Author Correction: Novel mode of defective neural tube closure in the non-obese diabetic (NOD) mouse strain

gene

from the uterus at 8.5 days of gestation using a Leica MZ6 stereomicroscope (Leica Microsystems, Buffalo Grove, IL). Extraembryonic endoderm and amnion were removed, and embryos were either fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) at 4°C, or placed in Tissue-Tek O.C.T. compound (Fisher Scientific, Pittsburgh, PA), frozen, and stored at -80°C.

Imaging of embryos.
For imaging, fixed embryos were stained in PBS/4% paraformaldehyde containing DAPI (4',6-diamidino-2-phenylindole) in order to visualize cell nuclei. Embryos were washed in PBS, dehydrated in a graded alcohol series, cleared in BABB (a 1:2 mixture of benzyl alcohol and benzyl benzoate), and imaged on a Leica SP5 Confocal microscope using 2-photon technology. Optical sections were used to generate threedimensional reconstructions of individual embryos using Imaris software (Bitplane Inc., South Windsor, CT). Computationally generated embryo surfaces as well as virtual sections were used to evaluate the anatomy of dysmorphologies.
were used to produce cryosections at 16µm nominal thickness on a Microm HM560 cryostat. Sections were mounted on glass slides with a PEN (polyethylene naphthalate) membrane, dried at 50°C for 30 minutes, and stored in a vacuum desiccator. Laser microdissection was performed on a Leica LMD 6000 laser microdissection system. In a set of serial sections from a single embryo, the site of neural tube closure (closure 1) was identified. At the anterior side of the closure, open neural tube tissues was microdissected and collected from the 10 sections preceding the first section showing a closed neural tube. All microdissected tissue segments from an individual embryo were collected into Trizol (Life Technologies, Grand Island, NY) for RNA preparations; RNA was quantified on a Qbit fluorometer (Life Technologies, Grand Island, NY), and stored at -80°C. Ectopic tissue from embryos showing morphogenetic deficiencies of the neural plate was dissected using sterile #55 forceps (Ted Pella Inc., Redding, CA); care was taken not to penetrate the neural plate during removal of the ectopic material. Tissue was directly transferred to Trizol for RNA extraction; RNA was quantified on a Qbit fluorometer, and stored at -80°C.

Molecular analyses.
Gene expression profiling was performed by expression tag sequencing (SAGE) on an AB SOLiD 5500XL next-generation sequencing instrument using reagent kits from the manufacturer (Applied Biosystems, Foster City, CA). Sequence reads were aligned to mouse RefSeq transcripts (version mm9) as the reference, utilizing the program SOLiDSAGE (Applied Biosystems). Only uniquely mapped sequence reads were counted to generate the expression count level for each respective RefSeq gene.
Sequencing libraries were constructed from 7ng of laser-microdissected open neural tube material, and yielded an average of 10.2 million mapped reads after quality control. Sequencing libraries from protrusion material were generated from 12ng of total RNA as input material, and yielded an average of 8.9 million mapped reads after alignment and quality control. Differential expression between open neural tube and protrusion tissue was analyzed using the R/Bioconductor program DESeq version 1.81 72 ; genes were considered differentially expressed if the adjusted p-value (Benjamini & Hochberg procedure) for the respective comparison was below 0.1.
Hierarchical clustering analyses were performed using MeV (Multi Experiment Viewer, http://www.tm4.org/mev/), using Manhattan distance as the correlation metric. Pathway analyses and biological annotation were done using DAVID 73,74 . Validation of sequencing-based expression data was done by qPCR 71,75 using cDNA equivalent to 50pg total RNA per reaction, 4 technical replicates per gene, and normalization to Polε4 as internal control.

Histological analyses.
In situ hybridizations on cryosections using digoxigenin-labelled antisense riboprobes were performed as described 71,76 . For immunohistochemistry 77 , cryosections were stained with antibodies against Laminin (Abcam, Cambridge, MA) and Phospho-Histone 3 (directly labeled with Alexa Fluor 488; Biolegend, San Diego, CA); Laminin was detected by indirect immunofluorescence using a goat-anti rabbit Alexa Fluor 594labeled secondary antibody (Life Technologies, Grand Island, NY). Sections were counterstained with Hoechst 33342 (Life Technologies, Grand Island, NY), and fluorescence was recorded on an Everest digital microscopy workstation using Slidebook software (Intelligent Imaging Innovations, Inc., Denver, CO).

Explant cultures.
Protrusions were microdissected from diabetes-exposed NOD embryos, and posterior tissues containing open neural plate, mesoderm and underlying endoderm were dissected from normal or diabetes-exposed NOD or FVB embryos using glass needles 32 as depicted in Supplemental Figure 3. The explants were placed into 12-well tissue culture plates coated with Matrigel, and cultured in DMEM containing 20% FCS and 2-Mercaptoethanol. Images were taken 6 hours after initiation of the cultures, and again 20 hours later, after a total of 26 hours in culture. At this time point, explants from normal E7.5 NOD embryos are in a phase of active outgrowth (which continues for at least another 3 days without obvious signs of cell death), with the margin of the explant extending outward at an average speed of 0.32 µm/min. (±0.12 µm/min., n=10), as determined from time-lapse videos, using the Leica TIRF DMI6000 system. These conditions support cell migration and differentiation of mesoderm, including cardiac mesoderm, as evidenced by the presence of rhythmically contractile areas in 7 out of 14 outgrowths from primitive streak explants of E7.5 NOD embryos from normal pregnancies at 26 hours in culture (time-lapse video provided in Supplemental Material).
Outgrowth from an explant is expressed as distance migrated by the margin of the explant and was determined as follows: for each image, the size of the area covered by the explant was determined, and the radius was calculated for a circle of the same area.
The radius at the 6h time point of the explant culture was subtracted from the radius for the area covered by the same explant at the 26h time point of culture. Net outgrowth over 20 hours of explants from E8.5 diabetes-exposed NOD embryos was significantly slower (p=6.7x10 -4 ) at an average of 0.27 µm/minute (±0.10 µm/min., n=39) than of explants from normal E8.5 NOD embryos at 0.44 µm/minute (±0.09 µm/min., n=8).

Legends to Supplementary Figures
Supplementary Figure 1: NOD diabetic pregnancies and folinic acid.

Legends to Supplementary Tables
Supplementary Table 1: Protrusion Incidence Data.
Incidence of protrusions in E8.5 embryos from diabetic pregnancies of the NOD strain compared with diabetic pregnancies in the FVB strain after chemical induction of diabetes. Referring to pregnancies, the term 'affected' indicates presence of a protrusion in at least one of the embryos of the dam; in case of embryos, the term refers to at least one protrusion in the respective embryo. Blood glucose and litter size values represent the mean plus/minus the standard deviation. Neither maternal blood glucose levels (Anova; NOD: p=0.89; FVB: p=0.14) nor litter sizes (Anova; NOD: p=0.20; FVB: p=0.57) were significantly different between protrusion-affected and non-affected pregnancies.
Supplementary The data indicate that protrusions express extensive gene regulatory networks driven by key mesodermal transcription factors.