|
Endothelial Apoptosis as the Primary Lesion Initiating Intestinal Radiation Damage in Mice
François Paris, Zvi Fuks, Anthony Kang, Paola Capodieci, Gloria Juan, Desiree Ehleiter, Adriana
Haimovitz-Friedman, Carlos Cordon-Cardo, and Richard Kolesnick
|
Supplementary Material
Supplemental Figure 1. Apoptosis in the lamina propria of irradiated GI villae assessed by H&E staining. C57BL/6 mice received 15 Gy whole-body radiation (WBR), whereas
asmase-/- mice were treated with 16 Gy. Small intestinal specimens were obtained 4 hours after irradiation and stained by H&E. Scoring of apoptotic nuclei was carried out by using the criteria proposed by A. J. Merritt
et al. [in
Techniques in Apoptosis, T. G. Cotter, S. J. Martin, Eds. (Portland Press, London, 1996), pp. 269-299]. Apoptotic nuclei are identified by single arrows. Scale bars: Main panels, 80
m; insets, 30
m.
Medium version | Full size version
| Supplemental Table 1. ASMase deficiency protects against the GI syndrome. The median survival was derived from product-limit Kaplan-Meier survival curves. P values were calculated by the Mantel log-rank test. Quantification of endothelial cell (EC) apoptosis per crypt-villus unit was as described in Fig. 2C. Student's t test and chi-square test for small populations were performed for statistical comparisons. |
| Genotype | |
| Measurement | Wild type | asmase-/- | Significance |
| Median survival (days) | 6 | 9 | P < 0.001 |
Units with  3 apoptotic cells | 77.44% | 32.90% | P < 0.001 |
| Units with 11 apoptotic cells | 42.07% | 3.95% | P < 0.001 |
| Units with >20 apoptotic cells | 12.81% | 0% | P < 0.001 |
| Mice with GI death | 4/5 | 1/6 | P < 0.006 |
| Supplemental Table 2. p53 deficiency does not protect against the GI syndrome. The median survival was derived from product-limit Kaplan-Meier survival curves. P values were calculated by the Mantel log-rank test. Quantification of endothelial cell (EC) apoptosis in villus-crypt units was as described in Fig. 2C. Student's t test and chi-square test for small populations were performed for statistical comparisons. |
| Genotype | |
| Measurement | Wild type | p53-/- | Significance |
| Median survival (days) | 6 | 6 | |
| Units with 3 apoptotic cells | 93.81 | 87.28 | P = 0.06 |
| Units with 11 apoptotic cells | 44.09 | 57.44 | P = 0.09 |
| Units with >20 apoptotic cells | 23.61 | 28.72 | P = 0.12 |
| Mice with GI death | 5/5 | 5/5 | |
| Supplemental Table 3. Incidence of stem cell apoptosis at position 4 to 5 from the crypt base. Irradiated animals were exposed to 15 Gy WBR. Small bowel specimens were stained with TUNEL 4 hours after irradiation. Three animals were scored at each point, and 200 crypts were scored for each animal. Student's t test and chi-square test for small populations were performed for statistical comparisons. |
| Apoptosis for radiation dose of |
| 0 Gy | 15 Gy | Significance |
| Wild type | 6 ± 1% | 64 ± 8% | P < 0.001 |
| p53-/- | 6 ± 8% | 8 ± 2% | P = 0.12 |
| asmase-/- | 7 ± 3% | 57 ± 3% | P < 0.001 |
| Wild type + bFGF | 8 ± 1% | 57 ± 6% | P < 0.001 |
Supplemental Figure 2. Murine recombinant PECAM-1 blocks immunohistochemical detection of endothelial cells in the lamina propria. Small intestine specimens were incubated with a rat antibody against CD31 (clone MEC 13.3, PharMingen, San Diego, CA), diluted 1:100, at 4°C overnight in the presence or absence of murine PECAM-1 (at five times the concentration of antibody against CD31; a kind gift from W. Muller). Staining was developed by using biotinylated rabbit anti-rat antibodies, the avidin-biotin-peroxidase complex, and the true blue peroxidase substrate (Kirkegaard and Perry Laboratory, Gaithersburg, MD). Tissues were counterstained with eosin. Note that preincubation of antibody against CD31 with PECAM-1 blocks detection of the numerous blue-staining endothelial cells in the lamina propria. Similar results were obtained with irradiated tissue. Scale bars: Main panels, 70 m; insets, 30 m.
Medium version | Full size version
Supplemental Figure 3. Frequency histograms of apoptotic cells in the lamina propria of crypts and villae in the mucosa of irradiated small intestines. Specimens were obtained from 0 to 8 hours after 15 Gy WBR (Fig 3a) or 4 hours after 0 to 15 Gy WBR (Fig. 3b) and stained for apoptosis by using TUNEL. At least 400 crypts and villae were scored for apoptotic cells at each point. Data represent means from three experiments.
(Figure 3a.)
Medium version | Full size version
(Figure 3b.)
Medium version | Full size version
Supplemental Figure 4. bFGF rescues C57BL/6 mice receiving whole-abdominal irradiation. 8- to 12-week-old mice were treated with 17 Gy whole-abdominal irradiation. Animals receiving bFGF were injected intravenously (iv) with 3.2 g human recombinant bFGF in 800-ng doses delivered 30 min before, and 5 min, 1 and 2 hours after irradiation. Actuarial survival was calculated by the product-limit Kaplan-Meier method (27). Numbers in parentheses indicate the number of animals per group. Death in the group of irradiated animals pretreated with bFGF was associated with leukopenia and sepsis that developed despite the fact that marrow in unirradiated bones remained intact. The mean peripheral white blood cell (WBC) count before treatment with 17 Gy + bFGF was 10.05 ( 103 WBC/µl. In the five mice that died, the mean WBC count decreased shortly before death to 1.95 ± 0.37 ( 103/µl. In contrast, a nadir value of only 3.97 ± 0.64 ( 103 WBC/µl was observed in the six surviving mice (P < 0.05) on day 8 after treatment, rapidly rising to normal levels in the subsequent 2 days. A similar pattern of leukopenia has been observed in patients with abdominal and pelvic tumors treated with whole-abdominal irradiation [Z. Fuks, Semin. Oncol. 2, 253 (1975)]. These patients were found to develop potentially lethal leukopenia and thrombocytopenia purportedly due to radiation exposure of the hematopoietic tissue of the abdominal spine and pelvic bones.
Medium version | Full size version
Supplemental Figure 5. Intravenously injected bFGF localizes within the lamina propria. Eight- to 12-week-old C57BL/6 mice received 800 ng bFGF by iv injection at 0, 30, 60, and 120 min (total 3.2 µg). Small intestinal specimens were harvested 30 min after the last injection and fixed in formalin. Sections were incubated with a rabbit antibody against human bFGF (Ab-2, Oncogene Res. Products, Boston, MA), diluted 1:100, at 4°C overnight. Staining was developed using biotinylated anti-rabbit antibodies and the avidin-biotin-peroxidase complex. Tissues were counterstained with H&E. bFGF
Medium version | Full size version
