Abstract
The distinct subjective effects that define psychedelics such as LSD, psilocybin or DOI as drug class are causally linked to activation of the serotonin 2A receptor (5-HT2AR). However, some aspects of 5-HT2AR pharmacology remain elusive, such as what molecular drivers differentiate psychedelic from non-psychedelic 5-HT2AR agonists. We developed an ex vivo platform to obtain snapshots of drug-mediated 5-HT2AR engagement of the canonical Gq/11 pathway in native tissue. This non-radioactive methodology captures the pharmacokinetic and pharmacodynamic events leading up to changes in inositol monophosphate (IP1) in the mouse brain. The specificity of this method was assessed by comparing IP1 levels in homogenates from the frontal cortex in DOI-treated wild-type and 5-HT2AR-KO animals compared to other brain regions, namely striatum and cerebellum. Furthermore, we encountered that head-twitch response (HTR) counts and IP1 in the frontal cortex were correlated. We observed that IP1 levels in frontal cortex homogenates from mice treated with LSD and lisuride vary in magnitude, consistent with LSD’s 5-HT2AR agonism and psychedelic nature, and lisuride’s lack thereof. MDMA evoked an increase of IP1 signal in the frontal cortex that were not matched by the serotonin precursor 5-HTP or the serotonin reuptake inhibitor fluoxetine. We attribute differences in the readout primarily to the indirect stimulation of 5-HT2AR by MDMA via serotonin release from its presynaptic terminals. This methodology enables capturing a snapshot of IP1 turnover in the mouse brain that can provide mechanistic insights in the study of psychedelics and other serotonergic agents pharmacodynamics.
Introduction
The last decade has witnessed a burgeoning renewed interest in psychedelic drugs. Neglected for decades outside the underground and counter-cultural movements, drugs such as lysergic acid diethylamide (LSD) N,N-dimethyltryptamine (DMT), psilocybin, mescaline, and other synthetic derivatives reappear as a research priority as the result of preliminary evidence suggesting that they may serve as treatments in mental health diagnoses for which therapeutic options are limited1. Albeit subjective in nature, the characteristic altered mental state elicited by psychedelics is attributable to discrete molecular interactions with specific neuronal receptors2. Mounting evidence from studies in humans, animal models and in vitro assays demonstrate the central role of serotonin (or 5-hydroxytryptamine) 2A (5-HT2AR) receptor activation in the action of classical psychedelics3–6.
5-HT2AR is a G protein-coupled receptor (GPCR) that canonically engages upon activation Gq/11 proteins primarily6. Downstream, recruitment of these mainly excitatory heterotrimeric G proteins results in the production of inositol triphosphate (IP3), with subsequent release of calcium from intracellular compartments. This mechanism is consistent with the hyperexcitable state that neurons enter when exposed to psychedelics such as LSD, and with the apparent increase in cortical metabolic activity associated with psilocybin administration in healthy human volunteers7–9. However, 5-HT2AR activation is not exclusively attained by psychedelics. Lisuride, 2-Br-LSD, Ariadne and 6-F-DET are a small but representative sample of compounds known to activate 2A but lack in the subjective effects of psychedelics10–13
There is an ongoing need for functional benchmarks that can link psychedelics and their manifestations to their underlying molecular actions. Connecting these domains can aid to identify potential therapeutics pathways and provide much clarity on the mechanistic drivers that differentiate psychedelic from non- or lesser psychedelic 5-HT2AR agonists14,15. With this goal in mind, we adapted a commercial a methodology based on homogenous-time resolved fluorescence (HTRF) for the determination of inositol phosphate (IP1)—a downstream metabolite of Gq and related G protein-dependent signaling pathways—from mouse brain dissects. The mouse brain samples are subject to minimal post-extraction manipulation involving only homogenization. In doing so, we managed to obtain a molecular readout from the mouse brain that, like in vitro systems, is representative of 5-HT2AR activation by psychedelics. But unlike in in vitro methods is also subject to the pharmacokinetics that determine drug action in vivo.
Results
IP1 detection in mouse brain samples
IP1 is a near-terminal metabolite product resulting from the sequential hydrolysis of IP3; a key second messenger molecule in the signal transduction of receptors like 5-HT2AR that couple to Gq/10/11/14/15 (Gq for brevity)16. HTRF detection of IP1 can readily be employed as a reporter of Gq activity in vitro in cultured cells using commercially available kits. Compelled by previous attempts to study IP1 production by muscarinic receptors in vivo17, we aimed to adapt detection of IP1 via HTRF to detect changes in levels of this metabolite in mouse brain homogenates from small samples from 5-HT2AR-rich brain regions(Fig 1).
HTRF detection of IP1 relies on immunodetection coupled with a fluorescence resonance energy transfer (FRET) system18. Two main components integrate this system: a donor Tb-cryptate tagged anti-IP1 antibody and an acceptor-tagged IP1 molecule. FRET signal is maximal when both components are in aqueous media forming an immunocomplex. When non-labeled IP1 is present (i.e., from a biological sample), it displaces the acceptor-tagged IP1 from the system, and the FRET effect is suppressed (Fig 1). By virtue of the long-lived fluorescence emission of a Tb cryptate donor in the system, the assay readout is time-resolved. This millisecond-scale delay in the readout is sufficient to outlast the almost-immediate decay in the fluorescence signal of endogenous biological fluorophores; a common source of interference that precludes the use of fluorescence-based techniques to native tissue preparations18,19. The experimental assay window (signal/baseline in standard curve) was >10 (Supp Fig 1.).
Brain region-specificity and 5-HT2AR involvement
DOI, or 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane, is a potent psychedelic drug and valuable tool in the study of psychedelics’ pharmacology20. As a proof-of-concept study in the involvement of 5-HT2AR in the action of psychedelics, wild-type (WT) and 5-HT2AR knockout (5-HT2AR-KO) animals were treated with 2 mg/kg of DOI; a dose known to be active at the molecular and behavioral level, and present in the mouse brain at the time (1h) of sample harvesting 21,22.
In the WT cohort, frontal cortex homogenates from DOI-treated animals showed an increased IP1 signal as exemplified by the quotient of emission at 615 nm and 655 nm. To further demonstrate specificity of the IP1 signal, we compared frontal cortex homogenates between WT and 5-HT2AR-KO animals, both treated with DOI (Fig 2A). The robust increase in IP1 signal in WT frontal cortex homogenates was not replicated in 5-HT2AR-KO; which suggests that the main driver of the IP1 signal detected in tissue samples from the mouse frontal cortex homogenates is a consequence of 5-HT2AR stimulation (two-way ANOVA, genotype: F[1,8]=57.50, P<0.001; treatment: F[1,8]=82.20, P<0.001. Bonferroni’s post hoc (veh vs DOI): WT, P<0.001; 5-HT2AR-KO, P>0.05). A similar demonstration of 5-HT2AR involvement was previously shown by blockage of IP1 signal by the putative psychedelic drug quipazine with pre-treatment with the 5-HT2AR antagonist M10090723.
These results are also consistent with the dense expression of 5-HT2AR-KO in the mouse anterior cortex24, and further supported by the absence of changes in the IP1 signal in the cerebella of the same animals, where 5-HT2AR expression is merely absent (Fig 2B) (two-way ANOVA, region: F[1,8]=1421, P<0.001; treatment: F[1,8]=116.0, P<0.001. Bonferroni’s post hoc (veh vs. DOI): F. cortex, P<0.001; Cerebellum, P>0.05). It is also worth noting that baseline levels of IP1 are greater in the frontal cortex than in the cerebellum.
We also assessed receptor specificity per brain region by analyzing the striatum. DOI produced a minimal increase in IP1 signal in both WT and 5-HT2AR-KO animals of comparable magnitude, albeit only significant in 5-HT2AR-KOs (Fig 2C) (two-way ANOVA, genotype: F[1,8]=54.24, P<0.001; treatment: F[1,8]=9.35, P<0.05. Bonferroni’s post hoc (veh vs DOI): WT, P=0>0.05; 5-HT2AR-KO, P<0.05). Interestingly, baseline levels of IP1 appeared to be lower in the striatum of 5-HT2AR-KO mice compared to WT controls. Responsiveness to DOI and baseline levels greatly differed between the frontal cortices and striata of WT mice (Fig 2D) (two-way ANOVA, region: F[1,8]=2785, P<0.001; treatment: F[1,8]=374.5, P<0.001. Bonferroni’s post hoc (veh vs DOI): F. cortex, P<0.001; Striatum, P>0.05). In contrast to the cortex, 5-HT2AR expression is much less abundant in the striatum24, a region known for the abundance of 5-HT2CRs25, another Gq-coupled serotonin receptor activated by psychedelics, including DOI.
Time-course of IP1 in the mouse brain
As a downstream readout in vivo, we expected that the IP1 signal would be influenced by the pharmacokinetic constraints that modulate the action of the stimulating drug21. To gain further insights on the temporal dynamics of DOI on this readout, we evaluated the evolution of IP1 signal over time in mice administered with 2 mg/kg of DOI (Fig 3). IP1 signal increased sharply during the first 1h post administration to then progressively decrease and return to baseline levels at >6h). Compared to the characteristic biphasic pharmacokinetic curve21, the time-course of IP1 in frontal cortex homogenates appeared to show truncated peak of effect (i.e., flattened maximum). This could indicate saturation of the biological system having reached the maximal IP1 production possible, rather than saturation of the signal (see Supp Fig 1. as an example of assay window).
Dose-response evaluation and behavioral correlates
DOI robustly increases HTR in mice and is a potent 5-HT2AR agonist in heterologous expression models in vitro21. To further explore the molecular and behavioral link of DOI 5-HT2AR action, we sought to determine HTR correlativeness to IP1 signal induced by different doses of DOI within the same subjects (Fig 4A).
As expected, increasing doses of DOI resulted in higher HTR counts; as determined by their electromagnetic signature. HTR plateau was reached at ∼1 mg/kg after 60 min (Fig 4B) (Supp Fig 2) (one-way ANOVA, F[5,6]=23.10, P<0.001. Bonferroni’s post hoc (veh vs DOI): 0.2 mg/kg, P>0.05; 0.5 mg/kg, P>0.05; 1 mg/kg P<0.001; 2 mg/kg, P<0.001; 5 mg/kg, P<0.01).
Immediately after recording HTR for 60 min after administration of DOI, the animals were sacrificed and the frontal cortices harvested. IP1 signal at 60 min followed an asymptotic dose-response pattern (Fig 4C) (one-way ANOVA, F[5,6]=142.3. Bonferroni’s post hoc (veh vs DOI): 0.2 mg/kg, P=0.08; 0.5 mg/kg, P<0.001; 1 mg/kg P<0.001; 2 mg/kg, P<0.001; 5 mg/kg, P<0.001). Furthermore, IP1 signal readout was highly correlated with the 60 min aggregated HTR counts (Fig 4D)(F[1,10]=23.91, P<0.001).
Evaluation of other serotonergic drugs
LSD is among the most paradigmatic examples of classical psychedelics. Its core ergoline structure is shared by lisuride. Lisuride, despite being a non-psychedelic analogue, shares with LSD its ability to potently bind to and activate the 5-HT2AR in vitro6,26. We then sought to evaluate this apparent paradox in the mechanism of action of psychedelics by determining in parallel IP1 signal in the mouse frontal cortex for both drugs at different doses (Fig 5A) (two-way ANOVA, dose: F[3,8]=103.7, P<0.001, treatment: F[3,8]=442.9. Bonferroni’s post hoc (veh vs LSD): 0.2 mg/kg, P<0.001; 0.4 mg/kg, P<0.001; 1 mg/kg, P<0.001. Bonferroni’s post hoc (veh vs lisuride): 0.2 mg/kg, P<0.05; 0.4 mg/kg, P=0.0397; 1 mg/kg, P<0.001). Fitting IP1 signal to a non-linear regression model showed a ∼6-fold differences in the saturation magnitude between both drugs (LSD, ED50=0.1191 mg/kg, span=58.28 ratiometric arbitrary units (A.U.), R-square=0.9837; lisuride, ED50=0.3401 mg/kg, span=20.82 A.U., R-square=0.8475). The meager production of IP1 signal by lisuride contrast with the robust increase shown by LSD.
To further explore the versatility of the platform, we sought to evaluate the IP1 signal for different serotonergic drugs neighboring the classical psychedelics chemical space (Fig 5B): MDMA, fluoxetine and 5-hydroxytryptophan (5-HTP). MDMA is a serotonin releaser that reverses serotonin transport thus leading to increases in the neurotransmitter in the synaptic space27. The traditional antidepressant fluoxetine directly blocks the serotonin transport back to the synapse28. 5-hydroxytryptophan (5-HTP) is metabolic intermediate in the biosynthesis of serotonin. Despite its lack of psychedelic effects in human at doses commonly used as a dietary supplement, 5-HTP induces HTR in mice at high doses, thus constituting one paradigmatic example of false positives in this predictor model of psychedelic effects in human29,30.
Out of the different drugs tested, only MDMA produced a substantial increase in IP1 signal (Fig 5B) (one-way ANOVA, F[3.8]=36.08, P<0.001. Bonferroni’s post hoc (vehicle vs drug): MDMA, P<0.001, fluoxetine, P>0.05; 5-HTP, P>0.05). The matched dose of fluoxetine, a dose known to produce behavioral effects in mice31, did not produce any apparent changes in the levels of IP1. Surprisingly, neither did 5-HTP at a dose known to induce HTR increases and intense diarrhea in mice29.
To ensure proper coverage of 5-HTP pharmacokinetics, in a separate experiment, 5-HTP was administered to WT mice at a higher dose (200 mg/kg) and samples collected at an earlier time-point (30 min) following recording of HTR (Fig 6A). 5-HTP produced a sharp increase of HTR events compared to vehicle, as expected (Fig 6B) (Supp. Fig 3)(P<0.01, t=4.656, df=4). However, the HTR induction profile was not paralleled by an increase in IP1 signal in the frontal cortex homogenates (Fig 6C)(one-way ANOVA, F[2,5]=87.21, P<0.001; Bonferroni’s post hoc (vehicle vs DOI), P<0.001).
Discussion
Activation of 5-HT2AR by classical serotonergic psychedelic drugs is the necessary pharmacological driver for their distinct effects in human psyche5,32. Canonically, 5-HT2AR activation results in Gq protein isoforms coupling to the receptor triggering a cascade of downstream events that culminates with the hydrolysis of IP1 6,33. Herein, we present a methodology that enables capturing a snapshot of IP1 levels in mouse brain homogenates as a surrogate readout of 5-HT2AR activation in mouse cortex. As a proof of concept, we demonstrate that increases in IP1 can readily be detected in frontal cortex brain homogenates following administration of psychedelics LSD and DOI to the alive animal. We also demonstrate the involvement of 5-HT2AR in this readout for DOI by genetic deletion of the receptor and by differential expression of the receptor in different neuroanatomical regions. DOI produced a distinct dose-dependent increase of IP1 in the 5-HT2AR-dense frontal mouse cortex in comparison with the cerebellum, a region known to lack in expression of 5-HT2AR25. The receptor specificity shown is in alignment with our previous study showing that IP1 signal in frontal cortex of mice treated with the 5-HT2AR agonist quipazine was blocked by pre-treatment with the selective antagonist M10090723.
Our results demonstrate that IP1 measured by HTRF in cortical homogenates can afford sufficient receptor and anatomical specificity to be utilized as a potential biomarker of 5-HT2AR activation in the frontal cortex. This prospective use is not without limitations. Psychedelics target several other GPCRs expressed in the central nervous system that upon activation couple Gq proteins. These considerations are important when considering potential experimental designs. We observed a small but statistically significant effect of DOI on IP1 signal in the striatum of 5-HT2AR-KO that could be attributed to activation of the 5-HT2CR. As it pertains to the use of frontal cortex as a region of interest to characterize the pharmacology of 5-HT2AR agonists in general, and psychedelics in particular, 5-HT2CR did not appear to contribute to the frontal cortex turnover of IP1 as experimentally shown with DOI in WT and 5-HT2AR-KO mice. This is likely possible due to the relative abundance of 5-HT2AR in the cortex relative to 5-HT2CR expression25,34.
We encountered that HTR and IP1 in frontal cortex homogenates from the same mice treated with DOI were highly correlated at 1h post administration. We appreciated divergences in the kinetics, however. Consistent with the inverted U-shape of HTR dose responses35, HTR counts appeared to drop at higher doses of DOI. Conversely, the IP1 signal in the frontal cortex of the animals followed an asymptotic fate akin to a saturation curve. It is plausible that behavioral disruption limits the expression of HTR at doses that might continue to stimulate 5-HT2AR before reaching saturation of the biological system. In the time domain, IP1 turnover increase was delayed relative to manifestation of HTR and our previously reported DOI brain pharmacokinetics21. IP1 is a near-terminal metabolite of the Gq pathway so it is possible that its formation can be subject to downstream catabolic bottlenecks. As we recently reported, exposure to DOI leads to downregulation of 5-HT2ARs in the frontal cortex and development of HTR tolerance36. IP1 formation might thus continue intracellularly after clearance of the drug-bound receptor from the cell surface.
Furthering in the study of psychedelic effect correlative to 5-HT2AR activation, we also tested LSD and lisuride in our IP1 model. Both compounds are brain penetrant ergolines and 5-HT2AR agonists, yet LSD is one of the most paradigmatic representatives of the psychedelic drug family, whereas lisuride is not psychedelic in human or animal models6,37. Increasing doses of LSD produced a saturable IP1 signal in the 5-HT2AR-rich mouse frontal cortex that was not matched in magnitude by lisuride. The lesser ability of lisuride to activate 5-HT2AR receptors in vivo might ultimately answer its lack of LSD-like effects as well as its apparent 5-HT2AR antagonist profile in numerous paradigms such as spontaneous HTR6,36.
We also screened several other serotonergic compounds and measured their effect on IP1 in mouse frontal cortex homogenates following systemic administration. Only MDMA at 10 mg/kg produced a statistically significant increase in IP1 compared to vehicle treated animals. Both MDMA and fluoxetine target the serotonin transporter primarily but differ in their primary effect on serotonin’s homeostasis. Fluoxetine blocks its uptake38 whereas MDMA produces serotonin efflux from its presynaptic storage39. It is plausible that the latter mechanism could contribute to greater levels of 5HT capable of attaining meaningful stimulation of cell signaling via 5-HT2AR in the mouse frontal cortex. Unlike fluoxetine, MDMA shares certain aspects of its phenomenological experience with classic psychedelics40 some of which can be blocked by the 5-HT2R antagonist ketanserin41.
HTR is one of the best classifiers of psychedelic effect in human for serotonergic drugs, but it is not exempt from the interference from false positives35. One of such cases is 5-HTP29,30. Devoid of psychedelic effect in human at doses used as a ‘nutritional supplement’, 5-HTP induces in mice HTR sensitive to the action of 5-HT2AR antagonists30. However, we showed that a HTR-inducing dose of 5-HTP did not stimulate 5-HT2AR-dependent IP1 signaling in the mouse frontal cortex. Our findings suggests that 5-HTP-induced HTR could be extra-cortical in nature. This also demonstrates how IP1 detection in mouse brain samples can complement the predictive power of HTR in the determination of psychedelic potential in human.
The study of 5-HT2AR pharmacodynamics at the receptor level generally starts with the use of in vitro approaches to avoid confounders in the readout. Singling out the receptor in heterologous expression in these models detach functional readouts from its physiological context. For identical compounds, historical divergences in reported maximal efficacies are common even for readouts that belong in the same Gq-dependent signaling pathway and heterologous expression system. One fundamental advantage of our model is that it provides a representative snapshot of IP1 levels in the brain. The readout integrates a wide array of intervening neurobiological and pharmacokinetic processes that occur in the entire animal that are not accounted for in cell systems. While animal intensive, the method does not require any additional experimental interventions besides drug administration during the in-life period and sample harvesting, thus avoiding introducing secondary sources of variability that can operate as external confounders. Other ex vivo techniques employed in the quantitation of inositol phosphates in the rodent brain involve incubation with the test compound in an organ bath42. In our case, sample processing is limited to homogenization following brain extraction and dissection to ensure that IP1 levels are representative of the drug effect during the in-life period. It also does not require the use of radioactive reagents and affords a readout of IP1 levels in less than 1h from sample collection.
The methodology is not exempt from some limitations, that while addressable, are worth considering. For instance, IP1 quantification alone provides no hints on the stoichiometry or receptor and Gq protein isoforms responsible for the readout. In vitro approaches using heterologous expression systems are better suited to study such a degree of molecular precision 43. Another potential caveat in our approach is that 5-HT2AR is not the only receptor capable of coupling Gq proteins in the frontal cortex with the consequential increase in IP metabolites. This calls for a careful evaluation of the drug-selectivity profile, neuroanatomical expression of target and off-target receptors and ultimately the test drug pharmacokinetics17. However, as exemplified here with DOI, these confounders can be controlled through experimental design in a way that can also produce valuable mechanistic insights. Quantitation of IP1 in the mouse brain, as shown here, can complement in vitro approaches and expand our understanding of 5-HT2AR pharmacodynamics for psychedelics and related compounds.
Conclusions
As psychedelic research gains traction, the need for complementary models that can afford a better understanding of their pharmacology becomes increasingly apparent. In response, we developed a platform capable of capturing a snapshot of drug mediated. 5-HT2AR activation of the Gq pathway in the mouse cortex via detection of IP1 in tissue homogenates. This methodology aims to bridge the gap between 5-HT2AR in vitro and in vivo pharmacodynamics by using a model that employs a molecular readout in the mouse brain that subjects the drug to the same pharmacokinetic constraints that drive behavioral pharmacology. We demonstrated a high degree of 5-HT2AR specificity in DOI stimulation of IP1 in the mouse frontal cortex that enables the exploration of this readout as a reference for the quintessential molecular interaction driving the subjective effects of psychedelics. It also serves as a comparator for other serotonergic drugs with related mechanisms of action. The platform can aid the exploration of relationships between drug-mediated 5-HT2AR activation, pharmacokinetics and dynamics of behavioral manifestations in the same animals.
Methods
Animals
Wild-type C57BL/6J male mice and 5-HT2AR-KO 6 male mice backcrossed from 129S6/SvEv onto C57BL/6J for several generations (F5) were randomly allocated into the different treatment groups (12-16 weeks old). Animals were housed on a 12 h light/dark cycle at 23°C with food and water ad libitum. All procedures were conducted in accordance with NIH guidelines and were approved by the Virginia Commonwealth University Animal Care and Use Committee. All efforts were made to minimize animal suffering, and the number of animals used.
Drugs
Drugs were administered dissolved in 0.9% saline as vehicle intraperitoneally (5 ul/g, i.p.) and sourced from authorized vendors: (±)-1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane ((±-DOI) hydrochloride (MilliporeSigma); LSD (Lipomed); lisuride maleate (Tocris); 5-HTP (MilliporeSigma); fluoxetine hydrochloride (Tocris); (±)-MDMA hydrochloride (Lipomed). An equimolecular amount of HCl was added to form the hydrochloride salt in situ for LSD and 5-HTP. In the case of LSD and lisuride, the vehicle contained an equivalent volume of DMSO (<1% final volume).
Quantification of head-twitch responses (HTR)
HTR assessment was performed in mice with magnetic ear tags designed for automated HTR detection21,44. Briefly, neodymium magnets bearing-ear tags (∼50 mg) were placed bilaterally through the pinna antihelix under isoflurane (2%)44. All animals were allowed to recover for a week prior to testing21. During the testing session, animals were transferred from their home cage to the testing chamber where they were allowed to habituate to the environment. After 30 min, DOI or 5-HTP or vehicle was administered i.p.. Data acquisition in the magnetometer was performed for 60 min for DOI and 30 min for 5-HTP as previously described21. After completion of the recording, the animals were sacrificed by cervical dislocation, and the brain samples harvested. Data was processed offline using a previously described signal analysis protocol44. To refine HTR detection, the signal was also processed using a deep learning-based protocol based on scalograms45. Mismatches between both detection methods were inspected visually without clues relative to the timestamp of the event or treatment group as previously described21,44.
IP1 detection in brain sample homogenates
Sample collection
Animals treated with test drug or vehicle were sacrificed at the time specified on each experimental design by cervical dislocation. All animals in each experiment and time point were sacrificed simultaneously (∼3 animals/min), decapitated, and the heads cooled down on wet ice. The brain extraction and dissection of the region/s of interest (Fig 1.) was performed sequentially. During this process, only saline 0.9% was employed for washing of specimens as the use of PBS can influence the HTRF readout. For consistency, during processing of a batch, samples and heads to be processed remained on ice until the dissection process was completed, and then samples were frozen at - 80°C for storage. The carryover of saline, blood or any other debris was minimized during this process. Quality and homogeneity in the dissection process proved crucial to obtain accurate readouts.
Sample processing
Each individual sample was transferred frozen to a 1.5 ml conical tube with a safelock cap (#3456 ThermoFisher) in a semi-micro balance (MS105DU, Mettler Toledo) to weight the sample with precision. A volume of 0.5 mm diameter glass beads (Biospec products #11079105) roughly equivalent to the sample volume and an exact amount of chilled homogenization buffer (10 μl/mg of sample) was added to the same tube. Homogenization buffer was prepared as a mix of 10% ‘Lysis & Detection Buffer’ and 90% ‘Stimulation Buffer’ 1× from the IP-One Gq kit (Perkin Elmer). While the exact formula of this buffers is proprietary, they contain LiCl to prevent IP1 enzymatic hydrolysis as outlined in the user manual. The precision in the volume-to-weight dilution proved to be crucial for consistency and narrowing of standard error of the mean. The tubes caps were secured and the samples homogenized (NextAdvance Bullet Blender 24) at 4C for 5 min, speed 6, after which they were centrifuged for 15 min microcentrifuge (4C, 17,000 g). The clarified homogenate supernatant was used directly or frozen at −80C until further use.
Platting and reading
The detection reagents: donor Tb cryptate antibody (K) and d2-labelled acceptor (d2) were reconstituted according to the IP-One Gq kit manufacturer instructions in distilled water (6X) and then diluted in ‘Lysis & Detection Buffer’ (1X). Aliquots were kept frozen and 20C. Platting was performed in white opaque HTRF 96 well low volume plates (66PL96005). Each well contained 18 μl of a mastermix composed by 12 μl of ‘Stimulation Buffer’ 1×, 3 μl of K reagent, 3 μl of d2 reagent. To reach the 20 μl final volume per plate, 2 μl of clarified homogenate supernatant (sample) or an aliquot of known concentration of IP1 (standard) were added. Mastermix was prepared fresh before each experiment and standard curves generated (See Supp. Fig. 1) Each experiment had, at least, a 0 μM and a 22 μM IP1 standard as internal control. The plate was covered and incubated for 30 min in darkness at room temperature and read in a VICTOR Nivo plate reader (Perkin Elmer). The incubation is not involved in the production of IP1, addition of DOI 10 μM prior to incubation to control wells did not change IP1 levels (results not shown). The settings employed for the reading were: excitation filter 320/8 nm, emission filter #1 615/8 nm, emission filter #2 665/8 nm, dichroic mirror D400, delay time 70 μs, emission time 200 μs, flash energy low, measurement time 500 ms, two reads per well, Z-focus 12 mm. The ratiometric signal was calculated for each well as the reading from Emission filter #1 divided by the reading from Emission filter #2 and multiplied by 100.
Statistical analysis
Statistical significance involving three or more treatments or different doses of the same treatment was assessed by one-way ANOVA followed by Bonferroni’s post hoc test. Statistical significance of experiments involving different treatments and doses was assessed by two-way ANOVA followed by Bonferroni’s post hoc test. The level of significance was set at p = 0.05. All values in the figure legends represent mean ± s.e.m. Statistical analysis was performed with GraphPad Prism software version 9 (La Jolla, CA).
Author contributions
M.F.R. conceived the methodology, performed the experiments, analyzed the data and wrote the manuscript along with J.G.-M. who supervised the research and obtained funding.
Conflict of interests
M.F.R. is the owner of GONOGO solutions LLC.
Supp Fig 1. Standard curve corresponding to different concentrations of IP1 from a standard used as internal controls from three different experiments. Shown are the corresponding non-linear fit curves and parameters.
Supp Fig 2. Time-course of vehicle and DOI-induced HTR at different doses. Arrow shows the point at which the drug is administered.
Supp Fig 3. Sum of HTR events for 30 min post-administration of 5-HTP (200 mg/kg) or vehicle. **P<0.01.
Acknowledgements
This work was supported by the National Institute of Health (NIH) grants T32MH020030 (M.F.R.) and R01MH084894 (J.G.-M.).