Increased hepatic expression of TLR2 and TLR4 in the hepatic inflammation-fibrosis-carcinoma sequence

Abstract

We evaluated expression of TLR2, TLR4 and proinflammatory genes [NF-κB, TNF-α, cyclooxygenase-2 (COX-2)] in liver samples of patients in different stages of liver disease. Fifteen patients with unexplained transaminases elevation (reference group), 22 with viral chronic hepatitis (hepatitis group), 14 with virus-induced severe fibrosis/cirrhosis (cirrhosis group) and 10 with hepatocarcinoma (hepatocarcinoma group) were consecutively included in the study. Quantification of TLR2, TLR4, NF-κB, TNF-α and COX-2 mRNA was done by real-time RT-PCR and TLR2 and TLR4 protein expression was evaluated by immunohistochemistry. Compared with reference, TLR2 and TLR4 mRNA was increased in hepatitis (TLR2: 2.66 ± 0.69; TLR4: 3.11 ± 0.79; P < 0.05) and cirrhosis (TLR2: 2.14 ± 0.5; TLR4: 1.74 ± 0.27; P < 0.05) and decreased in hepatocarcinoma (TLR2: 0.48 ± 0.15; TLR4: 0.54 ± 0.10; P < 0.05). This associated with increased TNF-α and COX-2 mRNA in hepatitis (TNF-α: 3.24 ± 0.79; COX-2: 2.47 ± 0.36; P < 0.05) and cirrhosis (TNF-α: 1.73 ± 0.28; COX-2: 1.8 ± 0.35, P < 0.05), whereas NF-κB mRNA was increased in hepatitis (2.42 ± 0.31; P < 0.05) and unchanged in cirrhosis (1.34 ± 0.17; P = 0.3). Hepatocarcinoma presented increased COX-2 mRNA (1.63 ± 0.15; P < 0.05) and maintained (at decreased levels) mRNA of NF-κB (0.52 ± 0.12) and TNF-α (0.52 ± 0.12; P < 0.05, all genes). Immunohistochemistry confirmed increased expression of TLR2 and TLR4 in hepatitis and cirrhosis and maintained expression in hepatocarcinoma. Upregulation of TLR2, TLR4 and their proinflammatory mediators is associated with virus-induced hepatic IFC sequence.

Keywords

Chronic hepatitis cirrhosis hepatocarcinoma TLR2 TLR4

Introduction

Chronic liver inflammation, irrespective of the underlying cause (metabolic, immune-driven or virus-induced), leads to fibrosis and/or cirrhosis, which are precancerous states in which the development of hepatocarcinoma is more likely. Some authors call this sequence the hepatic inflammation-fibrosis-carcinoma (IFC) sequence.¹ Nevertheless, the cellular and molecular effectors mediating the interplay between the components of hepatic IFC sequence continue to be largely unknown.

Recently, several studies implicated TLRs as potential key orchestrators of the hepatic IFC sequence.^2–5 TLRs are one of the most representative classes of pathogen-associated molecular patterns (PAMPs) receptors that play a critical role in innate immunity activation.^6,7 The human TLR family consists of 10 members that enable the innate immunity system to recognize different groups of pathogens while initiating appropriate and distinct immunological responses according to the recognized PAMP.^6,7 Besides immune cells, most liver cells (hepatocytes, Kupffer cells and stellate cells) also express TLRs and respond to their ligands.^8,9 TLR2 and TLR4 have been the most studied TLRs in liver diseases as they sense bacterial components and, thus, may mediate liver injury associated with increased bacterial translocation that is present in many liver diseases.^10–12 TLR2 is essential for the innate immune response to Gram-positive bacteria, being activated by bacterial lipoproteins and peptidoglycan.^6,7 TLR4 acts as a receptor for LPS, a cell wall component of Gram-negative bacteria.^6,7 Besides exogenous ligands, TLR2 and TLR4 may also sense endogenous ligands initiating danger signals, such as high mobility group box 1, hyaluronan and heat shock protein 60, inducing an inflammatory response in the absence of microbial challenge.¹³ Stimulation of these two receptors initiates a signaling cascade that promotes activation of NF-κB and MAPK and, consequently, production of different pro-inflammatory mediators, such as TNF-α and cyclooxygenase-2 (COX-2). ^14–18

Recent animal studies and in vitro hepatocyte culture models suggest that TLR2 and TLR4 may play a key role in the hepatic IFC sequence. Modulation of TLR2 and/or TLR4 function was shown to influence liver inflammation in chronic liver diseases, such as alcoholic liver disease (ALD), nonalcoholic fatty liver disease (NAFLD), chronic hepatitis C and chronic hepatitis B.^2,3

There is also accumulating evidence that TLR4-induced activation and sensitization of hepatic stellate cells (HSCs) may constitute an important molecular link between hepatic inflammation and fibrogenesis.^19–22 Moreover, a recent study has revealed TLRs, in particular TLR4, as major factors linking hepatic chronic inflammation and hepatocarcinoma.²³

However, to date, the suggested implication of TLR2 and TLR4 in the pathogenesis of hepatic IFC sequence is principally based on evidence obtained from animal studies or in vitro hepatocyte culture models. Studies using diseased human liver tissue to confirm or refute the in vitro and animal findings are scarce and have evaluated TLR2 and TLR4 in each stage of IFC sequence separately.

Therefore, in the present study, we evaluated the expression of TLR2 and TLR4 in liver samples from patients in each stage of virus-induced hepatic IFC sequence. The expression of NF-κB, TNF-α and COX-2 was also evaluated in order to characterize their association with TLR2 and TLR4 expression.

Materials and methods

Patients and biological samples

This study included patients from two hospitals of the North of Portugal (Braga Hospital and Portuguese Oncology Institute of Porto). The study protocol respected the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the Ethics Committee of Braga Hospital and Portuguese Oncology Institute of Porto. Informed consent was obtained from each patient.

Patients were recruited consecutively during 2009. We defined four groups: reference, hepatitis, cirrhosis and hepatocarcinoma. Reference group included patients followed in the Hepatology Outpatient Clinic of Braga Hospital who underwent liver biopsy because of chronic unexplained transaminase elevation. We excluded from this group patients with alcohol abuse (>30 g/d in males; >20 g/d in females); analytical or histologic findings favoring hemocromatosis; autoimmune hepatitis; primary biliary cirrhosis; primary sclerosing cholangitis; and HIV infection or clinical, analytical, imagiological or histologic evidence of severe fibrosis/cirrhosis (METAVIR F3–4). The hepatitis group included chronic hepatitis B or C patients followed in the Hepatology Outpatient Clinic of Braga Hospital who underwent staging liver biopsy. The cirrhosis group was selected from the same group of patients but with histologic evidence of severe fibrosis/cirrhosis (METAVIR F3–4). The hepatocarcinoma group included chronic hepatitis B or C patients with diagnosis of hepatocarcinoma (according to the EASL 2000 Barcelona Guidelines²⁴) followed in the outpatient clinic of Portuguese Oncology Institute of Porto who underwent surgical resection of hepatocarcinoma. In these groups, patients must have had >18 years serological evidence of chronic hepatitis B (HBsAg+) or C (HCVAb+) and clinical stability. Histologic evidence of cirrhosis and hepatocarcinoma was required in the hepatocarcinoma group. Patients with HIV infection or analytical or histologic findings suggestive of liver disease other than viral chronic hepatitis were excluded.

Before liver biopsy or surgical intervention, blood samples were drawn from fasting patients for routine analysis (complete blood count, glucose, electrolytes, renal and liver function tests, and coagulation study) and viral load quantification. Liver tissue was obtained by percutaneous biopsy using a 16-gauge Menghini needle or by transjugular biopsy. In the hepatocarcinoma group, we obtained hepatocarcinoma tissue [for mRNA quantification (in 6 patients) and immunohistochemical evaluation (in 10 patients)] and adjacent liver tissue (for cirrhosis confirmation) from the surgical specimen. The collected tissue was divided into two fragments: one was immediately placed in RNAlater (Ambion, Madrid, Spain) and stored at −80°C for mRNA isolation and quantification; the other was fixed in 10% buffered formalin and embedded in paraffin for histologic and immunhistochemical analyses.

mRNA isolation and quantification of TLR2, TLR4, NF-κB, TNF-α and COX-2

Total mRNA was extracted from tissue samples using the TriPure isolation reagent according to the manufacturer’s instructions (Roche, Mannheim, Germany). Concentration and purity were assayed by spectrophotometry (Eppendorf 6131000.012, Madrid, Spain). Two-step real-time RT-PCR was used to perform relative quantification of mRNA. For each studied mRNA molecule, standard curves were generated from the correlation between the amount of starting total mRNA and the PCR threshold cycle of graded dilutions from a randomly-selected sample from reference group. For relative quantification of specific mRNA levels, 100 ng of total mRNA from each sample underwent two-step real-time RT-PCR. GAPDH mRNA levels were similar in all experimental groups, which enabled the use of this gene as an internal control. RT (20 µl; 10 min at 22°C, 50 min at 50°C and 10 min at 95°C) was performed in a standard thermocycler (Whatman Biometra 050-901, Goettingen, Germany). Five percent of the cDNA yield was used as a template for real-time PCR (LightCycler II, Roche) using SYBR green (Qiagen 204143, Madrid, Spain), according to the manufacturer’s instructions. Specific PCR primers pairs for the studied genes (GAPDH, TLR2, TLR4, NF-κB, TNF-α and COX-2) are presented in Table 1. Results of mRNA quantification were expressed as an arbitrary unit (AU) set as the average value of reference group, after normalization for GAPDH.

Table 1.

Specific PCR primers pairs for the studied genes.

Gene	Primers
GAPDH	F: 5′ – GGT GGT CTC CTC TGA CTT CAA CA – 3′
GAPDH	R: 5′ – GTT GCT GTA GCC AAA TTC GTT GT – 3′
TLR2	F: 5′ – GAT CCC AAC TAG ACA AAG ACT – 3′
TLR2	R: 5′ – CTG CGG AAG ATA ATG AAC ACC – 3′
TLR4	F: 5′ – CCA TAA AAG CCG AAA GGT GAT TGT – 3′
TLR4	R: 5′ – AGA TGT GCC GCC CCA GGA C – 3′
NF-κB	F: 5′ – CCT GGA TGA CTC TTG GGA AA – 3′
NF-κB	R: 5′ – TCA GCC AGC TGT TTG ATG TC – 3′
COX-2	F: 5′ – ACC GGG GGT ATA CTA CGG TC – 3′
COX-2	R: 5′ – ACG GGC CCT ATT TCA AAG AT – 3′
TNF-α	F: 5′ – GGT TTG CTA CAA CAT GGG CTA – 3′
TNF-α	R: 5′ – AAG AGT TCC CCA GGG ACC TCT C – 3′

F = Forward primer, R = Reverse primer.

Immunohistochemical evaluation of TLR2 and TLR4

Tissue specimens were fixed in 10% neutral buffered formalin for 24 h and paraffin embedded. De-paraffinized tissue slides were submitted to antigen retrieval using a high temperature antigen unmasking technique in a water bath, 95°C in citrate buffer pH 6.0, for 20 min. Endogenous peroxidase activity was blocked by incubating the slides with freshly prepared 0.5% H₂O₂ in distilled water for 20 min. After washing the slides in distilled water and PBS/0.05% Tween 20 solution, immunostaining was performed using an immunoperoxidase method according to the manufacturer’s instructions. The slides were incubated with normal horse serum (Vector Laboratories, Burlingame, CA, USA) 1/50 in PBS-BSA 1% at room temperature (21–23°C) for 20 min in humid chamber. Sections were then incubated with primary Ab at 4°C for 16–18 h. The following primary Abs were used: rabbit polyclonal Ab anti-TLR2 (H-175, 1 : 50 dilution, Santa Cruz Biotechnology, CA, USA) and rabbit polyclonal anti-TLR4 (H-80, 1 : 100 dilution, Santa Cruz Biotechnology). The slides were then rinsed in PBS/0.05% Tween 20 solution and bound Ab was detected by applying biotinylated secondary Ab (Vectastain Universal Elite ABC Kit, Vector Labs, Peterborough, UK) for 30 min. After washing the slides with PBS/0.05% Tween 20 solution the slides were incubated with ABC reagent (Vectastain Universal Elite ABC Kit) for 30 min. The slides were washed in PBS and incubated for 7 min in 3,3-diaminobenzidine (DAB; Sigma-Aldrich, St Louis, MO, USA) 0.05 g/PBS, 0.03% H₂O_2. Following counterstaining with hematoxylin for 20 s, the slides were washed for 4 min in water, dehydrated and mounted with Entellan (Merck KGaA, Darmstadt, Germany). Normal gastric mucosa and lymph node tissue were used as negative and positive controls, respectively. An Ab diluent (non-immune IgG, TA-125-UD; Thermo Scientific, Erembodegem, Belgium) was used in some samples as additional negative control, confirming the specificity of our protocol. We evaluated immunostaining of hepatocytes for TLR2 and TLR4 in all the samples. In order to quantify TLR expression in tissue samples, three parameters were considered: (i) sample positivity—a sample was considered positive if hepatocytes were clearly marked by the Ab; (ii) grade of expression—a score of 0–3 was considered according to the number of epithelial cells marked (0 = no cells; 1 = less than 10% of epithelial cells; 2 = 10–75% cells; 3 = more than 75% cells); and (iii) intensity of expression—a score of 0–3 was considered according to a subjective evaluation of the intensity of marked cells (0 = no immunostaining; 1 = weak positive staining; 2 = moderate positive staining; 3 = strong positive staining). Immunohistochemical evaluation was performed independently by two experienced pathologists.

Statistical analysis

Data analysis was performed using the computer software SPSS for Windows (version 17.0; Chicago, IL, USA). Data are presented as mean ± standard error of mean (SEM) or as median and range, according to the type of distribution. Student’s t-test was used for comparison between groups. When necessary, the test was preceded by a natural logarithm transform to obtain a normal distribution. Correlation between TLR2 and TLR4 mRNA expression and viral load, necroinflammatory activity or transaminases levels was evaluated by univariate analysis. Statistical significance was set at P < 0.05.

Results

Baseline characteristics of patients

The baseline characteristics of patients are shown in Table 2. A total of 61 patients were included in the study: 15 patients in reference group, 22 (10 HBsAg+ and 12 HCVAb+) patients in hepatitis group, 14 (7 HBsAg+ and 7 HCVAb+) patients in cirrhosis group and 10 (4 HBsAg+ and 6 HCVAb+) patients in hepatocarcinoma group. Histologic findings in reference group included: steatohepatitis (seven patients); macrovesicular steatosis (three patients); perivenular cholestasis (two patients); granulomatous hepatitis (two patients); normal findings (one patient). Patients with steatohepatitis or macrovesicular steatosis were considered to have NAFLD. As expected, the cirrhosis and hepatocarcinoma groups had significantly higher levels of bilirubin and international normalised ratio (INR) and lower levels of albumin.

Table 2.

Baseline characteristics of patients.

Group parameter	Reference	Hepatitis	Cirrhosis	Hepatocarcinoma
n	15	22	14	10
Age	48 ± 5	41 ± 2	48 ± 3	72 ± 2
Male/female	8/7	10/12	9/5	7/3
AST (U/l) (10-36 U/l)¹	50 ± 7^#	33 ± 4^#	80 ± 14	35 ± 6^#
ALT (U/l) (10-30 U/l)¹	81 ± 10	44 ± 6*^,#	108 ± 26	29 ± 5*^,#
Bilirubin (mg/dl) (0.2-1.0 mg/dl)¹	0.57 ± 0.15	0.59 ± 0.07	1.20 ± 0.18^§	1.43 ± 0.10^§
Albumin (g/dl) (3.5–5.2 g/dl)¹	4.4 ± 0.1	4.5 ± 0.1	3.9 ± 0.2^§	3.6 ± 0.3^§
INR	1.04 ± 0.02	1.08 ± 0.02	1.20 ± 0.03^§	1.22 ± 0.02^§
HBsAg+	–	10	7	4
HBeAg+/HBeAg-	–	1/9	0/7	0/4
HBV DNA load (IU/ml)	–	3091 (<200– >20000000)	2325340 (3300– >20000000)	4300340 (2900– >20000000)
HCVAb+	–	12	7	6
HCV genotype 1/2/3/4	–	7/3/0/2	4/1/1/1	5/0/0/1
HCV RNA load (IU/ml)	–	834037 (9232 –19907580)	706025 (174044 –4803266)	804029 (182055 −4803266)
METAVIR Grade A0/A1/A2/A3	–	0/13/8/1	0/0/8/6	–
METAVIR Stage F0/F1/F2/F3/F4	–	6/10/6/0/0	0/0/6/8	–

Values are presented as mean ± SEM or as median and range according to the type of distribution. ¹Normal range. *P < 0.05 vs reference group; ^#P < 0.05 vs cirrhosis group; ^§P < 0.05 vs reference and hepatitis groups.

ALT = alanine aminotransferase, AST = aspartate transaminase, INR = international normalized ratio, HBV = hepatitis B virus, HCV = hepatitis C virus.

mRNA expression of TLR2, TLR4, NF-κB, TNF-α and COX-2

Quantifications of TLR2, TLR4, NF-κB, TNF-α and COX-2 mRNA are shown in Figure 1. In the hepatitis group, expression of TLR2 (2.66 ± 0.69, P = 0.04) and TLR4 (3.11 ± 0.79, P = 0.03) were greatly increased. This was associated with increased expression of NF-κB (2.42 ± 0.31, P = 0.0003), TNF-α (3.24 ± 0.79, P = 0.02) and COX-2 (2.47 ± 0.36, P = 0.003). Compared to the reference group, this increased inflammatory profile (with exception of NF-κB) persisted in the cirrhosis group (TLR2: 2.14 ± 0.5, P = 0.04; TLR4: 1.74 ± 0.27, P = 0.008; NF-κB: 1.34 ± 0.17, P = 0.3; TNF-α: 1.73 ± 0.28, P = 0.009; COX-2: 1.8 ± 0.35, P = 0.04), despite a global, but not significant (except for NF-κB), decrease in expression of all the genes when compared with the hepatitis group. In the hepatocarcinoma group, all samples were positive for all studied genes. Compared with the reference group, the hepatocarcinoma group presented a higher mRNA expression of COX-2 (1.63 ± 0.15; P = 0.02), similar to the hepatitis and cirrhosis groups, and lower mRNA expression of TLR2 (0.48 ± 0.15), TLR4 (0.54 ± 0.10), NF-κB (0.52 ± 0.12) and TNF-α (0.52 ± 0.12; P = 0.01, all genes). There were no differences between chronic hepatitis B or C patients (P > 0.05, all genes). We did not find any difference or tendency when comparing the genetic profile between F0, F1 or F2 patients (P > 0.05, all genes). No correlation was seen between mRNA expression of any gene and viral load, necroinflammatory activity or transaminases levels.

Figure 1.

mRNA quantification of TLR2, TLR4, NF-κB, TNF-α and COX-2, in the reference (mainly composed of patients with NAFLD), hepatitis (composed of patients with chronic hepatitis B or C), cirrhosis (composed of patients with post-chronic hepatitis B or C cirrhosis) and hepatocarcinoma (composed of patients with post chronic hepatitis B or C cirrhosis-related hepatocarcinoma) groups. Levels of mRNA are expressed as arbitrary unit (AU) set as the average value of control group after normalization for GAPDH. Results are presented as mean ± standard error of mean (SEM). *P < 0.05 vs reference group; ^#P < 0.05 vs cirrhosis group; ^§P < 0.05 vs hepatitis group.

Immunohistochemical evaluation of TLR2 and TLR4

All the samples, including those from the hepatocarcinoma group, were positive for TLR2 and TLR4. Compared with the reference group, TLR2 and TLR4 expression grade was maintained in the hepatitis group and reduced in the cirrhosis and hepatocarcinoma groups, while TLR2 and TLR4 expression intensity was increased in the hepatitis and cirrhosis groups. Moreover, in the reference group, cytoplasmic staining of hepatocytes for TLR2 and TLR4 was very heterogeneous, while in the hepatitis and cirrhosis groups, hepatocytes showed a diffuse cytoplasmic staining for TLR2 and TLR4. Compared with the hepatitis and cirrhosis groups, TLR2 expression grade and intensity and TLR4 expression grade were reduced, while TLR4 expression intensity was maintained in the hepatocarcinoma group. In most samples of hepatocarcinoma tissue there were well-differentiated areas with high staining alternating with poorly differentiated areas with low staining. There were no differences between chronic hepatitis B or C patients (P > 0.05, all proteins). These data are presented in Figure 2 and Table 3.

Table 3.

Immunohistochemichal evaluation of TLR2 and TLR4: grade and intensity of expression.

Group parameter	Reference (n = 15)	Hepatitis (n = 22)	Cirrhosis (n = 14)	Hepatocarcinoma (n = 10)
TLR2 grade	3.0	3.0	2.87 (2.80–2.94)	2.2 (2.00–2.40)
TLR2 intensity	1.73 (1.55–1.91)	2.27 (2.17–2.37)	2.37 (2.25–2.49)	1.9 (1.73–2.07)
TLR4 grade	3.0	3.0	2.87 (2.80–2.94)	2.6 (2.44–2.76)
TLR4 intensity	1.8 (1.63–1.97)	2.41 (2.31–2.51)	2.44 (2.22–2.36)	2.1 (1.87–2.33)

Values are presented as mean (95%CI).

Figure 2.

Immunohistochemical evaluation of TLR2 (left) and TLR4 (right). (A, B) Low power field magnification for TLR2 and TLR4 in the reference group (mainly composed of patients with NAFLD): there is a higher intensity staining in acinar zones 1 and 3 than in acinar zone 2. (C, D) High power field magnification for acinar zone 2 in the reference group showing sparse cytoplasmatic staining. (E, F) High and low power field magnification, respectively, in chronic hepatitis B or C (acinar zones 1 and 2) showing diffuse cytoplasmatic staining (in contrast with A–D images). (G, H) Low power field magnification in post-chronic hepatitis B or C cirrhosis: all hepatocytes are stained with a diffuse cytoplasmic staining. (I, J) Low power field magnification in post-chronic hepatitis B or C cirrhosis-related hepatocarcinoma: low intensity staining in poorly-differentiated areas of hepatocarcinoma with multiple bizarre cells (I) in contrast to nodular well-differentiated areas with high intensity staining in hepatocarcinoma cells (J). In most samples of hepatocarcinoma tissue there were well-differentiated areas with high staining (J) alternating with poorly-differentiated areas with low staining (I).

Discussion

In the present study, we evaluated the expression of TLR2 and TLR4 in liver samples from patients in each stage of virus-induced hepatic IFC sequence. We found increased TLR2 and TLR4 mRNA and protein expression in virus-induced chronic hepatitis and cirrhosis, and a maintained TLR2 and TLR4 protein expression in virus-induced hepatocarcinoma.

Although there are several animal and in vitro studies implicating TLRs in the pathogenesis of hepatic IFC sequence, studies using diseased human liver tissue are scarce. Hepatic expression of TLR2 was shown to be maintained in early- and late-stage biliary atresia²⁵ and HCV cirrhosis,^26,27 increased in primary biliary cirrhosis (PBC) and non-alcoholic steatohepatitis (NASH),²⁸ and decreased²⁷ or maintained²⁶ in alcoholic cirrhosis. Hepatic expression of TLR4 was shown to be maintained in early- and late-stage biliary atresia²⁵, alcoholic and HCV cirrhosis,^26,27 and increased in PBC²⁸ and NASH.^28,29 Besides contradictory results, none of these studies have evaluated simultaneously TLR expression in different stages of liver disease.

To our knowledge the present study is the first to evaluate hepatic TLR2 and TLR4 expression at different stages of the virus-induced hepatic IFC sequence. We found increased TLR2 and TLR4 mRNA and protein expression in virus-induced chronic hepatitis and cirrhosis and a maintained TLR2 and TLR4 protein expression in virus-induced hepatocarcinoma. Thus, upregulation of TLR2 and TLR4 is an early, and persistent, event in the virus-induced hepatic IFC sequence.

Regarding TLR2 and TLR4 protein expression, the differences between the groups were more evident in terms of intensity of expression (which reflects the level of expression per cell) than in terms of grade of expression (which reflects the number of cells expressing the protein). The intensity of TLR2 and TLR4 proteins expression was in line with TLR2 and TLR4 mRNA expression, while the grade of TLR expression changed little between the groups, not accompanying the changes in mRNA expression. This finding suggests that the virus-induced hepatic IFC sequence is associated with changes in the level of TLR2 and TLR4 protein expression per cell and not with changes in the number of cells expressing these proteins.

Interestingly, we found reduced TLR2 and TLR4 mRNA and protein expression in the hepatocarcinoma group when compared with the hepatitis and cirrhosis groups. Although we have no definite explanation for the reduced expression of TLR2 and TLR4 in hepatocarcinoma cells, we believe it is likely a consequence of loss of differentiation of hepatocarcinoma cells. This is suggested by the finding that in most samples of hepatocarcinoma tissue there were well-differentiated areas with high staining alternating with poorly differentiated areas with low staining. This finding likely means that at late stages of the hepatic IFC sequence, the role of these receptors in this sequence becomes smaller.

However, our study has some limitations. Firstly, most patients included in the reference group have evidence of NAFLD and it was demonstrated that NAFLD is associated with increased hepatic TLR2 and TLR4 mRNA expression.^28,29 This suggests that the increase in hepatic expression of TLR2 and TLR4 in chronic hepatitis, cirrhosis and hepatocarcinoma may, in fact, be underestimated. Moreover, the hepatitis, cirrhosis and hepatocarcinoma groups included both patients with HBV infection or HCV infection. Nonetheless, statistical analysis revealed no difference between HBV and HCV patients. Moreover, as we included only patients with virus-induced chronic hepatitis in this study, our data cannot be generalized to other chronic hepatic diseases that follow IFC sequence. Another limitation of our study is that the method we used for quantification of protein expression was semi-quantitative. Although this could contribute to the lack of difference in terms of grade of protein expression, we were able to obtain significant differences in terms of intensity of protein expression. The validity of the protein quantification by this method is supported by the overall agreement between protein (grade of expression) and mRNA data. Furthermore, the method was validated in a previous study with consistent results.³⁰

In the present study we have not explored the mechanisms underlying increased hepatic expression of TLR2 and TLR4. Nonetheless, previous studies have shown that HBV and HCV may upregulate TLR2 and TLR4 through direct and indirect mechanisms. In vitro studies have shown that HCV nonstructural protein NS5A upregulates TLR4 expression and that HBeAg upregulates TLR2 expression.^23,31 In chronic hepatitis B and C, besides up-regulated expression of TLR2 and TLR4 by the virus, other factors, such as augmented exposure to their ligands, can also contribute to increased activation of these TLRs, especially in later stages of hepatic fibrosis and cirrhosis. In fact, several studies have demonstrated that bacterial translocation is increased in patients with cirrhosis, resulting in augmented exposure of hepatic TLRs to their ligands.^10,11

Herein, we did not search for liver cell-specific expression of TLR2 and TLR4, but instead we have focused on total hepatic mRNA expression and protein expression of hepatocytes. This may be an important issue as it has been demonstrated that HCV and HBV may affect TLRs expression in a cell-specific manner.^2–5 Nevertheless, the immunohistochemistry that we performed in this study suggests that, at least in part, this increase in TLRs expression occurs significantly in hepatocytes.

Previous studies on human samples have shown that hepatic expression of TLR2 and TLR4 in HCV cirrhosis was unchanged compared with the reference group, which is in disagreement with our results.^26,27 The reasons for this disagreement are unclear, but our data are consistent with a previous in vitro study showing that hepatocyte-specific transgenic expression of the HCV nonstructural protein NS5A upregulates TLR4 expression.²³ Regarding chronic hepatitis B, Visvanathan et al.³¹ have shown that expression of TLR2 on hepatocytes and Kupffer cells was significantly reduced in patients with HBeAg-positive chronic hepatitis B in comparison with HBeAg-negative chronic hepatitis B and controls, whereas it was significantly increased in HBeAg-negative chronic hepatitis B compared with controls. The level of TLR4 expression did not differ significantly among the groups. Downregulation of TLR2 was also demonstrated in HepG-2 cells transduced with wild-type HBV (HBeAg-positive) but not in cells transduced with pre-core mutant HBV (HBeAg-negative). Regarding TLR2, our data are consistent with the study by Visvanathan et al.³¹ as most of our chronic hepatitis B patients were HBeAg-negative and have increased TLR2. We could not compare HBeAg-positive and HBeAg-negative patients owing to the limited number of HBeAg-positive patients in our study.

In order to clarify TLR2- and TLR4-induced proinflammatory genes expression we also studied NF-κB, TNF-α and COX-2 mRNA expression. We found that hepatic TNF-α and COX-2 mRNA expressions are increased in virus-induced chronic hepatitis and cirrhosis, whereas hepatic NF-κB mRNA is increased in virus-induced chronic hepatitis, but maintained in virus-induced cirrhosis. This is an interesting finding as these proinflammatory genes have been implicated in hepatic inflammation, fibrogenesis and carcinogenesis interplay.^1,32,33 Interestingly, hepatocarcinoma samples presented increased COX-2 expression, despite lower expression of other studied genes, suggesting, in line with previous studies, that this enzyme may have an important role in hepatocarcinogenesis.³³ Although we have not investigated the functionality of TLR2 and TLR4, increased expression of TLR2 and TLR4 proteins and of NF-κB, TNF-α and COX-2 (key mediators of TLR2 and TLR4 signaling pathway) mRNA expression suggest augmented signaling of TLR2 and TLR4. When comparing hepatitis with cirrhosis we found a tendency towards lower expression of inflammatory genes mRNA; however, we cannot dismiss the possibility that this tendency is not related to a higher inflammatory cell infiltrate observed in the hepatitis group. In fact, we did not find any clear tendency in the hepatocyte immunohistochemistry results, suggesting that the difference, if any, between the two groups is not significant.

In summary, in patients with HCV or HBV chronic infection, hepatic expression of TLR2 and TLR4 is increased in chronic hepatitis and cirrhosis and is maintained in hepatocarcinoma. This is associated with increased TLR2- and TLR4-induced proinflammatory gene expression. Overall, this study suggests that TLR2 and TLR4 may be key players in the human hepatic IFC sequence associated with viral chronic hepatitis.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Acknowledgments

We are sincerely grateful to Antónia Teles for her technical support in this study.

Conflict of interest statement

The authors declare that there is no conflict of interest.

References

Elsharkawy

Mann

. Nuclear factor-kappaB and the hepatic inflammation-fibrosis-cancer axis. Hepatology 2007; 46: 590–597.

Seki

Brenner

. Toll-like receptors and adaptor molecules in liver disease: update. Hepatology 2008; 48: 322–335.

Mencin

Kluwe

Schwabe

. Toll-like receptors as targets in chronic liver diseases. Gut 2009; 58: 704–720.

Pimentel-Nunes

Soares

Roncon-Albuquerque

Jr . Toll-like receptors as therapeutic targets in gastrointestinal diseases. Expert Opin Ther Targets 2010; 14: 347–368.

Soares

J-B

Pimentel-Nunes

Roncon-Albuquerque

Leite-Moreira

. The role of lipopolysaccharide/toll-like receptor 4 signaling in chronic liver diseases. Hepatology Int 2010; 4: 659–672.

Akira

Uematsu

Takeuchi

. Pathogen recognition and innate immunity. Cell 2006; 124: 783–801.

Beutler

. TLRs and innate immunity. Blood 2009; 113: 1399–1407.

Zarember

Godowski

. Tissue expression of human Toll-like receptors and differential regulation of Toll-like receptor mRNAs in leukocytes in response to microbes, their products, and cytokines. J Immunol 2002; 168: 554–561.

Nishimura

Naito

. Tissue-specific mRNA expression profiles of human toll-like receptors and related genes. Biol Pharm Bull 2005; 28: 886–892.

10.

Gratz

Mykkanen

El-Nezami

. Probiotics and gut health: a special focus on liver diseases. World J Gastroenterol 2010; 16: 403–410.

11.

Wiest

Garcia-Tsao

. Bacterial translocation (BT) in cirrhosis. Hepatology 2005; 41: 422–433.

12.

Pimentel-Nunes

Roncon-Albuquerque

Jr Goncalves

. Attenuation of toll-like receptor 2-mediated innate immune response in patients with alcoholic chronic liver disease. Liver Int 2010; 30: 1003–1111.

13.

Beutler

. Neo-ligands for innate immune receptors and the etiology of sterile inflammatory disease. Immunol Rev 2007; 220: 113–128.

14.

Akira

Takeda

. Toll-like receptor signalling. Nat Rev Immunol 2004; 4: 499–511.

15.

Fukata

Chen

Klepper

. Cox-2 is regulated by Toll-like receptor-4 (TLR4) signaling: Role in proliferation and apoptosis in the intestine. Gastroenterology 2006; 131: 862–877.

16.

Chang

Lin

Chen

. Helicobacter pylori-induced invasion and angiogenesis of gastric cells is mediated by cyclooxygenase-2 induction through TLR2/TLR9 and promoter regulation. J Immunol 2005; 175: 8242–8252.

17.

Spitzer

Zheng

Kolls

. Ethanol and LPS modulate NF-kappaB activation, inducible NO synthase and COX-2 gene expression in rat liver cells in vivo. Front Biosci 2002; 7: a99–a108.

18.

Lee

Tung

. Cooperation of TLR2 with MyD88, PI3K, and Rac1 in lipoteichoic acid-induced cPLA2/COX-2-dependent airway inflammatory responses. Am J Pathol 2010; 176: 1671–1684.

19.

Guo

Loke

Zheng

. Functional linkage of cirrhosis-predictive single nucleotide polymorphisms of Toll-like receptor 4 to hepatic stellate cell responses. Hepatology 2009; 49: 960–968.

20.

Huang

Shiffman

Friedman

. A 7 gene signature identifies the risk of developing cirrhosis in patients with chronic hepatitis C. Hepatology 2007; 46: 297–306.

21.

Chang

Abar

. Multiple variants in toll-like receptor 4 gene modulate risk of liver fibrosis in Caucasians with chronic hepatitis C infection. J Hepatol 2009; 51: 750–757.

22.

Seki

De Minicis

Osterreicher

. TLR4 enhances TGF-beta signaling and hepatic fibrosis. Nat Med 2007; 13: 1324–1332.

23.

Machida

Tsukamoto

Mkrtchyan

. Toll-like receptor 4 mediates synergism between alcohol and HCV in hepatic oncogenesis involving stem cell marker Nanog. Proc Natl Acad Sci USA 2009; 106: 1548–1553.

24.

Bruix

Sherman

Llovet

. Clinical management of hepatocellular carcinoma. Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver. J Hepatol 2001; 35: 421–430.

25.

Huang

Chou

. Expression of toll-like receptors and type 1 interferon specific protein MxA in biliary atresia. Lab Invest 2007; 87: 66–74.

26.

Manigold

Bocker

Hanck

. Differential expression of toll-like receptors 2 and 4 in patients with liver cirrhosis. Eur J Gastroenterol Hepatol 2003; 15: 275–282.

27.

Starkel

De Saeger

Strain

. NFkappaB, cytokines, TLR 3 and 7 expression in human end-stage HCV and alcoholic liver disease. Eur J Clin Invest 2010; 40: 575–584.

28.

Singh R, Bullard J, Kalra M, et al. Status of bacterial colonization, Toll-like receptor expression and nuclear factor-kappa B activation in normal and diseased human livers. Clin Immunol 2010.

29.

Thuy

Ladurner

Volynets

. Nonalcoholic fatty liver disease in humans is associated with increased plasma endotoxin and plasminogen activator inhibitor 1 concentrations and with fructose intake. J Nutr 2008; 138: 1452–1455.

30.

Pimentel-Nunes

Afonso

Lopes

. Increased expression of toll-like receptors (TLR) 2, 4 and 5 in gastric dysplasia. Pathol Oncol Res 2011; 17: 677–683.

31.

Visvanathan

Skinner

Thompson

. Regulation of Toll-like receptor-2 expression in chronic hepatitis B by the precore protein. Hepatology 2007; 45: 102–110.

32.

Giannitrapani

Ingrao

Soresi

. Cyclooxygenase-2 expression in chronic liver diseases and hepatocellular carcinoma: an immunohistochemical study. Ann N Y Acad Sci 2009; 1155: 293–299.

33.

Martin-Sanz

Mayoral

Casado

Bosca

. COX-2 in liver, from regeneration to hepatocarcinogenesis: what we have learned from animal models? World J Gastroenterol 2010; 16: 1430–1435.