Principle of the FRET-based ATP sensor, after Zhang et al., 2018. When not bound to ATP, the mApple and Clover fluorochromes are physically far apart. Clover is a form of green fluorescent protein that is excited by blue light and emits green light. mApple is a form of red fluorescent protein that is excited by green light and emits red light. The blue excitation light used for Clover produces only minimal excitation of mApple. However, when ATP is bound, the two fluorescent portions of the sensor are brought close together, permitting emission from Clover to directly excite mApple by FRET. Hence, the amount of red emission from the sensor protein, due to blue excitation, is an indirect measurement of the fraction of protein that is bound to ATP. As a control, excitation/emission of the Clover or mApple portions can detect the sensor protein independent of its interaction with ATP. In our hands, detection of mApple produced a much higher signal-to-noise ratio because of lower red autofluorescence in the gut. B. Image acquisition and analysis. The anterior gut was imaged twice with a 63x objective. The first image used direct excitation and emission of mApple using a TRITC filter set to assess where the sensor protein was present. The second image was obtained using a FRET filter set. This allows excitation of Clover, which produces green emission that excites mApple which is detected as red emission. Because the FRET emissions are of a much lower intensity, the image shown here was enhanced for contrast. Using a Python script, the TRITC image was used to identify 100x100 pixel blocks that contain the sensor protein, and hence define a region of interest (ROI). Pixel values were obtained in the red channel for the TRITC and FRET images across the ROI to generate an average red pixel value ratio. Abbreviations: ph, pharynx;
int1, first intestinal ring. C. Histograms showing the data for control and test cases. Error bars denote SEM. (a.u. = arbitrary units)