Protein Tyrosine Phosphatase β/ζ and alcohol use disorder: A Commentary

The need to find new pharmacological targets for treating alcohol use disorder (AUD) has been an extensive effort in alcohol research. Changes in immune function due to AUD may represent an exploitable target for developing new medications to treat AUD, since the cytotoxic effect of alcohol has shown to alter the expression of inflammatory mediators in the periphery and in the central nervous system (CNS) (Leclercq, de Timary, Delzenne, &Starkel, 2017).

The elegant study provided by Calleja-Conde and colleagues imparts valuable support towards targeting pleiotrophin (PTN) and midkine (MK), two cytokines involved in repairing damaged brain tissue. These two cytokines selectively bind to receptor protein tyrosine phosphatase (RPTP) β/ζ (PTPRZ1, RPTP β, PTP ζ) and inhibit its enzymatic activity (Herradon & Perez-Garcia, 2014).

The work itself is supported by the strong scientific premise that the receptor for PTN and MK, RPTPβ/ζ, is extensively expressed in parts of the brain, such as the prefrontal cortex (Abernathy, Chandler, & Woodward, 2010) and the amygdala (Gorka, Fitzgerald, King, & Phan, 2013) which are relevant to the development of AUD. Furthermore, MK protein expression is up-regulated both in mice bred for high alcohol consumption (Mulligan et al., 2006) and in the prefrontal cortex of individuals with AUD (Flatscher-Bader & Wilce, 2008).

The Calleja-Conde et al work has several strengths. First, they have designed, synthesized and tested MY10, a new CNS penetrant compound, that acts as the two cytokines: PTN and MK. Selectively, MY10 inhibits the tyrosine phosphatase activity by interacting with RPTPβ/ζ intracellularly (Pastor et al., 2018). The scientific rigor of the Calleja-Conde et al work is further supported by their prior research showing that MY10 reduced alcohol consumption in mice, with limited effects on the alcohol-induced ataxia and the potentiation of the sedative effects of alcohol. Their previous work evidenced a decrease in alcohol consumption that was exclusive for alcohol, as sucrose consumption was not influenced by MY10 administration (Fernandez-Calle et al., 2019; Fernandez-Calle et al., 2018).

The Calleja-Conde et al work tested MY10 in rats, a stronger preclinical model to study addictive behaviors than the mice model (Spanagel, 2017). They also used two validated alcohol consumption paradigms: the alcohol operant self-administration and the Drinking In the Dark-Multiple Scheduled Access paradigm (DID-MSA). The operant self-administration procedure was used to test the voluntary, reinforcing and motivational properties of alcohol within a constrained time to alcohol access. The DID-MSA evaluated effects felt due to alcohol over an extended period of time where rats could choose whether or not to drink.

In the DID-MSA model, rats treated with MY10 showed an overall decrease in alcohol consumption (28.1%) compared to the vehicle arm. This effect was specific for alcohol since there was no effect on the consumption of saccharine solution. In the alcohol preference analysis, the authors also demonstrated that, from baseline, the rats in the MY10 arm decreased alcohol preference. Furthermore, the rats in the MY10 arm showed a lower alcohol preference compared to the rats in the vehicle arm at the third day of medication treatment. When MY10 was discontinued, data also show a beautiful rebound effect during the washout period. Finally, as observed in the mice model, water consumption was not affected by the MY10 administration, which further substantiated the selectivity of MY10 for alcohol.

Finally, to provide additional support of the role of cytokines in the context of AUD, the authors elegantly analyzed the expression profile of the RPTPβ/ζ-related gene after administration of MY10 on genes in the PTN signaling pathway associated with alcohol-related behaviors. In the rat control arm (vehicle) after alcohol administration, the gene expression analysis showed that protein tyrosine phosphatase (RPTP) β/ζ Ptprz1 and anaplastic lymphoma kinase (Alk), were downregulated in the PFC. Conversely, the rats in the MY10 arm, restored the levels of expression of Ptprz1 and Alk, suggesting that MY10 may decrease alcohol consumption by preventing their downregulation induced by alcohol.

The only limitation of the Calleja-Conde et al work is that they tested only one dose of MY10 (100mg/kg) in the DID-MSA paradigm. This dose was reliable with their former DID work in mice (Fernandez-Calle et al., 2018) and it was accurately tested preceding the MY10 dose-response curve in the voluntary alcohol administration procedure. However, the 100 mg/kg dose was the highest dose tested in this experiment and it produced a significant decrease in alcohol consumption only on the second day of MY10 administration. Questions inquiring if a higher dose of MY10 would have either a stronger effect earlier in treatment and/or a higher cumulative alcohol consumption reduction are left unanswered.

Additionally, the authors found a modest overall treatment effect (26.1%) compared to vehicle in the operant self-administration study design. Cleverly, they tested the hypothesis that the inhibitory effect of MY10 on alcohol consumption was stronger during extended alcohol access (as observed in the DID-MSA procedure). However, when they protracted the alcohol access time in the operant self-administration procedure from 30 minutes to three 1-hour access sessions, they found a significant effect on time, but no effect on treatment. Similarly, questions remain if higher doses of MY10 could have engendered an effect on treating alcohol consumption when compared to vehicle.

At a 100 mg/kg dose, rats’ locomotor activity was not impaired and a higher MY10 dose should be evaluated in future work. It will be very valuable for translational purposes to determine if the higher dose of MY10 has a profound effect immediately after administration. A higher dose of MY10 should be administered under the operant self-administration procedure, with extended periods for alcohol access, to determine if alcohol consumption is reduced due to prior evidence suggesting that among preclinical models, longer exposure to alcohol paradigms has high predictive validity in individuals with AUD (Koob et al., 2003).

Finally, sex is a biological variable and should be considered in immunological studies within the alcohol research field. The effects of MY10 should be evaluated in both sexes, since sex differences occur in both innate and adaptive immune responses (Klein & Flanagan, 2016). Furthermore, a difference in the distribution of RPTPβ and its ligand PTN has been observed between male and female mice (Adthapanyawanich et al., 2013).

Conclusion and Future Directions

Recently, converging evidence identified specific alcohol-related phenotypes that underlie individual alcohol-related behavior(Morozova, Goldman, Mackay, & Anholt, 2012). This observation not only extends our knowledge of neurobiological adaptations occurring among the brains of alcohol induced preclinical models, but also highlights the complexity of the diverse spectrum of AUD (Heilig, Goldman, Berrettini, & O'Brien, 2011).

In this regard, testing medications in the early stage of development among different preclinical alcohol paradigms supports research efforts on the use of bidirectional translational paradigms. This would enhance identification of a specific subtype of the AUD population that may benefit from a specific pharmacological action of a drug. Calleja-Conde et al used an ad hoc alcohol self-administration protocol to better identify which AUD subgroup may benefit from the administration of a medication targeting the inhibition of receptor protein tyrosine phosphatase (RPTP) β/ζ (Ptprz1). At the early stage of a drug’s development, the authors astutely used an animal model that was able to test alcohol-related key variables (enhanced exposure to alcohol and or the reinforcing/motivational properties of alcohol) identified in individuals with alcohol use disorder (AUD).