Optical Biosensor Rapidly Detects Monkeypox Virus

Researchers at UC San Diego and Boston University have developed an innovative optical biosensor called PD-IRIS (Pixel-Diversity Interferometric Reflectance Imaging Sensor) that can rapidly detect monkeypox virus (MPXV). Here are the key points about this new technology:

Key Features:
- Can detect MPXV within 20 minutes from sample collection to results
- Achieves very high sensitivity - can detect concentrations as low as 200 PFU/ml (about 3.3 attomolar)
- Is about 9 times more sensitive than traditional lab-based ELISA tests
- Does not require complex sample preparation or preprocessing
- Uses specialized antibodies that bind to the A29 protein on the virus surface
- Can distinguish MPXV from similar viruses like herpes simplex and cowpox

How it Works:
- Uses interferometry and multi-spectral imaging to detect individual virus particles
- Captures virus particles on a specialized silicon chip using specific antibodies
- Creates a digital signature by illuminating particles with different wavelengths of light
- Can count individual virus particles bound to the sensor surface
- Provides real-time monitoring of virus binding

Advantages for Clinical Use:
- Could be used at point-of-care rather than requiring lab facilities
- Works with heat-inactivated (non-infectious) samples
- Can detect both major clades of MPXV
- Sensor chips can be stored at room temperature
- Compatible with testing complex biological samples
- Simple enough for non-expert users

The researchers suggest this technology could help control MPXV outbreaks, especially in resource-limited settings, by enabling quick and accurate diagnosis. The system could potentially be adapted to detect other viruses as well.

The development seems particularly timely given the recent MPXV outbreaks, including a severe outbreak of Clade I MPXV in the Democratic Republic of the Congo that has caused over 1,000 deaths since January 2023.

Figure 1 from the paper illustrates the complete workflow of the PD-IRIS device for detecting monkeypox virus (MPXV). Here's a step-by-step breakdown:

1. Sample Collection & Preparation:

- Shows a patient's arm with MPXV lesions
- Sample from lesions is collected and mixed with MPXV-A29 monoclonal antibodies (green Y-shaped structures)
- These antibodies specifically bind to the A29 protein on the virus surface (shown as blue sphere with green antibodies attached)

2. Device Setup:

- The central portion shows the PD-IRIS device's optical system
- Uses a multi-color light source (red and blue beams)
- Has a microscope-like setup with specialized optics
- The sample flows through a microfluidic channel on a chip
- Chip has spots of:
* Protein G (yellow spots) - captures virus-antibody complexes
* Streptavidin (pink spots) - serves as negative control

3. Detection System:

- Uses a Bayer filter (colored checkerboard pattern) on the camera sensor
- Captures different wavelengths of light
- Automated image processing converts raw data into usable signals
- Shows progression from raw image to processed result

4. Results:

- Graph shows real-time results
- Yellow line: Protein G signal increases as virus particles bind
- Black line: Streptavidin (negative control) shows no significant binding
- Complete process takes 20 minutes from sample to result

The figure effectively demonstrates how the device combines optical, biological, and computational elements to achieve rapid MPXV detection.

Here are the key researchers involved in developing the PD-IRIS sensor:

Lead Researchers also included:

1. Partha Ray
- Location: UC San Diego School of Medicine
- Role: Co-principal investigator
- Contribution: Provided biological expertise and authenticated samples

2. Selim Ünlü
- Position: Distinguished Professor of Engineering
- Location: Boston University
- Role: Co-principal investigator
- Contribution: Led the lab that developed the optical biosensor technology
- Has over 10 years of research experience developing optical biosensors for virus detection

Other Key Team Members:

At UC San Diego:
- Howard Brickner
- Alex E. Clark
- Aaron F. Carlin

At Boston University:
- Mete Aslan (PhD student in electrical and electronics engineering)
- Iris Celebi
- Elif Seymour

CDC Contributors:
- Michael B. Townsend
- Panayampalli S. Satheshkumar

The research was funded by:
- UCSD Pandemic Response to Emerging Pathogens, Antimicrobial Resistance and Equity (PREPARE) Grant
- San Diego Center for AIDS Research (SD CFAR)
- National Science Foundation (NSF-TT PFI grant)
- National Institute of Allergy and Infectious Diseases at the National Institutes of Health

The collaboration between UC San Diego's biological expertise and Boston University's engineering capabilities appears to have been crucial to the project's success. The team bridged biological and engineering disciplines to create this new diagnostic tool.

The study was published in November 2024 in the journal Biosensors and Bioelectronics. This appears to be a culmination of work building on both Ünlü's decade-plus experience in optical biosensors and the urgent need for better MPXV diagnostics highlighted by recent outbreaks.

From the documents, the PD-IRIS device appears to still be in the research and development phase. While the researchers demonstrate its potential for clinical applications, there are no direct mentions of current manufacturing plans or clinical approvals.

Key points about its current status:

Development Status:
- Successfully tested in laboratory conditions
- Described as a "prototype" throughout the paper
- Need for further optimization mentioned for point-of-care implementation

Changes Needed for Clinical Implementation:
1. Fluidic System Modifications:
- Current prototype uses an expensive syringe pump
- Researchers suggest replacing it with "a passive and disposable cartridge that includes an absorbing pad"

2. Stage System Simplification:
- Currently uses an expensive 3-axis NanoMax stage
- Could be replaced with "a custom stage with a differential driver" for manual focus adjustment

3. Manufacturing Considerations:
- The microarray nature of the biosensor substrates is described as "readily scalable for chip manufacturing"
- Researchers note this could lower the cost per test
- Sensor chips have demonstrated ~6 month shelf life at room temperature

Future Steps Mentioned:
- Need to establish calibration curves correlating surface density to PFU/mL using known samples
- Development of conversion algorithms for clinical test results
- Further validation with clinical specimens

Barriers to Implementation:
- Authors note there is "little market for diagnostics addressing future threats"
- Suggest government support would be needed for commercialization

The researchers express that while the technology shows promise for point-of-care use, particularly in resource-limited settings, it still requires optimization and support to move from prototype to manufactured product. There's no mention of FDA approval processes or specific timelines for clinical implementation.

A new variant of human mpox has claimed the lives of approximately 5% of people with reported infections in the Democratic Republic of the Congo since 2023, many of them children. Since then, it has spread to several other countries. The World Health Organization declared the outbreak a Public Health Emergency of International Concern on August 14. In addition, a different but rarely fatal mpox variant was responsible for an outbreak that has spread to more than 100 countries since 2022.

There is an urgent need for faster and more cost-effective diagnostic tools to curb the spread of mpox and to prepare for the possibility of a future global pandemic. Researchers from University of California San Diego School of Medicine, Boston University, and their colleagues have now developed an optical biosensor that can rapidly detect monkeypox, the virus that causes mpox. The technology could allow clinicians to diagnose the disease at the point of care rather than wait for lab results. The study was published on November 14, 2024 in Biosensors and Bioelectronics.

In the clinic, mpox symptoms such as fever, pain, rashes and lesions resemble those of many other viral infections, says Partha Ray, an associate project scientist at UC San Diego School of Medicine and co-principal investigator on the study. “So just by looking at the patient, it is not easy for clinicians to distinguish monkeypox from these other diseases.”

What’s more, polymerase chain reaction (PCR) is currently the only approved method of diagnosing mpox. It is expensive, requires a laboratory, and can take days or weeks to get results. “A deadly combination when there is a fast-spreading epidemic or pandemic,” said Ray.

The search for a better molecular diagnostic for mpox draws on more than 10 years of research in the lab of Selim Ünlü, a distinguished professor of engineering at Boston University (BU) and co-principal investigator on the study. The lab has developed optical biosensors for detecting the viruses that cause Ebola hemorrhagic fever and COVID-19, among others. Ray’s team at UC San Diego collaborated with Ünlü’s lab, providing biological expertise and authenticated samples to Ünlü’s engineering team.

The study, led by first author Mete Aslan, a Ph.D. student in electrical and electronics engineering at BU, used a digital detection platform called Pixel-Diversity interferometric reflectance imaging sensor, or PD-IRIS, to detect the virus.

“Just by looking at the patient, it is not easy for clinicians to distinguish monkeypox from these other diseases.”

Partha Ray

The researchers used samples collected from the lesions of a patient at UC San Diego Health with laboratory-confirmed mpox. They briefly incubated the samples with monoclonal monkeypox antibodies provided by Ray’s lab that bind to proteins on the surface of the virus. The virus-antibody complex was then transferred into tiny chambers on the surface of silicon chips on the sensor that were treated to fix these nanoparticles.

Shining precise wavelengths of red and blue light simultaneously on the chips caused interference, which resulted in slightly different responses when the virus-antibody nanoparticles were present. A color camera was used to detect this small signal and count individual particles with high sensitivity.

“You're not trying to see the scattered light from the virus particle itself, but you're looking at the interferometric signature of the field of scattered light mixed with the field that is reflected from the surface of the chip,” said Ünlü. He likens the process to FM radio, which mixes a weak signal containing information with a more powerful carrier signal at the same frequency, which, in turn, amplifies the weak signal.

The scientists also analyzed herpes simplex virus and cowpox virus samples, which have similar clinical presentations to mpox. The biosensor assay easily discriminated mpox samples from these other viruses, demonstrating that the specificity of the assay is essential for distinguishing mpox from these common viral diseases.

“Within two minutes, we can tell whether someone has monkeypox or not,” said Ray. “From collecting the virus samples to getting the real-time data takes around 20 minutes.”

In the clinic, the rapidity of the test would allow healthcare providers to diagnose mpox cases much more quickly than sending samples out to a lab. This is especially important for slowing community spread in countries where healthcare resources are sparse. Clinicians could also start treatment, if available, more quickly.

Ray envisions the tests being mass-produced as kits and sold to clinics, further reducing costs. A single boxed kit could be used to test for a variety of viruses, such as syphilis or HIV.

“The chip would be the same,” said Ray. “The only thing that would be different here is the binding antibody that would be specific for a particular virus.”

Ray and Ünlü are working together toward the goal of commercialization, not only to address the urgent need for rapid mpox tests in the Democratic Republic of the Congo but also to keep outbreaks from turning into pandemics. However, the researchers say this effort will require government support because there is little market for diagnostics addressing future threats.

“If we don't take care of this particular epidemic right now, it is not going to be limited within Africa,” said Ray.

Additional co-authors on the study include: Howard Brickner, Alex E. Clark, Aaron F. Carlin, UC San Diego; Elif Seymour, iRiS Kinetics, Boston University Business Incubation Center; Michael B. Townsend, Panayampalli S. Satheshkumar, Centers for Disease Control and Prevention; Iris Celebi, Boston University; Megan Riley, axiVEND.

The study was funded, in part, by the National Institute of Allergy and Infectious Diseases at the National Institutes of Health (P30 AI036214), and the National Science Foundation (NSF-TT PFI 2329817).

Read the full study.

Disclosures: All authors completed and submitted the International Committee of Medical Journal Editors form to disclose potential conflicts of interest. No other potential conflicts of interest were disclosed.

Under a Creative Commons license

open access

ABSTRACT

Diagnostic approaches that combine the high sensitivity and specificity of laboratory-based digital detection with the ease of use and affordability of point-of-care (POC) technologies could revolutionize disease diagnostics. This is especially true in infectious disease diagnostics, where rapid and accurate pathogen detection is critical to curbing the spread of disease. We have pioneered an innovative label-free digital detection platform that utilizes Interferometric Reflectance Imaging Sensor (IRIS) technology. IRIS leverages light interference from an optically transparent thin film, eliminating the need for complex optical resonances to enhance the signal by harnessing light interference and the power of signal averaging in shot-noise-limited operation ~~to achieve virtually unlimited sensitivity~~. In our latest work, we have further improved our previous 'Single-Particle' IRIS (SP-IRIS) technology by allowing the construction of the optical signature of target nanoparticles (whole virus) from a single image. This new platform, 'Pixel-Diversity' IRIS (PD-IRIS), eliminated the need for z-scan acquisition, required in SP-IRIS, a time-consuming and expensive process, and made our technology more applicable to POC settings. Using PD-IRIS, we quantitatively detected the Monkeypox virus (MPXV), the etiological agent for Monkeypox (Mpox) infection. MPXV was captured by anti-A29 monoclonal antibody (mAb 69-126-3) on Protein G spots on the sensor chips and were detected at a limit-of-detection (LOD) - of 200 PFU/ml (∼3.3 attomolar). PD-IRIS was superior to the laboratory-based ELISA (LOD - 1800 PFU/mL) used as a comparator. The specificity of PD-IRIS in MPXV detection was demonstrated using Herpes simplex virus, type 1 (HSV-1), and Cowpox virus (CPXV). This work establishes the effectiveness of PD-IRIS and opens possibilities for its advancement in clinical diagnostics of Mpox at POC. Moreover, PD-IRIS is a modular technology that can be adapted for the multiplex detection of pathogens for which high-affinity ligands are available that can bind their surface antigens to capture them on the sensor surface.

KEYWORDS

Monkeypox (Mpox)

Intact Virus detection

Label-free biosensor

Pixel Diversity Interferometric Reflectance Imaging Sensor (PD-IRIS)

Point of Care (POC) diagnostics

INTRODUCTION

Monkeypox virus (MPXV) is an enveloped double-stranded DNA virus belonging to the Orthopoxvirus genus in the Poxviridae family (Sklenovská and Van Ranst, 2018). MPXV causes Monkeypox (Mpox), an infectious zoonotic disease first reported in non-human primates in 1958 (von Magnus et al., 1959) and in humans in 1970 in the Democratic Republic of Congo (DRC) (Ladnyj et al., 1972). In the past, Mpox was considered a rare sporadic disease with a limited capacity to spread among humans (Organization and others, 1984). However, the recent rapid global spread of this infection, which is now recognized as the most critical orthopoxvirus infection in humans in the post-smallpox eradication era, has brought this neglected disease back into the spotlight.

Based on their genome sequence, MPXV has two major types: clade I and II (Likos et al., 2005). The clade I virus, endemic to central Africa, is particularly virulent, with human case fatality rates during some outbreaks estimated to be around 10% but could be below 2% with basic medical care (Pittman et al., 2023). A current Mpox (Clade I) outbreak in the DRC is a testament to this. The country has reported the most significant surge of Mpox cases ever recorded, with over 20,000 suspected cases and more than 1000 deaths since January 2023 (CDC, 2024a). The Clade II virus, formally known as the West African clade, can be further categorized into two phylogenetically distinct subclades: Clade IIa and IIb. Clade II results in less severe infection than Clade I; however, the global Mpox outbreak that started in 2022, caused by Clade IIb, serves as a stark reminder of the potential worldwide impact of these viruses. The rapid spread of clade II Mpox initially led the WHO to declare a Public Health Emergency of International Concern (PHEIC) on July 23, 2022, which was later lifted on May 10, 2023, following the recommendation from the International Health Regulations (IHR) Emergency Committee noting that progress was made in managing the disease outbreak. However, it should be noted that clade II Mpox cases are still being detected in non-endemic countries, and as of September 2024, 102,000 confirmed cases in over 122 countries had been reported, resulting in 222 deaths (CDC, 2024b).

MPXV is transmitted primarily through close contact with infected wild animals or individuals and direct contact with body fluids (CDC, 2024c). There is currently no treatment approved by the U.S. Food and Drug Administration (FDA) for Mpox. However, a clinical trial is underway to test the efficacy of Tecovirimat (an antiviral drug FDA-approved to treat smallpox infection) in treating Mpox (Stomp, 2024). A smallpox vaccine (JYNNEOS™) has also been recently approved in the USA by the FDA under Emergency Use Authorization (EUA) for the prevention of Mpox infection in adults (FDA, 2024); however, due to the limited supply of the vaccine early during the outbreak, vaccine hesitancy and its equitable distribution and access are significant concerns (Kahn et al., 2023).

Therefore, one measure that should be utilized to prevent the further spread of this infection is to make a timely diagnosis of the MPXV and isolate infected individuals. MPox's clinical manifestations, such as skin lesions, fever, headache, muscle soreness, and lymphadenopathy (swollen lymph nodes), are often atypical and can be easily mistaken for other prevalent infections (Hussain et al., 2022). Thus, diagnostics based on clinical criteria alone are challenging and underscore the need for molecular-based diagnosis, which can assist physicians in managing the disease and help health authorities implement effective countermeasures (Bourner et al., 2024).

Confirmatory tests of Mpox infection are performed by quantitative real-time PCR (qPCR) on the isolated virus DNA collected from the patient’s lesion specimens. However, the large central region of the Orthopoxvirus genus is highly conserved. Therefore, choosing proper target regions on the Mpox genome for PCR is essential to prevent cross-reactivity of the tests and differentiate between other orthopoxvirus species. Also, rare deletions in the MPXV genome result in false negative results in some PCR detection methods when targeting non-essential areas of the genome (Garrigues et al., 2022). Moreover, PCR-based tests, while accurate, can only be conducted in dedicated laboratories, making them unsuitable for rapid diagnostic tests at the point of care (POC). Few recent studies based on other nucleic acid amplification and detection methods, e.g., Loop-Mediation Isothermal Amplification (LAMP), Recombinase Polymerase Amplification (RPA), and Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-Cas system, have demonstrated the Mpox diagnostic capability at the POC (Wang et al., 2024; Yu et al., 2023). However, factors like pre-processing steps for nucleic acid extraction from specimens, complicated designing of multiple primers for the viral genome amplification, high costs of reagents, and non-specific amplification producing false-positive results are some of the technical and logistical drawbacks that need to be addressed before their deployment at POC.

Antigen-antibody-based POC assays like the Lateral Flow Immuno Assay (LFIA) are affordable, can be conducted at home or outpatient clinics by untrained personnel, and are thus ideal for rapid testing of viruses (Ince and Sezgintürk, 2022). However, compared to PCR-based assays, they often lack sensitivity, specificity, and quantitative readout to detect viral antigens (Posthuma-Trumpie et al., 2009). There is no FDA-approved commercial antigen-based assay for detecting Mpox (CDC, 2024d). The serological assay that detects antibodies developed in patients in response to Mpox infection can also be used for retrospective analysis at the point of care. However, the kinetics of antibody production in response to infection varies in patients, especially in immunocompromised patients like people living with AIDS. Additionally, these tests are not specific for Mpox, often resulting in false positive results due to the patients’~~prior smallpox vaccination history (~~Matusali et al., 2023).

These limitations of currently available nucleic acid or antigen-based tests highlight the urgent need to develop better point-of-care diagnostic tests for Mpox. An ideal diagnostic test would combine the sensitivity and specificity of quantitative nucleic acid-based tests with the user-friendly nature of rapid antigen-based point-of-care tests. This is particularly crucial in low-resource settings, where untrained personnel can utilize a simple diagnostic system that does not require expensive equipment to provide rapid, reliable, and affordable test results.

Optical sensors have been widely applied to detect viruses due to their simple use, cost-effectiveness, and direct detection capability (Maddali et al., 2021). An optical biosensor can detect viral nucleic acids, antigens, whole viruses, or antibodies produced in response to the viral infection for viral diagnostics. Some examples of these sensors include fluorescence-based optical sensors, surface plasmon resonance (SPR), optical resonators, and interferometry-based methods. Fluorescence-based sensors utilize fluorophore-attached detection molecules to bind to the viral target molecules captured on the surface via a virus-specific capture probe. Although fluorescence-based systems offer multiplexed detection and simpler workflows compared to lab-based tests such as ELISA, issues such as limited sensitivity, photobleaching, and non-specific binding affect the performance of these sensors (Seymour et al., 2023a). Label-free biosensors like SPR and optical resonator-based platforms are promising for POC viral diagnostics and can offer simple, direct, and sensitive detection (Baaske and Vollmer, 2012; Pandey et al., 2022). However, despite being good candidates for POC diagnostic devices, these platforms have certain limitations. SPR sensors suffer from the bulk effect (background signal due to buffer and temperature changes), low selectivity, limited multiplexing, and high substrate costs (Hassan et al., 2021; Nguyen et al., 2015). Optical resonator-based sensors are sensitive to temperature changes, and their substrates require complicated fabrication processes (Rho et al., 2020).

In contrast to these techniques, interferometry-based sensors gained attention owing to their simple and robust transduction mechanisms and minimal dependence on external factors. One such interferometric platform is a single-particle interferometric reflectance imaging sensor (SP-IRIS) composed of an LED-based imaging system and a layered silicon/silicon dioxide microarray chip (Daaboul et al., 2010). SP-IRIS can visualize single nanoparticles (NPs), both synthetic (e.g., polystyrene, gold NPs) and natural (e.g., intact viruses, extracellular vesicles), captured on the sensor surface via immobilized capture probes (Daaboul et al., 2016, 2014; Scherr et al., 2016a, Scherr et al., 2016b). Digital detection by SP-IRIS is achieved by collecting all the light emanating from the sensor surface in the presence of virus particles. The corresponding signal correlates to the particle's polarizability (or size). Discerning the optical signatures becomes particularly challenging as the size gets smaller. Therefore, SP-IRIS utilizes z-scan acquisition (a stack of images taken from different focal positions) to capture the defocus signature unique to sub-diffraction limited scattering object (viruses on sensor surface). However, z-scan acquisition imposes two significant drawbacks: (i) acquiring z-stacks requires repeatable and high-resolution scanning optics. (ii) the computationally expensive algorithms are required to process the z-stacks.

In this work, we present the first demonstration of multi-spectral Pixel Diversity IRIS (PD-IRIS) for label-free and rapid detection of MPXV, where the necessary optical signature is encoded into multiple wavelengths. The major improvement introduced in PD-IRIS is that target particles can be detected at a single snapshot, making the precise z-scanning parts obsolete. This is achieved by exciting the particles under multi-parametric light and collecting the resulting waves on a CMOS array decorated with a filter array. The term pixel-diversity stems from this multi-parametric imaging technique. We utilize a monoclonal antibody (mAb 69-126-3), developed by the Centers for Disease Control and Prevention (CDC) (Roumillat et al., 1984), to capture intact MPXV on the PD-IRIS sensor chip surface spotted with Protein G (Figure 1). The antibody binds to the A29 protein with high affinity and specificity (Hughes et al., 2014a, Hughes et al., 2014b). MPXV A29L is a homolog of vaccinia virus A27 associated with a mature virus membrane that binds to the host cell surface heparan sulfate and is essential for membrane fusion (T.-H. Chang et al., 2013). We demonstrate that the new PD-IRIS prototype can rapidly detect MPXV (sample-to-result within 20 minutes) with higher sensitivity than the traditional laboratory-based Enzyme-Linked Immunosorbent Assay (ELISA) method. The setup can also be reconfigured compactly to fit in a table-top device, allowing easy and quick result interpretation; thus, our assay, independent of any cell culture or sample pre-processing before testing like in SP-IRIS (Monroe et al., 2013), can be performed by untrained personnel and adapted to detect Mpox infection at the POC. Thus, PD-IRIS enables the practical implementation of digital virus detection, allowing for an automated, compact, and robust POC configuration.

MATERIALS AND METHODS

Viral growth, titration, and inactivation

MPXV was isolated in 2022 from a Mpox-positive patient sample on VeroE6 cells (ATCC, CRL-1586). Serial dilutions of the patient sample were made in DMEM (Corning, Catalog number 10-013CM) + 2% FBS (Biowest, Catalog number S1520) + 1x Antibiotic/Antimycotic (Thermo Fisher Scientific, Catalog number: 15240062) + 10mM HEPES (Thermo Fisher Scientific, Catalog number: 15630080) and applied to cells in the same medium but with 10% FBS. Cells were scraped upon the appearance of the Cytopathic Effect (CPE). The virus was confirmed to be Clade II by amplification and Sanger sequencing of two diagnostic regions (positions 46,239-46,737 and 133,388-133,984 in Reference sequence NC_063383) as previously described (Clark et al., 2023) using Q5 Hot-Start High-Fidelity Polymerase (New England Biolabs, Catalog number M0493S) with the following primers: 1-F 5’-ACAGGGTTAACACCTTTCCAATA-3’ + 1-R 5’-AATCTCCAGAACCAGCATCAC-3’ and 2-F 5’ TACAGTTGAACGACTGCG 3’ (Ghate et al., 2023) + 2-R 5’-CTCTCTTGCTTCTTCGTCATAG-3’. The isolated virus was propagated by infection of VeroE6 cells at a Multiplicity of Infection (MOI) of 0.2 and culturing until CPE was observed. Cells were scraped and subjected to 3 cycles of freeze/thaw with vortexing, then clarified by centrifugation, aliquoted, and stored at -80°C for experiments.

MPXV viral stock was titered by plaque assay on VeroE6 cells. Cells were plated in 12-well plates one day before infection. The virus was tenfold serially diluted and incubated on cells with rocking for 1h, then removed and replaced with 0.8% methylcellulose in MEM (Thermo Fisher Scientific, Catalog number:12492013) supplemented with 2% FBS and 1x Penicillin/Streptomycin (Thermo Fisher Scientific, Catalog number: 15140122). After three days of cell growth at 37°C and 5% CO₂, monolayers were fixed in 4% formaldehyde and stained with crystal violet to visualize plaques.

MPXV was heat-inactivated at 65°C for 40 min in a thermocycler. Before removal to BSL2, complete inactivation was confirmed by plaque assay and by culturing 10% of the material on VeroE6 cells for > 1 week, followed by blind passaging. Cells were monitored for CPE, and media was collected to track copies of the viral genome using the above primers to assay for an increase in viral copies. The quantitative PCR (qPCR) of the heat-inactivated MPXV samples was also conducted to estimate the copies/mL (Supplementary Table 2).

HSV-1 from a positive patient sample was isolated on BHK-21 cells (ATCC, CCL-10) by applying serial dilutions of the sample made in RPMI (Thermo Fisher Scientific, Catalog number: 11875093) + 20mM HEPES to monolayers of cells with rocking for 1h, then adding DMEM + 10% heat-inactivated FBS + 1x Antibiotic/Antimycotic. Supernatants were harvested when CPE spread throughout. As described above, the virus was titered by a plaque assay on VeroE6 cells, except that the overlay was 0.43% agarose in DMEM supplemented with 2% FBS and 1x Penicillin/Streptomycin. HSV-1 was heat-inactivated as described for MPXV, and inactivation was confirmed by plaque assay.

MPXV and HSV-I were isolated from consented patient samples under UC San Diego IRB #160524. All work with infectious MPXV was conducted under Biosafety Level (BSL) 3 conditions following guidelines approved by the Institutional Biosafety Committee.

Cowpox virus (CPXV), strain Germany_1998_2, was grown in confluent flasks of BSC-40 cells. (ATCC, CRL-2761). Cells were infected at an MOI of 0.1 and incubated for 48-72 hours before harvesting. Cells were scraped, pelleted at ∼1000 x g, the supernatant removed, and the cell pellet resuspended in RPMI + 2% FBS before undergoing 3x freeze/thaws and sonication before aliquoting. The virus was titered using serial dilutions in 6-well plates of confluent BSC-40 cells. After incubation for 48 hours, plaques were visualized with a crystal violet stain and counted. The resulting titer was 1.1 x 10⁸ PFU/mL. Virus aliquots were then inactivated with gamma irradiation (4.4 x 10⁶ rads) while frozen on dry ice. The CPXV work was conducted at the BSL2+ facility, and the CDC provided samples.

Enzyme-Linked Immunosorbent Assay (ELISA)

The ELISA assay was performed as described previously (Ray et al., 2024). Briefly, the inactivated viruses (MPXV, HSV-1, and CPXV) at the indicated PFU/mL were diluted in the carbonate-bicarbonate (pH 9.4) buffer (ThermoFisher Scientific Catalogue: 28382). For coating the 96-well microtiter plates (Sarstedt, Catalogue: 82.1581.100), a 100 μL volume of these diluted viruses was added directly to each well, sealed with adhesive strips, and incubated overnight at 4ºC. The contents of the plates were discarded the next day, and the wells were washed with 200 μL PBS buffer. Following this, 200 μL of blocking buffer (1x PBS with 1% BSA) was added to each well, covered with adhesive strips, and incubated at room temperature for two hours to block the remaining well surface unoccupied by the antibodies. This step improves the assay's sensitivity by reducing the background signal and increasing the signal-to-noise ratio. The blocking buffer was discarded, and 100 μl of the anti-A29 monoclonal antibody (mAb 69-126-3-7) (Hughes et al., 2014a, Hughes et al., 2014b) was added to each well at 1:3000 dilution in 1x PBS with 1% BSA buffer. The concentration of the antibody stock was 2.228mg/ml. The plates were then covered with adhesive strips and incubated for 90 min at room temperature on the rocker. The samples were discarded, and the wells were washed four times with 200 μL of PBS buffer. Next, 100 μL secondary antibody, Goat anti-mouse-HRP (Thermo Fisher Scientific Catalogue: 31430) at 1:4000 dilution, was added, and the plates were sealed with adhesive strips and incubated at room temperature for one hour on the rocker. The samples were discarded, and the wells were washed four times with 200 μL of PBS buffer. 100 μL of 3,3′,5,5′-Tetramethylbenzidine (TMB) substrate solution (Thermo Fisher Scientific Catalogue: N301) was added to each well, waited for two minutes, and the 100 μL TMB stop solution (Thermo Fisher Scientific Catalogue: N600) was added. The plates were scanned within 15 minutes in the Tecan Multimode microplate reader (Spark®) at 450 nm.

All the assays were conducted in triplicate (n=3) sets at every concentration for statistical significance and Limit of Detection (LOD) calculations. The threshold signal is calculated as an average signal from the negative control (HSV-1) plus three standard deviations. LOD is calculated as the concentration value corresponding to the point where the dilution curve intersects the threshold line.

PD-IRIS Chip Preparation

The surface of the PD-IRIS chips was activated with oxygen plasma (Plasma Etch, PE-25). Subsequently, a polymer-based coating (MCP-4, Lucidant Polymers) was applied to the activated surface of the chips, which covalently binds amine groups. The chips were immersed in a 1X polymer solution (1% w/v polymer in 20% saturated ammonium sulfate) for 30 min. The chips were then thoroughly rinsed with DI water, dried with nitrogen, and baked in a vacuum oven at 80 °C for 15 min. The chips were stored in a desiccator until microarray spotting.

The capture probes immobilized on the surface were 0.5 mg/mL protein G (Millipore Sigma, Catalog number: 08062) and 0.5 mg/mL streptavidin (Prospec, Catalog Number: Pro-791-b). The protein solutions were spotted in 200 mM sodium phosphate buffer with 0.01% Trehalose, pH 8.0, on the MCP-4 coated chips using an M2-Automation iONE-600 spotter. The probes were then incubated in a high-humidity chamber overnight (18 hours) to allow immobilization. After incubation, any remaining active groups on the polymer surface were blocked using a 50 mM ethanolamine solution in 150 mM Tris-HCl, pH 9.0. The chips were incubated in the blocking solution for 1h at room temperature and then washed with 1X PBS with 0.05% Tween. After rinsing in DI water, the chips were dried with nitrogen before assembly.

PD-IRIS Prototype:

PD-IRIS is a wide-field imaging technique that utilizes epi-illumination, i.e., reflected light microscopy. A novel illumination device with two LEDs (410 nm and 660 nm) provides a simple, cost-effective illumination source for uniformity corrections (Çelebi et al., 2023). This light source, EUCLID (Efficient Uniform Color-Light Integration Device), uses an adjustable hollow cavity that enables uniform light mixing from two different input ports. The uniformly mixed light is then introduced to the optical system after it passes through an adjustable iris. This output is imaged on the back focal plane of the objective lens (Nikon, Super Plan Fluor, 20x, 0.45NA) by two achromatic lenses (AC254-075-A, Thorlabs). The diameter of the iris is set to 4 mm to provide low-NA Koehler illumination to the sample chip surface. The reflected light is collected by the same objective lens, and it is imaged onto a CMOS sensor (Flir, BFS-U3-244S8C-C) using a tube lens (TTL 200-A, Thorlabs). The sensor's exposure time is set to 16 ms and runs at a speed of 16 frames per second.

PD-IRIS substrates, 60 nm-oxide silicon chips, are functionalized as described in the “PD-IRIS Chip Preparation” section to capture and immobilize the virions. The chip has two laser-cut holes to introduce sample buffer in and out over the active area. The side and top border of the chip channel are built by attaching a glass cover on the chip using a pressure-sensitive adhesive gasket. The gasket has a 1.5 mm rectangular opening in the center, allowing for imaging of the active area during the incubation. The chip is inserted into a custom chip holder with fluidic connections that provide samples to the chip through its laser-cut holes. The custom holder is mounted onto a 3-axis Nanomax stage (MAX312D). The optimal focus is determined by inspecting the silicon-etched marks on the chip, which are adjusted by a differential drive and locked to avoid any drift.

The flow in the channel is governed by a programmable 500 μL syringe pump (Hamilton, PSD4) with an 8-input valve to select between wash and sample channels, and the channels are connected by 0.01” ID tubes. The pump speed is 1500 μL/min and 10 μL/min for wash and sample channels. The wash channel is connected to the chip through the pump valve, whereas the sample channel is directly connected to the chip to decrease the dead volume introduced by the valve. 100 μL viral samples flow back and forth 3 times, corresponding to ∼1-hour incubation.

The prototype is controlled by custom-written Python code. The camera and pump can be addressed simultaneously, enabling acquisition when the pump runs for incubation. At the end of the experiment, the acquired images are processed by another custom-written Python code to detect and count individual virions.

Limit of Detection Determination for MPXV detection with PD-IRIS

A homogenous assay protocol is employed to optimize the capture efficiency of the chip surface (Seymour et al., 2015). In a homogenous assay, the virions are statically incubated with their corresponding antibody outside the system's fluidic channel (i.e., in an Eppendorf tube), and a molecule that has a high affinity with the antibody is spotted on the chip surface. Then, antibody-decorated virions are flow-incubated over the spotted chip. Different MPXV dilutions are statically incubated for our experiments with the anti-A29 mAb, and protein G is spotted on the chip surface to capture flowing antibody-virus complexes.

For the limit-of-detection (LOD) experiments, three different virus samples are prepared at 5x10⁴ PFU/mL, 1x10⁴ PFU/mL, and 1x10³ PFU/mL, which also contains anti-A29 mAb at 2.5 ng/mL. To prepare homogeneous virus-mAb mixtures, the stock mAb solution (at 2.228 mg/mL) is diluted to 0.01 μg/mL in 1x PBS, filtered with 0.02 μm Whatman Anotop filter. The stock MPXV sample (4.6x10⁶ PFU/mL) is diluted in 1x PBS solution to 1x10⁵ PFU/mL, 2x10⁴ PFU/mL, and 2x10³ PFU/mL. Experimental samples are prepared by mixing 50 μL diluted MPXV sample with 25 μL of diluted antibody solution and 25 μL of filtered 1x PBS to make a homogenous sample of 100 μL. Then, the prepared sample is left for static incubation for 5 minutes.

The flow incubation protocol consists of three steps. First, filtered 1x PBS solution is flowed over the protein G spotted chip twice at 1500 μL/min as an initial wash step for 20 seconds per wash. Then, the waste is replaced by a homogenous sample, and the syringe pulls and pushes the virus solution three times consecutively over the chip at 10 μL/min, corresponding to 65 minutes of flow incubation. After the incubation, the chip is washed twice at the same speed to remove non-specific bindings. During the flow incubation of the homogenous sample, the camera acquires spot images constantly at 16 frames per second. The acquired images are averaged every 30 seconds to create one data point in a real-time binding curve. After the ∼1 hour incubation and washing, the surface-bound particles are counted for both the protein G spots and negative spots and divided by the spot area to express the bound particle density as particles/mm².

The detection threshold is calculated from the HSV-1 homogenous assay, in which the HSV-1 sample is mixed with anti-A29 mAb and flowed over the PD-IRIS chip using the same steps described for the MPXV assay. The detection threshold is calculated as the mean particle density on the protein G spots (n=6) plus three standard deviations. LOD for MPXV detection is determined by linearly extrapolating the two lowest concentration data points and finding the concentration value at the point where the line intersects the threshold signal line.

Specificity Experiments using HSV-1 and CPXV

The specificity experiments are also performed using the homogenous assay protocol described in the previous section. The stock virion samples (9.67x10⁷ PFU/mL for HSV-1 and 1x10⁷ PFU/mL for CPXV) are diluted in filtered 1x PBS down to 2x10⁴ PFU/mL. Then, 50μL of diluted samples are mixed with 25 μL of monoclonal antibody solution and 25 μL of filtered 1x PBS to top the volume of the homogenous sample up to 100 μL. The final viral concentration of the homogenous samples is 1x10⁴ PFU/mL.

The same incubation protocol is followed for specificity experiments. The protein G spotted chip surface is washed twice at 1500 μL/min with filtered 1x PBS. Then, homogenous virus-mAb mixtures are incubated for 65 minutes at 10 μL/min. Finally, the incubation is followed by two wash steps. The data points for the flow incubation are calculated by averaging the accumulated images every 30 seconds.

Proof-of-Concept Experiments

Anti-IgG-coated 80 nm gold nanospheres (Nanopartz) are used and immobilized on the IgG-spotted 60 nm oxide PD-IRIS chip for the proof-of-concept experiment. The chip surface is washed with PBS before and after flowing the GNS-mAb sample at 50 μl/min for 10 min with the syringe pump. The concentration of the GNS is 10⁷ particles/mL (∼10fM), which is diluted in 1x PBS.

Particle Detection and Tracking Algorithm

In PD-IRIS setup, the camera acquires images constantly before, after, and during the flow incubation. The captured 8-bit images are added onto a float-64 array. That array is averaged every 30 seconds, creating one averaged image, and the resulting image is passed to the particle detection and tracking algorithm to quantify the bound particles to the surface after some pre-processing steps. In the pre-processing part, the first step is to extract spot locations from their corresponding circular etched regions. After the spots are located for every frame, the background levels of different color channels, corresponding to reflected light from the spots, are mapped to the same readout value by dividing each channel by its mod. After the peak values of different color channels are set to the same value, the fixed pattern noise (or pixel-to-pixel variation) is removed by applying the look-up table to the normalized image.

The signal extraction from pre-processed Bayer images has two different methods for experiments. In the proof-of-concept experiment, an individual gold nanosphere image is convolved with a variation filter, resulting in the following constructed signal ,(1)Where superscripts i and j represent the pixels of the constructed signal and the raw image. The second operation involves convoluting the raw image with two different 2x2 matrices in parallel. The element-wise square of the resulting matrices was summed to construct signal , which can be written as,(2)

Once the signal is constructed, the particles must be detected and counted for every frame. To do this, the signal is correlated with a 32 x 32 Gaussian function whose width is dictated by the optical resolution . Then, an arbitrary threshold is applied to segment the highlighted features in the correlation result. All of the white regions in the resulting segmented image are detected by OpenCv “findContours” function (Itseez, 2015), and the particle candidate locations are identified by filtering the detected contours given the optical properties of the setup (i.e., diffraction-limited size of the particle, circularity of the contour, etc.). Those locations are passed into the tracking algorithm.

The particle tracking algorithm is based on SPANDEX, developed by Sevenler et al. (Sevenler et al., 2019). It accepts the candidate particle locations and tracks each particle's appearance within a pre-defined region in the time stack using Trackpy (Allan et al., 2023). The tracking enables kinetic assay screening and eliminates the non-specifically bound or falsely detected particles, given that such artifacts cannot appear in the same location for multiple frames. All detection and tracking codes are written in Python, which facilitates the POC applicability of the design.

RESULTS

Multi-Spectral Pixel-Diversity IRIS:

Interferometry is a powerful tool for sensing infinitesimal changes, and its implementation in optical microscopy enables the label-free characterization of biological nanoparticles. Interferometric imaging systems are configured as common path interferometers where light scattered from the sample, , and reflected from the surface, , travel along the same optical path until the imaging sensor on which they interfere. Thus, the resulting intensity read by the sensor is,(3)

Since these methods are designed to detect nanoparticles whose diameter is much smaller than the working wavelength, the scattering intensity term, , is much less than the interferometric term. Thus, the reading depends only on the scattering amplitude and the phase difference between two interfering waves. Many interferometric imaging techniques create their signal by adjusting the phase to achieve the maximum interferometric contrast. Measuring the weight of individual proteins (Young et al., 2018) to detect a single virion (Nava et al., 2023) is demonstrated with interferometric imaging. However, only nonspecific binding events were measured for iSCAMS, and both techniques require precise alignment or auto-focusing systems.

Pixel-diversity IRIS improves upon the existing Single Particle Interferometric Reflective Imaging Sensor (SP-IRIS). SP-IRIS is a well-established technology that detects and characterizes individual biological particles such as viruses, bacteria, and exosomes in a label-free fashion, as well as biomarkers with the use of nanoparticle labels (Seymour et al., 2023b; Xia et al., 2023; Zaraee et al., 2020). The method utilizes a silicon-silicon dioxide chip on which a microarray is spotted to screen multiple target particles. The captured particles are excited with a precise light wavelength determined by the target particle and the oxide thickness in a common-path interferometric system. ~~The optical response of the particle,~~ ~~, and reflected light from the chip surface,~~ ~~, are imaged on a CMOS sensor on which they interfere. In the presence of small biological particles~~ ~~, the interferometric cross term dominates the scattering intensity.~~ Thus, the signal contrast is defined as,(4)

As Eq.4 suggests, the signal relies not on scattering intensity but its amplitude () ~~in SP-IRIS~~. Although there is a tremendous signal enhancement, distinguishing target particles from the background becomes challenging with wide-field, low-magnification objective lenses. Therefore, the focus dependence of the phase angle (Eq.3) is exploited in SP-IRIS by taking multiple defocus acquisitions, which eases the optical alignment requirements. The signal is constructed by calculating the pixel-wise variance of those acquisitions. In this case, the signal-to-noise ratio (SNR) depends on the number of captured images. This creates a tradeoff between temporal resolution and SNR. Another disadvantage of z-acquisitions is that they still require ~~repeatable, high-resolution scanning optics, like a piezo scanner,~~ a piezoelectric objective scanner, significantly increasing the cost.

In multi-spectral PD-IRIS, the need for the high-resolution scanning optics is eliminated. The wavelength dependence of the phase angle in the interferometric cross-term (Eq.3) is exploited to increase the faint signal contrast of a small particle. As a result, the sample can be screened at higher speeds, the data takes up less space on the computer and can be processed with more efficient algorithms. This makes the configuration more cost-efficient and rapid for point-of-care testing. Detection at a single snapshot consists of two steps: (i) illuminating the particles with multiple colors simultaneously and (ii) recording the response with a conventional color camera, resulting in checkerboard patterns in the camera readout in the presence of particles.

The differences between signal construction methods are demonstrated in Figure 2. Anti IgG-coated gold nanospheres are captured on the IgG spots to generate SP- and PD-IRIS signals. In SP-IRIS, 31 defocus images are taken across the optimal focal plane with 20x magnification using 530 nm dominant LEDs and stored in a 3D cube. Then, pixel-wise variations are calculated along the z-dimension. In this method, some spot features, like edges, may generate false signals, which increase the background level. Figure 2b demonstrates a PD-IRIS measurement using the same particles in Figure 2a. The particles are simultaneously excited under RGB light (460 nm, 523 nm, and 630 nm). The standard deviation within each superpixel is calculated to generate the processed image. With this new detection technique, PD-IRIS yields an improved SNR, and some image artifacts due to defocus, like spot edges, can be removed.

The optical signature of the particle is encoded in the high-frequency components of the captured PD-IRIS image. Other high-frequency features/noise must be removed to recover this signal with high SNR. One possible noise originates from the offset of the color channels on the spots. The reflected light from the spots depends on the spot thickness and the illumination wavelength. Since the thicknesses can vary, the value distribution of the spot pixels would be different in every color channel when the LED intensities are not adjusted properly, resulting in a significant background level. This effect is eliminated by detecting every spot in the field of view and correcting the pixel readings with respect to the channel’s mod value. To facilitate this post-processing step, a circular array is etched on the active area of the Si-SiO₂ chip, indicating the location of the spots. Due to the huge contrast difference, the spots can be detected by simple thresholding, and channels are normalized once an individual spot is cropped from the FOV.

Multi-spectral illumination's spatial and spectral non-uniformity is another noise source that would reduce the SNR. Similar to channel offsets, any non-uniformities in the illumination would increase the background level in the constructed PD-IRIS signal. Thanks to our group's recently developed light source, EUCLID (Çelebi et al., 2023), multiple colors can be spatially mixed with exquisite uniformity and inputted to the imaging optics from one output port. EUCLID's simplicity, cost, and size make it appealing for a POC setup.

The last two high-frequency features we studied are shot and fixed pattern noise. The shot noise can be reduced by collecting more electrons by averaging sequential frames. However, SNR cannot be improved further by simply averaging more frames because fixed pattern noise or pixel-to-pixel variations (Zapata-Pérez et al., 2020) become the dominating factor. Reading errors due to pixel-to-pixel variations introduce a significant noise source when particle signature is expressed in the sudden changes between the adjacent pixels. Thus, this must be removed for high SNR PD-IRIS images. This issue can be mitigated by using a uniformly illuminated mirror image as a look-up table and correcting every image accordingly.

Proof-of-concept Experiments with PD-IRIS:

To test the optical performances of different objective lenses and determine the optimal focal position for PD-IRIS, we measured the defocus signature of 80 nm GNSs and the immobilized MPXV particles. MPXV virions are oval, brick-shaped particles with sizes ranging from 220-450 nm in length and 140-260 nm in width (Moss, 2013)

First, we tested color aberrations introduced by the imaging optics. We imaged the immobilized GNSs with 10x (Nikon Plan Apo λD) and 20x (Nikon Super Plan Flour) objective lenses and compared the focal shifts of different colors (Figure 3a-b). Since the 10x objective was designed to minimize the color dispersion, the focal positions for different wavelengths were maintained. The 20x objective lens, however, yielded a –1.5μm shift for the 530 nm dominant green and a +1 μm shift for the 405 nm dominant blue LEDs. Unlike in a regular wide-field microscope, this focal shift increases the signal contrast difference between different wavelengths and improves PD-IRIS performance.

Finally, we determined the optimal focus location for MPXV particles by comparing its defocus profile at 410 nm and 660 nm. The selected wavelengths were also used for all virus detection experiments. Selecting two well-separated wavelengths was to reduce the crosstalk between color channels. As indicated in Figure 3c, there is ∼5 μm optimal defocus region. This focus region can be found manually by using differential drivers instead of piezo stages, which would be a one-time adjustment for the user and reduce the cost significantly. Particle size only affects the amplitude of the defocus curves (Figure 3c), but the defocus signature relies only on the phase difference, thus the particle’s height from the surface (Avci et al., 2016). The immobilized Protein G captures the antibody-occupied MPXV particles at a constant height so that PD-IRIS can be operated within this broad optimal focus range.

MPXV Detection Experiments with PD-IRIS:

Next, we determined our PD-IRIS assay's sensitivity and specificity for MPXV detection and compared our sensor data with ELISA results. The experiments are performed as described in the Materials and Methods section. We opted for a homogenous assay procedure in the experimental design due to its advantages in increasing the assay sensitivity and decreasing the assay time. The virus solutions are first incubated with an anti-A29 monoclonal antibody to allow for virus-mAb complex formation in this assay type. Virus experiments are performed in the prototype PD-IRIS setup (Figure 4a). The PD-IRIS chips are printed with a microarray consisting of protein G (positive), which has a higher affinity for mouse monoclonal antibody than protein A (Fishman and Berg, 2019), and streptavidin spots as the negative control (Figure 4c). The fluidic channel for the flow incubation is built in three layers (Figure 4b). A cover glass encapsulates the buffer solution when assembled to the functionalized PD-IRIS chip via a pressure adhesive tape.

First, we optimized the antibody concentration in homogeneous virus detection experiments to maximize capture efficiency. Unlike heterogenous assays, where the antibodies and virions are incubated sequentially on the active area of the chip, antibody-covered virions are introduced into the channel in a homogenous assay. In this assay format, since the antibodies are added in excess of the virus particles, the antibody concentration needs to be adjusted so that free antibodies in the solution do not saturate the protein G spots. An optimized homogenous assay is more efficient than a heterogenous assay in terms of capture efficiency (Seymour et al., 2021), and sample-to-result time is much shorter, considering it only involves one flow incubation. We mixed the virus samples (1x10⁵ PFU/mL) for the antibody optimization experiments with three different antibody concentrations (2.5 μg/mL, 2.5 ng/mL, 0.25 ng/mL). Given those concentrations, the antibody-to-viral particle ratio is determined and provided in Supplementary Table 1. The 2.5 ng/mL mAb concentration was selected as it yielded the highest signal for detected viral particles (Supplementary Fig. 1).

PD-IRIS utilizes a 1’x0.5’ chip format with a 5 x 25 circular etched array to indicate the spot positions. For our preliminary results with the MPXV, we consider two FOVs covering six protein G (positive) and four streptavidin (negative) spots. The number of captured virions on the positive and negative spots within an FOV is averaged separately to eliminate the spot-to-spot variations due to printing imperfections. First, the Limit of Detection (LOD) is estimated, the results of which are shown in Figure 5a. We conducted experiments with three serial dilutions of MPXV (5 x10⁴, 1 x 10⁴, and 1 x 10³ PFU/mL) and a blank sample (HSV-1 at 10⁴ PFU/mL) on separate chips. PD-IRIS could detect viral particles at 10³ PFU/mL, showing a signal well above the threshold. PD-IRIS achieved a calculated LOD of ∼200 PFU/mL from the extrapolated curve, outperforming ELISA by almost one order of magnitude (LOD ∼1800 PFU/mL) (Figure 5b). It should be noted that in the case of ELISA, although we are coating the plates with inactivated MPXV, there may be free A29 protein in the solution that can contribute to the signal. Whereas, with the PD-IRIS system, we only detected the whole MPXV, not the free A29 protein. For MPXV (Orthopoxvirus), the Viral Particle (VP)-to-PFU ratio is 10 (Americo et al., 2017). Therefore, the PD-IRIS’s LOD - 200 PFU/mL corresponds to 2000 VP/mL. Considering 1 mole is 6.022 x10²³ (N_A = Avogadro Constant) viral particles, the calculated MPXV LOD translates into 2x10⁶/ N_A = 3.3 x 10^-18 mol/L (∼3.3 attomolar).

We also perform kinetic binding measurements with PD-IRIS by real-time data acquisition (Figure 5c). Images are constantly collected with the camera during the incubation at 15 Frames per second (FPS). (A movie and the dynamic graph created from acquired images are shown in Supplementary Video 1, demonstrating real-time virus binding to a protein G (positive) and streptavidin (negative) spot. Obtained images are averaged every 30 seconds to create a data point in Figure 5c. With the help of kinetic measurements, we screen the incubation and validate that it is done without any experimental artifacts such as air bubbles. Real-time screening also enables us to identify all particles, record their interactions with the chip surface, and distinguish between specific and nonspecific bindings. We incubated the virus samples for over 1 hour and monitored the binding as it reached saturation. Signal saturation observed for the two highest virus concentrations is caused by the absence of available capture probes (Protein G) for further virus binding. At these virus concentrations, Protein G spots are occupied with antibody-bound viruses and excess antibody molecules initially added to the virus sample to the extent allowed by steric effects.

Finally, the specificity of the PD-IRIS MPXV assay is characterized using distractor viruses. For this purpose, two homogenous assays, including the Cowpox virus (CPXV), another species of orthopoxvirus genus, and Herpes Simplex Virus (HSV-1), are prepared. The skin lesions typical of Mpox are similar in clinical presentation to those of CPXV and HSV1 infections. Therefore, for proper clinical management, an ideal POC assay should be able to identify Mpox and distinguish it from these prevalent infections specifically. The same incubation protocol used for MPXV is followed. Inactivated CPXV and HSV-1 at 10⁴ PFU/mL are incubated with the A29 mAb at 2.5 ng/mL. After incubation, the tests are loaded to the PD-IRIS sensor, as described previously, and the particles on Protein G spots are counted to obtain the signal. A low background signal level is detected at the protein G and Streptavidin spots for these distractor viruses, with particle densities below the threshold signal (Figure 6a, Supplementary Fig. 2). This indicates that the A29 mAb is specific in capturing only the MPXV on the sensor surface to generate the optical signal. The A29 mAb does not capture CPXV or HSV-1 and, consequently, is not immobilized on the sensor surface to generate signals. The ELISA results corroborate the specificity of the A29 mAb against MPXV compared to CPXV and HSV1 (Figure 6b).

DISCUSSION

Digital detection, or single molecule counting, is an exciting recent development in diagnostics that provides resolution and sensitivity far beyond ensemble measurements. However, most emerging digital detection techniques, like digital PCR, bead-based single molecule assays (e.g., SiMoA), or digital-ELISA, are based on complicated particle confinement/isolation strategies that increase the cost and complexity of the entire process for POC applications. Pixel Diversity Interferometric Reflectance Imaging Sensor (PD-IRIS) is a technology that utilizes light interferometry from an optically thin film for label-free, high-sensitivity detection of individual nanoparticles (such as viruses or exosomes). As opposed to our previous technology, SP-IRIS, PD-IRIS utilizes multi-parametric particle excitation and special imaging sensors to extract the information encoded in this multi-parametric illumination. In multi-spectral PD-IRIS, the particles are excited with spatially and spectrally uniform light, and a conventional color CMOS sensor is used to extract the encoded particle signal. This novel imaging technique enables particle detection from a single snapshot, eliminating the need for the complicated and time-consuming z-scan process used in SP-IRIS. Moreover, in PD-IRIS, SNR is improved by using a uniform light source (EUCLID) and post-processing steps to decrease the noise from various sources. These advancements offered by PD-IRIS make the system more cost-efficient and faster compared to SP-IRIS, increasing its potential for POC diagnostic applications. The selection of illumination wavelength is limited by the Bayer filter's performance in the current PD-IRIS demonstration. Implementing another illumination wavelength might be necessary for smaller biological nanoparticles to improve the PD-IRIS signal for detection. This can be achieved by creating a sensor-specific look-up table to correct for cross-talks between color channels. Moreover, differential data acquisition methods (Foley et al., 2021; Young et al., 2018) can also be employed to improve the SBR further.

Using PD-IRIS, we demonstrated sensitive and specific detection of MPXV with a calculated LOD of 200 PFU/mL, providing a nine-fold improvement in sensitivity compared to a lab-based alternative test, ELISA. This LOD corresponds to ∼3.3 attomolar when 1 mol is taken as 6.022x10²³ (Avogadro Number) viral particles, considering that the viral particle-to-PFU ratio of MPXV solution is 10 (Americo et al., 2017). Moreover, our real-time virus binding experiment shows a significant PD-IRIS signal is achieved in less than 10 min of incubation time for 5 x10⁴ and 1 x 10⁴ PFU/mL samples. For a workflow similar to the one shown in Figure 1, the total assay time is expected to be about 20 min, including the initial virus sample-antibody incubation step. Thus, our results indicate that PD-IRIS offers a clinically relevant assay platform for detecting early Mpox infections (Paran et al., 2022). Since PD-IRIS detects the virus particles' surface density (particles/mm²), we will establish a calibration curve to correlate the surface density to PFU/mL using Mpox samples with known PFU/mL values. When reporting clinical test results, this information will be incorporated into a conversion algorithm to determine the PFU/mL in the test specimen. Although our MPXV detection experiments were performed in buffer media in the current study, our previous work utilizing SP-IRIS demonstrated that our platform can be employed to perform virus detection in complex media such as serum and blood without affecting its sensitivity (Daaboul et al., 2014; Scherr et al., 2016a, Scherr et al., 2016b).

We tested MPXV (Clade IIb) in our prototype PD-IRIS system. However, the amino acid sequence of the antigen epitope to which the anti-A29 monoclonal antibody (mAb 69-126-3) binds is conserved between Clade I and II MPXV (Davis et al., 2023). Therefore, our POC assay can most likely be used to diagnose Mpox infection caused by Clade I and II MPXV. Our ELISA data (Supplementary Figs. 3 and 4) supports this assumption by demonstrating the detection of A29 protein from Congo basin isolate MPXV-ZAI-96-I-16 (Clade I) using the mAb 69-126-3. (Azzi, 2023).

Heat treatment of protein antigens denatures them; consequently, they lose the ability to bind to antibodies. However, using ELISA and PD-IRIS sensors in this study, we found that the heat-inactivated MPXV could bind to the anti-A29 monoclonal antibody (mAb 69-126-3). We think this is due to the coiled-coil (CC) structure of the A29 protein (T. H. Chang et al., 2013); the CC structures are known to be heat resistant (Liu et al., 2020). Moreover, this would help in the clinical testing of non-infectious, heat-inactivated mpox patient samples in the lab without the requirement of BSL3 facilities.

In our PD-IRIS prototype, we employ a syringe pump to control the flow of the samples. Although the pump can adjust the flow speed precisely, it increases the total cost significantly, which is not desirable for a POC system, and is vulnerable to sample contamination for clinical diagnostic purposes. One alternative to the active fluidic system in our PD-IRIS prototype would be having a passive and disposable cartridge that includes an absorbing pad to conduct the sample flow over the chip (Scherr et al., 2017). Another component that needs to be optimized in our prototype for a POC setup is the 3-axis NanoMax (Thorlabs) stage on which the sample holder is mounted. For an initial prototype and optical characterization of the PD-IRIS setup, we used an expensive 3-dimensional stage and open-loop piezo to tune the focus position precisely. However, it is demonstrated that even with the highest magnification objective lens, 20x, there is ∼ 5 μm of optimal focus position that yields a detectable signal. Thus, a custom stage with a differential driver is enough to adjust the focal distance manually for the experimental acquisition. This will further enable widespread POC applicability of PD-IRIS.

In the recently published article by Song et al. (Song et al., 2024), the authors used an optical biosensor (Fiber Optic Biolayer Interferometer) for label-free detection of Mpox A29 protein at a LOD of 0.62 ng/mL in buffer and 0.77 ng/mL in spiked serum samples. Another study by Zhang et al. (Zhang et al. 2023) used Surface-Enhanced Raman Spectroscopy (SERS) to detect A29 protein at an LOD of 5 ng/mL. However, BLI and SERS are laboratory-based assays that require expensive equipment and trained personnel to run and interpret the tests and are not conducive to POC application (Supplementary Table 3). Also, in our study, we decorated the surface of MPXV virions with an anti-A29 antibody and captured the whole virus on the sensor chips for detection. Thus, compared to the other studies that detected free A29 protein, our assay detected the whole MPXV particles and did not require any sample pre-processing steps, e.g., protein extraction.

The PD-IRIS technology could revolutionize the diagnosis of infectious diseases at the point of care for several reasons. The sensor chips can be stored at room temperature in a dry format and do not require refrigeration. Our earlier work demonstrated that dried antibodies have a shelf life of approximately six months at room temperature (Seymour et al., 2021). The stability of Protein G spotted chips can also be enhanced by storing them with stabilizers such as trehalose. Moreover, the microarray nature of the biosensor substrates allows for readily scalable chip manufacturing, thus lowering the cost per test. Due to its robust signal transduction mechanism, environmental effects such as humidity and temperature do not affect PD-IRIS measurements. Since only virus particles captured by high-affinity probes on the sensor surface are visualized and counted, the PD-IRIS platform is compatible with complex sample matrixes such as saliva, whole blood, serum, or other biofluids. Thus, clinical specimens can be tested directly without requiring preprocessing steps to extract the test material or remove interfering biomolecules. Easy sample preparation and automated data acquisition and analysis make PD-IRIS an easy-to-use platform that simple instructions can operate. Finally, single-particle detection makes the system highly sensitive, enabling detection even at the early stages of the infection. Overall, our platform offers highly sensitive detection, with sensitivity levels exceeding that of antibody-based laboratory tests, in a relatively short time (∼20 min), comparable to the duration of lateral flow assays.

The recent COVID-19 pandemic has taught us the catastrophic consequences of failing to detect infectious diseases early in their path to curb their global spread. We certainly want to avoid repeating this mistake with the current surge in Mpox infection. Therefore, rapid detection followed by isolation and treatment of the infected patient is our best current arsenal against this rapidly spreading infectious disease. We envision deploying the PD-IRIS as a diagnostic test for Mpox as a partner in this battle. With its high sensitivity and specificity, this test can significantly improve healthcare outcomes by enabling early Mpox detection and substantially reducing the burden of the disease by preventing its spread.

Moreover, our PD-IRIS system is a versatile and modular technology. It can be adapted to detect other pathogens with high sensitivity and specificity. This can be achieved by recruiting other high-affinity ligands (aptamers or antibodies) that could specifically bind the pathogens' surface proteins and capture them on the sensor surface for digital detection. Various pathogens have been detected using the SP-IRIS modality, such as Ebola, Vaccinia, and Zika viruses (Daaboul et al., 2017). Similar virus detection assays can be efficiently designed for the PD-IRIS platform and extended to include newly emerging viruses that require rapid and sensitive POC diagnostics.

CONCLUSION

Digital detection of label-free bioparticles provides sensitivity beyond that of most ensemble measurements, and if made accessible at an affordable price, it can potentially revolutionize disease diagnostics. This work introduced a rapid multi-spectral, label-free detection tool called PD-IRIS. With the PD-IRIS modality, Monkeypox virus detection and enumeration were achieved from single-color images in a robust instrument with no moving parts, bringing us closer to our goal of making this advanced yet easy-to-use disease diagnostic tool available at a low cost. The microfluidic integration reduces sample volume requirements and allows accessible clinical specimen collection. The sensitivity of PD-IRIS was also superior to the lab-based diagnostic gold-standard ELISA; thus, it can be deployed to diagnose Monkeypox infection at point-of-care, especially in resource-poor areas. The microarray nature of PD-IRIS substrates also makes it possible to create desired viral diagnostics multiplex panels based on need using specific capture probes.

Michael B. Townsend: Writing – review & editing, Writing – original draft, Resources. Panayampalli S. Satheshkumar: Writing – review & editing, Writing – original draft, Resources. Alex E. Clark: Writing – original draft, Investigation. Iris Celebi: Methodology, Investigation, Data curation. M. Selim Ünlü;: Writing – review & editing, Writing – original draft, Supervision, Resources, Project administration, Investigation, Funding acquisition, Conceptualization. Megan Riley: Writing – original draft, Resources, Methodology. Aaron F. Carlin: Writing – review & editing, Resources. Elif Seymour: Writing – review & editing, Writing – original draft, Validation, Investigation, Formal analysis. Howard Brickner: Investigation, Formal analysis, Data curation. Partha Ray: Writing – review & editing, Writing – original draft, Validation, Supervision, Resources, Project administration, Methodology, Funding acquisition, Formal analysis, Data curation, Conceptualization. Mete Aslan: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization

Uncited reference

Declaration of Competing Interest

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

ACKNOWLEDGEMENTS

The project was partially funded by UCSD Pandemic Response to Emerging Pathogens, Antimicrobial Resistance and Equity (PREPARE) Grant#: 2P30AI036214-29 (PR) and the San Diego Center for AIDS Research (SD CFAR), an NIH-funded program (P30 AI036214) supported by the following NIH Institutes and Centers: NIAID, NCI, NHLBI, NIA, NICHD, NIDA, NIDCR, NIDDK, NIMH, NIMHD, NINR, FIC, and OAR. M.S.Ü. received the National Science Foundation, NSF-TT PFI (2329817) award, which partially funded the research work presented in this study. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC.

Data availability

Data will be made available on request.

REFERENCE

Allan et al., 2023
Allan, D.B., Caswell, T., Keim, N.C., der Wel, C.M., Verweij, R.W., 2023. soft-matter/trackpy: v0.6.1. https://doi.org/10.5281/zenodo.7670439

Daaboul et al., 2017
Daaboul G.G., Freedman D.S., Scherr S.M., Carter E., Rosca A., Bernstein D., Mire C.E., Agans K.N., Hoenen T., Geisbert T.W., Selim Ünlü M., Connor J.H.
Enhanced light microscopy visualization of virus particles from Zika virus to filamentous ebolaviruses
Daaboul et al., 2016
Daaboul G.G., Gagni P., Benussi L., Bettotti P., Ciani M., Cretich M., Freedman D.S., Ghidoni R., Ozkumur A.Y., Piotto C., others
Digital detection of exosomes by interferometric imaging
Sci Rep, 6 (2016), Article 37246
Daaboul et al., 2014
Daaboul G.G., Lopez C.A., Chinnala J., Goldberg B.B., Connor J.H., Unlu M.S.
Digital sensing and sizing of vesicular stomatitis virus pseudotypes in complex media: a model for Ebola and Marburg detection
ACS Nano, 8 (2014), pp. 6047-6055
Daaboul et al., 2010
Daaboul G.G., Yurt A., Zhang X., Hwang G.M., Goldberg B.B., Unlu M.S.
High-throughput detection and sizing of individual low-index nanoparticles and viruses for pathogen identification
Nano Lett, 10 (2010), pp. 4727-4731
Davis et al., 2023
Davis I., Payne J.M., Olguin V.L., Sanders M.P., Clements T., Stefan C.P., Williams J.A., Hooper J.W., Huggins J.W., Mucker E.M., Ricks K.M.
Development of a specific MPXV antigen detection immunodiagnostic assay
FDA, 2024
FDA
FDA approves first live, non-replicating vaccine to prevent smallpox and monkeypox
FDA (2024)
Fishman and Berg, 2019
Fishman J.B., Berg E.A.
Protein A and Protein G Purification of Antibodies
Foley et al., 2021
Foley E.D.B., Kushwah M.S., Young G., Kukura P.
Mass photometry enables label-free tracking and mass measurement of single proteins on lipid bilayers
Garrigues et al., 2022
Garrigues J.M., Hemarajata P., Lucero B., Alarcón J., Ransohoff H., Marutani A.N., Kim M., Marlowe E.M., Realegeno S.E., Kagan R.M., others
Identification of human monkeypox virus genome deletions that impact diagnostic assays
J Clin Microbiol, 60 (2022), Article e01655
Ghate et al., 2023
Ghate S.D., Suravajhala P., Patil P., Vangala R.K., Shetty P., Rao R.S.P.
Molecular detection of monkeypox and related viruses: challenges and opportunities
Hassan et al., 2021
Hassan M.M., Sium F.S., Islam F., Choudhury S.M.
A review on plasmonic and metamaterial based biosensing platforms for virus detection
Sens Biosensing Res, 33 (2021), Article 100429
Hughes et al., 2014a
Hughes Laura J., Goldstein J., Pohl J., Hooper J.W., Lee Pitts R., Townsend M.B., Bagarozzi D., Damon I.K., Karem K.L.
A highly specific monoclonal antibody against monkeypox virus detects the heparin binding domain of A27
Hughes et al., 2014b
Hughes Laura J., Goldstein J., Pohl J., Hooper J.W., Pitts R.L., Townsend M.B., Bagarozzi D., Damon I.K., Karem K.L.
A highly specific monoclonal antibody against monkeypox virus detects the heparin binding domain of A27
Virology, 464 (2014), pp. 264-273
Hussain et al., 2022
Hussain A., Kaler J., Lau G., Maxwell T.
Clinical conundrums: differentiating monkeypox from similarly presenting infections
Cureus, 14 (2022)
Ince and Sezgintürk, 2022
Ince B., Sezgintürk M.K.
Lateral flow assays for viruses diagnosis: Up-to-date technology and future prospects
Itseez, 2015
Itseez, 2015. Open Source Computer Vision Library.
Kahn et al., 2023
Kahn P.A., Ying X., Virata M., Magahis P., Li S., Mathis W.S.
Availability and Accessibility of Live Nonreplicating Smallpox/Mpox Vaccine
JAMA Netw Open, 6 (2023), Article e237873
Ladnyj et al., 1972
Ladnyj I.D., Ziegler P., Kima E.
A human infection caused by monkeypox virus in Basankusu Territory, Democratic Republic of the Congo
Bull World Health Organ, 46 (1972), p. 593
Likos et al., 2005
Likos A.M., Sammons S.A., Olson V.A., Frace A.M., Li Y., Olsen-Rasmussen M., Davidson W., Galloway R., Khristova M.L., Reynolds M.G., others
A tale of two clades: monkeypox viruses
Journal of General Virology, 86 (2005), pp. 2661-2672
Liu et al., 2020
Liu Y., Zhou X., Liu W., Miao W.
The stability of the coiled-coil structure near to N-terminus influence the heat resistance of harpin proteins from Xanthomonas
Maddali et al., 2021
Maddali H., Miles C.E., Kohn J., O’Carroll D.M.
Optical biosensors for virus detection: prospects for SARS-CoV-2/COVID-19
ChemBioChem, 22 (2021), pp. 1176-1189
Matusali et al., 2023
Matusali G., Petruccioli E., Cimini E., Colavita F., Bettini A., Tartaglia E., Sbarra S., Meschi S., Lapa D., Francalancia M., others
Evaluation of Cross-immunity to the Mpox Virus due to historic smallpox vaccination
Vaccines (Basel), 11 (2023), p. 1541
Monroe et al., 2013
Monroe M.R., Daaboul G.G., Tuysuzoglu A., Lopez C.A., Little F.F., Ünlü M.S.
Single nanoparticle detection for multiplexed protein diagnostics with attomolar sensitivity in serum and unprocessed whole blood
Moss, 2013
Moss B.
Poxvirus DNA replication
Nava et al., 2023
Nava G., Casiraghi L., Carzaniga T., Zanchetta G., Chiari M., Damin F., Bollati V., Signorini L., Delbue S., Bellini T., Buscaglia M.
Digital Detection of Single Virus Particles by Multi-Spot, Label-Free Imaging Biosensor on Anti-Reflective Glass
Nguyen et al., 2015
Nguyen H.H., Park J., Kang S., Kim M.
Surface plasmon resonance: a versatile technique for biosensor applications
Sensors, 15 (2015), pp. 10481-10510
Organization and others, 1984
Organization W.H., others
The current status of human monkeypox: memorandum from a WHO Meeting
Bull World Health Organ, 62 (1984), pp. 703-713
Pandey et al., 2022
Pandey P.S., Raghuwanshi S.K., Shadab A., Ansari M.T.I., Tiwari U.K., Kumar S.
SPR based biosensing chip for COVID-19 diagnosis—A review
IEEE Sens J, 22 (2022), pp. 13800-13810
Paran et al., 2022
Paran N., Yahalom-Ronen Y., Shifman O., Lazar S., Ben-Ami R., Yakubovsky M., Levy I., Wieder-Feinsod A., Amit S., Katzir M., Carmi-Oren N., Levcovich A., Hershman-Sarafov M., Paz A., Thomas R., Tamir H., Cherry-Mimran L., Erez N., Melamed S., Barlev-Gross M., Karmi S., Politi B., Achdout H., Weiss S., Levy H., Schuster O., Beth-Din A., Israely T.
Monkeypox DNA levels correlate with virus infectivity in clinical samples, Israel, 2022
Pittman et al., 2023
Pittman P.R., Martin J.W., Kingebeni P.M., Tamfum J.J.M., Mwema G., Wan Q., Ewala P., Alonga J., Bilulu G., Reynolds M.G., Quinn X., Norris S., Townsend M.B., Satheshkumar P.S., Wadding J., Soltis B., Honko A., Güereña F.B., Korman L., Patterson K., Schwartz D.A., Huggins J.W.
Clinical characterization and placental pathology of mpox infection in hospitalized patients in the Democratic Republic of the Congo
Posthuma-Trumpie et al., 2009
Posthuma-Trumpie G.A., Korf J., van Amerongen A.
Lateral flow (immuno) assay: its strengths, weaknesses, opportunities and threats. A literature survey
Anal Bioanal Chem, 393 (2009), pp. 569-582
Ray et al., 2024
Ray P., Ledgerwood-Lee M., Brickner H., Clark A.E., Garretson A., Graham R., Van Zant W., Carlin A.F., Aronoff-Spencer E.S.
Design and Development of an Antigen Test for SARS-CoV-2 Nucleocapsid Protein to Validate the Viral Quality Assurance Panels
Rho et al., 2020
Rho D., Breaux C., Kim S.
Label-free optical resonator-based biosensors
Sensors, 20 (2020), p. 5901
Roumillat et al., 1984
Roumillat L.F., Patton J.L., Davis M.L.
Monoclonal antibodies to a monkeypox virus polypeptide determinant
J Virol, 52 (1984), pp. 290-292
Scherr et al., 2016a
Scherr Steven M., Daaboul G.G., Trueb J., Sevenler D., Fawcett H., Goldberg B., Connor J.H., Ünlü M.S.
Real-time capture and visualization of individual viruses in complex media
ACS Nano, 10 (2016), pp. 2827-2833
Scherr et al., 2016b
Scherr Steven M., Daaboul G.G., Trueb J., Sevenler D., Fawcett H., Goldberg B., Connor J.H., Ünlü M.S.
Real-Time Capture and Visualization of Individual Viruses in Complex Media
Scherr et al., 2017
Scherr S.M., Freedman D.S., Agans K.N., Rosca A., Carter E., Kuroda M., Fawcett H.E., Mire C.E., Geisbert T.W., Ünlü M.S., Connor J.H.
Disposable cartridge platform for rapid detection of viral hemorrhagic fever viruses
Sevenler et al., 2019
Sevenler D., Trueb J., Selim Ünlü M.
Beating the reaction limits of biosensor sensitivity with dynamic tracking of single binding events
Seymour et al., 2015
Seymour E., Daaboul G.G., Zhang X., Scherr S.M., Ünlü N.L., Connor J.H., Ünlü M.S.
DNA-Directed Antibody Immobilization for Enhanced Detection of Single Viral Pathogens
Seymour et al., 2023a
Seymour E., Ekiz Kanik F., Diken Gür S., Bakhshpour-Yucel M., Araz A., Lortlar Ünlü N., Ünlü M.S.
Solid-Phase Optical Sensing Techniques for Sensitive Virus Detection
Sensors, 23 (2023), p. 5018
Seymour et al., 2023b
Seymour E., Ünlü M.S., Connor J.H.
A high-throughput single-particle imaging platform for antibody characterization and a novel competition assay for therapeutic antibodies
Seymour et al., 2021
Seymour E., Ünlü N.L., Carter E.P., Connor J.H., Ünlü M.S.
Configurable Digital Virus Counter on Robust Universal DNA Chips
Sklenovská and Van Ranst, 2018
Sklenovská N., Van Ranst M.
Emergence of monkeypox as the most important orthopoxvirus infection in humans
Front Public Health, 6 (2018), Article 383729
Song et al., 2024
Song X., Tao Y., Bian S., Sawan M.
Optical biosensing of monkeypox virus using novel recombinant silica-binding proteins for site-directed antibody immobilization
Stomp, 2024
von Magnus et al., 1959
von Magnus, P., Andersen, E.K., Petersen, K.B., Birch Andersen, A., 1959. A pox-like disease in cynomolgus monkeys.
Wang et al., 2024
Wang Y., Chen H., Lin K., Han Y., Gu Z., Wei H., Mu K., Wang D., Liu L., Jin R., others
Ultrasensitive single-step CRISPR detection of monkeypox virus in minutes with a vest-pocket diagnostic device
Nat Commun, 15 (2024), p. 3279
Xia et al., 2023
Xia Q., Guo Z., Zong H., Seitz S., Yurdakul C., Ünlü M.S., Wang L., Connor J.H., Cheng J.X.
Single virus fingerprinting by widefield interferometric defocus-enhanced mid-infrared photothermal microscopy
Young et al., 2018
Young G., Hundt N., Cole D., Fineberg A., Andrecka J., Tyler A., Olerinyova A., Ansari A., Marklund E.G., Collier M.P., Chandler S.A., Tkachenko O., Allen J., Crispin M., Billington N., Takagi Y., Sellers J.R., Eichmann C., Selenko P., Frey L., Riek R., Galpin M.R., Struwe W.B., Benesch J.L.P., Kukura P.
Quantitative mass imaging of single biological macromolecules
Yu et al., 2023
Yu C., Zuo L., Miao J., Mao L., Selekon B., Gonofio E., Nakoune E., Berthet N., Wong G.
Development of a Novel Loop-Mediated Isothermal Amplification Method for the Rapid Detection of Monkeypox Virus Infections
Zaraee et al., 2020
Zaraee N., kanik F.E., Bhuiya A.M., Gong E.S., Geib M.T., Lortlar Ünlü N., Ozkumur A.Y., Dupuis J.R., Ünlü M.S.
Highly sensitive and label-free digital detection of whole cell E. coli with Interferometric Reflectance Imaging
Zhang et al., 2023
Zhang Z., Jiang H., Jiang S., Dong T., Wang X., Wang Y., Li Y.
Rapid Detection of the Monkeypox Virus Genome and Antigen Proteins Based on Surface-Enhanced Raman Spectroscopy

Cited by (0)

Americo et al., 2017

Droplet digital PCR for rapid enumeration of viral genomes and particles from cells and animals infected with orthopoxviruses

Avci et al., 2016

Avci O., Adato R., Ünlü M.S., Ozkumur A.Y.

Physical modeling of interference enhanced imaging and characterization of single nanoparticles

Azzi, 2023

Azzi A.

Unusual Monkeypox virus outbreak in 2022: Phenotypic and molecular characteristics

Baaske and Vollmer, 2012

Baaske M., Vollmer F.

Optical resonator biosensors: molecular diagnostic and nanoparticle detection on an integrated platform

ChemPhysChem, 13 (2012), pp. 427-436

Bourner et al., 2024

Bourner J., Garcia E., Mbrenga F., Boum Y., Paterson A., Jones B., Olliaro P., Nakouné E., Rojek A.

Challenges in Clinical Diagnosis of Clade I Mpox: Highlighting the Need for Enhanced Diagnostic Approaches

medRxiv (2024), pp. 2023-2024

Çelebi İ., Aslan M., Ünlü M.S.

A spatially uniform illumination source for widefield multi-spectral optical microscopy

Chang et al., 2013a

Chang T.-H., Chang S.-J., Hsieh F.-L., Ko T.-P., Lin C.-T., Ho M.-R., Wang I., Hsu S.-T.D., Guo R.-T., Chang W., others

Crystal structure of vaccinia viral A27 protein reveals a novel structure critical for its function and complex formation with A26 protein

PLoS Pathog, 9 (2013), Article e1003563

Chang et al., 2013b

Chang T.H., Chang S.J., Hsieh F.L., Ko T.P., Lin C.T., Ho M.R., Wang I., Hsu S.T.D., Guo R.T., Chang W., Wang A.H.J.

Crystal structure of vaccinia viral A27 protein reveals a novel structure critical for its function and complex formation with A26 protein

Clark et al., 2023

Clark A.E., Furst A., Sejane K., Stellwagen L., Proost M., Pride D., Smith D.M., Carlin A.F., Bode L.

Validating Tools to Detect and Inactivate Monkeypox Virus in Human Milk, 18 (2023), pp. 785-789, 10.1089/BFM.2023.0175

Researchers from the University of California San Diego School of Medicine have created an optical biosensor that can quickly identify monkeypox, the virus that causes mpox, according to a study published in Biosensors and Bioelectronics on November 14^th, 2024.

Optical Biosensor for Rapid Detection of Monkeypox Virus — *Image Credit: Irina Starikova1811/Shutterstock.com*

A new variant of human mpox has claimed the lives of approximately 5 % of those infected in the Democratic Republic of the Congo since 2023, with many of the victims being children. Since its emergence, this variant has spread to several other countries, prompting the World Health Organization to declare the outbreak a Public Health Emergency of International Concern on August 14, 2024. Meanwhile, another mpox variant, though rarely fatal, has been responsible for an outbreak spanning more than 100 countries since 2022.

Addressing the urgent need for faster, more cost-effective diagnostic tools to curb the spread of mpox and prepare for potential future pandemics, researchers from the University of California San Diego School of Medicine, Boston University, and collaborators have developed an innovative optical biosensor.

This device can rapidly detect monkeypox, the virus that causes mpox, enabling point-of-care diagnosis and bypassing the need for laboratory testing. The study, published on November 14, 2024, in Biosensors and Bioelectronics, presents a significant advancement in molecular diagnostics.

Mpox symptoms, including fever, pain, rashes, and lesions, often resemble those of other viral infections, making visual diagnosis unreliable.

So just by looking at the patient, it is not easy for clinicians to distinguish monkeypox from these other diseases.

Partha Ray, Study Co-Principal Investigator an Associate Project Scientist, School of Medicine, University of California San Diego

Currently, polymerase chain reaction (PCR) is the only approved diagnostic method for mpox. However, it is expensive, requires laboratory infrastructure, and may take days or even weeks to yield results.

Ray noted, “A deadly combination when there is a fast-spreading epidemic or pandemic.”

The newly developed biosensor utilizes a digital detection platform called Pixel-Diversity Interferometric Reflectance Imaging Sensor (PD-IRIS). The technology builds on over a decade of research led by Selim Ünlü, a distinguished professor of engineering at Boston University, who has developed optical biosensors for detecting viruses such as Ebola and COVID-19. Collaborating with Ray’s team at UC San Diego, which provided biological expertise and authenticated samples, the researchers applied PD-IRIS to detect monkeypox.

The team collected samples from lesions of a patient with laboratory-confirmed mpox at UC San Diego Health. These samples were incubated briefly with monoclonal antibodies, provided by Ray’s lab, that bind to viral proteins. The virus-antibody complex was then transferred to tiny chambers on silicon chips designed to fix nanoparticles.

When red and blue light were shone on the chips, interference patterns revealed differences caused by the presence of virus-antibody nanoparticles. A color camera detected these signals with high sensitivity, enabling the counting of individual particles.

“You’re not trying to see the scattered light from the virus particle itself, but you’re looking at the interferometric signature of the field of scattered light mixed with the field that is reflected from the surface of the chip,” said Ünlü, likening the process to FM radio, which amplifies weak signals using a more powerful carrier frequency.

The biosensor assay proved capable of distinguishing mpox from other clinically similar viruses, such as herpes simplex and cowpox. “Within two minutes, we can tell whether someone has monkeypox or not,” Ray said. The entire process, from sample collection to results, takes approximately 20 minutes.

This rapid diagnostic capability could allow healthcare providers to identify mpox cases more quickly, especially in regions with limited healthcare resources. Early diagnosis would also facilitate timely treatment and help slow community transmission.

Ray envisions the biosensor being mass-produced as diagnostic kits for use in clinics, further driving down costs. The platform could be adapted for other viruses, such as HIV or syphilis, by simply changing the antibodies used in the test.

“The chip would be the same. The only thing that would be different here is the binding antibody that would be specific for a particular virus,” Ray stated.

Ray and Ünlü are working toward commercializing the biosensor to address the current mpox outbreak and prevent future pandemics. However, they emphasize that government support will be essential to advance these efforts, given the limited market for diagnostics targeting future threats.

Ray concluded, “If we don't take care of this particular epidemic right now, it is not going to be limited within Africa.”

Collaboration and Funding

The study was led by first author Mete Aslan, a Ph.D. student in electrical and electronics engineering at Boston University, and included co-authors from UC San Diego, the Centers for Disease Control and Prevention, and Boston University’s iRiS Kinetics. The work was funded by the National Institute of Allergy and Infectious Diseases at the National Institutes of Health and the National Science Foundation.

This innovative diagnostic tool represents a major step forward in managing mpox outbreaks and highlights the importance of leveraging interdisciplinary collaboration to address global health challenges.

Journal Reference:

Aslan, M. et. al. (2024) A Label-free Optical Biosensor-Based Point-of-Care Test for the Rapid Detection of Monkeypox Virus. Biosensors and Bioelectronics. doi.org/10.1016/j.bios.2024.116932