254 ^- ^- KREBS VOL. 4 (iQSO) 



lie processes in isolated material which have so far been studied have usually been found 

 to be of the same order in serum and in suitable saline media, at least for the usual 

 experimental periods of under two hours. But some tissues, in particular brain, retina, 

 choroid plexus, and foetal membranes, assume an opaque appearance on incubation in 

 saline and tend to break up into fragments whilst appearance and texture remain 

 unchanged in serum. The use of serum may therefore be advantageous in some investi- 

 gations. 



The earlier serum substitutes, such as Ringer's solution, were designed on an 

 empirical basis. Ringer" tested the effect of various saline media on the beat of the 

 isolated frog heart, and found that solutions containing certain quantities of Ca and 

 K ions, in addition to NaCl maintained the beat for longer periods than NaCl solutions. 

 Later, when precise data on the chemical composition of blood serum became available, 

 saline media were modelled on these data^^' ^^' ^"' ''^. It has been found repeatedly that 

 the closer the medium resembles serum the better does it maintain tissue activities 

 in vitro. The previous attempts to copy the composition of serum, however, considered 

 only the inorganic constituents and glucose. 



The saline medium of Krebs and Henseleit'^^ closely reproduces the inorganic 

 constituents of mammalian serum except that the concentration of CI is about 20% 

 higher. A discrepancy of this kind is unavoidable in a purely inorganic medium because 

 in serum a fraction of the anions, amounting to about 22 milliequivalents, consists of 

 organic substances. Replacement of part of the NaCl by the Na salts of pyruvic (or 

 L-lactic), fumaric and glutamic acids and addition of glucose eliminates the discrepancy 

 in the chloride concentration and introduces the 'relevant' metabolites. The following 

 composition is suggested for the saline serum substitute. Mix 



1. 80 parts of 0.9% NaCi (0.154 M) 



2. 4 parts of 1.15% KCl (0.154 M) 



3. 3 parts of o.ii M CaCl., 



4. I part of 2.11% KH2P64 (0.154 M) 



5. I part of 3.82% MgS04.7H20 



6. 21 parts of 1.3% NaHCOg (0.154 M) ; treated with CO2 until pn is 7.4 



7. 4 parts of 0.16 M Na-pyruvate (or L-lactate) 1 Prepared by 



8. 7 parts of o.i M Na-fumarate 

 g. 4 parts of 0.16 M Na-L-glutamate 



10. 5 parts of 0.3 M (5.4%) glucose 



neutralizing a 



solution of the acids with M 



NaHCOg solution 



The mixture must be saturated with a gas mixture containing about 5% COg. The 

 stock solutions are approximately isotonic. 



Solutions 7 to 10, unless sterilized, cannot be kept at room temperature. In the 

 refrigerator they keep for about a week if gross bacterial infections are avoided. 



Solutions I to 6 are mixed in the same proportion as the medium of Krebs and 

 Henseleit'^1, except that 80 parts NaCl solution instead of 100 parts are used. The 

 difference of 20 ml is made up by the solutions 7 to 10. The concentrations of the con- 

 stituents of this medium are shown in Table II. For comparison, data for human and 

 rat sera are also given and it will be seen that the concentration of the electrolytes in 

 the sera and the 'serum substitute' are very similar. 



Sera of different mam.Tialian species show relatively small variations except in the 

 case of inorganic sulphate. Normal human serum is reported to contain i to 1.5 mg SO4 

 References p. 26y—26g. 



