Primary C57bl/6 mouse cerebral endothelial cells (mBMEC) were purchased from Cell Biologics Inc. (C57-6023; Chicago, IL). mBMEC were cultured on 0.1% gelatin-coated flasks using endothelial cell medium (M1168, Cell Biologics, Inc.) with growth factor supplement additives (Cell Biologics, Inc.) and 5% (v/v) fetal bovine serum (Cell Biologics, Inc.). mBMEC were used at passages 2–5, and at least two different lots were used for each experiment. Mouse cerebellar astrocytes were purchased from ATCC (C8-D1A; Manassas, VA) and cultured in DMEM with 10% fetal bovine serum for trans-endothelial electrical resistance (TEER) experiments. Lasmiditan was purchased from Tocris (Ellisville, MO).

DNA was isolated from mBMEC, wild-type C57bl/6NJ spinal cord tissue, and 5-HT_1F receptor knockout mouse (B6N(Cg)-Htr1f^{tm1.1(KOMP)Vlcg}/J; Jackson Laboratories) spinal cord tissue using the Qiagen DNeasy Blood and Tissue Kit (Valencia, CA). To determine the presence of the 5-HT_1FR gene, DNA was amplified using Promega 2x PCR Master Mix (Promega, Madison, WI) in accordance with the manufacturer’s protocols. Amplified DNA was separated on a 1.5% agarose gel and visualized by ethidium bromide fluorescence. Primers used were sense: 5′-GCCGTGATGATGAGTGTGTC-3′ and antisense: 5′-ACTATCCGACTCGCTTGTCT-′3.

The oxygen consumption rate (OCR) of mBMECs was measured using the Seahorse Bioscience XF-96 Extracellular Flux Analyzer as previously described (Agilent, Santa Clara, CA) (Beeson et al., 2010). Cells were plated at a density of 1×10⁴ cells/well in 96-well plates. Each plate was treated with either vehicle (DMSO; <0.5%) or lasmiditan (0-100 nM) for 48 h in fetal bovine serum-free media. Basal OCR was measured before injection of carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP; 2 μM; Sigma-Aldrich, St. Louis, MO) to measure uncoupled OCR, a marker of functional MB (Beeson et al., 2010).

mBMECs were plated at a density of 5×10⁴ on D35 cell culture dishes and treated with vehicle or 3 nM lasmiditan for 24 h. mBMECs were then fixed and sectioned for transmission electron microscopy (TEM) as previously described (Dupre et al., 2019; Scholpa et al., 2019). Images were viewed using the FEI Tecnai Spirit microscope (FEI, Hillsboro, OR) operated at 100 kV and captured using an AMT 4 Mpixel camera (Advanced Microscopy Techniques, Woburn, MA) at a magnification of 16,500X. Mitochondrial count and morphology were analyzed using the analyze particles plug-in in ImageJ FIJI. In all cases, four to five images were analyzed blindly per sample.

Protein was extracted using RIPA buffer with protease inhibitor cocktail (1:100), 1 mM sodium fluoride, and 1 mM sodium orthovanadate (Sigma-Aldrich, St. Louis, MO) as described previously (Dupre et al., 2019; Scholpa et al., 2019). Protein was quantified using a bicinchoninic acid assay, and up to 10 μg of protein was separated via electrophoresis using 4–15% SDS-PAGE gels, then transferred to nitrocellulose membranes (Bio-Rad, Hercules, CA). Membranes were blocked in 5% milk or BSA in TBST and incubated overnight with primary antibodies with constant agitation at 4°C. Membranes were incubated with the appropriate horseradish peroxidase-conjugated secondary antibody and visualized using chemiluminescence (Thermo Scientific, Waltham, MA) on a GE ImageQuant LAS-4000 (GE Life Sciences, Pittsburg, PA). Optical density was determined using Image Studio Lite software. Primary antibodies used were PGC-1α (1 μg/mL; Abcam ab191838, Cambridge, MA), ATPSB (1 μg/mL; Abcam ab14730), CD31 (0.523 μg/mL; Abcam ab222783), claudin-5 (0.5 μg/mL; Invitrogen 35–2,500, Carlsbad, CA), ZO-1 (0.22 μg/mL; Abcam ab96587), occludin (0.71 μg/mL; Abcam ab216327), VE-cadherin (0.238 μg/mL; Abcam ab205336), p-VE-cadherin (Tyr658; 0.2 μg/mL; Invitrogen 44-1144G), p-Akt (Ser473; 0.091 μg/mL; Cell Signaling 9,271, Danvers, MA), Akt (0.031 μg/mL; Cell Signaling 9,272), p-eNOS (Ser1177; 0.485 μg/mL; Abcam ab215717), eNOS (0.25 μg/mL; Abcam ab76198), Nop-FoxO1 (Ser256; 0.107 μg/mL; Cell Signaling 9,461 s), FoxO1 (0.088 μg/mL, Cell Signaling 2,880 s), and tubulin (0.065 μg/mL, Abcam ab52866).

Migration was assessed and analyzed as previously described (Liang et al., 2007; Dupre et al., 2019) via MRI wound healing plug-in in ImageJ FIJI. Briefly, mBMEC were plated at a density of 5×10⁴ on D35 cell culture dishes and grown to confluence. Cells were then treated with either vehicle or 3 nM lasmiditan for 24 h in serum-free media. Scratches were applied using a 10-μL pipette tip, and media were replaced with complete media containing lasmiditan or vehicle. Images of five randomly selected areas of the scratch were taken for each treatment at each time point. The percent migration was determined by the following formula: % area of migration 6 h post-scratch = (scratch area 0 h post-scratch − scratch area 6 h post-scratch) / (scratch area 0 h post-scratch) * 100.

mBMEC were plated at a density of 2.2×10⁶ on D100 cell culture dishes pretreated with Gibco attachment factor and grown to approximately 80% confluence in endothelial cell media. Cells were then treated with either vehicle or 3 nM lasmiditan for 24 h. After 24 h, mBMEC were collected via Accutase dissociation and centrifuged at 300 x g for 10 min at 4°C to form a pellet. Cells were resuspended in endothelial cell media containing either vehicle or lasmiditan, counted, and diluted to 1.5×10⁵ cells/mL, and 50 μL of cell suspension was added to an Ibidi (Fitchburg, WI) μ-Slide 15 Well 3D cell culture slide precoated with 10 μL of Corning reduced growth factor phenol red free Matrigel solution (Corning, NY), as per manufacturer’s recommended protocol. mBMEC were allowed to grow in Matrigel-coated wells for 24 h, and wells were imaged with a 10x LWD objective on an EVOS M5000 microscope under phase contrast with a light diffuser insert to help prevent uneven illumination.

Ten images from each lot/passage were analyzed using Ibidi FastTrack AI software (provided by MetaVi Labs).

mBMEC were plated at a density of 1.8×10⁵ cells/well on polyester-coated Transwell inserts (CLS3460; Corning) of 0.4 μm pore size in endothelial cell media and grown to confluence. For monoculture experiments, membranes were then placed in 12-well plates containing endothelial cell media and treated with 3 nM lasmiditan or vehicle. For co-culture experiments, 2×10⁵ astrocytes/well were plated and grown 48 h in 12-well plates with DMEM. Inserts containing mBMEC were then added to the astrocyte-containing wells and treated with 3 nM lasmiditan or vehicle in endothelial cell media. TEER was assessed using an EVOM² Epithelial Voltohmmeter with chopsticks electrode set (World Precision Instruments) after 24 h (day 1) and again after 48 h (day 2). Net resistance (Ω) was determined by subtracting that of media-only blank wells, and TEER was calculated based on the area of the well as Ω·cm² (Srinivasan et al., 2015).

Primary mBMEC isolated from a single passage represents an individual experiment (n = 1) and n = 5-6/group was used for each experiment. Data were normalized to respective vehicle-treated samples. Differences between two groups were analyzed using a two-tailed paired t-test, while that of three or more groups was analyzed using a one-way ANOVA followed by Tukey’s post-hoc test. In all cases, GraphPad Prism software (La Jolla, CA) was used and a p < 0.05 was considered statistically different between mean values.

Prior to experimentation, expression of the 5-HT_1FR was confirmed in mBMEC via PCR (Figure 1A). FCCP-uncoupled oxygen consumption rate (OCR) is a marker of maximal ETC activity and potential MB (Beeson et al., 2010; Wills et al., 2012; Garrett et al., 2014; Cleveland et al., 2020; Simmons et al., 2021). mBMEC were exposed to various concentrations of lasmiditan for 48 h and then assessed for FCCP-OCR. Lasmiditan had a concentration-dependent effect on FCCP-OCR, with a ~ 25% increase with a 3 nM, the lowest effective concentration (Figure 1B). Based on this, 3 nM lasmiditan was used for all experiments. Transmission electron microscopy (TEM) was used to assess mitochondrial content in mBMEC (Figure 1C). Following lasmiditan exposure, mBMEC exhibited increased mitochondrial number and area (Figure 1D) per field compared to vehicle controls, further indicating MB. Immunoblot analysis revealed increased PGC-1α, the master regulator of MB, and ATP synthase β (ATPSB), a subunit of the ETC, with lasmiditan treatment compared to vehicle (Figure 1E).

A wound healing model was used to determine whether lasmiditan stimulates migration in mBMEC (Liang et al., 2007) (Figure 2A). Cells were grown to confluence and treated with vehicle or 3 nM lasmiditan for 24 h in serum-free media, after which a scratch was applied (0 h). Lasmiditan treatment decreased the area of damage by ~60% compared to vehicle control by 6 h (Figure 2B), equating to a 2-fold increase in cell migration (Figure 2C). Importantly, no difference in wound area was observed upon injury.

A Matrigel-based tube formation assay was used to assess the angiogenic potential of lasmiditan in mBMEC (Arnaoutova and Kleinman, 2010; Almeria et al., 2019; Hunter et al., 2022) (Figure 3A). Following 24 h of treatment, mBMEC were added to Matrigel-coated wells (0 h) and allowed to grow for an additional 24 h in the continued presence of vehicle or 3 nM lasmiditan. Lasmiditan treatment increased loop count 3-fold (Figure 3B), total tube length 1.5-fold (Figure 3C), and branch count 2-fold (Figure 3D) in mBMEC. Figures 3E–G depict the loop count, total tube length, and branch count for all images collected.

Trans-endothelial electrical resistance (TEER) was used to examine the effect of lasmiditan on mBMEC barrier function with and without astrocyte co-culture. When grown in monoculture, mBMEC TEER increased nearly 2-fold when treated with lasmiditan compared to vehicle on both day 1 (43 v 24 Ω·cm²) and day 2 (58 v 31 Ω·cm²) (Figure 4A). Similar results were seen in the presence of astrocytes, where lasmiditan treatment increased mBMEC TEER from 33 to 57 Ω·cm² on day 1 and from 45 to 78 Ω·cm² on day 2 (Figure 4B). Regardless of the presence or absence of astrocytes or lasmiditan, mBMEC TEER increased from day 1 to day 2. As expected, TEER values were inherently higher with astrocyte co-culture (Abbott, 2002); however, these data suggest that astrocytes have no effect on the lasmiditan-induced increase in mBMEC barrier integrity. Protein was collected from the mBMEC that had undergone TEER assessment in the presence of astrocytes (Figure 4B), and the expression of tight junction proteins was assessed. Claudin-5 and occludin increased 2- and 1.5-fold with lasmiditan treatment, respectively, while no effect was observed with ZO-1 (Figure 4C).

To begin to discern the pathway of the effect of lasmiditan on endothelial cell function in mBMEC, cells were exposed to lasmiditan or vehicle for 48 h, and protein was isolated for immunoblot analysis. Vascular markers CD31 and VE-cadherin were increased in the lasmiditan-treated compared to vehicle-treated cells (Figure 5A). Additionally, phosphorylation of VE-cadherin at Tyr658, which has been found to disrupt VE-cadherin-mediated junctions (Potter et al., 2005), was decreased with lasmiditan treatment. VE-cadherin can directly contribute to the expression of claudin-5, in that adhesion triggers phosphorylation of Akt and, subsequently, phosphorylation of FoxO1, which prevents its translocation into the nucleus where it can repress claudin-5 transcription (Gavard and Gutkind, 2008). Importantly, lasmiditan-treated mBMEC exhibited increased Akt, eNOS, and FoxO1 phosphorylation compared to vehicle-treated cells (Figure 5B).