Regulation

ATP, and divalent cations

The activity of hSR is modulated at different levels (catalysis, cellular localization, degradation) emphasizing the pivotal role played by this enzyme in the glutamatergic transmission in the central nervous system. The first modulators of hSR to be discovered were ATP and divalent cations (21, 34) (Table 1 and Figure 4). ATP and Mg²⁺ act synergistically and induce a 5-fold increase in the racemisation reaction. In the absence of the cofactors, no formation of pyruvate from L-serine was detectable (7). However, in the presence of ATP and Mg²⁺ the ratio between pyruvate and D-serine produced from L-serine is four. Subsequently, it was demonstrated that SR is activated also by nucleotides different from ATP, like ADP, GTP and, to a minor extent, CTP and UTP, with nucleosides monophosphates being almost ineffective (21). Ca²⁺ and Mn²⁺ also activate SR  (20, 21, 34) (Table 1), with Ca⁺⁺ showing an affinity similar to that of Mg²⁺. However, the physiological significance of a Ca²⁺-dependent regulation of SR is questionable since SR in resting cells is already fully saturated by Mg²⁺ (34) and an increase in Ca²⁺ concentration induced by a stimulus would probably not result in a further activation of SR. Thus ATP and Mg²⁺ are believed to be the in vivo natural effectors of mammalian SR (34) and bind to the enzyme as Mg-ATP complex. Regulation by nucleotides was not detected in SRs from plants (57, 58), whereas it was demonstrated for some bacterial dehydratases where AMP and/or ADP binding affects the oligomerization state and the activity, e. g., threonine dehydratase (59, 60). In the case of mammalian SR, ATP binding does not cause any effect on the aggregation state of the protein (36). In addition, ATP is not hydrolysed during catalysis and, consequently, non-hydrolysable analogs are effective (21, 34, 55). ATP and Mg²⁺ should thus exert their function as allosteric effectors of SR. In SpSR, the binding site of ATP  was found to be

connected to the active site by a conserved residue, namely Gln89 (18). Mg-ATP binding might act causing a conformational change that is transmitted to the active site (12, 20). The resulting effect is a decrease of the K_M for L-serine with negligible effects on k_cat (see also (12) for further comments on this point).

Nitrosylation and phosphorylation

ATP binding and the resulting activation of SR seem to be modulated by cysteine nitrosylation (Table 1). In fact, one out of the seven cysteine residues of hSR (Cys2, Cys6, Cys46, Cys113, Cys128, Cys217 and Cys309), namely Cys113, proved to be nitrosylated by S-nitroso glutathione (GSNO) (61). This post-translation modification leads to a decrease of SR activity with respect to controls and the SR mutant Cys113Ser showed no change in activity upon exposure to GSNO. ATP was found to compete with nitrosylation and, concurrently, nitrosylation was found to affect ATP binding (61). Because Cys113 is localized within the ATP binding site, Cys113 nitrosylation likely interferes with ATP binding and enhancement of SR activity, thus explaining the effect of Cys113 nitrosylation on SR activity. It has been speculated that the reaction of NO with Cys113 represents a feedback regulation of NMDAR activity. In fact, it is known that NMDAR activation stimulates neuronal NO-synthase. NO can react with both SR and DAAO causing a decrease in activity and an activation, respectively. Both chemical modifications result in a decrease of D-serine level, that, in turn, leads to a decrease of NMDAR-dependent neurotransmission (61).  It might also be that glutathione, and not NO, reacts with Cys113 (12, 13), accounting for the observed effects on activity. In fact, in a work where a different NO donor, DETA NONOate, was used no formation of thionitrosyl derivatives was detected (20). A definite answer might come from experiments using free NO and from a full characterization of cysteine reactivity of hSR. Furthermore, it has been reported that reducing agents, such as dithiothreitol or reduced glutathione (51, 61) are essential for SR activity (20) suggesting the detrimental effect of cystine formation. In the crystallization trials for hSR, in order to improve expression yields, Cys2 and Cys6 were both replaced with aspartate residues (17). A disulfide bond between Cys6 and Cys113 was identified in 3-morpholinosydnonimine hydrochloride (SIN-1)-treated SR monomer and dimer (23) whereas Cys217 was found to be involved in metal binding (31).

Activation by ATP participates in a more complex scenario, where subcellular localization of SR also plays an important role. A recent work has demonstrated that in resting glial cells SR is inactive and localized in the cell membrane (62). SR associated with the membrane is inactive because the ATP binding site is occupied by phosphatidylinositol (4, 5) bisphosphate (PIP2) (Table 1). Occupation of the ATP binding site by PIP2 was demonstrated in vitro by competition experiments, where the effect of Mg-ATP on SR activity decreases in the presence of PIP2 (62). Stimulation of astrocytes by glutamate released in the synaptic space through the metabotropic glutamate receptors mGluR5, causes the activation of phopholipase C that relieves inhibition on SR by degrading PIP2. Activation of SR increases the intracellular D-serine concentration that is subsequently released into the synaptic space where it will bind to NMDA receptors activating glutamatergic transmission. In a work on rat primary neuronal cultures the association of SR to cell membranes was also demonstrated. However, enzyme activity was proposed to be modulated by a completely different mechanism (63). In the absence of NMDAR stimulation SR is mainly located in the cytosol, whereas SR is associated to membranes when NMDAR are activated. Association to the membrane is mediated by palmitoylation of the protein and strengthened by phosphorylation of Thr227. Because membrane-bound SR is inactive (Table 1), the biological role of membrane association in neurons should be the prevention of NMDAR overactivation in vicinal cells. By LC-MS analysis another site of phosphorylation, namely Thr71, was discovered (64) (Table 1). The main effect of this post-translational modification on enzyme catalysis seems to be an increase in V_max, Further investigations are needed to fully establish the  role of  Thr71 phosphorylation.

Read Also

• Biji fenugreek: Makanan super untuk kesehatan dan kebugaran
• Spirulina menjanjikan dalam memerangi penyakit jantung dan diabetes
• Penelitian memberikan bukti bahwa penggunaan antidepresan pada kehamilan mempengaruhi perkembangan otak anak
• Pola makan kaya flavonol dikaitkan dengan penurunan angka kematian dan risiko penyakit, menurut penelitian
• Bagaimana Genomik Spasial Merevolusi Pemahaman Kita tentang Kesehatan dan Penyakit
• Mengevaluasi kemanjuran iradiasi seluruh payudara ultrahipofraksionasi dalam uji coba acak FAST-Forward
• Merokok, minum minuman keras, dan kebiasaan makan dikaitkan dengan risiko kanker kepala dan leher, demikian temuan penelitian
• Keragaman bakteri usus terkait dengan berat badan, atlet menunjukkan profil yang lebih sehat
• Memerangi Bugs dan Blues: Interaksi antara Infeksi dan Emosi
• Imunoterapi sel T CAR pada kanker kolorektal
• Terobosan metode gene silencing kolesterol selama satu tahun pada tikus, tanpa kerusakan DNA
• Nutrigenomik: Bagaimana Gen Anda Mempengaruhi Kebutuhan Gizi dan Kesehatan Anda

Interactome

In addition to allosteric effectors and post-translational modifications, SR function is regulated in vivo by changes in subcellular localization and interaction with specific proteins. Indeed, SR appears to have been molded by evolution to interact with a range of other proteins and subcellular structures, thanks, for example, to the occurrence of a peculiar C-terminus motif (36) that characterizes the human and murine enzyme but is absent in rat and bovine counterparts (9, 65). Although the complete set of interactors of SR has probably yet to be described, several interactions have been uncovered using yeast two-hybrid screening.  Partner proteins were shown to modulate SR function by directly affecting the specific activity of the enzyme but also by altering its subcellular localization and its half-life via the control of the rate of protein degradation by the ubiquitin-proteosome system.

In a pioneering study, Kim et al (66) showed that rat or mouse SR interacts with the protein Glutamate Receptor Interacting Protein 1 (GRIP1) and that such an interaction is instrumental to a glutamate-triggered SR activation process. GRIP1 is a large synaptic protein that interacts with AMPA glutamate receptors and contains several PDZ domains (67). One of these, PDZ6, was found to bind to the C-terminal region of SR (36, 66). Activation of AMPA receptors releases GRIP1 from its attachment sites, making it available to bind SR. In turn, this enhances SR activity and D-serine release (36, 66) (Table 1). Such a mechanism contributes to the physiologic regulation of cerebellar granule cell migration by SR (66).

SR was also found to bind to Protein Interacting with C-Kinase (PICK1) (68). This is another PDZ domain-containing scaffold that can bind to a large number of interactors, including AMPA receptors (69). Again, interaction with SR involves the PDZ domain of PICK1 and the C-terminus of SR (68). PICK1 knockout mice showed decreased concentrations of D-serine in the brain, suggesting that PICK1 may be involved in the regulation of SR function in a spatially and temporally specific manner (70) (Table 1). However, the mechanism by which this effect takes place is unclear, since PICK1 does not directly activate SR and the level of SR protein is unchanged in the PICK1-deficient animals (54, 70). A recent study has pointed out the involvement of protein kinase Ca in the PICK1-dependent modulation of mouse SR activity (71).

Another characterized interactor of SR is Golgin subfamily A member 3 (GOLGA3) (72). This is a protein with unknown function that associates to the cytosolic face of the Golgi apparatus. In contrast to the interactions described above, it was found that GOLGA3 binds to SR at the N-terminus of the racemase (72). Interaction with GOLGA3 decreases SR protein degradation through the ubiquitin pathway and indirectly affects D-serine levels (72). The study that characterized the interaction between SR and GOLGA3 also showed that a fraction of SR was strongly bound to the membrane fraction and resisted even washing with high salt concentration (72). The initial study also showed, by colocalization experiments, the presence of SR in the perinuclear region corresponding to the Golgi apparatus of both neurons and astrocytes, suggesting that GOLGA3 could mediate SR binding to the Golgi membrane (72). However, a subsequent study showed that in neurons most membrane-bound SR is located at the plasma membrane and dendrites (63), an interaction mediated by palmitoylation of the protein. As described above, NMDAR activation in primary neuronal cell cultures induces translocation of SR from the cytosol to intracellular and dendritic membranes, where it is inactive toward D-serine synthesis, thus providing a mechanism for feedback regulation of SR and NMDAR activity (63). In glial cells, conversely, the association of SR to membranes depends on SR interaction with PIP2 (62). Mutations in a potential lipid-binding region of SR abolished SR binding to the membranes and increased the specific activity of the isolated enzyme (62).

A further interaction of SR has been found with Disrupted-In-Schizophrenia-1 (DISC1), one of the few proteins whose mutations have been unquestionably linked to the genetics of schizophrenia (73). A very recent work by Ma et al. (74) has shown that the scaffold protein DISC1 binds to and stabilizes SR. Mutant DISC1 truncated at its C-terminus failed to bind to SR, facilitating ubiquitination and degradation of SR, with a consequent decrease in D-serine production (Table 1). These results suggest a very direct pathway by which DISC1 can modulate the production of D-serine and NMDA neurotransmission, relevant to the pathophysiology of schizophrenia and other neuropsychiatric disorders.