298 R. J. GOODLOW, L. A. MIKA, AND W. BRAUN [VOL. 60 



The reasons for past failures to detect this phenomenon in nonsynthetic media 

 are still obscure. That the complexity of such media may be responsible is in- 

 dicated by the recent results of Schuhardt, Rode, and Oglesby (1949), who 

 demonstrated that many amino acids, including alanine, are toxic to Brucella, 

 whereas other amino acids may neutralize their effects. Whether such inter- 

 actions may have masked the influence of filtrates when complex media were 

 used remains questionable since population changes similar to those in simple 

 media regularly occur in complex media, even though they remain unaffected 

 by the addition of filtrates from old cultures. 



A neutralization of the toxic effects of alanine accumulation in Brucella cul- 

 tures, possibly with the help of specific antagonists, should be capable of pre- 

 venting population changes in smooth cultures during prolonged periods of 

 growth in liquid media. Actually, such interference with the accumulation of 

 metabolites favoring the establishment of mutants may have been practiced 

 unwittingly in prior experiments that were concerned with specific selective 

 agents that suppressed the establishment of nonsmooth types in smooth popula- 

 tions. Thus it has been demonstrated (Braun, 1949) that the addition of small 

 amounts of normal serum from Z?rwceZZa-susceptible hosts to smooth broth cul- 

 tures will prevent the establishment of nonsmooth types. Preliminary data sug- 

 gest that in such serum cultures the accumulation of alanine is actually reduced 

 or that in addition to alanine a different amino acid, possibly antagonistic to 

 alanine, may accumulate. Similarly, the prevention of population changes in 

 smooth Brucella cultures containing 0.03 m sodium pyrophosphate (Braun, 1950) 

 may involve an interference with enzymatic processes leading to alanine produc- 

 tion through the removal of trace metals that may act as catalysts for these 

 reactions. This possibility is further supported by the fact that the addition of 

 Mn++ or Mg++ to smooth cultures results in a considerable enhancement of the 

 establishment of nonsmooth types (Cole and Braun, 1950). 



Attempts are being made to determine the pathways that may be involved 

 in alanine production. At least three mechanisms may be suggested, namely: 

 (1) There may be transamination between aspartic and pyruvic acids yielding 

 oxalacetic acid and alanine. The pyruvic acid may be formed as an end product 

 in the oxidation of lactic acid initially present in the medium, whereas the hy- 

 drolysis of asparagine, also initially present, would yield aspartic acid and free 

 ammonia. (2) There may be direct amination of pyruvic acid with the free am- 

 monia to yield alanine. (3) There may be a combination of pathways (1) and 

 (2). 



It has been ascertained that the accumulation of alanine in smooth cultures 

 and the simultaneous establishment of alanine-resistant nonsmooth mutants 

 described for B. abortus occurs in cultures of B. suis, B. melitensis, and strain 

 19 of B. abortus as well. This phenomenon, therefore, appears to be of general 

 significance for brucellae. To what extent it may also apply to other bacterial 

 species remains to be determined. However, there are a number of observations, 

 reported in the recent literature, which indicate the metabolic accumulation of 

 specific amino acids and their toxicity for certain cell types. If tested by methods 



140 



