A Transgenic Mouse Model for Adult Acute Chest Syndrome.

Acute chest syndrome (ACS) is a common cause of death amongst adults with sickle cell disease and a significant threat to children; however, there is no animal model that will allow pathogenic mechanisms and potential interventions to be tested. To help develop an animal model, we used the observations of Vichinsky et al that, in adults, ACS episodes frequently follow vaso-occlusive crisis (VOC) and are characterized by fat and bone marrow embolism at autopsy and/or on bronchoalveolar lavage. Fat emboli may be acted on by PLA₂ generating free fatty acids that have been demonstrated to elicit pulmonary pathology similar to acute respiratory distress syndrome (ARDS) in animals treated with oleic acid. We have tested several protocols for producing a mouse model of ACS. Initial studies utilizing injection of a mixture of bone marrow and mouse fat (BM/fat) into the tail vein of sickle transgenic BERK mice with HbF expressing exclusively human hemoglobins (BERK- γ) resulted in variable pulmonary pathology characterized by increased congestion (red cell retention), cellularity (WBC retention), and wall thickness (edema). Serum levels of VCAM-1 and creatine kinase, as well as markers of hemolysis were increased concomitantly in some animals. A more consistent and statistically significant response was obtained by first subjecting the animals to one hour of hypoxia (8% O₂) followed by one hour of re-oxygenation in room air (hypoxia/reperfusion) prior to injection of a mixture of BM/fat into the tail vein. In this study, three conditions were evaluated: no treatment (NT), hypoxia/reperfusion only (H), and hypoxia/reperfusion followed by BM/fat injection (HBF). Three independent observers evaluated all specimens. At baseline (NT), the BERK- γ had significantly more congestion than the C57BL6 mice, p<0.001. C57BL6 mice subjected to H-only had a significant increase in cellularity (p>0.0008) compared to baseline, and those subjected to HBF had both increased cellularity (p>0.0001) and wall thickness (p>0.02). Congestion did not increase in C57 mice under either treatment condition. BERK- γ mice had a statistically significant increase in cellularity (p>0.0001), congestion (p>0.01), and wall thickness (p>0.003) after HBF compared to baseline, but the response to H-only was not statistically significant. Cellularity and congestion were significantly correlated in both C57 (p>0.005) and BERK- γ mice (p>0.0001), but the increase in congestion vs increase in cellularity (slope) was twice as large for BERK- γ (0.55) mice as for C57 (0.23), suggesting that congestion was the more prominent pathologic feature. After HBF, immunostaining of the BERK- γ lungs for P-Selectin demonstrated increased expression in individual platelets (p<0.03), platelet aggregates (p<0.04), and the vascular endothelium (p<0.04). In conclusion, we found that both hypoxia/reperfusion and hypoxia/reperfusion with fat/BM increased cellularity but not congestion in C57BL6 mice. In contrast, in BERK- γ mice, hypoxia/reperfusion followed by fat/BM injection increased both cellularity and congestion, suggesting that this is a successful model of ACS. We speculate that platelet aggregation, release of P-Selectin and leukocyte recruitment may initiate an inflammatory response in this model and represent one aspect of the observed pathology.