OZITX, a pertussis toxin-like protein for occluding inhibitory G protein signalling including Gαz

Keen, Alastair C.; Pedersen, Maria Hauge; Lemel, Laura; Scott, Daniel J.; Canals, Meritxell; Littler, Dene R.; Beddoe, Travis; Ono, Yuki; Shi, Lei; Inoue, Asuka; Javitch, Jonathan A.; Lane, J. Robert

doi:10.1038/s42003-022-03191-5

Download PDF

Article
Open access
Published: 23 March 2022

OZITX, a pertussis toxin-like protein for occluding inhibitory G protein signalling including Gα_z

Communications Biology volume 5, Article number: 256 (2022) Cite this article

3669 Accesses
5 Citations
24 Altmetric
Metrics details

Subjects

Abstract

Heterotrimeric G proteins are the main signalling effectors for G protein-coupled receptors. Understanding the distinct functions of different G proteins is key to understanding how their signalling modulates physiological responses. Pertussis toxin, a bacterial AB₅ toxin, inhibits Gα_i/o G proteins and has proven useful for interrogating inhibitory G protein signalling. Pertussis toxin, however, does not inhibit one member of the inhibitory G protein family, Gα_z. The role of Gα_z signalling has been neglected largely due to a lack of inhibitors. Recently, the identification of another Pertussis-like AB₅ toxin was described. Here we show that this toxin, that we call OZITX, specifically inhibits Gα_i/o and Gα_z G proteins and that expression of the catalytic S1 subunit is sufficient for this inhibition. We identify mutations that render Gα subunits insensitive to the toxin that, in combination with the toxin, can be used to interrogate the signalling of each inhibitory Gα G protein.

Genetically detoxified pertussis toxin displays near identical structure to its wild-type and exhibits robust immunogenicity

Article Open access 05 August 2020

A toxin-deformation dependent inhibition mechanism in the T7SS toxin-antitoxin system of Gram-positive bacteria

Article Open access 28 October 2022

Alternate subunit assembly diversifies the function of a bacterial toxin

Article Open access 15 August 2019

Introduction

Heterotrimeric guanine nucleotide-binding proteins (G proteins) are important signalling transducers that link cell-surface receptors such as G protein-coupled receptors (GPCRs) to intracellular effectors^1,2,3. They consist of a Gα subunit as well as Gβ and Gγ subunits that function as an obligate dimer. There are four subfamilies of Gα subunits (Gα_s, Gα_i, Gα_q and Gα₁₂) based on sequence similarity. Their functions can be broadly generalised based on this classification. The stimulatory (Gα_s) and the inhibitory (Gα_i) subfamilies stimulate and inhibit adenylate cyclases, respectively^1,4. The Gα_q subfamily activates phospholipase C-β leading to increases in cytosolic Ca²⁺, and the Gα₁₂ subfamily activates Rho family GTPases that regulate cytoskeletal processes^2,5. Understanding the distinct signalling roles of individual members of each subfamily is central to our comprehension of how they control different physiological processes.

The Gα subunit, and in particular its carboxy tail, is largely responsible for determining the specificity of the interaction with an activated GPCR^6,7. The GPCR acts as a guanine nucleotide exchange factor, promoting the exchange of bound GDP for GTP at the guanine nucleotide-binding domain of the Gα subunit. This causes the Gα subunit to dissociate from, or rearrange relative to, the Gβγ dimer, and both then act on downstream effectors^8,9. The Gα subunit is a GTPase, hydrolysing GTP to restore GDP to the binding domain, allowing the Gβγ dimer to reassociate and completing the cycle.

AB₅-type toxins have proved to be useful tools for the interrogation of G protein signalling. These toxins are characterised by a hetero-hexameric structure consisting of the enzymatically active A subunit and pentameric ring of B subunits, which are responsible for recognition of host cell-surface receptors and facilitate cellular entry. In order to modulate host cell behaviour, AB₅ toxins have varied actions on their targets, including protease activity¹⁰, RNA N-glycosidase activity¹¹ and ADP-ribosylation¹². Of relevance to G-protein signalling, cholera toxin acts on the Gα_s subfamily¹³ and Pasteurella multocida Pasteurella multocida toxin acts on the Gα_i, Gα_q and Gα₁₂ family to render them constitutively active¹⁴. Pertussis toxin (PTX), from Bordetella pertussis, ADP ribosylates all members of the Gα_i subfamily, except for Gα_z¹⁵. Researchers have exploited these actions to identify the Gα subunits responsible for particular cell signalling processes. PTX-driven ADP ribosylation occurs on a cysteine residue of four residues from the carboxy terminus of Gα_i subunits, rendering them incapable of coupling to GPCRs. PTX-insensitive Gα_i/o subunit mutants, in which the cysteine modified by PTX is replaced by another residue, have been used to understand the role of individual Gα_i-subfamily members in vitro. One inhibitory G-protein family member, Gα_z, lacks this cysteine and is thus insensitive to PTX. Gα_z has a slow GDP-GTP exchange rate, slow GTP hydrolysis rate, and a restricted pattern of expression^16,17,18,19. Despite these unique characteristics, relatively little is known about the physiological role of Gα_z signalling, although evidence has been provided for its roles in circadian behaviours, dopaminergic signalling, and pancreatic islet β cell biology^19,20,21,22. Its function in other physiological processes remains elusive, in part due to its insensitivity to PTX. Indeed, there may be cases in which inhibitory G protein signalling has been ruled out based on a lack of PTX effect while neglecting the potential role of Gα_z.

A recent publication reported the identification and structural characterisation of a PTX-like protein derived from a uropathogenic Escherichia coli²³. The toxin has an active A subunit homologous to that of PTX and has a similar overall structural fold (Supplementary Fig. 1). Application of this toxin to HEK 293 cells, African green monkey kidney cells or bovine brain lysate revealed its substrates as heterotrimeric G proteins²³. Using Gα_i2 as a substrate in vitro, the toxin was shown to have distinct site(s) of ADP ribosylation from that of PTX—an asparagine residue and a lysine residue eight and ten residues from the carboxy terminus, respectively²³. The asparagine is conserved across several Gα subunits, suggesting that the toxin may have broader substrate specificity than PTX.

In this study, we show that this toxin inhibits the coupling of all Gα_i/o/z G proteins, including Gα_z. Thus, we refer to it as Gα_O, Gα_Z and Gα_i inhibiting ToXin, or in short; OZITX. The active A subunit is functional when expressed in mammalian cells, bypassing the need for toxin purification. Moreover, we generate mutants of the members of the Gα_i subfamily that are OZITX insensitive, and hence, can serve as tools in combination with OZTIX treatment to investigate the role of individual Gα_i/o/z G proteins.

Results

OZITX treatment abolishes GPCR-mediated activation of all Gα_i subfamily members, including Gα_z

We hypothesised that OZITX may display a broader selectivity as compared to PTX because the Asn³⁴⁸ residue that is ADP ribosylated by OZITX is conserved in a greater number of Gα subunits as opposed to the cysteine modified by PTX, which is only present in Gα_i and Gα_o family members (Fig. 1a, b)²³. We first sought to determine whether OZITX inhibits coupling to members of the inhibitory G protein subfamily. To achieve this, we used a previously described bioluminescence resonance energy transfer (BRET) assay that measures the release of Gβγ subunits from the Gα subunits upon activation of the heterotrimer (Fig. 2a)^24,25. While this assay provides a method for rapidly assessing G protein activation, the signal may be partially contaminated by endogenously expressed Gα subunits^25,26. We, therefore, adapted the assay for use in HEK293A CRISPR/Cas ΔGα-all cells in which all the Gα subunits had been genetically knocked out²⁷. This allowed us to monitor the Gβγ release specifically from the activation of one Gα subtype of interest that had been exogenously transfected.

**Fig. 1: Identification of Gα carboxy-tail amino acid residues that are putatively ADP-ribosylated by OZITX.**

**Fig. 2: Activation of members of the Gαi subfamily in the presence of OZITX and PTX.**

The dopamine D₂ receptor (D₂R) promiscuously couples to Gα_i/o and Gα_z G proteins^28,29. Cells transiently expressing the D₂R were pre-incubated with PTX or OZITX followed by stimulation with the D₂-like receptor-selective agonist ropinirole³⁰. We observed that OZITX completely blocked the activation of Gα_i1, Gα_i2, Gα_i3, Gα_oA and Gα_oB (Fig. 2b). Further, as predicted from the carboxy-tail Asn³⁴⁸ residue presented in Gα_z, Gα_z could no longer couple to the D₂R following OZITX treatment as well (Fig. 2b). In contrast, Gα_z was insensitive to inhibition by pre-treatment of cells with PTX, consistent with the absence of the required cysteine residue (Fig. 1a). This finding extends the initial characterisation of OZITX, showing that unlike PTX it can inhibit Gα_z as well as Gα_i/o²³.

Next, analogous experiments were performed with another Gα_i/o/z-coupled GPCR; the μ opioid receptor. HEK293A CRISPR/Cas ΔGα cells transiently expressing the MOPR were pre-incubated with either OZITX or PTX and then stimulated with the agonist DAMGO (Fig. 2c). OZITX inhibited coupling to each of the Gα_i/o/z subunits completely (Fig. 2c). This showed, as expected, that OZITX does not discriminate between GPCRs when inhibiting Gα_i/o/z G-protein activation.

We then sought to further characterise the toxin by measuring the activation of Gα_i2 by the D₂R after exposure to OZITX at different timepoints. Gα_i2 activation decreased with increasing time of OZITX exposure until Gα_i2 activation was completely abolished approximately sixteen hours after the addition of OZITX (Fig. 2d). This is consistent with the characteristics of PTX and suggests that OZITX, like PTX, would be best utilised in the laboratory by incubating with the cells for more than 16 h (Supplementary Fig. 2).

OZITX does not ablate Gα_s, Gα_q or Gα₁₂ subfamily coupling

In addition to the inhibitory Gα G protein subfamily, the asparagine eight residues from the carboxy terminus is conserved in other Gα subfamily members (Fig. 1a). We therefore sought to further assess the substrate selectivity of OZITX across all Gα subunits. The Gα_s subfamily possesses a histidine residue instead of an asparagine in this position. In accordance with OZITX’s proposed mechanism of action, overnight incubation with OZITX did not inhibit Gα_s or Gα_olf activation by the dopamine D₁ receptor, a Gα_s/olf-coupled receptor, stimulated with the D₁R-selective agonist SKF83822 (Fig. 3a)^31,32,33,34.

**Fig. 3: Gαs, Gαq and Gα12 subfamily activation in presence of OZITX and PTX.**

Gα_q, Gα₁₁ and Gα₁₄, but not Gα₁₅, possess an asparagine residue eight residues from their C termini, so one might expect these three subunits to be substrates for OZITX (Fig. 1a). We measured the activation of the Gα_q subfamily proteins by the Gα_q-coupled neurotensin receptor 1 stimulated by the agonist NT8-13 with and without OZITX treatment^35,36. OZITX pre-treatment did not inhibit the activation of Gα_q, Gα₁₁ or Gα₁₅ although Gα₁₄ activation was slightly (~25%) decreased (vehicle control = 0.0840, OZITX-treated = 0.0644, P = 0.0012, one-way ANOVA with Dunnett’s multiple comparisons test) (Fig. 3b).

Both members of the Gα₁₂ subfamily; Gα₁₂ and Gα₁₃, also have asparagine as their eighth to last residue (Fig. 1a). The neurotensin receptor 1 is known to couple to the Gα₁₂ subfamily³⁷. While we were successful in detecting robust activation of Gα₁₂ and Gα₁₃, there was no inhibitory effect on the activation of either subunit when the cells were treated with OZITX (Fig. 3c). Taken together, we conclude that despite the presence of this asparagine residue at the C-terminus of Gα_q and Gα₁₂ subfamily members, no detectable inhibitory action of OZITX was observed, with the exception of the limited inhibition of Gα₁₄.

Inhibition of cAMP production by Gα_i2-, Gα_oA- and Gα_z is inhibited by OZITX

Cell-surface receptor signalling is commonly amplified in subsequent steps down the signalling cascade. We wanted to confirm that the apparent complete blockade of Gα_i/o/z signalling at the level of G-protein coupling would concord with measurements further downstream. We assessed the effect of OZITX treatment in measurements of intracellular cAMP levels using an intramolecular conformational BRET sensor of cAMP (CAMYEL), since the Gα_i subfamily bind and inhibit adenylate cyclases^4,38. In these experiments, we used HEK293A cells that harbour a genetic knockout of all the Gα_i subfamily members using CRISPR/Cas (HEK293A CRISPR/Cas ΔGα_i)²⁷. Individual Gα_i subunits of interest were then transfected into this cell background. Cells were treated with forskolin to stimulate adenylate cyclase, followed by treatment with ropinirole to stimulate the D₂R, leading to activation of the Gα_i/o/z subunit of interest. In the absence of a transfected Gα subunit, there was no detectable drug-induced inhibition of cAMP production (Fig. 4a). When Gα_i2 or Gα_oA were transfected, activation of the D₂R produced a decrease in relative cAMP levels (indicated by an increase in BRET ratio) and this was completely abolished in cells treated with OZITX (Fig. 4b, c). Cells transfected with Gα_z also produced a decrease in cAMP, albeit to a slightly smaller degree, and this was again blocked in the presence of OZITX (Fig. 4d). This confirms that OZITX-mediated ADP ribosylation inhibits downstream Gα_i/o/z-mediated signalling.

**Fig. 4: The effect of OZITX on Gαi2-, GαoA- and Gαz-mediated inhibition of cAMP production.**

The active A subunit of OZITX can be transfected into mammalian cells to act as an inhibitor

In order to treat cells with AB₅ toxin protein complexes, both expression and purification of this toxin are required²³. The active A subunit of PTX alone can be transiently expressed to inhibit Gα_i/o signalling^39,40. Accordingly, we tested whether the OZITX would be functional upon transfection of the cDNA encoding the active A subunit alone (OZITX-S1), thus increasing its accessibility and utility to laboratories. The cDNA sequence of OZITX-S1 was codon-optimised for high expression in human cells and co-transfected into HEK293T cells along with the D₂R, the WT Gα_i/o/z subunits and the G-protein activation sensors. Upon activation of the D₂R with the agonist quinpirole the responses in cells transfected with Gα_i1-3, Gα_oA and Gα_oB were inhibited in cells transfected with the positive control PTX-S1 cDNA as well as the OZITX-S1 cDNA (Fig. 5a–c and Supplementary Fig. 3). Importantly, while transfection of cells with OZITX-S1 inhibited Gα_z activation, transfection of PTX-S1 cDNA did not. Having shown that transfected OZITX-S1 is functional, we then confirmed the pattern of OZITX selectivity across the Gα subfamilies was in accord with our previous results using treatment with the complete OZITX protein complex (Supplementary Fig. 3). Consistent with these findings, the OZITX-S1 transfection was ineffective in abolishing the activation of Gα_s, Gα_q and Gα₁₂ subfamilies (Supplementary Fig. 3). In contrast to the partial inhibition of Gα₁₄ that we observed when using the purified OZITX (Fig. 3b), we did not observe inhibition of Gα₁₄ in experiments transiently expressing the OZITX-S1 (Supplementary Fig. 3).

**Fig. 5: OZITX sensitivity of Gαi subfamily carboxy-tail Asn347/348 mutants.**

ADP ribosylation of the C-terminal cysteine of Gα_i/o subunits by PTX is thought to inhibit the functional interaction between these G proteins and an activated GPCR. To test whether ADP ribosylation might also inhibit such an interaction we measured the recruitment of Gα_oA:G_β:G_γ-venus G protein heterotrimers to the D_2LR in the presence of either PTX-S1 or OZITX-S1 expression. Our data clearly show that both PTX and OZITX inhibit Gα_oA:G_β:G_γ-venus G-protein heterotrimer recruitment to the D₂R but that only OZITX inhibits Gα_z:G_β:G_γ-venus G-protein heterotrimer recruitment, as expected. This is consistent with a mechanism of action whereby ADP ribosylation of the C-terminal asparagine of Gα_i/o/z by OZITX prevents their coupling to GPCRs akin to the action of PTX at the C-terminal cysteine of Gα_i/o (Supplementary Fig. 4).

Gα_i subunits can be made OZITX insensitive for dissection of Gα_i/o/z subtype signalling specificity

Understanding the actions of a single Gα_i/o/z subtype can be challenging because there are usually multiple Gα_i/o/z members expressed within any given cell type. A method that has permitted the investigation of the role of individual Gα_i/o/z subunits in a particular signalling process, such as coupling to a specific GPCR, is the use of PTX-insensitive Gα_i mutants in combination with PTX to uncouple any endogenously expressed PTX-sensitive Gα_i subunits⁴¹. We, therefore, wanted to generate OZITX-insensitive Gα_i/o/z mutants in the hope of increasing the scope of OZITX applications. To render the Gα_i/o/z subunits insensitive to OZITX, we replaced the asparagine eight residues from the carboxy terminus to an alanine (Gα_i1-N347A, Gα_i2-N348A, Gα_i3-N347A, Gα_oA-N347A, Gα_oB-N347A and Gα_z-N348A) as this residue was previously identified as the most likely substrate site (Fig. 1a)²³. We then performed G protein-activation assays using the D₂R to activate each Gα_i mutant in the presence or absence of PTX-S1 or OZITX-S1 (Fig. 5 and Supplementary Figs. 5 and 6). In contrast to the activation of the wild-type Gα_i3, Gα_oA and Gα_z that are all abolished by OZITX (Fig. 5a–c), activation of Gα_i3-N347A, Gα_oA-N347A and Gα_z-N348A were OZITX insensitive (Fig. 5d–f). When these mutant Gα subunits were transfected the potency of quinpirole was similar to that observed in the case of the WT Gα subunit, suggesting that these mutations did not affect the efficiency with which they couple to the D₂R. In addition, it was observed that the N347A/N348A mutation did not impact the PTX sensitivity of the Gα_i subunits (Fig. 5d–f). Likewise, the well-characterised PTX-insensitive mutation (C351I) introduced into Gα_i3 and Gα_oA, did not disturb the ability of OZITX to act on them (Fig. 5g, h). Having identified that the N347A/N348A mutation renders these Gα_i members insensitive to OZITX without perturbation, the mutations were extended into the remaining Gα_i subunits and validated (Supplementary Figs. 5 and 6).

The C-terminal ten residues of Gα_i are sufficient to confer OZITX selectivity

The asparagine residue important for the inhibitory action of OZITX is present in both the G_i/o/z and the G_q subfamilies (with the exception of Gα₁₅). Thus, the selective action of OZITX at the G_i/o/z G-protein α subunits appears not to be determined solely by the presence or absence of this residue. Similarly, a previous study has shown that replacement of the 5 C-terminal residues of Gα_q with that of Gα_i (which includes the cysteine that is modified by PTX) allows this chimeric Gα_qi5 G protein to couple to G_i protein family-coupled receptors but, importantly, does not confer sensitivity to PTX⁴². This suggests that there are determinants of PTX in addition to the presence of this cysteine residue. To explore other determinants of OZITX selectivity we generated two chimeric G proteins in which the last 10 (Gα_qi10) or 13 (Gα_qi13) residues of Gα_q were replaced with those of Gα_i3, a region that includes the asparagine residue modified by OZITX. In an assay measuring Ca²⁺ mobilisation, OZITX was unable to inhibit the Ca²⁺ response of the muscarinic M₃ acetylcholine receptor co-expressed with Gα_q when activated by the agonist carbachol (Supplementary Fig. 7). In this Ca²⁺ mobilisation assay we were unable to detect a measurable response to the agonist ropinirole when the D₂R was co-expressed with Gα_q, but we observed responses when the D₂R was co-expressed with both chimeric G proteins, meaning the D₂R can couple to both Gα_qi10 and Gα_qi13. Interestingly, both PTX and OZITX could inhibit these responses, indicating that the last 10 residues of Gα_i are sufficient to confer the sensitivity to both PTX and OZITX inhibition on a Gα_q background (Supplementary Fig. 7). While both PTX and OZITX partially inhibited the Gα_qi10 Ca²⁺ signal, the expression of both toxins completely inhibited Gα_qi13 signalling. This pattern is consistent with the idea that the greater the amount of C-terminal Gα_i amino acids swapped with those of Gα_q, the better the chimeric Gα_qi proteins become as substrates for both αβ₅ toxins. However, the lower potency of ropinirole in the Gα_qi13 Ca²⁺ assay suggests that the coupling of the D₂R to this chimera is less efficient, which may also account for the apparently greater inhibitory effect of PTX and OZITX (Supplementary Fig. 7).

Discussion

For decades, PTX and CTX have proven to be useful tools in GPCR signalling research to interrogate the Gα protein subfamilies or even specific Gα proteins responsible for particular physiological processes. Here, we characterise and demonstrate the utility of OZITX, a recently identified AB₅ toxin, for the inhibition of GPCR-mediated activation of the Gα_i/o/z subfamily. Importantly, unlike PTX, this activity extends to include Gα_z. OZITX acts to ADP-ribosylate an asparagine in the C-terminus of Gα_i/o/z proteins, a site distinct from the cysteine modified by PTX, accounting for this broader specificity. We found that OZITX displays a selective action to completely inhibit Gα_i/o/z proteins with no activity at Gα_s, Gα_q or Gα₁₂ proteins, with the exception of limited inhibition of Gα₁₄. The catalytic subunit of PTX (PTX-S1) can be expressed in mammalian cells to effectively inhibit Gα_i/o signalling, avoiding the time and cost associated with acquiring the purified protein^39,40. We demonstrate that the catalytic OZITX-S1 subunit can be used in a similar manner, increasing the utility of this tool. We identify mutations within Gα_i/o/z subfamily members that render them insensitive to OZITX and maintain their ability to couple to GPCRs. Together, these tools can be used to identify the Gα_i/o/z subunits participating in defined signalling pathways.

PTX played an important role in identifying the Gα_i subfamily by distinguishing it from the Gα_s subfamily⁴. PTX was shown to block the inhibitory effect that Gα_i proteins have on adenylyl cyclases, thus building evidence for a separate Gα species with distinct functionality to Gα_s. Since then, PTX has been widely used with the same rationale, that is, to differentiate GPCR responses mediated by Gα_i proteins from other signal transducers⁴³. However, such an approach cannot exclude the possibility that Gα_z might contribute to a particular response since it is insensitive to PTX^17,18. A clear advantage of OZITX, then, is that it can inhibit Gα_z in addition to inhibiting Gα_i1, Gα_i2, Gα_i3 and the Gα_o isoforms. We have not evaluated whether OZITX inhibits the coupling of the visual and taste Gα subunits; Gα_t1, Gα_t2 and Gα_gust. One might expect ADP ribosylation by OZITX to occur on Gα_t1, Gα_t2 and Gα_gust since they harbour an asparagine as their eighth to last amino acid residue in addition to having high sequence homology to the other Gα_i subunits, although given our findings that not all Gα subunits that contain this asparagine are inhibited by OZITX, inhibitor activity at Gα_t1, Gα_t2 and Gα_gust must be determined experimentally.

Our findings suggest that OZITX could serve as a replacement for PTX in most experimental paradigms aimed at interrogating Gα_i/o/z G-protein signalling. There are, however, cases where PTX and OZITX can be used in parallel due to their different Gα specificities, for example when disentangling the functions of Gα_z from the other Gα_i/o subunits. OZITX-treated, PTX-treated and -untreated conditions run in parallel would enable the signalling mediated by Gα_z, PTX-sensitive Gα_i subunits and toxin-insensitive Gα subunits to be isolated.

Previous studies aimed at interrogating Gα_z signalling have relied on other strategies, including overexpression of Gα_z-specific RGS proteins¹⁹, Gα_z-directed siRNA⁴⁴, and Gα_z de-activation via PKC phosphorylation⁴⁵. However, unlike OZITX, these approaches do not completely block activation of Gα_z so the influence of residual Gα_z signalling cannot be excluded, particularly when looking at an effect further down an amplified signalling cascade. Genetic knockouts of the gene that encodes Gα_z have been used for this reason but are technically challenging as compared to OZITX treatment^20,46. In addition, the results of such knockout approaches may be confounded by adaptive changes to the cell and/or circuit over time that compensate for the loss of that particular protein. The advantage of OZITX is that it can be used in a relatively acute manner following overnight treatment, so its use is less likely to be confounded by compensatory changes in cell function.

The substrate site that is ADP ribosylated by OZITX was shown to be an asparagine eight residues from the C-terminus of the Gα subunit²³. In agreement with this, we showed that Gα_i subunits can be made OZITX insensitive through mutation of the aligned asparagine in this position. Within our set of experiments, we observed that these mutations did not affect the potency or magnitude of the measured response as compared to when the WT Gα was used. This indicates that this mutation has not changed the coupling efficiency between the receptor and Gα subunit. It should be acknowledged, however, that these observations may be both receptor and downstream effector dependent. It may be that other mutations at this position may be superior to the alanine mutation for a particular combination, as has been observed for analogous studies using PTX and PTX-insensitive Gα mutants⁴⁷. Nonetheless, the OZITX-insensitive mutants can serve as a useful tool in combination with OZITX treatment to investigate the signalling of particular Gα_i/o/z proteins in isolation. In our hands, mutation of the Asn^347/348 residue alone was sufficient to render Gα_i1, Gα_i2 and Gα_i3 resistant to OZITX. These Gα subunits contain a lysine residue as their tenth-to-last residue (Lys^345/346) that was suggested to also be a site for OZITX-mediated ADP ribosylation by Littler and colleagues (Fig. 1a)²³. Our results suggest that this Lys^345/346 site is either a secondary substrate site that is minimally ADP ribosylated by OZITX, that ADP ribosylation of this site has no effect on G-protein coupling despite this residue being in close proximity to the GPCR upon coupling, or that the ribosylation of this lysine occurs sequentially to that of Asn^347/348 (Fig. 1b), such that the mutation of the asparagine residue is sufficient to abrogate reaction.

We hypothesised that the presence of an asparagine residue eight residues from the C-terminus of various Gα subunits would confer sensitivity to OZITX in a similar manner to the way in which the presence of Cys^351/352 confers sensitivity to PTX. In agreement with this, Gα_i/o/z proteins were inhibited by OZITX whereas Gα_s, which lacks this asparagine, was not. We observed, however, that OZITX only had a small inhibitory effect on Gα₁₄ activation and had no effect on the remaining Gα_q and Gα₁₂ subunits despite the presence of the aligned asparagine. Thus, this asparagine is not the only determinant of selectivity. This lack of OZITX sensitivity can be reconciled either with OZITX not ADP-ribosylating the asparagine residue in G_q and G₁₂ family Gα subunits or with ADP ribosylation of this residue in Gα_q and Gα₁₂ not affecting GPCR coupling and signalling. Prior studies have shown that swapping the five carboxy-terminal residues of Gα_i2 or Gα_oA onto Gα_q is not sufficient to confer sensitivity to PTX, even though the modified Gα_q contains the required cysteine residue four amino acids from the carboxy-termini⁴². This indicates that carrying the required substrate amino acid site is not sufficient to render the Gα subunit sensitive to PTX-like AB₅ toxins. Our own experiments using chimeras in which the last ten or thirteen amino acids of Gα_q were swapped with those of Gα_i2 revealed that exchange of this region was sufficient to confer both PTX and OZITX sensitivity. We computationally explored the feasibility of ADP-ribosylation of Gα subunits in the D₂R-Gα_i and the 5HT_2A–Gα_q complexes, by covalently docking the ADP-ribose moiety on the asparagine residue using the available cryo-EM structures of these complexes^48,49. Our results reveal that the 5HT_2A–Gα_q complex can easily accommodate the ADP-ribosylated asparagine whereas this covalent docking approach could not identify any feasible pose for the D₂R–Gα_i complex without steric clash (Supplementary Fig. 8). Even though ADP-ribosylation of the asparagine can be sterically accommodated in the 5HT_2A–Gα_q complex, we cannot rule out the possibility that this might still impact coupling, and as we have noted above, the residue may simply not be ADP-ribosylated due to the absence of other key determinants beyond the asparagine itself. Further studies are required to understand the additional structural basis for the recognition of specific Gα subunits by AB₅-type toxins such as OZITX and PTX and to understand the selective action of OZITx for G_i/o/z family G proteins.

Our study illustrates the continuing value in the characterisation and use of AB₅ toxins as laboratory tools. Host–pathogen interactions are hotspots of molecular evolution that result in proteins with extraordinary functionality. This is exemplified in the diversity of actions of ADP-ribosylating AB₅ toxins including PTX and CTX and now OZITX and their resulting value as research tools.

Methods

Materials

Polyethylenimine (PEI), Linear (MW 25,000) was purchased from Polysciences, Inc. Ropinirole was purchased from BetaPharma (Shanghai) Co. Ltd. DAMGO ((_D-Ala², N-Me-Phe⁴, Gly-ol⁵)-enkephalin) was purchased from Mimotopes. SKF83822, neurotensin residues 8-13 (NT8-13), (−)-quinpirole hydrochloride (#1061), acetylcholine chloride (#A2661), carbachol (C4382), D-glucose (#G8270) and pertussis toxin (PTX) were purchased from Sigma-Aldrich. Isoproterenol (#1747) and endothelin-1 (#1160) were purchased from Tocris Bioscience (Bristol, UK). Coelenterazine-h was purchased from both NanoLight™ Technology and Dalton research molecules (#50303-86-9). Forskolin was purchased from Cayman Chemicals (#11018). Nano-Glo^® luciferase assay system, containing the furimazine substrate, was purchased from Promega.

Plasmids

pcDNA3.1(+) encoding human constructs of long isoform of the dopamine D₂ receptor (D_2LR), μ opioid receptor (MOPR), dopamine D₁ receptor (D₁R), neurotensin receptor 1 (NTS₁R), M₁ muscarinic acetylcholine receptor (M₁R), β₂-adrenergic receptor (β₂AR), endothelin A receptor (ET_AR), Gα_i1, Gα_i2, Gα_i3, Gα_oA, Gα_oB, Gα_z, Gα_sS, Gα_sL, Gα_olf, Gα_q, Gα₁₁, Gα₁₄, Gα₁₅-EE, Gα₁₂ and Gα₁₃ were from the cDNA Resource Centre (cDNA.org). pcDNA3L-His-CAMYEL was purchased from ATCC (ATCC MBA-277). masGRK3ct-Nluc, masGRK3ct-Rluc8, venus-1-155-Gγ₂ and venus-156-239-Gβ₁ were from Nevin Lambert, Augusta University. pCAGGS-Ric8A and pCAGGS-Ric8B were from Asuka Inoue, Tohoku University. The active S1 subunit of OZITX (EcPltAB) was codon-optimised, synthesised and inserted into pcDNA3.1 + (see Supplementary Note 1 for sequence). OZITX-resistant mutations were made in Gα_i1, Gα_i2, Gα_i3, Gα_oA Gα_oB and Gα_z using site-directed mutagenesis. Primer sequences that were used for the mutagenesis can be found in Supplementary Table 1.

OZITX-resistant mutations were made by changing the eighth to last amino acid to an alanine (indicated in red) by using site-directed mutagenesis with the reverse primers used to the right, the alanine mutation change is shown in red and restriction sites chosen in blue (XhoI) or green (XbaI). The constructs were inserted into pcDNA3.1+ with KpnI and XhoI or XbaI as indicated. The two chimeric proteins Gα_qi10 and Gα_qi13 were generated using the Q5 site-directed mutagenesis kit from NEB. Primer sequences used for the mutagenesis can be found in Supplementary Table 2. PCR products were then treated with the KLD enzyme mix (kinase, ligase and DpnI) provided with the kit and then transform into NEB Turbo E. coli competent cells.

Cell culture

HEK293T cells were purchased from ATCC (CRL-3216). HEK293A ΔGα-all CRISPR/Cas knockout cells and HEK293A ΔGα_i/o CRISPR/Cas knockout cells were generated as described before²⁷. HEK293T cells, HEK293A ΔGα-all cells and HEK293A ΔGα_i/o cells were cultured in T175 flasks with Dulbecco’s Modified Eagle Medium (DMEM) + GlutaMAX^TM-I (Gibco, Invitrogen, Paisley, UK) with 10% foetal bovine serum (Corning #35-010) and 1% penicillin/streptomycin (Corning #30-002). All Cells were grown in a humidified incubator in 5% CO₂ at 37 °C and sub-cultured at a ratio of 1/10-1/20.

Transfection

Briefly, cells were harvested from T175 flasks and plated into six-well Nunc™ tissue culture plates at a density of 500,000 cells per well. The following day, the media was removed and replaced with fresh media and transfected using PEI as the transfection reagent. The corresponding amounts of PEI and DNA were added to the buffer separately before mixing together, incubating for 20 minutes, and then adding dropwise on top of the cells in the fresh media.

For the G-protein-activation assays where the toxin was added exogenously: The HEK293A ΔGα-all CRISPR knockout cells were transfected using PEI in a ratio of 6:1 PEI:DNA (w/w) in PBS. The cells were transfected with a cDNA mixture consisting of: 0.143 µg GPCR, 0.286 µg Gα, 0.143 µg Gβ₁-venus, 0.143 µg Gγ₂-venus, 0.143 µg masGRK3ct-Nluc and 0.143 µg Ric8A or Ric8B or pcDNA3.1. The chaperone Ric8A was transfected together with Gα₁₄ and Gα₁₅ and Ric8B was transfected with Gα_olf.

For the cAMP BRET assays, where the toxins were exogenously added: The HEK293A ΔGα_i/o CRISPR knockout cells were transfected using PEI in a ratio of 6:1 PEI:DNA (w/w) in PBS. The cells were transfected with a cDNA mixture consisting of: 0.143 µg D_2LR, 0.286 µg Gα_i2/Gα_oA/Gα_z/pcDNA3.1 and 0.429 µg CAMYEL sensor.

Assays, where the active A subunits of the toxins were transiently transfected, had the following conditions: HEK293T cells were transfected using PEI in a ratio of 1.5 PEI:1 DNA (w/w) mixed in 150 mM NaCl For the G-protein activation assays the cells were transfected with 0.500 µg β₁, 0.500 µg Venus-γ₂ and 0.100 µg masGRKctRluc8 as well as 1 µg of the G protein of interest together with 0.375 µg of a receptor suited for the specific G-protein class and 0.375 µg of the helper proteins Ric8A for Gα₁₄ and Gα₁₅ and Ric8B for Gα_olf and finally 0.200 µg of either the active subunit of PTX (PTX-S1), OZITX (OZITX-S1) or pcDNA3.1+ as a control. For the cAMP production inhibition assays the cells were co-transfected with 1 µg of the CAMYEL sensor (ATCC MBA-277). For the Ca²⁺ assay, the cells were transfected with 0.3 µg of the receptor (M3R or D2LR), 0.3 µg of Gα_q/qi10/qi13 and 0.2 µg of either S1-PTX/S1-OZITX or pcDNA3.

G-protein activation

G-protein activation was measured using a BRET assay that monitors Gβγ release^24,25. The HEK293A ΔGα-all cells were first transfected as described earlier and the following day the cells were harvested and transferred into white 96-well CulturPlates (PerkinElmer) in DMEM + 10% FBS. In the conditions where the cells were treated with OZITX or PTX, the cells were left to adhere before being treated in the 96-well plate 16–20 h before performing the assay. The G-protein activation assays were then performed ~24 h after plating out the transfected cells. The media in each well was aspirated, washed with Hank’s balanced salt solution pH 7.4 (HBSS), replaced with HBSS and then kept at 37 °C for the remainder of the assay. Furimazine was added with a multi-stepper pipette 15 min before agonist addition and left to equilibrate. The agonist was then added, and cells were incubated in a LUMIstar Omega (BMG Labtech) plate reader. The BRET measurements were then taken 2.5 min after agonist addition. Simultaneous dual emission filters were used in the LUMIstar Omega for detection of the luciferase at 445–505 nm and venus at 505-565 nm, all measured at 37 °C. For G-protein activation assays where the toxin active A subunit cDNAs were transfected, the same protocol was followed with some exceptions: HEK293T cells were used instead of CRISPR/Cas gene-edited cells, DPBS + 5 mM glucose was used as the assay buffer, 96-well black-white isoplates were used, and the plate was detected five minutes after agonist stimulation in a PHERAstar FS (BMG Labtech). After acquiring the data, the ratio of the venus emission channel was then divided by the luciferase emission channel to determine the BRET ratio. The vehicle-subtracted raw BRET ratio (drug-induced increase in BRET) is plotted for the G-protein activation assay data.

Gγ-mVenus recruitment assay

For the D₂R-mediated Gγ-mVenus recruitment assay, HEK293T cells were seeded onto six-well plates and transfected with a 1:6 ratio of DNA:polyethylenimine with plasmids encoding D2R-nluc (50 ng), Gα_oA/z WT (cDNA resource centre, Bloomsburg University, PA, 125 ng), human Gβ1 (250 ng) and human Gγ2- mVenus (250 ng). Cells were harvested from six-well plates 24 h after transfection and plated into poly-D-lysine coated (Sigma-Aldrich) white-bottom 96-well optiplates (Wallac, PerkinElmer Life and Analytical Sciences) at a density of 50,000 cells per well. Twenty-four hours after cells were transferred to plates, media was aspirated, cells washed once with DPBS and 80 μL DPBS containing 5 mM glucose was added to each well. In all, 10 μL of coelenterazine was added to each well and the plate read on the PHERAstar FS (BMG Labtech) for 5 min, paused for the addition of 10 μL agonists and read again for 10 min. After acquiring the data, the ratio of the venus emission channel was then divided by the luciferase emission channel to determine the BRET ratio. The vehicle-subtracted raw BRET ratio (drug-induced increase in BRET) is plotted for the G-protein recruitment assay data.

Gα_i/o/z-mediated inhibition of cAMP production

The cAMP production inhibition assays’ principle is based on the ability of a genetically encoded conformational BRET sensor to detect the relative concentrations of intracellular cAMP⁵⁰. Initially, the transfected HEK293A ΔGα_i/o cells were harvested and transferred into white 96-well CulturPlates in DMEM + 10% FBS 24 h after the transient transfection. When the cells were treated with OZITX or PTX, this occurred in the 96-well plate after adherence and about 18 h before the assay. Next, the cAMP inhibition assays were performed the following day after plating out the transfected cells and toxin or control treatment. On the day of the assay, the plate media was aspirated, washed once with HBSS pH 7.4 and replenished with HBSS pH 7.4 and then held at 37 °C for the rest of the experiment. In total, 5 μM coelenterazine-h was added 15 min before agonist addition. 10 µM Forskolin was added 10 min before agonist addition and the readings were then continuously taken in the live cells. Bioluminescence was detected on a LUMIstar Omega set to 37 °C. Simultaneous dual emission filters were used for the BRET donor at 445–505 nm and the acceptor at 505-565 nm. The ratio of the acceptor channel was then divided by the donor channel to determine the BRET ratio. The data was then baseline-corrected to the vehicle control wells over time. A slightly modified protocol was followed for the assays where the active subunit cDNAs of the toxins were transfected: HEK293T cells were used instead of the HEK293A ΔGα_i/o cells, 96-well black-white isoplates were used, DPBS + 5 mM glucose was used as the assay buffer, a higher concentration of 30 μM forskolin was used and this was co-added with the coelenterazine-h ten minutes prior to the addition of the agonist. The plate was then detected 20 min after agonist addition in a PHERAstar FS.

Ca²⁺ mobilisation assays

Cells were seeded in a clear bottom black 96-well plate coated with poly-d-lysine (50 μg/ml) at 100,000 cells per well. The following day, cells were washed with 100 μl of 1× HBSS assay buffer supplemented with 10 mM glucose, 4 mM probenic acid at pH 7.4 and brilliant black, and then loaded with 100 μl Fluo-4 AM (1 μM) (prepared in DMSO and pluronic acid) for 45 min at 37 °C (no CO₂). The release of Ca²⁺ was measured using a Flexstation 3 (Molecular Devices; Sunnyvale, CA). Drug dilutions were prepared in assay buffer (without Fluo-4) at 6x required concentration (volume 20 μl in 100 μl in Flexstation protocol) and transferred to a loading plate (transparent flat-bottom 96-well plate). Fluorescence was detected for 3 min30 s at 485 nm excitation and 525 nm emission. Relative fluorescence units were normalised to the fluorescence stimulated by ionomycin to account for differences in cell number and loading efficiency.

Data analysis, statistics and reproducibility

GraphPad Prism 8 was used for data analysis and performing statistical tests. Statistical analysis was carried out with a Student’s t test or one-way ANOVA followed by a post hoc test where appropriate. P values <0.05 were considered statistically significant. Data sets were of at least n = 3 and the experimental n number is stated for each data set in the corresponding figure legend. Figures depicting molecular structures were constructed using ICM-Browser (MolSoft LLC) and UCSF Chimera⁵¹. The covalent docking was carried out with the covDock module of Schrodinger suite (version 2021-1), assuming a SN2 nucleophilic substitution reaction, which results in the a-orientation of the attached ADP-ribose moiety on the asparagine⁵².

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

All original data have been deposited with Communications Biology in Supplementary Data 1 and are also available from the corresponding authors upon reasonable request.

Materials availability

The cDNA for the S1 subunit of OZITX as described in the manuscript will be made available on reasonable request.

References

Northup, J. K. et al. Purification of the regulatory component of adenylate cyclase. Proc. Natl Acad. Sci. USA 77, 6516–6520 (1980).
Article CAS PubMed PubMed Central Google Scholar
Taylor, S. J., Smith, J. A. & Exton, J. H. Purification from bovine liver membranes of a guanine nucleotide-dependent activator of phosphoinositide-specific phospholipase C. Immunologic identification as a novel G-protein alpha subunit. J. Biol. Chem. 265, 17150–17156 (1990).
Article CAS PubMed Google Scholar
Hescheler, J., Rosenthal, W., Trautwein, W. & Schultz, G. The GTP-binding protein, Go9 regulates neuronal calcium channels. Nature 325, 445 (1987).
Article CAS PubMed Google Scholar
Hildebrandt, J. D. et al. Stimulation and inhibition of adenylyl cyclases mediated by distinct regulatory proteins. Nature 302, 706 (1983).
Article CAS PubMed Google Scholar
Buhl, A. M., Johnson, N. L., Dhanasekaran, N. & Johnson, G. L. Gα12 and Gα13 stimulate Rho-dependent stress fiber formation and focal adhesion assembly. J. Biol. Chem. 270, 24631–24634 (1995).
Article CAS PubMed Google Scholar
Sullivan, K. A. et al. Identification of receptor contact site involved in receptor–G protein coupling. Nature 330, 758 (1987).
Article CAS PubMed Google Scholar
Hamm, H. E. et al. Site of G protein binding to rhodopsin mapped with synthetic peptides from the alpha subunit. Science 241, 832–835 (1988).
Article CAS PubMed Google Scholar
Lee, E., Taussig, R. & Gilman, A. G. The G226A mutant of Gs alpha highlights the requirement for dissociation of G protein subunits. J. Biol. Chem. 267, 1212–1218 (1992).
Article CAS PubMed Google Scholar
Digby, G. J., Lober, R. M., Sethi, P. R. & Lambert, N. A. Some G protein heterotrimers physically dissociate in living cells. Proc. Natl. Acad. Sci. USA 103, 17789–17794 (2006).
Article CAS PubMed PubMed Central Google Scholar
Ng, N. M. et al. EcxAB is a founding member of a new family of metalloprotease AB5 toxins with a hybrid cholera-like B subunit. Structure 21, 2003–2013 (2013).
Article CAS PubMed Google Scholar
Endo, Y. et al. Site of action of a Vero toxin (VT2) from Escherichia coli O157: H7 and of Shiga toxin on eukaryotic ribosomes: RNA N‐glycosidase activity of the toxins. Eur. J. Biochem. 171, 45–50 (1988).
Article CAS PubMed Google Scholar
Katada, T. & Ui, M. Direct modification of the membrane adenylate cyclase system by islet-activating protein due to ADP-ribosylation of a membrane protein. Proc. Natl. Acad. Sci. USA 79, 3129–3133 (1982).
Article CAS PubMed PubMed Central Google Scholar
Cassel, D. & Pfeuffer, T. Mechanism of cholera toxin action: covalent modification of the guanyl nucleotide-binding protein of the adenylate cyclase system. Proc. Natl. Acad. Sci. USA 75, 2669–2673 (1978).
Article CAS PubMed PubMed Central Google Scholar
Orth, J. H. et al. Pasteurella multocida toxin activation of heterotrimeric G proteins by deamidation. Proc. Natl. Acad. Sci. USA 106, 7179–7184 (2009).
Article CAS PubMed PubMed Central Google Scholar
Katada, T. & Ui, M. ADP ribosylation of the specific membrane protein of C6 cells by islet-activating protein associated with modification of adenylate cyclase activity. J. Biol. Chem. 257, 7210–7216 (1982).
Article CAS PubMed Google Scholar
Casey, P. J., Fong, H. K., Simon, M. I. & Gilman, A. G. Gz, a guanine nucleotide-binding protein with unique biochemical properties. J. Biol. Chem. 265, 2383–2390 (1990).
Article CAS PubMed Google Scholar
Fong, H., Yoshimoto, K. K., Eversole-Cire, P. & Simon, M. I. Identification of a GTP-binding protein alpha subunit that lacks an apparent ADP-ribosylation site for pertussis toxin. Proc. Natl. Acad. Sci. USA 85, 3066–3070 (1988).
Article CAS PubMed PubMed Central Google Scholar
Matsuoka, M., Itoh, H., Kozasa, T. & Kaziro, Y. Sequence analysis of cDNA and genomic DNA for a putative pertussis toxin-insensitive guanine nucleotide-binding regulatory protein alpha subunit. Proc. Natl. Acad. Sci. USA 85, 5384–5388 (1988).
Article CAS PubMed PubMed Central Google Scholar
Kimple, M. E. et al. A role for Gz in pancreatic islet β-cell biology. J. Biol. Chem. 280, 31708–31713 (2005).
Article CAS PubMed Google Scholar
Kimple, M. E. et al. Gαz negatively regulates insulin secretion and glucose clearance. J. Biol. Chem. 283, 4560–4567 (2008).
Article CAS PubMed Google Scholar
Brill, A. L. et al. Synergy between Gαz deficiency and GLP-1 analog treatment in preserving functional β-cell mass in experimental diabetes. Mol. Endocrinol. 30, 543–556 (2016).
Article CAS PubMed PubMed Central Google Scholar
Fenske, R. J. et al. The inhibitory G protein α-subunit, Gαz, promotes type 1 diabetes-like pathophysiology in NOD mice. Endocrinology 158, 1645–1658 (2017).
Article CAS PubMed PubMed Central Google Scholar
Littler, D. R. et al. Structure–function analyses of a pertussis-like toxin from pathogenic Escherichia coli reveal a distinct mechanism of inhibition of trimeric G-proteins. J. Biol. Chem. 292, 15143–15158 (2017).
Article CAS PubMed PubMed Central Google Scholar
Hollins, B., Kuravi, S., Digby, G. J. & Lambert, N. A. The c-terminus of GRK3 indicates rapid dissociation of G protein heterotrimers. Cell. Signal. 21, 1015–1021 (2009).
Article CAS PubMed PubMed Central Google Scholar
Masuho, I., Martemyanov, K. A. & Lambert, N. A. in G Protein-Coupled Receptors in Drug Discovery: Methods and Protocols (ed. Marta, F.) 107–113 (Springer New York, 2015).
Masuho, I., Xie, K. & Martemyanov, K. A. Macromolecular composition dictates receptor and G protein selectivity of regulator of G protein signaling (RGS) 7 and 9-2 protein complexes in living cells. J. Biol. Chem. 288, 25129–25142 (2013).
Article CAS PubMed PubMed Central Google Scholar
Hisano, Y. et al. Lysolipid receptor cross-talk regulates lymphatic endothelial junctions in lymph nodes. J. Exp. Med. 216, 1582–1598 (2019).
Article CAS PubMed PubMed Central Google Scholar
Lane, J. R., Powney, B., Wise, A., Rees, S. & Milligan, G. G protein coupling and ligand selectivity of the D2L and D3 dopamine receptors. J. Pharmacol. Exp. Therapeutics 325, 319–330 (2008).
Article CAS Google Scholar
Donthamsetti, P. et al. Arrestin recruitment to dopamine D2 receptor mediates locomotion but not incentive motivation. Mol. Psychiatry 25, 2086–2100 (2018).
Eden, R. et al. Preclinical pharmacology of ropinirole (SK&F 101468-A) a novel dopamine D2 agonist. Pharmacol. Biochem. Behav. 38, 147–154 (1991).
Article CAS PubMed Google Scholar
Dearry, A. et al. Molecular cloning and expression of the gene for a human D1 dopamine receptor. Nature 347, 72 (1990).
Article CAS PubMed Google Scholar
Zhou, Q.-Y. et al. Cloning and expression of human and rat Dt dopamine receptors. Nature 347, 76 (1990).
Article CAS PubMed Google Scholar
Seeman, P. & Niznik, H. Dopamine D1 receptor pharmacology. ISI Atlas Sci. Pharm. 2, 161–170 (1988).
CAS Google Scholar
Andersen, P. H. & Jansen, J. A. Dopamine receptor agonists: selectivity and dopamine D1 receptor efficacy. Eur. J. Pharmacol.: Mol. Pharmacol. 188, 335–347 (1990).
Article CAS Google Scholar
Wu, T., Li, A. & Wang, H.-L. Neurotensin increases the cationic conductance of rat substantia nigra dopaminergic neurons through the inositol 1, 4, 5-trisphosphate-calcium pathway. Brain Res. 683, 242–250 (1995).
Article CAS PubMed Google Scholar
Wang, H.-L. & Wu, T. Gαq11 mediates neurotensin excitation of substantia nigra dopaminergic neurons. Mol. brain Res. 36, 29–36 (1996).
Article CAS PubMed Google Scholar
Besserer-Offroy, É. et al. The signaling signature of the neurotensin type 1 receptor with endogenous ligands. Eur. J. Pharmacol. 805, 1–13 (2017).
Article CAS PubMed Google Scholar
Hildebrandt, J., Hanoune, J. & Birnbaumer, L. Guanine nucleotide inhibition of cyc-S49 mouse lymphoma cell membrane adenylyl cyclase. J. Biol. Chem. 257, 14723–14725 (1982).
Article CAS PubMed Google Scholar
Vivaudou, M. et al. Probing the G-protein regulation of GIRK1 and GIRK4, the two subunits of the KACh channel, using functional homomeric mutants. J. Biol. Chem. 272, 31553–31560 (1997).
Article CAS PubMed Google Scholar
Castro, M. G., McNamara, U. & Carbonetti, N. H. Expression, activity and cytotoxicity of pertussis toxin S1 subunit in transfected mammalian cells. Cell. Microbiol. 3, 45–54 (2001).
Article CAS PubMed Google Scholar
Senogles, S. E. The D2 dopamine receptor isoforms signal through distinct Gi alpha proteins to inhibit adenylyl cyclase. A study with site-directed mutant Gi alpha proteins. J. Biol. Chem. 269, 23120–23127 (1994).
Article CAS PubMed Google Scholar
Joshi, S. A., Fan, K. P., Ho, V. W. & Wong, Y. H. Chimeric Gαq mutants harboring the last five carboxy‐terminal residues of Gαi2 or Gαo are resistant to pertussis toxin‐catalyzed ADP‐ribosylation. FEBS Lett. 441, 67–70 (1998).
Article CAS PubMed Google Scholar
Reisine, T. Pertussis toxin in the analysis of receptor mechanisms. Biochemical Pharmacol. 39, 1499–1504 (1990).
Article CAS Google Scholar
Doi, M. et al. Gpr176 is a Gz-linked orphan G-protein-coupled receptor that sets the pace of circadian behaviour. Nat. Commun. 7, 10583 (2016).
Article CAS PubMed PubMed Central Google Scholar
Gonzalez-Iglesias, A. E., Murano, T., Li, S., Tomić, M. & Stojilkovic, S. S. Dopamine inhibits basal prolactin release in pituitary lactotrophs through pertussis toxin-sensitive and-insensitive signaling pathways. Endocrinology 149, 1470–1479 (2007).
Article PubMed PubMed Central CAS Google Scholar
van den Buuse, M. et al. Enhanced effect of dopaminergic stimulation on prepulse inhibition in mice deficient in the alpha subunit of Gz. Psychopharmacology 183, 358–367 (2005).
Article CAS PubMed Google Scholar
Bahia, D. S. et al. Hydrophobicity of residue351 of the G protein Gi1α determines the extent of activation by the α2A-adrenoceptor. Biochemistry 37, 11555–11562 (1998).
Article CAS PubMed Google Scholar
Kim, K. et al. Structure of a hallucinogen-activated Gq-coupled 5-HT2A serotonin receptor. Cell 182, 1574–1588 (2020).
Article CAS PubMed PubMed Central Google Scholar
Zhuang et al. Structural insights into the human D1 and D2 dopamine receptor signaling complexes. Cell 184, 931–942 (2021).
Article CAS PubMed PubMed Central Google Scholar
Jiang, L. I. et al. Use of a cAMP BRET sensor to characterize a novel regulation of cAMP by the sphingosine 1-phosphate/G13 pathway. J. Biol. Chem. 282, 10576–10584 (2007).
Article CAS PubMed Google Scholar
Pettersen, E. F. et al. UCSF Chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612 (2004).
Article CAS PubMed Google Scholar
Sung, V. M. “Mechanistic overview of ADP-ribosylation reactions.”. Biochimie 113, 35–46 (2015).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

This work was supported by the National Health and Medical Research Council (NHMRC, Australia, Project Grant 1049564), a Schaefer Research Scholars Program Award (Columbia University, New York, USA) and the Biotechnology and Biological Sciences Research Council (BB/T013966/1) to J.R.L. and National Institute of Health (NIH) grant MH54137 to J.A.J.

Author information

These authors contributed equally: Alastair C. Keen, Maria Hauge Pedersen.
These authors jointly supervised this work: Jonathan A. Javitch, J. Robert Lane.

Authors and Affiliations

Drug Discovery Biology, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, VIC, 3052, Australia
Alastair C. Keen
Division of Physiology, Pharmacology and Neuroscience, School of Life Sciences, Queen’s Medical Centre, University of Nottingham, Nottingham, UK
Alastair C. Keen, Laura Lemel, Meritxell Canals & J. Robert Lane
Centre of Membrane Proteins and Receptors, University of Birmingham and University of Nottingham, Nottingham, UK
Alastair C. Keen, Laura Lemel, Meritxell Canals & J. Robert Lane
Departments of Psychiatry and Molecular Pharmacology and Therapeutics, Vagelos College of Physicians and Surgeons, Columbia University, New York, NY, USA
Maria Hauge Pedersen & Jonathan A. Javitch
Division of Molecular Therapeutics, New York State Psychiatric Institute, New York, NY, USA
Maria Hauge Pedersen & Jonathan A. Javitch
NNF Center for Basic Metabolic Research, Section for Metabolic Receptology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
Maria Hauge Pedersen
Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, VIC, 3052, Australia
Daniel J. Scott
The Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, 3052, VIC, 3052, Australia
Daniel J. Scott
Infection and Immunity Program and Department of Biochemistry and Molecular Biology, Biomedicine Discovery Institute, Monash University, Clayton, VIC, 3052, Australia
Dene R. Littler
Department of Animal, Plant and Soil Science and Centre for AgriBioscience, La Trobe University, Bundoora, VIC, 3086, Australia
Travis Beddoe
Graduate School of Pharmaceutical Sciences, Tohoku University, Sendai, Miyagi, 980-8578, Japan
Yuki Ono & Asuka Inoue
Computational Chemistry and Molecular Biophysics Section, National Institute on Drug Abuse - Intramural Research Program, National Institutes of Health, Baltimore, MD, USA
Lei Shi

Authors

Alastair C. Keen
View author publications
You can also search for this author in PubMed Google Scholar
Maria Hauge Pedersen
View author publications
You can also search for this author in PubMed Google Scholar
Laura Lemel
View author publications
You can also search for this author in PubMed Google Scholar
Daniel J. Scott
View author publications
You can also search for this author in PubMed Google Scholar
Meritxell Canals
View author publications
You can also search for this author in PubMed Google Scholar
Dene R. Littler
View author publications
You can also search for this author in PubMed Google Scholar
Travis Beddoe
View author publications
You can also search for this author in PubMed Google Scholar
Yuki Ono
View author publications
You can also search for this author in PubMed Google Scholar
Lei Shi
View author publications
You can also search for this author in PubMed Google Scholar
Asuka Inoue
View author publications
You can also search for this author in PubMed Google Scholar
Jonathan A. Javitch
View author publications
You can also search for this author in PubMed Google Scholar
J. Robert Lane
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

A.C.K., M.H.P., J.A.J. and J.R.L. designed experiments. A.C.K., M.H.P. and L.L. performed experiments. A.C.K., M.H.P., L.L., J.A.J. and J.R.L. analysed the data. A.C.K., M.H.P., L.L., L.S., J.A.J., M.C. and J.R.L. wrote the manuscript. D.R.L., Y.O., T.B., A.I. and D.J.S. provided reagents.

Corresponding authors

Correspondence to Jonathan A. Javitch or J. Robert Lane.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Communications Biology thanks Ka Young Chung and the other, anonymous, reviewers for their contribution to the peer review of this work. Primary Handling Editors: Nicolas Fanget and Karli Montague-Cardoso. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Peer Review File

Supplementary Information

Description of Additional Supplementary Files

Supplementary Data 1

Reporting Summary

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Keen, A.C., Pedersen, M.H., Lemel, L. et al. OZITX, a pertussis toxin-like protein for occluding inhibitory G protein signalling including Gα_z. Commun Biol 5, 256 (2022). https://doi.org/10.1038/s42003-022-03191-5

Download citation

Received: 05 January 2021
Accepted: 17 February 2022
Published: 23 March 2022
DOI: https://doi.org/10.1038/s42003-022-03191-5

This article is cited by

Generation of Gαi knock-out HEK293 cells illuminates Gαi-coupling diversity of GPCRs
- Yuki Ono
- Kouki Kawakami
- Asuka Inoue
Communications Biology (2023)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.