The early time course of calcium-binding protein induction by 1,25-dihydroxyvitamin D3 as determined by computer analysis of two-dimensional electrophoresis gels.

The time course of calcium-binding protein induction by 1,25-dihydroxyvitamin D3 was examined in embryonic chick duodena by single-label autoradiography. Duodena were excised from 19-day-old embryos, cultured for 24 h in defined medium, and exposed to 1,25-dihydroxyvitamin D3 at 1/2, 1, 1 1/2, 2, 2 1/2, 4, 6, and 20 h before the end of the culture period. Control duodena were identically handled but were not exposed to the hormone. All duodena were pulsed with [14C]leucine during the final 30 min of incubation. Cytosolic proteins, extracted from each radiolabeled tissue, were resolved by two-dimensional electrophoresis and autoradiographs were prepared from the resulting gels. Calcium-binding protein was detectable by autoradiography in duodena cultured with hormone for 1 h or more but not in the control duodena. The autoradiographs were converted to digital images by a scanning densitometer and entered into the Man-computer Interactive Data Access System. The total radioactivity incorporated into calcium-binding protein during the 30-min pulse was determined by computer analysis of the digitized autoradiographs, and the rate of calcium-binding protein biosynthesis was calculated. Duodena cultured with hormone for 1-20 h synthesized calcium-binding protein at rates of 129 +/- 17 to 8800 +/- 110 pg/mg of cytosolic protein/30 min. These rates demonstrate that calcium-binding protein is induced within the first hour of exposure to 1,25-dihydroxyvitamin D3, but in amounts too small to be detected by commonly used immunoassays.

5 To whom all correspondence should be addressed. No reprints will be available from the authors.
appears to facilitate nuclear localization of the hormone (24-26). Studies of cultured duodena (27-31) have further demonstrated that hormonal stimulation of calcium transport can be blocked by a variety of inhibitors of transcription and translation. In the case of cycloheximide, this blockage is completely reversible (31).' Finally, one protein has been identified which is synthesized de novo in response to 1,25-(OH)'D3 administration. This gene product, calcium-binding protein (32,33), is induced by the nuclear release of newly synthesized mRNA which codes for the protein (34)(35)(36)(37)(38)(39)(40).
As calcium-binding protein is presently the only known gene product that results from the action of 1,25-(OH)2D3, considerable research effort has focused on its possible involvement in the calcium transport process. Early studies in which rachitic chicks were dosed with vitamin D3 or 25hydroxyvitamin D3 showed that increases in intestinal calcium transport were closely paralleled by the synthesis of calcium-binding protein (32,(41)(42)(43)(44). Experiments conducted with cultured embryonic duodena produced similar results (29,45). Later studies in which rachitic chicks were dosed with 1,25-(OH).JI3 indicated that the stimulation of calcium transport preceded the induction of calcium-binding protein, leading to the conclusion that this protein is not involved in the initiation of the calcium transport process (17,(46)(47)(48)(49). Consistent with this conclusion is the hypothesis that calcium-binding protein is synthesized in response to hormonal stimulation of calcium transport and that it functions to buffer changes in intracellular calcium concentration (17, 36,  Animals-Live chicken embryos (White Leghorn) were obtained from Sunnyside Hatcheries (Oregon, WI) a t 18 days of incubation. Before use in duodenal organ culture experiments, the embryos were maintained overnight at 37 "C in a humidified incubator.
Media-Waymouth's 752/1 medium, containing 380 p~ leucine, was obtained from GIBCO (Grand Island, NY). Similar medium, containing 300 p~ I4C-leucine, was prepared according to the published formulation for Waymouth's 752/1 medium (53) by reducing the leucine concentration. Medium with 150 nM 1,25-(OH)2D3 was prepared by adding the hormone in ethanol. Equivalent volumes of ethanol alone were added to control medium. In all cases, the ethanol concentration in the medium never exceeded 0.1%. Penicillin (50 units/ml) and streptomycin (50 pg/Fl), also obtained from GIBCO, were added to all media before use. Duodenal Organ Cultures-Twenty-seven duodena excised from 19-day-old embryos were cultured for 24 h by the method of Corradino (45) as modified by Franceschi (23). All duodena were cultured in Waymouth's 752/1 medium for the first 23% h and in similar medium containing ['4C]leucine for the remaining 30 min. One group of three duodena was exposed to 150 nM 1,25-(OH)zDs3 for the last 20 h of incubation, while seven other groups of three duodena were exposed to this concentration of hormone for the last l/z, 1, 1% 2, 2' /2, 4, or 6 h. A final group of three duodena was never exposed to the hormone and served as a control group. At the end of the culture period, all duodena were individually rinsed with 4 "C buffer (50 mM PO,,150 mM NaC1, pH 7.4), placed in separate vials, and frozen on dry ice.
Preparation of Cytosolic Extracts-Each radiolabeled duodenum was transferred to a Potter-Elvehjem homogenizer and individually homogenized in 1.5 ml of buffer (5 mM Tris, pH 7.4). The homogenates were centrifuged at 100,000 X g for 45 min a t 4 "C and the supernatants containing the extracted cytosolic proteins were recovered. Incorporation of ["C]leucine into newly synthesized proteins was estimated by analysis of protein precipitated by addition of trichloroacetic acid. An aliquot of each extract was combined with an equal volume of 10% trichloroacetic acid prechilled to 4 "C and the mixture was allowed to stand for 10 min. The precipitated proteins were washed twice with 5% trichloroacetic acid and dissolved in 0.2 N NaOH. Aliquots of this solution were analyzed for protein content by the Lowry method (54) and for radioactivity by liquid scintillation spectrometry. Radiolabel incorporation was found to be 808 X lo3 L 40 X IO3 disintegrations/min/mg of precipitated protein for the 27 extracts. Electrophoresis and Autoradiography-Each cytosolic extract was analyzed for total protein content by the Lowry method (54), lyophilized, and resuspended in sodium dodecyl sulfate-free lysis buffer (55) to a final concentration of 6 mg/ml. A 25-p1 sample of each extract (150 pg of protein; 1.2 X lo5 I4C dpm) was resolved in two dimensions using the O'Farrell method of polyacrylamide gel electrophoresis (55) by Kendrick Laboratories (Madison, WI). Isoelectric focusing was carried out a t 400 V for 12% h and 800 V for the following 45 min, and molecular weight separation was completed in 9% acrylamide gels (120 X 150 X 0.75 mm). To reduce the loss of protein during electrophoresis, the isoelectric focusing gels were loaded directly onto the second dimension gels without pre-equilibration in Buffer 0 (55), and the second dimension gels were dried under vacuum without prior fixing or staining. All gels were exposed to nonpreflashed Kodak X-Omat AR films for 10 and 40 days at 22 "c.
Calibration strips were included on the films to allow the conversion of film density to 14C disintegrations. These strips were 1/2 inch wide vertical slices of a dried slab gel composed of 10 horizontal bands. All bands contained a uniform level of radioactivity/unit area and were formed from 9% solutions of acrylamide containing progressively greater concentrations of 14C-labeled protein. The level of "C contained in each band was determined by cutting out replicate polyacrylamide discs with a 7-mm diameter cork borer, burning each 3 A concentration of 150 nM 1,25-(OH)zD3 is approximately one order of magnitude greater than the minimum concentration required to elicit a maximal calcium-uptake response in these organ cultures (23).
" -disc in a model 306 TriCarb sample oxidizer (Packard Instrument Co.), and measuring the released "C02 by liquid scintillation spectrometry. The measured radioactivity, adjusted for recovery from the sample oxidizer and for counting efficiency, was expressed as '*C disintegrations/O.Ol mm2 area of gel/lO or 40 days.
Image Processing-Autoradiographs produced from the two-dimensional gels were digitized by an Optronics PI000 rotating drum densitometer (Chelmsford, MA) connected t o a tape drive via an online microprocessor. The density of each 0.01 mm2 of film area (pixel) was converted to an eight bit grey scale value ranging from 0 to 255 (0.0-3.0 optical density units) and recorded on magnetic tape. Digitized images were entered into the Man-computer Interactive Data Access System a t the Space Science and Engineering Center, University of Wisconsin, Madison, WI and displayed on an image terminal. The Man-computer Interactive Data Access System, an IBM 4341 (mod 1, OS/VS 1) with two dual density tape drives and 8 640 Mbyte disks, has terminals which can display an image of 480 X 640 pixels. These terminals have the capability of overlaying graphics on the image and sequencing a series of images in a loop. They also have a variable size cursor, controlled by joysticks, which enables the operator to examine the digital data a t any image location. The film density produced by each band of a calibration strip was determined by accessing the grey scale values for a 10 X 10 array of pixels within the bands image and calculating the mean and standard error. Disintegrations of 14C contained within the calcium-binding protein spots were determined as described under "Results." The boundary for each calcium-binding protein spot was defined as mean local film background plus two standard deviations except in film areas where the absence of base-line separation from neighboring protein spots necessitated a subjective estimation of the boundary's position. RESULTS A simple organ culture experiment, was completed in order to determine how soon calcium-binding protein biosynthesis can be detected by direct autoradiography after administration of 1,25-(OH)2D3. Duodena were excised from 19-day-old embryos and cultured for 24 h by a method similar to that published by Corradino (45). Groups of three duodena were exposed to medium containing ~,~F I -( O H )~D~ a t various times before the end of the culture period, and were pulsed with ['4C]leucine during the final 30 min. A control group of duodena, not exposed to the hormone, was similarly pulsed with ['4C]leucine. Cytosolic proteins were extracted from each radiolabeled duodenum by gentle homogenization and subsequent recovery of the supernatant remaining after ultracentrifugation. An aliquot of each protein extract was resolved by two-dimensional polyacrylamide gel electrophoresis, and the resulting gels were securely positioned by Kodak X-Omat AR film for 10 and 40 days.
An autoradiograph produced by a 10-day exposure to one of the gels is shown in Fig. 1. This film was exposed to a gel containing cytosolic proteins extracted from a duodenum cultured with 1,25-(OH)2D3 for 20 h. The grey scale standards to the left of the gel image were produced by calibration strips containing known amounts of radioactivity. Calcium-binding protein, identified by the arrow at the lower right, was located by reference to molecular weight standards and by co-migration with purified calcium-binding protein as previously described (51). The lower right portion of an autoradiograph produced by a &day exposure to this same gel is included in Fig. 2. This autoradiograph (Fig. 2a) is displayed with the comparable portions of eight other 40-day films showing proteins from duodena cultured with 1,25-(OH)& for 6-0 h ( Fig. 2, b-i). These films show that calcium-binding protein, identified by arrows, was detectable in all duodena cultured with hormone for 1 h or more. Calcium-binding protein, however, was not detectable in duodena cultured with hormone for 112 h or in control duodena not exposed to the hormone.
In order to determine the amount of calcium-binding pro- Autoradiograph of a "C-labeled two-dimensional gel. "C-labeled cytosolic proteins were extracted from a duodenum exposed to 1,25-(OH)2Da for 20 h and resolved by two-dimensional polyacrylamide gel electrophoresis. The resulting gel was exposed to x-ray film for 10 days to produce the autoradiograph displayed (see "Experimental Procedures"). The molecular weights of the resolved proteins range from 220,000 (top) to 17,000 (bottom), and the isoelectric points range from 8 (left) to 4 (right). Calcium-binding protein is located at the lower right (arrow) and grey scale standards produced from "C-calibration strips are located at the extreme left.
tein produced during the 30-min labeling period, the autoradiographs were converted to digital images by a scanning densitometer. The density of each 0.01 mm2 unit of film area (pixel) was assigned a grey scale value ranging from 0 to 255, corresponding to an optical density of 0-3.0. Each pixel's grey scale value was further converted by computer to disintegrations of I4C by reference to the grey scale standards included on the films (Fig. 3). The total radioactivity incorporated into calcium-binding protein was determined by integrating the I4C disintegrations calculated for each pixel contained within the calcium-binding protein spot. Total radioactivity was converted to picograms/mg of cytosolic protein by knowledge of the specific activity of [L4C]leucine in the culture medium, the amino acid composition and molecular weight of calciumbinding protein (56), and the amount of cytosolic protein applied to each gel, as summarized in Table I. This conversion relies on the assumptions that 1) the intracellular pool of nonradiolabeled leucine is insignificant relative to that of [14C]leucine during the 30-min labeling period, and 2) there is little loss of calcium-binding protein during two-dimensional electrophore~is.~ As shown in Fig. 4, the amount of calcium-binding protein synthesized in the 30-min labeling period is small during the first 2% h of culturing with 1,25-(OH)2Ds. Duodena cultured with hormone for 1-2% h synthesized calcium-binding protein at a rate of 129-365 pg/mg of cytosolic protein/30 min. These rates were greater (p < 0.05) than those determined for control duodena not exposed to the hormone. As shown in Fig. 5, the rate of calcium-binding protein biosynthesis rose to nearly 9000 pg/mg of cytosolic protein/30 min after 20 h of continuous exposure to hormone. The increase in biosynthetic rate was linear with time after the 1% h time point, indicating that the total amount of calcium-binding protein produced by the embryonic duodenum increased exponentially with continuous exposure to hormone.
From these data, an estimate of total calcium-binding protein production could be calculated as a function of incubation Losses of calcium-binding protein during two-dimensional electrophoresis have been determined to be less than 20% under the conditions used in this experiment (N. C. Kendrick, C. W. Bishop, and H. F. DeLuca, unpublished results).

FIG. 2. Early time course of calcium-binding protein induction by 1,25-(0H)*D~.
The "C-labeled proteins seen in these 40day autoradiographs were extracted from duodena cultured with 1,25-(OHhD3 for 20 (a), 6 (b), 4 (c), 2% ( 4 , 2 (e), 1% 0, 1 k), 1/2 (h), and 0 h (i). Calcium-binding protein (arrows) can be detected in all duodena cultured with hormone for 1 h or more. Close inspection of these films reveals that calcium-binding protein appears as a cluster of two or more closely migrating proteins. These multiple forms of intestinal calcium-binding protein are presently being examined in our laboratory to determine whether they truly exist within the embryonic duodenum (presumably to effect some physiological function) or are produced artifactually during tissue fractionation and subsequent electrophoresis. (See "Experimental Procedures" for technical detail.) time in hormone (Fig. 6). This calculation was accomplished with the equation where CaBP,,, is the total production of calcium-binding protein from the moment of initial exposure to hormone (picogram/mg of cytosolic protein), R is the increase in biosynthetic rate of calcium-binding protein (947 pg/mg of cytosolic protein/h? see legend for Fig. 5), and t is the incubation time with 1,25-(OH)2Ds (h). The value for t was reduced by 1.5 in this equation because the rate of calcium-binding protein biosynthesis did not increase linearly until 1.5 h after hormone administration. Total calcium-binding production after 20 h of exposure to 150 nM 1,25-(OH)2D3 was estimated to be 160 ng/mg of cytosolic protein. After 4 h of exposure to hormone, just before the onset of the calcium uptake response (23, 511, total calcium-binding protein production was estimated to be 3 ng/mg of cytosolic protein.

Course
of Calcium-binding Protein Induction DISCUSSION nuclear localization of 1,25-(OH)2D3 is half-maximal in chick As an extension to our previous work (51), this experiment demonstrates that calcium-binding protein is synthesized in the duodena of 19-day-old chick embryos within the first hour of exposure to 1,25-(OH)*D3. As this protein is a gene product of 1,25-(OH)2D3 action (34)(35)(36)(37)(38)(39)(40), we conclude that the hormone reaches the nuclei of the enterocytes rapidly enough to initiate gene transcription, nuclear release of mRNA for calciumbinding protein, and polysomal translation of this mRNA within 1 h. This conclusion is consistent with the finding that Calibration strips containing 10 different levels of "C radioactivity were exposed with the "C-labeled two-dimensional gels to x-ray films. These strips produced bands of different densities on the autoradiographs (see Fig. 1) which allowed film density to be precisely related to I4C disintegrations for both 10 ( duodena by 15-30 min (7, 11); but it is inconsistent with reports of calcium-binding protein induction which show, without exception, that the synthesis of this protein begins no earlier than 2 h after hormone administration (17, 29, 32, Previous studies of the time course of calcium-binding protein induction were based on the Chelex ion exchange assay (33) or on various immunoassays. The Chelex assay is characterized by poor sensitivity and early detection of the protein was better accomplished with the more sensitive immunoassays. By analyzing duodenal cytosolic extracts with a radial immunodiffusion assay, calcium-binding protein biosynthesis was found to start at 4-12 h after pharmacological levels of vitamin DB or 1,25-(OH)2D3 were administered to chicks (17,41,58) and embryonic organ cultures (29,45,60). Small amounts of calcium-binding protein were also found in similarly prepared extracts at 6 h by radioimmunoassay (40,49) and at 5-8 h by rocket immunoelectrophoresis (47,48). All three of these assays had roughly comparable sensitivities: the lower limit of detection was 5-32 ng/sample for radial immunodiffusion (61,62), 1-5 ng/sample for the radioimmunoassay (49,63), and 20 ng/sample for immunoelectrophoresis (64). To detect calcium-binding protein more sensitively, Spencer and coworkers (36,37,47) administered 125-(OH)PDB to rachitic chicks and, at subsequent intervals, examined the ability of duodenal polysomes to synthesize the protein in a cell-free system. Calcium-binding protein was consistently found among the translation products at 2 h by immunoprecipitation and subsequent sodium dodecyl sulfatepolyacrylamide gel electrophoresis. Using similar procedures, Siebert and coworkers (40) detected the protein at 3 h, which was the earliest time point they examined. By improving the sensitivity of an immunocytochemical technique with which calcium-binding protein was found in duodena at 6-8 h after 1,25-(OH)zD3 administration (57), Taylor (64) was also able ~~ 40- 49,[57][58][59][60].

Rate of calcium-binding protein biosynthesis in duodena exposed to 1,25-(oH)zD3 for different lengths of time
Eight groups of three duodena were cultured with 1,25-(OH)2D3 for the last X-20 h of a 24-b culture period and were pulsed with [14C]Leu during the final 30 min of incubation, A control group of duodena, not exposed to the hormone, was similarly pulsed with [14C]Leu. Cytosolic proteins were extracted from each radiolabeled duodenum, separated by two-dimensional polyacrylamide gel electrophoresis, and detected by direct autoradiography (see "Experimental Procedures"). The amount of calcium-binding protein synthesized during the 30-min pulse was calculated from the "C disintegrations incorporated into the protein as determined by computer analysis of the autoradiographs. These calculations are summarized below.  Fig. 3) and integrating the disintegrations calculated for all pixels within the calcium-binding protein spot.
* Calculated by dividing 14C disintegrations/min incorporated into calcium-binding protein by the specific activity of [14C]Leu in the culture medium (7.46 X 10" dpm/mmol of Leu) and by the number of leucine residues in the calcium-binding protein molecule (29 mmol of Leu/mmol of calcium-binding protein), and multiplying the result by the molecular weight of calcium-binding protein (2.70 X l O I 3 pg/mmol of calcium-binding protein).  Table I. Synthesis of ["C] calcium-binding protein is increased (p < 0.05) in all duodena exposed to hormone for 1 h or more relative to the controls.  Table I. The rate of ('4C]calcium-binding protein synthesis rises linearly with time after 1% h. The increase in the biosynthetic rate was calculated to be 947 t 4 pg/mg of cytosolic protein/hz by doubling the slope of the best fit line drawn through the 1%-20 h time points (n = 6). The equation of the line is y = (473 f 2) (X) -710 ( r = 0.999).
Each tine point was calculated as described under "Results" and represents the estimated mean & S.E.
to detect the presence of protein by 2-2% h.
We inferred from these time course studies that calciumbinding protein is rapidly synthesized after hormone administration in both the 4-week-old chick and the embryo, but its early detection is strongly dependent on the sensitivity of the assay selected (51). To test this inference, we developed a highly sensitive autoradiographic assay that is capable of detecting changes in the biosynthetic rate of any protein and applied it to the study of calcium-binding protein induction.
The assay has a lower detection limit of 10 pg of calciumbinding protein/sample (Table I) which is 2-3 orders of magnitude more sensitive than the three immunoassays previously used. Use of the assay revealed that calcium-binding protein was not detectable at 1/2 h after hormone administration, but was detectable at 1 h in an amount too small to be detected by immunoassay (129 pg/mg of cytosolic protein). If calciumbinding protein induction follows an identical time course in both the 4-week-old chick and embryo, we believe that the inconsistency between our findings regarding calcium-binding protein induction and those of other investigators (17,29,32,(40)(41)(42)(43)(44)(45)(46)(47)(48)(49)(57)(58)(59)(60) stems from the differences in assay sensitivity.
We expect that our findings in the embryo may agree with the results of a similar experiment now in progress with the rachitic chick. The cultured embryonic duodenum is indistinguishable from the rachitic chick duodenum by three criteria (23,45,60). First, calcium uptake is saturable in both tissues and is maximal in low concentrations of sodium. Second, the relative abilities of vitamin D3 metabolites and analogs to elicit calcium uptake and calcium-binding protein responses are similar in both tissues. Third, the receptor protein found in the embryonic duodenum is identical with that found in the chick duodenum with regard to sedimentation coefficient, equilibrium dissociation constant, and affinity for various vitamin D3 analogs. These similarities indicate that calciumbinding protein biosynthesis begins in the rachitic chick duodenum within 1 h of 1,25-(OH)*D3 administration, as it does in the embryonic duodenum.
By comparison, studies in the rachitic chick have shown that 1,25-(OH)& does not elicit a significant increase in calcium transport until 2 or more hours after hormone administration. The onset of the calcium transport response after a dose of 1,25-(OH)& has been demonstrated at 2-5 h by monitoring blood levels of orally administered 45Ca (8,15,17,48), 6 h by measuring ileal calcium fluxes in a Ussing-type chamber (7), 2%-6 h by in situ techniques in the duodenum (17,46,65), 2 h by duodenal accumulation of * T a (17), and 2 h by the everted ileal sac technique (47, 48). Spencer and coworkers (47,48) cite their findings with everted sacs as evidence that the hormonal response begins at 1 h since a trend towards increased calcium transport is apparent at this time; however, these investigators were unable to detect a statistically meaningful increase until 2 h. Matsumoto and coworkers (66) have similarly shown a rise in calcium uptake by brush-border membrane vesicles prepared from chicks given hormone 1 h before killing, but the statistical significance was not reported.
Since 1,25-(OH)2D3 stimulation of calcium transport is observed in rachitic chicks before calcium-binding protein is detectable by immunoassay, some investigators have concluded that the protein is not involved in the initiation of the calcium transport process (17, [46][47][48][49]. Postulating that calcium-binding protein may alternatively function to buffer changes in intracellular calcium, these researchers have hypothesized that the protein is synthesized in response to increases in calcium transport (17,36,(47)(48)(49)(50). Our previous work with the embryonic duodenum (51) has demonstrated that this hypothesis cannot be correct. Using an autoradiographic technique similar to the one described above, we demonstrated that calcium-binding protein biosynthesis begins within 2 h of exposure to 1,25-(OH)2D3 and precedes the onset of the calcium-uptake response by a t least 2 h. Our present work with the embryonic duodenum, presented here, shows that measurable amounts of calcium-binding protein are synthesized within 1 h of hormone exposure. This finding indicates that calcium-binding protein biosynthesis may also