By Camella J. Carlson

Introduction

Mycobacterium tuberculosis, a dangerous bacterium typically affecting the lungs, has proved challenging to treat and cure. Tuberculosis (TB) is one of the top 10 causes of death worldwide. About one-fourth of the world population is infected with latent TB. In 2017 alone, there were an estimated 10 million new active TB cases and 1.3 million deaths attributed to TB.¹ In fact, deaths caused by TB just recently surpassed those by HIV, making it the leading infectious disease killer. Individuals with suppressed immune systems, for example, those living with HIV, are particularly susceptible to developing active TB. Tuberculosis usually affects the lungs, but it can also present as extrapulmonary TB infecting other areas of the body.

TB IS ONE OF THE TOP 10 CAUSES OF DEATH WORLDWIDE

TB’s rapidly evolving strains present a particularly formidable problem. Many strains become drug resistant and even multidrug resistant (MDR), a problem worsened by patients carrying out their treatment regimen incorrectly.² The worldwide strategy to combat this disease is multifaceted; efforts include increasing healthcare access and education in areas affected by TB, developing more accurate diagnostic methods, and improving drug research for more powerful treatments. Globally, the TB incidence rate is falling by 2% each year. However, this rate needs to increase in order to meet the goals of the World Health Organization’s End TB Strategy.

Expedited drug development methods are necessary to adequately address TB, particularly MDR-TB.³ Treating these strains increases the complexity of the treatment, adding time and cost. An average treatment could last at least 20 months. Our project aims to establish a faster and more cost-effective drug development process. Drug development typically requires longitudinal animal studies to assess effectiveness. These studies involve monitoring a new TB drug in animal subjects; they require many animals in order to sacrifice a few animals and analyze TB progression at multiple time points throughout the study. The Maitland lab proposes utilizing an optical detection method to simplify these studies and to require fewer animal subjects. This method would be more statistically relevant because the same animal could be measured at multiple points in time, resulting in studies with more scientific rigor. Additionally, a more sensitive analysis method could detect changes in the amount of TB sooner, providing quicker feedback about the likelihood of success for a treatment candidate.

Background

The optical detection of TB would employ a technology called reporter enzyme fluorescence.⁴ Reporter enzyme fluorescence uses a molecule that is cleaved only when it contacts a specific bacterial enzyme in Mycobacterium tuberculosis. After the fluorescent portion of the molecule is freed from the quenching portion, the molecule can be excited by light. Excitation causes the molecule to emit light at a wavelength longer than that of excitation. This light can be detected, and the amount of signal corresponds to the number of bacteria. This method provides a faster alternative to the current standard of analyzing the growth of bacteria on an agar plate; optical detection can deliver results in a matter of hours whereas culture takes 6 to 8 weeks.

Our system design is tailored to monitor TB in a mouse model, because mice are commonly used to assess potential drugs. We are utilizing both computer simulations of light transport and physical models of mouse lungs in order to optimize the system setup and further enhance its performance.

The specific scope of this project is to investigate how mouse lungs can be accurately modeled by an optical tissue phantom. Essentially, an optical tissue phantom is a plastic model that interacts with light in the same way as the biological tissue—sharing both similar shape and optical properties. Phantoms are commonly used to evaluate and calibrate imaging systems. Most are made of one or two materials with rather simplistic geometry. Adding complexity by replicating complicated anatomy, adding texture, or using more materials can mimic tissue with greater accuracy. Our phantom will both validate the computer simulation and test potential system configurations without requiring further animal studies.

Table 1. Summary of corresponding lung and phantom composition.

Figure 1. Simplified lung geometry depicting the airways and alveoli.

Methods

Because our main application is pulmonary TB, the phantom was made with particular attention to the properties and structure of the lungs. The airways, the lung tissue, and the surfactant, a mucus-like substance that coats the airways, were the three primary components depicted in the model because light interacts with each differently. Optical properties characterizing how light interacts with a specific medium were considered when picking the representative materials. Air was selected for the airways; a polymer, polydimethylsiloane (PDMS), for the lung tissue; and glass for the surfactant (Table 1). These materials all had a similar refractive index, which describe the speed and path of light passing through, to the biological tissue.

Lung airways branch out ending in tiny air sacs called alveoli (Figure 1). The alveoli are coated in surfactant and compose the majority of the lung volume. This geometry must also be represented in the model. Molds were fabricated to recreate the lung and airway shapes. Spherical glass microbubbles—small, round, thin glass shells filled with air—were employed to replicate the alveoli structures. Varying amounts of microbubbles were used to represent different points throughout the respiratory cycle, because the lung has different optical properties depending on the volume of air inside.⁵

MICROBUBBLES WERE USED TO REPRESENT DIFFERENT POINTS THROUGHOUT THE RESPIRATORY CYCLE

Sample phantoms were created and characterized to confirm their ability to model the alveoli. Prior to phantom fabrication, the microbubbles were sieved to narrow their diameter size distribution to 45–100 µm, an approximation of mouse alveoli size. First, the PDMS was measured and mixed. Next, the microbubbles were added and dispersed evenly throughout the mixture. The amounts of microbubbles were calculated based on average microbubble size and density to represent 30%, 40%, 50%, and 60% volume ratios. The phantoms required degassing to remove any air bubbles left from mixing. Then, they were solidified by curing in an oven.

Thick and thin phantom samples were made in order to be analyzed by two different methods: The thick samples were fabricated in petri dishes and removed once cured (Figure 2). They included a fluorescent chemical for viewing with a confocal microscope which captures an image by scanning a laser over the sample. Images were taken from the center of each phantom. The resulting images were used to verify the volume ratio and confirm successful removal of unwanted air bubbles.

For the thin phantoms, I used an integrating sphere to measure the transmittance and reflectance of light through each sample (Figure 3). Transmittance measures the amount of light that passes through a sample, and reflectance measures the amount of light reflected off of the sample surface. Measurements were taken at multiple points on the sample and averaged. These values were then used to calculate the absorption and scattering coefficients using inverse adding-doubling software.⁶ Both describe how light interacts with a specific material. Absorption measured how likely light is to get absorbed, and scattering how likely to get scattered instead of passing straight through.

Figure 2. Cured thick phantoms shown in order of increasing volume ratio (30%, 40%, 50%, 60%).

Figure 3. Thin phantoms were cured between microscope slides shown in order of increasing volume ratio (30%, 40%, 50%, 60%).

Results

The confocal microscope images validated that the intended volume percentages for each thick phantom was achieved (Figure 4) by comparing the area inside the bubbles to the entire field of view.

For each thin phantom, the scattering coefficient was consistent across the spectrum of visible light, so it could be approximated as a constant value for each volume percentage (Figure 5). The scattering showed a linear relationship to the volume of air bubbles (Figure 6). This trend corresponded to, and thus, validated the light transport computer simulation.⁷ Additionally, these phantoms successfully represented different points throughout the respiratory cycle.

Figure 4. Confocal microscope images: 30% (a) and 50% (b) volume ratios.

Figure 5. Scattering of microbubbles in PDMS across the visible light spectrum.

Figure 6. Comparing the scattering of the computer simulation and optical phantoms as it relates to microbubble volume.

A PHANTOM OFFERS A GREATER ADVANTAGE OVER AN ANIMAL BECAUSE IT HAS KNOWN CHARACTERISTICS

Discussion

These microbubble compositions can now be incorporated in molds replicating the mouse lung geometry. These phantoms can be used not only to optimize our system for improved TB detection, but also in a broader context. Because this research provides further insight into how light interacts with the complex geometry and materials composing the lung, it could aid in advancing diagnostic and treatment technologies for other pulmonary pathologies.

Optical tissue phantoms can be used to calibrate, optimize, and validate imaging systems. A phantom offers a greater advantage over an animal because it has known characteristics, whereas animals have unknown and varying properties. By increasing the complexity of the phantom, we are further developing means to minimize animal testing and carry out research responsibly.

Acknowledgments

This work was supported by the National Institute of Allergy and Infectious Diseases and the National Science Foundation. I would like to Dr. Maitland and Dr. Durkee for their mentorship and guidance throughout the project.

---

References

1. World Health Organization, Global Tuberculosis Report 2018 (Geneva: World Health Organization, 2018): 1-4.

2. Nazir Ismail et al., “Drug Resistant Tuberculosis in Africa: Current Status, Gaps and Opportunities,” African Journal of Laboratory Medicine 7, no. 2 (December 2018): 781, https://doi.org/10.4102/ajlm.v7i2.781.

3. Mahesh S. Vasava et al., “Drug Development Against Tuberculosis: Past, Present and Future,” Indian Journal of Tuberculosis 64, no. 4 (October 2017): 252–75, https://doi.org/10.1016/j.ijtb.2017.03.002.

4. Ying Kong and Jeffrey D. Cirillo, “Reporter Enzyme Fluorescence (REF) Imaging and Quantification of Tuberculosis in Live Animals,” Virulence 1, no. 6 (November 2010): 558–62, https://doi.org/10.4161/viru.1.6.13901.

5. Johan F. Beek et al., “The Optical Properties of Lung as a Function of Respiration,” Physics in Medicine and Biology 42, no. 11 (December 1997): 2263–72, https://doi.org/10.1088/0031-9155/42/11/018.

6. Prahl SA, et al., “Determining the optical properties of turbid media by using the adding-doubling method,” Applied Optics 32, no. 1 (1993): 559–68.

7. Durkee MS, et al., “Light scattering by pulmonary alveoli and airway surface liquid using a concentric sphere model,” Optics Letters 43, no. 20 (2018): 5001–5004.

Camella J. Carlson ‘20

Camella J. Carlson ‘20 is a biomedical engineering major from Olathe, Kansas. Camella performed this research under the guidance of Dr. Kristen Maitland out of a desire to further understand the role of research in the medical field and to contribute to a meaningful project. She plans to pursue further education as preparation to conduct translational medical research at a research hospital.

Vertical Divider