Content uploaded by Elena Sorrentino
Author content
All content in this area was uploaded by Elena Sorrentino
Content may be subject to copyright.
481
Lait 85 (2005) 481–490
© INRA, EDP Sciences, 2005
DOI: 10.1051/lait:2005031
Original article
Use of donkey’s milk for a fermented beverage
with lactobacilli
Cristiana CHIAVARIa*, Fabio COLORETTIa, Mauro NANNIb,
Elena SORRENTINOc, Luigi GRAZIAa
a Dipartimento di Protezione e Valorizzazione Agroalimentare (DIPROVAL),
Alma Mater Studiorum-Università di Bologna, Via F.lli Rosselli, 107, 42100 Reggio Emilia, Italy
b Tinval s.r.l., Spin Off Alma Mater Studiorum-Università di Bologna, Italy
c Dipartimento di Scienze e Tecnologie Agroalimentari, Ambientali e Microbiologiche (DISTAAM)
dell’Università del Molise, Campobasso, Italy
Received 28 January 2005 – Accepted 27 April 2005
Published online 1 October 2005
Abstract – The possibility of producing a fermented beverage from donkey’s milk using the pro-
biotic bacterial strains Lactobacillus rhamnosus AT 194, CLT 2/2, and Lactobacillus casei LC 88,
isolated from Parmigiano Reggiano cheese was investigated. The chemical-physical and microbio-
logical properties of the raw milk demonstrated that it has a low microbiological load and an eleva-
ted content of lysozyme. The bacterial strains employed for fermentation had a good growth
capacity in donkey’s milk only after an initial adaptation phase. An extremely high percentage of
viable bacteria were found in the final beverage, even after a 30-day shelf life. Likewise, the activity
of lysozyme was virtually unchanged with respect to initial values. Sensorial analysis permitted the
individuation of differences between the three bacterial strains used for fermentation in terms of
descriptors relative to aromatic-olfactory qualities. Based on the above results, technology can be
proposed for production of a fermented beverage from donkey’s milk that can be utilized by small
producers. This would allow the production of a beverage that would be well accepted by consumers
interested in a product with favorable therapeutic properties integrated with probiotic bacteria.
donkey’s milk / fermented beverage / Lactobacillus rhamnosus / Lactobacillus casei
摘要 – 乳杆菌发酵驴奶生产发酵乳饮料。本文主要探讨了从 Parmigiano Reggiano 干酪（帕
尔马·勒佐安诺干酪）中分离的益生菌菌株鼠李糖乳杆菌 (Lactobacillus rhamnosus) AT194,
CLT2/2 和干酪乳杆菌 (Lactobacillus casei) LC88, 以驴奶为原料应用这三株益生菌生产发酵
乳饮料的可行性。经理化和微生物性质研究表明，原料乳的菌数较低而溶菌酶含量较高。
发酵用的菌株在驴奶培养基中经过适当的驯化后在驴奶中具有很好的生长能力。在最终的
产品中，活菌数极高，即使经过了 30d 的保藏，活菌数变化不显著。同时，溶菌酶活性与
原始样品相比也基本上没有变化。在感官分析过程中，依据香气和香味的特性来描绘三株
菌发酵制得的饮料之间的区别。基于上述结果，认为驴奶发酵生产乳饮料在技术上是可行
的，这种技术可用于小规模的乳饮料生产。这种乳饮料尤其结合了益生菌特定的治疗作用，
可以很好的被消费者所接受。
驴奶 / 发酵饮料 / 鼠李糖乳杆菌 / 干酪乳杆菌
Résumé – Utilisation du lait d’ânesse pour la production d’une boisson fermentée avec des
souches de lactobacilles. Cette étude concerne l’utilisation du lait d’ânesse pour la production
d’une boisson fermentée en utilisant les souches probiotiques Lactobacillus rhamnosus AT 194,
* Corresponding author ( 通讯作者): cristiana.chiavari@unibo.it
Article published by EDP Sciences and available at http://www.edpsciences.org/laitor http://dx.doi.org/10.1051/lait:2005031
482 C. Chiavari et al.
CLT 2/2 et la souche Lactobacillus casei LC 88 toutes isolées du fromage Parmigiano Reggiano.
Le lait, caractérisé du point de vue physico-chimique et microbiologique, a comme caractéristiques
prédominantes une charge microbienne faible et un contenu en lysozyme élevé. Les souches utili-
sées ont montré une bonne capacité de développement dans le lait d’ânesse seulement après une
phase d’adaptation. Les bactéries inoculées sont restées vivantes plus de trente jours dans les bois-
sons obtenues, garantissant ainsi une longue durée de conservation. Au terme de l’essai, l’activité
du lysozyme n’a présenté aucune variation sensible par rapport à l’activité initiale. L’analyse sen-
sorielle à laquelle ont été soumises les boissons fermentées a permis de discriminer les différences
induites par les trois souches utilisées en ce qui concerne quelques descripteurs relatifs au tableau
aromatique et olfactif. Une technologie de production d’une boisson fermentée applicable à de petits
ateliers artisanaux a donc été proposée, permettant d’obtenir un produit en mesure d’être accepté par
un consommateur intéressé à la fois par les caractéristiques thérapeutiques du lait d’ânesse et les
propriétés bénéfiques pour la santé des bactéries probiotiques.
lait d’ânesse / boisson fermentée / Lactobacillus rhamnosus / Lactobacillus casei
1. INTRODUCTION
The properties of equine milk differ from
that of other mammals in many ways that
include important differences in nutritional
value. Moreover, its composition does not
permit the production of cheese due to the
presence of whey proteins that represent
35–50% of the nitrogen fractions [14, 20].
It also contains lysozyme, which is practi-
cally absent in the milk of cows, ewes and
goats [5, 18]; this enzyme possesses bacte-
ricidal properties as it hydrolyzes the
polysaccharides of bacterial cell walls and
inhibits development.
Among various equine milks, the best
known is the mare’s milk that is tradition-
ally produced and consumed by the
nomadic peoples of Central Asia (Mongo-
lia, Siberia and Kazakhstan) after lactic acid
and alcoholic fermentation called Koumiss
[16]. Even to date, Koumiss is the national
drink of the people in that area and is also
quite popular in countries bordering the
Russian federation. It has been suggested
that Koumiss has probiotic properties and
has even been prescribed as a cure for
patients with various diseases in Russian
hospitals [17]. It is practically unknown in
the rest of Europe and the Western world.
Donkey’s milk has a composition simi-
lar to that of mare’s milk [20], but is char-
acterized by a higher lysozyme [7] content,
which makes it somewhat selective with
regards to the bacteria it can host. Its con-
sumption, though extremely limited, has
increased due to pediatric studies carried
out over the last decade for the treatment of
intolerance to cow’s milk in young children
[6, 8, 11]. It is not used, either fresh or after
fermentation, in the diets of adults even if
it has been suggested to have favorable
pharmaceutical and nutritional properties
[9, 10, 20]. Considering the numerous ben-
efits of donkey’s milk, including its health-
promoting characteristics and probiotic
effects, Coppola et al. [7] suggested the pos-
sibility of using donkey’s milk for probiotic
purposes. Accordingly, these authors investi-
gated the fermentative properties of don-
key’s milk in addition to investigating the
survival of selected Lactobacillus rhamno-
sus strains.
The aim of the present study was to
establish the methodology for the produc-
tion of a fermented beverage made from
donkey’s milk using only lactic acid bacte-
ria with probiotic activity. This would
allow for the retention of the beneficial
properties of the milk in addition to main-
taining the presence of microorganisms.
We also evaluated the shelf life of the prod-
uct with particular regard to the survival of
the inoculated lactic acid bacteria and the
resulting sensorial characteristics as a func-
tion of the bacteria used for fermentation.
2. MATERIALS AND METHODS
2.1. Milk
The m ilk was taken fr om a herd of around
150 donkeys in the province of Reggio
Emilia, Italy. In particular, four jennets that
Donkey’s milk for a fermented beverage 483
had given birth one month previously were
selected and mechanically milked for 150 d.
On each day, the first milking was carried
out 4 h after the colt was removed and the
second milking was 3 h later. The colt was
then returned to its mother 2 h after the sec-
ond milking. During 5 months of milking,
a sterile milk sample (250 mL) was sub-
jected to microbiological and physical-
chemical analyses every month.
2.2. Starter cultures
The strains Lactobacillus rhamnosus
AT 194 and CLT 2/2, isolated from Parmi-
giano Reggiano cheese, were from the
Department of Agro-industrial, Environmen-
tal and Microbiological Sciences’ (DIS-
TAAM) collection (University of Molise),
classified as probiotic [22]. The strain
Lactobacillus casei LC 88, isolated from
Parmigiano Reggiano cheese, was from the
Department of Agri-food Protection and
Improvement (DIPROVAL) at the Univer-
sity of Bologna. All strains were preserved
in MRS agar (Oxoid, Unipath, Basingstoke,
UK) at 4 °C.
To prepare starter cultures, cells were
grown in 10 mL of MRS broth for 24 h, har-
vested by centrifugation (5000 rpm for
10 min), washed twice in a solution of
0.9 g·L–1 NaCl and re-suspended in 10 mL
of sterile skimmed milk (Oxoid) at 37 °C for
4 h. Heat-treated (110 °C for 10 min) pas-
teurized whole milk (90 mL) was then inoc-
ulated with approximately 108 cfu·mL–1of
each strain and incubated at 37 °C for 12 h
prior to the next culture passage (pre-culture).
2.3. Production of fermented milk
After milking, the raw milk was poured
into sterile 100-mL Sovirel flasks, pasteur-
ized at 63 °C for 30 min, rapidly cooled at
37 °C, inoculated with different amounts of
pre-culture (3 mL for the strain AT194,
4 mL for CLT 2/2 and 4 mL for L. casei) to
obtain about 106–107 cells·mL–1 , and incu-
bated at 37 °C. The rate of fermentation was
monitored by measuring the pH every 12 h
after inoculation. As soon as a pH of 4.5–
4.6 was reached, the flasks were chilled and
placed in a refrigerator at 4 ± 0.2 °C. At the
moment of inoculation (t = 0), upon reach-
ing a pH of 4.5–4.6 (t = 2) and after 7, 15
and 30 d, microbiological analyses were
carried out in order to evaluate the shelf life.
Sensory characterization was carried out on
the samples after 30 d of storage at 4 ± 0.2 °C.
2.4. Microbiological analyses
Serial dilutions of each sample in
0.9 g·L–1 NaCl were prepared and then
plated in duplicate on agar plates [21]. Total
aerobic bacteria were counted on PC Agar
(Oxoid) incubated at 30 °C for 48 h; lactic
acid bacteria were counted in MRS agar
(Oxoid) in anaerobic jars for 5 d at 30 °C;
thermoduric bacteria were counted after
pasteurization (80 °C for 10 min) on PC
Agar (Oxoid) incubated at 30 °C for 72 h.
Enterobacteriaceae were counted after
plating on Violet Red Bile Glucose Agar
(Oxoid) after incubation for 24 h at 37 °C,
while fecal coliform bacteria were plated on
Violet Red Bile Lactose Agar (Oxoid) for
24 h at 42 °C.
2.5. Lysozyme content
The concentration of lysozyme was
measured in milk before and after pasteur-
ization, and in the fermented beverage dur-
ing the storage phase at 7, 15 and 30 d after
inoculation.
Lysozyme was measured as described
by Lodi et al. [13], through the evaluation
of lytic activity on Micrococcus lysodeikti-
cus of the samples compared with standard
purified enzyme solutions extracted from
eggs. The method used is characterized by the
linear relationship (semi-logarithmic curve)
between the diameter of the zone of inhibi-
tion around the wells at which concentra-
tions of lysozyme between 1 and 4 ppm
were layered.
2.6. Chemical and physical analyses
Chemical and physical analyses were
carried out using the standard methods:
titratable acidity [19], pH was measured
using a Hanna Instruments (Padova, Italy)
484 C. Chiavari et al.
pHmeter, density at 20 °C by Quevenne’s
densitometer, dry matter was determined
by the gravimetric method at 102 °C at con-
stant weight, ash was determined in a fur-
nace overnight at 550 °C, fat was measured
using the Rose-Gottlieb method [2], and
lactose was determined by the phenol sulfuric
method [15]. Total nitrogen and nitrogen
fractions were determined using the Kieltec
Tecator System following the procedure
described by Aschaffenburg and Drewry [3].
2.7. Sensory characterization
Sensory characterization of the fermented
milks produced with the probiotic strains
was performed by a group of 15 persons
(7 male and 8 female). An evaluation card
previously studied by trained tasters (staff
from DIPROVAL highly familiar with fer-
mented dairy products) was used for the
assessment on the basis of a methodology
suggested by Bérodier et al. [4] for cheeses.
As suggested by the authors, the intensity of
each descriptor was evaluated on an arbi-
trary scale with values from 1 to 7. For the
evaluation of the significance of the result-
ing averages the ANOVA test was applied.
The significance levels were set at P < 0.05.
3. RESULTS AND DISCUSSION
3.1. Physical-chemical traits
of the raw milk
Table I shows the composition of don-
key’s milk used for production of the fer-
mented beverage. Compared with cow’s
milk [1], which is the raw material most
commonly used in the production of fer-
mented milks, donkey’s milk has a notably
higher concentration of lactose and lower
levels of fat and protein.
From observation of the distribution of
nitrogen fractions donkey’s milk also has a
lower protein content, and in particular
casein (Tab. II). However, seroproteins rep-
resent 37% of the proteic content in don-
key’s milk compared with 17% in cow’s
milk. The richness in whey protein content
of donkey’s milk, as in all one-stomached
animals, makes it more favorable for human
nutrition [8, 11]. As expected, lysozyme
was present at very high concentrations
(> 3 g·L–1). Pasteurization carried out at
63 °C for 30 min did not have any effect on
the anti-microbial activity (Fig. 1, wells D
and E). This finding is in agreement with
those previously reported in donkey’s milk
[7, 12] and in horse’s milk [20], confirming
the elevated thermostability of the enzyme.
3.2. Microbiological and
pasteurization traits
Bulk milk contained a low microbial
content, which as affirmed by Coppola et al.
Tab le I. Main characteristics of raw donkey’s
milk (means and standard deviation for 5
milkings of 4 donkeys) in comparison with
cow’s milk [1].
Donkey Cow
Density at 20 °C 1.029 ± 0.02 1.032
pH 7.01 ± 0.07 6.68
Dry matter g·kg–1 89.06 ± 0.31 125.0
Ash g·kg–1 3.24 ± 0.24 8.0
Lactose g·kg–1 67.31 ± 0.63 47.0
Fat g·kg–1 2.82 ± 0.45 35.0
True protein g·kg–1 15.88 ± 0.24 31.0
Figure 1. Plate diffusion for milk and donkey’s
milk fermented beverage. Lysozyme determi-
nation: wells A, B, C lysozyme standard solutions
(1, 2, 4 ppm); wells D, E diluted milk, respec-
tively before and after pasteurization at 63 °C
for 30 min; wells F, G, H diluted fermented bev-
erage 7, 15 and 30 d after inoculation.
Donkey’s milk for a fermented beverage 485
[7], can be attributed to the anti-bacterial
activity of lysozyme. The raw milk used for
production of the fermented beverage was
nonetheless subjected to pasteurization (63 °C
for 30 min) as an additional safety measure.
The results achieved after heat treatment
are shown in Table III.
3.3. Fermentations
The strains used showed remarkable fer-
mentative vigor in cow’s milk (data not
shown), but also demonstrated difficulties
at the adaptation stage when inoculated
either in small quantities or directly inocu-
lated into pasteurized donkey’s milk. This
could be overcome by using a 3% inoculum
prepared in donkey’s milk, previously treated
at 110 °C for 10 min, for production of a fer-
mented beverage containing viable cells in
quantities greater than 108mL–1. The fer-
mentations took place in a rapid and regular
manner, as was also evident by the changes
in titratable acidity, and were stopped by
refrigeration after reaching pH values
between 4.5 and 4.6 after about 48 h (Fig. 2).
3.4. Shelf life
After 2 days following inoculation, the
microbial load was about 108 cfu·mL–1.
One month after inoculation the lactic
microbe content was still found to be above
108 cfu·mL–1, demonstrating that the
strains used remained viable over time and
even thrive, representing almost all of the
bacterial content (Tab. IV). The presence of
108 cfu·mL–1 of probiotic lactic bacteria is
sufficient to ensure the daily intake sug-
gested by Vanderhoof and Young [23],
even with limited consumption of the bev-
erage. Enterobacteria and coliforms were
completely absent. The activity of lysozyme
measured after 7, 15 and 30 d did not show
any significant changes with respect to raw
milk and thermally-treated milk (Fig. 1,
wells F, G and H).
Table II. Nitrogen fraction distribution in donkey’s milk (means and standard deviation for 5
milkings of 4 donkeys) in comparison with cow’s milk [1].
Donkey Cow
Tot al N g· L–1 2.56 ± 0.03 5.01
Non – casein N g·L–1 1.40 ± 0.03 1.09
Caseinic N g·L–1 1.15 ± 0.07 3.91
True whey protein g·L–1 0.96 ± 0.03 0.84
Lysozyme mg·L–1 3750.0 ± 250.0 0.09*
NPN g·L–1 0.44 ± 0.03 0.25
Distribution total N
Caseinic N % 45.26 ± 2.05 78.0
True whey protein % 37.24 ± 1.55 17.0
NPN % 17.50 ± 1.32 5.0
* Mean of data reported by the author [1].
Table III. Microbial counts (log cfu·mL–1) of
donkey’s milk before and after pasteurization
(means and standard deviation for 5 milkings
of 4 donkeys).
Raw milk Pasteurized
milk 63 °C
for 30 min
Total aerobic bacteria 4.24 ± 0.12 NR
Lactic acid bacteria <1.00 NR
Thermoduric bacteria <1.00 <1.00
Enterobacteriaceae 2.01 ± 0.07 NR
Fecal coliform bacteria NR NR
NR = none recovered.
486 C. Chiavari et al.
3.5. Sensorial testing
The various taste analyses by the trained
panel allowed the identification of the
descriptors normally used for the testing of
cheeses, which were most useful to describe
the fermented beverage from a sensorial
point of view. Evaluation of the fermented
beverages was carried out according to a
precise order: (1) visual characteristics;
(2) odor; (3) aroma; (4) four basic tastes;
(5) trigeminal sensations; (6) aftertaste and
persistence; and (7) pleasantness.
On the basis of the table drawn up by the
trained panel, the fermented beverages
were evaluated by a group of 15 consumers.
Table V contains each descriptor and the
average of the taste evaluations. No signif-
icant differences were found among the
beverages produced with the three strains of
lactic bacteria in terms of the visual test, the
four basic tastes, the trigeminal sensations,
or the aftertaste and persistence. However,
the individual strains did bring to light notable
differences in some descriptors regarding
the range of smells, as shown in Figure 3.
Figure 2. pH (continous line)
and titratable acidity (dotted
line) variation during fermen-
tation of the three strains
used.(●: L. rhamnosus AT
194; ■: L. rhamnosus CLT 2/2,
▲: L. casei LC 88).
Figure 3. Presentation of the 8 descriptors (average of the evaluation by 15 consumers), which allows
the differentiation of beverages fermented with the three strains (—— L. rhamnosus AT 194,
---- L. rhamnosus CLT 2/2, ········ L. casei LC 88).
Donkey’s milk for a fermented beverage 487
Overall, the L. casei strain proved to be
most capable of producing a beverage with
a more balanced aromatic olfactory profile
in the individual descriptors. The milk fer-
mented with strain AT194 demonstrated a
clear hint of boiled vegetables and acidic
milk, whereas strain CLT 2/2 gave a higher
average relating to the smell of fresh milk,
grasses and animal odors.
Among the elementary tastes more com-
monly perceived, albeit at slightly different
levels for each strain, were sweetness and
acid. The fermented milk that was preferred
from an organoleptic point of view by the
15 tasters was produced with strain AT194,
which was also the best for production of
the fermented beverage from donkey’s milk.
3.6. Suggested technology
The tests carried out permit us to suggest
production technology for a fermented bev-
erage with probiotic strains that can be easily
applied to a local setting in conjunction
Table I V. Monitoring of microbic flora (log cfu·mL–1) in fermented milks at different days (0, 2,
7,15 and 30 d).
L . rhamnosus AT 194
0 d 2 d 7 d 15 d 30 d
Total aerobic bacteria 6.85 8.63 8.97 8.43 8.11
Lactic acid bacteria 6.77 8.54 8.94 8.40 8.06
Thermoduric bacteria <1.00 <1.00 <1.00 <1.00 <1.00
Enterobacteriaceae NR NR NR NR NR
Fecal coliform bacteria NR NR NR NR NR
NR = none recovered.
L . rhamnosus CLT 2/2
0 d 2 d 7 d 15 d 30 d
Total aerobic bacteria 6.27 7.61 8.64 8.11 8.07
Lactic acid bacteria 6.17 7.30 8.56 8.11 8.02
Thermoduric bacteria <1.00 <1.00 <1.00 <1.00 <1.00
Enterobacteriaceae NR NR NR NR NR
Fecal coliform bacteria NR NR NR NR NR
NR = none recovered.
L. casei LC88
0 d 2 d 7 d 15 d 30 d
Total aerobic bacteria 6.70 8.42 8.25 8.15 8.07
Lactic acid bacteria 6.10 8.28 8.14 8.00 8.02
Thermoduric bacteria <1.00 <1.00 <1.00 <1.00 <1.00
Enterobacteriaceae NR NR NR NR NR
Fecal coliform bacteria NR NR NR NR NR
NR = none recovered.
488 C. Chiavari et al.
Tab le V. List of descriptors and sensorial analysis of the fermented beverages with the three strains
used. Values are means and standard deviation for n = 15 consumers.
AT 194 CLT 2/2 LC 88
Visual characteristics
Color (white) 5.45 ± 0.85 5.33 ± 0.98 5.33 ± 0.98
Homogeneity 4.90 ± 2.09 4.80 ± 1.85 4.95 ± 1.01
Grittiness 0.73 ± 2.02 0.73 ± 2.10 0.80 ± 1.42
Odor
Fresh lactic 1.77 a ± 1.98 3.56 b ± 1.93 3.65 b ± 1.93
Acidified lactic 6.30 a ± 0.66 5.70 b ± 0.82 3.78 c ± 0.44
Grass 2.50 a ± 1.92 4.91 b ± 2.25 2.70 a ± 1.17
Fermented grass 2.56 a ± 0.52 4.12 b ± 1.48 3.92 b ± 1.33
Boiled vegetables 4.61 a ± 2.10 3.50 b ± 0.58 3.00 b ± 0.90
Toasted seeds 0.36 ± 0.78 0.58 ± 0.38 0.40 ± 0.81
Very toasted 0.71 ± 0.46 0.80 ± 0.60 0.90 ± 0.56
Cow-herd 2.19 a ± 1.02 4.09 b ± 1.91 3.35 c ± 1.78
Butyric, rancid 2.00 ± 0.35 2.20 ± 0.67 2.47 ± 1.05
Putrid, sulphuric, silage 0.44 ± 1.31 0.50 ± 1.27 0.56 ± 1.33
Aroma
Fresh lactic 0.79 ± 0.46 0.78 ± 0.69 0.89 ± 0.66
Acidified lactic 5.30 ± 1.32 5.06 ± 0.75 4.96 ± 0.58
Grass 2.71 ± 0.46 2.75 ± 0.76 2.92 ± 0.94
Fermented grass 2.81 ± 0.52 3.25 ± 0.53 3.22 ± 0.51
Boiled vegetables 4.88 a ± 1.27 2.86 b ± 1.59 2.18 c ± 0.31
Toasted seeds 0.60 ± 1.04 0.50 ± 1.07 0.50 ± 0.98
Very toasted 0.75 ± 1.12 0.78 ± 1.39 0.89 ± 1.69
Cow-herd 1.89 ± 0.56 2.04 ± 0.46 2.24 ± 1.05
Butyric, rancid 2.57 ± 0.74 2.42 ± 0.64 2.42 ± 0.64
Putrid, sulphuric, silage 0.67 ± 1.07 0.70 ± 1.34 0.55 ± 1.21
Basic tastes
Sweet 5.50 ± 1.35 5.36 ± 1.77 4.97 ± 1.45
Salty 0.10 ± 0.18 0.12 ± 0.30 0.09 ± 0.31
Acid 5.15 ± 0.83 4.85 ± 0.71 4.88 ± 1.10
Bitter 0.20 ± 0.60 0.20 ± 1.26 0.36 ± 1.15
Trigeminal sensations
Astringent 0.46 ± 0.41 0.40 ± 0.25 0.30 ± 0.12
Aftertaste and persistence 2.86 ± 0.91 2.46 ± 1.11 2.22 ± 1.04
Pleasantness 5.11 a ± 1.62 2.96 b ± 1.49 2.17 c ± 1.15 c
Values in the same row with different letters are significantly different: P < 0.05.
Donkey’s milk for a fermented beverage 489
with a donkey herd. The most delicate stage
of the technique is adaptation of the bacte-
rial strains to donkey’s milk, which was
overcome by adaptation in cow’s milk and
sterilized donkey’s milk. After adaptation
in donkey’s milk, the strains were able to
undergo rapid fermentation, especially fol-
lowing abundant inoculation.
After eventual deodorization and homog-
enization, the procedure outlined in Figure 4
entails thermal pasteurization at 63 °C for
30 min. Inoculation at 10% is done with pre-
viously obtained fermented donkey’s milk.
The fermentation takes place at 37 °C and
its progress is followed by measuring the
pH. Upon reaching pH 4.5–4.6, fermenta-
tion is stopped by refrigeration at 4 °C,
which may be followed by homogenization,
bottling, and storage at 4 °C. The process c an
be easily performed by small businesses
and does not require complex machinery.
4. CONCLUSION
Donkey’s milk can be easily used for the
production of a beverage fermented with
probiotic strains of lactic bacteria that retain
a high degree of viability after fermentation
and storage. Such a product is obtained by
setting up appropriate measures during
inoculation that allow the bacterial cultures
to replicate despite the high concentration
of lysozyme present in the raw material.
Moreover, strains of lactic bacteria that con-
fer pleasant sensorial traits to the fermented
product are used.
In fact, using the described procedure, a
product is obtained that would be well
accepted from a sensorial standpoint by
consumers looking for the therapeutic qual-
ities of donkey’s milk integrated with pro-
biotic bacteria. While the method described
within is suitable for small-scale produc-
tion, in the case of industrial production it
would be worthwhile to apply additional
procedures such as the standardization of
the raw material, homogenization, and the
addition of flavorings such as fruit juice.
The data obtained confirms that donkey’s
milk is a possible basis for a fermented
beverage as it contains several advantageous
qualities, such as low microbial activity and
high amounts of lysozyme, as well as being
a vehicle for the consumption of probiotic
bacteria.
Further studies may be warranted in
order to select other bacterial strains with
probiotic properties that are even better
adapted to donkey’s milk.
Figure 4. Technology proposed for the produc-
tion of a beverage made from donkey’s milk fer-
mented with probiotic lactic acid bacteria.
490 C. Chiavari et al.
Acknowledgements: We thank G. Borghi and
his family, Azienda Agrituristica Montebaducco,
Salvarano di Quattrocastella (Reggio Emilia,
Italy) for providing the milk used in this trial.
REFERENCES
[1] Alais C., Science du lait. Principes des Tech-
niques Laitières, 4th edn., Sepiac, Paris, France,
1984.
[2] AOAC, Official methods of analysis of the
association of official analytical chemists,
13th edn., Washington, DC, USA, 1980.
[3] Aschaffenburg R., Drewry J., New procedure
for the routine determination of the various
non casein proteins of milk, in: 15th Proc. Int.
Dairy Congress, London, England, 1959,
pp. 1631–1637.
[4] Bérodier F., Zannoni M., Herrero L., Lavanchy
P., Casals J., Adamo C., Guide to the smell,
aroma and taste evaluation of hard and semi-
hard cheese, European Research Program
AIR-CT94-2039, GECOTEFT, Poligny,
France, 1997.
[5] Blanc B., Les protéines du lait à activité enzy-
matique et hormonale, Lait 62 (1982) 350–395.
[6] Carroccio A., Cavataio F., Montalto G.,
D’Amico D., Alabrese L., Iacono G., Intole-
rance to hydrolysed milk proteins in infants:
clinical characteristics and dietary treatment,
Clin. Exp. Allergy 30 (2000) 1598–1603.
[7] Coppola R., Salimei E., Succi M., Sorrentino
E., Nanni M., Ranieri P., Belli Blanes R.,
Grazia L., Behaviour of Lactobacillus rham-
nosus strains in ass’s milk, Ann. Microbiol.
52 (2002) 55–60.
[8] Curadi C., Giampietro P.G., Lucenti P.,
Orlandi M., Use of mare’s milk in pediatric
allergology, in: 14th Proc. ASPA Congress,
Firenze, Italy, 2001, pp. 647–649.
[9] Doreau M., Le lait de jument, Prod. Anim. 4
(1991) 297–302.
[10] Herrouin M., Mollé D., Fauquant J., Ballestra
F., Maubois J.L., Léonil J., New genetic
variants identified in donkey’s milk whey
proteins, J. Protein Chem. 19 (2000) 105–115.
[11] Iacono G., Carroccio A., Cavataio F., Montalto
G., Soresi M., Balsamo V., Use of ass’s milk
in multiple food allergy, J. Pediatr. Gastroen-
terol. Nutr. 14 (1992) 177–181.
[12] Jauregui-Adell J., Heat stability and reactiva-
tion of mare milk lysozyme, J. Dairy Sci. 58
(1975) 835–839.
[13] Lodi R., Todesco R., Carini S., Indagine pre-
liminare sull’impiego del lisozima nella tec-
nologia del formaggio Provolone, Ind. Latte
20 (1984) 37–50.
[14] Malacarne M., Martuzzi F., Summer A.,
Mariani P., Protein and fat composition of
mare’s milk: some nutritional remarks with
reference to human and cow’s milk, Int.
Dairy J. 12 (2002) 869–877.
[15] Marier J.R., Boulet M., Direct analysis of lac-
tose in milk and serum, J. Dairy Sci. 42
(1959) 1390–1391.
[16] Montanari G., Zambonelli C., Grazia L.,
Kamesheva G.K., Shigaeva M.K.H., Saccha-
romyces unisporus as the principal alcoholic
fermentation microorganism of traditional
Koumiss, J. Dairy Res. 63 (1996) 327–331.
[17] Nassal J., Rembalski C., Hygienische Forde-
rungen bei der Produktion von Stutenmilk
und Kumyß, Arch. Lebensm. Hyg. 31 (1980)
209–213.
[18] Pagliarini E., Solaroli G., Peri C., Chemical
and physical characteristics of mare’s milk,
Ital. J. Food. Sci. 5 (1993) 323–332.
[19] Politi I., Chimica del latte e dei latticini,
Centro Sperimentale del Latte, Milano, Italy,
1958.
[20] Solaroli G., Pagliarini E., Peri C., Composi-
tion and nutritional quality of mare’s milk,
Ital. J. Food. Sci. 5 (1993) 3–10.
[21] Speck M.L., Compendium of methods for the
microbiological examination of food, Ame-
rican Public Health Association, Washing-
ton, D.C., USA, 1984.
[22] Succi M., Sorrentino E., Coppola R., Nanni
M., Chiavari C., Grazia L., I formaggi a lunga
stagionatura prodotti con latte crudo: una
fonte di batteri utili, in: Proc. Conference on
“The cheese: technology, ripening and cha-
racterization”, Parma, Italy, 2002, pp. 89–98.
[23] Vanderhoof J.A., Young R., Use of probiotics
in childhood gastrointestinal disorderes, J.
Pediatr. Gastroenterol. Nutr. 27 (1998) 323–
332.
