Safety and efficacy of a feed additive consisting of endo‐1,4‐beta‐xylanase, endo‐1,3(4)‐beta‐glucanase and endo‐1,4‐beta‐glucanase produced by Trichoderma reesei ATCC 74444 (Ronozyme® Multigrain) for use in poultry for fattening, poultry for laying and piglets (weaned) (DSM Nutritional Products)

3.1.1. Characterisation of the additive

The additive is authorised in two different formulations, a granular form (Ronozyme® MultiGrain (GT), referred to as GT) and a liquid form (Ronozyme® MultiGrain (L), referred to as L), both with a minimum enzymatic activity per gram or ml of additive of 2,700 U ⁷ xylanase, 800 U cellulase and 700 U glucanase.

As regards the manufacturing process, the enzymes contained in the additive are produced by fermentation with the production strain; ■■■■■. ⁸ The applicant declares that no antibiotics are used during the manufacturing process. ⁹

The applicant has declared changes in the composition of the additive since the last authorisation. The original composition of the additive was: ■■■■■ ¹⁰ ■■■■■ ¹¹ ■■■■■ ¹² The changes made to the composition of the additive are not expected to introduce safety concerns.

Eight batches of each form of the additive were tested for the enzyme activities. ¹³ The GT form showed per gram of product: 3,354 U xylanase (range: 2,930–3,630 U), 1,318 U cellulase (range: 1,220–1,450 U) and 1,003 U glucanase (range: 886–1,160 U). The L form showed per ml of product: 3,384 U xylanase (range: 3,290–3,420 U xylanase), 1,233 U cellulase (range: 1,170–1,410 U cellulase) and 941 U glucanase (range: 895–1,070 U glucanase). The measured enzyme activities are in compliance with the specifications set in the authorisation.

The same batches of the two forms of the additive were also analysed for chemical impurities (eight batches of the GT form, five of the L form) and microbiological contamination (eight batches of both forms).

In the GT form, mercury and cadmium values were below the limit of quantification (LOQ) of the analytical method, while levels of arsenic were < 0.3–0.51 mg/kg, and lead were < 0.5–0.92 mg/kg. In the L form, mercury, arsenic, cadmium and lead concentrations showed levels below the LOQ or LOD of the analytical methods. ¹⁴ Escherichia coli or Salmonella spp. were not detected in 25 g of product, Bacillus cereus counts were ≤ 10 CFU/g, total viable counts were ≤ 200 CFU/g, counts of coliforms were < 4 CFU/g (L form) or < 10 CFU/g (GT form), counts of yeasts and filamentous fungi were < 10 CFU/g, and counts of Enterobacteriaceae (only done in three batches of each form) were < 10 CFU/g. ¹⁵

The FEEDAP Panel considers that the microbial contamination and the amounts of the detected impurities do not raise safety concerns.

The applicant provided data on the physico‐chemical properties, stability, and homogeneity of the additive based on the new formulations.

The bulk density of the GT form is 1,070 kg/m³. The dusting potential of three batches of the GT form measured by Stauber–Heubach method showed a mean value of 3.3 mg/m³ (range: 0–10 mg/m³). Particle size distribution of the same batch of the additive tested by laser diffraction showed that the mean particle size is 539 μm and no particles below 250 μm were found. ¹⁶

Three batches of the GT form were kept in glass vials sealed with metal caps at a range of temperatures of 10–50°C for up to 24 months, or in glass vials at 40°C/60% relative humidity (RH) for 1 month. Samples kept at −18°C were used as baseline. ¹⁷ At 10–35°C after 24 months, average enzyme activities were in the range of 66–98% xylanase, 74–103% cellulase and 81–99% glucanase compared to the initial; at 40°C after 12 months, average enzyme activities were 63% xylanase, 90% cellulase and 80% glucanase compared to the initial; at 50°C after 3 months, average residual activities were 72% xylanase, 91% cellulase and 82% glucanase; and in glass vials after 1 month at 40°C/60 RH% average residual activities were 96% xylanase, 98% cellulase and 84% glucanase.

For the L form, three batches were kept in sealed glass vials at a range of temperatures of 10–50°C for up to 24 months. Samples kept at −18°C were used as baseline. ¹⁸ Compared to the baseline values, average enzyme activities were: in the range of 84–97% xylanase, 91–99% cellulase and 95–104% glucanase after 24 months at 10–25°C; 75% xylanase, 91% cellulase and 92% glucanase after 18 months at 35°; 76% xylanase, 91% cellulase and 95% glucanase after 6 months at 40°C (based on two batches); and 68% xylanase, 93% cellulase and 97% glucanase after 1 month at 50°C (based on two batches).

The stability of the GT form of the additive (three batches) was tested when mixed with a vitamin‐mineral premixture for chickens for fattening and stored in plastic bags at 25°C for 6 months. ¹⁹ No losses were observed at the end of the experimental period.

The stability of the GT form (three batches) when added at 150 mg/kg to feedingstuffs was tested in feed for chickens for fattening based on maize and soy. ²⁰ From each batch, a mash feed and two pelleted feeds (pelleting temperatures 80°C and 90°C) were manufactured and stored in plastic bags for 3 months at 25°C. Average enzyme activities after pelleting were higher than 82% in all cases. Storage for 3 months did not modify substantially the post‐pelleting enzyme activities.

The stability of the L form (three batches) was studied when sprayed onto pelleted feed for chickens for fattening at the intended inclusion level of 150 mg/kg feed. ²¹ Feed samples were kept for 3 months in paper bags at 25°C. Average enzyme activities were 97% xylanase, 88% cellulase and 98% glucanase of the initial ones.

The enzyme activities present in the additive are produced by a non‐genetically modified strain of T. reesei, originally isolated from a cotton duck shelter and subjected to further selection process by exposure to chemical/physical mutagenesis. The production strain has been deposited at the American Type Culture Collection (ATCC) with the accession number ATCC 74444. ²²

The identification of the production strain as T. reesei was confirmed by sequencing the Internal Transcribed Spacer (ITS) regions and the 18S rRNA gene and comparing the sequences with those existing in the NCBI RefSeq Targeted Loci Project and UNITE databases, respectively. ²³

The capacity of the production strain to produce substances with antimicrobial activity was not investigated. However, the potential presence of antimicrobial activity was tested in five batches of each formulation using a diffusion test. ²⁴ The reference strains used were: Staphylococcus aureus ATCC 6538, Streptococcus pyogenes ATCC 12344, Bacillus cereus ATCC 2, Bacillus circulans ATCC 4516, Escherichia coli ATCC 11229 and Serratia marcescens ATCC 14041. No antimicrobial activity was detected.

No data were provided regarding the capacity of the production strain to produce secondary metabolites. Some Trichoderma species are known to be capable of producing various mycotoxins and antifungal metabolites. T. reesei seems to be unable to produce mycotoxins (EFSA, 2007; Frisvad et al., 2018; EFSA BIOHAZ Panel, 2020) but it is known to produce peptaibols (e.g. paracelsin A, C and D (Frisvad et al., 2018)). ²⁵ Peptaibols are peptides with antimicrobial activity (primarily against Gram‐positive bacteria), which are mostly produced under stress conditions (Frisvad et al., 2018). The lack of antimicrobial activity in the additive under assessment (described above) would indicate that if peptaibols are produced under the fermentation conditions, their concentration would be of no concern.

The presence of viable cells of the production strain was investigated in triplicate in an intermediate concentrate ■■■■■ ²⁶ ■■■■■. A proper positive control was included in the analysis. A negative control was also considered (0.9% NaCl solution). The controls performed as expected. No viable cells of the production strain were found in an intermediate product representative of the final formulations.