Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Jun;10(2):e99.
doi: 10.1002/ctm2.99. Epub 2020 Jun 21.

Evidence for endothelial-to-mesenchymal transition in human brain arteriovenous malformations

Affiliations
Free PMC article

Evidence for endothelial-to-mesenchymal transition in human brain arteriovenous malformations

Lorelei D Shoemaker et al. Clin Transl Med. 2020 Jun.
Free PMC article

Abstract

Background: Brain arteriovenous malformations (AVMs) are rare, potentially devastating cerebrovascular lesions that can occur in both children and adults. AVMs are largely sporadic and the basic disease biology remains unclear, limiting advances in both detection and treatment. This study aimed to investigate human brain AVMs for endothelial-to-mesenchymal transition (EndMT), a process recently implicated in cerebral cavernous malformations (CCMs).

Methods: We used 29 paraffin-embedded and 13 fresh/frozen human brain AVM samples to profile expression of panels of EndMT-associated proteins and RNAs. CCMs, a cerebrovascular disease also characterized by abnormal vasculature, were used as a primary comparison, given that EndMT specifically contributes to CCM disease biology. AVM-derived cell lines were isolated from three fresh, surgical AVM samples and characterized by protein expression.

Results: We observed high collagen deposition, high PAI-1 expression, and expression of EndMT-associated transcription factors such as KLF4, SNAI1, and SNAI2 and mesenchymal-associated markers such as VIM, ACTA2, and S100A4. SMAD-dependent TGF-β signaling was not strongly activated in AVMs and this pathway may be only partially involved in mediating EndMT. Using serum-free culture conditions, we isolated myofibroblast-like cell populations from AVMs that expressed a unique range of proteins associated with mature cell types and with EndMT. Conditioned medium from these cells led to increased proliferation of HUVECs and SMCs.

Conclusions: Collectively, our results suggest a role for EndMT in AVM disease. This may lead to new avenues for disease models to further our understanding of disease mechanisms, and to the development of improved diagnostics and therapeutics.

Keywords: arteriovenous malformation (AVM); cerebrovascular disease; endothelial-to-mesenchymal transition (EndMT); myofibroblast.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

FIGURE 1
FIGURE 1
EndMT‐associated transcription factors are expressed in human brain AVMs. A, Vertical scatter plots of KLF4, SNAI1, SNAI2, TWIST1, TWIST2, and ZEB1 expression in normal brain (NB), AVM‐1‐5, and CCM‐1‐4 by qRT‐PCR. Each point represents a unique sample and each value is represented as the mean ± SD (n = 3). The horizontal line indicates the median. B, Representative images of SNAI1/2 expression by IHC in NB, AVMs, and CCMs. Tissue within dashed boxes is shown at higher magnification to the right. Arrows indicate regions of PECAM‐/SNAI1/2‐positive cells, whereas arrowheads indicate SNAI1/2‐positive cells present in the perivascular tissue. Scale bar for a‐d, g, h, k, and l, 100 μM; scale bar for e, f, i, j, m, and n, 50 μM
FIGURE 2
FIGURE 2
Expression of EndMT markers in human brain AVMs. A, Vertical scatter plots of expression of ACTA2 and VIM by qRT‐PCR in AVM‐1‐5 and CCM1‐4, compared to normal brain (NB). Each point represents a unique sample, with the sample value represented as the mean ± SD (n = 3). The horizontal line indicates the median. B, S100A4 is highly expressed in AVMs in PECAM‐positive ECs (arrows) as well as in the perivascular tissue (arrow heads) (d, e, g, and h). CCM tissue expressed S100A4, primarily in the PECAM‐positive ECs (arrows), although some S100A4 was observed proximal to the vessel lumen (arrowhead) (j and k). Both AVMs and CCMs demonstrated collagen deposition (blue) throughout the lesions (f, i, and l), including in regions of S100A4‐positive cells. NB expressed low levels of S100A4 and collagen (a‐c). Scale bar for a‐c, 100 μM; scale bar for d‐l, 50 μM
FIGURE 3
FIGURE 3
AVMs expressed PAI‐1, a protein transcriptionally regulated by TGF‐β. PAI‐1 was present in both PECAM‐positive ECs within the AVMs (arrows) and in PECAM‐negative regions (arrowheads) surrounding the vessels (c‐j). PAI‐1 was expressed throughout the CCM tissue (k‐n). Normal brain (NB) expressed low levels of PAI‐1 (a‐b). Regions in dotted boxes are shown at higher magnification to the right. Scale bar for a‐d, g, h, k, and l, 100 μM; scale bar for e, f, i, j, m, and n, 50 μM
FIGURE 4
FIGURE 4
SMAD4 expression in AVMs. Normal brain (NB) and AVM1‐5 samples expressed similarly lower levels of SMAD4 mRNA by qRT‐PCR, as compared to CCM‐1‐2 samples, as shown in the vertical scatter plot (A). Each point represents a unique sample, with the sample value represented as the mean ± SD (n = 3). The line indicates the median. B, SMAD4 protein expression by IHC was weak in the AVM samples and limited to discrete regions of the tissue, whereas CCMs expressed robust SMAD4 protein. NB showed low expression throughout the sample. Region 1 and Region 2 indicate two areas of the tissue that is representative of the expression pattern for that sample. Tissue within dashed boxes is shown at higher magnification to the right. Scale bar for a‐e and k‐o, 100 μM; scale bar for f‐j and p‐t, 50 μM
FIGURE 5
FIGURE 5
Protein profiling of AVM cell lines by immunocytochemistry. Expression of (A) mature cell‐type proteins, αSMA and PECAM, and (B) EndMT‐associated proteins SNAI1/2, S100A4, and PAI‐1, in AVM cell lines, aSMCs and HUVECs. Hoechst was used to label nuclei. Scale bars, 100 μm
FIGURE 6
FIGURE 6
Serum‐free NBA AVM‐CM supported proliferation of HUVECs and SMCs, and induced phenotypic changes in HUVECs. HUVECs (A) and SMCs (B) were grown in EGM‐2 and SmGM, (respectively), serum‐free NBA, and serum‐free NBA AVM‐CM for 24 h. Random fields were scored for total number of cells (Hoechst 33342‐positive) and of proliferating cells with EdU incorporation. Representative bright field images of HUVECs grown in EGM‐2 (C), unconditioned, serum‐free NBA (D), and serum free NBA AVM‐CM (E). Error bars: mean ± SD; T‐test was performed to determine significance: * P ≤ .05; ** P ≤ .001; scale bar, 200 μm

Similar articles

Cited by

References

    1. Laakso A, Hernesniemi J. Arteriovenous malformations: epidemiology and clinical presentation. Neurosurg Clin N Am. 2012;23(1):1‐6. - PubMed
    1. Meyer‐Heim AD, Boltshauser E. Spontaneous intracranial haemorrhage in children: aetiology, presentation and outcome. Brain Dev. 2003;25(6):416‐421. - PubMed
    1. Shoemaker LD, Fuentes LF, Santiago SM, et al. Human brain arteriovenous malformations express lymphatic‐associated genes. Ann Clinical Transl Neur. 2014;1:(12):982‐995. - PMC - PubMed
    1. Weinsheimer S, Bendjilali N, Nelson J, et al. Genome‐wide association study of sporadic brain arteriovenous malformations. J Neurol Neurosurg Psychiatry. 2016;87(9):916‐923. - PMC - PubMed
    1. Nikolaev SI, Vetiska S, Bonilla X, et al. Somatic activating KRAS mutations in arteriovenous malformations of the brain. N Engl J Med. 2018;378(3):250‐261. - PubMed