SEARCH

SEARCH BY CITATION

Keywords:

  • olfactory ensheathing cells;
  • olfactory bulb;
  • immuno-histochemistry;
  • human fetus;
  • CNS restoration

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED

Clinical studies have expanded the therapeutic olfactory ensheathing cells (OECs) transplantation to different human Central Nervous System (CNS) diseases. In fact, the OEC transplantation in clinic is a mixture of olfactory bulb cells; they even have not demonstrated that they have such a subpopulation yet. However, as a source of OECs transplantation, the development and identification of human fetal OECs are still need more understanding, because some surgery try to restoration CNS injury with a more purity of OEC cultures generated by a number of different procedures. In this article, twelve human fetal olfactory bulb (OB) samples were obtained from six fetuses in 20 weeks of gestation, it was studied by immunofluorescence on histological sections and cultured cells with multiple antibodies under confocal microscopy. The P75NTR positive OB-OECs (olfactory ensheathing cell from the olfactory bulb) were present in both outer olfactory nerve layers and glomerular layer. The percentage of OB cells in culture, about 22.31 was P75NTR positive, 45.77 was S100β, and 31.92 was GFAP. P75NTR and GFAP were coexpressed with S100β, respectively; however, P75NTR was not coexpressed with GFAP in human fetal OECs. It is suggested that the localization and development of human OECs in OB are different to those in rodent, and the P75NTR immunohistological staining is still necessary to identify and characterize human fetal OECs in culture before transplantation. Anat Rec, 2010. © 2009 Wiley-Liss, Inc.

Since olfactory ensheathing cells (OECs) have been reported to enhance axonal regeneration in rat spinal cord when transplanted into lesion sites (Li et al.,1997; Ramon-Cueto et al.,1998), OECs have become a prime candidate for cell-mediated repair following a variety of Central Nervous System (CNS) lesions (Richter and Roskams,2008) not only in animal model but also in clinical situation (Li et al.,1998; Dobkin et al.,2006; Franssen et al.,2007; Bauchet et al.,2008). More and more literature reviews give us more and more hope that OEC transplantation to be one of the most promising therapeutic strategies (Barnett and Riddell,2007; Sasaki et al.,2007; Bauchet et al.,2008; Bunge,2008; Radtke et al.,2008; Richter and Roskams,2008; Kawaja et al.,2009); OECs have been successfully transplanted in acute (Resnick et al.,2003; Polentes et al.,2004; Collazos-Castro et al.,2005; Lopez-Vales et al.,2006; Andrews and Stelzner,2007; Sasaki et al.,2007) and chronic (Andrews and Stelzner,2004; Lopez-Vales et al.,2007) models of rodent spinal cord injury. Data obtained using various injury models support the view that OEC transplants can reduce cavitation, increase axonal sprouting and regeneration, and a moderate degree of functional motoric recovery (Li et al.,1997; Ramon-Cueto et al.,2000; Collazos-Castro et al.,2005; Lopez-Vales et al.,2007; Sasaki et al.,2007). Moreover, transplanted OECs promote neuroplasticity in murine models of stroke (Shyu et al.,2008), restore functional deficits in rat model of Parkinson's disease (Agrawal et al.,2004), and OECs have been transplanted to an amyotrophic lateral sclerosis (ALS) mouse model (Morita et al.,2008). In clinic, Lima et al. (2006) and Feron et al. (2005) used adult autologous olfactory mucosa OECs from patients to repair spinal cord injury. Huang and colleagues (Beijing, China) treated hundreds of patients with spinal cord injury by transplantation of the cultured OECs from human fetal olfactory bulbs (Huang et al.,2003). Recently, they made a controlled pilot study and transplanted the cultured human fetus OECs in 15 patients with ALS (Huang et al.,2008). Despite this progression, several important issues have yet to be thoroughly addressed: for instance, what is the phenotype feature of human fetal OECs in situ? And what is the classification of human olfactory bulb cells in vitro before transplanting in patients with CNS disease? Recently, Kawaja provided an objective evaluation of many methods currently being used to generate cultures of OECs from rodent and non-rodent species, highlighting the pros and cons of each technique with regards to the validation of the cellular composition of the resulting cultures (Kawaja et al.,2009).

More than hundreds of patients have received the clinical transplantation of OB-OECs (olfactory ensheathing cells from the olfactory bulb) obtained from human fetus cadaver in the age of 4–6 gestating months (Huang et al.,2003;Huang et al.,2008). In fact, the OECs transplantation in clinic is a mixture of olfactory bulb cells, even though, the purity of OEC cultures, generated by a number of different procedures, is most often defined by the percentage of cells that express the P75NTR (Ramon-Cueto et al.,1993). This phenotypic marker is, however, common to both OECs and Schwann cells in vitro (Kawaja et al.,2009). However, we are still lacking a well-designed, controlled trial of using transplanted olfactory tissue in spinal cord injury: for instance, the distribution and phenotypic features of OECs in olfactory bulb in human fetus, the immunohistochemical characterization, and the quantification of the human embryonic OB-OECs in culture are still unclear. The idea that OB cell cultures possess three different types was first presented by Ramon-Cueto and colleagues (Ramon-Cueto et al.,1993): (1) fusiform Schwann cell-like OECs, (2) flattened multipolar astrocyte-like OECs, and (3) macrophage-like cells. These authors described three cell types in their cultures but only one of them was identified as OECs. Fusiform cells were identified as endothelial cells, and macrophage-like cells as microglia. Only P75NTR, GFAP positive multipolar cells were identified as OECs in rodent.

The goal of this article is to identify the immunohistological characterization of human fetal OB-OECs in OB cell cultures for clinical transplantation; we will focus on the expression of P75NTR and its coimmunolabeling with other cell markers in human fetal OECs from the 20th gestation week in vivo and in vitro.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED

Source of Tissue Samples

Twelve olfactory bulb samples in the 20th gestation week were obtained from six human fetuses aborted in the second trimester and autopsied within 2 hr of death. The tissues were obtained with the informed consent of the donors, as part of the clinical study in which OECs are transplanted into the patients with spinal cord injured (Huang et al.,2003; Huang et al.,2008) in Beijing Rehabilitation Center (Beijing, China). The clinical study protocol is approved by the Ethics Review Committee of Beijing Rehabilitation Center (Beijing, China).

Immunohistology

Six Olfactory bulbs from human fetuses in the 20th gestation week were fixed in freshly prepared 4% paraformaldehyde for 2 hr. Following fixation, specimens were cryoprotected into 20% sucrose solution for 24 hr and frozen. To determine the characterization of OECs in human fetal OB, the bulbs were cut into coronal in the rostral-caudal axis. 8 μm coronal sections were cut with a cryostat, mounted the sections on glass slides (Fisher), and put slides into −70°C for storage. To stain the sections, we placed the slides containing the sections on a slide warmer at 37°C for 2 hr. Then we washed them thrice with PBS for 10 min each time, incubated the sections in a blocking solution containing 10% normal goat serum (Vector Laboratories) with 0.3% Triton x-100 for 2 hr, then incubated with primary antibody solutions overnight at 4°C, secondary antibody solution for 45 min. To stain nuclei, we used Hoechst 33342 (10 μg/mL, Sigma-Aldrich).

The detailed information of primary antibodies used in present paper was listed in Table 1. Secondary antibodies were goat-anti-mouse Alexa 488 or 546, goat-anti-rabbit Alexa 488 or 546 (1:400, Molecular Probes), goat-anti-mouse Cy5 (1:400, Jackson Immuno Laboratory).

Table 1. The detail information of first antibodies
AntigenImmunogenManufacturer, species, type, catalog number, dilutionCharacterization
FibronectinFibronectin Ab-11Lab Vision, Mouse monoclonal #MS1351-P,1:200Reacts with exrracellular matrix of human, mouse, rat
GAP-43Anti-Growth-Associated Protein-43Sigma-Aldrich, monoclonal #G9264, 1:500Exhibit a wide interspecies cross-reactivity
GFAPAnti-GFAP cytoplasmic domainDakocytomation, Rabbits polyclonal, #Z0334, 1:200Reacts with cat, dog, mouse, rat, sheep, human, cow
GFAPAnti-GFAP cytoplasmic domainChemicon, mouse monoclonal #MAB360, 1:400Exhibit a wide interspecies cross-reactivity
LamininAnti-human lamininHybridoma bank, mouse monoclonal #2E8, 1:20Reacts specificity with human, rat
Low-affinity NGF receptor anti-p75Recombinant human p75 transmembrane domainPromega, Rabbits, polyclonal, #G3231, 1:200Reacts with, mouse, rat, human, chicken
Low-affinity NGF receptor anti-p75Recombinant human p75 transmembrane domainMillipore, Mouse, monoclonal, #MAB5592X, 1:500Reacts with mouse, rat, human
MAP-2Anti-MAP 2, with GTP polymerized tubulinAbcam, Chicken polyclonal #ab5392, 1:500Reacts with human, mouse, rat, cow
NestinAnti-Nestin human specific nestin proteinChemicon, mouse Monoclonal #SC33677, 1:300Staining in the developing human CNS. Antibody reacts only with human
O4O-antigens, markers on the surface of olgodendrocytesChemicon, mouse Monoclonal #MAB345, 1:50Stains only single band of 28 kD m.w. on Western blot of rat brain
S100βAnti-S100 β-subunitSigma-Aldrich, mouse monoclonal #S2532, 1:1,000Reacts with Cat, rat, human, cow, porcine, rabbit

Cultures of Human Fetal Olfactory Ensheathing Cells

For cell culture, another six of olfactory bulbs from human fetuses in the 20th gestation week were isolated, representing cells left over from transplantation procedures. The bulbs were placed in ice-cold Hank's Balanced Salt Solution (HBSS). After washing the bulbs three times with HBSS, the bulbs were diced into small fragments and digested in trypsin-EDTA (0.05% w/v) at 37°C for 10 min. An equal volume of DMEM-F12 with 10% certified fetal bovine serum (D/F-10S) was added to stop the digestion. The cells were cultured on cover slips with D/F-10S media for 7 days before fixation and staining.

Immunostaining to Identify OECs in Culture

Cell cultures were fixed with 4% paraformaldehyde for 10 min and then washed with PBS. After incubating the cells in the blocking solution, the primary antibody was added for 1 hr and secondary antibody for 30 min. The nuclei were stained with Hoechst 33342. To label different OB cultured cells, we used the following antibodies: monoclonal anti-P75NTR (1:500, Sigma-Aldrich), polyclonal anti-P75NTR (1:200, Promega), monoclonal anti-GFAP (1:400, Chemicon), polyclonal anti-GFAP (1:200, Dako), monoclonal anti-S100β (1:1000, Sigma-Aldrich), monoclonal anti-Nestin (1:300, Chemicon), monoclonal anti-Fibronectin (1:200, Lab Vision), monoclonal anti-Fibronectin (1:20, hybridoma bank). The detailed information of primary antibodies used in this article was listed in Table 1.

Photomicrography and Cell Counting

The sections and the culture dishes were imaged with a Zeiss LSM510 confocal microscope. Some of the confocal images were deconvolved to reduce light scattered background (Bitplane AG, Zurich, Switzerland). Large images were scanned in high resolution and stitched together using Zeiss Axio Vision software. The cell counting in cell culture was performed by recognized different colors marked cells on the photomicrographs on the screen of Zeiss LSM510 confocal microscope, using Zeiss Axio Vision software. In detail, a 24 dishes cell culture plate was plated per olfactory bulb, the area of each dish is about 20 mm2. The photomicrographs were taken under 40× magnification in random three fields per dish (72 fields per olfactory bulb). The positive cell quantification was expressed as the Positive cell Mean ± SEM vs. the Mean ± SEM of Hoechst nucleus and the percentage in per field (Table 2).

Table 2. Cell numbers and percentage of marked cells in human fetal olfactory bulb in culture
AntibodiesCounted image (n)Tissue cultureCell cultureTotal %
Positive cell Mean ± SEMHoeschst nucleus Mean ± SEM%Positive cell Mean ± SEMHoeschst nucleus Mean ± SEM%
P752042.91 ± 13.3200 ± 2.021.45 ± 12.846.32 ± 4.8200 ± 5.323.17 ± 10.322.31 ± 11.5
GFAP2063.03 ± 11.9200 ± 2.532.51 ± 10.562.68 ± 4.2200 ± 7.731.33 ± 13.231.92 ± 10.0
S100β2092.72 ± 14.7200 ± 2.246.04 ± 13.490.52 ± 5.1200 ± 10.345.50 ± 10.845.77 ± 12.1

Statistics

Data were expressed as mean ± SE. Statistical significance was verified by one-way ANOVA followed by the Tukey test for multiple comparisons. A probability value of P < 0.05 was considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED

The Distribution of OECs in the Olfactory Nerve Layer of Human Fetal OB

The whole image of olfactory bulb in the coronal section of human fetal in the 20th gestation weeks and the expressions of different cell markers were shown in Fig. 1. The olfactory nerve layer (ONL) is the outermost layer of the main olfactory bulb, in which a multitude of axonal fascicles penetrate. The entire ONL expressed P75NTR positive immunostaining, the outer layer expressed intensely and the inner layer expressed dimly (Fig. 1A1–E1). The location of Fig. 1B1–B4 at higher magnification was pane in Fig. 1A1–A4 at low magnification. There were intense P75NTR-positive cells surrounding the large axonal bundles in the outer layer, but weakly labeled in the inner layer (Fig. 1B1–E1). Double labeling with different cell markers showed P75NTR appeared in the outer part of OBL layer In A1 and B1, whereas S100 was located in the inner part of ONL layer in A2 and B2. It was even clearer in the merge images (A4 and B4) where both markers were seen in red and green in the outer and inner parts of ONL, respectively. There were very dim laminin immunohistochemical labeling in the outer layer (Fig. 1C2), and nearly no tightly association with P75NTR (Fig. 1C4). Nestin was also expressed in both inner and outer layers(Fig. 1D2)as well as in the deep of nerve layer, some of them were tightly associated with P75NTR (Fig. 1D4). GFAP was intensely expressed in both inner and outer layers (Fig. 1E2); however, they clearly showed no tight association with P75NTR (Fig. 1E4). Again, there were more weakly immunolabeled O4 (oligodendrocytes progenitor)-positive cells in olfactory nerve layer and fewer were tightly associated with P75NTR (figures were not shown). The P75NTR-positive cells did not express fibronectin (the fibroblast marker), and MAP2 (the neuronal marker) (figures were not shown).

thumbnail image

Figure 1. Trible fluorescence immunohistochemic label in the olfactory nerve layer of human fetal (gestation week 20) olfactory bulb in coronal sections. A: Low magnification showed the location of B. The OECs express P75NTR (A1-D1, red, E1 green), as well as S100β (A2, B2, red). Nuclei were stained by Hoeschst 33342 (A3-E3, blue). Some OB cells show patchy staining for laminin (C2, green). The OB cells weakly express nestin (D2, green). Relatively few OEC cells express GFAP (E2, red). Bar: A = 500 μm, B-E = 50 μm.

Download figure to PowerPoint

The Distribution of OECs in the Glomerular Layer of Human Fetal OB

The glomerular layer is just below nerve fiber layer, the primary olfactory axons project into the glomerular layer, where they synapse with the olfactory mitral and tufted neurons. The triple fluorescence staining of P75NTR and MAP-2 with GAP-43 or Nestin in the glomerular layer was showed in Fig. 2. At low magnification, the glomerular layer was located in the deep of ONL (Fig. 2A1–A4 and C1–C4). The P75NTR-positive OECs were clearly present around the glomeruli (Fig. 2B1, B4, and D1–D4). The intensity for GAP-43 was homogeneous around the glomeruli (Fig. 2B2).

thumbnail image

Figure 2. Trible fluorescence immunohistochemic label in the glomerular layer of human fetal (gestation week 20) olfactory bulb. A and C, low magnification shown the location of B and D in the whole section of OB. B: Many P75NTR-positive OEC cells (B1, green) surround the developing glomeruli (B2, red) with MAP2-positive dendrites (B3, blue) in the center of the glomeruli. B4 showed the merged B1-B3, and the location of B5. B5. High magnification showed the developing glomeruli have central cores of MAP-2 positive dendrites (blue), a ring of GAP-43 positive axons (red), and surrounding P75NTR positive OECs (green). D: A fully formed glomerulus with a nucleus-free core, surrounding P75NTR-positive OECs (D1, red), nestin-positive radial glial processes (D2, green), glomerulus with MAP2-positive dendritic processes (D3, blue), D4 shown the merged B1-B3, and the location of D5. D5: The glomerulus with MAP2-positive dendritic processes (blue), GFAP-positive processes (red) and nestin-positive processes (green). P75NTR-positive OEG processes ensheathing the glomerulus (red). Bar: A and C = 500 μm, B and D = 50 μm. B5 and D5 = 5 μm.

Download figure to PowerPoint

At high magnification, it was observed that P75NTR-positive OECs were clearly accompanied with the olfactory axons in the glomerular (Fig. 2B5). P75NTR-positive OECs encircled a glomerulus and sent P75NTR-positive processes into the glomerulus (Fig. 2B5). The intense GAP-43 labeling was also found inside glomeruli (Fig. 2B5). MAP2 seemed more intense in the plexiform, granular layer and glomerular layer (Fig. 2D3, D4, and D5).

Nestin staining revealed numerous radial glial cells processed outside and between the glomerulis but not inside. (Fig. 2C2, C4, D2, D4, and D5). Only at high magnification, a few nestin positive processes can be observed inside the glomeruli (Fig. 2D5). The P75NTR-positive OECs in glomerular layer surrounded the developing glmeruli with MAP-2 marked neuronal dendrites in the center of glomeruli, especially in the medial-anterior region of the olfactory bulb.

Immunohistochemic Characteristics of Human Fetal OB cells in Culture

The human fetal OB cell culture was separated into two groups. One was the small pieces of OB tissue culture (Fig. 3), another was OB cells culture (Fig. 4). After 7 days of culture, the quantification of both tissue and cell culture were shown in Table 2.

thumbnail image

Figure 3. Tissue culture of human fetal olfactory. A: P75NTR and GFAP. A1: P75NTR-positive cells (red). A2: hoechest stained cellular nucleus (blue). A3: GFAP-positive cell (green). A4: Merge of A1 + A2 + A3, seldom of red P75NTR-positive OECs were well coimmunolable with GFAP-positive cells. B: P75NTR-positive OECs were coimmunolable with S100β. C: S100β-positive cells were coimmunolable with GFAP. D: P75NTR-positive cells were coimmunolable with NCAM. E: few P75NTR-positive cells were coimmunolable with nestin. F: no P75NTR-positive cells were coimmunolable with FBN-1. Bar = 50 μm.

Download figure to PowerPoint

thumbnail image

Figure 4. Primary human fetal olfactory bulb cells in culture at low and high magnification. A and B: OECs express both P75NTR (green) and S100β (red). C and D: P75NTR-positive cells (red) are present but none of these express GFAP (green). E and F: majority GFAP (green) positive cells were co-immunolable with S100β (red). Bar: A, C, E = 50 μm, B, D, F = 5 μm.

Download figure to PowerPoint

Tissue Culture

The small piece of OB tissue grew bloomily and the cells burgeon radiated from inner of the tissue to the outer with bi-, tri-, or multipolar processes (Fig. 3). To identify the characteristics of OEC, several specific antibodies were selected (Table 1). Under confocal microscopy, the relationship between different markers, which include glial markers P75NTR, GFAP, S100β and no glial markers NCAMs (neuronal cell adhesion molecule), FBN-1(Fibrillin-1), and nestin (intermediate filament) were confiscated. The Hoechest 33342 was used to show the nuclei.

P75NTR and GFAP

About 21.45% of P75NTR-positive and 32.51% of GFAP-positive of OB-cells were shown in Fig. 3A and Table 2. They do not show any coexpression after merged (Fig. 3A4). They were completely different especially in the peripheral of tissue piece, even though both cells congregated closely inside the piece of tissue.

P75NTR and S100β

The P75NTR-positive and about 46.04% of S100β-positive OB-cells was shown in Fig. 3B and Table 2. The P75NTR-positive OECs were coimmunolabeled with a part of S100β in both of peripheral and inside of tissue piece (Fig. 3B4).

GFAP and S100β

Nearly all of the GFAP-positive OB-cells (Fig. 3C1) were coimmunolabeled with a part of S100β (Figs. 3C3, 4).

P75NTR and NCAMs

Neural cell adhesion molecules (NCAMs) are a family of closely related cell surface glycoprotein involved in cell to cell interactions during growth. Nearly, all of the P75NTR-positive OB-OECs (Fig. 3D1) were coimmunolabeled with NCAMs (Fig. 3D3,D4), both in the peripheral and inside of the tissue piece.

P75NTR and Nestin

When the cultured OB tissue was double immunofluorescence stained, some of the P75NTR-positive OB-OECs (Fig. 3E1) as well as other OB cells were coimmunolabeled with nestin (Fig. 3E3,E4).

P75NTR and FBN-1

Fibrillin-1 (FBN-) has generally been recognized as a marker of fibroblasts. Double fluoresce immuno- histochemical staining showed that no P75NTR-positive OB-OECs (Fig. 3F1) was coimmunolabeled with FBN-1 (Fig. 3F3,F4).

Cell Culture

OB cells attached on the 2nd day of cell culture. Some cells protruded small short prominences and had a good refraction. The cells with OEC-like morphology (small nucleus, thin cytoplasm, and fine processes) were detected at 3 days after seeding onto PLL-coated flasks and were observed after 7 days of culture. To determine the characterization of human OECs in OB cell culture, the glial markers of P75NTR, S100β, and GFAP were confiscated.

P75NTR and S100β

About 23.17% of P75NTR-positive OB-OECs and 45.50% of S100β OB-cells were shown in Fig. 4 and Table 2. At low magnification, the P75NTR positive OECs (Fig. 4A1) were fewer than S100β positive cells (Fig. 4A2,A4). At high magnification, the P75NTR-positive OECs with small nucleus, thin cytoplasm, and fine processes were intensely coimmunolabeled with S100β (Fig. 4B1–B3).

P75NTR and GFAP

About 31.33% of GFAP-positive OB-cells were shown in Fig. 4C2,D2. Whether at low or high magnification, the P75NTR positive OECs (Fig. 4C1,D1) were clearly not coimmunolabeled with GFAP (Fig 4.C3,D3).

GFAP and S100β

At low and high magnification, nearly all the GFAP-positive OB-cells (Fig. 4E1,F1) were coimmunolabeled with S100β (Fig. 4E2, E3, F2, and F3). However, a few S100β positive cells were not coimmunolabeled with GFAP (Fig. 4E3,F3).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED

In this article, we observed the distribution of OECs in the human fetal olfactory bulb, the classification of cultured human fetal OB cells. We believe this data will offer a useful reference for cellular strategy in neural restoration.

The Distribution of P75NTR-Positive OECs in Human Fetal Olfactory Bulb

In this study, we observe the expression of P75NTR is stronger in the outer layer, and weaker in the inner layer of olfactory nerve layer. The strongest expression of P75NTR is just in the area of olfactory axons approaching the PNS-CNS boundary. In this region, P75NTR positive OECs tend to be flat, almost sheet-like, but have long processes and attached with each other loosely. At the outer olfactory nerve layer, they are more compact and orderly.

We found that P75NTR-positive OECs were week present in putative glomerular layer. There was no special P75NTR-negative layer throughout the entire olfactory nerve layer. On the contrary, the expression of P75NTR was stronger close to the boundary of the outer nerve layer. This evidence supported the notion that OECs were a homogeneous P75NTR-positive population in developing human olfactory bulb. Our results indicated that OECs of olfactory bulb expressed P75NTR in the course of glomerular formation.

The ingrowth of olfactory axons accompanied by P75NTR-positive OECs was observed in the 20th gestation week. The glomeruli started to form indicated by interaction of olfactory axons and interneuron dendrites, especially in the anterior part of olfactory bulb where most olfactory axons were entering the bulb (Fig. 2). Typically, multiple P75NTR-positive OECs formed continuous rims around the neuropil and clearly encircle each putative glomerulus, separating glomeruli from each other. However, some investigators have proposed that the P75NTR-negative glial cells surrounding glomeruli are peripheral OECs, whereas GFAP-positive astrocytes processes in the center of the glomeruli derive from the radial glia (Raisman,1985;Valverde and Lopez-Mascaraque,1991;Valverde et al.,1992). On the contrary, other investigators have interpreted the absence of P75NTR from the glomerular layer as indicating absence of OECs in the glomerular layer throughout development (Treloar et al.,1999; Au et al.,2002). It was found in our present study that OECs sent long P75NTR-positive processes into the glomeruli in the development of human OB. It is suggested that OECs might play a role in glomerular formation and participate in the stabilization of the glomeruli in humans.

The Signification of Multiple Immunohistochemical Expression in Human Fetal OB

To clarify the characterization of OECs in human OB, several specific antibodies were used in this study. We found that human P75NTR-positive OECs in both olfactory nerve layer and putative glomerular layer were high associational with S100β (one of the family of EF-hand type Ca2+-binding proteins, which localizes in the cytoplasm and nuclei of astrocytes and Schwann's cells), seldom or a few association with O4 (O-antigens as differentiation markerson the surface of oligodendrocytes), laminin (a peripheral basal membrane protein, which contributes to cell differentation, cell shape and migration, maintenance of tissue phenotypes, and survival) (Fig. 1C2,C4), nestin (a intermediate filament) (Fig. 1D2,D4), and no association with GFAP (Glial fibrillary acidic protein, which is an immunohistochemical marker for astrocyte) in the development of human fetal olfactory bulb (Fig. 1E2,E4). It is indicated that the P75NTR positive phenotype of human fetal OECs in the 20th gestation weeks are Schwann's-like OECs in the central neuronal system, but different from astrocyte-like OECs. Fewer P75NTR-positive OECs coimmunolabeled with O4 and nestin means the human fetal in the 20th gestation weeks might have some stem/progenitors of glia and neuronal cells, which is also indicated in adult mouse (Wang et al.,2008). Human OECs are high associationed with NCAMs (Fig. 3D3,D4), it means that the human fetal OB-OECs have cell adhesion molecules. However, they do not express fibronectin (an extracellular matrix glycoprotein, fibroblast marker), GAP43 (neuron growth associated protein 43, as neuronal growth cones marker), or MAP2 (an agent of microtubule depolymerization, as neuronal dendrites marker), but the GAP43 and MAP2 positive nerve fibers in olfactory nerve are observed in glomerular layer (Fig. 2). It is indicated again that the human OECs in the 20th gestation weeks have the characterization of special glia, but neither characterization of fibroblasts, nor neurons or neuronal stem cells.

The Characterization of Human Fetal OB-OECs in Cell Culture before CNS Restoration

It is generally agreed that OECs in vivo coexpress P75NTR, S100β, and GFAP in rodent (Kawaja et al.,2009). However, it is clearly shown in this article that in human fetal OB cells cultures, the P75NTR positive OECs are only coimmunolabeled with S100β, but not with GFAP (Fig. 4). Immunohistochemically, the Schwann cell-like OECs and astrocte-like OECs in rats display different patterns of p75NTR and GFAP, the Schwann cell-like OECs have strong p75NTR but weak GFAP, and the astrocyte-like OECs have strong GFAP but weak P75NTR (Pixley,1992; Franceschini and Barnett,1996; Franklin and Barnett,1997). It means that the characterization of human OB-OECs is different to rats.

Recently, most studies have defined the purity of different animal as well as human OEC cultures based on sole on the proporting of p75NTR immunostaining (Barnett et al.,2000; Kato et al.,2000). We also determined the percentage of the fetal human OB cells in culture in this article, the P75NTR-positive OECs were 22.31, GFAP-positive OB-cells were 31.93, and S100β-positive OB-cells were 45.77 (Table 2). Double fluorescence staining shown by confocal microscopy indicated that both P75NTR and GFAP positive OB-cells were coimmunolabeled with the S100β, respectively, but the P75NTR-positive OB-OECs were clearly not coimmunolabeled with GFAP (Fig. 4), that is, different to in rodent. When compared with other authors, Barnett et al. (2000) primary cultured the human olfactory bulb, the data possessed 33–66% p75NTR. However, nearly at the same time, Kato et al. (2000) used human olfactory nerves as a tissue source over 90% of the human cell was GFAP. Such controversy results are still in different labs (Feron et al.,2005; Savchenko et al.,2005).

It is also shown that Nestin is coimmunolabeled by OB-OEC either in vitro or in vivo (Figs. 1–4). Nestin is not only a neural stem cell marker but also a marker for microglia (Takamori et al.,2009); reactive astrocytes (Wilhelmsson et al.,2006) including in humans (Tamagno and Schiffer,2006) and fibroblasts (Savchenko et al.,2005). Nestin and P75NTR can also label glial cells from the rostral migratory stream and in the case of the fetal olfactory bulb, nestin also labels glial precursors (Fig. 1D2 and Fig. 2D2). It further supports that fetal olfactory bulb human cultures contain precursor cells and not only OECs. Considering the different expression to other research, it may be due to they are in different stages (in embryonic or in adult). So, no single marker that distinguishes OECs from other glia and expression of many established markers is contextually dependent (Richter and Roskams,2008). We suggest that P75NTR is still a key specific marker of human fetal Schwann cell-like OECs, and the P75NTR, GFAP, and S100β antibodies are still needed to identify the human fetal OECs in OB cell cultures when preparing transplantation in clinic.

In conclusion, our data indicated that P75NTR-positive OECs present in the outer nerve layer of OB, and participate in glomerular formation in fetal human (20th week) olfactory bulb, entering the glomerular layer with ingrowing olfactory axons, encircling and sending processes into developing glomeruli. Our study further confirmed the classification and percentage of OB cell cultures that were predominantly immunopositive for P75NTR, GFAP, and S100β in human fetus OB cell culture at mid-gestation. P75NTR, GFAP, and S100β immunohistochemic staining are still necessary to identify OB-OECs in cell culture before CNS restoration.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED