Elevation in plasma tRNA fragments precede seizures in humanepilepsy

Marion C. Hogg, … , David C. Henshall, Jochen H.M. Prehn

J Clin Invest. 2019. https://doi.org/10.1172/JCI126346.

In-Press Preview

Transfer RNAs (tRNAs) are a major class of noncoding RNA. Stress-induced cleavage of tRNA is highly conserved andresults in tRNA fragments. Here we find specific tRNA fragments in plasma are associated with epilepsy. Small RNAsequencing of plasma samples collected during video-EEG monitoring of focal epilepsy patients identified significantdifferences in three tRNA fragments (5′GlyGCC, 5′AlaTGC, and 5′GluCTC) from controls. Levels of these tRNAfragments were higher in pre-seizure than post-seizure samples, suggesting they may serve as biomarkers of seizure riskin epilepsy patients. In vitro studies confirmed that production and extracellular release of tRNA fragments was lower afterepileptiform-like activity in hippocampal neurons. We designed PCR-based assays to quantify tRNA fragments in a cohortof pre- and post-seizure plasma samples from focal epilepsy patients and healthy controls. Receiver operatingcharacteristic analysis indicated that tRNA fragments potently distinguished pre- from post-seizure patients. ElevatedtRNA fragments levels were not detected in patients with psychogenic non-epileptic seizures, and did not result frommedication tapering. This study identifies a novel class of epilepsy biomarker and reveals the potential existence ofprodromal molecular patterns in blood that could be used to predict seizure risk.

Concise Communication Neuroscience

Find the latest version:

https://jci.me/126346/pdf

1

Elevation in plasma tRNA fragments precede seizures in human epilepsy 1

Marion C. Hogg1,2, Rana Raoof1,3, Hany El Naggar4, Naser Monsefi1, Norman Delanty2,4, Donncha 2

F. O’Brien5, Sebastian Bauer6,7,8, Felix Rosenow6,7,8, David C. Henshall1,2 & Jochen H.M. Prehn1,2* 3

1. Department of Physiology and Medical Physics, Royal College of Surgeons In Ireland, St. 4

Stephen’s Green, Dublin D02 YN77, Ireland. 5

2. FutureNeuro Research Centre, Royal College of Surgeons In Ireland, St. Stephen’s Green, 6

Dublin D02 YN77, Ireland. 7

3. Department of Anatomy, Mosul Medical College, University of Mosul, Mosul, Iraq. 8

4. Department of Neurology, Beaumont Hospital, Beaumont, Dublin, Ireland. 9

5. Department of Neurosurgery, Beaumont Hospital, Dublin, Ireland. 10

6. Epilepsy Center Hessen, Department of Neurology, Baldingerstr, 35043, Marburg, Germany. 11

7. Epilepsy Center Frankfurt Rhine-Main, Neurocenter, Goethe-University, Schleusenweg 2-16, 12

Haus 95, 60528, Frankfurt, Germany 13

8. LOEWE Center for Personalized Translational Epilepsy Research (CePTER), Frankfurt, Germany. 14

15

*Corresponding author: Jochen H. M. Prehn, Department of Physiology and Medical Physics, 16

Royal College of Surgeons in Ireland, 123 St. Stephen’s Green, Dublin 2, Ireland. 17

Email: prehn@rcsi.ie 18

Tel: + 353 1 402 2255 19

Fax: + 353 1 402 2447 20

21

Word count: 4,000 22

2

Abstract: 191 23

Figures: 4 figures 24

Supplementary data: 1 document 25

26

The authors have declared that no conflict of interest exists. 27

28

Keywords 29

Epilepsy, seizure, tRNA fragment, biomarker 30

3

Abstract 31

Transfer RNAs (tRNAs) are a major class of noncoding RNA. Stress-induced cleavage of tRNA is 32

highly conserved and results in tRNA fragments. Here we find specific tRNA fragments in plasma 33

are associated with epilepsy. Small RNA sequencing of plasma samples collected during video-34

EEG monitoring of focal epilepsy patients identified significant differences in three tRNA 35

fragments (5’GlyGCC, 5’AlaTGC, and 5’GluCTC) from controls. Levels of these tRNA fragments 36

were higher in pre-seizure than post-seizure samples, suggesting they may serve as biomarkers 37

of seizure risk in epilepsy patients. In vitro studies confirmed that production and extracellular 38

release of tRNA fragments was lower after epileptiform-like activity in hippocampal neurons. We 39

designed PCR-based assays to quantify tRNA fragments in a cohort of pre- and post-seizure 40

plasma samples from focal epilepsy patients and healthy controls. Receiver operating 41

characteristic analysis indicated that tRNA fragments potently distinguished pre- from post-42

seizure patients. Elevated tRNA fragments levels were not detected in patients with psychogenic 43

non-epileptic seizures, and did not result from medication tapering. This study identifies a new 44

class of epilepsy biomarker and reveals the potential existence of prodromal molecular patterns 45

in blood that could be used to predict seizure risk. 46

4

Introduction 47

Temporal lobe epilepsy (TLE) is the most prevalent form of focal epilepsy in adults, with around 48

0.1% of the world’s population affected. Diagnosis of epilepsy is complex and challenging. 49

Seizures are rarely witnessed by epileptologists and diagnosis is primarily based on clinical 50

history, supported by electroencephalography (EEG) and imaging (MRI) data. Accordingly, there 51

is an unmet need for biomarkers of epilepsy. To date, few studies have focussed on biomarkers 52

of seizure onset, and seizure prediction studies mainly rely on EEG recordings to provide a 53

patient-specific profile of seizure activity, which has limited predictive power in a subset of 54

patients (1). From a patient’s perspective, the ability to forecast seizures would allow greater 55

control in daily life and improve patient safety. 56

Circulating blood-based molecules represent an attractive source of epilepsy biomarker due to 57

the potential for fast analysis of easy-to-collect samples. Serum levels of High-Mobility Group Box 58

1 (HMGB1) protein have been shown to indicate epileptogenesis in animals and drug-refractory 59

epilepsy in patients (2). Small noncoding RNAs have recently emerged as potential biomarkers, 60

with several studies reporting differences in serum or plasma levels of miRNAs in patients with 61

epilepsy (3). Research into predictive biomarkers for seizure onset or imminence has been 62

limited. Interestingly, a recent study sampling blood from epilepsy patients in a video-EEG 63

monitoring unit identified four miRNAs that were significantly elevated immediately after onset 64

of generalised tonic-clonic seizures (GTCS) (4). 65

Transfer RNAs (tRNA) are ubiquitous noncoding RNAs that deliver amino acids to the ribosome 66

during protein synthesis. Cleavage of tRNAs occurs as part of a highly conserved stress response 67

5

present in single-celled organisms (5). In humans, tRNA fragments are generated by 68

ribonucleases including Dicer and Angiogenin (6). Numerous functions have been attributed to 69

tRNA fragments, including inhibition of protein translation (7), initiation of stress granule 70

formation (8), and regulation of gene expression (9, 10). tRNA fragments have also been 71

identified circulating in blood (11), indicating they are protected from degradation, which makes 72

them ideal candidates for investigation as biomarkers. 73

Here we identified plasma tRNA fragments in RNA sequencing (RNA-seq) data from focal 74

epilepsy patients (12). We found three specific tRNA fragments that were significantly elevated 75

in pre-seizure samples compared to post seizure samples and healthy controls. We performed 76

mechanistic and clinical validation studies to explore their potential as novel biomarkers of 77

seizure risk. Together, our study suggests that specific tRNA fragments may be a novel class of 78

epilepsy biomarker that could support prediction of seizure risk in patients diagnosed with 79

epilepsy. 80

81

6

Results and Discussion 82

Blood samples were collected from 16 refractory focal epilepsy patients (majority with TLE) 83

upon arrival for video-EEG monitoring at the Epilepsy Monitoring Unit (EMU) at the Epilepsy 84

Centre in Hessen, Department of Neurology, Marburg. All patients had drug-refractory focal 85

epilepsy and were on polytherapy. Demographics and medications are shown in Supplementary 86

Table 1. Continuous video-EEG monitoring was performed for each patient using the 10-20 87

standard international electrode placement system. Recordings were manually reviewed by a 88

neurologist with special training in epilepsy. A second blood sample was collected 24 hours after 89

a confirmed electro-clinical seizure. Plasma samples were obtained at the same site from age and 90

gender-matched healthy volunteers. 91

Small RNA-seq (< 50 nt) was performed on pooled samples from 16 controls and pre-seizure or 92

post-seizure samples from 16 focal epilepsy patients. A custom tRNA library was used to quantify 93

reads aligning to tRNAs, summarised in supplementary Table 2. Reads were pooled for tRNAs of 94

the same isoacceptor type, and tRNAs were ranked by fold-change between pre-seizure and 95

control samples (Table 1). Three tRNA fragments were chosen for further investigation based on 96

abundance (counts per million, CPM) and fold-change: Glycine GCC, Alanine TGC, and Glutamic 97

acid CTC (Table 1). Read coverage plots revealed reads aligned to the 5’ end of the tRNA (Figure 98

1A, D, & G). The cleavage site was mapped onto the predicted secondary structure of each of the 99

tRNAs (Figure 1B, E, & H). Predicted secondary structures of the tRNA fragments indicated they 100

form short hairpin structures (Figure 1C, F, & I). Of note, tRNA fragments levels were highest in 101

pre-seizure samples and closer to control levels in post-seizure samples (Figure 1A, D, & G). 102

7

As these tRNA fragments are novel, no existing assays were available; therefore we designed 103

custom Taqman assays. These assays utilise standard qPCR techniques and allow fast, specific 104

quantification of short RNA fragments. As tRNAs are highly conserved, the assays can be used 105

on both mouse and human samples (see supplementary Figure 1). Next, we aimed to 106

experimentally model the observation that tRNA fragments levels decreased following epileptic 107

seizure. Seizure-like activity was induced in cultured mouse hippocampal neurons by withdrawal 108

of magnesium and assessed by measuring intracellular calcium levels with FLUO-4 (Figure 2A) 109

(13). Representative FLUO-4 traces in the presence (Mg+: 1 mM MgCl2) and absence (Mg-: 0 mM 110

MgCl2) of magnesium indicated spikes of increased intracellular calcium levels corresponding to 111

increased neuronal activity (Figure 2B). Intracellular 5’GluCTC levels were significantly decreased 112

following epileptiform activity (Figure 2C). As tRNA fragments are secreted from neurons (14), 113

we next quantified tRNA fragments in the extracellular medium, and found that both 5’GlyGCC 114

and 5’GluCTC were significantly reduced following increased neuronal activity (Figure 2D). This 115

confirmed tRNA fragments are generated in neurons, secreted from neurons, and that secretion 116

of tRNA fragments is reduced following neuronal hyperexcitation (mimicking epileptiform 117

activity). tRNA fragment levels were normalised to U6 RNA, which remained unchanged 118

throughout experiments (supplementary Figure 2). 119

120

We next used the custom Taqman assays to validate the RNA-seq data in the 16 focal epilepsy 121

patients and 16 healthy controls recruited in Marburg, with an additional 16 focal epilepsy 122

patients recruited at the EMU in Beaumont Hospital, Dublin (Supplementary Table 1), and 16 123

healthy controls. We found that levels of all three tRNA fragments were significantly elevated in 124

8

pre-seizure samples (Figure 3A-C). Wilcoxon signed-rank test indicated there was a highly 125

significant change in tRNA fragment levels between pre and post seizure samples, 5’GlyGCC: 126

3.17-fold change (p=0.0002), 5’AlaTGC: 1.93-fold change (p<0.0001), and 5’GluCTC: 2.89-fold 127

change (p=0.0003) (Figure 3D-F), with the majority of samples showing a decrease in tRNA 128

fragment levels post-seizure. Receiver Operating Characteristic (ROC) analysis indicated tRNA 129

fragments could distinguish pre- and post- seizure samples (Figure 3G-I). 5’GlyGCC had an area 130

under the curve (AUC) of 0.816 (p = 0.000027, Figure 3G), 5’AlaTGC AUC = 0.916 and was highly 131

significant (p = 1.86x10-8, Figure 3H), and 5’GluCTC AUC = 0.802 (p = 0.000069, Figure 3I). 132

Youden’s J statistic was used to determine the optimal cut-off for distinguishing pre and post 133

seizure samples and this indicated a value of 1.33 was most discriminatory for 5’AlaTGC with a 134

sensitivity of 96.8% and specificity of 87.1% (Figure 3H), for 5’GlyGCC 1.36 was optimal (Figure 135

3G) and for 5’GluCTC a 2.13 cut-off performed best (Figure 3I), ROC analyses are summarised in 136

supplementary Table 3. Analysing males and females separately revealed 5’AlaTGC was 137

significantly elevated in pre-seizure samples from both groups (supplementary Figure 3). 138

Importantly, levels of two other tRNA fragments detected in the plasma RNA-seq data (5’ValAAC 139

& 5’ProAGG) were validated by custom Taqman assay and were not elevated in pre-seizure 140

samples (supplementary Figure 4). Plasma collected at additional time points were available for 141

24/32 patients (T1: upon arrival to the EMU and T3: 1 hour post-seizure). Analysis of extra time 142

points indicates 5’AlaTGC levels spike pre-seizure and fall 1 hour post-seizure before rising again 143

by 24 hours post-seizure (supplementary Figure 5). Analysis of correlation between time interval 144

(pre-seizure blood collection to seizure onset) and tRNA fragment level indicates 5’GluCTC is 145

significantly elevated prior to seizure onset (Spearman r=-0.59, p=0.0035) when analysis is limited 146

9

to under 50 hours (supplementary Figure 6). 5’GlyGCC and 5’AlaTGC showed weaker correlation 147

that did not reach statistical significance. These analyses indicate specific tRNA fragments can 148

discriminate between pre- and post-seizure samples and may be of use as epilepsy biomarkers 149

in advance of seizure. 150

151

Having demonstrated tRNA fragments are elevated in focal epilepsy patient plasma in advance 152

of seizures occurred we sought to determine whether tRNA fragments were detectable in brain 153

tissues. Five patients from the Dublin cohort had undergone surgical resection in an attempt to 154

alleviate their seizures. Fresh frozen hippocampal and cortical tissue was cryosectioned, with 155

neighbouring sections mounted for histological analysis and collected for RNA analysis. All three 156

tRNA fragments were detectable in both hippocampus and cortex tissue (Supplementary Figure 157

7A-C); no significant difference in tRNA levels between tissues was detected. Histological analysis 158

indicated tissue samples were intact although components of the hippocampal formation were 159

only visible in two samples (Supplementary Figure 7D). Analysis of MRI data revealed 9 focal 160

epilepsy patients had no structural lesions detected (20 positive for lesions, 3 indeterminate), 161

and interestingly tRNA fragments were higher in MRI negative patients (supplementary Figure 8) 162

indicating tRNA fragments are not released as a result of structural damage. 163

164

One caveat to interpreting this data is the patients’ medication is withdrawn upon arrival at the 165

EMU and may therefore be higher in the pre-seizure than the post-seizure samples; however, we 166

do not believe that medication levels influence tRNA fragment levels for several reasons. Firstly, 167

10

we noted that four patients did not have their medication reduced and went on to experience 168

seizures, however we found no significant difference between pre-seizure tRNA fragment levels 169

in patients with and without AED reduction (supplementary Figure 9). Secondly, we have 170

analysed tRNA fragment levels in a cohort of patients where AEDs were reduced but patients did 171

not experience seizures and here we find no change in tRNA fragment levels (supplementary 172

Figure 10). Finally, we analysed plasma tRNA fragment levels in six patients subsequently 173

diagnosed with psychogenic non-epileptic seizures (PNES); here we found no change in tRNA 174

fragment levels between pre and post “seizure-like” event (supplementary Figure 11), indicating 175

elevated tRNA fragments are linked to electro-clinical seizures. 176

177

To investigate the mechanism leading to increased pre-seizure tRNA fragment levels we 178

examined whether abnormal inter-ictal neuronal activity may be occurring, which is not sufficient 179

to trigger an epileptic seizure but may instigate tRNA cleavage and secretion. One study using 180

continuous intra-cranial EEG-monitoring of a small group of patients identified long-term energy 181

bursts up to 7 hours before seizure onset, which increased in frequency in the hours immediately 182

preceding seizure onset (15). To examine this a Neurologist reviewed video-EEG recordings from 183

a period of 18-24 hours upon arrival to the EMU and patients were classified into three groups 184

based on v-EEG activity: rare, occasional, and frequent. There was no significant difference in 185

pre-seizure tRNA fragment levels between groups (supplementary Figure 12). Eight patients 186

developed GTCS; however, no difference in pre-seizure tRNA fragment levels were detected in 187

these patients compared to those who did not develop GTCS (supplementary Figure 13). 188

11

189

There are currently no reliable biomarkers of seizure onset and patients with drug refractory 190

epilepsy consistently report the unpredictable nature of seizures to be a challenging facet of their 191

disease (16). The ability to forecast seizure activity would allow patients to regain control over 192

their condition. Here we investigated RNA-seq data from patients with focal epilepsy, and 193

identified three tRNA fragments that were elevated in pre-seizure plasma samples. We showed 194

they are expressed and secreted from neurons, and that tRNA fragment levels are regulated in 195

response to neuronal activity. Finally, we validated the RNA-seq data results showing tRNA 196

fragments are elevated in pre-seizure plasma samples from a cohort of focal epilepsy patients 197

compared to healthy controls. We present an exciting proof-of-concept study, which indicates 198

plasma tRNA fragments warrant further investigation as prodromal biomarkers that could be 199

used to predict seizure risk. 200

201

The underlying stress that precipitates tRNA cleavage in our study is unknown. Seizures are 202

generally defined as excessive electrical activity in the brain, which can lead to involuntary 203

movement, changes in behaviour, and loss of consciousness. Many underlying causes can lead to 204

seizures such as infections, tumours, brain trauma, oxygen deprivation and exposure to drugs or 205

toxins. However, in most patients diagnosed with epilepsy the underlying cause of seizures is 206

unknown. At the cellular level, numerous changes during epileptogenesis have been identified 207

including neurodegeneration, gliosis, inflammation, neuronal growth and angiogenesis 208

(reviewed in (17)). tRNA cleavage occurs in response to stress in organisms from yeast to humans 209

12

indicating it is a highly conserved process (5, 6). tRNA cleavage can occur at several sites on the 210

tRNA molecule; Dicer cleaves tRNAs between the D-loop and the anticodon loop, whereas 211

Angiogenin primarily cleaves tRNAs within the anticodon loop (6). Interestingly Dicer levels are 212

significantly lower in TLE patients with hippocampal sclerosis (HS) than those without (18). tRNA 213

cleavage has been identified in response to infection (19) and ischaemia (20), indicating these 214

epileptogenic processes could contribute to the observed tRNA cleavage. A recent study has 215

shown that Kainic Acid (KA) treatment of synaptosome fractions induced spontaneous release of 216

specific miRNAs, which was inhibited by depolarisation, indicating dynamic regulation of non-217

coding RNA secretion by neuronal activity (21). An important limitation of the present study is 218

that despite detecting the tRNA fragments in resected brain tissue and cultured hippocampal 219

neurons, we cannot exclude that the tRNA fragments we detected in patient plasma may have 220

originated in another tissue, as both acute seizures and chronic epilepsy are system disorders, 221

with physical and biochemical effects on organs and tissues outside the CNS (22). 222

223

tRNA fragments share many features with miRNAs, which make them both ideal molecules for 224

use as biomarkers. In order to be of use to patients a biomarker of seizure imminence would 225

require a portable quantification device capable of assessing tRNA fragment levels rapidly in 226

whole blood. Recent advances in this area include development of the TORNADO device (23) 227

which is capable of accurately quantifying miRNA-134 in plasma from focal epilepsy patients 228

using an electrochemical quantification approach (12). This might complement other new 229

wearable technologies, such as seizure detection watches (24). Our data indicate that tRNA 230

fragments are quantifiable using standard laboratory-based PCR techniques and may provide 231

13

increased specificity over miRNAs as epilepsy biomarkers, which are the current focus of many 232

studies. 233

14

Methods 234

Study Approval 235

Samples were collected at the Philipps University of Marburg, Germany (MAR) and the 236

Department of Neurology, Beaumont Hospital, Dublin, Ireland (DUB). Ethical approval was 237

obtained from the local medical ethics committees (MAR, 17/14; DUB 13/75), and written 238

consent was obtained from all participants according to the Declaration of Helsinki. Animal work 239

was approved by the Research Ethics Committee of the Royal College of Surgeons in Ireland (REC 240

1122). 241

242

Additional methods are provided in Supplementary data. 243

15

Author Contributions 244

Study design: M.C.H., D.C.H., & J.H.M.P. 245

Sample collection and clinical evaluation: R.R., H.E.N., N.D., D.O’B., S.B., & F.R. 246

Performing and analysing experiments: M.C.H, R.R., & N.M. 247

Writing manuscript: M.C.H., D.C.H, & J.H.M.P 248

249

Acknowledgements 250

The authors wish to thank the patients that participated in this study, the neuropathologists who 251

assessed surgically resected patient tissue (Drs Jane Cryan, Francesca M. Brett, Alan Beausang, 252

and Michael A. Farrell), and Dr Heiko Duessman (RCSI) for help with imaging. This publication has 253

emanated from research supported in part by a research grant from Science Foundation Ireland 254

(SFI) under Grant Number 16/RC/3948 and co-funded under the European Regional 255

Development Fund and by FutureNeuro industry partners. This study was supported by SFI grants 256

SFI/13/IA/1891, financial support of the European Union’s ‘Seventh Framework’ Programme 257

(FP7) under Grant Agreement No. 602130 (EpimiRNA), and a fellowship from the Iraqi Ministry 258

of Higher Education and Scientific Research. 259

260

16

References 261

1. Kuhlmann L, Lehnertz K, Richardson MP, Schelter B, Zaveri HP. Seizure prediction - ready for a 262 new era. Nat Rev Neurol. 2018. 263 2. Walker LE, Frigerio F, Ravizza T, Ricci E, Tse K, Jenkins RE, et al. Molecular isoforms of high-264 mobility group box 1 are mechanistic biomarkers for epilepsy. J Clin Invest. 2017;127(6):2118-32. 265 3. Enright N, Simonato M, Henshall DC. Discovery and validation of blood microRNAs as molecular 266 biomarkers of epilepsy: Ways to close current knowledge gaps. Epilepsia Open. 2018;3(4):427-36. 267 4. Sun J, Cheng W, Liu L, Tao S, Xia Z, Qi L, et al. Identification of serum miRNAs differentially 268 expressed in human epilepsy at seizure onset and post-seizure. Mol Med Rep. 2016;14(6):5318-24. 269 5. Lee SR, Collins K. Starvation-induced cleavage of the tRNA anticodon loop in Tetrahymena 270 thermophila. J Biol Chem. 2005;280(52):42744-9. 271 6. Li S, Xu Z, Sheng J. tRNA-Derived Small RNA: A Novel Regulatory Small Non-Coding RNA. Genes 272 (Basel). 2018;9(5). 273 7. Ivanov P, Emara MM, Villen J, Gygi SP, Anderson P. Angiogenin-induced tRNA fragments inhibit 274 translation initiation. Mol Cell. 2011;43(4):613-23. 275 8. Emara MM, Ivanov P, Hickman T, Dawra N, Tisdale S, Kedersha N, et al. Angiogenin-induced 276 tRNA-derived stress-induced RNAs promote stress-induced stress granule assembly. J Biol Chem. 277 2010;285(14):10959-68. 278 9. Chen Q, Yan M, Cao Z, Li X, Zhang Y, Shi J, et al. Sperm tsRNAs contribute to intergenerational 279 inheritance of an acquired metabolic disorder. Science. 2016;351(6271):397-400. 280 10. Sharma U, Conine CC, Shea JM, Boskovic A, Derr AG, Bing XY, et al. Biogenesis and function of 281 tRNA fragments during sperm maturation and fertilization in mammals. Science. 2016;351(6271):391-6. 282 11. Dhahbi JM, Spindler SR, Atamna H, Yamakawa A, Boffelli D, Mote P, et al. 5' tRNA halves are 283 present as abundant complexes in serum, concentrated in blood cells, and modulated by aging and 284 calorie restriction. BMC Genomics. 2013;14:298. 285 12. Raoof R, Bauer S, El Naggar H, Connolly NMC, Brennan GP, Brindley E, et al. Dual-center, dual-286 platform microRNA profiling identifies potential plasma biomarkers of adult temporal lobe epilepsy. 287 EBioMedicine. 2018. 288 13. Weisova P, Alvarez SP, Kilbride SM, Anilkumar U, Baumann B, Jordan J, et al. Latrepirdine is a 289 potent activator of AMP-activated protein kinase and reduces neuronal excitability. Transl Psychiatry. 290 2013;3:e317. 291 14. Li H, Wu C, Aramayo R, Sachs MS, Harlow ML. Synaptic vesicles contain small ribonucleic acids 292 (sRNAs) including transfer RNA fragments (trfRNA) and microRNAs (miRNA). Sci Rep. 2015;5:14918. 293 15. Litt B, Esteller R, Echauz J, D'Alessandro M, Shor R, Henry T, et al. Epileptic seizures may begin 294 hours in advance of clinical onset: a report of five patients. Neuron. 2001;30(1):51-64. 295 16. Dalic L, Cook MJ. Managing drug-resistant epilepsy: challenges and solutions. Neuropsychiatr Dis 296 Treat. 2016;12:2605-16. 297 17. Pitkanen A, Lukasiuk K. Mechanisms of epileptogenesis and potential treatment targets. Lancet 298 Neurol. 2011;10(2):173-86. 299 18. McKiernan RC, Jimenez-Mateos EM, Bray I, Engel T, Brennan GP, Sano T, et al. Reduced mature 300 microRNA levels in association with dicer loss in human temporal lobe epilepsy with hippocampal 301 sclerosis. PLoS One. 2012;7(5):e35921. 302 19. Wang Q, Lee I, Ren J, Ajay SS, Lee YS, Bao X. Identification and functional characterization of 303 tRNA-derived RNA fragments (tRFs) in respiratory syncytial virus infection. Mol Ther. 2013;21(2):368-79. 304 20. Li Q, Hu B, Hu GW, Chen CY, Niu X, Liu J, et al. tRNA-Derived Small Non-Coding RNAs in Response 305 to Ischemia Inhibit Angiogenesis. Sci Rep. 2016;6:20850. 306

17

21. Xu J, Chen Q, Zen K, Zhang C, Zhang Q. Synaptosomes secrete and uptake functionally active 307 microRNAs via exocytosis and endocytosis pathways. J Neurochem. 2013;124(1):15-25. 308 22. Devinsky O. Effects of Seizures on Autonomic and Cardiovascular Function. Epilepsy Curr. 309 2004;4(2):43-6. 310 23. McArdle H, Jimenez-Mateos EM, Raoof R, Carthy E, Boyle D, ElNaggar H, et al. "TORNADO" - 311 Theranostic One-Step RNA Detector; microfluidic disc for the direct detection of microRNA-134 in 312 plasma and cerebrospinal fluid. Sci Rep. 2017;7(1):1750. 313 24. Bruno E, Simblett S, Lang A, Biondi A, Odoi C, Schulze-Bonhage A, et al. Wearable technology in 314 epilepsy: The views of patients, caregivers, and healthcare professionals. Epilepsy Behav. 2018;85:141-9. 315 25. Chan PP, Lowe TM. GtRNAdb 2.0: an expanded database of transfer RNA genes identified in 316 complete and draft genomes. Nucleic Acids Res. 2016;44(D1):D184-9. 317

318

18

Figures 319

320

Figure 1: 5’tRNA fragments are differentially expressed in plasma from epilepsy patients. Small 321

RNA-seq on pooled samples from 16 focal epilepsy patients pre and post-seizure and 16 healthy 322

controls. Coverage plots showing reads aligned to A) GlyGCC, D) AlaTCG, & G) GluCTC tRNAs, in 323

counts per million (CPM, y-axis) and tRNA sequences (x-axis). B, E, & H the cleavage sites are 324

indicated (red triangle) on the mature tRNA structures (downloaded from GtRNAdb 2.0 (25)). C, 325

F, & I Predicted secondary structures of the tRNA fragments. 326

19

327

Figure 2: tRNA fragments are regulated by epileptiform activity in hippocampal neurons. A) 328

Primary mouse hippocampal neurons were incubated with 5 M FLUO4 for 45 mins at 37oC and 329

imaged at 5 Hz. Scale bar = 40 m. B) Representative FLUO4 traces are shown for neurons in 1 330

mM Mg (Mg+, upper panel) or 0 mM Mg (Mg-, lower panel). Following 2 hours in Mg+ or Mg- 331

media, media was collected and total RNA extracted. Both intracellular (C) and extracellular (D) 332

tRNA fragment levels were lower following culture in Mg-free media, where extracellular 333

5’GlyGCC and 5’GluCTC levels were significantly lower (p < 0.05, Students t-test). In C & D 334

individual data points are plotted with mean indicated in red and error bars +/- SEM.N= 15-16 335

replicates from 5 independent preps. 336

20

337

Figure 3: tRNA fragments are elevated in plasma from epilepsy patients pre-seizure. 32 epilepsy 338

patients had blood sampled pre and post-seizure and 32 healthy controls were analysed. A) 339

5’GlyGCC, B) 5’AlaTGC and C) 5’GluCTC tRNA fragments were significantly elevated in pre-seizure 340

samples, Kruskal-Wallis test indicated p = 0.0008, p < 0.0001, and p = 0.0001 respectively. No 341

significant difference was detected between post-seizure samples and controls. Mean +/- SEM 342

indicated. Most tRNA fragments decreased significantly post-seizure, Wilcoxon signed-rank test 343

results shown in D) 5’GlyGCC (p=0.0002), E) 5’AlaTGC (p<0.0001), and F) 5’GluCTC (p=0.0003). 344

ROC analysis indicated G) 5’GlyGCC had an area under the curve (AUC) of 0.816 (p = 0.000027), 345

H) 5’AlaTGC AUC = 0.916 (p = 1.86x10-8), and I) 5’GluCTC AUC = 0.802 (p = 0.000069). 346

21

347

Table 1: Reads aligned to tRNAs ranked by fold change between pre-seizure and control. Reads 348

aligning to tRNAs were pooled by isoacceptor type and ranked by fold-change between pre-349

seizure and control. Highlighted tRNAs were investigated further. 350

351