Excitatory/Inhibitory Balance across Ontogeny Contribute to Age-Specific Behavioral Outcomes of Ethanol-Like Challenge in Conditioned Taste Aversion

Abstract

Adolescent-typical sensitivities to ethanol (EtOH) are characterized in part by reduced sensitivity to EtOH’s aversive effects. Rodent studies have shown that adolescents are less sensitive than adults to aversive properties of EtOH in a conditioned taste aversion (CTA) paradigm. To the extent that EtOH exerts antagonist-like actions upon glutamate receptors and/or agonist-like actions upon GABA receptors, age differences in excitatory/inhibitory balance may regulate age-specific EtOH sensitivities, such as attenuated sensitivity of adolescents to EtOH aversion. In our experiments, adolescent and adult Sprague-Dawley rats were tested for CTA following challenge with one of the following pharmacological agents: glutamatergic AMPA1 receptor antagonist NBQX, glutamatergic NMDA NR2B receptor antagonist ifenprodil, and extrasynaptic GABA_A receptor agonist THIP to determine whether these induced age-specific aversive sensitivities like those seen with EtOH. NBQX administration did not induce CTA. The highest dose of extrasynaptic GABA_A agonist THIP induced CTA in adolescents but not adults, an opposite ontogenetic profile as seen following EtOH. Ifenprodil exerted an age-specific pattern of CTA similar to that seen with EtOH in males, with adolescents being insensitive to ifenprodil’s aversive effects relative to adults. Thus, only antagonism of NR2B receptors in male rats mimicked age-specific sensitivities to the aversive effects of EtOH.

Introduction

Adolescent alcohol use is a prevalent behavior within the United States, with as many as 2 out of 5 individuals aged 18 to 25 being binge drinkers and as many as 4.5 million 12–20 year old individuals reported as binge drinkers according to the 2016 National Survey on Drug Use and Health (Bose et al., 2016). This high rate of alcohol consumption is not nearly as prevalent in adults, which may be in part due to biological factors such as neural alterations occurring during this transient period of rapid maturation (Spear, 2016). Age differences in alcohol consumption are conserved across species, with similarly elevated rates of alcohol consumption seen in adolescents relative to adults in studies using animal models such as the rat (Doremus et al., 2005; Vetter et al., 2007). Using such animal models, adolescents have been shown to differ from adults in their sensitivity to alcohol (or ethanol (EtOH)) in ways that could permit higher consumption levels – e.g., showing less sensitivity to various EtOH effects that could serve as feedback cues to moderate intake (e.g., Spear, 2011; Spear, 2016).

One such developmental difference is the decreased sensitivity displayed by adolescent relative to adult rats to the aversive effects of EtOH indexed via taste aversions (Schramm-Sapyta et al., 2010). Aversive effects of EtOH and other drugs can be measured via a conditioned taste aversion (CTA) wherein an unconditioned stimulus (US) such as EtOH is paired with a conditioned stimulus (CS), resulting in decreased consumption of the CS tastant when the animal is subsequently allowed CS access (Welzl et al., 2001). Our lab and others have shown that adolescents are less sensitive to the aversive effects of EtOH in a CTA paradigm relative to adults (Anderson et al., 2010; Doremus-Fitzwater et al., 2010; Schramm-Sapyta et al., 2014; Vetter-O’Hagen et al., 2009), with adolescents requiring a higher dose of EtOH to produce an aversive effect relative to adults (Saalfield & Spear, 2016). What remain unclear are neural contributors to this age-specific attenuation of adolescents to the aversive properties of EtOH.

Across ontogeny, the general balance of excitation to inhibition (E/I) alters, with a shift toward inhibition increasing as adulthood approaches, an increase in inhibitory tone thought to contribute to developmental declines in plasticity (Selemon, 2013; Sturman & Moghaddam, 2011a). Specifically, glutamatergic (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and N-methyl-ᴅ-aspartate (NMDA)) activities diminish while γ-aminobutyric acid (GABA) activity increases across ontogeny (Caballero et al., 2014). As a result, adolescents tend to exhibit greater E/I balance – reflecting immature, hyperglutamatergic and/or hypoGABAergic states - when compared to adults. There is evidence to suggest that this reduced inhibition during adolescence contributes to age differences in reward sensitivity (Sturman & Moghaddam, 2011a; 2011b); however, it is not clear if this reduction in inhibition contributes to aversion sensitivity. While EtOH alters activity of a variety of neurotransmitter/neuromodulator systems, particularly notable effects include an agonist-like effect on the GABA system (Lobo & Harris, 2008) and an antagonist-like action on glutamatergic receptor systems, including both NMDA (Allgaier, 2002) and AMPA receptor (Möykkynen et al., 2003) subtypes. Previous research has shown that glutamatergic and GABAergic systems modulate a wide variety of effects of EtOH including self-administration (Stephens & Brown, 1999), intoxication (Jones et al., 2008), and sedation (Silveri & Spear, 2002), as well as the aversive properties of EtOH. There is some evidence to suggest that a CTA can be elicited following repeated pairings with the NMDA antagonist MK-801 in rats (Traverso et al., 2012), with infusion of the AMPA antagonist, 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), into the basolateral amygdala (BLA) likewise reported to produce an aversive response in rats (Yasoshima et al., 2005). Conversely, a GABA_A postsynaptic antagonist, picrotoxin, has been shown to attenuate EtOH-induced CTA, but not amphetamine-induced CTA (Smith et al., 1989). Together, these findings suggest a potential role of developmental differences in GABAergic activation and/or antagonism of NMDA/non-NMDA receptor systems in contributing to the resilience of adolescents to the aversive properties of EtOH.

In order to parse the potential role of developmental alterations in the E/I system in the ontogeny of EtOH’s aversive effects, we examined the effects of GABAergic and glutamatergic manipulations on induction of CTA in adolescent and adult rats to determine whether these ontogenetic patterns reflect those observed previously with EtOH. Drugs examined included: the AMPA1 receptor antagonist, NBQX; an agonist of extrasynaptic GABA_A receptors, THIP; and the NMDA NR2B antagonist, ifenprodil. Preliminary data conducted in male rats from our laboratory suggest that adolescents are less sensitive than adults to the aversive effects of ifenprodil, a similar ontogenetic pattern as to that seen with EtOH (Ramirez et al., 2010), although effects of AMPA receptor blockade on production of CTA in adolescents and adults has yet to be examined. Moreover, it is apparent that EtOH exerts some of its effects on extrasynaptic GABA_A receptors; however, the behavioral outcomes studied have focused primarily on sedation (Lobo & Harris, 2008) and mitigation of EtOH aversion (Smith et al., 1989). It is unclear if antagonism of these receptors alone will produce an aversion similar to that of EtOH. We propose that, if greater excitatory and/or reduced inhibitory actions of these neurotransmitter systems among adolescents relative to adults contributes to the attenuated aversive sensitivity of these systems in adolescents, then adolescents will show a similarly blunted sensitivity to the aversive effects of NBQX, THIP and ifenprodil relative to adults as seen with EtOH.

Method

Subjects

Animals bred and reared at Binghamton University were used in the present experiments (N=546; 282 male, 264 female). All animals were housed in a temperature-controlled (22˚C) vivarium maintained on a 12-/12-hr light/dark cycle with lights on at 0700h, and ad libitum access to food (Purina Rat Chow, Lowell, MA) and water (with the exception of water restriction prior to testing). Animal body weights were recorded on both conditioning and test days. Litters were culled 8–10 pups on postnatal [P] day 1 (P1). Weaning occurred on P21, with subjects pair-housed with same-sex non-littermates to allow assignment of housing pairs to the same dose condition without placing more than one animal/litter into a given group (see below). All maintenance and handling of animals followed the guidelines for animal care established by the National Institutes of Health, using protocols approved by Binghamton University’s Institutional Animal Care and Use Committee.

Experimental Design

The design for Experiment 1 was a 2 age (adolescent, adult) × 2 sex (male, female) × 6 dose (0.0, 1.0, 2.0, 4.0, 6.0 and 8.0 mg/kg NBQX) factorial, with 2 age (adolescent, adult) × 2 sex (male, female) × 4 dose (0.0, 1.0, 2.0 and 4.0 mg/kg THIP) and 2 age (adolescent, adult) × 2 sex (male, female) × 4 drug dose (0.0, 1.5, 3.0 and 6.0 mg/kg ifenprodil) factorials being used for Experiments 2 and 3, respectively. A broader range of doses was assessed with NBQX in Experiment 1 given the large dose range studied in the literature with this drug in terms of EtOH consumption (Ruda-Kucerova et al., 2017; Stephens & Brown, 1999) and social behavior (Vekovischeva et al., 2007). No more than one animal of the same sex was assigned to a given drug dose, with same-sex littermates assigned semi-randomly to different drug doses in order to avoid the possible confounding of litter with drug effects (Holson & Pearce, 1992; Zorrilla, 1997).

Drugs

NBQX disodium salt (2,3-Dioxo-6-nitro-1,2,3,4-tetrahydrobenzo[f]quinoxaline-7-sulfonamide disodium salt) (Tocris Bioscience) was dissolved in 0.9% saline and administered intraperitoneally (i.p.) at doses of 0.0 (vehicle control), 1.0, 2.0, 4.0, 6.0, and 8.0 mg/kg. THIP hydrochloride (4,5,6,7-Tetrahydroisoxazolo[5,4-c]pyridine-3-ol hydrochloride) (Tocris Bioscience) was dissolved in 0.9% saline, and administered i.p. at doses of 0.0 (vehicle control), 1.0, 2.0 and 4.0 mg/kg. The non-competitive NMDA receptor antagonist ifenprodil hemitartrate ((1R*,2S*)-erythro-2-(4-Benzylpiperidino)-1-(4-hydroxyphenyl)-1-propanol hemi-(DL)-tartrate) (Tocris Bioscience) was dissolved in double-distilled sterile water (ddH2O) and administered i.p. at doses of 0 (vehicle control), 1.5, 3.0 and 6.0 mg/kg. All doses were administered at a volume of 2 ml of drug solution/kg body weight.

Materials

A palatable tastant, “supersaccharin” (supersac: 3% sucrose, 0.125% saccharin in water, modified from Ji et al. (2008) (see Dannenhoffer & Spear, 2016; Saalfield & Spear, 2016) was used as the CS. All CTA conditioning and testing was conducted within the animals’ home cages under 15–20 lux, with animals separated via a wire mesh divider during conditioning and test, thereby allowing assessment of individual intakes while minimizing the stressor of social isolation (Hall, 1998).

Procedure

Animals were 50% water deprived one day prior to conditioning and testing in order to encourage consumption of the novel CS. This was accomplished using procedures in routine use in our laboratory (e.g., Anderson et al., 2010). Two days prior to conditioning (P33 or P68), water intake of the housing pair was measured. On the following day (P34 or P69), cage mates were provided with 50% of their previous day’s water intake for the next 24 hours, at the end of which conditioning was conducted (P35 or P70). Animals were then given ad libitum access to water for 24 hours, followed another 24 hour period of 50% water restriction prior to intake assessment on test day (P37 or P72).

On conditioning day, animals were weighed and returned to their side of the divided home cage with no food or water present. Ten minutes later, each animal was given access to one bottle containing supersac for a 30-min drinking period. Immediately after the drinking period, bottles were removed, animals were injected with their assigned dose of NBQX, THIP or ifenprodil and returned to the divided home cage for ten more minutes. Dividers were then removed and standard cage tops providing food and water returned. On test day, animals were again separated in their home cage with a divider for a 10-min habituation period (without food and water present) followed by each animal being given 30-min access to one bottle containing the supersac solution. All bottles were weighed before and after each session to determine intake of the CS on conditioning and test days, with standard cage tops containing food and water returned to the cages following cessation of each drinking period. In order to ensure sufficient sampling of the CS to assess development of an aversive response to the tastant, animals that did not consume ≥ of 0.5 mls on conditioning day were eliminated from the study. No more than 5 animals were removed from any group, with final sample sizes of 8–10 per group. Final sizes of each of these groups are listed in Tables 1, 2, and 3 in the order that the experiments are presented.

Table 1.

Group sizes and conditioning day intake (mean ± SEM) for all ages, sexes, and doses in Experiment 1 (NBQX).

NBQX	DOSE	N	Conditioning Day Intake
AF	0	9	4.79 ± 0.44
AF	1	8	4.51 ± 0.33
AF	2	8	3.36 ± 0.53
AF	4	10	4.47 ± 0.75
AF	6	9	4.50 ± 0.64
AF	8	10	5.72 ± 0.51
AM	0	10	5.03 ± 0.62
AM	1	9	6.49 ± 0.38
AM	2	10	5.48 ± 1.02
AM	4	10	4.15 ± 0.85
AM	6	9	5.79 ± 0.63
AM	8	8	4.84 ± 0.76
BF	0	8	4.01 ± 0.74
BF	1	9	3.65 ± 0.92
BF	2	9	4.92 ± 0.90
BF	4	10	2.86 ± 0.72
BF	6	8	4.75 ± 1.07
BF	8	8	4.65 ± 1.02
BM	0	10	7.07 ± 1.36
BM	1	10	5.15 ± 1.06
BM	2	9	7.10 ± 1.51
BM	4	10	8.01 ± 1.00
BM	6	8	8.73 ± 0.55
BM	8	8	6.78 ± 1.37

AF = adolescent female; AM = adolescent male; BF = adult female; BM = adult male

Table 2.

Group sizes and conditioning day intake (mean ± SEM) for all ages, sexes, and doses in Experiment 2 (THIP).

THIP	DOSE	N	Conditioning Day Intake
AF	0	8	4.19 ± 0.65
AF	1	9	3.63 ± 0.74
AF	2	8	3.70 ± 0.69
AF	4	9	3.74 ± 0.75
AM	0	9	3.90 ± 0.68
AM	1	10	3.83 ± 0.87
AM	2	10	5.34 ± 0.66
AM	4	9	5.08 ± 0.82
BF	0	8	3.73 ± 0.72
BF	1	9	3.30 ± 0.63
BF	2	8	3.68 ± 0.81
BF	4	9	4.32 ± 0.69
BM	0	10	5.18 ± 1.12
BM	1	8	5.53 ± 1.01
BM	2	8	3.76 ± 1.23
BM	4	10	4.48 ± 0.97

AF = adolescent female; AM = adolescent male; BF = adult female; BM = adult male

Table 3.

Group sizes and conditioning day intake (mean ± SEM) for all ages, sexes, and doses in Experiment 3 (Ifenprodil).

IFEN	DOSE	N	Conditioning Day Intake
AF	0	10	2.55 ± 0.71
AF	1.5	9	3.02 ± 0.50
AF	3	10	3.83 ± 0.68
AF	6	8	3.65 ± 0.81
AM	0	10	4.81 ± 0.79
AM	1.5	10	4.56 ± 0.85
AM	3	10	6.10 ± 0.64
AM	6	10	4.96 ± 0.70
BF	0	8	3.23 ± 0.47
BF	1.5	9	2.39 ± 0.45
BF	3	9	3.52 ± 0.66
BF	6	8	3.79 ± 1.08
BM	0	10	5.77 ± 0.93
BM	1.5	10	4.12 ± 0.91
BM	3	9	4.54 ± 0.88
BM	6	9	4.33 ± 1.21

AF = adolescent female; AM = adolescent male; BF = adult female; BM = adult male

Data Analysis

A 2 age × 2 sex × “n” dose × 2 (conditioning/test) repeated measures ANOVA was conducted to assess CTA for each of the test drugs. Fisher’s post hocs were used to assess changes from conditioning to test day as well as age, sex and dose effects. Test day data were also analyzed following transformation into % change from vehicle control (%Control), using 2 age × 2 sex × “n” dose ANOVAs and the same post hoc analyses. In order to minimize the possibility of type 2 errors in these large factorial ANOVAs, repeated measure (conditioning/test day) and %Control analyses were also conducted within each age and sex group, with Fisher’s post hocs used to determine the locus of significant dose effects.

Results

Experiment 1: NBQX CTA

There was neither a main effect nor interactions involving dose in the repeated measure analysis of the CTA raw data with NBQX (see Table 4A). Hence there was no evidence of an aversion to the AMPA1 antagonist in the raw data, and the ontogenetic pattern of this drug did not resemble that seen following EtOH administration. There was a significant main effect of sex (F (1, 193) =23.517, p<0.001) in this ANOVA, with males not surprisingly consuming more than females, as well as significant main effects of age (F (1,193=16.679, p<0.001) and day (F (1, 193) =174.933; p<0.001), tempered by an age by day interaction (F (1, 193) =19.716; p<0.001); regardless of dose, adult consumption was greater than that of adolescents, a difference particularly notable on test day. In the overall ANOVA of the test day data presented as %Control, again no dose effects emerged (see Table 4B); only a significant age by sex interaction (F (1, 193) =21.136; p<0.001) was evident, with age differences in both males and females as well as sex differences in both age groups in opposing directions (adolescent males: 93.76 % ± 5.95 %; adolescent females: 132.94 % ± 10.61 %; adult males: 116.51 % ± 7.43 %; adult females: 84.91 % ± 6.60 %).

Table 4.

Experiment 1 (NBQX) statistical analyses results

Table 4A. NBQX
2 sex X 2 age X 6 dose Repeated Measure ANOVA
Age	F(1,193)=16.679; p=0.000
Sex	F(1.193)=23.517; p=0.000
Dose	F(5,193)=0.529; p=0.754
Age*Sex	F(1,193)=5.202; p=0.024
Age*Dose	F(5,193)=1.438; p=0.212
Sex*Dose	F(5,193)=0.928; p=0.464
Age*Sex*Dose	F(5,193)=1.395; p=0.228
Day	F(1,193)=174.933; p=0.000
Day*Age	F(1,193)=19.716; p=0.000
Day*Sex	F(1,193)=0.854; p=0.356
Day*Dose	F(5,193)=2.187; p=0.057
Day*Age*Sex	F(1,193)=0.416; p=0.520
Day*Age*Dose	F(5,193)=1.003; p=0.417
Day*Sex*Dose	F(5,193)=0.784; p=0.562
Day*Age*Sex*Dose	F(5,193)=2.130; p=0.064
Table 4B. NBQX
2 sex X 2 age X 6 dose ANOVA of %Control Intake
Age	F(1,193)=2.518; p=0.114
Sex	F(1,193)=0.338; p=0.561
Dose	F(5,193)=1.286; p=0.271
Age*Sex	F(1,193)=21.136; p=0.000
Age*Dose	F(5,193)=1.979; p=0.083
Sex*Dose	F(5,193)=1.343; p=0.248
Age*Sex*Dose	F(5,193)=1.844; p=0.105
Table 4C. NBQX
Repeated Measure ANOVAs within each age/sex group
Adolescent Females	Dose: F(5,48)=2.243; p=0.065
	Day: F(1,48)=16.729; p=0.000
	Day*Dose: F(5,48)=3.204; p=0.014
Adolescent Males	Dose: F(5,50)=0.821; p=0.541
	Day: F(1,50)=26.899; p=0.000
	Day*Dose: F(5,50)=0.689; p=0.634
Adult Females	Dose: F(5,46)=1.249; p=0.302
	Day: F(1,46)=71.366; p=0.000
	Day*Dose: F(5,46)=1.196; p=0.326
Adult Males	Dose: F(5,49)=0.828; p=0.536
	Day: F(1,49)=69.498; p=0.000
Day*Dose: F(5,49)=1.447; p=0.224
Table 4D. NBQX
%Control one-way ANOVA within each age/sex group
Adolescent Females	F(5,48)=2.748; p=0.029
Adolescent Males	F(5.50)=0.633; p=0.675
Adult Females	F(5,46)=1.373; p=0.252
Adult Males	F(5,49)=0.880; p=0.502

Due to the age and sex differences both in the repeated measure and %Control ANOVAS, dose effects were assessed within each age and sex group. These analyses revealed significant dose effects only in adolescent females, with adolescent females given 8 mg/kg consuming more than all other groups on test day [significant effects of day (F (1, 48) =16.729; p<0.001) and day by dose (F (5, 48) =3.204; p<0.005)]. The %Control ANOVA within adolescent females revealed the same dose response pattern (F (5, 48) =2.748; p<0.05; see Figure 1B). The repeated measure ANOVAs for adult males and females yielded only main effects of day (F (1, 49) =69.498; p<0.001, and F (1, 46) =71.366; p<0.001, respectively), reflecting higher consumption on test day. Analysis of the %Control data in adolescent males, adult males and females yielded no significant dose effects (Figures 1A, 1C-D; Tables 4C-D).

Figure 1.

NBQX-induced CTA results in adolescent males (a), adolescent females (b), adult males (c), and adult females (d). NBQX did not induce a CTA at any dose for any age/sex group.

Experiment 2: THIP CTA

Reminiscent of the NBQX data, administration of THIP did not generally produce an ontogenetic pattern similar to that seen with EtOH. In the case of THIP, however, some aversive effects of THIP emerged in adolescent but not adult males and females - an opposite ontogenetic pattern as to that seen with EtOH. At low doses, there was also some evidence for an (EtOH-like) appetitive effect of THIP in adolescent males. The omnibus repeated measure ANOVA revealed significant main effects of sex (F (1, 126) =15.498; p<0.001), and day (F (1, 126) =40.521; p<0.001), along with significant interactions involving day by sex (F (1, 126) =11.982; p<0.001), day by dose (F (3, 126) =4.180; p<0.01), and day by sex by dose (F (3, 126) =5.299; p<0.005) wherein test day intake after a training dose of 4 mg/kg was significantly reduced from controls in males only (p=0.030) (see Table 5A).

Table 5.

Experiment 2 (THIP) statistical analyses results

Table 5A. THIP
2 sex X 2 age X 4 dose Repeated Measure ANOVA
Age	F(1,126)=1.629; p=0.204
Sex	F(1,126)=15.498; p=0.000
Dose	F(3,126)=0.850; p=0.469
Age*Sex	F(3,126)=1.478; p=0.226
Age*Dose	F(3,126)=0.681; p=0.565
Sex*Dose	F(3,126)=2.511; p=0.062
Age*Sex*Dose	F(3,126)=2.215; p=0.090
Day	F(1,126)=40.521; p=0.000
Day*Age	F(1,126)=3.543; p=0.062
Day*Sex	F(1,126)=11.982; p=0.001
Day*Dose	F(3,126)=4.180; p=0.007
Day*Age*Sex	F(1,126)=2.423; p=0.122
Day*Age*Dose	F(3,126)=2.575; p=0.057
Day*Sex*Dose	F(3,126)=5.299; p=0.002
Day*Age*Sex*Dose	F(3,126)=1.010; p=0.391
Table 5B. THIP
2 sex X 2 age X 4 dose ANOVA of %Control Intake
Age	F(1,126)=2.233; p=0.138
Sex	F(1,126)=2.747; p=0.100
Dose	F(3,126)=1.254; p=0.293
Age*Sex	F(1,126)=14.177; p=0.000
Age*Dose	F(3,126)=3.161; p=0.027
Sex*Dose	F(3,126)=5.613; p=0.001
Age*Sex*Dose	F(3,126)=1.981; p=0.120
Table 5C. THIP
Repeated Measure ANOVAs within each age/sex group
Adolescent Females	Dose: F(3,30)=2.361; p=0.091
	Day: F(1,30)=2.152; p=0.153
	Day*Dose: F(3,30)=3.433; p=0.029
Adolescent Males	Dose: F(3,34)=2.951; p=0.046
	Day: F(1,34)=8.919; p=0.005
	Day*Dose: F(3,34)=6.110; p=0.002
Adult Females	Dose: F(3,30)=1.608; p=0.208
	Day: F(1,30)=3.024; p=0.092
	Day*Dose: F(3,30)=0.944; p=0.432
Adult Males	Dose: F(3,32)=0.840; p=0.482
	Day: F(1,32)=38.404; p=0.000
	Day*Dose: F(3,32)=2.016; p=0.131
Table 5D. THIP
%Control one-way ANOVA within each age/sex group
Adolescent Females	F(3,30)=4.122; p=0.015
Adolescent Males	F(3,34)=5.436; p=0.004
Adult Females	F(3,30)=1.885; p=0.153
Adult Males	F(3,32)=1.281; p=0.298

The %Control ANOVA revealed interactions of age by sex (F (1, 126) =14.177; p<0.001), age by dose (F (3, 126) =3.161; p<0.05), and sex by dose (F (3, 126) =5.613; p<0.005) which were further explored via planned ANOVAs and corresponding post-hocs within each age/sex group (see Table 5B). Within adolescent males, the repeated measure ANOVA revealed significant main effects of dose (F (3, 34) =2.951; p<0.05) and day (F (1, 34) =8.919; p<0.01), and their interaction (F (3, 34) =6.110; p<0.005): 2 mg/kg was appetitive (i.e., test day consumption in this group was significantly greater than that of saline control animals); however, 4 mg/kg significantly reduced consumption relative to both 1 and 2 mg/kg (but not saline controls) on test day, indicating a trend for the emergence of an aversive effect at this dose. The analysis of %Control revealed a main effect of dose (F (3, 34) =5.436; p<0.005), with a similar dose response pattern emerging as in the repeated measure test day comparisons (Figure 2A). Within adolescent females, a day by dose interaction (emerged in the repeated measure (F (3, 30) =3.433; p<0.05) and %Control analyses (F (3, 30) =4.122; p<0.05), with both analyses revealed that 2 and 4 mg/kg resulted in a CTA relative to control animals (Figure 2B). No effects or interactions of dose emerged in either analysis of adult males or females (Figure 2C-D; Table 5C-D).

Figure 2.

THIP-induced CTA results in adolescent males (a), adolescent females (b), adult males (c), and adult females (d). The highest dose of THIP (4.0 mg/kg) induced a CTA in adolescent males (a), while the two highest doses induced a CTA in adolescent females (b).

Experiment 3: Ifenprodil CTA

Similar to what has been previously reported with EtOH (e.g., Saalfield & Spear, 2016), a high dose of ifenprodil (6 mg/kg) was aversive in adult males, whereas adolescent males showed no aversive effect. The omnibus ANOVA revealed main effects of sex (F (1, 133) =20.764; p<0.001), and day (F (1, 133) =68.021; p<0.001), along with day by age (F (1, 133) =6.862; p<0.01), day by dose (F (3, 133) =2.972; p<0.05), and day by age by dose interactions (F (3, 133) =2.787; p<0.05). Adult males exhibited significant CTA to both 1.5 and 6 mg/kg ifenprodil, but not to 3 mg/kg. Adolescents were not sensitive to ifenprodil’s aversive effects at any dose (see Table 6A).

Table 6.

Experiment 3 (Ifenprodil) statistical analyses results

Table 6A. Ifenprodil
2 sex X 2 age X 4 dose Repeated Measure ANOVA
Age	F(1,133)=1.069; p=0.303
Sex	F(1,133)=20.764; p=0.000
Dose	F(3,133)=1.457; p=0.229
Age*Sex	F(1,133)=1.215; p=0.272
Age*Dose	F(3,133)=1.631; p=0.185
Sex*Dose	F(3,133)=0.857; p=0.465
Age*Sex*Dose	F(3,133)=0.053; p=0.984
Day	F(1,133)=68.021; p=0.000
Day*Age	F(1,133)=6.862; p=0.010
Day*Sex	F(1,133)=1.763; p=0.187
Day*Dose	F(3,133)=2.972; p=0.034
Day*Age*Sex	F(1,133)=1.270; p=0.262
Day*Age*Dose	F(3,133)=2.787; p=0.043
Day*Sex*Dose	F(3,133)=0.405; p=0.749
Day*Age*Sex*Dose	F(3,133)=0.696; p=0.556
Table 6B. Ifenprodil
2 sex X 2 age X 4 dose ANOVA of %Control Intake
Age	F(1,133)=5.427; p=0.021
Sex	F(1,133)=0.800; p=0.373
Dose	F(3,133)=0.930; p=0.428
Age*Sex	F(1,133)=0.193; p=0.661
Age*Dose	F(3,133)=1.451; p=0.231
Sex*Dose	F(3,133)=0.396; p=0.756
Age*Sex*Dose	F(3,133)=0.071; p=0.975
Table 6C. Ifenprodil
Repeated Measure ANOVAs within each age/sex group
Adolescent Females	Dose: F(3,33)=0.236; p=0.871
	Day: F(1,33)=3.643; p=0.065
	Day*Dose: F(3,33)=0.464; p=0.709
Adolescent Males	Dose: F(3,36)=0.436; p=0.729
	Day: F(1,36)=14.610; p=0.001
	Day*Dose: F(3,36)=1.786; p=0.167
Adult Females	Dose: F(3,30)=0.531; p=0.664
	Day: F(1,30)=46.296; p=0.000
	Day*Dose: F(3,30)=1.856; p=0.158
Adult Males	Dose: F(3,34)=2.727; p=0.059
	Day: F(1,34)=21.354; p=0.000
	Day*Dose: F(3,34)=2.312; p=0.094
Table 6D. Ifenprodil
%Control one-way ANOVA within each age/sex group
Adolescent Females	F(3,33)=0.091; p=0.965
Adolescent Males	F(3,36)=0.823; p=0.490
Adult Females	F(3,30)=0.722; p=0.547
Adult Males	F(3,34)=3.996; p=0.015

These findings were confirmed in the analyses of %Control within each age/sex combination where a dose effect was found only in adult males, with significant CTA evident following doses of 1.5 and 6.0 mg/kg relative to controls (F (3, 34) =3.996; p<0.05; Figure 3C). No dose effects in the %Control data were evident in adolescent males or females at either age (Figures 3A-B, 3D; Table 6C-D).

Figure 3.

Ifenprodil-induced CTA results in adolescent males (a), adolescent females (b), adult males (c), and adult females (d). CTA was only present in adult males (c) at the low (1.5 mg/kg) and high (6.0 mg/kg) but not moderate (3.0 mg/kg) dose of ifenprodil.

Discussion

To the extent that attenuated sensitivity to the aversive effects of EtOH in adolescence relative to adulthood is in part due to ontogenetic differences in EtOH’s antagonistic actions on AMPA1 receptors, stimulation of extrasynaptic GABA_A receptors or antagonism of NMDA NR2B receptors, we predicted to find similar ontogenetic patterns of aversion to NBQX, THIP and ifenprodil, respectively. The present studies revealed no evidence for an adolescent insensitivity to the aversive effects of NBQX and THIP, though some age dependent effects reminiscent of attenuated adolescent aversive effects were evident with ifenprodil in males. In Experiment 1, NBQX did not mimic the effects typically seen with EtOH in our CTA paradigm. Repeated measure analyses demonstrated that supersac intake was higher on test day when collapsed across all other variables, an increase over initial intake probably related to some initial neophobia on conditioning day when exposed to the novel tastant that is diminished by the second exposure (Lin et al., 2012). Adults consumed more than adolescents possibly due to intake being proportional to their body weight and caloric intake. Moreover, within adolescent females, the highest dose (8 mg/kg) was appetitive in that their intake was significantly higher than their controls.

It is unclear why adolescent females developed a preference to 8 mg/kg of NBQX. There is little evidence to suggest that NBQX is appetitive; rather, there are several studies that suggest that NBQX contributes to decreases in EtOH consumption (Ruda-Kucerova et al., 2018; Stephens & Brown, 1999). It is possible that the increase in consumption in these females could have been driven by estradiol-mediated kainate receptor antagonism given that NBQX is not specific to AMPA receptors alone (Todd et al., 2007). Why this effect should be evident in adolescent but not adult females, however, is unclear. Aside from this effect, there were no effects of dose, suggesting that NBQX exerted no aversive effects. There has been little evidence to suggest that NBQX elicits an aversive response directly; however, research has shown that LiCl-induced taste aversion can be interrupted by the memory retrieval impairment induced by administration of NBQX into the BLA of rats (Garcia-de la Torre et al., 2014; Rodriguez-Ortiz et al., 2012). This suggests a possible role for AMPA-type receptors in the consolidation of aversive (Garcia-de la Torre et al., 2014; Rodriguez-Ortiz et al., 2012) or gustatory (Yasoshima et al., 2000) memory, but not an aversive response following AMPA receptor antagonism per se.

In Experiment 2, aversive responses to THIP were opposite of those seen with EtOH, with adolescents generally being more sensitive to CTA than adults. This effect was driven by adolescent females showing an aversion at 2 and 4 mg/kg THIP; some sign of an aversion was also evident in adolescent males who showed attenuated intake in response to 4 mg/kg relative to lower doses. Previous research in our lab has shown that THIP induces a sedative-like effect in adolescents, but not adults, at higher doses; e.g., 4 mg/kg (Dannenhoffer et al., 2018) which could contribute to the apparent signs of an aversive response seen in the present study. Thus, the aversion observed may be specific to the sedation caused by a higher dose of the drug rather than aversion related to gastrointestinal malaise per se. However, because THIP was injected following intake on conditioning day, it is not the case that the sedative effects of THIP interfered with the animals’ ability to consume the tastant. Indeed, there is some evidence to suggest that THIP may induce a place aversion following 6 mg/kg when assessed via place conditioning, with locomotor activity also significantly decreased during the conditioning sessions (Vashchinkina et al., 2012). Although only adults were tested for place conditioning in that study, a separate test of sedative properties in both adolescents and adults conducted in the same experimental series revealed age-specific sedative patterns, with doses 3 and 6 mg/kg THIP sedating adolescents more readily than adults. The lack of attenuated CTA to THIP in adolescents relative to adults in the present study could also be related in part to the potential role of this agonist in the rewarding properties of EtOH, with several studies showing that pretreatment with THIP elevates EtOH consumption (Boyle et al., 1992; Smith et al., 1992) and preference (Boyle et al., 1993). Given that adolescents are more sensitive to the rewarding properties of EtOH than are adults (Spear & Varlinskaya, 2010), it is possible that low doses of THIP may have mimicked this effect in our adolescent rats. Effects of this drug on both sedation and reward could complicate interpretation of age differences in responsiveness to THIP. Overall, the role of extrasynaptic GABA_A receptors in the rewarding and aversive effects of EtOH needs further investigation in developmental studies.

Of the three potential candidates explored, ifenprodil more clearly induced an ontogenetic pattern similar to EtOH. In Experiment 3, adult males were more sensitive to the aversive effects of ifenprodil at doses that did not produce aversive effects in adolescent males, confirming preliminary finding obtained in male rats in our lab previously (Ramirez et al., 2010). When CTAs were analyzed in both males and females in the present study, the aversive effects of ifenprodil evident in adulthood were found to be specific to adult males and not females, findings congruent with sex differences in EtOH CTA in adults (Schramm-Sapyta et al., 2014). Thus, age-specific sensitivities to the aversive effects of ifenprodil were evident only in males, with females being resistant to ifenprodil’s aversive effects at both ages. There is some evidence to suggest that females may be less sensitive to the analgesic effects of NMDA antagonists (Kavaliers & Choleris, 1997) as well as more sensitive to elevations in locomotor activity (Frantz & Van Hartesveldt, 1999) than males. Estrogen has also been suggested to moderate responses to ifenprodil (Dong et al., 2007). Interestingly, the effect of ifenprodil dose on CTA in adult males was biphasic, with doses of 1.5 and 6 mg/kg inducing aversive effects that were not evident at a moderate dose (3 mg/kg). To date, there is little other evidence suggesting that NMDA antagonism elicits age-specific biphasic effects, and hence this effect would need replication before exploring contributors to potential biphasic actions of this drug.

These findings of similarity between EtOH and ifenprodil’s actions upon NMDA NR2B receptors confirm the hypothesis that age differences in response to EtOH may be mediated in part by the ontogeny of this receptor system. EtOH exerts the greatest antagonist-like action upon NR2A and NR2B relative to NR1, 2C-D receptors (Nagy, 2004). Moreover, antagonism of these receptors have been associated with a signal of intoxication in human subjects (Krystal et al., 2003a), with EtOH-dependent patients being less sensitive to the aversive properties of the NMDA receptor antagonist, ketamine (Krystal et al., 2003b). General antagonists such as MK-801 have also elicited CTA in rats (Traverso et al., 2012). We hypothesized that adolescents may be less sensitive to the aversive effects of NMDA receptor antagonism in part related to the greater expression of NR2B receptors in adolescents relative to adults (Wenzel et al., 1997). To the extent that adolescents exhibit a greater E/I balance relative to adults, it would be anticipated that adolescents might be less sensitive to some behavioral outcomes of NMDA antagonism such as CTA.

There are several limitations to the present experiments. The chosen drugs do not have complete specificity for the receptors targeted. NBQX is considered both an AMPA and a kainate receptor antagonist (Ruda-Kucerova et al., 2017); therefore, we cannot conclude that the results are purely driven by AMPA receptor antagonism. Similarly, ifenprodil has been shown to antagonize serotonin subtype 3 and α1 noradrenergic receptors (Malinowska et al., 1999; McCool & Lovinger, 1995) as well as NR1A and NR2A subunits (Williams, 1993), therefore our conclusions may not be specific to NR2B antagonism. Studies have shown that the serotonin precursor 5-HTP can induce a taste aversion to ethanol (Zabik & Roache, 1983). Given that ifenprodil modulates serotonin receptors (McCool & Lovinger, 1995), it is possible that ifenprodil- and ethanol-induced CTA may be regulated through more than the NR2B system. It is also possible that the doses chosen for investigation may not have been optimal. For instance, NBQX has been primarily investigated in the anticonvulsant literature and hence doses used have focused primarily on those producing such effects; there are, however, a few studies that used doses within our range to assess the rewarding properties of hedonic stimuli including stimulants (Li et al., 1997), EtOH (Ruda-Kucerova et al., 2017) and non-drug reinforcers such as sucrose (Stephens & Brown, 1999). Another limitation is that manipulation of any one of these three receptors may not solely parallel the effects of EtOH given that EtOH exerts its effects through interaction with a myriad of receptor types simultaneously (e.g., Koob et al., 1998). It is possible that co-administration of two or all three substances could induce synergistic effects reminiscent of the ontogeny of EtOH’s aversive effects.

The present experiments suggest that age-sensitive aversive response to EtOH may be in part due to antagonism of NMDA NR2B receptors, more so than through antagonism of AMPA1 receptors or stimulation of extrasynaptic GABA_A receptors. The attenuated aversion of adolescents to ifenprodil relative to adults is sex-specific in that only males displayed this ontogenetic pattern. These data correspond with prior data from our lab suggesting that NR2B receptor antagonism and acute EtOH exposure produce similar age-specific behavioral effects in terms of dose-dependent effects on social facilitation and social inhibition (Morales & Spear, 2014; Morales et al., 2013). Overall, the present findings support the suggestion that enhanced glutamatergic, specifically NMDA, activity in adolescents may contribute to a dampened aversive response to EtOH in a sex-specific manner.