An ex vivo antennae-attached locust brain-based gas sensor for the detection of volatile organic compounds associated with lung cancer

Introduction:: Lung cancer is a leading cause of death globally and the development of new approaches for early detection is critical. The analysis of volatile organic compounds (VOCs) in exhaled human breath has been proposed as a means of noninvasively detecting metabolically linked diseases such as cancer. The obstacles in applying a gas-based cancer screening technology in clinical settings is that current VOC sensors lack sensitivity and reliability of detecting complex gas mixtures in natural settings. While gas chromatography-mass spectrometry is the current gold standard, it is generally slow and requires preprocessing of samples. To address these barriers, portable engineered chemical sensors, or electronic noses (e-noses), which mimic biological principles of chemical sensing have been developed and tested in medical diagnostics. However, e-noses still lack specificity, sensitivity, and ability to work in non-laboratory settings. Here we propose to overcome the above challenges by implementing a forward engineering approach by developing a robust insect-based biosensor that is able to detect VOCs associated with lung cancer. Our approach utilizes the superior biological olfactory system, harnessing the full extent of a biological chemical sensory array (insect antenna) and odor-specific neural computational schemes (antennal lobe processing) that have evolved to rapidly detect minute differences in complex gas mixtures. We have developed an ex vivo antennae-attached locust brain preparation, electrophysiological recording technique, and data analysis platform that can successfully differentiate between VOCs associated with lung cancer. This study is the first step in the creation of a portable insect-based biosensor for the detection of lung cancer.


Materials and Methods:: Odor-evoked neuronal voltage responses were recorded from isolated, antennae-attached, ex vivo locust brain placed in a petri dish. Recordings were performed from the olfactory neurons in the locust antennal lobe using 16-channel silicon probes (NeuroNexus Tech.). A miniature, multi-channel amplifier system (Intan Technologies, RHD 32) was used to collect odor-evoked voltage responses from projection neurons and were stored in a computer at a 20 kHz sampling rate using a Labview data acquisition system (PCI-MIO-16E-4 DAQ cards; National Instruments). Precise timing, volume, and concentration of the delivered biomarkers were controlled by a multi-channel olfactometer (Aurora Scientific, 220A Olfactometer).


Results, Conclusions, and Discussions:: Neural recordings from the projection neurons in the insect antennal lobe (analogous to the olfactory bulb in vertebrates) were obtained from the electrode array inserted in the locust brain. By developing an ex vivo recording technique, we were able to keep the insect brain viable for several hours. At the individual neuron level, distinct responses to individual VOCs are shown by the raster and peri-stimulus time histogram plots (Figure 1).


An ex vivo locust brain electrophysiological recording technique is successfully developed for an insect-based bioelectronic sensor. Neural responses from locust projection neurons from this ex vivo set up indicate that this insect-based biosensor can elicit distinct responses to different volatiles which could be used for VOC detection in natural environments (e.g., clinical settings).


While this technique is the first step towards engineering a portable insect-based bioelectronic sensor, several more milestones are needed before viable clinical application. First, exhaled human breath needs to be analyzed and successfully classified by this insect-based biosensor to show proof of concept for clinical use. Second, while we have achieved an ex vivo locust brain preparation, this method still uses a laboratory electrophysiological set-up and more work needs to be done to incorporate a wireless module to digitize and transmit neural spiking responses to a computer for data collection and analysis. Third, a micro electrode array with a high number of channels is needed to record from more projection neurons, thereby increasing the recorded neuronal response for one-shot and real-time classification of gas mixtures. Although more work needs to be done to move this biosensor into clinical settings, the ex vivo locust brain electrophysiological recording technique is novel and to our knowledge this study is the first to achieve this milestone. This work lays the foundation for the creation for a novel portable insect-based biosensor for the detection of lung cancer.






Acknowledgements (Optional): : This work was supported by the National Science Foundation Graduate Research Fellowship.

References (Optional): :