Pheromone/Receptor/G protein interactions

From Yeast Pheromone Model

< Pheromone

Jump to: navigation, search

Back to main model page

Receptor interactions with pheromone

Coexpression of a Ste2 mutant that is defective in pheromone binding (S184R) and a Ste2 mutant that is defective in G protein binding (specific alanine replacement in any of the 3 intracellular loops along with C terminal truncation at residue 303) does not restore sensitivity to pheromone in a ste2Δ strain. This was judged by halo formation and Fus1-lacZ expression. Chinault et al. 2004 PMID 14764600
- This suggests that receptors in a dimer/oligomer are independently activated by binding to their own pheromone peptide, rather than signaling in trans by allowing binding of pheromone to one Ste2 in a dimer to activate the other Ste2 in the dimer.
- These mutants were shown to form hetero-oligomers by FRET.

Measured Binding Affinities

Kd = 6.4 nM when Ste2 is expressed off a multicopy plasmid (measured at 0°C). The strains used were bar1^- to prevent pheromone degradation. Bajaj et al. 2004 PMID 15491163
- Using a fluorescent pheromone analog [K⁷(NBD),Nle¹²] α-factor, Kd = 3.6 nM when Ste2 is expressed from a CEN plasmid, and Kd = 7.4 nM when Ste2 is expressed from a multicopy plasmid. For Ste2 expressed off the multicopy plasmid, the on and off rates were also measured: kon = 1.6 * 10⁵ M^-1s^-1, koff = 1.1 * 10^-3 s^-1. These values were all measured at 0°C.
- The fluorescent pheromone analog's binding kinetics fits better to a double exponential than to a single exponential.

Kd = 2 nM. Experiment was done using ³⁵S-labeled pheromone at 22°C. Experiments done on membrane preparations. Cells were treated with NaN₃ and KF to prevent energy-dependent processes, and lysed. Blumer et al. 1988 PMID 2839507

Kd = 6 nM. Experiment was done using ³⁵S-labeled pheromone at 22°C. TAME was used to prevent pheromone degradation, and cells were treated with NaN₃ and KF to prevent growth and other energy-dependent processes. Jenness et al. 1986 PMID 3023832

Kd = 10 nM. Experiment was done using ³⁵S-labeled pheromone at 30°C. Cells were bar1^- (to prevent pheromone degradation), and cells were treated with NaN₃ and KF to prevent growth and other energy-dependent processes. Dube and Konopka. 1998 PMID 9819407

koff = 9 * 10^-4 s^-1. Experiment was done using 35S-labeled pheromone at 22°C. TAME was used to prevent pheromone degradation, and cells were treated with NaN₃ and KF to prevent growth and other energy-dependent processes. Kd was also measured, but later discounted by the same group. Jenness et al. 1983 PMID 6360378

Kd = 7 nM +/-1 nM (measured in triplicate). Experiments were done with ³H-labeled pheromone at 22°C. TAME was used to prevent pheromone degradation, and cells were treated with NaN₃ and KF to prevent growth and other energy-dependent processes. David et al. 1997 PMID 9182592

Kd = 4.2 nM. Experiments were done with ³⁵S-labeled pheromone, and the cells were treated with NaN₃ and KF. The strains used contained a non-functional bar1-1 allele to prevent pheromone degradation. The experiment was performed at room temperature. Dosil et al. 2000 PMID 10866688

Kd = 4.5 nM. Experiments were done with ³⁵S-labeled pheromone, and the cells were treated with NaN₃ and KF. The strains used contained a non-functional bar1-1 allele to prevent pheromone degradation. The temperature at which the experiment was performed is unclear. Chen an Konopka 1996 PMID 8524302

Kd = 6.8 nM and Kd = 5.6 nM. Experiments were done with ³H-labeled pheromone, and the cells were treated with NaN₃ and KF. The strains used contained a non-functional bar1-1 allele to prevent pheromone degradation. The temperature at which the experiment was performed is unclear. Lee et al. 2001 PMID 11495900

Kd = 2.3 nM. Experiments were done with ³⁵S-labeled pheromone at 22°C, and the cells were treated with NaN₃ and KF. Weiner et al. 1993 PMID 8385135

Kd = 0.8 nM. Experiments were done with ³⁵S-labeled pheromone at 22°C, and the cells were treated with NaN₃ and KF. Clark et al. 1994 PMID 8132618

Kd = 22 nM, k_on = 4.9 * 10⁴ M^-1s^-1, koff = 1.1 * 10^-3 s^-1. Experiments were done with ³H-labeled pheromone at 22°C, and the cells were treated with NaN₃ and KF. The strain used is bar1Δ. Raths et al. 1988 PMID 2846561

kon = 2 * 10⁶ M^-1s^-1, koff = 1 * 10^-2 s^-1, though it is unclear whether these values were measured or inferred from literature. Yi et al. 2003 PMID 12960402

From a computational model, Kd = 2 nM was used, though no source was given. Hao et al. 2003 PMID 12968019

From a computational model. kon = 2 * 10⁴ M^-1s^-1, koff = 1 * 10^-2 s^-1 (corresponding to a Kd = 500nM) was used. This these numbers were referenced to Yi et al 2003 PMID 12960402 who used the same koff, but a kon 100x greater. Kofahl and Klipp 2004 PMID 15300679

A number of Ste2 mutants and synthetic pheromone analogs have been identified with different Kd's and activities (measured by EC₅₀). These may be relevant in some experimental setups where faster pheromone dissociation is desired. Lee et al. 2001 PMID 11495900
- See also Naider and Becker. 2004 PMID 15374647

Effect of Gα on receptor/pheromone binding affinity

There is no significant difference in Kd for deletion of GPA1; Kd = 3.5 nM for WT cells, and Kd = 4.8 nM for gpa1Δ cells (with full length STE2 on CEN, and far1Δ). Bajaj et al. 2004 PMID 15491163
- This is contrary to most GPCRs where G protein increases receptors affinity for ligand.
- gpa1Δ far1Δ cells have altered growth and morphology, so we have no idea what else is happening (i.e. number of Ste2/cell?)
- For Ste2 expressed off of a multicopy plasmid, Kd is 2-fold higher than WT cells. gpa1Δ in this same strain results in ~WT receptor/ligand Kd.
- The authors argue that reason that they don't see change in Kd with deletion of G protein is that when they delete the G protein, the receptor is able to maintain it's high affinity state via interactions with other proteins. Other people have performed these experiments in reconstituted membranes, and so in the absence of the G protein there aren't other proteins to interact with the receptor.

There is a ~2x difference in Kd upon deleting Gpa1 and Ste4 (Kd = 4.2 nM, 3900 receptors per cell for WT cells, Kd = 9.1 nM, 1900 receptors/cell for gpa1Δ ste4Δ cells). Dosil et al. 2000 PMID 10866688
- This suggests that the G protein has a minimal effect on the affinity between the receptor and pheromone.

Using partially purified membranes containing Ste2 and the G protein, it was shown that addition of GTP[γ-S], a non-cleavable GTP analog, converts the receptor from a high affinity state (Kd = 17 nM), to a low affinity state (Kd = 150 nM). Blumer and Thorner 1990 PMID 2161538
- This suggests that the G protein has a large effect on the affinity between the receptor and pheromone.

Purified receptors that were reconstituted in vesicles in the absence of the G protein had an affinity of 155 nM for pheromone. David et al. 1997 PMID 9182592

Using membrane fractions, dissociation of pheromone in the presence of the G protein and saturating amounts of GTPγS is much faster (half-time < 2 min) than in the absence of GTPγS (half-time > 10 min). Weiner et al. 1993 PMID 8385135

Membrane preparation experiments showed that mAChR (a GPCR) affinity for one agonist, [³H]oxotremorine-M, is greatly reduced by GTP, while it's affinity for another agonist, L-[benzilic-4,4'-³H]quinuclidinyl benzilate, was unaltered by GTP. Waelbroeck et al. 1982 PMID 7110116
- This suggests that not all agonist/receptor interactions are affected by the state of the G protein.

Overexpression of Ste2 to approximately 160 000 receptors per cell does not affect the binding affinity of pheromone for Ste2 (~6 nM). David et al. 1997 PMID 9182592
- Since the total amount of G protein per cell (Gpa1_num and Ste4_num) is much less than 160 000, pretty much all of the Ste2 molecules are unbound by G proteins.

Reaction Definition

Here we have conflicting information. Membrane prep assays suggest that the G protein has a large effect (factor of 10) on receptor/ligand affinity. Cell assays suggest that the G protein has minimal effect (factor of 2 or less). We will assume that the cell assays more accurately reflect the effect of the G protein on ligand affinity. Since these studies show a less than 2x decrease in receptor/ligand in the absence of Gpa1, we will assume this is experimental noise, and not model any G protein effect on receptor/ligand affinity.

In general, the Kd and koff are measured directly, and the kon is calculated from these values. Thus, we will use the average measured Kd_Pheromone_Ste2 and the average measured koff_Pheromone_Ste2 to calculate the resulting kon_Pheromone_Ste2.

Assumptions:

Pheromone/receptor interaction is independent of whether Ste2 is bound to Sst2.

Pheromone(Ste2_site) + Ste2(Pheromone_site) <-> Pheromone(Ste2_site!1).Ste2(Pheromone_site!1)

Forward rate constant kon_Pheromone_Ste2

Reverse rate constant koff_Pheromone_Ste2

moleculizer-pheromone-Ste2-binding

Receptor interactions with G protein

Gpa1 contains a Ste2 binding domain. Kallal et al. 1997 PMID 9111362

The accepted model for G protein coupled receptors is that when ligand-bound, the receptor acts as a guanine nucleotide exchange factor (GEF) for the G protein. Nucleotide exchange results in the binding of GTP to the Gα subunit (Gpa1), which actvates the G protein leading to Gα (Gpa1) dissociation from both Gβγ (Ste4:Ste18) and the receptor (Ste2). Preininger and Hamm 2004 PMID 14762218

G protein coupling to Ste2 in vivo requires the presence of functional Ste4:Ste18 and functional Gpa1. Blumer and Thorner 1990 PMID 2161538

Ste2 mutated in the C-terminal cytosolic tail (N388S) exhibits reduced interaction with Ste4 and Gpa1, as determined by 2-hybrid assay. Duran-Avelar et al. 2001 PMID 11287148

Upon pheromone treatment, less Gpa1 is associated with Ste2 than prior to treatment (co-IP). Wu et al. 2004 PMID 15197187

Coexpression of a Ste2 mutant (alanine substitutions in the 1st intracellular loop) that is thought to be defective in Gpa1 binding/activation with another Ste2 mutant (alanine substitutions in the 3rd intracellular loop) that is thought to be defective in Gpa1 binding/activation partially restores sensitivity to pheromone. Chinault et al. 2004 PMID 14764600
- The authors conclude from this that Ste2 monomers in a dimers cooperate to activate Gpa1.

Reaction Definition

We know that Gpa1 contains a Ste2 binding domain, so presumably the coupling of the G protein to the receptor is through Gpa1. We also know that Gpa1 couples inefficiently to Ste2 in the absence of Ste4:Ste18. This leads to a model where Gpa1 binds Ste2 weakly, and Ste4:Ste18 greatly increases this binding efficiency. In other words, Gpa1(GDP), which is usually present in the heterotrimeric Gpa1:Ste4:Ste18 complex, binds to Ste2 with much higher affinity than Gpa1(GTP), which is usually present free of binding to Ste4:Ste18. A consequence of this relative affinity is that a single molecule or dimer of Ste2 could act enzymatically to catalyze the nucleotide exchange on many Gpa1 molecules.

Assumptions:

Above we've assumed that α-factor binding is not affected by the G protein, which is equivalent to assuming that G protein binding to Ste2 is not affected by pheromone.

Ste2's phosphorylation state has no effect on G protein coupling and Yck binding does not affect G protein coupling.

Gpa1's nucleotide state only indirectly affects G protein/receptor coupling by affecting Gpa1's affinity for Ste4:Ste18.

Sst2 binding to Ste2 does not affect G protein binding to Ste2 (see RGS(Sst2)/Galpha(Gpa1)/Receptor(Ste2) interactions)

The difference in affinity between Gpa1/Ste2 and Gpa1:Ste4:Ste18/Ste2 arises solely as differences in off-rates.

Ste2(Gpa1_site) + Gpa1(Ste2_site, Ste4_site) <->  Ste2(Gpa1_site!1).Gpa1(Ste2_site!1, Ste4_site)

Forward rate constant kon_Ste2_Gpa1

Reverse rate constant koff_Ste2_Gpa1

Ste2(Gpa1_site) + Gpa1(Ste2_site, Ste4_site!+) <-> Ste2(Gpa1_site!1).Gpa1(Ste2_site!1, Ste4_site!+)

Forward rate constant kon_Ste2_Gpa1Ste4Ste18

Reverse rate constant koff_Ste2_Gpa1Ste4Ste18

There are specific constraints on these rate constants.

moleculizer-Ste4-enhances-Gpa1-Ste2-binding

moleculizer-Gpa1-Ste2-binding

Gα (Gpa1) interactions with Gβ:Gγ (Ste4:Ste18)

Gpa1 and Ste4 interact by yeast two-hybrid, whereas Gpa1 and Ste18 do not interact. Clark et al 1993 PMID 8417317

A fusion protein between Gpa1 and Ste4 restored mating ability to a ste4Δ gpa1Δ strain. Klein et al. 2000 PMID 10725354
- It's unclear whether this means that the G protein does not actually dissociate upon activation and just undergoes an conformational shift, or that the linker that they used between Gpa1 and Ste4 is long enough to allow sufficient separation to mimic dissociation.

Crystal structures have shown the differences in structure of the portions of Gα that interact with Gβγ when Gα is in the GTP- versus the GDP-bound state. Mixon et al. 1995 PMID 7481799; Wall et al. 1995 PMID 8521505; Lambright et al. 1996 PMID 8552184

Rate constant for Gpa1(GDP) binding Ste4:Ste18 was estimated to be 1 (molecule per cell)^-1 s^-1. It is not clear how this rate was measured or chosen from literature. Yi et al. 2003 PMID 12960402

Loss of FRET is observed between CFP-Gpa1 and Ste18-YFP upon exposure to pheromone, suggesting dissociation of Gpa1 from Ste4:Ste18. Yi et al. 2003 PMID 12960402

Increase in FRET is observed between rat Gαi₁-YFP and human Gβ₁ upon exposure to agonist, suggesting that the G protein does not dissociate but mearly undergoes a conformational change upon activation. Bunemann et al. 2003 PMID 14673086

An association rate of 5.8 * 10^-4 molec^-1 s^-1 was used in a model for Gpa1(GDP) binding Ste4:Ste18. This was referenced to Yi et al. 2003 who used 1 (molecule per cell)^-1 s^-1, but no justification was given for the 1000x discrepancy. Yildirim et al. 2004 PMID 15313578

An association rate of 3.33 * 10¹⁰ M^-1 s^-1 was used in a model for Gpa1(GDP) binding Ste4:Ste18. This was referenced to Yi et al. 2003 who used 1 (molecule per cell)^-1 s^-1. This number is probably right at the limit of being diffusion-limited. Kofahl and Klipp 2004 PMID 15300679

Overexpression of Sst2 is able to inhibit mating. Konopka 1993 PMID 8413281
- Since over-expressed Sst2 does not appear to out compete Ste4:Ste18 for binding to Gpa1 (as this would result in activation of mating response rather than suppression of mating response), it seems likely that Sst2 is able to bind Gpa1 independently of Ste4:Ste18.

Reaction Definition

Despite the limited evidence to the contrary, we will assume that the G protein dissociates upon activation. We assume that Gpa1(GTP) does retain some limited affinity for Ste4:Ste18. As described above, Ste4:Ste18 increases Gpa1's affinity for Ste2, so in a reciprocal fashion Ste2 increases Gpa1's affinity for Ste4:Ste18.

Assumptions:

The Sst2-binding state of Ste2 does not affect Gpa1's affinity for Ste4:Ste18.

The difference in affinity between Gpa1/Ste4:Ste18 and Ste2:Gpa1/Ste4:Ste18 arises solely as differences in off-rates.

Gpa1(Ste2_site, Ste4_site, nucleotide~GDP) + Ste4(Gpa1_site, Ste5_site, Ste20_site) <->    Gpa1(Ste2_site, Ste4_site!1, nucleotide~GDP).Ste4(Gpa1_site!1, Ste5_site, Ste20_site)

Forward rate constant kon_Gpa1GDP_Ste4Ste18

Reverse rate constant koff_Gpa1GDP_Ste4Ste18

Gpa1(Ste2_site, Ste4_site, nucleotide~GTP) + Ste4(Gpa1_site, Ste5_site, Ste20_site) <->    Gpa1(Ste2_site, Ste4_site!1, nucleotide~GTP).Ste4(Gpa1_site!1, Ste5_site, Ste20_site)

Forward rate constant kon_Gpa1GTP_Ste4Ste18

Reverse rate constant koff_Gpa1GTP_Ste4Ste18

Gpa1(Ste2_site!+, Ste4_site, nucleotide~GDP) + Ste4(Gpa1_site, Ste5_site, Ste20_site) <->    Gpa1(Ste2_site!+, Ste4_site!1, nucleotide~GDP).Ste4(Gpa1_site!1, Ste5_site, Ste20_site)

Forward rate constant kon_Ste2Gpa1GDP_Ste4Ste18

Reverse rate constant koff_Ste2Gpa1GDP_Ste4Ste18

Gpa1(Ste2_site!+, Ste4_site, nucleotide~GTP) + Ste4(Gpa1_site, Ste5_site, Ste20_site) <->    Gpa1(Ste2_site!+, Ste4_site!1, nucleotide~GTP).Ste4(Gpa1_site!1, Ste5_site, Ste20_site)