Functional imaging of primary visual cortex using flavoprotein autofluorescence - PubMed (original) (raw)

Comparative Study

Functional imaging of primary visual cortex using flavoprotein autofluorescence

T Robert Husson et al. J Neurosci. 2007.

Abstract

Neuronal autofluorescence, which results from the oxidation of flavoproteins in the electron transport chain, has recently been used to map cortical responses to sensory stimuli. This approach could represent a substantial improvement over other optical imaging methods because it is a direct (i.e., nonhemodynamic) measure of neuronal metabolism. However, its application to functional imaging has been limited because strong responses have been reported only in rodents. In this study, we demonstrate that autofluorescence imaging (AFI) can be used to map the functional organization of primary visual cortex in both mouse and cat. In cat area 17, orientation preference maps generated by AFI had the classic pinwheel structure and matched those generated by intrinsic signal imaging in the same imaged field. The spatiotemporal profile of the autofluorescence signal had several advantages over intrinsic signal imaging, including spatially restricted fluorescence throughout its response duration, reduced susceptibility to vascular artifacts, an improved spatial response profile, and a faster time course. These results indicate that AFI is a robust and useful measure of large-scale cortical activity patterns in visual mammals.

PubMed Disclaimer

Figures

Figure 1.

Figure 1.

Responses in mouse primary visual cortex to a drifting horizontal bar. A, Mouse cortex imaged through the skull using green light to highlight the cortical vascular pattern. The dashed blue oval shows the approximate boundary of primary visual cortex. The blue line shows the transect along which the measurements shown in F and G were made. B, The pattern of flavoprotein autofluorescence in mouse V1 generated in response to a horizontal bar stimulus near the top of the visual field. Bright areas are active. C, The pattern of intrinsic signals in the same imaged field in response to the same bar stimulus. Dark areas are active. Images in B and C were linearly contrast stretched to fill the dynamic range. D, The retinotopic organization of mouse V1 mapped using autofluorescence imaging. The color bar shows the mapping of the stimulus monitor onto the cortical surface. E, The same retinotopic organization mapped with intrinsic signal imaging. F, Relative response strength measured along the path shown in A. Responses are normalized to the peak along the same transect. G, Phase of responses as a function of distance. The phase represents the retinotopic location that best activated a given point in the map. H, Phase of responses for three identical bar stimuli in the same mouse. I, Phase of responses for three identical bar stimuli in three other mice. The traces have been shifted in the _y_-axis to demonstrate slope.

Figure 2.

Figure 2.

Single-condition responses in cat primary visual cortex imaged by flavoprotein autofluorescence. A, Fluorescence changes in response to a horizontal square-wave grating (inset). Bright areas correspond to increases in activity and therefore show 0° iso-orientation domains. B, Fluorescence changes in response to a vertical grating; bright regions show 90° iso-orientation domains. X's mark identical locations in A and B.

Figure 3.

Figure 3.

Maps of orientation preference in cat primary visual cortex generated using AFI and ISI. A, Maps of orientation preference in cat area 17 generated with AFI in the same field shown in Figure 2. B, An AFI orientation map in one hemisphere from a different animal. The imaged field has been cropped to show the region used to compare orientation preference between AFI and ISI (“templated region”). C, The same field as A mapped with ISI. D, The same field as B mapped with ISI. E, Comparison of orientation preference in the images shown in A and C. Plotted is the distribution of pixels with a given difference in orientation preference measured by AFI and by ISI (solid line). For comparison, the same distribution is plotted after shuffling the ISI map (dashed line; average distribution over 100 random shuffles). F, Comparison of orientation preference in the images shown in B and D. The insets show cumulative distribution plots for orientation difference for the measured (solid line) and shuffled (dashed line) orientation maps.

Figure 4.

Figure 4.

Visually driven response time courses in cat area 17. A, Fluorescence intensity changes during and after a 6 s visual stimulus (black bar shows the duration of the stimulus). Responses are plotted for a pixel in a horizontal-preferring orientation domain. The visual stimulus consisted of either a horizontal (thick line) or vertical (thin line) square-wave grating drifting through the visual field. The baseline intensity in the first frame of each stimulus presentation was subtracted. B, Reflectance changes at 610 nm (ISI) at a different location than the traces in A, but that had the same orientation preference. Response amplitudes have been inverted such that increases in absorbance are upward deflections. C, Plots of autofluorescence changes from the same point as in A, but after spatial high-pass filtering the images. D, Plots of reflectance changes from the same point as in B after spatial high-pass filtering the images. The traces show the mean response over 15 trials, and error bars are SEM.

Figure 5.

Figure 5.

Vascular artifacts in ISI and AFI functional maps. A, The vascular pattern of the imaged field (same field as Fig. 2). The image was high-pass filtered and contrast stretched to better visualize blood vessels. B, The SD of fluorescent light intensity in AFI images generated in response to four oriented square-wave gratings. The brighter a pixel, the greater is its SD in response to the stimuli. The scale bar shows the range of SDs in blank-normalized images, multiplied by 10,000; the scale bar is applicable to B and C. C, The SD of reflected light intensity in intrinsic signal images generated in response to four oriented square-wave gratings. D, The same image as in B, but SDs have been log-scaled for better visualization. E, The same image as in C, but log-scaled. The scale bar is applicable to both D and E.

Figure 6.

Figure 6.

Spatial correlations in ISI and AFI orientation responses. A, B, Two-dimensional autocorrelograms averaged from three AFI (A) and three ISI (B) data sets (single-condition images were normalized but not spatially filtered). The color bar applies to A and B, and shows correlation levels from 0 (uncorrelated) to 1 (perfectly correlated). C, A plot of one transect in the spatial autocorrelogram (diagonal line in A). D, A plot of the transect through the unfiltered ISI autocorrelogram. Error bars are SEM. E, The same transects after images were high-pass filtered (AFI data in blue; ISI data in red). F, Difference in orientation preferences (in degrees) between an orientation map generated from a minimally filtered set of single-condition images (high-pass cutoff frequency, 0.185 mm−1) and an orientation map generated from the same images high-pass filtered with different spatial filter cutoffs.

Figure 7.

Figure 7.

AFI and absorption signal time courses in response to brief electrical stimulation in mouse. A–D, Time course of a single pixel in mouse visual cortex in response to electrical stimulation demonstrating autofluorescence signal (420–490 nm excitation filter; 520+ nm emission filter) (A), blue absorption signal (420–490 nm excitation filter, no emission filter) (B), green absorption signal (530–550 nm excitation filter; no emission filter) (C), and red absorption signal (600–620 nm excitation filter; no emission filter) (D). Each time course is from a single pixel and is the average of 22 cycles of stimulation. Error bars are SEM.

Figure 8.

Figure 8.

AFI and absorption signal time courses in response to brief electrical stimulation in cat. A–D, Time course of a single pixel in cat visual cortex in response to electrical stimulation demonstrating autofluorescence signal (A), blue absorption signal (B), green absorption signal (C), and red absorption signal (D). Each time course is from a single pixel and is the average of 22 cycles of stimulation. Error bars are SEM.

Figure 9.

Figure 9.

Visually and electrically driven AFI and absorption signal profiles. A, B, Time courses for the autofluorescence signal and potential intrinsic signal contaminants (all of which were inverted for direct comparison) for electrical (A) and visual (B) stimulation. Each trace is the average response over a 3 × 3 pixel region and is averaged over 22 cycles of stimulation.

Similar articles

Cited by

References

    1. Berwick J, Johnston D, Jones M, Martindale J, Redgrave P, McLoughlin N, Schiessl I, Mayhew JE. Neurovascular coupling investigated with two-dimensional optical imaging spectroscopy in rat whisker barrel cortex. Eur J Neurosci. 2005;22:1655–1666. - PubMed
    1. Blasdel GG, Salama G. Voltage-sensitive dyes reveal a modular organization in monkey striate cortex. Nature. 1986;321:579–585. - PubMed
    1. Bonhoeffer T, Grinvald A. Iso-orientation domains in cat visual cortex are arranged in pinwheel-like patterns. Nature. 1991;353:429–431. - PubMed
    1. Bonhoeffer T, Grinvald A. The layout of iso-orientation domains in area 18 of cat visual cortex: optical imaging reveals a pinwheel-like organization. J Neurosci. 1993a;13:4157–4180. - PMC - PubMed
    1. Bonhoeffer T, Grinvald A. Optical imaging of the functional architecture in cat visual cortex: the layout of direction and orientation domains. Adv Exp Med Biol. 1993b;333:57–69. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources