 Otto Warburg, "On The Origin of Cancer Cells," SCIENCE, 
 (24FEB1956), Volume 123, Number 3191, pp. 309-314.

 ----------------------------------------------------------------

 Professor Warburg is director of the Max Planck Institute for 
 Cell Physiology, Berlin-Dahlem, Germany.  This article is based 
 on a lecture delivered at Stuttgart on 25 May 1955 before the 
 German Central Committee for Cancer Control.  It was first 
 published in German [Naturwissenschaften 42, 401 (1955)].  This 
 translation was prepared by Dean Burk, Jehu Hunter, and W. H. 
 Everhardy of the U.S. Department of Health, Education, and 
 Welfare, Public Health Service, National Institutes of Health, 
 Bethesda, Md., with permission of Naturwissenschaften and with 
 collaboration of Professor Warburg, who has introduced 
 additional material.
 ----------------------------------------------------------------


   Our principal experimental object for the measurement of the 
 metabolism of cancer cells is today no longer the tumor but the 
 ascites cancer cells (1) living free in the abdominal cavity, 
 which are almost pure cultures of cancer cells with which one 
 can work quantitatively as in chemical analysis.  Formerly, it 
 could be said of tumors, with their varying cancer cell content, 
 that they ferment more strongly the more cancer cells they 
 contain, but today we can determine the absolute fermentation 
 values of the cancer cells and find such high values that we 
 come very close to the fermentation values of wildly 
 proliferating Torula yeasts.

   What was formerly only qualitative has now become 
 quantitative. What was formerly only probably has now become 
 certain.  The era in which the fermentation of the cancer cells 
 or its importance could be disputed is over, and no one today 
 can doubt that we understand the origin of cancer cells if we 
 know how their large fermentation originates, or, to express it 
 more fully, if we know how the damaged respiration and the 
 excessive fermentation of the cancer cells originate.

 Energy of Respiration and Fermentation

   We now understand the chemical mechanism of respiration and 
 fermentation almost completely, but we do not need this 
 knowledge for what follows, since energy alone will be the 
 center of our consideration.  We need to know no more or 
 respiration and fermentation here than that they are energy-
 producing reactions and that they synthesize the energy-rich 
 adenosine triphosphate through which the energy of respiration 
 and fermentation is then made available for life.  Since it is 
 known how much adenosine triphosphate can be synthesized by 
 respiration and how much by fermentation, we can write 
 immediately the potential, biologically utilizable energy 
 production of any cells if we have measured their respiration 
 and fermentation.  With the ascites cancer cells of the mouse, 
 for example, we find an average respiration of 7 cubic 
 millimeters of oxygen consumed per milligram, per hour, and 
 fermentation of 60 cubic millimeters of lactic acid produced per 
 milligram, per hour.  This, converted to energy equivalents, 
 means that the cancer cells can obtain approximately the same 
 amount of energy from fermentation as from respiration, whereas 
 the normal body cells obtain much more energy from respiration 
 than from fermentation.  For example, the liver and the kidney 
 of an adult animal obtain about 100 times as much energy from 
 respiration as from fermentation.

   I shall not consider aerobic fermentation, which is a result 
 of the interaction of respiration and fermentation, because 
 aerobic fermentation is too labile and too dependent on external 
 conditions. Of importance for the considerations that follow are 
 only the two stable independent metabolic processes, respiration 
 and anaerobic fermentation - respiration, which is measured by 
 the oxygen consumption of cells that are saturated with oxygen, 
 and fermentation, which is measured by the formation of lactic 
 acid in the absence of oxygen.

 Injuring of Respiration

   Since the respiration of all cancer cells is damaged, our firm 
 question is, How can the respiration of body cells be injured?  
 Of this damage to respiration, it can be said at the outset that 
 it must be irreversible, since the respiration of cancer cells 
 can never returns to normal.  Second, the injury to respiration 
 must not be so great that the cells are killed, for then no 
 cancer cells could result.  If respiration is damaged when it 
 forms too little adenosine triphosphate, it may be either that 
 the oxygen consumption has been decreased or that, with 
 undiminished oxygen consumption, the coupling between 
 respiration and the formation of adenosine triphosphate has been 
 broken, as was first pointed out by Feodor Lynen (2).

   One method for the destruction of the respiration of body 
 cells is removal of oxygen.  If, for example, embryonal tissue 
 is exposed to an oxygen deficiency for some hours and then is 
 placed in oxygen again, 50 percent more or more of the 
 respiration in destroyed.  The cause of this destruction of 
 respiration is lack of energy.  As a matter of fact, the cells 
 need their respiratory energy to preserve their structure, and 
 if respiration is inhibited, both structure and respiration 
 disappear.

   Another method for destroying respiration is to use 
 respiratory poisons.  From the standpoint of energy, this method 
 comes to the same result as the first method.  No matter whether 
 oxygen is withdrawn from the cell or whether the oxygen is 
 prevented from reacting by a poison, the result is the same in 
 both cases - namely, impairment of respiration from lack of 
 energy.
   I may mention a few respiratory poisons.  A strong, specific 
 respiratory poison is arsenious acid, which as every clinician 
 knows, may produce cancer.  Hydrogen sulfide and many of its 
 derivatives are also strong, specific respiratory poisons.  We 
 know today that certain hydrogen sulfide derivatives thiourea 
 and thioacetamide, with which citrus fruit juices have been 
 preserved in recent times, induce cancer of the liver and gall 
 bladder in rats.

   Urethane is a nonspecific respiratory poison.  It inhibits 
 respiration as a chemically indifferent narcotic, since it 
 displaces metabolites from cell structures.  In recent years it 
 has been recognized that subnarcotic does of urethane cause lung 
 cancer in mice in 100 percent of treatments.  Urethane is 
 particularly suitable as a carcinogen, because in contrast to 
 alcohol, it is not itself burned up on the respiring surfaces 
 and, unlike ether or chloroform, it does not cytolize the cells.  
 Any narcotic that has these properties may cause cancer upon 
 chronic administration in small doses.

   The first notable experimental induction of cancer by oxygen 
 deficiency was described by Goldblatt and Cameron (3), who 
 exposed heart fibroblasts in tissue culture to intermittent 
 oxygen deficiency for long periods and finally obtained 
 transplantable cancer cells, whereas in control cultures that 
 were maintained without oxygen deficiency, no cancer cells 
 resulted.  Clinical experiences along these lines are 
 innumerable: the production of cancer by intermittent irritation 
 of the outer skin and of the mucosa of internal organs, by the 
 plugging of the excretory ducts of glands, by cirrhoses of 
 tissues, and so forth.  In all these cases, the intermittent 
 irritations lead to intermittent circulatory disturbances.  
 Probably chronic intermittent oxygen deficiency plays a greater 
 role in the formation of cancer in the body than does the 
 chronic administration of respiratory poisons.

   Any respiratory injury due to lack of energy, however, whether 
 it is produced by oxygen deficiency or by respiratory poisons, 
 must be cumulative, since it is irreversible.  Frequent small 
 doses of respiratory poisons are therefore more dangerous than a 
 single large dose, where there is always the chance that the 
 cells will be killed rather than that they will become 
 carcinogenic.

 Grana

   If an injury of respiration is to produce cancer, this injury 
 must, as already mentioned, be irreversible.  We understand by 
 this not only that the inhibition of respiration remains after 
 removal of the respiratory poison but, even more, that the 
 inhibition of respiration also continues through all the 
 following cell divisions, for measurements of metabolism in 
 transplanted tumors have shown that cancer cells cannot regain 
 normal respiration, even in the course of many decades, once 
 they have lost it.

   This originally mysterious phenomenon has been explained by a 
 discovery that comes from the early years of cell physiology 
 (4).  When liver cells were cytolized by infusion of water and 
 the cytolyzate was centrifuged, it was found that the greater 
 part of the respiration sank to the bottom with the cell grana.  
 It was also shown that the respiration of the centrifuged grana 
 was inhibited by narcotics at concentrations affecting cell 
 structures, from which it was concluded -already in 1914- that 
 the respiring grana are not insoluble cell particles but 
 autonomous organisms, a result that has been extended in recent 
 years by the English botanist Darlington (5) and particularly by 
 Mark Woods and H.G. du Buy (6) of the National Cancer Institute 
 in Bethesda, Md.  Woods and du Buy have experimentally expanded 
 our concepts concerning the self-perpetuating nature of 
 mitochondrial elements (grana) and have demonstrated the 
 hereditary role of extranuclear aberrant forms of these in the 
 causation of neoplasia. The autonomy of the respiring grana, 
 both biochemically and genetically, can hardly be doubted today.

   If the principle Omne granum e grano is valid for respiring 
 grana, we understand why the respiration connected with grana 
 remains damaged when it has once been damaged;  it is for the 
 same reason that properties linked with genes remain damaged 
 when the genes have been damaged.

   Furthermore, the connection of respiration with the grana (7) 
 also explains carcinogenesis that I have not mentioned 
 previously, the carcinogenesis by x-rays.  Rajewsky and Pauly 
 have recently shown that the respiration linked with the grana 
 can be destroyed with strong doses of x-rays, while the small 
 part of the respiration that takes place in the fluid protoplasm 
 can be inhibited very little by irradiation.  Carcinogenesis by  
 x-rays is obviously nothing else than destruction of respiration 
 by elimination of the respiring grana.

   It should also be mentioned here that grana, as Graffi has 
 shown (8), fluoresce brightly if carcinogenic hydrocarbons are 
 brought into their surroundings, because the grana accumulate 
 the carcinogenic substances.  Probably this accumulation is the 
 explanation for the fact that carcinogenic hydrocarbons, 
 although almost insoluble in water, can inhibit respiration and 
 therefore have a carcinogenic effect.

 Increase of Fermentation

   When the respiration of body cells has been irreversibly 
 damaged, cancer cells by no means immediately result.  For 
 cancer formation there is necessary not only an irreversible 
 damaging of the respiration but also an increase in the 
 fermentation -- indeed, such an increase of the fermentation 
 that the failure of respiration in compensated for 
 energetically.  But how does this increase of fermentation come 
 about?

   The most important fact in this field is that there is no 
 physical or chemical agent with which the fermentation of cells 
 in the body can be increased directly;  for increasing 
 fermentation, a long time and many cell divisions are always 
 necessary.  The temporal course of this increase of fermentation 
 in carcinogenesis has been measured in many interesting works, 
 among which I should like to make special mention of those of 
 Dean Burk (9).

   Burk first cut out part of the liver of healthy rats and 
 investigated the metabolism of the liver cells in the course of 
 ensuing regeneration, in which, as is well known, the liver 
 grows more rapidly than a rapidly growing tumor.  No increase of 
 fermentation was found. Burk then fed rats for 200 days on 
 butter yellow, whereupon liver carcinomas were produced, and he 
 found that the fermentation slowly increased in the course of 
 200 days toward values characteristic of tumors.

   The mysterious latency period of the production of cancer is, 
 therefore. nothing more than the time in which the fermentation 
 increases after a damaging of the respiration.  This time 
 differs in various animals;  it is essentially long in man and 
 here often amounts to several decades, as can be determined by 
 the cases in which the time of the respiratory damage is known -
 - for example, in arsenic cancer and irradiation cancer.

   The driving force of the increase of fermentation, however, is 
 the energy deficiency under which the cells operate after 
 destruction of their respiration which forces the cells to 
 replace the irretrievably lost respiration energy in some way.  
 They are able to do this by a selective process that makes use 
 of the fermentation of the normal body cells.  The more weakly 
 fermenting body cells perish, but the more strongly fermenting 
 ones remain alive, and this selective process continues until 
 respiratory failure is compensated for energetically by the 
 increase in fermentation.  Only then has a cancer cell resulted 
 from the normal body cell.

   Now we understand why the increase in fermentation takes such 
 a long time and why it is possible only with the help of many 
 cell divisions. We also understand why the latency period is 
 different in rats and in man.  Since the average fermentation of 
 normal rat cells is much greater that the average fermentation 
 of normal human cells, the selective process begins at a higher 
 fermentation level in the rat and, hence is completed more 
 quickly than it is in man.

   It follows from this that there would be no cancers if there 
 were no fermentation of normal body cells, and hence we should 
 like to know, naturally, from where the fermentation of the 
 normal body cells stems and what its significance is in the 
 body.  Since, as Burk has shown, the fermentation remains almost 
 zero in the regenerating liver growth, we must conclude that the 
 fermentation of the body cells has nothing to do with normal 
 growth.  On the other hand, we have found tat the fermentation 
 of the body cells is greatest in the very earliest stages of 
 embryonal development and that it then decreases gradually in 
 the course of embryonal development.  Under these conditions, it 
 is obvious --since ontogeny is the repetition of phylogeny-- 
 that the fermentation of body cells is the inheritance of 
 undifferentiated ancestors that have lived in the past at the 
 expense of fermentation energy.

 Structure and Energy

   But why -- and this is our last question -- are the body cells 
 differentiated when their respiration energy is replaced by 
 fermentation energy?  At first, one would think that it is 
 immaterial to the cells whether they obtain their energy from 
 respiration or from fermentation, since the energy of both 
 reactions is transformed into the energy of adenosine 
 triphosphate, and yet adenosine triphosphate=adenosine 
 triphosphate.  This equation is certainly correct chemically and 
 energetically, but it is incorrect morphologically, because, 
 although respiration takes place for the most part in the 
 structure of the grana, the fermentation enzymes are found for a 
 greater part in the fluid protoplasm.  The adenosine 
 triphosphate synthesized by respiration therefore involves more 
 structure than the adenosine triphosphate synthesized by 
 fermentation.  Thus, it is as if one reduced the same amount of 
 silver on a photographic plate by the same amount of light, but 
 in one case with diffused light and in the other with patterned 
 light.  In the first case, a diffuse blackening appears on the 
 plate, but in the second case, a picture appears; however, the 
 same thing happens chemically and energetically in both cases.  
 Just as one type of light energy involves more structure than 
 the other type, the adenosine triphosphate energy involves more 
 structure when it is formed by respiration than it does when it 
 is formed by fermentation.

   In any event, it is one of the fundamental facts of present-
 day biochemistry that adenosine triphosphate can be synthesized 
 in homogeneous solutions with crystallized fermentation enzymes, 
 whereas so far no one has succeeded in synthesizing adenosine 
 triphosphate in homogeneous solutions with dissolved respiratory 
 enzymes, and the structure always goes with oxidative 
 phosphorylation.

   Moreover, it was known for a long time before the advent of 
 crystallized fermentation enzymes and oxidative phosphorylation 
 that fermentation --the energy supplying reaction of the lower 
 organisms-- is morphologically inferior to respiration.  Not 
 even yeast, which is one of the lowest forms of life, can 
 maintain its structure permanently by fermentation alone;  it 
 degenerates to bizarre forms.  However, as Pasteur showed, it is 
 rejuvenated in a wonderful manner, if it comes in contact with 
 oxygen for a short time.  "I should not be surprised," Pasteur 
 said in 1876 (10) in the description of these experiments, if 
 there should arise in the mind of an attentive hearer a 
 presentiment about the causes of those great mysteries of life 
 which we conceal under the words youth and age of cells."  
 Today, after 80 years, the explanation is as follows:  the 
 firmer connection of respiration with structure and the looser 
 connection of fermentation with structure.

   This, therefore, is the physiochemical explanation of the 
 dedifferentiation of cancer cells.  If the structure of yeast 
 cannot be maintained by fermentation alone, one need not that 
 highly differentiated body cells lose their differentiation upon 
 continuous replacement of their respiration with fermentation.

   I would like at this point to draw attention to a consequence 
 of practical importance.  When one irradiates a tissue that 
 contains cancer cells as well as normal cells, the respiration 
 of the cancer cells, already too small, will decline further.  
 If the respiration falls below a certain minimum that the cells 
 need unconditionally, despite their increased fermentation, they 
 die;  whereas the normal cells, where respiration may be harmed 
 by the same amount, will survive because, with a greater initial 
 respiration, they will still possess a higher residual 
 respiration after irradiation.  This explains the selective 
 killing action of of x-rays on cancer cells.  But still further:  
 the descendants of the surviving normal cells may in the course 
 of the latent period compensate the respiration decrease by the 
 fermentation increase and, thence, become cancer cells.  Thus it 
 happens that radiation which kills cancer cells can also at the 
 same time produce cancer or that urethane, which kills cancer 
 cells, can also at the same time produce cancer.  Both events 
 take place from harming respiration:  the killing, by harming an 
 already harmed respiration;  the carcinogenesis by the harming 
 of a not yet harmed respiration.

 Maintenance Energy

   When differentiation of the body cells has occurred and cancer 
 cells have thereby developed, there appears a phenomenon to 
 which our attention has been called by the special living 
 conditions of ascites cancer cells.  In extensively progressed 
 ascites cancer cells of the mouse, the abdominal cavity contains 
 so many cancer cells that the latter cannot utilize their full 
 capacity to respire and ferment because of the lack of oxygen 
 and sugar.  Nevertheless, the cancer cells remain alive in the 
 abdominal cavity, as the result of transplantation proves.

   Recently, we have confirmed this result by direct experiments 
 in which we placed varying amounts of energy at the disposal of 
 the ascites outside the body, in vitro, and then transplanted 
 it.  This investigation showed that all cancer cells were killed 
 when no energy at all was supplied for 24 hours at 38 degrees C 
 but that one-fifth of the growth energy was sufficient to 
 preserve the transplantability of the ascites.  This result can 
 also be expressed by saying that cancer cells require much less 
 energy to keep them alive than they do for growth.  In this they 
 resemble other lower cells, such as yeast cells, which remain 
 alive for a long time in densely packed packets -- almost 
 without respiration and fermentation.

   In any case, the ability of cancer cells to survive with 
 little energy, if they are not growing, will be of great 
 importance for the behaviour of the cancer cells in the body.

 Sleeping Cancer Cells

   Since the increase in fermentation in the development of 
 cancer cells takes place gradually, there must be a transitional 
 phase between normal body cells and fully formed cancer cells.  
 Thus, for example, when fermentation has become so great that 
 dedifferentiation has commenced, but not so great that the 
 respiration defect has been fully compensated for energetically 
 by fermentation, we may have cells which indeed look like cancer 
 cells but are still energetically insufficient.  Such cells, 
 which are clinically not cancer cells, have lately been found, 
 not only in the prostate, but also in the lungs, kidney, and 
 stomach of elderly persons.  Such cells have been referred to as 
 "sleeping cancer cells." (11,12)

   The sleeping cancer cells will possibly play a role in 
 chemotherapy.  From energy considerations, I could think that 
 sleeping cancer cells could be killed more easily than growing 
 cancer cells in the body and that the most suitable test objects 
 for finding effective agents would be the sleeping cells of the 
 skin -- that is, precancerous skin.

 Summary

   Cancer cells originate from normal body cells in two phases.  
 The first phases is the irreversible injuring of respiration.  
 Just as there are many remote causes of plague --heat, insects, 
 rats-- but only one common cause, the plague bacillus, there are 
 a great many remote causes of cancer --tar, rays, arsenic, 
 pressure, urethane-- but there is only one common cause into 
 which all other causes of cancer merge, the irreversible 
 injuring of respiration.

   The irreversible injuring of respiration is followed, as the 
 second phase of cancer formation, by a long struggle for  
 existence by the injured cells to maintain their structure, in 
 which a part of the cells perish from lack of energy, while 
 another part succeed in replacing the irretrievably lost 
 respiration by fermentation energy.  Because of the 
 morphological inferiority of fermentation energy, the highly 
 differentiated body cells are converted by this into 
 undifferentiated cells that grow wildly -- the cancer cells.

   To the thousands of quantitative experiments on which these 
 results are based, I should like to add, as a further argument, 
 the fact that there is no alternative today.  If the explanation 
 of a vital process is its reduction to physics and chemistry, 
 there is today no other explanation for the origin of cancer 
 cells, either special or general. From this point of view, 
 mutation and carcinogenic agent are not alternatives, but empty 
 words, unless metabolically specified.  Even more harmful in the 
 struggle against cancer an be the continual discovery of 
 miscellaneous cancer agents and cancer viruses, which, by 
 obscuring the underlying phenomena, may hinder necessary 
 preventive measures and thereby become responsible for cancer 
 cases.


 TECHNICAL CONSIDERATIONS AND COMMENTS

 Metabolism of the ascites cancer cells.

   The high fermentation of ascites cancer cells was discovered 
 in Dahlem in 1951 (12) and since then has been confirmed in many 
 works (13,14) For best measurements, the ascites cells are not 
 transferred to Ringer's solution but are maintained in their 
 natural medium, ascites serum, which is adjusted physiologically 
 at the beginning of the measurement by addition of glucose and 
 bicarbonate.  Because of the very large fermentation, it is 
 necessary to dilute the ascites cells that are removed from the 
 abdominal cavity rather considerably with ascites serum;  
 otherwise the bicarbonate would be used up within a few minutes 
 after addition to the glucose, and hence the fermentation would 
 be brought to a standstill.

 Under physiological conditions of pH and temperature, we find 
 the following metabolic quotients in ascites serum (15):

   Q(O2)    = -5 to -10
   Q(M)(O2) = 25 to 35
   Q(M)(N2) = 50 to 70

 where Q(O2) is the amount of oxygen in cubic millimeters that 1 
 milligram of tissue (dry weight) consumes per hour at 38 degrees 
 C with oxygen saturation, Q(M)(O2) is the amount of lactic acid 
 in cubic millimeters that 1 milligram of tissue (dry weight) 
 develops per hour at 38 degrees C in the absence of oxygen.

 Even higher fermentation quotients have been found in the United 
 States with other strains of mouse ascites cancer cells (13,14).

   All calculations of the energy-production potential of cancer 
 cells should now be based on quotients of the ascites cancer 
 cells, since these quotients are 2 or 3 times as large 
 anaerobically as the values formerly found for the purest solid 
 tumors.  The quotients of the normal body cells, however, remain 
 as they were found in Dahlem in the years from 1924 to 1929 (16-
 19).  It is clear that the difference in metabolism between 
 normal cells and cancer cells is much greater than it formerly 
 appeared to be on the basis of measurements of solid tumors.

 Utilizable energy of respiration and fermentation

   Since the discovery of the oxidation reaction of fermentation 
 in 1939 (20), we have known the chemical reactions by which 
 adenosine diphosphate is phosphorylated to adenosine 
 triphosphate in fermentation;  and since then we have found that 
 1 mole of fermentation lactic acid produces 1 mole of adenosine 
 triphosphate (ATP).

   The chemical reactions by which ATP is synthesized in 
 respiration are still unknown, but it can be assumed, according 
 to the existing measurements (21), that 7 moles of ATP can be 
 formed when 1 mole of oxygen is consumed in respiration.

 ATP quotients.

 If we multiply Q(O2) by 7 and Q(M)(N2) by 1, we obtain the 
 number of cubic millimeters of ATP that 1 milligram of tissue 
 (dry substance) can synthesize per hour (22,400 cubic 
 millimeters=1 millimole of ATP).  We call these quotients 
 Q(ATP)(O2) and Q(ATP)(N2), according to whether the ATP is 
 formed by respiration or by fermentation, respectively.

 Energy production of cancer cells and normal body cells.

 In Table 1, the Q values of some normal body cells are 
 contrasted with the Q values of our ascites cancer cells.

 The cancer cells have about as much energy available as the 
 normal body cells, but the ratio of the fermentation energy to 
 the respiration energy is much greater in the cancer cells than 
 it is in the normal cells.

 ----------------------------------------------------------------
 Table 1.  Contrast of the Q values of some normal body cells 
 with the  Q values of ascites cancer cells.
 ----------------------------------------------------------------
 Cells    Q(O2)  Q(M)(N2)  Q(ATP)O2 Q(ATP)N2    Q(ATP)O2+Q(ATP)N2
 Liver      -15    1         105      1           106
 Kidney     -15    1         105      1           106
 Embryo (*) -15    25        105      25          130
 Cancer     -7     60        49       60          109

 * young
 ----------------------------------------------------------------

 Uncoupling of respiration.

   If a young rat embryo is transferred from the amniotic sac to 
 Ringer's solution, the previously transparent embryo becomes 
 opaque and soon appears coagulated (17).  At the same time, the 
 connection between respiration and phosphorylation is broken;  
 that is, although oxygen is still consumed and carbon dioxide is 
 still developed, the energy of this combustion process is lost 
 for life.  If the metabolism quotients had previously been

     Q(O2) = 15, Q(M)(O2) = O, Q(M)(N2) = 25
     Q(ATP)(O2) = 105, Q(ATP)(N2) = 25

 in the amniotic fluid, afterward in Ringer's solution they are

     Q(O2) = -15, Q(M)(O2) = 25, Q(M)(N2) = 25
     Q(ATP)(O2) = O, Q(ATP)(N2) = 25

 Because of uncoupling of respiration and phosphorylation, the 
 energy production of the embryo has fallen from Q(ATP)(O2) + 
 Q(ATP)(N2) = 130, to 25;  since the uncoupling is irreversible, 
 the embryo dies in the Ringer's solution.

   This example will show that the first phase of carcinogenesis, 
 the irreversible damaging of respiration, need not be an actual 
 decrease in the respiration quotient but merely an uncoupling of 
 respiration, with undiminished over-all oxygen consumption.  
 Ascites cancer cells, which owe their origin primarily to an 
 uncoupling of respiration, could conceivably have the following 
 metabolism quotients, for example:

     Q(O2) = -50, Q(M)(O2) = 100, Q(M)(N2) = 100
     Q(ATP)(O2) = O, Q(ATP)(N2) = 100

 which would mean that, despite great respiration, the usable 
 energy production would be displaced completely toward the side 
 of fermentation.  One will now have to search for such cancer 
 cells among the ascites cancer cells.  Solid tumors --and 
 especially solid spontaneous tumors-- need no longer be 
 subjected to such examinations today, of course, since the solid 
 tumors are usually so impure histologically.

 Aerobic fermentation

 Aerobic fermentation is a property of all growing cancer cells, 
 but aerobic fermentation [p. 313 -->] without growth is a 
 property of damaged body cells -- for example, embryos that have 
 been transferred from amniotic fluid to Ringer's solution.  
 Since it is always easy to detect aerobic fermentation but 
 generally difficult to detect growth, or lack thereof, of body 
 cells, aerobic fermentation should not be used as a test for 
 cancer cells, as I made clear in 1928 (19).

 Nevertheless, misuse is still made of aerobic fermentation.  
 Thus, O'Connor (22) recently repeated our old experiments on the 
 aerobic fermentation of the embryo that has been transferred in 
 to Ringer's solution, but he drew the conclusion that the growth 
 of normal body cells is completed at the expense of the aerobic 
 fermentation, even though it has long been established that the 
 embryo does not ferment aerobically when it grows in the 
 amniotic fluid.

 Respiratory poisons.

   The specific respiration-inhibiting effect of arsenious acid 
 and the irreversibility of its inhibitions were discovered in 
 the first quantitative works on cell respiration (23,24).  There 
 is abundant literature on the carcinogenesis by arsenic, 
 particularly on arsenic cancer after treatment of psoriasis and 
 on the cancer of grape owners who spray their vineyards with 
 arsenic.  The specific respiration-inhibiting effect of hydrogen 
 sulfide has likewise been described by Negelein (25), and 
 carcinogenesis by derivatives of hydrogen sulfide has been 
 recently described by D. N. Gupta (26).

   The irreversible inhibition of cell respiration by urethane 
 was discovered early (27) as well as the fact that the urethane 
 inhibition is more irreversible, the higher the temperature.  In 
 sea urchin eggs, the effect of urethane was investigated, not 
 only on the metabolism, but also on cell division in studies 
 (28) from which the later urethane treatment of leukemia was 
 developed.  The physiochemical mechanism by which urethane and 
 other indifferent narcotics inhibit cell respiration was cleared 
 up in 1921 (29).  Only much later did the carcinogenic effect of 
 urethane become known.  Actually, multiple lung adenomas can 
 often be produced in 100 percent of the mice treated with small 
 doses of urethane (30).

 Oxygen deficiency.

   Short-period oxygen deficiency irreversibly destroys the 
 respiration of embryos (16) without thereby inhibiting the 
 anaerobic fermentation of the embryo.  If such embryos are 
 transplanted, teratomas are formed (31).  It has recently been 
 reported that, in the development of the Alpine salamander, 
 malformations occurred when the respiration was inhibited by 
 hydrocyanic acid in the early stages of embryonal development 
 (32).

   Goldblatt and Cameron (3) reported that, in the in vitro 
 culturing of fibroblasts, tumor cells appeared when the cultures 
 were exposed to intermittent oxygen deficiency for long periods, 
 whereas, in the control cultures, no tumor cells appeared.  In 
 the discussion at the Stuttgart convention, Lettre cited against 
 Goldblatt and Cameron the fact that another American tissue 
 culturist, Earle, had occasionally obtained tumor cells from 
 fibroblasts for reasons unknown to him and in an unreproduceable 
 manner, but this objection does not seem weighty, and the latter 
 part is untrue (33).  In any event, here is an area in which the 
 methods of tissue culture could prove useful for cancer 
 research.  But warnings must be given against metabolism 
 measurements in tissue cultures, if and when the tissue cultures 
 are mixtures of growing and dying cells, especially under 
 conditions of malnutrition. An example of the latter type of 
 confusion is involved in the discussion by Albert Fischer (34), 
 especially in the chapter "Energy exchange of tissue cells 
 cultivated in vitro."

 Rous agent

   If the Rous agent is inoculated into the chorion of chick 
 embryos, tumors originate in the course of a few days -- as 
 rapidly as the transplantation of cancer cells.  The tumors 
 formed are not chorion tumors but Rous sarcomas.  The Rous 
 agent, to which a particle weight of 150 million is ascribed at 
 present, is therefore capable of transmitting the morphological 
 properties of the Rous sarcoma;  and whatever we call the Rous 
 agent -- "hereditary unit," cell fragment, microcell, or spore -
 - the transmission of the Rous sarcoma by the Rous agent is, in 
 any case, nothing more than a transplantation and is to be 
 differentiated strictly from the production of a chicken sarcoma 
 by methylcholanthrene, which is a neoformation of a tumor from 
 normal body cells and as such takes a long time.

   The metabolism of the chicken sarcomas, whether produced by 
 the Rous agent or by methylcholanthrene, is the same and does 
 not differ in any way from the metabolism of the tumors of other 
 animals (35).  In the first case, however, the fermentation 
 potential has been transplanted with the Rous agent, whereas in 
 the second case, the fermentation has been intensified by 
 selection from normal body cells under the action of 
 methylcholanthrene.

 Addendum: in vitro Carcinogenesis and metabolism

   Since this paper was prepared, striking confirmation and 
 extension of its main conclusions have been obtained from 
 correlated metabolic and growth studies of two lines of tissue 
 culture cancer cells of widely differing malignancy that were 
 both derived from one and the same normal, tissue-culture cell 
 (36).  The single cell as isolated some 5 years ago from a 97-
 day old parent culture of a strain C3H/He mouse by Sanford, 
 Likely and Earle (33) of the National Cancer Institute.  Up to 
 the time that the single-cell isolation was made, no tumors 
 developed when cells of the parent culture were injected into 
 strain C3H/He mice. Injections of in vitro cells of the lines 
 1742 and 2049 (formerly labelled substrains VII and III, 
 respectively) first produced tumors in normal C3H/He mice after 
 the 12th and 19th in vitro transplant generations, respectively;  
 after 1.5 years, the percentage production of sarcomas was 63 
 and 0 percent, respectively;  and after 3 years, it was 97 and 1 
 percent, respectively,with correspondingly marked differences in 
 length of induction period.  Despite such gross differences in 
 "malignancy" in vivo, the rates of growth of the two lines of 
 cells maintained continuously in vitro have remained nearly 
 identical and relatively rapid.  Nevertheless, the metabolism of 
 the two lines of cancer cells, whose malignancy was developed in 
 vitro, has been found by Woods, Hunter, Hobby, and Burk to 
 parallel strikingly the differences in malignancy observed in 
 vivo, in a manner in harmony with the predications and 
 predictions of this article.

   The metabolic values were measured following direct transfer 
 of the liquid cultures from the growth flasks into manometric 
 vessels, without notable alteration of environmental 
 temperature, pH, or medium composition (horse serum, chick 
 embryo extract, glucose, bicarbonate, balanced saline).  The 
 values obtained this accurately represent the metabolism of 
 growing, adequately nourished, pure lines of healthy cancer 
 cells free of admixture with any other tissue cell type.  The 
 anaerobic glycolysis of the high-malignancy line 1742 was Q(M)N2 
 = 60 to 80, which is virtually maximum for any and all cancer 
 cells previously reported, including ascites cells (12-14).  The 
 anaerobic glycolysis of the low-malignancy line was, however, 
 only one-third as great, Q(M)N2 = 20 to 30.  The average aerobic 
 glycolysis values for the two lines were in the same order, 
 Q(M)O2 = 30 and 10, respectively, but of lower magnitude because 
 of the usual, pronounced Pasteur effect, greater in line 1742 
 than in line 2049 [Q(M)N2 - Q(M)O2 = about 40 and 15].  On the 
 other hand, the rates of oxygen consumption were in converse 
 order, being smaller in line 1742 [Q(O2) = 5 to 10] than in line 
 2049 [Q(O2) = 10 to 15], corresponding to a greater degree of 
 respiratory defect in line 1742.  The respiratory defect in both 
 lines was further delineated by the finding of little or no 
 increase in respiration after the addition of succinate to 
 either line of cells, in contrast to the considerable increases 
 obtained with virtually all normal tissues (9);  and the 
 respiratory increase with paraphenylenediamine was likewise 
 relatively low, compared with normal tissue responses.

   A further notable difference between the two cell lines was 
 the very much lower inhibition of glycolysis by podophyllin 
 materials (anti-insulin potentiators) observed with line 1742 
 compared with line 2049 (for example, 10 and 70 percent, 
 respectively, at a suitably low concentration).  This result 
 would be expected on the basis of the much greater loss of anti-
 insulin hormonal restraint of glucose metabolism, at the 
 hexokinase phosphorylating level, as the degree of malignancy is 
 increased, just as was reported for a spectrum of solid tumors 
 (14).

   Finally, the high-malignancy line 1742 cells have been found 
 by A. L. Schade to contain 3 times as much aldolase as the low-
 malignancy line 2049 cells (11,300 versus 3700 Warburg activity 
 units per millimeter of packed cells extracted), and about 2 
 times as much a-glycerophosphate dehydrogenase [2600 versus 1400 
 Schade activity units (13) per millimeter of packed cells 
 extracted].  The potential significance of these indicated 
 enzymic differences in relation to the parallel glycolytic 
 differences, measured with aliquots of the same cell cultures, 
 is evident, and may well be connected with the corresponding 
 hexokinase system differences.

   The new metabolic data on the two remarkably contrasting lines 
 of cancer cells, which originated from a single, individual cell 
 and have been maintained exclusively in vitro over a period of 
 years, epitomize and prove finally the main conclusions of this 
 article, which are based on decades of research.  Such metabolic 
 analyses provide promise of a powerful tool for diagnosis of 
 malignancy in the ever-increasing variety of tissue culture 
 lines now becoming available in this rapidly expanding 
 biological and medical field, where characterization of 
 malignancy by conventional methods (animal inoculation or 
 otherwise) may be difficult or impracticable.  This metabolic 
 tool should be especially important in connection with the use 
 of tissue cultures for the evaluation of chemotherapeutic agents 
 or other control procedures.

 REFERENCES AND NOTES

 1.   The transplantable ascites cancer was discovered by H. 
      Loewenthal and G. Jahn [Z. KREBSFORSCH, 37, 439 (1932)].  
      G. Klein (Stockholm) expanded our knowledge about the 
      physiology and morphology of the ascites tumors and showed 
      their great advantages as experimental material.  [EXPTL. 
      CELL RESEARCH, 2, 518 (1951)]

 2.   F. Lynen, NATURWISSENSCHAFTEN 30, 398 (1942); ANN. CHEM. 
      JUSTUS LIEBIG 573, 60 (1951).

 3.   H. Goldblatt and G. Cameron, J. EXPTL. MED. 97, 525 (1953).

 4.   O. Warburg, PFLUGER'S. ARCH. GES. PHYSIOL. 154, 599 (1913); 
      158, 19 (1914).

 5.   C. D. Darlington, BRIT. J. CANCER 2, 118 (1948).

 6.   M. W. Woods et al., J. NATL. CANCER INST, 11, 1105 (1951); 
      9, 311 (1949); 9, 325 (1949); 16, 351 (1955).  SCIENCE 102, 
      591 (1945); 111, 572 (1950).  AAAS RESEARCH CONF. ON CANCER 
      (Science Press, Lancaster, Pa., 1945), p. 162.  J. WASH. 
      ACAD. SCI. 42, 169 (1952).  PIGMENT CELL GROWTH, M. Gordon, 
      Ed. (Academic Press, New York, 1953), p. 335.  BIOCHEM. ET 
      BIOPHYS. ACTA 12, 329 (1953). PROC. AM. ASS. CANCER 
      RESEARCH 1, 7 (1954).  PROC. SOC. EXPTL. BIOL. MED. 83, 62 
      (1953).  PHYTOPATHOLOGY 33, 637, 766 (1943); 36, 472 
      (1946); 31, 978 (1941); 32, 288 (1942).  AM. J. BOTANY 33, 
      12a (1946); 38, 419 (1951).

 7.   A compilation of American works on the grana, in which my 
      results of 1914 (4) have been confirmed, is given by G. 
      Hogeboom, W. Schneider, and M. Striebich in CANCER RESEARCH 
      [13, 617 (1953)].  In a very special case -- nucleated red 
      blood cells of birds, which contain no grana or only poorly 
      visible ones -- the entire respiration can be centrifuged 
      off with the cell nuclei [O. Warburg, HOPPE-SEYLER'S Z. 
      PHYSIOL. CHEM. 70, 413 (1913)].

 8.   A. Graffi, Z. KREBSFORSCH, 49, 477 (1939).

 9.   D. Burk, SYMPOSIUM ON RESPIRATORY ENZYMES (University of 
      Wisconsin Press, Madison, 1942), p. 235.  J. G. Kidd, R. J. 
      Winzler, D. Burk, CANCER RESEARCH 4, 547 (1944).

 10.  L. Pasteur, ETUDES SUR LA BIERE (Masson, Paris, 1876), p. 
      240.

 11.  J. Craigie, J. PATHOL. BACTERIOL. 63, 177 (1951); 64, 251 
      (1952). H. Hamperl, VERHANDL. DEUT. GES. PATHOL. 35, 29 
      (1951).

 12.  O. Warburg and E. Hiepler, Z. NATURFORSCH 7b, 193 (1952).

 13.  A. L. Schade, BIOCHIM. ET. BIOPHYS. ACTA 12, 163 (1953).

 14.  M. Woods, J. Hunter, D. Burk, J. NATL. CANCER. INST. 16, 
      351 (1955).

 15.  I am indebted to Georg Klein of the Karolinska Institute,
      Stockholm, Sweden, for his Ehrlich strain of mouse ascites 
      cells.

 16.  O. Warburg, K. Posener, E. Negelein, BIOCHEM. Z. 152, 309 
      (1924).

 17.  E. Negelein, IBID, 165, 122 (1925).

 18.  O. Warburg et al., IBID. 189, 114, 175, 242 (1924); 193, 
      315 (1928); 197, 175 (1928); 204, 475, 479 (1929).

 19.  O. Warburg, IBID. 204, 482 (1929).

 20.  O. Warburg et al., IBID. 303, 40, 132 (1939).

 21.  H. A. Krebs et al., BIOCHEM. J. LONDON 54, 107 (1953).

 22.  R. J. O'Connor, BRIT. J. EXPTL. PATHOL.  31, 390 (1951).

 23.  O. Warburg, HOPPE-SEYLER'S Z. PHYSIOL. CHEM. 70, 413 
      (1911);  M. Onaka, IBID. 70, 433 (1911); 71, 193 (1911).

 24.  K. Dresel, BIOCHEM. Z. 178, 70 (1926).

 25.  E. Negelein, IBID.  165, 203 (1925).

 26.  D. N. Gupta, NATURE 175, 257 (1955).

 27.  R. Usui, PFLUGER'S ARCH. GES. PHYSIOL. 147, 100 (1912).

 28.  O. Warburg, HOPPE-SEYLER'S Z. PHYSIOL. CHEM. 66, 305 
      (1910); 70, 413 (1911).

 29.  ----------, BIOCHEM.  119, 143 (1921).

 30.  C. D. Larsen, J. NATL. CANCER INST. 8, 63 (1947).

 31.  O. Warburg, BIOCHEM. Z. 228, 257 (1930).

 32.  H. Tiedemann and H. Tiedemann, Z. NATURFORSCH. 9b, 371 
      (1954).

 33.  K. K. Sanford, G. D. Likely, W. R. Earle, J. NATL. CANCER 
      INST. 15, 215 (1954).

 34.  A. Fischer, BIOLOGY OF TISSUE CULTURE (Copenhagen, Denmark, 
      1946).

 35.  O. Warburg, BIOCHEM. Z.  160, 307 (1925);  D. Burk et al.,  
      J. NATL. CANCER INST.  2, 201 (1941).

 36.  This summary of studies of various collaborative groups of
      investigators was prepared by Dean Burk at Professor 
      Warburg's request.

