*4.5. Intracellular Ca2+ Imaging*

In order to measure acute opioid-induced changes in Ca2+ levels, cells were incubated for 30 min with Fura Red (Invitrogen), a ratiometric Ca2+ fluorescent indicator [78,79] stimulated with 750 nM opioids and Ca2+ levels were monitored using sequential, i.e., dual-track time-lapse CLSM imaging. In the first track, the 488 nm line of the Ar/ArKr laser was used to excite eGFP and Fura Red. The eGFP signal was collected using the band pass 505-530 nm emission filter, and the Fura Red signal was collected using the long pass 680 nm emission filter. In response to changes in Ca2+ concentrations, the excitation wavelength of Fura Red shifts from 472 nm at low Ca2+ concentration to 436 nm at high Ca2+ concentration [78]. As a result, the intensity of the Fura Red fluorescence signal decreases when intracellular Ca2+ levels increase and increases as they fall again. In the second track, the 543 nm HeNe laser was used to excite Tomato, and fluorescence was collected using the band pass 580–620 nm emission filter. The images were collected every 30 s for 40 min (in some cases up to 80 min). The pixel dwell time was 51.2 µs, and the images were collected without averaging.

#### *4.6. Western Blotting*

Transfected HEK293 cells expressing MOP-eGFP and 5-HT1A-Tomato were cultured in collagen-coated T25 flasks (Sarstedt) as described in the *Cell culture and transfection* section. At around 90% confluence, the cells were treated with opioids following the protocol described in the *Opioid treatment* section. Adherent cells were removed from the flasks with trypsin-EDTA (0.05%, Thermo Fisher Scientific), washed with ice-cold PBS and centrifuged at 1500 rpm for 5 min at 4 ◦C. The cell pellets were solubilized in RIPA lysate buffer (10 <sup>×</sup> <sup>10</sup><sup>6</sup> cells/mL) containing protease inhibitor and phosphatase inhibitor (Sigma-Aldrich) and incubated for 30 min on ice. The cell solution was transferred to Eppendorf tubes and centrifuged at 10 000 rpm for 10 min. The total protein content was determined colorimetrically using the BioRad RC DC Protein Assay (Bio-Rad). Samples (20 µg protein) were denatured at 70 ◦C for 10 min with 4X LDS Sample Buffer, 10X Sample Reducing Agent (Invitrogen) and water was added to a final volume of 30 µL. Samples were loaded on precast polyacrylamide gels (Invitrogen) along with 5 µL of pre-stained standard protein ladder (Thermo Fisher Scientific), electrophoresed and transferred onto a nitrocellulose membrane (Thermo Fisher Scientific). The membrane was probed overnight with antibodies against ERK1/2 (Invitrogen and Cell Signaling Technology, Inc., Danvers, MA, USA), phospho-ERK1/2 (Invitrogen and Cell Signaling Technology, Inc.), p38 (Cell Signaling Technology, Inc.), phospho-p38 (Cell Signaling Technology, Inc.) or β-actin (Invitrogen and Cell Signaling Technology, Inc.). Only one primary antibody was used at a time and the membrane was stripped for 15 min with a western blot stripping buffer (Thermo Fisher Scientific) in between the different primary antibodies. Depending on the primary antibody, either biotin or horseradish peroxidase conjugated secondary antibody (Invitrogen) was used for detection. Western blot experiments were repeated three times, starting from different cell cultures. The same trend was observed in all repetitions. Data from one representative experiment are shown (Supplementary Materials, Section S4. Western blotting (Figure S5)).

### *4.7. LC-MS/MS Opioid Metabolite Analysis*

In order to assess whether opioids were active compounds in their own right, HEK293 cells expressing MOP-eGFP and 5-HT1A-Tomato were cultured in collagen-coated T25 flasks (Sarstedt) as described in the Section 4.1. At 90% confluence, the cell media was supplemented with 750 nM of morphine, oxycodone, codeine or fentanyl. After 18 h of incubation, the cell media was collected, the cells were lysed, and the cell culture medium

and the lysate were stored at −20 ◦C. Prior to LC-MS/MS analysis, the samples were thawed and allowed to reach room temperature, diluted 10× with MilliQ water and filtered using Agilent/Whatman MiniUniPrep vials 0.2 µm PP (p/n 5190-1421). The samples were aliquoted (in triplicate) and subjected to LC-MS/MS for metabolite analysis. To this aim, a general protocol initially developed for urine analysis was used [80]. An LC-MS/MS analysis tested for the presence of morphine, morphine-3-glucuronide, morphine-6 glucuronide, normorphine, codeine, codeine-6-glucuronide, norcodeine, 6-acetylmorphine and ethylmorphine (Supplementary Materials, Section S5. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis (Tables S1 and S2)).

The analysis was performed using an ACQUITY UPLC I-Class system from Waters (Milford, MA, USA) coupled to a Waters Xevo TQD (Waters). All systems were controlled by MassLynx (Waters, version 4.1 SCN 940). Chromatographic separation was achieved on an ACE Excel 2 C18-PFP column (100 mm × 2.1 mm, 1.8 µm, Aberdeen, Scotland) kept at 60◦C. Mobile phase A consisted of 0.001% formic acid in 10 mM ammonium formate pH 5.2 and mobile phase B consisted of 0.001% formic acid in methanol. Initial gradient conditions were 1% B held for 1.5 min, then increasing to 5% B during 0.1 min following a ramping of B to 41% until 7.5 min. The following gradient steps were 95% B until 8 min following 95% B during 1 min before reaching 1% B again and equilibration during 1 min. Total run time was 10.1 min and LC flow was 0.5 mL/min. The injection volume was 3 µL. The electro spray ionization (ESI) interface was operating in positive mode and the mass spectrometer was operating in multiple reaction monitoring (MRM) mode with two transitions for each analyte and one transition for each internal standard according to the table below. Both quadrupoles were set in unit resolution. Data were processed using TargetLynx TM (MassLynx version 4.1 SCN940).

Analytes were identified by their retention time and transition ratio. Quantification of the analytes in the samples was performed using calibration samples with eleven concentrations of the analyte, as shown in Supplementary Materials (Section S6, Table S1).

#### *4.8. Statistical Analysis*

A student's *t*-test was used to determine whether the difference between the mean values measured in untreated and treated cells, or in cells treated with different opioids, are significantly different from each other. The results are reported using a two-tailed *p*-value (*p*). The Benjamini–Hochberg method to control the False Discovery Rate (FDR) in sequential modified Bonferroni correction for multiple hypothesis testing was thereafter applied. At an FDR value of 5%, *p* ≤ 0.012 was determined to be statistically significant.

#### **5. Conclusions**

The long-term usefulness of opioids in chronic pain treatment is hampered by side effects. Drugs targeting 5-HT1A have been shown to alleviate the adverse effects of prolonged opioid use, suggesting interactions between MOP- and 5-HT1A-mediated pathways. However, details of underlying mechanisms remain obscure. The aim of our study was to investigate whether these pathways can be integrated at the single-cell level by MOP- and 5-HT1A heterodimerisation. Our quantitative characterization of MOP-eGFP and 5-HT1A-Tomato interactions in live cells shows that these receptors can indeed form heterodimers in the plasma membrane of cells expressing physiologically relevant levels of these receptors. Our data show that under the conditions of our study, a surface density of (130 ± 10) molecules/µm<sup>2</sup> , about 33% of the 5-HT1A-Tomato receptors are bound in heterodimer complexes with MOP-eGFP in untreated cells.

In line with our hypothesis, we found an increase in MOP–5-HT1A heterodimer formation in cells simultaneously expressing MOP and 5-HT1A receptors following 18 h of incubation with fentanyl, morphine, codeine and oxycodone. All tested non-peptide opioids stabilized the heterodimer complexes and elicited a distinct down-stream cellular signalling response, as evidenced by Ca2+ imaging and ERK1/2 and p38 activation. An opioid metabolites analysis did not show any traces of common opioid metabolites, indicating that codeine was an active compound with a similar strength to morphine.

Taken together, our findings suggest that treatments hindering MOP–5-HT1A heterodimer formation could provide potentially new strategies to treat opioid induced hyperalgesia and help to preserve the analgesic effects of opioids. The development of new drugs targeting these mechanisms is therefore of interest.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/molecules27072350/s1. Subsections: S1. Transfection, positive and negative control cells, S2. Calculation of the apparent dissociation constant (derivation of Equation (6) given in the main text), S3. Relative Cross-Correlation Amplitude (RCCA) increased upon opioid treatment. Verification by switching FCCS, S4. Western blotting and S5. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. Figures: Figure S1. Transfection and cellular model generation; Figure S2. Determination of the RCCA for positive and negative control cells; Figure S3. RCCA increases upon opioid treatment; Figure S4. Switching FCCS; Figure S5. Western blots showing the activation of extracellular signal-regulated kinase (Erk1/2 and p-Erk1/2) and the p38 mitogen-activated protein kinase (p38/p-p38 MAPK) in HEK293 cells expressing MOPeGFP and 5-HT1A-Tomato. Loading control: β-actin. Tables: Table S1. Calibration standards for LC-MS/MS; Table S2. LC-MS/MS data for morphine (MOR), normorphine (NMOR), Morphine-3 glucuronide (M3G), Morphine-6-glucuronide (M6G), codeine (COD), norcodeine (NCOD), Codeine-6-glucuronide (C6G), Ethyl-morphine (EMOR) and 6-acetyl-morphine (6MAM). For every analyte two transitions (denoted 1 and 2) are measured, where one determines the concentration (underlined) and the other identity. In the measurements internal standards are also measured (denoted D3).

**Author Contributions:** Conceptualization, V.V. and E.K.; methodology, V.V., L.T., H.D. and E.K.; investigation, V.R., G.J., V.P. and A.N.; data analysis, V.R., G.J., V.P. and A.N.; writing—original draft preparation, V.R., V.V. and E.K.; writing—review and editing, all authors; resources and funding, V.V. and E.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement no 602919. Funding from the Swedish Research Council (VR 2016-01922 and VR 2018-05337), the Foundation for Strategic Research (SBE13-0115), is gratefully acknowledged.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors thank Caroline Grönwall, Peter Sahlström and Camilla Svensson for expert guidance with western blotting experiments and for providing key antibodies.

**Conflicts of Interest:** The authors have no conflicts of interest to declare.

**Sample Availability:** Samples of the compounds are not available from the authors.

#### **References**

