Neural tube formation is the process by which a flat sheet of cells on the back of the early embryo folds up and seals itself into the hollow tube that becomes the brain and spinal cord. In humans, this folding begins around the third week after conception and is mostly finished by about day 28, often before a person even knows they are pregnant. When the process fails at any point along the body, the result is a neural tube defect, one of the most common categories of serious birth defects worldwide. The biology behind this folding is remarkably intricate, and the consequences of even small disruptions help explain why prevention strategies like folic acid supplementation have become a cornerstone of prenatal care.
How a Flat Sheet Becomes a Tube
Early in development, a region of the embryo’s outer cell layer thickens into what is called the neural plate. This plate is essentially a patch of cells destined to become the entire central nervous system. For it to become a tube, the plate has to bend along its midline, raise its edges upward like two walls, and then bring those walls together until they fuse at the top. The whole sequence is called primary neurulation, and it depends on coordinated forces generated by the cells themselves.
The bending starts when certain cells along the midline and at specific lateral positions constrict their top surfaces, making them wedge-shaped. This squeezing at the top, called apical constriction, creates hinge points where the tissue buckles inward. Research in frog embryos has shown that a transcription factor called Mrtfa drives the assembly of the contractile protein machinery needed for this squeezing; when Mrtfa is knocked down, cells fail to constrict and the neural plate does not fold properly.1PubMed Central. Myocardin-related transcription factor regulates actomyosin contractility and apical junction remodeling during vertebrate neural tube closure Not every cell constricts at once. Studies of the frog neural plate reveal an alternating pattern of tightly constricted and expanded cells, creating a kind of tug-of-war that helps the tissue bend without shrinking along the head-to-tail axis.2PubMed Central. Mechanical control of neural plate folding by apical domain alteration
A separate but equally important force is convergent extension, a process in which cells rearrange themselves to make the tissue longer and narrower. This narrowing brings the two edges of the neural plate closer together so they can meet. Both apical constriction and convergent extension are coordinated by the same set of molecular signals, known as the planar cell polarity pathway. In mouse embryos, mutations in genes belonging to this pathway prevent the neural plate from narrowing, leaving the neural folds too far apart to fuse.3PubMed Central. Convergent extension, planar-cell-polarity signalling and initiation of mouse neural tube closure Work in the same system has shown that the polarity pathway drives directional contraction of cell-to-cell junctions, simultaneously promoting the squeezing and the midline convergence needed for closure.4PubMed. Planar cell polarity links axes of spatial dynamics in neural-tube closure When different components of this pathway are disrupted, different parts of the cellular machinery break down: one mutation may block cell intercalation at the base of the tissue while another blocks neighbor exchange near the surface, but either failure alone can prevent closure.5PubMed Central. Distinct apical and basolateral mechanisms drive planar cell polarity-dependent convergent extension of the mouse neural plate
Where and When Closure Happens in Humans
A common simplification is that the neural tube zips shut from the middle outward like a jacket. The reality, at least in humans, is that closure starts at multiple sites. Studies of human embryos have identified at least three initiation points. The first (Site A) is in the future neck region and is widely accepted as the primary starting point. A second site (Site B) is at the boundary between the midbrain and the hindbrain, and a third (Site C) is at the very front of the neural groove.6PubMed. Neural tube closure in humans initiates at multiple sites: evidence from human embryos and implications for the pathogenesis of neural tube defects Closure from Site B extends in both directions: toward the back of the head where it meets the wave from Site A, and toward the front where it meets closure coming from Site C. The last openings to seal are called the anterior neuropore (at the brain end, closing around day 25) and the posterior neuropore (at the tail end, closing around day 28).
This multi-site model matters because it explains the geography of neural tube defects. If the closure wave from Site B fails, you can end up with a defect specifically in the midbrain or upper hindbrain region. If the anterior neuropore never seals, the forebrain is left exposed, leading to anencephaly. If the posterior neuropore stays open, the result is spina bifida in the lower back. The pattern of closure sites predicts which stretches of the nervous system are most vulnerable.
The Tail End Works Differently
Everything described so far applies to primary neurulation, which builds the neural tube from the brain down through most of the spinal cord. But the very lowest part of the spinal cord, roughly corresponding to the sacral and coccygeal segments, forms through a separate process called secondary neurulation. Instead of folding a sheet of surface cells, the embryo starts with a clump of mesenchymal cells in the tail bud. These cells undergo a transition back into an organized epithelial state, line up into a solid rod called the medullary cord, and then hollow out from the inside to create a lumen that becomes continuous with the tube above.7PubMed Central. Cell Lineage, Self-Renewal, and Epithelial-to-Mesenchymal Transition during Secondary Neurulation8PubMed. In Vivo Analysis of the Mesenchymal-to-Epithelial Transition During Chick Secondary Neurulation
Defects of secondary neurulation tend to produce a different set of problems than those of primary neurulation. Rather than the open lesions seen in classic spina bifida, problems during secondary neurulation can lead to closed spinal defects, where the skin is intact but the underlying spinal cord is abnormally tethered or contains fatty tissue. These closed defects often have a subtler presentation and can be harder to detect prenatally.
What Goes Wrong in Neural Tube Defects
Neural tube defects are among the most common congenital malformations of the central nervous system. They range widely in severity. Anencephaly, where much of the brain and skull fail to form, is incompatible with life. Craniorachischisis, in which the neural tube remains open from the brain through much of the spine, is similarly lethal.9PubMed Central. Lethal Neural Tube Defects: Reports of Anencephaly and Craniorachischisis Cases and Literature Review Spina bifida, particularly the form called myelomeningocele where the spinal cord and its coverings protrude through an opening in the spine, is survivable with treatment but can lead to lifelong challenges with mobility, bladder control, and sensation below the level of the defect.
Closed neural tube defects, where the skin covers the lesion, encompass conditions like lipomyelomeningocele and myelocystocele. Even though the skin is intact, the underlying spinal cord can be tethered or compressed. Research on these closed defects has found that the presence of a fluid-filled sac worsens outcomes: children with a visible sac are less likely to walk independently by age five and more often need catheterization for bladder management compared to those without one.10PubMed Central. Clinical impact of fetal sac size on closed neural tube defects Bladder dysfunction is one of the most persistent complications of spina bifida regardless of type, and treating it aggressively from birth is considered essential to prevent kidney damage over time.
Why Folate Matters So Much
The link between folic acid and neural tube defects is one of the most successful stories in preventive medicine, though the underlying biology is still not completely settled. The prevailing explanation centers on one-carbon metabolism, the biochemical cycle that uses folate to shuttle single-carbon groups needed for DNA synthesis, repair, and methylation reactions. The hypothesis is that inadequate folate disrupts methylation, which in turn disrupts the gene regulation needed for proper neural tube closure.11PubMed Central. Neural tube defects and folate: case far from closed
The picture is more nuanced than “folate feeds methylation and methylation closes the tube.” Studies in mouse embryos have shown that both the folate cycle and the methionine cycle (which handles methylation) need to be active for normal closure, but the transfer of one-carbon units between the two cycles is not strictly necessary. In genetically susceptible mice, blocking the transfer between cycles by deleting the enzyme MTHFR actually rescued neural tube closure by keeping one-carbon units available within the folate cycle itself.12PubMed Central. Partitioning of One-Carbon Units in Folate and Methionine Metabolism Is Essential for Neural Tube Closure This finding suggests that the balance of carbon flow between the two cycles matters more than simply having enough methylation. It also helps explain why the mechanism behind folate’s protective effect has remained elusive for decades.
Regardless of the precise mechanism, the population-level results are clear. Countries that have mandated the addition of folic acid to grain products have seen reductions of roughly 25 to 50 percent in the prevalence of neural tube defects compared to countries without fortification.13PubMed Central. Global heterogeneity in folic acid fortification policies and implications for prevention of neural tube defects and stroke Some individual programs have reported even steeper drops; a review of fortification initiatives across 71 countries documented reductions of up to 78 percent in some settings.14PubMed Central. Folic Acid and the Prevention of Birth Defects: 30 Years of Opportunity and Controversies In Brazil, the prevalence of neural tube defects dropped from about 0.79 per thousand births to 0.55 per thousand after mandatory flour fortification was introduced.15PubMed. Prevention of neural tube defects by the fortification of flour with folic acid: a population-based retrospective study in Brazil These numbers underscore why public health agencies worldwide recommend that anyone capable of becoming pregnant consume supplemental folic acid, ideally starting before conception.
Genetic and Environmental Risk Factors Beyond Folate
Folate status is the best-known modifiable risk factor, but neural tube defects are genuinely multifactorial. Genetic variation in the MTHFR gene, which encodes the enzyme that channels one-carbon units from the folate cycle into methylation, is the most studied genetic contributor. The C677T variant of MTHFR reduces the enzyme’s stability, and a meta-analysis of 25 case-control studies found that mothers carrying two copies of this variant had roughly twice the odds of having a child with a neural tube defect compared to mothers without it.16PLoS ONE. Association of the Maternal MTHFR C677T Polymorphism with Susceptibility to Neural Tube Defects in Offsprings: Evidence from 25 Case-Control Studies A second variant in the same gene, 1298A→C, also increases risk, though to a lesser degree.17PubMed Central. Methylenetetrahydrofolate reductase mutations, a genetic cause for familial recurrent neural tube defects These variants are common in many populations and are not deterministic on their own; they raise risk on a background of other genetic and dietary factors.
On the environmental side, maternal diabetes is a well-established risk factor. High blood sugar appears to trigger oxidative stress and abnormal cell death in the developing neural tube. Research in mouse models has traced one pathway from high glucose to the activation of an enzyme called ASK1, which triggers a cell-death cascade that can prevent proper tube closure.18PubMed Central. Maternal hyperglycemia activates an ASK1-FoxO3a-caspase 8 pathway that leads to embryonic neural tube defects19Journal of Diabetes, Metabolism and its Complications. Potential Mechanisms Involved in Diabetic Teratogenesis During Embryonic Neural Tube Closure Certain medications also carry risk. Valproic acid, an anti-seizure drug, is associated with roughly a two-and-a-half-fold increase in major malformations compared to other anti-epileptic drugs, and neural tube defects are among the most characteristic of those malformations.20PubMed Central. Major malformations with valproic acid Obesity, high fever in early pregnancy, and low socioeconomic status have also been linked to increased risk, though the mechanisms are less clearly defined.
How Neural Tube Defects Are Detected Before Birth
For decades, the standard screening method was a blood test measuring maternal serum alpha-fetoprotein (MSAFP) during the second trimester, typically between weeks 15 and 18. Alpha-fetoprotein is a fetal protein that leaks into the mother’s blood at higher levels when the fetal skin is not intact, as in open neural tube defects. Early large-scale studies found that MSAFP screening could detect the vast majority of anencephaly cases and roughly four out of five spina bifida cases.21PubMed. Maternal serum AFP screening
Over the past two decades, however, high-resolution ultrasound has largely taken over as the primary screening tool. One study found that among nearly 200 neural tube defect cases, more than 60 percent were first detected by routine second-trimester ultrasound before any blood-test results were available. Of the cases that had received MSAFP screening, a quarter came back negative, meaning those diagnoses would have been missed without ultrasound.22PubMed. Routine ultrasonography compared with maternal serum alpha-fetoprotein for neural tube defect screening Ultrasound is especially useful for detecting closed defects, which do not elevate alpha-fetoprotein because the skin is intact.23PubMed Central. Prenatal screening for neural tube defects: from maternal serum alpha-fetoprotein to ultrasonography Many practices now use a combination of ultrasound and, where available, MSAFP as a complementary test.
Prenatal Surgery for Myelomeningocele
One of the most dramatic developments in treating neural tube defects has been fetal surgery for myelomeningocele. The rationale is straightforward: if you can close the spinal defect before birth, you may reduce the secondary damage caused by prolonged exposure of the spinal cord to amniotic fluid. A landmark randomized trial (known as the MOMS trial) compared prenatal repair performed before 26 weeks of pregnancy to standard postnatal surgery. The results were striking: only 40 percent of babies in the prenatal surgery group needed a shunt to drain excess cerebrospinal fluid, compared to 82 percent in the postnatal group. Prenatal repair also improved motor outcomes and reduced the severity of hindbrain herniation, a common complication where the lower brain is pulled downward into the spinal canal.24PubMed Central. A randomized trial of prenatal versus postnatal repair of myelomeningocele
Follow-up data have reinforced these benefits. A subsequent analysis of the same trial cohort found that children who had prenatal repair were more likely to be community ambulators, meaning they could walk functionally in everyday life, at roughly 51 percent compared to 23 percent in the postnatal group. They also performed better on timed walking tests and higher-level mobility tasks.25JAMA Pediatrics. Prenatal Repair and Physical Functioning Among Children With Myelomeningocele: A Secondary Analysis of a Randomized Clinical Trial A systematic review of studies published after the original trial confirmed these patterns: prenatal repair cuts shunt placement by about half and doubles the rate of independent walking, with hindbrain herniation reversing in up to three-quarters of prenatal cases.26PubMed. Fetal myelomeningocele repair after the MOMS trial: a systematic review of neurosurgical outcomes and evolving techniques
The tradeoff is real, though. Prenatal surgery carries a significant risk of preterm delivery (babies arrived around 34 weeks on average), premature rupture of membranes, and weakening of the uterine wall at the surgical site. These risks mean the decision is a careful weighing of benefit to the fetus against risk to both the fetus and the mother, and the procedure is performed only at specialized centers.
Research Models and the Push Toward Organoids
Much of what we know about neural tube closure comes from animal models, especially mice. Over the years, researchers have identified dozens of mouse mutations that cause neural tube defects, and several of these have proven responsive to folic acid supplementation, mirroring the human experience. Models like the splotch mouse (which has a mutation in the Pax3 gene) and the crooked tail mouse develop neural tube defects at high rates that can be reduced by giving the mothers extra folic acid.27PubMed. Mouse models of neural tube defects: investigating preventive mechanisms More recent work with the Zic2 mutant mouse, which develops severe spina bifida, has revealed that at least two independent signaling pathways (BMP and RhoA) need to go wrong simultaneously to prevent closure, suggesting that spina bifida in mammals is fundamentally a multi-pathway problem.28PubMed Central. Dual mechanism underlying failure of neural tube closure in the Zic2 mutant mouse
A limitation of animal models is that mouse neurulation differs from human neurulation in the number and location of closure initiation sites, and not every mouse finding translates directly. This has spurred interest in human neural tube organoids, miniature tube-like structures grown from human stem cells in the lab. Recent work has produced organoids that develop the kind of top-to-bottom patterning seen in a real embryonic spinal cord and, at the molecular level, resemble a human embryo at roughly Carnegie stage 12 (about four to five weeks post-fertilization) more closely than they resemble mouse tissue.29PubMed. Establishing dorsal-ventral patterning in human neural tube organoids with synthetic organizers Researchers have already used these organoids to model patterning defects caused by mutations in the gene TCTN2 and to test potential rescue strategies. Other groups have shown that applying mechanical forces to human stem-cell-derived organoids enhances the patterning and growth of the resulting neural tube structures, reinforcing the idea that physical forces and molecular signals work hand in hand during real closure.30PubMed Central. Actuation enhances patterning in human neural tube organoids
Epigenetic Layers of Complexity
Beyond straightforward genetic mutations, there is growing recognition that epigenetic changes play a role in neural tube defects. Epigenetic modifications are chemical tags added to DNA or to the histone proteins that package it, and they influence which genes get turned on or off without changing the DNA sequence. Studies of neural tissue, placental tissue, and blood cells from affected pregnancies have identified differences in DNA methylation patterns, histone modifications, and small regulatory RNA molecules called microRNAs. These changes can alter the expression of genes critical for neural tube closure, and they may help explain why many neural tube defect cases occur without an obvious genetic mutation or nutritional deficiency. Because folate metabolism directly feeds the methylation cycle, there is a plausible loop: insufficient folate could produce epigenetic changes that impair closure, while certain epigenetic states could make embryos more vulnerable to low folate in the first place. Untangling these layers is an active area of research, and it is likely that multi-omics approaches combining genomic, epigenomic, and transcriptomic data will reshape our understanding of neural tube defect risk in the coming years.
Living With Spina Bifida
For the thousands of children born with spina bifida each year, management extends far beyond the initial surgical closure. Bladder dysfunction is almost universal, and renal damage from poorly managed neurogenic bladder remains one of the most serious long-term threats. Current practice emphasizes treating bladder overactivity from birth rather than waiting for problems to develop, because damage to the upper urinary tract can begin in the first months of life. Most children with myelomeningocele eventually use clean intermittent catheterization to empty the bladder, and many achieve social continence by school age when treatment is started early. Mobility varies enormously depending on the level of the spinal lesion: children with lower sacral defects often walk independently, while those with higher lumbar or thoracic lesions typically rely on wheelchairs. Advances in prenatal repair have shifted this distribution somewhat, increasing the proportion of children who can walk functionally, but the condition still requires coordinated care across neurosurgery, urology, orthopedics, and rehabilitation throughout life.