Frontoparietal connectivity as a product of convergent evolution in rodents and primates: useful connectivity topologies in gray squirrels, rats, and marmosets

Subjects

Grey squirrels and rats

Data were collected from five (1 female, 4 melanistic) adult wild-caught Eastern grey squirrels (Sciurus carolinensis) of unknown age weighing 590–700 g under a wildlife scientific collector’s authorization from the Ontario Ministry of Natural Resources. Once a squirrel was caught inside a commercial trap (Havahart Squirrel & Chipmunk Live Trap), the trap was placed in a large plastic storage box and immediately transported to the Centre for Functional and Metabolic Mapping at the University of Western Ontario (5-min transport time). Squirrel anesthesia was induced by administering 4–5% isoflurane and oxygen with a flow rate between 1 and 1.5 l/min using a precision vaporizer attached to the plastic storage box, which served as an induction chamber. Data were collected from five adult male Wistar rats aged 8–12 weeks and weighing from 250–350 g. Rat anesthesia was induced by placing the animals in an induction chamber with 4–5% isoflurane and oxygen with a flow rate between 1 and 1.5 l/min.

Both species were lightly anesthetized via spontaneous inhalation of 1.5–2% isoflurane mixed with oxygen flowing between 1.5 and 2 L/min throughout the scan. Respiration, SpO2, and heart rate were continuously monitored using a pulse oximeter and were observed to be within the normal range throughout the scans (squirrels: respiration = 35–80 (mean 57) bpm, SpO2 = 95–100%, heart rate = 230–318 (mean 274) bpm). Body temperature was also measured and recorded throughout, maintained using warm-water circulating blankets, thermal insulation, and warmed air. All animals were head-fixed in stereotactic position using a custom-built MRI bed with ear bars and a species-specific palate bar as part of the anesthesia mask22. Upon completion of imaging, 3–5 ml of warmed NaCl was administered subcutaneously. A small ear punch was made in squirrels to mark the animal and avoid repeated study of the same squirrel. A blanket and supplemental heat were provided for recovery. After recovery, squirrels were transported in the plastic storage/induction box back to the capture location and released. One squirrel died during recovery. Its brain was removed, fixed by submersion in formalin, and subsequently imaged for ultra-high resolution structural MRI at the University of Pittsburgh Brain Institute. All experimental procedures were in accordance with the Canadian Council of Animal Care policy and protocols approved by the Animal Care Committee of the University of Western Ontario Council on Animal Care.

Marmosets

To compare the squirrel and rat data to marmoset data, we used fMRI data from our open-access resource (https://marmosetbrainconnectome.org)41. This database contains resting-state fMRI data from 31 awake marmosets (Callithrix jacchus, 8 females; age: 14–115 months; weight: 240–625 g) that were acquired at the University of Western Ontario (5 animals) and the National Institutes of Health (26 animals).

Imaging

Image acquisition

For both rats and grey squirrels, data were acquired on a 9.4 T 31 cm horizontal bore MRI scanner (Varian/Agilent, Yarnton, UK) and Bruker BioSpec Avance III console with the software package Paravision-6 (Bruker BioSpin Corp, Billerica, MA) and a custom-built, high-performance 15-cm-diameter gradient coil with 400-mT/m maximum gradient strength42 at the Centre for Functional and Metabolic Mapping at the University of Western Ontario. The animal holders and radiofrequency receive arrays were built in-house, and design files for these stereotactic holders have been made available22 with geometrically optimized phased array receive coil designs for the two species. The rat coil was made up of 6 channels, while a larger marmoset coil with 8 channels was used for the squirrels. Both coils have a similar signal-to-noise ratio (SNR). Preamplifiers were located behind the animals, and the receive coil was placed inside a quadrature birdcage coil (12-cm inner diameter) used for transmission.

For the rats, functional imaging was acquired during one session for each animal, with 6 functional runs (at 600 volumes each) with the following parameters: TR = 1,500 ms, TE = 15 ms, field of view = 38.4 × 38.4 mm, matrix size = 96 × 96, voxel size = 0.4 × 0.4 × 0.4 mm, slices = 35, bandwidth = 280 kHz, GRAPPA acceleration factor: 2 (anterior-posterior). T2-weighted structural scans were acquired for each animal with the following parameters: TR = 7,000 ms, TE = 44 ms, field of view = 38 × 38 mm, matrix size = 192 × 192, voxel size = 0.2 × 0.2 × 0.4 mm, slices = 35.

The squirrel imaging was similar to rat imaging with functional imaging data acquired from one session for each animal with 3–4 functional runs (at 600 volumes each) with the following parameters: TR = 1500 ms, TE = 15 ms, flip angle = 40 degrees, field of view = 64 × 64 mm, matrix size = 128 × 128, voxel size = 0.5 × 0.5 × 0.5 mm, slices = 42, bandwidth = 400 kHz, GRAPPA acceleration factor: 2 (anterior-posterior). T2-weighted structural scans were acquired for each animal with the following parameters: TR = 7000 ms, TE = 55 ms, field of view = 64 × 64 mm, matrix size = 384 × 384, voxel size = 0.133 × 0.133 × 0.5 mm, slices = 45. As an in vivo registration template, we also acquired a structural T2-weighted image from squirrel #5 with an isotropic voxel resolution with the following parameters: TR = 15,000 ms, TE = 48 ms, field of view = 51 × 51 mm, matrix size = 256 × 256, voxel size = 0.2 × 0.2 × 0.2 mm, bandwidth = 50 kHz, slices = 100, number of averages = 3.

For the squirrels, an ultra-high-resolution ex vivo template was also generated from squirrel #5 – the same brain used as the in vivo registration template. The sample was fixed by submersion in 10% formalin for one week, then imaged at the University of Pittsburgh Brain Institute on a 9.4 T 30 cm horizontal bore MRI scanner (Bruker BioSpin Corp, Billerica, MA) equipped with a Bruker BioSpec Avance Neo console and the software package Paravision-360 (version 3.2; Bruker BioSpin Corp, Billerica, MA) and a custom high performance 17 cm gradient coil (Resonance Research Inc, Billerica, MA) performing at 450 mT/m gradient strength. To maximize sensitivity, a custom 30 mm inner diameter millipede quadrature coil (ExendMR LLC, Milpitas, CA) was used. The sample was placed in a 50 ml conical centrifuge tube, filled with Fomblin oil (Solvay Solexis, West Deptford, NJ), and placed under vacuum (−27 inHg) for 30 min to remove air bubbles. A T2*-weighted structural scan was acquired with the following parameters: TR = 100 ms, TE = 8 ms, field of view = 35 × 30 × 20 mm, matrix size = 700 × 600 × 400, voxel size = 0.05 × 0.05 × 0.05 mm, bandwidth = 50 kHz, number of averages = 2, total scan time = 14 h, 48 min. The T1 relaxation time of the tissue was measured (mean T1 = 1047 ms), and an optimum flip angle (Ernst angle) of 24.7 degrees was set for a TR of 100 ms.

For the marmosets, functional imaging was acquired from two sites. The University of Western Ontario data were acquired using the same 9.4 T scanner as the rat and squirrel data. A custom 5-channel receive coil was used, which was rigidly fixed to the head implant43. Radiofrequency transmission was accomplished with a quadrature birdcage coil (12-cm inner diameter) built in-house. Functional imaging was performed over multiple sessions (days) for each animal, with 4-6 functional runs (at 600 volumes each) per animal with the following parameters: TR = 1500 ms, TE = 15 ms, flip angle = 35 degrees, field of view = 64 × 64 mm, matrix size = 128 × 128, voxel size = 0.5 × 0.5 × 0.5 mm, slices = 42, bandwidth = 500 kHz, GRAPPA acceleration factor: 2 (anterior-posterior). T2-weighted structural scans were acquired for each animal during one of the awake sessions with the following parameters: TR = 5,500 ms, TE = 53 ms, field of view = 51.2 × 51.2 mm, matrix size = 384 × 384, voxel size = 0.133 × 0.133 × 0.5 mm, slices = 42, bandwidth = 50 kHz, GRAPPA acceleration factor: 2.

The NIH marmoset data were acquired using at 7 T 30 cm horizontal bore magnet (Bruker BioSpin Corp, Billerica, MA, USA) with the software package Paravision (Bruker BioSpin Corp, Billerica, MA, USA), a 15-cm-diameter gradient coil with 450-mT/m maximum gradient strength (Resonance Research Corp, Billerica, MA, USA) and custom 10-channel phased-array receive coil which conformed to the 3D printed head holder. Radiofrequency transmission was accomplished with a 16-rung high-pass birdcage coil. Functional imaging was performed in a single session with 4-8 functional runs (at 512 volumes each) with the following parameters: TR = 2000 ms, TE = 22.2 ms, flip angle = 70.4, field of view = 28 × 36 mm, matrix size = 56 × 72, voxel size = 0.5 × 0.5 × 0.5 mm, slices = 38. Two sets of spin-echo EPI with an opposite phase-encoding direction (left-right and right-left) were collected for the EPI-distortion correction (TR = 3000 ms, TE = 0.44 ms, flip angle = 90 degrees, FOV = 28 × 36 mm, matrix size = 56 × 72, voxel size = 0.5 × 0.5 × 0.5 mm, axial slices = 38). T2-weighted structural image scans were acquired for each animal during one of the awake sessions with the following parameters: TR = 6000 ms, TE = 9 ms, flip angle = 90°, FOV = 28 × 36 mm, matrix size = 112 × 144, slices = 38, voxel size = 0.25 × 0.25 × 0.5 mm, number of averages = 8.

Image preprocessing

For squirrels, rats, and marmosets, data were similarly processed with custom preprocessing pipelines using the Analysis of Functional NeuroImages (AFNI)44 and FMRIB Software Library (FSL)45 software packages. Raw functional images were converted to NifTI format using dcm2niix46 and reoriented from the sphinx position using FSL. The images were then despiked (AFNI’s 3dDespike), and volume registered to the middle volume (AFNI’s 3dvolreg). The motion parameters from volume registration were stored for later use with nuisance regression. For the rats, images were smoothed by a 1 mm full-width at half-maximum Gaussian kernel to reduce noise (AFNI’s 3dmerge); for the larger squirrel and marmoset brains, a 1.5 mm kernel was used. An average functional image was then calculated for each run and registered (FSL’s FLIRT) to each animal’s T2-weighted image—the 4D time-series data was transformed using this matrix. T2-weighted images were manually skull-stripped (including the olfactory bulb in all three species), and this mask was applied to the functional images.

Each individual animal’s T2-weighted images were non-linearly registered to high-resolution anatomical templates: for squirrels, images were registered to the higher-resolution anatomical image from squirrel 5 (voxel size = 0.2 × 0.2 × 0.2 mm). For rats, images were registered to the anatomical image provided in template space (voxel size = 0.05 × 0.05 × 0.05 mm)47. For marmosets, images were registered to the NIH marmoset brain atlas (voxel size = 0.2 × 0.2 × 0.2 mm)48. Functional images from all three species were bandpass filtered between 0.01 and 0.1 Hz.

Fingerprinting

For frontal regions, seed analyses were conducted between the mean time course within each seed region and every other voxel in the brain (with the nuisance regressors described above). Group functional connectivity maps (Z score maps) were then calculated for each of the frontal seeds in the three species. For rats, seeds (0.9 mm isotropic cubes) were placed in frontal areas Fr3, FrA, M2, LO, and MO based on the atlas (see Fig. 2, red squares). For squirrels, no atlas is available; therefore, 6 seeds (1.6 mm isotropic cubes) were placed in the frontopolar cortex (see Fig. 2, red squares) across area F based on a paper from the Kaas lab26. For marmosets, seeds (single voxels) were placed in frontal area 6 VA, 6DR, 8AD, 8AV, 45, 47 L, 46D, and 46 V (see Fig. 2, red squares) using our open-access resource (https://marmosetbrainconnectome.org)41—given the high statistical power of this functional connectivity dataset, averaging across multiple voxels was not necessary.

We then specified eight common regions extrinsic to the frontal cortex in all three species (placement shown in Fig. 2 as blue squares). For each species, regions of interest were manually drawn in the posterior parietal cortex (PPC), posterior cingulate cortex (PCC), S1, M1, insula (Ins), striatum, superior colliculus (SC), and pulvinar (Pul). For the rats, these regions of interest were 0.9 mm isotropic cubes; in the squirrels and marmosets, 1.5 mm isotropic cubes. The placement of the cubes was based on definitions of these areas in the rat and marmoset atlas and the squirrel cortical parcellation by Wong and Kaas23. We also systematically translated the seed regions about the original anatomically defined centroid in each species and found that the placement had minimal impact on the overall fingerprint pattern (Supplementary Figs. 7, 8 and 9 for rats, squirrels, and marmosets respectively).

With the eight regions of interest defined in each respective species, we then extracted the mean connectivity values within these regions (with variance shown in Supplementary Fig. 10)—these values constituted the fingerprints, with a separate fingerprint for each seed region, for each species. To compare across the species, we normalized the fingerprint to a range between 0 (weakest connection with any of the target regions) and 1 (strongest connection with any of the target regions). We thus compare a connectivity pattern with target areas rather than absolute strength in any given species23. For the comparisons, we calculated the multidimensional cosine similarity across the matrix of functional connectivity fingerprints23—intuitively similar to a correlation value. The cosine similarity analysis provided an index of how similar or different the interareal fingerprint patterns were. By comparing the cosine of the angle between vectors (i.e., fingerprints), the cosine similarity metric indexes how similar the orientation of a set of vectors are in normalized space, with high similarity values indicating similar fingerprints (i.e., vectors in the same direction) and low scores indicating dissimilar fingerprints (i.e., vectors of diverging direction). By plotting the fingerprints in spider plots, we show the specific regions where the fingerprints differ. We applied this technique to compare squirrel, rat, and marmoset frontal FC patterns with the eight extrinsic regions.

Statistics and reproducibility

Permutation testing was used to test for statistical differences between fingerprints across the species. Permutation tests were performed via in-house code written in Matlab. Pair-wise comparisons of fingerprints were performed for each seed region of interest by first randomly dividing individual (i.e., scan-level) fingerprints into two groups (with fingerprints from each of two species randomly forming the comparison distributions), group-wise averaging, then normalization of the fingerprints to a range of 0 (weakest connection with any of the target regions) and 1 (strongest connection with any of the target regions). Cosine similarity was then calculated across species. This process was iterated 10,000 times, yielding a distribution of cosine similarity values that trended towards 1. A cosine similarity value smaller than the lowest 1 percentile of permuted cosine similarity values was defined as a significantly different fingerprint comparison.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.