Journal of Chromatography A

Volume 1217, Issue 29, 16 July 2010, Pages 48344840

Proteomic characterization of donkey milk caseome

Lina Chianesea, , ,
Maria Grazia Calabresea,
Pasquale Ferrantia,
Rosalba Maurielloa,
Giuseppina Garroa,
Carmela De Simonea,
Maria Quartoa,
Francesco Addeoa,
Gianfranco Cosenzab,
Luigi Ramunnob
Show more
doi:10.1016/j.chroma.2010.05.017
Get rights and content

Abstract
At present, compared with bovine milk, the characterization of donkey milk
caseins is at a relatively early stage progress, and only limited data are related to
its genetic polymorphism. In this work, the heterogeneity of donkey caseome was
investigated using a proteomic approach, based on one- (PAGE, UTLIEF) and
two-dimensional (PAGE UTLIEF) electrophoresis, stained with either
Coomassie Brilliant Blue or specific polyclonal antibodies, and structural MS
analysis. These combined methodologies allowed the contemporary identification
of donkey s1, s2, and -CN with their related heterogeneity due to
phosphorylation (s1, s2 and -CN), glycosylation (-CN) and incorrect splicing of
RNA in mRNA (deleted forms of s1-CN and -CN). The results achieved showed
11 components for -CN, six phosphorylated components for and s1-CN and
three main phosphorylated components for s2-CN, each accounting for 10, 11
and 12 P/mole. At this regard, for the first time, the primary structure of the

expressed protein corresponding to the only available donkey s2-CN cDNA

sequence was determined. Furthermore -CN was found in homozygous and
heterozygous state for the occurrence of a genetic -CN variant having a MW
value 28 mass units higher than the common -CN phenotype.

Keywords

Donkey caseome;
Mass spectrometry;
Proteomics;
Two-dimensional electrophoresis

1. Introduction
The main studies available on mammal milk proteins (caseins and whey proteins)
are focused on the four ruminant species (cow, water buffalo, sheep and goat)
because of the worldwide availability and use of their milk or derived products
(fermented milks, cheeses or dry powders) in human diet.
In general, the studies carried out on milk proteins are aimed to the
characterization of their molecular composition (primary structure, disulphide
bridges and other post-translational modifications as phosphorylation and
glycosylation), which can justify both their functional properties (solubility, clotting
aptitude, thermal denaturation) and nutritional quality (amino acid composition,
digestibility and bioactivity).
The most effective analytical procedures for structural analysis of casein and/or
whey protein fractions are based on one-dimensional electrophoretic techniques
(PAGE, UTLIEF and PAGE-SDS) or, more efficiently, on their orthogonal
combination (2-DE), coupled either to immunospecific staining and/or to MS
analysis [1], [2], [3], [4],[5] and [6]. In fact, more frequently, overlapping among
components of different casein fractions can take place in one-dimensional
electrophoretic analysis, caused by their very similar net charge or pI values.
These events can be amplified by the high heterogeneity of each single protein
fraction due to post-translational phenomena (phosphorylation, glycosylation),
genetic polymorphism (one or two variants in individual samples, much more in

bulk milk), non-allelic deleted forms [7], [8] and [10] and proteolysis by action of
endogenous proteases.
In addition to ruminants, same mono-gastric animal species produce milk
destined to human consumption, such as mare and donkey [11], [12] and [13]. At
this regard, recently, donkey milk intake showed a positive effect in diet therapy
of cow's milk protein allergy (CMPA) patients [14], [15] and [16] and an antiproliferative and anti-tumor in vitroeffects on human lung cancer cells [17]. From
a quantitative point of view, the composition of donkey milk is more similar to
human milk than bovine because of the higher lactose content
(6.5 vs. 5 g/100 mL), the lower protein content (1.2 vs.3.2 g/100 mL), the lower
casein (CN)/whey protein (WP) ratio (about 1 vs. 4) and the higher level of the
non-protein nitrogen (NPN) fraction (0.29 vs. 0.18%) [18].
The available knowledge on donkey milk protein structures mainly concerns the
primary structure and genetic polymorphism of whey proteins, such as lactalbumin (-La)[19] and [20], -lactoglobulin (-Lg) I and
II [21], [22], [23] and [24], lysozyme [19] and [25]and SA (serum albumin)
[UniProtKB/Swiss-Prot Q5XLE4]. Regarding donkey casein composition, the
occurrence of -CN and s1-CN has been reported, by means of IEF and MS
analysis through fitting of their tryptic hydrolysates with the known mare
counterparts [26] and the entire -CN and s1-CN sequences has been
determined[27] and [28].
A more recent study [29] was based on N-terminal sequencing and peptide mass
fingerprinting of 33 spots excised from 2-DE of donkey milk using horse protein
as reference.
In this paper, the donkey caseome has been characterized using a proteomic
approach, coupling one-dimensional (PAGE, UTLIEF) and 2-DE
(PAGE UTLIEF) profiles, stained with either Coomassie Brilliant Blue (CBB) or
specific polyclonal antibodies, with structural MS analysis.

2. Materials and methods

2.1. Materials
All chemicals were of the highest purity commercially available and were used
without further purification. TEMED, ammonium persulphate, glycine, urea,
ammonium bicarbonate (AMBIC), HPLC grade H2O, formic acid (FA) and CH3CN
were purchased from Carlo Erba (Milan, Italy). TFA, sinapinic acid (SA), -cyano4-hydroxycinnamic acid (CHCA), iodoacetamide, dithiothreitol were obtained

from Aldrich (St. Louis, MO, USA). Modified trypsin, sequencing grade was from
Promega (Madison, WI, USA), alkaline phosphatase (AP) was from Roche
(Mannheim, Germany). Zip-tip C18 microcolumns were from Millipore (Bedford,
MA, USA). Acrylamide, Bis, Ampholine buffers were from GE Healthcare
Amersham Bioscience (Buckinghamshire, UK). Coomassie Brilliant Blue R 250
and G 250 were purchased from Bio-Rad (Richmond, CA, USA).
2.2. Donkey milk sampling and casein extraction
Milk samples collected from 63 donkeys reared in Italy were analysed
individually. Each casein sample was prepared by acid precipitation from
skimmed milk, followed by centrifugation at 2500 g for 15 min, as described by
Aschaffenburg and Drewry [30]. The casein sample was freeze-dried and stored
at 20 C before use.
2.3. Electrophoretic and immunoblotting analysis
Casein samples (20 g/L) for electrophoretic analysis were dissolved in a 9 M
urea solution, containing 2-mercaptoethanol (1 mL/L). Polyacrylamide gel
electrophoresis (PAGE) at pH 8.6 was carried out with a vertical electrophoretic
apparatus (Protean II, Bio-Rad, Richmond, CA, USA) at 200 V and 6 C for 7 h
according to the procedure described by Chianese et al. [3]. Staining was
performed with Coomassie Brilliant Blue R-250.
Ultra-thin layer isoelectric focusing (UTLIEF) on polyacrylamide gels (0.25 mm)
was carried out as previously described [3]. Briefly, the pH gradient in the range
2.56.5 was obtained by mixing Ampholine buffers (GE Healthcare Amersham
Bioscience, Buckinghamshire, UK), 2.55, 4.55.4 and 46.5 (1.6:1.4:1 by
volume). The gel was stained with CBB G-250 as described by Krause et al. [31].
The 2-DE procedure was achieved by combining the first dimension PAGE gel
with the second dimension UTLIEF gel; in particular, an unstained strip of gel
PAGE, rinsed twice in distilled water and in 9 M urea with mercaptoethanol
0.01% (v/v), was applied along the cathode of prefocused UTLIEF plate.
For immunoblotting analysis, the casein fractions separated either by PAGE,
UTLIEF or 2-DE analysis were transferred by capillary diffusion from the gel onto
a nitrocellulose membrane (0.45 m, Trans-Blot, Bio-Rad). Immunodetection was
carried out according to the procedure already described by Chianese et
al. [3] using rabbit polyclonal antibodies against bovine peptides s1-CN (187
199) and -CN (195199) and porcine and s2-CN.

2.4. RP-HPLC analysis

Casein samples were fractionated by RP-HPLC on a 214TP54, 5 m Vydac C4,
250 mm 4.6 mm internal diameter column (Vydac, Hesperia, CA, USA) at a
detection wavelength of 220 nm. Solvent A was 0.1% trifluoroacetic acid in ultra
pure water (v/v) and solvent B 0.1% (v/v) TFA in acetonitrile. 200 L of a solution
containing 1 mg (casein sample)/mL (solvent A) were loaded onto a C 4 column,
equilibrated with solvent A. The elution program involved a gradient from 30 to
50% solvent B in 40 min, then from 50 to 100% B in 2 min, at a flow rate of
1 mL/min. Each eluted casein fraction was manually collected, freeze-dried, and
stored at 20 C.
2.5. In-gel digestion of protein spots
In-gel digestion of the protein spots was carried out on selected gel pieces
manually excised from the CBB-stained two-dimensional electrophoresis gels,
following the procedure reported by Mamone et al. [4]. Analysis of intact proteins
eluted from gels was carried out according to Cohen and Chait [32].
2.6. Matrix-assisted laser desorption/ionization-time-of-flight-mass spectrometry
analysis
MALDI-TOF-MS experiments were carried out on a Voyager DEPRO mass
spectrometer (PerSeptive Biosystems, Framingham, MA) equipped with delay
extraction technology and N2 laser at 337 nm. Mass spectra were acquired both
in positive linear or in reflector mode and 10 mg/mL SA and CHCA both dissolved
in 50% ACN/0.1% TFA, were used as matrices for the analysis of proteins and
peptides, respectively. MALDI-TOF analysis of intact proteins were obtained in
linear positive ion mode over the m/z range 10,00030,000 and were averaged
from about 150 laser shots.
The mixtures of tryptic peptides were subjected to a desalting/concentration step
with Zip-Tip C18 microcolumns prior to analysis by MALDI-TOF. Spectra were
obtained in reflectron positive ion mode over an m/z range 6004000 and were
averaged from about 150 laser shots. External calibration was performed by
acquiring separate spectra of a mixture of standard peptides (Perseptive
Biosystems).
2.7. Electrospray quadrupole-time-of-flight-mass spectrometry

Peptides and proteins were analysed by micro-LC-ESI Q-TOF MS using a

CapLC Waters high-throughput configuration directly connected to a Q-TOF
Ultima Mass Spectrometer (Waters Corporation, Manchester, UK). Samples
(1 L) were loaded on 5 mm 100 m ID Zorbax 300 SB C18 trap columns
(Agilent Technologies), and the peptides were separated on 15 cm 100 m ID
Atlantis C18 capillary columns at a flow rate of approx. 1 L/min. Solvent A
contained an aqueous 0.1% formic acid solution and solvent B contained 84%
ACN in 0.1% FA. The gradient consisted of isocratic conditions at 5% B for
10 min, a linear gradient to 30% B over 40 min, a linear gradient to 100% B over
10 min, and then a linear gradient back to 5% B over 5 min. MS analyses were
performed in positive mode using ESI. LC-MS was performed with the Q-TOF
Ultima operating in either (continuum) MS mode or in MS/MS mode for data
dependent acquisition (DDA) of MS/MS peptide fragmentation spectra.

3. Results
3.1. PAGE and immunoblotting analysis of donkey casein
The casein samples from 63 donkeys were analysed individually. In Fig. 1, the
two most common individual donkey casein samples Coomassie Brilliant Blue
stained (A) were compared to the cow counterparts on the basis of their relative
net charge at alkaline pH by PAGE analysis. The results obtained after specific
immunostaining with polyclonal antibodies against s1-, s2-, - and -CN (Fig.
1(B)(E), respectively), allowed to detect the four donkey casein fractions. By
comparing the two species each other, cow and donkey -CN exhibited a very
similar anodic mobility, while donkey s1 and s2-CN a lower and reversed anodic
mobility than its cow counterpart. The immunostained profiles showed: (i) an
overlap between s1 and s2 components and (ii) proteolysis of s2- and -CN,
giving rise to derived peptides having higher ( s2-CN derived fragments) and
lower anodic mobility (-CN derived fragments) than the parent protein.

Fig. 1.
PAGE analysis at pH 8.6 of the two common donkey's casein samples (1 and 2)
compared to cow casein, after CBB staining (A) and identification of the four casein
fractions by immunoblotting with polyclonal antibodies against s1 (B), s2 (C), (D) and CN(E).
Figure options

Considering the NPN peptide composition of donkey milk, consisting mainly in CN and s2-CN derived peptides generated by plasmin-like specific cleavages
(De Simone, submitted for publication), it can be supposed that the same
enzymes produced the above -CN and s2-CN derived insoluble peptides at pH
4.6.
Finally, as shown in CBB-stained PAGE analysis (Fig. 1A), it was very difficult to
detect the -CN components without specific immunostaining (Fig. 1E), likely
depending either on the low amount of this fraction in the milk or on its lower
reactivity to CBB staining. These results showed: (i) the lowest mobility of -CN
towards the anode compared to the other donkey casein fractions and (ii) a
different electrophoretic mobility of the two -CN (sample 1 > sample 2) likely due
to the occurrence of genetic polymorphism at this locus (Fig. 1E).
3.2. UTLIEF analysis and immunoblotting of donkey casein
The UTLIEF profiles of the two donkey casein samples, with the four casein
fractions labelled in brackets after specific immunostaining, are shown in Fig.
2(B)(E). The results underlined a more complex overlapping phenomenon, due

to the similar pI of casein fractions components, with respect to the above PAGE
analysis. In particular, donkey and cow -CN exhibit a very similar pI value, but a
different compositional heterogeneity (89 components in donkey vs. 23 in
cow), due to the different phosphorylation degrees of each donkey -CN
component and to the presence of deleted forms, as in mare counterpart [5].

Fig. 2.
UTLIEF analysis of the two common donkey casein samples, 1 and 2, compared to cow
casein as reference, after CBB staining (A) and identification of the four casein fractions
by immunoblotting with polyclonal antibodies against s1 (B), s2 (C), (D) and -CN (E).
Figure options

At this regard, Cunsolo et al. [27] found three phosphorylated components (5P,
6P and 7P) by MS analysis only in the full-length protein, while no information on
the phosphorylation degree has been provided for the shorter -CN form lacking
the -CN fragment (2734) [27].
Unlike PAGE, UTLIEF analysis allowed the separation of s1-CN and s2-CN
focalising at different pI values. However, s2-CN components partially overlapped
the -CN ones, and all these caseins fractions ( s1-CN, s2-CN, and -CN) were
overlapped by -CN components (1012) focalising throughout the pH gradient
( Fig. 2). Concordantly with PAGE analysis ( Fig. 1) s2-CN and -CN derived
fragments focused at lower and higher pI values than their parent casein,
respectively.
3.3. 2-DE analysis (PAGE UTLIEF)
To resolve the protein mixture components, the most commonly used 2-DE
analysis consists in the vertical PAGE-SDS separation of in-gel horizontally

focalised proteins, where the effectiveness of separation depends on the

differences among the protein MWs. In ruminants, it is known, casein MW
values, unlike these of whey proteins, are very close [33]. For this reason, the 2DE combination applied for donkey caseome resolution was the vertical
UTLIEF analysis of the two donkey caseins in-gel PAGE at pH 8.6 separated
(Fig. 3), as already reported for sheep and goat caseome [2] and [4]. By means
of specific immunostaining (Fig. 3(A)(D)) the four zones, each containing a
casein family (s1, s2, and -CN), were outlined with circles in Fig. 3.

Fig. 3.
2-D separation (PAGE UTLIEF) of donkey casein samples 1 and 2 showed in Fig.
1 and Fig. 2. Staining with CBB and with antibodies preparation raised against s2 (A),
s1 (B), (C) and -CN (D).
Figure options

These results allowed to conclude:

in both samples the s1-CN composition consisted of two doublet bands (a,
b and c, d), having decreasing pI values towards the anode;

in sample 2, the number of -CN components was doubled (six) with

respect to sample 1 (three), likely due to occurrence of genetic
polymorphism;

the most heterogeneous donkey casein fraction was -CN, since at least
11 components were specifically immunostained in the 2-D map, focalising
throughout the entire working pH range.
3.4. LC/MS analysis of donkey caseins
Comparison of the HPLC profiles of samples 1 and 2 (Fig. 4) showed that the
latter was more heterogeneous than the former for the presence of the additional
peak 4* eluted at shorter elution time than peak 4, common to the two samples.
In Table 1, the MW of eluted components, determined by means of LC/ESI/MS
analysis either before or after PA action, was reported. These data showed the
same eluted components in the four common peaks. In particular in peak 1 -CN

fragments, originated by hydrolysis of both full protein length (226 a.a.) and from
its deleted form lacking peptide (2734) [27], were identified. The MW values of
components eluted in peak 2 (Table 1) and those of their derived tryptic peptides
(results not shown), both fitting with the reported s1-CN sequence [28], allowed
to conclude the presence of donkey s1-CN at different phosphorylation degrees
(5P and 6P) as well as their counterpart deleted of a Gln residue (128 Da). On
the same way in peak 3 donkey s1-CN, deleted of fragment s1(3438; 621 Da)
carrying 6P and 5P, was identified together with its deleted form lacking both Gln
residue and s1 (3438) fragment. These results suggested that also in donkey
the compositional heterogeneity of s1-CN is depending on discrete
phosphorylation (5P, 6P) and on the presence of non-allelic forms generated by
RNA incorrect splicing, as in sheep and goat caseins [8] and [10].

Fig. 4.
UV profile ( = 220 nm) of LC chromatogram of donkey whole casein samples 1 and 2.
Figure options

Table 1.
Donkey casein identification in samples 1 and 2 by LC/ESI-MS analysis. The additional
peak 4* is only referred to sample 2.
Molecular mass (Da)
Retention time
(min)
Peak
Protein identificationa
Native Dephosphorylated
19.2
1
16,959
-CN fragment (922 Da) nPb
17,039
-CN fragment (922 Da) nPb
17,119
-CN fragment (922 Da) nPb
18,042
-CN fragment nPb
18,120
-CN fragment nPb
18,202
-CN fragment nPb

Molecular mass (Da)

Retention time
(min)

Peak

Protein identificationa
Native
18,279

21.6

Dephosphorylated
-CN fragment nPb

24,761
24,681
24,283
24,890
24,810

22.1

24,412

24,269
24,189
23,786
24,141
24,061

28.0

4*

25,034
25,113
25,193
26,050
26,131

28.8

25,013
25,099
25,178
25,938
26,023
26,101

23,662

24,632
25,570

24,619
25,542

s1-CN 6P (128 Da)

s1-CN 5P (128 Da)
s1-CN (128 Da)
s1-CN 6P
s1-CN 5P
s1-CN
s1-CN 6P (621 Da)
s1-CN 5P (621 Da)
s1-CN (621 Da)
s1-CN 6P (621 Da) (128 Da)
s1-CN 5P (621 Da) (128 Da)
s1-CN (621 Da) (128 Da)
-CN* 5P (923 Da)
-CN* 6P (923 Da)
-CN* 7P (923 Da)
-CN* (923 Da)
-CN* 6P
-CN* 7P
-CN*
-CN 5P (923 Da)
-CN 6P (923 Da)
-CN 7P (923 Da)
-CN (923 Da)
-CN 5P
-CN 6P
-CN 7P
-CN

Protein identification is based on 1 and 2 or more peptides.

b

nP, unknown number of phosphate groups.

Table options

In peak 4, the full-length -CN and its deleted form lacking of fragment (2734),
both having 5, 6 and 7 P/mole, eluted. These forms have been already detected
by Cunsolo et al. [27] only in the full-length protein. These data, in addition to
pI theoretical values of the two -CN forms ( Table 2), allowed us to identify them

on the 2-DE map ( Fig. 3) as the main and the minor components of donkey CN, respectively. In peak 4*, a novel -CN was found having a MW 28 mass
units higher than the most common -CN present in peak 4. This showed the
occurrence of genetic polymorphism at this casein locus.
Table 2.
Experimentally measured molecular mass of full-length and deleted forms occurring in CN. The theoretical molecular weight and isoelectric points are calculated using the
algorithm from ExPASy's Scansite Compute pI/MW program [34].
Measured molecular
Theoretical molecular
P/mole Length, a.a. mass (Da)
Protein
mass (Da)
pI
7
226
26,101
(-CN 7P
26,075
4.74
6
226
26,023
(-CN 6P
25,997
4.82
5
226
25,943
(-CN 5P
25,919
4.91
7

218

25,178

218

25,099

218

25,013

(-CN 7P
(923)
(-CN 6P
(923)
(-CN 5P
(923)

25,151

4.64

25,073

4.72

24,995

4.80
Table options

Although no HPLC-eluted component was detected to confirm the presence of

donkey s2-CN, it was possible to identify this fraction by MS analysis of the
excised spots 1, 2 and 3 from 2-DE map (sample 2 in Fig. 3). As an example,
in Fig. 5 the MALDI MS spectrum of the component excised from spot 1, before
and after AP action, was reported. These procedure (results in Table 3) showed
that donkey s2-CN occurred in spots 1, 2 and 3 with 10, 11 and 12 P/mole
respectively. These assignments were confirmed by MS analysis of tryptic
digests (Table 4). The alignment of the tryptic peptides gave a 90% coverage
corresponding to a protein 221 a.a. in length, in agreement with the cDNA
sequence reported by Ramunno [34].

Fig. 5.
MALDI-TOF/MS spectrum relative to spot 1 excised from the 2-D gel of donkey casein
sample 2 shown inFig. 3 before (A) and after (B) AP action.
Figure options

Table 3.
MALDI TOF/MS analysis relative to spots 3, 2 and 1 excised from the 2-D gel of donkey's
casein sample 2 shown in Fig. 3. A database search using the Mascot software retrieved
phosphopeptides from in-gel digest of each band with trypsin before and after AP action.
Molecular mass (Da)
Spot
1

Protein identification
Native
26,829

Dephosphorylated
26,029

s2-CN 10P
s2-CN

26,029

s2-CN 11P
s2-CN

26,029

s2-CN 12P
s2-CN

26,909

26,989

Table 4.

In-gel s2-CN tryptic digest of spot 1 excised from the 2-D gel of donkey casein sample 2
shown in Fig. 3, by ESI-Q-TOF MS analysis.
Identification
115147
3763 + 2P
193219
3763
122 + 2P
125148
89109
125148 + 1
P
2646
2646 + 1P
5672 + 6P
101114
111124
822
1023
209221
3446 + 1P
149161
174184
3446
209219
918 + 2P
112121
115124
918
185192
7380
4755
17

Peptide sequence s2 casein

IVLTPWDQTKTGASPFIPIVNTEQLFTSEEIPK
CSTSCEEATRNINEMESAKFPTEVYSS
IVHQHQTTMDPQSHSKTNSYQIIPVLR
CSTSCEEATRNINEMESAKFPTEVYSS + 2P
KHNMEHRSSSEDSVNISQEKFK + 2P
TGASPFIPIVNTEQLFTSEEIPKK
KQLNKINQFYEKLNFLQYLQA
TGASPFIPIVNTEQLFTSEEIPKK + 1P

Theoretical
mass
3701.25
3174.29
3159.59
3014.29
2777.85
2647.04
2644.09
2727.04

Measured
mass
3702.25
3177.17
3161.03
3015.81
2777.89
2646.03
2646.83
2728.50

YVVIPTSKESICSTSCEEATR
YVVIPTSKESICSTSCEEATR + 1P
FPTEVYSSSSSSEESAK + 6P
LNFLQYLQALRQPR
RQPRIVLTPWDQTK
SSSEDSVNISQEKFK
SEDSVNISQEKFKQ
TNSYQIIPVLRYF
ESICSTSCEEATR + 1P
TVDMESTEVVTEK
LLNKINQYYEK
ESICSTSCEEATR
TNSYQIIPVLR
SSEDSVNISQ + 2P
QPRIVLTPWD
IVLTPWDQTK
SSEDSVNISQ
FTLPQYFK
FPTEREEK
NINEMESAK
KHNMEHR

2417.71
2497.71
2301.88
1760.08
1738.03
1684.79
1638.46
1613.89
1495.52
1467.62
1425.66
1415.52
1303.53
1225.06
1224.43
1200.41
1065.06
1043.24
1035.13
1035.15
951.08

2419.03
2499.33
2301.50
1761.10
1738.83
1685.78
1639.64
1613.97
1496.50
1468.60
1427.77
1415.50
1304.87
1226.54
1224.49
1200.85
1066.67
1043.70
1036.68
1036.55
951.61
Table options

4. Discussion
This work reports for the first time the experimental evidence of the four casein
fractions in donkey milk, by coupling one-(PAGE, UTLIEF) and two-dimensional
(PAGE vs.UTLIEF) electrophoretic techniques with CBB and specific
immunostaining to mass spectrometry analysis (MALDI-TOF MS, ESI-Q-TOF MS
and LC-ESI-MS).

In previous studies [1] and [3], one-dimensional electrophoresis did not provide
efficient separation of caseins because of their partial overlapping. On the other
side, vertical UTLIEF analysis of proteins horizontally in-gel PAGE separated
after immunostaining was the most resolving 2-DE. This technique revealed a
characteristic protein pattern for the two donkey casein samples consisting in: (i)
at least four s1-CN main components; (ii) three -CN components, doubled in
sample 2; (iii) three s2-CN components and; (iv) eleven -CN components.
By means of LC-ESI-MS analysis, the heterogeneity of s1-CN was assigned to
either discrete phosphorylation (5, 6 and 7 P/mole) or non-allelic deleted forms
generated by incorrect RNA splicing, as already shown in the homologous goat
and sheep casein [7],[8] and [9]. The full-length -CN and its deleted form were
both found equally phosphorylated with 5, 6 and 7 P/mole and a novel -CN
variant was found in sample 2. The molecular mass determined by ESI/MS
before and after dephosphorylation of the variant showed a MW 28 mass units
higher than the most common -casein and the same phosphorylation pattern.
Donkey s2-CN contained 10, 11 and 12 P, and its structural characterization
confirmed for the first time the correctness of the cDNA-derived sequence [34].
Thus, the combined methodologies applied in this study were able to disclose the
heterogeneity of each donkey casein fraction due to post-translational
phenomena, as well as the occurrence of polymorphism at -CN locus. These
results may be useful when programming breeding strategies for preservation
and selection of donkey biodiversity, aimed to production of donkey milk for
human consumption. In fact, the variability associated with genetic polymorphism
complemented with a major or minor complexity in the expression of single
casein components may influence qualitative and quantitative donkey milk
composition and, as a consequence, its allergenic properties.

Acknowledgements
This research was supported by the Italian Ministry of Agriculture, MIPAF
(SELMOL project).
This work was carried out with equipment available within the Rete di
Spettrometria di Massa (Contract Fondo Europeo di Sviluppo Regionale
94.05.09.103, AINCO 94.IT.16.028).

References

1.
o

[1]

L. Chianese, G. Garro, R. Mauriello, P. Laezza, P. Ferranti, F. Addeo

J. Dairy Res., 63 (1996), p. 49

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (61)

2.
o

[2]

C. Bevilacqua, P. Ferranti, G. Garro, C. Veltri, R. Lagonigro, C. Leroux, E.

Pietrol, F. Addeo, F. Pilla, L. Chianese, P. Martin

Eur. J. Biochem., 269 (2002), p. 1293

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (45)

3.
o

[3]

L. Chianese, M. Quarto, F. Pizzolongo, M.G. Calabrese, S. Caira, R. Mauriello, S.

De Pascale, F. Addeo

Int. Dairy J., 19 (2009), p. 181

Article
|
PDF (695 K)
|
View Record in Scopus
|
Citing articles (14)

4.
o

[4]

G. Mamone, S. Caira, G. Garro, M.A. Nicolai, P. Ferranti, G. Picariello, A.

Malorni, L. Chianese, F. Addeo

Electrophoresis, 24 (2003), p. 2824

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (37)

5.
o

[5]

G. Miranda, M.F. Mah, C. Leroux, P. Martin

Proteomics, 4 (2004), p. 2496

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (69)

6.
o

[6]

C. DAmbrosio, S. Arena, A.M. Salzano, G. Renzone, L. Ledda, A. Scaloni

Proteomics, 8 (2008), p. 3657

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (48)

7.
o

[7]

P. Ferranti, A. Malorni, G. Nitti, P. Laezza, R. Pizzano, L. Chianese, F. Addeo

J. Dairy Res., 62 (1995), p. 281

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (91)

8.

[8]

P. Ferranti, F. Addeo, A. Malorni, L. Chianese, C. Leroux, P. Martin

Eur. J. Biochem., 249 (1997), p. 1

View Record in Scopus

|
Citing articles (49)

9.
o

[9]

P. Ferranti, L. Chianese, A. Malorni, F. Migliaccio, V. Stingo, F. Addeo

J. Agric. Food Chem., 46 (1998), p. 411

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (21)

10.
o

[10]

P. Ferranti, S. Lilla, L. Chianese, F. Addeo

J. Protein Chem., 18 (1999), p. 595

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (28)

11.
o

[11]

M. Malacarne, F. Martuzzi, A. Summer, P. Mariani

Int. Dairy J., 12 (2002), p. 869

Article
|
PDF (136 K)
|
View Record in Scopus
|

Citing articles (106)

12.
o

[12]

C. Chiavari, F. Coloretti, M. Nanni, E. Sorrentino, L. Grazia

Lait, 85 (2005), p. 481

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (26)

13.
o

[13]

E. Pagliarini, G. Solaroli, C. Peri

Ital. J. Food Sci., 4 (1993), p. 323

View Record in Scopus

|
Citing articles (41)

14.
o

[14]

L. Businco, P.G. Giampietro, P. Lucenti, F. Lucaroni, C. Pini, G. Di Felice, P.

Lacovacci, C. Curadi, M. Orlandi

J. Allergy Clin. Immunol., 10 (2000), p. 1031

Article
|
PDF (242 K)
|
View Record in Scopus
|
Citing articles (105)

15.
o

[15]

A. Carroccio, F. Cavataio, G.D. Montaldo, D. Amico, L. Alabrese, G. Iacono

Clin. Exp. Allergy, 30 (2000), p. 1597

View Record in Scopus

|
Citing articles (97)
16.
o

[16]

G. Monti, E. Bertino, M.C. Muratore, A. Coscia, F. Cresi, L. Silvestro, C. Fabris,

D. Fortunato, M.G. Giuffrida, A. Conti

Pediatr. Allergy Immunol., 18 (2007), p. 258

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (77)

17.
o

[17]

X. Mao, J. Gu, Y. Sun, S. Xu, X. Zhang, H. Yang, F. Ren

Int. Dairy J., 19 (2009), p. 703

Article
|
PDF (583 K)
|
View Record in Scopus
|
Citing articles (16)

18.
o

[18]

H.Y. Guo, K. Pang, X.Y. Zhang, L. Zhao, S.W. Chen, M.L. Dong, F.Z.J. Ren

J. Dairy Sci., 90 (2007), p. 1635

Article
|
PDF (406 K)
|
View Record in Scopus
|
Full Text via CrossRef
|

Citing articles (65)

19.
o

[19]

A. Conti, L. Napolitano, P. Lai, W. Pinna, J. Godovac-Zimmerman

Milchwissenschaft, 44 (1989), p. 138

o
20.
o

[20]

M.G. Giuffrida, A. Cantisani, L. Napolitano, A. Conti, J. Godovac-Zimmermann

Biol. Chem. Hoppe-Seyler, 373 (1992), p. 931

J. Godovac-Zimmermann, A. Conti, L. James, L. Napolitano

Biol. Chem. Hoppe-Seyler, 369 (1988), p. 171

View Record in Scopus

1.

|
Full Text via CrossRef
|
Citing articles (17)
2.
o

[22]

J. Godovac-Zimmermann, A. Conti, M. Sheil, L. Napolitano

Biol. Chem. Hoppe-Seyler, 371 (1990), p. 871

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (20)

3.
o

[23]

V. Cunsolo, R. Saletti, V. Muccilli, S. Foti

J. Mass Spectrosc., 42 (2007), p. 1162

View Record in Scopus

|

Citing articles (19)

4.
o

[24]

V. Cunsolo, A. Costa, R. Saletti, V. Muccilli, S. Foti

Rapid Commun. Mass Spectrosc., 21 (2007), p. 1438

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (10)

5.
o

[25]

M. Herrouin, D. Molle, J. Fauquant, F. Ballestra, J.L. Maubois, J. Leonil

J. Protein Chem., 19 (2000), p. 105

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (25)

6.
o

[26]

A. Criscione, V. Cunsolo, S. Bordonaro, A.M. Guastella, R. Saletti, A. Zuccaro, G.

DUrso, D. Marletta

Int. Dairy J., 19 (2009), p. 190

Article
|
PDF (481 K)
|
View Record in Scopus
|
Citing articles (27)

7.
o

[27]

V. Cunsolo, E. Cairone, R. Saletti, V. Muccilli, S. Foti

Rapid Comm. Mass Spectrosc., 23 (2009), p. 1907

View Record in Scopus

o
|

Full Text via CrossRef

|
Citing articles (14)
8.
o

[28]

V. Cunsolo, E. Cairone, D. Fontanini, A. Criscione, V. Muccilli, R. Saletti, S. Foti

J. Mass Spect., 44 (2009), p. 1742

View Record in Scopus

|
Citing articles (18)

9.
o

[29]

E. Bertino, D. Gastaldi, G. Monti, C. Baro, D. Fortunato, L. Perono Garoffo, A.

Coscia, C. Fabris, M. Mussap, A. Conti

Front Biosci., E2 (2010), p. 526

View Record in Scopus

|
Full Text via CrossRef
|
Citing articles (8)

10.
o

[30]

R. Aschaffenburg, J. Drewry

15th International Dairy Congress, vol. 3 (1959), p. 1631

o
11.
o

[31]

I. Krause, J. Buchberg, G. Weib, M. Pfluger, H. Klostermeyer

Electrophoresis, 9 (1988), p. 609

View Record in Scopus

|
Full Text via CrossRef

|
Citing articles (27)
12.
o

[32]

S.L. Cohen, B.T. Chait

Anal. Biochem., 247 (1997), p. 257

Article
|
PDF (156 K)
|
View Record in Scopus
|
Citing articles (146)

13.
o

[33]

H.E. McSweeney

P.F. Fox, P.L.H. McSweeney (Eds.), Advanced Dairy Chemistry. 1. Proteins,

Kluwer Academic/Plenum publishers, USA (2003), p. 139

o
14.
o

[34]

L. Ramunno

Direct Submission in EMBL/GenBank/DDBJ Databases

(December 2008)

o
Corresponding author. Tel.: +39 81 2539011; fax: +39 81 7762580.
Copyright 2010 Elsevier B.V. All rights reserved.