Figure 1.
Induction of hematoma and clot formation in the bone marrow.
(A) Induction of bleeding results in clot formation within 5 minutes as evidenced by Masson-Goldner staining. Bar represents 500 µm. The enlarged inset below shows the blood clot. Bar represents 100 µm. (B) The clot is still present, albeit partially reorganized at 24 hours after bleeding induction. Bar represents 500 µm. Enlarged inset is shown below. Bar represents 100 µm. (C) After 72 hours the clot resolved. (D) Thrombin staining of the bone shown in B confirms the presence of thrombin (in red) in the clot area. (E) Quantification of the change in hematoma size at different time points. After induction of bleeding in the bone marrow, mice were euthanized at the times mentioned, and tibiae examined. n = 3–4 mice/time point.
Figure 2.
Bone histomorphometry after clot induction.
(A–B) After three days from the time of induction of a blood clot Masson-Goldner staining (on the left) and calcein labeling (on the right) showed no difference between control (CT) in panel A and hematoma induction in panel B in the same mouse. Neither osteoblast number (C), nor bone formation rate (D), nor osteoclast number (E), nor erosion surface (F) were different between the two tibiae. CT represents the right tibia without clot induction and hematoma represents the left tibia in which bone marrow was flushed and a blood clot was induced. Tibiae were obtained three days after clot induction in the left tibia, embedded in polymethylmethacrylate and stained using Masson-Goldner. n = 4 mice.
Figure 3.
Enzymatic staining of bone sections.
(A) Osteoblast surface as evidenced by evaluating the surface stained positive for alkaline phosphatase did not differ between CT and hematoma bones. A representative stained pair of tibiae is shown on the left and quantification is shown on the right. (B) The number of osteoclasts as evidenced by counting the cells that stained positive for tartrate-resistant acid phosphatase and correcting to bone surface was not affected by clot induction. A representative stained pair of tibiae is shown on the left and quantification is shown on the right. CT represents the right tibia without clot induction and hematoma represents the left tibia in which bone marrow was flushed and a blood clot was induced. Tibiae were obtained three days after clot induction in the left tibia, fixed in 3,7% PFA, cryo-sectioned using adhesive film and stained as outlined in the methods. n = 4 mice.
Figure 4.
Infiltration of blood clots by cancer cells.
(A) Induction of a blood clot in the left tibia does not result in an increase in the number of infiltrating cancer cells compared to the right control tibia (CT) in the same mouse when cancer cells are injected 1 minute before blood clot induction using three cell lines (Breast cancer selected to home to the bone marrow: MDA-MB-231B/luc+; prostate cancer able to form bone metastases: PC3/luc+; and hepatoma cells not reported to form bone metastases: Huh-7). Cancer cells were injected 1 minute before clot induction. 24 hours later the bone marrow was isolated from the upper third of both tibiae and the number of cancer cells was evaluated by quantitative PCR of a cancer cell specific sequence and corrected to the total number of murine cells in the sample. n = 4–5/group. (B) The number of MDA cancer cells evaluated by flow cytometry was similar between the CT and hematoma group. MDA cancer cells were injected 1 minute before clot induction. 24 hours later the bone marrow was isolated from the upper third of both tibiae, red cells lysed, stained with a labeled human-specific CD49e (integrin α5) antibody and at least 3 million bone marrow cells were counted. n = 10 mice. (C) The use of surgical wax in the tibia following clot induction does not affect homing of cancer cells. 1 minute before clot induction MDA cancer cells were injected. The hole performed in the tibia in order to induce the blood clot in the bone marrow was either closed with surgical wax or left until bleeding stopped spontaneously (3–5 minutes) before closing the wound. n = 4 pairs. (D) Injection of MDA cancer cells 15 minutes before, 1 minute before and 5 minutes after clot induction did not affect the number of cancer cells in the bone marrow detected after 24 hours. Samples were prepared as in A. p = ns for each time point. (E) Evaluation of cancer cell numbers when injected 1 minute before clot induction did not reveal a difference in the number of cancer cells detected in the bone marrow at different time points (1, 4, 24 and 48 hours after clot induction). p = ns and n = 4–5 per time point. (F) Evaluation of cancer cell numbers when injected 5 minutes after clot induction showed a significant decrease in the number of cancer cells detected in the bone marrow at 1 hour after clot induction (p<0.05) but not at later time points (4, 24 and 48 hours after clot induction) (p = ns). n = 4–8 per time point.
Figure 5.
Bioluminescence and x-ray imaging in vivo.
Representative bioluminescence images from the last measurements before death and x-ray images at the time of death from 5 CT and 5 experimental mice in the group injected with cancer cells 5 minutes after clot induction. Upper panel represents the paired pictures from the CT group and the lower panel represents the paired pictures from the experimental group.
Figure 6.
Comparison of control and experimental groups.
(A) In the control group 13 lesions developed in the left tibia while in both experimental groups 11 developed in the left tibia. The percentages of mice with lesions in the left tibia is shown. n = 18, 14 and 17 mice. (B) The percentage of left tibia lesions in comparison to the total number of lesions is similar in both CT-experimental group pairs. n as in A. (C) Total bioluminescence signal per mouse expressed in relative light units (RLU) was similar between the control and the experimental groups at 7 weeks. n = 12, 11 and 15 mice. (D) Total bioluminescence signal per lesion was similar too. n as in C. (E) The size of the lytic lesions at 7 weeks was comparable. n as in A. (F) Survival curves of the control group (CT) and the experimental +5 min hematoma group shows that there is no difference in median survival (7.5 vs. 8.5 weeks, p = ns). n as in A. For ease of presentation CT was compared with the group with the seemingly larger median survival only.
Table 1.
Comparison of control and experimental groups.