Comments from curator
Phenotyping data come from
Supplementary Table S1 of the paper (neu.2024.0314_supp_tables1.docx;
journal page,
PMC12677224), which gives for each DGRP line the percent mortality 24 h after traumatic brain injury (TBI) of flies not exposed or pre-exposed to an anesthetic, and the resulting percent difference in mortality (anesthetic preconditioning). Only these line-level values were published (the paper says that the primary data are available from the authors' lab), so the study has a summary dataset only.
Design: DGRP lines from the Bloomington stock center were reared at 25 °C on cornmeal-molasses-yeast food. Flies were collected under CO
2 at 0-7 days old and aged 24 h (1-8 days old). Mixed-sex groups (about 50:50 females:males) were used, because female and male
w1118 flies have the same early mortality after TBI, so the sex is NA. Flies were exposed for 15 min to an anesthetic in air in a multi-chamber anesthesia apparatus, or not exposed, then transferred to empty vials without CO
2, immediately injured with the High-Impact Trauma (HIT) device and kept on food at 25 °C. Flies that remained immobile at the bottom of the vial 24 h later were scored dead ("early mortality"). At least 6 samples of 15-30 flies were tested per line and condition.
Two separate screens:
-
Isoflurane: 171 lines in the paper (170 distinct lines in Table S1, see below), pre-exposure to 2.0% isoflurane, injury by 4 strikes at 5-min intervals on a first HIT device.
-
Sevoflurane: 144 lines, pre-exposure to 3.5% sevoflurane, injury by 3 strikes at 5-min intervals on a second HIT device (the first one broke). The number of strikes was chosen to give the same mortality to
w1118 flies on both devices, but unexposed DGRP flies died more in the isoflurane screen than in the sevoflurane screen.
139 lines are in both screens according to the paper (140 in Table S1, see below). The unexposed flies of each screen are therefore separate phenotypes, named after their TBI protocol (TBI4: 4 strikes, TBI3: 3 strikes). The anesthetic doses are the ones that gave the strongest preconditioning in the
w1118 dose-response experiments of the paper.
The summary dataset has 6 phenotypes (unit %) for 174 lines:
-
mn_Mort24_TBI4 and
mn_Mort24_TBI4_Iso: mean percent mortality of unexposed and isoflurane-exposed flies, isoflurane screen.
-
mn_Mort24_TBI3 and
mn_Mort24_TBI3_Sevo: the same for the sevoflurane screen.
-
mn_Precond_Iso and
mn_Precond_Sevo: anesthetic preconditioning, as published: percent difference in mortality = 100 × (mean % mortality unexposed − mean % mortality exposed) / mean % mortality unexposed. Positive values mean that the anesthetic protects against TBI, negative values that it increases mortality. These are the phenotypes of the authors' GWAS (DGRP2 web server, 1,920,276 SNPs with MAF ≥ 0.05): 10 SNPs in 7 genes for isoflurane (Table 1, e.g.
Ir41a,
klu) and 4 SNPs in 3 genes for sevoflurane (Table 2, e.g.
Ca-alpha1D,
Dip-ζ), at P < 10
-5.
Notes on Table S1 (values uploaded as published):
- Table S1 has three blocks: the 139 lines of both screens, then 5 lines of the sevoflurane screen only and 32 lines of the isoflurane screen only.
- Line 359 is listed twice with different isoflurane values: in the block of both screens (58.16, 51.97 and 10.64% for unexposed, exposed and percent difference) and again in the isoflurane-only block (73.86, 67.64 and 8.42%). The first row, one of the 139 lines of Figure 6, is uploaded; the second is not. The isoflurane phenotypes therefore have 170 lines instead of 171.
- Line 362 is listed both in the sevoflurane-only and in the isoflurane-only block; the two rows are merged, so 140 lines have values for both screens (139 in the paper).
- RAL-539 and RAL-691 (sevoflurane screen) are not in the sequenced DGRP Freeze 2 panel; they were added to DGRPool as DGRP_539 and DGRP_691 for this study.
- The percent differences agree with the two published means (recomputed to within 0.005), except for line 892 with sevoflurane (published 19.7, recomputed 19.90).
- The paper's mean percent difference for isoflurane (14.11) and its correlation with unexposed mortality (R = 0.19) are those of the 139 lines of both screens; over the 170 uploaded isoflurane lines they are 14.43 and 0.08. The ranges (-19.16 to 52.58 with isoflurane, 52.57 in the text; -32.07 to 52.76 with sevoflurane), the sevoflurane mean (12.99 over the 139 lines, 12.88 over 144 lines) and the correlation between the isoflurane and sevoflurane percent differences (R = -0.12 over the 139 lines, -0.11 over 140) agree with the paper.
- The paper cites line 584 as preconditioned by both anesthetics, but Table S1 gives it a percent difference of -1.34 with isoflurane.
- The unexposed mortalities are new measurements: they differ from the mortality index of
study 20 (R = 0.21, no identical values).
Not uploaded: the dose-response experiments on
w1118 flies (Figures 1-3 and Supplementary Figure S1: isoflurane, sevoflurane and ether), which do not involve DGRP lines, and the GWAS results.
Related DGRPool studies by the same groups:
study 20 (Katzenberger et al. 2015, mortality index after TBI without anesthetic),
study 118 (Schiffman et al. 2020, isoflurane
post-exposure after TBI in 141 lines) and
study 110 (Scharenbrock et al. 2020, ten of those lines). This study uses anesthetic
pre-exposure instead.