Biomagic was used on the farms to improve the biodegradation of waste products and decrease environmental contamination. It is a granulated enzyme complex containing nitrogen, phosphorus, and potassium in a typical 1:1:1 ratio and produced through the fermentation of useful ingredients extracted from fruits such as orange, lemon, and papaya. It was treated with 1,000 kg feces/1 L (0.1%), once a day. This study involved a crossbreed Yorkshire and Landracer growing pigs with an initial average body weight of 60 kg (± 3.5 kg), housed at 3 different farms (n = 8) for 4 months. Seven points were measured from a 2 m point on the entrance wall of the pig pen to a 2 m point on the opposite wall, divided into 7 equal parts, and the average value was calculated. All animal care and experimental procedures were carried out according to internationally accepted guidelines on the use of laboratory animals (Institutional Animal Care and Use Committee permit no. 2022-0334) and the protocols were approved by the Animal Care Committee of the College of Veterinary Medicine, Kyungpook National University.

Before and after processing with the Biomagic, a MultiRAE gas meter (RAE Systems, Honeywell International, Inc., USA) was used to measure the representative harmful gases ammonia (NH₃) and hydrogen sulfide (H₂S), as well as the carbon dioxide (CO₂) and oxygen (O₂) concentrations. While odor intensity and offensiveness were scaled as described previously [20].

The blood samples were collected via venepuncture from the jugular vein. The collected blood was transferred to K3-EDTA-treated tubes and stored on ice until analysis. A complete blood cell count, which includes white blood cells, red blood cells, hemoglobin, hematocrit, and mean corpuscular volume, was measured using an automatic hematology analyzer, Hemavet. After blood collection, 1 mL of each sample was dispensed into a 1.5 mL microtube and centrifuged at 3,000 rpm for 20 minutes to obtain plasma for evaluating the cytokine profile using a standard commercial kit.

A 99-m sterilized anaerobic diluent containing 1 g of fresh test animal feces was homogenized, followed by a 10-fold anaerobic dilution process (NaCl, 0.9 g; KCl, 0.042 g; CaCl₂ 6H₂O, 0.048 g; NaHCO₃, 0.02 g; Cysteine HCl, 0.2 g; and 400 mL of distilled water were used to select the lactic acid strains for livestock probiotics). The selection and fractionation media used are listed in Table 1 and were gradually diluted in a solid medium to determine the number of bacteria in the dilution solution. On the solid medium that was plated with incremental dilutions, aerobic or anaerobic cultures were performed. Using a GasPak system (Becton, Dickinson and Company, USA), the anaerobic culture was performed for 48 hours at 37°C (refer to Table 1).

Microbial DNA was extracted and purified from fecal samples using a QIAamp DNA Stool Mini Kit (QIAGEN Inc., Germany) to analyze the microbiota from the stool samples, as previously described [21]. The DNA samples’ quality control requirement was met with an A260/A280 ratio of ≥ 1.8 and a DNA concentration of ≥ 15 ng/µL. Following the DNA extraction from the farm’s environmental feces, microbiome analysis was conducted at Kyungpook National University’s Sequencing Centre. The 16S rRNA gene sequences obtained from the fecal samples were used as the basis for microbiome profiling.

Viral RNA and DNA were extracted and purified from stool samples following the manufacturer’s protocol using a Patho Gene-spin RNA/DNA Extraction kit (iNTRON Biotechnology). To ensure the quality control, requirements, the extracted viral RNA and DNA were required to have an A260/A280 ratio of ≥ 1.8 and an RNA/DNA concentration of ≥ 15 ng/µL.

The extracted viral RNA and DNA were amplified to determine the presence of transmissible gastroenteritis virus (TGEV), porcine reproductive and respiratory syndrome (PRRS), and PCV2 using commercial kits for reverse transcription-poly­merase chain reaction (RT-PCR, TGEV and PRRS) and PCR reaction (PCV2) according to the manufacturer’s protocols (LiliF TGEV, PRRS-RT-PCR, and PCV2-PCR kit, iNtRON Bio­technology). The product was run on a 1.5% electrophoresis gel to assess the presence of viral RNA and DNA.

Acquired data were subjected to a one-way analysis of variance and post hoc Dunnett’s test (SAS Institute Inc., USA) to determine the statistical significance of the observed differences. The information is presented as the mean ± standard deviation (SD). Statistical significance was defined as p < 0.05.

MultiRAE gas meter was used to measure the representative harmful gases ammonia (NH₃) and hydrogen sulfide (H₂S) that cause odor complaints, as well as the carbon dioxide (CO₂) and oxygen (O₂) concentrations, which are environmental indicators. A significant reduction in ammonia concentrations was observed across the 3 farms after the application of Biomagic in August, as analyzed using repeated measures analysis of covariance (ANOVA) (Fig. 2). Farm 1 showed a significant decline in NH₃ levels from August (mean ± SD, 12 ± 0.5 ppb) to December (mean ± SD, 8 ± 0.5 ppb); F (1, 4) = 22.36, p < 0.01. Farm 2’s NH₃ levels halved from August (mean ± SD, 13 ± 0.7 ppb) to December (mean ± SD, 6 ± 0.5 ppb), F (1, 4) = 35.07, p < 0.001, indicating a highly significant effect. Farm 3 presented the most rapid reduction with levels decreasing from 22 ppb in August to 7 ppb in September, F (1, 4) = 45.85, p < 0.001, and stabilizing thereafter, demonstrating a substantial immediate response to Biomagic treatment. In contrast to NH₃, ozone levels remained relatively constant across all 3 farms throughout the observed period. Farm 1's ozone concentration averaged around 20 ppb, while Farms 2 and 3 maintained a consistent level of approximately 25 ppb. The stability in ozone concentrations indicates that either farm operations do not significantly influence ozone levels or external factors affecting ozone concentrations are uniform across these locations. Carbon dioxide emissions demonstrated an upward trend in 2 out of the 3 farms. Farm 1 shows an increase from about 2,000 ppb to nearly 3,500 ppb by December. Similarly, CO₂ emissions in Farm 2 rose from approximately 1,000 ppb to 2,000 ppb within the same period. In contrast, Farm 3 displayed a peak in CO₂ emissions in September at about 2,500 ppb, followed by a reduction to below 2000 ppb by December. This variation suggests that factors such as seasonal energy use, biomass decomposition, or heating requirements could influence CO₂ levels. Overall, our results demonstrated that after treatment with the enzyme complex, ammonia concentration was reduced and remained > 25 ppm, ozone concentration remained constant, CO₂ remained > 3,000 ppm, and H₂S was not detected. The odor intensity and offensiveness were characterized by the non-objective scaling method (Dynamic Olfactometry method) as reported previously [22]. Odor intensity and offensiveness were characterized as faint (2) and fairly offensive (3) from very strong (5) and strongly offensive (4), respectively (Table 2 and Fig. 2).

The blood biochemical indices (red blood cell count, white blood cell count, platelet count, hemoglobin concentration, hematocrit, and mean red blood cell volume) were measured to monitor changes in blood biochemical indices and normal homeostasis of the animals. Our results indicated no significant changes in the blood biochemical indices observed after using Biomagic (p < 0.05) (Supplementary Fig. S1). Additionally, we measured the cytokines to confirm whether it affect the pig's immune system. The blood inflammatory cytokines (interferon [IFN]-γ, interleukin [IL]-1β, IL-6, and tumor necrosis factor [TNF]-α) and immunoreactive cytokines (IL-2, IL-4, and IL-10; Supplementary Table 1) in the plasma were measured. Our results indicated that IFN-γ, IL-1β, IL-6, and TNF-α were below detectable concentrations, and no significant changes were observed in the levels of IL-2 and IL-4 (Fig. 3, Supplementary Table 1). However, IL-10 showed an increasing trend starting from the third month (November). Overall, the results indicate that the treatment with the Biomagic-Enzyme complex does not significantly affect the immune response of pigs.

The changes in the distribution of harmful and beneficial bacteria in the feces of pigs after treatment with the complex enzyme were compared from month to month. The number of bacteria in the feces was determined using selective media. Escherichia coli, Staphylococcus aureus, Clostridium, and Salmonella exhibited a decrease and then maintained a steady level, but Lactobacillus, a disease-defense bacterium, was consistently maintained at a certain level. Alternatively, Bifidobacteria showed a decrease in conjunction with E. coli, but at a certain level, they exhibited an increasing trend of recovery (Fig. 4). Changes in the number of harmful bacteria on selective media showed that after 4 months of complex enzyme treatment, the number of coliforms (MAC media) decreased in all 3 farms. S. aureus (mannitol salt [MS] medium) and Clostridium spp. (tryptone sulfite neomycin [TSN] medium) showed a decreasing trend. Salmonella spp. (SS medium) was not present in the 3 farms tested. Unusually, the level of Lactobacillus spp. (MRS medium), a beneficial bacterium often used as a probiotic remained steady. However, Bifidobacteria spp. (TOS medium) showed a decline along with E. coli, but then recovered to a steady level. These results show that the complex enzyme treatment suppressed the harmful enterobacteria while maintaining the beneficial lactobacilli.

Following the complex enzyme treatment, a marked decrease in E. coli populations was observed in all 3 farms. Initially, high counts were significantly reduced, indicating that the treatment effectively suppressed this group of harmful enterobacteria. The consistent reduction across different environmental conditions and farm management practices suggests a robust effect of the enzyme treatment against E. coli. In contrast to E. coli, the levels of Lactobacillus spp., a beneficial bacterium widely recognized for its probiotic properties, remained stable over the same period. This stability was observed despite the varying conditions across the farms, which underscores the selective efficacy of the enzyme treatment in preserving beneficial bacterial populations while targeting harmful ones.

We detected 21 bacterial phyla, 390 genera, and 148 families. The top 10 of these were selected from each level for statistical evaluation of the microbiota and are shown in the representative figure (Fig. 5A). The microbiota results revealed that the relative abundance of Spirochetes, Proteobacteria, and Campylobacteria appears to be different at the phylum level. The relative abundance of the families Lachnospiraceae, Spirochaetaceae, and Bacteroidaceae appear to be different at the family level. An alpha diversity analysis was performed to determine the differences in diversity (Shannon), richness (Chao1), and evenness (Evenness) based on time (Fig. 5B). The Shannon index is a measure of diversity that considers both the richness (the number of different species) and the evenness (how evenly the individuals are distributed across those species) of a community. A higher Shannon value indicates a more diverse community, where both species richness and evenness are higher. Chao1 is an estimate of species richness. It is used to predict the total number of species in a community, including those that are rare or undetected in the sampling process. This index is especially useful for identifying whether an observed sample has captured the full diversity of a community or if there may be more species that were not sampled. Evenness measures how evenly individuals are distributed among the species present. It is calculated from the Shannon index but focuses solely on the distribution aspect, not on species richness. An evenness value close to 1 indicates that species are nearly equally abundant, while lower values indicate some species dominate over others. The same graph (parking group) showed a statistically significant difference in diversity and uniformity between the first month and third month compared to the initial period. However, in terms of abundance, there was a tendency to decrease slightly in the first month compared to the initial period, and then an increase in the third month; however, this was not statistically significant. In the beta diversity, grouping according to time (initial, 1 month, 3 months) and the presence or absence of the agent treatment (yes, no) was applied. While the samples were mixed as a whole, the samples corresponded to 1 month, initial, and 3 months. This data demonstrates that the samples are a little further apart than the samples that correspond to the month group. Applying the unweighted UniFrac method, the difference in the microbial community between the samples was confirmed. As a result, it can be seen that the microbial community improved without affecting the overall population, which remained statistically non-significant (Fig. 5B). When the complex enzyme was treated in 3 test farms for 3 months, it was confirmed that it was a very safe biomaterial because it did not affect the composition of intestinal flora, unlike antibiotics.

Nucleic acid was isolated from feces samples and analyzed for TGEV, PRRS, and PCV2 using commercial assay kits. In this test system, TGEV was not detected, while PRRS and PCV2 showed a positive infection. TGEV and PRRS caused significant economic losses and deadly infection while PCV2 has been reported as non-pathogenic in swine. The infection rate for PRRS was insignificant, while the infection for PCV2 was to be 40% to 50% (Fig. 6).