445 - Effect of early-life circadian rhythm disruption on myeloid cell reprogramming and immune response in neonatal mouse pups

Publication Number: 2433.445

Sarah G. Cioffi, Case Western Reserve University School of Medicine, Rocky Hill, CT, United States; Katherine Wen, Case Western Reserve University School of Medicine, Cleveland, OH, United States; Asha Thomas, Indiana University, Indianapolis, IN, United States; Lalitha Nayak, Indiana University School of Medicine, INDIANAPOLIS, IN, United States; Devashis Mukherjee, UH Rainbow Babies & Children's Hospital, Cleveland, OH, United States

Background: Circadian rhythm (CR) cycles are evolutionarily conserved mechanisms governed by the Earth's 24-hour (h) rotation and entrained by light. CR cycles drive oscillations of the levels of two core transcription factors, BMAL1 and CLOCK, which regulate >10K genes and most key physiologic processes, including the immune response. In utero, fetuses lack an intrinsic CR due to no light cues and are entrained by maternal circadian signals. Preterm birth disrupts this by transitioning from a light-dim in-utero environment to a light-intensive NICU environment during a critical time of immune development. It is unclear how this alters the preterm immune system's response to bacterial antigens. We have shown that the anti-inflammatory Krüppel-like factor 2 (KLF2) regulates neonatal responses to bacterial antigens. Postnatal day (P)-4 pups lacking KLF2 in myeloid cells (KO) have increased endotoxemic mortality (lipopolysaccharide, LPS) compared to controls, driven by pro-inflammatory circulating neutrophils.

Objective: To investigate (a) how disruption of external CR cues in early life alters neonatal immune development (b) whether KLF2 controls the neonatal immune response in a circadian-dependent manner.

Design/Methods: KO and control (Cre) mouse pups were randomized at birth to standard lighting (SC, 12h light/12h dark), constant light (LC), or constant dark (DC) till P4. Endotoxemia was induced at P4 by 5ug/gm LPS injection, and pups were followed for mortality or used for serum procurement. Peripheral blood was collected at P4 for flow cytometry for immune profiling.

Results: LC Cre pups show significantly more LPS mortality, and DC confers near 100% survival (Fig. 1). In contrast, KO pups have uniformly high mortality across all conditions, exceeding Cre arms, with no protective effect of DC conditions. In Cre pups, LC increases circulating young neutrophils (known pro-inflammatory cells) and decreases monocytes, compared to SC and DC. This CR disruption-dependent change in myeloid cell subsets is lost in KO pups, which have higher young neutrophils and significantly lower circulating monocytes across all conditions (Fig. 2), resembling LC Cre. Loss of BMAL1 in myeloid cells can lead to loss of circadian-dependent rhythmicity of the immune response and can increase LPS-induced mortality. Correspondingly, RNA-seq of adult Cre and KO neutrophils showed significantly reduced BMAL1 gene expression in KO (p=0.0043).

Conclusion(s): We show, for the first time, that neonatal CR disruption reprograms myeloid development and alters the immune response to bacterial antigens, and that the KLF2-BMAL1 axis regulates this.

Figure 1
Fig. 1. Mortality within 48 hours after intraperitoneal LPS injection (5ug/gm) on P4 after pups were housed in SC, LC, or DC conditions from birth (P0) till P4. Pups were continued in the same conditions for 48h after LPS. Whereas LC increases mortality significantly in Cre pups and DC confers an almost 100% protection, KO pups exhibit uniformly high mortality across all circadian conditions and higher mortality than Cre pups for all respective circadian conditions. N=20-25 pups per condition; Chi-square test for trend used for analysis. ns=not significant.

Figure 2
Fig. 2. (A) Percentage of circulating young neutrophils out of all CD45+ leukocytes in P4 Cre (control) and P4 KO mice under standard lighting (SC), constant light (LC), or constant dark (DC) conditions. P4 Cre mice show a significant increase in young neutrophils in LC compared to SC and DC. P4 KO mice show a significant increase in young neutrophils in both LC and DC compared to SC. Additionally, the P4 KO DC young neutrophil population shows a significant increase compared to P4 Cre DC, indicating a loss of DC protection from the inflammatory young neutrophils and increased mortality. One-way ANOVA with multiple comparisons was used. (B) Percentage of circulating monocytes in P4 Cre and P4 KO mice under the same circadian conditions. P4 Cre mice show a significant decrease in monocyte levels in LC compared to SC, and a considerable increase in monocyte levels in DC compared to LC. P4 KO mice show a reduction in circulating monocytes across all light conditions, and a significant decrease in monocyte levels in P4 KO DC compared to P4 Cre DC. One-way ANOVA with multiple comparisons was used. (C) Bone marrow neutrophil BMAL1 gene expression in adult Cre and KO mice. BMAL1 expression is significantly reduced in KO neutrophils compared to Cre controls (p = 0.0043); a T-test was used.