Endothelial cells occupy one of the most mechanically demanding positions in the body. They form the interface between circulating blood and the vessel wall, enduring shear stress and cyclic strain from flow while anchoring themselves to the underlying extracellular matrix and to their neighbors. Detecting and responding to these forces is essential for vascular development and for maintaining vessels in a healthy, quiescent state, and when the responses go awry the result is abnormal remodeling, leaky barriers, and diseases such as atherosclerosis and aortic aneurysm. The nucleus itself is now recognized as a force sensor: it often bulges into the vessel lumen, where blood flow drags across it, and it receives mechanical input from the matrix through cytoskeletal linkages. Yet how mechanical information is shuttled between the nucleus and distant cellular structures has remained poorly understood.

The key structure in this study is the LINC complex, the Linker of Nucleoskeleton and Cytoskeleton, which physically bridges the nuclear interior to the cytoskeleton. SUN proteins anchored in the inner nuclear membrane bind nesprin proteins that span the outer membrane and connect to microtubules, actin, and intermediate filaments. Most cells express two SUN proteins, SUN1 and SUN2, which form trimeric complexes with nesprins. The team had previously shown that SUN1 stabilizes endothelial cell-cell junctions via microtubule regulation, so they turned to SUN2 expecting a complementary role. What they found instead was a protein with a distinct and, in some respects, opposite job: SUN2 governs how endothelial cells interact with the matrix beneath them, far from where the protein resides.

Using siRNA to deplete SUN2 from human umbilical vein endothelial cells, the researchers observed that the cells became abnormally elongated even without flow, and hyper-elongated further when exposed to homeostatic laminar shear stress of 15 dyn/cm2 for 72 hours. Under flow, SUN2-depleted cells also failed to polarize properly relative to the flow direction, as judged by the axis between nucleus and Golgi. More strikingly, laminar flow normally drives endothelial cells into a non-proliferative quiescent state, but SUN2-depleted cells kept incorporating the DNA label EdU, indicating continued S-phase entry. The flow-induced transcription factors KLF2 and KLF4, which mediate many protective responses to laminar shear, were significantly reduced at the mRNA level, and KLF4 protein was also diminished. SUN2, in other words, is required for cells to hear what flow is telling them.

To confirm the findings in living animals, the team generated an endothelial-specific, tamoxifen-inducible Sun2 knockout mouse, deleting the gene in pups during the first three postnatal days and analyzing the aorta and retinal vasculature at postnatal day 7. The aortic endothelium of knockout mice showed the same hyper-elongation seen in culture, along with irregular, wavy cell-cell junctions with reduced linearity. Nuclear KLF4 staining was reduced, mirroring the in vitro flow-response defect. The retinal vascular network showed reduced radial expansion, and in zebrafish, depletion of sun2 by morpholino impaired growth of intersegmental vessels and their connection to the dorsal longitudinal anastomotic vessel, with similar trends in a sun2 mutant line. Together these results establish that SUN2 regulates angiogenic expansion and flow responses in vivo, not merely in a dish.

Perhaps the most disease-relevant discovery came from the transcriptome. Bulk RNA sequencing of SUN2-depleted cells under flow identified 116 differentially expressed genes, the majority downregulated, with gene ontology enrichment in cell-substrate adhesion and extracellular matrix organization. Among the affected genes were FBN1, encoding fibrillin-1, along with FBN2, COL3A1, FBLN2, FBLN5, LTBP1, and the integrins ITGA9 and ITGB3. Fibrillin-1 protein itself was reduced. These are not random targets: FBN1 mutations cause Marfan syndrome, and COL3A1 mutations underlie vascular Ehlers-Danlos syndrome, both connective tissue disorders characterized by aneurysm and vessel rupture. Consistent with the gene expression changes, the aortas of Sun2 knockout mice displayed significantly more breaks in the internal elastic lamina, a phenotype that closely resembles that of mice carrying a Marfan-associated Fbn1 mutation.

The functional consequences of this altered matrix were tested directly with elegant ECM swap experiments. Control and SUN2-depleted cells were seeded onto decellularized matrices derived from either cell type. SUN2-depleted cells placed on control matrix largely recovered their normal shape, while even control cells seeded on matrix from SUN2-depleted cells became somewhat hyper-elongated. The conclusion is that the matrix produced by SUN2-deficient endothelial cells is itself abnormal, and that this defective matrix cooperates with cell-intrinsic changes to drive the observed phenotypes. Matrix metalloproteinase genes were not globally mis-regulated, but MMP9 RNA was elevated, and fibronectin protein levels were reduced, pointing to a shift in the balance of matrix deposition and degradation.

Barrier function suffered as well. Electrical impedance measurements across endothelial monolayers dropped when SUN2 was depleted, a biotin-accessibility assay revealed increased exposure of the underlying matrix, and microfabricated three-dimensional vessels seeded with SUN2-depleted cells leaked fluorescent 70-kDa dextran from their lumens. Notably, two broad-spectrum matrix metalloproteinase inhibitors, Marimastat and Ilomastat, rescued the barrier defects in both two-dimensional and three-dimensional models, implicating excessive matrix degradation, likely through enhanced delivery of MMPs by over-stabilized microtubules, as a proximal cause of the leakiness.

The mechanistic thread tying these phenotypes together turned out to be the microtubule cytoskeleton. SUN2-depleted cells had increased microtubule density, elevated acetylated tubulin, a marker of microtubule stabilization, and live tip-tracking of EB3-labeled growing ends showed that microtubules spent more time growing and less time shrinking. Focal adhesions, the cell-matrix contacts on the cell’s basal surface, became abnormally long and resistant to turnover, and traction force microscopy showed that force-generating contacts were concentrated at two poles rather than distributed evenly, a configuration associated with non-migrating cells. Cell migration was correspondingly impaired, with SUN2-depleted cells extending from both ends simultaneously before abruptly retracting. When the researchers partially depolymerized microtubules with low-dose nocodazole, the extended focal adhesions, the leaky barrier, the hyper-elongation, and the loss of flow-mediated quiescence were all rescued to control levels, placing microtubule dynamics downstream of SUN2 and upstream of nearly everything else.

The final piece of the puzzle was nesprin-1, the LINC complex partner that links SUN proteins to the cytoskeleton. Co-depleting nesprin-1 rescued the hyper-elongation, the increased microtubule density, the barrier defect, the elevated EdU incorporation under flow, and the reduced KLF4 expression caused by SUN2 loss, even though nesprin-1 depletion alone had no effect. This places the SUN2/nesprin-1 LINC complex at the top of the regulatory hierarchy. The authors propose a balanced model in which SUN1/nesprin-1 and SUN2/nesprin-1 complexes normally counterweight each other at the nuclear envelope: SUN1 stabilizes microtubules to protect cell-cell junctions, while SUN2 destabilizes them to permit proper focal adhesion turnover, matrix maintenance, and flow sensing. When that balance tips, as it appears to in cardiovascular aging and in diseases such as Marfan syndrome and Hutchinson-Gilford progeria, vessel wall integrity suffers. The work suggests that the nuclear envelope, long viewed as a passive container for the genome, is in fact a control hub whose molecular composition sets the mechanical tone of the entire vessel, and it points to LINC complex components and microtubule dynamics as potential therapeutic targets for diseases of the vessel wall.

Subject of Research: Role of the nuclear LINC complex protein SUN2 in endothelial cell-matrix interactions, microtubule dynamics, and vascular barrier function

Article Title: Nuclear SUN2 coordinates endothelial cell-matrix interactions to regulate blood vessel homeostasis and barrier function

Article References: Nuclear SUN2 coordinates endothelial cell-matrix interactions to regulate blood vessel homeostasis and barrier function. (n.d.). https://doi.org/10.1007/s10456-026-10098-w

Image Credits: AI Generated

DOI: 10.1007/s10456-026-10098-w

Keywords: SUN2, LINC complex, endothelial cells, mechanotransduction, microtubules, focal adhesions, extracellular matrix, barrier function, angiogenesis, Marfan syndrome, vascular homeostasis, nesprin-1