Transcriptomic data from a non-human primate model of stroke reproducing endovascular thrombectomy

Background and purpose: Rodent models have demonstrated the key pathogenic role of the lectin pathway (LP) of complement activation in brain ischemic injury. Here we assessed the LP activation and its association with stroke severity and outcome in a well-characterized model of stroke mimicking mechanical thrombectomy in non-human primates. Methods: Transient middle cerebral artery occlusion was performed in Macaca fascicularis randomly allocated in cyclosporine A or placebo group. Neurological status was serially assessed. The occlusion lesion, the ischemic penumbra, the final lesion, the lesion growth between occlusion and day 7 as well as the blood-brain barrier permeability using quantitative transfer constant Ktrans ratio and inflammation by quantitative [11C]PK11195 imaging were assessed using positron emission tomography–magnetic resonance (PET-MRI) imaging. Functional residual LP-specific C3b deposition assay was performed in plasma. C3b/iC3b, mannose binding lectin (MBL), MBL associated serine proteases-2 (MASP-2) and interleukin-6 were measured in plasma, cerebrospinal fluid and brain homogenates. Brain tissue was used to perform immunofluorescence and confocal analyses (C3b/iC3b, MBL and MASP-2 and glycocalyx) as well as transcriptomic analysis. Results: Twenty-three subjects were included with a mean age of 6.7 +/- 0.5 years. LP residual activation in plasma collected late after recanalization was significantly decreased compared to baseline conditions. C3b/iC3b peaked at day 7 and this increase correlated with a more severe neurological deficit at day 1 and day 7, a larger final lesion and lesion growth volumes as well as a greater Ktrans ratio. C3b/iC3b, MBL and MASP-2 were found in microvascular thrombi at day 30. At this time point, 760 genes were differentially expressed within the lesion, including 13 LP-related genes. Conclusions: Our results support that the LP of complement activation is involved in thromboinflammation in a non-human primate model of stroke mimicking mechanical thrombectomy in humans. Transient middle cerebral artery occlusion was performed in Macaca fascicularis randomly allocated in cyclosporine A or placebo group. Brains were obtained at day-30. Tissue samples were collected into (“lesion”) and around (“peri lesional”) the ischemic lesion using anatomical landmarks from FLAIR MRI at day 7 as well as in the contralateral mirror area (“contralateral”). RNA samples from snap-frozen data collection were sequenced. Quality and quantity checks were performed by means of Fragment Analyzer (Agilent). Library construction was carried out using NextFlex Rapid Directional mRNA-Seq v2.0 (Bioo-Scientific, PerkinElmer) according to the manufacturer protocol. Libraries were applied to an Illumina flow cell High Output v2.5 and run on the Illumina Nextseq 500 as a single end read for 75 pb. Image analysis and base calling was carried out using the NCS 4.0.1 and RTA 2.4.11 Illumina software suite implemented on the Illumina sequencing machine. Final file formatting, demultiplexing, and fastq generation were carried out using Bcl2fastq v2.17.1.14 and Cutadapt v1.9.1. For trimming the threshold quality score was set to 30, minimum length of 50 bases after trimming, the maximum rate of undetermined bases (“N”s) was set to 30% and adapters were trimmed with 3 bases overlap and a maximum error rate of 10%. Alignment and quantification were performed using Salmon v1.6.0 with Macaca fascicularis (Ensembl Release 106).