Elevated homocysteine is an important risk factor that increases cerebrovascular and neurodegenerative disease morbidity. In mammals, B vitamin supplementation can reduce homocysteine levels. Whether, and how, hibernating mammals, that essentially stop ingesting B vitamins, maintain homocysteine metabolism and avoid cerebrovascular impacts and neurodegeneration remain unclear. Here, we compare homocysteine levels in the brains of torpid bats, active bats and rats to identify the molecules involved in homocysteine homeostasis. We found that homocysteine does not elevate in torpid brains, despite declining vitamin B levels. At low levels of vitamin B6 and B12, we found no change in total expression level of the two main enzymes involved in homocysteine metabolism (methionine synthase and cystathionine β-synthase), but a 1.85-fold increase in the expression of the coenzyme-independent betaine-homocysteine S-methyltransferase (BHMT). BHMT expression was observed in the amygdala of basal ganglia and the cerebral cortex where BHMT levels were clearly elevated during torpor. This is the first report of BHMT protein expression in the brain and suggests that BHMT modulates homocysteine in the brains of hibernating bats. BHMT may have a neuroprotective role in the brains of hibernating mammals and further research on this system could expand our biomedical understanding of certain cerebrovascular and neurodegenerative disease processes.
Citation: Zhang Y, Zhu T, Wang L, Pan Y-H, Zhang S (2013) Homocysteine Homeostasis and Betaine-Homocysteine S-Methyltransferase Expression in the Brain of Hibernating Bats. PLoS ONE 8(12): e85632. doi:10.1371/journal.pone.0085632
Editor: Michelle L. Baker, CSIRO, Australia
Received: July 28, 2013; Accepted: December 5, 2013; Published: December 23, 2013
Copyright: © 2013 Zhang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was supported by grants of Shanghai Excellent Academic Leaders Project (Grant No. 11XD1402000) to SZ, the National Natural Science Foundation of China (Grant No. 31100273/C030101) to YP. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Homocysteine is a sulfur-containing amino acid and a risk factor involved in cerebrovascular and neurodegenerative diseases [1–5]. The elevation of homocysteine may result in the dysfunction of endothelial and smooth muscle cells in the vascular wall. Endothelial injuries such as the inhibition of cellular binding sites for tissue plasminogen activator , decreasing expression of thrombomodulin [1,7], and production of endoplasmic reticulum stress and growth arrest  are observed in hyperhomocysteinemic animals. The effect of hemostasis, induced by homocysteine, promotes blood clotting and reduces fibrinolysis . Elevated homocysteine impairs smooth muscle cells by inducing a proliferative state, and migration from the media to the intima of the vessel . These homocysteine-induced changes may initiate the pathogenesis of atherosclerosis  which can lead to vascular diseases in the brain. Elevated homocysteine expression is also known to play an important role in neurodegenerative diseases including brain atrophy, dementia and cognitive impairment [4,11,12]. For example, homocysteine is elevated in patients with confirmed Alzheimer’s disease [11,13,14]. Homocysteine causes excitotoxic and oxidative injury to hippocampal neurons in cell cultures and in vivo , and hyperhomocysteinemia due to dietary folate deficiency endangers dopaminergic neurons in models of Parkinson’s disease .
Homocysteine metabolism involves remethylation or transsulfuration (Figure 1) [17,18]. During remethylation, homocysteine is methylated to methionine by methionine synthase (MS, EC 126.96.36.199), which is ubiquitous, or by betaine-homocysteine S-methyltransferase (BHMT, EC 188.8.131.52), whose expression is mainly restricted to the liver and kidney . During transsulfuration, homocysteine is irreversibly converted into cysteine by cystathionine β-synthase (CBS, EC 184.108.40.206) and cystathionine γ-lyase (CγL, EC 220.127.116.11) in the liver, kidney, pancreas and small intestine . Both the remethylation and transsulfuration of homocysteine involve B vitamins: methionine synthase requires folic acid and vitamin B12 as substrates or cofactors, and cystathionine β-synthase is a vitamin B6-dependent heme protein. Inadequate levels of one or more B vitamins contribute to elevated homocysteine levels and neurological damage [20,21]
Figure 1. Homocysteine metabolism.
Homocysteine is metabolized to methionine by remethylation and cystathionine by transsulfuration. Coenzymes are shown in gray. BHMT, betaine-homocysteine S-methyltransferase; DMG, dimethylglycine; MAT, methionine adenosyltransferase; SAM, S-adenosylmethionine; SAH, S-adenosylhomocysteine; SAHH, S-adenosylhomocysteine hydrolase; MS, methionine synthase; THF, tetrahydrofolate; SHMT, serine hydroxymethyltransferase; CH2THF, methylene tetrahydrofolate; CH3THF, methyl tetrahydrofolate; CBS, cystathionine β-synthase; CγL, cystathionine γ-lyase.doi:10.1371/journal.pone.0085632.g001
Some mammals reduce body temperature, metabolic rate and other physiological processes during hibernation, which is an adaptive strategy in response to winter [22–24]. Hibernation involves fasting  and may therefore be a critical time during which the metabolism of homocysteine is inhibited because of B vitamin deficiency. How fasting mammals avoid elevated homocysteine and its negative neurological impacts remains unknown.
Bats are the second most abundant mammal species on earth and the majority of microbats hibernate . Recent research has revealed protective mechanisms in the brain tissue of hibernators [26–28], and hibernating species are emerging as ideal research models for neurological disease [29–32]. In this study, we investigate brain tissue in torpid and active Rickett’s big-footed bats (Myotis ricketti) to determine patterns of homocysteine homeostasis and metabolism during fasting. We aimed to (1) identify whether BHMT, MS and CBS are involved in homocysteine metabolism in the brain and describe expression patterns in different brain regions, and (2) give the likely role and importance of these enzymes in homocysteine regulation, describe the nature of selection operating on underlying genes. While expanding our understanding of adaptations to hibernation and fasting, this research will also contribute to the broader biomedical prevention and treatment of human cerebrovascular and neurodegenerative diseases.
Materials and Methods
All procedures involving the capture of bats and collection of samples were carried out in strict accordance with the Guidelines and Regulations for the Administration of Laboratory Animals (Decree No. 2, the State Science and Technology Commission of the People’s Republic of China on November 14, 1988) and were approved by the Animal Ethics Committee at East China Normal University (ID no: AR2012/03001).
Collection of animals and tissues
Hibernating male M. ricketti (n = 14) were captured from Fangshan Caves (39°48′ N, 115°42′ E), Beijing, China. Seven animals were immediately euthanized and their core body (Tb) and surface (Ta) temperatures were 13 ± 2°C and 9 ± 1°C, respectively. Seven M. ricketti were euthanized 48 h after arousal and their core body temperature (Tb) was 35 ± 2 °C. Non-hibernating Rousettus leschenaultii (n = 4) were captured in Mashan county (23°55′ N, 108°26′ E), Guangxi, China. These animals (Tb = 35 ± 2 °C) were sacrificed 48 h after capture by cervical dislocation. Male rats (n = 4) were obtained from Sino-British Sippr/BK Lab Animal Ltd (Shanghai, China). Brain tissue was rapidly removed and collected in 2 ml cryotubes and stored at -80°C.
B vitamins, homocysteine and betaine assays
Vitamin B6, vitamin B12, folate, homocysteine and betaine levels in brain tissue and blood were determined using vitamin B6 (TSZ-E80574), vitamin B12 (TSZ-E80166), folate (TSZ-E80542), homocysteine (TSZ-E30175) and betaine (TSZ- S25623) assay kits (Yanyu Chemical Reagent Inc., Shanghai) according to the manufacturer’s instructions. Brain tissue (20 mg) was homogenized in 200 μl PBS (137 mM NaCl, 2.7 mM KCl, and 10 mM Na2HPO4). The homogenate was then centrifuged at 14,000 xg, 4°C for 15 min, and the supernatant was assayed for B vitamins, homocysteine and betaine levels . Blood was obtained by cardiac puncture and centrifuged for 15 min, and plasma was frozen at -80°C for use. The 96-well microplate was incubated at 37°C for 30 min each with sample and HRP-reagent. After adding A and B developer buffers, the reaction last 10 min, and was stopped using stop solution. OD450 was measured within 15 min. B vitamins, homocysteine and betaine concentrations in samples were quantified by comparing OD450 values against those of known concentration. Results (mean ± SD) are from three separate experiments (n = 3 per group) and analyzed using Student’s t-tests (two-tailed). A P value < 0.05 was considered significant.
Brain proteins were extracted from bats and rats. The tissue was homogenized by lysis buffer (0.22 M Tris-HCl (pH 6.8), 8.8% SDS, 44.4% glycerol) and centrifuged at 12,000 xg, 4°C for 10 min. The supernatant was heated at 100°C for 10 min and then used for western blotting. Brain proteins of the rat were used as a positive loading control. Proteins in each sample (10 μg/lane) were separated by 10% SDS-PAGE and then transferred onto 0.2 μm PVDF membranes (Millipore, USA) with an electro-blotting apparatus (GenePure, Taiwan). The PVDF membranes were blocked in blocking solution containing 5% skim milk and 1% BSA at 4°C for 12 h, and then reacted with a series of primary antibodies including anti-BHMT (1:2500), anti-MS (1:250), and anti-CBS (1:1000). The anti-BHMT (ab96415) and anti-MS (ab9209) were acquired from Abcam Corporation, and anti-CBS (sc-67154) was purchased from Santa Cruz Biotechnology, Inc. Antibodies were selected based on the ability to combine with conserved epitopes of target proteins of many mammalian species. After washing, blots were reacted with appropriate secondary antibodies and visualized according to the instructions of the ImmobilonTM Western Chemiluminescence HRP substrate kit (Millipore, USA). Images were captured using ImageQuantTM LAS-4000 (Amersham Biosciences, USA), and detected bands were quantified with ImageQuantTM TL (v 7.0, Amersham Biosciences, USA). The intensity of each band was normalized to β-actin (sc-47778, 1:5000, Santa Cruz Biotechnology Corporation). Results (mean ± SD) were calculated from four repeats and then analyzed using Student’s t-tests (two-tailed). A P value < 0.05 was considered statistically significant.
Sample preparation and BHMT activity measurement were carried out following standard protocols [34–36] with several modifications. Brain tissue (0.1 g) was homogenized in 500 μl potassium phosphate buffer (40 mM, pH 7.5) containing 1 mM DTT. The homogenate was centrifuged at 18,000 xg, 4°C for 15 min, and the supernatant fraction was used for BHMT assay. The protein concentration of each sample was assayed with the Quick StartTM Bradford protein assay kit (Bio-Rad, USA) according to the manufacturer’s instructions.
The 280 μl standard mixture contained 10 mM DL-Hcy, 3.25 mM betaine, 50 mM Tris-HCl (pH 7.5) and preparing sample (80 μl). Assays were started by transferring the tubes to a 37°C water bath for 1–2 h. Following incubation, the mixture was chilled in ice water and centrifuged at 12,000 xg, 4°C for 15 min. Phenyl isothiocyanate as the derivatization reagent, was added to the supernatant. After standing at ambient temperature for 10 min, n-hexane was used to remove organic substances, and the water-soluble substances were filtered (0.22-μm filter) for use.
Samples were analyzed on a HPLC system equipped with separations module (Waters e2695) and a PntulipsTM BP-C18 (5 μm, 4.6 x 250 mm) with a 0.8 ml/min flow rate. Methionine was monitored by a photodiode array detector (Waters 2998) with an excitation wavelength of 254 nm. HPLC data were collected and analyzed using the Great Resource Health Standard Inc. (Shanghai) database. Methionine in a sample was identified and quantified by comparing the peak retention time and peak volume of the sample to internal standard methionine. Blanks contained all of the components except for the reactants DL-Hcy and betaine, and their values were subtracted from sample values. We compared the amount of methionine to reflect brain BHMT activity in torpid and active M. ricketti (n = 3 per group). All samples were assayed in three repeats, and data analyzed using two-tailed Student’s t-tests. Enzyme activity is expressed as nmol/h/mg protein.
Brain tissue was fixed in 4% paraformaldehyde buffer solution for 24 h after sampling. A series of graded alcohols and xylene were used for dehydration and clearing of tissue. Brain serial-sections (6 μm) were prepared after being paraffin-embedded. Hydrogen peroxide solution (3%) was used to stop endogenous peroxidase activity after the brain sections were rehydrated, and then blocked with bull serum albumin (0.3%). After rinsing with phosphate buffer solution (PBS, 0.1 M/L), brain sections were incubated in affinity-purified primary antibody (anti-BHMT, 1:1000, abcam, ab96415) diluent overnight at 4°C. Simultaneously, normal rabbit IgG (sc-2027, 1:2000, Santa Cruz, USA) and PBS were used to replace the primary antibody in control groups. After washing with 0.1 M/L PBS, slices were reacted with corresponding secondary antibody (about 45 min at 37°C) and then antigen-antibody complexes were detected using ChemMateTM EnVisonTM/HRP complex (including diaminobenzidine, DAB) as a peroxidase substrate (GK500705, Gene Tech). Results were visualized under an optical microscope (Lecia DM2500, Germany). All pictures were taken under the same microscope with uniform parameters after brain sections were mounted with xylene sealant.
Because bat brain atlases are not available, the mouse brain atlas  was used to localize the distribution of BHMT. We evaluated immunostaining for BHMT in the brains obtained from torpid and active M. ricketti bats (n = 3 per group). Images of the slices at 50× magnification were captured and analyzed using Image Pro Plus (Immagine Computer, Italy). Staining intensity was computed as integrated optical density (IOD) from four sections of expressed regions. The IOD was calculated from three arbitrary fields with the same area for each section. Data were analyzed using Student’s t-tests (two-tailed).
Molecular evolution analyses and homology modeling of bat BHMT
The coding regions of Bhmt were sequenced for 12 bat species from seven families (Table S1), including three non-hibernating species: Eonycteris spelaea, Rousettus leschenaultii and Cynopterus sphinx (Pteropodidae); and nine hibernating species: Taphozous melanopogon (Emballonuridae), Miniopterus fuliginosus (Miniopteridae), Pipistrellus pipistrellus and Myotis ricketti (Vespertilionidae), Artibeus lituratus and Leptonycteris yerbabuenae (Phyllostomatidae), Hipposideros armiger and Hipposideros pratti (Hipposideridae), and Rhinolophus ferrumequinum (Rhinolophidae). The available published sequences (in Ensembl database) of two other species, Pteropus vampyrus (Pteropodidae) and Myotis lucifugus (Vespertilionidae), were also incorporated in molecular analyses. Details (BHMT accession numbers and thermal physiology) of all species are listed in Table S1.
We selected M. ricketti to identify sequences of Bhmt from the brain and liver, and found they were consistent. Some bat species do not express BHMT in brain tissue, and so all cDNA sequences are cloned from liver tissue. Following standard protocols, total RNAs were isolated from liver tissue using Trizol reagent (Invitrogen) and were reverse-transcribed to cDNA with the RNAiso Plus kit (TaKaRa). Using primers (Table S2), all PCR products were isolated and puriﬁed using the Gel Extraction kit (Qiagen), and then ligated into pGEM-T Easy Vector (Promega). Correct recombinant clones were sequenced using the Terminator kit on an ABI 3730 DNA sequencer (Applied Biosystems).
The BHMT nucleotide sequences of 14 species were aligned using Clustal X . The pairwise dN/dS ratio (ω value, non-synonymous substitution rate/synonymous substitution rate) was calculated using Swaap v1.0.3 to determine selective pressure acting on Bhmt [39,40]. Data were analysed using two-tailed Student’s t-tests and a P value < 0.05 was considered significant.
To determine whether amino acid residue divergence between hibernating and non-hibernating bats is related to BHMT function we simulated BHMT structure in bats. The amino acid sequences of bat BHMT were deduced from corresponding nucleotide sequences. Structure-based amino acid sequence alignments were carried out by T-Coffee (Expresso mode) to determine the amino acid residue conserved in hibernators but diverged in non-hibernators . The BHMT amino acid sequence of Rattus norvegicus with known structure (1UMY, O09171), was selected as the template for alignment. The corresponding amino acid sequence files of M. ricketti were imported into SWISS-MODEL for homology modeling of BHMT structures ; PyMol was used to display the 3D structures and illustrate model results.
Vitamin B, homocysteine and betaine levels in torpid M. ricketti
We measured levels of vitamin B6, vitamin B12 and folate in torpid M. ricketti and non-torpid/non-fasting M. ricketti, R. leschenaultia and rats. We found that torpid M. ricketti had lower levels of B6 and B12 (vitamin B6: 4.09 ± 0.08 ng/g; vitamin B12: 15.03 ± 1.11 ng/g) compared to non-torpid M. ricketti (vitamin B6: P < 0.001; vitamin B12: P < 0.001) (Table 1). There was no difference in the folate concentration in both groups of M. ricketti (torpid: 194.95 ± 3.75 ng/g; active: 191.46 ± 8.17 ng/g) (Table 1).
|Animal groupa,b||Vitamin B6||Vitamin B12||Folate|
|M. ricketti (AF)||4.80 ± 0.09||17.49 ± 0.38||194.95 ± 3.75|
|M. ricketti (H-NF)||4.09 ± 0.08d||15.03 ± 1.11d||191.46 ± 8.17|
|R. leschenaultii (AF)||5.60 ± 0.16||21.77 ± 0.75||224.34 ± 27.45|
|R. norvegicus (AF)||4.80 ± 0.36||17.46 ± 0.30||197.48 ± 10.23|
Brain homocysteine levels in M. ricketti during torpor were approximately 10% lower than active M. ricketti (0.137 ± 0.003 vs. 0.153 ± 0.003 μmol/g, P < 0.001) (Figure 2A). Plasma homocysteine was also decreased in torpid M. ricketti compared to active M. ricketti (10.277 ± 0.774 vs. 11.826 ± 0.662μmol/L, P < 0.001) (Figure 2B). No difference was found between active M. ricketti and rats (0.150 ± 0.013 μmol/g, 12.122 ± 0.453 μmol/L) in brain tissue and blood. Homocysteine levels in R. leschenaultia (0.203 ± 0.017 μmol/g) were approximately 33–48 % higher than torpid and active M. ricketti and rats in the brain (P < 0.001), but there was no significant difference in plasma homocysteine among R. leschenaultia (12.345 ± 0.604 μmol/L), active M. ricketti and rats.
Figure 2. Expression levels of homocysteine in brain and plasma.
Homocysteine expression levels were determined in the brains (A) and plasma (B) of torpid M. ricketti (MT), active M. ricketti (MA), R. leschenaultii (R) and rats. Data are expressed as mean ± SD from three separate experiments with n=3 per group. The letters (a,b) between groups represent statistical differences P < 0.001. ** denotes statistical significance (P < 0.001) between torpid and active M. ricketti.doi:10.1371/journal.pone.0085632.g002
The expression pattern of betaine was different between brain tissue and plasma (Figure 3A, B). Brain betaine levels in torpid M. ricketti (6.239 ± 0.626 ng/g) were similar to content in active animals (5.859 ± 0.430 ng/g), but torpid M. ricketti had lower levels of plasma betaine than active M. ricketti (1.879 ± 0.501 vs. 3.415 ± 0.292 ng/mL, P < 0.001). Betaine levels in the brains of active M. ricketti were higher compared to the rat brain (4.212 ± 0.413 ng/g, P < 0.001), but betaine levels in plasma of M. ricketti were lower than rats (6.042 ± 0.731 ng/mL, P < 0.001). Betaine levels of R. leschenaultia in the brain (3.015 ± 0.560 ng/g) and plasma (1.556 ± 0.122 ng/mL) were lower than active M. ricketti and rats (P < 0.001).
Figure 3. Betaine levels in brain and plasma.
Brain betaine (A) and plasma betaine (B) were tested in torpid M. ricketti (MT), active M. ricketti (MA), R. leschenaultii (R) and rats. Three (n = 3) animals in each group were analysed. Results are expressed as mean ± SD of three repeats. The letters (a,b) between groups represent statistical differences P < 0.001. ** denotes statistical significance (P < 0.001) between groups.doi:10.1371/journal.pone.0085632.g003
Western blot validation of enzymes involved in homocysteine metabolism
The expression levels of BHMT, MS and CBS were investigated in the brain of hibernating bats (Figure 4A). BHMT levels were 1.85-fold higher in torpid M. ricketti compared to active M. ricketti (Figure 4B); BHMT was not expressed in non-hibernating R. leschenaultia and rats (Figure 4A). There was no difference in total expression level of MS and CBS of torpid and active M. ricketti, and levels were similar to those in rats. Two clear bands of CBS near 63 kDa were detected and the amount of the bands showed reciprocal expression patterns between torpid and active states of M. ricketti bats. In R. leschenaultia, MS expression levels were higher and CBS expression levels lower compared to M. ricketti and rats (Figure 4B).
Figure 4. Western blot analysis.
Expression levels of BHMT, MS and CBS were determined by Western blot (A). MT, torpid M. ricketti; MA, active M. ricketti; R, R. leschenaultii. All samples are from brain tissue. Arrows indicate predicated molecular weight (kDa) of proteins. (B) Relative expression levels of the proteins are represented as mean ± SD. Statistical significance by two tailed Student’s t-tests: *P<0.05, **P<0.001.doi:10.1371/journal.pone.0085632.g004
BHMT activity assay
More methionine was produced in torpid M. ricketti compared to active M. ricketti (0.432 ± 0.105 vs. 0.201 ± 0.054 nmol/h/mg protein) in brain tissue. Relative activity of BHMT was calculated by dividing the production amount of methionine by the relative expression level of BHMT, and revealed that there was no significant difference in BHMT relative activity between torpid and active M. ricketti (0.233 ± 0.057 vs. 0.201 ± 0.054 nmol/h/mg protein) (Figure 5). In addition, we noticed that the product level was lower in brain tissue compared to liver tissue , which may be due to BHMT expression levels in the brain being at least 100-fold lower than in the liver (unpublished data).
Figure 5. BHMT activity assay.
The production amount of methionine represents BHMT relative activity. All results (mean ± SD) are from three separate experiments with n = 3 per group. Torpid and active states of M. ricketti are tested. Statistical significance is determined by Student’s t-tests (two tailed). * denotes nmol/h/mg protein.doi:10.1371/journal.pone.0085632.g005
MS and CBS is usually expressed in mammalian brains [3,43], and the expression levels in the brain tissues of M. ricketti were found to be similar to that in rats. However, BHMT is only expressed in M. ricketti brains, and previous experiments showed that BHMT is mainly detected in the liver and kidney . To identify the distribution of BHMT in neuroanatomical locations that may be involved in the hibernation process, brain tissues from torpid and active M. ricketti were stained for BHMT expression. We found that BHMT was abundantly expressed in the telencephalon, including some regions of the cerebral cortex (such as the retrosplenial granular cortex, retrosplenial agranular cortex, primary motor cortex, secondary motor cortex, medial parietal association cortex, lateral parietal association cortex and primary somatosensory cortex, hindlimb region) (Figures 6A) and the amygdala of basal ganglia (Figures 6C). Integrated optical density of BHMT staining was increased in the cerebral cortex and no significant change was found in the amygdala of torpid M. ricketti (Figure 6B, D).
Figure 6. Immunohistochemical detection of BHMT in the brains of M. ricketti.
After series of coronal sections, BHMT expression was identified in the cerebral cortex (Cx) and the amygdala (Am) of basal ganglia between torpid and active states in M. ricketti. Immunopositivity was dark-brown in the cytoplasm of cells. The four photomicrographs indicate encephalic regions and states: cerebral cortex, torpid state (A, left lane); cerebral cortex, active state (A, right lane); amygdala, torpid state (C, left lane); amygdala, active state (C, right lane). Quantitative analysis of BHMT immunostaining in cerebral cortex (B) and amygdala (D). Three (n = 3) animals in each group were analysed. Relative integrated optical densities are shown as mean ± SD. *P < 0.05 (two tailed Student’s t-tests). CxT, the cerebral cortex of torpid M. ricketti; CxA, the cerebral cortex of active M. ricketti; AmT, the amygdala of torpid M. ricketti; AmA, the amygdala of active M. ricketti.doi:10.1371/journal.pone.0085632.g006
Selection pressure on Bhmt and structure analysis
To investigate the evolution track of BHMT, the selection pressures on Bhmt of hibernating and non-hibernating bats was calculated. The encoding regions of Bhmt were cloned, sequenced and analysed from four non-hibernating and ten hibernating species of bat to determine Bhmt adaptation to hibernation (Table S1). To compare differences in selection pressure, we divided sequences into two groups (non-hibernation and hibernation) based on clear physiological characteristics (Figure S1 and Table S1). We found that BHMT is under strong purifying selection and the ω (dN/dS) value was lower than 0.18. A stronger selection constraint was found in hibernators than non-hibernators (Figure 7).
Figure 7. Selective pressure on Bhmt.
Selective pressure on BHMT was determined by comparing the ω value (ω=dN/dS, dN: non-synonymous substitution rate, dS: synonymous substitution rate) for several species of non-hibernating (black bar) and hibernating (gray bar) bats. Data are presented as mean ± SD. Statistical significance was determined by Student’s t-tests (two tailed), **P < 0.001.doi:10.1371/journal.pone.0085632.g007
We aligned corresponding amino acid sequences of BHMT on the basis of required sequences (Figure S1). This showed high similarity between BHMT in the bat and rat whereby M. ricketti BHMT had a 96% similarity to rat BHMT (Figure 8A, Figure S1). Homology modeling revealed bat BHMT is composed of four identical subunits, which consisted mostly of a (α/β)8 barrel (Figure 8A) . Most amino acids were conserved in bats, but four amino acid sites (at positions 78, 149, 150 and 330) were different between non-hibernating and hibernating bats (Figure 8). Position 78 was changed from Ala in hibernating bats to Gly in non-hibernating bats (Figure 8B). Position 149 and 150 locating in α3 helix were variable from Leu/Met and Lys to Ile and Arg, respectively. Ser330 was situated in segment (residues 316-349) in hibernating bats, and was substituted by Thr330 or Pro330 in non-hibernating bats (Figure 8C).
Figure 8. Homology modeling of bat BHMT.
(A) BHMT tetramer of bat consists of chain A (red), B (green), C (orange) and D (purple). Bat BHMT was superposed to the 3D-structure of BHMT of rat (Rattus norvegicus, O09171) (gray). Four amino acid residues (Ala78, Leu/Met149, Lys150 and Ser330) that are conserved in all hibernators of bats but are varied among non-hibernators of bats are shown in blue. (B) Amino acid residue at position 78 linking the residues Phe76 and Tyr77 for substrate binding and loop L2 (residues 79–95) is Ala78 in hibernating bats (upper lane) but is Gly78 in non-hibernating bats (lower lane), in which the side chain distances between positions 78 and their adjacent residues Phe76 and Tyr77 are rearranged. (C) Ser330 in hibernating bats is different from Thr330 or Pro330 in non-hibernating bats, which may influence binding of the dimers AB to CD, especially the substitution of Pro330. Glu391 and Arg395 of chain D were stacked to the segment 316–349 of chain B, and dotted lines indicate the distance from Glu391/Arg395 to residues at position 330. All structures are represented by PyMol.doi:10.1371/journal.pone.0085632.g008
This study confirms that homocysteine does not accumulate in brain tissue of torpid M. ricketti, despite declining B vitamin levels from fasting. One reason for this pattern could be that the key catalyzing homocysteine enzymes BHMT, MS and CBS do not decline during hibernation. Another explanation is the role of other molecules in the control of hibernation. For example, one of the products of homocysteine metabolism, methionine, is an essential amino acid in mammals because it is not synthesized de novo . Methionine and its metabolic products are involved in multiple fundamental biological processes (Figure 1)  and methionine is catalized to become S-adenosylmethionine (SAM, an important methyl donor) by methionine adenosyltransferase (MAT). Several studies have demonstrated that the transcription and translation of many genes are regulated by methylation during mammalian hibernation. Methylated DNA has a strong impact on gene expression . Hibernation-nonresponsive genes show enhanced methylation of their promoter regions to switch off transcription . In addition, some enzymes, such as protein phosphatase 2A (PP2A) involved in tau phosphorylation in hibernation, can be activated by methylation of its catalytic subunit . Therefore, methylation may require more SAM, which promotes homocysteine methylation and results in a decrease in homocysteine in brain tissue.
The amount of homocysteine in the brain of non-hibernating R. leschenaultia is higher than for M. ricketti and rats. It is possible that the normal range of homocysteine is wider in R. leschenaultia or there exists other mechanisms to prevent homocysteine injury. BHMT was not detected in the brains of R. leschenaultia and rats, but MS is higher in R. leschenaultia brain tissue than M. ricketti, suggesting that MS is crucial to the homocysteine methylation pathway when BHMT is absent. In addition, we found that brain homocysteine, MS and CBS levels in torpid M. ricketti were similar to rats (Figures 2, 4) except for BHMT expression. Whether the normal range of homocysteine homeostasis in hibernating bats can be judged using parameters from rats needs to be verified.
Hyperhomocysteinemia has been diagnosed based on the plasma homocysteine concentration in clinical medicine . In this study, we found that plasma homocysteine is not increased in M. ricketti during hibernation and indicate that homocysteine concentration is maintained at a normal level in blood and brain tissues regardless of whether M. ricketti hibernates, which is beneficial to protect hibernators from atherothrombosis and thrombosis. The amount of plasma homocysteine in non-hibernating R. leschenaultia is close to active M. ricketti and rats, which is different from the condition in the brain between these species. The data reported here suggest that homocysteine levels are not consistent between blood and local tissue, and it is necessary to measure homocysteine in both.
Moreover, BHMT expression is elevated in the brains of torpid M. ricketti, compared to active animals, and its relative activity does not differ across the two states. The BHMT/betaine system exhibits a protective effect from homocysteine-induced injury  and BHMT deficiency leads to the elevation of total homocysteine concentration in Bhmt-/- mice . BHMT is a key modulator of homocysteine when expression or activity of other enzymes is abnormal [48,49]. The inherent kinetic properties of the enzyme are likely important  and compared with MS and CBS, BHMT has a lower Km for sulfur-containing substrates . BHMT can utilize homocysteine at relatively low concentrations and has a high affinity for homocysteine. Elevated BHMT in the brain during hibernation strongly suggests that BHMT play a role in the regulation of homocyteine during this state.
Hibernation inhibits betaine sourced from food and overall reduced metabolism results in a decrease in plasma betaine during hibernation (Figure 3B). However, brain betaine levels were not different between torpid and active M. ricketti (Figure 3A). The physiologic function of betaine is either as the methyl donor involved in BHMT-catalyzed homocysteine metabolism or as an osmolyte protecting cells, proteins and enzymes from environmental stress (such as extreme temperature in hibernation) . The difference in betaine levels in the brain and plasma suggests that hibernators adjust the betaine distribution for homocysteine metabolism and brain protection. Furthermore, several studies have reported that the synthesis and reserves of acetylcholine are lower during hibernation than when active . Acetylcholine (as a neurotransmitter) may be broken down to choline and oxidized to betaine when nerve conduction is reduced during hibernation [19,22]. Therefore, BHMT-catalyzed reactions play an important role in betaine homeostasis in addition to homocysteine levels.
The total amount of MS and CBS did not vary between torpid and active M. ricketti, but their function could be influenced during hibernation. B vitamin supplementation (vitamin B6, vitamin B12 and folate) and strengthening of the transulfuration and transmethylation pathways could lower homocysteine levels and slow the accelerated rate of neurological disease [12,52,53]. However, B vitamins are not synthesized by hibernators themselves, and fasting during hibernation leads to reduced intake of vitamin B6 and vitamin B12 in torpid bats. The brain is not the main storage organ of B vitamins in bats , and they are not quickly transported to the brain because of the lower flow velocity of blood [32,55]. These factors may inhibit the involvement of MS and CBS in transmethylation and transsulfuration. Meanwhile, the transsulfuration pathway is inactivated when methyl transfer (such as the process involving SAM) requirements are high , consistent with the idea that homocysteine levels are usually determined by remethylation rather than transsulfuration during fasting [57,58]. Interestingly, following declining vitamin B6 and vitamin B12, the level of folate did not vary between torpid and active bats (Table 1). This observation could be explained by the “methyl trap” considering that vitamin B12 deficiency reduces the activity of MS. It leads to the accumulation at other folate forms [59,60] and could inhibit folate converting to tetrahydrofolate (THF). These processes may discount the roles of MS and CBS during hibernation and make the role of BHMT more important in homocysteine homeostasis. However, two bands of CBS detected in M. ricketti bats exhibit a reciprocal expression pattern between torpid and active states (Figure 4A), which suggests the occurrence of post-translational modifications (PTMs) on CBS. PTMs (e.g., sumoylation and methylation) are known to be intensively involved in regulation of CBS activities [61–64]. Whether catabolism of homocysteine by transsulfuration pathway is reduced in torpid bats due to decreased B vitamins and PTMs of CBS remains to be investigated. Thus, the increased level of BHMT, reduced activity of transsulfuration pathway, and lower production of homocysteine could together maintain the homocysteine level during bat torpor.
Immunocytochemistry revealed that the cerebral cortex and the amygdala of basal ganglia were the main encephalic regions of BHMT distribution. Elevated homocysteine levels may result in the formation of neurofibrillary tangles (NFTs), which are severe in the cerebral cortex and amygdala in AD patients . BHMT expression in these two regions may modulate homocyteine levels and decrease production of homocystine-induced NFTs in order to prevent relative neurological damage during hibernation.
Furthermore, Bhmt is found under strong purifying selection (ω < 0.18) and there exists a stronger selective constraint in hibernating bats (Figure 7), indicating that this gene is very conservative and reflects a higher functional constraint on BHMT protein . The significant difference in selective pressure between hibernators and non-hibernators may be the result of environmental stressors, such as low temperature and food shortage . This analysis supported the functional importance of BHMT during hibernation.
The functional importance of BHMT in control of homocysteine levels in the brain of torpid bats is further supported by the observation that the amino acid residues of BHMT is more conserved in hibernating bats than that in non-hibernating bats (Figure 8). Three (positions 78, 150 and 330) of four different sites (positions 78, 149, 150 and 330) were 100% conserved in all ten hibernating bats but were diverged in four non-hibernating bats compared to other mammal species (Figure S1). Amino acid Ala78 links loop L2 (residues 79–95) and residues Phe76 and Tyr77 that are critical for substrate binding (Figure 8B) [44,66]. Phe76 has been included in the betaine-binding site and involved in properly orienting the Hcy, and Tyr77 is contributing to the hydrophobic core for betaine and provides a hydrogen bond for betaine . On the other hand, loop L2 presents mobility by playing an essential role in Hcy binding . Therefore, Ala78 as a linker between the two regions may play a role to influence substrate binding. Amino acid site 330 located in segment 316–349 (chain B) is stacked to a helix (residues 381–407) of the interacting subunit (chain D), which contributes to the tetramerization of AB and CD . The amino acid change at position 330 may influence BHMT structural stability, especially when Ser330 was substituted by Pro330 (Figure 8C). The substitutions at positions 78 and 330 change the side chain distance between these residues and their adjacent amino acids, whether the alteration of van der Waals’ forces among the residues affects BHMT function remains to be investigated (Figure 8B, C).
We found for the first time that BHMT is expressed in the brains of mammals. Expression in the cerebral cortex and amygdala, and strong purifying selection in BHMT, suggest that hibernators regulate BHMT expression to decrease homocysteine-induced brain damage. This research builds on our understanding of homocysteine homeostasis and this is essential to develop treatments for human cerebrovascular and neurodegenerative diseases. It is clear that BHMT is important to homocysteine homeostasis in brain tissue and the distribution and function of BHMT in other tissues and species is an exciting area of further research.
Alignment of amino acid sequences of BHMT. Amino acid sequences ranging from 22 to 354 were aligned from bats and human. Amino acid site numbers are referenced to mature human BHMT. The amino acid sites indicated by asterisk (*) mean that the corresponding amino acids are different between hibernating (H) bat group and non-hibernating (N) bat group.
The accession numbers and heterothermy condition in bat species.
Primers used for amplification of Bhmt gene from 12 bat species.
We thank Jinyao Zeng for helpful discussion and Junpeng Zhang and Guangjian Zhu for sample collection. We thank Dr. Chen-Chung Liao for providing valuable advices. We also thank Qin Liu for comments on an earlier version of this manuscript.
Conceived and designed the experiments: YZ YP SZ. Performed the experiments: YZ TZ LW. Analyzed the data: YZ TZ YP. Contributed reagents/materials/analysis tools: SZ. Wrote the manuscript: YZ.
- 1. Lentz SR (2005) Mechanisms of homocysteine-induced atherothrombosis. Journal of Thrombosis and Haemostasis 3: 1646-1654. doi:10.1111/j.1538-7836.2005.01364.x. PubMed: 16102030.
- 2. Jakubowski H (2004) Molecular basis of homocysteine toxicity in humans. Cell Mol Life Sci 61: 470-487. doi:10.1007/s00018-003-3204-7. PubMed: 14999406.
- 3. Zou C-G, Banerjee R (2005) Homocysteine and redox signaling. Antioxid Redox Signal 7: 547-559. PubMed: 15890000.
- 4. Sachdev P (2004) Homocysteine, cerebrovascular disease and brain atrophy. J Neurol Sci 226: 25-29. doi:10.1016/j.jns.2004.09.006. PubMed: 15537514.
- 5. Van den Berg M, Van der Knaap MS, Boers GH, Stehouwer CD, Rauwerda JA et al. (1995) Hyperhomocysteinaemia; with reference to its neuroradiological aspects. Neuroradiology 37: 403-411. doi:10.1007/s002340050119. PubMed: 7477843.
- 6. Hajjar KA, Mauri L, Jacovina AT, Zhong F, Mirza UA et al. (1998) Tissue plasminogen activator binding to the annexin II tail domain direct modulation by homocysteine. J Biol Chem 273: 9987-9993. doi:10.1074/jbc.273.16.9987. PubMed: 9545344.
- 7. Lentz SR, Sadler JE (1991) Inhibition of thrombomodulin surface expression and protein C activation by the thrombogenic agent homocysteine. J Clin Invest 88: 1906–1914. doi:10.1172/JCI115514. PubMed: 1661291.
- 8. Outinen PA, Sood SK, Pfeifer SI, Pamidi S, Podor TJ et al. (1999) Homocysteine-induced endoplasmic reticulum stress and growth arrest leads to specific changes in gene expression in human vascular endothelial cells. Blood 94: 959-967. PubMed: 10419887.
- 9. Perła-Kaján J, Twardowski T, Jakubowski H (2007) Mechanisms of homocysteine toxicity in humans. Amino Acids 32: 561-572. doi:10.1007/s00726-006-0432-9. PubMed: 17285228.
- 10. Ross R (1993) The pathogenesis of atherosclerosis: a perspective for the 1990s. PubMed: 8479518.
- 11. Seshadri S, Beiser A, Selhub J, Jacques PF, Rosenberg IH et al. (2002) Plasma homocysteine as a risk factor for dementia and Alzheimer's disease. N Engl J Med 346: 476-483. doi:10.1056/NEJMoa011613. PubMed: 11844848.
- 12. Smith AD, Smith SM, de Jager CA, Whitbread P, Johnston C et al. (2010) Homocysteine-lowering by B vitamins slows the rate of accelerated brain atrophy in mild cognitive impairment: a randomized controlled trial. PLOS ONE 5: e12244. doi:10.1371/journal.pone.0012244. PubMed: 20838622.
- 13. Clarke R, Smith AD, Jobst KA, Refsum H, Sutton L et al. (1998) Folate, vitamin B12, and serum total homocysteine levels in confirmed in Alzheimer disease. Archives of Neurology 55: 1449.
- 14. Miller JW (1999) Homocysteine and Alzheimer's disease. Nutr Rev 57: 126–129. PubMed: 10228350.
- 15. Kruman II, Culmsee C, Chan SL, Kruman Y, Guo Z et al. (2000) Homocysteine elicits a DNA damage response in neurons that promotes apoptosis and hypersensitivity to excitotoxicity. J Neurosci 20: 6920-6926. PubMed: 10995836.
- 16. Duan W, Ladenheim B, Cutler RG, Kruman II, Cadet JL et al. (2002) Dietary folate deficiency and elevated homocysteine levels endanger dopaminergic neurons in models of Parkinson's disease. J Neurochem 80: 101-110. doi:10.1046/j.0022-3042.2001.00676.x. PubMed: 11796748.
- 17. Brosnan JT, Jacobs RL, Stead LM, Brosnan ME (2004) Methylation demand: a key determinant of homocysteine metabolism. ACTA BIOCHIMICA POLONICA-ENGLISH EDITION 51: 405-414.
- 18. Selhub J (1999) Homocysteine metabolism. Annu Rev Nutr 19: 217-246. doi:10.1146/annurev.nutr.19.1.217. PubMed: 10448523.
- 19. Pajares MA, Pérez-Sala D (2006) Betaine homocysteine S-methyltransferase: just a regulator of homocysteine metabolism? Cell Mol Life Sci 63: 2792-2803. doi:10.1007/s00018-006-6249-6. PubMed: 17086380.
- 20. Obeid R, Herrmann W (2006) Mechanisms of homocysteine neurotoxicity in neurodegenerative diseases with special reference to dementia. FEBS Lett 580: 2994-3005. doi:10.1016/j.febslet.2006.04.088. PubMed: 16697371.
- 21. Ubbink JB, Vermaak WJ, van der Merwe A, Becker PJ (1993) Vitamin B-12, vitamin B-6, and folate nutritional status in men with hyperhomocysteinemia. Am J Clin Nutr 57: 47-53. PubMed: 8416664.
- 22. Carey HV, Andrews MT, Martin SL (2003) Mammalian hibernation: cellular and molecular responses to depressed metabolism and low temperature. Physiol Rev 83: 1153-1181. PubMed: 14506303.
- 23. Geiser F (2013) Hibernation. Curr Biol 23: R188-R193. doi:10.1016/j.cub.2013.01.062. PubMed: 23473557.
- 24. Morin P Jr, Storey KB (2009) Mammalian hibernation: differential gene expression and novel application of epigenetic controls. Int J Dev Biol 53: 433–442. doi:10.1387/ijdb.082643pm. PubMed: 19412897.
- 25. Neuweiler G (2000) Biology of bats. Oxford University Press, USA.
- 26. Schwartz C, Hampton M, Andrews MT (2013) Seasonal and Regional Differences in Gene Expression in the Brain of a Hibernating Mammal. PLOS ONE 8: e58427. doi:10.1371/journal.pone.0058427.
- 27. Epperson LE, Rose JC, Russell RL, Nikrad MP, Carey HV et al. (2010) Seasonal protein changes support rapid energy production in hibernator brainstem. J Comp Physiol B 180: 599-617. doi:10.1007/s00360-009-0422-9. PubMed: 19967378.
- 28. Kondo N, Sekijima T, Kondo J, Takamatsu N, Tohya K et al. (2006) Circannual control of hibernation by HP complex in the brain. Cell 125: 161-172. doi:10.1016/j.cell.2006.03.017. PubMed: 16615897.
- 29. Ma YL, Zhu X, Rivera PM, Tøien Ø, Barnes BM et al. (2005) Absence of cellular stress in brain after hypoxia induced by arousal from hibernation in Arctic ground squirrels. Am J Physiol Regul Integr Comp Physiol 289: R1297-R1306. doi:10.1152/ajpregu.00260.2005. PubMed: 15976308.
- 30. Härtig W, Stieler J, Boerema AS, Wolf J, Schmidt U et al. (2007) Hibernation model of tau phosphorylation in hamsters: selective vulnerability of cholinergic basal forebrain neurons–implications for Alzheimer's disease. Eur J Neurosci 25: 69-80. doi:10.1111/j.1460-9568.2006.05250.x. PubMed: 17241268.
- 31. Zhou F, Zhu X, Castellani RJ, Stimmelmayr R, Perry G et al. (2001) Hibernation, a model of neuroprotection. Am J Pathol 158: 2145-2151. doi:10.1016/S0002-9440(10)64686-X. PubMed: 11395392.
- 32. Dave KR, Christian SL, Perez-Pinzon MA, Drew KL (2012) Neuroprotection: Lessons from hibernators. Comp Biochem Physiol B Biochem Mol Biol 162: 1-9. doi:10.1016/j.cbpb.2012.01.008. PubMed: 22326449.
- 33. Pan Y-H, Zhang Y, Cui J, Liu Y, McAllan BM et al. (2013) Adaptation of Phenylalanine and Tyrosine Catabolic Pathway to Hibernation in Bats. PLOS ONE 8: e62039. doi:10.1371/journal.pone.0062039. PubMed: 23620802.
- 34. Finkelstein JD, Mudd SH (1967) Trans-sulfuration in Mammals The Methionine-Sparing Effect Of Cystine. J Biol Chem 242: 873-880. PubMed: 6067003.
- 35. Garrow TA (1996) Purification, kinetic properties, and cDNA cloning of mammalian betaine-homocysteine methyltransferase. J Biol Chem 271: 22831-22838. PubMed: 8798461.
- 36. Wang J, Dudman NP, Lynch J, Wilcken DE (1991) Betaine: homocysteine methyltransferase—a new assay for the liver enzyme and its absence from human skin fibroblasts and peripheral blood lymphocytes. Clinica Chimica Acta 204: 239-249. doi:10.1016/0009-8981(91)90235-5.
- 37. Paxinos G, Franklin KB (2004) The mouse brain in stereotaxic coordinates. Gulf Professional Publishing.
- 38. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25: 4876-4882. doi:10.1093/nar/25.24.4876. PubMed: 9396791.
- 39. Pride DT, Wassenaar TM, Ghose C, Blaser MJ (2006) Evidence of host-virus co-evolution in tetranucleotide usage patterns of bacteriophages and eukaryotic viruses. BMC Genomics 7: 8. doi:10.1186/1471-2164-7-8. PubMed: 16417644.
- 40. Nei M, Gojobori T (1986) Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol Biol Evol 3: 418-426. PubMed: 3444411.
- 41. Di Tommaso P, Moretti S, Xenarios I, Orobitg M, Montanyola A et al. (2011) T-Coffee: a web server for the multiple sequence alignment of protein and RNA sequences using structural information and homology extension. Nucleic Acids Res 39: W13-W17. doi:10.1093/nar/gkr245. PubMed: 21558174.
- 42. Kiefer F, Arnold K, Künzli M, Bordoli L, Schwede T (2009) The SWISS-MODEL Repository and associated resources. Nucleic Acids Res 37: D387-D392. doi:10.1093/nar/gkn750. PubMed: 18931379.
- 43. Finkelstein JD (1998) The metabolism of homocysteine: pathways and regulation. Eur J Pediatr 157: S40-S44. doi:10.1007/PL00014300. PubMed: 9587024.
- 44. González B, Pajares MA, Martı ́nez-Ripoll M, Blundell TL, Sanz-Aparicio J (2004) Crystal Structure of Rat Liver Betaine Homocysteine S-Methyltransferase Reveals New Oligomerization Features and Conformational Changes Upon Substrate Binding. J Mol Biol 338: 771-782. doi:10.1016/j.jmb.2004.03.005. PubMed: 15099744.
- 45. Su B, Wang X, Drew KL, Perry G, Smith MA et al. (2008) Physiological regulation of tau phosphorylation during hibernation. J Neurochem 105: 2098-2108. doi:10.1111/j.1471-4159.2008.05294.x. PubMed: 18284615.
- 46. Ji C, Shinohara M, Kuhlenkamp J, Chan C, Kaplowitz N (2007) Mechanisms of protection by the betaine homocysteine methyltransferase/betaine system in HepG2 cells and primary mouse hepatocytes. Hepatology 46: 1586-1596. doi:10.1002/hep.21854. PubMed: 17705221.
- 47. Teng Y-W, Mehedint MG, Garrow TA, Zeisel SH (2011) Deletion of betaine-homocysteine S-methyltransferase in mice perturbs choline and 1-carbon metabolism, resulting in fatty liver and hepatocellular carcinomas. J Biol Chem 286: 36258-36267. doi:10.1074/jbc.M111.265348. PubMed: 21878621.
- 48. Ratnam S, Wijekoon EP, Hall B, Garrow TA, Brosnan ME et al. (2006) Effects of diabetes and insulin on betaine-homocysteine S-methyltransferase expression in rat liver. Am J Physiol Endocrinol Metab 290: E933-E939. doi:10.1152/ajpendo.00498.2005. PubMed: 16352668.
- 49. Williams KT, Schalinske KL (2007) New insights into the regulation of methyl group and homocysteine metabolism. J Nutr 137: 311-314. PubMed: 17237303.
- 50. Craig SAS (2004) Betaine in human nutrition. Am J Clin Nutr 80: 539-549. PubMed: 15321791.
- 51. Melichar I, Brozek G, Janský L, Vyskocil F (1973) Effect of hibernation and noradrenaline on acetylcholine release and action at neuromuscular junction of the golden hamster (Mesocricetus auratus). Pflügers Archiv 345: 107-122. doi:10.1007/BF00585834. PubMed: 4797954.
- 52. Lonn E, Yusuf S, Arnold MJ, Sheridan P, Pogue J et al. (2006) Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med 354: 1567–1577. doi:10.1056/NEJMoa060900. PubMed: 16531613.
- 53. Cook S, Hess O (2005) Homocysteine and B vitamins. Atherosclerosis: Diet and Drugs. Springer. pp. 325-338.
- 54. van Tonder SV, Metz J, Green R (1975) Vitamin B12 metabolism in the fruit bat (Rousettus aegyptiacus). The induction of vitamin B12 deficiency and its effect on folate levels. Br J Nutr 34: 397-410. PubMed: 1201264.
- 55. Storey KB (2010) Out cold: biochemical regulation of mammalian hibernation–a mini-review. Gerontology 56: 220-230. PubMed: 19602865.
- 56. Medina MA, Urdiales JL, Amores-Sánchez MI (2001) Roles of homocysteine in cell metabolism. European Journal of Biochemistry 268: 3871-3882. doi:10.1046/j.1432-1327.2001.02278.x. PubMed: 11453979.
- 57. Brattström L, Israelsson B, Norrving B, Bergqvist D, Thörne J et al. (1990) Impaired homocysteine metabolism in early-onset cerebral and peripheral occlusive arterial disease effects of pyridoxine and folic acid treatment. Atherosclerosis 81: 51-60. doi:10.1016/0021-9150(90)90058-Q. PubMed: 2407253.
- 58. Verhoef P, Stampfer MJ, Buring JF, Gaziano JM, Allen RH et al. (1996) Homocysteine metabolism and risk of myocardial infarction: relation with vitamins B6, B12, and folate. Am J Epidemiol 143: 845-859. doi:10.1093/oxfordjournals.aje.a008828. PubMed: 8610698.
- 59. Herbert V, Zalusky R (1962) Interrelations of vitamin B12 and folic acid metabolism: folic acid clearance studies. J Clin Invest 41: 1263–1276. doi:10.1172/JCI104589. PubMed: 13906634.
- 60. Shane B, Stokstad ER (1985) Vitamin B12-folate interrelationships. Annu Rev Nutr 5: 115-141. doi:10.1146/annurev.nu.05.070185.000555. PubMed: 3927946.
- 61. Skovby F, Kraus JP, Rosenberg LE (1984) Biosynthesis and proteolytic activation of cystathionine beta-synthase in rat liver. J Biol Chem 259: 588-593. PubMed: 6706953.
- 62. Agrawal N, Banerjee R (2008) Human polycomb 2 protein is a SUMO E3 ligase and alleviates substrate-induced inhibition of cystathionine β-synthase sumoylation. PLOS ONE 3: e4032. doi:10.1371/journal.pone.0004032. PubMed: 19107218.
- 63. Gadalla MM, Snyder SH (2010) Hydrogen sulfide as a gasotransmitter. J Neurochem 113: 14-26. doi:10.1111/j.1471-4159.2010.06580.x. PubMed: 20067586.
- 64. Kery V, Poneleit L, Kraus JP (1998) Trypsin cleavage of human cystathionine β-synthase into an evolutionarily conserved active core: structural and functional consequences. Arch Biochem Biophys 355: 222-232. doi:10.1006/abbi.1998.0723. PubMed: 9675031.
- 65. Esiri MM, Pearson RC, Steele JE, Bowen DM, Powell TP (1990) A quantitative study of the neurofibrillary tangles and the choline acetyltransferase activity in the cerebral cortex and the amygdala in Alzheimer's disease. J Neurol Neurosurg Psychiatry 53: 161-165. doi:10.1136/jnnp.53.2.161. PubMed: 2313304.
- 66. Evans JC, Huddler DP, Jiracek J, Castro C, Millian NS et al. (2002) Betaine-homocysteine methyltransferase: zinc in a distorted barrel. Structure 10: 1159-1171. doi:10.1016/S0969-2126(02)00796-7. PubMed: 12220488.