Peripheral immune circadian variation, synchronisation and possible dysrhythmia in established type 1 diabetes

Aims/hypothesis The circadian clock influences both diabetes and immunity. Our goal in this study was to characterise more thoroughly the circadian patterns of immune cell populations and cytokines that are particularly relevant to the immune pathology of type 1 diabetes and thus fill in a current gap in our understanding of this disease. Methods Ten individuals with established type 1 diabetes (mean disease duration 11 years, age 18–40 years, six female) participated in a circadian sampling protocol, each providing six blood samples over a 24 h period. Results Daily ranges of population frequencies were sometimes large and possibly clinically significant. Several immune populations, such as dendritic cells, CD4 and CD8 T cells and their effector memory subpopulations, CD4 regulatory T cells, B cells and cytokine IL-6, exhibited statistically significant circadian rhythmicity. In a comparison with historical healthy control individuals, but using shipped samples, we observed that participants with type 1 diabetes had statistically significant phase shifts occurring in the time of peak occurrence of B cells (+4.8 h), CD4 and CD8 T cells (~ +5 h) and their naive and effector memory subsets (~ +3.3 to +4.5 h), and regulatory T cells (+4.1 h). An independent streptozotocin murine experiment confirmed the phase shifting of CD8 T cells and suggests that circadian dysrhythmia in type 1 diabetes might be an effect and not a cause of the disease. Conclusions/interpretation Future efforts investigating this newly described aspect of type 1 diabetes in human participants are warranted. Peripheral immune populations should be measured near the same time of day in order to reduce circadian-related variation. Graphical abstract Supplementary Information The online version contains peer-reviewed but unedited supplementary material available at 10.1007/s00125-021-05468-6.

Samples were fixed and data acquired on a LSR II flow cytometer equipped with 407, 488, 561 and 630nm lasers at the Indiana University School of Medicine Simon Cancer Center flow

Statistical Methods
Preparatory to analysis (data not shown), examination of ICC ("intraclass correlation coefficient", described below) values showed that percentage-based immune parameters tended to have higher ICC than count data indicating that percentage data has removed some of the within-subject circadian variation. This is a reasonable conclusion since some of the variation of the raw count data reflects between-subject variations in population numbers while the percent data factors that out, reporting results essentially as a count per standardized population count (i.e. population count=100) to yield a percent. As discussed below, we have established a rule for determining whether within-subject variation is "biologically significant" based on the ICC and therefore concluded that the use of percentage data would be a more conservative approach to selection of immune parameters for further study. We therefore restricted our primary circadian analysis to percentage data for the immune populations.
We followed study design protocols and analysis methodologies that are commonly found in human circadian studies, including sample size selection. Standardized versions of the variables were analyzed in order to increase the precision of circadian estimates by removing differences in subject-level dynamic ranges. For each subject, the measurements of each variable (cell populations, cytokine level and clock gene expression) were transformed into "z-scores" by subtracting the mean of the variable and dividing by the standard deviation of the variable. These values can then be intrepreted as the number of standard deviations from the mean and translated back to the original, untransformed values, using the means and standardard deviations of the variable.
Circadian rhythmicity patterns were estimated using "COSINOR" analysis 43 implemented in SAS (v9.4, Cary, NC) as a mixed linear model with random subject effect. Following Cornelissen 43 , a linear expression of a single phase Cosinor model of a value measured at time "t" is given by: In fitting the linear model above, we included a random effect for subject in order to account for the repeated measurements taken from each subject. This was accomplished using SAS Procedure "Mixed" using a "Variance Components" covariance structure. In addition, we tested whether cortisol or sex influenced circadian rhythmicity by fitting the linear model both with and without these covariates.
Results from the analysis of circadian clock genes are presented in ESM  Figure 1. Schematic of the study design. Adult volunteers with type 1 diabetes diagnosed for at least 12 months admitted for 24 hours at the Research Center University Hospital at IU. Blood samples were collected every 4 hr. at indicated times. Samples were aliquoted and one set processed immediately on site by the Flow Cytometry Resource Facility at IU. The other aliquots were shipped to the University of Florida for processing. Identified cell populations were tested for circadian rhythmicity and peak levels and times of peak and of trough were estimated for each cell population.

ESM Figure 2. Synchronization
Correlations of immune-related circulatory cell populations and cytokines after ordering on the sequence of their peak appearance during a 24-hour observation period. Correlations between pairs of immune variables whose peaks occur one after the other (indicating phase synchronization) are highlighted in bold text and border and appear above/below the diagonal. Significant (FDR-adjusted) correlations are indicated by shading. Red=negative, blue=positive correlation. Time of day of peak occurrence during the day of the study is indicated in the left-most column.