Therapy resistance remains a major barrier to improving outcomes in HER2-positive breast cancer, with dormant tumor cells (DTCs) and cancer stem cells (CSCs) playing critical roles in recurrence and treatment failure. Herein, we investigated the interplay between dormancy and CSCs features in HER2-amplified breast cancer cell models and evaluated the role of the JAK1–STAT3 axis in sustaining these therapy-resistant phenotypes. Using an in vitro dormancy model induced by HER2 inhibition, we observed a reversible quiescent state characterized by decreased proliferation and viability, accompanied by a significant increase in the CSC marker CD44. CD44 expression was rapidly induced following HER2 inhibition, preceding measurable effects on cell viability, and persisted throughout the dormancy phase. CD44-positive populations showed reduced sensitivity to HER2 inhibition and displayed robust proliferative recovery upon therapy withdrawal. Functional studies revealed that the inhibition of JAK1, but not STAT3, impaired the recovery of CD44-positive populations and decreased their proliferative capacity, suggesting a critical role for JAK1 in maintaining the CSC phenotype during therapy. These findings underscore the importance of CD44 as a marker and mediator of therapy resistance and suggest that targeting CD44-positive cells or the JAK1 signaling axis could improve the efficacy of HER2-targeted therapies. Our study provides novel insights into the mechanisms underlying dormancy and CSC induction in HER2-positive breast cancer and highlights potential strategies to mitigate therapy resistance and prevent disease recurrence.

The type VI secretion system (T6SS) is a contact-dependent, contractile multiprotein complex widely distributed among Gram-negative bacteria. It mediates the translocation of effector proteins into bacterial competitors and eukaryotic host cells, promoting environmental fitness and contributing to virulence. In Salmonella, five pathogenicity islands encoding T6SSs (SPI-6, SPI-19, SPI-20, SPI-21, and SPI-22) have been described, along with an expanding repertoire of associated effector proteins. However, their global diversity and distribution remain incompletely resolved due to limited genomic sampling. To address this, we analyzed a curated dataset of 490 Salmonella genomes representing 45 serotypes. T6SS regions were identified using SecreT6, revealing that SPI-6 is widely distributed, whereas SPI-19, SPI-20, and SPI-21 are restricted to a subset of serotypes. SPI-20 and SPI-21 were exclusively found in S. enterica subsp. arizonae and diarizonae, while SPI-22 was absent from all analyzed genomes. All open reading frames within T6SS clusters were then analyzed for effector prediction and functional annotation. This approach recovered 32 out of 45 previously described T6SS effectors and identified several novel candidates. These included a cytidine deaminase with predicted DNase activity in SPI-6: two candidate nuclease effectors in SPI-19 with DNase and RNase activities, and four putative effectors in SPI-21, including enzymes with predicted peptidoglycan hydrolase activity, a potential inhibitor of eukaryotic ATPases, and a membrane pore-forming toxin. Additionally, a putative phospholipase effector was identified within a VgrG-associated genomic island in a subset of S. enterica subsp. diarizonae isolates. Collectively, these findings expand the known repertoire of Salmonella T6SS effector proteins and highlight their functional diversity.