Time of day influences cytokine and clock gene response to immune stimulation in equine whole blood

Authors

  • O. F. McGlynn,

    1. Veterinary Sciences Centre, School of Agriculture, Food Science and Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland
    Search for more papers by this author
  • J. A. Browne,

    1. Veterinary Sciences Centre, School of Agriculture, Food Science and Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland
    Search for more papers by this author
  • C. M. Blake,

    1. Veterinary Sciences Centre, School of Agriculture, Food Science and Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland
    Search for more papers by this author
  • B. A. Murphy

    1. Veterinary Sciences Centre, School of Agriculture, Food Science and Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland
    Search for more papers by this author

B. A. Murphy, Veterinary Sciences Centre, University College Dublin, Belfield, Dublin 4, Ireland.
E-mail: Barbara.murphy@ucd.ie

Summary

The immune system demonstrates regularly recurring rhythmic variation that has important medical and veterinary clinical implications. Emerging evidence supports the existence of a bidirectional immune–circadian relationship during homeostatic challenge. In peripheral blood, circadian rhythmicity manifests in terms of clock gene expression, circulating levels and function of leucocyte cell populations and cytokine production. The involvement of interleukin 6 (IL6) in the transition from innate to acquired immunity is known, and inflammatory mediators can, in turn, phase shift peripheral molecular clocks. This study provides an initial investigation of diurnal variation in circadian clock and immune mediator response to antigenic challenge in the horse. Blood samples were collected at 4-h intervals at circadian times (CT) 0, 4, 8, 12, 16 and 20 from healthy Thoroughbred fillies maintained under winter photoperiod. Heparinized blood was cultured at 37 °C for 6 h with or without lipolysaccharide (LPS) (1 μg/ml). Cells were harvested by the addition of PAXgene™ RNA stabilizer solution to immediately preserve the RNA profile. Analyses of Taqman qPCR expression data reveal a significant CT X LPS interaction for the canonical clock genes PER2, CRY1, ARNTL, NR1D2 (P < 0.001, P < 0.05, P < 0.05, P < 0.01 respectively) and the immunomodulatory cytokine IL6 (P < 0.0001). LPS upregulated IL1B levels (P < 0.01), but there was no effect of CT. These results confirm that peripheral blood differentially responds to antigenic challenge over the 24-h cycle, impacting on our understanding of the pathophysiology of inflammatory responses. Further research in this area should provide valuable information regarding the optimum time for drug delivery and vaccination.

Equine chronobiology investigates how natural and artificially imposed environmental variation influences health and welfare in the horse (Murphy 2009). Cytokines play a critical role in the induction and regulation of immune effector functions, and these in turn are regulated by the circadian clock (Keller et al. 2009), which functions at the molecular level via autoregulatory transcription–translation feedback cycles of the core clock genes (Ripperger & Schibler 2001). It has been suggested that the direction of an immune response may depend on the cytokine environment at time of day of antigen presentation (Petrovsky & Harrison 1997). Light is the primary synchronizer of circadian (approximately 24 h) rhythms, but accumulating evidence highlights a bidirectional regulatory relationship between the immune and circadian systems (Coogan & Wyse 2008). Previously, equine core clock genes were identified (Murphy et al. 2007a), and it was shown that acute inflammation synchronized clock gene expression in equine peripheral blood in vivo (Murphy et al. 2007b). However, it was hitherto undetermined whether the observed coordinated upregulation of clock genes was time of day sensitive or simply an acute response to antigenic stimulus. Equine viral infections are responsible for disease epidemics around the world, resulting in major economic loss to the equine industry. This study aims to make initial inroads into unravelling the complex relationships between the immune and circadian systems in the horse, such that vaccine efficacy can be improved by choosing the optimum time of day for administration.

Four healthy Thoroughbred fillies (Equus caballus) (3–4 years of age) were housed under a light/dark (LD) cycle that mimicked the natural external photoperiod of early December (8 h L and 16 h D) at longitude W6.8, latitude N53.2 (County Kildare, Ireland). Blood samples were collected via jugular venipuncture at 4-h intervals beginning at lights on or circadian time (CT) 0. All procedures involving animals were approved by the Animal Research Ethics Committee of University College Dublin. Two hundred microlitres of heparinized blood was aliquotted into 24-well plates at each time point and treated with lipolysaccharide (LPS) (1 μg/ml) or left untreated. Plates were incubated at 37 °C with 5% CO2 for 6 h prior to harvesting by the addition of PAXgene™ RNA stabilizer solution (Qiagen), to immediately stabilize transcription and preserve the RNA profile. The success of this protocol was previously reported in human clinical studies (Carrol et al. 2007). Total RNA was isolated using the PAXgene™ Blood RNA kit, quality-checked using an Agilent 2100 Bioanalyzer (Agilent Technologies), and then 70 ng of RNA from each sample was converted to cDNA using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems) and stored at −20 °C. Taqman qPCR assays were designed using Genbank (NCBI) equine sequences and previously published assays to detect expression of the core clock genes PER2, CRY1, ARNTL (Murphy et al. 2006) and NR1D2 and the cytokines IL6 and IL1B (Quinlivan et al. 2007) (Table 1). Oligonucleotide primers were commercially synthesized by Eurofins MWG Operon and dual-labelled fluorescent probes by Biosearch Technologies. Each 20-μl qPCR  mixture contained 5 μl cDNA (1.5 ng RNA equivalents), 300 nm of each forward and reverse primer, 250 nm of probe, 10 μl Taqman Universal Master Mix (2×; Applied Biosystems) and nuclease-free H2O (Sigma-Aldrich) and was run on the ABI 7500 Real Time PCR System (Applied Biosystems). For each PCR run, a standard curve was generated using twofold serial dilutions of pooled cDNA. The PCR programme consisted of one cycle of 50 °C for 2 min and 95 °C for 10 min, followed by 45 cycles of 95 °C for 15 s and 60 °C for 1 min. All samples were run in duplicate. ACTB was selected for its stability as a reference gene using GeNorm.

Table 1.   qPCR primer and probe sequences for equine genes.
GeneAccession numberForward primer (5′-3′)Reverse primer (5′-3′)Probe (5′-3′)
IL6NM_001082496.1GGATGCTTCCAATCTGGGTTCAATTCCGAAAGACCAGTGGTGATTTTATCAGGCAGGTCTCCTG
IL1BNM_001082526CCGACACCAGTGACATGATGAATCCTCCTCAAAGAACAGGTCATTCATTGCCGCTGCAGTAAG
CRY1DQ988039.1CGGTTTGGGTGTCTGTCGTGTCCGCAGATGGGGTTTCCTTCCATTTTATCATGGGCAACTGTTATGGCGTGAATTTTTCTACACGGCAGCAAC
PER2EF015879.1CCAGCAAATATTTCGGAAGCATCGAGCCATCAGCAGCCAGACAGGAGCGAAAGCGAAGGTGGACGTGGACGGAAG
ARNTLNM_001081921CCACCAATCCATACACAGAAGCAAACTCTTCCCTCGGTCACATCCTCACCTCATTCTCAGGGCAGCAGATGGATTTTTGTTTGTCG
ACTBNM_001081838GCCGTCTTCCCCTCCATGCCCACGTATGAGTCCTTCTGGGCACCAGGGCGTGATGGTGGGC
NR1D2XM_001492478ACGGGTTCCTAAGAACATGGAGCAAGATGCTGCTGGCTCTCACTACAAACCATTGTGGCAGTGGGCTTAGCAGCCATT

Two-way repeated measures anova were conducted using graphpad Prism Version 4.0. The values of the relative expression of mRNA are presented as means ± SE, and a P-value of <0.05 was considered significant. A significant CT × LPS interaction for the canonical clock genes PER2, CRY1, ARNTL, NR1D2 (P < 0.001, P < 0.05, P < 0.05, P < 0.01 respectively) and the immunomodulatory cytokine IL6 (P < 0.0001) was observed (Fig. 1). There was a significant main effect of CT for PER2, CRY1 and ARNTL (P < 0.0001, P < 0.001, P < 0.01 respectively), supporting previous findings of a ‘phase shifting’ effect of inflammatory mediators on the circadian clock (Tsuchiya et al. 2005). The main effects of CT and LPS were both significant for NR1D2 (P < 0.01). LPS treatment significantly upregulated IL1B expression but was unaffected by time of day. These results support the view that the molecular clockwork differentially responds to an inflammatory stimulant over the 24-h cycle, which in turn regulates the transcriptional response of immune cells to an antigen (Keller et al. 2009).

Figure 1.

 Diurnal variation in gene expression from equine whole blood collected at 4-h intervals over the 24-h LD cycle and cultured for 6 h with (solid line) or without (dotted line) lipolysaccharide. Data are presented as means ± SE (n = 4 per timepoint). Within-time point significant Bonferoni statistical post hoc differences are indicated by superscript lettering; a, b = P < 0.05; c, d = P < 0.01; e, f = P < 0.001.

Interestingly, IL6 is differentially upregulated at CT 20 in equine circulation, in opposing phase to the temporal pattern observed in mice (Marpegan et al. 2009). This likely reflects a contrasting temporal immune surveillance regulation between diurnal and nocturnal species. Our finding suggests that equine Th1 humoral responses may be favoured when antigen exposure occurs in the evening. This has clear implications regarding the potential optimal time of day for vaccination in the horse, emphasizing the importance of further research in this area. In summary, this work highlights the complexity of circadian–immune regulation in a large diurnal mammal and potentially explains phenomena such as diurnal variation in resolution of inflammatory insult (Marpegan et al. 2009).

Conflicts of interest

The authors have declared no potential conflicts.

Ancillary