PQBP1: A New Player in Metabotropic Glutamate Receptor Signaling and Synaptic Plasticity

Group 1 metabotropic glutamate receptors (Gp1 mGluRs), comprising mGluR1 and mGluR5, are predominantly located at postsynaptic terminals. Gp1 mGluRs induce the activation of phospholipase C-γ, generate inositol triphosphate and diacylglycerol, and trigger a local Ca²⁺ rise and protein kinase C activation. It is well established that Gp1 mGluRs play broad roles in synaptogenesis, neuronal development, learning and memory, and neurodegeneration. The best-characterized function of Gp1 mGluRs is that they induce the long-term depression (LTD) and long-term potentiation (LTP) of synaptic strength in cerebellar Purkinje cells, the hippocampus, neocortex, dorsal and ventral striatum, and spinal cord [1]. It is noteworthy that mGluR-LTD is involved in many brain diseases, including Alzheimer's disease, Parkinson's disease, drug addiction, mental retardation, and autism. For example, the absence of mGluR-LTD results in a shift toward LTP within the striatum, and this leads to enhanced activity of the indirect pathway and excessive inhibition of movements in Parkinson’s disease [2]. Another example is that mGluR-dependent LTD is facilitated by amyloid β [3], which may partly explain the loss of structural and functional synapses and spines, and the early learning and memory deficits in Alzheimer’s disease [4]. For clinical trials targeting brain disorders related to Gp1 mGluRs, pharmaceutical companies have made great efforts to synthesize positive and negative allosteric modulators that sensitize or desensitize Gp1 mGluRs. Unfortunately, most of the trials failed or did not achieve the ideal effect, because the functional manipulation of Gp1 mGluRs has unpredictable effects on the development and/or function of neurons, indicating insufficient understanding of the pathogenesis of Gp1 mGluR-related diseases.

Renpenning syndrome is a type of X-linked intellectual disability that shows significant growth defects such as microcephaly, short stature, small testes, and specific facial dysmorphism [5]. It has been demonstrated that mutations of the polyglutamine-binding protein 1 (PQBP1) gene in humans are associated with Renpenning syndrome, and the depletion or knockdown of Pqbp1 in mice result in a microcephaly phenotype and impaired cognition. Although PQBP1 is predominant in the nucleus, it may function in the cytoplasm as well [6]. However, its mechanism of action remains unknown. To answer this question, a recent study demonstrated that deletion of Pqbp1 interrupts the eEF2K/eEF2 pathway and impairs mGluR-LTD and related behaviors [7]. This work provides convincing data to not only reveal the mechanism underlying Renpenning syndrome but also identify PQBP1 as a new player in mGluR signaling and synaptic plasticity.

It has been shown that Arc/Arg3.1 is translationally induced by mGluR activation and is essential to mGluR-LTD [8]. An increase in Arc/Arg3.1 translation requires eEF2K that binds to mGluRs but dissociates upon mGluR activation. eEF2K then phosphorylates eEF2 and acts to increases Arc/Arg3.1 translation [8]. While this study indicates a critical role of eEF2-Arc/Arg3.1 signaling in mGluR-LTD, it has been unknown for more than a decade whether this pathway is related to diseases. The study by Shen et al. sheds new light on this question [7]. They found that PQBP1 protects eEF2 from eEF2K-mediated phosphorylation and then inhibits the de novo translation of Arc/Arg3.1. Pqbp1 deletion elevates the basal protein level of Arc/Arg3.1, and prevents a DHPG treatment-induced increase in de novo translation of Arc/Arg3.1 as well as mGluR-LTD. Using electrophysiological recordings in CaMKIIα-Cre;Pqbp1^fl/Y mice, Shen et al. compared the input-output relationship and paired-pulse facilitation [7]. Their results showed that basal transmission at excitatory synapses is normal in the hippocampus of CaMKIIα-Cre;Pqbp1^fl/Y mice; however, the protein synthesis-dependent DHPG-induced late-phase mGluR-LTD is impaired in CaMKIIα-Cre;Pqbp1^fl/Y mice [7]. The authors further found that CaMKIIα-Cre;Pqbp1^fl/Y mice are defective in spatial recognition and object recognition tasks. Interestingly, re-introduction of the eEF2-interacting domain of PQBP1 rescues mGluR-LTD and spatial learning [7]. Interference peptides corresponding to the residues of PQBP1 in the interaction with eEF2 reduce the activity of eEF2, mGluR-LTD, and spatial learning [7]. In summary, Shen et al. have identified a novel regulator for mGluR-LTD and spatial learning, in which PQBP1 binds non-phosphorylated eEF2 and protects eEF2 from eEF2K-mediated phosphorylation (Fig. 1) [7]. Importantly, both PQBP1 and eEF2 are ubiquitously expressed and their interaction is preserved in different cell types. Hence, the current work may offer an excellent therapeutic target for other diseases, such as various neural diseases, viral infection, and cancer [9], as well as Renpenning syndrome.

Fig. 1.

Schematic summarizing the new model of how PQBP1 regulates eEF2-Arc/Arg3.1 signaling and mGluR-LTD (Dyn, dynein). This figure was modified from references [7, 8].

The exciting findings by Shen et al. yet raise several questions that need to be investigated in the future [7]. First, it is unknown whether CaMKIIα-Cre;Pqbp1^fl/Y mice have deficits in motor learning while they show normal motor ability. This question is important because impaired mGluR-LTD in the cerebellum causes defects in motor learning behaviors, including the eyeblink conditioning reflex and the vestibulo-ocular reflex. Second, is there a synergistic effect between PQBP1 and fragile X mental retardation protein (FMRP) (Fig. 1)? Previous work has shown that the translation of Arc/Arg3.1 is disrupted in Fmr1-knockout mice, suggesting the coordinated control of FMRP on de novo synthesis of Arc/Arg3.1 and mGluR-LTD [8]. This question becomes more interesting as the same group reported that PQBP1 mutants promote the degradation of FMRP [10]. Third, PQBP1 is a multifunctional protein that is involved in general transcription and splicing in the nucleus, so PQBP1 mutations may also contribute to the defective behaviors of patients. In fact, some patients carrying PQBP1 mutations that do not affect the interaction with eEF2 also show clinical features. Therefore, the detailed mechanisms underlying Renpenning syndrome need further investigation.