A Mouse Brain-based Multi-omics Integrative Approach Reveals Potential Blood Biomarkers for Ischemic Stroke

Abstract

Stroke remains a leading cause of death and disability worldwide. Despite continuous advances, the identification of key molecular signatures in the hyper-acute phase of ischemic stroke is still a primary interest for translational research on stroke diagnosis, prognosis, and treatment. Data integration from high-throughput -omics techniques has become crucial to unraveling key interactions among different molecular elements in complex biological contexts, such as ischemic stroke. Thus, we used advanced data integration methods for a multi-level joint analysis of transcriptomics and proteomics data sets obtained from mouse brains at 2 h after cerebral ischemia. By modeling net-like correlation structures, we identified an integrated network of genes and proteins that are differentially expressed at a very early stage after stroke. We validated 10 of these deregulated elements in acute stroke, and changes in their expression pattern over time after cerebral ischemia were described. Of these, CLDN20, GADD45G, RGS2, BAG5, and CTNND2 were next evaluated as blood biomarkers of cerebral ischemia in mice and human blood samples, which were obtained from stroke patients and patients presenting stroke-mimicking conditions. Our findings indicate that CTNND2 levels in blood might potentially be useful for distinguishing ischemic strokes from stroke-mimicking conditions in the hyper-acute phase of the disease. Furthermore, circulating GADD45G content within the first 6 h after stroke could also play a key role in predicting poor outcomes in stroke patients. For the first time, we have used an integrative biostatistical approach to elucidate key molecules in the initial stages of stroke pathophysiology and highlight new notable molecules that might be further considered as blood biomarkers of ischemic stroke. Stroke remains a leading cause of death and disability worldwide. Despite continuous advances, the identification of key molecular signatures in the hyper-acute phase of ischemic stroke is still a primary interest for translational research on stroke diagnosis, prognosis, and treatment. Data integration from high-throughput -omics techniques has become crucial to unraveling key interactions among different molecular elements in complex biological contexts, such as ischemic stroke. Thus, we used advanced data integration methods for a multi-level joint analysis of transcriptomics and proteomics data sets obtained from mouse brains at 2 h after cerebral ischemia. By modeling net-like correlation structures, we identified an integrated network of genes and proteins that are differentially expressed at a very early stage after stroke. We validated 10 of these deregulated elements in acute stroke, and changes in their expression pattern over time after cerebral ischemia were described. Of these, CLDN20, GADD45G, RGS2, BAG5, and CTNND2 were next evaluated as blood biomarkers of cerebral ischemia in mice and human blood samples, which were obtained from stroke patients and patients presenting stroke-mimicking conditions. Our findings indicate that CTNND2 levels in blood might potentially be useful for distinguishing ischemic strokes from stroke-mimicking conditions in the hyper-acute phase of the disease. Furthermore, circulating GADD45G content within the first 6 h after stroke could also play a key role in predicting poor outcomes in stroke patients. For the first time, we have used an integrative biostatistical approach to elucidate key molecules in the initial stages of stroke pathophysiology and highlight new notable molecules that might be further considered as blood biomarkers of ischemic stroke. Stroke remains a major leading cause of death and disability worldwide (1Benjamin E.J. Blaha M.J. Chiuve S.E. Cushman M. Das S.R. Deo R. de Ferranti S.D. Floyd J. Fornage M. Gillespie C. Isasi C.R. Jiménez M.C. Jordan L.C. Judd S.E. Lackland D. Lichtman J.H. Lisabeth L. Liu S. Longenecker C.T. Mackey R.H. Matsushita K. Mozaffarian D. Mussolino M.E. Nasir K. Neumar R.W. Palaniappan L. Pandey D.K. Thiagarajan R.R. Reeves M.J. Ritchey M. Rodriguez C.J. Roth G.A. Rosamond W.D. Sasson C. Towfighi A. Tsao C.W. Turner M.B. Virani S.S. Voeks J.H. Willey J.Z. Wilkins J.T. Wu J.H. Alger H.M. Wong S.S. Muntner P. American Heart Association Statistics Committee and Stroke Statistics SubcommitteeHeart disease and stroke statistics-2017 update: a report from the American Heart Association.Circulation. 2017; 135: e146-e603Crossref PubMed Scopus (6264) Google Scholar) and is a major public health problem that contributes to rising costs of healthcare in many developed countries (2Wang G. Zhang Z. Ayala C. et al.Costs of hospitalization for stroke patients aged 18-64 years in the United States.J. Stroke Cerebrovasc. Dis. 2016; 3: 973-982Google Scholar). Effective therapies based on reperfusion mechanisms are currently used (3Fiehler J. Gerloff C. Mechanical thrombectomy in stroke.Dtsch. Arztebl. Int. 2015; 112: 830-836PubMed Google Scholar, 4Lansberg M.G. Bluhmki E. Thijs V.N. Efficacy and safety of tissue plasminogen activator 3 to 4.5 hours after acute ischemic stroke: a metaanalysis.Stroke. 2009; 40: 2438-2441Crossref PubMed Scopus (191) Google Scholar), but only a small percentage of ischemic stroke patients currently benefit from them. The reason is the narrow therapeutic time window and absolute contraindications for those who of S. L. M. P. M. A. C. E. A. M. Stroke and thrombectomy in and the from a stroke 2015; PubMed Scopus Google Scholar, S. Willey J.Z. P. American Heart Association Stroke on and Stroke on and on of and and of after in acute ischemic stroke: a for healthcare from the American Heart Stroke 2017; PubMed Scopus Google Scholar). Thus, new therapeutic are to the ischemic and of ischemic stroke might the of patients that benefit from stroke is based on and techniques of acute 2015; Google Scholar), which are still in primary healthcare and biomarkers are to become for these L. stroke biomarkers in PubMed Scopus Google Scholar, in ischemic 2015; PubMed Scopus Google Scholar, S. of biomarkers for acute stroke in the PubMed Scopus Google Scholar) might the of stroke from that Furthermore, blood biomarkers might also stroke for and to poor at stages of the disease A. D. M. C. M. C.J. M. D. J. M. J. J. of blood in the of after stroke: and PubMed Scopus Google Scholar, A. A. D. G. L. A. E. A. A. J. new in human PubMed Scopus Google Scholar). of the stroke pathophysiology is to the identification of molecular of ischemic stroke and therapeutic could to the of ischemic -omics techniques have as that the identification of biological changes in the at different molecular levels L.C. proteomics and Google Scholar). of data sets from different techniques are crucial to the among different molecular and findings in a complex biological analysis of data for and of complex human public 2016; PubMed Scopus Google Scholar). Thus, the of techniques integrative is to the of the molecular ischemic stroke and might a the of the molecular over the of The of to the transcriptomics and proteomics changes that in the brains of mice the hyper-acute phase of cerebral ischemia. The is to biomarkers of ischemic stroke. We to data from techniques an integrative analysis to of genes and proteins in the very first stages of the stroke integrative approach has to a of molecules that are and could be further in the brains and blood of mice at different time after cerebral ischemia. we the of as biomarkers for the of ischemic stroke in a The is 3 a first phase and a in mouse samples, and a phase on mice and human and blood for the initial phase based on for the proteomics and transcriptomics were for the to the from the phase for of a of at of the a of 6 The of mouse and human used for the phase to and and data are were in the and in the of the and were the Committee of the mice were used for the were in a on a and were to to and for for in were in a and in to the C. M. research The for PubMed Scopus Google Scholar). of were used to the the were 3 were after the reperfusion after of the and death the the were 6 were of reperfusion after of the death the poor in the cerebral for an as M. cerebral in the PubMed Scopus Google Scholar). that a of of after and a of after were in the could for 4.5 h 2 h 6 h from the of the to their the of the were at their and blood were in and at at for The the at further were to blood from after mouse brains were and in conditions. The to the the of the tissue S. M. J. cerebral methods in mouse of cerebral ischemic PubMed Scopus Google to the and in and at were in and of and were the and were at further The phase on and from from ischemic 2 h after the were also in phase to the of and differentially of the ischemic from mouse were a and the The used to expression on a on a of of from the to the and to the the and and to expression and of data were in after a A. R. for of 2009; PubMed Scopus Google Scholar), and of PubMed Scopus Google Scholar) used for data to and of expression genes the of a and a to the were from the data and the of genes considered for the of differentially expressed elements based on a analysis for the and methods for expression in 3: Scopus Google Scholar). for were also the a and approach to Scholar). the to the expression for Data on the after stroke from mouse were the of were in conditions and in that were to of the were small and 10 for at for at and were in and at at and further The were in a and at further from were a approach in a were an a were at a and from the 3 a phase of in and and a from to of for at a of were a of to a The in a a to in the a of The were for in the their a of were to for and for The time and in the and The to and to a window of 10 were from further for For a window of and of 2 were the for to for proteomics used for data analysis The from of the were and were to a were and to first of based on the of of the the for proteomics within the to from to of time, and 2 to were considered for identification and were and the identification the to 10 and to considered as and of as as the to for identification to for the and were further to as a the were and a the to for different conditions. based on the of the within the of identified sets were to the of those proteins and identified at 2 and to to the as were considered for the analysis and are in Data were further in were and M.J. and the biological content of PubMed Scopus Google Scholar). of differentially expressed proteins based on a analysis in the M.E. Wu D. C.W. expression for and 2015; PubMed Scopus Google Scholar). for were also the a and approach to Scholar). the to the expression for Data on the after stroke For the integrative only from ischemic were and a were in the of differentially expressed and data sets in the identification of to the and from as which were for the integrative Thus, the integrative of differentially expressed genes and proteins were only from the from these integrative expression in the ischemic a data were and and proteins data sets were used for the integrative different integrative were of of C. approach to the integration of of PubMed Scopus Google in J. G. an for analysis of expression PubMed Scopus Google to the and data sets for of that data were to the and data were and a of the to a of and and within the used to a S. an to 2009; PubMed Scopus Google and data data were and and the of the were a of and 2 network for a correlation of the C. A. A. integrative analysis of 2016; to and genes and proteins data sets sets from the analysis based on the of the J. analysis Statistics Data Scopus Google Scholar) and the sets to in for A. C. M. P. The 2015; PubMed Scopus Google and molecular were of the data sets to data sets different the those sets in joint data and that of from the integrative network are The also and to to be C. A. A. integrative analysis of 2016; Scholar). were further to data and The phase on from 6 ischemic obtained 2 h after ischemia. were further evaluated on from 6 ischemic 6 h after ischemic and were obtained from the mouse brains the as in the The at the The to were for genes from the 3 genes and of the from mouse the the of were the the and to the for and data of were at of Data analysis and the were to the of the a to the of the were expressed as of the the for in levels and were at 2 h and at 6 h were evaluated the the were the as a 2 as S.E. C. R. Liu E. L. R. L. R. J. E. J. M. D. E. C. R. J. J. Rodriguez R.R. J. G. A. S. in and for a PubMed Scopus Google Scholar). from mouse were to to in content the and of were were 10 for and for at of for at were to a of in a of at The and the at further from the 3 proteins and were from the on the phase from data of were from of the a of the correlation and of over a to of from of were from which were were in the were at a 3 and from the 3 the used for the phase analysis The in used to of the from which a based on their for in The time to and For an of 2 data were to to the in a of the 3 to that the were used to in the For of different were for of the the were used as a to the and the were used to levels within of were from their and to the of the For those proteins the of also for were and the for were expressed as of the the in levels and were at 2 h and at 6 h were evaluated the the were the and were evaluated of is in The phase on and blood from the ischemic and obtained 2 h and 6 h after ischemia and from were further evaluated as stroke blood biomarkers on human blood from ischemic stroke patients and patients from stroke-mimicking conditions. human were the Committee of and obtained from in the of and patients were to the of the from to within the first 6 h after patients a of to differentially ischemic stroke and stroke-mimicking conditions stroke the of Stroke and obtained and data from patients. ischemic stroke patients and patients from stroke-mimicking conditions were in of the stroke patients the and thrombectomy to the 3 ischemic stroke patients The of at evaluated to the patients a 2 were as and patients a from 2 to 6 as from patients were on h from and of treatment. were in and at at for The the at further were evaluated of and were used for and for levels the and and different time were levels after were those after For human used to for expressed as continuous were were used to the of GADD45G and CTNND2 circulating levels and for the biomarkers potentially and the of the used to stroke stroke prognosis, were at were as and were and at the levels of were also on of their to and to new The used to the of of the and the integrated used to the of to the a considered at a of the is in We the and changes that early in the after cerebral ischemia obtained at 2 h after a to ischemia a The changes were identified transcriptomics and proteomics cerebral genes and proteins were to be differentially and only 2 genes and proteins were and after the of the and after cerebral ischemia is also the of genes and proteins to key deregulated after stroke, as in the mechanisms cerebral ischemia we a integration analysis of -omics data sets from genes and proteins that were deregulated after ischemia methods were used to data in these integrative we the and in the We to which that integrated data from data sets the to changes of cerebral ischemia. data were different in the and only a We correlation and data which we identified an integrated network elements are and were among genes and were and and 10 an role of these in the pathophysiology of stroke. integrated were to their in the hyper-acute phase of cerebral ischemia in in Furthermore, 3 were also of their leading in the analysis of the data and and in the data genes and were to be further in a new of ischemic and data sets from were used to the of of cerebral ischemia were to be deregulated of cerebral ischemia of were after the expression and of the in and are for are in h after h after over h after h 6 expression in a new also obtained the biological and molecular in which these were the that of have a role in the are in and the of the which are in acute cerebral ischemia. changes in the were at early time as in the acute pathophysiology of stroke. and of were also in of those activator and of these that genes were proteins to the of the at the time the of the and the and were evaluated in a new of the and of the genes were in new of 2 h after and in the the and in the the 2 of the proteins were also and CTNND2 levels a in the their and were also in the at 6 h after The expression of and at levels at 6 h after the ischemic and in the were the 2 time the genes and a in their expression at 6 h after that at 2 but the expression in the the the early the changes and at 6 h after ischemia. time the expression of also to that of over of levels further in the the at 6 h after CTNND2 levels their initial at time the further on and molecules an and deregulated expression in the after ischemia. The and were first evaluated at the to expression levels of in the at 2 h after ischemia and to to levels at a time the GADD45G content the and at 2 h after the ischemic but a in the were in levels and at 2 h and 6 h after ischemia. their role as stroke the levels of CLDN20, GADD45G, RGS2, BAG5, and CTNND2 were also evaluated in blood from mice to cerebral ischemia 2 h and 6 and levels were in of mouse blood levels of GADD45G and were 2 h after cerebral ischemia in and circulating GADD45G levels 6 h blood content and from levels in the the CTNND2 content in in mice that stroke the only at 6 h after ischemia 6 h among the further the of the biomarkers for ischemic stroke in we evaluated their circulating levels in human blood from patients from ischemic stroke and stroke-mimicking conditions The and of these patients are in ischemic stroke patients at and a of the and were for CTNND2 blood levels were in ischemic stroke patients and very and to the of patients blood levels of CTNND2 to be in ischemic stroke the at the of The ischemic stroke strokes to the and a of the but in the circulating levels of CTNND2 could be of these we a for CTNND2 and to the ischemic stroke from the and analysis to the of these to stroke. The analysis the at the of and for and The that as the only of ischemic stroke. the the of the we CTNND2 as a to We that the also as an to patients stroke-mimicking and the ischemic stroke the of the CTNND2 also the of the and the to for ischemic stroke and poor at For and are were to the is used as a is for and for the of are in Stroke at in a new GADD45G blood levels were in ischemic stroke patients and patients from the were these and GADD45G blood levels at were in patients who poor outcomes those who outcomes patients those poor outcomes were strokes to the and a of a of and a of in the circulating levels of GADD45G could be these and we a for GADD45G that poor at which and analysis for and at that and were the only of poor of the patients at The of GADD45G as a to that GADD45G as an of at the time The of to the the of the to the Furthermore, GADD45G blood levels the of the the of patients who poor the translational of stroke has an integrative analysis of transcriptomics and proteomics data from mouse brains very after cerebral ischemia. a an integrated network genes and proteins a in the hyper-acute stages of stroke pathophysiology has CTNND2 and GADD45G were identified as blood biomarkers for acute ischemic stroke and prognosis, have at unraveling the molecular changes that in the cerebral ischemia. of these have based on the of only have the pathophysiology of stroke to the of a still of the mechanisms of cerebral ischemia at different molecular the of of and to stroke of their complex and is to further the identification of of stroke. Thus, we used developed to transcriptomics and proteomics data sets to the genes and the the and of the which were for genes and The of data that genes and proteins might different but to the integrated at the time biological might also be the correlation such as the of the proteins and and which are to be D. G. A. D. and and 2016; PubMed Scopus Google Scholar). We the biostatistical to the the differentially genes and which complex among different molecular elements of stroke We that of the were in biological that are early after cerebral the of and as as the of the C. K. M. in ischemic J. PubMed Scopus Google Scholar). of these have in the as of the stroke and the of findings and to integrative approach R. J. a role for PubMed Scopus Google Scholar, Effective a of PubMed Scopus Google Scholar, M. M. R. S. E. of in 2016; PubMed Scopus Google Scholar, D. S. M. R. M. of a that death and ischemic PubMed Scopus Google Scholar, S. M. S. M. the cerebral ischemia in PubMed Scopus Google Scholar). be that that genes a to the molecular for might be the early time after ischemia at which we for many different that after expression that might also these the and data sets C. the of from and PubMed Scopus Google Scholar). the we on and the of ischemic stroke as blood a deregulated in the the hyper-acute phase after cerebral which the Thus, research could be to their levels at the therapeutic GADD45G and CTNND2 to be blood biomarkers of ischemic stroke and diagnosis, GADD45G is a of the of which are in acute and A. J. genes as therapeutic in 2009; Google Scholar, in and PubMed Scopus Google Scholar). We have a in expression in the early after ischemia. in the levels GADD45G as J. K. M. K. of the in the PubMed Scopus Google Scholar, M. A. K. S. of genes in mouse after ischemia PubMed Scopus Google Scholar, C. P. E. C. G. E. S. as a of ischemic PubMed Scopus Google Scholar). The of GADD45G in the after stroke is still the role of and D. levels of in PubMed Scopus Google Scholar, mice in PubMed Scopus Google Scholar), GADD45G is also to in and to ischemic A. J. D. A. P. death and 2015; PubMed Scopus Google Scholar). in has a of mechanisms which GADD45G could and which on the and the conditions a new that is in PubMed Scopus Google Scholar). role still in the of stroke, of the of GADD45G to death could also be as a therapeutic approach in the of stroke. therapeutic GADD45G has as a blood GADD45G levels the of ischemic stroke from stroke-mimicking but in the of stroke in the phase of the is for and the of GADD45G blood levels were in patients that early poor outcomes after stroke. the of has circulating GADD45G and stroke GADD45G has only the of disease in S. J. expression is of J. 2009; PubMed Scopus Google Scholar) and has as a of in tissue C. expression as a and in 2015; PubMed Scopus Google Scholar). the of circulating GADD45G levels are still is only that GADD45G is to the of expression in circulating is S. P. M. L. GADD45G and in 3: PubMed Scopus Google Scholar). GADD45G also further the of such as and and in PubMed Scopus Google Scholar). the of GADD45G blood levels as a is and further to these new findings be of CTNND2 is a that is in the and C. M. Liu J.T. for the of and in 2009; PubMed Scopus Google Scholar), a role as a and K. M. M. L. A. leading to PubMed Scopus Google Scholar). CTNND2 also and to the and to a of the and the M. M. K. and from PubMed Scopus Google Scholar). an acute to cerebral we have a in CTNND2 expression and in ischemic Our are those of et who that the of CTNND2 in also to from and very after the is The of in CTNND2 levels after ischemia has CTNND2 is to the which and is still which the of genes in and J. R. Z. R. J. L. to mice PubMed Scopus Google Scholar, J. Zhang J. D. C. a and 2015; PubMed Scopus Google Scholar, S. K. new of the complex that in Google Scholar). might be of interest to the of CTNND2 levels early after stroke to these to the we also that CTNND2 levels in the blood a to the ischemic stroke from patients. CTNND2 expression is to the could that in CTNND2 blood content might be a of is from the ischemic cerebral different have in blood the CTNND2 to ischemia in mice and to the CTNND2 in be CTNND2 has as a in for Zhang J. R. G. S. of for 2009; PubMed Scopus Google Scholar). has a role for CTNND2 as a blood of disease. Thus, for stroke to be considered further to the of the only CTNND2 the of the which might also have a role in strokes as early as and patients to a cerebral We have of a and biostatistical to and data from transcriptomics and proteomics and as acute biomarkers for the and of ischemic stroke. We that integrative the to in and in the biostatistical used to -omics data which could have notable and data from analysis these the integration of -omics techniques might the of stroke. For could on the of these after stroke. of the in the hyper-acute phase after stroke might the of the complex stroke and new blood biomarkers for disease. further might also be the of elements that we have as therapeutic for stroke as as their role as stroke also that be the integrative analysis identified correlation of genes and from which notable were these could be as a biological among have used as a and to highlight among of the stroke but the biological of these still to be in the time mice and are still and over time might have to be considered in a Furthermore, of we were to in the ischemic at the time which have also an to and the integrative analysis and integrative approach to and further genes and proteins in the ischemic brains of mice over we of approach to the of GADD45G and CTNND2 as blood biomarkers of stroke and diagnosis, The transcriptomics data in has in the The proteomics data has to the the the data and the proteomics data has to the We to the patients in the and their 3 the 3 and 6 cerebral cerebral of Stroke analysis of 2 2

Topics & Concepts