Bad to the bone: New light on the brain’s venous system… and on craniosynostosis

cerebral veins and skull development in a normal child
Normal skull and brain venous development in a young child (courtesy Tischfield et al).

A recent study rocked the neuroscience world by demonstrating what in retrospect seems obvious: the brain has its own lymphatic system to help remove waste. A new study, from the laboratory of Elizabeth Engle, MD, at Boston Children’s Hospital, sheds light on another critical, little-studied part of the brain’s drainage system: the dural cerebral veins that remove and reabsorb excess cerebrospinal fluid.

The story of these vessels, the cover article in the next Developmental Cell, is a great example of lab scientists and physicians joining to make fundamental discoveries in biology. Strangely, critical clues come from children with craniosynostosis, a congenital malformation in which the skull plates fuse together too early in prenatal development, resulting in abnormal head shapes and, often, neurologic complications.

The skullbone-vein connection

abnormal cerebral veins in a child with craniosynostosis
A child with craniosynostosis and venous malformations.

Five years ago, Max Tischfield, PhD, was doing postdoctoral research with Jeremy Nathans at Johns Hopkins on the development of the blood-brain barrier and how blood vessels pattern themselves in the brain. He became interested in a gene called Twist1, part of the canonical Wnt developmental signaling pathway he was studying.

Reports in the literature suggested Twist1 might be involved in blood vessel development, but when Tischfield inactivated the gene in the blood vessels of mice, nothing happened.

But Tischfield also had mice in which Twist1 was inactivated in embryonic cells that give rise to a variety of tissues in the head, including bone. The resulting mice had malformed skull bones, but Tischfield noticed something else. “After carefully examining the heads of these mice, I began to realize that the venous vasculature was malformed,” he says.

He decided to try alternate ways of deleting Twist1 in osteoprogenitor cells, which arise from the embryonic mesoderm and build skull bone. These mice, too, had malformations in both the skull bones and the adjacent veins on the surface of the brain.

Tischfield was intrigued. The mutation didn’t affect arteries at all. Yet everywhere else in the body, veins and arteries are closely aligned.

“That suggests that the brain’s cerebral veins are patterned independently of the arteries,” he says. “But there was nothing in the literature about how the cerebral veins develop or originate.”

Insights from embryology

Tischfield returned to Engle’s lab, where he had conducted his graduate studies. Engle recognized the venous malformations in the mice as dural cerebral vein malformations. The observation led Tischfield to dig up a book chapter published in 1921. Renowned embryologist George Streeter had made detailed sketches of human brains at different stages of embryonic development — including the vasculature.

cerebral veins and arteries in a human embryo
Vascular system of the brain of the human embryo (Streeter 1921). Hill, M.A. 2017. Embryology Streeter-plate04.jpg. Retrieved from

Streeter suspected an embryologic link between the cerebral veins and the skull bone: “As the brain becomes more complicated, and as the skull-membranes form,” he wrote, “there occur, step by step, the necessary adaptations on the part of the blood vessels.”

Tischfield’s observations of cerebral vein development in mice mirrored Streeter’s observations in humans.

“Development of the cerebral veins is spatially and temporally coupled with skull development,” he reports. He believes that bone progenitors sense pressure from the growing brain, stimulating growth of skull bone and, along with it, the veins.

Craniosynostosis: Beware the vessels

Infant skull plates (Sobotta’s Atlas and Textbook of Human Anatomy 1909)

For decades, craniofacial and brain surgeons have been making observations of their own. Most children with craniosynostosis have surgery to release their fused skull plates, allowing the brain to grow. Pre-surgical imaging studies in these children often show tortuous cerebral veins taking unusual paths.

Tischfield knew from his reading that craniosynostosis, the most common craniofacial anomaly after cleft lip/cleft palate, has been associated with Twist1 mutations. Through Engle, he met with Boston Children’s neuroradiologist Caroline Robson, MB, ChB, who images many of the hospital’s craniosynostosis patients.

Tischfield asked Robson if she’d ever heard of venous abnormalities in craniosynostosis patients. “‘Have I?” she replied. “Yes! I co-authored a paper on it!”

Tischfield had missed the 2000 paper because it didn’t involve Twist. Instead, Robson studied 33 patients with mutations in the gene encoding fibroblast growth receptor (FGFR). All had craniosynostosis related to complex genetic syndromes such as Crouzon syndrome and Apert syndrome.

Images of the cerebral veins were available from 12 patients. Compared with healthy controls, their jugular veins, the major exit pathway for blood leaving the head, were markedly narrowed or even absent. The opposite was true for the emissary veins, which connect veins outside and inside the skull and are normally tiny.

Brain drainage: Bad to the bone

Tischfield and Engle teamed up with Robson to launch a parallel study in craniosynostosis patients with Twist mutations. (Their condition is known as Saethre-Chotzen syndrome).

Anatomy of the cerebral vessels (Max Tischfield)

Robson imaged 10 patients and found that the large dural veins, which drain into the jugular vein, were stunted or absent. These abnormalities had never been noted before, in part because no one looked for them and in part because they’re not easy to see on CT scans, especially in very young children, Robson says. For the study, she used a higher-resolution technique known as magnetic resonance venography.

Meanwhile, the genetics have helped Tischfield establish how skull cells influence the brain’s venous development. Twist1 and FGFR are implicated in many cases of craniosynostosis. So is BMP, the gene for bone morphogenic protein.

“With disruption of any of these genes, the brain can’t develop optimal venous networks or, in many cases, optimally absorb cerebrospinal fluid,” says Tischfield.

The BMP part of the story emerged from prior observations that zebrafish lacking BMP had malformations of their tail veins. When Tischfield inactivated BMP in the bone-forming tissues and blood vessels of mice, with help from experts at Harvard Dental School, he found skull and venous malformations much like those in the Twist1 mutants.

The team further showed that pre-osteoblasts — the actual bone-forming cells — normally secrete BMPs, and that the nearby veins have BMP receptors. But in mice with Twist1 mutations, pre-osteoblasts fail to develop from osteoprogenitor cells — and as a result, BMP is never made.

Three craniosynostosis genes. Twist1, produced by mesenchymal stem cells, is active in the osteoprogenitor cell phase. FGFR2 is thought to regulate proliferation and differentiation of osteoprogenitors. When the Twist1 gene is expressed, BMP2 and BMP4 are secreted by preosteoblasts and bind to receptors on the cerebral veins.

Relieving pressure

Until now, surgeons have viewed the venous abnormalities in craniosynostosis as an attempt to work around the skull malformations

“What this work points out is that the same gene mutations causing the skull problems are also causing malformations of the veins,” says Mark Proctor, MD, neurosurgeon-in-chief at Boston Children’s, who often operates on children with craniosynostosis and was a consultant on the study. “We had assumed that the veins had to find a different pathway because the skull bones were closed.”

We know a lot about arteries, but we can’t skip the drainage system.”Robson hopes the study will make physicians more aware of the cerebral vessels in children with craniosynostosis.

“Understanding the abnormalities of intracranial venous drainage is very important for recognizing the risk at surgery, because most patients will require surgery,” she says.

Study co-author Linda Dagi, MD, a Boston Children’s ophthalmologist who evaluates many children with craniosynostosis, sees implications for managing increased intracranial pressure (ICP). Elevated ICP is common in craniosynostosis and can cause severe headaches and put pressure on the optic nerve, ultimately compromising vision. Dagi’s eye exam is often the first line of detection.

The study suggests that even when patients have surgery to open up their fused skull bones, poorly formed cerebral veins could, by themselves, increase pressure on the brain by allowing cerebrospinal fluid (CSF) to build up. Robson and Proctor note that CSF cannot drain into the veins if the venous pressure is too high due to narrowing or other malformations.

Putting cerebral veins on the map

While that question has yet to be explored, the study lays a foundation for understanding a part of our brains that’s been largely overlooked for the past 100 years. Tischfield thinks his paired observations in mice and people will advance research by helping define what’s a true venous defect versus a normal variation in vessel patterning.

“Now we can argue what a malformation is,” Tischfield says. “We want to provide a handbook to help people understand how veins develop in the brain and how this fits with the lymphatic system. We know a lot about arteries, but we can’t skip the drainage system.”