DOT1-like histone lysine methyltransferase is critical for adult vessel maintenance and functions

Objective Disruptor of telomeric silencing 1-like (DOT1L) is the only known histone H3K79 methyltransferase essential for the development of the embryonic cardiovascular system, including the heart, blood vessels, and lymphatic vessels, through transcriptional regulation. Our previous study demonstrated that Dot1l deletion results in aberrant lymphatic development and function. However, its precise function in the postnatal cardiovascular system remains unknown. Methods Using conditional and inducible Dot1l knockout (KO) mice, along with a reporter strain carrying the Geo gene at the Dot1l locus, DOT1L expression and its function in the vascular system during postnatal life were investigated. To assess vessel morphology and vascular permeability, we administered Latex or Evans blue dye to KO mice. In addition, in vitro tube formation and cell migration assays were performed using DOT1L-depleted human umbilical vein endothelial cells (HUVECs). Changes in the expression of vascular genes in HUVECs were measured by quantitative polymerase chain reaction. Results Our findings demonstrate that conditional Dot1l knockout in the Tg (Tie2-cre) strain results in abnormal blood vessel formation and lymphatic anomalies in the intestine. In a mouse model of Rosa26-creER-mediated inducible Dot1l knockout, we observed vascular phenotypes, including increased vascular permeability and brain hemorrhage, when DOT1L was deleted in adulthood. Additionally, DOT1L depletion in cultured HUVECs led to impaired cell migration and tube formation, likely due to altered gene transcription. These findings highlight the essential role of DOT1L in maintaining vascular integrity and function during embryonic development and postnatal life. Conclusion Our study revealed that DOT1L is required for the maintenance of adult vascular function through the regulation of gene expression.


INTRODUCTION
The cardiovascular system is a complex network comprised of the heart, blood vessels, and lymphatic vessels.These components work together as the driving force behind blood circulation, ensuring the supply of nutrients and oxygen throughout the body while maintaining the water balance [1][2][3].Recent evidence indicates that blood and lymphatic vessels are regulated [4][5][6].Notably, lymphangiogenesis studies have revealed that lymphatic endothelial cells (LECs) are not solely derived from a subset of blood endothelial cells (BECs) in the cardinal vein (CV) as previously thought [7].Instead, LECs originate from diverse types of (B)ECs, indicating a more complex origin model [8][9][10][11][12][13].Recent studies have highlighted that LECs in the CV are embryologically derived from PAX3(+) cells originating from the lateral plate mesoderm (LPM) [14,15].Migration of PAX3(+) cells to the CV is regulated by a centrifugal mechanism.However, specific molecular mechanisms underlying this regulation remain unclear.Epigenetic mechanisms are crucial for ensuring proper functioning and maturation of endothelial cells during development and for maintaining the integrity of the cardiovascular system [16,17].
Disruptor of telomeric silencing 1-like (DOT1L) was shown in previous studies performs critical functions during cardiovascular development.DOT1L plays vital roles in cardiac development and function [18][19][20][21].Additionally, DOT1L is particularly important for the development and function of lymphatic vessels (LECs) [18,20,21], and DOT1L was identified as a key protein regulating LECs migration over blood endothelial cell (BECs) migration in a genome-wide gene knockdown study [22].Our previous study demonstrated that DOT1L plays an important role in the development and function of lymphatics in developing mouse embryos by modulating histone H3K79 methylation and transcription [23].Furthermore, DOT1L promoted BEC migration, tube formation, and sprout formation in vitro and in vivo, primarily by regulating VEGFR2 expression [20].Remarkably, recent research has demonstrated that DOT1L has a different transcriptional regulatory network depending on the type of endothelial cell, whether LEC or BEC [24].This indicates the complexity of DOT1L function and highlights its significance in cardiovascular development and lymphangiogenesis.
However, whether DOT1L is essential for the function (s) of BECs and/or LECs during adult life remains unclear.In the present study, we generated DOT1L knockout and reporter alleles and analyzed their expression patterns and vascular functions.Our analysis revealed that DOT1L is broadly expressed in various organs of adult mice, and its loss results in brain hemorrhage and increased vascular permeability.Furthermore, DOT1L silencing in vitro led to impaired tube formation and migration, likely owing to aberrant transcriptional regulation.
4-Hydroxytamoxifen (HTM H6278, Sigma) was dissolved in DMSO (2 mg/25 g) and intraperitoneally injected into 7 to 8 weeks-old Dot1l 2f/2f ; ROSA26-creER and control (Dot1l 2f/2f ) mice every other day for 2 weeks.One month after the last HTM injection, Evans Blue (EB) dye (1 mL of 3% [in 0.9% saline]/kg of mouse) was injected into the tail vein of Dot1l 2f/2f ; ROSA26-creER and control mice.Thirty minutes after EB injection, mice were anesthetized with isoflurane and perfused with 1% paraformaldehyde in 0.05M citrate buffer through the left ventricle.Approximately 4 cm of the small intestine was collected and placed between two pieces of Whatman filter paper for 10 seconds.and weighed.The tissues were then placed in 1 mL of formamide overnight at 56°C to extract EB.The following day, after removing the tissues, the amount of EB extracted was measured using a spectrophotometer (A 620 ).Values were expressed as ng EB/mg of tissue.

Analysis of cell migration and tube formation
For the tube formation assay, the HUVECs were counted and seeded into Matrigel-coated 48-well plates.Approximately 5×10 3 cells were plated in each well.The plates were imaged 12 h after seeding to assess tube formation.The number of tube branches originating from the center of each well was counted as a measure of angiogenic potential.
For scratch wound closure analysis, approximately 2×10 5 HUVECs were seeded in a 6-well plate.Once the cells reached approximately 95% confluence, scratches or wounds were created across the cell monolayer using a p200 pipette tip.After 17 h, the wound closure was imaged using a phasecontrast microscope.The distance of wound closure was measured using the ImageJ software, which allowed quantification of the extent of cell migration and wound healing.
For hematoxylin and eosin (H&E) staining, tissues were dehydrated with differential concentration of ethanol series and xylene, embedded into paraffin, and cut with 6 μmthickness.Tissue sections were rehydrated with different concentrations of ethanol and H 2 O.They were then stained with eosin for the cytoplasm and hematoxylin for the nuclei, and mounted.Images were captured using an Olympus stereo microscope.

Reverse transcription quantitative polymerase chain reaction
Total RNAs was extracted using a RNeasy Plus Mini Kit (74104; Qiagen, USA), and cDNA was synthesized using Maxime RT PreMix cDNA synthesis kit (iNtRON; Seongnam, Korea).Quantitative polymerase chain reaction (qPCR) was performed on a StepOnePlus System (Applied Biosystems, Foster City, CA USA) using the Fast SYBR Green Master Mix (4385616; Applied Biosystems, USA).Primer sequences used in the qPCR analysis are presented (Table 1).

Statistical analysis
All statistical analyses were performed using Prism 8.0.2 (GraphPad Software, San Diego, CA, USA).A t-test was used to determine the statistical significance in tube formation, cell migration, and qPCR assays.Mean values with the standard error of the mean (SEM) are presented as graph errors.Statistical significance was determined at a p-value lower than 0.05 (*), 0.01 (**), 0.001 (***) or NS (not significant).

Dot1l expression during post-natal life
In this study, we first assessed Dot1l expression in adult organs using Dot1l Geo/+ mice, as previously described in another study [28].Similar to the embryonic expression pattern, ubiquitous Dot1l expression in adult organs, including the heart, lungs, liver, sternum, and thymus, was detected using X-Gal staining (Figure 1).These results demonstrated that Dot1l is widely expressed in various organs during adulthood, mirroring its embryonic expression pattern.

Analysis of postnatal vessel phenotypes in Dot1l knockout mice
In our previous study [23], we demonstrated that most conditional Dot1l knockout (cKO) mice using the Tg (Tie2-cre) strain exhibited late gestational embryonic lethality due to lymphatic anomalies such as blood-mixing, lymphatic hypoplasia, and chylous ascites.However, few cKO mice are viable after birth and display lymphatic phenotypes.Viable cKO mice were used to assay vessel phenotypes after birth.On approximately PD5, these mice were injected with latex dye into the left ventricle, which revealed a focal diffuse and nodular pattern of dye distribution around the hemorrhagelike spots of the intestinal tubes.To further examine DOT1L function during adulthood, we employed inducible knockout (iKO) mice using the ROSA26-creER strain.As depicted in (Figure 2E, F) hemorrhagic spots were found in iKO brains and increased vascular permeability was observed in these mice.

Dot1l depletion impairs endothelial cells migration, tube formation, and expression of vascular genes
To investigate Dot1l function in the regulation of vessel formation, in vitro tube formation and cell migration assays were performed.The results shown in (Figure 3A, C) indicate a significant reduction in the capacity for both tube formation and cell migration in DOT1L-depleted HUVECs.Next, we examined the gene expression patterns in DOT1L-depleted HUVECs.As shown in (Figure 3E), a subset of the genes examined, including PROX1, SOX18, NR2F2, VEGFR3, VEGFC, NRP2, TBX1, CALCRL, SYK, PDPN, RAC1, PPP1R13B, PLCG2, C1GALT1, DLL4, EFNB2 and DOT1L were repressed in KD HUVECs.

DISCUSSION
During embryonic development, precise control of gene expression is essential for the formation of various tissues and organs [29,30].Various signaling pathways and transcription factors regulate the expression of specific genes at different stages of development [31,32].Epigenetic modifications, such as histone methylation, play a key role in this process by regulating chromatin structure and accessibility, thereby influencing gene expression during cardiovascular development and function [16,33,34].The present study examined DOT1L function in adult mice and found that DOT1L loss induced cerebrovascular hemorrhage and increased vascular permeability.In addition, in vitro experiments showed that deletion of DOT1L impaired tube formation and migration in HUVECs.Furthermore, they found that this was due to the transcriptional regulation by DOT1L.As expected, Dot1l was widely expressed in the cells of various organs, and this expression pattern was likely associated with essential DOT1L functions.
Our analysis revealed that DOT1L loss in adults leads to increased vascular permeability and eventual cerebrovascular bleeding.Proper regulation of vascular permeability is crucial for tissue homeostasis and immune responses, and dysregulation can lead to pathological conditions such as edema and inflammation [38][39][40].Similarly, histone methylation plays a critical role in regulating vascular integrity, which is essential for maintaining the barrier function of blood vessels and preventing vascular leakage and inflammation.Epigenetic modification of histones, particularly methylation, can affect the expression of genes involved in endothelial cell adhesion, junctional complexes, and cytoskeletal organization, thereby influencing the overall vascular integrity [41][42][43].For example, histone methyltransferase Suv39h1 repress eNOS transcription via binding to its proximal promoter, in which reduced H3K9me3 level induces IFN-γ-induced eNOS repression [44].In general, eNOS is critical for the regulation of endothelial function and vascular tone [45,46].Jmjd3, a histone H3K27me3 demethylase, expression is upregulated upon oxygen-glucose deprivation/reperfusion injury in endothelial cells, in which the elevation of Jmjd3 leads to increase its interactions with Nf-κb (p65/p50) and CCAAT-enhancerbinding protein β in interleukin 6 (Il-6) gene promoter, and then decrease H3K27me3 levels to promote Il-6 expression [47].In addition, Jmjd3 expression is enhanced in endothelial cells after lipopolysaccharide treatment to activate target gene expression via demethylation of H3K27me3 [48].Overexpression of vascular endothelial cadherins increases claudin-5 expression by preventing the binding of PRC2 to the CLDN5 gene [49].Recently, it was shown that common and/or diverse gene signatures and epigenetic landscapes exist across organotypic BECs in mice [34].
Consistent with our analysis, silencing DOT1L in HUVECs leads to cell death and reduced capillary sprout formation, which are associated with altered H3K79 methylation and transcription factor ETS-1 binding, resulting in the regulation of VEGFR2 expression [20].Our previous study revealed that DOT1L regulates distinct sets of transcripts in different types of endothelial cells, likely through H3K79 methylation [24].One of these gene sets included cell-cell adhesion molecules (CLDN1, CLDN11, and PCDHs), which were found to be differentially expressed by DOT1L depletion or overexpression.Consequently, it is plausible that the cerebrovascular hemorrhage observed in the induced DOT1L KO mice was due to the repressed expression of these cell-cell adhesion molecules.
Our study, along with others, showed that DOT1L plays a crucial role in maintaining vessel integrity in adults through transcriptional regulation.Cerebrovascular hemorrhage observed in DOT1L KO mouse could potentially open new avenues for utilizing DOT1L in the treatment of cerebral hemorrhage.However, further research is required to explore this issue in more detail and gain a deeper understanding of the therapeutic potential of DOT1L in the treatment of cerebral hemorrhage.

Figure 2 .
Figure 2. Phenotypic analysis of Dot1l knockout (KO) strains.(A) Visualization of latex dye-injected Dot1l cKO intestine.Note that a diffused vessel structure was detected in the Dot1l cKO intestine.(B) Representative images of Evans blue dye leakages in face, dorsal skin and intestine.(C) Representative images of Evans blue dye-injected intestines.(D) Quantification of leaked Evans blue dye extracted from intestine.Graphs represent the mean and standard error mean.***, p<0.001.(E) Representative photographs of hemorrhage-like spots in the inducible Dot1l KO brain.Arrow head: hemorrhagic spot(s).(F) Hematoxylin and eosin (H&E) staining in the inducible Dot1l KO brain.Arrow head: hemorrhage.Scale bar = 200 μm.

Figure 3 .Figure 3 .
Figure 3. Analysis of cellular effects of DOT1L silencing in HUVECs.(A) Assay of in vitro tube formation in DOT1L-knock down HUVECs, and quantification (B).(C) Assay of cell migration in the DOT1L-knock down HUVECs, and quantification (D).Graphs in (B) and (D) represent the mean and standard error mean.* p<0.05.*** p<0.001.(E) qPCR analysis of vascular gene expression in the DOT1L-knock down HUVECs.Scale bar = 200 μm.HUVECs, human umbilical vein endothelial cells; RT-qPCR, reverse transcription quantitative polymerase chain reaction.

Table 1 .
Primer and knockdown (KD) sequences used in this study