Fig 1.
Overview of lipoarabinomannan (LAM) detection strategies.
When LAM is associated with a host lipoprotein carrier such as HDL, detection can be performed using lipoprotein capture, which requires two antibodies, as well as prior knowledge of LAM-lipoprotein carrier associations. An antibody targeting apolipoprotein A1, the coat protein of HDL, is used to capture the nanodiscs on the assay surface, followed by detection with a fluorescently labeled antibody targeting LAM. In the absence of sequestration by a host lipoprotein carrier, LAM can be directly detected by membrane insertion, which requires only one antibody. The amphiphilic antigen, LAM, is allowed to partition into a supported lipid bilayer interface, followed by detection with a specific fluorescently labeled antibody. Graphic representations are not drawn to scale. Figure created with BioRender.com.
Fig 2.
Assay optimization for the detection of LAM in human serum by lipoprotein capture assay.
Measurement of LAM by lipoprotein capture assay, as a function of concentration. (a) Representative spectral measurement of LAM (1.5 μM) incubated overnight at 4 °C in control human serum, with the specific signal (relative fluorescence units, RFU) from the detection of α-LAM antibody (15 nM) as a function of emission wavelength (nm). The background and non-specific signals are measured before the addition of LAM. (b) Lipoprotein capture assay was performed for the detection of LAM spiked into control serum at various concentrations and incubated overnight to allow incorporation of the amphiphile into carrier assemblies. Results are plotted as RFU as measured on the waveguide-based optical biosensor, at increasing concentrations of LAM. All values given in (b) are the mean ± standard deviation derived from at least two independent determinations (n = 2). Statistical significance was determined by Welch’s t-test using GraphPad Prism 8.
Fig 3.
Assay optimization for the detection of LAM in human serum by membrane insertion assay.
Measurement of LAM by membrane insertion assay, as a function of concentration. (a) Representative spectral measurement of LAM (0.5 μM) incubated overnight at 4 °C in control human serum, with the specific signal (relative fluorescence units, RFU) from detection of α-LAM antibody (15 nM) as a function of emission wavelength (nm). The background and non-specific signals are measured before the addition of LAM. (b) Membrane insertion assay was performed for the detection of LAM spiked into control serum at various concentrations and incubated overnight to allow incorporation of the amphiphile into carrier assemblies. Sample processing was done to remove lipoproteins. Results are plotted as RFU as measured on the waveguide-based optical biosensor, at increasing concentrations of LAM. All values given in (b) are the mean ± standard deviation derived from at least two independent determinations (n = 2). Statistical significance was determined by Welch’s t-test using GraphPad Prism 8.
Fig 4.
Direct detection of LAM in patient serum samples.
Detection of LAM in clinical serum samples using the lipoprotein capture assay. Data are presented as the Signal/Noise (S/N) ratio with a value above 1.0 indicating a positive result. The measured S/N in sera from 54 patients from 3 different categories (see legend) is plotted. Results of urine testing using the Determine™ TB LAM Ag assay at the time of enrollment into the clinical study: patients 1–20 were positive, patients 21–48 were negative, and healthy controls 49–54 were not tested.
Table 1.
Signal to noise ratios, by clinical group, for the lipoprotein capture assay.
Fig 5.
Comparison of lipoprotein capture and membrane insertion.
(a) Representative measurement of LAM (0.5 μM), by lipoprotein capture (black bars) and membrane insertion assay (grey bars), incubated overnight at 4 °C in control human serum, with the specific signal (RFU) from the detection of α-LAM antibody (15 nM). Values are the mean ± standard deviation derived from at least two independent determinations (n = 2). Statistical significance was determined by Welch’s t-test (*P < 0.05). (b) Comparison of LAM detection signal by lipoprotein capture (black bars) and membrane insertion assay (grey bars) in patient serum samples. Data are presented as the S/N ratio with a value over 1 indicating a positive result. Samples 14, 3, 4, 27, 22 and 24 are positive for LAM by either blood or sputum culture methods, whereas samples 52, 53 are healthy controls.