Mycoplasma Ocular Infection in Subretinal Graft Transplantation of iPS Cells-Derived Retinal Pigment Epithelial Cells

Abstract

Purpose: To report occurrence of acute severe inflammation after surgical implantation of mycoplasma-infected induced pluripotent stem cell–derived RPE (iPS-RPE) cells into the eyes of healthy primates, and determine the immunopathological mechanisms of the inflammation.

Methods: Ophthalmic allogeneic transplantation of iPS-RPE cells was performed in the subretina of major histocompatibility complex (MHC)-matched (two eyes) and MHC-mismatched (one eye) healthy cynomolgus monkeys. The clinical course after transplantation was observed using color fundus photography, fluorescence angiography, and optical coherence tomography. After the animals were killed at 1 month after surgery, eyeballs were removed and pathologically examined. Microorganisms were analyzed by PCR methods and BLAST analysis using preserved graft iPS-RPE cells and the recipients' vitreous humor. Mixed lymphocyte-RPE assay was performed on the mycoplasma-infected and noninfected iPS-RPE cells in vitro.

Results: In tested eyes, abnormal findings were observed in the grafted retina 2 weeks after surgery. Here, we observed retinal vasculitis and hemorrhage, retinal detachment, and infiltration of inflammatory cells into the retina of the eyes. One month after surgery, animals were killed due to the severe immune responses observed. Using PCR methods, sequence analysis detected mycoplasma-DNA (Mycoplasma arginini species) in both the grafted RPE cells and the collected vitreous fluids of the monkeys. Mixed lymphocyte-RPE assay revealed that the infected iPS-RPE cells enhanced the proliferation of inflammatory cells in vitro.

Conclusions: Transplantation of graft iPS-RPE cells contaminated with mycoplasma into the subretina caused severe ocular inflammation. Mycoplasma possesses the ability to cause immune responses in the host.

Recent developments in the field of regenerative medicine have made it possible to transplant cultured cells into patients for the treatment of diseases such as retinal disorders; however, to achieve a successful transplantation, it is necessary to be able to detect the presence of infectious agents. In a previous study, we reported on examination methods that can be used to study postoperative endophthalmitis caused by bacteria species¹; however, infectious endophthalmitis caused by microorganisms other than bacteria or fungal species has yet to be examined in detail. It has also been previously reported that ocular inflammation, such as endophthalmitis and uveitis, occurs in a rodent model in which mycoplasma is directly administered into the eye.^2,3 These mycoplasma animal models were shown to exhibit severe immune responses in the eye. On the other hand, ocular inflammation associated with Mycoplasma pneumoniae infections in humans has been reported.^4–6 In several other cases, mycoplasmas have been suggested to have the ability to cause ocular inflammation. One of the remarkable infection routes reported for mycoplasma involves transplantation surgery in which mycoplasma from a donor ends up causing an infection in the transplant recipient. In fact, there have been reports documenting the transfer of mycoplasma from the donor via a transplanted heart or lung^7,8 and blood vessels⁹ that ultimately caused pleurisy, surgical site infection, and sepsis. Moreover, it is also possible that cultured cells often can be infected with mycoplasma. Another study reported that mycoplasma infection altered the gene expression of many inflammatory cytokines.¹⁰ However, as far as we know, there have yet to be any reports that have shown that mycoplasma can cause severe ocular inflammation by directly invading the eye during intraocular surgery. 

In the present study, we detected mycoplasma in cases of severe ocular inflammation that occurred after induced pluripotent stem cell–derived RPE (iPS-RPE) cell transplant surgery in the vitreous of the recipient and in the stock of the grafted iPS-RPE cells used in the transplant. To verify the presence of mycoplasma in the eye, we conducted the following evaluations: detection of the mycoplasma genome by performing molecular biological analysis using vitreous fluids and transplanted RPE cells, and performed detailed clinical and pathological examinations of inflamed eyes to elucidate the pattern of inflammation. In addition, to elucidate the immune responses of the inflammatory cells against mycoplasma, we also performed the mixed lymphocyte-RPE assay using iPS-RPE cells infected with mycoplasma and recipient blood cells. 

Materials and Methods

Preparation of Monkey iPS-RPE Cells

We prepared iPS cells (iPSCs) from normal cynomolgus monkeys (Macaca fascicularis), 1121A1 iPSCs from the HT-1 major histocompatibility complex (MHC) homozygote monkey, and 46a iPSCs from the Cyn46 MHC heterozygote monkey.¹¹ The monkey iPS-RPE cells were established from the iPSCs as has been previously described.¹¹ All of the animal experiments were approved by the RIKEN BDR Animal Experiment Committee. The care and maintenance of the monkeys conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and the Use of Laboratory Animals, as well as to the Guidelines of the RIKEN BDR Animal Experiment Committee. 

Transplantation of iPS-RPE Cells into the Subretinal Space of Monkeys

MHC-controlled monkeys (DrpZ11 and 12: adult cynomolgus monkeys; Ina Research, Nagano, Japan) and a normal control cynomolgus monkey (TLHM-6) were used in this study. For transplantation of the iPS-RPE cell suspension (MHC homozygote 1121A1 iPS-RPE cells or MHC heterozygote 46a iPS-RPE cells), we injected the subretinal space with 400 μL of the iPS-RPE cell suspension (2.4 × 10⁶/mL) as per our previous reports.^11,12 RPE cells were stained with fluorescent dye PKH (PKH26GL; fluorescent at 567 nm; Sigma-Aldrich, St. Louis, MO, USA) to trace the cells after the transplantation. The graft cells were monitored by color fundus photographs, fluorescence angiography (FA) with RetCamII (Clarity Medical Systems, Pleasanton, CA, USA), and optical coherence tomography (OCT) (Nidek, Aichi, Japan) at 1, 2, and 4 weeks after the surgery. To avoid visual loss in the subject animal, the transplantation site was positioned out of the macular region. To avoid infectious endophthalmitis related to the surgery, preoperative ocular disinfection treatment was performed. In all tested primates, transplantation of the fellow eye was carried out after confirming that there were no complications in the first eye at 1 week after surgery. For example, because there were no problems found during the medical examination of the right eye of the DrpZ12 monkey at 1 week after the initial surgery, we then performed left eye surgery. Subsequently, however, inflammation unexpectedly appeared in both eyes (at approximately 2 weeks). Although before the monkey was killed there were no changes in the behavior of the monkey regarding daily routines (i.e., feeding and drinking water), the animal was killed after taking into consideration the possibility of binocular visual impairment and the importance of the overall investigation. 

MHC Typing

Genotyping of MHC-1 and MHC-II genes in cynomolgus monkeys was performed by pyrosequencing as previously described.¹³ MHC information for the 1121A1 MHC homozygote iPS-RPE cells, Cyn 46a RPE cells (MHC heterozygote: control), and the TLHM-6 monkey (MHC-mismatched monkey) has been described in our previously published report.¹¹ Supplementary Table S1 shows the MHC profiles of the MHC-matched monkeys (DrpZ11 and DrpZ12). 

PCR and Basic Local Alignment Search Tool (BLAST) Analysis

DNA was extracted from the samples using a DNA Mini Kit (Qiagen, Valencia, CA, USA). Genomic DNA of bacteria, fungi, and mycoplasma in the iPS-RPE cells and vitreous fluids was measured using real-time quantitative PCR assays. PCR was performed using a LightCycler 480 II instrument (Roche, Basel, Switzerland). The primers and probes (targeting the DNA of the region encoding the 16S ribosomal RNA (rRNA) (16S ribosomal DNA [rDNA]) used in this study for detection of mycoplasma-DNA were purchased from Nihon Techno Service Co., Ltd. (Tokyo, Japan). These samples were used for the quantitative PCR analysis. The Hokkaido System Sciences Co., Ltd., was contracted to perform the BLAST analysis. DNA was sequenced and aligned with data available from the GenBank at the National Institutes of Health with BLAST, a computer alignment program. After the alignment results for the 16s rRNA gene were matched to the database, the homology analysis was then performed. 

Detection of Anti-Mycoplasma Antibody in the Serum From Mycoplasma-Infected Monkey

Sera (n = 2) of the DrpZ12 monkey that was transplanted with mycoplasma-infected iPS-RPE cells was collected. The first sample was obtained before surgery, and the other was collected at 4 weeks after the transplantation. Cultured monkey iPS-RPE cells (1121A1) that were infected with mycoplasma and primary monkey RPE without infection (1 × 10⁴ cells/well) were recultured in a 96-well culture plate. As per our previous report,¹² RPE cells were incubated with the serum (×50 with PBS) overnight at 4°C. The cells were then incubated with 4′,6-diamidino-2-phenylindole (DAPI; ×1000; Invitrogen, Carlsbad, CA, USA) and a secondary antibody Alexa Fluor 488 anti-human IgG (×2000; Invitrogen) for 1 hour at room temperature. Primary RPE cells were used as the control. Images were acquired with a confocal microscope (LSM700; Zeiss, Oberkochen, Germany). Relative fluorescence intensity was analyzed using image analysis software (ZEN; Zeiss). At least three independent experiments were performed for the in vitro data. To test the fluorescence intensity difference, statistical analyses were performed using the paired Student's t-test. Values were considered statistically significant if P < 0.05. 

Mixed Lymphocyte-RPE Assay With iPS-RPE Cells and Blood Cells, and Flow Cytometry

Peripheral blood mononuclear cells (PBMCs) were isolated from a healthy adult MHC control monkey donor (DrpZ11), with the allogeneic immune responses assessed for the proliferation by Ki-67 incorporation in the PBMC. PBMCs were cultured with MHC-matched 1121A1 iPS-RPE cells (mycoplasma infected or not). The culture medium used was RPMI-1640 medium containing 10% fetal bovine serum (LONZA, Basel, Switzerland), human recombinant IL-2 (BD Biosciences, San Jose, CA, USA), 10 mM HEPES (Sigma-Alrich), 0.1 mM nonessential amino acids (Sigma-Alrich), 1 mM sodium pyruvate (Sigma-Alrich), penicillin-streptomycin (Thermo Fisher Scientific, Waltham, MA, USA), and 1 × 10⁻⁵ M 2-mercaptoethanol (Sigma-Alrich). Before the assay, the RPE cells were irradiated (20 Gy). After 96 to 120 hours, PBMCs were analyzed by flow cytometry (Ki-67 proliferation assay by fluorescence-activated cell sorting [FACS]).¹¹ For the Ki-67 proliferation assay by FACS analysis, the following antibodies were prepared: APC-labeled anti-CD4 (helper T cells, no. 130-098-133; Miltenyi Biotech, Auburn, CA, USA), APC-labeled anti-CD8 (cytotoxic T cells, no. 17-0088; eBioscience, San Diego, CA, USA), APC-labeled anti-CD11b (macrophages/monocytes, no. 130-091-241; Miltenyi Biotech), FITC-labeled anti-CD20 (B cells, no. 130-091-108; Miltenyi Biotech), APC-labeled anti-NKG2A (natural killer [NK] cells, no. 130-098-812; Miltenyi Biotech), and phycoerythrin (PE)-labeled anti-Ki-67 (no. 350504; BioLegend, San Diego, CA, USA). The harvested PBMCs were stained with the above antibodies at 4°C for 30 minutes. The intracellular staining for Ki-67 was performed after cell fixation and permeabilization (BioLegend). All samples were analyzed on a FACSCanto II Flow Cytometer (BD Biosciences). Data were analyzed using FlowJo software (version 9.3.1; Tree Star, Ashland, OR, USA). 

Immunohistochemistry (IHC)

Monkey eyes collected at 1 month were fixed and embedded in paraffin (Sigma-Aldrich). Paraffin sections were sliced into 10-μm-thick sections. Detailed information on the procedure has been presented in our previous reports.^11,12 The same immunochemical techniques and photographing methods, as mentioned below, were applied to all sections. Additional primary antibodies against the following proteins were used: ionized calcium-binding adapter molecule 1 (Iba1) (host: rabbit, ×1000, no. 019-19741; Wako, Osaka, Japan), CD3 (host: rabbit, ×100, no. ab16669; Abcam, Cambridge, UK), MHC-II (host: mouse, ×100, no. Nr.M0775; Dako Cytomation, Santa Clara, CA, USA), Ly6G (host: rat ×100, no. ab25024; Abcam), NKG2A (host: rabbit, ×100, no. ab93169; Abcam), and Alexa Fluor 488 anti-Human IgG (host: goat, ×2000, no. A11013; Invitrogen). All sections were incubated at 4°C overnight with the pertinent primary antibodies.^11,12 Images were acquired with a confocal microscope (LSM700, Zeiss; http://www.zeiss.com). 

Measurements of Cytokines

Vitreous fluids (160 to 200 μL) were collected from the transplanted left eye of the monkey (TLHM-6) at the time of transplantation (0 weeks) and at 1, 2, and 4 weeks after transplantation. Cytokine array experiments on the vitreous fluids were conducted at Filgen Incorporated (Nagoya, Japan) using the Monkey Cytokine Magnetic 29-Plex Panel (Thermo Fisher Scientific, Waltham, MA, USA). The measurement proteins were epidermal growth factor (EGF), eotaxin, FGF-basic, granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), hepatocyte growth factor (HGF), IFN-γ, IL-1β, IL-1RA, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12, IL-15, IL-17, interferon-inducible T-cell alpha chemoattractant (I-TAC), monocyte chemotactic protein-1 (MCP-1), macrophage-derived chemokine (MDC), macrophage migration inhibitor factor (MIF), monokine induced by interferon-γ (MIG), macrophage inflammatory protein 1 (MIP)-α, MIP-β, regulated on activation, normal T-cell expressed and secreted (RANTES), TNF-α, VEGF, and gamma interferon inducible protein 10 (IP-10). 

Results

Case 1: Severe Ocular Inflammation in MHC-Matched iPS-RPE Cell Transplantation

In the first step, we transplanted MHC homozygous iPS-RPE cells (1121A1 lines: cell suspension) into an MHC heterozygote, which in this case was the MHC-matched monkey (right eye of the DrpZ12 monkey). All clinical symptoms were followed after the initial transplantation. At 2 weeks after the transplantation, a whitish infiltrating mass was found in the subretinal space at the site of the graft (Fig. 1A). FA images clearly showed that there was fluorescence leakage in the retinal vein on the bleb of the grafted site (Fig. 1B). OCT evaluations indicated that there were deposits under the retina and subretinal fluids (Fig. 1C). At 4 weeks after the transplantation, vitreous opacity caused the fundus to be invisible (Supplementary Fig. S1A). At postoperative day 33, fibrin was seen in the anterior chamber and iris rubeosis was also observed (Supplementary Fig. S1B). 

Figure 1

 

Inflammation after allogeneic transplantation of MHC homozygote iPS-RPE cells into the subretinal space of the right eye of the MHC-matched monkey. Without using immunosuppression, we transplanted monkey 1121A1 iPS-RPE cells (5 × 10⁵ cells with single-cell suspension) into the subretinal space in the DrpZ12 MHC-matched monkey. (A) At 2 weeks (2W) after surgery, the fundus color photograph revealed a white subretinal mass infiltrating at the site of the graft (arrow). (B) FA (arrow) revealed the leakage from the retinal vein at the grafted site. (C) OCT showed the presence of cell infiltration (arrow) in the subretinal space of the grafted site. Presence of subretinal fluid was also seen (arrowhead). (D) Retinal hemorrhages were observed in split eyeballs extracted from the DrpZ12 monkey. Scale bar: 1 cm. (E) H&E staining for histological interpretation in the right eye of the DrpZ12. Retinal edema (arrowhead) and intraretinal and subretinal hemorrhages (arrow) were observed in the macula. (F) Infiltration of inflammatory cells in the subretinal space (arrow) and thickened choroid were observed in the transplanted area. Scale bar: 200 μm. (G) The graft iPS-RPE cells were stained with PKH to trace the cells after transplantation. Photomicrographs of the specimens collected from the paraffin sections show labeling of infiltrating cells in the DrpZ12 monkey retina in the right eye. The markers used included Iba1 (microglia/macrophage marker), MHC class II (MHC-II) (antigen-presenting cell marker), CD3 (T-cell marker), Ly6G (neutrophil marker), NKG2A (NK-cell marker), and IgG (antibody and B-cell marker). Many infiltrating cells were observed, including Iba1⁺ cells, MHC-II⁺ cells, CD3⁺ cells, Ly6G⁺ cells, and NKG2A⁺ cells, and there were IgG deposits around the PKH-positive iPS-RPE cell graft. Scale bar: 40 μm.

Although the vitreous and retina in the eyeball of the DrpZ12 monkey were clouded due to vitreous hemorrhage (Supplementary Fig. S1C), the retina and vitreous were transparent in the control monkey eye (Supplementary Fig. S1D). After removal of the vitreous, retinal hemorrhage was observed in the retina (Fig. 1D). 

To examine inflammation by IHC, we conducted hematoxylin and eosin (H&E) staining and immune staining of inflammatory cells in the retinal sections. H&E staining revealed the presence of hemorrhagic retinal detachment and retinal hemorrhage, which suggested a retinal circulation disorder (Fig. 1E). Results also showed there was a large amount of inflammatory cell infiltration centered on the site of the graft along with choroidal vasodilation (Fig. 1F). Immune staining additionally revealed large infiltrations of Iba1⁺ cells, MHC class II⁺ cells, CD3⁺ cells, Ly6G⁺ cells, and NKG2A⁺ cells on the site of the graft (Fig. 1G). In addition, deposits of IgG were also noted at the graft site (Fig. 1G). The left eye, which underwent the transplantation at 1 week after the right eye, also exhibited similar findings, thereby indicating that the transplanted eye also had severe ocular inflammation (Supplementary Fig. S2). 

Case 2: Severe Ocular Inflammation in MHC-Mismatched iPS-RPE Cell Transplantation

We also observed a case of severe inflammation at the transplanted site after MHC-mismatched iPS-RPE transplantation in a primate. Although remarkable changes were observed in the color fundus photographs at 2 weeks after the transplantation (Fig. 2A), FA showed there were fluorescence leakages in the grafted area and the macula (Fig. 2B). A high signal deposit under the retina was observed in the OCT images (Fig. 2C). H&E staining showed that there were infiltrating cells in the subretinal space of the transplanted area. Similar to the first case, we also found choroidal vasodilation (Fig. 2D). IHC of the infiltrating cells collected at the grafted site revealed infiltration of Iba1⁺ cells, CD3⁺ cells, MHC class II⁺ cells, Ly6G⁺ cells, IgG⁺ tissues, and NKG2A⁺ cells (Fig. 2E). 

Figure 2

 

Inflammation after MHC-mismatched allogeneic transplantation in the normal monkey. Monkey 46a iPS-RPE cells (5 × 10⁵, single-cell suspension) were transplanted into the subretinal space of a normal monkey (TLMH-6) without using immunosuppression. (A) At 2 weeks after surgery of the left eye, the results of fundus color photo revealed no obvious abnormality at the grafted site or macula area (arrow). (B) FA revealed leakages from the subretinal space at the grafted site and macula (arrow). (C) OCT showed the presence of cell infiltration in the subretinal space of the grafted site (white arrow). (D) TLHM-6 was killed at 33 days after transplantation. Infiltration of inflammatory cells in the subretinal space and a thickened choroid were observed within the grafted site. Scale bar: 100 μm. (E) Photomicrographs show immune staining of the infiltrating inflammatory cells in the grafted site. There were many infiltrating cells observed around the PKH⁺ iPS-RPE cell grafts, including Iba1⁺ cells, CD3⁺ cells, MHC-II⁺ cells, IgG, NKG2A⁺ cells, and Ly6G⁺ cells. Scale bar: 40 μm.

Subsequently, we then investigated the differences in inflammation between the mycoplasma infection and posttransplant immune rejections after IPS-RPE cell transplantation (without infection). We used IHC to examine the anti-Ly6G (neutrophil) and anti-NKG2A (NK cells) in the retinal sections of a monkey that had immune attacks after iPS-RPE transplantation and in a normal control monkey. In the RPE-related rejection retina, although infiltrating cells were primarily observed along the grafts, Ly6G⁺ and NKG2A⁺ cells were not found in these areas (Fig. 3A). In addition, there were also no Ly6G⁺ and NKG2A⁺ cells observed in the normal control retina (Fig. 3B). 

Figure 3

 

Immunohistochemistry of the eye in the iPS-RPE cell–related immune rejection and negative control monkeys. (A) Photomicrographs showing labeling of the K247 monkey retina (at 4 months after surgery) with Ly6G and NKG2A. There was an inflammatory nodule (arrow) due to rejection along with many infiltrating cells in the retina. However, Ly6G⁺ cells (left: neutrophils) and NKG2A⁺ cells (right: NK cells) were not seen. INL, inner nuclear layer; ONL, outer nuclear layer. (B) Although the DrpZ10 monkey underwent a vitrectomy, we only injected medium without RPE cells in this animal. Photomicrographs show labeling of the DrpZ10 monkey retina with Ly6G and NKG2A. We failed to find either Ly6G⁺ cells (left) or NKG2A⁺ cells (right). Scale bar: 40 μm.

Detection of Mycoplasma Genomic DNA From Donor iPS-RPE Cells and Recipient Vitreous Samples

Because ocular inflammation was fulminant as compared with the RPE cell–related rejection that we observed in our previous studies,^11,12 we suspected this inflammation was due to an infection caused by microorganisms. Quantitative PCR has been previously performed using primers and probes for detecting bacterial 16S rDNA,¹ fungal 28S rDNA,¹⁴ and mycoplasma species. In the present study, we prepared DNA from stocks of transplanted iPS-RPE cells (1121A1 or 46a lines). We detected 2.27 × 10¹⁰ copies/μg·DNA of mycoplasma-DNA from the 1121A1 iPS-RPE cells, whereas 9.82 × 10⁹ copies/μg·DNA were detected from the 46a iPS-RPE cells. We also demonstrated that bacteria 16S rDNA of the mycoplasma species was positive, whereas 16S rDNA of the other bacteria species and fungi 28S rDNA were not detected (data not shown). Interestingly, there was no obvious difference in the optical microscopic findings between the mycoplasma-noninfected and infected iPS-RPE cells (Fig. 4A). We also performed quantitative PCR of vitreous fluids, and detected 7.37 × 10⁴ copies/mL of mycoplasma-DNA in samples from the right eye at 2 weeks after transplantation and 1.80 × 10⁴ copies/mL from the left eye at 1 week after transplantation (Fig. 4B). Homology analysis using the BLAST database for the 16S rRNA gene in the infected iPS-RPE cells detected a homology of 99.276 to 99.783% with the strain of Mycoplasma arginini (Table 1). In addition, vitreous fluids from the recipient monkeys also exhibited similar results (Table 1). 

Figure 4

 

Detection of mycoplasma genome and anti-mycoplasma antibodies in mycoplasma-infected iPS-RPE cell transplantation. (A) Optical microscopic findings of monkey iPS-RPE cells (1121A1 lines). There were no differences noted in the microscopic image between the mycoplasma-noninfected (upper) and infected (lower) RPE cells. Scale bar: 100 μm. (B) We performed quantitative PCR to detect the mycoplasma-DNA in the vitreous collected from the DrpZ12 monkey. Quantitative levels of mycoplasma-DNA in the vitreous of the right eye (at 2 weeks after transplantation) were 7.37 × 10⁴ copies/mL, and 1.80 × 10⁴ copies/mL for the left eye (at 1 week after transplantation). (C) To detect anti-mycoplasma antibody from the infected monkey's serum, IHC was performed on the infected iPS-RPE cells and control primary monkey RPE cells using the anti-IgG antibody. RPE cells were incubated with the DrpZ12 serum collected before 0 weeks (0W) and at 4 weeks (4W) after transplantation. Monkey primary RPE cells with no infection that were stained with the DrpZ12 serum at 0 and 4 weeks showed low intensity for the IgG staining, with the cell nuclei clearly stained with DAPI. Monkey iPS-RPE cells infected with mycoplasma incubated with DrpZ12 serum showed high intensity for the IgG staining. Nonspecific staining between and around the nucleus of RPE cells was observed in addition to cell nuclei staining with DAPI. Scale bar: 50 μm. (D) The graphs show the mean fluorescence intensity of the IgG staining. Open bars show the intensity of the primary RPE cells. Black bars show the intensity of the iPS-RPE cells with the mycoplasma infection. Fluorescence intensity of iPS-RPE cells with the mycoplasma infection incubated with the serum at 4 weeks was significantly higher than that observed for the control serum (0 weeks, P = 0.021).

Table 1

 

Results of BLAST Analysis of Bacterial 16S rRNA Genes

 

Detection of Mycoplasma-Specific Antibodies in Recipient Sera

To demonstrate that mycoplasma-specific antibody was produced in the recipient by the mycoplasma infection, in the next step, we conducted immune staining of mycoplasma-infected and noninfected iPS-RPE cells using sera. The sera were collected before transplantation and at 4 weeks after transplantation from the recipient monkeys (DrpZ12). Nuclei of the noninfected cells were clearly stained with DAPI. In infected cells, in addition to the cell nuclei, nonspecific staining between and around the nuclei of RPE cells was also observed (Fig. 4C). Immunocytochemistry with fluorescently labeled anti-IgG antibody tended to show that the intensity of fluorescence in the infected iPS-RPE cells was much higher than that for the noninfected cells. In the infected iPS-RPE cells, fluorescent intensity for the serum at 4 weeks was significantly higher than that seen for the baseline serum (P = 0.021; Fig. 4D). Thus, we suspected that mycoplasma-specific antibody might be present in the serum following the initial transplantation. 

Inflammatory Cells Respond to Mycoplasma-Infected iPS-RPE Cells In Vitro

To investigate whether mycoplasma stimulates the immune responses, we analyzed PBMCs by the mixed lymphocyte-RPE assay. When PBMCs from the DrpZ11 MHC-matched monkey were co-cultured with MHC-matched 1121A1 iPS-RPE cells, there was clear suppression of the proliferation as compared with that observed for the culture of the control PBMCs alone (Fig. 5A). On the other hand, when PBMCs from the DrpZ11 monkey were co-cultured with mycoplasma-infected 1121A1 iPS-RPE cells, there was clear enhancement of the proliferation as compared with the cultures with PBMCs only (i.e., all types of the inflammatory cells responded against the RPE cells in vitro) (Fig. 5B). 

Figure 5

 

Mixed lymphocyte-RPE assay with fresh PBMC plus iPS-RPE cells. In the mixed lymphocyte-RPE assay with allogeneic 1121A1 iPS-RPE cells, PBMCs (2 × 10⁶ cells/well in the DrpZ11 MHC-matched monkey) were cultured with allogeneic iPS-RPE cells for 5 days. Before the assay, iPS-RPE cells were irradiated with 20 Gy, with 1 × 10⁴ cells then cultured in a 24-well plate. (A) Mycoplasma-infected iPS-RPE cells, (B) mycoplasma-noninfected iPS-RPE cells. Harvested PBMCs were stained with anti-CD4, anti-CD8, anti-CD11b, anti-CD20, anti-NKG2A, anti-Ki-67, with each isotype control antibody at 4°C for 30 minutes. The samples were analyzed on a FACS flow cytometer. Numbers (%) in the scatterplots indicate double-positive cells.

Increased Various Inflammatory Cytokines and Chemokines From Vitreous in Mycoplasma-Infected iPS-RPE Cell Transplantation

To examine whether ocular fluids in the mycoplasma-infected iPS-RPE cell transplantation monkey (TLHM-6) contain inflammatory cytokines and chemokines, we collected vitreous fluids at 0, 1, 2, and 4 weeks after transplantation. After collection, we then measured the inflammatory proteins using the cytokine beads array. Among the 29 proteins tested, significant increases were observed in the IL-1β, IL-1RA, IL-6, IL-12, IL-15, IFN-γ, MIF, eotaxin, IP-10, I-TAC, MCP-1, MDC, MIG, RANTES, and VEGF (Table 2). As compared with the vitreous data obtained before surgery (0 weeks), there was a significant increase in the Th1-related cytokines (IL-12, IL-15, and IFN-γ) and Th1-related chemokines (IP-10, I-TAC, MIG, and RANTES) in the mycoplasma-infected vitreous samples, especially at 1 or 2 weeks after transplantation (Table 2). These results indicated that various inflammatory cytokines/chemokines presented in the retina and vitreous (similar to a “cytokine storm” in the eye). 

Table 2

 

Measurements of Cytokines and Chemokines From Vitreous Fluids in Mycoplasma-Infected iPS-RPE Cell Transplantation Monkey

 

Taken together, the in vivo and in vitro results demonstrated that mycoplasma-infected iPS-RPE cells can stimulate immune responses, thereby causing severe inflammation in the recipient eye after transplantation. 

Discussion

Results of the present study showed that severe inflammation mainly occurred within the grafted area, especially in the subretinal space and choroid, with the inflammation then spreading throughout the eyeballs. These results suggested that the mycoplasma that directly invades the eye may cause strong immune responses. Moreover, the mixed lymphocyte-RPE assay showed that MHC-matched iPS-RPE cells suppressed proliferation of inflammatory cells, whereas RPE cells infected with mycoplasma enhanced the proliferation of inflammatory cells in vitro. These results demonstrate that mycoplasma can cause inflammatory reactions. 

We performed quantitative PCR to try to detect Mycoplasma spp. DNA collected from ocular samples and explanted RPE cells. In addition, we also tried to identify the species of the mycoplasma using homology analysis of the amplified 16S rDNA with the BLAST database. We detected more than a 99% homology with the strain of M. arginini. It has been previously reported that a 98.7% homology to 16S rDNA corresponded to the criterion of the 70% of the DNA-DNA hybrid, which has been considered to be the conventional standard of judgment for species.¹⁵ Based on these findings, the mycoplasma that infected the iPS-RPE cells and inflamed eyeballs was judged to be M. arginini. It has been previously reported that M. arginini is one of major species that causes contamination of cell cultures.¹⁶ Furthermore, it has been suggested that PCR can be useful for the early diagnosis of Mycoplasma hominis infections that occur after cardiac chest transplantations.⁸ Because these infections are known to progress with the passage of time, a rapid diagnosis is important to be able to provide an effective treatment. In some species, however, it often can be difficult to successfully perform cultures. Moreover, we also showed that the optical microscopic findings indicated that there was no obvious change in the iPS-RPE cells infected with M. arginini. Thus, when trying to detect microorganisms in a clinic, rapid diagnostic tests, such as PCR, are necessary. 

As shown in the present study, mycoplasma-infected cells caused intensive ocular inflammation when the cells were placed in the subretinal space of the recipient's eye. Initial signs included retinal vasculitis and subretinal fluorescence leakages on FA examination, whereas OCT revealed there was subretinal inflammatory cell infiltration at the grafted area. In addition, we observed retinal hemorrhage and iris rubeosis, which indicates the presence of an impaired retinal circulation. Furthermore, these severe inflammatory findings suggested that intraocular invasion by mycoplasma might lead to a poor prognosis for visual function. Ocular inflammation also can develop in humans in association with M. pneumoniae infection. Several cases also have been reported in patients found to have edema of the optic papilla or anterior uveitis.⁴ There are also studies that have reported finding retinal inflammation and circulatory disorders, including a case with panuveitis accompanied by retinal hemorrhage,⁶ and a case with frosted branch angiitis and macular edema.⁵ The inflammation and hemorrhage in both cases were mild, with no decrease in the final visual acuity observed. Recently, Narita¹⁷ classified the pathology of mycoplasma infection into three categories that included a direct type in which the mycoplasma directly invades the lesion part, a vascular occlusion type due to vasculitis, and an indirect type such as an autoimmune reaction. Narita¹⁷ considered that these three elements overlapped to form the overall disease state. In human uveitis cases, the mycoplasma antigen has yet to be detected in any eyes. Thus, the pathology of the mycoplasma-related ocular inflammation cases in humans may reflect the indirect type. In a rodent model in which mycoplasma was directly administered intraocularly, the animals exhibited severe endophthalmitis-like symptoms such as choroidal edema and exudative retinal detachment.^2,3 Because both our present cases and this rodent mycoplasma model exhibited severe ocular inflammation, this suggests that both of these can be classified as examples of direct-type infections. 

The entire genome of M. arginini, which was detected in our present cases, has been previously analyzed. Overall, it is thought that the virulence of M. arginini in living organisms is low, as it lacks the capsular synthetic genes and active oxygen production genes that are considered to be the cause of pathogenicity for mycoplasma.¹⁸ However, the findings for our present cases revealed that mycoplasma caused severe intraocular inflammation and circulatory disorder by means other than a capsule and active oxygen. Moreover, our in vitro experiments additionally revealed that mycoplasma has the ability to stimulate inflammatory cells. In the mixed lymphocyte-RPE assay, noninfected iPS-RPE cells suppressed proliferation of inflammatory cells such as CD4⁺ T helper cells, CD8⁺ cytotoxic T cells, CD11b⁺ monocytes/microglia, CD20⁺ B cells, and NKG2A⁺ NK cells. Conversely, in the mycoplasma-infected iPS-RPE cells, the proliferation of these inflammatory cells was greatly enhanced. It has been shown that M. arginini both stimulates human PBMC to produce IL-1β, IL-6, and TNF-α¹⁹ and causes mesenchymal stem cells to produce complement factors.²⁰ Mycoplasma does not produce toxins, unlike other bacteria, but the cell membrane lipoproteins of mycoplasma can cause an immune response via TLR2, 4 and autophagy.²¹ The findings of these previous reports support the concept that mycoplasma directly stimulates leukocytes, thereby initiating the inflammatory responses by a mechanism that differs from the other bacteria. In contrast, RPE infected with chlamydia has been reported to be able to upregulate secretion of IL-8,²² IL-6, and VEGF.²³ Thus, mycoplasma infection might also have influenced the cytokine production of RPE cells, thereby enhancing the ocular inflammation and neovascularization in our current study cases. 

In our present cases, infiltration of inflammatory cells including Ly-6G⁺ cells and NKG2A⁺ cells were observed in the grafted area. In the retina of the posttransplant immune rejections, although Iba1⁺ cells, MHC class II⁺ cells, and CD3⁺ T cells invaded the grafted area,¹¹ Ly6G⁺ cells and NKG2A⁺ cells were not found. Infiltration of neutrophils is known to be a characteristic pathological finding in lungs with M. pneumoniae pneumonia.²⁴ Resistance to mycoplasma has been shown to be mediated by activated NK cells.²⁵ The presence or absence of infiltrating Ly6G⁺ cells and NKG2A⁺ cells may reflect the difference in the pathophysiology between mycoplasma infection and immune rejection. 

The importance of taking precautionary measures to prevent mycoplasma contamination into cells that will be used for transplantation has been recognized, with mycoplasma testing now defined within the pharmacopoeia of several countries.^26,27 For human retinal cell transplantations, multiple sterility tests for mycoplasma have been recommended.²⁸ However, the possibility still exists that mycoplasma ocular infection could potentially occur by accident. Thus, the importance of our current research is that it provides additional knowledge that is necessary for diagnosing cases of mycoplasma ocular infection. 

In the present study, although we did not consider using any treatments in the experimental animals, as the monkeys were to be killed, it is likely that they probably would have been unable to see due to the severe ocular inflammation. Protocols using fluoroquinolones, tetracyclines, and macrolides have been proposed for removing mycoplasmas that have infected cultured cells.^29,30 Although quinolones, macrolides, and tetracyclines are conventionally considered to be effective for mycoplasma infections in humans, the proportion of mycoplasmas that are resistant to quinolones and macrolides is increasing.^31,32 Therefore, cases that are difficult to treat are becoming more of a problem. There have been a few reports regarding the use of doxycycline to treat M. arginini–infected humans.^33,34 However, presently the number of cases reported is small and as of yet, the drug resistance remains unknown. To our knowledge, there have yet to be any reports on the efficacy and safety of these antibiotics in the intraocular infections caused by mycoplasma. Furthermore, verification using an infected animal model is also necessary to develop a treatment protocol. Because transplantation using MHC-matched primates can be used to help avoid inflammation and potential immune rejections,¹¹ this model is considered to be useful for studies on infection, as this may make it possible to specifically analyze the mechanism responsible for inflammation due to infection. 

This is the first study to show that mycoplasma-infected explanted cells are able to cause severe ocular inflammation. Circulatory insufficiency caused by inflammation and thrombosis, followed by angiogenesis, might be associated with the pathological mechanisms responsible for mycoplasma intraocular inflammation. In fact, the present study demonstrated the presence of retinal and vitreous hemorrhages in an eye in conjunction with a large number of inflammatory cells. Therefore, the present results highlight the potential problem of infections that might be associated with transplantations, such as during regenerative medicine–associated cell-based therapy. 

Acknowledgments

The authors thank Naoko Hayashi, Kyoko Iseki, Shoko Fujino, and Masayuki Kawahara (Laboratory for Retinal Regeneration, RIKEN BDR, Kobe, Japan) for their expert technical assistance. 

Supported by the Research Center Network for Realization of Regenerative Medicine from the Japan Agency for Medical Research and Development (AMED) and by the RIKEN Junior Research Associate Program. This work was also supported by a Scientific Research Grant (B, 18H02959) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. 

Disclosure: K. Makabe, None; S. Sugita, None; A. Hono, None; H. Kamao, None; M. Takahashi, None