the characterization of adult human myoglobin*

THE JOURNAL OF BIOI.OGICAL CHEMISTRY Vol. 237, No. 9, September 1962

Printed in U.S.A.

The Characterization of Adult Human Myoglobin*

GERALD T. PERKOFF,? ROBERT L. HILL,$ DOUGLAS M. BROWN, AND FRANK H. TYLER

From the Laboratory for the Study of Hereditary and Metabolic Disorders and the Departments of Medicine and Biochemistry, University of Utah College of Medicine, Xalt Lake City, Utah

(Received for publication, May 17, 1962)

It has been reported that crystalline human myoglobin can be resolved by electrophoresis on paper into three components (1,2). Two of these possess nearly identical absorption spectra between 500 and 600 rnp and give the same reactions with oxygen. Al- though it was suggested that these components differ structurally (2), no evidence has been presented to support this view.

In the present study the heterogeneity of human metmyoglobin has been examined with the aid of ion exchange chromatography. Five chromatographically distinct fractions, two of which represent approximately 75% of the myoglobin in skeletal muscle, have been found. Chemical and physical analyses of these fractions indicate that the globin is identical in the two myoglobins that are present in major amount. The differences in chromatographic and electrophoretic behavior of these fractions result from differences in the state of the iron in the heme prosthetic group. At least one of the other myoglobin components, which represents only a minor amount of the myoglobin of normal muscle, appears to differ in the primary structure of the globin moiety. The availability of large amounts of chromatographi- tally pure myoglobin has prompted further characterization of this protein.

EXPERIMENTAL PROCEDURE

Materials and Methods-The myoglobin that was used as the starting material for the chromatographic experiments was prepared from skeletal muscle by a modification of the methods of Drabkin (3) and Ginger, Wilson, and Schweigert (4) as described previously (5). The muscle was obtained at autopsy, within 4 to 14 hours of death, from patients who died of nonmuscular disorders and was stored at -5” until used. Crystalline myoglobin was prepared by the method of Luginbuhl (6). Spectral analysis showed that the myoglobin in these preparations was in the ferri- or met- form.

Dilute solutions of myoglobin were concentrated by one of three means. Saturation of aqueous solutions with ammonium sulfate at pH 7 completely precipitates myoglobin, which then can be collected by gravity filtration on fluted filter paper, sus- pended in a small amount of water, and dialyzed until free of ammonium sulfate. Concentration also was obtained by centrifugation at 96,000 X g in the Spinco model L ultracentrifuge

* This work was supported by research grants from the Na- tional Institute of Arthritis and Metabolic Diseases. National Institutes of Health, United States Public Health Service, and the Muscular Dystrophy Associations of America, Inc.

t Present address: Salt Lake Veterans Administration Hosnital. 1 I Salt Lake City, Utah.

1 Present address: Department of Biochemistry, Duke Uni- versity, Durham, North Carolina.

at 4” for 72 hours. Under these conditions, myoglobin sediments to the bottom of the centrifuge tube and can be removed from the tube with a needle and syringe. The most effective means for concentrating myoglobin from very dilute solutions (< 0.2%) was by adsorption on DEAE-cellulose. Myoglobin is adsorbed quantitatively from solutions (< 0.001 M in salt concentration, pH 8.9) at 6” on DEAE-cellulose that has been equilibrated with 0.001 M Tris buffer, pH 8.9. Adsorption is most effective when the cellulose is packed in columns with dimensions of approximately 1 X 10 cm. The myoglobin is eluted from the cellulose with 1 M Tris buffer, pH 7.85, or 1 M sodium chloride (unbuffered). Buffer salts or sodium chloride are removed by dialysis.

The concentration of myoglobin was measured spectrophotometrically. The extinction coefficient (E&zEs,liter) of human myoglobin was determined for several preparations from simul- taneous dry-weight determinations at 105” and spectrophoto- metric analysis at 280 rnp. For a molecular weight of 17,500, the molar extinction coefficient is 3.4 X 104 between pH 5 and 9.

Zone electrophoresis on starch gels was performed by the method of Smithies (7). Starch gels were stained with Amido Black B (7) and with benzidine hydrochloride (8). Two-dimensional paper electrophoresis-chromatography of tryptic peptides was performed as described earlier (9). Spectral analyses were made with a Carey Model 14 automatic recording spectrophotom- eter. Amino terminal end group analyses were performed by the phenyl isothiocyanate and the fluorodinitrobenzene techniques (10). Carboxgl-terminal end group analysis was made with carboxypeptidase A. Amino acid analyses were performed with the Spinco model MS automatic amino acid analyzer (11, 12).’ Iron was estimated by the method of Gubler et al. (13). Car- boxypeptidase A (Worthington Biochemical Corporation, 3 X crystallized) was treated with diisopropyl fluorophosphate before use. Crystalline trypsin was a commercial preparation (Worth- ington) .

Chromatography of Myoglobin-Myoglobin was chromatographed at 6” on DEAE-cellulose that had been equilibrated with 0.005 M Tris buffer, pH 7.85. Jacketed columns (0.9 x 12 cm) which were cooled by circulation of water at 6” were used for analysis of small (5 to 15 mg) amounts of myoglobin. Chro- matography on a preparative scale (0.5 to 1.2 g) was achieved in the same manner with 5 x 15.cm jacketed columns. Typical results of a chromatographic separation of myoglobin performed on preparative columns are shown in Fig. 1. Four fractions (FA, Fl, Fg, Fz) were eluted with the dilute Tris buffer, whereas

1 We would like to thank Mr. Boyd Lythgoe for his valuable assistance in performing these analyses.

2820

by guest on February 14, 2018http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

September 1962 G. T. Perko$, R. L. Hill, D. M. Brown, and F. H. Tyler 2821

Volume a liters Volume fi liters

Volume * liters

FIG. 1. Chromatography of human myoglobin on DEAE-cellulose. Myoglobin, 800 mg, dissolved in 50 ml of water was dialyzed for 18 hours at 5” against 50 volumes of 0.005 M Tris buffer, pH 7.85, and applied to a column (5 X 12 cm) of DEAE-cellulose which was previously equilibrated with the same buffer. The flow rate was 60 ml per hour; the temperature was maintained at 6” until 8.2 liters were collected. The column then was warmed to 22” and developed with 1 M Tris buffer, pH 7.85. Ten-milliliter fractions were collected and analyzed spectrophotometrically at 415 and 280 rnp. Further analysis of each component was made by pooling the myoglobin emerging in the peak tubes.

one (F3) was eluted with 1 M Tris buffer, pH 7.8, at 25”. Frac- tions F1, Fs, and Fa contained the majority of the heme protein. Fraction A, which emerged from the column at the hold-up volume, contained only none-heme protein. The yield of each fraction varied among different preparations; however, F1, Fz, and FI generally accounted for 55 to 65%, 15 to 20%, and 15 to 30%, respectively, of the total myoglobin applied to the column. It is noteworthy that the fractions had visibly different colors: F1 was brown, FP was dark red, Fa was reddish-brown, and FB was greenish-brown. Crystalline myoglobin gave the same chromatographic pattern as the noncrystalline myoglobin shown in Fig. 1.

When F1 and Fz were rechromatographed separately under the same conditions, two components were observed. Eighty per cent of F1 emerged from the column in the volume expected from its behavior as shown in Fig. 1, whereas the remainder behaved like FP. Fraction 2 on the other hand revealed approximately 25% of FI on rechromatography. Because the rechromatographed samples could not be contaminated with one another to this extent, these results suggest that the myoglobin in each fraction can be converted into the other chromatographic form. When FI and Fs were chromatographed as oxymyoglobin or cyanmetmyoglobin, they could not be distinguished from one another. This is seen in Fig. 2A (FI and Fz oxymyoglobins) and in Fig. 2B (FI and Fz cyanmetmyoglobins). The small differences in the point of emergence probably reflect variations in the size of the columns used in the experiments. When unfractionated preparations were chromatographed as oxy- or cyanmetmyoglobin, only one peak was observed in addition to Ft.

Fraction 3 could not be purified further under the conditions described in Fig. 1. The most successful means for separating the myoglobins in this fraction was by chromatography as shown in Fig. 3. A gradient of sodium chloride formed with the cone-

sphere arrangement (1850 ml, total volume) described by Sober and Peterson (14) was used in conjunction with a DEAE-cellulose column and 0.005 M Tris, pH 7.85, as the starting buffer. Six distinct fractions were obtained, only three of which were present in significant amounts (F3B, FE and F3D). Because each component except FB,, had a smaller (415/280 mp) absorption ratio than & and Fz, it is likely that significant amounts of non-heme proteins contaminated all fractions but FSD. At- tempts to purify the other heme proteins in F3 by varying the chromatographic conditions were unsuccessful.

Electrophoresis of Myoglobin on Starch Gels-Zone electrophoresis on starch gels proved to be a useful means of characterizing unfractionated myoglobin. Fig. 4 shows the electrophoretic behavior of unfractionated myoglobin and of the chromatographically purified fractions. Three heme proteins could be distinguished. Occasionally a fourth, which migrated more rapidly than the others, was observed. Each of the three major heme proteins in unfractionated myoglobin was similar to one of the major chromatographic fractions (FI, Fz, and F3). The heme protein that moves more rapidly than the major components in unfractionated preparations was found in Fa and presumably represents one of the minor components of this fraction (Fig. 3). Although it is evident that FI shows a considerable amount of Fz and Fz shows significant amounts of PI, this result is analogous to the behavior of these fractions on rechromatography. Frac- tion B could not be distinguished from Ft by this method. It is noteworthy that FB has a greater electrophoretic mobility than any of the other fractions. When unfractionated myoglo-

I I I I I I A. Oxymyoglobin B. Cyanmetmyoglobins

.25 - 0 Fraction 1 0 Fraction 1 - l Fraction 2 Fraction 2

Volumednl.

FIG. 2. Rechromatography of purified fractions of myoglobin. Samples of F1 and Fz, which were obtained as shown in Fig. 1, were applied to 0.9 X 12 cm columns of DEAE-cellulose and developed at 6” with Tris buffer at a flow rate of 20 ml per hour. Fractions, 3 ml, were collected, and the myoglobin was estimated spectrophotometrically at 415 rnp. Each fraction was chromatographed on separate columns. A, Chromatography of F1 and Ff as oxymypglobin. Eight milligrams of each fraction, dissolved in 4 ml of Tris buffer, were treated with 0.5 mg of sodium dithionate and oxygenated for 20 minutes with purified oxygen. Each solution was then dialyzed against buffer which was previously satu- rated with oxygen. B, Chromatography of FI and FZ as cyanmetmyoglobins. Ten to twelve milligrams of each fraction, dissolved in 4 ml of Tris buffer, were converted to the cyanide derivative by mixing with 0.5 ml of 0.0012 M potassium cyanide. Each fraction was analyzed as described above except that the columns were developed with the Tris buffer containing 0.001 M potassium cyanide.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

2822 Human Myoglobin Vol. 237, No. 9

I.9 1.8 t

0.6 0.5 0.4 : 0.3 t FxA ”

I I : 0.2 i's 8 .̂ WA r'. I

0. I 0.2 0.3 0.4 0.5 0.6 Volume-liters

FIG. 3. Rechromatography. of myoglobin Fa. Fa, 60 mg, which was obtained as shown in Fig. 1, was converted to cyanmetmyoglobin as described in Fig. 2B. The myoglobin was then applied to a 1 X 10 cm column of DEAE-cellulose which had been equilibrated with 0.005 M Tris buffer, pH 7.8, containing 0.001 M potas-

sium cyanide at 6”. After the sample was absorbed at the top of the column, the column was developed at 6” by gradient elution as described in the text. Fractions of 5 ml were collected at a flow rate of 20 ml per hour.

bin and each of the chromatographic components were converted to cyanmetmyoglobin and then analyzed electrophoretically, only 2 heme proteins (“PI, Fz” and FJ were demonstrable. This observation is consistent with the finding that PI and FP in the cyanmet- form emerge from columns of DEAE-cellulose in the same volume.

Comparison of Tryptic Peptides from Chromatographically Purified MyoglobinsIn order to determine whether the chromatographically pure myoglobins differed from one another in primary structure, tryptic digests of each myoglobin were analyzed on paper by a two-dimensional electrophoretic-chromatographic technique (9). Myoglobin, 25 mg, dissolved in 2 to 3 ml of water, were treated at - 10’ with 25 volumes of an acetone- HCl mixture according to the methods of Rossi-Fanelli and An- tonini (15). The resulting globin was dissolved in 2.5 ml. of water, and the pH of the solution was adjusted to 8.0. Trypsin, 0.5 mg, dissolved in 0.1 ml of 0.001 N HCl, was added to start

FIG, 4. Zone electrophoresis of myoglobin on starch gels. Un- fractionated myoglobin and the chromatographic components derived from myoglobin were analyzed at pH 8.6 by the methods of Smithies (7). Electrophoresis was performed for 18 hours with a potential gradient of 4 volts per cm. After longitudinal cuts of these gels were made, the proteins were detected by staining with Amido Black B. When cyanmetmyoglobins were analyzed, potassium cyanide 0.001 M was added to the buffer. In the top half of the figure, metmyoglobins are shown. The lower half, showing cyanmetmyoglobins, illustrates the fact that electrophoretic differences between Fi and Fz disappear when the cyanmetmyoglobins are used.

METMYOGLOBINS FRACTION 3

* s-t-t”

i FRACTION I

*_i_\ Ag FRACTION 2

FRACTION B =

‘: ‘: WHOLE Mb.

:YANMETMYOGLOBINS

‘RACTION 3

‘RACTION B

‘RACTION 2

‘RACTION I

\ ORIGIN


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

September 1962 G. T. Perkof, R. L. Hill, D. M. Brown, and F. H. Tyler 2823

FIG. 5. Two-dimensional peptide patterns of the tryptic peptides from human mvoelobins F1 and F,. An aliauot of the tryptic digest which contained the peptides from 1 *to 2 mg of myoglobin was resolved electrophoretically on Whatman 3MM paper for 3 hours at 10 volts per cm in pyridine acetate buffer,

the digestion and the mixture was incubated at 40”. Throughout the digestion, the pH was maintained between pH 7.8 and 8.2 by the addition of 0.1 N NaOH. The precipitate of globin disap- peared during the first 15 to 30 minutes and hydrolysis was complete after 90 minutes. No differences were observed in the rates of hydrolysis of the different myoglobin fractions. Analy- sis of aliquots of the tryptic digests obtained in this manner gave peptide patterns like those shown in Fig. 5. Fractions 1, 2, and B were indistinguishable by this method of analysis. At least 24 distinct peptides were found, a number close to that expected from the lysine and arginine content of human myoglobin (Table I). Fractions 1, 2, and B also gave identical patterns when the papers were sprayed with reagents which specifically detect histidine (16), tryptophan (17), arginine (18), tyroside (19) and sulfur-containing peptides (20). Fraction 3D (Fig. 3), which was the only myoglobin from Fraction 3 that was not contaminated with non-heme protein, gave patterns that differed from those in Fig. 5. Several spots in addition to those peptides found in Fractions 1, 2, and B were observed. This result, together with the difference in electrophoretic mobility noted when F) and FI were compared by zone electrophoresis, suggests that the primary structure of the myoglobin in FSD may differ from that of FI, Ft, and FI. However, since FsD may contain small amounts of non-heme protein, no definite conclusions about its primary structure can be drawn at this time. Additional experiments will be necessary to clarify the nature of this myoglobin.

PHYSICAL CHARACTERIZATION

Spectral AnalysisThe ultraviolet and visible spectra of each of the myoglobin components were determined. The wave

pH 6.5. The peptides then were resolved chromatographically in the second dimension for 26 hours with butanol-acetic acid-water (260:30:75). Peptides were located with a ninhydrin spray re- agent.

lengths of maximal absorption for four of the chromatographi- tally purified myoglobins are listed in Table II. FI has absorption characteristics of acid metmyoglobin, whereas Fz and Fa have those of alkaline metmyoglobin (21). A portion of the spectra obtained with these types of metmyoglobin is’ shown in Fig. 6. The spectra of FI and Ft were identical when they were converted to either the reduced, oxy-, cyanmet-, or carboxy- forms.

Sedimentation and Di$usion Studies---These analyses were made in 0.1 M Verona1 buffer, pH 8.5, with the Spinco model E ultracentrifuge; each run was made at 59,780 r.p.m., 20”. The photographic plates were read with a microcomparator. All other operations were the same as reported in earlier studies from this laboratory (22, 23).

F1 and Fz sedimented as single, homogenous peaks. Their rate of sedimentation was a function of the protein concentration. Although the red color of myoglobin at concentrations greater than 6 mg per ml prevented more extensive studies, the S2,,u, value at infinite dilution was calculated from the data available to be 1.815X.

Diffusion studies were performed in the Spinco model H electrophoresis-diffusion apparatus with the aid of the Rayleigh interference optical system. Diffusion constants were calculated by the methods of Longsworth (24) and Schachmann (25).

Fz was analyzed in 0.1 M Verona1 buffer, pH 8.5, at a protein concentration of 1.1 mg per ml. The color of the myoglobin prevented measurements at higher concentrations. Photo. graphs of the interference patterns were calculated between 20 and 56 hours after formation of the boundary. The patterns obtained were read on a microcomparator. The results were corrected for viscosity, density, .and temperature in the usual


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

2824 Human ikfyoglobin Vol. 237, No. 9

TABLE I Amino acid composition of human myoglobin

Amino acid \mino acid per 100 g of protein

fWn, residue er 100 g f protein

:a1cu1atec no. of

residues

No. of .esidues ,Ref. 33)

Alanine. Arginine Aspartic

acid..... Glutamic

acid..... Glycine. Histidine. Isoleucine Leucine Lysine Methionine Phenylala-

nine.. Proline. Serine..... Threonine Trypto-

phan Tyrosine Valine.. . Ammonia.. Glutamine. Asparagine

8 g 5.71 4.56 2.65 2.38

5.22 11.2 11 11 5.10 2.7 3 2

Similar results were obtained when globin was analyzed on paper strips (1 X 10 cm) with the phenyl isothiocyanate method described by Shelton and Schroeder (27). Only glycine phenyl isothiohydantoin was detected on Solvents A, D, and F of Sj6- quist (28) and Edman and SjGquist (29). Although no quanti- tative estimation was made of the amount of this derivative, the absence of other detectable phenyl thiohydantoins precludes the possibility of other amino end groups.

9.21 7.97 5.80 12.1 12 10

17.31 15.20 9.87 20.6 21 19 6.25 4.75 6.98 14.6 15 13 8.21 7.26 13.31 9.3 9 8 5.03 4.34 3.21 6.7 7 6

12.19 10.52 7.00 16.3 16 18 16.12 14.14 18.50 19.3 19 22 2.45 2.16 1.38 2.9 3 3

6.22 5.54 3.16 6.6 7 8 3.99 3.37 2.94 6.1 6 8 4.55* 3.77 3.63 7.6 8 7 2.94* 2.50 2.07 4.3 4 4

3.60t 2.35 5.26 1.083

3.28 2.12 4.45

2.96 3.0 1.09 2.3 3.77 7.8 5.66 11.1

3 2 8

11 7t, t 3t, $

3 2 6

13

Carboxyl Terminal End-group AnalysisGlobin is not suffi- ciently soluble at pH 8 to allow uniform digestion with carboxypeptidase A. Digestion was achieved, however, under conditions similar to those used by Guidotti for the ar- and P-chains of human hemoglobin (30). Globin, 25 mg, dissolved in 2.5 ml of water was mixed with 0.125 ml of 0.2 M sodium lauryl sulfate and the pH of the mixture was adjusted to 8. After addition of 0.6 mg of carboxypeptidase A (0.05 ml) the mixture was incubated at 40". Aliquots, 0.7 ml, were removed at 15, 30, and 55 minutes after addition of the enzyme, and acidified to stop hydrolysis. The amino acid content of each aliquot was measured by the methods of Spackman, Stein, and Moore (11, 12). Fig. 7 shows the rate of release of the amino acids. Although it is impossible to distinguish between glutamine, aspara.gine, and serine on the columns used for analysis, only glutamine was seen on paper chromatograms at 15,30, and 55 minutes. It is evident that the

TABLE II Absorption maxima of human myoglobins

Total...... 114.04 98.31 102.45 154 150 Myoglobin

1

* These values obtained by extrapolation to zero time hydrolysis .

t These values obtained from analysis of complete enzymic hydrolysates.

$ These values omitted from the total.

manner to give values of D20,w. Under these conditions the value for D20,w was calculated to be 9.6 X lO+ cm2 per second.

The molecular weight of myoglobin, computed from the sedimentation and diffusion constants, is 17,900. The partial specific volume of 0.743, which was used in calculating the molecular weight, was estimated from the amino acid composition reported in Table I.

CHEMICAL CHARACTERIZATION

Amino Terminal End-group AnalysisGlobin, prepared from F1 and Fz, gave glycine as the only detectable end group. Flua- rodinitrobenzene, 0.05 ml, in 5 ml of ethanol was added to 2 ml of 1% sodium bicarbonate containing 25 mg of globin. After the mixture had reacted for 2 hours at 25”, the insoluble dinitrophenylglobin was collected by centrifugation, washed three times successively with water, ethanol, and ether, and then air-dried for 5 days. Samples of the dinitrophenylglobin were hydrolyzed with the formic acid-acetic anhydride-perchloric acid mixture of Hanes, Hird, and Isherwood (26) for 2 hours at 100”. Chro- matographic analysis of hydrolysates (10) revealed only dinitro- phenylglycine in an amount corresponding to 0.73 moles per mole of dinitrophenylglobin (mol. wt. 17,500). Because this value was not corrected for the weight of the dinitrophenyl

groups in the protein, it is evident that the yield of dinitrophenyl glycine is close to 1 mole per mole of globin.

Absorption maxima (wave length, mr)

Fraction l........................... Fraction 2 ........................... Fraction 3 ........................... Fraction 3D. ........................ Fraction 1, 2 (ferromyoglobin) ....... Fraction 1, 2 (cyanmetmyoglobin) .... Fraction 1, 2 (oxymyoglobin) ........ Fraction 1, 2 (carboxymyoglobin) ....

634,500,410,278 580, 543, 412, 279 580, 540, 414, 278 540, 416, 273 551, 424, 275 545, 424, 279 580, 540, 417, 282 580, 542, 522, 274

I I I 0.4

0.3

500 600 700

WAVE LENGTH - mu FIG. 6. The absorption spectra of myoglobin Fractions 1 and 2.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

September 1962 G. T. Perkof, R. L. Hill, D. M. Brown, and F. H. Tyler 2825

amino acids are released in the order glycine > glutamine > phenylalanine > leucine, results which suggest a carboxyl terminal sequence similar to that found on analysis of horse

NH2

myoglobin by similar methods (31, 32), namely, Leu-Phe-Glu- Gly. However, it is apparent that more than one residue of glycine is released, which suggests that this amino acid occurs in more than one position in the carboxyl terminal sequence. Indeed, unpublished studies,2 have shown that a peptide from a tryptic digest of globin possesses the sequence Glu-Leu-Gly-

NH2 I

Phe-Gly-Glu-COOH- and must represent the carboxyl terminal sequence of human myoglobin.

Amino Acid Composition-The myoglobin in FI was used for these studies. The protein was exhaustively dialyzed against water, lyophilized, washed with acetone and ether, and air- dried for two weeks before analysis. The nitrogen content of this material, determined by a micro-Kjeldahl method, was 16.7x, a value in agreement with previous analyses (26). The iron content was 0.318%. Moisture content was 7.3%. Sam- ples were prepared for hydrolysis for 23, 44, and 70 hours as previously described (34). Duplicate analyses were performed on each hydrolysate by the met,hods of Moore, Spackman, and Stein (12). When destruction of amino acids occurred during acid hydrolysis, the yields at different times of hydrolysis were corrected by linear extrapolation to zero time (34). Globin obtained from the myoglobin of F1 also was hydrolyzed completely with enzymes as described by Hill and Schmidt (35). Analysis of this enzymic hydrolysate provided an estimate of the tryptophan, asparagine, and glutamine content.

The composition of human myoglobin estimated by these methods is given in Table II. The weight recovery of amino acid residues and the nitrogen recovery, which are not corrected for the heme content of myoglobin, are within the precision of the analytical methods employed. The number of residues was calculated from the minimal molecular weight that was obtained with each of the amino acids on the basis of a molecular weight of 17,450. The molecular weight computed from the assumed number of residues is 17,330. When corrected for the heme content of myoglobin, a molecular weight of 17,920 is obtained, in close agreement with the value 17,900 estimated from sedimentation-diffusion measurements, and the value 17,450 calculated from the iron content.

It is noteworthy that the composition obtained by these methods differs somewhat from that repdrted earlier by Rossi- Fanelli, Cavallini, and DeMarco (33). The improved methods of chromatographic analysis of amino acids that were used here were not available when the earlier analyses were made, nor was a chromatographically purified myoglobin preparation examined at that time. These factors may account for the small differences observed. One analysis of a hydrolysate of F2 (see above) gave essentially the same composition shown in Table I for F1. FI was not analyzed since its composition would be altered by the non-heme proteins that are present in this material.

* Unpublished studies, R. L. Hill. With the aid of the paper strip phenyl isothiocyanate end-group method (26), the amino terminal sequence in human myoglobin was found to be Gly-Leu- Ser-(Asp- or Glu-).

I I I Glycine 0 Glutamine 0 /” I Phuylohnine A

1.0 - Leucine X

Time -Minutes

FIG. 7. The rate of release of amino acids from globin by car- boxypeptidese. Experimental details are described in the text.

DISCUSSION

It is evident that the various components in human metmyoglobin which can be distinguished by electrophoretic or chromatographic means differ structurally either in their heme prosthetic group or in their globin moiety. The two major components, FI and Fs, appear to differ only in their prosthetic group. F1 has the spectral properties of acid metmyoglobin and F2 of alkaline metmyoglobin. These forms are in equilibrium with one another as shown schematically by the following equation, where MbFe+-OH2 represents acid metmyoglobin and MeFe- OH represents alkaline metmyoglobin.

MbFef-OH 2 = MbFe-OH + H+

The pK’,, of this reaction for horse myoglobin at 20” in the low salt concentration used in this study is approximately 8.9 (36). Because the acid and alkaline species are in equilibrium, and the amount of each depends on the pH of the solution, it is strik- ing that one species can be isolated from the other by chromatographic or electrophoretic means. In the chromatographic experiments that employ Tris buffers at pH 7.8 to 7.9, approximately 85% of the myoglobin (excluding Fs) is in the acid form, the remainder in the alkaline form. These are close to the amounts expected at this pH if the pK of the reaction for human myoglobin is close to that of horse myoglobin.

FB, a chromatographic component which is present in minor amounts, appears to possess the same globin moiety as FL and F2. Although its electrophoretic mobility on starch gels is unaffected by reaction with cyanide, other observations suggest that it has an altered prosthetic group. It possesses a distinct brown-green color, absorbs slightly at 620 rnp, and has a lower (415/280 rnp) absorption ratio than FI or Fi. Only small amounts of thii material were available and an exact spectral characterization was impossible. How-ever, the color and spectral properties of FB suggest that it contains a modified protoporphyrin ring system similar to choleglobin or verdoglobin (20). This form of the protoporphyrin ring probably would not be expected to exist in Y~UO but could be formed either after death before the muscle is removed and frozen, or during the preparative procedure.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

2826 Human Myoglobin Vol. 237, No. 9

In contrast to F1, Fz, and FB, Ft appears to contain more than one myoglobin component as well as non-heme proteins. When FS is analyzed electrophoretically on starch gels (Fig. 4), it mi- grates at pH 8.6 faster than either FI or Fz. As isolated from DEAE-cellulose (Fig. I), it has a spectrum similar to that of alkaline metmyoglobin. Although it readily reacts with cyanide, as well as with oxygen or carbon monoxide after reduction with dithionate, its electrophoretic mobility is unaffected. Rechro- matography, with the use of a salt gradient with DEAE-cellulose (Fig. 3) at pH 7.8, partially resolves F3 into six different components. Despite these differences in the properties of Fa, the presence of small quantities of contaminating protein in FaD, the only F1 component with a 415/280 rnp ratio similar to F1 or FP, prevent definite conclusions to be made about primary structural differences.

It is interesting to compare the chromatographic behavior of human myoglobin with the behavior of myoglobin from other species. Brown (37) reported recently the chromatographic analysis of whale, bovine, and tuna myoglobins on DEAE- cellulose under conditions similar to those developed here. At pH 8.2 to 8.6 in 0.005 to 0.05 M Tris buffer, the bovine and whale myoglobins chromatographed as single, apparently homogenous proteins. Bluefin tuna myoglobin, however, shows three peaks, each of which is analogous both in point of emergence and in amount to one of the three major peaks in human myoglobin. Lack of detailed studies of the three tuna components prevents more detailed comparison. Chromatographic characterization of the myoglobin from other species has been achieved with carboxylic acid ion-exchange resins or celluloses. Rumen (38) has shown that five distinct components can be resolved from whale myoglobin at pH 8.6, all of which are reported to possess the same spectral properties. From the limited data presented it is impossible to judge whether any of these components correspond to acid or alkaline metmyoglobins. Edmundson and Hirs (39), with Amberlite IRC-50 at pH 5.86, also found five components in whale myoglobin. Because only acid metmyoglobin would be present at this pH, the heterogeneity of whale myoglobin appears to result at least in part from primary structural differences (39). Similarly, Akeson and Theorell (40) have shown that equine myoglobin can be resolved into at least three components at pH 6.9 on carboxymethyl cellulose. This heterogeneity also appears to result from primary structural differences.

TABLE III Amino and carboxyl-terminal sequences in human,

whale, horse, and seal myoglobins

Species

Human.

Whale.

Horse. Seal

Amino terminal sequence Carboxyl terminal sequence

NH2

Gly-Leu-Ser- -Leu-Gly-Phe-Gly-Glu- NH2

Val-Ala-Gly- -Leu-Gly-Tyr-Gly-Giu- NH2

I Gly-Leu- Gly-

-Leu-Asp-Phe-Gly-

Reference

This paper*

(40)

~ (30,33)

(38)

* Detailed sequence analysis of human myoglobin will be pub-

, iuman I \ Nhale

FOrigin + 0 0

07 g,, Q?J I/ 012

O 4

F#

0 V

0 A

FIG. 8. Schematic representation of the tryptic peptide patterns of human and sperm whale myoglobin. Patterns were prepared as described in the text.

Although the properties of human myoglobin that have been determined in this study are similar to those of myoglobins from other species, certain structural differences between human myoglobin and other myoglobins are evident. The amino terminal and carboxyl terminal sequences of four species are shown in Table III. Certain species variations occur although none of these would be predicted to alter markedly properties of the myoglobin. The tryptic peptide patterns show distinct differences from those of other myoglobins that have been analyzed. Stockell (41) has compared the peptide patterns of seven species (5 mammals, 1 bird, and 1 reptile) and found considerable simi- larity in certain unique peptides. The methods employed here for the preparation of peptide patterns differ from those used by Stockell and direct comparison is impossible. The patterns of whale3 and human myoglobin prepared by our methods are shown in Fig. 8. Several similarities are evident among the peptides. On the basis of specific color reactions which certain peptides give for histidine, tryptophan, arginine, or tyrosine, at least eight peptides would seem to correspond to one another in the two myoglobins. The similar peptides are given identical numbers in the figure. The other peptides cannot be related easily to one another although the fact that human myoglobin does not yield an insoluble residue after tryptic digestion accounts for the larger number of peptides seen on the human pattern. Whereas these results are not unexpected, it is surprising that the amino acid composition of whale (39) and human myoglobin differs by at least 35 residues of the approximately 153 residues in each molecule.

SUMMARY

1. Human metmyoglobin can be resolved chromatographically on DEAE-cellulose into at least four distinct heme protein components.

2. The two chromatographic components that account for 75 to 80% of the total myoglobin of normal human muscle appear to differ from one another only in their heme prosthetic group.

3. Some of the chromatographic components present in minor amounts in human myoglobin appear to differ in their primary structures. The exact differences have not been established at this time.

4. The physical and chemical properties of chromatographi-

3 Whale myoglobin (crystalline) was kindly supplied by Dr. -_ -- . lished later. John Kendrew.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

September 1962 G. T. Perko$, R. L. Hill, D. M. Brown, and F. H. Tyler 2827

tally pure human myoglobin are similar to those reported previ- 19. ACHER, R., AND CROCKER, C., Biochim. et Biophys. Acta, 9,

ously. 704 (1952).

5. Certain primary structural differences between human 20. TOENNIES, G., AND KOLB, J. J., Anal. Chem., 23, 823 (1951).

globin and the globin from other species have been shown. 21. LEMBERG, R., AND LEGEE, J. W., Hematin compounds and bile

viaments. Interscience Publ.. Inc.. New York. 1949.

REFERENCES

ROSSI-FANELLI, A., AND ANTONINI, E., Arch. Biochem. Bio- vhus.. 66, 587 (1956).

R&&$ANELLI, ‘A., .AND ANTONINI, E., in A. NEUBERGER (Editor), Symposium on protein structure, John Wiley and Sons, Inc., New York, 1958, p. 140.

DRABKIN, D. L., J. Biol. Chem., 183, 317 (1950). GINGER. L. D.. WILSON. G. D.. AND SCHWEIGERT. B. S.. J.

Agr. And Fooh Chem., i, 1037 (1954). PERKOFF, G. T., AND TYLER, F. H., Metabolism, 7, 751 (1958). LUGINBIJHL, W. H., Proc. Sot. Exptl. Biol. Med., 106, 504

(1960).

22. S&&H, E. ‘L., AND BROWN, D: M.,‘J. Biol. &em., 196, 525 (1952).

1.

2.

3. 4.

5. 6.

7. 8.

9.

10.

11.

12.

13.

14.

15.

S&T&S, O., Biochem. J., 71, 585 (1959). HAUT. A.. TUDHOPE. G. R.. CARTWRIGHT. G. E.. AND WIN-

TROBE, &. M., J. kin. Invest., 41, 579, (1962). ’ HILL, R. L., SWENSON, R. T., AND SCHWARTZ, H. C., J. Biol.

Chem., 236, 3182 (1960). FRANKEL-CONRAT, H., HARRIS, J. L., AND LEVY, A. L., in

D. GL~CK (Editor), Methods of biochemical analysis, vol. II, Interscience Publ., Inc., New York, 1955, p. 359.

SPACKMAN, D. H., STEIN, W. H., AND MOORE, S., AnaZ. Chem., 30, 1190 (1958).

MOORE, S., SPACKMAN, D. H., AND STEIN, W. H., Anal. Chem., 30, 1185 (1958).

GUBLER, C. J., LAHEY, M. E., CHASE, M. S., CARTWRIGHT, G. E.,. AND WINTROBE, M. M., Blood, 7, 1075 (1952).

SOBER. H. A.. AND PETERSON. E. A.. Federation Proc.. 17. 1116’(1958).’

I

ROSSI-FANELLI, A., AND ANTONINI, E., Arch. Biochem. Bio-

23. SMITH, E. L., BROWN, D. M., AND LASKOWSKI, M., J. Biol. Chem., 191, 639 (1951).

24. LONGSWORTH, L. G., J. Am. Chem. Sot., 74, 4155 (1953). 25. SCHACHMANN, H. K., in S. P. COLOWICK AND N. 0. KAPLAN

(Editors), Methods in enzumoloau. vol. IV. Academic Press, I%ew York, 1957, p. 32. ” ““’ ’

26. HANES. C. S.. HIRD. F. J. R.. AND ISHERWOOD. F. A.. Biochem. J., Si, 25 (1952). ’ ’

I I

27. SHELTON, R., AND SCHROEDER, W. A., J. Am. Chem. Sot., 82, 3342 (1960).

28. SJ~QUIST, T., Acta Chem. &and., 7, 447 (1953). 29. EDMAN, P., AND SJ~QUIST, T., Acta Chem. i&and., 10, 1507

(1956). 30. GUIDOTTI, G., Biochim. et Biophys. Acta, 42, 177 (1960). 31. HOLLEMAN, J. W., AND BISERTE, G., Biochim. et Biophys. Acta,

33, 143 (1959). 32. DAUTREVAUX, M., BOULANGER, Y., AND BISERTE, G., Bull.

Sot. Chim. Biol., 43, 533 (1961). 33. ROSSI-FANELLI, A., CAVALLINI, D., AND DEMARCO, C., Bio-

chim. et Biophys. Acta, 17, 377 (1955). 34. SMITH, E. L., AND STOCKELL, A., J. Biol. Chem., 207,501 (1954). 35. HILL, R. L., AND SCHMIDT, W. R., J. Biol. Chem., 237, 389

(1962). 36. GEORGE, P., AND HANANIA, G., Biochem. J., 62, 517 (1952). 37. BROWN, W. D., J. Biol. Chem., 236,2238 (1961). 38. RUMEN, N. M., Acta Chem. &and., 13, 4 (1959). 39. EDMUNDSON, A. B., AND HIRS, C. H. W., Nature, 199, 663

phys., 72, 243 (1957). 16. BLOCK, R. J., Arch. Biochem. Biophys., 31, 266 (1951). 17. SMITH, I., Nature, 171, 43 (1953). 18. JEPSON, J. B., AND SMITH, L., Nature, 172, 1100 (1953).

40. A::::: d., AND THEORELL, H., Arch. Biochem. Biophys., 91, 319 (1960).

41. STOCKELL, A., J. Mol. Biol., 3, 362 (1961)


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Gerald T. Perkoff, Robert L. Hill, Douglas M. Brown and Frank H. TylerThe Characterization of Adult Human Myoglobin

1962, 237:2820-2827.J. Biol. Chem.

http://www.jbc.org/content/237/9/2820.citation

Access 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/237/9/2820.citation.full.html#ref-list-1

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


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/content/237/9/2820.citation

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;237/9/2820&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/237/9/2820.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=237/9/2820&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/237/9/2820.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/237/9/2820.citation.full.html#ref-list-1

http://www.jbc.org/

the characterization of adult human myoglobin*

Documents