ORIGINAL ARTICLE

Effects of chronic hepatitis C genotype 1 and 4 on serum activinsand follistatin in treatment naı̈ve patients and their correlationswith interleukin-6, tumour necrosis factor-a, viral load and liverdamage

Bassem Refaat • Ahmed Mohammed Ashshi •

Adel Galal El-Shemi • Adnan AlZanbagi

Received: 14 February 2014 / Accepted: 10 May 2014

� Springer-Verlag Italia 2014

Abstract The importance of activins and follistatin in

liver diseases has recently emerged. The aim of the present

study was to measure the influence of chronic infection

with viral hepatitis C (CHC) genotype 1 and 4 on serum

levels of activin-A, activin-B and follistatin, and to deter-

mine their correlations with viral load, liver damage,

interleukin-6 (IL-6) and tumour necrosis factor (TNF)-a.

Sera samples collected from 20 male and 20 female

treatment naı̈ve CHC genotype 1 and 4 Saudi patients (ten

males and ten females for each genotype), and 40 gender-

and age-matched healthy participants were analysed for

activin-A, activin-B and follistatin using enzyme-linked

immunosorbent assay and their levels were correlated with

IL-6, TNF-a, viral load and AST platelet ratio index

(APRI). Serum activin-A, activin-B, IL-6 and TNF-a were

significantly increased, while serum follistatin was signif-

icantly decreased, in both genders of CHC patients com-

pared with control subjects, In both viral genotypes,

activin-A was strongly and positively correlated with the

viral load, APRI, IL-6 and TNF-a, and negatively with

albumin (P \ 0.01). Activin-B showed the same correla-

tions of activin-A only in CHC genotype 1 patients, but it

was weaker than activin-A. No correlation was detected

with follistatin. Serum activins, particularly activin-A, and

follistatin are significantly altered by CHC genotype 1 and

4. This dysregulation of activins/follistatin axis may be

associated with viral replication, host immune response and

liver injury. Further studies are needed to illustrate the

definite role(s) and clinical value of activins and follistatin

in CHC.

Keywords Hepatitis C virus � Activin-A � Activin-B �Follistatin � Interleukin-6 � Tumour necrosis factor-a

Introduction

The World Health Organization has estimated that 170

million persons are infected with hepatitis C virus (HCV)

worldwide with an estimated annual incidence of 3–4

million newly acquired infections [1]. The HCV species

have been classified into six major genotypes and each

genotype contains different subtypes [1]. The distribution

of HCV genotypes varies geographically throughout the

world [2], and genotypes 4 and 1 are the most predominant

HCV species in Saudi Arabia [3, 4].

About 70–80 % of HCV infection cases progress to

chronic stage, due to viral escape from the immune system

[5], and those patients are at significant risk of developing

liver fibrosis, cirrhosis and hepatocellular carcinoma [1].

The progression to chronic inflammation and the devel-

opment of liver fibrosis is regulated by several cytokines

including IL-6, TNF-a and transforming growth factor

(TGF)-b [5, 6].

Activins are members of the TGF-b family and they are

homodimers of two b-subunits (bA and bB). The different

dimerization of subunits gives rise to three proteins: acti-

vin-A (bA–bA), activin-B (bB–bB) and activin-AB (bA–

B. Refaat (&) � A. M. Ashshi � A. G. El-Shemi

Laboratory Medicine Department,

Faculty of Applied Medical Sciences, Umm Al-Qura University,

PO Box 7607, Al Abdeyah, Makkah, KSA

e-mail: Bassem.refaat@yahoo.co.uk

A. G. El-Shemi

Department of Pharmacology, Faculty of Medicine,

Assiut University, Assiut, Egypt

A. AlZanbagi

Gastroenterology Department, King Abdullah Medical City,

Makkah 21955, KSA

123

Clin Exp Med

DOI 10.1007/s10238-014-0297-2

bB) [7]. The coordinated synthesis of follistatin with ac-

tivins is the main regulator of the bioactivity of the dif-

ferent activin isoforms [7]. Activins and follistatin were

initially recognised as gonadal proteins that regulate the

secretion of follicle-stimulating hormone [7]. Later, they

have been shown to be essential for cell survival and they

are involved in the control of many biological processes by

orchestrating and modulating the developmental pro-

gramme, growth and differentiation profile and functional

homoeostasis of most cell types [8].

Activin subunits and follistatin are secreted by the

hepatocyte, and they play an important role in the regula-

tion of liver regeneration and repair [9, 10]. Moreover,

pathological alterations in the serum concentrations of

activin-A and follistatin have been associated with

numerous acute and chronic liver diseases [11]. However,

little is known about the role of the other isoforms of

activin (e.g. activin-B and activin-AB) in liver pathology

despite the reported results from gene knockout studies,

which have demonstrated that each subunit has distinct

functions in vivo [12, 13].

Activin-A increases in liver fibrosis in rat, and it is

mainly localised around the fibrotic areas [14]. Follistatin is

also altered in tissues during fibrosis, and therefore, a role

for endogenous follistatin in controlling the fibrotic effects

of activin has been suggested [15, 16]. The dysregulation in

activin-A/follistatin axis has also been proposed to con-

tribute and promote the growth of pre-neoplastic hepato-

cytes [11, 17]. Hence, activin-A and follistatin, including

serum activin/follistatin ratio, have been suggested as

markers for the diagnosis/prognosis of a variety of liver

pathologies [18, 19].

Currently, the published studies on activins and chronic

hepatitis C (CHC) are few and the results are not consistent

[20–22]. Additionally, none of the previously published

reports examined the effect of viral genotypes on activins

and follistatin. Therefore, the aim of the present study was

to measure the concentrations of activin-A, activin-B and

follistatin in serum samples collected from treatment naı̈ve

male and female patients diagnosed with CHC genotype 1

and 4, and the results were compared with those obtained

from age-matched healthy individuals. We also investi-

gated the correlation of serum concentrations of the can-

didate molecules with the viral load, liver function

parameters, IL-6 and TNF-a.

Subjects and methods

Ethical approval

The study was approved by the Institutional Review Board

and Ethics Committee of King Abdullah Medical City

(IRB 12-028). All samples were collected following

obtaining informed written consent from all the

participants.

Study design

This trial is a case–control study. Serum samples were

collected from 40 treatment naı̈ve Saudi patients diagnosed

with CHC and for whom polymerase chain reaction fol-

lowing reverse transcription was positive for HCV RNA.

The case group included 20 males and 20 females. The

cases positive for genotype 1 (n = 20) and 4 (n = 20)

according to gender were 10 males and 10 females for each

genotype. The inclusion and exclusion criteria are sum-

marised in Table 1. Serum samples from 20 male and 20

female healthy Saudi blood donors with no current acute or

chronic illness were also collected and served as control

group.

Serum concentrations of activin-A, activin-B and fol-

listatin were analysed as described below and their values

were compared between case and control groups, and the

results were also stratified according to viral genotype and

patient gender. The results of liver function parameters and

viral load at the time of sample collection were performed

as part of the routine laboratory workup.

Calculation of AST/platelet ratio index (APRI)

APRI was calculated using the following equation: (AST/

upper limit of normal)/platelet count (9109/L) 9 100. The

Table 1 Inclusion and exclusion criteria

Principal inclusion criteria Principle exclusion criteria

Patient age C18 B45 years Patient age \18 or [45 years

HCV RNA positive Previous non-responders/relapse

No concurrent infection with

HBV or HIV

Solid organ transplant (renal,

heart, or liver, etc.)

Proven fertility Primary or secondary infertility

Not taking exogenous hormones/

oral contraceptive pills for at

least 3 months prior to

enrolment

Taking exogenous hormones/

oral contraceptive pills

Treatment naı̈ve patients Autoimmune condition (e.g.

type 1 DM and rheumatoid

arthritis)

Compensated liver disease (e.g.

no liver cirrhosis, failure or

cancer) & APRI B1.2

History or current thyroid

disease

Acceptable haematological and

biochemical indices

Uncontrolled type 2 DM and

HTN

No or controlled type 2 diabetes

mellitus and hypertension

Concurrent chronic disease

(Renal failure, coronary heart

disease, etc.)

Clin Exp Med

123

interpretation of the APRI results was performed according

to previously published studies [23, 24]. Briefly, index

B0.5 indicated normal liver with no or minimal fibrosis,

[0.5 and B1.5 indicated progressive fibrotic stages (e.g.

Metavir F1–F4) and [1.5 indicated liver cirrhosis. Only

samples with B1.2 APRI were included into the study to

avoid patients with liver cirrhosis.

Enzyme-linked immunosorbent assay (ELISA)

ELISA was used for quantitative measurement of human

activin-A (R&D systems, Minneapolis, USA), activin-B

(Uscn Life Science Inc., Hubei, China), follistatin (R&D

systems, Minneapolis, USA), IL-6 (R&D systems, Min-

neapolis, USA) and TNF-a (R&D systems, Minneapolis,

USA). All samples were processed in duplicate and

according the manufacturer’s instructions.

Activin(s)/follistatin ratio index

Activin-A/follistatin ratio index (AFRI) and activin-B/fol-

listatin ratio index (BFRI) and activins/follistatin ratio

index (AsFRI) were calculated as follow, respectively:

(activin-A/follistatin 9 100), (activin-B/follistatin 9 100)

and [(activin-A ? activin-B)/follistatin 9 100].

Statistical analysis

Statistical analysis of the results was performed using SPSS

version 16. One-way ANOVA followed by LSD post hoc

test was used to compare between the different groups.

Correlations were determined using Pearson’s test.

P value \ 0.05 was considered significant.

Results

Liver function parameters

There was no significant difference in the mean age

between the control and study groups either according to

patient gender or viral genotype (Table 2). By contrast,

significant differences in serum albumin, ALP, AST, ALT

and APRI were detected between the CHC groups and

control group. Nonetheless, there was no significant dif-

ference (P [ 0.05) either between both genders or between

the viral genotype in these parameters within the CHC

group (Table 2).

Serum concentrations of IL-6 and TNF-a

Serum level of both IL-6 (9.5 ± 3.5 Vs. 3.4 ± 1.2;

P = 0.01 9 10-7) and TNF-a (11.2 ± 1.2 Vs. 5.3 ± 0.9; Ta

ble

2D

emo

gra

ph

ican

dla

bo

rato

rych

arac

teri

stic

so

fth

ep

atie

nts

acco

rdin

gto

vir

alg

eno

typ

ean

dg

end

ero

fth

ep

arti

cip

ants

Co

ntr

ol

mal

e(C

M)

(n=

20

)

Co

ntr

ol

fem

ale

(CF

)

(n=

20

)

Mal

eG

1(M

G1

)

(n=

10

)

Mal

eG

4(M

G4

)

(n=

10

)

Fem

ale

G1

(FG

1)

(n=

10

)

Fem

ale

G4

(FG

4)

(n=

10

)

Ag

e(y

ears

)3

9±

5.4

36

±8

.34

0.3

±4

39

.2±

5.7

37

.9±

6.2

34

.8±

8.8

Vir

allo

adat

dia

gn

osi

s

(IU

/mL

)

ND

ND

86

6,7

84

.8±

23

7,0

56

.29

45

,14

8.7

±3

81

,18

1.2

1,0

07

,45

8.5

±3

18

,74

2.4

98

8,4

47

.8±

37

2,3

27

.8

AL

P(I

U/L

)7

9.4

±2

1.6

69

.2±

16

.11

25

±4

5.8

a,b

13

3.3

±3

5.4

a,b

13

0.6

±6

3a,b

12

4±

32

.3a,b

AL

T(I

U/L

)2

8±

11

.22

2.2

±5

.68

9.3

±1

8.1

a,b

81

.6±

19

a,b

71

.9±

22

.7a,b

72

.8±

20

.1a,b

AS

T(I

U/L

)2

1±

4.6

22

.1±

3.1

50

.5±

14

.3a,b

54

.6±

12

.5a,b

57

.8±

13

.4a,b

51

.9±

15

.5a,b

Alb

um

in(g

/dL

)4

.4±

0.2

44

.5±

0.2

73

.7±

0.5

a,b

3.6

±0

4a,b

3.6

±0

.4a,b

3.7

±0

.5a,b

AP

RI

0.3

7±

0.0

70

.36

±0

.06

0.7

8±

0.3

a,b

0.8

4±

0.2

5a,b

0.8

5±

0.2

9a,b

0.8

9±

0.3

2a,b

ND

no

td

on

e;a

P\

0.0

5co

mp

ared

toC

Man

db

P\

0.0

5co

mp

ared

toC

Fg

rou

ps

Clin Exp Med

123

P = 0.04 9 10-14) increased significantly in the case

group compared with control. However, there was no sig-

nificant difference in the serum concentrations of both

cytokines either between the viral genotypes or gender

groups (P [ 0.05).

Serum levels of activins and follistatin

Overall, CHC significantly increased serum concentrations

of activin-A (832.07 ± 252.2 Vs. 294.6 ± 71.4;

P = 0.03 9 10-10), activin-B (227.01 ± 86.6 Vs.

143.2 ± 58.02; P = 0.001), AFRI (189.5 ± 61.8 vs.

34.3 ± 11.5; P = 0.01 9 10-8), BFRI (51.7 ± 27.4 Vs.

17.8 ± 9.3; P = 0.0002), AsFRI (243.2 ± 89.4 Vs.

52.1 ± 21.2; P = 0.01 9 10-10) and significantly

decreased follistatin (500.9 ± 203.8 Vs. 923.02 ± 273.5;

P = 0.03 9 10-5) compared with the control.

By further disseminating the results according to viral

genotype and gender of the patients, both viral genotypes

significantly altered the serum concentrations of candidate

molecules compared with control (P \ 0.05). However,

there was no significant difference between genotype 1 and

4 groups (Fig. 1).

Activin-A and follistatin were significantly higher in

male (331.1 ± 65.5 and 1,111.3 ± 267.8, respectively)

compared with female (256.8 ± 70.1 and 734.7 ± 185.9,

respectively) participants in the control group (P \ 0.05).

However, there was no significant difference (P [ 0.05)

between both genders either between viral genotype 1 and

genotype 4 or within each genotype group (Table 3).

Correlation of serum activins and follistatin with pro-

inflammatory cytokines, liver function parameters

and viral load

Generally, a significant positive correlation between serum

activin-A and activin-B (r = 0.674, P = 0.007 9 10-8)

was detected. Both activin-A (r = -0.481, P = 0.000006)

and activin-B (r = -0.445, P = 0.0003) correlated inver-

sely and significantly with serum follistatin.

When only results from the CHC group (n = 40) were

analysed, serum activin-A was still strongly and positively

correlated with activin-B (r = 0.699, P = 0.00001).

However, both activins did not correlate with serum fol-

listatin (P [ 0.05). Additionally, the strongest positive

correlation between activin-A and activin-B was detected

in patients with genotype 1 (r = 0.911, P = 0.01 9

10-12). Nonetheless, there was no correlation between both

proteins in patients with genotype 4 (r = 0.378, P = 0.1).

Moreover, both proteins did not correlate with follistatin in

genotype 1 and genotype 4 patients (P [ 0.05).

By analysing the results of the 80 study participants

(control and CHC groups), activin-A significantly

correlated positively with IL-6 (r = 0.889, P = 0.03 9

10-20), TNF-a (r = 0.806, P = 0.01 9 10-17), viral load

(r = 0.882, P = 0.01 9 10-15), ALP (r = 0.510,

P = 0.00001), ALT (r = 0.338, P = 0.0003), AST

(r = 0.538, P = 0.00002) and APRI (r = 0.838,

P = 0.03 9 10-12), and negatively with albumin (r =

-0.746, P = 0.02 9 10-10). Activin-B, AFRI, BFRI,

AsFRI and follistatin also significantly correlated with

IL-6, TNF-a, viral load, liver enzymes and APRI but to a

lesser extent than activin-A (Table 4).

Furthermore, results from the CHC group (n = 40)

showed that serum activin-A was significantly (P \ 0.05)

and positively correlated with, IL-6, TNF-a, viral load and

APRI, and negatively with albumin (Fig. 2). However,

there was no correlation between activin-A and liver

enzymes (P [ 0.05). Similar results for activin-A with

viral load, liver function parameters and APRI were also

observed when the patients were disseminated according to

viral genotype (Table 5).

Activin-B was only correlated with IL-6 (r = 0.576,

P = 0.001), viral load (r = 0.417, P = 0.007) and APRI

(r = 0.430, P = 0.006) but not with TNF-a, albumin and

liver enzymes in the CHC group. Similar results were also

observed in patients with genotype 1. However, there was a

positive correlation between serum activin-B and TNF-a(r = 0.627, P = 0.003). In genotype 4 group, there was no

correlation for activin-B with serum cytokines viral load,

liver enzymes, albumin and APRI.

There was no significant correlation (P [ 0.05) for

follistatin with IL-6, TNF-a, viral load, liver enzymes,

albumin and APRI either in the CHC group or by strati-

fying the patients according to viral genotype.

Discussion

Significant dysregulation in serum activin-A/follistatin axis

may occur during liver diseases and therefore could be

utilised as diagnostic/prognostic bio-makers for liver injury

[18]. In the present study, the influence of chronic infection

with hepatitis C virus (CHC) of genotype 1 and 4 on serum

levels of activins and follistatin was measured in Saudi

male and female patients. Data showed a significant

increase in serum activin-A and activin-B, and significant

decrease in follistatin in CHC patients of both genotype 1

and 4, which was independent from the gender of the

patient. More interestingly, the serum levels of activin-A

were strongly and directly correlated with those of IL-6,

TNF-a, viral load, as well as with the liver damage asso-

ciated with CHC infection. These results may suggest that

HCV and/or its associated liver damage modulate the

serum concentrations of activins and follistatin. Additionally,

it appears that pathological alteration in serum activin-A, but

Clin Exp Med

123

not activin-B or follistatin, is common with both HCV

genotype 1 and 4 and is associated with the pro-inflam-

matory cytokine expression pattern and degree of liver

damage.

Over the last decade, there were few reports that had

spotlighted the influence of CHC on serum activin-A. The

first study was reported among Australian patients by

Patella et al. [21] and demonstrated a significant increase in

the serum levels of activin-A with non-significant change

in follistatin in archived samples of 47 CHC patients.

Additionally, there was no significant correlation between

these proteins and liver enzymes. Meanwhile, the alliter-

ating effect of CHC on serum activin-A was further

observed among Egyptian CHC patients by El-Sammak

et al. [20], who did not include follistatin, and the authors

have demonstrated a significant positive correlation

between activin-A and the Child-Pugh scores in 30 patients

[20]. By contrast and more recently, Voumvouraki et al.

Fig. 1 Values of (1) serum activin-A, (2) serum activin-B, (3) serum follistatin, (4) AFRI, (5) BFRI and (6) AsFRI among the studied groups

(Data are represented as Mean ± standard deviation; a = P \ 0.05 compared to control)

Clin Exp Med

123

[22] did not detect significant changes in serum activin-A

in 18 Greece patients with CHC [22].

In comparison with these previous reports, the results

of the present study can support the findings of Patella

et al. [21] and El-Sammak et al. [20] and disagree with

Voumvouraki et al. [22] as there was a significant

increase in serum activin-A in CHC genotype 1 and 4.

Nevertheless, we observed a significant decrease in fol-

listatin and significant correlations for activin-A with the

viral load, which disagree with the findings reported by

the Patella et al. [21].

In addition to the controversial outcomes of these pre-

vious studies, neither the possible role of viral genotype nor

the serum levels of other activin isoforms have been

reported. Genotype 1 and 4 are known to be more

aggressive, harder to treat and are associated with more

severe form of the disease compared with genotype 2 and 3

[1]. Hence, the inconsistency between both studies could

be due to differences in the viral genotypes as the most

prevalent genotypes in Australia are genotype 1 (54 %) and

3 (37 %) [25], while in Saudi Arabia genotype 4 (69 %)

and 1 (25 %) are most common [3, 4]. Furthermore, the

discrepancies in the results reported from Greece with the

other two reports and our study regarding activin-A could

be due to either their small sample size (18 patients) or

differences in viral genotype as the most predominant

genotypes in Greece are 1 and 3 (40 % each) [26, 27].

The insertion of other viral genotype (genotype 2 and 3)

in our study would have revealed the true effect(s) of the

different genotypes on serum activins and follistatin.

However, the addition of other genotypes according to the

inclusion and exclusion criteria of the present study

appeared not achievable as the prevalence of genotype 2

and 3 in Saudi Arabia is about 1 % [3, 4]. Therefore, future

studies should include the different genotypes to measure

the definite effect(s) of CHC on serum activins and

follistatin.

CHC induces several mechanisms of liver regenerating

and is a major cause of HCC [1]. Several reports have

indicated that activin-A is an important negative regulator

of liver growth under physiological conditions and it also

contributes to terminate liver regeneration [11]. Addition-

ally, activin-A is involved in the inflammatory and repair

phases and it induces early scar formation [28, 29]. The

expression of activin bA-subunit increases in liver fibrosis

in rat and it is mainly localised around the fibrotic areas

[14]. Moreover, activin-A increases significantly in liver

cirrhosis and HCC due to different aetiological factors,

including CHC [22]. Alternatively, follistatin induces liver

growth and regeneration in normal rat livers [30]. Disrup-

tion of the activin/follistatin axis was linked to viral hep-

atitis B and C replication but did not correlate with

hepatocyte apoptosis [21], and therefore, it has been sug-

gested that increased production of activin-A could lead to

impaired liver regeneration [31].

APRI has recently been proposed as a predictor of liver

fibrosis and cirrhosis in CHC to replace liver biopsy in a

substantial proportion of patients [32]. Several studies have

confirmed the high sensitivity and specificity and significant

correlation of APRI with both the stage of liver fibrosis and

the grade of activity [23, 24]. Data of the present study are

in agreement with the previous observations as demon-

strated by the presence of significant positive correlations

for serum activin-A and activin-B with APRI, and signifi-

cant negative correlation with serum albumin, suggesting

that the observed increase in activins and decrease in fol-

listatin could be due to liver damage induced by CHC.

Nevertheless, the correlations of activin-B with the liver

function parameters, APRI and viral load were weaker

compared with activin-A, and it was inconsistent in the

different genotypes. Therefore, we suggest that serum

activin-A could be a potential serum marker for the diag-

nosis of liver fibrosis/cirrhosis associated with CHC. Fur-

ther studies are compulsory to support our hypothesis.

Table 3 Mean ± standard deviation of activin-A, activin-B, follistatin, AFRI, BFRI and AsFRI in the different study groups

Control male (CM)

(n = 20)

Control female (CF)

(n = 20)

Male G1 (MG1)

(n = 10)

Male G 4 (MG4)

(n = 10)

Female G1 (FG1)

(n = 10)

Female G4 (FG4)

(n = 10)

Activin-A

(pg/mL)

331.1 ± 65.5 256.8 ± 70.1a 898.9 ± 204a,b 797.6 ± 283.7a,b 874.7 ± 304.4a,b 756.9 ± 221.1a,b

Activin-B

(pg/mL)

117.8 ± 55.1 168.6 ± 62.4 244.6 ± 79.1a,b 259.9 ± 95.3a,b 215.2 ± 40.5a 188.2 ± 32.7a

Follistatin

(pg/mL)

1,111.3 ± 267.8 734.7 ± 185.9a 502.7 ± 107.7a,b 462.3 ± 181.4a,b 540.8 ± 214.4a,b 497.9 ± 193.2a,b

AFRI 33.2 ± 15.7 35.4 ± 8.8 149.9 ± 60.4a,b 172.6 ± 56.7a,b 207.1 ± 86a,b,c 228.2 ± 90a,b,c

BFRI 12.6 ± 4.2 22.9 ± 8.6 39.4 ± 16.5a,b 60.4 ± 23.1a,b,c 44.2 ± 19.3a,b,d 61.2 ± 21a,b,c,e

AsFRI 45.8 ± 20.1 58.4 ± 10.5 228.7 ± 10.6a,b 245.2 ± 9.5a,b 241.5 ± 15.2a,b 257.1 ± 13.9a,b

a P \ 0.05 compared to CM, b P \ 0.05 compared to CF; c P \ 0.05 compared to MG1; d P \ 0.05 compared to MG4 and e P \ 0.05 compared

to FG4 groups

Clin Exp Med

123

IL-6 and TNF-a are well-known pro-inflammatory

cytokines that play a significant role in the host immune

response as well as in the prognosis of chronic infection

with HCV [5, 6]. There is compelling evidence that there is

a remarkable increase in the serum level of both cytokines

in patients with CHC and their levels return to normal

following treatment success [33–36]. Our results are in

agreement with these previous findings as both cytokines

were significantly elevated in the CHC group compared

with control. More importantly, the expression pattern of

both cytokines has also been reported to be regulated by

activin-A in several acute and chronic inflammatory dis-

eases [37–39]. Coherently, the observed positive correla-

tions for activin-A with both IL-6 and TNF-a may suggest

that the increased levels of serum activin-A in CHC could

be a part of the host immune response against HCV.

At the present time, there is no strong evidence indi-

cating the direct involvement of activins and/or follistatin

in the immune response to HCV infection. However,

cumulative data have shown that activin-A and follistatin

are regulatory elements of both innate and humoral

immune responses [40, 41], and both proteins have been

described in immune response to several pathogens

including viruses [42]. Activin-A and follistatin are also

important regulators of regulatory T cells [43], natural

killer cells [44] and dendritic cells [45], which are known

to play an important role in controlling CHC infection [5].

Furthermore, activin-A has been shown to promote the

production of Th2 cytokines, which inhibit Th1 cytokine-

mediated response and promote the development of

humoral response and the subsequent development of liver

fibrosis [5, 46, 47]. Hence, we suggest that activin-A could

have a role in the regulation of the host immune response

against HCV. However, additional studies are needed to

explore the role(s) of activins and follistatin in the immune

response to HCV.

In the current study, we measured the effect of CHC on

serum concentrations of activin-A and activin-B as results

from gene knockout studies have shown that activin bA-

and bB-subunits have distinct functions, and these subunits

do not functionally overlap in all settings in vivo [12, 13].

Our results suggest that activin-A plays a more important

role in the pathophysiology of CHC genotype 1 and 4,

while activin-B is only involved with genotype 1. How-

ever, more studies are compulsory on the role of activin-B

in CHC as the development of specific ELISA kits for

activin-B was not available until recently [48].

Although follistatin is regarded as the activin-binding

protein or an activin antagonist that regulates the majority

of activins’ physiological and pathological actions [7], it

appears that the production of follistatin is not only driven

by activin during inflammation [49, 50]. The synthesis and

secretion of follistatin are also modulated by otherTa

ble

4C

orr

elat

ion

anal

ysi

su

sin

gP

ears

on

’ste

stin

all

stu

dy

par

tici

pan

ts(n

=8

0)

for

acti

vin

-A,

acti

vin

-B,

foll

ista

tin

,A

FR

I,B

FR

Ian

dA

sFR

Iw

ith

IL-6

,T

NF

-a,

vir

allo

ad,

alb

um

in,

liv

er

enzy

mes

and

AP

RI

IL-6

TN

F-a

Vir

allo

adA

lbu

min

AL

PA

LT

AS

TA

PR

I

Act

ivin

-AR

Val

ue

0.8

89

*0

.80

6*

0.8

82

*-

0.7

46

*0

.46

2*

0.3

88

*0

.53

8*

0.8

38

*

PV

alu

e0

.03

91

0-

20

0.0

19

10

-17

0.0

19

10

-15

0.0

29

10

-10

0.0

00

01

0.0

00

30

.00

00

20

.03

91

0-

12

Act

ivin

-BR

Val

ue

0.5

68

*0

.48

5*

0.5

10

*-

0.2

54

*0

.28

6*

0.1

13

0.2

74

*0

.56

5*

PV

alu

e0

.03

91

0-

50

.00

05

0.0

00

01

0.0

20

.01

0.3

0.0

10

.04

91

0-

6

Fo

llis

tati

nR

Val

ue

-0

.42

5*

-0

.60

8-

0.5

55

0.3

62

*-

0.1

22

-0

.20

1-

0.2

04

-0

.36

6*

PV

alu

e0

.00

08

0.0

29

10

-7

0.0

09

91

0-

60

.00

10

.20

.07

0.0

60

.00

1

AF

RI

RV

alu

e0

.69

0*

0.7

62

*0

.76

2*

-0

.62

1*

0.2

75

*0

.23

3*

0.3

32

*0

.54

2*

PV

alu

e0

.01

91

0-

13

0.0

29

10

-15

0.0

29

10

-14

0.0

07

91

0-

80

.01

0.0

30

.00

30

.00

00

02

BF

RI

RV

alu

e0

.46

0*

0.6

03

*0

.55

0*

-0

.31

4*

0.1

99

0.1

11

0.2

18

0.3

46

*

PV

alu

e0

.00

10

.03

91

0-

50

.00

00

01

0.0

05

0.0

70

.30

.05

20

.00

2

AsF

RI

RV

alu

e0

.74

4*

0.7

59

*0

.75

1*

-0

.57

9*

0.2

61

*0

.21

60

.33

3*

0.5

17

*

PV

alu

e0

.02

91

0-

15

0.0

39

10

-12

0.0

19

10

-10

0.0

00

00

10

.01

0.0

54

0.0

03

0.0

00

00

08

Clin Exp Med

123

cytokines, including IL-1b, TNF-a and IFN-c [51, 52]. Our

results agree with these findings as they have shown a

significant decrease in serum concentrations of follistatin

associated with a significant increase in serum activins.

However, there was no significant correlation between the

serum levels of activins and follistatin in the CHC group,

suggesting that each of these proteins have a distinct role

during CHC and that the activities of activins are regulated

by other mechanisms [53], besides follistatin, during

chronic infection with HCV.

In conclusion, serum concentrations of activin-A, acti-

vin-B and follistatin are modulated by CHC. However, it

appears that pathological alterations in serum activin-A are

genotype and gender independent and that activin-A

plays a more important role in the immune response and

pathophysiology of CHC. Additionally, the significant

1

R = 0.836 P = 0.01 X 10-9

2

R = 0.593 P = 0.0005

3

R = 0.745 P = 0.03 X 10-6

4

R = 0.734 P = 0.006 X 10-4

5

R = -0.586 P = 0.0007

6

R = 0.007 P = 0.9

Fig. 2 Correlation of serum activin-A in CHC group (n = 40) with (1) IL-6, (2) TNF-a, (3) viral load, (4) APRI, (5) albumin and (6) ALT by

Pearson’s correlation test

Clin Exp Med

123

correlations of serum activin-A with the viral load and

APRI suggest that it could represent a potential serum

marker for the diagnosis/prognosis of liver damage asso-

ciated with CHC genotype 1 and 4. Further studies are

needed to explore the possible diagnostic/prognostic value

of these proteins in CHC disease.

Acknowledgments The authors thank KACST for the financial

support of the study (12-MED2302-10) under the National Science,

Technology and Innovation Plan. The study was funded by a Grant

(12-MED2302-10) under the National Science, Technology and

Innovation Plan from King Abdul Aziz City for Sciences and Tech-

nology (KACST), Riyadh, KSA.

Conflict of interest None.

References

1. Averhoff FM, Glass N, Holtzman D. Global burden of hepatitis

C: considerations for healthcare providers in the United States.

Clin Infect Dis. 2012;55(Suppl 1):S10–5.

2. Tezcan S, Ulger M, Aslan G, Yaras S, Altintas E, Sezgin O, et al.

Determination of hepatitis C virus genotype distribution in

Mersin province, Turkey. Mikrobiyol Bul. 2013;47:332–8.

3. Al Ashgar HI, Khan MQ, Al-Ahdal M, Al Thawadi S, Helmy AS,

Al Qahtani A, et al. Hepatitis C genotype 4: genotypic diversity,

epidemiological profile, and clinical relevance of subtypes in

Saudi Arabia. Saudi. Saudi J Gastroenterol. 2013;19:28–33.

4. Madani TA. Hepatitis C virus infections reported in Saudi Arabia

over 11 years of surveillance. Ann Saudi Med. 2007;2007(27):

191–4.

5. Fallahi P, Ferri C, Ferrari SM, Corrado A, Sansonno D, Antonelli

A. Cytokines and HCV-related disorders. Clin Dev Immunol.

2012;2012:468107.

6. Ramadori G, Saile B. Inflammation, damage repair, immune

cells, and liver fibrosis: specific or nonspecific, this is the ques-

tion. Gastroenterology. 2004;127:997–1000.

7. Refaat B, Ledger W. The expression of activins, their type II

receptors and follistatin in human Fallopian tube during the

menstrual cycle and in pseudo-pregnancy. Hum Reprod. 2011;26:

3346–54.

8. Hedger MP, de Kretser DM. The activins and their binding protein,

follistatin-diagnostic and therapeutic targets in inflammatory dis-

ease and fibrosis. Cytokine Growth Factor Rev. 2013;24:285–95.

9. Yndestad A, Haukeland JW, Dahl TB, Halvorsen B, Aukrust P.

Activin A in nonalcoholic fatty liver disease. Vitam Horm.

2011;85:323–42.

10. Tashiro S. Mechanism of liver regeneration after liver resection

and portal vein embolization (ligation) is different? J Hepatobil-

iary Pancreat Surg. 2009;16:292–9.

11. Rodgarkia-Dara C, Vejda S, Erlach N, Losert A, Bursch W,

Berger W, et al. The activin axis in liver biology and disease.

Mutat Res. 2006;613:123–37.

12. Brown CW, Houston-Hawkins DE, Woodruff TK, Matzuk MM.

Insertion of Inhbb into the Inhba locus rescues the Inhba-null

phenotype and reveals new activin functions. Nat Genet.

2000;25:453–7.

13. Brown CW, Li L, Houston-Hawkins DE, Matzuk MM. Activins

are critical modulators of growth and survival. Mol Endocrinol.

2003;17:2404–17.

14. Gold EJ, Francis RJ, Zimmermann A, Mellor SL, Cranfield M,

Risbridger GP, et al. Changes in activin and activin receptor

subunit expression in rat liver during the development of CCl4-

induced cirrhosis. Mol Cell Endocrinol. 2003;201:143–53.

15. Abe S, Soejima M, Iwanuma O, Saka H, Matsunaga S, Sakiyama

K, et al. Expression of myostatin and follistatin in Mdx mice, an

animal model for muscular dystrophy. Zool Sci. 2009;26:315–20.

16. Mukhopadhyay A, Chan SY, Lim IJ, Phillips DJ, Phan TT. The

role of the activin system in keloid pathogenesis. Am J Physiol

Cell Physiol. 2007;292:C1331–8.

17. Grusch M, Drucker C, Peter-Vorosmarty B, Erlach N, Lackner A,

Losert A, et al. Deregulation of the activin/follistatin system in

hepatocarcinogenesis. J Hepatol. 2006;45:673–80.

18. Filik L. Activin as a promising test in differential diagnosis of

chronic liver diseases. Eur J Clin Invest. 2012;42:1147 Author

reply 1148.

19. Tomoda T, Nouso K, Miyahara K, Kobayashi S, Kinugasa H,

Toyosawa J, et al. Prognotic impact of serum follistatin in

patients with hepatocellular carcinoma. J Gastroenterol Hepatol.

2013;28:1391–6.

20. Elsammak MY, Amin GM, Khalil GM, Ragab WS, Abaza MM.

Possible contribution of serum activin A and IGF-1 in the

development of hepatocellular carcinoma in Egyptian patients

suffering from combined hepatitis C virus infection and hepatic

schistosomiasis. Clin Biochem. 2006;39:623–9.

21. Patella S, Phillips DJ, de Kretser DM, Evans LW, Groome NP,

Sievert W. Characterization of serum activin-A and follistatin and

their relation to virological and histological determinants in

chronic viral hepatitis. J Hepatol. 2001;34:576–83.

22. Voumvouraki A, Notas G, Koulentaki M, Georgiadou M, Klir-

onomos S, Kouroumalis E. Increased serum activin-A differen-

tiates alcoholic from cirrhosis of other aetiologies. Eur J Clin

Invest. 2012;42:815–22.

23. Fouad SA, Esmat S, Omran D, Rashid L, Kobaisi MH. Non-

invasive assessment of hepatic fibrosis in Egyptian patients with

chronic hepatitis C virus infection. World J Gastroenterol.

2012;18:2988–94.

Table 5 Results of correlation analysis using Pearson’s test for activin-A in all CHC patients (n = 40), genotype 1 patients (n = 20) and

genotype 4 patients (n = 20) with IL-6, TNF-a, viral load, albumin, liver enzymes and ASPRI

IL-6 TNF-a Viral Load Albumin ALP ALT AST APRI

Activin-A (CHC all) R Value 0.836* 0.593* 0.745* -0.586* 0.277 0.007 0.297 0.734*

P Value 0.01 9 10-9 0.0005 0.03 9 10-6 0.0007 0.08 0.9 0.06 0.006 9 10-4

Activin-A (Genotype 1) R Value 0.850* 0.660* 0.869* -0.516* 0.126 -0.105 0.327 0.637*

P Value 0.0002 0.002 0.005 9 10-3 0.02 0.5 0.6 0.1 0.0001

Activin-A (Genotype 4) R Value 0.832* 0.493* 0.718* -0.683* 0.442 0.139 0.337 0.828*

P Value 0.005 9 10-3 0.02 0.00003 0.001 0.06 0.5 0.1 0.000006

Clin Exp Med

123

24. Guzelbulut F, Sezikli M, Akkan-Cetinkaya Z, Yasar B, Ozkara S,

Kurdas-Ovunc AO. AST-platelet ratio index in the prediction of

significant fibrosis and cirrhosis in patients with chronic hepatitis

B. Turk J Gastroenterol. 2012;23:353–8.

25. Bowden DS, Berzsenyi MD. Chronic hepatitis C virus infection:

genotyping and its clinical role. Future Microbiol. 2006;1:

103–12.

26. Stamouli M, Panagiotou I, Kairis D, Michopoulou A, Skliris A,

Totos G. Genotype distribution in chronic hepatitis C patients in

Greece. Clin Lab. 2012;58:173–6.

27. Karatapanis S, Tsoplou P, Papastergiou V, Vasiageorgi A,

Stampori M, Saitis I, et al. Hepatitis C virus genotyping in

Greece: unexpected high prevalence of genotype 5a in a Greek

island. J Med Virol. 2012;84:223–8.

28. Fumagalli M, Musso T, Vermi W, Scutera S, Daniele R, Alotto

D, et al. Imbalance between activin A and follistatin drives

postburn hypertrophic scar formation in human skin. Exp Der-

matol. 2007;16:600–10.

29. McLean CA, Cleland H, Moncrieff NJ, Barton RJ, de Kretser

DM, Phillips DJ. Temporal expression of activin in acute burn

wounds–from inflammatory cells to fibroblasts. Burns. 2008;34:

50–5.

30. Takabe K, Wang L, Leal AM, Macconell LA, Wiater E, Tomiya

T, et al. Adenovirus-mediated overexpression of follistatin

enlarges intact liver of adult rats. Hepatology. 2003;38:1107–15.

31. Hughes RD, Evans LW. Activin A and follistatin in acute liver

failure. Eur J Gastroenterol Hepatol. 2003;15:127–31.

32. Wai CT, Greenson JK, Fontana RJ, Kalbfleisch JD, Marrero JA,

Conjeevaram HS, et al. A simple non-invasive index can predict

both significant fibrosis and cirrhosis in patients with chronic

hepatitis C. Hepatology. 2003;38:518–26.

33. Guzman-Fulgencio M, Jimenez JL, Berenguer J, Fernandez-

Rodriguez A, Lopez JC, Cosin J, et al. Plasma IL-6 and IL-9

predict the failure of interferon-alpha plus ribavirin therapy

in HIV/HCV-coinfected patients. J Antimicrob Chemother.

2012;67:1238–45.

34. Falasca K, Ucciferri C, Dalessandro M, Zingariello P, Mancino P,

Petrarca C, et al. Cytokine patterns correlate with liver damage in

patients with chronic hepatitis B and C. Ann Clin Lab Sci.

2006;36:144–50.

35. Ren Y, Duan ZH, Meng OH, Li Z, Li J. Relationship between the

level of serum TNF-alpha of patients with chronic hepatitis C and

treatment with interferon-alpha and the influencing factors.

Zhonghua Shi Yan He Lin Chuang Bing Du Xue Za Zhi.

2009;23:129–31.

36. Martinez D, Palmer C, Simar D, Cameron BA, Nguyen N, Ag-

garwal V, Lloyd AR, Zekry A. Characterisation of the cytokine

milieu associated with the up-regulation of IL-6 and suppressor of

cytokine 3 in chronic hepatitis C treatment non-responders. Liver

Int. 2014. doi:10.1111/liv.12473.

37. Ofstad AP, Gullestad L, Orvik E, Aakhus S, Endresen K, Ueland

T, et al. Interleukin-6 and activin A are independently associated

with cardiovascular events and mortality in type 2 diabetes: the

prospective Asker and Baerum cardiovascular diabetes (ABCD)

cohort study. Cardiovasc Diabetol. 2013;12:126.

38. Yoshino O, Izumi G, Shi J, Osuga Y, Hirota Y, Hirata T, et al.

Activin-A is induced by interleukin-1beta and tumor necrosis

factor-alpha and enhances the mRNA expression of interleukin-6

and protease-activated receptor-2 and proliferation of stromal

cells from endometrioma. Fertil Steril. 2011;96:118–21.

39. Tsumoto K, Ejima D, Nagase K, Arakawa T. Arginine improves

protein elution in hydrophobic interaction chromatography. The

cases of human interleukin-6 and activin-A. J Chromatogr A.

2007;1154:81–6.

40. Jones KL, Mansell A, Patella S, Scott BJ, Hedger MP, de Kretser

DM, et al. Activin A is a critical component of the inflammatory

response, and its binding protein, follistatin, reduces mortality in

endotoxemia. Proc Natl Acad Sci USA. 2007;104:16239–44.

41. Phillips DJ, de Kretser DM, Hedger MP. Activin and related

proteins in inflammation: not just interested bystanders. Cytokine

Growth Factor Rev. 2009;20:153–64.

42. Nishino Y, Ooishi R, Kurokawa S, Fujino K, Murakami M,

Madarame H, et al. Gene expression of the TGF-beta family in rat

brain infected with Borna disease virus. Microbes Infect.

2009;11:737–43.

43. Semitekolou M, Alissafi T, Aggelakopoulou M, Kourepini E,

Kariyawasam HH, Kay AB, et al. Activin-A induces regulatory T

cells that suppress T helper cell immune responses and protect

from allergic airway disease. J Exp Med. 2009;206:1769–85.

44. Robson NC, Wei H, McAlpine T, Kirkpatrick N, Cebon J,

Maraskovsky E. Activin-A attenuates several human natural

killer cell functions. Blood. 2009;113:3218–25.

45. Scutera S, Riboldi E, Daniele R, Elia AR, Fraone T, Castagnoli

C, et al. Production and function of activin A in human dendritic

cells. Eur Cytokine Netw. 2008;19:60–8.

46. Moser M, Murphy KM. Dendritic cell regulation of TH1–TH2

development. Nat Immunol. 2000;1:199–205.

47. Ogawa K, Funaba M, Chen Y, Tsujimoto M. Activin A functions

as a Th2 cytokine in the promotion of the alternative activation of

macrophages. J Immunol. 2006;177:6787–94.

48. Ludlow H, Phillips DJ, Myers M, McLachlan RI, de Kretser DM,

Allan CA, et al. A new ‘total’ activin B enzyme-linked immu-

nosorbent assay (ELISA): development and validation for human

samples. Clin Endocrinol (Oxf). 2009;71:867–73.

49. Wilson KM, Smith AI, Phillips DJ. Stimulatory effects of lipo-

polysaccharide on endothelial cell activin and follistatin. Mol

Cell Endocrinol. 2006;253:30–5.

50. Blount AL, Vaughan JM, Vale WW, Bilezikjian LM. A Smad-

binding element in intron 1 participates in activin-dependent

regulation of the follistatin gene. J Biol Chem. 2008;283:

7016–26.

51. Abe M, Shintani Y, Eto Y, Harada K, Fujinaka Y, Kosaka M,

et al. Interleukin-1 beta enhances and interferon-gamma sup-

presses activin A actions by reciprocally regulating activin A and

follistatin secretion from bone marrow stromal fibroblasts. Clin

Exp Immunol. 2001;126:64–8.

52. Keelan JA, Zhou RL, Evans LW, Groome NP, Mitchell MD.

Regulation of activin A, inhibin A, and follistatin production in

human amnion and choriodecidual explants by inflammatory

mediators. J Soc Gynecol Investig. 2000;7:291–6.

53. Ethier JF, Findlay JK. Roles of activin and its signal transduction

mechanisms in reproductive tissues. Reproduction. 2001;121:

667–75.

Clin Exp Med

123