Originally posted 17 March 2016; revised 28 April 2016

www.sciencemag.org/content/352/6285/555/suppl/DC1

Supplementary Materials for

Unconventional endocannabinoid signaling governs sperm activation via the sex hormone progesterone

Melissa R. Miller, Nadja Mannowetz, Anthony T. Iavarone, Rojin Safavi,

Elena O. Gracheva, James F. Smith, Rose Z. Hill, Diana M. Bautista, Yuriy Kirichok, Polina V. Lishko*

*Corresponding author. E-mail: lishko@berkeley.edu

Published 17 March 2016 on Science First Release Published 29 April 2016, Science 352, 555 (2016)

DOI: 10.1126/science.aad6887

This PDF file includes

Materials and Methods Supplementary Text Figs. S1 to S16 Movie legends S1 to S6 Database captions S1 and S2 References

Other Supplementary Material for this manuscript includes the following: (available at www.sciencemag.org/content/352/6285/555/suppl/DC1)

Movies S1 to S6 Databases S1 and S2 as Excel files

Revised 28 April 2016: In this revision, the authors have made changes to the methods, including a reagent name, timing, and units of measure. The cover page was in error, lacking Figures S15 and S16. The originally posted supplementary materials are available here.

1

Materials and Methods

Reagents Progesterone was purchased from CalBiochem (EMD Millipore, Darmstadt,

Germany); 2AG was from Enzo Life Sciences Inc. (Farmingdale, NY); 1AG, methyl arachidonyl fluorophosphanate (MAFP), and prostaglandin E1 were from Cayman Chemical Company (Ann Arbor, MI). Progesterone-3-CMO-(2-biotinyl-PEG2-aminopropylamido-4-trifluoromethyl diazirinyl benzamide), a modified P4 analog (P4*) was synthesized by Cayman Chemical Company. All other compounds were from Avanti Polar Lipids (Alabaster, AL) or Sigma Aldrich (St. Louis, MO) unless otherwise specified. ABHD2 antibodies (labeled in the text “AbA”) were obtained from One World Lab: C14214 (San Diego, CA; affinity-purified anti-ABHD2 polyclonal rabbit IgG). ABHD2 antibodies (labeled in the text as “AbB”) were purchased from Proteintech 14039-1-AP (Rosemont, IL). ACPP antibodies (TmPAP) were obtained from Aves lab (Tigard, OR). Nonspecific rabbit IgG was purchased from Jackson ImmunoResearch lab (111-005-003; West Grove, PA). Monoclonal acetylated alpha-tubulin antibody (clone 6-11B-1) was purchased from Sigma.

Animals and animal tissues

Male C57BL/6 mice and Wistar rats were purchased from Harlan Laboratories (Livermore, CA) and were kept in the Animal Facility of the University of California, Berkeley. All experiments were performed in strict accordance with NIH Guidelines for Animal Research and approved by UC Berkeley Animal Care and Use Committee under the approved protocol MAUP #R352-012. Animals were humanely killed according to ACUC guidelines, and sperm were collected as described previously (30). Bovine and boar ejaculated semen were obtained from UC Davis via collaboration with Dr. Stuart Meyers’ laboratory.

Healthy donors and isolation of human ejaculated spermatozoa

A total of 25 healthy volunteers aged 21-38 were recruited to this study. Freshly ejaculated semen samples were obtained by masturbation. Protocols for the human sperm studies were approved by the Committee on Human Research at the University of California, Berkeley, IRB protocol number 2013-06-5395. Sperm samples were allowed to liquefy for 30-60 min at room temperature before processing. Human spermatozoa were purified by the swim up method (7) in the artificial human tubal fluid solution (HTF, in mM): 98 NaCl, 4.7 KCl, 0.3 KH2PO4, 2 CaCl2, 0.2 MgSO4, 21 HEPES, 3 glucose, 21 lactic acid, and 0.3 sodium pyruvate, pH 7.4 (adjusted with NaOH).

Electrophysiology

All CatSper recordings were performed as described in (7, 18). Briefly, a gigaohm seal between the patch pipette and human spermatozoon was formed at the cytoplasmic droplet. Spermatozoa were patched either at the cytoplasmic droplet or, if the cytoplasmic droplet was inconspicuous, at the neck region. Seals were formed in high saline (HS) solution containing: 130 mM NaCl, 5 mM KCl, 1 mM MgSO4, 2 mM CaCl2, 5 mM glucose, 1 mM sodium pyruvate, 10 mM lactic acid, and 20 mM HEPES; pH 7.4 was adjusted with NaOH, 320 mOsm/liter. Transition into the whole-cell mode was

2

performed by applying short voltage pulses. Access resistance was 25-40 M. Cells were stimulated every 5 s. Data were sampled at 2-5 kHz and filtered at 1 kHz. Pipettes (11-17 MΩ) for whole-cell patch-clamp recordings of monovalent CatSper currents were filled with (in mM): 130 Cs-methanesulfonate, 70 HEPES, 3 EGTA, 2 EDTA, 0.5 TrisHCl, pH 7.4 adjusted with CsOH. Bath divalent free (DVF) solution for recording of monovalent CatSper currents contained (in mM): 140 Cs-methanesulfonate, 40 HEPES, 1 EDTA, pH 7.4 adjusted with CsOH. HEPES was substituted for MES in the event when acidic extracellular pH was used. HS solution was used to record baseline current while measuring monovalent CatSper currents (Ca2+ in HS solution inhibits monovalent CatSper currents and causes Ca2+-dependent inactivation of CatSper channels). Osmolarities of above mentioned solutions were approximately 321 mOsm/liter and 335 mOsm/liter for bath and pipette solutions, respectively. All electrophysiology experiments were performed at ambient temperature. ICatSper were elicited by voltage ramps from a holding potential of 0 mV. Ramps were applied from –80 mV to +80 mV.

Immunocytochemistry

Purified spermatozoa (~107 cells/ml) were plated onto 20-mm coverslips in HS and allowed to attach for 20 min. The cells were fixed with 4% paraformaldehyde (PFA) in 1× PBS for 10 min and washed twice with PBS. Additional fixation was performed with 100% ice-cold methanol for 1 min with two washes in 1× PBS. Cells were permeabilized for 5 min with 0.5 g/ml saponin in PBS followed by two PBS washes and blocked by 1-hour incubation in PBS supplemented with 10% immunoglobulin-γ (IgG)–free BSA. Immunostaining was performed in blocking solution. Detergent-treated cells were incubated with primary antibodies overnight at 4°C. After extensive washing in PBS, secondary antibodies were added for 60 min. After vigorous washing, cells were mounted with ProLong Gold Antifade with DAPI reagent (Life Technologies, Carlsbad, CA). Images were acquired on an Olympus IX71 microscope equipped with differential interference contrast (DIC) and coupled to an XCite fluorescence source. Alexa488-conjugated antirabbit antibodies were obtained from Thermo Scientific.

Electrophoresis and immunoblotting

Protease inhibitors were used throughout the following procedure. Highly motile sperm swim-up fraction was lysed with 2× Laemmli buffer/-mercaptoethanol and boiled at 100 °C for 5 min. An additional 5-min sonication at 100 °C for 5 min was performed. Total crude cell lysate was loaded onto a 4-20% gradient Tris-HCl Criterion SDS-PAGE gel (BioRad). ABHD2, ABHD12 and empty vector-transfected CHO cells were lysed in 2× Laemmli sample buffer, and 104 cells per well were subjected to SDS-PAGE. After transfer to PVDF membranes, blots were blocked in 0.1% PBS-Tween20 with 3% IgG-free BSA for 15 min and incubated in primary antibodies overnight at 4 °C. Blots were probed with one of the following: anti-ABHD2, anti-FLAG, or mouse monoclonal anti-actin C4 antibodies (Abcam, Cambridge, MA). After subsequent washing and incubation with secondary HRP-conjugated antibodies (Abcam, Cambridge, MA), membranes were developed with an ECL SuperSignal West Pico kit (Pierce, Life Technologies, Carlsbad, CA).

3

Glycerol assay Free glycerol quantification was performed as described in (31). Briefly, transfected

cells were washed with phosphate buffered saline (PBS), collected by scraping, resuspended in 50 l PBS and lysed by freeze-fracture and sonication. Total protein was quantified using Pierce 660 reagent, and lysates were divided into aliquots for storage in –80 oC until use. For assay, 0.3 g of protein was added per well to 100 l of assay buffer (50 mM Tris-HCl, 1 mM EDTA, 5 mM MgCl2, 100 mM NaCl, 0.5% BSA plus progesterone [100 M in ethanol] or vehicle) in a 96-well plate. To lysate solution, 100 l of quantification buffer was added (per well: 0.4 U glycerol kinase, 0.4 U glycerol-3-phosphate oxidase, 0.4 U horseradish peroxidase, 0.25 mM ATP, and 20 M Ampliflu Red in assay buffer with substrate or vehicle. For these experiments, AG was used as substrate at a final concentration of 25 M). Fluorescence (emission 532 nm and excitation 580 nm) was monitored every 10 min for a total of 90 min. Background fluorescence was subtracted by including cell- and substrate-free wells. A glycerol titration (63-500 pmol/well) was also included for each assay run to generate standard curves by which free glycerol production could be determined. Presented data are the average of two individual cell transfections assayed in duplicate plus or minus S.E.M.

ABHD cloning and expression

The complete open reading frame of ABHD2 was amplified used gene specific primers (FWD 5’- ggg gac aag ttt gta caa aaa agc agg ctt cAT GAA TGC CAT GCT GGA G– 3’; REV 5’ – ggg gac cac ttt gta caa gaa agc tgg gtc AGA GTC CGG AGG CCT CA - 3’) from cDNA libraries generated from purified human sperm samples. The gateway cloning system (Life Technologies, Carlsbad, CA) was used to insert the sequence-confirmed full-length gene in to mammalian expression vector pEZYflag (Addgene plasmid # 18700), a gift from Yu-Zhu Zhang (32). This expression vector allows for expression of an N-terminal FLAG-tagged protein of interest. For recombinant expression, HEK293 or CHO cells were transfected using Lipofectamine 2000 reagent (Life Technologies, Carlsbad, CA) according to the manufacturer’s recommendation, and overexpression was verified by the presence of FLAG-tagged proteins using M2 anti-FLAG antibody F1804 (Sigma, St. Louis, MO) in Western blot analysis. Human ABHD12 ORF was obtained from PlasmID (Harvard Medical School, MA) and subcloned into the previously mentioned mammalian expression vector, pEZYflag. All of our PCR products obtained from cloning experiments were DNA sequence verified against canonical RefSeq sequences.

PI-PLC treatment and progesterone binding protein pull-down

Sperm cells: After collection, sperm cells were treated with 0.5 units (U) of PI-PLC (P6466; Life Technologies, Carlsbad, CA) and incubated at 37 oC for an hour. Cells were then washed twice with DVF solution and treated for 10 min in the presence of 250 nM P4*, kept at ambient temperature in the dark for 10 min under gentle agitation and UV irradiated (312 nm) for an additional 3 min. Competition with excess estrogen was used to determine whether P4* binding was the result of non-specific steroid interaction, and Ru486 competition removed genomic P4 receptor-like binding. Samples were then washed three times with HS solution and lysed on ice for 30 min. Lysates were added to streptavidin coated magnetic beads (65001; Life Technologies, Carlsbad, CA) on ice for

4

another 30 min with gentle agitation. Bound proteins were eluted by boiling in Laemmli Sample Buffer in the presence of β-mercaptoethanol and subjected to SDS–polyacrylamide gel electrophoresis. The proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane, blocked in PBS with 0.05% Tween 20 (PBS-T) containing 3% IgG-free bovine serum albumin and then incubated with peroxidase (POD)-conjugated streptavidin to detect proteins with P4* bound. For proteomic identification of bands of interest, proteins pulled down by streptavidin and P4* were visualized in SDS–polyacrylamide gels using imperial protein stain (24615, Thermo Scientific). Bands were excised and subjected to in-gel trypsinization after which peptides were extracted and dried for use in proteomic analysis. CHO cells: cells were transfected with ABHD2 plasmid for 24 to 48 hours prior to P4* treatment, and P4* binding proceeded in the same manner as for sperm cells. Briefly, P4* binding to recombinant ABHD2 was performed by applying 250 nM P4* in the absence or presence of unlabeled steroid to live cells for 10 min prior to 3 min UV irradiation. Cells were then washed, lysed, and loaded into a polyacrylamide gel for Western blot analysis.

Protein mass spectrometry

Mass spectrometry was performed by the Vincent J. Coates Proteomics/Mass Spectrometry Laboratory at UC Berkeley. A nano-liquid-chromatography (LC) column was packed in a 100 μm inner diameter glass capillary with an emitter tip. The column consisted of 10 cm of Polaris c18 5-μm packing material (Varian). The column was loaded by use of a pressure bomb and washed extensively with buffer A (see below). The column was then directly coupled to an electrospray ionization source mounted on a Thermo-Fisher LTQ XL linear ion trap mass spectrometer. An Agilent 1200 high-performance LC (HPLC) column equipped with a split line to deliver a flow rate of 300 nl/min was used for chromatography. Peptides were eluted with a linear gradient from 100% buffer A to 60% buffer B in 2.5 hours. Buffer A was 5% acetonitrile and 0.02% heptaflurobutyric acid (HBFA); buffer B was 80% acetonitrile and 0.02% HBFA. Protein identification was done with Integrated Proteomics Pipeline software (IP2, Integrated Proteomics Applications, Inc., San Diego, CA) by using ProLuCID/Sequest and DTASelect2. Tandem mass spectra were extracted into ms1 and ms2 files from raw files using RawExtract 1.9.9 and were searched against the human protein database plus sequences of common contaminants, concatenated to a decoy database in which the sequence for each entry in the original database was reversed. LTQ data was searched with 3000.0 milli-atomic mass unit precursor tolerance, and the fragment ions were restricted to a 600.0 ppm tolerance. All searches were parallelized and searched on the VJC proteomics cluster. Search space included all tryptic peptide candidates with no missed cleavage restrictions. Carbamidomethylation (+57.02146) of cysteine was considered a static modification and oxidation of methionine (+15.9949) was a variable modification. We required 1 peptide per protein and two tryptic termini for each peptide identification. The ProLuCID search results were assembled and filtered using the DTASelect program (version 2.0) with a general peptide FDR of 0.005 and a peptide FDR of 0.001 for inclusion at the protein level. Under such filtering conditions, the global peptide false discovery rate was 0.37 or less for all samples.

Lipid extraction

5

In order to selectively extract lipids from the plasma membrane, we developed a novel cyclodextrin-based lipid extraction protocol that allows removal of steroids and single-tailed lipids, such as AGs, and leaves multi-tailed lipids in the membrane. Once extracted, the lipids are separated from the cyclodextrin via a hexane:isopropanol technique, and purified extracellular lipids are analyzed using an LTQ-Orbitrap-XL mass spectrometer equipped with an ESI source and operated in the negative ion mode. Briefly, between 33 and 37 million sperm cells from five different human donors were assessed. Sperm cells were washed with DVF and treated with 500 nM P4* or vehicle for 10 min and then 5 mM β-cyclodextrin C4805 (Sigma, St. Louis, MO, USA) for another 10 min. Cells were pelleted down and supernatant added to streptavidin magnetic beads (Dynabeads MyOne Streptavidin C1, #65001; Life Technologies, Carlsbad, CA) to remove P4* from extract. Flow through was treated for 10 min at 55 oC with maltogenic amylase (A2986, Sigma, St. Louis, MO) to hydrolyze β-cyclodextrin. Lipids were extracted from live cells by using hexane:isopropanol method(33), dried under gentle nitrogen flow and samples reconstituted in 7:1 methanol/water prior to analysis. A cell free control was also run to account for lipid contamination present within any of the treatment, pull-down or extraction steps. This condition was used to determine the baseline for our lipid detection (below 100 fmol).

Liquid chromatography-mass spectrometry

Methanol (Fisher Optima grade, 99.9%), ammonium formate (Alfa Aesar, 99%), and water purified to a resistivity of 18.2 MΩ·cm (at 25 °C) using a Milli-Q Gradient ultrapure water purification system (Millipore, Billerica, MA), were used to prepare mobile phase solvents. Lipid extracts were analyzed using an Agilent 1200 liquid chromatograph (LC; Santa Clara, CA) that was connected in-line with an LTQ-Orbitrap-XL mass spectrometer equipped with an electrospray ionization source (ESI; Thermo Fisher Scientific, Waltham, MA). The LC was equipped with a C4 analytical column (Viva C4, 150 mm length × 1.0 mm inner diameter, 5-µm particles, 300 Å pores, Restek, Bellefonte, PA). Solvent A was water with 20 mM ammonium formate and solvent B was methanol. The sample injection volume was 50 µl. The elution program consisted of isocratic flow at 1% B for 3 min, a linear gradient to 99% B over 32 min, isocratic flow at 99% B for 5 min, and isocratic flow at 1% B for 19.9 min, at a flow rate of 150 µl/min. The column and sample compartments were maintained at 40 ºC and 4 ºC, respectively. The injection needle was rinsed with a 1:1 methanol:water (v/v) solution after each injection to avoid cross-contamination between samples. The column exit was connected to the ESI probe of the mass spectrometer using PEEK tubing (0.005” inner diameter × 1/16” outer diameter, Agilent). Mass spectra were acquired in the negative ion mode over the range m/z = 100 to 1000 using the Orbitrap mass analyzer, in profile format, with a mass resolution setting of 100,000 (at m/z = 400, measured at full width at half-maximum peak height. Mass spectra were processed and analyzed using Xcalibur software (version 2.0.7, Thermo) and LIPID MAPS Online Tools. To confirm the identity of the 2AG ion, the synthetic compound was analyzed under identical conditions as the isolated lipid mixtures. We determined from the lipid standard that 2AG occurs as a formate adduction, [M+45 daltons]-, in negative ion ESI, with a chromatographic retention time in the range of 30-40 min (Figure S5).

6

Mouse sperm capacitation and acrosome exocytosis assessment in live sperm Acrosome status was assessed using transgenic mice B6D2-Tg(CAG/Su9-

DsRed2,Acr3-EGFP)RBGS002Osb (kindly provided by Dr. Masaru Okabe, Osaka University, Japan). These mice express EGFP that is targeted to acrosome, and is lost at the time of acrosomal exocytosis. Spermatozoa from the cauda epididymides were capacitated by isolating and perforating the cauda epididymis and suspending it in 500 μl of HTF media (Irvine scientific) for 30 min at 37 °C under 5% CO2. After this treatment the cauda was removed and free swimming sperm were allowed to incubate for another 30 min in the same culture conditions. To assess induced acrosome exocytosis, capacitated sperm cells were allowed to adhere to poly-D-lysine–coated coverslips and then incubated for 15 min with drug or vehicle, after which time coverslips were imaged for the presence of fluorescently labeled acrosome. For MAFP effect, cells were pre-incubated for 10 min with 2 μM MAFP or vehicle. At least 100 cells were scored for each condition by two independent researchers. Data are presented as the mean of three individual experiments ± S.E.M.

RNA deep sequencing analysis of the human sperm and human monocytes,

Total RNA was extracted from human spermatozoa purified from 32 human semen samples of seven different donors. Sperm purification was carried out by gradient centrifugation using ISolateTM (Irvine Scientific, CA) density gradient medium. Two ISolate concentrations were made: 80 and 40%, both were prepared with HS solution. Spermatozoa were initially purified from semen samples by the spin-down method described in (34) and then overlayed on top gradient (40% ISolate) and centrifuged for another 30 min at 300g. Pure spermatozoa were collected at the bottom and visually examined for purity and contamination under phase-contrast microscope. Only 100% pure sperm samples were used for RNA extraction. Because monocytes are often found in human ejaculates, comparison of both transcriptomes should be used to determine sperm-specific transcripts. Peripheral blood mononuclear cells (PBMCs) were purchased from AllCells (Emeryville, CA), product number PB004F. PBMCs were from the age-matched male donor. Total RNA was extracted from cells using Qiagen RNeasy kit, and sequencing libraries were prepared poly(A)+-enriched RNA using Illumina mRNA-Seq Sample Prep Kit according to the manufacturer’s instructions. Libraries were sequenced on the Illumina Genome Analyzer II using a 36-cycle sequencing kit v3 by standard protocols. Single-read FC flow cell was used for sequencing. Sequences were aligned to human genome GRCh37. Bioinformatics support was provided by Centrillion Biosciences Inc. (Palo Alto, CA).

Sperm motility analysis

Viable human spermatozoa were selected by swim-up, and a subset of cells was allowed to capacitate at 37°C in normal capacitation media for 3.5 h as reported in (7). Cells were then divided into aliquots and pre-incubated for 15 min in the presence of vehicle or MAFP (2 M) prior to exposure to progesterone (3 M). Human CASA was done with 3 M of P4, which is meant to mimic physiological concentrations of progesterone, which can be found up to 10 μM. Sperm motility was analyzed by computer-assisted semen analysis (CASA, HTM-IVOS sperm analysis system, version 12.3, Hamilton Thorne Biosciences, Beverly, MA) system that measured average path

7

velocity (VAP, m/s), straight-line velocity (VSL, m/s,) and curvilinear velocity (VCL, m/s). From these measurements, we can determine linearity of progression [LIN = (VSL/VCL) × 100], and straightness [STR = ( VSL/VAP) × 100]. We chose to focus on VCL as no significant progesterone induced effect was observed for LIN or STR consistent with Calogero et al. (35) (Fig S1C). Data were normalized to vehicle matched controls and presented as the average of (n) individual experiments ± S.E.M. For all experimental conditions, a minimum of 10 fields of view was analyzed, and all experiments were performed at 37°C.

Calcium imaging

Human spermatozoa purified by swim-up method as described above were loaded with 10 g/ml of Fluo-4, AM (ThermoFischer) in HS solution and 0.05% Pluronic in DMSO for 30 min at room temperature. Cells were then allowed to adhere to poly-D-lysine-coated imaging chambers for 10 min and then washed with imaging medium (HS supplemented with 15 mM NaHCO3). Washed sperm were left in 200 l of imaging buffer for 20 min prior to fluorescence recording. Imaging was performed using Olympus IX-71 microscope equipped with Lambda XL (Sutter, CA), and fluorescence was recorded at 2 Hz over a 25-s time frame; fluorescence change over time was determined as ΔF/F0 were ΔF is the change in fluorescence intensity (F - F0) and F0 is the baseline as calculated by averaging the fluorescence signal 5 s prior to stimulus application. Calcium imaging results are presented as the average ± the standard deviation of >50 individual sperm flagella, and the regions of interest (ROI) were selected manually using MetaFluor Imaging software (Molecular Devices).

Prostatic acid phosphatase (TmPAP or ACPP) cloning and functional assay

The complete open reading frame of the transmembrane containing isoform of ACCP, termed TmPAP, was amplified with gene-specific primers (FWD 5’- aca aca ggt acc ATG AGA GCT GCA CCC CTC - 3’; REV 5’- aca aca gcg gcc gcT CAG ATG TTC CCA TAG GAT TCT C - 3’). A cDNA library generated from purified ejaculated human sperm cells was used as a template, and we inserted the resulting PCR product into the mammalian expression vector, pTRACER (Invitrogen Inc., CA) using conventional cloning methods. After sequence confirmation, the TmPAP-containing plasmid was used for transfection of HEK293 cells and phosphatase activity assessed using p-Nitrophenyl phosphate as a substrate in the presence or absence of 0.5 μM progesterone, according to the manufacturer’s instructions (NEB Ipswich, MA; P0757L). Phosphatase activity was determined by recording the absorbance of the assay buffer at 410 nm.

Data analysis

Data were analyzed with Origin 7.0 and Clampfit 9.2. Statistical data were calculated and represented everywhere in text as the mean S.E.M., and (n) indicates number of experiments. To determine ICatSper inactivation rate current densities extracted from voltage ramps and acquired at +80 pA were plotted against time. The data were fit with Boltzmann equation y = 1/(1 + exp[(x - )/dx)], and mean numbers were determined from (n) independent experiments. Hill equation was used to build dose-response curve: y = Vmax*xn/(kn + xn) , where n defines Hill slope factor, k defines IC50,

8

and Vmax is close to 100% inhibition. Statistical significance (t test) is indicated by (****P < 0.0001, ***P < 0.001, * P < 0.005, and *P < 0.05).

Supplementary Text

Extended list of acknowledgements: The authors thank Steven Mansell and Divya Patel for the help with mouse sperm,

Dr. Dan Feldman (UC Berkeley) for sharing rat tissue, and Lisa Jalalian (UCSF) for the help with human CASA. We thank Dr. Stuart Meyers (UC, Davis) for sharing bovine and boar specimens.

9

Fig. S1. CatSper activation by P4 requires mSH activity; however, PGE1 effect is mSH-independent.

(A) ICatSper amplitudes recorded at +80 mV (triangles) and –80 mV (circles) plotted against time and the duration of treatment indicated above plot. (B) Representative ICatSper recorded from control human spermatozoa in the absence (black) and in the presence of methyl arachidonyl fluorophosphanate (MAFP, green). (C) Table comparing sperm motility parameters: linear percent (LIN) and straight percent (STR). Data are the average ± S.E.M. of 4 independent experiments. (D) Representative ICatSper recorded from control human spermatozoa in the absence (black) and presence of 1 M prostaglandin E1 (red; left panel). ICatSper recorded from human spermatozoa pre-treated with 2 M of MAFP for 30 min (MAFP, right panel). Initial ICatSper (black) and ICatSper after addition of 1 M PGE1 with 2 M MAFP (red) is shown. In both cases ICatSper was potentiated.

10

Fig. S2. Pharmacological profile of human CatSper.

(A) ICatSper densities recorded at +80 mV (positive scale) or –80 mV (negative scale) extracted from ramps shown in (B). Recordings were performed in the absence or presence of indicated compounds. Dotted red lines indicate control ICatSper densities. Mean data were obtained from (n) independent experiments, statistical significance is indicated by **P < 0.005. (B) Representative ICatSper sensitivity to indicated compounds and treatments. Compounds: lysophosphatidylinositol (lyso-PI), lysophosphatidylethanolamine (lyso-PE), anandamide (AEA), oleamide (OEA), progesterone (P4). Treatment: PF750 selective FAAH inhibitor.

11

Fig. S3. Inhibition kinetics of CatSper by endocannabinoids.

(A) Percent inhibition of ICatSper by 1AG and 2AG. Percent inhibition of outward ICatSper was 92.4 ± 3.5 and 85.0 ± 0.5 for 2AG and 1AG, respectively; cells from at least three different human donors were used. (B) Representative ICatSper amplitudes recorded at +80 mV (triangles) and –80 mV (circles) as shown in Figure 1E and plotted against time. (C) Representative rates of ICatSper inactivation curves upon application of endocannabinoids. Normalized ICatSper densities recorded at +80 mV were plotted against time and fitted with Boltzmann equation as described in methods. (D) ICatSper inactivation rates upon addition of either endocannabinoid or removal of P4 (upon CatSper washout). (E) Representative recordings of control ICatSper before and after addition of various 2AG concentrations. (F) Dose-response of ICatSper inactivation by 2AG. Both inward (at –80 mV) and outward (at +80 mV) inactivation curves are shown. ICatSper amplitudes were extracted from ramps shown in (E) at corresponding voltages and normalized to controls. Data were averaged and fitted with Hill equation as described in methods. The steepness of curves and Hill slope factor (k > [1]) is common for lipid compounds and reflect cooperative effect of 2AG molecules toward the channel via their slow incorporation into a lipid bilayer. Between 4 and 15 cells were used to build the dose-response.

12

Fig. S4. Monoacylglycerols are endogenous cannabinoids and can be detected in human sperm.

(A) 2AG does not affect ICatSper or its potentiation by P4 if applied intracellularly. Representative recording is shown. (B) Averaged ICatSper densities recorded at –80 mV extracted from ramps shown in (A) and (C). Data are represented as means ± SEM of (n) independent experiments. (C) Extracellular glycerol does not affect ICatSper or its potentiation by P4 indicating that the active part of 2AG is its lipid tail. Representative recording is shown. (D) 1AG/2AG content of sperm plasma membrane as measured by LC-MS. Data are representative from one of two independent mass-spectrometry runs. AGs concentration varied between 98 and 240 fmol/sample (of roughly 33-37 million purified sperm cells, from five different human donors).

13

Fig. S5. Mass spectrometry analysis of the sperm plasma membrane outer leaflet.

(A) Extracted ion chromatogram (EIC) between 20 and 50 min of full mass spectra acquisition for lipid extracts showing the relative abundance of m/z = 423.2754±0.006. Lipids were extracted by β-cyclodextrin treatment of live sperm cells in the absence (middle) or presence (lower) of P4*. The upper panel shows the EIC for β-cyclodextrin alone (blank). The peaks between times of 27 and 43 min were integrated to obtain area under the curve (AUC) (B) Shown is the EIC for 50 nmol of synthetic 2AG showing a retention time of 35.3 min. (C) Electrospray Ionization mass spectra reveals the formation of a formate adduct as the major ionization product of 2AG when acquired in the negative ion mode. This m/z was used to assess 2AG presence in experiments mentioned above. Note that in these experiments we are unable to differentiate between the 1AG and 2AG stereoisomers based on either m/z or retention time.

14

Fig. S6. Sperm P4 receptor is not GPI- anchored.

(A) Representative ICatSper and their P4 sensitivity recorded from control human spermatozoa (left panel) and human spermatozoa treated with 0.5 U of PI-PLC (right panel). (B) Averaged ICatSper densities recorded at +80 mV (positive scale) or –80 mV (negative scale) extracted from ramps shown in (A) and (C). Data are represented as means ± SEM of (n) independent experiments. (C) Representative ICatSper recorded from spermatozoa in the absence (black) and presence of either P4 (red) or P4* (green).

15

Fig. S7. UV-activated chemical reaction involving diazirinine moiety.

16

top candidates # of different peptides % coverage sequence #

ABHD2 1 4.0% K.SPTAPPDLYFQDSGLSR.F 3

FAAH 3 5.9% R.RTARGAVVR.A

R.ELAPEAVLFTYVGK.A R.LDPTVPPLPFR.E

1 2 2

TmPAP 1 3.0% R.SPIDTFPTDPIK.E 1

Fig. S8. Proteomic analysis of ~50 kDa band.

Top candidates revealed by selection criteria were: fatty-acid amide hydrolase 1 (FAAH), hydrolase domain-containing protein 2 (ABHD2), and prostatic acid phosphatase which has a transmembrane domain containing splice isoform (TmPAP).

17

Fig. S9. TmPAP is not involved in progesterone regulation of CatSper.

(A) Cloning of ACPP gene using primers specific towards the transmembrane domain containing isoform. cDNA library was generated from purified human ejaculated sperm cells. (B) Representative western blot of HEK293 cell lysates from control cells (NT) and cells transfected with TmPAP-containing plasmid. Westerns were probed with an anti-TmPAP (ACPP) antibody. Actin was used as a loading control. (C) anti-TmPAP antibody does not affect nascent ICatSper or its potentiation by P4. Representative ICatSper is shown. Human spermatozoa were treated with 1:100 dilutions of the antibodies for at least 30 min. (D) TmPap activity assay shows that recombinantly expressed TmPap is an active phosphatase; however, its activity is not regulated by progesterone addition.

18

Fig. S10. ABHD2 expression in heterologous system (CHO cells) and human sperm cells.

(A) ABHD2 was heterologously expressed within CHO cells. Heterologous expression was confirmed by two different specific anti-ABHD2 antibodies: AbA (1:300) and AbB (1:200) as indicated. (B) Immunostaining of human spermatozoa with two different anti-ABHD2 antibodies as indicated AbA (1:100) AbB (1:400).

19

Fig. S11. ABHD2 is expressed in human spermatozoa.

(A) Amplification of abhd2 from human sperm cDNA library reveals a band corresponding to the full-length ORF. (B) The full western blot image of ABHD2 staining of sperm lysate probed with different anti-ABHD2 antibody “AbB”; used in 1:10k dilution. (C) ABHD2 was cloned from purified human spermatozoa and heterologously expressed in HEK cells. Heterologous expression was confirmed by anti-ABHD2 antibodies AbB. (D) Nonspecific rabbit IgG (left), nor anti-ABHD2 (AbA) antibody alone (right) had no effect on nascent ICatSper. Human spermatozoa were treated with 1:100 dilutions of either antibodies for at least 30 min. (E) ABHD2 and ABHD12 were heterologously expressed with Flag-tag in CHO cells. Heterologous expression was confirmed by either Flag or anti-ABHD2 antibodies AbA.

20

Fig. S12. Acrosome intact and acrosome reacted murine spermatozoa in response to indicated treatments.

Live capacitated mouse spermatozoa are shown. Top row, fluorescently labeled mouse spermatozoa (these mice express EGFP that is targeted to acrosome, and is lost at the time of acrosomal exocytosis); middle row, DIC images; and bottom row, overlaid images. Scale bar, 20 m. Acrosome intact spermatozoa show bright green acrosomal staining (white arrowhead), whereas acrosome-reacted sperm cells are dark (yellow arrowhead).

21

Fig. S13. Endocannabinoids are also inhibitors of murine and human CatSpers.

(A) Representative ICatSper recorded from mouse spermatozoon in the absence (black) and in the presence of 2AG (green). (B) Current densities of mouse ICatSper recorded at +80 mV (positive scale) or –80 mV (negative scale) in the absence or presence of 2AG. Percent inhibition of mouse CatSper is 73.9 ± 4.4, measured at +80 mV.

22

Fig. S14. Inactivation of CatSper channel upon progesterone removal is likely due to AGs replenishment.

(A) Representative ICatSper amplitudes when plotted over time show slow ICatSper inactivation upon removal of P4. (B) The rate of ICatSper inactivation (washout) after removal of P4 was calculated as mean ± SEM extracted from ICatSper densities obtained at +80 mV. Mean rates were obtained from three independent experiments. This rate matches the rate of 2AG inhibition of CatSper (fig. S3D).

23

Fig. S15. Model of non-genomic P4 signaling in human sperm cells. P4 binds ABHD2 and activates the hydrolysis of 2AG (indicated as triangle with a tail in the figure). Basal levels of 2AG inhibit calcium channel CatSper. Once 2AG is metabolized into glycerol and arachidonic acid (AA), CatSper inhibition is eliminated and Ca2+ enters the cytoplasm. Elevation of [Ca2+] in the flagellum leads to sperm hyperactivation. The latter helps sperm overcome the egg’s protective vestments and leads to fertilization. AA, a product of ABHD2 hydrolysis, may act as a negative regulator of ABHD2.

24

Fig. S16. Arachidonic acid acts as a dual regulator of CatSper.

(A) ICatSper amplitudes recorded at +80 mV (triangles) and –80 mV (circles) were plotted against time to show time-dependent AA effect towards CatSper and removal of P4 activation. (B) ICatSper densities recorded at +80 mV (positive scale) or –80 mV (negative scale) extracted from three independent experiments.

25

Movie S1

Elevation of intracellular [Ca2+] in human sperm in response to 1 M P4.

Movie S2

Elevation of intracellular [Ca2+] in human sperm in response to 1 M P4 (zoomed area).

Movie S3

Flagellar intracellular [Ca2+] fails to rise in response to 1 M P4 when sperm were pretreated with 2 M MAFP.

Movie S4

Flagellar intracellular [Ca2+] fails to rise in response to 1 M P4 when sperm were pretreated with 2 M MAFP (zoomed area).

Movie S5

Flagellar intracellular [Ca2+] fails to rise in response to P4 when sperm were preincubated with anti-ABHD2 antibody. Experiment was done in the presence of 1:100 anti-ABHD2 AbA.

Movie S6

Flagellar intracellular [Ca2+] fails to rise in response to P4 when sperm were preincubated with anti-ABHD2 antibody. Experiment was done in the presence of 1:100 anti-ABHD2 AbA (zoomed area).

Additional Data table S1 (separate file, Database 1: Peptide Mass Spectrometry database)

Additional Data table S2 (separate file, Database 1: Transcriptomes)

Comparative gene expression profile (mRNA-Seq reads) of peripheral blood mononuclear cells (PBMC) vs human spermatozoa. Signals <1 reads are within statistical noise and therefore scored as non-expressed sequences.

26

References and Notes

1. R. M. Evans, The steroid and thyroid hormone receptor superfamily. Science 240, 889–895 (1988). Medline doi:10.1126/science.3283939

2. R. Lösel, M. Wehling, Nongenomic actions of steroid hormones. Nat. Rev. Mol. Cell Biol. 4, 46–55 (2003). Medline doi:10.1038/nrm1009

3. A. Revelli, M. Massobrio, J. Tesarik, Nongenomic actions of steroid hormones in reproductive tissues. Endocr. Rev. 19, 3–17 (1998). Medline

4. P. F. Blackmore, S. J. Beebe, D. R. Danforth, N. Alexander, Progesterone and 17 alpha-hydroxyprogesterone. Novel stimulators of calcium influx in human sperm. J. Biol. Chem. 265, 1376–1380 (1990). Medline

5. E. Baldi, R. Casano, C. Falsetti, C. Krausz, M. Maggi, G. Forti, Intracellular calcium accumulation and responsiveness to progesterone in capacitating human spermatozoa. J. Androl. 12, 323–330 (1991). Medline

6. P. Poisbeau, A. F. Keller, M. Aouad, N. Kamoun, G. Groyer, M. Schumacher, Analgesic strategies aimed at stimulating the endogenous production of allopregnanolone. Front. Cell. Neurosci. 8, 174 (2014). Medline doi:10.3389/fncel.2014.00174

7. P. V. Lishko, I. L. Botchkina, Y. Kirichok, Progesterone activates the principal Ca2+ channel of human sperm. Nature 471, 387–391 (2011). Medline doi:10.1038/nature09767

8. J. F. Smith, O. Syritsyna, M. Fellous, C. Serres, N. Mannowetz, Y. Kirichok, P. V. Lishko, Disruption of the principal, progesterone-activated sperm Ca2+ channel in a CatSper2-deficient infertile patient. Proc. Natl. Acad. Sci. U.S.A. 110, 6823–6828 (2013). Medline doi:10.1073/pnas.1216588110

9. T. Strünker, N. Goodwin, C. Brenker, N. D. Kashikar, I. Weyand, R. Seifert, U. B. Kaupp, The CatSper channel mediates progesterone-induced Ca2+ influx in human sperm. Nature 471, 382–386 (2011). Medline doi:10.1038/nature09769

10. T. A. Quill, D. Ren, D. E. Clapham, D. L. Garbers, A voltage-gated ion channel expressed specifically in spermatozoa. Proc. Natl. Acad. Sci. U.S.A. 98, 12527–12531 (2001). Medline doi:10.1073/pnas.221454998

11. D. Ren, B. Navarro, G. Perez, A. C. Jackson, S. Hsu, Q. Shi, J. L. Tilly, D. E. Clapham, A sperm ion channel required for sperm motility and male fertility. Nature 413, 603–609 (2001). Medline doi:10.1038/35098027

12. A. E. Carlson, R. E. Westenbroek, T. Quill, D. Ren, D. E. Clapham, B. Hille, D. L. Garbers, D. F. Babcock, CatSper1 required for evoked Ca2+ entry and control of flagellar function in sperm. Proc. Natl. Acad. Sci. U.S.A. 100, 14864–14868 (2003). Medline doi:10.1073/pnas.2536658100

13. Y. Kirichok, B. Navarro, D. E. Clapham, Whole-cell patch-clamp measurements of spermatozoa reveal an alkaline-activated Ca2+ channel. Nature 439, 737–740 (2006). Medline doi:10.1038/nature04417

27

14. P. V. Lishko, Y. Kirichok, D. Ren, B. Navarro, J. J. Chung, D. E. Clapham, The control of male fertility by spermatozoan ion channels. Annu. Rev. Physiol. 74, 453–475 (2012). Medline doi:10.1146/annurev-physiol-020911-153258

15. R. A. Osman, M. L. Andria, A. D. Jones, S. Meizel, Steroid induced exocytosis: The human sperm acrosome reaction. Biochem. Biophys. Res. Commun. 160, 828–833 (1989). Medline doi:10.1016/0006-291X(89)92508-4

16. J. J. Chung, S. H. Shim, R. A. Everley, S. P. Gygi, X. Zhuang, D. E. Clapham, Structurally distinct Ca2+ signaling domains of sperm flagella orchestrate tyrosine phosphorylation and motility. Cell 157, 808–822 (2014). Medline doi:10.1016/j.cell.2014.02.056

17. H. Qi, M. M. Moran, B. Navarro, J. A. Chong, G. Krapivinsky, L. Krapivinsky, Y. Kirichok, I. S. Ramsey, T. A. Quill, D. E. Clapham, All four CatSper ion channel proteins are required for male fertility and sperm cell hyperactivated motility. Proc. Natl. Acad. Sci. U.S.A. 104, 1219–1223 (2007). Medline doi:10.1073/pnas.0610286104

18. P. Lishko, D. E. Clapham, B. Navarro, Y. Kirichok, Sperm patch-clamp. Methods Enzymol. 525, 59–83 (2013). Medline doi:10.1016/B978-0-12-397944-5.00004-3

19. J. Z. Long, B. F. Cravatt, The metabolic serine hydrolases and their functions in mammalian physiology and disease. Chem. Rev. 111, 6022–6063 (2011). Medline doi:10.1021/cr200075y

20. J. L. Blankman, G. M. Simon, B. F. Cravatt, A comprehensive profile of brain enzymes that hydrolyze the endocannabinoid 2-arachidonoylglycerol. Chem. Biol. 14, 1347–1356 (2007). Medline doi:10.1016/j.chembiol.2007.11.006

21. G. Cobellis, G. Ricci, G. Cacciola, P. Orlando, S. Petrosino, M. G. Cascio, T. Bisogno, L. De Petrocellis, T. Chioccarelli, L. Altucci, S. Fasano, R. Meccariello, R. Pierantoni, C. Ledent, V. Di Marzo, A gradient of 2-arachidonoylglycerol regulates mouse epididymal sperm cell start-up. Biol. Reprod. 82, 451–458 (2010). Medline doi:10.1095/biolreprod.109.079210

22. G. Catanzaro, N. Battista, G. Rossi, M. Di Tommaso, M. Pucci, V. Pirazzi, S. Cecconi, M. Maccarrone, Effect of capacitation on the endocannabinoid system of mouse sperm. Mol. Cell. Endocrinol. 343, 88–92 (2011). Medline doi:10.1016/j.mce.2011.01.022

23. H. Rejraji, B. Sion, G. Prensier, M. Carreras, C. Motta, J. M. Frenoux, E. Vericel, G. Grizard, P. Vernet, J. R. Drevet, Lipid remodeling of murine epididymosomes and spermatozoa during epididymal maturation. Biol. Reprod. 74, 1104–1113 (2006). Medline doi:10.1095/biolreprod.105.049304

24. A. Lenzi, L. Gandini, V. Maresca, R. Rago, P. Sgrò, F. Dondero, M. Picardo, Fatty acid composition of spermatozoa and immature germ cells. Mol. Hum. Reprod. 6, 226–231 (2000). Medline doi:10.1093/molehr/6.3.226

25. V. Di Marzo, M. Bifulco, L. De Petrocellis, The endocannabinoid system and its therapeutic exploitation. Nat. Rev. Drug Discov. 3, 771–784 (2004). Medline doi:10.1038/nrd1495

26. P. Grimaldi, M. Pucci, S. Di Siena, D. Di Giacomo, V. Pirazzi, R. Geremia, M. Maccarrone, The faah gene is the first direct target of estrogen in the testis: Role of histone

28

demethylase LSD1. Cell. Mol. Life Sci. 69, 4177–4190 (2012). Medline doi:10.1007/s00018-012-1074-6

27. M. Maccarrone, M. Bari, M. Di Rienzo, A. Finazzi-Agrò, A. Rossi, Progesterone activates fatty acid amide hydrolase (FAAH) promoter in human T lymphocytes through the transcription factor Ikaros. Evidence for a synergistic effect of leptin. J. Biol. Chem. 278, 32726–32732 (2003). Medline doi:10.1074/jbc.M302123200

28. S. P. Williams, P. B. Sigler, Atomic structure of progesterone complexed with its receptor. Nature 393, 392–396 (1998). Medline doi:10.1038/30775

29. K. Miyata, Y. Oike, T. Hoshii, H. Maekawa, H. Ogawa, T. Suda, K. Araki, K. Yamamura, Increase of smooth muscle cell migration and of intimal hyperplasia in mice lacking the alpha/beta hydrolase domain containing 2 gene. Biochem. Biophys. Res. Commun. 329, 296–304 (2005). Medline doi:10.1016/j.bbrc.2005.01.127

30. G. Wennemuth, D. F. Babcock, B. Hille, Calcium clearance mechanisms of mouse sperm. J. Gen. Physiol. 122, 115–128 (2003). Medline doi:10.1085/jgp.200308839

31. D. Navia-Paldanius, J. R. Savinainen, J. T. Laitinen, Biochemical and pharmacological characterization of human α/β-hydrolase domain containing 6 (ABHD6) and 12 (ABHD12). J. Lipid Res. 53, 2413–2424 (2012). Medline doi:10.1194/jlr.M030411

32. F. Guo, M. Y. Chiang, Y. Wang, Y. Z. Zhang, An in vitro recombination method to convert restriction- and ligation-independent expression vectors. Biotechnol. J. 3, 370–377 (2008). Medline doi:10.1002/biot.200700170

33. A. Hara, N. S. Radin, Lipid extraction of tissues with a low-toxicity solvent. Anal. Biochem. 90, 420–426 (1978). Medline doi:10.1016/0003-2697(78)90046-5

34. P. V. Lishko, I. L. Botchkina, A. Fedorenko, Y. Kirichok, Acid extrusion from human spermatozoa is mediated by flagellar voltage-gated proton channel. Cell 140, 327–337 (2010). Medline doi:10.1016/j.cell.2009.12.053