The ability of the producing strain to maintain its high productivity during both culture maintenance and a fermentation is a very important quality. Yield decay during culture storage may be avoided by the use of maintenance techniques such as those discussed earlier in this chapter, but loss of productivity during the fermentation is far more difficult to control. A decrease in the productivity of a commercial strain is normally due to the occurrence of lower-yielding, spontaneous revertant mutants which frequently have a higher growth rate than the high-producing parent, so that yield decay is especially problematical in long-term fermentations such as fed-batch or continuous culture where the faster-growing, lower producer may predominate, or even replace, the high-producing original strain. This situation is illustrated by the commercial, amino acid-producing organisms, many of which are insufficiently stable to be used in continuous-culture processes. Many workers have attempted to control the stability of amino acid-producing strains. As may be seen from the previous section on the isolation of mutants overproducing primary metabolites the amino acid producers tend to be auxotrophs, analogue-resistant mutants or revertants; such mutations removing the normal mechanisms controlling the production of amino acids. The introduction of more than one mutation giving the same phenotype may give a more stable strain since all the mutations would have to revert for the strain to lose its productivity.
Woodruff and Johnson (1970) selected a double auxotrophic mutant of Micrococcus glutamicus requiring both homoserine and threonine and compared its ly-sine-producing properties with those of a homoserine auxotroph. The authors claimed that the double mutant had a two-fold advantage in that it produced higher levels of lysine compared with the single auxotroph and was also less susceptible to reversion to low productivity, the stability of the double auxotroph being such that it was a suitable organism for the production of lysine by continuous culture.
Nakayama (1972) discussed the problem of strain reversion in the lysine fermentation and cited an example of a fermentation where 87% of the cells from a 60-88-hour culture were revertants. This very high level of revertants was probably due to the faster growth rate of the revertant compared with the single auxotroph in homoserine-limited culture. This situation was controlled, to a certain extent, by the incorporation of antibiotics, such as erythromycin, which are more inhibitory to the rapidly growing revertants than to the homoserine limited auxotrophs. However, the use of mutants which have multiple markers appears to be a better solution, as this considerably reduces the probability of reversion to the wild type as reversion of all the markers must occur, which should be an extremely rare event. Sano and Shiio (1970) cited the use of a C. glutamicum mutant for the industrial production of lysine which was auxotrophic for homoserine and leucine and was resistant to S-(2-aminoethyl)-L-cy-steine and produced 39 g dm*3 lysine. This strain would have had to have several reversions before it was restored to anything approaching the wild-type.
The stability of fungal diploids used in commercial fermentations may be controlled by a technique discussed by Ball (1973). Ball claimed that it may be possible to control the degeneration of a diploid into haploids by incorporating non-homologous recessive lethal mutations on separate chromosomes of an homologous pair in the diploid. The deleterious effects of the mutations would be repressed by the dominant alleles in the diploid but would be expressed in a haploid derived from the diploid, resulting in any haploids being non-viable.
A very simple but effective technique for selecting stable strains of P. chrysogenum was used in the Panlabs strain development programme (Lein, 1986). Final evaluation of a culture was made using the second slant of a slant-to-slant transfer. If the culture was unstable then the yield of a fermentation from the second slant would be poor, resulting in it being rejected. This procedure was followed sequentially through the programme with the result that the later strains showed less tendency to degenerate after subculturing.
Recombinant plasmids used as vectors in genetic manipulation are susceptible to two types of instability. Segregational instability is due to uneven partitioning of the plasmids at replication resulting in the production of plasmid-free daughter cells. Structural instability is a result of recombination events occurring within the vector, sometimes resulting in disruption of the desired gene. It is essential for acceptable stability that a plasmid vector contains a partitioning locus and that segregation of the plasmid is not simply a random event. The problem has been overcome in some systems by the integration of the vector into the chromosomal DNA, as already seen for filamentous fungal vectors. The disadvantage of this approach is that only one (or a few) copy of the desired gene is present whereas the use of an autonomously replicating plasmid would result in many gene copies.
Was this article helpful?