DEVELOPMENTAL GENETICS 12:299-307 (1991)

Developmental Regulation of the Sheep P-Lactoglobulin Gene in the Mammary Gland of Transgenic Mice S. HARRIS, M. McCLENAGHAN, J.P. SIMONS, S. ALI, AND A.J. CLARK AFRC Znstitute of Animal Physiology and Genetics Research, Edinburgh Research Station, Roslin, Midlothian, Scotland

ABSTRACT p-Lactoglobulin (BLG) is the most abundant whey protein in sheep milk but it is not present in mouse milk. We have previously shown that transgenic mice carrying the BLG gene express it specifically in the mammary gland and secrete BLG into milk at high concentrations. Here we demonstrate that BLG transcription is correctly initiated in mice and that BLG synthesis is restricted to the secretory epithelial cells of the mammary gland. We have also determined the temporal pattern of milk protein gene expression and find that the BLG transgene is regulated coordinately with mouse p-casein and that the patterns of regulation of BLG in mouse and sheep share some similarities. Key words: p-lactoglobulin, transgenic, mammary gland development, milk protein gene expression

INTRODUCTION The mammary gland is the specialised secretory organ that provides essential nourishment to mammalian young in the form of milk. Mammary gland development and milk protein gene expression are regulated by a number of peptide and steroid hormones, as well as cell-cell and cell-substratum interactions within the gland [Topper and Freeman, 1980; Levine and Stockdale, 1985; Li et al., 19871. The contribution of different hormones and morphological factors on mammary gland development and milk protein gene expression have previously been examined in the whole animal and in mammary explant and epithelial cell culture systems [reviewed by Topper and Freeman, 19801. A number of genomic clones encoding milk proteins have been described and DNA sequence comparisons have suggested the existence of common regulatory elements in the promoter region of several casein and whey protein genes [for refs. see Harris et al., 19901. Furthermore, a n in uitro DNA-protein interaction has been observed between part of this DNA sequence motif and nuclear protein(s) isolated from the mammary gland and other tissues [Lubon and Hennighausen,

0 1991 WILEY-LISS, INC.

1987,19881. However, a s yet, the in uiuo significance of these observations remains unclear. The ability to generate transgenic animals containing manipulated DNA sequences provides a n opportunity to investigate the regulation of milk protein gene expression under the influence of all contributory factors. We have chosen to investigate expression of the ovine p- lactoglobulin (BLG) gene in transgenic mice. BLG is a n 18 kd whey protein present in the milk of ruminants, horses, pigs, dogs, and dolphins, but not that of rodents and humans [for refs. see Ali and Clark, 19881. Ovine genomic DNA encompassing the BLG structural gene is capable of directing the expression of high levels of BLG protein to mouse milk [Simons et al., 19871. However, from the previous study it was not known how precisely the sheep gene was regulated in the mouse. In the present study we compare the transcriptional start-site of the BLG gene in mouse and sheep and define the cell type in which transgene expression occurs in the mouse mammary gland. We also compare the expression profile of the transgene during pregnancy and lactation to that of two endogenous milk protein genes, p-casein and whey acidic protein (WAP), and to the BLG expression profile in sheep.

MATERIALS AND METHODS 1. Animals The generation and maintenance of transgenic mice containing DNA fragments encompassing the ovine

Received for publication October 22, 1990; accepted February 12, 1991. Address reprint requests to Dr. A.J. Clark, AFRC Institute of Animal Physiology and Genetics Research, Edinburgh Research Station, Roslin, Midlothian EH25 9PS, Scotland.

S. Ali is now a t Inserm U184, CNRS LGME, Institute de Chimie Biologique, Faculte de Medecine, 11 Rue Humann, 67085-Strasbourg Cedex, France. S. Harris is now a t Glaxo Group Research Ltd., Greenford Road, Greenford, Middlesex UB6 OHE England.

300

HARRIS ET AL. 3’

5’ (SalI)

(SalI)

SphI

Animal

Copy n o / c e x

BLG level

protein m g h l Line 7 Line 14 Sheep

15-20

23. 1

RNA

3.3

1-2

5-0

0.75

2

4.6

1.0

Fig. 1. Organisation of the ovine beta-lactoglobulin gene. Schematic representation of the ovine BLG clone used to generate transgenic mice by pronuclear injection. The bold line represents 5’ and 3‘

flanking DNA and intronic sequences; the exons are depicted as solid boxes. Tabulated are a number of relevant features of the transgenic lines described here compared to sheep [data from Simons et al., 19871.

BLG gene has been described previously [Simons et al., water, using a 96 well manifold (BRL); total RNA sam19871. In the present study two of these lines of trans- ples were transferred in a final volume of 50 p1 congenic mice (7 and 14) were employed. Both contain a taining 2.2 M formaldehyde after heating to 65°C for 5 16.2 kb SalI fragment comprising the 4.9 kb BLG tran- min. Hybridisations were performed with oligolabeled scription unit, 4 kb of 5’ flanking sequences, and 7.7 kb probes [Feinberg and Vogelstein, 1983,19841 using the of 3’ flanking sequences (Fig. 1). The founder (GO) method of Church and Gilbert [1984]. After autoraditransgenic mice were F, (C57BL/6 x CBA), and the ography at -70°C using preflashed film and intensifylines were propagated by backcrossing to F, (C57BL/6 ing screen [Laskey, 19801 various exposures were x CBA) mice. In order to facilitate the breeding of scanned on a Schimadzu CS-9000 dual wavelength flylarge numbers of transgenic mice, we used mice ho- ing spot densitometer. S1 mapping was carried out usmozygous for the transgenes obtained by sib matings of ing the protocol described in Davis et al. [1986]; samG1 or G2 mice. Homozygous males were mated with F, ples were run on denaturing urea-polyacrylamide gels females to generate the hemizygous transgenic females before autoradiography with intensifying screen at used in the expression analysis; the transgenic status -70°C. of these animals was confirmed on tail DNA samples taken when sacrificed. Age-matched non-transgenic 3. Quantification of RNA DotlBlots control mice were C57BL/6 x CBA F,. Sheep material For each time point, duplicate samples of three difwas obtained from virgin and first pregnancy “gim- ferent dilutions of total RNA were transferred to nylon mer” ewes [Ali, 19891. membranes before hybridisation to individual 32P-labeled DNA probes (see Fig. 4). An integrated density 2. RNA Analysis value for each dilution, above background, was obTotal RNA was isolated from tissue by the guanidi- tained by densitometric scanning of autoradiographs. nium thiocyanate/CsCl gradient method [Chirgwin et For each probe values a t each dilution (in duplicate) al., 19791. Northern blots were performed on 10 pg of were combined. These values were corrected for variatotal RNA size-fractionated on 1.5% formaldehyde/ tions in RNA loading by normalising the ribosomal MOPS agarose gels before transfer in 20 x SSC to Hy- RNA values and then multiplying the value for indibond-N membrane (Amersham). Dot-blots were loaded vidual probes a t each time point by the resultant coronto Hybond-N membranes, pre-wetted with distilled rection factor. For each probe the level of RNA was

REGULATION OF P-LACTOGLOBULIN calculated relative to day 11 lactation. The sheep time course was established as described by Ali [1989].

A)

a G A

T C E w

4. DNA Probes

All DNA fragments used for probing blots were recovered from agarose gels using the DE81 method [Dretzen et al., 19811. The BLG probe was isolated as a 424 bp PstI fragment from the ovine P-lactoglobulin cDNA p931 [Gaye et al., 19861. Similarly, the murine p-casein probe was isolated as a 160 bp EcoRI-Hind111 fragment from pSMPC-10, a pSP64 recombinant containing the PstI-EcoRI fragment from the 6-casein cDNA clone pCas51 [Mehta et al., 19811. The murine WAP probe was isolated a s a 380 bp EcoRI-Hind111 fragment from pSP51-9, a pSP65 recombinant containing the 360 bp SalI-Hind111 fragment from pmWAPl [Hennighausen and Sippel, 19821. The -12 kb Xenopus ribosomal DNA probe was isolated as a Hind111 fragment from pXlrlOl [Sollner-Webb and Reeder, 19791. Plasmids pSMPC-10, pSP51-9, and pXlrlOl were the gifts of Dr. R. Lathe.

RESULTS Initiation of BLG Transcription BLG transcript initiation was investigated by S1 protection analysis (Fig. 2). RNA preparations from the mammary gland of sheep, control, and transgenic mice were hybridised with a 187 n t SphI-TaqI probe which encompasses the 5’ end of the BLG transcription unit (Fig. 2B). In sheep samples and in samples from the transgenic mouse lines, the major protected fragment was 136 nt. No protection of the probe was observed with RNA from control mouse mammary gland. This confirms the assignment of the major BLG transcription initiation site at position -40 made by primer extension [Gaye et al., 1986; Ali and Clark, 19881 and demonstrates that BLG transcription is initiated a t the same site in sheep and transgenic mice. Localisation of BLG Within the Mammary Gland To determine in which cell-type the BLG transgene was expressed, mammary gland sections were analysed by immunohistochemistry on tissue taken from 12-day pregnant females. Increased BLG and p-casein mRNAs are first apparent a t this time (see below) and small

Tg line

1 2 714

c*

5. Immunohistochemistry

Mammary tissue was fixed in a modification of Tellyesniczky’s fluid [Dils and Forsyth, 19811 and embedded in paraffin wax. Sections of tissue, 4 pm thick, were stained using the method described by Finlayson et al. [1985]. Briefly, tissue samples were serially reacted with rabbit-anti-ovine BLG (a gift of Dr. P. Gaye) and donkey-anti-rabbit Ig horseradish peroxidase linked F(ab‘)2 (Amersham), developed with 3,3’-diaminobenzidine (Sigma) and counterstained with haematoxylin.

Sheep

NTP

301

Sphl

*

Taql

187 bp probe 136 bp protected

fragment

Fig. 2. Correct initiation of transcription from the BLG promoter in transgenic mice. A Autoradiographic visualisation of the probe fragment protected from S1 digestion by BLG transcripts after electrophoresis on a denaturing polyacrylamide gel. Total RNA samples were analysed from mammary tissue of 11 day lactating transgenic mice (lines 7 and 14); a non-transgenic mouse (NT); and a lactating sheep (day 20; 1and 2). Also shown are the products of probe digestion in the presence of tRNA, and undigested probe (P). The sequence ladder (G,A,T,C) was used to determine the size of fragments. B Schematic representation of the region encompassing exon 1 of the ovine BLG gene. The bold line signifies 5’ flanking and intron sequences while translated and non-translated sequences of exon 1 are represented by shaded and open boxes, respectively. Below, the 136 bp product expected after S1 digestion of the 187 bp probe, if transgene transcription is initiated a t the previously described major BLG cap site [*, Gaye et al., 1986; Ali and Clark, 19881.

quantities of milk proteins have been found in secretory epithelial cells a t this stage [Vonderhaar and Bhattacharjee, 19851. When mammary gland tissue sections from transgenic mice are probed with antisera raised against ovine BLG, immuno-staining is clearly visible in the columnar secretory epithelial cells of the alveolus, particularly toward the luminal surface of these cells (Fig. 3). Staining of other cell types was not observed. The

302

HARRIS ET AL.

”

.

Non-transgenic Fig. 3. Localisation of BLG expression to the secretory epithelial cells of the mammary gland. Sections of mammary gland (from control and line 14)taken a t day 12 of gestation are shown after immunohistochemical staining with a n antibody raised against BLG and

-

.

-

Transgenic: line 14 counterstained with haematoxylin (see “Materials and Methods”). Within the lobulo-alveolar structures shown, columnar secretory-epithelial cells (S) can be distinguished from myoepithelial cells (MI.

lished data) and fluctuations in BLG RNA levels were also observed during the time-course in this line. It is possible that the transgenic locus in line 7 is particularly sensit,ive to the heterogeneous genetic background of these mice (see Materials and Methods), and this is currently being investigated. BLG transgene expression does not appear to grossly interfere with the time-course of induction of expression of endogenous milk protein genes as shown by the comparison to samples derived from control mice (Fig. 5c). It should be Regulation of @-Caseinand BLG During noted, however, that the control mice were C57BW6 x Mammary Gland Development in CBA F1, and thus genetically homogeneous, whereas Transgenic Mice the transgenic mice had a more heterogeneous genetic The developmental profiles of milk protein gene ex- background. Most of the induction of BLG and @-caseingene expression were established by determining the level of specific mRNA transcripts at various stages during pression occurs during the second half of pregnancy; pregrlancy and lactation. The profiles of the sheep BLG from day 10 to day 18 the relative proportions of BLG transgene and the endogenous mouse @-caseingene are and @-caseinRNA transcripts increase some 10-40very similar in transgenic mice (Figs. 4, 5a). Both fold. During this time the duct system increases in size genes are expressed at low levels in the virgin gland, due to a n approximately 4-fold increase in the proporand during the first 10 days of pregnancy there is a tion of epithelialhtromal cells as the alveolar strucsmall, gradual accumulation of RNA transcripts. tures are formed. At this stage there is also a n approxThereafter, there is a rapid accumulation of specific imately 6-fold increase of RNA in the mammary gland, mRNAs which continues until after parturition. From as estimated from the recovery of total RNA/g tissue day 10 until parturition the levels of expression in- (Fig. 5b) [see also Denamur, 19741. Mammary gland crease from approximately 5% to 65-80% of the levels epithelial cells divide with a doubling time of about 6 a t mid-lactation. Following parturition there is a fur- days during this period [Knight and Peaker, 19821, so ther 1.5-2-fold increase in the relative levels of expres- there must be a substantial increase in the RNA consion. Toward the end of lactation and immediately af- tent of each cell. From parturition to mid-lactation there is a 1.5- to ter weaning expression levels decrease precipitously. Similar expression profiles of WAP and @-caseinare 2-fold increase in the proportion of BLG and @-casein obtained with the two lines of transgenic mice; how- transcripts. This is accompanied by a 1.5-to 2-fold furever, there are ’significant variations in BLG protein ther increase in the amount of total RNA/g tissue (Fig. levels in milk collected a t mid-lactation from animals 5b). In the mouse, mammary cell proliferation continof line 7 (results not shown; M. McClenaghan, unpub- ues into lactation; however, there is no major induction

specificity of the BLG antiserum is demonstrated by the absence of any immuno-staining on mammary gland tissue sections from control non-transgenic mice processed in a n identical manner on the same slide. Thus BLG expression is tightly regulated within the mammary gland, being restricted to the columnar secretory epithelial cells which are known to synthesise and secrete the major milk proteins [see Mercier and Gaye, 19831.

REGULATION OF P-LACTOGLOBULIN

V

9-2 9-3 9-5 9-7 g-10 g-11 g-12

V

9-2 9-3 9-5 9-7 g-10 g-11 g-12

9-14 9-16 b 1-11 b cs

cm b

g-I 4 g-16 b 1-1 1 b cs cm b

v 9-2 9-3 9-5 9-7 g-10 g-11 g-12

V

9-2 9-3 9-5 9-7 g-10 g-11 g-12

303

9-14 g-16 b 1-1 1 b cs cm b

g-14 g-16 b 1-1 1 b cs cm b

Fig. 4. Gene expression during gestation in transgenic mice. Dot blots of total RNA isolated from mammary gland tissue after hybridisation to 32P-labeled ovine BLG (A), murine p-casein (B), murine WAP (C) and Xenopus ribosomal (D) probes. Shown are duplicate samples (from left to right, 1.0, 0.5, and 0.25 pg) of mammary gland total RNA from a mature virgin (v) and individual females sacrificed

at days 2,3,5,7,10,11,12,14, and 16 gestation (g-2 to g-16), respectively. Detection of the post-copulation plug was taken to represent day 0 gestation. Control sheep (cs), control mouse (C57BL/6; cm), and transgenic (line 7) day 11 lactation (1-11)RNA samples are diluted 10-fold compared to others; b = blank.

of BLG and p-casein gene expression at this time, the high levels induced during the latter half of pregnancy being sustained during lactation. The overall increase in the proportion of BLG or endogenous p-casein transcripts represents a substantial induction of the expression of these genes on a per cell basis, at least 150300-fold.

the relative proportions of BLG and p-casein transcripts are about 45% of those at mid-lactation. Interestingly, this is the time (after sterile mating) a t which pseudopregnancy ends. At this stage in pregnant females, the concentration of prolactin has fallen to that in non-pregnant females, and placental lactogen concentrations increase dramatically [Thordarson and Talamantes, 19871. Thus the onset and induction of WAP expression may be critically dependent on, and/or require significantly higher levels of, this mammotrophic hormone than are required by BLG and p-casein.

Regulation of WAP During Mammary Development In contrast to both endogenous p-casein and BLG, which are expressed at low levels in the gland of virgins and then rapidly induced from day 10 of gestation, the expression of WAP was not detected in the virgin gland and the onset of expression was delayed until day 14 of pregnancy (Figs. 4, 5a) [see also Pittius et al., 19881. Thereafter expression is rapidly induced, with about the same rate of increase observed for BLG and p-casein. Consequently, a significantly greater proportion of WAP induction occurs during lactation than for the other two genes. By the time WAP is first detected

Developmental Regulation of BLG in Sheep In sheep, mammary growth is slow until 3 months after the establishment of pregnancy; thereafter rapid growth occurs but, in contrast to mice, there is little cell proliferation after parturition [Anderson, 19751. The expression profile of BLG in transgenic mice was compared to that in sheep after compensating for the different lengths of pregnancy in the two species (Fig. 5d). During gestation the expression profile for BLG is

5c)

a, U

3

L

.->

m

w

X

v)

.-5 ln Ea

a, A

5

00

02

04

06

08

10

1 2

.l

0O t

0 2-

04‘

06-

08.

1 0.

12-

Lactation cycle (days)

-

Lactation cycle (days)

Fig. 5 .

1

0

l

2

S

4

I

I

6

8

1

I

r

l

r

r

1

1

Lactation cycle

10 12 14 1618202224

1

CornDarison of SheeD & mouse BLG

Lactation cycle (days)

.

.

.

.

.

. 35 3 7 3 9 41 4 3 45

REGULATION OF P-LACTOGLOBULIN

305

enous mouse milk protein gene, p-casein. Thus the BLG transgene used in these experiments includes the cis-acting DNA sequences required for the correct cellspecific and developmental regulation of this gene in mice. Furthermore, mice retain the ability to interpret these sequences, despite the absence of an equivalent gene in this species. By contrast, the endogenous pcasein and WAP genes are expressed in a non-coordinate manner in transgenic and control mice, both in terms of basal level of expression and the point at which RNA first accumulates during gestation. In rodents, the non-coordinate expression of these and other milk proteins has been noted previously [Nakhasi and Qasba, 1979; Vonderhaar and Nakhasi, 19861, reflectDISCUSSION ing subtle differences in the regulation of specific We have previously shown that expression of the genes. The mammary gland is a target tissue for many difgene encoding BLG in transgenic mice can result in a significant alteration in the composition of milk [Si- ferent hormones which act synergistically to ensure mons et al., 19871. Recently it has also been shown that that milk production coincides with the birth and nurthe bovine alpha-lactalbumin and rat WAP genes can turing of the young. The similarity between the BLG be expressed specifically and at significant levels in expression profile in sheep and transgenic mice during transgenic mice [Vilotte et al., 1989; Bayna and Rosen, pregnancy suggests that the endocrine control of mam19901. In contrast, transgenic mice containing the rat mary gland development and milk protein gene expresgenomic p-casein gene express the transgene at only sion are similar in these two species during this period 0.01-0.1% of the level of the endogenous murine p- (Fig. 5d). An increase in BLG expression occurs at an casein gene [Lee et al., 19881. This modest level of ex- equivalent stage in both animals and this is correlated pression presumably reflects either the absence from with an increase in cell number, RNA content, and the the transgene of an important cis-acting regulatory el- first appearance of milk proteins [Denamur, 1974; ement(s) required for efficient expression in the mouse Shuster et al., 19761. Both an increase in transcription andior differences between the transcriptional machin- rate and an increase in mRNA stability have been invoked as mechanisms underlying the induction of milk ery of these two species. protein gene expression [Rosen et al., 19751. We report here further characterization of transBefore conception the mammary gland consists of a genic mice containing the ovine BLG gene. We demonstrate that transcription is correctly initiated within partly developed duct system lying in a n extensive fat the promoter of the transgene and that expression is pad. It has been estimated that in the virgin gland only restricted to the epithelial cells of the mammary gland about 25% of the cells are epithelial [Nicoll and Tucker, involved in milk protein secretion. Transgene expres- 19651. In the virgin gland and during early pregnancy sion is coordinately regulated with at least one endog- the low level expression of both p-casein and BLG can be accounted for by the low proportion of active secretory epithelial cells that are present in the mammary gland. Considerable morphological changes occur to mammary epithelia during pregnancy and lactation. Fig. 5. Shown sideturned on facing page. Milk protein gene expresMost notable in the mouse is a major increase in cell sion during a lactation cycle. a: Relative levels of BLG, B-casein, and WAP RNA expression during the first lactation; levels are compared volume (approximately 5-fold) which commences durto lactation day 11 = 1. Detection of the post-copulatory plug defined ing late pregnancy and continues well into lactation day 0, mature virgin = -1, parturition = 19, and weaning = 40, [Foster, 19771. This presumably represents an expanrespectively. The gestation period is indicated by stippling. Data for sion by each epithelial cell of its capacity to synthesise hemizygous females offspring of males 14.24.a4 and 14.24.b4. At each time point expression levels were determined for a female offspring and secrete milk components. For the most part, this from each male. The curves are drawn between the averages of these increase occurs after the induction of BLG and p-casein values. b: Yield of total RNA per gram wet weight mammary tissue. gene expression. Thus the increase in milk protein proThe means and standard errors were estimated from the line 14, line duction that occurs at lactation is probably determined 7, and control mice used in this study. c: Relative levels of murine p-casein and WAP in mammary gland RNA isolated from control by changes other than the further induction of milk mice. In this case expression levels were determined for a single protein mRNAs. mouse a t each time-point. d Comparison of BLG RNA levels in mamOne difference between mice and sheep is that a mary gland RNA isolated from sheep and transgenic mice (average greater proportion of the induction of BLG expression values from line 14). Expression levels are depicted relative to day 20 takes place at parturition and during lactation in lactation (sheep) and day 11 lactation (transgenic mice). Time points sheep. This observation correlates with other indices of for mice as above. For sheep, the time-scale in weeks is shown below differentiation showing that by parturition mouse that for mice in days.

similar in the two species, with the possible exception of an early transient rise in BLG RNA levels in sheep. This transient increase in sheep may reflect a similar transient increase in the circulating level of prolactin following conception [Ali, 19891. In sheep, as in transgenic mice, low level BLG expression was detected in virgin animals and an increase in expression takes place during the second half of pregnancy. In contrast to the mouse, a considerably greater proportion of BLG induction takes place at parturition and during lactation. A similar increase in the levels of expression of other sheep milk protein genes has also been observed during this period [Ali, 19891.

306

HARRIS ET AL.

mammary cells are more completely differentiated than are those from ruminant species [e.g., see Wilde et al., 1986; Shipman et al., 19871. In transgenic mice the pattern of BLG expression would appear to be determined by the changing status of differentiation of the mouse cells rather than the origin of the gene. Finally, it has been proposed that manipulation of milk composition is a candidate for genetic manipulation in domestic dairy animals [Lathe et al., 1986; Clark et al., 1987; Jimenez-Flores and Richardson, 19881. We are investigating the use of transgenic sheep containing BLG-fusion genes as a source of authentic human protein in milk [Simons et al., 1988; Clark et al., 19891. In the absence of a suitable cell system in which to study milk protein gene expression transgenic mice will prove invaluable as a relatively rapid means of evaluating the expression of further BLG-hybrid constructs for their potential to direct the expression of foreign protein(s) into the milk of domestic ruminants [Archibald et al., 19901.

Feinberg AP, Vogelstein B (1983): A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6-13. Feinberg AP, Vogelstein B (1984): Addendum. A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 137:266,267. Finlayson J , Buxton D, Anderson IE, Donald KM (1985): Direct immunoperoxidase method for demonstrating Chlamydia psittaci in tissue sections. J Clin Pathol 38712-714. Foster RC (1977): Changes in mouse mammary epithelial size during mammary gland development. Cell Differ 6:l-8. Gaye P, Hue-Delahaie D, Mercier J-C, Soulier S, Vilotte J-L, Furet J-P (1986): Ovine p-lactoglobulin messenger RNA: nucleotide sequence and mRNA levels during functional differentiation of the mammary gland. Biochimie 68:1097-1107. Harris S , McClenaghan M, Simons J P , Ali S, Clark AJ (1990): Gene expression in the mammary gland. J Reprod Fertil 88:707-715. Hennighausen LG, Sippel AE (1982): Mouse whey acidic protein is a novel member of the family of ‘four-disulphide core’ proteins. Nucleic Acids Res 10:2677-2684. Jimenez-Flores R, Richardson T (1988): Genetic engineering of the caseins to modify behaviour of milk during processing: a review. J Dairy Sci 71:2640-2654. Knight CH, Peaker M (1982): Development of the mammary gland. J Reprod Fertil65:521-536. Laskey RA (1980): The use of intensifying screens or organic scintilACKNOWLEDGMENTS lators for visualizing radioactive molecules resolved by gel electroWe thank Mrs. S. Couperwhite for her technical asphoresis. Methods Enzymol 65363-371. sistance and those involved in generating and main- Lathe R, Clark AJ, Archibald AL, Bishop JO, Simons J P , Wilmut I (1986): Novel products from livestock. In Smith C, King JWB, taining mice, in particular Miss R. Wallace. This work McKay J C (eds): “Exploiting New Technologies in Animal Breedwas supported by the AFRC. ing: Genetic Developments.” Oxford: Oxford University Press, pp 91-102. REFERENCES Lee K-F, DeMayo FJ, Atiee SH, Rosen J M (1988):Tissue-specific expression of the rat p-casein gene in transgenic mice. Nucleic Acids Ali S (1989): Structure and expression of the gene encoding ovine Res 161027-1041. p-lactoglobulin. PhD thesis: University of Edinburgh, Scotland. Ali S, Clark J (1988): Characterisation of the gene encoding ovine Levine JF, Stockdale FE (1985): Cell-cell interactions promote mammary epithelial cell differentiation. J Cell Biol 100:1415-1422. p-lactoglobulin: similarity to the genes for retinol binding protein Li ML, Aggeler J , Farson DA, Hatier C, Hassell J, Bissel MJ (1987): and other secretory proteins. J Mol Biol 199:415-426. Influence of a reconstituted basement membrane and its compoAnderson RR (1975): Mammary gland growth in sheep. J Anim Sci nents on casein expression and secretion in mouse mammary epi41: 118-123. thelial cells. Proc Natl Acad Sci USA 84:136-140. Archibald AL, McClenaghan M, Hornsey V, Simons J P , Clark AJ (1990):High-level expression of biologically active human al-anti- Lubon H, Hennighausen L (1987): Nuclear proteins from lactating trypsin in the milk of transgenic mice. Proc Natl Acad Sci USA mammary glands bind to the promoter of a milk protein gene. Nu875178-5182. cleic Acids Res 152103-2121. Bayna EM, Rosen J M (1990): Tissue-specific high level expression of Lubon H, Hennighausen L (1988): Conserved region of the rat athe rat whey acidic protein gene in transgenic mice. Nucleic Acids lactalbumin promoter is a target site for protein binding in-uitro. Res 18:2977-2985. Biochem J 256:391-396. Chirgwin JM, Pryzbyla AE, MacDonald RJ, Rutter WJ (1979): Isola- Mehta NM, El-Gewely RM, Joshi J, Helling RB, Benejee MH (1981): tion of biologically active ribonucleic acid from sources enriched in Cloning mouse p-casein gene sequences. Gene 15285-288. ribonuclease. Biochemistry 185294-5299. Mercier J-C, Gaye P (1983): Milk protein synthesis. In Mepham TB Church GM, Gilbert W (1984): Genomic sequencing. Proc Natl Acad (ed): “Biochemistry of Lactation.” New York: Elsevier, pp 177-227. Sci USA 81:1991-1995. Nakhasi H-L, Qasba PK (1979): Quantitation of milk proteins and Clark AJ, Simons J P , Wilmut I, Lathe R (1987): Pharmaceuticals their mRNAs in rat mammary gland at various stages of gestation from transgenic livestock. Tibtech 5:20-24. and lactation. J Biol Chem 254:6016-6025. Clark AJ, Bessos H, Bishop JO, Brown P, Harris S, Lathe R, McClenaNicoll CS, Tucker HA (1965): Estimates of parenchymal, stromal and ghan M, Prowse C, Simons JP, Whitelaw CBA, Wilmut I (1989): lymph node DNA in mammary glands of C3HiCgrli2 mice. Life Sci Expression of human anti-hemophilic factor IX in the milk of trans4:993-1001. genic sheep. BioiTechnology 7:487-492. Davis LG, Dibner MD, Battey JF (1986): “Basic Methods in Molecular Pittius CW, Sankaran L, Topper YJ, Hennighausen L (1988): Comparison of the regulation of the whey acidic protein gene with that Biology.” New York: Elsevier. of a hybrid gene containing the whey acidic protein gene promoter Denamur R (1974): Ribonucleic acids and ribonucleoprotein particles in transgenic mice. Mol Endocrinol 2:1027-1032. of the mammary gland. In Larsen BL, Smith VR (eds): “The MamRosen JM, Woo SLC, Comstock J P (1975): Regulation of casein mesmary Gland, Development and Maintenance.” London: Academic senger RNA during development of the mammary gland. BiochemPress, pp 413-465. istry 14:2895-2903. Dils R, Forsyth IA (1981):Preparation and culture of mammary gland Shipman LJ,Docherty AH, Knight CH, Wilde CJ (1987): Metabolic explants. Methods Enzymol 72:724-742. adaptations in mouse mammary gland during a normal lactation Dretzen G, Bellard M, Sassone-Corsi P, Chambon P (1981): A reliable cycle and in extended lactation. Q J Exp Physiol 72:303-311. method for the recovery of DNA fragments from agarose and acryShuster RC, Houdebine LM, Gaye P (1976): Studies on the synthesis lamide gels. Anal Biochem 112:295-298.

REGULATION OF P-LACTOGLOBULIN of casein messenger RNA during pregnancy in the rabbit. Eur J Biochem 71:193-199. Simons JP, McClenaghan M, Clark AJ (1987):Alteration of the quality of milk by expression of sheep p-lactoglobulin in transgenic mice. Nature 328530-532. Simons JP, Wilmut I, Clark AJ, Archibald AL, Bishop JO, Lathe R (1988): Gene transfer into sheep. Biotechnology 6:179-183. Sollner-Webb B, Reeder RH (1979): The nucleotide sequences of the initiation and termination sites for ribosomal RNA transcription in X . laeuis. Cell 38:485-499. Thordarson G, Talamantes F (1987):Role of the placenta in mammary gland development and function. In Neville MC, Daniel CW (eds): “The Mammary Gland.” New York Plenum, pp 459-498. Topper YJ, Freeman CS (1980): Multiple hormone interactions in the

307

developmental biology of the mammary gland. Physiol Rev 60: 1049-1 106. Vilotte J-L, Soulier S, Stinnakre M-G, Massoud M, Mercier J-C (1989): Efficient tissue-specific expression of bovine a-lactalbumin in transgenic mice. Eur J Biochem 187:43-48. Vonderhaar BK, Bhattacharjee M (1985): The mammary gland: a model for hormonal control of differentiation and preneoplasia. In Mihich E (ed): “Biological Responses in Cancer,” Volume 4. New York: Plenum Press, pp 125-159. Vonderhaar BK, Nakhasi HL (1986): Bifunctional activity of EGF on a- and K-casein gene expression in rodent mammary gland in uitror Endocrinology 119:1178-1184. Wilde CJ, Henderson AJ, Knight CH (1986):Metabolic adaptations in goat mammary tissue during pregnancy and lactation. J Reprod Fertil 76289-298.