Gene Expression Analysis of the Irrigation Solution Samples Collected during Vitrectomy for Idiopathic Epiretinal Membrane

Purpose The analysis of gene expression in idiopathic epiretinal membranes (iERMs) may help elucidate ERM formation and its pathology. Here, we conducted a case-control study, in order to determine the expression levels of cytokines and other genes in eyes with macular hole (MH) or iERM. Methods Twenty eyes, obtained from seven male and 13 female patients, were included in the study. The average age of the study subjects was 69.1 ± 7.67 years, and 15 eyes had iERM, while five eyes had MH. Irrigation solution samples were collected during vitrectomy, centrifuged, and the levels of cytokine and other mRNAs in the sediment were assessed using real-time PCR. The expression level of 11 cytokine genes, four transcription factor genes, two cytoskeletal genes, and genes encoding two extracellular matrix proteins in eyes with MH or iERM were determined and compared. Results The expression levels of interleukin 6 (IL6), tumor growth factor B2 (TGFB2), vascular endothelial growth factor A (VEGFA), chemokine C-X-C motif ligand 1 (CXCL1), v-rel avian reticuloendotheliosis viral oncogene homolog A (RELA), glial fibrillary acidic protein (GFAP), and tenascin C (TNC) were significantly higher in eyes with iERM than in eyes with MH. The expression of these genes was not associated with the preoperative visual acuity of the investigated patients. Conclusions The obtained results indicate that real-time PCR analysis of irrigation solution samples collected during vitrectomy can help assess the expression levels of several genes, and that iERM is associated with the expression of pro-inflammatory genes and the genes expressed during angiogenesis and wound healing process (IL6, TGFB2, VEGFA, CXCL1, RELA, GFAP, and TNC).


Introduction
The development of epiretinal membranes (ERMs) can cause visual acuity reduction or vision distortion in the elderly. Idiopathic or secondary ERM can accompany the development of certain diseases such as proliferative vitreoretinopathy, ocular inflammation (e.g., uveitis), ocular trauma, and postoperative rhegmatogenous retinal detachment, as a response to changes in the vitreous humor [1,2]. ERM is thought to be caused by a fibrocellular proliferation of the inner limiting membrane (ILM) and the subsequent vitreoretinal adhesion and traction [3,4]. Although the precise pathogenesis remains unknown, the measurements of the levels of cytokines and other factors in the solid component of the irrigation solution may aid future studies and help elucidate the role of various proteins in the pathogenesis of ERM. Enzyme-linked immunosorbent assay (ELISA), immunoblotting, and traditional reverse transcription PCR (RT-PCR) have been used to assess the level of cytokines and/or other proteins, such as transcription factors, in the vitreous fluids [5][6][7][8][9].
In this study, we used real-time PCR to examine the expression of genes encoding cytokines and other proteins, and to measure their levels in the solid component of the irrigation solution collected during vitrectomy. ELISA requires the collection of dry vitreous humor, which necessitates the discontinuation of irrigation, and this may lead to a reduction in intraocular pressure. However, our method of choice allows normal vitrectomy, without the need for the discontinuation of the irrigation. Therefore, the mechanisms underlying the idiopathic epiretinal membrane (iERM) pathogenesis can be assessed using real-time PCR and subsequent analysis can be performed comparing the samples from eyes with macular hole (MH) and iERM.

Materials and Methods
This study was conducted according to the principles outlined in the Declaration of Helsinki and it was approved by the Institutional Review Board of Kyushu University Hospital, Japan. All study participants were informed regarding the potential consequences of the study and written informed consents were obtained from all of them.

Patients
Fifteen eyes obtained from 15 patients (six males and nine females) with iERM were included in the study. The mean age of the subjects was 71.5 ± 6.59 years. ERM was diagnosed by ophthalmoscopy and spectral-domain optical coherence tomography (OCT) (Cirrus HD-OCT, Carl Zeiss Meditec, Dublin, CA, USA). These patients underwent cataract surgery and vitrectomy at the Kyushu University Hospital, Fukuoka, Japan, between April and October 2013. The eyes of all patients were treated with an intravitreal vital dye Brilliant Blue G and triamcinolone acetonide. Patients with concomitant ocular lesions, such as those that underwent previous intravitreal injection and eye surgery, or suffered from ocular trauma, diabetic retinopathy, uveitis, retinal tears or holes, and retinal artery or vein occlusions, were excluded from the study. Five patients with ILM (one male and four females) who underwent vitrectomy for MH without ERM were included as a control group (mean age, 61.8 ± 5.84 years). One patient with MH had hyperlipidemia and asthma as well, while another patient had Sjögren's syndrome. Additionally, three patients with iERM had hypertension and four patients had heart diseases. Their treatment included warfarin in three of these patients and verapamil in one. The other patients had no systemic diseases, including autoimmune diseases.

Sample collection
Irrigation solution samples were collected during three-port 25-gauge closed vitrectomies. In eyes with cataract, phacoemulsification and aspiration were performed first, while the vitrectomy was performed after changing the irrigation solution cassette, and these irrigation solution samples were collected immediately after each surgery. Irrigation solution samples were centrifuged at 1,100 ×g for 30 min and afterward at 2,000 ×g for 10 min. Supernatants were discarded after each centrifugation. The pellets were homogenized in 300 μL of TRIzol reagent (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA), using Fast Prep System and Lysing Matrix D (FastRNA Green; QBiogene, Illkirch Cedex, France), according to the manufacturer's instructions. Following the addition of 60 μL of chloroform, samples were centrifuged at 18,000 ×g for 15 min at 4°C, and the aqueous phase was collected for a subsequent isolation of total RNA, using the Magtration nucleic acid purification kit (Precision System Science, Chiba, Japan). Reverse transcription was performed using the Transcriptor First Strand cDNA Synthesis Kit (Roche Applied Science, Basel, Switzerland).

Statistical analyses
All statistical analyses were performed using the JMP 9 software (SAS Institute Inc., Cary, NC, USA). The obtained data are presented as mean ± standard deviation (SD). For normally and equally distributed data, gene expression levels between groups were compared using t-test. For normally and unequally distributed data, Welch's t-test was used. Fisher's exact test was performed for the categorical data, because of the gender classification of patients with two investigated diseases (MH or iERM). The correlation between these two parameters and cytokine, transcription factor, cytoskeletal, or extracellular matrix gene expressions was determined using the Spearman's correlation coefficient. P < 0.05 was considered a statistically significant difference. For the analyses of the correlation between genes, ΔCt values obtained for each sample were used.

Results
The demographic characteristics of the patients belonging to the two disease groups (MH and iERM) are presented in Table 1. Significant difference in age, but no significant differences in gender and surgery time were observed between the MH and iERM groups. All investigated eyes had cataract with nuclear opacity of Emery grade 2, and they underwent phacoemulsification and aspiration. Among the eyes with MH, one eye was classified as stage 3 and four eyes were classified as stage 2, according to the Gass Classification System [10]. The average hole diameter was 337 ± 193 μm.

Correlation between the gene expression and visual acuity
A univariate analysis of correlation (Spearman's correlation coefficient) was performed for the seven genes upregulated in the iERM eyes (IL6, TGFB2, VEGFA, CXCL1, RELA, GFAP, and TNC) and the preoperative visual acuity (assessed using a logMAR chart) of the eyes with MH or iERM. No significant positive correlations were observed (Table 3). An additional Spearman's correlation coefficient analysis was performed in order to compare the expression of these seven genes between the eyes with iERM. As shown in Table 4, strong positive correlations between TGFB2 and IL6, CXCL1, or RELA, VEGFA and GFAP or TNC, and GFAP and TNC were detected. We showed scatterplots of raw data (S1 Fig), and presented p value acquired by Spearman's correlation coefficient analysis.
Here, the expression levels of four cytokine-encoding genes (IL6, TGFB2, VEGFA, and CXCL1) and three other genes (RELA, GFAP, and TNC) were shown to be significantly higher in the iERM eyes than in the MH control eyes. The upregulation of these genes may be associated with the iERM formation. Even though the number of females included in our study was 1/3 higher than the number of males, the genes shown to be upregulated did not significantly differ between them (S2 Table). Only RELA gene expression was shown to be positively correlated with age (S1 Fig). The relationship of advancing age and ERMs has been reported previously [2]. RELA is a subunit of the nuclear factor kappa B (NF-κB) heterodimer, which is a transcription factor that can be activated by various stimuli, such as hypoxia, bacterial and viral infections, and proinflammatory cytokines, including TNF-α and IL1β, and which plays significant roles in immune response, inflammation, apoptosis, and cellular proliferation [16,17]. NF-κB and IL8, an angiogenic factor, were found to be expressed in the glial and vascular endothelial cells, and their expression levels were associated with ERM formation [12]. This suggests that the advanced age may be associated with the upregulation of RELA in glial and endothelial cells, and, subsequently, ERM formation. Furthermore, positive correlations between TGFB2 and IL6, CXCL1 or RELA were determined. TGFB2 upregulation was reported in eyes with iERM, proliferative diabetic retinopathy (PDR), and proliferative vitreoretinopathy (PVR), and it is associated with intraocular fibrosis [5,7,13,15]. TGFB2 was shown to induce the transformation of the retinal pigment epithelial (RPE) cells, hyalocytes, and myofibroblastic cells [5,7,13]. Minchiotti et al. [8] showed that the expression of TGFB1 is elevated in ERM vitreous. However, the expression of TGFB2, but not TGFB1, was found to be higher in iERM samples than in MH samples in our study. Iannetti et al. [5] also reported that the correlation between TGFB2 expression and iERM is more significant than between TGFB1 expression and iERM development. Kohno et al. [7] showed a strong contractile activity of collagen gels and αSMA (ACTA1) overexpression in the hyalocytes in the presence of TGFB2. They also reported that iERM was composed of αSMAand GFAP-positive cells in the ERM foci and periphery, respectively. This suggests that ACTA1 (gene encoding αSMA) expression level may be too low to be possible to detect it by real-time PCR in the irrigation solution. IL6, a pro-inflammatory factor, was previously reported to be involved in the activation of Müller glial cells via microglia [18] and in the epiretinal formation in eyes, particularly those with PDR and PVR [19]. To the best of our knowledge, the correlation between CXCL1 expression and iERM in vivo has not been reported previously. However, CXCL1 expression may be associated with iERM formation as CXCL1 is involved in angiogenesis, inflammation, and wound healing process [20,21]. Iwona et al. [22] reported that RPE-stimulated Müller glial cells react to recombinant human CXCL1.
Additionally, we showed that the expression of T-cell cytokines and other T-cell associated factors, including IL17A, IFNG, RORC, and TBX21, did not correlate with iERM formation. Further investigations are needed to elucidate iERM pathogenesis, as the low number of subjects enrolled and a significant difference in the age of the subjects between two groups in this study represent the limitations of this study. Additionally, protein expression and the activation of signaling pathways should be further analyzed as well.
Three genes (IL6, CXCL1, and RELA) were positively correlated with TGFB2, while other three genes (VEGFA, GFAP, and TNC) were themselves correlated. Based on these findings, we hypothesized that TGFB2 and three factors (IL6, CXCL1, and NF-κB) might stimulate GFAPpositive cells related to fibrosis, and angiogenic factors (VEGFA and Tenascin C) can augment iERM formation.

Conclusions
We evaluated gene expression levels in the solid component of the irrigation solution sample collected during pars plana vitrectomies of eyes with MH or iERM. Previous studies utilized ELISA, immunoblotting, or traditional RT-PCR to compare the level of several cytokines and other factors in the vitreous humor or membranes of patients with ERM [5][6][7][8][9]. To the best of our knowledge, this is the first time real-time PCR analysis has been employed to assess the expression levels of cytokines and/or other genes in the irrigation solution samples collected during vitrectomy.
The factors associated with angiogenesis and wound healing processes were more upregurated in the eyes with iERM than with MH, and the factors associated with T-cell did not seem to affect ERM formation. This study may help elucidate the pathogenesis of iERM, but it may also represent a foundation for future studies of cytokines and other molecules involved in the development of secondary ERM.