Skip to main content
. 2024 Sep 17;7(12):e202402849. doi: 10.26508/lsa.202402849

Figure 1. Multi-isotope imaging mass spectrometry demonstrates intranuclear heterogeneity of new RNA synthesis.

Figure 1.

(A) Strategy for multiplexed measurement of new RNA and DNA architecture. 13C- or 2H-thymidine was administered to proliferating cells during serial passage to saturate DNA labeling, followed by pulse labeling with 15N-uridine as a tracer for new RNA. Cells were analyzed by multi-isotope imaging mass spectrometry with a representative example shown of a THP-1 cell administered with 15N-uridine for 60 min. 32S images show histological details, including cellular contours, the nucleus (N), and the cytoplasm (C). Hue, saturation, and intensity images display isotope ratio measurements, quantitatively mapping DNA labeling (13C/12C) and new RNA labeling (15N/14N). The lower bound of the scale is set to the background ratio (0%), and the upper bound is set to reveal differences in labeling (50% and 150% above background, respectively). Scale bar: 1 μm. (B) 2H-thymidine–labeled THP-1 cells pulsed with 15N-uridine for 120 min with/without RNA synthesis inhibitor (actinomycin D). 31P mass images (top) show nuclei in a pattern qualitatively similar to DAPI because of the high phosphorous content of chromatin. Scale bar: 5 μm. (C) 15N-uridine pulse-chase labeling of THP-1 cells. Scale bar: 5 μm. (D) Nuclear quantification of 15N-uridine pulse-chase labeling. (E) Cytoplasmic quantification of 15N-uridine pulse-chase labeling: pulse compared with chase timepoints by ANOVA with Dunnett’s adjusted P-values shown. (F) Isotope ratio data for intranuclear pixels from 120-min pulse and 120-min chase cells. Left: pixel distribution dot plots with mean line. Right: statistical metric of dispersion (median absolute deviation, MAD) for the pixel distributions with P-value reported for an unpaired t test. (G) Mean nuclear 15N-uridine labeling (120 min) in THP-1 cells stimulated with lipopolysaccharide and/or the bromodomain inhibitor (JQ1), with significance assessed by two-way ANOVA for overall lipopolysaccharide effect (P = 0.0009) and JQ1 effect (P = 0.023). Individual P-values for multiple comparisons are provided in the figure.