Biomedical Engineering Reference
In-Depth Information
probes the onset of denser or larger fibers (Strupler et al. 2007). Note that this image processing may be
applied to
z
-stacks of 2D images to obtain volume scores.
15.4.4 Biomedical Applications
Fibrosis quantization by means of SHG microscopy was first developed in animal models (Pena et al.
2007, Strupler et al. 2008, Sun et al. 2008, Tai et al. 2009a, He et al. 2010, Raub et al. 2010), and more
recently applied to human biopsies (Gailhouste et al. 2010, Sun et al. 2010). We illustrate here the poten-
tial of this technique with murine kidney fibrosis.
We studied a murine model of hypertensive fibrosis by subcutaneous injection of Angiotensin II
(AngII) for 4 or 7 weeks (Strupler et al. 2007, 2008). Physiological and biochemical parameters, in par-
ticular systolic blood pressure, were monitored to verify the progression of hypertension and fibrosis.
Kidneys were then harvested and prepared either for histology: a half kidney was fixed in formalin,
embedded in paraffin, sectioned, and stained with Masson's trichrome, or for SHG imaging: the other
half kidney was fixed in PFA and 200-μm thick coronal slices were cut using a vibrating microtome and
imaged without any staining (see Figure 15.3c).
Since fibrosis is heterogeneously distributed, we used a motorized stage to scan the kidney tissue
and image the cortical region in our sample (typically 5 × 3 mm
2
) by stitching many SHG/2PEF images
acquired with a 20×, 0.9 NA objective (512 × 512 μm
2
field of view). Typical results are displayed in
Figure 15.3. SHG images reveal the progression of fibrosis in AngII-infused mice, whereas endogenous
2PEF signals enable the visualization of the tissue morphology. We then took advantage of the 2PEF
image to develop a segmentation algorithm based on the kidney morphology and score the fibrosis only
in the cortical region that is relevant for biomedical reasons (see Figure 15.3b). The SHG density scores
are displayed in Figure 15.3e for all the mice under study. They are compared to anapathologist semi-
quantitative scoring of tubulointerstitial fibrosis in Masson's trichrome sections of the same kidney
(Spurney et al. 1992). Both scoring methods show a good agreement as expected. However, SHG imag-
ing advantageously provides continuous scoring with better sensitivity and can distinguish different
pathological grades at early stages (see the range of SHG scores in the 0 grade for anapathologist score).
We took advantage of this sensitivity to quantify the role of tissular transglutaminase (TG2) in the
fibrosis progression. This collagen cross-linking enzyme is expected to promote fibrosis through col-
lagen assembly. Accordingly, we observed that TG2-deficient mice exhibited significant lower intersti-
tial fibrosis than wild type mice, whereas they showed similar hypertension progression. However, we
observed no colocalization between TG2 activity and interstitial fibrosis (see Figure 15.8).
Our SHG images also proved efficient to look at the 3D distribution of collagen fibrosis within the
kidney tissue as exemplified in Figure 15.3d. Interestingly, we observed a continuity between perivascu-
lar, periglomerular, and tubulointerstitial fibrils (within our optical resolution) that may suggest com-
mon mechanisms of progression.
15.5 Discussion: Advantages and Limitations of SHG
Microscopy for Fibrosis Scoring
15.5.1 comparison to other imaging techniques
We have demonstrated in the former sections that SHG microscopy is a valuable tool for fibrosis imag-
ing and quantization. However, it is a complex and expensive method compared to histological tech-
niques and it is worth discussing the crucial advantages of this technique.
The main practical drawback of histological and immunochemical techniques stems from the stain-
ing/labeling procedure that limits the reproducibility and restricts these techniques to thin 2D sec-
tions. The most effective technique among histological and immunochemical techniques appears to be
Picrosirius Red staining when visualized with circularly polarized light microscopy (Junqueira et al.
