Differentiation of Myoglobin and Hemoglobin in Biological Fluids

ERNEST C. ADAMS, P h .D.

Research Products Division, Miles Laboratories, Inc., Elkart, IN 46515

ABSTRACT

The use of several methods to differentiate myoglobin from hemoglobin has been investigated. The immunochemical methods, particularly those of hemagglutination inhibition and radioimmunoassay, are the most useful. This report summarizes work in the Ames Research Laboratory over the past 17 years with the several methods.

Introduction haptoglobin-hemoglobin complex is Myoglobin, the oxygen binding pig large, so that it is not excreted in the urine. m ent of muscle, and hem oglobin, the oxy Thus, no hemoglobin appears in the urine gen binding pigment of erythrocytes, are until the blood level of hemoglobin ex alike, yet different. They perform similar ceeds the haptoglobin binding capacity. functions and undergo many of the same Once all the haptoglobin is bound, reactions. There are some physical and hemoglobin appears in the plasma, both chemical differences.3,11 Despite their bound to albumin as methemalbumin and many similarities, myoglobin and hemo as free hemoglobin. globin are immunologically different. This Myoglobinuria is often inferred from difference, and some of the physical and the clinical symptoms of muscle weakness chemical differences, can be used to dif and pain,11,20'21 when there is a ferentiate one from the other. With the peroxidase-like pigment in the urine with exception of the immunochemical few or no erythrocytes seen on the micro methods, this usually requires the pres scopic examination. A rise in serum ence of relatively large quantities of creatine phosphokinase (CPK) is often myoglobin and hemoglobin. used to bolster the suspicion.21 If the labo Myoglobin is released into the blood ratory is asked to confirm this presump plasma as a result of damage to muscle tive diagnosis, a classical method such as tissue. Because of its small size and lack of solubility in 80 percent saturated am binding to haptoglobin, it is rapidly re monium sulfate might be used. When moved from the blood by the kidney and there is insufficient pigment in the urine excreted into the urine. Hemoglobin, to be clearly visible, there generally is no when released from the red cell by in- suspicion of hemoglobinuria or myo travascular hemolysis, is rapidly and globinuria. The conditions would be de tightly bound to haptoglobin. The tected only by routinely running occult 493 0091-7370/80/1100-0493 $01.20 © Institute for Clinical Science, Inc. 494 ADAMS blood tests. The differentiation would munization, the portion of gel containing then have to be by the more sensitive im the myoglobin bands is electroeluted. munochemical tests. Colored eluted fractions are pooled and the absorbance in the Soret band region is M eth od s measured. After the elution, the myoglo bin is in the ferro form; at a concentration N o n -Immunological M e t h o d s of one mg per ml, this has an absorbance of The more classical methods of differ 5.8 at 422 nm. From this value and the entiation, such as ammonium sulfate sol volume of elute, the total amount of myo ubility, ultrafiltration, gel exclusion globin is calculated. The elute, which con chromatography, affinity chromatog tains buffer matetial as well as myoglobin, raphy, and spectra, are described and dis is lyophilized. For immunization, the cussed in the earlier publication.3 In some myoglobin can be either human or of these, the literature method was mod monkey. ified by using occult blood tests to in The hemoglobin immunogen is pre crease the sensitivity. pared from laked erythrocytes as de scribed earlier3 and purified by electro phoresis. I mmunochemical D ifferentiation The earlier immunization scheme3 has o f M y o g l o b in a n d H e m o g l o b in been modified in dose and route. The ini The most definitive differentiation of tial immunization in Freund’s complete hemoglobin and myoglobin in biological adjuvant is given into the four footpads of fluids can be achieved by using relatively rabbits or into the four dew claws of goats. simple immunochemical methods. These The boosters in Freund’s incomplete ad methods, which include immunodiffu juvant are given subcutaneously or in sion, hemagglutination inhibition, and tramuscularly (roughly equal amounts immunoelectrophoresis, are dependent into the four quarters of the animals). The upon the fact that specific antisera will initial dose is 45 fig per animal and boos react only with its homologous antigen. ter doses are 22.5 fig per animal. The Moreover, the methods will quantitate the schedule of boosting and bleeding remain hemoglobin or myoglobin present. the same. The antisera are centrifuged Preparation of Antisera. The isolation and processed as soon as a good clot has and purification of the immunogens, the formed. On immunodiffusion or im immunization schedules, and the process munoelectrophoresis, the antisera to ing of the antisera are essentially the same myoglobin should give a precipitin line as previously reported.3 Myoglobin is iso against a myoglobin solution or a urine lated from human or rhesus monkey mus containing myoglobin at levels of 40 to 50 cle according to the selective precipita fig per ml and no line against human al tion method of Luginbuhl.18 Hemoglobin bumin at levels of 1000,100, and 10 fig per and proteins other than myoglobin are ml or against human serum. The antisera precipitated from extracts of muscle at pH to hemoglobin should give a precipitin 8 with 80 percent ammonium sulfate sat line against hemoglobin at levels of 40 to uration. Under these conditions, the 50 ¡jug per ml and no lines against albumin myoglobin remains in solution. Myoglo or other serum proteins at levels of 1000, bin is precipitated at pH 7 with 100 per 100, and 10 fig p er ml. cent ammonium sulfate saturation. The Methods for Detecting Antigen- myoglobin is further purified by elec Antibody Reaction. In this laboratory, the trophoresis on acrylamide gel which re methods employed for the immunochem moves albumin and other serum protein ical reactions of myoglobin and hemoglo contaminants. To obtain material for im bin are immunodiffusion, immunoelec MYOGLOBIN AND HEMOGLOBIN DIFFERENTIATED IN BIOLOGICAL FLUIDS 495 trophoresis, and hemagglutination inhibi protein-buffer without antisera. One hori tion for urine, radioimmunoassay for zontal row is marked for a negative urine serum myoglobin, and latex agglutination (titer row) one is marked for each standard for special applications. Immunodiffusion urine, and one for each unknown urine. In and immunoelectrophoresis are the same each well of the first vertical row is placed, as described earlier,3 but with modifica with a Pasteur pipet (0.025 ml), one drop tions of media and buffer as in a later pa of the 1:20 antiserum dilution; in each per.5 Precipitin lines develop within four well of the second vertical row is placed to 16 hours. one drop of the 1:40 dilution. This is con Hemagglutination Inhibition. M yoglo tinued through the ninth vertical row. In bin is coupled to formalinized sheep red the tenth row is placed one drop of the blood cells as previously described3 with protein-buffer. One drop of negative urine modifications. After dialysis in the carbo is added to all the wells of the first hori nate buffer, the crude myoglobin solution zontal row; one drop of standard or un is diluted so that it has an absorbance of known urine is added to each well of the 0.79 at 410 nm. One half ml of this solution other horizontal rows. One drop of the ap is reacted with 6 mg pyrrole-2-carboxylic propriate conjugate is added to all the acid azide at pH 9.5. The packed cells wells. from 7 ml of the 10 percent formalinized The contents of the wells are mixed sheep erythrocytes are reacted with 2 ml either by vigorously rotating the whole bis-diazobenzidine at pH 3.5. The plate or by stirring each well with an myoglobin-pyrrole component is reacted applicator stick or toothpick. The plate is with the diazo-benzidine-cells at pH 7. placed either over a mirror or on a white Hemoglobin is coupled to formalinized background for one hour. A positive is in sheep red blood cells.3 dicated by a ring or button of cells in a well For both myoglobin and hemoglobin to the left of the first well with a ring in the conjugates, the stabilized diazonium salt, titer row. The greater the number of wells Fast Black B, can be substituted for the with rings, the greater the concentration bis-diazobenzidine.6 A solution of 1.2 mg of myoglobin or hemoglobin. Fast Black B per ml is equivalent to the The best standard for myoglobin is a prepared bis-diazobenzidine. The prep urine containing a large amount of myo aration of pyrrole-2 carboxylic acid azide globin but no hemoglobin. The amount of has been described.6 myoglobin can be determined by a The antisera are heated at 56° to destroy peroxidase activity method and then the complement and absorbed with for urine can be diluted with negative urine malinized sheep red blood cells as de to give appropriate standards. A standard scribed.3 Two fold dilutions of the anti for myoglobin can be prepared by sera (starting with 1:20) are made in 0.2 M, chromatography of a muscle extract on pH 8.5 Bicine buffer containing 1 percent Sephadex® G-100. The first peroxidase-, nonimmune rabbit serum. The number of like reactive band is catalase, the second dilutions will depend on the titer, but one is hemoglobin, and the third is myo nine dilutions are usually prepared. globin. This myoglobin is not as im- Microtiter plates* (the titer may vary munoreactive as myoglobin excreted in slightly depending on the particular urine. If treated with 6 M urea, which then brand) are used for the titration. Ten verti is removed on a column of Sephadex G-25, cal rows on the tray are marked for the the immunoreactivity is increased to that nine antisera dilutions and one for of excreted myoglobin. The amount of myoglobin is determined by the adsorp tion in the soret band region (about 410 * U bottom Ames Autotray, Cooke, or Linbro. nm): A (mg per ml) = 10. 496 ADAMS

Hemoglobin standards can be prepared ously mixed. After 30 minutes, 0.2 ml of from urines and hemolysates in a similar the second antibody is added to each tube. m anner. The tubes are vigorously mixed. After 30 Serum Myoglobin. Because of the rela minutes, the tubes are again mixed and tively low amounts of myoglobin in serum then centrifuged at 3000 rpm for 15 min compared to that in urine (nanograms vs. utes. The supernates are decanted; the micrograms) and because of interference lips of the tubes (while in an inverted by other serum proteins, the hemaggluti position) are wiped with a tissue. The nation inhibition method is generally not tubes are inverted on a towel. The tubes suitable for serum myoglobin. If the are counted in a well counter. The counts serum myoglobin concentration is high, as of the serum blank are subtracted from the may be caused by renal shutdown so that counts of the other tubes. The net counts the serum proteins can be diluted out, the of the standards and unknowns are ex hemagglutination inhibition method may pressed as percentages of the net count of be used. Serum myoglobin may be deter the 100 percent blank. The percentage of mined by separating myoglobin from most the standards is plotted on log logit paper of the serum proteins by passing it against the ng of myoglobin. From this through an Amicon XM 50 filter and then curve the myoglobin in the unknown concentrating it on a UM 10 filter before serums is calculated. A similar commer using the hemagglutination inhibition cial kit is available.* method. Radioimmunoassays are best for The use of the hemagglutination serum myoglobin. A column radioim method in detecting myoglobin in tissues munoassay was reported previously.5 A and tissue culture and a fluorescent anti double antibody method can be set up in body technique for detecting myoglobin the following manner. deposited in the kidney was described in For the first antibody, goat anti-human the earlier publication.3 The use of im myoglobin is diluted in goat gamma munochemical methods to distinguish globulin solution (1 mg per ml in 0.2 M, endogenous hemoglobin from myoglobin pH 8.5 Bicine buffer) so that 0.1 ml will and hemoglobin of dietary origin was de bind 30 to 40 percent of the labelled myo scribed at the 1973 seminar.4 globin. For the second antibody, the solu tion used is rabbit anti-goat gamma globu Results and Discussion lin in 0.2 M, pH 8.5 Bicine buffer, 0.03 M in ethylenediamine tetraacetic acid M any laboratories are d epending on the (EDTA), and containing 2 percent Brij 35. clinical history or the classical ammonium Labelled myoglobin and standards are sulfate precipitation method for differ prepared in the same manner as for the entiating hemoglobinuria from myo column method. globinuria. The original method used only Disposable tubes (75 x 12 mm) are the appearance of the urine before and labelled 100 percent blank, serum blank, after 80 percent saturation with am each standard, and each unknown serum. monium sulfate.8 The sensitivity is in Into each tube is placed 0.1 ml standard or creased by combining this with an occult unknown serum as labelled; 0.1 ml non- blood test.3 With this technique, many immune serum is used for both the blanks. urines which contain myoglobin when To each tube, except the serum blank, is analyzed by immunochemical methods, added 0.1 ml of the first antibody. To the are reported to contain only hemoglobin. serum blank is added 0.1 ml Bicine buffer. Chu et al9 suggest that the ammonium sul The tubes are vigorously mixed. After 15 fate method fails many times because the minutes, 0.1 ml of labelled myoglobin is added to each tube. The tubes are vigor * Nuclear Medical Systems. MYOGLOBIN AND HEMOGLOBIN DIFFERENTIATED IN BIOLOGICAL FLUIDS 497 pH is not adjusted above 7.5. This is not norm al levels to be below 5 ¡xg p er m l and the complete answer since pH 8 has been following myocardial infarction to be used routinely by us and yet these false above 50 /xg per ml. Both of these num positives for hemoglobin are obtained. In bers are 10 times higher than ours. It is the ultrafiltration method, myoglobin believed this is a reflection of an error in sometimes fails to pass the filter. standards. Both the ammonium sulfate precipita Using the radioimmunoassay, it was tion and the ultrafiltration methods also found in our laboratory that the levels of fail to detect a small amount of one of the myoglobin in the serum of 10 normal per pigments in the presence of a large sons ranged from 0 to 69 ng per ml with a amount of the other. Only the im mean value of 26 and a standard deviation munochemical methods will do this. of 26. This is in agreement with values reported by Jutzy13 and by Stone.23 There M in i m a l D e t e c t a b l e L e v e l s is a wide variation in reported hemoglo In addition to being more discriminat bin levels in plasma depending on the ing, the immunochemical tests have the care with which the blood is collected and lowest minimal detectable levels of myo the method of assay. Naumann19 reported globin or hemoglobin. The radioim values of 0.3 to 2.5 mg per dl with an aver munoassay will measure 5 to 10 ng per ml age of 1.3 mg per dl ± 0.12 S.D. of serum.5,13,23 The hemagglutination Originally, our immunochemical method will detect and measure from 0.3 methods were developed because the to 1 ¡xg of myoglobin or hemoglobin per question was asked, “Why are the occult ml of urine. The immunodiffusion and blood tests positive and yet no red blood immunoelectrophoresis systems are gen cells are seen under microscopic examina erally positive w ith 5 to 10 ¡xg p er ml. The tion?” Longfield17 reported that out of occult blood sticks will detect from 1 to 10 2,700 routine hospital urines, 145 gave fig of hemoglobin or myoglobin per ml of positive occult blood reactions when no urine, but this is without differentiation of red blood cell or only an occasional cell the two pigments. It would be expected was seen. In 1962,1 it was reported by us that when combined with the stick test, that of 2,050 urine specimens, 212 con the ammonium sulfate precipitation and tained occult blood and had sufficient the ultrafiltration methods should be able volume for further study. By behavior of to detect 10 ¡xg of hemoglobin or myoglo the reactive material on ultrafiltration, ion bin per ml. Chu9 however, found a mini exchange cellulose, coprecipitation with mal detectable level of 50 ¡xg per ml with albumin by ammonium sulfate, and pre the ammonium sulfate precipitation cipitation with nanonoic and decanoic m ethod. acid followed by examination of their spectra, circumstantial evidence of the N o r m a l L e v e l s presence of hemoglobin was obtained. The normal levels of hemoglobin or In retrospect, it is now realized that myoglobin in urine have never been some of the urines may have contained clearly defined. From our experience myoglobin. In the past two years, 6,888 with the hemagglutination inhibition urine specimens were tested for occult methods and the stick test, it was found blood by a research laboratory at a local that the normal amount of the pigments in hospital. Of these specimens, 1,025 gave urine is less than 0.3 (xg per ml. With vig positive occult blood tests and 187 of orous muscular activity, myoglobin in these were shown by the hemagglutina urine sometimes increases to detectable tion test to contain myoglobin. levels. Saranchak and Bernstein22 re Most of the studies have been carried ported using radial immunodiffusion out to show myoglobinuria following 4 9 8 ADAMS myocardial infarction. In table I are could be established by an objective sign shown the positives obtained with the var such as onset of chest pain, it was shown ious studies. The studies are not directly that the serum myoglobin level rose in the comparable to each other since the rea first six hours, peaked in about 12 hours, gents and the form of the test have varied and fell to normal within 24 hours. In over the period of time. The diagnosis of some cases the serum myoglobin level myocardial infarction was based on clini remained elevated, indicating continued cal symptoms, ECG changes, and enzyme damage to the heart. In other cases, the elevations, but ultimately on the willing level peaked, fell to normal, and then later ness of the participating cardiologist to rose again,—probably indicating a new at commit himself. When other conditions tack. Sometimes there were elevated known to cause myoglobinuria are noted, levels of myoglobin in the serum drawn positive tests in the absence of myocardial on admission. This suggests that the in infarction are negligible. farction had occurred some time earlier In a previous report on the radioim and was generally confirmed by the clini munoassay for serum myoglobin,5 there cal history. was a scattergram showing the distribu The importance of the timing of the col tion of values of serums from normal per lection of the specimen was also seen in sons, hospitalized patients without all the studies with detection of urinary myocardial infarction, and patients with myoglobin. Usually, the amount of uri diagnosed myocardial infarction. Nine nary myoglobin in the first 12 hours fol teen of the 24 patients diagnosed as hav lowing the insult was greater than in the ing myocardial infarction had at least one second 12 hours. For 15 of the 24 patients elevated serum myoglobin. For two of the with myocardial infarction, it was judged five patients without elevated serum that both urine and serum had been col myoglobin, the serum enzymes (total lected at the critical times. The sera was CPK, SGOT, LDH) were not elevated, the assayed by radioimmunoassay and the urine myoglobin was negative, and the urines by hemagglutination inhibition. ECG was questionable; however, the F ourteen of the patients had m yoglobin in diagnosis was myocardial infarction. For both serum and urine. One had myoglobin the other three patients without elevated in serum, but not in the urine. This may serum myoglobin, it is likely the critical reflect the differences in sensitivities of sera were not available. Values as high as the methods. 1120 ng per ml were found with most A variety of conditions resulting in myo- being in the 300 to 500 ng per ml range. globinemia and myoglobinuria or in hemo- When the time of myocardial infarction globinemia and hemoglobinuria were listed in earlier publications.3'10’14’20,21 T A B L E I During the various studies we, or our

Myoglobinuria in Myocardial Infarction cooperating investigators, have noted myoglobinuria in cases of dermato-

Number Percent with myositis, trichinosis, fever and virus in Institution of Cases Myoglobinuria fections, arterial occlusion, carbon monoxide intoxication, crush injuries EH-12 44 89 (both myoglobin and hemoglobin), drug EH-2 24 83 LB 7 86 overdose, heroin addiction, collagen vas B 12 33 67 P - l 15 13 85 cular disease, myositis, alcoholic binges, P-2 32 84 myasthenia gravis with collagen disease, B H 16 37 92 MM 3 100 electrical shock, malignant hypertension, K-l7 8 88 disorder of muscle lipid metabolism, and K-2 17 95 heart and arterial surgery. Some of these MYOGLOBIN AND HEMOGLOBIN DIFFERENTIATED IN BIOLOGICAL FLUIDS 499 have been associated with renal failure.10 Master of Science in Nursing. Indiana Univer sity, Indianapolis, 1975. In the case of the patient with trichinosis, 8. B l o n d h e i m , S. H ., M a r g o l ia s h , E., a n d enough serum myoglobin became availa Sh a f r i r , E.: A simple test for myohemo- ble, owing to renal shutdown, that the globinuria (myoglobinuria). J. Amer. M ed . As soc. 167:453-454, 1958. serum could be diluted to give a positive 9. C h u , S. Y., C u r t is , C ., and T u r k in g t o n , V. E.: indication in the hemagglutination test. Influence of pH on the simple solubility test for The urine of this patient was reported to myoglobinuria. Clin. Biochem. J 7 :2 3 0 -2 3 1 , 1978. contain only hemoglobin by a laboratory 10. C if u e n t e s , E., N o r m a n , M. E., S c h w a r t z , using the ammonium sulfete solubility M. W ., M a l e y , B ., and Ba s o n , W .: Myo method. The immunochemical methods globinuria with acute renal failure in children. Clin. Pediat. 75:63-66, 1976. showed only myoglobin. 11. C o m in g s , D. E. and Rosenfeld, H.: The presence of myoglobin in kidney Idiopathic paroxysmal myoglobinuria. Ann. Int. slices from one crush injury victim was Med. 55:647-661, 1961. 12. H a m m a c k , W . J. and S h e e h y , T. W .: shown by the indirect fluorescent stain. It Myoglobinuria—its value as a diagnostic test of was not possible to find myoglobin in the myocardial infarction. Clin. Res.X77:49A, 1975. urine from dystrophy patients. It may be 13. Ju t z y , R. V., N e v a t t , G . W., P a l m e r , F . J., and N e l s o n , J. C.: Radioimmunoassay of serum these patients had already lost too much myoglobin in acute myocardial infarction. muscle mass and thus were not secreting Amer. J. Cardiol. 35:147, 1975. detectable amounts of myoglobin. 14. K a g e n , L. J.: Myoglobin, Biochemical, Physiological, and Clinical Aspects. New York, The immunochemical tests for hemo Columbia University Press, 1973, pp. 86-116 globin were used to differentiate hemo 15. K e s s l e r , H. A ., L i e b s o n , P. R., globin of endogenous origin from gas M attenheiner , H., and Ad a m s , E. C.: Acute myocardial infarction diagnosed by myo trointestinal bleeding from the myoglobin globinuria. Arch. Intern. Med. 735:1181-1183, and hemoglobin of dietary origin.4 1975. 16. L e v in e , R. S., Al t e r m a n n , U., G u b n e r , R. S., and Ad a m s, E. C.: Myoglobinuria in myocardial References infarction. Amer. J. Med. Sci. 262:179-183, 1971. 1. A d a m s , E . C., F e t t e r , M . C., F r e e , H . M ., and 17. Longfield, G: M., H o l l a n d , D. E., L a k e , A. F r e e , A. H.: Hemolysis in heinaturia. J. Urol. J., and K n ig h t s , E. M.: Hematuria: Compari 88:427-430, 1962. son of chemical with microscopic examination. 2. Ad a m s , E . C. and E l l i o t t , T. A.: Urinary J. Mich. State Med. Soc. 59:785-786, 1960. myoglobin in myocardial infarction. J. Amer. 18. LUGINBUHL, W. H.: Immunologic investigation Med. Assoc. 2il:1013-1015, 1970. of preparations of human myoblogin. Amer. J. 3. Ad a m s , E. C.: Differentiation of myoglobin and Clin. Path. 38:487-493, 1962. hemoglobin in biological fluids. Ann. Clin. Lab. 19. N a u m a n n , H. N.: The measurement of hemo Sci. 7 :208-221, 1971. globin in plasma. Hemoglobin. Its Precursors 4. A d a m s E. C. and L a y m a | j , K. M.: Im and Metabolites. Sunderman, F. W. and Sun munochemical confirmation of gastrointestinal derman, F. W. Jr., eds. Philadelphia, J. B. Lip- bleeding. Ann. Clin. Lab. Sci. 4:343-349, 1974. pincott Co., 1964, pp. 40-48. 5. A d a m s , E. C., L a y m a n , K. M., and J a u n a k a is , 20. R o w l a n d , L. P., F a h n , S., Hirschberg, E., I.: Radioimmunoassay for serum myoglobin. and H a r t e r , D. H .: Myoglobinuria. Arch. Ann. Clin. Lab. Sci. 8:330-341,1978; Manual of Neurol. 70:537-562, 1964. Procedures for Seminar on Proteins and Pro- 21. R o w l a n d , P. and P e n n , A. S.: Myoglobinuria. teinopathies. Sunderman, F. W-, ed. Philadel Med. Clin. North Amer. 56:1233, 1972. phia, Institute for Clinical Science, 1977, pp. 22. Sa r a n c h a k , H. J. and B e r n s t e in , S. H.: A new 111-134. diagnostic for acute myocardial infarction. The 6. Ad a m s , E . C ., L a y m a n , K. M ., J o h n s o n , P. K., detection of myoglobinuria by radioim- and Pa p e n m e ie r , G. J .: Fast black B as a substi munodiffusion assay. J. Amer. Med. Assoc. tute for bis-diazobenzidine to couple proteins 228:1251-1255, 1974; 229:287, 1974. to erythrocytes. J. Immunol. Methods 25:383- 23. St o n e , N. J., W i l l e r s o n , J. T., G o m e z - 388, 1979. Sa n c h e z , C. E., and W a t e r m a n , M. R.: Radio 7. A l v o r d , C.: Myoglobinuria in the diagnosis of immunoassay of myoglobin in human serum. acute myocardial infarction. Thesis in partial Results in patients with acute myocardial in fulfillment of requirements for the Degree of farction. J. Clin. Invest. 56:1334-1339, 1975.