Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
PART IV. GENETIC ASPECTS OF ABNORMAL HEMOGLOBINS GENETIC ASPECTS OF ABNORMAL HEMOGLOBINS NAMES V. N EEL So much has been written recently concerning the genetic aspects of the abnormal hemoglobins that you are undoubtedly familiar with the main out- lines of our present knowledge in this area. Nevertheless, in keeping with my assignment, I should like to summarize once again, but as briefly as possible, some of the generally accepted facts and interpretations concerning hemo- globin genetics, and then devote much of the remainder of my time to a dis- cussion of a family which my collaborators and I have recently described. This discussion will illustrate some of the approaches and problems which arise in attempts to determine the genetic relationships of the various abnormal hemoglobins to one another. Finally, me shall deal briefly with the implications, for genetics and for human biology in general, of some current developments as regards the hemoglobins. The Genetics of Hemoglobin S. The hypothesis that the sickling phe- r~omenon (and the formation of hemoglobin S) is determined by a single gene with different ejects ire homozygotes and heterozygotes is well known to all Of vou 46, 4`, 4S, ~ TO ;nr1;~;A',olc hPt^'r~7`rccr~c friar that crone ~xr;1-h thy . ~ ~ ~ ~ ~ J. ~ 1 A ~ ~A ~ 1 ~ ~~ Arm ~ ~ V~} 8~ REV ~ V ~ ~~ ~ ~~AA~~ ~ ~ ~ %4 ~ sickle cell trait, some 30 to 40 per cent of the hemoglobin is type S. the re- mainder being hemoglobin A. In individuals homozygous for the gene, with sickle cell anemia, no hemoglobin A is present, the hemoglobin being pre- dominantly hemoglobin S. but with ~ variable component up to 20 per cent of the total of hemoglobin F.58 64, 6s <;nrP hr~m^~7`rctnilc ;nA;`r;` must have received ~ sickle cell gene from both parents, then, with certain very rare exceptions to be discussed later' both parents of a child with this type of sickle cell anemia must exhibit the sickling phenomenon. Tow, one of the appeals of genetics is that one can submit hypotheses to rather rigorous testing. In the case of a marriage of a sickle cell trait x normal, we expect sickle cell wait and normal children in the ratio of exactly 1 :1. In the case of a marriage of two persons with the sickle cell trait, we expect normal, sickle cell trait, and sickle cell anemia children in the ratio of exactly 1 :2 :1. These ratios may be distorted by the manner in which these families come to our attention. The most extensively studied type of sibship to date is that ~J111~ ~~ AA1~A V ~ w ~ Much of the original world herein referred to has been collaborative, involving particularly Dr. Wolf Zuelzer, Mr. Abner Robinson, Mr. Frank Livingstone, Dr. lIarvey Itano, and Dr. Eugene Kaplan. We are greatly indebted to the U. S. Public Health Service and the Rockefeller Foundation for financial support. ~ Reviews: Cherno$, 1955 ;~2 Gatto, 1956 ;~i Itano, 1955,~5 1956 ,~6 Itano, Bergren and Sturgeon, 1956;2S Lehmann, 1957;34 Neel, 1956,50 1957;;): Singer 1955~66 White and Heaven, 1954;79 Zuelzer, Neel, and Robinson, 1956.S° 253
254 PART IV. GENETIC ASPECTS in which the propositus for the sibship has sickle cell anemia and segregation is studied among the non-anemic siblings of the propositus, both parents being either known or assumed to have the sickle cell trait. Here a ratio of 2 sickle cell trait: ~ normal is expected. Table I summarizes the results in four series. Agreement with hypothesis is satisfactory. It will be noted that such de- parture from expectation as exists is in the direction of a deficiency of sicklers. TABLE I SEGREGATION FOR SICKLE CELL TRAIT VS. NON-SICKLE CELL TRAIT AMONG THE NON- ANEMIC SIBLINGS OF CHILDREN WITH SICKLE CELL ANEMIA, BOTH PARENTS KNOWN OR ASSUMED TO HAVE THE SICKLE CELL TRAIT. Author Number with s.c.t. Number examined Observed ~ Expected Number normal Observed ~ Expected Neel, 19514S Lambotte-Legrands, 195 13i Banks, Scott and Simmons, 19524 Vandepitte, 1955~0 98 62 65.4 36 70 45 46.7 25 23.3 33 15 22.0 18 208 141 138.7 67 32.6 110 69.3 409 i 263 ~ 272.8 ~ 146 ~ 13 6.2 X- 1.023 D.F. 1 0.50 > P > 0.30 In my own experience with a number of other types of segregating sib- ships,48 there was a similar deficiency, with x~ for the entire group of cases studied by myself significant at the 1 per cent level. However, it must be re- membered that both discrepancies between legal and biological paternity, and the occurrence in the series of some of the variants of sickle cell anemia (as sickle cell-thalassemia disease; see below) would be expected to procure departures in this direction, for which reason the over-all agreement with hypothesis should be regarded as satisfactory. Bunny and Sandier and Novitski63 have raised the possibility that there may be an abnormal segregation ratio with respect to the genes responsible for the sickling phenomenon and for thalassemia, in consequence of which there is an excess of individuals with the abnormality over expectation in segregating sibships. They have introduced this as a possible contributory factor in the relatively high frequency of these genes in certain parts of the world. The departure from expectation in the sickle cell data, such as it is, is actually opposite in direction to expectation on this hypothesis, although, as Sandler and Novitski point out, large numbers are necessary for a critical test of the hypothesis. In American Negroes with the sickle cell trait, the amount of abnormal
GENETICS OF ABNORMAL HEMOGLOBINS NEEL 255 hemoglobin revealed by moving boundary electrophoresis varies from approx- imately 22 to 45 per cent of the total, with modal values at 3~36 per cent and 40-42 per cent. There are significant intrafamily correlations ire this respect; it has been suggested these can be explained through the action of segregating genetic modifiers.7S 54 An alternative hypothesis, that the correla- tions may be due to the occurrence of different "iso-alleles" of the normal allele of the hemoglobin S gene, seems not to be borne out by the data. It is worth emphasizing that, if our limited experience is typical, these modifiers are common in the American Negro population. Allisont and Raper,6t using paper electrophoresis have also encountered two modal values in the amount of abnormal hemoglobin in African Negroes with the sickle cell trait. The modes in the case of Raper's data do not correspond with the values observed ir: the American Negroes, but this may be due to the different techniques employed. The role of the genetic mechanism responsible for this bimodality in the maintenance of the relatively high frequency of the sickle cell gene, a subject to be discussed later, is unknown. The Genetics of Hemoglobin C. Like hemoglobin S. the presence of hemoglobin C seems to be determined by a single gene which, when heterozy- gous, results in some 30 to 40 per cent of the hemoglobin being abnormal in type, and when homozygous, in all the hemoglobin being abnormal save for a variable but usually small component of hemoglobin F.'`' '9, ·~6' 7~, ·36 Because of the greater difficulties involved in studying the inheritance of hemoglobin C, in that one must rely on electrophoresis rather than such a simple procedure as a sickling test, there is not yet available a large body of data for a critical evaluation of the precise ratios in segregating families. Such data as are available, as summarized in Zuelzer, Neel, and Robinson, ~ 9 reveal no de- , . . . . . partures from the expected ratios. The Genetic Relationship of Hemoglobins 5 and C. The first papers on hemoglobin C raised the obvious question of the genetic relationship of this hemoglobin to hemoglobin S and pointed out the information necessary to a decision. Critical data were soon supplied by Ranney, Larson, and McCor- mack<;° and by Smith and Conley (unpublished), on the basis of which it may be concluded that the two genes concerned are either alleles or linked. In brief, this evidence consists of the observation that the children of a mar- riage involving a normal person and a person simultaneously heterozygous for both the abnormal genes all either possess the sickle cell trait or the hemo- globin C trait. The basis for this statement is indicated in figure 1. However, the number of children studied is small, and further data on this point are highly desirable. The Inheritance of Hemoglobins D, E, G. I, and I. The relatively limited material available concerning the inheritance of hemoglobins D, E, G. I and J (cf. reviews quoted in footnote to introduction) suggest that each of these abnormal hemoglobins owes its presence to a single gene which in the
256 PART INT. GENETIC ASPECTS ALLELiC GENES INDEPENDENT GENES R A r lo LINKED GENES FIG. 1. Diagrammatic representation of expectation on the basis of i) allelism of two genes, A and B; 2) independence (location on different chromosomes) of the two genes; or 3 ) linkage, with less than 50 per cent crossing-over, of the two genes. ( From Human Heredity, by J. V. Neel and W. J. Schull, University of Chicago Press, 1954). heterozygous state results in some 30 to 40 per cent of the hemoglobin being of an abnormal type. Hemoglobin ~ is an exception to this generalization, in that in the family reported by Thorup, Itano, Wheby, and Leavell,73 as well as in the material reported by us under the designation of Liberian 1,6-' the abnormal component in heterozygotes amounted to 60 per cent of the total. Individuals homozygous for the gene producing hemoglobin E are characterized by the fact that all of their hemoglobin is of the abnormal type save for a small and variable type F component.~3 39' 40 An individual all of whose hemoglobin was type D has been presented as the probable result of homozygosity for the hemoglobin D gene.8 While this may be the case, there there no family studies, and experience with the manner in which the thalas- semia and possibly other genes may interact with the genes responsible for the abnormal hemoglobins has emphasized the absolute necessity of family studies in reaching conclusions concerning genotype. The father of the proband for the family in which hemoglobin G was first discovered appeared on paper electrophoresis to have only hemoglobin G. Ten of his 11 children showed the hemoglobin G trait; the eleventh is a parentage exclusion on the basis of exhibiting the sickling phenomenon. These findings are highly suggestive of homozygosity as regards the gene responsible for this abnormal hemoglobin.~9 It is noteworthy that this presumed homozygote showed no anemia. On the basis of this finding, as well as our oven experience with hemoglobin G. to be discussed later, we feel that G is better designated a normal variant than an abnormal hemoglobin. The genetic basis of hemoglobin H appears to be somewhat different from that of the other abnormal hemoglobins discussed thus far. This hemoglobin has been observed only in individuals who are also heterozygous for the thalas- semia gene, suggesting that this latter gene somehow "potentiates" the appear- ance of hemoglobin H.44 It will be difficult to conduct detailed studies in man of the genetics of a trait which is only expressed on a specific genetic background.
GENETICS OF ABNORMAL HEMOGLOBINS NEEL 257 The Genetics of Thalassemia. Before turning to a consideration of the genetic relationships of the genes responsible for hemoglobins D, E, G. I, and ~ so one another and to the genes responsible for hemoglobins S and C, we must, for reasons which will become apparent, consider at this point the gene- tics of thalassemia. As in the case of the abnormal hemoglobins, thalassemia has been extensively reviewed recently (cf. refs. p. 2j3), and we shall con- fine our attention here to the main outlines. The data on the familial dis- tribution of thalassemia are in general consistent with the hypothesis that the mild form of this disease, known as thalassemia minor, is due to heterozygosity for a gene which in the homozygous condition produces the much more serious and fatal disorder known as thalassemia major.~°, 'I i4 75 53 Recently, as no- ted earlier, several workers have raised the possibility that there may be art abnormality in genetic segregation in connection with the thalassemia gene, comparable to that know-e for the T locus in the mouse, and resulting in an excess of thalassemia major and minor over expectation in segregating matings. The data as summarized by several authors53 ~ do not appear to lend support to . . t Ifs suggestion. In diagnosing the presence of an abnormal hemoglobin, the basis of the diagnosis is a rather specific alteration in the behavior of a well known and extensively studied protein. The diagnosis of thalassemia has until recently lacked this specificity, depending rather on morphological criteria involving .. . . . . ~ , . . . . . . . .. erythrocyte size arid shape. ~ or several years it has been recognized that ~nd~- ~-iduals with thalassemia minor might, on moving boundary electrophoresis, exhibit a minor hemoglobin component with mobility slightly greater than hemoglobin A.7~ ~' 60 More recently, Kunkel and Wallenius,30 using "starch block" electrophoresis, have observed the regular occurrence of a minor hemo- globin component, now termed A`', in normal blood, which component is increased in thalassemia minor; this is apparently identical with the just-men- tioned fast-moving shoulder seen in thalassemia minor with moving boundary eiectrophoresis. Here may be a "handle" which will make thalassemia minor somewhat easier to work with. Recently several investigators have emphasized on clinical and biochemical grounds the possibility that there may be not one but several thalassemias.79 28 This opinion is based in part on the wide range in the expression of the heter- ozygote, but to an even greater extent in the variability in the results of com- bining a thalassemia gene with the gene responsible for either hemoglobin S of hemoglobin C. ~ decision concerning the obvious question of the relationship of the thalassemia gene or genes to the locus associated with the production of hemo- globins S and C depends, of course, on the study of the type of children emerg- ing from a marriage of a normal individual and an individual heterozygous for the thalassemia gene and also either the gene resulting in hemoglobin S or C. Here again, we encounter a paucity of data This is in part because many
258 PART IV. GENETIC ASPECTS individuals simultaneously heterozygous for the genes responsible for thalas- semia and hemoglobin S suffer from a severe, chronic hemolytic anemia which may be clinically indistinguishable from sickle cell anemia, and thus results in early death and impaired reproduction. I have been able to find in the literature a description of only seven marriages involving a normal person and one with thalassemia-hemoglobin S disease. The results are shown in table II. Most of the evidence suggesting non-allelism is supplied by the TABLE II A SUMMARY OF THE TYPES OF CHILDREN RESULTING FROM 1\IARRIAGES BETWEEN A l\.CRMAL INDIVIDUAL AND ONE WITH BOTH THE THALASSEMIA GENE AND THE GENE RESPONSIBLE F R HEMCGL~BIN S. l l Children t Authors Ethnic background - Tha, assemia- sickle cell disea se Normal Thal- assem~a Sickle cell Trait Silvestroni and Bianco, 195265 Powell, Rodarte, and Neel, 1950;'$3 Neel, Itano, and Lawrence, 19535-' Neel, 1951,4S kin- dred 60 Ceppellinili Italian (Italy) 1 0 0 0 Italian (USA) 0 0 2 0 Greek (USA) 0 0 0 1 American Negro 1 ~ 1? 2 1 + 1 ? 1 7 American Negro 0 0 4 3 7 Greek (USA) 0 0 1 2 Italian (Italy) 0 0 1 2 2 + 1? 2 9 ~ 1? 9 2 1 3 3 filmily which I described in 1951. In this family the stigmata of thalassemia minor were very minor indeed, leading to some difficulties in classification. Furthermore, the interaction of the thalassemia and sickling genes was very much less than usual. However, even with the exclusion of the doubtful in- dividuals, the occurrence of normal children and some with thalassemia- hemoglobin S disease suggests non-allelism of the two genes. On the other hand, omitting the doubtful individuals, the departure from expectation on the basis of complete independence of the two genes is significant (expectation 11 thalassemia-hemoglobin S or normal: 11 thalassemia minor or sickle cell trait; probability of the observed or a greater deviation as computed from binomial distribution 0.004~. This suggests either linkage or heterc- geneity of the material (i.e., two, or more, thalassemia loci).
GENETICS OF ABNORMAL HEMOGLOBINS NEEL 259 In addition to the data of table II, Silvestroni and Bianco65 in 1952 des- cribed a marriage between an individual with sickle cell-thalassemia disease and an individual with the sickle cell trait, from which among 7 children who could be studied or on whom there were medical records, there was one normal child, again suggestive of non-allelism. Finally, Singer et al.69 have reported the birth of a normal child to a woman with thalassemia-hemoglobin C dis- ease whose husband was deceased and hence of unknown genotype. This sug- gests non-allelism of the gene responsible for these two traits, to be expected if hemoglobins S and C for their part are due to allelic genes. Further data on this point are highly desirable. At present the evidence for the heterogeneity of the thalassemia trait is growing. A comparable situation may exist with respect to the elliptocytosis trait, a morphological abnormality of the erythrocyte which in several dozens of pedigrees has behaved as if due to a dominant gene. But now it has been shown that in some pedigrees this gene appears to be rather closely linked to the Rh LOCUS' whereas in other pedigrees there is no evidence for linkage frets. in Morton43~. There is some evidence for linkage between the thalassemia gene and the genes responsible for the Lewis and Secretor characteristics;7 an ex- tension of these data, with particular attention to the question of hetero- geneity of the material, would seem indicated. It will also be of interest to determine whether the A2 component is accentuated in some pedigrees of thalassemia and not in others, in relation to this heterogeneity problem. The Relationship of the Genes Responsible for Hemoglobins D, E, G. 0,1, and I, to One Another and to the Genes Responsible for Hemoglobins S and C. By now we have established the high probability that the genes responsible for hemoglobins S and C are either alleles or linked, and the further probability that at least one of the several possible thalassemia genes is to be found at a different locus, with the possibility that another of the postulated thalassemia genes may either occur at or be linked with the S-C locus. We turn now to a consideration of the relationship of the genes responsible for the remaining abnormal hemoglobins to these two loci. A number of investigators have postulated that several more or even all of the genes responsible for the abnormal hemoglobins may fall into the S-C series of multiple alleles.24, 9' 45' 3~, 35 This opinion is based in part on the apparent absence of normal hemo- globin in persons heterozygous for two of these genes, as S-D. In the final analysis, however, these genetic relationships can only be established with certainty from the study of fertile marriages in which one of the marital partners is simultaneously heterozygous for two or more of the genes con- cerned. Indeed, as will become apparent in the family we are now to consider, we know too little about biochemical genetics to draw inferences from the type of factor interaction as to the formal relationships of the genes involved. Recently I have been so fortunate as to be associated in the study of a family in which the genes responsible for thalassemia and hemoglobins S and
260 PART IV. GENETIC ASPECTS G were present.64 Not only was it a stroke of luck to encounter such a family, but the types of individuals observed in the family were such as to yield almost a maximum of possible genetic information. The details of the family are complicated and will require constant reference to table III and figure 2. (I should mention that since our paper on this family went to press it has been possible for Mr. Robinson, one of the co-authors, to compare the hemo- globin G in this family with that identified as G by Edington, Lehmann, and Schneider in 1955,19 with agreement in all essential respects.) The propositus for this kindred was a 28-year-old male of Italian extraction who appeared on clinical grounds to have sickle cell anemia. Fifteen members of the family were available for study. Although it is customary to begin the description of a family with the propositus, it will in this case be somewhat less confusing if we begin with III-5. She was a 14-year-old girl with the hematological findings of thalassemia minor; her hemoglobin in both its solu- bility and electrophoretic behavior corresponded to type A. Her father, II-6, had no hematological abnormality. Her mother, II-7, showed in unequivocal fashion the findings of thalassemia minor. On both Tiselius and paper electro- phoresis, essentially all the hemoglobin appeared to be type G. although Tise- lius studies revealed a small and equivocal "shoulder" with a mobility cor- responding to 2.7 ~ 10-;' cm.~/sec./volt. This is the approximate position of the "thalassemia shoulder" of hemoglobin As which was referred to earlier. There was no significant increase in the amount of fetal hemoglobin. Although the electrophoretic findings would by analogy with what is known concerning the results of homozygosity for the genes responsible for hemoglobins S. C, and E suggest that II-7 was homonymous for the G gene. the fact that her 1 1 ~ ~ ~ J ~ ~ aaugnrer was nor sound to rave any hemoglobin G would seem to rule out that explanation. A second factor ruling out this explanation is the fact that I-3, II-7's father, is completely normal hematologically, and hence could scarcely have made the contribution of a hemoglobin G gene necessary if his daughter were homozygous. The most reasonable interpretation of these findings would seem to be that they are due to simultaneous heterozygosity for the thalassemia and G genes. We turn now to II-5, sister to II-7, whose findings differ from those of her sister only in the presence of a distinct, minor hemoglobin component which migrates with a speed of 2.7 x 10-' cm.~/sec./volt, the rate shown by hemoglobins D and S. This amounts to about 20 per cent of the total, and appears to be an accentuation of the "shoulder" observed in the electrophoretic pattern of II-7. We will assume on the basis of the cytological findings that this woman is heterozygous for the thalassemia gene, and on the basis of the O electrophoret~c findings that she Is either homozygous or heterozygous for the G gene. Again, however, as in the case of II-7, the absence of hemoglobin G in her father renders homozygosity unlikely, and suggests the most probable
GENETICS OF AB N:ORMAL HEMOGLOBINSNEEL 261 FIG. 2.Picture pedigree of a family in which hemoglobins S and G and thalas- semia were encountered. The crescent and dot-withir~-circle symbols represent the sickling and thalassemia traits, respectively. The letters refer to the hemoglobin phenotypes. (Reproduced from Blood 1~?: 239, 1957, by permission of Grune and Stratton, Inc.). TABLE III~ A SUMMARY OF THE HEMATOLOGICAL AND BIOCHEMICAL FINDINGS IN A FAMILY IN WHICH THE GENES RESPONSIBLE FOR HEMOGLOBINS S .AND G AND THALASSEMIA WERE SEGREGATING. o .~ :- ~ ~ or .~ .~ q~ c; ¢-Q I-3 II-7 II-5 II-4 II-6 III-1 III-2 III-3 III-4 III-5 IV-1 IV-2 5 => .. X _ ~ ~ F.C. hI B.H. F F.H. F~ F`.S. M M.H. M V.S. M C.S. F G.S. M M.S. F Fr.H. F S.S. M I.S. F~ . ~ .~ .= ~ . '= ~ ~, _ ~- ~ ulubility 76 _ _ _ A A oo A 39 + _ + G G oo A 45 ~ _ ~ G G oo A 52 _ + _ AS AS AS 45 _ _ _ A A oo 28 + + + S GS AS 24 _ _ _ A A oo A 26 _ + _ GS GS AS ~- 25 _ _ _ A A co A 14 + _ _ A A oo A 1 + _ _ A A _ 5 _ AG AG ~ A Heweglobin Type . . ._ . ~ 0 0 ~ ~ ~ tl - A A G G G G AS AS A A S GS A A GS GS A A A A A A AG AG ~0 o~ c, ._ ~ ~ .0 ,~ 0 . <= <1.0 <1.0 <1.0 1.1 1.2 3.9 <1.0 1.2 <1.0 1.8 4.1 1.4 * Reproduced from Blood 1~: 240, 1957, by permission o£ Grune and Stratton, Ine. b4 5= ~Q ~_ 112 114 85 241 96 121 130
262 PART IV. GENETIC ASPECTS interpretation to be that she, like her sister, is simultaneously heterozygous for the G and thalassemia genes. This woman (II-5) married a man who had the sickle cell trait, as did all three of his siblings available for testing. This marriage, then, involves three different genes responsible for erytnrocytic variation, namely, the genes responsible for hemoglobins G arid S. and the thalassemia gene. The marriage resulted in two sons. One, III-3, exhibits the sickling phenomenon but has no anemia, thus appearing to have only the sickle cell trait. However, electro- phoretically his hemoglobin is a mixture of S and G with G comprising about ~ r~ . r . 1 . . 1 T T · · . ~ 1 3 ou per cent or tne total. Bus marriage to a completely normal woman, III-4, has thus far resulted in one child whose hemoglobin on electrophoresis shows an AG pattern. The proper genetic interpretation of these findings would appear to be that III-3 has inherited the sickle cell gene from his father and the hemoglobin G gene from his mother, and has, in turn, transmitted the latter gene to his daughter. Note the ~nn~rent l~h~7sin1~ic;31 e~lliv~lence of hemoglobin G to hemoglobin A. Errs i- ~--~ -~- III-1, the individual who originally brought this family to our attention, has, in contrast to his brother, a severe, chronic hemolytic anemia in all respects answering to the description of sickle cell anemia, even though on Tiselius electrophoresis his hemoglobin is quite similar to his brother's, ap- parently consisting of approximately equal proportions of G and S. On paper electrophoresis the pattern is essentially that of S alone, thus illustrating the discrepancies which occasionally occur between paper and moving-boundary electrophoresis. The only son of III-1, aged 1 year, exhibits morphologic ab- normalities of his erythrocytes consistent with thalassemia minor, although it must be recognized that this is an age when mild iron deficiency anemias are not uncommon. Since III-2, the mother of this child, is hematologically normal, the thalassemia gene which appears to be present in the child must have been inherited from our propositus, his father. The presence of a thalas- semia gene in the propositus would also explain the marked difference between his findings and those of his brother, who has the S and C genes, but ap- parently not the thalassemia gene. We are thus led to the conclusion that I I I- 1 possesses three different genes resulting in hematologic abnormality, namely, those responsible for thalassemia and hemoglobins G and S. If this interpretation of the pedigree is correct, then two important genetic conclusions are automatic: 1. The genes responsible for hemoglobins G and S cannot be alleles otherwise IV-1, the son of the propositus, would have received one or the other of the genes from his father. 2. The genes responsible for hemoglobins G and thalassemia cannot be alleles otherwise the propositus could not have received both of these genes from his mother, as analysis of the pedigree seems to demonstrate. We have already considered briefly the evidence that the genes responsible
GENETICS OF ABNORMAL HEMOGLOBINS NEEL 263 for hemoglobins S and C, on the one hand, are not allelic with at least one of tile several possible thalassemia genes. It therefore becomes necessary to postulate at least three different genetic loci involved in hemoglobin produc- tion, with the relationship of the genes responsible for hemoglobins D, E, H. I; and ~ to these loci yet to be determined. There are two more points concerning this family which deserve emphasis. The finding that an individual simultaneously heterozygous for the genes responsible for hemoglobins S and G annarentlv produces no hemoglobin A, . . .. ~ ~ J AL ~ even though these genes are non-allelic, demonstrates the fallibility of this criterion of allelism. Secondly, we must consider the theoretical implications of the finding, in II-5 and II-7, of a minor hemoglobin component with the mobility of hemoglobin D or S. This component is unaccompanied by sickling. Although it could be the "thalassemia shoulder," it was not observed in III-5 c, IV-1, who also have the thalassemia gene. Conceivably, it might be the result of an inheritance from I-4, who was not studied. There is, on the other hand, the possibility that this component is not the result of a specific gene, but is rather a genetic interaction product. It has been demonstrated that when pigeons of two species are crossed, the hybrids possess certain erythrocyte antigens not present in either parent (cf. Irwin, 1951,23 for summary). Can this component be regarded as such a "hybrid substance?" And, in this same context, would hemoglobin H also qualify as a hybrid substance? As I pointed out earlier, this family has an almost incredibly high "infor- mation content." The segregants in each of the four marriages in the pedigree which could be studied in detail in each instance shed critical light on the hypothesis under test. Although I shall undoubtedly continue to pursue pedi- grees as long as I live, I cannot escape the feeling that no family I shall ever study in the future can be quite as exciting as this one. Since coming to these meetings, I have learned of a family being studied by Drs. Torbert and Smith,74 in which there occurs an individual possessing hemoglobin S in combination with a "fast component," married to an indi- vidual possessing neither of these components. The issue of the marriage in- cludes children of such a type as to suggest non-allelism of the genes respon- sible for the two hemoglobins. The relationship of this "fast component" gene to the thalassemia and hemoglobin G loci is unknown, but it is perhaps not too soon to start to think of the possibility of still another genetic locus involved in hemoglobin production. Terminology. There comes a time in the development of many genetic problems when a symbolic shorthand becomes a great aid to clarity of thought and brevity of expression. That time has arrived with respect to the abnormal hemoglobins. Thus far in this presentation, we have avoided the use of gene symbols only by rather cumbersome circumlocutions. Two years ago Allison2 and I'd independently advanced terminological systems which were quite similar. Last year we were able to meet and resolve such differences as did
264 PART IV. GENETIC ASPECTS exist. Subsequently, at the sixth meeting of the International Society of Hema- tology in 1956, I presented a compromise nomenclature which at a later dis- cussion seemed to find rather general acceptance.52 At least, it did not elicit any violent controversy, which in matters terminological is about as close as one can come to a seal of approval. The chief merits of the system are that it follows accepted usage in experimental genetics, and is elastic and descriptive. In brief, it is suggested that the first locus to be recognized as associated with hemoglobin production be given the locus symbol Hbi. We now recog- nize three different genes which may occur at that locus, the normal gene, HERA, and the abnormal genes associated with the production of hemoglobins S and C, namely, Hats and Half. The second locus to be identified will be designated as He,, with the two alleles thus far known being identified as HERA and Hb9G. Finally, it is suggested that the thalassemia locus be desig- nated by the symbol Th, with the thalassemia gene being known as Thy and its normal allele as ThN. When and if it is definitely shown that there are several thalassemia loci, then they may be designated as Thi, The, etc. With respect to these three loci, the genotype of a normal person is HbiA/Hbi~; Hb~/Hb.~; Th~/Th~-. The genotypes encountered in the pedigree we have just considered in some detail are interpreted to be as shown in Fig. 3. The woes and shortcomings of terminological systems are proverbial. In this instance, one is, for example, immediately confronted by the problem of the genetic basis of the minor hemoglobin components we have been dis- cussing yesterday arid today. Does each of these imply the existence of a ? 31 Hb 1'Hb, / \ Hb 2 / Hb 2 ( ) ThN/ThN ~ / I T m: ;~> 31 ' at ' 1 me| fib, /Hb, | ( Hb2/Hb2 ~ ( ~b2 /~b2 ~ Hb2 /Hb2 Hb ^/Rb ( ~b2C/Hb2 ~ Hb2 /Hb2 ( Hb2G/Hb2 ma/ Aim T h / T h ~~ T h / T h 111 IV | Hb, /Hb' | 35;: b1 1 ~: Hb2G/Hb2A (:i Hb 2G/Hb2A ~ ( Hb2 /~b2 T b / T h 4/ T h / r h \~/ my 1 ^z A Hb 2 / ~ b 2 Thr/ rh:Q me, ( Ht2/Hb2 ) FIG. 3. The genetic interpretation of the family shown in figure 2 and described in detail in the text.
GENETICS OF ABNORMAL HEMOGLOBINSNEEL 265 separate locus, and if so, what is the relation of these to the loci we have just been discussing? A related problem concerns the genetic control of the forma- tion of hemoglobin F. Thus far, no inherited abnormalities of hemoglobin F leave been identified, so that the possibility of a direct attack on the problem does not exist. We are left with some indirect considerations, which, as we have already seen, are to be pursued with caution in this field. Be that as it may, two a priori possibilities seem outstanding. On the one hand, the hy- pothesis may be entertained that the genetic basis of hemoglobin F produc- tion is the same as that for hemoglobin A, with the oxygen tension in the hematopoietic tissues controlling a switch-over mechanism. The persistence of small amounts of hemoglobin ~ through life could imply that this switch- over is never quite complete. The apparently normal adults with substantial proportions of hemoglobin F who have been encountered f rom time to timers 5' would then be individuals in whom the switch-over mechanism divas for some reason imperfect. On the other hand, we must consider the possi- biiity that there is a separate locus involved in the production of hemoglobin if, which locus is active (with some degree of suppression of activity at the loci responsible for hemoglobin A) under conditions of relative tissue anoxia. In view of the apparent differences in amino acid composition between fetal and adult hemoglobin, and in view of the "template" concept of protein syn- thesis, the biochemical evidence would seem to favor the latter hypothesis, a point of view previously expressed by Itano,24 Singer, Singer, and Goldberg,70 and Huisman, Jonxis, and van der Schaaf.2~ This locus might be designated as Hb3, with the only gene thus far known for this locus being Hb3F. Biochemical Genetics and the Abnormal Hemoglobins. Thus far we have been largely concerned with the formal genetics of the abnormal hemo- globins. Some time must be devoted to a consideration of the opportunities provided by this system for precisely defining the nature of gene action, and for the further opportunities for quantitative genetics. The elegant localization of the difference between sickle cell and normal hemoglobin to a single amino acid which Dr. Ingram has just presented constitutes a genetic milestone. As the precise biochemical basis for the other abnormal hemoglobins is worked out, there will emerge a detailed picture of the effects of gene substitutions which should open a new chapter in physiological genetics. Elsewhere, in discussing the manifestations of hemoglobins C and E, ~ have raised the possibility that "the basic defect is the same in these two conditions, the slight dissimilarities being due to small normal differences between the globin portion of the hemoglobin molecule in Negroes and Orientals, differences which only come to light under these circumstances."~° The opportunity to test this hypothesis with respect to these and other abnormal hemoglobins has come sooner than could be anticipated. This system provides an opportunity to test another very basic question. This has to do with the matter of "invisible" mutations, i.e., mutations which
266 PART IV. GENETIC ASPECTS at the time of their occurrence have no anT~arent effects heca~ise the. alte.rations 1 1 they induce are not reflected in the phenotype as it is usually studied. It is conceivable that some amino acid substitutions in tI,e chains which comprise a hemoglobin molecule are not critical, in the sense of producing clinical effects of altering hemoglobin behavior by any of our current tests. 1 heir importance lies in the fact that they may be cumulative or else may find expression when environmental factors alter. The task of analyzing the hemoglobin of as many as a thousand individuals by Dr. Ingram's techniques will be extremely la- borious, but yet would supply significant insight `-~ri~h respect to human vari- ability in regards to just one of the many proteins of the body. Furthermore, in the field of quantitative genetics, an obvious question is the nature and amount of the abnormal hemoglobin found in E'b~s/FIb~-i, Hk 0/Hk ~ indi- viduals. If hemoglobins G and S diffe, in amino acid substitutions at dif- fcrent points in the chain, will some hemoglobin molecules show both types of substitution, and some only one? These are very basic questions in biochemical genetics. The Population Genetics of the Abnormal HemGgIobins. In bringing this talk to a conclusion, I feel obligated to say a few words about the popu- lation genetics of the genes we have been discussing. Unlike so ready of the inherited biochemical abnormalities which have been a source of instruction to biochemist and geneticist alike, the genes under discussion today are no rare oddities, but in various parts of the world attain very considerable frequencies. Thus, in parts of Africa 40 per cent of the inhabitants are heterozygous for the sickle cell gene, and in other parts, 20 per cent for the hemoglobin C gene. I;requencies of heterozygotes for the hemoglobin E gene of 12 per cent have been reported from Thailand. In parts of Cyprus and Italy, thalassemia minor frequencies of 15 per cent or even higher are known. Gene frequencies of this magnitude do not arise by chance, but indicate that the gene in ques- tion makes some positive contribution to the fitness of the population. The search for this fitness factor is currently attracting considerable attention. The outstanding hypothesis to date, applicable at the moment only to the sickle cell gene, is that individuals possessing this gene enjoy a relative im- munity to P. falciparum malaria which, up to a point, can offset the increased death rate of persons homozygous for the gene (summary in Allison).3 If this is so, then in an analysis of the detailed mechanism whereby this immunity is conferred, we have an opportunity for the precise dissection of a problem in population genetics which is not to be exceeded in any other species. Three possible (non-alternative) mechanisms have been suggested: 1) Alli- soni and Mackey and Vivarelli4i have postulated that the falciparum parasite may not be able to metabolize hemoglobin S as readily as hemoglobin A. 2) Mackey and Vivarelli4i have suggested that the presence of the malaria para- site, with its oxygen demands, may induce sickling in the cells of individuals with the sickle cell trait, with destruction of both the cells and the parasites.
GENETICS OF ABNORMAL HEMOGLOBINS NEEL 267 Miller, Neel, and Livingstone42 have pointed out the tendency of erythrocytes containing the falcipar2~m parasite to adhere to blood vessel walls. Such cells, being handicapped in the renewal of their oxygen supply, would in parasitized persons with the sickle cell trait have an added reason for sickling and prema- ture destruction. 3) Livingstone'37 3s has suggested that, because of 2), the placentae of women with the sickle cell trait who have falciparz~m malaria will not be parasitized to the same extent as the placentae of normal women. This might be expected to result in a greater birth weight for the children of women with the sickle cell trait; there is some evidence that this is indeed the case. Under African conditions, one might (within limits, of course) expect heavier babies to have a better chance of survival than the lighter ones. A1- though no decision between the relative contributions of these three possibilities' can be reached at present, it is apparent that it should be feasible to collect the data necessary for a decision. Whatever the mechanism responsible for maintaining the relatively high frequencies of these genes, it is a fact that these genes must have a point of origin through mutation. Since it is a well established tenet of genetics that mutation tends to be recurrent, in any consideration of the population genetics of the abnormal hemoglobins there immediately arises the question of the frequency with which mutation resulting in the hemoglobin genes is occurring. There are a good 'many complications to establishing unequivocally the fact that a mutation has occurred.77 Be that as it may, there are now in the litera- ture a number of reports suggestive of mutation to the sickle cell gene.49' 77' 32 These involve for the most part the failure of a mother of a child with sickle cell anemia to exhibit the sickle cell trait or any other identifiable hemoglobin abnormality. It is very doubtful whether all these possible cases of mutation actually are examples of mutation. Some of them may be attributed to a well disguised thalassemia gene. However, taking the data at their face value, the observed frequency of possible mutation is still clearly less than the rate required to account for the gene frequencies observed in certain parts of Africa, but yet sufficient to suggest that a mutation rate of ~ x 10-5/locus/ generation is possible. On the other hand, there is increasing evidence to suggest that the distri- bution of the sickle cell gene in Africa is to be explained primarily, not in terms of local origins through mutation with subsequent increase in fre- quency through selection, but in terms of the introduction of the gene through migration with subsequent increase in frequency when conditions are favor- able. The case for spread through migration is especially clear in West Africa blest of Nigeria (see especially Livingstone3S). Last year I speculated on the possibility that "the mutations responsible for these hemoglobin types arise very infrequently but under special circumstances then become widely dis- seminated. Either the mutation is very rare or the 'special circumstances' are, geologically speaking, of recent origin otherwise, one would expect the
268 PART IV. GENETIC ASPECTS genes to be much more widely disseminated." Since that time, Livingstone38 has argued that the conditions which favor malaria endemicity are much more those that accompany the "slash and burn" type of agriculture than those which charactize a hunting economy. He postulates that the relatively recent spread of "slash and burn" agriculture through large parts of Africa has created an ecological situation favorable to an increased importance of the malaria parasite, and hence has lead to an extension of the conditions resulting in selection for the sickle cell gene. There is, then, something of a conflict in evidence, between findings indicating a sufficient mutation rate to account for numerous local origins of the sickle cell gene, and other findings suggesting that much of the present-day distribution of the sickle cell gene can be ac- cc~unted for through relatively recent gene flow and increase with changing selective pressures, with the inference that mutation is extremely rare. A very careful scrutiny of all apparent cases of mutation is called for. In concluding, I should like only to express my appreciation of the occasion which brings biochemist, clinician, and geneticist together. Under slightly different circumstances, our gathering might have included the epidemiologist and anthropologist. I know of no field of inquiry today which better illustrates the fundamental unity of science. REFEREN CES 1. Allison, A. C.: Protection afforded by sickle-cell trait against subtertian malarial infection, Brit. Med. J., ~ (Feb. 6) : 290-294, 1954. 2. Allison, A. C.: Notation for hemoglobin types and genes controlling their syn- thesis, Science 122: 640-641, 1955. 3. Allison, A. C.: Aspects of polymorphism in man, Cold Spring Harbor Symp. on 4. Quant. Biol. 20: 239-252, 1955. Banks, L. O., Scott, R. B., and Simmons, J.: Studies in sickle cell anemia, Am. Dis. Child. 84: 601-608, 1952. 5. Beet, E. A.: The genetics of the sickle cell trait in a Bantu tribe, Ann. Eug. 14: 279-284, 1949. 6. Bianco, I., Montalenti, G., Silvestroni, E., and Siniscalco, M.: Further data on genetics of microcythemia or thalassemia minor and Cooley's disease or thal- assemia major, Ann. Fug. 16: 299 - 315, 1952. 7. Bianco, I., Ceppellini, R., Silvestroni, E., and Siniscalco, M.: Data for the study of linkage in man. Microcythemia and the Lewis-Secretor characters, Ann. Human Genet. 19: 81-89, 1954. 8. Bird, G. W. G., and Lehmann, H.: The finding of haemoglobin D disease in a Sikh, Man 56: 1-3, 1956. 9. Bird, G. W. G., Lehmann, H., and Mourant, A. E.: A third example of haemo- globin D, Trans. Roy. Soc. Trop. Med. & Hyg. 49: 399-400, 1955. 10" Caminopetros, J.: Recherches sur l'anemie erythroblastique infantile, des peoples de la Mediteranee orientate, Am. Med. 43: 27-61, 104-125, 1938. 11. Ceppellini, R.: Genetica delle emoglobinopatie. Atti 14° Cong. Soc. It. Ematologia, Roma, Ottobre, 1955. Suppl. Folia Haematologica, in press. 12. Chernoff, A. I.: The human hemoglobins in health and disease, N. Engl. J. Med. 251: 322-331, 365-374, 416-423, 1955.
GENETICS OF ABNORMAL HEMOGLOBINSNEEL 269 13. Chernoff, A. I., Minnich, V. M., Chongchareonsuk, S., Na-Nakorn, S., and Cher- noff, R.: Clinical, hematological, and genetic studies of hemoglobin E, J. Lab. Sz Clin. Med. 44: 780, 1954. 14. Dameshek, W.: Familial Mediterranean target-oval cell syndromes, Am. J. Med. Sci. 205: 643-660, 1943. 15. Dunn, L. C.: Variations in the segregation ratio as causes of variation of gene frequency, Acta Genet. et Statist. lMed., 4: 139-147, 1953. 16. Dunn, L. C., and Morgan, A'. C., Jr.: Segregation ratios of mutant alleles from wild populations of Mus musculus, Am. Nat. 87: 327-329, 1953. 17. Edington, G. M.: The pathology of sickle-cell disease in West Africa, Trans. RoY. Soc. TroD. Med. & Hop. 49: 253-267. 1955. , ., .=, , 18. Edington, G. M., and Lehmann, H.: Expression of the sickle-cell gene in Africa, Brit. Med. J., Nov. 26, p. 1328, 1955. 1'>. Edington, G. M., Lehmann, H., and Schneider, R. G.: Characterization and genetics of haemoglobin G. Nature 175: 850, 1955. 20. Gatto, I.: Ricerche sui familiar) di bambini affetti da malattia di Cooley, Soc. ital. Ped.-sez. Sicilia occid. 14-XII-1947. Ref. medic. infant. 1~: 17 e Rivista di Clin. Ped., 40: 184, 1942. 21. Gatto, I.: Anemic emolitiche da anomalie ereditarie dell'hb, Acta Genet. NIed. et Gemell. 5: 427~61, 1956. 22. Huisman, T. H. J., Jonxis, J. H. P., and van der Schaaf, P. C.: Amino-acid com- position of four different kinds of human hemoglobin, Nature 175: 902, 1955. 23. Irwin, M. R.: Genetics and immunology. In: Gcrzetics in the Twentieth Century, L. C. Dunn, ea., Macmillan Co., New York, 1951, pp. xi and 634. 24. Itano, PI. A.: Qualitative and quantitative control of adult hemoglobin synthesis a multiple allele hypothesis, Am. J. Human Genet. 5: 34 45, 1953. 25. Itano, H. A.: Clinical states associated with alterations of the hemoglobin mole- cule, (The Minot Lecture), Arch. Int. Med. 96: 287-297, 1955. 26. Itano, H. A.: The hemoglobins, Ann. Rev. Biochem. 25: 331 - 348, 1956. 27. Itano, H. A., and Neel, J. V.: A new inherited abnormality of human hemoglobin, Proc. Nat. Acad. Sci. 36: 613 - 617, 1950. 28. Itano, H. A., Bergren, W. R., and Sturgeon, P.: The abnormal human hemo- globins, Medicine 35: 121-159, 1956. 29. Kaplan, E., Zuelzer, W. W., and Neel, J. V.: A new inherited abnormality of hemoglobin and its interaction with sickle cell hemoglobin, Blood 6: 1240-1259, 1951. 30. Kunkel, H. G., and Wallenius, G.: New hemoglobin in normal adult blood, Science 122: 288, 1955. 31. Lambotte-Legrand? J., and Lambotte-Legrand, C.: L'anemie a hematies falciforme~. chez l'enfant indigene do gas-Congo, Memoirs Institut Royal Colonial Belge, v. 19, no. 7, pp. 93, 1951. 3 A. 33. Lambotte-Legrand, J., and Lambotte-Legrand, C.: Unpublished data. Lehmann, H.: Human haemoglobins, St. Bartholomew s Hosp. Journal 60: 237- 242, 1956. 34. Lehmann, H.: Haemoglobin and its abnormalities, Practitioner 178: 198 - 214, 1957. 3 5. Lehmann, H., Story, P., and Thein, H.: Haemoglobin E in Burmese, Brit. Med. J., 1: 54 - 547, 1956. 36. Levin, W. C., Schneider, R. G., Cudd, J. A., and Johnson, R. E.: A family with homozygous hemoglobin C and sickle cell trait union: a clinical, hematological, and electrophoretic study, J. Lab. & Clin. Med. 42: 918-919, 1953. 37. Livingstone, F`. B.: Sickling and malaria, Brit. Med. J. 1: 762-763, 1957.
270 PART IV. GENETIC ~ SPECTS 38. Livingstone, F. B.: Ph.D. thesis: The explanation of the distribution of the sickle cell gene in West Africa, with particular reference to Liberia. University of Michigan, 1957. 42. 39. Luan Eng, L.-I.: Penjelidikan hemoglobin patologik di Indonesia, Thesis, Uni- versitas Indonesia di Djakarta, 1956. 40. Luan Eng, L., and Giok, O. H.: Homozygous hemoglobin-E disease in Indonesia, Lancet, Jan. 5, pp. 20-23, 1957. 41. Mackey, J. P., and Vivarelli, F.: Sickle-cell anaemia, Brit. Med. J., ~ (Jan. 30): 276, 1954. (From Dar es Salaam.) Miller, M. J., Neel, J. V., and Livingstone, F. B.: Distribution of parasites in the red cells of sickle cell trait carriers infected with Plasmodium falciparum, Tr. Roy. Soc. Trop. Med. & EIyg. 50: 294 - 296, 1956. 43. Morton, N. E.: The detection and estimation of linkage between the genes for elliptocytosis and the Rh blood type, Am. J. Human Genet. 3: 80-96, 1956. 44. Motulsky, A. G.: Genetic and haematological significance of haemoglobin II, Nature 178: 10551056, 1956. 45. Mourant, A. E.: Some aspects of the congenital abnormalities of hemoglobin syn- thesis, Proc. Fifth Int. Congr. Blood Transfusion, Paris, 1955. 46. Neel, J. V.: The clinical detection of the genetic carriers of inherited disease, Medicine, 26: 115-153, 1947. 47^ Neel, J. V.: The inheritance of sickle cell anemia, Science 110: 64 - 66, 1949. 48. Neel, J. V.: The inheritance of the sickling phenomenon, with particular refer- ence to sickle cell disease, Blood 6: 389-412, 1951. 49. Neel, J. V.: Data pertaining to the population dynamics of sickle cell disease, Am. J. Human Genet. 5: 154 - 167, 1953. 50. Neel, J. V.: The genetics of human hemoglobin differences: problems and per- spectives, Ann. Human Genet. 21: 1-30, 1956. 5 1. Neel, J. NT.: Human hemoglobin types: their epidemiological implications, New Engl. J. Med. 256: 161 - 171, 1957. 52. Neel, J. V.: The genetic control of hemoglobin synthesis. Proc. Sixth Meeting, International Soc. of Hematology, in press. 53. Neel, J. V., and Valentine, W. N.: Further studies on the genetics of thalassemia, Genetics 32: 38-63, 1947. 54. Neel, J. V., Wells, I. C., and Itano. H. A.: Familial differences in the proportion of abnormal hemoglobin present in the sickle cell trait, I. Clin. Invest. 30: 1120- 1124, 1951. 55. Neel, J. V., Itano, H. A., and Lawrence, J. S.: Two cases of sickle cell disease presumably due to the combination of the genes responsible for thalassemia and sickle cell hemoglobin, Blood 8: 43~443, 1953. 56. Neel, J. V., Kaplan, E., and Queller, W. W.: Further studies on hemoglobin C. I. A description of three additional families segregating for hemoglobin C and sickle cell hemoglobin, Blood 8: 720734, 1953. A. Neel, J. V., Hiernaux, J., Linhard, J., Robinson, A. R., Zuelzer, W. W., and Liv- ingstone, F'. B.: Data on the occurrence of hemoglobin C and other abnormal hemoglobins in some African populations, Am. J. Human Genet. 8: 138-150, 1956. 5g. Pauling, L., Itano, H. A., Singer, S. J., and Wells, I. C.: Sickle cell anemia, a molecular disease, Science 110: 543-548, 1949. 59. Powell, W. N., Rodarte, J. G., and Neel, J. V.: The occurrence in a family of Sicilian ancestry of the traits for both sickling and thalassemia, Blood 5: 887- 897, 1950.
GENETICS OF ABNORMAL HEMOGLOBINSNEEL 271 ^ Ranney, H. M., Larson, D. L., and McCormacl<, G. H., Jr.: Some clinical, bio- chem~cal, and genetic observations on hemoglobin 1277-1284, 1953. 61. Raper, A. 13.: The significance of sicklaemia in Uganda. Proc. Fifth Int. Congr. Of Blood Transfusion, Paris, 1955. 62. Robinson, A. R., Zuelzer, W. W., Neel, J. V., Lie ingstone, F. B., and Miller, NI. J.: Two "fast" hemoglobin components in Liberian blood samples, Blood 11: 902-906, 1956. Sandler, L., and Novitski, E.: Meiotic drive as an evolutionary force, Amer. Nat. 9f: 105-110, 1957. 64. Schwartz, H. C., Spaet, T. H., Zuelzer, W. W., Neel, l. V., Robinson, A. R., and Kaufman, S. F.: Combinations of hemoglobin G. hemoglobin S. and thalassemia occurring in one family, Blood 12: 238-250, 1957. 65. Silvestroni, E., and Bianco, I.: Genetic aspects of sickle cell anemia and micro- drepanocvtic disease, Blood 7: 429~35, 1952. 66. Singer, K.: Hereditary hemolytic disorders associated with abnormal hemo- globins, Am. J. Med. 15: 633-652, 1955. 67. Singer, K, Chernoff, A. I., and Singer, L.: Studies on abnormal hemoglobins. I. Their demonstration in sickle cell anemia and other hematologic disorders by means of alkali denaturation, Blood 6: 413~28, 1951. 68. Singer, E;:., Chernoff, A. I., and Singer, L.: Studies on abnormal hemoglobins. lI. Their identification by means of the method of fractional denaturation, Blood 6: 429~35, 1951. 69. Singer, K., Kraus, A. P., Singer, L., Rubinstein, H. M., and Goldberg, S. R.: Studies on abnormal hemoglobins. X. A new syndrome: hemoglobin C-thalas- semia disease, Blood 9: 1032-1046, 1954. 70. Singer, K., Singer, L., and Goldberg, S. R.: Studies on abnormal hemoglobins. XI. Sickle cell-thalassemia disease in the Negro. The significance of the S+A+F and S+A patterns obtained by hemoglobin analysis, Blood 10: 405~15, 1955. 71. Spaet, T. H., Alway, R. PI., and Ward, G.: lIomozygous type "C" hemoglobin, Pediatrics 1~?: 483~90, 1953. 72. Sturgeon, P., Itano, H. A., and Valentine, W. N.: Chronic hemolytic anemia associated with thalassemia and sickling traits, Blood 7: 350 - 357, 1952. 73. Thorup, O. A., Itano, H. A., Wheby, M., and Leavell, B. S.: Hemoglobin J. Science 123: 889890, 1956. 74. Torbert, J. V., Jr., and Smith, E. W.: Simultaneous inheritance of three adult hemoglobins determined at two genetic loci, Clin. Res. Proc. 5: 138, 1957. 75. Valentine, W. N., and Neel, I. V.: Hematologic and genetic study of the trans- mission of thalassemia, Arch. Int. Med. 74: 185196, 1944. 76. Vandepitte, J.: Aspects quantitatifs et genetiques de l'anomalie falciforme a Leopoldville. Proc. Fifth Int. Confer. Blood Transfusion, Paris, 1955. 77. Vandepitte, J. M., Zuelzer, W. W., Neel, J. V., and Colaert, J.: Evidence con- cerning the inadequacy of mutation as an explanation of the frequency of the sickle cell gene in the Belgian Congo, Blood 10: 341-350, 1955. 78. Wells, I. C., and Itano, If. A.: Ratio of sickle-cell anemia hemoglobin to normal hemoglobin in sicklemics, l. Biol. Chem. 188: 65-74, 1951. 79. White, J. C., and Beaven, G. H.: A review of the varieties of human haemoglobin in health and disease, J. Clin. Path. 7: 175 - 200, 1954. 80. Zuelzer, W. W., Neel' J. V., and Robinson, A. R.: Abnormal hemoglobins. In: Progress in Hematology, L. M. Tocantins, ea., pp. 91-137. Grune and Stratton, New Yor}:, 1956. C, J. Clin. Invest. 32:
272 PARIS IV. GENETIC ASPECTS < DISCUSSION Dr.. Salome G. f aelsch: It was a great pleasure for me to hear Dr. Neel's presentation of the genetics of abnormal hemoglobins, and I listened to it while still under the impact of the most impressive demonstration by Dr. Ingram on the nature of the gene effect in the case of sickling. Dr. Ingram seems to feel that the hemoglobins are rather direct and primary products of the gene. This reminded me of the time when the biochemical genetics of ne2~rostora was still young and when enzymes were considered primary gene products, a concept which since then has been modified exten- sively. I wonder whether perhaps some caution is not also indicated if one regards hemoglobins as primary gene products. Some of Dr. Neelts data which he presented here today I, at least, feel cannot be reconciled with the idea that hemoglobins are really primary gene products. For example, he reported the very interesting fact, of which I was not aware before, that individuals heterozygous for the sickling gene and for the hemoglobin G gene, do not show the presence of any normal hemoglobin. If hemoglobins are primary gene products, then one would expect normal hemoglobin which is produced by the normal gene in the case of the sickle heterozygote, to be found somewhere in these double heterozygotes, and the same should be true for the hemoglobin G. particularly if the difference be- tween normal and abnormal hemoglobin is only due to the substitution of one amino acid in one of the hemoglobin peptides. In this connection, I might perhaps just call attention to some work that Dr. Ranney and I have been doing on genetically-determined different hemo- globins in the mouse, which, I must admit after having listened to Dr. Neel, is not nearly as good an experimental subject as man. Nevertheless, some of our data seem to indicate at least that there is interaction of primary gene products, if you want to use that term, on a rather early level, and that hemo- globins may perhaps turn out not to be primary gene products after all. I shall be very much interested to hear what Dr. Neel has to say about this. Dr. Neel: I can only endorse wholeheartedly what Dr. Waelsch has just said. As a matter of fact, I think I was fairly careful not to identify hemo- globins with primary gene products. But if ever there was an opportunity to get pretty close to primary gene products, I think we have it in this system. I am as impressed as you are by the apparent failure of hemoglobin A to be formed in the double heterozygote for the genes responsible for hemoglobins S and G. It is that fact which has dissuaded me for the time being from attempting to formulate any biosynthetic sequences. However, when you say that the findings are not consistent with hemoglobins being primary gene products, I am not sure that we know enough about what is consistent with primary gene products to say that this is inconsistent. Dr. Harvey Ita~zo: I echo Dr. Waelsch's concern about this absence of A.
DISCUSSION 273 However, I would like to point out certain pertinent facts in these cases, namely, there is a pair of siblings, both of whom are supposed to have SG. However, they have different electrophoretic patterns. One of them has three components by moving boundary electrophoresis, and the other has two. More- over, the latter only shows an S spot on paper. Be that as it may, I am still inclined to the view that the amino acid sequence in a hemoglobin molecule reflects exactly the information carried in the genes. If two loci that segregate independently determine sequences in different parts of the molecule, the normal molecule would be an association product which contains both sequences. A doubly heterozygous red ceil would syn- thesize four polypeptides, including a normal and an abnormal polypeptide, corresponding to each of the two loci. These would be potentially capable of associating into four types of molecules, the normal form, two different ab- normal forms, and a doubly abnormal form. One of the SO individuals pre- sumably has a thalassemia allele. If the thalassemia allele is the absence or the complete malfunction of a locus for hemoglobin, at most only three different polypeptides can result from the taco loci. If the two loci correspond to dif- ferent parts of the hemoglobin molecule, the polypeptides corresponding to allelic genes would not associate, and the three nolvPentides could form only two different hemoglobin molecules. 1 , ~ Dr. Neel: With reference to the speculations of Dr. Itano, the possibilities for speculation here are almost infinite. I think most of us should have felt that if the genetic interpretation which I have presented is correct and there is ~ G locus and an S locus, the double heterozyg;otes should show an appreciable quantity of hemoglobin A. The fact is that we did not find appreciable quanti- ties of hemoglobin A. This observation has discouraged me from speculation as to what we should find. I can only express the hope that those of you who are in charge of hematological laboratories will help us dig out the families that will give us facts of this nature, so important in developing an understanding of the biochemical genetics of this situation. Dr. I. Herbert Scheinberg: I would like to ask Dr. Reel a question. If there is some appreciable incidence of spontaneous mutation to the sickle cell gene, is there any way of explaining or speculating as to why this should occur only, or almost only, in Negroes, and Why you should not get some mutation in white people? Dr. Neel: Why don't we find hemoglobin S in persons other than Africans ? This is something that I have speculated on under other circumstances. We have never had the kind of large-scale survey on Caucasians or Mongolians which would tell us just how frequent the hemoglobin S gene may be in these groups. But on indirect grounds, such as the paucity of cases of sickle cell anemia in Caucasians with no known or probable Negro admixture in in~- mediate or remote times, we can infer that it is quite rare. Certainly this sug- gests to me a mutation rate below 10-';, but I don't think we can exclude
274 PART IV. GENETIC ASPECTS mutation rate of 10-' in Caucasian and Mongolian populations. Increasingly, my oven thinking is that, despite certain cases of apparent mutation and I have published some of them probably the sickle cell mutation is a rela- tively rare one by our present standards, even in Negro populations. It must be regarded as a mutation which may have occurred infrequently in the history of the human race and which confers a special advantage, and so flourishes, under a particular set of circumstances, namely, those encountered in regions where malaria pressure is intense. Whether other factors than malaria pres- sure are of importance in maintaining a high f requency of the sickle cell gene is still an open question. This is the best that I think anyone can do at the present time. It is be- cause of this indirect evidence that the sickle cell mutation is rare that I em- phasized the necessity of studying apparent cases of mutation with every tool at our disposal. We have found too many of them to reconcile with this other evidence that makes us think the mutation must be a very rare one. At the moment, it appears as if some of these apparent mutations are instead to be explained by our old friend the thalassemia gene, very well disguised. Dr. DVilliam R. Bergren: Recently in our laboratory we have had another blood sample from Dr. Sturgeon's clinic containing hemoglobin H. This occurrence follows very closely the reported case of Rigas and his associates' in that there is no evidence in the parents of the presence of the fast hemoglobin. In this particular case, as with Rigas, there is thalassemia on one side of the family, with the hemoglobin ~ presumably coming from the other. Seventeen members of the non-thalassemia side have been studied, with no funding of hemoglobin H beyond that in the propositus. These results parallel what has been mentioned by Rigas, by Gouttas, and, at this Conference, by Motulsky. The studies on the present family add to the data from two previous occur- rences of hemoglobin H at our hospital. I would like to go back a couple of years to work in our laboratory which has not been reported as yet except informally. In late 1954 we observed on a paper electrophoresis run a hemoglobin with a mobility considerably higher at pH S.6 than that of the normal hemoglobin A. Naturally this was very exciting, but we were cautious in feeling that our observation was a new one, for we had heard through Dr. Itano that Dr. Jensen and Dr. Rucknagel at Duke University had been working with a fast hemoglobin. On cross-checking samples with Dr. Jensen, it was found that our hemoglobin was different from the Duke material (which is now known as hemoglobin I). Meanwhile, we had heard that Dr. Rigas had been studying a fast hemoglobin and that Dr. Motulsky also had two occurrences of a similar one. On crost;-checking sam- ples, it was found that our occurrence checked very closely with the hemo- globin H of Rigas. A second independent occurrence of a fast hemoglobin in our laboratory checked out identically. Meanwhile, cross checks also were
DISCUSSION 275 made against the fast hemoglobin found by Dr. Thorup and Dr. Itano (now hemoglobin J.). It is of interest that our early occurrences of hemoglobin H seem quite different in family relationship from our recent one or from those reported by others. In each of the two early cases, the hemoglobin H was found in elderly men, and in neither instance was there any Oriental connection in the family. The first case was a 76-year-old man of Scottish ancestry. He had been robust and healthy until age 70, at which time he underwent ~,enitourinary tract surgery. Soon thereafter an anemia developed, and the condition progressed until supportive transfusions there necessary every few weeks. We studied his three children, six grandchildren and a sister. In no case was the fast hemo- globin found, and similar results were had with two correlative lines of the family. Although a genetic relationship could not be established, neither divas it ruled out. The interesting point is that nothing in our study or in history of the family suggested the presence of thalassemia. The second occurrence was in ~ man of about the same age. No family re- mained to be studied, but nothing in our examination of him or of his history divas suggestive of thalassemia. Our assumption has been that some stress cir- cumstance is necessary for hemoglobin H to become manifest, such a cir- c~mstance being thalassemia on the other side of the family in our recent oc- currence and in the published instances. In the two early cases, we considered the possibility of the stress of age as an evocative circumstance for the fast hemoglobin production, provided that the genetic possibility is present. Of course, this concept of a geriatric hemoglobin has only speculative grounds on the basis of present information. It is apparent now that it was a mistake on my part to use an alphabetical designation as a provisional dame for our first occurrence of a fast hemo- globin. The term hemoglobin X has crept into the literature through personal communication, 2 and it has not been clear that hemoglobin X corresponds to hemoglobin H. It would have been much better to designate the new hemo- globin by a noncommittal term such as Los Angeles No. 1, pending compara- tive study. Recently this practice has been followed by Dr. Zuelzer and his colleagues with their fast hemoglobins from Africa, and it has been adopted by Dr. Chernoff for his apparently new fast hemoglobin, that which he has referred to in this Conference as Durham No. 1. . :REFEREN CES 1. Rigas, Demetrios A., Koler, Robert D., and Osgood, Edwin E.: Hemoglobin H: Clinical, laboratory, and genetic studies of a family with a previously un- described hemoglobin, J. Lab. Clin. Med. 47: 51, 1956. 2. Rucknagel, D. L., Page, E. B., and Jensen, W. N.: Hemoglobin I: An inherited hemoglobin anomaly, Blood 10: 999, 1955.