Computer Simulation of Metabolism
Consideration of Cycles - The GS/GOGAT Cycle
The following figure illustrates an application of the hypothetical stable isotope pulse-chase labeling kinetic simulation models described on the previous page, in the consideration of 15N flux via the glutamine synthetase (GS)
- glutamate synthase (GOGAT) cycle. The GS/GOGAT cycle is thought to be the primary pathway of ammonia assimilation in higher plants. In this cycle, 15NH3 is first incorporated into the amide-group of glutamine (Gln) in the reaction catalyzed by the chloroplast isoform of GS. The amide-group of Gln is then transferred to 2-oxoglutarate in the reaction catalyzed by the chloroplast-localized, ferredoxin-dependent GOGAT, producing 2 molecules of glutamate (Glu). One of these Glu molecules is randomly re-utilized in the synthesis of Gln [in this case transferring 15N label to the amino-group of Gln], and the second Glu is used in other reactions in the chloroplast, or
exported to the cytosol for net amino acid synthesis.
In the greatly simplified model shown above, we have allowed exactly one-half of the Glu formed in the GOGAT reaction in the chloroplast to be exported to the cytoplasm [i.e. no utilization of either Glu or Gln in other reactions is envisaged to occur
within the chloroplast]. Thus, the rate of Glu export from the chloroplast (r1) is the same as the rate of ammonia assimilation in the chloroplast (r1). The Glu that is exported to the cytosol is envisaged to be used in several reactions. In part, this
Glu is used to sustain the synthesis of Gln by a second cytosolic isoform of GS (r2) [in this particular scenario, the flux via the cytosolic Gln pool corresponds to 10% of the total ammonia assimilation rate]. Note that the cytosolic Gln pool receives
15N in the amide position from 15NH3, and in the amino position from the cytosolic Glu pool. The cytosolic Glu is, in part, envisaged to be used in the synthesis of two amino acids; proline (Pro) and gamma-aminobutyrate (GABA). In the case of Pro synthesis (r10), the pools of intermediates (gamma-glutamylphosphate, glutamic semialdehyde and pyrroline-5-carboxylate) are envisaged to be negligible. GABA synthesis is envisaged to occur in a single step (r13); decarboxylation of Glu. Both GABA and Pro can be metabolized back to Glu; in the former case by transamination (r14), and in the latter case by oxidation (r9). In this greatly simplified model we have not considered the potential role of the mitochondrion in processes such as proline oxidation.
The cytosolic Gln, Glu and Pro pools are envisaged to be utilized in the synthesis of protein in the cytosol, at rates r5, r6 and r11, respectively [note that since GABA is a non-protein amino acid, a similar fate for GABA is not considered]. Small
portions of the newly synthesized cytosolic Gln, Glu and Pro pools are also envisaged to be sequestered in the vacuole, at rates r4, r7 and r12. For simplicity, a vacuolar pool of GABA is not considered. "Other" unspecified metabolic fates of Gln, Glu
and GABA make up the balance of the fluxes (e.g. for Gln "other" (r3) could include synthesis of asparagine, histidine and tryptophan; for Glu "other" (r8) could include transamination to aspartate, glycine, alanine, and many other amino acids; for
GABA "other" (r15) could include transamination to alanine). Note that for each of the pools (except the vacuolar and protein pools) the total influx to each pool is equal to the total efflux from the corresponding pool (e.g. r2 = r3 + r4 + r5; r10 = r9
+ r11 + r12; r1 + r14 + r9 = r2+ r13 + r10 + r6 + r7 + r8). Thus, the chloroplastic and cytosolic pools of the amino acids remain constant, while the vacuolar and protein-bound pools expand with time.
In the figure shown above, the average isotope abundances and total pool sizes of the total free pools have been plotted; for example, the isotope abundance for Gln-amide corresponds to the average isotope abundance of the chloroplastic,
cytosolic and vacuolar Gln-amide pools. Because the Gln-amide and Gln-amino nitrogen pool sizes are of exactly the same size, only the Gln-amino nitrogen pool is shown.
When different option buttons are checked, this permits simulation of the individual isotope abundances and pool sizes of the chloroplastic (Chl), cytosolic (Cyt), vacuolar
(Vac), and protein-bound (Prot.) pools, using identical flux and pool size assumptions.
Note that it is only the chloroplastic pools which reveal the expected order of labeling of intermediates for the operation of the GS/GOGAT cycle; i.e. Gln-amide > Glu > Gln-amino. The presence of a second, more slowly-turning
over Gln pool in the cytosol, and a large vacuolar pool of Gln, substantially distorts the labeling patterns for the total free pool of Gln, such that Glu rather than Gln-amide, is the most heavily labeled product during the
later part of the "pulse" phase of the labeling time-course.
[Visual Basic program code used for simulation of the GS/GOGAT cycle]
Download the Visual Basic program illustrated above. To run this program you must have Visual Basic 5.0 (or greater) installed on your computer.
An on-line, interactive version of this program is available as a Java applet. In this version of the program, options for plotting isotope abundance and pool size of total free pools, chloroplastic, cytoplasmic,
vacuolar or protein-bound pools must be entered as a numerical value of 1 to 5, respectively.
The following references have used this type of model (written in Basic rather than Visual Basic) to simulate 15N flux via the glutamine synthetase (GS) - glutamate synthase (GOGAT) cycle, and the metabolism of glutamate to
other amino acids (e.g. proline and GABA). Rhodes et al (1980) showed that the labeling kinetics of Gln and Glu in Lemna minor supplied with 15NH4+ were consistent with
synthesis of Gln in two separate compartments, with a small, rapidly turning-over pool of Gln participating in the synthesis of Glu via the GS/GOGAT cycle. This type of model was subsequently extended to consider the flux of 15N
via Glu to Pro in proline-accumulating tomato and tobacco cells adapted to water and salinity stresses (Rhodes et al, 1986; Rhodes and Handa, 1989). The models were used to show that the elevated proline pools in the stress-adapted cell lines must be due
to increased Pro synthesis from Glu. Mayer et al (1990) used a similar approach to show that heat shock greatly stimulated the in vivo rate of synthesis of GABA from Glu in cowpea cells.
Mayer, R.R., Cherry, J.H. and Rhodes, D. 1990. Effects of heat shock on amino acid metabolism of cowpea cells. Plant Physiology 94: 796-810.
Rhodes, D. and Handa, S. 1989. Amino acid metabolism in relation to osmotic adjustment. In (J.H. Cherry, ed.) "Environmental Stress in Plants. Biochemical and Physiological Mechanisms", NATO ASI Series G: Ecological Sciences, Vol 19,
Springer-Verlag, Berlin, pp. 41-62.
Rhodes, D., Handa, S. and Bressan, R.A. 1986. Metabolic changes associated with adaptation of plant cells to water stress. Plant Physiology 82: 890-903.
Rhodes, D., Sims, A.P. and Folkes, B.F. 1980. Pathway of ammonia assimilation in illuminated Lemna minor. Phytochemistry 19: 357-365.