Pervasive environmental chemicals impair oligodendrocyte development

Abstract

Exposure to environmental chemicals can impair neurodevelopment^1-4. Oligodendrocytes that wrap around axons to boost neurotransmission may be particularly vulnerable to chemical toxicity as they develop throughout fetal development and into adulthood^5,6. However, few environmental chemicals have been assessed for potential risks to oligodendrocyte development. Here, we utilized a high-throughput developmental screen and human cortical brain organoids, which revealed environmental chemicals in two classes that disrupt oligodendrocyte development through distinct mechanisms. Quaternary compounds, ubiquitous in disinfecting agents, hair conditioners, and fabric softeners, were potently and selectively cytotoxic to developing oligodendrocytes through activation of the integrated stress response. Organophosphate flame retardants, commonly found in household items such as furniture and electronics, were non-cytotoxic but prematurely arrested oligodendrocyte maturation. Chemicals from each class impaired human oligodendrocyte development in a 3D organoid model of prenatal cortical development. In analysis of epidemiological data from the CDC's National Health and Nutrition Examination Survey, adverse neurodevelopmental outcomes were associated with childhood exposure to the top organophosphate flame retardant identified by our oligodendrocyte toxicity platform. Collectively, our work identifies toxicological vulnerabilities specific to oligodendrocyte development and highlights common household chemicals with high exposure risk to children that warrant deeper scrutiny for their impact on human health.

Extended Data Fig. 1:. Screening a library of environmental chemicals in developing oligodendrocytes identifies cytotoxic chemicals and modulators of oligodendrocyte generation.

a, Representative heatmaps of one of six primary screening 384-well plates depicting cytotoxic compounds (red), oligodendrocyte inhibitors (blue), and drivers (green). Viability and percent O1+ oligodendrocytes are normalized to vehicle control (DMSO). Thyroid hormone, a known driver of oligodendrocyte generation, is included as a positive control for oligodendrocyte development. b, Quantification of hits across 6 primary screening plates showing distribution of chemicals identified as cytotoxic (black), drivers (green), and inhibitors (blue).

Extended Data Fig. 2:. Quaternary compounds are specifically cytotoxic to oligodendrocyte development and induce apoptosis.

a, Table of the top use categories for the 206 validated cytotoxic chemicals and the number of chemicals belonging to each category. b, Venn diagram showing the overlap of 206 validated cytotoxic chemicals identified in the oligodendrocyte screen (in red) compared to cytotoxic hits identified in an identical screen performed in mouse astrocytes (in gray). c, Quaternary compounds tested in 10-point dose response from (40 nM to 20 μM), on developing oligodendrocytes quantifying cell number (DAPI+). Data are presented as the mean value from 3 biological replicates (OPC batches generated from independent mPSC lines). d, Representative immunohistochemistry images of mPSC-derived oligodendrocytes (O1+, in green) and primary mouse oligodendrocytes treated with methyltrioctylammonium chloride and tributyltetradecylphosphonium chloride. Nuclei are marked by DAPI (in blue). e, Quantification of viability of mouse primary and PSC-derived oligodendrocytes treated with 20µM methyltrioctylammonium chloride or tributyltetradecylphosphonium chloride. Data are presented as the mean ± standard deviation from three biological replicates, represented by closed circles (from independent primary OPC isolations or OPC batches from independent mPSC lines). f, Quantification of oligodendrocyte viability normalized to DMSO control. Developing oligodendrocytes were cultured for 3 days in the presence of 120 nM methyltrioctylammonium chloride or 100 nM tributyltetradecylphosphonium chloride at (approximate IC75 in mPSC-derived OPCs), and cell death inhibitors QVD-OPH, necrostatin-1, and ferrostatin-1, in 8-point dose response (80 nM to 10 µM). Data are presented as the mean ± standard deviation from two biological replicates, represented by closed circles (OPC batches generated from independent mPSC lines).

Extended Data Fig. 3:. Organophosphate flame retardants inhibit oligodendrocyte development.

a, Primary chemical screen of 1,539 non-cytotoxic environmental chemicals showing the effect of individual chemicals on oligodendrocyte generation, presented as percent O1+ cells normalized to the DMSO control, as shown in Fig. 2a. Two dotted lines show the hit cutoffs for identification of oligodendrocyte drivers and inhibitors. Drivers result in an increase of O1+ percentage by 22% (>3 standard deviations) compared to negative DMSO control. Inhibitors reduce O1+ percentage by more than 50% (>7 standard deviations) compared to negative DMSO control. Thyroid modulators are highlighted in yellow. b, Table shows IC50 concentrations, cytotoxicity median values, and use categories for three organophosphate esters identified as inhibitors of oligodendrocyte development. c, Representative immunohistochemistry images of oligodendrocytes, generated from mPSC-derived OPCs and mouse primary OPCs, tested with three organophosphate flame retardants at 20 μM. Generation of oligodendrocytes was evaluated using the oligodendrocyte marker O1 (green). Nuclei are marked with DAPI (in blue). d, Quantification O1+ mPSC-derived and primary oligodendrocytes, shown as a percentage of DAPI+ cell number, across three biological replicates, represented at closed circles (independent isolations of primary OPCs and OPC batches generated from independent mPSC lines). e, Immunohistochemistry images of early (O4+, in magenta), intermediate (O1+, in green), and late (MBP+, in yellow) oligodendrocytes treated with 20 µM TBPP or TMPP. Control images and TDCIPP treated oligodendrocytes are shown in Fig. 2e. Nuclei are marked with DAPI (in blue). f, Quantification of primary oligodendrocytes at the early (O4+), intermediate (O1+), and late (MBP+) stage, shown as a percentage of DAPI+ cell number, over three days of development. Data are presented as the mean ± standard deviation from three biological replicates (OPC batches generated from independent mPSC lines), indicated by closed circle data points. p-values were calculated using one-way ANOVA with Dunnett post-test correction for multiple comparisons.

Extended Data Fig. 4:. TDCIPP is associated with abnormal neurodevelopmental outcomes in children.

a, Venn diagram showing co-occurrence of three neurodevelopmental outcomes in the study population. b, Density plots showing the distribution of urine BDCIPP levels within individual quintiles. c, Adjusted odds ratio for the neurodevelopmental outcome “sought mental health treatment”. Significant odds ratios are highlighted in blue (BDCIPP Q5 v Q1 OR = 4.6 [95% CI = 1.785–12.104) and significant covariates are highlighted in gray (p < 0.001).

Fig. 1:. Quaternary compounds are potently cytotoxic to developing oligodendrocytes.

a, Schematic of the primary chemical screen in mPSC-derived oligodendrocytes. b, Pie chart of the number of cytotoxic chemicals (black), inhibitors of oligodendrocyte development (blue), and drivers of oligodendrocyte development (green) identified from the primary chemical screen, along with chemicals that had no effect (gray). c, Representative immunohistochemistry images after 3 days of oligodendrocyte development. Each image shows cells cultured with DMSO (vehicle control), or one of three chemicals with different effects on oligodendrocyte generation. Nuclei are marked using DAPI (in blue) and oligodendrocytes are marked using O1 (in green). d, Primary chemical screen showing the effect of 1,823 environmental chemicals on the viability of developing oligodendrocytes displayed as viability normalized to vehicle control. The solid line represents the average of the vehicle control set at 100%. The dotted line marks a reduction in viability of 30% (>3 standard deviations). The 206 cytotoxic hits that pass this threshold and were validated by MTS are colored in black. Non-cytotoxic chemicals and cytotoxic hits not validated by MTS are colored in gray. e, Table showing characteristics of 49 oligodendrocyte-specific cytotoxic hits tested in 10-point dose response from (40 nM to 20 μM). IC50 values were determined with curve-fitting and compared to median cytotoxicity values obtained from the EPA database for each chemical. Potency scores were calculated by dividing the cytotoxicity median by the experimentally determined IC50 in oligodendrocytes. Chemicals were ranked based on increasing potency score. Table also includes each chemical’s use category. f, Chemotype analysis for the 49 oligodendrocyte-specific cytotoxic compounds, with the most enriched structural domain, bond.quatN_alkyl_acyclic (p-value = 0.002, OR = 16.2), highlighted in red. p-values were generated using a one-sided Fisher’s exact test. g, Chemical structures for the four quaternary ammonium compounds and one quaternary phosphonium compound. The enriched cytotoxicity-associated bond for quaternary ammonium compounds is highlighted in red. The quaternary phosphonium bond is highlighted in orange.

Fig. 2:. Quaternary compounds activate the integrated stress response and are cytotoxic to human oligodendrocytes.

a, Gene set enrichment analysis (GSEA) of hallmark gene sets upregulated in OPCs in response to incubation with 20 µM quaternary ammonium (red) or phosphonium compounds (orange) for 4 hours. b, GSEA of an integrated stress response gene set in OPCs treated with methyltrioctylammonium chloride or tributyltetradecylphosphonium chloride compared with DMSO treated OPCs demonstrates significant enrichment (FDR<0.001) for genes involved in the integrated stress response (normalized enrichment scores [NES] = 1.91, 1.88 respectively). c, qRT-PCR of CHOP in OPCs treated with DMSO (gray), methyltrioctylammonium chloride (red), and tributyltetradecylphosphonium chloride (orange). Data are presented as the mean value ± standard deviation from three biological replicates, represented as closed circles. P values were calculated using one-way ANOVA with Dunnett post-test correction for multiple comparisons. d, Schematic depicting exposure of human cortical organoids to cytotoxic chemicals. e, Representative immunohistochemistry images of control human cortical organoids and organoids treated for 10 days with 360 nM methyltrioctylammonium chloride or 300 nM tributyltetradecylphosphonium chloride (approximate IC90 in mPSC-derived oligodendrocytes). Images show all cells (DAPI+, blue) and oligodendrocytes (SOX10+, in green) at day 70. f, Quantification of total cell number (DAPI+ per mm2) and oligodendrocytes (SOX10+ per mm2) in the whole cortical organoid. Data are presented as the mean value ± standard deviation from n ≥ 5 biological replicates (individual organoids) indicated by closed circle data points. p-values were calculated using one-way ANOVA with Dunnett post-test correction for multiple comparisons.

Fig. 3:. Organophosphate flame retardants arrest oligodendrocyte maturation.

a, Primary chemical screen showing the effect of 1,539 non-cytotoxic environmental chemicals on oligodendrocyte development displayed as percent O1+ cells normalized to DMSO. The dotted lines mark ± 3 standard deviations from the mean of control wells. The blue dotted line marks the inhibitor hit cutoff, a reduction in O1+ cells of 50% (>7 standard deviations) compared to DMSO. The 49 oligodendrocyte inhibitors that pass this threshold are colored in blue. All other non-cytotoxic chemicals that did not inhibit oligodendrocyte development are colored in gray. b, Chemotype analysis for oligodendrocyte inhibitors showing both the p-value and odds ratio. Among the top most significant structural domains, bond.P.O_phosphate_alkyl_ester (p-value = 0.02 , OR = 12.5) has the highest odds ratio, and is highlighted in yellow. p-values were generated using a one-sided Fisher’s exact test. c, Chemical structures for three organophosphate flame retardants containing the structure bond.P.O_phosphate_alkyl_ester, highlighted in yellow. d, Graph of eight-point dose response (30 nM to 20 µM) quantifying the effect of three organophosphate flame retardants on oligodendrocyte (O1+) generation from OPCs. Data are presented as the mean value ± standard deviation from three biological replicates (OPC batches generated from independent mPSC lines). e, Schematic showing stages of in vitro oligodendrocyte development and the markers for early (O4), intermediate (O1), and late (MBP) oligodendrocytes. f, Representative images of early (O4+, in magenta), intermediate (O1+, in green), and late (MBP+ in yellow) oligodendrocytes after treatment with DMSO vehicle control or 20 µM TDCIPP for 1, 2, and 3 days of maturation. Nuclei are marked using DAPI (in blue). Images for oligodendrocytes treated with TMPP and TBPP are shown in Extended Data Fig. 4e. g, Quantification of early (O4+), intermediate (O1+), and late (MBP+) oligodendrocytes, after day 1, 2, and 3 of development, normalized to DMSO vehicle control. Data are presented as the mean value ± standard deviation from three biological replicates (OPC batches generated from independent mPSC lines), indicated by closed circle data points. Data for oligodendrocytes treated with TBPP and TMPP are shown in Extended Data Fig. 4f. p-values were calculated using one-way ANOVA with Dunnett post-test correction for multiple comparisons.

Fig. 4:. TDCIPP inhibits human oligodendrocyte development and is associated with abnormal neurodevelopmental outcomes in children.

a, Representative immunohistochemistry images of human cortical organoids treated for 10 days with the flame retardant TDCIPP at 18 µM (approximate IC75 in mPSC-derived oligodendrocytes). Images show all cells (DAPI+, in blue), oligodendrocyte lineage cells (SOX10+, in green), and oligodendrocytes (CC1+, in magenta). b, Quantification of total cell number (DAPI+ per mm2), oligodendrocyte lineage cells (SOX10+ per mm2) and mature oligodendrocytes (SOX10+CC1+ per mm2) in whole cortical organoids. Data are presented as the mean value ± standard deviation from n ≥ 8 biological replicates (individual organoids) indicated by closed circle data points for TDCIPP-treated organoids. p-values were calculated using Student’s two-tailed t test. c, Pie chart showing the number of children ages 3–17 years old from the NHANES 2017–2018 dataset with undetectable and detectable levels of BDCIPP, the urine metabolite of TDCIPP. d, Density plot showing the range and quintiles of urine BDCIPP levels in children ages 3–17 years old from the NHANES 2017–2018 dataset. e, Boxplot showing creatinine-normalized levels of BDCIPP in children 3–17 years of age and adults aged 18 years and older. p-value was calculated using the Kruskal Wallis one-way ANOVA. f, Adjusted odds ratio for the neurodevelopmental outcome: requiring special education or early intervention. Significant odds ratios are highlighted in yellow (BDCIPP Q5 v Q1 OR = 2.7 [95% CI = 1.012–7.407) and gray (Female v Male OR = 0.376 [95% CI = 0.228–0.621]). g, Adjusted odds ratio for the neurodevelopmental outcome: gross motor limitations. Significant odds ratios are highlighted in yellow (BDCIPP Q5 v Q1 OR = 6.0 [95% CI = 1.243–29.426).