Development of a MR-visible compound for tracing neuroanatomical connections in vivo

Abstract

Traditional studies of neuroanatomical connections require injection of tracer compounds into living brains, then histology of the postmortem tissue. Here, we describe and validate a compound that reveals neuronal connections in vivo, using MRI. The classic anatomical tracer CTB (cholera-toxin subunit-B) was conjugated with a gadolinium-chelate to form GdDOTA-CTB. GdDOTA-CTB was injected into the primary somatosensory cortex (S1) or the olfactory pathway of rats. High-resolution MR images were collected at a range of time points at 11.7T and 7T. The transported GdDOTA-CTB was visible for at least 1 month post-injection, clearing within 2 months. Control injections of non-conjugated GdDOTA into S1 were not transported and cleared within 1-2 days. Control injections of Gd-Albumin were not transported either, clearing within 7 days. These MR results were verified by classic immunohistochemical staining for CTB, in the same animals. The GdDOTA-CTB neuronal transport was target specific, monosynaptic, stable for several weeks, and reproducible.

Figure 1. General paradigm

A) Typical example of an intra-cortical injection site. GdDOTA-CTB was injected into the forepaw representation of S1. B) Example of a brain section reacted for CO activity, showing the location and structure of somatosensory thalamic nuclei known to have direct connections with S1. At this anterior-posterior level, these nuclei include the VPL, Rt, Po. C) MR images show clear and consistent enhancement in the presumed thalamic targets, ipsilateral to the injection site (red arrows). Examples are shown from two different MRI sequences: T1-IR (left), and T1-W (right). MR images were acquired 7 days post-injection. Related data are shown in Figs. S1 and S2.

Figure 2. Continuity of labeling across thalamic targets

Each panel shows the enhancement of MR signals in presumptive VPL and Po across a series of slices 100 μm thick, spaced 200 μm apart. Rt and VM are included at the margins of this slice series. A) The average of 24 T1-IR scans, acquired during one scan session, from a single ex vivo case. B) An average of 9 T1-W scans in vivo, across 3 scan sessions. C) The same slices as in panel B, after normalization, co-registration and subtraction of baseline images. Voxels showing significant enhancement (p < 0.002, uncorrected) are shown in red. Intrinsic midline enhancement is the fornix, a white matter tract.

Figure 3. Spatial profile of MR enhancement in thalamic target zones

Images were acquired 7–30 days following GdDOTA-CTB injections into right S1, in three animals. In all cases, clear enhancements were found in VPL (white arrows); in some cases the slice plane also included Rt. In each panel, the slice of interest is shown at the top, and the bottom shows the corresponding measurements from mirror-reflected lines passing through the ipsilateral (red) vs. contralateral (green) thalamus. A, B) Data acquired using T1-W sequences. C, D) Data acquired using T1-IR sequences. In all cases, MR image intensity was enhanced in ipsilateral VPL compared to contralateral VPL. In each section, the MR signal intensity was sampled along a line (300 μm width) extending from the midline of the brain toward the internal capsule, in both directions.

Figure 4. Transport dynamics

A) Following GdDOTA-CTB injections into S1, the enhancement in VPL remained stable from 7 days through at least 1 month after injection. T1-IR images are illustrated. B) Detailed time course of MR signal changes in VPL, averaged from T1-W images (n=8). Most of the enhancement occurred between 5–7 days following injection. The enhancement remained in the thalamus for up to 1 month, and cleared by 2 months post-injection. Group data are presented as a mean ± SEM. Related data are shown in Figs. S4 and S5.

Figure 5. Histology of GdDOTA-CTB

Following S1 injections, putative GdDOTA-CTB transport into the thalamic nuclei was verified with histological staining for CTB. A) MSME (T1-W) images show the distinctive semicircular-shaped enhancement in the forepaw representation of VPL. B) In the corresponding location, a histological section verified the transport zones by CTB staining. The location of MR enhancement (from panel A) is outlined with reddish-brown dashed lines, and the border of VPL is outlined with shorter black dashed lines. The labeled terminals are shown at higher magnification in the inset to the right (panel C). D) CTB-labeled cell bodies and sparse presynatpic terminals in VPL (left), and labeled presynaptic terminals in Rt (right). In other sections, labeled terminals were also found in VPL, but cell bodies were not labeled anywhere in Rt. As described in the literature, we found that presynaptic labeling in VPL was regionally variable. Here the cell bodies were more prominently labeled, although presynaptic terminals are also evident. MR images were taken 7–10 days post-injection, and histological processing was done immediately thereafter. Scale bar = 0.1 mm. Overall, the GdDOTA-CTB confirms previous tracer studies of CTB, showing that the projections of S1 terminate in VPL, with collaterals terminating in Rt.

Figure 6. GdDOTA-CTB labeling of intrinsic cortical connections

Following cortical injection of GdDOTA-CTB into S1, T1-W (A–C) images revealed bands of MR enhancement in the middle layer(s) of cortex (arrows), parallel to the cortical surface. These enhanced bands were especially apparent when the injections involved the superficial cortical layers (e.g. A–C). C) Statistical data confirmed this observation from the corresponding slice at the same magnification in different animals, indicating extrinsic connections linking areas S1 and M1, as well as intrinsic connections linking different body representations within S1 (red arrows in A–B; black arrows in C). Statistic p value < 0.05, uncorrected. D–F) At higher magnification, CTB-labeled pyramidal neurons were found in layers III and V, with clear axonal processes and apical dendrites. This suggests that the enhancements in layers III and V arise from the apical dendrites of the pyramidal neurons in those layers, which extend into the superficial cortical layers. Scale bars in A–C: 1 mm and in D–F: 0.1 mm. Related data are shown in Fig. S3.

Figure 7. Spatial and temporal variations in thalamic transport targets

This figure shows the thalamic transport zones at different time points in the same animals. As controls, results are also shown from the injection sites, from the same cases. A) The thalamic transport zones following GdDOTA-CTB injections are patchy; transport is confined within subfields of the thalamic nuclei that have known somatotopic connections with S1 (C, D). These results are consistent and very stable for several weeks, until the transported compound clears. B) After manganese injections, the thalamic transport zones expand within a few hours. The expansion extends through the different subfields of thalamic nuclei, and to other neighboring nuclei that do not have known connections with S1 (C, D). Related results are shown in Fig. S6.

Figure 8. GdDOTA-CTB transport in the olfactory pathway

A) T1-W images in the sagittal plane, acquired two days after unilateral nostril injections. GdDOTA-CTB strongly enhanced the entire peripheral olfactory projection to its central target OB. Prominent signal increases were also detected in the olfactory epithelium (red arrowheads), the olfactory nerve tract (ON, long red arrows). Note that injection of GdDOTA-CTB in one nostril did not enhance signal in the contralateral nostril pathway. B) T1-W images in the coronal plane at a different level (a and b, as shown in the box), acquired 7 days after a right nostril injection. In addition to the olfactory epithelium (red arrowheads) and olfactory nerve tract (red arrows), in the outer glomeruli layer of the OB, some individual glomeruli were also enhanced (blue arrowheads). C) Signal enhancement in the central olfactory pathway 7 days following OB injection. MR levels were enhanced in the anterior olfactory nucleus (AON) and pyriform cortex (PCX).