Concurrent Glycogen and Lactate Imaging with ... - ACS Publications

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Concurrent Glycogen and Lactate Imaging with FTIR Spectroscopy to Spatially Localize Metabolic Parameters of the Glial Response following Brain Ischemia Mark John Hackett, Nicole J Sylvain, Huishu Hou, Sally Caine, Mariam Alaverdashvili, Michael Jake Pushie, and Michael E Kelly Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b02588 • Publication Date (Web): 03 Oct 2016 Downloaded from http://pubs.acs.org on October 11, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Concurrent Glycogen and Lactate Imaging with FTIR Spectroscopy to Spatially Localize Metabolic Parameters of the Glial Response following Brain Ischemia Mark J. Hackett*A, Nicole J. SylvainB, Huishu HouB, Sally CaineC, Mariam AlaverdashviliC, Michael J. PushieB, Michael E. Kelly** B *First author and co-corresponding author for spectroscopy related questions. Email: [email protected] Phone: +61-8-9266-3102 ** Corresponding author for stroke, neuroscience and animal model related questions. Email: [email protected] Phone: 1-306-844-1104 A

Nanochemistry Research Institute, Department of Chemistry, Curtin University, GPO Box U1987, Perth, Western Australia 6845 B Department of Surgery, Division of Neurosurgery, College of Medicine, University of Saskatchewan, 107 Wiggins Road, Suite B419 Health Sciences Building, Saskatoon, Saskatchewan S7N 5E5, Canada C

College of Pharmacy and Nutrition, College of Medicine, University of Saskatchewan, 107 Wiggins Road, Suite B419

Health Sciences Building, Saskatoon, Saskatchewan S7N 5E5, Canada

Key Words FTIR Imaging, Ischemia, Photothrombotic, Glycogen, Lactate, Astrocytes, Glia Abstract Imaging energy metabolites as markers of the energy shuttle between glia and neurons following ischemia is an ongoing challenge. Traditional microscopies in combination with histochemistry reveal glycogen accumulation within glia following ischemia, indicating an altered metabolic profile. Although semi-quantitative histochemical glycogen analysis is possible, the method suffers from typical confounding factors common to histochemistry, such as variation in reagent penetration and binding. In addition, histochemical detection of glycogen does not reveal information on the metabolic fate of glycogen (i.e., lactate production). Therefore, validation of a direct semi-quantitative method to simultaneously image both brain glycogen and lactate in the same tissue section would benefit this research field. In this study, we demonstrate the first application of FTIR spectroscopy for simultaneous direct spectroscopic imaging of brain glycogen and lactate, in situ within ex vivo tissue sections. Serial tissue sections were analysed with GFAP immuno-histochemistry to provide a comparison between the glycogen and lactate distribution revealed by FTIR and the glial distribution revealed by GFAP immunohistochemistry. The distribution of glycogen revealed by FTIR spectroscopic imaging has been further compared with histochemical detection of glycogen on the adjacent tissue sections. This approach was then applied to study spatio-temporal disturbances in metabolism, relative to glia and neuronal populations, following cerebral ischemia in a murine model of stroke.

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 19

Introduction Metabolic alterations are a hallmark of an ischemic insult to the brain, which occurs during stroke, and neurodegenerative disease.1-6 Both neurons and glial cells die due to insufficient oxygen and energy supply to the brain. Nevertheless, a substantial number of neurons and glial cells (specifically astrocytes) in the peri-infarct zone (PIZ) stay viable following stroke-induced metabolic perturbation.3 Neurons themselves can’t sustain metabolic hemostasis, since they can’t extract glucose from the blood stream or store “emergency” energy reserves, such a glycogen. On the contrary, glia can “self-regulate” and sustain a metabolic challenge, due to the presence of small (relative to other tissues) glycogen reserves.7-9 During an ischemic insult, glial glycogen is rapidly metabolized to glucose or lactate (i.e., astrocyte neuron lactate shuttle hypothesis),10 to maintain glia and neuron viability.9-13 Glucose transporters on glial cell processes and glia glycolytic activity is upregulated not only during a hypoxic or ischemic insult but stays upregulated for several days following the insult, which is postulated to ensure additional ‘metabolic security’ for glia and neurons.14-16 A concomitant increase in bulk brain glycogen is observed during this time frame, with numerous sub-cellular deposits of glycogen observed within astrocytes.17-19 A mild ischemic (or hypoxic) insult, sufficient to initiate a glial cell response, has been shown to be neuroprotective against a secondary larger ischemic episode (i.e., ischemic preconditioning).17, 20-22 Thus, it has been speculated that the increased glycogen content of glial cells, induced by an initial minor ischemic insult, may act as a therapeutic energy reserve during the second larger ischemic attack.17,

20, 21

Therefore, pathways involved in glial cell glycogen accumulation and

subsequent metabolism and supply to neurons and other brain cells may be of therapeutic interest. The fate of glycogen that accumulates in glia after an ischemic insult, and the distribution of glycogen metabolic products (e.g. lactate), remains largely unknown at the cellular or near cellular level (i.e., < 50 µm spatial resolution). The major challenge in detecting and visualizing metabolic products is simultaneous mapping of glycogen and small mobile metabolic products such as lactate. Although histochemical studies utilizing periodic acid-Schiff (PAS) staining and electron microscopy can identify the presence of glycogen deposits within glial cells,8, 18, 23-24 traditional histology and microscopies are not well suited to study small and mobile metabolic parameters, such as lactate. Although, direct in vivo imaging methods such as magnetic resonance imaging (MRI) and positron emission tomography (PET) have been instrumental to the field, spatial resolution is limited to approximately 1mm, preventing studies at the cellular level.5,

25

Therefore, an analytical technique capable of co-localizing the regional and

especially cellular distribution of glycogen and metabolic end products, such as lactate, would be of great benefit in research aimed at elucidating the biological role of glial cell glycogen pathways in stroke pathophysiology and post-stroke therapeutic interventions.


Page 3 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fourier transform infrared (FTIR) spectroscopic imaging permits stain free biochemical imaging of brain tissue. The technique has been previously used to study biochemical mechanisms of Alzheimer’s disease,26,

27

epilepsy,28 multiple sclerosis,29 cerebral malaria,30 ischemic stroke31-33 and hemorrhagic

stroke.34 FTIR imaging has also been used to detect lactate content in cerebellar purkinje neurons,35 in the murine cerebellum during cerebral malaria,30 in tumor tissue,36-38 and tumour spheroids,39 and glycogen content in cancerous tissue,40 and cells,41 via characteristic marker bands diagnostic of lactate at ~1127 cm-1 and glycogen at ~1152, ~1045 and ~1025 cm-1.30,

36-38, 40-42

Although other molecules, such as

carbohydrates and phosphates produce absorbance at similar regions to glycogen, none contain characteristic bands at all of these regions.

30, 36-38, 40,41

Despite this strong evidence from the published

literature to support the use of FTIR to image glycogen and lactate metabolic pathways in the brain, no previous study has evaluated simultaneous expression of glycogen and lactate within brain tissue and/or compared their spatial distribution with specific cell types, i.e., glia. Thus, in this study we have used FTIR imaging in combination with qualitative PAS histochemical glycogen detection, and immunohistochemical localization of glia, to semi-quantitatively study glycogen and lactate distribution during the glial response in a photothrombotic murine model of focal ischemic stroke. To the best of our knowledge, this is the first study to demonstrate glycogen accumulation in the photothrombotic focal ischemic stroke model. Further, the results reveal distinct regional variation in glycogen and lactate distribution, providing new insight into metabolic pathways occurring within the necrotic and anoxic ischemic infarct and the surrounding PIZ. As such, this study demonstrates a major advantage of FTIR imaging when used in combination with traditional microscopy and immunohistochemistry. Continued integration of FTIR spectroscopic imaging into the field of neuroscience, may pave the way to increase our mechanistic understanding of metabolic pathways in brain ischemia, allowing potential manipulation of these pathways as a novel therapeutic strategy.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 19

Materials and Methods

Animal model –Photothrombotic stroke targeting the primary somatosensory cortex (S1FL) in 11 week old Male Balb/c mice, was induced as previously described.31,43 Mice (n=5 per timepoint) were sacrificed 1 hour, 1 day, 2 days and 3 days after the induction of stroke (e.g. when the laser was turned off). All work was approved by the University of Saskatchewan’s Animal Research Ethics Board, and adhered to the Canadian Council on Animal Care guidelines for humane animal use.

Sample Preparation - To avoid introduction of chemical artifacts that can result during sample preparation of biological samples, all mice were sacrificed as described previously.32, 44, 45 Briefly, the mice were anaesthetized with isofluorane, decapitated, and the head was immediately submerged into liquid nitrogen.32 The frozen brain was chiseled out from the head on dry ice, and stored at -80 ˚C until required for analysis. For each brain, several adjacent 10-µm-thick coronal section of the brain were cut with a cryo-microtome at -18 ˚C from the central location of the lesion (+ 0.25 mm anterior to Bregma). Serial sections were either melted onto a CaF2 substrate for FTIR imaging analysis, or onto glass microscope slides for histochemical PAS staining and GFAP immunohistochemistry. Following tissue sectioning, slides were stored at –80 ˚C until analyzed, and air-dried at ambient temperature immediately before analysis.

Globar FTIR imaging - Globar-FTIR spectroscopic imaging was performed at the Canadian Light Source (CLS) with a Hyperion 3000 microscope fitted with an upper objective of 15× magnification and a numerical aperture of 0.4, and a lower condenser of 15× magnification and 0.4 numerical aperture. This arrangement yielded a pixel size of 2.65 × 2.65 µm, which was later subjected to 8×8 pixel binning to yield an effective image pixel size of 21.2 × 21.2 µm. Globar-FTIR images were acquired with a spectral resolution of 4 cm-1 with the co-addition of 16 scans. A background image was collected from blank substrate using 16 co-added scans immediately prior to each sample. The entire imaged areas consisted of approximately 20 x 15 tiles per sample and took approximately 2 hours per image to acquire.

Data processing and data analysis of FTIR spectra and images - All data processing and image generation was performed using Cytospec software (Cytospec, Version 1.2.04) and Opus software (Version 6.5, Bruker, Ettlingen, Germany). To minimize the influence of variation in tissue thickness, data was vector normalized to the amide I band (1600 – 1700 cm-1), i.e., normalized to “total protein”, similar to a biochemical assay approach. Second-derivative spectra were generated from normalized spectra using a 13 smoothing point Savitzky-Golay function. Images of relative glycogen and lactate 4 ACS Paragon Plus Environment

Page 5 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


distribution were determined from second-derivative intensity at 1152 and 1125 cm-1 respectively. The use of second-derivative spectra to study relative concentration of chemical species has been described in several studies,35,46,47, and readers are referred to Susi et. al., (1986) for a detailed description.47 The average glycogen and lactate content was determined from an average FTIR spectrum determined for the entire ischemic infarct (excluding any tears or cracks), peri-infarct zone (PIZ), and contra-lateral tissue. The average spectrum for these regions was determined via comparison of FTIR spectroscopic images of lipid ester distribution with H&E histology, as we have previously reported,31 and representative regions of interest are shown in Figure S1.

GFAP Immunohistochemistry – For anti-Glial Fibrillary Acidic Protein (GFAP, Dako, Z0334, Glostrup, Denmark) immunohistochemistry, frozen sections were stained as previously described.31 Briefly, adjacent slides were immuno-labelled using a Dako EnVision Plus reagents kit (Dako, Denmark), with the aid of a hydration chamber to prevent dehydration. Air dried sections were fixed with 10% methanol free formalin (prepared freshly from paraformaldehyde), rinsed with PBS with 0.1% Tween 20 (PBST), then blocked with Dako Peroxidase Block for 10 mins, and incubated for 30 mins in GFAP diluted at 1:1500 in Dako Antibody diluent. The slides were then incubated for 30 mins in Dako Labelled polymer (HRP anti-rabbit), followed by an incubation in freshly mixed Dako DAB Chromagen for 10mins, and incubated for 5 mins in fresh 2% copper sulfate. Finally, the slides were counterstained with Modified Harris Hematoxylin, dehydrated in ethanol, cleared in xylene and coverslipped. Semi-quantitative analysis was not performed, only qualitative visual comparison between FTIR images and GFAP stained tissue sections was performed.

PAS Histochemical Staining – Reagents and procedures for Periodic Acid – Shiff (PAS) staining were modified from Humason (1979).48 Briefly, slides were cryosectioned and allowed to air dry, fixed in Rossman’s fixative (9 parts absolute alcohol saturated with picric acid and 1 part concentrated formalin) at 4°C for 15 mins, and rinsed with 95% EtOH several times. Slides were rinsed quickly with water, then incubated in 1% periodic acid for 5 mins, followed by rinsing for 5 mins under running tap water, then incubated in Shiffs’s reagent for 15 mins (Shiff’s: 0.5g Basic Fuchsin, 100mL distilled water, 15mL 1N Hydrochloric acid, 1.5g Potassium metabisulphite and 0.25-0.5g activated charcoal). After being rinsed under running water for 10 mins, the slides were counterstained by incubating for 5 mins in modified Harris Hematoxylin, followed a dip in acid alcohol and 15 seconds in lithium carbonate and 3 mins of rinsing under running tap water. To finish, slides were dehydrated in 70%, 95% and 100% EtOH, then cleared in Xylene and coverslipped. To confirm the glycogen staining, glycogen in serial sections was digested with 1% Amylase in PBS (phosphate buffered saline) for 30 minutes at 37°C prior to incubation 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 19

in periodic acid. Semi-quantitative analysis was not performed, only qualitative visual comparison between FTIR images and PAS stained tissue sections was performed.

Statistical Analysis –A two tailed Student’s t-test was used to test for a statistically significant difference between lactate and glycogen content within the PIZ and infarct relative to the contra-lateral hemisphere, across the 4 different time points. Due to the use of multiple t-tests, the p-value from individual t-tests was multiplied by the number of comparisons performed for each group (i.e., multiplied by 4), to obtain a corrected p-value. A 95% confidence limited was used as the significance cut off (i.e., p-value corrected for multiple comparison < 0.05, or, uncorrected p-value < 0.0125).


Page 7 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Results and Discussion Although brain tissue micro-dissection and bulk biochemical assay has revealed substantial insight into the bulk metabolic alterations that occur due to cerebral ischemia insult,17, 49-51 this approach does not have the spatial resolution to probe inter-cellular metabolic relationships, such as those between glial cells and neurons, which remain to be fully elucidated.51 A lack of suitable imaging techniques to complement bulk assays can be suggested as one of the major reasons of insufficient knowledge of metabolic perturbations after stroke. For example, to the best of our knowledge, semi-quantitative biochemical imaging of metabolic parameters, such as glycogen and lactate, at the cellular or near cellular level (i.e., < 50 µm), following cerebral ischemia has not previously been performed. Staining of brain glycogen is particularly problematic as it is rapidly metabolized during ischemic conditions associated with disease, and also during ischemic conditions that are created within the brain following animal death.23,

24

Consequently, typical fixation methods, such as perfusion fixation, which create cerebral

ischemia during the fixation process partially or completely deplete brain glycogen.23, 24 Further, glycogen may be dissolved and leached from tissue sections by polar staining media (aqueous solutions and alcohol solvents) during post-fixation, or staining procedures.23,

24

In addition to sample preparation induced

alterations to brain glycogen, glycogen metabolites, such as lactate are also affected. While the concentration of brain glycogen rapidly decreases following animal death, the lactate concentration rapidly increases.52-54 Lactate, which is a small mobile polar molecular, is readily redistributed or removed from tissue sections during any form of fixation or staining.44 Therefore, rapid flash freezing cryo-preservation methods are required in order to preserve glycogen and lactate levels as close as possible to the in vivo condition.52, 53 Such methods are routinely used for bulk biochemical assays,5,2 53 and have recently been adopted for FTIR imaging studies.32 Therefore, the combination of FTIR imaging with suitable sample preparation protocols, as used in this study, provides the novel capability to simultaneously image glycogen and lactate at the near cellular level, and compare the distribution of these important metabolic markers with routine histology and immuno-histochemistry of glia and neuronal populations. Using PAS histochemical staining or electron microscopy for identifying glycogen deposits, it is well established that glial cells accumulate glycogen during the “glial response” following an ischemic insult to the brain.17-19 Although glycogen accumulation has not previously been investigated in the photothrombotic stroke model, it is well established that a pronounced glial response occurs over a 1 – 7 day period post ischemic insult in the photothrombotic stroke model (mouse and rat).55 In this study, to validate the FTIR spectroscopic imaging approach against standard histological methods, a qualitative comparison of FTIR spectroscopic images, PAS histochemistry and GFAP immuno-histochemistry was performed using tissue from mice 3 days post ischemic insult. GFAP immunohistochemistry was used to 7 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 19

confirm the presence of the glial response following photothrombotic insult in mice, similar to our previous report using this model.31 Mild GFAP antigenicity was observed in “healthy” brain regions (i.e., the contralateral hemisphere) with strong GFAP antigenicity observed in the PIZ that surrounds the ischemic infarct, and sparse antigenicity was observed within the ischemic infarct (Figure 1A, B). This distribution of GFAP antigenicity is consistent with the published literature for this model.31,

55

The

morphology of cells expressing GFAP antigenicity was consistent with astrocyte morphology. PAS histochemical staining revealed an intense staining pattern within the PIZ, similar to the intense GFAP antigenicity observed within the PIZ (Figure 1D, E). Treatment of tissue sections with amylase prior to PAS staining removes glycogen but not other sources of carbohydrates (i.e., proteoglycans). Thus the loss of intense PAS staining within PIZ in amylase treated tissue sections in this study, confirms glycogen as the source of the original intense PAS staining (Figure 1G, H). The ischemic infarct core was visualized in FTIR spectra based on the abundance of aggregated proteins, shown through second-derivative intensity at 1625 cm-1, as previously reported.31 FTIR spectroscopy has previously identified a characteristic marker band for glycogen/glucose hydroxyl groups, ν(C-O) at ~1025 cm-1, and ~1080 cm-1 with additional bands associated with glycolytic linkages at 1045 and 1152 cm-1.36-38, 40 To avoid spectra overlap with other carbohydrate and phosphate absorbance at ~1080 cm-1, glucose absorbance at 1025 cm-1, FTIR false colour functional group images of the relative glycogen concentration were generated from the second-derivative intensity at 1152 cm-1 in this study (Figure 1F). These images revealed that the relative concentration of glycogen is highest within the PIZ, co-localized with intense GFAP antigenicity and PAS staining. Both FTIR spectroscopic images and PAS stained tissue sections revealed a lack of glycogen within the ischemic infarct (Figure 1D, E, F). In contrast to glycogen, the ischemic infarct contained substantial lactate content, as observed from secondderivative intensity of the lactate marker bank at 1127 cm-1. Representative raw and second-derivative spectra highlighting the spectral alterations associated within increased glycogen in the PIZ and increased lactate within the ischemic infarct are shown in Figure 2. The abundance of lactate within the hypoxic and necrotic infarct bears a similar resemblance to the lactate distribution previously reported from FTIR spectroscopic analysis of tumour spheroid tissue sections.39 As the ischemic infarct is permanently occluded in the photothrombotic stroke model, there is no energy or oxygen supply from the blood stream to this tissue. Thus, any cells that survived the immediate ischemic insult, or any infiltrating cells (i.e., glia and macrophages) are likely only capable of anaerobic metabolism due to the hypoxic conditions within the infarct, which may explain the abundance of lactate. It is possible that the glial cells within the PIZ, accumulate glycogen as an energy reserve, which is then metabolized by glia, which can increase their rate of glycolysis,56 to prolong the life of adjacent neurons. It is unlikely that this would result in substantial benefit to cells within the core of the infarct; 8 ACS Paragon Plus Environment

Page 9 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


however, it could potentially prolong the life of cells at the ischemic infarct boundary. However, the glial cell glycogen accumulation may also reflect an abnormal response, induced within glial cells by damaging cytopatic events within the ischemic infarct. It is possible that the glycogen accumulation offers no benefit, and could potentially be harmful if the accumulated glycogen starves adjacent cells of their energy supply. To investigate this further, FTIR spectroscopic imaging was performed on tissue sections obtained from animals across a time course of 1hour, 1, 2 and 3 days post ischemic insult (Figure 3 and Figure 4). The results revealed a significant time dependent increase in glycogen within the PIZ 1, 2 and 3 days post ischemic insult, but not at 1 hour (Figure 4A). There was no difference in lactate concentration within the PIZ across the time course. However, lactate was increased within the ischemic infarct at all time points studied (Figure 4). Glycogen levels did not change significantly within the ischemic infarct across the time course studied. Although lactate was increased on average within the infarct, the images of lactate distribution within the infarct (Figure 3) show numerous pixels with high lactate signal (dark red) and also low lactate signal (dark blue). We speculate that the large number of cells that die following ischemic insult in this model results in release of intracellular lactate, contributing to the pixels with low lactate signal. Future higher resolution imaging of glycogen and lactate with a synchrotron source will be required to answer these questions. Such post-stroke time course metabolic mechanistic information, as described above, could not previously be obtained without the application of FTIR spectroscopic imaging in combination with histology and immuno-histochemistry, as performed in this study. At this stage it is not known if glycogen accumulation offers any benefit to neuron survival following ischemic insult, or if in fact accumulation may “starve” others cells of an energy supply, and therefore, contribute to cell death. In addition, the released lactate may serve as a beneficial energy source for cells, however it can also lower pH causing acidosis, propagating tissue damage within the infarct.51 Although much remains unknown, these mechanistic avenues can now be investigated in future studies using the novel ability of FTIR imaging to correlate the relative glycogen and lactate content at the cellular level. We anticipate that future work, particularly with increasing developments in the application of quantum cascade lasers,57 may allow resolution of lactate and glycogen within individual cells, on laboratory based benchtop instruments, which has the potential for many exciting applications to the field of neuroscience. Lastly, these results highlight that the glial response induced by photothrombotic stroke results in the typical glycogen accumulation observed in other models of brain ischemia. The photothrombotic stroke model has several advantages over other stroke models, including a relatively simpler and less invasive animal surgery, and the ability to form a highly reproducible ischemic infarct.43, 56, 58-59 However, several differences exist between the ischemic conditions created in this model and the human condition 9 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 19

of stroke, raising questions about the relevance of this model to study biochemical mechanism of human stroke. For example, the photothrombotic stoke model produces an ischemic infarct which contains almost no reperfusion and therefore, only a small penumbra.60, 61 In contrast, much larger regions of tissue reperfusion (or partial reperfusion) and a substantial ischemic penumbra are observed in clinical human stroke.62 In addition, edema occurs to a substantially greater extent in photothrombotic animal models than in human stroke.63, 64 As it is known that edema can have substantial effects on astrocytes, (i.e, endfeet and mitochondrial swelling), metabolic processes, such as glucose uptake from the blood stream and glycogen accumulation may occur differently in this model.65 To the best of our knowledge, this is the first study to demonstrate glycogen accumulation within the PIZ, concomitant with the glial response in the photothrombotic model. This is an important result as it demonstrates the photothrombotic model is a suitable model for future work aimed at elucidation of the biochemical mechanisms of glial glycogen accumulation, and the downstream physiological effects of glycogen accumulation, which reflects the human clinical condition.

Conclusions In this study we have demonstrated for the first time that FTIR spectroscopic imaging is a valuable approach capable of imaging at the cellular level, the relative glycogen and lactate content of tissue sections following ischemic insult. The results have revealed the spatio-temporal pattern of important metabolic disturbances after photothrombotic stroke in the mouse, which have not been studied before. Such a method provides immense potential to be adopted in future studies, specifically in studies designed to elucidate the metabolic fate and physiological function of glycogen after ischemic preconditioning. Such knowledge may allow future therapeutic manipulation of glial metabolic pathways, which could provide substantial health benefits to stroke victims, or victims of neurodegenerative diseases or disorders with a pathogenic ischemic component.

Acknowledgements

This work was supported by: a Saskatchewan Health Research Foundation (SHRF) Establishment Grant awarded to M.E.K.; a Heart and Stroke Foundation (HSF), SHRF, and University of Saskatchewan (U of S) Saskatchewan Research Chair for Clinical Stroke Research awarded to M.E.K.; and a joint Canadian Institutes of Health Research (CIHR)/Heart and Stroke Foundation of Canada (HSFC) Synchrotron Medical Imaging Team Grant #CIF 99472 awarded to Helen Nichol, M.E.K and others. M.J.H. received scholarship support as a CIHR Postdoctoral Fellow, and a Saskatchewan Health Research Foundation postdoctoral fellow. M.J.H and SC received support as SMI postdoctoral fellows and CIHR-Training 10 ACS Paragon Plus Environment

Page 11 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


grant in Health Research Using Synchrotron Techniques (CIHR-THRUST) Fellows. The authors would like to acknowledge Phyllis Paterson for the use of surgical equipment, Sharleen Weese-Maley for her administrative assistance, and Karen Yuen for her help with PAS staining. Research described in this paper was performed in part at the Mid-IR beamline at the Canadian Light Source (off-line nonsynchrotron instrumentation), which is supported by the Natural Sciences and Engineering Research Council of Canada, the National Research Council Canada, the Canadian Institutes of Health Research, the Province of Saskatchewan, Western Economic Diversification Canada, and the University of Saskatchewan.

Author Contributions *denotes first author of this manuscript. **denotes the co-author who acted as the primary supervisor for this research.

*MJH contributed 30 % of experimental research, writing and editing the manuscript. NJS contributed 15 % of the experimental research and contributed to editing the manuscript. HH contributed 15 % of the experimental research and contributed to editing the manuscript. MA contributed 10 % of the experimental research and contributed to editing the manuscript. SC contributed 10 % of the experimental research and contributed to editing the manuscript. MJP contributed 10 % of the experimental research and contributed to editing the manuscript. **MEK contributed 10 % of the experimental research design, manuscript editing and is the corresponding author.

Associated Content This material is available free of charge via the Internet at http://pubs.acs.org.

8.0 Conflicts of Interest Authors declare no competing financial conflicts of interest.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 19

Figure 1: Comparison of immuno-histochemical GFAP detection (A, D), PAS glycogen staining (B, C, E, F), and FTIR glycogen and lactate imaging (G-I). Overview images of ischemic infarct 3 days after photothrombotic stroke (A-C). High magnification view of ischemic infarct boundary, including periinfarct zone, PIZ (D-F). FTIR images of aggregated protein in the ischemic infarct and FTIR images of glycogen and lactate distribution, derived from second-derivative intensity (G-I). Arrows indicate PIZ, which displays increased GFAP antigenicity, increased glycogen observed in both PAS staining and FTIR imaging methods. Pre-treatment of tissue sections with amylase (B, E), confirms PAS staining is due to glycogen and not non-proteoglycan carbohydrates. Numbers 1-3in G indicates the anatomical location of contralateral hemisphere (1.), PIZ (2.), ischemic infarct (3.). Scale bar in A, G = 500 µm, in D = 100 µm.


Page 13 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2: (A) Representative raw FTIR spectra from contralateral hemisphere, peri-infarct zone (PIZ) and infarct, 3 days post-ischemic insult. Shading in A, highlights spectroscopic location of glycogen and lactate bands. (B) Comparison of second-derivative FTIR spectra from contra-lateral hemisphere and PIZ, showing location of glycogen bands (1-5), indicating increased glycogen levels in PIZ. Asterisk indicates lactate band. (C) Comparison of second-derivative spectra from the contralateral hemisphere and ischemic infarct, showing location of lactate band. Second-derivative spectra in B and C (top spectra), are offset above corresponding region of raw spectra (bottom spectra).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 19

Figure 3: Spatio-temporal comparison of glycogen and lactate following photothrombotic stroke in the mouse. Arrows indicate location of PIZ, with increased GFAP antigenicity and glycogen accumulation. Numbers 1, 2 and 3 indicate anatomical location of contralateral hemisphere (1.), PIZ (2.), and ischemic infarct (3.). FTIR second-derivative images of aggregated protein are shown to highlight location of ischemic infarct. The images show a time-dependent increase in glycogen accumulation within the PIZ following ischemic stroke. Lactate is increased within the ischemic infarct at all time points following stroke. No time dependent changes in the contra-lateral hemisphere were observed, a representative day 3 contralateral hemisphere is shown in this Figure. Scale bar = 500 µm.


Page 15 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4: Analysis of relative glycogen and lactate levels in contralateral hemisphere, ischemic infarct, and PIZ, across a 3 day time course post photothrombotic stroke. Relative glycogen and lactate levels were determined from second-derivative intensity at 1152 cm-1 and 1127 cm-1, respectively. Circles indicate mean second-derivative intensity, calculated from 5 animal replicates. Error bars indicate standard deviation. Red circles indicate a statistically significant difference relative to contralateral hemisphere, calculated with a Student’s t-test, and 95% confidence interval (p < 0.05). As 4 multiple comparisons were performed, a Bonferroni correction was applied.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 19

Nomenclature and Abbreviations FTIR Fourier Transform Infrared IAUC Integrated Area Under the Curve PIZ Peri-Infarct Zone PAS Periodic Acid-Schiff References

(1) Kalaria, R. N. Neurobiol. Aging 2000, 21, 321-330. (2) De Keyser, J.; Steen, C.; Mostert, J. P.; Koch, M. W. J. Cer. Blood Flow Metab. 2008, 28, 16451651. (3) Lipton, P. Physiol. Rev. 1999, 79, 1431-1568. (4) Bouma, G. J;, Muizelaar, J. P.; Choi, S. C.; Newlon, P. G.; Young, H. F. J. Neurosurg. 1991, 75, 685693. (5) Magistretti, P. J.; Pellerin, L.; Rothman, D. L.; Shulman, R. G. Science 1999, 283, 496-497. (6) Pellerin, L.; Magistretti, P. J. Proc. Natl. Acad. Sci. USA 1994, 91, 10625-10629. (7) Gruetter, R. J. Neurosci. Res. 2003, 74, 179-183. (8) Cataldo, A. M.; Broadwell, R. D. J. Electron Microsc. Tech. 1986, 3, 413-437. (9) Dringen, R.; Gebhardt, R.; Hamprecht, B. Brain Res. 1993, 623, 208-214. (10) Pellerin, L.; Magistretti, P. J. Cer. Blood Flow Metab. 2003, 23, 1282-1286. (11) Pellerin, L.; Bouzier-Sore, A.-K.; Aubert, A.; Serres, S.; Merle, M.; Costalat, R.; Magistretti, P. J. Glia 2007, 55, 1251-1262. (12) Pellerin, L.; Pellegri, G.; Bittar, P. G.; Charnay, Y.; Bouras, C.; Martin, J. L.; Stella, N. Magistretti, P. J. Dev. Neurosci. 1998, 20, 291-299. (13) Pellerin, L.; Magistretti, P. J. Neuroscientist 2004, 10, 53-62. (14) Iwabuchi, S.; Kawahara, K. Neurochem. Int. 2011, 59, 319-325. (15) Marrif, H.; Juurlink, B. H. J. Neurosci. Res. 1999, 57, 255-260. (16) Schurr, A.; Payne, R. S.; Miller, J. J.; Tseng, M. T.; Rigor, B. M. Brain Res. 2001, 895, 268-272. (17) Brucklacher, R. M.; Vannucci, R. C.; Vannucci, S. J. Dev. Neurosci. 2002, 24, 411-417. (18) Kajihara, H.; Tsutsumi, E.; Kinoshita, A.; Nakano, J.; Takagi, K.; Takeo, S. Brain Res. 2001, 909, 92-101. (19) Rossi, D. J.; Brady, J. D.; Mohr, C. Nat. Neurosci. 2007, 10, 1377-1386. (20) Liu, X.-q.; Sheng, R.; Qin, Z.-h. Acta Pharmacol. Sin. 2009, 30, 1071-1080. (21) Trendelenburg, G.; Dirnagl, U. Glia 2005, 50, 307-320. 16 ACS Paragon Plus Environment

Page 17 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(22) Gorgias, N.; Maidatsi, P.; Tsolaki, M.; Alvanou, A.; Kiriazis, G.; Kaidoglou, K.; Giala, M. Brain Res. 1996, 714, 215-225. (23) Kong, J.; Shepel, P. N.; Holden, C. P.; Mackiewicz, M.; Pack, A. I.; Geiger, J. D. J. Neuorsci. 2002, 22, 5581-5587. (24) Guth, L.; Watson, P. K. Exp. Neurol. 1968, 22, 590-602. (25) Öz, G.; Henry, P.-G.; Seaquist, E. R.; Gruetter, R. Neurochem. Int. 2003, 43, 323-329. (26) Liao, C. R.; Rak, M.; Lund, J.; Unger, M.; Platt, E.; Albensi, B. C.; Hirschmugl, C. J.; Gough, K. M. Analyst 2013, 138, 3991-3997. (27) Leskovjan, A. C.; Kretlow, A.; Miller, L. M. Anal. Chem. 2010, 82, 2711-2716. (28) Chwiej, J.; Dulinska, J.; Janeczko, K.; Dumas, P.; Eichert, D.; Dudala, J.; Setkowicz, Z. J. Chem. Neuroanat. 2010, 40, 140-147. (29) Heraud, P.; Caine, S.; Campanale, N.; Karnezis, T.; McNaughton, D.; Wood, B. R.; Tobin, M. J.; Bernard, C. C. A. NeuroImage 2009, 49, 1180-1189. (30) Hackett, M. J.; Aitken, J. B.; El-Assaad, F.; McQuillan, J. A.; Carter, E. A.; Ball, H. J.; Tobin, M. J.; Paterson, D.; de Jonge, M. D.; Siegele, R.; Cohen, D. D.; Vogt, S.; Grau, G. E.; Hunt, N. H.; Lay, P. A. Sci. Adv. 2015, 1, e1500911. (31) Caine, S.; Hackett, M. J.; Hou, H.; Kumar, S.; Maley, J.; Ivanishvili, Z.; Suen, B.; Szmigielski, A.; Jiang, Z.; Sylvain, N. J.; Nichol, H.; Kelly, M. E. Neurobiol. Dis. 2016, 91, 132-142. (32) Hackett, M. J.; Britz, C. J.; Nichol, H.; Paterson, P. G.; Pickering, I. J.; George, G. N. ACS Chem. Neurosci. 2015, 6, 226-238. (33) Hackett, M. J.; Smith, S. E.; Caine, S.; Nichol, H.; George, G. N.; Pickering, I. J.; Paterson, P. G. Free Radic. Biol. Med. 2015, 89, 806-818. (34) Hackett, M. J.; DeSouza, M.; Caine, S.; Bewer, B.; Nichol, H.; Paterson, P. G.; Colbourne, F. ACS Chem. Neurosci. 2015, 6, 761-770. (35) Hackett, M. J.; Borondics, F.; Brown, D.; Hirschmugl, C.; Smith, S. E.; Paterson, P. G.; Nichol, H.; Pickering, I. J.; George, G. N. ACS Chemi. Neurosci. 2013, 4, 1071-1080. (36) Petibois, C.; Deleris, G. Trends Biotechnol. 2006, 24, 455-462. (37) Petibois, C.; Drogat, B.; Bikfalvi, A.; Deleris, G.; Moenner, M. FEBS Lett. 2007, 581, 5469-5474. (38) Petibois, C.; Gionnet, K.; Goncalves, M.; Perromat, A.; Moenner, M.; Deleris, G. Analyst 2006, 131, 640-647. (39) Zhang, J. Z.; Bryce, N. S.; Siegele, R.; Carter, E. A.; Paterson, D.; de Jonge, M. D.; Howard, D. L.; Ryan, C. G.; Hambley, T. W. Integr. Biol. 2012, 4, 1072-1080. (40) Yano, K.; Ohoshima, S.; Shimizu, Y.; Moriguchi, T.; Katayama, H. Cancer Lett. 1996, 110, 29-34. (41) Kuimova, M. K., Chan, K. L. A., Kazarian, S.G. Appl. Spectrosc. 2009, 63, 164-171.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 19

(42) Chiriboga, L.; Xie, P; Yee, H.; Vigorita, V.; Zarou, D.; Zakim, D.; Diem, M. Biospectroscopy. 1998, 4, 47–53. (43) Winship, I. R.; Murphy, T. H. J. Neurosci. 2008, 28, 6592-6606. (44) Hackett, M. J.; McQuillan, J. A.; El-Assaad, F.; Aitken, J. B.; Levina, A.; Cohen, D. D.; Siegele, R.; Carter, E. A.; Grau, G. E.; Hunt, N. H.; Lay, P. A. Analyst 2011, 136, 2941-2952. (45) Hackett, M. J.; Smith, S. E.; Paterson, P. G.; Nichol, H.; Pickering, I. J.; George, G. N. ACS Chem. Neurosci. 2012, 3, 178-185. (46) Kneipp, J.; Miller, L. M.; Joncic, M.; Kittel, M.; Lasch, P.; Beekes, M.; Naumann, D. BBA Mol. Basis. Dis. 2003, 1639, 152-158. (47) Susi, H., Byler, D.M. Method. Enzymol. 1986, 130, 290-311. (48) Humason, G.L 1979, 4th Ed. WH Freeman and Co (Ed), Virginia, USA. (49) Nowak, T. S.; Fried, R. L.; Lust, W. D.; Passonneau, J. V. J. Neurochem. 1985, 44, 487-494. (50) Wagner, S. R.; Lanier, W. L. Anesthesiology 1994, 81, 1516-1526. (51) Rossi, D. J.; Brady, J. D.; Mohr, C. Nat. Neurosci. 2007, 10, 1377-1386. (52) Lust, W. D.; Passonneau, J. V.; Veech, R. L. Science 1973, 181, 280-282. (53) Veech, R. L.; Harris, R. L.; Veloso, D.; Veech, E. H., J. Neurochem. 1973, 20, 183-188. (54) Shank, R. P.; Aprison, M. H. J. Neurobiol. 1971, 2, 145-151. (55) Nowicka, D.; Rogozinska, K.; Aleksy, M.; Witte, O. W.; Skangiel-Kramska, J. Acta Neurobiol. Exp. 2008, 68, 155. (56) Herrero-Mendez, A; Almeida, A; Fernandez, E; Maestre, C; Moncada, S; Bolanos, J.P. Nat. Cell. Biol. 2009, 11, 747-752. (57) Yeh, K., Kenkel, S., Liu, J.N., Bhargava, R. Anal. Che. 2014, 87, 485-493. (58) Carmichael, S. T. NeuroRx. 2005, 2, 396-409. (59) Durukan, A.; Tatlisumak, T. Pharmacol. Biochem. Behav. 2007, 87, 179-197. (60) Katsman, D.; Zheng, J.; Spinelli, K.; Carmichael, S. T., J. Cer. Blood Flow Metab. 2003, 23, 9971009. (61) Kim, G. W.; Sugawara, T.; Chan, P. H. J. Cer. Blood Flow Metab. 2000, 20, 1690-1701. (62) Provenzale, J. M.; Jahan, R.; Naidich, T. P.; Fox, A. J. Radiology 2003, 229, 347-359. (63) Van Bruggen, N.; Cullen, B.; King, M.; Doran, M.; Williams, S.; Gadian, D.; Cremer, J. Stroke 1992, 23, 576-582. (64) Lee, V. M.; Burdett, N. G.; Carpenter, T. A.; Hall, L. D.; Pambakian, P. S.; Patel, S.; Wood, N. I.; James, M. F. Stroke 1996, 27, 2110-2119. (65) Manley, G. T.; Fujimura, M.; Ma, T.; Noshita, N.; Filiz, F.; Bollen, A. W.; Chan, P.; Verkman, A. Nat. Med. 2000, 6, 159-163. 18 ACS Paragon Plus Environment

Page 19 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC Graphic


Concurrent Glycogen and Lactate Imaging with ... - ACS Publications

Recommend Documents