G protein nucleotide hydrolysis/exchange

From Yeast Pheromone Model

Jump to: navigation, search

Back to main model page



Contents

Nucleotide Exchange

  • Using purified Gs (the full αβγ G protein), the dissociation rate of GDP at 30°C is 0.7 min-1. Brandt and Ross. 1985 PMID 2981206
    • The dissociation is actually biphasic, with a much slower dissociation rate for about 50% of the G proteins.
    • When excess Gβγ is added (there is already Gβγ present, although perhaps sub-stoichiometric), the dissociation rate increases 4-fold, suggesting that GDP dissociates at least 4-fold more slowly from Gβγ-bound Gα than it does from Gα by itself.
  • GDP dissociates from purified Goα (from bovine brain), both in the absence and presence of Gβγ, at a rate of 0.4 min-1. Higashijima et al. 1987 PMID 3027067
    • This was done at 20°C in the presence of 20mM Mg2+.
  • The rate of GDP dissociation from purified Goα (from bovine brain) is independent of Mg2+ concentration in the absence of Gβγ, and increases with Mg2+ concentration in the presence of Gβγ. Higashijima et al. 1987 PMID 3100519
    • At the concentration of Mg2+ found in synthetic media (~5 mM, see Sherman 2002 PMID 12073320), the dissociation rate of GDP is approximately the same (0.3 min -1) in the absence and presence of Gβγ.
    • In the absence of Mg2+, Gβγ increases Goα's affinity for GDP by 300-fold, but in the presence of 10 mM Mg2+, Gβγ increases Goα's affinity for GDP by only 7.5-fold. At 10 mM Mg2+, the difference in affinity manifests itself solely as a difference in association rates.
    • This was done at 20°C.
  • GDP dissociates from purified Gi1 at a rate of 0.53 min-1, from Gi2 at a rate of greater than 4 min-1, and Gi3 with a rate of 0.93 min-1. Carty et al. 1990 PMID 2108158
    • This was done in the presence of 2 mM Mg2+ at 32°C.
    • The proteins were all purified from human erythrocytes.
  • Yeast cells grown in either standard SD (unknown amount of Mg2+), or standard SD supplemented with 200 mM Mg2+ have about 3.1 milligrams of Mg2+ per gram dry cell weight. Graschopf et al. 2001 PMID 11279208
    • Using estimates for cell dry weight and cell volume from Sherman (2002 PMID 12073320), this corresponds to total cellular concentrations of ~30 mM.
  • G protein coupled receptors are known to act as catalysts where a single ligand-bound receptor is able to catalyse nucleotide exchange in multiple G proteins. Berstein et al. 1992 PMID 1341877
  • For Gq/11, agonist-bound receptor acts as a GEF (Guanine nucleotide Exchange Factor) to increase the nucleotide exchange rate to 1-2 min-1 (measured with reconstituted receptor and G protein in lipid vesicles at 30°C). Berstein et al. 1992 PMID 1341877
    • Using the same G proteins, the same group later published a much faster rate (see Biddlecome et al. 1996 PMID 8626481 and Mukhopadhyay and Ross 1999 PMID 10449728), so it is likely that there is something wrong with these results.
  • Nucleotide exchange on Gq/11 coupled to a liganded-receptor occurs faster than 20 min-1. This rate is thought to be limited by nucleotide dissociation rather than nucleotide binding. Biddlecome et al. 1996 PMID 8626481
    • This was done in vesicles with purified Gqα and Gβ1γ2 and purified receptor.
    • The same group later published a exchange rate of 1.5 s-1, and feel that this number is more corect than the 20 min-1 measured here.
  • GEF-stimulated nucleotide exchange occurs at a rate of 1.5 s-1 for Gq at 30°C. Mukhopadhyay and Ross 1999 PMID 10449728
    • This was done in vesicles with purified Gqα and Gβ1γ2 and purified receptor.
    • The GEF-stimulated nucleotide exchange rate was unaffected by the presence of the GAP.
  • Sst2 does not affect the rates of binding or release of GDP and GTP by Gpa1. Apanovitch et al. 1998 PMID 9537998
    • The innate nucleotide exchange rate was measured to be 0.037 min-1 at 30°C for purified Gpa1 (in the absence of Ste4 and Ste18).
  • Kinetics of in vitro GDP dissociation, apparent GTPγS binding, nucleotide turn-over and nucleotide hydrolysis were measured for recombinant rat Gα subunits G, Giα1, Giα2, and Giα3. Linder et al. 1990 PMID 2159473
    • It appears that GTP binding as well as the GTPase turnover cycle are limited by the rate of GDP dissociation.
    • These experiments were performed in the absence of the Gβγ subunit, and in 10 mM Mg2+.
    • All kcat's were measured at 30°C, all the rates for Goa were measured at 30°C, and the other rates were measured at 20°C.
GTPγS
binding
kcat (min-1)
GDP
release
koff (min-1)
Steady state
GTPase turnover
number (min-1)

kcat (min-1)
G 0.27 ± 0.04 0.19 ± 0.03 0.27 ± 0.03 2.2 ± 0.3
Giα1 0.029 ± 0.002 0.026 ± 0.003 0.028 ± 0.005 2.4 ± 0.1
Giα2 0.063 ± 0.002 0.072 ± 0.005 0.095 ± 0.005 2.7 ± 0.4
Giα3 0.022 ± 0.001 0.025 ± 0.004 0.036 ± 0.002 1.8 ± 0.2
  • Bulk rate of G protein activation by ligand-bound Ste2 was estimated via fitting a model to data. The rate constant of 1 * 10-1 (molecules per cell)-1 s-1 corresponds to the equation pheromone:Ste2 + Gpa1(GDP):Ste4:Ste18 --> pheromone:Ste2 + Gpa1(GTP) + Ste4:Ste18. This rate constant is thus a mixture of the rates of receptor binding to the G protein, ligand-bound receptor induced nucleotide exchange, and G protein dissociation. Yi et al. 2003 PMID 12960402
    • This suggests that a single pheromone-bound receptor can cause the population of GDP-bound Gpa1 to exchange GDP for GTP at a rate of 0.1 s-1, or a hundred pheromone-bound receptors would cause exchange at a rate of 10 s-1. Since they assume 10000 receptor molecules per cell, they are assuming a minimum nucleotide exchange rate of 1000 s-1 for pheromone:Ste2:Gpa1(GDP):Ste4:Ste18. This rate seems unreasonable.
  • Bulk rate of G protein activation by ligand-bound Ste2 was estimated via fitting a model to data. The rate constant of 6 * 104 M-1 s-1 corresponds to the equation pheromone:Ste2 + Gpa1(GDP):Ste4:Ste18 --> pheromone:Ste2 + Gpa1(GTP) + Ste4:Ste18. This rate constant is thus a mixture of the rates of receptor binding to the G protein, ligand-bound receptor induced nucleotide exchange, and G protein dissociation. They reference this number to Yi et al. 2003 PMID 12960402. Kofahl and Klipp 2004 PMID 15300679
    • If we follow the same logic that we used in the above case, the minimum rate constant for nucleotide exchange is (assuming 1666.67 nM Ste2 as the authors did) 1666.67 nM × 6 * 104 M-1 s-1 = 0.1 s-1. So somehow the authors failed to correctly calculate the rate constant when they converted units from the rate constant used by Yi et al.
  • Bulk rate of G protein activation in the absence of pheromone was given in a model (no source). The rate constant of 5.25 * 10-5 min-1 corresponds to the equation Gpa1(GDP):Ste4:Ste18 --> Gpa1(GTP) + Ste4:Ste18. This rate constant is thus a mixture of the rates of the innate nucleotide exchange rate, and G protein dissociation. Nucleotide exchange is likely limiting in this case. Hao et al. 2003 PMID 12968019
    • Assuming that nucleotide exchange is slower than G protein dissociation, this gives an innate nucleotide exchange rate of 5.25 * 10-5 min-1.
  • Bulk rate of G protein activation in the presence of pheromone was given in a model (no source). The rate constant of f * 1.75 min-1 (f = fractional ligand occupation of Ste2) corresponds to the equation Gpa1(GDP):Ste4:Ste18 --> Gpa1(GTP) + Ste4:Ste18. This rate constant is thus a mixture of the rates of receptor binding to the G protein, ligand-bound receptor induced nucleotide exchange, and G protein dissociation. Hao et al. 2003 PMID 12968019
  • In general, Gβγ's (Ste4:Ste18) regulate Gα's (Gpa1) by inhibiting GDP release (which is thought to be the limiting step in nucleotide exchange). Preininger and Hamm 2004 PMID 14762218
  • The dissociation of GDP from two rabbit Gα proteins (Gsα and Goα) is slowed 3-5 fold by the presence of the corresponding Gβγ subunits in vitro. Brandt and Ross 1985 PMID 2981206; Higashijima et al. 1987 PMID 3100519
    • However, at physiological concentrations of Mg2+ (15-50 mM?), dissociation of GDP from Gα appears only to be slightly affected by the presence of Gβγ.

Reaction Definition

We know that pheromone-bound Ste2 acts as a GEF for Gpa1. We also know that the presence of Sst2 appears to have no effect on nucleotide exchange. Evidence using homologous G proteins suggests that Ste4:Ste18 may not affect the rate of nucleotide exchange.

Assumptions:

  • Nucleotide dissociation is limiting in nucleotide exchange, so that dissociation of GDP results in near-immediate binding of GTP.


The measured innate nucleotide exchange rate for Gpa1 not bound by Ste4:Ste18 is 0.037 min -1 (6.17 * 10-4 s-1). We are not using the reported rate 5.25 * 10-5 min-1 because it is unclear how it was derived.

The rate of GEF mediated exchange for Gpa1 has not yet been reliably measured directly. The rate from Yi et al. seems suspiciously high (>1000 s-1), and the rate from Kofahl and Klipp was supposedly derived from this, but is actually 1000-fold less. For Gq, the measured GEF mediated nucleotide exchange rate (in the presence of Gβγ) is 1.5 s-1. This number, although for a different receptor/G protein system, seems to be more reliable.

Gpa1(Ste2_site, nucleotide~GDP) -> Gpa1(Ste2_site, nucleotide~GTP)

Ste2(Pheromone_site, Gpa1_site!2).Gpa1(Ste2_site!2, nucleotide~GDP) -> 
    Ste2(Pheromone_site,Gpa1_site!2).Gpa1(Ste2_site!2, nucleotide~GTP)


Ste2(Pheromone_site!+, Gpa1_site!2).Gpa1(Ste2_site!2, nucleotide~GDP) -> 
   Ste2(Pheromone_site!+,Gpa1_site!2).Gpa1(Ste2_site!2, nucleotide~GTP)


Nucleotide Hydrolysis

  • Sst2 is a GAP (GTPase Activating Protein), which acts to increase Gpa1’s rate of GTP hydrolysis. Apanovitch et al. 1998 PMID 9537998
  • GAP protein RGS4 increases the GTP hydrolysis rate of Giα1 by at least 40-fold at 4°C, with half-time of hydrolysis of less than 7s in the presence of RGS4. Berman et al. 1996 PMID 8756726;
  • Although GTP hydrolysis by Giα1 proteins is dependent on the presence of Mg2+ (i.e. cannot proceed in the presence of EDTA), the presence of RGS4 allows for rapid GTP hydrolysis in the presence of EDTA, prehaps because RGS4 (in combination with Giα1-GTP) can bind Mg2+ sufficiently strongly. Berman et al. 1996 PMID 8756726
  • GAP protein 6His-RGS10 increases the GTP hydrolysis rate of Giα3 by at least 10-fold at 4°C. Hunt et al. 1996 PMID 8774883
    • The presence of 6His-RGS10 had no effect on the nucleotide exchange rate of Giα3.
  • RGS1 binds Goα weakly in the presence of GDP or GTPγS, and strongly in the presence of GDP and AlF4-. Watson et al. 1996 PMID 8774882
    • RGS1, RGS4 and GAIP can copurify Goα and Giα that are treated with GDP and AlF4-, but not when thet are treated with GDP alone.
    • This suggests that RGS1 preferentially binds/stabilizes the transition state leading to GTP hydrolysis.
  • RGS1 increases the GTP hydrolysis rate of both Gαo and Gαi1 by at least 10 fold at 5°C, with half-time of hydrolysis of less than 10s. Watson et al. 1996 PMID 8774882
  • Sst2 binds to Gpa1. In pulldown experiments performed at 0°C with partially purified proteins, binding was similar with GDP and GTPγS and was increased with GDP-AlF4-, suggesting that Sst2 has highest affinity for transition state intermediate of Gpa1-GTP hydrolysis. Apanovitch et al. 1998 PMID 9537998
  • Sst2 binds to the C-terminal tail of Ste2. Genetic and biochemical evidence suggests that the DEP domain of Sst2 (distant from the RGS domain that acts upon Gpa1) must bind to the C-terminal tail of Ste2 in order to be targeted to Gpa1. Ballon et al. 2006 PMID 16990133
    • Mutations in the DEP domain of Sst2 abolish its ability to negatively regulate Gpa1 and impair its interaction with Ste2.
    • There is no evidence that Sst2 is allosterically activated by its binding to Ste2, so it is likely that binding to Ste2 is essential for increasing the local concentration of Sst2 in the vicinity of Gpa1.
  • Goα hydrolyses GTP at a rate of 1.8 min-1 in the absence of Gβγ, and at a rate of 2.1-2.4 min-1 in the presence of Gβγ. Higashijima et al. 1987 PMID 3027067
    • This was done at 20°C in the presence of 20mM Mg2+.
  • Kinetics of in vitro GDP dissociation, apparent GTPγS binding, nucleotide turn-over and nucleotide hydrolysis were measured for recombinant rat Gα subunits Goa, Gia1, Gia2, and Gia3. Linder et al. 1990 PMID 2159473
    • It appears that GTP binding as well as the GTPase turnover cycle are limited by the rate of GDP dissociation.
    • These experiments were performed in the absence of the Gβγ subunit, and in 10 mM Mg2+.
    • All kcat's were measured at 30°C, all the rates for Goa were measured at 30°C, and the other rates were measured at 20°C.
GTPγS
binding
kcat (min-1)
GDP
release
koff (min-1)
Steady state
GTPase turnover
number (min-1)

kcat (min-1)
Goa 0.27 ± 0.04 0.19 ± 0.03 0.27 ± 0.03 2.2 ± 0.3
Gia1 0.029 ± 0.002 0.026 ± 0.003 0.028 ± 0.005 2.4 ± 0.1
Gia2 0.063 ± 0.002 0.072 ± 0.005 0.095 ± 0.005 2.7 ± 0.4
Gia3 0.022 ± 0.001 0.025 ± 0.004 0.036 ± 0.002 1.8 ± 0.2
  • The innate nucleotide hydrolysis rate of purified Gq/11 was measured indirectly to be 0.6 min-1, and directly to be 0.8 min-1 at 30°C. Berstein et al. 1992 PMID 1341877
  • Gq/11 GAP-mediated nucleotide hydrolysis occurs at a rate of 66-400 min-1 (~100- to 600-fold increase over the innate rate). Biddlecome et al. 1996 PMID 8626481
    • Follow-up work from the same lab shows that RGS (Regulators fo G protein Signaling) proteins can accelerate nucleotide hydrolysis by between 400-10,000 fold over the rate in the absence of GAPS. For Gq, the appropriate RGS proteins can increase this rate to 9-27 s-1 at 30°C. Mukhopadhyay and Ross 1999 PMID 10449728
  • The rate of GTP hydrolysis by Gpa1 was measured to be 0.006 min-1 at 0°C. Apanovitch et al. 1998 PMID 9537998
  • I estimate the Sst2:Gpa1(GTP) rate of hydrolysis to be >4 min-1 (at 0°C). This is a Sst2-dependent acceleration of more than 667-fold, which fits with the general RGS stimulation above. See figure below from Apanovitch et al. 1998 PMID 9537998
    • High concentrations of Sst2 (2μM) were used, probably to compensate for the weak affinity between Sst2 and Gpa1 as shown by Ballon et al. (2006 PMID 16990133)

Image:Figure from Apanovitch et al., Biochem (37) 4815.gif

  • Gpa1's innate GTP hydrolysis rate was estimated at 0.004 s-1 by fitting a model to data Yi et al. 2003 PMID 12960402
    • This same value was used (and referenced to Yi et al. 2003) by Kofahl and Klipp 2004 PMID 15300679
  • Gpa1's GTP hydrolysis rate was estimated at 0.11 s-1 (or 6.6 min-1) in cells with WT amounts of Sst2. This bulk rate does not take into account how many of the Gpa1 molecules are bound to Sst2, and was determined by fitting a model to data. Yi et al. 2003 PMID 12960402
  • Bulk rate of G protein deactivation in the absence of Sst2 was given in a model (no source). The rate constant of 3.5 * 10-3 min-1 corresponds to the equation Gpa1(GTP) + Ste4:Ste18 --> Gpa1(GDP):Ste4:Ste18. This rate constant is thus a mixture of the rates of innate nucleotide hydrolysis and G protein association. It seems likely that the nucleotide hydrolysis is rate limiting. Hao et al. 2003 PMID 12968019
  • Bulk rate of G protein deactivation in the presence of Sst2 was given in a model (no source). The rate constant of 3500 min-1 mM-1 corresponds to the equation Sst2 + Gpa1(GTP) + Ste4:Ste18 --> Sst2 + Gpa1(GDP):Ste4:Ste18. This rate constant is thus a mixture of the rates of Sst2 association with Gpa1, Sst2 accelerated nucleotide hydrolysis, and G protein association. Hao et al. 2003 PMID 12968019
    • Unfortunately, the authors list the species concentrations as in the units of molecules per cell rather than concentration, and do not state the assumed cell volume. If I assume a cell volume of 40 fL (Jorgensen et al. 2002 PMID 12089449), using 2000 molec per cell of Sst2 (83 nM), we get a minimum GAP mediated nucleotide hydrolysis rate of 0.3 min-1 (5 * 10 -3 s-1), which would be an 86-fold increase in rate over the rate in the absence of Sst2 in the same paper. Hao et al. 2003 PMID 12968019
  • Bulk rate of G protein deactivation in the presence of Sst2 was given in a model. The rate constant of 1.35 * 10-5 molec-1 s-1 corresponds to the equation Sst2 + Gpa1(GTP) --> Sst2 + Gpa1(GDP). This rate constant is thus a mixture of the rates of Sst2 association with Gpa1 and Sst2 accelerated nucleotide hydrolysis. Reference was given to first order rates of 0.11 s-1 from Yi et al (PMID 12960402) and 24 s-1 from Mukhopadhyay and Ross (PMID 10449728). Yildirim et al. 2003 PMID 15313578
    • Using their value of 2000 molecules per cell of Sst2, we get a minimum GAP mediated nucleotide hydrolysis rate of 0.027 s-1.
  • Innate nucleotide hydrolysis rate of 1 * 10-3 s-1 in the absence of Sst2 was used in a model. Reference was given to the value of 4 * 10-3 s-1 from Yi et all (PMID 12968019). Yildirim et al. 2003 PMID 15313578
  • Bulk rate of G protein deactivation in the presence of Sst2 was given in a model. The rate constant of 0.33 nM-1 min-1 corresponds to the equation Sst2 + Gpa1(GTP) --> Sst2 + Gpa1(GDP). This rate constant is thus a mixture of the rates of Sst2 association with Gpa1 and Sst2 accelerated nucleotide hydrolysis. Reference was given to first order rates of 0.11 s-1 from Yi et al (PMID 12960402). Kofahl and Klipp 2003 PMID 15313578
    • The authors start their model with no Sst2 prior to pheromone treatment, and have its synthesis controlled exclusively by Fus3. This makes it difficult to map their rate constant on to the rate constant that we want.

Reaction Definition

We know that Sst2 acts as a GAP to accelerate nucleotide hydrolysis. Evidence from Ballon et al (2006 PMID 16990133) suggests that Sst2 does not bind Gpa1 directly, but rather likely associates with Gpa1 through binding to Ste2 (see RGS(Sst2)/Galpha(Gpa1)/Receptor(Ste2) interactions for more details). Thus we will assume that Sst2:Ste2:Gpa1(GTP) has an accelerated GTP hydrolysis rate.

Assumptions:

  • Ste2 and Ste4 binding to Gpa1 do not affect nucleotide hydrolysis rates.
  • Phosphorylation of Sst2 does not affect its GAP ability.
  • Binding of Sst2 to Ste2 does not modulate its GAP ability per se (no change in kcat), but does promote interaction between Sst2 and Gpa1 and thereby indirectly increase the overall rate of nucleotide hydrolysis.

For the innate GTPase activity, we have a measured minimum rate of 0.006 min-1 (1 * 10-4 s-1), as measured at 0°C. The values used in models range from 5.8 * 10-5 s-1 (slower than the rate measured at 0°C) to 4 * 10-3 s-1. The measured rates of nucleotide hydrolysis for Gq are 1-1.5 * 10-2 s-1. We will assume a rate constant of 5 * 10-3 s-1, which is 50 times greater than the measured rate at 0°C.

The Sst2 mediated nucleotide hydrolysis rate was measured to be 6.7 * 10-2 s-1 at 0°C. The rate constants used in models vary from slower than the rate measured at 0°C (5 * 10-3 s-1 and 2.7 * 10-2 s-1) to 0.11 s-1. Measured values for Gq vary from 1.1 s-1 to 27 s-1. We will assume a rate constant of 4 s-1. (This is ~50-fold greater than the rate measured at 0°C. For innate GTPase rate, we picked a rate that is 50-fold greater than the measured value at 0°C.) Thus, we assume Sst2 accelerates GTP hydrolysis by 800-fold.

Gpa1(Ste2_site, nucleotide~GTP) -> Gpa1(Ste2_site, nucleotide~GDP)

Ste2(Gpa1_site!2, Sst2_site).Gpa1(Ste2_site!2, nucleotide~GTP) -> 
    Ste2(Gpa1_site!2, Sst2_site).Gpa1(Ste2_site!2, nucleotide~GDP)


Ste2(Gpa1_site!2, Sst2_site!+).Gpa1(Ste2_site!2, nucleotide~GTP) -> 
   Ste2(Gpa1_site!2, Sst2_site!+).Gpa1(Ste2_site!2, nucleotide~GDP)


moleculizer-Gpa1-auto-hydrolysis

Personal tools