Attention-deficit/hyperactivity disorder (ADHD) is a highly prevalent neurodevelopmental disorder characterized by impulsivity, inattention, and hyperactivity (American Psychological Association [APA], 2013). ADHD is a commonly diagnosed in childhood but can continue into adulthood (Weibel et al., 2020) and is often comorbid with other externalizing disorders (Faraone et al., 2003; Palacio et al., 2004; Frick and Nigg, 2012; Hansen et al., 2018). Based on the 2016 National Parent Survey, about 9.8% of children are diagnosed with ADHD and 6 in 10 children with ADHD had at least one other mental, emotional, or behavioral disorder (Bitsko et al., 2022).

Although many factors in the environment can contribute to the development of ADHD, the literature has shown that genetics can help explain ADHD variability (Spencer et al., 2007). Recent data has shown a linkage of ADHD and other externalizing behaviors with markers on chromosome 4q13.2 (Arcos-Burgos et al., 2004; Acosta et al., 2008; Arcos-Burgos and Muenke, 2010). Mapping of this region has revealed that variants in the Lphn3 (Adgrl3) gene (OMIM 616417) predispose individuals to ADHD (Acosta et al., 2008, 2011, 2016) and predict ADHD severity and response to treatment (Arcos-Burgos et al., 2010; Acosta et al., 2011; Bruxel et al., 2015). Similar studies in other populations have also found that Lphn3 gene variants contribute to ADHD susceptibility (Ribases et al., 2011; Hwang et al., 2015; Gomez-Sanchez et al., 2016; Martinez et al., 2016; Huang et al., 2019; Kappel et al., 2019; Puentes-Rozo et al., 2019).

Research in animal models has also provided some corroborating evidence for the role of Lphn3 in ADHD. For example, zebrafish that lack lphn3.1 (one of two Lphn3 orthologs) were found to be hyperactive (Lange et al., 2012, 2018)–an effect that was attenuated by the ADHD medications methylphenidate and atomoxetine (Lange et al., 2012). The down-regulation of lphn3.1 in zebrafish caused a misplacement of dopamine (DA) (but not norepinephrine or serotonin) neurons (Lange et al., 2012) and decreased locomotor sensitivity to DA agonists and antagonists (Lange et al., 2018). Likewise, Lphn3^–/– knockout (KO) mice and rats are also hyperactive (Wallis et al., 2012; Regan et al., 2019), and they have been shown to be impaired on a facet of impulsivity called impulsive action (Mortimer et al., 2019; Sable et al., 2021) which is the inability to inhibit a prepotent motor response (Bari and Robbins, 2013; MacKillop et al., 2016). The exact mechanism whereby alterations in Lphn3 gene expression alters catecholamine neurotransmission is still being investigated. However, in both Lphn3 mutant mice and rats, the expression levels of the DA transporter (DAT) gene (Slc6a3) and the protein itself differ from wildtype controls. Adult Lphn3 KO rats have increased DAT expression (Regan et al., 2019) and increased reuptake of DA (i.e., functional implication for increase in DAT) (Regan et al., 2020) in the dorsal striatum. Likewise, adult Lphn3^–/– mice demonstrate overexpression of Slc6a3 in whole brain (Wallis et al., 2012) and in the dorsal striatum (Pramod et al., 2013), but downregulation of Slc6a3 in the prefrontal cortex (PFC) (Orsini et al., 2016; Mortimer et al., 2019). DAT expression is associated with ADHD (see Pramod et al., 2013 for review), including the site of action for many ADHD medications (Fone and Nutt, 2005; Gerlach et al., 2013; Faraone, 2018), so these results are particularly noteworthy.

While previous research has demonstrated Lphn3 contributes to hyperactivity and deficits in impulsive action, this project examined the impact of Lphn3 deletion in rats on a different facet of impulsivity. Using a delay-discounting (DD) task, we assessed impulsive choice, which is the inability to delay gratification (Reynolds et al., 2002). We also report “positive control” measures for the same behavioral assay in spontaneously hypertensive rats (SHRs) which have also been proposed as an animal model of ADHD (Prediger et al., 2005; Sagvolden et al., 2005b,2009; Kantak et al., 2008; Meneses et al., 2011; Sagvolden and Johansen, 2012; Garcia and Kirkpatrick, 2013; Natsheh and Shiflett, 2018) and have previously exhibited DD deficits (Fox et al., 2008; Aparicio et al., 2019; Sjoberg et al., 2021). Compared to their control strain, Wistar-Kyoto (WKY) rats, SHRs also exhibit hyperactivity and inattention (Russell, 2011; Sagvolden and Johansen, 2012) as well as deficits in impulsive action (Sable et al., 2021; González-Barriga and Orduña, 2022). Here, we expected to observe more impulsive choice in the SHR and Lphn3 KO rats represented by their tendency to choose the small, immediate reward more often that the larger, delayed reward relative to their control strains, WKY and Lphn3 WT rats, respectively.

If a rat did not demonstrate 60% choice for the larger reward at the 0 s delay during the final maintenance phase, it was determined that the rat had not learned to differentiate between the levers associated with the small versus large reward (i.e., the rat had not learned the task). Thus, these data were not included in the final analyses. Specifically, data from 4 SHRs (1 male, 3 female), 3 WKY rats (3 male, 0 female), 3 Lphn3 KO rats (2 male, 1 female), and 6 Lphn3 WT rats (4 male, 2 female) were not included leaving final n’s of 20 SHRs (11 male, 9 female), 21 WKYs (9 male, 12 female), 29 Lphn3 KOs (14 male, 15 female), and 27 Lphn3 WTs (12 male, 15 female).

If a sphericity violation was found for any within-subjects effect, a Greenhouse–Geisser correction was used to reduce the risk of a Type I error because ε < 0.75 in all cases (Maxwell and Delaney, 1999). In the interest of brevity, only significant genotype- and strain-related main effects and interactions are reported. There were no significant strain × sex or genotype × sex interactions for any of the dependent variables, so the results are presented collapsed across sex.

We hypothesized that impulsive choice would be greater in the SHR and Lphn3 KO rats relative to their control strains–WKY and WT rats, respectively. However, the results only partially supported this hypothesis. While the SHRs demonstrated a decrease in the percent choice for the larger reward as well as a more negative slope and decreased area under the curve compared to WKY rats, the Lphn3 KO and WT rats did not differ on these measures. These results were surprising as both SHRs and Lphn3 KO rats have been shown to be impulsive when the task assessed impulsive action (Sable et al., 2021).

One possibility is that both the SHRs and Lphn3 KO rats exhibit impulsive choice, but they present the impairment differently. As previously mentioned, DD deficits have been previously reported in SHRs (Fox et al., 2008; Sjoberg et al., 2021). These results and ours indicate that the SHRs have a substantial problem with delay of gratification. When the delay between response and reward was too long, the greater magnitude of the delayed reinforcer was not enough to entice them to choose that lever. Rather, the delivery of the reinforcer needed to occur soon after lever selection, so they choose the smaller, but immediate reward. As mentioned by Sjoberg et al. DD performance by SHRs strongly supports Dynamic Developmental Behavioral Theory, which argues the salience of a reinforcer decreases as it is separated in time from the response made to achieve it (Sagvolden et al., 2005a; Sjoberg et al., 2021).

Notably, in the version of the task conducted above, a subsequent trial did not begin until the end of the delay period, regardless of which lever was selected. In other words, because rats were required to “wait” for the next trial to start even if they picked the immediate lever, it is possible the Lphn3 KO rats may have opted to pick the larger magnitude reward. We decided to investigate this possibility in Experiment II.

Research that has examined whether consistency in trial length affects choice of the lever associated with the larger delayed reward has provided mixed results. For SHRs, the length of the inter-trial interval (ITI) does not appear to influence choice behavior within trials (Sjoberg et al., 2021). However, among ADHD children, when a subsequent trial begins as soon as the reinforcer from the previous trial is delivered/retrieved, this increase in relative response rate has been shown to shift an even greater percentage of responding to the immediate lever, thereby minimizing the impact of reward magnitude (Marco et al., 2009). This finding has also been shown to occur in research animals, especially when the post-reward delay was cued (Pearson et al., 2010). In Experiment II, we incorporated a 1-s ITI immediately after delivery of either the immediate (1 pellet) or delayed (3 pellet) reinforcer. Because previous research has already shown the length of the ITI does not appear to influence choice behavior in SHRs (Sjoberg et al., 2021), we only tested Lphn3 KO and WT rats in Experiment II. We predicted that the Lphn3 KO rats would choose the small, immediate reward more often that the larger, delayed reward relative to the WT rats, thereby demonstrating an increase in impulsive choice associated with an increase in the rate of reinforcement.

The inclusion criterion was the same as for Experiment 1. Data from 2 male Lphn3 KO rats and 3 Lphn3 WT rats (2 male, 1 female) were not included as they did not demonstrate 60% choice for the larger reward at the 0 s delay during the final maintenance phase. Thus, 20 Lphn3 KOs (9 male, 11 female) and 22 Lphn3 WTs (10 male, 12 female) were included in the final analyses. Greenhouse–Geisser corrections were again used for sphericity violations. Analysis of the percent choice revealed that discounting was evident, as the percent choice for the larger reward decreased overall with increasing delay [F(2.034, 77.301) = 337.275, p < 0.001]. However, analysis of the percent choice for the larger reward did not reveal a significant main effect of genotype [F(1, 38) = 1.462, p = 0.234] or sex [F(1, 38) = 0.007, p = 0.934], nor any significant genotype- or sex-related interactions (see Figure 5). The analyses of the slope and area under the discounting curve also did not reveal significant main effects of genotype [F(1, 38) = 0.162, p = 0.690 and F(1, 38) = 2.070, p = 0.158, respectively], nor any other significant genotype- or sex-related differences (see Figure 6).

The results of Experiment II indicated that removing the post-reward buffer following selection of the lever associated with a smaller, but immediate reward did not differentially affect DD behavior for either genotype. Thus, the rate of reinforcement did not appear to affect impulsive choice in the KO rats. These results are in line with previous research demonstrating that the length of the ITI had little influence on impulsive choice in SHRs (Sjoberg et al., 2021).

Overall, SHRs had increased impulsive choice due to their inability to delay gratification after a response to obtain a reinforcer had been elicited. Lphn3 KO rats, on the other hand, did not appear to exhibit impulsive choice, even when the option to increase the rate of reinforcer delivery was available. Thus, while both ADHD models exhibit hyperactivity (Russell, 2011; Sagvolden and Johansen, 2012; Regan et al., 2019) and impulsive action deficits (Sable et al., 2021; González-Barriga and Orduña, 2022), impulsive choice appeared to be differentially affected between the models. Notably, our previous research found that while both models exhibited a deficit in impulsive action, the degree of impairment was much more profound in the SHRs than in the Lphn3 KO rats (Sable et al., 2021). Thus, the overall degree of impulsivity appears to be much more substantial in the SHRs.

Dopamine regulation within the PFC is critically involved in impulsive behavior (Logue and Gould, 2014), and medications targeting the dopamine system are routinely prescribed to ADHD patients in an attempt to reduce impulsive behavior (Arnsten and Pliszka, 2011; Sharma and Couture, 2014). However, as previously mentioned impulsivity is a multi-faceted construct and behavioral deficits on tasks of impulsive choice do not always coincide with deficits on tasks of impulsive action (or vice-versa) in rats or in humans (Solanto et al., 2001; Broos et al., 2012; van den Bos et al., 2014).

Notably, there appears to be some degree of regional specificity that mediates impulsive action versus impulsive choice. While this is not yet entirely understood, in human subjects, gray matter volume in the right frontal pole (RFP) and left middle frontal gyrus (LMFG) were predictive of DD performance, while gray matter volume in the right inferior frontal gyrus (RIFG), supplementary motor area (SMA), and anterior cingulate cortex (ACC) predicted performance on an impulsive action task (Wang et al., 2016). Preclinical research also suggests the RIFG mediates impulsive action along with the ventrolateral prefrontal cortex (VLPFC), while the dorsal lateral prefrontal cortex (DLPFC) and is heavily involved in mediating impulsive choice (see Kim and Lee, 2011; Bari and Robbins, 2013 for reviews).

Our observed differences among the ADHD animal models has the potential to promote a better understanding of the underlying mechanism(s) responsible for the differential behavioral effects observed. The discrepant results between the SHRs and Lphn3 KO rats presented above suggest more widespread disruption of frontal cortical regions involved in both impulsive action and impulsive choice in SHRs, with disruption limited only to those regions involved in impulsive action in the Lphn3 KO rats. Ongoing research in our lab is currently investigating this possibility. These findings will be very important as they will have the potential to inform medication development, leading to a more targeted approaches to curb the facets of impulsivity that an ADHD individual may present (i.e., impulsive actions and/or impulsive choice), while sparing those that are not affected. This would undoubtedly reduce side effects and increase compliance.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

This animal study was reviewed and approved by the University of Memphis, Institutional Animal Care and Use Committee–protocol #0875.

MC: data collection, analysis, and manuscript preparation. AB, OH, and HN: experimental design, data collection, and manuscript preparation. SR: experimental design, data collection, and analysis. MW, CV, and HS: research idea, experimental design, data collection, analysis, and manuscript preparation. All authors contributed to the article and approved the submitted version.