Formation of neurodegenerative aggresome and ...

Formation of neurodegenerative aggresome and death-inducing signaling complex in maternal diabetes-induced neural tube defects Zhiyong Zhaoa,b,1,2, Lixue Caoa,1, and E. Albert Reecea,b a Department of Obstetrics, Gynecology, and Reproductive Sciences, University of Maryland School of Medicine, Baltimore, MD 21201; and bDepartment of Biochemistry and Molecular Biology, University of Maryland School of Medicine, Baltimore, MD 21201

Diabetes mellitus in early pregnancy increases the risk in infants of birth defects, such as neural tube defects (NTDs), known as diabetic embryopathy. NTDs are associated with hyperglycemia-induced protein misfolding and Caspase-8–induced programmed cell death. The present study shows that misfolded proteins are ubiquitinylated, suggesting that ubiquitin-proteasomal degradation is impaired. Misfolded proteins form aggregates containing ubiquitin-binding protein p62, suggesting that autophagic-lysosomal clearance is insufficient. Additionally, these aggregates contain the neurodegenerative disease-associated proteins α-Synuclein, Parkin, and Huntingtin (Htt). Aggregation of Htt may lead to formation of a death-inducing signaling complex of Hip1, Hippi, and Caspase-8. Treatment with chemical chaperones, such as sodium 4-phenylbutyrate (PBA), reduces protein aggregation in neural stem cells in vitro and in embryos in vivo. Furthermore, treatment with PBA in vivo decreases NTD rate in the embryos of diabetic mice, as well as Caspase-8 activation and cell death. Enhancing protein folding could be a potential interventional approach to preventing embryonic malformations in diabetic pregnancies. diabetic embryopathy chemical chaperone

| protein folding | protein aggregation | Caspase-8 |

P

regnancies complicated by diabetes mellitus have a higher risk of resulting in infants with congenital birth defects, a complication known as diabetic embryopathy (1). The defects in the central nervous system are the result of perturbed neural tube formation (neurulation) during early embryogenesis and are thus referred to as neural tube defects (NTDs) (2). Neurulation is a process that begins with development of the neural folds at about week 2 of gestation in humans and embryonic day 8 (E8) in the mouse, and concludes with fusion of the neural tissues at the dorsal midline, to form the neural tube, by about week 4 in humans and E11.5 in the mouse (3, 4). The failure of neural tube closure in diabetic embryopathy is associated with excessive programmed cell death (apoptosis) in the neural epithelium (1). Diabetes-induced apoptosis is mediated by Caspase-8 (5), which is activated by a death-inducing signaling complex (DISC) (6, 7). Maternal hyperglycemia, the major teratogen of diabetesinduced embryonic malformations, exerts profound effects on intracellular metabolic homeostasis and organelle function (1). Perturbation of rough endoplasmic reticulum (ER) function has been implicated in diabetic embryopathy by the observations that factors involved in the unfolded protein response are up-regulated (8, 9), due to accumulation of improperly folded proteins in the ER lumen (10). Misfolded proteins are usually degraded through the ubiquitin-proteasome system (UPS), including the ER-associated protein degradation pathway (11–13). However, if the UPS is impaired, misfolded proteins escape degradation and are released into the cytosol (14). Accumulation of misfolded cytotoxic proteins within a cell perturbs organelle functions, such as inducing the mitochondria to generate an abundance of reactive oxygen species www.pnas.org/cgi/doi/10.1073/pnas.1616119114

(ROS), which results in oxidative stress (15, 16) and disrupts intracellular signaling, thus altering cellular activities (17). Misfolded proteins are prone to forming aggregates that are resistant to UPS degradation (18). Aggregation of α-Synuclein (α-Syn), Parkin, and Huntingtin (Htt) has been found to be involved in the pathogenesis of Alzheimer’s, Parkinson’s, and Huntington’s diseases (19–22). Protein aggregates can be eliminated by the autophagic-lysosomal system (ALS) (23, 24). The transfer of ubiquitinylated proteins to the ALS is mediated by a ubiquitin-binding protein, p62, also known as Sequestosome-1 (25, 26). Protein aggregates containing p62 are commonly seen in neurodegenerative diseases, indicating deficiency of the ALS in the pathogenesis of these diseases (27). These neural degenerative disease-associated proteins are expressed in the developing neural tube (28–30); however, it is unknown whether they are involved in hyperglycemia-induced NTDs. In the present study, we aimed to investigate the fate of these misfolded proteins and their potential effects on apoptotic regulation and NTD formation. We also explored an approach to alleviate protein aggregation by enhancing protein folding using chemical chaperones to provide insights into the mechanisms of diabetic embryopathy, and to help identify potential strategies to prevent diabetes-induced birth defects. Results Protein Ubiquitinylation and Aggregation. Misfolded proteins are

usually ubiquitinylated and degraded by the UPS (13). However, when the UPS is impaired, ubiquitinylated proteins can accumulate in the cell. Because we hypothesized that UPS dysfunction Significance Neural tube defects in infants of women with diabetes are associated with disrupted protein folding and increased programmed cell death in embryonic cells. Misfolded proteins are ubiquitinylated and form aggregates that contain the neurodegenerative disease-associated proteins α-Synuclein, Parkin, and Huntingtin (Htt). Aggregation of Htt leads to formation of a Hip1–Hippi–Caspase-8 complex to activate Caspase-8. Treatment with chemical chaperones alleviates protein aggregation and decreases neural tube defects in the embryos of diabetic mice. Targeting misfolded proteins could be a potential approach to preventing birth defects in diabetic pregnancies. Author contributions: Z.Z. and E.A.R. designed research; Z.Z. and L.C. performed research; L.C. analyzed data; and Z.Z. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. T.M.D. is a Guest Editor invited by the Editorial Board. 1

Z.Z. and L.C. contributed equally to this work.

2

To whom correspondence should be addressed. Email: [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1616119114/-/DCSupplemental.

PNAS Early Edition | 1 of 6

MEDICAL SCIENCES

Edited by Ted M. Dawson, Institute for Cell Engineering, Johns Hopkins University School of Medicine, Baltimore, MD, and accepted by Editorial Board Member Gregg L. Semenza March 13, 2017 (received for review September 27, 2016)

6 4 2 0

β-actin

C

F

J

M

D

G

K

N

validated to specifically detect protein aggregates and aggresomelike inclusion bodies in cells and cell lysates (31). In the tissue sections of the dorsal brain (SI Appendix, Fig. S1), more intense Proteostat signals were observed in the DM group than in the ND group (Fig. 1 C and F). Quantification of the levels of protein aggregation showed significant increases in the DM group, compared with those in the ND group (Fig. 1I). Aggregates of ubiquitinylated proteins are subject to ALS clearance, mediated by ubiquitin-binding protein p62 (25, 26). Doublelabeling with Proteostat to detect protein aggregates and an antibody to detect p62 revealed colocalization of protein aggregates and p62 (Fig. 1 F–H). To determine whether high glucose induces the protein aggregation observed in embryos of diabetic dams, we used neural stem cells, derived from E9 mouse neural tubes, as a model system (32). The cells were cultured in a concentration of 27.8 mM (500 mg/dL) glucose [high glucose (HG)] or 6 mM (100 mg/dL) glucose [normal glucose (NG)] and assessed for protein aggregation. Significant increases in protein aggregation and colocalization with p62 were observed in the cells of the HG group, compared with those in the cells of the NG group (Fig. 1 J–P).

8

ND

DM

E

H

L

O

I 150 % (Aggregates/total protein)

B

100

50

0

ND

DM

P 40 % (Positive/total cells)

DM

Ub-protein/β-actin

ND

Ub-proteins

A

30 20 10 0

NG

HG

Fig. 1. Protein ubiquitinylation and aggregation in cells of the embryonic neural tube of diabetic mice and cultured neural stem cells treated with high glucose. (A) Western blot of ubiquitinylated (Ub) proteins in the dorsal region of the brain of E9.5 embryos, using an anti–Ub-protein antibody. (B) Quantification of the Western blot bands of the whole lane. β-Actin, loading control. Data are presented as mean ± SD, *P = 0.005, n = 3 embryos. (C, F, J, and M) Proteostat detection (red). (D, G, K, and N) p62 immunofluorescence (green). (E, H, L, and O) Merged images of Proteostat, p62, and DAPI. (C–H) Dorsal brain regions of E9.5 embryos. (Inset in C) Diagrammatic representation of a crosssection of the neural tube. The box marks the proximal area shown in C–H. (C–E) Nondiabetic (ND). (F–H) Diabetes mellitus (DM). Arrows indicate colocalization of Proteostat and p62. (I) Levels of protein aggregation in neural tissues, assessed by measuring Proteostat fluorescence intensity. Data are presented as mean ± SD, *P = 0.003, n = 6 embryos (three repeats). (J–O) Neural stem cells. (J, K, and L) Normal glucose (NG). (M–O) High glucose (HG). (P) Quantification of neural stem cells with protein aggregates. Data are presented as mean ± SD, *P = 0.003, n = 4 repeats. (Scale bar = 10 μm for all images.)

plays a role in diabetes-induced NTDs, we generated diabetic pregnant mice (10 wk old) and collected neural tissues of embryos at E9.5, an important stage in murine neurulation (3, 4), to quantify changes in levels of ubiquitinylated proteins. In the dorsal brain regions (SI Appendix, Fig. S1), where cranial NTDs occur, the levels of global protein ubiquitinylation were significantly higher in the diabetes mellitus (DM) group, compared with the nondiabetic (ND) group (Fig. 1 A and B). Accumulation of misfolded proteins has been implicated in diabetic embryopathy (8, 9), but the fate of these misfolded proteins, which can aggregate and become cytotoxic (18), has not been previously explored. We used a Proteostat system to examine protein aggregates in the neural tube of embryos at E9.5 and E10.5. Proteostat is a rotor molecule that emits fluorescence when it binds to the tertiary structure of aggregated proteins. It has been 2 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1616119114

Aggregation of Neurodegenerative Disease-Associated Proteins.

Aggregation of α-Syn, Parkin, and Htt is seen in Alzheimer’s, Parkinson’s, and Huntington’s diseases (19). Because these proteins are also expressed in the neural epithelium of the developing embryo (SI Appendix, Fig. S2) (28–30), we sought to determine whether these neurodegenerative disease-associated proteins also aggregate in the neural tube of embryos from diabetic mice. We used a proximity ligation assay (PLA) to examine interactions between p62 and α-Syn, Parkin, and Htt in situ on tissue sections. When two proteins physically interact, the antibodies that recognize them are in a close proximity so that conjugated oligonucleotides can serve as guides to ligate additional oligonucleotides to form a circular template for DNA synthesis. The DNA amplified from the template is then detected using a fluorescent dye-labeled probe (33). PLA signals in the dorsal region of the neural tube at E9.5 in the DM group were significantly higher than those in the ND group (Fig. 2 A–I), as well as in the neural stem cells cultured in high glucose (SI Appendix, Fig. S3). Such protein–protein interactions were also detected in high glucose-treated neural stem cells, using a coimmunoprecipitation assay (Fig. 2 J and K). Formation of Death-Inducing Signaling Complex. High levels of apoptosis in NTDs in diabetic embryopathy are mediated by Caspase-8 (34). Caspase-8 can be activated by a DISC. One such DISC contains Htt-interacting protein 1 (Hip1) and Hip1 protein interactor (Hippi), which forms when Htt aggregates to release Hip1 (35, 36). We investigated if Hip1, Hippi, and Caspase-8 form this unique DISC in embryos from diabetic dams. Interactions between these proteins were detected, using PLA, in the dorsal region of the neural tube in E9.5 embryos. The levels of these protein complexes were significantly higher in the DM group than in the ND group (Fig. 3). Effects of Chemical Chaperones on Protein Aggregation in Vitro and in Vivo. To examine if protein aggregation in diabetic embryopathy

is the result of aberrant protein folding, we treated embryonic neural stem cells with chemical chaperones, including sodium 4-phenylbutyrate (PBA), sodium taurochenodeoxycholate (TUDCA), and trimethylamine N-oxide (TMAO). The treatments significantly decreased protein aggregation in neural stem cells cultured in high glucose (HG+PBA: 25 μM and 50 μM; HG+TUDCA: 25 μM and 50 μM; HG+TMAO: 100 μM and 250 μM), compared with the HG+VEH group (Fig. 4 A–C). To investigate whether chemical chaperones can alleviate the hyperglycemia-induced protein-folding crisis, and can reduce diabetes-induced embryonic malformations in vivo, we used PBA to treat diabetic pregnant mice (100 mg/kg, daily) from E6.5 to Zhao et al.

DM

B

C5 J

Dots/cell

p62-α-Syn

4 3

E

F

70 kDa

p62

2 50 kDa

1 0

D

Input (5%) IgG α-Syn

ND

A

ND

DM

1.0

0.6 0.4

K

Htt

p62-Parkin

Dots/cell

0.8

0.2 0.0

G

H

I

70 kDa

ND DM

2.5

p62 50 kDa

1.5

MEDICAL SCIENCES

Dots/cell

p62-Htt

2.0

1.0 0.5 0.0

ND

DM

Fig. 2. Interactions of p62 with neurodegenerative proteins. (A–I) PLA to detect interaction between p62 and α-Syn (A and B), Parkin (D and E), or Htt (G and H) in the dorsal brain regions of E9.5 embryos. PLA signals are red dots. DAPI counterstaining is blue. (Inset in A) Diagrammatic representation of a cross-section of the neural tube. The box marks the proximal area shown in the figures. (C, D, and I) Quantification of PLA signals. Data are presented as mean ± SD, *P = 0.002, n = 3 embryos (C); *P = 0.0009, n = 4 embryos (F); *P = 0.0003, n = 3 embryos (I). (J and K) Coimmunoprecipitation assay of neural stem cells treated with a high concentration of glucose (27.8 mM or 500 mg/dL) for 18 h. Pull-down with antibodies recognizing α-Syn (J) and Htt (K); Western blot detection of p62 with an antip62 antibody. Arrows and open stars indicate full-length and degraded products of p62, respectively. Arrowheads indicate molecular weight markers (70 kDa and 50 kDa). n = 3 repeats. ND, nondiabetic; DM, diabetes mellitus. (Scale bar = 20 μm in A–H.)

E9.5, the period of early neurulation. After treatment, neural tissues were collected at E9.5 and E10.5 for examination of protein modification and aggregation. In the embryos of the DM+PBA group, the levels of global protein ubiquitinylation and aggregation were significantly decreased, compared with those in the DM+VEH group (Fig. 4 D and E). The involvement of α-Syn, Parkin, and Htt in the p62-containing inclusion bodies was also significantly decreased, compared with that in the DM+VEH group (Fig. 4 F–H).

(p-eIF2α). Maternal diabetes also compromises cell survival regulation, indicated by decreased expression of protein kinase Bα (p-Pkbα) and mechanistic target of rapamycin (p-mTOR) (8, 15). In the present study, PBA treatment significantly decreased the levels of Atf6 (Fig. 6A), p-Perk (Fig. 6B), and p-eIF2α (Fig. 6C) and increased the activation of p-Pkbα (Fig. 6D) and p-mTOR (Fig. 6E), compared with the levels of the proteins in the DM+VEH group, at E9.5 and E10.5.

Effects of PBA on NTDs, Apoptosis, and Caspase-8 Activation. In human diabetic pregnancies, NTDs in offspring are present in the brain and spinal cord, and are frequently manifested as exencephaly, anencephaly, and spina bifida (37). Our mouse model system recapitulates these anomalies. Embryos from the DM+VEH group displayed NTDs in the forebrain (presumptive anencephaly), midbrain and/or hindbrain (presumptive exencephaly), and spinal cord (spina bifida) (Fig. 5B) (38). The NTD rate, examined at E10.5, in the DM+VEH group was significantly higher than that in the ND group (Table 1). When diabetic pregnant mice were treated with PBA, the NTD rate was significantly lower than that in the DM+VEH group, but similar to that in the ND group (Table 1 and Fig. 5 A–C). PBA treatment also significantly reduced the levels of apoptosis in the dorsal region of the neural tube at E9.5 and E10.5 (Fig. 5 D–F) and also reduced the levels of Caspase-8 activation (Fig. 5 G and H).

Discussion Maternal diabetes mellitus perturbs protein folding in cells of the embryo, leading to accumulation of improperly folded proteins (8, 9). However, the fate of these abnormally folded proteins, which likely interfere with normal intracellular signaling and cellular activities, has not been explored. Here, we show that misfolded proteins form aggregates and are resistant to degradation via the UPS and ALS. In addition, the protein aggregates contain α-Syn, Parkin, and Htt, which are also involved in the pathogenesis of neurodegenerative diseases, including Alzheimer’s, Parkinson’s, and Huntington’s diseases (19–22). The presence of these aggregates in maternal diabetes-induced embryonic NTDs suggests that similar molecular processes may be occurring in diabetic embryopathy and neurodegenerative diseases. To control the quality of proteins, abnormally folded proteins are usually flagged with ubiquitin and degraded by the UPS (12). In the neural tube of embryos of diabetic mice, global protein ubiquitinylation is increased, indicating a significant accumulation of misfolded proteins. Abnormally folded proteins can aggregate and form inclusion bodies in the cell, thereby perturbing intracellular signaling and cell activity. Although these aggregates can be eliminated by p62-mediated ALS (25, 26), the accumulation of

Effects of PBA on Cellular Stress and Cell Survival Signaling. It was previously shown that maternal diabetes exacerbates ER stress, indicated by an increased expression of activating transcription factor 6 (Atf6), phosphorylated (p) protein kinase RNA-like ER kinase (PERK), and α-subunit of eukaryotic initiation factor 2 Zhao et al.

PNAS Early Edition | 3 of 6

Ub-proteins/β-actin

8 6 4 2 0

ND

BA

H

E +V

2

ND

H 2.5

0.8

2.0

0.6 0.4

0.0

DM

BA

DM

+P

µM 0

µM

+P

DM

1.0

0

H

BA

EH

+V

DM

0.2

E +V

0

0

1

ND

25

50

Dots/cell

3

HG+TMAO

100

G

4

HG+TUDCA

150

+P

F5

0

E

DM

DM

10

G

EH +V

H

G

N

4 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1616119114

HG+PBA

D

Dots/cell

p62-containing aggregates indicates an impairment of the ALS in embryonic cells of diabetic mice. Three proteins associated with neurodegenerative diseases, α-Syn, Parkin, and Htt, also form aggregates in the embryos of diabetic mice. These proteins are expressed in the neural epithelium during neurulation (28–30). Aggregation of these proteins may abolish their normal functions and also disturb intracellular signaling, as seen in neurodegenerative diseases (19–22). Protein aggregates containing α-Syn, known as Lewy bodies (LBs), are seen in many diseases, including neurodegenerative diseases, dementia with LBs, and multiple system atrophy, and thus are referred to as syncleinopathies (39, 40). The role of α-Syn inclusion bodies in the pathogenesis of neurodegenerative diseases is not completely understood; however, α-Syn interacts with phospholipids on plasma membranes and may be involved in altered phospholipid metabolism (41). Aberrant phospholipid metabolism has also been suggested as an important process in diabetic embryopathy (15). α-Syn is involved in regulation of mitochondrial dynamics, such as fusion and fission (42), and LBs localized in the mitochondria can influence electron transport to generate ROS (43, 44). Both perturbed mitochondrial dynamics and ROS overproduction are associated with hyperglycemia-induced embryonic malformations in diabetic embryopathy (1, 15); therefore, the role of LBs in diabetic embryopathy deserves further investigation. Parkin is an E3 ubiquitin ligase, which plays a role in protein degradation via the UPS (45, 46). Mutations in the PARK2 gene cause PARKIN aggregation and loss of function in neurons, which are hallmarks of Parkinson’s disease (47). In the present study, we observed Parkin aggregation in the neural cells of embryos from diabetic dams, suggesting that an impaired Parkin-mediated UPS may be involved in diabetic embryopathy. PARKIN interacts with PTEN-induced putative kinase 1 (PINK1) on the mitochondrial membrane (48, 49). The action of the PARKIN–PINK1 complex

0

20

EH

0

10

30

+V

10

20

40

G

20

30

H

30

C

40

10

B

40

% (Positive/total cells)

A

G

Fig. 3. Protein interactions in the embryonic neural tube of diabetic mice. PLA assay to detect interaction between Hip1 and Hippi (A and B), Hip1 and Caspase-8 (D and E), and Hippi and Caspase-8 (G and H) in the dorsal brain of E9.5 embryos. PLA signals are red dots. DAPI counterstaining is blue. (Inset in A) Diagrammatic representation of a cross-section of the neural tube. The box marks the proximal area shown in the figures. (A, D, and G) Nondiabetic (ND). (B, E, and H) Diabetes mellitus (DM). (Scale bar = 20 μm.) (C, F, and I) Quantification of PLA signals. Data are presented as mean ± SD, *P = 0.0004, n = 3 embryos (C); *P = 0.0007, n = 3 embryos (F); *P = 0.0001, n = 3 embryos (I).

N

DM

µM

0

Dots/cell

ND

2

µM

4

50

DM

EH

6

25

I

G

H

Dots/cell

Hippi-Casp8

0

G

ND

1

+V

2

G

F

DM

H

E

ND

% (Positive/total cells)

D

3

Dots/cell

Hip1-Casp8

0

N

1

µM

2

% (Aggregates/total protein)

3

µM

4

is to regulate mitochondrial morphogenesis and remove damaged mitochondria through a form of ALS (50, 51). Mutations in PINK1 are associated with mitochondrial dysmorphogenesis and dysfunction, which are also characteristics of Parkinson’s disease (52, 53). In diabetic embryopathy, the aggregation of Parkin may perturb mitochondrial dynamics and protein quality control, thereby leading to embryonic malformations (15). Htt aggregation causes neuronal death via multiple mechanisms, including activating Caspase-8 (54). Caspase-8–dependent apoptosis has been observed in diabetes-induced NTDs (5). The DISC, which activates Caspase-8 in the tumor necrosis factor receptor (TNFR) and Fas receptor signaling systems, consists of Fasassociated death domain-containing protein, TNFR-associated factor 2, and TNFR-associated death domain protein (6, 7). However, in Huntington’s disease, Caspase-8 can be activated by a unique DISC, consisting of HIP1 and HIPPI (55). HTT aggregation releases HIP1 from the cell membrane to make it available for DISC formation (36). In this study, we observed that Hip1, Hippi, and Caspase-8 form a complex in the neural tissues of embryos from diabetic dams, suggesting that Htt aggregation serves as one mechanism of Caspase-8 activation in diabetic embryopathy. The pathway that begins with disrupted protein folding and ends with protein aggregation in embryonic cells subjected to

50

Dots/cell

C

5

25

B

Hip1-Hippi

A

DM

% (Positive/total cells)

ND

1.5 1.0 0.5

ND

EH

DM

+V

BA

DM

+P

0.0

ND

BA

EH

+V

DM

+P

DM

Fig. 4. Effects of chemical chaperones on protein aggregation in neural stem cells and embryonic neural tube of diabetic mice. (A–C) Neural stem cells cultured in high glucose and treated with chemical chaperones for 18 h. Data are presented as mean ± SD. (A) PBA, 25 μM and 50 μM. *P < 0.0001, n = 4 repeats. (B) TUDCA, 25 μM and 50 μM. *P < 0.0001, n = 4 repeats. (C) TMAO, 100 μM and 250 μM. *P < 0.0001, n = 4 repeats. (D–H) Embryos from diabetic mice treated with PBA (100 mg/kg) from E6.5 to E9.5. Data are presented as mean ± SD. (D) Levels of protein ubiquitinylation in the brain tissues of E9.5 embryos, assessed using Western blot assay. *P = 0.005, n = 3 embryos. (E) Levels of protein aggregation in brain tissues of E9.5 embryos, assessed by measuring Proteostat fluorescence intensity. *P = 0.0002, n = 6 embryos (three repeats). (F–H) PLA assay of interaction of p62 with α-Syn (F: *P = 0.005, n = 3 embryos), Parkin (G: *P = 0.0001, n = 3 embryos), and Htt (H: *P = 0.0001, n = 3 embryos) in the dorsal region of neural tube at E9.5. DM, diabetes mellitus; HG, high glucose; ND, nondiabetic; NG, normal glucose; VEH, vehicle.

Zhao et al.

A

B

C

G

1

2

3 50 kDa 30 kDa

18 kDa

-actin

F

H

0.8 0.6 0.4 0.2 0.0

1

2

3

Fig. 5. Effect of PBA on NTDs, apoptosis, and Caspase-8 activation in the embryos of diabetic mice. Diabetic pregnant dams were treated with PBA (100 mg/kg) from E6.5 to E9.5. (A–C) Embryos at E10.5. Arrow indicates open neural tube. (D–F) TUNEL assay of brain region of the neural tube of E9.5 embryos. (A and D) ND. (B and E) DM+VEH. (C and F) DM+PBA. (G) Western blot of Caspase-8 cleavage in the brain region of the neural tube. Arrowheads indicate cleaved products of Caspase-8 (43 kDa and 18 kDa). 1, ND; 2, DM+VEH; 3, DM+PBA; β-actin, loading control. (H) Quantification of the 18-kDa bands. Data are presented as mean ± SD, *P = 0.0003, n = 4 embryos. (Scale bar = 0.5 mm in A–C; 10 μm in D–F.)

hyperglycemic conditions is demonstrated by the treatment with three different chemical chaperones (PBA, TUDCA, and TMAO). The involvement of protein folding and aggregation in the pathogenesis of diabetic embryopathy is further demonstrated by in vivo treatment with PBA. More importantly, PBA treatment significantly reduces NTDs in embryos of diabetic dams. Such a significant effect is associated with reductions of Caspase-8 activation and apoptosis, alleviation of ER stress, and restoration of cell survival regulation. PBA, a Food and Drug Administrationapproved drug, has been shown to protect neuronal death by enhancing proper protein folding (56) and is used to ameliorate protein folding-associated diseases (57). These observations suggest that targeting the protein-folding crisis is a potential interventional approach to prevent diabetes-associated birth defects. Previous efforts to reduce embryonic malformations in diabetic embryopathy have focused on amelioration of oxidative stress using antioxidants (1). However, the effectiveness of antioxidants in humans remains questionable, as multiple large-scale human trials that have used antioxidants (i.e., vitamin C and vitamin E) to treat other diseases (i.e., preeclampsia, cardiovascular disease) have revealed unsatisfactory results (58–60). Although the reasons for this failure have not been identified, it is speculated that other cellular stress conditions (i.e., ER stress), which may not be affected by antioxidant therapies, may be involved (60). Chemical chaperones have been shown to improve diabetes in animal

models by restoring protein folding, thereby reducing cellular stress (61, 62). Chemical chaperones have also been tested in humans to treat protein-folding–related diseases (57, 63, 64). The observation that PBA reduces embryonic malformations in diabetic mice suggests that targeting aberrant protein folding could be a means to prevent birth defects in diabetic pregnancies. In addition to enhancing protein folding, targeting protein aggregation could mitigate the cytotoxicity of protein inclusion bodies (65). For example, efforts have been taken to block or disrupt α-Syn oligomization in Parkinson’s disease with promising outcomes (66). The delineation of the protein-folding crisis and aggregation in hyperglycemia-induced embryonic malformations provides opportunities for developing interventions at multiple levels to prevent birth defects in diabetic pregnancies.

Table 1. NTD rate in the embryos of diabetic mice treated with PBA

Groups ND DM+VEH DM+PBA

Total no. of embryos (litters)

Embryos with NTDs

NTD rate, %

Blood glucose, mg/dL; mean ± SD

85 (11) 96 (13) 123 (17)

1 23 5

1.18 23.96 4.06

121.7 ± 16.8 311.5 ± 7.7 291.7 ± 9.7

DM, diabetes mellitus; ND, nondiabetic; NTDs: open forebrain, midbrain, hindbrain, and/or spinal cord. PBA, phenylbutyrate; VEH, vehicle. P = 0.009 (DM+PBA vs. DM+VEH); P = 0.0016 (DM+VEH vs. ND); P = 0.17 (DM+PBA vs. ND).

Zhao et al.

Fig. 6. Changes in ER stress and cell survival factors by PBA treatment in the embryonic neural tube of diabetic mice. Data are presented as mean ± SD. (A–C) ER stress factors Atf6 (*P = 0.045, n = 3 embryos), p-Perk (*P = 0.04, n = 3 embryos), and p-eIF2α (*P = 0.042, n = 5 embryos). (D and E) Cell survival factors p-Pkbα (*P = 0.0013, n = 4 embryos) and p-mTOR (*P = 0.032, n = 4 embryos). β-Actin, loading control; DM, diabetes mellitus; ND, nondiabetic; PBA, phenylbutyrate; VEH, vehicle.

PNAS Early Edition | 5 of 6

MEDICAL SCIENCES

E

Casp8/ -actin

D

Materials and Methods Diabetes mellitus was induced in female mice (C57BL/6J) by i.v. injection of streptozotocin. The use of animals was approved by the Institutional Animal Care and Use Committee of University of Maryland, Baltimore. Detailed procedures about the generation of diabetic mice, cell culture, protein aggregation assay, TUNEL, proximity ligation assay, coimmunoprecipitation,

1. Zhao Z, Reece EA (2013) New concepts in diabetic embryopathy. Clin Lab Med 33: 207–233. 2. Greene ND, Copp AJ (2014) Neural tube defects. Annu Rev Neurosci 37:221–242. 3. O’Rahilly R, Muller F (1994) Neurulation in the normal human embryo. Ciba Found Symp 181:70–82. 4. Copp AJ, Greene ND, Murdoch JN (2003) The genetic basis of mammalian neurulation. Nat Rev Genet 4:784–793. 5. Zhao Z, Yang P, Eckert RL, Reece EA (2009) Caspase-8: A key role in the pathogenesis of diabetic embryopathy. Birth Defects Res B Dev Reprod Toxicol 86:72–77. 6. Tibbetts MD, Zheng L, Lenardo MJ (2003) The death effector domain protein family: Regulators of cellular homeostasis. Nat Immunol 4:404–409. 7. Sheikh MS, Huang Y (2003) Death receptor activation complexes: It takes two to activate TNF receptor 1. Cell Cycle 2:550–552. 8. Zhao Z, Eckert RL, Reece EA (2012) Reduction in embryonic malformations and alleviation of endoplasmic reticulum stress by nitric oxide synthase inhibition in diabetic embryopathy. Reprod Sci 19:823–831. 9. Zhao Z (2012) Endoplasmic reticulum stress in maternal diabetes-induced cardiac malformations during critical cardiogenesis period. Birth Defects Res B Dev Reprod Toxicol 95:1–6. 10. Walter P, Ron D (2011) The unfolded protein response: From stress pathway to homeostatic regulation. Science 334:1081–1086. 11. Meusser B, Hirsch C, Jarosch E, Sommer T (2005) ERAD: The long road to destruction. Nat Cell Biol 7:766–772. 12. Glickman MH, Ciechanover A (2002) The ubiquitin-proteasome proteolytic pathway: Destruction for the sake of construction. Physiol Rev 82:373–428. 13. Jarosch E, Lenk U, Sommer T (2003) Endoplasmic reticulum-associated protein degradation. Int Rev Cytol 223:39–81. 14. Tsai B, Ye Y, Rapoport TA (2002) Retro-translocation of proteins from the endoplasmic reticulum into the cytosol. Nat Rev Mol Cell Biol 3:246–255. 15. Cao L, et al. (2016) Identification of novel cell survival regulation in diabetic embryopathy via phospholipidomic profiling. Biochem Biophys Res Commun 470:599–605. 16. Yang X, Borg LA, Eriksson UJ (1995) Altered mitochondrial morphology of rat embryos in diabetic pregnancy. Anat Rec 241:255–267. 17. Ron D, Walter P (2007) Signal integration in the endoplasmic reticulum unfolded protein response. Nat Rev Mol Cell Biol 8:519–529. 18. Moreno-Gonzalez I, Soto C (2011) Misfolded protein aggregates: Mechanisms, structures and potential for disease transmission. Semin Cell Dev Biol 22:482–487. 19. Ross CA, Poirier MA (2004) Protein aggregation and neurodegenerative disease. Nat Med 10:S10–S17. 20. Selkoe DJ (2011) Alzheimer’s disease. Cold Spring Harb Perspect Biol 3:a004457. 21. Irwin DJ, Lee VM, Trojanowski JQ (2013) Parkinson’s disease dementia: Convergence of α-synuclein, tau and amyloid-beta pathologies. Nat Rev Neurosci 14:626–636. 22. Ross CA, Tabrizi SJ (2011) Huntington’s disease: From molecular pathogenesis to clinical treatment. Lancet Neurol 10:83–98. 23. Boya P, Reggiori F, Codogno P (2013) Emerging regulation and functions of autophagy. Nat Cell Biol 15:713–720. 24. Mizushima N, Komatsu M (2011) Autophagy: Renovation of cells and tissues. Cell 147: 728–741. 25. Rogov V, Dotsch V, Johansen T, Kirkin V (2014) Interactions between autophagy receptors and ubiquitin-like proteins form the molecular basis for selective autophagy. Mol Cell 53:167–178. 26. Katsuragi Y, Ichimura Y, Komatsu M (2015) p62/SQSTM1 functions as a signaling hub and an autophagy adaptor. FEBS J 282:4672–4678. 27. Zatloukal K, et al. (2002) p62 is a common component of cytoplasmic inclusions in protein aggregation diseases. Am J Pathol 160:255–263. 28. Dragatsis I, Efstratiadis A, Zeitlin S (1998) Mouse mutant embryos lacking huntingtin are rescued from lethality by wild-type extraembryonic tissues. Development 125: 1529–1539. 29. Kuhn K, Zhu XR, Lubbert H, Stichel CC (2004) Parkin expression in the developing mouse. Brain Res Dev Brain Res 149:131–142. 30. Zhong SC, et al. (2010) Expression and subcellular location of α-synuclein during mouse-embryonic development. Cell Mol Neurobiol 30:469–482. 31. Shen D, et al. (2011) Novel cell- and tissue-based assays for detecting misfolded and aggregated protein accumulation within aggresomes and inclusion bodies. Cell Biochem Biophys 60:173–185. 32. Schlett K, Madarasz E (1997) Retinoic acid induced neural differentiation in a neuroectodermal cell line immortalized by p53 deficiency. J Neurosci Res 47:405–415. 33. Soderberg O, et al. (2008) Characterizing proteins and their interactions in cells and tissues using the in situ proximity ligation assay. Methods 45:227–232. 34. Cao L, et al. (2016) Amelioration of intracellular stress and reduction of neural tube defects in embryos of diabetic mice by phytochemical quercetin. Sci Rep 6:21491.

6 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1616119114

and Western blot assay, as well as statistical analyses, can be found in the SI Appendix, Supplementary Materials and Methods. ACKNOWLEDGMENTS. We thank Hua Li for technical assistance, Dr. Min Zhan for statistical analyses, and Dr. Julie Rosen for critical reading and editing. The work was supported by the National Institutes of Health under Award Numbers HD076245 (to Z.Z.) and HD075995 (to Z.Z.).

35. Bates G (2003) Huntingtin aggregation and toxicity in Huntington’s disease. Lancet 361:1642–1644. 36. Gervais FG, et al. (2002) Recruitment and activation of caspase-8 by the Huntingtininteracting protein Hip-1 and a novel partner Hippi. Nat Cell Biol 4:95–105. 37. Correa A, et al. (2008) Diabetes mellitus and birth defects. Am J Obstet Gynecol 199: 237.e1–9. 38. Zhao Z (2010) Cardiac malformations and alteration of TGFb signaling system in diabetic embryopathy. Birth Defects Res B Dev Reprod Toxicol 89:97–105. 39. Spillantini MG, et al. (1997) α-Synuclein in Lewy bodies. Nature 388:839–840. 40. Goedert M (2001) α-Synuclein and neurodegenerative diseases. Nat Rev Neurosci 2:492–501. 41. Dikiy I, Eliezer D (2012) Folding and misfolding of α-synuclein on membranes. Biochim Biophys Acta 1818:1013–1018. 42. Norris KL, et al. (2015) Convergence of Parkin, PINK1, and α-Synuclein on stressinduced mitochondrial morphological remodeling. J Biol Chem 290:13862–13874. 43. Reeve AK, et al. (2015) Aggregated α-synuclein and complex I deficiency: Exploration of their relationship in differentiated neurons. Cell Death Dis 6:e1820. 44. Devi L, Raghavendran V, Prabhu BM, Avadhani NG, Anandatheerthavarada HK (2008) Mitochondrial import and accumulation of α-synuclein impair complex I in human dopaminergic neuronal cultures and Parkinson disease brain. J Biol Chem 283: 9089–9100. 45. Trempe JF, et al. (2013) Structure of parkin reveals mechanisms for ubiquitin ligase activation. Science 340:1451–1455. 46. Zhang Y, et al. (2000) Parkin functions as an E2-dependent ubiquitin-protein ligase and promotes the degradation of the synaptic vesicle-associated protein, CDCrel-1. Proc Natl Acad Sci USA 97:13354–13359. 47. Sriram SR, et al. (2005) Familial-associated mutations differentially disrupt the solubility, localization, binding and ubiquitination properties of parkin. Hum Mol Genet 14:2571–2586. 48. Kim Y, et al. (2008) PINK1 controls mitochondrial localization of Parkin through direct phosphorylation. Biochem Biophys Res Commun 377:975–980. 49. Kawajiri S, et al. (2010) PINK1 is recruited to mitochondria with parkin and associates with LC3 in mitophagy. FEBS Lett 584:1073–1079. 50. Poole AC, et al. (2008) The PINK1/Parkin pathway regulates mitochondrial morphology. Proc Natl Acad Sci USA 105:1638–1643. 51. Vives-Bauza C, et al. (2010) PINK1-dependent recruitment of Parkin to mitochondria in mitophagy. Proc Natl Acad Sci USA 107:378–383. 52. Lim KL, Ng XH, Grace LG, Yao TP (2012) Mitochondrial dynamics and Parkinson’s disease: Focus on parkin. Antioxid Redox Signal 16:935–949. 53. Bueler H (2009) Impaired mitochondrial dynamics and function in the pathogenesis of Parkinson’s disease. Exp Neurol 218:235–246. 54. Labbadia J, Morimoto RI (2013) Huntington’s disease: Underlying molecular mechanisms and emerging concepts. Trends Biochem Sci 38:378–385. 55. Mattson MP (2002) Accomplices to neuronal death. Nature 415:377–379. 56. Mimori S, et al. (2013) 4-Phenylbutyric acid protects against neuronal cell death by primarily acting as a chemical chaperone rather than histone deacetylase inhibitor. Bioorg Med Chem Lett 23:6015–6018. 57. Rubenstein RC, Zeitlin PL (1998) A pilot clinical trial of oral sodium 4-phenylbutyrate (Buphenyl) in ΔF508-homozygous cystic fibrosis patients: Partial restoration of nasal epithelial CFTR function. Am J Respir Crit Care Med 157:484–490. 58. Villar J, et al. (2009) World Health Organisation multicentre randomised trial of supplementation with vitamins C and E among pregnant women at high risk for preeclampsia in populations of low nutritional status from developing countries. Br J Obstet Gynaecol 116:780–788. 59. Briasoulis A, Tousoulis D, Antoniades C, Stefanadis C (2009) The oxidative stress menace to coronary vasculature: Any place for antioxidants? Curr Pharm Des 15: 3078–3090. 60. Steinhubl SR (2008) Why have antioxidants failed in clinical trials? Am J Cardiol 101: 14D–19D. 61. Ozcan U, et al. (2006) Chemical chaperones reduce ER stress and restore glucose homeostasis in a mouse model of type 2 diabetes. Science 313:1137–1140. 62. Engin F, et al. (2013) Restoration of the unfolded protein response in pancreatic beta cells protects mice against type 1 diabetes. Sci Transl Med 5:211ra156. 63. Parenti G, Andria G, Valenzano KJ (2015) Pharmacological chaperone therapy: Preclinical development, clinical translation, and prospects for the treatment of lysosomal storage disorders. Mol Ther 23:1138–1148. 64. Ebrahimi-Fakhari D, Wahlster L, McLean PJ (2011) Molecular chaperones in Parkinson’s disease: Present and future. J Parkinsons Dis 1:299–320. 65. Ignatova Z, Gierasch LM (2006) Inhibition of protein aggregation in vitro and in vivo by a natural osmoprotectant. Proc Natl Acad Sci USA 103:13357–13361. 66. Dehay B, et al. (2015) Targeting α-synuclein for treatment of Parkinson’s disease: Mechanistic and therapeutic considerations. Lancet Neurol 14:855–866.

Zhao et al.