422

CHEMISTR Y: CARSON A ND R UBEN

PROC. N. A. S.

4Ruben, Hassid and Kamen, Jour. Am. Chem. Soc., 61, 661 (1939). ' Ruben, Kamen, Hassid and DeVault, Science, 90, 570 (1939). s Additional evidence to be published shortly. 7 McAlister, Jour. Gen. Physiol., 22, 613 (1939). 8 Emerson and Lewis, Am. Jour. Botany, 26, 808 (1939). 9 Cf. van Niel, Ann. Rev. Biochem., 6, 606 (1937); Gaffron, Ibid., 7, 986 (1939).

C2 A SSIMILA TION BY PROPIONIC ACID BACTERIA STUDIED

Communicated May 15, 1940

The studies of Wood and Werkman1 2, 3 as well as the work of Phelps, Johnson and Peterson4 have shown that propionic acid bacteria can utilize CO2 during the fermentation of glycerol. In the absence of CO2 this fermentation can be adequately represented by the equation:5 CH20H CH3

-ICHOH

= CH2 + H20

CH20H COOH In the presence of CO2 the formation of propionic acid is accompanied by the appearance of succinic acid in amounts closely equimolar with the quantity of absorbed CO2. This made it seem possible that CO2 becomes converted into succinic acid by combination with a 3-carbon compound. The formation of succinic acid during the fermentation of pyruvate, dextrose and galactose by Escher. coli, particularly its dependence upon the CO2 partial pressure,6 supports
this view. It is apparent that important information regarding the mechanism through which CO2 is utilized can be obtained by the use of radioactive CO2.7 We have employed this approach in a study of the fermentation of glycerol by Propionibacterium pentosaceum. The bacteria were grown anaerobically in yeast extract-glycerol media in the presence of CO2 for 3 to 6 days. For the experiments cells from 250 to 500 ml. of such cultures were centrifuged, washed and suspended in 0.5 per cent phosphate buffer at pH 7.0 with and without added substrates. The suspensions were shaken at 30C. in the presence of N2 and C*028 for

VOL. 26, 1940

CHEMISTRY: CARSON AND RUBEN

423

30 to 40 minutes. After centrifuging off the cells, propionic acid and succinic acid were added to the supernatant liquid to provide carriers for the small quantities of radioactive metabolic products. The volatile and nonvolatile acids were separated by exhaustive steam distillation; the cells were boiled gently with acid for a few minutes and separated from the acid extract by centrifugation. Special care was taken to remove all traces of dissolved C*02 before radioactivity measurements were made. The results of an experiment carried out in the presence of 2 per cent glycerol are summarized in table 1.
TABLE 1

BACTERIA + 2 PER CENT GLYCEROL + C*O2 FOR 40 MINUTES

72.0 Volatile acids 10.0 Non-volatile acids 17.5 Boiled cell extract 0.5 Cells after extraction All measurements are corrected for decay and are therefore comparable.

The above table shows that 72 per cent of the C*02 taken up by the cells can be recovered in the volatile acid fraction and only 10 per cent in the non-volatile acid fraction. The boiled cell extract consisted mainly of acids carried down with the cells. The activity in the volatile fraction was found to be restricted to propionic acid by a Duclaux distillation. The C* content of successive distillates is compared in table 2 with the titration values obtained with pure propionic acid distilled under the same conditions.
TABLE 2
PER CBNT

PER CENT VOLUMB OF DISTILLATE

PER CENT RADIOACTIVITY FOUND

ACIDITY BY TITRATION FOR PURB PROPIONIC ACID

20.0 40.0 60.0 80.0 100.0

23.0 45.0 65.0 84.0 100.0

24.5 46.7 67.0 85.1 100.0

The agreement between the radioactivity and titration measurements is remarkably good. This result excludes the presence of more than 5%o of C* in other fatty acids (formic, acetic, etc.). The radioactive component of the non-volatile acid fraction was identified as succinic acid in two ways. The distribution coefficients of the non-

424

CHEMISTRY: CARSONAND RUBEN

PROC N. A. S.

volatile material between water and ethyl ether by titration and radioactivity measurements are compared in table 3.
TABLE 3

DISTRIBUrION COEFFICIENTS (10 MI. Aqueous Phase + 90 Ml. Ether)

7.4

6.0

In addition, the non-volatile material was sublimed and the specific ac-

tivity (

\succinic acid

. ) of the various fractions is shown in table 4.

SUBLIMATION
SPBCIFIC ACTIVITY

(RADIOACTIVITY
FRACTION \SUCCNIC

ACID)

1 (210-2200C.) 2 (220-240C.)

9.5 9.9

No radioactive material came off below 210C. It seems quite reasonable to conclude that the radioactivity is mainly due to succinic acid. The remarkable result that so much of the radioactive carbon is found in propionic acid might be explained by assuming a reversible reaction CH8CH2COOH + CO2 a COOHCH2CH2COOH. Experiments were performed to test this hypothesis by the addition of propionic acid and succinic acids to cell suspensions in the presence of C*O2. In order to reduce the amount of glycerol and reserve products in the cell material, the bacteria were centrifuged from the medium in which they were grown, and suspended in phosphate buffer for 3 hours; this procedure was then repeated, and immediately before exposure to C*02 they were again washed. The results of this experiment are shown in table 5.
TABLE 5

CELL SUSPENSION IN PHOSPHATE BUFFER EXPOSED TO C*Os FOR 30 MINUTES

SUBSTRATE

C* ASsILATBD (ARBITRARY UNITS)

RADIOACTIVITY OF VOLATILB ACID NON-VOL. ACID

Nothing
0.4 per cent Na propionate 0. 2 per cent Na succinate 2. O per cent glycerol

5.3 4.8 4.5 100.0

0.7 0.78 0.2 3.1

In a further experiment the cells were suspended in phosphate buffer for 7 hours and then centrifuged and washed as above. The results are summarized in table 6.

VL2610 CHEMISTRYY: CARSON AND R UBEN VOL. 26, 1940

425

TABLE 6

CELL SUSPENSION IN PHOSPHATE BUFFER EXPOSED SUBSTRATE Nothing 2 per cent glycerol 2 per cent glycerol + 0.4 per cent Na propionate

TO

C*O2 FOR 30 MINUTES

TOTAL C* ASSIMATBD

(AtBITRARY UNITS) 2.3

VOLATILB ACID
NON-VOL.

ACID

0.41
3.0,3.8 2.4

The addition of propionic acid in the absence of glycerol has little if any effect on the distribution of the C* between the propionic and succinic acids or on the total C* taken up. This is evidence against the formation of succinic acid from CO2 and propionic acid. The presence of glycerol not only has an enormous effect on the C*02 uptake, but also increases the ratio of radioactive carbon in propionic acid and succinic acid. It seems quite likely in view of these results that propionic and succinic acids are formed via reactions between CO2 and glycerol or intermediate products arising during its fermentation. Pyruvic acid, which could be a possible intermediate, was found to be inactive (< 0.5%). The data so far presented could find an explanation on the basis of a mechanism such as:

glycerol >- C-acidl d (intermediate product)

propionic acid

11+ Co2
C4-dibasic acid z succinic acid
In this case one would expect the radioactive carbon atom to be located in the carboxyl groups. Further experiments are in progress to test this hypothesis. The results of the above experiments are of general interest also in connection with a major problem encountered in tracer experiments with labeled carbon, namely, the synthesis of radioactive molecules starting with C*O2. In many cases synthetic organic methods result in poor yields despite time-consuming and painstaking effort. In such instances the use of appropriate micro6rganisms to perform the desired synthesis may be highly desirable. We may cite the above experiments as an example. In 30 minutes the bacteria converted over 80 per cent of the assimilated C*02 into propionic and succinic acids, which are thus made readily available for tracer experiments. It is a pleasure to thank Dr. M. D. Kamen and Professor E. 0. Lawrence

426

CHEMISTR Y: H. A. BARKER, ET AL.

PROC. N. A. S.

of the Radiation Laboratory for their interest, and for making the radioactive carbon available. We are indebted to Professor C. B. van Niel and Professor H. A. Barker for much advice and assistance. 1 Biochem. Jour., 30, 48 (1936). 2 Ibid., 32, 1262 (1938).
3Ibid., 34, 7 (1940). 4Biochem. J., 33, 726 (1939). 6 C. B. van Niel, The Propionic Acid Bacteria, Haarlem (1928). 6 Elsden, Biochem. Jour., 32, 187 (1938).
of some of its atoms. The isotope is C" and was produced in the Berkeley cyclotron; it has a half-life of 21 minutes.
7 Ruben, Hassid and Kamen, Jour. Am. Chem. Soc., 61, 661 (1939). 8 The symbol C* will be used to represent carbon labeled by the radioactivity

THE REDUCTION OF RADIOACTIVE CARBON DIOXIDE BY METHANE-PRODUCING BACTERIA By H. A. BARKER, S. RUBEN AND M. D. KAMEN
DIVISION OF PLANT NUrRITION, DEPARTMENT OF CHEmMISTRY, RADIATION LABORATORY, UNIVERSIIY OF CALIFORNIA

Communicated May 15, 1940

Several years ago' it was shown that the formation of methane in the fermentation of ethyl and butyl alcohols and of butyric acid by impure cultures of methane-producing bacteria is accompanied by the disappearance of an equi-molecular quantity of carbon dioxide. In the fermentation of ethyl alcohol, for example, the reaction can be rather closely described by the equation: (1) 2C2H5OH + C02 -2CH3COOH + CH4 The same reaction has recently been shown2 to be carried out by pure cultures of Methanobacterium Omelianskii. As to the mechanism of the above reaction, it seemed quite likely that the acetic acid arises by an oxidation of ethyl alcohol, carbon dioxide being simultaneously reduced to methane. This would be in accordance with the view originally advanced by Professor C. B. van Niel that the methane fermentation of organic as well as of inorganic compounds is a process of oxidation by means of carbon dioxide which can be represented by the general equation: 4H2A + C02 -4A + CH4 + 2H20 (2) where H2A is an oxidizable molecule and A is its oxidation product. Conclusive proof in favor of this mechanism, however, was lacking. Furthermore, no direct experimental evidence could be obtained in support of the