Acute opioid responses are modulated by dynamic interactions of Oprm1 and Fgf12
- University of Tennessee Health Science Center, United States
- University of California, San Diego, United States
- Washington University School of Medicine, United States
- Marshall University, United States
- Ball State University, United States
- University of Pennsylvania, United States
- Nash Family Department of Neuroscience and Friedman Brain Institute, Icahn School of Medicine at Mount Sinai, United States
Figures
Figure 1 with 2 supplements
Time series of quantitative trait loci (QTLs) for morphine-induced locomotor response.
Locomotion data were quantile-normalized and mapped against whole genome sequencing (WGS)-based genotype using GEMMA in https://genenetwork.org/. QTLs for male (a) and female (b) are stacked up by increasing 15 min time intervals. The dotted lines represent the genome-wide significance level of ~3.77. The color shaded areas indicate consistent associations across time bins for the Chr 10 locus (red) and the Chr 16 locus (blue). Strong correlation in morphine-induced locomotor response between male and female BXD strains is shown in Pearson’s correlation scatter plots during the 45–60 min (c) and 165–180 min (d) time frames.
Figure 1—figure supplement 1
Distribution of raw locomotion data after morphine injection.
Distribution of raw data for locomotion after morphine injection at different time intervals in females (orange A–K) and males (blue L–V). Data for 135–150 min were lost for both sexes.
Figure 1—figure supplement 2
Distribution of quantile-normalized locomotion data after morphine injection.
Data are plotted for each time bin for both females (orange A–K) and males (blue L–V). Data for 135–150 min were lost for both sexes.
Figure 2
Quantitative trait loci (QTLs) of morphine-induced locomotion between 45 and 60 min on Chr 10.
(a) QTL for males with a peak –logp of 10.06 (n=63 strains). (b) QTL for females with a peak –logp of 9.60 (n=64 strains). (c) Zoomed-in view of QTL in males. (d) Zoomed-in view of QTL in females. (e) Cis-eQTL for Oprm1 in the nucleus accumbens (NAc) of the BXDs, with a peak –logp of 10.01 (n=34 strains). (f) Cis-eQTL for Oprm1 in the hippocampus of the BXDs, with a peak –logp of 9.7 (n=67 strains) on Chr 10 at 5.6 Mb. (g) Oprm1 neighborhood in BXD family with SNP densities. (h) Haplotype map of the eQTL region. The ‘B’ of BXD is the mother, and ‘D’ is the father. GEMMA with LOCO was used for all association mapping.
Figure 3
Quantitative trait loci (QTLs) of morphine-induced locomotion between 165 and 180 min on Chr 16.
(a) QTL in males with a peak –logp of 4.28 (n=63 strains). (b) QTL for females with a peak –logp of 10.56 (n=64 strains). (c) Zoomed-in view of the QTL in males. (d) Zoomed-in view of the QTL in females. (e) Cis-eQTL in the striatum of the BXDs, with a peak –logp of 4.43. (f) Cis-eQTL in the ventral tegmental area (VTA) of the BXDs, with a peak –logp of 4.06. (g) Histogram of the normalized expression of Fgf12 in striatum and zoomed-in view of the QTL region. (h) Histogram of the normalized expression of Fgf12 in VTA and zoomed-in view of the QTL region. GEMMA with LOCO was used for all association mapping.
Figure 4
Chr 16 locus at 165–180 min after morphine injection.
(a) Zoomed-in view of the Chr 16 locus, from 26 to 29 Mb, showing individual SNPs (blue dots) and genes (purple horizontal lines) in the region. (b) A screenshot of GeneNetwork showing SNPs that inhabit this region. Almost all B vs. D SNPs (orange hash along x-axis) are restricted to two regions.
Figure 5 with 1 supplement
Epistatic interaction among genotype combinations.
(a) Males and (b) females with different combinations of the ‘B’ (i.e. B/B) or ‘D’ (i.e. D/D) genotypes yield different distances traveled after morphine injection over 120 min. Distance traveled (m) between the two genotypes for each loci is shown for each time point in (c–f) males and (g–j) females. Error bars represent standard errors.
Figure 5—figure supplement 1
Pairwise linkage statistics across the genome.
(a) Genome-wide two-dimensional heat map showing all pairwise marker combinations, illustrating both additive (main) effects and epistatic (interaction) effects. The upper-left triangle displays LOD scores for interaction terms, while the rightmost column shows the main-effect (single-locus) genome scan. (b) Zoomed-in view of linkage statistics between chromosomes 10 and 16. The red band along the bottom represents a strong additive effect on chromosome 16.
Figure 6 with 2 supplements
Oprm1 and Fgf12 are positively correlated in rat nucleus accumbens (NAc) suggested by single-nucleus RNA-seq (snRNA-seq).
(a) UMAP visualization of cell clusters from 4495 nuclei from rat NAc core. (b) Dot plot showing the expression level of cell-type marker genes in cell clusters. The shade of dots denotes normalized and scaled average expression, and the size of dots denotes the percentage of cells expressing the gene in each cell cluster. (c, d) Violin plots indicating the normalized and scaled expression level of Oprm1 (c) and Fgf12 (d) across cell clusters. (e–g) Scatter plots showing the correlation relationship between Oprm1 and Fgf12 in all cells (e), D1-MSN-3 (f), and D1-MSN-2 (g). Nuclei counts (n), Pearson’s correlation coefficients (r), and p-values (p) are labeled on each panel. UMAP, uniform manifold approximation and projection; D1-MSN, Drd1-expressing medium spiny neuron; D2-MSN, Drd2-expressing medium spiny neuron; GABA, GABAergic inhibitory neuron; Glut, glutamatergic excitatory neuron; Ol, oligodendrocyte; OPC, oligodendrocyte precursor cell; Ast, astrocyte; Mg, microglial cells.
Figure 6—figure supplement 1
Quality control (QC) of single-nucleus RNA-seq (snRNA-seq) and number of cells per cluster.
(a–e) Violin plots showing QC parameters used for selecting high-quality nuclei in each sample: (a) Unique gene numbers per nuclei (nFeature_RNA). (b) Log-transformed total read numbers per nuclei (log-nCount_RNA). (c) Percentage of mitochondrial gene reads. (d) Percentage of small 40S ribosomal (Rps) gene reads. (e) Percentage of large 60S ribosomal (Rpl).
Figure 6—figure supplement 2
Gene expression of D1- and D2-type medium spiny neuron subtypes.
Expression of Oprm1 and Fgf12 in acute and repeated nucleus accumbens (NAc) and in primary striatal culture from the Ratlas database. (a) Expression of Oprm1 in acute NAc. (b) Expression of Fgf12 in acute NAc. (c) Expression of Oprm1 in both acute and repeated NAc. (d) Expression of Fgf12 in both acute and repeated NAc. (e) Expression of Oprm1 in primary striatal culture. (f) Expression of Fgf12 in primary striatal culture.
Figure 7 with 1 supplement
Modeling the mechanistic interactions between genetic variants and phenotypes using Bayesian network.
This network illustrates the relationship between genetic variants in the Chr 10 and Chr 16 quantitative trait locus (QTL) regions and the differential expression of candidate genes Oprm1 and Fgf12 in the BXD family. Additional gene expression data of MAP kinases were retrieved from https://genenetwork.org/. Morphine-induced locomotion responses at different time bins are also included in the network. This causal hypothesis was developed using the Bayesian network framework available in GeneNetwork, where the arrow is indicative of the direction of the causal relationship. Additional genes included in the input of the network construction but had no connection to the final network were excluded in the illustration.
Figure 7—figure supplement 1
Levels of Oprm1, Fgf12, and MAP kinases proteins in the brains of BXD parental strains and F1s.
Effect of genotype on associated MAP kinases and genes for both parental strains and F1 generations. The genotype effect shows the B allele is associated with high expression of Oprm1 and the MAP kinases Mapk8ip2, Map3k11, and Map3k12, but low expression of Fgf12, when compared to the D allele.
Figure 8 with 1 supplement
Quantitative trait loci (QTLs) for naloxone-induced morphine withdrawal responses in male and female BXD mice.
Behavior data were quantile-normalized and mapped against whole-genome sequencing (WGS)-based genotypes using GEMMA in https://genenetwork.org/. (a) Number of jumps 15 min after naloxone injection in males. (b) Number of jumps 15 min after naloxone injection in females. (c) Change in locomotion measured by the last 15 min of morphine locomotion response minus the first 15 min after naloxone injection in males, with a peak on Chr 10 and a peak on Chr 16. (d) Change in locomotion measured by the last 15 min of morphine locomotion response minus the first 15 min after naloxone injection in females, with a peak on Chr 16. (e) Horizontal activity (distance traveled) 0–15 min after naloxone injection for males. (f) Horizontal activity (distance traveled) 0–15 min after naloxone injection for females, with a peak on Chr 6 and a peak on Chr 12. (g) Number of beam breaks in an open field 0–15 min after naloxone injection in females, with a peak on Chr 6 and a peak on Chr 12. (h) Salivation level in females. Dashed lines: genome-wide significance threshold. Dotted lines: threshold for suggestive significance.
Figure 8—figure supplement 1
Naloxone phenotypes vs. genotypes.
The turquoise points show morphine-induced locomotion (distance traveled in cm) between 165 and 180 min after injection. Orange points show the change in locomotion measured by the last 15 min of morphine locomotion response minus the first 15 min after naloxone injection. (a) Effect of genotype on two naloxone phenotypes in males. (b) Effect of genotype on the number of beam breaks in an open field 0–15 min after naloxone injection in males. (c) Effect of genotype on two naloxone phenotypes in females. (d) Effect of genotype on the number of beam breaks in an open field 0–15 min after naloxone injection in females. Error bars represent standard errors.
Author response image 1
Tables
Table 1
Genome-wide significant loci morphine-induced locomotion or naloxone-induced withdrawal.
| Locus position (Mb, GRC38) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Locus | Sex† | Peak –logp‡ | Chr | Prox | Peak | Distal | 1.5 –logp interval | Add. effect§ | GN short phenotype description | Time period (min) | GN BXD trait |
| Mor1a | F | 4.42 | 1 | 72.3 | 77.6 | 78.4 | 6.1 | –0.42 | Morphine-induced locomotion, distance traveled | 0–15 | 11583 |
| Nalx1a | MF | 3.96 | 1 | 124.0 | 128.5 | 141.8 | 17.9 | –0.43 | Naloxone-induced withdrawal, salivation | NA | 11875 |
| Nalx1b | MF | 4.14 | 1 | 163.3 | 167.2 | 168.2 | 4.8 | –0.44 | Naloxone-induced withdrawal, salivation | NA | 11875 |
| Mor1b | F | 4.32 | 1 | 173.4 | 173.5 | 176.2 | 2.8 | –0.38 | Morphine-induced locomotion, distance traveled | 150–165 | 11584 |
| Mor4a | M | 4.06 | 4 | 19.7 | 21.7 | 33.1 | 13.4 | –0.39 | Morphine-induced locomotion, distance traveled | 45–60 | 11331 |
| Mor5a | F | 5.01 | 5 | 9.8 | 9.8 | 13.4 | 3.5 | 0.52 | Morphine-induced locomotion, distance traveled | 105–120 | 11581 |
| Mor5b | F | 3.86 | 5 | 76.1 | 82.2 | 90.4 | 14.2 | 0.41 | Morphine-induced locomotion, distance traveled | 30–45 | 11587 |
| Nalx6a | MF/F | 4.69 | 6 | 112.1 | 116.9 | 122.8 | 10.7 | –0.42 | Naloxone-induced withdrawal, horizontal activity | 0–15 | 11871 |
| Mor7a¶ | M/F | 5.28 | 7 | 96.0 | 96.8 | 97.5 | 1.5 | –0.44 | Morphine-induced locomotion, distance traveled | 0–15 | 11326 |
| Mor9a | F | 3.90 | 9 | 83.1 | 83.9 | 86.1 | 3.0 | 0.45 | Morphine-induced locomotion, distance traveled | 120–135 | 11582 |
| Mor10a* | M/F | 10.53 | 10 | 5.6 | 5.6 | 9.6 | 4.0 | –0.59 | Morphine-induced locomotion, distance traveled | 30–45 | 11330 |
| Nalx12a | MF/F | 4.91 | 12 | 16.6 | 27.6 | 28.5 | 11.9 | 0.46 | Naloxone-induced withdrawal, horizontal activity | 0–15 | 11871 |
| Mor12a | F | 4.03 | 12 | 69.6 | 70.8 | 74.7 | 5.2 | –0.42 | Morphine-induced locomotion, distance traveled | 0–15 | 11583 |
| Mor12b | F | 5.71 | 12 | 98.6 | 99.6 | 102.5 | 3.9 | –0.44 | Morphine-induced locomotion, distance traveled | 30–45 | 11587 |
| Nalx14a | M | 4.00 | 14 | 49.1 | 52.1 | 61.2 | 12.1 | –0.41 | Naloxone-induced withdrawal, number of jumps | NA | 11336 |
| Nalx14b | MF | 4.36 | 14 | 118.4 | 118.5 | 119.4 | 1.0 | –0.25 | Naloxone-induced withdrawal, horizontal activity | 0–15 | 11871 |
| Mor16a* | F/M | 10.56 | 16 | 27.0 | 27.5 | 29.9 | 2.8 | –0.61 | Morphine-induced locomotion, distance traveled | 165–180 | 11585 |
| Nalx16b | MF | 4.02 | 16 | 57.5 | 59.7 | 62.2 | 4.6 | 0.42 | Naloxone-induced withdrawal, salivation | NA | 11875 |
-
*
Significant for morphine and naloxone treatments; best candidate for Mora10a is Oprm1; best for Mor16a is Fgf12.
-
†
MF = male and female joint mapping. M/F=significant in both sexes, but –logp value applies to first entry.
-
‡
–logp values given for quantile normalized traits but verified on other scales. Values are genome-wide p≤0.05 at ≥3.80.
-
§
Positive additive effects mean higher trait values linked to D genotypes. Negative values linked to high B values.
-
¶
Strong candidate gene for Mor7a is Tenm4.
Table 2
List of candidate genes.
| Candidate gene | Chr | Start (Mb) | Stop (Mb) | Male trait ID | Female trait ID | Phenotype |
|---|---|---|---|---|---|---|
| Tenm4 | 7 | 96 | 97.5 | BXD_11326 | BXD_11572 | Locomotion 0–15 min after morphine injection |
| Oprm1 | 10 | 5.6 | 9.6 | BXD_11331 | BXD_11588 | Locomotion 45–60 min after morphine injection |
| Oprm1 | 10 | 5.6 | 9.6 | BXD_11339 | BXD_11596 | Differences in locomotion before vs. after injection |
| Slc7a7/slc7a8 | 14 | 49.1 | 61.2 | BXD_11336 | BXD_11850 | Number of jumps after naloxone-induced withdrawal |
| Fgf12 | 16 | 27.0 | 29.9 | BXD_11328 | BXD_11585 | Locomotion 165–180 min after morphine injection |
| Fgf12 | 16 | 27.0 | 29.9 | BXD_11339 | BXD_11596 | Differences in locomotion before vs. after injection |
-
Potential candidate genes in QTL regions that were further investigated in this study are provided. For details on any given trait, please visit the corresponding GN website by embedding the trait ID in the URL such as https://genenetwork.org/show_trait?trait_id=11331&dataset=BXDPublish (for trait BXD_11331, morphine response (50 mg/kg ip), locomotion (open field) from 45 to 60 min after injection in an activity chamber for males [cm]).
Additional files
-
Supplementary file 1
Genome-wide significant loci associated with morphine and naloxone responses.
- https://cdn.elifesciences.org/articles/108845/elife-108845-supp1-v1.xlsx
-
Supplementary file 2
Detailed information on all morphine traits analyzed in this work.
- https://cdn.elifesciences.org/articles/108845/elife-108845-supp2-v1.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/108845/elife-108845-mdarchecklist1-v1.pdf
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Acute opioid responses are modulated by dynamic interactions of Oprm1 and Fgf12
eLife 14:RP108845.
https://doi.org/10.7554/eLife.108845.3
