University of Southern California Dissertations and Theses

Mechanisms of P2XR-mediated ethanol consumption

(USC Thesis Other)

Mechanisms of P2XR-mediated ethanol consumption

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Mechanisms of P2XR-mediated Ethanol Consumption
by
Larry Rodriguez
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHARMACEUTICAL SCIENCES)
August 2020
Copyright 2020 Larry Rodriguez

ii

Epigraph

“I'm not saying I'm gonna rule the world or I'm gonna change the world, but I guarantee
you that I will spark the brain that will change the world. And that's our job, it’s to spark
somebody else watching us.”
Tupac Shakur

iii

Acknowledgments
I want to thank my parents Adrian Rodriguez and Gloria Sanchez for supporting
me throughout my academic career, and for inspiring me to pursue my passions. I want
to thank my girlfriend Lourdes Johanna Avelar Portillo for all her love and care, and her
family for their support. I want to thank my advisor, Dr. Daryl Davies for his guidance,
mentorship, and patience, and for always being willing to take a chance on me. I want to
thank Drs. John Woodward, Mark Brodie, and Eric Boué-Grabot for their advice and
feedback. I want to thank Dr. Chang You for her patience in teaching me
electrophysiology. I want to acknowledge my committee members, Dr. Liana Asatryan for
her unwavering willingness to help, and Dr. Curtis Okamoto for always being available to
talk about science and other issues.
The work in Chapter 4 would not have been possible without the support of my
family: Gabriel and Silvia Delgado, Coraima Delgado, Claricela Delgado, Bernabe
Rodriguez, Edgar Rodriguez, Soccorro Valenzuela, and Freddy Valenzuela. I want to
thank all my friends who have made this dissertation possible with their support: George
Martinez, Johnny Varela, Anthony Ybarra, Kenny Varela, Roberto Varela, Luis Sanchez,
Jeffrey Andrade, Lazaro Escalante, Jaylan Padilla, Tyriq Herbin, Juan Banuelas, Rodrigo
Gonzalez, Octaviano Mares, Jose Cabrales and Clinton Donkers.
I also want to thank Alan Guan, Tahmina Hasan, Ishtiak Aziz, and Nirali Patel, my
Bravo Medical Magnet High School student research assistants (STAR program). I want
to thank my undergraduate students, Jesse Chen, Brandon Reyes, Aleksandra
Konovnitsyna, Megan Ryu, Edward Jeon, Catherine Yi, Cameron Chu, Michael
Bloomfield, Sharyse Watanabe, and Brent Han. It was a pleasure mentoring you all, and

iv

I’ve learned just as much from you all as you have learned from me.
I want to thank so many members of the community at USC that made this
possible, especially Wade Thompson Harper, Rosie Soltero, and Juana Prieto. Sorry to
graduate affairs for all the headaches that came with getting me registered in classes,
paid, etc. Thanks to my good friend and colleague Albert Lam for helping me troubleshoot
molecular biology experiments, allowing me to vent, and generally making graduate
school bearable. I want to thank Vijaya Pooja Vaikari, Santosh Peddi, and Aida Kouhi for
being great friends, and for being part of the (best) 2017 AAPS-USC board. I also want
to acknowledge The Joe Budden Podcast, Joe Budden, and Shawn Carter for helping me
get through [many] late nights, early mornings, and tough times in the lab.
Arturo Rodriguez, sigues tú.
Authorship
Published and in preparation works by the author incorporated into the dissertation
Rodriguez, Larry; Yi, Catherine, Chu, Cameron; Watanabe, Sharyse; Asatryan, Liana; and
Davies, D. L. ; Cross-talk between P2X and NMDA receptors: a novel mechanism for
regulation. Molecular Pharmacology (2020, in preparation)

Rodriguez, Larry; You, Chang; Brodie, Mark; Asatryan, Liana; and Davies, Daryl. L. ;
Ethanol antagonizes P2X4 receptors in ventral tegmental area neuron. [published online ahead
of print, 2020 Jul 10]. Neuroreport. 2020;10.1097/WNR.0000000000001504.
doi:10.1097/WNR.0000000000001504

Popova, M.*; Rodriguez, L.*; Trudell, J.R.; Nguyen, S.; Bloomfield, M.; Davies, D.L.; Asatryan,
L. Residues in Transmembrane Segments of the P2X4 Receptor Contribute to Channel
Function and Ethanol Sensitivity. Int. J. Mol. Sci. 2020, 21, 2471.
* Co-first authors
Additional works by the author relevant to this dissertation but not forming part of it
Cayla NS, Dagne BA, Wu Y, Lu Y, Rodriguez L, Davies DL, Gross ER, Heifets BD, Davies MF,
MacIver MB, Bertaccini EJ. A newly developed anesthetic based on a unique chemical core.
Proc Natl Acad Sci U S A. 2019 Jul 30;116(31):15706-15715. doi: 10.1073/pnas.1822076116.
Epub 2019 Jul 15. PMID: 31308218; PMCID: PMC6681746.

v

Yang S, Chen Y, Feng M, Rodriguez L, Wu JQ, Wang MZ. Improving Eflornithine Oral
Bioavailability and Brain Uptake by Modulating Intercellular Junctions With an E-cadherin
Peptide. J Pharm Sci. 2019;108(12):3870–3878. doi:10.1016/j.xphs.2019.09.015
Research Support
NIAAA/NIH R01 AA022448 (D.L.D.), the American Foundation for Pharmaceutical Education
(AFPE) predoctoral fellowship (L.R.) and the USC School of Pharmacy.

vi

Table of Contents
Epigraph ................................................................................................................................................... ii
Acknowledgments ................................................................................................................................. iii
List of Tables ........................................................................................................................................ viii
List of Figures ........................................................................................................................................ ix
Abstract .................................................................................................................................................... x
Chapter 1 Introduction ................................................................................................................ 1
Alcohol as a drug of abuse ................................................................................................................. 1
The Neurobiology of motivation and addiction ............................................................................... 3
Dopamine, reinforcement, and addiction .................................................................................... 5
Ligand-Gated Ion channels ................................................................................................................ 6
The intoxicating effects of ethanol on LGICS............................................................................. 8
GABA receptors ................................................................................................................................ 9
Glutamatergic receptors .................................................................................................................. 9
Purinergic receptors (P2XRs) ...................................................................................................... 10
Dissertation Hypothesis and Outline .............................................................................................. 13
Chapter 2 Residues in Transmembrane Segments of the P2X4 Receptor Contribute to
Channel Function and Ethanol Sensitivity^ .................................................... 14
Chapter 2 Abstract ............................................................................................................................... 14
Materials and Methods ........................................................................................................................ 17
Results ................................................................................................................................................... 21
TM1 Alanine Scan Revealed Residues That Are Important for Receptor Function .............. 21
Alanine Scan of the TM1 Revealed Sites for Ethanol Action .................................................... 24
Physical-Chemical Properties of Residues at Position 33 Determine Receptor
Function and Ethanol Sensitivity .................................................................................................... 26
Residues That Are Important for Agonist Sensitivity and Ethanol Effects Identified
in Homology Models ........................................................................................................................ 27
Arginine 33 and Aspartic Acid 354 Interactions ........................................................................... 29
Discussion ............................................................................................................................................. 32
Function of valine at position 49 in the upper portion of the TM1 segment ............................ 32
Arginine at position 33 in the TM1 segment in ethanol action ................................................... 34
Interaction between TM1 arginine at position 33 and TM2 aspartic acid at position
354 in channel functioning .............................................................................................................. 35
Conclusion ............................................................................................................................................. 38
Chapter 3 Cross-talk between P2X and NMDA receptors: a novel form of neuromodulation ..... 40
Chapter 3 Abstract ............................................................................................................................... 40
Introduction ............................................................................................................................................ 41
Materials and Methods ........................................................................................................................ 44
Results ................................................................................................................................................... 46
Co-expression of P2XRs and NMDARs does not affect the function of individual
channels............................................................................................................................................. 46
Co-activation of P2XRs and NMDARs produces non-additive (inhibitory) responses .......... 48
P2X4R-NMDAR cross-talk is independent of Ca
2+
influx ........................................................... 51
Recovery of P2X4R inhibition is GluN2-subunit specific ............................................................ 52

vii

P2XR-NMDAR cross-talk depends on a shared C-terminal motif ............................................ 54
Resolving the P2XR-NMDAR cross-talk motif ............................................................................. 57
Disrupting P2XR-NMDAR cross-talk ............................................................................................. 59
Discussion ............................................................................................................................................. 61
P2XR modulation of NMDARs ....................................................................................................... 62
P2XR intracellular domains mediate NMDAR cross-talk ........................................................... 64
Resolving the role of P2XRs in the brain ...................................................................................... 66
Chapter 4 Ethanol antagonizes P2X4 receptors in ventral tegmental area neurons^ ................... 68
Chapter 4 Abstract ............................................................................................................................... 68
Introduction ............................................................................................................................................ 69
Materials and Methods ........................................................................................................................ 70
Results ................................................................................................................................................... 74
Reduction of P2X4Rs after incubation with siRNA ..................................................................... 74
VTA neurons were significantly inhibited by ATP........................................................................ 76
Discussion ............................................................................................................................................. 79
Chapter 5 Conclusion .............................................................................................................................. 82
Summary of overall findings ............................................................................................................... 82
Limitations ............................................................................................................................................. 84
Future directions ................................................................................................................................... 85
List of References ................................................................................................................................ 89

viii

List of Tables
Table 1. ...................................................................................................................................................... 22
Table 2 ....................................................................................................................................................... 23
Table 3 ....................................................................................................................................................... 30

ix

List of Figures
Figure 1.1 .................................................................................................................................................... 4
Figure 1.2 .................................................................................................................................................... 8
Figure 1.3 .................................................................................................................................................. 11
Figure 2.1 .................................................................................................................................................. 23
Figure 2.2 .................................................................................................................................................. 26
Figure 2.3 .................................................................................................................................................. 28
Figure 2.4 .................................................................................................................................................. 30
Figure 2.5 .................................................................................................................................................. 32
Figure 3.1 .................................................................................................................................................. 48
Figure 3.2 .................................................................................................................................................. 50
Figure 3.3 .................................................................................................................................................. 52
Figure 3.4 .................................................................................................................................................. 54
Figure 3.5 .................................................................................................................................................. 56
Figure 3.6 .................................................................................................................................................. 58
Figure 3.7 .................................................................................................................................................. 61
Figure 4.1 .................................................................................................................................................. 75
Figure 4.2 .................................................................................................................................................. 78

x

Abstract
Alcohol use disorder (AUD) affects over 18 million people in the United States alone and
can be attributed to an economic burden of over $200 billion per year. Despite
considerable efforts toward developing new therapies for AUD, there are currently only
three FDA-approved medications to pharmacologically treat AUD, none of which are
considered to be particularly effective in reducing alcohol consumption or craving. The
field of neuroscience has an abundance of research dedicated to the roles of individual
receptors involved in alcohol addiction, such as gamma-aminobutyric acid (GABA) and
glutamate (NMDA) receptors, although attempts to target these receptors has failed to
produce promising results in the clinic. More recently, the purinergic (P2XR) family of
receptors has become an emerging target for various diseases. In particular, the
purinergic receptor family member P2X4 has been linked to the voluntary alcohol
consumption pathway in AUD mouse models. Genetic, pharmacological and behavioral
mouse studies report an inverse relationship between ethanol (EtOH) intake and P2X4R
activity, although the underlying mechanism for this phenomenon is not understood.
Interestingly, several members of the P2XR family have been shown to cross-regulate
the activity of another receptor type (e.g. GABAA or AMPA receptors.) P2X4Rs are
expressed in the Ventral Tegmental Area (VTA) of the brain, along with ionotropic NMDA
receptors (NMDARs), a central component of addiction circuitry. The goal of this
dissertation is to investigate molecular and cellular mechanisms by which P2X4Rs
regulate ethanol consumption and addiction. This is accomplished using complementary
techniques in electrophysiology, molecular biology, and biochemistry. Chapter 1
describes our understanding of AUD as it relates to treatment and pharmacotherapy.

xi

Chapter 2 describes a novel interaction between residues that is required for P2X4R
channel function and a site for ethanol action. Chapter 3 describes an interaction between
P2XRs (either P2X2R or P2X4R) and NMDARs, which provides a previously
unrecognized link to addiction targets. Chapter 4 describes the role of P2X4Rs in
mediating the effects of ethanol in the ventral tegmental area (VTA) of the brain, a region
known to play a key role in reward and addiction. Chapter 5 summarizes the results of
my findings, describes the limitations of my studies, and provides future directions of
investigation. Overall, my findings 1) indicate that P2X4Rs play a previously unrecognized
role in neuronal cell firing within the VTA 2) suggest a novel mechanism for the regulation
of NMDARs, and 3) provide potential targets for the treatment of AUD and other
neurological diseases.

1

Chapter 1 Introduction
Alcohol as a drug of abuse
Humans have long been aware of alcohol’s intoxicating properties, as cultures
from around the world have been consuming alcoholic beverages for thousands of years,
either recreationally or ceremoniously. In fact, up until the 20
th
century, alcohol could have
been considered a full-fledged medicine, as it was given to patients undergoing surgery,
due to its perceived anesthetic properties. However, for those unable to control their
consumption of alcohol, the line between recreation and addiction is blurred. Ethanol, the
agent responsible for the intoxicating effects of alcoholic beverages, is a drug with
potential for dependence and devastating health and societal consequences.
Individuals that are consuming repeated patterns of unhealthy drinking are
diagnosed as suffering from an Alcohol use disorder (AUD; described in greater detail in
the next paragraph). Unhealthy drinking can be generally categorized into two distinct
patterns of problematic alcohol consumption: binge drinking in shorter periods of time or
heavy drinking over a long period of time. AUD is a serious health problem; it affects over
18 million (or 1 in 20) people (WHO, 2014) in the United States and is linked to an
economic burden of over $250 billion per year (NIAAA, 2020). On a global scale, roughly
1 in 5 (18.4%) of adults (or 39.6% of all adult alcohol consumers) report heavy episodic
drinking (≥60 g ethanol in one occasion) in a given 30 day time frame (Peacock et al.,
2018). Compared to other drugs, alcohol is the most prevalent substance of dependence
in the world, with 63.5 million reported cases in 2015 (Peacock et al., 2018).
AUD, as described by the Diagnostic and Statistical Manual of Mental disorders

2

(DSM-V), is defined as a person that has experienced two or more of the drinking
symptoms listed in the DSM-V within a 12 month period of time (2013). In the United
States, a “drink” is considered a US is 12 oz of beer, 5 oz of wine, or 1.5 oz of [80 proof]
liquor. According to National Institute on Alcohol Abuse and Addiction (NIAAA) a binge
session is defined as a 2 hour period of drinking, where a woman suffering from AUD
consumes ≥4 alcoholic beverages, while men suffering from AUD consume ≥5. On the
other hand, a heavy drinker consumes alcohol more frequently; for women, this means
drinking >1 alcoholic beverage per day, while for men, this means >2 beverages per day.
While there may be different classifications of AUD, AUD follows the general mechanism
by which many addictions develop: a repetitive cycle of binge/intoxication,
withdrawal/negative affect, and preoccupation/anticipation (i.e. craving). These stages
can worsen and grow more intense over time, resulting in changes in the brain circuitry
(plasticity) and how it responds to reward, stress, and executive function, which makes
recovery/abstinence difficult (Koob and Volkow, 2010).
While we may have some understanding of how the brain is affected by ethanol
consumption and addiction, much remains unknown. This paucity of understanding of
AUD is exemplified by the lack of FDA approved drugs to treat AUD. Presently, there are
only 3 FDA-approved drug therapies to treat AUD, and these compounds are ineffective
for most patients suffering from AUD. Disulfiram, the oldest AUD “treatment” option,
negatively affects patients for their ongoing drinking, as consumption of alcohol in the
presence of disulfiram causes nausea and sickness (the drug produces a buildup of
acetaldehyde in the body by blocking the process of ethanol metabolism) (Wright and
Moore, 1990). Naltrexone is a µ-opioid receptor antagonist originally developed and

3

approved for opioid dependence, and has been found to have modest effects alcohol
consumption and the consumption frequency, although it was not found to reduce the
number of people suffering from AUD when compared to control groups (Rösner et al.,
2010). Acamprosate is another drug used to treat AUD, that was first approved in Europe.
However, additional clinical trials have not demonstrated clinical benefit in either short,
medium, or long-term treatment (Plosker, 2015). Furthermore, data from The National
Survey on Drug Use and Health indicates that only 8.2% of individuals with a 12-month
history of AUD had received any pharmacological treatment, and even then, those who
did receive pharmacological treatment faced a high rate of relapse. These low rates of
beneficial treatment reflect the fact that current therapies for AUD are mostly inadequate
for the management of AUD; physicians are unaware/unfamiliar with treatment options
and may even doubt their efficacy (Kim et al., 2018). This skepticism is not unfounded as
none of the FDA-approved medications to treat AUD (disulfuram, naltrexone, and
acamprosate) were specifically developed to treat AUD, nor are they particularly effective
in reducing alcohol consumption or craving. The continued failure of traditional AUD drug
targets in the clinic underscores the need for better mechanistic insights and
consequently, the discovery of novel AUD targets.
The Neurobiology of motivation and addiction
All organisms possess motivation, a process by which they respond to a stimulus
in a way that reflects how a predicted result impacts the survival of itself, as well as its
species. An underlying facet of motivation is learning about how environmental or
biological stimuli are related to and might predict the stimuli/responses that follow or
precede them. Learning how stimuli and responses form predictive relationships allows

4

an organism to seek out and achieve biologically valuable events. Interestingly, various
brain regions/circuits that play a role in how organisms learn and develop/process
motivation have been shown to be affected by drug consumption/addiction (Di Chiara,
2002). Additionally, the drug-taking cycle shares characteristics of impulse control
disorders, which are associated with positive reinforcement, and compulsive disorders,
which associated with negative reinforcement. For example, these positive
reinforcements can be the pleasure derived from committing the act of drug taking or the
increased sense of tension/arousal before the drug taking act, while the negative
reinforcements reduce the stress and anxiety felt before performing a compulsive [drug-
taking] act.

Figure 1.1
Schematic diagram of circuits and their role in the development addiction (Koob and

5

With this in mind, early stages of drug consumption can be thought of as primarily
driven by impulsivity, while the later stages involve (and are subsequently dominated by)
compulsive motivations. As the addiction cycle progresses, these stages become more
intense, leading to more drug intake and ultimately full-fledged addiction (Koob and
Volkow, 2010). Understanding how different brain circuits are affected by ethanol and
their role in the development of addiction (Figure 1) are important for developing safe,
effective medication to treat AUD.
Dopamine, reinforcement, and addiction
It is well established that drugs of abuse disrupt the brain’s reward system, which
has long been theorized to play a crucial role in the development of addiction (Di Chiara,
2002). Furthermore, dopamine changes have been shown to play a role in each of the
three stages of the alcohol addiction cycle. For example, studies have shown that alcohol
activates the reward system in ways that contribute to the binge/intoxication stage; by
directly activating DA neurons in the ventral tegmental area (resulting in an increase in
firing rate) or by increasing the amount of DA released into other areas of the brain,
including the shell of the nucleus accumbens (Koob, 1992). Human imaging studies by
Volkow et al showed that during the withdrawal stage, subjects suffering from AUD
release lower amounts of DA than control subjects (Volkow et al., 2007.) Attempts to
Volkow, 2010). Left Several neuroadaptations that occur within brain circuits that promote drug
seeking behavior once addiction has developed addicted state. Arrows originate at regions that
affect other circuits. Right The addiction cycle (preoccupation/anticipation in green,
binge/intoxication in blue, withdrawal/negative affect in red) with the brain regions responsible
for mediating drug seeking behaviors. Green and blue arrows denote glutamatergic projection,
orange arrows denote dopaminergic projections, pink arrows denote GABAergic projections.
Brain regions: Acb indicates nucleus accumbens, BLA indicates basolateral amygdala, BNST
indicates bed nucleus of the stria terminalis, CeA indicates central nucleus of the amygdala,
DGP indicates dorsal globus pallidus, VGP indicates ventral globus pallidus, VTA indicates
ventral tegmental area, SNc indicates substantia nigra pars compacta. Neurotransmitters: NE
indicates norepinephrine, CRF indicates corticotropin-releasing factor, DA indicates dopamine.

6

“compensate” for this DA deficit puts AUD patients at higher risk for consuming excess
alcohol. These results translate well with animal studies, which found significant
reductions in DAergic neuron firing in the VTA (Shen et al., 2007) and decreased DA in
NAc (Weiss et al., 1996) after withdrawal from chronic alcohol. Additionally, studies have
suggested that, once the symptoms of acute symptoms of withdrawal have subsided,
hypofunction in DA pathways (reductions in both DA release and D2 receptor expression)
may contribute to a reduction in sensitivity to a rewarding stimulus (anhedonia.) (Martinez
et al., 2005; Volkow et al., 2007). Even during the preoccupation/anticipation stage, the
negative state associated with acute drug abstinence is thought to be related to
reductions in reward neurotransmitter function, causing long-term biochemical changes
that contribute to high rates of AUD relapse (Weiss et al., 1996.)
One could speculate that, if dopamine and reward dysregulation is involved in the
development of AUD, then therapies to treat AUD should target the reward system.
Indeed, naltrexone and acamprosate have been shown to affect the mesolimbic DA
system. However, while studies suggest that dopamine contributes in the reinforcing
properties of drugs of abuse, dopamine-independent reward properties have been shown
to exist for both opiates and ethanol. An alternative view suggests that dopamine serves
as a critical link for reward function (Koob, 1992) and that ligand-gated ion channels are
more promising targets for AUD, given their role in alcohol intoxication/addiction.
Ligand-Gated Ion channels
Ligand gated ion channels (LGICs) perform important signaling and regulatory
functions in neurons, and serve as targets for the intoxicating and addictive effects of
ethanol (EtOH). Upon ligand binding, they facilitate the depolarization of cell membranes

7

by forming channels for ions (typically cations such as Na
+
, Ca
+
, and K
+
or anions such
as Cl
-
) to pass through and enter cells (Figure 2). These channels can be activated by
various chemical species, such as glutamate (Glu), adenosine triphosphate (ATP), and
glycine (Gly), with multiple receptors being found at the postsynaptic clefts of neurons
(Barria and Malinow, 2002). Of these, glutamate receptors are one of the largest and most
widely expressed family of excitatory LGICs found in the CNS.

8

The intoxicating effects of ethanol on LGICS
In the U.S., a person is considered legally intoxicated, and unable to operate a
vehicle, if their blood ethanol concentration is >0.08%, or 17 mM. At physiologically

Figure 1.2
Representative diagram of Ligand-gated ion-channels (LGICS) on neurons Top Diagram
demonstrating a synapse. Neurotransmitters are released from the presynaptic membrane
(purple) into the synaptic junction. These molecules diffuse and bind to LGICs on the
postsynaptic membrane (blue) resulting in their activation. Bottom Illustrations demonstrating
LGIC activation. In the absence of agonist, LGICs remain closed (left), which prevents entry of
ions. Upon binding of agonist (or in some cases, drugs), LGICs open and allow charged ions
into the cell, changing the membrane potential from rest (e.g. -70 mV) if cation channels, or
preventing a change in membrane potential if anion channels.

9

relevant BEC, LGICS have been shown to be affected by ethanol, resulting in either
potentiation or inhibition of their activity. At the same time, LGICs have also been shown
to play a role in alcohol consumption. Examples include the gamma-aminobutyric acid
(GABA), Glutamate (NMDA), and more recently, the purinergic family of receptors. In
particular, the purinergic receptor family member or P2X4R has been linked to the
voluntary alcohol consumption pathway in AUD mouse models (Huynh et al., 2017; Wyatt
et al., 2014a; Yardley et al., 2012).
GABA receptors
It is well established that gamma-aminobutyric acid-A receptors (GABAARs) are a
target for the action of ethanol: ethanol can directly activate GABAARs and induce GABA
release in the VTA, nucleus accumbens, and amygdala. As such, GABAARs have been
the focus of ethanol research and drug development for the past twenty plus years.
However, no direct acting GABAAR drugs have been approved to treat AUD. Baclofen, a
GABABR inhibitor, has been used off-label to treat AUD (Winslow et al., 2016), as studies
have suggested that it might reduce the symptoms of alcohol withdrawal, dependence,
and craving, although results have been mixed. Additionally, Gabapentin also has been
shown to potentiate inhibitory GABABRs, although it is important to note that gabapentin
and baclofen have potential for abuse (Kim et al., 2018). Nonetheless, because of the
importance of GABA neurotransmission, the scientific community continues to investigate
the importance of GABARs and the action of ethanol.
Glutamatergic receptors
Glutamate receptors are a second class of LGICs that are known to be targets for
ethanol action. Within the superfamily of Glutamate receptors, three different subtypes of

10

ionotropic glutamate receptors exist, differentiated by their ability to be stimulated by
selective agonists: Kainate, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (or
AMPA), and N-methyl-D-aspartate receptors (or NMDARs). NMDARs are widely held to
play a key role in alcohol addiction (Wang et al., 2007; Xiao et al., 2008a) and several
rodent models of addiction have demonstrated that NMDAR modulation plays a
significant role in the development of alcohol dependence (Creed et al., 2016; Eisenhardt
et al., 2015; Nagy et al., 2005). NMDARs are heteroteteramers, usually consisting of two
GluN1 subunits and either two GluN2 or two GluN3 subunits. Within the NMDA type of
glutamate receptors, there exist several subtypes of GluN2 (i.e., GluN2A-D), each with a
different cytoplasmic domain, resulting in differences in functional and physiological
activities (Barria and Malinow, 2002; Kopec et al., 2006). For example, the recycling fate
of NMDARs differs depending on the subunit composition (i.e., GluN2B containing
tetramers traffic to recycling endosomes, while GluN2A tetramers associate to late
endosomes (Lin et al., 2006; Prybylowski et al., 2005; Prybylowski et al., 2002). Removal
of NMDARs from the cell surface can induce long term depression, a form of synaptic
plasticity that impacts learning and memory(Sans et al., 2003; Yi et al., 2007). NMDARs
are 1) a key targets for EtOH 2) primary regulators of dopamine neuronal activity 3)
important components of addiction circuitry and 4) expressed in the VTA with dopamine
receptors and P2X4Rs (Bo et al., 2003; Khakh and North; Sheraz Khoja, 2016; Xiao et
al., 2008a).
Purinergic receptors (P2XRs)
Another superfamily of LGICS that is becoming recognized as important for
neuronal signaling and alcohol addiction is the purinergic family of receptors, named for

11

their agonists: purines. Building evidence indicates that the ionotropic subset of this
receptor superfamily, or “P2XR” are linked to the regulation of neuroinflammation, pain,
and other neurological conditions (Khakh and North). One member of this family of ATP-
gated channels, P2X4R, is suggested to represent a target for the development of drugs
to prevent and/or treat alcohol
use disorder (AUD). For example,
in our laboratory we demonstrated
that male P2X4R KO mice (i.e.,
p2rx4 gene deleted) consumed
significantly more EtOH as
compared to wildtype controls
(Yardley et al., 2012) (Fig 3A)
suggesting that P2X4Rs play a
role in the regulation of EtOH
intake (Yardley et al., 2012). In
agreement with this hypothesis,
two related macrocyclic lactones,
ivermectin (IVM) and moxidectin
(MOX) that potentiate P2X4R function, can antagonize EtOH-inhibition of ATP gated
currents in P2X4Rs in vitro and significantly reduce EtOH intake in mice (Asatryan et al.,
2014b; Huynh et al., 2017).Investigations testing P2X4R KO mice identified not only
cognitive behavioral changes, but also dysregulation of several receptors within various
regions of the brain as well as aberrant signaling within the mesolimbic pathway, including
a)
b)
Figure 1.3
P2X4R genotype contributes to altered EtOH intake
in male C57BL/6 mice using a 24-h-two-bottle
choice paradigm 3a): P2X4R KO mice exhibited
significant increase in EtOH intake as compared to WT
littermates [F(1,16)=4.88, p<0.05]. Values represent
mean ± SEM for 8 WT and 10 P2X4R KO mice,
*P<0.05 versus WT for entire period, #P<0.05 versus
WT in week 5. 3b): Knockdown of p2rx4 gene via LV-
shRNA treatment significantly increased EtOH as
compared to baseline levels prior to surgery and
compared to saline injected mice.
[F(5,48)=4.368,p<0.01]. Values represent mean ±
SEM for 9 mice, **P<0.01 vs baseline.

12

glutamate receptors (Letisha R Wyatt, 2013; Sheraz Khoja, 2016). While these data
indicate that P2X4Rs are integral in neuronal signaling and disease states, including
AUD, there remains a paucity of information regarding how P2X4Rs and ethanol interact
(for a more detailed description of P2X4R sites of ethanol action, see Chapter 2
introduction).
Receptor cross-talk
Neurological signaling through multiple receptors and pathways is recognized as
being an important mechanism for various cognitive functions, such memory, learning,
and behavioral responses. Electrophysiology studies are a recognized method to
investigate individual receptors, which explore the physiological and functional roles of
those receptors in transmitting electrical signals under normal and diseased conditions.
On the other hand, few studies have investigated how these families of receptors interact
(communication between receptor families) and influence one another within neurons.
Recent investigations suggest that P2XRs interact with other receptor
superfamilies, including GABAA and AMPA receptors (Jo et al., 2011; Pougnet et al.,
2014). These multi-receptor interactions provide a novel form of neuronal signaling
(cross-talk) and are suggested to play a role in the regulation of synaptic plasticity, either
through direct protein-protein interactions, or by co-trafficking (Boué-Grabot et al., 2004a;
Jo et al., 2011; Pougnet et al., 2016; Pougnet et al., 2014). However, the effects of
receptor cross-talk between P2X4Rs and NMDARs has not yet been explored, nor have
any P2XR cross-talk studies focused on the action of EtOH. These two issues represent
a significant gap in our knowledge in neuroscience and alcohol investigations. A better
understanding of how these receptors interact should provide a novel function for P2X4Rs

13

and identify new roles for P2X4Rs in alcohol addiction.
Dissertation Hypothesis and Outline
A primary goal of my dissertation is to present the results of studies that investigate
the molecular and cellular mechanisms by which P2X4Rs are affected by and contribute
to ethanol effects. This will be accomplished using a multidisciplinary approach that
combines electrophysiology, molecular biology, and biochemistry. In Chapter 2 of my
dissertation, I present data based on my investigation focusing on the structure-function
relationship of P2X4Rs and how ethanol affects the function of these channels. In Chapter
3, I present data on my investigation into on how interactions between P2XRs and
NMDARs produce a distinct form of signaling (crosstalk), which could serve as the
mechanism by which P2X4Rs contribute to alcohol consumption. In Chapter 4, I present
data focused on my studies investigating the effects of ethanol on P2X4Rs at the cellular
level: in dopaminergic neurons from the ventral tegmental area, a key component of the
mesolimbic (reward) system. Chapter 5 of my dissertation, I summarize my studies as
they relate to AUD and drug discovery, discuss the limitations of my studies, and provide
future directions of research.

14

Chapter 2 Residues in Transmembrane Segments of the
P2X4 Receptor Contribute to Channel Function and Ethanol
Sensitivity^
Popova, M.*; Rodriguez, L.*; Trudell, J.R.; Nguyen, S.; Bloomfield, M.; Davies, D.L.; Asatryan,
L. Residues in Transmembrane Segments of the P2X4 Receptor Contribute to Channel
Function and Ethanol Sensitivity. Int. J. Mol. Sci. 2020, 21, 2471.
* Co-first authors
^ Note: This chapter is taken from the previously mentioned publication.
Chapter 2 Abstract
Mouse models of alcohol use disorder (AUD) revealed purinergic P2X4 receptors
(P2X4Rs) as a promising target for AUD drug development. We have previously
demonstrated that residues at the transmembrane (TM)–ectodomain interface and within
the TM1 segment contribute to the formation of an ethanol action pocket in P2X4Rs. In
the present study, we tested the hypothesis that there are more residues in TM1 and TM2
segments that are important for the ethanol sensitivity of P2X4Rs. Using site ‐directed
mutagenesis and two electrode voltage ‐clamp electrophysiology in Xenopus oocytes, we
found that arginine at position 33 (R33) in the TM1 segment plays a role in the ethanol
sensitivity of P2X4Rs. Molecular models in both closed and open states provided
evidence for interactions between R33 and aspartic acid at position 354 (D354) of the
neighboring TM2 segment. The loss of ethanol sensitivity in mixtures of wild ‐type (Bristow
et al.) and reciprocal single mutants, R33D:WT and D354R:WT, versus the WT ‐like
response in R33DD354R:WT double mutant provided further support for this interaction.

15

Additional findings indicated that valine at TM1 position 49 plays a role in P2X4R function
by providing flexibility/stability during channel opening. Collectively, these findings
identified new activity sites and suggest the importance of TM1 ‐TM2 interaction for the
function and ethanol sensitivity of P2X4Rs.
Introduction
Alcohol abuse and alcoholism (alcohol use disorder: AUD) continues to present a
serious health and economic burden worldwide (Sacks et al., 2015; WHO, 2014), due in
part to the lack of effective medications for treatment and/or prevention. However,
medication development is complicated by a lack of understanding of the targets, target
sites, and mechanisms of ethanol action (i.e., alcohol intoxication) in the central nervous
system (CNS). Growing evidence suggests that ethanol affects purinergic P2X4 receptors
(P2X4Rs), a member of the P2XR superfamily [for review, see (Franklin et al., 2014)]. Of
the seven family members, P2X4Rs are the most abundant P2XR subtype in the CNS
(Buell et al., 1996; Soto et al., 1996). P2X4Rs have been found in reward circuitry,
specifically the mesolimbic and mesocortical pathways (Gonzales et al., 2004). Gene
expression profiling studies have suggested that P2X4R expression may be involved in
innate alcohol preference, and, indeed, reduced regional levels of p2rx4 mRNA were
found in alcohol-preferring vs. alcohol non-preferring rats (Kimpel et al., 2007) as well as
high alcohol drinking vs. low alcohol drinking rats (Tabakoff et al., 2009).
In agreement with these findings, studies from our group revealed that ethanol intake
was higher in P2X4R-knockout male mice compared to their wild-type counterparts, in
both intermittent limited access and 24-hr access drinking models (Wyatt et al., 2014a).
Additionally, within the P2XR family, P2X4Rs are the most ethanol-sensitive subtype; in

16

vitro, P2X4Rs are inhibited by behaviorally relevant concentrations of ethanol (e.g., 17
mM which is equal to 0.08% blood ethanol, or the legal limit to drive in the United States
of America) (Davies et al., 2005; Olga Ostrovskaya, 2011; Popova et al., 2010; Xiong et
al., 2000). Together, these findings indicate that P2X4Rs and ethanol are related in the
behavioral, cellular, and molecular level, and therefore represent an important therapeutic
target for pharmacological strategies aimed at the prevention and/or treatment of AUD.
Functional P2XRs are homomeric (formed from a single P2X subtype) or heteromeric
(formed from different P2X subtypes) trimers. Structurally, each subunit consists of two
intracellular N- and C-termini, two transmembrane (TM) segments, and a large
extracellular domain (ectodomain) (Young et al., 2008). Interactions between the TM1
and TM2 segments have been shown to play a major role in the function of P2X4Rs.
Tryptophan and cysteine scanning studies of both TM1 and TM2 revealed that the upper
regions of each segment undergo substantial rearrangement and adopt α-helical
secondary structures during channel gating (Jelínkova et al., 2008; Silberberg et al., 2005;
Stojilkovic et al., 2010). Furthermore, interactions between the TM1 region of one subunit
and the TM2 region of another subunit have been shown to regulate the gating and ion
conductance of P2X4Rs, specifically the lower ends of the TM segments (Heymann et
al., 2013). Studies have suggested that homomeric P2X4Rs contain a putative ethanol
activity site formed by residues located at the ectodomain–TM interfaces (Asatryan et al.,
2010), aspartic acid at position 331, and methionine at 336 located in the upper part of
the TM2 segment (Popova et al., 2010), and two tryptophan residues at position 50 at
TM1–ectodomain interface and position 46 in the TM1 segment (Popova et al., 2010). In
this regard, homology models of the rat P2X4R [(Popova et al., 2010; Popova et al., 2013)

17

built using open and closed conformations of zebrafish P2X4R as templates (Hattori and
Gouaux, 2012; Kawate et al., 2009) suggest that residues located in the TM1 segment
influence ethanol interactions with P2X4Rs (Popova et al., 2010) and resolved potential
interactions between residues at the TM2–ectodomain interface. These findings provide
a strong support for the notion that there are multiple sites in the TM1 and TM2 segments
of P2X4Rs responsible for receptor function and ethanol sensitivity. Despite these
advances, we still lack a clear understanding of how other residues in the TM1 and TM2
segments function and/or interact during ethanol action in the P2X4R.
In this study, we employed an alanine scanning strategy on the TM1 segment of the
P2X4R to identify novel residues responsible for ethanol sensitivity and receptor function.
We explored possible interactions between the residues at the lower portions of TM1 and
TM2 segments with additional site-directed mutagenesis and molecular modeling. To
further understand the role of these residues in receptor function and ethanol response,
we also tested the effects of ivermectin (IVM) on several mutant receptors based on our
previous studies that found partially overlapping sites for ethanol and IVM, and its
recognition for interacting favorably with the open P2X4R conformation (Asatryan et al.,
2010; Egan et al., 1998).
Materials and Methods
Materials
Adenosine 5′-triphosphate disodium salt, ethanol (190 proof, USP), and other
chemicals were purchased from Sigma Co. (St. Louis, MO, USA). All other chemicals
were of reagent grade.
Isolation of Xenopus Laevis oocytes and cRNA injections

18

Xenopus oocytes were isolated from Xenopus Laevis (Nasco, CA, USA) and
maintained as described previously (Popova et al.). Stage V and VI oocytes were
selected, rinsed and stored in incubation medium containing (in mM), NaCl 96, KCl 2,
MgCl2 1, CaCl2 1, HEPES 5, theophylline 0.6, pyruvic acid 2.5, with 1% horse serum and
0.05 mg/mL gentamycin. The following day, oocytes were injected (32 nl, 20 ng/oocyte)
into the cytosol with cRNA encoding rat WT or mutant P2X4R or the mixtures of WT and
non-functional mutants (GenBank accession no. X87763). The injections were performed
with Nanoject III Nanoliter injection system (Drummond Scientific, Broomall, PA, USA).
The injected oocytes were stored in incubation medium at 17 °C and used in
electrophysiological recordings for 1–4 days after cRNA injections.
Site-directed mutagenesis and cRNA synthesis
The cDNA of rat P2X4R (GenBank accession no. X87763) was subcloned into
pcDNA3 vector (Invitrogen, Carlsbad, CA, USA). The mutations were generated by using
QuikChange II XL Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA). The
primers were purchased from Integrated DNA Technologies (Coralville, Iowa). Once the
mutations were sequence verified (Genewiz, La Jolla, CA, USA), plasmid DNA was
linearized by restriction digestion with XhoI and cRNA was synthesized using
mMESSAGE mMACHINE T7 Kit (Applied Biosystems, Foster City, CA, USA) and stored
at −70 °C until injection.
Whole-cell voltage clamp recordings
To increase the number of mutant receptors that can be efficiently screened, we used
an automated computer-controlled 8-channel two electrode voltage-clamp system,
OpusXpress (Molecular Devices, Union City, CA, USA). Oocytes were placed in 8

19

recording chambers (volume 200 µL), superfused with P2X buffer solution (in mM) (NaCl
110, KCl 2.5, BaCl2 1.8, HEPES 10, pH 7.5) at a rate of 3–4 mL⁄min, and impaled with 2
glass electrodes filled with 3 M KCl (0.5 to 2 MΩ). The membrane potential was held at
−70 mV, and the currents were sampled at 5 KHz and filtered at 1 KHz. The currents
acquired were stored on the computer for off-line analysis using Axon pClamp 9 software
(Axon Instruments, Union City, CA, USA).
Experimental procedures
All experiments were performed at room temperature (20–23 °C). To generate ATP-
concentration curves, oocytes were exposed to 0.05–100 µM ATP range for 5 sec
followed by 5–15 min washout. The effect of ethanol on P2X4R function is more robust
and reliable when tested in the presence of sub-maximal concentrations of ATP (typically
EC5–20) (Popova et al., 2010). Therefore, ethanol was co-applied with EC10 ATP. ATP
EC10-activated currents were measured before and after each ethanol application to take
into account possible shifts in the baseline current values. Ethanol did not affect the
resting membrane currents in oocytes expressing P2X4Rs in the absence of agonists or
in un-injected oocytes. IVM was pre-applied to the oocytes for 1 min and then co-applied
with EC10 of ATP.
Cell surface biotinylation and immunoblotting
The study was performed according to procedures established in our laboratory
(Perkins et al., 2009). Briefly, 2–5 days after injections with WT or mutant P2X4R cRNAs,
oocytes (15 oocytes/group) were incubated with 1.5 mg/mL membrane-impermeable
sulfosuccinimidyl 2-(biotinamido)-ethyl-1,3-dithiopropionate (Thermo Fisher Scientific
Inc., Rockford, IL, USA) for 30 min at room temperature. Oocytes were homogenized in

20

500 μL of lysis buffer (40 mM Tris (pH 7.5), 110 mM NaCl, 4 mM EDTA, 0.08% Triton X-
100, and 1% protease inhibitor cocktail (Sigma, St. Louis, MO, USA), and after the cellular
debris and yolk were removed by centrifugation at 10,000× g for 10 min, 50 μL aliquots
were stored at −20 °C to assess total receptor fraction. Biotinylated proteins were
captured by overnight incubation with streptavidin beads (Thermo Fisher Scientific Inc.,
Rockford, IL, USA) at 4 °C and eluted by heating at 95 °C for 10 min in SDS loading
buffer. The surface (biotinylated) and total proteins were then separated on SDS-PAGE
and transferred to PVDF membranes. The membranes were then incubated overnight
with polyclonal antibodies against P2X4Rs (1:2000 dilution; Alomone Labs, Jerusalem,
Israel), followed by incubation with HRP-conjugated secondary antibodies. Protein bands
were visualized using enhanced chemiluminescence (Thermo Fisher Scientific Inc.,
Rockford, IL, USA).
Homology Modeling
We built two new homology models of rat P2X4R based on the closed (PDB ID 4DW0)
and open (PDB ID 4DW1) zebrafish structures (Hattori and Gouaux). We used the
‘modeler’ module of Discovery Studio 3.5 (DS 3.5; Biovia, San Diego, CA, USA) to make
50 models of both the open and closed form, essentially as we previously described
(Popova et al., 2013). We chose the ‘best’ models based on force field energy. All side
chain rotamers of these selected models were optimized with the ‘side chain refinement’
module of DS 3.5 while the backbone atoms were fixed. Then harmonic backbone
restraints of 10 kcal/(mol x Angstrom) were applied and the models were optimized using
the Biovia version of the CHARMm force field in DS 3.5. Each model was ‘relaxed’ with a
brief (10 ps) molecular dynamics simulation at 300 K using the same backbone restraints.

21

Data analysis
The results are presented as a percentage change in ATP EC10-activated currents
(nA) after normalizing them with the response obtained with agonist alone. All the results
are expressed as mean ± SEM. The data were obtained from oocytes from at least two
different frogs. The n refers to the number of different oocytes tested. Significant
differences were determined by non-parametric one-way ANOVA. Prism (GraphPAD
Software, San Diego, CA, USA) was used to perform statistical analyses and curve fitting.
The ATP concentration response data were fitted to a concentration-response curve
using the logistic equation I/Imax = 100*[drug]
n
/([drug]
n
+

(EC50)
n
), where I/Imax is the
percentage of the maximum obtainable response, EC50 is the concentration producing a
half-maximal response, and n is the Hill coefficient (nH). Statistical significance was
defined as p < 0.05.
Results
TM1 Alanine Scan Revealed Residues That Are Important for Receptor Function
Amino acid residues within the TM1 segment of rat P2X4Rs, positions 29 through 49,
were individually mutated to alanine (A), excluding alanine residues at positions 34 and
41, which were mutated to tryptophan (W). All mutant receptors were tested for changes
in agonist response (Table 1). Apart from valine (V) 49 to alanine (V49A), all mutants
produced measurable ATP-gated currents. One-way ANOVA found a significant effect of
mutations on the maximum inducible currents (Imax). Individual t-tests indicated
significant differences in Imax values between the WT P2X4R and number of mutants
(Table 1). Further analysis found significant changes in EC50 values for F48A, W46A, and
L37A mutants and significantly lower values for the Hill slopes for F48A, W46A, G45A,

22

and R33A mutants when compared to that of the WT P2X4R (Table 1, see also Appendix
A Figure A1A for some of the mutants).
Table 1.
Imax, EC50, and Hill slope values for ATP concentration-response curves of wild-type and TM1
segment mutant P2X4Rs. Imax represents the peak currents generated by 100 µM ATP. EC50 and
Hill slope values were determined from ATP concentration-response curves. The data shown are
mean ± SEM; individual t-test (non-parametric, Mann–Whitney),
#
p < 0.05,
##
p < 0.01 compared
to WT P2X4Rs. One-way ANOVA, * p < 0.05, ** p < 0.01 compared to WT P2X4Rs; n.f. – non-
functional; n.d. - not determined.
Receptor ATP
Imax, nA EC50, μM Hill slope N
WT P2X4 4215 ± 678 5.74 ± 0.45 1.48 ± 0.10 9
V49A n.f. n.d. n.d. 7
F48A 8716 ± 1533
#
3.53 ± 0.73* 1.04 ± 0.09* 8
V47A 2049 ± 319
#
5.0 ± 0.76 1.27 ± 0.06 7
W46A 3230 ± 498 3.58 ± 0.57* 1.02 ± 0.08* 10
G45A 5994 ± 1100 4.65 ± 0.54 1.16 ± 0.05* 10
I44A 4025 ± 687 5.68 ± 0.42 1.21 ± 0.05 11
V43A 6987 ± 1602 4.83 ± 0.59 1.34 ± 0.13 7
Y42A 1372 ± 119
##
4.33 ± 0.7 1.11 ± 0.25 5
A41W 1667 ± 233
##
4.62 ± 0.47 1.34 ± 0.07 12
L40A 1633 ± 282
##
6.41 ± 0.73 1.50 ± 0.10 8
I39A 5544 ± 1510 5.87 ± 0.70 1.39 ± 0.07 7
L38A 5719 ± 1211 7.33 ± 1.05 1.32 ± 0.06 8
L37A 5688 ± 749 8.91 ± 0.84** 1.20 ± 0.08 7
Q36A 1568 ± 129
#
6.35 ± 1.81 1.33 ± 0.14 4
V35A 7079 ± 1491 6.35 ± 0.55 1.36 ± 0.10 6
A34W 4386 ± 608 6.50 ± 1.02 1.21 ± 0.06 11
R33A 3545 ± 748 7.04 ± 0.51 1.03 ± 0.02* 5
N32A 2986 ± 397 5.40 ± 1.00 1.44 ± 0.08 7
M31A 3456 ± 375 7.07 ± 1.64 1.49 ± 0.23 7
L30A 7161 ± 1188 6.76 ± 0.84 1.37 ± 0.19 6
G29A 1175 ± 44
##
6.43 ± 0.34 1.33 ± 0.10 8
n.f.
– non-functional;
n.d.
- not determined.
Based on the Imax, EC50, and Hill slope values, more notable changes occurred in
V49A, F48A, and W46A mutants. We have discussed changes and the role of W46
residue in P2X4R function in our previous publication (Popova et al., 2013). Not
undermining the role of residue at position 48, which can be studied in more detail in the
future, in this study we focused on V49 as mutation of this residue to alanine resulted in
a loss of function. To determine if this effect in V49A mutant was due to a lack of receptor
expression, we performed surface biotinylation and subsequent Western Blot analysis.

23

We found that the surface expression of this mutant receptor was similar to that of the
WT receptor (Figure 1). We next tested whether position 49 was involved in the function
of a putative hinge region theorized to play a role in the stability of the alpha-helical
structure of the TM1 segment (Li and Deber, 1992). This was accomplished by mutating
the valine at position 49 to proline (P) or glycine (G), as proline and glycine are known to
be alpha-helix destabilizing residues (Li and Deber, 1992; Silberberg et al., 2007). In
addition, leucine (L), an alpha-helix stabilizing residue(Li and Deber, 1992), was
substituted for valine (V49L).

Figure 2.1
Mutations at positions 49 and 33 do not significantly affect total and surface expression of mutant
P2X4Rs. Representative Western Blots showing the P2X4R bands at ~60 kDA for total and
biotinylated fractions for WT and mutant receptors.
ATP elicited a negligible response in V49P but measurable currents in V49G and
V49L mutants (Table 2). There was a right-shift in ATP-concentration response curves
for V49G and V49L mutants with significantly higher EC50 values (Table 2; Appendix A
Figure A1B). Like the V49A mutant, the V49P mutant was readily expressed on the
surface of the oocytes (Figure 1), suggesting that trafficking of the mutant receptor to the
surface was not affected by mutagenesis.
Table 2

24

Imax, EC50, and Hill slope values for ATP concentration-response curves of WT, position 49 and
position 33 mutant P2X4Rs. Imax represents the peak currents generated by 100 µM ATP. The
EC50 and Hill slope values were determined from ATP concentration-response curves. The data
shown are mean ± SEM; one-way ANOVA, * p < 0.05 compared to WT P2X4Rs.
Receptor ATP
Imax, nA EC50, μM Hill slope N
WT P2X4 5425 ± 1235 5.45 ± 0.4 1.47 ± 0.13 8
V49P n.f. n.d. n.d. 5
V49G 2131 ± 292* 9.60 ± 0.9* 1.23 ± 0.10 6
V49L 2365 ± 277 10.60 ± 1.3* 1.54 ± 0.18 4
R33K 5428 ± 986 6.33 ± 0.93 1.35 ± 0.09 9
R33S 2492 ± 912 4.94 ± 1.09 1.35 ± 0.22 6
R33V 5947 ± 1396 3.67 ± 1.28 0.86 ± 0.08* 6
R33C 5214 ± 560 3.96 ± 2.03 1.24 ± 0.36 7
R33F 4270 ± 791 4.68 ± 1.14 1.03 ± 0.03 7
R33L 2539 ± 418* 5.06 ± 0.70 0.85 ± 0.12* 9
R33Y n.f. n.d. n.d. 11
R33Q n.f. n.d. n.d. 3
R33E n.f. n.d. n.d. 3
n.f.
– non-functional;
n.d.
- not determined.
Alanine Scan of the TM1 Revealed Sites for Ethanol Action
Continuing our alanine scanning mutagenesis strategy, we next investigated how
mutations in the TM1 region affected the inhibitory effects of ethanol on P2X4R function.
We focused on two concentrations of ethanol (10 mM—low, behaviorally relevant; 100
mM—high, pharmacologically challenging). In good agreement with the previously
reported values (Popova et al., 2010; Popova et al., 2013), ethanol significantly inhibited
EC10 ATP-activated currents in WT P2X4Rs (Figure 2A,B). The degree of ethanol
inhibition in WT receptors was 16 ± 3% for 10 mM ethanol and 44 ± 4% for 100 mM
ethanol.
In the tested mutants, we found that the inhibitory effect of ethanol at 10 mM was
significantly increased compared to the WT receptor in R33A and A34W mutants
(Figure 2A). However, the inhibitory effects of ethanol at 100 mM were similar in WT
and most of the mutant receptors, including R33A and A43W (Figure 2A), with the
exception of the W46A mutant. Substitution of tryptophan at 46 with alanine

25

significantly reduced the inhibitory response to 100 mM ethanol (Figure 2A). We have
previously studied the role of position 46 in ethanol action (Popova et al.). For this
study, we chose to focus on arginine at position 33 because of its positive charge,
and its potential for interactions with other residues.

26

Figure 2.2
Effects of 10 mM and 100 mM ethanol in WT and TM1 segment mutant P2X4Rs. (A) A bar graph
of ethanol responses. The response to 10 mM ethanol was significantly increased in the R33A
and A34W mutants’ receptors, and the inhibitory effect to 100 mM ethanol was significant
decreased in the W46A mutant receptor. The data are presented as mean ± SEM, n = 5–18. * p<
0.05 and ** p < 0.001 compared to ethanol effects in the WT P2X4R. (B) Representative ATP-
induced current tracings for the WT and R33A mutant P2X4Rs; the effects of 10 and 100 mM
ethanol are shown.

Physical-Chemical Properties of Residues at Position 33 Determine Receptor
Function and Ethanol Sensitivity
The WT arginine residue at position 33 was mutated to amino acids with different
physical-chemical properties in order to examine whether the polarity and/or size of the
individual side-chains play a role in the 1) agonist properties and/or the 2) ethanol
sensitivity of the receptor.
Receptor Function
Mutating the positively charged bulky residue arginine at position 33 to another
positively charged residue, lysine (K), produced a mutant receptor with Imax currents and
an EC50 value similar to WT P2X4Rs (Table 2). Substitutions with small polar serine or
non-polar residues valine, cysteine (C), leucine (L), or phenylalanine (F) also produced
mutants with WT-like agonist responses (Table 2, Appendix A Figure A1C). Mutating the
positively charged arginine at position 33 to the negatively charged glutamic acid (E) or
non-charged, bulky polar residues, such as glutamine (Q) or tyrosine (Y), yielded very
small Imax currents compared to WT, precluding the determination of their EC50. The total
and surface expression of these mutants were not different compared to the WT P2X4R
(Figure 1), suggesting that the charge, rather than the polarity or size, of the residue at
position 33 is important for the normal receptor function.
Ethanol Response

27

We tested the effects of increasing ethanol concentrations (10, 25, 50, and 100 mM)
of ethanol in position 33 mutants (Figure 3). As reported above, alanine substitution at
position 33 (R33A) produced a mutant receptor that had a significantly greater degree of
inhibition at low 10 mM ethanol concentration compared to WT P2X4Rs. Consistently,
substituting serine (S), another small, polar amino acid, for large arginine (R33S) resulted
in a similar increase in the inhibitory response to 10 mM ethanol compared to the WT
(Figure 3). Further increases in ethanol inhibition in R33A and R33S mutants were also
observed at 25 mM ethanol; however, these effects were not significant. No differences
between the effects of these two mutants and WT receptors were observed at 50 and 100
mM ethanol. The substitution of non-polar amino acids valine (R33V), cysteine (R33C),
leucine (R33L), and phenylalanine (R33F), for the large, positively charged arginine at
position 33 produced mutant receptors with WT-like ethanol sensitivities at all tested
ethanol concentrations (Figure 3). Finally, substituting the positively charged lysine (K), a
residue with similar electrostatic properties as WT arginine (R33K), also resulted in WT-
like ethanol responses. The molecular weight of the residue at position 33 negatively
correlated with the ethanol effect (Pearson’s r = −0.47 and −0.54 for 10 and 25 mM
ethanol, respectively). These data suggested that the size is the main property at position
33 that determines the ethanol sensitivity of P2X4Rs.
Residues That Are Important for Agonist Sensitivity and Ethanol Effects Identified
in Homology Models
We used homology models of the rat P2X4Rs, based on the structures of the
zebrafish P2X4R (Hattori and Gouaux), as a template to visualize the potential
interactions of TM1 segment V49 and R33 in both closed and open configurations (Figure

28

4A,B). As hypothesized, the closed conformation of the P2X4R shows that V49 is located
in a hinge region of an alpha-helical structure, within close proximity to W46 of the same
TM1 segment (Figure 4A). In the open conformation, a shift appears to occur, which
results in these two residues facing one another (Figure 4B). The model also illustrates
that, in the closed conformation, positively charged R33 in the TM1 segment of one
P2X4R subunit and negatively charged residue D354 at the end of the TM2 of the
neighboring subunit are in close proximity (estimated distance at 8 Å), potentially forming
a salt bridge (Figure 4A). The interaction between

Figure 2.3
Effects of ethanol (10–100mM) in mutant receptors at position 33. (A) A bar graph of ethanol
responses. Alanine or serine substitution at position 33 increased the inhibitory effect of ethanol
at low concentrations (10 mM and 25 mM). The increases were significant for 10 mM but not 25
mM ethanol. The data are expressed as mean ± SEM, n = 4–17. * p < 0.05, ** p < 0.005 compared
to ethanol effects in the WT P2X4R. (B) Representative ATP-induced current tracings for the WT,
R33S, and R33K mutant P2X4Rs; the effects of 10 mM ethanol are shown.

29

R33 and D354 seems to be conserved but is weaker in the open conformation, where
both residues are farther apart (10 Å), as illustrated in Figure 4B. We theorized that this
putative salt bridge may be stabilizing the lower end of each subunit during the closed
state and that this interaction is somewhat weakened during the transition from the closed
to the open state, allowing D354 to alter the flow of positively charged ions.
Arginine 33 and Aspartic Acid 354 Interactions
To test the potential interactions between R33 and D354, we reciprocally mutated the
R33 and D354 residues to their putative interaction partner, generating single mutants,
R33D, and D354R, as well as the double reciprocal mutant R33D-D354R. In all mutants,
ATP evoked negligible currents, and while the surface expression of these mutants was
lower, mutant receptors were expressed in oocytes (Table 3; Figure 5A), suggesting that
these mutations hindered receptor activity. To overcome these functional issues, we
adopted a strategy from Silberberg et al. (Silberberg et al.) where WT and non-functional
mutant cRNAs are mixed at an equal ratio and injected into oocytes.
While this mixing strategy resulted in substantially smaller Imax values, we saw no
significant changes in EC50 and/or Hill slopes from those of the WT P2X4R (Table 3,
Appendix A Figure A1D). Surprisingly, the inhibitory effects of ethanol were profoundly
different in these mutant receptors (Figure 5B). When compared to the WT responses,
ethanol inhibition in the range of 10–50 mM was abolished in 354R:WT and 33D:WT
receptor mixes (Figure 5B). One-way ANOVA showed a significant effect of the mutations
on ethanol inhibition across all ethanol concentrations, with significant individual
differences between mutant (354R:WT and 33D:WT mixes) and WT receptors at 25 and
50 mM ethanol. Remarkably, the double reciprocal mutant and WT receptor mix, i.e., 33D-

30

354R:WT, consistently showed ethanol inhibition similar to that of the WT receptor. These
data suggest that the 33D and 354R mutant receptors heteromerized with the WT P2X4R
to form functional channels, and that these mutations likely affect ethanol sensitivity.

Figure 2.4
Homology models of rat P2X4Rs based on the closed (4DW0) and open (4DW1) zebrafish
structures (Hattori and Gouaux). (A) A model illustrating the transmembrane (TM) segments in
the closed state (red solid ribbon of backbone with residues at positions 33, 46, and 49 in the
TM1 segment of one subunit as well as 354 in the TM2 segment of the neighboring subunit
depicted as a ball and stick. Residue 33 of subunit 1 (S1) and 354 of subunit 2 (S2) face each
other on the closed structure. (B) A model illustrating the TM segments in the open state (blue
line ribbon of backbone) with the same residues shown in A. The residues at positions 33 (S1)
and 354 (S2) face away from each other.
Table 3
Imax, EC50, and Hill slope values for the ATP concentration-response curves of WT, position 33
and position 354 reciprocal mutants injected without and with the WT P2X4R. Imax represent peak
currents generated by 100 µM ATP. EC50 and Hill slope values were determined from ATP
concentration-response curves. The data shown are mean ± SEM.
Receptor ATP
Imax, nA EC50, μM Hill slope N
WT P2X4 2886 ± 614 5.34 ± 0.11 1.11 ± 0.08 7
R33D n.f. n.d. n.d. 5
D354R n.f. n.d. n.d. 6
R33D-354R n.f. n.d. n.d. 6
R33D:WT,1:1 763.3 ± 519 5.84 ± 0.11 1.45 ± 0.11 7
D354R:WT, 1:1 560 ± 195 7.24 ± 1.07 1.2 ± 0.95 5
R33D-D354R:WT, 1:1 540 ± 143 4.94 ± 1.06 1.2 ± 0.09 6
n.f.
– non-functional;
n.d.
- not determined.
To further characterize the interaction between R33 and D354, we also tested the
effects of IVM on mutant-WT receptor mixtures. Studies have shown that IVM potentiates

31

the WT P2X4Rs by interacting favorably with the open receptor conformation. If R33
and/or D354 were involved in an interaction that affects the transition between the closed
and open states, the effects of IVM would be significantly augmented for one or both
mutants. The potentiating effect of 3 µM IVM in the 33D:WT mixture was similar to the
effect seen for the WT, while the ATP-evoked response was potentiated approximately
5-fold in 354R:WT mixture (Figure 5C). In contrast, the double reciprocal 33D-354R:WT
mixture responded to IVM similar to the WT receptor (Figure 5C). These results suggest
that D354 plays a role in regulating ion conductance.

32

Figure 2.5
Effects of reciprocal mutations at position 33 and/or 354 as well as the mixtures of reciprocal and
WT on ethanol and ivermectin (IVM) responses. (A) A representative Western Blot showing the
total and surface expression (~60 kDA) for R33D, D354R, and R33D-D354R mutant receptors.
The expression of the mutants was not much different compared to the WT receptor. (B) A bar
graph of the responses to 10–200 mM ethanol for mutant:WT receptor mixtures (at a ratio of 1:1).
The data are expressed as mean ± SEM; * p< 0.05, ** p< 0.01 compared to ethanol effects in the
WT P2X4R. (C) A bar graph of the responses to 3 µM IVM for mutant:WT receptor mixtures. The
data are expressed as mean ± SEM; * p< 0.05, ** p< 0.01 compared to IVM effects in the WT
P2X4R.
Discussion
We previously identified residues at the ectodomain–transmembrane (TM) interface
(aspartic acid at position 331 and methionine at 336) and the TM1 segment (tryptophan
at position 46) of P2X4Rs that play a role in the ethanol sensitivity of the receptor (Popova
et al., 2010; Popova et al., 2013; Prichard et al., 2012). The present study revealed new
residues in the TM1 segment of P2X4Rs that play a role in the receptor function and
resolved a potential interaction between the TM1 and TM2 segments that affects ethanol
sensitivity. Among these new residues, stronger and more consistent changes in Imax,
EC50, and/or Hill slope values were found for V49, F48, and W46 residues. We further
focused on V49 responses to ATP.
Function of valine at position 49 in the upper portion of the TM1 segment
The alanine scan of TM1 and follow-up mutational studies provide insight into the role
of valine at position 49 in receptor function in P2X4Rs. Alanine substitution at this position
caused a loss of function despite no change in surface expression of the mutant receptor.
These findings are in agreement with previous studies, which showed that mutating the
valine residue at position 49 to either alanine or cysteine affected the response to agonist
(Jelínkova et al., 2008).
Previous studies, using circular dichroism and nuclear magnetic resonance
spectroscopy, found that valine, a β-branched residue found in the TM segments of

33

membrane proteins, may contribute to receptor function by providing conformational
flexibility and/or helix destabilization at various stages of the protein activation cycle (Li
and Deber, 1992). Thus, it is possible that in the P2X4R, valine at position 49 provides
the upper portion of the TM1 segment with the flexibility needed for reorganization during
the opening or closing of the channel. Studies on other proteins found that substituting an
alanine for valine caused tighter protein–protein packing and increased α-helical stability
due to the reduction in side-chain volume and strain (Deber et al., 1993). The changes in
flexibility and packing could reflect, in part, changes in the orientation and interaction of
the neighboring aromatic residues produced by substituting alanine for valine. In addition,
previous studies found that the aromatic residues in the upper half of the TM1 segment
near position 49, which includes tryptophan 46 and tryptophan 50, play an important role
in the three-dimensional organization of P2X4Rs (Jindrichova et al., 2009). Therefore,
position 49 could alter the orientation of these nearby aromatic residues and thus affect
the channel function. A shift in the position of V49, noted in the molecular model of the
open conformation of the P2X4R, further supports this hypothesis.
To test whether position 49 was in fact important for the stability of the alpha-helical
structure of the TM1 segment, we mutated the valine to known alpha-helix destabilizing
residues (proline or glycine) and or to an alpha-helix stabilizing residue (leucine) (Li and
Deber, 1992),(Polinsky et al., 1992). Substitutions to alpha-helix destabilizing residues
either reduced (glycine) or eliminated (proline) the agonist response. The presence of an
alpha-stabilizing leucine also resulted in decreased agonist response. Previous studies
have shown that the substitution of V49 with the hydrophobic residue tryptophan had no
effect on ATP-induced currents or IVM potentiation when compared to the WT receptor

34

(Silberberg et al., 2007), which suggests that the effects we see in V49 are not directly
related to P2X4R gating. Valine, most probably, provides a certain level of stability to the
alpha helix of the TM1 segment, and this may be due to its interactions with neighboring
residues, in part from the TM1 segment, such as F48, V47, and W46. It is likely that a
higher or lower level of stability respectively introduced by the mutations to glycine or
leucine may have similarly affected the receptor function towards lower activity, as
reflected in the EC50/Imax values of the corresponding mutants. Collectively, these
findings suggest that position 49 provides flexibility/stability to the upper portion alpha-
helix of the TM1 segment, which is required for the proper function of the P2X4R channel.
Arginine at position 33 in the TM1 segment in ethanol action
Prior studies suggested that there are multiple sites of ethanol action in P2X4Rs. We
identified D331 and M336 in the ectodomain–TM2 interface and W46 in the TM1 segment
as sites of ethanol action and/or modulation in P2X4Rs (Popova et al., 2010),(Popova et
al., 2013). Mutations at positions 331 and 336 decreased the sensitivity of P2X4Rs to a
broad range of ethanol concentrations extending from the behaviorally relevant or
intoxicating (10–50 mM) to potentially toxic concentrations (100–200 mM). In contrast,
mutations at position 46 only affected the sensitivity of the receptor to high ethanol
concentrations (> 50mM). Consistent with previous findings, the results of the current
study suggest that there are additional sites with different sensitivities to ethanol in the
lower region of the TM1 segment of P2X4Rs, i.e., positions 33 and 34. Increased
sensitivity to 10 mM ethanol suggests that these residues may be important for the effects
of the lower, behaviorally relevant ethanol concentrations.

35

Initial alanine substitution studies affected ethanol responses at two low but
behaviorally relevant concentrations, 10 and 25 mM. Therefore, in the present study, we
performed an in-depth investigation of the physical-chemical requirements of position 33
for the ethanol sensitivity of P2X4Rs. We found an inverse relationship between the
molecular size of the amino acid residue at position 33 and ethanol sensitivity. This
response was independent of the polarity of the residue. Despite the fact that we could
not assess the ethanol sensitivities of all tested residues due to the minimal Imax currents
of several of the mutants (i.e., glutamic acid, glutamine, and tyrosine), these findings
support the conclusion that the size of the residue at position 33 is a key factor influencing
the ethanol sensitivity of P2X4R. Our previous findings also suggested a key role for the
molecular size, and not polarity, of the residue at position 46, with the difference that there
was a positive correlation with ethanol inhibition (Popova et al., 2013). In contrast,
polarity, not molecular size, at positions 331 and 336 in the ectodomain–TM2 interface of
P2X4Rs plays a role in ethanol sensitivity (Popova et al., 2010). Together, these findings
suggest that different physical-chemical properties influence the ethanol sensitivity of
P2X4Rs and that these differences may be region specific. Similar regional differences in
the effects of physical-chemical properties on ethanol sensitivity have been reported in
glycine receptors [[Ye, 1998 #249], (Crawford et al., 2007), (Naito et al., 2014; Perkins et
al., 2008), which suggests that this is a general phenomenon that extends across ligand-
gated ion channel super-families.
Interaction between TM1 arginine at position 33 and TM2 aspartic acid at position
354 in channel functioning

36

Studies have shown that the TM1 segment of the P2X4R has a limited contribution
to ion conduction by itself (Haines et al., 2001), (Jiang et al., 2001), (Samways et al.,
2012). However, prior work has also shown a conserved mechanism within the P2XR
family where the TM segments of the receptor subunits rearrange during channel opening
or closing, allowing for residues from different subunits to interact with one another
[[Kawate, 2009 #244](Hattori and Gouaux, 2012), (Silberberg et al., 2007), (Jiang et al.,
2001), (Samways et al., 2012), (Egan et al., 1998). We hypothesized that the positively
charged arginine at position 33 of the TM1 segment may interact with a negatively
charged aspartic acid of the TM2 segment of a neighboring subunit. To visualize this
potential interaction, we built homology models with the zebrafish P2X4R serving as a
template. Both the closed and open structures of the P2X4Rs highlight the close proximity
of arginine (R) at position 33 in the TM1 segment of one P2X4R subunit and aspartic acid
(D) at position 354 in the TM2 subunit of another subunit (Figure 4A), supporting the idea
of a potential interaction between the two residues. We tested this finding by making
reciprocal interaction mutations, i.e., R33D and D354R mutants, however those turned to
be non-functional. We then adopted a strategy previously used by Silberberg et al.
(Silberberg et al., 2005) which produced responses from non-functional mutations via the
incorporation/mixing of WT receptor cRNA, mixed in equal ratios. These functional mutant
receptor mixtures, R33D:WT and D354R:WT, exhibited WT-like agonist properties (as
observed by EC50 and Hill slope values), although the changes seen in the ethanol
sensitivity of these receptor mixtures indicate successful incorporation of mutant subunits
in these functional receptors. The inhibitory effects of sub-100 mM ethanol were abolished
in the single reciprocal mutations, suggesting that the interaction between the lower

37

portions of the TM segments plays a role in the ethanol sensitivity of P2X4Rs. More
importantly, the double mutant R33D-D354R:WT mixture demonstrated WT-like
responses to ethanol over the whole range of concentrations. Interestingly, the physical
site for ethanol activity did not seem to be disturbed by a double-reciprocal mutation of
these residues.
We have previously found that sites for ethanol and IVM partially overlap, i.e., that
IVM antagonizes the inhibitory action of ethanol (Asatryan et al., 2010). Therefore, we
tested whether IVM potentiation would be disturbed in these reciprocal mutant mixtures.
Consistent with the response to ethanol, there was no change in IVM potentiation in the
double reciprocal mutant or the single R33D:WT mutant, which retained WT-like
responses to IVM. Only the reciprocal mutation of position 354 significantly increased IVM
potentiation (seen in D354R:WT receptor mixtures). These results are consistent with
previous studies where the tryptophan scanning mutation of R33 did not affect IVM
potentiation and the tryptophan substitution of D354 was non-functional (Silberberg et al.,
2005). It is unlikely that mutating the TM2 segment residue (D354) creates a
conformational change affecting IVM binding pocket, as the double reciprocal mutation
retains WT-like IVM responses, suggesting a WT-like interaction between the TM1 and
TM2 segment during channel opening.
These findings provide new insights into the importance of interactions between the
lower portions of the TM1 and TM2 segments in P2X4Rs for ethanol action. In
combination with the findings from our molecular models, these results suggest that the
interactions between R33 and D354 residues affect the transition from closed to open
conformation during channel opening. The presence of an aspartic acid residue in TM2

38

is conserved among all P2XR subtypes (Kawate et al., 2009) and has been shown to play
a significant role in ion conductance (Cao et al., 2009). Moreover, recent studies suggest
that the lower part of the TM2 segment changes shape during activation (Kracun et al.,
2010). The systemic probing of the TM2 domain using the substituted cysteine
accessibility method (SCAM) presented evidence that the TM2 domain was the primary
pore-forming part of P2XRs, which is in agreement with the crystal structure (Kawate et
al., 2009), (Hattori and Gouaux, 2012). Using combination of SCAM with a thiol-reactive
probe Cd
2+
, Kracun et al. identified D349 in P2X2Rs that provides movement to the pore-
lining regions of the TM2 during channel opening (Kracun et al., 2010). Interestingly,
P2X2R residue D349 corresponds to D354 in P2X4R (Kracun et al., 2010). As such, it is
plausible that during channel opening, the interaction between R33 and D354 is
weakened, thus allowing D354 to alter the flow of positively charged ions. With these
considerations in mind, it is logical to conclude that non-polar substitutions at position 33
would not change the nature of the interaction with the residue at position 354 and
therefore, these mutants would behave like WT in terms of ion conductance (Imax). In the
case of substitutions of position 33 with negatively charged or polar residues, an
interaction with D354 is altered, causing a loss of receptor function. Collectively, these
results suggest that an interaction between TM1 and TM2 segments of P2X4Rs occurs
in the closed state, which is altered during channel opening, with D354 driving the
conformational change and ion conductance and position 33 contributing to the channel
stability.
Conclusion

39

The findings of the present study provide new insight into the role of residues in the
TM1 segment in receptor activity and ethanol action in P2X4Rs. We demonstrate that
position 49 contributes to the channel function by providing flexibility/stability of the upper
portion of the alpha-helix during channel opening. These findings also suggest that R33
in the lower part of the TM1 segment is involved in ethanol sensitivity at lower,
behaviorally relevant ethanol concentrations. Moreover, interactions between R33 in the
TM1 segment of one subunit and D354 in the TM2 segment of the neighboring subunit
may be important in affecting the channel transition from closed to open conformation
and thus affect ion conduction, as well ethanol sensitivity. These results identify new
residues that are important for ethanol action on P2X4Rs and, in combination with
modeling studies, provide new information for the development of a pharmacophore for
AUD drug discovery.

40

Chapter 3 Cross-talk between P2X and NMDA receptors: a
novel form of neuromodulation
Chapter 3 Abstract
Purinergic receptors (P2XRs) are non-selective, cation channels that are gated by ATP
with multiple subtypes (P2X1-7). Recently the P2X4R subtype become recognized as a
target for the development of drugs to prevent and/or treat alcohol use disorder (AUD).
This hypothesis is derived from genetic, pharmacological and behavioral evidence
reporting an inverse relationship between ethanol (EtOH) intake and P2X4R activity
where a reduction in P2X4R expression leads to an increase in EtOH intake. EtOH also
affects excitatory ionotropic glutamate receptors (NMDARs), where EtOH also leads to a
reduction in cation flux. Both receptors are expressed in areas of the brain associated
with 1) ethanol intake and 2) behaviors that are strongly associated with reward, memory
and addiction (e.g., ventral tegmental area, nucleus accumbens, hippocampus). NMDAR
ethanol sensitivity has been shown to depend on the combination of the GluN2 subunits,
which have distinct physiological and behavioral functions. Much is known regarding the
effects of EtOH on P2XR and NMDARs -- tested individually. However, there is minimal
information regarding P2XR and NMDAR interactions (i.e. “cross-talk” between
receptors) or how interactions of these two receptors affects receptor/ethanol-induced
signaling. We hypothesized that receptor cross-talk plays an important role in modulating
behavioral and/or sensorimotor functions associated with addiction. We tested this
hypothesis by investigating interactions between P2X4Rs or P2X2Rs and NMDARs
expressed in X. laevis oocytes using two-electrode voltage-clamp electrophysiology, site-

41

directed mutagenesis, and biochemical studies. We found that P2XRs and NMDARs
functioned properly (e.g., expected individual channel activity, EC50, etc.) when co-
expressed in oocytes. However, when co-applying agonists, we report a significantly
lower response than the sum of the response to each agonist. Further investigation
showed that, when agonists are applied sequentially, (glutamate followed by ATP, or vice
versa) ATP activity via either P2X2Rs or P2X4Rs appeared to suppress signaling in
NMDARs containing the GluN2A GluN2B, and even the GluN2C subunit. On the other
hand, NMDAR induced signaling did not appear to significantly alter P2X4R activity but
did reduce P2X2R activity. Furthermore, we found that when P2X4R stimulation precedes
NMDAR stimulation, NMDAR responses were significantly reduced in a time-dependent
manner. Taken together, this initial work provides preliminary insight into unrecognized
signaling interactions between P2X4Rs and NMDARs.
Introduction
Ionotropic receptors are ligand-gated ion channels (LGICS) responsible for various
cognitive and physiological processes. These channels can be activated by various
chemical species, such as glutamate, adenosine triphosphate (ATP), and gamma
aminobutyric acid (GABA), with multiple receptor types being found expression in various
cells types, mainly in neurons (Barria and Malinow, 2002). Of these, ionotropic glutamate
receptors are one of the largest and most widely expressed family of excitatory LGICs
found in the CNS. Three different classes of ionotropic glutamate receptors exist,
differentiated by their ability to be stimulated by selective agonists: Kainate, α-amino-3-
hydroxy-5-methyl-4-isoxazolepropionic acid (or AMPA), and N-methyl-D-aspartate
receptors (or NMDARs). NMDARs are heteroteteramers, usually consisting of two

42

obligate GluN1 subunits, and either two GluN2 or two GluN3 subunits. Within the NMDA
type of glutamate receptors, there exist several subtypes of GluN2 (i.e., GluN2A-D), each
with a different cytoplasmic domain, resulting in differences in functional and physiological
activities (Barria and Malinow, 2002; Kopec et al., 2006). Additionally, the effects of
ethanol have been shown to depend on the GluN2 subunit composition of NMDARs
[Mirshahi, 1995 #49], and chronic intermittent exposure to ethanol leads to increased
GluN2A and GluN2B subunit expression [Roberto, 2017 #283].
Studies into the ATP-gated ionotropic subset of the purinergic receptor superfamily,
or P2X receptors (P2XRs), have shown a functional role with respect to
neuroinflammation, pain, and other neurological conditions. Of the members of the P2XR
family (consisting of P2X1-P2X7), studies have demonstrated that P2X2R, P2X4R, and
P2X6R subtypes, are normally expressed on the majority of neurons, often located at the
edge of the postsynaptic density synapses in the peri- and extrasynaptic space (Khakh
and North, 2012). The P2XR subtypes also show significant structural homology, as the
P2X4R crystal structure (PDB: 4DW1) has been used to build P2X2R models. A role for
P2XRs in behavior and cognitive function is beginning to emerge. Early studies have
found that male P2X4R-knockout (P2X4R KO) mice consumed significantly more ethanol
as compared to wildtype controls (Wyatt et al., 2014a). Moreover, ivermectin (IVM) and
moxidectin (MOX), two related drugs that positively potentiate P2X4R function, were
found to reduce ethanol-inhibition in vitro and ethanol intake in vivo (Huynh et al., 2017;
Yardley et al., 2012). With respect to P2X4R KO mice, our group, we reported cognitive-
behavioral deficits, dysregulation of several receptors within various regions of the brain
and aberrant signaling within the mesolimbic pathway (Sheraz Khoja, 2016; Wyatt et al.,

43

2014a). More recently, using an internalization-deficient P2X4R knock-in mice, it had
been demonstrated that increased surface P2X4R in forebrain excitatory regulate anxiety
and memory processes (Bertin et al., 2020). While these studies indicate that P2XRs are
integral in neuronal signaling and cognitive disease states, determining how P2XRs
mediate these effects is challenging, given that neuronal synapses do not rely on ATP
alone as P2X are rarely found at post-synaptic densities, and that direct P2XR stimulation
has only shown modest contribution to excitatory synapses. (Pankratov et al., 1998) (Mori
et al., 2001) (Jo and Schlichter, 1999).
Although evidence supporting the role of P2XRs in synaptic transmission is limited,
there is a larger body of evidence suggesting that one function of P2XRs involves
interaction with other LGICs. This interaction, termed cross-talk, has been shown to
modulate the activity of GABA, nicotinic and 5-HT3 and AMPA receptors, which is
theorized to play a role in synaptic plasticity. For example, direct interactions between
P2X4R and GABAARs in the ventromedial nucleus of the hypothalamus have been shown
to regulate the synaptic strength of GABAergic neurons(Jo et al., 2011), while P2XRs
(including P2X2R and P2X4R) have been shown to reduce AMPA receptor trafficking and
synaptic strength at glutamatergic synapses in the hippocampus (Pougnet et al., 2016;
Pougnet et al., 2014). P2XR have even been implicated in modulation of N-methyl-D-
aspartate receptors (NMDARs), which is notable considering that NMDARs are the most
widely expressed ionotropic glutamate receptor subtype. Presently, the role of P2XRs in
modulating NMDAR activity remains controversial. In support of this hypothesis, previous
work suggested that P2X4Rs play a role in the induction of long-term potentiation (LTP)
via NMDAR modulation (Sim et al., 2006). Additionally, in more recent work P2XRs were

44

reported to down-regulate NMDARs in a calcium-dependent manner (Lalo et al., 2016).
On the other hand, another study has discounted the possibility that calcium influx from
P2XRs alone was responsible for changes in synaptic transmission (Baxter et al., 2011).
In this study, we investigated the interaction between P2X2R or P2X4R and NMDAR.
This was accomplished using two-electrode voltage clamp (TEVC) electrophysiology and
biochemical studies in Xenopus laevis oocytes co-expressing P2XRs and NMDARs. We
found significant forms of cross-talk between P2XR and NMDARs, with subunit-
dependent properties. We use mutagenesis and sequence analyses to delve deeper into
the domains responsible for this interaction. In this latter work, we found evidence for a
common sequence motif that confers NMDAR cross-talk in P2X2Rs and P2X4Rs,
supporting the electrophysiology and biochemical analyses. Overall, the findings from this
work supported the hypothesis that P2XRs play a role in modulating NMDAR function
and provide direct evidence of a novel function for P2XRs.
Materials and Methods
Molecular Biology
Rat GluN receptor subunits were a kind gift from Dr. John Woodward. P2X4R cDNA
was cloned into the pCDNA3.1 vector as previously described (ref). pUNIV backbone was
a gift from Cynthia Czajkowski (Addgene plasmid # 24705 ; http://n2t.net/addgene:24705
; RRID:Addgene_24705) and was modified for subcloning of rat GluN subunits (to
enhance RNA expression.) Mutant receptors were either available from previous studies
(ref) or mutated using the SuperFi PCR kit and transformed into Zymo Mix&Go competent
cells. Single colonies were inoculated into Luria Broth and after 16-20 hours, minipreps
were performed using the ZymoPure miniprep kit. Plasmids were then restriction digested

45

with NotI-HF (New England Biolabs) and purified using the Zymo DNA clean-up kit. All
constructs were sequence verified via sanger sequencing (Genewiz; La Jolla,CA). The
11C peptide was synthesized by Genscript, reconstituted in ultrapure water, and diluted
to 10 mM using HEPES (10 mM; pH 7.2).
Xenopus laevis oocyte injection and electrophysiology
cRNA for experiments were synthesized with the Ambion message machine T7 kit
(ThermoFisher Scientific), purified using the Ambion MegaClear kit (ThermoFisher
Scientific), and injected into Xenopus laevis oocytes (Ecocyte Biosciences, Austin, TX).
Previous studies have reported functional receptors using an injection concentration of
approximately 10 ng of total NMDAR RNA (5 ng of GluN1 and 5 ng of GluN2), 10-20pg
of P2X2R RNA, or 20ng of P2X4R RNA. For P2X4R oocytes studies, 10-20ng per 40nL
of P2X4R RNA , and/or 5-10ng per 40nL of NMDAR RNA, was combined and injected
into each cell. For P2X2R studies, 10-20pg of P2X2R RNA per 40nL and/or 0.25-1ng of
total NMDAR RNA was combined and injected into each cell. Peptide (15 nL) injections
were performed 30 minutes before TEVC recordings. All injections were performed using
an NanoJect III system (Drummond). Recordings were performed 1-3 days after cRNA
injection. Two-electrode voltage clamp recordings were performed using previously
established methods (Davies et al., 2003; Perkins et al., 2009). In brief, oocyte membrane
potentials were clamped at -70 mV using oocyte clamp OC-725C (Warner Instruments,
Hamden, CT), and the oocyte recording chamber was continuously perfused with
Ringer’s solution ± agonist using a Dynamax peristaltic pump (Rainin Instrument Co.,
Emeryville, CA) at 3 ml/min using an 18-gauge polyethylene tube (Becton Dickinson,
Sparks, MD). All perfusion solutions contain a buffer solution consisting of 115mM NaCl,

46

2.5mM KCl, 1.8mM CaCl (or 1.8 mM BaCl to avoid calcium induced current generated by
Ringer’s solution), and 10mM HEPES, with a final pH of 7.2, unless indicated otherwise.
Glutamate and/or ATP were applied for 10 seconds (to reach a peak current response),
unless otherwise stated. A wait time of at least 5 minutes of perfusion buffer occurred
between any agonist applications as to ensure complete washout of agonist. The resulting
currents were filtered at 5kHz and recorded using an analog chart recorder (Linear). All
current values obtained were normalized to the stable response obtained immediately
before agonist/drug applications began.
Data Analysis
Data were obtained from several batches of oocytes from at least three different frogs,
and are expressed as mean ±S.E.M. The effects of co-stimulation are presented as
percentage of the stable currents evoked by ATP and glutamate alone on individual
oocytes. The Prism 8 software suite (GraphPad Software, Inc., San Diego, CA) was used
for data analysis and curve fitting. Statistical analysis was performed using student’s
paired t test, one-way ANOVA followed by a Bonferroni post hoc comparison or Kruskal-
Wallis test, as noted. Significance was set at p< 0.05.
Results
Co-expression of P2XRs and NMDARs does not affect the function of individual
channels
To evaluate the potential functional interaction between P2XR and NMDARs using
two-electrode voltage-clamp (TEVC) we co-expressed both receptor types in Xenopus
laevis oocytes. We performed mRNA injection of P2XRs or NMDARs at previously
reported concentrations, titrating injections until each receptor system produced

47

comparable currents. We then generated 8-point concentration response curves for
oocytes expressing either P2X4Rs or NMDARs (GluN1 and GluN2A or GluN2B or
GluN2C). The EC50 values for P2X4Rs and NMDARs calculated from ATP and gutamate
concentration response curves (Fig. 1a-d, solid lines), were consistent with previously
reported values (Ogata et al., 2006; Toulmé et al., 2006). We then generated an 8-point
ATP or glutamate concentration response curve for oocytes co-expressing both P2XRs
and NMDARs. There were no shifts in concentration response curves when P2X4Rs and
NMDARs were co-expressed (Fig. 1a-d, dotted lines vs solid lines). Consistently, there
were no statistically significant differences in the EC50 values for receptors regardless
those were expressing individually or in combination (Fig.1). These studies demonstrate
that co-expressing both P2XR and NMDARs does not change the function of either
receptor.

(a)
(b)
P2X4
ATP ( M)
Normalized Response (%)
0.01 0.1 1 10 100
0
20
40
60
80
100
EC
50
= 5.6 M
EC
50
= 4.9 M
P2X4R
NMDAR + P2X4R
GluN2A
Glut ( M)
Normalized Response (%)
0.01 0.1 1 10 100
0
20
40
60
80
100
EC
50
= 3.1 M
NMDAR
NMDAR + P2X4R
EC
50
= 3.0 M

48

(c)

(d)
Figure 3.1
Comparison of the concentration-response curves for P2X4Rs or NMDARs expressed
individually (depicted with solid lines) or together (depicted with dotted lines) in Xenopus
oocytes. (a) ATP concentration-response curves. EC 50 values obtained from ATP-
concentration curves of individual P2X4Rs and P2X4Rs co-expressed with NMDARs
were not significantly different. (b-d) Glutamate-concentration response curves. EC 50
values were not statistically significantly different for glutamate-concentration response
curves for individual GluN2A, GluN2B, GluN2C (solid lines) and each NMDAR subtype
co-expressed with P2X4Rs (dotted lines). P2X4Rs and NMDARs were injected at
respectively 20 ng and 10 ng cRNAs. Data represent Mean ± SEM. Statistical analysis
performed using Exact sum-of-squares F-test. (a) p > 0.5; n=9-12; (b) p > 0.5; n=9-12;
(c) p > 0.3; n=9-12; (d) p > 0.5; n=9-12.
Co-activation of P2XRs and NMDARs produces non-additive (inhibitory)
responses
Next, we sought to characterize the effects of activating both receptor types at the
same time (co-activation). In principle, if there is no interaction between two different
receptors, then coactivation should equal the sum of the response of each receptor when
activated individually(Boué-Grabot et al., 2003b; Boué-Grabot et al., 2004b; Jo et al.,
2011). On the other hand, a putative interaction could be identified if receptor coactivation
produces a synergistic (greater than additive) or inhibitory (less than additive) response.
As illustrated in Figure 2A, the simultaneous co-activation of P2X4R and NMDARs
(containing GluN2B subunits) produced a signficantly smaller response (solid line)
GluN2B
Glut ( M)
Normalized Response (%)
0.01 0.1 1 10 100
0
20
40
60
80
100
EC
50
=1.9 M
EC
50
= 2.2 M
NMDAR
NMDAR + P2X4R
GluN2C
Glut ( M)
Normalized Response (%)
0.01 0.1 1 10 100
0
20
40
60
80
100
NMDAR
NMDAR + P2X4R
EC
50
= 1.3 M
EC
50
= 1.2 M

49

compared to the arithmetic sum of each individual agonist responses (dotted line.) Figure
2B shows bar graph representations of the means of actual and predicted peak currents
for co-activation of P2X4Rs and different GluN2-containing NMDARs: when coactivated
with P2X4R, GluN2A produced 73.6 ± 3.1 % (p <0.0001; paired t-test; n = 9), GluN2B
produced 77.7 ± 3.9 % (p<.001; paired t-test; n = 9), and GluN2C produced 82.2 ± 4.7 %
(p<.01; paired t-test; n = 10) of the predicted response, set as 100%. Similar cross-talk
was observed for co-activation of P2X2R and NMDARs (Figure 2C).These results
suggest that an inhibitory interaction occurs between NMDAR and P2XR subtypes.
To investigate the directional nature of this phenomenon, we added the agonists
sequentially; e.g. ATP was applied when the glutamate response got to its maximum
and vice versa – glutamate was applied when ATP response reached its peak. As
shown in Figure 2C application of either ATP or glutamate reduced the subsequent co-
activation response. This suggests that both P2X2Rs and NMDARs inhibit the response
of one another in a reciprocal manner. However, for P2X4Rs and NMDARs, only
P2X4R activation appeared to significantly reduce the coactivation response (Figure 2C,
lower panel). These results suggest that, unlike reciprocal P2X2R-NMDAR cross-talk,
P2X4R-NMDAR cross-talk is unidirectional.

50

(a)

(b)
(c)

(d)
Figure 3.2
The results of P2XR-NMDAR co-activation. (a) Representative tracings from an
individual oocyte co-expressing GluN2B-containing NMDARs and P2X4Rs responding
to glutamate (left), ATP (middle), or glutamate + ATP (right) are shown. The predicted
additive response (dotted lines) is the sum of the individual glutamate and ATP induced
currents. (b) Bar graphs comparing the predicted and actual responses obtained from
co-application of agonists, normalized to the sum of the individual glutamate and ATP
responses for each oocyte. The data are expressed as mean ± SEM; Statistical analysis
performed using paired t-test * p< 0.05, ** p< 0.01, *** p< 0.001, **** p< 0.0001. The
same results are seen using a saturating concentration of 100 mM glutamate (data not
shown).(c) Representative tracings from an individual oocyte co-expressing NMDARs
Glut
+
ATP
Glut
+
ATP
Glut
+
ATP
0
50
100
P2X4+NMDAR
ATP
Glu
ATP
Glu
ATP
Glu
✱✱✱✱ ✱✱✱✱ ✱✱
Peak Current
(% of Predicted)
GluN2A GluN2B GluN2C
Glut
+
ATP
Glut
+
ATP
Glut
+
ATP
0
50
100
P2X2+NMDAR
GluN2A GluN2B GluN2C
ATP
Glu
ATP
Glu
ATP
Glu
✱✱✱✱ ✱✱✱✱ ✱
Peak Current
(% of Predicted)

51

and P2X4Rs (Top) or NMDARs and P2X2Rs (bottom) For sequential activation of
P2X4Rs and NMDARs, application of ATP followed by glutamate only produces what
resembles an ATP response, suggesting that P2X4R activation prevents the subsequent
activation of NMDARs.
P2X4R-NMDAR cross-talk is independent of Ca
2+
influx
Calcium influx has been shown to regulate NMDAR activity, either by mediating
protein interactions (Krupp et al., 1996) or activating downstream modulators such as
protein kinase C (Michailidis et al., 2007; Zheng et al., 1997). To determine whether
P2X4R-NMDAR cross-talk is mediated by calcium influx, we utilized a calcium-free
Ringers’ solution (CfRS) which substitutes barium chloride for the calcium chloride. In the
absence of calcium, co-activation of GluN2A-P2X4R produced 65.5 ±6.5 % (p<0.0001;
paired t-test; n = 10), GluN2B-P2X4R produced 48.2 ±5.4 % (p<.0001; paired t-test; n =
9), and GluN2C 85.7 ± 6.2 % (p<0.0001; paired t-test; n = 28) of the predicted additive
responses (Figure 3A). The actual coactivation responses were statistically significantly
lower than the predicted values and the extents of inhibtion were comaprable to those
obtained in Ca
2+
-containing medium (Figure 2) indicating that the Ca
2+
-influx does not
play a role in the observed inhibitory cross-talk between P2XRs and NMDARs. These
data suggest that P2X4R-mediated cross-talk of NMDARs is independent of Ca
2+
influx.

52

(a)
Figure 3.3
P2X4R-NMDAR cross-talk is independent of calcium. (a) Bar graphs representing the
predicted and actual responses obtained from co-application of agonists, normalized to
the sum of the individual glutamate and ATP responses for each oocyte. For GluN2A,
co-activation produced a statistically lower response than predicted response (p<0.0001;
paired t-test; n = 10). The same result was observed for GluN2B (p<.0001; paired t-test;
n = 9) and GluN2C (p<0.0001; paired t-test; n = 28) coactivation. The data are expressed
as mean; error bars represent SEM; statistical analysis performed using paired t-test, *
p< 0.05, ** p< 0.01, *** p< 0.001, **** p< 0.0001 ;
Recovery of P2X4R inhibition is GluN2-subunit specific
Activating P2X4Rs and NMDARs either simultaneously or sequentially (Figure 2)
suggests that P2X4R-NMDAR cross-talk is a rapidly occurring phenomenon, but does not
indicate whether this interaction is short-lived or sustained. To resolve the duration of this
inhibitory interaction we first obtained stable glutamate-induced currents, followed by
activation of P2X4Rs with ATP, and then applied glutamate again, with 5-minute washout
periods between agonist applications (see Figure 4A). As shown in Figure 4B, the
glutamate response for all GluN2A-C obtained after P2X4R activation were signficantly
Glut
+
ATP
Glut
+
ATP
Glut
+
ATP
0
50
100
ATP
ATP
Glu
ATP
Glu
ATP
Glu
✱✱✱✱ ✱✱✱✱ ✱✱✱✱
GluN2A GluN2B GluN2C
Peak Current
(% of Predicted)
Ca
2+
-free P2X4+NMDAR responses

53

lower than the baseline glutamate response obtained before P2X4R activation (p<0.0001;
One-way ANOVA; n=9-14). Immediately after P2X4R activaiton, we saw a reduction in
GluN2A and GluN2B responses to glutamate; 57.8 ± 6.6 % and 69.1 ± 4.04% of baseline,
respectively. Consistent with the co-activation response, GluN2C showed the lowest
effect-size 5 minutes after P2X4R activation with 87.5 ± 1.9 % of the baseline response
to glutamate. Furthermore, we saw that both recovery (i.e. time to return to baseline) and
degree of inhibition depended on NMDAR subunit composition. For example, despite a
15 minute washout period, NMDARs containing GluN2A or GluN2C produced 64.3 ± 5.0
% and 87.3 ± 3.8 % of the baseline responses to glutamate . On the other hand, NMDARs
containing GluN2B seemed to recover from inhibition more easily, although even 15
minutes after P2X4R activation, responses to glutamate were statistically significantly
inhibited (p<.05) with 87.9 ± 3.1 % baseline response. These results indicate that recovery
from the inhibitory interaction between P2X4Rs NMDARs depends upon the GluN2
subunit.

(a)

54

(b)
Figure 3.4
The time-course for recovery of P2X4R-mediated NMDAR inhibition. (a) Representative
tracing from an individual oocyte co-expressing both NMDARs and P2X4Rs during an
application of glutamate (Glut) or ATP. NMDAR current response by glutamate is stable
and reproducible, before ATP activation of P2X4Rs. NMDAR response to glutamate is
lower after P2X4R activation. (b) Bar graphs representing the glutamate responses of
NMDARs before and after P2X4R activation by ATP. Glutamate responses after P2X4R
activation were significantly lower for 3 NMDAR containing GluN2A-C subunits (* p<
0.05, ** p< 0.01, *** p< 0.001, **** p< 0.0001 , one-way ANOVA with Dunnet’s post hoc
test; n=10-17 oocytes). The time course of glutamate current recovery appears GluN2-
subunit specific. The data are expressed as mean ± SEM. All values were normalized to
the glutamate response obtained immediately before P2X4R stimulation. The same
results are seen using a saturating concentration of 100 mM glutamate (data not shown).
P2XR-NMDAR cross-talk depends on a shared C-terminal motif
Cross-talk studies often identified residues responsible for these interactions on
the intracellular c-terminal domains (CT) of receptors (Boué-Grabot et al., 2003b; Boué-
Grabot et al., 2004b; Emerit et al., 2016; Jo et al., 2011). Given that there is significant
sequence and structural homology between P2X2Rs and P2X4Rs (Khakh and North,
2012), we hypothesized that a shared CT motif between P2X2Rs and P2X4Rs facilitates
NMDAR inhibition. To test this hypothesis, we performed a sequence alignment and
focused on the CT of P2X2Rs and P2X4Rs. As illustrated in Figure 5A, we found nearly
a dozen conserved residues between the P2X2R and P2X4R CT. We theorized that if the
mechanism of cross-talk relies on the residues located in the P2XR CT, expression of the
P2X4R CT (K373-Q388) would interfere with the interaction between P2X2Rs and
NMDARs. In other words, addition of the P2X4R CT should preclude inhibitory P2X2R

55

cross-talk responses. Similarly, the P2X2R CT should also interfere with P2X4R-NMDAR
interactions. As presented in Figure 5B, we found that the P2X2R CT was able to ablate
the interaction between GluN2A containing NMDARs and P2X4Rs, as there was no
statistically signifcant difference (104.95±5.5%; p>0.4, paired t-test; n=7) between the
predicted response and the coactivation response. Conversely, we saw that the P2X4R
CT was able to ablate the interaction between GluN2A containing NMDARs and P2X2Rs,
as there was no statistically signifcant difference (p>0.3, paired t-test; n=10) between the
predicted response and the coactivation response (96.21±3.9%.) As a positive control,
we injected mRNA coding for the P2X2R CT into oocytes coexpressing P2X2Rs and
GluN2A-containing NMDARS, and also saw no statistically signifcant difference (p>0.5,
paired t-test; n=2) between the predicted response and the coactivation response
(105.54±6.21%) These results support the hypothesis that cross-talk between P2XRs and
NMDARs relies on a common CT motif. In subsequent studies, we chose to proceed with
mutagenic experiments on only the P2X4R because the P2X4R CT has been shown to
tolerate mutation (Royle et al., 2002; Toulmé et al., 2006) and because it is shorter than
the P2X2R CT by over 100 amino acid residues.

56

(a)

(b)
Figure 3.5
P2XR CT mediate interactions with NMDARs. (a) Top A representative illustration of a
homotrimeric P2XR. Each P2XR subunit contains an N-terminal intracellular domain, two
transmembrane domains, an extracellular domain, and a CT intracellular domain. Bottom
A sequence alignment of the P2X2R (upper) and P2X4R (lower) intracellular CT
domains. Residues in red are common between the two different P2XRs; (b) Bar graphs
representing the predicted and actual responses obtained from co-application of
glutamate and ATP in oocytes expressing a P2XR CT domain, in combination with
P2XRs and NMDARs. Agonist responses were normalized to the sum of the individual
glutamate and ATP responses for each oocyte. There was no statistically significant
difference between the predicted responses and the actual responses produced
Glut
+
ATP
Glut
+
ATP
Glut
+
ATP
0
50
100
ATP
Glu
ATP
Glu
ATP
Glu
ns ns ns
P2X2+X2CT P2X4+X2CT P2X2+X4CT
Peak Current
(% of Predicted)
P2XR+NMDAR+P2XCT

57

GluN2A-containing NMDARs and P2X4Rs in the presence of the P2X2R CT
(104.951±5.507% , p>0.4, n=7) Similarly, there was no statistically significant difference
between the predicted responses and the actual responses produced GluN2A-
containing NMDARs and P2X2Rs in the presence of the P2X4R CT (96.21±3.946% ,
p>0.3, n=10 ) or the P2X2R CT (105.54±6.21% , p>0.5, n=2 ). The data are expressed
as mean ± SEM. Statistical analysis performed using paired t-test.
Resolving the P2XR-NMDAR cross-talk motif
P2X4R trafficking has been shown to rely on a non-canonical internalization motif
(Y378XXGL382) (Royle et al., 2002; Royle et al., 2005) present on the P2X4R C-terminus.
Interestingly, several key residues of this motif are present in our consensus alignment of
P2X2Rs and P2X4Rs (Figure 5A). As such, we hypothesized that these residues might
be responsible P2X4R cross-talk. Its important to note that previously, GABAR-P2X4R
cross-talk was shown to be independent of this region, relying solely on two c-terminal
residues: Y374 and V375 (Jo et al., 2011). To examine whether residues in the P2X4RCT
were responsible for the interaction with NMDARs, we truncated or replaced the P2X4R
the internalization motif, as illustrated in Figure 6A. We hypothesized that, if residues in
the internalization domain of the P2X4R is responsible for mediating NMDAR cross-talk,
then truncating the P2X4R at residue 377 (P2X4-377TR) or replacing the internalization
domain (YEQGL) of the wildtype P2X4R with a FLAG epitope (P2X4-FlagIn) would
abolish the inhibitory effects of receptor co-stimulation. Similarly, if only the internalization
motif were driving the interaciton with NMDARs, then truncating the P2X4Rs after the
internalization motif, corresponding to residue 382, (P2X4-382TR) would still show
inhibitory coactivation responses (crosstalk.) In support of our hypothesis, Figure 6B
shows that the inhibitory response previously shown (i.e., Figure 2) was no longer present
when coactivating either P2X4-377TRs or P2X4-FlagIns. Furthermore, Supplementary
Figure 1B shows that P2X4-377TR failed to produce the long-lasting inhibition previously
seen by full-length P2X4Rs during sequential activation of each receptor shown in Figure

58

4B. Surprisingly, we did not see any inhibitory P2X4R-NMDAR responses (i.e. cross-talk)
when P2X4Rs were truncated at residue 382, despite the inclusion of the P2X4R
internalization motif in these mutant receptors. Collectively, these results suggest that the
complete P2X4R C-terminus, starting with residue 377, is necessary for an interaction
with NMDARs to occur.

(a)

(b)
Figure 3.6
Glu+ATP 377TR
Glu+ATP FlagIn
Glu+ATP 382TR
Glu+ATP 377TR
Glu+ATP FlagIn
Glu+ATP 382TR
Glu+ATP 377TR
Glu+ATP FlagIn
Glu+ATP 382TR
0
50
100
P2X4R Mutants
ns ns ns
ATP
Glu
ATP
Glu
ATP
Glu
GluN2A GluN2B GluN2C
Peak Current
(% of Predicted)

59

The P2X4R internalization motif and the residues after confer NMDAR crosstalk. (a) An
illustration of the mutations performed on the P2X4R internalization motif compared to
the wildtype P2X4R; (b) Bar graphs representing the predicted and actual responses
obtained from coapplication of glutamate and ATP in oocytes coexpressing different
P2X4R mutants and NMDARs. Agonist responses were normalized to the sum of the
individual glutamate and ATP responses for each oocyte. None of the P2X4R mutations
were statistically significantly different from the predicted responses for any GluN2
subunit. The data are expressed as mean ± SEM. Statistical analysis performed using a
Kruskal-Wallis test.
Disrupting P2XR-NMDAR cross-talk
Sequence alignment of the P2X2R and P2X4R c-terminii indicates that common
residues between the two receptors exists, (Figure 5A) and that those residues
correspond to the last 11 amino acids of the P2X4R C-terminus, YXQXLXXXMXQ. This
sequence is notable, as studies have shown that a small peptide form of these 11 c-
terminal residues (designated 11C; Figure 7A) can modulate P2X4R desensitization and
potentiates P2X4R responses, both in vitro and ex vivo(Jo et al., 2011). We theorized that
if these common P2X2R-P2X4R residues did impart the ability to interact with NMDARs,
then injection of 11C into oocytes coexpressing P2XRs and NMDARs would eliminate or
reduce P2XR-mediated cross-talk. In support of this hypothesis, Figure 7B demonstrates
that injection of the 11C peptide effectively ablated all previously reported inhibitory
responses in both P2X2R-NMDAR and P2X4R-NMDAR co-expressing oocytes (see
Figure 2B and 2D.) These data suggest that residues common to both P2X2R and P2X4R
are responsible for cross-talk with NMDARs.

60

(a)

(b)
Glut
+
ATP
Glut
+
ATP
Glut
+
ATP
0
50
100
P2X4+NMDAR+11C
ATP
Glu
ATP
Glu
ATP
Glu
ns ns ns
GluN2A GluN2B GluN2C
Peak Current
(% of Predicted)

61

(c)
Figure 3.7
11C disrupts P2XR-NMDAR cross-talk: (a) An illustration of the WT P2X4R and the 11C
peptide; (b) Bar graphs representing the predicted and actual responses obtained from
coapplication of glutamate and ATP for oocytes injected with the 11C peptide,
coexpressing P2X4R and NMDARs. (c) Bar graphs representing the predicted and
actual responses obtained from coapplication of glutamate and ATP for oocytes injected
with the 11C peptide, coexpressing P2X2Rs and NMDARs. Agonist responses were
normalized to the sum of the individual glutamate and ATP responses for each oocyte.
No inhibitory currents (i.e. cross-talk) were detected. The data are expressed as mean ±
SEM.
Discussion
P2XR are commonly found at postsynaptic areas and at dendritic spines of
neurons, although P2XR-mediated excitatory synapses are either sparsely reported or
require excessive stimulation ((Lalo et al., 2016; Rubio and Soto, 2001; Sim et al., 2006).
If P2XR receptors do not contribute significantly to excitatory transmission, then what
purpose do they serve in the brain? Evidence from brain slice electrophysiology studies
has suggested that P2XRs serve a regulatory purpose, but providing a detailed answer,
particularly at the molecular level, has proved challenging. ATP acts on multiple targets,
including metabotropic and ionotropic receptors, and the exact composition of neuronal
P2XRs is difficult to determine, as pharmacological tools either lack selectivity or provide
Glut
+
ATP
Glut
+
ATP
Glut
+
ATP
0
50
100
P2X2+NMDAR+11C
ATP
Glu
ATP
Glu
ATP
Glu
ns ns ns
GluN2A GluN2B GluN2C
Peak Current
(% of Predicted)

62

mixed results. It is for these reasons that we chose to investigate the potential interaction
between purinergic receptors (either P2X2Rs or P2X4Rs) and NMDARs (consisting of
GluN2A-C subunits) using Xenopus laevis oocytes. These P2XR subtypes, are widely
expressed in nearly all areas of the brain, and evidence shows that they colocalize with
and regulate AMPA and GABA receptors in the hippocampus,(Boué-Grabot et al., 2004b;
Pougnet et al., 2014). As a model system, X. laevis oocytes allow one to ask more
succinct questions: can P2X2Rs or P2X4Rs regulate the activity of NMDARs?
P2XR modulation of NMDARs
Our results are the first to provide direct evidence for P2XR-NMDAR cross-talk.
We found that, when heterologous expressed in X. laevis oocytes, P2X4Rs can interact
with NMDARs consisting of GluN2A-C subunits (Figure 2C). Furthermore, we observe a
similar phenomenon between P2X2Rs and NMDARs (Figure 2B). Interestingly, this
P2X2R-NMDAR interaction seems to be reciprocal in nature (each receptor inhibits the
other), while the P2X4R interaction is unilateral (i.e. P2X4Rs can inhibit NMDARs, but
NMDARs do not affect P2X4Rs; Figure 2E). These results indicate that P2XR modulation
of NMDARs may be more complicated and robust than early reports (Baxter et al., 2011;
Lalo et al., 2016; Pankratov et al., 2002). However, while X. laevis oocytes are a useful
model for receptor activity, its important to understand the limits of the system in the
context of synaptic activity and P2XR function. Specifically, P2XRs have not been shown
to elicit EPSCs. As such, it’s important to recognize that our reported cross-talk
physiologically represents NMDAR currents and P2XR activity, rather than P2XR current.
Our results indicate that P2XRs serve an important role in modulating NMDAR
activity and provide new context for which to interpret the function of P2XRs, which has

63

remained elusive, if not controversial. Early studies reported that P2XR could contribute
to synaptic transmission, albeit sparsely, and suggested that P2XRs could function as a
“low-frequency filter”, suppressing NMDAR-mediated LTP under weak stimuli. With the
development of P2X4R KO mice came more support for a role for P2XRs in synaptic
plasticity: 1) in the absence of P2X4Rs, hippocampal neurons exhibited reduced LTP
facilitation, 2) Ivermectin, a P2X4r positive allosteric modulator, could increase LTP in
wildtype mice, but not P2X4R KO mice, and 3) an NMDAR inhibitor completely blocked
induction of LTP in wildtype mice, but had no effect on P2X4R KO mice (Sim et al., 2006).
These results lead to the hypothesis that P2XRs in the postsynaptic membrane modulate
NMDAR activity and LTP induction via calcium influx, rather than through synaptic
transmission. However, studies have also demonstrated that P2X4Rs themselves must
contribute to NMDAR modulation at postsynaptic densities, as intracellular administration
of the calcium chelator BAPTA (via recording electrode) could block NMDAR facilitation
in WT mice, but had no effect in P2X4R KO mice. (Baxter et al., 2011). Surprisingly, our
results are also consistent with this hypothesis, as we found that calcium-free Ringer’s
solution did not preclude P2X4-NMDAR cross-talk during receptor coactivation (Figure
3A). How can P2XRs play a neuromodulatory role with these opposing results? We argue
that P2XR modulation is multifaceted, containing a receptor and calcium component.
Until recently, the modulatory nature of P2XRs (including regulation of NMDARs)
was inextricably linked to P2XRs and their ability to permeate calcium. However, with the
characterization of a novel knock-in mouse strain, where P2X4R is fluorescently labeled
and internalization-deficient (P2X4mCherryIn), a more nuanced role for P2XRs has
emerged (Bertin et al., 2020). This study demonstrated that in CA1 synapses,

64

P2X4mCherryIn mice displayed no changes in basal excitatory transmission, affected
LTP and LTD induction. Considering that, in CA1 hippocampus neurons, LTP and LTD
have been shown to rely on post synaptic NMDARs (Mizuno et al., 2001)(for review on
LTP see(Huganir and Nicoll, 2013)), these results suggest that increased P2X4R activity
in CA1 neurons alters NMDAR function, supporting the idea that P2X4Rs are involved in
regulating synaptic plasticity.
P2XR intracellular domains mediate NMDAR cross-talk
Studies have shown that P2X4Rs and P2X2Rs both interact with 1) GABARs
physically (Boué-Grabot et al., 2004a) 2) AMPARs in a calcium-dependent manner
(Pougnet et al., 2016), and 3) 5-HT3A receptors physically (Boué-Grabot et al., 2003a)
and subcellularly (Emerit et al., 2016), although only the physical interaction between
P2X2R and 5-HT3A was shown to be mediated by the P2X2R c-terminus. P2X2Rs are
similar in sequence and structure to P2X4Rs, with the most prominent differences being
found between the intracellular c-terminus domains (Figure 3?). Furthermore, P2XR-
mediated regulation of other ionotropic receptors has been found to rely on 1) physical
protein-protein interactions, 2) receptor co-trafficking, or 3) signaling cascades, which can
all be found within the receptor c-terminus (i.e. intracellular; see Figure 3 diagram) region.
In an effort to better understand the receptor domains that drive this interaction, we chose
to focus on the P2X4R c-terminus, which is significantly shorter and has been shown to
tolerate mutagenesis (Royle et al., 2002; Toulmé et al., 2006). Our studies identified that
truncation of P2X4Rs before the start of their internalization motif ablated NMDAR cross-
talk, which leads one to believe that this interaction might be dependent upon P2X4R
trafficking. Replacing only the internalization motif (YEQGL) with a flag epitope

65

(DYKDDDKD) also ablated cross-talk, which confirms the significance of these residues.
However, the P2X4R internalization motif does not seem to mediate this interaction
entirely, as P2X4Rs truncated after the internalization motif also fail to interact with
NMDARs (Figure 4). It is important to note that this non-canonical tyrosine-based sorting
motif YXXGL has been well characterized (Royle et al., 2002; Royle et al., 2005) (Toulmé
et al., 2006), but was not previously shown to be responsible for the cross-talk interaction
between P2X4Rs and GABARs (Jo et al., 2011)
While this resolves the interaction between P2X4Rs and NMDARs, do P2X2Rs
interact with NMDARs via a common or similar mechanism? Sequence alignment of
P2X2R and P2X4R c-termini revealed multiple consensus sites, several of which span
residues 378-388 of P2X4R. Based on the P2X4R truncation results, we reasoned that,
if the P2X2R and P2X4R c-termini did share residues that mediated an interaction with
NMDARs, co-expressing the P2X4R c-terminus in oocytes would interfere with P2X2R-
NMDAR cross-talk, and vice versa. As shown in Figure 5B, our results confirmed that
residues in P2X c-terminus mediate NMDAR cross-talk, as each c-terminus disrupted the
inhibitory responses we previously reported (see Figure 2). These results also supported
the hypothesis that consensus c-terminal residues (Figure 3, residues in red) might be
mediating the interaction between P2XR and NMDARs. It is important to recognize that
previous P2X4R-GABAR cross-talk studies by Jo et al found that, while the P2X4R
internalization motif did not mediate GABAR cross-talk (Jo et al., 2011), a peptide
containing the 11 P2X4R c-terminal residues (11C) could potentiate P2X4R activity. Since
11C contains the internalization motif, we wanted to see if the P2X4R c-terminus could
be used as a tool to modulate P2XR-NMDAR cross-talk. We synthesized and injected the

66

11C peptide as previously reported (Jo et al., 2011), and found that injection of 11C
ablated all P2XR-mediated NMDAR cross-talk (Figure 6) suggesting that this interaction
relies on several key residues found on the intracellular domains of P2XRs. Further
studies will determine which specific residues mediate the P2XR and NMDAR cross-talk
by using a point mutation approach.
How could the c-termini of both P2X4Rs and P2X2Rs mediate inhibition of NMDAR
receptors, given the size difference of these domains? Studies on the P2X4R c-terminus
and its non-canonical internalization motif revealed that, when co-crystalized with the u2
AP subunit, residues 374 to 380 do not adopt a rigid structure (Royle et al., 2005).
Furthermore, the non-canonical YXXGL motif functions at the same canonical YXXΦ site
due to the flexibility imparted by the glycine residue. When looking at the composition of
the P2X2R CT, proline and other hydrophobic residues are prevalent. This is notable
because proline can disrupt secondary protein structures and limits flexibility, while
hydrophobic residues can promote a more “buried” conformation. It is possible that these
residues allow the P2X2R c-terminus to adopt a conformation that favors an interaction
with NMDARs, like P2X4Rs, but prevents P2X2Rs from functionally uncoupling from the
multi-protein complex, hence the reciprocal nature of this interaction. Unfortunately, no
structural information exists for any P2XR c-termini, which would provide more insight
into their function.
Resolving the role of P2XRs in the brain
Distinct forms of P2XR cross-talk might serve discrete regulatory functions and
arise from P2XR mobility and localization, which has been shown to be subunit
dependent. For example, P2X2Rs are highly mobile and stable at the cell surface, but

67

rarely found on synaptic densities (Bobanovic et al., 2002; Richler et al., 2011). On the
other hand, P2X4Rs are primarily found within intracellular compartments (due to
constitutive internalization). Evidence has already shown that P2X4Rs play a role in
integrating ATP signaling from astrocytes in the tripartite synapse specifically by inhibiting
GluN2B-containing NMDARs, an interaction that involves a multiprotein complex (Lalo et
al., 2016). As such, P2X2Rs at extra synaptic densities may serve as molecular “trap”,
inhibiting NMDARs via an interaction that prevents their inclusion into the postsynaptic
densities. The reciprocal nature of this interaction might act as a negative feedback loop
and allow for more diverse responses or fine tuning. In contrast, P2X4Rs can act in a
more targeted manner, waiting inside the cell and mobilizing into the postsynaptic density
when stimulated.

68

Chapter 4 Ethanol antagonizes P2X4 receptors in ventral
tegmental area neurons^
Rodriguez, Larry; You, Chang; Brodie, Mark; Asatryan, Liana; and Davies, Daryl. L. ;
Ethanol antagonizes P2X4 receptors in ventral tegmental area neuron. Neuroreports (2020,
accepted)
^ Note: This chapter is taken from the previously mentioned publication.
Chapter 4 Abstract
P2X4 receptors (P2X4Rs) are found throughout the central nervous system, and studies have
shown that these purinergic receptors are important regulators of alcohol intake. The ventral
tegmental area is an important region for the rewarding and reinforcing properties of alcohol, but
the role of P2X4R in this region is unknown. Using both immunohistochemical and
electrophysiological methods, we examined the interaction between P2X4Rs and alcohol on VTA
neurons. Incubation of brain slices containing the VTA for two hours with siRNA targeting P2X4Rs
resulted in about a 25% reduction in P2X4R immunoreactivity in tyrosine hydroxylase positive
VTA neurons. In electrophysiological experiments, ATP (0.5-3 mM) produced a reduction in
spontaneous firing rate, and ethanol significantly reduced this inhibition. Exposure to siP2X4R for
two hours via the recording micropipette resulted in a suppression of the response of VTA neurons
to ATP but no significant reduction in ethanol-inhibition of the ATP response was observed after
this P2X4R downregulation. These results support the idea that VTA neurons are inhibited by
ATP, that ethanol antagonizes this inhibition, and that the ethanol-sensitive component of ATP
inhibition is mediated by P2X4Rs. This interaction of ethanol with P2X4Rs may be an important
regulator of the rewarding effects of ethanol, making P2X4Rs an intriguing target for the
development of agents to treat alcohol use disorders.

69

Introduction
There have been limited studies of the role of purinergic receptors, including
P2X4R receptors (P2X4Rs), in brain areas related to drug and alcohol abuse. The fact
that alcohol inhibits these receptors (Ostrovskaya et al., 2011) makes them an intriguing
target for investigation. Interactions between alcohol and P2X4Rs in the central nervous
system may play an important role in the mediation of alcohol-related behaviors and
neurochemical effects of ethanol (Asatryan et al., 2011; Franklin et al., 2014; Wyatt et al.,
2014b). Specifically, alcohol drinking is altered in P2X4R knockout mice (Wyatt et al.,
2014b), and positive modulators of P2X4Rs reduce alcohol intake (Khoja et al., 2018).
Examination of the mechanisms of the effect of P2X4Rs on alcohol drinking may be useful
in the search for novel agents to treat alcoholism.
The fact that ATP, an agonist at purinergic receptors, is co-released with
monoamine neurotransmitters like dopamine from synaptic vesicles (Szalay et al., 1998)
suggests that studies of ATP-ethanol interactions on purinergic receptors in monoamine-
containing brain areas like the ventral tegmental area (VTA) is relevant. The VTA supplies
dopamine to a number of brain areas that mediate the effects of alcohol on reward and
reinforcement processes (Koob and Volkow, 2016; Wise, 1996; Wise and Morales, 2010).
Projections from the VTA to the nucleus accumbens , amygdala and prefrontal cortex are
crucial for the development of addiction disorders, including alcoholism (Koob, 2003;
Oliva and Wanat, 2016). Evidence from decades of research link VTA dopamine release
and the administration of alcohol (Morikawa and Morrisett, 2010). Pharmacological
interference with dopamine neurotransmission has been shown to reduce alcohol intake
(Samson et al., 1993; Samson et al., 1992). Ethanol-induced place-preference is

70

dependent on the ventral tegmental area (Bechtholt and Cunningham, 2005). In addition,
rats will self-administer alcohol directly into the VTA (Rodd et al., 2008). Examination of
agents that act in the VTA to alter the actions of alcohol may lead to identification of new
mechanisms by which alcohol intake can be regulated.
While selective positive modulators and antagonists of purinergic receptors are
available (Franklin et al., 2014), all current agents, like most pharmacological agents,
have off-target effects that may confound interpretation of the results. While P2X4R
knockout mice are useful, developmental compensation in global or conditional knockout
mice may add complexity to the interpretation of the results. We have developed a novel
method of specific downregulation of gene expression using in vitro administration of
interference RNA via the recording micropipette during brain slice electrophysiology
(Nimitvilai et al., 2013a). Here, we demonstrate that ethanol significantly antagonizes the
inhibitory response to bath-applied ATP in VTA neurons, and furthermore that
downregulation of P2X4Rs using siRNA reduces that antagonism. These results implicate
P2X4Rs in the ethanol modulation of purinergic responses in the VTA.
Materials and Methods
Animals
Male C57BL/6 J mice (C57) were purchased from the Jackson Laboratories (Bar
Harbor, ME). Mice were 4–6 weeks old on arrival. All mice were treated in strict
accordance with the NIH Guide for the Care and Use of Laboratory Animals, and all
experimental methods were approved by the Animal Care Committee of the University of
Illinois at Chicago.
Electrophysiological Recording

71

Preparation of brain slices containing the ventral tegmental area (VTA) has been
described previously (You et al., 2019). Following rapid removal of the brain, the tissue
was blocked coronally to contain the VTA and submerged in chilled cutting solution and
coronal sections (250 μm thick) were cut using a vibratome, and then placed in the
recording chamber submerged in artificial cerebrospinal fluid (aCSF) maintained at a flow
rate of 2 ml/min and a temperature of 35°C. The composition of the aCSF in these
experiments was (in mM): NaCl 126, KCl 2.5, NaH2PO4 1.24, CaCl2 2.4, MgSO4 1.3,
NaHCO3 26, glucose 11. The composition of the cutting solution in the vibratome was (in
mM): KCl 2.5, CaCl2 2.4, MgSO4 1.3, NaHCO3 26, glucose 11, and sucrose 220. Both
solutions were saturated with 95% O2/5% CO2 (pH = 7.4). Recordings were done at least
1 hour after cutting to allow equilibration.
During recording, electrodes were placed in the VTA under visual control under
low power light microscopy. In this method, we did not use infrared or other methods to
visualize the cells that were recorded. Only those neurons which were anatomically
located within the lateral VTA and which conformed to the electrophysiological criteria for
pDAergic neurons established in the literature and in this laboratory (Calabresi et al.,
1989; Mueller and Brodie, 1989) were studied. At the beginning of the day, caffeine (1
µM) was added to the superfusate to block adenosine receptors. At the end of recording,
whenever possible, cells were tested with baclofen to confirm that the VTA neurons from
which we recorded were dopaminergic (Margolis et al., 2006). Extracellular recording
electrodes were made from 1.5 mm diameter glass tubing with filament. Tip resistance of
the microelectrodes ranged from 2 to 4 MΩ. A Fintronics amplifier was used in conjunction
with an IBM-PC-based data acquisition system (ADInstruments, Inc.). Offline analysis

72

was used to calculate, display, and store the frequency of firing 1 min intervals. Firing rate
was determined before and during drug application. The change in firing rate is expressed
as a percentage of the initial firing rate, which controls for small changes in firing rate
which may occur over time.
Ethanol (10mM-120mM), caffeine (1 µM), ATP (200 - 3000 µM), Baclofen (100nM)
were added from calibrated infusion pumps from stock solutions 100–1000 times the
desired final concentrations and mix and diluted completely with aCSF before this mixture
reached the recording chamber. Final concentrations were calculated from aCSF flow
rate, pump infusion rate, and concentration of drug stock solution. The small volume
chamber (about 300 μl) used in these studies permitted the rapid application and washout
of drug solutions. Typically, drugs reach equilibrium in the tissue after 2–3 min of
application. siRNA against P2X4R was delivered through microelectrode filling solution.
Briefly, 5nmol siP2X4R (Dharmacon) was mixed with 1.5 n-Fect transfection reagent
(Neuromics), and then diluted 1:10 with n-Fect DNA Diluent (Neuromics). This solution is
further mixed 1:1 with normal saline to make final microelectrode filling solution. The
method of microelectrode delivery of RNA interference and pharmacological antagonists
has been described previously (Mameli et al., 2007; Nimitvilai et al., 2013a; Nimitvilai et
al., 2013b; Pesavento et al., 2000) with the advantage of more localized drug application
and reduced expense.
Thick section Immunofluorescence
Thick (250 µm) section immunofluorescence staining was adapted from Brain
BLAQ method (Kupferschmidt et al., 2015). Briefly, 250 µm brain slices containing VTA
were cut similarly to those used for electrophysiological recording. To prepare siRNA

73

stock to treat slices for immunohistochemistry, 5nmol siP2X4R (Dharmacon) was mixed
with 1.5 n-Fect transfection reagent (Neuromics), and then diluted 1:10 with n-Fect DNA
Diluent (Neuromics). The slices were then immediately submerged and incubated for 2
hours in artificial cerebrospinal fluid (aCSF as described above) containing either siRNA
(1:10 siRNA stock in aCSF) or DNA Diluent (control: 1:10 in aCSF) and maintained at a
temperature of 35°C and saturated with 95% O2/5% CO2 (pH = 7.4). At the end of 2
hours, slices were transferred into chilled 4% PFA and stored in 4 °C overnight.
At the beginning of staining, slices were washed with phosphate buffered saline
with triton (PBST;0.2% triton) at room temperature for 1 hour, followed by 2 washes with
distilled water (diH2O) for 2 minutes each. Then slices were incubated with sodium
borohydride NaBH4 (5mg/ml) in diH2O twice with 10 min each to quench aldehyde-
induced catecholamine fluorescence and reduce free aldehyde groups. After rinsing with
diH2O twice for 1 minute each time, slices were then incubated with Sudan Black B
solution (0.2% in 70% ethanol) for 30 min, to block myelin fluorescent. After washing two
times with PBS for 30 minutes each, the tissue was incubated in PBST with BSA for 4
hours before incubation with primary antibodies (P2X4: Alomone Labs 1:200, TH:
Millipore 1:1000) for 72 hours at 4 °C. Washing after primary antibody incubation was
done 4 times within 24 hours. Then slices were incubated with secondary antibodies for
48 hours in dark at 4 °C. Washing after secondary antibody incubation was done 4 times
within 24 hours, followed by 1-hour wash with PBS, before mounting on slides.
Images were taken by confocal LSM 710 (Zeiss, Thornwood, NY). For
quantification of knockdown of P2X4Rs in the VTA after by siP2X4R, the intensity of
P2X4Rs was quantified and normalized to the intensity of TH staining in the VTA using

74

the National Institutes of Health ImageJ software. The intensity of P2X4R and TH was
quantified by averaging the fluorescent intensity of 3 TH+ cells per picture, 2-4 pictures
per section, 2 sections per mouse and 3 mice per group; 14-15 pictures per group were
summarized.
Results
Reduction of P2X4Rs after incubation with siRNA
We first determined whether 2-hour incubation of brain slices containing the VTA
with siRNA targeting P2X4Rs significantly altered the expression of P2X4Rs. As noted
above, brain slices of the same thickness as those used for electrophysiology were
incubated at 35 °C for two hours with either control solution or solution with siRNA
targeting P2X4. Figure 1A illustrates pictures comparing the VTA from slices incubated
with control solution (top) or siRNA targeting P2X4. Immunohistochemistry labeling of
tyrosine hydroxylase (TH) was used to control for differences in tissue viability.
Expression of P2X4R was normalized as a proportion of TH immunoreactivity. The mean
level of relative P2X4R immunoreactivity (84.1 ± 0.06 % of TH expression in control slices,
n= 14) was reduced to 63.2 ± 0.05 % (n=15) after incubation with siP2X4R (Figure 1B).
This difference was statistically significant (two-sample t-test, t = 2.623, DF=27, P < 0.02),
and indicates that 2-hour exposure to siP2X4R decreased P2X4R immunoreactivity.

75

Figure 4.1
A: Images of P2X4R Immunoreactivity in the VTA of C57/BL6J mice. Immunohistochemistry
was performed using antibodies for P2X4RS and tyrosine hydroxylase (TH). In 250 µm
sections containing the ventral tegmental area from C57BL/6J mice were cut with the same
procedure as used for electrophysiological recording, and sections were then incubated with
ACSF containing either P2X4R siRNA or DNA Diluent (control) at 35°C and saturated with
95% O2/5% CO2 (pH = 7.4) for 2 hours. After incubation, sections were fixed with chilled 4%
PFA and then proceed for immunohistochemistry. Left panel: Red-stained neurons are TH-
positive. Middle panel: Green stained neurons are P2X4-positive. Right panel: Merging the
images of TH and P2X4R immunoreactivities. B: Quantitative assessment of P2X4R
knockdown by siP2X4R in the VTA of C57/BL6J mice . Immunoreactivity of P2X4R was
quantified and normalized to that of the TH, using Image J. The intensity of P2X4R and TH
from 3 TH-positive cells per picture was measured and averaged, with n=14-15 (3 mice per
group, 2 sections per mouse, 2-4 pictures per section) pictures were taken from each group.
Relative expression of P2X4RS (normalized to TH) was significantly decreased in sections
that were incubated with siP2X4R, compared to control. (P<0.02 two sample t-test)

76

VTA neurons were significantly inhibited by ATP.
In order to assess an important correlate of these immunohistochemical results,
we assessed the effect of 2-hour exposure to P2X4R siRNA on electrophysiological
response to ATP. VTA neurons were recorded from brain slices with electrodes
containing siP2X4. Initial responses to ATP alone and to ATP in the presence of ethanol
(40mM) were assessed and compared to responses to the same concentration of ATP
and ATP plus ethanol two hours later. As the responses were quite variable, we adjusted
the ATP concentration for each cell; the mean ATP concentration for Control cells was
2.1 ± 0.6 mM (n=4), and the concentration for siP2X4-treated cells was 1.21 ± 0.5 mM;
these values were not statistically different (two-sample t-test, t=1.10, DF=9, P > 0.05).
As immunohistochemical results shown in Figure 1 above indicate that P2X4R expression
was reduced after 2-hour exposure to siP2X4, we examined whether there were also
functional changes in the response to ATP. An example of the experimental design for
the electrophysiology experiments is shown in Figure 2A. In this recording, 800 mM ATP
produced a large inhibition of firing rate (-95.8%), and the magnitude of this inhibition was
reduced to -35.1% in the presence of 40 mM ethanol. After two hours, the response to
ATP was reduced to -42.0%, whereas the response to ATP in the presence of ethanol
was reduced to -27.4%. The mean results shown in Figure 2B and C indicate a similar
pattern of responses for the pool of cells examined. Figure 2B illustrates the mean
responses to ATP in the absence of ethanol. The initial response to ATP in the control
cells (n=4) was -63.3 ± 12.3% and in the siP2X4R cells (n=7) was -82.0 ± 9.5%; after 2
hours, ATP inhibited the firing rate of control cells by 66.4 ± 14.8%, but the siP2X4R
exposed cells were only inhibited by 39.5 ± 8.9%. There was a statistically-significant

77

difference among the groups (One-way ANOVA, F3,21=3.32, P<0.05), and post-hoc
comparison of means indicated there was a significant difference between the initial ATP
response and the 2-hour ATP response of the siP2X4R cells (Tukey test, P<0.05 for that
comparison; P>0.05 for all other comparisons). Therefore, the response of VTA neurons
to ATP is significantly reduced after 2-hour exposure to P2X4R siRNA via the recording
pipette, consistent with the reduction in P2X4R immunoreactivity when the whole slice
was incubated with P2X4R siRNA.
Figure 2C illustrates the magnitude of the responses to ATP in the presence of 40
mM ethanol. Note that the initial responses to ATP were reduced in comparison to the
initial responses in the absence of ethanol (compare to Figure 2B), in line with our
observation that ethanol decreased the response to ATP. In the presence of ethanol, the
initial response to ATP in the control cells (n=3) was -34.2 ± 9.2% and in the siP2X4R
cells (n=7) was -21.7 ± 3.3%. After 2 hours, in the presence of ethanol ATP inhibited the
firing rate of control cells by 27.2 ± 14.6%, and the siP2X4R exposed cells were inhibited
by 18.0 ± 3.6%. There was a no statistically significant difference among the groups (One-
way ANOVA, F3,19=1.17, P>0.05). The reduction in ATP response by P2X4R
downregulation (Figure 2B) is not observed when ethanol is present (Figure 2C),
indicating that ethanol inhibits P2X4Rs. The inhibition seen in the presence of ethanol,
therefore, is mediated by non-P2X4Rs.

78

Figure 4.2
A: Representative ratemeter for electrophysiological recordings. A typical rate meter represents the
experimental protocol used for the electrophysiological recording experiments. After baseline firing was
stable (4~5 min after getting a cell), ATP dose was tested until a dose that produces 50%~100% inhibition
on cell firing (in the case of the example on the rate meter the dose was 800 µm). For initial test, ATP was
applied for 7 minutes then after washing for 10 minutes, 40mM Ethanol was applied for 10 minutes. With
Ethanol’s continuing present, apply the same dose of ATP for 7 minutes before wash. The identical protocol
as above was applied a second time at 2 hour point of the recoding, to allow siRNA (siP2X4R as shown on
rate meter example, or control) be delivered to the cell. Results of ATP response with or without ethanol

79

present was compared between initial test and test at 2 hour point. B: Effect of siP2X4R on ATP Inhibition.
The ATP response of VTA neurons was recorded at the beginning of recoding and after 2 hours to allow
delivery of siRNA (siP2X4, n=7 cells from 6 mice; or control, n=4 cells from 4 mice). C: Effect of siP2X4R
on Ethanol and ATP interaction. The ATP response of VTA neurons with presents of 40mM ethanol was
recorded at the beginning of recoding and after 2 hours to allow delivery of siRNA (siP2X4, n=7 cells from
6 mice; or control, n=4 cells from 4 mice).
Discussion
The role of P2X4Rs in the ventral tegmental area, and the interaction of those
receptors with ethanol, was examined in the current study using siRNA and
electrophysiology. The findings reinforce the idea of the sensitivity of P2X4Rs to ethanol
in a brain area important for reward and reinforcement. The response of VTA neurons to
ATP was a decrease in spontaneous firing rate. In the presence of ethanol (40 mM), there
was a decrease in the ATP response. With a 2-hour exposure to siRNA targeting P2X4Rs,
there was a significant decrease in the response to ATP which is correlated with
decreased P2X4R expression after 2-hour incubation with siP2X4. Interestingly, the
ethanol-induced reduction in the response to ATP did not occur following the reduction in
P2X4Rs. Taken together it appears that the VTA is inhibited by the action of ATP at two
distinct sites: P2X4Rs and at least one other receptor. Ethanol appears to antagonize the
effect of ATP on P2X4Rs, as reduction in the level of P2X4Rs did not alter the ethanol-
induced decrease in response to ATP (Figure 2C), even though the response to ATP in
the absence of ethanol was decreased by about half by siRNA (Figure 2B). Dopamine
system interaction with P2X4R systems was demonstrated previously; ivermectin, a
P2X4R positive modulator disrupts prepulse inhibition of acoustic startle reflex in a
dopamine-receptor dependent manner (Khoja et al., 2019). In that study, it was shown
that activation of P2X4Rs was associated with increased dopamine neurotransmission
and disrupted information processing through D1 and D2 mechanisms (Khoja et al.,

80

2019). Ethanol-P2X4R interactions could interfere with dopamine-dependent information
processing as well.
Studies performed thus far support the importance of P2X4Rs in the regulation of
alcohol intake. Ivermectin, a broad-spectrum anthelmintic that has affinity for P2X4Rs,
interfered with ethanol-induced inhibition of P2X4Rs (Yardley et al., 2012). Avermectins
that interfere with ethanol action on P2X4R signaling also reduce alcohol intake (Asatryan
et al., 2014a). The roles of P2X4Rs also have been studied using a P2rx4 knockout
mouse model. Knockout of P2X4Rs transiently increases ethanol intake during 24 hour
or limited access paradigms (Wyatt et al., 2014b). These results strongly support the role
of P2X4Rs in regulation of drinking behavior. The present study supports the idea that
one of the brain regions that may be important for ethanol-P2X4R interactions is the VTA.
We have previously demonstrated a downregulation of Gq function after 2-hour
exposure to shRNA that was incorporated in the recording micropipette (Nimitvilai et al.,
2013a). Other groups have shown a decrease in metabotropic glutamate receptor
function following delivery of siRNA via whole-cell patch clamp recording electrodes
(Mameli et al., 2007).In the current study, we used siRNA with a lipid transfection agent
(n-Fect) rather than a viral vector. While it might be possible to identify the precise cells
that receive the siRNA, numerous technical difficulties made it impossible to perform
accurate identification of those cells following electrophysiology. One difficulty was that
the histochemical methods needed to perform immunohistochemistry in 250 µm thick
brain slices appeared to eliminate dyes incorporated in the pipette to label the cells in the
vicinity of the recording micropipette. As an alternative strategy to demonstrate
downregulation of protein by siRNA in vitro, we show here that 2-hour exposure of the

81

entire brain slice to siRNA in n-Fect results in a decrease in P2X4R immunohistochemical
labeling. This demonstrates that there is sufficient synthesis and turnover of P2X4R
protein in the brain slice preparation to alter the immunolabeling of these receptors within
a relatively short time frame. While the exposure to siRNA via the recording pipette could
be more or less effective than siRNA incubation in a static chamber, the
electrophysiological results shown above indicate that such siRNA exposure can reduce
P2X4R function, as indicated by the reduced inhibitory response to ATP.
Interestingly, after the 2-hour exposure to siP2X4, the ATP response was not
further reduced by 40 mM ethanol. This finding is important in that it indicates that the
ethanol-sensitive ATP response is driven by P2X4Rs. This is important confirmation of
observations that P2X4Rs are sensitive to inhibition by ethanol (Ostrovskaya et al., 2011).
Furthermore, it supports the idea that P2X4Rs can influence the activity of VTA neurons,
and that this influence is reduced by ethanol.
VTA neurons are affected by numerous neurotransmitter systems, including dopamine,
GABA and glutamate(Oliva and Wanat, 2016), and numerous peptides, including
substance P (Xia et al., 2010), corticotropin release factor (CRF)(Wise and Morales,
2010) and cholecystokinin (Brodie and Dunwiddie, 1987). Purinergic control of VTA
neurons adds another possible neurotransmitter system that can be exploited for the
development of agents to control ethanol actions, as the influence of purinergic inhibition
by P2X4Rs is sensitive to ethanol. Additional studies will be needed to establish the
importance of ethanol-induced reduction in P2X4R function to the rewarding effects of
ethanol in vivo, and whether chronic alcohol exposure produces crucial changes in the
interaction between ethanol and P2X4Rs on dopamine neurons.

82

Chapter 5 Conclusion
Summary of overall findings
P2X4 receptors (P2X4Rs) have emerged as promising targets for the treatment of
alcohol use disorder (AUD). However, while transgenic mouse models provide evidence
that P2X4Rs play a key role in alcohol consumption, the underlying process by which this
occurs remains largely unknown. My Dissertation addressed this important question by
investigating the role of P2X4Rs at the cellular and the molecular level. Over the course
of this work I sought to address three questions: 1) how ethanol affects P2X4R function,
2) what regulatory functions are mediated by P2X4Rs, and 3) what the role of P2X4Rs
within addiction circuitry is. The findings from my investigations provide a novel
understanding of the mechanisms by which P2X4Rs function and support the hypothesis
that P2X4Rs are a promising target for AUD therapies.
In Chapter 2, I found that an interaction between two residues, arginine 33 (R33)
and aspartic acid 354 (D354), is necessary for P2X4R channel activity. This interaction is
also a site for ethanol action, as mutating R33 to aspartic acid, or D354 to arginine,
renders the P2X4Rs insensitive to ethanol. Additionally, D354, but not R33, likely plays a
role in channel gating, as IVM, a P2X4R positive allosteric modulator, elicited a
statistically significantly higher degree of potentiation (>10 fold) for D354R mutations, but
had no effect on R33D or R33D-D354R (double mutant) potentiation, both of which
behaved similarly to the WT receptor with respect to IVM potentiation (~5 fold).
Collectively, these findings provide a mechanistic understanding of how ethanol affects
P2X4R activity and provide new sites for drug targeting.

83

In Chapter 3, my investigations uncovered a novel inhibitory interaction between
P2XRs (either P2X2Rs or P2X4Rs) and NMDARs (consisting of GluN1 and GluN2A-C).
NMDARs have long been targets of interest for AUD, although drugs targeting NMDARs
have failed in clinical trials, either due a lack of efficacy or due to safety issues. Findings
from this study provide a novel and robust mechanism for NMDAR regulation: P2XR
signaling. Furthermore, this study provided direct evidence for a link between protein
targets for AUD: NMDARs and P2X4Rs. I demonstrated that this interaction was found to
rely on the c-terminus of P2X2R and P2X4R, and could be modulated via a small peptide
derived from the P2X4R c-terminus: 11C. Collectively, these findings provided insight into
the regulatory function of P2X4Rs as they relate to AUD targets, and suggest that 11C
could serve as a drug discovery platform for modulating P2XR activity.
In Chapter 4, I used a new tool to investigate the role of P2X4Rs and alcohol intake:
SiRNA-Loaded Electrodes kNocksdown Target, or SiLENT slice. This technique reduces
receptor function during extracellular brain slice recordings by knocking down targets
(SiRNA is incorporated into recording electrodes.) Using brain slice electrophysiology, we
found that ATP reduces the spontaneous firing rate of DA neurons in the VTA, the brain
region associated with the rewarding and reinforcing properties of ethanol. Using SiLENT
slice, I was able to provide new evidence supporting the hypothesis that P2X4Rs play a
key role in mediating this response to ATP. I found that using siRNA on neurons targeting
P2X4Rs produced statistically significantly lower ATP responses as compared to negative
controls (scrambled siRNA). Lastly, I found that ethanol inhibits P2X4R activity in a dose-
dependent manner, consistent with our animal and in vitro studies. However, ethanol did
not produce a statistically significant change in response to ATP following the reduction

84

in P2X4Rs, indicating that at least one other ATP-site affects DA VTA neurons, which is
ethanol insensitive. Collectively, these findings indicate that P2X4Rs play a crucial role in
mediating the intoxicating effects of ethanol, supporting their potential as therapeutic
targets for AUD.
Limitations
While these findings provide new evidence for the role of P2X4Rs in AUD, it is
important to understand the limitations of this work. Additional studies should be
performed in order to fully corroborate these findings, as several issues occur at the
cellular and molecular level which place restrictions on the interpretation and translation
of these results.
First, heterologous expression systems do not always model physiological
responses. For example, while both P2X3R and P2X4Rs have been shown to be
sensitive to ethanol at physiologically relevant concentrations in Xenopus laevis oocytes
(Davies et al., 2005), P2X3R studies utilizing human embryonic kidney (HEK) 293 cells
failed to show any ethanol effects at high ethanol concentrations (100 mM) (Fischer et
al., 2003) while studies in HEK 293 cells on P2X4Rs have been consistent with oocyte
findings (Ostrovskaya et al., 2011). Furthermore, X. laevis oocytes may not fully represent
the intracellular environment, possibly due to species differences in protein expression.
For example, NMDARs have been shown to rely on intracellular scaffold proteins, such
as PSD95 and HOMER, for trafficking and regulation (Lin et al., 2006) (Smothers et al.,
2016) which limit the scope of interpretation. Future studies incorporating brain slice
electrophysiology recordings could be used to enhance and better define the
physiological relevance of the interactions identified in my oocyte work.

85

A second limitation arises from the non-selective nature of purinergic signaling.
When released, ATP has the potential to activate not only ionotropic P2XRs, but also the
metabotropic family of purinergic receptors, P2YRs. Indeed, studies on VTA neurons
have demonstrated that both P2XRs and P2YRs can play a role in the regulation of
GABA-releasing terminals on DA neurons. Furthermore, even the effects of ethanol on
VTA neurons has been shown to depend on the P2R class (Xiao et al., 2008b). In the
past, members of ionotropic families of receptors have been characterized using selective
agonists or allosteric modulators (e.g. NMDA, AMPA, and Kainate receptors, which are
all glutamate receptors, but whose names are derived from the molecules that selectively
activate them). Unfortunately, selective P2XR pharmacological agonists/antagonists that
can differentiate between receptor subtypes have yet to be developed, and
positive/negative modulators such as IVM have off-target activity (e.g. GABAR activity).
This is a growing area of research, and future work incorporating additional methods
could be performed to help confirm specific P2XR subtypes that participated in the
outcomes. This could include immunohistochemistry, fast-scanning voltammetry, and
proteomic analyses.
Despite such caveats, these studies do provide important mechanistic information
regarding 1) EtOH-P2X4R interactions seen in behavioral models; 2) NMDA-P2X4R
crosstalk interactions seen in in vitro models; and 3) the role of P2X4R in the VTA region
of the brain.
Future directions
While the results from my studies have begun to answer mechanistic questions
about P2XR signaling, they also provide exciting new directions to advance our

86

understanding of P2XR signaling. These findings provide new opportunities for
investigation, which combine cellular and molecular approaches to further uncover the
role of P2X4Rs in health and disease.
P2XRs are increasingly being recognized for their regulatory role. GABA, AMPA,
and now NMDA receptors (NMDARs) have been shown to be regulated by P2XRs in a
subunit dependent manner (Boué-Grabot et al., 2004b; Jo et al., 2011; Pougnet et al.,
2016; Pougnet et al., 2014). Given that all of these interactions have relied on c-terminal
intracellular domains, one new direction to identify additional interaction partners is
proteomics. For example, a given P2XR c-terminus could be recombinantly expressed
(e.g. E. coli or HEK293 cells) and purified for use in immunoprecipitation (IP) studies.
Once the P2XR c-terminus was conjugated to IP resin, mouse/rat brain lysate could be
applied to the resin, washed, and eluted, and the bound fraction could be and analyzed
via mass spectroscopy (MS). This approach would provide a higher-throughput strategy
for discovering interaction partners compared to electrophysiology methods and achieve
a level of selectivity not currently possible in brain slices.
Additionally, it would be interesting to investigate the molecular nature of the
P2XR-NMDAR interaction. In my studies, I was able to demonstrate that 11C, a peptide
composed of the 11 c-terminal residues of the P2X4R, could disrupt the inhibitory
interaction between P2X2Rs or P2X4Rs and NMDARs. However, an important question
remains: is the interaction physical, or is it mediated by intracellular machinery (e.g.
scaffold proteins or protein kinases/phosphatases). As such, another direction is to
determine the nature of this interaction. I previously conjugated 11C to NHS biotin-
agarose beads; future experiments could use this column to attempt to pulldown NMDAR

87

from oocyte lysate samples. If NMDARs are present in western blot experiments, then
this interaction is likely physical. If not NMDARs are not present in western blot
experiments, then the 11C column be used to identify the proteins mediating this
interaction using mouse/rat brain lysate (see previous paragraph).
Recently, a novel knock-in mouse model was developed, which also provides new
opportunities for investigation. In this mouse model, the P2X4R gene has been replaced
with a fluorescently labeled internalization deficient P2X4R gene. The authors did not
report dramatic changes to basal excitatory transmission but did find alterations to long-
term potentiation and depression in CA1 hippocampal synapses. NMDARs have been
shown to play a key role in regulating these processes, and with my findings that P2X4Rs
inhibit NMDARs, a physiological role/mechanism begins to emerge. With this new mouse
model, the interaction between P2X4Rs and NMDARs could be further investigated, and
greater insight may be gleaned at the cellular and molecular level, while also providing
physiological insights.
My findings not only indicate that P2X4Rs regulate NMDAR activity, but also that
this interaction can be modulated 11C. Given its size, one possible new direction is to
create a small molecule mimetic of the 11C peptide. Previous brain slice
electrophysiology studies have shown that, when placed into a recording electrode
solution, 11C could modulate GABAergic neuron activity. As such, a small molecule
mimetic could serve as a more specific tool for modulating P2X4R activity with a
mechanism distinct from ivermectin: preventing receptor internalization.
Until recently, the physiological role of P2XRs was unclear. Their study was
hindered by their unconventional nature and their relatively recent discovery. However,

88

the findings from my investigations provide new molecular and cellular context for
P2X4Rs and, with the advent of new tools and complementary techniques, open new
avenues of investigation. Indeed, as the evidence for the importance of P2XRs continues
to accumulate, greater advancements and understanding regarding the importance and
role of P2X4Rs as a drug target for the development of therapies for AUD.

89

List of References
1. WHO, Global status report on alcohol and health. 2014: World Health
Organization.
2. NIAAA, Alcohol Facts and Statistics, N.I.o.A.A.a. Alcoholism, Editor. 2020,
Department of Health and Human Services. p. 5.
3. Peacock, A., et al., Global statistics on alcohol, tobacco and illicit drug use: 2017
status report. Addiction, 2018. 113(10): p. 1905-1926.
4. Substance-Related and Addictive Disorders, in Diagnostic and Statistical Manual
of Mental Disorders 5. 2013.
5. Koob, G.F. and N.D. Volkow, Neurocircuitry of addiction.
Neuropsychopharmacology, 2010. 35(1): p. 217-38.
6. Wright, C. and R.D. Moore, Disulfiram treatment of alcoholism. Am J Med, 1990.
88(6): p. 647-55.
7. Rösner, S., et al., Opioid antagonists for alcohol dependence. Cochrane
Database Syst Rev, 2010(12): p. Cd001867.
8. Plosker, G.L., Acamprosate: A Review of Its Use in Alcohol Dependence. Drugs,
2015. 75(11): p. 1255-68.
9. Kim, Y., et al., Practical outpatient pharmacotherapy for alcohol use disorder.
Drugs in context, 2018. 7: p. 212308-212308.
10. Di Chiara, G., Nucleus accumbens shell and core dopamine: differential role in
behavior and addiction. Behavioural Brain Research, 2002. 137(1): p. 75-114.
11. Koob, G.F., Drugs of abuse: anatomy, pharmacology and function of reward
pathways. Trends Pharmacol Sci, 1992. 13(5): p. 177-84.

90

12. Volkow, N.D., et al., Profound decreases in dopamine release in striatum in
detoxified alcoholics: possible orbitofrontal involvement. J Neurosci, 2007. 27(46): p.
12700-6.
13. Shen, R.Y., K.C. Choong, and A.C. Thompson, Long-term reduction in ventral
tegmental area dopamine neuron population activity following repeated stimulant or
ethanol treatment. Biol Psychiatry, 2007. 61(1): p. 93-100.
14. Martinez, D., et al., Alcohol Dependence Is Associated with Blunted Dopamine
Transmission in the Ventral Striatum. Biological Psychiatry, 2005. 58(10): p. 779-
786.
15. Weiss, F., et al., Ethanol self-administration restores withdrawal-associated
deficiencies in accumbal dopamine and 5-hydroxytryptamine release in dependent
rats. J Neurosci, 1996. 16(10): p. 3474-85.
16. Barria, A. and R. Malinow, Subunit-Specific NMDA Receptor Trafficking to
Synapses. Neuron, 2002. 35(2): p. 345-353.
17. Yardley, M.M., et al., Ivermectin reduces alcohol intake and preference in mice.
Neuropharmacology, 2012. 63(2): p. 190-201.
18. Wyatt, L.R., et al., Contribution of P2X4 receptors to ethanol intake in male
C57BL/6 mice. Neurochemical research, 2014. 39(6): p. 1127-1139.
19. Huynh, N., et al., Preclinical development of moxidectin as a novel therapeutic for
alcohol use disorder. Neuropharmacology, 2017. 113, Part A: p. 60-70.
20. Winslow, B.T., M. Onysko, and M. Hebert, Medications for Alcohol Use Disorder.
Am Fam Physician, 2016. 93(6): p. 457-65.

91

21. Xiao, C., et al., Ethanol Facilitates Glutamatergic Transmission to Dopamine
Neurons in the Ventral Tegmental Area. Neuropsychopharmacology, 2008. 34(2): p.
307-318.
22. Wang, J., et al., Ethanol Induces Long-Term Facilitation of NR2B-NMDA
Receptor Activity in the Dorsal Striatum: Implications for Alcohol Drinking Behavior.
The Journal of Neuroscience, 2007. 27(13): p. 3593-3602.
23. Eisenhardt, M., et al., Glutamate Receptors within the Mesolimbic Dopamine
System Mediate Alcohol Relapse Behavior. The Journal of Neuroscience, 2015.
35(47): p. 15523-15538.
24. Creed, M., et al., Cocaine Exposure Enhances the Activity of Ventral Tegmental
Area Dopamine Neurons via Calcium-Impermeable NMDARs. The Journal of
Neuroscience, 2016. 36(42): p. 10759-10768.
25. Nagy, J., et al., Role of Altered Structure and Function of NMDA Receptors in
Development of Alcohol Dependence. Current Neuropharmacology, 2005. 3(4): p.
281-297.
26. Kopec, C.D., et al., Glutamate receptor exocytosis and spine enlargement during
chemically induced long-term potentiation. J Neurosci, 2006. 26(7): p. 2000-9.
27. Prybylowski, K., et al., Relationship between Availability of NMDA Receptor
Subunits and Their Expression at the Synapse. The Journal of Neuroscience, 2002.
22(20): p. 8902-8910.
28. Prybylowski, K., et al., The Synaptic Localization of NR2B-Containing NMDA
Receptors Is Controlled by Interactions with PDZ Proteins and AP-2. Neuron, 2005.
47(6): p. 845-857.

92

29. Lin, Y., et al., PSD-95 and PKC converge in regulating NMDA receptor trafficking
and gating. Proceedings of the National Academy of Sciences, 2006. 103(52): p.
19902-19907.
30. Yi, Z., et al., The Role of the PDZ Protein GIPC in Regulating NMDA Receptor
Trafficking. The Journal of Neuroscience, 2007. 27(43): p. 11663-11675.
31. Sans, N., et al., NMDA receptor trafficking through an interaction between PDZ
proteins and the exocyst complex. Nat Cell Biol, 2003. 5(6): p. 520-530.
32. Khakh, Baljit S. and R.A. North, Neuromodulation by Extracellular ATP and P2X
Receptors in the CNS. Neuron. 76(1): p. 51-69.
33. Sheraz Khoja, V.S., Damaris Garcia, Liana Asatryan, Michael W. Jakowec, and
Daryl L Davies, Role of purinergic P2X4 receptors in regulating striatal dopamine
homeostasis and dependent behaviors. Journal of Neurochemistry, 2016. 139(1): p.
14.
34. Bo, X., et al., Tissue distribution of P2X4 receptors studied with an ectodomain
antibody. Cell and Tissue Research, 2003. 313(2): p. 159-165.
35. Asatryan, L., et al., Avermectins differentially affect ethanol intake and receptor
function: implications for developing new therapeutics for alcohol use disorders.
International Journal of Neuropsychopharmacology, 2014. 17(6): p. 907-916.
36. Letisha R Wyatt, S.C.G., Sheraz Khoja, Michael W Jakowec, Ronald L Alkana,
Marco Bortolato, and Daryl L Davies, Sociocommunicative and Sensorimotor
Impairments in Male
P2X4-Deficient Mice. Neuropsychopharmacology, 2013. 38: p. 10.

93

37. Jo, Y.-H., et al., Cross-talk between P2X4 and γ-Aminobutyric Acid, Type A
Receptors Determines Synaptic Efficacy at a Central Synapse. The Journal of
Biological Chemistry, 2011. 286(22): p. 19993-20004.
38. Pougnet, J.-T., et al., ATP P2X Receptors Downregulate AMPA Receptor
Trafficking and Postsynaptic Efficacy in Hippocampal Neurons. Neuron, 2014. 83(2):
p. 417-430.
39. Boué-Grabot, É., et al., Cross-talk and Co-trafficking between ρ1/GABA
Receptors and ATP-gated Channels. Journal of Biological Chemistry, 2004. 279(8):
p. 6967-6975.
40. Pougnet, J.-T., et al., P2X-mediated AMPA receptor internalization and synaptic
depression is controlled by two CaMKII phosphorylation sites on GluA1 in
hippocampal neurons. Scientific Reports, 2016. 6: p. 31836.
41. Bristow, L.J., et al., Preclinical Characterization of (R)-3-((3S,4S)-3-fluoro-4-(4-
hydroxyphenyl)piperidin-1-yl)-1-(4-methylbenzyl)pyrrolidin-2-one (BMS-986169), a
Novel, Intravenous, Glutamate N-Methyl-D-Aspartate 2B (GluN2B) Receptor
Negative Allosteric Modulator with Potential in Major Depressive Disorder. Journal of
Pharmacology and Experimental Therapeutics, 2017.
42. Sacks, J.J., et al., 2010 National and State Costs of Excessive Alcohol
Consumption. Am J Prev Med, 2015. 49(5): p. e73-e79.
43. Franklin, K.M., et al., P2X4 receptors (P2X4Rs) represent a novel target for the
development of drugs to prevent and/or treat alcohol use disorders. Front Neurosci,
2014. 8: p. 176.

94

44. Buell, G., G. Collo, and F. Rassendren, P2X receptors: an emerging channel
family. Eur J Neurosci, 1996. 8(10): p. 2221-8.
45. Soto, F., et al., P2X4: an ATP-activated ionotropic receptor cloned from rat brain.
Proc Natl Acad Sci U S A, 1996. 93(8): p. 3684-8.
46. Gonzales, R.A., M.O. Job, and W.M. Doyon, The role of mesolimbic dopamine in
the development and maintenance of ethanol reinforcement. Pharmacol Ther, 2004.
103(2): p. 121-46.
47. Kimpel, M.W., et al., Functional gene expression differences between inbred
alcohol-preferring and -non-preferring rats in five brain regions. Alcohol, 2007. 41(2):
p. 95-132.
48. Tabakoff, B., et al., Genetical genomic determinants of alcohol consumption in
rats and humans. BMC Biol, 2009. 7: p. 70.
49. Xiong, K., C. Li, and F.F. Weight, Inhibition by ethanol of rat P2X(4) receptors
expressed in Xenopus oocytes. Br J Pharmacol, 2000. 130(6): p. 1394-8.
50. Davies, D.L., et al., Ethanol differentially affects ATP-gated P2X(3) and P2X(4)
receptor subtypes expressed in Xenopus oocytes. Neuropharmacology, 2005. 49(2):
p. 243-53.
51. Popova, M., et al., A POINT MUTATION IN THE ECTODOMAIN-
TRANSMEMBRANE 2 INTERFACE ELIMINATES THE INHIBITORY EFFECTS OF
ETHANOL IN P2X4 RECEPTORS. Journal of neurochemistry, 2010. 112(1): p. 307-
317.
52. Olga Ostrovskaya, L.A., Letisha Wyatt, Maya Popova, Kaixun Li, Robert W.
Peoples, Ronald L. Alkana, and Daryl L. Davies, Ethanol Is a Fast Channel Inhibitor

95

of P2X4 Receptors. JThe Journal of Pharmacology and Experimental Therapuetics,
2011. 337(1): p. 8.
53. Young, M.T., et al., Molecular shape, architecture, and size of P2X4 receptors
determined using fluorescence resonance energy transfer and electron microscopy.
J Biol Chem, 2008. 283(38): p. 26241-51.
54. Silberberg, S.D., T.H. Chang, and K.J. Swartz, Secondary structure and gating
rearrangements of transmembrane segments in rat P2X4 receptor channels. J Gen
Physiol, 2005. 125(4): p. 347-59.
55. Jelínkova, I., et al., Identification of P2X(4) receptor transmembrane residues
contributing to channel gating and interaction with ivermectin. Pflugers Arch, 2008.
456(5): p. 939-50.
56. Stojilkovic, S.S., et al., Structural insights into the function of P2X4: an ATP-
gated cation channel of neuroendocrine cells. Cell Mol Neurobiol, 2010. 30(8): p.
1251-8.
57. Heymann, G., et al., Inter- and intrasubunit interactions between transmembrane
helices in the open state of P2X receptor channels. Proc Natl Acad Sci U S A, 2013.
110(42): p. E4045-54.
58. Asatryan, L., et al., Ivermectin Antagonizes Ethanol Inhibition in Purinergic P2X4
Receptors. Journal of Pharmacology and Experimental Therapeutics, 2010. 334(3):
p. 720-728.
59. Popova, M., et al., Tryptophan 46 is a site for ethanol and ivermectin action in
P2X4 receptors. Purinergic Signalling, 2013. 9(4): p. 621-632.

96

60. Kawate, T., et al., Crystal structure of the ATP-gated P2X(4) ion channel in the
closed state. Nature, 2009. 460(7255): p. 592-8.
61. Hattori, M. and E. Gouaux, Molecular mechanism of ATP binding and ion
channel activation in P2X receptors. Nature, 2012. 485(7397): p. 207-12.
62. Egan, T.M., W.R. Haines, and M.M. Voigt, A domain contributing to the ion
channel of ATP-gated P2X2 receptors identified by the substituted cysteine
accessibility method. J Neurosci, 1998. 18(7): p. 2350-9.
63. Li, S.C. and C.M. Deber, Glycine and beta-branched residues support and
modulate peptide helicity in membrane environments. FEBS Lett, 1992. 311(3): p.
217-20.
64. Silberberg, S.D., M. Li, and K.J. Swartz, Ivermectin Interaction with
transmembrane helices reveals widespread rearrangements during opening of P2X
receptor channels. Neuron, 2007. 54(2): p. 263-74.
65. Prichard, R., C. Ménez, and A. Lespine, Moxidectin and the avermectins:
Consanguinity but not identity. International Journal for Parasitology: Drugs and
Drug Resistance, 2012. 2: p. 134-153.
66. Deber, C.M., et al., Val-->Ala mutations selectively alter helix-helix packing in the
transmembrane segment of phage M13 coat protein. Proc Natl Acad Sci U S A,
1993. 90(24): p. 11648-52.
67. Jindrichova, M., et al., Functional relevance of aromatic residues in the first
transmembrane domain of P2X receptors. J Neurochem, 2009. 109(3): p. 923-34.
68. Polinsky, A., et al., Minimum energy conformations of proline-containing helices.
Biopolymers, 1992. 32(4): p. 399-406.

97

69. Crawford, D.K., et al., Evidence that ethanol acts on a target in Loop 2 of the
extracellular domain of alpha1 glycine receptors. J Neurochem, 2007. 102(6): p.
2097-2109.
70. Perkins, D.I., et al., Targets for ethanol action and antagonism in loop 2 of the
extracellular domain of glycine receptors. J Neurochem, 2008. 106(3): p. 1337-49.
71. Naito, A., et al., Glycine and GABA(A) ultra-sensitive ethanol receptors as novel
tools for alcohol and brain research. Mol Pharmacol, 2014. 86(6): p. 635-46.
72. Haines, W.R., et al., The first transmembrane domain of the P2X receptor
subunit participates in the agonist-induced gating of the channel. J Biol Chem, 2001.
276(35): p. 32793-8.
73. Jiang, L.H., et al., Amino acid residues involved in gating identified in the first
membrane-spanning domain of the rat P2X(2) receptor. J Biol Chem, 2001. 276(18):
p. 14902-8.
74. Samways, D.S., B.S. Khakh, and T.M. Egan, Allosteric modulation of Ca2+ flux in
ligand-gated cation channel (P2X4) by actions on lateral portals. J Biol Chem, 2012.
287(10): p. 7594-602.
75. Cao, L., et al., Polar residues in the second transmembrane domain of the rat
P2X2 receptor that affect spontaneous gating, unitary conductance, and rectification.
J Neurosci, 2009. 29(45): p. 14257-64.
76. Kracun, S., et al., Gated access to the pore of a P2X receptor: structural
implications for closed-open transitions. J Biol Chem, 2010. 285(13): p. 10110-21.

98

77. Mirshahi, T. and J.J. Woodward, Ethanol sensitivity of heteromeric NMDA
receptors: Effects of subunit assembly, glycine and NMDAR1 Mg2+-insensitive
mutants. Neuropharmacology, 1995. 34(3): p. 347-355.
78. Roberto, M. and F.P. Varodayan, Synaptic targets: Chronic alcohol actions.
Neuropharmacology, 2017. 122: p. 85-99.
79. Khakh, Baljit S. and R.A. North, Neuromodulation by Extracellular ATP and P2X
Receptors in the CNS. Neuron, 2012. 76(1): p. 51-69.
80. Bertin, E., et al., Increased surface P2X4 receptor regulates anxiety and memory
in P2X4 internalization-defective knock-in mice. Mol Psychiatry, 2020.
81. Pankratov, Y., et al., A purinergic component of the excitatory postsynaptic
current mediated by P2X receptors in the CA1 neurons of the rat hippocampus. Eur
J Neurosci, 1998. 10(12): p. 3898-902.
82. Mori, M., et al., Fast synaptic transmission mediated by P2X receptors in CA3
pyramidal cells of rat hippocampal slice cultures. J Physiol, 2001. 535(Pt 1): p. 115-
23.
83. Jo, Y.H. and R. Schlichter, Synaptic corelease of ATP and GABA in cultured
spinal neurons. Nat Neurosci, 1999. 2(3): p. 241-5.
84. Sim, J.A., et al., Altered Hippocampal Synaptic Potentiation in P2X4 Knock-Out
Mice. The Journal of Neuroscience, 2006. 26(35): p. 9006-9009.
85. Lalo, U., et al., ATP from synaptic terminals and astrocytes regulates NMDA
receptors and synaptic plasticity through PSD-95 multi-protein complex. 2016. 6: p.
33609.

99

86. Baxter, A.W., et al., Role of P2X4 receptors in synaptic strengthening in mouse
CA1 hippocampal neurons. European Journal of Neuroscience, 2011. 34(2): p. 213-
220.
87. Toulmé, E., et al., Functional Properties of Internalization-Deficient P2X4
Receptors Reveal a Novel Mechanism of Ligand-Gated Channel Facilitation by
Ivermectin. Molecular Pharmacology, 2006. 69(2): p. 576-587.
88. Ogata, J., et al., Effects of Anesthetics on Mutant N-Methyl-d-Aspartate
Receptors Expressed in Xenopus Oocytes. Journal of Pharmacology and
Experimental Therapeutics, 2006. 318(1): p. 434-443.
89. Boué-Grabot, É., et al., Intracellular Cross Talk and Physical Interaction between
Two Classes of Neurotransmitter-Gated Channels. The Journal of Neuroscience,
2003. 23(4): p. 1246-1253.
90. Boué-Grabot, É., et al., Subunit-specific Coupling between γ-Aminobutyric Acid
Type A and P2X2 Receptor Channels. Journal of Biological Chemistry, 2004.
279(50): p. 52517-52525.
91. Krupp, J.J., et al., Calcium-dependent inactivation of recombinant N-methyl-D-
aspartate receptors is NR2 subunit specific. Molecular Pharmacology, 1996. 50(6):
p. 1680-1688.
92. Zheng, X., et al., Ca2+ Influx Amplifies Protein Kinase C Potentiation of
Recombinant NMDA Receptors. The Journal of Neuroscience, 1997. 17(22): p.
8676-8686.

100

93. Michailidis, I.E., et al., Phosphatidylinositol-4,5-Bisphosphate Regulates NMDA
Receptor Activity through α-Actinin. The Journal of Neuroscience, 2007. 27(20): p.
5523-5532.
94. Emerit, M.B., et al., A New Mechanism of Receptor Targeting by Interaction
between Two Classes of Ligand-Gated Ion Channels. The Journal of Neuroscience,
2016. 36(5): p. 1456-1470.
95. Royle, S.J., L.K. Bobanović, and R.D. Murrell-Lagnado, Identification of a Non-
canonical Tyrosine-based Endocytic Motif in an Ionotropic Receptor. Journal of
Biological Chemistry, 2002. 277(38): p. 35378-35385.
96. Royle, S.J., et al., Non-canonical YXXGΦ endocytic motifs: recognition by AP2
and preferential utilization in P2X4 receptors. Journal of Cell Science, 2005. 118(14):
p. 3073-3080.
97. Rubio, M.E. and F. Soto, Distinct Localization of P2X Receptors at Excitatory
Postsynaptic Specializations. The Journal of Neuroscience, 2001. 21(2): p. 641-653.
98. Pankratov, Y.V., U.V. Lalo, and O.A. Krishtal, Role for P2X Receptors in Long-
Term Potentiation. The Journal of Neuroscience, 2002. 22(19): p. 8363-8369.
99. Mizuno, T., I. Kanazawa, and M. Sakurai, Differential induction of LTP and LTD is
not determined solely by instantaneous calcium concentration: an essential
involvement of a temporal factor. Eur J Neurosci, 2001. 14(4): p. 701-8.
100. Huganir, R.L. and R.A. Nicoll, AMPARs and synaptic plasticity: the last 25
years. Neuron, 2013. 80(3): p. 704-17.

101

101. Boué-Grabot, E., et al., Intracellular cross talk and physical interaction
between two classes of neurotransmitter-gated channels. J Neurosci, 2003. 23(4): p.
1246-53.
102. Richler, E., E. Shigetomi, and B.S. Khakh, Neuronal P2X2 receptors are
mobile ATP sensors that explore the plasma membrane when activated. J Neurosci,
2011. 31(46): p. 16716-30.
103. Bobanovic, L.K., S.J. Royle, and R.D. Murrell-Lagnado, P2X Receptor
Trafficking in Neurons Is Subunit Specific. The Journal of Neuroscience, 2002.
22(12): p. 4814-4824.
104. Ostrovskaya, O., et al., Ethanol is a fast channel inhibitor of P2X4
receptors. J Pharmacol Exp Ther, 2011. 337(1): p. 171-9.
105. Wyatt, L.R., et al., Contribution of P2X4 receptors to ethanol intake in
male C57BL/6 mice. Neurochem Res, 2014. 39(6): p. 1127-39.
106. Asatryan, L., et al., Implication of the purinergic system in alcohol use
disorders. Alcohol Clin Exp Res, 2011. 35(4): p. 584-94.
107. Khoja, S., et al., Reduced expression of purinergic P2X4 receptors
increases voluntary ethanol intake in C57BL/6J mice. Alcohol, 2018. 68: p. 63-70.
108. Szalay, K.S., et al., Local non-synaptic modulation of aldosterone
production by catecholamines and ATP in rat: implications for a direct neuronal fine
tuning. Horm Metab Res, 1998. 30(6-7): p. 323-8.
109. Wise, R.A. and M. Morales, A ventral tegmental CRF-glutamate-dopamine
interaction in addiction. Brain Res, 2010. 1314: p. 38-43.

102

110. Wise, R.A., Neurobiology of addiction. Curr. Opin. Neurobiol, 1996. 6(2):
p. 243-251.
111. Koob, G.F. and N.D. Volkow, Neurobiology of addiction: a neurocircuitry
analysis. Lancet Psychiatry, 2016. 3(8): p. 760-773.
112. Koob, G.F., Alcoholism: allostasis and beyond. Alcohol Clin. Exp. Res,
2003. 27(2): p. 232-243.
113. Oliva, I. and M.J. Wanat, Ventral Tegmental Area Afferents and Drug-
Dependent Behaviors. Front Psychiatry, 2016. 7: p. 30.
114. Morikawa, H. and R.A. Morrisett, Ethanol action on dopaminergic neurons
in the ventral tegmental area: interaction with intrinsic ion channels and
neurotransmitter inputs. Int Rev Neurobiol, 2010. 91: p. 235-88.
115. Samson, H.H., et al., Effect of dopamine agonists and antagonists on
ethanol- reinforced behavior: the involvement of the nucleus accumbens. Brain. Res.
Bull, 1993. 30: p. 133-141.
116. Samson, H.H., et al., Alcohol self-administration: role of mesolimbic
dopamine. Ann. N. Y. Acad. Sci, 1992. 654: p. 242-253.
117. Bechtholt, A.J. and C.L. Cunningham, Ethanol-induced conditioned place
preference is expressed through a ventral tegmental area dependent mechanism.
Behav Neurosci, 2005. 119(1): p. 213-23.
118. Rodd, Z.A., et al., The reinforcing properties of salsolinol in the ventral
tegmental area: evidence for regional heterogeneity and the involvement of
serotonin and dopamine. Alcohol Clin. Exp. Res, 2008. 32(2): p. 230-239.

103

119. Nimitvilai, S., et al., Suppression of Gq Function Using Intra-Pipette
Delivery of shRNA during Extracellular Recording in the Ventral Tegmental Area.
Front Cell Neurosci, 2013. 7: p. 7.
120. You, C., et al., Ethanol acts on KCNK13 potassium channels in the ventral
tegmental area to increase firing rate and modulate binge-like drinking.
Neuropharmacology, 2019. 144: p. 29-36.
121. Calabresi, P., M.G. Lacey, and R.A. North, Nicotinic excitation of rat
ventral tegmental neurones in vitro studied by intracellular recording. Br. J.
Pharmacol, 1989. 98: p. 135-140.
122. Mueller, A.L. and M.S. Brodie, Intracellular recording from putative
dopamine-containing neurons in the ventral tegmental area of Tsai in a brain slice
preparation. J. Neurosci. Methods, 1989. 28: p. 15-22.
123. Margolis, E.B., et al., The ventral tegmental area revisited: is there an
electrophysiological marker for dopaminergic neurons? J. Physiol, 2006. 577(Pt 3):
p. 907-924.
124. Pesavento, E., et al., Blocking the NGF-TrkA interaction rescues the
developmental loss of LTP in the rat visual cortex: role of the cholinergic system.
Neuron, 2000. 25(1): p. 165-175.
125. Mameli, M., et al., Rapid synthesis and synaptic insertion of GluR2 for
mGluR-LTD in the ventral tegmental area. Science, 2007. 317(5837): p. 530-533.
126. Nimitvilai, S., M.A. McElvain, and M.S. Brodie, Reversal of dopamine D2
agonist-induced inhibition of ventral tegmental area neurons by Gq-linked

104

neurotransmitters is dependent on protein kinase C, G protein-coupled receptor
kinase and dynamin. J. Pharmacol. Exp. Ther, 2013. 344(1): p. 253-263.
127. Kupferschmidt, D.A., et al., Brain BLAQ: Post-hoc thick-section
histochemistry for localizing optogenetic constructs in neurons and their distal
terminals. Front Neuroanat, 2015. 9: p. 6.
128. Khoja, S., et al., Dopamine Receptor Blockade Attenuates Purinergic
P2X4 Receptor-Mediated Prepulse Inhibition Deficits and Underlying Molecular
Mechanisms. Front Cell Neurosci, 2019. 13: p. 331.
129. Asatryan, L., et al., Avermectins differentially affect ethanol intake and
receptor function: implications for developing new therapeutics for alcohol use
disorders. Int J Neuropsychopharmacol, 2014. 17(6): p. 907-16.
130. Xia, Y.F., E.B. Margolis, and G.O. Hjelmstad, Substance P inhibits
GABAB receptor signalling in the ventral tegmental area. J. Physiol, 2010. 588(Pt 9):
p. 1541-1549.
131. Brodie, M.S. and T.V. Dunwiddie, Cholecystokinin potentiates dopamine
inhibition of mesencephalic dopamine neurons in vitro. Brain. Res, 1987. 425: p.
106-113.
132. Fischer, W., et al., Characterization of P2X3, P2Y1 and P2Y4 receptors in
cultured HEK293-hP2X3 cells and their inhibition by ethanol and trichloroethanol. J
Neurochem, 2003. 85(3): p. 779-90.
133. Smothers, C.T., et al., Altered NMDA Receptor Function in Primary
Cultures of Hippocampal Neurons from Mice Lacking the Homer2 Gene. Synapse
(New York, N.Y.), 2016. 70(1): p. 33-39.

105

134. Xiao, C., et al., Purinergic type 2 receptors at GABAergic synapses on
ventral tegmental area dopamine neurons are targets for ethanol action. J
Pharmacol Exp Ther, 2008. 327(1): p. 196-205.

Abstract (if available)

Abstract Alcohol use disorder (AUD) affects over 18 million people in the United States alone and can be attributed to an economic burden of over $200 billion per year. Despite considerable efforts toward developing new therapies for AUD, there are currently only three FDA-approved medications to pharmacologically treat AUD, none of which are considered to be particularly effective in reducing alcohol consumption or craving. The field of neuroscience has an abundance of research dedicated to the roles of individual receptors involved in alcohol addiction, such as gamma-aminobutyric acid (GABA) and glutamate (NMDA) receptors, although attempts to target these receptors has failed to produce promising results in the clinic. More recently, the purinergic (P2XR) family of receptors has become an emerging target for various diseases. In particular, the purinergic receptor family member P2X4 has been linked to the voluntary alcohol consumption pathway in AUD mouse models. Genetic, pharmacological and behavioral mouse studies report an inverse relationship between ethanol (EtOH) intake and P2X4R activity, although the underlying mechanism for this phenomenon is not understood. Interestingly, several members of the P2XR family have been shown to cross-regulate the activity of another receptor type (e.g. GABAA or AMPA receptors.) P2X4Rs are expressed in the Ventral Tegmental Area (VTA) of the brain, along with ionotropic NMDA receptors (NMDARs), a central component of addiction circuitry. The goal of this dissertation is to investigate molecular and cellular mechanisms by which P2X4Rs regulate ethanol consumption and addiction. This is accomplished using complementary techniques in electrophysiology, molecular biology, and biochemistry. Chapter 1 describes our understanding of AUD as it relates to treatment and pharmacotherapy. Chapter 2 describes a novel interaction between residues that is required for P2X4R channel function and a site for ethanol action. Chapter 3 describes an interaction between P2XRs (either P2X2R or P2X4R) and NMDARs, which provides a previously unrecognized link to addiction targets. Chapter 4 describes the role of P2X4Rs in mediating the effects of ethanol in the ventral tegmental area (VTA) of the brain, a region known to play a key role in reward and addiction. Chapter 5 summarizes the results of my findings, describes the limitations of my studies, and provides future directions of investigation. Overall, my findings 1) indicate that P2X4Rs play a previously unrecognized role in neuronal cell firing within the VTA, 2) suggest a novel mechanism for the regulation of NMDARs, and 3) provide potential targets for the treatment of AUD and other neurological diseases.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Role of purinergic P2X4 receptors in regulation of dopamine homeostasis in the basal ganglia and associated behaviors

PDF

The preclinical evaluation of moxidectin as a platform for drug development for alcohol use disorder

PDF

The study of DHM effects on counteracting ethanol intoxications

PDF

Development of lentiviral vector mediated gene transfer techniques as tools to overexpress/knockdown P2X4 receptors in vitro

PDF

Role of purinergic P2X7 receptors in inflammatory responses in the brain and liver: a study using a mouse model of chronic ethanol and high-fat diet exposure

PDF

Preclinical investigation of ivermectin as a novel therapeutic agent for treatment of alcohol use disorders

PDF

Development of glycine and GABAA ultra-sensitive ethanol receptors (USERs) as novel tools for alcohol and brain research

PDF

Regulation of ethanol intake by purinergic P2X4 receptors

PDF

Positive and negative modulatory sites for ethanol in alpha1 glycine receptors

PDF

Ethanol induced modulation of microglial P2X7 receptor expression and its role in neuroinflammation

PDF

Developing ultrasensitive ethanol receptors (USERS) as novel tools for alcohol research: optimizing Loop 2 mutations in α1 GlyRs

PDF

Investigation of the anti-alcohol potential of avermectin drugs in in vitro studies of P2X4 and GABA(A) receptors

PDF

Development of dihydromyricetin (DHM) as a novel therapy for alcoholic liver disease (ALD) and alcohol use disorder (AUD)

PDF

DHM reduces cognitive impairment induced by social isolation and ethanol consumption

PDF

Sodium butyrate prevents antibiotic-induced increase in ethanol drinking in C57BL/6J mice by modulating neuroinflammatory response

PDF

Development of ivermectin as a platform for the treatment and/or prevention of alcohol use disorders

PDF

Beneficial effect of antibiotic treatment on alcohol-related liver pathology in mice is not due to reduction in the butyrate-producing gut microbial phyla

Asset Metadata

Creator Rodriguez, Larry (author)

Core Title Mechanisms of P2XR-mediated ethanol consumption

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Pharmaceutical Sciences

Publication Date 08/03/2020

Defense Date 06/03/2020

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag alcohol use disorder,cross-talk,dopaminergic neuron,drug discovery,ethanol,NMDA receptor,OAI-PMH Harvest,P2X2 receptor,P2X4 receptor,pharmacology,purinergic receptor,two-electrode voltage-clamp electrophysiology,ventral tegmental area,VTA

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Davies, Daryl (committee chair), Okamoto, Curtis (committee chair), Asatryan, Liana (committee member)

Creator Email larryrod@usc.edu,lrodriguez@scripps.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-357711

Unique identifier UC11666321

Identifier etd-RodriguezL-8876.pdf (filename),usctheses-c89-357711 (legacy record id)

Legacy Identifier etd-RodriguezL-8876.pdf

Dmrecord 357711

Document Type Dissertation

Rights Rodriguez, Larry

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA