Characterization of Bone Marrow–Derived Cells in the Substantia Propria of the Human Conjunctiva

Abstract

purpose. To characterize the main population of bone marrow–derived cells (BMCs) in human normal subconjunctiva and make a comparison with BMCs in the corneal stroma and epithelium.

methods. Normal human donor corneas with attached conjunctiva were examined by fluorescence microscopy after single and double staining for multiple markers. CD68⁺ cells were separated from the conjunctival tissues by using magnetic beads, and the expression of toll-like receptor (TLR) 2 and TLR4 was examined. Surface markers of CD68⁺ cells were compared with those of BMCs from the corneal stroma and epithelium.

results. CD45⁺ cells were detected in the substantia propria of the conjunctiva, and approximately 60% of these cells were CD68⁺. All the CD68⁺ cells expressed HLA-DR and CD14. CD68⁺ cells isolated from conjunctival tissues expressed TLR2 and TLR4 on flow cytometry. BMCs in both the corneal stroma and the subconjunctiva expressed scavenger receptor CD163. Macrophage mannose receptor CD206 was expressed by BMCs in the substantia propria of the conjunctiva, but not by BMCs in the corneal stroma or epithelium.

conclusions. These findings demonstrated that the main population of BMCs in the substantia propria of normal human conjunctiva is CD68⁺CD14⁺HLA-DR⁺ cells. These BMCs express scavenger receptor, macrophage mannose receptor, TLR2, and TLR4 and may play a role in adaptive and innate immune responses in the human ocular surface. These cells are phenotypically different from the CD68⁻CD206⁻ monocyte- lineage cells found in the corneal stroma and the CD11c⁺CD68⁻CD163⁻CD206⁻ dendritic cells residing in the corneal epithelium.

The conjunctiva is a thin mucous membrane that lines the eyelids and covers part of the sclera and the inner side of eyelids, ¹ and the conjunctival epithelium is continuous with that of the cornea. The bulbar conjunctiva is loosely adherent to underlying fibrous connective tissue, known as the substantia propria, which varies in depth and density compared with the tarsal plate. Vessels are distributed on the surface of the conjunctiva, while deeper vessels supply the peripheral corneal arcades, iris, and ciliary body. ¹  

In addition to the existence of bone marrow–derived cells (BMCs) in normal mouse cornea, ^{2 3 4} we have previously characterized BMCs that reside in the normal human corneal stroma and epithelium. ^{5 6 7} Monocyte-lineage BMCs exist constitutively in the corneal stroma, ⁵ whereas BMCs from the corneal epithelium are myeloid-lineage dendritic cells (DCs) that can be classified into at least three phenotypes (CD11c⁺CD16⁻ DCs form the main population along with a small number of CD11c⁺CD16⁺ DCs and CD11c⁺CD1c⁺ DCs). ⁵ The leukocytes that normally reside in human conjunctival tissues were used as normal controls for investigation of pathologic conditions, such as cicatricial pemphigoid, ^{8 9 10} vernal keratoconjunctivitis, ^{11 12} pterygium, ^{13 14} trachoma, ¹⁵ and Behçet disease. ¹⁶ However, characterization of the main BMC population in the conjunctiva and comparison with other BMC phenotypes in the cornea remains poorly understood in the normal human eye. 

In the present study, the subconjunctival region (i.e., the substantia propria) of normal human donor corneas was examined by immunofluorescence of cross-sections and fluorescence microscopy with the use of various leukocyte markers. We attempted to characterize the leukocytes of the subconjunctiva based on recent advances in knowledge about the cell surface markers of various leukocyte subsets. 

Methods

Human Donor Corneas and Immunohistochemistry

This study was conducted in accordance with the Declaration of Helsinki. Corneas were obtained from the Rocky Mountain Lions’ Eye Bank at 3 to 5 days postmortem (donor ages, 56–69 years) and were kept in corneal preservation medium (Optisol GS; Bausch & Lomb, Rochester, NY) at 4°C until use. Corneal tissues (n = 6) were embedded in optimal cutting temperature (OCT) compound. Then frozen cross-sections (10 μm) were cut on a cryostat, air-dried for 10 minutes, and immunostained without fixation. Primary monoclonal antibodies (mAbs) used for the immunohistochemical and flow cytometric staining procedures and their specificity are shown in Table 1 . All primary monoclonal antibodies (mAbs) for immunohistochemistry were obtained from BD Biosciences (San Diego, CA), except HLA-DR-FITC (TAL.1B5; Chemicon International, Temecula, CA), TLR2 (Alexis Biochemicals, San Diego, CA), TLR4 (Monosan, Uden, The Netherlands), and CD204 and CD207 (R&D Systems, Minneapolis, MN). Sections were blocked with an anti-Fc receptor (FcR) blocker (Miltenyi Biotec, Bergisch Gladbach, Germany) and with isotype-matched immunoglobulin for each antibody (5 μg/mL mouse IgG1, 5 μg/mL IgG2a, or 5 μg/mL IgG2b; Dako, Carpinteria, CA) diluted in phosphate-buffered saline (PBS) for 30 minutes to eliminate nonspecific staining. Then fluorescein isothiocyanate (FITC)- or phycoerythrin (PE)-conjugated mAbs or isotype-matched control antibodies (mouse IgG1, κ-FITC isotype control, MOPC-31C, mouse IgG2a, κ-FITC, G155–178, mouse IgG2b, κ-PE 27–35) were applied for 30 minutes. Nonspecific staining with FITC- or PE-conjugated isotype controls (IgG1, IgG2a, and IgG2b) was not detected after pretreatment with the anti-FcR blocker and the isotype-matched immunoglobulin for each antibody. Next, the sections were covered with mounting medium (Vector Laboratories, Burlingame, CA) and were examined under a fluorescent microscope (model BH2-RFL-T3 or BX50; Olympus, Tokyo, Japan). Sections stained with FITC-conjugated anti-CD45 mAb (HI30) were coverslipped using an antifading mounting medium that contained propidium iodide (PI; Vectashield; Vector Laboratories). In double staining, nuclei were stained with Hoechst 33342. As secondary Abs, FITC-conjugated bovine anti-goat IgG antibody and Alexa 488–conjugated anti-mouse IgG2b antibody were used. All staining procedures were performed at room temperature. Cell numbers in the conjunctival loose connective tissue layer under the multilayered epithelium were counted on both sides of the sclera of normal human donor corneas, and the average number of cells with positive staining was calculated. However, the cell count of the conjunctival epithelium was not determined because the epithelium did not remain intact in the cross-sections. 

Isolation of CD68⁺ Human Corneal Epithelial Cells

Before conjunctival cells were isolated, the peripheral cornea (including the limbal region) was dissected away from the stroma to avoid possible contamination by limbal epithelial cells. The dissected conjunctiva was incubated overnight at 37°C in serum-free basal medium containing trypsin/EDTA (Sigma-Aldrich, St. Louis, MO). After 3 washes with PBS, single cells were dissociated by trituration with a fire-polished Pasteur pipette. Then CD68⁺ cells were positively isolated with a magnetic-activated cell sorter (MACS; Miltenyi Biotec) according to the manufacturer’s instructions. For separation of CD68⁺ cells, three to five conjunctival specimens were processed together. In total, 16 donor corneas were used for isolation of the CD68⁺ cells used in the flow cytometry and reverse transcription–polymerase chain reaction (RT-PCR) experiments. 

Flow Cytometry

For flow cytometry, cells isolated by MACS were blocked in 3% normal human serum before incubation for 30 minutes with anti-human TLR2 mAb (TL2.1; Alexis Biochemicals, Lausen, Switzerland), anti-human TLR4 mAb (HTA125; Monosan, Uden, Netherlands), or the isotype control (mouse IgG2a, κ). After three washes in PBS, the cells were incubated with the FITC-conjugated goat anti-mouse secondary antibody for 30 minutes at 4°C. Then flow cytometric analysis was performed using a flow cytometer (FACSCalibur; Becton Dickinson, Raleigh, NC). All experiments were performed independently in duplicate. 

RNA Preparation and RT-PCR

Total RNA was isolated from CD68⁺ cells (Isogen; Nippon Gene, Tokyo, Japan) according to the manufacturer’s instructions. Water was used as the negative control. First-strand cDNA was synthesized from the total RNA with reverse transcription (Reverse Transcription System^; Promega Corporation, Tokyo, Japan). The PCR reaction mixture was composed of 1% cDNA, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl₂, 0.2 mM dNTPs, 20 pmol oligonucleotides, and 2.5 U polymerase (AmpliTaq Gold; Perkin Elmer, Wellesley, MA) in a total volume of 50 μL. After incubation at 95°C for 9 minutes, amplification was performed at 94°C for 30 seconds, and then at 60°C for 30 seconds in a thermal cycler (I Cycler; Bio-Rad Laboratories, Hercules, CA). Samples were separated on 2% agarose gel, and the products were detected by ethidium bromide staining. TLR2 and TLR4 expression was detected with probe (Dual PCR kits; Maxim Biotech, Inc., South San Francisco, CA) according to the manufacturer’s instruction. 

Results

CD45⁺ Cells in the Corneal Epithelium

For fluorescence microscopy, cross-sections of the subconjunctiva were stained with an FITC-conjugated anti-CD45 (panleukocyte) mAb, and nuclear staining was performed with PI (red). As a result, a significant number (272 ± 96 cells/mm²) of CD45⁺ cells were detected in the subepithelial region of the bulbar conjunctiva (Fig. 1A) . Among these CD45⁺ cells, 56% ± 18% (152 ± 49 cells/mm², n = 6) were also CD68⁺, as shown in Figures 1B 1C 1D(representative photographs of double staining). The CD68⁻ cell population contained a variety of small subpopulations, such as CD11b⁺ or CD11c⁺ cells. CD3⁺ (T cells; UCHT1) and CD19⁺ (B cells; SJ25C1) cells were focally detected, suggestive of conjunctiva-associated lymphoid tissues as described elsewhere. ^{17 18} CD66 (granulocytes; B6.2/CD66) cells were detected in the interior of blood vessel. There was no staining with the mAbs for CD56 (NK cells; B159), CD80 (B7.1; L307.4), CD86 (B7.2; FUN-1), CD1a (HI149; cutaneous Langerhans cells), and Langerin (CD207; cutaneous Langerhans cells) in the substantia propria of the conjunctiva (data not shown). 

Characterization of CD68⁺ Cells

To investigate the characteristics of the main cell population (CD68⁺ cells) in the substantia propria of the conjunctiva, we performed double staining with various markers and CD68. Nuclei were stained with Hoechst 33342. As shown in Figure 2 , all the CD68⁺ cells were stained by HLA-DRa (MHC class II, G46–6; Figs. 2A 2B ) and CD14 (monocytes, macrophages, or pre-DC, M5E2; Fig. 2C ). Among the CD68⁺ cells, 87% ± 9% (n = 6) were also CD11b⁺ and 18% ± 12% (n = 6) were CD11c⁺. 

TLR2 and TLR4 Expression by CD68⁺ Cells

TLR2 and TLR4 expression was examined using CD68⁺ cells isolated from the substantia propria of the conjunctiva with magnetic beads. More TLR2⁺ and TLR4⁺ cells were detected by flow cytometry with anti-TLR2 and -TLR4 antibodies than with the isotype-matched control antibody (nonimmunized mouse IgG2a; Fig. 3A ). When RT-PCR was performed, GAPDH mRNA was detected in the positive control and in CD68⁺ cells, but not in negative control samples (30 cycles). TLR2 mRNA (30 cycles) and TLR4 mRNA (35 cycles) were detected in the CD68⁺ cells by RT-PCR (Fig. 3B) . 

Comparison between Conjunctival CD68⁺ Cells and Corneal Stromal/Epithelial BMCs

BMCs in the human corneal stroma are CD14⁺ cells that include monocytes, macrophages, and dendritic cells, as described elsewhere, ⁵ wheres BMCs in the corneal epithelium are myeloid-lineage DCs. ⁵ These cells in the corneal stroma and epithelium were compared with the main population of BMCs in the substantia propria of the conjunctiva. BMCs in the corneal stroma and epithelium were CD68⁻ (data not shown). In addition, macrophage mannose receptor CD206 was expressed by BMCs in the substantia propria of the conjunctiva (Figs. 4A 4B 4C 4D)but not by BMCs in the corneal stroma and epithelium (data not shown). Furthermore, corneal stromal BMCs (Figs. 4E 4F)and conjunctival BMCs, but not corneal epithelial BMCs, expressed scavenger receptor CD163. Representative photographs of CD16⁺ corneal stromal BMCs (Fig. 4G)and CD16⁺ and CD68⁺ conjunctival BMCs (Fig. 4H)are shown. CD16⁺ cells accounted for 32% ± 17% (n = 6) of stromal BMCs, whereas 48% ± 14% (n = 6) of CD68⁺ subconjunctival BMCs were positive for CD16. Scavenger receptor CD204 (clone 351615) was negative in all three BMC populations (data not shown). 

Discussion

We detected a substantial number of CD45⁺ cells in the substantia propria, which is the subepithelial region of the normal human conjunctiva. The main population of these cells was CD14⁺ CD68⁺HLA-DR⁺ BMCs. Anti-CD68 (Y1/82A) antibody reacts with a glycoprotein expressed in the cytoplasmic granules of monocytes/macrophage, dendritic cells, and granulocytes. Lack of any expression of granulocyte marker CD66 and the low expression of CD11c⁺ by CD68⁺cells indicated that most of these cells were from the monocyte/macrophage lineage. In addition to CD14⁺ MHC class II⁺CD163 (scavenger receptor) staining of CD68⁺ cells, these cells expressed macrophage mannose receptor CD206. The fact that CD206 is expressed by macrophages and DCs suggests that CD68⁺ cells residing in the substantia propria of the conjunctiva are macrophage-like rather than monocyte-like cells. Therefore, these CD68⁺CD14⁺CD163⁺CD206⁺ macrophage-like cells are phenotypically different from the CD11c⁺CD68⁻ CD163⁻CD206⁻ DCs residing in the corneal epithelium ⁶ and CD11b⁺CD11c⁺CD14⁺CD68⁻CD206⁻ monocyte-lineage cells in the corneal stroma. ⁵ In light of the generally considered function of BMCs, they can have roles in each part of the ocular surface. Macrophage-like cells in the substantia propria of the conjunctiva have a strong phagocytic function in the outer part of ocular surface, DCs in the corneal epithelium respond to foreign antigens quickly, and monocyte-lineage cells in the corneal stroma work as DC and macrophage precursors in addition to having phagocytic activity. The mechanisms that control the tissue-specific distribution of BMCs are unknown. However, specific BMCs may be distributed accurately to each tissue by local chemokines, or, more likely, prototype cells may be differentiated into tissue-specific cells by the local effects of cytokines and chemical mediators. Interestingly, cutaneous representative DCs—Langerhans cells—express CD1a ¹⁹ and Langerin (CD207), ²⁰ but these markers were negative not only in the corneal epithelium ⁶ and stroma ⁵ but also in the substantia propria, as shown in this study, suggesting that ocular BMCs are at least phenotypically different from so-called Langerhans cells in the skin. Birbeck granules detected in epidermal Langerhans cells, however, should be investigated in the BMCs of ocular surface under the transmission electron microscope. ²¹  

CD163 is a glucocorticoid-inducible member of the scavenger receptor cysteine-rich family of proteins, which is highly expressed on human macrophages and monocytes. ²² CD163 expression is up-regulated during phagocytic differentiation of monocytes in response to cytokine stimulation and is downregulated during dendritic differentiation. ²³ The mannose receptor (CD206) on the surfaces of macrophages and dendritic cells binds terminal mannoses expressed on fungi and other pathogens ²⁴ and has been shown to mediate the cellular immune response to fungal antigens in vitro. ²⁵ The mannose receptor is involved in phagocytosis and endocytosis of antigens as part of the innate host defenses and also participates in signal transduction and the adaptive immune response. ²⁶ CD68⁺ cells in the substantia propria of the conjunctiva express both markers and reside at the surface of the eye, where these cells may actively protect the host by the adaptive immune response and innate defenses and by phagocytosis of foreign antigens and pathogens. 

CD16 (FcγRIII) was originally identified as a human NK cell–associated antigen, but it is also expressed by DCs, monocytes/macrophages, and granulocytes. ²⁷ CD16⁺ DCs show greater phagocytic and oxidative activity than CD16⁻ DCs, produce significant amounts of cytokines, ²⁸ and have a marked ability to activate naive T cells in vitro. ²⁷ CD14⁺CD16⁺ monocytes/macrophages are thought to be precursors of DCs. The finding that a significant population of CD14⁺CD16⁺ cells resides in the corneal stroma and conjunctival substantia propria suggests a role in the rapid immune response to foreign antigens and pathogens because DCs are more effective than macrophages for the initiation and expansion of secondary immune responses. ²⁹  

TLRs have been identified as part of a large family of pathogen recognition receptors, which play a decisive role in the induction of innate and adaptive immunity. ³⁰ DCs in the corneal epithelium may represent an efficient system that prevents or treats certain inflammatory microbial conditions at the ocular surface. TLR2 and TLR4 are receptors not only for bacteria and fungi but also for heat-shock proteins and viruses. ³¹ Expression of TLR2 and TLR4 by CD68⁺ cells from the substantia propria indicates that these cells are primed for various foreign antigens, suggesting a protective mechanism against exogenous proteins and foreign bodies that exists in the normal conjunctiva. 

The limitations of this study were as follows. We found dendritic CD45⁺ cells in the conjunctival epithelium but could not determine the characteristics of these cells. Because the conjunctival epithelium did not remain completely intact in corneal preservation medium (Optisol GS; Bausch & Lomb), various leukocyte phenotypes were not definitively determined in this study. The finding that the ocular surface is protected by DCs as a first-line defense suggests a critical role of such cells in protection of the eye. The number of BMCs in the substantia propria of the conjunctiva varied between donor corneas and between different parts of the conjunctiva. The variation depended on donor corneal condition and was greater than that of cells in the corneal stroma. To minimize this variation, two areas (on average) per donor cornea were examined to obtain representative data for an eye. However, donor corneas have an advantage as research material compared with the limited size of specimens that can be obtained by biopsy in healthy human volunteers. These specimens were donor corneas 3 to 5 days postmortem, and we cannot deny the possibility that preservation in corneal preservation medium (Optisol GS; Bausch & Lomb) affected the staining patterns or that BMCs migrated out of the tissue or underwent apoptosis during this time. Therefore, we compared the staining patterns obtained with various markers after bisected donor corneas were preserved in corneal preservation medium (Optisol GS; Bausch & Lomb) for 4 and 7 days. In contrast to the well-maintained expression of CD45, CD11b, CD68, and DRa (TAL.1B5), expression of CD14 and DRa (G46–6) decreased after 7 days of storage. Thus, the storage period of donor corneas should be considered when immunohistochemical studies are performed. However, evident leukocyte loss in donor corneas was not detected between 4 and 7 days in our preliminary study (data not shown). 

In summary, we characterized leukocytes in the conjunctival substantia propria of donor human corneas and determined that major histocompatibility complex class II–positive BM-derived CD68⁺CD11b⁺CD14⁺ cells of the myeloid lineage were the primary population. These cells expressed scavenger receptor CD163, macrophage mannose receptor CD206, TLR2, and TLR4 and may play roles in adaptive and innate immune responses in the ocular surface of human eye. These cells show phenotypic differences from the CD11b⁺CD11c⁺CD68⁻CD163⁺CD206⁻ monocyte-lineage cells of the corneal stroma and the CD11b⁻CD11c⁺CD68⁻CD163⁻CD206⁻ DCs of the corneal epithelium, suggesting that they have different roles in each part of the ocular surface. 

 Supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

 Submitted for publication December 30, 2006; revised May 30 and June 28, 2007; accepted August 17, 2007.

 Disclosure: S. Yamagami, None; S. Yokoo, None; S. Amano, None; N. Ebihara, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Antibodies Used in Labeling

Antibodies (Clone)	Specificity	Antibody Composition
CD1a-PE (HI149)	Langerhans cell, DCs	Mouse IgG1, κ
CD3-FITC (UCHT1)	T-lymphocytes	Mouse IgG1, κ
CD11b-PE (ICRF44)	Activated lymphocytes, Mo granulocytes, a subset of NK cells	Mouse IgG1, κ
CD11c-PE (B-ly6)	NK cells, a subset of B and T cells Mo, granulocytes, macrophages	Mouse IgG1, κ
CD14-FITC, PE (M5E2)	Mo, macrophages, DCs, LCs	Mouse IgG2a, κ
CD19-FITC (SJ25C1)	Mature and immature B cells	Mouse IgG1, κ
CD45-FITC (H130)	Panleukocytes	Mouse IgG1, κ
CD56-PE (B159)	Pan NK-cell	Mouse IgG1, κ
CD66-FITC (B6.2/CD66)	Granulocytes	Mouse IgG1, κ
CD68-PE (Y1/82A)	Mo/macrophage, DCs, Granulocytes, myeloid progenitor cells	Mouse IgG2b, κ
CD80-FITC (L307.4)	B7-1, costimulatory molecules	Mouse IgG1, κ
CD86-FITC (FUN-1)	B7-2, costimulatory molecules	Mouse IgG1, κ
CD163-PE (GHI/61)	Scavenger receptor	Mouse IgG1, κ
CD204 (351615)	Macrophage scavenger receptor	Mouse IgG2b
CD206-FITC (19.2)	Macrophage mannose receptor	Mouse IgG1, κ
CD207	Langerhans cell-restricted protein	Goat IgG
HLA-DR-FITC (G46-6)	HLA-DR antigens	Mouse IgG2a, κ
HLA-DR-FITC (TAL.1B5)	HLA-DR antigens	Mouse IgG1, κ
Mouse IgG1, κ-FITC (MOPC-31C)	Isotype control	Mouse IgG1, κ
Mouse IgG2a, κ-FITC (G155-178)	Isotype control	Mouse IgG2a, κ
Mouse IgG2b, κ-PE (27–35)	Isotype control	Mouse IgG2b, κ
TLR2 (TL2.1)	Toll-like receptor 2	Mouse IgG2a, κ
TLR4 (HTA125)	Toll-like receptor 4	Mouse IgG2a, κ

 DC, dendritic cells; LC, Langerhans cell; Mo, monocyte; NK, natural killer.

Figure 1.

 

Immunohistochemical staining of the substantia propria of normal human conjunctiva. (A) Transverse cross-sections stained with anti-CD45 mAb were coverslipped using antifading mounting medium that contained PI (red) for nuclear staining. Positive staining of CD45⁺ cells (green) is detected in the subepithelial region of the conjunctiva. (B–D) Among the CD45⁺ cells (B, green), 56% ± 18% (n = 6) were CD68⁺ cells (C, red). A merged image is shown in (D). Yellow cells are CD45⁻ (green) and CD68⁺ (red). Original magnifications, (A) ×200, (B–D) ×100. Scale bar, 50 μm.

Figure 2.

 

Representative photomicrographs of CD68⁺ cells stained with various markers. (A) All the CD68⁺ cells (red) are also stained for HLA-DRa (clone G46–6, green). Double-stained cells are observed as yellow cells (arrows). Some cells are HLA-DRa⁺ in the substantia propria (green, arrowheads). (B) Nuclear staining with Hoechst 33342 was added (blue) in the double stainings of CD68 and HLA-DRa. (C) All the CD68⁺ cells (red) are CD14⁺ (green). In the merged image, yellow cells are double-positive for CD68 and CD14 (arrows). One cell is CD14⁺ but CD68⁻ (green, arrowhead). Nuclei are stained with Hoechst 33342. Original magnifications, (A, B) ×100. Scale bar, 50 μm.

Figure 3.

 

TLR2 and TLR4 expression revealed by flow cytometry and RT-PCR. TLR2 and TLR4 expression was examined in CD68⁺ cells isolated from the substantia propria of the conjunctiva with magnetic beads. (A) More TLR2⁻ and TLR4⁺ cells were detected by flow cytometry using anti-TLR2 and -TLR4 antibodies than using isotype-matched control antibody. (B) GAPDH mRNA is detected in the positive control and in CD68⁺ cells but not in the negative control (563 bp, 30 cycles). Both TLR2 mRNA (302 bp, 30 cycles) and TLR4 mRNA (631 bp, 35 cycles) were detected in the positive control and in CD68⁺ cells but not in the negative control. GAPDH, TLR2, and TLR4 cDNAs (provided by the manufacturer) served as the positive controls. SSC-H, side scatter height; M, size markers; P, positive control; S, sample; N, negative control.

Figure 4.

 

Comparison between conjunctival CD68⁺ BMCs and corneal stromal or epithelial BMCs. CD68⁺ cells (red, A) but not BMCs in corneal stroma and epithelium (data not shown) express the macrophage mannose receptor CD206 (green, B) in the substantia propria (merged, C). (D) Nuclear staining with Hoechst 33342 was added (blue) in the double stainings of CD68 and CD206. (E) Corneal stromal CD45⁺ cells (BMCs, green) express scavenger receptor CD163 (red, merged yellow). (F) Nuclei in (E) are stained with Hoechst 33342. Conjunctival BMCs, but not corneal epithelial BMCs, express CD163 (data not shown). CD16⁺ (red) cells are present among corneal stromal BMCs of CD45⁺ cells (green, merged yellow, arrows, G) and conjunctival CD68⁺ (green) BMCs (merged yellow, arrows, H). Original magnification, ×100. Scale bar, 50 μm.

The authors thank Toshiya Osawa and Kayo Aoyama for excellent technical support.