The precise mechanisms of action of PBA are

The precise mechanisms of action of 4-PBA are as yet ambiguous. Its capacity as a chemical chaperone has been thoroughly documented for the ABC transporter family in particular (Prulière-Escabasse et al., 2007; Iram and Cole, 2014; Gordo-Gilart et al., 2016; Pomozi et al., 2017). 4-PBA has been classified as a hydrophobic chaperone, which interacts with hydrophobic segments of unfolded proteins and protects them from aggregation. In addition, 4-PBA is known to reduce ER stress through molecular mechanisms, which apparently affect many levels of regulation. Some studies allude to synergistic actions of 4-PBA, such as e.g. downregulation or modulation of HSC70 expression (Rubenstein and Zeitlin, 2000), increase in HSP70 levels (possibly as a secondary effect of decreasing HSC70 levels), reversible histone deacetylase (HDAC) class I and IIa inhibition (Cousens et al., 1979; Konsoula and Barile, 2012), and signaling via Elp2 and STAT-3 (Suaud et al., 2011). 4-PBA-induced inhibition of HDACs leads to transcriptional regulation of genes in the unfolded protein response (UPR) system. This in turn induces the synthesis of molecular chaperones (HSPs) and downregulation of protein synthesis (Cortez and Sim, 2014). In fact, the neuroprotective effects of 4-PBA may be accounted to epigenetic regulators related to HDAC inhibition related to enhanced synaptic plasticity, learning and memory (Konsoula and Barile, 2012). Accordingly, 4-PBA showed beneficial effects in various cell culture and animal models of neurodegenerative disorders, e.g. Alzheimer's (Ricobaraza et al., 2009, 2012) and Parkinsons's disease (Inden et al., 2007; Ono, 2009). It is worth mention that several pathological variants found in the hCRT-1 also trigger folding diseases at equivalent positions in other SLC6 family members. For instance, the replacement of glycine 132 by valine in hCRT-1 (i.e G132V-hCRT-1 variant) causes severe mental retardation in boys (Lion-François et al., 2006), and the mutation to glutamine of the equivalent glycine in Drosophila melanogaster DAT (i.e. G108Q-dDAT), leads to a sleepless phenotype reminiscent of a DAT knock-out in flies (Kasture et al., 2016, 2017). Similarly, the mutation of P554 to leucine (L) elicits misfolding in both transporters and consequently leads to disease: infantile/juvenile parkinsonism-dystonia in hDAT (Kurian et al., 2009, 2011; Ng et al., 2014; Asjad et al., 2017), and severe mental retardation in hCRT-1 (Rosenberg et al., 2004, 2007; Mancini et al., 2005). The same holds true for the mutation to leucine (L) of P390 in hCRT-1 and the equivalent residue P395 in hDAT. P544 and P554 are located within TMD11 and the extracellular loop between TMDs 11 and 12, respectively. The proline residue P390-hDAT/P395-hCRT-1 is located in TMD8. At least seven CTD-causing hCRT-1 mutations reported to date are substitutions of prolines by other residues, leucine being the most frequent substitution (i.e. P382L, P295R, P390L, P434L, P544L, P554L and P597T). Prolines are known to induce kinks in proteins and their replacement by other residues is likely to perturb the protein structure, which in many instances accounts for folding defects. Proline is the only natural amino geldanamycin whose side chain connects to the protein backbone twice, forming a five-membered nitrogen-containing ring. The main chain conformations that can be adopted by prolines are hence limited in comparison to any other amino acid. On a clinical level, CTD encompasses a vast range of symptoms contingent on the particular variant. Some mutations elicit much graver phenotypes than others. The R391W variant is accompanied by speech delay, hyperactivity and abnormal behaviour, as well as myoclonic seizures (Mencarelli et al., 2011). Another CTD variant associated with epilepsy is hCRT1-P544L; it is manifested by moderate mental retardation, delayed language and motor development, but accompanied by multifocal epileptic waves (Mancini et al., 2005; Betsalel et al., 2012). The P554L variant leads to hypotonia, severe intellectual disability and drug-resistant epilepsy, and sudden death of one patient at the age of 17 (Rosenberg et al., 2004; Nozaki et al., 2015). One clinical study reported significantly higher levels of guanidinoacetic acid (GAA) in the brains of CTD patients, with creatine being virtually absent (Sijens et al., 2005). CRT-1 is known to play a role in the transport of GAA in the brain and the blood cerebrospinal fluid barrier, albeit with a K value 10-fold greater than that of creatine (Tachikawa et al., 2008). The lack of functional CRTs in CTD leads to cerebral accumulation of GAA. Because GAA is an endogenous convulsant, this may account for at least some of the epileptic seizures observed in many CTD patients. At this point, none of the currently available therapeutic interventions can ameliorate the clinical outcome in any of the diagnosed CTD patients. One likely rationalisation for this is that creatine is released from central neurons and exerts its action as a neuromodulator (Almeida et al. 2006). In effect, hCRT-1 must play a major role in taking up the previously released creatine (Peral et al. 2010) and/or releasing creatine from neurons (Mak et al. 2009). Consequently, even though supplementation with creatine itself might improve creatine levels, it cannot compensate for the neuromodulatory action accomplished by the plasmalemmal hCRT-1. We hence postulate that in CTD, hCRT-1 is absent from its physiological site of action in neuronal cells. The localization of hCRT-1 in the human brain was mapped out using immunohistochemistry: the transporter is robustly expressed in the large projection neurones of the brain and spinal cord (i.e in the pyramidal neurones in the cerebral cortex, Purkinje cells in the cerebellar cortex and motor neurones of the somatic motor and visceromotor cranial nerve nuclei and the ventral horn of the spinal cord), while only negligible levels of CRT-1 were mapped to substantia nigra and locus coeruleus, i.e. regions typically implicated in neurodegenerative diseases (Lowe et al., 2015). The absence of functional CRTs in brain cells portends that the energy normally conferred by creatine is no longer attainable, since creatine is taken up and/or released from these cells via membrane-bound CRTs. Since many CTD-triggering variants of hCRT-1 are confined to the ER due to protein folding defects, 4-PBA treatment may be the key to tackling CTD, by rectifying their folding, trafficking and transport activity. Proper trafficking, e.g. axonal targeting of SERT in rat dorsal raphe neurons is specified by SEC24C-dependent ER export (Montgomery et al., 2014). SERT mutants, which are either folding-deficient or harbour a disrupted SEC24C-recognition motif are also trapped in the ER (Sucic et al., 2011, 2013). The same holds true for misfolded variants of hDAT associated with infantile Parkinsonism/dystonia; several of which were functionally rescued by pharmacochaperoning, both in HEK293 cells and in living flies (Kasture et al., 2016; Asjad et al., 2017).