In vivo Manganese-enhanced MRI for Visuotopic Brain Mapping

Abstract

This study explored the feasibility of localized manganese-enhanced MRI (MEMRI) via 3 different routes of Mn²⁺ administrations for visuotopic brain mapping of retinal, callosal, cortico-subcortical, transsynaptic and horizontal connections in normal adult rats. Upon fractionated intravitreal Mn²⁺ injection, Mn enhancements were observed in the contralateral superior colliculus (SC) and lateral geniculate nucleus (LGN) by 45–60% at 1–3 days after initial Mn²⁺ injection and in the contralateral primary visual cortex (V1) by about 10% at 2–3 days after initial Mn²⁺ injection. Direct, single-dose Mn²⁺ injection to the LGN resulted in Mn enhancement by 13–21% in V1 and 8–11% in SC of the ipsilateral hemisphere at 8 to 24 hours after Mn²⁺ administration. Intracortical, single-dose Mn²⁺ injection to the visual cortex resulted in Mn enhancement by 53–65% in ipsilateral LGN, 15–26% in ipsilateral SC, 32–34% in the splenium of corpus callosum and 17–25% in contralateral V1/V2 transition zone at 8 to 24 hours after Mn²⁺ administration. Notably, some patchy patterns were apparent near the V1/V2 border of the contralateral hemisphere. Laminar-specific horizontal cortical connections were also observed in the ipsilateral hemisphere. The current results demonstrated the sensitivity of MEMRI for assessing the neuroarchitecture of the visual brains in vivo without depth-limitation, and may possess great potentials for studying the basic neural components and connections in the visual system longitudinally during development, plasticity, pharmacological interventions and genetic modifications.

I. INTRODUCTION

The rodents are an excellent model for understanding the mechanisms of development, plasticity and functional specialization in the visual system [1–13]. In normal adults, more than 90% of axons of retinal ganglion cells (RGCs) project contralaterally to the superior colliculus (SC) and lateral geniculate nucleus (LGN) [14, 15]. The superficial gray layer of the SC receives about 90% of its excitatory input from the retina [16] and the remainder 10% from the visual cortex [17], whereas about 30% of RGC axons project to the contralateral dorsal LGN [18], and over 90% of cells projected from the primary visual cortex (V1) lie in the ipsilateral dorsal LGN [19]. The topographic layout of the retina is represented in the SC, LGN, visual callosal fibers and each cortical visual area [20–23]. Ocular dominance plasticity is also present in rodent visual cortex [21, 24, 25]. To date, limited tools have been available for in vivo, high-resolution mapping of neuroarchitecture in the visual brains globally and longitudinally [3, 26, 27]. Mn²⁺ has been increasingly used as a T₁-weighted contrast agent for in vivo neuronal tract tracing [28–38], detection of neuroanatomy and pathophysiology [39–43] and functional brain mapping at lamina levels [33, 44–47]. In this study, we explore the capability of Mn-enhanced MRI (MEMRI) via 3 different routes of Mn²⁺ administration for in vivo assessments of retinal, callosal, transsynaptic, corticothalamic, corticocollicular and horizontal connections in normal adult rat brains.

II. Materials and Methods

A. Animal Preparation

Adult Sprague-Dawley rats (N=17) were divided into 3 groups. In Group 1 (n=4), a fractionated dose of Mn²⁺ at 3µL and 50mM each was injected intravitreally into the left eye every day for a total of 3 days; In Group 2 (n=6), Mn²⁺ was injected unilaterally into the left lateral geniculate nucleus (LGN) at 30nL and 100mM; In Group 3 (n=7), Mn²⁺ was injected intracortically to the V1/V2 transition zone of the right visual cortex at 100nL and 100mM. For Group 1, MEMRI was performed before, and at 1, 2 and 3 days after initial Mn²⁺ intravitreal injection. For Groups 2 and 3, MEMRI was performed at 1 hour, 8 hours and 1 day after Mn²⁺ administration.

B. MRI Protocols

All in vivo MRI measurements were acquired utilizing a 7 T Bruker scanner using a 72 mm birdcage transmit-only RF coil and an actively decoupled receive-only quadrature surface coil. Under inhaled isoflurane anaesthesia (3% induction and 1.5% maintenance), animals were kept warm under circulating water at 37°C with continuous monitoring of the respiration rate. 2D T1-weighted (T1W) spin-echo RARE pulse sequence was acquired with repetition time/echo time (TR/TE) = 475/8.8 ms, field of view/slice thickness (FOV/th) = 32x32 mm²/0.8 mm, matrix resolution = 256x256, acquired resolution = 125x125 µm², number of slices = 10, RARE factor = 4 and total scan time = 15 mins. 3D T1WI was acquired using the MPRAGE sequence covering the entire visual pathway, with TI/TR/TE = 2500/9/3ms, FOV = 32x32x11mm, acquisition resolution = 200x200x240μm³, 1 segment, number of averages = 8 and total scan time = 15 mins.

C. Data Analysis

T1W signal intensities (SI) in the superior colliculi (SC), lateral geniculate nuclei (LGN), primary visual cortex (V1), and V1/V2 transition zone of each hemisphere, and in the splenium of corpus callosum (CC) were measured using ImageJ v1.43u, and were normalized to the surrounding muscles. Mn²⁺ enhancement was quantified by calculating the ratio between left and right visual components in Groups 1 and 2, and the rate of signal increase at Hour 8 and Day 1 compared to Hour 1 in Group 3. Values at each time point were compared to the first time point using two-tailed paired t-tests. Results were considered significant when p<0.05.

III. Results

A. Intravitreal, fractionated MEMRI of Retinal and Transsynaptic Connections

In normal adult brains, fractionated, intravitreal Mn²⁺ injection resulted in significant Mn enhancements in contralateral SC and LGN by 45–60% at 1–3 days after initial Mn²⁺ injection, and in contralateral V1 (arrows) by about 10% at 2–3 days after initial Mn²⁺ injection (Fig. 1). Contralateral SC appeared to enhance slightly more than contralateral LGN at all times after initial Mn²⁺ injection.

Fig. 1. Intravitreal, fractionated Mn²⁺ injection.

(Top) MEMRI at 3 days after initial Mn²⁺ administration to the left eye. (Bottom) Quantitative analyses of T1W SI between right and left SC, LGN and V1 before, and at 1, 2 and 3 days after initial Mn²⁺ injection (Two-tailed paired t-tests with 1^st time point, *p<0.05; **p<0.01; ***p<0.001)

B. Subcortical, single-dose MEMRI of Thalamo-cortical Connections

Direct, single-dose Mn²⁺ injection to LGN resulted in Mn enhancement by 13–21% in ipsilateral V1 (arrows), and 8–11% in ipsilateral SC at 8–24 hours after Mn²⁺ injection (Fig. 2).

Fig. 2. Subcortical, single-dose Mn²⁺ injection. injection.

(Top) MEMRI at 8 hours and 1 day after Mn²⁺ administration to the left LGN. (Bottom) Quantitative analyses of T1W SI between left and right SC, LGN and V1 at 1, 8 and 24 hours after Mn²⁺ injection. (Two-tailed paired t-tests with 1st time point, *p<0.05; **p<0.01; ***p<0.001)

C. Intracortical, single-dose MEMRI of Callosal, Cortico-subcortical and Horizontal Connections

Intracortical, single-dose Mn²⁺ injection to the visual cortex resulted in Mn enhancement by 17–25% in contralateral V1/V2 transition zone (closed arrows), 32–34% in CC (open arrows), 53–65% in ipsilateral dorsal LGN (dashed arrows) and 15–26% in ipsilateral SC at Hours 8–24 (Fig. 2). Notably, some patchy patterns were apparent near the V1/V2 border of the contralateral left hemisphere (Fig. 3), which might be indicative of the ocular dominance domains recently suggested in rodents [48]. Some horizontal cortical connections were also observed in the superficial and middle layers of the ipsilateral hemisphere (Fig. 3)

Fig. 3. Intracortical, single-dose Mn²⁺ injection.

(Top) MEMRI at 1, 8 and 24 hours after Mn²⁺ administration to the right V1/V2 transition zone. (Bottom) Quantitative analyses of T1W SI of visual components normalized to Hour 1. (Two-tailed paired t-tests with Hour 1, *p<0.05; **p<0.01; ***p<0.001)

IV. Discussions

A. In vivo MEMRI of Retinal, Subcortico-cortical and Transsynaptic Connections

Upon intravitreal Mn²⁺ injection, Mn²⁺ ions were taken up by the RGCs, underwent anterograde axonal transport along retinocollicular and retinogeniculate projections at a rate of 2–6 mm/h, and accumulated at the axonal terminals in the contralateral SC and LGN of the adult rat brain [28, 46]. Although transneuronal transport has been shown to occur in the brain [31], transsynaptic illumination of the visual cortex via intravitreal Mn²⁺ injection has been difficult [49]. Olson et al recently demonstrated that the degree of Mn enhancement in the visual pathway is determined by the duration of availability of Mn²⁺ from the vitreous body but not the injected dose [50]. Given the rapid clearance of Mn from the vitreous body [50], this study evaluated the Mn enhancement in V1 upon (i) fractionated intravitreal injections or (ii) direct, single-dose injection to the LGN. As shown in Figs. 1 and 2, daily, fractionated intravitreal injection resulted in about 10% of Mn enhancement in contralateral V1 starting at Day 2, whereas direct, single-dose injection into the LGN resulted in more than 10% of enhancement in the ipsilateral V1 as early as at 8 hours after injection. The results of these experiments demonstrated the feasibility of MEMRI to evaluate both monosynaptic and polysynaptic anterograde transport in the retinal projections longitudinally upon prolonged Mn input to localized visual components.

B. In vivo MEMRI of Cortico-cortical and Cortico-subcortical Connections

Upon intracortical Mn²⁺ injection, the ipsilateral LGN appeared to enhance at the highest rate than other measured visual components at Hour 8, possibility because of the short distance and high density of cortico-geniculate connections in the visual brain [19]. While the Mn accumulation appeared to slow down from Hours 8 to 24 in the ipsilateral LGN and the CC, the contralateral V1/V2 and ipsilateral SC, whose connections to ipsilateral V1/V2 were more distant, continued to enhance steadily at Hour 24. Recent anatomical and functional studies suggested the presence of ocular dominance domains in the rodent visual cortex [48, 51]. In particular, both histological tracing and electrophysiological experiments demonstrated patches of ipsilateral retinal input in close correlation with callosal patches in the binocular zone of the rat V1 [48]. Our MEMRI observation of the patchy patterns in the callosal projection might provide a new model system for imaging how the precise neuronal connections are shaped by experience in order to study the cellular and molecular mechanisms of ocular dominance plasticity in the critical period [48, 52, 53]. The laminar-specific enhancement in the ipsilateral cortex in Fig. 4 suggested that Mn transport occurred along horizontal connections faster than the diffusion of Mn from the injection site.

Fig. 4.

Reconstructed 3D T1WI of the dorsal brain at 8 hours after right intracortical Mn²⁺ injection. Note the patchy patterns of Mn enhancement in the V1/V2 transition zone of the contralateral left hemisphere (arrow) projected along the visual callosal pathway. The horizontal cortical connections were also observed in the superficial and middle layers of the ipsilateral hemisphere (arrowheads).

V. Conclusions

The results of this study demonstrated the sensitivity of MEMRI for assessing the neuroarchitecture of the visual brains in vivo via 3 different routes of Mn administration without depth-limitation, and may possess great potentials for studying the basic neural components and connections in the visual system longitudinally during development, plasticity, pharmacological interventions and genetic modifications in future studies.