necrosis factor-α

Author contributions

Sujie Liu (Investigation, Writing—original draft, Writing—review & editing), Jian Wang (Conceptualization, Data curation, Formal analysis), Shuang Dong (Conceptualization, Validation), and Yonggai Duan (Conceptualization, Methodology), Yongxi Ma (Funding acquisition, Project administration, Supervision, Writing—review & editing)

Supplementary Data

Supplementary data are available at Journal of Animal Science online.

Conflict of interest statement:

The authors have no conflicts of interest to declare.