The liver of animals treated with CCl₄ had a finely granulated surface macroscopically. As anticipated, the individual response to the fibrosis-induction protocol varied widely from animal to animal. Hence, based on histological analysis, CCl₄-treated rats were staged into three groups according to the percentage of fibrotic area compared with the total area of the LB: mild and moderate fibrosis (defined as the fibrotic area being < 6% of the total; n = 33), severe fibrosis (6 to 11%; n = 15) and cirrhosis (> 11%; n = 20). Control rats displayed no appreciable alterations in liver histology, and had an almost negligible amount of fibrous tissue. However, in fibrotic rats, there was progressive accumulation of ECM as a consequence of the continuous liver injury, evolving from a slight deposition of fibrosis mainly in the portal area (mild/moderate fibrosis) to numerous and thicker septa resulting from the longer exposure to CCl₄ (severe fibrosis). Finally, most of the animals exposed to the toxin for the longest periods developed cirrhosis, characterized by the formation of regenerative nodules of liver parenchyma separated by fibrotic septa. Figure 1 shows representative Sirius red staining of tissue from a control rat, a rat with mild/moderate fibrosis, a rat with severe fibrosis and a rat with cirrhosis.

The fibrotic/cirrhotic rats included in the study had important alterations in liver-function tests, being more pronounced in the group of rats with cirrhosis (Table 1). These animals also had a moderate but significant alteration in extracellular fluid homeostasis, as reflected by hyponatremia.

Western blotting analysis of serum samples from cirrhotic and control rats precipitated with the anti-CO3-610 antibody was carried out (Figure 2). No signal was detected in healthy animals, but a clear specific band at a molecular mass of approximately 25 kDa was visible in blots of serum from cirrhotic rats. These results indicate that a single molecular entity is detected with the CO3-610 antibody, in contrast to other serum collagen ELISA assays reported previously, in which several fragments of collagen were measured simultaneously 15 .

No difference was found between serum levels of CO3-610 between control rats and those with mild/moderate fibrosis (26.6 ± 1.3 ng/ml versus 28.5 ± 1.6 ng/ml, respectively) (Figure 3A). However, progression of liver fibrosis was associated with a parallel increase in the circulating levels of the collagen III-derived epitope; values significantly increased in rats with severe fibrosis (43.5 ± 3.3 ng/ml, P < 0.001) and reached maximum levels in rats with cirrhosis (60.6 ± 4.3 ng/ml, P < 00.01). There was a close direct relationship between CO3-610 values and hepatic collagen content in CCl₄-treated rats (r = 0.78; P < 0.001) (Figure 3B). These differences were clearly greater than those obtained by calculating the correlation coefficient between serum hyaluronic acid (HA) (a well-established serum marker of liver fibrosis) and the hepatic collagen content in fibrotic/cirrhotic rats (r = 0.49; P < 0.05).

Further insight into the suitability of CO-610 as a surrogate indicator of hepatic fibrosis was gained by measuring mean arterial pressure (MAP) and portal pressure (PP) in 20 of the 68 CCl₄-treated rats included in the protocol. This subset of animals did not differ from the whole group of fibrotic/cirrhotic rats in terms of fibrosis staging (seven rats with mild/moderate fibrosis, three rats with severe fibrosis and ten rats with cirrhosis), range of CO3-610 values (37.3 ± 5.6 ng/ml, 55.1 ± 6.3 ng/ml and 72.2 ± 6 ng/ml, respectively) or correlation coefficient with hepatic collagen content (r = 0.71; P < 0.001). As previously observed, the increased disruption in the hepatic architecture was associated with a progressive deterioration in systemic hemodynamics. In fact, animals with cirrhosis had frank hypotension (MAP 96.5 ± 2.4 mmHg; P < 0.001) compared with control rats (MAP 124.6 ± 2.4 mmHg). Accordingly MAP inversely correlated with serum CO3-610 values in CCl₄-treated rats (r = -0.59; P < 0.01) (Figure 4A). However, the most interesting finding of this investigation was that the serum concentration of CO3-610 depicted a higher direct relationship with the degree of portal hypertension in fibrotic/cirrhotic animals (r = 0.84; P < 0.001) (Figure 4B).

As expected, the expression of collagen 1α2 (Col1α2) and 3α1 (Col3α1) mRNA was generally enhanced in fibrotic/cirrhotic rats, but the pattern and degree of changes clearly differed between the two transcripts (Figure 5A). Collagen 1α2 mRNA expression increased in parallel with the worsening of the liver disease. In cirrhotic rats this enhancement was around 20 times higher than that in controls. This increase in the expression of collagen 3α1 mRNA reached its peak in the animals with fibrosis, and there was no worsening of this level in the liver of rats with cirrhosis. Because CO3-610 is a degradation product of collagen type III, we next examined the hepatic content of collagen type III in the liver of CCl₄-treated rats. Gene activation of collagen 3α1 mRNA was associated with a progressive deposition of collagen III fibers in the hepatic tissue of animals with induced fibrosis/cirrhosis protocol (Figure 5B),

Following hepatic injury, increased matrix deposition of collagen fibers results from an altered balance between the activity of cleavage proteases such as MMPs and their endogenous inhibitors, the TIMPs. Therefore, we sought to determine expression of the rat genes Mmp2, Mmp9, Timp1 and Timp2 in the liver of CCl₄-treated rats with different degrees of cirrhosis (Figure 5C). Reverse transcriptase PCR analysis revealed marked Mmp2 and Mmp9 mRNA overexpression in fibrotic rats compared with control animals. A similar degree of Mmp2 mRNA expression was apparent in cirrhotic rats; however, the abundance of the Mmp9 transcript was clearly attenuated in this group of animals. Timp1 and -2 mRNAs were also significantly overexpressed in the liver of CCl₄-treated rats, with the higher intensity being found in cirrhotic animals.

The study was performed on 68 male Wistar rats with different degrees of fibrosis and 20 control Wistar rats (Charles-River, Saint Aubin les Elseuf, France). Fibrosis was induced by repetitive CCl₄ inhalation 34 . The rats were fed ad libitum with standard chow and given water containing phenobarbital 0.3 g/L as drinking fluid. Animals were exposed to a CCl₄ vapor atmosphere twice a week, starting with 0.5 minutes per exposure. The duration of exposure was increased by 1 minute after every three session until it reached 5 minutes, which was used until the end of the investigation. To examine the effects of variable degrees of hepatic fibrosis, the CCl₄-treated rats were examined at weeks 8 (n = 13), 13 (n = 26), 16 (n = 13) and 19 (n = 16) after starting the fibrosis-induction protocol. For the analysis of results, all treated animals were grouped according to thresholds of histological fibrosis (Figure 1). Control rats were studied after similar periods of phenobarbital administration alone.

When scheduled, animals were anesthetized and a hemodynamic study was performed when indicated. A subgroup of 20 rats was selected to perform these measurements before termination to measure MAP and PP. A blood sample was then obtained from all animals to measure osmolality, electrolytes, standard parameters of liver function, and serum concentrations of HA and CO3-610. Thereafter, the animals were killed by isofluorane overdose (Forane, Abbott Laboratories S.A., Madrid, Spain), and hepatic samples were obtained from the middle liver lobe, which were immediately frozen in liquid nitrogen until further analysis of mRNA expression, or fixed in 10% buffered formalin for further analysis with hematoxylin and eosin stain, Sirius red stain and immunohistochemistry.

Rats were anesthetized with barbiturate 100 mg/kg body weight (Inactin^®; Sigma-Aldrich Chemie Gmbh, Steinherim, Germany) and a polyvinyl catheter (PE-50; ; Becton-Dickinson, Franklin Lakes, NJ, USA) was implanted in the left femoral artery. The arterial catheter was connected to a very sensitive transducer (Hewlett Packard, Avondale, PA, USA) that was calibrated before each study. Hemodynamic parameters were allowed to equilibrate for 30 minutes, and MAP and heart rate values were recorded for two periods of 30 minutes. Each value represented the average of two measurements. A midline abdominal incision (20 mm) was made, and a PE-50 catheter was placed in the portal vein through an ileocolic vein. After verifying free blood reflux, the catheter was fixed to the mesentery with cyanoacrylate glue, and PP was measured. Animals were killed by isofluorane overdose, and tissue specimens were collected.

Liver sections 4 μm thick were stained in 0.1% Sirius red F3B (Sigma-Aldrich, St. Louis, MO, USA) in saturated picric acid (Sigma-Aldrich). The relative fibrotic area, expressed as a percentage of total liver area, was assessed by analyzing 36 fields of Sirius red-stained liver sections per animal. Each field was acquired at 10× magnification with a microscope ( E600; Nikon, Tokyo, Japan) and digital camera (RT-Slider Spot; Diagnostic Instruments, Sterling Heights, MI, USA). Results were analyzed using imaging software (ImageJ, NIH). To evaluate the relative fibrosis area, the measured collagen area was divided by the net field area and then multiplied by 100. Subtraction of the vascular luminal area from the total field area yielded the net fibrosis area 35 .

Serum samples from rats treated with CCl₄ for 19 weeks and from non-treated animals were analyzed in parallel. Briefly, 1.5 μg/well of biotinylated anti-CO3-610 antibody was used to coat streptavidin 96-well plates (Thermo Scientific Pierce, Rockford, IL, USA). Serum samples (125 μL) were diluted 1:1 in immunoprecipitation buffer, added to the wells, and incubated for 1 hour at room temperature. Antibody-antigen complexes were removed from non-specific binding, and the antigen eluted from the plates according to the manufacturer's instructions. For western blotting analysis, 1.4 μl of DTT 1 mol/l was added to 35 μL of each eluted sample, and then heated for 5 minutes at 95°C. The heated samples were loaded onto 15% SDS gels, resolved, and transferred to nitrocellulose membranes using 60 mA for 90 minutes. Membranes were blocked overnight in 5% milk, and incubated for a further 2.5 hours with horseradish peroxidase-labeled anti-CO3-610 antibody (diluted 1:100). After three washes of 10 minutes each, membranes were incubated for a further 1 hour with anti-mouse IgG HRP-linked secondary antibody (diluted 1:5000) followed by another three washes of 10 minutes each. Membranes were then developed with autoradiography films (GE Healthcare, Piscataway, NJ, USA) for 30 minutes.

Details on the design and production of the monoclonal antibody against CO3-610 and the development of the serum CO3-610 ELISA assay have been previously published 13 . In brief, 100 μl of 2.5 ng/mL C-term biotinylated peptide NB51 (KNGETGPQGP) in PBS-Tris-borate-EDTA were used to coat streptavidin plates for 30 minutes at 20°C on a 300 rpm shaker, and then removed with washing buffer. Serum samples (20 μl) were diluted eightfold in incubation buffer (10 mmol/l: 400 mmol/l Tris:Bis-Tris buffer) and added to the plates. Next, 100 μL of CO3-610 peroxidase conjugated antibody solution (1:40,000 dilution) was added to the plates and left to incubate overnight at 4°C, on a 300 rpm shaker. The next day, plates were washed five times in washing buffer, then 100 μl of 3,3',5,5'-tetramethylbenzidine was added, and the plates incubated in darkness for 15 minutes at 20°C on a 300 rpm shaker. The reaction was stopped with the addition of 100 μL of stop solution, and plates were read on an ELISA reader at 450 nm, with 650 nm as reference.

Total RNA was extracted from frozen liver specimens of control and fibrotic rats (RNeasy Kit; Qiagen, Hilden, Germany), then 1 μg of total RNA was reverse transcribed using a complementary DNA synthesis kit (High-Capacity cDNA Reverse Transcription Kit; Applied Biosystems, Foster City, CA, USA). RNA concentration was determined by spectrophotometric analysis (ND-100 Spectrophotometer, Nanodrop Technologies Inc., Wilmington, DE, USA). Total RNA (1 μg) was reverse transcribed using the same complementary DNA synthesis kit described above. Specific primers and probes used for the different genes studied were designed to include intron spanning, using the Universal Probe Library Assay Design Center through the ProbeFinder software (version 2.45; Roche Diagnostics, Indianapolis, IN, USA; https://www.roche-applied-science.com/sis/rtpcr/upl/index.jsp). The primers used are shown in Table 2. Primers for rat were designed according to GenBank sequences (NM_053356.1, NM_032085.1, NM_053819.1, NM_021989.2, NM_031054.2, NM_031055.1 and NM_012583.2, respectively) and probes were designed using ProbeFinder software as above.

Real-time quantitative PCR reactions were performed in duplicate and analyzed (Light Cycler 480; Roche Diagnostics). The PCR reaction mix was 10 μl total volume, comprising 1:8 cDNA dilution, 200 nmol/l primer dilution, 100 nmol/l pre-validated nine-mer probe (Universal ProbeLibrary, Roche Diagnostics) and a master kit (FastStart Taqman Probe Master Kit; Roche Diagnostics). The hypoxanthine phosphoribosyltransferase 1 gene (Hprt1) was used as a reference gene for normalization (Table 2), and water was used as negative control.

Relative quantification was calculated using the comparative threshold (CT) cycle, which is inversely related to the abundance of mRNA transcripts in the initial sample. The mean of CT duplicate measurements was used to calculate ΔCT as the difference in CT for target and reference. The relative quantity of product was expressed as fold-induction of the target gene compared with the reference gene according to the formula 2^-ΔΔCT, where ΔΔCT represents ΔCT values normalized with the mean ΔCT of control samples.

Sections of frozen liver tissue 5 μm thick were cut and fixed in acetone for 10 minutes, then dried overnight at room temperature. Slides were washed in water for 5 minutes. Anti-type III collagen ab7778 primary antibody (Abcam, Cambridge, UK) was diluted 1:50 with 1% bovine albumin, and incubated on the slides for 30 minutes at room temperature, then removed with 0.1% Triton X-100 in PBS (two washes of 5 minutes each). Slides were then incubated with secondary horseradish peroxidase Envision anti-rabbit IgG antibody (Dako, Glostrup, Denmark) for 30 minutes, and washed with 0.1% Triton X-100 (2 × 5 minutes). Red staining was produced by incubation with 3-amino-9-ethyl carbazole substrate (Vector Laboratories, Burlingame, CA, USA) for 30 minutes. Slides were counterstained with Meyers hematoxylin.

Serum levels of HA were quantified with an enzyme-linked binding-protein assay according to the manufacturer's instructions (Corgenix Inc, Westminster, CO, USA). Serum osmolality was determined from osmometric depression of the freezing point (Osmometer 3300; Advanced Instruments, Needham Heights, MA, USA) and Na⁺ and K⁺ concentrations by flame photometry (IL 943; Instrumentation Laboratory, Lexington, MA, USA). Serum albumin, alanine transaminase and lactate dehydrogenase were measured by a chemical analyzer (ADVIA 2400; Siemens Healthcare Diagnostics, Tarrytown, NY, USA). Quantitative data were analyzed using GraphPad Prism 5 (GraphPad Software, Inc., San Diego, CA, USA). Differences between mean values obtained from CCl₄-treated and control groups were analyzed using one-way ANOVA with Newman-Keuls post hoc test, and Kruskal--Wallis test with Dunn post hoc tests when appropriate. Correlations between serum CO3-610 values and the rest of the variables studied were analyzed with the Pearson two-tailed test. Data was expressed as mean ± SEM, and P≤0.05 was considered significant.