Thiol/disulphide homeostasis in pregnant women with Familial Mediterranean fever

Abstract

Objective: To investigate the presence of oxidative stress (OS) in pregnant women with Familial Mediterranean fever (FMF) in the first trimester by evaluating thiol/disulphide homeostasis.

Study design: A total of 31 pregnant women with a diagnosis of FMF, between 11⁰ and 13⁶ weeks of gestation, were compared with 51 healthy pregnant controls at the same gestational weeks. A recently defined method was used to measure plasma native thiol, total thiol and disulphide levels.

Results: There were no differences between groups in terms of maternal age, body mass index and numbers of gravida and parity. Antenatal complications (45.2% vs. 9.8%, P = 0.001) and primary caesarean section (22.6% vs. 5.9%, P = 0.037) were higher in the FMF group. Pregnant women with FMF had significantly lower first trimester serum levels of native thiol (297.5 μmol/l (153.2–441.8) vs. 366.1 μmol/l (288.7–432.4), P = 0.000), total thiol (327.2 μmol/l (171.0–471.0) vs. 389.9 μmol/l (317.1–449.8), P = 0.000) and higher levels of disulphide (14.2 ± 4.5 μmol/l vs. 12.4 ± 3.4 μmol/l, P = 0.045). No differences were found in these parameters among FMF patients with and without antenatal complications.

Conclusions: The main outcome demonstrates a relation between OS and pregnant women with FMF in the first trimester of gestation. OS in the first trimester may be a major aetiological factor of unfavourable pregancy outcomes in this group of patients.

Keywords:

• Disulphide
• FMF
• Oxidative stress
• Pregnancy
• Thiol

Introduction

Familial Mediterranean fever (FMF) is an autoimmune disease characterized by acute recurrent fever attacks, inflammatory peritonitis, pleuritis and arthritis.¹ It is regarded as a systemic disease clinically affecting the kidneys, heart, liver, thyroid and adrenal glands.² Some unfavourable effects of FMF on gestational outcomes were formerly reported.³

Human cells are embellished with enzymatic and nonenzymatic systems for maintaining redox balance, which is mandatory for growth and survival.⁴ The term oxidative stress (OS) indicates an imbalance between the production and removal of reactive oxygen species that results in the accumulation of oxidative damage. OS is reported to be involved in many diseases and syndromes.^5–7 Nonetheless, until recently in trials aimed at evaluating OS, usually quite complicated methods like high-sensitivity fluorometric high-performance liquid chromatography,⁸ capillary electrophoresis with laser-induced fluorescence detection⁹ and enzymatic recycling,¹⁰ and substances such as 5,5′-dithiobis-(2-nitrobenzoic acid) (DTNB)¹¹ and dithiodipyridine (4-DPS)¹² were used.

Thiols are organic molecules comprising a sulfhydryl group that play important roles in redox homeostasis through oxidation and reduction reactions. In the presence of oxygen, thiol groups are reversibly oxidized to form disulphide linkages. In other words, interconversions between thiols (the reduced state) and disulphide groups (the oxidized state) maintain oxidative balance.¹³ However, for the analysis of OS, only the thiol side of the balance could be measured, and quantification of the disulphide side could not be performed thus far. In 2014, Erel and Neselioglu defined a novel, simple and practical technique that enables the detection of both sides of the the balance in a practical way.¹⁴ We performed this study in order to investigate OS in pregnant women with FMF by evaluating thiol/disulphide homeostasis via the technique recently defined.

Materials and methods

The protocol for this prospective case control study was approved by the institutional review board for human studies of Zekai Tahir Burak Women's Health Care, Training and Research Hospital and informed consent was obtained from all participants.

Between March 2014 and March 2015, a total of 31 pregnant women who were referred to our perinatology department with a diagnosis of FMF according to the Tel-Hashomer criteria¹⁵ were compared with 51 healthy pregnant women. The exclusion criteria for the study and control groups were the presence of multiple pregnancies, any systemic diseases other than FMF, any gestational complications such as gestational diabetes, hypertensive disorder, preterm labour, etc. in the previous or the current pregnancy, alcohol or smoking addiction and any chronic drug use other than for FMF (e.g. colchicine).

All venous blood samples of 10 cc were collected in tubes containing ethylenediaminetetraacetic acid (EDTA) just after ultrasonographic measurement of crown-rump length and nuchal translucency in the first trimester (between 11⁰ and 13⁶ weeks). Plasma samples were separated from cells by centrifugation at 1500 × g for 10 minutes. The samples were run immediately or stored at −80°C.

A new and fully automated method developed by Erel and Neselioglu was used for the measurement of plasma native thiol, total thiol and disulphide levels.¹⁴ The method was based on the reduction of dynamic disulphide bonds to functional thiol groups by sodium borohydride (NaBH4). Formaldehyde was used to remove all of the unused NaBH4, in order to prevent extra reduction of 5,5′-dithiobis-2-nitrobenzoic acid (DTNB) and further reduction of disulphide bonds formed after the DTNB reaction. Total thiol content of the sample was measured using modified Ellman's reagent. Native thiol content was subtracted from total thiol content and half of the obtained difference gave the disulphide bond amount. In addition, disulphide/thiol, disulphide/total thiol and thiol/total thiol ratios were calculated automatically at the same time.

Measured values were examined and reported as mean ± standard deviation, median (min–max) and number, % using the SPSS 21.0 statistical program. The normality of the distribution of groups was determined with the Kolmogorov–Smirnov test. Parametric data were analysed by independent two-sample t test and non-parametric data were compared using Mann Whitney-U test. Chi-square test was used for the comparison of percentages. A P value < 0.05 was considered as statistically significant.

Results

There were no significant differences between groups in terms of maternal age, body mass index and numbers of gravida and parity (Table 1). The rates of antenatal complications (45.2% vs. 9.8%, P = 0.001) and primary caesarean section (22.6% vs. 5.9%, P = 0.037) were higher in the study group (Table 1). Characteristics of pregnant women with FMF are shown in Table 2. Pregnant women with FMF had significantly lower first trimester serum levels of native thiol (297.5 μmol/l (153.2–441.8) vs. 366.1 μmol/l (288.7–432.4), P = 0.000) (Fig. 1), total thiol (327.2 μmol/l (171.0–471.0) vs. 389.9 μmol/l (317.1–449.8), P = 0.000) (Fig. 2) and higher levels of disulphide (14.2 ± 4.5 μmol/l vs. 12.4 ± 3.4 μmol/l, P = 0.045). In addition, disulphide/native thiol and disulphide/total thiol ratios were higher and native thiol/total thiol were lower in pregnant women suffering from FMF. No differences were measured in serum albumin, total protein, creatinine and blood urea levels (Table 3). Comparison of native thiol, total thiol and disulphide levels, ratios and biochemical results among FMF patients with and without antenatal complications did not reveal any significant difference (Table 4).

Figure 1 Native thiol levels of groups. FMF, Familial Mediterranean fever

Figure 2 Total thiol levels of groups. FMF, Familial Mediterranean fever

Table 1 Comparison of patient characteristics, antenatal complications and pregnancy outcomes

	FMF (n = 31)	Control (n = 51)	P
Maternal age (years)	27 (19–39)	26 (17–42)	0.874^a
Gravida (number)	2 (1–4)	2 (1–6)	0.538^a
Parity (number)	1 (0–3)	1 (0–3)	0.727^a
Body mass index	27.1 ± 2.7	25.7 ± 3.9	0.072^b
Antenatal complications (n, %)	14 (45.2%)	5 (9.8%)	0.001^c
Abortus imminens	1 (3.2%)	1 (2%)
Threatened preterm labour	7 (22.6%)	1 (2%)
<34 weeks labour	1 (3.2%)	1 (2%)
Pyelonephritis	1 (3.2%)	0 (0%)
Intrahepatic cholestasis	3 (10%)	0 (0%)
Hyperemesis gravidarum	1 (3.2%)	2 (3.9%)
Birth week	39 (33–41)	38 (30–41)	0.253^a
Birth weight (g)	3150 ± 319	3250 ± 507	0.276^b
NICU admission (n, %)	2 (6.5%)	2 (3.9%)	0.631^c
Primary C/S (n, %)	7 (22.6%)	3 (5.9%)	0.037^c

^aMann–Whitney U test.

^bT test.

^cChi-square test.

Data are expressed as median (min–max), mean ± standard deviation and number, (%) where appropriate. P < 0.05 indicates significant difference. FMF, Familial Mediterranean fever; NICU, Neonatal intensive care unit; C/S, Caesarean section.

Table 2 Characteristics of pregnant women with Familial Mediterranean fever (n = 31).

Mean age at first attack (years)	14.7 (minimum:7, maximum:40)
Peritonitis	93.5%
Arthritis	58%
Pleuritis	3.2%
Presence of attacks during pregnancy	64.5%
Mean number of attacks during pregnancy	2 (minimum:1, maximum:5)
Mean interval of days between attacks	23.5 (mininum:6, maximum:60)
Pain score during attacks	7 (minimum:4, maximum:10)
Hospitalization due to attacks during pregnancy	6.5%
Previous abdominal surgery	16.1%

Table 3 Comparison of disulphide levels, ratios and biochemical results

	FMF (n = 31)	Control (n = 51)	P
Disulphide (μmol/l)	14.2 ± 4.5	12.4 ± 3.4	0.045^b
Disulphide/nat. thiol × 100	4.5 (2.2–8.4)	3.3 (1.4–5.6)	0.002^a
Disulphide/total thiol × 100	4.1 (2.1–7.5)	3.1 (1.4–5.0)	0.002^a
Native thiol/total thiol × 100	91.8 (85.0–95.8)	93.8 (90.0–97.3)	0.002^a
Albumin (g/dl)	4.10 (2.90–4.61)	4.04 (3.66–4.66)	0.494^a
Total protein (g/dl)	6.88 ± 0.60	7.05 ± 0.34	0.144^b
Creatinine (mg/dl)	0.60 ± 0.10	0.64 ± 0.09	0.079^b
Blood urea nitrogen (mg/dl)	16.6 ± 5.4	17.9 ± 5.1	0.349^b

^aMann–Whitney U test.

^bT test.

Data are expressed as median (min–max) and mean ± standard deviation where appropriate. P < 0.05 indicates a significant difference. FMF, Familial Mediterranean fever.

Table 4 Comparison of native thiol, total thiol and disulphide levels, ratios and biochemical results among FMF patients with and without antenatal complications

	FMF patients with antenatal complications (n = 14)	FMF patients without antenatal complications (n = 17)	P
Native thiol (μmol/l)	294.1 (153.2–441.8)	328.1 (194.8–394.8)	0.922^a
Total thiol (μmol/l)	325.7 (171.0–471.0)	357.4 (228.8–418.8)	0.891^a
Disulphide(μmol/l)	15.3 ± 4.8	12.7 ± 3.8	0.131^b
Disulphide/nat. thiol × 100	5.2 (2.7–8.4)	4.5 (2.2–8.2)	0.262^a
Disulphide/total thiol × 100	4.7 (2.6–7.5)	4.1 (2.1–7.4)	0.262^a
Native thiol/total thiol × 100	91.8 (85.1–95.8)	90.7 (85.0–95.0)	0.262^a
Albumin (g/dl)	4.10 (2.99–4.61)	4.02 (2.90–4.50)	0.544^a
Total protein (g/dl)	6.92 ± 0.60	6.84 ± 0.62	0.781^b
Creatinine (mg/dl)	0.60 ± 0.09	0.60 ± 0.12	0.535^b
Blood urea nitrogen (mg/dl)	16.0 ± 5.8	17.7 ± 4.7	0.631^b

^aMann–Whitney U test.

^bT test.

Data are expressed as median (min–max) and mean ± standard deviation where appropriate. P < 0.05 indicates significant difference. FMF, Familial Mediterranean fever.

Discussion and conclusions

FMF is assumed to be an ethnic specific disease with an estimated prevalence of 1/250–1000 in non-Ashkenazi Jews, 1/73 000 in Ashkenazi Jews, 1/500 in Armenians and 1/3000 in Turkish population.^16–18 Furthermore, new cases in unexpected communities (e.g Japanese, Korean) have been recently reported with improvements in diagnostic genetics.^19–21 Up to now, very limited data about the interactions between FMF and pregnancy have been defined. Thus, new studies of pregnant women with FMF may put forward novel findings with an expanded population and their perinatal characteristics.

OS ensues from a disproportion among the systemic manifestations of reactive oxygen species and the detoxification processes. OS has been evaluated as an aetiological factor in many disorders, including FMF. Karaguezyan et al.²² showed patterns of differences in membrane phospholipid composition that were consistent with increased OS in patients with FMF. Sarkisian et al.²³ demonstrated OS in FMF by evaluating clastogenic plasma factors in neutrophils. In addition, Karakurt Ariturk et al.²⁴ showed lower paraoxonase-1 activity, pointing to OS in FMF patients. And recently in a study by Savran et al.,²⁵ FMF seemed to be associated with increased concentrations of macrophage migration inhibitory factor that were significantly correlated with OS. In our view, the results of these studies provide convincing evidence to form a conclusion about the relationship between FMF and OS. However, a careful look at the methods of OS analysis performed in these studies indicates that these procedures cannot easily be integrated into daily clinical practice.

In 2014, Erel and Neselioglu disclosed a new method to establish the dynamic thiol/disulphide homeostasis.¹⁴ According to the authors, this technique was less expensive, less time-consuming and less complex than other methods. In our opinion, it is a valuable alternative, particularly with the escalating interest in OS and thiol/disulphide balance, and for studies aiming to understand unknown aspects of the aetiology of disease.

In our study, there was an altered thiol/disulphide balance in pregnant women with FMF, which was previously reported to be an indicator of OS.²⁶ The ratios (disulphide to native thiol × 100 and disulphide to total thiol × 100) that were used to show the overall status of the balance more precisely indicated there was OS in the study group. However, further studies should be conducted in order to clarify the issue of whether OS in pregnant women with FMF is a primary aetiological factor of the disease or secondary to the FMF itself. Previous data, confirming a drop in the ratio of thiol/disulphide as an indicator of cellular apoptosis,²⁷ which is a similar finding to our study, may also suggest that apoptosis is a cause of the unfavourable antenatal complications in pregnant women with FMF. In our opinion, the insignificant differences in OS markers among our FMF patients with and without antenatal complications may be due to the small numbers of participants. Besides supplementation of a thiol donor, such as N-acetylcysteine, may have positive effects on gestational outcomes by boosting the levels of native thiols and protein adjustment, the benefit of which has been proven previously.^28,29

In this study, the levels of disulphide were significantly higher in pregnant women with FMF compared to controls. Similar (higher) levels versus controls were reported in patients with degenerative disorders such as diabetes, pneumonia, obesity and smoking.¹⁴ Based on this observation, future trials exploring the similarities between FMF and degenerative disorders may reveal more conclusive data about the aetiology of FMF.

Nevertheless, there are some limitations of our study. The drawbacks can be addressed by performing new studies that evaluate other OS markers in addition to thiols and disulphides, that include a relatively larger number of participants, and that have a prospective cohort design targeting more concrete conclusions about the relation between OS and FMF in the gestational period. Moreover, serum analysis during attacks may also provide further data on this issue. Thiol/disulphide examination in all three trimesters, as a longitudinal study design, might manifest not only the alterations of OS markers during gestation but also the percentile ranks for the gestational weeks.

In conclusion, the main outcome of our study demonstrates a relation between OS and pregnant women with FMF in the first trimester of gestation. OS may be a major aetiological factor of unfavourable pregancy outcomes in this group of patients. The method used in our study seems to be a promising, practical and daily applicable test for evaluating thiol/disulphide homeostasis in pregnant women with FMF.

Disclaimer statement

Contributors All authors contributed equally.

Funding None.

Conflict of interest The authors report no conflicts of interest.

Ethics approval The protocol of this prospective case control study was approved by the institutional review board for human studies of Zekai Tahir Burak Women's Health Care Training and Research Hospital.