Human Reproduction, Vol.30, No.4 pp. 853– 860, 2015 Advanced Access publication on March 3, 2015 doi:10.1093/humrep/dev031
ORIGINAL ARTICLE Infertility
Serum miR-155 as a potential biomarker of male fertility Christos Tsatsanis 1,2,*, Johannes Bobjer 1,3, Hamideh Rastkhani 4, Erini Dermitzaki2, Marianna Katrinaki 2, Andrew N. Margioris 2, Yvonne Lundberg Giwercman 4, and Aleksander Giwercman 1,3 1
Molecular Reproductive Research Group, Department of Translational Medicine, Lund University, Malmo¨, Sweden 2Department of Clinical Chemistry, School of Medicine, University of Crete, Heraklion, Crete, Greece 3Reproductive Medicine Centre, Ska˚ne University Hospital Malmo¨, Malmo¨, Sweden 4Molecular Genetic Reproductive Medicine, Department of Translational Medicine, Lund University, Malmo¨, Sweden
Submitted on September 12, 2014; resubmitted on January 23, 2015; accepted on February 3, 2015
study question: Are serum levels of micro-RNAs miR-155 and miR-146a associated with male fertility, low-grade systemic inflammation (LGSI) and androgens?
summary answer: miR-155 was associated with male subfertility independent of LGSI or androgens while miR-146a was only weakly associated with subfertility and LGSI. what is known already: Male subfertility has been associated with LGSI as well as with androgen deficiency. miR-155 and miR-146a are central regulators of inflammation and their level in cells and in the serum has been associated with several inflammatory conditions. study design, size, duration: In this case –control study, two independent groups of 60 subjects each (exploratory and confirmatory cohort) were randomly selected from an ongoing study on subfertile men (in total: hypogonadal; n ¼ 40, eugonadal; n ¼ 40 and control group n ¼ 39) at a University Hospital Reproductive Medicine Centre. Individuals were matched for age.
participants/materials, setting, methods: Total RNA was isolated from cell-free serum. As internal control a synthetic miRNA, UniSp6, was added to each sample prior to extraction. miRNA expression levels were measured by real-time RT –PCR and presented as fold difference (arbitrary units, U) from control. Sera from these individuals had been previously analyzed for hormone and cytokine levels. main results and the role of chance: Serum levels of miR-155 were associated with levels of miR-146a (P , 0.0001), but only miR-146a was associated with inflammatory markers. miR-155 was strongly associated with subfertility (for subfertile group 1.88 U, 95% confidence interval (CI) 1.6 –2.1 U versus 1.15, 95% CI 1.0 –1.2 U in controls; P ¼ 0.001). Receiver operating characteristic curve analysis indicated that miR-155 but not miR-146a can be used as a marker of subfertility. MiR-155 with a cutoff value of 1.77 had 47% sensitivity and 95% specificity for identifying subfertility and a positive predictive value (PPV) and negative predictive value (NPV) of 95 and 47%, respectively. When used in combination with FSH, sensitivity and specificity were 80 and 100%, respectively, while PPV and NPV were 100 and 71%, respectively, those values being higher than for the FSH alone. Repeating the results obtained in the exploratory cohort in an independent confirmatory cohort reduced the risk of a chance finding.
limitations, reasons for caution: Although the results from the exploratory cohort were confirmed in the confirmatory cohort, studies from other centers are needed to establish the role of miR-155 as a new biomarker of male fertility. Furthermore, the role of this marker in distinguishing between different groups of male subfertility is to be elucidated. wider implications of the findings: Association of the inflammatory miRNA miR-155 with male fertility contributes to our understanding of the pathophysiology of subfertility and suggests a novel biomarker. Serum miR-155 in combination with FSH has higher diagnostic specificity and sensitivity compared with FSH alone.
study funding/competing interest(s): This work was supported by grants from Swedish Governmental Grant (ALF), Skane county council research and development foundation, Skane University Hospital Fonds and by the EU and Greek funds under the action ‘Education and lifelong learning’ program THALIS-FAT-VESSEL (No 379527). The authors have no competing interests to disclose. Key words: micro-RNA / male fertility / testosterone / FSH / inflammation & The Author 2015. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email:
[email protected]
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 15, 2017
*Correspondence address. Department of Clinical Chemistry-Biochemistry, School of Medicine, University of Crete, PO BOX 2208, Heraklion 71003, Crete, Greece. Tel:+30-2810-394833; Fax: +30-2810-394581; E-mail:
[email protected]
854
Introduction
Materials and Methods Ethics statement The study was approved by the regional ethical review board at Lund University, Sweden. All subjects provided written informed consent.
Study design Subjects for this nested study were selected from participants of an ongoing case –control study on hypogonadism among men seeking medical care due
to subfertility, as described previously (Bobjer et al., 2013). The primary subfertility cohort comprised a consecutive group of men from couples unable to conceive after 1 year of regular sexual activity, in the absence of an explanatory female factor (ovulatory disturbance, tubal obstruction and/or endometriosis). Furthermore, the following inclusion criteria were applied to the subfertile men: age 18 – 50 years at inclusion and sperm concentration ,20 million/ml. Of the 286 eligible men, 115 (40%) did not reply despite a reminder by letter, 52 (18%) declined to take part in the study, and 119 (42%) gave their consent. The control group was selected from the Swedish Population Register by matching their date of birth with that of the subfertile men. One hundred men were included at the time of selection of subjects for the present study, from which the controls were randomly selected (see below). None of the selected controls were previously or currently receiving any fertility or cancer treatment.
Inclusion Physical examination including assessment of height (without shoes, with a stadiometer to the nearest 1 mm) and weight (in light clothing, with an electronic scale to the nearest 0.1 kg) was undertaken. Information on other diseases as well as active medication was retrieved from interviews at inclusion. None of the participants reported any serious medical illness, other than subfertility. Among the randomly selected subgroups, namely hypogonadal infertile, eugonadal infertile and control, one subject presented with Klinefelter syndrome (47, XXY karyotype), one reported two previous occasions of epididymitis, one had had mumps orchitis as a child and one who was unilaterally orchidectomized due to tuberculosis in childhood was also included. One subject was on current lipid lowering therapy and two were on antihypertensive treatment. Neither therapy was discontinued prior to inclusion and these subjects were not excluded because of this. None of the subjects or controls were on any anti-diabetic/insulin therapy, opioids, glucocorticoids or other anti-inflammatory drugs. A free online random number generator (random.org) was used to select 20 hypogonadal subfertile men, 20 eugonadal subfertile men and 20 controls. Hypogonadism was strictly biochemically defined according to our laboratory’s normal levels as testosterone ,8.0 nmol/l and/or LH . 8.6 IU/l. No men, in any of the groups, received androgen replacement therapy. After excluding the subjects in the exploratory (first) cohort we repeated the same random selection process for a cohort of similar size and combination of eugonadal, hypogonadal and subfertile subjects (confirmatory cohort).
miRNA expression Fasting blood samples were drawn between 8 and 10 am from all participants. Serum was isolated and stored at 2808C immediately. No internal reference miRNAs exist in the serum therefore, for the accurate measurement of circulating miRNAs a control miRNA (UniSp2miRNA) was spiked in 200 ml of serum according to the manufacturer instructions (Exiqon, Denmark). RNA was extracted according to the manufacturer’s protocol (Exiqon, Denmark) using the MirCury RNA isolation columns for biofluids, which selectively isolate small nucleic acids. Contaminating cells or cell debris were carefully removed by two centrifugations: one at the stage of serum isolation and a second one by centrifugation at 3000g for 5 min each, prior to isolation of serum RNA, and supernatants were carefully removed to avoid contamination with the pellet. Samples were not treated with RNAse or DNAse to avoid degradation of the circulating nucleic acids. RNA is not quantifiable in the serum preparations and quantification was provided by adding exogenous miRNA. cDNA was synthesized and levels of miR-155, miR-146a were measured by real-time PCR according to the manufacturer instructions using primers validated by the manufacturer for selective amplification of the target miRNAs (Exiqon, Denmark). Briefly, cDNA prepared with the
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 15, 2017
Male subfertility is a heterogeneous condition for which the pathogenic mechanisms include abnormal androgen levels and systemic inflammation (Bobjer et al., 2013; Rusz et al., 2013). Several attempts have been made to define serum biomarkers that can be used to identify subfertile men and a series of those are being used, including reproductive hormone levels, but they can only identify a fraction of the subfertile or infertile population (Andersson et al., 2004). Micro-RNAs (miRNAs) are small RNAs that control mRNA stability and translation. miRNAs play important roles in the regulation of cell activation, proliferation and apoptosis (O’Neill et al., 2012). In immune cells miRNAs have been identified as central regulators of inflammation. Among miRNAs, miR-155 and miR146a levels are up-regulated in macrophages upon activation and have been causally involved in multiple inflammatory pathologies (O’Neill et al., 2012). MiRNAs are also secreted from cells into exosomes as paracrine or endocrine messengers and can be detected in the serum as circulating miRNAs (De Guire et al., 2013). Circulating miRNAs are increasingly being identified as potential biomarkers for several conditions, including inflammatory diseases and low-grade systemic inflammation (LGSI)-related pathologies. MiR-155 and miR-146a have been found up-regulated in the serum of patients who are chronically infected with hepatitis C virus (Bala et al., 2012) and down-regulated in the serum of systemic lupus erythematosus patients (Wang et al., 2010). Serum miR-155 has also been found elevated in breast cancer patients, as it is overexpressed in malignant cells and biopsies from cancer specimens (Eichelser et al., 2013). Thus, circulating miRNAs may be of functional significance in disease development and progression. Male subfertility is associated with low testosterone levels (Andersson et al., 2004). Several studies have linked low testosterone levels with the development of metabolic syndrome (Laaksonen et al., 2004) and LGSI-related disorders such as cardiovascular disease (Phillips et al., 1994). Recent studies from our group have analyzed serum from subfertile and control men for the expression of hormones, including androgens and estrogens, and a series of inflammatory markers, including cytokines and chemokines, showing that LGSI in young men is associated with low androgens (Bobjer et al., 2013). Additional studies support that inflammation, apparent in either inflammatory diseases (Moretti et al., 2009; Rusz et al., 2013) or in the metabolic syndrome (Morgante et al., 2011; Michalakis et al., 2012), may be linked to male subfertility. In the present study we investigated whether inflammatory miRNAs in the serum were associated with hypogonadism and/or fertility in men. For this purpose we analyzed the serum from 120 individuals, including healthy male controls and hypogonadic subfertile and eugonadic subfertile men, for the presence of the inflammatory miRNAs miR-155 and miR-146a.
Tsatsanis et al.
855
Micro-RNA-155 in male subfertility
miRCURY cDNA synthesis kit II (Exiqon, Denmark) was diluted 1:20 and 4 ml were added in Exilent SYBRgreen PCR master mix (Exiqon, Denmark) and subjected to real-time PCR under the following conditions: 1 cycle at 958C for 10 min, 45 cycles of 958C for 10 s, 608C for 1 min, followed by a dissociation cycle in a Stratagene Mx3000p instrument. The no-template control did not produce any product. The results were analyzed using the Stratagene MxPro software. The relative levels of miRNAs in the serum were measured using the 22(DDCt) method (Pfaffl, 2001) using as baseline Ct the mean of the Ct values of the control group.
Hormone and cytokine/chemokine assessment
Statistical analysis Since real-time PCR is a semi-quantitative method, we used the 22(DDCt) method (Pfaffl, 2001) to compare the Ct values obtained in each sample, using as baseline the mean Ct from the ones in the control population. Therefore, levels of miR-155 and miR-146a are shown in arbitrary units (U) defined as fold difference from the control. Statistical analysis was performed as follows: First, using Spearman correlation test, we tested following correlations: (i) between expression of miR-155 and miR-146a; (ii) each miRNA versus levels of TT, free testosterone (fT), LH, SHBG, estradiol and FSH; (iii) miRNA levels versus those of following inflammatory markers that have been previously analyzed: high sensitivity C-reactive protein (hs-CRP), TNFa, MIP1a, MIP1B, MCP1, EGF, FGF2, IFNG, IL12p40, IL12p70, IL1B, IL1RA, IL4, IL6, IL7, IL10, IL13, IL17, IP10, IL9 and IL8. Thereafter, using a non-parametric Mann – Whitney test we did the following: (i) evaluated differences in the levels of miR-155 and miR-146a between the subfertile and control groups; (ii) to determine potential association of the miRNA levels with hormonal status in subfertile men we compared the levels of miR-155 and miR-146a between the control population and eugonadal subfertile or hypogonadal subfertile subjects. These analyses were primarily carried out in the exploratory cohort and, thereafter, the findings were tested in the confirmatory cohort and the results were subsequently merged. Finally, we performed receiver operating characteristic curve (ROC) analysis for each miRNA in relation to fertility status and we determined specificity and sensitivity as well as prediction values for subfertility
Results miR-155 and miR-146a are present in male serum and are correlated with each other Both miR-155 and miR-146a were detectable in all serum samples (Table I). Correlation analysis between the levels of these two miRNAs showed that they were correlated with each other (Spearman r ¼ 0.2, P , 0.0001) (Fig. 1A).
Association of serum miR-155 and miR146a levels with circulating hormones Correlation analysis between expression of each miRNA and expression of fT, TT, LH, FSH, SHBG or estradiol showed that the levels of these miRNAs were not associated with any of these hormones, with the exceptions of miR-155 that was associated with FSH (r ¼ 0.30, P ¼ 0.001) and to a lesser extent with LH (r ¼ 0.23, P ¼ 0.01), and miR-146a with TSH (r ¼ 20.26, P ¼ 0.006) (Fig. 1B and Table II).
Association of serum miR-155 and miR146a levels with inflammatory markers Analyses of miR-155 or miR-146a revealed association of miR-155 only with IL-8. In contrast, miR-146a was statistically significantly associated with EGF, IL-10, IL-12p40, IL-13, IL-1RA, IL-7 and hs-CRP (Table III).
Association of serum miR-155 and miR146a levels with fertility status Comparing the relative expression levels of miR-155 or miR-146a with fertility status in the exploratory cohort revealed that miR-155 was associated with subfertility (P ¼ 0.001, control group mean value 1.17 U, 95% confidence interval (CI) 0.98– 1.4 U, and subfertile group mean value 2.23 U, 95% CI 1.74 –2.7 U) while miR-146a was not (P ¼ 0.23, control group mean value 1.18 U, 95% CI 0.99 –1.38 U and subfertile group mean value 1.42 U, 95% CI 1.2 –1.65 U). When we divided the groups according to their hormonal status, we found that miR-155 levels were statistically significantly higher in subfertile eugonadal men than in the control group (eugonadal subfertile group mean value 2.25 U, 95% CI 1.8 –2.7 U) (P ¼ 0.0008). MiR-155 values also differed between hypogonadal subfertile male and the control group (hypogonadal group mean value 2.18 U, 95% CI 0.38 –3.7 U) (P ¼ 0.03). MiR-146a was not associated with subfertility in either eugonadal or hypogonadal individuals. Analyzing the second independent confirmatory cohort we found similar results, further supporting the initial observations (Table I). Combining the data of the two cohorts increased the power of the analysis and revealed a stronger association between miR-155 and subfertility. Specifically, miR-155 was associated with subfertility (P , 0.0001, control group mean value 1.15 U, 95% CI 1.03 –1.27 U and subfertile group mean value 1.88 U, 95% CI 1.6 –2.1 U) while miR-146a was marginally significant (P ¼ 0.03, control group mean value 1.17 U, 95% CI 1.04 – 1.30 U and subfertile group mean value 1.48 U, 95% CI 1.33 –1.64 U). When dividing the groups according to hormonal status, miR-155
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 15, 2017
Hormone levels in plasma were assessed at the Department of Clinical Chemistry, Ska˚ne University Hospital, Malmo¨, Sweden. Serum values of total testosterone (TT) were assessed by a two-step competitive immunoassay with a luminometric technique (Electro ChemiLuminescence Immunoassay, ECLI); lower detection level (LDL) 0.087 nmol/l; imprecision (CV%) 2.4% at 1.9 nmol/l and 4.0% at 25.5 nmol/l. LH, FSH and sex hormonebinding globulin (SHBG) concentrations were determined with a one-step immunometric sandwich assay with a luminometric technique (ECLI); LH 0.10 IU/l; CV%, 2.0% at 5.0 IU/l and 2.2% at 55.2 IU/l; SHBG 0.35 nmol/ l; CV%, 1.0% at 22.0 nmol/l and 1.1% at 49.5 nmol/l. Estradiol concentrations in serum were assessed by an immunofluorometric method (DELFIA Estradiol, Wallace OY); FSH 8 pmol/l; CV%, 20% at 30 pmol/l and 10% at 280 pmol/l. Free testosterone was calculated from TT, SHBG and fixed albumin levels (Vermeulen et al., 1999). The cytokines and chemokines interferon gamma (IFNG), interleukin 12p40 (IL12p40), IL12p70, IL1B, interleukin 1 receptor antagonist (IL1RA), IL4, IL7, IL10, IL13, IL17, interferon gamma-induced protein 10 (IP10), IL9, endothelial growth factor (EGF), fibroblast growth factor 2 (FGF2), macrophage inflammatory protein 1 beta (MIP1B) were measured using a bead-based enzyme-linked immunosorbent assay (ELISA) method (Lincoplex, Millipore) for the exploratory cohort and tumor necrosis factor alpha (TNFa), macrophage inflammatory protein 1 alpha (MIP1a), monocyte chemotactic protein 1 (MCP1), IL6 and IL8 were measured by sandwich ELISA (e-Bioscience) for both the exploratory and confirmatory cohorts.
for each miRNA alone and by combining them with the established marker of FSH. Analyses were performed using SPSS and GraphPad Prizm software, using P , 0.05 as minimum level of significance.
856
Tsatsanis et al.
Table I Median levels of micro-RNAs (miRNAs) in the two subgroups of subfertile men (20 hypo- and 20 eugonadal) as compared with the 19 controls in the exploratory and confirmatory cohorts as well as in a combination of the cohorts. miRNA (U)*
Exploratory cohort (n 5 59)
Subfertile men
................................................
HG (n 5 20)
EG (n 5 20)
Mean
Mean
.............................. Mean
Min
Max
All subfertile (n 5 40) Mean
Controls (n 5 19)
P, HG versus control
P, EG versus control
P, Subf. versus control
P, HG versus EG
Mean
............................................................................................................................................................................................. miR-155
1.87
0.56
7.66
2.18
2.25
2.23
1.17
0.03
0.0008
0.001
NS
miR146a
1.17
0.62
3.85
1.32
1.78
1.42
1.18
NS
NS
NS
NS
............................................................................................................................................................................................. miRNA (U)
Confirmatory cohort (n 5 60)
Subfertile men
................................................
HG (n 5 20)
EG (n 5 20)
Mean
Mean
.............................. Mean
Min
Max
All subfertile (n 5 40) Mean
Controls (n 5 20)
P, HG versus control
P, EG versus control
P, Subf. versus control
P, HG versus EG
Mean
............................................................................................................................................................................................. 1.44
0.53
2.97
1.37
1.80
1.60
1.12
0.02
0.0002
0.0005
NS
miR146a
1.42
0.65
4.00
1.79
1.35
1.56
1.15
0.003
NS
0.0033
0.02
............................................................................................................................................................................................. All (n 5 119)
miRNA (U)
Subfertile men
................................................ HG (n 5 40)
EG (n 5 40)
Mean
Mean
.............................. Mean
Min
Max
All subfertile (n 5 80) Mean
Controls (n 5 39)
P, HG versus control
P, EG versus control
P, subfertile versus control
P, HG versus EG
Mean
............................................................................................................................................................................................. miR-155
1.68
0.53
7.66
1.58
2.05
1.88
1.15
0.04
,0.0001
,0.0001
0.002
miR146a
1.39
0.62
4.00
1.83
1.45
1.48
1.17
0.0001
NS
0.030
0.015
NS, not statistically significant; HG, hypogonadal; EG, eugonadal. *miRNA levels are presented as fold difference (arbitrary units, U) from control. Significance was calculated by non-parametric Mann– Whitney test.
Figure 1 Correlation analyses for micro (mi)RNA expression levels in the serum of men. (A) Correlation of micro-RNA (miR)-155 relative expression levels with those of miR-146a and (B) correlation of miR-155 relative expression levels with FSH (nM), using linear regression analysis.
levels were higher in subfertile eugonadal men than in the control group, confirming the previous result (eugonadal subfertile group mean value 2.05 U, 95% CI 1.76 –2.33 U, P , 0.0001). miR-155 values also differed between the hypogonadal subfertile male and the control group (hypogonadal group mean value 1.58 U, 95% CI 1.1 –2.1 U, P ¼ 0.04) (Table I). Analysis of miR-146a levels revealed a strong association between hypogonadal and control men (hypogonadal group mean value 1.83 U, 95% CI 1.5–2.1 U, P ¼ 0.0001). The relative levels of miR-155 were higher in subfertile men compared with the control group regardless of their testosterone levels (mean value
1.88 U, 95% CI, 1.6 –2.1 U, P , 0.0001) (Fig. 2A and Table I) while a modest difference was observed in the levels of miR-146a between the same groups (mean value 1.48 U, 95% CI, 1.31 –1.64 U, P ¼ 0.03) (Fig. 2B and Table I).
MiR-155 is a potential marker of subfertility ROC analysis of the relative expression of miR-155 in the total population analyzed (n ¼ 119) showed statistical significance (P , 0.0001, area under the curve (AUC) 0.76) (Fig. 2C). ROC analysis for the
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 15, 2017
miR-155
857
Micro-RNA-155 in male subfertility
Table II Associations between levels of miRNAs and hormone concentrations in serum from all subjects included in the study. Hormone
miR-155
miR-146a
................................................................................. ..................................................................................
n
Spearman rho
95% CI
P
Significance
n
Spearman rho
95% CI
P
Significance
............................................................................................................................................................................................. FSH
119
0.30
0.11 to 0.46
0.001 **
118
0.16
20.032 to 0.34
0.16
NS
LH
119
0.23
0.05 to 0.41
0.01
*
118
0.11
20.08 to 0.29
0.11
NS
Total testosterone
119
20.12
20.20 to 0.17
0.89
NS
118
0.12
20.06 to 0.31
0.19
NS
Free (calculated) testosterone
119
20.11
20.28 to 0.09
0.28
NS
118
0.05
20.14 to 0.24
0.92
NS
SHBG
119
20.005
20.19 to 0.18
0.95
NS
118
0.03
20.15 to 0.22
0.45
NS
PSA
119
20.05
20.24 to 0.13
0.53
NS
118
0.09
20.09 to 0.28
0.32
NS
119
20.05
20.23 to 0.14
0.60
NS
118 20.26
20.43 to 20.07
0.006
**
119
20.08
20.26 to 0.11
0.38
NS
118 20.08
20.26 to 0.11
0.40
NS
F-T4
119
20.15
20.33 to 0.04
0.11
NS
118
0.13
20.05 to 0.31
0.11
NS
Estradiol
119
0.10
20.08 to 0.28
0.27
NS
118
0.16
20.02 to 0.34
0.27
NS
Spearman correlation analysis, CI, confidence interval; NS, not statistically significant. SHBG, sex hormone-binding globulin; PSA, prostate serum antigen; FT3, free tri-iodothyronine; FT4, free thyroxine. *P , 0.05. **P , 0.01.
same population using the established subfertility markers FSH and LH showed similar values (FSH P , 0.0001 AUC 0.84; LH P , 0.0001 AUC 0.82). In contrast, the levels of miR-146a did not reveal significance in ROC analysis. miR-155 with a cutoff value of 1.77 U conferred 47% sensitivity and 95% specificity, a positive predictive value (PPV) 95% and negative predictive value (NPV) 47% for diagnosis of subfertility (Table IV). FSH, an established marker of subfertility with a cutoff value of 8 nM had 62% sensitivity (95% CI: 50– 73%) and 94% specificity, PPV 96% and NPV 56%. Combination of FSH with a cutoff of 8 nM and miR-155 with a cutoff value of 1.77 U, meaning having either of the two markers positive, resulted in 80% sensitivity and 100% specificity, a PPV 100% and NPV 71% for subfertility and AUC 0.88 (P , 0.0001) (Table IV).
Discussion Subfertility is a growing problem in western societies and the factors that contribute to the condition are diverse and include hypogonadism, obesity and associated LGSI. In the present study we identified circulating miRNAs as being associated with subfertility, providing a novel potential biomarker which performs similarly well when compared with the current marker FSH and can effectively be used in combination with FSH providing increased specificity and sensitivity. Circulating miRNAs are promising biomarkers for several disorders. These miRNAs function as endocrine messengers and are involved in the regulation of cellular responses. miR-155 and miR-146a are miRNAs found in multiple cell types and involved in diseases including inflammation (Androulidaki et al., 2009; Arranz et al., 2012; Vergadi et al., 2014). miRNA expression is induced in macrophages and T-cells and their levels in the serum are elevated in different inflammatory diseases (Murakami et al., 2012; Wang et al., 2012; Olivieri et al., 2013; Tacke
et al., 2014) while their presence in the circulation in exosomes may have functional significance as endocrine messengers (McDonald et al., 2014). Surprisingly our analysis did not reveal any strong association of miR-155 with inflammatory indices other than a weak negative association with IL-8, while miR-146a was weakly correlated with CRP and the anti-inflammatory mediators IL-10 and IL1-RA, the immunomodulatory cytokines IL-13 and IL-7 and the growth factor EGF, suggesting a link with anti-inflammatory events and the M2-type of innate immune responses. In the present study we measured the levels of miR-155 and miR-146a in a nested population of healthy and subfertile male individuals. Earlier studies from our group have shown that subfertility is associated with low androgens and LGSI (Bobjer et al., 2013). The population analyzed was a nested group of men who were identified as a consecutive group of men from couples unable to conceive after 1 year of regular sexual activity, in the absence of an explanatory female factor. The subfertility group included equal numbers of eugonadal and hypogonadal men using a previously described biochemical definition of hypogonadism based on low serum testosterone concentrations as well as high LH levels, thereby including men with compensated testicular endocrine dysfunction (Bobjer et al., 2012). Even though hypogonadism was associated with LGSI, in the present analysis circulating miR-155 or miR-146a were not associated with testostosterone or LH levels. Only miR-155 was weakly associated with FSH and LH. In our previous analysis testosterone levels were associated with pro-inflammatory cytokines and chemokines, particularly TNFa, IL-6 and MIP1a (Bobjer et al., 2013). Herein no association was observed between miR-146a and miR-155 and testosterone or LH, indicating that their levels may not be affected by androgens. While these two miRNAs are causally linked with increased inflammation they were not
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 15, 2017
TSH F-T3
858
Tsatsanis et al.
Table III Associations between levels of miRNAs and inflammatory marker concentrations in serum from all groups of men included in the study. Inflammatory marker
miR-155
miR-146a
............................................................................... .................................................................................. n
Spearman rho
95% CI
P
Significance
n
Spearman rho
95% CI
P
Significance
.............................................................................................................................................................................................. EGF
58
20.04
20.31 to 0.22
0.72
NS
57
0.39
0.12 to 0.60
0.003
**
FGF2
59
0.02
20.24 to 0.28
0.86
NS
58
20.03
20.30 to 0.24
0.818
NS
IFNG
59
20.08
20.34 to 0.19
0.54
NS
58
20.25
20.49 to 0.02
0.070
NS
IL-10
59
20.25
20.49 to 0.01
0.05
NS
58
20.39
20.60 to 20.12
0.003
**
IL-12p40
59
20.14
20.39 to 0.12
0.28
NS
58
20.32
20.55 to 20.04
0.018
*
IL-12p70
59
0.06
20.20 to 0.32
0.63
NS
58
20.13
20.40 to 0.14
0.322
NS *
59
20.08
20.34 to 0.18
0.50
NS
58
20.30
20.53 to 20.03
0.025
59
0.07
20.19 to 0.33
0.57
NS
58
20.08
20.35 to 0.19
0.550
NS
IL-1B
59
20.07
20.33 to 0.19
0.57
NS
58
20.22
20.47 to 0.05
0.098
NS
IL-1RA
59
20.11
20.37 to 0.15
0.38
NS
58
20.27
20.51 to 20.0007 0.043
*
IL-4
59
20.08
20.34 to 0.18
0.53
NS
58
20.11
20.37 to 0.17
0.421
NS
IL-6
119
20.09
20.27 to 0.09
0.31
NS
118
20.14
20.32 to 0.049
0.133
NS
IL-7
59
20.18
20.42 to 0.09
0.17
NS
58
20.33
20.56 to 20.06
0.013
*
IL-8
119
20.26
20.42 to 20.07
0.004
**
118
20.01
20.20 to 0.17
0.88
NS
IL-9
59
20.08
20.34 to 0.18
0.51
NS
58
20.26
20.50 to 0.01
0.057
NS
IP-10
59
0.05
20.21 to 0.31
0.68
NS
58
0.12
20.16 to 0.38
0.389
NS
MCP1
119
20.13
20.31 to 0.05
0.15
NS
118
0.001
20.19 to 0.19
0.99
NS
MIP1a
119
0.04
20.14 to 0.22
0.63
NS
118
0.07
20.11 to 0.26
0.097
NS
MIP1b
59
20.12
20.37 to 0.14
0.35
NS
58
20.05
20.33 to 0.22
0.676
NS
TNFa
119
20.04
20.22 to 0.15
0.68
NS
118
20.04
20.23 to 0.14
0.64
NS
hs-CRP
59
0.04
20.22 to 0.30
0.74
NS
58
0.33
0.06 to 0.56
0.013
*
Spearman correlation analysis. IL, interleukin; EGF, epidermal growth factor; FGF2, fibroblast growth factor; IFNG, interferon gamma; IP10, interferon gamma-induced protein 10; MCP1, monocyte chemotactic protein 1; MIP1a, macrophage inflammatory protein 1-alpha; MIP1B, macrophage inflammatory protein 1-beta; TNFa, tumor necrosis factor alpha; hs-CRP, high sensitive c-reactive protein; NS, not statistically significant. *P , 0.05. **P , 0.01.
associated with these particular cytokines. CRP is a widely used marker of both acute and LGSI. miR-146a but not miR-155 was associated with CRP, suggesting that miR-155 may be affected by multiple factors other than inflammation and therefore no association is observed. miR-155 is a pleiotropic miRNA and has been associated with regulation of cell proliferation and apoptosis, in addition to inflammation (Huang et al., 2013; Zhang et al., 2013; Zhu et al., 2013). It is, thus, possible that increased levels of miR-155 may reflect conditions that affect fertility other than LGSI. When comparing each miRNA with all subfertile men, including both eugonadal and hypogonadal, miR-155 showed association with subfertility, and the association was stronger when only eugonadal men were included. MiR-146a showed a weaker association which was stronger when comparing hypogonadal infertile men with control individuals, suggesting that it may reflect LGSI present in hypogonadic infertile individuals (Bobjer et al., 2013). ROC analysis revealed that miR-155 but not miR-146a is a potential biomarker of subfertility, independent of androgens or LGSI. To date, androgens, FSH and accordingly LH are used as serum biomarkers to diagnose subfertility. The combination of FSH
and miR-155 in our group conferred a PPV of 100% and a negative predictive value of 71%. Elevated serum miR-155 may be either an indirect marker of subfertility that reflects tissue damage (i.e. in the blood-testis barrier) or may also have a functional role in suppressing spermatogenesis. Even though no information exists on the role of miRNAs in the regulation of blood-testis barrier, it has been shown that miR-155 directly affects the blood –brain barrier by increasing its permeability through de-stabilizing cell-to-cell adhesion by suppressing claudin-1 and the adhesion molecules DOCK-1 and syntenin-1 (Lopez-Ramirez et al., 2014). Micro-RNAs are also important in Sertoli cell differentiation and function (Papaioannou et al., 2011), suggesting that miR-155 may be a product of Sertoli cells released in the circulation. Alternatively, elevated miR-155 may reflect systemic tissue damage, and therefore indirectly impact fertility, as it is known to be involved in different types of tissue damage (Ammari et al., 2013; Yao et al., 2014). Performing studies on rather limited numbers of subjects implies a risk of selection bias and, thereby, type I error. However, since our primary results obtained in an exploratory cohort were repeated in an
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 15, 2017
IL-13 IL-17
859
Micro-RNA-155 in male subfertility
Table IV Diagnostic value of miR-155, FSH and the combination of miR-155 with FSH to predict subfertility in men. miR-155 (cutoff 1.77 U)
FSH (cutoff 8 nM)
Combination of miR-155 (cutoff 1.77 U) and FSH (cutoff 8 nM)
............................................................................................................................................................................................. Sensitivity
¼ 47%
95% CI: 35–58%
Sensitivity
¼ 62%
95% CI: 50– 73%
Sensitivity
¼ 80%
95% CI: 70– 88%
Specificity
¼ 95%
95% CI: 82–99%
Specificity
¼ 94%
95% CI: 82– 99%
Specificity
¼ 100%
95% CI: 91– 100%
PPV
¼ 95%
95% CI: 83–99%
PPV
¼ 96%
95% CI: 87– 99%
PPV
¼ 100%
95% CI: 94– 100%
NPV
¼ 47%
95% CI: 35–59%
NPV
¼ 56%
95% CI: 43– 69%
NPV
¼ 71%
95% CI: 57– 82%
AUC
¼ 0.76
P , 0.0001
AUC
¼ 0.84
P , 0.0001
AUC
¼ 0.88
P , 0.0001
PPV, positive predictive value; NPV, negative predictive value; AUC, area under the curve; N/A, not applicable.
independently selected confirmatory cohort, the risk of a chance finding is significantly reduced. In conclusion, circulating miRNAs are potential biomarkers of subfertility. Our results indicate that miR-155 may be biomarker of fertility, which is independent from androgens, estrogens or LGSI markers and can potentially be used in combination with FSH.
Acknowledgements We would like to thank the research nurses Irene Leijonhufvud, Gunilla H Lundberg and Matilda Engquist, Reproductive Medicine Centre Malmo¨, Sweden for help with the patient inclusion and Susanne Lundin for administration of blood samples.
Authors’ roles C.T. designed the study, performed analyses and drafted the manuscript. J.B. collected samples, performed analyses and drafted the manuscript. H.R. performed analyses. E.D. and M.K. performed analyses and drafted the manuscript. A.N.M. and Y.L.G. designed the study and drafted the manuscript. A.G. designed the study, performed analyses and drafted the manuscript.
Funding This work was supported by grants from Swedish Governmental Grant (ALF), Skane county council research and development foundation, Skane University Hospital Fonds and by the EU and Greek funds under
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 15, 2017
Figure 2 Association of miRNAs with subfertility in men. (A and B) Comparison of the levels of miR-155 (n ¼ 119) (A) and miR-146a (n ¼ 115) (B) between the control and the subfertile groups using non-parametric Mann – Whitney test. (C) Receiver operating characteristic curve analysis for miR-155 as a marker for subfertility.
860 the action ‘Education and lifelong learning’ program THALIS-FAT-VESSEL (No 379527). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Conflict of interest None declared.
References
Michalakis K, Mintziori G, Kaprara A, Tarlatzis BC, Goulis DG. The complex interaction between obesity, metabolic syndrome and reproductive axis: a narrative review. Metabolism 2012;62:457 –478. Moretti E, Cosci I, Spreafico A, Serchi T, Cuppone AM, Collodel G. Semen characteristics and inflammatory mediators in infertile men with different clinical diagnoses. Int J Androl 2009;32:637– 646. Morgante G, Tosti C, Orvieto R, Musacchio MC, Piomboni P, De Leo V. Metformin improves semen characteristics of oligo-teratoasthenozoospermic men with metabolic syndrome. Fertil Steril 2011; 95:2150 – 2152. Murakami Y, Toyoda H, Tanahashi T, Tanaka J, Kumada T, Yoshioka Y, Kosaka N, Ochiya T, Taguchi YH. Comprehensive miRNA expression analysis in peripheral blood can diagnose liver disease. PLoS One 2012; 7:e48366. O’Neill LA, Sheedy FJ, McCoy CE. MicroRNAs: the fine-tuners of Toll-like receptor signalling. Nat Rev Immunol 2012;11:163 – 175. Olivieri F, Rippo MR, Procopio AD, Fazioli F. Circulating inflamma-miRs in aging and age-related diseases. Front Genet 2013;4:121. Papaioannou MD, Lagarrigue M, Vejnar CE, Rolland AD, Kuhne F, Aubry F, Schaad O, Fort A, Descombes P, Neerman-Arbez M et al. Loss of Dicer in Sertoli cells has a major impact on the testicular proteome of mice. Mol Cell Proteomics 2011;10:M900587MCP900200. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 2001;29:e45. Phillips GB, Pinkernell BH, Jing TY. The association of hypotestosteronemia with coronary artery disease in men. Arterioscler Thromb 1994; 14:701 – 706. Rusz A, Pilatz A, Wagenlehner F, Linn T, Diemer T, Schuppe HC, Lohmeyer J, Hossain H, Weidner W. Influence of urogenital infections and inflammation on semen quality and male fertility. World J Urol 2013; 30:23 – 30. Tacke F, Roderburg C, Benz F, Cardenas DV, Luedde M, Hippe HJ, Frey N, Vucur M, Gautheron J, Koch A et al. Levels of circulating miR-133a are elevated in sepsis and predict mortality in critically ill patients. Crit Care Med 2014;42:1096– 1104. Vergadi E, Vaporidi K, Theodorakis EE, Doxaki C, Lagoudaki E, Ieronymaki E, Alexaki VI, Helms M, Kondili E, Soennichsen B et al. Akt2 deficiency protects from acute lung injury via alternative macrophage activation and miR-146a induction in mice. J Immunol 2014;192:394– 406. Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 1999;84:3666 – 3672. Wang G, Tam LS, Li EK, Kwan BC, Chow KM, Luk CC, Li PK, Szeto CC. Serum and urinary cell-free MiR-146a and MiR-155 in patients with systemic lupus erythematosus. J Rheumatol 2010;37:2516 – 2522. Wang H, Peng W, Shen X, Huang Y, Ouyang X, Dai Y. Circulating levels of inflammation-associated miR-155 and endothelial-enriched miR-126 in patients with end-stage renal disease. Braz J Med Biol Res 2012; 45:1308 – 1314. Yao M, Wang X, Tang Y, Zhang W, Cui B, Liu Q, Xing L. Dicer mediating the expression of miR-143 and miR-155 regulates hexokinase II associated cellular response to hypoxia. Am J Physiol Lung Cell Mol Physiol 2014; 307:L829– L837. Zhang CM, Zhao J, Deng HY. MiR-155 promotes proliferation of human breast cancer MCF-7 cells through targeting tumor protein 53-induced nuclear protein 1. J Biomed Sci 2013;20:79. Zhu GF, Yang LX, Guo RW, Liu H, Shi YK, Wang H, Ye JS, Yang ZH, Liang X. miR-155 inhibits oxidized low-density lipoprotein-induced apoptosis of RAW264.7 cells. Mol Cell Biochem 2013;382:253 – 261.
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 15, 2017
Ammari M, Jorgensen C, Apparailly F. Impact of microRNAs on the understanding and treatment of rheumatoid arthritis. Cur Opin Rheumatol 2013;25:225 – 233. Andersson AM, Petersen JH, Jorgensen N, Jensen TK, Skakkebaek NE. Serum inhibin B and follicle-stimulating hormone levels as tools in the evaluation of infertile men: significance of adequate reference values from proven fertile men. J Clin Endocrinol Metab 2004;89:2873– 2879. Androulidaki A, Iliopoulos D, Arranz A, Doxaki C, Schworer S, Zacharioudaki V, Margioris AN, Tsichlis PN, Tsatsanis C. The kinase Akt1 controls macrophage response to lipopolysaccharide by regulating microRNAs. Immunity 2009;31:220– 231. Arranz A, Doxaki C, Vergadi E, Martinez de la Torre Y, Vaporidi K, Lagoudaki ED, Ieronymaki E, Androulidaki A, Venihaki M, Margioris AN et al. Akt1 and Akt2 protein kinases differentially contribute to macrophage polarization. Proc Natl Acad Sci USA 2012;109:9517– 9522. Bala S, Tilahun Y, Taha O, Alao H, Kodys K, Catalano D, Szabo G. Increased microRNA-155 expression in the serum and peripheral monocytes in chronic HCV infection. J Transl Med 2012;10:151. Bobjer J, Naumovska M, Giwercman YL, Giwercman A. High prevalence of androgen deficiency and abnormal lipid profile in infertile men with non-obstructive azoospermia. Int J Androl 2012;35:688– 694. Bobjer J, Katrinaki M, Tsatsanis C, Lundberg Giwercman Y, Giwercman A. Negative association between testosterone concentration and inflammatory markers in young men: a nested cross-sectional study. PLoS One 2013;8:e61466. De Guire V, Robitaille R, Tetreault N, Guerin R, Menard C, Bambace N, Sapieha P. Circulating miRNAs as sensitive and specific biomarkers for the diagnosis and monitoring of human diseases: promises and challenges. Clin Biochem 2013;46:846 – 860. Eichelser C, Flesch-Janys D, Chang-Claude J, Pantel K, Schwarzenbach H. Deregulated serum concentrations of circulating cell-free microRNAs miR-17, miR-34a, miR-155, and miR-373 in human breast cancer development and progression. Clin Chem 2013;59:1489– 1496. Huang C, Li H, Wu W, Jiang T, Qiu Z. Regulation of miR-155 affects pancreatic cancer cell invasiveness and migration by modulating the STAT3 signaling pathway through SOCS1. Oncol Rep 2013;30:1223–1230. Laaksonen DE, Niskanen L, Punnonen K, Nyyssonen K, Tuomainen TP, Valkonen VP, Salonen R, Salonen JT. Testosterone and sex hormonebinding globulin predict the metabolic syndrome and diabetes in middleaged men. Diabetes Care 2004;27:1036 – 1041. Lopez-Ramirez MA, Wu D, Pryce G, Simpson JE, Reijerkerk A, King-Robson J, Kay O, de Vries HE, Hirst MC, Sharrack B et al. MicroRNA-155 negatively affects blood-brain barrier function during neuroinflammation. FASEB J 2014;28:2551– 2565. McDonald MK, Tian Y, Qureshi RA, Gormley M, Ertel A, Gao R, Lopez EA, Alexander GM, Sacan A, Fortina P et al. Functional significance of macrophage-derived exosomes in inflammation and pain. Pain 2014; 155:1527– 1539.
Tsatsanis et al.
