Disruption of Shmt1 Impairs Hippocampal Neurogenesis and Mnemonic Function in Mice

Elena V Abarinov; Anna E Beaudin; Martha S Field; Cheryll A Perry; Robert H Allen; Sally P Stabler; Patrick J Stover

doi:10.3945/jn.113.174417

. 2013 May 22;143(7):1028–1035. doi: 10.3945/jn.113.174417

Disruption of Shmt1 Impairs Hippocampal Neurogenesis and Mnemonic Function in Mice^¹^,^²^,^³

Elena V Abarinov ^7,^4,⁶, Anna E Beaudin ^7,^5–7,⁶, Martha S Field ⁷, Cheryll A Perry ⁷, Robert H Allen ⁸, Sally P Stabler ⁸, Patrick J Stover ^7,^*

PMCID: PMC3681542 PMID: 23700346

Abstract

Impaired folate-mediated one-carbon metabolism (OCM) has emerged as a risk factor for several diseases associated with age-related cognitive decline, but the underlying mechanisms remain unknown and thus hinder the identification of subpopulations most vulnerable to OCM disruption. Here we investigated the role of serine hydroxymethyltransferase 1 (SHMT1), a folate-dependent enzyme regulating de novo thymidylate biosynthesis, in influencing neuronal and cognitive function in the adult mouse. We observed Shmt1 expression in the hippocampus, including the granule cell layer of the dentate gyrus (DG), and examined hippocampal neurogenesis and hippocampal-dependent fear conditioning in mice deficient for Shmt1. We used a 3 × 3 design in which adult male Shmt1^+/+, Shmt1^+/−, and Shmt1^−/− mice were fed folic acid control (2 mg/kg), folic acid–deficient (0 mg/kg), or folic acid–supplemented (8 mg/kg) diets from weaning through the duration of the study. Proliferation within the DG was elevated by 70% in Shmt1^+/− mice, yet the number of newborn mature neurons was reduced by 98% compared with that in Shmt1^+/+ mice. Concomitant with these alterations, Shmt1^+/− mice showed a 45% reduction in mnemonic recall during trace fear conditioning. Dietary folate manipulations alone did not influence neural outcomes. Together, these data identify SHMT1 as one of the first enzymes within the OCM pathway to regulate neuronal and cognitive profiles and implicate impaired thymidylate biosynthesis in the etiology of folate-related neuropathogenesis.

Introduction

Low folate status is associated with several adverse neurocognitive outcomes, but the underlying mechanisms remain poorly defined (1). In particular, memory function has emerged as acutely sensitive to perturbations in folate homeostasis (2, 3). Mnemonic performance relies on the integrity of the hippocampus, and it has been suggested that decreased hippocampal neurogenesis as a result of aging may contribute to hippocampal dysfunction and memory impairment (4, 5). Impaired hippocampal proliferation affects performance on certain memory tasks in rodents (6, 7), and altered hippocampal proliferation has also been observed in the context of several neurological disorders (8–10). Interestingly, adult mice fed an FD diet exhibit impaired hippocampal proliferation (11, 12). The effect of folate deficiency on hippocampal proliferation may, therefore, underlie the association between disrupted one-carbon metabolism (OCM)⁹ and memory impairment; however, studies clearly identifying the responsible pathways and mechanism(s) are lacking.

Folates function as cofactors for the remethylation of homocysteine to methionine and for the de novo synthesis of purines and thymidylate (dTMP) (13) (Supplemental Fig. 1). Methionine can be adenosylated to form S-adenosylmethionine (AdoMet), the primary methyl donor for methylation of DNA and RNA and for neurotransmitter synthesis. Impaired dTMP biosynthesis causes uracil misincorporation into the genome, which contributes to hippocampal neurodegeneration (12). Elevated homocysteine induces oxidative stress and neurodegeneration (14) and impairs hippocampal proliferation (15). Thus, impaired folate metabolism may promote neuropathogenesis by affecting homocysteine levels, cellular methylation, and/or genomic stability. Unfortunately, dietary manipulation of folate status affects the entire OCM network and is thus insufficient to discriminate among these potential causal pathways.

Serine hydroxymethyltransferase 1 (SHMT1) is a folate-dependent enzyme that regulates homocysteine remethylation and de novo dTMP biosynthesis (16, 17) (Supplemental Fig. 1). In the cytoplasm, SHMT1 sequesters 5-methyltetrahydrofolate (5-methylTHF), a cofactor necessary for homocysteine remethylation and subsequent AdoMet production (Supplemental Fig. 1). During the S phase, SHMT1 translocates into the nucleus to provide thymidine triphosphate for DNA replication (18–20). Interestingly, Shmt1 heterozygosity, but not homozygosity, elevates genomic uracil misincorporation and cellular methylation potential (16); however, only impaired dTMP biosynthesis mediates colon cancer risk and neural tube closure defects in Shmt1^+/− mice (21, 22). Studying the consequences of Shmt1 disruption may, therefore, clarify the underlying mechanisms connecting impaired folate metabolism and neurocognitive disorders.

In this study, we used Shmt1-deficient mice to determine the mechanism(s) by which impaired folate metabolism influences hippocampal function at both the cellular and behavioral levels. We hypothesized that disruption of dTMP biosynthesis in Shmt1^+/− mice would affect hippocampal neurogenesis and related behavioral outcomes, but we examined both Shmt1^+/− and Shmt1⁻^/− mice as each has been associated with a distinct metabolic and disease phenotype. Together, these experiments aimed to identify a causal relationship between disruption of specific pathways in folate-mediated OCM and folate-associated neuropathogenesis.

Materials and Methods

Experimental mice and diets.

The generation and characterization of Shmt1^−/− mice have been previously described (16). Shmt1^+/− mice were maintained as a congenic heterozygous breeding colony on a C57BL/6 background (Jackson Laboratories). Mice received ad libitum access to food and water and were housed under a 12-h light/dark cycle under specific pathogen–free conditions. All experimental procedures were approved by the Cornell Institutional Animal Care and Use Committee according to the guidelines of the Animal Welfare Act and all applicable state and federal laws. At weaning, male Shmt1^+/+, Shmt1^+/−, and Shmt1^−/− mice were randomly assigned to an AIN93G folate-control (FC; 2 mg/kg folic acid; Dyets), folate-deficient (FD; 0 mg/kg folic acid), or folate-supplemented (FS; 8 mg/kg folic acid) diet. Assignment to each genotype and diet condition was balanced across litters. All mice remained on their respective diet until they were killed. Mice were housed up to 4 to a cage and grouped according to diet with the exception of mice used for behavioral analysis, which were housed individually at least 1 wk prior to training. Unless otherwise noted, experiments were conducted on male mice between 14–16 mo of age. All experiments were conducted by an investigator blinded to genotype and diet group.

Experimental design.

Mice were assigned to 1 of 3 experimental groups, and group assignment was balanced across litters. Experimental group A underwent behavioral testing as described below. Three weeks after behavioral analysis, these same mice were killed by cervical dislocation, and their brains, livers, plasma, and RBCs were flash-frozen for biochemical analysis, including measurement of metabolite concentrations and genomic uracil content. Hippocampal proliferation was assessed in mice assigned to group B 24 h after pulse treatment with the thymidine analog 5′-bromodeoxyuridine (BrdU; Sigma). A 50-mg/kg dose of BrdU was injected intraperitoneally once daily for 4 consecutive days. Survival and neuronal differentiation of proliferating cells were assessed in mice assigned to group C 3 wk after this same BrdU pulse.

In situ hybridization.

Localization of Shmt1 mRNA by in situ hybridization was performed by using digoxigenin-labeled RNA probes as previously described (23). Probes were synthesized to the 5′ end of Shmt1 with the following primers: 1) forward, 5′-TGGCGGGACCGAGTTCCATC-3′ and 2) reverse, 5′-CCAGGTTCCGGGAGTAGCAG-3′. Expression was analyzed in adult mice that were 6–8 wk of age.

BrdU immunofluorescent staining.

Twenty-four hours (group B) or 3 wk (group C) after the last BrdU injection, mice were transcardially perfused with PBS followed by 4% paraformaldehyde. Thirty-micrometer sections were treated for 20 min at 95°C in 1× Antigen Retrieval Solution (Dako), and DNA was denatured with 2 N HCl for 1 h at 37°C. Sections were incubated with rat anti-BrdU (1:50; Abcam) overnight at 4°C and with Alexa Fluor 488 conjugated goat anti-rat secondary antibody (1:400; Molecular Probes) for 1 h at room temperature (RT).

BrdU/NeuN immunofluorescence.

Free-floating sections were processed by using the same procedure as for BrdU-alone immunostaining but without antigen retrieval. Sections were incubated with rat anti-BrdU (1:600; Abcam) and mouse anti-NeuN (1:1000; Chemicon) for 1 h at RT, followed by an incubation of Alexa Fluor 488 conjugated goat anti-rat (1:400; Molecular Probes) and Cy5 conjugated goat anti-mouse (1:300; Chemicon) for 1 h at RT.

Imaging and cell counts.

Every 10th (BrdU alone, group B) or 20th (BrdU/NeuN, group C) section throughout the rostral-caudal extent of the hippocampus was visualized on a confocal microscope (Leica TCS SP2) at a magnification of 20× . A z-series was taken throughout the depth of each section and reconstructed in Volocity 4.4 (PerkinElmer). BrdU-labeled cells were counted in the subgranular zone (SGZ) of the DG, defined as a 2-cell width spanning the intersection between the granule cell layer and the hilus. BrdU/NeuN-labeled cells were counted within the entirety of the granule cell layer. Cell counting of BrdU+ cells was performed automatically in Volocity 4.4. Colabeling of BrdU+ cells with NeuN was determined manually.

Volume measurements.

The volume of the DG was measured in every section adjacent to those used in immunofluorescent staining. The area of the DG was measured in Nissl-stained sections by manual tracing using Metamorph Image Analysis (Molecular Devices). Volume was calculated by multiplying the area of each section with its depth (30 μm) and summing these values across the rostral-caudal extent of the hippocampus.

Auditory trace fear conditioning.

Auditory trace fear conditioning was used to assess associative memory deficits related to hippocampal proliferation (6). Whereas trace fear conditioning measures subtle defects associated with altered hippocampal proliferation, contextual conditioning serves as a measure of general hippocampal integrity (24, 25). The protocol closely followed Smith et al. (26) and is summarized in Figure 3. Briefly, mice were first placed individually into the training chamber (context A, vanilla scent; Coulbourn) and acclimated for 720 s. Percentage freezing, defined as complete immobility for at least 1 s other than that required for respiration, was recorded by Actimetrics FreezeFrame software (Coulbourn) throughout the entirety of all procedures. Twenty-four hours after acclimation, mice were returned to context A and presented with six 100-s trials consisting of a 20-s sound (80 db, 2 kHz) and a 2-s shock (0.7 mAmp), separated by an interval of 15 s (the “trace”) between the offset of the tone and the onset of the shock. The first tone was presented after a 3-min acclimation period, and subsequent trials were separated by a 63-s intertrial interval. The following day, mice were returned to context A for 300 s to assess contextual conditioning. Freezing behavior in response to the context was measured across the entirety of the 300-s session. Approximately 2 to 4 h later, mice were placed into a novel context with a different set of visual and olfactory cues (context B, bedding scent; Coulbourn) and presented with 4 trials that were identical to those from training, except that the foot-shock was not delivered. Baseline freezing was measured during the first 30 s after placement into context B, prior to the presentation of the first tone. Trace conditioning was measured as freezing during the 35-s interval including the presentation of the tone and the trace period. All testing occurred during the light phase of the light/dark cycle.

Determination of metabolites.

Blood was collected by cardiac puncture into heparin-coated tubes (BD Vacutainer) and centrifuged at 2000 × g for 5 min to isolate plasma. Tissues were immediately flash-frozen in liquid nitrogen and stored at −80°C. Plasma metabolite concentrations were determined by stable isotope dilution capillary GS-MS as described elsewhere (27, 28). Folate concentrations in RBCs and cortex samples were measured by a Lactobacillus casei assay as previously described (29).

Statistical analysis.

Analyses of cell counts, genomic uracil content, and metabolite measurements were conducted in JMP, version 8.02 (SAS Institute). Cell counts and metabolite measurements were analyzed by 2-way ANOVA with Tukey’s post hoc analysis. Shmt1 genotype and diet were included as independent variables. BrdU cell counts and serine concentrations were log-transformed in order to normalize the data. For BrdU/NeuN quantification, 3 mice (2 Shmt1^+/− mice and 1 Shmt1^−/− mouse) were excluded from the analysis due to the complete absence of BrdU+ cells. Analyses of behavior data were performed in SAS (version 9.0) by using a mixed-model procedure (PROC Mixed) followed by post hoc analysis with Student’s t test and Bonferroni’s correction. Values in the text indicate means ± SEMs. Differences were considered significant at P ≤ 0.05.

Results

Shmt1 is specifically expressed in regions of constitutive neurogenesis.

Examination of Shmt1 expression revealed restricted expression within the adult mouse brain. In situ hybridization localized Shmt1 mRNA to the hippocampus, including the DG and cornu ammonis regions (Fig. 1D–F), as well as the subventricular zone, rostral migratory stream, and the cerebellum (data not shown). Thus, Shmt1 mRNA is expressed in regions of cellular proliferation, as well as in regions implicated in memory function.

*Shmt1* is expressed in the CA regions and dentate gyrus of the hippocampus. (A–C) Schematics showing coronal sections at 3 rostro-caudal levels corresponding to −1.60, −2.10, and −5.70 mm relative to Bregma. (D–F) Photomicrographs indicating in situ hybridization signal with antisense *Shmt1* probes in coronal brain sections. (G–I) Corresponding photomicrographs depicting in situ hybridization signal with a sense (control) probe. CA, cornu ammonis; cc, corpus callosum; CP, caudate putamen; DG, dentate gyrus; dhc, dorsal hippocampal commisure; ec, entorhinal cortex; fi, fimbria; int, internal transverse artery; *Shmt1*, serine hydroxymethyltransferase 1; sg, stratum granulosum; sp, stratum pyramidale; SUB, subiculum; VL, lateral ventricle.

Shmt1 disruption enhances proliferation in the SGZ of the DG.

Analysis of hippocampal cell proliferation, as determined by the total number of BrdU+ cells in the SGZ of the DG, revealed that Shmt1^+/− mice exhibited 70% more BrdU+ cells compared with the Shmt1^+/+ and Shmt1^−/− mice (P = 0.012; Fig. 2A, B). Two-way ANOVA also revealed a significant gene × diet interaction (P = 0.0011); however, post hoc analysis revealed that none of the groups were significantly different from Shmt1^+/+ mice fed an FC diet (P > 0.10), and the limited sample size of a few groups obstructed clear interpretation of the interaction effect (see Materials and Methods). Dietary folate intake alone did not independently affect hippocampal proliferation (P > 0.10). Together, these data indicate that partial loss of Shmt1 expression results in greater proliferation in the SGZ independent of dietary folic acid intake.

*Shmt1* heterozygosity, independent of diet, results in greater hippocampal cellular proliferation but reduced numbers of mature neurons in the DG. (A) Representative images of BrdU+ cells (green) in the DG of *Shmt1^+/+* and *Shmt1^+/−* mice 24 h after BrdU pulse. (B) Quantification of BrdU-labeled cells from panel A. Values are means ± SEMs. Groups designated with different letters are significantly different from one another: *Shmt1^+/−* mice are significantly different than *Shmt1^+/+* and *Shmt1^−/−* mice (P = 0.012, n = 9–10 per genotype). (C) A representative image of a BrdU/NeuN-labeled cell in the DG 3 wk after BrdU pulse. (D) Quantification of BrdU+ cells and BrdU/NeuN-labeled cells 3 wk after BrdU pulse. Values are means ± SEMs. No significant differences in the number of BrdU+ cells were observed among groups (P > 0.1, n = 9–12 per genotype). Analysis of BrdU/NeuN-labeled cells, however, showed a trend toward a significant effect of genotype, and post hoc analysis revealed that *Shmt1^+/−* mice were significantly different from *Shmt1^+/+* mice (*P = 0.05, n = 8–10 per genotype). BrdU, 5-bromodeoxyuridine; DG, dentate gyrus; NeuN, Neuronal Nuclei; *Shmt1*, serine hydroxymethyltransferase 1.

Shmt1 disruption decreases the number of newly born, mature neurons in the DG.

To determine whether the increased proliferation in the DG of Shmt1^+/− mice resulted in greater net neurogenesis, the survival and differentiation of BrdU-incorporating cells was analyzed 3 wk after a BrdU pulse. In this paradigm, BrdU+ cell counts reflected both initial proliferation and subsequent survival of newly generated cells. Further phenotypic analysis of BrdU-labeled cells was performed by costaining with the mature neuronal marker NeuN. Analysis throughout the granule cell layer of the DG revealed that the total number of BrdU-labeled cells remaining 3 wk after BrdU pulse did not differ across Shmt1 genotypes or diet conditions (P > 0.10; Fig. 2D). However, analysis of the number of BrdU/NeuN-labeled cells showed a trend toward a significant effect of Shmt1 genotype (P = 0.05; Fig. 2D). Post hoc analysis revealed that Shmt1^+/− mice exhibited significantly fewer BrdU/NeuN-labeled cells compared with Shmt1^+/+ mice (Fig. 2D). Diet and gene × diet interactions did not significantly affect the total number of BrdU/NeuN-labeled cells. These results indicate that despite the high levels of proliferation initially observed in the Shmt1^+/− mice, the majority of these newborn cells did not persist after their generation. Indeed, of the newborn cells that did remain in the DG of Shmt1^+/− mice, significantly fewer mature neurons were observed than in wild-type mice. Despite this reduction in newborn neurons, no differences in total volume of the DG were observed among groups (P > 0.10; data not shown).

Shmt1 disruption impairs trace fear conditioning.

We next investigated whether altered proliferation and neurogenesis observed in Shmt1^+/− mice correlated with behavioral outcomes in a fear-conditioning paradigm (Fig. 3A). Two-way ANOVA revealed that Shmt1^+/− mice froze significantly less compared with Shmt1^+/+ mice during presentation of the tone alone when freezing was averaged across all trials (freezing was analyzed during the tone and trace; P = 0.02; Fig. 3B). No significant effect of diet or gene × diet interaction was observed (P > 0.10). Acquisition freezing, baseline freezing, or freezing behavior during the context test did not differ among groups (P > 0.10; Fig. 3B and data not shown). These results demonstrate that Shmt1^+/− mice exhibit a specific deficit in hippocampal function consistent with the alterations observed in hippocampal neurogenesis.

Mice heterozygous for *Shmt1* exhibit deficits in auditory trace fear conditioning independent of diet. (A) Schematic representation of the behavior protocol. (B) *Shmt1^+/−* mice showed a significant reduction in percentage freezing during the trace test compared with *Shmt1^+/+* mice (*P = 0.020, n = 7–10 per condition). No differences in freezing were observed at baseline (measured during the first 30 s in context A on training day 2, prior to presentation of the first tone) or across all 5 min of testing during the context test (P > 0.10). Values are means ± SEMs. Shmt1, serine hydroxymethyltransferase 1.

Shmt1 disruption and folate intake influence folate status in brain and RBC tissue.

We examined the effect of Shmt1 deficiency and dietary folate manipulation on blood and brain folate concentrations. Dietary folic acid deficiency significantly reduced RBC folate concentrations by 40% (P < 0.0001; Table 1), but dietary folate supplementation did not affect RBC folate concentrations. In the brain, 2-way ANOVA revealed a significant gene × diet interaction on folate concentratoins (P = 0.037; Table 1). Student’s t test post hoc analysis indicated significantly reduced brain folate concentrations in both Shmt1^+/+ and Shmt1^+/− mice in response to the FD diet, but the FS diet did not significantly elevate folate concentrations relative to the FC diet (Table 1). In contrast, brain folate concentratons in Shmt1^−/− mice did not differ between the FC and FD diets (P > 0.10; Table 1); only the FS diet significantly elevated folate concentrations above those observed with the FD diet in these mice (Table 1). Overall, these results suggest that the FD diet reduced brain folate concentrations, whereas the FS diet did not significantly elevate brain folate concentrations. These results are consistent with both a significant main effect of diet and the absence of an effect of the FS diet on blood folate concentrations across all 3 genotypes. Additionally, Shmt1^−/− mice were differentially sensitive to dietary manipulation of folate concentrations.

TABLE 1.

Total RBC and brain folate concentrations in Shmt1^+/+, ^+/−, and ^−/− mice fed FD, FC, or FS diets¹

	RBCs					Brain
	Diet			P value		Diet			P value
Genotype	FD	FC	FS	Diet × genotype effect	Diet effect	FD	FC	FS	Diet × genotype effect	Diet effect
	fmol/μg protein					fmol/μg protein
Shmt1^+/+	11.4 ± 5.0	36.9 ± 5.0	53.4 ± 7.0	NS	N/A	1.56 ± 0.2^b	3.05 ± 0.2^a	2.73 ± 0.2^a	0.037	N/A
Shmt1^+/−	11.0 ± 5.0	33.4 ± 5.0	42.5 ± 5.0	NS	N/A	1.00 ± 0.2^b	2.66 ± 0.2^a	3.46 ± 0.2^a	0.037	N/A
Shmt1^−/−	20.0 ± 5.0	42.9 ± 5.0	41.4 ± 5.0	NS	N/A	1.21 ± 0.3^b	2.16 ± 0.2^b	3.21 ± 0.2^a	0.037	N/A
All	14.1 ± 3.2^b	37.7 ± 3.2^a	41.8 ± 3.4^a	N/A	<0.0001	1.26 ± 0.2^c	2.62 ± 0.1^b	3.14 ± 0.1^a	N/A	<0.0001

Open in a new tab

Values are means ± SEMs; n = 4 per group. P ≤ 0.05 was considered significant; NS = P > 0.10. Means in a row with superscripts without a common letter differ, P ≤ 0.05. No significant genotype effects were detected. FC, folate control; FD, folate deficient; FS, folate supplemented; N/A, not applicable; Shmt1, serine hydroxymethyltransferase 1.

Shmt1 and dietary folate deficiency influence metabolic markers of homocysteine remethylation.

Both dietary folate status and Shmt1 disruption significantly influenced several biomarkers of homocysteine metabolism. The FD diet resulted in a doubling of homocysteine concentrations compared with the FC and FS diets (P < 0.0001; Table 2), whereas homocysteine concentrations were significantly reduced by 25% in Shmt1^−/− mice compared with the Shmt1^+/+ mice (P = 0.032; Table 2). ANOVA also revealed significant effects of both diet (P = 0.005; Table 2) and Shmt1 genotype (P = 0.034; Table 2) on plasma concentrations of cystathionine, a product of homocysteine degradation via the folate-independent transsulfuration pathway, catalyzed by the enzyme cystathionine β-synthase. Cystathionine concentrations were ∼20% higher in mice fed the FD diet compared with those fed the FC and FS diets (Table 2), consistent with the observed increase in homocysteine concentrations associated with the FD diet. In contrast, cystathionine concentrations were reduced by 15% in Shmt1^+/− mice, suggesting that Shmt1^+/− mice may metabolize less homocysteine through the transsulfuration pathway (Table 2). Serine concentrations were also significantly affected by diet and Shmt1 genotype. Post hoc analysis of a significant gene × diet interaction (P = 0.034; Table 2) revealed that serine concentrations were significantly elevated in Shmt1^+/− males fed the FS diet compared with those fed the FC and FD diets (Table 2).

TABLE 2.

Metabolic profile of plasma isolated from Shmt1^+/+, Shmt1^+/−, and Shmt1^−/− mice fed FD, FC, or FS diets¹

	FD				FC				FS				P value
Metabolite	Shmt1^+/+	Shmt1^+/−	Shmt1^−/−	All	Shmt1^+/+	Shmt1^+/−	Shmt1^−/−	All	Shmt1^+/+	Shmt1^+/−	Shmt1^−/−	All	Genotype	Diet	Diet × Genotype
Homocysteine, μmol/L	8.96 ± 0.72^a	7.70 ± 0.72^a,b	6.13 ± 0.72^b	7.59 ± 0.41^c	3.77 ± 0.72^a	4.41 ± 0.63^a,b	3.47 ± 0.72^b	3.88 ± 0.40^d	4.12 ± 0.72^a	3.61 ± 0.72^a,b	2.88 ± 0.72^b	3.54 ± 0.41^d	0.0322^a,b	<0.0001^c,d	NS
Cystathionine, nmol/L	1040 ± 77^a	1030 ± 77^a,b	1120 ± 73^b	1060 ± 44^c	976 ± 77^b	665 ± 67^a,b	871 ± 77^b	837 ± 43^d	936 ± 77	782 ± 77^a,b	831 ± 73^b	850 ± 44^d	0.0337^a,b	0.0005^c,d	NS
Cysteine, μmol/L	208 ± 8	209 ± 8	210 ± 7	209 ± 4	211 ± 8	216 ± 7	194 ± 8	207 ± 4	199 ± 8	209 ± 8	203 ± 7	204 ± 4	NS	NS	NS
α-Aminobutyric acid, μmol/L	13.0 ± 1.6^a	13.1 ± 1.6^a	13.6 ± 0.9^a	13.3 ± 1.5	14.5 ± 1.6^a	14.0 ± 1.4^a	17.2 ± 1.6^a	15.2 ± 0.9	10.7 ± 1.6^a	11.2 ± 1.6^a	16.3 ± 1.5^a	12.7 ± 0.9	NS	0.0292	NS
Methionine, μmol/L	44.7 ± 10.9	38.0 ± 10.9	44.0 ± 10.4	43.7 ± 6.2	66.0 ± 10.9	43.7 ± 9.5	43.1 ± 10.9	59.3 ± 6.0	42.7 ± 10.9	44.7 ± 10.9	49.7 ± 10.4	49.0 ± 6.2	NS	NS	NS
Glycine, μmol/L	201 ± 16	183 ± 16	213 ± 16	203 ± 9	257 ± 16	210 ± 14	201 ± 16	231 ± 9	204 ± 16	222 ± 16	209 ± 16	231 ± 9	NS	NS	NS
Serine, μmol/L	149 ± 19^a,b	130 ± 19^b	154 ± 18^a,b	144 ± 11	196 ± 19^a	129 ± 17^b	136 ± 19^a,b	151 ± 11^a,b	137 ± 19^a,b	145 ± 19^a,b	152 ± 18	144 ± 11	N/A	N/A	0.0340
Dimethylglycine, μmol/L	6.40 ± 0.49	5.20 ± 0.49	5.21 ± 0.49	5.60 ± 0.28	6.04 ± 0.49	6.29 ± 0.43	5.42 ± 0.49	5.92 ± 0.28	5.10 ± 0.49	5.34 ± 0.49	5.14 ± 0.49	5.18 ± 0.31	NS	NS	NS
Methylglycine, μmol/L	1.0 ± 0.10^a	0.85 ± 0.10^a	1.0 ± 0.10^a	1.0 ± 0.06	1.1 ± 0.10^a	0.89 ± 0.10^a	0.78 ± 0.10^a	0.93 ± 0.06	0.86 ± 0.10^a	0.77 ± 0.10^a	0.85 ± 0.10^a	0.83 ± 0.06	NS	0.0640	NS

Open in a new tab

Values are means ± SEMs; n = 10–14 per group. P ≤ 0.05 was considered significant; NS = P > 0.10. Means in a row with superscripts without a common letter differ, P ≤ 0.05. All mice within a genotype are marked to indicate genotype effects irrespective of diet designation. FC, folate control; FD, folate deficient; FS, folate supplemented; N/A, not applicable; Shmt1, serine hydroxymethyltransferase 1.

Discussion

Epidemiological studies have identified impaired folate metabolism resulting from either dietary deficiency and/or genetic variation as a possible risk factor for age-related cognitive decline and neurodegenerative diseases [for review see (1)]. However, study of the mechanisms by which impaired folate metabolism affects neurocognitive function has been obstructed by the lack of available genetic mouse models that target individual folate-dependent pathways without causing severe disruption to the broader OCM network. In this study, we report that deficiency of the folate-metabolizing enzyme SHMT1, which regulates folate-dependent de novo dTMP biosynthesis, affects hippocampal function at both the cellular and behavioral levels in adult mice. Shmt1 showed enriched expression within the hippocampus (Fig. 1), and adult Shmt1^+/− mice exhibited elevated hippocampal cellular proliferation (Fig. 2A, B). However, the majority of proliferating cells did not persist after their generation. Although we did not explore the alternative fates of these newborn cells, we demonstrated that of the BrdU+ cells remaining after 3 wk, significantly fewer expressed the mature neuronal marker NeuN (Fig. 3C, D). Concomitant with these alterations in neurogenesis, Shmt1^+/− mice performed significantly worse on a trace fear-conditioning task, a test of associative memory that has been demonstrated to reflect changes in hippocampal proliferation (6, 7). We have previously reported that Shmt1^+/− mice are impaired in their ability to synthesize dTMP de novo and, therefore, accumulate uracil within nuclear DNA (16). Together, these data implicate SHMT1 in regulating hippocampal function and suggest that impaired dTMP biosynthesis may contribute to neurocognitive dysfunction.

In the present study, we observed neural and cognitive deficits in mice that were heterozygous for Shmt1 disruption, but mice that were homozygous null for Shmt1 were not affected. Our laboratory has recently shown that Shmt1 heterozygosity correlates with both elevated genomic uracil misincorporation and disease susceptibility (21, 22). Shmt1^+/− mice maintained on an APC^min background were susceptible to colon cancer that was associated with greater nuclear uracil accumulation (21). Importantly, neither elevated genomic uracil misincorporation nor cancer susceptibility was observed in Shmt1 null mice. This absence of disease vulnerability in Shmt1^−/− mice was associated with an upregulation of dTMP synthase levels, as well as enzymes within the nucleotide salvage pathway, which ameliorated the disruption of dTMP biosynthesis caused by the complete lack of SHMT1 protein (21). Shmt1^+/− mice, on the other hand, did not exhibit these compensatory mechanisms. The presence of both neurobehavioral and cellular phenotypes associated with Shmt1 heterozygosity in the present study therefore suggests that impaired dTMP biosynthesis in newborn hippocampal neurons underlies the neurocognitive pathology observed in response to Shmt1 deficiency. Although we did not detect differences in the genomic uracil content of whole hippocampus preparations from Shmt1-deficient mice (data not shown), this result is likely due to the limited population of proliferating Shmt1-expressing cells within the hippocampus. Nevertheless, newly generated neurons have been shown to be more sensitive to telomere-associated DNA damage than either neuronal progenitors or mature neurons (30). Furthermore, uracil misincorporation in uracil-DNA-glycosylase–deficient mice impairs hippocampal neurogenesis and also diminishes survival of hippocampal neurons in vivo (12, 14). Together, these results suggest that disrupted folate status may affect hippocampal function by altering dTMP biosynthesis and/or genomic uracil levels in proliferating cells within the DG.

In addition to regulating dTMP biosynthesis, SHMT1 also influences cellular methylation potential by binding 5-methylTHF in the cytoplasm and preventing its entrance into the homocysteine remethylation cycle (16, 17) (Supplemental Fig. 1). Although hippocampal methylation potential was not directly measured in this study, Shmt1^+/− mice exhibited lower concentrations of cystathionine, a product of homocysteine catabolism through the folate-independent transsulfuration pathway, suggesting greater availability of folate-derived methyl groups for homocysteine remethylation and subsequent AdoMet production. AdoMet-dependent methylation is required for methylation of DNA, RNA, proteins, lipids, and neurotransmitters (31), and the effects of chromatin hypomethylation on hippocampal proliferation and neurocognitive function have been well documented (32). However, it is unclear whether a higher cellular methylation potential affects methyltransferase activity or whether chromatin hypermethylation affects neurocognitive function. Furthermore, no differences in methylation potential or gene expression have been observed in the colons of cancer-susceptible Shmt1-deficient mice (21). Together, these data suggest that the changes in methylation potential resulting from Shmt1 disruption contribute minimally toward neurocognitive disease.

In contrast to previous reports (11, 12, 14, 33), we did not observe an effect of dietary folic acid deficiency on hippocampal neurogenesis or on hippocampal-dependent memory function. The deficient diets used in our study differed from those in previous investigations in that they did not include the addition of the antibiotic succinyl sulfathiazole in the diet to eliminate folate-producing bacteria found in the murine digestive system. Although mice fed the FD diet exhibited significant reductions in RBCs and brain total folate concentrations, as well as elevated plasma homocysteine concentrations (∼1.5-fold), reports of mice fed deficient diets with the addition of succinyl sulfathiazole indicate more extreme deficiency (e.g., ∼9-fold elevated homocysteine concentrations) (11, 33). Dietary folate supplementation also showed little effect on neurocognitive outcomes in our study, but analysis of brain folate concentrations revealed that the FS diet did not significantly raise brain folate concentrations above those observed in mice fed the FC diet within each genotype group (Table 1).

The absence of an effect of dietary folate deficiency on neurobehavioral outcomes in our study also is in contrast to aforementioned reports of folate deficiency affecting cognitive function in humans. One possible explanation for this discrepancy is that human dietary folate deficiency, particularly in aging humans, is often comorbid with deficiency of other micronutrients, including vitamin B-12, which influences folate metabolism. Additionally, genetic diversity in the human population, including common polymorphisms in folate-related genes, leaves certain subpopulations more vulnerable to the effect of folate deficiency (34). Our study with mice does not address such genetic variation. Nevertheless, our dietary manipulations, which resulted in moderate folate deficiency, may better model human dietary folate fluctuations and highlights the necessity of studying the causal mechanisms underlying OCM disruption as folate supplementation alone may not ameliorate every condition.

Homocysteine is a biomarker of impaired folate metabolism, and increased concentrations of homocysteine have also been identified as an independent risk factor for age-related cognitive decline and neurodegenerative disease [for review see (35)]. Whether homocysteine is a direct causal factor in age-related neurological disorders remains to be determined. Both in vitro and in vivo studies have indicated that highly elevated homocysteine concentratons, caused either by dietary manipulation or direct treatment with homocysteine, can lead to impaired hippocampal proliferation, impaired cognitive function, and neurodegeneration (11, 12, 14, 33). In our study, mice fed an FD diet exhibited elevated plasma homocysteine, but elevated homocysteine was not associated with any of our behavioral or molecular outcomes. Furthermore, homocysteine was not significantly elevated in affected Shmt1^+/− mice (Table 2). Another recent study that used equivalent dietary manipulations reported that folate deficiency did not affect hippocampal morphology or cell density, despite elevated homocysteine concentratons (36). Together, these results indicate that, in our model, homocysteine does not play a causal role in the neural and cognitive alterations associated with Shmt1 heterozygosity and suggest that homocysteine may be a biomarker of impaired folate status as opposed to a causal factor in neurocognitive pathology.

Human epidemiological studies have identified memory function as particularly sensitive to folate status. Here, by using a novel mouse model to understand the role of folate metabolism in normal brain function, we have shown that memory performance is sensitive to perturbations in folate metabolism independent of dietary folate intake. We have demonstrated that disruption of Shmt1 is associated with specific cellular changes within the hippocampus that underlie the observed cognitive impairments. Although further investigation is required to elucidate the mechanism(s) whereby Shmt1 disruption affects hippocampal function, we present the first, to our knowledge, genetic evidence implicating altered de novo thymidylate biosynthesis in the etiology of neurocognitive dysfunction. Future work will be required to assess the impact of dTMP supplementation on the prevention and treatment of neurocognitive disease.

Supplementary Material

Online Supporting Material

supp_143_7_1028__index.html^{(723B, html)}

Acknowledgments

The authors thank Francoise Vermeylen for assistance with statistical analysis and Rachel Slater, Sanchit Gupta, and Dina Diskina for technical assistance. E.V.A., A.E.B., P.J.S. designed the research; E.V.A., A.E.B., M.S.F., C.A.P., R.H.A., and S.P.S. conducted the research; E.V.A. and A.E.B. analyzed the data; E.V.A., A.E.B., and P.J.S. wrote the manuscript; and P.J.S. had primary responsibility for final content. All authors read and approved the final manuscript.

Footnotes

⁹

Abbreviations used: AdoMet, S-adenosylmethionine; BrdU, 5-bromodeoxyuridine; DG, dentate gyrus; dTMP, thymidylate; FC, folate control; FD, folate deficient; FS, folate supplemented; NeuN, neuronal nuclei; OCM, one-carbon metabolism; RT, room temperature; SGZ, subgranular zone; SHMT1, serine hydroxymethyltransferase 1; 5-methylTHF, 5-methyltetrahydrofolate.

Literature Cited

1.Kronenberg G, Colla M, Endres M. Folic acid, neurodegenerative and neuropsychiatric disease. Curr Mol Med. 2009;9:315–23. [DOI] [PubMed] [Google Scholar]
2.Ramos MI, Allen LH, Mungas DM, Jagust WJ, Haan MN, Green R, Miller JW. Low folate status is associated with impaired cognitive function and dementia in the Sacramento Area Latino Study on Aging. Am J Clin Nutr. 2005;82:1346–52. [DOI] [PubMed] [Google Scholar]
3.Durga J, Van Boxtel MP, Schouten EG, Kok FJ, Jolles J, Katan MB, Verhoef P. Effect of 3-year folic acid supplementation on cognitive function in older adults in the FACIT trial: a randomised, double blind, controlled trial. Lancet. 2007;369:208–16. [DOI] [PubMed] [Google Scholar]
4.Kuhn HG, Dickinson-Anson H, Gage FH. Neurogenesis in the dentate gyrus of the adult rat: age-related decrease of neuronal progenitor proliferation. J Neurosci. 1996;16:2027–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Bondolfi L, Ermini F, Long JM, Ingram DK, Jucker M. Impact of age and caloric restriction on neurogenesis in the dentate gyrus of C57BL/6 mice. Neurobiol Aging. 2004;25:333–40. [DOI] [PubMed] [Google Scholar]
6.Shors TJ, Townsend DA, Zhao M, Kozorovitskiy Y, Gould E. Neurogenesis may relate to some but not all types of hippocampal-dependent learning. Hippocampus. 2002;12:578–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Shors TJ, Miesegaes G, Beylin A, Zhao M, Rydel T, Gould E. Neurogenesis in the adult is involved in the formation of trace memories. Nature. 2001;410:372–6. [DOI] [PubMed] [Google Scholar]
8.Na YJ, Jin JK, Lee YJ, Choi EK, Carp RI, Kim YS. Increased neurogenesis in brains of scrapie-infected mice. Neurosci Lett. 2009;449:66–70. [DOI] [PubMed] [Google Scholar]
9.Jin K, Peel AL, Mao XO, Xie L, Cottrell BA, Henshall DC, Greenberg DA. Increased hippocampal neurogenesis in Alzheimer's disease. Proc Natl Acad Sci USA. 2004;101:343–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Liu BF, Gao EJ, Zeng XZ, Ji M, Cai Q, Lu Q, Yang H, Xu QY. Proliferation of neural precursors in the subventricular zone after chemical lesions of the nigrostriatal pathway in rat brain. Brain Res. 2006;1106:30–9. [DOI] [PubMed] [Google Scholar]
11.Kruman II, Mouton PR, Emokpae R, Jr, Cutler RG, Mattson MP. Folate deficiency inhibits proliferation of adult hippocampal progenitors. Neuroreport. 2005;16:1055–9. [DOI] [PubMed] [Google Scholar]
12.Kronenberg G, Harms C, Sobol RW, Cardozo-Pelaez F, Linhart H, Winter B, Balkaya M, Gertz K, Gay SB, Cox D, et al. Folate deficiency induces neurodegeneration and brain dysfunction in mice lacking uracil DNA glycosylase. J Neurosci. 2008;28:7219–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Fox JT, Stover PJ. Folate-mediated one-carbon metabolism. Vitam Horm. 2008;79:1–44. [DOI] [PubMed] [Google Scholar]
14.Kruman II, Kumaravel TS, Lohani A, Pedersen WA, Cutler RG, Kruman Y, Haughey N, Lee J, Evans M, Mattson MP. Folic acid deficiency and homocysteine impair DNA repair in hippocampal neurons and sensitize them to amyloid toxicity in experimental models of Alzheimer's disease. J Neurosci. 2002;22:1752–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Rabaneda LG, Carrasco M, Lopez-Toledano MA, Murillo-Carretero M, Ruiz FA, Estrada C, Castro C. Homocysteine inhibits proliferation of neuronal precursors in the mouse adult brain by impairing the basic fibroblast growth factor signaling cascade and reducing extracellular regulated kinase 1/2-dependent cyclin E expression. FASEB J. 2008;22:3823–35. [DOI] [PubMed] [Google Scholar]
16.MacFarlane AJ, Liu X, Perry CA, Flodby P, Allen RH, Stabler SP, Stover PJ. Cytoplasmic serine hydroxymethyltransferase regulates the metabolic partitioning of methylenetetrahydrofolate but is not essential in mice. J Biol Chem. 2008;283:25846–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Herbig K, Chiang EP, Lee LR, Hills J, Shane B, Stover PJ. Cytoplasmic serine hydroxymethyltransferase mediates competition between folate-dependent deoxyribonucleotide and S-adenosylmethionine biosyntheses. J Biol Chem. 2002;277:38381–9. [DOI] [PubMed] [Google Scholar]
18.Anderson DD, Stover PJ. SHMT1 and SHMT2 are functionally redundant in nuclear de novo thymidylate biosynthesis. PLoS ONE. 2009;4:e5839. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Anderson DD, Eom JY, Stover PJ. Competition between sumoylation and ubiquitination of serine hydroxymethyltransferase 1 determines its nuclear localization and its accumulation in the nucleus. J Biol Chem. 2012;287:4790–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Woeller CF, Anderson DD, Szebenyi DM, Stover PJ. Evidence for small ubiquitin-like modifier-dependent nuclear import of the thymidylate biosynthesis pathway. J Biol Chem. 2007;282:17623–31. [DOI] [PubMed] [Google Scholar]
21.Macfarlane AJ, Perry CA, Mcentee MF, Lin DM, Stover PJ. Shmt1 heterozygosity impairs folate-dependent thymidylate synthesis capacity and modifies risk of Apc(min)-mediated intestinal cancer risk. Cancer Res. 2011;71:2098–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Beaudin AE, Abarinov EV, Noden DM, Perry CA, Chu S, Stabler SP, Allen RH, Stover PJ. Shmt1 and de novo thymidylate biosynthesis underlie folate-responsive neural tube defects in mice. Am J Clin Nutr. 2011;93:789–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Williams EO, Sickles HM, Dooley AL, Palumbos S, Bisogni AJ, Lin DM. Delta protocadherin 10 is regulated by activity in the mouse main olfactory system. Front Neural Circuits. 2011;5:9. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Kim JJ, Fanselow MS. Modality-specific retrograde amnesia of fear. Science. 1992;256:675–7. [DOI] [PubMed] [Google Scholar]
25.Phillips RG, Ledoux JE. Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behav Neurosci. 1992;106:274–85. [DOI] [PubMed] [Google Scholar]
26.Smith DR, Gallagher M, Stanton ME. Genetic background differences and nonassociative effects in mouse trace fear conditioning. Learn Mem. 2007;14:597–605. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Stabler SP, Lindenbaum J, Savage DG, Allen RH. Elevation of serum cystathionine levels in patients with cobalamin and folate deficiency. Blood. 1993;81:3404–13. [PubMed] [Google Scholar]
28.Allen RH, Stabler SP, Savage DG, Lindenbaum J. Elevation of 2-methylcitric acid I and II levels in serum, urine, and cerebrospinal fluid of patients with cobalamin deficiency. Metabolism. 1993;42:978–88. [DOI] [PubMed] [Google Scholar]
29.Suh JR, Oppenheim EW, Girgis S, Stover PJ. Purification and properties of a folate-catabolizing enzyme. J Biol Chem. 2000;275:35646–55. [DOI] [PubMed] [Google Scholar]
30.Cheng A, Shin-Ya K, Wan R, Tang SC, Miura T, Tang H, Khatri R, Gleichman M, Ouyang X, Liu D, et al. Telomere protection mechanisms change during neurogenesis and neuronal maturation: newly generated neurons are hypersensitive to telomere and DNA damage. J Neurosci. 2007;27:3722–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Axelrod J. Methylation reactions in the formation and metabolism of catecholamines and other biogenic amines. Pharmacol Rev. 1966;18:95–113. [PubMed] [Google Scholar]
32.Day JJ, Sweatt JD. Epigenetic mechanisms in cognition. Neuron. 2011;70:813–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Troen AM, Chao WH, Crivello NA, D'anci KE, Shukitt-Hale B, Smith DE, Selhub J, Rosenberg IH. Cognitive impairment in folate-deficient rats corresponds to depleted brain phosphatidylcholine and is prevented by dietary methionine without lowering plasma homocysteine. J Nutr. 2008;138:2502–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Beaudin AE, Stover PJ. Folate-mediated one-carbon metabolism and neural tube defects: balancing genome synthesis and gene expression. Birth Defects Res C Embryo Today. 2007;81:183–203. [DOI] [PubMed] [Google Scholar]
35.Smith AD. The worldwide challenge of the dementias: a role for B vitamins and homocysteine? Food Nutr Bull. 2008;29:S143–72. [DOI] [PubMed] [Google Scholar]
36.Partearroyo T, Perez-Miguelsanz J, Ubeda N, Valencia-Benitez M, Alonso-Aperte E, Varela-Moreiras G. Dietary folic acid intake differentially affects methionine metabolism markers and hippoccampus morphology in aged rats. Eur J Nutr. 2013;52:1157–67. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Online Supporting Material

supp_143_7_1028__index.html^{(723B, html)}

jn.113.174417_nutrition174417SupplementaryData1.pdf^{(259.8KB, pdf)}

[bib1] 1.Kronenberg G, Colla M, Endres M. Folic acid, neurodegenerative and neuropsychiatric disease. Curr Mol Med. 2009;9:315–23. [DOI] [PubMed] [Google Scholar]

[bib2] 2.Ramos MI, Allen LH, Mungas DM, Jagust WJ, Haan MN, Green R, Miller JW. Low folate status is associated with impaired cognitive function and dementia in the Sacramento Area Latino Study on Aging. Am J Clin Nutr. 2005;82:1346–52. [DOI] [PubMed] [Google Scholar]

[bib3] 3.Durga J, Van Boxtel MP, Schouten EG, Kok FJ, Jolles J, Katan MB, Verhoef P. Effect of 3-year folic acid supplementation on cognitive function in older adults in the FACIT trial: a randomised, double blind, controlled trial. Lancet. 2007;369:208–16. [DOI] [PubMed] [Google Scholar]

[bib4] 4.Kuhn HG, Dickinson-Anson H, Gage FH. Neurogenesis in the dentate gyrus of the adult rat: age-related decrease of neuronal progenitor proliferation. J Neurosci. 1996;16:2027–33. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib5] 5.Bondolfi L, Ermini F, Long JM, Ingram DK, Jucker M. Impact of age and caloric restriction on neurogenesis in the dentate gyrus of C57BL/6 mice. Neurobiol Aging. 2004;25:333–40. [DOI] [PubMed] [Google Scholar]

[bib6] 6.Shors TJ, Townsend DA, Zhao M, Kozorovitskiy Y, Gould E. Neurogenesis may relate to some but not all types of hippocampal-dependent learning. Hippocampus. 2002;12:578–84. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib7] 7.Shors TJ, Miesegaes G, Beylin A, Zhao M, Rydel T, Gould E. Neurogenesis in the adult is involved in the formation of trace memories. Nature. 2001;410:372–6. [DOI] [PubMed] [Google Scholar]

[bib8] 8.Na YJ, Jin JK, Lee YJ, Choi EK, Carp RI, Kim YS. Increased neurogenesis in brains of scrapie-infected mice. Neurosci Lett. 2009;449:66–70. [DOI] [PubMed] [Google Scholar]

[bib9] 9.Jin K, Peel AL, Mao XO, Xie L, Cottrell BA, Henshall DC, Greenberg DA. Increased hippocampal neurogenesis in Alzheimer's disease. Proc Natl Acad Sci USA. 2004;101:343–7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib10] 10.Liu BF, Gao EJ, Zeng XZ, Ji M, Cai Q, Lu Q, Yang H, Xu QY. Proliferation of neural precursors in the subventricular zone after chemical lesions of the nigrostriatal pathway in rat brain. Brain Res. 2006;1106:30–9. [DOI] [PubMed] [Google Scholar]

[bib11] 11.Kruman II, Mouton PR, Emokpae R, Jr, Cutler RG, Mattson MP. Folate deficiency inhibits proliferation of adult hippocampal progenitors. Neuroreport. 2005;16:1055–9. [DOI] [PubMed] [Google Scholar]

[bib12] 12.Kronenberg G, Harms C, Sobol RW, Cardozo-Pelaez F, Linhart H, Winter B, Balkaya M, Gertz K, Gay SB, Cox D, et al. Folate deficiency induces neurodegeneration and brain dysfunction in mice lacking uracil DNA glycosylase. J Neurosci. 2008;28:7219–30. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib13] 13.Fox JT, Stover PJ. Folate-mediated one-carbon metabolism. Vitam Horm. 2008;79:1–44. [DOI] [PubMed] [Google Scholar]

[bib14] 14.Kruman II, Kumaravel TS, Lohani A, Pedersen WA, Cutler RG, Kruman Y, Haughey N, Lee J, Evans M, Mattson MP. Folic acid deficiency and homocysteine impair DNA repair in hippocampal neurons and sensitize them to amyloid toxicity in experimental models of Alzheimer's disease. J Neurosci. 2002;22:1752–62. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib15] 15.Rabaneda LG, Carrasco M, Lopez-Toledano MA, Murillo-Carretero M, Ruiz FA, Estrada C, Castro C. Homocysteine inhibits proliferation of neuronal precursors in the mouse adult brain by impairing the basic fibroblast growth factor signaling cascade and reducing extracellular regulated kinase 1/2-dependent cyclin E expression. FASEB J. 2008;22:3823–35. [DOI] [PubMed] [Google Scholar]

[bib16] 16.MacFarlane AJ, Liu X, Perry CA, Flodby P, Allen RH, Stabler SP, Stover PJ. Cytoplasmic serine hydroxymethyltransferase regulates the metabolic partitioning of methylenetetrahydrofolate but is not essential in mice. J Biol Chem. 2008;283:25846–53. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib17] 17.Herbig K, Chiang EP, Lee LR, Hills J, Shane B, Stover PJ. Cytoplasmic serine hydroxymethyltransferase mediates competition between folate-dependent deoxyribonucleotide and S-adenosylmethionine biosyntheses. J Biol Chem. 2002;277:38381–9. [DOI] [PubMed] [Google Scholar]

[bib18] 18.Anderson DD, Stover PJ. SHMT1 and SHMT2 are functionally redundant in nuclear de novo thymidylate biosynthesis. PLoS ONE. 2009;4:e5839. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib19] 19.Anderson DD, Eom JY, Stover PJ. Competition between sumoylation and ubiquitination of serine hydroxymethyltransferase 1 determines its nuclear localization and its accumulation in the nucleus. J Biol Chem. 2012;287:4790–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib20] 20.Woeller CF, Anderson DD, Szebenyi DM, Stover PJ. Evidence for small ubiquitin-like modifier-dependent nuclear import of the thymidylate biosynthesis pathway. J Biol Chem. 2007;282:17623–31. [DOI] [PubMed] [Google Scholar]

[bib21] 21.Macfarlane AJ, Perry CA, Mcentee MF, Lin DM, Stover PJ. Shmt1 heterozygosity impairs folate-dependent thymidylate synthesis capacity and modifies risk of Apc(min)-mediated intestinal cancer risk. Cancer Res. 2011;71:2098–107. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib22] 22.Beaudin AE, Abarinov EV, Noden DM, Perry CA, Chu S, Stabler SP, Allen RH, Stover PJ. Shmt1 and de novo thymidylate biosynthesis underlie folate-responsive neural tube defects in mice. Am J Clin Nutr. 2011;93:789–98. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib23] 23.Williams EO, Sickles HM, Dooley AL, Palumbos S, Bisogni AJ, Lin DM. Delta protocadherin 10 is regulated by activity in the mouse main olfactory system. Front Neural Circuits. 2011;5:9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib24] 24.Kim JJ, Fanselow MS. Modality-specific retrograde amnesia of fear. Science. 1992;256:675–7. [DOI] [PubMed] [Google Scholar]

[bib25] 25.Phillips RG, Ledoux JE. Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behav Neurosci. 1992;106:274–85. [DOI] [PubMed] [Google Scholar]

[bib26] 26.Smith DR, Gallagher M, Stanton ME. Genetic background differences and nonassociative effects in mouse trace fear conditioning. Learn Mem. 2007;14:597–605. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib27] 27.Stabler SP, Lindenbaum J, Savage DG, Allen RH. Elevation of serum cystathionine levels in patients with cobalamin and folate deficiency. Blood. 1993;81:3404–13. [PubMed] [Google Scholar]

[bib28] 28.Allen RH, Stabler SP, Savage DG, Lindenbaum J. Elevation of 2-methylcitric acid I and II levels in serum, urine, and cerebrospinal fluid of patients with cobalamin deficiency. Metabolism. 1993;42:978–88. [DOI] [PubMed] [Google Scholar]

[bib29] 29.Suh JR, Oppenheim EW, Girgis S, Stover PJ. Purification and properties of a folate-catabolizing enzyme. J Biol Chem. 2000;275:35646–55. [DOI] [PubMed] [Google Scholar]

[bib30] 30.Cheng A, Shin-Ya K, Wan R, Tang SC, Miura T, Tang H, Khatri R, Gleichman M, Ouyang X, Liu D, et al. Telomere protection mechanisms change during neurogenesis and neuronal maturation: newly generated neurons are hypersensitive to telomere and DNA damage. J Neurosci. 2007;27:3722–33. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib31] 31.Axelrod J. Methylation reactions in the formation and metabolism of catecholamines and other biogenic amines. Pharmacol Rev. 1966;18:95–113. [PubMed] [Google Scholar]

[bib32] 32.Day JJ, Sweatt JD. Epigenetic mechanisms in cognition. Neuron. 2011;70:813–29. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib33] 33.Troen AM, Chao WH, Crivello NA, D'anci KE, Shukitt-Hale B, Smith DE, Selhub J, Rosenberg IH. Cognitive impairment in folate-deficient rats corresponds to depleted brain phosphatidylcholine and is prevented by dietary methionine without lowering plasma homocysteine. J Nutr. 2008;138:2502–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib34] 34.Beaudin AE, Stover PJ. Folate-mediated one-carbon metabolism and neural tube defects: balancing genome synthesis and gene expression. Birth Defects Res C Embryo Today. 2007;81:183–203. [DOI] [PubMed] [Google Scholar]

[bib35] 35.Smith AD. The worldwide challenge of the dementias: a role for B vitamins and homocysteine? Food Nutr Bull. 2008;29:S143–72. [DOI] [PubMed] [Google Scholar]

[bib36] 36.Partearroyo T, Perez-Miguelsanz J, Ubeda N, Valencia-Benitez M, Alonso-Aperte E, Varela-Moreiras G. Dietary folic acid intake differentially affects methionine metabolism markers and hippoccampus morphology in aged rats. Eur J Nutr. 2013;52:1157–67. [DOI] [PubMed] [Google Scholar]

PERMALINK

Disruption of Shmt1 Impairs Hippocampal Neurogenesis and Mnemonic Function in Mice1,2,3

Elena V Abarinov

Anna E Beaudin

Martha S Field

Cheryll A Perry

Robert H Allen

Sally P Stabler

Patrick J Stover