Neurotoxicity Research

, Volume 20, Issue 2, pp 150–158

Flavin-Containing Monooxygenase mRNA Levels are Up-Regulated in ALS Brain Areas in SOD1-Mutant Mice

Authors

    • Lab of Experimental NeurobiologyIRCCS National Neurological Institute “C. Mondino”
    • Department of Neurological SciencesUniversity of Pavia
  • Paolo Ogliari
    • Lab of Experimental NeurobiologyIRCCS National Neurological Institute “C. Mondino”
  • Annalisa Davin
    • Lab of Experimental NeurobiologyIRCCS National Neurological Institute “C. Mondino”
  • Manuel Corato
    • Lab of Experimental NeurobiologyIRCCS National Neurological Institute “C. Mondino”
    • Department of Neurological SciencesUniversity of Pavia
  • Emanuela Cova
    • Lab of Experimental NeurobiologyIRCCS National Neurological Institute “C. Mondino”
  • Kenneth Abel
    • Human BioMolecular Research Institute
  • John R. Cashman
    • Human BioMolecular Research Institute
  • Mauro Ceroni
    • Lab of Experimental NeurobiologyIRCCS National Neurological Institute “C. Mondino”
  • Cristina Cereda
    • Lab of Experimental NeurobiologyIRCCS National Neurological Institute “C. Mondino”
Article

DOI: 10.1007/s12640-010-9230-y

Cite this article as:
Gagliardi, S., Ogliari, P., Davin, A. et al. Neurotox Res (2011) 20: 150. doi:10.1007/s12640-010-9230-y

Abstract

Flavin-containing monooxygenases (FMOs) are a family of microsomal enzymes involved in the oxygenation of a variety of nucleophilic heteroatom-containing xenobiotics. Recent results have pointed to a relation between Amyotrophic Lateral Sclerosis (ALS) and FMO genes. ALS is an adult-onset, progressive, and fatal neurodegenerative disease. We have compared FMO mRNA expression in the control mouse strain C57BL/6J and in a SOD1-mutated (G93A) ALS mouse model. Fmo expression was examined in total brain, and in subregions including cerebellum, cerebral hemisphere, brainstem, and spinal cord of control and SOD1-mutated mice. We have also considered expression in male and female mice because FMO regulation is gender-related. Real-Time TaqMan PCR was used for FMO expression analysis. Normalization was done using hypoxanthine–guanine phosphoribosyl transferase (Hprt) as a control housekeeping gene. Fmo genes, except Fmo3, were detectably expressed in the central nervous system of both control and ALS model mice. FMO expression was generally greater in the ALS mouse model than in control mice, with the highest increase in Fmo1 expression in spinal cord and brainstem. In addition, we showed greater Fmo expression in males than in female mice in the ALS model. The expression of Fmo1 mRNA correlated with Sod1 mRNA expression in pathologic brain areas. We hypothesize that alteration of FMO gene expression is a consequence of the pathological environment linked to oxidative stress related to mutated SOD1.

Keywords

FMOAmyotrophic lateral sclerosisqPCRSOD1Exposure to toxins

Introduction

Amyotrophic lateral sclerosis (ALS) is an adult-onset, progressive, and fatal neurodegenerative disease with unknown pathogenesis due to a selective loss of motor neurons in the brainstem and spinal cord. The majority (i.e., 90%) of cases presents sporadic onset disease (i.e., SALS), while 10% are described as familial onset (i.e., FALS). Twenty percent of FALS are linked to mutations in the gene encoding Cu–Zn superoxide dismutase (SOD1), an enzyme converting superoxide anion into hydrogen peroxide (Rosen et al. 1993). Familial and sporadic forms of ALS present an overlapping clinical picture and disease course suggesting a common pathogenesis that involves the same cellular pathways and mechanisms such as oxidative stress and protein aggregation (Cleveland and Rothstein 2001; Kruman et al. 1999).

It has been suggested that the onset of the sporadic ALS (SALS) may be related to exposure to toxic environmental factors (Steele and McGeer 2008). In 1954, epidemiological studies showed that ALS was 100-fold more prevalent in the Chamorro indigenous people of Guam Island (Mulder et al. 1954). Guam ALS is clinically and pathologically identical to classic ALS, and different environmental factors were hypothesized to be involved. The most frequently cited hypothesis of the environmental cause of Guam syndrome has been the exposure to toxins from the Cycas micronesica plant (Steele and McGeer 2008).

Environmental hypothesis has been supported also by an association between the risk of developing SALS and the exposure to heavy metals, industrial solvents, and pesticides, especially organophosphates (Johnson and Atchison 2009; Weisskopf and Ascherio 2009). In particular, studies have suggested associations between occupational exposure to organophosphates related to playing professional soccer and farming, and the risk of developing ALS (Li and Sung 2003; Abel 2007).

Variations of genetic background involving the single nucleotide polymorphisms (SNPs) of genes related to detoxication pathways may also have a role in SALS. Data from different populations have shown associations between some SNPs belonging to the paraoxonase genes (PON 1, 2, 3) and SALS (Slowik et al. 2006; Mahoney et al. 2006; Saeed et al. 2006; Ricci et al. 2010). PON genes encode high-density lipoprotein-associated enzymes that play a role in the detoxication of a large number of organophosphorus compounds (Costa et al. 2003). Other reports have suggested a role of Flavin-containing monooxygenase (FMO) genes in ALS, underling a correlation between polymorphisms located in the 3′untranslated region of the FMO1 gene and SALS (Malaspina et al. 2001; Cereda et al. 2006).

FMOs constitute a family of enzymes located in the endoplasmic reticulum (ER) that catalyze the oxygenation of a variety of endogenous and exogenous nucleophilic compounds including organophosphates and pesticides (Hines et al. 1994; Elfarra 1995; Cashman 1995) in detoxication processes (Venkatesh et al. 1992; 1991; Siddens et al. 2008; Leoni et al. 2008).

FMO enzymes are primarily expressed in tissues that carry out detoxication processes (e.g., liver, kidney, and lung) and they are also present in mammalian brain (Zhang and Cashman 2006; Janmohamed et al. 2004), although their role in the nervous system is not clear. Nine FMO genes located in two clusters on chromosome 1 are present in mouse. The first cluster contains five genes named Fmo1, 2, 3, 4, and 6; the second comprises the Fmo9, 12, and 13 genes. The Fmo5 gene is located on chromosome 3 (Cashman 1995).

The most highly expressed Fmo genes in mouse brain are reported to be Fmo5 and Fmo1, while Fmo2 and Fmo4 are transcribed at relatively lower levels and Fmo3 mRNA was not detectable (Janmohamed et al. 2004). Experimental studies with ALS model mice carrying mutated SOD1 have shown modified cellular metabolism and development of multiple pathogenic cellular processes similar to sporadic human ALS, including oxidative stress and protein aggregation (Shaw 2005) especially in the brainstem and spinal cord (Mahoney et al. 2006; Garbuzova-Davis et al. 2007; Watanabe et al. 2001). Moreover, Liu et al. (1998) showed that ALS model mice transgenic for mutant SOD1-G93A showed increased levels of free radical-derived products from hydrogen peroxide in spinal cord. Instead, in brain, oxygen radical content in both SOD1-G93A and non-transfected mice did not show statistically significant differences, suggesting that the ROS increase in spinal cord was specific. Increased oxidative stress is well documented in brain and spinal cord of sporadic and familial ALS patients and seems to be both a hallmark of the pathology and the SOD1 mRNA increase (Liu 1996; Gagliardi et al. 2010).

To examine a potential role for Fmo in ALS, we characterized Fmo gene expression levels in different brain regions and in spinal cord in both control C57BL/6J mice and in mutated SOD1-G93A ALS model mice. We also compared Fmo brain expression in female and male mice because evidence exists that regulation of human and mice Fmo gene expression is gender-dependent (Falls et al. 1997; Coecke et al. 1998). Moreover, we also searched for a correlation between FMO and SOD1 expression to understand better the pathogenetic mechanisms related to the environmental factors and gender.

Materials and Methods

Animals

A group of 12 SOD1-G93A transgenic mice (6 males and 6 females C57BL/6J mice carrying the human SOD1-G93A mutation) and a control group of 20 C57BL/6J mice (8 males and 12 females) were used. Transgenic SOD1-G93A mice originally obtained from Jackson Laboratories (Bar Harbor, Maine, USA) and carrying about 20 copies of the mutant human hSOD1 gene with a Gly93Ala substitution (B6SJL-TgNSOD-1-SOD1G93A-1Gur) were bred and maintained on a C57BL/6 genetic background at Harlan Italy S.R.L., Bresso (MI), Italy. Transgenic mice were identified by PCR and were killed at the end stage of the disease characterized by the complete paralysis of the hind limbs and the difficulty to right through their own effort within 30 s after being placed on both sides (approximately 21 weeks for males and 23 weeks for females). All mice were housed at a temperature of 21 ± 1°C with relative humidity 55 ± 10% and 12 h of light. Food (standard pellets) and water were supplied ad libitum. Procedures involving animals and their care were conducted in conformity with institutional guidelines that are in compliance with national (D.L. No. 116, G.U. Suppl 40, Feb 18, 1992, Circolare No. 8, G.U., 14 luglio 1994) and international laws and policies (EEC Council Directive 86/609, OJL 358, 1 Dec 12, 1987; NIH Guide for the Care and use of Laboratory Animals, U.S. National Research Council, 1996). Brains were isolated by surgical ablation and were split into right and left halves. Also spinal cords were isolated. The right brain sections were used to study the expression of FMO mRNAs in total brain, and the left sections were utilized to obtain distinct brain areas (cerebellum, cerebral hemisphere, and brainstem). All samples were stored in 1 ml of Trizol® (Sigma-Aldrich, Milan, Italy) at −80°C.

RNA Isolation

Samples were homogenized, and total RNA was isolated by Trizol® reagent (Sigma-Aldrich, Milan, Italy) following the manufacturer’s specifications and quantified by spectrophotometric analysis (Nanodrop®, Celbio, Milan, Italy).

Reverse Transcription

RNA (11 μl at 550 ng/μl) was reverse-transcribed using High Capacity cDNA Archive Kit (Applera, Monza, Italy) according to the manufacturer’s recommendations. The cDNA samples were stored at −20°C.

Validation of q-PCR Conditions

We designed, optimized, and validated primer and probe systems with a quantitative real-time PCR (q-PCR) technique, using RNA from mouse liver because FMO expression in this tissue was well characterized. The same protocols were used to study the FMO mRNA expression in brain. In addition, we checked specificity of the primers by melting curve analysis to identify the presence of possible primer dimers or non-specific products. An assay for the housekeeping Hypoxanthine–guanine phosphoribosyl transferase (Hprt) q-PCR technique gene was chosen as a control because of its relatively low mRNA levels in brain that are similar to those of FMO genes observed in other species (Janmohamed et al. 2004; Shaw 2005). Hprt expression was invariable in all brain areas and spinal cord in G93A transgenic and no transgenic control female and male mice (data not shown).

Concerning SOD1 mRNA analysis, we evaluated mRNA levels using Sybr Green Real-Time PCR. Sybr-Green primers were designed with identical annealing and melting temperatures using Primer Express Software (Applera, Monza, Italy). The primers were designed spanning introns to avoid amplification of genomic DNA. We checked specificity of the primers by melting curve analysis.

Real-Time PCR Primers and Probes

FMO quantitative experiments were done using Taq-Man Real-Time PCR, while SOD1 mRNA level evaluation was tested by Sybr Green Real-Time PCR. Taq-Man primers and probes (Eurofins MWG Operon, Ebersberg, Germany) were designed with identical annealing and melting temperatures using Primer Express Software (Applera, Monza, Italy). Primers, probes, sequences, and amplification efficiencies are listed in Table 1. Real-Time PCR was done on an iCycler PCR Detection System (iQ5 vers.2.0, Bio-Rad, Segrate, Italy) using iQ Supermix (Bio-Rad, Segrate, Italy). Each sample was tested with PCR using three replicates with 5 μl of cDNA at a concentration of 300 ng/μl, forward and reverse primers at a final concentration of 500 nM, and Taq-Man probe at a final concentration 200 nM, 12.5 μl of iQ buffer reaction (100 mM KCl, 40 mM Tris–HCl, pH 8.4, 1.6 mM dNTPs, 50 U/ml Taq DNA polymerase, 6 mM MgCl2) in 15 μl total volume. The PCR protocol started with denaturation at 95°C for 5 min followed by 40 cycles (95°C × 15″ and 59°C × 1′). Concerning SOD1, oligonucleotides were used at a concentration of 150 nM in a total volume reaction of 15 μl. Control reactions included a template from the cDNA synthesis reaction, ± the enzyme reverse transcriptase (RT), as well as the no-template controls (NTC). The qPCR reaction was 95°C for 5 min, 95°C for 15 s, and 55°C for 30 s for 40 cycles, with a melting curve starting at 55°C and increasing 0.5°C for each 30 s. All the experiments were done twice with three replicates. The average of the Ct values from the three replicates for each sample was reported, and a standard deviation equal to or less than 0.15 was considered a valid result. The relative mRNA levels were displayed as rq values normalized to Hprt expression using the 2−ΔΔCt comparative method related with an internal standard (Livak and Schmittgen 2001).
Table 1

q-PCR primers and probes

Primer

Gene

Sequence

Amplification efficiency (90–110%)

Primer forward

Hprt1 F

5′-TCC CAG CGT CGT GAT TAG C-3′

 

Primer reverse

Hprt1 R

5′-CGG CAT AAT GAT TAG GTA TAC AAA ACA-3′

99.7%

Probe

Hprt1

5′-6-FAM-TGA TGA ACC AGG TTA TGA CC-3′

 

Primer forward

Fmo1 F

5′-CGT GGA GGC CAG CCA CT-3′

 

Primer reverse

Fmo1 R

5′-CAC CCA TGC CCC TCC AG-3′

97.5%

Probe

Fmo1

5′-6-FAM-CAA AAA AGG TGT TCC TCA GC-3′

 

Primer forward

Fmo2 F

5′-CAC ATC CAG CCT CAC CTG C-3′

 

Primer reverse

Fmo2 R

5′-GGC CTC GGA ACC TCT CAA TAC-3′

98.1%

Probe

Fmo2

5′-6-FAM-ACT CAA GTC ATT CCC-3′

 

Primer forward

Fmo3 F

5′-TGA TTT GTT CTG GGC ATC ACA-3′

 

Primer reverse

Fmo3 R

5′-AAC GGT TCA GTC CTG GAA AGG-3′

101.6%

Probe

Fmo3

5′-6-FAM-CCA TGT ACC AAA AGA C-3′

 

Primer forward

Fmo4 F

5′-TGG TTT GCA CTG GGC AAT T-3′

 

Primer reverse

Fmo4 R

5′-TGG ATT CCA GGA AAG GAC TCC-3′

99.4%

Probe

Fmo4

5′-6-FAM - TGA GCC CAC ATT TAC CTC-3′

 

Primer forward

Fmo5 F

5′-AGA CTA CAG TGT GCA GCG TGA AG-3′

 

Primer reverse

Fmo5 R

5′-GCC ATT GGC CCG AGG TA-3′

100.9%

Probe

FMO5

5′-6-FAM-AGC AGC CTG ATT TC-3′

 

Primer forward

Sod1 F

5′-ACT TCG AGC AGA AGG CAA GC-3′

 

Primer reverse

Sod1 R

5′-TTA GAG TGA GGA TTA AAA TGA GGT C-3′

99.7%

PCR primer and probe sequences for the real-time q-PCR analysis of Hprt housekeeping gene, Fmo and Sod1 genes are listed, with the corresponding assay amplification efficiencies (90–110%)

Statistical Analysis

Kruskal–Wallis and Mann–Whitney tests were used to compare expression of FMO genes in total brain and in different brain sub regions. The Kruskal–Wallis test was used to compare different FMO genes, and the Mann–Whitney test was used to compare the expression differences between genders. Statistical analyses were done with GraphPad Prism version 3.0 (GraphPad Software Inc, San Diego, California). A P-value of 0.05 or lower was considered statistically significant.

Results

Fmo mRNA Levels in SOD1-G93A and Control Mouse Total Brain

Fmo gene-specific expression was defined in total brain in ALS model and control mice and also compared in distinct central nervous system (CNS) areas (cerebellum, cerebral hemisphere, brainstem, and spinal cord). All Fmo genes were detectably expressed in the murine CNS with the exception of Fmo3. In total brain, Fmo mRNAs showed generally higher expression in the ALS model mice than in controls. A significant increase in Fmo4 expression was observed between ALS and control mice (P < 0.01) (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs12640-010-9230-y/MediaObjects/12640_2010_9230_Fig1_HTML.gif
Fig. 1

Fmo mRNA levels in C57BL/6J and SOD1-G93A mouse total brain. The mRNA levels (combined males and females) were normalized to Hprt expression using the 2−ΔΔCt comparative method related with a calibrator (Livak and Schmittgen 2001). Comparing total brain RNAs from C57BL/6J (black bar) and SOD1-G93A (gray bar) mice showed a higher expression of Fmo in the ALS model. A statistically significant difference was observed for Fmo4 (**, P < 0.01 SOD1-G93A versus C57BL/6J mice)

Fmo mRNA Levels in Specific CNS Areas in SOD1-G93A and Control Mice

ALS model mice showed a different profile of Fmo gene expression in the CNS areas examined when compared with control mice (Fig. 2, Table 2). The most significant difference was seen for Fmo1 expression in spinal cord, where Fmo1 mRNA was detected at levels 16-fold greater in SOD1-G93A than in control mice (P < 0.001) (Fig. 2). As in spinal cord, Fmo1 showed relatively higher expression in brainstem of SOD1-G93A mice (P < 0.05). No significant differences were detected in Fmo25 gene expression in these areas, although the Fmo2, 4, 5 genes were up-regulated in spinal cord (P < 0.05) (Fig. 2). In cerebellum, we observed 3-fold greater relative expression of Fmo4 (P < 0.001), and 2-fold greater Fmo2 and Fmo5 expression (P < 0.01 and P < 0.05, respectively) in the ALS model than in C57BL/6J mice. In the cerebral hemisphere, Fmo2, Fmo4, and Fmo5 were also expressed at significantly greater levels in SOD1-G93A compared to C57BL/6J mice (P < 0.05) (Fig. 2, Table 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs12640-010-9230-y/MediaObjects/12640_2010_9230_Fig2_HTML.gif
Fig. 2

Fmo mRNA levels in specific mouse nervous regions. In SOD1-G93A mouse spinal cord, Fmo1 expression was observed to be 16 times higher than in C57BL/6J mice (***, P < 0.001) and 3 times in brain stem (*, P < 0.05). Fmo2 and Fmo4 levels were significantly different in all areas (cerebellum:Fmo2 **, P < 0.01; Fmo4 ***, P < 0.001; cerebral hemisphere and spinal cord: Fmo2 and Fmo4 *, P < 0.05). Fmo5 showed statistically significant differences between controls and ALS mice in all tested regions (*, P < 0.05)

Table 2

Fold-expression differences in specific CNS areas in SOD1-G93A and C57BL/6J mice

 

Fmo1

Fmo2

Fmo4

Fmo5

Spinal cord

+16

+1

+0.5

Brain stem

+3

Cerebellum

+2

+3

+2

Cerebral hemisphere

+1

+1

+1

Fold differences are presented as related increases (positive values in the table, SOD1-G93A versus C57BL/6J mice) or decreases in ALS (negative values, SOD1-G93A versus C57BL/6J mice)

Fmo Gender-Specific Expression in SOD1-G93A and Control Mice

Fmo1 expression showed the greatest gender-specific differences in ALS and control mice. In male ALS model mice, Fmo1 mRNA was detected at the highest level in spinal cord (P < 0.01), at levels 4-fold greater compared to C57BL/6J brainstem (P < 0.05) (Fig. 3a). In contrast, Fmo1 expression was significantly decreased in spinal cord of female SOD1-G93A compared to C57BL/6J female mice (P < 0.05) (Fig. 3b).
https://static-content.springer.com/image/art%3A10.1007%2Fs12640-010-9230-y/MediaObjects/12640_2010_9230_Fig3_HTML.gif
Fig. 3

Comparison of brain Fmo genes expression between male and female mice. In SOD1-G93A male mice (a), Fmo1 expression was observed to be 3-fold greater than in C57BL/6J male mice (**, P < 0.01), while it was lower in SOD1-G93A females compared to normal female mice (*, P < 0.05) (b). Fmo2 was more highly expressed in cerebral hemisphere of C57BL/6J male mice than in SOD1-G93A mice (**, P < 0.01) (c). An opposite trend we observed in females, where SOD1-G93A mice showed higher Fmo2 expression in cerebral hemisphere (*, P < 0.05) and cerebellum (**, P < 0.01) than in control females (d). Although Fmo4 and Fmo5 mRNA levels did not show statistically significant differences between male mice (e, g). In females, they were more highly expressed in all areas in SOD1-G93A than in C57BL/6J mice but we have statistically differences only in cerebellum Fmo4 gene expression (*, P < 0.05) (f, h)

Fmo2 showed significantly higher expression in male spinal cord of C57BL/6J than in SOD1-G93A mice (P < 0.01) (Fig. 3c). In female SOD1-G93A mice, Fmo2 was more highly expressed in cerebellum and cerebral hemisphere than in female C57BL/6J mice (P < 0.01, P < 0.05, respectively) (Fig. 3d). In male mice, the data of Fmo4 expression showed no statistically significant differences between control and ALS mice (Fig. 3e), while in female SOD1-G93A mice Fmo4 expression showed the greatest difference in cerebellum compared with control mice (P < 0.05) (Fig. 3f). Fmo5 did not show statistically significant differences in mRNA levels between control and ALS mice in both males and females (Fig. 3g, h).

Sod1 mRNA Expression in Pathological Areas and Gender in G93A Mice

Sod1 mRNA was found to be increased in pathological areas such as spinal cord and brainstem compared to non-pathological regions such as cerebellum and cerebral hemisphere.

ALS model mice showed a different profile of Sod1 gene expression between females and males in the CNS areas examined when compared with control mice (Fig. 4). In all nervous areas examined, the Sod1 mRNA level was greater in males than in female mice. In particular, Sod1 showed increased expression in males compared to female mice although the difference was not statistically significant.
https://static-content.springer.com/image/art%3A10.1007%2Fs12640-010-9230-y/MediaObjects/12640_2010_9230_Fig4_HTML.gif
Fig. 4

Comparison of brain Sod1 gene expression between SOD1-G93A male and female mice. Sod1 expression was observed to be greater in SOD1-G93A male mice than in SOD1-G93A female mice in cerebellum, cerebral hemisphere, spinal cord, and brain stem. The mRNA levels were normalized to Hprt expression using the 2−ΔΔCt comparative method related with a calibrator (Livak and Schmittgen 2001)

Discussion

In this study, we examined for the first time the expression of FMO genes in specific subregions of the CNS in ALS model and control mice. We also compared FMO mRNA levels between male and female mice of the two groups.

Our data showed that FMO genes were up-regulated in ALS CNS areas studied relative to control C57BL/6J mice. The major finding was the up-regulation of the Fmo1 gene in spinal cord and brainstem, where Fmo1 mRNA levels were 16- and 3-fold greater, respectively, in the ALS model compared with control mice. The contrasting data reported by Malaspina et al. (2001) documenting down-regulation of FMO1 in human ALS spinal cord compared to healthy controls may be explained by species-specific gene regulation differences. Alterations in the CNS of FMO gene expression in ALS model mice may be a consequence of the pathological environment linked to effects of the mutant SOD1. Experimental studies with mutant SOD1 animal models revealed modified cellular metabolism and development of multiple pathogenic processes including oxidative stress, oxidation of proteins, and protein aggregation (Shaw 2005), particularly in brainstem and spinal cord (Garbuzova-Davis et al. 2007; Watanabe et al. 2001; Mahoney et al. 2006). In particular, free radical products from hydrogen peroxide were found to be increased in spinal cord of ALS model mice transfected with mutant SOD1-G93A and in autoptic brain samples of the ALS patients (Liu et al. 1998; Liu 1996). Moreover, altered FMO expression likely impacts oxidative-redox systems, as described for the yeast flavin-containing monooxygenase (yFMO) that is vital to the response of yeast to reductive stress (Suh and Robertus 2000a, b). Interestingly, Fmo1 mRNA up-regulation is strongly associated with ALS pathological regions because it is up-regulated only in spinal cord and brainstem and because it is the only Fmo gene showing increased expression in brainstem. It is possible that in ALS the activation of the FMO detoxication system may be increased by reactive oxygen species (ROS) that have been found to be present in high concentrations in spinal cord extracts prepared form SOD1-G93A mice (Liu et al. 1998).

In SALS not involving SOD1 mutations, the pathological environmental may be due to exposure to toxic factors that increase free radicals and activate the FMO system (Suh et al. 1999; Suh and Robertus 2000a, b). Studies have shown that metals including iron, copper, mercury, nickel, and lead deplete protein-bound sulfhydryl groups, resulting in ROS production as superoxide ions, hydrogen peroxide, and hydroxyl radicals. Also, it has been reported that exposure to mercury is selectively toxic to spinal cord and motor neurons in mouse and in rats, increasing ROS in nervous regions affected by the metal (Su et al. 1998; Pamphlett and Patricia 1996; LeBel et al. 1990).

The connection between genetic background and the ability of detoxication processes to respond to toxic agents has been shown previously. Polymorphisms in PON genes have been associated with SALS, although results from different populations are not conclusive (Wills et al. 2009; Ricci et al. 2010). Cereda et al. (2006) also showed a significant association between SALS disease and a genetic polymorphism within the FMO1 gene. These data highlight a likely role of at least one detoxication pathway in ALS and may help to understand the involvement of FMO enzymes in this disease. Moreover, both FMO and PON enzymes metabolize organophosphate compounds that are commonly present in farming and gardening. Studies have shown increased incidence of ALS in specific occupational categories such as professional soccer and farming (Li and Sung 2003; Abel 2007).

Finally, we studied FMO gender-regulated expression because the literature has shown the importance of hormonal regulation of both human and mouse FMO genes mediated by sex steroids (Falls et al. 1997; Coecke et al. 1998). Coecke et al. (1998) demonstrated that the human sex hormone 17β-estradiol caused a significant decrease in FMO activity both in cultured male rat hepatocytes and also in male rat liver.

In this study, we showed that Fmo1 gene expression in spinal cord and in brainstem was up-regulated in male ALS mice but down-regulated in female ALS mice. Interestingly, Fmo1 gene expression correlated with Sod1 mRNA with regard to gender and specificity of affected areas. We showed that Sod1 expression is decreased in ALS female mice relative to ALS males. In particular, both Fmo1 and Sod1 expression were down-regulated in female relative to male spinal cord.

It was previously shown that SOD1 concentration in the CSF is decreased in female relative to male ALS patients (Frutiger et al. 2008). The epidemiological data presented a different onset ratio between males and females in ALS as a function of the fecund age of the females: the ratio between males and females was 2.5 in the younger female group and 1.4 in the older female group (Manjaly et al. 2010).

In SOD1-G93A female mice, Sod1 mRNA may be down-regulated because sex hormones may have a protective function against oxidative stress damage (Behl et al. 1995). Choi et al. (2008) reported that 17β-estradiol appeared to have a protective effect in ALS. Treatment with 17β-estradiol of G93A ovariectomized female mice did not delay the onset of disease, but did retard the progression of ALS motor dysfunctions. Recently, it was shown that SOD1 interacts with estrogen receptor α (ER-α) and this complex enhances the binding of ER-α to ERE (estrogen response element)-containing DNA, representing a regulation point of the various pathways (Rao et al. 2008). ER-α and SOD1 were found to be associated with regions of the progesterone receptor gene, involved in conferring estrogen-responsive gene expression. We suggest that in female ALS mice, sex hormones may play a protective role to maintain down-regulation of the Sod1 and Fmo system, while in ALS male mice Sod1 and the Fmo system were up-regulated and their activation may be necessary in response to the ROS increase. The potential involvement of sex hormones in ALS pathology is particularly interesting in light of previous genetic studies showing a significant association between FMO1 3′UTR SNP frequencies in female SALS patients (Cereda et al. 2006). The different genetic arrangements, as differential interaction with FMO1 gene promoter regions by sex hormone-regulated transcription factors gene in males and females (Cereda et al. 2006), may influence FMO1 mRNA levels observed to be gender-dependent in ALS model.

Overall, our findings of altered Fmo gene expression within specific ALS brain regions in the SOD1-G93A mouse model provide additional support for a role for FMO enzymes in cellular response to ALS neurodegeneration. Although specific functions have not been defined for mammalian FMO isozymes in the CNS, the known functions of certain FMO enzymes in humans and other species suggest a role for detoxication in mutant SOD1-mediated ALS, likely in response to increasing oxidative stress in susceptible motor neurons.

Future work will help to understand the role of FMO-mediated detoxication and redox balance in the CNS of sporadic and familial ALS patients.

Acknowledgments

We thank Dr. Caterina Bendotti, Mario Negri Farmacological Institute, Milano, Italy, for supplying SOD1-G93A mice tissues and for her help in writing the materials and methods paragraph. This study was supported by the Ministry of Scientific Research and Ministry of Health (Ex art.56/2003-Neurodegenerative diseases).

Copyright information

© Springer Science+Business Media, LLC 2010