Turning Blood into Brain: Cells Bearing Neuronal Antigens Generated in Vivo from Bone Marrow Éva Mezey, Karen J. Chandross, Gyöngyi Harta, Richard A. Maki, and Scott R. McKercher

Supplementary Material

Supplemental Figure 1. Cultured bone marrow cells express nestin in vitro. Bone marrow cells were isolated from adult mice as described [see (20) in report] and treated with ACK lysing buffer (BioWhittaker) for 30 s to eliminate red blood cells. Cells were plated onto plastic tissue culture dishes and grown in DMEM-10% FCS supplemented with platelet-derived growth factor (10 ng/ml). Cultures were analyzed for nestin immunoreactivity 4 hours and 15 days after plating. The nestin antibody [#130 polyclonal IgG, 1:1000 (generous gift of R. McKay, NIH, Bethesda, MD)] was visualized using a goat anti-rabbit IgG antibody conjugated to RITC (red). Nuclei were labeled with DAPI (blue) to identify individual cells. After 4 hours in culture, bone marrow cells did not express nestin. However, after 15 days, nestin-immunoreactive cells were present and were morphologically identical to adult neural stem cells. In this preparation, 50% of all cells were nestin-positive. Scale bar represents 30 mm.

Medium version | Full size version

---

Supplemental Figure 2. Y chromosome staining in the CNS. Coronal sections from a 4-month-old nontransplanted (A to E) female and (F to J) male brains that were mounted on the same slide and processed together. The Y chromosome was detected with the tyramide-FITC conjugate. (A) and (F) show the overlay of the NeuN (red) immunostaining of neuronal nuclei and the subsequent Y chromosome hybridization (green). In (B) and (G), cell nuclei were stained with DAPI (blue). The overlay of the DAPI and Y chromosome staining is shown in (B) and (G). (C) and (H) show the nonradioactive ISH, which was performed after the NeuN staining to detect the Y chromosome in cell nuclei. (D) and (I) are overlays of the three colors in the female (A to C) and male (F to H), respectively. Although the NeuN and DAPI staining were present in both male and female sections, the Y chromosome was apparent only in the male brain, demonstrating the specificity of the hybridization. (E) and (J) are higher magnification views of the cerebral cortex in a female and a male brain, respectively. The images are triple exposures showing all three colors (red, neuronal nuclei; blue, cell nuclei; green, Y chromosome) together. Note the lack of the Y chromosome staining in (E), the female brain. As shown in (J), the location of the Y chromosome at the periphery of the nucleus is due to its heterochromatic nature. Low-magnification view of a sagittal section from (K) a nontransplanted female brain and (L) a female PU.1 knockout mouse brain 1 month after receiving male bone marrow cells at birth. In (L), the Y chromosome hybridization was visualized using BCIP/NBT (purple-black dots) in order to identify anatomical landmarks. Note the homogenous distribution of the Y chromosome, which is due to the fact that all microglia in these mice were also derived from the transplanted (male) bone marrow. cc, corpus callosum; cx, cerebral cortex; Cpu, caudate putamen; fi, fimbria hippocampi; hi, hippocampus; LV, lateral ventricle. Scale bar in (L) represents the following sizes: 25 mm, (A) to (D) and (F) to (I); 20 mm, (E), (J), and (M); 200 mm, (K) and (L). Similar results were observed with three different animals for each experimental condition.

Medium version | Full size version

---

Supplemental Figure 3. Y chromosome-positive neuronal nuclei in homozygous PU.1 female brains following transplantation of male bone marrow. High-magnification images of NeuN and Y chromosome double staining in (A) the olfactory cortex of 2-month-old, (D) the frontopolar cortex of 1-month-old, and (G) the cortical amygdala of 2-month-old females transplanted at birth with male bone marrow. The same fields were overlaid to show the colocalization of Y chromosome hybridization [(detected with tyramide-FITC conjugate (green)] to NeuN immunopositive (red) nuclei. (B), (E), and (H) represent the identical fields to (A), (D), and (G), respectively, taken through the UV filter to show all (not only neuronal) DAPI-positive cell nuclei (blue). The corresponding Y chromosome hybridization images were overlaid onto (B), (E), and (H) to demonstrate that each Y chromosome was associated with a cell nucleus. (C), (F), and (I) are the triple-color images of the same fields. (C) shows the overlays of (A) and (B), (F) shows the overlay of (D) and (E), and (I) shows the overlay of (G) and (H). Arrows point to the DAPI-positive cell nuclei that were associated with NeuN-immunopositive neurons and had the Y chromosome. There were no other cell nuclei adjacent to the NeuN- and Y chromosome-positive cells, strongly indicating that the these neurons arose from the bone marrow. Scale bar, 10 m.

Medium version | Full size version

---

Supplemental Figure 4. A Y chromosome- and NeuN-positive neuron in the parietal cortex (1.2 mm behind the bregma) of a 1-month-old homozygous female PU.1 knockout transplanted at birth with male bone marrow. The images were obtained using a Zeiss confocal microscope (×63 oil objective and ×2 digital magnification). (A to E) Five different levels through the section (1 mm thick each), overlaying the NeuN (red) and Y chromosome [visualized with tyramide-FITC conjugate (green)] fluorescence. (a to e) Overlays of the corresponding DAPI (blue) and Y chromosome staining. (a' to e') Overlays of the corresponding NeuN, Y chromosome, and DAPI fluorescence. The Y chromosome hybridization was localized to a NeuN-immunopositive neuron and was not associated with any neighboring nuclei in the x, y, or z planes. Scale bar, 10 mm. These results were observed with five independent Z series from three different animals.

Medium version | Full size version

---

---

Supplemental Figure 5. Z series of a Y chromosome- and NeuN-positive neuron in the somatosensory cortex (1.2 mm behind the bregma) of a 2-month-old homozygous female PU.1 knockout transplanted at birth with male bone marrow. The images were obtained with a Zeiss confocal microscope (×63 oil objective and ×2 digital magnification). (A to I) Nine different levels through the section (1 mm thick each), overlaying the NeuN (red), Y chromosome [detected with the tyramide-FITC conjugate (green)], and DAPI (blue) (to visualize cell nuclei) fluorescence. (a to i) Overlays of the corresponding DAPI and Y chromosome staining. (a' to i') Overlays of the corresponding NeuN and Y chromosome fluorescence. The Y chromosome was localized to a NeuN-immunopositive neuron and was not associated with any neighboring nuclei in the x, y, or z planes. Scale bar, 10 mm.

Medium version | Full size version

---

Supplemental Figure 6. Examples of the presence of Y chromosome-positive cells in the ventricular system of transplanted mice. (A, D, and E) The sections are stained with ethidium bromide to show cell nuclei, and the Y chromosome is detected using nonradioactive in situ hybridization and TSA amplification with FITC-tyramide. The Y chromosome is seen as a green dot in the periphery of the nuclei. (A) is a sagittal view of the lateral ventricle of a female recipient 3 months after transplant. (D) is a coronal section of the ventral portion of the lateral ventricle from another 3-month-old recipient. Note the high density of Y chromosome-positive cells in the choroid plexus. (E) demonstrates a high number of Y chromosome-positive nuclei in the basal subarachnoideal space. (B) and (C) are representative coronal sections of the lateral and the dorsal third ventricles, respectively, both from a 1-month-old recipient. The Y chromosome riboprobe was labeled with 35S-UTP and a MaxiScript T7 kit (Ambion). The hybridization was carried out at 80°C for 10 min, then at 55°C overnight. Washes were done as described previously. At the end, the sections were coated in Kodak NTB3 emulsion and developed a week later. The slides were then stained with DAPI to identify the nuclei. (B) shows a coronal image of the lateral ventricle. The autoradiographic grains were photographed in dark-field illumination through a red color filter. There is a high density of Y chromosome-positive cells in the choroid plexus, in the ependyma, and in the subependymal areas. (C) demonstrates the presence of many Y chromosome-positive cells in the dorsal part of the third ventricle. Abp, basal posterior amygdala; F, fimbria hippocampi; Ha, medial habenula; Hi, hippocampus; LHth, lateral hypothalamus; Lv, lateral ventricle; Ot, optic tract; 3rdVD, dorsal portion of the third ventricle. Scale bar in (E) represents the following sizes: 80 mm, (A), (C), and (D); 100 mm, (B); 40 mm, (E).

Medium version | Full size version

---

Supplemental Methods

Immunostaining of acutely isolated bone marrow cells. For the NeuN immunostaining, cells were immediately spin-seeded onto eight-well glass chamber slides (10,000 cells per well) (Lab-Tek) that were precoated with poly-D-lysine (10 mg/ml) (Sigma). Immediately after plating, cells were fixed with 2% paraformaldehyde for 10 min. Additionally, bone marrow cells isolated from our previously characterized muCNP-bgeo transgenic mice [K. J. Chandross et al., J. Neurosci. 19, 759 (1999)] were used to determine whether bone marrow preparations contained Schwann cells or oligodendrocytes. In these mice, the murine 2'3'-cyclic nucleotide 3'-phosphodiesterase (muCNP) promoter drives the expression of both a detectable and selectable marker in oligodendrocytes and Schwann cells. The muCNP-bgeo construct was made by ligating a 3.7-kb region of the muCNP-promoters I and II [M. Gravel, A. Di Polo, P. B. Valera, P. E. Braun, J. Neurosci. Res. 53, 393 (1998)] into a plasmid containing bgeo, a 4.2-kb reporter gene encoding a protein with both b-Gal and NPT activity [G. Friedrich, P. Soriano, Genes Dev. 5, 1513 (1991)]. Transgenic mice on a ([C57BL/6J × C3H] F1 × [C57BL/6J × C3H] F1) background were generated by pronucleus injection of the 8-kb muCNP-bgeo fragment. The animals were bred with C57BL/6J F1 mice, the same strain as the PU.1 mice. Bone marrow cells were flushed from the femurs of muCNP-bgeo mice, as described above, and were immediately spin-seeded onto eight-well glass chamber slides (800,000 cells per well) (Lab-Tek) that were precoated with poly-D-lysine (10 mg/ml) (Sigma). For live staining, cells were immediately incubated with an affinity-purified monoclonal IgM O4 antibody [I. Sommer, M. Schachner, Dev. Biol. 83, 311 (1981)] [hybridoma cells were provided by S. Pfeiffer (University of Connecticut, Farmington)] that was directly conjugated to either Marina blue (used at 1:50 dilution) (Molecular Probes) or Alexafluor-488 (used at 1:100 dilution) (Molecular Probes) [directly conjugated antibodies were provided by R. Cohen (NIH-NINDS, Bethesda, MD)]. Live staining was done in DMEM-10% fetal calf serum at 4°C for 30 min, and cells were subsequently fixed with 4% paraformaldehyde for 10 min at room temperature. Alternatively, cells were immediately fixed and permeabilized/blocked (0.1% triton-X, 1% IgG-free bovine serum albumin, 20% goat serum). Subsequently, double immunostaining was performed using b-Gal (monoclonal IgG2A, 1:200) (Promega) and NG2 (polyclonal, 1:500) [provided by W. Stallcup (La Jolla Cancer Research Foundation, La Jolla, CA)] antibodies. Subsequently, these antibodies were visualized using a goat anti-mouse IgG2a antibody conjugated to FITC (b-Gal) or a goat anti-rabbit IgG antibody conjugated to RITC (NG2). Cell nuclei were stained with DAPI. Negative controls included incubation with no secondary antibodies or incubation with secondary antibodies alone. In parallel, two eight-well chamber slides were incubated in X-Gal solution and examined for b-gal activity, as previously described [K. J. Chandross et al., J. Neurosci. 19, 759 (1999)]. Results were obtained from two independent preparations.