Epiphytic lactic acid bacteria succession during the pre-ensiling Epiphytic lactic acid bacteria succession during the pre-ensiling and ensiling periods of alfalfa and corn and ensiling periods of alfalfa and corn

the fermentation explained by the epiphytic LAB count or population succession. Rather, the well-fermented alfalfa silages were those ensiled at a high dry matter (DM) content (>36%) and low buffering capacity (<450 meq/kg of DM). Only a few of the LAB strains were consistently present, thus indicating that populations changed during fermentation to fit an ecological niche

Financial assistance was provided by Chr.Hansen's Bio Systems, Milwaukee, Wisconsin; Kemin 1 Industries, Inc., Des Moines, Iowa; and Pioneer Hi-Bred International, Inc., North American Seed Division, Johnston, Iowa.

Summary
Twenty three species and 306 strains of epiphytic lactic acid bacteria (LAB) were found for two cuttings of alfalfa, each harvested at three stages of maturity, and three whole-plant corn hybrids.Epiphytic LAB counts were low and variable on the standing crops, particularly on alfalfa.Wilting increased LAB numbers slightly for alfalfa, but the chopping process increased counts dramatically for both crops.
Lactobacillus plantarum, Pediococcus pentosaceus, Enterococcus faecium, and E. faecalis were predominant on both standing crops.The changes in LAB caused by wilting or chopping were mainly proportional changes in the four dominant species.Once the crops were ensiled, total LAB counts increased rapidly, reached a maximum within 1 day, and then declined after 7 days of fermentation.Enterococcus species decreased sharply or disappeared during the early fermentation.The species most prominent through day 7 were L. plantarum and P. pentosaceus.After 7 days, more species, i.e., L. homohiochii, L. brevis, and L. gasseri, joined the succession and became prevalent, depending on the crop.
Only two of the six alfalfa silages were adequately preserved, whereas all three corn hybrids fermented normally.No relationship was found between epiphytic LAB numbers or species and adequacy of fermentation.Neither were pH changes during the fermentation explained by the epiphytic LAB count or population succession.Rather, the wellfermented alfalfa silages were those ensiled at a high dry matter (DM) content (>36%) and low buffering capacity (<450 meq/kg of DM).Only a few of the LAB strains were consistently present, thus indicating that populations changed during fermentation to fit an ecological niche.

Introduction
Epiphytic LAB (i.e., lactobacilli, lactococci, enterococci, pediococci, streptococci, and leuconostocs) play a major role in silage fermentation.Their absolute and relative numbers might be important in predicting fermentation adequacy and in deciding whether or not to apply a silage bacterial inoculant.Epiphytic LAB counts are usually low and variable on silage crops, and LAB counts usually increase coincident to the chopping process.
Only limited information is available on epiphytic LAB succession during the ensiling of alfalfa and corn, the two major silage crops in the United States.Our objective was to investigate the epiphytic LAB succession during the pre-ensiling and ensiling periods for alfalfa and whole-plant corn.

Experimental Procedures
A second-year stand of Cody alfalfa was harvested at the 2nd and 4th cuttings and at the late-bud, 10% bloom, and 50% bloom stages of maturity within each cutting in the 1989 growing season.Following mowing, the alfalfa was wilted in the windrow for 5 to 6 hr prior to chopping.Three corn hybrids (Pioneer 3377, 3379, and 3389) were grown under irrigation in 1989 and directly chopped at the 2/3 milk line stage of kernel maturity.Samples were aseptically taken from standing crops, windrowed alfalfa prior to chopping, and chopped forages.The forages were ensiled in laboratory silos, stored, and sampled as described for the alfalfa and corn silages on pages 118 and 124, respectively, of this report.
Isolation of lactic acid bacteria.Each sample was homogenized with phosphate buffer (0.7 mM, pH 7.0) and serially diluted in the same buffer.Lactobacilli, pediococci, and leuconostocs were counted on Rogosa SL medium (Difco #480).Plates were overlaid with the same medium and incubated at 35 C for 2 days.Streptococci were counted on Slanetz & Bartley medium (Oxoid CM 377) following incubation at 35 C for 2 days.
Approximately 30 colonies from Rogosa SL medium and 20 from the Slanetz & Bartley medium were picked at random.Each colony was purified twice on Lactobacilli MRS medium (Difco #881) containing 1.5% agar (Difco #140).The pure cultures were grown in Lactobacilli MRS broth at 35 C for 20 hr, mixed with sterile glycerol in a ratio of 2:3, and stored as stock cultures at -22 C for further examination.
Morphology, soluble proteins, and biochemical tests.The cultures were examined for Gram reaction, morphology, and catalase production.Gram-positive, catalasenegative rods and cocci were studied further.For the isolates from Slanetz & Bartley me-dium, only Gram-positive, catalase-negative cocci were tested further.
Soluble cellular proteins from the pure cultures were analyzed by polyacrylamide gel electrophoresis, using a commercial Pediococcus strain as a control.
Cultures with different soluble protein patterns underwent final tests for lactic acid configuration, growth at 15 and 45 C in Lactobacilli MRS broth, gas production from glucose, fermentation of 32 carbohydrates, hydrolysis of esculin, and deamination of arginine.
Identification and counts of lactic acid bacteria.The identity of each isolate was determined by a dBASE III computer program that compared the tested characteristics of the LAB to respective phenotypic information in Bergey's Manual of Systematic Bacteriology.The percentage of each species or strain was calculated from the total LAB counts for each sample.Total LAB count was defined as count from Rogosa SL medium plus streptococci count from the Slanetz & Bartley medium.

Results and Discussion
LAB species succession during the preensiling and ensiling periods.Twenty three species and 306 strains of LAB were identified out of 3,400 colonies isolated from the crops and their silages (Table 1).Of the total LAB colonies isolated, more than 90% were homofermentative (produce only lactic acid).Lactobacillus plantarum was the predominant species, followed by Pediococcus pentosaceus.Enterococcus faecium, E. faecalis, L. brevis, and L. homohiochii contributed about 33% of the isolates.Remaining species were each 1% or less of the total.More LAB species were recovered from alfalfa than corn, although two of the species on corn (L.casei and Streptococcus bovis) were absent on alfalfa.Heterofermentive LAB were more numerous on alfalfa than corn, whereas the proportion of homofermentative L. plantarum was higher on corn.Only a few strains of L. plantarum and P. pentosaceus were the same between the crops.
The LAB species or strains and their percentages of the total isolates at each period varied considerably among silage and showed profound changes as fermentation progressed (Tables 2 and 3).The second cutting, 10% bloom, standing alfalfa was inhabited only by heterofermentative L. brevis and Leuconostoc mesenteroides subsp.
Dramatic changes in epiphytic LAB occurred during the ensiling period.E. faecalis disappeared within 1 day.E. faecium decreased and vanished within 3 days, except for the fourth cutting, late-bud alfalfa.
L. plantarum predominated throughout the ensiling period of the second cutting, 10 and 50% bloom alfalfa silages and, with P. pentosaceus, dominated the other silages through day 7.After 7 days of fermentation, L. brevis and L. homohiochii increased and, along with L. plantarum and P. pentosaceus, became prevalent in several silages at 42 days.At the end of the ensiling period, two of the silages had L. brevis and one had L. gasseri as predominant LAB species.The other six were dominated by L. plantarum.P. pentosaceus disappeared from seven of the nine silages and was minor in the other two.None of the LAB strains remained predominant throughout the ensiling period in either alfalfa or corn.
Fermentation changes during the ensiling period (data not shown).For the second cutting alfalfa ensiled at 10 and 50% bloom, pH declined rapidly to 4.9 on the second day and dropped an additional .2 to .4 pH units by the end of the ensiling period.For the other four alfalfas, pH decreased to about 5.0 at 1 day but did not decline further.For the corn silages, the minimum pH was reached on day 3 of fermentation for hybrids 3377 and 3389; however, 3379 silage did not reach its lowest pH until day 7.
Lactic acid levels varied with both fermentation time and crop.For the second cutting, 10 and 50% bloom alfalfas, lactic acid increased rapidly during the first 3 days of fermentation and remained high (5.5 to 7.5% of the silage DM).For the other four alfalfa silages, lactic acid increased initially, then declined to only 2 to 3% of the silage DM at the end of the ensiling period.Lactic acid content in the second cutting, late-bud alfalfa silage remained almost constant during the final 87 days of fermentation, but in the fourth cutting, 50% bloom silage, lactic acid decreased from 8.5% of the silage DM on day 42 to only 2.5% on day 90.In all three corn silages, lactic acid concentration continued to increase throughout the ensiling period.
Ammonia-nitrogen increased throughout the ensiling period for all six alfalfa silages.However, only second cutting alfalfa ensiled at either 10 or 50% bloom had low enough ammonia-nitrogen values at the end of the fermentation to be acceptable (less than 12% of the total silage nitrogen).Before ensiling, these two alfalfas had the highest DM contents and lowest buffering capacities (Table 4).For all three corn hybrids, ammonia-nitrogen was produced slowly during fermentation and was less than 12% of the total nitrogen at 120 days.
Conclusions.These results indicate that the numbers and species of epiphytic LAB varied between alfalfa and corn and during the pre-ensiling and ensiling periods.Because knowing the numbers and species of LAB did not predict the outcome of the fermentations, further characterization of the epiphytic LAB strains and chemical composition of the ensiled crops, particularly their water soluble carbohydrate profiles and buffering capacity, is necessary.