Methodologies for Labeling Stem Cells 258
Imaging Modalities 258
Optical Imaging 260
Direct Labeling 260
Indirect Labeling 260
Application of Optical Imaging in the Kidney 260
Magnetic Resonance Imaging 262
Direct Labeling 262
Special Cell-labeling Approaches for Magnetic Resonance Imaging 262
Indirect Labeling: Magnetic Resonance Reporter Genes 263
Magnetic Resonance Stem Cell Imaging in the Kidney 263
Radionuclide Labeling (PET/SPECT) 265
Direct Labeling PET/SPECT Imaging 265
Indirect Labeling PET/SPECT Reporter Genes 265
Other Imaging Modalities 267
Multimodality Approach 267
Imaging of Native Stem Cells 267
Over the past decade, major strides have been made towards improvement of stem cell imaging techniques for better understanding of cell biology and tracking the fate of stem cells after transplantation. This chapter outlines some of the most important imaging techniques used for the detection and monitoring of progenitor and stem cells. This chapter focuses on imaging technologies including histology, optical imaging, magnetic resonance imaging and radionuclide imaging, which enable short-term or long-term monitoring of stem cell biology in vivo or in vitro. So far, no single technique can answer all the research questions and address all the needs. Indeed, in the future, the use of multimodality imaging tools may be found to be the most useful and instructive in this exciting and rapidly evolving field. Applying these techniques to clinical use is still challenging, but is likely to be resolved over the next few years.
Utilization of stem cells for therapeutic approaches is becoming an attractive alternative to conventional treatments, especially for diseases refractory to other treatments. The main objective of cell-based therapies is to repopulate the damaged tissue with functional cells, or to use the cells to administer therapeutic agents to the target tissue. Use of stem cells in animal models has been shown to regenerate renal tubular , mesangial and endothelial cells in kidney diseases such as acute kidney ischemia or glomerulopathy. Endothelial progenitor cells (EPCs) have been used in the stenotic kidney of a swine model of chronic renal artery stenosis and it was demonstrated that EPCs integrated into renal vascular and tubular structures and improved renal function . Stem cells have also been shown to regenerate different organ and systems, such as pancreas , joints , musculoskeletal system and components of the cardiovascular system , and have also been used as adjuvant treatment for malignancies . In addition to EPCs, a wide range of cell types, including mesenchymal stem cells (MSCs) , embryonic stem cells (ESCs) , bone marrow-derived hematopoietic stem cells (HSCs) and neural stem cells , have been applied in different systems.
A significant body of knowledge has been acquired with regard to the biology of stem cells and the potential benefit of their use in tissue and organ regeneration by ex vivo imaging (histology). However, such techniques cannot address questions regarding the biology of stem cells in living subjects, such as temporal and longitudinal patterns of cell location (due to migration), viability and functional status (interactions with tissue, differentiation, etc.) after delivery. Thus, the need for an imaging technique to track the biology and behavior of stem cells before and after transplantation becomes of paramount importance for the advancement of the field of cell-based regenerative medicine.
The capacity to study cell biology and cell survival has been limited in part owing to limitations in imaging technology. Traditional ex vivo histology/immunohistochemistry assays are useful for ex vivo cell biology study, but for stem cell tracking they are limited to biopsy specimens or to information obtained at the time of euthanasia of the animals. A non-invasive approach would permit a longitudinal study (in the same subject) of the biology of stem cells, at the same time minimizing the interference of biological variables (because the same subject can be used for multiple studies) . Furthermore, a non-invasive approach has the potential to disturb the microenvironment minimally, allowing a more physiological study. Moreover, such strategies have the potential to be translated for clinical use, and some non-invasive imaging has indeed been adapted to image stem cells in clinical trials.
In addition to histological technique, current available mainstream non-invasive image modalities for stem cell tracking include optical imaging (OI), magnetic resonance imaging (MRI) , radionuclide positron emission tomography (PET) and single-photon emission computed tomography (SPECT). These techniques not only provide structural and functional information after cell therapy, but may also be capable of tracking stem cells in vivo longitudinally, and following their migration and transdifferentiation. Figure 17.1 shows the potential role of imaging in cell-based therapy of renal diseases. This chapter will discuss using these modalities for stem cell imaging and their application in kidney diseases.
Methodologies for Labeling Stem Cells
To render stem cells visible under any imaging modality, proper labeling of cells is important. There are two major strategies for the labeling of stem cells for imaging: direct labeling and indirect labeling (reporter genes). In a direct labeling strategy, labeling agents are introduced into the cells by simple incubation with the tracer before transplantation, and stem cells are transplanted and then followed in vivo. In an indirect labeling system, protein-encoding genes can be transiently or stably introduced in the cell, thus becoming reporters of gene expression and enabling cell-tracking following transduction. Stem cells incorporated with reporter gene ex vivo are then transplanted into living subjects. These stem cells express a special protein (e.g. enzyme, cell surface receptor) that interacts with an exogenously given substrate or light excitation, resulting in signal which can be detected non-invasively in vivo or by histology.
Histology is a conventional tool to track stem cells mainly using microscopy and, because of its invasiveness, especially in animal research. Stem cells need to be labeled with tracers before transplantation, which enable them to be distinguished from host cells.
The most effective tracer of cell morphology is a fluorescent probe, which has the capacity for localized introduction into a cell or organelle, as well as long-term retention within that structure, and is biologically inert and non-toxic. Fluorescent tracers can be used to investigate the movements of labeled cells in culture, tissues or intact organisms. There are four major different types of fluorescent probes for stem cell tracking.
Membrane tracers : DiI, DiO, DiD, DiR, DiA and their analogs. Di dyes can be introduced into membranes by direct application of dye onto the cell by loading from culture media . Lateral diffusion of the dye within the membrane eventually stains the entire cell. These probes are widely used for stem cell tracking and long-term assays of cell–cell association . DiI dye has been applied in EPC labeling and tracking in the kidney. EPCs were incubated with CM-DiI for 30 min, then transplanted into the stenotic kidney of pigs with unilateral renal artery stenosis. Red fluorescence EPCs could be visualized in kidney sections as late as 4 weeks after transplantation ( Fig. 17.2 ). Furthermore, combining Di dye with other immunohistological staining, it was possible to track stem cell change in phenotype (transdifferentiation). For example, a histological section from a kidney injected with Di dye-labeled EPCs (fluorescent red) was stained with an antibody against the endothelial marker CD31 (labeled fluorescent green). The cells showing both red and green fluorescence (i.e. yellow) were mainly confined to perivascular capillary structures, indicating the ability of EPCs to transdifferentiate into mature vascular endothelial cells ( Fig. 17.2 ).
Fluorescent microspheres . Those microspheres contain approximately 10 2 to 10 10 fluorescent dyes per bead and are the most intensely fluorescent tracers available . Furthermore, they are often biologically inert and physically durable, which makes fluorescent beads particularly useful as long-term markers for transplantation studies. Submicrometer microspheres can be taken up by phagocytosis. This feature enabled EPCs to be labeled by incubation with microspheres in the culture medium, and then EPCs (yellow fluorescence) were tracked in the kidney 4 weeks after injection ( Fig. 17.2 ).
Cell-permeant cytoplasmic labels (thiol-reactive CellTracker probes) yield fluorescent products that are retained in many live cells through several generations and are not transferred to adjacent cells, except possibly by transport through gap junctions. These probes constitute excellent long-term tracers for transplanted cells or tissues .
Fluorescent dextran conjugates . Dextran conjugates are ideal cell lineage tracers because they are relatively inert, exhibit low toxicity and are retained in cells for long periods . However, they are membrane-impermeant probes, and usually need to be loaded into cells by invasive techniques such as microinjection, whole-cell patch clamping or scrape loading. This may limit the application of dextran conjugates in stem cell tracking.
Histological techniques are also used to track iron-labeled stem cells in the kidney in conjunction with MRI studies. Prussia blue staining is specific for iron, and comparing MRI images with histological sections is useful to confirm and determine the accuracy of MRI tracking.
For histology, cells can be labeled with a fluorescent reporter gene, and then when the tissue is excised, the labeled cells that remained in the tissue can be easily identified using histological methods . Similar to in vivo conditions, different reporter genes emit light in different spectra permitting detection of corresponding fluorophores in the sample. These applications play a very important role in the study of stem cell biology in cell cultures and ex vivo. Furthermore, owing to their light emission and detection characteristics, they are commonly used as a basic strategy when designing novel reporter gene constructs. A good example for reporter genes is green fluorescent protein (GFP) . Stem cells isolated from GFP transgenic animals can be allogenically transplanted into the host body; those cells consistently express GFP and can be visualized using fluorescence microscopy.
In summary, although histology is a traditional ex vivo imaging technique, which can only provide information obtained by invasive techniques, it is still a basic and powerful tool that allows accurate cell tracking, and provides cell fusion information and cell–tissue interaction, especially in animal studies.
The principle of OI is similar to that underlying the acquisition of a photograph. It uses a light that excites a fluorophore and a detector, such as a highly sensitive charge-coupled device (CCD), which captures the probe’s emission once resolved by an appropriate filter. The ability to image at single cell level, lack of radiation and low cost make this modality a basic imaging technique for stem cell tracking, especially in small animals.
This method employs an incident of light that excites a fluorophore and emitted light is captured by a CCD camera. So far, no single ideal optical contrast agent is available. In general, all OI contrast agents need to be biocompatible, possess a tolerable toxicity profile, be small in size and exude a bright signal.
Fluorescent dyes . These fluorochromes are continuously being designed (hundreds are commercially available) and include DiI, DiO, DiD and CM-DiI. They have proven effective for in vitro cell labeling (as discussed in the last section) and in vivo cell tracking with OI.
Targeted probes . Colloidal quantum dots (QDs) have a narrow emission and continuous broad absorption spectrum (i.e. broad excitation spectrum), which allows fluorescent excitation by any wavelength below the emission maximum.
One of the main advantages of fluorescent reporter genes is that different fluorophores (e.g. red or green fluorescent proteins) can be used to label different cell populations, permitting their concomitant imaging .
Bioluminescence . Bioluminescent imaging typically involves the luciferase (luc) gene and based on luc-mediated oxidation of D-luciferin (with emission spectra: 400–620 nm). A chemical process following intravenous administration of luciferin generates light emission. Photon generation takes place exclusively at the site of luciferase expression; therefore, the target-to-background signal ratio is extremely high, thus allowing for detection at lower wavelengths with a maximal penetration depth of 3 cm.
Fluorescent proteins . The first representative of this class of proteins was GFP, which has been commonly used for cell tracking with OI in small animal models.
Perhaps the most frequent application of fluorescent reporter genes in cell imaging is for ex vivo analysis, where they can be used for cell sorting. For example, not all stem cells transduced with a bioluminescent or PET reporter gene incorporate the gene of interest into their genome. A common method of sorting and selecting the cells that did incorporate the gene of interest takes advantage of fluorescent reporter gene cell markers (e.g. Gfp, Rfp) .
Application of Optical Imaging in the Kidney
Few studies have applied OI to track stem cells in the kidney. Tögel et al. tested bioluminescence to track MSCs in a mice acute kidney injury (AKI) model using the OI technique. Bone marrow-derived MSCs were infected with luciferase/neomycin phosphotransferase. The transinfection had no impact on MSC biology, as tested by a multidifferentiation assay. Transinfected MSCs (1 × 10 5 ) were injected into mice through the jugular vein or carotid artery. Immediately or 24 h after injection, animals were imaged using the Xenogen IVIS 100 system. MSCs showed distinct accumulation in the kidney only in AKI mice, mainly after the cells had been administered by intra-arterial injection. Importantly, intravenous injection of MSCs resulted in substantial accumulation in the lung ( Fig. 17.3 ), so that fewer cells reached the kidney. This study demonstrated that bioluminescence is sensitive for in vivo MSC tracking, and is suitable for non-invasive kidney localization of injected MSCs in small animals.
The fluorescence direct imaging and reporter gene approaches have limited tissue penetration (around 2 mm), restricting the use of these techniques to superficial tissues in small animals (e.g. mice) . Furthermore, most common fluorescence imaging devices do not have tomographic capabilities, which further limits the identification of deeper organs. Several studies are underway to attempt to provide tomographic fluorescence imaging .
There are also some other drawbacks to the use of fluorescent reporter genes. Similar to other imaging modalities, the detected signal constitutes only a small fraction of the emitted signal. Significant efforts are underway to use multiple cameras, which may enable detection of a higher fraction of the emitted signal . Novel developments, such as time-domain imaging , incorporate the time domain in the analysis, and have the potential to provide depth information of the fluorescent signal. However, such strategies are under development and not ready to be routinely applied by mainstream imagers. In addition, fluorescence techniques are associated with background autofluorescence, which necessitates a stronger signal to overcome the noise in the data. The presence of autofluorescence could be a significant disadvantage, especially when the emitted signal is not high, leading to a low signal-to-noise ratio.
Magnetic Resonance Imaging
MRI provides excellent soft tissue contrast with high resolution. Thanks to a large number of available MRI techniques to generate contrast in the images, a number of anatomical, physiological and metabolic variables can be measured almost simultaneously and under physiological conditions. Partly because of this versatility, MRI has evolved into one of the most powerful imaging tools in radiology and biomedical sciences. Furthermore, the high resolution of MRI enables its use for the visualization of single cells against a homogeneous background. The availability of responsive and targeted contrast agents extends applications of MRI from visualization of cell location to characterization of molecular and cellular signaling events and their functional status, such as enzyme or receptor expression.
To visualize cells against the background of host cells and to increase the likelihood and sensitivity of their detection with MRI, cells need to be labeled with specific MRI contrast agents. Four types of agents are currently available for MRI cell imaging, and selection of a specific agent depends on the purpose of imaging.
Iron oxide-based contrast agents. Superparamagnetic iron oxide particles (SPIOs) are the most commonly used strategy in MRI cell labeling. These highly magnetic particles induce changes in T2 relaxivity, which makes them detectable in vivo . In a direct labeling strategy, stem cells are prelabeled with SPIOs, before transplantation and imaging, so that the signal originating from the SPIOs is considered to reflect number of cells retained in the tissue.
Lanthanide chelates and Mn 2+ -based contrast agents. Paramagnetic compounds such as Mn 2+ ions or lanthanide chelates usually affect T2 less than SPIOs, but are commonly used for their increase in T1 relaxation, resulting in a hyperintense (bright) contrast in T1-weighted images. Manganese-based contrast agents such as manganese chloride (MnCl 2 ) or manganese oxide (MnO 2 ) nanoparticles have also been used for cell labeling. However, their toxicity prevents their widespread application.
Contrast agents based on heteronuclear MRI. The majority of MRI contrast agents used for clinical and experimental applications is based on modulating the contrast in 1 H-based MRI . Although relatively sensitive, this presents a major challenge for discriminating labeled cells from the background signal and motivates the search for a less abundant contrast molecule. Most heteronuclear approaches to MRI-based cell imaging have focused on 19 F-MRI and spectroscopy. One advantage of 19 F-MRI and spectroscopy is the ability to quantify the signal as long as the signal-to-noise ratio is sufficiently high. Other potential applications of heteronuclear MRI and spectroscopy include 13 C and 31 P.
Responsive contrast agents. Contrast agents that change contrast owing to their changed magnetic properties (e.g. relaxation, chemical shift) in response to dynamic changes in physiological, enzymatic and other metabolic properties are frequently referred to as responsive contrast agents . Contrast agents that are (chemically) modified by cellular changes have initially been developed for reporting on the physiological status and metabolic activity of cells. Most of these contrast agents are lanthanide chelates with one or more potential coordination sites for water that can be blocked in the inactive state of the contrast agent. Using these agents enables detection of the interaction between transplanted cells and host organs. A potential limitation is the remaining influence of the “inactive” contrast agent on T1 relaxation owing to the secondary coordination sphere of the water, which further decreases its sensitivity.
Special Cell-labeling Approaches for Magnetic Resonance Imaging
Useful approaches for stable labeling of cells with MRI contrast agents include incorporation of the contrast agent in vitro before transplantation and engraftment into the host, systemic application and subsequent specific internalization of the contrast agent by the targeted cells, receptor-specific binding or internalization of targeted contrast agent to targeted cells, or the generation and accumulation of contrast by genetically modified cells.
In vivo cell labeling. In vivo cell labeling applications include localized injection of iron oxide particles, which allows monitoring of the migration of endogenous stem cells. In a recent study , iron oxide particles were injected in the area of stem cell generation, and labeled cells that translocated along the rostral migratory stream to the target area, assumed to represent stem cells.
Targeted contrast agents for in vivo cell labeling. To achieve more specific labeling of particular cell types by systemic administration of a contrast agent, targeted agents are the most promising approach. Targeted contrast agents are chemically modified so that they accumulate specifically in certain tissue types, usually owing to the presence of particular cell types. Active accumulation occurs by specific binding or uptake due to biomarkers bound to the contrast agents, such as antibodies or small peptides. Antibodies bound to iron oxide particles or lanthanide chelates accumulate at sites of high expression of the targeted receptor. The most common approach is to use Tat peptides to derivatize either coated nanoparticles or lanthanide chelates. Biological applications of targeted contrast agents include inflammation, angiogenesis, apoptosis, tumors and atherosclerosis. These approaches are often hampered by insufficient accumulation of the contrast agent owing to either low cell densities or low receptor expression on the cell surface.
Indirect Labeling: Magnetic Resonance Reporter Genes
MRI-based reporter genes strategies are based on the production of intracellular metalloproteins . As previously described, iron is a paramagnetic substance that induces changes in relaxivity (i.e. T2 ∗ effect) that can be detected using specific imaging sequences, and the goal of these strategies is to accumulate large quantities of iron intracellularly for its detection. Two major metalloproteins have been used in MRI reporter gene techniques: ferritin and tyrosinase. Ferritin is a metalloprotein that functions as an iron depot and can contain up to 4000 iron atoms. For example, cells transduced with the ferritin reporter gene were delivered to the brain of mice . After several days, enough iron signal was “collected” inside the transplanted cells (due to the overexpression of ferritin) to enable the non-invasive monitoring of transplanted cells using MRI. In a recent study , the investigators used a replication-defective adenovirus vector to deliver the ferritin transgenes. Following focal inoculation of the vector into the mouse brain, they monitored the reporter activity using in vivo time-lapse MRI. They observed robust contrast in virus-transduced neurons and glia for several weeks ( Fig. 17.4 ). This technology is adaptable to monitor transgene expression in vivo in many tissue types and has numerous biomedical applications, such as visualizing preclinical therapeutic gene delivery.