Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Jul 22;2(7):e12088.
doi: 10.14814/phy2.12088. Print 2014 Jul 1.

Automated Quantification of Renal Fibrosis With Sirius Red and Polarization Contrast Microscopy

Affiliations
Free PMC article

Automated Quantification of Renal Fibrosis With Sirius Red and Polarization Contrast Microscopy

Jonathan M Street et al. Physiol Rep. .
Free PMC article

Abstract

Interstitial fibrosis is commonly measured by histology. The Masson trichrome stain is widely used, with semiquantitative scores subjectively assigned by trained operators. We have developed an objective technique combining Sirius Red staining, polarization contrast microscopy, and automated analysis. Repeated analysis of the same sections by the same operator (r = 0.99) or by different operators (r = 0.98) was highly consistent for Sirius Red, while Masson trichrome performed less consistently (r = 0.61 and 0.72, respectively). These techniques performed equally well when comparing sections from the left and right kidneys of mice. Poor correlation between Sirius Red and Masson trichrome may reflect different specificities, as enhanced birefringence with Sirius Red staining is specific for collagen type I and III fibrils. Combining whole-section imaging and automated image analysis with Sirius Red/polarization contrast is a rapid, reproducible, and precise technique that is complementary to Masson trichrome. It also prevents biased selection of fields as fibrosis is measured on the entire kidney section. This new tool shall enhance our search for novel therapeutics and noninvasive biomarkers for fibrosis.

Keywords: Chronic kidney disease; Masson trichrome; fibrosis.

Figures

Figure 1.
Figure 1.
Kidney fibrosis. Color images of fibrotic kidney sections stained with (A) Masson trichrome (bright field) or stained with Sirius Red viewed under (B) bright field or (C) polarization contrast. Grayscale images of Sirius Red staining in polarization contrast (D) before and (E) after a background threshold is applied. Scale bar is 100 μm.
Figure 2.
Figure 2.
Threshold intensity is set by the distribution of pixel intensities in the image. Images of Sirius Red stained sections were recorded using 12‐bit resolution (possible pixel intensity values 0–4095). The location and width at half maximum of the background peak on a histogram are measured. The threshold is then set to the pixel intensity of the peak plus 1.5 times the peak width at half of the maximum intensity.
Figure 3.
Figure 3.
Focus and polarizer alignment influence the measurement of fibrosis. (A) A series of images both in focus and progressively out of focus were recorded for individual high‐power fields by varying the height of the stage. The percentage of fibrosis was measured for each image from high‐power fields with minimal fibrosis (open circles) or substantial fibrosis (closed circles). Loss of focus reduced fibrosis percentage area for fields with minimal fibrosis and increased fibrosis percentage area for fields with substantial fibrosis. (B) Signal‐to‐noise ratio for a HPF decreases as misalignment increases between the polarization filters.
Figure 4.
Figure 4.
Masking perivascular collagen. Perivascular collagen was removed from the ROI using a mask. (A) Perivascular collagen appears as large clusters of high‐intensity pixels. (B) Applying a high threshold to the image preferentially identifies perivascular collagen. (C) A Gaussian blur assigns each pixel a weighted average of neighboring pixels. (D) Preservation of pixel intensity in large clusters enables a second threshold to identify perivascular collagen. Scale bar is 100 μm. (E) Comparison of the perivascular and interstitial fibrosis percentage areas using the automated masking technique.
Figure 5.
Figure 5.
Severity of fibrosis varies among fields. Adjacent kidney sections from seven mice were stained with Masson trichrome or Sirius Red. Ten 20× objective fields from the Masson trichrome stained sections were scored for kidney fibrosis using a scale ranging from 0 to 5. Twenty 40× objective fields from Sirius Red stained sections were imaged using polarization contrast microscopy and the fibrotic area was measured. Horizontal lines represent the mean ± SEM. Both Masson trichrome fibrosis score (A) and Sirius Red percentage fibrotic area (B) varied between different high‐powered fields for the same animal.
Figure 6.
Figure 6.
Correlation between fibrosis severity as measured by Masson trichrome and Sirius Red stains. For the folic acid model (A and B) and UUO model (C and D) adjacent kidney sections stained with Masson Trichrome or Sirius Red were scored and measured, respectively. The Masson Trichrome and Sirius Red fibrosis values were plotted, and a Pearson correlation coefficient calculated. (A and C) The entire region of interest was included when calculating the fibrotic area percentage using Sirius Red. Although statistically significant the relationship between Masson Trichrome score and Sirius Red fibrosis percentage was weak. (B and D) Vessels were excluded from the Sirius Red analysis by an automated three‐step threshold‐filter‐threshold process (Methods). The collagen surrounding many blood vessels is strongly birefringent and can erroneously increase the fibrosis percentage and weaken the correlation with the Masson Trichrome score.
Figure 7.
Figure 7.
Correlation for Masson trichrome and Sirius Red stains between different sessions by a single observer. Two weeks after folic acid injection adjacent kidney sections were stained with Masson trichrome or Sirius Red. The sections were measured twice by the same observer for fibrosis. The correlation between the first and second observations for (A) Masson trichrome and (B) Sirius Red is plotted and a Pearson correlation coefficient calculated for each. Sirius Red was much more consistent between sessions than Masson trichrome, exhibiting a stronger correlation and less bias.
Figure 8.
Figure 8.
Correlation for Masson trichrome and Sirius Red stains between two observers. Two weeks after folic acid injection adjacent kidney sections were stained with Masson trichrome or Sirius Red. Two different observers then measured the sections for fibrosis. The correlation between the scores given by the two observers for (A) Masson trichrome and (B) Sirius Red is plotted and a Pearson correlation coefficient calculated for each. Sirius Red was more consistent between observers illustrating a reduced subjectivity of the automated three‐step threshold‐filter‐threshold process used with Sirius Red.
Figure 9.
Figure 9.
Correlation for Masson trichrome and Sirius Red stains between two kidneys from the same mouse. Two weeks after folic acid injection adjacent sections from the left and right kidneys stained with Masson trichrome or Sirius Red were measured for fibrosis. The correlation between the two kidney sections for (A) Masson trichrome and (B) Sirius Red is plotted and a Pearson correlation coefficient calculated for each. Although both techniques performed well, Masson trichrome scores were slightly more consistent between kidneys.

Similar articles

See all similar articles

Cited by 32 articles

See all "Cited by" articles

References

    1. Doi K., Leelahavanichkul A., Hu X., Sidransky K. L., Zhou H., Qin Y. 2008. Pre‐existing renal disease promotes sepsis‐induced acute kidney injury and worsens outcome. Kidney Int.; 74:1017-1025. - PMC - PubMed
    1. Farris A. B., Colvin R. B. 2012. Renal interstitial fibrosis: mechanisms and evaluation. Curr. Opin. Nephrol. Hypertens.; 21:289-300. - PMC - PubMed
    1. Farris A. B., Adams C. D., Brousaides N., Della Pelle P. A., Collins A. B., Moradi E. 2011. Morphometric and visual evaluation of fibrosis in renal biopsies. J. Am. Soc. Nephrol.; 22:176-186. - PMC - PubMed
    1. Grimm P. C., Nickerson P., Gough J., McKenna R., Stern E., Jeffery J. 2003. Computerized image analysis of Sirius Red‐stained renal allograft biopsies as a surrogate marker to predict long‐term allograft function. J. Am. Soc. Nephrol.; 14:1662-1668. - PubMed
    1. He W., Wang Y., Zhang M. Z., You L., Davis L. S., Fan H. 2010. Sirt1 activation protects the mouse renal medulla from oxidative injury. J. Clin. Invest.; 120:1056-1068. - PMC - PubMed

LinkOut - more resources

Feedback