Applications of Upconversion Nanoparticles in Imaging, Detection and Therapy

Lei Yin Ang; Meng Earn Lim; Li Ching Ong; Yong Zhang


Nanomedicine. 2011;6(7):1273-1288. 

In This Article

UCNs in Disease-related Detection & Sensing

The applications of UCNs are not limited to bioimaging. In the following section, the various UC luminescence-based detection and sensing methods are reviewed. For some of these examples, the submicron UC particles (~200–400 nm in diameter) are not really small enough to be classified as nanoparticles. Nevertheless, these studies are still valid in illustrating their capabilities in detection as, ultimately, it is the particles' luminescent property that is used.

Point-of-care Diagnostics

UC particles have been used as labels in point-of-care (POC) diagnostics, which revolves around the use of simple devices or assays that can give rapid results. The particles have helped to improve the techniques in aspects such as detection limits, signal outputs, shelf life and robustness (Table 4).[60–62] The various UC particles-based strategies can be broadly classified into three groups depending on the analyte types. Within each, the particle or device modifications, detection strategies and their applications will be discussed.

Nucleic Acid-based Detection Nucleic acid-based detection is applicable when a disease manifests by sequence changes. Table 5 & Table 6 summarize the UC particle-based strategies. In all techniques, the particles tag the double-stranded DNA complexes formed between the disease sequences and the probe DNA(s). These assays can be either in solution format or within devices. In the solution format, DNA-modified UC particles are often used in conjunction with another acceptor dye such as TAMRA,[63] SYBR Green I[64] or Alexa Fluor.[65] In the presence of disease sequences, luminescence resonance energy transfer (LRET) occurs. This is observed as an increase in acceptor emission and a corresponding decrease in particle emission upon NIR excitation. Currently, there are two slight variations of using LRET for disease detection (Table 5). In the first case, the acceptor dye is modified with another probe DNA. Hence, the disease sequences will be labeled with both particles and dyes.[63,65] The acceptor dye can also be 'label-free', meaning it is used without modification. A fluorescent DNA binding dye, such as SYBR Green, is used. The quantum yield of the dye is enhanced upon binding to double-stranded DNAs. Hence, significant LRET only occurs when the double-stranded DNAs are formed in the presence of disease sequences.[64] These solution-based assays were applied in the detection of genetic diseases such as sickle cell anemia[63,64] and ankylosing spondylitis[65] with low detection limits of nanomolar concentrations.

Upconversion particles have also been applied as labels in devices such as lateral flow test (LFT) strip and microarrays. Typically, in a LFT system, a sandwich assay format is used when analyzing nucleic acids. The labeled target nucleic acids are captured on the test line while the control line binds the excess labels. The capturing of these targets on the test line can either be mediated by antibodies (nucleic acid lateral flow immunoassay [NALFIA]) or other binding moieties (nucleic acid lateral flow test [NALFT] strip) (Table 6). PCR often precedes the detection, then, in the presence of disease sequences, double-stranded amplicons are generated.[66] In the examples involving UC particles, the amplicons can bind to both particles and test lines via the modifications on the forward and reverse primers. As such, the modifications of the particles and strips will be different depending on the primers (The UC-related schemes are summarized in Table 6). For NALFIA, the primers are tagged with avidin and digoxigenin. Modifications on the primers allow binding of the amplicons to the test line and UC particles. Biotin-avidin interactions link the amplicons to the biotin-coated particles while digoxigenin binds the labeled amplicons to the antidigoxigenin antibodies on the test line. The readout is based on the presence of the particles' luminescence on the test and control lines upon NIR excitation. Such a setup had been combined with microfluidics in a cassette for the detection of a food-borne pathogen, Bacillus cereus.[67] Conversely, in a UC-based NALFT setup, biotin binds the amplicons to the test lines and the antibody-antigen interactions link the amplicons and the antidigoxigenin antibody-modified UC particles. Similarly, this setup has been combined with microfluidics to make cassettes capable of detecting B. cereus[68,69] and HIV, the causative agent for AIDS.[68]

In a typical microarray setup, the probe DNAs specific to the disease sequences are immobilized on a solid surface (Table 6). Here, PCR also precedes the detection on the microarray but the amplicons are typically functionalized with only one binding group, such as biotin. In the presence of the disease sequences, the resulting amplicons will hybridize to probe DNAs on the chip surface. Since these amplicons have the biotin tag, they are labeled by the streptavidin-functionalized particles. Hence, particles' luminescence on the microarrays upon NIR excitation will indicate the presence of disease sequences. The disease detection ability of such a microarray-based platform using 30–50 nm UCNs was demonstrated using a model pathogen, Escherichia coli.[70]

Protein-based Detection & Direct Detection of PathogensTable 7 summarizes representative UC particle-based detection strategies for protein markers and whole pathogens. They are typically based on the LFT system and involve antibody-antigen interactions for target recognition and labeling. For majority of the analytes (i.e., the first case in Table 7) two antibodies are used and they will form a sandwich complex with the target. The antibodies on the particles direct the binding to the target, forming particles-target complexes, which are captured by another antibody on the test line. The luminescence signals on the test and control lines upon NIR excitation are then read and interpreted. Such a setup has proved to be useful in clinically relevant examples of brucellosis,[71] asymptomatic leprosy,[72] plague,[73] respiratory syncytial virus infection[74] and schistosomiasis.[75] For most of these, the antibodies were directly immobilized onto the test line. The only exception is the asymptomatic leprosy example in which the test line was decorated with avidin. Hence, the capture on the test line was via avidin-biotin interactions.[72]

Similar antibody-antigen interactions can be used to detect antibodies generated by the host immune response. As seen from Table 7, the target antibody is labeled by the particles modified with their specific antigens or a general antibody-binding protein such as protein A. Immobilized antigens on the test line mediates target capturing. Here, the target antibody serves as a linker between the UC particles and the test line. Quantitative detection of hepatitis B surface antibodies has been achieved in both the laboratory and clinical settings. This platform has also proved its detection capability with its high sensitivity and good detection agreement.[76] Moreover, detection of HIV antibody has been accomplished by the use of protein A-conjugated UC particles in a LFT platform.[69] In another case, these generic antibody-binding UC particles allowed simultaneous detection of immune response antibodies against two disease pairs, HIV/TB and HIV/HCV, illustrating their multiplexing capability.[77]

Other Disease-related Sensing & Monitoring

Other than detecting the explicit disease markers (as discussed above), UC particles, more specifically UCNs, have also been applied in the detection of other disease- or human health-related analytes. Novel UCN-based sensors for the monitoring and sensing of key health parameters such as CO2[78] and pH[79] have been developed. Furthermore, mercuric ions (Hg2+), a type of water pollutant, have also been detected by such sensors. Here, the general strategy is to modify the setup to enable detectable changes in the particles' luminescence upon target recognition. LRET and the inner filter effect are the two different approaches that were used to achieve these changes. For the LRET approach, the 'label-free' detection scheme for nucleic acid-based detection (Table 5) was used to detect Hg2+. The probe DNA was designed such that a hairpin structure would be formed in the presence of Hg2+ due to thymine-Hg2+-thymine interactions. The long stem of this hairpin structure would provide the double-stranded DNA environment for LRET to occur. This Hg2+ sensor was shown to be selective and highly sensitive, having a detection limit of 0.06 nM in an aqueous buffer.[80] In another strategy, the inner filter effect generates observable luminescence intensity changes of the particles in the presence of targets.[78,79] For instance, a UCN-based sensor for CO2 makes use of a pH probe, bromothymol blue (BTB). In the presence of CO2, the color of BTB changes from blue to yellow, thereby exerting an effect on the observed emission color from the UCNs. When incorporating UCNs and the sensor cocktail with BTB into a polystyrene film, the changes in particles' luminescence gave an indication of the CO2 concentration.[78] A similar principle to BTB's color change was also used in another optical sensor made up of UC nanorods to monitor pH changes directly.[79]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.