ABNORMALITY OF HEMOGLOBIN MOLECULES IN HEREDITARY HEMOLYTIC ANEMIAS* LINUS PAULING Professor of Chemistry, California Institute of Technology, Pasadena, California T WENTY years ago, after having worked for a decade on the determination of the structure of relatively simple inorganic and organic molecules, I became interested in hemoglobin. This interest arose from the consideration of the structural origin of the sigmoid oxygen equilibrium curve. 1 It was soon extended to include the denaturation of hemoglobin and other proteins2 and the mag- netic properties of hemoglobin and its derivatives.3-11 The study of magnetic properties has been especially fruitful in providing in- formation about the nature of the bonds formed by the iron atoms in hemoglobin with the neighboring atoms of the porphyrin ring system, the globin, and attached molecules such as the oxygen mo1ecule.3~12-1" The discovery of the abnormal hemoglobins was the result of the consideration of hypothetical molecular mechanisms of the disease. In the spring of 1945 I, together with eight men from medical schools of the country, was serving as a member of the Medical Advisory Committee which assisted in the preparation of the Bush Report. I6 One evening Dr. William B. Castle, Professor of Medicine in Harvard University, mentioned to the other mem- bers of the Committee the disease sickle-cell anemia, with which he had had some experience. He told about the discovery of the disease by Dr. J. B. Herrick, in 1910, I' and described the character- istic change in shape of the red corpuscles and the effect of oxygen in preventing the sickling and of carbon dioxide in accelerating it. I suggested that the action of carbon dioxide was to accelerate the dissociation of oxygen from oxyhemoglobin, through the Bohr- Hasselbalch effect (it had in fact been clearly stated by Hahn and Gillespie18 in 1927 that sickling occurs only when the partial pres- * Lecture delivered April 29, 1954. 216 OXIDATIVE PHOSPHORYLATION 215 48. Sanadi, D. R., and Littlefield, J. W. 1953. J. Biol. Chem. 201, 103. 49. Judah, J. D. 1951. Biochem. J. 49, 271. 50. Slater, E. C. Personal communication. 51. Slater, E. C. 1950. Nature 166, 982. 52. Lehninger, A. L., Hassan, M. u., and Suddutb, H. C. J. Biol. Cbem. In press. 53. Ahmad, K., Schneider, H. G., and Strong, P. M. 1950. Arch. Biorhem. 28, 281. 54. Nielsen, S. O., and Lehninger, A. L. 1954. J. Am. Cbem. Sot. 76, 3860. 55. BorgstrGm, B., Sudduth, H. C., and Lehninger, A. L. In press. 56. Vernon, L. P., Mahler, H. R., and Sarkar, N. K. 1952. J. Biol. Chem. 199, 585, 599. 57. Lardy, H. A., and Wellman, H. 1952. J. Biol. Cbem. 195, 215. 58. Chance, B., and Williams, G. R. 1954. Federation Proc. 13, 190. 59. Krebs, H. A. 1953. Briz. Med. Bull. 9, 97. 60. Ball, E. G. 1944. Ann. N.Y. Acad. Sri. 45, 363. 61. Slater, E. C. 1953. Nature 172, 975. 62. Ogston, A. G., and Smithies, 0. 1948. Physiol. Revs. 28, 283. 63. Lardy, H. A., and Wellman, H. 1953. J. Biol. Cbem. 201, 357. 64. Cohn, M. 1953. J. Biol. Cbem. 201, 735. 65. Boyer, P. D., Harrison, W. H., Falcone, A. B., and Gander, J. G. 1954. Federation Proc. 13, 185. 66. Pinchot, G. B. 1953. J. Biol. Chem. 205, 65. 67. Bartley, W., and Davies, R. E. 1954. Biocbem. J. 57, 37. 68. Martius, C., and Hess, B. 1951. Arch. Biochem. Biophys. 33, 486. 69. Maley, G. F., and Lardy, H. A. 1953. J. Biol. Chem. 204, 435. 70. Lynen, F., and Kcnigsberger, R. 1950. Ann. Chem. 569, 129; Ziillner, N. 1952. 2. Pbysiof. Cbem. 291, 12. 71. Ennor, A. H., and Rosenberg, H. 1954. Biocbem. J. 56, 302, 308. 72. Barkulis, S. S., and Lehninger, A. L. 1951. J. Biol. Cben. 193, 597. ABNORMALITY OF HEMOGLOBIN MOLECULES IN ANEMIA 217 sure of oxygen is small), and I pointed out that the relation of sickling to the presence of oxygen clearly indicated that the hemo- globin molecules in the red cell are involved in the phenomenon of sickling, and that the difference between sickle-cell-anemia red corpuscles and normal red corpuscles could be explained by postu- lating that the former contain an abnormal kind of hemoglobin, which when deoxygenated has the power of combining with itself into long rigid rods, which then twist the red cell out of shape. The opportunity to test this idea arose when Dr. Harvey A. Itano came to the California Institute of Technology, in the fall of 1946. He had been a student of Professor Edward A. Doisy, of St. Louis University SchooI of Medicine, where Dr. Itano had received his M.D. degree in 1945. Dr. Doisy suggested that he work with me, and the opportunity for doing so arose in the course of his year as an intern, when he was awarded an American Chemical Society Predoctoral Fellowship in Chemistry, for the three years 1946 to 1949. In a letter to Dr. Itano I suggested that he investigate the hemoglobin from the red cells of sickle-cell-anemia patients, in order to see whether it was different from normal adult human hemoglobin. On his arrival in Pasadena in September, 1946, he began this investigation. He verified the published reportslB that carbonmonoxyhemoglobin, like oxyhemoglobin, prevents sickling of the red cells, and found that some other hemoglobin derivatives, including alkyl isocyanide-ferrohemoglobin, ferrihemoglobin, ferrihemoglobin azide, and ferrihemoglobin cyanide similarly pre- vent sickling. He developed a rapid diagnostic test for sickle-cell anemia and sickle-cell trait, based on the use of a chemical reducing agent.lD Most of the properties of the hemoglobin from the blood of sickle-cell-anemia patients were found to be the same, to within the error of determination, as those of hemoglobin from normal individuals, but it was finally clearly shown, by careful measure- ment of electrophoretic mobility, that the blood of the patients contains nearly 100 per cent of an abnormal hemoglobin, differing from normal adult human hemoglobin, and that the blood of the parents of patients contains an approximately half-and-half mixture of the abnormal hemoglobin and normal adult human hemo- globin. 20 This electrophoretic work was carried out with the col- laboration of Dr. S. J. Singer and Dr. Ibert C. Wells. 218 LINUS PAULING THE INHERITANCE OF SICKLE-CELL ANEMIA The electrophoretic patterns reported in the first publication20 are shown in Fig. 1. They were made by electrophoresis for 20 hours at a potential gradient of 4.73 volts per centimeter of solu- tions of carbonmonoxyhemoglobins in phosphate buffer of 0.1 ionic strength and pH 6.90. The peaks R and b, representing normal hemoglobin and hemoglobin from the red cells of patients with sickle-cell anemia, are single peaks, corresponding in each (a) Normal (c) Sickle - cell trait lb) Sickle -cell anemia (d) Mixture of (a) and (b) FIG. I. Longsworth scanning diagrams of carbonmonoxyhemoglobin in phosphate buffer of 0.1 ionic strength and pH 6.90, taken after 20 hours electtophoresis at a potential gradient of 4.73 volts per centimeter. case to an electrophoretically homogeneous material. The electro- phoretic mobilities are different for the two hemoglobins; in fact, at this pH the molecules of normal hemoglobin move toward the anode, showing that they have a negative electric charge, and those of sickle-cell-anemia hemoglobin move toward the cathode, show- ing that they have a positive charge. The isoelectric points in phos- phate buffer of ionic strength 0.1 were found to be 6.87 (in pH units) for normal adult human carbonmonoxyhemoglobin and 7.09 for sickle-cell-anemia hemogiobin. The difference between these values is nearly the same as that between the observed values 6.68 for normal ferrohemoglobin and 6.91 for sickle-cell-anemia ferro- hemoglobin. ABNORMALITY OF HEMOGLOBIN MOLECULES IN ANEMIA 219 The electrophoretic diagram of a solution containing a mixture of normal carbonmonoxyhemoglobin and sickle-cell-anemia car- bonmonoxyhemoglobin, in equal amounts, is shown as d in Fig. 1, and that of hemoglobin from the red cells of a parent of a patient is shown as c. The blood from which this hemoglobin was obtained showed the characteristic properties of sickle-cell trait (sicklemia) ; the cells could be made to sickle, on removal of oxygen, but less readily than the cells of a sickle-cell-anemia patient. It is seen that the sicklemic hemoglobin is a mixture of two hemoglobins, pre- sumed to be normal adult human hemoglobin and sickle-cell- anemia hemoglobin, with the normal hemoglobin present in an amount somewhat greater than 50 per cent. The indication of a genetic basis for the sickling of erythrocytes had been recognized by EmmeP in 1917; and Taliaferro and Huck,22 at a time when the distinction between sicklemia and sickle-cell anemia was not clearly understood, suggested that a single dominant gene was involved. The inheritance of sickle-cell disease was then clarified by Neel, who in 15~47~~ had suggested "that there is present in the colored population a certain factor which, when heterozygous, may have no discernible effect, but usually results in sickling, and, when homozygous, tends to result in sickle cell anemia." In 1949 he reported2' that every one of 42 tested parents of children with sickle-cell anemia was found to be sicklemic, their blood containing red cells which could be made to sickle, though less readily than that of the sickle-cell-anemia pa- tients. He concluded that sickle-cell anemia is the result of the homozygous condition of the sickle-cell gene, and sicklemia the result of the heterozygous condition. Beet25 arrived at the same conclusion independently and almost simultaneously. The electro- phoretic patterns shown in Fig. 1 had permitted this inference to be drawn before Neel's paper and Beet's paper were published. Moreover, the gene responsible for the sickling process could be identified with an alternative pair of alleles of which neither one is recessive or dominant, one allele being responsible for a part of the process of manufacture of normal adult human hemoglobin, and the other for the manufacture of sickle-cell-anemia hemoglobin. The fact that all the red cells of a sicklemic individual can be made to sickle by removal of oxygen shows that the cells are not of two 220 LINUS PAULING classes, one containing normal hemoglobin and the other abnormal hemoglobin, but that each cell contains a mixture of the two kinds of hemoglobin. The presence of a larger amount of normal than of abnormal hemoglobin in the blood of sicklemic individuals indicates that the process of manufacture of the abnormal hemo- globin is somewhat less efficient than that of normal hemoglobin. It was suggested in the first paper on sickle-cell-anemia hemo- globinzO that the two genes in the heterozygous individual might compete for a common substrate in the synthesis of two different enzymes essential to the production of the two different hemo- globins, or that competition for a common substrate might occur at a later stage in the series of reactions leading to the synthesis of the two hemoglobins themselves. An investigation of the amount of abnormal hemoglobin in the blood of sicklemic individuals was carried out by Wells and Itan0,26 who found, using the electrophoretic method, that the amount of sickle-cell-anemia hemoglobin varied from 24 per cent to 45 per cent in 42 individuals with sicklemia. Neel, Wells, and Itanoz7 reported a study of 32 sicklemic individuals who were mem- bers of 7 Negro families, comprising 74 individuals altogether. The amounts of abnormal hemoglobin, ranging from 22.3 per cent to 45.2 per cent, showed significant differences between family means. A postulate to explain the apparent inheritance of a factor determining the amount of abnormal hemoglobin in sicklemic blood was made by Itano.z8 He suggested that the differ- ences can be attributed to differences in the rate of synthesis of normal hemoglobin, and that the evidence requires that there be at least three rate-determining modifications of the mechanism of synthesis of normal hemoglobin. Additional contributions to the problem of the genetics of nor- mal and abnormal hemoglobins have been made by Nee12g-31 and other workers.32,33 THE PROPERTIES OF SICKLE-CELL-ANEMIA HEMOGLOBIN AND THE MECHANISM OF SICKLING Sickle-cell-anemia hemoglobin is closely similar to normal adult human hemoglobin in most of its properties.20 The two proteins have approximately the same sedimentation and diffusion constants, ABNORMALITY OF HEMOGLOBIN MOLECULES IN ANEMIA 221 and hence nearly the same molecular weights. The acid-base titra- tion curves of both hemoglobins in the neighborhood of neutrality are linear, a change of 1 pH unit of the solution being associated with the change in charge of the hemoglobin of about 13 electronic charges per molecule. The normal molecule has about three more negative charges than the abnormal molecule in this region. In the search for the structural basis for this difference samples of porphyrin dimethyl esters were prepared from the two hemo- globins, and the samples were shown by their x-ray powder photographs and by identity of their melting points and mixed melting points to be identical. The difference in structure was hence attributed to a difference in the globins. An investigation by Schroeder, Kay, and Wells34 of the amino acid composition of normal adult human hemoglobin (from nor- mal Negro individuals) and sickle-cell-anemia hemoglobin gave results indicating that the hemoglobins do not differ with respect to their content of basic and acidic amino acids; the investigators concluded that sickle-cell-anemia hemoglobin probably contains slightly less leucine and more serine than normal hemoglobin, and possibly less valine and more threonine. These amino acids do not contribute directly to the net charge of the proteins, but they might affect the folding or coiling of the polypeptide chains in such a way as to change the acid or basic constants of other groups. Havinga investigated the phosphorus content, optical rotation, ease of separation of hemes and globin, and number of terminal amino acid residues of normal adult human hemoglobin and sickle-cell-anemia hemoglobin, and found no significant differences between the two proteins. Globins carefully prepared from the two hemoglobins were investigated electrophoretically by Havinga and ItanosE and found to have the same difference in electrophoretic mobility as the hemoglobins themselves. On denaturation by treat- ment with 4 N guanidinium chloride for 1 hour at 4O C. and removal of the guanidinium chloride by dialysis, the globins were found to have increased markedly in heterogeneity, and to have essentially the same electrophoretic properties. These results in- dicate that the normal and abnormal hemoglobin molecules might be composed of the same polypeptide chains, folded, however, in different ways, and that on denaturation with guanidinium ion the 222 LINUS PAULING resulting denatured proteins have the same complex of configura- tions. The interesting possibility exists that the gene responsible. for the sickle-cell abnormality is one that determines the nature of the folding of polypeptide chains, rather than their composition, It was pointed out by Sherman3" in 1940 that sickled red cells are observed under the polarizing microscope to be birefringent, whereas normal cells are optically isotropic. Ponders8 suggested, on the basis of this observation, that in sickled cells the hemoglobin molecules assume an orderly or paracrystalline arrangement, which is responsible for the sickling. A detailed mechanism of the sickling process was suggested in the first paper on sickle-cell- anemia hemoglobin,20 as follows: "We can picture the mechanism of the sickling process in the following way. It is likely that it is the globins rather than the hemes of the two hemoglobins that are different. Let us propose that there is a surface region on the globin of the sickle-cell-anemia hemoglobin molecule which is absent in the normal molecule and which has a configuration complementary to a different region of the surface of the hemoglobin molecule. This situation would be somewhat analogous to that which very probably exists in antigen-antibody reactions.3g The fact that sickling occurs only' when the partial pressures of oxygen and carbon monoxide are low suggests that one of these sites is very near to the iron atom of one or more of the hemes, and that when the iron atom is combined with either one of these gases, the com- plementariness of the two structures is considerably diminished. Under the appropriate conditions, then, the sickle-cell-anemia hemoglobin molecules might be capable of interacting with one another at these sites sufficiently to cause at least a partial align- ment of the molecules within the cell, resulting in the erythrocyte's becoming birefringent, and the cell membrane's being distorted to accommodate the now relatively rigid structure within its con- fines. The addition of oxygen or carbon monoxide to the cell might reverse these effects by disrupting some of the weak bonds between the hemoglobin molecules in favor of the bonds formed between gas molecules and iron atoms of the hemes." A more detailed discussion of the effect of oxygen was made possible by the results of an investigation of the combination of hemoglobin with alkyl isocyanides.40 It was found that ethyl iso- ABNORMALITY OF HEMOGLOBIN MOLECULES IN ANEMIA 223 cyanide, isopropyl isocyanide, and tertiary butyl isocyanide differ greatly in their combining powers with hemoglobin, although they have essentially the same combining power with heme; and this fact was interpreted as showing that the four hemes in the hemo- globin molecule are buried within the globin: "Our postulate provides an obvious explanation of the action of oxygen in pre- venting the sickling of sickle-cell-anemia erythrocytes. We have visualized the sickling process*O as one in which complementary sites on adjacent hemoglobin molecules combine. It was suggested that erythrocytes containing oxyhemoglobin or carbonmonoxy- hemoglobin do not sickle because of steric hindrance of the at- tached oxygen or carbon monoxide molecule. This steric-hindrance effect might be distortion of the complementary sites through forcing apart of layers of protein, as is suggested by the isocyanide experiments." Substantiation of this picture was soon obtained through micro- scopic investigations. Rebuck, Sturrock, and Monaghan4' sub- stantiating the work of Sherman3? observed that in the early stages of sickling the intracellular hemoglobin forms anisotropic aggre- gates, suggestive of incipient crystallization. Perutz and Mitchi- son,42 at the suggestion of Dr. C. A. Stetson of the Rockefeller Institute, compared the dichroism of sickled cells and hemoglobin crystals, and additional studies of the same sort were reported by Peru@ Liquori, and Eirich.43 These investigators found that the dichroism of the sickled cells corresponds to an orientation of the hemoglobin molecules such that the normal to the plane of the heme groups is perpendicular to the long axis of the crystal needles and of the sickled cells. (This statement is based on the paper of Peru& Liquori, and Eirich, *43 there is some conflict with the earlier paper.42) They also pointed out 42,43 that the solubility of sickle- cell-anemia hemoglobin is much smaller than that of normal hemoglobin or of either normal oxyhemoglobin or sickle-cell- anemia oxyhemoglobin. A detailed study of the solubilities of mixtures of sickle-cell-anemia hemoglobin and other hemoglobins has been made by Itano,44 who has shown that a solubility meas- urement provides a simple way of determining roughly the amount of sickle-cell-anemia hemoglobin present in a mixture of hemo- globins. A most significant investigation was then reported by 224 LINUS PAULING FIG. 2. Hemoglobin formed in stroma-free solutions of deoxygenated sickle- cell-anemia hemoglobin. Phase photomicrography, X 375. From John W. Harris, Pm-. Sot. Exprl. Biol. and Med. 75, 197 (1950). AUNORMALITY OF HEMOGLOBIN MOLECULES IN ANEMIA 225 FIG. 3. Sickled erythroqtes in oxygen-depleted whole blood from a patient with sickle-cell anemia, demonstrating the similarities in shape to tactoids formed in stroma-free solutions of their deoxygenated hemoglobin. Phase photomicrography, X 375. From John W. Harris, Proc. Sm. Exptl. Biol. and Med. 75, 197 (1950). 226 LINUS PAULING Harris.45 He showed that a stroma-free solution of sickle-cell- anemia hemoglobin with concentration 15.2 or 23.5 g. per 100 ml. on deoxygenation forms birefringent spindle-shaped bodies vary- ing in length from 1 to 15 p. He identified these bodies as tactoids (liquid crystals) of the nematic type. His photomicrograph is shown as Fig. 2, which may be compared with Fig. 3, a similar photomicrograph (375 x magnification) of sickled erythrocytes. It hence seems probable that sickle-cell anemia can be described as a molecular disease, resulting from the difference in molecular structure of sickle-cell-anemia hemoglobin and normal adult human hemoglobin. The properties of the abnormal hemoglobin are such that when deoxygenated the molecules combine with one another to form long molecular strings, which, through inter- molecular attraction, aggregate into tactoids. These tactoids have enough mechanical strength to distort the red cell, changing the viscosity of the blood, and causing the clinical and pathological manifestations of the disease. SS>SD>AS+Th>SC>CC FIG. 6. A chart representing possible combinations of the alleles A, S, C, D, and E. The horizontal row at the bottom represents simultaneous occurrence with thalassemia minor. The amount of fetal hemoglobin usually present is also indicated: F means a few per cent, F means 10 per cent or more. Ob- served conditions are shown within heavy borders. At the top the seriousness of different kinds of anemia is indicated. The seriousness of these diseases (not including AC + Th and E + Th, for which the number of patients is too small to permit an estimate) is indicated at the top of Fig. 6. There is some variability in the seriousness of the diseases, in part as the result of the extent to which compensation of abnormal hemoglobins is achieved through the manufacture of fetal hemoglobin. The mildest anemias are SC and CC. Not all SC and CC individuals are anemic. Al- though they probably have a greater than normal rate of hemolysis, they may be able to compensate completely with a greater than normal rate of production of red blood cells.52~72 The conditions CD, DD, SE, CE, DE, EE, and DA + Th have not yet been observed; it may be found that they are associated with anemia. iSNORMALITY OF HEMOGLOBIN MOLECULES IN ANEMIA 233 Except for the fetal-hemoglobin fraction, the hemoglobin of patients with thalassemia major has properties reported to be the same as those of normal adult human hemoglobin,58 and it is usu- ally considered that thalassemia is not associated with the produc- tion of an abnormal hemoglobin. There are some facts, however, that indicate that thalassemia hemoglobin is a fifth abnormal kind of adult human hemoglobin, with properties so closely similar to those of normal adult human hemoglobin that the differences have escaped detection. It has been observed that thalassemia minor in- volves a greater interference with the synthesis of the normal hemoglobin than of abnormal hemoglobin. For example, whereas in sicklemic individuals the ratio A/S is always greater than 1, this ratio is much less than 1 in thalassemia:sickle-cell disease. Two patients reported by Neel, Itano, and Lawrence68 were found to have the ratios A: S: F equal to 20 : 61: 19 and 11: 84: 5, respec- tively. The hemoglobin of the patient with thalassemia: hemo- globin-E disease contains no detectable amount of normal adult hemoglobin, but only E and F. The thalassemia allele interferes with the manufacture of normal adult hemoglobin, and seems not to interfere seriously with the manufacture of the abnormal hemo- globins. The simplest explanation of this fact is that the thalassemia allele occupies the same locus in the chromosome as the alleles for the other abnormal hemoglobins, and is itself responsible for an abnormal globin -in which the abnormality is of such a nature as to interfere with the step of inclusion of the hemes in the molecule. If this postulate is correct it should be possible to show a difference . between thalassemia hemoglobin and normal adult human hemo- globin. Under this circumstance there would be six alleles occupying the same locus, A, S, C, D, E, and Th (more than six if there is more than one thalassemia allele). The six states at the bottom of the triangle in Fig. 6 should then be written ATh, STh, CTh, DTh, ETh, and ThTh. Of the possible twenty-one combinations of the six alleles, fourteen would be ascribed to known individuals. The possibility that the thalassemia gene is allelomorphic with the sickle-cell gene has been discussed by Neel.30 He pointed out that both of the children of a patient with thalassemia:sickle-cell dis- ease and married to a normal woman exhibited thalassemia minor, 234 LINUS PAULING and mentioned that if either of these children had been normal (without either thalassemia minor or sicklemia) the hypothesis of allelomorphism would have to be abandoned. Silvestroni and Bianco, who a number of years ago described the disease resulting from simultaneous inheritance of thalassemia minor and sick- lemia,73 have also discussed the genetic aspects of thalassemia and sickling, and have concluded that the two are not allelomorphic.T4 SICKLE-CELL-ANEMIA HEMOGLOBIN AND MALARIA A number of interesting questions about the origin and heredity of the abnormal hemoglobins remain to be answered. It has, for example, been pointed out by LehmannT5 that sickle-ceil anemia seems to be a far less serious disease in Africa than in America. A possible explanation may be that the patients in Africa for some reason manufacture a larger amount of fetal hemoglobin than those in America; the question can presumably be answered by a thorough investigation of the hemoglobins of Africans, which has not yet been carried out. The question of the continued high incidence of the sickle-cell allele, despite its continued loss because of the lethal character of the homozygous condition, has been raised by NeeLso who has suggested three alternative explanations: (1) continued produc- tion of the allele through mutation; (2) the existence of an ab- normal genetic mechanism that favors the heterozygous condition, AS, over the normal condition, AA; (3) a positive selection of the heterozygote, perhaps through increased fertility. The first ex- planation has to be rejected because the rate of mutation that would be required is far greater than any that has ever been observed for any organism. There now exists evidence indicating that the third alternative provides the correct explanation, and that malaria is involved. It was first suggested by Brain76 that the presence of S in the red cells might give protection against malaria parasites, and thus confer an advantage to the sicklemic individual that would balance the disadvantage of the lethal homozygosity. Lehmann" wrote that "The lethal tendency of a gene potentially causing sickle cell anemia may thus be counteracted by its conferring a re- sistance to malaria similar to that found in early infancy." A test of the hypothesis was carried out recently by Allison,`* who in- ABNORMALITY OF HEMOGLOBIN MOLECULES IN ANEMIA 235 fected 15 healthy adult Africans with the sickle-cell trait and 15 similar healthy adult Africans without the sickle-cell trait with P. fdcipawm, by subinoculation with 15 ml. of blood containing a large number of trophozoites or by being bitten by heavily in- fected anopheles mosquitoes, in which the presence of sporozoites was confirmed by dissection of the mosquito. The infection was established in 14 out of the 15 Africans without the sickle-cell trait, and in only 2 of the 15 with this trait. It was concluded by Allison that the abnormal erythrocytes of individuals with the sickle-cell trait are less easily parasitized by P. fdcipavum than are normal erythrocytes, and that accordingly those who are het- erozygous for the S allele have a selective advantage over normal individuals in regions where malaria is hyperendemic. It is, of course, not unreasonable that the abnormal hemoglobin may be less effective than normal hemoglobin in nourishing the parasites. The question of the origin of sickle-cell anemia has been in- vestigated especially by Lehmann, who has studied the incidence of the ability of red cells to sickle in several parts of the world.7e-s1 In Africa the incidence of the sickle-cell allele is highest in the north and east, diminishing somewhat toward the west and south, and being virtually zero in South Africa itself. In India the major groups of the population, classed as Dravidians, show no sickling, and among the pre-Dravidians living in the hills of Southern India the Veddoids alone have a high incidence of sickling. A high incidence of sickling was found also in a Veddoid community of Southern Arabia, although sickling is absent among the Semitic Arabs. Some sicklemic individuals are found in Italy, and in Greece there are a few communities, each of a few hundred individuals, with a high incidence of sickling. Lehmann concIuded7' that it is not unlikely that the sickle-cell gene originated, presumably through a mutation, in a Veddoid community in Southern Arabia, and that it spread from this point to India, the Mediterranean region, and especially to Africa. MOLECULAR DISEASES Sickle-cell anemia has been described as a molecular disease.*O It may be that all diseases can be described as molecular diseases, inasmuch as the human body and the vectors of disease are all com- 236 LINUS PAULING posed of molecules. For example, carbon monoxide poisoning is the result of combination of molecules of carbon monoxide with molecules of hemoglobin. Erythroblastosis fetalis involves the in- teraction of the molecules or haptenic groups of an Rh antigen on red cells with molecules of the homologous antibody. An inborn error of metabolism such as alcaptonuria results from the failure of the body to manufacture molecules of the enzyme that, if present, would catalyze the oxidation of homogentisic acid in the body. Any hereditary disease may be said to be a molecular dis- ease, if the genes are described as molecules, in that it involves an abnormality of a gene. There is, however, one sense of the expression molecular dis- ease that permits it to be applied to sickle-cell anemia and the other diseases associated with the abnormal hemoglobins. These diseases have been shown to result from the manufacture by the patient of abnormal molecules, in the place of normal molecules which are manufactured by normal individuals, and the abnormal molecules have been characterized. I think that it is not unlikely that many diseases will in the course of time be found to be molecular dis- eases in this sense. The discovery of the abnormal molecules re- sponsible for other molecular diseases may be far more difficult than the discovery of the abnormal hemoglobins. Hemoglobin is unique among the proteins of the human body because of its presence in very large amount, about 1 per cent by weight of the body, and because of the ease with which it can be obtained from the individual and characterized. Other substances in general are present in far smaller amounts. I have estimated that there are of the order of 100,000 different kinds of proteins in the human body. If the number of substances making up the human body is about 100,000, the average amount of each substance present (aside from water) is about 100 mg., and some important sub- stances may be present in far smaller amounts. The isolation and identification of abnormal forms of molecules of these substances might be extremely difficult. Because of the insight into the nature of disease that would be provided by the discovery of additional molecular diseases, however, it seems important to prosecute investigations along this line with much vigor. ABNORMALITY OF HEMOGLOBIN MOLECULES IN ANEMIA 237 ACKNOWLEDGMENTS It is a pleasure for me to acknowledge the financial support of the work on hemoglobin received from The Rockefeller Founda- tion during the entire score of years that the work has been carried on in our laboratories, and from the Public Health Service of the National Institutes of Health (Research Grant 257, on Investiga- tions of the Chemistry of Blood) during the period 1946 to 1953. I am glad also to have an opportunity to express my thanks to my co-workers, Alfred E. Mirsky, Charles D. Coryell, Fred Stitt, Donald S. Taylor, Richard W. Dodson, Charles D. Russell, S. J. Singer, Ibert C. Wells, Robert B. Corey, Ray D. Owen, W. A. Schroeder, Lois M. Kay, Alexander Rich, E. Havinga, D. H. Camp- bell, Morris Goodman, Robert C. C. St. George, William R. Bergren, Phillip Sturgeon, and especially Harvey A. Itano, who has been the discoverer or co-discoverer of each of the four known abnormal forms of adult human hemoglobin. REFERENCES I. Pauling, L. 1935. The oxygen equilibrium of hemoglobin and its structural interpretation. Proc. Natl. Ad. Sri. U.S. 21, 186. 2. Mirsky, A. E., and Pauling, L. 1936. On the structure of native, denatured, and coagulated proteins. Proc. Nat2. Ad. Sri. U.S. 22, 439. 3. Pauling, L., and Coryell, C. D. 19%. The magnetic properties and structure of the hemochromogens and related substances. PCOC. Nd. Ad. Sci. U.S. 22, 159. 4. Pauling, L., and Coryell, C. D. 1936. The magnetic properties and structure of hemoglobin, oxyhemoglobin, and carbonmonoxyhemoglobin. Proc. N&l. Ad. Sri. U.S. 22, 210. 5. Coryell, C. D., Stitt, F., and Pauling, L. 1937. The magnetic properties and structure of ferrihemoglobin (methemoglobin) and some of its compounds. J. Am. Chem. Sot. 59, 633. 6. Taylor, D. S., and Cotyell C. D. 1938. The magnetic susceptibility of the iron in ferrohemoglobin. J. Am. Chem. Sot. 60, 1177. 7. Stitt, F., and Coryell, C. D. 1939. The magnetic study of the equilibrium between ferrohemoglobin, cyanide ion, and cyanide ferrohemoglobin. J. Am. Chem. Sot. 61, 1263. 8. Taylor, D. S. 1939. The magnetic properties of myoglobm and ferrimyo- globin, and their bearing on the problem of the existence of magnetic interactions in hemogtobin. J. Am. Gem. Sot. 61, 2150. 9. Coryell, C. D., and Stitt, F. 1940. Magnetic studies of ferrihemoglobin 238 LINUS PAULING reactions. II. Equilibria and compounds with azide ion, ammonia, and ethanol. J. Am. Cbem. Sot. 62, 2942. 10. Coryell, C. D., Pauling, L., and Dodson, R. W. 1939. The magnetic prop erties of intermediates in the reactions of hemoglobin. J. Pbys. Cbem. 43, 825. 11. Russell, C. D., and Pauling, L. 1939. The magnetic properties of the compounds ethylisocyanide-ferrohemoglobin and imidazole-ferrihemoglobin. Proc. Natl. Acad. Sci. U.S. 25, 517. 12. Coryell, C. D. 1939. The existence of chemical interactions between the hemes in ferrihemoglobin (methemoglobin) and the role of interactions in the interpretation of ferro-ferrihemoglobin electropotential measurements. J. Pbys. Chem. 43, 841. 13. Coryell, C. D., and Pauling, L. 1940. A structural interpretation of the acidity of groups associated with the hemes of hemoglobin and hemo- globin derivatives. J. Biol. Chem. 132, 769. 14. Pauling, L. 1948. The interpretation of some chemical properties of hemo- globin in terms of its molecular structure. Stanford Med. Bull. 6, 215. 15. Pauling, L. 1949. The electronic structure of haemoglobin, in "Haemo- globin," pp. 57-65, Butterworths Scientific Publications. London. 16. Bush, V. 1945. Science, the Endless Frontier: A Report to the President, U.S. Government Printing Office, Washington, D.C. 4 Xr. Herrick, J. B. 1910. Peculiar elongated and sickle shaped red corpuscles in a case of severe anemia. Arch. Internal Med. 6, 517. 18. Hahn, E. V., and Gillespie, E. B. 1927. Sickle cell anemia: Report of a case greatly improved by splenectomy. Experimental study of sickle cell formation. Arch. internal Med. 39, 233. 19. Itano, H. A., and Pauling, L. 1949. A rapid diagnostic test for sickle cell anemia. Blood 4, 66. I* M. Pauling, L., Itano, H. A., Singer, S. J., and Welis, I. C. 1949. Sickle cell anemia, a molecular disease. Science 110, 543. b p Emmel, V. E. 1917. A study of the erythrocytes in a case of severe anemia with elongated and sickle-shaped red blood cells. Arch. Internal Med. 20, 586. 7 %% Taliaferro, W. H., and Huck, J. G. 1923. The inheritance of sickle cell anemia in man. Genetics 8, 594. g ti Neel, J. V. 1947. The clinical detection of the genetic carriers of inherited disease. Medicine 26, 115. 9 & Neel, J. V. 1949. The inheritance of sickle cell anemia. Scienre 110, 64. 25. Beet* E. A. 1949. The genetics of the sickle cell trait in a Bantu tribe. Ann. Eugen. 14, 279. lo MWeIIs, I. C., and Itano, H. A. 1951. Ratio of sickle-cell anemia hemoglobin to normal hemoglobin in sicklemics. J. Biol. Cbem. 188, 65. 27. Neel, J. V., Wells, I. C., and Itano, H. A. 1951. Familial differences in the proportion of abnormal hemoglobin present in the sickle cell trait. J. Clin. Invest. 30, 1120. ABNORMALITY OF HEMOGLOBIN MOLECULES IN ANEMIA 239 &! Itano, H. A. 1953. Qualitative and quantitative control of adult hemoglobin I\ synthesis-a multiple allele hypothesis. Am. J. Human Genet. 5, 34. 29. Neel, J. V. 1951. The inheritance of the sickling phenomenon, with par- F titular reference to sickle cell disease. Blood 6, 389. Neel, J. V. 1951. The population genetics of two inherited blood dys- crasias in man. Cold Spring Harbor Sympo.ria errant. Biol. 15, 141. 8 1'. Neel, J. V. 1933. Data pertaining to the population dynamics of sickle cell disease. Am. J. Human Genet. 5, 154. 32. Singer K. 1951. The pathogenesis of sickle cell anemia: A review. Am. J, Clirz. Patbol. 21, 858. 33. Silvestroni, B., and Bianco, I. 1952. Genetic aspects of sickle cell anemia `h Schroeder, W. A., Kay, and microdrepanocytic disease. Blood 7, 429. L. M., and Wells, I. C. 1950. Amino acid compo- 1% sition of hemoglobins of normal negroes and sickle-cell anemics. J, Biol. Cbem. 187, 221. 35. Havinga, E. 1953. Comparison of the phosphorus content, optical rotation, separation of hemes and globin, and terminal amino acid residues of normal adult human hemoglobin and sickle cell anemia hemoglobin. PYOC. Natl. Acad. Sci. U.S. 39, 59. 36. Havinga, E., and Itano, H. A. 1953. Electrophoretic studies of globins prepared from normal adult and sickle cell hemoglobins. Proc. Natl. Acad. \g;rSh Sri. U.S. 39, 65. erman, I. J. 1940. The sickling phenomenon, with special reference to the \# differentiation of sickle cell anemia from the sickle cell trait. Bull. Johns llkt Hopkins Hosp. 67, 309. Ponder, E. 1948. "Hemolysis and Related Phenomena," p. 145, Grune and \ 3 Stratton, New York. H Pauling, L., Pressman, D., and Campbell* D. H. 1943. The nature of the e forces between antigen and antibody and of the precipitation reaction. ` ) /: I&- Pbysiol. Revs. 23, 203. 40. St. George, R. C. C., and Pauling, L. 1951. The combining power of hemoglobin for alkyl isocyanides, and the nature of the heme-heme inter- actions in hemoglobin. Science 114, 629. ti Rebuck, J. W., Sturrock, R. M., and Monaghan, D. A. 1950. Sickling 17' fl P recesses in anemia and trait erythrocytes with the electron microscopy of their incipient crystallization. Federation Proc. 9, 340. r Perutz, M. F., and Mitchison, J. M. 1950. State of haemoglobin in sickle- `1 . cell anemia. Nature 166, 677. VI