The ultradian transcriptome of the human prefrontal cortex and abnormalities in schizophrenia

Twelve-hour (12 h) ultradian rhythms are a well-known phenomenon in coastal marine organisms. While 12 h cycles are observed in human behavior and physiology, no study has measured 12 h rhythms in the human brain. Here we identify 12 h rhythms in transcripts which either peak at sleep/wake transitions (~9 AM/PM) or static times (~3 PM/AM) in the dorsolateral prefrontal cortex, a region involved in cognition. Subjects with schizophrenia lose 12 h rhythms in genes associated with the unfolded protein response and neuronal structural maintenance. Moreover, genes involved in mitochondrial function and protein translation, which normally peak at sleep/wake transitions, peak instead at static times in schizophrenia, suggesting suboptimal timing of these essential processes that may contribute to cognitive deficits commonly found in schizophrenia.


Introduction
abnormalities in brain function.  (58 and 67% of 24 and 12 h RCs, respectively) in the SZ cohort (Fig 3C-D). As an example of what 186 these data show, expression of transcripts in the electron transport chain complexes across time 187 indicated mitochondria-associated genes had 12 h rhythms in the mNP cohort, while in SZ these 188 transcripts had a combination of 12 and 24 h RCs (Fig 3E-F). We propose that this may explain 189 why the NLR analysis identifies this group of transcripts as either a 12 h or 24 h rhythm, or both, 190 in SZ subjects.   The superchiasmatic nucleus (SCN) is the master pacemaker of circadian rhythms in the 300 brain, but whether ultradian rhythms are regulated across regions by a dedicated system is unknown.

301
While not specific to 12 h rhythms, levels of dopamine in the striatum fluctuate in synchrony with 302 ultradian locomotor activity cycles and dopaminergic transmission directly regulates ultradian cycle 303 length in mice [33]. Interestingly, these dopamine-driven ultradian cycles in locomotor activity 304 harmonize with circadian rhythms coordinated by the SCN, but if this relationship is disrupted it 305 can lead to altered patterns of arousal and disrupted sleep/wake cycles when desynchronized [33], 306 which are commonly observed in subjects with SZ. SZ has long been associated with altered 307 dopaminergic transmission in cortico-striatal pathways [34]. Abnormalities in dopaminergic 308 signaling, therefore, may be a potential mechanism by which ultradian rhythms are disrupted across 309 the brain in SZ.
Human postmortem brain tissue research presents limitations that we have addressed as 311 much as possible through our study design including exclusion of subjects older than 65, strict TOD 312 criteria, and cohorts matched for a number of important biological, clinical and technical factors.

313
Future studies in animal models will be necessary to determine the impact of features such as 314 antipsychotic medication on the ultradian and circadian transcriptomes.

315
In conclusion, this study is the first to identify 12 h rhythms in transcript expression in the 316 human brain. These rhythms are associated with fundamental cellular processes. However, in SZ, 317 there is a strong reduction in the number of transcripts with 12 h rhythms, along with altered timing 318 of transcripts important in mitochondrial function. Future studies will determine the functional 319 consequences of these findings to optimal brain health and the pathophysiology of brain disorders.

322
Human postmortem brain samples 323 Human postmortem RNA-sequencing data for 613 samples were obtained from the 324 CommonMind Consortium (https://nimhgenetics.org/available_data/commonmind/) and filtered 325 based off previously described criteria [8,10]. Briefly, subjects were included if they met the criteria 326 of rapid death (<2 h elapsed time between precipitating event and death announcement) and had a 327 postmortem interval (PMI) of <30 h. In addition, subjects with age >65 years were removed as our 328 lab has previously observed significant differences in molecular rhythms between young and 329 elderly subjects [6]. 104 non-psychiatric (NP) subjects and 46 subjects with schizophrenia (SZ) met 330 these criteria (S1 Table). The larger sample size of NP subjects may result in more statistical power 331 for rhythm detection. Therefore, 46 non-psychiatric subjects that were best matched to the 46 SZ 332 subjects by age, sex, race, TOD, PMI, site of collection, and pH (match NP cohort (mNP); S1 Table). The subjects in these cohorts are the same as those used in a previous study performed by  17.9 ± 6.0 17.3 ± 6.2 17.0 ± 8.3 Brain pH (mean ± SD) 6.7 ± 0.2 6.6 ± 0.2 6.5 ± 0.3 TOD (mean ± SD) 8.6 ± 5.7 7.0 ± 6.4 7.8 ± 5.2 Site (Pitt/MSSM) 61/43 22/24 22/24 S1 Table. Description of CommonMind Consortium cohorts Abbreviations: y = years, h = hours; PMI = postmortem interval; SD = standard deviation; TOD = time of death; p < 0.05 p < 0.01 p < 0.001