Novel approaches to avoiding spoilage of perennial ryegrass silage: minimizing fungal and bacterial spoilage

Using this information, the most promising additives were chosen and tested in a silage model system. <br><br>212 microbes were isolated from NI PRG silage to create a representative silage culturome. A novel screen was developed to investigate how 17 potential additives (14 of which are novel to PRG silage) qualitatively affected the growth and acid production of 168 of these isolates (those which grew on nutrient agar and were non-filamentous).

Six additives were identified (abietic acid, epigallocatechin gallate (EGCG), lysozyme, Fisher gum rosin (FGR), manganese sulfate and Tween 80) which increased the proportion of growing acid producing species by up to 38.9% compared to non-acid producing species in the screen. Interestingly sodium benzoate and potassium sorbate which are currently used in silage additives appeared to inhibit acid producing species more than spoilage microbes. <br><br>These six additives, alongside sodium benzoate to act as a comparison for what is currently available, were then trialled to see how they quantitatively affected the growth and acid production of 12 representative acid producing silage isolates as well as three model pathogens (Candida albicans, Listeria monocytogenes and Salmonella typhimurium).

All the additives had different effects across the microbes with differences even appearing between strains of the same species. Overall, it appeared that additives that inhibited the growth of pathogenic microbes inhibited acid producing species as well. Sodium benzoate was a prime example of this, inhibiting the pathogenic species but also severely inhibiting acid producing species including Lactobacillus plantarum, a common silage lactic acid bacterium (LAB).

Some of the non-LAB acid producing isolates identified from the NI silage culturome were genera not normally associated with silage acidification. To investigate if some these could potentially replace or support LAB during silage acidification, three isolates (Citrobacter sp. AH04, Klebsiella pneumoniae AA02 and Proteus sp.

AH10) were tested for growth and acid production with the addition of sodium benzoate. FGR and abietic acid were also tested with these isolates as they were similarly shown to inhibit the growth of L. plantarum. Abietic acid and FGR were much less inhibitory to the growth and acid production of the three isolates compared to sodium benzoate.

Although all three silage isolates could produce lactic acid, the most productive, Citrobacter AH04, only produced 0.51g/l of lactic acid in 24 hours compared to L. plantarum which produced 6.81g/l meaning that although they could theoretically replace LAB during ensiling it would likely come at the cost of a much slower fermentation. <br><br>Fungi are also an important source of silage spoilage where air is present such as at the edges of the silage clamps, surface of bales or during feed out.

Therefore, it is important that the potential additives do not increase the growth of the fungi and ideally, they should inhibit it. The effect the 17 potential additives from Chapter 2 had on the growth of 8 fungi isolated from NI silage in Chapter 2 was examined through in vitro culture alongside Penicillium roqueforti DSMZ 11974, a common Irish silage spoilage fungus. Out of the 17 additives tested 15 were inhibitory to the growth of fungi.

The most inhibitory was EDTA which inhibited 8 of the 9 the fungi tested by over 70%. The currently used additives potassium sorbate and sodium benzoate, alongside abietic acid, dehydroabietic acid, both gum rosins, curcumin and sodium acetate were effective at inhibiting many of the fungi. This indicated they should be beneficial for the aerobic stability of silage (a silage’s ability to remain unspoiled after oxygen is introduced). <br><br>From the 17 additives tested seven were chosen to be taken forward to a vacuum bag model to examine how they affected the PRG silage through fermentation and during aerobic stability testing.

Abietic acid, EGCG, FGR and manganese sulfate, which have never been trialled as PRG silage additives, were selected for their ability to increase the proportion of acid producing species in the aerobic microbial screen in Chapter 2 and their antifungal abilities. Lysozyme was also selected for its ability to increase the proportion of acid producing species in the aerobic and anaerobic microbial screen in Chapter 2 and its antifungal abilities; Tween 80 for its ability to enhance the acid production of L. plantarum ATCC 11974 and its previous positive results in corn, orchard grass and alfalfa silage; and sodium benzoate to act as a benchmark for currently used silage additive components and to investigate its effect in acid production during silage fermentation.

No additive had a significant effect on the aerobic stability of the silage tested. Only sodium benzoate and lysozyme had any significant effects on the fermentation of the silage that was carried through to the final timepoint. Lysozyme increased the final lactic acid concentration by 8.8% but it did not increase the overall rate of pH decline.

This combined with it not lowering the concentration of any of the molecules associated with spoilage microbe growth suggest that lysozyme is not effective enough at improving fermentation to warrant the cost of its addition. Sodium benzoate reduced the amount of butyric (-100%) and succinic acid (-30.7%) present indicating its use may reduce silage spoilage. These two additives did have the largest effect on the proportion of acid producing species in the culturome screen indicating the screen may help to identify additives which can have a significant impact on silage fermentation.<br><br>Interestingly none of the additives which inhibited the growth and acid production of L. plantarum in Chapter 3 (abietic acid, FGR or sodium benzoate) significantly reduced the amount of lactic acid in the silage according to HPLC analysis.

To investigate if non-LAB may be contributing to the acidification of silage treated with sodium benzoate and abietic acid the number of anaerobic microbes, LAB and yeast and fungi were tracked throughout a silage fermentation with these treatments. Abietic acid was found to have no effect on the silage culturome, possibly due to it being metabolised early in fermentation. Sodium benzoate however significantly reduced the number of LAB from day 2-100.

Previous studies had shown that sodium benzoate could reduce the number of LAB at the end of fermentation but did not examine the effect during fermentation. By assessing these factors throughout fermentation, the effect of sodium benzoate on the culturome was shown to be almost immediate and reduced the number of LAB throughout. Sodium benzoate treated silage had a lower pH on day 0,2,4 and 16 however it only reduced the amount of lactic acid present on day 2 and 56.

This combined with the reduced LAB numbers potentially suggests that non-LAB microbes are producing a significant amount of lactic acid during this fermentation, contributing more to acidification than previously thought. The sodium benzoate treated silage had a slowed fermentation but even with this still had lower amounts of ethanol and succinic acid present on day 56 and fewer yeasts and fungi throughout the fermentation.

This indicated that although it slowed fermentation the antimicrobial ability of sodium benzoate still led to an overall better, less spoiled silage. <br> While abietic acid, FGR, and sodium benzoate were shown to have potentially negative effects on silage fermentation, they were also very inhibitory to spoilage fungi. Therefore, they were trialled as post-fermentation additives for improving aerobic stability, which would avoid their negative effects on fermentation.

Abietic acid had no effect up to day 5 of aerobic exposure but by day 10 abietic acid-treated silage was significantly more spoiled than the control silage. Fisher gum rosin appeared to reduce fungal growth by day 5 indicating it might aid in short term silage storage. Sodium benzoate had an immediate effect reducing the number of culturable fungi present, this continued to day 5 indicating it may help reduce spoilage during short term aerobic spoilage but by day 10 there was no significant difference between the sodium benzoate treated silage and the control.<br><br>In conclusion, it was found an in vitro microbial screen using pH indicating agar may be able to predict how an additive will affect silage fermentation, but further validation is required.

Lysozyme was found to improve PRG silage fermentation, but its benefits are unlikely to warrant its costs. FGR was identified as a potential novel post fermentation silage additive to slow aerobic spoilage. Sodium benzoate appears to inhibit silage fermentation and LAB growth, but its natural antimicrobial properties mean silage spoilage is still reduced with its use.

It also appears that non-LAB microbes may be able to contribute more to silage acidification than previously thought. These alternative microbes may provide novel drivers of fermentation for biotechnology and agriculture in changing climates.