AUTHORS
Yichan Yang, Jeremy T Baeten, PhD2, Imene Tabet, PhD3, Yujie Zhang4, Elaheh Nasrollahzadeh, MD5, Jun Xia, PhD6, Marc Herman7, John Krais, PhD8, Daniel C. Link, MD9
AFFILIATIONS
1 Washington University in St. Louis, St. Louis, Missouri, United States
2 Washington University in St. Louis, St. Louis, Missouri, United States
3 Washington University in St. Louis, St. Louis, Missouri, United States
4 Washington University in St. Louis, St. Louis, Missouri, United States
5 Washington University in St. Louis, St. Louis, Missouri, United States
6 Washington University in St. Louis, St. Louis, Missouri, United States
7 University of Missouri, Kansas City, Missouri, United States
8 Washington University in St. Louis, St. Louis, Missouri, United States
9 Washington University in St. Louis, St. Louis, Missouri, United States
ABSTRACT
Poly(ADP-ribose) polymerase inhibitors (PARPi) have transformed the treatment of homologous recombination-deficient (HRD) cancers, particularly those with BRCA1 or BRCA2 mutations. However, hematologic toxicities, most notably anemia, remain among the most common and dose-limiting adverse events. These toxicities often lead to dose reduction, and treatment interruption and/or discontinuation. The clinical and molecular determinants of PARPi-induced erythroid toxicity are not well understood. Bi-allelic mutations of BRCA1 are associated with impaired homologous recombination (HR) and genomic instability. However, patients with germline heterozygous BRCA1 mutations retain one functional BRCA1 allele and are generally considered HR proficient. A previous study showed that reduced BRCA1 levels in BRCA1-haploinsufficient primary epithelial cells limit replication fork protection under replication stress. Of note, in addition to DNA damage, PARPi induce replication stress. Here, we explore the hypothesis that germline heterozygous pathogenic mutations of BRCA1 increase PARPi-induced erythroid toxicity by accentuating replication stress.
Germline Brca1+/− mice or wildtype (WT) littermates were treated with talazoparib (0.33 mg/kg/day) or vehicle alone for 3 weeks. We confirmed that this regimen completely suppressed PARylation by immunoblotting of bone marrow cells. At baseline, blood counts were similar in WT and Brca1+/− mice. However, treatment with talazoparib induced significantly greater anemia in Brca1+/− mice versus WT mice (RBCs: 4.73 x 106 per µL vs 6.21 x 106 per µL, p<0.05). Talazoparib significantly reduced total bone marrow cellularity in both WT and Brca1+/- mice without affecting spleen cellularity. Whereas the absolute number of HSCs and GMPs were preserved, talazoparib resulted in a marked decrease in MEPs and erythroid precursors. These reductions were consistently greater in Brca1+/- mice, reaching statistical significance with late erythroid precursors.
To characterize molecular mechanisms for the increased sensitivity to PARPi, we sorted various HSPCs subsets from WT or Brca1+/- mice 16 hours after a single dose of talazoparib. Talazoparib resulted in a significant increase in the percentage of γH2AX positive lineage- Sca1+ Kit+ (LSK) cells from Brca1+/- but not WT mice (WT: 12.1 ± 14.6 % vs Brca1+/-: 44.3 ± 35.0%, mean ± SD; p<0.0001). Likewise, talazoparib induced significantly more γH2AX foci in Brca1+/- MEPs (WT: 5.5 ± 4.2 vs Brca1+/-: 13.4 ± 11.1 foci/cell, p<0.0001). These data suggest that Brca1 haploinsufficiency increases PARPi-induced DNA damage in HSPCs. We next performed single-molecule DNA fiber assays on sorted MEPs to assess PARPi-induced replication stress. Briefly, sorted MEPs from vehicle alone or talazoparib treated mice were sequentially labeled with 5’-iodo-2’-deoxyuridine (IdU) for 20 min followed by labeling with second thymidine analog 5’-chloro-2’-deoxyuridine (CIdU) for additional 20 min. In vehicle-treated MEPs, no difference in replication fork progression rate was observed between WT and Brca1+/- cells. However, after talazoparib treatment, replication tract length was increased in both WT and Brca1+/- MEPs, with a significantly greater increase in Brca1+/- MEPs (WT: 13.2 ± 3.85 vs Brca1+/-: 14.4 ± 3.45 µM, p=0.0001). Prior studies have implicated BRCA1 and PARP1 in replication fork protection. Thus, we next treated cells with hydroxyurea following DNA labeling to assess replication fork protection. Indeed, Brca1+/- MEPs displayed impaired protection of nascent DNA compared to WT cells (ratio of CldU to IdU, WT: 1.16 ± 0.35 vs Brca1+/-: 1.02 ± 0.34, p<0.0001). Finally, since RAD51 participates in stalled fork protection, we measured RAD51 foci. Again, talazoparib induced increased RAD51 foci in MEPs, with a significantly greater increase in Brca1+/- MEPs (WT: 16.1 ± 13.3 vs Brca1+/-: 26.7 ± 19.1 foci/cell, p<0.0001).
Collectively, these data show that germline BRCA1 pathogenic mutations inhibit replication stress responses to PARPi, rendering erythroid progenitors more sensitive to PARPi-induced death. Studies to assess the impact of PARP1-selective inhibitors on hematopoiesis in WT and Brca1+/- mice are underway.
DISCLOSURES
Yichan Yang: No relevant conflicts of interest to declare.