Real-Time Imaging of Immune Modulation by Cannabinoids Using Intravital Fluorescence Microscopy

Abstract

Introduction: The endocannabinoid system (ECS) is an endogenous regulatory system involved in a wide range of physiologic and disease processes. Study of ECS regulation provides novel drug targets for disease treatment. Intravital microscopy (IVM), a microscopy-based imaging method that allows the observation of cells and cell-cell interactions within various tissues and organs in vivo, has been utilized to study tissues and cells in their physiologic microenvironment. This article reviews the current state of the IVM techniques used in ECS-related inflammation research.

Methodological Aspects of IVM: IVM with focus on conventional fluorescent microscope has been introduced in investigation of microcirculatory function and the behavior of individual circulating cells in an in vivo environment. Experimental setting, tissue protection under physiologic condition, and microscopical observation are described.

Application of IVM in Experimental Inflammatory Disorders: Using IVM to investigate the effects of immune modulation by cannabinoids is extensively reviewed. The inflammatory disorders include sepsis, arthritis, diabetes, interstitial cystitis, and inflammatory conditions in the central nervous system and eyes.

Conclusion: IVM is a critical tool in cannabinoid and immunology research. It has been applied to investigate the role of the ECS in physiologic and disease processes. This review demonstrates that the IVM technique provides a unique means in understanding ECS regulation on immune responses in diseases under their physical conditions, which could not be achieved by other methods.

Introduction

Over the last 50 years, the endocannabinoid system (ECS) has emerged as an important endogenous system involved in a wide range of physiological and pathological processes. The ECS is composed of cannabinoid receptors (CBRs), their endogenous lipid-based ligands (endocannabinoids), and their cognate synthetic and degradative enzymes. Two well-characterized CBRs are the cannabinoid 1 receptor (CB1R) and the cannabinoid 2 receptor (CB2R). CB1R is densely located in the central nervous system (CNS), while CB2R is mainly expressed on central and peripheral immune cells, as well as in the gut, liver, and bones.^1–3 Among human immune cell populations, CB2Rs are highly expressed in neutrophils and macrophages, as well as T cells, B cells, and natural killer (NK) cells.⁴ The two major physiological CB1R and CB2R ligands are the endocannabinoids anandamide (AEA) and 2-arachidonoyglycerol (2-AG), and they are synthesized on demand.^5,6 In the cell, fatty acid amide hydrolase (FAAH) hydrolyses both AEA and 2-AG, whereas monoacylglycerol lipase (MAGL) metabolizes only 2-AG.⁷

One way to study the ECS is through in vitro or ex vivo methods, such as cell cultures or histology/immunohistochemistry. Although cell cultures can be manipulated genetically and pharmacologically to mimic components of human biological systems, they often do not provide a true representation of the biology of complex multicellular organisms in a living physiologic system.⁸ To overcome this, intravital microscopy (IVM) can be used to study tissues in vivo in their microenvironment, but within the greater physiologic system. IVM is a microscopy-based imaging method that allows the observation of cells and cell-cell interactions within various tissues and organs.^8,9 IVM has been widely used in small animals, such as rodents, in particular, in the fields of neurology and immunology research.^8,10 Recently, IVM has been applied to investigate the role of the ECS in physiologic and disease processes,^11–13 which has proven valuable in both cannabinoid drug discovery and target validation.

Methodological Aspects of IVM

Studies using IVM as a research tool have been established after the development of fluorescence microscopes and injectable fluorophores in early 20th century.¹⁰ The conventional wide-field fluorescent microscope, such as an epifluorescent microscope, has been used extensively to investigate parameters of microcirculatory function and the behavior of individual circulating cells in an in vivo setting. In addition, this technique allows the temporal observation of changes in a tissue, spanning seconds to days.

The light source of a conventional fluorescence microscope emits a specific wavelength, which excites a fluorescent chemical, the fluorophore, causing it to emit a lower energy wavelength light, thus signaling its presence.⁸ Using fluorescence IVM, researchers have studied leukocyte-endothelium interactions in the microcirculation and capillary perfusion in various tissues and organs, including the intestine, brain, eye, joint, pancreas, and bladder.^11–18 Multiphoton and laser scanning confocal microscopy have also been developed to allow visualization of deeper tissues, such as in brain parenchyma.¹³ In this review, we will focus on conventional IVM.

IVM is usually performed with a fluorescence microscope equipped with a digital camera. The image from the camera is displayed on a computer monitor, captured, and recorded for offline data analysis (Fig. 1). Before examination with the microscope, most experiments first require a surgical procedure under anesthesia to expose the tissue of interest and prepare an image window. The surgery can be performed either shortly before IVM (i.e., for viewing intestine, pancreas, bladder, and brain)^11,14,16,19 or several days before (i.e., for viewing brain).^17,18 IVM can also be performed directly on some organs, without the need for surgical procedures, such as for the eye (including the iris and retina) or the tongue.^20–24

FIG. 1.

An intravital microscopy setting for performing the examination of mouse intestinal microcirculation. Images from the microscope (M) were captured and recorded by the videotape and transferred into the computer (C) and analyzed offline. Images depict leukocyte rolling and adhesion (L) within venules and capillary perfusion (P).

To facilitate IVM imaging, animals need to receive an injection of one or more fluorophore solutions or fluorescence-labeled cells to allow target cell visualization. For example, rhodamine 6G solution can be administered intravenously to stain mitochondrion-containing cells, for example, for the visualization of leukocytes (Fig. 2), whereas fluorescein isothiocyanate (FITC) conjugated albumin or dextran is administered for observation of capillary perfusion and vascular permeability.^11,25 Fluorescence-conjugated antibodies specific to an antigen of interest expressed on cells have also been used to visualize the target cells, such as neutrophils.²⁶ In some experiments, transgenic fluorescent animals were used, such as fluorescent zebra fish or transgenic mice expressing green fluorescent protein.²⁷

FIG. 2.

A still-frame image of adherent leukocytes in the microcirculation of a mouse bladder. Images obtained by intravital microscopy (magnification=200×) with intravenously injected rhodamine 6G solution. White arrows indicate adherent leukocytes.

During IVM, it is important to maintain the observed tissue and organ as close as possible to their physiological condition—such as through maintenance of temperature, pH, and humidity. Since small blood vessels are extremely sensitive to changes in temperature and humidity, cooling and drying of the observed tissue may generate erroneous results. Many researchers have developed their own devices to protect against tissue damage during IVM. For example, to study the intestinal microcirculation, Pavlovic et al. created a thermostatic tissue platform that consisted of an organ bath filled with circulating warm (37°C) physiological solution, an organ-support hand for holding the intestinal tissue, and a coverslip with plastic ring (Fig. 3).²⁸ During IVM, the intestinal segment of interest was gently exteriorized and placed on the “hand” of the organ support. It was then covered with a coverslip and perfused with warm physiological solution from a small tube, which was fixed on the organ bath and pointed to the lower surface of the coverslip, forming a “hanging drop” (Fig. 3). This tissue platform with a “hanging drop” technique provided stable temperature and humidity to prevent tissue quenching and undisturbed microcirculation.²⁸

FIG. 3.

An image of a thermostatic tissue platform modified from Pavlovic et al.²⁸ It consists of an animal platform (A), an organ bath (B), an organ support hand (C) for holding the intestinal tissue, and a circulating warm solution (W). An exteriorized intestine segment is placed on the “hand” of the organ support and covered with a coverslip. With the addition of a physiological solution (37°C) coming from a tube, it will form a “hanging drop.”

To observe the microcirculation on the surface of the brain, a cranial window is often utilized, which is created by circular or rectangle craniotomy using a high-speed drill, located over the right or left parietal cortex, and aligned to the middle of the sagittal suture.^19,29 The cranial window can be filled with warm physiological solution and the pial microvasculature can be reviewed by IVM immediately thereafter.¹⁹ In other experiments, researchers placed a coverslip over the exposed brain, which produced an airtight seal using tissue glue, before allowing the animal to recover for 4–7 days before IVM observation.^17,18,29 This recovery period allows the pial vessels to be studied multiple times, and further acts to reduce inflammation that may be generated by the surgery.

To observe the microcirculation of the pancreas and bladder, these organs must be externalized from the abdomen and placed on an organ holder, before being covered with a coverslip and warm normal saline.^14,16 For studies of the joint microcirculation, the joint is surgically opened and covered with warm normal saline before being examined directly under the microscope.¹⁵ For observation of the eye microcirculation, lubricating gel is generously applied to the ocular surface to prevent corneal desiccation, as well as to assist with placement of a coverslip and enhance optical resolution.^23,30

Application of IVM on ECS Research on Experimental Inflammatory Disorders

Intestinal microcirculation in sepsis

Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection.³¹ It involves a systemic inflammatory response with characteristics of an early hyperacute proinflammatory response followed in some patients by the development of systemic immunosuppression. One of the remarkable characteristics of the inflammatory response is the dysregulation of microcirculatory blood flow with hypoperfusion leading to organ dysfunction and failure.^8,32 Among the organs, dysregulation of intestinal microcirculation is critical because of the potential translocation of bacteria from the gut into circulation, which is an important event in sepsis pathology.^33,34 Therefore, researchers often examine the intestinal microcirculation, using IVM, to investigate the immune response and microcirculatory function in animals with sepsis.^11,12,25,35

Since CB1R and CB2R are extensively expressed on intestinal tissue, targeting the ECS may have a potential therapeutic effect on inflammatory disorders.^34,36 A group of researchers have studied the effects of ECS modulation in the setting of sepsis by visualizing leukocyte-endothelial interactions and functional capillary density (FCD) in the intestinal microcirculation of experimental animals with sepsis^{11,12,25,35,37} (summarized in Table 1). It has been shown that increased leukocyte adhesion was present in the intestinal venules in both lipopolysaccharide (LPS)-induced endotoxemia and colon ascendens stent peritonitis (CASP), a polymicrobial abdominal sepsis model, in rats.¹¹ The increased leukocyte adhesion was corresponding to an increased inflammatory cytokine production and soluble adhesion molecule levels in the plasma (e.g., interleukin 1 [IL-1], tumor necrosis factor-alpha [TNF-α], RANTES, macrophage inflammatory protein-2 [MIP-2], intercellular adhesion molecule-1 [ICAM-1], and vascular cell adhesion molecule-1 [VCAM-1]).¹¹ These data suggest that hyperimmune responses generated in experimental septic models can be detected by IVM. A single intravenous administration of the CB2R agonist, HU308, 15 min post-LPS injection or CASP surgery significantly attenuated increased immune responses in these animals with sepsis.¹¹ A high dose of HU308 (10 mg/kg) was required to attain similar effects in the CASP model compared to a low dose (2.5 mg/kg) for LPS models, suggesting that the effect of CB2R activation on inflammation is dose dependent and varies according to the experimental conditions.¹¹

Table 1.

Summary of Main Intravital Microscopy Findings in the Microvasculature of Various Organs After Drug Treatment

Organ	Trigger of inflammation	Drug treatment	IVM results	Reference
Intestine	LPS i.v. or CASP in rats	HU308	↓ Leukocyte adhesion	Lehmann et al.11
	LPS i.v. in mice	JZL184	↓ Leukocyte adhesion	Sardinha et al.12
	LPS i.v. in rats	URB597	↓ Leukocyte adhesion	Kianian et al.37
		URB597+AM630	Reversed the effect of URB597
	LPS i.v. in mice	CID16020046 or O-1918	↓ Leukocyte adhesion	Zhou et al.25
		LPI or O-1602	No effect on adhesion
	LPS i.v. in rats	AM281	↓ Leukocyte adhesion ↑FCD	Kianian et al.35
		AM281+ACEA	Reversed the effect of AM281 on FCD
Brain	LPS i.p. in mice	JWH133 or O-1966	↓ Leukocyte adhesion ↓ Leakage of FITC-dextran	Ramirez et al.17
		Co-administration with SR144528	Reversed the effect of JWH133 or O-1966
	LPS i.p. in mice	Transferring CB2 agonistsa treated leukocytes isolated from WT or CB2KO	From WT: ↓ Leukocyte adhesion From CB2KO:↑ Leukocyte adhesion	Rom et al.41
	LPS i.v. in mice	CBD	↓ Vasodilation ↓ Leukocyte margination ↓ Dextran extravasation	RuizValdepenas et al.42
	LPS i.p. in mice	Transferring CD4+ T cells treated with CB1 inverse agonistsbex vivo	Inhibits antigen specific T cell adhesion in LPS treated mouse brain venules	Rossi et al.43
	MOG s.c. & PTX i.p. in mice	WIN55212-2	↓ Leukocyte rolling and firm adhesion.	Ni et al.46
		Co-administration of SR144528, not SR141716A	Reversed the effect of WIN55212-2
Eye	LPS i.vi. in mice	HU308	↓ Leukocyte adhesion	Toguri et al.20,21, Porter et al.60
	Dispase i.vi. in mice	HU308	↓ Leukocyte adhesion	Szczesniak et al.24
Joint	MIA i.k. in rats	CBD	↓ Leukocyte rolling ↓ Leukocyte firm adhesion	Philpott et al.15
		CBD+AM630	Blocked the effect of CBD
	MIA i.k. in mice	URB597	↓ Leukocyte rolling ↓ Leukocyte firm adhesion ↓ Microvascular perfusion	McDougall et al.65
	C/K i.k. in mice	URB597	↓ Leukocyte rolling ↓ Leukocyte firm adhesion ↓ Microvascular perfusion	Krustev et al.64
		URB597+AM215 or AM630	Blocked the effect of URB597 on leukocyte rolling & microvascular perfusion
Pancreas	NOD mouse	CBD	↓ Leukocyte adhesion ↑FCD	Lehmann et al.14
Bladder	LPS i.ve. in mice	BCP or HU308	↓ Leukocyte adhesion ↑FCD	Berger et al.16

^aCB2 agonists: GP1a, JWH133 or AM1241; ^bCB1 inverse agonist: SR1416716A, AM251, AM281; CD4+ T cells were isolated from the mice immunized by PLP139-151 peptide, which is specific for generation of experimental encephalomyelitis.

↓, decrease; ↑, increase; ACEA, CB1 agonist; AM281, CB1 antagonist; AM630, CB2R antagonist; BCP, beta-caryophyllene; C/K, carrageenan + kaolin; CASP, colon ascendens stent peritonitis; CB1R, cannabinoid 1 receptor; CB2R, cannabinoid 2 receptor; CBD, cannabidiol; CID16020046 and O-1918, GPR55 antagonists; FCD, functional capillary density; FITC, fluorescein isothiocyanate; GPR55, G protein-coupled receptor 55; HU308, CB2 agonist; i.k., knee joint space; i.p., intraperitoneal injection; i.v., intravenously injection; i.ve., intravesical injection; i.vi., intravitreal injection; JWH133 and O-1966, CB2 agonists; JZL184, monoacylglycerol lipase inhibitor; KO, knockout; LPI, lysophosphatidylinositol; LPI and O-1602, GPR55 agonists; MIA, sodium monoiodoacetate; MOG, myelin oligodendrocyte protein; NOD, nonobese diabetic; PTX, pertussis toxin; s.c., subcutaneous injection; SR144528, CB2 antagonist; URB597, fatty acid amide hydrolase inhibitor; WIN55212–2, agonist for both CB1R and CB2R; WT, wild type.

To further study the effects of ECS modulation on sepsis, researchers have assessed the effects of inhibitors of endocannabinoid-degrading enzymes, such as JZL184 (MAGL inhibitor) or URB597 (FAAH inhibitor), in endotoxemic animals to increase endocannabinoid levels.^12,37 They demonstrated that either inhibitor ablated the LPS-induced increase of adherent leukocytes in the intestinal venules 2 h post-LPS injection.^12,37 The effect of URB597 was reversed by co-administration of the CB2R antagonist, AM630, suggesting that the anti-inflammatory effect of FAAH is through CB2R.³⁷

G protein-coupled receptor 55 (GPR55) is a lysophosphatidylinositol (LPI) receptor and was considered a potential novel CBR due to its high affinity to natural and synthetic cannabinoids, and endocannabinoids.³⁸ In a recent IVM study in LPS-induced endotoxemia, it has been shown that administration of GPR55 antagonists, CID16020046 or O-1918, post-LPS injection significantly decreased the numbers of adherent leukocytes in the intestinal submucosal venules.²⁵ The decreased leukocyte adhesion was corresponding to a reduction of IL-6, a common proinflammatory cytokine present in animals with endotoxin-induced sepsis.²⁵ Administration of GPR55 agonists (LPI and O-1602) did not increase the numbers of adherent leukocytes and IL-6 production significantly in the endotoxemic animals. Taken together, these data suggest that GPR55 plays an inhibitory role in the immune response in experimental endotoxemia.²⁵

Using IVM, researchers also investigated the role of CB1R on intestinal microcirculation in experimental endotoxemia.³⁵ Administration of a CB1R antagonist (AM281) in this model significantly reduced the numbers of adherent leukocytes and improved FCD in the intestinal microcirculation. Although administration of a CB1R agonist (ACEA) alone did not further increase leukocyte adhesion in the endotoxemic mice, co-administration of ACEA with AM281 led to a reversal of the beneficial effect of the CB1R antagonist on FCD.³⁵

Inflammatory disorders of the CNS

Neuroinflammatory disorders of CNS arise from diverse etiologies, including CNS infections, injury, autoimmune processes, and conditions involving neurodegeneration. Examples include encephalomyelitis, multiple sclerosis, CNS vasculitis, and Huntington's disease. Under physiological conditions, the blood–brain barrier (BBB) prevents neurotoxic plasma components, blood cells, and pathogens from entering the brain, in addition to acting as a regulator of molecular transport into and out of the CNS, and acting as a mediator of communication between the CNS and the periphery. Breakdown of the BBB occurs in the setting of inflammatory disorders of CNS.³⁹

To examine the role of CB2R function in the setting of both BBB dysfunction and immune cell function under inflammatory conditions, Ramirez et al. used IVM through a cranial window to directly observe the cerebral vascular changes and leukocyte-endothelial cell interactions in real time in a model of LPS-induced encephalitis mice.¹⁷ They found that administration of either of the CB2R-selective agonists JWH133 or O-1966 significantly attenuated leukocyte adhesion in pial vessels and in deep ascending cortical postcapillary venules. In addition, JWH133 or O-1966 reduced leakage of FITC-dextran (70 kDa) from the cerebral vessels into the brain parenchyma—a phenomenon characteristic in the setting of neuroinflammation. Co-administration of the CB2R-selective antagonist SR144528 reversed the aforementioned results¹⁷ (Table 1). Together, these data suggest protective effects of CB2R activation on leukocyte activation and BBB function. In their experiments, the pial vessels were observed using a conventional wide-field fluorescence (epifluorescence) microscope, while the deep-tissue imaging was studied by a multiphoton microscope due to its improved depth penetration and reduced photodamage.⁴⁰ Their IVM finding corresponded with their ex vivo data, in that, administration of CB2R agonist O-1966 following LPS injection decreased ICAM-1 and VCAM-1 expression in the brain sections examined by immunofluorescence histology. These data suggested that CB2R may provide a therapeutic target, not only for their effects on attenuating immune cell activation but also in protecting the integrity of the BBB during neuroinflammation.

Using an adoptive cell transfer experiment with ex vivo CB2R-activated leukocytes and IVM, Rom et al. examined the impact of CB2R activation on leukocyte-endothelial cell interactions in LPS-induced encephalitis.⁴¹ They isolated leukocytes from mice lacking CB2R (CB2KO) and wild-type (WT) CB2R-expressing animals, and then treated the leukocytes with or without CB2R agonists GP1a, JWH133, or AM1241 ex vivo. These treated cells were then stained with a fluorescent dye, calcein-AM, and injected intraorbitally into the WT mice, which had previously received a cranial window implant and intraperitoneal LPS.⁴¹ After injection of rhodamine 6G, leukocytes were visualized in cerebral vessels through the cranial window by IVM (fluorescent light 495 nm excitation for calcein-AM and 601 nm for rhodamine 6G). CB2R activation in leukocytes (stained by calcein-AM) isolated from WT animals significantly inhibited endothelial leukocyte adhesion in cortical vessels, while the autologous leukocytes (labeled by rhodamine 6G) showed a uniformly increased adhesion, indicative of the expected inflammatory response. Adoptive transfer of the leukocytes isolated from CB2KO mice showed increased adhesion, confirming the role of CB2R activation in the attenuation of brain inflammation.⁴¹

In other experiments of LPS-induced brain inflammation, Ruiz-Valdepeñas et al. demonstrated that CBD, a phytocannabinoid with a diverse receptor-binding profile, prevented pial arteriolar and venular vasodilation, as well as leukocyte margination (immobilized leukocyte inside the vessels).⁴² In their experiments, the 70,000 MW Texas red-conjugated dextrane solution was injected to stain blood plasma, while leaving nucleated cells unstained, which appears as so-called “ghosts” inside the vessels. With this approach, only cells that are stationary or dramatically slowed by adhesive interactions with the vessel wall can be detected. In addition, BBB integrity can be evaluated by assessing the fluorescence intensity across the brain blood vessels, since 70,000 MW dextrane is unable to leave the blood vessels under normal conditions, but diffuses into the brain parenchyma when BBB integrity is compromised.⁴² It was demonstrated that a single dose of CBD reduced LPS-induced vasodilation and leukocyte margination. In addition, CBD treatment significantly reduced the extent of dextrane extravasation, suggesting a protective role of CBD on BBB integrity.⁴³

To examine the effects of CB1R and CB2R modulation on T cell function in the setting of brain inflammation, Rossi et al. performed adoptive transfer experiments with antigen-specific T cells in the setting of LPS-induced neuroinflammation in mice.⁴³ Antigen-specific CD4⁺ T cells were isolated from draining lymph nodes that were pre-immunized with PLP139–151 peptide (specific for generation of experimental encephalomyelitis) 10 days prior. These were then expanded in vitro by re-stimulation with PLP139–151 peptide. The CD4⁺ T cells were treated with a CB2R inverse agonist (SR144528) or a CB1R inverse agonist (one of SR1416716A, AM251, and AM281) and labeled with fluorescent dye CMTMR or CMFDA before being adoptively transferred to the mice that had previously received LPS.⁴³ IVM examination of the brain vasculature showed that each of SR1416716A, AM 251, and AM281 was able to inhibit the antigen-specific (encephalitogenic) T cell adhesion in LPS-treated mouse brain venules; however, that treatment with the CB2R inverse agonist, SR144528, had no effect.⁴³ These data suggested that selective CB1R inverse agonism interfered with lymphocyte trafficking in the brain. Ex vivo studies further suggest that this may be mediated by blocking of signal transduction mechanisms, which control encephalitogenic T cell adhesion in inflamed brain venules by a PKA-dependent mechanism, rather than adhesion molecule expression on the cell surface.⁴³

It has been suggested that activation of CB2R by synthetic cannabinoids has a protective effect in experimental autoimmune encephalomyelitis (EAE), an experimental model of multiple sclerosis.^44,45 EAE, induced by injection of myelin oligodendrocyte protein (MOG) and pertussis toxin (PTX), leads to increased numbers of leukocyte rolling and firm adhesion in the brain venules by observation using IVM.⁴⁶ WIN55212–2, an agonist for both CB1R and CB2R, decreased leukocyte rolling and firm adhesion induced in EAE.⁴⁶ Administration of the CB2R antagonist, SR144528, but not the CB1R antagonist, SR141716A, reversed the effect of WIN55212–2. This suggests that CB2R activation had anti-inflammatory effects in the setting of EAE.⁴⁶

Taken together, these in vivo IVM data suggest that the ECS plays a critical role in inflammatory CNS disorders. Activation of CB2R inhibits the inflammatory response in the CNS, while activation of CB1R also prevents T lymphocyte adhesion in inflamed brain and contributes to an anti-inflammatory response through a PKA-dependent mechanism.

Ocular inflammatory conditions

The eye is a complex organ with structures that are readily accessible for visualization. This presents a unique ability to use noninvasive imaging techniques to assess eye health and diagnose ocular pathology. It also imparts the ability to quantify various parameters rapidly and reliably in ocular disease models—such as through quantification of leukocytes to assess levels of ocular inflammation. With both CB1R and CB2R identified in ocular tissues,^47–51 modulation of the ECS has a range of potential therapeutic benefits in this setting. A major therapeutic target and research focus are ECS-based anti-inflammatory actions, where many ocular pathologies include some component of inflammation. Specifically, ocular inflammation can arise from infectious or noninfectious etiologies and can lead disabling ocular pain and irritation, in addition to potentially damaging the delicate ocular tissues involved in vision. Any level of inflammation in the eye has the potential to be vision-threatening. Other ECS-based ocular therapeutic strategies include various CBR-based and ECS metabolic enzyme-based treatment strategies targeting pain, inflammation, and other aspects of ocular physiology such as intraocular pressure.

The eye is a so-called “immune-privileged” environment, whereby the tissues are shielded from the systemic immune system by physical barriers, including the blood-ocular barrier, soluble and cell-bound immunosuppressive factors, and direct modulation of immune responses by ocular tissues.⁵² This reflects a physiologic adaptation aimed at avoidance of immune activation to prevent inflammation-mediated damage of ocular structures. The range of potential clinical applications for ECS modulation-based therapeutics in the setting of ocular disease are reviewed in detail by Toguri et al., Cairns et al., and Lafreniere and Kelly.^30,53–55 In terms of inflammation, “uveitis” describes ocular inflammation of various etiology and of vast presentation and structural involvement. The inflammation in uveitis can range from mild to severe and can be anatomically isolated (e.g., anterior uveitis) or involve all ocular structures (so-called “panuveitis”). Disability assessments demonstrate the marked impact on a patient's quality of life, with symptoms including potentially severe pain, irritation, redness, and loss of visual acuity.⁵⁶ Compounding the need for effective treatment options, experts acknowledge that uveitis as a disease is critically underserved in terms of available pharmacological agents.⁵⁷

A range of inflammation-triggering substances can be used to initiate animal models of ocular inflammation, which are generally administered by intravitreal (IVT) injection. A commonly used acute inflammatory trigger is LPS. LPS rapidly and reliably generates an endotoxin-induced uveitis (EIU), with inflammation first detectable within 1 h. Other substances include peptidoglycan,⁵⁸ from Gram-positive bacteria, or antigenic substances such as interphotoreceptor retinoid-binding protein (IRBP) to generate experimental autoimmune uveitis.^59,60 Due to their ready access for imaging purposes, vessels of the iris present an ideal target for the use of IVM, allowing direct quantification of leukocytes. Anatomically, the iris and anterior chamber are bathed in the IVT-administered inflammatory trigger due to posterior-to-anterior direction of flow of aqueous humor and as such function as reliable indicators of ocular inflammatory status.

Specifically, the EIU model induces inflammation primarily through Toll-like receptor 4 (TLR-4) activation, in turn inducing the production and release of various inflammatory mediators, including cytokines, chemokines, and cellular adhesion molecules, through increased signaling by the MyD88/NF-κB pathway.²⁰ Using this model, Toguri et al. assessed the effect of a single dose of the selective CB2R receptor agonist, HU308, delivered topically following IVT injection of LPS. Inflammation was assessed by IVM of rhodamine 6G-labeled leukocytes adhering to the vascular endothelium of iris venules at 6 h post-IVT, demonstrating significantly reduced leukocyte levels following HU308 treatment. Furthermore, HU308 led to significantly reduced levels of various proinflammatory mediators (TNF-α, IL-1ß, IL-6, IFN-γ, CCL5, and CXCL2) in ocular tissues at 6 h. Along this line, a decrease in messenger RNA (mRNA) levels of NF-κB, a key transcription factor for proinflammatory mediators, was also noted following HU308 treatment. The effects of common clinical treatments used for ocular inflammation, including the corticosteroids dexamethasone and prednisolone, and the nonsteroidal anti-inflammatory drugs (NSAID), nepafenac, were also assessed using the same dosing regimen and route as HU308, and found to be inferior to the anti-inflammatory effect of HU308. The EIU model has been used in other studies examining the efficacy of anti-inflammatory ECS modulators,⁶¹ including with LPS administered through the intravenous route that establishes systemic inflammation, in addition to a uveitis.²¹

Although the use of IVM to study uveitis may be of particular focus in the literature, there are a range of other applications, including other ocular inflammatory conditions. For example, Szczesniak et al. used IVM to assess inflammation in a dispase-induced model of proliferative vitreoretinopathy (PVR) in mice.²⁴ PVR is characterized by the formation of ectopic membranes above and below the retina and clinically occurs as a complication of rhegmatogenous retinal detachment. Key in the pathogenesis of PVR are growth factors and pro-inflammatory mediators that drive cell recruitment and membrane growth. Dispase is a neutral protease that cleaves the basement membrane of the retina, inducing retinal detachment and the associated inflammatory response marked by inflammatory cell infiltration and the release of inflammatory mediators. In this model, treatment with topical HU308 attenuated ocular tissue pathology, and decreased microglial and macrophage recruitment to the retina. This effect was lost with the addition of a selective CB2R antagonist, suggesting that the therapeutic actions of HU308 were a result of its agonist activity at the CB2R. IVM indicated that HU308 treatment led to decreased levels of leukocytes adhering to the endothelium in iris venules. Interestingly, Szczesniak et al. reported that the addition of a CB2R antagonist alone to WT mice or examining CB2R KO mice supported the role of CB2R in ocular immune regulation, as both of these groups were found to have significantly increased levels of proinflammatory mediators.⁶¹

The use of IVM techniques has the potential for imaging of other ocular structures, namely the retina. Various retinal pathologies, including age-related macular degeneration and diabetic macular edema, involve a component of inflammation, in addition to pathologic angiogenesis—leading to antivascular endothelial growth factor (anti-VEGF) biologics such as bevacizumab (Avastin^®) being the clinical gold standard of treatment. Similarly, with described anti-inflammatory and antiangiogenic actions, ECS modulators are exciting novel agents with therapeutic potential in this setting. IVM of the retinal microvasculature would likely be useful in animal models of these conditions, to quantify leukocytes and assess the FCD. Looking to the ocular surface, conjunctival/episcleral vessels may be imaging targets in models of dry eye disease, episcleritis, and scleritis. Similar to the iris vasculature used as an imaging target in the studies described above, the vessels on the anterior ocular surface would be readily accessible.

Arthritis of joints

Arthritis describes inflammation of the joint, with symptoms including joint pain, swelling, stiffness, and decreased range of motion. The most common forms of arthritis are osteoarthritis (OA) and rheumatoid arthritis, which involve joint degeneration. Recent research has demonstrated that CBRs are found on chondrocytes and bone.^62,63 Using IVM, the effects of ECS modulation on joint inflammation have been studied in several experimental models. In one study, OA was induced in rats by injecting sodium monoiodoacetate (MIA) into the knee joint space.¹⁵ Acute joint inflammation was observed by IVM 1 day after MIA injection, which induced an increased number of leukocytes rolling and firm adhesion in the microvasculature of joints. This leukocyte trafficking was reduced by local application (through a close distal saphenous intra-arterial administration) of CBD to the joint. Co-administration of AM630, a CB2R antagonist, blocked the CBD-induced reduction in leukocyte trafficking, indicating that the protective effect of CBD is CB2R mediated.¹⁵

In another experiment, acute joint inflammation in mice was generated by injection of carrageenan and kaolin into the joint.⁶⁴ Local treatment with URB597 (a selective FAAH inhibitor) locally decreased leukocyte rolling, adhesion, and microvascular perfusion. Co-administration of either a CB1R antagonist (AM251) or a CB2R antagonist (AM630) blocked the anti-inflammatory effects of URB597 on leukocyte rolling and vascular perfusion, but not leukocyte adhesion. These data suggest that the anti-inflammatory effect of URB597 on leukocyte rolling and microvascular perfusion is CB1R and CB2R mediated, while the effect on leukocyte adherence is independent of the receptor activation.⁶⁴ A similar reduction in leukocyte adherence following treatment with URB597 was also demonstrated in the model of MIA-induced knee inflammation.⁶⁵

Diabetes

Type 1 diabetes (T1D) is an autoimmune disease that involves a loss of insulin-producing β-cells in the pancreas, leading to an inability to maintain glucose regulation. Histopathologically, T1D is characterized by immune cell infiltration of pancreatic tissue and associated destruction of pancreatic β cells.⁶⁶ The “nonobese diabetic (NOD) mouse” is the most frequently used animal model for studying T1D.^14,67 Using NOD mice and IVM, Lehmann et al. demonstrated the correlation between inflammatory damage of the pancreas and the numbers of adherent leukocytes in pancreatic microcirculation.⁶⁷ Inflammatory damage was evaluated by the level of blood glucose.⁶⁷ The NOD mice with low-degree inflammatory changes (indicated by low level of blood glucose) did not show signs of leukocyte activation, while increased numbers of adherent leukocytes were present in the NOD mice with high-degree inflammation.⁶⁷ NOD mice treated with daily CBD for 10 weeks showed delayed onset of pancreatic inflammation with a significant reduction in the number of adherent leukocytes as well as increased FCD.¹⁴ These data were consistent with decreased plasma levels of soluble adhesion molecules ICAM-1 and P-selectin in CBD-treated NOD mice. These data suggested that experimental CBD treatment is able to reduce immune cell activation within the pancreatic microcirculation in early T1D.

Interstitial cystitis

Interstitial cystitis (IC) is a chronic inflammatory disorder of the urinary bladder with unclear etiology. Both CB1R and CB2R have been identified on the sensory nerves of the detrusor (smooth muscle in bladder wall) and mucosa (urothelium and suburothelium) within the bladders of various species, including rats, mice, monkeys, and humans,^68,69 suggesting that modulation of the ECS may have potential in terms of developing novel therapeutics for IC. Evidence showed that the CB2R agonist, JWH015, reduced cellular infiltration and mRNA expression of proinflammatory cytokines in LPS-induced mouse IC bladder tissue.⁷⁰ Using IVM, the effects of CB2R activation on immune responses and microcirculatory function were also revealed in vivo in an LPS-induced IC model in mice.¹⁶ It was demonstrated that instillation into the urinary bladder (intravesical) of beta-caryophyllene (BCP), a natural dietary sesquiterpenoid, which has agonist actions at the CB2R,⁷¹ significantly reduced the number of adhering leukocytes in submucosal bladder venules, and improved bladder capillary perfusion.¹⁶ The effects of BCP were comparable to that of the selective CB2R agonist, HU308, and superior to intravesical dimethyl sulfoxide (DMSO), an agent with FDA approval for local application in the clinical treatment of IC.¹⁶ Oral administration of BCP reduced bladder inflammation and significantly reduced mechanical allodynia in experimental IC. These data suggest that CB2R activation may represent a viable therapeutic target for IC, and drugs that activate CB2R, such as the FDA-designated “generally regarded as safe—GRAS” dietary sesquiterpenoid, BCP, may serve as an adjunct and/or alternative treatment option for alleviating symptoms of inflammation in the management of IC.

Conclusion

In this review, we have provided information on the methodologies used to image specific organs and on the application of IVM on studying the role of ECS in inflammatory disorders in vivo in their physiologic microenvironment. IVM has been used as an important tool in cannabinoid and immunology research, providing unique advances in our understanding of immune responses in diseases in their physical condition, which could not be achieved by other methods. The IVM data included in this review suggest that the ECS plays a critical role in experimental inflammatory disorders, including sepsis, arthritis, diabetes, interstitial cystitis, and inflammatory conditions in the central nervous system and eyes. Modulation of ECS can be used for drug discovery and therapeutic validation.

Abbreviations Used

2-AG

2-arachidonoyglycerol

AEA

anandamide

BBB

blood–brain barrier

BCP

beta-caryophyllene

CASP

colon ascendens stent peritonitis

CB1R

cannabinoid 1 receptor

CB2R

cannabinoid 2 receptor

CBD

cannabidiol

CBR

cannabinoid receptor

CNS

central nervous system

EAE

experimental autoimmune encephalomyelitis

ECS

endocannabinoid system

EIU

endotoxin-induced uveitis

FAAH

fatty acid amide hydrolase

FCD

functional capillary density

FITC

fluorescein isothiocyanate

GPR55

G protein-coupled receptor 55

IC

interstitial cystitis

ICAM-1

intercellular adhesion molecule-1

IL

interleukin

i.p.

intraperitoneal injection

i.v.

intravenously injection

IVM

intravital microscopy

IVT

intravitreal

KO

knockout

LPI

lysophosphatidylinositol

LPS

lipopolysaccharide

MAGL

monoacylglycerol lipase

MIA

monoiodoacetate

mRNA

messenger RNA

NOD

nonobese diabetic

NSAID

nonsteroidal anti-inflammatory drugs

OA

osteoarthritis

PVR

proliferative vitreoretinopathy

s.c.

subcutaneous injection

T1D

type 1 diabetes

TNF-α

tumor necrosis factor-alpha

VCAM-1

vascular cell adhesion molecule-1

WT

wild type

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this review.

Cite this article as: Zhou J, Kamali K, Lafreniere JD, Lehmann C (2021) Real-time imaging of immune modulation by cannabinoids using intravital fluorescence microscopy, Cannabis and Cannabinoid Research 6:2, 1–12, DOI: 10.1089/can.2020.0179.