Longitudinal Cell Tracking and Simultaneous Monitoring of Tissue Regeneration after Cell Treatment of Natural Tendon Disease by Low-Field Magnetic Resonance Imaging

Abstract

Treatment of tendon disease with multipotent mesenchymal stromal cells (MSC) is a promising option to improve tissue regeneration. To elucidate the mechanisms by which MSC support regeneration, longitudinal tracking of MSC labelled with superparamagnetic iron oxide (SPIO) by magnetic resonance imaging (MRI) could provide important insight. Nine equine patients suffering from tendon disease were treated with SPIO-labelled or nonlabelled allogeneic umbilical cord-derived MSC by local injection. Labelling of MSC was confirmed by microscopy and MRI. All animals were subjected to clinical, ultrasonographical, and low-field MRI examinations before and directly after MSC application as well as 2, 4, and 8 weeks after MSC application. Hypointense artefacts with characteristically low signal intensity were identified at the site of injection of SPIO-MSC in T1- and T2 (∗) -weighted gradient echo MRI sequences. They were visible in all 7 cases treated with SPIO-MSC directly after injection, but not in the control cases treated with nonlabelled MSC. Furthermore, hypointense artefacts remained traceable within the damaged tendon tissue during the whole follow-up period in 5 out of 7 cases. Tendon healing could be monitored at the same time. Clinical and ultrasonographical findings as well as T2-weighted MRI series indicated a gradual improvement of tendon function and structure.

Figure 1

Exemplary transverse T1-weighted gradient echo magnetic resonance image of the equine distal limb in the metacarpal region, with arrows indicating the metacarpal bone (A), the hypointense deep digital flexor tendon (DDFT, B), and the superficial digital flexor tendon (SDFT, C) displaying a hyperintense intratendinous lesion (D) surrounded by hypointense healthy tendon tissue. The rectangle marks the region relevant to this study, which is displayed in all other figures; lateral is displayed on the left.

Figure 2

Microscopic and magnetic resonance images of labelled cells. Prussian blue staining of intracellular iron oxide particles (a); orange fluorescence of the intracellular rhodamine, nuclei in blue (b); T1-weighted (T1w) and T2^∗-weighted (T2^∗w) gradient echo (GRE) images (c) of the gel phantoms 1 week (1 wk), 2 weeks (2 wks), and 4 weeks (4 wks) after cell labelling. The upper wells in each gel contain 10⁶ labelled cells (MSC), the lower wells 10⁵ labelled cells (MSC). The shape of the wells is indicated exemplarily by circles in the first upper image in (c). The hypointense artefacts caused by the labelled cells are visible in both sequences until 2 weeks after labelling. Here, the higher cell concentration led to artefacts exceeding the shape of the well in which the MSC were localized. Four weeks after labelling, no artefacts were seen in the T1w GRE sequence and only 10⁶ MSC induced weak hypointense artefacts in the T2^∗w GRE sequence.

Figure 3

(Up): boxplot of score points obtained in ultrasonographic assessment in all cases at the day of treatment (pre) and 2 weeks (2 w), 4 weeks (4 w), and 8 weeks (8 w) after treatment, with “a” indicating lower score points (improvement) compared to 2 w (P < 0.05). (Down): exemplary cross-sectional ultrasonographic images of a severe tendon lesion with an extensive hypoechoic defect (arrow) of the superficial digital flexor tendon at the day of treatment (A), 2 weeks after treatment (B), and 4 (C) and 8 (D) weeks after treatment when showing gradual filling of the defect and decrease of tendon enlargement.

Figure 4

Boxplots displaying the lesion percentage and lesion signal intensity quantified by ultrasonography and/or magnetic resonance imaging (MRI) in all cases and exemplary transversal MRI images from the different examinations before (pre, A–D) and directly (post) as well as 2 (2 w), 4 (4 w), and 8 (8 w, E–H) weeks after cell application. Displayed MRI images are T1-weighted gradient echo (T1w, A + E), T2^∗-weighted gradient echo (T2^∗w, B + F), T2-weighted fast spin echo (T2w, C + G), and short tau inversion recovery fast spin echo sequences (STIR, D + H). In the boxplots, “a,” “b,” and “c” indicate lower values compared to T1w, T2^∗w, or STIR images, respectively (P < 0.05); stars indicate differences between T2w images from different examinations. The MRI images demonstrate the decrease in signal intensity and lesion percentage in the T2w and STIR sequences; lateral is displayed on the left.

Figure 5

Exemplary transverse T2^∗-weighted gradient echo images of a horse with severe tendinopathy (upper row, A–E) and a horse with moderate tendinopathy (lower row, F–J) that were both treated with labelled cells, and boxplots displaying the signal intensities of regions of interest measured for cell tracking and the areas covered by hypointense artefact in T1-weighted gradient echo (T1w) and T2^∗-weighted gradient echo (T2^∗w) images. All separate examinations before injection (A + F, pre) and directly (B + G, post) as well as 2 (C + H, 2 w), 4 (D + I, 4 w), and 8 (E + F, 8 w) weeks after injection are displayed. Before injection, no hypointense areas were visible within the lesion except for 1 case. Directly after injection, hypointense artefacts were located within the lesion and surrounding tissue (arrows). At week 2, week 4, and week 8, the hypointense areas were decreasing gradually but still visible. Furthermore, in the case displayed in the lower row, the artefacts appeared to be located more within the tendon lesion at week 8 compared to week 2 or week 4. Lateral is displayed on the left. The upper boxplot illustrates that signal intensities within hypointense areas relatable to labelled cells (MSC) were low in all cases and mostly within the same range as signal intensity of the corresponding healthy deep digital flexor tendon (DT). The plot also displays the one case in which a false-positive judgement had been made (F-P), showing that this case could be discriminated by the higher signal intensity within the area that had been mistaken for being induced by labelled cells. The lower boxplot shows the decreasing trend in the area covered by hypointense artefacts.

Figure 6

Exemplary transverse T2^∗-weighted gradient echo images of a tendon lesion in which labelled cells could not be discriminated until week 8, before (A) and directly after injection of labelled cells (B) and after 8 weeks (C). The image obtained before injection displays the tendon lesion as well as adjacent hypointense tendon fibres (white arrows). Directly after injection, the hypointense artefacts induced by the labelled cells are clearly visible as additional hypointense areas (grey arrows). After 8 weeks, it was not possible to discriminate these hypointense artefacts from the adjacent hypointense tendon fibers anymore.