- Email: [email protected]
Functional mapping of cortical areas with optical imaging
www.elsevier.com/locate/ynimg NeuroImage 37 (2007) 440 – 448
Functional mapping of cortical areas with optical imaging K. Holthoff,⁎ E. Sagnak, and O.W. Witte Klinik und Poliklinik für Neurologie, Friedrich-Schiller-Universität Jena, Erlanger Allee 101, 07747 Jena, Germany Received 22 December 2006; revised 24 April 2007; accepted 30 April 2007 Available online 18 May 2007
Sensory areas in mammalian cortex compute sensory inputs of different modalities in order to perceive the environment. Much is known about the anatomical pattern of inter-laminar connections, which form the basis of the computational process. Nevertheless, less is known about the functional relevance of these wiring patterns. We used intrinsic optical signals (IOSs) in vitro to investigate functional properties of inter-laminar connections in cortical brain slices of rat sensory cortex. By electrical stimulation in layer VI, a columnarshaped IOS in all cortical areas was found. We detected different laminar patterns of activation in different cortical areas. In primary sensory areas, like primary visual cortex and primary somatosensory cortex, the peak intensity of IOSs occurred in layer IV, which receives the main thalamic input. In secondary sensory areas, like the secondary visual cortex or the secondary somatosensory cortex, the maximum of IOSs amplitude was shifted to layer II/III. In motor areas, IOS peak amplitude is located in layer II/III. In the hind limb area, considered as amalgam between sensory and motor function, a mixture of the activity patterns observed in primary sensory and a motor area occurred with a peak amplitude in layers II and IV. At different stimulation sites within one cortical area, the shape of columnar IOSs remained very similar, reflecting a canonical architecture of functional micro-circuitry. We conclude that both primary and secondary sensory cortical areas display their characteristic functional activation pattern, regardless of their sensory modalities. © 2007 Elsevier Inc. All rights reserved. Keywords: Intrinsic optical signals; Cortical micro-circuit; Column; Rat; Brain slice
Introduction Whereas much is known about the anatomical connections and specific cytoarchitectonic properties in different cortical areas, our knowledge about the functionally relevant micro-circuitry in
⁎ Corresponding author. Present address: Institut für Neurowissenschaften, TU München, Biedersteinerstr. 29, 80802 München, Germany. Fax: +49 89 41403352. E-mail address: [email protected] (K. Holthoff). Available online on ScienceDirect (www.sciencedirect.com). 1053-8119/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2007.04.059
mammalian cortex is much more limited. In recent years, new insights in functional cortical circuitry have been revealed by optical techniques in vitro (Aizenman et al., 1996; Yuste et al., 1997) and in vivo (Bakin et al., 1996; Bonhoeffer and Grinvald, 1991; Ohki et al., 2005; Stosiek et al., 2003). In primary sensory cortex, the main thalamic input to cortical layer IV represents the first and most effective functional projection in the hierarchy of information processing (Bruno and Sakmann, 2006; Miller et al., 2001; Yuste et al., 1997). The functional projections into the secondary sensory areas have been studied less intensively. The introduction of new optical methods, have greatly enhanced our ability to analyze the functional cortical microcircuitry. Both membrane potential-sensitive and Ca2+-sensitive fluorescent dyes confirmed the columnar shape of neuronal information processing units (Yuste et al., 1992, 1997) already discovered several decades before (Hubel and Wiesel, 1963; Powell and Mountcastle, 1959). The use of intrinsic optical signals (IOSs) in vivo led to a breakthrough in analyzing cortical information processing in sensory systems (Bonhoeffer and Grinvald, 1991; Malonek et al., 1994). In vitro, IOSs reflect activity-induced changes of extracellular space (ECS) volume (Fayuk et al., 2002; Holthoff and Witte, 1996, 1998; Witte et al., 2001). Nevertheless, several lines of evidence suggest that IOSs are strictly coupled to synaptic transmission. Firstly in cortical brain slices, the amplitude of optical signals correlates with the amplitude of extracellular field potentials, which reflect excitatory postsynaptic potentials (Holthoff et al., 1994). Secondly in hippocampus, IOSs occur only in layers in which synaptic transmission takes place (MacVicar and Hochman, 1991). And thirdly, the density of synaptic connections and the amplitude of activation determine the amplitude of IOSs (Dodt et al., 1993; Holthoff and Witte, 1996; MacVicar and Hochman, 1991). Using IOSs we investigated the functional properties of different cytoarchitectonic cortical areas. Specifically, we asked the following questions: • Do different cytoarchitectonic cortical areas display characteristic and therefore distinct activation columns? • Are these patterns independent of the sensory modalities? • Do the activation patterns change with intensity of activation?
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Materials and methods Slice preparation Experiments were done on brain slices of 14-day-old male Wistar rats. After ether anesthesia, animals were decapitated, and the brains were quickly removed and were cooled down to 4 °C in artificial cerebral spinal fluid (ACSF) [containing (in mM) NaCl 124, NaHCO3 26, KCl 3, CaCl2 2, MgSO4 2, NaH2PO4 1.25, and glucose 10, equilibrated with 5% CO2 in O2 to pH 7.4]. Coronal brain slices of 400-μm thickness were cut from occipital and parietal regions using a vibratome (Campden Instruments). To ensure that nerve fibers in the cortex ran parallel to the cutting plane, brain slices from parietal regions were cut with an angle of 20° to the coronal plane. Imaging Intrinsic optical signals (IOSs) were recorded as previously described (Holthoff and Witte, 1996). Brain slices were perfused submerged with warmed ACSF (32 °C) at 3 ml/min and equilibrated for at least 20 min before each experiment. Slices were illuminated in the dark-field configuration of an upright microscope (Axioskop FS, Zeiss) with near-infrared light (750 ± 50 nm). IOSs were recorded at low optical magnification (Plan-Neofluar 2.5×, Zeiss) using a CCD-camera (C2400-77, Hamamatsu). Using the CCDcamera control device (Hamamatsu), the camera signal was contrast enhanced by a factor of 3. The shading correction unit of the camera control device was used to balance the overall brightness of the resulting image. Further image processing procedures were done using a video processing unit (DVS 3000, Hamamatsu, Herrsching). To detect IOSs, representing small changes in brightness of the dark-field illuminated slices, images were background subtracted and digitally enhanced. Therefore, the background intensity was measured before each stimulation by averaging 64 subsequent images. This background intensity image was subsequently subtracted from the current video signal. The resulting difference images were digitally enhanced by a factor of 4 to 8. The achieved video signal was stored on an S-VHS video recorder (Panasonic, Hamburg) and analyzed off-line with NIH-Image Software using a Macintosh Computer equipped with a frame-grabber card (Scion Corporation, Frederick MA). Slices from occipital and parietal region were stimulated successively at different positions in layer VI at the border of the white matter. Each brain slice contained several different cortical areas, which were demarcated after the experiment (see below). The electrical stimulation was composed of a train (50 Hz for 2 s) of single stimuli, each 200 μs long. Beginning at the medial part of the slices, intrinsic optical signals were elicited successively every 500 μm moving laterally. The procedure was repeated going into the opposite direction from lateral to medial parts of the slices and stimulating interstitially of the positions of the first stimulation series. This procedure allowed us to get an estimate of the reproducibility of the optical signals in different cortical areas. Cortical area demarcation In a selected number of slices, the cortical areas were demarcated with two different methods. First, in both occipital (n = 4) and parietal (n = 5) slices cortical area boundaries were determined using the unprocessed microscopic dark-field images. Because the field of view did not cover the entire slices, only parts of the slices were
441
recorded successively and the images of the whole slices were reconstructed off-line using Photoshop (Adobe). Using anatomical atlases (Paxinos and Watson, 1986; Zilles and Wree, 1995), cortical areas were identified by analyzing the relative thickness of the cortical layers and the characteristic appearance of different cortical and subcortical areas. An important criterion for cortical area demarcation was the relative thickness and appearance of layers IV and VI. In the parietal cortex, layer IV showed its largest width in area Par 1. In area Hl, layer IV was clearly identifiable but smaller than in Par 1. In the motor area Fr and the secondary sensory area Par 2, lamina IV was almost not detectable because of its small width. In parietal slices, the different appearance of layer VI in the dark-field images of different cortical areas was also used for demarcation. Area Par 1 could be distinguished from Par 2 and Hl because in this area layer VI appeared darker. A similar procedure was used to demarcate cortical areas in occipital slices. The primary visual cortical area Oc 1 was delimited by its thick layer IV (Fig. 1). In contrast, layer IV was smaller in the secondary visual areas Oc 2L and Oc 2M. The granular (RSG) and agranular (RSA) retrosplenial cortical areas were easily demarcated from neocortical areas by their four-layered structure. The results were verified in a second step by mapping the cortical areas in cresyl-violet-stained slices. In this case, the same slices were fixated, resliced with a kryotome in 50-μm steps, and stained with cresyl violet (Zilles and Wree, 1995). Both methods gave identical results with respect to the localization of cortical area boundaries (Figs. 1A and B). Therefore, in all other experiments only the dark-field images were used to demarcate cortical area boundaries. For better illustration, images of IOSs were smoothed (7 × 7 Gaussian filter, NIH-Image) and were colorcoded. Red colors represent high optical signal amplitude, and blue colors represent low optical signal amplitude. Examples of unfiltered optical data of an occipital and a parietal region are shown in Figs. 4 and 5, respectively. All quantitative measurements were performed on unprocessed optical data. Statistical values are reported as mean ± standard deviation of the mean. Results All intrinsic optical signals (IOSs) elicited in 25 coronal brain slices of 17 animals showed a columnar shape. As described before (Holthoff and Witte, 1996), the IOS time course was slow. Signals peaked within the first 4 s after onset of stimulation and fell off to resting level during the following minute. After stimulating layer VI electrically, changes in IOS could be recorded in layers II to VI as shown before (Holthoff and Witte, 1996, 1998). Layer I was never involved and the signal stayed at pre-stimulus level. We found different laminar patterns of IOSs determined at peak amplitude depending on the cortical area, in which optical signals were elicited. Primary sensory areas In primary visual cortex, IOS amplitude was maximal in layer IV (Figs. 2 and 7A). The maximal IOS intensity in layer IV was clearly demarcated as a separate peak from adjacent layers. Intensity in layer II/III declined from the border of layer IV to layer I. At the edge to layer I, optical signals dropped suddenly to pre-stimulus levels. IOSs in primary somatosensory area (Par 1) showed features similar to those in primary visual cortex (Figs. 2 and 7B). In Par 1,
442
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Fig. 1. Demarcation of cortical areas. Cortical areas of an occipital brain slice were demarcated using the unprocessed dark-field image (A) and the fixed and cresyl-violet-stained slice (B). One typical occipital slice is shown out of five similar experiments. Area borders were determined by analyzing characteristic laminar pattern. Differences between the position of area borders determined by these independent methods were negligible. Calibration bars represent 1 mm. Abbreviations: occipital cortex, area 1 (Oc 1); occipital cortex, area 2, lateral (Oc 2 L); occipital cortex, area 2, medial (Oc 2 M); temporal cortex (Te); retrosplenial agranular cortex (RSA); retrosplenial granular cortex (RSG) (Paxinos and Watson, 1986).
the maximum of IOSs at peak intensity occurred in layer IV. The maximum IOSs in layer IV were not as clearly demarcated from signals in layer II/III as in Oc 1. The intensity maximum in layer IV formed a plateau-like shape and declined in layer II/III before IOS intensity dropped to resting levels at the border to layer I. Some of the IOSs elicited in Par 1 (8 out of 65) showed lateral activation in layer IV beyond the column, probably reflecting intra-laminar projections to adjacent columns (Fig. 2, arrow). Secondary sensory areas In the lateral secondary visual cortex (Oc 2L), the maximum IOS intensity was observed in layer II/III (Fig. 3). Layer IV and upper layer V displayed a homogeneous IOS which declined in lower layer V and layer VI. Similar laminar patterns of IOS occurred in secondary somatosensory cortex (Par 2) (Fig. 3). A remarkable IOS maximum in layer II/III and a second hump in layer V were separated by lower levels of IOS in layer IV. Some of the IOSs in Par 2 (3 out of 38) displayed lateral activation beyond the principle column. In contrast to lateral activation in Par 1 described above, lateral activation in Par 2 occurred in supragranular layer II/III (Fig. 3, arrow). The magnitude of these lateral activations was roughly a factor of five smaller than the peak amplitude in the center of activated column. Motor cortex and hind limb area Activation in frontal cortex (Fr), a motor cortical area, was dominated by an IOS maximum in layer II/III (Fig. 3). A second activation hump in layer V was not as pronounced as in the secondary sensory areas Oc 2L and Par 2. The hind limb area Hl, which is known as an amalgam of sensory and motor function, showed a mixture of features typical for a motor area and a primary
sensory area (Fig. 2). A pronounced IOS maximum in layer II/III was combined with a similar maximum in layer IV. IOS dependence on stimulation strength To compare optical signals from different cortical areas in one slice, we determined whether the shape of optical signals varies with different stimulation strengths. Optical signals were elicited by subsequent stimulation with different stimulation strength at the same position in the slice. As previously described (Holthoff et al., 1994), in all cortical areas signal amplitude increased with increasing stimulation intensity in a linear fashion. Additionally in primary sensory areas (Oc 1 and Par 1), the laminar specificity of optical signal amplitude remained unchanged for all tested stimulation intensities (Fig. 4). On the other hand, in the secondary sensory areas Par 2 and Oc 2M, laminar profiles changed with stimulation intensity, developing the characteristic profiles at higher stimulation intensities (Fig. 5). Variability of IOS pattern To determine the variability of optical signals within individual cortical areas, laminar intensity plots of single optical signals elicited in one cortical area of an individual slice were compared to each other (Fig. 6A). To get an idea about the variability of optical signals between different slices, mean optical signals elicited in the same cortical area but in different slices or animals were compared (Fig. 6B). Both sets of experiments show that variability in the laminar pattern of individual stimulation in a cortical area of an individual slice as well as the differences between the averaged laminar patterns from different slices were small. This is true for all cortical areas we examined. We concluded that characteristic laminar intensity distribution of IOSs in different cortical areas
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
443
Fig. 2. Laminar pattern of intrinsic optical signals of cortical areas in occipital slices. Left: optical signals at peak intensity 4 s after onset of stimulation. Dotted line indicates longitudinal axis of intrinsic optical signal, along which an intensity plot was performed. Calibration bars represent 250 μm. Right: mean intensity plots along longitudinal axes of intrinsic optical signals (mean ± SD). For better comparability, cortical depth was labeled on the ordinate in addition to cortical laminae extension.
were detectable even in individual experiments and were not the result of averaging variable single trials. Discussion Understanding the functional micro-circuitry of the brain is one of the major aims in neuroscience. A lot detailed information is known about the wiring of cortical micro-circuitry, which forms the basis for neuronal computation. Nevertheless, more information about the functional relevance of neuronal connections is crucial to understand how the brain is able to perform the incredible tasks of sensory perception. We chose functional imaging of neuronal activity using intrinsic optical signals (IOSs) in vitro because this technique allows to monitor reliably neuronal activity with high spatial resolution without any damage to the tissue. Therefore, it is possible with this technique to determine activation patterns in different cortical areas by stimulating different locations in an individual brain slice. Origin of IOS It is now well established that activity-induced IOSs in vitro recorded with the techniques used in the present study are caused by
changes in extracellular space volume (Andrew and MacVicar, 1994; Holthoff and Witte, 1996, 1998). Interestingly, only synaptic activity seems to be able to induce changes in IOSs. If synaptic transmission is blocked, either by pharmacological treatment or removal of extracellular calcium, IOSs are not detectable, although antidromically induced activity is still present (Dodt et al., 1996; Holthoff and Witte, 1996). Additionally, there is no correlation between IOS amplitude and cell density. In hippocampal slices, stimulating Schaffer collaterals induces IOSs in stratum radiatum of area CA1 (MacVicar and Hochman, 1991), the region where Schaffer collaterals make synaptic contacts with CA1 pyramidal neurons and where cell density is very low. In contrast, the region with the highest cell body density, the stratum pyramidale, does not show an appreciable signal. Nevertheless after Schaffer collateral activation, high spiking activity of pyramidal cells can be assumed. On the other hand in primary visual cortex after stimulation in layer VI, the peak amplitude of IOSs occurs in layer IV (Fig. 2), the layer with the highest cell density of this area (Braitenberg and Schuetz, 1998). Additionally in neocortical slices, the amplitude of IOSs in vitro are highly related to field potential amplitude, a means of postsynaptic excitation strength (Holthoff et al., 1994). All these data are consistent with the idea that IOSs reflect functional synaptic activity and IOS amplitude is proportional to synaptic density and/or
444
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Fig. 3. Laminar pattern of intrinsic optical signals of cortical areas in parietal slices. Left: optical signals at peak intensity 4 s after onset of stimulation. Calibration bars represent 250 μm. Right: mean intensity plots along longitudinal axes of intrinsic optical signals (mean ± SD and see Fig. 2). For better comparability, cortical depth was labeled on the ordinate in addition to cortical laminae extension.
strength of synaptic activity. Therefore, IOSs can be used to evaluate the functional connectivity in brain slices in vitro.
Primary sensory areas We found that primary sensory areas show a typical pattern of activation with a clear maximum in layer IV (Figs. 2 and 7), regardless of their sensory modality. That is exactly what we would expect because by stimulating layer VI electrically, we excite both thalamocortical and callosal fibers. The main afferents coming from the thalamus to primary visual and somatosensory cortex project to layer IV (Burkhalter, 1989; Garey and Powell, 1971). However, only a small fraction of the excitatory synapses on layer IV spiny cells, the main target for thalamocortical projections, comes from thalamic afferents (Ahmed et al., 1994; Benshalom and White, 1986; Peters and Payne, 1993). Nevertheless, layer IV neurons respond to sensory stimulation with large depolarization (Brecht and Sakmann, 2002; Ferster et al., 1996). Two different mechanisms have been proposed to explain how a sparse thalamic input could nevertheless dominate the activity of its target cells. Firstly, it has been shown that thalamocortical and intra-cortical synapses exhibit different features leading to a nominal stronger impact of the thalamocortical pathway (Bruno and Sakmann, 2006;
Gil et al., 1999). Secondly, according to the so-called amplifier model, an amplification of the thalamic input by recurrent excitatory intra-cortical connections is supposed to enhance the influence of the thalamocortical pathway (Douglas et al., 1995; Lubke et al., 2000; Stratford et al., 1996). For primary sensory areas, we can confirm from our data that projections into layer IV coming via layer VI are stronger than those projecting into other layers. We cannot distinguish, whether this is due to a higher number of functionally active synapses or a stronger activation. We also have evidence for intra-laminar activity in layer IV. A fading activation can be observed in layer IV beyond the activated column (Fig. 2, arrow). It is unlikely that this activation is induced by accidentally exciting an adjacent column. Increasing stimulation intensities would lead to an evenly distributed lateral spread of IOS in both directions and should include all layers homogeneously (Dodt et al., 1993; Holthoff et al., 1994).
Secondary sensory areas The activation patterns in secondary sensory areas are different from those in primary sensory areas. The peak amplitude of IOSs is shifted to layer II/III. It is interesting that both secondary sensory areas considered here, Par II and Oc 2L, show a similar laminar
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
445
Fig. 4. Variability of laminar pattern of intrinsic optical signals elicited with different stimulation strength in primary visual cortex (Oc 1). Optical signals are displayed at peak intensity 4 s after stimulating with medium (A1) and high stimulation intensities (B1). Calibration bars represent 250 μm. Corresponding intensity plots were analyzed along longitudinal axes of optical signals. Plotted are means and standard deviations for medium (A2) and high stimulation intensities (B2).
Fig. 5. Variability of laminar pattern of intrinsic optical signals elicited with different stimulation strength in parietal cortex, area 2 (Par 2). Optical signals are displayed at peak intensity 4 s after stimulating with medium (A1) and high stimulation intensities (B1). Corresponding intensity plots were analyzed along longitudinal axes of optical signals. Plotted are means and standard deviations for medium (A2) and high stimulation intensities (B2).
446
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Fig. 6. Variability of laminar pattern of intrinsic optical signals elicited in parietal cortex, area 1 (Par 1). Intensity plots along longitudinal axes of optical signals are shown. (A) Intensity plots of optical signals elicited in one individual slice at different stimulation sites in cortical area Par 1. (B) Intensity plots of optical signals elicited in cortical area Par 1 in different slices. Each plot represents an average of all signals elicited in the respective area of an individual slice.
activation pattern. It is known that in layer II/III there is a large proportion of collateral projections spanning up to millimeters in length (Burkhalter and Charles, 1990; Kisvarday et al., 1997). Sometimes we saw a fading activation in layer II/III beyond the actually activated column (Fig. 3, arrow). This activation could be
due to long-range cortico-cortical connections in layer II/III. The smaller amplitude of this activation may be due to polysynaptic rather than monosynaptic activation. In the hind limb area, known as an amalgam between sensory and motor function, laminar pattern of IOS resemble this mixture
Fig. 7. Stimulation-induced intrinsic optical signals (IOSs) in individual slices. In subsequent stimulations, IOSs were induced at different stimulation sites in an occipital (A) and a parietal (B) neocortical slice with medium stimulation intensities. IOSs were captured 4 s after begin of stimulation when amplitude peaked. Stimulation sites were marked by white circles. Cortical areas, demarcated with white arrowheads, were mapped using unprocessed dark-field images of the slice, taken before each stimulation. Abbreviations: frontal cortex (Fr); hindlimb (Hl); occipital cortex, area 1 (Oc 1); occipital cortex, area 2, lateral (Oc 2 L); occipital cortex, area 2, medial (Oc 2 M); parietal cortex, area 1 (Par 1); parietal cortex, area 2 (Par 2); temporal cortex (Te); retrosplenial agranular cortex (RSA); retrosplenial granular cortex (RSG) (Paxinos and Watson, 1986).
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
of functionality. IOSs in Hl showed both the characteristic intensity maximum in layer II/III for a motor area and another hump of activation in layer IV characteristic for a primary sensory area. Reliability and reproducibility of IOS The laminar patterns of activation showed excellent reproducibility. In between the same cortical area, the laminar patterns of IOS were very similar. Besides good reproducibility, this reflects that functional architecture in cortical areas seems to be homogeneous. Cortical areas have distinct patterns of activation that change abruptly at the boundaries. We have never seen mixtures of different laminar patterns at the cortical area boundaries suggesting that the borders of cortical areas are very abrupt without smooth transitions. Because the smallest distance between our stimulations was around 250 μm, we cannot exclude that below this resolution transition zones between adjacent cortical areas exist. Functional imaging using intrinsic optical signals can provide important insights in functional interlaminar connectivity in cortex. Using different preparations like thalamocortical or callosal slices, it will be possible to distinguish between different cortical inputs more specifically. Additionally, stimulating cortical layer V or III will lead to a more detailed analysis of functional micro-circuitry in different brain areas. Acknowledgments This work was supported by Sonderforschungsbereich 194 B2. Gratitude is expressed to G. Heide, H.J. Freund, A. Majewska, E.J. Speckmann, and K. Zilles for helpful comments.
References Ahmed, B., Anderson, J.C., Douglas, R.J., Martin, K.A., Nelson, J.C., 1994. Polyneuronal innervation of spiny stellate neurons in cat visual cortex. J. Comp. Neurol. 341, 39–49. Aizenman, C.D., Kirkwood, A., Bear, M.F., 1996. A current source density analysis of evoked responses in slices of adult rat visual cortex: implications for the regulation of long-term potentiation. Cereb. Cortex 6, 751–758. Andrew, R.D., MacVicar, B.A., 1994. Imaging cell volume changes and neuronal excitation in the hippocampal slice. Neuroscience 62, 371–383. Bakin, J.S., Kwon, M.C., Masino, S.A., Weinberger, N.M., Frostig, R.D., 1996. Suprathreshold auditory cortex activation visualized by intrinsic signal optical imaging. Cereb. Cortex 6, 120–130. Benshalom, G., White, E.L., 1986. Quantification of thalamocortical synapses with spiny stellate neurons in layer IVof mouse somatosensory cortex. J. Comp. Neurol. 253, 303–314. Bonhoeffer, T., Grinvald, A., 1991. Iso-orientation domains in cat visual cortex are arranged in pinwheel-like patterns. Nature 353, 429–431. Braitenberg, V., Schuetz, A., 1998. Cortex: Statistics and Geometry of Neuronal Connectivity, 2nd ed. Springer-Verlag, Berlin. Brecht, M., Sakmann, B., 2002. Dynamic representation of whisker deflection by synaptic potentials in spiny stellate and pyramidal cells in the barrels and septa of layer 4 rat somatosensory cortex. J. Physiol. 543, 49–70. Bruno, R.M., Sakmann, B., 2006. Cortex is driven by weak but synchronously active thalamocortical synapses. Science 312, 1622–1627. Burkhalter, A., 1989. Intrinsic connections of rat primary visual cortex: laminar organization of axonal projections. J. Comp. Neurol. 279, 171–186. Burkhalter, A., Charles, V., 1990. Organization of local axon collaterals of
447
efferent projection neurons in rat visual cortex. J. Comp. Neurol. 302, 920–934. Dodt, H.U., Holländer, H., Zieglgaensberger, W., 1993. Visualization of the spread of excitation in brain slices with infrared-darkfield microscopy. Abstr.-Soc. Neurosci. 19, 272. Dodt, H.U., Darcangelo, G., Pestel, E., Zieglgansberger, W., 1996. The spread of excitation in neocortical columns visualized with infrareddarkfield videomicroscopy. NeuroReport 7, 1553–1558. Douglas, R.J., Koch, C., Mahowald, M., Martin, K.A., Suarez, H.H., 1995. Recurrent excitation in neocortical circuits. Science 269, 981–985. Fayuk, D., Aitken, P.G., Somjen, G.G., Turner, D.A., 2002. Two different mechanisms underlie reversible, intrinsic optical signals in rat hippocampal slices. J. Neurophysiol. 87, 1924–1937. Ferster, D., Chung, S., Wheat, H., 1996. Orientation selectivity of thalamic input to simple cells of cat visual cortex. Nature 380, 249–252. Garey, L.J., Powell, T.P., 1971. An experimental study of the termination of the lateral geniculo-cortical pathway in the cat and monkey. Proc. R. Soc. Lond., B Biol. Sci. 179, 41–63. Gil, Z., Connors, B.W., Amitai, Y., 1999. Efficacy of thalamocortical and intracortical synaptic connections: quanta, innervation, and reliability. Neuron 23, 385–397. Holthoff, K., Witte, O.W., 1996. Intrinsic optical signals in rat neocortical slices measured with near-infrared dark-field microscopy reveal changes in extracellular space. J. Neurosci. 16, 2740–2749. Holthoff, K., Witte, O.W., 1998. Intrinsic optical signals in vitro: a tool to measure alterations in extracellular space with two-dimensional resolution. Brain Res. Bull. 47, 649–655. Holthoff, K., Dodt, H.U., Witte, O.W., 1994. Changes in intrinsic optical signal of rat neocortical slices following afferent stimulation. Neurosci. Lett. 180, 227–230. Hubel, D.H., Wiesel, T.N., 1963. Shape and arrangement of columns in cat's striate cortex. J. Physiol. 165, 559–568. Kisvarday, Z.F., Toth, E., Rausch, M., Eysel, U.T., 1997. Orientationspecific relationship between populations of excitatory and inhibitory lateral connections in the visual cortex of the cat. Cereb. Cortex 7, 605–618. Lubke, J., Egger, V., Sakmann, B., Feldmeyer, D., 2000. Columnar organization of dendrites and axons of single and synaptically coupled excitatory spiny neurons in layer 4 of the rat barrel cortex. J. Neurosci. 20, 5300–5311. MacVicar, B.A., Hochman, D., 1991. Imaging of synaptically evoked intrinsic optical signals in hippocampal slices. J. Neurosci. 11, 1458–1469. Malonek, D., Tootell, R.B., Grinvald, A., 1994. Optical imaging reveals the functional architecture of neurons processing shape and motion in owl monkey area MT. Proc. R. Soc. Lond., B Biol. Sci. 258, 109–119. Miller, K.D., Pinto, D.J., Simons, D.J., 2001. Processing in layer 4 of the neocortical circuit: new insights from visual and somatosensory cortex. Curr. Opin. Neurobiol. 11, 488–497. Ohki, K., Chung, S., Ch'ng, Y.H., Kara, P., Reid, R.C., 2005. Functional imaging with cellular resolution reveals precise micro-architecture in visual cortex. Nature 433, 597–603. Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates, 2nd ed. Academic Press. Peters, A., Payne, B.R., 1993. Numerical relationships between geniculocortical afferents and pyramidal cell modules in cat primary visual cortex. Cereb. Cortex 3, 69–78. Powell, T.P., Mountcastle, V.B., 1959. Some aspects of the functional organization of the cortex of the postcentral gyrus of the monkey: a correlation of findings obtained in a single unit analysis with cytoarchitecture. Bull. Johns Hopkins Hosp. 105, 133–162. Stosiek, C., Garaschuk, O., Holthoff, K., Konnerth, A., 2003. In vivo twophoton calcium imaging of neuronal networks. Proc. Natl. Acad. Sci. U. S. A. 100, 7319–7324. Stratford, K.J., Tarczy-Hornoch, K., Martin, K.A., Bannister, N.J., Jack, J.J., 1996. Excitatory synaptic inputs to spiny stellate cells in cat visual cortex. Nature 382, 258–261.
448
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Witte, O.W., Niermann, H., Holthoff, K., 2001. Cell swelling and ion redistribution assessed with intrinsic optical signals. An. Acad. Bras. Cienc. 73, 337–350. Yuste, R., Peinado, A., Katz, L.C., 1992. Neuronal domains in developing neocortex. Science 257, 665–669.
Yuste, R., Tank, D.W., Kleinfeld, D., 1997. Functional study of the rat cortical microcircuitry with voltage-sensitive dye imaging of neocortical slices. Cereb. Cortex 7, 546–558. Zilles, K., Wree, A., 1995. Cortex: areal and laminar structure, In: Paxinos, G. (Ed.), The Rat Nervous System, 2nd ed. Academic Press, pp. 649–685.
Functional mapping of cortical areas with optical imaging K. Holthoff,⁎ E. Sagnak, and O.W. Witte Klinik und Poliklinik für Neurologie, Friedrich-Schiller-Universität Jena, Erlanger Allee 101, 07747 Jena, Germany Received 22 December 2006; revised 24 April 2007; accepted 30 April 2007 Available online 18 May 2007
Sensory areas in mammalian cortex compute sensory inputs of different modalities in order to perceive the environment. Much is known about the anatomical pattern of inter-laminar connections, which form the basis of the computational process. Nevertheless, less is known about the functional relevance of these wiring patterns. We used intrinsic optical signals (IOSs) in vitro to investigate functional properties of inter-laminar connections in cortical brain slices of rat sensory cortex. By electrical stimulation in layer VI, a columnarshaped IOS in all cortical areas was found. We detected different laminar patterns of activation in different cortical areas. In primary sensory areas, like primary visual cortex and primary somatosensory cortex, the peak intensity of IOSs occurred in layer IV, which receives the main thalamic input. In secondary sensory areas, like the secondary visual cortex or the secondary somatosensory cortex, the maximum of IOSs amplitude was shifted to layer II/III. In motor areas, IOS peak amplitude is located in layer II/III. In the hind limb area, considered as amalgam between sensory and motor function, a mixture of the activity patterns observed in primary sensory and a motor area occurred with a peak amplitude in layers II and IV. At different stimulation sites within one cortical area, the shape of columnar IOSs remained very similar, reflecting a canonical architecture of functional micro-circuitry. We conclude that both primary and secondary sensory cortical areas display their characteristic functional activation pattern, regardless of their sensory modalities. © 2007 Elsevier Inc. All rights reserved. Keywords: Intrinsic optical signals; Cortical micro-circuit; Column; Rat; Brain slice
Introduction Whereas much is known about the anatomical connections and specific cytoarchitectonic properties in different cortical areas, our knowledge about the functionally relevant micro-circuitry in
⁎ Corresponding author. Present address: Institut für Neurowissenschaften, TU München, Biedersteinerstr. 29, 80802 München, Germany. Fax: +49 89 41403352. E-mail address: [email protected] (K. Holthoff). Available online on ScienceDirect (www.sciencedirect.com). 1053-8119/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2007.04.059
mammalian cortex is much more limited. In recent years, new insights in functional cortical circuitry have been revealed by optical techniques in vitro (Aizenman et al., 1996; Yuste et al., 1997) and in vivo (Bakin et al., 1996; Bonhoeffer and Grinvald, 1991; Ohki et al., 2005; Stosiek et al., 2003). In primary sensory cortex, the main thalamic input to cortical layer IV represents the first and most effective functional projection in the hierarchy of information processing (Bruno and Sakmann, 2006; Miller et al., 2001; Yuste et al., 1997). The functional projections into the secondary sensory areas have been studied less intensively. The introduction of new optical methods, have greatly enhanced our ability to analyze the functional cortical microcircuitry. Both membrane potential-sensitive and Ca2+-sensitive fluorescent dyes confirmed the columnar shape of neuronal information processing units (Yuste et al., 1992, 1997) already discovered several decades before (Hubel and Wiesel, 1963; Powell and Mountcastle, 1959). The use of intrinsic optical signals (IOSs) in vivo led to a breakthrough in analyzing cortical information processing in sensory systems (Bonhoeffer and Grinvald, 1991; Malonek et al., 1994). In vitro, IOSs reflect activity-induced changes of extracellular space (ECS) volume (Fayuk et al., 2002; Holthoff and Witte, 1996, 1998; Witte et al., 2001). Nevertheless, several lines of evidence suggest that IOSs are strictly coupled to synaptic transmission. Firstly in cortical brain slices, the amplitude of optical signals correlates with the amplitude of extracellular field potentials, which reflect excitatory postsynaptic potentials (Holthoff et al., 1994). Secondly in hippocampus, IOSs occur only in layers in which synaptic transmission takes place (MacVicar and Hochman, 1991). And thirdly, the density of synaptic connections and the amplitude of activation determine the amplitude of IOSs (Dodt et al., 1993; Holthoff and Witte, 1996; MacVicar and Hochman, 1991). Using IOSs we investigated the functional properties of different cytoarchitectonic cortical areas. Specifically, we asked the following questions: • Do different cytoarchitectonic cortical areas display characteristic and therefore distinct activation columns? • Are these patterns independent of the sensory modalities? • Do the activation patterns change with intensity of activation?
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Materials and methods Slice preparation Experiments were done on brain slices of 14-day-old male Wistar rats. After ether anesthesia, animals were decapitated, and the brains were quickly removed and were cooled down to 4 °C in artificial cerebral spinal fluid (ACSF) [containing (in mM) NaCl 124, NaHCO3 26, KCl 3, CaCl2 2, MgSO4 2, NaH2PO4 1.25, and glucose 10, equilibrated with 5% CO2 in O2 to pH 7.4]. Coronal brain slices of 400-μm thickness were cut from occipital and parietal regions using a vibratome (Campden Instruments). To ensure that nerve fibers in the cortex ran parallel to the cutting plane, brain slices from parietal regions were cut with an angle of 20° to the coronal plane. Imaging Intrinsic optical signals (IOSs) were recorded as previously described (Holthoff and Witte, 1996). Brain slices were perfused submerged with warmed ACSF (32 °C) at 3 ml/min and equilibrated for at least 20 min before each experiment. Slices were illuminated in the dark-field configuration of an upright microscope (Axioskop FS, Zeiss) with near-infrared light (750 ± 50 nm). IOSs were recorded at low optical magnification (Plan-Neofluar 2.5×, Zeiss) using a CCD-camera (C2400-77, Hamamatsu). Using the CCDcamera control device (Hamamatsu), the camera signal was contrast enhanced by a factor of 3. The shading correction unit of the camera control device was used to balance the overall brightness of the resulting image. Further image processing procedures were done using a video processing unit (DVS 3000, Hamamatsu, Herrsching). To detect IOSs, representing small changes in brightness of the dark-field illuminated slices, images were background subtracted and digitally enhanced. Therefore, the background intensity was measured before each stimulation by averaging 64 subsequent images. This background intensity image was subsequently subtracted from the current video signal. The resulting difference images were digitally enhanced by a factor of 4 to 8. The achieved video signal was stored on an S-VHS video recorder (Panasonic, Hamburg) and analyzed off-line with NIH-Image Software using a Macintosh Computer equipped with a frame-grabber card (Scion Corporation, Frederick MA). Slices from occipital and parietal region were stimulated successively at different positions in layer VI at the border of the white matter. Each brain slice contained several different cortical areas, which were demarcated after the experiment (see below). The electrical stimulation was composed of a train (50 Hz for 2 s) of single stimuli, each 200 μs long. Beginning at the medial part of the slices, intrinsic optical signals were elicited successively every 500 μm moving laterally. The procedure was repeated going into the opposite direction from lateral to medial parts of the slices and stimulating interstitially of the positions of the first stimulation series. This procedure allowed us to get an estimate of the reproducibility of the optical signals in different cortical areas. Cortical area demarcation In a selected number of slices, the cortical areas were demarcated with two different methods. First, in both occipital (n = 4) and parietal (n = 5) slices cortical area boundaries were determined using the unprocessed microscopic dark-field images. Because the field of view did not cover the entire slices, only parts of the slices were
441
recorded successively and the images of the whole slices were reconstructed off-line using Photoshop (Adobe). Using anatomical atlases (Paxinos and Watson, 1986; Zilles and Wree, 1995), cortical areas were identified by analyzing the relative thickness of the cortical layers and the characteristic appearance of different cortical and subcortical areas. An important criterion for cortical area demarcation was the relative thickness and appearance of layers IV and VI. In the parietal cortex, layer IV showed its largest width in area Par 1. In area Hl, layer IV was clearly identifiable but smaller than in Par 1. In the motor area Fr and the secondary sensory area Par 2, lamina IV was almost not detectable because of its small width. In parietal slices, the different appearance of layer VI in the dark-field images of different cortical areas was also used for demarcation. Area Par 1 could be distinguished from Par 2 and Hl because in this area layer VI appeared darker. A similar procedure was used to demarcate cortical areas in occipital slices. The primary visual cortical area Oc 1 was delimited by its thick layer IV (Fig. 1). In contrast, layer IV was smaller in the secondary visual areas Oc 2L and Oc 2M. The granular (RSG) and agranular (RSA) retrosplenial cortical areas were easily demarcated from neocortical areas by their four-layered structure. The results were verified in a second step by mapping the cortical areas in cresyl-violet-stained slices. In this case, the same slices were fixated, resliced with a kryotome in 50-μm steps, and stained with cresyl violet (Zilles and Wree, 1995). Both methods gave identical results with respect to the localization of cortical area boundaries (Figs. 1A and B). Therefore, in all other experiments only the dark-field images were used to demarcate cortical area boundaries. For better illustration, images of IOSs were smoothed (7 × 7 Gaussian filter, NIH-Image) and were colorcoded. Red colors represent high optical signal amplitude, and blue colors represent low optical signal amplitude. Examples of unfiltered optical data of an occipital and a parietal region are shown in Figs. 4 and 5, respectively. All quantitative measurements were performed on unprocessed optical data. Statistical values are reported as mean ± standard deviation of the mean. Results All intrinsic optical signals (IOSs) elicited in 25 coronal brain slices of 17 animals showed a columnar shape. As described before (Holthoff and Witte, 1996), the IOS time course was slow. Signals peaked within the first 4 s after onset of stimulation and fell off to resting level during the following minute. After stimulating layer VI electrically, changes in IOS could be recorded in layers II to VI as shown before (Holthoff and Witte, 1996, 1998). Layer I was never involved and the signal stayed at pre-stimulus level. We found different laminar patterns of IOSs determined at peak amplitude depending on the cortical area, in which optical signals were elicited. Primary sensory areas In primary visual cortex, IOS amplitude was maximal in layer IV (Figs. 2 and 7A). The maximal IOS intensity in layer IV was clearly demarcated as a separate peak from adjacent layers. Intensity in layer II/III declined from the border of layer IV to layer I. At the edge to layer I, optical signals dropped suddenly to pre-stimulus levels. IOSs in primary somatosensory area (Par 1) showed features similar to those in primary visual cortex (Figs. 2 and 7B). In Par 1,
442
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Fig. 1. Demarcation of cortical areas. Cortical areas of an occipital brain slice were demarcated using the unprocessed dark-field image (A) and the fixed and cresyl-violet-stained slice (B). One typical occipital slice is shown out of five similar experiments. Area borders were determined by analyzing characteristic laminar pattern. Differences between the position of area borders determined by these independent methods were negligible. Calibration bars represent 1 mm. Abbreviations: occipital cortex, area 1 (Oc 1); occipital cortex, area 2, lateral (Oc 2 L); occipital cortex, area 2, medial (Oc 2 M); temporal cortex (Te); retrosplenial agranular cortex (RSA); retrosplenial granular cortex (RSG) (Paxinos and Watson, 1986).
the maximum of IOSs at peak intensity occurred in layer IV. The maximum IOSs in layer IV were not as clearly demarcated from signals in layer II/III as in Oc 1. The intensity maximum in layer IV formed a plateau-like shape and declined in layer II/III before IOS intensity dropped to resting levels at the border to layer I. Some of the IOSs elicited in Par 1 (8 out of 65) showed lateral activation in layer IV beyond the column, probably reflecting intra-laminar projections to adjacent columns (Fig. 2, arrow). Secondary sensory areas In the lateral secondary visual cortex (Oc 2L), the maximum IOS intensity was observed in layer II/III (Fig. 3). Layer IV and upper layer V displayed a homogeneous IOS which declined in lower layer V and layer VI. Similar laminar patterns of IOS occurred in secondary somatosensory cortex (Par 2) (Fig. 3). A remarkable IOS maximum in layer II/III and a second hump in layer V were separated by lower levels of IOS in layer IV. Some of the IOSs in Par 2 (3 out of 38) displayed lateral activation beyond the principle column. In contrast to lateral activation in Par 1 described above, lateral activation in Par 2 occurred in supragranular layer II/III (Fig. 3, arrow). The magnitude of these lateral activations was roughly a factor of five smaller than the peak amplitude in the center of activated column. Motor cortex and hind limb area Activation in frontal cortex (Fr), a motor cortical area, was dominated by an IOS maximum in layer II/III (Fig. 3). A second activation hump in layer V was not as pronounced as in the secondary sensory areas Oc 2L and Par 2. The hind limb area Hl, which is known as an amalgam of sensory and motor function, showed a mixture of features typical for a motor area and a primary
sensory area (Fig. 2). A pronounced IOS maximum in layer II/III was combined with a similar maximum in layer IV. IOS dependence on stimulation strength To compare optical signals from different cortical areas in one slice, we determined whether the shape of optical signals varies with different stimulation strengths. Optical signals were elicited by subsequent stimulation with different stimulation strength at the same position in the slice. As previously described (Holthoff et al., 1994), in all cortical areas signal amplitude increased with increasing stimulation intensity in a linear fashion. Additionally in primary sensory areas (Oc 1 and Par 1), the laminar specificity of optical signal amplitude remained unchanged for all tested stimulation intensities (Fig. 4). On the other hand, in the secondary sensory areas Par 2 and Oc 2M, laminar profiles changed with stimulation intensity, developing the characteristic profiles at higher stimulation intensities (Fig. 5). Variability of IOS pattern To determine the variability of optical signals within individual cortical areas, laminar intensity plots of single optical signals elicited in one cortical area of an individual slice were compared to each other (Fig. 6A). To get an idea about the variability of optical signals between different slices, mean optical signals elicited in the same cortical area but in different slices or animals were compared (Fig. 6B). Both sets of experiments show that variability in the laminar pattern of individual stimulation in a cortical area of an individual slice as well as the differences between the averaged laminar patterns from different slices were small. This is true for all cortical areas we examined. We concluded that characteristic laminar intensity distribution of IOSs in different cortical areas
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
443
Fig. 2. Laminar pattern of intrinsic optical signals of cortical areas in occipital slices. Left: optical signals at peak intensity 4 s after onset of stimulation. Dotted line indicates longitudinal axis of intrinsic optical signal, along which an intensity plot was performed. Calibration bars represent 250 μm. Right: mean intensity plots along longitudinal axes of intrinsic optical signals (mean ± SD). For better comparability, cortical depth was labeled on the ordinate in addition to cortical laminae extension.
were detectable even in individual experiments and were not the result of averaging variable single trials. Discussion Understanding the functional micro-circuitry of the brain is one of the major aims in neuroscience. A lot detailed information is known about the wiring of cortical micro-circuitry, which forms the basis for neuronal computation. Nevertheless, more information about the functional relevance of neuronal connections is crucial to understand how the brain is able to perform the incredible tasks of sensory perception. We chose functional imaging of neuronal activity using intrinsic optical signals (IOSs) in vitro because this technique allows to monitor reliably neuronal activity with high spatial resolution without any damage to the tissue. Therefore, it is possible with this technique to determine activation patterns in different cortical areas by stimulating different locations in an individual brain slice. Origin of IOS It is now well established that activity-induced IOSs in vitro recorded with the techniques used in the present study are caused by
changes in extracellular space volume (Andrew and MacVicar, 1994; Holthoff and Witte, 1996, 1998). Interestingly, only synaptic activity seems to be able to induce changes in IOSs. If synaptic transmission is blocked, either by pharmacological treatment or removal of extracellular calcium, IOSs are not detectable, although antidromically induced activity is still present (Dodt et al., 1996; Holthoff and Witte, 1996). Additionally, there is no correlation between IOS amplitude and cell density. In hippocampal slices, stimulating Schaffer collaterals induces IOSs in stratum radiatum of area CA1 (MacVicar and Hochman, 1991), the region where Schaffer collaterals make synaptic contacts with CA1 pyramidal neurons and where cell density is very low. In contrast, the region with the highest cell body density, the stratum pyramidale, does not show an appreciable signal. Nevertheless after Schaffer collateral activation, high spiking activity of pyramidal cells can be assumed. On the other hand in primary visual cortex after stimulation in layer VI, the peak amplitude of IOSs occurs in layer IV (Fig. 2), the layer with the highest cell density of this area (Braitenberg and Schuetz, 1998). Additionally in neocortical slices, the amplitude of IOSs in vitro are highly related to field potential amplitude, a means of postsynaptic excitation strength (Holthoff et al., 1994). All these data are consistent with the idea that IOSs reflect functional synaptic activity and IOS amplitude is proportional to synaptic density and/or
444
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Fig. 3. Laminar pattern of intrinsic optical signals of cortical areas in parietal slices. Left: optical signals at peak intensity 4 s after onset of stimulation. Calibration bars represent 250 μm. Right: mean intensity plots along longitudinal axes of intrinsic optical signals (mean ± SD and see Fig. 2). For better comparability, cortical depth was labeled on the ordinate in addition to cortical laminae extension.
strength of synaptic activity. Therefore, IOSs can be used to evaluate the functional connectivity in brain slices in vitro.
Primary sensory areas We found that primary sensory areas show a typical pattern of activation with a clear maximum in layer IV (Figs. 2 and 7), regardless of their sensory modality. That is exactly what we would expect because by stimulating layer VI electrically, we excite both thalamocortical and callosal fibers. The main afferents coming from the thalamus to primary visual and somatosensory cortex project to layer IV (Burkhalter, 1989; Garey and Powell, 1971). However, only a small fraction of the excitatory synapses on layer IV spiny cells, the main target for thalamocortical projections, comes from thalamic afferents (Ahmed et al., 1994; Benshalom and White, 1986; Peters and Payne, 1993). Nevertheless, layer IV neurons respond to sensory stimulation with large depolarization (Brecht and Sakmann, 2002; Ferster et al., 1996). Two different mechanisms have been proposed to explain how a sparse thalamic input could nevertheless dominate the activity of its target cells. Firstly, it has been shown that thalamocortical and intra-cortical synapses exhibit different features leading to a nominal stronger impact of the thalamocortical pathway (Bruno and Sakmann, 2006;
Gil et al., 1999). Secondly, according to the so-called amplifier model, an amplification of the thalamic input by recurrent excitatory intra-cortical connections is supposed to enhance the influence of the thalamocortical pathway (Douglas et al., 1995; Lubke et al., 2000; Stratford et al., 1996). For primary sensory areas, we can confirm from our data that projections into layer IV coming via layer VI are stronger than those projecting into other layers. We cannot distinguish, whether this is due to a higher number of functionally active synapses or a stronger activation. We also have evidence for intra-laminar activity in layer IV. A fading activation can be observed in layer IV beyond the activated column (Fig. 2, arrow). It is unlikely that this activation is induced by accidentally exciting an adjacent column. Increasing stimulation intensities would lead to an evenly distributed lateral spread of IOS in both directions and should include all layers homogeneously (Dodt et al., 1993; Holthoff et al., 1994).
Secondary sensory areas The activation patterns in secondary sensory areas are different from those in primary sensory areas. The peak amplitude of IOSs is shifted to layer II/III. It is interesting that both secondary sensory areas considered here, Par II and Oc 2L, show a similar laminar
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
445
Fig. 4. Variability of laminar pattern of intrinsic optical signals elicited with different stimulation strength in primary visual cortex (Oc 1). Optical signals are displayed at peak intensity 4 s after stimulating with medium (A1) and high stimulation intensities (B1). Calibration bars represent 250 μm. Corresponding intensity plots were analyzed along longitudinal axes of optical signals. Plotted are means and standard deviations for medium (A2) and high stimulation intensities (B2).
Fig. 5. Variability of laminar pattern of intrinsic optical signals elicited with different stimulation strength in parietal cortex, area 2 (Par 2). Optical signals are displayed at peak intensity 4 s after stimulating with medium (A1) and high stimulation intensities (B1). Corresponding intensity plots were analyzed along longitudinal axes of optical signals. Plotted are means and standard deviations for medium (A2) and high stimulation intensities (B2).
446
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Fig. 6. Variability of laminar pattern of intrinsic optical signals elicited in parietal cortex, area 1 (Par 1). Intensity plots along longitudinal axes of optical signals are shown. (A) Intensity plots of optical signals elicited in one individual slice at different stimulation sites in cortical area Par 1. (B) Intensity plots of optical signals elicited in cortical area Par 1 in different slices. Each plot represents an average of all signals elicited in the respective area of an individual slice.
activation pattern. It is known that in layer II/III there is a large proportion of collateral projections spanning up to millimeters in length (Burkhalter and Charles, 1990; Kisvarday et al., 1997). Sometimes we saw a fading activation in layer II/III beyond the actually activated column (Fig. 3, arrow). This activation could be
due to long-range cortico-cortical connections in layer II/III. The smaller amplitude of this activation may be due to polysynaptic rather than monosynaptic activation. In the hind limb area, known as an amalgam between sensory and motor function, laminar pattern of IOS resemble this mixture
Fig. 7. Stimulation-induced intrinsic optical signals (IOSs) in individual slices. In subsequent stimulations, IOSs were induced at different stimulation sites in an occipital (A) and a parietal (B) neocortical slice with medium stimulation intensities. IOSs were captured 4 s after begin of stimulation when amplitude peaked. Stimulation sites were marked by white circles. Cortical areas, demarcated with white arrowheads, were mapped using unprocessed dark-field images of the slice, taken before each stimulation. Abbreviations: frontal cortex (Fr); hindlimb (Hl); occipital cortex, area 1 (Oc 1); occipital cortex, area 2, lateral (Oc 2 L); occipital cortex, area 2, medial (Oc 2 M); parietal cortex, area 1 (Par 1); parietal cortex, area 2 (Par 2); temporal cortex (Te); retrosplenial agranular cortex (RSA); retrosplenial granular cortex (RSG) (Paxinos and Watson, 1986).
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
of functionality. IOSs in Hl showed both the characteristic intensity maximum in layer II/III for a motor area and another hump of activation in layer IV characteristic for a primary sensory area. Reliability and reproducibility of IOS The laminar patterns of activation showed excellent reproducibility. In between the same cortical area, the laminar patterns of IOS were very similar. Besides good reproducibility, this reflects that functional architecture in cortical areas seems to be homogeneous. Cortical areas have distinct patterns of activation that change abruptly at the boundaries. We have never seen mixtures of different laminar patterns at the cortical area boundaries suggesting that the borders of cortical areas are very abrupt without smooth transitions. Because the smallest distance between our stimulations was around 250 μm, we cannot exclude that below this resolution transition zones between adjacent cortical areas exist. Functional imaging using intrinsic optical signals can provide important insights in functional interlaminar connectivity in cortex. Using different preparations like thalamocortical or callosal slices, it will be possible to distinguish between different cortical inputs more specifically. Additionally, stimulating cortical layer V or III will lead to a more detailed analysis of functional micro-circuitry in different brain areas. Acknowledgments This work was supported by Sonderforschungsbereich 194 B2. Gratitude is expressed to G. Heide, H.J. Freund, A. Majewska, E.J. Speckmann, and K. Zilles for helpful comments.
References Ahmed, B., Anderson, J.C., Douglas, R.J., Martin, K.A., Nelson, J.C., 1994. Polyneuronal innervation of spiny stellate neurons in cat visual cortex. J. Comp. Neurol. 341, 39–49. Aizenman, C.D., Kirkwood, A., Bear, M.F., 1996. A current source density analysis of evoked responses in slices of adult rat visual cortex: implications for the regulation of long-term potentiation. Cereb. Cortex 6, 751–758. Andrew, R.D., MacVicar, B.A., 1994. Imaging cell volume changes and neuronal excitation in the hippocampal slice. Neuroscience 62, 371–383. Bakin, J.S., Kwon, M.C., Masino, S.A., Weinberger, N.M., Frostig, R.D., 1996. Suprathreshold auditory cortex activation visualized by intrinsic signal optical imaging. Cereb. Cortex 6, 120–130. Benshalom, G., White, E.L., 1986. Quantification of thalamocortical synapses with spiny stellate neurons in layer IVof mouse somatosensory cortex. J. Comp. Neurol. 253, 303–314. Bonhoeffer, T., Grinvald, A., 1991. Iso-orientation domains in cat visual cortex are arranged in pinwheel-like patterns. Nature 353, 429–431. Braitenberg, V., Schuetz, A., 1998. Cortex: Statistics and Geometry of Neuronal Connectivity, 2nd ed. Springer-Verlag, Berlin. Brecht, M., Sakmann, B., 2002. Dynamic representation of whisker deflection by synaptic potentials in spiny stellate and pyramidal cells in the barrels and septa of layer 4 rat somatosensory cortex. J. Physiol. 543, 49–70. Bruno, R.M., Sakmann, B., 2006. Cortex is driven by weak but synchronously active thalamocortical synapses. Science 312, 1622–1627. Burkhalter, A., 1989. Intrinsic connections of rat primary visual cortex: laminar organization of axonal projections. J. Comp. Neurol. 279, 171–186. Burkhalter, A., Charles, V., 1990. Organization of local axon collaterals of
447
efferent projection neurons in rat visual cortex. J. Comp. Neurol. 302, 920–934. Dodt, H.U., Holländer, H., Zieglgaensberger, W., 1993. Visualization of the spread of excitation in brain slices with infrared-darkfield microscopy. Abstr.-Soc. Neurosci. 19, 272. Dodt, H.U., Darcangelo, G., Pestel, E., Zieglgansberger, W., 1996. The spread of excitation in neocortical columns visualized with infrareddarkfield videomicroscopy. NeuroReport 7, 1553–1558. Douglas, R.J., Koch, C., Mahowald, M., Martin, K.A., Suarez, H.H., 1995. Recurrent excitation in neocortical circuits. Science 269, 981–985. Fayuk, D., Aitken, P.G., Somjen, G.G., Turner, D.A., 2002. Two different mechanisms underlie reversible, intrinsic optical signals in rat hippocampal slices. J. Neurophysiol. 87, 1924–1937. Ferster, D., Chung, S., Wheat, H., 1996. Orientation selectivity of thalamic input to simple cells of cat visual cortex. Nature 380, 249–252. Garey, L.J., Powell, T.P., 1971. An experimental study of the termination of the lateral geniculo-cortical pathway in the cat and monkey. Proc. R. Soc. Lond., B Biol. Sci. 179, 41–63. Gil, Z., Connors, B.W., Amitai, Y., 1999. Efficacy of thalamocortical and intracortical synaptic connections: quanta, innervation, and reliability. Neuron 23, 385–397. Holthoff, K., Witte, O.W., 1996. Intrinsic optical signals in rat neocortical slices measured with near-infrared dark-field microscopy reveal changes in extracellular space. J. Neurosci. 16, 2740–2749. Holthoff, K., Witte, O.W., 1998. Intrinsic optical signals in vitro: a tool to measure alterations in extracellular space with two-dimensional resolution. Brain Res. Bull. 47, 649–655. Holthoff, K., Dodt, H.U., Witte, O.W., 1994. Changes in intrinsic optical signal of rat neocortical slices following afferent stimulation. Neurosci. Lett. 180, 227–230. Hubel, D.H., Wiesel, T.N., 1963. Shape and arrangement of columns in cat's striate cortex. J. Physiol. 165, 559–568. Kisvarday, Z.F., Toth, E., Rausch, M., Eysel, U.T., 1997. Orientationspecific relationship between populations of excitatory and inhibitory lateral connections in the visual cortex of the cat. Cereb. Cortex 7, 605–618. Lubke, J., Egger, V., Sakmann, B., Feldmeyer, D., 2000. Columnar organization of dendrites and axons of single and synaptically coupled excitatory spiny neurons in layer 4 of the rat barrel cortex. J. Neurosci. 20, 5300–5311. MacVicar, B.A., Hochman, D., 1991. Imaging of synaptically evoked intrinsic optical signals in hippocampal slices. J. Neurosci. 11, 1458–1469. Malonek, D., Tootell, R.B., Grinvald, A., 1994. Optical imaging reveals the functional architecture of neurons processing shape and motion in owl monkey area MT. Proc. R. Soc. Lond., B Biol. Sci. 258, 109–119. Miller, K.D., Pinto, D.J., Simons, D.J., 2001. Processing in layer 4 of the neocortical circuit: new insights from visual and somatosensory cortex. Curr. Opin. Neurobiol. 11, 488–497. Ohki, K., Chung, S., Ch'ng, Y.H., Kara, P., Reid, R.C., 2005. Functional imaging with cellular resolution reveals precise micro-architecture in visual cortex. Nature 433, 597–603. Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates, 2nd ed. Academic Press. Peters, A., Payne, B.R., 1993. Numerical relationships between geniculocortical afferents and pyramidal cell modules in cat primary visual cortex. Cereb. Cortex 3, 69–78. Powell, T.P., Mountcastle, V.B., 1959. Some aspects of the functional organization of the cortex of the postcentral gyrus of the monkey: a correlation of findings obtained in a single unit analysis with cytoarchitecture. Bull. Johns Hopkins Hosp. 105, 133–162. Stosiek, C., Garaschuk, O., Holthoff, K., Konnerth, A., 2003. In vivo twophoton calcium imaging of neuronal networks. Proc. Natl. Acad. Sci. U. S. A. 100, 7319–7324. Stratford, K.J., Tarczy-Hornoch, K., Martin, K.A., Bannister, N.J., Jack, J.J., 1996. Excitatory synaptic inputs to spiny stellate cells in cat visual cortex. Nature 382, 258–261.
448
K. Holthoff et al. / NeuroImage 37 (2007) 440–448
Witte, O.W., Niermann, H., Holthoff, K., 2001. Cell swelling and ion redistribution assessed with intrinsic optical signals. An. Acad. Bras. Cienc. 73, 337–350. Yuste, R., Peinado, A., Katz, L.C., 1992. Neuronal domains in developing neocortex. Science 257, 665–669.
Yuste, R., Tank, D.W., Kleinfeld, D., 1997. Functional study of the rat cortical microcircuitry with voltage-sensitive dye imaging of neocortical slices. Cereb. Cortex 7, 546–558. Zilles, K., Wree, A., 1995. Cortex: areal and laminar structure, In: Paxinos, G. (Ed.), The Rat Nervous System, 2nd ed. Academic Press, pp. 649–685.