36a On Respiratory Impairment in Cancer Cells' 

 By Dean Burk und Arthur L. Schade 



A mass of experimental data published diiring the past 33 years has established 

 the phenomenon of a metabolism characteristic of living Cancer cells, including 

 high anaerobic and aerobic glycolysis (formation of lactic acid from glucose) and an 

 impaired respiration 1 , 2 - The impaired respiration may involve any combination — 

 usually at least three — of the following experimental quantities readily measurable 

 under appropriate conditions: (i) a high ratio of glycolysis to respiration; (ii) a low 

 absolute value for oxygen consumption (<2o 2 ); (iü) an inefficient or uncoupled 

 respiration; or (iv) a low paraphenylenediamine (succinate) oxidative response 3 . 

 This list could be expanded. 



Respiratory impairment in living Cancer cells, first discovered by Otto War- 

 burg in 1923, is an experimental fact, and not, as described by Weinhouse 4 , 5 a 

 hypothesis based on "essentially fallacious reasoning". Statements of Weinhouse 

 indicating that the author of "Burk's extensive tables" has at any time ever denied 

 the factual existence of an impaired respiration in Cancer cells are categorically 

 untrue. 



The overwhelming evidence for the occurrence of various forms of respiratory 

 impairment in neoplastic cells, available in some 1000 experimental papers, but 

 denied by Weinhouse, obviously cannot be recapitulated in this note. It is, how- 

 ever, possible to outline here decisive errors in the most tangible support offered 

 by Weinhouse for his central, underlying view that "by and large, a representative 

 group of tumors absorb oxygen about as rapidly as a comparable group of nonneo- 

 plastic tissues . . . oxygen consumption is not quantitavely diminished" 6 . Curious- 

 ly, this longdiscarded but now exhumed view, which specifically denies a low<2o 2 

 in most Cancer cells (impairment ii), is alleged to derive its main support from 

 "Burk's extensive tables" 7 . Weinhouse 4 ' (p 271) has selectively Condensed parts 

 of these extensive tables, on some incompletely defined basis, and arrived at the 

 following average tissue slice<2o 2 values for three groups of tissues, each of which 

 has a wide spread of individual values: 15 types of malignant tumors, — 11.8; seven 

 growing tissues (three types), — 9.7; 14 nongrowing tissue types, — 9.3. In short, 

 by these statistics, the tumors would appear to have an average ("by and large") 

 respiration that is not lower, but even higher, than that of either the growing or 

 nongrowing tissues selected as "comparable". 



But such average Ooo values are meaningless as they stand, both statistically and 

 otherwise. They have not been adjusted for important species differences among 

 the three groups, for medium effects, for cellular impurity in the tumors (normal 

 tissues, ordinarily being 100 percent nonneoplastic, need no such adjustment), for 

 differences of technique among the different investigators cited, and so forth. A 

 remarkable error is Weinhouse's inclusion, in the already small group of growing 



* Republished from Science 124 (1956): 270 by permission. 



