Behavioral impairment in SHATI/NAT8L knockout mice via dysfunction of myelination development

We have identified SHATI/NAT8L in the brain of mice treated with methamphetamine. Recently, it has been reported that SHATI is N-acetyltransferase 8-like protein (NAT8L) that produces N-acetylaspatate (NAA) from aspartate and acetyl-CoA. We have generated SHATI/NAT8L knockout (Shati−/−) mouse which demonstrates behavioral deficits that are not rescued by single NAA supplementation, although the reason for which is still not clarified. It is possible that the developmental impairment results from deletion of SHATI/NAT8L in the mouse brain, because NAA is involved in myelination through lipid synthesis in oligodendrocytes. However, it remains unclear whether SHATI/NAT8L is involved in brain development. In this study, we found that the expression of Shati/Nat8l mRNA was increased with brain development in mice, while there was a reduction in the myelin basic protein (MBP) level in the prefrontal cortex of juvenile, but not adult, Shati−/− mice. Next, we found that deletion of SHATI/NAT8L induces several behavioral deficits in mice, and that glyceryltriacetate (GTA) treatment ameliorates the behavioral impairments and normalizes the reduced protein level of MBP in juvenile Shati−/− mice. These findings suggest that SHATI/NAT8L is involved in myelination in the juvenile mouse brain via supplementation of acetate derived from NAA. Thus, reduction of SHATI/NAT8L induces developmental neuronal dysfunction.


Results
Expression of Shati/Nat8l mRNA increases with brain development in mice. We collected the whole brain for the measurement of Shati/Nat8l mRNA during brain development, because the brain development began from embryonic state and Shati/Nat8l mRNA is expressed in whole brain. To analyze the expression of Shati/Nat8l, we first quantified Shati/Nat8l mRNA levels in the whole brain on various days at the age of embryonic (E) 15 Deletion of Shati/Nat8l altered the MBP level in the brain of juvenile, but not adult mice. We next investigated whether the deletion of SHATI/NAT8L affected the myelin basic protein (MBP) level in the prefrontal cortex, which is involved in the myelination and emotional behaviors in mice 17,18 . To account for several isoforms of MBP, we detected protein bands from 10 kDa to 25 kDa in the Western blot analysis as previous report 20 . Immunohistochemical analysis and Western blotting showed the expression of MBP in the prefrontal cortex of juvenile (3 weeks old) mice was decreased in Shati −/− mice compared with that of Shati +/+ mice ( Fig. 1b and d; Shati +/+ mice = 1.00 ± 0.13, Shati −/− mice = 0.66 ± 0.09, t 4 = 4.754, p < 0.05). On the other hand, the expression of MBP in the prefrontal cortex of the adult (10 weeks old) mice was unchanged in Shati −/− mice (Fig. 1c,e; Shati +/+ mice = 1.00 ± 0.10, Shati −/− mice = 0.92 ± 0.13, t 4 = 2.483, n.s.). These findings indicate that SHATI/NAT8L plays an important role in the regulation of myelin state in the brain of juvenile mice.

Discussion
Although only SHATI/NAT8L was identified as N-acetyltransferase in brain 2,3 , NAA and NAAG are detected in the frontal cortex of mice brains. We also investigated the NAA and NAAG contents in Shati +/+ and Shati −/− mice by liquid chromatography-mass spectrometry (LC-MS), which is much more sensitive than HPLC methods 9,12 . NAA was markedly decreased and NAAG was completely knocked out in Shati −/− mice (Sumi, Tomohiro, Hatanaka, Nitta unpublished data). Although the synthase of NAA is considered to be only SHATI/NAT8L, the result from LC-MS experiments also indicates the role of an unknown enzyme in NAA production in the brain. Further analysis is needed to elucidate the mechanisms of NAA production.
In the present study, we observed expression of Shati/Nat8l mRNA increases with brain development in mice (Fig. 1a). This result indicates that the function of SHATI/NAT8L is important for the brain after birth. To investigate the function of SHATI/NAT8L in the brain development, we checked the level of MBP in the juvenile and adult brain. Interestingly, deletion of SHATI/NAT8L altered the MBP level in the brain of juvenile, but not adult, mice (Fig. 1d,e). Our result is consisted with previous report that the MBP level in the adult brain of Shati −/− mice is no change compared with Shati +/+ mice 23 . These results suggest that SHATI/NAT8L could be involved in the myelination in the juvenile mouse brain.
We have previously reported that a shorten immobility of Shati +/− mice in the forced swimming test was ameliorated by a single intracerebroventricular injection of NAA 12 . On the other hand, the same treatment of NAA could not completely normalize the behavioral deficits seen in Shati −/− mice, the reason for which is still unclear 12 . In the present study, we treated NAA into intracerebroventricules for adult Shati −/− mice repeatedly, but NAA did not improve behavioral deficit (Fig. 3b). These results indicate that treatment of NAA for adult mice dose not rescue the behavioral deficits of Shati −/− mice. We had also considered it is important to investigate the effect of NAA on rescuing myelination in juvenile mice. It was an important point whether treatment of NAA from jugular stage could rescue the behavioral deficits in adult Shati −/− mice. However, NAA could not be penetrated from periphery to brain by using intraperitoneal injection although acute and repeated NAA oral treatment did not show toxicity 24,25 . Moreover, it is technically difficult to inject to ventricles of mice in juvenile stage, since their brains are too small and brittle. NAA is metabolized to aspartate and acetate by ASPA in oligodendrocytes in the brain. Then acetate is converted to acetyl-CoA and used for lipid synthesis and myelination. Moreover, GTA is metabolized to acetate and rapidly distribute to the brain after oral administration as previous reports 21,22 . Hence, we investigated the roles of NAA in behavioral deficits using Shati −/− mice treated with GTA, the acetate trimester of glycerol, from P7 to P56. Interestingly, GTA treatment normalized MBP level in the brain of juvenile mice and ameliorated reduced social interaction caused by deletion of SHATI/NAT8L in adulthood ( Fig. 2). Furthermore, to investigate the critical period that affects the behavioral deficits in Shati −/− mice, we treated GTA for mice from P7 to P21 until weaning. GTA treatments improve reduced social interaction of adult Shati −/− mice. Furthermore, the level of acetate in the juvenile prefrontal cortex was found to be decreased in Shati −/− mice compared with Shati +/+ mice (Fig. 5c), suggesting that the presence of acetate in the juvenile period is important for social behavior.
There are several reports that impaired or delayed myelination in the prefrontal cortex induces reduced social interaction in adult mice 18,19 . Also absence of NAA and NAAG is involved in delayed myelination in patients with hypoacetylaspartia 26 . On the other hand, it was reported that BDNF signaling in the developmental brain is involved myelination 27,28 . Previously, it was reported that the levels of BDNF mRNA in the prefrontal cortex were decreased in Shati −/− mice 13 , GTA treatment for Shati −/− mice normalized the decrease of BDNF mRNA and MBP level (Figs 3c, 4a,b). Moreover, by using electron microscopy analysis, myelinated nerve filers of Shati −/− mice were decrease compared with that of Shati +/+ mice although g-ratio of each groups were not changed. (Fig. 4c,d,e). As shown in Supplemental Figs 5-7, the dysfunction of myelination did not induce apoptosis in Shati −/− mice. Hence, these results show possibility that the recovery of behavior deficits and delayed myelination in Shati −/− mice by GTA treatment is associated with normalization of BDNF mRNA level. There is no report that acetate or GTA could directly affect expression of BDNF. Therefore, we assume that the normalization of BDNF mRNA level in Shati −/− mice treated with GTA might indicate amelioration of neuronal activity due to impairment of myelination in the jugular stage of the Shati −/− mice. BDNF mRNA expression is regulated neuronal activity 29 .
We have previously reported that SHATI/NAT8L is associated with neurite elongation and the ATP synthetic pathway via NAA synthesis 14 . SHATI/NAT8L is expressed in the mitochondria of neuronal cells, and NAA synthesized by SHATI/NAT8L is associated with the tricarboxylic acid cycle related to metabolism in neurons 14 . Further, NAA is metabolized to acetate and aspartate in the oligodendrocytes. Hence, the ameliorative effect of acetate derived from GTA on the behavioral deficits is hypothesized that it acts directly at oligodendrocytes. In the present study, we checked the expression of ASPA and ATP citrate lyase mRNA and acetate contents to investigate the effect on utilization of NAA in oligodendrocyte (Fig. 5a-f). Surprisingly, the levels of ATP citrate lyase mRNA and acetate contents were decreased in the PFC of juvenile, but not adult Shati −/− mice. These results suggested that utilization ability of NAA was decreased in the oligodendrocytes of Shati −/− mice. The reduced levels of acetate in the brain of Shati −/− mice are consisted with previous report that knock-down of the NAA-cleaving enzyme reduces acetate levels in adipocytes 30 . On the other hand, it was reported that the levels of ATP citrate lyase mRNA is increased in the adipocytes of Shati −/− mice and this report is inconsistent with our results 31 . The reasons of discrepancy between ATP citrate lyase mRNA in the brain and adipocytes of Shati −/− mice is unclear at the present. We estimate that the differences between the organs cause the result, because ASPA expressed in the adipocytes, but not the neurons. Further study is needed to clarify the detail mechanism of this discrepancy.
The findings of the current study and those of previous studies show that deletion of SHATI/NAT8L alters MBP level in the brain of juvenile, but not adult mice, suggesting that SHATI/NAT8L is involved in myelination via its role in NAA synthesis. Furthermore, Shati −/− mice showed several behavioral deficits, and these deficits were ameliorated by GTA treatment during the juvenile stage, suggesting that the behavioral deficits occurred due to decreased acetate. These findings suggest that SHATI/NAT8L is involved in myelination in the juvenile mouse brain via supplementation of acetate derived from NAA. It is well known that defects in NAA metabolism result in impaired postnatal myelination, most notably in Canavan disease 23 , SHATI/NAT8L might be involved in brain development, especially, in myelination, and may be therapeutic targets for developmental disorders. The number of patients with developing disorders is much more than Canavan disease. The pharmaceutical therapy is required for developing disorders, but we have no means at the present. Then our results will contribute the development of the medical tools for developing disorders. The absence of NAA and NAAG is involved in delayed myelination in humans. Therefore, it is possible that these molecules participate in other developmental disorders 26 . We expect that SHATI/NAT8L will become a novel therapeutic target for the treatment of cryptogenic developmental disorders.

Materials and Methods
Animals. We have previously described the generation of Shati −/− mice 18 . Animals were housed in a room with a 12 h light/dark cycle (light cycle starting at 8:00 AM.). Food and water were available ad libitum. All experiments followed the National Institute of Health Guidelines for the Care and Use of Laboratory Animals and were approved by the committee for Animal Experiments of the University of Toyama (A2015PHA-23, G2015PHA-15).

Administration of glyceryltriacetate (GTA) and N-acetylaspatate (NAA). Glyceryltriacetate
(GTA; Wako, Osaka, Japan) treatment was performed as previously described 21,22 . GTA was treated orally to Shati +/+ and Shati −/− pups from day 7 after birth until day 14 at a dose of 4.2 g/kg. 5.8 g/kg GTA was administered from day 15 to day 21. After weaning (after day 22), the pups received GTA in their water (5% GTA by weight). Intracerebroventricular injection of NAA was performed as previously described 12 . Briefly, a microsyringe with a 28-gauge stainless-stell needle (3 mm in length) was used for the microinjection. The mice were lightly anesthetized and the needle was implanted into the lateral ventricle (AP −0.6 mm, ML +1.0 mm, DV −2.5 mm). NAA was solubilized in Saline to obtain a concentration of 20 µg/µL. The i.c.v. injection volume was 3 µL, 30 min before each Three-chambered social interaction test trail, and the injection speed was 20 s. Schedule of behavioral tests and sampling for brain tissues. All behavioral tests were performed from the age of 8-9 weeks old in the following order so as to reduce the stress on the mice; locomotor activity, Y-maze test, three-chambered social interaction test, and elevated plus maze test. After the behavioral tests, brain samples were collected and used for Western blotting or acetate assay. The brains used for the experiments with electron microscopy were separately prepared. Behavioral tests were finished during the ages of 9-10 weeks old, and sampling was performed when the mice became 10 weeks old (Fig. 2a).
Immunostaining of mice brains. Immunostaining was performed as previously described 14  Western blotting. Brains were isolated and cut into 1 mm-thick sections. The prefrontal cortex was isolated from the brain section and fractured in RIPA buffer (50 mM Tris-HCl pH 7.5, 152 mM NaCl, 5 mM EDTA, 1% TritonX-100, 0.5% sodium deoxy cholate, 1 mM PMSF, 2% protease inhibitor cocktail, and 1% phosphatase inhibitor cocktail). After centrifugation, the supernatant was collected in a fresh tube and the protein concentration was measured (BCA kit, Wako). Equal amounts of protein from each sample were mixed with loading buffer (50 mM Tris-HCl pH 7.5, 5% 2-mercaptoethanol, 2% sodium dodecyl sulfate (SDS), 5% sucrose, and 0.005% bromophenol blue) and then denatured at 100 °C. The protein extracts were subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) (10% acryl amide gel) in electrophoresis running buffer and electrophoresed for 1 h at room temperature at 0.2 mA, and then transferred onto a membrane (Millipore, Darmstadt, Germany) for 1 h at 100 V. The membrane was blocked with 5% skim milk powder in TBS-T for 1 h. After washing by TBS-T, the membrane was incubated with primary antibodies (MBP, BioLegend, 1:1000; GAPDH, MBL, Nagoya, JAPAN) overnight. After extensive rinsing, the membrane was incubated with a secondary antibody (Anti-Mouse IgG HRP-Linked Fragment, Cell Signaling Technology, Danvers, MA) for 1 h at room temperature. The corresponding bands were detected using an ECL-plus Western Blotting Detection System (GE Healthcare, Little Chalfont, UK). Densitometry of western blot data was performed using image j software. To account for several isoforms of MBP, protein bands from 10 kDa to 25 kDa were used 17 .
Measurement of locomotor activity. Measurement of locomotor activity was performed as previously described 9 . Mice were placed individually in a transparent acrylic cage with a black frosted Plexiglas floor (45 × 25 × 40 cm), and locomotor activity was measured every 5 min for 60 min using digital counters with infrared sensors (Scanet MV-40; MELQEST, Toyama, Japan). Y-maze test. Measurement of spontaneous alternation behavior was performed as previously described 13 .
The percentage alternation was calculated using the following formula: (number of alternations)/(total number of arm entries-2) × 100 (%).
Three-chambered social interaction test. The social interaction test was performed using a three-chambered plastic box (60 × 40 × 22 cm, MELQEST), as described in a previous report 32 . An unfamiliar C57BL/6J male (Stranger) that had no contact with the subject mice were placed in one side of the chamber, and an object was placed on the other side. The stranger mouse and the object were enclosed in a small, round wire cage, which allowed olfactory, visual, auditory, and tactile contact, but did not allow for deep contact. The subject mouse was first placed in the middle chamber and allowed to explore the entire social test box for a 10 min session. Measurement of the interaction time was taken from the amount of time spent around the wire cage.
Elevated plus maze test. The elevated plus maze test was performed as previously described 13 . In brief, this maze is comprised of two open arms (25 × 5 × 5 cm), two closed arms (25 × 5 × 27 cm), and a home platform (5 × 5 cm). It was elevated to a height of 55 cm above the floor. The time spent in open arm was measured.
Electron microscopy. Electron microscopy was performed as previously described 11 . In brief, Shati +/+ and Shati −/− mice at postnatal day 21 were anesthetized and perfused intracardially with 2.0% glutaraldehyde in 0.1 M cacodylate buffer pH 7.4 for 15 min. The brains were removed and dissected 1-2 mm thick section which included the prefrontal cortex. The sections were left in fixative overnight at 4 °C then washed in 0.1 M cacodylate buffer, dehydrated with graded ethanols and infiltrated with propylene oxide. After infiltration of propylene oxide, the section was oriented and embedded with epoxy resin. Sections (1 µm) of the specimen block were cut on Ultracut micotome (Leica), stained with 0.5% toluidine blue in 1% sodium borate in water and prefrontal cortex was then identified by light microscopy and areas were selected for thin sectioning. Thin sections (100 nm) were collected on copper grids, stained with uranyl acetate and lead citrate. The samples were viewed at electron microscope.