Crossing the Blood–Brain Barrier: Molecular Imaging Illuminates Transplanted Stem Cells in the Brain

Abstract

Stem cell therapy offers a unique opportunity to repair the damaged brain in neurological conditions such as stroke, Parkinson's disease, Alzheimer's disease, and traumatic brain injury [1]. This strategy of replacing lost neurons and supporting surviving tissue with transplanted cells is compelling because of its potential to reconstruct both the structure and function of the injured brain. Preclinical studies and early-phase clinical trials have yielded encouraging results, demonstrating functional improvements and evidence of graft survival in animal models and initial human cohorts [2]. However, translating these promising findings into reliable and approved therapies for clinical practice demands a deeper understanding of the in vivo processes that transplanted cells undergo after entering the brain.

Molecular imaging represents an important strategy for visualizing transplanted stem cells in vivo. By transferring cell behaviors into imaging data, it may solve the key issues in stem cell therapy, including cell fate tracking, functional recovery evaluation, and safety monitoring [3]. Benefitting from different modalities including optical imaging, magnetic resonance imaging, positron emission tomography (PET), and hybrid multimodal imaging, imaging agent labeling strategies contribute to the in vivo tracking of grafted stem cells within the brain. However, the blood–brain barrier (BBB) severely restricts the application of existing molecular imaging methods for the visualization of transplanted stem cells within the brain. The BBB protects the brain but also prevents the entry of most contrast agents, radiotracers, and reporter probes designed to track transplanted stem cells in the brain parenchyma [4]. Consequently, once stem cells are transplanted into the brain, they cannot be tracked, which prevents any observations of their in vivo behavioral processes. To advance stem cell therapy for neurological diseases, there is an urgent need to develop new molecular imaging tools or strategies that can cross the BBB to visualize transplanted stem cells in the brain.

Direct in vitro labeling of stem cells prior to transplantation offers a straightforward way to avoid BBB issues. Imaging agents, commonly including superparamagnetic iron oxide nanoparticles for magnetic resonance imaging [5], radionuclides such as 2-deoxy-2-[¹⁸F]fluoro-D-glucose for PET [6], and organic fluorophores for optical imaging, can be loaded before cell transplantation, with no need to cross the BBB. However, several inherent shortcomings of these techniques limit their reliability for long-term in vivo tracking. For example, the intracellular label is continuously diluted with each round of cell division, eventually falling below the detection threshold. Critically, when labeled cells die, the imaging agent (particularly superparamagnetic iron oxide) is often retained at the site of death, generating persistent false-positive signals that mimic the surviving graft. Additionally, radionuclide labels undergo physical decay that imposes a short detection window, thereby compounding the dilution problem. These limitations suggest that simply labeling cells outside the BBB is insufficient and motivate a shift toward indirect reporter gene systems that produce signals in living cells only and delivery of corresponding imaging probes across the BBB.

The temporary and regional regulation of BBB permeability through physical or pharmacological approaches may be an effective strategy for achieving the intracranial delivery of molecular imaging tracers. A recent report has established an advanced strategy for monitoring transplanted stem cells in the brain after ischemic stroke [7]. A clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated nuclease 9 (Cas9)–engineered triple-fusion reporter gene imaging system combined with a noninvasive adenosine agonistic micelle–based probe delivery strategy was introduced. Beyond green fluorescent protein for fluorescence imaging and luciferase for bioluminescence, a reporter gene expressing herpes simplex virus type 1 thymidine kinase was adopted to visualize transplanted stem cells using PET, paired with 9-(4-[¹⁸F]-fluoro-3-hydroxymethylbutyl)guanine ([¹⁸F]FHBG). Conventionally, [¹⁸F]FHBG has poor BBB penetration and was previously only used for imaging transplanted cells in the treatment of extracranial diseases [8]. This limitation is solved by the application of an adenosine A2A receptor–targeted agonistic micelle. It can target blood vessels in ischemic areas, thus reversibly and transiently opening the BBB and increasing the accumulation of PET tracer in the brain. By combining bioluminescence and [¹⁸F]FHBG PET with this delivery strategy, transplanted stem cells in the brain have been dynamically monitored for at least 8 weeks, achieving in vivo imaging with high sensitivity, specificity, and quantifiability. Additionally, focused ultrasound—particularly combined with microbubble technology—has been demonstrated to reversibly open the BBB in preclinical studies and clinical trials [9], thereby allowing molecular imaging tracers that were previously inaccessible to the brain to enter specific brain regions. Studies of non-human primates and patients with Parkinson's disease have confirmed that focused ultrasound can open the BBB; increased [¹⁸F]choline uptake was observed only in regions in which BBB opening occurred [10]. With the help of physical or pharmacological approaches to open the BBB and deliver imaging agents, transplanted stem cells within the brain can be successfully observed using molecular imaging.

Although this strategy can be used to cross the BBB and visualize transplanted stem cells in the brain, challenges remain for its clinical translation. Non-human reporter genes may trigger immune rejection in the human body, thus affecting cell survival and imaging authenticity; however, this can be avoided by using a humanized reporter gene or endogenous sequences [11]. This system also faces important ethical and biosafety concerns regarding the use of gene editing and tracers in clinical practice, meaning that unlabeled imaging technology without genetic modification and imaging tracers may be a valuable research direction. Moreover, it is worth noting that opening the BBB may result in the transient exposure of brain tissue to plasma proteins and peripheral immune components, which may have long-term effects on the survival and functional integration of transplanted stem cells.

Utilizing the transport system of the BBB itself to achieve the physiological internalization of imaging tracers is a possible alternative for visualizing transplanted stem cells in the brain. It avoids damage to the structural integrity of BBB, thus maintaining homeostasis of the brain microenvironment. By conjugating imaging tracers with receptor ligands of endogenous molecular channels within the BBB, tracers can be delivered into the brain through receptor-mediated transcytosis (RMT) [12]. Transferrin receptor (TfR) is a rapidly progressing target in RMT delivery research in recent years, and the delivery of imaging agents into the brain has been achieved by conjugating them with TfR-binding antibody or fragments [13]. Low-density lipoprotein receptor associated protein 1 (LRP1) and glucose transporter 1 (GLUT1) have been also used to deliver imaging agents via RMT. Imaging agents modified with angiopep-2 can bind to LRP1 with high affinity and specificity, and efficiently cross the BBB through RMT with superior brain entry efficiency to TfR [14]. GLUT1 is highly expressed in brain microvascular endothelial cells, thereby allowing glycosylation-modified imaging agents to be transported into the brain. A recent study used GLUT1 as the specific target to deliver iron oxide nanoparticles functionalized with glucuronic acid into the brain for magnetic resonance imaging [15]. However, the delivery efficiency of this approach requires improvements to increase the signal-to-noise ratio, which is largely limited by receptor expression abundance and endosome escape efficiency. Furthermore, lipophilic small molecule imaging tracers can directly cross the BBB because they can penetrate the lipid membranes of brain endothelial cells through passive diffusion [16]. However, these tracers are affected by molecular size, efflux pump, and metabolic stability. The development of optimized tracer structures, prodrugs, and activation strategies is likely to help solve the current limitations of small molecule imaging tracers. The use of physiological transport routes or small molecule tracers can preserve BBB integrity while enabling the entry of imaging agents to visualize transplanted stem cells.

Guided by these approaches for crossing the BBB to deliver imaging agents, the in vivo processes of transplanted stem cells to treat neurological conditions can be successfully visualized and monitored. In ischemic stroke models, [¹⁸F]FHBG PET with A2A receptor–targeted agonistic micelle can monitor the survival, proliferation, and migration of transplanted stem cells in vivo [7]. In clinical trials of treatments for Parkinson's disease, transplanted dopaminergic progenitors can be successfully imaged using [¹⁸F]fluorine-L-dihydroxyphenylalanine and dopamine transporter PET for both the monitoring of grafts and the evaluation of therapeutic efficacy [2, 17]. These imaging agents directly cross the BBB for delivery. In stem cell–based therapies for Alzheimer's disease beyond conventional direct labeling with superparamagnetic iron oxide, neural stem cells can be directly delivered to clear beta-amyloid by regulating the BBB using combined focused ultrasound and microbubbles [18]; these can then be visualized by adding reporter gene systems or delivering imaging agents.

In summary, crossing the BBB is the first step toward visualizing transplanted stem cells in the brain using molecular imaging (Figure 1). In the future, we should focus on achieving safe long-term visualization through clinically compatible imaging technologies, humanized reporter systems, and precise BBB-crossing strategies that preserve brain homeostasis. Molecular imaging transforms transplanted stem cells in the brain from invisible to visible and quantifiable, thereby accelerating the translation of stem cell therapy for neurological diseases.

Peili Cen: writing – original draft. Yan Zhong: writing – review and editing. Rui Zhou: writing – review and editing. Jing Wang: writing – review and editing. Han Jiang: visualization. Mei Tian: writing – review and editing, supervision. Hong Zhang: writing – review and editing, supervision.

This work was supported by the National Key Research and Development Program of China (Grant 2021YFA1101700), National Natural Science Foundation of China (NSFC) (Grant 82302267, 82302263), the Fundamental Research Funds for the Central Universities (Grant 226-2024-00059), and Joint Funds for the innovation of Science and Technology, Fujian province (Grant 2023Y9143).

The authors have nothing to report.

The authors have nothing to report.

Prof. Mei Tian and Prof. Hong Zhang are members of the iRADIOLOGY Editorial Board. To minimize bias, they were excluded from all editorial decision-making related to the acceptance of this article for publication.

Data sharing is not applicable to this article as no datasets were generated or analyzed.