Detection and relative quantitation of mRNA for creatine kinase isoenzymes in mRNA from myogenic cell cultures and embryonic chicken tissues.

The presence of mRNA coding for creatine kinase M (Mck) and creatine kinase B (B-CK) in RNA from myogenic and fibrogenic cell cultures, embryonic muscle, and embryonic brain tissue was demonstrated by "in vitro" translation in a heterologous cell-free protein-synthesizing system from rabbit reticulocytes. The products were isolated by sensitive immunochemical methods and their identity with isolated M-CK and B-CK was shown by the following criteria: (a) the in vitro synthesized creatine kinases react with the specific antibody against these antigens; (b) the labeled peptides co-migrate with purified creatine kinase on sodium dodecyl sulfate gels in single bands; (c) the labeled peptides form homo- and heterodimers with isolated enzymatically active creatine kinase, thus behaving like authentic creatine kinases. The assay was shown to be reproducible and gave a linear response with increasing amounts of RNA, allowing relative quantitation of mRNA in polysomal RNA for the creatine kinases M and B. MRNA for M-CK was detected in polusomal RAN and total cellular RNA from myogenic cells. It is also present in polysomal RNA from enbryonic muscle and the fraction binding to oligo(dT)-cellulose. mRNA for B-CK could be found in RNA extracted from young myogenic cultures and the fraction of polysomal embryonic brain RNA binding to oligo(dT)-cellulose.

The presence of mRNA coding for creatine kinase M (M-CK) and creatine kinase B (B-CK) in RNA from myogenic and fibrogenic cell cultures, embryonic muscle, and embryonic brain tissue was demonstrated by "in vitro" translation in a heterologous cell-free protein-synthesizing system from rabbit reticulocytes. The products were isolated by sensitive immunochemical methods and their identity with isolated M-CK and B-CK was shown by the following criteria: (a) the in vitro synthesized creatine kinases react with the specific antibody against these antigens; (6) the labeled peptides co-migrate with purified creatine kinase on sodium dodecyl sulfate gels in single bands; (c) the labeled peptides form homo-and heterodimers with isolated enzymatically active creatine kinase, thus behaving like authentic creatine kinases.
The assay was shown to be reproducible and gave a linear response with increasing amounts of RNA, allowing relative quantitation of mRNA in polysomal RNA for the creatine kinases M and B. mRNA for M-CK was detected in polysomal RNA and total cellular RNA from myogenic cells. It is also present in polysomal RNA from embryonic muscle and the fraction binding to oligo(dT)-cellulose. mRNA for B-CK could be found in RNA extracted from young myogenic cultures and the fraction of polysomal embryonic brain RNA binding to oligo(dT)-cellulose.
In the course of differentiation of myogenic tissue from the chicken embryo (1) as well as of myogenic cells, many structural and biochemical changes take place (2, 3).' Most of the studies of the regulation of the biosynthesis of cell-specific proteins have dealt with the major proteins of the contractile apparatus, the outstanding feature of differentiated muscle cells. Myosin heavy chains and light chains as well as the different forms of actin have been studied at the level of the biosynthesis of these proteins and at the level of their mRNAs (4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16). These studies were facilitated by the rather unique size of myosin heavy chain or the relative abundance of actin and light chains of myosin. For some of these proteins, however, polymorphic forms have been found not all of which are specific for myogenic cells and muscle tissue (7,(17)(18)(19)(20) and some seem to be ubiquitously distributed (see Ref. 21). Hence the advantage from the quantitative considerations cannot be * This work was supported by a grant from the Muscular Dystrophy Association of America (H. M. E.). The costs of publication of this article were defrayed in part by the payment of page charges. ?'his article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 17>4 solely to indicate this fact, ' M. Caravatti and J. C. Perriard (1978) submitted for publication. fully exploited and probably will require reinterpretation of earlier results.
Other changes that have been observed comprise the a(:cumulation of muscle-specific enzymes, which could easily be measured with the available specific assays for enzymatic activity (for review, see Refs. 2 and 3), but usually only total activity was accounted for. In the case of creatine kinase, an isoenzyme transition was observed to take place simultaneously with the differentiation of both the embryonic tissue (22) and the cultured myogenic cells (23-25)." Mononucleated cells from presumptive avian muscle and the undifferentiated tissue contain only the B form of creatine kinase while M-W' becomes the predominant form in the differentiated myotubes or tissue (25,26).' The M-CK is characteristic for differentiated muscle and is only found in significant concentrations in striated muscle (22,27). This isoenzyme transition provides a good model for one of the processes involving differential gene expression associated with myogenic terminal differentiation.
The two polypeptides are very probably the products of at least two different genes. Although the peptides have identical or very similar molecular weights of about 40,000, their amino acid composition and their tryptic peptide maps are clearly different (28). Furthermore, the antibodies against the two homomeric enzymes do not cross-react with the heterologous antigen (28,29) and can be used to distinguish between the two polypeptides.
Studies that concern differential gene expression in myogenic cells must necessarily also concentrate on the identification and the metabolism of the mRNA of such molecules. Since these mRNAs have no distinct size or structural features which can be used to separate them and furthermore, since they are expected to occur only in small concentrations in myogenic tissue or cells, we characterized these mRNAs by their property to be translated in a heterologous cell-free "in uitro" protein-synthesizing system. This allowed not only the demonstration of the distribution of the mRNA for M-CK and B-CK but also their relative quantitation, providing the basis for further studies of the levels of regulation of the creatine kinase isoenzymes. Lysis of cells was completed by 10 strokes with loose fitting pestle A in a Dounce homogenizer and the homogenate was kept in ice for 10 min before removing cellular debris by centrifugation for 15 min at 10,000 x g at 0-2'C.
The resulting supernatant was processed as described above for embryonic tissue. Oligo( dT)-cellulose Chromatography-Purification of RNA containing poly (A) tracts from embryo polysomes was carried out according to Krystosek et al. (34).

Reticulocyte
Lysates-Rabbits 2 to 3 kg were made anemic following the schedule of Villa-Komaroff et al. (35) by subcutaneous injection of a neutralized solution of 1.2% acetylphenylhydrazine starting with 2 ml at Day 1 and at the following days 1.6, 1.2, 1.6, and 2.0 ml.
At Day 1, an injection of 1 ml of 0.01% (w/v) of vitamin B12 and 0.1% (w/v) of folic acid according to Crystal et al. (36) was given intramuscularly.
Reticulocytes were stained in blood smears prepared at Day 7 or 8 with an ethanolic solution of brillant cresyl blue. The blood from animals with high reticulocyte counts (70 to 90% of the circulating cells) was obtained by terminal bleeding from the carotid artery.
Blood was collected in ice cold Reticulocyte Standard Buffer (RSB) (140 mM NaCl, 1.5 mM MgCl,, and 5 mM KCl) to which a few drops of a heparin solution (1 mg/ml) were added to prevent clotting. The cells were washed three times by repeated suspension in RSB buffer and centrifugation in a Sorvall GSA rotor at 8,000 x g, 4"C, for 10 min, the buffy coat was removed by suction and the cells were finally lysed by the addition of an equal volume of ice cold water. Membranes and other cellular debris were removed by centrifugation in a Sorvall SS-34 rotor at 23,500 x g at 0-2°C for 20 min and the resulting lysate supernatant was immediately frozen and stored in liquid nitrogen in l-ml portions. Cell-free Translation-The lysates were incubated according to the method of Pelham and Jackson (37) to yield a mRNA-dependent protein-synthesizing system. (23) and the acrylamide gel was stained for protein and processed for autoradiography as described before.

Cell-free in Vitro Translation
of M-CK and B-CK-The total cellular or polysomal RNA obtained frommyogenic cells or embryonic tissue by the CsCl/lauroyl sarcosinate centrifugation method (33) did not show any signs of degradation as judged by SDS/sucrose density gradient centrifugation or agarose-polyacrylamide gel electrophoresis (data not shown). These RNA preparations were effective in priming the synthesis of peptides in a mRNA-dependent in vitro proteinsynthesizing system derived from rabbit reticulocytes (37) and the synthesis of M-CK and B-CK could be demonstrated.

Aliquots taken from different in vitro incubations
were analyzed on 10% SDS-polyacrylamide slab gels (39) and the radioactively labeled peptides synthesized were revealed by fluorography.
In Fig. 1, the data are compiled for the comparison of the different steps. The part labeled a displays the gels stained for protein, while Part b displays the fluorograph of the same gel. Lane 1 displays a sample of a lysate treated with micrococcal nuclease but without added RNA and demonstrates that the digestion procedure effectively removes the background of peptides synthesized due to the presence of endogenous mRNA, leaving only two faintly labeled bands, already described by Pelham and Jackson (37). If, however, polysomal RNA from myogenic cell cultures was added, the system yields a host of radioactively labeled, different products, presumably reflecting all different kinds of mRNA species. The size of these products extends well above a M, of 100,000 and the autoradiography of the SDS gel (Fig. 1, Lane  2) shows the presence of peptides electrophoresing in sharp bands, indicating the synthesis of defined peptides. The M, of the subunits of creatine kinase are around 40,000 (43) and these molecules migrate, therefore, in the same region as many other peptides, like actin, synthesized in the cell-free system primed by RNA derived from myogenic cells. Preliminary experiments using direct immunoprecipitation methods to purify creatine kinase peptides had shown that some peptides unspecifically co-precipitated along with the creatine kinase and were contaminating the specific immunoprecipitate, giving rise to a substantial background. This could be prevented by treatment of incubated lysates with an antibody (rabbit anti-guinea pig serum) not reacting with any of the chicken muscular proteins and its corresponding antigen (guinea pig serum). The autoradiograph of the resulting im- munoprecipitate analyzed on a SDS gel demonstrates the unspecific precipitation of a variety of proteins, some of them migrating in the critical region of the creatine kinase peptides, around the M, of 40,000 (Fig. 1, Lane 3). We have made sure that no creatine kinase is precipitated in these immunoprecipitates by immunoreplica of SDS gels (44).' There is not even a trace of radioactively labeled M-CK or B-CK detectable as demonstrated by the autoradiographs of such immunoreplicas (not shown). The in uitro synthesized M-CK and B-CK peptides could be isolated by subsequent specific immunoprecipitation using the specific antibodies and the purified enzymes as carrier. The autoradiograph of the SDS gels shows that both species of creatine kinase were synthesized and could be specifically precipitated (Fig. 1, Lane 4). If the products were precipitated with only one kind of creatine kinase antibody only one kind of creatine kinase was precipitated (Fig. 1, Lanes 5 and 6). If, however, in all the above mentioned specific immunoprecipitation steps, preimmune serum or antiguinea pig antiserum were substituted for the specific anti-creatine kinase antibody preparation, no precipitation of creatine kinase peptides was observed (Fig. 1,  Lanes 8 and 9).
After the removal of the in vitro synthesized creatine kinase, new carrier protein and the corresponding anti-creatine kinase antibodies were added and this second immunoprecipitate was analyzed for the presence of either creatine kinase peptides. As shown in Fig. 1, Lane 10, no radioactivity is detectable by autoradiography.
We conclude that all in vitro synthesized creatine kinase was precipitated in the specific immunoprecipitation step. In Vitro Synthesized Creatine Kinase Peptides Are Genuine Creatine Kinase-As demonstrated before, peptides with some properties of creatine kinases are synthesized in a cellfree system. There are, however, some examples, especially of secretory proteins, synthesized in vitro in cell-free systems with properties slightly different from those of the authentic molecules isolated from the respective tissue as discussed in Palmiter et al. (45). We have already demonstrated that the in vitro synthesized creatine kinase react with the specific antibodies and co-migrate on SDS gels with the molecules purified from adult chicken tissues. In the following experiment, the evidence is given that the in vitro synthesized creatine kinase are capable of interacting with purified creatine kinase in the same way as the subunits of the active enzymes do.
Radioactively labeled B-CK and M-CK were prepared and isolated in the manner described above and the immunoprecipitates containing only B-CK and M-CK were treated with high ionic strength buffer to dissociate both the enzyme dimers and the immunocomplexes.
The resulting solution was divided into two parts of equal size. To one part, a lo-fold excess of homologous, enzymatically active creatine kinase was added together with twice as much of the heterologous creatine kinase species which was not present as radioactive creatine kinase. The other half of each preparation was supplemented with a 20-fold excess of the homologous enzyme present in the immunoprecipitate as the radioactive protein. The dissociated enzymes were allowed to reassociate and regain enzymatic activity by dialysis against low ionic strength buffer containing ,&mercaptoethanol.
All four samples were subsequently analyzed on polyacrylamide slab gels under nondenaturing conditions, allowing the separation of all three native, enzymatically active creatine kinase species, MM-CK, MB-CK, and BB-CK. The enzymatic activity of the samples was revealed by placing a detector gel on top of the polyacrylamide gel; after this incubation, it was stained for protein and autoradiographed.
The enzyme activity stain is shown in the right portion of Fig. 2. In Lanes 1 and 3, the enzymatically active bands of BB-CK and MM-CK coincide with the autoradiograph of the same samples and the protein staining pattern (left portion of Fig. 2), indicating that the in vitro synthesized creatine kinase molecules took part in the formation of the homologous creatine kinase species. If, however, in vitro synthesized molecules were dissociated and reassociated in the presence of both kinds of homomeric creatine kinase, the formation of all three creatine kinase species could be observed as shown in Lanes 2 and 4. Due to the interaction of the radioactively labeled peptides with the heterologous enzyme subunit, the radioactivity appeared also in the position of the heteromeric enzyme MB-CK, but never in the position of the heterologous dimeric enzyme. This experiment further indicates that the in vitro synthesized creatine kinase peptides behave like the enzymatically active, purified enzymes and take part in the interaction of the subunits. This adds addi-tional evidence that the observed peptides are identical with genuine creatine kinase subunits.
Reproducibility of Assays for mRNA Coding for Creatine Kinase and Their Quantitation-Since one of the goals of the present study was to establish methods to determine relative amounts of mRNA present in complex mixtures of RNA such as total cellular RNA or polysomal RNA, it was necessary to test the reproducibility of the assays. After the isolation of the synthesized creatine kinase peptides on SDS gels, the corresponding bands were cut out, solubilized in a liquid scintillation mixture, and analyzed for radioactivity in a liquid scintillation spectrometer. Table I shows the result of a series of assays primed by similar RNA preparations at two concentrations.
The results demonstrate that even at this low level of incorporation into creatine kinase peptides (0.1 to 0.4% of total), the observed radioactivity measurements were within a reasonable range of the average value. Again, reprecipitation of the supernatant with anti-B-CK and anti-M-CK did not yield any radioactivity in the bands of the gels representing M-CK and B-CK as was shown in the autoradiograph of a similar precipitate ( Fig. 1, Lane 10).
In the next experiment, duplicate assays were run with increasing amounts of RNA per standard reticulocyte lysate For each RNA preparation, four assays were mn in order to demonstrate the reproducibility of the procedure. Assays 1 to 4 were primed by 1.8 pg of polysomal RNA from 4-day-old cultures, Assays 5 to 8 by 1.15 ,ag of a similar preparation. The assays were run on different days, but with the same lysate and identical label concentration. assay and the M-CK and B-CK produced were isolated and their radioactivity was measured as described above. The RNA, derived from 4-day-old myogenic cell cultures obviously contained both kinds of mRNA for creatine kinase as shown in Fig. 3. The curves show a linearly increasing incorporation with increasing amount of RNA assayed and saturation of incorporation into creatine kinase around 120 pg of RNA/ml. The procedure allows, therefore, not only the detection of mRNA but also the measurements of relative amounts of mRNA species and thus will be useful in studying the regulation of gene expression in differentiating myogenic cells.
Occurrence of mRNA for M-CK and B-CK-As already shown in Fig. 1 (Lane 5) and Fig. 3 as well as in Table I, the occurrence of mRNA for both kinds of creatine kinase has been established for RNA from 4-day-old myogenic cell cultures. We wanted to test further different RNA preparations from a variety of sources for the occurrence of mRNA for the creatine kinase peptides. Total polysomal RNA was extracted from the breast muscle and brain of 14-day-old chicken em-bryos and a crude poly(A)-containing RNA fraction was prepared by chromatography on oligo(dT)-cellulose. The resulting RNA binding to the adsorbent was assayed for creatine kinase mRNA in reticulocyte lysates as described above. In Fig. 4, the fluorographs of these assays are compiled and show the mRNA for M-CK to occur in myogenic cells and embryonic muscle tissue but not in brain tissue polysomal RNA. Furthermore, the experiment established that both kinds of mRNA have a poly (A) tract as judged by the activity of the RNA retained on oligo(dT)-cellulose to code for both M-and B-CK peptides (Fig. 4, Lanes 4 and 5). Brain RNA contains only mRNA for B-CK as expected from the accumulation patterns of this tissue (27); embryonic muscle, however, contains mRNA for the M-CK and only a trace of B-CK mRNA activity is left in the total polysomal RNA from this tissue, indicating that the isoenzyme transition at the level of the mRNA is already quite advanced.
In a second experiment, we tested for the disappearance of the mRNA for M-CK after subculturing tibrogenic cells derived from fibrogenic tissue of 11-day-old chicken embryonic breast muscle. Total cellular RNA was extracted from primary fibrogenic cultures and from cells that were subcultured two 12345 -a.JB ---l M  for Chicken Creatine Kinases times. Earlier studies had shown that MM-CK and MB-CK activity was present in primary cultures of fibrogenic cells but that the M-CK subunit-containing enzymes were removed upon subculture of these cells (23) due to the selection against the myogenic cells in the subculturing process. The autoradiographs of the corresponding SDS gels show the presence of both kinds of mRNA in confluent primary fibrogenic cultures (Fig. 4, Lane 2) but the absence of the mRNA for M-CK in subcultures of the same cells (Fig. 4, Lane 3). Thus, we conclude that we are able to pick up the mRNA from contaminating cells, due to remaining myogenic cells in such fibrogenie primary cultures. Quantitative analysis demonstrated also the removal of mRNA for M-CK in the process of the subculture of fibroblast-like cells from myogenic tissue as shown in Fig. 5. Thus, the procedure is sensitive enough to pick up the mRNA derived from some contaminating myogenic cells, comprising 5 to 10% of the nuclei in myotubes, (Fig. 5a) in the primary fibroblast cultures and their removal after subculturing the cells twice (Fig. 56).

DISCUSSION
The occurrence of isoenzyme transitions concomitant with other differentiative events has been observed in a variety of tissues and organisms (see Ref. 46) and is not restricted to myogenic cells undergoing differentiation.
It is, however, interesting to note that myogenic cell cultures display a variety of isoprotein transitions. Not only enzymes like creatine kinase and aldolase (23, 47) go through such transitions but also the different forms of actin (7,19) and myosin proteins display similar features (18). Since it is rather difficult to distinguish between the different forms of some of these molecules and their mRNAs, the advantages conferred by the relatively high concentration of these species are not easy to exploit. In the case of creatine kinase however, the isoprotein molecules can be identified relatively easily, in spite of their low abundance, on the basis of their immunological and biochemical properties.
Another feature of the creatine kinases of cultured myogenic cells is that B-CK, the form of creatine kinase in young embryonic muscle, never completely disappears" in culture (25) and therefore, this molecule represents an internal control for basic cell function and allows experimental manipulation of the cells. The levels of creatine kinases in nonmyogenic tissues are much lower; for example, fibroblast cultures synthesize only about 0.031% of total cellular protein compared to 0.5% devoted to total creatine kinase synthesis of differentiated myogenic cells' and therefore do not represent a major source of mRNA for B-CK, while M-CK was almost undetectable.
The approach to the study of these mRNAs in the present communication was to use the translatability of the mRNA coding for the creatine kinase proteins, as a criterion for their occurrence. Of course, if these molecules were accumulated in some untranslatable form, we would be unable to detect these and molecular hybridization methods would be required. The purification of the mRNA coding for these enzymes and the subsequent synthesis of a cDNA probe seems within the reach of modern technology, but still represents a major effort for mRNA available in such small concentrations.
The fact that the mRNA can be assayed for in purified, but unfractionated RNA allows us to test the RNA in larger numbers of samples without the need to purify the RNA further, thus avoiding problems of recovery, poly(A) content and the completeness of retention on affinity adsorbents like oligo(dT)-cellulose or poly(U)-Sepharose and the purity of the eluted material. These considerations may be important since recent studies (48) have indicated that ,&actin mRNA fails to bind to oligo(dT)-cellulose, the mRNA for the isoproteins y-actin and oc-actin, however, are retained. Considering such heterogeneity, we thought of using unfractionated cellular or polysomal RNA in our experiments, since it is not clear that such differences are restricted to the mRNA for actin.
The criteria for the authenticity of the in vitro synthesized peptides have been described under "Experimental Procedures" and "Results," but some features of the demonstration of authenticity deserve further attention. Since the relatively abundant actins almost co-migrate with B-CK on SDS gels, we had been worried about possible contamination of the immunoprecipitates with such molecules. Pretreatment of incubated lysates with unrelated antibody and its antigen removed the contaminating molecules without precipitating any creatine kinase peptides. The fact that the two creatine kinase peptides could be resolved on 10% SDS-polyacrylamide slab gels is astonishing since earlier studies on cylindrical gels did not show a clear separation. Further details will be discussed elsewhere.' Of course, the usefulness of an assay for mRNA based on in vitro translatability is limited if the assay is either very complicated to carry out or if it is not reproducible and thus limits the application only to qualitative investigations. With the present study, we hope to show clearly that even for a mRNA species present at low concentrations, it is possible to find suitable assay conditions to allow at least a relative quantitation of the two mRNA species involved. Lodish (49) already pointed out that LY-and P-globin mRNA were translated with different efficiencies in cell-free protein-synthesizing systems. We assume that this could be the case in our system; therefore, the quantitative determinations must be considered as relative, but reproducible, if determinations are considered within one experiment measured under identical conditions.
It will depend on the availability of a purified mRNA preparation for both kinds of creatine kinase to allow the standardization of these assays, in order to calibrate the assays and give the results absolute values.
We want to apply these procedures to the study of gene expression at the level of production and accumulation of mRNA for B-CK and muscle-specific M-CK. The methods described here will provide the basis for characterizing more carefully the mRNA and the polysomes of the creatine kinase isoproteins, a prerequisite of their enrichment and future purification.