Differentiating human osteoclasts

To characterize transcriptional reprogramming during human osteoclastogenesis, we first determined the dynamic gene expression patterns of myeloid progenitors differentiating into mature osteoclasts. Therefore, human peripheral blood CD14+ monocytes from eight anonymous female blood donors aged 18–49 years were isolated and differentiated into osteoclasts using concomitant treatment with M-CSF for 9 days (Days 0-9) and RANKL for 7 days (Days 2–9). RNA was harvested on Day 0 and at Days 2, 5, and 9 after induction of differentiation (Fig. 1a). The activity of the osteoclast marker tartrate-resistant acid phosphatase (TRAcP) in the cell culture media was analyzed on Days 7 and 9. Furthermore, resorptive activity was determined by seeding and incubating osteoclasts on Day 9 in bovine bone slices for 72 h.

Fig. 1

RNA collection during human osteoclast differentiation. a Schematic representation of the experimental setup. b Light microscopy images of mature osteoclasts on Day 9 and of resorption pits on Day 12, absorbance-based TRAcP activity in media at Days 7 and 9, and quantification of the percentage of eroded surface per bone surface resorbed by 50 000 mature osteoclasts from Days 9 to 12 for each of the eight donors. c Box plot of RNA-seq-based gene expression levels for monocyte (upper panel) and osteoclast-specific (lower panel) genes. d Box plot of RNA-seq-based gene expression levels for lysophosphatidylcholine acyltransferase 2 (LPCAT2) and cytochrome c oxidase copper chaperone (COX11). e Histogram reporting the frequency of genes grouped by the number of donors with a Pearson correlation greater than 0.8 to the average expression level. Genes within black bars were considered for further analysis. f Principal component analysis plot based on genes with differential (FDR < 0.000 1 between at least two timepoints) and reproducible (≥6 donors with Pearson’s correlation >0.8 to the average) expression during human osteoclast differentiation. g Heatmap showing the Pearson correlation for log twofold changes in gene expression during osteoclast differentiation according to Rashid et al.20h Scatter plot comparing the log twofold changes between OC-like cells and PBMCs from Rashid et al.20 with log twofold changes occurring between Day 9 and Day 2 of osteoclast differentiation in the present study. Genes were selected based on upregulation (FDR < 0.01) within expression data from Rashid et al.20. i Box plot (band: mean; box: first and third quartiles; whiskers: 1.5 times the interquartile range) of RNA-seq-based gene expression levels for solute carrier family 6 member 7 (SLC6A7) in the present study (left panel) and from Rashid et al.20 (right panel)

We found that cells from all donors differentiated into TRAcP-expressing multinucleated osteoclasts capable of resorbing bone as previously reported.16,17 TRAcP and resorptive activity varied between donors (Fig. 1b). RNA-seq analyses revealed time-dependent downregulation of monocyte-specific genes (e.g., TREM1, SELL, and CLEC10A)18 and monocyte-macrophage markers (e.g., ADGRE1)19 and a similar time-dependent increase in the expression of osteoclast-specific marker genes (e.g., CTSK, ACP5, MMP9, and CA2) from Day 0 to Day 9 (Fig. 1c). Analyses of the variance in gene expression at different time points revealed 10 849 genes that were differentially expressed; i.e., the level of gene expression changed between at least two time points across donors. For some of these genes, such as COX11, the donors exhibited a highly similar temporal expression pattern, whereas the expression of other genes, e.g., LPCAT2, exhibited substantial interdonor variation (Fig. 1d). To avoid donor-dependent gene expression patterns, we included only differentially expressed genes with high similarity across donors (Fig. 1e, f). Thus, we obtained 8 980 genes that were differentially expressed during osteoclast differentiation and formed the basis of subsequent analyses.

By comparing our dataset with recently published gene expression data from human myeloid progenitors and differentiated osteoclasts,20 we found a strong correlation between the datasets for upregulated genes (Fig. 1G), particularly when comparing Days 2 and 9 of our differentiation protocol, for which we also reached similar magnitudes of gene induction (Fig. 1H). This finding aligns well with the fact that the authors used mononucleated cells after 2 days of stimulation with M-CSF as a starting point, while we used CD14+ monocytes.20 Therefore, we did not identify nonmonocyte genes, such as the CD8+-T-cell related gene EOMES (Fig. 1i, left panel), and we did not observe early gene regulation upon M-CSF stimulation, as exemplified by SLC6A7, which was subsequently strongly upregulated in mature osteoclasts in both datasets (Fig. 1i, right panel).

RNA-seq reveals temporal gene expression patterns with distinct cellular functions and implications for human bone biology

Using k-means clustering, we observed that the 8 980 genes could be grouped into eight temporal gene expression patterns (clusters) (Fig. 2a). Among these, two clusters were characterized by a transient decrease or increase in gene expression levels, and six clusters were characterized by early (peak on Day 2), middle (peak on Day 5), or late (peak on Day 9) changes in gene expression levels. Gene Ontology (GO) and Reactome pathway analyses revealed cluster-specific enrichment of distinct biological processes (Fig. 2b, c), such as metabolic reprogramming and mitochondrial activation, which were linked to early upregulated genes (Cluster 2) (Fig. 2b). In line with this, we found distinct temporal profiles for metabolic genes involved in glucose metabolism and the tricarboxylic acid cycle (TCA)-mediated metabolism of fatty acids (Fig. 2c). The downregulated genes were barely linked to metabolic processes but important for cytokine production and immune cell activation. Canonical markers of osteoclast function, such as CTSK, ACP5, DCSTAMP, and CA2 (ref.16,21,22,23,24), were among the late upregulated Cluster 4 genes that were involved in mature osteoclast-related processes, such as cytoskeleton organization, pH regulation, bone remodeling, cell migration, and cell‒cell fusion.25,26,27

Fig. 2

Temporal changes in gene expression patterns link osteoclast function to bone biology. a Heatmap showing scaled expression levels of the 8 446 differentially expressed genes among the eight k-means clusters for each sample. b Heatmap showing the false discovery rate (GOseq) for the enrichment of the gene clusters for biological process-annotated Gene Ontology (GO) terms. c Heatmap showing the false discovery rate (GOseq) for the enrichment of the gene clusters for pathways of the Reactome database. d Heatmap showing the P value (hypergeometric test) for the enrichment of the gene clusters for genes that increase or decrease bone mineral content or bone mineral density or genes causing abnormal bone structure, mineralization, and morphology in knockout mouse models from the International Mouse Phenotyping Consortium (IMPC). e Box plot of RNA-seq-based (with cluster membership, lines represent individual donors) gene expression levels for actinin alpha 2 (ACTN2) during human osteoclast differentiation and microarray-based (limma-based statistics) mRNA expression of ACTN2 in iliac crest biopsies from healthy (n = 39) and osteoporotic (n = 27)29 subjects. f Heatmap showing the P value (hypergeometric test) for the enrichment of the gene clusters for genes up- or downregulated in iliac crest biopsies of osteoporotic patients (op) versus healthy controls.29g Heatmap showing the enrichment of estimated bone mineral density (eBMD)-associated SNPs33 near genes whose expression changes dynamically during osteoclast differentiation. h Heatmap showing the P value (hypergeometric test) for the enrichment of the gene clusters for genes up- or downregulated during full or stress fracture in mice.34

To determine the associations of the differentially expressed genes with bone biology, we investigated whether there was an association between osteoclast gene expression profiles and genes associated with bone mineral density (BMD), bone mineral content (BMC), and/or bone morphology in knockout mouse models. Based on data from the International Mouse Phenotyping Consortium (IMPC),28 knockout phenotypes characterized by increased BMC and BMD (based on dual-energy X-ray absorptiometry [DXA]) were linked to genes repressed early during osteoclast differentiation, i.e., from monocytes to macrophages (Clusters 5, 6, and 7), while knockout phenotypes characterized by decreased BMC and BMD were linked to genes repressed during late osteoclast differentiation (Cluster 8) (Fig. 2d). Gene knockout phenotypes characterized by other skeletal defects on X-ray were linked to genes induced in mature osteoclasts (Cluster 4) and genes with decreased expression during osteoclastogenesis (Clusters 5-8). While disrupting osteoclast function is linked to bone phenotype-causing mutations in mice, these data indicate the existence of specific associations between subgroups of bone phenotypes and temporal gene expression patterns during osteoclast differentiation.

Next, we tested whether the differentially expressed gene patterns were aberrantly expressed in patients with osteoporosis. Using published microarray data on RNA in iliac crest bone tissue from 27 osteoporotic and 39 nonosteoporotic postmenopausal women,29 we found that the expression of ACTN2, a gene related to osteoclast fusion30 and a member of Cluster 4, was greater in bone from osteoporotic women (Fig. 2e). In a genome-wide context, we found an overlap between genes whose expression was upregulated in samples from osteoporotic women and genes whose expression was induced (Clusters 2 and 4) during osteoclast differentiation and vice versa for genes whose expression was repressed in both datasets (Clusters 7 and 8) (Fig. 2f). This finding suggested an increased abundance of osteoclast-specific markers and the absence of progenitor-related genes in bone in osteoporotic women. To further elucidate this phenomenon, we used published microarray data from peripheral blood monocytes (PBMs) from 73 nonosteoporotic pre- and postmenopausal Caucasian women grouped into low or high hip BMD groups assessed using DXA.31,32 However, in contrast to the clear enrichment patterns observed in the expression data from postmenopausal osteoporotic iliac crest biopsies, we could not find an association between genes whose expression was upregulated (Clusters 1–4) or downregulated (Clusters 5–8) during osteoclast differentiation and genes with altered expression in PBMs from nonosteoporotic pre- and postmenopausal women with low or high BMD (data not shown).

To test the association of differentiation-associated osteoclast genes with human genetics of bone mineral density, we used summary statistics from previously published GWAS data33 on heel quantitative ultrasound (eBMD) and tested the distribution of significant SNPs (P < 5 × 10–8) around the transcription start site of our clustered genes. We found that the density of SNPs associated with eBMD was greater for nearby genes with differential expression during osteoclast differentiation than for those with a genomic background (random distribution in the genome). This was especially the case for genes repressed after M-CSF stimulation (Cluster 7), late-induced genes (Cluster 4) and genes transiently repressed (Cluster 5) (Fig. 2g). These data indicate that genes linked to monocyte progenitors and mature osteoclasts, i.e., those whose expression strongly changes following differentiation from the macrophage stage, are likely to be affected by human sequence variations associated with alterations in eBMD.

Finally, we expanded our comparisons by moving from the steady-state analysis above to dynamic processes such as fracture healing. Using gene expression data34 from mice exposed to stress or a complete fracture, reflecting intramembranous and endochondral ossification processes, we observed a stronger and subsequently more coordinated association of the RNA-seq clusters with genes that are upregulated during the time course of fracture healing, especially in bones undergoing a full fracture (Fig. 2h). Taken together, these analyses show that distinct gene expression profiles throughout human osteoclastogenesis are implicated in bone development, bone remodeling during disease and fracture repair, as well as the genetics of bone density.

Transcriptional networks drive osteoclast differentiation

We next investigated whether the temporal changes in gene expression were linked through regulatory networks, i.e., whether gene induction during the late stages of osteoclast differentiation was a consequence of early changes in gene expression. We applied the machine learning algorithm “Integrated System for Motif Activity Response Analysis” (ISMARA)35 to model transcription factor activity and to predict target genes of each transcription factor using gene expression data and motif occurrences in promoter regions. Importantly, the algorithm predicted a strong increase in the motif activity of two key transcription factors involved in osteoclast differentiation, NFATC1 and JUN36 (Fig. 3a). We also found that more than half of the 682 transcription factors that were found in the ISMARA motif database and our expression dataset exhibited changes in activity during differentiation (Fig. 3b); these genes included many of the transcription factors annotated in the GO terms “osteoclast differentiation” or “bone resorption”. In addition to the changes in activity, ISMARA predicts target genes for a given transcription factor, i.e., genes that are very likely to depend on the presence of that specific factor in the given expression dataset. In accordance with the continuous increase in JUN and NFATC1 activity during osteoclast differentiation and their documented role in the maturation of osteoclasts,36 we found that most of the predicted target genes, i.e., members of Cluster 4, were upregulated during the late phase of differentiation (Fig. 3c). To test this model for target gene prediction, we performed coexpression analysis of transcription factors and their predicted target genes on published single-cell RNA-seq (scRNA-seq) data from in vitro differentiated human osteoclasts.37 Due to the sparse nature of the scRNA-seq data, we first performed deep clustering to estimate the correlation between the coexpression of transcription factors and predicted targets at the cluster level (Fig. 3d). As exemplified by the predicted target genes for MYB-related protein 2 (MYBL2), a transcription factor known to regulate cell cycle genes,38 which were enriched in Cluster 3 (data not shown), we found strong overlap between the expression of MYBL2 and the predicted target genes at both the single-cell (Fig. 3e) and cluster (Fig. 3f) levels. Importantly, similar associations were observed only among transcription factors with cluster-specific expression patterns (Fig. 3g), suggesting that the expression of target genes follows the expression pattern of the regulating transcription factors.

Fig. 3

Machine learning highlights the transcriptional networks involved in human osteoclastogenesis. a Box plot of RNA-seq-based motif activity using ISMARA for NFATC1 and JUN during human osteoclast differentiation. Lines represent individual donors. b Heatmap showing the motif activity of transcription factors with differential activity (P value < 0.001) during human osteoclast differentiation. c Circular plot showing the ISMARA-predicted target genes of NFATC1 and JUN. d UMAP plot of scRNA-seq data from in vitro differentiated human osteoclasts on Day 14 of differentiation. e Gene expression levels of MYBL2 and the sum of MYBL2 target genes in a UMAP plot of differentiated human osteoclasts at the single-cell level. f Average cluster expression levels of MYBL2 versus the sum of MYBL2 targets in differentiated human osteoclasts at the single-cell level. g Gene set enrichment analysis of 26 transcription factors with cluster-specific expression patterns among the 329 transcription factors that were ranked according to Spearman’s correlation for transcription factor and target gene expression at the cluster level (as illustrated for MYBL2 in 3 F). h Network enrichment analysis (NEAT) showing significantly enriched regulatory relationships within and between RNA-seq clusters based on the ISMARA-predicted target genes. i Genome-wide associations and their predicted causal genes for estimated bone mineral density (eBMD) were filtered for transcription factor information

We next constructed a directed transcriptional network by combining the predicted target genes of all 682 transcription factors from the ISMARA database via network enrichment analysis.39 This analysis demonstrated that members of the early, transiently induced gene cluster (Cluster 1) were particularly important for the regulation of genes in the early and middle-induced clusters (Clusters 2 and 3), which in turn controlled the induction of osteoclast-specific genes (Cluster 4) (Fig. 3h). Using the transcriptional network approach, we showed that the proportions of transcription factors highly relevant for the regulation of mature osteoclast genes (Cluster 4) and/or causal eBMD genes33 were independent of the factor itself, which changed expression levels or was a causal eBMD gene (Fig. 3i). For example, we identified transcription factors, including members of the homeobox A and myocyte enhancer family, such as HOXA10 and MEF2A, that are predicted to regulate genes related to mature osteoclasts and bone mineral density despite not being differentially expressed or associated with eBMD SNPs (Fig. 3I). Taken together, these network analyses demonstrated associations in the stepwise remodeling of the transcriptional networks regulating osteoclast differentiation, i.e., feedforward loops in gene regulation to overcome transitions between states of cellular differentiation in osteoclasts. In addition, our data suggested that network-based analysis can predict transcription factors important for osteoclast function based on the level of transcription factor activity, i.e., factors that contribute to the regulation of osteoclast genes through post-transcriptional mechanisms.

Subgrouping of osteoclast transcriptional networks

As it was recently suggested that mature murine osteoclasts can undergo fission into osteomorphs, which are transcriptionally distinct from osteoclasts and macrophages and can be recycled back into osteoclasts,40 we questioned whether our gene expression data could be used to subgroup osteoclast transcriptional networks and provide insight into the transcriptional regulation of osteomorph-related genes. Using previously published gene signatures of osteomorphs, osteoclasts and monocytes in vivo,40 we found that the human orthologs of the 132 genes that were selectively expressed at high levels in osteomorphs (only) and of the 448 genes that were highly expressed in both osteomorphs and osteoclasts (commonly) were often dynamically expressed within our dataset and among the variable genes of the previously published human osteoclast single-cell RNA-seq data37 (Fig. 4a). Specifically, osteomorph-selective genes showed distinct enrichment from common osteomorph and osteoclast genes when aligned with our time course RNA-seq clusters (Fig. 4b) and marker genes of the single-cell RNA-seq clusters (Fig. 4c). Furthermore, osteomorph-selective genes were linked to cell proliferation, as highlighted by the strong overlap with RNA-seq Cluster 3 (Figs. 4b and 2b) and by the expression of osteomorph-selective genes in scRNA-seq clusters (Fig. 4d) with cells in proliferating cell cycle states (Fig. 4e). Using hypergeometric tests and our transcriptional network, we identified members of the E2F and HOX gene families that specifically regulate osteomorph-selective genes (Fig. 4f). However, this association was lost when correcting for multiple testing. Although neither our study nor that of Omata et al.37 provided culture conditions that support the presence of osteomorphs, we found that osteomorph-selective genes exhibit distinct dynamic expression patterns throughout human osteoclastogenesis, are partially linked to the cell cycle, and are likely regulated by transcription factors distinct from those regulating common osteoclast genes.

Fig. 4

Subpopulation specificity of osteoclast transcriptional networks. a Bar plot showing overlap of osteomorph-selective (only) and osteomorph-osteoclast-selective (common) genes, with genes being differentially expressed throughout osteoclast differentiation at the bulk level (Fig. 2a – dynamic RNA-seq) and genes being cluster-specifically expressed in the scRNA-seq of mature osteoclasts (Fig. 3d – scRNA cluster marker). b Heatmap showing the P value (hypergeometric test) for the enrichment of the bulk RNA-seq gene clusters in Fig. 2a for the osteomorph-selective (only) and osteomorph-osteoclast-selective (common) genes. c Heatmap showing the P value (hypergeometric test) for the enrichment of the scRNA-seq cluster markers in Fig. 3d for the osteomorph-selective (only) and osteomorph-osteoclast-selective (common) genes. d Dot plot showing the cluster expression levels of the osteomorph-selective genes that overlap with the markers of scRNA-seq Cluster 6 in Fig. 3d. e Bar plot quantifying the cell cycle score according to the scRNA-seq data of mature osteoclasts. f Heatmap showing the unadjusted P values (hypergeometric test) for the enrichment of ISMARA-predicted transcription factor target genes for the osteomorph-selective (only) and osteomorph-osteoclast-selective (common) genes. g CD14 and CTSK gene expression levels on a UMAP plot of differentiated human osteoclasts at the single-cell level. h UMAP plot of scRNA-seq data from in vitro differentiated human osteoclasts on Day 14 of differentiation clustered into two subpopulations of differentiating cells (left and right) and one population of undifferentiated cells (middle connective piece). i Heatmap of gene expression levels for left-, right- and mature-specific gene groups. j Heatmap showing the false discovery rate (GOseq) for the enrichment of the scRNA-seq signatures in 4I for biological process-annotated GO terms. k Heatmap showing the P value (hypergeometric test) for the enrichment of the bulk RNA-seq gene clusters in 2 A for the scRNA-seq signatures in 4I. l Heatmap showing the Benjamini–Hochberg adjusted P values (hypergeometric test) for the enrichment of ISMARA-predicted transcription factor target genes for the scRNA–seq signatures in 4I

Expanding our analysis with previously published scRNA-seq data,37 we found that the scRNA-seq dataset contained two groups of mature osteoclasts at the top of two arms that emerged from monocytes (Fig. 4g, h). Therefore, we assessed the biological functional differences between the two arms and found that the left arm-selective genes were enriched among biological processes such as lipid metabolism, cytokine production and cell migration, whereas the right arm-selective genes were enriched among mitochondrial processes, ATP transport, and translation (Fig. 4i, j). Notably, the left- and right-arm selective genes also exhibited distinct temporal expression patterns in our bulk RNA-seq profiles (Fig. 4k) and, importantly, could be clustered separately according to our gene regulatory network (Fig. 4l). Here, the well-known osteoclast transcription factors JUN and NFATc1 were important for mature osteoclast genes independent of the subpopulation, while other osteoclast-associated transcription factors (GO terms ‘osteoclast differentiation’ and ‘bone resorption’) showed specificity for the genes of the left (e.g., CEBPB, FOS, BCL6) or right arm (e.g., ATF1, NRF1, SIX5). Although it is unknown whether these differences were caused by one arm containing immature and nonfused mature osteoclasts or the other containing mainly fully fused osteoclasts, these analyses demonstrated that subpopulation-selective osteoclast genes have distinct temporal expression patterns and that osteoclast subtype specificity is defined by the interaction of a core osteoclast transcriptional network (e.g., NFATc1) with subtype-specific transcription factors.

Differentially expressed GPCRs during human osteoclast differentiation

From the perspective of identifying potential antiosteoporotic treatment targets, we speculated whether our dataset could be used to detect and test previously unrecognized regulators of osteoclast differentiation and activity. To test this possibility, we focused on the role of GPCRs in the transcriptional reprogramming of osteoclast differentiation, as GPCRs are implicated in bone biology and represent feasible therapeutic targets accounting for approximately 34% of FDA-approved drugs (2017) that target a total of 108 different GPCRs.41 An example of this is teriparatide, a bone anabolic antiosteoporotic drug that targets the parathyroid hormone (PTH) receptor on osteoblasts or their precursors.42 Furthermore, 36 GPCRs are known to be associated with altered BMD, bone morphology, and bone-related diseases such as arthritis in humans.43 For example, activating the calcitonin receptor, a bona fide osteoclast marker, reduces osteoclast activity.44

Using the GPCR database (GPCRdb),45 we identified 144 GPCRs and 43 gene-encoded ligands expressed during osteoclast differentiation (Fig. 5a). Interestingly, to a minor extent, GPCRs but mostly ligand‒receptor pairs were enriched among the differentially expressed genes (Fig. 5b, c). Focusing on all human-encoded nonolfactory GPCRs, we did not find them to be enriched among the 835 genes predicted to be causally related to eBMD.33 However, this change occurred when GPCRs that were expressed or differentially expressed during osteoclast differentiation were grouped (Fig. 5d). Among the differentially expressed GPCRs, we found well-described regulators of osteoclast function, such as CALCR as well as LGR4, which encodes a leucine-rich repeat containing G-protein coupled receptor 4 that acts as a receptor for RANKL, thereby inhibiting osteoclast differentiation and activity;46 and GIPR, which encodes the receptor of the incretin hormone glucose-dependent insulinotropic polypeptide (GIPR), which inhibits human osteoclast differentiation and activity.17 Taken together, these data indicate a pivotal role for differentially expressed GPCRs and their ligands in the regulation of osteoclast differentiation and activity and, in turn, bone mineral density in humans.

Fig. 5

G-protein coupled receptors are highly differentially expressed during osteoclast differentiation. a Bar plot showing the frequency of nonolfactory G protein-coupled receptor (GPCR), gene-encoded GPCR ligand and receptor‒ligand (G-L) pairs from the GPCRdb expressed at the mRNA level during human osteoclast differentiation. b Bar plot quantifying the enrichment of GPCRs, gene-encoded ligands and pairs divided into one or both among the genes differentially expressed during human osteoclast differentiation. c Heatmap quantifying the gene expression levels of gene-encoded ligands (left panel) and GPCRs (right panel) during human osteoclast differentiation grouped according to gene clusters. d Bar plot quantifying the enrichment of nonolfactory GPCRs among the genes predicted to be causal genes in the eBMD GWAS, grouped into all and those expressed or differentially expressed during human osteoclast differentiation. e Heatmap quantifying the preferential coupling of GPCRs from RNA-seq Clusters 1–8 to G proteins according to the GPCRdb

Based on these findings, we hypothesized that other GPCRs could affect osteoclast differentiation or function. To test this hypothesis, we focused on differentially expressed GPCRs with known downstream G protein-coupling,47 which left us with 64 GPCRs (Fig. 5e). Of these, we selected three (marked in red in Fig. 5e), namely, C5AR1 (complement 5a receptor 1), SSTR2 (somatostatin receptor 2), and FFAR4 (free fatty acid receptor 4/GPR120), because all these receptors are commercially available agonists and because their regulatory effects on human osteoclast differentiation and activity have not been previously investigated.

C5AR1 activation impairs osteoclast differentiation due to its progenitor-restricted expression

The expression of C5AR1, a Cluster 7 member, was strongly decreased during osteoclast differentiation (Fig. 6a). C5AR1 expression was detected via scRNA-seq in human monocytes and osteoclasts,37 in which C5AR1 expression was limited to undifferentiated cells (Fig. 6b), indicating that C5AR1 is not a functional receptor on mature osteoclasts. C5AR1 is a class A GPCR that predominantly couples to inhibitory G proteins (Gi) following ligand binding. This change leads to a decrease in intracellular cAMP levels