Regulation of Protein Synthesis in Chick Oviduct I. INDEPENDENT REGULATION OF OVALBUMIN, CONALRUMIN, AND INDUCTION*

SUMMARY Egg white protein synthesis is induced and maintained in oviduct magnum of immature female chicks by administering gonadal steroid hormones. The relative rate of synthesis of four egg white proteins (ovalbumin, conalbumin, ovomucoid, and lysozyme) was measured by culturing magnum explants with radioactive amino acids followed by specific immuno-precipitation of each of the egg white proteins. Induction of these proteins was studied as a function of time, hormonal dose, and hormonal combination. A combination of estrogen, progesterone, and testosterone was found which would induce the relative rate of synthesis of each of the four egg white proteins to a level that closely approximates that observed in laying hens; this combination also promotes maximal magnum growth. The synthesis of these proteins is not strictly coordinated, since the rate of synthesis of one or more of these proteins can change relative to the others, although they are all synthesized in the same cell. Noncoordinate protein is exemplified by the changing ratio of conalbumin to ovalbumin synthesis during both primary and secondary stimulation with The ratio of synthesis of these two proteins a function the


SUMMARY
Egg white protein synthesis is induced and maintained in oviduct magnum of immature female chicks by administering gonadal steroid hormones.
The relative rate of synthesis of four egg white proteins (ovalbumin, conalbumin, ovomucoid, and lysozyme) was measured by culturing magnum explants with radioactive amino acids followed by specific immunoprecipitation of each of the egg white proteins. Induction of these proteins was studied as a function of time, hormonal dose, and hormonal combination.
A combination of estrogen, progesterone, and testosterone was found which would induce the relative rate of synthesis of each of the four egg white proteins to a level that closely approximates that observed in laying hens; this combination also promotes maximal magnum growth.
The synthesis of these proteins is not strictly coordinated, since the rate of synthesis of one or more of these proteins can change relative to the others, although they are all synthesized in the same cell. Noncoordinate protein synthesis is exemplified by the changing ratio of conalbumin to ovalbumin synthesis during both primary and secondary stimulation with estrogen. The ratio of synthesis of these two proteins also changes as a function of the dosage of hormone administered. Furthermore, different combinations of hormones can produce noncoordinate synthesis; for example, the synthesis of ovomucoid and conalbumin is increased relative to ovalbumin when either progesterone or testosterone are administered along with estrogen. The most likely explanation for the change in the ratio of ovalbumin to conalbumin synthesis observed during hormonal stimulation is the 2-fold difference in the rates of degradation of their mRNAs.
In contrast, changes in mRNA synthesis or activation best account for the preferential induction of conalbumin synthesis by low doses of estrogen, and the increased rates of conalbumin and ovomucoid synthesis which are produced by supplementing estrogen with either progesterone or testosterone. control of egg white synthesis is unlikely since several of these proteins appear to be under independent control.
Avian egg white consists of several prominent protein species that provide both nutrition and protection for the developing embryo.
The four proteins studied in this paper, ovalbumin, conalbumin, ovomucoid, and lysozyme, which together account for 85 to 90% of the egg white proteins, have been shown by immunofluorescent techniques to be localized within the tubular gland cells of the magnum (see "Appendix").
These cells constitute roughly SO7o of the magnum in a hormone-stimulated chicken.
The synthesis of egg white proteins can be induced in sexually immature chicks by administering gonadal steroid hormones; the most important hormone is estrogen, but its action is modified by progesterone and testosterone (1)(2)(3)(4)(5)(6)(7)(8)(9). Normally these proteins would not be synthesized until the onset of sexual maturity which occurs in chicks after 3 to 4 months when the gonads commence the production and secretion of steroids (1).
An unanswered question about the regulation of protein synthesis in eukaryotic organisms is whether there is any equivalent to the bacterial operon, such that the induction of several proteins is strictly coordinated by a common regulatory factor. The induction of egg white proteins was used to test this possibilit: because several of these proteins are synthesized in the same cell and because these proteins are packaged so that eggs with a fairly constant composition are produced. Thus, the question is whether the synthesis of the mRNAs for each of these proteins is coordinate, i.e. iu constant proportion to one another. Since techniques were not available to measure coordinate synthesis of the mRNAs directly, the translatable mRNA concentrations were estimated by measuring the relative rates of synthesis of the proteins for which they encode. Relating protein synthesis to mRNA synthesis must be considered tentative because of the many potential regulatory steps between these two events. For instance, the relative rate of protein synthesis is related to mRNA concentration only if all mRNAs are translated at the same rate. Furthermore, the concentration of mRNAs is not only a function of synthesis but also of activation and degradation.
After taking the above considerations into account, the results suggest that several of the egg white proteins are regulated independently rather than strictly coordinately.
However, there are many aspects of the synthesis of egg white proteins which indicate that their induction is interrelated and under similar control.

Hormone Injection Regimen-Hormones
were administered to 4-to lo-day-old, female, White Leghorn chicks according to a standardized regimen consisting of an optimal dose of estrogen (1 mg) administered daily for 10 days (a period called primary stimulation), then 15 to 30 days without hormone administration (called withdrawal), followed by several days of secondary stimulation with various steroid hormones, at a dose of 2 mg per day unless otherwise noted.
A dose of 2 mg per day was chosen because that dose of estrogen, progesterone, or testosterone produces maximal effects when administered to 200-to 400-g chicks (5,6,9).
The effect of this regimen on various magnum parameters, viz. growth, protein concentration and rate of synthesis, ribosome concentration and aggregation into polysomes, has been published (10).
17&Estradiol benzoate was a gift from Schering; 2 a-methyl dihydrotestosterone proprionate was a gift from Syntex; progesterone was purchased from Schering as "Proluton." All hormones were dissolved in sesame oil at a concentration of 10 mg per ml (10 min in a boiling water bath was required to dissolve the estrogen) and injected into the lower leg muscles.
Purification of Antigens-The protein peaks from the columns described below were localized by measuring the absorption at 280 nm and samples from the center and periphery of the peaks were subjected to electrophoresis as in Fig. 1. Only those fractions which were uncontaminated were pooled and concentrated. The extinction coefficients and approximate molecular weights are given in Table I. Ovalbumin (five times crystallized, Nutritional Biochemicals) was purified on DEAE-cellulose as previously described (13) Only the A, form of ovalbumin which carries 2 phosphate residues was used as antigen (Fig. I).
Conalbumin was purified by modifying the method of Warner (13). Egg whites were blended for 40 s in a Waring Blendor, filtered through glass wool, and centrifuged 10 min at 15,000 X g. Ethanol was added to the supernatant to 20y0 and the temperature lowered to 0". The precipitate which formed was removed by centrifugation and resuspended in 20 InM glycine, dialyzed against 20 rnrvr glycine overnight, applied to a DEAE-cellulose column, and eluted with the buffer system proposed by Mandeles (14). The peak fractions cont.aining conalbumin were concentrated by ammnoium sulfate precipitation (70yc), dialyzed against distilled water, and frozen. Fig. 1 shows the electrophoresis of conalbumin in two systems.
Ovomucoid was purified from egg white trypsin inhibitor (Sigma).
The crude protein was dissolved in 10 mm sodium phosphate, pH 7.4, and applied to a DEAE-cellulose column equilibrated with the same buffer. After washing with 160 ml of phosphate buffer a gradient of NaCl from 0 to 0.8 M was started. The peak ovomucoid fractions were pooled, the pH adjusted to 3.5 with HzS04, and then an equal volume of 10% trichloroacetic acid (pH 3) was added. The precipit,ate which formed was removed by centrifugation and discarded. The supernatant was dialyzed against distilled water and lyophilized.
Since ovomucoid does not precipitate with trichloroacetic acid it diffused out of the acrylamide gels during the destaining procedures following electrophoresis and therefore is not shown in Fig. 1. However, the protein could be detected as a diffuse band in the gels during the early st'ages of destaining; the mobility of this band was similar to ovalbumin in SDS gels and migrated approximately halfway between ovalbumin and conalbumin in the pH 8.9 gels shown in Fig. 1. In neither case was the presence of any other protein detected.
Lysozyme (three times crystallized, Nutritional Biochemicals) was further purified with methods adapted from Warner (13). The protein was dissolved in 10 mM sodium phosphate, pH.7.4, and applied to a CM-cellulose column equilibrated with the same buffer and eluted with a gradient of NaCl up to 0.5 M. The peak lysozyme fractions were adjusted to pH 10 and NaCl added to 1 M. The precipitate which formed was collected by centrifugation, then dissolved in, and dialyzed against, distilled water, and finally lyophilized.
Lysozyme migrates toward the cathode at pH 8.9 and hence the electrophoresis of this protein is only shown in the SDS gels ( Fig. 1).
Bovine serum albumin (crystalline, Pentex) was used without further purification.
Antibody Preparation-Rabbits were injected subcutaneously with approximately 5 mg of purified antigen dissolved in 1 ml of distilled water and sonicated with an equal volume of Freund's complete adjuvant.
The rabbits were bled several weeks later and a crude y-globulin fraction prepared by ammonium sulfate precipitation as before (12). Rabbits were given booster immunizations when necessary. As a precaution, purified ovalbumin, or conalbumin, or both, were added to all nonimmune sera prior to ammonium sulfate fractionation and any precipitate which formed was removed by centrifugation.
The approximate titre of the antibody preparations used, expressed as micrograms of y-globulin necessary to precipitate 1 pg of antigen at the equivalence point, is given in Table I.
Labeling Oviduct Proteins and Preparation of Homogenates-Oviducts were removed from decapitated chicks with sterile procedures, the magnum portion isolated and cut into four to ten pieces. These were incubated in Hanks' balanced salt solution containing penicillin and streptomycin, buffered with NaHC03, and continuously gassed with 95% 0~5% CO2 as previously described (12). The explants were labeled with either 3H-amino acid mixtures (12.5 PCi per ml, Schwarz) or 14C-amino acid mixtures (1.25 PCi per ml, Schwarz) for 1 to 3 hours. After labeling, The homogenate WAS then diluted to 2o/, with the above buffer without detergents and centrifuged 90 min at 40,000 rpm (Spinco 40 rotor).
The supernatants were filtered through glass wool if much lipid was present and used for the determination of total protein radioactivity and immunoprecipitable radioactivity as described below.
When magnums are homogenized in the presence of detergents more than 90% of the incorporated amino acids are solubilized. Less protein is solubilized when detergents are not used, resulting in relative rates of ovalbumin synthesis about 1.6 times higher than those reported here (8,12 I~~zmunoprecipitation-Immunoprecipitation of radioactive oviduct proteins was performed essentially as before (12). At least two different dilutions of supernatant were always precipitated to ensure antibody excess, and 5 pg of unlabeled purified antigen were added to all samples which did not contain at least that amount of endogenous antigen (immunoprecipitation of [3I-I]ovalbumin was incomplete when less than 4 pg of ovalbumin was present).
The precipitation of 5 pg of bovine serum albumin added to the radioactive oviduct homogenate by anti-bovine serum albumin was used as a control for nonspecific coprecipitation. This value (usually 0.2. f O.lyO of t,otal acid-precipitable radioactivity) was determined for each homogenate and subtracted from the experimental values. Total acid-precipitable radioactivity was determined as before (8). The counting efficiency for both acid precipitates and immunoprecipit'ates was between 50 to 55% for 3H and 80 to 85% for W. measuring relative rates synthesis of specific protein in culture probably reflects the in viuo condition.
The radioactive profile of magnum proteins separated by SDS acrylamide gel electrophoresis from chicks given 8 days of primary stimulation and from chicks after several weeks of Tvithdrs\val are shown in Fig. 2 The profile of magnum proteins from withdrawn chicks is similar to that from unstimulated chicks (8). Estrogen treatment changes completely the profile of proteins synthesized; the two major peaks correspond to conalbumin and ovalbumin.

Validity of Methods
Antibody Specijcity-The specificity of each antibody preparation used was ascertained by SE-acrylamide gel electrophoretic analysis of the immunoprecipitates formed when an excess of antibody was added to a magnum homogenate of tissue which had been labeled in culture with radioactive amino acids (12). Fig. 2 shows the specificity of the anti-ovalbumin, anti-connlbumin, anti-ovomucoid, and anti-lysozymepreparat,ions. Ineach case the precipitate was formed in 3%labeled homogenate and [%]ovalbumiu was added as an internal marker. Each panel shows a single tritium radioactive peak which accounts for at least 907; of the total immunoprecipitable radioactivity. Labeling Oviduct Proteins-Synthesis of magnum protein in All the proteins migrated according to the log of their molecular culture has been shown to be qualitatively and quantitatively weight (Table I) except ovomucoid. The anomalous migration similar to that in viva (12). Over 95% of the incorporated of ovomucoid may be due to the fact that this protein contains amino acids are recovered in the tissue when labeling times are approrimately 25 7;; carbohydrate (15). Since the mobility of less than 5 hours; and the homogenization procedures used solu-ovomucoid is the same a.s ovalbumin on SDS-acrylamide gels, bilize more than 90% of the newly synthesized protein.
Thus, the specificity of the anti-ovomucoid preparation was also tested

Measuring
Relative Rate of XpeciJc Protein Syn,thesis-The rates of egg white protein synthesis are expressed as relative rates of protein synthesis, i.e. the percentage of the total acid-precipitnble radioact,ivity which is specifically immunoprecipitable. This form of data presentation was chosen because: (a) it eases t,he comparison of one tissue with another by avoiding the problems inherent in measuring the actual rates of protein synthesis which entails knowing the specific activity of the amino acid pools; and (b) the relative rate of protein synthesis may be directly related t,o the relative mRKA concentration if the rates of polypeptide initiat,ion and elongation on all mRNAs are t,he same (see "Discussion"). Fig. 3 illustrates the met'hod used to measure t,he relative rate of synthesis of four of t,he egg white proteins.
The amounts of supernatant used for each curve depend on the endogenous antigen concentration and the antibody titres. After 8 days of primary stimulation with estrogen the relat,ive rates of synthesis of each of the proteins, except ovomucoid, approximate t,he relative amount of each of the proteins in egg white (Table I). Together these four proteins account for about 68s; of total protein synt,hesis after this hormonal treatment. induction of Egg White Protein Synthesis with Estrogen Tilne Course-The relative rates of ovalbumin and conalbumin synthesis at various times during a standardized regimen of hormonal stimulation and withdrawal are shown in Fig. 4. Oyalbumin synthesis is undetectable in unstimulated immature chicks and is first detectable approximately 24 hours after primary stimulation (8). Fig. 4  magnums aft'er 10 days of primary stimulation. During withdrawal the relative rat'e of ovalbumin synthesis declines until it is again undetectable after 10 to 15 days. Light and electron micrographs, and measurements of DNA content, indicate that during withdrawal most of the tubular gland cells are retained although they cease to synthesize the characteristic secretory proteins (6,10). After commencing secondary stimulation, ovalbumin synthesis rises to the same maximal value as during primary stimulatiou but in about one-half the time.
In addition, there is only a 5-hour lag before ovalbumin is first detectable after secondary stimulation. Fig. 4 (bottom) shows that the rate of conalbumin synthesis reaches a value of 10 to 12% of the total protein being synthesized but this value is reached in about one-half the time it takes ovalbumin to reach its maximal values. Also, the rate of decay of conalbumin synthesis during withdrawal is slower than that of ovalbumin synthesis.
Thus, the ratio of ovalbumin to conalbumin synthesis changes from about 1 during early primary and secondary stimulation and late withdrawal to approximately 6 during late primary and secondary stimulation.
The maximal ovalbumin to conalbumin synthesis ratios are similar to that in magnum tissue from laying hens (16) and resemble the ratio of these proteins in egg white (Table I). Possible mechanisms underlying the noncoordinate synthesis of these two proteins are examined below.
Estrogen Dose Response-The effect of varying doses of estrogen during secondary stimulation on the relative rate of specific protein synthesis, as well as magnum growth, is shown in Fig. 5. After either 1 day (Fig. 5A) or 2 days (Fig. 5B), half-maximal stimulation of specific protein synthesis and growth occurs with doses of 0.1 to 0.2 mg of estrogen per day to 400-g chicks. Maxi- ma1 stimulation occurs with 1 to 2 mg of estrogen per day. The ratio of ovalbumin to conalbumin synthesis changes as a function of dose as well as a function of time (above). With less than 50 pg of estrogen per day the relative rate of conalbumin synthesis exceeds ovalbumin synthesis, whereas at maximal doses of estrogen the ovalbumin to conalbumin ratio is >2. Fig. 5B also shows the dose response for the induction of ovomucoid and lysozyme with estrogen. However, the relative rates of synthesis of these two proteins are too low to determine whether their dose response is similar to either ovalbumin or conalbumin. The time course of ovalbumin and conalbumin induction with suboptimal levels of estrogen (250 pg per day) is shown in Fig. 6; for comparison the curves from Fig. 4 (optimal estrogen, 2 mg per day) are superimposed.
The induction of conalbumin synthesis is similar with either dose of estrogen, whereas the induc- were given up to 4 days of secondary stimulation with 2 mg of progesterone per day. The relative rates of synthesis were determined as in Fig. 3. For reference the relative rate of ovalbumin and conalbumin synthesis dllring secondary stimulntion with estrogen ( Fig. 4) is shown by clotted lines.
for this discrepancy is that the estrogen concentration in the serum may decline slowly during withdrawal and may be s&icient to maintain secretory protein synthesis for some time.
This explanation is possible because estrogen is administered as a supersaturated solution in oil, thus creating a depot for slow release.
This possibility was tested by inducing egg white protein synthesis with a lower dose of estrogen (0.1 mg per day) and then measuring the relabive rate of ovalbumin and conalbumin synthesis during withdrawal (Fig. 7). Comparison of the results with those in Fig. 4 (Figs. 3,4,9). Induction with estrogen will be used as a standard for comparison in the sections below. Testosterone Alone-Testosterone does not induce the synthesis of any of the egg white proteins tested when given alone, nor does it affect magnum growth (Table II).
The values given are from chicks given 4 days of secondary stimulation; they are not significantly different from those obtained in control chicks (29 days withdrawn).
Progesterone Alone-Although incapable of inducing ovalbumin synthesis (and presumably the synthesis of conalbumin, ovomucoid, and lysozyme as well) during primary stimulation (8), progesterone induces the synthesis of all four proteins when administered as a secondary stimulation (Table  II). Fig. 8 shows that conalbumin synthesis is induced with progesterone at the same rate as with estrogen, but ovalbumin is induced more rapidly and then plateaus at a level 25% lower than with estrogen alone. Progesterone stimulates magnum growth less than estrogen. A large portion of the wet weight increase observed with progesterone (from 70 to 390 mg, Table II), is caused by edema and cellular hypertrophy since magnum DNA content increases only l&fold after 6 days compared to 5-fold with estrogen (6). Because progesterone has little effect on tubular gland cell cyto- were given 4 days of secondary stimulation after 29 days withdrawal.
Two milligrams of each hormone was administered daily to these 420-g chicks as indicated in the first column as a subscript.
Relative rates of protein synthesis were determined as in Fig. 3 that the time course of induction during secondary stimulation is not significantly influenced by the cytodifferentiation of new tubular gland cells.
Combination of Testosterone and Progesterone-This combination results in a slight inhibition of egg white protein synthesis compared to progesterone alone (Table II).
Combination of Testosterone with Estrogen-In contrast to above, combination of testosterone with estrogen has a marked synergistic effect on magnum growth and on the induction of ovomucoid synthesis, and to a lesser extent conalbumin synthesis, compared to estrogen alone (Table II).
The synergistic effect of testosterone on growth is partly due to cellular hypertrophy and partly due to an increase in cell number since the DNA content does not increase proportionately with wet weight (Table III). The augmentation in conalbumin synthesis and content, as well as magnum growth, that is produced when testosterone is administered with estrogen has been noted previously under slightly different experimental conditions (1,2,9). Combination of Progesterone and Estrogen-With this combination growth is inhibited, compared to estrogen alone, but the synthesis of ovomucoid and conalbumin is significantly enhanced, even more than with estrogen plus testosterone (Tables II and  III).
After 4 to 5 days of secondary stimulation with estrogen plus progesterone, conalbumin synthesis is almost double, and ovomucoid synthesis is 3 to 4 times greater than with estrogen stimulation alone. The time course of the induction of all four proteins is shown in Fig. 9. In these experiments the magnums by guest on March 24, 2020 http://www.jbc.org/ Downloaded from from estrogen-plus-progesterone-treated chicks were labeled with "C-amino acids while magnums from estrogen-treated chicks were labeled with Samino acids; the immunoprecipitations were performed with supernat'ants derived from the pooled tissues, thus eliminating possible procedural differences.
The data for kinetics of ovalbumin, ovomucoid, and lysozyme induction are all similar in that it takes 4 to 5 days to reach maximal induction with either estrogen or estrogen plus progesterone.
The induc-t#ion of conalbumin synthesis is peculiar since maximal values of 10% of total protein synthesis are reached in 2 days with estrogen, and values of 20% are reached in 5 days with estrogen plus progesterone; furthermore, the curves are not linear. Combinations of Estrogen, Progesterone, and Testosterone-The data of Table II suggest that administering some combination of estrogen, progesterone, and testosterone, each at the appropriate concentration, might produce relative rates of synthesis of each egg white protein similar to those achieved naturally in laying hens which have been reported (16) and are: ovalbumin, 64.2%; conalbumin, 12 l"/c; ovomucoid, 8.57%; lysozyme, 1.5%. Two series of experiments were tried: (a) a saturating dose of estrogen (2 mg) with varying concentrations of progesterone, and (b) saturating doses of estrogen and testosterone (2 mg) with varying amounts of progesterone; in each case the hormones were administered for 5 days as a secondary stimulation. Table III shows that increasing the dose of progesterone augments ovomucoid and conalbumin synthesis in both series of experiments. With higher doses of progesterone, conalbumin synthesis becomes greater than physiological.
In addition, growth is progressively inhibited with increasing progesterone concentration.
As noted above, testosterone has a synergistic effect on growth whether measured as wet weight, DNA, or protein (Table III).
The combination of 2 mg each of estrogen and testosterone and 0.1 mg of progresterone appears to be the most physiological combination of hormones for inducing egg white protein synthesis in chicks. After 5 days of secondary stimulation with this combination, over 80 7; of the proteins being synthesized are secretory proteins, and the magnums are almost lo/;, of the body weight, values similar to those of actively laying hens. mRNA Half-lije One method of estimating the half-life of mRNA is to stop RNA synthesis and then measure the decay in the rate of specific protein synthesis.
The rates of degradation of several egg white protein mRNAs have been measured in this may by pulse labeling separate cultures of magnum tissue with radioactive amino acids at various times after addition of actinomycin G (at 10 pg per ml, a concentration which inhibits RNA synthesis greater than 98%; (12)). The tip of ovalbumin mRNA has been estimated in magnum tissue from chicks receiving 5 days of primary stimulation, 1 and 5 days secondary stimulation; in each case the tlp was between 12 to 14 hours (12). Ovomucoid mRNA also has a t,lP of about 14 hours, but conalbumin mRNA has a faster rate of degradation with a tllz of 6 to 8 hours3 The average mRNA tli2 for nonsecretory proteins (total protein minus immunoprecipitated egg white proteins) is even shorter, t1/2 of 4 to 5 hours (I 2).
To explore the mechanism by which estrogen plus progesterone augments ovomucoid and conalbumin synthesis compared to tstrogen alone, the relative rates of mRNA degradation were compared after hormonal stimulation with the two treatments. These experiments were conducted as described above, although a double label technique was employed; tissue from estrogen-plusprogesterone-treated chicks was labeled with 1Gamino acids while tissue from estrogen-treated chicks was labeled with ?I-Iamino acids. The results are plotted as percentage of total protein; thus after blocking RNA synthesis with actinomycin D, those proteins with relatively long lived mRNAs will comprise a progressively greater percent of the total protein synthesized, and vice versa. For comparative purposes this seemingly indirect method of measuring mRNA half-lives is simpler and more reliable than attempting to measure actual rates of protein synthesis.
The results of two experiments, one conducted after 1 day of secondary stimulation with estrogen or estrogen plus progesterone treatment (Fig. 10A) and the other after 3 days secondary stimulation (Fig. IOB), are shown.
In both experiments all the secretory proteins measured become a larger percentage of the tot,al protein synthesized while nonsecretory proteins become a smaller percentage.
The actual slopes of the curves depend on the relative amounts of mRSh of tach half-life; thus, in the experiment illustrated in Fig. IOA the slopes for all t'he secretory proteins are greater t,han in Fig. 1OR because there are more mRNAs with short half-lives after 1 day than after 3 days of secondary stimulation.
However, within an experiment the slopes should be proportional to the mRNA half-lives. Ovalbumin and ovomucoid mRNAs have the longest half-lives, followed by lysozyme, conalbumin, and nonsecretory protein mRiYAs. Comparison of the slopes of each specific protein after estrogen or estrogen plus progesterone treatment in either experiment reveals little difference in the rates of mRr\'A degradation.
If anything, there are slightly shorter mRNh half-lives for ovomucoid and conalbumin after 3 days of estrogen plus progesterone treatment; thus the increased synthesis of these proteins after this treatment is not due to mRNA stabilization.

Increases i n Specijk
Translatable mRNA Concentration---As a result of hormonal stimulat'ion of immature chicks, egg white protein synthesis rises from undetectable levels to 80 to 907; of the total protein being synthesized in oviduct magnum (Tables  II and III).
These increases in specific protein synthesis proba- doses which presumably saturate all the specific binding sites. All the proteins are induced at the same time aft.er secondary stimulation and for three of them, the relative rates of synthesis increase in a nearly linear fashion, for 4 days and then begin to plateau thereby keeping their rates of synthesis in constant proportion (Fig. 9). The exception to this pattern, conalbumin, whose synthesis plateaus earlier, (Figs. 4,6,8,9) can be explained by its shorter mR?;A half-life.
In culture, the absolute rate of conalbumin synthesis declines with a tip of about 7 hours compared to 14 hours for ovalbumin, whether RNA synthesis is inhibited with actinomycin I> or not.3 Also, the relative rate of conalbumin synthesis increases less than the other secretory proteins after actinomycin I> treatment (Fig. 10). Theoretically, the time course of the change in mRNh concentration, which results from a change in mRNA synthesis/activation, should reflect the tljz of the mRNA, just as the approach of protein concentrations to new steady state levels after a change in the rate of protein synthesis depends on the rate of protein degradation (al), i.e. the shorter the tljz, the faster the new steady state level should be reached.
Differential translation of conalbumin mRNA is ruled out because of results indicating that after both 1 and 3 days of secondary stimulation the relative rate of conalbumin and ovalbumin polypeptide initiation and elongation stay in nearly constant proportion to each other.3 Thus, after secondary stimulation with saturating doses of estrogen, and perhaps with other hormonal regimens as well, the working model is that there is essentially a step change in the rate of synthesis/activation of the mRNAs for each of the egg white proteins studied.
There is a lag of about 5 hours, then these mRNAs enter the pool of translatable mRNAs at nearly constant rates for the next several days. I%ut, because of its greater rate of degradation, conalbumin mRNA concentration plateaus sooner than the others.

Regulation oj Egg IlXite Protein
Synthesis-This model suggests that the induction of mRXA synthesis/activation of the various egg white proteins could be linked via some common regulatory factor.
However, several experimental findings indicate that the model must be more complex since the synthesis/activation of the various mRNAs appear to be controlled independently.
If the model were correct, then one would predict that during withdrawal conalbumin synthesis would decline faster than ovalbumin synthesis because of its shorter tl/s, contrary to that observed in Fig. 4. This discrepancy suggested the experiments which showed that conalbumin synthesis is preferentially induced at low estrogen concentrations (Figs. 5 to 7). Thus, a likely reconciliation of these facts can be envisaged if the concentration of estrogen in the serum falls gradually during withdrawal and reaches the threshold for ovalbumin synthesis before reaching the threshold for conalbumin synthesis.
The results illustrated in Fig. 7 support this contention.
Another test would be to measure the decay in ovalbumin and conalbumin synthesis after treatment of chicks with a potent anti-estrogenic compound. Other evidence for independent regulation of egg white mRNA synthesis/activation comes from experiments measuring induction with different steroids. For example, Fig. 8 shows that with progesterone treatment the ratio of ovalbumin to conalbumin synthesis during secondary stimulation is different from that with estrogen treatment.
Also, with estrogen plus testosterone the induction of ovalbumin is not significantly altered whereas ovomucoid and conalbumin are induced to greater extents (Tables II and III).
The comparison of estrogen treatment with estrogen plus progesterone treatment was studied most thor-oughly because of the prominent effects of the latter treatment on ovomucoid and conalbumin synthesis (Fig. 9). Translational differences are an unlikely explanation because the rate of conalbumin polypeptide elongation and initiation remain in constant proportion to ovalbumin with both hormonal treatments, and the average rate of initiation and elongation for all proteins is similar with either treatment.3 Comparison of the two treatments reveals no significant differences in the rates of mRNA degradation for any of the egg white proteins (Fig. 30). Thus, differences in the rates of mRNA synthesis/activation seems to be the most likely explanation of increased ovomucoid and conalbumin synthesis when progesterone is present with estrogen. The similarity in the time course of ovomucoid induction with either hormonal treatment (Fig. 9) is further evidence that there is an effect on mRNA synthesis/activation rather than degradation. With conalbumin, the time courses are not linear and they plateau at different times ( Fig. 9) suggesting that the rate of conalbumin mRn'A synthesis/activation may not be constant throughout secondary stimulation. Relative Number of Egg lb'hite Protein mRNAs and Genes-By knowing the relative rate of synthesis for each of the proteins synthesized in a cell and the relative rates of mRNA degradation one can calculate (a) relative concentrations of each translatable mRNA and (b) relative gene frequencies.
The main assumptions are that: (a) all genes are transcribed at the same rate, (b) all mRNAs are activated at the same rate, and (c) all mRNAs are translated at the same rate. There is some evidence to support the third assumption,3 but the others are unfounded. Table  IV shows that the relative number of mRNAs is a function of the relative rate of synthesis and polypeptide length, whereas the relative gene frequency is the product of the relative rate of synthesis and degradation.
If the assumptions are correct then for every lysozyme gene there are about 30 ovalbumin, 12 conal- Assumes all genes transcribed at the same relative rate, all mRNAs activated and translated at same rate. bumin, and 5 ovomucoid genes. If there are equal numbers of each of t&se genes, then there must be large differences in the rates of transcription or activation of the different mRSAs. Steroid Receptors-Although this paper does not deal dir&l> with steroid receptors, it indicates that analysis of the interaction of these receptors with their target sites will be complex.
Part of this complexity arises from the fact that there are at least two cell types in oviduct magnum which respond to these hormones, the tubular gland cells which synthesize the proteins studied here (see "Appendix") and the goblet cells that synthesize avidin (22). Another aspect of the problem is that thele are at least three different steroids that affect the magnum.
Both estrogen and progesterone receptors have been identified in chick oviduct and both have been found attached to chromatin (23, 24). O'Malley et al. have shown that there is specificity of the progesterone receptor for oviduct chromatin and this specificity appears to reside in the acidic proteins (25). The two receptors appear to be distinct proteins because of their different sedimentation properties in sucrose gradients.
I-Iowever, there is some competition for each receptor by the other active steroids (23, 26).
This study allows several predictions about the number and action of tubular gland cell steroid receptors.
The facts that progesterone is inactive alone as a primary stimulation (8) and testosterone is always inactive alone indicates that there are at least three distinct receptors, one each for estrogen, progesterone, and testosterone.
The progesterone receptors may be induced by estrogen since only after tubular gland cells cytodifferentiate do they respond to progesterone by synthesizing egg white proteins (26). There are probably multiple target sites for the receptors, rather than a single regulatory site which mediates many pleiotropic responses. The independent regulation of the induction of the egg white proteins is one piece of evidence, while the fact that as a secondary stimulation estrogen induces the synthesis of egg white proteins, ribosomes, and DNA whereas progesterone affects mainly the former, is another (6,10). Furthermore, thedifference in thethrcshold for conalbuminand ovalbumin induction suggests that the affinity of receptors for different targets may not be the same. The regulation of egg white synthesis is even more complicated by the suggestion that full nctivation of some receptor targets requires either receptors with both estrogen and progesterone bound, or both receptors bound at the same time, e.g. in the induction of ovornucoid. Finally, the testosterone receptor seems to act at a site distinct from either estrogen or progesterone receptors because it is synergistic with estrogen, but not progesterone, and inactive alone.
Contrast with Bacterial Operon-The regulation of egg \\hite protein synthesis by steroids has certain gross similarities to t'he inducible enzyme systems in bacteria.
In both cases the synthesis of a set of specific proteins is induced by low molecular weight effecters. Furthermore, both are dynamic control systems in that the effect is modulated by effector concentration rather than a simple '(or1-off" switch.
However, in the inducible bacterial systems the tffector binds to a repressor and thereby releases it from the 11KA resulting in activation of gene transcription, whereas in the steroid-induced systems the effector binds to a receptor which presumably promotes mRKA synthesis/ activation by binding to a target site, e.g. chromatin.
hlorcover, in eit,her operon-or regulorl-inducible systems there is a single regulator protein but in the oviduct there are several regulators involved, the different hormone receptors and their target sites. Thus, the synthesis/activation of egg white protein mRNAs can be perturbed to alter the ratio of synthesis of the corresponding proteins but such perturbations do not occur in the bacterial controls.
Another prominent difference is that in bacteria the effector action is limited to a particular set of related proteins whereas in the oviduct and many other systems mediated by steroids, the effector not only induces a particular set of proteins but also stimulates the entire protein synthetic apparatus resulting in accumulation of tRNA (27), ribosomes (10, 28), endoplasmic reticulum (8,12), translation factorq3 etr. Concomitantly steroids may also promote cell division leading to more competent cells capable of synthesizing the specific proteins (4)(5)(6)(7)(8).
Ackncwledg??:ents---I sincerely thank 11r. R. T. Schimke who provided both excellent research facilities and a stimulating atmosphere for conducting this research.
1 am grateful to Doctors R. Schimke, R. Cos, N. Carey, and N. Stebbing for constructive criticisms during the preparation of this manuscript. The help of 1'. LeTourneau and M. Holzer, who purified ovomucoid, lysozyme, and conalbumin, is also appreciated.