- Email: [email protected]
There is selective accumulation of a growth factor in chicken skeletal muscle
DEVELOPMENTAL
BIOLOGY
103,267-275 (1934)
There Is Selective Accumulation of a Growth Factor in Chicken Skeletal Muscle I. Transferrin Accumulation in Adult Anterior Latissimus Dorsi RYOICHI MATSUDA,’ DENNIS SPECTOR, AND R. C. STROHMAN~ Department
of Zoology, University of California, Berkeley, California 94720 Recehed August 3, 1989; accepted Janoucry 9, 1984
Chick embryo myoblaste in culture will respond to extracts of adult anterior latiseimus dorsi muscle with an increase in cell number and an increase in total protein and in myosin heavy chain in fused myotubes. Extracts of adult pectoralis major and of posterior latissimus muscles are only marginally active. The active adult muscle extracts are fractionated by DEAE-cellulose column chromatography and transferrin is identitled as the active component based on the following findings: (1) the active fractions are shown to contain an 30K protein that comigratee with chicken traneferrin on SDS-PAGE, (2) the active extract from the anterior latissimus dorsi completely replaced embryo extract in the culture medium and supported normal myogenesia, (3) the active extract requires iron for its ability to support myogenesia, (4) the peptide map of the 30K protein is identical to a peptide map of transferrin. Under conditions where the 80K protein is detected in adult anterior latissimue dorei muscles it is s,hown that the protein is nevertheless not synthesized in the muscle. These results support the idea that tissues of selective muscles in the adult chicken accumulate transferrin. An accompanying paper shows that transferrin also accumulates in early developmental stages of fast muscle tissue but that accumulation ceases after hatching in these muscles in normal chickens but not in animals of congenic strains with inherited muscular dystrophy. INTRODUCTION
In a previous communication (Matsuda et aa, 1983) we showed that the anterior latissimus dorsi muscle (ALD) of the adult chicken yielded a significantly and persistently higher number of satellite cells than could be obtained from the pectoralis major (PM) muscle of the same animal. In pursuing this matter we discovered that extracts of adult ALD but not of adult PM had a stimulatory effect on myogenesis of both ALD and PM embryonic myoblasts in cell culture. In the present communication we are able to show that this stimulatory agent in the ALD is, in all probability, identical to transferrin. Transferrin (Tf) is a normal component of chicken serum and is found as well in chick embryo extract (Ii et aL, 1982). It is an iron-binding protein and is known to be required for the growth of a great variety of cells in culture (Barnes and Sato, 1989). Chick embryo extract, a normal requirement for myogenesis in cell culture, can be completely replaced with small amounts of chicken serum and ultimately with microgram quantities of purified chicken Tf (Kimura et aL, 1982; Saito et aL, 1982). Avian Tf and mammalian Tf appear to be ’ Present address: Department of Biology, Tokyo Metropolitan versity, Setagaya-Ku, Tokyo 158, Japan. *To whom correspondence should be addressed.
Uni-
specific in their respective abilities to stimulate myogenesis in titm and mammalian Tf is ineffective in avian muscle cell cultures (Ozawa and Hagiwara, 1981; Hagiwara and Ozawa, 1982). In the absence of embryo extract but in the presence of horse serum, Tf is not only required for the growth and fusion of myoblasts, it is required as well for the maintenance of mature myotubes (Ozawa and Hagiwara, 1982). Interestingly, chick embryo cells in general are stimulated to divide by hemoglobin (Verger, 1979) and by iron salts in the absence of Tf (Imbenotte and Verger, 1980). Tf is a necessary component in a completely defined myogenic cell medium constructed by Dollenmeier et al. (1981). A myogenic trophic factor has also been isolated from sciatic nerves by Markelonis and Oh (1979). This factor, sciatin, is also in chick embryo extract (Oh and Markelonis, 1980) and is an absolute requirement for normal myogenesis. Like transferrin, it is able to replace embryo extract and, in the presence of minimum essential medium and horse serum, supports myoblast growth, cell fusion, and the development of cross-striated myotubes. Sciatin has been extensively characterized. It is a glycoprotein of approximately 84K molecular weight (Markelonis and Oh, 1981; Markelonis et d, 1980a). In addition to its myogenic-stimulating activity it also stimulates general protein synthesis in cultured muscle in a manner that is independent of muscle cell activity 267
0012-1606/34 $3.00 Copyright All rights
0 1994 by Academic Pm, Inc. of reproduction in any form mmwd.
268
DEVELOPMENTAL
BIOLOGY
and of any variation in cyclic nucleotides (Markelonis et aL, 1980b). Sciatin also increases the number of acetylcholine receptors and receptor clusters in cultured myotubes (Markelonis et c& 1982a). It has also been demonstrated that sciatin has a strong resemblance to transferrin in molecular weight, amino acid composition, immunological crossreactivity, iron-binding ability and myogenic stimulating activity (Markelonis et al, 1982b). Bischoff (1981) has reported a mitogen from crushed muscle that stimulates satellite cell growth in vitro and this factor may be related to mammalian Tf but there is no biochemical description of the factor available. The muscle growth factors described by Linkhart et al. (1981) and by Florini and Roberts (1979) are not related to transferrin. In this paper we show that a muscle trophic factor is accumulated in tissues of adult ALD, a slow muscle, but is only marginally present in two fast muscles, the pectoralis and the posterior latissimus dorsi and that this factor, in all aspects we have examined, is identical to Tf. In an accompanying paper in this issue we are also able to demonstrate, however, that this trophic factor is persistently present in fast muscles of chickens carrying the gene for muscular dystrophy. MATERIALS
AND
METHODS
Muscle ceU cultures Mononucleated cells were obtained from 12-day chicken muscle as described previously (Bandman et al, 1982). Cells were cultured in 10% horse serum (selected lots), 1.5% embryo extract, and Eagle’s minimum essential medium (MEM) at 37°C and in 5% COz at 100% humidity. The embryo extract was prepared in this laboratory from ll- to 12-day decapitated embryos (Paterson and Strohman, 1972) and was stored at -20°C until use. In some experiments the embryo extract was omitted as part of the assay for Tf. Amounts of protein, myosin heavy chain (MHC), and DNA in culture were determined as described previously (Strohman et a& 1981). Briefly, total protein was determined with a Bradford (1976) assay; MHC was first isolated on 5% SDS-PAGE and concentration determined by scanning the MHC band directly on stained gels (Bosshard and Datyner, 1977); DNA was determined according to a modified Hinegardner (1971) procedure (Bandman and Strohman, 1982). Muscle extra&a One-year-old female White Leghorn chickens were anesthetized with pentabarbital and bled extensively through neck blood vessels. Pectoralis major (PM), posterior latissimus dorsi (PLD), and anterior latissimus dorsi (ALD) muscles were excised under sterile conditions and cleaned to remove connective tissue. The minced muscle was suspended in 4 vol of MEM and centrifuged immediately at 100,OOOg for 1 hr to remove particulate matter and lipid material. Extensive
VOLUME
103, 1984
incubation of the mince or homogenization of the mince prior to centrifugation resulted in the extraction of myofibrillar proteins and complicated the measurements of myotrophic activity in the extracts. The resulting supernatants were stored at -80°C prior to use. In some cases we removed the iron from ALD extracts or from transferrin or from ALD fractions from DEAEcellulose chromatography. The solutions were dialyzed against Fe-depleted buffer (0.1 M Na acetate, 0.1 M Na phosphate, 10 mM EDTA, pH 4.5) overnight at 4°C. For saturating with iron the solutions were dialyzed against 1 mM ferric chloride, 0.1 M Na citrate, 0.1 M Na bicarbonate, pH 8.6, overnight at 4°C. The solutions were then dialyzed again against Eagle’s MEM in order to remove excess reagents. Activity could also be restored to iron-depleted extracts by simple addition of iron to dialyzed extracts. Gel electrophoresis. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed according to Laemmli (1970). Peptide mapping procedures were done as described by Cleveland et al. (1977). Purified Tf (see below) and the 80K protein (610 pg) were applied to 10% SDS gels. Following SDSPAGE the respective bands were cut from the gels and reelectrophoresed into 17.5% SDS gels in the presence of 100-250 ng of Staph. A.V8 protease (Miles). Digestion was performed in the stacking gel for 30 min at room temperature. After electrophoresis the gel was stained with Coomassie brilliant blue. DEAE-cellulose ion-exchange chromatography. DEAE-cellulose (DE 52, Whatman) was equilibrated with 10 mMTris-HCl, pH 8.0, and was packed into 1.8 X 7-cm columns. Protein (30 mg) from a 50-80% saturation ammonium sulfate fraction of muscle extract was chromatographed by elution with a lOO-ml linear gradient of O-O.4 M NaCl in 10 mM Tris-HCl, pH 8.0. Fractions of 1.5 ml were collected and protein content measured by the method of Bradford (1976). All procedures for chromatography were performed under sterile conditions at 4°C. Fractions were stored at -80°C in small aliquots prior to use. Pur$icatim of tramfmh Serum was obtained from l-year-old female White Leghorn chickens and Tf was prepared according to the method described by Ii et al. (1982). An acetone powder of serum was dissolved in 10 mM Tris-HCl, pH 8.0, and the 55-80% saturated ammonium sulfate fraction was subjected to DEAE-cellulose column chromatography as described above except that a linear gradient of O-O.2 M NaCl was used for elution. Eluted protein was estimated by absorbance at 280 nm. Purity of Tf in various fractions from the column was determined by SDS-PAGE as described in the figure legends below. Pure Tf fractions were pooled and frozen at -80°C.
Tm@rrinSelective Accumulation in NormalMuscle MATSUDA, SPECTOR, ANDSTROHMAN
269
In wivo Iabeling with [35S]methimine.Anesthetized by phasemicroscopyandPLDextractshadonlymarchickenswere usedand the ALD musclewas exposed ginal effects when total protein, myosin heavy chain under sterile conditions. Radioactive methionine (0.5 (MHC), andDNA weremeasured(Fig. 2). ALD extract mCi) in 50 ~120 mM potassium acetate,0.1%2-mer- effectswere dosedependentwith large effectsclearly captoethanol was injected directly into the ALD. The wound was stitched and after 2 hr the animal was sacrificed by bleeding. The ALD extract was prepared as described above. RESULTS
ALD Extracts
Stimulate
iklyogenesis
in Vitro
In initial experiments we measured the ability of various muscle extracts to stimulate muscle fiber formation and growth in culture in medium containing embryo extract. As shown in Fig. 1, ALD extracts, when applied to myoblast cultures from PM embryo muscle, stimulate cell number, number of myotubes, and thickness of myotubes compared to cultures matched for initial cell density and time after plating. Extracts from the PLD and PM, however, had no obvious effects when viewed
measurable at 0.5 mg/ml of extract protein in total culture medium volume. Total protein was more than doubled as were DNA values and MHC accumulation increased fourfold. ALD extracts also contained mitogenic activity when added to purified skin fibroblast cultures (data not shown) so we cannot say at this time what the increase in myoblasts might have been relative to muscle fibroblasts also present in our myogenic cell cultures. The MHC stimulation by ALD extracts in the absence of embryo extract in the culture is even more profound and will be discussed below. The ALD Maotrophic to Tramferrin
Factor Appears to Be Identical
DEAE-cellulose column chromatography of the ALD extract yielded the elution profile shown in Fig. 3. Fol-
FIG. 1. Effect of adult muscle extracts on myogenesis in vitro. Myogenic cells derived from 1Bday chick embryo pectoralis major muscle were cultured in complete medium containing extracts from (a) no extract added, (b) extract from adult PM, (c) extract from adult PLD, (d) extract from adult ALD. Extracts were prepared as described under Methods. The final concentration of each extract in the culture medium was 0.1 mg/ml. Extracts were added at the time of plating. Cells were fed on Day 4 with extracts added where appropriate and cells were taken for measurement or photographed usually at Days 6-7 when myotube formation was well developed. All extracts were obtained from muscles of l-year-old White Leghorn chickens. The same results were obtained with extracts from adult muscle of an inbred ..c-.,.:.. l-m .~.l.Lh :a -la,. Wh:+a 1 ndmm OI .tr.in 413 urhbh icr Now Hamnuhir~ The bar inrlirntpn 1IK) u,,-.
270
DEVELOPMENTAL BIOLOGY
C
as
0 Q142
45
i
0 a1 a2
FIG. 2. Dose-dependent effect of muscle extracts on myogenesis in vitro Cells were cultured as described in Fig. 1. Extracts of ALD, PLD, and PM were added at the time of cell plating and were included in feeding medium. As a control, additional horse serum (HS) was added as indicated. Cells were harvested on Day 7. (A) Amount of total protein, (B) amount of myosin heavy chain, and (C) amount of DNA. Values are for individual 60-mm dishes.
lowing elution of unabsorbed protein, a major peak appears at about fraction 29 and a second major peak at fraction 43. The first peak corresponds to the Tf-containing peak obtained by Ii et al. (1982) on fractionation of embryo extract. We tested each fraction for its ability to stimulate myogenesis in culture as outlined in Figs. 1 and 2. The only fractions showing activity are those
z I
VOLUME 103,19&i
from fractions 29 through 33. These assays were conducted in the presence of embryo extract but the results were the same and even more convincing when ALD column fractions were assayed in the absence of embryo extract (see Fig. 5). Each fraction was then subjected to SDS-PAGE in order to survey the pattern and identity of proteins present in the active versus inactive fractions. The active fractions, 29-33, show several important features compared to the inactive fractions eluting immediately before and after. First, a heavy band comigrating with chicken transferrin begins to appear at fraction 29 and is much reduced in all fractions following 33. Second, the only other bands appearing in the active fractions that are entirely missing in the inactive fractions are the lower-molecular-weight proteins marked with a double arrow in Fig. 4. The myotrophic activity could possibly then be ascribed to either the Tf or to these lower MW proteins. We have no further information on the low-molecular-weight proteins at this time. That the myogenic-stimulating factor of the ALD extract is actually Tf is based on the following information. First, purified Tf is known to replace embryo extract as a requirement for myogenesis in cell culture. The active ALD fractions also completely replace embryo extract and give over 300-fold increases in MHC com-
I &4 I 0.26 P EL”
30
40
50
60
numbor
FIG. 8. Chromatography of ALD extract on DEAE-cellulose. Approximately 80 mg protein from a 50-80% eaturated ammonium sulfate fraction of total ALD extract wae applied to the column as described in the rection on methods. The column buffer was 20 m&f Tris-HCl, pH 8.0, and a O-O.4 MNaCl linear gradient was started with fraction 12 Each fraction collected was 1.5 ml. Biological activity of each fraction was messwed as described above. The shaded area indicates those fractione with myogenic etimulating activity in cell cultures growing in complete medium.
FIG. 4. SDS-PAGE pattern of chromatographic fractions of ALD extract, Fractions 25-35 of ALD extract chromatographed as shown in the previous figure were analyzed by 12.5% SDS-PAGE; 20 rl of each fraction was applied to respective slots of the gel. Following electrophoresie the gel was stained with Coomassie brilliant blue. Purified chicken Tf was run as a marker in the slot following fraction 36.
MATSUDA,
SPECTOR,
Fraction
AND STROHMAN
!l’ransferrin
number
FIG. 5. Stimulation of myosin heavy chain accumulation by fractionated ALD extract. Two-hundred microliters of each fraction was added to breast muscle cultures growing in MEM-10% horse serum but without embryo extract. Extract was added at the time of plating and the cells were prepared for analysis on Day 5 of culture without any intervening feeding. All 60 fractions shown in Fig. 3 were analyzed. Only fractions 33-33 contained MHC stimulating activity.
Selective Accumulation
in Normal
Muscle
271
at levels detectable by our method, but is taken up from the blood circulation. We measured, therefore, the relative blood supply to the ALD compared with the PM of the adult chicken. [%]Methionine was injected intravenously and allowed to circulate for 2 min at which time the amino acid is found to be mostly in the blood and not taken up into intracellular tissue protein as determined by counts incorporated into TCA precipitates. By measuring CPM of radioactive methionine in muscle pieces we could estimate the relative blood volume. The ALD showed about three times more radioactivity per milligram wet weight of muscle than did the PM. Roughly, then, the blood supply is estimated to be three times greater in the ALD than in the PM. In addition, when we prepared extracts from identical pieces (wet weight) of ALD and PM we found that the PM extract protein concentration was about three times greater than that of the ALD. Assuming that Tf in the blood supply of the muscle is extracted together with soluble muscle protein, our ALD extracts are estimated to be roughly nine times more
pared to myogenic cells growing in horse serum and MEM but without embryo extract (Fig. 5). Second, when the major band from the ALD extract that corn&rates with chicken Tf on SDS-PAGE is prepared for peptide mapping the outcome is a peptide map that is almost if not completely identical to Tf (Fig. 6). Finally, when the active fractions are tested for myogenic stimulating activity they are found to have an absolute requirement for iron (Fig. ‘7). That is, if the fractions are dialyzed against Fe-containing buffer or buffer containing no Fe (EDTA-buffer) then only the material dialyzed against Fe displayed activity. The Source of Tranqfkrrin
in Specific Muscles
Because of the selective accumulation of Tf or Tf-like myogenic stimulating factor by ALD as opposed to the PM or the PLD we wanted to know whether the material was synthesized by the ALD. We therefore injected the living ALD muscle with radioactive methionine as described in the methods section. Following a 2-hr in tivo incubation, the ALD muscle was taken and analyzed as shown in Fig. 8. There is no radioactivity that comigrates with chicken Tf as determined by an autoradiograph of the stained gel even though Tf is clearly present in the muscle (compare slots a and b in Fig. 8). We must assume that Tf is not synthesized by ALD tissue cells, at least
FIG. 6. Peptide maps of chicken tranbferrin and of the 80K protein of the ALD extract. The two proteins were isolated from respective 10% gels of SDS-PAGE and subjected to peptide mapping as described above. Lanes a and d are maps of chicken serum Tf; lanes c and f are maps of the 80K protein of the ALD extract, and lanes b and e are maps of a mixture of the two proteins. In,a-c the Staph A.V8 proteaee concentration was 100 ng/elot and in d-f it was 350 rig/slot. T, Undigested Tf or 8OK protein; V, Staph A.V8 protease. The gels were stained with Coomaesie brilliant blue.
272
DEVELOPMENTAL
BIOLOGY
VOLUME
103, 1084
FIG. 7. Iron requirement for ALD extract activity. The total ALD extract and fraction 31 (see Fig. 3) were tested for ability to stimulate myogeneeis in cultures growing in MEM-10% horse serum but without embryo extract. Panels a and b show effect of adding purified chicken W, c and d show addition of total ALD extract, and e and f show addition of fraction 31. Prior to addition to the cultures the various proteins were first dialyzed to remove iron as described under the Methods section. Solutions a, c, and e were dialyzed in iron-containing buffer and solutions b, d, and f were dialyzed against EDTA-buffer containing no iron. Added to the cultures was 20 pg/ml of Tf or 200 pg/ml of ALD extract or 200 pg/ml of fraction 31. Cultures were allowed to grow for 5 days prior to photography.
active in blood Tf in the ALD compared with the PM. Thus the high levels of Tf found in the ALD compared with the PM could possibly be accounted for by differences in vascularization of the muscles. At the present time, however, we cannot say whether vascularization differences account for the entire’ Tf difference or whether additional uptake mechanisms also exist that serve to concentrate Tf in particular cells of the muscle tissue (see accompanying paper for immunocytochemical studies of Tf localization).
DISCUSSION
Our ALD extract and active fractions derived from it have many characteristics identical to Tf. First, the ALD extract will stimulate myogenesis in cell culture even in the presence of embryo extract (Figs. 1, 2). In the absence of embryo extract, the ALD extract yields fractions from DEAE-cellulose chromatography that completely substitute for embryo extract and that produce several-hundred-fold increases in accumulation of
MATSUDA, SPECTOR,
AND STROHMAN
!l’ransferrin
a b
Selective Accumulation
tract contains a factor
in Normal
Muscle
273
or factors in addition to Tf that
work synergisticallywith Tf to supportmusclecell growth and myogenesis
in vitro.
Our results in Fig. 2 indicate that starting at about
FIG. 8. ALD extract after in tivo [86S]methionine pulse for 2 hr and 12.5% SDS-PAGE. Radioactive methionine was injected directly into the ALD of an adult chicken as described above. After 2 hr the ALD muscle was taken for analysis. (a) Stained gel of the labeled ALD extract after SDS-PAGE; (h) autoradiogram of slot a. The arrow indicates the position of migration of a purified Tf marker. Approximately 5000 cpm of TCA insoluble extract was subjected to electrophoresis. Dried gels were then exposed to Kodak AR X-O mat film for 2 days at room temperature.
myosin heavy chain (Fig. 5). The only fractions from the chromatography that are active are ones containing distinct bands in SDS-PAGE which comigrate with Tf (Fig. 4). When these bands are analyzed they yield peptide maps that are identical to maps prepared from purified chicken Tf. Finally when the ALD extracts or active fractions from it are dialyzed to remove bound iron then the preparations lose biological activity (Fig. 7). We conclude therefore that the myogenic-stimulating activity of the ALD is in fact Tf. We were somewhat surprised to discover that the ALD extract would stimulate myogenic activity in the presence of embryo extract. The embryo extract contains Tf and normally the amount of embryo extract that we use (1.5%) is optimal for muscle growth and differentiation. We must consider therefore that the ALD ex-
0.1 mg/ml of ALD extract protein in the culture medium there is a measurable effect on myogenic cell cultures. At 0.5 mg/ml there is a 2.5-fold increase in total protein, a 3.5-fold increase in myosin heavy chain accumulation, and a 4-fold increase in DNA. Since these cultures contain fibroblasts and since the ALD extract is also somewhat mitogenic for chicken skin fibroblasts in culture (data not shown) we cannot be certain about the degree to which myoblast growth is stimulated relative to total single cell population. We do know, however, that stimulation of MHC synthesis by ALD extract is most probably due to enhanced myoblast cell division and ultimately to larger numbers of myotubes. If, for example, we add ALD extract to cultures after cell fusion has already taken place then there is no stimulation of MHC synthesis (data not shown). An alternative explanation for this finding would be that myotubes somehow lose ability to bind Tf but this is highly improbable since Ozawa and Hagiwara (1982) have reported that if Tf is withdrawn from myotube cultures the muscle degenerates. The ALD extract, as might be expected, has a profound effect on myogenesis when embryo extract is omitted from the cultures. In accordance with many previous observations (for example, Konigsberg, 1979) we note that cell division, cell fusion, and the whole of muscle cell development is completely arrested if the medium contains only MEM and horse serum. When whole ALD extract or an active fraction of ALD is added there is restoration of myogenic activity. This activity is completely dependent on the presence of iron since when either extract or fraction is dialyzed against changes of buffer containing no iron but containing EDTA the myogenic activity is abolished. The questions of how and why the ALD, and perhaps other slow muscles, accumulate Tf while fast muscles like breast and PLD either do not or do so only marginally remain unanswered. With regard to the first question, we have demonstrated in Fig. 8 that the adult ALD does not synthesize any protein comigrating on SDS-PAGE with Tf even though the same stained gel shows that the muscle has accumulated TF. We conclude that the ALD is somehow able to concentrate Tf from the blood or perhaps from its nerve supply. Clearly, chicken nerves are a source of sciatin and other molecules similar to Tf (Markelonis and Oh, 19’79; Popiela and Ellis, 1981; Stamatos et aL, 1983). The ALD, in contrast to the PM, has multiple innervation and small amounts of Tf could be delivered to the muscle via the nerve. It is interesting to note that Oh et al. (1981) did
274
DEVELOPMENTAL
BIOLOGY
VOLUME
1~,1984
not observe any immunocytochemicallocalization of and muscleactivity. If tonic muscleslike the soleusin
sciatin on muscle in viva. But they apparently restricted their observations to the fast PLD muscle. As we have shown in Fig. 2, adult chicken PLD has a small but almost insignificant Tf-like ability to stimulate muscle formation in cell culture. In addition, as we show in an accompanying paper (Matsuda et ak, 1984, this issue), extracts of adult PLD contain only minor amounts of a protein comigrating with Tf in SDS-PAGE. Assuming that Tf and sciatin are identical, this result would agree with the negative finding by Oh et al. (1981) concerning sciatin localization in the PLD muscle. We would expect therefore that immunocytochemical studies would show a high level of Tf in ALD compared to PM or PLD and indeed under conditions where the adult ALD shows heavy localization of anti-Tf around muscle fibers the adult PM is almost without reaction to the antibody (Matsuda et &, 1984, this issue). The higher degree of vascularization we have found in the ALD is a more likely explanation for the high Tf levels found in normal ALD muscle. It is useful to speculate on the resulting physiological effects of such differences in blood supply and attending high Tf levels. First of course, the higher Tf levels will support the increased need of the oxidative ALD for iron. But our results also show that Tf, or a factor that coisolates with Tf, promotes growth in myogenic cell cultures. On this basis we might also expect high Tf levels in particular muscles to support a higher growth rate for satellite cells or that muscle under appropriate conditions such as muscle injury or chronic (tonic) use. We have not yet done any histochemical survey of the ALD in which we attempt to show a correlation between Tf localization and satellite cell activity. However, Gibson and Schultz (1982,1983) have shown in therat that different muscles display wide variation in satellite cell density. Slow oxidative muscles like the soleus (rat) have significantly higher levels of satellite cells than do fast muscles like the EDL. We have made some preliminary calculations from our earlier work on satellite cell cultures derived from chicken ALD and PM muscle (Matsuda et aL, 1983) and these roughly agree with the more quantitative study of Gibson and Schultz (1983). They find approximately 33,000 satellite cells per milligram rat slow soleus muscle and 1200 satellite cells in the (mostly) fast EDL muscle. Using very rough cell counts on material cultured from adult muscle we calculate that adult chicken ALD muscle yields about 1300 satellite cells per milligram wet weight whereas the PM yields only 400. In both the rat and the chicken therefore the slow, oxidative muscle has approximately three times the number of satellite cells as the fast glycolytic muscle. The work of Gibson and Schultz (1983) also points to a clear positive correlation between satellite cell number
the rat or the ALD in the chicken, which must be used over long periods of time, are required to replace muscle mass (higher turnover rates) more frequently than fast muscles like the EDL (rat) and PM (chicken) then the tonic muscles would indeed require a greater supply of satellite cells. The relationship between vascularization and growth factor concentration in a tissue, on the one hand, to muscle differentiation and maturation, on the other, remains speculative and, in fact, this area is not one about which we are well informed. Studies reviewed by Vrbova (1980), however, make it clear that when fast muscle is stimulated with a slow muscle frequency the very first of the changes in the direction of conversion to slow muscle is growth of capillaries. Clearly we need to consider more thoughtfully the possible role of vascularization in determining the course of muscle development and maturation. As we discuss further in the accompanying paper, there are radical shifts in Tf accumulation during muscle development and there are very clear differences found between normal and dystrophic muscle with regard to Tf. Many of these differences appear to be associated with differences in single cell populations between muscle fibers and may be related as well to differences in satellite cell activity. REFERENCES E., MATSUDA, R., and STROHMAN,R. C. (1982). Developmental appearance of myoein heavy and light chain isoforms in vkvo and in vitro in chicken skeletal muscle. Dev. Bid 93, 568-518. BANDMAN, E., and STROHMAN, R. C. (1982). Increased K+ inhibits spontaneous contractions and reduces myosin accumulation in cultured chick myotubes. J. CeU Bid 93,698-704. BARNES, D., and SATO, G. (1989). Serum-free cell culture: A unifying approach. Cell 22.649-655. BISCHOFF, R. (1981). Activation and proliferation of muscle satellite cells on isolated fibers. J. CeU Bid 91,342a. BOSSARD, H. F., and DATYNER, A. (1977). The use of a new reactive dye for quantitation of prestained proteins on polyacrylamide gels. BANDMAN,
And
Bimhem.
33,327-333.
BRADFORD,M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye-binding. And Biochsm 72, 248-252. DOLLENYEIER. P., TURNER, D. C., and EPPENBERGER, H. M. (1981). Proliferation and differentiation of chick skeletal muscle cells cultured in a chemically defined medium. Exp. Cell Rea 136,47-61. FLORINI, J. R., and ROBERTS, S. B. (1979). A serum free medium for the growth of muscle cells in culture. In Vitro 16.983-992. GIBSON, M. C., and SCHIJL~~, E. (1982). The distribution of satellite cells and their relationship to specific fiber types in soleua and extensor digitorum longus muscles. Anat. Rec. 202,329-337. GIBBON, M. C., and SCHULZ, E. (1983). Age-related differencee in absolute numbers of skeletal muscle satellite cells. Muscle Nerve 6, 574-580. HAGIWARA, Y., and OZAWA, E. (1982). Class specificity of avian and mammalian sera in regards to myogenic cell growth in vitro. Deu. &o~th Diger. 24(l), 115-123. HINEGARDNER, T. T. (1971). An improved fluorometric assay for DNA. Amcl
Bioch
39.197-261.
MATSUDA, SPECTOR, AND STROHMAN
TransferA
II, I., KIMURA, I., and OZAWA, E. (1982). A myotrophic protein from chick embryo extract: Its purification, identity to transferrin, and indizpenzibility for avian myogenesis. De-v. Bid 94,36&3’77. IYBENOTTE, J., and VERGER, C. (1980). Nature of the iron requirement for chick embryo cells cultured in the presence of horse serum. Cell BioL Int. Rep. 4.447-452. KIMURA, I., HASEGAWA, T., and OZAWA, E. (1982). Indispensability of iron-bound transferrin for chick myogenesis in vitro. Den Grozuth D@r. 24(4), 369-380. KONIGSBERG, I. R. (1976). The role of the environment in the control of myogenesis in vitro. In “Pathogenesis of Human Muscular Dystrophy” (L. P. Rowland, ed.), pp. 7’79-798. Excerpta Medica, Amsterdam. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Landon) 277. 680-685. LINKHART, T. A., CLEGG, C. H., and HAUSCHKA, S. D. (1981). Myogenic differentiation in permanent clonal mouse myoblast cell lines: Regulation by macromolecular growth factors in the culture medium. Lkv. Bid 86,19-3(X MARKELONIS, G., and OH, T. H. (1979). A sciatic nerve protein has a trophic effect on development and maintenance of skeletal muscle cells in culture. PTOC Natl. AC&. Sci. USA 76,2470-2474. MARKELONIS, G., KEMERER, V. F., and OH, T. H. (1989a). Sciatin: Purification and characterization of a myotrophic protein from chicken sciatic nerve. J. Bid Chem 266,8967-8970. MARKELONIS, G., OH, T. H., and DE& D. (1989b). Stimulation of protein synthesis in cultured skeletal muscle by a trophic protein from sciatic nerves. Exp. NewoL 70,598-612. MARKELONIS, G., and OH, T. H. (1981). Purification of seiatin using affinity chromatography on concanavalin A-agarose. J. Neurochem 37.95-99. MARKELONIS, G., OH, T. H., ELDEFRAWI, M. E., and GUTH, L. (1982a). Sciatin: A myotrophic protein increases the number of acetylcholine receptors and receptor clusters in cultured skeletal muscle. Deu. Bid 89,353-361. MARKELONIS, G., BRADSHAW, R. A., OH, T. H., JOHNSON, J. L., and BATES, 0. J. (1982b). Sciatin is a transferrin-like polypeptide. J. Neurochem 39,315-329.
Selective Accumulation
in Normal
Muscle
275
MATSUDA, R., SPECTOR,D. H., and STROHMAN,R. C. (1983). Regenerating adult chicken skeletal muscle and satellite cell cultures express embryonic patterns of myosin and tropomyosin isoforms. Dev. BioL 199,4’78-488. MATSUDA, R., SPECTOR,D. H., MICOU-EASTWOOD, J., and STROHMAN, R. C. (1984). There is selective accumulation of a growth factor in chicken skeletal muscle. II. Transferrin accumulation in dystrophic fast muscle. L)ev. Biol 103, 276-284. OH, T. H., and MARKELONIS, G. (1989). Dependence of in vitro myogenesis on a trophic factor present in chicken embryo extract. Proc. NatL Ad. Sci USA 77,6922-6925. OH, T. H., SOFIA, C. A., KIM, Y. C., CARROLL,C., KIM, H. H., MARKELONIS, G., and REIER, P. J. (1981). Sciatin: Immunocytochemical localization of a myotrophic protein in chicken neural tissues. J. Hz&x& Q/to&m. 29,1205-1212. OZAWA, E., and HAGIWARA, Y. (1981). Avian and mammalian transferrins are required for chick and rat myogenic cell growth in vitro, respectively. Proc Japan Ad Ser. B 67,406-409. OZAWA, E., and HAGIWARA, Y. (1982). Degeneration of large myotubes following removal of transferrin from culture medium. Bimed Rex 3(l), 16-23. POPIELA, H., and ELLIS, S. (1981). Neurotrophic factor: Characterization and partial isolation. Dev. BioL 33.266-277. SAITO, K., HAGIWARA, Y., HASEGAWA, T., and OZAWA, E. (1982). Indispensibility of iron for the growth of chick cells. Den Gmzuth D@r. 24(6), 571-580. STAMATOS, C., SQUICCIARINI,J., and FINE, R. E. (1983). Chick embryo spinal cord neurons synthesize a transferrin-like myotrophic protein. FEBS Lett. 153,387-390. STROHMAN, R. C., BANDMAN, E., and WALKER, C. (1981). Regulation of myosin accumulation by muscle activity in cell culture. J. Musck Res. Cell Motil 2,269-282. VERGER, C. (1979). Proliferation and morphology of chick embryo cells cultured in the presence of horse serum and hemoglobin. In Vitro l&587-592. VRBOVA, G. (1989). Innervation and differentiation of muscle fibers. In “Development and Specialization of Skeletal Muscle” (D. F. Goldspink, ed.). Cambridge Univ. Press, London.
BIOLOGY
103,267-275 (1934)
There Is Selective Accumulation of a Growth Factor in Chicken Skeletal Muscle I. Transferrin Accumulation in Adult Anterior Latissimus Dorsi RYOICHI MATSUDA,’ DENNIS SPECTOR, AND R. C. STROHMAN~ Department
of Zoology, University of California, Berkeley, California 94720 Recehed August 3, 1989; accepted Janoucry 9, 1984
Chick embryo myoblaste in culture will respond to extracts of adult anterior latiseimus dorsi muscle with an increase in cell number and an increase in total protein and in myosin heavy chain in fused myotubes. Extracts of adult pectoralis major and of posterior latissimus muscles are only marginally active. The active adult muscle extracts are fractionated by DEAE-cellulose column chromatography and transferrin is identitled as the active component based on the following findings: (1) the active fractions are shown to contain an 30K protein that comigratee with chicken traneferrin on SDS-PAGE, (2) the active extract from the anterior latissimus dorsi completely replaced embryo extract in the culture medium and supported normal myogenesia, (3) the active extract requires iron for its ability to support myogenesia, (4) the peptide map of the 30K protein is identical to a peptide map of transferrin. Under conditions where the 80K protein is detected in adult anterior latissimue dorei muscles it is s,hown that the protein is nevertheless not synthesized in the muscle. These results support the idea that tissues of selective muscles in the adult chicken accumulate transferrin. An accompanying paper shows that transferrin also accumulates in early developmental stages of fast muscle tissue but that accumulation ceases after hatching in these muscles in normal chickens but not in animals of congenic strains with inherited muscular dystrophy. INTRODUCTION
In a previous communication (Matsuda et aa, 1983) we showed that the anterior latissimus dorsi muscle (ALD) of the adult chicken yielded a significantly and persistently higher number of satellite cells than could be obtained from the pectoralis major (PM) muscle of the same animal. In pursuing this matter we discovered that extracts of adult ALD but not of adult PM had a stimulatory effect on myogenesis of both ALD and PM embryonic myoblasts in cell culture. In the present communication we are able to show that this stimulatory agent in the ALD is, in all probability, identical to transferrin. Transferrin (Tf) is a normal component of chicken serum and is found as well in chick embryo extract (Ii et aL, 1982). It is an iron-binding protein and is known to be required for the growth of a great variety of cells in culture (Barnes and Sato, 1989). Chick embryo extract, a normal requirement for myogenesis in cell culture, can be completely replaced with small amounts of chicken serum and ultimately with microgram quantities of purified chicken Tf (Kimura et aL, 1982; Saito et aL, 1982). Avian Tf and mammalian Tf appear to be ’ Present address: Department of Biology, Tokyo Metropolitan versity, Setagaya-Ku, Tokyo 158, Japan. *To whom correspondence should be addressed.
Uni-
specific in their respective abilities to stimulate myogenesis in titm and mammalian Tf is ineffective in avian muscle cell cultures (Ozawa and Hagiwara, 1981; Hagiwara and Ozawa, 1982). In the absence of embryo extract but in the presence of horse serum, Tf is not only required for the growth and fusion of myoblasts, it is required as well for the maintenance of mature myotubes (Ozawa and Hagiwara, 1982). Interestingly, chick embryo cells in general are stimulated to divide by hemoglobin (Verger, 1979) and by iron salts in the absence of Tf (Imbenotte and Verger, 1980). Tf is a necessary component in a completely defined myogenic cell medium constructed by Dollenmeier et al. (1981). A myogenic trophic factor has also been isolated from sciatic nerves by Markelonis and Oh (1979). This factor, sciatin, is also in chick embryo extract (Oh and Markelonis, 1980) and is an absolute requirement for normal myogenesis. Like transferrin, it is able to replace embryo extract and, in the presence of minimum essential medium and horse serum, supports myoblast growth, cell fusion, and the development of cross-striated myotubes. Sciatin has been extensively characterized. It is a glycoprotein of approximately 84K molecular weight (Markelonis and Oh, 1981; Markelonis et d, 1980a). In addition to its myogenic-stimulating activity it also stimulates general protein synthesis in cultured muscle in a manner that is independent of muscle cell activity 267
0012-1606/34 $3.00 Copyright All rights
0 1994 by Academic Pm, Inc. of reproduction in any form mmwd.
268
DEVELOPMENTAL
BIOLOGY
and of any variation in cyclic nucleotides (Markelonis et aL, 1980b). Sciatin also increases the number of acetylcholine receptors and receptor clusters in cultured myotubes (Markelonis et c& 1982a). It has also been demonstrated that sciatin has a strong resemblance to transferrin in molecular weight, amino acid composition, immunological crossreactivity, iron-binding ability and myogenic stimulating activity (Markelonis et al, 1982b). Bischoff (1981) has reported a mitogen from crushed muscle that stimulates satellite cell growth in vitro and this factor may be related to mammalian Tf but there is no biochemical description of the factor available. The muscle growth factors described by Linkhart et al. (1981) and by Florini and Roberts (1979) are not related to transferrin. In this paper we show that a muscle trophic factor is accumulated in tissues of adult ALD, a slow muscle, but is only marginally present in two fast muscles, the pectoralis and the posterior latissimus dorsi and that this factor, in all aspects we have examined, is identical to Tf. In an accompanying paper in this issue we are also able to demonstrate, however, that this trophic factor is persistently present in fast muscles of chickens carrying the gene for muscular dystrophy. MATERIALS
AND
METHODS
Muscle ceU cultures Mononucleated cells were obtained from 12-day chicken muscle as described previously (Bandman et al, 1982). Cells were cultured in 10% horse serum (selected lots), 1.5% embryo extract, and Eagle’s minimum essential medium (MEM) at 37°C and in 5% COz at 100% humidity. The embryo extract was prepared in this laboratory from ll- to 12-day decapitated embryos (Paterson and Strohman, 1972) and was stored at -20°C until use. In some experiments the embryo extract was omitted as part of the assay for Tf. Amounts of protein, myosin heavy chain (MHC), and DNA in culture were determined as described previously (Strohman et a& 1981). Briefly, total protein was determined with a Bradford (1976) assay; MHC was first isolated on 5% SDS-PAGE and concentration determined by scanning the MHC band directly on stained gels (Bosshard and Datyner, 1977); DNA was determined according to a modified Hinegardner (1971) procedure (Bandman and Strohman, 1982). Muscle extra&a One-year-old female White Leghorn chickens were anesthetized with pentabarbital and bled extensively through neck blood vessels. Pectoralis major (PM), posterior latissimus dorsi (PLD), and anterior latissimus dorsi (ALD) muscles were excised under sterile conditions and cleaned to remove connective tissue. The minced muscle was suspended in 4 vol of MEM and centrifuged immediately at 100,OOOg for 1 hr to remove particulate matter and lipid material. Extensive
VOLUME
103, 1984
incubation of the mince or homogenization of the mince prior to centrifugation resulted in the extraction of myofibrillar proteins and complicated the measurements of myotrophic activity in the extracts. The resulting supernatants were stored at -80°C prior to use. In some cases we removed the iron from ALD extracts or from transferrin or from ALD fractions from DEAEcellulose chromatography. The solutions were dialyzed against Fe-depleted buffer (0.1 M Na acetate, 0.1 M Na phosphate, 10 mM EDTA, pH 4.5) overnight at 4°C. For saturating with iron the solutions were dialyzed against 1 mM ferric chloride, 0.1 M Na citrate, 0.1 M Na bicarbonate, pH 8.6, overnight at 4°C. The solutions were then dialyzed again against Eagle’s MEM in order to remove excess reagents. Activity could also be restored to iron-depleted extracts by simple addition of iron to dialyzed extracts. Gel electrophoresis. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed according to Laemmli (1970). Peptide mapping procedures were done as described by Cleveland et al. (1977). Purified Tf (see below) and the 80K protein (610 pg) were applied to 10% SDS gels. Following SDSPAGE the respective bands were cut from the gels and reelectrophoresed into 17.5% SDS gels in the presence of 100-250 ng of Staph. A.V8 protease (Miles). Digestion was performed in the stacking gel for 30 min at room temperature. After electrophoresis the gel was stained with Coomassie brilliant blue. DEAE-cellulose ion-exchange chromatography. DEAE-cellulose (DE 52, Whatman) was equilibrated with 10 mMTris-HCl, pH 8.0, and was packed into 1.8 X 7-cm columns. Protein (30 mg) from a 50-80% saturation ammonium sulfate fraction of muscle extract was chromatographed by elution with a lOO-ml linear gradient of O-O.4 M NaCl in 10 mM Tris-HCl, pH 8.0. Fractions of 1.5 ml were collected and protein content measured by the method of Bradford (1976). All procedures for chromatography were performed under sterile conditions at 4°C. Fractions were stored at -80°C in small aliquots prior to use. Pur$icatim of tramfmh Serum was obtained from l-year-old female White Leghorn chickens and Tf was prepared according to the method described by Ii et al. (1982). An acetone powder of serum was dissolved in 10 mM Tris-HCl, pH 8.0, and the 55-80% saturated ammonium sulfate fraction was subjected to DEAE-cellulose column chromatography as described above except that a linear gradient of O-O.2 M NaCl was used for elution. Eluted protein was estimated by absorbance at 280 nm. Purity of Tf in various fractions from the column was determined by SDS-PAGE as described in the figure legends below. Pure Tf fractions were pooled and frozen at -80°C.
Tm@rrinSelective Accumulation in NormalMuscle MATSUDA, SPECTOR, ANDSTROHMAN
269
In wivo Iabeling with [35S]methimine.Anesthetized by phasemicroscopyandPLDextractshadonlymarchickenswere usedand the ALD musclewas exposed ginal effects when total protein, myosin heavy chain under sterile conditions. Radioactive methionine (0.5 (MHC), andDNA weremeasured(Fig. 2). ALD extract mCi) in 50 ~120 mM potassium acetate,0.1%2-mer- effectswere dosedependentwith large effectsclearly captoethanol was injected directly into the ALD. The wound was stitched and after 2 hr the animal was sacrificed by bleeding. The ALD extract was prepared as described above. RESULTS
ALD Extracts
Stimulate
iklyogenesis
in Vitro
In initial experiments we measured the ability of various muscle extracts to stimulate muscle fiber formation and growth in culture in medium containing embryo extract. As shown in Fig. 1, ALD extracts, when applied to myoblast cultures from PM embryo muscle, stimulate cell number, number of myotubes, and thickness of myotubes compared to cultures matched for initial cell density and time after plating. Extracts from the PLD and PM, however, had no obvious effects when viewed
measurable at 0.5 mg/ml of extract protein in total culture medium volume. Total protein was more than doubled as were DNA values and MHC accumulation increased fourfold. ALD extracts also contained mitogenic activity when added to purified skin fibroblast cultures (data not shown) so we cannot say at this time what the increase in myoblasts might have been relative to muscle fibroblasts also present in our myogenic cell cultures. The MHC stimulation by ALD extracts in the absence of embryo extract in the culture is even more profound and will be discussed below. The ALD Maotrophic to Tramferrin
Factor Appears to Be Identical
DEAE-cellulose column chromatography of the ALD extract yielded the elution profile shown in Fig. 3. Fol-
FIG. 1. Effect of adult muscle extracts on myogenesis in vitro. Myogenic cells derived from 1Bday chick embryo pectoralis major muscle were cultured in complete medium containing extracts from (a) no extract added, (b) extract from adult PM, (c) extract from adult PLD, (d) extract from adult ALD. Extracts were prepared as described under Methods. The final concentration of each extract in the culture medium was 0.1 mg/ml. Extracts were added at the time of plating. Cells were fed on Day 4 with extracts added where appropriate and cells were taken for measurement or photographed usually at Days 6-7 when myotube formation was well developed. All extracts were obtained from muscles of l-year-old White Leghorn chickens. The same results were obtained with extracts from adult muscle of an inbred ..c-.,.:.. l-m .~.l.Lh :a -la,. Wh:+a 1 ndmm OI .tr.in 413 urhbh icr Now Hamnuhir~ The bar inrlirntpn 1IK) u,,-.
270
DEVELOPMENTAL BIOLOGY
C
as
0 Q142
45
i
0 a1 a2
FIG. 2. Dose-dependent effect of muscle extracts on myogenesis in vitro Cells were cultured as described in Fig. 1. Extracts of ALD, PLD, and PM were added at the time of cell plating and were included in feeding medium. As a control, additional horse serum (HS) was added as indicated. Cells were harvested on Day 7. (A) Amount of total protein, (B) amount of myosin heavy chain, and (C) amount of DNA. Values are for individual 60-mm dishes.
lowing elution of unabsorbed protein, a major peak appears at about fraction 29 and a second major peak at fraction 43. The first peak corresponds to the Tf-containing peak obtained by Ii et al. (1982) on fractionation of embryo extract. We tested each fraction for its ability to stimulate myogenesis in culture as outlined in Figs. 1 and 2. The only fractions showing activity are those
z I
VOLUME 103,19&i
from fractions 29 through 33. These assays were conducted in the presence of embryo extract but the results were the same and even more convincing when ALD column fractions were assayed in the absence of embryo extract (see Fig. 5). Each fraction was then subjected to SDS-PAGE in order to survey the pattern and identity of proteins present in the active versus inactive fractions. The active fractions, 29-33, show several important features compared to the inactive fractions eluting immediately before and after. First, a heavy band comigrating with chicken transferrin begins to appear at fraction 29 and is much reduced in all fractions following 33. Second, the only other bands appearing in the active fractions that are entirely missing in the inactive fractions are the lower-molecular-weight proteins marked with a double arrow in Fig. 4. The myotrophic activity could possibly then be ascribed to either the Tf or to these lower MW proteins. We have no further information on the low-molecular-weight proteins at this time. That the myogenic-stimulating factor of the ALD extract is actually Tf is based on the following information. First, purified Tf is known to replace embryo extract as a requirement for myogenesis in cell culture. The active ALD fractions also completely replace embryo extract and give over 300-fold increases in MHC com-
I &4 I 0.26 P EL”
30
40
50
60
numbor
FIG. 8. Chromatography of ALD extract on DEAE-cellulose. Approximately 80 mg protein from a 50-80% eaturated ammonium sulfate fraction of total ALD extract wae applied to the column as described in the rection on methods. The column buffer was 20 m&f Tris-HCl, pH 8.0, and a O-O.4 MNaCl linear gradient was started with fraction 12 Each fraction collected was 1.5 ml. Biological activity of each fraction was messwed as described above. The shaded area indicates those fractione with myogenic etimulating activity in cell cultures growing in complete medium.
FIG. 4. SDS-PAGE pattern of chromatographic fractions of ALD extract, Fractions 25-35 of ALD extract chromatographed as shown in the previous figure were analyzed by 12.5% SDS-PAGE; 20 rl of each fraction was applied to respective slots of the gel. Following electrophoresie the gel was stained with Coomassie brilliant blue. Purified chicken Tf was run as a marker in the slot following fraction 36.
MATSUDA,
SPECTOR,
Fraction
AND STROHMAN
!l’ransferrin
number
FIG. 5. Stimulation of myosin heavy chain accumulation by fractionated ALD extract. Two-hundred microliters of each fraction was added to breast muscle cultures growing in MEM-10% horse serum but without embryo extract. Extract was added at the time of plating and the cells were prepared for analysis on Day 5 of culture without any intervening feeding. All 60 fractions shown in Fig. 3 were analyzed. Only fractions 33-33 contained MHC stimulating activity.
Selective Accumulation
in Normal
Muscle
271
at levels detectable by our method, but is taken up from the blood circulation. We measured, therefore, the relative blood supply to the ALD compared with the PM of the adult chicken. [%]Methionine was injected intravenously and allowed to circulate for 2 min at which time the amino acid is found to be mostly in the blood and not taken up into intracellular tissue protein as determined by counts incorporated into TCA precipitates. By measuring CPM of radioactive methionine in muscle pieces we could estimate the relative blood volume. The ALD showed about three times more radioactivity per milligram wet weight of muscle than did the PM. Roughly, then, the blood supply is estimated to be three times greater in the ALD than in the PM. In addition, when we prepared extracts from identical pieces (wet weight) of ALD and PM we found that the PM extract protein concentration was about three times greater than that of the ALD. Assuming that Tf in the blood supply of the muscle is extracted together with soluble muscle protein, our ALD extracts are estimated to be roughly nine times more
pared to myogenic cells growing in horse serum and MEM but without embryo extract (Fig. 5). Second, when the major band from the ALD extract that corn&rates with chicken Tf on SDS-PAGE is prepared for peptide mapping the outcome is a peptide map that is almost if not completely identical to Tf (Fig. 6). Finally, when the active fractions are tested for myogenic stimulating activity they are found to have an absolute requirement for iron (Fig. ‘7). That is, if the fractions are dialyzed against Fe-containing buffer or buffer containing no Fe (EDTA-buffer) then only the material dialyzed against Fe displayed activity. The Source of Tranqfkrrin
in Specific Muscles
Because of the selective accumulation of Tf or Tf-like myogenic stimulating factor by ALD as opposed to the PM or the PLD we wanted to know whether the material was synthesized by the ALD. We therefore injected the living ALD muscle with radioactive methionine as described in the methods section. Following a 2-hr in tivo incubation, the ALD muscle was taken and analyzed as shown in Fig. 8. There is no radioactivity that comigrates with chicken Tf as determined by an autoradiograph of the stained gel even though Tf is clearly present in the muscle (compare slots a and b in Fig. 8). We must assume that Tf is not synthesized by ALD tissue cells, at least
FIG. 6. Peptide maps of chicken tranbferrin and of the 80K protein of the ALD extract. The two proteins were isolated from respective 10% gels of SDS-PAGE and subjected to peptide mapping as described above. Lanes a and d are maps of chicken serum Tf; lanes c and f are maps of the 80K protein of the ALD extract, and lanes b and e are maps of a mixture of the two proteins. In,a-c the Staph A.V8 proteaee concentration was 100 ng/elot and in d-f it was 350 rig/slot. T, Undigested Tf or 8OK protein; V, Staph A.V8 protease. The gels were stained with Coomaesie brilliant blue.
272
DEVELOPMENTAL
BIOLOGY
VOLUME
103, 1084
FIG. 7. Iron requirement for ALD extract activity. The total ALD extract and fraction 31 (see Fig. 3) were tested for ability to stimulate myogeneeis in cultures growing in MEM-10% horse serum but without embryo extract. Panels a and b show effect of adding purified chicken W, c and d show addition of total ALD extract, and e and f show addition of fraction 31. Prior to addition to the cultures the various proteins were first dialyzed to remove iron as described under the Methods section. Solutions a, c, and e were dialyzed in iron-containing buffer and solutions b, d, and f were dialyzed against EDTA-buffer containing no iron. Added to the cultures was 20 pg/ml of Tf or 200 pg/ml of ALD extract or 200 pg/ml of fraction 31. Cultures were allowed to grow for 5 days prior to photography.
active in blood Tf in the ALD compared with the PM. Thus the high levels of Tf found in the ALD compared with the PM could possibly be accounted for by differences in vascularization of the muscles. At the present time, however, we cannot say whether vascularization differences account for the entire’ Tf difference or whether additional uptake mechanisms also exist that serve to concentrate Tf in particular cells of the muscle tissue (see accompanying paper for immunocytochemical studies of Tf localization).
DISCUSSION
Our ALD extract and active fractions derived from it have many characteristics identical to Tf. First, the ALD extract will stimulate myogenesis in cell culture even in the presence of embryo extract (Figs. 1, 2). In the absence of embryo extract, the ALD extract yields fractions from DEAE-cellulose chromatography that completely substitute for embryo extract and that produce several-hundred-fold increases in accumulation of
MATSUDA, SPECTOR,
AND STROHMAN
!l’ransferrin
a b
Selective Accumulation
tract contains a factor
in Normal
Muscle
273
or factors in addition to Tf that
work synergisticallywith Tf to supportmusclecell growth and myogenesis
in vitro.
Our results in Fig. 2 indicate that starting at about
FIG. 8. ALD extract after in tivo [86S]methionine pulse for 2 hr and 12.5% SDS-PAGE. Radioactive methionine was injected directly into the ALD of an adult chicken as described above. After 2 hr the ALD muscle was taken for analysis. (a) Stained gel of the labeled ALD extract after SDS-PAGE; (h) autoradiogram of slot a. The arrow indicates the position of migration of a purified Tf marker. Approximately 5000 cpm of TCA insoluble extract was subjected to electrophoresis. Dried gels were then exposed to Kodak AR X-O mat film for 2 days at room temperature.
myosin heavy chain (Fig. 5). The only fractions from the chromatography that are active are ones containing distinct bands in SDS-PAGE which comigrate with Tf (Fig. 4). When these bands are analyzed they yield peptide maps that are identical to maps prepared from purified chicken Tf. Finally when the ALD extracts or active fractions from it are dialyzed to remove bound iron then the preparations lose biological activity (Fig. 7). We conclude therefore that the myogenic-stimulating activity of the ALD is in fact Tf. We were somewhat surprised to discover that the ALD extract would stimulate myogenic activity in the presence of embryo extract. The embryo extract contains Tf and normally the amount of embryo extract that we use (1.5%) is optimal for muscle growth and differentiation. We must consider therefore that the ALD ex-
0.1 mg/ml of ALD extract protein in the culture medium there is a measurable effect on myogenic cell cultures. At 0.5 mg/ml there is a 2.5-fold increase in total protein, a 3.5-fold increase in myosin heavy chain accumulation, and a 4-fold increase in DNA. Since these cultures contain fibroblasts and since the ALD extract is also somewhat mitogenic for chicken skin fibroblasts in culture (data not shown) we cannot be certain about the degree to which myoblast growth is stimulated relative to total single cell population. We do know, however, that stimulation of MHC synthesis by ALD extract is most probably due to enhanced myoblast cell division and ultimately to larger numbers of myotubes. If, for example, we add ALD extract to cultures after cell fusion has already taken place then there is no stimulation of MHC synthesis (data not shown). An alternative explanation for this finding would be that myotubes somehow lose ability to bind Tf but this is highly improbable since Ozawa and Hagiwara (1982) have reported that if Tf is withdrawn from myotube cultures the muscle degenerates. The ALD extract, as might be expected, has a profound effect on myogenesis when embryo extract is omitted from the cultures. In accordance with many previous observations (for example, Konigsberg, 1979) we note that cell division, cell fusion, and the whole of muscle cell development is completely arrested if the medium contains only MEM and horse serum. When whole ALD extract or an active fraction of ALD is added there is restoration of myogenic activity. This activity is completely dependent on the presence of iron since when either extract or fraction is dialyzed against changes of buffer containing no iron but containing EDTA the myogenic activity is abolished. The questions of how and why the ALD, and perhaps other slow muscles, accumulate Tf while fast muscles like breast and PLD either do not or do so only marginally remain unanswered. With regard to the first question, we have demonstrated in Fig. 8 that the adult ALD does not synthesize any protein comigrating on SDS-PAGE with Tf even though the same stained gel shows that the muscle has accumulated TF. We conclude that the ALD is somehow able to concentrate Tf from the blood or perhaps from its nerve supply. Clearly, chicken nerves are a source of sciatin and other molecules similar to Tf (Markelonis and Oh, 19’79; Popiela and Ellis, 1981; Stamatos et aL, 1983). The ALD, in contrast to the PM, has multiple innervation and small amounts of Tf could be delivered to the muscle via the nerve. It is interesting to note that Oh et al. (1981) did
274
DEVELOPMENTAL
BIOLOGY
VOLUME
1~,1984
not observe any immunocytochemicallocalization of and muscleactivity. If tonic muscleslike the soleusin
sciatin on muscle in viva. But they apparently restricted their observations to the fast PLD muscle. As we have shown in Fig. 2, adult chicken PLD has a small but almost insignificant Tf-like ability to stimulate muscle formation in cell culture. In addition, as we show in an accompanying paper (Matsuda et ak, 1984, this issue), extracts of adult PLD contain only minor amounts of a protein comigrating with Tf in SDS-PAGE. Assuming that Tf and sciatin are identical, this result would agree with the negative finding by Oh et al. (1981) concerning sciatin localization in the PLD muscle. We would expect therefore that immunocytochemical studies would show a high level of Tf in ALD compared to PM or PLD and indeed under conditions where the adult ALD shows heavy localization of anti-Tf around muscle fibers the adult PM is almost without reaction to the antibody (Matsuda et &, 1984, this issue). The higher degree of vascularization we have found in the ALD is a more likely explanation for the high Tf levels found in normal ALD muscle. It is useful to speculate on the resulting physiological effects of such differences in blood supply and attending high Tf levels. First of course, the higher Tf levels will support the increased need of the oxidative ALD for iron. But our results also show that Tf, or a factor that coisolates with Tf, promotes growth in myogenic cell cultures. On this basis we might also expect high Tf levels in particular muscles to support a higher growth rate for satellite cells or that muscle under appropriate conditions such as muscle injury or chronic (tonic) use. We have not yet done any histochemical survey of the ALD in which we attempt to show a correlation between Tf localization and satellite cell activity. However, Gibson and Schultz (1982,1983) have shown in therat that different muscles display wide variation in satellite cell density. Slow oxidative muscles like the soleus (rat) have significantly higher levels of satellite cells than do fast muscles like the EDL. We have made some preliminary calculations from our earlier work on satellite cell cultures derived from chicken ALD and PM muscle (Matsuda et aL, 1983) and these roughly agree with the more quantitative study of Gibson and Schultz (1983). They find approximately 33,000 satellite cells per milligram rat slow soleus muscle and 1200 satellite cells in the (mostly) fast EDL muscle. Using very rough cell counts on material cultured from adult muscle we calculate that adult chicken ALD muscle yields about 1300 satellite cells per milligram wet weight whereas the PM yields only 400. In both the rat and the chicken therefore the slow, oxidative muscle has approximately three times the number of satellite cells as the fast glycolytic muscle. The work of Gibson and Schultz (1983) also points to a clear positive correlation between satellite cell number
the rat or the ALD in the chicken, which must be used over long periods of time, are required to replace muscle mass (higher turnover rates) more frequently than fast muscles like the EDL (rat) and PM (chicken) then the tonic muscles would indeed require a greater supply of satellite cells. The relationship between vascularization and growth factor concentration in a tissue, on the one hand, to muscle differentiation and maturation, on the other, remains speculative and, in fact, this area is not one about which we are well informed. Studies reviewed by Vrbova (1980), however, make it clear that when fast muscle is stimulated with a slow muscle frequency the very first of the changes in the direction of conversion to slow muscle is growth of capillaries. Clearly we need to consider more thoughtfully the possible role of vascularization in determining the course of muscle development and maturation. As we discuss further in the accompanying paper, there are radical shifts in Tf accumulation during muscle development and there are very clear differences found between normal and dystrophic muscle with regard to Tf. Many of these differences appear to be associated with differences in single cell populations between muscle fibers and may be related as well to differences in satellite cell activity. REFERENCES E., MATSUDA, R., and STROHMAN,R. C. (1982). Developmental appearance of myoein heavy and light chain isoforms in vkvo and in vitro in chicken skeletal muscle. Dev. Bid 93, 568-518. BANDMAN, E., and STROHMAN, R. C. (1982). Increased K+ inhibits spontaneous contractions and reduces myosin accumulation in cultured chick myotubes. J. CeU Bid 93,698-704. BARNES, D., and SATO, G. (1989). Serum-free cell culture: A unifying approach. Cell 22.649-655. BISCHOFF, R. (1981). Activation and proliferation of muscle satellite cells on isolated fibers. J. CeU Bid 91,342a. BOSSARD, H. F., and DATYNER, A. (1977). The use of a new reactive dye for quantitation of prestained proteins on polyacrylamide gels. BANDMAN,
And
Bimhem.
33,327-333.
BRADFORD,M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye-binding. And Biochsm 72, 248-252. DOLLENYEIER. P., TURNER, D. C., and EPPENBERGER, H. M. (1981). Proliferation and differentiation of chick skeletal muscle cells cultured in a chemically defined medium. Exp. Cell Rea 136,47-61. FLORINI, J. R., and ROBERTS, S. B. (1979). A serum free medium for the growth of muscle cells in culture. In Vitro 16.983-992. GIBSON, M. C., and SCHIJL~~, E. (1982). The distribution of satellite cells and their relationship to specific fiber types in soleua and extensor digitorum longus muscles. Anat. Rec. 202,329-337. GIBBON, M. C., and SCHULZ, E. (1983). Age-related differencee in absolute numbers of skeletal muscle satellite cells. Muscle Nerve 6, 574-580. HAGIWARA, Y., and OZAWA, E. (1982). Class specificity of avian and mammalian sera in regards to myogenic cell growth in vitro. Deu. &o~th Diger. 24(l), 115-123. HINEGARDNER, T. T. (1971). An improved fluorometric assay for DNA. Amcl
Bioch
39.197-261.
MATSUDA, SPECTOR, AND STROHMAN
TransferA
II, I., KIMURA, I., and OZAWA, E. (1982). A myotrophic protein from chick embryo extract: Its purification, identity to transferrin, and indizpenzibility for avian myogenesis. De-v. Bid 94,36&3’77. IYBENOTTE, J., and VERGER, C. (1980). Nature of the iron requirement for chick embryo cells cultured in the presence of horse serum. Cell BioL Int. Rep. 4.447-452. KIMURA, I., HASEGAWA, T., and OZAWA, E. (1982). Indispensability of iron-bound transferrin for chick myogenesis in vitro. Den Grozuth D@r. 24(4), 369-380. KONIGSBERG, I. R. (1976). The role of the environment in the control of myogenesis in vitro. In “Pathogenesis of Human Muscular Dystrophy” (L. P. Rowland, ed.), pp. 7’79-798. Excerpta Medica, Amsterdam. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Landon) 277. 680-685. LINKHART, T. A., CLEGG, C. H., and HAUSCHKA, S. D. (1981). Myogenic differentiation in permanent clonal mouse myoblast cell lines: Regulation by macromolecular growth factors in the culture medium. Lkv. Bid 86,19-3(X MARKELONIS, G., and OH, T. H. (1979). A sciatic nerve protein has a trophic effect on development and maintenance of skeletal muscle cells in culture. PTOC Natl. AC&. Sci. USA 76,2470-2474. MARKELONIS, G., KEMERER, V. F., and OH, T. H. (1989a). Sciatin: Purification and characterization of a myotrophic protein from chicken sciatic nerve. J. Bid Chem 266,8967-8970. MARKELONIS, G., OH, T. H., and DE& D. (1989b). Stimulation of protein synthesis in cultured skeletal muscle by a trophic protein from sciatic nerves. Exp. NewoL 70,598-612. MARKELONIS, G., and OH, T. H. (1981). Purification of seiatin using affinity chromatography on concanavalin A-agarose. J. Neurochem 37.95-99. MARKELONIS, G., OH, T. H., ELDEFRAWI, M. E., and GUTH, L. (1982a). Sciatin: A myotrophic protein increases the number of acetylcholine receptors and receptor clusters in cultured skeletal muscle. Deu. Bid 89,353-361. MARKELONIS, G., BRADSHAW, R. A., OH, T. H., JOHNSON, J. L., and BATES, 0. J. (1982b). Sciatin is a transferrin-like polypeptide. J. Neurochem 39,315-329.
Selective Accumulation
in Normal
Muscle
275
MATSUDA, R., SPECTOR,D. H., and STROHMAN,R. C. (1983). Regenerating adult chicken skeletal muscle and satellite cell cultures express embryonic patterns of myosin and tropomyosin isoforms. Dev. BioL 199,4’78-488. MATSUDA, R., SPECTOR,D. H., MICOU-EASTWOOD, J., and STROHMAN, R. C. (1984). There is selective accumulation of a growth factor in chicken skeletal muscle. II. Transferrin accumulation in dystrophic fast muscle. L)ev. Biol 103, 276-284. OH, T. H., and MARKELONIS, G. (1989). Dependence of in vitro myogenesis on a trophic factor present in chicken embryo extract. Proc. NatL Ad. Sci USA 77,6922-6925. OH, T. H., SOFIA, C. A., KIM, Y. C., CARROLL,C., KIM, H. H., MARKELONIS, G., and REIER, P. J. (1981). Sciatin: Immunocytochemical localization of a myotrophic protein in chicken neural tissues. J. Hz&x& Q/to&m. 29,1205-1212. OZAWA, E., and HAGIWARA, Y. (1981). Avian and mammalian transferrins are required for chick and rat myogenic cell growth in vitro, respectively. Proc Japan Ad Ser. B 67,406-409. OZAWA, E., and HAGIWARA, Y. (1982). Degeneration of large myotubes following removal of transferrin from culture medium. Bimed Rex 3(l), 16-23. POPIELA, H., and ELLIS, S. (1981). Neurotrophic factor: Characterization and partial isolation. Dev. BioL 33.266-277. SAITO, K., HAGIWARA, Y., HASEGAWA, T., and OZAWA, E. (1982). Indispensibility of iron for the growth of chick cells. Den Gmzuth D@r. 24(6), 571-580. STAMATOS, C., SQUICCIARINI,J., and FINE, R. E. (1983). Chick embryo spinal cord neurons synthesize a transferrin-like myotrophic protein. FEBS Lett. 153,387-390. STROHMAN, R. C., BANDMAN, E., and WALKER, C. (1981). Regulation of myosin accumulation by muscle activity in cell culture. J. Musck Res. Cell Motil 2,269-282. VERGER, C. (1979). Proliferation and morphology of chick embryo cells cultured in the presence of horse serum and hemoglobin. In Vitro l&587-592. VRBOVA, G. (1989). Innervation and differentiation of muscle fibers. In “Development and Specialization of Skeletal Muscle” (D. F. Goldspink, ed.). Cambridge Univ. Press, London.