Captopril mitigates splenomegaly and myelofibrosis in the Gata1 low murine model of myelofibrosis

Abstract Allogeneic stem cell transplantation is currently the only curative therapy for primary myelofibrosis (MF), while the JAK2 inhibitor, ruxolitinib. Has been approved only for palliation. Other therapies are desperately needed to reverse life‐threatening MF. However, the cell(s) and cytokine(s) that promote MF remain unclear. Several reports have demonstrated that captopril, an inhibitor of angiotensin‐converting enzyme that blocks the production of angiotensin II (Ang II), mitigates fibrosis in heart, lung, skin and kidney. Here, we show that captopril can mitigate the development of MF in the Gata1 low mouse model of primary MF. Gata1 low mice were treated with 79 mg/kg/d captopril in the drinking water from 10 to 12 months of age. At 13 months of age, bone marrows were examined for fibrosis, megakaryocytosis and collagen expression; spleens were examined for megakaryocytosis, splenomegaly and collagen expression. Treatment of Gata1 low mice with captopril in the drinking water was associated with normalization of the bone marrow cellularity; reduced reticulin fibres, splenomegaly and megakaryocytosis; and decreased collagen expression. Our findings suggest that treating with the ACE inhibitors captopril has a significant benefit in overcoming pathological changes associated with MF.

frequently be withdrawn due to side effects, such as anaemia, thrombocytopenia and infections. Thus, novel, non-toxic therapies are desperately needed for this molecularly heterogeneous disorder. Primary MF is characterized by abnormal megakaryocytes, aberrant cytokine production and bone marrow failure with extramedullary haematopoiesis. 5 Stem cell-derived myeloproliferation and abnormal cytokine production lead to the dysregulation of megakaryocytes and fibrotic remodelling of the bone marrow. 6 The degree of collagen fibrosis in the bone marrow can be correlated with the severity of primary MF. 6 Several genetically engineered mouse models based on JAK2, MPL or CALR mutations are available to study MF. [7][8][9] Patients with idiopathic MF were found to harbour reduced levels of the transcription factor GATA1 in megakaryocytes. 10 GATA1 is a haematopoietic master transcription factor that provides regulation for both erythroid and myeloid lineages. 11 Due to a deletion in the hypersensitive site of its promoter, which drives its transcription in megakaryocytes, GATA1 deficiency results in aberrant megakaryocytopoiesis resulting in hyperproliferative progenitors, defective terminal differentiation, impaired erythropoiesis and transient anaemia. 11,12 The Gata1 low mouse strain has been especially useful to study MF because fibrotic remodelling of the bone marrow microenvironment also occurs. 13,14 A final common pathway that leads to MF is thought to involve aberrant regulation of TGF-b1 and the subsequent deposition of reticulin and collagen. 15 Recent work suggests that malignant and non-malignant cells cooperate in this inflammatory process and subsequent fibrosis and that fibrocytes may play an important role in this process. 16,17 However, the identity of the cell types and the inflammatory cytokines directly responsible for myelofibrotic remodelling are not known, but might be important in developing more effective, non-transplant therapies.
A number of studies have demonstrated the role of Ang II in fibrotic remodelling of the lung, heart, kidney, skin and liver. [18][19][20][21] It has been demonstrated in a number of animal models that inhibitors of angiotensin-converting enzyme (ACE) can block or reverse fibrotic remodelling through the reduction in Ang II maturation. [22][23][24][25][26] Therefore, we hypothesized that captopril, an ACE inhibitor, could reverse MF. We tested this hypothesis in the Gata1 low mouse model of primary MF.

| Chemicals
Reagents were obtained from Sigma-Aldrich (St. Louis, MO) except where indicated.

| Animals and captopril treatment
All animal handling procedures were performed in compliance with guidelines from the National Research Council for the ethical handling of laboratory animals and were approved by the Uniformed Services University of the Health Sciences Institutional Animal Care and Use Committee. Male and female Gata1 low and wild-type CD1 mice were purchased from Jackson Laboratories (Bar Harbor, ME).
Quantitative PCR confirmed low expression of Gata1 (results not shown). The mice were crossed to a CD1 background as previously described to establish a line of homozygous mutant mice. 14 Mice were kept in a barrier facility for animals accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. Mice were housed in groups of four. Animal rooms were maintained at 21 AE 2°C, 50% AE 10% humidity and 12-hour light/dark cycle with commercial freely available rodent ration (Harlan Teklad Rodent Diet 8604, Frederick, MD, USA). Captopril (USP grade; Sigma-Aldrich, St Louis, MO, USA) was dissolved in acidified water at 0.6 g/L and provided to animals starting at 10 months of age until 12 months of age, as previously described. 27 An earlier study established the stability of captopril in acidified water. 28 Based on previously measured volumes of water consumed per day by the mice, we determined that daily water consumption resulted in a dose of 79 mg/kg/d. 27 Control animals received acidified water (vehicle) without captopril. Animals were killed at 13 months of age.

| Blood cell analysis
Complete blood counts (CBC) with differentials were obtained using a Baker Advia 2120 Hematology Analyzer (Siemens, Tarrytown, NY, USA). Separate mice were used for each point (n = 5-6 per group).

| Histology and myelofibrosis scoring
Sternebrae, humeri and femurs were surgically removed from killed animals and fixed in 10% neutral formalin overnight. Tissues were paraffin blocked and stained using standard methods for haematoxylin and eosin (H&E), Masson's trichrome and Gomori reticulin stain by Histoserve (Germantown, MD). Stained slides were evaluated by a pathologist who was blinded to the identity of the treatment groups using a published system for scoring MF. 29 Bone marrow sections were digitally scanned using the Zeiss Axio Scan and images for publication were produced with Zen Lite software (Carl Zeiss, USA).

| Reverse transcription polymerase chain reaction (RT-PCR)
Total RNA was extracted from cells isolated from bone marrow or spleen cells using phenol-chloroform extraction with silicone lubricant using a modified protocol. 30

| RESULTS
To determine the efficacy of captopril in reversing MF, we evaluated morphologic and phenotypic changes in the Gata1 low mouse model. Untreated Gata1 low mice at 13 months of age exhibited classic features of marrow MF as compared to wild-type CD1 mice ( Figure 1A Figure 1D,E). Consistent with previous reports of splenomegaly in Gata1 low mice, we observed that the splenic weight was increased sixfold in untreated Gata1 low mice as compared to wt CD1 mice (P value < .05). Captopril treatment for 2 months induced a~2-fold decrease (P < .05) in splenic weight in Gata1 low mice as compared to untreated Gata1 low mice ( Figure 1F). Peripheral blood counts were studied in captopril-treated and untreated Gata1 low mice and their wild-type littermates. As shown in Figure 2  Flow cytometric analysis also showed that Gata1 low mice had a trend towards higher levels of splenic megakaryocytes as compared to wt CD1 mice ( Figure 3F), although this did not reach significance.
We observed a~2-fold decrease in the frequency of megakaryocytes as a percentage of total live cells in response to captopril treatment (P < .05). This decrease in megakaryocytes as determined by FACS was also reflected in qRT-PCR detection of CD41 and CD61 markers, which decreased~6-fold and~5-fold, respectively, in captopril-treated Gata1 low mice (P < .05) ( Figure 3G Our study also demonstrated a marked reduction in abnormal megakaryoctyes in the Gata1 low mice after captopril treatment. Ang II, as a part of the renin-angiotensin system, is a master regulator of blood pressure and blood volume homeostasis. 56 This system has also been demonstrated to regulate cell proliferation and F I G U R E 3 Effects of captopril administration of megakaryocytes and collagen. Wild-type (wt) or Gata1 low mice were treated from 10 to 12 mo with either captopril 72 mg/kg/d or vehicle in drinking water. Mice were killed at 13.5 mo, and tissues were harvested. (A) Flow cytometric analysis was performed on femur bone marrow cells to measure percentage of CD45 + cells expressing CD41 + cells. Representative FACS data are presented for wt, Gata1 low -untreated mice and Gata1 low captopril-treated mice. (B-E) qPCR of mRNA isolated from bone marrow of Gata1 low mice treated AE captopril, as described above. Interrogated transcripts were CD41, CD61, Col 1a and Col 3a. Data show Gata1 low qPCR transcript levels from untreated mice compared to the ratio of Gata1 low transcript levels from mice treated with captopril relative to Gata1 low -untreated mice. *P value .05. (F) Flow cytometric analysis was performed on spleen-derived cells to measure percentage of CD45 + cells expressing CD41 + cells. Representative FACS data are presented for wt, Gata1 low -untreated mice and Gata1 low captopril-treated mice. (G-J) qPCR of mRNA isolated from spleens of Gata1 low mice treated AE captopril, as described above. Interrogated transcripts were CD41, CD61, Col1a and Col3a. Data show Gata1 low qPCR transcript levels from untreated mice compared to the ratio of Gata1 low transcript levels from mice treated with captopril relative to Gata1 low -untreated mice. *P value .05 differentiation of specific haematopoietic lineages. 57 Ang II was shown to be required for normal myelopoiesis and erythropoiesis. 58 ACE knockout mice, in which Ang II levels are 10-fold lower than in wt mice, have several myelopoietic abnormalities resulting in a reduction in normal, mature macrophages and have an accumulation of myeloblasts and myelocytes. 59 Additionally, Ang II peptide administration in mice was shown to increase levels of megakaryocyte precursors and megakaryocytes in the blood after radiation exposure. 60 Findings from our laboratory and others indicated that captopril increased survival from radiation-induced haematopoietic injuries suggesting that ACE inhibition can also reduce injuries to the haematopoietic system. 27,61-63 ACE inhibitors were also shown to cause a reduction in granulocyte colony-forming and erythroid burstforming units which were accompanied by an increase in undifferentiated cells, including granulocyte, erythroid, macrophage and megakaryocyte colony-forming units (CFU). 58,64 Investigation of the direct effects of Ang II on bone marrow colony formation demonstrated that the addition of Ang II to bone marrow cultures resulted in the stimulation of immature CFU granulocyte/macrophage and CFU granulocyte/erythrocyte/monocyte/megakaryocyte under panmyeloid culture conditions. 65 However, it was later demonstrated that the addition of Ang II did not affect CFU megakaryocyte colony formation in a lineage assay in culture. 59 Captopril's ability to reverse fibrosis in this murine model is novel and future studies are needed to assess its feasibility for clinical use.
The JAK2 inhibitor ruxolitinib reduces splenic haematopoiesis but does not reverse MF in the Gata1 low mice, 66 and ruxolitinib is currently approved by the Food and Drug Administration (FDA) only for palliation of splenomegaly and MF-associated symptoms. Results of several clinical trials have thus far failed to demonstrate its reversal of fibrosis. 67 Because captopril is a FDA-registered drug with widespread use, low cost and little toxicity, our studies provide compelling evidence to initiate a phase I/II trial in patients with primary MF aimed at reducing marrow fibrosis, replacement blood product usage and MF-associated symptoms. The human equivalent dose to 110 mg/kg/d captopril (0.55 g/L in the drinking water) is~330 mg/ d. 68 Captopril's maximally tolerated dose of 500 mg/d , 69 which makes our dosage feasible. Our initial treatments with captopril were based on our findings of prevention of bone marrow injury by total body irradiation in mice. 27,63 We have since found that reduction in captopril levels to as low as 13 mg/kg/d is sufficient for the prevention of radiation-induced bone marrow injury in mice (R.M. Day, unpublished findings). We wish to repeat our work in the Gata1 low myelofibrosis model also using this reduced dosage of captopril. In addition, we are currently investigating the molecular mechanism of captopril-mediated reduction in fibrosis and identifying the cytokine (s) responsible.