1 3.19. The control of the production of an end product P.

Mutants may be isolated which are resistant to the inhibitory effects of the analogue and, if the site of toxicity of the analogue is the mimicing of the control properties of the natural product, such mutants may overproduce the compound to which the analogue is analogous. To return to the example of the biosynthesis of P where P* is inhibitory due to its mimicing the control properties of P; a mutant may be isolated which may be capable of growing in the presence of P* due to the fact that the first enzyme in the pathway is no longer susceptible to inhibition by the analogue. The modified enzyme of the resistant mutant may not only be resistant to inhibition by the analogue but may also be resistant to the control effects of the natural end product, P, resulting in the uninhibited production of P. If the control system were the repression of enzyme synthesis, then the resistant mutant may be modified such that the enzyme synthesis machinery does not recognize the presence of the analogue. However, the site of resistance of the mutant may not be due to a modification of the control system; for example, the mutant may be capable of degrading the analogue, in which case the mutant would not be expected to overproduce the end product. Thus, analogue resistant mutants may be expected to overproduce the end product to which the analogue is analogous provided that:

(i) The toxicity of the analogue is due to its mimicing the control properties of the natural product.

(ii) The site of resistance of the resistant mutant is the site of control by the end product.

Resistant mutants may be isolated by exposing the survivors of a mutation treatment to a suitable concentration of the analogue in growth medium and purifying any colonies which develop. Sermonti (1969) described a method to determine the suitable concentration. The organism was exposed to a range of concentrations of the toxic analogue by inoculating each of a number of agar plates containing increasing levels of the analogue with 106 to 109 cells. The plates were incubated for several days and examined to determine the lowest concentration of analogue which allowed only a very few isolated colonies to grow, or completely inhibited growth. The survivors of a mutation treatment may then be challenged with the pre-determined concentration of the analogue on solid medium. Colonies which develop in the presence of the analogue may be resistant mutants.

Szybalski (1952) constructed a method of exposing the survivors of a mutation to a range of analogue concentrations on a single plate. Known as the gradient plate technique, it consists of pouring 20 cm3 of molten agar medium, containing the analogue, into a slightly slanted petri dish and allowing the agar to set at an angle. After the agar has set, a layer of medium not containing the analogue is added and allowed to set with the plate level. The analogue will diffuse into the upper layer giving a concentration gradient across the plate and the survivors of a mutation treatment may be spread over the surface of the plate and incubated. Resistant mutants should be detected as isolated colonies appearing beyond a zone of confluent growth, as indicated in Fig. 3.20. Whichever method is used for the isolation of analogue-resistant mutants, great care should be taken to ensure that the isolates are genuinely resistant to the analogue by streaking them, together with analogue-sensitive controls, on both analogue-supplemented and analogue-free media. The resistant isolates should then be screened for the production of the desired compound by overlayering them with a bacterial strain requiring the compound; producers may then be recognized by a halo of growth of the indicator strain.

Sano and Shiio (1970) investigated the use of lysine analogue-resistant mutants of Brevibacterium flauum for the production of lysine. The control of the biosynthesis of the aspartate family of amino acids in B. flauum is as illustrated for C. glutamicum in Fig. 3.14. The main control of lysine synthesis is the concerted feedback inhibition of aspartokinase by lysine and threonine. Sano and Shiio demonstrated that S-(2 ami-noethyl) cysteine (AEC) completely inhibited the growth of B. flauum in the presence of threonine, but only partially in its absence. Also, the inhibition by AEC and threonine could be reversed by the addition of lysine. This evidence suggested that the inhibitory effect of AEC was due to its mimicing lysine in the concerted inhibition of aspartokinase. AEC-resistant mutants were isolated by plating the survivors of a mutation treatment on minimal agar containing 1 mg cm~3 of both AEC and threonine. A relatively large number of the resistant isolates accumulated lysine, the best producers synthesizing more than 30 g dm-3. Investigation of the lysine producers indicated that their aspartokinases had been desensitized to the concerted inhibition by lysine and threonine.

The development of an arginine-producing strain of B. flauum by Kubota et al. (1973) provides an excellent example of the selection of a series of mutants resistant to increasing levels of an analogue. The control of the

___il Unsupplemented

— —— Medium supplemented __** " with analogue

Decreasing analogue concentration in the upper layer Side view of an agar plate prepared for the isolation of analogue-resistant mutants

Zone of 'no growth' due to the high level of toxic analogue

A possible resistant mutant

Confluent growth below the critical inhibitory level of the analogue

Surface view of a gradient plate after inoculation and incubation.

Fig. 3.20. The gradient plate technique for the isolation of analogue-resistant mutants.

biosynthesis of arginine in B. flauum is similar to that shown for C. glutamicum in Fig. 3.15. Kubota et al. selected mutants resistant to the arginine analogue, 2-thiazolealanine, and the genealogy of the mutants is shown in Fig. 3.21. Strain number 352 produced 25.3 g dm~3 arginine. Presumably, the mutants were altered in the susceptibility of the second enzyme in the pathway to inhibition by arginine.

A classic example of the rationale of analogue resistant mutant isolation is seen in the development of biotin overproducing strains. Currently, biotin is produced commercially by chemical synthesis, but considerable effort has been (and is being) expended to develop a competitive biotechnological process. The screening of natural isolates for their ability to accumulate biotin-vitamers led to the isolation of Bacillus sphaericus (Ogata et al, 1965). Repression by biotin was shown to be an important control mechanism and, thus, attempts were made to isolate mutants resistant to biotin analogues. Mutants of B. sphaericus resistant to acidomycin (ACM) and/or 5-(2-thienyl)-«-valeric acid (TVA) were capable of synthesizing up to 11 times (0.4 mg cm"3) the biotin level of the wild-type (Tanaka et al., 1988). More recently, Sakurai et al. (1993) isolated dual ACM/TVA resistant strains of Setratia marcescens capable of synthesizing 20 mg cm"3 biotin. The use of genetic engineering techniques to further develop these

Zone of 'no growth' due to the high level of toxic analogue

A possible resistant mutant

Confluent growth below the critical inhibitory level of the analogue

B. flavum ATCC 14067

| X-ray irradiation No. 33038 (guanine")

| NG treatment, selection with TA at 5 mg dm-3 No. 112 ¡guanine", TA resistant)

| NG treatment, selection with TA at 10 mg drrr3 No. 179 (guanine", TA resistant) L-arginine producer at 14.3 g drrr3

| Diethyl sulphate treatment No. 352 (guanine", TA resistant) L-arginine producer at 25.3 g drrr3

Fig. 3.21. The genealogy of L-arginine-producing mutants of B. flavum. TA, thiazolealanine; NG, W-methyl-/V'-nitro-A'-nitroso-guani-dine (Kubota et al., 1973).

strains is considerd in a later section of this chapter.

The second technique used for the isolation of mutants altered in the recognition of control factors is the isolation of revertant mutants. Auxotrophic mutants may revert to the phenotype of the mutant 'parent'. Consider the hypothetical pathway illustrated in Fig. 3.19 where P controls its own production by feedback inhibiting the first enzyme (a) of the pathway. A mutant does not produce the enzyme, a, and is, therefore, auxotrophic for P. However, a revertant of the mutant produces large concentrations of P. The explanation of the behaviour of the revertant is that, with two mutations having occurred at loci concerned with the production of enzyme a, the enzyme of the revertant is different from the enzyme of the original prototrophic strain and is not susceptible to the control by P. Rever-tants may occur spontaneously or mutagenic agents may be used to increase the frequency of occurrence, but the recognition of the revertants would be achieved by plating millions of cells on medium which would allow the growth of only the revertants, i.e. in the above example, on medium lacking P.

Shiio and Sano (1969) investigated the use of prototrophic revertants of B. flavum for the production of lysine. These workers isolated prototrophic revertants from a homoserine dehydrogenase-defective mutant. The revertants were obtained as small-colony forming strains and produced up to 23 g dm*3 lysine. The overproduction of lysine was shown to be due to the very low level of homoserine dehydrogenase in the revertants which, presumably, resulted in the synthesis of threonine and methionine in quantities sufficient for some growth, but insufficient to cause inhibition or repression.

Mutant isolation programmes for the improvement of strains producing primary metabolites have not relied on the use of only one selection technique. Most projects employed a number of methods including the selection of natural variants and the selection of induced mutants by a variety of means. The selection of bacteria overproducing threonine provides a good example of the use of a variety of selection techniques. Attempts to isolate auxotrophic mutants of C. glu-tamicum producing threonine were unsuccessful despite the fact that productive auxotrophic strains of Escherichia coli had been isolated. Huang (1961) demonstrated threonine production at a level of 2-4 g dm~3 by a diaminopimelate and methionine double auxotroph of E. coli. Kase et al. (1971) isolated a triple auxotrophic mutant of E. coli which required diaminopimelate, methionine and isoleucine and produced between 15 and 20 g dm *3 threonine. The control of the production of the aspartate family of amino acids in E. coli is shown in Fig. 3.22 and that in C. glu-tamicum in Fig. 3.14. The mechanism of control in E. coli involves a system of isoenzymes, three isoenzymic forms of aspartokinase and two of homoserine dehydrogenase, under the influence of different end products. However, in C. glutamicum control is effected by the concerted inhibition of a single aspartokinase by threonine and lysine; by the inhibition of homoserine dehydrogenase by threonine and the repression of homoserine dehydrogenase by methionine. Thus, the control of homoserine dehydrogenase may not be removed by auxotrophy without the loss of threonine production. However, in E. coli methionine auxotrophy would remove control of the methionine-sensitive homoserine dehydrogenase and aspartokinase which would allow threonine production, despite the control of the threo-nine-sensitive isoenzymes by threonine. E. coli mutants

Fig. 3.22. Control of the aspartate family of amino acids in Escherichia coli.

also lacking lysine and isoleucine would be relieved of the control of the lysine-sensitive aspartokinase and the degradation of threonine to isoleucine.

The production of threonine by C. glutamicum has been achieved by the use of combined auxotrophic and analogue resistant mutants. A good example of the approach is given by Kase and Nakayama (1972) who obtained stepwise improvements in productivity by the imposition of resistance to a-amino-j8-hydroxy valeric acid (a threonine analogue) and S-( /3-aminoethyl)-L-

cysteine (a lysine analogue) on a methionine auxotroph of C. glutamicum. The genealogy of the mutants is shown in Fig. 3.23. The analogue-resistant strains were shown to be altered in the susceptibility of aspartokinase and homoserine dehydrogenase to control, and the lack of methionine removed the repression control of homoserine dehydrogenase. The use of recombinant DNA technology has resulted in the construction of far more effective threonine producers and these strains are considered in a later section of this chapter.

C. glutamicum KY 9002 (wild type)

KY 10290 (met-, AHV resistant) Produced 0.6 g dm-3 threonine

KY 10484 (met-, AHV resistant) Produced 2.3 g dm~3 threonine

KY 10440 (met-, AHV resistant,

AEC resistant) Produced 9,5 g dm"3 threonine

KY 10184 (met-, AHV resistant) Produced 1.5 g dm-3 threonine

KY 10230 (met-, AHV resistant) Produced 3.7 g dm"3 threonine

KY 10251 (met-, AHV resistant,

AEC resistant) Produced 9.0 g dm~3 threonine and 5.5 g dm~3 lysine

Fig. 3.23. The genealogy of mutants of C. glutamicum producing L-threonine or l-threonine plus l-lysine. AHV, a-amino-0-hydroxy valeric acid: AEC, S-( j8-aminoethyl)-L-cysteine (Kase and Nakayama, 1972).

The development of strains producing guanosine in high levels provides a further example of the use of auxotrophs and analogue-resistant mutants in the same programme. Guanosine production has been achieved using auxotrophic and analogue resistant mutants of B. subtilis. The control of the production of purine nucleotides in B. subtilis is shown in Fig. 3.16 (Demain, 1978). The major points of control to be modified to achieve guanosine overproduction are:

(i) Removal of AMP inhibition of phosphoribosyl pyrophosphate amidino transferase by adenine auxotrophy.

(ii) Elimination of GMP reductase.

(iii) Bypassing the inhibition of IMP dehydrogenase by GMP.

This approach is demonstrated in the work of Shiio (cited by Demain, 1978) who obtained analogue-resistant mutants from an adenine auxotroph of B. subtilis. Mutants were isolated in two stages — first, for resistance to low levels of 8-azaguanine (an analogue of GMP) and then for resistance to high levels of the analogue. The resulting mutant produced 9 g dm 3 guanosine due to the removal of control of phosphoribosyl pyrophosphate amidinotransferase and IMP dehydrogenase.

The isolation of induced mutants producing improved yields of secondary metabolites where directed selection is difficult to apply

The discussion so far has considered the isolation of mutants producing products whose biosynthesis and control have been sufficiently understood to prepare 'blueprints' of the desirable mutants which have enabled the construction of suitable selection procedures. In contrast, important secondary metabolites were being produced long before their biosynthetic pathways, and certainly the control of those pathways, had been elucidated. Thus, strain improvement programmes had to be developed without this fundamental knowledge which meant that they depended on the random selection of the survivors of mutagen exposure. Elander and Vournakis (1987) described these techniques as "hit or miss methods that require brute force, persistence and skill in the art of microbiology". However, despite the limited knowledge underlying these approaches they were extremely effective in increasing the yields of antibiotics, as illustrated in Table 3.5 (Riviere, 1977).

Table 3.5. Improvement of antibiotic yields during the first 20 years of antibiotic development (Riviere, 1977)


Initial yield at

Improved yield in

Vinegar For Your Health

Vinegar For Your Health

A resource for the many ways you can use Vinegar to improve your health! In today's society of miracle medicine, we often overlook things that have been around hundreds of years! Things like Vinegar!

Get My Free Ebook

Post a comment