Moderate prenatal alcohol exposure increases total length of L1-expressing axons in, E15.5 mice

Abstract

Public health campaigns broadcast the link between heavy alcohol consumption during pregnancy and physical, cognitive, and behavioral birth defects; however, they appear less effective in deterring moderate consumption prevalent in women who are pregnant or of childbearing age. The incidence of mild Fetal Alcohol Spectrum Disorders (FASD) is likely underestimated because the affected individuals lack physical signs such as retarded growth and facial dysmorphology and cognitive/behavioral deficits are not commonly detected until late childhood. Sensory information processing is distorted in FASD, but alcohol’s effects on the development of axons that mediate these functions are not widely investigated. We hypothesize that alcohol exposure alters axon growth and guidance contributing to the aberrant connectivity that is a hallmark of FASD. To test this, we administered alcohol to pregnant dams from embryonic day (E) 7.5 to 14.5, during the time that axons which form the major forebrain tracts are growing. We found that moderate alcohol exposure had no effect on body weight of E15.5 embryos, but significantly increased the length of L1+ axons. To investigate a possible cause of increased L1+ axon length, we investigated the number and distribution of corridor cells, one of multiple guidance cues for thalamocortical axons, which are involved in sensory processing. Alcohol did not affect corridor cell number or distribution at the time which thalamocortical axons are migrating. Future studies will investigate the function of other guidance cues for thalamocortical axons, as well as lasting consequences of axon misguidance with prenatal alcohol exposure.

1. Introduction

1.1. Fetal Alcohol Spectrum Disorders (FASD)

Fetal alcohol spectrum disorders (FASD) refers to a range of cognitive defects, distinct facial morphology (on the severe end of the spectrum), and growth deficiencies due to prenatal alcohol exposure (PAE)¹. Symptoms vary widely in severity and are dependent upon degree of gestational alcohol exposure². In a study of alcohol consumption among pregnant women between the ages of 18–44 in the United States, it was found that 1 in 9 pregnant women reported drinking with about 1/3 of these engaging in binge drinking³. The prevalence of alcohol consumption during pregnancy has been escalating from 2011–2018⁴. An earlier report found consumption was highest during the first trimester, but about 8% of women continued to drink during the second and third trimesters⁵.

Sensory processing deficits are one symptom that can present in cases of mild and severe PAE, and may lead to abnormal behaviors in children with FASD^{5, 6}. In particular, children with an FASD may demonstrate under-responsiveness to sensory stimuli or impaired ability to filter auditory stimuli⁷. In fact, children with varying degrees of FASD exhibited delayed auditory processing, supporting audition as a sensitive measure of PAE⁸.

Structural abnormalities have also been observed in PAE and related alcohol abuse disorders. Alcohol use disorders and comorbid mental disorders have been linked to disruptive volumetric changes and deterioration of the thalamus^{9, 10}. There are similar anatomical and circuit pathologies within the thalamus in both alcohol use disorder and developmental alcohol exposure which may underlie cognitive dysfunction¹¹. The mediolateral and laterodorsal thalamus in rats were particularly vulnerable to third trimester alcohol exposure, displaying increased cell death and reduced thalamic volume¹². While studies have shown that the structural reduction of the thalamus remains a constant within children prenatally exposed to alcohol¹³, both structural and functional thalamic changes demonstrate sex differences. The magnitude of differences in thalamic volume is larger in males than females¹⁴. In addition to volumetric changes, PAE reduced the neural connectivity of the cortex, hypothalamus and thalamus in rats¹⁵. At the behavioral level, sex-specific changes can also occur. Mice prenatally exposed to alcohol demonstrate sex-specific changes in anxiety and social behaviors^{16, 17}. In human studies, adolescent males with FASD also exhibit reduced thalamic activation and poorer performance on a virtual water maze task when compared to females, further demonstrating an increased male susceptibility to PAE¹³.

1.2. Thalamocortical Axons

Thalamocortical axons (TCAs) carry sensory information from the thalamus to the cortex for the information to be processed¹⁸. Correct migration of TCAs to their laminar position is a complex, multistep process that relies on multiple guidance cues¹⁹. Studies have shown that PAE disrupts TCA guidance, subsequently leading to a disorganized cortex. This disorganization results in altered communication between brain regions, specifically the thalamocortical connections and crossed corticothalamic connections, contributing to the observed behavioral differences of PAE²⁰. A low dose of PAE reduced the number of serotonergic TCAs which reach the cortex by embryonic day (E) 18.5²¹.

TCAs migrate to the cortex through the subpallium, crossing two main checkpoints: the diencephalic-telencephalic boundary (DTB) by E13.5 and pallial-subpallial boundary (PSPB) by E15^{22, 23}. They are repelled from the hypothalamus and midline by Slit1 and Slit2 signaling, leading them to cross the DTB²⁴. After crossing the DTB, TCAs must pass through the medial ganglionic eminence (MGE) of the subpallium, an area that is repulsive to TCAs^{25, 26}. The MGE is an important intermediate target during TCA pathfinding²⁵. TCAs, along with other axons traversing the MGE²⁵, express L1, a neural cell adhesion molecule found throughout the developing forebrain²⁷. L1 interactions with the cytoskeleton, growth factors, and axon guidance molecules facilitate cell migration and neurite extension and are thus critical determinants of synaptic targeting in vivo²⁷. TCAs require additional guidance cues to direct them towards the PSPB, including a cell population known as corridor cells.

1.3. Corridor Cells

Corridor cells provide a necessary pathway through the MGE via neuregulin-1 signaling²⁶. Corridor cells originate in the lateral ganglionic eminence (LGE) around E12 and migrate tangentially to the MGE to guide TCAs²⁶. They express genes typical of LGE-born cells, including islet1²⁶. Pax6 expression and Slit2 signaling are required for the proper placement of corridor cells in the MGE²⁸. TCA migration is altered in Pax6 mutant mice, possibly due to a widened Islet1+ corridor region²⁹.

Corridor cells share features with other cell populations including cortical interneurons. Both populations are GABAergic, and they pass each other in the MGE while corridor cells migrate to the MGE and cortical interneurons migrate to the cortex^{26, 30}. Following PAE, the number of GABAergic interneurons in all cortical layers is increased³⁰. Tangential migration of GABAergic interneurons from the MGE to the cortex is also increased by PAE³¹. We hypothesize that corridor cells may be sensitive to PAE because of their similarities with cortical interneurons.

The current study operated within the paradigm of moderate PAE, equivalent to 3–4 drinks per occasion³² within the second trimester, similar to a gavage injection paradigm used in C57BL/6J mice³³. L1 is a well-established target of alcohol and studies using cultured cortical and cerebellar cells have demonstrated alcohol exposure disrupts L1-mediated axon extension^{34, 35}, but no studies have explored these effects in vivo. Thus, we stained E15.5 brains for L1, then measured the length of L1+ axons using the Axon Tracer plugin of ImageJ³⁶. AxonTracer is an open-source algorithm which provides automated detection and quantification of labeled axons. This algorithm is advantageous because it requires minimal user input, reduces bias, produces comparable results to the widely used (but semi-automated) NeuronJ plugin, is able to efficiently trace large quantities of axons in each image, and is amenable to high-throughput image analysis. To focus on the guidance of TCAs, we measured the number and distribution of corridor cells in the MGE at E13.5 using immunostaining for Islet1. Despite labeling LGE-derived cells more broadly, Islet 1 is a suitable marker for corridor cells because Islet1 expression in the MGE is limited to the corridor region²⁶. By E13.5, corridor cells have migrated to the MGE and TCAs are traveling through the corridor region, making this timepoint suitable for analysis of this region²⁶. We hypothesized that PAE would reduce the number of corridor cells and widen their distribution, preventing L1+ axons from correctly extending into the cortex at E15.5. Differences in embryonic weight due to sex and treatment were also investigated. Unfortunately, our sample size did not give sufficient power to determine sex differences for axon length.

2. Methods

2.1. Mice and Alcohol Injections.

Timed pregnant Swiss Webster mice, which were E2.5 upon arrival, were purchased from Charles River Laboratories. Mice were group-housed and maintained on a 12-hour reverse light-dark cycle. All procedures were approved by the Ursinus College Institutional Animal Care and Use Committee and are in accordance with the National Institute of Health Guide for Care of Laboratory Animals. Pregnant dams were randomly assigned to receive daily subcutaneous injections of 15μL per gram of body weight (2.3g/kg) of either saline or 20% ethanol (EtOH) diluted in saline beginning on E7.5. This dose results in a maternal blood ethanol concentration of 154.47 +/− 33.70 mg/dl (mean +/− standard deviation) on E14.5 (n = 3 dams) as determined by gas chromatography of trunk blood samples (Danforth Center, St. Louis, MO), consistent with reports in the literature³³. Trunk blood samples were collected on E14.5 only, at time of euthanasia. Prior to injections, mice were anesthetized with isoflurane to minimize distress of daily injections. Subcutaneous injections allow us to control exactly how much EtOH is administered into the dams without endangering embryonic offspring. Each injection was just under 1.5 minutes of isofluorane exposure.

Dams for Islet1 analysis were euthanized on E13.5. Dams for L1 analysis were injected with saline or EtOH until E14.5 and euthanized on E15.5. On their euthanasia day, dams at E13.5 also received intraperitoneal injections of 50mg per kg of body weight of bromodeoxyuridine (BrdU) for another study. Thirty minutes after receiving the BrdU injections, dams were euthanized with CO₂. Embryos were removed via Caesarian section and preserved in 4% paraformaldehyde (PFA) at 4°C. Embryos were selected randomly for analysis. At time of selection, we did not notice any gross morphological differences in degree of digit differentiation between embryos.

2.2. Embryo Preparation

2.2.1. Sex Genotyping.

We considered sex in our analysis because of sex-specific effects of PAE in both humans and rodents^{14, 16, 17}. To distinguish between males and female embryos, the Jarid1C (on X chromosome) and Jarid1D (on Y chromosome) genes were amplified³⁷. Tails were removed from each embryo for DNA isolation. DNA extraction was performed using the Epoch Life Sciences GenCatch Genomic DNA Extraction kit, following an adjusted version of the kit’s protocol (Epoch 2460050–1/13). The attached protocol suggests using 200μL of heated water or Tris as an elution buffer, but we used 20μL of heated elution buffer from the Qiagen DNeasy Blood & Tissue Kit to maximize DNA yield (Qiagen 69506). Once DNA was isolated, concentrations were measured using a Nanodrop 2000 (Thermo Fisher). We prepared a stock solution of sex primers: 1μL of forward primer (5’-CTGAAGCTTTTGGCTTTGAG-3’), 1μL of reverse primer (5’-CCACTGCCAAATTCTTTGG-3’), and 100μL of sterile H₂O. For each sample, we prepared this mixture for PCR: 10μL of Taq polymerase (Sigma P0982), 1μL of the premade sex primer stock, 200ng of DNA, and enough sterile H₂O to fill the volume to 20μL. Samples ran through a PCR program of 35 cycles in the following: denaturing for 20 seconds at 94°C, elongation for 60 seconds at 54°C, and extension for 40 seconds at 72°C. Following PCR, the samples were run on a 2% agarose gel diluted in 0.5X Tris/borate/EDTA (TBE) at 100V for approximately 90 minutes.

2.2.2. Microtome Sectioning.

One male and one female embryo from each litter were used in this study. After determining the sex, chosen embryos were sent to AML Laboratories (Jacksonville, FL) and embedded in paraffin in the coronal orientation. Some embryos were also sectioned in-house with a Leica microtome. All embryos were sectioned at 5-micron thickness.

2.2.3. Nissl Staining.

From each sectioned embryo, every fifth slide was selected for Nissl staining to identify the region of interest for immunostaining. Appropriate slides contained the medial and lateral ganglionic eminences. To visualize the region of interest and apply the Nissl stain, wax was removed by a 5-minute bath in Histoclear (Electron Microscopy Sciences #64110). Tissue was then rehydrated by a 5-minute bath in 100% ethanol, followed by 2-minute baths in 95% ethanol, 70% ethanol, and distilled water, in that order. Slides were stained by a 2-minute bath in 0.1% cresyl violet acetate (Electron Microscopy Sciences #26089–01), followed by a 2-minute rinse in distilled water. The tissue was then dehydrated by a series of 2-minute baths in 70% ethanol and 95% percent ethanol, followed by a 5 minute bath in 100% ethanol and a 5 minute bath in Histoclear. Slides were dried in open air for at least 1 hour prior to coverslipping with Cytoseal (Electron Microscopy Sciences #18007).

2.3. Immunohistochemistry and Analysis

2.3.1. L1 and Islet1 Immunostaining, Microscopy, and Axon Tracing.

To visualize axons of interest, E15.5 brains were prepared for L1 staining. E13.5 brains were Islet1 stained to visualize corridor cells. Dewaxing and tissue rehydration procedure was performed as previously described. Slides incubated in 1% Antigen Unmasking Solution, citric acid based (Vector H-3300) in a microwave (Islet1 brains) or 95°C water bath (L1 brains) for 20 minutes. Antigen unmasking solution was rinsed in a 2-minute incubation in distilled water and a 5-minute incubation in 1X phosphate-buffered saline (PBS). After these rinses, each slide was treated with 200μL of antigen blocking solution, consisting of 10% normal goat serum and 0.1% Triton (Electron Microscopy Sciences #22145) in 1X PBS. Slides were blocked for 30 minutes in a humidified chamber, humidified with water. Following the block, slides rinsed for 5 minutes in 1X PBS. Slides were rinsed in 1X PBS. Islet1 antibody (Developmental Studies Hybridoma Bank 39.4D5) was diluted 1:100 in 1X PBS. Anti-L1 NCAM antibody (AbCam #ab208155) was diluted at a concentration of 1:250 in 1X PBS. Each slide received 200μL of its respective primary antibody and incubated overnight at 4°C in a humidified chamber. The next day, L1 slides were given three 5-minute washes in 1X PBS, while Islet1 slides were rinsed through three 10-minute washes in 1x PBS + 0.1% Tween-20. Slides were treated with 200 μL each of ImmPRESS goat anti-mouse secondary antibody (Islet1 slides; Vector MP-7452) or ImmPRESS goat anti-rabbit secondary antibody (L1 slides; Vector #MP-7401) and incubated for 30 minutes in a humidified chamber. Slides were also covered with parafilm during the incubation. Parafilm was then removed and slides were rinsed for 5 minutes in a 1X PBS bath, then 5 minutes in a distilled water bath. Diaminobenzidine (DAB; Vector #SK-4105) was used to visualize staining. Slides were covered with 200μL each of DAB and incubated for about 1 minute before being submerged in fresh distilled water. Slides were then dehydrated as previously described, and allowed to dry before coverslipping with Cytoseal.

2.3.2. Axon Tracer Analysis of E15.5 L1 immunostaining

Axon Tracer was utilized to quantify axon length. Images of L1-stained E15.5 brains were collected using a Nikon Eclipse 80i Compound Microscope at 10X magnification equipped with a Nikon C2-SH digital camera. NIS-Elements AR software was used to capture images. As some sections exhibited tissue damage in the region of staining, we were not able to have the same sample size between groups (rostral and middle: n=3 alcohol, n=4 saline; middle caudal: n=4 for each; caudal: n=8 alcohol, n=4 saline). One section was analyzed from each litter. L1-stained axons from each hemisphere were analyzed using ImageJ with AxonTracer to determine total axon length³⁶ (see supplemental figure). The algorithm used by AxonTracer does not measure the length of individual axons, but rather the total length of all the axons in a particular section. All axon tracing data is saved as raw data (total pixels per image). Investigators were not aware of sex or treatment during analysis.

2.3.2. Islet1 Binning Analysis

Following immunostaining, we used a binning protocol to quantify the distribution of Islet1+ corridor cells in the MGE. Images of L1-stained E15.5 brains were collected using a Nikon Eclipse 80i Compound Microscope at 10X magnification equipped with a Nikon C2-SH digital camera. NIS-Elements AR software was used to capture images. Several images were required to capture the entire ganglionic eminence area. These images were stitched together using the PhotoMerge Panorama function of Photoshop Elements. Merged images were loaded into Fiji software. The images were rotated so that the pial surface was horizontal. The width of the ganglionic eminences was measured and divided in half (representing the MGE and LGE). A grid was placed over the merged images and sized so that the MGE was covered by 10 rows and 7 columns of bins. The Cell Counter plugin of Fiji was used to count the cells in each bin. This binning procedure was adapted from another study which used binning to quantify the distribution of Islet1+ corridor cells in Pax6 knockouts²⁹.

3. Results

3.1. Sex and/or moderate prenatal alcohol exposure do not affect E15.5 mice weight

There was no effect of sex (p = 0.46) or treatment (p = 0.17) on embryonic body weight nor was there an interaction between sex and treatment (p = 0.895, Two-way ANOVA; Fig 1). Our results do not support our hypothesis that sex and treatment changes body weight at E15.5.

Figure 1. Moderate prenatal alcohol exposure has no impact on body weight in E15.5 mice.

There is no effect of sex (p = 0.46) or treatment (p = 0.17) on embryonic body weight nor is there an interaction between sex or treatment (p = 0.895, Two-way ANOVA; alcohol males: n = 7, 0.38 ± 0.06g, alcohol females: n = 5, 0.36 ± 0.08g, saline males: n = 10, 0.34 ± 0.095g, saline females: n = 8, 0.32 ± 0.04g). Data are represented as the mean; error bars represent standard deviation. Statistical analysis was performed with t-test and with a two-way analysis of variance (ANOVA).

3.2. Moderate PAE increases L1+ axon length increases at E15.5

In order to explore the effects of prenatal alcohol exposure on thalamocortical axon (TCA) guidance and formation, we immunostained axons with LI. Utilizing an atlas of an E16 mouse brain³⁸ in coronal sections, we refer to each level in relative rostral to caudal directions of the embryonic brain. After visually analyzing L1+ axons and finding no visibly distinct differences across the atlas levels examined (Fig. 2a), we quantified the axon lengths using the AxonTracer plugin for ImageJ³⁶. A two-way ANOVA indicated that there was no significant main effect of atlas level, F(3, 16) = 1.17, p = 0.35, while there was a significant main effect of treatment on axon length, F(1, 16) = 7.51, p = 0.01. There was not a significant interaction between atlas level and treatment on axon length, F(3, 16) = 0.19, p = 0.90. Our results do not support our hypothesis that axon length would be decreased due to moderate prenatal alcohol exposure but instead show increased axon length.

Figure 2. Moderate prenatal alcohol exposure increases total length of L1+ axons in E15.5 mice.

(a) Representative pictures of E15.5 coronal brain sections of alcohol- and saline-exposed groups. Atlas levels are in reference to an E16 mouse brain atlas (Schambra, 2008) in coronal sections. There were no visibly distinct differences across the atlas levels examined. Scale bar, 100um (b) A two-way ANOVA indicated that among the 4 atlas levels there were no significant differences (p = 0.35), while between treatments there was a significant difference (p = 0.0145). However, there was no interaction (p = 0.90). Rostral (n =3 alcohol, n = 4 saline) and middle caudal (n = 4 alcohol, n = 4 saline) sections showed the greatest difference between treatment groups. Data are represented as the mean; error bars represent standard deviation. Statistical analysis was performed with a two-way analysis of variance (ANOVA).

3.3. PAE does not alter the number or distribution of corridor cells in the MGE at E13.5

Because PAE increased the length of L1+ axons at E15.5, we decided to investigate PAE’s effects on corridor cells at E13.5. PAE did not alter the number or distribution of corridor cells in the MGE at E13.5 (Figure 3). Representative Islet1 staining images are shown in Figure 3a. Using a linear mixed-effects model fit, and accounting for random effect due to measuring both hemispheres of each individual, we did not find an effect of sex (p = .45) or treatment (p = .57) on total corridor cell count. A linear mixed-effects model did not indicate a significant effect of alcohol treatment (p = .67) or sex (p = .91) on corridor cell distribution (Figure 3b, c).

Figure 3. Prenatal alcohol exposure does not affect the number or distribution of Islet1+ nuclei in the medial ganglionic eminence at E13.5.

(a) Immunohistochemistry for Islet1 labels corridor cells in the MGE of saline-injected (left) and EtOH-injected (right) embryos. Both embryos shown are female. Scale bar, 100um (b) Following immunohistochemistry, Islet1+ nuclei were counted using the Cell Counter plugin of Fiji. There was no significant effect of sex (p = .45) or treatment (p = .57) on total corridor cell count. (c) To analyze distribution, the MGE was divided into ten equal horizontal bins. Islet1+ cells were counted in each bin. There was no significant effect of sex (p = .67) or treatment (p = .91) on corridor cell distribution. Data are represented as the mean; error bars represent standard deviation. (MGE: medial ganglionic eminence; LGE: lateral ganglionic eminence; EtOH: ethanol)

4. Discussion

We found that a moderate level of PAE increased the length of L1+ axons at E15.5. The increased length of L1+ axons in alcohol-exposed embryos did not correspond to altered corridor cell count or distribution at E13.5. One important caveat in our interpretation of data has to do with the fact that pregnant dams were subjected to isoflurane anesthesia to minimize distress due to daily injections. It is possible that the changes we observed represent an interaction between anesthesia and maternal alcohol exposure. However, our findings are in line with those of Godin et al. that showed a single high dose alcohol exposure on gestational day 7 resulted in a thickened subpallium with aberrant projections in offspring³⁹. We are unaware of any study investigating axon length in embryos in vivo, so we cannot determine if our findings are within range. We are also unable to determine length of individual axons using the AxonTracer program with the type of sections we analyzed (coronal sections immunostained using DAB chromogen). These limitations could be overcome with additional studies using a different plane of section to capture the full axon trajectory and/or co-immunostaining with a molecular marker for cell bodies. While previous studies investigating the effect of prenatal alcohol exposure have demonstrated decreased birth weights for exposed embryos, it is likely that there was no observable weight difference in our study due to the moderate level of alcohol exposure compared to the higher level other studies have utilized².

Corridor cell count and distribution are just two of many features which could lead to the PAE-induced increase in L1+ axon length. It is well known that alcohol exposure prevents growth cone collapse⁴⁰ and thus influences axonal projection by causing a decrease in cortical growth cone responsiveness³⁴. It is possible that the L1+ axons within our study display a lack of guidance cue responsiveness, resulting in the observed increase in axon length. This would match our findings that PAE did not alter the number or distribution of corridor cells in the MGE at E13.5, the time at which TCAs are migrating through the MGE. However, it is possible that corridor cell function, not TCA guidance cue responsiveness, is altered by PAE. Impaired corridor cell neuregulin-1 signaling could alter TCA growth without altering corridor cell placement. In addition, other TCA guidance factors may be more sensitive to PAE than corridor cells. Other necessary guidance cues include Slit/Robo signaling, Pax6 expression, and OL-protocadherin^{24, 29, 41}. We are currently studying PAE’s effects on Slit/Robo expression. Changes to Slit/Robo signaling could directly and indirectly impact TCA migration, as Slit/Robo repel axons from the midline and place corridor cells in the MGE^{24, 28}. Pax6 is necessary for TCA migration and correct formation of the corridor region²⁹. EtOH injections reduced Pax6 expression in rats⁴²; however, Pax6 expression in mice following PAE is unknown. OL-protocadherin is required for proper striatal axon migration, not TCA migration; however, striatal axons may act as a necessary scaffold for TCAs to cross through the subpallium⁴¹. PAE may affect one or more of these other guidance cues, impacting subsequent axon migration without altering corridor cell location.

Another target of future work includes further analysis of the other forebrain axon populations which express L1. L1 is an important cell adhesion molecule involved in axon migration that is expressed in many populations of forebrain axons, including corticothalamic and corticospinal axons⁴³. Further research into the specific populations of developing L1 neurons affected by PAE would garner a more thorough understanding into the etiology of PAE. It is unknown whether PAE’s effects on axon length at E15.5 would affect forebrain axon positioning and organization within the cortex in older mice. This is especially pertinent, as it has been hypothesized that organization of TCAs may first begin within the subpallium, long before the projections reach the cortex⁴⁴.

Our paradigm of PAE models a moderate level of maternal alcohol consumption during the late first trimester and early second trimester of human gestation. Alcohol consumption during this time may cause lasting behavioral effects in children. Consumption of at least one alcoholic beverage per day during the second trimester correlated with a significant decrease in spelling and math performance in children⁴⁵. Although moderate maternal alcohol consumption is enough to produce effects in children, public health campaigns often focus on more severe cases of FASD and higher levels of consumption. Despite questions surrounding the dangers of mild to moderate alcohol consumption during pregnancy, these results indicate that these levels maternal consumption can affect development. Pre-clinical studies using a full range of alcohol doses and timings are necessary to fully understand the impacts of drinking during pregnancy.

Supplementary Material

1

Supplemental Figure. Automatic quantification of L1-immunostained axons in E15.5 mouse embryos. Coronal brain sections were immunolabeled for L1. Left image shows L1-expressing axons in the region of interest (rostral section from a saline-exposed embryo). Middle image shows blue channel greyscale image of L1-expressing axons. Right image shows automated tracing by AxonTracer (yellow lines) superimposed on blue channel greyscale image which confirms accurate axon tracing. Scale bar, 100um

Highlights.

• Moderate prenatal alcohol exposure did not affect embryo weight at E15.5

• Moderate prenatal alcohol increases length of L1+ axons in embryonic mice

• Increased L1+ axon length not due to altered corridor cell count or distribution