*3.2. Depletion of Fibrocytes Attenuated Hepatic Fibrogenesis*

TAA-administration induced a marked perilobular fibrosis in mice of the control (Ctrl)- and fibrocyte-ablated group (FC-Abl., Figure 2a,b). Irregular liver architecture and regenerative nodules indicated beginning cirrhosis in some mice. Total liver hydroxyproline content was reduced by 7.8% (95% CI: 0.7%–14.8%; Figure 2c) in fibrocyte-ablated mice, denoting a reduced deposition of fibrillar collagens. The difference is considered statistically significant (*p* = 0.033). Neither the pathologist's staging depicted in Table 1, nor the morphometric analysis of histological samples (Figure 2d) showed differences between the control- and fibrocyte-ablated group.


**Table 1.** Staging according to Ishak et al.

<sup>1</sup> *p* = 0.476; Mann-Whitney *U* test was applied.

**Figure 2.** Fibrocyte ablation attenuated hepatic fibrogenesis. Fibrillar collagen distribution was visualized by (**a**) Sirius Red/Fast Green staining and (**b**) immunohistochemical staining of collagen I on formalin fixed, paraffin-embedded liver sections. TAA-treatment caused pronounced periportal and bridging fibrosis as well as faint chicken wire sinusoidal fibrosis in the control- and fibrocyte-ablated group. Dotted boxes are shown in enlarged panels on the right side. Magnification 200×, bars 100 and 25 μm. (**c**) Quantitative assessment of hepatic hydroxyproline content revealed a reduction of fibrillar collagens in mice lacking fibrocytes. The assay was performed three times. Mean values of each individual mouse are depicted by black triangles (control) or blue dots (fibrocyte ablation). The solid line depicts the group mean, the dotted line the mean hydroxyproline level in untreated supercontrols (SC). (**d**) Morphometric analysis of Sirius Red/Fast Green-stained sections displayed a comparable extent of red-stained areas in TAA-treated mice with and without fibrocyte ablation. A total of 1361 images were analyzed (2–134 per mouse). (**e**) The transcriptional levels of *Col1a1* were equal throughout both groups. (**f**) Relative protein levels of MMP-2, MMP-8, MMP-9, MMP-12, TIMP-1, TIMP-4, and Serpin E1 were assessed utilizing a multiplex ELISA and remained constant as a result of fibrocyte ablation. Absolute concentrations and individual *p*-values are provided in Figure S4.

Next, a high-throughput analysis of 84 fibrosis-related genes was conducted; a table of the strongest regulated genes can be found in Figure S3. The gene expression of *Col1a2* (fold-change 0.88) and *Col3a1* (fold-change 0.88) was not relevantly altered in this array. Quantitative real-time PCR, moreover, demonstrated an unchanged expression of *Col1a1* in result of the fibrocyte ablation (fold-change 0.97; 95% CI: 0.74–1.26; *p* = 0.803; Figure 2e).

The extent of fibrosis is largely influenced by the degradation of ECM components. Quantitative analysis of the protein levels of several MMPs and TIMPs showed an about equal expression throughout both groups (Figure 2f). The mean group difference of all but one analyte (MMP-9) yielded 95% confidence intervals whose border values were considered irrelevant in the context of this study. For MMP-9, the high scatter in the data, possibly due to the short half-life of MMP-9, impeded a conclusive interpretation. While there is a negligible mean difference, the 95% confidence interval ranges from a considerable increase (40.0%) to a noteworthy decrease (50.4%) in the mean concentration. Absolute concentrations and individual *p*-values are provided in Figure S4. The gene expression of selected MMPs and TIMPs was not relevantly altered in the gene expression array either.
