Robust detection reagents are of increasing necessity for developing new malaria diagnostic tools. An iridium(III) probe was designed that emits long-lasting luminescent signal in the presence of a histidine-rich malarial protein biomarker. Detection of the protein either in solution or immobilized on a magnetic particle affords flexibility in application.
This work outlines the synthesis of a non-emissive, cyclometalated Ir(III) complex, Ir(ppy)2(H2O)2+ (Ir1), which elicits a rapid, long-lived phosphorescent signal when coordinated to a histidine-containing protein immobilized on the surface of a magnetic particle. Synthesis of Ir1, in high yields,is complete O/N and involves splitting of the parent cyclometalated Ir(III) chloro-bridged dimer into two equivalents of the solvated complex. To confirm specificity, several amino acids were probed for coordination activity when added to the synthesized probe, and only histidine elicited a signal response. Using BNT-II, a branched peptide mimic of the malarial biomarker Histidine Rich Protein II (pfHRP-II), the iridium probe was validated as a tool for HRP-II detection. Quenching effects were noted in the BNT-II/Ir1 titration when compared to L-Histidine/Ir1, but these were attributed to steric hindrance and triplet state quenching. Biolayer interferometry was used to determine real-time kinetics of interaction of Ir1 with BNT-II. Once the system was optimized, the limit of detection of rcHRP-II using the probe was found to be 12.8 nM in solution. When this protein was immobilized on the surface of a 50 µm magnetic agarose particle, the limit of detection was 14.5 nM. The robust signal response of this inorganic probe, as well as its flexibility of use in solution or immobilized on a surface, can lend itself toward a variety of applications, from diagnostic use to imaging.
Colorimetric and fluorescent labeling is an important method for the detection and tracking of biochemical molecules and processes1,2. The most common fluorescent markers are low molecular weight organic dyes3,4, but these molecules do not always have ideal optical properties. Fluorescent dyes are prone to photobleaching, often have small Stokes shifts, and may have overlapping excitation and emission spectra. Colorimetric labeling is often achieved by the use of enzymatic labels, which have an amplified signal useful in immunoassay quantification5,6. These enzymes also have their drawbacks, including photosensitivity, reaction conditions, and short substrate shelf life. These properties tend to require immunoassays and protein labeling methods to be done under well-controlled conditions using costly reagents.
Emissive transition-metal complexes have been explored as an alternative labeling approach for biochemical detection. In particular, cyclometalated Ir(III) has been studied in the context of organic light emitting diodes (OLEDs)7-9 oxygen sensing10, catalysis11, and protein/cell staining12-14. High photostability and quantum efficiency make this class of probes a good candidate for biomolecule detection15,16. It was previously found that cyclometalated Ir(III) complexes, of the form [Ir(C^N)2(solv)2]+, irreversibly bind histidine and elicit a blue-green signal response12,17. These complexes are non-emissive in the solvento state, but when histidine displaces the solvent molecules and binds to the metal center, they release an intense phosphorescent signal after long-wave UV irradiation. This signal only occurs after ligand substitution, and is the result of a triplet state electron relaxing to the ground state through the activation of metal ligand charge transfer (3MLCT) and ligand centered transfer (3LC) pathways8,15. These complexes can potentially be used as detection probes for histidine-rich proteins.
Histidine-rich proteins and their regulation levels are important in many diseases, including liver cirrhosis, cancer, and thrombic disorders.18Plasmodium falciparum Histidine Rich Protein II (pfHRP-II) in particular is a well-validated biomarker for malaria parasite infection. This protein is 67 kDa and contains 34% histidine, mostly within characteristic AHHAHHAAD repeating motifs19. These histidine repeats can bind free metal ions19 and heme complexes20 in host blood. pfHRP-II is commonly detected in low-resource settings using immunochromatographic rapid diagnostic tests (RDTs), but these tests are often inaccurate due to sample conditions, low biomarker concentration, poor manufacturing standards, and antibody degradation21.
Metal-based phosphorescent probes such as the cyclometalated Ir(III) complexes described above are attractive options for the detection of pfHRP-II due to their selective binding of histidine and their stabile and efficient emission properties. In this paper, the use of [Ir(ppy)2(H2O)2]+ (Ir1) to detect BNT-II, a branched peptide mimic of pfHRP-II is explored22. Kinetics of this interaction were monitored in real-time using biolayer interferometry techniques. The assay was also adapted to an on-bead ELISA-type format, where nanomolar limits of detection were achieved. This assay holds advantages over traditional ELISAs because it can be performed in under 2 hr with antibody-free reagents, instead of the 4-5 hr and biological reagents required with typical ELISAs.
1. Synthesis of Iridium(III) Complex
2. Interactions of Ir1 with Various Amino Acids
3. Titration of Ir1 with BNT-II
4. Real-time Kinetic Analysis of the Ir1/BNT-II System Using Biolayer Interferometry
5. On-bead Detection of BNT-II and rcHRP-II with Ir1
As shown in Figure 1, synthesis of the Ir1 solvento-complex involved splitting of the chloride bridge in the parent dimer by precipitation of the chloride in the form of insoluble AgCl and association of water molecules to the metal center. The formation of Ir1 was confirmed by H1NMR and ESI. Additionally, UV-visible bands characteristic of metal ligand charge transfer and π-π* transitions were assigned to the spectrum, further validating the formation of Ir1. This solvated complex exhibits no emissive character when excited at 365 nm.
Figure 1: Synthesis and Characterization of Ir1. Chloride bridge splitting reaction of Dichlorotetrakis(2-(2-pyridinyl)phenyl)diiridium(III) to create the aquo complex Ir(ppy)2(H2O)2+ (Ir1). Upon addition of L-His to the aquo complex in aqueous buffer, luminescent signal is switched on.
Once the complex was synthesized and characterized, the amino acid selectivity was analyzed, as previously described using a similar iridium analog17. Figure 2A shows that only histidine elicited a signal response at 510 nm, when excited at 365 nm.
Figure 2: Amino Acid Selectivity and Titration of Ir1 with Histidine Rich Peptides. (A) Interaction of 200 µM of various amino acids (Cys, Ser, Asp, Glu, Phe, Lys, Arg, Tyr, Trp, His) with 50 µM Ir1 in HBS. Spectral scans were taken from 400-700 nm at an excitation wavelength of 365 nm in a black 96 well plate. Signal in relative fluorescence units (RFU) from all amino acids besides histidine (dashed black trace) was negligible. (B) Nanomolar concentrations of BNT-II (▪) and rcHRP-II (♦) were titrated with Ir1 in HBS. The histidine rich peptides were incubated with the probe for 15 min in solution before reading the emission at 510 nm after excitation with 365 nm light. Limit of detection was calculated as the value of x when y = 3σblank.
This “on-switch” of the phosphorescence of the iridium probe occurs because the singlet excited state (1MLCT/1LC) of the Ir(III) bioconjugate undergoes intersystem crossing to the triplet excited state (3MLCT), when histidine is coordinated to the metal center. This probe was applied to BNT-II a peptide mimic of the malarial biomarker Plasmodium falciparum Histidine Rich Protein II (pfHRP-II). In a titration of BNT-II with Ir1, a concentration dependent signal response was observed (Figure 2B). Recombinant HRP-II (rcHRP-II) also exhibited a similar response in solution. The limits of detection of BNT-II and rcHRP-II in solution were 54.8 nM and 12.8 nM, respectively. Using biolayer interferometry techniques, the KD of Ir1 binding to BNT-II immobilized on a Ni(II)NTA surface was found to be 2.05 µM (Figure 3).
Figure 3: Real-time Kinetic Analysis of Ir1 with BNT-II. Biolayer interferometry for kinetic analysis of various concentrations of Ir1 binding to BNT-II on the surface of a Ni(II)NTA glass sensor. After equilibrating the sensors in KB (Region A), the sensors are loaded with 0.5 µM BNT-II (Region B). Once the peptide is loaded on the sensors, a baseline is established (Region C) prior to measuring the association of Ir1 to the BNT-II (Region D). After a period of association, the sensors are placed back into KB to measure dissociation (Region E). The whole kinetic analysis process takes less than 30 min.
When transitioning from a solution-based assay to a magnetic particle platform, the added complexity of the particle in the system had to be taken into account. It was known from previous work, that these particles efficiently bind the malaria biomarker pfHRP-II.24 Black fluorescent plates worked well for the solution assay describe above, but white plates were better suited for the on-particle detection platform, as seen in Figure 4A. In a white plate, the light is reflected back into the sample, thus allowing for better absorbance by the Ir1 bound to the surface of the particle. In the on-particle assay, the limit of detection of rcHRP-II was determined to be 14.5 nM (Figure 4B). The limit of detection for rcHRP-II in-solution and on-bead were statistically the same based on an unpaired t-test (p = 0.731).
Figure 4: On-bead Detection of BNT-II and HRP-II. (A) Difference between the signal detected from Ir1 bound to BNT-II on the surface of 50 µm Ni(II)NTA magnetic agarose particles in a black 96-well plate (solid line) versus a white 96-well plate (dashed line). The particles were excited with 365 nm light, and emission was measured at 510 nm. (B) Titration of rcHRP-II immobilized on the magnetic particles and detected using Ir1. Limit of detection was calculated as the value of x when y = 3σblank.
Figure 5: Schematic Representation of the On-bead Detection of rcHRP-II with Ir1. General scheme of HRP-II binding to the surface of Ni(II)NTA particles and labeled with Ir1. The particles are incubated with a histidine rich peptide for 15 min. After this incubation period, the particles are washed with HBST using a magnet, in order to pull the particles out of solution. Finally, the peptide bound particles are incubated with Ir1 for 1 hr prior to reading the emission at 510 nm after excitation with 365 nm light.
As microparticle-based diagnostic tools come to the forefront of modern diagnostic technology, detection of target biomolecules directly on the surface of the particle is a major advantage. Traditional molecular detection methods, such as ELISA and PCR, are advantageous, in that they can achieve very low limits of detection25. However, these assays require extensive reagents and lengthy protocols. This assay was designed to mimic ELISA-type sandwich interactions without the time and reagent requirements (Figure 5). Additionally, the cost of the Ir1 probe per sample was estimated to be around a penny, whereas the cost of antibodies for a typical ELISA is $0.20-0.30 per sample.
Since the methods described above rely on a cyclometalated iridium probe for detection of the malarial biomarker, this probe would not be susceptible to failure modes common to other detection methods (i.e. fluorophore quenching and antibody/enzyme degradation). Histidine is unique in its metal binding properties. The outlined work takes advantage of this phenomenon by replacing the antibodies in a typical ELISA sandwich with metal containing complexes. Ni(II)NTA captures rcHRP-II on the surface of the particle and Ir1 signals the presence of the protein. The most critical step in the assay is allowing the magnetic particles to mix with the sample and the probe. In a 96-well plate ELISA, the biomolecules and reagents reach equilibrium with the two-dimensional surface of the bottom of the well. Magnetic particles have the tendency to settle out of solution due to their dense iron-oxide core, which would reduce the available binding sites. Thus, the particles must be mixed during the assay time to ensure maximal binding.
When designing a reagent for disease diagnostics, one must keep in mind the form of patient sample as well as the physiological concentration of the biomarker. For malaria, the concentration of pfHRP-II in a patient’s blood can vary from low picomolar to high nanomolar. While this assay is clinically relevant for higher levels of infection, the limit of detection needs to be improved in order to detect asymptomatic patients with low picomolar circulating pfHRP-II. In the methods outlined above, the “switch-on” signal generated by Ir1 occurs in the presence of histidine. While the malarial biomarker is rich in histidine, other serum proteins, such as human serum albumin and histidine rich glycoprotein, would elicit a signal response with the probe. This would in turn lead to false positive diagnoses. This makes the capture of the protein on the surface of the particle a beneficial step, in that the protein of interest can be rapidly extracted from a patient sample. Additionally, designing a bifunctional probe, that couples a histidine rich peptide to a robust molecular recognition element (i.e. aptamer), could add another layer of specificity to the assay. The peptide would be loaded with iridium before coupling to the aptamer. In such a design, the stable Ir1 probe can still be utilized while achieving target specificity with the aptamer. This would allow for application of the probe to detect native HRP-II in a complex matrix (i.e. plasma or whole blood) and reduce non-specific binding effects. In order to further enhance the signal of rapid diagnostic tests, the assay could be incorporated into an electrochemiluminescent (ECL) system, where pfHRP-II can be detected on the RDTs as a electrochemical readout26. This next generation iridium probe would greatly enhance our on-bead ELISA assay for the detection of disease biomarkers.
The authors have nothing to disclose.
Support for this work was provided by the Bill and Melinda Gates Foundation Grand Challenges in Global Health: Develop Technologies That Allow Assessment of Multiple Conditions and Pathogens at Point-of-Care. K.M.D. was supported by an NSF Graduate Research Fellowship (2012095464).
Dicholortetrakis(2-(2-pyridinyl)phenyl)diiridium(III) | Sigma-Aldrich | 658383 | |
silver trifluoromethanesulfonate | Sigma-Aldrich | 176435 | |
Silica gel | Sigma-Aldrich | 6858 | |
L-Amino acids | Fisher Scientific | varies | |
HEPES | Sigma-Aldrich | H3375 | |
BNT-II peptide | Synthesized in house | N/A | |
recombinant HRPII | Immunology Consultants Laboratory inc. | AGPF-55 | |
Costar black polypropylene 96-well plate | Fisher Scientific | 07-200-567 | |
Costar white polypropylene 96-well plate | Fisher Scientific | 07-200-589 | |
Grenier black polypropylene 96-well plate | VWR | 89089-582 | |
Ni(II)NTA magnetic agarose particles | Qiagen | 36111 | |
Ni(II)NTA biolayer interferometry sensors | ForteBio | 18-5101 | |
Octet Red96 Biolayer Interferometry Instrumet | ForteBio | ||
Biotek Synergy H4 Microplate Reader | Biotek | ||
Fisher Biotech Transilluminator | Fisher Scientific | FBDLT-88 |