Short Communication: Few Liver-Infiltrating Cells Express CXCR3 in HIV/HCV Patients Commencing Antiretroviral Therapy

Abstract

Coinfections with Hepatitis C virus and human immunodeficiency virus accelerate the progression of both conditions and hamper effective treatment. Here we describe expression of CXCR3 on liver-infiltrating cells and peripheral T cells from coinfected patients commencing antiretroviral therapy (ART) in Indonesia. CXCR3 was expressed by small number of intrahepatic inflammatory cells, mostly in the portal areas. The number of cells did not change on ART and was markedly lower than the number of CD4⁺ and CD8⁺ cells in the liver. Data suggest that CXCR3 may contribute to liver infiltration but demonstrate a dynamic situation, changing as the immune system recovers on ART.

Hepatitis C virus (HCV)/human immunodeficiency virus (HIV) coinfection is common in intravenous drug users as the viruses share a route of transmission. Coinfection accelerates hepatic cirrhosis and liver failure,¹ so it is plausible that the mechanisms driving hepatic inflammation and fibrosis may differ from those invoked by HCV alone, and may change in HIV patients beginning antiretroviral therapy (ART).

CXCR3 has been identified as the chemokine receptor most clearly expressed in cirrhotic liver, and mechanisms implicating this receptor in hepatic inflammation and fibrosis have been proposed.² Of the CXCR3 reactive chemokines, circulating CXCL10 levels correlated most clearly with fibrosis.^3,4 Elevated expression of CXCR3 has also been demonstrated on resting peripheral CD8⁺ T cells from HIV/HCV-infected patients stable on ART.⁵ CXCL10 levels in plasma and polymorphisms in the encoding gene also affected fibrosis and a response to therapy in HCV/HIV coinfection.^6,7 However, CXCL10 levels decline on ART,⁸ so the impact of this cytokine on HCV may change.

Understanding the effect of ART on liver histopathology in asymptomatic HCV coinfected patients requires liver biopsies collected before and after ART in well-characterized cohorts. Here we describe further studies of a prospective cohort,⁹ in which we obtained sequential liver biopsies at baseline and after 48 weeks to monitor the infiltration of cells expressing CD4 and CD8.¹⁰ Here, histopathological changes in the liver were correlated with expression of CXCR3 on circulating and intrahepatic cells and levels of CXCL10 in plasma.

We recruited HIV/HCV coinfected patients as they began ART in the HIV/AIDS clinic of Cipto Mangunkusumo Hospital, Jakarta, in 2008–09. Patient demographics have been described.^9,10 Inclusion criteria were HCV seropositivity, age 17–50 years, no previous ART or therapy for HCV disease, and <200 CD4⁺ T cells/μL blood. Exclusion criteria were hepatitis B virus seropositivity, clinically apparent liver cirrhosis, contraindications for liver biopsy, kidney or heart failure, pregnancy, or alcohol excess. Liver biopsies were performed at baseline (48 patients) and week 48 (34 patients). Plasma HIV RNA and HCV RNA levels were measured using a Cobas Amplicore Monitor (Roche Molecular Systems, Pleasanton, CA). Lower limits of detection were 400 copies HIV RNA/ml and 200 copies HCV RNA/ml. The study was approved by the Ethics Committee for Research on Human Subjects, University of Indonesia. Informed consent was obtained for all procedures.

Biopsies used modified Menghini needles, fixed with formalin, and stained to evaluate necroinflammation (score 0–18) and fibrosis (score 0–6) using Ishak's scale.¹⁰ Expression of CXCR3 was assessed using sections stained with antimouse anti-CXCR3 antibody (2Ar1; Abcam, Cambridge, MA), followed by the Starr Trek Universal HRP Detection System (Biocare, Hague, The Netherlands) and diaminobenzidine tetrahydrochloride substrate. Specimens were counterstained with hematoxylin. Seventeen samples (nine baseline and eight from 48 weeks) were excluded as insufficient tissue remained. CXCR3 expression on infiltrating cells was counted over three to five portal areas and the average number of cells per field was determined. Positive controls were human lymphoid tissue and negative controls were liver biopsies stained without primary antibodies. Hepatic expression of CXCL10 was evaluated in representative samples using a rabbit polyclonal antibody (Abcam).

Plasma levels of CXCL10 were measured by Cytometric Bead Array (BD Biosciences, SanJose, CA). Chemokine receptor expression was assessed using peripheral blood mononuclear cells (PBMCs) cryopreserved in liquid nitrogen. Expression of CXCR3 on monocytes and CD4⁺ T cells was assessed using anti-CD14 PE-Cy7 (BioLegend, San Diego, CA), anti-CD4 FITC, and anti-CXCR3 APC. Expression of CXCR3 on CD8⁺ T cells was assessed using anti-CD3 PerCP-Cy5.5, anti-CD8 FITC, and anti-CXCR3 APC (BD Biosciences). A total of 100,000 events per sample were analyzed on an FACS Canto cytometer and analyzed with FlowJo v7.6 (TreeStar, Ashland, OR). Mononuclear cells were gated by forward and side light scatter.¹¹

Data were compared using nonparametric Mann–Whitney tests. Statistical significance was defined as p < .05. Correlations were assessed using Spearman statistics.

The number of intrahepatic CD4⁺and CD8⁺cells and the necroinflammatory and fibrosis scores have been reported.¹⁰ Here we show that CXCR3 was expressed by small number of inflammatory cells before ART [1.4 (0.67–1.7) cells per field] (Table 1), with most positive cells being in the portal areas (Supplementary Fig. S1A, B; Supplementary data are available online at www.liebertpub.com/aid). The two cases with a few positive cells in the lobular area had the largest number of positive cells in the portal area (data not shown). In these cases, some CXCR3⁺ cells resembled endothelial cells. All CXCR3⁺ cells were morphologically distinct from hepatocytes. The number of cells did not change on ART and was markedly lower than the number of cells expressing CD4 or CD8. Before ART, the number of CXCR3⁺ cells in the liver correlated directly with the number of CD4⁺ cells (r = 0.5, p = .01), but not significantly with CD8⁺ cells (r = 0.15, p = .36). There was no correlation between the number of CXCR3⁺ cells and necroinflammatory scores (r = 0.08, p = .62) or fibrosis scores (r = −0.02, p = .87), but all scores were low to moderate (i.e., none had severe HCV disease). There was also no correlation between plasma HCV RNA and CXCR3⁺ cells in the liver before and after ART (r = 0.02, p = .86 and r = 0.16, p = .42). Expression of CXCL10 was sought but yielded only weak diffuse staining that could not be quantified. We also detected CXCR3⁺-infiltrating cells in liver biopsies from HCV monoinfected patients with relatively severe disease, but had insufficient samples for a quantitative study (Supplementary Fig. S1C).

Table 1.

Intrahepatic Inflammation and CXCR3 Expression

	Week 0 (n = 39)	Week 0 (n = 26)	Week 48 (n = 26)
Systemic markers
CD4⁺ T cells/μl	28 (7–65)	35 (8–56)	122 (85–148)
Plasma HIV RNA, log₁₀ copies/ml	5.3 (4.9–5.5)	5.3 (4.9–5.5)	>2.7 in three patients
Plasma HCV RNA, log₁₀ copies/ml	5.6 (3.0–5.9)	5.7 (4.8–6.0)	5.6 (5.6–6.0)
Histopathological markers
Necroinflammatory score (0–18)	5 (4–7)	5 (4–6)	4 (3–6)
Fibrosis score (0–6)	1 (1–1)	1 (1–2)	1 (1–2)
Portal CD4⁺ lymphocytes^a	29 (24–32)	28 (24–31)	55 (45–70)
Portal CD8⁺lymphocytes^a	46 (40–55)	44 (38–53)	33 (30–42)
Portal CXCR3⁺ lymphocytes^a	1.4 (0.67–1.7)	1.3 (0.57–1.7)	1.0 (0.65–2.0)

For each individual, average number of cells per field were determined by examination of three to five fields in the portal region. Data are presented as median (interquartile range) reflecting individual variation.

At week 48, the number of CD4⁺ cells in the liver had risen whereas the number of CD8⁺ cells declined,¹⁰ paralleling changes seen in the blood.¹¹ Correlations between the number of CD4⁺ and CD8⁺ cells and cells expressing CXCR3 in the liver were not significant, with a trend toward inverse relationships (r = −0.33, p = .16 and r = −0.20, p = .42; respectively).

PBMCs cryopreserved at week 36 were used to assess expression of CXCR3 on circulating cells, as this may influence which cells are in the liver at week 48. CXCR3 was expressed on 16 (9–30)% of CD4⁺ T cells, 24 (17–30)% of CD8⁺ T cells, and 0.5 (0.3–1.1)% of CD14⁺ cells (monocytes) in peripheral blood at week 36. The number of CXCR3⁺ cells in the liver at 48 weeks correlated inversely with the proportion of circulating CD4⁺ (r = −0.44, p = .045) and CD8⁺ (r = −0.40, p = .07) T cells expressing CXCR3 at week 36, but did not associate with the small number of monocytes expressing CXCR3 (r = 0.025, p = .91).

Plasma levels of CXCL10 were 1,426 (1,105–2,406) pg/ml at baseline, 464 (246–922) pg/ml at 36 weeks, and 691 (356–1,064) pg/ml at 48 weeks when expressed as median (interquartile range). Hence, levels declined between weeks 0 and 36 (Mann–Whitney, p < .0001), but stabilized by week 48 (p = .34). There was no correlation between levels of CXCL10 at week 0 and any population assessed in the liver at that time (r = −0.05–0.14, p = .47–.79). The number of CXCR3⁺ cells in the liver at week 48 correlated with plasma CXCL10 at week 36 (r = 0.45, p = .03), suggesting a role for CXCL10 in T-cell infiltration into the liver. However, CXCL10 levels at week 48 did not show this correlation (r = 0.02, p = .90).

Overall, in this well-characterized cohort of coinfected patients, intrahepatic CXCR3⁺ cells exist in low numbers in the portal areas without significant changes over the first year on ART. A portal distribution is consistent with recent arrival from peripheral blood, as demonstrated previously in cases of chronic hepatic inflammation.⁴ Here, it was rare to find positive cells in lobular regions. This may reflect the distribution of CXCR3-binding chemokines in the liver as differential expression of CXCL9, CXCL10, and CXCL11 in portal and lobular areas colocalized with CXCR3 expression in advanced HCV monoinfections.⁴

We noted a weak positive correlation between the number of CXCR3⁺ cells and CD4⁺ or CD8⁺ cells in the liver at baseline, which switched to weak negative correlations at week 48. This suggests a change in the role of CXCR3 and is consistent with the marked decline in CXCL10 levels in plasma over time on ART.

CXCR3 expression was demonstrated on circulating CD4⁺ and CD8⁺ T cells at levels similar to those reported previously, with slightly lower expression in HIV/HCV patients than in HCV monoinfected patients.¹² Here there was also a negative correlation between the number of CXCR3⁺ cells in the liver and CXCR3 expression on circulating CD4⁺ T cells at week 48. This could reflect migration of CXCR3⁺ T cells from the blood to the liver or may arise randomly if CXCR3 was irrelevant to liver infiltration. Detailed longitudinal studies are needed to determine how the chemokine is relevant to liver infiltration at specific times on ART.

Footnotes

Acknowledgments

The authors thank all patients who participated in this study and Dr. Henny Saraswati for the cryopreservation of blood leukocytes. This work was supported by the Strategic Research Scheme of the University of Indonesia and AbbVie Pty. Ltd.

Author Disclosure Statement

No competing financial interests exist.

References

Hernandez

, Sherman

. HIV/hepatitis C co-infection natural history and disease progression. Curr Opin HIV AIDS, 2011; 6:478–482.

Larrubia

, Benito-Martinez

, Calvino

, et al. Role of chemokines and their receptors in viral persistence and liver damage during chronic hepatitis C virus infection. World J Gastroenterol, 2008; 14:7149–7159.

Tacke

, Zimmermann

, Berres

, et al. Serum chemokine receptor CXCR3 ligands are associated with progression, organ dysfunction and complications of chronic liver diseases. Liver Int, 2011; 31:840–849.

Zeremski

, Petrovic

, Chiriboga

, et al. Intrahepatic levels of CXCR3-associated chemokines correlate with liver inflammation and fibrosis in chronic Hepatitis C. Hepatology, 2008; 48:1440–1450

Kimball

, McDougan

, Stirling

. CXCR3 Expression elevated on peripheral CD8⁺ lymphocytes from HIV/HCV coinfected individuals. Viral Immunol, 2011; 24:441–448

Pineda-Tenor

, Berenguer

, García-Álvarez

, et al. Single nucleotide polymorphisms of CXCL9-11 chemokines are associated with liver fibrosis in HIV/HCV-coinfected patients. J Acquir Immune Defic Syndr, 2015; 68:386–395.

Pineda-Tenor

, Berenguer

, Jiménez-Sousa

, et al. CXCL9, CXCL10 and CXCL11 polymorphisms are associated with sustained virologic response in HIV/HCV-coinfected patients. J Clin Virol, 2014; 61:423–429.

Chew

, Cherry

, Kamarulzaman

, et al. A longitudinal study of the effects of ART on plasma chemokine levels in Malaysian HIV patients. Dis Markers, 2011; 31:303–309.

Yunihastuti

, Lee

, Gani

, et al. Antibody and markers of T-cell activation illuminate the pathogenesis of HCV immune restoration disease in HIV/HCV co-infected patients commencing ART. Clin Immunol, 2011; 139:32–39.

10.

Gani

, Yunihastuti

, Krisnuhoni

, et al. Periportal CD4⁺ cell infiltration increases in HIV/hepatitis C virus-coinfected patients commencing ART, whereas CD8⁺ cells clear from the liver. J Infect Dis, 2014; 210:405–409.

11.

Tanaskovic

, Fernandez

, Saraswati

, et al. Naive and memory CD4⁺ T-cells are differentially affected in Indonesian HIV patients responding to ART. Viral Immunol, 2016; 29:176–183.

12.

Roe

, Coughlan

, Dean

, et al. Phenotypic characterization of lymphocytes in HCV/HIV co-infected patients. Viral Immunol, 2009; 22:39–48.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB