Regulation of glutamine synthetase and ITS role in citric acid fermentation by aspergillus niger

Abstract

Glutamine synthetase (L Glutamate:ammonia ligase EC 6.3.1.2), which catalyzes the synthesis of L glutamine from L glutamate, ATP and NH , plays a central role in the nitrogen metabolism of all living organisms. Elegant and detailed studies on bacterial glutamine synthetases have established that this enzyme is under rigorous cellular control. Intensive investigations have established that the enzyme is regulated by repression, rapid inactivation, covalent modification, feedback inhibition, etc. Although divalent metal ions have been implicated in the regulation of this enzyme, to date, there is very little evidence indicating the role of divalent cations in vivo.

In addition to its role in the regulation of nitrogen metabolism, GS functions as a part of the connecting link between carbon and nitrogen metabolism. Consequently, it can be visualized that derangement in this link could affect carbon metabolism. A special case of altered carbon metabolism which has attracted considerable attention, especially in view of its industrial importance, is the excretion of citric acid by A. niger. Based on the following observations, namely, the growth of the organism at low pH, the requirement of Mn² for GS activity and the cessation of growth before citric acid production, and accumulation of glutamate and metabolites derived from it, it was postulated that inhibition of GS would result in the derangement of nitrogen metabolism leading to a back up of the citric acid cycle causing an accumulation of citric acid.

This investigation was aimed at testing the above hypothesis by measuring GS levels and its regulation under a wide variety of growth conditions such as in different carbon and nitrogen sources and under limiting availability of metal ions as well as at different pH values; purifying the enzyme to homogeneity; establishing the mechanisms of its regulation by metal ions, end product inhibition; locating the amino acid residues at the active site and finally, establishing the operation of these types of mechanisms under in vivo conditions for the production of citric acid.

The materials and methods pertaining to the growth of the organism, estimations of various metabolites, purification and characterization of the enzyme, etc. are described in Chapter II.

When the fungus was grown on an easily assimilable nitrogen source such as NH or glutamine, GS levels were very low, whereas on poor nitrogen sources like NO and glutamate, a 3–5 fold increase in the enzyme levels was obtained (Table 2). These results suggested that the enzyme levels were regulated by the availability of nitrogen sources.

200 6. The enzyme appeared to be repressed by NH as shown by the low levels of GS activity when the fungus was grown on NH , and the enzyme content was dependent on the concentration of NH (Table 4). At higher concentrations of NH (100 mM), decreased levels of GS were observed. GS obtained from cells grown on different nitrogen sources e.g., NO , glutamate and glutamine was identical in its electrophoretic mobility (Figure 7), gel filtration profile on Bio Gel A5M (Figure 8), ratio of GS S Mg:GS T Mn (Table 2) suggesting that the enzyme in this organism was probably not regulated by covalent modification, rapid in vivo inactivation and proteolysis or by dissociation into oligomeric states as observed in the case of bacteria and yeasts.

The enzyme from glutamate grown A. niger mycelia was purified by protamine sulfate treatment, ammonium sulfate fractionation, DEAE Sephacel chromatography, affinity chromatography on AMP Sepharose and gel filtration on Sepharose 4B column. This procedure (Table 6) resulted in a 50 fold purification with 8% recovery. A. niger GS was extremely labile and considerable protection was provided by inclusion of 5% glycerol in all the buffers during purification and storage (Figure 13).

The homogeneity of the enzyme was indicated by the presence of a single band corresponding to the enzyme activity on polyacrylamide gel electrophoresis (Figure 10).

201 Molecular weight of the enzyme (Figure 11) was determined to be approximately 385,000 ± 25,000. The enzyme was octameric and shown to possess identical subunits of molecular weight 53,000 ± 5,000 (Figure 12).

The enzyme catalyzed both the biosynthetic (Eq. 5) and glutamyl transferase (Eq. 6) reactions. The biosynthetic activity was supported by either Mg² or Mn² whereas the glutamyl transferase activity specifically required Mn² . The GS S Mg activity was optimal at pH 7.8, while the GS S Mn activity was maximal at pH 5.5 and the GS T Mn activity was optimal at pH 6.0 (Figure 15).

The pH dependence of the Km value for L glutamate indicated that a group with pK of 6.5 was probably involved in catalysis. This pK value corresponded with pK of a histidine side chain (Figure 17).

The sigmoid saturation curves for the enzyme when the Mg² concentration was varied in the presence of different fixed concentrations of ATP (Figure 25), the increased sigmoidicity when the higher concentrations of ATP were used and the absence of inhibition by excess Mg² , suggested that the enzyme was interacting in a kinetically significant manner with Mg·ATP complex and with free ATP.

Peaks in the velocity profiles observed when ATP concentration was varied in the presence of different fixed levels of Mg² further supported the suggestion that excess free ATP was inhibitory. The ascending limbs of ATP saturation curves (Figure 26) were not apparently sigmoid. A Km value of 1.5 mM was obtained for Mg·ATP (Figure 27). From the isovelocity replot (Figure 29) analysis (224), a Ki for ATP was calculated to be 2.4 mM.

Based on these results a mechanism was proposed (Figure 35) to explain the activation by Mg² of the reaction (Eq. 5) catalyzed by the A. niger GS.

At varying concentrations of Mn² in the presence of different fixed levels of ATP, bell shaped velocity curves (Mn² profiles) were obtained (Figure 30) for reaction (Eq. 5). This suggested that unlike Mg² , excess Mn² inhibited the enzyme activity. The ascending limbs of all the velocity profiles were sigmoid and the sigmoidicity increased with increasing fixed concentration of ATP.

When ATP concentration was varied at different fixed concentrations of Mn² (ATP profiles, Figure 31), bell shaped velocity curves were again obtained, showing that excess ATP was inhibiting the GS S Mn activity. The sigmoidicity of the ascending limbs of these curves increased with increasing fixed concentrations of Mn² .

Using isovelocity method (224), kinetic constants for ATP (1.1 mM), Mn² (0.6 mM) and Mn·ATP (0.9 mM) were obtained from both Mn² (Figure 30) and ATP (Figure 31) profiles. The graphical analysis also provided the dissociation constant (Kd) value of 74 M (Figure 33B) for the dissociation of Mn·ATP complex. When GS S Mn was monitored at different equimolar Mn² and ATP concentrations a sigmoid curve (Figure 34) was obtained showing that free ATP, Mn² and Mn·ATP were interacting with the enzyme.

These results were explained by postulating a set of equilibria shown in Figure 36.

The interaction of the enzyme with ATP, Mg² , Mg·ATP and Mn² , ATP and Mn·ATP were explained by fitting the data to the general model proposed by London and Steck (224).

The results discussed above showed that A. niger GS was activated by either Mg² or Mn² , however their mechanism of activation was quite distinct as indicated by the model. Supporting evidence for the above mentioned mechanisms was obtained by differential protection offered by these ligands against inactivation by NEM and phenylglyoxal (Figures 44 & 54).

A. niger GS was completely inactivated by phenylglyoxal and NEM suggesting the requirement of arginine and cysteine for catalytic activity. From the first order plots and from the double log plot (Figure 52), the number of NEM molecules required to inactivate one active centre was calculated to be one. The second order rate constant for NEM inactivation was calculated to be 760 M ¹·min ¹.

204 21. Inactivation of GS activity by phenylglyoxal followed first order kinetics (Figure 42) and two molecules of this reagent were required for inactivating one active site (Figure 43) suggesting the occurrence of at least one arginine at the active site. The rate constant for phenylglyoxal inactivation was calculated to be 3.8 M ¹·min ¹.

Manganese and Mg·ATP protected the enzyme against the inactivation by both NEM and phenylglyoxal (Figures 44 & 54). From the concentration dependence of the protection of GS at fixed concentration of NEM or phenylglyoxal, binding constants were obtained for Mn² (14 M, Figures 56 and 50, respectively) and Mg·ATP complex (0.9 mM, Figure 47). Free Mg² protected the enzyme against both the inactivations only marginally, whereas free ATP enhanced the rate of inactivation due to NEM and phenylglyoxal.

The binding constant for Mg·ATP obtained by protection experiments (0.9 mM, Figure 47) agreed reasonably well with the kinetic constant (1.5 mM, Figure 27) for the ligand. The dissociation constant obtained for Mn² from protection experiments was at least an order of magnitude less than the value obtained from kinetic analysis. This result suggested the presence of a high affinity site for Mn² on the enzyme and hence the possible in vivo occurrence of A. niger GS as a Mn(II) enzyme.

The specificity of the enzyme towards divalent cations (p. 72), potent inhibition by Cu² and Zn² of the synthetase activity (Tables 9 & 10), the dependence of the reaction rate on the ratio of ATP to metal activator and the protection of the enzyme by free Mn² against inactivation suggested that these metal ions interact at catalytic as well as at regulatory sites.

Glutamine synthetase activity was probably not modulated by many of the end products of glutamine metabolism such as CTP, GMP, tryptophan, anthranilic acid, NAD , glucosamine, etc. (Tables 14 & 15). Metabolites which showed significant inhibition of A. niger GS include histidine, carbamyl phosphate, AMP and ADP. The inhibitions were more pronounced when Mn² was the metal ion activator and greater inhibition was obtained at lower pH values (Tables 14 & 15).

Histidine and carbamyl phosphate were complete inhibitors of the enzyme (Figures 60 & 66). They both competed with glutamate as the varied substrate (Figures 61, 62 & 67). The pH dependence of histidine inhibition (Figure 63) suggested that the amino acid with protonated imidazole side chain was the potent inhibitor of the enzyme.

AMP and ADP competed with ATP for the enzyme and both were complete inhibitors suggesting interaction at the active site as classical competitive inhibitors rather than as cumulative, partial feedback inhibitors.

206 28. Glutamine synthetase bridges carbon and nitrogen metabolism and therefore its role in citric acid fermentation assumes importance. The physiological status of GS and the effect of various conditions of fermentation on its activity showed that GS levels were very significantly lowered (Tables 17 & 18). Citric acid fermentation was favoured by low pH, low buffering capacity, high concentration of carbon source and manganese deficiency (< 2 ppb).

Intracellular pH of A. niger cells was measured to determine whether Mn² or Mg² dependent activity was physiologically important. The bromophenol blue dye distribution method as well as the use of fluorescence probe showed that intracellular pH was acidic (Figures 73 & 74) around 6.0–6.5.

A. niger GS was inactivated under acidic pH condition and Mn² ions protected against this inactivation to some extent (Figure 76).

The GS S Mn activity was higher at acidic pH values (6.0) compared to GS S Mg activity. The amount of Mg² required to activate GS at pH 6.0 was very high and non physiological (> 50 mM, Figure 75). Also, small amounts of Mn² inhibited the Mg² dependent GS activity at pH 7.8–8.0 (Figure 24).

Both GS S Mn and GS S Mg reactions were inhibited by citrate, 2 oxoglutarate, ATP as well as by EDTA (Figures 77–83 and Table 19). These results suggested that inhibition by citrate, etc. was due to their ability to chelate metal ions rather than to any direct interaction with the enzyme.

The results discussed above showed that during citric acid fermentation the GS is almost completely inhibited. A block at GS leads to decreased biosynthetic potential and back up of the tricarboxylic acid cycle resulting in the excretion of large amounts of citric acid by the fungus (Figure 84).

The results presented in this thesis provide strong experimental support to the hypothesis proposed (p. 29) namely, that GS is implicated in the production of citric acid by the organism. These observations pertain to decreased levels of GS under fermentation conditions; acidic intracellular pH; Mn² deficiency; inhibition of GS activity by citrate, etc., and inactivation of the enzyme due to acidic intracellular pH.

It can be concluded that, although A. niger GS was not probably regulated by covalent modification, rapid in vivo inactivation as in the case of E. coli, its activity was predominantly modulated by metal ions, intracellular pH, availability of nitrogen source(s) in the medium, etc. This enzyme had an important role in the excretion of citric acid by this organism under special conditions of fermentation.