www.fems-microbiology.org

FEMS Microbiology Letters 247 (2005) 131–135

Fungi isolated from contaminated baled grass silage on farms inthe Irish Midlands

Martin O�Brien a,c,*, Padraig O�Kiely a, Patrick D. Forristal b, Hubert T. Fuller c

a Teagasc, Grange Research Centre, Dunsany, Co Meath, Irelandb Teagasc, Crops Research Centre, Oak Park, Co Carlow, Ireland

c Department of Botany, University College Dublin, Belfield, Dublin 4, Ireland

Received 30 March 2005; accepted 28 April 2005

First published online 23 May 2005

Edited by G.M. Gadd

Abstract

The incidence of fungal growth on baled grass silage was recorded on 35 farms in the Irish Midlands in 2003. Fungal colonies

were visible on 58 of 64 bales examined and the number of colonies per bale ranged from 1 to 12. On average, 5% of bale surface

areas were affected. The fungus most prevalent on bales was Penicillium roqueforti, present on 86% of bales and representing 52% of

all isolates. Other moulds isolated were Penicillium paneum, Geotrichum, Fusarium and mucoraceous species. Schizophyllum com-

mune was observed protruding through the plastic film on bales on 17 of the 35 farms.

� 2005 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

Keywords: Baled silage; Fungus; Mould; Penicillium roqueforti

1. Introduction

Currently, silage is made on 82% of all farms in Ire-

land to feed livestock when grass is in short supply.

Baled silage accounts for one-third of this total, and is

made on 80% of all silage-making farms [1]. In 2002,

some 12 million bales of grass silage were harvested

for subsequent feeding. The preservation of grass as si-

lage depends on storing the harvested crop under anaer-

obic conditions. In baled silage, baling the grass andimmediately wrapping each bale with at least four layers

of plastic stretch-film accomplished this. If the integrity

of the plastic film seal is damaged during storage, the in-

gress of oxygen will permit the growth of fungi and bac-

0378-1097/$22.00 � 2005 Federation of European Microbiological Societies

doi:10.1016/j.femsle.2005.04.037

* Corresponding author. Tel.: +353 4690 61100; fax: +353 4690

26154

E-mail address: mobrien@grange.teagasc.ie (M. O�Brien).

teria, resulting in quantitative and/or qualitative

nutrient losses in the silage [2]. Apart from these losses,fungal growth can reduce feed intake in livestock by

compromising the palatability of the forage, with conse-

quent losses in animal performance. Equally problem-

atic is the development of respiratory allergies and

mycoses in livestock, and also in farm workers, arising

from exposure to large numbers of mould spores, often

in poorly ventilated buildings [3–5]. Additionally, myco-

tic abortions may account for up to 15% of all abortionsin housed cattle, with fungi/fungal products gaining en-

try by inhalation or via the digestive tract [6]. Evidence

is increasing that mycotoxins are regularly formed under

ensiling conditions [7]. Recently, zearalenone and afla-

toxin B1 have been detected in maize silage [8], myco-

phenolic acid in grass and maize silages [9] and

roquefortine C in wilted grass and whole-crop maize si-

lages [10]. A comparative study on the preservation of

. Published by Elsevier B.V. All rights reserved.

132 M. O�Brien et al. / FEMS Microbiology Letters 247 (2005) 131–135

lucerne or grass from natural pastures in Italy showed

that wrapped round bales contained more mycotoxins

than hay made with the same forage [11]. Carry over

of mycotoxins into edible animal products such as milk

or meat could be a route to possible human exposure

[12].Only a few studies to date have researched the myco-

biota of baled grass silage. One such study recorded 76

fungal species in baled grass silage in Norway [13]. In

that survey, the fungus most frequently isolated was

Aspergillus fumigatus and the number of species re-

corded was highest in the genus Penicillium, of which

Penicillium roqueforti was the most common. Other spe-

cies commonly found in bales were Rhizopus stolonifer,Mucor circinelloides, Aspergillus flavus and Geotrichum

candidum. P. roqueforti is the most common contami-

nant mould found in various silage types [10,13–15].

On occasions when the mycotoxins roquefortine C and

mycophenolic acid were found in silage, P. roqueforti

was present [9,10]. In addition to moulds, macrofungi

are also found in baled silage; Schizophyllum commune

has been recorded on 53% of bale collections on Irishfarms [16].

This study was instigated because of concerns relating

to the perceived high incidence of visible mould on baled

grass silage in Ireland. The purpose of this survey was to

identify the visible fungal growths on baled silage, deter-

mine the extent of their occurrence and assess the risk

they might pose to livestock.

2. Materials and methods

In March 2003, baled grass silage was surveyed on 35

farms along a 150 km route (from 85�63 0 N to 12�10 0 S

(Irish National Grid)) in the Irish Midlands. The route

was sub-divided into five sections and seven farms were

surveyed per quintile. A detailed questionnaire was com-pleted on each farm visited, with information being

sought on harvesting and bale management practices.

One to three bales were examined in detail on each farm;

the bales chosen were those in readiness for feeding. The

number of bales examined on each farm represented the

usual number of bales fed daily to livestock. The total

number of bales sampled was 64. Prior to removing

the plastic film surrounding each bale, it was examinedcarefully for visible holes/damage. On removal of the

plastic film, all visible mould and yeast colonies on the

bale surface were located, numbered, sampled and pho-

tographed. Mould colonies were defined as areas of vis-

ible mycelial growth and yeast colonies as a diffuse

creamy growth on the surface of the bale. For yeasts

and visibly sporulating mould colonies, a sterile moist

cotton swab was gently touched off the fungal materialand replaced into a sterile sealed container. For mould

colonies that were not visibly sporulating, a small frag-

ment of foliage colonised with fungal material was asep-

tically transferred to an individual sterile sealable

container. The surface area of each colony was deter-

mined by placing a plastic grid (individual square si-

ze = 5 · 5 cm) over the colony and visually estimating

its area. The percentage of the total surface area affectedwith fungal growth was then calculated for each bale.

Bale collections on the farms were also assessed for

the occurrence of the macrofungus Schizophyllum com-

mune. The proportion of bales on each farm where S.

commune was visibly evident protruding through the

plastic film and its location on the bale surface was

recorded.

Dry matter (DM) and pH were determined using si-lage samples (four composited sub-samples per bale)

that had no visible mould. These samples were stored

at �18 �C until required for analyses and then thawed

and finely comminuted. DM was assayed by drying

(98 �C for 16 h) in an oven with forced air circulation

and pH was determined from the juice of a silage:dis-

tilled water (1:1) mixture. The pH of silage contami-

nated with visible mould was also recorded on site onfarms. The readings were taken from the centre of large

fungal colonies on bales by inserting a pH probe (Schott

Blueline 12 pH probe) to a depth of 5–6 cm into the fun-

gal contaminated silage.

The direct-plating method was used to isolate the

mould and yeast samples. Foliage bearing fungal mate-

rial was transferred aseptically to the surface of malt ex-

tract agar (MEA, Oxoid) and dicloran rose bengalchloramphenicol agar (DRBC, Oxoid). DRBC was used

to allow the growth of slow-growing fungi by inhibiting

the growth of rapid growers. The antibiotics, chlortetra-

cycline (50 lg ml�1, Sigma) and chloramphenicol

(100 lg ml�1, Sigma) were added to MEA to inhibit bac-

terial growth. DRBC contained 100 lg ml�1 chloram-

phenicol (Sigma). Four foliage fragments (<1 cm in

length) from each sample were placed onto each of thetwo media at four equidistant points. Cotton swabs with

adhering fungal material were gently touched against

the surface of the two media (four points per plate).

Plates were incubated for between 3 to 14 days at

25 �C depending on the growth rate of the fungi iso-

lated. Fungi were sub-cultured onto MEA (Oxoid),

incubated for 5–10 days (depending on the fungal spe-

cies) and stored at 4 �C for later identification. Fungiwere identified to genera or species by their macro-

and micromorphology features using appropriate identi-

fication keys [17–19]. Yeasts were not identified further.

As an additional confirmation tool for Penicillium iso-

lates, a representative number were screened for their

secondary metabolite profiles using HPLC-UV with

LC-MS.

Statistical analysis was performed using the generallinear model (GLM) procedure of the statistical analysis

system version 8.2 (SAS Inst. Inc., Cary, NC, USA). For

M. O�Brien et al. / FEMS Microbiology Letters 247 (2005) 131–135 133

significant F test, the least square means were compared

and the statistical difference determined using the prob-

ability of differences (PDIFF) procedure of SAS.

3. Results and discussion

The majority of the bales examined had been har-

vested in June and July 2002 and the bale storage char-

acteristics were similar to those observed in a previous

study [20]. Bales had a DM content of 286 (SD

93.9) g kg�1 and a pH of 4.3 (SD 0.56). The mean pH

of fungal-contaminated silage was 6.5 (SD 1.28), which

was significantly (P < 0.001) less acidic than clean silagefrom the same bales. Utilisation of lactic and other si-

lage acids by colonising moulds would result in a raised

silage pH and further reduction in silage preservation

quality. On average baled silage in Ireland contains

324 g DM kg�1 [1]. The lower than normal bale DM

content recorded in this survey may be explained by

the adverse conditions at ensiling. Meteorological data

at Grange Research Centre (located on the route sur-veyed) recorded mean rainfall in June and July 2002 that

was 17% above average and the corresponding mean to-

tal sun hours for these two months was 29% below aver-

age. The DM of baled silage is an important factor at

ensiling because grass with a lower DM facilitates the

harvesting of more densely compact bales, which helps

in excluding air and thus, discourages the growth of

fungi.Visible fungal growth was present on 58/64 (91%)

bales examined. On average, there were six visible fungal

colonies on each affected bale and this ranged from 1 to

12 colonies per bale (Fig. 1). The extent of fungal growth

on the bale surfaces ranged from 0.1% to 14.9% cover-

age, with a mean coverage of 5.1%. Most bales had an

area in the range 0.01–0.2 m2 covered with visible fungal

Fig. 1. Mould colonies on a bale of grass silage from which the plastic

film has just been removed. Bale Size = 1.2 m · 1.2 m (W · L).

growth (Fig. 2). Fungi appeared to be able to colonize

any part of the bale surface. The percentage of the bale

area affected with fungal growth when bales were stored

on their ends (n = 20 bales) and on their curved side

(n = 44 bales) was 5.4 (SD 4.7)% and 4.2 (SD 3.8)%,

respectively. As this difference was not statistically sig-nificant (P > 0.05), it indicates that the extent of fungal

colonisation is not contingent on how a bale is stored on

the ground.

The plastic film of 25/64 (40%) bales examined was

visibly damaged. This level of damage is probably an

underestimation as there may be micro-damage that is

not readily visible to the human eye. Damaged bales

had proportionally higher fungal coverage (7.0 (SD4.0)%) than where the film appeared intact (4.0 (SD

3.4)%) and this was statistically significant (P < 0.05).

While damage was observed on all parts of bales, the

plastic film was mostly observed to be damaged on the

bales curved side. Bird and cat damage accounted for

35% and 17% respectively of all damage. Farm machin-

ery, livestock and rodents caused other damage. Only

4% of damaged bales were repaired and these were gen-erally repaired using adhesive plastic repair patches.

Damaged plastic film on bales is normally repaired in

the interval between wrapping the bales and transport-

ing them to the storage site [20]. Anecdotal evidence

indicates that any subsequent damage that occurs dur-

ing storage is usually not repaired.

A total of 332 visible mould and yeast colonies were

sampled on 58 bales, resulting in 444 fungal isolates. Asingle pure fungal culture was isolated from 209 samples

taken from fungal colonies on bales. In the case of the

other samples, two, three or even four different fungi

grew from a single silage sample. On occasion, two fungi

were observed co-existing with each other on the same

0

10

20

30

40

50

absent 0.01-0.2 0.2-0.4 0.4-0.6 0.6-0.8 0.8-1.0 1.0-1.2

Area of fungal coverage (m2)

Perc

enta

ge o

f ba

les

Fig. 2. Extent of visible fungal growth on the surface of baled grass

silage (n = 64) on a sample of Irish farms (n = 35). Fungal growth on

bale surfaces was estimated using a plastic grid of a known area. Each

bar represents the percentage of bales contaminated with fungal

growth corresponding to the area range (m2) of the bale surface

affected with visible fungi. Total bale surface area = 6.78 m2; 0.2 m2

represents 3% of bale surface area.

134 M. O�Brien et al. / FEMS Microbiology Letters 247 (2005) 131–135

part of the bale. In most cases the target fungus/fungi on

the bale could be recognised in culture, isolated and

identified. Occasionally fungi that were not obviously

present at the time of sampling, grew out from silage

samples in culture. These fungi may have arisen from

dormant propagules in the silage. Ten fungal samplesfrom bales failed to grow in vitro, representing 2.7%

of total isolates.

The predominant genus isolated was Penicillium and

the species encountered most frequently was P. roque-

forti Thom (Table 1). This mould was present on 55/

64 (86%) bales examined. P. roqueforti was first isolated

and identified from both baled and clamp silage in Ire-

land in 2000 (O�Brien, M., unpublished). Although theincidence of P. paneum Frisvad was low (4.5% of all iso-

lates), the known ability of this fungus to produce patu-

lin would give cause for concern [21]. The identity of

both P. roqueforti and P. paneum was confirmed by their

secondary metabolite profiles. Other frequently isolated

fungi included Geotrichum, mucoraceous species S. com-

mune, and yeasts (Table 1). In a survey of big bale grass

silage in Norway, Aspergillus fumigatus was the speciesmost frequently isolated [13]. In contrast, S. commune

was absent from Norwegian baled silage, and to date

Aspergillus spp. have not been isolated from Irish bales.

Differences in the bale mycobiota of grass silage between

these two countries could be possibly due to climatic fac-

tors or to differences in bale management practices.

Bale collections on the 35 farms were surveyed to

establish the extent of occurrence of the macrofungusS. commune. This fungus was present in the form of a

bracket mushroom visibly protruding through the plas-

tic film on bales on 49% (17/35) of the farms surveyed.

This survey paralleled the results of a national survey

conducted five years previously that established the

widespread occurrence of S. commune on baled silage

in Ireland [16]. When present, Schizophyllum was ob-

served on less than 10% of the bale collection on a farm;the curved sides and shoulders of bales were the areas

most frequently affected.

Table 1

Fungi isolated from contaminated baled grass silage on a sample of 35

Irish farms in Spring 2003

Fungal genera/species Number of

isolates

Percentage of

total isolates

Penicillium roqueforti 231 52.0

Penicillium paneum 20 4.5

Yeasts 60 13.5

Geotrichum 35 7.9

Mucoraceous species 27 6.1

Schizophyllum commune 19 4.3

Fusarium 5 1.1

Trichoderma 2 0.5

Corprinus 1 0.2

Other unidentified isolates 44 9.9

Total 444

This survey showed that fungal contamination of

ruminant feed in the form of baled silage is widespread

on farms in the Irish Midlands. Although a relatively

small number of fungal species was responsible for most

of the contamination, at least two of these fungi (P.

roqueforti and P. paneum) are capable of reducing silagequality and potentially causing health problems in live-

stock by their known ability to produce harmful myco-

toxins. Air ingress to baled silage is a major factor in

facilitating mould colonisation in a substratum other-

wise inhibitory to fungal growth. Livestock owners

who feed baled silage need to be aware of the potentially

harmful effects that certain fungi pose to livestock and

to employ effective measures to maintain the integrityof the plastic film in order to minimise fungal contami-

nation on bales.

Acknowledgements

The authors wish to thank Messrs B. Burke, J. Ha-

mill, J. Marron and J. Moran for their technical assis-tance. The authors are also grateful to farmers for

facilitating sampling on their farms. For the screening

of Penicillium isolates for secondary metabolites, the

authors would like to sincerely thank Dr. K.F. Nielsen

and Prof. J.C. Frisvad (Centre for Microbial Biotech-

nology, Technical University of Denmark, Denmark)

and funding provided by the EU 5th Framework

Award. A Teagasc Walsh Fellowship Research Scholar-ship awarded to M. O�Brien supported this study.

References

[1] O�Kiely, P., Forristal, P.D., Lenehan, J.J., Fuller, H., Brady, K.

and Barlow, M. (1998) Characteristics of baled silage on Irish

farms. Farm Food 8, 29–32.

[2] Mc Namara, K.M., O�Kiely, P., Whelan, J., Forristal, P.D. and

Lenehan, J.J. (2002) Simulated bird damage to the plastic stretch-

film surrounding baled silage and its effects on conservation

characteristics. Ir. J. Agric. Res. 41, 29–41.

[3] Jarvis, B., Seiler, D.A.L., Ould, A.J.L. and Williams, A.P. (1983)

Observations on the enumeration of moulds in food and feeding

stuffs. J. Appl. Bacteriol. 55, 325–336.

[4] Sargison, N. (1993) Health hazards associated with the feeding of

big bale silage. In practice, 291–297.

[5] Lacey, J. (1975) Potential hazards to animals and man from

microorganisms in fodders and grain. Trans. B. Mycol. Soc. 65,

171–184.

[6] Scudamore, K.A. and Livesey, C.T. (1998) Occurrence and

significance of mycotoxins in forage crops and silage: a review.

J. Sci. Food Agric. 77, 1–17.

[7] Fink-Gremmels, J., (2003) Moulds and mycotoxins in silage, in:

The Second World Mycotoxin Forum, Noordwijk, The Nether-

lands, 17–18 February 2003.

[8] Cavallarin, L., Borreani, G., Tabacco, E. (2004) Mycotoxin

occurrence in farm maize silages in northern Italy, in: Proceedings

of 20th General Meeting of the European Grassland Federation,

Luzern, Switzerland, 21–24 June 2004.

M. O�Brien et al. / FEMS Microbiology Letters 247 (2005) 131–135 135

[9] Schneweis, I., Meyer, K., Hormansdorfer, S. and Bauer, J. (2000)

Mycophenolic acid in silage. Appl. Environ. Microbiol. 66, 3639–

3641.

[10] Auerbach, H., Oldenburg, E. and Weissbach, F. (1998) Incidence

of Penicillium roqueforti and roquefortine C in silage. J. Sci. Food

Agric. 76, 565–572.

[11] Tomasi, L., Horn, W., Roncada, P., Zaccaroni, A., Ligabue, M.,

Battini, F. and Stracciari, G.L. (1999) Recherches preliminaires

sur la presence d�aflatoxin B1 dans les fourrages fanes de la

province de Reggio Emilia (Italie). Fourrages 158, 179–186.

[12] Veldman, B., (2003) Effects of dietary mycotoxins on the quality

of animal products and on human health, in: The 2nd World

Mycotoxin Forum, Noordwijk, The Netherlands, 17–18 February

2003.

[13] Skaar, I., (1996) Mycological survey and characterisation of the

mycobiota of big bale grass silage in Norway, PhD thesis,

Norwegian College of Veterinary Medicine, Norway.

[14] Nout, M.J.R., Bouwmeester, H.M., Haaksma, J. and Van Dijk,

H. (1993) Fungal growth in silages of sugarbeet press pulp and

maize. J. Agric. Sci. Camb. 121, 323–326.

[15] Tuller, G., Armbruster, G., Wiedenmann, S., Hanichen, T.,

Schams, D. and Bauer, J. (1998) Occurrence of roquefortine in

silage – toxicological relevance to sheep. J. Anim. Physiol. Anim.

Nutr. 80, 246–249.

[16] Brady, K.C., O�Kiely, P., Forristal, P.D. and Fuller, H. (2005)

Schizophyllum commune on big-bale grass silage in Ireland.

Mycologist 19, 30–35.

[17] Cooke, W.B. (1961) The genus Schizophyllum. Mycologia 53,

575–599.

[18] Pitt, J.I. (2000) A Laboratory Guide to Common Penicillium

Species. Food Science Australia, North Ryde, NSW, Australia,

197 pp..

[19] Samson, R.A., Hoekstra, E.S., Frisvad, J.C. and Filtenborg,

O. (2002) Introduction to Food- and Airborne Fungi. Centra-

albureau voor Schimmelcultures, Utrecht, The Netherlands,

389 pp..

[20] Mc Namara, K.M., O�Kiely, P., Whelan, J., Forristal, P.D.,

Fuller, H. and Lenehan, J.J. (2001) Vertebrate pest damage

to wrapped baled silage in Ireland. Int. J. Pest. Manag. 47,

167–172.

[21] Boysen, M., Skouboe, P., Frisvad, J. and Rossen, L. (1996)

Reclassification of the Penicillium roqueforti group into three

species on the basis of molecular genetic and biochemical profiles.

Microbiology 142, 541–549.