Production of pharmaceutical proteins by transgenic animals.

Proteins are involved in a majority of the biochemical events that take place in all living organisms. Protein synthesis is directed by genes. All genes contain two major DNA regions. The region containing the genetic message proper (the 'coding region') is preceded by a regulatory region ('the promoter'), which determines when and in which organs a given gene must produce the corresponding protein. The techniques of genetic engineering allow the association of the coding region from one gene with the regulatory region from another gene. The expression of these recombinant genes may be achieved in cultured cells, in transgenic animals or in plants. This leads to the production of the corresponding proteins, including pharmaceutical proteins. Milk from transgenic animals is one potential source of pharmaceutical proteins. To achieve this, the promoters from milk protein genes are bound to DNA fragments containing the coding region of the genes of interest. The desired proteins are then taken from the milk and purified. Two human pharmaceutical proteins are on the market and about 20 projects are in development. One of the proteins produced in milk, antithrombin III, is an anticoagulant and the other, human C1-esterase inhibitor, is an anti-inflammatory. Several human proteins have been produced in the egg white of transgenic chickens and one has been approved by the United States Federal Drug Administration. This process has also been used to modify antibodies in cows. The genes that code for antibodies in the cow were deleted and replaced by human antibody genes. These cows, immunised by various antigens, then secreted purely human antibodies in their blood. Antibodies from such cows were able to attenuate the effects of Ebola virus in human patients.

Les protéines sont des acteurs moléculaires majeurs des organismes vivants. Leur synthèse résulte du décodage des gènes. Essentiellement, tous les gènes contiennent deux régions d'ADN majeures. La région contenant le message génétique proprement dit (région codante) est précédée d'une région de régulation (le promoteur) qui détermine quand et dans quels organes un gène donné doit produire la protéine correspondante. Les techniques de génie génétique permettent l'association de la région codante d'un gène donné à une région régulatrice d'un autre gène. L'expression des gènes recombinants peut être réalisée dans des cellules cultivées, dans des animaux ou des plantes transgéniques. Ceci conduit à la production des protéines correspondantes comprenant des protéines pharmaceutiques. Le lait d'animaux transgéniques a été retenu comme l'une des sources potentielles de protéines pharmaceutiques. Pour atteindre cet objectif, les promoteurs des gènes des protéines du lait sont liés à des fragments d'ADN contenant les gènes d'intérêt. Les protéines sont ainsi retrouvées dans le lait et purifiées. Deux protéines pharmaceutiques humaines sont sur le marché et une vingtaine de projets sont en cours de développement. L'une des protéines produites dans le lait, l'antithrombine III, est un anticoagulant, tandis que l'autre, l'inhibiteur de la C1-estérase humaine, est un anti-inflammatoire. Plusieurs protéines humaines ont été produites dans du blanc d'oeuf de poulet transgénique et l'une d'entre elles a été approuvée par l'Agence américaine des produits alimentaires et médicamenteux. Cette méthode a également été utilisée pour modifier l'expression génétique des anticorps chez des vaches. Des vaches dont les gènes codant les anticorps ont été inactivés et remplacés par les gènes d'anticorps humains ont été obtenues. Ces vaches immunisées par divers antigènes sécrètent dans leur sang des anticorps purement humains. Certains anticorps de telles vaches ont été capables d'atténuer les effets du virus Ebola chez les patients.

Las proteínas participan en la mayoría de las reacciones bioquímicas que tienen lugar en todos los organismos vivos. La síntesis de proteínas está regida por los genes. Todos los genes contienen dos regiones principales de ADN. La región que contiene el mensaje genético propiamente dicho («región codificante¼) viene precedida de una región reguladora («promotora¼), que determina cuándo y en qué órganos un determinado gen debe expresar la correspondiente proteína. Las técnicas de ingeniería genética permiten asociar entre sí la región codificante de un gen y la región reguladora de otro gen. Estos genes recombinantes pueden ser llevados a expresarse en cultivos celulares, animales transgénicos o plantas, lo que resulta en la producción de las correspondientes proteínas, entre ellas proteínas farmacéuticas. La leche de animales transgénicos es una posible fuente de proteínas farmacéuticas. Para obtenerlas se unen las secuencias promotoras de los genes de proteínas lácteas con fragmentos de ADN que contienen la región codificante de los genes que interesan. Después se extraen de la leche y se purifican las proteínas deseadas. Ahora mismo hay dos proteínas farmacéuticas humanas que ya están comercializadas, además de unos 20 proyectos en curso. Una de las proteínas fabricadas en leche, la antitrombina III, es un anticoagulante, y la otra, inhibidora de la C1 esterasa humana, un antiinflamatorio. Varias proteínas humanas, una de ellas ya aprobada por la Administración de Alimentos y Medicamentos de los Estados Unidos, han sido producidas en la clara de huevo de gallinas transgénicas. Este proceso también ha sido utilizado para modificar anticuerpos en vacas. Los genes que codifican los anticuerpos de la vaca son suprimidos y sustituidos por genes de anticuerpos humanos. Estas vacas, expuestas a varios antígenos, secretan y envían al torrente sanguíneo anticuerpos humanos puros. Con los anticuerpos obtenidos a partir de tales vacas se pudieron atenuar los efectos del virus Ebola en pacientes humanos.

Prolactin-like activity of antiprolactin receptor antibodies in rat mammary tumor explants.

The effects of prolactin and a serum containing antiprolactin receptor antibodies on some lactogenic and mammogenic responses were investigated in nitrosomethylurea-induced mammary tumors in organ cultures. Prolactin was able to induce an increase in lactose synthetase activity, DNA synthesis, and prolactin-binding sites at moderate (i.e., physiological) concentrations of prolactin; at higher concentrations, desensitization of the tissues was observed for DNA synthesis and lactose synthetase activity, whereas the "down-regulation" of prolactin receptors occurred. Prolactin had no effect on casein synthesis. Antiprolactin receptor serum was capable of inducing an increase in lactose synthetase activity, DNA synthesis, and prolactin-binding sites, however, lower than that observed with prolactin, thus mimicking prolactin action. The antiserum did not induce any change in casein synthesis. These findings suggest that, in rat mammary tumors, as it has been observed in the normal mammary gland, the prolactin molecule is not required beyond the initial binding to its receptor for its action to be attained. It appears also that a differentiated function has lost its hormonal dependence in these tumors, although their hormonal dependence for other functions and growth is maintained.

Role of prolactin and glucocorticoids in the expression of casein genes in rabbit mammary gland organ culture. Quantification of casein mRNA.

Milk synthesis is initiated solely by prolactin in the pseudopregnant rabbit and glucocorticoids potentiate this action of prolactin. In organ culture, prolactin, in the presence or in the absence of insulin, enhances casein synthesis and cortisol (inactive alone) amplifies this action. Measurements of casein mRNA concentration in total cellular RNA, by hybridization with DNA complementary to casein mRNA, revealed that the stimulation of casein synthesis by the glucocorticoid is accompanied by an increase in the amount of casein mRNA. A systematic comparison of variations of these two parameters indicated that the major effect of glucocorticoids on lactogenesis in the rabbit at this stage of mammary gland development is mediated through an increase in the quantity of casein mRNA available for translation. No simultaneous control of casein mRNA translation by cortisol was observed.

Cloning, structural organization and tissue-specific expression of the rabbit transferrin gene.

We cloned the rabbit transferrin (rTf) cDNA and gene, and quantified the expression of the rTf gene at the RNA level in various organs. The tissue-specific pattern of expression of rTf gene is different to those in other species, with a high expression in mammary gland and kidney. The exon/intron structure of the rTf gene (17 exons/16 introns) is similar to those of transferrins from other species. The sequence of the rTf cDNA already published is corrected and lengthened in the 5' region, and a likely polymorphism is documented.

Beneficial effects of fibrates on apolipoprotein A-I metabolism occur independently of any peroxisome proliferative response.

BACKGROUND: In humans, fibrates are frequently used normolipidemic drugs. Fibrates act by regulating genes involved in lipoprotein metabolism via activation of the peroxisome proliferator-activated receptor-alpha (PPARalpha) in liver. In rodents, however, fibrates induce a peroxisome proliferation, leading to hepatomegaly and possibly hepatocarcinogenesis. Although this peroxisome proliferative response appears not to occur in humans, it remains controversial whether the beneficial effects of fibrates on lipoprotein metabolism can occur dissociated from such undesirable peroxisomal response. Here, we assessed the influence of fenofibrate on lipoprotein metabolism and peroxisome proliferation in the rabbit, an animal that, contrary to rodents and similar to humans, is less sensitive to peroxisome proliferators. METHODS AND RESULTS: First, we demonstrate that in normal rabbits, fenofibrate given at a high dose for 2 weeks does not influence serum concentrations or intestinal mRNA levels of the HDL apolipoprotein apoA-I. Therefore, the study was continued with human apoA-I transgenic rabbits that overexpress the human apoA-I gene under control of its homologous promoter, including its PPAR-response elements. In these animals, fenofibrate increases serum human apoA-I concentrations via an increased expression of the human apoA-I gene in liver. Interestingly, liver weight or mRNA levels and activity of fatty acyl-CoA oxidase, a rate-limiting and marker enzyme of peroxisomal beta-oxidation, remain unchanged after fenofibrate. CONCLUSIONS: Expression of the human apoA-I transgene in rabbit liver suffices to confer fibrate-mediated induction of serum apoA-I. Furthermore, these data provide in vivo evidence that the beneficial effects of fibrates on lipoprotein metabolism occur mechanistically dissociated from any deleterious activity on peroxisome proliferation and possibly hepatocarcinogenesis.

A combination of distal and proximal regions is required for efficient prolactin regulation of transfected rabbit alpha s1-casein chloramphenicol acetyltransferase constructs.

In the rabbit, alpha s1-casein is the major casein secreted in the milk. Transcription of the alpha s1-casein gene is induced by PRL. To define the positions of the cis-sequences involved in the control of rabbit alpha s1-casein gene expression by PRL, chimeric genes containing upstream regions of alpha s1-casein gene linked to the chloramphenicol acetyltransferase gene were cotransfected into Chinese hamster ovary cells with the plasmid expressing the rabbit mammary PRL receptor. It was observed that a distal fragment -3442/-3118 was responsible for a high induction of PRL sensitivity when linked in the 5'-position to a chimeric construct (-391/1774)-chloramphenicol acetyltransferase. A cooperation between distal and proximal regions of the alpha s1-casein gene is responsible for the PRL-dependent enhancer activity of the distal fragment. The mammary gland-specific nuclear factor-like binding sequence found around position -90 in the proximal promoter of the alpha s1-casein gene is involved in this cooperation. The distal fragment was further studied to determine the position of regulatory regions. A -3442/-3385 fragment was sufficient to induce a PRL sensitivity similar to that conferred by the larger -3442/-3118 distal fragment, but multiple interactions are likely to exist between other regulatory regions included in this distal fragment. Four DNA-binding regions (I-IV) have been identified within the reduced -3442/-3385 fragment by footprint experiments using rabbit mammary gland or liver nuclear extracts (NE). Protected area III is observed using both NE. Protected areas I, II, and IV are specific for lactating mammary gland NE. The sequences of areas I and IV share several homologies with the sequence of the mammary gland-specific nuclear factor-binding site.

[Confinement and consumption of cloned and transgenic animals]. / Confinement et consommation des animaux clonés et transgéniques.

Reproduction by cloning can eliminate some of the problems inherent to sexual reproduction, but it creates others. The genetic heritage of nucleus donor cells and the genetic status of clones are not precisely known. Furthermore, reprogramming of the genome of nucleus donor cells by the ovocyte cytoplasm is often incomplete. Animals obtained through cloning are thus essentially genetically identical to their genitors, but they are often epigenetically modified, with unpredictable effects. Transgenesis results in most cases from the addition to a genome of one or more known genes. The direct and indirect effects of transgenesis cannot all be predicted. Specific confinement measures make it possible to raise animals in high-security conditions, preventing their dissemination in the human food chain, in animal feed or in the environment. The toxicity, allergenicity and infectiousness of cloned ortransgenic animals can be evaluated by means of tests.

Effect of various lysosomotropic agents and microtubule disrupting drugs on the lactogenic and the mammogenic action of prolactin.

Various drugs added to the culture medium of rabbit mammary gland were assayed for their capacity to affect the lactogenic and the mammogenic activities of prolactin. Three lysosomotropic agents NH4Cl, chloroquine and methylamine which were previously demonstrated to inhibit the degradation of the hormone-receptor complex after its internalization (down-regulation) did not prevent the initiation of casein synthesis, of lactose synthetase activity and of DNA synthesis. Five microtubule disrupting drugs, colchicine, colcemid, vinblastin, podophyllotoxin and nocodazole inhibited the induction of casein and DNA synthesis by prolactin whereas two inactive analogues, trimethylcolchicinic acid and lumicolchicine had no effect. None of these drugs exhibited any general cytotoxic effect as judged by the capacity of the tissue to incorporate 14C aminoacids into total proteins and 3H-uridine into total RNA. The microtubule disrupting drugs did not greatly reduce the rate of casein synthesis in the cultured mammary tissue explanted from lactating rabbits. The data suggest that the down-regulation of prolactin receptor is not strictly required for the two considered prolactin activities. By contrast, the integrity of microtubules, or at least of structures in which tubulin is involved, is necessary to ensure a normal transmission of the prolactin information responsible for the initiation of milk and DNA synthesis. In addition, the fact that the lactogenic and the mammogenic activities of prolactin are affected by the same drugs suggests that these two properties of the hormone are mediated by cellular mechanisms which have at least one common step.

[Prolactin internalisation by the epithelial mammary cell: effects of lysosomotropic agents and transglutaminase inhibitors]. / Endocytose de la prolactine dans la cellule épithéliale mammaire: effets des agents lysosomotropes, et des inhibiteurs de la transglutaminase.

Prolactin endocytosis was studied by electron microscopy with 125I-prolactin 125I-hGH (human growth hormone) and prolactin-ferritin. Endocytosis and intracellular transit of the labelled hormone proceeded identically in epithelial cells isolated from the mammary glands of pseudopregnant rabbits and in surviving fragments from mammary glands of lactating rabbits. After binding of the hormone to its receptor, the labelled material was rapidly detectable in vesicles showing an homogeneous aspect; 15 min later part of the labelled material was still localized within the same kind of vesicles, but in addition it appeared to have migrated into microvesicles of the Golgi region and into vesicles of heterogeneous aspect tentatively identified with lysosomes. Endocytosis of bovine serum albumin, labelled with ferritin followed the same intracellular pathway. Native ferritin accumulated in vesicles of various sizes, but seemed excluded from the microvesicles of the Golgi zone. In the presence of lysosomotropic agents labelled prolactin accumulated in cytoplasmic vesicles. In the presence of dansylcadaverine, endocytosis of the labelled material proceeded unimpaired. Conversely, in the presence of bacitracin, the internalisation of labelled prolactin seemed to be reduced. These observations show that the endocytosis of the hormone/receptor complex is linked to membrane movements, which eventually lead to its location within both the Golgi apparatus and the lysosomes.

The effects of various kinase and phosphatase inhibitors on the transmission of the prolactin and extracellular matrix signals to rabbit alpha S1-casein and transferrin genes.

In all species, milk protein genes are specifically expressed in the mammary gland under the control of lactogenic hormones and extracellular matrix. In rabbit, casein gene expression is induced by prolactin alone and this induction is amplified by extracellular matrix. Transferrin gene expression is induced by extracellular matrix in the absence of hormones. The transduction mechanisms of prolactin and extracellular matrix to milk protein genes is only partly known. The present study has been undertaken to determine if protein kinases and phosphatases are involved in these mechanisms. Rabbit primary mammary cells were cultured in three different conditions (i) directly on floating collagen I, (ii) on plastic after a trypsinization to remove endogenous extracellular matrix, and (iii) on floating collagen I after a trypsinization to restore a functional extracellular matrix. In these culture conditions, prolactin and several protein kinase and phosphatase inhibitors were added to the medium. The expression of alpha S1-casein and transferrin genes was evaluated using Northern blotting analysis. In cells cultured directly on collagen I, staurosporine, quercetin and 6-dimethylaminopurine strongly inhibited prolactin action of alpha S1-casein gene whereas herbimycin A was only partly inhibitory. An erbstatin analogue, tyrosine phosphate, 1(5 isoquinolylsulphonyl) 2-methylpiperazine and GF 109 203 X did not alter prolactin action. The inhibitors which inhibited prolactin action when cells were directly cultured on collagen I were also those which prevented the induction of alpha S1-casein gene expression when cells were cultured on plastic in the absence of extracellular matrix. The induction of transferrin gene by the extracellular matrix was inhibited slightly by quercetin. Okadaic acid, phenylarsine oxide and sodium pervanadate which inhibit Ser/Thr and Tyr phosphatase inhibitors were unable to mimic prolactin action on alpha S1-casein gene expression. On the contrary, these inhibitors prevented prolactin action. These data suggest that a cascade including protein kinases and phosphatases for Ser/Thr and Tyr phosphate is involved in the transduction of the prolactin message from its receptor to casein genes. The signal delivered to the mammary cells by the extracellular matrix is quite different, possibly involving another cascade of protein kinases.

Correlation between prolactin-receptor interaction, down-regulation of receptors, and stimulation of casein and deoxyribonucleic acid biosynthesis in rabbit mammary gland explants.

Increasing concentrations of PRL were added to culture media of rabbit mammary gland explants. After 24 h of culture, free and total PRL receptors, casein synthesis, casein messenger (m) RNA concentrations and DNA synthesis were estimated. In the PRL concentration range of 0-100 ng/ml, receptors were progressively down-regulated by the hormone, but no occupied desaturable receptors could be detected. At concentrations between 1,000 and 20,000 ng/ml, the PRL receptors were increasingly occupied without being quantitatively down-regulated. There exists a reciprocal correlation between the rate of casein synthesis, casein mRNA concentration, DNA synthesis and the down-regulation of the receptors as a function of PRL concentration. The maximal responses were reached around a PRL concentration of 100 ng/ml. Interestingly, with a large excess of PRL a desensitization process was observed, the responses to the hormone being attenuated. This desensitization was more pronounced for DNA synthesis than for casein synthesis or for casein mRNA accumulation. The data suggest that down-regulation of PRL receptors occurs even at physiological concentrations of PRL. These observations are compatible with the hypothesis that each receptor can generate a limited number of intracellular relays eliciting hormonal action before being irreversibly inactivated and degraded. In addition, high concentrations of PRL induce a refractory state in the mammary gland that is possibly related to a low level of peripheral PRL receptors.

The effect of chloroquine on lysosomal prolactin receptors in rat liver.

PRL receptors have been previously identified in purified rat liver plasma membrane and Golgi vesicle preparations. In this study, we report on PRL receptors located in highly purified lysosome preparations. These lysosomal PRL receptors were characterized using Scatchard analysis and compared to other intracellular and cell surface receptors. We have identified two classes of lysosomes. Lighter lysosome-like vesicles, which are greatly enriched in acid phosphatase activity (the marker enzyme of lysosomes), contain a great deal of binding activity. This PRL binding was only slightly increased by pretreatment of animals with the lysosomotropic agent chloroquine. In contrast, mature lysosomes showed very little binding activity in control animals, but chloroquine treatment increased binding 7- to 8-fold in these mature lysosomes. We suggest that the lysosome-like structures are immature lysosomes (namely prelysosomes) toward which the hormone-receptor complex is internalized: they appear to bear little proteolytic activity. These structures could play a role in PRL receptor recycling. Lysosomal PRL receptors showed curvilinear Scatchard plots, in contrast to plasma membrane and Golgi counterparts, which were linear over the same range of hormone concentrations. The high affinity site in lysosomes had a Kd comparable to the cell surface and Golgi receptors. The number of binding sites per mg protein in prelysosomes and lysosomes was 3 times greater than that in the homogenate, but Golgi preparations were 3 times as rich as lysosomes. The great number of PRL receptors in prelysosomes could be attributed, in large part, to the low affinity sites. The internalization of PRL into rat liver was examined after in vivo injection of [125I]iodoovine PRL. The labeled hormone was found initially in the plasma membrane fraction, after which it localized preferentially in the Golgi fraction, with maximum incorporation 15 min postinjection. Substantial radioactivity was observed in both classes of lysosomes (L-1 and L-2). In contrast to the Golgi fraction, maximum incorporation of [125I]iodoovine PRL in lysosomes occurred at 30 min. This suggests either that during internalization, PRL first reaches Golgi elements and is then transferred to the lysosomal compartment, or that there are two independent pathways of internalization, one rapid toward the Golgi complex (may be a path of receptor recycling) and the other toward lysosomes (probably leading to receptor degradation).

Maintenance of prolactin (PRL) binding sites in rat liver cells in suspension culture: effect of PRL and of inhibitors of various cellular functions.

An in vitro method to study the regulation of PRL receptors has been established using adult rat liver cells cultured in a continuous suspension in L-15 medium. PRL binding averaged 28.2 +/- 1.8% of the added labeled hormone per 10(6) cells in freshly isolated liver cells prepared from female rats treated with 17 beta-estradiol. When these cells were incubated at 37 C, binding rapidly declined by 50% at 10 h and 90% at 48 h. This rapid decline could be counteracted by the inclusion of ovine PRL (50 nM), which maintained initial PRL receptor levels up to 48 h of culture. Higher concentrations of PRL (2.5 microM) induced a rapid down-regulation, apparent at 2 and 10 h of culture. Cycloheximide (50 micrograms/ml) induced a slight diminution of control PRL receptor levels and partially reversed the effect of 50 nM PRL. Approximately 60% of the PRL receptors were resistant to the effect of cycloheximide. On the other hand, actinomycin D (10 micrograms/ml) had no effect on PRL receptor levels in control and only a very slight effect in PRL-treated cells. Dinitrophenol, which blocks metabolic oxidation, also partially reversed the effect of 50 nM PRL although it was without any significant effect on control levels. Chloroquine (100 microM) and colchicine (1 microM) failed to alter PRL binding either in the absence or presence of 50 nM PRL. Our results suggest that the existence of regulatory factors occurring in vivo, which are absent in the culture medium, could be responsible for the decline in PRL receptor levels in the control hepatocytes. PRL itself could be one of these factors. On the other hand, and in agreement with the putative actions of the drugs utilized, the mechanism of the PRL-induced maintenance of receptor levels appears to lie in part with an effect on receptor synthesis at the translational (ribsomal) level but to be independent of the internalization or of lysosomal degradation.

Differential biological activities between mono- and bivalent fragments of anti-prolactin receptor antibodies.

Previous studies have established that antibodies against PRL receptors can mimic PRL effects on casein gene expression and on thymidine incorporation into DNA in the mammary gland. In the present work, bivalent F(ab')2 and monovalent Fab' fragments of the anti-PRL receptor antibodies were prepared. Both inhibited the binding of 125I-labeled PRL to rabbit mammary gland membranes. F(ab')2 as well as the unmodified antibodies were able to enhance casein synthesis and thymidine incorporation into DNA in cultured rabbit mammary gland explants. Moreover, when added to isolated membranes, both were able to induce the generation of the PRL relay which specifically stimulates caseins gene transcription in isolated mammary nuclei. In contrast, monovalent fragments were totally devoid of any of these PRL-like activities. However, bivalent and monovalent antibodies were equipotent in inducing a down-regulation of PRL receptors in mammary explants. These data indicate that the biological PRL-like activity of antibodies against PRL receptors is strictly related to their bivalent structure. This fact indicates a possible crucial role of a microaggregation of PRL receptors in the transmission of the PRL message across the membranes. In addition, these experiments reinforce the idea that internalization and down-regulation are not directly related to PRL action on casein or DNA synthesis in mammary gland.

Structure of the gene encoding rabbit alpha s1-casein.

The complete nucleotide sequence of the entire rabbit alpha s1-casein-encoding gene Aslca and its flanking regions was determined. These data represent the first complete primary sequence of an Aslca gene. The gene consists of 19 exons spread over 16 kb. Highly conserved sequences were found between this gene and other casein-encoding genes mainly upstream from the gene from position -180 to -10. Several repeated interspersed elements of unknown function were also identified within introns.

Cloning and functional expression of the rabbit transferrin gene promoter.

The transferrin gene is expressed in all mammals, mainly in the liver. A rabbit genomic library was screened using cDNA probes, and 8kb of 5' flanking sequence of the rabbit transferrin gene was cloned upstream of the cat reporter gene. The first 200 nucleotides of this promoter were sequenced. The rabbit transferrin promoter is highly homologous to the human and murine ones. Its functional activity was tested in the human hepatic cell line HepG2. Using transitory transfections in these cells, a proximal positive region, a negative region and a distal positive region located between -3.6 and -4.0kb were identified. This distal positive region sequence is highly conserved with the the human gene enhancer sequence, and contains an HNF3alpha binding site, the mutation of which totally abolished its effect in HepG2 and HuH7 cell lines. The rabbit transferrin 5' flanking sequence thus shows a promoter organization very similar to that of the human gene, and the HNF3alpha binding site in the distal positive region presents the same functional importance in both genes.

Structure of the gene encoding rabbit beta-casein.

The entire rabbit beta-casein-encoding gene and 400 bp upstream were sequenced. Eight introns, located essentially at a position similar to the corresponding gene in other species, were found. Strong homology with several casein-encoding genes from rabbit and from other species was observed in the upstream region of the gene. Repeated sequences of unknown function were also located within introns.

Structure of the rabbit kappa-casein encoding gene: expression of the cloned gene in the mammary gland of transgenic mice.

The rabbit kappa-casein (kappa-Cas) encoding gene has been isolated as a series of overlapping DNA fragments cloned from a rabbit genomic library constructed in bacteriophage lambda EMBL3. The clones harboured the 7.5-kb gene flanked by about 2.1 kb upstream and 9 kb downstream sequences. The cloned gene is the most frequently occurring of two kappa-Cas alleles identified in New Zealand rabbits. Comparison of the corresponding domains in rabbit and bovine kappa-Cas shows that both genes comprise 5 exons and that the exon/intron boundary positions are conserved whereas the introns have diverged considerably. The first three introns are shorter in the rabbit, the second intron showing the greatest difference between the two species: 1.35 kb instead of 5.8 kb in the bovine gene. Repetitive sequence motives reminiscent of the rabbit C type repeat and the complementary inverted C type repeat were identified in the fourth and first introns, respectively. Transgenic mice were produced by microinjecting into mouse oocytes an isolated genomic DNA fragment which contained the entire kappa-Cas coding region, together with 2.1-kb 5' and 4.0-kb 3' flanking region. Expression of transgene rabbit kappa-Cas mRNA could be detected in the mammary gland of lactating transgenic mice and the production of rabbit kappa-Cas was detected in milk using species-specific antibodies. The cloned gene is thus functional.

Effect of sodium butyrate on the stimulation of casein gene expression by prolactin.

Sodium butyrate, but not isobutyrate, inhibits prolactin action on the induction of casein synthesis and casein mRNA accumulation in rabbit mammary explants. Sodium butyrate specifically prevents the generation of the prolactin relay which can be released from isolated membranes incubated with prolactin and which stimulates directly casein gene transcription when added to isolated mammary nuclei. This indicates that sodium butyrate exerts its inhibitory action essentially at the membrane level.

A MGF/STAT5 binding site is necessary in the distal enhancer for high prolactin induction of transfected rabbit alpha s1-casein-CAT gene transcription.

The rabbit alphas1-casein gene contains a distal prolactin-dependent enhancer 3442-3285 bp 5' to the site of initiation of transcription. We have reported previously that four DNA/protein-binding sites (F1-F4) are located within this distal enhancer. We now show that one of this binding site (the F4 site) binds in vitro a MGF/STAT5-like factor. The functional importance of the F4 site was estimated by cotransfection of CHO cells with a chimeric gene containing or not the F4 sequence linked to the (-391/+1774)CAT gene and a plasmid encoding the rabbit mammary prolactin receptor. The F4 site is necessary for maximal response, of the enhancer to prolactin. However, this site has to be associated to the Fl-F3 fragment. It can be replaced by a genuine MGF/STAT5-binding site. A mutational analysis indicates that F4 and F1 sites are simultaneously involved to confer a high prolactin sensitivity.