Skip to main content
NCBI home page

This is a preprint.

It has not yet been peer reviewed by a journal.

The National Library of Medicine is running a pilot to include preprints that result from research funded by NIH in PMC and PubMed.

medRxiv logoLink to medRxiv
[Preprint]. 2026 Sep 10:2026.09.08.26362462. [Version 1] doi: 10.64898/2026.09.08.26362462

Effects of sex, treatment resistance, and antidepressant treatment on monoamine oxidase B total distribution volume

Emily JY Kuik 1,2, Catalina Baena-Tan 1, Sho Moriguchi 3,4, Joeffre Braga 1,2, Laura Miler 1, Pablo Rusjan 5, Michael Bagby 6,7,8, M Ishrat Husain 6,9,10, Neil Vasdev 1, Toshifumi Tomoda 2,6,11, Mounira Banasr 2,6,11, Stefan Kloiber 6,12, Jeffrey H Meyer 1,2,6,12
PMCID: PMC13622770  PMID: 42818885

Abstract

Monoamine oxidase B (MAO-B) is a high-density protein located mainly in astrocytes that influences mitochondrial function and transition to astrogliosis. MAO-B produces hydrogen peroxide while metabolizing non-serotonergic monoamines. Elevated MAO-B level is implicated in the pathophysiology of several common neuropsychiatric diseases. The aims are to examine the effect of treatment resistant major depressive episodes, sex, and antidepressant treatment on MAO-B total distribution volume ([11C]SL25.1188 VT), index of MAO-B level. [11C]SL25.1188 PET was applied in 95 adults (29 with MDE and no history of treatment-resistance (NTR-MDE), 26 with MDE and history of treatment-resistance (TRD), and 40 healthy control (HC)). A subset of NTR-MDE and TRD (n=13) were scanned before and after treatment with phenelzine, rasagiline, or duloxetine. [11C]SL25.1188 VT was higher in grey matter regions in females (10%, p<.001). [11C]SL25.1188 VT was also higher in prefrontal cortex in TRD compared to HC (23%, p<.001) and TRD compared to NTR-MDE (6%, p=.033). Duloxetine had minimal effect on [11C]SL25.1188 VT, but occupancy of rasagiline and phenelzine was ~94%. This study discovered sex differences in MAO-B level, which may explain sex differences in prevalence or trajectory of illnesses with greater MAO-B level like Alzheimer’s disease, traumatic brain injury and major depressive disorder. Given our findings of greater MAO-B level in prefrontal cortex of TRD and its insensitivity to serotonin reuptake inhibition, combined with potential harm of highly elevated MAO-B level, development of MAO-B inhibitors could be considered for a subset of TRD. Rasagiline and phenelzine demonstrate similarly potent occupancy in vivo at clinical dosage.

Introduction

Monoamine oxidase B (MAO-B), a high-density enzyme in the human brain (1.8 to 34 pmol/mg in grey matter), is primarily located on the outer mitochondrial membrane in astrocytes as well as the cell bodies of serotonergic and dopaminergic neurons [1, 2]. In brain tissue, MAO-B level is of considerable interest because it is highly correlated with MAO-B activity [2, 3]. MAO-B is an ongoing target of drug development [4, 5], as it plays important roles in the pathophysiology of psychiatric and neurological diseases. Illnesses with greater MAO-B level include the prefrontal cortex (PFC) of major depressive disorder (MDD) [6], the grey matter of Alzheimer’s disease [7], the grey matter of long COVID [8], the cortex of traumatic brain injury with chronic persistent symptoms [9], and the midbrain of Parkinson’s disease [10]. Previous studies demonstrated a number of mechanisms by which MAO-B mediates pathophysiological consequences. MAO-B produces reactive oxygen species (i.e., hydrogen peroxide) while metabolizing non-serotonergic monoamines, such as dopamine, norepinephrine, and phenylethylamine [11, 12]. Greater production of reactive oxygen species could account for negative impacts of overexpressed MAO-B on mitochondrial complex I activity [12] and hence oxidative phosphorylation. MAO-B also contributes to the development and persistence of astrogliosis, a pathologically activated state of glial fibrillary acidic protein (GFAP)-positive astrocytes [12–15]. In addition, stress and the stress-associated steroid hormones, glucocorticoids, upregulate MAO-B transcription via direct binding of the glucocorticoid receptor to the glucocorticoid response element 4 (GRE4) in the MAO-B promoter. An alternative mechanism of glucocorticoid effect involves transcriptional repressors (i.e., E2F-associated phosphoprotein (EAPP) and R1), which compete with a transcriptional activator, Simian virus promoter factor 1 (Sp1), for binding to the Sp1 sites in the MAO-B promoter region. Under stress conditions, reduced occupancy of EAPP and R1 at the Sp1 site in the MAO-B promoter therefore leads to elevated level of MAO-B and hence greater MAO-B activity [16–18].

[11C]SL25.1188 VT, an index of MAO-B level, may be measured in vivo in humans applying the positron emission tomography (PET) radiopharmaceutical [11C]SL25.1188. This radiopharmaceutical has outstanding qualities, including very high specific binding to non-displaceable binding ratio (>8), excellent reversibility, high brain uptake, selectivity for MAO-B, and an absence of brain-penetrant radioactive metabolites [19, 20].

The first aim of this study is to investigate whether there are differences in PFC MAO-B VT during major depressive episodes with a history of treatment resistance (TRD) compared to major depressive episodes with no history of treatment resistance (NTR-MDE). Greater PFC MAO-B VT was previously found during MDE of MDD, and MAO-B is not targeted directly by first- and second-line serotonin and serotonin and norepinephrine reuptake inhibitor medication treatments [21]. It is hypothesized that the highest level of PFC MAO-B VT is associated with TRD because overexpression of MAO-B is associated with greater removal of non-serotoninergic monoamines, greater production of reactive oxygen species, and loss of mitochondrial complex I [11, 12]. Each of these sequelae are also not direct targets of serotonin reuptake inhibitor treatments, yet may contribute to MDE symptoms: In humans, depletion of non-serotonergic monoamines, such as through α-methylparatyrosine administration, may rapidly induce depressive symptoms [22]. Greater production of reactive oxygen species has been proposed to create risk for depressive symptoms [23, 24] and genetic disorders with loss of mitochondrial complex I level or activity are associated with high risk of depressive symptoms [25, 26].

The second aim of this study is to determine whether there are any differences in MAO-B VT in females compared to males across grey matter regions, including the PFC, anterior cingulate cortex (ACC), dorsal putamen, ventral striatum, thalamus, and hippocampus. This study measures an index of MAO-B level, and MAO-B level is consistently correlated with MAO-B activity in brain tissue, yet the effect of sex on either had not been prioritized in previous studies of the human brain. The present study is better suited to address this question due to its sampling, being balanced between males and females, and of relatively large sample size, including about twice as many females as the largest samples among previous studies. Since estrogens lower the expression of MAO-B transcription by downregulating the expression of oestrogen-related receptors, which constitutively upregulate MAO-B gene transcription [27, 28], it is hypothesized that MAO-B VT will be lower in females in the grey matter regions prioritized. In addition, as mechanisms to account for potential differences in MAO-B VT between sex have had minimal investigation, we also explored gene expression differences between females and males in a publicly-available microarray dataset derived from postmortem human brain tissue, focusing on those related to MAO-B expression.

The third aim of this study is to investigate the effects of rasagiline, an MAO-B inhibitor; phenelzine, an MAO-A and MAO-B inhibitor; and duloxetine, a selective serotonin and norepinephrine reuptake inhibitor (SNRI), on MAO-B VT, and characterize MAO-B occupancy for the first two medications. Occupancy may be derived from VT measurements in which there is a substantial lowering in VT at the time of the second PET scan, which is expected for the first two medications [29]. Rasagiline occupancy is of interest to justify potential repurposing because it is a well-tolerated, selective, and irreversible MAO-B inhibitor that typically does not require the dietary restriction of low tyramine intake for inhibitors of both MAO-A and MAO-B [30].

Tyramine is metabolized by both MAO-A and MAO-B, so substantial inhibition of only one enzyme still allows adequate metabolism of tyramine, whereas medications with substantial inhibition of both MAO-A and MAO-B require dietary restriction to avoid hypertensive crisis from elevated plasma tyramine level [11]. Rasagiline is primarily used for Parkinson’s disease and has not been advanced for repurposing in other illnesses [30]. Phenelzine represents a positive control, as it is a nonselective irreversible MAO inhibitor that has an indication for MDE of MDD but is not commonly used because it requires dietary restriction of tyramine and cannot be given with other monoamine-influencing antidepressants [31]. Other advantages of determining phenelzine occupancy as compared to the other option of tranylcypromine is that the occupancy of phenelzine for MAO-A is known at common treatment dose, and phenelzine is sufficiently well-tolerated to reach the clinical dosing associated with its initial development in open trial, whereas tranylcypromine is difficult to tolerate at dosage supported by clinical trial evidence [32–34]. Non-quantitative assessment of rasagiline with [11C]deprenyl PET suggests that rasagiline has detectable occupancy [35], but the present study will provide quantitative results. Duloxetine is anticipated to be a negative control as it is classified as a SNRI with no direct MAO-B inhibition or binding.

Patients and methods

Participants

Participants were either healthy, had NTR-MDE or had TRD. For healthy participants, the main inclusion criteria was no history of psychiatric illness based on the Structured Clinical Interview for DSM-4 or DSM-5 (SCID) [36] and good physical health based on a structured health questionnaire. The key inclusion criteria for NTR-MDE and TRD participants included primary diagnosis of MDE with early-onset MDD, with first MDE prior to age 40, verified by the SCID for DSM-4 or DSM-5 [36] and a score of greater than or equal to 17 on the 17-item Hamilton Depression Rating Scale (HDRS). Categorization as TRD versus NTR-MDE was additionally based on history of non-response to at least two different classes of antidepressants or at least one SNRI. Important exclusion criteria included use of antidepressants within the past 4 weeks for NTR-MDE and TRD, and lifetime use for healthy. Exclusion criteria common to all participants were history of psychotic symptoms, antisocial or borderline personality disorders based on the Structured Clinical Interview for DSM-IV (SCID-II) [37] or Structured Clinical Interview for DSM-5 Personality Disorders Version (SCID-5-PD), and neurodegenerative illness; cigarette smoking for the past 6 months or current use of recreational drugs with recent use verified by urine drug screen; current alcohol use disorder; electroconvulsive therapy or mechanical brain stimulation treatment within the previous 6 months; and use of MAO-B inhibitor treatments within the last 4 weeks.

Additional inclusion and exclusion criteria can be found in Supplementary Materials. Participants provided written informed consent after all procedures were fully explained, and all components of the study were approved by the Research Ethics Board of the Centre for Addiction and Mental Health (CAMH) and Health Canada.

[11C]25.1188 VT Measurement

PET imaging data was acquired from each participant for 90 minutes with a 3D high-resolution research tomograph (HRRT; CPS/Siemens) PET scanner system as previously described [20]. Arterial sampling was done concurrently during the PET scan using an automatic blood sampling system and manual sampling during the emission PET scan. An MRI image (Discovery MR750 3T GE scanner) was acquired for delineation of regions of interest (ROI). [11C]SL25.1188 VT was calculated using a 2-tissue compartment model previously validated for [11C]SL25.1188 PET [20]. Both PET and MRI scans were completed in Toronto. See Supplementary Materials for additional details.

Treatment Interventions on [11C]SL25.1188 VT

A subset of participants also partook in the open trial treatment portion of the study. NTR-MDE received duloxetine. Assignment of TRD participants was substantially consistent with usual treatment algorithms, such that participants received duloxetine if they had no history of non-response to SNRI duloxetine or venlafaxine. Otherwise, TRD participants received phenelzine if they felt they could follow the dietary restriction requirement for phenelzine or rasagiline if they felt they could not follow the dietary restriction requirement for phenelzine. Administration of medication was open label and began the day following the first [11C]SL25.1188 PET scan. Phenelzine was administered at 15 mg once daily and raised every 4 to 7 days by 15 mg to achieve a total daily dose of 45 to 60 mg given over two doses based on tolerability. Rasagiline was administered at 0.5 mg daily for 4 to 7 days, then raised to 1 mg daily for 6 to 8 weeks if well tolerated. Otherwise, dosing remained at 0.5 mg daily. Duloxetine was given at 30 mg daily in week 1, then raised to 60 mg daily for 6 weeks. The follow-up PET scan occurred 4 to 6 weeks after titration to the optimized dose and was further timed to avoid peak plasma level for the MAO-B inhibitors (i.e., scanning occurred 6 to 24 hours after last dose for each drug). Non-responders to medication received a tapering dose for one week before the medication was stopped (unless they were already taking the minimum dose of rasagiline at 0.5 mg daily). Responders were given the option to continue treatment at the dose associated with response, with a transition to their regular treating physician. Compliance was assessed via pill count done halfway through the trial and at the end of the trial.

Exploration of Sex Differences in mRNA

Microarray data for 18 females and 20 males from a sample of individuals with major depressive disorder (n=19) and controls (n=19) from the Gene Expression Omnibus dataset GSE53987, sampling the hippocampus, dorsolateral prefrontal cortex, and associative striatum regions [38, 39], was used to explore the presence of sex differences in the expression of several mRNA related to the MAO-B protein level, including MAO-B, regulators of MAO-B transcription (EAPP, R1), astrogliosis (GFAP), and mitochondrial density (citrate synthase) (Supplemental Materials).

Statistical Analyses

The analysis determining the effect of sex on [11C]SL25.1188 VT was a linear mixed-effects model (LMM) evaluating sex, group, and region (PFC, ACC, dorsal putamen, ventral striatum, thalamus, hippocampus) as fixed effects and participant as a random effect, with [11C]SL25.1188 VT as the dependent variable. Sex was assessed first so as to consider its inclusion in the comparisons of illness groups. The analysis comparing [11C]SL25.1188 VT in PFC between TRD and NTR-MDE applied two similar methodologically overlapping approaches. One was an ANCOVA with group and sex as predictor variables and whole PFC [11C]SL25.1188 VT as the dependent variable. The second analysis applied a LMM evaluating sex, group, region (dorsolateral PFC, ventrolateral PFC, medial PFC, and orbitofrontal cortex), and group*region as fixed effects and participant as a random effect, with [11C]SL25.1188 VT as the dependent variable.

To assess whether there was a change in [11C]SL25.1188 VT before and after antidepressant treatment, the impact for each drug was determined separately by applying a LMM evaluating time and region (PFC, ACC, dorsal putamen, ventral striatum, thalamus, hippocampus) as fixed effects and participant as a random effect, with regional [11C]SL25.1188 VT as the repeated dependent variable. For interventions with a substantial change of greater than 30% in [11C]SL25.1188 VT, a Lassen plot was applied to determine a mean occupancy of [11C]SL25.1188 VT across brain regions and a 95% confidence interval for occupancy was found for each intervention for which a substantive change in [11C]SL25.1188 VT was present.

Results

Participants

Ninety-five participants, age 18 to 65 years (52 females, 43 males), were included in the analyses. Of these, 40 were healthy, 29 had NTR-MDE and 26 had TRD. Age and sex were not significantly different across all three groups (Table 1). There were no significant differences in HDRS score at screening, number of previous MDEs, and duration of illness across NTR-MDE and TRD groups (Table 1).

Table 1.

Demographic and Clinical Variables

Participants
NTR-MDE (n = 29) TRD (n = 26) Healthy Controls (n = 40) p
Age (years) 31.3 ± 10.4 35.7 ± 12.9 31.9 ± 10.9 .294
Sex (Female:Male) 18:11 11:15 23:17 .305
Hamilton Depression Rating Scale 19.0 ± 4.2 19.9 ± 3.7 - .428
Number of past MDEsi 4.3 ± 3.0 6.1 ± 3.4 - .055
Age of onset (years) 23.5 ± 8.6 20.9 ± 8.7 - .323
Open in a new tab

Data presented as mean ± standard deviation, unless stated otherwise.

Statistical analyses were performed using analysis of variance (ANOVA), with the exception of sex, which was performed using χ2 test.

i

The maximum number of MDEs counted is 10.

Effect of sex on [11C]SL25.1188 VT

[11C]SL25.1188 VT was significantly higher in females compared to males in the PFC, ACC, ventral striatum, dorsal putamen, thalamus and hippocampus (LMM; F1,86 = 17.57, p <.001; Fig. 1 and Table 2). An additional analysis extending to include more grey matter regions (parietal cortex, occipital cortex, temporal cortex and midbrain) with the same predictor variables also found that [11C]SL25.1188 VT was significantly elevated in females (LMM; F1,88 = 16.30, p <.001; Table 2).

Figure 1: [11C]SL25.1188 total distribution volume (VT) in brain regions of interest in females vs. males.

Figure 1:

Open in a new tab

A linear mixed-effects model found an effect of sex on [211C]SL25.1188 VT, with greater [11C]SL25.1188 VT in females (F1,86 = 17.57, p <.001). Filled triangles represent males (n = 43). Hollow triangles represent females (n = 52). Black triangles represent healthy controls (n = 40). Blue triangles represent MDE participants, which includes both NTR-MDE and TRD participants (n = 29). Red horizontal bars represent means.

Table 2.

Regional [11C]SL25.1188 VT in females and males.

Females (n = 52) Males (n = 43)
Region of Interest Mean SD Mean SD Difference F p
Prefrontal Cortex 32.7 5.6 28.9 5.5 12.3% 11.352 .001
Dorsolateral Prefrontal Cortex 32.7 5.2 28.6 6.2 13.4% 12.517 <.001
Ventrolateral Prefrontal Cortex 29.7 5.1 26.4 5.4 11.8% 9.477 .003
Medial Prefrontal Cortex 36.4 6.0 32.7 6.1 10.7% 8.668 .004
Orbitofrontal Cortex 29.6 5.7 27.3 4.9 8.1% 4.410 .038
Anterior Cingulate Cortex 39.5 6.2 35.7 6.8 10.1% 7.944 .006
Hippocampus 37.0 6.2 32.5 6.1 13.0% 13.023 <.001
Dorsal Putamen 50.3 7.2 44.7 8.3 11.8% 12.339 <.001
Ventral Striatum 60.3 12.7 53.6 11.2 11.8% 7.290 .008
Thalamus 54.9 9.5 50.2 9.6 8.9% 5.824 .018
Midbrain 37.9 7.1 34.6 6.7 9.1% 5.404 .022
Parietal Cortex 31.8 4.7 29.3 5.4 8.2% 5.933 .017
Temporal Cortex 31.9 4.8 29.8 6.0 6.8% 3.677 .058
Occipital Cortex 27.3 3.9 25.0 4.4 8.8% 7.337 .008
Open in a new tab

Means and standard deviations presented in table are derived from raw unadjusted VT values.

P values are derived from an analysis of variance (ANOVA).

Effect of TRD versus NTR-MDE or Health on [11C]SL25.1188 VT

[11C]SL25.1188 VT in PFC was significantly elevated in TRD compared to NTR-MDE (ANCOVA, factors of group and sex; effect of group: F1,52 = 4.78, p = 0.033; Fig. 2 and Table S1). Similar results were found comparing [11C]SL25.1188 VT in TRD to NTR-MDE across subregions of the PFC, as there was a group effect but no subregion-specific difference (LMM; predictors of group, region, sex, group by region; effect of group: F1,52 = 4.45, p = .040; Fig. 2 and Table S1). Consistent with these findings, [11C]SL25.1188 VT in PFC was significantly elevated in TRD compared to the healthy group (ANCOVA, factors of group and sex; effect of group: F1,63 = 35.39, p < 0.001; Fig. 2 and Table S2). There was also an effect of group comparing subregions of the PFC but no subregion-specific findings (LMM; predictors of group, region, sex, group by region; effect of group: F1,63 = 19.23, p < 0.001; Fig. 2 and Table S2).

Figure 2: [11C]SL25.1188 total distribution volume (VT) in brain regions of interest in participants with NTR-MDE vs. TRD vs. healthy controls.

Figure 2:

Open in a new tab

A linear mixed-effects model found an effect of group on [311C]SL25.1188 VT, with greater [11C]SL25.1188 VT in PFC subregions in TRD compared to NTR-MDE (F1,52 = 4.45, p = .040) and healthy controls (F1,63 = 19.23, p < 0.001). Black triangles represent healthy controls (n = 40). Blue triangles represent NTR-MDE participants (n = 29). Purple triangles represent TRD (n = 26). Filled triangles represent males (n = 43). Hollow triangles represent females (n = 52). Red horizontal bars represent means.

Change in [11C]SL25.1188 VT after treatment

Compliance was high, with 100%, 97.5%, and 99.6% of the medication taken on pill counts for rasagiline, phenelzine and duloxetine respectively.

When including all participants in the LMM, including treatment, region, and medication group, there was an effect of medication by treatment (F2,57 = 84.63, p <.001). When analyzing each individual treatment, [11C]SL25.1188 VT was significantly lower in the PFC, ACC, dorsal putamen, ventral striatum, thalamus and hippocampus after treatment with rasagiline or phenelzine (LMM; rasagiline: F1,10 = 502.547, p < 0.001; phenelzine: F1,7 = 1091.052, p < 0.001; Fig. 3 and Table S3). The residuals in the model for the data of each treatment were not normally distributed, so results were confirmed with ANOVA within individual regions (Table S3). The LMM comparing [11C]SL25.1188 VT before and after participants took duloxetine found an effect of treatment (F1,18 = 8.439, p =.010; Fig. 3), but none of the ANOVAs within individual regions had uncorrected p values below 0.05, reflecting that the change after duloxetine was modest. Mean occupancy values, calculated across brain regions for each participant using a Lassen plot, was 95.4% for rasagiline (95% CI (92.15, 98.60)), and 93.7% for phenelzine (95% CI (86.19, 101.28)).

Figure 3: [11C]SL25.1188 total distribution volume (VT) in brain regions of interest in participants with MDE pre- and post-treatment.

Figure 3:

Open in a new tab

A linear mixed-effects model including treatment, region, and medication group found an effect of medication by treatment (F42,57 = 84.630, p <.001), reflecting greater effect of phenelzine and rasagiline compared to duloxetine. Blue circles represent participants who took phenelzine (n = 4). Purple circles represent participants who took rasagiline (n = 5). Grey circles represent participants who took duloxetine (n = 4). Red horizontal lines represent group means.

Consistency of imaging agent administered, and free fraction across groups analyzed

[11C]SL25.1188 was of high radiochemical purity, with a minimum threshold of 95% and high specific activity (mean ± SD = 76.40 ± 42.18 TBq/mmol at the time of injection). There was no significant difference in specific activity between the two sexes (ANOVA, F1,93 = .838, p = .362) or diagnosis (ANOVA, F2,92 = 1.10, p = .338). There was also no significant difference between % free fraction between the two sexes (ANOVA, F1,86 = .951, p = .332) or across diagnostic subgroups (ANOVA, F2,84 = .003, p = .997) (Table S4).

Exploration of Sex Differences in mRNA

None of the exploratory comparisons of mRNA across sex in the previously published microarray dataset yielded results at an uncorrected significance of less than 0.05 (Table S5).

Discussion

To our knowledge, this is the first study to report sex differences in MAO-B levels in human brain. Arguably, the most novel finding was that females had significantly higher [11C]SL25.1188 VT compared to males, which influences our understanding of sex differences for neuropsychiatric illnesses in which differential MAO-B level is implicated in modulating their pathophysiology. To our awareness, this is also the first study to investigate MAO-B in TRD, and we found significantly greater [11C]SL25.1188 VT in the PFC in TRD compared to NTR-MDE and healthy controls, prompting consideration for targeting MAO-B in treatment resistance. Finally, we also found that rasagiline and phenelzine, in contrast to duloxetine, lowered [11C]SL25.1188 VT across grey matter regions; and that rasagiline and phenelzine have high MAO-B occupancy. This confirms high potency of phenelzine and rasagiline at the treatment doses in clinical settings, pointing to the opportunity for future clinical development of rasagiline.

Our finding of greater [11C]SL25.1188 VT in females was unexpected and has potential implications for sex-related differences in neuropsychiatric illnesses. MAO-B has a critical role in promoting GFAP-labelled astrogliosis, since overexpression of MAO-B promotes astrogliosis, knockdown or blockade of MAO-B reduces astrogliosis and MAO-B expression is often increased during astrogliosis [12–15]. A general, albeit simplified literature perspective is that acute astrogliosis tends to have protective advantages whereas chronic astrogliosis tends to be harmful [40, 41]. Illnesses such as Alzheimer’s disease and long COVID have higher prevalence in females [42, 43], and chronic traumatic brain injury symptoms have a longer time course in females [44]. Chronic astrogliosis with heightened level of MAO-B has been raised as an aggravating factor in the pathophysiology of all of these illnesses. We also note that greater MAO-B level can occur independently of astrogliosis, and that greater MAO-B binding was identified in the PFC of MDD, another illness that has a higher prevalence ratio of females to males. In regards to the mechanism for sex-related differences, the present finding is opposite to the original hypothesis that differential estrogen level between sex would account for a lower MAO-B level in females. Another possibility for further study is that given the MAO-B gene is located on the X chromosome, it is possible that it escapes X-chromosome inactivation. A previous report indicated that females have both greater MAO-A and MAO-B mRNA expression in grey matter throughout their lifespan [45]. However, it is acknowledged that greater mRNA level does not necessarily translate to measures of greater protein level [45], as we previously found that [11C]harmine VT, an index of MAO-A level, did not differ between males and females prior to age 41 outside of the immediate postpartum period [46–48]. Also, an exploratory analysis of a publicly available microarray dataset derived from postmortem human brain tissue yielded no difference in mRNA for MAO-B, or mRNA corresponding to markers of astrogliosis or mitochondrial density within the hippocampus, dorsolateral prefrontal cortex, or associative striatum. This raises the possibility of sex differences in MAO-B translation or stability as other explanations for sex related differences (see Supplementary Discussion for further detail).

Our finding that individuals with TRD have elevated PFC [11C]SL25.1188 VT compared to NTR-MDE as well as healthy controls raises the question as to whether greater PFC MAO-B level may contribute to treatment resistance. Elevated MAO-B level beyond healthy control level is implicated in causing MDE symptoms, as generalized overexpression of MAO-B in mice leads to reduced locomotor speed and activity, shorter distances traveled and decreased time spent in motion in open field assessments [12]. Also, as presented in the introduction, sequelae of greater MAO-B level led to greater removal of non-serotoninergic monoamines, greater production of reactive oxygen species and loss of mitochondrial complex I [11, 12], processes also implicated in causing depressive symptoms. The present study also demonstrated that [11C]SL25.1188 VT is not substantively affected by SNRI treatment. So, given that MAO-B level is not well targeted by SNRIs, is more prominently elevated in TRD, and elevation in MAO-B level from the usual state predisposes towards symptoms and sequelae that may contribute to depressive symptoms, it may be worth assessing whether MAO-B inhibitors may have therapeutic efficacy in TRD.

Non-selective MAO inhibitors like phenelzine have some evidence for benefit in clinical trials for treatment resistant depression [32], but as the rationale was largely ascribed to MAO-A inhibition [11], selective MAO-B inhibitors have not had similar investigation. Non-selective MAO inhibitor tranylcypromine was assessed in treatment resistance in a STAR*D trial, but unfortunately, poor tolerability of tranylcypromine in that trial led to high non-completion rates and dosing that may have been lower than optimal for treatment response [34, 49]. It is also uncertain whether the relatively low dose used in that trial achieved a substantive MAO-B occupancy [34].

Both phenelzine and rasagiline caused high magnitude reductions of [11C]SL25.1188 VT with occupancy of 94 to 95 per cent in contrast to duloxetine. One implication is that phenelzine at a total daily dose of 45 to 60mg is associated with substantive occupancy, suggesting that such dose is often adequate for target engagement of MAO-B inhibition. A second implication is that high occupancy of rasagiline, being similar to the antidepressant phenelzine, argues for consideration of development of rasagiline to target MAO-B in mood disorders, moving beyond its approvals presently obtained only for Parkinson’s disease. For example, there are reports that rasagiline improves depression outcomes in Parkinson’s disease even when controlling for motor function [50, 51], and decreases avolition, a symptom also commonly found in MDD [52]. A third consideration is that new and more selective MAO-B inhibitors in development [4, 5] should consider the high occupancies of phenelzine and rasagiline as comparative, achievable thresholds when determining occupancy for therapeutic trials.

There are some limitations to the present study, some common to PET imaging investigations. First, [11C]SL25.1188 VT reflects both specific binding of the radiotracer to MAO-B and non-specific binding. However, based on the Lassen plots, 88 to 95 per cent of VT is represented by specific binding, so changes in non-displaceable binding are unable to account for sex differences or effects of phenelzine or rasagiline, and would require large improbable changes to account for the effect of treatment resistance. Second, while typical of PET occupancy studies, the number of participants enrolled in the treatment component of the study was low in the study plan because the variability in the blocked condition is small. Third, the magnitude of difference in [11C]SL25.1188 VT, while substantive comparing TRD to healthy, was modest in the comparison of TRD to NTR-MDE, so the notion of investigating MAO-B inhibitors as clinical interventions in TRD presented earlier might only be relevant to a subset of TRD, requiring a stratification approach in future investigations.

In conclusion, the main findings were greater [11C]SL25.1188 VT in females and in TRD, and that both phenelzine and rasagiline strongly reduced [11C]SL25.1188 VT, unlike duloxetine. The interpretation of greater MAO-B level in brain grey matter in females raises important implications as to whether this contributes to sex differences observed in illnesses with prominent MAO-B-labelled astrogliosis like Alzheimer’s disease, traumatic brain injury or long COVID and/or illnesses with only elevated MAO-B level like major depressive disorder. The interpretation that greater MAO-B level occurs in TRD, combined with lack of effect of SNRI duloxetine and roles of greater MAO-B level in predisposition to symptoms of depression, raises consideration for greater development of well-tolerated MAO-B inhibitors with substantial occupancy for TRD. Consistent with this direction, we demonstrate that the occupancy for two MAO-B inhibitor medications with clinical indications is high, and comparable across less selective MAO-B inhibitor phenelzine and better tolerated, more selective MAO-B inhibitor rasagiline.

Acknowledgments

The authors acknowledge the valuable contributions of the study participants as well as research staff at the Centre for Addiction and Mental Health.

Funding

This study was supported by the National Institutes of Mental Health (grant 1R01MH115014–01), Brain and Behavior Research Foundation, and Canadian Institutes of Health Research (CIHR). Key infrastructure support was from the Azrieli Foundation, Canada Foundation for Innovation, and Ontario Ministry of Research and Innovation. Dr. Meyer, Dr. Husain and Dr. Vasdev were supported by the Canada Research Chair program. Ms. Kuik was supported by a Doctoral Award funded by the Peterborough K.M. Hunter Charitable Foundation. Dr. Meyer and Dr. Tomoda were supported in part by Womenmind seed fund initiative at CAMH.

Footnotes

Competing interest

Dr. Meyer has applied for patenting of dopaminergic agents, including rasagiline in long COVID (and inflammatory bowel disease is mentioned as an example of another illness with elevated monoamine oxidase B for which the patent could be relevant) and is inventor of a dietary supplement to reduce severity of postpartum blues and postpartum depression manufactured and sold by Exeltis. Dr. Husain has provided consultancy to Mindset Pharma, PsychEd Therapeutics, and Wake Network, and led contracted research for COMPASS Pathfinder Limited. The other authors report no relationships with competing interests.

Data Availability Statement

The data for the main figures and graphs of this study are available within the supplementary material.

References

  • 1.Saura J, Andres N, Andrade C, Ojuel J, Eriksson K, Mahy N. Biphasic and region-specific MAO-B response to aging in normal human brain. Neurobiology of Aging. 1997;18(5):497–507. [DOI] [PubMed] [Google Scholar]
  • 2.Tong J, Meyer JH, Furukawa Y, Boileau I, Chang L-J, Wilson AA, et al. Distribution of monoamine oxidase proteins in human brain: implications for brain imaging studies. Journal of Cerebral Blood Flow & Metabolism. 2013;33(6):863–871. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Saura J, Kettler R, Da Prada M, Richards JG. Quantitative enzyme radioautography with 3H-Ro 41–1049 and 3H-Ro 19–6327 in vitro: localization and abundance of MAO-A and MAO-B in rat CNS, peripheral organs, and human brain. J Neurosci. 1992;12(5):1977–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Georgieva M, Sharkov M, Mateev E, Mateeva A, Kondeva-Burdina M. Evaluation of Pyrrole Heterocyclic Derivatives as Selective MAO-B Inhibitors and Neuroprotectors. Molecules (Basel, Switzerland). Jan 4 2026;31(1)doi: 10.3390/molecules31010186 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Nam MH, Park JH, Song HJ, Choi JW, Kim S, Jang BK, et al. KDS2010, a Newly Developed Reversible MAO-B Inhibitor, as an Effective Therapeutic Candidate for Parkinson's Disease. Neurotherapeutics : the journal of the American Society for Experimental NeuroTherapeutics. Jul 2021;18(3):1729–1747. doi: 10.1007/s13311-021-01097-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Moriguchi S, Wilson AA, Miler L, Rusjan PM, Vasdev N, Kish SJ, et al. Monoamine oxidase B total distribution volume in the prefrontal cortex of major depressive disorder: an [11C] SL25. 1188 positron emission tomography study. JAMA psychiatry. 2019;76(6):634–641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Saura J, Luque J, Cesura A, Da Prada M, Chan-Palay V, Huber G, et al. Increased monoamine oxidase B activity in plaque-associated astrocytes of Alzheimer brains revealed by quantitative enzyme radioautography. Neuroscience. 1994;62(1):15–30. [DOI] [PubMed] [Google Scholar]
  • 8.Braga J, Kuik EJ, Lepra M, Rusjan PM, Kish SJ, Vieira EL, et al. Astrogliosis marker [11C] SL25. 1188 after COVID-19 with ongoing depressive and cognitive symptoms. Biological Psychiatry. 2025;97(8):816–824. [DOI] [PubMed] [Google Scholar]
  • 9.Koshimori Y, Cusimano MD, Vieira EL, Rusjan PM, Kish SJ, Vasdev N, et al. Astrogliosis marker 11C-SL25. 1188 PET in traumatic brain injury with persistent symptoms. Brain. 2023;146(11):4469–4475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Tong J, Rathitharan G, Meyer JH, Furukawa Y, Ang L-C, Boileau I, et al. Brain monoamine oxidase B and A in human parkinsonian dopamine deficiency disorders. Brain. 2017;140(9):2460–2474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Youdim MB, Edmondson D, Tipton KF. The therapeutic potential of monoamine oxidase inhibitors. Nature Reviews Neuroscience. Apr 2006;7(4):295–309. [DOI] [PubMed] [Google Scholar]
  • 12.Mallajosyula JK, Kaur D, Chinta SJ, Rajagopalan S, Rane A, Nicholls DG, et al. MAO-B elevation in mouse brain astrocytes results in Parkinson's pathology. PLoS One. Feb 20 2008;3(2):e1616. doi: 10.1371/journal.pone.0001616 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Chun H, Lim J, Park KD, Lee CJ. Inhibition of monoamine oxidase B prevents reactive astrogliosis and scar formation in stab wound injury model. Glia. Feb 2022;70(2):354–367. doi: 10.1002/glia.24110 [DOI] [PubMed] [Google Scholar]
  • 14.Jaisa-Aad M, Munoz-Castro C, Healey MA, Hyman BT, Serrano-Pozo A. Characterization of monoamine oxidase-B (MAO-B) as a biomarker of reactive astrogliosis in Alzheimer's disease and related dementias. Acta Neuropathol. Apr 3 2024;147(1):66. doi: 10.1007/s00401-024-02712-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ekblom J, Jossan SS, Oreland L, Walum E, Aquilonius SM. Reactive gliosis and monoamine oxidase B. J Neural Transm Suppl. 1994;41:253–8. [DOI] [PubMed] [Google Scholar]
  • 16.Udemgba C, Johnson S, Stockmeier CA, Luo J, Albert PR, Wang J, et al. The expression of KLF11 (TIEG2), a monoamine oxidase B transcriptional activator in the prefrontal cortex of human alcohol dependence. Alcohol Clin Exp Res. Jan 2014;38(1):144–51. doi: 10.1111/acer.12229 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Chen K, Ou XM, Wu JB, Shih JC. Transcription factor E2F-associated phosphoprotein (EAPP), RAM2/CDCA7L/JPO2 (R1), and simian virus 40 promoter factor 1 (Sp1) cooperatively regulate glucocorticoid activation of monoamine oxidase B. Mol Pharmacol. Feb 2011;79(2):308–17. doi: 10.1124/mol.110.067439 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Raitsin S, Tong J, Kish S, Xu X, Magomedova L, Cummins C, et al. Subchronic glucocorticoids, glutathione depletion and a postpartum model elevate monoamine oxidase a activity in the prefrontal cortex of rats. Brain Res. Jul 1 2017;1666:1–10. doi: 10.1016/j.brainres.2017.03.032 [DOI] [PubMed] [Google Scholar]
  • 19.Vasdev N, Sadovski O, Garcia A, Dollé F, Meyer JH, Houle S, et al. Radiosynthesis of [11C] SL25. 1188 via [11C] CO2 fixation for imaging monoamine oxidase B. Journal of Labelled Compounds and Radiopharmaceuticals. 2011;54(10):678–680. [Google Scholar]
  • 20.Rusjan PM, Wilson AA, Miler L, Fan I, Mizrahi R, Houle S, et al. Kinetic modeling of the monoamine oxidase B radioligand [11C] SL25. 1188 in human brain with high-resolution positron emission tomography. Journal of Cerebral Blood Flow & Metabolism. 2014;34(5):883–889. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Owens MJ, Morgan WN, Plott SJ, Nemeroff CB. Neurotransmitter receptor and transporter binding profile of antidepressants and their metabolites. J Pharmacol Exp Ther. 1997;283(3):1305–22. [PubMed] [Google Scholar]
  • 22.Bremner JD, Vythilingam M, Ng CK, Vermetten E, Nazeer A, Oren DA, et al. Regional brain metabolic correlates of α-methylparatyrosine–induced depressive symptoms: implications for the neural circuitry of depression. Jama. 2003;289(23):3125–3134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Correia AS, Cardoso A, Vale N. Oxidative Stress in Depression: The Link with the Stress Response, Neuroinflammation, Serotonin, Neurogenesis and Synaptic Plasticity. Antioxidants (Basel). Feb 13 2023;12(2)doi: 10.3390/antiox12020470 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Ait Tayeb AEK, Poinsignon V, Chappell K, Bouligand J, Becquemont L, Verstuyft C. Major Depressive Disorder and Oxidative Stress: A Review of Peripheral and Genetic Biomarkers According to Clinical Characteristics and Disease Stages. Antioxidants (Basel). Apr 17 2023;12(4)doi: 10.3390/antiox12040942 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Morava E, Gardeitchik T, Kozicz T, de Boer L, Koene S, de Vries MC, et al. Depressive behaviour in children diagnosed with a mitochondrial disorder. Mitochondrion. Aug 2010;10(5):528–33. doi: 10.1016/j.mito.2010.05.011 [DOI] [PubMed] [Google Scholar]
  • 26.Fattal O, Link J, Quinn K, Cohen BH, Franco K. Psychiatric Comorbidity in 36 Adults with Mitochondrial Cytopathies. CNS Spectrums. 2007 2014;12(6):429–438. [DOI] [PubMed] [Google Scholar]
  • 27.Zhang Z, Chen K, Shih JC, Teng CT. Estrogen-related receptors-stimulated monoamine oxidase B promoter activity is down-regulated by estrogen receptors. Molecular Endocrinology. 2006;20(7):1547–1561. [DOI] [PubMed] [Google Scholar]
  • 28.Gundlah C, Lu NZ, Bethea CL. Ovarian steroid regulation of monoamine oxidase-A and B mRNAs in the macaque dorsal raphe and hypothalamic nuclei. Psychopharmacology. 2002;160(3):271–282. [DOI] [PubMed] [Google Scholar]
  • 29.Cunningham VJ, Rabiner EA, Slifstein M, Laruelle M, Gunn RN. Measuring drug occupancy in the absence of a reference region: the Lassen plot re-visited. Journal of Cerebral Blood Flow & Metabolism. 2010;30(1):46–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Finberg JP. Pharmacology of rasagiline, a new MAO-B inhibitor drug for the treatment of Parkinson’s disease with neuroprotective potential. Rambam Maimonides Medical Journal. 2010;1(1) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Matveychuk D, MacKenzie EM, Kumpula D, Song M-S, Holt A, Kar S, et al. Overview of the neuroprotective effects of the MAO-inhibiting antidepressant phenelzine. Cellular and Molecular Neurobiology. 2022;42(1):225–242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.McGrath PJ, Stewart JW, Nunes EV, Ocepek-Welikson K, Rabkin JG, Quitkin FM, et al. A double-blind crossover trial of imipramine and phenelzine for outpatients with treatment-refractory depression. Am J Psychiatry. 1993;150(1):118–23. [DOI] [PubMed] [Google Scholar]
  • 33.Chiuccariello L, Cooke RG, Miler L, Levitan RD, Baker GB, Kish SJ, et al. Monoamine Oxidase-A Occupancy by Moclobemide and Phenelzine: Implications for the Development of Monoamine Oxidase Inhibitors. Int J Neuropsychopharmacol. Aug 27 2015;19(1)doi: 10.1093/ijnp/pyv078 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.McGrath PJ, Stewart JW, Fava M, Trivedi MH, Wisniewski SR, Nierenberg AA, et al. Tranylcypromine versus venlafaxine plus mirtazapine following three failed antidepressant medication trials for depression: a STAR*D report. Am J Psychiatry. Sep 2006;163(9):1531–41; quiz 1666. doi: 10.1176/ajp.2006.163.9.1531 [DOI] [PubMed] [Google Scholar]
  • 35.Freedman NM, Mishani E, Krausz Y, Weininger J, Lester H, Blaugrund E, et al. In vivo measurement of brain monoamine oxidase B occupancy by rasagiline, using 11C-L-deprenyl and PET. Journal of Nuclear Medicine. 2005;46(10):1618–1624. [PubMed] [Google Scholar]
  • 36.First MB. Structured clinical interview for the DSM (SCID). The Encyclopedia of Clinical Psychology. 2014:1–6. [Google Scholar]
  • 37.First MB. Structured clinical interview for DSM-IV axis II personality disorders,(SCID-II). (No Title). 1997; [DOI] [PubMed] [Google Scholar]
  • 38.Edgar R, Domrachev M, Lash AE. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Research. 2002;30(1):207–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Lanz TA, Reinhart V, Sheehan MJ, Rizzo SJS, Bove SE, James LC, et al. Postmortem transcriptional profiling reveals widespread increase in inflammation in schizophrenia: a comparison of prefrontal cortex, striatum, and hippocampus among matched tetrads of controls with subjects diagnosed with schizophrenia, bipolar or major depressive disorder. Translational Psychiatry. 2019;9(1):151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Pekny M, Pekna M. Astrocyte reactivity and reactive astrogliosis: costs and benefits. Physiological Reviews. 2014;94(4):1077–1098. [DOI] [PubMed] [Google Scholar]
  • 41.Sofroniew MV. Astrogliosis. Cold Spring Harbor Perspectives in Biology. 2015;7(2):a020420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Aggarwal NT, Mielke MM. Sex Differences in Alzheimer's Disease. Neurol Clin. May 2023;41(2):343–358. doi: 10.1016/j.ncl.2023.01.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Greenhalgh T, Sivan M, Perlowski A, Nikolich JZ. Long COVID: a clinical update. Lancet. Aug 17 2024;404(10453):707–724. doi: 10.1016/S0140-6736(24)01136-X [DOI] [PubMed] [Google Scholar]
  • 44.Arachchi S, Daly E, Dwivedi A, Ryan L. Sex Differences in Severity and Recovery Following Mild Traumatic Brain Injury: A Systematic Review. Brain Sci. Jan 6 2026;16(1)doi: 10.3390/brainsci16010077 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Sanfilippo C, Castrogiovanni P, Imbesi R, Lazzarino G, Di Pietro V, Li Volti G, et al. Sex-dependent monoamine oxidase isoforms expression patterns during human brain ageing. Mech Ageing Dev. Jul 2021;197:111516. doi: 10.1016/j.mad.2021.111516 [DOI] [PubMed] [Google Scholar]
  • 46.Sacher J, Wilson AA, Houle S, Rusjan P, Hassan S, Bloomfield PM, et al. Elevated brain monoamine oxidase A binding in the early postpartum period. Archives of General Psychiatry. 2010;67(5):468–474. [DOI] [PubMed] [Google Scholar]
  • 47.Rekkas PV, Wilson AA, Lee VWH, Yogalingam P, Sacher J, Rusjan P, et al. Greater monoamine oxidase A binding in perimenopausal age as measured with carbon 11–labeled harmine positron emission tomography. JAMA Psychiatry. 2014;71(8):873–879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Meyer JH, Wilson AA, Sagrati S, Miler L, Rusjan P, Bloomfield PM, et al. Brain monoamine oxidase A binding in major depressive disorder: relationship to selective serotonin reuptake inhibitor treatment, recovery, and recurrence. Arch Gen Psychiatry. Dec 2009;66(12):1304–12. doi:66/12/1304 [pii] 10.1001/archgenpsychiatry.2009.156 [DOI] [PubMed] [Google Scholar]
  • 49.Ricken R, Ulrich S, Schlattmann P, Adli M. Tranylcypromine in mind (Part II): review of clinical pharmacology and meta-analysis of controlled studies in depression. European Neuropsychopharmacology. 2017;27(8):714–731. [DOI] [PubMed] [Google Scholar]
  • 50.Korchounov A, Winter Y, Rössy W. Combined Beneficial Effect of Rasagiline on Motor Function and Depression in De Novo PD. Clinical Neuropharmacology. 2012;35(3):121–124. doi: 10.1097/WNF.0b013e31823b1da8 [DOI] [PubMed] [Google Scholar]
  • 51.Barone P, Santangelo G, Morgante L, Onofrj M, Meco G, Abbruzzese G, et al. A randomized clinical trial to evaluate the effects of rasagiline on depressive symptoms in non-demented Parkinson's disease patients. European Journal of Neurology. 2015;22(8):1184–1191. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Buchanan RW, Weiner E, Kelly DL, Gold JM, Keller WR, Waltz JA, et al. Rasagiline in the Treatment of the Persistent Negative Symptoms of Schizophrenia. Schizophrenia Bulletin. 2014;41(4):900–908. doi: 10.1093/schbul/sbu151 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Data Availability Statement

The data for the main figures and graphs of this study are available within the supplementary material.

Articles from medRxiv are provided here courtesy of Cold Spring Harbor Laboratory Preprints

RESOURCES