Exp. Anim. 61(1), 67–70, 2012

—Note— Identification of Whey Acidic Protein (WAP) in Dog Milk

Mami SEKI1, 7), Rina MATSURA2), Tokuko IWAMORI3), Naoko NUKUMI4), Keitaro YAMANOUCHI5), Kiyoshi KANO1), Kunihiko NAITO1), and Hideaki TOJO6)

1)Laboratory of Applied Genetics, The University of Tokyo, 1–1–1 Yayoi, Bunkyo-ku, Tokyo 113-8657, 2)Department of Research and Development, Taisho Pharmaceutical Co., Ltd., 3–24–1 Takada, Toshima-ku, Tokyo 170-8634, Japan, 3)Department of Pathology, Baylor College of Medicine, S217 One Baylor Plaza, Houston, TX 77030, USA, 4)Pharmaceuticals and Medical Device Agency, 3–3–2 Kasumigaseki, Chiyoda- ku, Tokyo 100-0013, 5)Laboratory of Veterinary Physiology, The University of Tokyo, 1–1–1 Yayoi, Bunkyo-ku, Tokyo 113-8657, 6)Yamazaki College of Animal Health Technology, 4–7–2 Minamioosawa, Hachioji, Tokyo 192-0364, and 7)Present address: Tokyo New Drug Research Laboratories, Kowa Co., Ltd., 2–17–43 Noguchimachi, Higashimurayama, Tokyo 189-0022, Japan

Abstract: Whey acidic protein (WAP) has been identified as a major whey protein in milk of a wide range of species and reportedly plays important roles in regulating the proliferation of mammary epithelial cells. However, in some species including humans, WAP is not synthesized in the mammary gland. The presence of WAP in carnivore species has not been reported. We searched the National Center for Biotechnology Information (NCBI) database for the dog WAP gene and tried biochemically to identify WAP in dog milk. The nucleotide sequence of the examined dog genomic DNA was completely identical to that in the NCBI database and showed that the dog WAP gene, like other known functional WAP genes, has four exons. Biochemical analysis of milk protein by reverse-phase HPLC and Western blotting demonstrated the presence of WAP in dog milk. Key words: dog milk, NCBI database, WAP

Whey acidic protein (WAP) has been identified in mammary gland. It would be interesting to investigate various species as a major whey protein in milk [1, 3–6, the biological meaning of the absence of WAP synthesis 9, 15, 21]. WAPs share structural similarity with serine (loss of WAP gene function) in the mammary gland of protease inhibitors containing WAP motif domains char- such species from the standpoint of the molecular evolu- acterized by a four-disulfide core (4-DSC) [9]. Possible tion of the WAP gene. It is unknown whether or not WAP physiological functions of WAP have been proposed is synthesized in carnivore species. In the present study, based on its similarity to protease inhibitor [6]. Our pre- we searched the National Center for Biotechnology In- vious studies on WAP function utilizing in vivo [7, 10, formation (NCBI) database for the dog WAP gene and 12, 16] and in vitro [11, 16, 19] systems showed that WAP attempted to biochemically identify WAP in dog milk. plays important roles in regulating the proliferation of First, we searched the Canis familiaris-WGS trace mammary epithelial cells. The function of WAP has led archive database with the Discontiguous MegaBLAST to speculation that its synthesis and secretion are wide- program using the pig WAP DNA sequence as a query spread in a range of species. However, in some species and obtained 24 hits. The dog WAP gene sequence was including humans [18], WAP is not synthesized in the predicted by assembling the matched sequence. Then,

(Received 12 July 2011 / Accepted 31 July 2011) Address corresponding: H. Tojo, Yamazaki College of Animal Health Technology, 4–7–2 Minamioosawa, Hachioji, Tokyo 192-0364, Japan

© 2012 Japanese Association for Laboratory Animal Science 68 M. Seki, ET AL.

Fig. 1. Computational and biochemical analyses of dog WAP. A, Alignments of nucleotide and amino acid sequences predicted by using the BLAST Web server of NCBI and the ExPASy Translate Tool (http:// kr.expasy.org/tool.dna.html). The amino acid sequence of the predicted protein is shown above the DNA sequence. The nucleotide sequence of the examined dog genomic DNA was completely identical to the dog WAP sequence in the NCBI database. The arrowhead indicates the signal peptide cleavage site. Phosphorylation sites are boxed. The nucleotide of each exon is marked by a forward arrow below the sequence. WAP domains are underlined. The grey box corresponds to 100% of the 15 amino ter- minal amino acid residues determined by N-terminal amino acid sequencing of dog WAP. B, Separation of WAP from dog milk by reverse-phase HPLC. WAP above peak 2 indicates the fraction that showed a positive signal in immunoblotting analysis in D. C, SDS-polyacrylamide gel (15%) electrophoresis and Coomassie Brilliant Blue G250 staining of elutes. D, Immunoblotting of WAP in elutes. we searched the National Center for Biotechnology In- miliaris cont2.035360, whole genome shotgun sequence formation (NCBI) nucleotide sequence database with the (GenBank AAEX02035361). The alignment of the dog BLASTIN program, and the predicted dog WAP gene WAP DNA sequence, like other WAP DNA sequences sequence matched position 25184–23606 of Canis fa- from the mouse [9], rat [3], rabbit [5], and pig [21], WHEY ACIDIC PROTEIN IN DOG MILK 69 showed that the consensus sequences such as the TATA tems) and a 15–100% gradient of acetonitrile in 0.1% box (TTTAAA), translation initiation site (ATG), stop trifluoroacetic acid. The absorbance of the protein peaks codon (TGA), GT-AG at the exon-intron boundaries, and was measured at 215 nm. Each elute fraction was con- poly-A additional signal were conserved in the dog WAP centrated by a centrifugal evaporator, dissolved in 20 µl gene, respectively. The NCBI database also showed that of Laemmli buffer, denatured, and stored at –80°C until the functional WAP gene, like other known functional use. The fractions were subjected to SDS polyacrylam- WAP genes, of the dog has four exons (Fig. 1A). Next, ide gel electrophoresis in 15% polyacrylamide gel. The to confirm the putative dog WAP gene sequence, we proteins were stained with Coomassie Brilliant Blue amplified the WAP DNA sequences from dog genomic R250. Western blotting was performed as follows to DNA extracted from Saint Bernard blood cells by PCR determine whether or not WAP was present in the dog using the following primer set: 5’-GCCACCATGCGCT- milk. Rat whey [13] was used as a positive control. GTCTTGCT-3’ (forward) and 5’-GAAGGTTC TTGC- Whey fractions isolated from the milk were separated CAAGCAGACT-3’ (reverse). The PCR conditions were on SDS-PAGE (15%) gel and transferred to a membrane 40 cycles for 30 s at 94°C, for 30 s at 66°C, and for 90 by semidry blotting. The membrane was blocked in s at 72°C. The PCR product was inserted into pGEM-T Tris-HCl-buffered saline (TBS, pH 7.4) containing 3% Easy (Promega, Madison, WI, USA). DNA sequencing skim milk for 20 min and incubated with anti-rat WAP of the PCR product was performed by using a commer- antibody (1:200) (Santa Cruz Biotechnology, Santa Cruz, cial sequencing kit (Applied Biosystems, Carlsbad, CA, CA, USA) overnight at 4°C. After washing, the mem- USA) and a DNA sequencer (ALFexpress, Amersham- brane was incubated with rabbit anti-goat horseradish Pharmacia, Tokyo, Japan) according to the manufac- peroxidase-conjugated secondary antibody (1:5,000) turer’s instructions. It was confirmed that the nucleotide (Jackson ImmunoResearch Laboratories, West Balti- sequence of the PCR product from dog genomic DNA more, PA, USA). The membrane was developed with completely corresponded to the predicted dog WAP gene an ECLTM Western Blotting Detection Reagents Kit (Am- sequence. ersham Biosciences, Tokyo, Japan) according to the The doge WAP amino acid sequence was predicted manufacturer’s instructions. Nine fractions were eluted using the Translate Tool of ExPASy (http://kr.expasy. by reverse-phase HPLC of whey protein from dog milk, org/tools/dna.html), and the protein structure was pre- as shown in Fig. 1B. From the SDS-polyacrylamide gel dicted using SignalP (http://www.cbs.dtu.dk/services/ electrophoresis and Coomassie Brilliant Blue staining SignalP/) and InterProScan (http://www.ebi.ac.uk/Inter- of elutes, the fractions 3–6 each showed a single band, ProScan). Dog WAP cDNA encodes 136 translated while other fractions showed multiple bands (Fig. 1C). amino acids with conserved residues composed of a Western blotting analysis of each elute indicated the 19-amino acid signal peptide coding region and two WAP positive signal only for fraction 2, shown in Fig. 1B and domains (Fig. 1A). These results account for the pres- 1C, as well as for the rat whey fraction (Fig. 1D). The ence of WAP from dog milk found by reverse-phase molecular size of dog WAP was predicted to be 14–18 HPLC and Western blotting analysis, which are described kDa (Fig. 1D). in the latter section of this paper. Lastly, determination of the amino-terminal amino Furthermore, we attempted to identify WAP in dog acid sequence (15 residues) of putative dog WAP was milk. Dog milk was collected from a lactating Saint consigned to APRO Life Science Institute, Inc. (Naruto, Bernard at 4 days of lactation. The milk was diluted Tokushima, Japan). Determination of the amino acid twice with chilled distilled water, and the fat was re- sequence of the N-terminus of the fraction 2 confirmed moved by centrifugation for 15 min at 3,100 × g at 4°C. that 15 amino acid residues (from a.a20 to a.a34) coin- The supernatant was adjusted to pH 4.6, and caseins were cided with 100% of the predicted amino acid sequence removed in the same manner. The supernatant was ad- of dog WAP, as shown in Fig. 1A. justed to pH 7.2 and subjected to a reverse-phase HPLC WAP has been found to be a major component of whey using a POROS R2 10 micron column (Applied Biosys- in the milk of various species and has two WAP domain 70 M. Seki, ET AL. structures at the four-disulfide core (4-DSC). A number References of proteins containing 4-DSC domains have been identi- fied as protease inhibitors [8, 14, 22]. Based on its struc- 1. Beg, O.U., Von Bahr-Lindstrom, H., Zaidi, Z.H., and Jornvall, H. 1986. Eur. J. Biochem. 159: 195–201. tural similarity to proteinase inhibitors, WAP has been 2. Burdon, T., Wall, R.J., Shamay, A., Smith, G.H., and proposed to have a possible function. Overexpression Hennighausen, L. 1991. Mech. Dev. 36: 67–74. of the WAP transgene in the mammary glands of trans- 3. Campbell, S.M., Rosen, J.M., Hennighausen, L.G., Strech- Jurk, U., and Sippel, A.E. 1984. Nucleic Acids Res. 12: genic animals has been shown to inhibit the proliferation 8685–8697. and advanced functional differentiation of mammary 4. Demmer, J.S., Stasiuk, J., Grigor, M., Simpson, K.J., and epithelial cells [2, 10, 16]. We previously reported that Nicholas, K.R. 2001. Biochim. Biophys. Acta 1522: 187– . the enforced expression of exogenous WAP significant- 194 5. Devinoy, E., Hubert, C., Schaerer, E., Houdebine, L.M., and ly inhibited the proliferation of HC11 cells, a mouse Kraehenbuhl, J.P. 1988. Nucleic Acids Res.16: 8180. mammary epithelial cell line, through the suppression 6. Grabowski, H.D., Chêne, N., Attal, J., Maliénou-Ngassa, of cyclin D1 [16]. Ikeda et al. [11] reported, in a study R., Puissant, C., and Houdebine, L.M. 1991. J. Dairy Sci. 74: 4143–4150. using EpH4/K6 mouse mammary epithelial cells, that 7. Hase, W., Seo, B.B., Tojo, H., Tanaka, S., and Tachi, C. WAP plays a negative regulatory role in the cell-cycle 1996. J. Reprod. Dev. 42: 265–271. progression of mammary epithelial cells through an au- 8. Heinzel, R., Appelhans, H., Gassen, G., Semuller, U., , . , . , . . . . tocrine/paracrine mechanism. In addition, in a study Machleidt W , Fritz H , and Steffens G 1986 Eur J Biochem. 160: 61–67. using MCF-7 cells, a human breast cancer cell line, Nu- 9. Hennighausen, L.G. and Sippel, A.E. 1982. Nucleic Acids kumi et al. [17, 20] suggested that WAP has a protease Res. 10: 2677–2684. inhibitory function that affects the progression of breast 10. Ikeda, K., Kato, M., Yamanouchi, K., Naito, K., and Tojo, H. 2002. Exp. Anim. 51: 395–399. cancer cells. Furthermore, Iwamori et al. [13] isolated 11. Ikeda, K., Nukumi, N., Iwamori, T., Osawa, M., Naito, K., WAP from rat milk by column chromatography and re- and Tojo, H. 2004. J. Reprod. Dev. 50: 87–96. ported that WAP inhibited the growth of Staphylococcus 12. Iwamori, T., Osawa, M., Nukumi, N., Sudo, K., Kano, K., , ...... aureus JCM2413 but did not inhibit the growth of Es- Naito K , and Tojo, H 2005 J Reprod Dev 51: 579– 592. cherichia coli. 13. Iwamori, T., Nukumi, N., Itoh, K., Kano, K., Naito, K., From these studies, Nukumi et al. [19] concluded that Kurohmaru, M., Yamanouchi, K., and Tojo, H. 2010. J. Vet. WAP plays an important role in regulating the prolifera- Med. Sci. 72: 621–625. 14. Kirchoff, C., Habben, I., Ivell, R., and Krull, N. 1991. Biol. tion of mammary epithelial cells by preventing elastase- Reprod. 45: 350–357. type serine proteases from carrying out laminin degrada- 15. Nicholas, K.R., Fisher, J.A., Muths, E., Trott, J., Janssens, tion and thereby suppressing the MAP kinase signal P.A., Reich, C., and Shaw, D.C. 2001. Comp. Biochem. . . . . pathway in the cell cycle. Thus, WAP plays an important Physiol, Part A Mol Integr Physiol 129: 851–858 16. Nukumi, N., Ikeda, K., Osawa, M., Iwamori, T., Naito, K., role in regulating the proliferation of mammary epithe- and Tojo, H. 2004. Dev. Biol. 274: 31–44. lial cells. However, in some species including humans, 17. Nukumi, N., Iwamori, T., Naito, K., and Tojo, H. 2005. J. WAP is not synthesized in the mammary gland [18]. It Reprod. Dev. 51: 649–656. 18. Nukumi, N., Seki, M., Iwamori, T., Yada, T., Naito, K., and is unknown whether or not WAP is synthesized in car- Tojo, H. 2006. J. Reprod. Dev. 52: 315–320. nivore species. In this study, we have demonstrated the 19. Nukumi, N., Iwamori, T., Kano, K., Naito, K., and Tojo, H. presence of WAP in dog milk. In the future, we need to 2007. J. Cell Physiol. 213: 793–800. . , . , . , . , . , . investigate whether or not dog WAP, like mouse WAP, 20 Nukumi N , Iwamori T , Kano K , Naito K , and Tojo H 2007. Cancer Lett. 252: 65–74. is physiologically functional. 21. Rival, S., Attal, J., Delville-Girud, C., Yerle, M., Laffont, P., Rogel-Gaillard, C., and Houdebine, L.M. 2001. Gene Acknowledgment 267: 37–47. 22. Wiedow, O., Schroder, J.M., Gregory, H., Young, J.A., and Christophers, E. 1990. J. Biol. Chem. 265: 14791–14795. This study was supported in part by a Grant-in-Aid (No.16380197) to HT from the Ministry of Education, Culture, Sports, Science and Technology of Japan.