Scientific Research
Researching at NIH
1. Inroads into Cancer Research
Although my initial intention at NIH was to work on amino acid antagonism, I soon found out that it would be preferable if I would work on some cancer-related problem. I got my initiation into the field by having read the classical book, Biochemistry of Cancer (1), written by my sponsor Dr. Jesse P. Greenstein ( 2).
Before I developed a project of my own, Dr. Greenstein introduced me to a young physician, Dr. Howard Radwin from the Clinical Center, who wanted to conduct research in his lab, using Greenstein’s newly developed liquid synthetic diet. He was spending a two-year assignment in Public Health Service at NIH, in lieu of his military service. At Greenstein’s suggestion, we worked together on a collaborative study of metabolic changes associated with cancer surgery. The study was based on the following premise. Major exenterative surgery, such as was performed by the Surgery Branch of the Cancer Institute, was disruptive to the general metabolism of the patient, and recovery was a burden to the reserves of those who were frequently already debilitated. The demands of convalescence might be eased if the patient were provided with the proper nutrition in the postoperative period. However, this would depend on the ability of the patient to utilize these nutrients. In this study an attempt was made to define the effect of major trauma on the metabolism of young animals and the ability of different nutritional programs to influence the changes observed. A chemically defined, synthetic diet, developed in Greenstein’s laboratory, was utilized to make the quantitative aspects of the data more precise. The diet was water soluble and contained the “L” forms of all essential amino acids (3) and eight nonessential amino acids, as well as vitamins, salts, trace elements, and glucose. The composition of the diet was rigidly controlled, which permitted accurate daily nitrogen balance (4) determinations. The experimental groups were subjected to a crush injury of sufficient severity to produce major metabolic reaction without proving fatal.
The ability of major trauma to alter the protein metabolism in the young animal was clearly manifested by retardation of growth rate and diminution of the positivity of nitrogen balance. Nevertheless, it appeared that these effects can be moderated to the benefit of the organism by a balanced diet supplied in adequate amounts. The effect of injury on the growth rate was influenced by the nutritional status of the animal, thus, when the growth rates of injured and uninjured animals on an 18-cc. diet were compared, a marked lag in growth of the former was observed, while the difference in those receiving 4 cc. was considerably less in extent and duration. The data on nitrogen balance support the hypothesis that the injured animal can utilize balanced dietary supplements for growth. The carcass analyses, peformed after the experiment was terminated, clearly indicated identical proportions of nitrogen and water in injured animals receiving an increased intake, in uninjured animals, and in injured animals not receiving added nutrients. It thus cannot be argued that the weight gain was due simply to water retention. While the diet may not be utilized as efficiently in the postoperative period as it is under normal conditions, there is an advantage to the maintenance of a high level of an intake. There was also strong indication that ad libitum ingestion is preferable to 4 daily tube feedings. (5).
Endnotes
1. National Institutes of Health. Bethesda, MD
2. Jessie P. Greenstein, Biochemistry of Cancer (New York: Academic Press, 1954).
3. Jesse P. Greenstein (1902–1959) held the position of Chief of Laboratory of Biochemistry in the National Cancer Institute.
4. An essential amino acid or indispensable amino acid is an amino acid that cannot be synthesized de novo by the organism (usually referring to humans), and therefore must be supplied in the diet.
5. The state of the body in regard to ingestion and excretion of nitrogen. In negative nitrogen balance, the amount excreted is greater than the quantity ingested; in positive nitrogen balance, the amount excreted is smaller than the quantity ingested.
6. Howard M. Radwin, Miloslav Rechcigl, Jr., and Jesse P. Greenstein, The Effect of Diet on Growth and Nitrogen Balance after Injury, Surgery 48, No. 1 (October 1960), pp. 687-
2. Killing a Mockingbird (1)
While my collaborative study with Radwin was proceeding, I became friends with two US PHS Commissioned Officers (2) Dr. Vincent E. Price and Robert E. Greenfield, members of the Enzymes and Metabolism Section, who occupied the labs next to mine. Through them I became familiar with the problems and literature on the tumor-host relations, and particularly on anemia (3) and cachexia (4) that accompany cancer.
They carried out metabolic studies on etiology and nature of anemia of tumor-bearing animals. Their studies demonstrated that the anemia of rats with various tumors could be quantitatively correlated with the loss of cells by hemorrhage into the area of the tumor. I became very much interested in this work and began investigating the nutritional aspects. In the course of this work, it was observed that rats with marked anemia lost weight more rapidly, became cachectic and died while the tumors were in the range of 20-30% of the body weight, whereas in animals without anemia the onset of weight loss and cachexia was much slower, and the tumor frequently reached 50-65% of the body weight. However, in further study on the relationship of the anemia to the cachectic process, it was observed that during the period of most rapid tumor growth, and despite the onset of anemia, rats bearing the lymphosarcoma gained appreciably more weight that the normal controls although they both had almost identical food consumption.
I then went on to study the nature of the body weight gain of the tumor-bearing animals as compared to normal controls, by utilizing the carcass and tumor analysis, and compared the efficiency with which the two groups of animals utilize their diet by the use of metabolic studies.
During the period of most rapid tumor growth, rats bearing lymphosarcoma (5) gained more body weight than the control animals although their food consumption was almost identical. Rats bearing hepatoma, on the other hand, consumed less food and grew at equal or slower rate than the controls. Both types of tumors, however, exerted similar effects on the host with regard to the development of anemia, elevation of plasma aldolase, depression of liver catalase, adrenal enlargement, thymus involution and changes in weight of other organs. Carcass and tumor analyses showed that the lymphosarcoma-bearing animals had an increased percentage of water sufficient to explain their increased weight gain. An equally high degree of hydration was observed in rats bearing hepatoma whose body weight gains were lower than those of the normal controls. The combined nitrogen content of the tumor and the carcass of both groups of cancerous rats was similar to, and smaller than, the body nitrogen of the control animals, while the fat content of the tumor-bearing rats was markedly reduced. In confirmation of these findings, balance studies showed nearly the same retention of nitrogen in rats bearing the lymphosarcoma as in the controls, but there was a greater retention of water and sodium, concomitant with the increased weight gains in the tumor-bearing animals.
During the terminal stages the lymphosarcoma-bearing rats were observed to suffer from anorexia (6) loss in total body weight and finally died. In this period the animals were in negative nitrogen and potassium balance, the water retention was reduced below the normal level while the sodium balance together was not appreciably affected. The positive sodium balance, together with the decrease in the potassium retention, indicates the possibility that, to a considerable extent, the terminal dehydration of the tumor- bearing animals was due to loss of intracellular fluid.
The onset of anorexia and the negative nitrogen balance was not prevented by increasing the sodium chloride intake of the animals. Moreover, in other experiments in which the rats were given a choice of water and saline, neither the tumor-bearing rats, nor the control animals, showed preference for fluid containing sodium chloride. A substitution of physiological saline for water in rats bearing lymphosarcoma resulted in greater fluid consumption, greater fluid retention and greater body weight gains than was observed with water ingestion. Consumption of saline did not cause any improvement in the nitrogen retention and in fact it resulted in greater loss of urinary nitrogen in both normal and lymphosarcoma-bearing rats when compared with the groups ingesting water. Ingestion of saline by the control and the hepatoma-bearing animals, while increasing the fluid consumption in both groups, had no apparent effect on the animals’ weight gain. Decrease in survival time among the tumor-bearing animals ingesting saline was also observed (7 )
When we first reported on our findings (8 ) it created quite a sensation at NIH, and the cancer community as a whole. At that time several cancer investigators, particularly the associates of Dr. Julius White, Chief of the Laboratory of Physiology, were promoting the idea that tumor tissues contain a specific growth factor that stimulates appetite and growth in tumor-bearing animals ( 9 ). Our findings, showing that the increased weight gain of tumor-bearing animals could be accounted for entirely by an increased water and sodium retention in the tumor mass and the carcass, puts the idea of some new growth factor in tumor tissues into jeopardy.
Endnotes
1. The title of Ernest Hemingway’s popular book. Here it is used in a metaphoric sense, comparable to the saying “killing a sacred cow.”
2. The United States Public Health Service Commissioned Corps (PHSCC) is the federal uniformed service of the United States Public Health Service (PHS) and is one of the seven uniformed services of the United States. They only consist of commissioned officers and have no enlisted men. Officers of the PHS are classified as noncombatants. Members of the PHSCC wear the same uniforms as the United States Navy with special corps insignia, and hold ranks equivalent to those of naval officers.
3. A condition in which the blood is deficient either in quantity or in quality.
4. A profound and debilitating poor state of general health and malnutrition.
5. A malignant neoplasm arising in lymphatic tissue from proliferation of atypical lymphocytes. ‘
6. Lack or loss of the appetite for food.
7. Miloslav Rechcigl, Jr., Flora Grantham and Robert E. Greenfield, Studies on the Cachexia of Tumor-bearing Animals. I. Body Weight Changes, Carcass Composition, and Metabolic Studies, Cancer Research 21, No. 2 (February 1961), pp. 238-251.
8. Rechcigl, Miloslav, Jr., and R. E. Greenfield, Studies on Nitrogen and Water Metabolism in Rats Bearing Lymphosarcoma R2788 and Hepatoma 3683, Proc. Am. Assoc. Cancer Res. 3 (1960), p. 144.
9. F. K. Millar, J. White, R. H. Brooks, and G. B. Mider, Walker Carcinoma 256 Tissues as a Dietary Constituent. I. Stimulation of of Appetite and Growth in the Tumor-bearing Rat, Journal of the National Cancer Institute 19 (1957), pp. 957-967.
.
3. In Pursuit of Toxohormone
One of the classical effects of tumor on its host is the production of lowered catalase (1) activity in the liver and kidneys. This phenomenon was actually discovered by the sponsor of my fellowship, Jesse P. Greenstein. (2) Later it was postulated that a substance, referred to as toxohormone,502 is produced by tumor cells which passes through circulation to the liver and kidneys causing reduction in catalase activity of these organs. If such a substance were produced in tumor cells, presumably it would also have a direct effect on a catalase present in these cells. This was consistent with the findings that tumor tissues, in general, have only a negligible level of the enzyme activity.
These observations, about which I learned soon after my arrival at NCI after reading Greenstein’s classical book, Biochemistry of Cancer (3), intrigued me very much. As a part of my work on the tumor-host relations and after I was introduced to enzyme kinetics, using catalase as a model, it was natural that I began exploring the validity of this hypothesis. Just about this time, it was reported that certain hepatomas (4) possess significant levels of several enzymes505 that were previously thought to be low or virtually absent in the tumor tissues. These reports prompted me to carry out a series of experiments on these hepatomas to see whether they also possess catalase.
Several rat and mouse hepatomas and the liver and kidneys of the host animals were thus assayed for catalase activity. Little or no activity could be demonstrated in the subcutaneously transplanted Novikoff Hepatoma, Hepatoma LC18 and Hepatoma 3685 of rats and in a number of transplanted hepatomas of mice and hamsters. In contrast, a relatively high activity was found, however, in the transplanted Hepatoma 5123 and the ethionine-induced hepatomas of rats, and in the spontaneous hepatomas of C3H mice, in which other enzymes were recently noted in significant amounts. Interestingly enough, the relatively high level of catalase existed despite the observed depression of catalase activity in the liver and kidneys of these tumor-bearing animals.
When the total catalase activities of host liver and the Hepatoma 5123 were plotted against tumor weight, there was a progressive decrease in total liver catalase activity with increasing tumor size, in contrast with the total tumor catalase activity which was increased. An even more striking example of the liver and kidney catalase depression in the presence of high tumor catalase levels was found in animals bearing the HC (“High Catalase”) Hepatoma. The catalase activity of the tumor was at times so high that it greatly exceeded the level observed in the normal liver and kidneys.
Another study was carried out to determine whether the site of tumor transplantation would affect the enzymatic activity of the Hepatoma 5123. There was no apparent effect of site of implantation, since the subcutaneous and intraperitoneal implants grown in the same animal possessed approximately the same levels of enzyme activity. Single subcutaneous or intraperitoneal inoculations in different animals gave similar results. Moreover, the catalase activity of these tumors was not affected by the sex of the host animal. A similar stability in the tumor catalase activity was observed in male and female host animals of varying ages. On the other hand, both sex and age significantly altered the level of liver catalase activity. In the case of the kidneys, the catalase activity was affected by the sex of the animal, whereas the age of the animal was without apparent influence.
These findings were not consistent with the previously held view that tumor tissues, in general, contain no or negligible amounts of catalase. The enzymatic assays of the liver and kidneys indicate that a depression can be observed in the catalase activity of the tissues of the tumor-bearing host, even in the presence of significant levels of catalase activity in tumors. These observations throw a serious doubt on the concept of toxohormone, since it seems unlikely that such a factor released from tumors would depress the enzyme level of the liver and kidneys without markedly affecting the catalase activity of the tumor itself.
There remains a possibility, however, that toxohormone is not released in an active state within the tumor cell and that this might account for its failure to completely depress the catalase activity of certain hepatomas. Another possibility is that there may be two or more biologically, if not physically, distinct species of a catalase within the liver. It is possible that during carcinogenesis one form of catalase may be retained in some hepatomas but not in others, and this form may be inaccessible to toxohormone and therefore remain active. This would be consistent with my findings showing the relative stability of a tumor catalase toward the effects of the host, i.e., sex and age of the host, tumor site, and the lack of the response to a protein deficiency.506
Endnotes
1. Miloslav Rechcigl, Jr., Flora Grantham and Robert E. Greenfield, Studies on the Cachexia of Tumor-bearing Animals. I. Body Weight Changes, Carcass Composition, and Metabolic Studies, Cancer Research 21, No. 2 (February 1961), pp. 238-251.
2. Rechcigl, Miloslav, Jr., and R. E. Greenfield, Studies on Nitrogen and Water Metabolism in Rats Bearing Lymphosarcoma R2788 and Hepatoma 3683, Proc. Am. Assoc. Cancer Res. 3 (1960), p. 144.
3. F. K. Millar, J. White, R. H. Brooks, and G. B. Mider, Walker Carcinoma 256 Tissues as a Dietary Constituent. I. Stimulation of Appetite and Growth in the Tumor-bearing Rat, Journal of the National Cancer Institute 19 (1957), pp. 957-967.
4. The enzyme which catalyzes the decomposition of hydrogen peroxide into water and oxygen. The enzyme, in turn, is any of numerous proteinaceous substances that are produced by living cells and bring about or accelerate reactions without themselves undergoing destruction in the process.
5. J. P. Greenstein and H. B. Andervont, Catalase Activity of Tumor-bearing Mice and the Effect of Spontaneous Regression and Removal of Certain Tumors. J. National Cancer Institute 2 (1942), pp. 345-353.
6. W. Nakahara and F. A. Fukuoka, Toxic Cancer Tissue Constituent as Evidenced by its Effect on Liver Catalase Activity, Japanese Medical Journal 1 (1948), 271-277.
7. Jesse P. Greenstein, Biochemistry of Cancer, op. cit.
8. Hepatomas is a usually malignant tumor occurring in the liver.
9. Any of a group of complex proteins or conjugated proteins that are produced by living
cells and act as catalysts in specific biochemical reaction.
10. Miloslav Rechcigl, Jr., Vincent E. Price, and Harold P. Morris, Studies on the
Cachexia of Tumor bearing Animals. II. Catalase Activity in the Tissues of Hepatoma-bearing Animals. Cancer Research 22, No. 7 (August 1962), pp. 874-880.
4. Working with Slow-Growing Hepatomas
As has been initially reported by Greenstein, it has been generally thought that tumors contain no or negligible level of enzymes that are normally present in the liver. This conclusion underwent a drastic revision when Pitot et al.508 reported the presence of enzymes in several slow-growing hepatomas, particularly the Morris 5123 hepatoma. Since the source of these hepatomas was our own Laboratory of Biochemistry, this gave me an opportunity to ascertain whether or not the slow-growing hepatoma contains the enzyme catalase which was found lacking in all the previous work.
In confirmation with the earlier studies, little or no activity could be demonstrated in Dunning, Novikoff, and the 3683 Hepatomas. In mark contrast, however, high activity was found in the Morris 5123 Hepatoma, as well as in the ethionine-induced hepatomas of
rats, and in the spontaneous hepatomas of C3H mice. It was further observed that the catalase level of the 5123 Hepatoma is not altered under the stress of a protein-free diet which produces a marked lowering in the liver catalase of both normal and tumor-bearing animals.
In continuation of this work, I collaborated with Dr. Herschel Sidransky, NCI pathologist, to determine whether the catalase activity of a given tumor tissue is a fixed entity, using ethionine-induced hepatomas which he developed, The primary hepatoma and its subsequent transplants were transplanted subcutaneously by trocar into the backs of rats of the inbred strain of Osborne-Mendel, raised at NIH. At the time of transplantation the rats were killed by decapitation and exsanguination. All were autopsied, and one kidney and representative pieces of liver and of tumor were weighed and used for catalase assay. Care was taken to select viable, non-necrotic areas of tumor. Pieces of tumor were routinely fixed in Zenker-formol solution and occasionally in absolute alcohol and 10 percent formalin. Additional tissue was taken from any area that appeared abnormal. All sections were routinely stained with hematoxylin and cosin, and some were also treated with periodic acid-Schiff, Best’s carmine, and Oil red O stains.
The subcutaneous implantation of the primary ethionine-induced hepatoma in two weanling rats of the OM/N strain gave rise to single solid tumors in these animals. They were first observed around the 12th week after transplantation. The animals were killed approximately 18 weeks after tumor transplantation, at which time one tumor weighed 4.3 g and the other 7.3 g. Analysis revealed that the two tumors differed markedly in the catalase activity, the first possessing 30 units and the second 474 units per g of tissue as compared to 63 units found in the primary tumor. The value of 474 units was entirely unexpected in that it not only exceeded the activity of the primary tumor but also that of the normal liver. In fact, this was the highest figure ever reported for a rat in our laboratory. The tumors described were transplanted for a number of generations and the tumor transplants were assayed for their catalase activities. The catalase activity of both lines of hepatomas, established in the first transplant generation, decreased in the third and the later transplant generations, when compared with the first-generation transplants. However, the two lines of tumors, which I designated as HC and LC, with different catalase activity established in the first transplant generation remained distinct from each other throughout the subsequent transplantations. All HC tumors that arose from the high catalase transplant possessed high catalase activity, while those (LC) from the low catalase transplant had low catalase activity.
There was no apparent relationship between the tumor catalase activity of either line or the sex and age of the animals. Moreover, the extremely large tumors did not differ much from the small tumors in the catalase activity, which indicated that cachexia of the tumor-bearing host had no apparent effect on tumor catalase. Similarly, very young umors had levels of catalase activity comparable to those seen in other tumors, which indicated that the time of transplantation did not significantly influence the tumor catalase activity. On the other hand, the catalase activity of the host liver, and to a lesser extent the kidney, was influenced by these parameters.
With subsequent transplantations it became more and more apparent, however, that the low catalase tumors grew faster than those of high catalase line.
To get a more dramatic demonstration of the difference in the growth rate and the catalase activity of the two lines of hepatomas, equal amounts of tumor tissue from each line were transplanted in the same animal. Tumors arising from the LC transplants were 2 to 20 times larger than the HC tumors, while the catalase activity of the former group was 4 to 10 times lower than that of the latter.
To ascertain whether the low catalase activity observed in all tumors of the LC catalase line was due to the presence of some inhibitors of the enzyme, the homogenates of both tumor lines were mixed, and the mixtures were assayed for catalase activity. The resulting values were identical with the calculated average figure of the sum of individual determinations of the low and high catalase activities. These results speak against the presence of active inhibitors of catalase in the LC line.
In an attempt to find out whether the two established lines of hepatomas with widely different catalase activity would also differ in some other biochemical parameter, the tumors were analyzed for a number of common enzymes and tissue constituents. Our data showed that the two lines of hepatomas do not differ greatly with the parameters used. Actually, there were greater differences among the tumors within the same line than between the two lines. This was true for the nitrogen, water, lipids, glycogen, ribonucleic acid (RNA), and deoxyribonucleic acid (DNA), as well as for various enzymes, including glucose-6-phosphatase, amylase, tryptophan pyrrolase, arginase, aldolase, lactic dehydrogenase, glutamic-oxalacetic transaminase, D-amino acid oxidase, uricase, acid and alkaline ribonuclease, and choline oxidase.
From photographs of the primary tumor from which the two lines arose, it was evident that the liver contained multiple small and large gray nodules. On histologic examination, some of these nodules showed atypical hyperplastic changes, while others, especially the larger ones, showed changes diagnostic of hepatocellular carcinoma. Other areas of liver showed fatty change, cholangiofibrosis, and cysts.
Equal amounts of each line of tumor were initially inoculated subcutaneously into the back of a rat, the HC line on the left and the LC line on the right. The LC line of tumor grew much more rapidly than the HC line. Grossly, both lines appeared as grey to yellowishgray lobulated tumor masses with frequent red areas.
Tissues from all tumors transplanted were examined histologically. No differences that could be related to the two different levels of catalase activity were observed in any tumors of the HC and LC lines. The tumors of both lines resembled the primary tumor and were typical trabecular hepatocellular carcinomas.
This study509 demonstrated the establishment of two lines of hepatomas, with widely different catalase activity, which arose from the same primary tumor. The change occurred in the first-generation transplants and is apparently a constant one, since it persisted for over 10 generations.
Changes in the enzyme activity and the establishment of different cell lines have been reported by other researchers in the tissue-culture field. Our study was, however, the first report of such phenomenon among transplanted tumors.
The establishment of two lines of hepatomas with widely different catalase activity, which arose from the same primary tumor, suggests that primary tumors are not homogeneous but rather mixtures of cell populations that may change in size during subsequent transplantations. The data also emphasize that tumor transplants may differ biochemically, though they appear similar morphologically.
Since the change already occurred in the first-generation transplants and since most researchers in the cancer field use tumors established a number of years ago, it is likely that many of these tumors do not represent the original cell population.
The absence of low levels of catalase in many tumors, often reported in the literature, apparently are not typical of, but rather coincidental with, the neoplastic process.
To an enzymologist it is also of special interest to find that the two lines of tumors possessing such greatly different levels of catalase activity have similar levels of uricase and no demonstrable D-amino acid oxidase. The presence of these three enzymes was reported in microbodies. Considering that the mentioned enzymes are involved in the metabolism of hydrogen peroxide, these data would indicate that they behave independently of each other, even though they may be located in the same organelle.
Endnotes
1. J. P. Greenstein, Biochemistry of Cancer, op. cit. 508 Henry C. Pitot, “The Comparative Enzymology and Cell Origin of Rat Hepatomas. II. Glutamate Dehydrogenase, Choline Oxidase, and Glocusoe-6-phosphotase,” Cancer Res. 20 (1960), pp. 1262-126
5. Getting into Enzyme Kinetics
After the death of my sponsor, Jessie P. Greenstein, Vincent E. Price became my new sponsor. He was the head of Metabolism and Enzymes Section in the Laboratory of Biochemistry at NCI who introduced me into a new field of enzyme kinetics. Almost immediately I became his right-hand man on a joint project to develop simple techniques for measuring the rates of enzyme synthesis and destruction in vivo, the kinetics of these processes, and how they change during development, and how they are altered in disease states.
Although changes in the activity of various enzymes have been routinely studied in various physiological and pathological states, little attention has been given to the question whether the alteration in the activity of tissue enzymes is the result of the changes in their rate of synthesis or destruction. The demonstration that 3,amino-1,2,4- triazole (AT) irreversibly inactivates enzyme catalase in the liver without interfering with its re-synthesis, and allylisopropyl-acetamide (AIA) blocks synthesis of new catalase suggested that these compounds could be used as tools for studying the kinetics of catalase synthesis and destruction in vivo.
The basic techniques consisted in injecting a drug 3-amino-1,2,4-triazole (AT) in dose of 1 gram per kilogram intraperitoneally into a group of animals. At various periods thereafter the animals were killed, the liver homogenized in 49 volumes of cold water, and catalase assays were performed on the homogenates using a Cary Recording Spectrophotometer with a Log-absorbance attachment.
Initially, it was found that the rapid fall in catalase activity following the injection of AT resulted from destruction of the catalase, and that the return in catalase activity during the following 3-4 days resulted from the synthesis of new catalase. It was further shown by use of radioiron that administration of the AT produced no significant alteration in the rate of catalase synthesis.
Kinetic studies have been carried out on the rate of catalase return following administration of AT to rats. It has been shown that the curve could be described by an equation which allowed one to calculate the rate in which the enzyme is synthesized, as well as the rate at which it is destroyed (1).
When such equation was calculated for a particular breed of male rats (M520 x A x C), weighing approximately 230 gm., the rate of synthesis per unit time was found to be 4.54 units, while the comparable figure for the rate of destruction was 0.0214. This meant that 4.54 units or 26.6 micrograms of catalase were being synthesized each hour per gram of liver and 2.14% of the catalase molecules present at a given time were being destroyed during the next hour. With Sprague-Dawley male rats weighing approximately 210 grams the rate of catalase synthesis was 24.5 micrograms per hour in the liver and 6.6 micrograms per hour in the kidney, and 2.27% of the catalase present was being destroyed each hour in the liver and 2.13% in the kidney. This meant that the rate of synthesis per gram of tissue was 4 times as high in the liver as in the kidney, but the fraction of catalase destroyed per unit of time was nearly the same in both tissues. The kinetics indicated that newly-formed catalase molecules did not have a finite life span, but instead had the same risk of destruction in each unit of time as an older molecule. This was not unexpected since most data on the incorporation of carbon labeled- amino acids into proteins provided a similar interpretation.
The AT technique has some distinct advantages over other methods. Until then the principal technique by which the rate of protein synthesis and destruction could be estimated was by use of radioisotopes. This necessitated rather extensive fractionation at each experimental point and purifying the given protein to a constant specific activity. With the AT technique this has not been necessary, for one has only to prepare a homogenate for a catalase assay.
The AT technique is complicated somewhat by the fact that one is injecting a drug which does alter the body metabolism. Although radioiron studies indicated that the single injection of AT has little or no effect on the rate of catalase synthesis, the kinetics data did show that for a period of 7-12 hours, depending on the organ and strain of an animal, any catalase that was formed would be destroyed. Similar periods would presumably exist for tryptophan peroxidase, and with the “iodine peroxidase” in the thyroid this period may be somewhat longer because of the apparent greater sensitivity of this enzyme to AT. Nevertheless, the net effect of these changes on the rate of catalase synthesis in the liver appears to be negligible.
The above technique is of some significance. Although considerable attention has been given to the synthesis of proteins, relatively little is known about their destruction. In problems on cachexia in the tumor-bearing host it is important to develop a fund of knowledge about the destruction of proteins so as to understand the more intimate nature of the cachectic process. Since the reduction in catalase activity of the liver and kidney is one of the most prominent features of the tumor-bearing host, studies on the kinetics of synthesis and destruction of this enzyme is of particular importance in the cancer field.
Although these studies have been limited to a certain family of enzymes it is considered likely that other drugs could bind rather specifically to other enzymes so as to eliminate their activity rather abruptly. It is therefore thought possible that similar studies could be extended to these enzymes.
Endnotes
1. V. E. Price, M. Rechcigl, Jr. and R.W. Hartley, Jr., Methods for Determining the Rates of Catalase Synthesis and Destruction, in Vivo, Nature 189 (1961), pp. 62-63; Vincent E. Price, William R. Sterling, Vincent A. Tarantola, Robert W. Hartley, Jr. and Miloslav Rechcigl, Jr., The Kinetics of Catalase Synthesis and Destruction in Vivo, J. Biol. Chem. 237 (1962), pp. 3468-3475.
6. Regulation of Enzyme Synthesis and Degradation
In starvation and in the cachexia of the tumor-bearing host, as well as in many nutritional deficiencies, there is a marked reduction in many of the proteins which make up the various organs and tissues. Biochemists have long been intrigued by these observations but have lacked specific tools by which to pursue the question of the mechanism by which such reduction or a decrease rate of synthesis, of the respective proteins occurs.
The following studies were conducted in an attempt to throw some light on this question, utilizing albeit simple yet unique methodology (1). The techniques permit determination f the rates of synthesis and degradation of a single protein, the enzyme catalase in the living animal. They are based on the use of two drugs, 3-amino-1,2,4- triazole (AT), which acts by destroying catalase without interfering with its resynthesis, and allylisopropylacetamide (AIA), which acts by blocking catalase synthesis without interfering with the activity of the previously formed enzyme.
Normal Physiological Conditions - The level of catalase activity of the liver is about 3.5 times that found in the kidney. In order to compare the relative rates of catalase synthesis and destruction in the two organs, a number of male rats weighing 200 gm, were injected with AT and equal number of rats were used as controls at intervals for 4 days. Five animals from each group were killed and assays were performed on the liver and kidney. The AT values were then subtracted from the normal value for that day and the data were plotted semilogarithmically. During each hour 4.18 units of catalase were being synthesized per gram of liver as compared to 1.18 units in the kidney. During the same l hour period 2.25 and 2.27% of the catalase molecules present were destroyed. It is of interest that the rate constant for catalase destruction is almost identical in the liver and kidneys, so that in both tissues the same constant fraction of the catalase molecules are destroyed each hour, although the rate of synthesis in the liver is nearly four times that of the kidney.
In a similar study on the effect of sex, the calculated rate constant for catalase decay in the liver of young male rats did not differ significantly from that of young females. Similarly, the higher levels of the enzyme activity in the kidney of male rats, when compared to females, were found to be due to different rates of enzyme synthesis rather than to those of its breakdown.
Comparable results were obtained with adult rats. Interestingly enough, the hepatic catalase of the adult animals underwent a greater decay when compared to the younger rats, while the reverse was found in the case of renal enzyme.
Effect of Starvation – At the onset of the second decade of the 20th century it was reported that prolonged intermittent starvation led to a marked reduction in the tissue catalase activity. Some thirty years later Greenstein et al. found that a 3-day starvation had no apparent effect on the rat hepatic catalase activity. And yet, in 1948, Miller observed a significant decrease in the enzyme level following a 7-day fast. Several other studies, in turn, reported an actual increase in the enzyme activity.
In order to shed some light on these conflicting results, I decided to revisit this problem. In addition to methodically following the time course of changes which occur in the levels of catalase activity in the liver and the kidneys during the course of starvation, attempts were made to measure the turnover of the enzyme and to evaluate the relative role of its synthesis and degradation. Use has been made of the kinetic techniques developed in our laboratory which allow the determination of the rate constants of the latter under in vivo conditions (2).
The animals were divided into four groups. Two of the groups were injected with 3- amino-1,2,4-triazole (AT) and the two remaining groups were kept as controls. Eighteen hours after AT, enzymic assay showed that the expected rate of synthesis was occurring, at which time food was removed from one of the control and one of the AT groups. At various points during the experiments 5 rats in each of the four groups were killed and their livers and kidneys analyzed for catalase activity. Starvation produced progressive loss in the weight of the rats, irrespective of the presence or absence of AT. The body weight changes were reflected in the reduced weight of the tissues. While the alterations in the weights of the kidneys were proportional to the changes of the total weights of the animals, the diminution of the hepatic mass was far greater than that which was elicited in the carcass weights.
The total liver catalase fell off sharply soon after starvation was begun and there was actually less total enzyme being generated within the liver. When the data were expressed in terms of units per gram of tissue, an entirely different situation presented itself. Thus, during the first five days of starvation, in the AT animals, the starved groups had nearly the same rates of enzyme synthesis and destruction as those observed in the control animals. During these first five days of starvation, therefore, the rate of catalase synthesis per gram of liver proceeded as rapidly as in the controls; however, since the livers were becoming gradually smaller in size there was less total enzyme formed.
In the control animals, which were starved, there was, in fact, some elevation in the hepatic catalase activity per gram of tissue during the period from 24 to 48 hours, suggesting that as the liver shrinks in size the catalase concentration within the cell rises until the enzyme can be destroyed. With a heightened enzyme concentration, the amount of catalase dissipated per unit tissue could magnify although the fraction of enzyme molecules degraded during that time remained constant. After five days, the animal entered a second phase, in which the enzyme activity per gram of the organ commenced to descend and this endured onward until death of the animal on the 7th to 10th day.
When the data were expressed in terms of units of the enzyme activity per mg. nitrogen, the rate of synthesis of catalase in the starved animals seemed to progress at lower pace when compared to the controls, already during the first stage of fasting.
In order to get an estimate of the synthetic rate of the enzyme during prolonged starvation, another kinetic study was conducted, using pre-fasted animals at the point of their entry into the second stage, referred to above. For this trial, rats that were subjected to four days of starvation, previously, were divided into four groups. One group of animals was injected with a single dose of AT; the second groups received AIA twice daily during the entire experiment; the third group was used for controls.
By plotting semilogarithmically the rate of which the AT curve approaches the normal level of the group given both AT and AIA, straight lines were obtained, from which the first order constants for catalase destruction can be determined. The inspection of the data disclosed that during one-hour period only 2.6 units of catalase were being formed per gram of tissue on prolonged starvation which can be contrasted with the figure of 4.3 units in the case of normal controls. These results substantiate the earlier conclusions that the rates of synthesis are indeed reduced after the animal enters the second phase of starvation. The rate constant of the enzyme degradation, on the other hand, remained unaltered during the course of the entire starvation period.
The time course of inflections in the renal catalase activity, generally speaking, brought forth comparable results to the hepatic data, except that they were of lesser magnitude. The analysis of the renal kinetic data revealed one salient disclosure in that the severe starvation, which elicits a rigorous repression in the synthesis of hepatic catalase, has no such effect on the renal enzyme.
Results of this study offered explanation for the seemingly conflicting data of earlier investigators, some having reported a decrease, while others showed an increase in the activity of catalase as a consequence of fasting. The examination of the time course of changes in the catalase levels following starvation reveals a two-stage pattern on the enzyme behavior. During the first stage the total activity fell progressively with the weights of the organs, yet the rate of its synthesis, as well as of its breakdown, remained essentially unaltered. In the second stage, however, the rate of the enzyme formation declined until death ensued.
Effect of Protein-free Diet - The above study led us also to reexamine the biochemical changes that take place during protein deficiency, which also include the lowering of the enzyme catalase. Specifically, we wanted to learn whether the observed reduction in the enzyme activity was evoked by an increased breakdown of the enzyme, or by a decreased synthetic rate of the latter. The basic design of the experiments was patterned after the previous work on the enzyme kinetics during inanition.
The results 513 showed that liver weights of the protein-depleted rats reflected, in general, the overall changes in the weights of the animals. With respect to catalase activity, a marked drop was found in the total level of the hepatic enzyme in animals which were depleted of protein. Furthermore, only a limited return of catalase occurred in those animals which were pre-treated with 3-amino-1,2,4-triazole. The level of enzyme reached a shallow plateau in the course of two days, after which no further changes occurred.
Essentially the same results were obtained when the data were expressed per gram of tissue. This was not at all surprising since the reduction in the weight of the liver was relatively small, reflecting, for the most part, the changes in the animals’ body weight.
In order to arrive at an estimate of the rate of the catalase synthesis and degradation during protein deficiency, we carried out another kinetic study, in which animals were previously deprived of protein for a period of 4 days, making use of a combination of the two drugs,3-amino-1,2,4-triazole (AT) and allylisopropylacetamide (AIA). There was a typical lowering of catalase activity following the injection of AIA, as a result of the stoppage of the enzyme synthesis, representing catalase breakdown. The return of catalase to the low plateau after treatment with AT was comparable to the earlier experiment.
The results of this kinetic study provided sound evidence for the conclusion that the depression in the enzyme level of rats fed a diet devoid of protein was a consequence of a decreased rate of synthesis. The rate of degradation, which is first order in respect to the enzyme present, interestingly, was unaffected by the deficiency.
Effect of Carcinogen - In order to see how prevalent the changes in enzyme synthesis were in comparison with the seemingly little effect on the rates of enzyme degradation, we have conducted a study using the glucoside cycasin, a potent carcinogen, isolated from endosperm of Cycas circinalis L, which evokes a conspicuous reduction in catalase activity.514
From kinetic studies on the rate of return of catalase activity after administration of 3-amino-1.2.4-triazole, it was calculated that in the cycasin-fed animals only approximately 1.2 units of hepatic catalase were synthesized per hour as compared with 5.0 units in pair-fed controls. The rates of destruction, which are first order in respect to the catalase present, were also lower in cycasin-fed rats than in the controls, i.e., 1.4% and 2.9% per hour respectively, which were less pronounced than the changes observed in the rates of enzyme synthesis.
In the case of the kidneys, which possess about one-fourth to one-third of the catalase activity found in the liver, the level of the enzyme was not affected by the cycasin feeding. Yet, the kinetic studies had shown that there was less than one unit of the enzyme synthesized per hour in the cycasin-treated rats, while in the control animals 1.6 units were formed. The rates of destruction in the renal enzyme of the respective groups were 1.8% and 3.3% in the same period. This was believed to be the first demonstration of the alteration of both the rates of enzyme synthesis and degradation by a carcinogen.
Effect of Neoplasia - Similar studies were initiated in certain rat hepatomas possessing unusually high levels of catalase activity. Contrary to the situation in normal tissues, the catalase level of these hepatomas was neither affected by cachexia in the tumor-bearing host nor altered under the stress of a protein-free diet.515
Studies on catalase synthesis and destruction were undertaken with rats bearing Morris 5123 Hepatoma. On administration of AT, the tumor catalase activity fell to a minimal value as in the liver of the host animal. Over the next few days, the catalase activities then rose progressively toward the expected values.
On the other hand, the lowering of tumor catalase activity following the daily injection of AIA, was small when compared with that observed in the liver. In animals which received both drugs, the catalase activity was not reduced to the low plateau seen in the liver, showing that AIA, has less effect on catalase synthesis in the tumor than in the liver. However, the absolute amount of the AIA-resistant catalase in the hepatoma was almost the same as in the liver.
In subsequent studies, the daily dosage of AIA was increased two-fold by injection twice daily, under which conditions a greater response of tumor catalase to the drug was observed, although it was still smaller than that observed in the liver. In the case of the liver, on the other hand, no further reduction in the level of catalase activity was observed by increasing the AIA dosage.
On the basis of the AT data, it was calculated that the rate of catalase synthesis was about four times that observed in the tumor. The first order rate constant for catalase destruction, however, was practically the same in both types of tissues.
These studies provide a conclusive proof for the existence of protein turnover in neoplastic tissues. Although there is some evidence for the overall protein turnover in cancerous tissues, our studies are believed to be the first demonstration of such regulatory process, using a single protein.
Endnotes
1. M. Rechcigl, Jr. and V. E. Price, The Rates and the Kinetics of Enzyme Formation and Destruction in the Living Animal, in: Albanese, Newer Methods of Nutritional Biochemistry. New York: Academic Press, 1963, pp. 185-197; M. Rechcigl, Jr., Relative Role of Synthesis and Degradation in the Regulation of Catalase Activity, in: Regulatory Mechanisms for Protein Synthesis in Mammalian Cells. New York: Academic Press, 1968, pp. 399-455.
2. M. Rechcigl, Jr. Studies on Regulatory Aspects of Nutrition at Molecular Level. I. Enzyme Turnover during Fasting, Int. Arch. Physiol. Biochem. 76 (1968), pp. 693-706.
3. M. Rechcigl, Jr., Studies on Regulatory Aspects of Nutrition at Molecular Level. II. Enzyme Turnover during Protein Deficiency, Nutr. Diet 1 (1969), pp. 214-227. 514 M. Rechcigl, Jr. and G. L. Laqueur, Carcinogen-Mediated Alteration of the Rates Enzyme Synthesis and Degradation, Enzym. Biol Clin. 9 (1968), pp. 276-286.
4.. M. Rechcigl, Jr. and V. E. Price, Studies on the Turnover of Catalase in Vivo. Progr. Exp. Tumor Res. 10 (1968), pp. 112-132; M. Rechcigl, Jr., The Roles of Synthesis and Degradation in the Regulation of Catalase Levels in the Neoplastic Tissues, Enzym. Biol. Chem. 10 (1969), pp. 161-80.
7. Genetic Control of Enzyme Activity
My interest in the role of genetics as a determinant of an enzyme activity stems from an accidental observation of a high liver catalase activity in one of the substrains of C57BL mice which have been known to possess a relatively low level of the enzyme activity. This led me to a systematic survey of catalase activity in mice in collaboration with Walter E. Heston, a prominent geneticist, who held the position of Chief of Laboratory of Biology in NCI.
Tissue Catalase Activity in Various Strains and Substrains of Mice - Variation between inbred strains of mice in respect to liver catalase activity was originally reported by Jessie P. Greenstein in 1942 (1). He noted that the catalase activity in the liver of mice of strain C57BL was only approximately half that of mice of other strains studies. It was, therefore, surprising when in a survey of a number of strains we noted that the activity in the livers of mice of Dr. Heston’s substrain (C57BL/He) was at a high level equal to that of mice of other strains (2).
We then investigated substrain C57BL/6 and found that the results for this substrain confirmed Greenstein’s earlier observation in that the level of liver catalase was approximately that in C57BL/He and in the other inbred strains.
These observations prompted a more thorough survey of the liver and kidney catalase activity of various C57 strains and substrains and of strain C58, all of which belong to the family of strains derived from the original Lathrop stock. We found that these strains and substrains did not vary in respect to kidney catalase activity. All C57 substrains, as well the strain C58, were low in liver catalase activity except for C57BL/An and C57BL/He, which were high. Because C57BL/He had been derived from C57BL/An which had come from C57BL/Sp it was postulated, under the assumption of a major gene affecting liver catalase activity, that a mutation had occurred in the progenitors of the C57BL/An substrain.
Genetic Analysis of Liver Catalase - A thorough genetic analysis of the variation between substrains C57BL/He and C57BL/6 was, therefore, undertaken. Reciprocal crosses were made between these parental substrains to produce the reciprocal F1s (3). These F1s were, in turn, mated to produce reciprocal F2s (4) and were also backcrossed (5) to the parental substrains to produce reciprocal backcrosses. In addition, progeny tests were made of C57BL/He - first backcross segregant males by mating them with C57BL/He females.
Approximately fifty males and fifty females of the two parental substrains, the reciprocal F1, and first backcross groups were produced for measuring catalase activity. Somewhat greater numbers were produced in the F2 groups. The animals were weaned and individually marked at 4 weeks of age, segregated as to sex and set aside in plastic cages with eight mice to the cage. Throughout the study the animals were fed NCI pellets. Determination of liver and kidney catalase was made on all mice at 6 weeks of age.
There was no variation between the two parental substrains and the hybrid groups in level of kidney catalase activity. Throughout the groups, however, there was a very striking uniform sex difference as had been noted previously, the level for the females being only about half that of the males. A sex difference in level of liver catalase was also noted in all groups, although not nearly as great as the sex differences in kidney catalase activity.
The difference between the two substrains in respect to liver catalase was clearly shown. The means for the male and female groups of substrains C57BL/6 were approximately half the means for the male and female groups of substrain C57BL/He and there was no overlap between the two substrains.
Dominance of low liver catalase activity over high was evident in that the F1 groups were in the range of the C57BL/6 groups and did not overlap with the C57BL/He groups. There was no evidence of any maternal influence for the reciprocal F1s were alike.
Evidence of segregation was apparent in that there was much greater variation in the F2 groups than in the F1 or parental groups. Furthermore, the distribution in the F2 groups was distinctly bimodal, indicating segregation of single pair of genes. The distribution within the groups resulting from backcrossing the F1s reciprocally to the C57BL/6 was unimodal and approximately within the range of two C57BL/6 and F1 groups, confirming the dominance of low-level activity. The variation within the groups resulting from the backcrossing the F1s reciprocally to the C57BL/He parental substrain like that in the F2 groups was greater than that in either parental substrain or the F1s and the distribution within each of these backcross groups was bimodal, also suggesting that but a single pair of genes was involved.
The cutoff values for high and low level of liver catalase activity for the F2 generation and the recessive-parent backcross were determined by rather a complicated posterior odds method for univariate normal classification which I won’t go into detail here. The analysis was carried out in two steps; first, for each catalase activity value the likelihood that activity value belongs to each genotypic group separately was computed. Then for each activity value a likelihood ratio of the predictive densities was computed, and the cutoff value for high and low catalase activity was set where the likelihood ratios of the activity values were maximum for each genotypic group. The predictive density was then combined with the a priori probability of segregating class in order to obtain the probability that the activity value belonged to a specific genotypic group and segregating class.
Confirming single factor inheritance with low dominant to high are the results of the progeny tests of the seventeen C57BL/He first backcross segregant males. With this pattern of inheritance the offspring of these first backcross males that were in turn backcrossed to C57BL/He females should identify these males belonging to two groups, homozygous (6) 521 high males that should have high offspring only and heterozygous (7) 522 low males that should have offspring half of which were low and half of which were high. This is indeed what we found. Five males had nothing but offspring with high liver catalase activity, whereas approximately half the offspring of each of the remaining twelve were high and half low. The total number of offspring of the twelve males was 230 and of these 117 were high and 113 low, a very close approximation of the expected 1:1 ratio.
Confirmation of the genotypes (8) of the seventeen males, as determined by the progeny tests, was later obtained when they were killed, and their livers were analyzed for catalase activity. The five whose breeding tests had indicated that they were homozygous high all showed a high level of liver catalase activity, whereas the twelve indicated as heterozygous low all had low liver catalase activity.
The data of this study showed and confirmed that the low level of liver catalase activity in the mouse is controlled by a single pair of genes with low dominant to high. We proposed Ce for the gene symbol. Thus, C57BL/6 would be CeCe, C57BL/HE cece, and the F1 hybrid Cece (9).
In considering the possible mode of action of this gene one notes several points that suggest that the gene does not act directly in the synthesis of catalase. First, the presence of the gene reduces catalase activity to one-half the normal amount rather than eliminating activity; the homozygous and heterozygous low both being expressed at this level. Second, low is dominant to high in contrast with genetic control over direct synthesis of enzyme. Third, the effect of the gene is apparently limited to the liver. Thus, the gene must be providing a regulator of the amount of catalase activity in the liver. This is in sharp contrast with acatalasemia (10) in the Japanese where the gene is apparently involved directly in the synthesis of catalase, for homozygous recessive individuals show no evidence of catalase.
Inasmuch as most strains have a high level of liver catalase activity and the wild mice, we have tested likewise had a high activity, the recessive high should be considered the wild type. The original mutation then would have been a dominant mutation to the low level that appeared in the original Lathrop stock from which the C57 and C58 lines arose. Since all of these lines were low it is probable that the original females number 57 and 58 and their littermate male number 52 that were mated in 1921 were homozygous for the mutation. If not, the mutation would have been fixed in the homozygous state through the subsequent years if brother x sister inbreeding. A reverse mutation back to the recessive wild-type must then have occurred in the immediate progenitors of Andervont’s subline of strain C57BL, and by selection the homozygous recessive was established in Andervont’s C57BL/An, and in the C57BL/He substrain derived from it.
Linkage studies to locate the gene would be the next appropriate step. With biochemical genetics in the mouse developing so rapidly this gene should provide an excellent biochemical marker.
Genetic Regulation of Enzyme Activity - As demonstrated above, the original mutation had been to the dominant gene for low level catalase activity. There is a question concerning the mechanism by which the mutant gene brings about the observed reduction in the enzyme activity. We took advantage of the recently developed techniques in our laboratory, using two drugs: 3-amino-1,2,4-triazole (AT) and allylisopropylacetamide (AIA) to determine the kinetics of catalase synthesis and destruction in the two C57BL substrains (11).
Male C57BL/HE and C57BL.6 mice, 6 weeks of age, were used. Half of the animals of both substrains were injected intraperitoneally with AT, the others being kept as un injected controls. At various points thereafter 5 animals each of the control and the AT treated groups were killed and their liver catalase activity determined.
After reaching the minimal op point, within 2 hours after the injection of AT, catalase activity started to return, reaching a maximum rate at about 12 hours. Initially the rates of catalase return were almost identical on the two C57BL substrains. Following the 22- hour point, however, the catalase return of C57BL/6 mice began to slow down, reaching the maximal value of 92 units/ gram, approximately 3 days after the injection of AT. In C57BL/He mice, on the other hand, the initial rapid rate of catalase return remained uninterrupted, for at least 36 hours after AT. From this point on the catalase activity mounted at somewhat slower pace until the maximal value of 202 units/gram was reached, 5 days after the administration of AT.
Using the previously established techniques, it was calculated that in C57BL/HE substrain there were 3.60 units liver catalase synthesized per hour as compared with 3.79 units in CF57BL/6 substrain. The calculated rate of constant for catalase destruction in the former group was 1.95% per hour, while in the latter group it was 4.5%. The alteration in the rate of destruction is a unique funding since in all previous experiments in our laboratory under various physiological and pathological conditions, the rates of catalase destruction seemed to be almost constant.
Although alteration in the rate of enzyme degradation had been observed under certain physiological conditions with other enzyme systems, our studies are believed to be the first demonstration of such regulatory mechanism under genetic control.
Erythrocyte Catalase - We had also conducted a separate study to determine the erythrocyte5 (12) catalase activity for inbred mice and to determine if a relationship exists between erythrocyte, liver, and kidney catalase levels (13). The 18 inbred strains of mice used were chosen to represent the major lines of independent origin and subsequent crossings between these lines. Blood samples were obtained from the orbital sinus from mice 6 weeks of age. Erythrocytes were washed with cold phosphate-buffered saline and packed erythrocytes were lysed in ice-cold distilled water. Catalase activity was determined spectrophotometrically. In strains BALB/c, C57BL/An, C57/He, C57L, C58, and DD the difference between the male and female erythrocyte catalase activity was statistically significant. There did not appear to be a relationship between the difference in male and female enzyme levels and the magnitude of enzyme activity, the 6 strains which showed significant sex differences were equally distributed throughout the observed range of enzyme activities.
In strains with a catalase activity above 25 U/ml RBC females had higher enzyme levels than males, while in strains with levels less than 25 U/ml RBC males had higher enzyme levels than females.
A perfect correlation was found between the erythrocyte catalase activity of inbred strains of independent origin and the erythrocyte catalase activity of the strains and substrains derived from them. An analysis of variance of the erythrocyte catalase activities of the 18 strains indicated that the enzyme levels were significant. Thus, the groupings of related strains close to the diagonal line above the parent strains indicated a major genetic component for the variation observed between the inbred strains. C57 BL strains exhibited 50% of the enzyme activity of strains A. ARK and BALB/c, and 70% of the enzyme activity of C57BR.
The quantitative levels of kidney catalase activity of strains BALB/c, C3Hf, YBR, C57BL/An, C57BL/He, C57BL/6, C57BL/10 and NBL fell into a distribution similar to that for erythrocyte catalase levels with the exception of that in NBL the activity was the same as in the other C57BL sublines. The liver catalase activity in the above strains followed a similar distribution with the exception that C57BL/An and C57BL/He exhibited higher enzyme activities than the other C57BL sublines.
The data on erythrocyte, liver, and kidney catalase activity demonstrate that the enzyme activity levels are specific for each inbred strain and that so-called normal mouse level does not exist. The strain, sex, and age of animal should be included in the quantitative estimate of enzyme activity.
Endnotes
1. J. P. Greenstein and H. B. Andervont, The Liver Catalase Activity of Tumor-bearing Mice
and the Effect of Spontaneous Regression and of Removal of Certain Tumors, Journal of the National Cancer Institute 2 (1942), pp. 345-355.
2. Miloslav Rechcigl, Jr. and W E. Heston, Tissue Catalase Activity in Several C57BL Substrains and in Oher Strains of Inbred Mice, Journal of the National Cancer Institute 30 (1963), pp. 855-864
3. F1 hybrid is a term used in genetics and selective breeding. F1 stands for Filial 1, the first filial generation seeds/plants or animal offspring resulting from a cross mating of distinctly different parental types. The term is sometimes written with a subscript, as F1 hybrid. The offspring of distinctly different parental types produce a new, uniform variety with specific characteristics from either or both parents.
4. The F2 generation is the result of a cross between two F1 individuals (from F1 generation).
5. Backcrossing is a crossing of a hybrid with one of its parents or an individual genetically similar to its parent, in order to achieve offspring with a genetic identity which is closer to that of the parent. It is used in horticulture, animal breeding and in production of gene knockout organisms.
7. Homozygous refers to having identical alleles for a single trait.
7. Heterozygous refers to having two different alleles for a single trait.
8 The genotype is the genetic makeup of a cell, an organism, or an individual (i.e. the specific allele makeup of the individual) usually with reference to a specific character under consideration.
9. W. E. Heston, Harold A. Hoffman and Miloslav Rechcigl, Jr., Genetic Analysis of Liver Catalase Activity in wo Substrains of C57BL Mice, Genetic Research 6 (1965), pp. 387-397.
10 Hereditary deficiency of catalase in the blood, often manifested by recurrent infection or ulceration of the gums and related oral structures.
11.M. Rechcigl, Jr. and W. E. Heston, Genetic Regulation of Enzyme Activity in Mammalian System by the Alteration of the Rates of Enzyme Degradation, Biochem. Biophys. Res. Comm. 27, No. 2 (1967), pp. 119-124.
12. A red blood cell (RBC).
13. H. A. Hoffman and M. Rechcigl, Jr., Erythrocyte Catalase in Inbred Mice, Enzyme 12 (1971), pp. 219-225.
8. Enzymatic Changes in Acute Porphyria
It must have been some time in 1961 when I became first acquainted with Donald P. Tschudy, Senior Investigator in the Metabolism Service of the National Cancer Institute. He was a physician with training and interests in biochemistry. Because of mutual interests we became good friends and scientific collaborators, as well. Several papers came out of this, mostly in the area of experimental porphyria (1).
Acute intermittent porphyria (AIP) is a well-recognized clinical disorder characterized chemically by the excretion of increased amounts of porphobilinogen (PBG) and deltaaminolevulinic acid (ALA) in the urine, and clinically by neurologic dysfunction. A metabolic disorder resembling the human disease was produced in animals by the administration of Sedormid and later by the more soluble analogue allylisopropylacematide (AIA) (2). Administration of these agents is followed by the urinary excretion of increased amounts of porphyrin precursors.
Since the enzymatic defect in both human and experimental porphyria was unknown, we decided to survey various enzymes and metabolic pathways related to the biosynthesis and utilization of porphyrin precursors in the livers of porphyric (AIA-induced) animals (3).
There are several possible defects in metabolic pathways that could result in increased production or decreased utilization of ALA and PBG. One such lesion could be caused by a block beyond PBG in the biosynthetic pathway of heme. The increase in free porphyrins in the liver and the marked diminution in liver catalase in experimental porphyria could result from a decrease in the synthesis of its heme component as a result of a block beyond PBG in the porphyrin biosynthetic pathway. Previously obtained evidence against this hypothesis is the fact that glycine appeared to be incorporated into the heme moiety of other heme containing enzymes in porphyric liver to about the same extent as in controls, and Fe59 was incorporated into heme of porphyric liver to a greater extent than in controls.
In order to obtain additional evidence of whether or not porphobilinogenuria results from a block in the biosynthetic pathway beyond PBG, an attempt was made to measure the disappearance rate of PBG in normal and porphyric liver slices. Since PBG enters cells poorly or not at all it is not possible to measure the hepatic rate of utilization of PBG by incubating liver slices with preformed porphobilinogen. PBG was formed intracellularly by the addition of a small quantity of ALA to liver slices. After 1 hour of aerobic incubation, delta-oximinolevulinic acid, an analogue of ALA which inhibits the conversion of ALA to PBG, was added to prevent any further formation of PBG. Half of the flasks remained aerobic, and the other half was continued as anaerobic incubations. Since PBG is not metabolized anaerobically, the difference in PBG content between anaerobic and aerobic flasks should be a measure of the disappearance rate of porphobilinogen.
The results of this study in rabbits showed that (1) the ability of liver slices to produce PBG from ALA far exceeded their ability to remove it, as measured by this method, and (2) the removal rate is barely detectable in normal and porphyric liver, with porphyric liver showing a tendency toward a greater removal rate. Because of the limitations of this method in measuring small differences, it cannot be concluded with certainty, but the data suggest, that a decreased rate of utilization of porphobilinogen is not the cause of porphobilinogenuria in porphyria.
The levels of hepatic ALA dehydrase activity in control and porphyric rats was determined next. The activity of homogenates was usually only slightly greater than that of slices. This degree of activity in slices indicates that ALA enters liver cells well. The activity per unit weight of liver is decreased in the rats manifesting porphobilinogenuria below that of controls. However, the total hepatic activity is greater in the porphyric rats
than in controls since hepatic enlargement occurs quite regularly during the administration of AIA to both rats and rabbits.
An increase in ALA dehydrase activity per unit weight of liver and a greater increase per total liver was seen in the porphyric rabbits as compared to controls. Enlargement of the livers of the AIA treated animals is the reason for the greater degree of rise of ALA dehydrase activity in total liver than per unit weight of liver. Total kidney ALA dehydrase is considerably increased in the AIA treated animals.
Liver catalase, as previously shown, was profoundly depressed in the AIA treated animals, both on the basis of activity per unit liver weight and per total organ. Kidney catalase was not altered greatly, but usually showed some increase per total organ. Liver lactic dehydrogenase in the few animals where it was studied was considerably decreased per unit weight, but not significantly decreased per total organ. In these same animals, renal lactic dehydrogenase was elevated on the basis of activity per unit weight and considerably more so per total organ.
Catalase and ALA dehydrase have previously been shown to behave similarly in rats and mice under the influence of a number of variables. Both enzymes are considerably decreased in activity in the livers of tumor bearing hosts. In the livers of C57 black mice both are normally about one-half the activity seen in the rivers of other strains of mice. Both enzymes are decreased in activity in liver following administration of 3-amino-l,2,4- triazole (AT). Although the effect of these variables has not been studied in rabbits, the administration of AIA to rabbits produces a fall in liver catalase activity and a rise in ALA dehydrase activity, the first example of a difference in response of the two enzymes.
The reason for the fall in liver catalase and rise in hepatic ALA dehydrase activity during administration of AIA is not apparent. Both Sedormid and AIA are thought to block the synthesis of catalase. However, increased incorporation of Fe59 into liver heme in AIA treated rats and other data along with the present study suggest that the decreased liver catalase activity is not a result of a block in the synthesis of the porphyrin component of the heme moiety.
The lesion could also represent a block in the known alternative pathway of ALA utilization. The utilization of the delta carbon of ALA for incorporation as a “one carbon fragment” following removal of the amino group and formation of a ketoglutaraldehyde was previously shown. If the activity of this pathway were normally of sufficient magnitude, a block in this pathway might produce some decrease in production of “one carbon fragment” derivatives such as purines and excessive accumulation of ALA and hence porphobilinogen. A decreased production of uric acid in Sedormid treated chicks has been demonstrated.
Any alpha-ketoglutaraldehyde which was oxidized directly through alphaketoglutarate in the tricarboxylic acid cycle, or any succinate formed following the emoval of the aldehyde carbon as a “one carbon fragment” would in either case ultimately cause the 4 carbon of ALA to be oxidized to CO2 via the tri-carboxylic acid cycle. The molar production of CO2 would not necessarily correspond to the number of moles of succinate produced from ALA, since tricarboxylic acid cycle intermediates are partially converted to amino acids, etc.
The rate of conversion of C14-delta aminolevulinic acid to C14-O2 was used as a measure of the activity of this alternative pathway of ALA metabolism. The rates of reaction as measured over various intervals (for total liver) are not consistently different in porphyric and normal liver, suggesting that a block in this alternative pathway of ALA metabolism is not present in experimental porphyria. While the above data suggest that the succinate-glycine cycle is intact as far as CO2 production from ALA is concerned, a block in the utilization of the aldehyde carbon of a ketoglutaraldehyde for one carbon fragment incorporation might not affect CO2 production from the 4 carbon of ALA.
Another possible lesion could be brought about by a partial block in the utilization of succinyl CoA. Such a lesion might lead to the increased synthesis of porphyrin precursors. In order to estimate the oxidative activity of the tricarboxylic acid cycle, the rate of formation of C14O2 from 5-C14 alpha-ketoglutarate was studied. C14O2 is produced from 5-C14 alpha-ketoglutarate by decorboxylation of uro- and copro-porphorinogen and in the reactions of the tricarboxylic acid cycle. C14O2 is not produced from labeled alpha ketoglutarate in the succinate glycine cycle. The contribution of porphyrinogen decarboxylation to C14O2 production as judged by PBG disappearance as a measure of maximum activity of this pathway is small compared to tricarboxylic acid cycle activity. The sites of C14O2 production from 5-C14 alpha-ketoglutarate are at and beyond oxaloacetate. Since no difference in the capacity of control and porphyric liver to oxidize 5-C14 alpha-ketoglutarate to C14O2 is seen, a block in the oxidative utilization of succinyl CoA and in the total activity of the tri-carboxylic acid cycle appears unlikely in the liver in experimental porphyria.
Another possible lesion could be the result of a block in one or more of the pathways of utilization of glycine for conversion to serine, C02, purines, etc. A defect in glycine utilization for any of these pathways might allow glycine to tend to accumulate with consequent overflow in the direction of porphyrin synthesis. Data are already available to suggest the possibility of defects in glycine utilization in experimental porphyria. A decreased production of uric acid in Sedormid treated chick embryos has been demonstrated. This might result from any of a number of possible biochemical defects including defective glycine utilization for purine synthesis. Defective conversion of glycine to serine has been demonstrated in certain individuals in a series of porphyric patients.
The rate of conversion of the alpha-carbon of glycine to CO2 by normal and porphyric liver slices was studied. The animals (controls and porphyries) were starved 72 hours, and presumably their livers were depleted of glycogen. Incubation with glucose produced a depression of the conversion of the alpha carbon of glycine to CO2 in both normal and porphyric liver, perhaps due to a “sparing” effect of carbohydrate. Since the oxidation of the alpha carbon of glycine to CO2 was greater in the absence of glucose, tissue from starved animals was incubated with glycine as the only substrate to approach more nearly the maximum capacities of the oxidative pathways of glycine to CO2.
The oxidation of the alpha carbon of glycine to CO2 is seen to be usually depressed in porphyric liver as compared to normal liver, the relative differences being more apparent at the later time points. This difference is not due to the earlier denaturation or “death” of porphyric tissue since other metabolic pathways in porphyric liver retain an activity comparable to those of normal liver. The disparity between the capacities of porphyric and normal liver to oxidize the alpha carbon of glycine to CO2 per unit weight of liver is greater than on the basis of total liver due to the greater weight of the liver in the porphyric animal.
The demonstrated effect of ingestion of carbohydrate in depressing PBG excretion in porphyric animals and the porphyric human prompted an evaluation of the capacity for glycolysis and certain accessory reactions in normal and porphyric liver.
The capacity of porphyric liver to convert the alpha carbon of pyruvate to CO2 is comparable to that of normal liver. The capacities of porphyric and normal liver to convert the alpha carbon of acetate to CO2 are comparable, but are considerably less than their abilities to convert the alpha carbon of pyruvate to CO2, possibly because of a ratelimiting step at the “activation” of acetate to acetyl-Co A.
The conversion rates of the 6 carbon of glucose to CO2 in normal and porphyric liver are likewise comparable at the time points evaluated.
Therefore, no defect in the capacity for glycolysis and in the capacity to oxidize the alpha carbon of pyruvate and acetate could be detected in the porphyric as compared to the normal liver. The normal activity of these and certain other metabolic pathways discussed above demonstrates that defects induced by AIA are specific and not related to generalized enzymatic or hepatic toxicity.
The above data suggest that there is no defect in PBG utilization, the “alternative pathway” of ALA utilization via the succinate-glycine cycle, the function of the tricarboxylic acid cycle, or glycolysis in the liver in experimental porphyria. However, present and previous data raise the question of alterations in glycine metabolism which may be related to overproduction of porphyrin precursors in experimental and human porphyria.
The oxidation of the alpha carbon of glycine to CO2 can probably proceed via various pathways, such as: (1) conversion of glycine to ALA and oxidation of ALA via the succinateglycine cycle as discussed above, (2) conversion of glycine to aminoacetone and its conversion to methylglyoxal and pyruvate, (3) conversion of glycine to glyoxylic acid which might have its alpha carbon utilized either as a “one carbon fragment” or conversion to formate and then to CO2 and (4) conversion to serine and thence to pyruvate. Only the first of these four pathways (that portion beyond ALA) has been studied experimentally in the present report, and no difference was seen between normal and porphyric liver. From these findings it would seem that one cannot invoke the tempting hypothesis of a block in the “alternative pathway” (ALA to a ketoglutaraldehyde, etc.) of ALA utilization to explain both an excessive accumulation of ALA and a decreased oxidation of the alpha carbon of glycine.
The pathways which have been previously suspected as defective in porphyria, i.e., purine synthesis and conversion of glycine to serine are both known to require “one carbon fragments.” The first three of the above mentioned pathways of oxidation of the alpha carbon of glycine are known to generate “one carbon fragments.” Thus, the oxidation of the alpha carbon of glycine, and generation of “one carbon fragments” from it, occurs via the same pathways initially. A defect in one or more of these pathways might result in a decrease of both the oxidation of the alpha carbon of glycine and formation of certain “one carbon fragment” end products such as purines or serine.
The relationship between the decreased liver catalase and the diminished ability of liver to oxidize the alpha carbon of glycine in AIA induced porphyria is not known. Catalase is probably involved in the oxidation of the alpha carbon of glycine. It has been suggested that glyoxylate formed from glycine forms a Schiff base with glutamate, which is cleaved to form CO2 and formylglutamate. The formyl group (originally the alpha carbon of glycine) can be transferred as a “one carbon fragment” or converted to formate, which can be oxidized to CO2 by catalase. Probably only this one of the above four pathways utilizes catalase in the oxidation of the alpha carbon of glycine to CO2. If this pathway were a major route of glycine oxidation, it is possible that the diminished catalase activity in AIA induced porphyria might be the cause of the diminished ability of porphyric liver to oxidize the alpha carbon of glycine.
None of the above speculations offers a mechanism to explain the profound decrease in liver catalase in experimental porphyria, unless this could be related to defective synthesis of its protein moiety as a result of alteration in amino acid metabolism.,
Endnotes
1. The porphyrias are a group of rare inherited or acquired disorders of certain enzymes that normally participate in the production of porphyrins and heme. They manifest with either neurological complications or skin problems, or occasionally both.
2. The same compound I was using in my kinetic enzyme studies.
3. Donald P. Tschudy, James Rose, Emanuel Hellman, Anne Collins and M. Rechcigl, Jr., “Biochemical Studies of Experimental Porphyria,” Metabolism 11, No. 12 (December 1962), pp. 1287-1301.
9. Aminoketone Metabolism in Tumor-bearing Host
Aminoacetone is formed metabolically from threonine, a reaction that is catalyzed by the enzyme threonine dehydrogenase. Aminoacetone can also be synthesized by condensation of glycine and acetyl CoA to form alpha-amino-beta-ketobutyric acid. This
reaction is analogous to that involved in the biosynthesis of delta-aminolevulenic acid, the aliphatic precursor of porphyrin.
Aminoacetone may be thus a key intermediate at least in some tissues in a pathway which is involved in the oxidation of both threonine and glycine and the utilization of their alpha-carbon atoms as “one-carbon fragments.”
Since the presence of a tumor affects the level of hepatic Delta-aminolevulinic acid dehydrase (ALA dehydrase), an enzyme involved in aminoketone metabolism, investigation of other aspects of aminoketone metabolism in tumor-bearing host was undertaken, in cooperation with my NIH colleague and friend Donald Tschudy. (1)
Aminoacetone production from threonine in whole liver homogenates was found to be quantitatively comparable to that of mitochondrial suspensions when both were calculated in terms of wet weight of whole liver. Attempts to increase threonine dehydrogenase activity by freezing and thawing or by use of detergents (Triton 100) produced only a marked decrease in activity of the enzyme. It was found that the conversion of threonine to aminoacetone was much more active in Tris buffer than in phosphate buffer. The optimal pH was 7.3-7.4, and aminoacetone formation increased with increasing concentration of L-threonine up to 0.4 M. Under these conditions in both mitochondrial preparation and whole liver homogenates, zero order kinetic conditions with regard to substrate were found to prevail for more than 30 minutes in all cases. A number of other variables which might influence the reaction were investigated. With increasing amounts of added DPN in homogenates, aminoacetone formation decreased.
To determine whether homogenates could be used as a measure of only the production of aminoacetone from threonine and not the difference between production and removal, it was necessary to determine whether aminoacetone might be metabolized by liver homogenates. When synthetically prepared aminoacetone was added to rat and mouse liver homogenates in amounts of 20, 100, and 1000 m/micromoles under the same conditions used in the experiments, no significant decrease in the amount of aminoacetone could be demonstrated at 60 and 120 minutes. Since no significant disappearance of aminoacetone occurred at a concentration equal to that which was produced from threonine during incubations, it was concluded that homogenates could be utilized to measure aminoacetone production from threonine.
To investigate whether a humoral factor produced by tumors might be responsible for the decrease of hepatic threonine dehydrogenase activity, the level of this enzyme in the livers of tumor-bearing parabiotic (2) partners was examined. In all four pairs of animals the hepatic threonine dehydrogenase activity per unit weight of liver was lower in the tumor-bearing animal than in the normal parabiotic partner. The hepatic threonine dehydrogenase activity was considerably decreased below that of their normal parabiotic partners. The normal animals which were anastomosed to the tumor-bearing animals were not significantly different from the pairs of normal controls. A decrease in hepatic threonine dehydrogenase in tumor-bearing animals was present when the tumor weight was in the range of 6-9 per cent of total body weight. This decrease was not seen in mice bearing Sarcoma 37 until much later in the course of tumor growth.
In another experiment the hepatic levels of catalase, ALA dehydrase, and threonine dehydrogenase were measured during the progressive growth of Sarcoma 37. Catalase and ALA dehydrase activity were both decreased by day 6 of tumor growth, whereas threonine dehydrogenase activity was normal. The decrease of catalase activity was more profound than that of ALA dehydrase by day 6. After day 10 catalase had reached a level slightly less than one-third of normal, beyond which it did not appear to decrease further. ALA dehydrase decreased between days 12 and 18 to about one-third the normal level. Threonine dehydrogenase activity remained normal through day 12 and then virtually disappeared between days 12 and 18.
The possibility, that failure to detect significant levels of aminoacetone after incubation with threonine in the livers of the tumor-bearing animals from day 18 might be a result of rapid catabolism of any aminoacetone formed, was investigated in another experiment by incubation of liver homogenates with synthetically prepared aminoacetone. Mice bearing Sarcoma 37 were sacrificed, 20 and 21 days after transplant, and homogenates of these livers were incubated with aminoacetone under the same conditions as in the experiments with threonine. Homogenates of these same livers were also incubated with threonine. No aminoacetone disappeared for periods up to 2 hours in homogenates of livers in which no aminoacetone formation from threonine could be demonstrated. It was thus concluded that failure to demonstrate aminoacetone after incubation with threonine was a result of markedly diminished or absent threonine dehydrogenase activity and not a rapid utilization of aminoacetone.
Because of the recent work which has raised the question of bacterial contamination of tumors in relation to “toxohormone” and its effect on liver catalase, a tumor which caused a decrease in both hepatic threonine dehydrogenase and ALA dehydrase was examined for bacterial contamination. Sarcoma 37 was grown as an ascites tumor and a solid tumor in CAF1 female mice, and the ascites tumor were cultured in thioglycollate broth at the time of sacrifice. Multiple cultures were taken from each tumor and incubated for more than 1 month. In these tumor-bearing mice, whose livers were shown to have significantly reduced threonine dehydrogenase and ALA dehydrase activity, the tumor cultures were uniformly negative. In these animals, as in those of day 18, hepatic threonine dehydrogenase activity was almost completely absent.
It is apparent from this study that the pathway of aminoacetone —> methylglyoxal —> D-lactate —> pyruvate, suggested by earlier investigators, does not appear to be active in normal rat or mouse liver homogenates or in tumor-bearing mouse liver homogenate, since no significant aminoacetone disappearance could be demonstrated during incubation of homogenates of these livers. The possibility remains that in the present experiment mitochondrial permeability to exogenous aminoacetone may limit or prevent the demonstration of its catabolism by liver, if conversion of aminoacetone to methylglyoxal occurred only in mitochondria.
Our experiments have demonstrated that hepatic level of threonine dehydrogenase is decreased in tumor-bearing hosts. The enzyme appears to decrease to barely detectable levels in mice bearing advanced Sarcoma 37. This may be the first example of an enzyme which completely disappears from the liver of a tumor-bearing host. The present data do not permit detailed conclusions concerning the relationships between biological characteristics of the tumor and the rate decrease of the enzyme. The enzyme was decreased in the livers of rats bearing adrenocortical Carcinoma 494, when it was only 6- 9 per cent of body weight, but Sarcoma 37 in mice caused no effect on the hepatic level of this enzyme, until it had exceeded one-third the carcass weight. Adrenocortical Carcinoma 494 grows much more slowly than Sarcoma 37 and was present for a much longer period of time when it had reached 6-9 per cent of body weight than was required for the Sarcoma 37 to exceed one-third of the carcass weight. It is thus possible that slowly growing tumor may affect the hepatic level of the enzyme at a relatively smaller tumor size than rapidly growing tumors. Another possibility is a species difference in tumorhost relationships, since Sarcoma 37 grows in mice and adrenocortical: Carcinoma 494 grows in rats.
Hepatic enzyme activities may be affected in a specific order during progressive tumor growth. The data suggest that catalase and ALA dehydrase are affected before threonine dehydrogenase. If the presence of a tumor produces specific alterations in hepatic amino acid and protein metabolism, the stage of tumor growth at which certain hepatic enzymes decline and their rate of decrease would be related to variables such as their turnover rates.
The possibility of a humoral transmission of a material from tumors which might cause a decrease in hepatic threonine dehydrogenase was investigated by a study of the activity of this enzyme in the livers of parabiotic rats with one partner bearing a tumor. Other investigators performed this same type of experiment in studying the mechanism by which tumors cause a depression of liver catalase. They found the liver catalase activity of both the tumor-bearing animal and the normal parabiotic partner to be reduced. This is in contrast to the present experiments on threonine dehydrogenase, wherein the hepatic activity of this enzyme was reduced in the tumor-bearing animal but was normal in the tumor-free parabiotic partner. The capacity of a humoral factor to produce an effect in the normal partner is a function of the difference of its rate of catabolism (and excretion) by the normal partner and its rate of entry into the circulation of the normal partner. The blood flow through the anastomosis534 is small in comparison to the blood volume of each of the animals. It was estimated that in this type of parabiotic preparation about 0.66 per cent of the blood volume per minute was transferred from one animal to another. Substances which are present in high concentration and are not “cleared” rapidly from the blood can pass from one parabiont to the other to achieve sufficient concentration in both animals necessary to exert an effect. On the other hand, substances whose concentration is low or whose turnover is high with respect to the blood flow across the anastomosis will not be detected across the parabiotic barrier. If a humoral material produced by tumors is the cause of the decreased hepatic threonine dehydrogenase, it would be of this latter category. It is not necessary, however, to invoke the existence of a humoral factor produced by tumors which causes specific enzyme activities to decrease in the liver, since tumors might accelerate the removal from the circulation of materials necessary for the maintenance of certain enzymes in the liver.
A decrease of hepatic threonine dehydrogenase can occur in the tumor-bearing animal in the absence of anemia. The animals bearing adrenocortical Carcinoma 494 were markedly polycythemic (hematocrits over 60 per cent), even though hepatic threonine dehydrogenase was decreased.
Endnotes
1. 532 Donald P. Tschudy, Frederick H. Weeland, Annie R. Collins, and Miloslav Rechcigl, Jr. The Effect of Tumor Growth on Hepatic Threonine Dehydrogenase. Cancer Research 24 (1964), 2033-2037.
2. 533 Parabiosis refers to the union of the individuals, as of joined twins, or by surgical operation.
3. 534 The connection of normally separate parts or spaces so they intercommunicate.
4. 535 Inborn errors of metabolism comprise a large class of genetic diseases involving disorders of metabolism. The majority are due to defects of single genes that code for enzymes that facilitate conversion of various substances (substrates) into others (products). In most of the disorders, problems arise due to accumulation of substances which are toxic or interfere with normal function, or to the effects of reduced ability to synthesize essential compounds. Inborn errors of metabolism are now often referred to as congenital metabolic diseases or inherited metabolic diseases.
10.‘Overproduction Disease’ Localized to a Specific Gene
Acute intermittent porphyria is an inborn error of metabolism (1) characterized chemically by the excretion in the urine of excessive amounts of porphyrin precursors [delta-aminolevulinic acid (ALA) and porphobilinogen (PBG)] and clinically by neurologic dysfunction. It is transmitted as an autosomal dominant disorder with an incidence of about 1.5 per 100,000 in most areas. Chemical manifestations which resemble those of acute intermittent porphyria can be produced in animals by administration of any one of several compounds. The increased excretion of porphyrin precursors seen in experimental porphyria produced by at least one of these compounds (allylisopropylacetamide) is now thought to result from the induction in liver of delta-aminolevulinic acid synthetase (ALA synthetase), the rate-controlling enzyme of porphyrin biosynthesis. The enzyme occurs exclusively in mitochondria (2). The induction of this enzyme can be inhibited by carbohydrate administration - a phenomenon previously demonstrated with certain other inducible enzymes and termed the “glucose effect.” The inhibition of induction of hepatic ALA synthetase by carbohydrate explains the observation that the ability to produce experimental porphyria is reciprocally related to the intake of carbohydrate in the diet. The fact that changes in diet in the human genetically determined disease produced similar effects on porphyrin precursor excretion to those seen in experimental porphyria suggested that ALA synthetase was induced in the livers of patients with the disease and that the induction in man was subject to the “glucose effect.”
Jointly with my colleague Donald Tschudy, I had an opportunity to collaborate on a study to carry out biochemical (3) analyses on the tissues of a patient who died of acute intermittent porphyria (AIP) which brought additional light on the understanding of this disease.
The levels of ALA synthetase, ALA dehydrase, catalase, and lactic dehydrogenase in the liver of a patient with acute intermittent porphyria and those of nonporphyric were determined and compared with control patients. ALA synthetase activity in the porphyric liver exceeded that in the biopsy control livers approximately 7-fold and in the postmortem controls by at least 14-fold. The porphyric liver was obtained about 30 min after death, whereas the liver biopsy samples from controls were utilized immediately after removal from live patients.
The values of the biopsy control specimens which were obtained under general anesthesia were about the same as those of normal unanesthetized animals. Hence it is unlikely that the differences between porphyric and control livers are related to administration of anesthetics. The values of ALA synthetase obtained by the present technique do not represent the measurement of the activity of the isolated enzyme, but rather the ALA produced from added glycine and endogenously generated succinyl coenzyme A. A number of previous observations using liver homogenates from other species suggest that the ALA obtained by this method is a good measure of ALA synthetase activity:
(1) Under the conditions of homogenate (4) incubation, ALA utilization does not occur to a significant extent at the concentrations at which it is produced.
(2) That the generation of succinyl coenzyme A is not the rate-limiting process in the formation of ALA in the homogenate incubations described above is supported by several lines of evidence, (a) Addition of various tricarboxylic acid cycle (TCA cycle) intermediates at different concentrations does not increase ALA production above that of endogenous TCA cycle substrates; (b) The enzyme is probably readily saturated by endogenously generated succinyl coenzyme A, since the Km for this substrate is probably very much lower than for glycine in both liver and nucleated erythrocytes; (c) The concentration of succinyl coenzyme A which is the optimal range for ALA synthetase activity in rat liver mitochondria is approximately that concentration actually found for TCA cycle intermediates in fasting rat liver; (d) Estimates of TCA cycle capacities in rat liver by several techniques show that only a small fraction (probably < 1%) of the succinyl coenzyme A generated is required for maximal ALA production. These findings indicate that the level of ALA synthetase is the rate-limiting step in the formation of ALA by liver homogenates under the conditions described above.
(3) Earlier methods for measuring ALA synthetase have used isolated mitochondria with added glycine and TCA cycle intermediates. The present homogenate method produces somewhat greater amounts of ALA per unit liver than the previous methods.
The findings which are the basis for the chemical diagnosis of the disease, i.e., increased excretion of ALA and PBG, are shown by the present data to result from an increased ability to synthesize ALA in the livers of patients with acute intermittent porphyria. The evidence cited above indicates that this overproduction is caused by a high level of ALA synthetase, previously shown to be an inducible enzyme. The relationship of the increased ALA synthetase in the liver to the neurologic aspects of the disease remains unknown, since some patients who are asymptomatic produce large amounts of porphyrin precursors.
The fact that acute intermittent porphyria is a genetic disease with a high hepatic level of ALA synthetase raises the question of whether it could be caused by a defect of a regulator gene, i.e., a constitutive regulator mutation. Evidence against this hypothesis is the fact that such genetic defects are recessive, whereas acute intermittent porphyria is a dominant disorder. A constitutive operator mutation could explain the genetic and biochemical findings. This requires that the neurologic symptoms result from overproduction of porphyrin precursors. However, if this etiologic relationship does not exist and one postulates a single genetic defect as the explanation of both chemical and clinical aspects of the disease, then the defect may not be in a gene involved in the primary regulation of ALA synthetase production.
Another possibility is that a defect in a structural gene in the heme biosynthetic pathway beyond PBG might result in defective end-product repression. For example, if the end product of the porphyrin biosynthetic pathway (heme) were involved in the repression of ALA synthetase, as has been suggested in microorganisms, then a partial block in this pathway could explain the induction of ALA synthetase. Evidence against this hypothesis is the fact that hepatic ALA synthetase can be induced in high levels in one type of experimental porphyria where ability for heme synthesis appears to be increased. Therefore, it is possible that the genetic defect lies outside the heme synthetic pathway and secondarily leads to induction of ALA synthetase.
Assuming ALA synthetase to be the rate-controlling enzyme in human hepatic porphyrin synthesis, then the maximal ability for heme synthesis in a normal adult liver is approximately 1/4 to ½ the normal unstressed marrow production per day.
A slight increase in ALA dehydrase activity in acute intermittent porphyria is suggested by the data. As measured by the present techniques, the maximal capacity to utilize ALA exceeds the ability to generate it by a factor of about 80 in normal liver. The reason for the normally high ratio of ALA dehydrase to ALA synthetase probably lies in the fact that ALA is formed in mitochondria and diffuses out into the soluble portion of the cell where it is acted on by ALA de-hydrase to form PBG. This means that ALA is distributed over a relatively large intracellular volume at a low concentration before it is acted on by ALA dehydrase. Because of the low concentration of ALA normally presented to ALA dehydrase, a high level of the latter enzyme is required to ensure maximal utilization of the ALA which is formed and to minimize the escape of ALA from the cell. As the ratio of ALA synthetase to ALA dehydrase rises, as in acute intermittent porphyria, the increasing amount of ALA which is formed is partitioned between escape from the cell and conversion to PBG. Both of these processes are increased above those of normal liver cells. This explains the pattern of porphyrin precursor excretion seen in this disease. The large excess of ALA dehydrase capacity above ALA synthetase capacity demonstrated in vitro may be greater than actually exists in vivo, since the ALA dehydrase may be functioning under conditions which do not permit maximal activity.
Examination of three other enzymes outside the porphyrin synthetic pathway (AA synthetase, catalase, and lactic dehydrogenase), reveals a variable degree of decrease in activity in the porphyric liver as compared with biopsy samples from normal liver. Some or all of the decrease may be artifactual in that the porphyric liver was obtained shortly after death, whereas the control liver biopsy samples were studied immediately.
It has been proposed that the increased production of ALA in acute intermittent porphyria might result from inability to convert glycine to AA. The production of AA from glycine in the porphyric patient is only slightly lower than in controls. The conditions used for the incubation were those for optimal ALA production and are not optimal for AA production. Hence, the AA produced cannot be considered a measure of maximum ability to produce AA from glycine. The present study clearly demonstrates, however, that significant amounts of AA can be produced from glycine and endogenously generated acetyl coenzyme A by the liver in acute intermittent porphyria. As previously demonstrated in guinea pig liver, normal human liver appears to produce much more AA than ALA from glycine.
The hepatic catalase of the porphyric patient is less than that of biopsied controls; it is not certain to what extent this decrease results from rapid post-mortem breakdown of the enzyme.
An increased level of ALA synthetase in the liver of a patient with acute intermittent porphyria provides an explanation of the increased porphyrin precursor production and excretion seen in this disease. This is thought to represent the first example in which the biochemical manifestations of a disease are directly attributed to an increased level of a specific enzyme. The mechanism responsible for the induction of ALA synthetase remains unknown as does the relationship of the high hepatic ALA synthetase to the neurologic manifestations of the disease.
A follow-up morphologic study (5) on the same patient provided evidence suggesting existence of the syndrome of inappropriate antidiuretic hormone (ADH) release. Examination of the brain revealed injury to the median eminence and bilateral loss of neurons of the supraoptic and paraventricular nuclei. This represents the first correlation of a specific hypothalamic lesion with the syndrome of inappropriate ADH release.
Endnotes
1. Inborn errors of metabolism comprise a large class of genetic diseases involving disorders of metabolism. The majority are due to defects of single genes that code for enzymes that facilitate conversion of various substances (substrates) into others (products). In most of the disorders, problems arise due to accumulation of substances which are toxic or interfere with normal function, or to the effects of reduced ability to synthesize essential compounds. Inborn errors of metabolic diseases.
2. In cell biology, a mitochondrion (plural mitochondria) is a membrane-enclosed organelle found in most eukaryotic cells (nucleated cells).
3. Donald P. Tschudy, Mark G. Perlroth, Harvey S. Marver, Annie Collins, George Hunter, Jr., and M. Rechcigl, Jr., Acute Intermittent Porphyria: The First “Overproduction Disease Localized to a Specific Gene,” Proceedings of the National Academy of Sciences 53, No. 4 (April 1965), pp. 841-847.
4. A tissue that is or has been made homogenous, as by grinding cells into a creamy consistency for laboratory studies,
5. Harvey S. Marver, Anne Collins, Donald P. Tschudy and Miloslav Rechcigl, Jr. Delta-Aminolevulinic Acid Synthetase. II. Induction in Rat Liver. Journal of Biological Chemistry 241 (1966), pp. 4323-4329.
11. Induction of ALA Synthetase
The increased excretion of porphyrin precursors, such as delta-aminolevulenic acid (ALA) and porphobilinogen in porphyria has been linked with the induction of deltaaminolevulinic acid synthetase (ALA synthetase) in the liver. This rate-controlling enzyme of porphyrin biosynthesis catalyzes the reaction between pyridoxal phosphate-activated glycine and succinyl coenzyme A to form delta-aminolevulinic acid. The induction can be brought about experimentally by administration of allylisopropylacetamide (AIA).
Together with Don Tschudy and his colleagues (1), we investigated the process of ALA synthetase induction in rat liver. Twenty-four hours after the last of two daily injections of AIA to fasted rats, the liver size and liver protein were both increased by 55%. With the possible exception of a 15% increase in mitochondria, the protein concentrations of the various subcellular fractions were not altered significantly. The increase was distributed in approximately normal proportions among the subcellular components, again except
possibly for the mitochondria, in which the fractional protein content was increased by 20%.
Following a single injection of AlA there was a pronounced rise of hepatic ALA synthetase, the maximum level being attained with a dosage of 300-400 mg AlA. The rise in ALA synthetase, following AIA, was blocked by a number of inhibitors of protein synthesis. Succinyl-CoA failed to enhance Delta-amino-levulinic acid production when added to liver homogenates of rats receiving AIA together with these compounds. Addition of 1 mM concentrations of these compounds to porphyric rat liver homogenates did not interfere with Delta-aminolevulinic acid production. The effects of these compounds in vivo were not related to the production of inhibitors, as evidenced by the fact that addition of equal quantities of liver homogenates of rats treated with these compounds to porphyric rat liver homogenates did not reduce Delta-aminolevulinic acid production.
The half-life of ALA synthetase as determined by the use of puromycin was independent of the level of the enzyme, suggesting that the degradation of the induced enzyme is a first order process. In a total of four similar experiments, the half-life of ALA varied from 67 to 72 min following administration of puromycin.
The injection of actinomycin D or 5-fluorouracil resulted in rapid declines in the hepatic levels of induced ALA synthetase. Actinomycin D, given 14 hours after AIA, and 5- fluorouracil, given 12 hours after AIA, produced a 47% and 60% decrease, respectively, in ALA synthetase in 4 hours. The more detailed study of ALA synthetase levels following actinomycin D revealed the decline to be composed of several components.
Glucose administration 3 hours prior to and together with AIA was as effective in blocking the induction of ALA synthetase as actinomycin D. The addition of 1mM glucose or liver homogenates of glucose-treated rats did not reduce Delta-aminolevulinic acid production by porphyric rat liver homogenates.
When glucose was administered after ALA synthetase had been induced by AIA, a significant decline occurred in the subsequent 4 hours as compared with the continued rise observed in the controls. The curves of decline of ALA synthetase following glucose resembled those following actinomycin D in that they also appeared to contain several components. The more profound inhibition of the induction of ALA synthetase by glucose given at zero time may be related to the fact that the animals used for the experiment had been additionally treated with glucose 3 hours before AIA.
Glucose modified other effects of AIA on the liver in addition to its inhibition of the AIA-mediated induction of ALA synthetase. When AIA was administered to glucose-fed rats it produced a smaller increase in liver weight (28% after 3 days of AIA) and total liver nitrogen (19% after 3 days of AIA) per g of body weight than occurs when the drug is given to fasted animals (54% id 47%, respectively, after 3 days of AIA). The 56% increase of hepatic Delta-aminolevulinic acid dehydratase (per mg of nitrogen) which results from the administration of AIA to fasted rats for 3 days was completely eliminated by feeding 2.5 g of glucose twice daily during that time. As compared with control values, hepatic catalase levels (per mg of nitrogen) decreased by 70% in glucose-fed rats treated with AIA for 3 days and by 79% in fasted rats similarly treated with AIA.
The administration of glucagon by several different schedules, including injection of 1.0 mg per kg of crystalline glucagon every 3 hours for the 12 hours before the animals were killed, failed to increase the levels of hepatic ALA synthetase. Crystalline glucagon (1 mg per kg) given every 3 hours for the 12 hours before the animals were killed together with AIA given 12 hours before killing slightly decreased (17%) the induction of ALA synthetase produced by AIA alone.
Bovine growth hormone injected every 6 hours or 12 hours before killing the rats slightly enhanced (17%) the increase in ALA synthetase resulting from AIA given 12 hours before the rats were killed.
As shown in this study, in addition to its enhancement of ALA synthetase levels, AIA produced a rapid and marked increase in the protein content without changing liver protein concentration. The possible increase of mitochondrial protein resulting from AIA administration was of particular interest since ALA synthetase is a mitochondrial enzyme.
The mechanism for the AIA-mediated increase of hepatic ALA synthetase had been investigated by the use of a number of compounds which interfere with different steps of protein synthesis. The fact that the increase of ALA synthetase following AIA is blocked by puromycin and DL-p-fluorophenylalanine suggests that it is dependent upon protein synthesis. Puromycin is known to interfere with the transfer of amino acids to ribosomes, whereas DL-p-fluorophenylalanine may result in nonfunctional enzyme. At the dosage used in these experiments, puromycin blocked protein synthesis in rat liver by the above mechanism, and DL-p-fluorophenylalanine interfered with the formation of a number of enzymes in regenerating rat liver. The inhibition of the AIA-mediated increase of ALA synthetase by actinomycin D and 5-fluorouracil suggests that this rise is dependent upon RNA synthesis. Both actinomycin D and 5-fluorouracil are inhibitors of mammalian as well as bacterial enzyme induction. 5-Fluorouracil is inhibitory as a consequence of its corporation in vivo into RNA. Actinomycin D inhibits the DNA-dependent synthesis of RNA. However, recent studies have suggested that this compound may have other modes of action. Three lines of evidence suggest that in the present experiments actinomycin D blocks the AIA-stimulated rise of ALA synthetase as a result of inhibition of RNA synthesis: (a) 5-fluorouracil, which affects RNA synthesis by a mechanism different from that of actinomycin D, similarly interferes with the AIA-mediated increase of ALA synthetase; (b) at the dosage of actinomycin D used (1.5 mg per kg) there is reported to be both selective inhibition of RNA synthesis and retention of normal liver histology; and (c) there is no evidence that under these conditions actinomycin D significantly interferes with energy production, since addition of succinyl-CoA to incubation mixtures containing liver homogenates of AIA-actinomycin D-treated rats did not augment 5-aminolevulinic acid synthesis.
In view of the mechanism by which puromycin inhibits protein synthesis, the decline of ALA synthetase following administration of the antibiotic is a measure of the rate of degradation of the induced enzyme. This method for estimating enzyme half-lives has been used previously in studies of tryptophan pyrrolase and it is supported by the fact that the rate of loss in vivo of isotope from prelabeled tryptophan pyrrolase is very similar to the rate of decay of the induced enzyme after puromycin. The half-life of ALA synthetase of 67 to 72 min following puromycin is contrasted by a mean half-life of 2 to 3 days for total liver protein.
‘The prompt decline of induced hepatic ALA synthetase after administration of actinomycin D differs markedly from the unresponsiveness of gross incorporation of C14- leucine into liver protein of intact rats. Actinomycin D also fails to reduce the hepatic levels of that fraction of microsomal RNA which, like bacterial messenger RNA, is capable of stimulating amino acid incorporation in vitro. Therefore, the delay in the effect of actinomycin D on protein synthesis in liver, as well as certain other tissues, is thought to result from relatively stable messenger RNAs. The rapidity with which ALA synthetase falls after actinomycin D may be explained on the basis of a relatively unstable messenger RNA for that enzyme. This explanation is consistent with the fall of induced ALA following administration of 5-fluorouracil.
If in the present studies actinomycin D blocks the synthesis of the messenger RNA for ALA synthetase, then the decrease of ALA synthetase after administration of actinomycin D would be a function of the degradation rate of both the specific messenger RNA and the enzyme. This may explain the multiple components of the curve of ALA synthetase activity when actinomycin D is administered after induction of the enzyme. A recently reported model of protein synthesis permits the calculation of the t1/2 of a specific messenger RNA from this type of curve when the % of the protein (enzyme) is known. With this model, it can be shown that the half-life of the messenger RNA for ALA synthetase is approximately the same as that of the enzyme, that is, 65 to 75 min.
There has been a recent accumulation of evidence supporting the idea that there is considerable variation in the stability of individual proteins and messenger RNAs within the same cell. ALA synthetase and its messenger RNA may account for a fraction of relatively unstable protein and RNA within the hepatocyte.
It would appear that the AIA stimulation of RNA synthesis continues for at least 14 hours after administration of the drug. This is evidenced by the finding that actinomyci D given 14 hours after AIA produces a more rapid decline of ALA synthetase than occurs after administration of AIA alone.
The present experiments indicate that glucose profoundly inhibits the AIA-stimulated induction of hepatic ALA synthetase. Previous studies have shown that the increase of porphyria precursors in acute intermittent porphyria is reciprocally related to carbohydrate intake in the diet, and that this increase results from markedly elevated levels of ALA synthetase. These findings are best explained by the hypothesis that glucose inhibits the genetically mediated induction of ALA synthetase in acute intermittent porphyria, which suggests a general effect of glucose on the induction of hepatic ALA synthetase. This inhibitory effect of glucose on hepatic enzyme induction has previously been described only for rat liver threonine dehydrase and ornithine transaminase.
The presence of several components in the curves of decline of ALA synthetase following both glucose and actinomycin D suggests that these compounds may affect the induction of ALA synthetase by a similar mechanism. That is, the two components of the curves of decline of ALA synthetase after glucose may be related to the degradation of ALA synthetase and its messenger RNA, as has been postulated for the curves following actinomycin D. This would imply that, in the liver, glucose may interfere with the induced synthesis of certain RNAs.
Ornithine transaminase and threonine dehydrase can be induced by the administration of glucagon to intact rats. In contrast, glucagon failed to increase the hepatic levels of ALA synthetase, nor did it augment the AIA-mediated induction of that enzyme.
Endnotes
1. Mark G. Perlroth, Donald P. Tschudy, Harvey S. Marver, Costan W. Berard, Robert F. Zeigel, M. Rechcigl, Jr. and Annie Collins, Acute Intermittent Porphyria. New Morphologic and Biochemical Findings, American Journal of Medicine 41 (July 1966), pp. 149-162.
2. Harvey S. Marver, Anne Collins, Donald P. Tschudy and Miloslav Rechcigl, Jr. Delta-Aminolevulinic Acid Synthetase. II. Induction in Rat Liver. Journal of Biological Chemistry 241 (1966), pp. 4323-
12. Role of Catalase and Peroxidase in Leukocytes
During my early years in the Laboratory of Biochemistry, I became acquainted with a young fellow in Dean Burk’s lab, Warren H. Evans, who carried out metabolic studies on cellular preparations using the Warburg apparatus. Since each of us could contribute a different expertise and knowledge, we decided to collaborate on the study of the metabolism of leukocytes (1) which are rich in both catalase and peroxidase.
Despite the large concentration of these two enzymes in leukocytes, relatively little is known about their respective roles. Since hydrogen peroxide is known to be present in leukocytes, an obvious physiological function of catalase and peroxidase might be to regulate intracellular concentration of hydrogen peroxide which, if allowed to accumulate, could become toxic to these cells.
In order to assess the possible role of catalase in the metabolism of leukocytes, we took the advantage of a relatively specific inhibitor of catalase activity, 3-amino-1,2,4-triazole
(AT). This compound has been shown to inhibit catalase activity of liver and kidney but not of the erythrocyte. We examined the effect of AT on leukocyte catalase in vivo and in vitro in comparison with other tissues. An attempt was also made to ascertain a role of catalase in respiration and aerobic glucose utilization of leukocytes.
We used guinea pigs as the source of tissues. Erythrocytes were obtained by centrifuging heparinized blood (2). After the removal of the plasma and the buffy layer (3), the erythrocytes were washed with saline to remove any trace of AT, and then lysed (4) in distilled water.
Suspensions rich in polymorphonuclear leukocytes were obtained from peritoneal exudates (5) of guinea pigs injected intraperitoneally with a solution of sodium caseinate prior to killing of the animal. The leukocytes were removed from the exudates by centrifugation, washed with saline and finally suspended in distilled water and homogenized, using a motor-driven pestle. For enzymatic essays the homogenates were diluted with distilled water.
Respiration and uptake of glucose were measured in Warburg flasks in air. For enzymatic essays the cells were collected quantitatively from the Warburg flasks at the end of the incubation period, washed with saline and homogenized, as described above.
Our in vivo experiments showed that the injection of AT produced 98 per cent inhibition of catalase activity of the liver, 91 per cent inhibition of the kidney catalase but was without any apparent effect on the erythrocyte catalase. AT also caused a profound depression of the leukocyte catalase, 96 per cent, comparable to that seen in the liver.
The leukocyte catalase was also markedly inhibited by AT in vitro. The inhibition was proportional to the AT concentration used. The leukocyte preparations treated to remove the few contaminating red cells had a level of catalase activity comparable to that seen in the untreated leukocytes.
Inhibition of the leukocyte catalase occurred regardless of whether glucose was present or absent in the incubation medium. Despite the marked effect on the leukocyte catalase activity, AT had no apparent influence on the leukocyte respiration rate and glucose utilization.
Leukocytes were also assayed for their myeloperoxidase activity. The incubation of leukocytes with AT, in contrast to catalase, produced no apparent change in the leukocyte peroxidase activity.
On the basis of this evidence, with respect to the effect of AT on catalase activity of different tissues, it is evident that leukocyte behavior is similar to that of the liver and kidney rather than that of the erythrocyte.
Hydrogen peroxide in leukocytes is generated as a product of glucose oxidation via hexosemonophosphate shunt pathway. Our results clearly showed that hydrogen peroxide is formed in leukocytes since the presence of hydrogen peroxide is a prerequisite for the inhibition of catalase by AT. It is evident that the rates of oxygen and glucose uptake by leukocytes are relatively independent of catalase activity down to as little as 9 per cent of the original activity of this enzyme. It is possible that the residual activity in the presence of AT is sufficient to protect the cell from hydrogen peroxide accumulation. However, more likely explanation is that some other hydrogen peroxide destroying system, such as myeloperoxidase, which is found in high concentrations in leukocytes, is of greater significance than catalase in protecting the cell from any toxic effects of hydrogen peroxide (6).
Endnotes
1. A colorless blood corpuscle capable of ameboid movement, whose chief function is to protect the body against microorganisms causing disease.
2. Blood treated with heparin to prevent clotting.
3. The coagulated plasma of blood when the red corpuscles have so settled out that the coagulum appears nearly colorless.
4. The disintegration of a cell resulting from destruction of its membrane by a chemical substance, especially an antibody or enzyme.
5. Exudate is fluid that leaks out of blood vessels into nearby tissues. The fluid is made of cells, proteins, and solid materials. Exudate may ooze from cuts or from areas of infection or inflammation.
6. Miloslav Rechcigl, Jr. and Warren H. Evans, Role of Catalase and Peroxidase in Metabolism of Leukocytes, Nature 199, No. 4897 (September 1963), pp. 1001-1002.
3. Factors Influencing Leukocyte Myeloperoxidase and Catalase
In the previous studies it was not possible to determine whether the small residual catalase activity in the presence of 3-amino-1,2,4-triazole (AT) was sufficient to protect leukocytes from any toxic effects of H2O2 or whether myeloperoxidase which is found in high concentration in polymorphonuclear leukocytes, might also be involved. The localization of myeloperoxidase in the lysosome-like granule of leukocytes suggested to us the possibility that this enzyme, like hydrolases, might not function in resting cells but would be active in phagocytizing cells following granule lysis. (1).
In contrast to our previous results with intact cells, the data indicated that AT inhibits myeloperoxidase and catalase in leukocytes homogenized in a sucrose-free medium containing an H2O2 - generating system. Both enzymes were inhibited to about the same extent at various concentrations of AT. The data showed that inhibition of myeloperoxidase is dependent on the presence of the added H2O2 -generating system. The inhibition of catalase by AT in the absence of any added glucose oxidase suggests the presence of endogenous H2O2, since it is known that the inhibition of catalase by AT requires H2O2.
Increasing the level of the H2O2 -generating system in the absence of AT results in an inhibition of myeloperoxidase but not of catalase. In control studies, glucose oxidase added in the absence of glucose has no effect on myeloperoxidase or catalase. The value for O2 uptake were greater in the presence of AT than in the absence of AT when the glucose oxidase concentration exceeded 3.3 microgram /ml.
This effect is presumably due to a decrease in the conversion of H2O2 to O2 and water by catalase. The finding of an inhibitory effect of AT on myeloperoxidase in crude homogenates was confirmed by studies on the purified enzyme. The effect of AT on myeloperoxidase and catalase in intact cells incubated in the presence of polystyrene particles, methylene blue and phenazine methosulfate was also studied. These factors are known to greatly enhance the hexose-monophosphate shunt pathway in leukocytes which could lead to an increase in endogenous H2O2 production. In our studies AT had no effect on myeloperoxidase in intact cells exposed to methylene blue or phenazine methosulfate. In leukocytes phagocytizing polystyrene particles 55% inhibition of myeloperoxidase by AT was observed. Phenazine methosulfate also caused significant inhibition of catalase and myeloperoxidase. AT had no significant effect on the O2 uptake of leukocytes in the presence or absence of any of the factors studied. An experiment was also carried out in which glucose and glucose oxidase were added to intact cells in the presence and absence of AT. Neither the H2O2 generated by glucose oxidase, nor the AT had any effect on myeloperoxidase activity.
The possibility that the granules which contain myeloperoxidase might protect this enzyme from inhibition by AT or H2O2 in intact cells was investigated by comparing the effects of AT and H2O2 on myeloperoxidase in intact versus disrupted granules present in the supernatant fraction of leukocytes homogenized in sucrose. Catalase assays were also performed on these preparations. The results showed that the enzyme in the absence of AT or H2O2 is 18 times more active in granule preparations disrupted by freezing and thawing as compared to intact preparations. It is clear that from the data that myeloperoxidase in intact granules is not inhibited by H2O2, AT or both whereas in granules disrupted by freezing and thawing myeloperoxidase is inhibited 28% by H2O2 and 83% by AT plus H2O2. In the absence of H2O2 or AT, catalase activity was essentially the same as in the unfrozen supernatant fraction as in the frozen-thawed preparation. In the unfrozen preparation, AT inhibited catalase to the same extent in the presence and absence of H2O2, whereas in the frozen-thawed preparations the effect of AT in the absence of H2O2 was only about one-half of that in the presence of H2O2.
Myeloperoxidase and catalase were readily separated by centrifuging the uninoculated supernatant fraction. Myeloperoxidase apparently behaves like lactoperoxidase, fatty acid peroxidase and iodide peroxidase with respect to inhibition by AT. Other peroxidases, such as horseradish peroxidase and the peroxidase of pea seedlings are not inhibited by AT.
The failure of AT to inhibit myeloperoxidase in intact cells is probably not due to the lack of an adequate concentration of H2O2 in the cell. A more likely explanation is that the granules that contain myeloperoxidase prevent either H2O2 or AT or both of these compounds from interacting with the enzyme.
In keeping with their lysosome-like nature, intact leukocyte granules appear to have a low permeability to substrates. Thus, we have observed that myeloperoxidase is about 18 times more active in disrupted granules as compared to intact granules. The fact that the enzyme is protected from inhibition by H2O2 in intact granules but not in disrupted granules further suggests a low permeability of substrates for hydrolases found in intact granules.
On the basis of these data one could postulate that myeloperoxidase, like lysosomal enzymes, is probably inactive in resting leukocytes in which the granules are intact. Myeloperoxidase is more likely to be active in phagocytizing leukocytes in which lysis of the granules occurs. Though the function of myeloperoxidase in phagocytizing leukocytes is far from being understood, evidence is accumulating in the literature that suggests that the enzyme might participate in the inactivation of various bacteria and bacterial toxins.
Whatever role myeloperoxidase may play in phagocytizing leukocytes, the possibility arises that this enzyme, once released from the granules, might be inactivated by the increased endogenous H2O2 produced by ingested bacteria. That the former probably does not occur is indicated by our data showing that essentially no inactivation of myeloperoxidase occurs in leukocytes ingesting polystyrene particles. One possible explanation for the finding is that catalase in leukocytes regulates the endogenous H2O2 concentration at a level which is non-inhibitory to myeloperoxidase, a similar role of catalase in protecting myeloperoxidase from H2O2 generated by bacteria is also conceivable/
Endnotes
1. Warren H. Evans and Miloslav Rechcigl, Jr., Factors Influencing Myeloperoxidase and Catalase in Polymorphonuclear Leukocytes, Biochimica et Biophysica Acta 148 (1967), pp. 243-250.
14. Structure-Function Relationships
I became interested in this subject after I became acquainted at one of the Federation Meetings with a pathologist of Czech extraction, Dr. Zdeněk Hruban, who was a professor of pathology at the University of Chicago. We eventually became good friends and because of mutual scientific interests, with each of us having complimentary expertise, he in pathology and I in biochemistry, we decided to collaborate.
Fine Structure of Transplantable Hepatomas - The opportunity arose with the discovery of slow-growing tumors, hepatomas (1), which had rather unique biochemical properties and which were available in our laboratory. Transplanted hepatomas derived from the liver are of particular importance in the study of neoplasia. The biological and biochemical properties of these tumors can be compared in some respect with properties of adult, embryonic, and regenerating livers. Zdeněk Hruban was an expert on electron microscopy which gave us an opportunity to investigate the ultrastructure of several transplantable hepatomas in relation to their biochemical properties (2).
These studies included two fast-growing tumors, Novikoff and Morris 3683, two ethionone-induced tumors with intermediate growth rates, and two slow-growing tumors, Reuber H-35 and Morris 5123. The hepatomas were propagated by serial subcutaneous transplantation in appropriate strains of rats, 1 to 4 months old at the time of transplantation. Several generations of the same tumor were used, and biopsies of tumors of the same generation. Liver biopsies were also obtained from rats bearing large hepatomas. Tissue for studies was obtained under light ether anesthesia from regions grossly free of necrosis. The tissue samples were fixed in osmium tetroxide buffered with buffer and embedded in benzoyl peroxide as catalyst. Thin sections were cut with microtome and stained with lead Karnovsky method or with aqueous uranyl acetate. The sections embedded in methacrylate were coated with carbon. Selected areas were photographed in an RCA EMU electron microscope. The results showed that the fine structure of each hepatoma is characteristic. It does not change from transplant to transplant. Common to all hepatomas are large nucleoli, large Golgi complex, and the presence of numerous free ribosomes arranged in clusters. The fast-growing hepatomas have a simple organization of the cytoplasm. The slowgrowing hepatomas have better developed and more organized cytoplasmic membrane system, including well-developed rough and smooth endoplasmic reticulum. Stacks of rough cisternae are best developed in the slow-growing Reuber H-35 hepatoma. Similarly, mitochondria are smaller and have fewer cristae in the rapidly growing hepatomas and are larger and more hepatocyte-like in the slow-growing tumors. Nevertheless, the structures of slow-growing hepatomas never resemble that of normal hepatocytes in complexity. These variations suggest that certain cytoplasmic organelles respond in a coordinated fashion to factors associated with the rate of growth. Other characteristics, such as the absence of microbodies or the presence of fibrils, may be attributable to inherent differences in the genetic makeup or developmental history of the cell. Still other characteristics are more likely associated with indirect factors, working through the altered physiological condition of the tissue. Many abnormal characteristics of hepatoma cells can be mirrored by metabolic inhibitors, toxic substances, or deficiency states. Certain mitochondrial changes, such as the presence of only one limiting membrane in
some areas, and herniation of the mitochondrial matrix have also been seen in hepatocytes of rats fed protein-depleted diets. A series of cellular changes, such as the swelling of cisternae and vacuolar elevation of the nuclear envelope, are probably secondary indications of osmotic imbalance of the cell. Variation in cellular hydration is also probably responsible for the formation of “dark” or shrunken and “light” or swollen cells. The variability encountered in microbodies of the different hepatomas is of particular interest. This organelle is a characteristic component of hepatocytes of vertebrates, being in most cases somewhat smaller than mitochondria, with a moderately dense matrix confined by a single smooth limiting membrane. The presence of typical microbodies in tumors of the liver would therefore help to confirm the tumors hepatocellular origin. Microbodies are absent from both the Novikoff and Morris 3683 Hepatomas, which represents either a loss of ability to form microbodies, possibly in relation to the rapid growth of these tissues, or a nonhepatocytic origin of the tumors. In the four hepatomas studied, where microbodies are present, the hepatoma type can be specifically identified solely on the basis of the morphology of these inclusions. There is evidence that microbodies in normal hepatocytes contain certain enzymes, i.e., uricase, catalase, and D-amino acid oxidase. Correlation of enzyme activities, uricase, catalase, and D-amino acid oxidase, with the presence of microbodies shows that only hepatomas with a relatively high catalase activity contain microbodies. The activity of uricase also correlates well with the presence of the crystalloid. Little is known about the distribution of D-amino acid oxidase, but it apparently is absent in the two ethionine-induced tumors despite the presence of microbodies. The correlation between uricase and catalase and the structure of microbodies in hepatomas can serve as a model of the kind of combined biochemical and morphological approach to neoplastic tissues that should provide new insights into tumor biology.
Fine Structure of Opossum Prostate - Together with Zdeněk Hruban we also became intrigued with the complex structure of the prostate gland of opossum which could be an excellent model for the study of structure-function relationships. The prostate gland of the opossum is subdivided into three segments, each of which has a distinct color in life. Although earlier biochemical studies of this gland did not reveal any major differences from the prostates of various placental mammals, later investigations on the cytological level, however, showed that the prostate gland of the opossum is an unusually complex tissue, each segment having a characteristic cellular morphology and secretion.
We therefore had undertaken studies of the structure of the opossum prostate in which cytochemical observations were correlated with electron microscopy ( ). 550 The first segment contains two types of cells, one of which secretes an iron containing product. The second segment contains three cell types, each having a different type of secretory granule. The lumina of the third segment contain glycogen, fragments of lipoprotein membranes, one or more phosphatases and a peroxidase. The dark cells of the first segment and the cells in the middle zone of the third segment show apocrine secretion. These studies emphasize the extreme cytological complexity of the organ in which a number of specialized cell types contribute varied, specific cell secretions to the seminal plasma.
The results showed that the opossum prostate is indeed highly complex cytologically, containing six different secretory cell types, each possessing a recognizably different secretory product. Secretory granules in cells of the first segment contained numerous particles 60 A in diameter of high electron density. They are transferred to the lumen by apocrine secretion. Both granules and lumina contain iron. Segment two possessed three cell types. Type A cells have spherical secretory granules containing acid mucopolysaccharides. In type B cells, secretory granules are elongate and possess densely packed protein filaments, 80 A in diameter. Secretory granules in type C cells contain crystalloids of hexagonally packed cylinders with a spacing of 600 A. Third segment cells contain few organelles and have a broad irregular micro-villar border. Most of the cytoplasm is filled with glycogen particles. The secretion is apocrine. The lumen contains lipoprotein lamellae, probably derived from shed microvilli, and also membrane bound glycogen globules originating from the cytoplasm. The contents are positive for acid, diand tri-phosphatases and peroxidase. Catalase activity in the third segment is unusually high.
Unusual features of our findings were the secretion of iron containing granules by the dark cells in the first segment and membrane containing material by the cells of the third segment. The lumina of the third segment contained peroxidase and phosphatases.
The dark cells of the first segment form granules containing several enzymes, including acid phosphatase and iron particles. The high content of iron in this segment is probably responsible for its brown color. Since the first segment of the sexually immature opossum is pale, the iron accumulation is apparently related to sexual maturity. The release of iron particles apparently represents secretion of iron containing material.
Histologically demonstrable iron in macrophages of various other organs is probably of hemolytic origin but in the prostate, since iron levels are very much higher, its presence is probably associated with a specific physiological function. The physiological significance of iron secretion is unknown, but it may regulate the activity of enzymes in the seminal fluid either as a cofactor or by inhibitory or stimulatory effects. Iron is known to have an inhibitory effect on sulphoglutathione reductase and uricase, but it reverses the cysteine inhibition of aldolase. It is of interest that zinc, and copper have similar effects in some cases and that the zinc content of the opossum prostate is somewhat lower than found in other animals. Histologically demonstrable iron is also present in the mammary gland. It is possible, but less likely, that the high iron content is associated with the presence of cytochrome c, catalase or peroxidases.
The nature of the light cells in the first segment is not clear. They may represent dark cells which have extruded their secretion. The presence of characteristic secretory granules on the other hand suggests that these cells have a specific function; their granules morphologically resemble those described in the ventral lobe of the rat prostate.
The secretory granules of the C-cells of the second segment have a complex structure, possibly suggesting the presence of an enzyme or enzyme complex. The crystalline or purified enzymes uricase, pyruvate kinase and glutamic dehydrogenase have, in our experience, a tubular or cylindrical substructure. The dimensions of these crystals are, however, different from those of the cylinders seen in the granules of C cells. The possibility that these intragranular structures are of viral origin has also to be considered. We have found virus-like panicles in microvilli of the salivary gland of one opossum. Ccell granules of identical structure were found, however, in prostates of opossums obtained from three different localities. Microcylinders occurring within mitochondrial cristae in the rat pinealocytes have a very similar structure.
The secretory granules of B-cells may represent stored enzymes possibly related to the metabolism of glycogen of phospholipids secreted by the third segment. The seminal fluid is also known to contain proteolytic enzymes. The carbohydrate containing granules of A-cells may represent storage granules of material utilized in formation of the vaginal plug.
The third segment has abundant glycogen, demonstrable by staining techniques. The presence of glycogen in the other two segments also can be demonstrated biochemically. It is possible that the glycogen, liberated by apocrine secretion, and the phospholipid membranes, which are probably derived from cellular microvilli, or their breakdown products, are metabolized by the sperm. The lumen of this segment contains peroxidase, diphosphatases and ATP-ase. Biochemically demonstrable catalase in the third segment is unusually high. The histologically demonstrable peroxidase may be related to the high activity of catalase. Some enzymatic proteins are inactivated by peroxidase, particularly in the presence of uric acid. The peroxidase of the opossum prostate may perform a similar function in regard to the protein secretions of the second segment. Peroxidase has been demonstrated previously in the thyroid gland and associated with hormone metabolism.
The complex dense bodies seen in some cells were reminiscent of bodies found in the coagulating gland of the rat. The opossum does not have a separate coagulating gland and some of its functions may be performed by the prostate. Mixture of homogenates of individual prostatic segments, however, in our hands failed to produce coagulation. Other granules were closely attached to the smooth portions of rough cisternae in a fashion reminiscent of microbodies of mammalian liver. It is, however, surprising that, in spite of high catalase activity, only a few microbodies were found.
A striking difference between the fine structure of the opossum prostate and that of other species is the relatively poor development of the rough endoplasmic reticulum. It is possible that the prostate of the sexually inactive adult opossum is concerned primarily with storage of secretory products and its structure may be different following mating or in young animals.
Our findings suggest that the secretory products of the opossum prostate gland differ qualitatively from those of placentals. Whether these differences are common to all marsupials, or are only present in the primitive opossum, could not be answered by our study.
Endnotes
1. Liver tumors.
2. Z. Hruban, J. Martan and A. Slesser, D. F. Steiner, M. Lubran and M. Rechcigl, Jr., Fine Structure of the Prostatic Epithelium of the Opossum (Didelphis virginiana Kerr, J. Exp. Zool. 160, No. 1 (October 1965), pp. 81-106.
