Reduced Circulating Soluble Receptor for Advanced Glycation End-products in Chronic Hepatitis B Are Associated with Hepatic Necroinflammation

The diagnosis and disease management of chronic hepatitis B (CHB) remain challenging due to the elusive assessment of disease severity. Recently, soluble receptor for advanced glycation end-products (sRAGE) has been implicated in the inflammatory-immune response initiated by liver injury. Nonetheless, its natural behavior and clinical importance in CHB remain elusive. One hundred and twenty CHB patients and forty healthy controls (HCs) were enrolled, and the serum sRAGE as well as RAGE expression in biopsy specimens from these subjects was analyzed, and correlation of sRAGE with clinical features as well as its potential predictive value for monitoring the CHB was also evaluated. Reduced serum sRAGE levels and decreased tissular RAGE expression were observed in CHB patients. sRAGE and RAGE were inversely correlated with gradually increased grades of hepatic necroinflammation as well as the routine indicator ALT. Furthermore, receiver operating characteristic (ROC) analysis showed that combination of ALT and sRAGE exerted better predictive power (area under the ROC curve (AUC) of 0.86) for hepatic necroinflammation than that of ALT (AUC of 0.82), sRAGE (AUC of 0.81), or sRAGE-to-ALT ratio (sRAGE/ALT) (AUC of 0.85) alone. More importantly, circulating sRAGE alone exerted valuable predictive power for hepatic moderate-to-severe necroinflammation in CHB patients but with normal ALT (AUC of 0.81) or minimally elevated ALT (AUC of 0.85). In conclusion, reduced serum sRAGE levels may imply an increased severity for necroinflammation, and it may serve as a potential alternative biomarker for monitoring hepatic necroinflammation in CHB.


INTRODUCTION
Persistent HBV infection leads to adverse outcomes including acute and chronic hepatitis B (CHB), hepatitis B-associated liver cirrhosis (HBLC), and hepatocellular carcinoma (HCC), which remain a public concern that causes considerable liver-related morbidity and mortality especially in China [1,2]. Since active hepatic necroinflammation is the dominant risk factor for driving liver cirrhosis and HCC in CHB patients, a precise evaluation during the initial stage of hepatic inflammation is vital for subsequent treatment and surveillance, representing a high priority and growing medical need.
Presently, wide arrays of tests such as invasive or noninvasive procedures, including liver biopsy or serum biomarkers such as alanine aminotransferase (ALT), have been used to assess disease activity [3]. Since a liver biopsy is an invasive and painful procedure and carries a risk of potential complications, its application has been limited to mass screening [4]. Serum biomarkers such as ALT are extensively used due to their simple operation but have several limitations due to their poor specificity, low sensitivity, or unsatisfactory accuracy, which are easily affected by environmental and lifestyle-related factors [5]. The clinical diagnosis for active CHB as well as the time points for antiviral treatment is predominantly dependent on persistent or intermittent elevated values of ALT [6]. Nevertheless, it is unfeasible for patients to receive therapy when they have a high viral load but with normal or minimal ALT levels. Therefore, there is still a requirement for more reliable, noninvasive, and cost-effective biomarkers for CHB disease activity interpretation.
The receptor for advanced glycation end-products (RAGE), a multiligand pattern recognition receptor, can be stimulated by a variety of damage-associated molecular patterns (DAMPs), activating downstream cascades and triggering an inflammatory-immune response implicated in host defense against infections, inflammation, and cardiometabolic disorders [7][8][9]. RAGE-ligand interactions also contribute to the progression of numerous types of hepatic disorders by inducing oxidative stress and subsequently evoking inflammation, including liver injury, steatosis, fibrosis, and even hepatocarcinogenesis [10]. These reports suggest that blocking the RAGEligand interactions as well as downstream cascades could be a novel therapeutic target for various liver diseases.
Soluble RAGE (sRAGE), a splicing variant of the fulllength receptor that contains only the extracellular domain of RAGE formed by proteolytic cleavage, counteracts RAGE-ligand interactions by sequestering and eliminating RAGE ligands, thereby reducing chronic inflammatory stresses and aggravating tissue injury [7]. A mouse model study showed that treatment with sRAGE could attenuate hepatic injury and inflammation, in parallel with increased expression of pro-regenerative cytokines [11]. Therefore, sRAGE can function as a "decoy" by binding to RAGE ligands and preventing their ligation, which attenuates the inflammatory-immune response. Several investigations have indicated that sRAGE levels are reduced in some disease states and are negatively associated with disease risk and adverse outcomes, including coronary artery disease [12], systemic lupus erythematosus [13], axonal Guillain-Barré syndrome [14], chronic obstructive lung disease [15], and cancer [16,17]. CHB is also a chronic disease with hepatic necroinflammation caused by HBV-induced immune cell death. However, the serum levels of sRAGE in patients with CHB and their underlying relationship with disease progression remain elusive.
In this study, we determined sRAGE levels in CHB and investigated their correlation with clinical parameters in a well-defined cohort of CHB patients, aiming to explore whether sRAGE can be used as a potential disease biomarker during CHB progression.

Patients
A total of one hundred and twenty treatment-naive CHB patients were enrolled in the present study between May 2016 and May 2020 at the Second Affiliated Hospital of Chongqing Medical University. Diagnosis was primarily established by histology, serology, imagiology, and medical history. Patients were not included if they were detected with other disease entities, including infection with hepatitis viruses C, A, D, and E; autoimmune liver disease; and alcoholic liver disease. Clinical characteristics of patients were recorded at the time of CHB diagnosis. CHB patients were consisted of four subgroups (HBeAg( −) inactive HBV carrier, HBeAg( −) immune reactivation phase, HBeAg( +) immune-tolerant phase, and HBeAg( +) immune-active phase) based on the natural course of chronic HBV infection according to the American Association for the Study of Liver Diseases (AASLD) [18]. Hepatic necroinflammation grading was assessed according to the Scheuer staging system [19]. The obtained serum and biopsy specimens are stored at − 80 °C until further processing. In addition, forty serum samples from age-and gender-matched healthy volunteers were enrolled as healthy controls (HCs). Five normal biopsy specimens from HCs who underwent liver biopsy to exclude liver disease were also collected and stored until use. Informed written consent was obtained from all patients and the study was approved by the Institutional Ethics Committee for human studies at the Second Hospital affiliated to Chongqing Medical University, Chongqing, China. All procedures were in accordance with the declaration of Helsinki. Patient characteristics are summarized in Table 1.

Immunohistochemical (IHC)
The expression of RAGE in liver biopsy specimens was examined by IHC. The sections from the formalinfixed, paraffin-embedded tissues were deparaffinized and dehydrated. Then, the sections were boiled for 10 min in a 0.01-M citrate buffer and incubated with 0.3% hydrogen peroxide in methanol for 15 min to block endogenous peroxidase. The sections were then incubated with the anti-RAGE B2513 (Santa Cruz Biotechnology, USA), overnight at 4 °C, following incubation with secondary antibody tagged with the peroxidase enzyme (SP-9000, Zhongshan Golden Bridge, China) for 30 min at room temperature and were visualized with 0.05% 3,3-diamino-benzidine tetrachloride until the desired brown reaction product was obtained. The sections were finally counterstained with hematoxylin. All the slides were observed under an OLYMPUS microscope, and representative photographs were taken.

Hepatic Biochemical and Serological Indexes
Hepatic biochemical indexes such as alanine aminotransferase (ALT) were measured by HITACHI 7600 (HITACHI, Japan). HBsAg, HBeAg, and antibodies against HBsAg (anti-HBs), HBeAg (anti-HBe), and hepatitis B core antigen (anti-HBc) were determined using an Abbott i2000 immunoassay analyzer. HBV DNA was quantified by an ABI 7500 PCR analyzer.

Enzyme-Linked Immunosorbent Assay
Serum samples were analyzed by commercial human sRAGE enzyme-linked immunosorbent assay kits (JYM, China) used according to the manufacturer's instructions. Samples were run in duplicate.

Statistical Analysis
Data were analyzed using SPSS 17.0 and GraphPad Prism 7. The Mann-Whitney or Kruskal-Wallis test was performed to determine significance of sRAGE in CHB patients with various clinical and biochemical features. Dunn's multiple comparison test was used to compare two groups. Correlation coefficients (r) were calculated using Spearman's correlation. Receiver operating characteristic (ROC) curves were generated to classify patients in different groups, as well as for the evaluation of predicting power for serum sRAGE, ALT, and sRAGE/ALT via calculation

Serum Levels of sRAGE in CHB
We detected and analyzed the serum levels of sRAGE in CHB patients to explore whether its levels are abnormally altered in CHB patients. CHB patients exhibited markedly lower serum sRAGE levels than HCs (Fig. 1a). We then detected RAGE expression in tissue sections from biopsy specimens in patients with CHB using IHC analysis. The results showed an attenuated signal of RAGE staining in hepatocyte of patients with CHB (Fig. 1b). According to AASLD guidelines, CHB patients were classified into four subgroups: HBeAg( −) inactive CHB phase (IC), HBeAg( −) immune reactivation phase (IR), HBeAg( +) immune-tolerant phase (IT), and HBeAg( +) immune-active phase (IA). We then assessed sRAGE levels in the four subgroups. In the two CHB with HBeAg( −) subgroups, the HBeAg( −) IR phase showed significantly lower sRAGE levels than the HBeAg( −) IC  (Fig. 1b). In the two HBeAg( +) subgroups, the HBeAg( +) IA phase showed significantly lower sRAGE levels than the HBeAg( +) IA phase (Fig. 1c).

Determinants Linked with sRAGE in CHB Patients
Univariate analysis was performed to identify the determinants of sRAGE. Among the studied variables, phases of CHB, necroinflammation grades (G), necroinflammation parameters (ALT and AST), and the hepatic indicator albumin were significantly linked to sRAGE ( Table 2). To determine the independent determinants of sRAGE, multiple linear regression analysis was performed. We found that necroinflammation G and ALT were independently related to serum levels of sRAGE ( Table 2).

Distribution of Serum sRAGE in CHB Patients with Different Viral Loads
We explored the distribution of serum sRAGE levels in CHB patients with different viral loads, including high HBV DNA levels (≥ 7 log 10 IU/mL), intermediate HBV DNA levels (≥ 5-7 log 10 IU/mL), and low HBV DNA levels (< 5 log 10 IU/mL). No significant difference was found among the three subgroups (Fig. 2a). We further analyzed the correlation between serum sRAGE levels and HBV DNA levels in CHB patients. sRAGE levels were not found to be associated with HBV DNA levels in CHB patients (Fig. 2b). Since high viral loads mainly existed in the three subgroups of the HBeAg( −) IR phase, the HBeAg( +) IT phase, and the HBeAg( +) IA phase, the correlation between serum sRAGE levels and HBV DNA levels was also analyzed in the three subgroups. We found that there was no correlation between serum sRAGE levels and viral loads in all these three subgroups (Fig. 2c-e).

Correlation of Serum sRAGE with Hepatic Necroinflammation in CHB
We observed lower sRAGE levels in patients with moderate-to-severe hepatic necroinflammation (G2-4) than those in patients with no or minimal hepatic necroinflammation (G0-1) (Fig. 3a). Serum sRAGE was inversely correlated with necroinflammation grades in patients with CHB (Fig. 3b). We then analyzed the association between serum sRAGE levels and the hepatic necroinflammation parameter ALT. An inverse correlation between serum sRAGE and ALT was verified in patients with CHB (Fig. 3c). Since elevated levels of ALT mainly existed in subgroups of the HBeAg( −) IR and the HBeAg( +) IA, the correlation between serum sRAGE and ALT levels in these two phases was also analyzed. A negative correlation of sRAGE with ALT levels was verified in these two subgroups (Fig. 3d, e). Moreover, RAGE expression was also analyzed in biopsy specimens from CHB patients with various grades of necroinflammation. We found that RAGE staining was gradually attenuated along with the increased hepatic necroinflammation grades (G0-G4) (Fig. 3f).

DISCUSSION
CHB is a dynamic liver disease characterized by hepatic necroinflammation that is influenced by host and virological factors. Elucidation of the host inflammatoryimmune response involved in CHB pathogenesis provides new perspectives for the identification of biomarkers and therapeutic targets. RAGE functions as a crucial transducer in the inflammatory-immune response by interacting with its ligand and activating downstream cascades. sRAGE, a circulating soluble isoform of RAGE, has recently gained interest due to its potential to ameliorate inflammation by competing with cell surface RAGE for ligand binding under pathological conditions. sRAGE can be detected in human serum and has been implicated in liver diseases, including nonalcoholic fatty liver disease (NAFLD) [20], autoimmune hepatitis [21], and HCC [22]. Nevertheless, its natural behavior and relationship with disease progression as well as its clinical predictive values in CHB development remain elusive. Here, we assessed the serum levels of sRAGE and its association with hepatic necroinflammation parameters during the dynamic course of CHB, aiming Fig. 4 Differentiating power of serum sRAGE for hepatic necroinflammation in CHB patients. a ROC curves of serum sRAGE, ALT, sRAGE/ALT, and combination of sRAGE and ALT (ALT + sRAGE) for predicting necroinflammation in CHB patients. b ROC curve of sRAGE, ALT, sRAGE/ ALT, and ALT + sRAGE for identifying liver moderate-to-severe hepatic necroinflammation (G2-4) from no or minimal hepatic necroinflammation grade (G0-1). c ROC curves of serum sRAGE for discriminating patients with moderate-to-severe necroinflammation (G ≥ 2) but ALT < ULN or ALT ≤ 2 ULN. to determine its potential clinical significance for clinical judgment and surveillance of CHB.
The presence of HBV replication is closely associated with persistent hepatitis activity or intermittent hepatitis flares and subsequent disease progression, including hepatic decompensation, liver cirrhosis, or HCC [23]. Generally, the HBV life cycle is heavily dependent upon and regulated by multiple host functions including intrinsic host innate defensive factors and gene products that provide functions necessary for the virus to complete its life cycle [24]. The present study demonstrated that there was no significant difference among CHB patients with different viral loads, which was confirmed in the three subgroups with high viral loads: HBeAg( −) IR phase, HBeAg( +) IT phase, and HBeAg( +) IA phase. These findings indicate that sRAGE may not be a factor of the host innate defense system that participates in HBV replication, consistent with a previous report showing that serum sRAGE levels were linked to the chronic inflammation in lung disease rather than pathogen infection [25].
The present study demonstrated that CHB patients in a well-defined cohort exhibited reduced sRAGE levels, implying that sRAGE may exert opposite functions in regulating CHB pathogenesis. This finding supports those of previous studies showing low sRAGE levels in other inflammatory-immune disorders [26][27][28]. It has been reported that sRAGE is mainly generated from the proteolytic cleavage of membrane RAGE by metalloproteases such as MMP9 and ADAM1 [8,29], while MMP9 levels were elevated in patients with varying degrees of CHB [30], suggesting that reduced serum sRAGE levels in CHB may not be mainly influenced by RAGE cleavage. The assessment of sRAGE levels in the CHB subgroups revealed that the HBeAg( −) IR and HBeAg( +) IA phases exhibited lower sRAGE levels than the HBeAg( −) IC and HBeAg( +) IA phases, respectively, suggesting an inverse correlation between sRAGE and hepatic necroinflammation. Since active hepatic necroinflammation in CHB patients represents one of the important risk factors for developing liver fibrosis and even HCC, an evaluation of necroinflammation severity as well as its early control is imperative to determine the need for surveillance and treatment in patients with CHB [31]. Here, patients with moderate-to-severe necroinflammation (G ≥ () had lower sRAGE levels than those with no or minimal hepatic necroinflammation (G0-1). Serum sRAGE levels were inversely correlated with ALT levels, a classic indicator of hepatic necroinflammation and prior CHB treatment. Similar results were also observed in the HBeAg( −) IR and HBeAg( +) IA phases, which exhibited hepatic necroinflammation. These observations suggest that reduced sRAGE levels may be identical to elevated ALT, reflecting inflammatory activity. Furthermore, RAGE expression in biopsy specimens from CHB patients was reduced along with hepatic necroinflammation grades, while RAGE binds to its ligands, leading to the activation of downstream cascades, and maintains persistent inflammation [9]. With the inconsistence of reduced RAGE but its supporting role in inflammation, we hypothesize that persistent inflammation elicited by RAGE-ligand systems may be mainly dependent on RAGE's ligands, but this hypothesis still needs to be clarified in future research.
Many investigators have been trying to develop noninvasive tests for predicting liver inflammation. To date, ALT remains widely used as a routine indicator for hepatic necroinflammation or therapeutic judgment for antiviral therapy [32]. Nonetheless, growing evidence has revealed that ALT alone is not as good as an indicator of the grade of hepatic necroinflammation. One statistical study in China revealed that nearly half of CHB patients with ALT < 2 ULN showed moderate-to-severe necroinflammation (G ≥ 2) [33]. Another biopsy report in the USA similarly showed that one-third of CHB patients with normal ALT levels showed severe inflammation [34]. Therefore, CHB patients may exhibit normal ALT levels, leading to an inaccurate assessment of disease severity or even misdiagnosis. The present data revealed that combination of sRAGE and ALT value was superior to ALT, sRAGE, or sRAGE/ALT alone for predicting hepatic necroinflammation. Therefore, sRAGE combined with ALT is necessary and may be used for patients who lack clear-cut indications for treatment. We also found that sRAGE has predictive power for CHB patients (G s pr with normal and minimally elevated ALT levels. This finding is meaningful that a CHB patient with a low serum sRAGE level may have significant hepatic necroinflammation even with a normal or minimally elevated ALT. However, further work on larger samples should be performed to substantiate the findings.
Nonetheless, several limitations to the present study exist. First, we did not assess the specificity of serum sRAGE in CHB compared to those with other liver disease entities such as chronic hepatitis C, nonalcoholic fatty liver disease, and alcoholic liver disease using the same set of analyses. Second, we investigated the sRAGE levels during the chronic phase of HBV infection, and it is insufficient to elucidate dynamic changes of sRAGE during the different phases of HBV infection from acute hepatitis, CHB, hepatic fibrosis, and HBLC to HCC. Last, the detailed molecular mechanism regarding the underlying role of sRAGE in RAGE signaling activation in CHB pathogenesis and whether sRAGE can serve as a therapeutic target still need further clarification.

CONCLUSION
In conclusion, the present study demonstrated that CHB patients exhibited reduced levels of sRAGE, and its levels were inversely correlated with hepatic necroinflammation. Furthermore, combination of ALT and RAGE exerted valuable predictive power for hepatic necroinflammation in CHB patients. More importantly, circulating sRAGE alone exerted valuable predictive power for hepatic moderate-to-severe necroinflammation in CHB patients but with normal or minimally elevated ALT levels. These data suggest that monitoring sRAGE and/or its combination with ALT may facilitate decision-making regarding prophylaxis and treatment, ultimately leading to the prevention of CHB progression.

AUTHOR CONTRIBUTION
All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Xiuyu Zhang, Yan You, Qiao Liu, Xiaoyu Sun, Weixian Chen, and Liang Duan. The first draft of the manuscript was written by Liang Duan and all authors commented on previous versions of the manuscript. All authors reviewed and approved the final manuscript.

FUNDING
This work was supported by the National Natural Science Foundation of China (82072364), Natural Science Foundation of Chongqing (cstc2019jcyj-msxmX0864), and Kuanren Talents Program of the Second Affiliated Hospital of Chongqing Medical University.

DATA AVAILABILITY
The authors confirm that the data supporting the findings of this study are available within the article and material.

DECLARATIONS
Ethics Approval All procedures were in accordance with the Declaration of Helsinki. Informed written con-sent was obtained from all patients and the study was approved by the Institutional Ethics Committee for human studies at the Second hospital affiliated to Chongqing Medical University, Chongqing, China.