Various molecular imaging techniques such as magnetic resonance imaging (MRI), positron emission tomography (PET), single photon emission computed tomography (SPECT), optical imaging (OI) and ultrasound (US) play a crucial role in the individualized diagnosis and treatment of diseases (Weissleder and Pittet, 2008). These molecular imaging techniques have been applied to evaluate specific molecular targets and visualize the internal structure of the human body. They have also been applied to the non-invasive study of biological processes in vivo at the cellular and molecular levels and play a key role in the diagnosis of diseases, patient management, and healthcare. However, among all the current molecular imaging techniques, there is not a single modality that can perfectly provide all the information needed. For example, optical fluorescence imaging is difficult to quantify, and has limited tissue penetration in vivo. MRI has high resolution but low sensitivity, while PET offers very high sensitivity but relatively poor resolution (Table 1) (Massoud and Gambhir, 2003; Gleich and Weizenecker, 2005; Alphandéry, 2019; Lu et al., 2021). Therefore, multimodal contrast agents and probes have been developed to solve this problem. Multimodal imaging is the combination of two or more imaging technologies, combining the advantages of different imaging modalities, while minimizing the disadvantages of those technologies. These contrast agents and probes make it possible to visualize, quantify, and trace the molecular processes. They can also detect abnormalities in the human body, obtain new information about some diseases, and achieving the effect of “1 + 1>2” to optimize diagnosis and treatment of diseases. For example, contrast agent and probes can be used to guide the scalpel during surgery (by fluorescence imaging), ensuring that all cancerous materials have been removed (by MRI), and tracking and identifying tumor cells and physiological processes (by PET or SPECT imaging). However, the synthesis of contrast agent and probes poses a huge challenge (Jennings and Long, 2009; Zhou and Brahme, 2010; Misri et al., 2012). Nanoparticles (NPs) have been extensively used as contrast agents for molecular imaging due to their potential in combining multimodal imaging, drug delivery, and targeted therapy into a single entity (Table 2 summarized the current application of advanced nanomaterials for multimodal imaging) (Yuan et al., 2021).

The application of MRI contrast agents based on magnetic iron oxide nanoparticles began in the 1990s for the clinical diagnosis of liver tumors (Stark et al., 1988). Currently, commercially available products include Feridex (superparamagnetic iron oxide [SPIO]; liver injury imaging), Gastromark (SPIO, gastrointestinal imaging) and Combidex (ultrasmall SPIO [USPIO]; lymphography) (Gao and Yang, 2009). The results of clinical application of these products demonstrate the excellent magnetic imaging performance and in vivo safety of magnetic iron oxide nanoparticles in MRI diagnosis. Thus, magnetic iron oxide nanoparticles as MRI contrast agents are a current domestic and international research hotpot. In recent years, a new generation of SPIO nanoparticles (SPIONs) as MRI contrast agents with complex modified structures and functions haven been developed along with the continuous development of nanoparticle preparation technology (Xing et al., 2014; Ajinkya et al., 2020). However, in vivo targeted imaging applications place high demands on the physicochemical properties of nanoparticles. SPIONs modified by branched polyethyleneimine (BPEI) exhibit good physicochemical properties and can provide functionalization platforms for further chemical modifications for targeting, drug delivering and other functions (Bao et al., 2014; Molaei et al., 2021).

Here, we briefly summarize the application of SPIONs in multimodal imaging, the synthetic approach to BPEI-modified iron oxide nanoparticles, and the prospects and challenges for their application use in multimodal imaging by searching in the PubMed, Web of Science databases based on the keywords “iron oxide, nanoparticles, polyethyleneimine, multimodal, MRI, etc.,”.

MRI is noninvasive, safe, and radiation-free modality and has a high-spatial resolution. Its applications in molecular and cellular imaging are growing rapidly and plays an essential role in diagnosing and staging of tumors. It is not restricted by the penetration depth of the signal, has no ionizing radiation, and exhibits high soft tissue resolution and wide range of clinical applicability. These advantages have made MRI an important imaging technique of clinical tumors diagnosis (Terreno et al., 2010). Since the approval of the first clinical magnetic resonance (MR) contrast agent, namely, Magnevist (Gd-DTPA), by the U.S. Food Surveillance Administration in 1998 up to the present (Boros et al., 2015), the most widely used agent clinically remains to be chelates based on the metal gadolinium (Gd³⁺). Particularly, the demand for gadolinium-based MR contrast agents has been increasing in recent decades, with growing concerns about their safety (Shen et al., 2018). Intravenous administration of gadolinium-based contrast agents (GBCAs) is used due to their ability to reduce T1 and T2 relaxation time. GBCAs are mainly excreted through glomerular filtration, with an excretion half-life of 90 min. Therefore, the potential toxicity of GBCAs is directly related to renal disease. In patients with chronic renal failure, the excretion half-life of GBCAs may be significantly prolonged to 24 h or even longer, which may lead to the retention of GBCAs in the body (Malikova and Holesta, 2017). Studies of the potential clinical implications of gadolinium retention have focused on neurologic and cognitive effects (Habermeyer et al., 2020; Solmaz et al., 2021), as gadolinium can cross the blood-brain barrier and deposit in brain tissue, particularly in the dentate nucleus and basal ganglia (Kanda et al., 2014; McDonald et al., 2015), together with the neurotoxicity of free gadolinium (Rogosnitzky and Branch, 2016). Moreover, intravenous administration of large amounts of GBCAs can result in extensive multiorgan deposition. In a study by Robert J. McDonald et al., healthy rats received 20 intravenous injections of 2.5 mmol gadolinium per kilogram (gadolinium-exposed group) or saline (control group) over a 26-day period. Their results demonstrated that the application of macrocyclic gadolinium chelates instead of linear chelates could reduce the deposition but could not eliminate it (McDonald et al., 2017). Therefore, the development of safe and efficient new contrast agents is of great significance and value. Inorganic nanomaterials, especially magnetic nanoparticles have been extensively studied and applied in the biomedical field (Lutz et al., 2006; Moradi Khaniabadi et al., 2017; Nosrati et al., 2019).

SPIONs are often used as MRI imaging probes in molecular imaging, these particles effectively shorten the T2 of water protons, particularly T2^* (Laurent et al., 2008). The mechanism of contrast generation is related to the magnetic properties of nanoparticles, which exert a strong magnetic induction effect on the water protons diffusing around the particles. The relaxation and pharmacological properties of SPIOs are mainly controlled by their size. SPIONs usually consist of a core of magnetite (Fe₃O₄) and γ-magnetohematite (Fe₂O₃) crystals. A study by Ajay Kumar Gupta et al. concluded that magnetic nanoparticles of 10–100 nm have the best stability and magnetization strength (Gupta and Gupta, 2005). The cores are coated with suitable materials and have total diameters ranging from approximately 60nm–250 nm. Small particles (diameters ranging from 20–50 nm), defined as USPIOs are characterized by a low r2/r1 ratio. Moreover, micrometer-sized particles (micrometer SPIOs) are useful in cell labeling and vascular targeting (Shapiro et al., 2005; Shapiro et al., 2006). Currently, SPIOs are widely used in varieties biomedical applications, such as cell separation (Liu et al., 2008), drug and gene delivery (Pan et al., 2007; Chen et al., 2009; Chen et al., 2010), multimodal imaging photothermal therapy (Cheng et al., 2011), and MRI (Corti et al., 2008; Zhang et al., 2015). For successful biomedical applications, Fe₃O₄ NPs are usually required to have good colloidal stability, low nonspecific phagocytosis by the reticuloendothelial system (RES), and active targeting specificity after targeting ligand functionalization. Hence, the surface modifications of Fe₃O₄ NPs with hydrophilic and biocompatible polymers are effective and important strategies. Various materials and polymers such as albumin (Berry et al., 2003), dextran (Berry et al., 2004), dendrimers (Shi et al., 2007), polyethylene glycol (PEG) (Kohler et al., 2006), and polyethyleneimine (PEI) (Chertok et al., 2010) are coated on the surface of Fe₃O₄ NPs to improve their stability and reduce the clearance by RES.

However, some studies have shown that SPIONs still have many limitations in molecular imaging, such as biocompatibility. Although SPIONs have been widely used in biomedical fields, their biocompatibility is still a problem that needs to be solved. Some studies have also demonstrated that SPIONs may cause toxic reactions in cells, which may impair the normal functions of the cells. Moreover, the stability of SPIONs in organisms is also another issue that requires attention. Under certain conditions, SPIONs may undergo oxidation, aggregation, or decomposition, which may not only hamper their imaging or drug delivery efficacy but may also cause negative implications on organisms. The metabolism of SPIONs in organisms remains another challenge. If SPIONs cannot been efficiently cleared from the organisms, their accumulation in tissues or organs may result in potential health risks. Finally, the preparation of SPIONs is often complicated. It requires precise control of reaction conditions to obtain the desired particle size and shape. Furthermore, surface modifications are also complex and necessary to improve their biocompatibility and functionality (Mahmoudi et al., 2011; Wu et al., 2015; Zhu et al., 2018).

Therefore, researchers have developed a series of surface engineering strategies to modify and functionalize the SPION surface with organic or inorganic materials, such as polymers, biomacromolecules, silica, and metals (Wu et al., 2008; Laurent and Mahmoudi, 2011; Wahajuddin and Arora, 2012). Table 3 summarizes the organic macromolecules which have been used for the iron oxide NPs functionalization and their advantages. Dendrimers are monodispersed polymers characterized by a dendritic structure consisting of oligomers that are repeated and linearly connected through branching units. Dendrimers form macromolecules with dendritic structures through repeated growth and branching, and the degree of branching expands as the number of polymerization generations increases, eventually forming closed three-dimensional spherical structures with embedded cavity structures, surfaces enriched with reactive functional groups, and controllable physicochemical properties. The application of dendrimers for the modification of iron oxide nanoparticles is a novel nanomaterial preparation strategy widely used in the pharmaceutical industry (Sharma et al., 2017). BPEI is a polymer with good water solubility and thermal stability. The branched chains of BPEI are rich in amino groups and have certain internal hydrophobic cavity structures, which provide the potential of multifunctional modification on the surfaces (Li et al., 2016).

BPEI is a cationic polymer with excellent water solubility and abundant imino and amine groups, and has been used in a wide range of applications (Jager et al., 2012); in particular, BPEI has been used as a modifier for the preparation of composites (Liu et al., 2014) and as delivery vehicles for biomedicine, drug delivery and gene transfection (An and Gao, 2007; Bao et al., 2014). BPEI modification can greatly improve the dispersibility of magnetic nanomaterials Fe₃O₄ superparamagnetic nanoparticles, while the surface amine groups enable their conjugations to ligands, antibodies, and drugs. However, the amine groups of BPEI or BPEI-modified nanoparticles may result in severe cytotoxicity and nonspecific cell membrane binding, which may cause undesirable consequence for biological applications. Hence, surface amine group neutralization has been employed to reduce positive charge on surfaces and overcome the above disadvantages. For example, the primary amine groups of BPEI-modified multicarbon nanotubes can be acetylated or carboxylated, which can significantly improve the biocompatibility of those nanotubes (Wen et al., 2013). Table 4 summarizes the advantages of BPEI-modified iron oxide nanomaterials.

The synthesis methods of BPEI-modified SPIONs include electrostatic adsorption, Covalent binding, Ligand exchange, Hydrothermal method, Photochemistry synthesis and other methods. We summarized the chemical and physical properties as well as applications of different types of BPEI -modified SPIONs synthesis methods (Table 5).

Multimodal imaging is a method that combines multiple imaging techniques to provide much more comprehensive and accurate information than single modality. BPEI-modified SPIONs have been used as highly efficient MRI contrast agents, which significantly improve the imaging resolution and contrast. BPEI-modified SPIONs are capable of accommodating different imaging modalities, such as MRI, fluorescence imaging, and photoacoustic imaging, by adjusting their surface properties and structure. While those kinds of modalities have a wide range of prospects for multimodal imaging applications. Apart from the applications in imaging, BPEI-modified SPIONs also demonstrated great potentials of targeted delivery and imaging of specific lesions by conjugating to specific targeting molecules, and these applications greatly facilitate the identification and localization of lesion areas. BPEI modification significantly improves the stability, biocompatibility and biosafety of SPIONs, and reduce the aggregation and clearance in vivo. SPIONs exhibit great prospects for applications in medical diagnosis and therapy.

However, challenges and limitations also exist with BPEI-modified SPIONs in multimodal imaging, such as imaging effectiveness, targeting and specificity, stability, and cost-effectiveness. In particular, the design and synthesis of molecular probes with excellent imaging performance are essential challenges. Moreover, the targeting and specificity of nanoparticles in multimodal imaging are vital for precise imaging. Studies have been conducted to achieve localized imaging of specific tissues and cells by introducing specific targeting molecules on the surface of BPEI-modified SPIONs.

Furthermore, the stability of nanomaterials is crucial for long-time imaging and storage. Various strategies have been proposed to improve the stability of BPEI-modified SPIONs, such as the synthesis of stable core-shell structures and the introduction of cross-linking agents. Finally, the cost of nanoparticles’ synthesis and application is also an important concern. Efforts have been conducted to improve the cost-effectiveness of BPEI-modified SPIONs through improving the synthesis methods.

In conclusion, BPEI-modified SPIONs have demonstrated promising applications in multimodal imaging, but further research and improvements are still needed to overcome the existing challenges and limitations, improve the imaging efficacy, targeting, biocompatibility and stability of the nanoparticles through continuously optimization of synthesis methods and surface modification strategies. Through persistent efforts, accurate and reliable multimodal imaging using BPEI-modified SPIONs can be achieved in the future.