Changes in Cytokine and Cytokine Receptor Levels During Postnatal Development of the Human Dorsolateral Prefrontal Cortex

In addition to their traditional roles in immune cell communication, cytokines regulate brain development. Cytokines are known to in�uence neural cell generation, differentiation, maturation, and survival. However, most work on the role of cytokines in brain development investigates rodents or focuses on prenatal events. Here, we investigate how mRNA and protein levels of key cytokines and cytokine receptors change during postnatal development in the human prefrontal cortex. We �nd that most cytokine transcripts investigated (IL1B, IL18, IL6, TNF, IL13) are lowest at birth and increase between 2–5 years old. After 5 years old, transcriptional patterns proceeded in one of two directions: decreased expression in teens and young adults (IL1B, p = 0.002; and IL18, p = 0.004) or increased mean expression with maturation, particularly in teenagers (IL6, p = 0.004; TNF, p = 0.002; IL13, p < 0.001). In contrast, cytokine proteins tended to remain elevated after peaking signi�cantly at age 5 (IL1B, p = 0.012; IL18, p = 0.026; IL6, p = 0.039; TNF, p < 0.001), with TNF protein being highest in young adults. Early developmental increases in cytokines were paralleled by increases in their receptor binding subunits, such as IL1R1 (p = 0.033) and IL6R (p < 0.001) transcripts. In contrast, cytokine receptor-associated signaling subunits, IL1RAP and IL6ST, did not change signi�cantly between age groups. Of the two TNF receptors, the ‘pro-death’ TNFRSF1A and ‘pro-survival’ TNFRSF1B, only TNFRSF1B was signi�cantly changed (p = 0.028), increasing �rst in toddlers and again in young adults. Finally, the cytokine inhibitor, IL13, was elevated �rst in toddlers (p < 0.001) and again in teenagers (p < 0.001). While the mean expression of interleukin-1 receptor antagonist (IL1RN) was highest in toddlers, this increase was not statistically signi�cant. The �uctuations in cytokine expression reported here support a role for increases in speci�c cytokines at two different stages of human cortical development. The �rst is during the toddler/preschool period (IL1B and IL18) and the other at adolescence/young adult maturation (IL6 and TNF).

However, activation of cytokine signaling alone, in the absence of infection, can induce pathological changes in the brain (24)(25)(26).Additionally, genetic variation in immune-associated signals can alter the expression and function of immune signals independent of pathogen exposure.These were described with the copy number variant of complement component C4 (27).Thus, genetic and epigenetic changes to immune signaling molecules have the potential to disrupt normal patterns of expression and may exhibit heightened responses during episodes of infection-associated and/or sterile in ammation.Furthermore, psychiatric disorders rst become evident during postnatal development.Therefore, we aimed to characterize the developmental timing of typical cytokine signaling during the normal postnatal period to determine when immune signaling perturbations would be expected to impact cortical development.
We previously investigated the trajectory of complement pathway members in neurotypical human prefrontal cortex across major periods of post-natal brain development.We found that the peak expression of many complement pathway members was in toddlers 2-5 years old (28).However, this study did not examine whether an increase in pro-in ammatory and anti-in ammatory cytokines occurs in tandem with complement pathway activation or whether the trajectory of cytokine change is distinct.
Therefore, given this existing association between complement and cytokine induction.We would expect pro-in ammatory cytokines and their associated receptors to coordinately regulated with complement during development.
To date, the studies investigating cytokine expression in human postnatal brain have been predominantly restricted to adults.Therefore, there is a gap in our understanding of immune expression in the brain at earlier, more dynamic ages.These include age periods coinciding with the typical age of onset for neurodevelopmental disorders such as autism spectrum disorders (ASD) (66, 67), which is concomitant with a peak cortical expression of the complement pathway (28).Additionally, abnormal cytokine levels have been implicated in ASD (68-70).Aberrant cytokine signaling could disrupt numerous processes during postnatal brain development.These include interneuron migration (71,72), developmentally programmed cell death (73-76), neuronal maturation (77-80), synaptogenesis (81, 82), synaptic pruning (83-87), and oligodendrocyte maturation/myelination (88-90).Therefore, for the rst time, we quanti ed the normal human cortical expression of an array of pro-in ammatory and anti-in ammatory cytokines implicated in neurodevelopment and characterized their expression trajectory from near birth until early adulthood using both mRNA and protein-based methods.We hypothesized that the mRNA and protein of pro-in ammatory cytokines IL1β, IL6, IL18 and TNF, and their associated receptors, would follow a similar trajectory to that of complement, peaking during early postnatal human brain development.Additionally, we hypothesized that the expression of associated inhibitory cytokines IL1RN, IL10, and IL13 would increase from school-age children through young adult, comparable to complement inhibitors (28).

Subjects
Postmortem samples from Brodmann's area 46 in the DLPFC were obtained from the University of Maryland Brain and Tissue Bank in conjunction with the National Institute of Child Health and Development (N=65).The individuals in this cohort ranged from 6 weeks old to 25 years old.Subjects (neonates, infants, toddlers, school-age children, teenagers, and young adults) were assigned to one of six age groups based on cognitive development stages in consultation with a pediatrician.These developmental groups roughly correspond with stages of sensory and cognitive development, i.e., visual acuity and grasping (Neonates: birth-3 months old); sitting upright and environmental engagement (Infants: 4 months-1 year old); preoperational, pre-conceptual rapid language development (Toddlers: 1.5-4 years old); concrete operations (School-age children: 5-12 years old); formal operations, reproductive competence and heightened peer interactions (Teenagers: 14-18 years old); social independence and early maturity (Young Adults: around 22 years old).Eight to thirteen samples were selected per developmental group as described: groups were established with care to balance male and female samples as best as possible, and groups were matched for RNA integrity number (RIN), pH, and postmortem interval (PMI) for most groups.However, PMI was signi cantly longer in neonates and infants than in older ages (ANOVA, F(5,51)=2.881,p=0.023).Age groups analyzed, and their respective demographic and tissue quality information, are reported in Table 1.We have also included several Sudden Infant Death Syndrome (SIDS) samples in our age groups under two years old.A role for in ammation in SIDS has been proposed (91).However, the syndrome remains poorly understood, with much potential for heterogeneous causes of death.We studied ages 25 and under to focus on developmental processes, rather than aging (92) -however, the time at which "aging" begins remains an area of debate (93)(94)(95)(96).All subjects were free of neurological and psychiatric symptoms at the time of death.Individual sample information, including age, sex, RIN, pH, PMI, and cause of death, are in Supplementary Table 1.
No signal in the no-template control or the no-RT control exceeded cycle threshold levels for any qPCR run.
The homogenate was then incubated on wet ice for an additional ten minutes.Microcentrifuge tubes of homogenate were then centrifuged for 10 min at 10,000g at 4°C to pellet tissue debris and insoluble material.The supernatant was removed and stored in 1.5 ml low-protein binding centrifuge tubes (Fisher Scienti c, Hampton, NH, USA, Cat# E925000090) at -80°C.Before use, the protein was quanti ed by bicinchoninic acid (BCA) assay.All samples were then diluted in N-PER with Halt protease and phosphatase inhibitor cocktail to a concentration of 2 mg/mL.

Multiplex immunoassay
For cytokine protein expression, multiplex immunoassays were performed.

Statistical analysis
Statistical tests of qPCR and multiplex immunoassay results were performed using SPSS statistics (Version 25, OSX, IBM, Armonk, NY, USA).Outliers by age group were identi ed using the ROUT test for outliers using GraphPad Prism (GraphPad Prism version 9.3.1 for Windows, GraphPad Software, San Diego, CA, USA) and then removed (~ 2 outliers per transcript).For TNF mRNA, 5 out of 11 samples in the neonate and infant groups were below the level of detection.Since the visual inspection of the data suggested the neonates and infants had similar levels of TNF mRNA (when detectable), the neonate and infant groups were combined for this analysis.Outliers were then removed using ROUT analysis.Data were tested for normality using Kolmogorov-Smirnoff testing and for homogeneity of variance using Levene's test.If the data were non-normally distributed, data were logged, and Cox Box transformed (Real Statistics Excel plug-in, Release 5.4.Copyright 2013 -2018 Charles Zaiontz.www.real-statistics.com).
Data were tested for correlation with pH, RIN, and PMI.
Data without covariates that were normally distributed were analyzed using ANOVA, and non-normally distributed data were analyzed using the Kruskal-Wallis test.Normally distributed data with covariates were analyzed by ANCOVA.Non-normally distributed data with covariates were analyzed using Quade's ranked test.A signi cant ANOVA, ANCOVA, Kruskal-Wallis, or Quade's ranked test was followed by a Fisher's Least Signi cant Difference (LSD) posthoc test.Welch's ANOVA was used for normally distributed data exhibiting non-homogenous variance between groups (99, 100), followed by the Games-Howell posthoc test.Sex differences were analyzed by independent sample 2-tailed t-tests for each transcript but could not be compared between age groups due to sample size limitations.All transcripts were tested for correlation with age under the age of 5 years old using Spearman's ρ.Statistical signi cance was set at p≤0.05.

Results
Increases in IL1B mRNA and protein early in life IL1B mRNA levels (Fig. 1a) were at a nadir just after birth in the neonatal prefrontal cortex and increased during postnatal life (F(5,46) = 4.451,p = 0.002).For IL1B mRNA, levels increased approximately 95% between neonates and infants (p = 0.038) and almost 380% between infants and toddlers (p = 0.006), ultimately peaking with an 830% net increase from the lowest age group (Neonates) to the highest (Toddlers) (Fig. 1a).When examining ages ranging from 2 months to 5 years old more speci cally, IL1B mRNA was found to be signi cantly positively correlated with age (Spearman's rho, ρ = 0.580, p < 0.001) (Table 2).Mean IL1B mRNA levels were then reduced back to roughly infant levels in school-age children but were still signi cantly elevated compared to neonates through school-age children (p = 0.022) and adolescence (p = 0.008).IL1B mRNA levels in young adults were intermediate compared to earlier ages and not signi cantly different from any other age group.
Analysis of protein also revealed signi cant increases in IL1B (Fig. 1b) levels across postnatal human life (F(5,52) = 3.265, p = 0.012) and resembled mRNA expression patterns overall.IL1B protein was at its lowest level in neonates and showed an approximate 70% increase in levels between neonates and infants (p = 0.008) (Fig. 1b).IL1B subsequently remained elevated, with toddler (p = 0.002), teenager (p = 0.004), and young adult (p = 0.005) levels of IL1B protein signi cantly higher than those for neonates.A signi cant linear increase in IL1B protein was detected in earlier age groups (ages 2 months old-5 years old) (Spearman's rho, ρ = 0.479, p = 0.006) (Table 2).Overall, the magnitude of changes in protein across age was not as great as those in the mRNA, never exceeding a 140% increase in the mean between any two groups.However, there were some interesting differences in the pattern of change between the IL1B protein and mRNA, in that the plateau seen between ages 5-18 years old in the mRNA seemed to extend in the IL1B protein into the young adult group, whereas the mean IL1B mRNA levels dropped in the young adult group.In fact, the young adult IL1B protein level was still signi cantly increased by 100% compared to the IL1B protein level in neonates (p=0.005).
Increases in IL6 mRNA and protein both early in life and at maturation Similar to IL1B, IL6 mRNA increased gradually, starting with low levels in neonates, which stabilized in the toddler age group and were sustained later in development.The rst signi cant increase in IL6 mRNA by age group (F(5,52) = 3.265, p = 0.012) was a 380% increase from neonates to toddlers (p = 0.003) (Fig. 2a), and IL6 mRNA was moderately correlated with age within the rst 5 years of life (Spearman's rho, ρ = 0.419, p = 0.033) (Table 2).After this, IL6 mRNA showed an additional signi cant 410% increase in the expression of teenagers compared to infants (p < 0.019) and a 520% increase compared to neonatal levels.There was a slight subsequent reduction in IL6 mRNA in young adults, which returned the mean expression to intermediate levels, increased by only 140% compared to neonates (p = 0.010).
Like IL6 mRNA, IL6 protein increased gradually early in life before plateauing in toddlers.However, IL6 protein had an overall greater magnitude of mean change than IL6 mRNA.As with IL6 mRNA, IL6 protein was rst signi cantly elevated in the toddler age group (F(5,44) = 2.585, p = 0.039), with an approximately 680% increase from neonatal levels (p = 0.006).Mean IL6 protein levels did start to increase in infants, which is re ected in the positive correlation of IL6 protein with age between 2 months and ve years old (Spearman's rho, ρ = 0.474, p = 0.011) (Table 2).While the early life IL6 mRNA increase was re ected in the IL6 protein level, the second period, where we found the highest mean IL6 mRNA in teenagers, was not.Mean protein level showed no signi cant difference between school-age children and teenagers (p = 0.514) (Fig. 2b).Thus, the highest IL6 protein levels were detected during the school-age period, which did increase by a large amount, 1200%, compared to neonates (p = 0.004).
In contrast to IL1R1 and IL6R, IL6ST mRNA did not change signi cantly across development (F(5,49) = 2.284, p = 0.061) (Fig. 2d).In spite of the modest and non-signi cant changes in gene expression represented by the group means, IL6ST mRNA was signi cantly correlated with age under 5 years old, suggesting that there is an early steady increase in expression for all three of the IL1/IL6 receptors studied (ρ = 0.498, p = 0.004) (Table 2).
Increases in IL18 mRNA and protein, but not IL18R, with age As with IL1B and IL6, IL18 mRNA levels were lowest just after birth in neonates and increased early in postnatal life (F(5,48) = 4.053, p = 0.004) (Fig. 3a).Like IL1B, IL18 mRNA then decreased in school-age children.IL18 mRNA decreased once more in young adults.Thus, IL18 mRNA peaked in toddlers with a signi cant 200% increase from neonates (p < 0.001), with mean IL18 mRNA levels in infants falling between neonates and toddlers (p = 0.229).IL18 mRNA was also signi cantly increased in toddlers by 90% when compared to infants (p = 0.001) (Fig. 3a).When focusing on the three youngest developmental groups, a fairly strong linear correlation between age and IL18 mRNA was detected between 2 months old and 5 years old (Spearman's rho, ρ = 0.580, p < 0.001) (Table 2).The mean level of IL18 mRNA then decreased slightly in school-age children where it was still signi cantly elevated by 130% compared to neonates (p = 0.025).Mean IL18 mRNA levels were reduced to approximately neonatal levels in the young adult group (p = 0.129).
IL18 protein levels followed a similar pattern as their encoding transcript, increasing during postnatal life (Fig. 3b) (F(5,49) = 2.825, p = 0.026).Unlike IL18 mRNA, IL18 protein remained stably elevated in the teenager and young adult groups compared to neonates and infants (all p > 0.05).IL18 protein was elevated in toddlers by 50% compared to neonates (p = 0.046) and by 130% compared to infants (p = 0.043).There was a slight mean decrease in the school-age group which was not signi cantly different from any other age group (all p > 0.05).
While mean levels of IL18R1 mRNA appeared to increase over the age period studied, the change in transcript levels did not differ signi cantly with age (F(5,42) = 1.957, p = 0.105) (Fig. 3c).The levels of IL18R1 mRNA were quite variable at every age.Unlike other cytokines and receptors, IL18R1 mRNA was not signi cantly correlated with age when only considering those under 5 years old (Spearman's rho, ρ = 0.254, p = 0.232) (Table 2).
Increases in TNFΑ and TNFR2, but not TNFR1, with age As with the three other cytokines studied, TNF mRNA levels were lowest earlier in postnatal life and signi cantly increased during development (F(4,40) = 5.104, p = 0.002) (Fig. 4a).TNF mRNA increased 360% between the neonate + infant group and the toddler group (p = 0.044).When focusing on changes within the rst few years of life, no signi cant correlation was found between TNF mRNA and age (Spearman's rho, ρ = 0.421, p = 0.057) (Table 2).Similar to IL18, TNF decreased by 64% from toddlers to school-age children, where it was not signi cantly different from neonatal levels (p = 0.650).Then, in a pattern distinct from all three other cytokines investigated, TNF mRNA increased again in teenagers, exceeding even toddler levels by 100% (Fig. 4a) (p = 0.046).
Levels of TNF protein resembled TNF mRNA levels during post-natal life, increasing signi cantly in toddlers and remaining elevated into young adulthood (Fig. 4b) (F(5,50) = 7.722, p < 0.001).The uctuations found in TNF mRNA expression levels between toddlers, school-age children, and teenagers could not be detected in TNF protein levels, but mean levels were elevated in toddlers, school-age children, teenagers, and young adults compared to neonates (All p < 0.001) and infants (Toddler, p = 0.007; schoolage children, p = 0.008; teenager, p < 0.001, young adult, p = 0.004) Fig. 4b).When focusing on changes within the rst few years of life, a strong positive correlation was found between TNF protein and age (Spearman's rho, ρ = 0.555, p < 0.001).In agreement with TNF mRNA, mean TNF protein levels peaked in teenagers, which were increased by almost 430% compared to neonates (p < 0.001).

Discussion
In this study, we have characterized changes in mRNA and protein expression of cytokine and cytokine receptors in the developing human prefrontal cortex, with su cient power to allow a quantitative rather than qualitative analysis.All cytokines studied were expressed in the post-natal human brain as both mRNA and protein (excepting the very little to no expression of TNF mRNA in some neonatal prefrontal cortices).We found that both transcript and protein levels of almost all pro-in ammatory cytokines were characterized by a steady increase under age 5, which was consistent with complement pathway factors previously reported in the same cohort (28).Given what we know about the interplay between complement and cytokines, we interpret that this concomitant increase could mean that these components of the immune system operate together or in a coordinated fashion to regulate normal development.However, following the initial increase in toddlers, the four pro-in ammatory cytokines studied here diverged in expression during brain maturation.IL1B and IL18 decreased in transcription in the school-age children and young adult age groups as predicted, similar to the changes previously seen in complement expression (28).In contrast, IL6 and TNF expression, increased a second time during adolescence.Thus, we saw peaks in cytokine expression during two distinct periods of development.
Previously, we found that complement inhibitor expression increased after complement peaked in toddlers, and that complement expression subsequently decreased (28).In contrast, here we found that the inhibitory cytokines IL1RN or IL10 were not as dynamically expressed across development as the inhibitors of complement we previously characterized.The only cytokine inhibitor to show any change in expression was IL13.As IL13 inhibits proin ammatory cytokines (101)(102)(103), the increase in IL13 mRNA in adolescence may represent regulatory pressure on proin ammatory cytokine expression in that age group.However, IL13 can also regulate the effects of complement-mediated lysis in the periphery by inhibiting the downstream effects of the membrane attack complex (104), raising the possibility that elevated IL13 may be more involved in the regulation of complement pathway activity than in IL6 or TNF signaling.
As with cytokines, certain cytokine receptors varied across development.In particular, IL1R1, IL6R and TNFRSF1B expression changed between age groups.Notably, when changes were found in receptor expression, those changes approximated the pattern of their associated cytokine ligand.IL1R1, for instance, increased in toddlers and decreased in older age groups.IL6R and TNFRSF1B, like IL6 and TNF protein and mRNA, increased in toddlers and remained relatively elevated during late development, including adolescence.Other receptors, IL1RAP, IL18R, IL6ST, and TNFRSF1A remained stable throughout the rst 2 decades of postnatal life.These results suggest that, like cytokines, the trajectory of cytokine receptor expression across development is not uniform, and that increases in ligand and receptor pairs are more often concomitantly increased when they activate signaling rather than inhibit it.
The evidence presented here suggests that cytokine expression is upregulated during two major periods in human development: early neurodevelopment, when processes such as interneuron migration (71,72) and synaptogenesis (81, 82) dominate brain development; and late neurodevelopment, which is associated with synaptic pruning (81, 82), myelination (89, 90), and nal GABAergic system maturation (105,106).Cytokines have the potential to affect these processes based on both the timing reported here, and on effects previously reported in the literature.For instance, IL1B has been previously associated with neuronal migration (41) and the survival of differentiating oligodendrocytes (107) in vitro and in vivo.
Based on the timing of the peak in IL1B, it is interesting to consider whether IL1B during postnatal development may be associated with maturation from preoligodendrocyte to mature oligodendrocyte, which drops around age 5 in humans (108).Similarly, the adolescent peak in IL6 and TNF mRNAs could be associated with maturation of neurotransmitter systems, i.e. the GABAergic system.The development of GABAergic signaling is a known developmental process that is estimated to extend into adolescence (105,106).Both IL6 and TNF are known regulators of GABA receptor levels through endocytosis of GABA receptors at the synapse (109,110).It is interesting to consider whether increased expression in IL6 and/or TNF could contribute to the normal regulation of GABA receptor levels at the synapse in adolescents.We speculate that exaggerated increases in either IL6 or TNF during maturation may throw off normal homeostasis.
The adolescent increase in IL6 and TNF may be signi cant when considered in light of the age of onset for schizophrenia, which commonly occurs just after adolescence.Notably, increased IL6 mRNA and protein levels are one of the most robust changes found in the brain and blood of people with schizophrenia (111)(112)(113)(114)(115)(116)(117)(118)(119)(120)(121).There is increasing evidence of the role of the immune system in neurodevelopmental psychiatric disorders.Epidemiological work in humans suggests that immune activation during gestation and early childhood is associated with the occurrence of neurodevelopmental disorders, in particular the development of ASD, ADHD (Attention-de cit/hyperactivity disorder), and schizophrenia (23,(122)(123)(124)(125)(126).Immune activation during development may have short-and long-term impacts on brain development, potentially by disrupting immune factor levels directly during development (127), by inducing changes that affect the systemic response to environmental triggers later in life (e.g.(128-130)), and/or by inducing long-term changes in immune signal expression (50,113,128,131).In this study, we have found evidence that cytokine expression is also active and regulated well after the fetal period in humans, from birth through young adulthood, indicating that cytokines likely have a role in normal postnatal human cortical development.Thus, the dysregulation of cytokine expression during early postnatal life could similarly disrupt brain development through direct or indirect effects on developmental processes.We can use this new knowledge of regulated cytokine expression in the postnatal human prefrontal cortex to expand upon the interaction between cytokine expression, speci c brain development processes, and behavioral outcomes using experimental data.In particular, future studies utilizing targeted knockdown, knockout, and/or overexpression of these cytokines with inducible rodent models may allow us to assess the contribution of speci c cytokines to development more directly.For instance, by inhibiting a given cytokine during its period of elevated expression reported here, it may be possible to tie normal cytokine activity to particular processes in postnatal development.
Characterizing the normal contribution of cytokines to development will help clarify both the effects of abnormal in ammation/cytokine expression on developmental processes, and whether those effects relate to distinct periods of vulnerability.
Unlike mRNA, cytokine proteins circulate freely in the blood serum under physiological conditions.
Unfortunately, studies regarding the expression patterns of cytokines in the serum at different developmental ages in humans are somewhat limited.In the case of IL1B, IL6, IL18, and TNF, a comparison of the protein levels detected in this study to serum protein levels reported under physiological conditions during childhood (132) suggests that the contamination of tissue with circulating cytokines would not fully explain the changes in cytokine protein expression seen in this study.
The reported pattern of IL18 blood serum protein expression, for instance, found that IL18 was highest between ages 1-7 years old and lowest in samples over 18 years old (132).IL18 brain protein levels were lowest in neonates and infants, so a contribution from the serum would presumably arti cially in ate apparent IL18 protein concentrations in this age group.If anything, excluding any serum protein contribution from the results in our study, would emphasize the difference between age groups by decreasing neonate and infant levels to diverge further from teenager and young adult levels.In short, serum protein levels may contribute to the total protein levels reported in this study; however, the patterns of protein expression reported herein were largely consistent with corresponding mRNA expression patterns and did not strongly resemble existing reports of serum cytokine levels during human development.

Conclusion
This work provides a foundation for understanding the normal changes in levels of cytokine, cytokine receptors, and inhibitors from early postnatal neurodevelopment to maturation in the human cortex.Our results suggest that there may be two distinct time windows (toddlers and teenagers) when cytokine signaling takes a more salient role in human cortical development, both of which are associated with an increased incidence of major psychiatric disorders.The increases seen in cytokine expression for IL1B, IL6, IL13, IL18, and TNF between birth and age 5 suggest a signi cant action around the time of onset of ASD.These results highlight toddlerhood not just as an overall transformative period in the brain but as a period of development strongly tied to immune-associated cytokine signaling as well as complement signaling.Given that late adolescence is considered a period of vulnerability to psychiatric disorders, especially schizophrenia, the role of IL6 and TNF in adolescent cortical development needs further consideration and study.Additionally, since adolescence is normally not associated with increased complement levels but rather is associated with increased complement inhibition, this suggests that increased cytokines may be acting independently of complement at maturation and without associated transcriptional increases in major cytokine inhibitors.

Figure 2 Expression
Figure 2

Figure 4 Expression
Figure 4

*
≤0.05; **≤0.01;***≤0.001all images.Colored asterisks denote comparisons to age group with a shared color scheme.Bars represent standard error of the mean.The horizontal line represents and the bars represent SEM.

Figure 5 Expression
Figure 5

Table 1 .
Summary of patient demographics.RIN, pH, and PMI are presented as mean ± standard deviation.

Table 2 .
Correlation of cytokine protein and transcript expression levels with age under 5. Bolded transcripts and proteins are signi cantly correlated with age under 5.