The Molecular Basis of Genetic Defects

First Inter-American Conference on Congenital Defects. Los Angeles, California, January 22 to 24, 1962

I am glad to have this opportunity to express my appreciation of the work that the National Foundation has been carrying on. I have been for many years the recipient of help from the National Foundation, which has enabled me and my coworkers to carry out many scientific studies that would not have been possible otherwise.

Just two days ago I had an illustration of the value of the work being done by the National Foundation. As my wife and I drove through a small hamlet in Southern Chile, we saw, on the wall of a small building, three placards exhorting all people to come in, and to bring their children, to be vaccinated against poliomyelitis with the Sabin oral vaccine.

The past fifty years have seen astounding developments in the field of science and in our understanding of the nature of the world. Fifty years ago it had been known for only one year that the atom consists of one or more electrons surrounding a heavy nucleus that is very small in size in comparison with the atom itself. The first significant theory about the electronic structure of atoms, the Bohr theory, was not developed until the following year. Fifty years ago, in 1912, the fundamental discovery of diffraction of x-rays of crystals was made, by Max von Laue, and during that year and the following year the technique of x-ray diffraction was used by W. L. and W. H. Bragg for the first time to determine the structures of crystals. Since then thousands of crystals and of gas molecules have been subjected to thorough experimental investigation by diffraction methods and also by spectroscopic methods of various sorts. Quantum mechanics, the fundamental theory of molecular structure, has been developed and extensively applied. The understanding of the nature of the physical world that exists now is so much greater and deeper than that that existed fifty years ago that it is hard to make a significant comparison of present knowledge and the knowledge of 1912.

I am confident that there is going to be similar progress in the field of biology and medicine during the next few decades, such that the period of fifty years beginning about ten years ago may well, in the year 2000, be called the golden age for biology and medicine.

I believe that we are beginning to understand the nature of congenital defects in a fundamental way, going back to the molecules that make up the body of the defective infant. This knowledge, which at the present time is fragmentary, will, I am sure, lead in a few years to improved therapy, to palliative treatments, to preventive measures that will be a great contribution to the solution of the problem of decreasing the amount of human suffering in the world.

Many congenital defects are the result of the inheritance by the child, or the presence in the child as a result of a new mutation, of one or more defective genes that manufacture molecules that do not operate in the same way as the molecules that are manufactured by the corresponding genes in normal human beings. The molecules that they manufacture are for the most part enzymes that catalyze certain chemical reactions. If, in the course of time, we obtain an understanding of the nature of the defective molecules, especially the defective enzymes, there is the possibility that this knowledge will permit something to be done to decrease the harm caused by the abnormal molecules. Nobody knows the detailed molecular structure of any enzyme, but there is little doubt that knowledge of the detailed structure of some enzyme molecules will be obtained within a few years, and that the difference in molecular structure of the defective enzymes associated with some congenital diseases in human beings and the normal enzymes will be discovered.

As an illustration of the way in which progress may be made, I shall discuss the structure of the hemoglobin molecule in relation to health and disease. I have been working on hemoglobin for 28 years, in collaboration with many able scientists, to whom I am grateful: Harvey Itano, Charles D. Coryell, A. E. Mirsky, Fred Stitt, Richard W. Dodson, Charles D. Russell, Robert B. Corey, Robert C. C. St. George, Allen Lein, S. J. Singer, Ibert C. Wells, Walter A. Schroeder, Lois M. Kay, Herbert S. Rhinesmith, Alex Rich, W. R. Bergren, Philip Sturgeon, Richard T. Jones, Jerome R. Vinograd, Y. Nishiwaki, and Emile Zuckerkandl.

A few years ago it was discovered (L. Pauling, H. A. Itano, S. J. Singer, and I. C. Wells, Science 110, 543 (1949)) that the disease sickle-cell anemia is a disease of the hemoglobin molecule. The patients with this disease manufacture abnormal molecules of hemoglobin, rather than normal adult human hemoglobin molecules. There is strong indication that the abnormal molecules of hemoglobin are directly responsible for the symptoms of the disease. Since then it has been discovered that there are many other hereditary hemolytic anemias that are caused by abnormal kinds of human hemoglobin.

The disease sickle-cell anemia was first described by Dr. J. B. Herrick in 1910. The red cells of patients with this disease undergo a change in shape when they are partially deoxygenated, as in venous blood. This change in shape, which is called sickling (deformation to resemble a sickle), can also be caused to occur in a specimen of blood from a parent of a sicklo-cell-anemia patient, by making the partial pressure of oxygon small enough; sickling does not occur to significant extent in the venous blood of these individuals, who are not anemic. It was suggested by Emmel in 1917 and by Taliaferro and Huck in 1923 that the disease is hereditary, and the Mendelian inheritance was made clear by Neel (J. V. Neel, Science 110, 64 (1949)). Neel presented evidence that there is a sickle-cell gene that causes some tendency to sickling when it is present in the heterozygous condition (one slckle-oell gene and one normal gene) but does not cause sickle-cell anemia except when it is present in the homozygous condition (two sickle-cell genes).

The fact that the red cells of sickle-cell-anemia patients undergo the change in shape in venous blood but not in arterial blood suggested that the hemoglobin molecule is responsible for the disease, and led to the experiments showing that sickle-cell anemia is a molecular disease and that the patients manufacture an abnormal kind of hemoglobin. It was found that at pH 6.90 normal adult human hemoglobin has a negative electric charge, and moves toward the anode in an electrophoretic cell, whereas sickle-cell-anemia hemoglobin moves toward the cathode, and has a positive charge, at this pH. The difference in electric charge of the two kinds of molecules is about two electronic charges. The hemoglobin from the red cells of sickle-cell heterozygotes is a mixture of the two kinds, usually with normal adult hemoglobin present in amount somewhat greater than fifty percent. It may be concluded that the gone for normal adult human hemoglobin and the gene for sickle-cell hemoglobin operate essentially independently of one another in manufacturing their hemoglobin molecules, and that the manufacture of normal adult hemoglobin is somewhat more efficient than that of sickle-cell hemoglobin.

It is believed that the molecules of sickle-cell-anemia hemoglobin, which we may represent by the symbol HbS, are self-complementary, so that they clamp on to one another, to form a long rod of these molecules. The rods then line up side by side to make a sort of needle-like crystal, a liquid crystal of the nematic type, which, as it grows, becomes longer than the diameter of the normal red cell. The red cell is then twisted out of shape, its surface becomes sticky, and the cells are recognized by the spleen as abnormal and are destroyed rapidly. Moreover, they tend to stick together and to clog the capillaries, interfering with the flow of blood to organs, and causing damage by anoxia. The oxygenation of the hemoglobin molecules destroys the self-complementariness in structure, and the sickling process is reversed.

The first paper on sickle-cell anemia as a molecular disease contained the statement that the four heme groups of HbS are identical with those of HbA, and that the difference between the two molecules is a difference in the globin. Amino-acid analyses of HbA (normal adult human hemoglobin) and HbS showed no detectable difference in composition.

During the last few years extensive studies of the chemical nature of hemoglobin molecules have been made. It was found by Rhinesmith, Schroeder, and collaborators (H. S. Rhinesmith, W. A. Schroeder, and L. Pauling, J. Am. Chem. Soc. 79, 4682 (1957); H. S. Rhinesmith, W. A. Schroeder, and N. Martin, ibid. 80, 3358 (1958)) that HbA contains two different kinds of polypeptide chains. It contains two alpha chains, which begin, at the free amino end, with the amino-acid residues valine-leucine-, and two beta chains, which begin with the sequence valine-histidine-leucine-. It had been shown by M. Perutz a number of years ago, through x-ray investigation of crystals, that the hemoglobin molecules of most animal hemoglobins have a two-fold axle of symmetry, which requires that polypeptide chains occur in identical pairs. The HbA alpha chain contains 142 amino acid residues, and the beta chain contains 147. The four chains, each carrying a heme group, are bonded together to form a normal adult human hemoglobin molecule. Vernon Ingram then developed a powerful new method of studying the composition of the polypeptide chains in hemoglobin, by enzymatic splitting of the chains into peptides and separation of the peptides (26 for the HbA molecule) by paper electrophoresis-chromatography, and subsequent analysis of the amino-acid sequence in the individual peptides. He was able to show that the half-molecule of HbS differs from that of HbA by only one amino-acid residue: a valyl residue occupies the position in HbS that is occupied by a glutamate residue in HbA (V. M. Ingram, Biochem. et Biophys. Acta 28, 539 (1958); 36, 402 (1959)). It was then shown by Ingram and by Schroeder that this difference lies in the beta chain; the alpha chains are the same for HbA and HbS; and, in fact, the residue that is different in the normal beta chain and the sickle-cell beta chain is the sixth residue from the free amino end of the chain.

It has accordingly been found that the difference in structure of the sickle-cell-hemoglobin molecule and the normal molecule amounts to only one part in 289; only one of the 289 amino-acid residues in the half-molecule is different in the two forms. This small difference in the nature of the hemoglobin molecule - a difference of about one third of one percent in chemical composition - is apparently enough to produce the complementariness in structure between the sickle-cell-hemoglobin molecules that leads to the aggregation of these molecules and the manifestations of the disease.

I estimate that a human being inherits 100,000 genes from his parents. The sickle-cell-anemia patient has inherited two abnormal genes out of the 100,000, and we may say that this genic abnormality is 0.00002. But each of the two abnormal genes is abnormal to the extent of only one part in 147 (the number of amino-acid residues in the beta chain - the alpha chain and the beta chain are manufactured under the control of separate genes). This abnormality of the gene, assumed to be the same as of the hemoglobin molecule whose manufacture it controls, amounts to about 0.7 percent. Hence the total abnormality in genetic character that dooms the sickle-cell infant to a short life of suffering and early death amounts to 0.7 percent of 0.00002; that is, to less than one part in a million of the genetic material that he has inherited.

The properties of a protein molecule are determined not only by the sequence of amino-acid residues in its polypeptide chains but also by the way in which the chains are folded. The principal ways of folding polypeptide chains have been determined, through studies of amino acids, simple peptides, and fibrous proteins by x-ray diffraction methods. In particular, a stable helical configuration, called the alpha helix, has been discovered and shown to be present in many proteins. In this type of folding the polypeptide chain is coiled into a helix, with 3.6 amino-acid residues per turn of the helix. Each amino-acid residue is attached by hydrogen bonds to the residue fourth removed from it in each direction along the chain. The alpha helix is the characteristic structural feature of fibrous proteins of the alpha-keratin type, including hair, horn, fingernail, and muscle.

Great progress toward the ultimate solution of the problem of determining the complete structure of the hemoglobin molecule has been made during recent years through the x-ray investigation of crystals of myoglobin by Kendrew and of hemoglobin by Perutz and collaborators. Although the structures have not yet been completely determined, it has been shown that in myoglobin and hemoglobin molecules the polypeptide chains are coiled in such a way that each chain consists of about seven segments with the configuration of the alpha helix, connected by portions of the chain that do not have this configuration. About 70 percent of the chain is in the alpha-helix configuration.

The structure of the molecule of normal adult human hemoglobin and of sickle-cell-anemia hemoglobin will, with little doubt, be completely determined within a few years. It will be possible then to understand not only the ordinary properties of hemoglobin - the reversible combination with oxygen; the heme-heme interactions in hemoglobin leading to the sigmoid oxygen equilibrium curve; the interaction of each heme group with two acidic groups to produce the Bohr-Hasselbalch effect (increase in acidity of the groups on oxygenation of the heme); and various other properties - but also the complementariness in structure of the HbS molecules, their difference in structure from molecules of HbA that leads to the formation of tactoids and to the manifestations of the disease sickle-cell anemia. The same technique of x-ray diffraction of protein crystals that has been developed so effectively by Kendrew and Perutz will without doubt also, during the next few years, be applied in the determination of the detailed structure of proteins with important enzyme activities, leading to a detailed understanding of the mechanism of catalytic action of these substances. I can think of no development in science of greater importance to our understanding of life than this one.

The possibility of obtaining a detailed molecular understanding of genic interference with physiological function can be illustrated by a discussion of the amino-acid substitutions that lead to methemoglobinemia, a disease in which the iron atoms of the hemoglobin molecule are oxidized to the tripositive state. The alterations of amino-acid sequence in the polypeptide chains of hemoglobin that lead directly to an interference with the oxyphoric function of hemoglobin are those of the forms of hemoglobin called HbM (P. S. Gerald, in "The Metabolic Basis of Inherited Disease," J. B. Stanbury, J. B. Wyngaarden, and D. S. Fredrickson, Editors, pp. 1068-1085, McGraw-Hill, New York). The structural studies of these forms of the hemoglobin molecule have revealed that the amino-acid substitutions leading to formation of HbM all affect a certain region of the hemoglobin polypeptide chains. This region is called the basic center I, to distinguish it from a second basic center further along the chain. The two basic centers have in common the property of containing a histidine residue that is thought to be closely connected with the iron atom of the heme group.

A series of seven consecutive amino-acid residues for the normal alpha chain, the normal beta chain, and three types of chains in human HbM are shown in Table I. The sequence of residues comprises residues number 56 to 62 in the alpha chain and 61 to 67 in the beta chain, counting from the amino end. It is believed that the third residue, a histidine residue, is involved in close interaction with the iron atom, in both the alpha chain and the beta chain. In HbM_Boston it is seen that this residue is replaced by a tyrosine residue in the alpha chain, and in HbM_Emory the residue is also replaced by a tyrosine, but in the beta chain rather than the alpha chain. In HbM_Milwaukee it is again the beta chain that is abnormal, but the abnormality does not involve the histidine residue; instead, the valine residue four removed from histidine along the chain is replaced by a glutamic acid residue. Other known substitutions in positions two removed from the histidine residue or one removed do not lead to methemoglobinemia. The possibility that the residue four removed from the histidine can interfere with the interaction between the histidine residue and the iron atom may be made reasonable by the consideration of the fact that in a polypeptide chain with the configuration of the alpha helix the fourth residue is closer to the histidine residue than are the intervening residues, because there are 3-6 residues per turn of the helix. [See Table I]

Detailed information bearing on the question of the mechanism of interference with function by a change in molecular structure is as yet scanty. We may be confident that the next few years will see a great increase in the amount of information and a consequent great increase in our understanding of the molecular basis of genetic diseases.

We may raise the question as to why there are so many people in the world, many millions, who carry the gene for sickle-cell-anemia hemoglobin, such that when two of these heterozygotes marry one another one quarter of their children, on the average, are born with a gross congenital defect, the disease sickle-cell anemia. The sickle-cell gone is clearly disadvantageous, because the homozygotes die without progeny. Through their death the genes are removed from the pool of human germ plasm. In the case of some deleterious genes we may accept, as an explanation of their incidence in the pool of human germ plasm, the existence of a steady state, involving the rate at which the defective genes are removed from the pool and the rate at which new defective genes are introduced into the pool through damage of normal genes by high-energy radiation, chemical mutagens, thermal agitation, or other mutagenic agents. The estimate made by geneticists of the probability of these events is that a mutation occurs in about one out of 25,000 or one in 50,000 genes for hemophilia, phenylketonuria, galactosemia, achondroplasia, and some other defects.

The incidence of the sickle-cell gene among certain populations is far too high, as great as one in 20, among certain populations to be accounted for reasonably in this way (J. V. Neel, Cold Spring Harbor Symposia Quant. Biol. 15, 141 (1951)). It was suggested by Brain (P. Brain, Brit. Med. J. ii, 880 (1952)) that the presence of hemoglobin S in the red cells might give protection against malaria, and thus confer an advantage on the sickle-cell heterozygote that would more than balance the disadvantage of the lethal character of the disease in the homozygote. A test of this hypothesis was carried out by Allison (A. C. Allison, Brit. Med. J. i, 290 (1954)), who infected 15 healthy adult Africans with sickle-cell heterozygosity and 15 similar healthy adult Africans without the sickle-cell gene with Plasmodium falciparum. They were infected by subinoculation with 15 ml of blood containing a large number of trophozoites or by being bitten by heavily infected 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 not carrying the sickle-cell gene, and in only two of the 15 heterozygotes. Allison concluded that the abnormal erythrocytes of individuals with one sickle-cell gene are less easily parasitized by P. falciparum than are normal erythrocytes, and that accordingly those individuals who are heterozygous for the HbS allele have a selective advantage over normal individuals in regions where malaria is hyperendemic. It is possible that many of the abnormal genes which in double dose lead to serious congenital defects are or have at some time in the past been advantageous to the heterozygotes. Another possible example is cystic fibrosis, which is responsible for a seriousdefect in about one child in 800.

In a malarial region the population might consist largely of sickle-cell heterozygotes. The children of pairs of heterozygotes would be viable to the extent of only fifty percent: a quarter would be normal individuals, who would die of malaria, and a quarter would be patients with sickle-cell anemia, who would die of this disease. It would, of course, be better if the mutated gene that provided the protection against malaria had not changed quite so much from the normal gene - had, we might say, changed only half as much, so that it would provide protection against malaria when present in double dose. Then all children would be homozygotes in this newly mutated gene, and would have the corresponding abnormal hemoglobin in their red cells, undiluted with HbA. But there is no good mechanism in an ordinary population for spreading the gene that provides protection when it is present in double dose. After the original mutation, there would be a few heterozygotee, but, except in a highly inbred community, the chance for homozygotes to be born would be small, and the incidence of the mutated gene would not increase.

We might well believe that if malaria had not been overcome in another way - by the development of drugs - another mutation would in the course of time have occurred, such that the newly mutated gene when present in double dose or with the sickle-cell gene would provide protection against malaria, and would not give rise in the homozygotes to a disease. It is possible that the change in genetic character of humans and other organisms has often involved two steps of this sort, rather than one.

Life is a relationship among molecules, and not a property of any one molecule. Disease, also, is a relationship among molecules. There are molecular diseases, but no diseased molecules. At the level of the molecule we find only variations in structure and in physicochemical properties, and rarely can we detect any criterion by means of which one molecule can be placed higher or lower on the evolutionary scale than another molecule. Human hemoglobin, although different from horse hemoglobin, seems to be in no way more highly organized.

It is possible to reach some conclusions about the evolutionary history of polypeptide chains of hemoglobin from consideration of the amino-acid sequences. The polypeptide chains of horse hemoglobin and human hemoglobin differ by about 18 point mutations per chain. From paleontological evidence it may be estimated that the common ancestor of man and horse lived until about 150 million years ago, and that accordingly one point difference has arisen on the average every seven million years.

The alpha chain for gorilla hemoglobin differs from that for human hemoglobin by two point mutations, and the beta chains differ by one. These numbers suggest, with use of the basic mutation rate mentioned above, that gorilla and human separated from their common precursor some time around seven million to fourteen million years ago. Further studies of animal hemoglobins may lead to a great amount of information about evolution.

Research on the molecular basis of disease is not only of fundamental importance to biology, but of great interest to medicine as well. We may hope that the knowledge that is obtained during the next decades about the molecular structure of the human body in health and disease will lead to a significant decrease in the amount of human suffering.

A most satisfying way of meeting the challenge of molecular disease would be the elimination of the disease-causing mutant genes from the human populations. I believe that no objection can be legitimately raised against the ambition to eliminate from the pool of human germ plasm those genes that lead to clearly pathological manifestations of disease and to great human suffering. For example, we know now that in the United States about ten percent of the Negro population, and a smaller percent of the remaining populations carry one HbS or HbC gene. About one child in 400 born to Negro parents has the deadly disease sickle-cell anemia. A simple test permitting the detection of the heterozygous carriers of the sickle-cell gene exists. As a first protective step there should, I believe, be a law requiring that ail persons submit themselves to this test, and that they have knowledge as to whether or not they carry the gene.

If people carrying the mutant gene wore to refrain from marrying one another, but married normal individuals, there would be no children born with the disease sickle-cell anemia, but the incidence of the gene would remain constant in the population, and the problem of eliminating the gene would not be solved. The following rules may be proposed to eliminate the mutant gene, with the minimum amount of human suffering. If two heterozygotes marry they should have no children of their own. If a heterozygote marries a normal person they should have a number of progeny smaller than the average. Thus the mutant gene would be eliminated in the course of time in a way not involving the suffering caused by the birth of the defective children. Similar measures should be taken in the case of phenylketonuria and other molecular diseases, when a satisfactory test for heterozygosity has been discovered.

I believe that we have now come to the epoch in the history of civilization when it has become essential that we take action that will lead to the diminution in incidence of defective genes in the pool of human germ plasm in a way that does not involve the cost of the suffering of hundreds of thousands of grossly defective children who are born each year, year after year. I believe that the increase in knowledge about the molecular structure of the human body and the nature of molecular diseases can permit a significant decrease to be achieved in the amount of human suffering in the world.