Advanced Dairy Chemistry

440 J.-L. Vilotte et al. a reduction (Liu et al., 1997). Indeed, there is no gene were reported to occur (Whitelaw and correlation between the expression of the STAT5a Webster, 1998). This implies a reorganization of and the b-casein genes (Kazansky et al., 1995). the chromatin in response to the interaction of Studies with STAT5b knockout mice indicate transcription factors. Furthermore, this assay has that this protein is associated with growth hor- detected species-specific differences in this reor- mone effects (Udy et al., 1997). ganization of the chromatin (Pena et al., 1998). These differences are important clues to under- Altogether, these observations suggested that stand the various levels of expression of the same STAT5 may facilitate the interaction of other gene between species or between alleles (Whitelaw, transcription factors to allow the transcriptional 2000). Similar studies have shown that chromatin activation of the promoter. Indeed, functional reorganization events also occur during the induc- interaction between STAT5 and the glucocorti- tion of the expression of the rat WAP gene (Li and coid receptor was reported, and this molecular Rosen, 1994b). However, the clearest evidence yet complex was demonstrated to cooperate in the describing a role for chromatin structure in the induction of the b-casein promoter, indepen- regulation of milk protein gene expression comes dently of the DNA-binding function of the gluco- from the analysis of the b-casein BCE-1 element. corticoid receptor (Stöcklin et al., 1996, 1997; This element is not activated in transient transfec- Lechner et al., 1997). tions, in which true chromatin is not formed, and is responsive to the state of acetylation of the his- Better understanding of the nature and struc- tones (Myers et al., 1998). tural arrangement of the cis-regulatory elements involved in the control of the transcription of the The transcription domains of the major milk major milk protein genes has given an insight protein genes are poorly defined. As yet, they into the differential developmental regulation of have generally been defined through functional their expression. However, many questions analyses only. Transgenic studies have revealed remain. Why is expression of WAP in preference that while the as1- and the b-casein genes from to b-casein affected in the STAT5a-null mice various species can be expressed at relatively given that the reciprocal requirement for prolac- high levels in mice, only weak expression of the tin is observed? If it is not STAT5, and it appears as2- and the k-casein transgenes was observed. It not to be, what confers mammary specificity to is thus hypothesized that a mammary-specific milk protein gene expression? How do the vari- locus control region (LCR) might control the ous transcription factors involved interact with expression of the casein locus and that this puta- each other to generate a production transcription tive element is located close to the as1- and the complex? How does this transcription complex b-casein loci. So far, however, no direct evidence interact with the underlying nucleosomes? for the existence of such an element has been pro- vided. Furthermore, as the casein gene clusters of 14.3.5 Milk Protein Gene Chromatin some species contain several genes which are not Domains expressed tissue specifically in the mammary gland (Fig. 14.1), the influence of a putative LCR It is now generally accepted that chromatin, or its would have to be selective for the casein genes. epigenetic regulation, plays a central role in the regulation of gene expression (Rijnkels et al., A consequence of chromatin gene domains is 2010). Studies on the ovine b-lactoglobulin gene that boundaries to these domains must exist. There brought indirect, although compelling, evidence are several documented examples (see Geyer, 1997, that chromatin reorganization is important for milk for review), but again, as yet, none has been con- protein gene expression. Concomitant to the clusively identified for a milk protein gene. increase in b-lactoglobulin gene expression during However, some predictions are possible using pregnancy, due to activation of STAT5, changes in genome comparison. The boundaries of the a-lac- DNase I hypersensitivity in the b-lactoglobulin talbumin gene are not known, but those of the related lysozyme gene are well defined (Phi-Van

14 Genetics and Biosynthesis of Milk Proteins 441 and Strätling, 1988). It is likely that the a-lactalbu- frame. Since part of the regulation of milk protein min gene contains similar elements, probably in a gene regulation is exerted at a post-transcriptional similar location. This hypothesis is supported by level (see Aggeler et al., 1988; Blum et al., 1989; the observed position-independent and copy num- Golden and Rillema, 1995, for examples), con- ber-related expression of a human 250 kb YAC and servation of the UTR might reflect their impor- of a 160 kb goat BAC-a-lactalbumin transgene tance for the mRNA processing. In many cases, (Fujiwara et al., 1997; Stinnakre et al., 1999). the rate of translation is influenced by the 5¢ UTR sequence and its secondary structure. Sequences At present, the transcription domain of the located in the 3¢ UTR are also known to interact b-lactoglobulin gene might be one of the best with proteins to either stabilize or degrade the studied within milk protein genes. Several DNase RNA, and/or alter mRNA translation efficiency, I-hypersensitive sites that reflect its expression and to be potential targets for miRNA. Involvement status have been located within the promoter, of miRNA in the biology of the mammary gland intronic and 3¢-flanking regions (Whitelaw and is still poorly documented. However, recent stud- Webster, 1998). These sites spatially reflect the ies suggested their implication during normal limit of the chromatin domain, as defined by mammary gland development and differentiation nuclease sensitivity. The 5¢ limit of this domain (Gu et al., 2007; Wang and Li, 2007; Avril-Sassen resides very close to the promoter. In addition, the et al., 2009; Sdassi et al., 2009; Tanaka et al., proximal 3¢-flanking b-lactoglobulin sequences 2009; Ucar et al., 2010). A direct role for miRNA can interact with the nuclear matrix in vitro. This in the synthesis of a milk protein was recently suggests that the b-lactoglobulin gene resides in a evidenced with the identification of a conserved very small chromatin domain (Whitelaw, unpub- region in the 3¢ UTR of the lactoferrin gene tar- lished results). Sequences that interact with the geted by miR-214 (Liao et al., 2010). nuclear matrix, usually AT-rich in nature, may be involved in regulating chromatin structure thereby As already mentioned, total or partial exon facilitating gene expression. Although several skipping during the splicing of the pre-mRNA studies have addressed the ability of such is responsible for the differences observed sequences to enhance milk protein gene expres- between casein variants and between homolo- sion, a clear picture has not yet emerged (e.g., gous proteins from different species. In this Attal et al., 1995; McKnight et al., 1996). Another section, we will just describe variations that well-studied locus is that encompassing the WAP affect the overall level of gene expression. gene, the DNA of which adopts different chroma- Deletion or transition of a single nucleotide tin loop structures according to the studied tissue within coding exons of a caprine as1-casein and developmental stage that allow the differen- allele and of two b-casein alleles creates non- tial expression of the genes it contains (Montazer- sense codons (Leroux et al., 1992; Persuy et al., Torbati et al., 2008) and regulation from distant 1996; Rando et al., 1996). These alleles are enhancers and/or repressors (Saidi et al., 2007). characterized by a much lower level of mRNAs compared to other alleles and, for some of them, 14.3.6 Milk Protein mRNAs with multiple exon-skipping events. This phe- nomenon has been observed for several other The structure of milk protein mRNAs has been genes where nonsense codons have been found investigated in numerous species (see Mercier to be associated with mRNA decay and/or exon and Vilotte, 1993, for review). In lactation, these skipping (see Valentine, 1998; Hentze and RNAs account for up to 60 %–80 % of the total Kulozik, 1999, for reviews). Goat and bovine RNA present in mammary epithelial cells. The as1-casein alleles, also associated with reduced calcium-sensitive casein mRNAs are character- amount of mRNA and milk protein, were found ized by a better interspecies conservation of the to contain a truncated inserted LINE element in UTRs and of the sequence encoding the signal the 3¢UTR (Perez et al., 1994; Rando et al., peptide compared to the mature protein-coding 1998). Insertion of these elements is supposed to reduce mRNA stability.

442 J.-L. Vilotte et al. 14.4 Milk Protein Synthesis cells that constitute the acini are separated from and Secretion the interstitial space by a basement membrane like in any epithelial tissue. Secretion of milk protein is obviously of interest 14.4.2 The Functional due to its physiological and economic importance Compartmentalisation but also from the point of view of the study of of the Biosynthetic-Secretory high-efficiency secretory pathways. The trafficking Pathway of Mammary and processing events leading to the secretion of Epithelial Cells milk proteins are known in general outline, but relatively little is established of the molecular cell During lactation, mammary epithelial cells are biology of milk protein transport in the secretory highly polarized and display features typical of pathway of mammary epithelial cells. cells specialized for secretion (Bargmann and Knoop, 1959; Wooding, 1977; Pitelka and 14.4.1 Morphological Organization Hamamoto, 1983; illustrated in Figs. 14.4 and of the Mammary Secretory 14.5). Their cytoplasm contains an extended net- Epithelium work of numerous parallel lamellar cisternae and branching tubules decorated with electron-dense Functional differentiation of the mammary gland ribosomal particles: the rough endoplasmic reticu- is linked to the development of its epithelial tis- lum (ER). As described in other polarized epithe- sues. From the onset of pregnancy, the duct cells lial cells, the Golgi apparatus is typically located enter a proliferation and differentiation period in the peri- and supranuclear region of the cell, leading to the development of a highly branched close to the centrosome. The Golgi cisternae are ductal tree which fills the entire mammary fat well developed, more or less distended, often con- pad. Alveolar structures, or acini, develop at the taining electron-dense particles and filamentous ends of the side branches, and terminal differen- materials. Of note, unusually large vesicles are tiation of the alveolar mammary epithelial cells is observed on the trans side of the Golgi apparatus. completed at the end of gestation with the start of It is not clear whether these dilated structures con- milk secretion at parturition. Functional acini are stitute the trans-Golgi network (TGN) or represent embedded in a stroma composed of connective newly formed secretory vesicles en route to the and adipose tissues, fibroblasts, plasma cells, apical plasma membrane (APM). In addition, blood vessels and nerve terminals. These mor- numerous smaller vesicles, some of them coated phological aspects are well documented at http:// with the typical spike structure characteristic of mammary.nih.gov/ clathrin, are associated with the trans side of the Golgi. Secretory vesicles that have pinched off The acini consist of a single layer of cuboidal from the TGN contain filamentous structures and mammary epithelial secretory cells sealed casein micelles similar to those found in milk, in together at the luminal border by tight junctions an electron-lucent fluid. so that passage of molecules from the interstitial space to the lumen of the acini by a paracellular 14.4.3 Intracellular Transport and route is restricted during lactation, confining such Co- and Post-translational molecular movements to the cellular trans-cytotic Modifications of Milk Proteins pathway. Progenitor cells are believed to reside below the monolayer of luminal cells. Contractile In mammary epithelial cells, transport of newly myoepithelial cells that have long spidery pro- synthesized proteins destined for secretion pro- cesses embrace the alveoli secretory cells. They ceeds according to the general biosynthetic-secre- participate in milk ejection by squirting milk out of the acini lumen into the ducts. Finally, these

14 Genetics and Biosynthesis of Milk Proteins 443 Fig. 14.4 The intracellular compartments of the mam- ported to the apex of the mammary epithelial cells. CLDs mary epithelial cell involved in the biosynthesis and secre- and caseins secretion may be partly (pathway B) or tion of milk proteins and lipids. Milk proteins undergo significantly (pathway C) coupled (dotted lines indicate various co- and post-translational modifications during that membrane of the casein vesicles has fused together) their translocation into the lumen of the rough endoplas- since the budding of the CLDs requires considerable mic reticulum (RER) and further transport to the apical amounts of APM. Pathway B may only involve hetero- cell surface, via the Golgi apparatus. Casein self-associa- typic fusion between the casein vesicles and the APM tion starts in the lumen of the RER and proceeds into long while pathway C may imply both homotypic between SVs loose linear aggregates in the trans-Golgi compartments and heterotypic fusion of SVs with the APM. Alternatively, and early secretory vesicles (SV). They further self-asso- CLDs may be secreted independently of casein-contain- ciate before release, and structures with the characteristic ing SV (pathway D), and some MLDs may also be directly morphological aspect of casein micelles are found in dis- secreted at the apical side of the MECs (not shown). Basal tal SV (see Fig 14.5). Casein-containing SV reach and plasma membrane (BPM), cis-Golgi network (CGN), fuse with the apical plasma membrane (APM) by exocy- ER-Golgi intermediate compartment (ERGIC), lipid tosis (pathway A). Microlipid droplets (MLDs) emerge droplet (LD), milk fat globule (MFG), tight junction (TJ), from the RER and may fuse with each other and with trans-Golgi network (TGN). Redrawn from Chat et al. larger cytoplasmic lipid droplets (CLDs) as they are trans- (2011) (Courtesy of S. Truchet) tory pathway (Palade, 1975). Following a 3-min Golgi region after 15 min, accumulated in apical pulse labelling with a radioactive amino acid, newly secretory vesicles after 45 min and are predomi- synthesized proteins are detectable in the ER 5 min nantly located in the lumen of the acini after 60 min after the beginning of the pulse, concentrated in the (Seddiki and Ollivier-Bousquet, 1991).

444 J.-L. Vilotte et al. Fig. 14.5 Transmission electron micrograph of part of trans-Golgi, whereas more condensated protein aggre- a mouse mammary epithelial cell during lactation. The gates and typical casein micelles are present in more major compartments of the secretory pathway are illus- distal SV and in the lumen of the acini. ER, endoplas- trated. Filamentous material is found in dilated Golgi mic reticulum (Courtesy of S. Chat and C. Longin) cisternae and in secretory vesicles (SV) close to the 14.4.3.1 Translocation into and is assisted by proteins with chaperone activities, Transport from the Endoplasmic including BiP/GRP78, Erp77, calnexine and Reticulum calreticulin (Pelham, 1989). These ER-resident proteins bind to misfolded newly synthesized Milk proteins are of two types: transmembrane proteins and retain them in the ER until they have proteins and water-soluble proteins. An example achieved their properly folded or oligomeric of milk transmembrane protein is butyrophilin state. Correct protein folding or oligomerisation which is first targeted to the plasma membrane is prerequisite for protein export from the ER and from which it is released into milk as part of the transport to the Golgi complex. This process has milk fat globule (MFG) membrane. The vast been named “quality control” (Hammond and majority of milk proteins, however, are water- Hellenius, 1995). Accumulation of improperly soluble proteins, for example, the caseins. Entry folded proteins in the ER causes a stress which into the ER requires a signal sequence (Blobel triggers a coordinated adaptive programme called and Dobberstein, 1975). Translocation of milk the unfolded protein response (UPR). protein polypeptide chains into the ER lumen and subsequent cleavage of their signal sequence Elucidation of the secondary structure of involve classical mechanisms (Schatz and a-lactalbumin and b-lactoglobulin has revealed Dobberstein, 1996). the presence of four and two disulfide bridges, respectively (Brew et al., 1970; Papiz et al., Once translocated into the ER lumen, proteins 1986). As to WAP, it contains numerous cysteine are in an oxidizing environment which promotes residues which form multiple intramolecular the formation of disulfide bonds between cysteine disulfide bonds (Hennighausen and Sippel, 1982, residues. Most proteins synthesized in the rough Devinoy et al., 1988). Moreover, native WAP is ER are glycoproteins. They are modified by a dimeric (Baranyi et al., 1995). Analysis of the common oligosaccharide on target asparagines primary structure of k-casein reveals the very residues (N-glycosylation). N-linked oligosac- high degree of conservation of at least one charides are added co-translationally and serve as cysteine residue in the N-terminal domain of the tags to monitor the state of protein folding. The protein (Bouguyon et al., 2006). Dimers of proper folding of proteins that are made in the ER

14 Genetics and Biosynthesis of Milk Proteins 445 k-casein were found in all milk studied, and at least some of the caseins in the ER lumen is a interchain disulfide bridges were also found for prerequisite for their forward transport to the any casein possessing a cysteine residue Golgi apparatus is not clear. However, investiga- (Bouguyon et al., 2006), for example, in mouse tion of the impact of the polymorphism at the as1- and rat which express four or five cysteine-con- casein locus on goat milk secretion has shown taining caseins, respectively. As expected, disul- that, in the absence of aS1-casein, other caseins phide bond formation between casein molecules accumulate in the ER (Chanat et al., 1999). The was demonstrated to occur within the ER lumen efficiency of casein transport from the ER to the (Le Parc et al., 2010). Finally, N-glycosylation of Golgi apparatus was strongly affected in this con- rat a-lactalbumin has been detected in mammary text. Data suggested that interaction of caseins in microsomal membranes (Lingappa et al., 1978). a yet-to-be-identified structure including aS1-ca- Although N-glycosylated forms of a-lactalbumin sein is required for efficient transport of these have been observed in several species, only rat proteins to the Golgi apparatus. More recently, a-lactalbumin is efficiently glycosylated (Prasad the existence of a membrane-associated form of et al., 1979; Chanat, 2006). aS1-casein in the ER and more distal compart- ments of the secretory pathway of mammary epi- Caseins are not N-glycosylated and were thelial cells has been reported (Le Parc et al., shown to lack appreciable amount of regular sec- 2010), further suggesting a key role of aS1-casein ondary structure. b- and k-caseins might possess in casein transport in the biosynthetic pathway. premolten or molten globule conformations whereas aS1- and aS2-caseins are intrinsically From the above considerations, it is now obvi- unstructured proteins (Farrell et al., 2006; see ous that first interactions between the caseins Chap. 5) or natively unfolded proteins (Holt and take place in the ER and that the exit of casein Sawyer, 1993). The characteristic structural fea- from this compartment is a key step in casein ture of natively unfolded proteins is a combina- micelle biogenesis and casein transport in the tion of low mean hydrophobicity and relatively secretory pathway. high proportion of charged residues at physiolog- ical pH (Uversky et al., 2000). Proteins with such 14.4.3.2 Transport Through the Golgi an open structure possess a peculiar aggregative Apparatus behaviour and are prone to interact with their specific ligand in vivo. The caseins, however, do Protein transport through the Golgi apparatus not fulfil these two criteria since they present may occur according to the vesicular transport relatively high hydrophobicities. model (stable Golgi cisternae) or the cisternal maturation model (Glick and Malhotra, 1998). To date, the question whether caseins are sub- Many lines of evidence now support this later jected to the ER quality control machinery has model in which the Golgi cisternae themselves not been directly addressed. Moreover, whether move through the Golgi stack. The observation or not soluble luminal cargo proteins need to be that electron-dense structures, most likely casein concentrated and/or require intrinsic sorting aggregates, are detectable in Golgi cisternae of information for loading into COPII-coated trans- lactating mammary epithelial cells but are port carriers at the ER exit sites to move to the excluded from Golgi-associated vesicles is in Golgi apparatus is still controversial (for review agreement with this cisternal maturation model see Lee et al., 2004). Some secretory proteins (Clermont et al., 1993). seem to enter a transport carrier without positive selection by a signal-independent mechanism Golgi enzymes carry out protein modifications known as “bulk flow” (Rothman and Wieland, that include glycosylation, phosphorylation, sul- 1996). Finally, as stated above, subunit oligo- phation and proteolytic processing (Fig. 14.4). In merisation is required for export of multimeric mammary epithelial cells, glycosyltransferases proteins from the ER (Copeland et al., 1986; Kim are notably involved in the O-glycosylation of and Arvan, 1991). Whether interaction between k-casein. Moreover, galactosyltransferase, with the help of a-lactalbumin, is also responsible for

446 J.-L. Vilotte et al. the synthesis of lactose (Ebner and Brodbeck, tent present in newly formed secretory vesicles in 1968). Galactosyltransferase activity has been clathrin-coated vesicles (Tooze, 1998). detected within the Golgi apparatus but also in secretory vesicles and milk (Witsell et al., 1990; Pre-micellar aggregates have been observed Boisgard and Chanat, 2000). The synthesis of by electron microscopy in the lumen of the Golgi lactose in the trans-most cisternae of the Golgi cisternae (Clermont et al., 1993). Consistent with apparatus, and most likely in secretory vesicles, the presence of a high calcium concentration in surely explains the swollen aspect of these organ- the trans-most Golgi cisternae (Neville and elles in mammary epithelial cells. The striking Watters, 1983), there is a drastic rearrangement reduction of the volume of casein-containing of the micellar structure during the formation of secretory vesicles in a-lactalbumin-deficient secretory vesicles at the TGN and their transport mice is in agreement with this hypothesis to the APM for exocytosis. In newly formed (Stinnakre et al., 1994). Phosphorylation of the transport vesicles, caseins are concentrated in the calcium-sensitive caseins on serine clusters form of long loose linear aggregates (Mather and allows calcium phosphate binding and further Keenan, 1983; Clermont et al., 1993). These pro- interactions between caseins. Like in other cell gressively self-associate and become bigger and systems, the kinases that phosphorylate the denser, and structures with the characteristic hon- caseins are located within the Golgi apparatus eycomb texture of casein micelles from milk are (Bingham and Farrell, 1974; West and Clegg, found in distal secretory vesicles (see Fig. 14.5). 1984). Notably, the phosphorylation of b-casein The mean size of casein micelles varies widely seems delayed compared to that of aS1-casein across species, and the relative proportion of (Turner et al., 1993; Boisgard and Chanat, 2000; k-casein was demonstrated to be a modulator of Péchoux et al., 2005). This suggests that strong micelle size (Gutierrez-Adan et al., 1996). interaction of b-casein with casein polymers Noteworthy, casein micelles, or at least big aggre- might be postponed until it is trafficked to trans- gates, still form in mammary epithelial cells from Golgi cisternae (see below). Protein sulphation is of aS1-, b- or k-casein-deficient animals (Kumar a ubiquitous TGN-specific post-translational et al., 1994; Chanat et al., 1999; Shekar et al., modification. Beside sulphated proteoglycans 2006). Interactions between the various caseins (Boisgard et al., 1999), the sulphation of both and minerals during micelle biogenesis in the a-lactalbumin and k-casein from rat has been secretory pathway might therefore involve rather described (Chanat, 2006). One the other hand, general physico-chemical and biochemical char- although no direct evidence for the cleavage of acteristics of these components. However, these milk proteins by endoproteases has been reported, characteristics are specific enough to avoid incor- furin has been shown to be relatively abundant in poration of whey proteins in the micelles. On the Golgi-derived clathrin-coated vesicles from lac- other hand, casein-containing secretory vesicles tating rabbit mammary epithelial cells (Pauloin might undergo maturation before exocytosis et al., 1999). (Pauloin et al., 1999). Finally, reports support the notion that secretory vesicles destined to fuse 14.4.3.3 Transport from the trans-Golgi with the apical cell surface transport both caseins Network and Secretory Vesicle and whey proteins (Devinoy et al., 1995; Neville Exocytosis et al., 1998; Ollivier-Bousquet, unpublished observation). Secretory proteins are segregated, highly concen- trated and packaged into appropriate transport A large body of evidence supports the concept vesicles in the TGN. Sorting and concentration that local modifications of the lipid composition are believed to involve the selective aggregation of membranous sub-domains by lipid-modifying of the secretory proteins in the ionic environment enzymes also contribute to vesicular traffic. In of the TGN (Chanat and Huttner, 1991), as well line with this, phospholipase D and calcium- as retrieval of excess membrane and luminal con- independent phospholipase A2 were reported to be involved in both the transport of milk proteins

14 Genetics and Biosynthesis of Milk Proteins 447 from the ER to the Golgi and in the formation of including oxytocin and prolactin are able, at least secretory vesicles from the TGN (Boisgard and in vitro, to increase casein secretion (secret- Chanat, 2000; Péchoux et al., 2005), as was agogue effect) in mammary epithelial cells from observed in other cell systems (Riebeling et al., rabbits and rodents (see Ollivier-Bousquet, 1993, 2009; Schmidt et al., 2010). 1997). Prolactin seems to act on a late step of casein trafficking, possibly exocytosis. In con- Trafficking steps within the secretory pathway trast, oxytocin was reported to also accelerate the of mammary epithelial cells and exocytosis of transport of newly synthesized proteins from the casein-containing vesicles with the plasma mem- ER to the Golgi apparatus and to secretory vesi- brane might involve SNARE (soluble cles (Lollivier et al., 2006). N-ethylmaleimide-sensitive fusion (NSF) attach- ment protein (SNAP) receptor) proteins, as 14.5 Amino Acid Transport already described in other cell types (Sollner by the Mammary Gland et al., 1993; Jahn and Scheller, 2006). To date, however, only a few studies have directly Amino acids, extracted from interstitial fluid, are addressed the functions of SNARE proteins in the major source of amino nitrogen for milk pro- mammary epithelial cells. One of these suggests tein synthesis; therefore, a knowledge of mam- that VAMP-8 (vesicle-associated membrane pro- mary tissue amino acid transport mechanisms tein 8) may be involved in casein secretion (Wang and their regulation is important if we are to et al., 2007). On the other hand, SNAP-23 (syn- understand fully the process of milk protein aptosomal-associated protein 23), syntaxin-3 and secretion. Such knowledge will help those wish- syntaxin-5, and Ykt6 have been described as ing to manipulate milk protein content via dietary being associated with lipid droplets (Boström means. The uptake of amino acids across the et al., 2007; Reinhardt and Lippolis, 2008). basolateral membranes of mammary secretory SNAP-23 is believed to play a role in homotypic cells, the major point of entry, is accomplished by fusion of intracellular lipid droplets. Moreover, an array of distinct transport mechanisms. The the large amount of membrane necessary for amino acid transporters differ from one another MFG secretion by budding of the APM could be with respect to kinetics, substrate specificity and partly provided by exocytosis of casein-contain- ion dependency; however, it is evident that they ing vesicles (Mather and Keenan, 1998; see operate in a coordinated fashion to supply amino Fig. 14.4). Recently, the endogenous expression nitrogen to support milk protein synthesis. levels of a large set of SNAREs were investigated in mouse mammary gland (Chat et al., 2011). The study of mammary epithelial amino acid This study points to SNAP-23 as a potential cen- transport is hampered by the relatively complex tral player for the coupling of casein and MFG anatomy of the mammary gland. Nevertheless, secretion during lactation. the use of mammary tissue explants and the per- fused mammary gland has enabled the transport 14.4.4 Hormonal Regulation of Milk of amino acids (using radiotracers) by mammary Protein Secretion tissue to be characterized. Mammary explants are easy to prepare and have the advantage that Proteins destined for the cell exterior are secreted the cellular architecture remains intact. by either the constitutive or the regulated secre- Furthermore, the preparation of mammary tory pathway (Glombik and Gerdes, 2000; explants does not require digestive agents: Morvan and Tooze, 2008). The fact that there is enzymes such as collagenase could ultimately no substantial storage of newly synthesized milk alter the properties of membrane transport pro- proteins in mammary epithelial cells does not teins. Although mammary tissue explants iso- support the later hypothesis (Devinoy et al. 1995; lated from lactating animals are comprised of Pauloin et al. 1997). On the other hand, hormones more than one cell type, it can be assumed that

448 J.-L. Vilotte et al. the vast majority of the surface area of explants dient to drive the movement of amino acids into is that of the basolateral membrane of the secre- mammary cells. The Na+-dependent mechanisms tory cells. Therefore, it is reasonable to assume which have been identified in mammary tissue that mammary tissue explants can be used to include systems A, ASC, XAG- and b. give a measure of amino acid transport across the blood-facing side of the mammary epithe- System A prefers short-chain neutral amino lium. The large tissue extracellular space associ- acids as substrates and is characterized by its tol- ated with mammary explants does, however, erance of N-methylated amino acids such as place limitations on the design of experiments. N-methylaminoisobutyrate (MeAIB). Indeed, The perfused mammary gland allows the trans- Na+-dependent amino acid uptake inhibited by port of amino acids across the blood-facing MeAIB is usually taken as a measure of transport aspect of the mammary epithelium to be mea- via system A. It appears that mouse, rat and sured under near physiological conditions: per- bovine mammary tissue possesses system A fusates can be delivered to the gland with a flow activity (Neville et al., 1980; Baumrucker 1984; and pressure profile similar to that found in vivo. Verma and Kansal 1993; Shennan and McNeillie The perfused mammary gland used in combina- 1994a). Lopez et al. (2006) provided convincing tion with a rapid, paired-tracer dilution technique evidence that SNAT2 may be the molecular cor- allows the transport of amino acids to be mea- relate of system A in rat mammary tissue. In con- sured over very short time courses (Mepham trast, no evidence for the presence of system A at et al., 1985; Calvert and Shennan 1996). the blood-facing aspect of the guinea pig mam- mary gland could be found (Mepham et al., 14.5.1 Mammary Tissue Amino Acid 1985). The limited kinetic data available suggest Transport Systems that system A in lactating mammary tissue oper- ates with relatively low affinity: the KM of methi- The identification of individual amino acid trans- onine and a-aminoisobutyric acid uptake via port systems is difficult on account of the fact system A in mouse mammary tissue is 0.47 mM that a single amino acid may be able to utilize and 2.0 mM, respectively (Neville et al., 1980; several transport systems (see Barker and Ellory Verma and Kansal, 1993). Several lines of evi- 1990). Furthermore, the difficulty in characteriz- dence suggest that system A is regulated by milk ing amino acid transport is compounded by the stasis, by starvation and by the stage of lactation lack of specific inhibitors of amino acid trans- (Neville et al., 1980; Shennan and McNeillie porters. In spite of these drawbacks, significant 1994c; Verma and Kansal 1995). In accordance progress has been made towards identifying with these findings, SNAT2 mRNA expression, mammary tissue amino acid transport systems respectively, increases and decreases during lac- together with their putative molecular correlates. tation and weaning. Moreover, the abundance of Mammary tissue amino acid transport mecha- SNAT2 mRNA is increased by oestrogen which nisms fall into two categories: Na+-dependent may explain the high levels observed during and Na+-independent systems. pregnancy (Lopez et al., 2006). System A activ- ity is regulated by prolactin: treating animals with 14.5.1.1 Na+-Dependent Transport bromocriptine, a drug which inhibits prolactin Mechanisms secretion from the pituitary gland, markedly reduces the arteriovenous concentration differ- Lactating mammary cells are able to concentrate ences of amino acids which are potential sub- free amino acids (particularly the non-essential strates of the A system (Vina et al., 1981). ones) with respect to plasma, suggesting that there must be an input of free energy (Shennan et al., Mammary tissue from several species (e.g., 1997). It is apparent that several amino acid trans- bovine, mouse, guinea pig) has been shown to pos- port systems utilize the electrochemical Na+ gra- ses system ASC (Baumrucker 1985; Mepham et al., 1985; Verma and Kansal 1993). It has been established in other tissues that this mechanism

14 Genetics and Biosynthesis of Milk Proteins 449 cotransports neutral amino acids such as alanine, Taurine, a nonprotein amino acid, is taken up threonine and cysteine with Na+. System ASC in by lactating rat and porcine mammary tissue via mammary tissue, like system A, operates with a high-affinity, Na+-dependent transport system relatively low affinity: the KM of methionine uptake analogous to system b (Shennan and McNeillie, by lactating mouse mammary tissue via system 1994b; Bryson et al., 2001). The mammary ASC is 0.46 mM (Verma and Kansal, 1993). At (Na++taurine) cotransport system also requires present, there appears to be a paucity of informa- Cl- for maximal activity. System b has narrow tion on the regulation of mammary tissue amino substrate specificity: only b-amino acids such as acid transport via system ASC except for the taurine and b-alanine interact with the trans- finding that starvation gives rise to a large increase porter. The activity of the rat mammary (Na+-Cl- in system ASC activity in mouse mammary tissue -taurine) cotransporter decreases as lactation (Verma and Kansal, 1995). There are at least two progresses (Millar and Shennan 1999). In accor- molecular isoforms of system ASC, both of which dance with this, Aleman et al. (2009) have have been detected in mammary tissue. Thus, the reported that the expression of rB16 (a cloned expression of ASCT1 and ASCT2 mRNA has taurine transporter) mRNA decreases between respectively been described in lactating rat and early and peak lactation. The lactating gerbil porcine mammary tissue (Aleman et al., 2009; mammary gland expresses a high-affinity (Na+- Laspiur et al., 2009). The expression of both tran- taurine) cotransport system which, unlike the rat scripts increases with the onset of lactation. mammary taurine transporter, is not dependent upon Cl- (Shennan, 1995). The transport of anionic amino acids by lactat- ing mammary tissue has been extensively studied 14.5.1.2 Na+-Independent Transport (Millar et al., 1996, 1997a, b). The predominant, Mechanisms if not the only, pathway for L-glutamate and L-aspartate transport is a high-affinity (KM = 18 System L has been identified in mouse, rat, guinea mM for L-glutamate), Na+-dependent system pig and bovine mammary tissue (Neville et al., analogous to system XAG- (Kanai et al., 1994). 1980; Mepham et al., 1985; Verma and Kansal, The Na+-dependent anionic amino acid carrier is 1993; Shennan and McNeillie, 1994c). System L very selective for anionic amino acids: it does not is a Na+-independent transport mechanism that interact with neutral or cationic amino acids has wide substrate specificity. Indeed, system L (Millar et al., 1996, 1997b). An unusual feature of may be the most important transport system for system XAG- is the ability to discriminate between the uptake of essential neutral amino acids by the the opical isomers of glutamate but not those of lactating mammary gland. Na+-independent aspartate. The high-affinity anionic amino acid amino acid transport sensitive to 2-aminobicyclo- carrier is able to act as an exchange system as well heptane-2-carboxylic acid (BCH) is taken as a as a cotransport mechanism suggesting that the measure of system L activity. It is generally transport of L-glutamate will affect the intracel- accepted that system L can act as an amino acid lular concentration of L-aspartate (and vice versa) exchange mechanism. Accordingly, methionine (Millar et al., 1997b). Several high-affinity anionic uptake by mouse mammary gland via system L amino acid carriers, which have identity with sys- can be trans-accelerated by intracellular amino tem X-AG, have been cloned and characterized acids. However, amino acid efflux from rat mam- (e.g., see Kanai and Hediger, 1992; Pines et al., mary tissue, via system L, is not trans-stimulated 1992; Storck et al., 1992). Three clones, EAAC1, by extracellular amino acids, suggesting that the GLAST and GLT-1, have been identified in rat transporter operates with asymmetric kinetics mammary tissue (Martinez-Lopez et al., 1999; which could favour the retention of substrates Aleman et al., 2009). However, the exact contri- within the gland. System L has been localized to bution of these isoforms to anionic amino acid the basolateral aspect of the lactating rat mam- transport across the basolateral membranes of mary gland (Shennan et al., 2002). There are at mammary secretory cells is unknown. least four molecular isoforms of system L

450 J.-L. Vilotte et al. (LAT 1–4), two of which have been described 1996). Thus, the moiety of tyrosine uptake by in the mammary gland. Thus, LAT1 and LAT2 mouse mammary tissue which is not dependent mRNA are expressed in the rat mammary gland upon extracellular Na+ and is not sensitive to (Shennan et al., 2002; Aleman et al., 2009). BCH has been ascribed to system T. This mecha- Interestingly, Aleman et al. (2009) have shown nism operates with low affinity; the KM for that LAT1 mRNA expression in rat mammary tyrosine transport is approximately 15 mM. tissue markedly increases during lactation. LAT mRNA is also expressed in the mouse and bovine 14.5.1.3 Volume-Activated Amino Acid mammary gland (Rudolph et al., 2007; Finucane Transport et al., 2008; Connor et al., 2008). The expression of LAT1 mRNA increases during the transition To survive, cells have to regulate their volume from pregnancy to lactation and is also up- within relatively narrow limits (see Hoffmann regulated by milking frequency. and Simonsen, 1989, for a review). Cell mem- branes are very permeable to water which means The transport of cationic amino acids by lac- that cell volume, otherwise termed the cellular tating mammary tissue is a Na+-independent pro- hydration state, will be determined by the osmo- cess (Baumrucker 1984; Shennan et al., 1994b; larity of the extracellular fluid and by the intrac- Hurley et al., 2000). The pathway for lysine and ellular content of osmotically active solutes. The arginine uptake by bovine mammary tissue is not cellular hydration state can be changed as a con- affected by replacing extracellular Na+ and sequence of anisosmotic conditions, cellular appears to be relatively specific for cationic accumulation of solutes or oxidative metabolism. amino acids. On this basis Baumrucker (1984) Cells are able to regulate their volume following concluded that cationic amino acids are trans- swelling or shrinking. Cell volume regulation ported via system y+. There is evidence suggest- involves the transmembrane movement of sol- ing that CAT-1 may be responsible for system y+ utes together with osmotically obliged water activity in rat mammary tissue (Aleman et al., (Hoffmann and Simonsen, 1989). 2009). CAT-1 mRNA expression is low in mam- mary tissue isolated from pregnant rats and A knowledge of volume-regulatory processes increases with the onset of lactation (Aleman in mammary tissue is of particular importance et al., 2009). Transcripts for two cloned Na+- given that cell volume changes markedly affect independent cationic amino acid transporters, the rate of milk protein synthesis (Millar et al., CAT-1 and CAT-2B, have been detected in lactat- 1997a). It has been established that cell swelling ing porcine tissue (Laspiur et al., 2009), However, activates the transport of amino acids in mam- it appears that the expression of both transcripts mary tissue (Shennan et al., 1994; 1996). Thus, is relatively unaffected by the stage of lactation. cell swelling, induced by a hypoosmotic shock, Cationic amino acid uptake by lactating rat mam- increases the efflux of amino acids such as tau- mary tissue is also facilitated by a transporter rine and glycine via a pathway which has the which interacts with neutral amino acids such as characteristics of a channel rather than a carrier. glutamine and leucine (Shennan et al., 1994a; The swelling-induced amino acid efflux pathway Calvert and Shennan 1996). Thus, certain neutral appears to be situated in the blood-facing aspect amino acids respectively inhibit and stimulate of the mammary epithelium (Calvert and lysine uptake by and efflux from rat mammary Shennan, 1998). The volume-activated efflux of tissue. It is likely that this pathway is y+L: taurine is dependent upon the extent of cell swell- y + LAT1 mRNA has been detected in lactating ing and can be inhibited by a number of pharma- rat mammary tissue (Boyd, Kudo and Shennan, cological agents such as niflumic acid (Shennan unpublished). et al., 1996). There is the strong possibility that volume-activated amino acid efflux may partici- System T, a mechanism specific for aromatic pate in mammary cell volume regulation given amino acids, has been described in lactating that amino acids are concentrated within mam- mouse mammary tissue (Kansal and Kansal, mary cells with respect to plasma.

14 Genetics and Biosynthesis of Milk Proteins 451 14.5.2 Transport and Metabolism demonstrated using both direct and indirect of Peptides experimental approaches (Shennan et al., 1998; 1999). Anionic dipeptides presented to rat mam- Although free amino acids are the major source mary tissue can trans-stimulate D-aspartate efflux of amino nitrogen for milk protein synthesis, it is via the high-affinity anionic amino acid carrier. If evident that the uptake of several essential amino it is accepted that dipeptides do not interact acids is less than their output in milk, suggesting directly with the amino acid carrier, then it can be that other circulating forms of amino acids, such assumed that anionic amino acids, produced as a as peptides, may be available for casein produc- consequence of extracellular hydrolysis, act to tion (e.g., see Backwell et al., 1996). In this con- stimulate D-aspartate efflux. In this connection, it nection, it has been demonstrated that the goat has been shown that mammary tissue is capable mammary gland is able to use intravenously of hydrolysing a variety of aminoacyl-p-nitroa- administered peptides for milk protein synthesis nilides (peptide analogues). Quantitatively, it (Backwell et al., 1994, 1996). The in vivo studies appears, at least in the rat, that hydrolysis of pep- were unable to show whether the mammary gland tides followed by uptake of the free amino acids transported peptides intact or whether the pep- may be more important than the transport of pep- tides were hydrolysed extracellularly followed tides (Shennan et al., 1998). by uptake of the liberated amino acids. It has been shown, albeit indirectly, that the mouse The identity of the peptidases involved in the mammary tissue does not significantly hydrolyse extracellular hydrolysis of peptides by mam- peptides extracellularly but is able to transport mary tissue has not been established. However, peptides intact (Wang et al., 1996). On the other g-glutamyltranspeptidase appears to be involved hand, studies using the rat mammary gland as a given that glutathione can stimulate D-aspartate model suggest that mammary tissue is able to efflux from rat mammary tissue in a fashion sen- both transport and hydrolyse dipeptides (Shennan sitive to acivicin (Shennan et al., 1998). The et al., 1998, 1999). It is evident that the perfused hydrolysis of glutathione may be an important lactating rat mammary is able to transport dipep- route for providing the mammary gland with tides which are resistant to hydrolysis (i.e., cysteine. There is evidence to suggest that amin- D-Phe-L-Gln; D-Phe-L-Glu); the nature of the opeptidase N plays a role in providing free pathway remains to be identified precisely. amino acids for protein synthesis in the goat However, it is apparent that the capacity of the rat mammary gland (Liu et al., 2010). mammary gland to transport peptides intact across the basolateral pole of the epithelium is References relatively limited (Shennan et al., 1998). Transcripts for two proton-dependent peptide Adachi, T., Ahn, J.Y., Yamamoto, K., Aoki, N., Nakamura, transporters, namely, PepT1 and PepT2, have R. and Matsuda, T. (1996). Characterization of the been identified in lactating rat and human mam- bovine kappa-casein gene promoter. Biosci. Biotech. mary epithelial cells (Alcorn et al., 2002; Biochem. 60, 1937–1940. Groneberg et al., 2002; Gilchrist and Alcorn, 2010). PepT2 protein is localized in the apical Aggeler, J., Park, C.S. and Bissell, M.J. (1988). Regulation membrane of rat ductal epithelial cells and is of milk protein and basement membrane gene expres- therefore not in a position to facilitate the uptake sion: the influence of the extracellular matrix. J. Dairy of peptides from the circulation. Rather, it is Sci. 71, 2830–2842. likely that PepT2 provides a route for the reuptake of peptides from milk as postulated by Shennan Alcorn, J., Moscow, J.A. and McNamara, P.J. (2002). and Peaker (2000). Transporter gene expression in lactating and nonlac- tating human mammary epithelial cells using real- Rat mammary tissue is able to extensively time reverse transcription-polymerase chain reaction. hydrolyse peptides extracellularly: this has been J. Pharmacol. Ther. 303, 487–496., Aleman, G., Lopez, A., Ordaz, G., Torres, N. and Tovar, A.R. (2009). Changes in messenger RNA abundance of amino acid transporters in rat mammary gland dur- ing pregnancy, lactation and weaning. Metabolism 58, 594–601.

452 J.-L. Vilotte et al. Alexander, L.J., Hayes, G., Bawden, W., Stewart, A.F. and Barker, G.A. and Ellory, J.C. (1990). The identification of MacKinlay, A.G. (1993). Complete nucleotide neutral amino acid transport systems. Exp. Physiol. sequence of the bovine beta-lactoglobulin gene. Anim. 75, 3–26. Biotechol. 4, 1–10. Baumrucker, C.R. (1984). Cationic amino acid transport by Alexander, L.J., Stewart, A.F., MacKinlay, A.G., bovine mammary tissue. J. Dairy Sci. 67, 2500–2506. Kapelinskaya, T.V., Tkach, T.M. and Gorodetsky, S.I. (1988). Isolation and characterization of the bovine Baumrucker CR. (1985). Amino acid transport systems in kappa-casein gene. Eur. J. Biochem. 178, 395–401 bovine mammary tissue. J. Dairy Sci. 68(9), 2436–2451. Ali, S. and Clark, J. (1988). Characterization of the gene Bingham, E.W. and Farrell, H.M. Jr (1974). Casein kinase encoding ovine beta-lactoglobulin. Similarity to the from the Golgi apparatus of lactating mammary gland. genes for retinol binding protein and other secretory J. Biol. Chem. 249, 3647–3651. proteins. J. Mol. Biol. 199, 415–426 Blobel, G. and Dobberstein, B. (1975). Transfer of pro- Altiok, S. and Groner, B. (1993). Interaction of two teins across membranes 1. Presence of proteolytically sequence-specific single-stranded DNA-binding pro- processed and unprocessed nascent immunoglobulin teins with an essential region of the beta-casein pro- light chains on membrane-bound ribosomes of murine moter is regulated by lactogenic hormones. Mol. Cell. myeloma. J. Cell Biol. 67, 835–851. Biol. 13, 7303–7310. Blum, J.L., Zeigler, M.E. and Wicha, M.S. (1989). Altiok, S. and Groner, B. (1994). Beta-casein mRNA Regulation of mammary differentiation by the extra- sequesters a single stranded nucleic acid-binding pro- cellular matrix. Environ. Health Perspect. 80, 71–83. tein which negatively regulates the beta-casein gene promoter. Mol. Cell. Biol. 14, 6004–6012. Boisgard, R. and Chanat, E. (2000). Phospholipase D-dependent and -independent mechanisms are Attal, J., Cajero-Juarez, M., Petitclerc, D., Theron, M.C., involved in milk protein secretion in rabbit mammary Stinnakre, M.G., Bearzotti, M., Kann, G. and epithelial cells. Biochim Biophys Acta 1495, 281–296. Houdebine, L.M. (1995). The effect of matrix attached regions (MAR) and specialized chromatin structure Boisgard, R., Charpigny, G. and Chanat, E. (1999). (SCS) on the expression of gene constructs in cultured Polymeric IgA are sulfated proteins. FEBS Lett 463, cells and transgenic mice. Mol. Biol. Rep. 22, 37–46. 250–254. Avril-Sassen, S., Goldstein, L.D., Stingl, J., Blenkiron, Boisnard, M., Hue, D., Bouniol, C., Mercier, J.C. and C., Le Quesne, J., Spiteri, I., Karagavriilidou, K., Gaye, P. (1991). Multiple mRNA species code for two Watson, C.J., Tavaré, S., Miska, E.A. and Caldas, C. non-allelic forms of ovine alpha-s2-casein. Eur. J. (2009). Characterisation of miRNA expression in Biochem. 201, 633–641 post-natal mouse mammary gland development. BMC Genomics 10, 548 Bonsing, J., Ring, J.M., Stewart, A.F. and MacKinlay, A.G. (1988). Complete nucleotide sequence of the bovine Backwell, F.R.C., Bequette, B.J., Wilson, D., Calder, beta-casein gene. Aust. J. Biol. Sci. 41, 527–537. A.G., Metcalf, J.A., Wray Cahen, D., MacRae, J.C., Beever, D.E. and Lobley, G.E. (1994). Utilization of Boström, P., Andersson, L., Rutberg, M., Perman, J., dipeptides by the caprine mammary-gland for protein- Lidberg, U., Johansson, B.R., Fernandez-Rodriguez, synthesis. Am. J. Physiol. 267, R1–R6. J., Ericson, J., Nilsson, T., Borén, J. and Olofsson, S.-O. (2007). SNARE proteins mediate fusion between Backwell, F.R.C., Bequette, B.J., Wilson, D., Metcalf, cytosolic lipid droplets and are implicated in insulin J.A., Franklin, M.F., Beever, D.E., Lobley, G.E. and sensitivity. Nat. Cell Biol. 9, 1286–1293. MacRae, J.C. (1996). Evidence for the utilization of peptides for milk protein synthesis in the lactating Boucheron, C., Dumon, S., Santos, S.C., Moriggl, R., dairy goat in vivo. Am. J. Physiol. 271, R955–R960. Hennighausen, L., Gisselbrecht, S. and Gouilleux, F. (1998). A single amino acid in the DNA binding Baranyi, M., Aszodi, A., Devinoy, E., Fontaine, M.L., regions of Stat5A and Stat5B confers distinct binding Houdebine, L.M. and Bosze, Z. (1996). Structure of specificities. J. Biol. Chem. 273, 33936–33941. the rabbit kappa-casein encoding gene: expression of the cloned gene in the mammary gland of transgenic Bouguyon, E., Beauvallet, C., Huet, J.C. and Chanat, E. mice. Gene 174, 27–34. (2006). Disulphide bonds in casein micelle from milk. Biochem. Biophys. Res. Commun. 343, 450–458. Baranyi, M., Brignon, G., Anglade, P. and Ribadeau- Dumas, B. (1995). New data on the proteins of rabbit Brew, K., Castellino, F.J., Vanaman, T.C., Hill, R.L. (Oryctolagus cuniculus) milk. Comp. Biochem. (1970). The complete amino acid sequence of bovine Physiol. B. Biochem. Mol. Biol. 111, 407–415. alpha-lactalbumin. J. Biol. Chem. 245, 4570–4582. Barash, I. (2006). Stat5 in the mammary gland: control- Bryson, J.M., Jackson, S.C., Wang, H. and Hurley, W.L. ling normal development and cancer. J. Cell Physiol. (2001). Cellular uptake of taurine by lactating porcine 209, 305–313 mammary gland. Comp. Biochem. Physiol. B. 128, 667–673. Bargmann, W. and Knoop, A. (1959). Morphology of lac- tation; light and electro-microscopic studies on the Burdon, T.G., Maitland, K.A., Clark, A.J., Wallace, R. mammary glands of rats. Z. Zellforsch. Mikrosk. Anat. and Watson, C.J. (1994). Regulation of the sheep beta- 49, 344–388. lactoglobulin gene by lactogenic hormones is medi- ated by a transcription factor that binds an interferon gamma activation site-related element. Mol. Endocrinol. 8, 1528–1536.

14 Genetics and Biosynthesis of Milk Proteins 453 Buser, A.C., Gass-Handel, E.K.., Wyszomierski, S.L., mediated by elements in the promoter and 3¢ untrans- Doppler, W., Leonhardt, S.A., Schaack, J., Rosen, lated region. Mol. Cell. Biol. 12, 905–914. J.M., Watkin, H., Anderson, S.M. and Edwards, D.P. Davies, M.S., West, L.F., Davis, M.B., Povey, S. and (2007). Progesterone receptor repression of prolactin/ Craig, R.K. (1987). The gene for human alpha-lactal- signal transducer and activator of transcription 5-medi- bumin is assigned to chromosome 12q13. Ann. Hum. ated transcription of the beta-casein gene in mammary Genet. 51, 183–188 epithelial cells. Mol Endocrinol.1, 106–25. Dawson, S.P., Wilde, C.J., Tighe, P.J. and Mayer, R.J. (1993). Characterization of two novel casein transcripts Calvert, D.T. and Shennan, D.B. (1996). Evidence for an in rabbit mammary gland. Biochem. J. 15, 777–784. interaction between cationic and neutral amino acids Demmer, J., Burdon, T.G., Djiane, J., Watson, C.J. and at the blood-facing aspect of the lactating rat mam- Clark, A.J. (1995). The proximal milk protein binding mary epithelium. J. Dairy Res. 63, 25–33. factor binding site is required for the prolactin respon- siveness of the sheep beta-lactoglobulin promoter in Calvert, D.T. and Shennan, D.B. (1998). Volume-activated Chinese hamster ovary cells. Mol. Cell. Endocrinol. taurine efflux from the in situ perfused lactating rat 107, 113–121. mammary gland. Acta Physiol. Scand. 162, 97–105. Derbinski, J., Pinto, S., Rösch, S., Hexel, K. and Kyewski, B. (2008). Promiscuous gene expression patterns in Campbell, S.M., Rosen, J.M., Hennighausen, L.G., single medullary thymic epithelial cells argue for a Strech-Jurk, U. and Sippel, A.E. (1984). Comparison stochastic mechanism. Proc. Natl. Acad. Sci. U.S.A. of the whey acidic protein genes of the rat and mouse. 105, 657–662 Nucleic Acids Res. 12, 8685–8697 Devinoy, E., Hubert, C., Schaerer, E., Houdebine, L.M. and Kraehenbuhl, J.P. (1988). Sequence of the rabbit whey Chanat, E. (2006). Sulphated proteins secreted by rat mam- acidic protein cDNA. Nucleic Acids Res. 16, 8180. mary epithelial cells. Reprod Nutr Dev 46, 557–566. Devinoy, E., Stinnakre, M.G., Lavialle, F., Thepot, D. and Ollivier-Bousquet, M. (1995). Intracellular routing Chanat, E. and Huttner, W.B. (1991). Milieu-induced, and release of caseins and growth hormone produced selective aggregation of regulated secretory proteins in into milk from transgenic mice. Exp. Cell Res. 221, the trans-Golgi network. J. Cell. Biol. 115, 1505–1519. 272–280. Dobie, K.W., Lee, M., Fantes, J.A., Graham, E., Clark, A.J., Chanat, E., Martin, P. and Ollivier-Bousquet, M. (1999). Springbett, A., Lathe, R. and McClenaghan, M. (1996). as1-Casein is required for the efficient transport of Variegated transgene expression in mouse mammary b- and k-casein from the endoplasmic reticulum to gland is determined by the transgene integration locus. the Golgi apparatus of mammary epithelial cells. Proc. Natl. Acad. Sci. U.S.A. 93, 6659–6664. J. Cell Sci. 112 (Pt 19), 3399–3412. Ebner, K. and Brodbeck, U. (1968). Biological role of alpha-lactalbumin: a review. J Dairy Sci 51, 317–322. Chat, S., Layani, S., Mahaut, C., Henry, C., Chanat, E. and Edlund, A., Johansson, T., Leidvik, B. and Hansson, L. Truchet, S. (2011). Characterisation of the potential (1996). Structure of the human kappa-casein gene. SNARE proteins relevant to milk product release by Gene 174, 65–69. mouse mammary epithelial cells. Eur. J. Cell. Biol. 90, Edwards, G.M., Wilford, F.H., Liu, X., Hennighausen, L., 401–413. Djiane, J. and Streuli, C.H. (1998). Regulation of mammary differentiation by extracellular matrix Clermont, Y., Xia, L., Rambourg, A., Turner, J.D. and involves protein-tyrosine phosphatases. J. Biol. Chem. Hermo, L. (1993). Transport of casein submicelles and 17, 9495–9500. formation of secretion granules in the Golgi apparatus Eisenstein, R.S. and Rosen, J.M. (1988). Both cell sub- of epithelial cells of the lactating mammary gland of stratum regulation and hormonal regulation of milk the rat. Anat. Rec. 235, 363–373. protein gene expression are exerted primarily at the posttranscriptional level. Mol. Cell. Biol. 8, Collet, C. and Joseph, R. (1995). Exon organization and 3183–3190. sequence of the genes encoding alpha-lactalbumin and Faerman, A., Barash, I., Puzis, R., Nathan, M., Hurwitz, beta-lactoglobulin from the Tammar Wallaby D.R. and Shani, M. (1995). Dramatic heterogeneity of (Macropodidae, Marsupialia). Biochem. Genet. 33, transgene expression in the mammary gland of lactat- 61–72. ing mice: a model system to study the synthetic activ- ity of mammary epithelial cells. J. Histochem. Collet, C., Joseph, R. and Nicholas, K. (1990). Cloning, Cytochem. 43, 461–470. cDNA analysis and prolactin-dependent expression of Finucane, K.A., McFadden, T.B., Bond, J.P., Kenelly, J.J. a marsupial alpha-lactalbumin. Reprod. Fertil. Dev. 2, and Zhao, F.-Qi. (2008). Onset of lactation in the 693–701. bovine mammary gland: gene expression profiling indicates a strong inhibition of gene expression in cell Connor, E.E., Siferd, S., Elsasser, T.H., Evock-Clover, proliferation. Funct. Integr. Genomics 8, 251–264. C.M., Van Tassel, C.P., Sonstegard, T.S., Fernandes, V.M. and Capuco, A.V. (2008). Effects of increased milking frequency on gene expression in the bovine mammary gland. BMC Genomics 9, 362 Copeland, C.S., Doms, R.W., Bolzau, E.M., Webster, R.G. and Helenius, A. (1986). Assembly of influenza hemagglutinin trimers and its role in intracellular transport. J. Cell. Biol. 103, 1179–1191. Dale, T.C., Krnacik, M.J., Schmidhauser, C., Yang, C.L.Q., Bissell, M.J. and Rosen, J.M. (1992). High- level expression of the rat whey acidic protein gene is

454 J.-L. Vilotte et al. Farrell, H. M. Jr., Malin, E.L., Brown, E.M. and Qi, P.X. Glick, B.S., Malhotra, Y. (1998). The curious status of the (2006). Casein micelle structure: what can be learned Golgi apparatus. Cell 95, 883–889. from milk synthesis and structural biology? Curr. Opin. Coll. Interf. Sci. 11, 135–147. Glombik, M.M. and Gerdes, H.H. (2000). Signal-mediated sorting of neuropeptides and prohormones: secretory Flower, D.R. (1996). The lipocalin protein family: struc- granule biogenesis revisited. Biochimie 82, 315–326. ture and function. Biochem. J. 318, 1–14. Golden, K.L. and Rillema, J.A. (1995). Effects of prolac- Folch, J.M., Coll, A. and Sanchez, A. (1994). Complete tin on galactosyl transferase and alpha-lactalbumin sequence of the caprine beta-lactoglobulin gene. mRNA accumulation in mouse mammary gland J. Dairy Sci. 77, 3493–3497. explants. Proc. Soc. Exp. Biol. Med. 209, 392–396. Folch, J.M., Coll, A., Hayes, H.C. and Sanchez, A. Groenen, M.A.M., Dijkhof, R.J.M., Verstege, A.J.M. and (1996). Characterization of a caprine beta-lactoglobu- van der Poel, J.J. (1993). The complete sequence of lin pseudogene, identification and localization by in the gene encoding bovine alpha-s2-casein. Gene 123, situ hybridization in goat, sheep and cow. Gene 24, 187–193. 87–91. Groneberg, D.A., Doring, F., Theis, S., Nickolaus, M., Fujiwara, Y., Miwa, M., Nogami, M., Okumura, K., Fischer, A. and Daniel, H. (2002). Peptide transport in Nobori, T., Suzuki, T., Ueda, M. (1997). Genomic the mammary gland: expression and distribution of organization and chromosomal localization of the PEPT2 mRNA and protein. Am. J. Physiol. Endocrinol. human casein gene family. Hum. Genet. 99, 368–373. Metab. 282, E1172–E1179. Fujiwara, Y., Miwa, M., Takahashi, R., Kodaira, K., Grosclaude, F. (1979). Polymorphism of milk proteins: Hirabayashi, M., Suzuki, T. and Ueda, M. (1999). some biochemical and genetical aspects, in, Proc.16th High-level expressing YAC vector for transgenic ani- International Conference Animal Blood Groups mal bioreactors. Mol. Reprod. Dev. 52, 14–20. Biochemical Polymorphisms, International Society for Animal Blood Group Research, ed., International Funder, J.W. (1989). Hormonal regulation of gene expres- Society for Animal Blood Group Research, Leningrad. sion. Biochem. Soc. Symp. 55, 105–114. pp. 54–92. Gallagher, D.S., Schelling, C.P., Groenen, M.M.A. and Grusby, M.J., Mitchell, S.C., Nabavi, N. and Glimcher, Womack, J.E. (1994). Confirmation that the casein L.H. (1990). Casein expression in cytotoxic T lympho- gene cluster resides on cattle chromosome 6. Mamm. cytes. Proc. Natl. Acad. Sci. U.S.A 87, 6897–6901. Genome 5, 524. Gu, Z., Eleswarapu, S. and Jiang, H. (2007). Identification Gallagher, D.S., Treadgill, D.W., Ryan, A.M., Womack, and characterization of microRNAs from the bovine J.E. and Irwin, D.M. (1993). Physical mapping of the adipose tissue and mammary gland. FEBS Lett. 581, lysozyme gene family in cattle. Mamm. Genome 4, 981–988. 368–373. Gupta, P., Rosen, J.M., D’Eustachio, P. and Ruddle, F.H. Geissler, E.N., Cheng, S.V., Gusella, J.F. and Housman, (1982). Localization of the casein gene family to a D.E. (1988). Genetic analysis of the dominant white- single mouse chromosome. J. Cell Biol. 93, 199–204 spotting (W) region on mouse chromosome 5: identification of clones DNA markers near W. Proc. Gutierrez-Adan, A., Maga, E.A., Meade, H., Shoemaker Natl. Acad. Sci. U.S.A. 85, 9635–9639 C.F., Medrano J.F., Anderson G.B. and Murray J.D. (1996). Alterations of the physical characteristics of Gellin, J., Echard, G., Yerle, M., Dalens, M., Chevalet, milk from transgenic mice producing bovine kappa- C. and Gillois, M. (1985). Localization of the alpha casein. J. Dairy Sci. 79, 791–799. and beta casein genes to the q24 region of chromo- some 12 in the rabbit (Oryctolagus cuniculus L.) by Guyette, W.A., Matusik, R.J. and Rosen, J.M. (1979). in situ hybridization. Cytogenet. Cell Genet. 39, Prolactin-mediated transcriptional and post-transcrip- 220–223. tional control of casein gene expression. Cell 17, 1013–1023 George, S., Clark, A.J. and Archibald, A.L. (1997). Physical mapping of the murine casein locus reveals Hajjoubi, S., Rival-Gerbier, S., Hayes, H., Floriot, S., the gene order as alpha-beta-gamma-epsilon-kappa. Eggen, A., Piumi, F., Chardon, P., Houdebine, L.M. DNA Cell. Biol. 16, 477–484. and Thepot, D (2006). Ruminants genome no longer contains whey acidic protein gene but only a pseudo- Geyer, P.K. (1997). The role of insulator elements in gene. Gene 370, 104–112 defining domains of gene expression. Curr. Opin. Genet. Dev. 7, 242–248. Hall, L., Emery, D.C., Davies, M.S., Parker, D. and Craig, R.K. (1987). Organization and sequence of the human Gilchrist, S.E. and Alcorn J (2010). Lactation stage- alpha-lactalbumin gene. Biochem. J. 242, 735–742 dependent expression of transporters in rat whole mammary gland and primary epithelial organoids. Halliday, J.A., Bell, K., McKenzie, H.A. and Shaw, D.C. Fundam. Clin. Pharmacol. 24, 205–214. (1990). Feline whey proteins: identification, isolation and initial characterization of alpha-lactalbumin, beta- Gilmour, K.C., Pine, R. and Reich, N. (1995). Interleukin lactoglobulin and lysozyme. Comp. Biochem. Physiol. 2 activates STAT5 transcription factor (mammary 95, 773–779. gland factor) and specific gene expression in T lym- phocytes. Proc. Natl. Acad. Sci. U.S.A. 92, Hammond, C. and Helenius, A. (1995). Quality control in the 10772–10776. secretory pathway. Curr. Opin. Cell. Biol. 7, 523–529.

14 Genetics and Biosynthesis of Milk Proteins 455 Hansson, L., Edlund, A., Johansson, T., Hernell, O., Jollès, J., Fiat, A.M., Schoentgen, F., Alais, C. and Jollès, Strömqvist, M., Lindquist, S., Lönnerdal, B. and P. (1974). The amino acid sequence of sheep kappa- Bergström, S. (1994). Structure of the human beta- A-casein. II. Sequence studies concerning the casein encoding gene. Gene 139, 193–199. kappa-A-caseinoglycopeptide and establishment of the complete primary structure of the protein. Biochim. Harris, S., Ali, S., Anderson, S., Archibald, A.L. and Biophys. Acta 365, 335–342. Clark, A.J. (1988). Complete nucleotide sequence of the genomic ovine beta-lactoglobulin gene. Nucleic Jones, W.K., Yu-Lee, L.Y., Clift, S.M., Brown, T.L. and Acids Res. 16, 10379–10380 Rosen, J.M. (1986). The rat casein multigene family. Fine structure and evolution of the beta-casein gene. Harris, S., McClenaghan, M., Simons, J.P., Ali, S. and J. Biol. Chem. 260, 7042–7050 Clark, A.J. (1990). Gene expression in the mammary gland. J. Reprod. Fertil. 88, 707–715 Kabotyanski, E.B., Huetter, M., Xian, W., Rijnkels, M. and Rosen, J.M. (2006). Integration of prolactin and Hayes, H., Petit, E., Bouniol, C. and Popescu, P. (1993a). glucocorticoid signaling at the beta-casein promoter Localization of the alpha-S2-casein gene (CASAS2) and enhancer by ordered recruitment of specific tran- to the homeologous cattle, sheep and goat chromo- scription factors and chromatin modifiers. Mol. somes 4 by in situ hybridization. Cytogenet. Cell. Endocrinol. 10, 2355–2368. Genet. 64, 282–285. Kabotyanski, E.B., Rijnkels, M., Freeman-Zadrowski, C., Hayes, H., Petit, E., Lemieux, N. and Dutrillaux, B. Buser, A.C., Edwards, D.P. and Rosen, J.M. (2009). (1992). Chromosomal localization of the ovine beta- Lactogenic hormonal induction of long distance inter- casein gene by non-isotopic in situ hybridization and actions between beta-casein gene regulatory elements. R-banding. Cytogenet. Cell. Genet. 61, 286–288. J. Biol. Chem. 284, 22815–22824. Hayes, H., Popescu, P. and Dutrillaux, B. (1993b). Kanai, Y. and Hediger, M.A. (1992). Primary structure Comparative gene mapping of lactoperoxidase, retino- and functional characterization of a high-affinity glu- blastoma and alpha-lactalbumin genes in cattle, sheep tamate transporter. Nature 360, 467–471. and goats. Mamm. Genome 4, 593–597. Kanai, Y., Smith, C.R. and Hediger, M.A. (1994). A new Hayes, H.C. and Petit, E.J. (1993c). Mapping of the beta- family of neurotransmitter transporters: the high affinity lactoglobulin gene and of an immunoglobulin M heavy glutamate transporters. FASEB J. 8, 1450–1459. chain-like sequence to homoeologous cattle sheep, and goat chromosomes. Mamm. Genome 4, 207–210. Kang, Y.K., Lee, C.S., Chung, A.S. and Lee, K.K. (1998). Prolactin-inducible enhancer activity of the first intron Heim, M.H. (1999). The Jak-STAT pathway: cytokine sig- of the bovine beta-casein gene. Mol. Cells 30, nalling from the receptor to the nucleus. J. Recept. 259–265. Signal Transduct. Res. 19, 75–120. Kansal, R. and Kansal, V.K. (1996). Discrimination of Hennighausen, L.G. and Sippel, A.E. (1982). Mouse whey transport systems of L-tyrosine in mouse mammary acidic protein is a novel member of the family of four- gland: characterisation of system T. Indian J. Exp. disulfide core proteins. Nucleic Acids Res. 10, Biol. 34, 750–757. 2677–2684. Kapelinskaia, T.V., Tkach, T.M., Smirnov, I.K. and Hentze, M. and Kulozik, A.E. (1999). A perfect message: Gorodetskii, S.I. (1989). The Bos taurus casein genes. RNA surveillance and nonsense-mediated decay. Cell Isolation and characterization of the kappa-casein 96, 307–310. gene. Genetika 25, 15–23. Hoffmann, E.K. and Simonsen, L.O. (1989). Membrane Kawasaki, K., Lafont, A.G. and Sire, J.Y. (2011). The mechanisms in volume and pH regulation in vertebrate evolution of milk casein genes from tooth genes cells. Physiol. Rev. 69, 315–382. before the origin of mammals. Mol Biol Evol. 28, 2053–2061. Holt, C. and Sawyer, L. (1993). Caseins as rheomorphic proteins: interpretation of the primary and secondary Kazansky, A.V., Raught, B., Lindsey, S.M., Wang, Y.F. structures of the as1-, b- and k-caseins. J. Chem. Soc. and Rosen, J.M. (1995). Regulation of mammary Farad. Trans. 89, 2683–2692. gland factor/Stat5 during mammary gland develop- ment. Mol. Endocrinol. 9, 1598–1609. Hurley, W.L., Wang, H., Bryson, J.M. and Shennan D.B. (2000). Lysine uptake by mammary gland tissue from Kim, P.S. and Arvan, P. (1991). Folding and assembly of lactating sows. J. Anim. Sci. 78, 391–395. newly synthesized thyroglobulin occurs in a pre-Golgi compartment. J. Biol. Chem. 266, 12412–12418. Jahn, R. and Scheller, R.H. (2006). SNAREs–engines for membrane fusion. Nat. Rev. Mol. Cell. Biol. 7, Koczan, D., Hobom, G. and Seyfert, H.M. (1991). 631–643. Genomic organization of the bovine as1-casein gene. Nucleic Acids Res. 19, 5591–5596. Jolivet, G., Devinoy, E., Fontaine, M.L. and Houdebine, L.M. (1992). Structure of the gene encoding rabbit Kolb, A.F., Günzburg, W.H., Albang, R., Brem, G., Erfle, alpha s1-casein. Gene 113, 257–262. V. and Salmons, B. (1994). Negative regulatory ele- ment in the mammary specific whey acidic protein Jolivet, G., L’Hotte, C., Pierre, S., Tourkine, N. and promoter. J. Cell. Biochem. 56, 245–261. Houdebine, L.M. (1996). The MGF/STAT5 binding site is necessary in the distal enhancer for high prolac- Kolb, A. F. (2002). Structure and regulation of the murine tin induction of transfected rabbit alpha-s1-casein- g-casein gene. Biochem. Biophys. Acta 1579, 101–116. CAT gene transcription. FEBS Lett. 8, 257–262.

456 J.-L. Vilotte et al. Kolb, A. F. (2003). The first intron of the murine b-casein Li, S. and Rosen, J.M. (1994a). Distal regulatory elements gene contains a functional promoter. Biochem. required for rat whey acidic protein gene expression in Biophys. Res. Commun. 306, 1099–1105. transgenic mice. J. Biol. Chem. 269, 14235–14243. Kumar, S., Clarke, A.R., Hooper, M.L., Horne, D.S., Law, Li, S. and Rosen, J.M. (1994b). Glucocorticoid regulation of A.J., Leaver, J., Springbett, A., Stevenson, E. and rat whey acidic protein gene expression involves hor- Simons, J.P. (1994). Milk composition and lactation of mone-induced alterations of chromatin structure in the beta-casein deficient mice. Proc. Natl. Acad. Sci. distal promoter region. Mol. Endocrinol. 8, 1328–1335. U.S.A 91, 6138–6142. Li, S. and Rosen, J.M. (1995). CTF/NF1 and mammary Laird, J.E., Jack, L., Hall, L., Boulton, A., Parker, D. and gland factor (Stat5) play a critical role in regulating rat Craig, R.K. (1988). Structure and expression of the whey acidic protein gene expression in transgenic guinea-pig alpha-lactalbumin gene. Biochem. J. 254, mice. Mol. Cell. Biol. 15, 2063–2070. 85–94 Liao, Y., Du, X. and Lönnerdal, B. (2010). miR-214 regu- Laspiur, J.P., Burton, J.L., Weber, P.S.D., Moore, J., lates lactoferrin expression and pro-apoptotic function Kirkwood, R.N. and Trottier, N.L. (2009). Dietary in mammary epithelial cells. J. Nutr. 140, 1552–1556. protein intake and stage of lactation differentially modulate amino acid transporter mRNA abundance in Lin, C.Q., Dempsey, P.J., Coffey, R.J. and Bissell, M.J. porcine mammary tissue. J. Nutr. 139, 1677–1684. (1995). Extracellular matrix regulates whey acidic protein gene expression by suppression of TGF-alpha Lear, T.L., Brandon, R., Masel, A., Bell, K. and Bailey, E. in mouse mammary epithelial cells: studies in culture (1998). Molecular cloning and chromosomal localiza- and in transgenic mice. J. Cell Biol. 129, 1115–1126. tion of horse alpha-1-antitrypsin (AAT), beta-lacto- globulin 1 and 2 (BLG, BLG2), lactotransferrin (LTF) Lingappa, V.R., Lingappa, J.R., Prasad, R., Ebner, K.E. and transferrin (TF). Anim. Genet. 29, 43–49. and Blobel, G. (1978). Coupled cell-free synthesis, segregation, and core glycosylation of a secretory pro- Lechner, J., Welte, T., Tomasi, J.K., Bruno, P., Cairns, C., tein. Proc. Natl. Acad. Sci. U. S. A. 75, 2338–2342. Gustafsson, J.A. and Doppler, W. (1997). Promoter- dependent synergy between glucocorticoid receptor Liu, H., Wang, L., Wang, L.-B., Li, S.-L. and Cao, Z.-J. and Stat5 in the activation of beta-casein gene tran- (2010). Responses of mammary amino acid metabo- scription. J. Biol. Chem. 272, 20954–20960. lism and aminopeptidase N gene expression to duode- nal soyabean small peptides and infusion of free amino Lee, C.S. and Oka, T. (1992). A pregnancy-specific mam- acids in lactating goats. J. Anim. Feed Sci. 19, 24–36. mary nuclear factor involved in the repression of the mouse beta-casein gene transcription by progesterone. Liu, X., Robinson, G.R., Wagner, K.U., Garrett, L., J. Biol. Chem. 267, 5797–5801. Wynshaw-Boris, A. and Hennighausen, L. (1997). Stat5a is mandatory for adult mammary gland devel- Lee, K.F., Atiee, S.H., Henning, S.J. and Rosen, J.M. opment and lactogenesis. Genes Dev. 11, 179–186. (1989). Relative contribution of promoter and intragenic sequences in the hormonal regulation of rat Lopez, A., Torres, N., Ortiz, V., Aleman, G., Hernandez- beta-casein transgenes. Endocrinology 3, 447–453. Pando, R. and Tovar, A.R. (2006). Characterization and regulation of the gene expression of amino acid trans- Lee, M.C., Miller, E.A., Goldberg, J., Orci, L. and port system A (SNAT2) in rat mammary gland. Am. J. Schekman, R. (2004). Bi-directional protein transport Physiol. Endocrinol. Metab. 291, E1059–E1066. between the ER and Golgi. Ann. Rev. Cell Dev. Biol. 20, 87–123. Lollivier, V., Marnet, P.G., Delpal, S., Rainteau, D., Achard, C., Rabot, A. and Ollivier-Bousquet, M. (2006). Lefèvre, C.M., Sharp, J.A. and Nicholas, K.R. (2009). Oxytocin stimulates secretory processes in lactating rab- Characterisation of monotreme caseins reveals lineage- bit mammary epithelial cells. J. Physiol. 570, 125–140. specific expansion of an ancestral casein locus in mammals. Reprod. Fertil. Dev. 21, 1015–1027. Martinez-Lopez, I., Garcia, C., Barber, T., Vina, J.R. and Miralles, V.J. (1999). The L-glutamate transporters Lemay D.G., Lynn D.J., Martin W.F., Neville M.C., GLAST (EAAT1) and GLT-1(EAAT2): expression Casey T.M., Rincon G., Kriventseva E.V., Barris and regulation in the rat lactating mammary gland. W.C., Hinrichs A.S., Molenaar A.J., Pollard K.S., Mol. Membr. Biol. 15, 237–242. Maqbool N.J., Singh K., Murney R., Zdobnov E.M., Tellam R.L., Medrano J.F., German J.B. and Rijnkels Maschio, A., Brickell, P.M., Kioussis, D., Mellor, A.L., M. (2009). The bovine lactation genome: insights Katz, D. and Craig, R.K. (1991). Transgenic mice car- into the evolution of mammalian milk. Genome Biol. rying the guinea-pig alpha-lactalbumin gene transcribe 10, R43 milk protein genes in their sebaceous glands during lactation. Biochem. J. 275, 459–467. Le Parc, A., Leonil, J. and Chanat, E. (2010). as1-Casein, which is essential for efficient ER-to-Golgi casein Mather, I.H. and Keenan, T.W. (1983). Function of endo- transport, is also present in a tightly membrane-associ- membranes and the cell surface in the secretion of organic ated form. BMC Cell Biol. 11, 65–76. milk constituents, in, Biochemistry of Lactation. T.B. Mepham, ed. Amsterdam, Elsevier IP. pp. 231–283. Leroux, C., Mazure, N. and Martin, P. (1992). Mutations away from splice site recognition sequences might cis- Mather, I.H. and Keenan, T.W. (1998). Origin and secre- modulate alternative splicing of goat alpha-S1-casein tion of milk lipids. J. Mammary Gland Biol. Neoplasia transcripts. Structural organization of the relevant 3, 259–273. gene. J. Biol. Chem. 267, 6147–6157. McConkey, E.H., Menon, R., Williams, G., Baker, E. and Sutherland, G.R. (1996). Assignment of the gene for

14 Genetics and Biosynthesis of Milk Proteins 457 beta-casein (CSN2) to 4q13 - > q21 in humans and Moore, A., Hall, L. and Hamilton, D.W. (1990). An 18-kDa 3p13 - > p12 in chimpanzees. Cytogenet. Cell Genet. androgen-regulated protein that modifies galactosyl- 72, 60–62. transferase activity is synthesized by the rat caput McKnight, R.A., Spencer, M., Dittmer, J., Brady, J.N., epididymidis, but has no structural similarity to rat Wall, R.J. and Hennighausen, L. (1995). An Ets site in milk alpha-lactalbumin. Biol. Reprod. 43, 497–506. the whey acidic protein gene promoter mediates tran- scriptional activation in the mammary gland of preg- Morvan, J. and Tooze, S.A. (2008). Discovery and prog- nant mice but is dispensable during lactation. Mol. ress in our understanding of the regulated secretory Endocrinol. 9, 717–724. pathway in neuroendocrine cells. Histochem. Cell McKnight, R.A., Spencer, M., Wall, R.J. and Biol. 129, 243–252. Hennighausen, L. (1996). Severe position effects imposed on a 1 kb mouse whey acidic protein gene Myers, C.A., Schmidhauser, C., Mellentin-Michelotti, J., promoter are overcome by heterogeneous matrix Fragoso, G., Roskelley, C.D., Casperson, G., Mossi, attachment regions. Mol. Reprod. Dev. 44, 179–184. R., Pujuguet, P., Hager, G. and Bissell, M.J. (1998). Meier, V.S. and Groner, B. (1994). The nuclear factor Characterization of BCE-1, a transcriptional enhancer YY1 participates in repression of the beta-casein gene regulated by prolactin and extracellular matrix and promoter in mammary epithelial cells and is counter- modulated by the state of histone acetylation. Mol. acted by mammary gland factor during lactogenic hor- Cell. Biol. 18, 2184–2195. mone induction. Mol. Cell. Biol. 14, 128–137. Mepham, T.B., Overthrow, J.I. and Short A.H. (1985). Neville, M.C., Chatfield, K., Hansen, L., Lewis, A., Epithelial cell entry and exit competition amongst Monks, J., Nuijens, J., Ollivier-Bousquet, M., amino acids in the perfused isolated lactating mam- Schanbacher, F., Sawicki, V. and Zhang, P. (1998). mary gland of guinea pig, in, Carrier Mediated Lactoferrin secretion into mouse milk. Development Transport of Solutes from Blood to Tissue, D.L. of secretory activity, the localization of lactoferrin in Yudilevich and G.E. Mann, eds., Longman, London. the secretory pathway, and interactions of lactoferrin pp 369–372. with milk iron. Adv. Exp. Med. Biol. 443, 141–153. Mercier, J.C. and Vilotte, J.L. (1993). Structure and func- tion of milk protein genes. J. Dairy Sci. 76, 3079–3098. Neville, M.C., Lobitz, C.J., Ripoll, E.A. and Tinney, C. Millar, I.D., Barber, M.C., Lomax, M.A., Travers, M.T. (1980). The sites of a-aminoisobutyric acid uptake and Shennan, D.B. (1997a). Mammary protein synthe- in normal mammary gland and ascites tumor cells. sis is acutely regulated by the cellular hydration state. J. Biol. Chem. 255, 7311–7316. Biochem. Biophys. Res. Commun. 230, 351–355. Millar, I.D., Calvert, D.T., Lomax, M.A. and Shennan, Neville, M.C. and Watters, C.D. (1983). Secretion of cal- D.B. (1996). The mechanism of L-glutamate transport cium into milk: review. J. Dairy Sci. 66, 371–380. by lactating rat mammary tissue. Biochim. Biophys. Acta 1282, 200–206. Ollivier-Bousquet, M. (1993). Secrétion des caséines: Millar, I.D., Calvert, D.T., Lomax, M.A. and Shennan, régulation hormonale, in, Biologie de la Lactation, D.B. (1997b). Substrate specificity of the mammary Martinet, J. and Houdebine, L.M. eds., pp. 367–387. tissue anionic amino acid carrier operating in the NRA-INSERM, France. cotransport and exchange modes. Biochim. Biophys. Acta 1326, 92–102. Ollivier-Bousquet, M. (1997). Milk protein transfer in the Millar, I.D. and Shennan, D.B. (1999). The regulation of mammary cell. Flem. Vet. J. 66 (Suppl.), 125–142. Na+-dependent anionic amino acid transport by the rat mammary gland. Biochim. Biophys. Acta 1421, Onoda, M. and Inano, H. (1997). Distribution of casein- 340–346. like proteins in various organs of rat. J. Histochem. Mizoguchi, Y., Kim, J.Y., Enami, J. and Sakai, S. (1997). Cytochem. 45, 663–674. The regulation of the prolactin receptor gene expres- sion in the mammary gland of early pregnant mouse. Palade, G. (1975). Intracellular aspects of the process of Endocr. J. 44, 53–58. protein synthesis. Science 189, 347–358. Molenaar, A.J., Davis, S.R. and Wilkins, R.J. (1992). Expression of alpha-lactalbumin, alpha-s1-casein Papiz, M.Z., Sawyer, L., Eliopoulos, E.E., North, A.C.T., and lactoferrin genes is heterogeneous in sheep and Findlay, J.B.C., Sivaprasadavao, R., Jones, T.A., cattle mammary tissue. J. Histochem. Cytochem. 40, Newcomer, M.E. and Kraulis, P.J. (1986). The struc- 611–618. ture of beta-lactoglobulin and its similarity to plasma Montazer-Torbati, M.B., Hue-Beauvais, C., Droineau, S., retinol-binding protein. Nature 324, 383–385. Ballester, M., Coant, N., Aujean, E., Petitbarat, M., Rijnkels, M. and Devinoy, E. (2008). Epigenetic Passey, R.J. and McKinlay, A.G. (1995). Characterization modifications and chromatin loop organization explain of a second, apparently inactive, copy of the bovine beta- the different expression profiles of the Tbrg4, WAP lactoglobulin gene. Eur. J. Biochem. 233, 736–743. and Ramp3 genes. Exp. Cell Res. 314, 975–987. Pauloin, A., Delpal, S., Chanat, E., Lavialle, F., Aubourg, A. and Ollivier-Bousquet M. (1997). Brefeldin A dif- ferently affects basal and prolactin-stimulated milk protein secretion in lactating rabbit mammary epithe- lial cells. Eur. J. Cell Biol. 72, 324–336. Pauloin, A., Tooze, S.A., Michelutti, I., Delpal, S. and Ollivier-Bousquet, M. (1999). The majority of clathrin coated vesicles from lactating rabbit mammary gland arises from the secretory pathway. J. Cell Sci. 112(Pt 22), 4089–4100. Pechoux, C., Boisgard, R., Chanat, E. and Lavialle, F. (2005). Ca(2+)-independent phospholipase A2 partic-

458 J.-L. Vilotte et al. ipates in the vesicular transport of milk proteins. Rando, A., Di Gregorio, P., Ramunno, L., Mariani, P., Biochim. Biophys. Acta 1743, 317–329. Fiorella, A., Senese, C., Marletta, D. and Masina, P. Pelham, H.R. (1989). Control of protein exit from the (1998). Characterization of the CSNAG allele of the endoplasmic reticulum. Ann. Rev. Cell Biol. 5, 1–23. bovine alpha-s1-casein locus by the insertion of a rel- Pena, R.N., Folch, J.M., Sanchez, A. and Whitelaw, ict of a long interspersed element. J. Dairy Sci. 81, C.B.A. (1998). Chromatin structures of goat and sheep 1735–1742. beta-lactoglobulin gene differ. Biochem. Biophys. Res. Commun. 252, 649–653. Rando, A., Pappalardo, M., Capuano, M., Di Gregorio, P. Pena, R.N., Sanchez, A., Coll, A. and Folch, J.M. (1999). and Ramunno, L. (1996). Two mutations might be Isolation, sequencing and relative quantification by responsible for the absence of beta-casein in goat milk. fluorescent-ratio PCR of feline beta-lactoglobulin I, II Anim. Genet. 27, 31. and III cDNAs. Mamm. Genome 10, 560–564. Perez, M.J., Leroux, C., Bonastre, A.S. and Martin, P. Raught, B., Liao, W.S.L. and Rosen, J.M. (1995). (1994). Occurrence of a LINE sequence in the 3¢ UTR Developmentally and hormonally regulated CCAAT/ of the goat alpha-s1-casein E-encoding allele associ- enhancer-binding protein isoforms influence beta-casein ated with reduced protein synthesis level. Gene 147, gene expression. Mol. Endocrinol. 9, 1223–1232. 179–187. Persuy, M.A., Printz, C., Medrano, J.F. and Mercier, J.C. Reinhardt, T.A. and Lippolis, J.D. (2008). Developmental (1996). One mutation might be responsible for the changes in the milk fat globule membrane proteome absence of beta-casein in two breeds of goats. Anim. during the transition from colostrum to milk. J. Dairy Genet. 27, 96–102. Sci. 91, 2307–2318. Persuy, M.A., Stinnakre, M.G., Printz, C., Mahé, M.F. and Mercier, J.C. (1992). High expression of the caprine Rhoads, R.E. and Grudzien-Nogalska, E. (2007). beta-casein gene in transgenic mice. Eur. J. Biochem. Translational regulation of milk protein synthesis at 205, 887–891 secretory activation. J. Mammary Gland Biol. Phi-Van, L. and Strätling, W.H. (1988). The matrix attach- Neoplasia 12, 283–292. ment regions of the chicken lysozyme gene co-map with the boundaries of the chromatin domain. EMBO Riebeling, C., Morris, A.J. and Shields, D. (2009). J. 7, 655–664. Phospholipase D in the Golgi apparatus. Biochim. Pierre, S., Jolivet, G., Devinoy, E. and Houdebine, L.M. Biophys. Acta 1791, 876–880. (1994). A combination of distal and proximal regions is required for efficient prolactin regulation of trans- Rijnkels, M. (2002). Multispecies comparison of the fected rabbit alpha-s1-casein chloramphenicol acetyl- casein gene loci and evolution of casein gene family. transferase constructs. Mol. Endocrinol. 8, J. Mammary Gland Biol. Neoplasia 7, 327–345. 1720–1730. Pines, G., Danbolt, N.C., Bjoras, M., Zhang, Y., Bendahan, Rijnkels, M., Kabotyanski, E., Montazer-Torbati, M.B., A., Eide, L., Koepsell, H., Storm-Mathisen, J., Seeberg, Beauvais, C.H., Vassetzky, Y., Rosen, J.M. and E. and Kanner, B.L. (1992). Cloning and expression of a Devinoy, E. (2010). The epigenetic landscape of mam- rat brain L-glutamate transporter. Nature 360, 464–467. mary gland development and functional differentia- Pitelka, D.R. and Hamamoto, S.T. (1983). Ultrastructure tion. J. Mammary Gland Biol. Neoplasia 15, 85–100. of the mammary secretory cell, in, Biochemistry of Lactation, T.B. Mepham, ed., Elsevier Science Rijnkels, M., Kooiman, P.M., de Boer, H.A. and Pieper, Publishers B.V., pp. 29–70. Amsterdam, New York. F.R. (1997b). Organization of the bovine casein gene Prasad, R., Hudson, B.G., Butkowski, R., Hamilton, J.W. locus. Mamm. Genome 8, 148–152. and Ebner, K.E. (1979). Resolution of the charge forms and amino acid sequence and location of a tryp- Rijnkels, M., Meershoek, E., de Boer, H.A. and Pieper, F.R. tic glycopeptide in rat alpha-lactalbumin. J. Biol. (1997c). Physical map and localization of the human Chem. 254, 10607–10614. casein gene locus. Mamm. Genome 8, 285–286. Provot, C., Persuy, M.A. and Mercier, J.C. (1995). Complete sequence of the ovine beta-casein-encoding gene and Rijnkels, M., Wheeler, D.A., de Boer, H.A. and Pieper, interspecies comparison. Gene 154, 259–263. F.R. (1997a). Structure and expression of the mouse Qasba, P.K. and Safaya, S.K. (1984). Similarity of the casein gene locus. Mamm. Genome 8, 9–15. nucleotide sequences of rat alpha-lactalbumin and chicken lysozyme genes. Nature 608, 377-380 Rival-Gervier, S., Thepot, D., Jolivet, G. and Houdebine, Qasba, P.K., Hewlett, I.K. and Byers, S. (1983). The pres- L.M. (2003). Pig whey acidic protein gene is sur- ence of the milk protein alpha-lactalbumin and its rounded by two ubiquitously expressed genes. mRNA in the rat epididymis. Biochem. Biophys. Res. Biochim. Biophys. Acta 1627, 7–14. Commun. 30, 306–312. Roberts, B., DiTullio, P., Vitale, J., Hehir, K. and Gordon, K. (1992). Cloning of the goat beta-casein-encoding gene and expression in transgenic mice. Gene 121, 255–262. Robinson, C. and Kolb, A. F. (2009). Analysis of mam- mary specific gene locus regulation in differentiated cells derived by somatic cell fusion. Exp. Cell Res. 315, 508–22. Robinson, G.W., Johnson, P.F., Hennighausen, L. and Sterneck, E. (1998). The C/EBPbeta transcription factor regulates epithelial cell proliferation and differentiation in the mammary gland. Genes Dev. 12, 1907–1916.

14 Genetics and Biosynthesis of Milk Proteins 459 Robinson, G.W., McKnight, R.A., Smith, G.H. and Sdassi, N., Silveri, L., Laubier, J., Tilly, G., Costa, J., Hennighausen, L. (1995). Mammary epithelial cells Layani, S., Vilotte, J.L. and Le Provost, F. (2009). undergo secretory differentiation in cycling virgins but Identification and characterization of new miRNA require pregnancy for the establishment of terminal cloned from normal mouse mammary gland. BMC differentiation. Development 121, 2079–2090. Genomics 10, 149. Rohrer, G.A., Alexander, L.J. and Beattie, C.W. (1997). Seagroves, T.N., Krnacik, S., Raught, B., Gay, J., Burgess- Mapping genes located on human chromosomes 2 and Beusse, B., Darlington, G.J. and Rosen, J.M. (1998). 12 to porcine chromosomes 15 and 5. Anim. Genet. 28, C/EBPbeta, but not C/EBPalpha is essential for ductal 448–450. morphogenesis, lobuloalveolar proliferation, and func- tional differentiation in the mouse mammary gland. Rosen, J.M. (1987). Milk protein gene structure and Gene Devp. 12, 1917–1928 expression, in, The Mammary Gland, Neville, M.C. and Daniel, C.W., eds., Plenum Publishing Corporation, Seddiki, T. and Ollivier-Bousquet, M. (1991). Temperature New York. pp. 301–322 dependence of prolactin endocytosis and casein exo- cytosis in epithelial mammary cells. Eur. J. Cell Biol. Rosen, J.M., Li, S., Raught, B. and Hadsell, D. (1996). 55, 60–70. The mammary gland as a bioreactor: factors regulating the efficient expression of milk protein-based trans- Sharp, J.A., Lefèvre, C. and Nicholas, K.R. (2008). Lack genes. Am. J. Clin. Nutr. 63, 627–632. of functional alpha-lactalbumin prevents involution in Cape fur seals and identifies the protein as an apop- Rosen, J.M., Mediona, D., Schlein, A.R., Eisenstein, R.S. totic milk factor in mammary gland involution. BMC and Yu-Lee, L.Y. (1988). Hormonal and cell-substratum Biol. 6, 48–57. regulation of casein gene expression at the posttran- scriptional level, in, Steroid Hormone Action, G. Ringold Shekar, P.C., Goel, S., Rani, S.D.S., Sarathi, D.P., Alex, (ed.) Alan R. Liss. Inc. New York. pp 269–278 J.L., Singh, S. and Kumar, S. (2006). kappa-Casein- deficient mice fail to lactate. Proc. Natl. Acad. Sci. Rosen, J.M., Rodgers, J.R., Couch, C.H., Bisbee, C.A., U.S.A 103, 8000–8005. David-Inouye, Y., Campbell, S.M. and Yu-Lee, L.Y. (1986). Multihormonal regulation of milk protein gene Shennan, D.B. (1995). Identification of a high affinity tau- expression. Ann. N. Y. Acad. Sci. 478, 63–76 rine transporter which is not dependent on chloride. Biosci. Rep. 15, 231–239. Rosen, J.M., Zahnow, C., Kazansky, A. and Raught, B. (1998). Composite response elements mediate hor- Shennan, D.B. and McNeillie, S.A. (1994a). Characteristics monal and developmental regulation of milk protein of a-aminoisobutyric acid transport by the lactating gene expression. Biochem. Soc. Symp. 63, 101–113. rat mammary gland. J. Dairy Res. 61, 9–19. Rudolph, M.C., McManaman, J. L., Phang, T., Russell, T., Shennan, D.B. and McNeillie, S.A. (1994b). High affinity Kominsky, D.J., Serkova, N.J., Stein, T., Anderson, (Na+ + Cl-)-dependent transport by lactating mammary S.M. and Neville, M.C. (2007). Metabolic regulation tissue. J. Dairy Res. 61, 335–343. in the lactating mammary gland: a lipid synthesizing machine. Physiol. Genomics 28, 323–336. Shennan, D.B. and McNeillie, S.A. (1994c). Milk accu- mulation down regulates amino acid uptake via sys- Rothman, J.E. and Wieland, F.T. (1996). Protein sorting tems A and L by lactating mammary tissue. Horm. by transport vesicles. Science 272, 227–234. Metab. Res. 26, 611. Saidi, S., Rival-Gervier, S., Daniel-Carlier, N., Thepot, Shennan, D.B. and Peaker, M. (2000).Transport of milk D., Morgenthaler, C., Viglietta, C., Prince, S., Passet, constituents by the mammary gland. Physiol. Rev. 80, B., Houdebine, L.M. and Jolivet, G. (2007). Distal 925–951. control of the pig whey acidic protein (WAP) locus in transgenic mice. Gene 401, 97–107. Shennan, D.B., Backwell, F.R.C. and Calvert, D.T. (1999). Metabolism of aminoacyl-p-nitroanilides by Schatz, G., Dobberstein, B. (1996). Common principles of rat mammary tissue. Biochim. Biophys. Acta 1427, protein translocation across membranes. Science 271, 227–235. 1519–1526. Shennan, D.B., Calvert, D.T., Backwell, F.R.C. and Boyd, Schmidhausser, C. Casperson, G.F., Myers, C.A., Sanzo, C.A.R. (1998). Peptide aminonitrogen transport by the K.T., Bolten, S. and Bissell, M.J. (1992). A novel tran- lactating rat mammary gland. Biochim. Biophys. Acta scriptional enhancer is involved in the prolactin- and 1373, 252–260. extracellular matrix-dependent regulation of beta- casein gene expression. Mol. Biol. Cell 3, 699–709. Shennan, D.B., Calvert, D.T., Travers, M.T., Kudo, Y. and Boyd, C.A.R. (2002). A study of L-leucine, Schmidt, J.A., Kalkofen, D.N., Donovan, K.W. and L-phenylalanine and L-alanine transport in the perfused Brown, W.J. (2010). A role for phospholipase A2 rat mammary gland: possible involvement of LAT1 and activity in membrane tubule formation and TGN LAT2. Biochim. Biophys. Acta 1564, 133–139. trafficking. Traffic 11, 1530–1536. Shennan, D.B., Cliff, M.J. and Hawkins, P. (1996). Schmitt-Ney, M., Doppler, W., Ball, R.K. and Groner, B. Volume-sensitive taurine efflux from mammary tissue (1991). Beta-casein gene promoter activity is regu- is not obliged to utilize volume-activated anion chan- lated by the hormone-mediated relief of transcriptional nels. Biosci. Rep. 16, 459–465. repression and a mammary-gland-specific nuclear fac- tor. Mol. Cell. Biol. 11, 3745–3755. Shennan, D.B., McNeillie, S.A. and Curran, D.E. (1994). The effect of a hyposmotic shock on amino acid efflux

460 J.-L. Vilotte et al. from lactating rat mammary tissue : stimulation of tau- Tang, Y. (1993). No alpha-lactalbumin-like activity rine and glycine efflux via a pathway distinct from detected in a low molecular mass protein fraction of rat anion exchange and volume-activated anion channels. epididymal extract. Reprod. Fertil. Dev. 5, 229–237. Exp. Physiol. 79, 797–808. Shennan, D.B., McNeillie, S.A., Jamieson, E.A. and Thépot, D., Fontaine, M.L., Houdebine, L.M. and Calvert, D.T. (1994). Lysine transport in lactating rat Devinoy, E. (1990). Complete sequence of the rabbit mammary tissue : evidence for an interaction between WAP gene. Nucleic Acids Res. 18, 3641. cationic and neutral amino acids. Acta Physiol. Scand. 151, 461–466. Thépot, D., Fontaine, M.L., Houdebine, L.M. and Shennan, D.B., Millar, I.D. and Calvert, D.T. (1997). Devinoy, E. (1991). Structure of the gene encoding Mammary-tissue amino acid transport systems. Proc. rabbit beta-casein. Gene 97, 301–306 Nutr. Soc. 56, 177–191. Sollner, T., Whiteheart, S.W., Brunner, M., Erdjument- Threadgill, D.W. and Womack, J.E. (1990). Genomic Bromage, H., Geromanos, S., Tempst, P. and Rothman, analysis of the major bovine casein genes. Nucleic J.E. (1993). SNAP receptors implicated in vesicle tar- Acids Res. 18, 6935–6942 geting and fusion. Nature 362, 318–324. Soulier, S., Lepourry, L., Stinnakre, M.G., Langley, B., Tomic S, Chughtai N. and Ali S. (1999). SOCS-1, -2, -3: L’Huillier, P.J., Paly, J., Djiane, J., Mercier, J.C. and selective targets and functions downstream of the pro- Vilotte, J.L. (1999). Introduction of a proximal Stat5 lactin receptor. Mol. Cell. Endocrinol. 158, 45–54. site in the murine alpha-lactalbumin promoter induces prolactin dependency in vitro and improves expression Tomlinson, A.M., Cox, R.D., Lehrach, H.R. and frequency in vivo. Transgenic Res. 8, 23–31. Dalrymple, M.A. (1996). Restriction map of two yeast Soulier, S., Mercier, J.C., Vilotte, J.L., Anderson, J., Clark, artificial chromosomes spanning the murine casein J. and Provot, C. (1989). Characterization of genomic locus. Mamm. Genome 7, 342–344. clones homologous to the alpha-lactalbumin gene in the bovine and ovine species. Gene 83, 331–338. Tooze, S.A. (1998). Biogenesis of secretory granules in Starr, R. and Hilton, D.J. (1999). Negative regulation of the trans-Golgi network of neuroendocrine and endo- the JAK/STAT pathway. Bioessays 21, 47–52. crine cells. Biochim. Biophys. Acta 1404, 231–244. Stinnakre, M.G., Soulier, S., Schibler, L., Lepourry, L., Mercier, J.C. and Vilotte, J.L. (1999). Position- Topcic, D., Auguste A., De Leo, A.A., Lefevre, C., Digby, independent and copy-number-related expression of a M.R. and Nicholas, K.R. (2009). Characterization of goat bacterial artificial chromosome alpha-lactalbu- the tammar wallaby (Macropus eugenii) whey acidic min gene in transgenic mice. Biochem. J. 339, protein gene; new insights into the function of the pro- 33–36. tein. Evol. Dev. 11, 363–375. Stinnakre, M.G., Vilotte, J.L., Soulier, S. and Mercier, J.C. (1994). Creation and phenotypic analysis of Triplett, A. A., Sakamoto, K., Matulka, L. A., Shen, L., alpha-lactalbumin-deficient mice. Proc. Natl. Acad. Smith, G. H. and Wagner, K. U. (2005). Expression of Sci. U.S.A 91, 6544–6548. the whey acidic protein (WAP) is necessary for ade- Stöcklin, E., Wissler, M., Gouilleux, F. and Groner, B. quate nourishment of the offspring but not functional (1996). Functional interactions between Stat5 and the differentiation of mammary epithelial cells. Genesis glucocorticoid receptor. Nature 383, 726–728. 43, 1–11. Stoecklin, E., Wissler, M., Morriggl, R. and Groner, B. (1997). Specific DNA binding of Stat5, but not of glu- Turner, M.D., Handel, S.E., Wilde, C.J. and Burgoyne, cocorticoid receptor, is required for their functional R.D. (1993). Differential effect of brefeldin A on phos- cooperation in the regulation of gene transcription. phorylation of the caseins in lactating mouse mam- Mol. Cell. Biol. 17, 6708–6716. mary epithelial cells. J. Cell Sci. 106, 1221–1226. Storck, T., Schulte, S., Hofmann, K. and Stoffel, W. (1992). Structure, expression and functional analysis Ucar, A., Vafaizadeh, V., Jarry, H., Fiedler, J., Klemmt, of a Na+-dependent glutamate aspartate transporter P.A., Thum, T., Groner, B. and Chowdhury, K. (2010). from rat brain. Proc. Natl. Acad. Sci. U.S.A. 89, miR-212 and miR-132 are required for epithelial 10959–10965. stromal interactions necessary for mouse mammary Streuli, C.H., Edwards, G.M., Delcommenne, M., Whitelaw, gland development. Nat. Genet. 42, 1101–1108. C.B.A., Burdon, T.G., Schindler, C. and Watson, C.J. (1995). Stat5 as a target for regulation by extracellular Udy, G.B., Towers, R.P., Snell, R.G., Wilkins, R.J., Park, matrix. J. Biol. Chem. 270, 21639–21644. S.H., Ram, P.A., Waxman, D.J. and Davey, H.W. Tanaka, T., Haneda, S., Imakawa, K., Sakai., S. and (1997). Requirement of STAT5b for sexual dimor- Nagaoka., K. (2009). A microRNA, miR-101a, controls phism of body growth rates and liver gene expression. mammary gland development by regulating cyclooxy- Proc. Natl. Acad. Sci. U.S.A. 94, 7239–7244. genase-2 expression. Differentiation 77, 181–187. Uversky, V.N., Gillespie, J.R. and Fink, A.L. (2000). Why are “natively unfolded” proteins unstructured under physiologic conditions? Proteins 41, 415–427. Valentine, C.R. (1998). The association of nonsense codons with exon skipping. Mutat. Res. 411, 87–117. Verma, N. and Kansal V.K. (1995). Characterisation and starvation induced regulation of methionine uptake sites in mouse mammary gland. Indian J. Exp. Biol. 33, 516–530. Verma, N. and Kansal, V.K. (1993). Characterisation of the routes of methionine transport in mouse mammary glands. Indian J. Med. Res. [B] 98, 297–304.

14 Genetics and Biosynthesis of Milk Proteins 461 Vilotte J.L., Whitelaw, C.B.A., Ollivier-Bousquet, M. and a milk protein gene promoter in vitro: identification of Shennan, D.B. (2002). Biosynthesis of milk proteins, a mammary gland-specific factor. Nucleic Acids Res. in, Advanced Dairy Chemistry 1, Proteins, 3rd edn., 19, 6603–6610. P.F. Fox and P.L.H. McSweeney, eds., Kluwer Webster, J., Wallace, R.M., Clark, A.J. and Whitelaw, Academic/Plenum Publishers. New York. pp.699–738. C.B.A. (1995). Tissue-specific, temporally regulated expression mediated by the proximal ovine beta-lacto- Vilotte, J.L. and Soulier, S. (1992). Isolation and charac- globulin promoter in transgenic mice. Cell. Mol. Biol. terization of the mouse alpha-lactalbumin-encoding Res. 41, 11–15. gene: interspecies comparison, tissue- and stage- West, D.W. and Clegg, R.A. (1984). Casein kinase activ- specific expression. Gene 119, 287–292. ity in rat mammary gland Golgi vesicles. Demonstration of latency and requirement for a transmembrane ATP Vilotte, J.L., Soulier, S. and Mercier, J.C. (1993). Complete carrier. Biochem. J. 219, 181–187. sequence of a bovine alpha-lactalbumin pseudogene: the Whitelaw C.B.A. (2000). Nucleosome organization of the region homologous to the gene is flanked by two directly beta-lactoglobulin gene. Transcription complex for- repeated LINE sequences. Genomics 16, 529–532. mation. Adv. Exp. Med. Biol. 480, 147–153 Whitelaw, C.B.A. and Webster, J. (1998). Temporal profiles Vilotte, J.L., Soulier, S., Mercier, J.C., Gaye, P., Hue- of appearance of DNAse I hypersensitive sites associated Delahaie, D. and Furet, J.P. (1987). Complete with the ovine beta-lactoglobulin gene differ in sheep nucleotide sequence of bovine alpha-lactalbumin and transgenic mice. Mol. Gen. Genet. 257, 649–654. gene. Comparison with its rat counterpart. Whitelaw, C.B.A., Harris, S., McClenaghan, M., Simons, Biochimie 69, 609–620 J.P. and Clark, A.J. (1992). Position-independent expression of ovine beta-lactoglobulin in transgenic Vilotte, J.L., Soulier, S., Printz, C. and Mercier, J.C. mice. Biochem. J. 286, 31–39. (1991). Sequence of the goat alpha-lactalbumin- Winklehner-Jennewein, P., Geymayer, S., Lechner, J., encoding gene: comparison with the bovine gene and Welte, T., Hansson, L., Geley, S. and Doppler, W. evidence of related sequences in the goat genome. (1998). A distal enhancer region in the human beta- Gene 98, 271–276 casein gene mediates the response to prolactin and glucocortocoid hormones. Gene 217, 127–139. Vina, J., Puertes, I.R., Saez, G.T. and Vina, J.R. (1981). Witsell, D.L., Casey, C.E. and Neville, M.C. (1990). Role of prolactin in amino acid uptake by the lactat- Divalent cation activation of galactosyl transferase in ing mammary gland of the rats. FEBS Lett. 126, native mammary Golgi vesicles. J. Biol. Chem. 265, 250–252. 15731–15737. Wolberger, C. (1998). Combinatorial transcription factors. Wang, C. and Li, Q. (2007). Identification of differen- Curr. Opin. Genet. Dev. 8, 552–559. tially expressed miRNA during the development of Wooding, F.B.P. (1977). Comparative mammary fine Chinese murine mammary gland. J. Genet. Genomics structure. Symp. Zool. Soc. Lond. 41, 1–41. 34, 966–973. Yoshimura, M. and Oka, T. (1989). Isolation and struc- tural analysis of the mouse beta-casein gene. Gene 78, Wang, C.C., Shi, H., Guo, K., Ng, C.P., Li, J., Gan, B.Q., 267–275 Chien Liew, H., Leinonen, J., Rajaniemi, H., Zhou, Yu-Lee, L.Y., Richter-Mann, L., Couch, C.H., Stewart, Z.H., Zeng, Q. and Hong, W. (2007). VAMP8/endo- A.F., MacKinlay, R.G. and Rosen, J.M. (1986). brevin as a general vesicular SNARE for regulated Evolution of the casein multigene family conserved exocytosis of the exocrine system. Mol. Biol. Cell 18, sequences in the 5¢ flanking and exon regions. Nucleic 1056–1063 Acids Res. 14, 1883–1902Martin, P. and Leroux, C. (1992) Exon-skipping is responsible for the 9 amino Wang S, Webb KE Jr, Akers MR. (1996) Peptide-bound acid residue deletion occurring near the N-terminal of methionine can be a source of methionine for the syn- human b-casein. Biochem. Biophys. Res. Commun. thesis of secreted proteins by mammary tissue explants 183, 750–757. from lactating mice. J. Nutr. 126, 1662–1672. Warner, B., Janssens, P. and Nicholas, K. (1993). Prolactin- independent induction of alpha-lactalbumin gene expression in mammary gland explants from pregnant Balb/C mice. Biophys. Biochem. Res. Commun. 194, 987–991. Watson, C.J., Gordon, K.E., Robertson, M. and Clark, A.J. (1991). Interaction of DNA-binding proteins with

Genetic Polymorphism 15 of Milk Proteins P. Martin, L. Bianchi, C. Cebo, and G. Miranda 15.1 Introduction of this account will be on advances made since the previous chapter in this series (Ng-Kwai-Hang Since the discovery, over half a century ago, of and Grosclaude, 2003) on the molecular bases for two electrophoretically distinct forms of b-lacto- genetic polymorphism at the genome level. globulin by Aschaffenburg and Drewry (1957), genetic polymorphism of milk proteins has been With the advent of molecular biology and extensively investigated in cattle. For obvious DNA analysis, search for polymorphism progres- economic reasons most of the efforts have been sively moved from qualitative to quantitative focused on the impact on milk processing (par- aspects (from protein to genome), and with the ticularly cheesemaking properties). However, availability of next-generation sequencing (NGS) genetic polymorphisms of milk proteins have technologies that have revolutionised genomics been also used to analyse, through association and genetics, this trend has been accentuated to studies, possible relationships with quantitative focus on a large number of outstanding issues and qualitative milk traits as well as evolutionary that previously could not be addressed effectively. and biodiversity issues. Most of these aspects Thus, we are now able to study genetic variation have been addressed in depth in the previous ver- on a genome-wide scale and characterise gene sions of the chapter in this series dedicated to regulatory processes at unprecedented resolution genetic polymorphism of milk proteins by (Gilad et al., 2009). Ng-Kwai-Hang and Grosclaude (1992, 2003). One of the main goals of genomics is to Since Caroli et al. (2009) recently reviewed determine the genetic differences among indi- milk protein polymorphisms in cattle, focusing viduals and to understand their relationships to mainly on the effect on animal breeding and the phenotypic differences within species. human nutrition, species other than bovine will Structural variations within the genome have be particularly considered here and the main focus been described in a number of species. These variations consist in single-nucleotide polymor- P. Martin • L. Bianchi • C. Cebo • G. Miranda phisms (SNPs) and structural variations (SVs) Institut National de la Recherche Agronomique, including short insertions/deletions (indels) and UMR1313, Génétique animale et Biologie intégrative other more complex ones such as duplications (GABI), Équipe “Lait, Génome & Santé”, Domaine de and translocations. SNPs and SVs have been Vilvert-Bâtiment 221,78350 Jouy-en-Josas, France shown to account for ca. 83% and 17%, respec- e-mail: [email protected]; leonardo. tively, of the total detected genetic variation in [email protected]; [email protected]; gene expression (Stranger et al., 2007). In [email protected] humans, SVs have been associated with com- plex human traits, such as autism, schizophrenia, P.L.H. McSweeney and P.F. Fox (eds.), Advanced Dairy Chemistry: Volume 1A: Proteins: Basic Aspects, 463 4th Edition, DOI 10.1007/978-1-4614-4714-6_15, © Springer Science+Business Media New York 2013

464 P. Martin et al. Crohn’s disease and susceptibility to HIV infec- spectrometry, which has increasing resolution and tion (Zhang et al., 2009). For most other species, is fast and efficient, are beginning to supplant IEF. including the major farm animals, the extent and biological consequences of SVs remain rather Our first purpose here is to introduce recent poorly documented (Kerstens et al., 2011). developments in mass spectrometry coupled with Because of the efficiency of genotyping methods separation techniques to analyse milk protein and the central role they play in the genome- polymorphisms which are the most powerful wide association studies, SNPs are currently the methods available nowadays to detect and analyse best known and useful genetic variations. genetic polymorphisms at the DNA level. However, SVs which have been much less stud- ied due to the lack of a cost-effective approach 15.2.1 Recent Developments for genotyping start to be considered and their in Phenotyping Methods genetic significance recognised (Zhang et al., 2011), but the effect that SVs have on gene The profiling of milk proteins is complicated by expression is likely underestimated given the the existence of genetic polymorphisms, alter- much less completeness and accuracy with native splicing variants and post-translational which SVs could be queried. modifications (PTMs). As a result, several forms of a same protein family are usually present Most of the studies reporting genetic poly- together in a given milk sample, rendering the morphisms of milk proteins are related to the analysis of the lacto-proteome a particularly com- colloidal and soluble fraction of milk, namely, plex matter. Over the years, several electropho- caseins and whey proteins. However, growing retic and chromatographic techniques have been attention is paid to milk-fat globule membrane applied (reviewed in detail by Ng-Kwai-Hang and (MFGM)-associated proteins. Although these Grosclaude, 2003) to the routine screening of proteins account for only 1–2% of total milk pro- individual milk samples. Electrophoresis and, to a teins, evidence that MFGM proteins possess lesser extent, chromatography have made it pos- techno-functional and nutritional properties is sible to detect the vast majority of the currently accumulating (Dewettinck et al., 2008). Since known milk protein variants. Over the last 15 data on the genetic polymorphism of these pro- years, however, techniques based on the mass teins is becoming more consistent, a part of this spectrometry analysis of proteins and peptides chapter will be devoted to this issue. have started to be applied to milk proteins. Mass spectrometry analyses offer the undoubted advan- 15.2 Methods of Detecting Genetic tage of detecting minimal differences in terms of Polymorphism molecular mass, even among residues having similar chemical properties. Thus, the possibility Methods of detecting genetic polymorphism have to detect new polymorphisms is greatly increased, been discussed in detail in the previous version of in that theoretically all amino acid modifications this chapter. Briefly, genetic polymorphisms can be imply a molecular mass shift in the mature pro- detected at the phenotypic level using classical tein, with the sole possible exception of isobaric techniques to analyse proteins such as electropho- amino acids, such as leucine/isoleucine (that resis, isoelectric focusing (IEF) or chromatography. remain undetectable using the current techniques) These techniques detect only genetic variations or lysine/glutamine. resulting in differences on the net charge, hydro- phobicity or molecular weight of the protein. For Remarkable progress has been made in top- the phenotyping of breeds and large populations, down (TD) MS that has gained a remarkable IEF is still the most effective and convenient space in proteomics, rapidly trespassing the clas- method since it gives simultaneously an overview sical bottom-up approaches and surpassing the of the phenotype expression of the six main milk limit between a promising approach and a solid protein genes. However, techniques using mass established technique (Armirotti and Damonte, 2010). Using the so-called “top-down” process,

15 Genetic Polymorphism of Milk Proteins 465 in which proteins are analysed in the gas phase as2-casein (Picariello et al., 2009a) variants as intact molecules, it is possible to derive struc- (see below). tural information on proteins with the level of accuracy that is impossible to achieve by classi- Mass spectrometric techniques have therefore cal bottom-up approaches. Complete maps of proved to be useful tools for the screening of milk PTMs and assessment of single amino acid poly- protein polymorphisms. On this basis, the more morphisms are only a few of the results that can recent trends aim at developing quantitative or be obtained with this technique. Despite some semi-quantitative tools to assess the presence of existing technical and economic limitations, TD given casein variants in bulk milk from the same analysis is at present the most powerful instru- species (Picariello et al., 2009b) or the fraudulent ment for MS-based proteomics. Such TD presence of milk from other species, as adultera- approaches have been used to analyse phospho- tion of milk for the dairy industry (Nicolaou proteins, including caseins (Wu et al., 2009), et al., 2011; Cuollo et al., 2010). and O-glycosylation of MUC-1 and k-casein (Hanisch, 2011). 15.2.2 Methods of Detecting Genetic Polymorphisms at the DNA Level Beforehand, however, classical mass spectro- metric approaches had been first applied in the Analyses of milk protein genes at the nucleotide detection of a milk protein variant by Visser et al. level started 25 years ago with cDNA cloning and (1995), and, then by Dong and Ng-Kwai-Hang Sanger’s method for DNA sequencing, providing (1998) and Senocq et al. (2002) who character- us with the most basic information of all: the ised variants G (P152L), F (P137L) and H (M93L) sequence of nucleotides. Few years later, with the of bovine b-casein, respectively. In particular, the advent of the polymerase chain reaction (PCR) peptides carrying the mutation in variants G and technique (Saiki et al., 1988), amplification of H were sequenced by tandem mass spectrometry specific gene regions (including exons) at the (MS/MS). Ferranti et al. (2001) and Pierre et al. genomic level has facilitated the analysis of (2001) proposed the use of ESI-MS for the polymorphisms within the coding sequences and routine screening of ovine and caprine caseomes, provided tools to genotype animals, including respectively. They were able to detect genetic males and females prior to lactation. variants, together with their non-allelic forms, of as1- and as2-caseins, and different phosphoryla- Capillary sequencing is no longer the tech- tion levels. Neveu et al. (2002) reported the exis- nology of choice for most ultra-high-throughput tence of a new variant of caprine b-casein by applications. We are now witnessing a genomic LC-MS, and the phosphorylation patterns of the revolution due to the continued advancements in protein were characterised by the combined use the next-generation sequencing technologies of peptide mass fingerprinting and sequencing by which assist solving complex biological prob- tandem mass spectrometry. The existence of a lems. Genome sequences are now available for a truncated form of caprine b-casein associated number of domestic species, including mam- with the O¢ allele and its amino acid sequence mals, and with them high-throughput tools was proposed by Cunsolo et al. (2005). The same including high-density single-nucleotide poly- research group described two truncated forms morphism (SNP) panels. As a result, domestic (204 instead of 207 AA) of caprine as2-casein animal populations are becoming invaluable and speculated that these proteins could be the resources for studying the molecular architecture product of a differential splicing of pre-messenger of complex traits. Recent progress in the posi- RNA encoding as2-casein alleles A and E during tional identification of genes underlying com- their maturation process (Cunsolo et al., 2006). plex traits in domestic animals has been reviewed LC-MS/MS and MALDI-PSD-TOF-MS were by Georges (2007), and the importance of com- applied, respectively, for the characterisation parative genomics for dissecting the genetic of ovine as1- (Chianese et al., 1997a, b) and basis of phenotypic variation has been stressed

466 P. Martin et al. (Andersson and Georges, 2004). Obviously, this variants), or not. Several alleles can actually give massive sequencing at the nucleotide level high- rise to a same variant (silent alleles), namely, lights the occurrence of new genetic polymor- when a nucleotide substitution does not modify phism of which molecular bases are easily the coding message (synonymous mutation) or identified. This situation differs from our current when mutations occur in a portion of the gene concept of genetic polymorphism which moves excluded from the mature transcript during the from qualitative variability (protein variants) to a course of the splicing process (introns) or located notion integrating a quantitative variability within untranslated regions of the messenger or dimension. Thus, emphasis is placed on new even when the sequence of the signal peptide polymorphisms impacting not only milk protein which is removed during the secretion process to structure but also its expression. Polymorphisms produce the mature protein is modified. occurring in cis-regulatory elements (mainly within the 5¢-flanking region of transcription units Most protein variants originate from single- encoding milk proteins) have been reported, as nucleotide mutations in the coding sequence of well as insertion/deletion (indel) and SNP within the parent DNA, thus leading to amino acid sub- exon, intron and/or 3¢-untranslated sequences. stitutions (non-synonymous mutation). The Mutations responsible for the occurrence of resulting amino acid, in turn, can modify the premature stop codons have been shown to be physico-chemical properties (net charge, isoelec- associated both with a decrease in the level of tric point, phosphorylation or hydrophobicity) of the relevant transcripts and the existence of the protein. Substitution of a single nucleotide multiple forms of messengers due to alternative occurring in intron consensus splice sequences splicing (exon skipping, usage of cryptic splice can also alter the maturation process of messen- sites). Such a situation, well exemplified by the gers, and as a result, one or several exons can be gene encoding as1-casein in the goat, may have skipped in the mature mRNA. Deletion of a single dramatic biological consequences (secretion nucleotide in coding exon sequences usually pathway, casein micelle structure, fat content, causes the occurrence of a premature stop codon etc.) by modifying the message and accordingly interrupting the reading frame and promoting the primary structure of the protein as well as its nonsense-mediated mRNA decay. In turn, these expression (Martin et al., 2002). events result in the appearance of internally deleted (exon skipping) or virtually truncated Rather than to make an exhaustive review of forms of the mature protein. This is frequently the abundant literature existing in this field, we associated with a strong reduction or even the have chosen to focus on some demonstrative absence of protein synthesis due to degradation examples, such as b-lactoglobulin in cattle and of the artefactual transcripts (see below). Deleted as1- and b-caseins in goats, to show how muta- forms have been described, so far in most spe- tions responsible for polymorphisms at the cies, in the three calcium-sensitive caseins (as1-, genomic level can influence milk protein compo- as2- and b-caseins), whereas a truncated form has sition, both at the qualitative and quantitative been to date reported and characterised only for levels. Beforehand, we will update the state of the goat b-casein (Cunsolo et al., 2005). To our our current knowledge of the molecular bases for knowledge, the k-casein gene has never been genetic polymorphism of milk proteins. reported to contain exon-skipping events. This is likely due to functional constraints. The gene 15.3 Molecular Basis for Genetic consists of five exons, of which only three are Polymorphism of Milk Proteins coding (exons 2, 3 and 4), the major part of the protein being encoded by exon 4 and the signal A genetic polymorphism is due to the occurrence peptide by both exons 2 and 3. An exception to of different alleles at the same locus, which may the general rule described above is the goat as1- code for different polypeptide chains (protein casein variant M, detected in the Italian Montefalcone breed which was suggested to

15 Genetic Polymorphism of Milk Proteins 467 originate from an interallelic recombination event enhancer sequences and the interaction between between two phylogenetically distinct parent multiple activator proteins and inhibitor proteins. alleles (Bevilacqua et al., 2002). Polymorphisms present in cis-regulatory elements, mainly within the 5¢-flanking region of transcrip- In their study on the so-called, at the time, “as- tion units encoding milk proteins, have been found casein” fraction in goat milk, Richardson and (reviewed by Martin et al., 2002). Our current Creamer (1975) did not detect a fraction corre- understanding of regulatory polymorphisms of sponding to as1-casein. These conclusions were milk protein genes is growing. Mechanisms are later corrected by Boulanger et al. (1984) who complex and regulation of milk protein expression not only detected and sequenced the protein but is mostly controlled by the non-coding portion of were able to describe at least three variants, the genome, through a series of complex mecha- named A, B and C. They stated that the apparent nisms acting at several levels, including pre-mRNA lack of as1-casein in some samples was in fact splicing (as already mentioned) and export, mRNA due to a genetically controlled strongly reduced stability and translation (Bevilacqua et al., 2006). rate of expression. It was the first example of genetic polymorphism associated with a quanti- Unstable transcripts have sequences (predom- tative variability (quantitative polymorphism), inately, but not exclusively, in the 3¢-untranslated reported regarding milk proteins. regions) that are signals for rapid degradation. Insertions of repetitive sequences, such as relicts Goat as1-casein is a paradigm of this kind of of long interspersed elements (LINE), have been polymorphism. Indeed, alleles associated with at also described to influence mRNA stability least four levels of synthesis have been described (Jansà-Perez et al., 1994). In addition, in the very so far, with the actual concentration in milk being recent years, a new model of gene regulation has the arithmetic sum of the contribution of each emerged that involves control exerted by small allele in homozygous and heterozygous subjects non-coding RNAs. This small RNA-mediated (Grosclaude and Martin, 1997). Depending on control can be exerted either at the level of the their level of expression, alleles are therefore translatability of the messengers or their stability. referred to “strong” (3.6 g/L per allele), “inter- A nice example has been provided by Liao et al. mediate” (1.6 g/L), “weak” (0.6 g/L) and “null” (2010), who showed that lactoferrin gene expres- (non-detectable amounts) alleles. At least one sion and function are directly targeted by miR- “null” allele has been described in the goat spe- 214 in HC11 and MCF7 cells. In the lactoferrin cies for the three calcium-sensitive caseins mRNA 3¢-untranslated region of human, mouse, (Leroux et al., 1990; Mahé and Grosclaude, 1993; rat, pig, bovine, camel and goat species, there is a Persuy et al., 1999; Ramunno et al., 2001a, 2005; conserved region that perfectly matches the seed Cosenza et al., 2003; Ådnøy et al., 2003) but, region of miR-214. interestingly, not for k-casein, not surprisingly however given its stabilising function of casein 15.4 Genetic Polymorphism of Milk micelles in milk. Quantitative polymorphisms Proteins in Dairy Ruminants were also described for bovine as1- and k-caseins (Rando et al., 1998; Damiani et al., 2000) and The aim of this section is to present the current b-lactoglobulin (BLG) in various breeds knowledge in terms of genetic polymorphism for (reviewed by Braunschweig and Leeb, 2006), but the major milk proteins of dairy ruminants, namely, none was reported so far in the ovine species, cow, goats and sheep. It will deal with the four except a moderate reduction in as1-casein secre- caseins (as1, b, as2 and k) and the major whey pro- tion associated with variant H (Giambra et al., teins (BLG and a-lactalbumin), and only muta- 2010a, b). tions leading to a well-characterised qualitative or quantitative protein polymorphism will be consid- General mechanisms controlling gene expres- ered. The multiple silent alleles coding for a same sion act both at the transcriptional and post- protein variant will therefore not be described. transcriptional levels. Multiple specific factors exert control of transcription: the strength of promoter elements, the presence or absence of

468 P. Martin et al. Since the last edition of this book, genetic poly- ThrP) and 64–68 (SerP-Ile-SerP-SerP-SerP). The morphism of bovine milk proteins was the subject bovine protein, unlike the mouse as-casein, con- of two comprehensive reviews (Farrell et al., 2004; tains no cysteine residues. Nine genetic variants Caroli et al., 2009), in which the current nomencla- have been characterised to date and are presented ture and recent findings were given. There have in Table 15.1. Protein variants differ mainly by also been two reviews by Marletta et al. (2007) and single amino acid substitutions, with the excep- Amigo et al. (2000) dealing with goat caseins and tions of A (Grosclaude et al., 1972) and H (Mahé the ovine major milk proteins, respectively. Over et al., 1999), which are internally deleted forms the last decade, however, nucleic acid-based lacking, respectively, sequences encoded by exon approaches which represent the highest throughput 4 and exon 8. Variant G that shares the same and best overall methods for obtaining information primary structure as variant B is characterised by a at the genome level as well as proteomics-based reduction of the as1-casein content in milk, reaching approaches relying on mass spectrometry methods ca. 55% (Mariani et al., 1995; Rando et al., 1998). for the detection and characterisation of milk pro- Moreover, the single amino acid substitution of a tein polymorphisms have led to significant advances potentially phosphorylated serine to a leucine, at on the subject, mainly concerning small ruminants position 66 in variant F (Prinzenberg et al., 1998), and, to a lesser extent, cattle. results in the loss of a further potential phosphory- lation site at position 64. It was postulated that Where possible, the description of a given pro- variant I is caused by a non-synonymous nucle- tein variant was integrated in tables with amino otide substitution in exon 11 of the gene and that it acid modification, GenBank (http://www.ncbi. originated within Bos indicus and spread subse- nlm.nih.gov/genbank/) and UniProt (http://www. quently to Bos taurus (Lühken et al., 2009). uniprot.org/) accession numbers, as well as with relevant bibliographic references. Nevertheless, 15.4.1.2 Bovine b-Casein such information was not always available. Often, The sequence of bovine b-casein was first estab- variants were detected and confirmed by classical lished at the protein level by Ribadeau Dumas electrophoresis techniques, but the primary struc- et al. (1972). Carles et al. (1988), using a new ture and the allele sequence were not character- strategy for primary structure determination of ised. Likewise, databases or reviews reported proteins, reported a sequence which currently sequence conflicts or misinterpretations that the refers to variant A1. The A2 variant sequence was present chapter aims at resolving. determined by cDNA sequencing (Stewart et al., 1987). The protein is secreted as a 209-residue Lastly, according to the current nomenclature, peptide chain, containing no cysteine and with the variants ascribed to bovine, caprine and ovine six potential phosphorylation sites (5 serine and 1 milk do include sequences found, respectively, in threonine), of which four are clustered at posi- the genera Bos, Capra and Ovis and not only in tions 15–19 (SerP-Leu-SerP-SerP-SerP). domesticated cattle, goat and sheep. Compared to previous reviews by Farrell et al. 15.4.1 Bovine Milk Proteins (2004) and Caroli et al. (2009), the number of variants (12) remains unchanged, although some 15.4.1.1 Bovine as1-Casein modifications are presented (Table 15.2). The Bovine as1-casein was first sequenced at the pro- variant previously named H1 (Han et al., 2000) tein level by Mercier et al. (1971) and Grosclaude was excluded, since the mutation leading to the et al. (1973) and then at the cDNA level by Nagao substitution R25C (Arg/Cys at position 25, previ- et al. (1984) and Stewart et al. (1984), who were ous GenBank accessions AF104928 and thus able to establish the sequence of the signal AF104929) was not confirmed by re-sequencing peptide. The mature protein is a 199-residue poly- (present GeneBank accession AH007287). The peptide chain, carrying ten potential phosphoryla- variant previously named H2, described by tion sites (9 serine and 1 threonine), of which seven Senocq et al. (2002), is therefore renamed H. are clustered at positions 46–49 (SerP-Glu-SerP- More recently, Miranda et al. (submitted) have

Table 15.1 Changes in bovine as1-casein variants (alleles) 15 Genetic Polymorphism of Milk Proteins Variant/allele Position 51–58 59 64 66 84 192 GenBank accession SwissProt accession References A 14–26 53 Gln SerP SerP Glu Glu X59856 P02662 B DEL Grosclaude et al. (1972) U862370/371 Mercier et al. (1971) and Grosclaude Ala (Bos indicus) et al. (1973) EU908730 Grosclaude et al. (1972) C Gly (Bos taurus) Grosclaude et al. (1972) D ThrP Grosclaude et al. (1972) E Lys Gly Prinzenberg et al. (1998) F Ser Leu Rando et al. (1998) G Mahé et al. (1999) H DEL Lühken et al. (2009, Balteanu et al. I Asp Gly (2008, 2010) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column DEL: internal deletion of the corresponding sequence in the mature protein 469

Table 15.2 Changes in bovine b-casein variants (alleles) 470 P. Martin et al. Position GenBank SwissProt 36 37 52 67 72 93 106 122 138 152 ? (114–169) accession accession References Variant/allele 18 35 A1 His X14711 Peterson and Kopfler (1966) and Carles et al. (1988) A2 SerP SerP Glu Glu Phe Pro Gln Met His Ser Pro Pro Gln M16645 P02666 Peterson and Kopfler (1966), Yan and Wold (1984) and Stewart et al. (1987) A3 Gln NM_181008 Peterson and Kopfler (1966) B His Arg BC111172 Aschaffenburg (1961) C Ser Lys His Aschaffenburg (1961) D Lys Aschaffenburg et al. (1968) E Lys Voglino (1972) F His Leu Visser et al. (1995) G His Leua Dong and Ng-Kwai- Hang (1998) H Glu Leu Glub Senocq et al. (2002) I Leu AY366419 Jann et al. (2002) J Ser Miranda et al. (submitted) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column aOriginally identified as P137L by Dong and Ng-Kwai-Hang (1998), the amino acid modification has been assigned to position 138 according to the given reference sequence bThe authors located a Gln to Glu substitution in the 114–169 sequence of the mature protein but could not establish more precisely its position

15 Genetic Polymorphism of Milk Proteins 471 Table 15.3 Changes in bovine as2-casein variants (alleles) Position GenBank SwissProt Variant/allele 8 33 47 51–59 130 accession accession References A SerP Glu Ala Thr M16644 P02663 Stewart et al. (1987) and M94327 Groenen et al. (1993) B Phe Ibeagha-Awemu et al. (2007) C Gly Thr Ile Mahé and Grosclaude (1982) D DEL Bouniol et al. (1993a, b) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column DEL: internal deletion of the corresponding sequence in the mature protein detected and characterised a novel variant (named (Bouniol et al., 1993a, b), results in the loss of an J), both at the protein and at the genomic DNA entire cluster of phosphorylated serines. level, differing from A2 for a single amino acid substitution (phenylalanine to serine at position 15.4.1.4 Bovine k-Casein 52) which does not result in a new potential phos- The primary structure of bovine k-casein and of phorylation site. its signal peptide was predicted by cDNA sequenc- ing (Stewart et al., 1984). The mature protein has 15.4.1.3 Bovine as2-Casein 169 amino acid residues, with two cysteines and The primary structure of bovine as2-casein was six potential phosphorylation sites (two serine and obtained by amino acid sequencing (Brignon four threonine), and up to six glycosylation sites et al., 1977) and subsequently corrected by cDNA have been described (Pisano et al., 1994). The sequencing (Stewart et al., 1987). The gene variants so far described for k-casein are shown in sequence is also available (GeneBank accession Table 15.4. No modifications are proposed to the M94327). The mature protein is a polypeptide of review article by Caroli et al. (2009), whereas the 207 amino acids, displaying 17 potential phos- amino acid substitutions ascribed to variants F2 phorylation sites (12 on serine and five on threo- and G1 by Farrell et al. (2004) are inconsistent nine residues), nine of which are clustered within with those published by the original authors. three short segments of the molecule: 8SerP- Compared to the previous reports, two additional SerP-SerP10, 56SerP-SerP-SerP58 and 129SerP- protein variants were included, here named K and ThrP-Ser131. Several as2-casein isoforms are L, both predicted from sequencing of exon 4 in present in bovine milk, differing in the level of domesticated yak samples. Variant K (GenBank phosphorylation (10–13 phosphate groups/mole- accession number AF194989) carries the substi- cule). Bovine as2-casein contains two cysteine tution already observed in G2 (D148A) with two residues involved in intra-chain disulphide additional amino acid substitutions (P36L and bridges in monomers of as2-casein isolated from P130R). Variant L has the same D148A mutation, milk (Rasmussen et al., 1994). which is common to all the yak sequences so far, but carries in addition a unique insertion of four Four genetic variants have been described, amino acids, corresponding to the duplication of named A to D, as shown in Table 15.3, and no sequence 148–151 (ASPE) found in bovine major modifications have been reported since the allele B. Therefore, the predicted primary struc- review by Caroli et al. (2009), except the Ser/Phe ture of the mature protein contains 173 amino substitution observed at position 8 (S8F) in vari- acid residues (Prinzenberg et al., 2008). The ant B, resulting in the loss of a potential phospho- description of variants will not be discussed in rylation site (Ibeagha-Awemu et al., 2007). detail, but attention can be drawn on mutations Variant C differs at three positions from variant A involving R10 (variant F2) and R97 (variants C, D (Mahé and Grosclaude, 1982). Lastly, the inter- and G1) and P130, resulting either in the loss nal deletion of eight residues encoded by exon 8

472 P. Martin et al. Table 15.4 Changes in bovine k-casein variants (alleles) Variant/ Position GenBank SwissProt allele 10 36 97 104 130 135 136 148 148–151 153 155 accession accession References A Arg Pro Arg Ser Pro Thr Thr Asp – Ile Ser AY380228 P02668 Robitaille et al. (2005) B Ile Ala AY380229 Robitaille et al. (2005) B2 Ile Ala Thr M36641 Gorodetskii and Kaledin (1987) C His Ile Ala Miranda et al. (1993) D His AJ619772 Q705V4 Caroli et al. (2009) E Gly AF041482 Miranda et al. (1993) F1 Val Sulimova et al. (1992) F2 His AF123250 Prinzenberg et al. (1996) G1 Cys Ile AF123251 Prinzenberg et al. (1996) G2 Ala AJ849456 Q5ZET1 Sulimova et al. AJ841941 (1996) H Ile AF105260 Prinzenberg et al. (1999) I Ala AF121023 Prinzenberg et al. (1999) J Ile Ala Arg Mahé et al. (1999) K Leu Arg Ala AH009225 L Ala INS AY095311 Prinzenberg et al. (2008) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column INS: 4 amino acid long insertion, corresponding to a duplication of sequence 148–151 (Ala-Ser-Pro-Glu) of the bovine allele B although 147–150 (Glu-Ala-Ser-Pro) is also possible (position 10 and 97) or in the appearance (posi- et al., 2004). The sequence of bovine BLG cDNA tion 130) of a site of cleavage by pancreatic trypsin (variant A) was established by Alexander et al. in the digestive tract. Furthermore, mutations (1989) and translated into a 162-amino-acid poly- involving T135 (G1) and T136 (B, B2 and C) peptide containing five cysteines, of which four determine the loss of a potential site of glycosyla- are involved in intramolecular disulphide bonds. tion. Lastly, the S104A modification occurring in A list of the 13 currently known variants is given the I variant alters the -Ser-Phe-Met-Ala- chy- in Table 15.5. Variants A and B are common in mosin-sensitive site described by Visser et al. most cattle breeds (Farrell et al., 2004). BLG vari- (1976) and can thus affect the rennet-induced ants A and B differ by two amino acid substitu- clotting of milk. tions: Asp64Gly (D64G) and Val118Ala (V118A). Since the last edition of this book, only one novel 15.4.1.5 Bovine b-Lactoglobulin variant having amino acid substitutions compared BLG is the major whey protein in cows’ milk. to B has been reported and was already included Eleven genetic variants have been reported, with in the lists by Farrell et al. (2004) and Caroli et al. variants A and B being the most frequent (Farrell (2009). It was named W and the authors proposed

Table 15.5 Changes in bovine b-lactoglobulin (BLG) variants (alleles) 15 Genetic Polymorphism of Milk Proteins Variant/ Position 108 118 126 129 GenBank SwissProt allele 28 45 50 56 59 64 70 78 Val Pro Asp 158 accession accession References A Asp B Asp Glu Pro Ile Gln Gly Lys Ile Glu Ala Tyr X14712 Braunitzer et al. (1973) B* Leu C His Glu Z48305 P02754 Braunitzer et al. (1973) DQ489319 Braunschweig and Leeb (2006) As reported by Ng-Kwai-Hang and Grosclaude (2003) D Gln As reported by Brignon and Ribadeau-Dumas (1973) Dr Asn As reported by Bell et al. (1981a) E Gly As reported by Ng-Kwai-Hang and Grosclaude (2003) F Ser Gly Bell et al. (1981a) G Met Gly Bell et al. (1981a) H Asp Asn Val Conti et al. (1988) and Davoli et al. (1987) I Gly Godovac-Zimmermann et al. (1996) J W Leu Godovac-Zimmermann et al. (1996) Godovac-Zimmermann et al. (1990) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column 473

474 P. Martin et al. that it could result from an isoleucine to leucine identified, including six single-nucleotide substi- substitution at position 56 (I56L), but this was not tutions, a single-nucleotide deletion, and a 7 bp further confirmed. This nomenclature is provi- duplication. Comparison of DNA sequences sionally retained here. Compared to previous showed that the investigated 5¢-flanking region is reports, variants Dr and B* are included as true highly conserved between ruminants, and the protein variants. The first (Dr) is quite unique in duplication g.-1885_-1879dupCTCTCGC as well that it can be found in a glycosylated form, as as the substitution g.-1888A > G is found only in described by Bell et al. (1970). The authors first the BLG*A and D alleles in cattle. However, no described a sequence identity with variant A, transcription binding site is predicted to corre- but in a subsequent report (Bell et al., 1981a), spond to the region around the duplication they mentioned an aspartic acid to asparagine (g.−1885_− 1879dupCTCTCGC). A cytosine at substitution at position 28 (D28N), which would position g.-1957 and two thymines at positions be consistent with the migration pattern observed g.-2008 and g.-2049 were found only in BLG*B and with the appearance of a glycosylation site. alleles. It is suggested that the described genetic On the other hand, the product of BLG allele B* variability contributes to the differential allelic (here named variant B*, accordingly) is added to expression of the BLG gene (Braunschweig, the list; notwithstanding, it has the same primary 2007). Recently, Ganai et al. (2009) detected 50 structure as variant B. In fact, the quantitative polymorphisms within the coding, intron and pro- polymorphism associated with BLG*B* (Kim moter regions of bovine BLG gene, of which 42 et al., 1996), resulting in 40% of the amount of were in complete linkage disequilibrium (LD) both transcripts and protein, as compared with the with BLG protein variants A and B. One of the B allele (BLG*B), has been resolved recently eight polymorphisms remaining (six segregating (Braunschweig and Leeb, 2006). It was well with variant A and two with variant B) had a established that the predominant two genetic significant effect on BLG protein concentration. variants, A and B, are differentially expressed This SNP, g.-731G > A, segregated only within (Cerbulis and Farrell, 1975). Numerous studies on cows homozygous for BLG variant A. These new various breeds reported a higher expression of the reported polymorphisms, including the 7 bp BLG A variant compared with the B variant. duplication in the BLG A 5¢-flanking region, may Extensive investigation of the genetic variation in contribute to the allele-specific differential expres- the promoter region of the BLG gene revealed the sion of BLG (Fig. 15.1). The story is far from existence of specific haplotypes associated with being written and the 5¢-flanking region and 5¢ the A and B variants, respectively. Two SNPs UTR of the BLG gene have to be scanned in addi- (g.1772G > A and g.3054C > T) lead to amino acid tional populations. NGS technologies will probably changes (G64D and A118V, respectively) and are provide some answers in the near future. the causal genetic polymorphisms of BLG vari- ants B (G64, A118) and A (D64, V118). However, 15.4.1.6 Bovine a-Lactalbumin the genetic basis for the differential expression of The primary structure of bovine a-lactalbumin BLG A and B alleles still remained elusive. Would was determined by Brew et al. (1970) at the the extremely weak BLG B* variant ultimately amino acid level and confirmed at the nucleic solve the puzzle? Comparative DNA sequencing acid level by Hurley and Schuler (1987) and of 7,670 bp of the BLG*B* allele and the estab- Wang et al. (1989). The mature protein is a poly- lished BLG*B allele revealed a unique differ- peptide of 123 amino acid residues, containing ence of a C to A transversion at position 215 bp eight cysteines that form four intramolecular dis- upstream of the translation initiation site (g.- ulphide bonds. Together with variant B, a variant 215C > A). This mutation segregated perfectly A (Bhattacharya et al., 1963; Grosclaude et al., with the differential phenotypic expression. 1974) and a variant C (Bell et al., 1981b) were Additional genetic variation further upstream in described at the protein level (Table 15.6), the the 5¢-flanking region of the BLG gene was first being present but rare in European cattle and

15 Genetic Polymorphism of Milk Proteins 475 Variant A MKCLLLALALTCGAQA LIVTQTMKGLDIQKVAGTWYSLAMAASDISLLDAQSAPLRVY VEELKPTPEGDLEILLQKWEND ECAQKKIIAEKTKIPAVFKIDALNENKVLVLDTDYKKYL LFCMENSAEPEQSLV CQCLVRTPEVDDEALEKFDKALKALPMHIRLSFNPTQLEEQCHI Allele A (Braunschweig, 2007) transcription x 3 relatively to Allele B -1885:-1879 dupli CTCTCGC TATA box AT –215 C A GC –731 G A AlleleB* (Braunschweig & Leeb, 2006) AlleleB (Ganai et al., 2009) Transcription factors (c-Rel and Elk-1) very low transcription MKCLLLALALTCGAQA LIVTQTMKGLDIQKVAGTWYSLAMAASDISLLDAQSAPLRVY Variant B* VEELKPTPEGDLEILLQKWENG ECAQKKIIAEKTKIPAVFKIDALNENKVLVLDTDYKKYL LFCMENSAEPEQSLA CQCLVRTPEVDDEALEKFDKALKALPMHIRLSFNPTQLEEQCHI MKCLLLALALTCGAQA LIVTQTMKGLDIQKVAGTWYSLAMAASDISLLDAQSAPLRVY Variant B VEELKPTPEGDLEILLQKWENG ECAQKKIIAEKTKIPAVFKIDALNENKVLVLDTDYKKYL LFCMENSAEPEQSLA CQCLVRTPEVDDEALEKFDKALKALPMHIRLSFNPTQLEEQCHI Fig. 15.1 Schematic representation of the bovine b-lac- part to alleles B and B*. The amino acid sequence of toglobulin gene (BLG) and the respective haplotypes the B variant is also given (in smaller frames according for alleles A, B and B* in the 5¢-flanking region of the to their expression level). The black arrow indicates transcription unit. Exons are depicted as large green the duplication g.-1885:-1879dupCTCTCGC. SNP (non-coding) and blue (coding) boxes. The white box g.-731G > A, segregating only within cows homozygous codes for the signal peptide. The yellow arrow upstream for BLG*A (Ganai et al., 2009), is indicated by a pink of exon 1 indicates the position of the TATA Box. The arrow whereas the mutation responsible for the very upper part of the figure is dedicated to variant A, of low transcription of allele B* (g.-215) is indicated by a which the amino acid sequence is framed and the bottom violet arrow Table 15.6 Changes in bovine a-lactalbumin variants (alleles) Variant/allele Position 65 GenBank accession SwissProt accession References A 10 ? Gln P00711 Bhattacharya et al. (1963) B Gln M18780 Brew et al. (1970) Arg Asp/Glu J05147 Bell et al. (1981b) C Asn/Glna His JN258330 Visker et al. (2012) D Gln Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column aThe authors suggested an Asp to Asn or Glu to Gln substitution in the sequence of the mature protein but could not establish more precisely its position the last having been observed but not confirmed (Visker et al., 2012). This SNP, responsible for in Bali cattle. Very recently, a single nucleotide the amino acid substitution Gln65His in this new polymorphism g.600G>T was detected in exon 2 variant (named D), is not expected to affect the of the gene (LALBA) coding for a-lactalbumin protein function.

476 P. Martin et al. 15.4.2 Caprine Milk Proteins between alleles A and F. Referring to phosphory- lation, it can be seen that the substitution Glu to a 15.4.2.1 Caprine as1-Casein Gln at position 77, in the A variant (and the Since the pioneering works by Boulanger et al. derived ones, namely, G, H, I and M), causes the (1984), goat as1-casein is known to possess the loss of a potential phosphorylation site on Ser75. highest degree of polymorphisms and represents The mutation Ser/Leu at position 66 causes the an exceptional paradigm among milk proteins loss of another potential site (Ser 64), in variant (Grosclaude and Martin, 1997; Bevilacqua et al., M. With respect to the chain length, variants G 2002). The protein was first sequenced by and F were described as deleted variants since Brignon et al. (1989) who found it in a 199 amino- they lack, respectively, 13 and 37 amino acid resi- acid (AA) peptide chain, carrying up to 13 poten- dues, due to exon 4 (G) and exons 9–11 (F) skip- tial phosphorylation sites on serine (as well as ping. The D variant has been removed and appears threonine) residues, of which five are clustered cited in brackets in the table. The genetic study by at position 64–68 in the mature protein and does Mahé and Grosclaude (1989) should be more cor- not have any cysteine. Along with its multiple rectly attributed to variant G. For the F allele, in phosphorylation degrees and up to eight so- particular, the deletion of a cytidine in the coding called non-allelic variants, the polymorphism of sequence at position 23 of exon 9 is responsible goat as1-casein is further increased by the pres- for a complex splicing process which in turn ence of at least 18 alleles encoding several results in the production of an array of RNA different polypeptide chains (qualitative poly- forms, five of which appeared to be original F morphism). A list of the known variants described allele, as first reported by Leroux et al. (1992). up to now is given in Table 15.7. In the case of N, These authors suggested that the previously O1, O2, O4-ON, described as null alleles, the described variant D (Brignon et al., 1990) could item is entered as allele, in italics, in Table 15.7. be in fact the result of an alternative splicing event The Z variant was predicted only from an iso- of allele F pre-messengers but failed to recover lated transcript and not confirmed at the protein the exactly corresponding RNA. They actually level. It has to be considered very likely as a found a transcript form in which exon 9 was lack- transcript arising from a cryptic splicing variant, ing as well as Gln78, known to be stochastically lacking glutamine residue at position 78 and lost during the splicing process of primary tran- showing two point mutations Ile/Thr and Lys/ scripts. This hypothesis was confirmed (Ramunno Asn at position 111 and 114 of the mature pro- et al., 2005). Finally, goat CSN1S1 alleles are tein, respectively. It is provisionally included in associated with at least four levels of as1-casein italics. synthesis (quantitative polymorphism). Alleles are therefore referred to as “strong” (A, B1, B2, As shown, variants may differ in primary struc- B3, B4, C, H, L, and M), averaging 3.6 g/L per ture (amino acid substitutions), degree of phos- allele; “intermediate” (E and I), yielding 1.6 g/L; phorylation (loss of potential phosphorylation “weak” (F and G), with 0.6 g/L; and “null” (N, sites) and even length (internally deleted forms). O1, O2, ON), leading to non-detectable amounts It is still to be ascertained whether the ancestral of as1-casein in milk. It must be noted that the E form is variant B2 (as assumed here) or B1. Most and I variants share the same primary sequence as of the other variants may have originated from B4 and A, respectively. They are considered as single-nucleotide mutations leading to amino acid different variants. Nevertheless, it would be more modifications, with the exception of variant M, consistent to consider that the relevant alleles (E for which an interallelic recombination between and B4, as well as A and I) are actually different, an A and a B2 ancestor has been suggested by showing mutations in non-coding sequences, Bevilacqua et al. (2002). More recently, a similar that translate into the same protein at different mechanism has been suggested for the genesis of expression levels (quantitative polymorphism). allele N by Ramunno et al. (2005) who described In the case of the E allele, this phenomenon was it as the possible result of a recombination event

Table 15.7 Changes in caprine as1-casein variants (alleles) 15 Genetic Polymorphism of Milk Proteins Variant/ Position GenBank SwissProt allele 1 8 14–26 16 59–69 59–95 64 66 75 77 78 90 100 111 114 195 accession accession References A Leu Ser Gln AJ504710 Q8MIH4 Brignon et al. (1989) and Ramunno et al. (2004) B1 Leu Grosclaude et al. (1994) and Grosclaude and Martin (1997) B2 Arg His Pro SerP SerP SerP Glu Gln Arg Arg Ile Lys Thr P18626a Grosclaude et al. (1994) and Grosclaude and Martin (1997) B3 Lys Grosclaude et al. (1994) and Grosclaude and Martin (1997) B4 Lys Ala Grosclaude et al. (1994) and Grosclaude and Martin (1997) C Ile Lys Ala Brignon et al. (1989) (D)b DEL Ramunno et al. (2005) and Brignon et al. (1990) E Lys Ala X72221c Jansà-Perez et al. (1994) F DEL AJ504711 Q8MIH3 Leroux et al. (1992) and Ramunno et al. (2005) G DEL Ser Gln Mahé and Grosclaude (1989) and Martin and Leroux (1994) H Lys Leu Ser Gln Chianese et al. (1997a, b) I Leu Ser Gln Chianese et al. (1997a, b) L His Chianese et al. (1997a, b) M Ser Leu Ser Gln Bevilacqua et al. (2002) ND AJ504712 Q8MIH2 Ramunno et al. (2005) O1 AJ252126 Cosenza et al. (2003) O2 Martin et al. 477 (unpublished) (continued)

Table 15.7 (continued) 478 P. Martin et al. Variant/ Position GenBank SwissProt allele 1 8 14–26 16 59–69 59–95 64 66 75 77 78 90 100 111 114 195 accession accession References ON D Ådnøy et al. (2003) and Hayes et al. (2006) Z DEL Thr Asn AY344966 Q69EZ6 Variants are presented in different rows; null alleles are presented in italics; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column DEL: internal deletion of the corresponding sequence in the mature protein D refers to a nucleotide deletion occurring in exons 9 and 12 aUniProt accession number P18626 refers to variant B2 except for a sequence conflict at position 9 (R/Q) in the mature protein. To our knowledge, this conflict is inconsistent with the available literature. All the published sequences show a Q residue at position 9 bPredicted form of alternatively spliced transcripts arising from the F allele, it is not a genetic variant cThere is a sequence conflict (T/A) regarding the amino acid residue at position 194 in the mature protein due to a mistake in the nucleotide sequence given in GenBank (acces- sion number X7222) referring to allele E, codon 209 should be ACT (Thr) instead of GCT (Ala)

15 Genetic Polymorphism of Milk Proteins 479 F B3 B4 E L C B2 + (11+3) nt + (11+3) nt H8 P16 S6P6 E77 R100 T195 H8 L16 S6P6 E77 R100 T195 RsaI O1 H8 L16 S6P6 Q77 R100 T195 H8 P16 S6P6 Q77 R100 T O2 N RsaI I ON H8 P16 L66 Q77 R100 T195 HG RsaI Fig. 15.2 A putative phylogeny integrating a possible these four potentially recombinant alleles (boxed) are interallelic recombination between two allelic lineages of circled. Arrows indicate a possible pathway of evolution as1-casein (adapted from Bevilacqua et al., 2002). Four to alleles associated with high (black) or reduced (pink) alleles (B2, A, B1, and W) putatively involved in a recom- amounts of casein synthesised. The M allele is derived bination event are schematically represented as a chain of from allele W by a single-nucleotide transition C > T six boxes (exons) on which are indicated polymorphic (nucleotide 23/exon 9) leading to the occurrence of a amino acid residues and their position in the peptide Leu residue (allele M) instead of the Ser (putative chain, thus providing a simplified haplotype formula allele W) in the multiple phosphorylation site of the pro- (e.g. HPSPERT and HLSPQRT for alleles B2 and A, tein. This new phylogeny has been enriched with three respectively). The RsaI polymorphic restriction site and novel variants (H, I and L) reported by Chianese et al. insertions occurring between exons 6 and 8 and within (1997a, b) and alleles N (Ramunno et al., 2005) and ON, intron 9, respectively, are indicated. Alleles deriving from the Norwegian “Null” allele (Ådnøy et al., 2003) explained as the result of the insertion of a long tion of transcripts (as the results of alternative interspersed repeated element (LINE) sequence in splicing events) was actually detected (Ramunno the 3¢ UTR that is thought to reduce the stability et al., 2005), amounting to 1/3 those of the F vari- of mRNA (Jansà-Perez et al., 1994). A similar ant (as total RNA concentration), could be in fact mechanism is supposed to explain the reduced a “false null” allele and one or more truncated expression rate of allele I, but has not been proved forms could be present in milk at hardly detect- to date. Interestingly, the F allele is a model of able levels. It seems that allele CSN1S1*N could both qualitative (internal deletion) and quantita- be the counterpart of CSN1S1*F from the allelic tive (rate of expression) polymorphism, as well as lineage A (Fig. 15.2). As far as CSN1S1*ON is of non-allelic variants. Furthermore, it could be concerned, the mutation responsible for the very speculated that allele N, for which a large popula- low expression, if any, of which the frequency is

480 P. Martin et al. very high (ca. 80%) in the Norwegian dairy goat into a variant, named E, differing from variant A population (Ådnøy et al., 2003), is a single-nucle- by the substitution Ser/Tyr at position 166 of the otide deletion occurring in exon 12. This deletion mature protein. Lately, Moatsou et al. (2007) results in the occurrence of a premature stop described a novel variant showing a Tyr instead codon leading to a theoretical truncated protein of an Asp residue at position 47. The authors also made of 137 amino acid residues which remains named this variant E. Here we propose to name it to be found. Given its allele haplotype, it may CSN2*F in order to avoid any possible ambiguity originate from the A lineage. and to respect the chronological order of identification. As mentioned above, the first evi- 15.4.2.2 Caprine b-Casein dence of milk samples lacking the electrophoretic The goat b-casein gene (CSN2), first sequenced band of b-casein was provided by Mahé and by Roberts et al. (1992), translates into a 222-AA Grosclaude (1993) who also suggested the exis- precursor which, after cleavage of the signal pep- tence of two alleles controlling this phenomenon tide, is secreted as a 207 AA polypeptide chain. in the Creole population analysed. Two distinct It may carry up to five phosphoserine residues alleles were then described by Persuy et al. (1996, (four of which are clustered at positions 15, 17, 1999) and by Rando et al. (1996) that could be 18 and 19) and two phosphothreonine residues associated with the absence of this protein in and has no cysteine. Table 15.8 summarises the milk. They were named, respectively, CSN2*O1 known protein variants and alleles. Compared to and CSN2*O2. In the CSN2*O1 allele, the dele- as1-casein, the protein is thought to possess a tion of a single nucleotide (A residue) in a AAAA lower degree of polymorphism, since only eight sequence between residues 16 and 19 of exon 7 alleles have been characterised so far, two of causes a frameshift resulting in a premature stop which were originally described as “null” alleles. codon in the coding sequence of the cDNA The B variant was the first to be described (Mahé (codon 73). In the CSN2*O2 allele, two single and Grosclaude, 1993), together with a null allele point mutations were identified: the first one is a (CSN2*O). The authors did not determine its pri- T to C transition at nucleotide −388 from the start mary structure, but speculated that the B variant site of transcription and the second is a C to T could differ from A by an additional phosphate transition at position 373 of exon 7, creating a group. Keeping in mind that the protein is present premature stop codon in the coding sequence mainly in its 6 and 5 phosphate forms, it still (codon 182) of the cDNA (Fig. 15.3). A 20-fold remains uncertain whether the B variant origi- reduction in the quantity of mRNA was observed nates from a mutation leading to one more poten- with CSN2*O1 (Persuy et al., 1999), whereas a tial phosphorylation site or is merely an A variant tenfold reduction was observed with CSN2*O2 at its maximum degree of phosphorylation. The (Ramunno et al., 1995). Despite the marked genomic DNA sequence of a novel allele was reduction in the amount of mRNA, Ramunno first submitted by Wang et al. (accession number et al. (1995) found that milk from homozygous AF409096). The primary structure and phospho- O2/O2 goats contained ca. 1.6% of b-casein, rylation sites of the corresponding protein were compared to 53% in the milk from homozygous subsequently given by Neveu et al. (2002a), who goats A/A at the CSN2 locus. Consistent with this named it variant C. A non-allelic variant D was result, Cunsolo et al. (2006) combining HPLC described by Galliano et al. (2004) in homozy- and MS/MS techniques, identified and sequenced gous C goats at the CSN2 locus and probably a 166 AA protein whose sequence corresponds originated from an incorrect translation of the to the truncated protein theoretically encoded by genetic information, leading to an Asn instead of the CSN2*O2 allele. Recently, a new CSN2 allele a Val residue at position 207. Caroli et al. (2006), (here named CSN2*O3) showing a SNP in a study on casein genetic polymorphisms in (g1311T > C) in the promoter region has been local Italian breeds, identified a novel mutation at reported as being associated with the absence of the CSN2 locus by PCR-SSCP, which translated b-casein in milk (Cosenza et al., 2007). In fact,

Table 15.8 Changes in caprine b-casein variants (alleles) 15 Genetic Polymorphism of Milk Proteins Variant/allele Position 47 166 167–207 177 207 GenBank accession SwissProt accession References A Asp Ser Ala Val AH001195.1 P33048 Roberts et al. (1992) B Mahé and Grosclaude (1993) C Val AF409096 (D)a Val Asn Galliano et al. (2004) E Tyr Caroli et al. (2006) F Tyr Moatsou et al. (2007) and Chianese et al. (2007a) O1 AF172260 Mahé and Grosclaude (1993) and Persuy et al. (1999) O2 DEL AJ011019 Rando et al. (1996) and Cunsolo et al. (2005) O3 AJ011018 Cosenza et al. (2007) Variants are presented in different rows; null alleles are presented in italics; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column DEL: internal deletion of the corresponding sequence in the mature protein aObserved as a non allelic variant from homozygous goat C/C at the CSN2 locus 481

482 P. Martin et al. Induction nucleotide deletion (A) Persuy et al. (1999) MGNF 0.15 STOP codon MGNF -90 0.15 -140 NF2 NF4 1.8 0.8 2.2 1.2 0.7 0.7 NF1 NF3 exon 7 8.5 kb Repression C>T Transition Rando et al. (1996) STOP codon Fig. 15.3 Schematic representation of the goat b-casein (yellow arrow) whereas the second one (Rando et al., encoding gene (CSN2). Two distinct mutations responsi- 1996) occurring in the 3¢ end of the same exon is due to a ble for the occurrence of premature STOP codons and C > T transition (green arrow). Nuclear regulatory factors consequently for the absence of b-casein in goat milk and their interaction sites with DNA in the 5¢ flanking have been described. The first one (Persuy et al., 1999) is region of the transcription unit are schematised due to a single-nucleotide deletion in the 5¢ part of exon 7 this SNP corresponds to the first mutation, T to C mature protein. Moreover, it contains two cysteine transition at nucleotide −388, previously reported residues that are available for intermolecular dis- by the same group. Therefore, one can expect ulphide bridges. At present, a total of seven that this single SNP in the promoter region, alone CSN1S2 alleles have been described in the goat in itself, accounts for the absence of b-casein in species (Table 15.9). Most of them differ by sin- milk. gle-nucleotide substitutions in the coding sequence with the exception of alleles CSN1S2*D 15.4.2.3 Caprine as2-Casein and E. CSN1S2*D apparently associated with The goat as2-casein was first sequenced by decreased synthesis of the protein is character- Bouniol (1993) at the cDNA level. The mature ised by an internal deletion spanning over 106 protein has 208 AA residues and the first detailed nucleotides, involving the last 11 nucleotides of data on the genetic polymorphism were reported exon 11 and the first 95 nucleotides of the follow- by the same group (Bouniol et al., 1994). Of the ing intron. This large deletion triggers a deep two alleles, A and B, previously described at the rearrangement of the messenger in which codon goat CSN1S2 locus, the former was resolved into 121 is modified to code for an Asn residue two alleles, named CSN1S2*A and CSN1S2*C. (instead of a Thr residue in variant A), and the as2-Casein C, which cannot be distinguished from following three codons (122–124) are lost as2-casein A by starch or polyacrylamide gel (Ramunno et al., 2001a, b). In the case of variant electrophoresis, was shown to differ by a single E, conflicting sequences have been reported for substitution Lys/Ile at position 167 of the mature the mRNA (GenBank accession number protein. Goat as2-casein, like its bovine counter- AJ249995) and the joined coding DNA (GenBank part, is the most phosphorylated casein with up to accession numbers AJ242526, AJ242527, 17 potential phosphorylation sites (11 serine and AJ242528, AJ242533, AJ242728, AJ242866, 6 threonine residues), nine of which occur in AJ249995, AJ297310, AJ297311, AJ297312, three clusters regularly distributed on the peptide AJ297313, AJ297314, AJ297315, AJ297316, chain at position 8SerP-SerP-SerP10, 57SerP- AJ298297) published by Lagonigro et al. (2001), SerP-SerP59 and 130SerP-ThrP-SerP132 of the and the possible amino acid modifications are

Table 15.9 Changes in caprine as2-casein variants (alleles) 122–124 144 146 167 193 GenBank SwissProt References 15 Genetic Polymorphism of Milk Proteins Position SerP Glu Lys Pro accession accession Bouniol (1993) X65160.1 P33049 Bouniol et al. (1993b) Variant/allele 7 62 64 110–208 121 DEL Ile Arg Bouniol et al. (1994) A Val SerP Glu Thr (Ser) (Lys) Ile S74171 Q9XSL1 Ramunno et al. (2001a) B Ser Lys AJ238684 Lagonigro et al. (2001) C AJ249995.1 Ramunno et al. (2001b) D Asn AJ289716.1 Ramunno et al. (2001a) (E) a AJ289715.1 F Ile O DEL Variants are presented in different rows; the null allele is presented in italics; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column DEL: internal deletion of the corresponding sequence in the mature protein aConflicting sequences concerning the Glu to Lys modification at position 146 were reported for mRNA and genomic DNA 483

484 P. Martin et al. provisionally presented into brackets. In the case mary structures of such alleles, however, do not of the null allele (CSN1S2*O), a transversion G always result in new protein variants, and there- to A occurs at the 80th nucleotide of exon 11, fore, the nomenclature proposed by these authors giving rise to a premature stop codon in the cDNA was modified accordingly to follow the chrono- sequence (codon 110). A truncated protein of 109 logical order of publication. Regarding the new AA could therefore be predicted, but none has variants reported by Gupta et al. (2009), several been detected in milk up to date. differences could be found between sequences provisionally submitted to GenBank and those 15.4.2.4 Caprine k-Casein presented and discussed later in the cited paper. Before the last edition of Advanced Dairy In case of conflict, the amino acid substitutions Chemistry-1: Proteins in 2003, k-casein was given in Gupta et al. (2009) were preferred. thought to be scarcely polymorphic in the goat species, as very little evidence had been found 15.4.2.5 Caprine b-Lactoglobulin and for amino acid modifications or differences in a-Lactalbumin electrophoresis patterns. Nevertheless, over the last 10 years, the systematic use of molecular To date, no protein polymorphisms have been biology approaches revealed an array of genetic described for these whey proteins in the goat spe- polymorphisms, many of which translated into cies, with the exceptions of isolated observations different protein variants. Most of these new by Boulanger (1976) and Russo et al. (1986), developments have been reviewed by Prinzenberg respectively, on BLG and a-lactalbumin. These et al. (2005) who proposed a new nomenclature findings, however, did not lead to the characteri- basically according to the chronological order sation of new variants nor were they confirmed in of GenBank accession numbering. Table 15.10 following studies. shows the variants detected to date in the goat species. Applying the principle proposed by 15.4.3 Ovine Milk Proteins Prinzenberg et al. (2005), the nomenclature was revised and variant L which is actually a dif- 15.4.3.1 Ovine as1-Casein ferent allele but coding for the same protein Ovine as1-casein was first sequenced by Ferranti variant as G was suppressed, and the successive et al. (1995). It is a 199 AA-residue polypeptide variants have been renamed accordingly. with up to ten potentially phosphorylated serine residues displaying a 97% identity with the pri- Caprine k-casein cDNA was first sequenced mary structure of its caprine counterpart. The two by Coll et al. (1993). The mature protein is a proteins share a common feature: a high molecu- 171-amino-acid chain (two residues more than its lar diversity in part due to alternative splicing bovine counterpart) with three serine and one variants. Nevertheless, unlike the goat, the exis- threonine residues being potentially phosphory- tence of quantitative polymorphisms or null lated. It has two cysteine residues at positions 10 alleles has not been proved so far in the ovine and 11 that are available for the formation of species, although Giambra et al. (2010a) reported intermolecular disulphide bridges with other recently an average 26% reduction in the expres- k-casein or as2-casein molecules. Moreover, in sion level of the CSN1S1*H allele. The first vari- analogy with the other k-casein, it can have vari- ants to be described, A, B, C, D and E, migrate ous degrees of glycosylation. with decreasing speed on basic electrophoresis gels and show the opposite behaviour in acidic Regarding its genetic polymorphism, since the capillary electrophoregrams (as reported by publication of the above-mentioned review by Amigo et al., 2000). With the exception of vari- Prinzenberg et al. (2005), a systematic sequenc- ant B, of which the primary structure remains to ing of exon 4 made it possible to identify seven be determined (but for which the modification of new alleles in Indian Jakhrana goats (Gupta et al., an acidic AA to a neutral one can be hypothe- 2009) and four in goat populations from East Africa (Kiplagat et al., 2010). The predicted pri-

Table 15.10 Changes in caprine k-casein variants (alleles) 15 Genetic Polymorphism of Milk Proteins Position Variant/ 53 61 65 77 90 119 145 156 159 GenBank SwissProt accession References allele 44 Val Ala Ser accession P02670 X60763 Coll et al. (1993), Gupta et al. (2009) and Kiplagat A Gln Asn Tyr Val Gln Asp Val Val Pro EF053353 Q540J1 et al. (2010) Pro AF485340 B Ile Pro AF434988 Yahyaoui et al. (2001), Jann et al. (2004) and Pro AY166705 Kiplagat et al. (2010) AY166706 C Ile Ile Ala Pro AY166707 Yahyaoui et al. (2001) and Prinzenberg et al. (2005) AY350425 D Arg Ile Ile Pro AF485341 Caroli et al. (2001) and Yahyaoui et al. (2001) Pro AY027868 E Gly Ile AY090465 Angiolillo et al. (2002) F Ile AF486523 Yahyaoui et al. (2003) G Ile Ile AY090466 Yahyaoui et al. (2003) and Jann et al. (2004) AY090467 H Ser Ile AY166708 Jann et al. (2004) I Ile Ile AF521022 Jann et al. (2004) J Cys Ile AY166710 Jann et al. (2004) K Arg Ile AY166711 Jann et al. (2004) AY166709 L Asn Ile EF053350 Prinzenberg et al. (2005) and Kiplagat et al. (2010) M Arg Ser Arg Ile AY428577 Gupta et al. (2009) EF053351 N Arg Arg EF053354 Gupta et al. (2009) O Ser Ile EF053352 Gupta et al. (2009) P Ser EF053355 Gupta et al. (2009) Q EF053356 Kiplagat et al. (2010) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column DEL: internal deletion of the corresponding sequence in the mature protein 485

Table 15.11 Changes in ovine as1-casein variants (alleles) 486 P. Martin et al. Position GenBank SwissProt accession accession References Variant/allele 12 13 43–50 51–58 64 66 68 70–77 194 A SerP Ser P04653 Ferranti et al. (1995) B Chianese et al. (1996) C¢ Thr AY444506 Ceriotti et al. (2004, 2005) C Ser Pro SerP SerP SerP Ile X03237a Ferranti et al. (1995) D Ser Ser Asn Ferranti et al. (1995) E Ser Ser Ser DEL Chianese et al. (1996, 2007b) F Pirisi et al. (1999) G Chianese (unpublished) H DEL FJ440846 Giambra et al. (2010a) I DEL FJ695513 Giambra et al. (2010b) FJ695515 Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column DEL: internal deletion of the corresponding sequence in the mature protein aThe original mRNA sequence refers to the alternative spliced variant lacking exon 16, coding for a 191 AA residues internally deleted protein, lacking residues 141 to 148

15 Genetic Polymorphism of Milk Proteins 487 sised), protein sequencing of the other four vari- Chianese (1997) first identified two variants, ants made it possible to explain this variability named B and C, by IEF, but only the sequence of with a progressive loss in potential phosphoryla- variant C is presently available. It contains a Gln tion sites, as shown in Table 15.11. In particular, instead of a Glu residue at position 2. Ceriotti the substitution Ser/Pro at position 13 would et al. (2004) and Chessa et al. (2010) identified interfere with the phosphorylation of Ser12 in two polymorphic patterns within exon 7 by PCR- variant C. Variant D and E lack, respectively, 2 SSCP and DNA sequencing that lead up to two and 3 phosphoserine residues in the phosphoryla- novel protein variants, here named D and E, tion cluster at position 64–68 due to an amino respectively. Compared to the reference variant, acid substitution at position 68 and the deletion the first one has a Val residue instead of a Met at of eight AA residues (encoded by exon 8) between position 183; in the latter, the Leu residue at posi- positions 70 and 77, respectively. Recently, tion 196 is replaced by an Ile residue. Ceriotti et al. (2004, 2005) found and character- ised a novel allele (named CSN1S1*C¢) coding 15.4.3.3 Ovine as2-Casein for a protein that differs from the C variant by a In the case of ovine as2-casein, a series of single-AA substitution, Ile/Thr, at position 194 sequence conflicts and inconsistent nomenclature (and not at position 186, as mentioned errone- make the identification of variants difficult. ously in the original manuscript by Ceriotti et al., Therefore, to clarify the situation, a new nomen- (2004), according to the published sequence. The clature is proposed in the present chapter, follow- authors suggested that this could be the ancestral ing the chronological order in which each variant form in the ovine species because a Thr residue is was detected, to unambiguously describe the conserved at the same position both in goat and state of the art regarding the genetic polymor- in cattle. Two novel variants named F and G were phism of this protein. found, respectively, by IEF and RP-HPLC tech- niques, but their primary structure has not yet A first sequence of ovine as2-casein was pro- been published. More recently, two novel vari- vided by Boisnard and Petrissant (1985) based on ants, named H and I, were identified by Giambra mRNA (GenBank accession number, X03238.1; et al. (2010a, b) and characterised at the protein UniProt accession number, P04654). Compared and nucleotide levels. Both are described as inter- to the bovine and caprine sequences (both having nally deleted proteins lacking the amino acid an Asn residue at position 49), the sequence by sequence encoded by exon 8 and exon 7, respec- Boisnard and Petrissant (1985) has an Asp resi- tively. Variant I, thus, loses two potential phos- due at this position, whereas all of the sequences phorylation sites encoded by exon 7. published later have an Asn residue instead. It is therefore referred to as variant A¢ in Table 15.13 15.4.3.2 Ovine b-Casein and will not be considered as the reference The sequence of ovine b-casein, first proposed by sequence hereafter. The actual sequence of vari- Richardson and Mercier (1979) according to ant A differs from that of variant A¢ solely for the automated and manual Edman degradation Asn49 residue, and the genomic DNA sequence sequencing of acid and enzymatic hydrolysates, corresponding to CSN1S2*A allele is now avail- was later corrected by Provot et al. (1989), who able in GenBank (Giambra et al., in press). In a sequenced the cDNA and the gene (Provot et al., subsequent report by Boisnard et al. (1991), an 1995). Similar to the caprine protein, the precur- allelic variant, here named B, was identified at sor of ovine b-casein is a 222 AA-residue poly- the transcript level and characterised by a Lys to peptide, made of a 15 AA-signal peptide and a Asn substitution at position 200. This mutation 207 AA-mature protein. Potential phosphoryla- was further confirmed by Chessa et al. (2010) by tion sites are the same as in the goat protein, and sequencing of exon 16. One more protein variant ovine b-casein likewise lacks cysteine residues. (here named C) was described by Chianese et al. The known variants are presented in Table 15.12. (1993) applying an extensive one- and two- dimensional electrophoresis approach, combined

Table 15.12 Changes in ovine b-casein variants (alleles) 488 P. Martin et al. Position Variant/allele 2 183 196 GenBank accession SwissProt accession References A Glu Met Leu X16482 P11839 Provot et al. (1989, 1995) X79703 B Chianese (1997) C Gln Chianese (1997) D Val AY444504 Ceriotti et al. (2004) E Ile Chessa et al. (2010) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column

Table 15.13 Changes in ovine as2-casein variants (alleles) 15 Genetic Polymorphism of Milk Proteins Position GenBank SwissProt accession accession References Variant/allele 45 46 48 49 75 105 161 200 A¢ Asp X03238.1 P04654 Boisnard and Petrissant (1985) A Val Arg Ala Asn Asp Ile Arg Asn GU169085 Boisnard et al. (1991) and Giambra et al. (in press) B Tyr Boisnard et al. (1991) and Chessa et al. (2010) C Chianese et al. (1993) and Recio et al. (1997b, c) D Tyr Val GU169086 Chessa et al. (2003), Picariello et al. (2009a) and Giambra et al. (in press) E Ile Ser(P) GU169087 Giambra et al. (in press) F Ser GU169088 Giambra et al. (in press) G His GU169089 Giambra et al. (in press) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column 489

490 P. Martin et al. Table 15.14 Changes in ovine k-casein variants (alleles) Variant allele 2 7 104 GenBank accession SwissProt accession References A¢ Gln A Glu Glu Jolles et al. (1974a, b) B Gln Ser X51822 P02669 Furet et al. (1990) Leu AY444505 Ceriotti et al. (2004) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column Table 15.15 Changes in ovine b-lactoglobulin variants (alleles) Variant allele 20 148 GenBank accession SwissProt accession References A Tyr X12817 P67976 Ali et al. (1990) B His Arg Gaye et al. (1987) C Gln Erhardt (1989a) Variants are presented in different rows; amino acids in the reference variant are in boldface; amino acid modifications are given in the relevant column with immunoblotting experiments, and was sub- et al. (1990) reported a cDNA sequence which sequently confirmed by Recio et al. (1997c). The therefore included the signal peptide. Like its authors named the variant “fast” due to its marked caprine counterpart, mature ovine k-casein is a anodic mobility. They concluded that it should be 171 residue polypeptide with four potential phos- an internally deleted variant with a greater nega- phorylation sites (three serine and one threonine), tive net charge and lower isoelectric point, but no two cysteine residues and several potential further characterisation was performed. Later, O-glycosylation sites. Unlike its bovine and Chessa et al. (2003) detected a novel protein vari- caprine counterparts, the ovine protein appears to ant showing higher isoelectric point than variant be scarcely polymorphic, since only a single vari- A on IEF gels and named it B (here renamed D). ant, detected on genomic DNA by PCR-SSCP, Picariello et al. (2009a ), using MALDI-TOF MS has been described so far (Ser to Leu substitu- and MALDI-PDS-TOF MS, were able to identify tion, at position 104) by Ceriotti et al. (2004). two amino acid substitutions, Ile for Val at posi- tion 105 and Asp for Tyr at position 75, the latter 15.4.3.5 Ovine b-Lactoglobulin being responsible for the newly observed IEF Ovine BLG is a 162 amino-acid polypeptide. pattern. As mentioned above, it is only recently Since the work of Bell and McKenzie (1967), that Giambra et al. (in press) were able to confirm two variants of ovine BLG are known and named the genomic DNA sequences of variants A A and B. The cDNA sequence for variant B was (Asn49) and D (Asn49, Tyr75, Val105) (accord- published by Gaye et al. (1987), and the mutation ing to the new nomenclature proposed) and responsible for the His to Tyr substitution at posi- describe three novel variants (here named E, F tion 20 in variant A was reported by Ali et al. and G). (1990), whereas a novel protein variant (named C) described by Erhardt (1989a) was sequenced 15.4.3.4 Ovine k-Casein (Prinzenberg and Erhardt, 1999). The three vari- The first sequence of ovine k-casein (named A¢ ants are presented in Table 15.15. in Table 15.14) was published by Jolles et al. (1974a, b) for para-k-casein and for the caseino- 15.4.3.6 Ovine a-Lactalbumin macropeptide, respectively, based primarily on Two nucleotide sequences are available for the trypsin digestion of the mature protein and direct ovine a-lactalbumin on GenBank (accession N-terminal sequencing of peptides. Later, Furet numbers X06367 and AB052168), the latter