Bone marrow transplantation temporarily improves pancreatic function in streptozotocin-induced diabetes: potential involvement of very small embryonic-like cells

Abstract

Background: The role of bone marrow (BM)-derived cells in pancreatic beta-cell regeneration remains unresolved. We examined whether BM-derived cells are recruited to the site of moderate pancreatic injury and contribute to beta-cell regeneration.

Methods: Low-dose streptozotocin (STZ) treatment was used to induce moderate pancreatic damage and hyperglycemia. Enhanced green fluorescent protein-positive (EGFP) BM chimeras were evaluated for beta-cell regeneration after STZ treatment.

Results: To test the hypothesis that pancreatic tissue injury induces a stromal cell-derived factor (SDF)-1 gradient to chemoattract the stem cells, we evaluated the expression of mRNA for SDF-1 in damaged pancreatic tissue. SDF-1 was significantly increased in the pancreas after damage, peaking at day 10. The majority of BM cells expressing mRNA for pancreatic development markers were detected in the subpopulation of CD45/Sca-1/Lin very small embryonic-like (VSEL) cells. VSEL cells mobilized from BM to peripheral blood in response to pancreatic damage, peaking in peripheral blood at day 5, and were enriched in the pancreas 10 to 15 days after STZ treatment. To confirm a role for BM-derived cells in pancreatic beta-cell regeneration, we prepared EGFP-->B6 chimeras. In the EGFP chimeras, EGFP cells were detected around duct and islets and were positive for insulin after STZ treatment. However, STZ-induced hyperglycemia was reduced only transiently (49-77 days) after pancreatic injury.

Conclusions: These data suggest that VSEL cells are mobilized into injured pancreatic tissue and contribute to beta-cell regeneration. Transplantation of BM-derived cells improves the function of injured pancreas, although the response is not sufficient to restore sustained normoglycemia.

Figure 1. Pancreas developmental markers (Nkx 6.1, Pdx 1, and Ptf 1) are significantly increased in the CD45⁻ fraction

(A) BM cells were harvested from B6 mice and sorted for CD45⁺/Sca-1⁺/Lin⁻ or CD45⁻/Sca-1⁺/Lin⁻ cells. The total RNA was isolated from BM cells, CD45⁺/Sca-1⁺/Lin⁻ or CD45⁻/Sca-1⁺/Lin⁻ cells, and analyzed for transcript for NKx 6.1, Pdx 1, and Ptf 1 by real-time RT-PCR. The CD45⁻/Sca-1⁺/Lin⁻ cell fraction was highly enriched in mRNA for markers of pancreas development. *P< 0.001 for CD45⁻/Sca-1⁺/Lin⁻ vs. BM cells or CD45⁺/Sca-1⁺/Lin⁻ cells. The pancreatic damage from using STZ treatment resulted in down-regulation of markers of pancreatic development (NKx6.1, Pdx1, and Ptfl) in BM (B) and upregulation of these markers in PB (C), spleen (D), and pancreas (E).

Figure 1. Pancreas developmental markers (Nkx 6.1, Pdx 1, and Ptf 1) are significantly increased in the CD45⁻ fraction

(A) BM cells were harvested from B6 mice and sorted for CD45⁺/Sca-1⁺/Lin⁻ or CD45⁻/Sca-1⁺/Lin⁻ cells. The total RNA was isolated from BM cells, CD45⁺/Sca-1⁺/Lin⁻ or CD45⁻/Sca-1⁺/Lin⁻ cells, and analyzed for transcript for NKx 6.1, Pdx 1, and Ptf 1 by real-time RT-PCR. The CD45⁻/Sca-1⁺/Lin⁻ cell fraction was highly enriched in mRNA for markers of pancreas development. *P< 0.001 for CD45⁻/Sca-1⁺/Lin⁻ vs. BM cells or CD45⁺/Sca-1⁺/Lin⁻ cells. The pancreatic damage from using STZ treatment resulted in down-regulation of markers of pancreatic development (NKx6.1, Pdx1, and Ptfl) in BM (B) and upregulation of these markers in PB (C), spleen (D), and pancreas (E).

Figure 2. Upregulation of SDF-1, HGF mRNA in damaged pancreas

Animals were treated with 35 mg/kg/day of STZ. The total RNA was isolated from STZ-damaged pancreatic tissue. The levels of SDF-1, HGF and LIF mRNA were analyzed by quantitative real-time RT-PCR. Data showed that the mRNA of SDF-1 and LIF were upregulated in damaged pancreatic tissue.

Figure 3. Dose titration of STZ treatment

(A) % of mice with hyperglycemia after treatment with STZ for 35 mg/kg/day × 5 days (group A) or 45 mg/kg/day × 6 days (Group B). (B) Blood glucose level after STZ treatment. Blood glucose concentrations of hyperglycemic animals are shown as the mean ± SD. (C) H & E staining of representative pancreatic section. The left panel shows islet; the right panel shows the islet and the duct. Bar = 50 μm. (D) Immunofluorescence of pancreatic section at 42 days after STZ treatment. Islets in STZ-treated mice showed altered morphology such as small size, disorganized architecture, and decreased insulin secretion compared to untreated controls.

Figure 4. Donor EGFP⁺ BMC contribute to regenerate

β-cells. (A) Preparation of EGFP⁺ chimeras. Donor BM cells were harvested from EGFP⁺ B6 mice and 5 × 10⁶ cells were transplanted into recipient B6 mice conditioned with 950 cGy TBI. (B) Donor EGFP⁺ chimerism was assessed 3 months using flow cytometry. (C) EGFP⁺ cells were observed in pancreas of EGFP⁺ mixed chimeras 42 days after STZ treatment. The pancreas of EGFP⁺ B6 and naïve B6 mice served as controls. (D) Top panel: sections were stained for EGFP (green) and insulin (red) and show co-localization of EGFP⁺ and insulin expressing cells in pancreatic ducts. The lower panel: sections stained for nuclei using DAPI (blue), insulin (red), and EGFP (green), and show co-localization of EGFP⁺ and insulin-expressing cells in the islet region at 42 days after STZ treatment.

Figure 5. FL + G-CSF treatment elevated blood glucose in STZ-treated mice

(A) EGFP⁺ mixed chimeras were treated with STZ for 6 days. Experimental mice were then treated with FL and G-CSF for 10 days. Untreated animals and STZ-treated animals served as controls. (B) Blood glucose and (C) weight were measured weekly. Blood glucose concentrations and weight were shown as the mean ± SD. *P< 0.05 for mice treated with FL + G-CSF vs. without FL + G-CSF.

Figure 6. Bone marrow transplantation temporarily reduces high blood glucose in STZ-treated mice

(A) Experimental design for the induction of hyperglycemia and transplantation of EGFP⁺ B6 bone marrow. (B) Transplantation of BM-derived cells into STZ-treated B6 mice conditioned with 350 cGy of TBI (n = 10). Animals received only media (n = 11) or animals without STZ treatment (n = 4) served as controls. Blood glucose concentrations were shown as the mean ± SD. *P < 0.05 for STZ-treated mice that received BM cells or media alone vs. mice not treated with STZ.