Redox Imaging of Inflammation in Asthma

To the Editor:

Glutathione (l-γ-glutamyl-l-cysteinylglycine) is found at exceptionally high levels in the human lung (1–3), where it serves as the primary redox couple in the reduced GSH and oxidized GSSG forms. We have identified that the airway epithelium of patients with asthma is shifted to greater reducing potential, recognized by a higher ratio of reduced to oxidized glutathione (4), which occurs in response to the episodic oxidative inflammation of asthma (1, 5–9). On this basis, we hypothesized that regional airway inflammation in asthma may be quantitatively evaluated by a nuclear imaging strategy based on uptake of a radiopharmaceutical (^99mTc-exametazime [HMPAO]) selectively retained in tissues depending on GSH levels (10–13) (see Figure E1 in the online supplement). To test this concept, airway redox, inflammation, and ^99mTc-HMPAO uptake determined by single-photon emission computed tomography (SPECT) were evaluated in individuals with asthma and in control subjects (Table E1). Some of the results of these studies have been previously reported in the form of abstracts (14–16).

^99mTc-HMPAO (Ceretec; GE Healthcare, Arlington Heights, IL) readily diffuses into cells, where it is reduced to a hydrophilic form in the presence of GSH and retained in tissues (Figure E1) (10–13). ^99mTc-HMPAO accumulation in the lung was confirmed to be glutathione dependent as indicated by markedly less uptake in mice depleted of glutathione with diethyl maleate (17) as compared with naive mice (Figure E1). Many studies report an increase in total glutathione in asthmatic lungs that occurs in an adaptive response to the episodic asthmatic exacerbations that are characterized by profound oxidative bursts of inflammatory cells in the airways (1, 5–9). Here, in anticipation of imaging, intracellular glutathione was determined in airway epithelial cells obtained at bronchoscopy from patients with asthma and control subjects to assess the magnitude of difference in redox potential between asthma and control lungs. Airway epithelium was obtained by gently brushing the airway, and the freshly obtained cells were immediately processed in the bronchoscopy suite for determination of reduced and oxidized glutathione and reducing potential in human airway epithelium of control subjects and patients with asthma in vivo. Similar to a prior report (4), GSH tended to be higher but GSSG lower in asthma (GSH, μM: patients with asthma, 17 ± 3 [n = 3]; control subjects, 8.9 ± 0.5 [n = 3]; P = 0.14; GSSG, μM: patients with asthma, 0.38 ± 0.04; control subjects, 0.48 ± 0.08; P = 0.2), so that the ratio of GSH to GSSG in the airway epithelium of patients with asthma was double that of control subjects (GSH/GSSG: patients with asthma, 43 ± 6; control subjects, 20 ± 4; P = 0.05) (Figure 1A). The reduction potential (Eh) of the GSH:GSSG couple calculated by Nernst equation (18, 19) revealed a significantly greater intracellular reducing state in the airway epithelium of patients with asthma as compared with control subjects (Eh, mV: patients with asthma, –323 ± 4; control subjects, –306 ± 3; P = 0.04) (Figure 1A). On this basis, we tested whether redox-based imaging might differentiate asthmatic from control lungs, and used SPECT and coregistered computed tomography (CT) imaging to quantitate ^99mTc-HMPAO concentration and to determine anatomic sites of ^99mTc-HMPAO retention (detailed imaging methods are presented in the online supplement). ^99mTc-HMPAO SPECT/CT imaging revealed visibly greater HMPAO uptake in asthmatic lungs as compared with control lungs (Figures 1C and 1D; Movie E1). Normalized uptake values (NUVs) were significantly higher in patients with asthma than control subjects within central and middle transverse areas of lower lung regions (Figure 1E). Interestingly, there was significant variance in the ^99mTc-HMPAO NUV of upper and lower lung regions of patients with asthma. Control lung uptake was similar among upper and lower lungs, and across the central, middle, and peripheral transverse regions of the lung. In contrast, patients with asthma had significantly greater ^99mTc-HMPAO uptake in lower lung as compared with upper lung regions, and the greatest uptake within central lung regions as compared with middle and peripheral transverse regions (Figure 1E and Table 1). The finding of greater uptake in central lung regions is consistent with a greater distribution of conducting airways in central regions, and the notion of greater reducing potential in asthmatic airways. Overall, the ^99mTc-HMPAO uptake findings indicate a greater than normal reducing potential in asthmatic airways as compared with normal airways, as well as significant heterogeneity of lung redox, with the most striking changes in the lower and central regions of the asthmatic lung. In this context, the lower and central regions tend to be the most prominent regions of air trapping (20, 21), which also supports the idea of more severe inflammation/airway disease in those regions.

Figure 1.

Redox imaging of asthmatic lungs. (A) Glutathione, GSH/GSSG, and glutathione (GSH:GSSG) redox potential (Eh) are greater in the airway epithelium of patients with asthma (n = 3) as compared with control subjects (n = 3). (B–D) Quantitation of ^99mTc-HMPAO uptake in patients with asthma (n = 7) and in control subjects (n = 7). Regions of interest were selected in the upper and lower lungs, and the transverse section of each region was segmented into central, middle, and peripheral areas for quantitation of ^99mTc-HMPAO normalized uptake values (NUVs). Coregistered SPECT and CT images of the lungs in coronal and transverse section show greater ^99mTc-HMPAO uptake in the lungs and thoracic spine of (C) patients with asthma as compared with (D) control subjects. Movie E1 shows three-dimensional ^99mTc-HMPAO SPECT images of the lungs of patients with asthma and control subjects. (E) Asthmatic lungs have greater NUV as compared with control lungs in the central and middle regions of the lower lobes. All *P < 0.05. ^99mTc-HMPAO = ^99mTc-exametazime; CT = computed tomography; GSH = glutathione, reduced; GSSG = glutathione, oxidized; SPECT = single-photon emission computed tomography.

Table 1:

Regional Distribution of Inflammation and ^99mTc-HMPAO Uptake in Asthmatic Lung

	Lower Lobe	Upper Lobe	P Value*
Bronchoalveolar lavage
Volume, ml	43 ± 4	44 ± 4	0.5
Protein, μg/ml	104 ± 15	100 ± 10	0.8
Cell count, × 106	3.6 ± 2.2	6.4 ± 0.9	0.16
Macrophage, %	88 ± 2	94 ± 6	0.4
Neutrophil, %	4.4 ± 3.6	1.3 ± 0.3	0.5
Lymphocyte, %	6.6 ± 2.0	4.2 ± 1.2	0.5
Eosinophil, %	0.8 ± 0.3	0.6 ± 0.1	0.08
IL-4, pg/ml	2.83 ± 0.06	2.72 ± 0.08	0.004
IL-13, pg/ml	31.7 ± 1.3	29.0 ± 1.3	0.05
IFN-γ, pg/ml	12.3 ± 2.2	17.2 ± 4.8	0.3
VEGF, pg/ml	245 ± 42	244 ± 41	0.9
99mTc-HMPAO SPECT uptake (NUV) in transverse sections
Central	10.1 ± 0.4	9.0 ± 0.6	0.04
Middle	9.9 ± 0.5	8.5 ± 0.5	0.04
Peripheral	9.2 ± 0.4	8.2 ± 0.4	0.05

Definition of abbreviations: NUV = normalized uptake value; SPECT = single-photon emission computed tomography; ^99mTc-HMPAO = ^99mTc-exametazime; VEGF = vascular endothelial growth factor.

All data are for patients with asthma; control subjects did not undergo bronchoscopy because only patients with asthma showed differences in HMPAO uptake in upper and lower lobes of the lung. Asthma, n = 9 participants for bronchoalveolar lavage studies, and n = 7 participants for SPECT studies.

^*Paired t test of upper versus lower lobe values, and across regions in transverse sections.

The greater ^99mTc-HMPAO uptake in lower regions of asthmatic lungs also suggested that inflammation might be greater in lower lung regions. To investigate this, bronchoalveolar lavage (BAL) was performed in the upper and lower lobes of patients with asthma to compare inflammatory cells and cytokines. BAL volume return, protein concentration, and total cell counts were similar among upper and lower lobes (Table 1). Eosinophils and cytokines that typify helper T-cell type 2 (Th2) inflammation, IL-4 and IL-13, were present in greater concentrations in lower lobes as compared with upper lobes (Table 1). In contrast, IFN-γ, a prototypic Th1-type cytokine, tended to be greater in upper lobes (Table 1). The IL-13 levels in BAL were inversely related to IFN-γ levels (R = –0.483, P = 0.02).

This letter reveals that there is a greater reducing potential in the asthmatic airway as compared with control lungs but that there is also substantial heterogeneity of reducing potential across the lung in asthma. This supports the general concept that functional imaging based on redox potential may be used to identify intensity and regionality of airway inflammatory disease. It is important to note that the greater uptake of HMPAO may not necessarily be due to the airway epithelium in asthma. In fact, the endothelium is also inflamed in asthma and might also account for the uptake of the HMPAO (22–26). However, vascular endothelial growth factor was similar among upper and lower lung regions in asthma (Table 1). Further, to assess whether the variation of HMPAO uptake in asthma was related to variation in vascular perfusion, the blood flow in the pulmonary circulation was measured in the supine position, using ^99mTc-macroaggregated albumin SPECT/CT (patients with asthma, n = 3; control subjects, n = 3). Pulmonary blood flows were not different in upper and lower regions of the lungs (all P > 0.5). Because all subjects underwent ^99mTc-HMPAO SPECT/CT in the supine position (during injection, uptake, and scan), this validated that blood flow to upper and lower parts of the lungs was equally distributed and unlikely to be the cause of variance in upper and lower lobe HMPAO uptake in asthma. Likewise, there may be potential effects and/or dependence of ^99mTc-HMPAO uptake on participant age or sex. Although numbers are small, there was no apparent association among ^99mTc-HMPAO uptake by age or sex (all P > 0.3).

Other studies have identified heterogeneity of remodeling and airflow obstruction in the lungs of patients with asthma as identified by anatomical and functional imaging modalities, for example, high-resolution CT imaging and hyperpolarized helium magnetic resonance imaging (21, 27). The participants in this study were patients with stable asthma, and further investigation is necessary to determine whether ^99mTc-HMPAO uptake changes with asthma exacerbations and/or with greater asthma severity. Resolution limitation of SPECT imaging precludes the assignment of uptake to specific airways, limiting the generalized application of ^99mTc-HMPAO SPECT/CT imaging in asthma. However, future development of redox-imaging compounds detectable by superior resolution modalities, for example, positron emission tomography, would be a valuable approach to visualize sites of remodeling by coregistered CT images and amounts of inflammation by redox imaging. Current studies rely on invasive bronchoscopic approaches to investigate regional asthma inflammation. The invasive nature and risk of these approaches can preclude enrollment of those patients with severe airflow limitation. A noninvasive functional nuclear imaging approach might offer an alternative quantitative solution to advance research in asthma and assess new biologic-based therapies.