In vivo evaluation of BBB integrity in the post-stroke brain

Abstract

The blood-brain barrier (BBB) is a dynamic interface responsible for maintaining central nervous system (CNS) homeostasis. An intact BBB protects the brain from undesired compounds and proteins from the blood, however, BBB impairment is involved in various pathological conditions including stroke. In vivo evaluation of BBB integrity in the post-stroke brain is important for investigating stroke-induced CNS pathogenesis and developing CNS-targeted therapeutic agents. In this chapter, we describe both quantitative and morphometric methods and tools to evaluate BBB integrity in vivo. These methods do not require expensive magnetic resonance imaging (MRI) and computed tomography (CT) imaging capabilities and can be conducted in research laboratories with access to a confocal microscope and fluorescence microplate reader.

1. Introduction

The blood-brain barrier (BBB) serves as a semipermeable interface between the vasculature system and the brain parenchyma, acting as a physical and metabolic [1] barrier that mediates the transport of selective substances in a bi-directional way [2] and regulates diverse physiological processes in the central neural system (CNS) [3]. The BBB is anatomically at the level of the brain endothelial cells (BMECs) and is supported by accessory cells, such as pericytes, astrocytes, and a basement membrane composed of extracellular matrix components (EMCs) [3–6]. These collective cells interact in a paracrine fashion and are often referred as the neurovascular unit. BMECs constitute the innermost luminal component of the BBB [4] surrounded by pericytes, astrocyte end-feet and basement membrane. BMECs show unique properties distinguishable from other peripheral endothelial cells [4], characterized by well-formed intercellular tight junctions (TJs) and a higher number of mitochondria.

TJs reduce the permeation of plasma solutes into brain parenchyma and limit the passage of proteins and lipids located at the apical-lateral membrane [7–9]. TJs consist of three types of integral membrane proteins (claudins, occludin, and junction adhesion molecules) and several cytoplasmic accessory proteins (zonula occludens [ZO]-1, −2, −3, cingulin, etc.). In the brain, claudins-1, −5, −11 are the major components to form the primary seal of the TJs [10–13]. Occludin is highly expressed in brain endothelial cells compared to nonneural tissues [14]. Both claudins and occludin are essential extracellular component to form the TJs and have a crucial role in regulating paracellular permeability of BBB [15]. Cytoplasmic proteins such as ZO-1, ZO-2, and ZO-3 link integral membrane proteins to the primary cytoskeleton protein actin which functions to stabilize the transmembrane elements [10, 16]. Disappearance and disorganization of those TJ proteins are evidences of BBB disruption.

The abundant mitochondria allow for generation of greater amounts of biological energy required to maintain BBB integrity [17] and augment the selective molecular permeability [4]. The selective exchanges of solutes and proteins between blood and CNS are through two manners: paracellular and transcellular [18, 19]. The paracellular transport happens by passive (energy-independent) diffusion via the TJs, whereas the transcellular transport occurs through the BMECs [20]. The transcellular pathway includes passive transcellular diffusion, passive or active (energy-dependent) carrier-mediated transport (CMT) [19], active receptor-mediated transcytosis (RMT), active efflux transport, and active adsorptive-mediated transcytosis [20, 21]. Among them CMT allows the molecule exchange via substrate-specific transporters [6, 19] such as glucose transporter 1 (GLUT-1) [22]. GLUT-1 is the principal glucose transporter of the BBB, and its deficiency leads to a stroke mimic condition [23]. GLUT-1 has been widely used to label vasculature in the brain [24]. Studies of our and other laboratories have shown downregulation of GLUT-1 in BBB disruption induced by ischemic stroke and smoking insults [25, 26]. The alteration of active transporters that play a critical role in the selective transportation in BBB could be used as a marker of BBB functional impairment.

BBB dysfunction is reported in many CNS pathological conditions including stroke [27]. Stroke is the second leading cause of mortality and morbidity worldwide [28, 29]. One of the major events taking place following stroke onset is the impairment and disruption of the BBB [2]. Rather than being solely a consequence of stroke, BBB damage also contributes to the pathogenesis of stroke [30] and is usually associated with poor clinical prognosis [2, 29, 31]. The raised interest in the critical role of BBB in the pathogenesis of stroke as well as other neurological disorders requires accurate and representative approaches to evaluate the BBB integrity in vivo [32].

To date, there is no single available approach that fulfills all the requirements for an accurate evaluation of BBB integrity in vivo. The most common tool to measure BBB permeability in vivo is the imaging technique, including magnetic resonance imaging (MRI) and computed tomography (CT), due to its non-invasive capability to quantitatively assess the brain [33]. Another common approach of evaluating BBB permeability in animal models is the application of various tracers (Table 1). In this approach, exogenous tracers are injected to the experimental animals and the fraction of tracers that extravasates is quantified as a biomarker of BBB disruption and opening of the paracellular barrier. Tracers with different molecular sizes can be used to evaluate the magnitude of the BBB opening. In general, tracers with low molecular weight are indicators of solute and ion permeability and tracers with high molecular weight are indicators for protein permeability [34–36]. Following injuries, disruption of BBB can change over time, and it is not an “all-or-nothing” phenomenon. Therefore, the molecular size of tracer should be considered based on the purpose of the study. Many tracers can be measured in the brain tissue with microscopical optical density (OD) measurement to semi-quantitatively evaluate the BBB opening [35–37]. Moreover, free tracers can be extracted from brain parenchyma and their concentration can be measured photometrically or fluorometrically, which is a more accurate manner compared to the semi-quantitative measurement [38–42]. Fluorescein isothiocyanate (FITC)-dextran tracers are available with a wide range of molecular sizes (3 – 2000 kDa), making them suitable for functionally evaluating the magnitude of BBB opening in vivo. Besides, morphometric biomarkers (Table 2), such as GLUT-1, claudin-5, occludin and ZO-1, could also enable the assessment of BBB’s functional and structural changes induced by various insults including stroke [43–46]. The latter approaches might be more useful for basic research since the MRI and CT instruments specialized for small animals are not assessable to all researchers.

Table 1.

Examples of tracers to quantify BBB disruption.

Tracer name	Molecular weight	Advantages	Disadvantages	Reference
Evans blue	Free form: 1 kDa Bounded form: 67 kDa	Binds to albumin and can be used to measure extravasation of albumin; Qualitative and quantitative assessment of BBB integrity.	Risk of overestimating BBB permeability to albumin; Extravasation kinetics cannot be determined.	[34]
Fluorescein isothiocyanat e (FITC)-dextran	3 – 2000 kDa	Can be coupled with different molecules; Based on the molecular weight can be used to test both ion and protein permeability.	Stability issue; Not suitable for small molecules permeability; Might be toxic at high concentrations.	[35, 47]
Horseradish peroxidase	44 kDa	Simultaneous assessment of BBB disruption and cellular morphology.	Extravasation kinetics cannot be determined.	[34]
Sodium fluorescein	376 Da	Inexpensive, nontoxic; Freely diffusible; Suitable for small molecules permeability prediction.	Interaction with BBB transporters; Weekly binds to plasma proteins.	[47]

Table 2.

Examples of morphometric markers to demonstrate BBB disruption

BBB markers	Cellular expression	Subcellular localization	Role in BBB	BBB transport Pathway	Reference
GLUT-1	Ubiquitous distribution with highest levels in erythrocytes and BBB endothelial cells [48]	Integral membrane protein	Glucose transportation	Transcellular	Downregulation [25, 26]
Claudin-5	Endothelial cell-specific expression [12]	Integral membrane protein	Essential TJs component for forming BBB	Paracellular	Downregulation/disorganization [45, 46]
Occludin	High expression in brain endothelial cells compared to nonneural tissues [14]	Integral membrane protein	Essential TJs component for forming BBB	Paracellular	Downregulation/disorganization [44, 45]
ZO-1	Expression in both epithelial and endothelial cells [49, 50]	Cytoplasm protein	Links integral membrane TJs proteins to actin cytoskeleton	Paracellular	Downregulation/disorganization [44, 45]

Overall, it is best to investigate the effects of short-term and long-term stroke injury by assessing both functional and structural changes in BBB characteristics. Herein, we describe a method using both quantitative and morphometric markers to evaluate BBB integrity in vivo.

2. Materials

2.1. Animals and reagents

1. CD-1 mice of 3–6 months old (Charles River Laboratories).

2. Fluorescein isothiocyanate-dextran (FITC-Dextran) of 4 kDa (Millipore Sigma, Cat# FD4), 70 kDa (Millipore Sigma, Cat# FD70S), 150 kDa (Millipore Sigma, Cat# FD150S,), and 2000 kDa (Millipore Sigma, Cat# FD2000S).

3. Ketamine (Butler Schein, Cat#010177).

4. Xylazine (Butler Schein, Cat# 33197).

5. Blunt scissors.

6. Micro dissection scissors.

7. Scalpel.

8. 27-gauge needle.

9. 21-gauge butterfly needle.

10. 1 mL syringe.

11. Glass beaker.

12. Graduated cylinder.

13. Stir bar.

14. Ice bucket.

15. Slide box.

16. Slide rack.

17. Slide moisture chamber.

18. Cryostat microtome (Leica, Cat# CM3050S).

19. Tissue-Tek O.C.T. Compound (Electron microscopy sciences, Cat# 62550–12).

20. Perfusion pump.

21. 96-well black plate (Fisher scientific, Cat# 07–200-340).

22. BioTek Synergy 2 microplate reader (Agilent).

23. Tabletop centrifuge.

24. Fisherbrand Homogenizer (Fisher scientific, Cat# 15–340-169).

25. Nikon A1RMP confocal microscope (Nikon Instrument).

26. 32% Paraformaldehyde (formaldehyde) aqueous solution (Fisher Scientific, Cat# 15714-S).

27. Triton^® X-100 (Acros Organics, Cat # 42235–5000).

28. Bovine serum albumin (BSA) powder.

29. Normal goat serum (Abcam, Cat # ab7481).

30. 10 X Phosphate Buffered Saline (PBS) Powder, pH 7.4 (Fisher Scientific, Cat # BP665–1).

31. Fisherbrand Tissue Path Superfrost Plus Gold Slides (Cat # 15–188-48).

32. Cover Glasses.

33. Distyrene plasticizer xylene (DPX) Mounting Medium for Microscopy (VWR international, Cat # 360294H).

34. DAPI (4’,6-diamidino-2-phenylindole, dihydrochloride) (Thermo Scientific, Cat # 62247).

35. Primary Claudin 5 Mouse Monoclonal Antibody (4C3C2) (Cat # 35–2500), ZO-1 Rabbit Polyclonal Antibody (Cat # 40–2200) and GLUT1 Rabbit Monoclonal Antibody (SA0377) (Cat # MA5–31960) were obtained from Invitrogen. Primary Occludin (E6B4R) Rabbit Monoclonal Antibody (Cat # 91131) was obtained from the Cell Signaling Technology.

36. Anti-rabbit secondary antibody, Alexa Fluor 488 (Cat # A-11008), anti-mouse secondary antibody, Alexa Fluor 488 (Cat # A-11001), anti-rabbit secondary antibody, Alexa Fluor 594 (Cat # A-11012), and anti-mouse secondary antibody, Alexa Fluor 594 (Cat # A-11005) were obtained from the Invitrogen.

37. Others: 70% alcohol, aluminum foil, dry ice, 1.5 mL and 2 mL Eppendorf tubes.

2.2. Solution preparation

1. 1 X PBS (0.137M NaCl, 0.0027M KCl, and 0.0119M phosphates), pH 7.4. Dissolve a pack (98.9 g) of 10 X PBS powder, pH 7.4, in 10 L deionized water.

2. Dilute all the FITC-Dextran tracers in sterile PBS into 2 mM stocks. Wrap the stock solution in aluminum foil for protecting from light and store at −20 °C (see Notes 4.1.1).

3. Cardiac perfusion buffer: 1 X PBS, pH 7.4.

4. Immunohistochemistry-fluorescence (IHC-F) washing buffer: 1 X PBS, pH 7.4.

5. IHC-F fixative Solution: 4% paraformaldehyde solution. Dilute 1 mL 32% paraformaldehyde aqueous solution in 7 mL 1 XPBS, pH 7.4 (dilution ratio 1:8).

6. IHC-F permeabilization solution: 0.1% Triton X-100 in 1 X PBS, pH 7.4. Dilute 10 μL Triton X-100 in 10 mL 1 X PBS, pH 7.4.

7. IHC-F blocking buffer: 1 X PBS, pH 7.4, containing 1% BSA and 2% normal goat serum. Dissolve 100 μg BSA in 10 ml 1X PBS, pH 7.4, and add 200 μL normal goat serum. Store any unused blocking buffer at 4 °C. The blocking buffer is also used as the antibody dilution buffer.

8. Antibody dilution buffer: same as the blocking buffer.

9. DAPI counterstain solution: dissolve 10 mg DAPI powder in 2 mL deionized water to make the stock solution. Wrap the stock solution in aluminum foil for protecting from light and store it at −20 °C. Dilute the stock solution at a ratio of 1: 20,000 in 1 X PBS, pH 7.4, to make a working solution.

3. Methods

All animals should be handled with the utmost care minimizing pain or discomfort during these procedures. The details of each step are described below.

3.1. Quantitative assays

3.1.1. Administration of tracers, cardiac perfusion, and serum and brain collection

1. After induction of stroke injury as previously described [51], intravenously (i.v.) inject 100 μL of each tracer (2 mM stock solution) to one mouse via tail vein with a 27-gauge needle. In addition, inject at least one animal with 1 X PBS alone to serve as sham control for autofluorescence background subtraction (see Notes 4.1.2).

2. Two minutes after tracer injection, anesthetize the mouse with an intraperitoneally (i.p) injection of Ketamine/Xylazine (100 mg/10 mg in saline solution per kilogram body weight, respectively).

3. Test paw retraction and ensure animal reaches a surgical plane of anesthesia.

4. Lay the animal on their back and sterilize the skin with 70 % ethanol. Open the abdominal wall with a small incision beneath the rib cage, and then cut the diaphragm to expose the pleural cavity.

5. Cut the rib cage bilaterally to expose the heart. Insert a 21-gauge butterfly needle connected to a perfusion system in the posterior of the left ventricle.

6. Puncture the right atrium and quickly collect 200 – 300 μL of blood released into the chest cavity using 1 mL pipette tip into 2 mL Eppendorf tubes and store the blood on ice. After 30 minutes, centrifuge the blood at 1500 g, 4 °C, for 10 minutes in a tabletop centrifuge to remove the clot. Collect the serum (top clear layer with light yellow color) for immediate fluorometry or stored at −80 °C for later fluorometry (see Notes 4.1.1).

7. Switch on the perfusion system (5 mL/minute) and perfuse the animal for 3 minutes with 1 X PBS. The total amount of PBS used for perfusion is in the range of 15 – 20 mL (see Notes 4.1.3 and 4.1.4). This perfusion procedure does not need to be performed in a fume hood.

8. At the end of perfusion, confirm death of the animal by cervical dislocation. Carefully harvest the brain and verify the successful brain perfusion by checking the color of the brains (no visible blood in vessels of the meninges) (see Notes 4.1.5).

9. Remove the olfactory lobes and cerebellum. Separate the cerebrum into 2 hemi-cerebrums (ipsilateral hemi-cerebrum could be larger than the contralateral hemi-cerebrum because of tissue edema after stroke injury) using a scalpel. Immediately put both hemi-cerebrums in 2 mL Eppendorf tubes on dry ice, respectively. After collecting all hemi-cerebrums, transfer them to −80°C freezer until further processing.

3.1.2. Homogenization and centrifugation

1. Weigh the hemi-cerebrum and thaw it on ice (see Notes 4.1.6). Add cold 1 X PBS at ratio 1: 1 (i.e., 100 mg tissue : 100 μL PBS) to the tube containing the hemi-cerebrum (see Notes 4.1.7).

2. Homogenize each sample in the original Eppendorf tube with a homogenizer. Rinse the homogenizer probe with 1 X PBS between samples and wipe it to dry before proceeding to the next sample.

3. Store the homogenized samples on ice protected from light (see Notes 4.1.1). After homogenizing all samples, centrifuge them together at 15,000 g, 4 °C, for 20 minutes in a tabletop centrifuge. Transfer homogenized tissue supernatants to a new 1.5 mL tubes on ice for immediate fluorometry or store at −80 °C for later fluorometry.

3.1.3. Fluorescence measurement and quantification

1. If the samples have been frozen at −80 °C in previous steps, thaw the homogenized tissue supernatants and serum samples on ice protecting from light.

2. Pipette 50 μL of homogenized tissue supernatants or diluted serum (20 μL serum + 30 μL 1 X PBS) into a 96-well black plate.

3. Insert the 96-well black plate into the microplate reader and set the excitation/emission (nm) values at 485 – 490 / 520 – 528 for FITC dye. Start fluorometry reading to obtain the raw fluorescence units (RFUs) of each sample.

4. Use the RFUs values after subtracting the corresponding sham autofluorescence background values to calculate the brain permeability index (PI) following the below formulas as previously described with minor modifications [42].

$$\text{Brain}\hspace{0.17em}\text{PI}\left(\text{mL/g}\right)\hspace{0.17em}\text{=}\hspace{0.17em}\text{Tissue}\hspace{0.17em}\text{RFUs/g}\hspace{0.17em}\text{tissue}\hspace{0.17em}\text{weight / Serum}\hspace{0.17em}\text{RFUs/mL}\hspace{0.17em}\text{serum}$$

3.1.4. Example of quantitation calculation

1. One trace-injected hemi-cerebrum with a weight of 200 mg (200 x 10⁻³ g) is homogenized in 200 μl 1 X PBS. Fifty microliters (50 μL) of the homogenized tissue supernatants is used for RFUs reading and have a RFUs value of 150. Twenty microliters (20 μL) of serum from the same animal plus 30 μL of 1 X PBS have a RFUs value of 30000.

2. One sham hemi-cerebrum with a weight of 180 mg (180 x 10⁻³ g) is homogenized in 180 μL 1 X PBS. Fifty microliters (50 μL) of the homogenized tissue supernatants is used for RFUs reading and have a RFUs value of 30. Twenty microliters (20 μL) of serum from the same animal plus 30 μL of 1 X PBS have a RFUs value of 20.

3. Tissue RFUs/g tissue weight = (150 – 30) x 200 μl / 50 μl x 200 x 10⁻³ g = 2.40 x 10³ RFUs/g tissue.

4. Serum RFUs/mL serum = (30000 – 20) / 0.02 mL = 1.50 x 10⁶ RFUs/mL serum.

5. Brain PI (mL/g) = 2.40 x 10³ RFUs/g tissue / 1.50 x 10⁶ RFUs/mL serum = 1.60 x 10⁻³.

3.2. Morphometric assays

3.2.1. Section slide preparation

1. After stroke injury, the mouse brain is rapidly dissected (avoiding damage the tissue) (see Notes 4.2.1).

2. Immediately put the whole brain on dry ice, and then store at −80°C freezer until further processing.

3. The frozen brain is placed at −20°C for 1 hour to achieve a temperature balance, then sectioned to 20 μM thickness using a cryostat instrument.

4. The tissue sections are mounted on the Fisherbrand Tissue Path Superfrost Plus Gold Slides. The section slides can be stored at −80°C if not used immediately.

3.2.2. IHC-F staining

1. Section slides are incubated in fixation solution for 15 minutes at room temperature.

2. Discard the fixation solution and incubate the sections in permeabilization solution for 10 minutes at room temperature (see Notes 4.2.2).

3. Wash sections with 1 X PBS three times for 5 minutes each.

4. Discard the wash buffer and block the sections with 100 μl blocking solution for 1 hour at room temperature.

5. Remove the blocking solution and drop 100 μl of the primary antibody of claudin-5, occludin, ZO-1 or GLUT-1 (diluted at 1: 100 for each) in the blocking solution to each section (see Notes 4.2.3).

6. Incubate the sections overnight at 4°C (see Notes 4.2.4).

7. Remove the primary antibodies and wash the sections with 1 X PBS three times for 5 minutes each.

8. Discard the wash buffer and drop 100 μl of the corresponding secondary antibody (diluted at 1: 200 for each) in blocking solution to each section.

9. Protect the sections from light and incubate them for 1 hour at room temperature (see Notes 4.2.5 and 4.2.6).

10. Remove the secondary antibodies and drop 100 μl of the DAPI working solution to each section.

11. Incubate the sections for 10 minutes at room temperature.

12. Remove DAPI working solution and wash the sections with 1 X PBS three times for 5 minutes each.

13. Discard the wash buffer and dry the sections at room temperature.

14. After the sections are dry, add 1 – 3 drops of mounting medium to each section and cover each section with a coverslip.

15. After the mounting medium solidifies, examine the sections under a laser Nikon A1RMP confocal microscope (see Notes 4.2.7).

4. Notes

4.1. Quantitative assays

1. The FITC-Dextran solution, homogenized tissue supernatants and serum samples should be protected from light to preserve the FITC signal.

2. Include at least one sample of serum and tissue supernatants from a sham animal, respectively, for autofluorescence background subtraction.

3. Perfusion pressure and speed should be consistent during the perfusion procedure to avoid microvascular disruption. Over hydraulic pressure may cause microvascular rupture leading to artificial BBB leakage.

4. Perfusion time should be no more than 3 – 4 minutes to avoid washing out tracers leaked into the brain parenchyma.

5. Check each brain after cardiac perfusion to make sure to achieve a successful perfusion. Brain without a successful perfusion should be excluded.

6. It is important to freeze the samples before proceeding to the homogenization steps as homogenization efficiency is increased after freeze-thaw cycle.

7. The use of detergents during brain tissue homogenization is avoided due to potential interference with fluorometry. Intracellular tracer is detectable in this protocol as the tissue is thoroughly homogenized.

4.2. Morphometric assays

1. To maximally preserve the morphology of BBB, the mouse brain should be dissected quickly and immediately put on dry ice.

2. The permeabilization time should not be more than 10 minutes. Over permeabilization may impact the membrane morphology.

3. If primary and secondary antibodies with different clones or vendors are used, the antibody dilution should be optimized for the best result.

4. The incubation time for the primary antibodies could be shortened to 2 – 4 hours at room temperature. However, incubation overnight at 4°C usually will give a better result.

5. The incubation time for the secondary antibodies could be extended to 2 hours if a stronger signal is preferred. However, over incubation of the secondary antibodies may increase the background noise.

6. From the step of incubation with the secondary antibodies, the sections should be protected from light. The immunofluorescent signal may last for two weeks if the sections are properly stored at 4°C.

7. To take a quality immunofluorescent image, a laser confocal microscope is required.