Mapunda JA, Pareja J, Vladymyrov M, Bouillet E, Hélie P, Pleskač P, Barcos S, Andrae J, Vestweber D, McDonald DM, Betsholtz C, Deutsch U, Proulx ST, Engelhardt B. VE-cadherin in arachnoid and pia mater cells serves as a suitable landmark for in vivo imaging of CNS immune surveillance and inflammation. Nat Commun. 2023 Sep 20;14(1):5837. PubMed.

Recommends

Please login to recommend the paper.

Comments

  1. The study from Britta Englehardt’s group provides a very clear analysis of the morphological, molecular, and functional characteristics of the arachnoid. It also challenges the recently described impermeable SLYM. The present study convincingly shows that Prox1-tdTomato+ cells are located within the VE-cadherin layers of inner arachnoid and they did not form a continuous layer. The present study uses double reporter mice (VE-cadherin-GFP crossed with Prox1-tdTomato mice), allowing thus a very clear and accurate analysis in vivo as well as ex vivo.

    Using transcriptomics as well as electron microscopy, Pietilä et al. provide further evidence that SLYM is not molecularly and morphologically defined as a different entity.

    These two studies are indeed contradicting the existence of SLYM. Together, using a very convincing state-of-the-art spectrum of techniques, they are very convincing. The significance of the studies is multiple, in the context of the fact that the leptomeninges are important in drug delivery to the brain as well as in the pathogenesis of dilated perivascular spaces, a major feature of vascular cognitive disorders and even amyloid related imaging abnormalities (ARIA).

    View all comments by Roxana Carare

  2. These two studies add new valuable insight regarding the molecular and functional organization of the CNS meninges—and lay ground for further debate and controversy.

    Since the discovery in 2015 of functional meningeal lymphatic vessels capable of draining substances from cerebrospinal fluid (CSF) (Louveau et al., 2015; Aspelund et al., 2015), emerging evidence suggests a crucial role of meninges in CNS immune surveillance and clearance of substances from subarachnoid CSF (Da Mesquita, 2022; Da Mesquita et al., 2018; Da Mesquita et al., 2021; Cugurra et al., 2021; Mazzitelli et al., 2022; Ding et al., 2021; Rustenhoven and Kipnis, 2022). The existence of two-way passage of cells and substances between CSF of the subarachnoid space and dura mater/skull bone marrow obviously has major implications for our understanding of normal CNS function, and for evolvement of diseases such as neurodegenerative disease, neuro-inflammatory disease, or consequences of traumatic brain injury or stroke.

    The two reports touch upon important research questions that need to be resolved, namely relating to the functional organization of the subarachnoid space, and the barrier function of the arachnoid membrane, primarily the outer arachnoid barrier cell layer.

    In the pioneering work by Pietilä and collaborators, the authors challenge the traditional, rather simplistic view of the meningeal organization; here, they report on six different single-cell transcriptomic variants of meningeal fibroblasts (type 5 in dura border cells, type 4 in arachnoid barrier cells, types 3-2 in inner arachnoid cells, type 1 in pia, and a sixth type in parenchymal perivascular fibroblasts). This molecular differentiation of fibroblasts in the various cell layers suggests a complex functional organization of the meninges. These results should inspire further work into molecular organization of the arachnoid, as the output might have obvious clinical relevance, given the role of fibroblasts in fibrotic reactions after inflammation and trauma. Other possible areas include events following subarachnoid hemorrhage and neuroinflammation associated with tumors or neurodegeneration. 

    Earlier this year, a fourth meningeal layer between the pia and outer arachnoid barrier cell layer, referred to as the subarachnoid lymphatic-like membrane (SLYM), was reported (Møllgård et al., 2023). With their experimental setup, Pietilä et al. found no evidence for this fourth membrane; they rather suggest their observations support the concept of “a single SAS bordered by pial cells on the brain side and by the arachnoid on the skull side.” This view compares with a traditional concept that the subarachnoid space is one compartment with no directionality of CSF flow. In my view, the authors are wrong at this point, and I don’t agree that their findings justify such a generalized view. Further research using a variety of methodologies, including both the macro- and micro-perspective, is required. The subarachnoid space is functionally compartmentalized. However, to which degree the previously reported SLYM captures this needs to be further explored.  

    Regarding the barrier function of the arachnoid, Pietilä et al. report that the arachnoid barrier cell layer consists of a two-layer cell arrangement where cells connect by tight junctions, adherens junction and tricellular junctions. Such an organization might give an anatomical basis for an impermeable arachnoid membrane. A traditional view states that the arachnoid is impermeable to CSF causing the CSF to be contained within the subarachnoid space (Weller et al., 2018), because cells of the outer barrier layer of the arachnoid are joined by tight junctions that were considered to establish a barrier to the passage of CSF out of the subarachnoid space.

    The study by Mapunda et al. supports this traditional view. They boldly state that the recent studies reporting on two-way passage of cells and substances between the subarachnoid space and dura mater/skull bone marrow have “omitted consideration of the barrier properties of the arachnoid mater, which forms a barrier between the dura mater and the SAS.” In support of their view, they used two-photon imaging in vascular endothelial (VE) cadherin-green fluorescent protein (GFP) knock-in mice to explore the barrier properties of arachnoid and pia to substances and cells. Cadherin is a cell adhesion molecule enabling cells to adhere. They identified a VE-cadherin-GFP+ cellular layer both just beneath the dura and at a location corresponding to the pia, and in trabecula crossing the subarachnoid space. From this, they claim that the previously reported SLYM (Møllgård et al., 2023), characterized by positive PROX-1 expressing meningeal cells, is instead part of the inner arachnoid cell layer without any barrier function, and state that the lack of simultaneous visualization of arachnoid membrane with PROX-1 expression caused a wrong interpretation of a fourth meningeal layer.

    All experimental setups have limitations, including the setup of Mapunda et al.; caution should therefore be made when criticizing previous reports about crosstalk of cells and substances between CSF and dura mater/skull bone marrow across the arachnoid. However, given the possible implications, this is a field for further intensive research.

    Lastly, Mapunda et al. found that the arachnoid barrier layer was impermeable to both low molecular weight molecules (tracers 3kDa) and large molecular weight tracers (40-66 kDa tracers), leading the authors to conclude that the outer arachnoid barrier layer due to its tight barrier properties is impermeable also to low molecular weight substances. In my view, this conclusion underlines the need for addressing research questions from different perspectives, including the macro perspective.

    In this regard, we have sstudied how a CSF tracer, the magnetic resonance imaging (MRI) contrast agent gadobutrol (a hydrophilic 604 molecular weight substance), injected to the intrathecal space enriches the subarachnoid space and passes further to the inside of the dura nearby the superior sagittal sinus, denoted the parasagittal dura (PSD) (Ringstad and Eide, 2020), and even passes to the skull bone marrow (Ringstad and Eide, 2022) and enriches extracranial lymph nodes (Eide et al., 2018). As can be seen in Figure 1 to the left, the parasagittal dura (PSD; yellow color) is irregular. We further showed that the volume of PSD shows a high degree of inter-individual variation (Melin et al., 2023). These observations clearly show that the arachnoid is not impermeable to substances in the CSF, at least not to low molecular weight substances. Therefore, human tracer studies do not support the traditional view of an impermeable arachnoid membrane. However, given the irregular form of PSD, the anatomical region of dura mater studied in experimental setups would have major impact on the observations.

     

    Figure. 1. In humans, a CSF tracer injected to the intrathecal space enriches the subarachnoid CSF space, as well as the dura mater itself nearby the superior sagittal sinus, denoted the parasagittal dura (PSD), demonstrating direct passage from the CSF (shown as turquoise color in the image to the left), via arachnoid to the dura mater (yellow color). As indicated in the graph to the right, enrichment of the PSD starts after a few hours and peaks after several hours (about 24 hours). The CSF tracer is an MRI contrast agent, gadobutrol, which is a hydrophilic 604 MW molecule. [Courtesy of Ringstad and Eide, Nature Communications, 2020.] 

    References:

    Louveau A, Smirnov I, Keyes TJ, Eccles JD, Rouhani SJ, Peske JD, Derecki NC, Castle D, Mandell JW, Lee KS, Harris TH, Kipnis J. Structural and functional features of central nervous system lymphatic vessels. Nature. 2015 Jul 16;523(7560):337-41. Epub 2015 Jun 1 PubMed.

    Aspelund A, Antila S, Proulx ST, Karlsen TV, Karaman S, Detmar M, Wiig H, Alitalo K. A dural lymphatic vascular system that drains brain interstitial fluid and macromolecules. J Exp Med. 2015 Jun 29;212(7):991-9. Epub 2015 Jun 15 PubMed.

    Da Mesquita S. Charting the meningeal lymphatic network. J Exp Med. 2022 Aug 1;219(8) Epub 2022 Jul 5 PubMed.

    Da Mesquita S, Louveau A, Vaccari A, Smirnov I, Cornelison RC, Kingsmore KM, Contarino C, Onengut-Gumuscu S, Farber E, Raper D, Viar KE, Powell RD, Baker W, Dabhi N, Bai R, Cao R, Hu S, Rich SS, Munson JM, Lopes MB, Overall CC, Acton ST, Kipnis J. Publisher Correction: Functional aspects of meningeal lymphatics in ageing and Alzheimer's disease. Nature. 2018 Dec;564(7734):E7. PubMed.

    Da Mesquita S, Papadopoulos Z, Dykstra T, Brase L, Farias FG, Wall M, Jiang H, Kodira CD, de Lima KA, Herz J, Louveau A, Goldman DH, Salvador AF, Onengut-Gumuscu S, Farber E, Dabhi N, Kennedy T, Milam MG, Baker W, Smirnov I, Rich SS, Dominantly Inherited Alzheimer Network, Benitez BA, Karch CM, Perrin RJ, Farlow M, Chhatwal JP, Holtzman DM, Cruchaga C, Harari O, Kipnis J. Meningeal lymphatics affect microglia responses and anti-Aβ immunotherapy. Nature. 2021 May;593(7858):255-260. Epub 2021 Apr 28 PubMed.

    Cugurra A, Mamuladze T, Rustenhoven J, Dykstra T, Beroshvili G, Greenberg ZJ, Baker W, Papadopoulos Z, Drieu A, Blackburn S, Kanamori M, Brioschi S, Herz J, Schuettpelz LG, Colonna M, Smirnov I, Kipnis J. Skull and vertebral bone marrow are myeloid cell reservoirs for the meninges and CNS parenchyma. Science. 2021 Jun 3; PubMed.

    Mazzitelli JA, Smyth LC, Cross KA, Dykstra T, Sun J, Du S, Mamuladze T, Smirnov I, Rustenhoven J, Kipnis J. Cerebrospinal fluid regulates skull bone marrow niches via direct access through dural channels. Nat Neurosci. 2022 May;25(5):555-560. Epub 2022 Mar 17 PubMed.

    Ding XB, Wang XX, Xia DH, Liu H, Tian HY, Fu Y, Chen YK, Qin C, Wang JQ, Xiang Z, Zhang ZX, Cao QC, Wang W, Li JY, Wu E, Tang BS, Ma MM, Teng JF, Wang XJ. Impaired meningeal lymphatic drainage in patients with idiopathic Parkinson's disease. Nat Med. 2021 Mar;27(3):411-418. Epub 2021 Jan 18 PubMed.

    Rustenhoven J, Kipnis J. Brain borders at the central stage of neuroimmunology. Nature. 2022 Dec;612(7940):417-429. Epub 2022 Dec 14 PubMed.

    Møllgård K, Beinlich FR, Kusk P, Miyakoshi LM, Delle C, Plá V, Hauglund NL, Esmail T, Rasmussen MK, Gomolka RS, Mori Y, Nedergaard M. A mesothelium divides the subarachnoid space into functional compartments. Science. 2023 Jan 6;379(6627):84-88. Epub 2023 Jan 5 PubMed.

    Weller RO, Sharp MM, Christodoulides M, Carare RO, Møllgård K. The meninges as barriers and facilitators for the movement of fluid, cells and pathogens related to the rodent and human CNS. Acta Neuropathol. 2018 Mar;135(3):363-385. Epub 2018 Jan 24 PubMed.

    Ringstad G, Eide PK. Cerebrospinal fluid tracer efflux to parasagittal dura in humans. Nat Commun. 2020 Jan 17;11(1):354. PubMed.

    Ringstad G, Eide PK. Molecular trans-dural efflux to skull bone marrow in humans with CSF disorders. Brain. 2022 May 24;145(4):1464-1472. PubMed.

    Eide PK, Vatnehol SA, Emblem KE, Ringstad G. Magnetic resonance imaging provides evidence of glymphatic drainage from human brain to cervical lymph nodes. Sci Rep. 2018 May 8;8(1):7194. PubMed.

    Melin E, Ringstad G, Valnes LM, Eide PK. Human parasagittal dura is a potential neuroimmune interface. Commun Biol. 2023 Mar 11;6(1):260. PubMed.

    View all comments by Per Kristian Eide

  3. Both these papers were submitted prior to the Science publication by Dr. Maiken Naadergard’s group reporting the fourth layer; I expect that both papers were in active revision. Therefore, the authors took an opportunity to address some of the controversial aspects of the Science paper with the tools and data that they uniquely had at their disposal due to their extensive ongoing work in this area. Therefore, these papers are not "reactive" but rather asked the question, “Can we find evidence in our ongoing studies that supports or does not support the claims of a functionally distinct fourth layer?”

    Building off the molecular marker Prox1 reported in the Science study and using some of the same technical approaches (Prox1-RFP, 2-photon live imaging, histology), both papers studied the identity and positioning of the Prox1+ layer, identified as a defining characteristic of the fourth layer. They conclude that Prox1+ cells are part of what is historically considered the arachnoid mater. They were unable, using 2-photon in vivo live imaging, electron microscopy and immunofluorescence, to detect a separate space created by Prox1+ cells reported in the Science paper.

    Notably, the single cell profiling by the Betsholtz’s group shows that tight junction expression and accessory proteins is limited to arachnoid barrier cells, supporting that arachnoid barrier cells are most likely the only meningeal cell layer with size-restrictive barrier properties.

    The Engelhardt group’s paper shows Prox1+ cells express adherens junction protein VE-Cadherin; however, pial cells also express VE-cadherin, form a near continuous sheet connected by adherens junctions yet the pial layer lacks size restrictive barrier properties. Collectively, these two papers provide important information about the Prox1+ arachnoid cell layer that allows scientists both in and outside the field to better evaluate the validity of the conclusions of the paper reporting a fourth meningeal layer.

    While the authors of these two papers were unable to find evidence of CSF-containing space between the Prox1+ layer and the arachnoid barrier layer, the experiment that would test this definitively is combined use of transgenic mouse lines that independently mark the E-cadherin+ arachnoid barrier and Prox1 layer for use in 2-photon studies, also ideally the dura. Further, combining tracer studies with transmission electron microscopy and immunogold to detect Prox1+ cells would permit high-resolution detection of where exactly CSF moves in or around cell layers and spaces in the leptomeninges, including Prox1+ cells. I think to better understand the functional relevance of Prox1+ arachnoid layer, or any meningeal layer, ablation studies to disrupt the layers would be ideal. 

    Both papers take different technical approaches to molecularly and functionally define the leptomeninges, an understudied structure with clear functional roles in CNS development, adult CNS function and a well-known site of neuroinflammation and barrier breakdown in disease states. The impact of these papers for the meninges biology field goes far beyond addressing recent controversies. In the paper by the Engelhardt group, the authors illustrate the utility of transgenic mouse lines to visualize meningeal cell layers and the cellular boundaries created by different layers. Their findings generated important new knowledge on immune cell surveillance, in particular around the pia, and movement of molecules in the leptomeninges. In the paper by the Betsholtz group, not only does this provide unprecedented insight into the molecular identity of leptomeningeal fibroblasts but also their positioning within the leptomeninges, including important new data on the organization of tricellular junctions in arachnoid barrier layer. The tools, datasets and knowledge gained from these two studies will be vital for advancing studies on functional relevance of meningeal layer subtypes.

    As I outlined in my rebuttal to the Science paper soon after it was published, I strongly suspected at the time that Prox1+ cells were most likely inner arachnoid cells based on extensive histological and electron microscopy data of leptomeninges cellular architecture that show the arachnoid has multiple cell layers, all connected by a variety of junctions (tight, adherens, gap). I had major doubts about this layer having barrier properties on its own, mostly due to the lack of tight junctions protein expression. But at the time, the characterization of Prox1-expressing cells was very limited. Now we have a lot more information because of these papers by the Engelhardt and Betsholtz groups and a much better picture of the molecular and cellular anatomy of the leptomeninges. I am now confident in my original suspicion that the Prox1+ cell layer is part of what is historically considered the arachnoid mater and lacks the functional components to be a barrier layer. I am also confident in my opinion that there are not large, previously unknown spaces in the leptomeninges nor a bifurcation of the subarachnoid space.

    That said, there is much more to learn about meninges structure and function. There is a lot more to learn about how the meningeal layers, defined both as the three main layers and as sublayers of the arachnoid and dura, are compartmentalized and structured to control CSF flow and exit from the CNS. All of these studies have stimulated a lot of interest in this important structure, and, hopefully, there will be more groups engaging in the meninges biology field. 

    View all comments by Julie Siegenthaler

  4. The meninges, a multilayered membrane structure that covers the surface of the brain and spinal cord, acts as a barrier to prevent uncontrolled molecular exchange and contributes to immune surveillance of the central nervous system. However, the cellular and molecular composition of the meningeal layers and their accessibility to immune cells are not well understood. These two studies took different approaches to shed light on (1) the anatomical structure and barrier properties of brain and spinal cord meningeal layers to CSF prefusion and immune cell infiltration using in vivo imaging in VE-cadherin GFP reporter mice (Mapunda et al.) and (2) the identification of transcriptionally distinct fibroblasts and their cellular localization in the individual layers of the meninges (R. Pietilä et al.).

    Traditionally, the meninges are thought to consist of three layers, the dura, arachnoid, and pia matter. Interestingly, both papers challenge the claims of a recently described fourth meningeal layer consisting of Prox1-positive cells subdividing the arachnoid space, described by Møllgård et al. Mapunda et al. provide evidence using imaging of VE-cadherin/Prox1 double reporter mice that Prox1-positive cells do not form a separate barrier within the subarachnoid space, but in fact are an integral and continuous part of the arachnoid mater. Pietilä et al. found transcriptomic and in situ hybridization evidence of tight and adherens junctions between the cells of the inner arachnoid layer, consistent with Mapunda et al.’s observations of tight and adherins junctions in VE-cadherin reporter mice. Pietilä et al. also explored the location of Prox1 mRNA, finding it specifically expressed in a subset of fibroblast-like cells (BFB3). Both studies mapped the location of Prox1-positive cells to the inner arachnoid layer. These observations argue against the existence of a fourth meningeal layer, and instead suggest a single coherent arachnoid cellular layer.

    These studies highlight that the anatomy and function of the meningeal layers may be highly complex, and do not rule out the possibility of structural, functional, or transcriptomic changes in disease or inflammatory conditions. Indeed, Mapunda et al. observed changes in the depth of the subarachnoid space in mice with experimental autoimmune encephalomyelitis (EAE), and Pietilä et al. observed ultrastructural changes in the meninges after experimental traumatic brain injury (TBI). Thus, upregulated inflammation in the CNS reported with various conditions including meningitis, Alzheimer’s disease and stroke, TBI, as well as potential crosstalk from systematic conditions, may affect the meningeal layer structure, integrity and functionality.

    Both manuscripts provide important contributions to our understanding of the transcriptomic, anatomical and functional composition of the meninges. Future studies should more thoroughly investigate the leptomeningeal barrier properties to physiological molecules such as cytokines, interactions with immune cells of the skull, meninges, and CNS. Understanding these functions may also aid in developing new methods of drug delivery into the CNS. Future research may further provide new insights into the debate around the meningeal layers and the role of Prox1-positive cells within the meninges.

    View all comments by Berislav Zlokovic

  5. These new works further elucidate the molecular anatomy of mouse meninges, focusing on dorsal brain regions. The thorough analysis by Pietilä et al. combines histologic and transcriptomic analyses and elucidates intercellular junctional protein expression among the various meningeal cell types and layers, documenting the multilaminar arachnoid barrier makeup and structure. Mapunda and Pareja et al. present additional in vivo data that depicts in detail the meningeal layers in situ. The resolution of the live 2p images approximates ex vivo data. The VE-cadherin marker in these studies is important, as it will allow for better correlation of future in vivo and ex vivo evidence which has been a limitation of prior studies.

    The current studies also highlight the importance of traditional and advanced histology techniques and careful interpretation for elucidation of meningeal structure and function. Undervaluation of histology data and interpretation is a mistake as histological and physiological evidence must be correlated to definitively and comprehensively conclude routes of intracranial fluid movement in vivo. 

    I must admit that I agree with these authors as well as with Dr. Julie Siegenthaler and others who have alluded earlier to the fact that data reported by Møllgård et al. failed to clearly depict the E-cadherin-positive arachnoid barrier cell layer. Thus, the interpretation of a double SAS compartment is highly problematic and unproven in the study by Møllgård and colleagues. Upon my interpretation​, I cannot exclude that the "outer subarachnoid space” ​reported in Figure 2A of that study ​is actually artifactual subdural space infiltration by experimentally administered tracer. This may not necessarily indicate that some amount of fluid does not naturally transit across the subdural space, but I do not feel the Møllgård​ et al. study has definitively proven that, either, as the ​technique was nonphysiological and molecular anatomy data ​in that study was lacking.

    View all comments by Rupal Mehta

  6. These studies lay a groundwork for analysis of leptomeningeal structure and function as well as alterations in aging and disease. As these works focus primarily on dorsal cerebral cortical brain regions of mice, additional investigations are needed to further map the meninges across various brain regions and species. It will be very interesting to see what is learned next using these novel approaches in mice and to understand how new knowledge of rodent leptomeninges will inform on neurofluid physiology and mechanisms of disease in humans.

    View all comments by Rashi Mehta

  7. Mapunda et al. used two-photon imaging with VE-cadherin-GFP knock-in mice and discovered that VE-cadherin, previously considered a marker for vascular endothelial cells, is actually expressed in leptomeninges as well and borders the subarachnoid space filled with CSF. Furthermore, they demonstrated that Prox1-positive cells are indeed part of the VE-cadherin-positive arachnoid mater.

    The paper by Pietila et al., through single-cell RNA sequencing analysis, examined the transcriptome of meningeal fibroblasts and revealed six distinct groups of brain and leptomeningeal fibroblast transcriptomes referred to as BFB1-6, with Prox1 aligning with BFB3, representing the inner arachnoid. Both papers confirm the alignment between Prox1-positive meninges and arachnoid markers.

    These new two papers have shed light on the molecular anatomies of leptomeninges, which were previously poorly characterized. It is now expected that further investigations into the functional roles of leptomeninges under physiological and pathological conditions will become feasible.

    View all comments by Takeshi Iwatsubo

  8. A team of human anatomists, clinicians, and basic scientists have come together to systematically examine the controversy of intermediate leptomeningeal layer or SLYM in the brain. A preprint of the document is at this open-access link: https://osf.io/5mhtu/?view_only=e5a2bf6004dd498db87587ec8c32df9e.

    View all comments by Ashutosh Kumar

Make a Comment

To make a comment you must login or register.

This paper appears in the following:

News

  1. Not So Fast—The Brain Has Three Meningeal Membranes After All