Mice

Male C57BL/6 J and BALB/c wild-type mice, aged 6–8 weeks, were purchased from Charles River Laboratories (Saint-Germain-sur-l’Arbresle, France). Mice were housed in the facilities of the iMM João Lobo Antunes (iMM JLA), in specific pathogen-free environment and given water and food ad libitum. Mice are kept in 22–24 °C and 45–65% humidity in a 14h-light/10h-dark cycle. All in vivo protocols were approved by the ORBEA committee of the iMM JLA and were performed according to national and European regulations.

Sporozoite production

The following parasite lines were used in this study: GFP-expressing P. berghei ANKA (clone 259cl2); P. berghei NK65 EXP2 FRT (WT), P. berghei NK65 EXP2 FRT UIS4 FLP Recombinase (EXP2 cKO), P. berghei NK65 EXP2 FRT TRAP FLP Recombinase and P. berghei NK65 EXP2-HA. The first was a kind gift from Chris Janse from Leiden University Medical Center from Leiden, The Netherlands, whereas the other lines were generated in the laboratory of Tania de Koning-Ward from Deakin University, Australia, and are described in a previous study that also outlines the protocol used to achieve excision of the EXP2 3′ UTR²⁸. Parasites were stored in frozen blood vials, containing 10⁷ blood stage parasites, and kept at −80 °C in our laboratory. To achieve sporozoites, 10⁷ infected red blood cells were injected intra-peritoneally into a BALB/c wild-type mouse. After 5 days of infection, exflagellation of the male gametes in the blood of infected mice was observed under a light microscope. If more than five events per field of view were observed, the infected mouse was used to feed naive Anopheles stephensi mosquitos, bred in the insectary of the iMM JLA, for 30 min. For the P. berghei NK65 EXP2 FRT TRAP FLP Recombinase parasite line, infected mosquitos were placed at 25 °C 18 days after the mosquito bite, to enhance the activity of the recombinase²⁹. In total, 22–35 days after the mosquito blood meal, salivary glands containing P. berghei sporozoites, were dissected from infected female Anopheles stephensi mosquitoes into simple Dulbecco’s Modified Eagles Medium (DMEM, Gibco, Thermo Fisher Scientific, Waltham, MA, USA) and collected into an Eppendorf tube (Eppendorf, Hamburg, Germany). Salivary glands were smashed with a plastic pestle and filtered through a 40-µm Falcon cell-strainer (Thermo Fisher Scientific) to release sporozoites. Sporozoites were counted using a hemocytometer (Marienfeld Superior, Lauda-Königshofen, Germany).

Hepatoma cells

HepG2 cells (obtained from ATCC, Manassas, VA, USA) were cultured at 37 °C, with a 5% CO₂ atmosphere, in DMEM (Gibco), supplemented with 10% FCS (Gibco), 1% glutamine (Gibco) and 1% penicillin/streptomycin (Gibco). Cells were seeded onto no. 1 grade 12 mm diameter glass coverslips (VWR, Radnor, PA, USA) in 24-well plates (Thermo Fisher Scientific) or onto black glass-bottom 96-well plates (Greiner, Kremsmünster, Austria). After plating, cells were incubated at 37 °C, with a 5% CO₂ atmosphere.

Recombinant EXP2 protein production and purification

Plasmid for recombinant EXP2 protein (rEXP2) production was a kind gift from Professor Masafumi Yohda from Tokyo University of Agriculture and Technology from Tokyo, Japan³⁹. In brief, the plasmid pGEX-3X-HRV 3C-EXP2 containing a truncated version of the EXP2 protein fused to Glutathione S-transferase (GST) and an HRV 3C protease cleavage site to excise the affinity tag, was transformed into E. coli B21, a gift from the laboratory of Gonçalo Bernardes at iMM JLA. Transformed bacteria were grown in LB broth supplemented with 100 µg/mL ampicillin at 37 °C and when the optical density (OD₆₀₀) of the culture reached 0.5, the protein expression was induced with isopropyl b-d-1thiogalactopyranoside to a final concentration of 0.5 mM, and further incubated for 24 h at 18 °C, after which they were pelleted by centrifugation.

The pelleted bacteria were resuspended in the PBS containing 137 mM NaCl (Sigma); 2.7 mM KCl (Sigma); 1 mM EDTA (Sigma), and 0.15 mM phenylmethylsulfonyl fluoride (Sigma) and disrupted by sonication. Cell debris was eliminated by centrifugation at 14,000 × g, the supernatant collected and filtered through a membrane with 0.45 µm pore size and incubated with 0.05% N-dodecyl-b-d-maltopyranoside (DDM, Sigma) at 4 °C for 1 h prior to loading onto a 1 mL GST Hitrap FF column (GE Healthcare, Chicago, IL, USA). The column was then washed with a 10-fold volume of washing buffer (PBS with 0.02% DDM). To release rEXP2, HRV 3 C protease (Takara Bio, Shiga, Japan) in 50 mM Tris-HCl, 200 mM NaCl and 0.02% DDM was applied to the column. After a 16 h incubation at 4 °C, rEXP2 was eluted from the column with PBS buffer. All the purification steps were performed in an AKTA Explorer chromatographic system (GE Healthcare, Chicago, IL, USA). rEXP2 concentration was estimated using Pierce BCA Protein Assay Kit (Thermo Fisher Scientific).

Calcium imaging using Gcamp6f

HepG2 cells were transfected with Gcamp6f calcium reporter fluorescent plasmid (Addgene Plasmid #40755, Watertown, MA, USA, a gift from Mafalda Pimentel at iMM JLA) using FuGene 6 HD (Promega, Madison, WI, USA) and OptiMEM (Gibco). In total, 48 h after transfection, cells were imaged in Zeiss Cell Observer widefield fluorescent microscope for 3 min, using ZEN 2 software (Blue version). After an initial 30 s of imaging, the appropriate concentration of rEXP2 or vehicle control was added to the culture medium and cells were imaged for another 150 s. Fluorescence signal through time was normalized to the fluorescence at time −30s (F_i/F₋₃₀).

Potassium imaging using GEPII sensor

HepG2 cells were transfected with GEPII potassium reporter fluorescent plasmid (NGFI, Graz, Austria) using FuGene 6 HD (Promega, Madison, WI, USA) and OptiMEM (Gibco). In total, 48 h after transfection, cells were imaged in Zeiss LSM 710 confocal fluorescent microscope for 3 min, using ZEN 2 software (Blue version). After an initial 30 s of imaging, the appropriate concentration of rEXP2, digitonin or vehicle control was added to the culture medium and cells were imaged for another 150 s. Cells were excited using the 405 nm laser and fluorescence was collected using the following bandwidths: 420–480 nm for CFP channel and 500–700 nm for FRET channel. To calculate the FRET ratio, the intensity of the FRET channel was divided by the intensity of the CFP channel through time. These values were further normalized to the fluorescence at time −30s (F_i/F₋₃₀).

Chemicals and proteins

Desipramine, recombinant α-hemolysin (α-HL) from Staphylococcus aureus, recombinant streptolysin O (SLO), Digitonin and recombinant acid Sphingomyelinase (aSMase) from Bacillus cereus was purchased from Sigma (Kawasaki, Japan).

Liver infection and parasite burden determination

For sporozoite infections, mice were injected with 2 × 10⁴ sporozoites, in 200 µL of DMEM via intravenous injection. At different timepoints after sporozoite infection (6, 24, 48 h after infection), livers of infected mice were collected into 3 mL of denaturing solution (4 M guanidium thiocyanite (Sigma), 25 mM sodium citrate (Sigma), 0.5% sarcosyl (Sigma) in MilliQ water treated with DEPC (Sigma). Livers were then mechanically homogenized using 1 mm diameter silica beads (BioSpec Products, Bartlesville, OK, USA) in MiniBeadBeater homogenizer (BioSpec Products) for 2 min. After mechanical disruption of the tissue, 100 µL of homogenate was used to extract RNA, using NZY Total RNA Isolation Kit (NZYTech, Lisboa, Portugal), as per manufacturer’s instructions. In all, 1 µg of extracted RNA was converted into cDNA using NZY First-Strand cDNA Synthesis Kit (NZYTech), as per manufacturer’s instructions. cDNA was then used for quantitative Polymerase Chain Reaction (qPCR), by measuring the abundance of Pb18s RNA, compared with MmHprt RNA using either ViiA 7 (384-well plates, using QuantStudio, v1.3 software) or 7500Fast (96-well plates, using 7500Fast, v2.3 software) Real-Time PCR Systems (Thermo Fisher Scientific) using iTaq Universal SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA, USA). Analysis of results was performed using the ΔΔC_T method after export of C_T values from the collection softwares:

Primers used for liver load determination:

Pb18s – forward primer: AAGCATTAAATAAAGCGAATACATCCTTAC

Pb18s – reverse primer: GGAGATTGGTTTTGACGTTTATGTG

MmHprt– forward primer: TTTGCTGACCTGCTGGATTAC

MmHprt– reverse primer: CAAGACATTCTTTCCAGTTAAAGTTG

Infection of cells

To infect cells, the required number of sporozoites were suspended in complete DMEM and incubated with cells and centrifuged for 5 min at 1 600 × g. Cells were incubated for 2 hours at 37 °C, with a 5% CO₂ atmosphere, after which the medium was replaced with complete DMEM supplemented with 0.3% Fungizone (Gibco). For some experiments, compounds or rEXP2 protein was added during infection. These materials were either added to the cells concomitantly with sporozoites or added to the cells 1 hour after the sporozoites, as described in the figure legends.

Expression of parasite genes during invasion

After sporozoites and cells were incubated for the appropriate amount of time, mRNA from samples was extracted using NZY Total RNA Isolation Kit (NZYTech), as per manufacturer’s instructions. In all, 1 µg of extracted RNA was converted into cDNA using NZY First-Strand cDNA Synthesis Kit (NZYTech), as per manufacturer’s instructions. cDNA was then used for qPCR, by measuring the abundance of PbEXP2, PbEXP1 and PbGAP45 was compared with Pb18s RNA using either ViiA 7 (384-well plates, using QuantStudio software, v1.3) or 7500Fast (96-well plates, using 7500Fast software, v2.3) Real-Time PCR Systems (Thermo Fisher Scientific) using iTaq Universal SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA, USA). Analysis of results was performed using the ΔΔC_T method as described above.

Primers used for determination of expression of Plasmodium invasion genes:

Pb18s – forward primer: AAGCATTAAATAAAGCGAATACATCCTTAC

Pb18s – reverse primer: GGAGATTGGTTTTGACGTTTATGTG

PbEXP2 – forward primer: ACGATCCAGGTTTGATTG

PbEXP2 – reverse primer: TGGTAATAGTGGGACATTC

PbGAP45 – forward primer: GTGGAGTAGTCTTTAAGG

PbGAP45 – reverse primer: GTGGAGTAGTCTTTAAGG

PbEXP1 – forward primer: AGGGAAGACATCCATTCCAAATTGG

PbEXP1 – reverse primer: TGAAGATTTGGCATGTTAAGTGGTG

IFA of hepatoma cells

To process cells and parasites for IFA, experiments were performed either on black glass-bottom 96-well plates or on no. 1 grade 12 mm diameter glass coverslips. Cells and parasites were fixed at the appropriate time point in 4% paraformaldehyde (Santa Cruz Biotechnology, Dallas, TX, USA) for 10–20 min at room temperature. Cell and parasite fixed material were then blocked and permeabilized with 5% bovine serum albumin (BSA, NZYTech), 0.2% saponin (Sigma) in PBS for at least 30 min at room temperature. Cells and parasites were stained with the appropriate primary antibodies (diluted in 5% BSA, 0.2% saponin in PBS) for 2 h at room temperature in a humid chamber. Samples were washed three times with PBS for 10 min. After washing, samples were stained with appropriate fluorescent secondary antibodies (Thermo Fisher Scientific) (diluted 1:300 in 5% BSA 0.2% saponin in PBS) and with 1 mg/mL Hoechst 33342 (Thermo Fisher Scientific) for 1 hr at room temperature in a humid chamber. Stained samples were washed three times with PBS for 10 min. After the final wash, coverslips were mounted on glass slides using a drop of Fluoromount-G (Thermo Fisher Scientific). If black glass-bottom 96-well plates were being used, 50 µL of Fluoromount G was placed on the stained well.

For in and out experiments, fixed samples were incubated with αPbCSP antibody before permeabilization was performed, to stain parasites that would be outside cells. After incubation with αPbCSP antibody, samples were washed with PBS and were permeabilized and blocked using 5% BSA, 0.2% saponin in PBS. After permeabilization, samples were incubated with αPbUIS4 antibody and the staining progressed as explained above. More than 40 sporozoites were imaged per replicate per parasite line.

For time-course analysis, samples were imaged using either Leica DM5000B (Leica, Wetzlar, Germany), Zeiss Axiovert 200 M or Zeiss Cell Observer (Zeiss, Oberkochen, Germany), which are all widefield fluorescence microscopes, using a ×20 dry objective. Data were collected using Leica DFC Twain (v7.7.1), Metamorph (v7.7.9.0), and ZEN 2 (Blue version), respectively.

The primary antibodies used were goat αPbUIS4 (1:1000 dilution, Sicgen, Cantanhede, Portugal); mouse αPbCSP (1:1000 dilution, clone 3D11, Malaria Research and Reference Reagent Resource Center, MR4, BEI Resources, ATCC); mouse αPfEXP2⁶⁸ (1:500 dilution, clone 7.7, The European Malaria Reagent Repository, Edinburgh, Scotland, UK); rabbit αPfEXP2⁵³ (1:500 dilution, a kind gift from the laboratory of Brendan Crabb).

Sporozoite staining

To stain for sporozoite proteins, 5 × 10⁴ freshly dissected sporozoites were incubated in DMEM medium or DMEM containing 10% FCS at 37 °C on glass coverslips. After 30 min of incubation, sporozoites were fixed in 4% paraformaldehyde for 10–20 min at room temperature. Fixed sporozoites were permeabilized using ice-cold methanol for 5 min at −20 °C. After washing, sporozoites were blocked and stained as described above. Coverslips were imaged using Zeiss LSM 710 confocal laser point-scanning fluorescence microscope (using ZEN 2 software, Blue edition), using ×63 or ×100 oil objective. The primary antibodies used were goat αPbUIS4 (1:500 dilution, Sicgen, Cantanhede, Portugal); mouse αPbCSP (1:1000 dilution, clone 3D11, MR4); mouse αPfEXP2 (1:300 dilution, clone 7.7, The European Malaria Reagent Repository, Edinburgh, Scotland, UK); rabbit αPfEXP2 (1:300 dilution, a kind gift from the laboratory of Brendan Crabb⁵³) and rabbit αPbTRAP (1:1000 dilution, a kind gift from the laboratory of Joana Tavares⁶⁹) and mouse αPbRON4 (1:300 dilution, a kind gift from the laboratory of Maryse Lebrun⁷⁰).

For colocalization experiments, the plug-in Coloc 2 from FIJI image analysis software was used.

Quantification of EXP2-positive and EXP2-negative sporozoites in EXP2 cKO sporozoite population

To distinguish which sporozoites would be EXP2-sufficient and EXP2-deficient, immunofluorescence analysis was performed in WT or EXP2 cKO freshly dissected sporozoites stained with αEXP2 antibody. To assess the background fluorescence intensity of the staining, we also imaged sporozoites that were not stained with αEXP2 antibody. This allowed us to create a cutoff value for the intensity of EXP2 staining, below which EXP2 cKO sporozoites would be considered as EXP2-deficient (see Supplementary Fig. 1a–b) for an example of how sporozoites were analyzed). More than 50 sporozoites or EEFs were imaged per replicate per parasite line.

Gliding assay

For gliding assays, 5 × 10⁴ sporozoites were incubated in complete DMEM medium at 37 °C on glass coverslips. After 30 min of incubation, sporozoites were fixed in 4% paraformaldehyde for 10–20 min at room temperature. Fixed sporozoites were blocked as described above and only stained with αCSP antibody (1:1000 dilution, clone 3D11) and an anti-mouse secondary antibody (1:300 dilution) and Hoechst. Coverslips were imaged using a Leica DM5000B widefield fluorescence microscope and ×40 dry objective, using Leica DFC Twain (v7.7.1) software. Since CSP is shed in a circular fashion during sporozoite gliding, sporozoite motility can be quantified by counting the circles of CSP (termed trails) produced by each individual sporozoite. Sporozoites were divided into three groups (no trails; 1–10 trails; >10 trails) based on the number of trails that it displayed (see Supplementary Fig. 1e for an example of how sporozoites were separated between these groups). More than 75 sporozoites were imaged per replicate per parasite line.

Flow cytometry analysis of cell traversal

To quantify the level of traversal, HepG2 cells and sporozoites were incubated in the presence of 0.5 mg/mL of 10 kDa Dextran-Rhodamine (Thermo Fisher Scientific) in complete DMEM at 37 °C. The dextran molecule is passively taken up by cells that have been traversed and is detected because of the Rhodamine dye, which has Excitation/Emission maxima at 570/590 nm. Data collection was performed using BD Accuri C6 cytometer (Franklin Lakes, New Jersey, USA) and Accuri C6 software (v1.0.264.21) software was used. Data analysis was performed using FlowJo X software (FlowJo LLC, Ashland, OR, USA) (see Supplementary Fig. 1f for the gating strategy used).

Sporozoite genomic DNA purification

After dissection from salivary glands, sporozoites were pelleted by centrifugation in a table top centrifuge (Eppendorf) at maximum speed for 15 min. Genomic DNA was purified using NZY Blood gDNA Isolation kit (NZYTech). Gene abundance in sporozoite genomic DNA was analyzed using either ViiA 7 (384-well plates, using QuantStudio v1.3 software) or 7500Fast (96-well plates using 7500Fast v2.3 software) Real-Time PCR Systems (Thermo Fisher Scientific) using iTaq Universal SYBR Green Supermix (Bio-Rad). Analysis of results was performed using the ΔΔC_T method described above.

Primers used for assessing EXP2 recombination/excision:

Pbdhfr – forward primer: GTTGGTTCGCTAAACTGCATC

Pbdhfr – reverse primer: CTGTTTACCTTCTACTGAAGAGG

Pbhsp70 – forward primer: TGCAGCAGATAATCAAACTC

Pbhsp70 – reverse primer: ACTTCAATTTGTGGAACACC

Blood stage parasite protein isolation

For preparation of samples from blood stages, BALB/c mice were infected with P. berghei NK65 EXP2-HA (PbEXP2-HA) and parasites multiplied until reaching a parasitemia of ~10%. Mice were sacrificed using isoflurane (Abbott Laboratories, Lake Bluff, IL, USA) overdose and blood was removed by heart puncture. Red blood cells were suspended in PBS and pelleted by centrifugation at 450 × g for 5 min without brake. Pelleted red blood cells were lysed using 0.05% Saponin containing cOmplete Protease Inhibitor Cocktail (Sigma) in PBS for 5 min of ice. Parasites were pelleted at 2500 × g for 5 min and were lysed in 500 µL of radioimmunoprecipitation assay (RIPA) buffer buffer with 1% sodium dodecyl sulphate (SDS) for 1 h on ice.

Sporozoite secretion assay

Sporozoites were dissected as described above and purified using 17% Accudenz protocol⁷¹, where dissected sporozoites were placed on top of a 17% Accudenz solution and were centrifuged for 20 min at 2500 × g at room temperature. In all, 1 mL of the top fraction was removed and centrifuged at 20,000 × g for 10 min. Pelleted sporozoites were resuspended in simple DMEM and counted. In all, 2 × 10⁵ sporozoites were then transferred into a tube that would be kept on ice or at 37 °C, with or without 10% FCS. After 30 to 120 min of incubation, sporozoites were pelleted again by centrifuging at 20,000 × g for 10 min. Supernatants was separated from the pelleted sporozoites and sporozoites were lysed with RIPA buffer with 1% SDS for 1 h on ice. RIPA buffer comprised 50 mM Tris-HCl (NZYTech), 150 mM NaCl (Sigma), 5 mM EDTA (Sigma), 10 mM NaF (Sigma), 1% Triton X-100 (USB Corporation, Cleveland, OH, USA), 0.5% sodium deoxycholate (AppliChem, Maryland Heights, MO, USA). Following lysis, NuPAGE loading dye (Thermo Fisher Scientific) and β-mercaptoethanol (to a final concentration of 10%, Sigma) were added to all samples. Samples were boiled for 15 min at 95 °C.

WB analysis

Samples were separated in 10% Acrylamide gels and transferred to 0.2 µm pore-sized Nitrocellulose membranes (Bio-Rad). Membranes were placed inside Falcon tubes and blocked with 5% skim milk (Nestle, Vevey, Switzerland) overnight at 4 °C, on a tube roller. Membranes were then incubated with the appropriate primary antibodies, diluted in 5% skim milk, for 1 hour at room temperature. Membranes were washed with 0.1% Tween 20 (Sigma) in PBS, three times for 10 minutes. Membranes were then incubated with the appropriate secondary antibodies, conjugated to Horseradish peroxidase diluted in 5% skim milk, for 1 hour at room temperature, after which they were washed again three times for 10 minutes. WB were revealed using Amersham ECL Prime Blotting Detection Reagent (GE Healthcare) on the ChemiDoc XRS + system (Bio-Rad) using ImageLab software (v5.2.1). The primary antibodies used were rabbit αHA tag (1:1000 dilution, clone C29F4, Cell Signaling Technology, Danvers, MA, USA); mouse αPbCSP (1:5000 dilution, clone 3D11, MR4) and rabbit αPbBiP (1:1000 dilution, GenScript Biotech. Corp, New Jersey, NJ, USA). The secondary antibodies used were goat anti-mouse IgG F(ab’)2, polyclonal antibody horseradish peroxidase (HRP) conjugate (1:5000 dilution, Enzo Life Sciences, Lausen, Switzerland); anti-rabbit IgG, HRP-linked antibody (1:3000 dilution, Cell Signaling Technology).

To assess the amount of secreted EXP2 protein, the intensity of the EXP2 band in the supernatant fraction was normalized to the intensity of the CSP band intensity at the respective time point in the pellet fraction. These values were further normalized to the ratio of EXP2/CSP in the pellet fraction at time zero (no stimulation).

All membranes used for this manuscript are presented in Supplementary Fig. 4.

Acid sphingomyelinase knockdown

Acid sphingomyelinase (gene name SMPD1) and Scramble short-hairpin RNAs were obtained from Mission shRNA (Sigma). In all, 800 ng/well of shRNAs were mixed with OptiMEM (Thermo Fisher Scientific) containing 0.3% Lipofectamine RNAiMax (Thermo Fisher Scientific) and added to 100,000 HepG2 cells. Cells were infected with sporozoites 48 hours after transfection. Samples were fixed 2 hours after infection and were stained as described above.

To assess the amount of knockdown of acid sphingomyelinase, mRNA was collected from uninfected cells, lysed at the time of infection. To quantify the amount of acid sphingomyelinase RNA, RNA was extracted using NZY Total RNA Isolation Kit, following manufacturer’s instructions. 1 µg of extracted RNA was converted into cDNA using NZY First-Strand cDNA Synthesis Kit (NZYTech), as per manufacturer’s instructions. cDNA was then used for qPCR, by measuring the abundance of HuSMPD1 RNA, compared with HuHPRT RNA using either ViiA 7 (384-well plates, using QuantStudio v1.3 software) or 7500Fast (96-well plates, using 7500Fast v2.3 software) Real-Time PCR Systems (Thermo Fisher Scientific) using iTaq Universal SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA, USA). Analysis of results was performed using the ΔΔC_T method described above.

Primers used to assess aSMase knockdown:

HsHPRT– forward primer: TTTGCTGACCTGCTGGATTAC

HsHPRT– reverse primer: CAAGACATTCTTTCCAGTTAAAGTTG

HuSMPD1 – forward primer: GGCCCACATTTGGGAAAGTT

HuSMPD1 – reverse primer: TTCACCGGATGATCTTGCCT

Electron microscopy

Immuno-electron microscopy of purified parasites was performed according to Tokuyasu technique. In brief, the pellet was chemically fixed in 0.1 M phosphate buffer containing 2% formaldehyde and 0.2% glutaraldehyde, embedded in food-grade gelatine and cryopreserved in 2.3 M sucrose. Gelatine blocks were shaped in cubes and froze in liquid nitrogen and sectioned at −110 °C using a cryo-ultramicrotome (UC7 and FC7, Leica) to generate 70 nm sections. Sections were collected and thawed in a mixture of 2.3 M sucrose and 2% methylcellulose. Immuno-labeling was done in 1% bovine serum albumin and 0.8% gelatine from cold water fish skin in PBS with rabbit αEXP2 primary antibody (1:500 dilution) and 15 nm gold coupled Protein A (CMC Utrecht, The Netherlands, 1:50 dilution). After immuno-labeling, the sections were stained and mounted in a mixture of 3% (aq.) uranyl acetate and 2% methylcellulose. Images were recorded using a Hitachi H-7650 electron microscope (Hitachi, Tokyo, Japan) at 100 kV acceleration.

Image analysis

All immunofluorescence images were processed using FIJI software (version 1.52i) and macros written for each analysis to automate it. For WB images, the software ImageLab was used (version 5.2.1)

Flow cytometry analysis

Flow cytometry data were analyzed using FlowJo version X software (FlowJo LLC, Ashland, OR, USA).

Statistical analysis

Statistical analysis was performed using GraphPad Prism 5 software (GraphPad, La Jolla, CA). Mann–Whitney U test was used to assess significance of differences observed between two groups and non-linear regression was performed to assess Desipramine inhibitory effects on invasion. Raw data and a description of statistical tests used is provided in Supplementary Information. Plots and figures were prepared using Adobe Illustrator (version CS4).

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.