6 7 Chemical modification of transcripts with 5' caps occurs in all organisms. Here we report a 8 systems-level mass spectrometry-based technique, CapQuant, for quantitative analysis of the 9 cap epitranscriptome in any organism. The method was piloted with 21 canonical caps -10 m 7 GpppN, m 7 GpppNm, GpppN, GpppNm, and m 2,2,7 GpppG -and 5 "metabolite" caps -NAD, 11 FAD, UDP-Glc, UDP-GlcNAc, and dpCoA. Applying CapQuant to RNA from purified dengue 12 virus, Escherichia coli, yeast, mice, and humans, we discovered four new cap structures in 13 humans and mice (FAD, UDP-Glc, UDP-GlcNAc, and m 7 Gpppm 6 A), cell-and tissue-specific 14 variations in cap methylation, and surprisingly high proportions of caps lacking 2'-O-methylation, 15 such as m 7 Gpppm 6 A in mammals and m 7 GpppA in dengue virus, and we did not detect cap 16 m 1 A/m 1 Am in humans. CapQuant accurately captured the preference for purine nucleotides at 17 eukaryotic transcription start sites and the correlation between metabolite levels and metabolite 18 caps. The mystery around cap m 1 A/m 1 Am analysis remains unresolved. 19 20 21 cap structure involves -phosphate methylation of unprocessed 5'-triphosphate (mPPPN) on 1 small RNAs such as mammalian U6 and 7SK, mouse B2, and plant U3 RNAs (7). 2 A variety of non-canonical caps involving nucleotide metabolites (Figure 1A) have also recently 3 been described (8,9). For example, nicotinamide adenine dinucleotide (NAD) and coenzyme A 4 (CoA) were found as cap-like structures in bacterial small RNAs (10) and the NAD cap was also 5 found in yeast and human mRNA and non-coding RNAs (11). Julius and Yuzenkova expanded 6 the potential repertoire of caps by demonstrating that a variety of nucleotide metabolites could 7 initiate transcription by bacterial RNA polymerase (RNA Pol) in vitro, including flavin adenine 8 dinucleotide (FAD), uridine diphosphate glucose (UDP-Glc), and uridine diphosphate N-9 acetylglucosamine (UDP-GlcNAc) (9). They also showed that capping with NAD and UDP 10 analogs by bacterial RNA Pol is promoter-specific and stimulates promoter escape (9), 11 suggesting a role for metabolite caps in regulating gene expression. For example, the NAD cap 12 has been shown to influence RNA stability and turnover, and is a substrate for decapping 13 enzymes (11). However, the lack of sensitive and specific analytical methods has hindered the 14 systematic study of the cap landscape dynamics in cells. 15Analysis of RNA cap structures has traditionally relied on radioisotope labeling and enzymatic 16 hydrolysis, followed by thin-layer and other types of chromatography to resolve cap structures 17 (12)(13)(14). While sensitive, the radiolabeling approach lacks specificity (12) and has the potential 18 to create cellular toxicity artifacts (15,16). While two-dimensional electrophoresis (14) allows 19 multiple caps analysis, it (i) lacks specificity for identifying intact cap structures, (ii) is limited to 20 NpppN caps, (iii) does not provide absolute quantification, and (iv) is semi-quantitative at b...