12/15-Lipooxygenase Inhibition Reduces Microvessel Constriction and Microthrombi after Subarachnoid Hemorrhage in Mice

Background and Purpose Impaired cerebral circulation, induced by blood vessel constrictions and microthrombi, leads to delayed cerebral ischemia after subarachnoid hemorrhage (SAH). 12/15-Lipooxygenase (12/15-LOX) overexpression has been implicated in worsening early brain injury outcomes following SAH. However, it is unknown if 12/15-LOX is important in delayed pathophysiological events after SAH. Since 12/15-LOX produces metabolites that induce inflammation and vasoconstriction, we hypothesized that 12/15-LOX leads to microvessel constriction and microthrombi formation after SAH, and thus 12/15-LOX is an important target to prevent delayed cerebral ischemia. Methods SAH was induced in C57BL/6 and 12/15-LOX−/− mice of both sexes by endovascular perforation. Expression of 12/15-LOX was assessed in brain tissue slices and in vitro. C57BL/6 mice were administered either ML351 (12/15-LOX inhibitor) or vehicle. Mice were evaluated for daily neuroscore and euthanized on day five to assess cerebral 12/15-LOX expression, vessel constrictions, platelet activation, microthrombi, neurodegeneration, infarction, cortical perfusion, and for development of delayed deficits. Finally, the effect of 12/15-LOX inhibition on platelet activation was assessed in SAH patient samples using a platelet spreading assay. Results In SAH mice, 12/15-LOX was upregulated in brain vascular cells and there was an increase in 12-S-HETE. Inhibition of 12/15-LOX improved brain perfusion on days 4–5 and attenuated delayed pathophysiological events, including microvessel constrictions, microthrombi, neuronal degeneration, and infarction. Additionally, 12/15-LOX inhibition reduced platelet activation in human and mouse blood samples. Conclusions Cerebrovascular 12/15-LOX overexpression plays a major role in brain dysfunction after SAH by triggering microvessel constrictions and microthrombi formation, which reduces brain perfusion. Inhibiting 12/15-LOX may be a therapeutic target to improve outcomes after SAH.


Introduction
Subarachnoid hemorrhage (SAH) affects about 30,000 individuals each year in the United States.Up to 30% of patients who survive aneurysm rupture develop delayed cerebral ischemia (DCI) 4 to 10 days following SAH, which accounts for the most common cause of morbidity and mortality in SAH survivors. 1Multiple clinical studies indicate the cause of DCI is multifactorial and may include microthrombi, brain blood vessel constriction, and in ammation. 2ter SAH, there is a signi cant enhancement in brain arachidonic acid metabolization, which triggers the formation of thrombi and vasospasm, worsening the pathogenesis of SAH. 3,4,512/15lipooxygenases (12/15-LOX) are enzymes that catalyze fatty acids, including arachidonic acid and linoleic acid, 6 into various bioactive lipid metabolites. 7Of note, 12/15-hydroxyeicosatetranoic acids (12/15-S-HETE), which are 12/15-LOX metabolites, are potent pro-in ammatory chemoattractants for neutrophils and leukocytes. 812/15-S-HETE induces the expression of IL-6, IL-12, TNF-α, MCP1, and adhesion molecules in macrophages and vascular cells, 9,10,11 and disrupts endothelial tight junction and barrier function. 1212/15-LOX has been shown to deleterious in cerebrovascular diseases, such as cerebral ischemia 13,14 and early brain injury after SAH. 15 12/15-LOX is known to trigger in ammation, 16 activating platelets, 17 promoting thrombosis 18 and constricting blood vessels. 19As such, 12/15-LOX may cause these events after SAH since it is overactivated after stroke. 20nce 12/15-LOX can cause in ammation, thrombi, and vascular dysfunction, and since it is upregulated after SAH, we hypothesized that inhibiting 12/15-LOX after SAH will reduce microvessel constrictions and formation of microthrombi, thus improving brain perfusion and preventing DCI after SAH in mice.

Materials and Methods
Study approval: All procedures performed on animals were approved by the UTHealth Animal Welfare Committee and conducted according to the NIH Guide for the Care and Use of Laboratory Animals.The results are reported in accordance with ARRIVE (Animal Research: Reporting in Vivo Experiments) guidelines.
Animal Study: Two hundred thirty adult C57BL/6 and fty-ve 12/15-LOX −/− mice (4-6 months old) of both sexes were used.Mice were housed under a 12-h day/12-h night cycle with free access to food and water.SigmaPlot 11.0 was used to estimate all sample sizes using data from previous experiments and preliminary data with α = 0.05 and β = 0.2. 21C57BL6 mice were electronically randomized into either Sham, SAH, Sham + Vehicle, SAH + Vehicle, or SAH + ML351, while 12/15-LOX −/− mice were randomized into sham or SAH.The same individual performed all surgeries and investigators responsible for functional assessment, outcome measurement, and data analysis were blinded to experimental groups, sex, and genotype.
Cerebral blood ow (CBF) and intracranial pressure (ICP) were monitored as previously described 22,23,24 using a laser Doppler probe (Perimed, Järfälla, Sweden) and microcatheter transducer (Millar, Houston, USA), respectively.SAH was induced using the endovascular perforation model as described previously. 21In brief, a 5 − 0 mono lament lament was inserted into the left external carotid artery and advanced towards the Circle of Willis.After con rming SAH induction (an immediate CBF drop of at least 85% or an immediate ICP increase from 1-3 mmHg to 40-80 mmHg), the lament was immediately removed, and the external carotid artery was ligated.Sham-operated animals were treated similarly, with the exception that the lament was not advanced far enough to induce SAH.After recovery from iso urane, mice were observed for up to four hours and only mice without hemiparesis were included in this study.Mice were allowed to survive for up to 5 days post-SAH.Health status was assessed a minimum of three times per day.Mice not surviving from day 1 to day 5 were excluded from all outcomes except mortality, behavior, laser speckle imaging, and delayed neurological de cits (DND).Excluded mice were replaced to satisfy sample size calculations for the primary outcomes (microvessel constriction, microthrombi, infarction, and 12-S-HETE levels).

ELISA
Sham or SAH C57BL/6 mice (n = 5/group/sex/time-point) were euthanized at day 1, 2, 5, or 6 post-SAH for assessment of 12-S-HETE plasma levels.A separate cohort of mice (n = 5/group/sex for C57BL/6 mice, n = 5/group for male 12/15-LOX −/− mice) was euthanized on day 5 to assess the effect of 12/-15-LOX inhibition/knockout on 12-S-HETE plasma levels.Blood was collected via cardiac puncture into a syringe containing 50µL 3.8% citric acid.Blood was centrifuged at 2000g for 15 minutes to obtain plasma.Plasma was stored at -80 o C until measurement.12-S-HETE concentration was measured using an ELISA kit (ab133034, Abcam) according to manufacturer guidelines.
The samples were carefully placed onto microscope slides using Fluoromount-G (0100-01, SouthernBiotech).Slices were observed for 12-LOX, 15-LOX, and laminin expression using a microscope and images were processed using THUNDER (Leica DMI8 Thunder microscope).
Microvessel constrictions: Five days after SAH, mice (n = 6-8/group/sex/strain) were euthanized via cardiac perfusion of PBS followed by gelatin-India ink (1:2 India ink:10% gelatin).Mice were stored at 4 o C overnight, then brains were removed and stored in 4% PFA for 48 hours then stored in PBS with 0.01% sodium azide until imaging.A Zeiss Discovery stereomicroscope was used to image the entire surface of the brain.Vessel constriction was evaluated by determining locations in the artery segments with had a diameter reduction of more than 10%.Large arteries and arterioles were de ned as > 50µm and 10-50µm, respectively.

Platelet morphology
Prior to processing blood samples, coverslips (12-545-101P, Fisher Scienti c) were coated with poly-Llysin (0.1% (v/v) in PBS, 0413, ScienCell) solution and incubated for 30 min at 37°C in 35mm tissue culture dishes (130180, Thermo Scienti c).Thereafter, cover glasses were washed with PBS and dried in air for 30 min.Slides were stored in a dust-free box until use.
Blood was collected from mice (n = 6/group/sex/strain) on day 5 via cardiac puncture into a syringe containing 50µL ACD buffer (8013-89-6, Sigma Aldrich).The blood was mixed with wash buffer (2:1 ratio, buffer contained 10 mM sodium citrate, 150 mM NaCl, 1 mM EDTA, and 1% (w/v) dextrose in Tyrode's buffer (11760-10, EMS) at pH 7.4) and centrifuged for 5 min at 800g and room temperature.The supernatant (platelet-rich plasma) was collected and then centrifuged for 20 min at 100g and room temperature.The supernatant was collected and centrifuged again for 20 min at 800g and room temperature.The supernatant was carefully aspirated until about 100 µL of the solution remained with the pellet.Then 1400 µl of glucose solution (0.1% glucose + 0.3% BSA dissolved in Tyrode's buffer) was added to the sample.The tubes were icked three times and then incubated at 37°C for 15 min.Each sample was placed onto the prepared poly-L-lysin coated glass slides, followed by incubation for 30 min at 37 o C. Slides were then washed 2 times with PBS.After aspiration of PBS, the platelet samples were xed with 4% PFA for 15 min.PFA was aspirated, samples were washed, then 0.3% Triton-X100 was added for 10 min, following by incubation with Phalloidin-647 (1:1000, ab176759, Abcam) for 2 hours.In select samples a CD42a antibody (1:1000, sc-166420, Santa Cruz Biotechnology) was also added to con rm the cells were platelets.Samples were washed and then allowed to dry in the dark overnight at room temperature.The samples were carefully placed onto microscope slides using Fluoromount-G.
Images were taken over the whole area using a uorescence microscope at 100x.Platelet morphology was determined by a blinded investigator for each image.Platelets were categorized as either inactiveted (discoidal shaped) or activated (spikey, partially spread, or spread) based on morphology. 26,27icrothrombi On day 5 post-SAH, mice (n = 6/group/sex/strain) were administered heparin (50µL/10g, 1000U/mL) 5 min before euthanasia which consisted of cardiac perfusion of PBS followed by 4% PFA.Brains were removed and stored in PFA in a 4 o C refrigerator before sectioning into 40µm thick slices.Slices at -2 from bregma were stained with Martius Scarlet and Blue (MSB) and microthrombi were counted throughout the entire slice as previously described. 21urodegeneration Fixed brain slices (n = 6/group/sex/strain) at -2 bregma were stained with Fluorojade C following the manufacturer's protocol (AG325, Millipore).Neurons undergoing degeneration (i.e.exhibited positive staining) were counted throughout the entire slice using a Leica DMi8 microscope.

Infarction
Mice (n = 6-8/group/sex/strain) euthanized on day 5 had one xed brain slice between − 1 and − 2 from bregma stained with crysel violet at room temperature following manufacturer methods.The entire slice was imaged to quantify the total area of infarcted tissue.

Laser speckle contrast imaging of brain perfusion
Male mice (n = 6/group/strain) were used to study cortical brain perfusion using Laser Speckle Contrast imaging.Mice received a cranial window implantation of stacked coverglass (three 3mm glasses stacked on a 5mm glass) 3-4 weeks before SAH.Brie y, the skull was exposed in anesthetized mice.Then a 4mm diameter circle was created using a microdrill, ensuring that the edges were at least 1mm away from the midline and the bregma/lambda sutures.After skull removal and bleeding cessation, the stacked glass coverslip was placed into the hole after being lled with aCSF.Dental cement was applied around the stacked 5mm glass edge to secure it to the skull.A custom-made head frame was attached using super glue and dental cement. 28The mice were allowed to recover at least 3 weeks before SAH.
On the day of SAH, the cranial windows were examined and mice which had windows that were opaque or not secure were excluded and replaced.Brain perfusion data was collected for 10 min before SAH (baseline perfusion) and for up to 80 min post-SAH using an in-house built laser speckle imaging system. 29At 15 min after SAH, recording was paused to inject ML351 or vehicle.Imaging was also performed for 10 min on days 1-5.Data was analyzed using the MatLab algorithms developed by the Functional Optical Imaging Laboratory. 29,30In brief, ROIs were drawn to measure cerebral perfusion of the MCA, ACA, and watershed territories.Within each ROI, the cerebral perfusion units were normalized to the baseline perfusion units and are presented as percent of baseline.

Neurobehavior and DND
All mice surviving more than 1 day are included in neurological assessment.Daily behavioral performance was assessed 1-5 days post-SAH using an 8-test sensorimotor neuroscore which evaluates functional performance in exploration, climbing, forelimb and hind limb use, whisker and side sensation, balance, and visual re ex. 31 In brief, the maximum score of 24 corresponds to no de cits and the minimum score of 0 is unresponsive.DND is classi ed as mice experiencing a reduction of more than 4 points in the neuroscore after recovery from the day 1 neuroscore.Delayed death on days 3-5 (if some functional recovery was observed on prior days) is also considered as developing DND as the neuroscore would be 0. Mice which exhibited continuous neuroscore decline from day 1 are not included in DND analysis.
Human Study: All procedures performed on human samples were approved by UTHealth and conducted according to the NIH Guidelines.Aneurysmal SAH patients and control human blood were enrolled (via written informed consent prior to participation) into the study under a protocol approved by the UTHealth Institutional Review Board.Blood was collected from fourteen aneurysmal SAH patients (con rmed via CT angiography) on days 1, 2, 4, and 7 post-rupture, and from nine control humans, into BD Vacutainer ACD-B (364816, Becton Dickinson and Company) between 9:00a and 4:00p.Blood was processed within 30 min after it was drawn.Brie y, the blood was mixed with wash buffer (1:1 ratio) in a 15mL falcon tube and processed as described in the "platelet morphology" outcome.After the three centrifugation steps as described in "platelet morphology", supernatant was vacuum sucked and platelet washing buffer (1500µL) was added to the pellet, and the tube was icked three times to resuspend the pellet.The solution was then centrifuged for 8 min at 100g and room temperature.Thereafter, the supernatant was collected and platelets counted (Hemavet 950 FS, Drew Scienti c).Following platelet counting, 750µL of platelet sample was put into two tubes.One tube had 100µL of ML351 ( nal concentration of 10µM) added and the other tube had 100µL of vehicle (10% solutol, 10% Chremophor EL, and 20% PGE400 in 0.9% NaCl saline) added.The tubes were icked three times and then incubated at 37°C for 15 min.Then each sample was placed onto the prepared poly-L-lysin coated glass slides, and processed as described in the "platelet morphology" outcome.

Statistical analysis:
Unless otherwise speci ed, data are presented as mean and SD with individual values.All outcomes were tested for normality and homoscedasticity, and if failed, the equivalent non-parametric tests were used.Multiple groups were analyzed using one-way ANOVA with Tukey post hoc or Kruskal-Wallis with Dunn's post hoc.Laser speckle contrast imaging data was analyzed using two-way ANOVA.Neuroscore data was analyzed using two-way ANOVA on ranks (Friedman) followed by Wilcoxon signed-rank posthoc then corrected using a Bonferroni correction.DND incidence was analyzed using a log-rank test.
Unpaired t-tests were used to compare sex differences, differences between 12/15-LOX −/− Sham and SAH, and for the human platelet morphology outcome.All calculations were performed using a SPSS v28 and Graphpad Prism 6. Differences were considered to be signi cant at p < 0.05.

SAH causes increased 12/15-LOX expression in brain vasculature
Before examining the impact of 12/15-LOX on SAH outcomes, we stained the brain for 12/15-LOX to identify where 12/15-LOX is expressed and if there are changes after SAH.Using immunostaining of male brains, we observed that SAH induces an increase in 12/15-LOX expression in the brain vasculature on day 5 post-SAH (Fig. 1A-C).The cortical vessels had the highest expression of 12/15-LOX.In male mice, the 12/15-LOX expression was signi cantly more elevated on day 5 compared to day 1 (Sham vs SAH D5 p = 0.0034 and SAH D1 vs SAH D5 p = 0.0003, Fig. 1D).Female mice did not have a statistically signi cant increase in brain 12/15-LOX expression in female mice (Fig. 1D).As 12/15-LOX expression was upregulated in the brain vasculature, to identify if brain endothelial cells or pericytes have more expression after SAH, we performed cell culture of human and mouse brain microvascular endothelial cells and human pericytes subjected to hemoglobin toxicity.Western blot analysis of 12/15-LOX expression by pericytes and microvascular endothelial cells indicates increased expression of 12/15-LOX after hemoglobin injury (Fig. 1E).

12/15-LOX inhibition reduces platelet activation and microthrombi formation on day 5
As 12-S-HETE is also known to promote activation and aggregation of platelets, 17,18 and since microthrombi are reported to be part of DCI after SAH, 21 we measured the effect of 12/15-LOX inhibition on platelet morphology changes and brain microthrombi.As assessed using the platelet spreading assay, 5 days after SAH, platelets are signi cantly activated in SAH + Vehicle mice compared to Sham + Vehicle mice (male: p < 0.0001; female: p < 0.0001) (Fig. 5A-C).SAH mice treated with ML351 showed a signi cant reduction in activated platelets compared to SAH mice treated with vehicle (male: p < 0.0001; female: p < 0.0001).
To test human relevance of platelet activation after SAH, and to test if 12/5-LOX inhibition could alter platelet morphology changes, we assessed platelet spreading in human platelets at several time-points post-SAH.Compared to control patient blood, SAH induces a signi cant increase in platelet activation for up to 7 days post-SAH (Fig. 5D).Since platelets use 12-S-HETE to self-regulate (i.e.promote activation via 12-S-HETE release), 32 we examined if ML351 could prevent platelet spreading.Human SAH platelets treated with ML351 showed signi cantly less platelet activation than platelets treated with vehicle 2 and 4 days after SAH (D1: p = 0.091, D2: p = 0.0399, D4: p = 0.0062, D7: p = 0.1749) (Fig. 5).

12/15-LOX inhibition improves brain perfusion on days 4 and 5
Following SAH, brain perfusion becomes compromised.To examine if 12/15-LOX inhibition could restore the lessened brain perfusion, we performed laser speckle contrast imaging in male mice and evaluated the MCA, ACA, and watershed territories via a cranial window.While the MCA and ACA areas had little improvement in perfusion with ML351 treatment, the watershed region displayed signi cantly improved brain perfusion in treated mice on days 4 (p = 0.0457 SAH + Vehicle vs SAH + ML351) and 5 (p = 0.0063 SAH + Vehicle vs SAH + ML351) (Fig. 8).
12/15-LOX inhibition improves neurological behavior and prevents the development of DND Neurological behavior was assessed using a neuroscore each day.Vehicle-treated mice with SAH had signi cant neuroscore de cits compared to Sham + Vehicle, whereas ML351 treated mice had improved behavior at day 1 (male p = 0.024, female p = 0.105) which remained for at least 3 days (Fig. 9).
As DCI is a major contributor to poor outcome for SAH, we measured cerebral infarction on day 5 and assessed for the development of delayed neurological de cits which are hallmarks of clinical DCI (Fig. 10).The infarct volume is signi cantly higher in SAH female mice than sham (p = 0.0032) which can be signi cantly reduced by ML351 treatment (p = 0.0457).However, there was no signi cant infarction in male SAH mice (Sham + Vehicle vs SAH + Vehicle p = 0.1622) and no signi cant reduction of infarction after treatment with ML351 (p = 0.6405 vs SAH + Veh) (Fig. 11).Inhibition of 12/15-LOX by ML351 lead to a signi cant reduction in DND incidence for male SAH mice (p = 0.0329) but not for female SAH mice (p = 0.4425).There is also a tendency for less DND incidence in male ML351-treated mice as compared to female ML351-treated mice (p = 0.0714) (Fig. 10).

Discussion
Approximately 30% of patients who survive aneurysm rupture develop DCI 4 to 10 days post-SAH, which is the major cause of morbidity and mortality among SAH survivors. 1Various clinical studies indicate that the cause of DCI is multifaceted, involving factors such as microthrombi, vasospasm, and in ammation. 2In this study, we examined the role of 12/15-LOX in delayed microvessel constrictions and microthrombi after SAH.For the rst time, we observed that 1) 12/15-LOX is upregulated in brain microvessels after SAH, 2) a 12/15-LOX metabolite, 12-S-HETE, expression peaks day 5, 3) inhibiting 12/15-LOX can reduce delayed vasospasm, microthrombi, and neuronal degeneration, resulting in improved cerebral perfusion which ultimately improved outcomes following SAH (including reducing DND incidence), 4) platelet spreading is increased by SAH and inhibition of ML351 can prevent platelet spreading in humans and mice, and 5) there are sex differences in the response to ML351 treatment.These ndings collectively demonstrate a critical role of 12/15-LOX in the pathophysiology of SAH and warrants investigation to determine its potential as a therapeutic target to mitigate the adverse effects of SAH.
Following an insult, such as SAH, arachidonic acid is released from cell membranes by phospholipases 33 and arachidonic acid metabolism leads to metabolites which promote in ammation. 34Arachidonic acid metabolism is enhanced, especially within the brain cortex, after SAH via both the cyclooxygenease and lipooxygenase pathways which may be responsible for arterial constrictions and blood clot formation. 3 was recently reported that 12/15-LOX is involved in early brain injury after SAH.Speci cally, 12/15-LOX was overexpressed in macrophages triggering in ammation leading to edema formation and neuronal cell death. 15Other studies have suggested that early brain injury may be a factor involved in causing DCI, 35,36 so we sought to investigate if 12/15-LOX could be a link between early brain injury and DCI.Herein, we observed that whole brain levels of 12/15-LOX are relatively unchanged at 1 day, but signi cantly elevated on day 5 (Fig. 1E, Supplemental Fig. 1).We also observed a high expression of 12/15-LOX within the penetrating arterioles of the brain cortex on day 5, which is likely due to increased expression by microvascular endothelial cells (Fig. 1A-C, Supplemental Fig. 1A-E).
12/15-S-HETE, which induces the expression of many in ammatory markers and adhesion molecules in macrophages and vascular cells, 9,10,11 and disrupts endothelial tight junction and barrier function, 37 is highly increased in blood plasma during the DCI phase.This increase may be a cause of delayed vessel constriction and microthrombi formation, potentially to its impact on platelets.
While large vessel vasospasm can be prevented with various pharmacological interventions after SAH, 38,39 DCI incidence remains high.There are limited studies identifying potential mechanisms responsible for microvessel constrictions, which are known to be present after SAH. 40As 12/15-LOX is elevated in microvessels, and as no study has addressed the impact of 12/15-LOX on the formation of cerebral small vessel constriction and microthrombi after SAH nor 12/15-LOX involvement in brain perfusion, we explored the pathological interaction between 12/15-LOX, microvessels, and platelets.We found that increased 12/15-LOX metabolites, as 12-S-HETE, may play an important role in platelet activation and small vessel constrictions, impacting brain perfusion, since 12/15-LOX inhibition attenuated these pathophysiological events.Despite providing reduced microvessel constriction and platelet spreading in females, only male mice exhibited a reduction in the DND incidence.There are several potential reasons.First, the higher 12-S-HETE levels in females even after ML351 treatment may impede recovery from SAH.Second, we chose a treatment regimen that worked to prevent EBI in male mice with SAH.So, it may be necessary to use a different treatment regimen (dosing, timing) better suited for female mice to address this disparity.
In patients developing DCI, cerebral perfusion is thought to be reduced in localized areas, leading to infarction or DND.To examine if 12/15-LOX impacts brain perfusion, we performed laser speckle contrast imaging.We found that inhibiting 12/15-LOX can signi cantly improve the perfusion of the watershed areas but had no effect on the large artery perfusion.Since watershed areas make up around 10% of all brain infractions after ischemic stroke, 41 they may also play an important role for SAH.The cause of improved watershed area perfusion by 12/15-LOX inhibitions remains unknown, but may be due to fewer microthrombi or less microvessel constrictions.Regardless, if 12/15-LOX inhibition improves watershed perfusion after SAH, it may be a therapeutic target to improve watershed perfusion for cerebrovascular diseases.

Clinical Implications
It is well-known that arachidonic acid metabolism is deleterious following cerebrovascular diseases, including SAH. 42 Arachidonic acid can be metabolized by three major classes of enzymes, namely cytochrome P450, cyclooxygenase (COX), and LOX. 43COX, which promotes in ammatory response and platelet function, 44,45,46,47,48 may not be able to decrease the incidence of DCI.COX inhibitors, such as aspirin, have been tested in clinical studies and overall do not decrease the incidence of DCI. 49ytochrome P450 is also unlikely to be a major factor in arachidonic acid metabolism following SAH as there are low brain levels of cytochrome P450 and restricted expression. 50,51Together with the study by Gaberel et al., 15 12/15-LOX may be a key player in the metabolism of arachidonic acid after brain injury.But one caveat to note is that arachidonic acid can be metabolized by any one of these three enzymes.Thus, inhibiting one enzyme may increase the metabolism of arachidonic acid by another enzyme, thereby producing more of its metabolites.So a question that remains is can inhibition of a single arachidonic acid enzyme be enough to promote recovery or do COX and LOX need to be inhibited simultaneously?
By discovering the critical role of 12/15-LOX, an in ammation regulator, 16 in microvessel constrictions and platelet activation, our data suggests that effective inhibition of 12/15-LOX may lead to a reduction of platelet activation, microthrombi and small vessel constrictions (Fig. 12), ultimately contributing to the reduction of DCI after SAH.

Limitations
This study is not without limitations.First, ML351 is cell-permeable and appears to be non-reductive and reversible. 52ML351 exhibits greater than 250-fold selectivity towards its primary target, 12/15-LOX (IC50 = 200nM), compared to related LOX isoenzymes such as 5-LOX.However, we cannot exclude the potential protective effect of inhibited 5-LOX with ML351 (IC50 = 50µM) on the outcome. 52,53ML351 inhibits 12/15-LOX throughout the body.Our data suggests that 12/15-LOX expression is increased in endothelial and pericyte cells, and Gaberel et al., reports an increased expression of 12/15-LOX in macrophages. 15Thus we cannot state whether the observed effects are due to one cell type or a combination; future studies should investigate selectively targeting 12/15-LOX within these cell types to identify which cell(s) are key regulators of SAH pathophysiological events.
Second, a component of the vehicle used for ML351, Cremophor EL, possesses a documented neuroprotective effect. 54In this study, we also observed that the vehicle may reduce platelet activity after SAH (Supplement Fig. 4).Yet, ML351 clearly has a protective effect since it signi cantly improved several outcomes compared to vehicle-treated mice.The protective effect by Cremophor EL, especially on platelets (Supplemental Fig. 4) may be an explanation for why female mice with SAH did not have elevated microthrombi counts (compared to sham) in this study which is different from our other studies. 55 mentioned above, there are multiple mechanisms of metabolizing arachidonic acid.In this study we only examined 12/15-LOX.When 12/15-LOX is inhibited, arachidonic acid may be metabolized through other ways as 5-LOX, metabolizing arachidonic acid to 5-HPETE which can be further metabolized to n-HETE, balancing out the missing HETE from 12/15-LOX. 43Also mentioned above, the arachidonic acid may also be metabolized by COX which also contributes vasoconstrictors and pro-in ammatory metabolites. 43,56Additional studies are needed to examine the effect of 12/15-LOX inhibition on arachidonic acid metabolism and arachidonic acid enzyme function.
Another limitation of this study is that we only used a single dose administered at an early time-point after SAH.Since there were mixed bene cial effects, especially in females, we may not have used the optimal dosing regimen.Testing other doses or timing of administration would aid in not only identifying the best regimen, but would also help to determine if there are sex-speci c bene ts by ML351.Furthermore, no long-term study was conducted.

Conclusion
In this study, we observed that inhibiting 12/15-LOX leads to a reduction in delayed microvessel constriction and microthrombi formation, enhances cerebral perfusion, and consequently results in less neurological de cits and DCI.While interference with the arachidonic acid-COX pathway has seen limited clinical translation, this study, along with the work by Gaberel et al. 15 indicates that arachidonic acid metabolism by 12/15-LOX may represent a crucial mechanism following SAH.

Figures
Figures

Figure 1 See
Figure 1

Figure 3 See
Figure 3

Figure 4 See
Figure 4

Figure 5 See
Figure 5

Figure 11 See
Figure 11