Download Associations between Cellular Immune Effector Function, Iron

1452
Associations between Cellular Immune Effector Function, Iron Metabolism,
and Disease Activity in Patients with Chronic Hepatitis C Virus Infection
Günter Weiss,1 Florian Umlauft,1 Martina Urbanek,1
Manfred Herold,1 Mark Lovevsky,3 Felix Offner,2
and Victor R. Gordeuk3
Departments of 1Internal Medicine and 2Pathology, University
Hospital, Innsbruck, Austria; 3Division of Hematology/Oncology,
George Washington University Medical Center, Washington, DC
We studied the associations of macrophage activity, T-helper cell types 1 and 2 (Th-1/Th2) responses, and iron status in 55 patients with hepatitis C virus (HCV)-related liver disease
and 28 control patients with noninfectious liver disease. Serum concentrations of soluble tumor
necrosis factor receptor type II (sTNFrec 75), a macrophage activation marker, were higher
in cirrhotic than in noncirrhotic patients (P ! .001 ) regardless of their HCV status, whereas
levels of neopterin, interleukin (IL)-4 and IL-10 did not differ significantly. In HCV-positive
patients, sTNFrec 75 levels and transferrin saturation (TfS) correlated positively with levels
of aspartate transaminase (P ! .001 for sTNFrec 75 and P = .028 for TfS) and alanine transaminase (P = .003 for sTNFrec 75 and P = .039 for TfS). Increased TfS correlated significantly
with both advanced liver disease and a predominant Th-2 pattern in HCV patients. Our data
suggest that an association exists between macrophage activation and hepatic dysfunction,
and that iron status may affect the clinical course of HCV infection by modulating Th-1/Th2 responses in vivo.
Hepatitis C virus (HCV) infection is responsible for the majority of cases of parenterally acquired non-A, non-B hepatitis.
This virus has a high propensity to cause chronic infection, and
a substantial proportion of patients progress to cirrhosis, liver
failure, and hepatocellular carcinoma [1]. Nowadays, it is well
established that there exist 2 subsets of T-helper (Th) cells in
humans, Th-1 and Th-2. Although Th-1–derived cytokines such
as interferon (IFN)-g and interleukin (IL)-2 are crucial for effective host defense in the acute phase of certain infections,
increased activity of Th-2–derived cytokines such as IL-4, IL10, and IL-13 heightens susceptibility to many infections and
causes exacerbations. The latter effects may be due to a function
of IL-4, IL-10, or IL-13 that inhibits the production of Th-1
cytokines and the activation of macrophages [2–4].
Recent reports have suggested that hepatic iron overload in
HCV-infected patients predicts a poor response to treatment
with interferon-a (IFN-a) [5–8]. This observation is of interest
because in vitro data show that iron availability affects cellmediated immune effector mechanisms involving macrophages
[9, 10]. Increased intracellular concentrations of nonferritinbound iron reduce the effects of the IFN-g signal toward macReceived 6 October 1998; revised 18 June 1999; electronically published
8 October 1999.
Informed consent to draw an additional blood sample was obtained from
all subjects who participated in this study.
Grant support: Austrian Research Funds (FWF 12186 to G.W.).
Reprints or correspondence: Dr. Günter Weiss, Dept. of Internal Medicine, University Hospital, Anichstr. 35, A-6020 Innsbruck, Austria (guenter
[email protected]).
The Journal of Infectious Diseases 1999; 180:1452–8
q 1999 by the Infectious Diseases Society of America. All rights reserved.
0022-1899/1999/18005-0006$02.00
rophages, as seen by reductions in major histocompatibility
complex (MHC) class II expression, neopterin formation, and
tumor necrosis factor (TNF)–a production; conversely, withdrawal of iron upon addition of the iron chelator desferrioxamine enhances these metabolic effects [11]. Therefore, ironloaded macrophages have a reduced cytotoxic potential toward
various intracellular pathogens, including viruses [12–15].
Iron directs the immune response toward a Th-2 response
pattern [16–18], which is unfavorable for fighting a bacterial
or viral infection. Moreover, intracellular iron availability modulates the expression of nitric oxide (NO) by regulating the
transcription of nitric oxide synthase type II (NOS II, inducible
NOS, iNOS) in macrophages [19]. NO formed by macrophages
or hepatocytes in response to stimulation with IL-1, IFN-g,
TNF-a, or lipopolysaccharide (LPS) [20, 21] plays a pivotal
role in the body’s defense against infectious agents, such as
viruses [22]. Therefore, it seems possible that increased availability of iron in HCV patients may reduce the cytotoxic immune effector potential of the host against the viruses [23]. To
test this hypothesis, we investigated the interactions among cellular immune effector function, iron availability, and clinical
course in patients suffering from HCV infection.
Materials and Methods
Patients.
Eighty-three patients were included in the study.
Fifty-five of these patients were HCV positive and 28 were HCV
negative. Blood samples were taken as part of a routine examination. Patients investigated in this study had negative tests for
autoimmune hepatitis, hepatitis B, Wilson’s disease, a 1-antitrypsin
deficiency, and a 1-fetoprotein; there was no ultrasonographic,
computed tomographic, or liver biopsy-based evidence for hepa-
JID 1999;180 (November)
Iron, Cellular Immunity, and HCV Infection
tocellular carcinoma. Phenotypic hereditary hemochromatosis was
excluded by performing genetic testing of patients who had either
transferrin saturation (TfS) 145% or serum ferritin levels 1400 ng/
mL (n = 17). Patients are either homozygous or heterozygous for
1 of the known mutations of the HFE gene (C282Y, H63A), as
checked by the polymerase chain reaction (PCR) method described
elsewhere [24], were excluded from the study (n = 5). Anti-HCV
was detected with a second-generation enzyme-linked immunosorbent assay and confirmed with nested PCR as described elsewhere
[25]. Twenty-two of the HCV-infected patients investigated in this
study had previously received anti-HCV treatment with IFN-a, but
none of them had been receiving therapy for at least 6 months
prior to their participation in this study. When we excluded those
22 patients who had previously received IFN-a, the statistical
trends and results were almost the same as those described later
for all patients (see Results). The diagnosis of cirrhosis was made
on the basis of ultrasonographic and histopathologic examination
after liver biopsy.
As controls, we recruited 28 patients who were examined at our
outpatient clinic because of suspected chronic liver disease. All of
these patients were negative for HCV and the laboratory tests listed
earlier; 26 of these patients had fatty liver disease of various stages,
and 2 had cryptogenic liver fibrosis. Ten of the 26 HCV-negative
patients were given a diagnosis of nonalcoholic steatosis hepatis.
In the control group, 17 (60.7%) of 28 patients regularly drank
alcoholic beverages (112 g daily alcohol intake), whereas in the
HCV group, 30 (54.5%) of 55 had a daily alcohol intake of 112 g.
Laboratory measurements. Serum specimens were drawn during a routine examination and stored at 2707C until cytokine assays were performed. Laboratory parameters, including alanine
transaminase (ALT), aspartate transaminase (AST), hemoglobin,
and iron content were determined by routinely used automated
laboratory tests; ferritin concentration was determined by an immunoassay, and transferrin concentration by a turbidometric
method. The sum of NO22/NO32 was determined after reduction
of nitrate with nitrate reductase with the Griess-Ilosvay’s reagent
as described elsewhere [26]. The level of neopterin was determined
by means of a radioimmunoassay (Brahms, Berlin, Germany) with
a detection limit of 2 nM. Determinations of serum concentrations
of IL-4, IL-6, and IL-10 were all done by commercially available
ELISA kits obtained from Genzyme (Cambridge, MA), Medgenix
(Fleurus, Belgium), and Biosource International (Camarillo, CA),
respectively. The detection limits were 2 ng/L for IL-4, 2 ng/L for
IL-6, and 2 ng/L for IL-10. According to the manufacturers’ information, normal serum values for neopterin are !10 nM: below
the detection limit of the assay for IL-4, !10 ng/L for IL-6, and
below the detection limit for IL-10.
Liver iron determination. Tissue iron concentrations were determined in a subgroup of 22 patients for whom liver biopsy specimens were still available; we used the method of Torrance and
Bothwell [27]. Of these patients, 18 were HCV positive and 4 were
HCV negative.
Statistical analysis. Analyses were done by use of the statistical
software package SYSTAT 7.01 (SPSS, Chicago, IL). Continuous
variables were compared according to HCV status, cirrhosis status,
or transferrin saturation category by Student’s t-test after Bonferroni correction for multiple tests. Proportions were compared by
use of Fisher’s exact test. Because these immunological parameters
1453
followed skewed rather than Gaussian distributions, they were also
evaluated by nonparametric statistical analysis (Kruskal-Wallis)
and a Bonferroni correction was applied when statistical significances were calculated. Logistic regression was used to examine
the relationship of iron status and immunologic markers to cirrhosis after adjustment for age, previous therapy with IFN-a, baseline hemoglobin levels, leukocyte count, and serum creatinine concentrations. Correlations among various measures were assessed
by use of Spearman ranks correlation technique.
Results
Iron and immune status in cirrhotic and noncirrhotic
patients. In the subset of 22 patients tested, liver iron concentration was highly significantly correlated to both TfS
(r = 0.599; P = .005) and serum ferritin levels (figure 1). This
observation validates the use of the indirect measures of TfS
and serum ferritin concentration to reflect iron status in this
study. As shown in table 1, panel A, serum ferritin concentrations and transferrin saturations (TfSs) were significantly higher
in cirrhotic than in noncirrhotic HCV-positive patients. No significant trends were found in markers of iron metabolism between noncirrhotic and cirrhotic HCV negative control patients
(table 1, panel B). In logistic regression models involving HCV
patients, serum ferritin and TfS levels were significantly associated with the finding of cirrhosis. Each 10 ng/mL increase in
serum ferritin concentration was associated with a 1.1-fold increase in the odds of cirrhosis (95% confidence interval [CI],
1.02–1.2; P = .02) after adjustment for age. Each 1% increase
Figure 1. Correlation between liver iron concentration and serum
ferritin levels as shown by dot-blot graph. Regression line and 95%
confidence intervals are shown. Axes are plotted logarithmically.
1454
Weiss et al.
Table 1. Comparison of patients’ baseline parameters according to
presence/absence of cirrhosis.
JID 1999;180 (November)
NOTE. Data are mean 5 SD. Statistical significance of differences between
groups were estimated by nonparametric Kruskal-Wallis test after Bonferroni
correction. ALT, alanine transaminase; AST, aspartate transaminase; HCV, hepatitis C virus; IL-4, IL-6, IL-10, interleukin-4, 6, and 10, respectively; sTNFrec
75, soluble tumor necrosis factor receptor type II; TfS, transferrin saturation.
ver, ALT activity (P ! .001 ) was significantly elevated in HCVpositive, noncirrhotic patients as compared with HCV-negative,
noncirrhotic patients (table 1, panel A and B, first rows). ALT
activity (P ! .01) and AST levels (P ! .05 ) were also significantly
different when comparing cirrhotic HCV-positive with cirrhotic
HCV-negative patients (table 1, panel A and B, last row).
Among immune activation markers only neopterin (P = .05)
was significantly higher, with HCV positivity among the cirrhotic patients (table 1, panel A and B, last row). TfS (P =
.01) was significantly higher with HCV-positivity among the
cirrhotic but not the noncirrhotic patients (table 1). Accordingly, when all patients (cirrhotic and noncirrhotic) were compared, HCV-negative patients still had significantly lower neopterin levels (P = .046) than did HCV-positive patients, which
was not true for sTNFrec 75 (P = .088).
Association between iron status and immune function in HCV
patients. Because the differences in iron status observed in
table 1 appeared to be primarily associated with HCV infection,
we next investigated the impact of TfS on immune and liver
enzymes in HCV-positive patients. We performed a dichotomization of patients according to whether or not their TfS was
130%, the median of normal values in central Europe [28].
Cirrhosis was more frequent in patients with TfS 130% (10
[43.5%] of 23) than in those with a TfS !30% (9 [26.5%] of 34;
P = .07). There was a trend toward higher levels of AST in
patients with TfS 130%. Interestingly, these patients had significantly higher serum levels of the Th-2 cytokine IL-4 than
patients with TfS !30%, whereas NO and neopterin concentrations tended to be lower when TfS was high (table 2).
in TfS was associated with an 1.3-fold increase in the likelihood
of cirrhosis (95% CI, 1.04–1.6; P = .03).
When we examined immune activation markers, we found
sTNFrec 75 concentrations significantly higher in cirrhotic than
in noncirrhotic patients in both the HCV-negative and HCVpositive groups (figure 2). The association between sTNFrec 75
and cirrhosis was confirmed by a logistic regression model: each
1000 ng/mL increase in sTNFrec 75 levels was associated with
a 2.9-fold higher risk for the presence of cirrhosis (95% CI,
1.6–5.2; P ! .001) in HCV patients and with a 2.7-fold higher
risk in non-HCV patients (95% CI, 1.6–4.5; P ! .001). sTNFrec
75 had a stronger association with the finding of cirrhosis than
did serum ferritin or TfS. Among measurements of liver enzyme
we found significantly higher levels of AST and gGT when
comparing cirrhotic with noncirrhotic HCV patients, whereas
in HCV-negative patients, only gGT levels were significantly
different (table 1).
Iron and immune status in HCV-positive and HCV-negative
patients. We next investigated possible differences according
to stratification of patients by HCV infection. In noncirrhotic
patients we found significantly higher levels of neopterin
(P = .01), sTNFrec 75 (P = .02), and IL-10 (P = .05) in HCVpositive as compared with HCV-negative individuals. Moreo-
Figure 2. Soluble tumor necrosis factor receptor type II (sTNFrec
75) levels (mean 5 SD) in patients with/without cirrhosis and/or hepatitis C virus (HCV) infection. See tables 1 and 2 for the significance
values of the differences.
Panel A (HCV-positive patients)
Patients (n)
AST (U/L)
ALT (U/L)
gGT (U/L)
Transferrin (mg/dL)
Ferritin (ng/L)
TfS (%)
Neopterin (nM)
sTNFrec 75 (ng/L)
IL-4 (ng/L)
IL-6 (ng/L)
IL-10 (ng/L)
NO22/NO32 (mM)
HCV quantity
Panel B (HCV-negative patients)
Patients (n)
AST (U/L)
ALT (U/L)
gGT (U/L)
Transferrin (mg/dL)
Ferritin (ng/L)
TfS (%)
Neopterin (nM)
sTNFrec 75 (ng/L)
IL-4 (ng/L)
IL-6 (ng/L)
IL-10 (ng/L)
NO22/NO32 (mM)
Noncirrhotic
Cirrhotic
P
22.9
54.7
77.5
320.4
154.7
25.1
11.3
4013
3.4
13.9
3.8
14.7
11.4
36
5
5
5
5
5
5
5
5
5
5
5
5
5
30.3
28.1
124.4
56.1
145.5
11.0
4.5
1052
5.7
7.6
2.8
12.5
17.6
19
59.4 5 28.6
71.6 5 31.7
122.1 5 137.9
299.4 5 71.0
404.4 5 205.9
39.4 5 0.16.2
13.2 5 5.9
6135 5 1302
5.3 5 7.9
18.4 5 11.9
3.1 5 4.4
13.0 5 8.4
6.5 1 9.7
.002
.09
.05
.1
.02
.007
.3
.0001
.5
.1
.7
.5
.7
25.7
21.2
60.7
293.3
142.5
26.0
8.1
2823
2.7
13.4
2.2
12.6
18
5
5
5
5
5
5
5
5
5
5
5
5
22.2
23.8
61.8
98.6
79.5
15.1
4.3
1646
3.5
10.3
3.9
5.9
33.6
19.3
205.6
294.1
288.1
24.4
9.9
5045
1.6
20.9
3.6
10.7
10
5 27.1
5 6.7
5 301.0
5 59.8
5 297.4
5 13.2
5 4.3
5 1249
5 1.2
5 16.2
5 8.0
5 4.8
.7
.5
.02
1.0
.2
.8
.4
.02
.4
.2
.9
.5
JID 1999;180 (November)
Iron, Cellular Immunity, and HCV Infection
Table 2. Comparison of HCV-positive patients with high and low
transferrin saturations.
No. of patients
AST (U/L)
ALT (U/L)
gGT (U/L)
Transferrin (mg/dL)
Ferritin (ng/mL)
Neopterin (nM)
sTNFrec 75 (ng/L)
IL-4 (ng/L)
IL-6 (ng/L)
IL-10 (ng/L)
NO22/NO32(mM)
HCV quantity
Cirrhosis (no. [% of pts.])
Transferrin low
(!.30)
Saturation high
(1.30)
P
34
40.0 5 28.7
50.3 5 36.2
83.4 5 122
323.8 5 72.4
124.2 5 99.1
13.8 5 4.6
4931 5 1716
2.8 5 5.3
13.3 5 8.0
3.4 5 5.0
15.6 5 10.1
9.9 5 13.5
9 (26.5%)
21
51.2 5 37.7
82.0 5 83.3
79.0 5 121
300.0 5 54.2
379.2 5 183.0
10.4 5 3.4
4489 51252
6.0 5 7.1
18.2 5 13.5
3.5 5 4.4
11.7 5 7.7
9.6 5 17.1
10 (47.6%)
.23
.1
.68
.18
.002
.09
.81
.01
.19
.95
.07
.93
.07
1455
in HCV patients, as well as interactions between iron parameters, liver damage, and immune effector function.
When comparing patients according to their HCV status and
presence or absence of cirrhosis (table 1), we found that serum
neopterin concentrations were significantly higher in HCV-positive compared with HCV-negative patients. Neopterin is produced and released in excess mainly by human macrophages
in response to stimulation by IFN-g [29]. Therefore, neopterin
NOTE. Data are mean 5 SD. Statistical significance of differences between
groups were calculated by nonparametric Kruskal-Wallis test after Bonferroni
correction. ALT, alanine transaminase; AST, aspartate transaminase; HCV, hepatitis C virus; IL-4, IL-6, IL-10, interleukin-4, 6, and 10, respectively; sTNFrec
75, soluble tumor necrosis factor receptor type II; TfS, transferrin saturation.
HCV-positive patients were dichotomized according to whether or not their TfS
was 130%, which is median of normal values in central Europe [28].
Interrelationship among liver damage, iron metabolism, and
immune function. To estimate the associations between the
various parameters in our study, we next calculated Spearman
rank correlations. As is evident from figure 3A, sTNFrec 75
concentrations significantly correlated with serum AST activity
(P ! .001), and they were also positively associated with ALT
levels (P = .003; data not shown). This was true in all patient
groups, but the trends were more pronounced in HCV-positive
individuals. Similarly, we found a significant positive association between serum measures of iron status and liver cell enzyme activity with the closest correlations between AST and
TfS (P = .02, figure 3B), followed by ALT and serum iron
(P = .02), ALT and TfS (P = .03), AST and serum ferritin
(P = .04), and ALT and serum ferritin (P = .09).
On the basis of these data we next investigated a possible
interrelationship between immune and iron markers in HCVpositive patients; our findings are shown in table 3. Interestingly, no significant interactions were found between TfS and
sTNFrec 75. However, TfS was significantly related to IL-4
levels, and a negative correlation was observed between TfS
and NO (table 3). Among immune activation markers the most
striking significance was found between sTNFrec 75 and neopterin (P ! .001 ); sTNFrec 75 also correlated significantly to
macrophage-derived cytokines, such as IL-6 and IL-10, whereas
neopterin was closely related to IL-10 and NO (table 3).
Discussion
In this study we found a highly significant association between sTNFrec 75 levels and cirrhosis both in HCV-positive
and HCV-negative individuals. Moreover, we identified correlations between iron status and more progressive liver disease
Figure 3. A, correlation between soluble tumor necrosis factor receptor type II (sTNFrec 75) and aspartate transaminase (AST), levels
in hepatitis C virus (HCV)–positive patients. B, correlation between
transferrin saturation (TfS) and AST levels in HCV-positive patients.
Regression lines and 95% confidence intervals are shown.
1456
Weiss et al.
Table 3. Spearman’s rank correlation coefficients, identifying associations between iron and immune parameters in patients positive for
hepatitis C virus.
Variables
TfS
Ferritin
Neopterin
sTNFrec 75
IL-4
IL-6
IL-10
Ferritin
a
.62
—
Neopterin
sTNFrec 75
IL-4
IL-6
IL-10
NO
2.07
.04
—
.19
a
.34
a
.55
—
.45
.15
.05
.03
—
a
.07
.11
.10
a
.33
a
.36
—
.14
.07
a
.44
a
.39
.24
.07
—
2.37
.06
a
.32
.21
2.18
.11
.25
a
NOTE. ALT, alanine transaminase; AST, aspartate transaminase; IL-4, IL6, IL-10, interleukin-4, 6, and 10, respectively; NO, nitric oxide; sTNFrec 75,
soluble tumor necrosis factor receptor type II; TfS, transferrin saturation.
a
Significantly correlated, P ! .05.
is a clinically valuable marker for monitoring individuals with
activated cellular immune response [30]. Other investigators
have found that serum neopterin concentrations are useful in
distinguishing patients with noninfectious liver disease from
those with viral hepatitis [31, 32], and, as mentioned earlier,
our findings tend to confirm this observation. Histologic data
are also supportive of this finding: an increased expression of
Th-1 cytokines, such as IFN-g, is observed in liver biopsy specimens from patients with chronic HCV infection, compared
with those from patients with noninfectious liver disease [33].
In contrast to neopterin, sTNFrec 75 is not only produced
in excess by macrophages upon stimulation with IFN-g, but
can also be formed by challenging of macrophages with LPS,
TNF-a, or IL-1, and presumably also by cell detritus products
[34]. Therefore, sTNFrec 75 levels appear to be a reflection of
an ongoing inflammatory response involving macrophages in
the liver due to infectious and noninfectious agents. Because
sTNFrec 75 levels were significantly different between cirrhotic
and noncirrhotic patients, independent of the underlying cause
of disease (figure 2), determination of sTNFrec levels may be
a good surrogate marker to estimate both disease activity and
progression to cirrhosis.
As with sTNFrec 75 levels, serum ferritin concentrations were
increased in cirrhotic as compared with noncirrhotic patients
(table 1). In addition to reflecting high liver iron stores (figure
1), elevated serum ferritin concentrations may also indicate an
ongoing inflammation and activity of liver disease, because ferritin expression is induced in hepatocytes upon stimulation with
proinflammatory cytokines such as IL-1 or TNF-a [35]. The
close correlation between serum ferritin concentration and hepatic iron concentration in a subset of 22 subjects (figure 1)
confirms the validity of serum ferritin concentration as an indicator of iron status in the present study.
The close correlation between liver enzymes (AST and ALT)
and sTNFrec 75 levels points to a possible relationship between
macrophage activation and liver damage. This could either imply that stimulation of macrophages by an infectious agent or
a toxic substance leads to destruction of liver cells via toxic
radical formation by the immune cells, for example, or that
JID 1999;180 (November)
just the opposite may be true. Apoptosis and destruction of
hepatocytes after challenge with microorganisms or toxins
might cause activation of Kupffer cells or invading monocytes
by hepatocyte detritus products [36].
Our finding that TfS levels are positively associated with with
liver enzyme concentrations but not with sTNFrec 75 levels is
consistent with the possibility that excessive iron may contribute to liver damage via its catalytic role for hydroxyl-radical
formation. The close relationship between TfS and hepatic iron
concentration in a subset of our patients supports this thesis.
Nevertheless, increased serum iron levels and TfS in such patients might also be a reflection of iron release by damaged
hepatocytes. In either case, our data are consistent with the
possibility that excess iron then affects immune function. In
macrophages, iron challenge causes an impaired response to
stimulation by the Th-1 cytokine IFN-g [11], thus reducing the
cytotoxic effector potential of such macrophages towards various intracellular bacteria or viruses [12–15]. Moreover, iron
perturbations caused by various infections change the balance
of Th-1/Th-2 cytokines by this basic mechanism. Although
withdrawal of iron upon addition of an iron chelator, such as
desferrioxamine, increases Th-1–mediated immune effector
mechanisms including the formation of NO [16, 17], Th-2 function appears to be weakened, as estimated by determination of
anti-inflammatory cytokines such as IL-4 [16–18]. According
to the Th-1/Th-2 paradigm by which Th-1 and Th-2 cytokines
negatively affect each other’s activity [2], this observation could
be due to macrophages’ increased response to stimulation by
IFN-g (Th-1) activity upon withdrawal of iron, which in turn
leads to decreased Th-2 cell activity and Th-2 cytokine (e.g.,
IL-4) production.
Although a Th-1–mediated immune response appears to be
necessary to gain control over acute viral infections, the damage
to hepatocytes by a virus such as HCV may enhance the destruction of other cells by causing iron to be released, which
subsequent increases the formation of radicals [36]. At the same
time, increased iron concentrations may also weaken the immune potential of macrophages against viruses by directing the
immune response from a Th-1 to a Th-2 pattern [3, 4]. This is
suggested by our data that show both a significant positive
correlation between TfS and IL-4 (Th-2) levels and a negative
trend between TfS and macrophage activation markers such as
NO (Th-1, table 3). This last observation might also be due to
a negative regulatory effect of iron on NOS II transcription,
as has been shown in murine macrophages [19]. These mechanisms could then account for impaired killing of viruses by
macrophages upon iron perturbations, [15] because NO is centrally involved in the body’s defense mechanism toward viral
infections [22] and may also be a key factor for the host response toward HCV infection [37, 38].
The impact of iron homeostasis on the immune response in
HCV infection has also been emphasized by data showing that
the clinical response to IFN-a is reduced in patients with in-
JID 1999;180 (November)
Iron, Cellular Immunity, and HCV Infection
creased hepatic iron stores [4–8]. This could possibly be due to
an inhibitory effect of iron on IFN-a action, in a fashion comparable to that shown for IFN-g [10], because type I and type
II interferons share similar signal transduction mechanisms in
cells [39]. The trend evident in our study, toward a higher prevalence of cirrhosis in patients with high TfS (table 3), could be
a reflection of the mechanism described earlier by which iron
directs the immune response to a Th-2 pattern (IL-4; table 3).
In turn, the Th-2 pathway downregulates the antiviral effector
mechanisms of macrophages and Kupffer cells, while at the
same time causes hepatocyte damage by catalyzing radical formation [36, 40]. Patients with higher TfS might also respond
poorly to IFN-a because of a dominant Th-2 response that
may inhibit some of the pathways induced by type I interferons
toward target cells [2–4].
It is tempting to speculate about whether patients could benefit from iron reduction therapy. Iron chelation could increase
the Th-1–mediated immune response and antimicrobial effector
mechanism of macrophages. Such an effect has been shown in
P. falciparum infection [16, 18], in which iron chelation contributes to more efficient clearance of parasites [41]. Such an
effect has also been suggested for viral infections by a recent
phlebotomy study [42]. Alternatively, iron chelation may reduce
liver damage via withdrawal of iron, thus preventing toxic hydroxyl-radical formation by the catalytic action of the metal
[43].
It will be interesting in future studies to investigate (1) the
impact of iron metabolism on cellular immune function, response to cytokine therapy, and progression of disease in patients with HCV infection in longitudinal investigations; (2) the
definitive mechanism and background for the association between macrophage activation and liver damage; (3) the cutoff
value of sTNFrec 75 levels to estimate the risk of cirrhosis and
the activity of a liver disease; and (4) the value of neopterin to
distinguish between infectious (HCV) and noninfectious liver
disease.
References
1. Alter MJ, Sampliner RE. Hepatitis C. And miles to go before we sleep
(editorial). N Engl J Med 1989; 321:1538-40.
2. Romagnani S. The Th1/Th2 paradigm. Immunol Today 1997; 18:263–6.
3. Bogdan C, Vodovotz Y, Paik J, Xie QW, Nathan C. Mechanism of suppression of nitric oxide synthase by interleukin-4 in primary mouse macrophages. J Leukoc Biol 1994; 55:227–32.
4. Powrie F, Menon S, Coffman RL. Interleukin-4 and interleukin-10 synergize
to inhibit cell-mediated immunity in vivo. Eur J Immunol 1993; 23:2223–9.
5. Akiyoshi F, Sata M, Uchimara Y, Suzuki H, Tanikawa K. Hepatic iron
stainings in chronic hepatitis C patients with low HCV RNA levels: a
predictive marker for IFN therapy. Am J Gastroenterol 1997; 92:1463–6.
6. Barton AL, Banner B, Cable ED, Bonkovsky HL. Distribution of iron in
the liver predicts the response of chronic hepatitis C infection to interferon
therapy. Am J Clin Pathol 1995; 103:419–24.
7. Farigon S, Fracabzani AL, Sampietro M, et al. Liver iron influences the
response to interferon alpha therapy in chronic hepatitis C. Eur J Gastroenterol Hepatol 1997; 9:497–503.
1457
8. Piperino A, Sampietro M, D’Alba R, et al. Iron stores, response to alpha
interferon therapy and effects of iron depletion in chronic hepatitis C.
Liver 1996; 16:248–54.
9. Brock JH. Iron in infection, immunity, inflammation and neoplasia. In: Brock
JH, Halliday JW, Pippard MJ, Powell LW, eds. Iron Metabolism in Health
and Disease. Philadelphia: WB Saunders, 1994:353–90.
10. Weiss G, Wachter H, Fuchs D. Linkage of cellular immunity to iron metabolism. Immunol Today 1995; 16:495–500.
11. Weiss G, Fuchs D, Hausen A, et al. Iron modulates interferon-gamma effects
in human myelomonocytic cell line THP-1. Exp Hematol 1992; 20:605–10.
12. Alford CE, King TE, Campell PA. Role of transferrin, transferrin receptors
and iron in macrophage listericidal activity. J Exp Med 1991; 174:459–67.
13. Barnewell RE, Rikihias Y. Abrogation of gamma interferon-induced inhibition of Ehrlichia chaffeensis infection in human monocytes with transferrin iron. Infect Immun 1994; 82:4804–10.
14. Byrd TF, Horwitz MA. Lactoferrin inhibits or promotes Legionella pneumophila intracellular multiplication in nonactivated and interferongamma activated human monocytes depending upon ist degree of iron
saturation: iron-lactoferrin and nonphysiologic iron chelates reverse
monocyte activation against Legionella pneumophila. J Clin Invest
1991; 88:1103–12.
15. Karupiah G, Harris N. Inhibition of viral replication by nitric oxide and its
reversal by ferrous sulfate and tricarboxylic acid cycle metabolites. J Exp
Med 1995; 181:2171–9.
16. Weiss G, Thuma P, Mabeza R, Werner ER, Herold M, Gordeuk V. Modulatory effects of iron chelation therapy on nitric oxide formation in children with cerebral malaria. J Infect Dis 1997; 175:226–30.
17. Menacci A, Cenci E, Boelaert JR, et al. Iron overload alters T helper cell
responses to Candida albicans in mice. J Infect Dis 1997; 175:1467–76.
18. Thuma PE, Weiss G, Herold M, Gordeuk VR. Serum neopterin, interleukin4 and interleukin 6 concentrations in cerebral malaria patients and the
effect of iron chelation therapy. Am J Trop Med Hyg 1996; 54:164–8.
19. Weiss G, Werner-Felmayer G, Werner ER, Grünewald K, Wachter H, Hentze
MW. Iron regulates nitric oxide synthase activity by controlling nuclear
transcription. J Exp Med 1994; 180:969–76.
20. Xie QW, Cho H, Calaycay J, et al. Cloning and characterization of inducible
nitric oxide synthase from mouse macrophages. Science 1992; 256:225–8.
21. Geller DA, Lowenstein C, Shapiro RA, et al. Molecular cloning and expression of inducible nitric oxide synthase from human hepatocytes. Proc
Natl Acad Sci USA 1993; 90:3491–5.
22. Karupiah G, Xie QW, Buller RM, Nathan C, Duarte C, MacMicking JD.
Inhibition of viral replication by interferon-gamma induced nitric oxide
synthase. Science 1993; 261:1445–8.
23. Weinberg ED, Weinberg GA. The role of iron in infection. Curr Opin Infect
Dis 1995; 8:164–9.
24. Jazwinska EC, Cullen LM, Busfield F, et al. Haemochromatosis and HLAH. Nat Genet 1996; 14:249–51.
25. Umlauft F, Grünewald K, Weiss G, et al. Patterns of Hepatitis C viremia in
hemodialysis patients. Am J Gastroenterol 1997; 92:73–8.
26. Green LC, Wagner DA, Glogowski J, Skipper PL, Wishnok JS, Tannenbaum
SR. Analysis of nitrite, nitrite, and [15N]nitrite in biological fluids. Ann
Biochem 1982; 126:131–8.
27. Torrance JD, Bothwell TH. Tissue iron stores. In: Cook JD, ed. Iron. New
York: Churchill Livingstone 1980: 90–115.
28. Thomas L. Labor und Diagnose. 3rd ed. Marburg, Germany: Medizinische
Verlagsgeschellschaft, 1988:387-400.
29. Huber C, Batchelor JR, Fuchs D, et al. Immune response associated production of neopterin. J Exp Med 1984; 160:310–6.
30. Fuchs D, Hausen A, Reibnegger G, Werner ER, Wachter H. Neopterin as
a marker of activated cell-mediated immunity. application on HIV-infection. Immunol Today 1988; 9:150–5.
31. Prior C, Fuchs D, Hausen A, et al. Potential of urinary neopterin excretion
in differentiating chronic non-A, non-B hepatitis from fatty liver. Lancet
1987; 2:1235–7.
1458
Weiss et al.
32. Reibnegger G, Weiss G, Werner-Felmayer G, Judmaier G, Wachter H. Neural
networks as a tool for utilizing laboratory information: comparison with
linear discriminant analysis and with classification and regression trees.
Proc Natl Acad Sci USA 1991; 88:11426–30.
33. Napoli J, Bishop A, McGuinness PH, Paintner DM, McCAughan GW. Progressive liver injury in chronic hepatitis C infection correlates with increased intrahepatic expression of Th-1 associated cytokines. Hepatology
1996; 24:759–65.
34. Diez-Ruiz A, Tilz GP, Zangerle R, Baier-Bitterlich G, Wachter H, Fuchs D.
Soluble receptors for tumor necrosis factor in clinical laboratory diagnosis.
Eur J Haematol 1995; 54:1–8.
35. Bomford AB, Munro HN. Ferritin gene expression in health and malignancy.
Pathobiology 1992; 60:10–18.
36. Pietrangelo A. Metals, oxidative stress, and hepatic fibrogenesis. Semin Liv
Dis 1996; 16:13–30.
37. Mihm S, Fayyazi A, Ramadori G. Hepatic expression of inducible nitric
oxide synthase transcripts in chronic hepatitis C virus infection: relation
to hepatic viral load and liver injury. Hepatology 1997; 26:451–8.
38. Sharara AI, Perkins DJ, Misukonis MA, Chan SU, Dominitz JA, Weinberg
39.
40.
41.
42.
43.
JID 1999;180 (November)
JB. Interferon (IFN)-alpha activation of human blood mononuclear cells
in vitro and in vivo for nitric oxide synthase (NOS) type 2 mRNA and
protein expression: possible relationship of induced NOS2 to the antihepatitis C effects of IFN-alpha in vivo. J Exp Med 1997; 186:1495–502.
Ivashkiv LB. Cytokine and STATs: how can signals achieve specificity? Immunity 1995; 3:1–8.
Beinker NK, Voigt MD, Arendse M, Smit J, Stander IA, Kirsch RE. Threshold effect of liver iron content on hepatic inflammation and fibrosis in
hepatitis B and C. J Hepatol 1996; 25:633–8.
Gordeuk VR, Thuma PE, Brittenham G, et al. Effect of iron chelation therapy
on recovery from deep coma in children with cerebral malaria. N Engl J
Med 1992; 327:1473–7.
Tsai NC, Zuckerman E, Han SH, Goad K, Redeker AG, Fong TL. Effect
of iron depletion on long term response to interferon-alpha in patients
with chronic hepatitis C who previously did not respond to interferon
therapy. Am J Gastroenterol 1997; 92:1831–4.
Pietrangelo A, Gualdi R, Casalgrandi G, Montosi G, Ventura E. Molecular
and cellular aspects of iron induced hepatic cirrhosis in rodents. J Clin
Invest 1995; 95:1824–31.