Lymphatic vessels around the body drain excess fluid and protein from tissues and serve as a conduit for trafficking immune cells. The central nervous system (CNS) was long thought to lack lymphatic drainage; fluid and macromolecules were instead thought to be cleared from the CNS by other routes, such as through absorption into the bloodstream, or through channels running along the outside of blood vessels or nerves to reach the lymphatic system outside the CNS1,2. A few years ago, lymphatic vessels were discovered in the dura mater3,4, the outermost of the three meningeal membranes that envelop the CNS. They were reported to provide a route for immune cells trafficking from the CNS and for clearing CNS waste3,4, and to thus represent potential therapeutic targets for neurological diseases5,6. However, exactly how they might enable drainage from the brain has remained unclear, and whether they even have a role in drainage has been questioned1,2,7.
Writing in Nature, Ahn et al.8 now demonstrate that meningeal lymphatic vessels at the base of the rodent skull (basal mLVs) provide a direct route for the clearance of proteins and other large molecules from the CNS to the peripheral lymphatic system. They further characterize age-related changes in these vessels that impair their drainage function and that might contribute to ageing-associated neurological diseases.
Previous studies have shown that, in mice, mLVs grow in the first month after birth from the base of the skull along blood vessels and nerves9 into an intricate network that extends to the upper (dorsal) part of the skull3,4,9. Ahn et al. used fluorescence microscopy to characterize the morphology of the mLVs in the dorsal and basal skull8, and thus extend previous studies3,4,9. They showed that, unlike dorsal mLVs, basal mLVs possess specialized features associated with both uptake and drainage of fluid (Fig. 1). For example, the authors observed that the basal mLVs have tiny, blunt-ended capillary branches where the endothelial cells that make up the vessel walls are joined together loosely by intermittent, ‘button-like’ junctions, enabling fluid uptake. Basal mLVs, but not dorsal mLVs, also have ‘pre-collecting’ vessels that drain the capillaries and have valves inside them that enable flow in one direction3,4,8,9. The endothelial cells in the pre-collecting vessels are joined by both button-like junctions and continuous ‘zipper-like’ junctions, suggesting that they have a dual function of taking up and transporting fluid.
It is unclear whether and how cerebrospinal fluid (CSF) — which surrounds the brain and spinal cord and fills cavities in the brain known as ventricles — can cross the arachnoid layer (the middle meningeal membrane) to access the mLVs in the dura mater. Ahn et al. found that basal, but not dorsal, mLVs were close enough to the CSF-filled space between the innermost layer of the meninges and the arachnoid layer (known as the subarachnoid space) to be accessible for fluid entry. Through technically challenging analysis of the skull base around the openings where nerves and vessels exit the cranial cavity, the authors showed that the basal mLVs are distinct from the nerves along which CSF drainage was thought to occur7. Instead, the basal mLVs drain directly into collecting lymphatic vessels outside the CNS, which have previously been shown to transport molecules from the CSF to lymph nodes in the neck called deep cervical lymph nodes3–8.
Next, the authors assessed the drainage function of mLVs. They infused contrast agents into the CSF-filled subarachnoid space in rats and tracked these using magnetic resonance imaging as they moved along the lymphatic vessels exiting the base of the skull and into the deep cervical lymph nodes. Similarly, the researchers used microscopy analyses to track a fluorescently labelled tracer that had been infused into the CSF or into the interstitial fluid of the brain tissue in mice. In both cases, they detected the fluorescent tracer inside basal mLVs (both in capillaries and in pre-collecting vessels) and in deep cervical lymph nodes in the neck.
A study last year reported that dorsal mLVs in a few specific areas can take up CSF6. However, consistent with another previous study7, Ahn et al. were unable to detect evidence of uptake of the tracer from the CSF to dorsal mLVs. Overall, on the basis of the functional experiments, combined with the anatomical and morphological observations, the authors conclude that basal mLVs provide the main route for macromolecule uptake and for drainage of CSF and interstitial fluid from the brain directly into the peripheral lymphatic system.
The flow in lymphatic vessels that carry fluid drained from the CSF to lymph nodes in the periphery was previously reported to be slower in older compared with younger mice7, and such a decline in drainage function might have implications for age-related neurological diseases1,2,5. The authors therefore compared basal and dorsal mLVs in young mice (3 months old) with those in aged mice (24–27 months old). Whereas dorsal mLVs in aged mice showed deterioration, basal mLVs in aged mice were enlarged and more numerous. The basal mLVs in older mice also had fewer luminal valves compared with younger mice, and the junctions between the endothelial cells that form the vessel walls in older mice showed signs of disintegration. The authors confirmed that these age-related changes in basal-mLV morphology correlated with reductions in the drainage of macromolecules from the CSF in the aged mice.
A decline in mLV function has been suggested to lead to a build-up of proteins in the brain and to contribute to cognitive deficits and brain pathology in Alzheimer’s disease5. One way of counteracting age-dependent reductions in drainage function might be to stimulate mLV growth and to increase the diameter of mLVs. The endothelial cells that compose the mLVs in adult mice express the receptor VEGFR3, which is activated by the growth factor VEGF-C, and treating adult mice with VEGF-C induces growth and widening of mLVs4,9. The authors used a genetic manipulation to remove VEGFR3 from all lymphatic vessels, including mLVs, in adult mice. This approach revealed that, consistent with previous findings9, when VEGF-C signalling is lost, dorsal mLVs deteriorate more rapidly than do basal mLVs. However, it remains unclear whether VEGF-C–VEGFR3 signalling is affected in ageing and whether it could be targeted to counteract the observed ageing-associated changes in mLV function.
Besides clearing CNS macromolecules, mLVs also drain immune cells to lymph nodes4,6, where immune responses are initiated. Indeed, dorsal mLVs were previously identified on the basis that they contained immune cells4, and Ahn et al. also observed such cells in the basal mLVs. A previous study6 showed that disrupting dorsal mLVs attenuated inflammatory responses in a mouse model of the neurological disorder multiple sclerosis, indicating that mLVs might have a role in neuroinflammatory diseases. Future experiments should investigate whether, independently of their drainage function, mLVs might also promote immune tolerance (that is, a dampening of immune responses to recognized substances), as do the lymphatic vessels in lymph nodes10.
We still need a better understanding of the mechanisms that enable entry of fluid to the basal mLVs and of how mLVs cooperate with the other systems that clear waste from the CNS. Nevertheless, the identification of the precise exit routes for fluids leaving the brain is a crucial step towards understanding how waste is cleared from the CNS. This finding might eventually enable the development of therapies that promote CNS drainage to combat pathological processes in neurological diseases.
Nature 572, 34-35 (2019)
Louveau, A. et al. J. Clin. Invest. 127, 3210–3219 (2017).
Bower, N. I. & Hogan, B. M. J. Mol. Med. 96, 383–390 (2018).
Aspelund, A. et al. J. Exp. Med. 212, 991–999 (2015).
Louveau, A. et al. Nature 523, 337–341 (2015).
Da Mesquita, S. et al. Nature 560, 185–191 (2018).
Louveau, A. et al. Nature Neurosci. 21, 1380–1391 (2018).
Ma, Q., Ineichen, B. V., Detmar, M. & Proulx, S. T. Nature Commun. 8, 1434 (2017).
Ahn, J. H. et al. Nature 572, 62–66 (2019).
Antila, S. et al. J. Exp. Med. 214, 3645–3667 (2017).
Maisel, K., Sasso, M. S., Potin, L. & Swartz, M. A. Adv. Drug Deliv. Rev. 15, 43–59 (2017).
Baluk, P. et al. J. Exp. Med. 204, 2349–2362 (2007).