Structural proteome of human colostral fat globule membrane proteins

Donatella Fortunato1

Maria Gabriella Giuffrida1

Maria Cavaletto2

Lorenza Perono Garoffo1

Giuseppina Dellavalle1

Lorenzo Napolitano1

Carlo Giunta3

Claudio Fabris4

Enrico Bertino4

Alessandra Coscia4

Amedeo Conti1

1CNR, Institute of Scienceof Food Production,Turin, Italy

2Dipartimento di Scienzee Tecnologie Avanzate,University of Piemonte Orientale,Alessandria, Italy

3Dipartimento di BiologiaAnimale e dell’Uomo,University of Turin,Turin, Italy

4Cattedra di Neonatologia,University of Turin,Turin, Italy

Structural proteome of human colostral fat globulemembrane proteins

Milk fat globule membrane (MFGM) contains proteins derived from the apical mem-brane of secreting epithelial cells of the mammary gland. Between 2–4% of totalhuman milk protein content is associated with the fat globule fraction, as MFGM pro-teins. While MFGM proteins have very low classical nutritional value, they play im-portant roles in various cell processes and defence mechanisms for the newborn.To date, fewer than 30 human MFGM proteins have been identified and characterized,either by immunological methods or by Edman sequencing and mass spectrometry.This study aimed to update the structural proteome of human colostral MFGM proteinsand to create an annotated two-dimensional electrophoresis (2-DE) MFGM proteindatabase available on-line. More than one hundred 2-DE spots derived from humancolostral MFGM proteins were investigated by matrix-assisted laser desorption/ioniza-tion-time of flight mass spectrometry and proteins were identified by three differentsoftware packages available on the web (PeptIdent, MS-Fit and ProFound); uncertainidentifications were solved by nanoelectrospray ionization-ion trap mass spectrometryusing SEQUESTsoftware.

Keywords: Bidimensional electrophoresis / Human milk fat globule membrane proteins / Identifi-cation / Mass spectrometry PRO 0367

1 Introduction

Colostral milk is generally defined as the mammary secre-tion collected in the first 2–3 days of lactation [1] or within1 week after parturition [2]; colostrum can be regarded asa casein micelles-free milk [3]. Despite the lack of caseinmicelles, the total protein content in human colostrum ishigher than in mature milk [4], mainly due to the very highcontent of immunocompetent or protective componentssuch as secretory immunoglobulin A (IgA) and lactoferrin[5], among the “soluble” whey proteins. Many other milkcomponents, such as glycoconjugates, are now recog-nized as inhibiting pathogens [6] and several are eithersoluble or bound to the milk fat globule membranes(MFGMs). The fat content of human colostrum is lowerthan that of mature milk but, in contrast, because of smal-ler diameter fat globules, the total surface area of the glo-bules is 3.3 m2/g in human colostrum compared to 1.4 m2/g in mature milk [7], thus enhancing contact between theMFGM proteins (MFGMPs) and their ligands. MFGMPsderive from the apical membrane of secreting epithelial

cells of the mammary gland and they represent 2–4% ofthe total protein content of human colostrum. Up to now,lessTo date fewer than 30 human MFGMPs have beenidentified and characterized, either by immunologicalmethods [8] or by Edman sequencing and mass spec-trometry [9]; they play important roles in various cell pro-cesses and defence mechanisms, e.g. mucins, lacthad-herin, adipophilin, butyrophilin, HLA class I proteins,rather than possessing classical nutritional value [10].

Formula-fed babies, and even babies fed in the veryfirst days of life with human milk from a Lactarium, com-pletely lack the immediate and long-term beneficialeffects of human colostral MFGMPs. To produce morefunctionally adequate formulas or fortified human milks,further knowledge of human MFGMP composition is re-quired.

This study aimed to update the structural proteome ofhuman colostral MFGMPs and to create an annotated2-DE MFGMP database available on-line. One hundredand seven 2-DE spots derived from human colostralMFGMPs were investigated by MALDI-TOF mass spec-trometry, and the proteins were identified by three differ-ent software packages available on the web (PeptIdent,MS-Fit and ProFound); uncertain identifications weresolved by nanoESI-ion trap mass spectrometry usingSEQUEST software.

Correspondence: Dr. Maria Gabriella Giuffrida, CNR c/o Bioin-dustry Park – via Ribes, 5 10010, Colleretto Giacosa (Torino), ItalyE-mail: [email protected]: +39-0125-564942

Abbreviation: MFGM, milk fat globule membrane

Proteomics 2003, 3, 897–905 897

2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 0173-0835/03/0606–897 $17.50�.50/0

DOI 10.1002/pmic.200300367

898 D. Fortunato et al. Proteomics 2003, 3, 897–905

2 Materials and methods

2.1 Preparation of the human milk fat globule

Human colostrum was collected from several healthymothers and pooled. A protease inhibitor (Complete;Roche Diagnostics, Mannheim, Germany) was immedi-ately added to the specimens to prevent proteolysis.Samples were centrifuged at 2000 g for 30 min at 10�Cto remove cells and then centrifuged again at 189 000 gfor 70 min at 6�C to remove caseins [11]. Fat globuleswere recovered in the supernatant layer and washedthree times with 0.9% NaCl by centrifugation at 3000 gfor 30 min at 10�C to remove water soluble proteins.Washed globules (WG) were stored at �20�C until use.

2.2 Sample preparation for 2-DE

To extract the MFGM proteins, 150 �L WG were mixed 2:3with an SDS-containing solution (63 mM Tris-HCl, pH 6.8,2% SDS, 5% �-mercaptoethanol) following Laemmli [12],incubated at room temperature for 60 min and centri-fuged at 10 000 g for 10 min. After removing the floatingcream layer, the supernatant was subjected to precipita-tion with methanol and chloroform following Wessel et al.[13]. A solution of 7 M urea, 2 M thiourea, 4% CHAPS, 1%Triton X-100, 20 mM Tris, 1% DTT and 0.5% IPG bufferwas then used as sample solution to solubilize the preci-pitated proteins.

2.3 2-DE

IEF was performed using ready-to-use Immobiline Dry-Strips, linear pH gradient 3–10 or 4–7, length 18 cm(Amersham Biosciences Uppsala, Sweden) and the in-gel sample rehydration method [14]. IEF was run on anIPGphore unit (Amersham Biosciences) at 20�C constanttemperature and 8000 V for 99 000 Vh. After IEF the IPGgel strips were incubated at room temperature for 15 minin 6 M urea, 30% w/v glycerol, 2% w/v SDS, 2% DTT,5 mM Tris-HCl, pH 8.6. A second equilibration step wascarried out for 15 min in the same solution, with theexception that DTTwas replaced by 2.5% iodoacetamideplus a trace of bromophenol blue as tracking dye. Thestrips were sealed at the top of the 1.0 mm vertical sec-ond dimensional gels (Ettan DALT II system, AmershamBiosciences) with 0.5% agarose in 25 mM Tris, 192 mM

glycine, 0.1% SDS, pH 8.3.

SDS-PAGE was carried out on homogeneous runninggels 11.7%T, 2.7%C (Duracryl; Genomic Solutions, AnnArbour, MI, USA). The running buffer was 25 mM Tris,192 mM glycine, 0.1% SDS, pH 8.3 and running con-

ditions were 15�C, 5 watt/gel for 45 min and 13 watt/geluntil the bromophenol blue reached the bottom of thegels. Molecular weight markers used were Low MolecularWeight Electrophoresis Calibration Kit (Amersham Bio-sciences). Gels were automatically stained using the Pro-cessor Plus (Amersham Biosciences) with Brilliant Blue Gcolloidal (Sigma, St. Louis, MO, USA) following the manu-facturer’s instructions.

2.4 Image acquisition and data analysis

2-DE gels were digitized at 600 dots per inch (Image-Scanner; Amersham Biosciences). Image analysis wasperformed using the software ImageMaster 2D Elite V3.1(Amersham Biosciences).

2.5 In-gel digestion

Spots of interest were cut from the gel and destainedovernight with a solution of 25 mM ammonium bicarbo-nate, 50% acetonitrile. The proteins were digested in-gelwith trypsin (Promega, Madison, WI, USA) as describedby Hellmann [15].

2.6 MS analysis

For MALDI-TOF MS, 0.5 �L of each peptide mixture wasapplied to a target disk and allowed to air-dry. Sub-sequently, 0.5 �L of the matrix solution (1% w/v �-cyano-4-hydroxycinnamic acid in 30% acetonitrile, 0.1% TFA)was applied to the dried sample and again allowed todry. Spectra were obtained using a Bruker ReflexIIIMALDI-TOF spectrometer (Bremen, Germany). To inter-pret the MS spectra of protein digests, PeptIdent (http://www.expasy.ch/tools/peptident.html), MS-Fit (http://falcon.ludwig.ucl.ac.uk/MS-Fit.html) and ProFound(http://prowl.rockefeller.edu/cgi-bin/ProFound) softwarepackages were used. For the MS/MS sequencing ex-periment, the tryptic peptide mixture was desalted andconcentrated by ZipTipC18 devices (Millipore, Bedford,MA, USA). After elution with 50% methanol, 1% formicacid, 4 �L of the tryptic peptide mixture was applied intoa gold coated borosilicate capillary (New Objective, Cam-bridge, MA, USA) and analyzed by an LCQ ThermoFinni-gan (San Jose, CA, USA) ion trap mass spectrometerfitted with a nanoESI source. The capillary voltage wasset at 46 V and spray voltage at 1.6 kV. The most stablesignals from the full-scan mass spectrum were trappedand fragmented by low-energy CID with normalized colli-sion energy ranging from 22–26%. MS/MS spectra wereautomatically analyzed using the SEQUEST program.

Proteomics 2003, 3, 897–905 Human colostral fat globule membrane proteins 899

3 Results and discussion

3.1 Improvement of the 2-DE map

In a previous paper we published the 2-DE map of humancolostral MFGM proteins and the identification of themost abundant proteins contained in 23 spots [9]. Ouraim is to build up a human colostral MFGM database atthe sensitivity level of colloidal Blue Coomassie staining.The present database includes the protein identificationof 107 spots obtained by MS analysis. Human MFGMswere obtained, as previously reported elsewhere [9],from healthy donors. The protein extract from 150 �L ofwashed globules was loaded each an 18 cm length IPGstrip. The broad pH 3–10 gels were used to obtainthe general map, while the pH 4–7 gels were used for amore precise detection of acidic proteins. To improveprotein focalization, an IEF program leading to 99 000 Vhwas used. Figure 1 shows the representative map ofhuman coloastral MFGM proteins and Fig. 2 the corre-sponding “zoomed” pH 4–7 MFGM map. Spot patterns

of different experimental gels were highly reproduciblebetween samples. Compared to the earlier part of thisstudy, a great improvement was obtained in both proteinseparation and detection. To achieve the identification oflow abundance spots, the gel shown in Fig. 1 was over-loaded and the major spots were intensely stained.

3.2 Protein identification by MS analysis

Protein identification was achieved by MALDI-MS and thepeptide mass fingerprinting method (PMF) [16]. Thismethod is fast and simple and is the method of choicefor large-scale protein identification studies. The specific-ity of the enzyme used and the mass accuracy of MSmeasurements are the major positive points of themethod. We used trypsin, the most commonly usedenzyme for in-gel digestion, which cuts rather specificallyafter lysine and arginine residues and affords small pep-tides, which can be efficiently eluted from the gel. More-over, some of its autoproteolytic fragments have been

Figure 1. 2-DE map of humancolostrums MFGMP on pH 3–10,18 cm IPG DryStrips.


Figure 2. 2-DE map of humancolostrums MFGMP pH 4–7,18 cm IPG DryStrips.

used for internal calibration of the mass spectra, allowingthe use of very narrow windows of mass tolerance(� 30 ppm) to be used in data searches. The search cri-teria used were a minimum of five matching peptides witha mass accuracy of better than 30 ppm. In most cases,the identification was based on more than nine matchingpeptides, giving a good identification confidence, al-though Fountoulakis and co-workers [17] reported thatproteins with a molecular mass ranging between 10–20 kDa are usually identified with a smaller number ofpeptides. The sequence coverage varied from 11–76%of the whole amino acid sequence and generally wasmore than 20% (Table 1). For unambiguous identificationof proteins with less than five matching peptides (spots89, 97, 101 and 107) or identification with six or sevenmatching peptides but mass accuracy above 30 ppm(spots 22, 38, 54, 67, 96, 98 and 103), ESI-IT/MS/MSwas used.

The 2-DE map of colostral MFGM proteins (Fig. 1) consistsof protein spots ranging from mass 9300–92 000 Da and pI

from 4 to 9. The 107 proteins identified in our studiesare derived from only 39 genes, or genes families. Manyproteins were found in multiple spots with different pIsand molecular masses. For human proteins it is morelikely that they represent post-translational modificationsof the polypeptides, rather than artifacts of the 2-DE (suchas, for example, carbamylation). Proteins showing a Mr

below the theoretical (from DNA sequence), may be pro-teolitic products, probably already generated in the mam-mary gland cells and with a possible biological signifi-cance. Considering the map of Fig. 1, lactadherin wasidentified in nine spots, butyrophilin in seven spots, lacto-ferrin in five spots, while adipophilin and carbonic anhy-drase were both identified in four spots each.

Spots 35 to 79 were identified in the pH 4–7 map (Fig. 2),where many isoforms for butyrophilin, apolipoproteinA-IV and clusterin were identified. Butyrophilin, one ofthe major proteins of the MFGM, was present in multipleglycoforms, as described in detail in our recent paper[18].


Table 1. List of identified human colostral MFGM proteins

No. Entry Full name Function Mr pI Matchingpeptides

Sequencecoverage

Identificationconfirmed byMS/MS

Figure

1a) Q08431 Lactadherin Phospholipid binding 40 791 8.22 9 27 – 12a) Q08431 Lactadherin Phospholipid binding 40 791 8.22 8 25 – 13a) Q08431 Lactadherin Phospholipid binding 40 791 8.22 4 15 – 14a) Q08431 Lactadherin Phospholipid binding 40 791 8.22 7 21 – 15a) Q99541 Adipophilin Triacylglycerol deposition 48 075 6.34 12 43 – 16a) Q99541 Adipophilin Triacylglycerol deposition 48 075 6.34 4 23 – 17a) Q99541 Adipophilin Triacylglycerol deposition 48 075 6.34 7 21 – 18a) Q99541 Adipophilin Triacylglycerol deposition 48 075 6.34 7 19 – 19a) P23280 Carbonic anhydrase Reversible hydration

of carbon dioxide35 377 6.53 4 16 – 1

10a) P23280 Carbonic anhydrase Reversible hydrationof carbon dioxide

35 377 6.53 9 39 – 1


35 377 6.53 6 33 – 1


35 377 6.53 5 17 – 1

13a) Q13410 Butyrophilin Fat globules secretion 56 196 5.32 – – – 114a) Q13410 Butyrophilin Fat globules secretion 56 196 5.32 – – – 115a) P30101 Disulfide isomerase Protein destination 54 265 5.61 – – – 116a) Q13410 Butyrophilin Fat globules secretion 56 196 5.32 – – – 117a) Q13410 Butyrophilin Fat globules secretion 56 196 5.32 – – – 118a) Q13410 Butyrophilin Fat globules secretion 56 196 5.32 – – – 119a) P10909 Clusterin Apo J

(�-chain)Apoptosis 24 197 6.27 – – – 1

20a) P10909 Clusterin Apo J(�-chain)

Apoptosis 24 197 6.27 – – – 1

21a) P05814 �-casein Micelles stability 25 382 5.5 – – – 122 P05814

P08729�-caseinkeratin type II (CK 7)

Micelles stabilityCytoskeletal structure

25 38251 286

5.55.4

410

1719

YES–

1

23a) P00709 �-lactalbumin Lactose synthesis 16 660 4.80 7 44 – 124 Q08431 Lactadherin Phospholipid binding 40 791 8.22 21 69 – 125 Q08431 Lactadherin Phospholipid binding 40 791 8.22 25 76 – 126 Q08431 Lactadherin Phospholipid binding 40 791 8.22 17 58 – 127 P15328 Folate-binding protein Folate-derivatives transport 24 656 8.32 13 41 – 128 P15328 Folate-binding protein Folate-derivatives transport 24 656 8.32 9 23 – 129 P15328 Folate-binding protein Folate-derivatives transport 24 656 8.32 6 23 – 130 Q9H1Z3 Lactoferrin Iron transport 76 366 8.52 20 45 – 131 Q9H1Z3 Lactoferrin Iron transport 76 366 8.52 30 58 – 132 Q0H1Z3 Lactoferrin Iron transport 76 366 8.52 24 47 – 133 Q9H1Z3 Lactoferrin Iron transport 76 366 8.52 31 62 – 134 Q9H1Z3 Lactoferrin Iron transport 76 366 8.52 30 57 – 135 P11021 GRP 78 Protein destination 70 479 5.01 13 22 – 236 P14625 Endoplasmin Protein destination 92 740 4.8 23 35 – 137 P07237 Disulfide isomerase Protein destination 55 294 4.69 21 39 – 238 Q13410

P01833ButyrophilinPolymeric Ig receptor

Fat globules secretionIg superfamily

56 19681 379

5.325.59

710

1627

YES 2


Table 1. Continued


Sequencecoverage


Figure

39 Q13410 Butyrophilin Fat globules secretion 56 196 5.32 19 31 – 240 P11021 GRP 78 Protein destination 70 479 5.01 14 27 – 241 P30101 Disulfide isomerase Protein destination 54 265 5.61 16 24 – 242 P30101 Disulfide isomerase Protein destination 54 265 5.61 17 26 – 243 Q13410 Butyrophilin Fat globules secretion 56 196 5.32 12 24 – 244 Q13410 Butyrophilin Fat globules secretion 56 196 5.32 11 26 – 245 Q13410 Butyrophilin Fat globules secretion 56 196 5.32 18 38 – 246 Q13410 Butyrophilin Fat globules secretion 56 196 5.32 9 17 – 247 Q13410 Butyrophilin Fat globules secretion 56 196 5.32 7 14 – 248 Q13410 Butyrophilin Fat globules secretion 56 196 5.32 20 39 – 249 Q13410 Butyrophilin Fat globules secretion 56 196 5.32 15 27 – 250 P02768 Serum Albumin Plasma protein 66 472 5.67 5 11 – 251 Q15084 PDI related protein Protein destination 46 171 4.95 7 25 – 252 O60664 Cargo selection

proteinReceptor 47 033 5.30 7 24 – 2

53 O60664

Q13410

Cargo selectionproteinButyrophilin

Receptor

Fat globules secretion

47 033

56 196

5.30

5.32

9

7

23

13

–

YES54 O60664

Q13410

Cargo SelectionproteinButyrophilin

Receptor

Fat globules secretion

47 033

56 196

5.30

5.32

16

7

43

15

–

YES

2

55 P06727 Apolipoprotein A-IV Lipoprotein metabolism 43 375 5.18 13 37 – 256 P06727 Apolipoprotein A-IV Lipoprotein metabolism 43 375 5.18 18 47 – 257 P06727 Apolipoprotein A-IV Lipoprotein metabolism 43 375 5.18 13 38 – 258 P06727 Apolipoprotein A-IV Lipoprotein metabolism 43 375 5.18 9 27 – 259 P02570 Actin Cell motility 41 606 5.29 12 32 – 260 P02570 Actin Cell motility 41 606 5.29 8 25 – 261 P10909 Clusterin Apo J

(�-chain)Apoptosis 25 883 5.66 5 23 – 2

62 P10909 Clusterin (�-chain) Apoptosis 24 197 6.27 7 18 – 263 P10909 Clusterin (�-chain) Apoptosis 24 197 6.27 6 12 – 264 P04901 GTP binding protein

(�-subunit)Signal transduction 37 377 5.60 7 20 – 2

65 P02649 Apolipoprotein E Transport and catabolismof lipoprotein

36 154 5.6 13 46 – 2

66 P02679 Fibrinogen Platelet aggregation 48 483 5.24 6 16 – 267 Q9H459 BTN2A1 unknown 42 430 6.37 7 22 YES 268 P02768

Q13410Serum AlbuminButyrophilin

Binding and transportFat globules secretion

66 47256 196

5.675.32

137

2314

– 2

69 P04003 C4BP Complement activation 67 034 7.2 8 13 – 270 P04003 C4BP Complement activation 67 034 7.2 10 17 – 271 P04003 C4BP Complement activation 67 034 7.2 15 29 – 272 Q13410 Butyrophilin Fat globules secretion 56 196 5.32 12 22 – 273 Q99541 Adipophilin Triacylglycerol deposition 48 075 6.3 16 45 – 274 Q99541 Adipophilin Triacylglycerol deposition 48 075 6.3 15 36 – 275 Q99541 Adipophilin Triacylglycerol deposition 48 075 6.3 12 33 – 2


Table 1. Continued


Sequencecoverage


Figure

76 Q99541 Adipophilin Triacylglycerol deposition 48 075 6.3 13 36 – 277 Q96GX7 Selenium binding

proteinunknown 52 391 5.93 19 51 – 2

78 P11142 HS7C unknown 70 898 5.37 14 29 – 279 Q13410 Butyprophilin Fat globules secretion 56 196 5.32 15 29 – 280 Q93097 WNT-2B protein Cellular development 41 830 9.31 10 36 – 181 Q13410 Butyprophilin C-term Fat globules secretion 56 196 5.32 14 22 – 182 P02647 Apolipoprotein A-I Transport of cholesterol 28 079 5.27 16 58 – 183 P02647 Apolipoprotein A-I Transport of cholesterol 28 079 5.27 24 67 – 184 Q13410 Butyprophilin C-term Fat globules secretion 56 196 5.32 8 16 – 185 Q9Y6B6

P06749

GTP-binding proteinSAR1bTransforming proteinRhoA

Transport

Signal transduction pathway

22 410

21 678

5.76

5.83

12

6

60

48

–

–

1

86 P19440 Gamma-glutamyltranspeptidase –chain 2

Glutathione metabolism 19 998 5.90 7 48 – 1

87 Q9NRV9 Heme-binding protein unknown 21 097 5.71 9 64 – 188 Q9HCU9 Breast cancer metas-

tasis-suppressor 1Mediator of metastasis

suppression28 461 4.69 4 21 – 1

89 P05814 �-casein Micelles stability 25 254 5.50 3 19 YES 190 P06749 Transforming protein

RhoASignal transduction pathway 21 678 5.83 6 37 – 1

91 Q9Y6B6 GTP-binding proteinSAR1b

Transport 22 410 5.76 5 24 – 1

92 Q08431 Lactadherin (N-term) Phospholipid binding 40 791 8.22 9 30 – 193 Q08431 Lactadherin Phospholipid binding 40 791 8.22 7 24 – 194 P05413 Fatty acid-binding

proteinLipid transport 14 727 6.34 13 81 – 1

95 P05413 Fatty acid-bindingprotein

Lipid transport 14 727 6.34 7 42 – 1

96 P29373 CRABP II Cell differentiation 15 562 5.43 6 53 YES 197 P36969 GSHH Protecting 18 700 8.44 4 26 YES 198 P05092 Rotamase Folding 17 881 7.72 5 43 YES 199 P05092 Rotamase Folding 17 881 7.72 5 37 – 1100 Q08431 Lactadherin (N-term) Phospholipid binding 40 791 8.22 6 19 – 1101 P02654 Apolipoprotein C1 Interaction with ApoE 9 332 8.00 4 16 YES 1102 P02654 Apolipoprotein C1 Interaction with ApoE 9 332 8.00 6 28 – 1103 P02649 Apolipoprotein E Lipid binding 34 237 5.52 7 21 YES 1104 P00709 �-lactalbumin Lactose synthesis 16 660 4.80 7 46 – 1105 Q08431 Lactadherin Phospholipid binding 40 791 8.22 15 40 – 1106 Q08431 Lactadherin Phospholipid binding 40 791 8.22 17 52 – 1107 P13987 CD59 glycoprotein Inhibitor 14 177 6.0 3 19 YES 1

a) Spot whose identification was described in the previous paper [9]


3.3 Functional distribution of identified proteins

The identified proteins are classified into different groupsaccording to their function (Fig. 3), as found in the SWISS-PROT (http://www.expasy.ch/) and PROSITE database(http://www.expasy.org/prosite/). In our study, 46% ofthe proteins are typical MFGM proteins. Among thesethe most abundant is lactadherin, a peripheral membraneprotein containing an endothelial growth factor-likedomain. As it is known, MFGM proteins have a very lownutritional value, while they play important roles indefense mechanisms for the newborn. Lactadherinresists being digested in the stomachs of milk-fed infants,and prevents symptomatic rotavirus infection in breast-fed infants [19]. Typical milk proteins, such as �-lactalbu-min or casein, account for 13% of the identified spots;they were found associated with MFGMs, even if theyare usually detected in the skimmed milk fraction. Thepresence of lactoferrin, a glycoprotein which accountsfor 20% of the total protein in human milk is quite interest-ing, it has a bacteriostatic effect by competing with bac-teria for iron [20].

Considering the high rate of protein synthesis of themammary gland during milk secretion, 10% of the identi-fied proteins are involved in folding and protein destina-tion. Among these, rotamase or cyclophilin A (CyPA) is apeptidylprolyl cis-trans-isomerase; although ubiquitouslydistributed as intracellular protein, it may be secreted bycells in response to inflammatory stimuli and is a potentneutrophil and eosinophil chemoattractant in vitro andin vivo [21].

Figure 3. Graphic representation of the distribution ofhuman MFGM proteins classified according to theirfunction.

Proteins involved in the intracellular transport and/orreceptorial activity were detected in 9% of the identifiedspots. In the case of the cargo selection protein or TIP47,the biological function appears to be linked to the bud-ding and secretion of MFGM, through the interaction be-tween lipid droplet surface, adipophilin and butyrophilin[22].

Low abundance proteins, identified as single or doublespots, are involved in cell development and differentia-tion, in cell motility and in signal transduction. They showthe mammary epithelial cell origin of the MFGM, and theycould be the markers of the pathophysiological state ofthe lactating mammary gland.

4 Concluding remarks

Basic knowledge derived from structural proteomic stud-ies is a prerequisite for understanding functional proteom-ics. In this case, an update of the proteins associated withhuman milk fat globules and, more remarkable, the avail-ability of this information on-line as an annotated 2-DEmap (http://www.csaapz.to.cnr.it/proteoma/2DE), may beof some use in functional proteomic studies dealing withthe physiology and pathology of the mammary gland, andwith achieving formulated milks that are more similar tothe “golden standard” represented by maternal milk.

This work was partially supported by Business Unit Nutri-tion (B.U.N.) of Nestlè – Italiana, S.p.A.

Received October 15, 2002

5 References

[1] Ferlin, M. L., Santoro, J. R., Jorge, S. M., Goncalves, A. L.,J. Perinat. Med. 1986, 14, 251–257.

[2] Lawrence, R. A., in: Bircher, S. (Ed.), Breastfeeding: a Guidefor Medical Profession, C. V. Mosby, St. Louis 1989, pp. 73–117.

[3] Cavaletto, M., Cantisani, A., Napolitano, L., Giuffrida, M. G. etal., Milkwissenshaft 1994, 49, 303–305.

[4] Jenness, R., in: Larson, B. L. (Ed.), Lactation, University ofIowa Press, Ames USA 1985, pp. 78–94.

[5] Lönnerdal, B., Atkinson, S., in: Jensen, R. G. (Ed.), Handbookof Milk Composition, Academic Press, San Diego 1995,pp. 351–368.

[6] Newburg, D. S., Curr. Med. Chem. 1999, 6, 117–127.

[7] Jensen, R. G., Blanc, B., Patton, S., in: Jensen, R. G. (Ed.),Handbook of Milk Composition, Academic Press, San Diego1995, pp. 50–62.

[8] Goldfarb, M., Electrophoresis 1997, 18, 511–515.


[9] Quaranta, S., Giuffrida, M. G., Cavaletto, M., Giunta, C. et al.,Electrophoresis 2001, 22, 1810–1818.

[10] Keenan, T. W., Patton, S., in: Jensen, R. G. (Ed.), Hand-book of Milk Composition, Academic Press, San Diego1995, pp. 5–50.

[11] Lönnerdal, B., Woodhouse, L. R., Glazier, C., J. Nutr. 1987,117, 1385–1395.

[12] Laemmli, U. K., Nature 1970, 227, 680–685.

[13] Wessel, D., Flugge, U. I., Anal. Biochem. 1984, 138, 141–143.

[14] Sanchez, J. C., Rouge, V., Pisteur, M., Ravier, F. et al., Elec-trophoresis 1997, 18, 324–327.

[15] Hellmann, U., Wernstedt, C., Gonez, J., Heldin, C. H., Anal.Biochem. 1995, 224, 451–455.

[16] Henzel, W. J., Billeci, T. M., Stults, J. T., Wong, S. C. et al.,Proc. Natl. Acad. Sci. USA 1993, 90, 5011–5015.

[17] Fountoulakis, M., Juranville, J. F., Berndt, P., Langen, H.,Suter, L., Electrophoresis 2001, 22, 1747–1763.

[18] Cavaletto, M., Giuffrida, M. G., Fortunato, D., Gardano, L.et al., Proteomics 2002, 2, 850–856.

[19] Peterson, J. A., Scallan, C. D., Ceriani, R. L., Hamosh, M.,Adv. Exp. Med. Biol. 2001, 501, 179–187.

[20] Newborg, D. S., J. Mammary Gland Biol. Neoplasia 1996, 1,271–283.

[21] Yurchenko, V., Zybarth, G., O’Conner, M., Dai, W. W. et al.,J. Biol. Chen. 2002, 277, 22959–22965.

[22] Mather, I. H., J. Dairy Sci. 2000, 83, 203–247.

Structural proteome of human colostral fat globule membrane proteins

Documents

Transcript of Structural proteome of human colostral fat globule membrane proteins