Effect of Direct-Fed Microbials and Monensin on In vitro Fermentation of a High-Forage Diet

An in vitro experiment was conducted to determine effects of directfed microbials (DFM) on rumen fermentation of a forage-based diet in the presence and absence of monensin (MON). Treatments included DFM (0 and 50,000 cfu; primarily Lactobacillus acidophilus and Enterococcus faecium) and monensin (0 and 5 ppm) and were applied directly to the appropriate flask. Substrates (80:20 forage: concentrate) were incubated for 30 h. Measurements included serial gas pressure, and terminal CH4 production, pH, and volatile fatty acid (VFA) and NH3 concentrations. Interactions between DFM and MON were absent (P>0.10) for gas production and fermentation end products, except for a tendency for DFM to increase total VFA concentration without, but not with MON (interaction between DFM and MON; P=0.07). Addition of DFM did not affect (P>0.10) gas production or fermentative end products except for a tendency (P=0.08) for a slight increase in proportion of isovalerate. >Contrariwise, MON decreased (P<0.001) total gas and CH4 production. Similarly, MON decreased (P<0.001) total VFA and NH3 concentrations, and molar proportions of acetate and butyrate and increased (P<0.001) proportions of propionate, valerate, isobutyrate, and isovalerate. Independently, DFM and MON increased (P<0.001) end point pH, resulting in an additive effect when the two treatments were combined. Addition of MON altered fermentation in a predictable fashion and was not influenced by the addition of DFM. In contrast, fermentation characteristics were largely unaffected by DFM with two exceptions: a slight increase in total VFA in the absence of MON and a small increase in pH that appeared to occur through a different mechanism than that for MON.


Introduction
Inclusion of DFM in both receiving and finishing diets in beef cattle has been shown to improve intake, as well as rate and efficiency of gain [1][2][3].Although the mechanism(s) responsible for these improvements in performance have not been completely elucidated, there is evidence that DFM alter rumen fermentation characteristics (i.e.VFA concentrations, methane production, microbial population) [4][5][6][7].Most commonly, a mixed bacterial culture of lactic acid producing and utilizing, gram positive bacteria are used as DFM in ruminants [4].Lactic acid producing bacteria such as Lactobacillus and Streptococcus [8], and lactate utilizing bacteria such as Megasphaera elsdenii [9] has been investigated individually as well as in combination to identify their specific beneficial effects on the rumen environment.Recent research from our laboratory has shown that mixed bacterial cultures consisting primarily of Lactobacillus acidophilus and Enterococcus faecium when used in combination with a high concentrate ration, decreases in vitro gas production as well as total VFA concentration [10].Additionally, in vivo, this mixed culture DFM decreased total VFA concentration and increased molar proportions of acetate leading to an increase in pH [11].However, little is known about this mixed culture DFM (consisting primarily of Lactobacillus acidophilus and Enterococcus faecium) in combination with a forage-based diet.
Microbial populations are altered by changes in dietary composition leading to higher counts of fibrolytic bacteria with increased forage levels [12].Fibrobacter succinogenes (gram -ve), Ruminococcus flavefaciens (gram +ve) and Ruminococcus albus (gram +ve) are considered representative fibrolytic bacteria of the rumen [13,14].These shifts in microbial population have the potential to alter VFA production and subsequently pH levels in the rumen.
Ionophores are a feed-grade antibiotic used in cattle diets to enhance feed efficiency and body weight gain.Ionophores selectively inhibit the metabolism of gram-positive bacteria, which lack a protective outer membrane, and protozoa in the rumen.Monensin is a carboxylic polyether ionophore [15] that has been shown to alter ruminal fermentation by selecting against hydrogen-producing bacteria reducing the substrate for methane production [16].Moreover, monensin decreases NH 3 production through inhibition of the hyper-ammonia-producing bacteria, a small group of ruminal bacteria that are responsible for the production of most of the ammonia [17].Additionally, monensin elicits changes in VFA profile, specifically increasing propionic acid production and reducing acetic acid [18][19][20].
Monensin is widely used in receiving cattle rations as well as in some forage-based production systems [21,22].The biological benefits of DFM have led to an increase in use.However, little is known about the effects of combined DFM and monensin treatments.Monensin's ability to selectively inhibit gram-positive bacteria in combination with the gram-positive nature of DFM suggests the possibility of a direct interaction between there dietary addition.However, other possibilities for interaction between the two exist, given that both have potential to modify the microbial ecosystem in a variety of ways.The objective of this study was to determine the effect of a mixed bacterial culture, previously used by our lab, consisting primarily of Lactobacillus acidophilus and Enterococcus faecium, on in vitro fermentation and methane production from a forage substrate, with and without the addition of monensin.

Materials and Methods
All procedures were approved by the University of Kentucky Institutional Animal Care and Use Committee.

Donor animals and diet
Four ruminally cannulated steers (396-440 kg) were housed indoors in individual pens (2.4 × 2.6 m) and fed an 80:20 forage: concentrate diet (Table 1) at 1.75 × NEm.The steers were adapted to the experimental diet over a 15 d span prior to sampling of rumen contents and were fed twice daily (8am and 5pm) during the experimental period.

Treatments
All treatments were applied in vitro and were arranges as a 2 × 2 factorial.Treatments included 2 levels of DFM (0 and 50,000 cfu) and 2 levels of monensin (0 and 5 ppm).The DFM contained a mixed culture of bacteria consisting of primarily Lactobacillus acidophilus and Enterococcus faecium and also included Pediococcus acidilaticii, Lactobacillus brevis, and Lactobacillus plantarum (10-G, Vit-E-Men Co., Norforlk, NE, USA; precise proportions are proprietary).Media DFM dosage (cfu/unit of dry matter) was based on previous in vivo studies with the same DFM that have shown enhanced animal performance and alterations in rumen fermentation [23].Direct-fed Microbial treatments were prepared by dissolving 0.05 g DFM plus carrier or carrier (lactose) into 100 mL distilled water and 1.0 mL of each solution was added to the appropriate vessels.Similarly, monensin treatments were prepared by dissolving 473.2 mg monensin sodium salt (CAS 22373-78-0; MilliporeSigma, St Louis, MO) into 47.32 mL 100% ethanol and 50 µL of the monensin containing solution or ethanol was added to the appropriate vessels [24].

In vitro procedures
Ruminal contents were collected from the ventral rumen of each steer approximately 1 h after morning feeding.Contents from each of the four steers were stored in separate insulated containers for transport into the lab.Prior to processing, the entire contents from each individual animal were combined into a large insulated container.The combined ruminal contents were strained through 4 layers of cheesecloth and the resulting fluid, along with a grab sample of whole contents, were processed using an immersion blender for two minutes under a CO 2 headspace.The blended contents were strained a second time through 4 layers of cheesecloth to form the inoculum source for the experiment.In vitro gas production was determined on 4 separate days, with each run consisting of 4 vessels/treatment.One run was determined to be an outlier (detailed in description of statistical analyses), resulting in 12 replications per treatment.Buffer solution, micro-and macro-mineral solutions, and reducing solutions were prepared as described previously [25].A combination of prepared solutions (1475 mL) and 350 mL of rumen inoculum (media solution) were maintained in a 39°C water bath under a CO 2 environment until added to the 250 mL fermentation vessels.Fermentation vessels were supplied with common substrate (400 mg of donor diet (Table 1); the diet was ground with a Wiley Mill to pass through a 1-mm screen.Each fermentation vessel also received 2 mL of H 2 O (to prevent suspension of feed particles outside of liquid solution), 100 mL of media solution and appropriate amounts of each treatment solution (1mL DFM solution or lactose carrier, and 50 µL monensin solution or ethanol).Subsequently, vessels were gassed with CO 2 for 30 seconds and then fitted with remote automatic pressure transducers (Ankom RF Wireless Gas Production System, Ankom Technology, Macedon, NY).Based on preliminary tests runs using the same substrate and conditions (data not shown), vessels were incubated in a water bath at 39°C for 30 hours, to ensure plateau gas pressure was reached, and gas pressure was measured at 5-minute intervals.At the completion of the 30-h fermentation, vessels were placed into an ice bath to cease fermentation and gas samples were drawn into10mL red-topped serum vacutainer tubes for methane analysis.After gas sampling, flasks were opened, pH was immediately determined using a portable pH meter (Acorn pH 6 Meter, Oakton Instruments, Vernon Hills, IL, USA), and samples of the culture broth were collected.A 5 mL aliquot of the sample was added to 15mL Nunc screw-cap centrifuge tubes containing 0.5 mL of metaphosphoric acid (25 g/100 mL) and 0.5 mL of volatile fatty acid (VFA) internal standard (1 g/100 mL 2-ethlybutyrate) and frozen for later VFA analysis.Additionally, a 100 µL sample was combined with 3.9 mL phosphoric acid (25 mM H 3 PO 4 ) and frozen for NH 3 analysis.

Results
The addition of DFM did not affect (P>0.10) the rate or total gas production (Table 2).In contrast, MON decreased (P<0.01)both total gas production as well as rate of gas production (Table 2).
Monensin decreased (P<0.01)total VFA concentration while DFM tended to increase (P=0.06)total VFA concentration (Table 2).However, there was a tendency (P=0.07) for a DFM × MON interaction for total VFA concentration where DFM increased total VFA concentration in the absence but not in the presence of MON.There were no other DFM × MON interactions (P>0.11) for any of the other variables.
The provision of MON altered the concentration of all VFAs (Table 2).Monensin decreased (P<0.01)acetate and butyrate concentrations, and increased (P<0.01) the concentrations of propionate, isobutyrate, isovalerate and valerate.In contrast to MON, there was no effect of DFM on VFA concentrations except for a tendency to decrease (P=0.08)isovalerate concentration.Monensin decreased (P<0.01) the molar proportions of acetate and butyrate and increased (P<0.01)molar proportions of propionate, valerate, isobutyrate and isovalerate.Similar to the effect of DFM on VFA concentrations, the effect of DFM was for molar proportion of VFAs with the exception that DFM slightly decreased (P=0.08) the molar proportion of isovalerate.Additionally, MON decreased (P<0.01) the acetate to propionate ratio (4.33 vs 3.15), while DFM had no effect (P>0.10).Ammonia-N was decreased (P<0.01) by MON (Table 2).In contrast, NH 3 -N was not impacted (P=0.57) by DFM.Direct-fed microbial and MON increased (P<0.01)end point pH similarly resulting in an additive effect when the two treatments were combined (Table 2).Monensin decreased (P<0.01)total amount and percent of methane produced (Table 2), while DFM had no effect (P=0.61).

Discussion
In vitro techniques attempt to simulate the rumen environment to allow for characterization of treatment-mediated changes within the rumen.Total gas produced is generally increased as substrate disappearance increases indicating a positive relationship between extent of feedstuff degradation and in vitro gas accumulation [31].In the current experiment, DFM treatment, with a mixed bacterial culture of lactate producing DFM primarily consisting of Lactobacillus acidophilus and Enterococcus faecium, resulted in no significant changes in total gas production or rate of gas production.Monensin decreased both rate and extent of gas production.Baah et al. [32] observed a linear decrease in total gas production after 12 h of fermentation, relative to control, with provision of increasing levels of lactate producing DFM (i.e.Lactobacillus casei and Lactobacillus lactis) in a 60:40 forage: concentrate barley-silage-based diet.However, similar to the current study, no differences in total gas production were observed for 6, 24, and 48 h fermentations.While limited data is available on gas production measures with the inclusion of DFM in forage-based diets, more is known about the effect of DFM in vitro with concentrate diets.Previous research from our lab using the same inoculum DFM treatment, with high concentrate substrate, resulted in a decrease in total gas production, indicating a decrease in substrate degradation and reduced rumen fermentation [11].Additionally, Baah et al. [32] reported a linear decrease in total gas production with inclusion of previously mentioned DFM in a 10:90 forage: concentrate barley-grain-based diet.In concert with the current findings, this suggests that DFM, particularly lactate producing bacteria, effects on gas production is dependent, at least in part, on time of incubation and substrate fermented.

Sample analysis
Donor diet and in vitro substrate was analyzed for chemical content by a commercial laboratory using wet chemistry methods (Dairy One Forage Laboratory, Ithaca, NY).The concentrations of VFA in the culture broth were determined by gas chromatography (6890 Hewlett-Packard, Avondale, PA), fitted with a Supelco 25326 Nukol fused silica capillary column (15m × 0.53mm × 0.05um film thickness; Sigma/Supelco, Bellefonte, PA).To accomplish the analysis, 0.2 µL of sample was injected at 110°C with a 2:1 split.After a 1-minute hold, the temperature will then be increased at 5°C / minute to 125 ºC for 2 minutes.The inlet and injector were set at 260 °C.Samples prepared as described previously [26].Ammonia-N concentration was determined using a photometric test with enzymatic assay through Konelab analysis (Model 20XTi, Thermo Fisher Scientific, Waltham, MA) following procedures previously described by Kun and Kearny [27].Additionally, gas samples were analyzed for methane concentration by gas chromatography (6890 Hewlett-Packard, Avondale, PA), fitted with a Supelco stainless steel 40/60 carboxen 1000 packed column (5ft × 1/8in × 2.1mm).Column head pressure was set at 10psi and oven temperature set point was 125°C [28].
Hay and supplement samples were analyzed for nutrient content by a commercial laboratory (Dairy One Forage Laboratory, Ithaca, NY).

Calculations
Head space volume for each incubation flask (206 ± 4.7 mL) was determined by subtracting additions of media, substrate, and treatments from total volume (water displacement).Cumulative gas pressure readings were converted to gas volumes at standard temperature and pressure using the ideal gas law.Converted gas volumes of individual modules were quantified using the best fit model from the evaluation of ten gas production models evaluated by Pitt et al. [29], in addition to an exponential model without a lag period (V=VF(1-e -kt ) where V=gas volume at time (t) and VF=gas volume at plateau [30].All model parameters and curve fit statistics were generated using nonlinear least squares methods in MATLAB (Version R2013a, Mathworks, Natick, MA).The best fit curves were determined as those with the lowest RMSE values.The exponential model without lag, described above, was determined to be the best-fit model for gas production data and was used to calculate rate and total production of gas.

Statistical analysis
Exponential model parameters of in vitro gas production (gas volume at plateau and rate of gas production) and fermentation end products (VFA, NH 3 and methane) were analyzed using incubation flask as the experimental unit (n=12) and a model appropriate for a randomized complete block design, with blocks representing separate runs of the in vitro procedures conducted on each of 4 separate days.The data were analyzed using the GLM procedure in SAS (SAS Inst.Inc., Cary, NC).The model statement included main effects (DFM and Monensin) and their interactions (DFM × MON), as well as block.Multiple replications of treatments within each run permitted initial analysis to evaluate run × treatment interactions.This preliminary analysis revealed run × treatment interactions (P<0.01) for either DFM and/or MON treatments with all fermentative end products, rate of gas production, and end point pH values which highlighted a single run containing spurious values.This run was removed from subsequent analysis.No other run × treatment interactions were detected (P>0.10) and thus, data were analyzed using only three runs and the interaction term was removed from final statistical analysis for all response variables.Citation: Wingard  Dissimilar to DFM, there was a positive relationship between gas production and total VFA concentrations with inclusion of MON treatments, indicating a decrease in dry matter disappearance.Although limited data is available on the effect of monensin on in vitro gas production with use of forage-based inoculum, more is known about monensin effects on concentrate diets.Quinn et al. [33] reported average total gas production was 5.9% less than control with monensin treatment on in vitro fermentation of steam-flaked corn and cottonseed meal substrate.These results were similar to those reported by Callaway and Martin [34], where in vitro culture with added monensin had less total gas production than those not receiving monensin.
Although DFM did not mediate a change in gas production parameters, it did tend to increase total VFA production in the absences of MON without altering VFA proportions.Baah et al. [32] observed a linear increase in total VFA production, as compared to control, with provision of increasing levels of Lactobacillus casei and Lactobacillus lactis after 6 and 12 h of in vitro fermentations; however, a linear decrease was observed at 24 h fermentation.In the same study, linear increase in acetate to propionate ratio was observed with increasing DFM provision after 12, 24 and 48 h of fermentation.It is unknown whether similar results would have been observed in the present study given analyses were only obtained on 30 h data.In contrast, Raeth-Knight et al. [35] observed no difference in total VFA concentrations or molar proportions of individual VFA, in vivo, among mixed lactate producing and utilizing DFM treatments (DFM1 Lactobacillus acidophilus strain LA747 and Propionibacterium freudenreichii; DFM2 Lactobacillus acidophilus strains LA747and LA45 and Propionibacterium freudenreichii) and control treatments.
In contrast to DFM, monensin decreased total VFA concentration and shifted the molar proportion of acetate and propionate such that there was a decrease in the acetate-propionate ratio.Additionally, monensin decreased butyrate levels and increased isobutyrate, valerate and isovalerate levels.Treatment effects were absent for total VFA concentrations in previous research with inclusion of varying levels of monensin in forage based diets [36][37][38].Similar to the findings of the present study, Richardson et al. [18] and Ramanzin et al. [37] observed decreased proportions of acetic and butyric acids with concurrent increased proportion of propionic acid with inclusion of monensin in vitro and in vivo (respectively).
Both DFM and MON are commonly incorporated into cattle diets.However, there is a paucity of information in the literature concerning the potential interaction between DFM and MON.In addition to the main effect of both additives there was a tendency for DFM × MON interaction on total VFA concentration, while no other interactions were observed.DFM tended to increase total VFA concentration in the absence but not in the presence of MON.The ability of DFM to alter total VFA concentrations in the absence but the not presence of MON, suggests that monensin may be impacting the ruminal effects of the DFM.The basic mode of action of an ionophore is to disrupt the movement of ions (Na + , Ca 2+ , K + , H + ) across biological membranes [39].Ionophores selectively inhibit the metabolism of gram-positive bacteria, which lack a protective outer membrane, and protozoa in the rumen.Monensin's ability to inhibit lactate-producing rumen bacteria, Citation: Wingard  S. bovis and Lactobacillus species, was observed by Dennis et al. [40] in sensitivity and growth rate trials.Similar results were reported in which monensin significantly decreased the lactobacilli and enterococci counts in in vitro incubations of the crop contents of chickens not previously exposed to monensin treatments [41].This research suggests that monensin may have an adverse effect on the DFM treatments in the present study given the DFM primarily consist of Lactobacillus acidophilus and Enterococcus faecium.Additionally, Van Nevel et al. [42] reported efficiency for both total and net microbial growth were significantly depressed by addition of monensin in an in vitro study as determined by measuring the incorporation of 32 P-labeled phosphate in microbial material, which is in agreement with other data obtained in vitro as well as in vivo [19].This would suggest that monensin could either directly inhibit the DFM bacteria and/or other bacteria that may be positively influenced by the DFM but inhibited by monensin.
In the present study, monensin decreased total VFA concentration and increased end point pH levels of culture broth.These observations are inconsistent with previous in vivo studies in cattle consuming high forage diets which have shown the inclusion of monensin to have no effect on total VFA concentration or rumen pH levels [38,43].The reason for this discrepancy is unclear.In contrast, inclusion of DFM treatment resulted in increased pH of the culture broth combined with increased total VFA concentrations, particularly in the absence of monensin.Nonetheless, it has been well demonstrated across a variety of diets that there is an inverse relationship between rumen pH and total VFA concentrations [44,45].Other research reported no difference in rumen pH with inclusion of various strains and combinations of Lactobacillus acidophilus and Propionibacterium in forage-based diets [35,46].Given the scope of this experiment it is difficult to ascribe a mechanism to account for the currently observed increase in both pH and total VFA concentrations.Inclusion of the current DFM has resulted in increased levels total VFA in combination with decreased pH, in vitro, with high concentrate substrate [11].This is an indication that DFM-modulated changes could be dependent on diet.
Monensin reduces NH 3 -N concentration through inhibition of the hyper-ammonia producing bacteria; a small group of ruminal bacteria that are responsible for the production of most of the NH 3 [17].In the present study monensin decreased levels of NH 3 when used in vitro with high forage substrate.Similar results were observed in vitro using a timothy hay substrate with the inclusion of monensin, where monensin caused a significant decrease in NH 3 accumulation [47].Additionally, monensin appeared to decrease NH 3 concentrations levels in vivo with a 90% orchardgrass diet although the differences among treatments were not significant [43].Ammonia-N did not differ with provision of DFM, which is in agreement with previous work that has found no differences in NH 3 concentrations or microbial N with DFM provision in vitro and in vivo [32,35].
Monensin decreased methane concentrations with inclusion in vitro with a high forage substrate.Similar results have been observed in vitro using a timothy hay substrate with the inclusion of monensin, where monensin caused a significant decrease in methane production [47].Additionally, monensin decreased methane production in vivo when included in forage-based diets consisting of corn silage/haylage and barley silage [48,49].Benefits of feeding monensin include a shift in the acetate to-propionate ratio toward more propionate and an associated decrease in methanogenesis [50].Consistent with these findings, monensin decreased methane and increased propionate in the present study.
There was no significant effect of DFM on methane production.Increased levels of acetate are indicative of increased methane production (as previously mentioned); consistent with our results neither end product was affected by DFM provision.Limited data is available with direct measure of methane as affected by inclusion of DFM.However, it has been suggested that some DFM may redirect H 2 reducing its availability for use in methane production [51].In terms of DFM containing yeast, there have been inconsistent results from studies both in vitro and in vivo, however, live yeast have shown beneficial effects on the growth and H 2 -utilization of acetogenic bacteria in vitro [52].Acetogenic bacteria reduce carbon dioxide to acetate and in this reduction they are competing with methanogens for hydrogen subsequently reducing CH 4 production [53].An additional strategy to mitigate CH 4 in ruminants is to increase competition for hydrogen by producing more propionate in the rumen [49,54].A recent study conducted by Alazzeh et al. [53] provides evidence those specific strains of Propionibacteria, which produce propionate as an end product of fermentation, could help mitigate methane production in vitro with forage diets.

Conclusion
The addition of monensin inhibited fermentation as indicated by reductions in both gas production and total VFA concentration.Monensin also altered VFA proportions and reduced methane production as has been previously reported.These actions were not influenced by the addition of DFM.In contrast, DFM had little influence on fermentation characteristics; exceptions being a slight increase in total VFA in the absence of MON and a small increase in pH that appeared to occur through a mechanism that differs from that for MON.Both DFM and MON increased pH, resulting in an additive effect when the two treatments were combined.

5
SM, Vanzant ES, Harmon DL, McLeod KR (2018) Effect of Direct-Fed Microbials and Monensin on In vitro Fermentation of a High-Forage Diet.J Anim Sci Res 2(3): doi dx.doi.org/10.16966/2576-6457.120Journal of Animal Science and Research Open Access Journal

Table 1 :
Ingredients and chemical composition of donor diet and in vitro substrate.

Table 2 :
Impact of direct-fed microbial and monensin on in vitro fermentative end products.