Research ArticleReproductive Biology

Temperature-activated ion channels in neural crest cells confer maternal fever–associated birth defects

See allHide authors and affiliations

Sci. Signal.  10 Oct 2017:
Vol. 10, Issue 500, eaal4055
DOI: 10.1126/scisignal.aal4055
  • Fig. 1 Hyperthermia-induced congenital defects in neural crest–dependent tissues.

    (A) Alcian blue– and alizarin red–stained craniofacial features of HH36 heads for control and hyperthermia-exposed chicks. Upper beak measurement extended from the quadratojugal (white arrowhead) to the tip of the upper beak (black arrowhead). (B) Upper beak length was normalized to femur lengths for control (black) and hyperthermia-exposed (red) chicks. (C) Graph of upper beak length–to–femur length ratios for control and hyperthermia chicks. (D) Whole-mount and histological sections of HH36 hearts from control (normothermia) and hyperthermia-exposed embryos. The white and black arrowheads in the whole-mount images highlight the alignment of the aorta (Ao) and pulmonary trunk (P) in the control heart compared to the hyperthermia-exposed heart. Histological sections of the whole-mount hearts at the level of the ventricular septum and the outflow vessel semilunar valves. Dashed lines indicate the plane of outflow tract septation. Sections through the hyperthermia-exposed DORV heart with a perimembranous VSD (*) and a rightward shift of the aorta in relation to the pulmonary trunk (dashed line). Gross and histological analysis of heart anatomy was performed in 60 normothermia hearts and 49 hyperthermia-exposed hearts. TV, tricuspid valve; MV, mitral valve. (E) Percentage of conotruncal defects in control (0%; n = 60 hearts) and hyperthermia groups (12%; n = 6 of 49 hearts). (F) Representative histological sections through the aorta and the pulmonary trunk used to compare the luminal areas (arrows) of normothermic and hyperthermia-exposed hearts distal to the valves at the level of the left coronary artery (white arrowhead). Scale bars, 200 μm. (G) Average Cavalieri probe estimates of the luminal cross-sectional areas of the aorta and pulmonary trunk immediately distal to the respective semilunar valves. Significance was determined using unpaired t test (C and G) or Fisher’s exact test (E). *P < 0.0001. The number of biological replicates is indicated by n in the graphs in (C), (E), and (G).

  • Fig. 2 Temperature-activated TRPV channels in avian and mammalian neural crest cells.

    (A) Images of chick primary explanted neural tubes (NT) expressing GCaMP6 after 24 hours in culture. The dashed box region shows migrating neural crest cells used in subsequent analyses (green arrowheads). (B) Change in GCaMP6 fluorescence in chick neural crest cells after exposure to 1 μM GSK101 (blue) or 1 μM GSK101 combined with 10 μM RN1734 (gray line). (C) GCaMP6 fluorescence in chick neural crest cells after exposure to 1 μM VaTx3 (arrow, blue line) or buffer alone (gray line). (D) Change in Fluo-4 fluorescence in mouse primary explanted neural crest cells from embryonic day 8.5 (E8.5) embryos in response to the indicated doses of GSK101. (E) Fluo-4 fluorescence in mouse primary explanted neural crest cells from E8.5 embryos in response to GSK101 (arrow, blue line) or to GSK101 and 10 μM RN1734 (gray line). (F) Changes in Fluo-4 fluorescence in mouse neural crest cells after exposure to 1 μM capsaicin (arrow, blue line) or to capsaicin and SB366791 (gray line). (G) Representative GCaMP6 fluorescence in chick primary neural crest cells after exposure to 40°C imaging buffer, followed by imaging buffer (pH 5). (H) Representative GCaMP6 fluorescence in chick primary neural crest cells after exposure to 40°C imaging buffer or GSK101 (blue) or in the presence of RN1734 inhibitor (gray line). (I) Averages of more than four separate experiments as in (H) analyzing the number of cells per condition, as indicated. Significance was determined using unpaired t test. *P < 0.05. The number of biological replicates is indicated by n, and/or the number of cells analyzed is indicated in the graphs in (B) to (F), (H), and (I).

  • Fig. 3 TRPV1 inhibition rescues hyperthermia-induced congenital defects.

    (A) Alcian blue and alizarin red stains of HH36 heads from hyperthermia-exposed and hyperthermia-exposed chicks after pretreatment with the TRPV1 inhibitor SB366791. Upper beak measurement extended from the quadratojugal (white arrowhead) to the tip of the upper beak (black arrowhead). (B) Comparison of upper beak length–to–femur length ratios in control, hyperthermia, or hyperthermia + SB366791 chicks. (C) Histological sections through the aorta (Ao) and the pulmonary trunk (P) comparing the luminal areas (arrows) of control hearts (top), hyperthermia-exposed hearts (middle), and hyperthermia-exposed hearts with SB366791 pretreatment (lower) distal to the semilunar valves at the level of the left coronary artery (white arrowhead). Scale bars, 200 μm. (D and E) Cavalieri probe estimates of the luminal cross-sectional area of the aorta (D) or the pulmonary trunk (E) immediately above the aortic valve. Significance was determined by one-way analysis of variance (ANOVA), followed by Bonferroni’s multiple comparisons test. *P < 0.05, **P < 0.004, and ***P < 0.0001. The number of biological replicates is indicated by n in the graphs in (B), (D), and (E).

  • Fig. 4 Ligand activation of TRPV4 replicates hyperthermia-induced birth defects.

    (A) Whole mounts of hearts from control and GSK101-treated embryos. The GSK101-treated heart showed DORV orientation of the aorta (Ao) and the pulmonary trunk (P). See fig. S9 for histological sections. Whole-mount and histological analysis of heart anatomy was performed in 38 hearts from DMSO-treated embryos and 15 GSK101-treated embryos. (B) Percentage of hearts with histologically confirmed conotruncal defects in GSK101-treated embryos compared to dimethyl sulfoxide (DMSO) controls. (C) Panel of histological sections through the aorta (Ao) and the pulmonary trunk (P) comparing the luminal areas (white arrows) at the level of the left coronary artery (*) in control, DMSO, and 20 or 50 μM GSK101 treatment. Scale bar, 200 μm. (D and E) Treatment with GSK101 reduced aortic luminal areas (D) [average coefficient of error (CE), 0.03] and pulmonary trunk luminal areas (E) (average CE, 0.03). (F) Alcian blue and alizarin red stains of DMSO control (top) and 20 μM GSK101-treated embryos (bottom) at HH36. (G) Normalization of upper beak length to femur lengths. (H) Graph of upper beak–to–femur ratios in untreated control, DMSO-treated control, and two GSK101 treatment groups. Significance was determined using Fisher’s exact test (B) or one-way ANOVA, followed by Bonferroni’s multiple comparisons test (D, E, and H). *P < 0.03, **P < 0.004, and ***P < 0.0001; NS, not significant. The number of biological replicates is indicated by n in the graphs in (D), (E), and (H).

  • Fig. 5 TRPV4 activation disrupts jaw extension in zebrafish larvae.

    (A) Activity of GSK101 at the indicated concentrations and of GSK1153218 (black tracing), a structurally related compound that lacks activity on mammalian TRPV4 (table S1), in cloned zebrafish TRPV4 in CHO cells using GCaMP6 to assess Ca2+ permeability. (B) Representative images of 1.4cola1:egfp transgenic zebrafish larvae treated with 20 μM GSK101, vehicle control, or GSK1153218A. Replicate batches were imaged live at 3 and 4 dpf. Scale bar, ~130 μm. (C and D) Quantification of the distance between Meckel’s cartilage and the ceratohyal was measured [red line in (B)]. Significance was determined using one-way ANOVA, followed by Bonferroni’s multiple comparisons test. *P < 0.0001 (GSK101-treated group compared to all other groups). The number of biological replicates is indicated by n, and/or the number of cells analyzed is indicated in the graphs in (A), (C), and (D).

  • Fig. 6 Development of remotely controlled TRPVFeRIC channels.

    (A) TRPV channels were tagged with D5 of kininogen-1 and were cloned into the PLVX vector with an internal ribosomal entry site (IRES) for mCherry (figs. S12 and S13). (B) FeRIC channels were designed to recruit endogenous cellular ferritin to the modified TRPV channel at the cell membrane. (C) Cytoplasmic distribution of ferritin heavy chain fused with mCherry (FTH1mCherry) in HEK293T cells expressing TRPV1WT and membrane redistribution of FTH1mCherry (white arrowheads) in TRPV1FeRIC-expressing cells. Images were representative of two independent experiments. (D) Representative immunoprecipitation (IP) and Western blot of HEK293T cells expressing FLAG-tagged TRPV1WT, TRPV1FeRIC, TRPV4WT, or TRPV4FeRIC. The blot was probed for FLAG and FTH1. Four independent experiments were conducted using TRPV1FeRIC, and three independent experiments were conducted using TRPV4FeRIC. (E) GCaMP6 fluorescence in TRPV1WT-expressing (blue) or TRPV1FeRIC-expressing (red) HEK293T cells after RF (gray box) and then 1 μM capsaicin (bar). Bar graphs are ΔF/F0 averages of four experiments with 50 to 100 cells per group analyzed. (F) GCaMP6 fluorescence in TRPV4WT-expressing (blue) or TRPV4FeRIC-expressing (red) HEK293T cells after RF (gray box) and then 1 μM GSK101 (bar). Bar graphs are ΔF/F0 averages of five experiments with 106 to 123 cells per group analyzed. Significance was determined using unpaired t test. *P < 0.05.

  • Fig. 7 Fever-associated heart defects after remote activation of TRPV1 in neural crest cells.

    (A) GCaMP6 fluorescence in primary chick neural crest cells electroporated with TRPV1WT (blue lines) or TRPV1FeRIC (red lines) in the absence or presence of the TRPV1 inhibitor SB366971 (gray and black lines) after RF (gray box) and then 1 μM capsaicin (bar). Bar graph shows cumulative responses representing three to four separate experiments with 22 to 138 cells per group analyzed. (B) mCherry+ neural crest streams migrating to the pharyngeal arches (white arrows) in HH14 control or TRPV1FeRIC electroporated embryos. Images are representative of six independent experiments. (C) Whole-mount hearts from RFNegTRPV1FeRIC embryo and RF10minTRPV1FeRIC embryo with arrows noting the position of the aorta (Ao) and the pulmonary trunk (P) with respect to the right ventricle (RV). Five independent experiments were performed. See movie S1 for an MRI reconstruction of an RF10minTRPV1FeRIC-induced DORV heart defect in chick. (D) Percentage of embryos with histologically confirmed conotruncal heart defects within indicated groups at HH36. (E) Representative histological sections through the aorta and the pulmonary trunk at the level of the coronary artery (*) in control RF, RF10minTRPV1WT, RF10minTRPV1FeRIC, and RF10minTRPVFeRIC with SB366791 pretreatment. The double arrowheads highlight the luminal areas of the aorta and the pulmonary trunk estimated using the Cavalieri probe. Scale bar, 200 μm. (F) Graph of the Cavalieri probe estimates of cross-sectional areas through the aorta at the level of the coronary artery in the indicated treatment groups (CE, 0.03). (G) Graph of the Cavalieri probe estimates of the cross-sectional areas through the pulmonary trunk distal to the semilunar valve in the indicated treatment groups (CE, 0.03). Significance was determined using unpaired t test (A), Fisher’s exact test (D), or one-way ANOVA, followed by Bonferroni’s multiple comparisons test (F and G). *P < 0.02, **P < 0.005, and ***P < 0.0001. The number of biological replicates is indicated by n in the graphs in (F) and (G).

  • Fig. 8 Fever-associated heart defects after remote activation of TRPV4 in neural crest cells.

    (A) GCaMP6 fluorescence in chick neural crest cells electroporated with TRPV4WT (blue lines) and TRPV4FeRIC (red lines) after RF (gray box) and then GSK101 (bar). TRPV4 inhibitor RN1734 inhibits response (dashed lines). Bar graph shows cumulative responses representing four separate experiments with 45 to 49 cells per group analyzed. (B) Whole-mount and histological sections of HH63 hearts from RFNegTRPV4FeRIC embryo and RF10minTRPV4FeRIC embryo with DORV and a persistent L4 arch artery. The white and black arrowheads in the whole-mount images highlight the alignment of the aorta (Ao) and the pulmonary trunk (P) in the RFNegTRPV4FeRIC and RF10minTRPV4FeRIC hearts. Histological sections of the whole mount in RFnegTRPV4FeRIC and RF10minTRPV4FeRIC hearts (above) at the level of the ventricular septum (IVS), at the level of the semilunar valves of the aorta (coronary arteries; black arrowheads) and the pulmonary trunk, and more distally through the smooth muscle walls of the aorta and the pulmonary trunk at the level of the left coronary artery (*). The dashed line indicates the plane of outflow tract septation. Black arrows indicate a VSD and persistent L4 arch artery in the section through the RF10minTRPV4FeRIC heart. Double-headed arrows indicate the luminal cross-sectional areas of the aorta and the pulmonary trunk measured in the Cavalieri estimates in (D) and (E). Scale bars, 200 μm. Three separate experiments were performed, and a total of nine RFNegTRPV4FeRIC and seven RF10minTRPV4FeRIC hearts were analyzed. av, aortic vestibule; pi, pulmonary infundibulum; RA, right atrium; LA, left atrium. (C) Percentage of histologically confirmed conotruncal defects in RFNegTRPV4FeRIC embryos compared to RF10minTRPV4FeRIC embryos. (D) Graph of the Cavalieri probe estimates of cross-sectional areas through the aorta at the level of the coronary artery in the indicated treatment groups (CE, 0.05). (E) Graph of the Cavalieri probe estimates of the cross-sectional areas through the pulmonary trunk distal to the semilunar valve in the indicated treatment groups (CE, 0.05). Significance was determined using unpaired t test (A), Fisher’s exact test (C), or one-way ANOVA, followed by Bonferroni’s multiple comparisons test (D and E). *P < 0.02, **P < 0.005, and ***P < 0.0001. The number of biological replicates is indicated by n in the graphs in (C) to (E).

Supplementary Materials

  • www.sciencesignaling.org/cgi/content/full/10/500/eaal4055/DC1

    Fig. S1. Hyperthermia causes severe craniofacial defects in chicks.

    Fig. S2. Hyperthermia causes aortic arch patterning defects in chicks.

    Fig. S3. TRPV channel mRNA expression in neural crest cells.

    Fig. S4. TRPV1 transcripts are detected in cranial and cardiac neural crest cells by in situ hybridization.

    Fig. S5. TRPV4 transcripts are detected in cranial and cardiac neural crest cells by in situ hybridization.

    Fig. S6. Activation of cloned chick TRPV4 and chick TRPV1 with channel-specific ligands.

    Fig. S7. Neural crest cell–related congenital defects in chicks upon TRPV4 inhibition.

    Fig. S8. TRPV1 inhibition partially rescues hyperthermia-induced conotruncal defects.

    Fig. S9. Pharmacological activation of TRPV4 induces fever-related heart defects.

    Fig. S10. TRPV4 activation disrupts palatogenesis in zebrafish larvae.

    Fig. S11. TRPV1 maps.

    Fig. S12. TRPV4 maps.

    Fig. S13. Expression of FeRIC-modified TRPV1 does not alter cellular Fe2+ homeostasis.

    Fig. S14. RF coil and SAR estimation.

    Fig. S15. Membrane integrity and cell viability are unchanged in TRPV1FeRIC-expressing cells after RF stimulation.

    Fig. S16. Remote activation of FeRIC is ferritin-dependent.

    Fig. S17. A temperature-insensitive TRPV1 mutant does not respond to RF.

    Fig. S18. Histological analysis of HH36 hearts confirms conotruncal defects after transient activation of TRPV1FeRIC by RF.

    Fig. S19. Fever-associated craniofacial defects after remote activation of TRPV1 in neural crest cells.

    Fig. S20. RF does not cause bulk tissue temperature changes in ovo.

    Table S1. Structural and functional comparison of GSK101 and GSK1153218 (GSK-dead).

    Movie S1. MRI reconstruction of a TRPV1FeRIC-induced DORV heart defect in chick.

    References (7379)

  • Supplementary Materials for:

    Temperature-activated ion channels in neural crest cells confer maternal fever–associated birth defects

    Mary R. Hutson, Anna L. Keyte, Miriam Hernández-Morales, Eric Gibbs, Zachary A. Kupchinsky, Ioannis Argyridis, Kyle N. Erwin, Kelly Pegram, Margaret Kneifel, Paul B. Rosenberg, Pavle Matak, Luke Xie, Jörg Grandl, Erica E. Davis, Nicholas Katsanis, Chunlei Liu,* Eric J. Benner*

    *Corresponding author. Email: eric.benner{at}duke.edu (E.J.B); chunlei.liu{at}berkeley.edu (C.L.)

    This PDF file includes:

    • Fig. S1. Hyperthermia causes severe craniofacial defects in chicks.
    • Fig. S2. Hyperthermia causes aortic arch patterning defects in chicks.
    • Fig. S3. TRPV channel mRNA expression in neural crest cells.
    • Fig. S4. TRPV1 transcripts are detected in cranial and cardiac neural crest cells by in situ hybridization.
    • Fig. S5. TRPV4 transcripts are detected in cranial and cardiac neural crest cells by in situ hybridization.
    • Fig. S6. Activation of cloned chick TRPV4 and chick TRPV1 with channel-specific ligands.
    • Fig. S7. Neural crest cell–related congenital defects in chicks upon TRPV4 inhibition.
    • Fig. S8. TRPV1 inhibition partially rescues hyperthermia-induced conotruncal defects.
    • Fig. S9. Pharmacological activation of TRPV4 induces fever-related heart defects.
    • Fig. S10. TRPV4 activation disrupts palatogenesis in zebrafish larvae.
    • Fig. S11. TRPV1 maps.
    • Fig. S12. TRPV4 maps.
    • Fig. S13. Expression of FeRIC-modified TRPV1 does not alter cellular Fe2+ homeostasis.
    • Fig. S14. RF coil and SAR estimation.
    • Fig. S15. Membrane integrity and cell viability are unchanged in TRPV1FeRIC-expressing cells after RF stimulation.
    • Fig. S16. Remote activation of FeRIC is ferritin-dependent.
    • Fig. S17. A temperature-insensitive TRPV1 mutant does not respond to RF.
    • Fig. S18. Histological analysis of HH36 hearts confirms conotruncal defects after transient activation of TRPV1FeRIC by RF.
    • Fig. S19. Fever-associated craniofacial defects after remote activation of TRPV1 in neural crest cells.
    • Fig. S20. RF does not cause bulk tissue temperature changes in ovo.
    • Table S1. Structural and functional comparison of GSK101 and GSK1153218 (GSK-dead).
    • Legend for movie S1
    • References (7379)

    [Download PDF]

    Technical Details

    Format: Adobe Acrobat PDF

    Size: 2.41 MB

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.mov format). MRI reconstruction of a TRPV1FeRIC-induced DORV heart defect in chick.

    [Download Movie S1]


    Citation: M. R. Hutson, A. L. Keyte, M. Hernández-Morales, E. Gibbs, Z. A. Kupchinsky, I. Argyridis, K. N. Erwin, K. Pegram, M. Kneifel, P. B. Rosenberg, P. Matak, L. Xie, J. Grandl, E. E. Davis, N. Katsanis, C. Liu, E. J. Benner, Temperature-activated ion channels in neural crest cells confer maternal fever–associated birth defects. Sci. Signal. 10, eaal4055 (2017).

    © 2017 American Association for the Advancement of Science

Navigate This Article