THE JOURNAL OF B~LOGICAL CHEMISTRY Vol. 251. No. 5. Issue of March 10. pp. 1505-lR10, 1976

Printed in U.S.A.

Identification and Subcellular Distribution of Adipocyte Peptides and Phosphopeptides”

(Received for publication, March 3, 1975)

JOSEPH AVRUCH,$ Gw R. LEONE, AND DONALD B. MARTIN

From the Diabetes Unit, Massachusetts General Hospital, Boston, Massachusetts, and the Department of Medicine, Harvard Medical School, Boston, Massachusetts 02114

Subcellular fractions of high purity (including plasma membrane, endoplasmic reticulum, mito- chondria, nuclei, and cytoplasm) were prepared from isolated adipocytes, and the peptide components were examined by detergent gel electrophoresis. Each fraction except the endoplasmic reticulum exhib- ited a unique and reproducible complement of major peptides. Although the endoplasmic reticulum was distinctive in its enzymic mark&s, its peptide components showed striking homologies with certain species in the plasma membrane and cytoplasm. The two major adipocyte glycopeptides appear to be contained in the plasma membrane, inasmuch as they followed the distribution of 5’-nucleotidase.

Incubation of adipocytes with extracellular $*P, led to a uniform rate of incorporation of szP into cellular peptides, with steady-state incorporation reached by 2 hours. Plasma membrane, mitochondria, nuclei, and cytoplasm all contained a distinctive complement of from two to five major phosphopeptides of different molecular weights. The majority of endoplasmic reticulum phosphopeptides exhibited molec- ular weights closely similar tb those of certain species in the plasma membrane and cytoplasm. The phosphopeptides of the plasma membrane exhibited the highest absolute s?p incorporation of all phos- phopeptides, next was the single major mitochondrial phosphopeptide. All fractions except the mito- chondria contained, in addition to the few major phosphopeptides, numerous minor **P-labeled phospho- peptides.

Phosphopeptides are widely distributed in animal tissues (1). Although no satisfactory functional classification is availa- ble, certainly the phosphorylenzyme intermediates (2) must be distinguished from the bulk of cellular phosphoproteins. The latter are both enzymic and nonenzymic in function, contain from one to several moles of phosphate per mole of protein, and their turnover is mediated by protein kinases and protein phosphatases of varying specificity. This class of phosphopep- tides is of great interest for two reasons. First, phosphorylation is known to greatly modify the functional properties of several proteins (3). Second, with the description of a cyclic AMP’- dependent protein kinase in 1968 (4), it became clear that the regulation of certain of these phospho-dephospho conversions was the site of hormonal control of glycogen metabolism, as well as of adipose tissue lipolysis. These findings, taken together with the widespread tissue distribution and apparent broad substrate specificity of cyclic AMP-dependent protein kinases (5), have led to the suggestion that many (perhaps all) of the effects of cyclic AMP on cellular function in mammalian cells are mediated by alterations in protein phosphorylation. Much of the data adduced in support of this hypothesis is indirect, and our knowledge of cellular phosphoprotein me-

* This work was suphrted by Grants AM 13774 and 17776 from the National Institutes of Health and by a grant from The John A. Hartford Foundation, Inc.

$ Investigator of the Howard Hughes Medical Institute. ‘The abbreviations used are: cyclic AMP, adenosine 3’:5’-mono-

phosphate; PAS, periodic acid-Schiff.

tabolism is fragmentary. We undertook the study of this problem in the isolated adipocyte. This report will describe modifications of existing techniques that allow the reproduci- ble isolation of the major subcellular fractions of adipose cells with high purity. In addition, we will describe the peptide and glycopeptide complements of these fractions as revealed by polyacrylamide gel electrophoresis in sodium dodecyl sulfate. Finally, the phosphopeptides generated by adipocytes incu- bated with NaH,**PO, will be identified and catalogued by molecular weight and subcellular distribution, and methods for their measurement will be compared.

MATERIALS AND METHODS

Male Sprague-Dawley rats, 100 to 125 g, were obtained from Gofmoor Farms and used within 2 weeks. Isolated fat cells were prepared according to Rcdbell (6). The adipocytes were washed five times in Krebs-Ringer bicarbonate buffer (with bovine serum albumin, 1.7 g/100 ml) prepared without inorganic phosphate. Twelve to

eighteen milliliters of packed cells (derived from 30 to 40 rats) were suspended in 40 ml of Krebs-Ringer bicarbonate buffer minus inor- ganic phosphate. D-Glucose was added to an extracellular concentra- tion of 5 mM. NaHf’PO, (5 mCi in 0.5 ml of water, 0.5 Ci/mmol) was added to the suspension yielding an inorganic phosphate concentration of approximately 0.25 mM and a radiochemical concentration of approximately 125 &i/ml. After gassing with 5% COJ95Y 0, the suspension was incubated at 37” with gentle shaking. Aliquots of the suspension were removed at intervals, and the cells were separated from the medium by centrifugation for 30 s. The cells then received two cycles of resuspension and centrifugation in 0.25 M sucrose, 0.01 M Tris-chloride (pH 7.4), and 0.001 M EDTA (Medium I) and were

1505

by guest on February 7, 2018http://w

ww

.jbc.org/D

ownloaded from

1506 Adipocyte Peptides and Phosphopeptides

resuspended in this medium at 4’. The washing, performed at room temperature, required 4 to 5 min. The adipocytes were then subfrac- tionated according to McKee1 and Jarett (7). modified as follows. The washed fat cells, suspended in 10 to 20 ml of iced Medium I, were homogenized 10 strokes at 1,800 rpm (Thomas 3431-E25). The suspen- sion was centrifuged at 16,000 x g for 15 min. yielding a fat cake (which was removed and discarded), a supernatant (Sl), and a pellet (Pl). Sl was centrifuged at 100,000 x g for 2 hours. The supematant thus obtained was designated the cytosol fraction; the pellet was resus- pended in Medium I and designated the endoplasmic reticulum fraction.

Pl was resuspended in 8 ml of Medium I, homogenized 6 strokes at 1,000 rpm (Thomas 3431-E15), and centrifuged at 16,000 x g for 20 min. yielding a supematant (S2), which was discarded, and a pellet (P2). P2 was suspended in 2 to 8 ml of Medium I, homogenized 6 strokes at 1,500 rpm (Thomas 3431-E15). and layered onto replicate, linear sucrose gradients (p = 1.1 to 1.2) containing Tris-chloride (0.01 M, pH 7.4) and EDTA (0.001 M). The gradients were centrifuged at 40,000 rpm in an SW 41 rotor for 45 min. After centrifugation, the plasma membrane (p = 1.10 to 1.14), mitochondria (near p = 1.18). and nuclear pellet were diluted with 7 to 8 volumes of Medium I and resedimented at 40,000 x g for 30 min. The nuclear fraction was subjected to brief sonication.

Polyacrylamide gel electrophoresis in 1% sodium dodecyl sulfate and staining for peptides and glycopeptides were performed according to Fairbanks et a[. (81. Samples were nrenared for electrouhoresis bv the addition of the following (final concentration): protein (0.25 to 1.0 mg/ml), sucrose (5 to lo%), dithiothreitol (40 mru), and sodium dodecyl sulfate (1%). Samples were incubated at 37” for 30 min. (Samples prepared at lOO”, on electrophoresis, exhibited peptide staining patterns identical with those of samples prepared at 37”.) Portions containing 50 pg of protein were subjected to electrophoresis on cylindrical gels (6-mm diameter). Molecular weight markers were prepared as previously described (8, 9).

After destaining, replicate gels were analyzed for radioactivity in two ways. Gels were sliced transversely into 2-mm segments with a razor blade stack. Individual slices were placed in 10 ml of water, and ‘,P was detected as Cerenkov radiation with an efficiency of approxi- mately 30% (10); all counting data were corrected for radioactive decay. The counting rate remained linear with increasing counts per peak to at least 2000 cpm. Analysis of replicate gels showed that the coefficient of variation of the count rate in a single peak was approximately 2 to 6%. Concurrently, replicate gels were sliced longitudinally, dried as previously described (ll), subjected to radio- autography, followed by densitometry. 100% transmission was fixed on a portion of the radioautograph not exposed to a gel slice. To obtain quantitatively accurate estimates using densitometry, the darkening of the film must be submaximal. Using Kodak blue-sensitive SB54 film,

the relationship between counts per min and optical density deviated from linearity if the exposure resulted in darkening to an absorbance of greater than 0.8 A. The area subtended by different bands within the same gel should most accurately approximate the counting data; however, if the same band was to be compared on replicate gels, the peak height served as well. The latter allowed more accurate estima- tions forphosphopeptides of similar mobility that were incompletely resolved by electrophoresis (Fig. 3). Despite the additional steps entailed by densimmetry of radioautographs, precision compared favorably (coefficient of variation 7%) with counting of transverse slices.

5’-Nucleotidase (EC 3.13.51, NADH oxidase (EC 1.699.31, and succinic cytochrome c reductase (EC 1.3.99.1) were assayed as previ- ously described (12). Protein was determined according to Lowry et al. (13). NaHZa*PO, in water was obtained from New England Nuclear. Partially purified [“Cluridine-labeled cytoplasmic RNA prepared from HeLa cells was kindly provided by Dr. Thoru Petersen of the Worcester Foundation for Experimental Biology.

RESULTS AND DISCUSSION

Enzymic Characterization of Subcellular Fractions-The method described by McKee1 and Jarett (7) was designed primarily to obtain a purified plasma membrane fraction. With the modifications described herein, relatively pure mitochon- drial and nuclear fractions were obtained conveniently, as well. The fractionation scheme was evaluated by assessing the distribution of enzymic markers at various steps (Table I), as well as by polyacrylamide gel electrophoresis in sodium do- decyl sulfate of the individual fractions (Fig. 1). Two steps in the fractionation are critical. First, the initial homogenization and centrifugation is of prime importance to the yield and purity of the plasma membrane and endoplasmic reticulum fractions. When the vigor of the initial homogenization was increased, the loss of plasma membrane into Sl was increased; conversely, the contamination of the final plasma membrane fraction by NADH oxidase was somewhat diminished. Less vigorous homogenization resulted in inadequate fragmentation of the endoplasmic reticulum. While Pl contained over 80% of the plasma membrane, mitochondria (Table I), and nuclei (by phase microscopy), Pl, however, was still significantly contam- inated by residual endoplasmic reticulum. Thus, the proper rehomogenization of Pl is the second step critical to the purity

TABLE I

Representative adipocyte fractionation

Adipocytes were prepared from the pooled epididymal fat pads of 20 activity units are micromoles hour-’ mg-’ for 5’nucleotidase; other- rata. Subcellular fractionation was performed as described under wise, micromoles -min- * mg-‘. 5’nucleotidase is present primarily in “Materials and Methods.” Aliquots of each intermediate and final the plasma membrane. NADH oxidase is present in the endoplasmic fraction (as described under “Materials and Methods”) ware analyzed reticulum and the outer membrane of the mitochondria. Succinic-cyto- for protein and enzymic markers and by polyacrylamide gel electro- chrome c reductase is located in the mitochondrial inner membrane phoresis in 1% sodium dodecyl sulfate (Fig. 1). The “Total” column (12, 14-16). represents all of the activity recovered in that fraction. Specific

Fraction Protein 5’-Nucleotidase

Specific Activity Total

NADH oxidase

Specific Activity Total

Succinic cytochrome c reductase

Specific Activity Total

Homogenate Pl Sl P2 P3 s3 Endoplasmic reticul Cytosol Plasma membrane Mitochondria Nuclei

mg 44.3 14.3 23.8 10.7

0.63 1.14

lum 6.00 11.5

1.66 3.10 2.17

0.20 8.86 0.56 24.8 0.043 0.56 8.01 0.83 11.9 0.107

0.05 1.19 0.75 17.9 0.003 0.72 7.70 0.81 8.66 0.173 0.94 0.59 1.59 1.00 0.003 0.08 0.09 0.39 0.44 0.017 0.30 1.80 1.59 9.54 0.007 0.01 0.12 0.10 1.15 0.006 2.12 3.52 0.54 0.90 0.006 0.49 1.52 1.66 5.15 0.182 0.59 1.28 0.79 1.71 0.018

1.90 1.53 0.07 1.85

0.02 0.04 0.07 0.01 0.56 0.04

by guest on February 7, 2018http://w

ww

.jbc.org/D

ownloaded from

FIG. 1. Polyacrylamide gel electrophoresis of adipocyte subcellular fractions. At each step during adipocyte subfractionation as described under “Materials and Methods” and in Table I, a sample of each fraction (50 fig of protein for all except S3) was removed for polyacryl- amide gel electrophoresis in 1% sodium dodecyl sulfate. Replicate gels were stained with Coomassie blue (upper panel) or PAS (lower panel). The abbreviations for the fractions are defined under “Materials and Methods.” The molecular weights of several major peptides are indi- cated at the sides of each panel. and their subcellular distribution may he judged by connecting the lines at either side of’ the panel. 7’D, trark- ing dye.

Adipocyte Peptides and Phosphopeptides 1507

lar weight 100,000 and 84,000. which coincide exactly with the two glycopeptides seen with PAS staining. Glycopeptides of similar mobility are also seen faintly in the endoplasmic reticulum fraction (Fig. l), to an extent consistent with the degree of 5’-nucleotidase contamination of this fraction. Other examples include peptides of molecular weights 14,000 and 25,000 found only in the cytosol, a mitochondrial peptide of molecular weight 135,000, and the major nuclear peptides between 13,500 and 17,000. By contrast, although the overall pattern of the endoplasmic reticulum peptides is unique, this fraction contains no major peptide not detectable in another fraction. In fact, several of these homologies are quite striking. A prominent endoplasmic reticulum peptide of molecular weight 216,000 is also prominent in the cytosol, as is another peptide of molecular weight 123,000. A peptide of similar mobility is seen also in the plasma membrane, but the greatest homology between the plasma membrane and the endoplasmic reticulum is in the peptides of molecular weight 62,000 and 79,000, both constituting major components of each of these subcellular fractions. The 62,000-dalton species in the plasma membrane is heterogeneous; although it rarely appears as a doublet in gels containing 1% sodium dodecyl sulfate, it is consistently resolved into two species in gels containing 0.2% sodium dodecyl sulfate. Conversely, the 62,000-dalton species in the endoplasmic reticulum displays no such splitting. The relatively low enzymic cross-contamination between these subcellular fractions suggests that these peptides are native to both fractions. (These findings should be compared with those reported by Czech and Lynn (17).) The presence of peptides of identical molecular weight in different subcellular fractions does not imply that these are the same peptides. The point to be emphasized is that in comparing the plasma membrane, mitochondrial, cytoplasmic, and nuclear fractions with each other, very few major peptides of similar mobility occur. In contrast, ‘only the endoplasmic reticulum contains major pep- tide species, the molecular weight of which is similar or identical with that of the major components in both the cyto- plasm and plasma membrane. These homologies might in part be fortuitous or an artifact of the fractionation technique; however, because the endoplasmic reticulum fraction as isolated is probably the main site of adipocyte protein syn- thesis, the possibility of a precursor-product relationship should be considered.

of the final plasma membrane fraction, as stressed by McKee1 and Jarett (7), and this was demonstrated in several ways. When the second cycle of homogenization and centrifugation of Pl was omitted, the NADH oxidase activity of the final plasma membrane fraction was increased more than 2-fold, and the calculated endoplasmic reticulum contamination was in- creased to over 50% (data not shown). Similarly, the particu- late material released by homogenization of Pl (P3) possessed the enzymic characteristics (Table I) and electrophoretic profile (Fig. 1) of the final endoplasmic reticulum fraction.

The “nuclear spin” suggested by McKee1 and Jarett (7) was omitted, since approximately 25% of the homogenate mito- chondria co-sedimented with the nuclei. These fractions were separated on the subsequent sucrose gradient (Table I). In summary, we estimate that the plasma membrane fraction is 70 to 75% pure, the major contaminant being endoplasmic reticulum. The endoplasmic reticulum fraction is approxi- mately 80 to 90% pure, with plasma membrane as the major contaminant. The mitochondrial fraction is approximately 15% contaminated by plasma membrane. The contamination of the mitochondrial fraction and the nuclear fraction by endoplasmic reticulum is uncertain, inasmuch as NADH-oxi- dase activity is known to be present in all three organelles in other tissues.

Peptides of Adipocyte Subcellular Fractions-Each subcel- lular fraction except the nuclear fraction contains an array of major peptides of molecular weight ranging from approxi- mately 15,000 to 200,000 without a single predominant species. The nuclear fraction, however, exhibits two predominant species of molecular weight 13,000 to 17,000, which we presume to be histones. Each subcellular fraction (except the endoplas- mic reticulum) contains at least one major peptide that is greatly enriched in that fraction (Fig. 1). For example, the plasma membrane contains two prominent peptides of molecu-

Identification of S2P-labeled Phosphopeptides-The aaP- labeled species in each subcellular fraction as detected by radioautography of dried stained gel slices are shown in Figs. 2 and 3. As to the chemical nature of these components, low molecular weight compounds containing alp were quantita- tively removed from the final gels through diffusion during the fixation, staining, and destaining steps. The broad zone of 8zP seen immediately behind the tracking dye in the particulate fractions corresponds to the mobility of phospholipid (8) and is selectively extracted by treatment of plasma membrane and endoplasmic reticulum with 90% acetone (data not shown). The contribution of the SzP-labeled polynucleotide to the nonphospholipid bands was asSessed by placing stained gels of each adipocyte fraction into 10% trichloroacetic acid at 90” for 35 min. This procedure reduced the s*P content of the major *?P bands by approximately 10 f 5%, whereas a gel containing [Yluridine-labeled cytoplasmic RNA prepared from HeLa cells lost over 95% of the [“Cluridine (Fig. 2). Thus, if the phospholipid is ignored, all of the remaining major azP-con-

by guest on February 7, 2018http://w

ww

.jbc.org/D

ownloaded from

Adipocyte Peptides and Phosphopeptides

FIG. 2. Effects of trichloroacetic acid on “*P-labeled adipocyte fractions. Isolated adipocytes, incubated 2 hours in the presence of extracellular NaH,azPO,, were subfractionated as described under “Materials and Methods.” Replicate aliquots of 50 gg of protein of plasma membrane (PM), endoplasmic reticulum (ER), mitochondria (M), nuclei (N), and cytosol (0, as well as an aliquot of [“Cluridine- labeled partially purified cytoplasmic RNA derived from HeLa cells ([“CIRNA), were subjected to polyacrylamide gel electrophoresis in sodium dodecyl sulfate and Coomassie blue staining. Replicate gels of each of these six species were immersed in 10% trichloroacetic acid at 90” for 35 min. The trichloroacetic acid-treated gels were then returned to 10% acetic acid (along with the untreated gels) and soaked 14 hours with several changes of 10% acetic acid. In no case did trichloroacetic acid treatment detectably alter the Coomassie blue-staining pattern. Control and trichloroacetic acid-treated gels of each fraction were analyzed both by radioautography and liquid scintillation counting of Z-mm slices.

For radioautography, dried longitudinal slices from control and trichloroacetic acid-treated gels from each adipocyte fraction, as well as [“CIRNA, were mounted together and exposed simultaneously. The right hand strip in each pair of radioautographs is from the trichloro- acetic acid-treated gel. A, radioautograph exposed 2 days. Note the essentially complete absence of (“Cluridine after trichloroacetic acid treatment. B, radioautograph exposed 12 days. 0, origin; TD, tracking dye.

taining bands would appear, by exclusion, to be phosphopeptides.* Consistent with this assumption was the lability of the s*P in these bands in boiling 1 N NaOH for 10 min; 90 to 95% of the $*P was discharged, with negligible losses of peptides that stain with Coomassie blue (data not shown). Direct confirmation of the phosphoprotein nature of four selected bands (plasma membrane: ‘79,000, 22,000 to 26,000; endoplasmic reticulum: 123,000, 69,000 to 79,000) was achieved by identification of the 82P-containing residues after elution from fixed gels, partial acid hydrolysis (6 N HCl, lOO”, 6

*Prosthetic groups containing phosphate, covalently linked to pro- tein, are also possible.

hours), and high voltage electrophoresis (18). Correcting for losses of phosphoserine and phosphothreonine during hydrol- ysis (18), we estimate that 75 * 10% of the ‘*P-labeled material subjected to hydrolysis can be accounted for by S*P-labeled phosphoserine (predominantly) and phosphothreonine. The similar behavior of these four bands and of the remaining major **P-containing species visible in the radioautographs with respect to treatment with hot acid, hot alkali, and 90% acetone indicates that, exclusive of phospholipid migrating near the tracking dye, they are phosphopeptides.

The unfractionated homogenate reveals 10 to 12 major phosphopeptides (Fig. 3). The number of major phosphopep- tides is far less than the number of major Coomassie blue- staining peptides in a given fraction; the total number of phosphopeptide species, however, is very large and is best shown in a radioautograph of *rP-labeled cytoplasm exposed for varying times (Fig. 2). At short exposure, two phosphopep- tides are quantitatively pre-eminent, the remainder of the gel appearing as a nondescript background. After prolonged expo- sure, this background is seen as a host of individual minor phosphopeptide bands. This pattern can also be seen in the nuclear and endoplasmic reticulum fractions. We cannot rigorously exclude the possibility that these minor species are artifacts generated through proteolysis or aggregation. The reproducibility of both the Coomassie blue and radioauto- graphic patterns, however, argues against this explanation. Although many peptides of considerable functional importance may be present among these minor components, their low absolute $*P incorporation prevents accurate quantitative examination of their behavior individually.

Kinetics of Phosphopeptide Generation and Turnover-The incorporation of szP into the major adipocyte phosphopeptides increases steadily to 120 min and is essentially unchanged between the 2nd and 3rd hour (Fig. 4). This pattern is seen with all the major phosphopeptides in the homogenate as well as within each subcellular fraction, regardless of the absolute $*P incorporation into that band. Thus, 3*P labeling of phosphopeptides to steady state is consistently achieved by 2 hours. If cells are labeled with szP for 2 hours, then washed free of the radioisotope, and resuspended in buffer containing rlP, the loss of B*P-phosphopeptide counts is relatively slow (15 to 35% over the initial 30 min), homogeneous, and approximates the initial rate of szP incorporation. These findings suggest that an early step in phosphate metabolism, e.g. phosphate entry or incorporation into ATP, is rate limiting in achieving 3*P incorporation into protein.

Phosphopeptides in Specific Subcellular Fractions-Each subcellular fraction has a distinct and characteristic array of phosphopeptides that is reproducible with great fidelity. The pattern of major phosphopeptides in the plasma membrane, mitochondrial, nuclear, and cytosol fractions shows no overlap and is further supporting evidence for the trivial cross-contam- ination between these fractions (Fig. 3). The overall pattern of major phosphopeptides in the endoplasmic reticulum is dis- tinctive but, as with the pattern of Coomassie blue-stained peptides, exhibits several homologies with both the plasma membrane and cytoplasm. The endoplasmic reticulum phos- phopeptides of molecular weight 94,000 and 17,000, although similar in mobility to species in the cytoplasm and plasma membrane, respectively, are clearly most highly labeled in the endoplasmic reticulum. However, phosphopeptides of molecu- lar weight 79,000, 62,000, 26,060, and 22,000 are present in both the endoplasmic reticulum and plasma membrane, and the

by guest on February 7, 2018http://w

ww

.jbc.org/D

ownloaded from

Adipocyte Peptides and Phosphopeptides 1509

0 IO 20 30 40 0 10 20 30 40 0 10 20 30 40

FIG. 3. Comparison of analytic techniques for *‘P in polyacrylamide gels. Adipocytes were incubated for 2 hours with NaH.“PO,. After subfractionation, replicate aliquote (each containing 50 ng of protein) of homogenate and each subcellular fraction were subjected to poly- acrylamide gel electrophoresis in sodium dodecyl sulfate. After stain- ing, **P was analyzed both by liquid scintillation counting of 2-mm gel segments (shaded area) and radioautography with densitometry. All

counts are corrected for **P decay. Numbers represent molecular weight x lo-’ at the indicated point on the densitometric scan. The abscissa indicates the gel slice number; the top of each gel is zero on the abscissa. The ordinate units are counts per 10 min per gel slice x 10’ for the shoded bars and optical density for the densitometric scan. Abbreviations are the same as those in Fig. 2.

EXPERIMENT I 1 EXPERIMENT 2

60

60

40

20

0

120

100

SO

60

40

60

40

20

/ 0 I 2 3 0 I 2 3

HOURS PIG. 4. ‘*P incorporation into selected phosphopeptides in adipo-

cyte subcellular fractions. Adipocytes were incubated with NaH,VO,

extent of sT incorporation into these species in the endoplas- mic reticulum is much less than in the plasma membrane. Because the endoplasmic reticulum is 10 to 20% contaminated by plasma membrane as judged by enzymic markers, some (perhaps all) of the **P in these endoplasmic reticulum bands represents contaminating plasma membrane. The two endo- plasmic reticulum phosphopeptide bands at molecular weights 216,000 and 123,000 are also the major phosphopeptide bands observed in the cytoplasm and correspond to major Coomassie blue-staining peptides. No estimate of cytoplasmic contamina- tion of the endoplasmic reticulum is available. We have observed that the addition of divalent cations to Sl (which contains the endoplasmic reticulum and cytoplasmic frac- tions), in excess of the EDTA present, led to the adsorption of both of these stained peptides and phosphopeptide bands onto the endoplasmic reticulum. Therefore, tRe subcellular localiza- tion of these phosphopeptides in situ is undetermined, and they may be entirely cytoplasmic or partitioned between the endoplasmic reticulum and cytoplasm. The four phosphopep- tides of the plasma membrane exhibited, by far, the highest absolute incorporation of *V’ observed (Fig. 3) and are cer- tainly native plasma membrane species. The next most highly labeled constituent was the major mitochondrial phosphopep- tide of molecular weight 41,000.

and aliquots removed for subcellular fractionation at 30, 60, 120, and 130 min. After polyacrylamide gel electrophoresis in sodium dodecyl sulfate of each fraction (50 cg of protein) at all time points, the stained gels were transected into 2-mm segments and counted. Slices contrib- uting to major peaks were summed, and the results plotted as per cent of the value at 2 hours. The phosphopeptides are designated o through d from the plasma membrane, e through g from the endoplasmic reticulum, h from the mitochondria, i from the cytosol, and j from the nucleus, with molecular weights x lo-’ as follows: o = 79, b = 62, c = 5O,d=22to27,e=17,f=123,g=27,/r=41,i=123,andj=l3to 17. The abbreviations are the same as those in Fig. 2.

by guest on February 7, 2018http://w

ww

.jbc.org/D

ownloaded from

1510 Adipocyte Peptides and Phosphopeptides

REFERENCES 11.

1. Taborsky, G. (1974) Ado. Protein Chem. 28.1-210 12. 2. Bell, R. M., and Koshland, D. E., Jr. (1971) Science 172.1253-1256 3. &gal, H. L. (1973) Science 180, 25-32 13. 4. Walsh, D. A., Perkins, J. P., and Krebs, E. G. (1968) J. Biol.

Chem. 243,3763-3765 14. 5. Kuo, J. F., and Greengard, P. (1969) Proc. Natl. Acad. Sci. U.S. A.

64.1349-1355 15. 6. Rodbell, M. (1964) J. Biol. Chem. 239, 375-380 7. McKeal, D. W., and Jarett, L. (1970) J. Cell Biol. 44, 417-432 16. 8. Fairbanks, G., Steck, T. L., and Wallach, D. F. H. (1971) 17.

Biochemistry 10, 2606-2616 9. Fairbanks, G., and Avruch, J. (1972) J. Supramol. Struct. 1.66-75 18.

10. Avruch, J., and Fairbanks, G. (1972) Proc. Natl. Acad. Sci. U. S. A. 69.1216-1220

Fairbanks, G., Levinthal, C., and Reeder, R. H. (1965) Biochem. Biophys. Res. Commun. 20. 393-399

Avruch, J., and. Wallach, D. F. H. (1971) Biochim. Biophys. Acta 233,334-347

Lowry, Cf. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275

Sottocasa, G. L., Kuylenstiema, B., Em&r, L., and Bergstands, A. (1967) J. Cell Biol. 32. 415-436

Schnaitman, C., and Greenwalt, J. W. (1968) J. Cell Biol. 38, 15S175

Kamat, V. B., and Wallach, D. F. H. (1965) Science 148,1343-1345 Czech, M. P., and Lynn, W. S. (1973) J. Biol. Chem. 248,

508-5088 Allerton, S. E., and Perlmann, G. E. (1965) J. Biol. Chem. 240,

3892-3898

by guest on February 7, 2018http://w

ww

.jbc.org/D

ownloaded from

J Avruch, G R Leone and D B Martinphosphopeptides.

Identification and subcellular distribution of adipocyte peptides and

1976, 251:1505-1510.J. Biol. Chem. 

  http://www.jbc.org/content/251/5/1505Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/251/5/1505.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on February 7, 2018http://w

ww

.jbc.org/D

ownloaded from