Differential regulation of the yeast isozymes of pyruvate carboxylase and the locus of action of acetyl CoA.

2.1 Chemicals and reagents

The sodium salt of acetyl CoA was purchased from Sigma and all other reagents were of analytical quality.

2.2 Yeast strains and constructs

S. cerevisiae DM18 (MATα, ura3, trp1, Δpyc1: LEU2, Δpyc2: HIS3) (Brewster, Val, Walker & Wallace, 1994) harboring pVT100-PYC1 (Branson, Nezic, Wallace & Attwood, 2002) was used as a source to prepare PYC1 isozyme.

The PYC2 construct was generated by PCR-amplification of PYC2 gene from S. cerevisiae W303 chromosomal DNA. The gene was cloned from two overlapping PCR fragments using two pairs of primer designed from the PYC2 gene sequence deposited in the Saccharomyces genome database (http://www.yeastgenome.org). [PYC2-F1 (5’cccaagcttaagtaaaatatgagcagtagcaag-3’, underline indicates HindIII restriction site, bold indicates an ATG initiation codon) and PYC2-R1 (5’-accaccggccttcttgacagcattc-3’) were synthesized to generate a 1.9 kb 5’-end of PYC2 gene while PYC2-F2 (5’-agtgctactggaaaagggaccatg -3’) and PYC2-R2 (5’-ccctctagattaacgagtaaaaattactttttttggggatgg -3’; underline indicates XbaI restriction site) were synthesized to generate a 1.8 kb 3’-end of PYC2 gene. These two fragments were sequenced and assembled by digestion with HindIII/BamHI and with BamHI/XbaI respectively, and cloned into HindIII/XbaI-digested pVT100u (Vernet, Dignard & Thomas, 1987), producing pVT100-PYC2 construct. Pyc2 construct with N-terminal 9His tag was constructed by PCR-amplifying a 0.8 kb 5’-end of Pyc1 gene using 9His F primer (5’-gcccctcgagatgcaccaccaccatcatcaccaccatcacagcagtagcaagaaattggccggtctt-3’, bold indicates codons for 9His) and Pyc2-R primer (5’-atagcgtcacgaacttcacgg-3’). This PCR product was cloned into pDrive cloning system (Qiagen) and sequenced. The insert was excised from pDrive vector by digesting with XhoI, located upstream of an initiation codon and an AccI, which located in the middle of PCR product, and replaced the corresponding fragment in pVT100-PYC2. Pyc1 construct with N-terminal 9His was generated by PCR-amplifying a 0.9 kb 5’-end of Pyc2 gene using 5’-His Pyc1 primer (5’ gccctcgagtaaaaataatgcatcatcatcaccaccatcatcatggctcgcaaagaaaattcgccggcttg-3’; bold indicates codons for 9His) and PYC1-R (5’-tctgtgatggtatgttccacttgg-3’) using pV100-PYC1 as template. This PCR product was cloned into pDrive cloning system and sequenced. Finally, the insert was excised from pDrive by digesting with XhoI, locating at the most 5’-end of initiation codon, and EcoRI, locating in the middle of the PCR product, and replaced the corresponding fragment in pVT100-PYC1.

Six Pyc1/Pyc2 chimeras were generated as follows: Chimera 1 containing the N-terminal 616 residues of Pyc1 and the C-terminal 562 residues of Pyc2 and chimera 2 containing the N-terminal 617 residues of Pyc2 and the C-terminal 562 residues of Pyc1 were generated by utilizing a unique BamHI site presence between codons 616 and 617 of Pyc1, and between codons 617 and 618 of Pyc2 genes, allowing these two fragments to be swapped in frame. These two chimeras were constructed by digesting Pyc1 and Pyc2 genes with XhoI, locating at the most 5’-end of the gene and BamHI. These XhoI-BamHI fragments were then replaced each other to form chimeras 1 and 2, respectively. Chimera3 containing the N-terminal 437 residues of Pyc1 and the C-terminal 741 residues of Pyc2 and chimera 4 containing the N-terminal 438 residues of Pyc2 and the C-terminal 741 residues of Pyc1 were generated by introducing a SacII site between codons 437 and 438 of Pyc1 and between codons 438, and 439 of Pyc2 genes, without altering their encoded amino acids and reading frames by site-directed mutagenesis. The mutagenic primers used to introduce SacII site in Pyc1 gene were PYC1/SacII-Forward (5’-aatcgagttcagaatccgcggtgtcaagaccaac-3’; underline indicates SacII restriction site) and PYC1/SacII-Reverse (5’-gttggtcttgacaccgcggattctgaactcgatt-3’) and in Pyc2 gene were PYC2/SacII-Forward (5’-atcgaattcagaatccgcggtgtgaagaccaac-3’) and PYC2-SacII-Reverse (5’-gttggtcttcacaccgcggattctgaattcgat-3’). These SacII-modified Pyc1 and Pyc2 constructs cloned in pBluescript SK II plasmid (Stratagene) were then excised with XhoI locating at the most 5’-end of both Pyc1 and Pyc2 genes, and SacII, and replaced each other to produce chimeras 3 and 4, respectively. Chimera 5 containing the N-terminal 95 residues of Pyc1 and the C-terminal 1,083 residues of Pyc2 and chimera 6 containing the N-terminal 96 residues of Pyc2 and the C-terminal 1,083 residues of Pyc1 were generated by introducing SalI site between codons 95 and 96 of Pyc1 and between codons 96 and 97 of Pyc2 genes, without altering their encoded amino acids and reading frames. The mutagenic primers used to introduce a SalI site in Pyc1 gene were Pyc1-MluI-Forward (5’-ccaaaaacaccaggtcgacttcatccatccaggt-3’) and Pyc1-MluI-Reverse (5’-acctggatggatgaagtcgacctggtgtttttgg-3’) and for Pyc2 gene are Pyc2-MluI-Forward (5’-aaagaagcataaggtcgacttcatccatccaggt-3’) and Pyc2-MluI-Reverse (5’-acctggatggatgaagtcgaccttatgcttcttt-3’). The SalI-modified Pyc1 and Pyc2 constructs cloned in pBluescript SK II plasmid were then excised with XhoI and SalI and replaced each other to produce chimeras 5 and 6, respectively. Finally, all of the above chimeric Pyc1-Pyc2 fragments were excised from pBluescript and cloned into the equivalent sites in pVT100u. The positions of various regions of Pyc1 and Pyc2 being swapped to form these six chimeras are shown in Table 2 and Table 3. All of these constructs were transformed into S. cerevisiae W303 where the pyc1 gene had been disrupted with a HIS3 cassette by homologous recombination (Brachmann et al., 1998).

Table 2.

Analysis of effects of NH₄Cl on Pyc1, Pyc1 and Chimeras 1 & 2

Isozyme/Chimera	Ka (mM)	KI (mM²x 10−4)	Vmax (u mg−1)	Vo (% Vmax)
	11.1 ± 0.6	7.8 ± 0.3	5.8	6.8
	7.5 ± 0.8	19.8 ± 1.7	7.9	10.4
	13.7 ± 1.7	4.5 ± 0.5	8.0	9.7
	9.2 ± 1.0	17.8 ± 1.5	3.8	11.6

Open in a new tab

Values of K_a, K_I and V_max are derived from the fits of equation (ii) to the data as shown in Fig. 3.

Values of v_o were obtained by expressing the velocity in the absence of NH₄Cl as percentage of V_max.

Table 3.

Analysis of acetyl CoA activation of Pyc1, Pyc2 and Chimeras 1–6 and the effects of aspartate on this activation

Isozyme/Chimerad	aKa (µM)	Vmax (u mg−1)	bvo (% max vel)	cn (0 mM Asp)	cn (2.6 mM Asp)	cn (5.0 mM Asp)
	16.5±0.6	8.3	6.1	1.20±0.04	1.86±0.04	2.13±0.04
	8.0±1.7	4.0	26.7	0.99±0.02	1.32±0.01	1.51±0.03
	6.9±0.3	5.1	11.3	1.18±0.04	ND	ND
	4.0±0.2	2.1	20.0	0.98±0.15	ND	ND
	10.9±0.7	4.5	10.5	1.12±0.06	ND	ND
	1.9±0.1	2.3	22.2	0.94±0.07	ND	ND
	7.6±0.4	1.4	15.2	1.18±0.07	ND	ND
	1.8±0.1	0.8	22.3	1.07±0.04	ND	ND
	5.7±0.4	0.06	17.1	1.03±0.06	ND	ND
	3.4±0.1	1.3	11.9	1.37±0.03	ND	ND

Open in a new tab

^a

K_a is the activation constant for acetyl CoA in the absence of aspartate ± standard error derived from the fit of equation (iv) to the data.

^b

v_o is the velocity of the reaction in the absence of acetyl CoA (and aspartate) given as a percentage of the maximum velocity of the reaction at saturating [acetyl CoA].

^c

n is the Hill coefficient of cooperativity ± standard error estimated by linear regression analysis of the Hill plot.

^d

Diagrams representing the polypeptide chains of the various constructs of Pyc1, Pyc2 and the chimeras where the numbers indicate the amino acid position in Pyc1 at the boundary between the Pyc1 and Pyc2 sequences.

ND is not determined

2.3 Preparation and purification of isozymes and chimeras

Preparation and purification of the untagged Pyc1 and Pyc2 isozymes and the untagged chimeras was essentially as described by Branson, Nezic, Jitrapakdee, Wallace & Attwood. (2004) using anion exchange chromatography on DEAE Sepharose CL-6B and affinity chromatography on avidin-Sepharose as a last purification step. The 9xHis-tagged chimeras were purified by affinity chromatography on Ni-NTA resin, using a linear imidazole gradient (0 – 250 mM). Purified enzymes were stored at −80°C in storage buffer comprising 20% (v/v) glycerol in 0.1M Tris-Cl, pH 7.8. Prior to assay, the enzyme was transferred to 0.1M Tris-Cl, pH 7.8 by centrifuging through Sephadex G-25 as described by Helmerhorst & Stokes (1980).

2.3 Assays of pyruvate carboxylase activity

The conditions for this spectrophotometric assay where oxaloacetate formation is measured in a coupled reaction with malate dehydrogenase, were essentially as described by Attwood & Cleland (1986). One unit of enzyme activity is defined as the amount of enzyme required to catalyze the formation of 1 µmol of oxaloacetate min⁻¹ at 30°C. All assays were performed at 30°C. Assays were performed in 0.1 mM Tris-Cl, pH7.8 with the following concentrations of substrates unless stated otherwise: 2.5 mM ATP; 20 mM NaHCO₃; 7 mM MgCl₂; 10 mM pyruvate; 0.21 mM NADH. Note that in the experiment where [ATP] was varied, the concentration of free Mg²⁺ was maintained at 4.5 mM and in all experiments where substrate concentrations were varied the reactions were performed in the presence of 0.25 mM acetyl CoA. Assays were initiated by the addition of either isozymes or chimeras and activities were calculated as units ml⁻¹ of enzyme solution.

2.4 Data analysis

The kinetic parameters given in Table 1 were obtained from non-linear regression fits of the Michaelis-Menten equation to the data. The ammonium chloride activation/inhibition data were analysed using non-linear least-squares regression analysis to fit of the following equations (i and ii) to the data:

where: v is the measured velocity of the reaction; v_o is the measured velocity of the reaction in the absence of NH₄Cl; V is the velocity of the reaction if all of the enzyme subunits had a single NH₄⁺ bound (E.NH₄, see Fig. 3); K_a is the activation constant for the binding of NH₄⁺ to the enzyme (see Fig. 3); K_I is the inhibition constant for the subsequent binding of NH₄⁺ (see Fig. 3) and has units of mM in (i) and mM² in (ii).

Equation (i) and Equation (ii) were derived based on the respective reaction schemes shown in Fig. 3 with the assumptions that E.NH₄ represents the maximally active form of the enzyme and that E.(NH₄)₂ and E.(NH₄)₃ have no enzymic activity.

The data for activation of the isozymes and chimeras by acetyl CoA, in the presence or absence of aspartate were initially analysed by linear regression according to the Hill equation (iii):

where: v is the velocity of the reaction at any acetyl CoA concentration; V_max is the velocity of the reaction at saturating acetyl CoA; n is the Hill coefficient of cooperativity; K_a is the activation constant.

In order to obtain an estimate of K_a with a standard error, the data were re-analysed according to equation (iv):

where the value of n used was taken from the Hill plot analysis.

3.1 Kinetic parameters for the reaction catalysed by Pyc1 and Pyc2

The values of the K_ms for the substrates and k_cat values are shown in Table 1. The value of k_cat in the presence of saturating concentrations of acetyl CoA and the substrates of the reaction for Pyc2 is 78% of that for Pyc1 indicating that Pyc2 is somewhat less catalytically active than Pyc1. In addition, the K_m for pyruvate of Pyc2 is somewhat higher than that of Pyc1. The largest differences between Pyc1 and Pyc2 however, occur in their K_m values for MgATP and HCO₃⁻.

The K_m for MgATP is 2.4 fold higher for Pyc2 compared to that of Pyc1 and is closer to the value of 0.2 mM reported by Myers, Tolbert & Utter. (1983) for a preparation of pyruvate carboxylase from S. cerevisiae, which may have contained a mixture of Pyc1 and Pyc2. This may indicate that there is a greater degree of control of the activity of Pyc2 than that of Pyc1 activity by the concentration of ATP.

The K_m for HCO_3⁻, of Pyc2 is nearly 16 times that of Pyc1. The value of the K_m for HCO₃⁻ of Pyc2 lies closer to the value reported by Myers Tolbert & Utter, (1983) (19.1 mM) in the presence of acetyl CoA than that measured by Ruiz-Amil, de Torrontegui, Palacián, Catalina & Losada, (1965) (2.4 mM) for enzyme isolated from S. cerevisiae, which in both cases may have contained a mixture of Pyc1 and Pyc2. However, in the presence of 8.4 mM K⁺ the enzyme preparation studied by Myers, Tolbert & Utter, (1983) exhibited a K_m for HCO₃⁻ of 3.1 mM and in the presence of a combination of acetyl CoA and K⁺, this K_m was reduced further. Thus in the yeast cell, where the concentration of K⁺ is estimated to be about 17 mM (Myers, Tolbert & Utter, 1983) and in the presence of acetyl CoA, the K_m for HCO₃⁻ may be much closer to that measured for Pyc1. When we repeated the kinetic experiments varying [NaHCO₃] in the presence of 17 mM KCl, we found that at concentrations of NaHCO₃ above about 50 mM, inhibition was evident (data not shown). Estimates of the K_ms obtained from a fit of an equation of the form similar to that of eq (i) to the data were 4.1 ± 1.8 mM for Pyc2 and 34 ± 10 mM for Pyc1. The inhibition was somewhat similar to that seen with increasing concentrations of NH₄⁺ (see below) suggests that there may be both activatory and inhibitory effects of the monovalent cations (K⁺ and Na⁺) (McClure, Lardy& Kneifel, 1971). that differ between Pyc11 and Pyc2 and that make determination of the true K_m for HCO₃⁻ difficult.

3.2 Effect of NH₄⁺ on the activity of Pyc1, Pyc2 and Chimeras 1 and 2

In view of the amino sequence similarity between Pyc1 and Pyc2 (92.8% sequence identity), it seemed feasible to prepare chimeras of Pyc1 and Pyc2 without incurring any great disruption of the structure or function of the enzyme whilst enabling the investigation as to which domain(s) was associated with the regulatory properties with respect to acetyl CoA and NH₄⁺. To perform an initial investigation, two chimeras were constructed, each containing roughly half the amino acid sequence of each isozyme (see Table 2). Both chimeras were expressed as 9His-tagged recombinant proteins.

Fig. 3 shows the effects of increasing concentration of NH₄⁺ on the activity of Pyc1, Pyc2 and chimeras 1 and 2. Enzyme activity increases with [NH₄⁺] up to about 75 mM. However, above this concentration, the NH₄Cl becomes inhibitory. The estimated K_as and K_Is for NH₄⁺ and velocities in the absence of NH₄⁺ are given in Table 2. Both Pyc1 and chimera 1 are somewhat less sensitive to activation by NH₄⁺, with their K_as being about 49% higher than those for Pyc2 and chimera 2 respectively. On the other hand, both Pyc2 and chimera 2 are less sensitive to inhibition by NH₄⁺, with their K_Is being 2.5 and 4 fold higher than those of Pyc1 and chimera 1 respectively. Whilst the v_o values for Pyc1 and chimera 1 are lower than those for Pyc2 and chimera 2 respectively, the maximum velocity for Pyc1 is lower than that for Pyc2 whilst the reverse is the case for the chimeras.

Huet, Menendez, Gancedo & Francois (2000) found a 4–5 fold up-regulation of the expression of Pyc1, but little change in Pyc2 expression when the yeast was grown with ammonium as the nitrogen source compared to aspartate. Huet, Menendez, Gancedo & Francois (2000) offered the explanation that with NH₄⁺ as the nitrogen source there would be a high demand for α-ketoglutarate as a substrate for glutamate dehydrogenase which catalyses the formation of glutamine, using NH₄⁺ as the other substrate. Pyruvate carboxylase activity would therefore need to be increased to replenish α-ketoglutarate in the tricarboxylic acid cycle removed for this purpose. One hypothesis that might explain the difference in regulation of expression Pyc1 compared to Pyc2 is that Pyc1 activity is more highly regulated by the intracellular concentration of NH₄⁺. However, our data do not really support this hypothesis, since the differences in activation of the two isozymes by NH₄⁺ are relatively small and indeed the K_a of activation for Pyc1 is higher than that for Pyc2.

We did find a larger difference between the isozymes in terms of their inhibition at high NH₄Cl concentrations, with Pyc2 being less sensitive to inhibition. However, the concentrations of NH₄Cl at which this inhibition has a pronounced effect on enzymic activity are unlikely to be of physiological relevance and may be a salt concentration effect rather than a specific effect of NH₄⁺ binding.

In general, chimera 2 is more like Pyc2 in being somewhat more sensitive to activation and less sensitive to inhibition by NH₄⁺ than chimera 1, which is more like Pyc1. In terms of the velocities in the absence of NH₄⁺ (v_o) chimera 1 is more similar to Pyc1 in having a somewhat lower v_o than chimera 2, however the maximum velocity of chimera 1 is lower than that of chimera 2 whilst the reverse is true for Pyc1 and Pyc2. Thus, in the majority of ways, the effects of NH₄⁺ on yeast pyruvate carboxylase appear to be determined by the N-terminal half of the amino acid sequence, although the maximum velocity may be influenced by the carboxy-terminal sequence.

3.3 Activation of Pyc1 and Pyc2 by acetyl CoA and the effects of aspartate

Fig. 4a and b show Hill plots of the activation of Pyc1 and Pyc2 respectively, by acetyl CoA in the presence and absence of aspartate. Table 3 shows the Hill coefficients of cooperativity calculated from these Hill plots, the K_as of activation by acetyl CoA in the absence of aspartate and the velocities of the reaction catalysed by Pyc1 and Pyc2 in the absence of acetyl CoA as a percentage of those in the presence of saturating acetyl CoA. In the absence of aspartate, the Hill coefficient for Pyc2 is not significantly different from 1 indicating that there is that there is no cooperativity in the binding of acetyl CoA. However, that for Pyc1 in the absence of aspartate is significantly higher than 1, indicating that there is some positive cooperativity in this case. With both Pyc1 and Pyc2, the Hill coefficients increase in the presence of increasing aspartate concentration however, those for Pyc1 are higher than for Pyc2 and indicate a higher degree of cooperativity in the binding acetyl CoA to Pyc1 in the presence of aspartate and thus a greater inhibitory effect of aspartate on Pyc1. The values of the Hill coefficient for Pyc1 are similar to those observed in pyruvate carboxylase prepared from baker’s yeast (Cazzulo & Stoppani, 1968).

The original acetyl CoA binding studies on the chicken liver enzyme indicated that there were four acetyl CoA binding sites per tetramer (Frey & Utter, 1977). However, a more recent study on the yeast enzyme in which more accurate determinations of protein concentration were performed, revealed only 2 binding sites per tetramer (Chapman-Smith, Booker, Clements, Wallace & Keech, 1991). If this latter figure is correct, this suggests that at 5 mM aspartate, acetyl CoA binding is completely cooperative in Pyc1.

If we consider the classical model of allostery, in the absence of any ligand there is an equilibrium between relaxed (R) and taut (T) conformers of the enzyme, with the affinity of the R form for the ligand being higher than that of the T form. In the case of Pyc1 and Pyc2, the ligand is acetyl CoA and in the absence of aspartate the equilibrium between R and T forms of the enzymes would appear to be very much in favour of the R form. Thus, in Pyc2, in the absence of aspartate, all the binding sites for acetyl CoA are already in the high affinity state and thus there is no cooperativity of acetyl CoA binding. Aspartate, however shifts the equilibrium between R and T conformers in favour of the T form and hence, as the concentration of acetyl CoA increases, this equilibrium is shifted back towards the R form, giving rise to the sigmoidal activation kinetics. In Pyc1 there is a significant presence of the T form in the equilibrium between R and T forms in the absence of aspartate and acetyl CoA, with increasing concentrations of aspartate shifting this equilibrium in favour of the T form. This difference between Pyc1 and Pyc2 may also, in part at least explain, the much higher activity (v_o) of Pyc2 in the absence of activators compared to Pyc1, with the R form having a higher intrinsic activity than the T form.

The activation constant for acetyl CoA (K_a) in the absence of aspartate for Pyc1 is about twice that for Pyc2 and v_o for Pyc2 is about 4 times greater than that for Pyc1 suggesting that Pyc2 is more active at zero and low concentrations of acetyl CoA than Pyc1. Huet, Menendez, Gancedo & Francois (2000) suggested that Pyc2 is important for glycolytic growth under anaerobic conditions, where acetyl CoA concentrations would be low, and a major role of Pyc2 would be to produce oxaloacetate that could then be reduced by malate dehydrogenase, with concomitant oxidation of NADH to NAD⁺. This supply of NAD⁺ would allow the continuation of glycolysis in the early stages of fermentative growth.

3.5 Activation of chimeras 1 and 2 by acetyl CoA

Fig. 5a shows Hill plots of the activation of chimeras 1 and 2 by acetyl CoA. Table 3 shows the Hill coefficients of cooperativity calculated from these Hill plots, the K_as of activation by acetyl CoA in the absence of aspartate and the velocities of the reaction catalysed by chimeras 1 and 2 in the absence of acetyl CoA as a percentage of those in the presence of saturating acetyl CoA. The Hill coefficient for chimera 2 is not significantly different from one, indicating that there is no cooperativity in the binding of acetyl CoA. However, the Hill coefficient for chimera 1 is significantly higher than one, indicating that there is some positive cooperativity in this case. In this respect, chimera 1 is similar to Pyc1 while chimera 2 is similar to Pyc2.

In terms of the values of v_o, chimera 1 is similar to Pyc1 and chimera 2 is more like Pyc2. Similarly, chimera 2 has a lower activation constant (K_a) than chimera 1, again making it more like Pyc2 than Pyc1. However, the absolute values of K_a for both chimeras 1 and 2 are 1.5 and 4 fold lower than those for Pyc1 and Pyc2 respectively. To investigate whether the His tag on the chimeras was affecting the activation constant for acetyl CoA, N-terminally 9His-tagged constructs of Pyc1 and Pyc2 were expressed and purified and their activation by acetyl CoA measured as for Pyc1 and Pyc2. The parameters for the activation of these constructs are also shown in Table 3. Thus the K_a for 9HisPyc1 was found to be less than that of Pyc1, suggesting that the 9His tag is affecting the activation of the enzyme by acetyl CoA, primarily by reducing the K_a. Similarly, the K_a for 9HisPyc2 was reduced compared to Pyc2. In addition, the values of V_max of the 9His-tagged constructs were lower than those of the untagged isozymes and similar to those of the corresponding chimeras whilst the v_o and n values appear not to be markedly affected.

From the studies of chimeras 1 and 2 and the comparisons with the Pyc1 and Pyc2 and their 9His-tagged forms, we can conclude that the major determinants of the response of the enzyme to acetyl CoA activation and the effects of NH₄Cl lie in the N-terminal half of the amino acid sequence. This is in agreement with the study by Islam, Sueda & Kondo, (2005) who made a chimera of the biotin carboxylation subunit (472 amino acids) of the Aquifex aeolicus enzyme and the C-terminal sequence of the Bacillus thermodenitrificans enzyme. The A. aeolicus enzyme activity is independent of acetyl CoA and Islam, Sueda & Kondo (2005) found that the activity of the chimera was also acetyl CoA-independent. From our study, however, it is important to note the marked effect of the 9His tag on the K_a for acetyl CoA in both Pyc1 and Pyc2 and illustrates the potential dangers of using such tags which, although extremely useful for the purposes of enzyme purification, can have unforseen effects on enzyme action.

3.6 Further investigation of the locus of activation of the Pyc isozymes by acetyl CoA

In order to further investigate the locus of action of acetyl CoA, two more sets of chimeras were created (see Table 3). As the 9His tag on both Pyc1 and Pyc2 appeared to affect K_a for acetyl-CoA, the next sets of chimeras were constructed without this N-terminal tag. In chimeras 3 and 4, the N-terminal section comprises the first 437 amino acids from Pyc1 and Pyc2 respectively. This N-terminal sequence encompasses the whole biotin carboxylation domain (see Fig. 1) and a sequence of about 100 amino acids, N-terminal to this. In chimeras 5 and 6, the N-terminal section comprises most of this N-terminal sequence i.e. the first 95 amino acids from Pyc1 and Pyc2 respectively.

Figs. 5b and c show the Hill plots for activation of these chimeras by acetyl CoA and Table 3 shows parameters of activation of these chimeras by acetyl CoA. The parameters for chimeras 3 and 4 most resemble those of Pyc1 and Pyc2, respectively with chimera 4 showing little cooperativity, having a higher v_o than chimera 3 and a lower K_a than chimera 3. The absolute values of K_a for both chimeras are however, lower than those for Pyc1 (about 50%) and Pyc2 (about 23%) and the v_o value for chimera 3 is 2.5 times that for Pyc1, whilst that for chimera 4 is similar to that of Pyc2.

The parameter values for chimeras 5 and 6 in Table 3, show that chimera 5 exhibits little cooperativity, while chimera 6 shows significant cooperativity. In addition, the v_o value for chimera 5 is larger than that for chimera 6. In these respects chimera 6 behaves like Pyc1 and chimera 5 more like Pyc2. However, the K_a value for chimera 5 is still larger than that for chimera 6, although there is less than a two-fold difference between them. The K_a value for chimera 6 is almost double that of chimera 4 and 43% of that of Pyc2, while that of chimera 5 is 75% of that of chimera 3 and 35% of that of Pyc1. The low V_max values observed with chimeras 3– 6 suggest that fusion of parts of the isozyme sequences has introduced some structural instability. This is especially severe in chimera 5, suggesting that there may be some structurally important interactions in Pyc1 that have been severely disrupted in this chimera.

Analysis of the activation of chimeras 3 and 4 by acetyl CoA indicate that, while overall they behave in a similar manner to Pyc1 and Pyc2 respectively, there are some marked differences. The Hill coefficients for chimeras 3 and 4 are essentially the same as for Pyc1 and Pyc2 respectively and the v_o value for chimera 4 is similar to that of Pyc2, the v_o value for chimera 3 is considerably higher than for Pyc1. In addition, the values of K_a for chimeras 3 and 4 are lower than for Pyc1 and Pyc2 respectively. This suggests that there is part of the amino acid sequence between amino acids 437 and 616 that contributes to the action of acetyl CoA on the enzyme. However, analysis of Pyc1 and Pyc2 sequences between these regions reveal only three amino acid differences i.e. V⁴⁴⁹ in Pyc1 (I⁴⁵⁰ in Pyc2), A⁶⁰⁴ (S⁶⁰⁵) and R⁶⁰⁸(K⁶⁰⁸) (see Fig. 6).

Chimera 5 shows essentially no cooperativity whilst chimera 6 exhibits positive cooperativity and the acetyl CoA-independent activity of chimera 5 is greater than that of chimera 6. In these respects chimera 5 is behaving more like Pyc2 and chimera 6 is more like Pyc1. On the other hand, the K_a for chimeras 5 is larger than that for chimera 6, although that for chimera 6 is almost twice that for chimera 4. Thus the main determinants for cooperativity and acetyl-CoA-independent activity appear to lie somewhere between amino acids 95 and 437. Analysis of Pyc1 and Pyc2 sequences shows 13 amino acid differences across this region including N¹⁴¹ (H¹⁴² in Pyc2), K¹⁴⁵ (R¹⁴⁶), E¹⁶² (Q¹⁶³), E²⁸⁵ (V²⁸⁶), A³³⁰ (S³³¹), S³⁴⁰ (T³⁴¹), P³⁴² (T³⁴³), F³⁴⁷ (L³⁴⁸), A³⁵⁷ (S³⁵⁷), A³⁶⁸ (S³⁶⁸), I³⁷⁸ (L³⁷⁹), T⁴⁰⁰ (A⁴⁰¹) and I⁴⁰¹ (T⁴⁰²). These sequence differences between Pyc1 and Pyc2 may form distinct structural motifs that mediate differential activation by acetyl-CoA.

Sueda, Islam & Kondo (2004) found that there is an interaction between the biotin carboxylase domains in the enzyme from B. thermodenitrificans and postulated that this was important for the formation of the tetrameric quaternary structure. Since cooperativity is dependent on inter-subunit interactions, this is in agreement with our data that the basis for the cooperativity lies in the biotin carboxylation domain.

The situation is more complex with respect to the K_a for acetyl CoA. The differences in the K_a values between all of the chimeras and the isozymes suggests that determinants of K_a are more dispersed and there is an interdependence between these determinants that affects K_a. The fact that the K_a values of chimeras 5 and 6 are still quite different from those of Pyc1 and Pyc2 indicates that one or more of these determinants lies in the N-terminal 95 amino acid sequence.

Fig. 6 shows a multiple sequence alignment between the N-terminal halves of the amino acid sequences of Pyc1, Pyc2 and pyruvate carboxylase from Aspergillus niger, whose activity is completely independent of acetyl CoA (Feir & Suzuki, 1969). As can be seen, there are relatively few differences between Pyc1 and Pyc2 alone and with the exception of the arginine in position 595 of Pyc1 where the corresponding residue in Pyc2 is alanine, the differences are fairly conservative in terms of amino acid properties. Where there are differences between the sequences of Pyc1 and Pyc2 that are not so conservative, these also coincide with different residues in the A. niger sequence. For example at the position corresponding to 71 in Pyc1 there is a valine in Pyc1, a glutamate in Pyc2 and lysine in the A. niger enzyme. Thus at this position, the range of amino acids varies from non-polar, to acidic to basic. There is a similar arrangement at position 88. At other positions there is a combination of polar, non-polar and basic or acidic residues (positions 1, 91, 94, 285,342 and 545) or aromatic, aliphatic and polar (positions 5 and 347). Thus in N-terminal region of the sequence these amino acid differences are scattered. The differences in the alignment at positions 285 and 342 of Pyc1 do lie within the span between positions 95-437 and will be of interest for future site-directed mutagenesis studies aimed at investigating the basis of acetyl CoA-independent activity and the cooperativity of acetyl CoA activation.

3.7 Conclusions

In conclusion, we have shown a number of differences between the kinetic properties of the isozymes Pyc1 and Pyc2 from S. cerevisiae, which may have physiological relevance in the differential control of their activities. However, in the current work all activation experiments were performed in the presence of high substrate concentrations and in the presence of a single activator and thus we are essentially studying activator effects on V_max. In the yeast cell there is likely to be a combination of the effects of different activator concentrations and changing substrate and inhibitor concentrations that contribute to the fine differential regulation of Pyc1 and Pyc2 activities.

The main determinants of acetyl CoA activation of the enzyme lie in the N-terminal half of the amino acid sequence, with those that affect the degree of acetyl-CoA-independent activity and the cooperativity of acetyl CoA action being more localised to the biotin-carboxylation domain.