Professional Documents
Culture Documents
Vargas Et Al. 2023 - Psychedelics Promote Neuroplasticity Through The Activation of Intracellular 5-HT2A Receptors - Science
Vargas Et Al. 2023 - Psychedelics Promote Neuroplasticity Through The Activation of Intracellular 5-HT2A Receptors - Science
Vargas Et Al. 2023 - Psychedelics Promote Neuroplasticity Through The Activation of Intracellular 5-HT2A Receptors - Science
D
plasticity in this assay (9). These structure-
ysregulation of the cortex has been hy- 5-HT2ARs (9, 10). The mechanism by which activity relationship (SAR) studies revealed
pothesized to play an important role in 5-HT2AR activation leads to changes in neu- that increasing N-methylation led to an en-
the pathophysiology of mental illnesses ronal growth is still poorly defined, although it hanced ability to promote neuronal growth,
such as depression and often manifests appears to involve TrkB, mechanistic target of with the N,N-dimethyl compounds increasing
as structural changes, including decreased rapamycin (mTOR), and AMPA receptor signal- dendritic arbor complexity to the greatest ex-
dendritic arbor complexity and reduced den- ing (11). It is currently unclear why some 5-HT2AR tent (Fig. 1, B to D).
dritic spine density (1–3). Traditional antide- ligands can promote neuroplasticity and pro- Increasing N-methylation is known to affect
pressants, such as selective serotonin reuptake duce sustained therapeutic behavioral responses the efficacy of 5-HT2AR signaling, so we at-
inhibitors (SSRIs), can rescue these deficits in the absence of hallucinogenic effects, where- tempted to correlate psychoplastogenic effi-
after chronic treatment, although it seems that as other 5-HT2AR agonists do not promote cacy across a range of 5-HT2AR ligands with
their effects may be independent of serotonin plasticity at all (12–15). Indeed, serotonin itself efficacy in a traditional [3H]–inositol phosphates
and perhaps involve the activation of tropomyosin does not produce psychedelic-like effects on (IP) accumulation assay (Fig. 1E) (22). Notably,
receptor kinase B (TrkB) signaling (4, 5). A class of neuronal growth when administered to corti- we did not observe a positive correlation be-
therapeutic compounds known as psychoplasto- cal cultures (9). This enigmatic finding can- tween psychoplastogenic effects and ligand
gens (6) are differentiated from SSRIs by their not be easily explained by traditional biased efficacy. Indeed, there seemed to be a non-
ability to produce both rapid and sustained ef- agonism because serotonin is a balanced agonist significant inverse correlation between [3H]-IP
fects on structural plasticity and behavior after a of the 5-HT2AR that exhibits high potency and accumulation and dendritogenesis efficacy
single administration (7). Psychoplastogens in- efficacy for activating both heteromeric guanine (Fig. 1E). To avoid potential issues associated
clude both ketamine and serotonergic psychedel- nucleotide–binding protein (G protein) and with the amplification of secondary messengers,
ics, although their primary targets are distinct (7). b-arrestin pathways (16, 17). we used psychLight2, a fluorescent biosensor
Psychedelics are 5-hydroxytryptamine (sero- Unlike psychedelics, the physicochemical that is capable of directly detecting changes in
tonin) 2A receptor (5-HT2AR) agonists that can properties of serotonin prevent it from enter- 5-HT2AR conformation (14). PsychLight2 effi-
lead to profound changes in perception, cogni- ing cells by passively diffusing across nonpolar cacy closely mirrored that observed by using
tion, and mood (8). Recent evidence suggests membranes (8). Thus, we reasoned that an- [3H]-IP accumulation assays, although psy-
that they promote cortical structural and func- other form of functional selectivity, known chLight2 efficacy exhibited an even stronger
tional neuroplasticity through activation of as location bias, might explain the difference anticorrelation with dendritogenesis efficacy
in cellular signaling elicited by serotonin and (P = 0.06) (Fig. 1F). Lastly, we used biolumines-
1
Neuroscience Graduate Program, University of California, psychedelics (18, 19). Here, we leveraged both cence resonance energy transfer (BRET) assays
Davis, Davis, CA 95618, USA. 2Institute for Psychedelics and chemical design and genetic manipulation to directly measure Gq activation or b-arrestin-2
Neurotherapeutics, University of California, Davis, Davis, CA to test the hypothesis that activation of an recruitment (fig. S2) (23). Both measures of
95618, USA. 3Department of Chemistry, University of
California, Davis, Davis, CA 95616, USA. 4Biochemistry,
intracellular population of 5-HT2ARs is nec- 5-HT2AR efficacy exhibited a strong negative
Molecular, Cellular, and Developmental Biology Graduate essary for 5-HT2AR ligands to induce cortical correlation with psychoplastogenicity (Fig. 1,
Program, University of California, Davis, Davis, CA 95616, structural plasticity and produce antidepressant- G and H). Moreover, both Gq activation and
USA. 5Center for Neuroscience, University of California,
Davis, Davis, CA 95618, USA. 6Department of Cell Biology,
like behavioral responses. b-arrestin recruitment correlated well with
Neurobiology, and Anatomy, Medical College of Wisconsin, psychLight efficacy [coefficient of determina-
Milwaukee, WI 53226, USA. 7Pharmacology and Toxicology Lipophilicity correlates with psychoplastogenicity tion (R2) = 0.9; P < 0.0001 and P = 0.0006, re-
Graduate Program, University of California, Davis, Davis,
To firmly establish the role of 5-HT2AR activ- spectively] (Fig. 1I). Negative correlation between
CA 95616, USA. 8Department of Neurology, School of
Medicine, University of California, Davis, Sacramento, CA ation in psychedelic-induced spinogenesis, we 5-HT2AR efficacy and psychoplastogenicity
95817, USA. 9Department of Biochemistry and Molecular administered5-methoxy-N,N-dimethyltryptamine should be interpreted with caution because
Medicine, School of Medicine, University of California, Davis, (5-MeO) to wild-type (WT) and 5-HT2AR knock- this relationship may only apply to tryptamine-
Sacramento, CA 95817, USA.
*Corresponding author. Email: deolson@ucdavis.edu out (KO) mice (20) and assessed structural and based ligands or compounds that exhibit a
†These authors contributed equally to this work. functional changes in layer 5 pyramidal neurons threshold level of 5-HT2AR activation.
Nitrogen methylation of 5-HT2AR ligands extent of overexpression was similar for 5- in the culturing media because serum contains
structurally related to serotonin is known to HT2ARs and b2ARs in both HEK293T cells serotonin, which can lead to agonist-induced
result in partial agonism (22), but it also has and cortical neurons (fig. S3A). We performed changes in trafficking and localization (fig. S3,
a profound effect on their physicochemical these localization experiments without serum B and C). In both neurons and HEK293T cells,
properties. These compounds display a wide
range of lipophilicities that ranged from A C Nmax
R NH2 R N H N Me
highly polar molecules such as serotonin to R
Me Me
relatively nonpolar compounds such as N,N- N N N
VEH
Number Of Crossings
Number Of Crossings
Number Of Crossings
serotonin, and 5-MeO–TRY scaffolds. The find- 6 6 6
5-HT *
ing that lipophilicity was a better predictor
of psychoplastogenicity than 5-HT2AR activa- 4 4 4
8
the plasma membrane, several exhibit substan- 20 μm Number of Crossings
tial intracellular localization (24–27). In vitro
E F I
5-HT
and ex vivo experiments have also established 100 100 100
DMT 5-MeO-NMT TRY 5-MeO-TRY
the existence of large intracellular pools of 5- 95
Gq Emax
DMT
N-Me-5-HT 5-MeO-DMT
80 BUF 80 BUF 90 BUF
HT2ARs in various cell types in the absence of 5-MeO-DMT
5-MeO-DMT
NMT p = <0.001
85
% Efficacy
% Efficacy
DMT
a ligand (28–31). To compare 5-HT2AR local- 60 5-HT
60 NMT 5-MeO-NMT
80
R2 = 0.9
NMT 5-HT
ization patterns between cell types, we expressed 5-MeO-NMT
-arrestin 2 Emax
40 40 N-Me-5-HT 120 5-MeO-TRY
5-MeO-TRY 5-MeO-TRY
a Myc–5-HT2AR–enhanced cyan fluorescent N-Me-5-HT TRY
TRY 100 5-MeO-DMT
5-HT
TRY
protein (ECFP) construct in both human em- 20 p = 0.2 20 p = 0.06 80 N-Me-5-HT 5-MeO-NMT
R2 = 0.3 R2 = 0.4 BUF p = 0.0006
bryonic kidney 293T (HEK293T) cells and cor- 60
0 0 DMT NMT R2 = 0.9
0 20 40 60 80 100 0 5 10 15 20 40
tical neurons, performed live-cell imaging, and
0 5 10 15 20
assessed colocalization with a membrane dye [3H]-IP accumulation Emax psychLight 2 F/F
psychLight 2 F/F
(Cellbrite Steady) that labels the extracellular G 100 H 100 J 100 DMT
side of the plasma membrane. Because over- DMT DMT
80 BUF 80 BUF 80 BUF 5-MeO-DMT
expression of tagged receptor constructs might 5-MeO-DMT 5-MeO-DMT
% Efficacy
% Efficacy
% Efficacy
Manders’ CC
Manders’ CC
0.6 0.6 0.6 0.6 (fig. S6B). In psychLight2 assays, the membrane-
0.4 0.4
impermeable analogs displayed comparable
0.4 0.4
efficacies to their uncharged parent molecules
0.2 0.2 0.2 0.2 with reduced potencies (fig. S6, C and D).
0.0 0.0 0.0 0.0 Next, we treated freshly dissected rat embry-
onic cortical neurons with DMT (1 mM) and PSI
R
R
R
AR
AR
FP
AR
FP
AR
FP
FP
2A
2A
2A
2A
C
T2
T2
T2
T2
C
H
H
5-
5-
5-
5-
A OH
O
D and 2.33, respectively), we reasoned that their
N Me N Me O
VEH abilities to displace [3H]–D-lysergic acid diethyl-
Me Me N **** amide (LSD) bound to the 5-HT2AR would de-
N F DMT
N DMT N PSI KTSN pend on whether those receptors were exposed
H H TMT
N O to the extracellular environment. Thus, we per-
O H O
OH PSI **** formed radioligand competition binding ex-
N Me P O
Me O O I PSY periments using intact HEK293T cells that
Me N Me N
I H N F expressed Myc–5-HT2AR or psychLight2 as
N Me Me
H well as membrane preparations obtained from
10
TMT N PSY N O MKTSN
H H spines / 10 μm
these systems. The inhibition constant (Ki)
B C E values for serotonin and 5-MeO–DMT were
VEH (–) VEH (+)
– Electroporation + Electroporation VEH nearly identical when using membrane prep-
arations or intact PSYLI2 cells, with HEK293T
VEH cells that stably expressed a 5-HT2AR con-
20 μm
DMT struct with an ER export sequence resulting in
DMT (–) DMT (+) DMT **** **** a large proportion of the 5-HT2ARs being ex-
posed to the extracellular environment (fig. S8).
TMT
However, 5-MeO–DMT was an order of magni-
+ DMT
6 display a large proportion of plasma membrane–
PSY (–) PSY (+) MKTSN bound 5-HT2ARs, serotonin and 5-MeO–DMT
VEH + DMT **** 3 had comparable potencies and efficacies (fig.
5-HT
S9A). When the same experiment was performed
0
VEH MKTSN in rat cortical neurons, serotonin failed to elicit
7
8
SERT – SERT –
Number of Crossings
CIT (10 mM), the plasticity-promoting effects of
D Blocking 5-HT Effects Blocking DMT Effects serotonin (1 mM) on SERT-positive neurons
5-HT KTSN CIT DMT KTSN CIT
– – – ns – – – ns were blocked (Fig. 4D). By contrast, CIT had no
effect on the ability of DMT (1 mM) to promote
+ – – + – – **** ns the growth of SERT-positive or SERT-negative
**** **** ****
cortical neurons (Fig. 4D). In the presence of
– + – ns – + – ns
KTSN (10 mM), neither serotonin nor DMT could
+ + – ns + + – ns
promote neuronal growth, which confirms that
DMT and intracellular serotonin promote plas-
– – + ns – – + ns ticity by means of 5-HT2AR receptors in vitro
(Fig. 4D).
+ – + ns + – + **** ns
To determine whether intracellular serotonin
****
could promote the growth of cortical neurons
4
8
4
Number of Crossings Number of Crossings in vivo, we injected the medial PFC (mPFC) of
SERT - SERT + SERT - SERT + Thy1–enhanced green fluorescent protein (EGFP)
mice with either CaMKII-SERT-mCherry or
E VEH DMT 5-HT F CaMKII-mCherry. The mPFC was chosen as the
VEH ns injection site because it exhibits high levels of
5-HT2AR expression (28) and has been impli-
DMT **** ns cated in the antidepressant-like effects of psycho-
**** plastogens (42). After 3 weeks to enable construct
expression (Fig. 5, A and B), both groups were
5-HT ****
**** administered (±)-para-chloroamphetamine
[PCA, 5 mg/kg intraperitoneally (ip)], a selec-
0
10
Discussion
Although GPCRs are traditionally viewed as
initiators of signal transduction that originates Fig. 5. Cellular uptake of serotonin produces antidepressant-like effects in vivo. (A) Schematic that
at the plasma membrane, increasing evidence displays experimental design for measuring spine density in Thy1-EGFP mice after administration of
suggests that GPCR signaling from intracellu- a serotonin-releasing agent. (B) Histology images of the mPFC of Thy1-EGFP mice that express CaMKII-
lar compartments can play important roles in mCherry or CaMKII-SERT-mCherry. (C) Confocal images of dendritic spines in the mPFC of mice treated with
cellular responses to drugs. Recently, location PCA (5 mg/kg ip). (D) Mice that express CaMKII-SERT-mCherry display increased dendritic spine density
bias has been proposed to explain signaling after PCA treatment. (E) Schematic that displays experimental design for determining the sustained
differences between endogenous membrane- antidepressant-like effects of serotonin in mice that express SERT in the mPFC. (F) NIL demonstrates no
impermeable peptide ligands and membrane- difference between CaMKII-SERT-EYFP– and CaMKII-GFP–expressing mice. (G) CaMKII-SERT-EYFP– and
permeable ligands of opioid receptors (38). CaMKII-GFP–expressing mice exhibit no differences in the FST. After PCA (5 mg/kg ip) administration,
Moreover, distinct ligand-induced signaling CaMKII-SERT-EYFP–expressing mice display a sustained antidepressant-like effect. ns is not significant, *p <
has been observed for plasma membrane– 0.05, and **p < 0.01, as compared between indicated pairs of data in (D) and (F) (two-tailed unpaired
localized and intracellular populations of Student’s t test) or (G) (two-way repeated measures ANOVA followed by Šídák’s multiple comparisons test).
d-opioid receptors (46). Here, we extend the Error bars in (D), (F), and (G) represent standard error of the mean.
concept of location bias to ligands of the
5-HT2AR.
A substantial proportion of 5-HT2ARs in neurons of the mPFC, a brain region that is determine whether intracellular 5-HT2AR sig-
cortical neurons are localized to the Golgi, and known to be essential for the HTR (49). Thus, naling is distinct from 5-HT2AR signaling at
intracellular compartments such as the Golgi activation of intracellular cortical 5-HT2ARs the plasma membrane.
are slightly acidic compared with the cytosol may play a role in the subjective effects of Although intracellular expression of 5-HT2ARs
and extracellular space. Thus, it is possible that psychedelics. This hypothesis is further sup- within cortical pyramidal neurons has been
protonation of psychedelics within the Golgi ported by previous work that demonstrates known for some time, it is unclear at present
leads to retention and sustained signaling, that a high dose of the serotonin precursor 5- why the subcellular localization of these re-
which results in neuronal growth, even after hydroxytryptophan (5-HTP) induces a HTR ceptors differs greatly between neurons and
transient stimulation (47). Persistent growth in WT mice, which can be blocked by an N- HEK293T cells (28, 29, 31). One possibility is
after the drugs have been removed from the methyltransferase inhibitor that prevents the that 5-HT2ARs form heterodimeric complexes
extracellular space is a hallmark of serotoner- metabolism of 5-HTP to N-methyltryptamines that affect cellular trafficking (31). Thus, by
gic psychoplastogens (47, 48). Although the (50). Inhibition of N-methyltransferase failed dictating 5-HT2AR localization, cellular con-
mechanistic details that link intracellular 5- to block the HTR induced by 5-MeO–DMT (50). text could influence responses to endogenous
HT2AR activation to cortical neuron growth Taken together, this work emphasizes that ac- neuromodulators and/or exogenous drugs,
have not been fully elucidated, they are likely cessing intracellular 5-HT2ARs is important for which potentially results in circuit-specific ef-
to involve AMPA receptor, TrkB, and mTOR 5-HT2AR agonists to produce a HTR. fects of 5-HT2AR ligands.
signaling, as previously established (9). Future Our results demonstrate that membrane per- Intracellular signaling has been hypothesized
studies should examine the detailed signaling meability is essential for a ligand to activate to contribute to the pharmacological proper-
interplay between these proteins. 5-HT2ARs in cortical neurons; however, our ties of a diverse range of compounds that in-
In addition to promoting psychedelic-induced experiments did not distinguish between in- cludes nicotine, ketamine, and SSRIs (51–53).
structural neuroplasticity, the intracellular tracellular signaling or the possibility of psy- Like psychedelics, these compounds are weak
population of 5-HT2ARs might also contribute chedelics acting as pharmacological chaperones. bases with pKa (where Ka is the acid dissoci-
to the hallucinogenic effects of psychedelics. Others have hypothesized that GPCR ligands ation constant) values ranging from 7 to 10.
When we administered a serotonin-releasing may act as pharmacological chaperones, which Given that the antidepressant mechanisms of
agent to WT mice, we did not observe a HTR. facilitates their export to the plasma membrane ketamine and SSRIs have not been definitively
However, the same drug was able to induce a where they could presumably engage in canon- established, it is intriguing to speculate that
HTR in mice that expressed SERT on cortical ical signaling (51). Thus, future studies should they also might promote cortical neuron growth
by binding to intracellular targets. Perhaps other 19. M. A. Mohammad Nexhady, J. C. Nezhady, S. Chemtob, reader. We also thank C. Nichols for advice about the radioligand
antidepressants affect the function of scaffold- iScience 23, 101643 (2020). binding experiments. Funding: This work was supported by
20. N. V. Weisstaub et al., Science 313, 536–540 (2006). funds from the National Institutes of Health (NIH) (R01GM128997
ing proteins within the cell interior to modu- 21. D. Ristanović, N. T. Milosević, V. Stulić, J. Neurosci. Methods to D.E.O., R35GM133421 to J.D.M., and U01NS120820,
late neuronal growth phenotypes. 158, 212–218 (2006). U01NS115579, and 2R01MH101214-06 to L.T.), three NIH training
Without facilitated transport across the 22. B. J. Ebersole, I. Visiers, H. Weinstein, S. C. Sealfon, Mol. Pharmacol. grants (T32GM099608 to M.V.V., T32GM113770 to R.J.T., and
63, 36–43 (2003). T32MH112507 to H.N.S.), Human Frontier (L.T.), the Camille and
plasma membrane, serotonin cannot induce 23. R. H. J. Olsen et al., Nat. Chem. Biol. 16, 841–849 (2020). Henry Dreyfus Foundation (D.E.O.), a sponsored research
psychedelic-like effects on neuronal morphol- 24. J. C. Bermak, M. Li, C. Bullock, Q. Y. Zhou, Nat. Cell Biol. 3, agreement with Delix Therapeutics (D.E.O.), and a University of
ogy. Although it is possible that serotonin could 492–498 (2001). California (UC) Davis Provost’s Undergraduate Fellowship (S.J.C.).
25. U. E. Petäjä-Repo, M. Hogue, A. Laperriere, P. Walker, This project used the Biological Analysis Core of the UC Davis
alter cortical neuron physiology by activating M. Bouvier, J. Biol. Chem. 275, 13727–13736 (2000). MIND Institute Intellectual and Development Disabilities Research
cell-surface 5-HT2ARs, this receptor pool does 26. S. E. Jacobsen et al., J. Biol. Chem. 292, 6910–6926 Center (U54 HD079125). The Nikon High Content Analysis
not seem to be involved in 5-HT2AR–induced (2017). Spinning Disk Confocal microscope used in this study was
27. C. A. Purgert et al., J. Neurosci. 34, 4589–4598 (2014). purchased using NIH Shared Instrumentation Grant 1S10OD019980-
structural plasticity. Our results raise the in-
28. V. Cornea-Hébert, M. Riad, C. Wu, S. K. Singh, L. Descarries, 01A1. We thank the MCB Light Microscopy Imaging Facility,
triguing possibility that serotonin may not be J. Comp. Neurol. 409, 187–209 (1999). which is a UC Davis Campus Core Research Facility, for the use of
the endogenous ligand for the population of 29. V. Cornea-Hébert et al., Neuroscience 113, 23–35 (2002). this microscope. Funding for the nuclear magnetic resonance
5-HT2ARs expressed inside cortical neurons. 30. C. L. Schmid, K. M. Raehal, L. M. Bohn, Proc. Natl. Acad. Sci. U.S.A. spectrometers was provided by the National Science Foundation
105, 1079–1084 (2008). (NSF CHE-04-43516) and NIH (08P0ES 05707C). Analysis
Alternative ligands could include methylated 31. R. Toneatti et al., Sci. Signal. 13, eaaw3122 (2020). for this project was performed in the UC Davis Campus Mass
congeners of serotonin or TRY because these 32. J. E. Saffitz, S. B. Liggett, Circ. Res. 70, 1320–1325 (1992). Spectrometry Facilities, with instrument funding provided by the
compounds have greater abilities to cross non- 33. Q. Fu, Y. K. Xiang, Prog. Mol. Biol. Transl. Sci. 132, 151–188 NIH (1S10OD025271-01A1). Several of the drugs used in this study
Science (ISSN 1095-9203) is published by the American Association for the Advancement of Science. 1200 New York Avenue NW,
Washington, DC 20005. The title Science is a registered trademark of AAAS.
Copyright © 2023 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim
to original U.S. Government Works