Regulation of Mouse Major Urinary Protein Production

A method was developed to quantitate the daily excretion of the three major urinary proteins (mups) to test which parameters of the mup phenotype are controlled by the the Mup-a gene. Electrophoretic separation of the mup proteins, followed by staining and spectrophotometric scanning was used to characterize the phenotypes of various inbred strains. The mup phenotype of a strain proved to have two components: the absolute levels and the relative proportions of the mups present in the urine. Testosterone treatment alters both components of the mup phenotype, increasing mup excretion and altering their relative proportions. The induced proteins are the same as the basal proteins as judged by electrophoretic mobility, molecular weight, and reactivity with antibody. All strains excrete all three mups when induced. The Mup-a gene appears to be a single, codominantly expressed regulatory locus that controls the induced proportions of the three proteins. However, other genes in addition to Mup-a participate in controlling the basal mup proportions, as well as individual and total mup levels before and after testosterone treatment. HE major urinary proteins (mups) of inbred mice are a family of three Tandrogen-inducible proteins that are synthesized in the liver and ultimately excreted into the urine (FINLAYSON, POTTER and RUNNER 1963; RUMKE and THUNG 1964; FINLAYSON et al. 1965). The three proteins are defined by their electrophoretic mobilities, and inbred strains of mice exhibit variation in their urinary mup electrophoretic patterns. This variation is controlled by the Mup-a locus on chromosome 4 (HUDSON, FINLAYSON and POTTER 1967; FINLAYSON, HUDSON and ARMSTRONG 1969). All mice excrete mup 3, but strains carrying the Mup-a' allele appear to excrete mup 1, but not mup 2; conversely strains carrying the Mup-az allele appear to excrete mup 2, but not mup 1. The mup system provides an excellent model for studying the hormonal regulation of specific gene expression. First and most significant, there is definable genetic variation in the production of the mups in response to androgenic hormones; consequently, this system may ultimately offer insight into the role of specific DNA sequences in the stimulation of gene expression by steroid hormones. Second, when induced by androgens, mice excrete milligram quantities of the mups daily. Therefore, mup synthesis must constitute a considerable portion of the daily liver protein synthesis. This large response, combined with Genetics 90: 597-612 November, 1978. 598 P. R. SZOKA A N D K. PAIGEN the small molecular weight of the mups, should facilitate development of techniques to study mup gene expression at the level of messenger RNA. Third, protein turnover should not be a complicating factor in studying mup induction by testosterone, since the mups are rapidly excreted rather than degraded. Although untreated female homozygotes excrete either mup 1 or mup 2, depending on the Mup-a allele carried by the strain, Mup-d/Mup-a2 heterozygotes excrete both mup 1 and mup 2. This suggests that mup 1 and mup 2 represent the allelic products of a structural polymorphism at the Mup-a locus (HUDSON, FINLAYSON and POTTER 1967). This hypothesis was supported by biochemical studies demonstrating that mup 1 and mup 2 are nearly identical proteins; they have identical amino acid sequences for the first 36 N-terminal residues (FINLAYSON et a1 1974), and differ by only a single tryptic peptide (FINLAYSON et aZ. 1968). The exact structural difference between mup 1 and mup 2 is still unknown. Although mup 3 is related to the other two proteins, it is distinct, since it differs in amino acid sequence from mup 1 and mup 2 (FINLAYSON et al. 1974). Our current finding that all three mups may be present after induction of homozygous animals suggests that this interpretation of the Mup-a locus as a structural polymorphism may be overly simplified. The genetics of the mups is also uncertain in the sense that it is not known whether genes other than Mup-a participate in the regulation of the mup production. In order to clarify these questions we have developed a method for quantitating the urinary mup phenotypes of mice and, using this method, asked what parameters of the phenotype are under the control of the Mup-a locus. The phenotype proved to be complex. Testosterone not only induces an increase in the absolute levels, but also alters the relative proportions of the mups in urine. A comparison of inbred strains makes it clear that the Mup-a gene does not define a structural polymorphism as was originally suggested, since all strains excrete all three mups when induced. Rather, Mup-a appears to be a regulatory locus that determines the relative proportions of the three mups, when induced. However, by itself, Mup-a does not control the basal proportions of the mups, nor the levels of mup excretion. The determination of these phenotypes includes additional genetic factors. MATERIALS A N D METHODS Animals: Female inbred mice were obtained from the Jackson Laboratory, Bar Harbor, Maine. All mice were between two and four months old. Whenever possible, within an experiment mice of each strain were age matched. Only female mice were used to avoid the problem of variation in endogenous androgen levels found among males. Urine collection and testosterone induction: For urine collections, animals were maintained on a schedule of 12 hr light alternated with 12 hr darkness and housed two to four mice per cage in stainless steel metabolism cages equipped with food and water. Food was presented as a paste of water and ground Rockland Mouse Breeder Diet (Teklab, Inc., Monmouth, Illinois) containing 17% minimum crude protein, 10% minimum crude fat, and 2.5% maximum crude fiber. Urine was collected over 24 hr periods into 12 ml conical test tubes containing mineral oil, but no preservative. Urine samples were clarified at 1,000 x g for 10 min and stored at -20". The same mice were used for both basal and testosterone-induced urine collections. 30 mg testosREGULATION O F MUP PRODUCTION 599 terone pellets (Schering Corp. or Department of Pharmaceutics, University of Tennessee College of Pharmacy) were implanted subcutaneously and maintained throughout the induction period. Induced urine samples were collected on days 17, 18, and 19 after implantation of testosterone pellets, unless otherwise noted. EZectrophoresis: Urine samples were dialyzed at 4" overnight against 100 volumes of 0.01 M tris, 0.01 M acetate, 0.25 M sucrose pI-1 5.4, then assayed for protein content by the microbiuret procedure of GOA (1953). Samples of 5 pg protein in a volume of 5 or 10 c1 were electrophoresed at 200 V for 90 min through a 10 cm x 1.5 mm 5% polyacrylamide slab gel containing 0.2% bisacrylamide and buffered with 0.01 M tris, 0.01 M acetate pH 5.5 (final concentrations). The electrophoresis buffer was 0.01 M tris, 0.01 M acetate pH 5.4. After electrophoresis gels were stained with either Coomassie brilliant blue G or Amido Schwartz. For Coomassie brilliant blue G staining, the gels were first fixed for 10 min in 12.5% TCA and then stained for two hr with stirring with 20:l (v/v) of 12.5% TCA: 0.25% Coomassie brilliant blue G (Sigma). The stain was further developed by incubating the gel overnight in 5% acetic acid. For Amido Schwartz staining, the gels were stained for 30 min with 1% Buffalo Black NBR (Allied Chemical) in 7.5% acetic acid with stirring and then destained with 7% acetic acid. Gels were scanned using a Corning 750 densitometer using a setting of 1 O.D., a slit of 0.2 x 3 mm, and wavelengths of 650 nm for Coomassie brilliant blue G or 625 nm for Amido Schwartz. By comparison of the peak areas of urine mup bands with that of a purified mup 3 standard, pg mup protein/band was determined. The p H 5.5 polyacrylamide gel system for separating the mfips was used in a modification of the method of FERGUSON (1964) that related molecular size with the dependence of electrophoretic mobility on acrylamide gel concentration. The concentration of acrylamide was varied appropriately. The gels were stained with the Coomassie brilliant blue G stain. The SDS polyacrylamide gel electrophoresis was done by the method of LAEMMLI (1970) using 4 to 6 pg of each standard protein. Gels were stained for 30 min with 0.25% Coomassie brilliant blue R (Sigma) in 5 parts methanol: 5 parts water: 1 part acetic acid, destained with 50% methanol, 7% acetic acid, followed by treatment for one hr with 5% methanol, 7% acetic acid. Gels were stored in 7% acetic acid. Purification of the mups: The mups were purified from urine of testosterone-treated C57BL/6J female mice, since this strain excretes considerable quantities of each mup when induced. The procedure used was a modification of that of FINLAYSON et a2. (1968). Approximately 120 mg of urinary protein in 0.1 M tris, 0.1 M acetate pH 5.5 was applied to a 1.5 x 81 cm column of Sephadex GI00 (Pharmacia) and eluted with the same buffer at a flow rate of 5 ml/hr. The peak containing the mups was dialyzed against distilled water, lyophilized, and reconstituted with 0.05 M tris, 0.05 M acetate pH 5.5; 22 mg of this protein was applied to a 1 X 25 cm column of DEAE cellulose (Whatman DE52, microgranular, preswollen) preequilibrated with the same buffer. After washing the column with the same buffer containing 0.05 M NaCl elution was performed by applying 400 ml of a linear 0.05 t o 0.10 M NaCl gradient in the column buffer; this gradient resulted in elution of mup 1, mup 2, and mup 3 in separate peaks. Application of another 400 ml linear gradient containing 0.10 to 0 . 1 5 ~ NaCl in the same buffer to the column eluted a second fraction containing mup 3. All of the peak fractions were dialyzed against distilled water, lyophilized, reconstituted in 0.02 M imidazole, 0.25 M sucrose pH 7.4, and stored at -20". Throughout the purification procedure, protein concentration were determined by the method of LOWRY et a2. (1951) and fractions containing the mups were identified by the pH 5.5 polyacrylamide gel electrophoresis. Preparation of antibody: Equal quantities of each purified mup in 0.02 M imidazole, 0.25 M sucrose pH 7.4 were combined as a pooled antigen. Rabbits were immunized with 90 pg of antigen mixed with an equal volume of Freund's complete adjuvant. Injections were given intradermally at multiple sites in the back and were administered every two weeks for a total of six injections; then blood was collected. The serum prepared from the blood was stored at -20". 600 P. R. SZOKA A N D K. PAIGEN Double diffusion analysis: The Ouchterlony technique was performed by the method of OUCHTERLONY (1958). Wells with a capacity of 15 pl were cut in plates containing 1% agarose (Fisher) dissolved in 0.85% saline, 0.09 M tris HCl pH 7.5, with 0.1% Na azide added as a preservative. After addition of the appropriate antiserum and antigens to the wells, the plates were allowed to develop overnight a t room temperature.