Innate immunity is an evolutionarily ancient mechanism that provides general host protection against pathogens (1
). In mammals, innate immunity functions alongside adaptive immunity and also plays a key role in its activation (2
). On the other hand, many organisms, including the bacterivore nematode Caenorhabditis elegans
, lack adaptive immunity and defend themselves solely with innate immune mechanisms (4
). Some core components of innate immunity are conserved, but the particular means of protection used by different animals vary. In mammals, the innate system depends on physical and anatomical barriers (e.g., the barrier epithelial cells, mucus, tears, earwax, and stomach acid), the humoral component (i.e., cytokines, chemokines, and defensins secreted by innate immune cells), and several types of phagocytic cells, among them are macrophages, which recognize and destroy pathogens (6
). C. elegans
has no specialized immune cells and relies on its barrier tissues, epidermal and intestinal cells, for defense (4
). An efficient immune response is provided by the high secretion capacity of the particular host cells, as they release a variety of antimicrobial peptides and enzymes that can directly attack pathogens (4
). In all organisms studied, the innate immune response is regulated at both the transcriptional and posttranscriptional levels (4
). The signaling pathways and transcriptional factors that control innate immunity in worms have been studied in detail, but less is known about the posttranscriptional mechanisms involved in this process.
In eukaryotes, most mRNAs are polyadenylated by the canonical polyadenylate [poly(A)] polymerase during mRNA 3′ end processing in the nucleus (8
). The poly(A) tail is essential for mRNA stability, export to the cytoplasm, translation, and turnover (10
). Poly(A) tails are gradually shortened in the cytoplasm by deadenylases, and their reduction to less than 20 nucleotides (nt) leads to mRNA degradation (11
). However, in some cases, a poly(A) tail can be extended in the cytoplasm by noncanonical poly(A) polymerases (ncPAPs). These enzymes belong to the family of terminal nucleotidyltransferases (TENTs) and are implicated in a range of physiological processes (12
). Cytoplasmic polyadenylation has been mostly studied in the context of the activation of dormant deadenylated mRNAs during gametogenesis (15
) and in neuronal processes (18
), in which GLD-2 (Germ Line Development 2)/TENT2 polyadenylates certain mRNAs in response to cellular signals. The recent discovery of the TENT5 family of cytoplasmic ncPAPs and the characterization of their functions expanded the repertoire of physiological processes that are affected by this type of posttranscriptional regulation (20
). Mammalian genomes encode four TENT5 proteins (TENT5A to TENT5D, also known as FAM46A to FAM46D), all of which are active ncPAPs (21
). Functional analysis revealed that TENT5C is a multiple myeloma growth suppressor (21
). In multiple myeloma cell lines, TENT5C polyadenylates and stabilizes numerous mRNAs that encode secreted proteins (21
). TENT5C also plays a crucial role in the regulation of immunoglobulin expression and the humoral immune response in mice through polyadenylation of mRNAs that encode immunoglobulins (23
). TENT5A polyadenylates mRNAs encoding collagens and is thus required for the proper bone formation (26
). However, TENT5 proteins are differentially expressed in mammalian tissues and organs with potential redundancy, which makes study of their functions difficult.
Here, we characterized the only TENT5 family member in C. elegans, F55A12.9/PQN-44, which we renamed TENT-5. Transcriptomic and proteomic analysis, along with functional studies, revealed that TENT-5 is an innate immune response regulator. Poly(A) tail profiling by direct RNA sequencing (RNA-seq) showed that TENT-5 polyadenylates and stabilizes mRNAs that encode defense proteins. tent-5 deficiency led to an impaired innate immune response in worms. The role of TENT5 proteins in innate immunity is evolutionarily conserved because murine macrophages devoid of TENT5A and TENT5C also exhibited defects in polyadenylation of mRNAs that encode proteins with a role in innate immunity. Together, we identified C. elegans TENT-5 and its mammalian orthologs TENT5A and TENT5C as previously unknown players that regulate innate host defenses.
In its natural habitat, C. elegans
encounters numerous and diverse pathogens. Worms’ fitness and rapid adaptation to an ever-changing microbial environment require dynamic and highly efficient modulation of the immune response. It is widely appreciated that in animals, innate immunity is orchestrated by a plethora of transcriptional and posttranscriptional mechanisms (7
). The posttranscriptional aspect of innate immune regulation in worms is a fast-growing field of research (58
). C. elegans
can sense bacterial noncoding RNAs to induce avoidance of a pathogen (85
). A few reports have also implicated microRNAs in host defense (86
), whereas viral RNA uridylation, by one member of the TENT family, CDE-1, has been shown to play a role in the antiviral response (89
). Cytoplasmic polyadenylation is a powerful posttranscriptional mechanism that shapes the transcriptome and consequently also the proteome, through the regulation of mRNA stability and translation efficiency. This work shows that cytoplasmic polyadenylation by ncPAP TENT-5 positively regulates innate immunity in C. elegans
TENT-5 is a cytoplasmic protein that is expressed through the whole life cycle of the worm in multiple cells of the body, including the intestine (Fig. 1
). In worms, the immune response and digestion are connected. Many enzymes responsible for the macromolecular degradation of food participate in the degradation of pathogen-derived macromolecules (61
). Upon infection with bacterial pathogens that infect through the gut, intestinal cells secrete a large amount of enzymes and antibacterial proteins. Intestinal cells must have an enormous capacity for protein synthesis and secretion. Our results show that TENT-5 preferentially polyadenylates and stabilizes mRNAs that encode short secreted proteins with a role in digestion and immunity (Figs. 5
). The length of the poly(A) tail is critical for mRNA stability, and the posttranscriptional lengthening of mRNAs’ poly(A) tails in the cytoplasm may thus extend their half-life. Such a mechanism, extending mRNA longevity and promoting translation, would be not only energy effective but also extremely fast, a valuable feature when a rapid reaction to changing environmental conditions is needed, for example, during immune or stress responses.
The relevance of TENT-5 in innate immunity is also supported by the reduced survival of tent-5
–deficient worms upon infection with a range of bacterial pathogens (Fig. 4
). Mutant worms display a moderately reduced life span also when grown on E. coli.
Furthermore, this survival defect is associated with the increased bacterial load in the intestine of the tent-5
–deficient worms. This observation may suggest that because of the decreased basal expression of genes encoding digestive enzymes and cytoprotective proteins, mutant worms have limited ability to deal even with relatively nonpathogenic food, exhibiting reduced life span. In such an interpretation, TENT-5 may not be solely dedicated to innate immunity but rather influence stability and potentially translation efficiency of many mRNAs encoding secreted proteins. The consequences of tent-5
deficiency are, however, most evident under physiological conditions when secretion plays a life-saving role. The fact that TENT-5 is up-regulated upon infection may be explained by specific pathogen–induced transcriptional regulation or by a mechanism that senses the demand for efficient protein secretion. Further research will be needed to determine what other secretion-dependent processes are regulated by TENT-5.
Direct sequencing of RNA samples prepared from tent-5
mutants and wild-type worms allowed us to uncover substrates of TENT-5 enzymatic activity (Fig. 5
). Our analysis of poly(A) tails lengths in L4 C. elegans
is consistent with previous results (52
). Those studies identified a negative correlation of poly(A) tail length with mRNA expression. Given the counterintuitive nature of these observations, we sought to take advantage of the high quality of our DRS data and analyzed the relationship between these features. In accordance with (52
), our data showed a similar phenomenon (fig. S5). However, TENT-5 substrate mRNAs showed significantly decreased expression levels in tent-5(tm3504)
mutants, suggesting that shortening of the poly(A) tail from its wild-type length lowers mRNAs expression (Fig. 5
). Thus, our results show that for TENT-5 substrates, an increase in poly(A) tail length sustains mRNA steady-state levels.
Our data indicate that TENT-5 affects mRNAs that encode short secreted proteins. Among the prominent TENT-5 targets are mRNAs of the nspc
family of genes for which mRNA polyadenylation and expression are strongly decreased in the tent-5
mutant (Fig. 5
). In our previous studies, we also found that TENT5 family members regulate the expression of genes encoding secreted proteins. In B cells, TENT5C polyadenylates mRNAs that encode immunoglobulins (23
), and in osteoblasts, the main substrates of TENT5A are mRNAs encoding collagens (26
). In line with that, we show that mRNAs that encode secreted proteins constitute a large fraction of TENT5A and TENT5C direct substrates in macrophages (Fig. 7
). In agreement with the nature of TENT5 substrates, these poly(A) polymerases are associated with the ER, and this is probably the main determinant of their substrate specificity. It was recently shown that TENT5C is actively recruited to ER in human cells through interaction with fibronectin type III domain–containing proteins FND3CA and FNDC3B (64
). Disturbance of TENT5C function leads to ER shrinking, destabilization of ER-translated mRNAs, and defects in ER-mediated protein folding and secretion (64
). It is probable that TENT5C’s binding to the ER, mediated by FNDC3A and FNDC3B, may be enough to target its substrates. Whether the worm’s TENT-5 target selection corresponds to an analogous mechanism requires further investigation, because among proteins with fibronectin type III domains, there are no obvious FNDC3 orthologs encoded in the genome. Controlling the secretory capacity of the ER is key for resistance to infection (90
). Cytoplasmic polyadenylation by TENT5 proteins of mRNAs encoding proteins that transit the ER provides a new layer to the regulation of secretion essential for the proper immune response and other physiological processes such as bone formation in vertebrates.
Last, we demonstrate that the role of TENT-5 in innate immunity is evolutionarily conserved. In murine macrophages, TENT-5 orthologs, TENT5A and TENT5C, polyadenylate mRNAs that encode lysozyme and cathepsin proteases (Fig. 7
), increasing their protein level. Lysozyme is one of the most abundant antimicrobial proteins secreted by macrophages, and deletion of Lyz2
increases susceptibility to infection with P. aeruginosa
, Micrococcus luteus
, and Klebsiella pneumoniae
due to impaired clearance of pathogen and can lead to higher host mortality (71
). Cathepsins not only play a notable role in lysosomal protein breakdown but also regulate the immune response (92
). BMDMs isolated from Ctsd
-deficient mice display enhanced susceptibility to Listeria monocytogenes
infection and increased intraphagosomal viability of bacteria (75
). Moreover, cathepsin D protein levels were up-regulated after infection of murine macrophages with Bacillus subtilis
, P. aeruginosa
, L. monocytogenes
, S. aureus
, and E. coli
, and again, its deficiency led to an increase in the amounts of each of these bacteria inside populations of macrophages (76
). Cathepsin B was also required for optimal posttranslational processing of tumor necrosis factor–α (TNF-α) in response to LPS, and BMDMs from Ctsb
-deficient mice secrete significantly less TNF-α than wild-type macrophages (77
Our work demonstrates the conserved role of mRNA polyadenylation and TENT5 family ncPAPs in the regulation of innate immunity in animals. Taking into account that, in worms, TENT-5 is expressed in multiple tissues, we expect that its functions go beyond protection against pathogens and may be generally important for physiological processes that involve protein secretion. In mammals, TENT5A to TENT5D are expressed in different tissues and developmental stages, opening the possibility for them to have broad biological significance and functional interactions too.
MATERIALS AND METHODS
Bacterial strains and culture
HB101 was cultured at 37°C in LB medium supplemented with streptomycin (final concentration of 0.1 mg/ml). P. aeruginosa
PAO1 and S. marcescens
Db10 were grown at 37°C in LB medium without antibiotics (48
). P. luminescens
Hb was grown at 30°C in LB medium without antibiotics (48
). S. aureus
NCTC 8325 (Argenta Ltd.) was cultured at 37°C on tryptic soy agar (TSA) or tryptic soy broth (TSB) (both from BD Biosciences) supplemented with nalidixic acid (Nal) (final concentration of 10 μg/ml). E. coli
DH5α and MH1 strains were cultured at 37°C in LB supplemented with appropriate antibiotics. All antibiotics were purchased from Sigma-Aldrich.
C. elegans strains and maintenance
C. elegans was maintained on nematode growth medium (NGM) plates seeded with E. coli HB101 at 20°C unless otherwise specified. Strains obtained from the National Bioresource Project of Japan (NBRP) and strains generated by the CRISPR-Cas9 were outcrossed two to nine times to the wild-type strain. The following C. elegans strains were used: N2 Bristol (wild type) and VK2664 (vkEx2664 [nhx-2p::CemOrange2::tram-1, myo-2p::gfp]) strains were obtained from the Caenorhabditis Genetics Center; tent-5(tm3504) I was obtained from NBRP; ADZ20 (tent-5(rtt5) I), ADZ21 (tent-5(rtt6[tent-5::gfp::3xflag]) I), and ADZ24 (vgln-1(rtt9[vgln-1::mKate2:::3xmyc] II) were generated in this study; ADZ87 (tent-5(rtt6[tent-5::gfp::3xflag]) I, vkEx2664 [nhx-2p::CemOrange2::tram-1, myo-2p::gfp]) strain was obtained by crossing ADZ21 with VK2664.
C. elegans transgenic strain generation
Transgenic worm strains were generated using standard microinjection protocols. Plasmids that were used for microinjections were purified with PureLink mini-prep kit (Thermo Fisher Scientific, K210002). Injections were conducted using the Axio Observer D1 inverted microscope (Zeiss) equipped with a Femto Jet 4i microinjection system (Eppendorf). For each transformation, at least two independent transgenic strains were obtained. All oligonucleotides and DNA constructs used for transgenic strain generation are listed in tables S2 and S3, respectively.
The KO strain ADZ20 tent-5(rtt5) I
that harbors a 2909–base pair deletion, which spans from the start to stop codon of tent-5
isoform d, was generated using an adapted version of the CRISPR-Cas9 protocol (94
). Mutation in the dpy-10
gene was used as a CRISPR co-conversion marker. Single-stranded DNA (ssDNA) encoding tracrRNA (trans-activating CRISPR RNA) and CRISPR RNA (crRNA) with 20 N of single-guide RNA (sgRNA) sequences was used for the preparation of templates for in vitro sgRNA synthesis. Briefly, 5 μl of 100 μM VL311 tracrRNA oligo was annealed with 5 μl of 100 μM sgRNA oligo (VL312, VL313, and VL315) in the 50-μl mix containing deoxynucleotide triphosphates, Phusion buffer, and Phusion Hot Start II Polymerase and incubated at 98°C for 3 min; 98°C for 10 s, 65°C for 20 s, and 72°C for 5 s for 10 cycles; and 72°C for 5 min. The reaction was purified with AMPure XP magnetic beads [1:1.4 (v/v) mix to beads; Beckman Coulter, A63882]. In vitro transcription was assembled by mixing 400 ng of DNA template in 35 μl of RNase-free water, 5 μl of ribonucleoside triphosphates (NTPs) mix (20 mM each), 10× transcription buffer [200 mM tris-HCl (pH 7.9), 30 mM MgCl2
, 50 mM dithiothreitol (DTT), 50 mM NaCl, and 10 mM spermidine], 1.25 μl of RiboLock RNase Inhibitor (40 U/μl) (Thermo Fisher Scientific, EO0384), and 4 μl of T7 polymerase (homemade). Eight reactions were set up for the single sgRNA transcription. Following incubation at 37°C for 3 hours, each sample was treated with 0.5 μl of TURBO deoxyribonuclease (DNase) (2 U/μl; Thermo Fisher Scientific, AM2239) at 37°C for 30 min. RNA was purified from the pooled reactions with the phenol/chloroform extraction and purified further through electrophoresis in 6% urea polyacrylamide gel electrophoresis (PAGE). Animals were injected with the following mix: 15.5 μM Cas9 (Cas9::NLSSV40
protein; homemade), 5.9 μM sgRNA–dpy-10, 11 μM sgRNA–tent-5-1, 11 μM sgRNA–tent-5-2, 0.44 μM single-stranded oligodeoxynucleotide (ssODN)–dpy-10, 0.88 μM ssODN–tent-5, 150 mM KCl, and 20 mM Hepes (pH 8.0). Worms showing dumpy phenotype in F1
progeny were screened for tent-5
deletion using PCR, and, later, the deletion was confirmed by Sanger sequencing. Mutant worms were backcrossed two times to wild-type worms to cross out the dpy-10
mutation and CRISPR off-targets.
The knockin strain ADZ21 tent-5(rtt6[tent-5::gfp::3xflag]) I
was generated by CRISPR-Cas9 according to (95
). The gRNA sequence (5′-TGCCACCAGATGCAGCTACA-3′) was cloned into pDD162 to generate pDD162_sgRNA425
. Homology arm regions were amplified by PCR using genomic DNA (gDNA) as a template and were inserted into pDD282, resulting in the pDD282-tent-5::gfp::3xflag
construct. N2 worms were injected with the following mix: pDD282-tent-5::gfp::3xflag
(10 ng/μl), pDD162_sgRNA425
(50 ng/μl), myo-2p::mCherry
pharyngeal coinjection marker pCFJ90 (2.5 ng/μl), and myo-3p::mCherry
body wall muscle coinjection marker pCFJ104 (5 ng/μl). The selection of positive knockin candidates was performed as described (95
). Animals with successful GFP-tag insertion were backcrossed three times with the wild-type strain to get rid of CRISPR off-target effects. The knockin strain ADZ24 vgln-1(rtt9[vgln-1::mKate2:::3xmyc] II
was generated by CRISPR-Cas9 using constructs encoding gRNA, pDD162_sgRNA456
(gRNA sequence: 5′-CGTTCTTTACCAACGACGAG-3′), and homology arm regions, pDD287-vgln-1::mKate2::3xmyc.
General cloning techniques were conducted according to the well-established protocols (96
) or manuals provided by the manufacturers of kits. All plasmids were generated using either classical restriction enzyme digestion and ligation or sequence- and ligation-independent cloning (SLIC) (97
) and validated by digestion with restriction enzymes and sequencing. All oligonucleotides and DNA constructs are listed in tables S2 and S3. To generate pDD162-sgRNA425
(pVL060), two PCRs were performed using pDD162 as a template, with primers VL342 and VL344 and with VL343 and VL345. Fragments were gel-purified and used in a 1:1 molar ratio as templates for PCR with primers VL386 and VL387. The product was purified from a gel and used for the SLIC with pDD162 that has been digested with Nde I and Sph I. pDD162-sgRNA456
(pVL069) was cloned in a similar way using primers VL365 and VL364 instead of VL344 and VL345. pDD282-tent-5::gfp::3xflag
(pVL062) was prepared as follows: Arms homologous to tent-5
were amplified with VL347 and VL348 and with VL349 and VL350 on gDNA isolated from N2. pDD282 was digested with Avr II and Spe I, and the reaction was purified using a Clean-Up kit (A&A Biotechnology). For pDD287-vgln-1::mKate2::3xmyc
construct (pVL070), homology arms were amplified from gDNA with primers VL366 and VL367 as well as VL368 and VL369. The pDD287 vector was digested with Avr II and Ngo MIV. In both cases, 200 ng of vector and 50 ng of each homology arm products were used for SLIC. The pClneo-NHA (N-terminal lambda symbolN boxB-binding domain and an HA-tag) constructs for tethering assays pCI-NHA-tent-5WT
were cloned as follows: PCR products were generated with primers NHATENT-5_fw and NHATENT-5_rev on plasmids carrying tent-5WT
(isoform a) genes and subsequently cloned into Sal I and Not I sites of the pClneo-NHA.
All mice lines were generated by CRISPR-Cas9 in the Genome Engineering Unit (https://crisprmice.eu/
) using methods described in (21
). Briefly, a cKO Tent5aFlox/Flox
/Tar) mouse line was created by insertion of LoxP
sites in introns flanking exon 2, which contains triplets encoding the catalytic center of the protein (D144N and D146N). Cas9-generated double-strand breaks in gDNA were targeted using two chimeric sgRNA. Bam HI restriction sites were inserted next to LoxP
sites to facilitate genotyping. Donor mice were handled and injected as described before (26
). The CRISPR cocktail consisted of mRNA Cas9 (25 ng/μl), sgRNAs (15 ng/μl), and ssDNA repair template (6 ng/μl). Correct integration of LoxP
sites was confirmed by Sanger sequencing and followed by routine mice genotyping. Sequences of the sgRNAs, ssDNA donor, and primers used for sequencing and genotyping of Tent5aFlox/Flox
mice can be found in table S2. Double Tent5aFlox/Flox Tent5c−/−
mouse line was obtained by crossing Tent5aFlox/Flox
cKO with the previously described Tent5c−/−
/Tar) KO line (21
/Tar) knockin mouse line was generated as described in (21
), with the exception that 3xFLAG was added instead of 1xFLAG. Tent5a
) knockin line were described previously (26
). Mice were bred in the animal house of Faculty of Biology, University of Warsaw and maintained under conventional conditions (21
). All animal experiments were approved by the First Local Ethical Committee in Warsaw affiliated to the University of Warsaw, Faculty of Biology (approval numbers: WAW/176/2016 and WAW/772/2018) and were performed according to Polish Law (act number 266/15.01.2015) and in agreement with the corresponding European Union directive.
Primary BMDM cell culture
The primary BMDM cell cultures were established from the bone marrow monocytes isolated from Tent5aFlox/Flox Tent5c−/−
-3xFLAG, and wild-type mice. Mice were euthanized by cervical dislocation at ages 13 to 22 weeks. Femurs and tibias were isolated, the ends of bones were cut, and the bone marrow was flushed with medium using a 25-gauge needle. Bone marrow cells were plated in Iscove’s modified Dulbecco’s medium (Thermo Fisher Scientific, 21980065) supplemented with 10% fetal bovine serum (FBS) (Gibco), penicillin (100 U/ml)/streptomycin (0.1 mg/ml) solution (Sigma-Aldrich), and macrophage colony-stimulating factor (10 ng/ml; PeproTech, 315-02) and cultured at 37°C in 5% CO2
as described previously (99
). For conditional gene targeting, BMDMs derived from Tent5aFlox/Flox Tent5c−/−
and wild-type mice were transduced on the 8th day after isolation with 1 ml of concentrated lentivirus solution per 1 million cells. The medium was changed 16 hours after transduction. The lentivirus production was performed as described previously (21
). Lentiviral packaging (pMD2.G) and envelope (psPAX2) plasmids were provided by D. Trono (École Polytechnique Fédérale de Lausanne, Switzerland). The pCAG-Cre-IRES2-GFP plasmid was a gift from J. Jaworski (International Institute of Molecular and Cell Biology, Warsaw, Poland). The floxed locus was genotyped 3 days after transduction using gDNA isolated from 0.5 million cells with a Genomic Mini kit (A&A Biotechnology). Sequences of primers used for genotyping are listed in table S2. For BMDM stimulation with LPS, on the 14th day after isolation, cells were treated with LPS (100 ng/ml; Santa Cruz Biotechnology, sc3535) for 4 to 16 hours depending on the experiment.
Sequences of TENT proteins used for the phylogenetic analysis have been obtained from the WormBase WS272 (C. elegans
) and UniProt (other organisms), and their IDs are listed in table S4. Sequences were aligned using the PROMALS3D server (100
). Input sequence alignment for phylogenetic analysis was performed with MUSCLE (101
). The phylogenetic tree was built using the neighbor-joining method and visualized with iTOL v5 (102
Worms’ brood size, body parameters, and locomotion analyses
For brood size analysis, four individual L4 larvae per replicate were placed onto single 35-mm NGM plates seeded with E. coli HB101 and were allowed to lay eggs at 20°C. Worms were transferred to fresh plates every 12 hours until they no longer produced embryos. Eggs were counted after the adult was moved. For each strain, 10 worms have been analyzed in two independent trials (two-tailed unpaired t test). For worm’s body and movement analysis, eight age-synchronized young adult worms per strain were placed onto 35-mm NGM plates seeded with E. coli, and the worm movement was recorded for 2 min using the WormLab system (MBF Bioscience). The frame rate, exposure time, and gain were set to 7.5 frames/s, 0.0031 s, and 1, respectively. The worms’ body length and width, track length, center point speed, and the overall track pattern of individual worms were analyzed using the WormLab software (MBF Bioscience). For each strain, 80 worms have been analyzed, and data were compared using the two-tailed unpaired Student’s t test with Welch’s correction and presented as mean values ± SD. A P < 0.05 was considered significantly different from control: ns, not significant; *P < 0.05, **P < 0.001, and ***P < 0.0001.
Worms were immobilized with tetramisole, placed on slides coated with 2% agarose, and immediately imaged. The confocal microscopy for Fig. 1
and fig. S1 was performed using an FV1000 system with a 60×/1.2 water immersion lens (Olympus). Images were processed using Fiji/ImageJ software (version 2.0.0-rc69/1.52p) (103
). For colocalization analysis presented in Fig. 6
, worms were imaged using Zeiss LSM800 confocal microscope with 40×/1.2 water immersion apochromatic objective. Z-stack images were processed using Imaris 8.3 software. Median filter with a 3 × 3 × 1 kernel was applied to remove noise. Analysis was restricted to cells expressing CemOrange2–TRAM-1 ER protein. Gating with polygon was used to exclude the strongest unspecific signal from the green channel (Fig. 6
, marked with asterisks). Pearson correlation coefficient was calculated with the ImarisColoc module based on three-dimensional data obtained from 18 worms.
Tethering assays were performed as previously described (21
). 293T cells (American Type Culture Collection, CRL-3216) were cultured in Dulbecco’s modified Eagle’s medium (Invitrogen) supplemented with 10% FBS (Gibco) and penicillin (100 U/ml)/streptomycin (0.1 mg/ml) (Sigma-Aldrich) at 37°C in 5% CO2
. Cells were seeded into six-well plates and allowed to grow until about 70 to 80% confluence. Next, cells were cotransfected with 0.1 μg of pRL-5Box plasmid carrying an RL
) and 2 μg of plasmid encoding tethered wild-type or catalytically inactive NHA–TENT-5 using 5 μl of Lipofectamine 2000 and Opti-MEM medium (Thermo Fisher Scientific, 31985047) according to the manufacturer’s instructions. Cells were collected 24 hours after transfection for RNA (Northern blot and DRS) and protein level analyses.
C. elegans cultures for RNA analysis
Animal populations were synchronized by bleaching of the gravid adults and starvation of L1 larvae for 16 hours. Synchronized worms were grown on NGM plates seeded with E. coli
HB101 at indicated temperatures until they reached the L4 stage. Worms were washed three times with 50 mM NaCl (900g
for 2 min at room temperature) and resuspended in 1 ml of TRI Reagent (Sigma-Aldrich, T9424). Samples were incubated for 5 min at room temperature and stored at −80°C. For RNA analysis after C. elegans
infection by S. aureus
, synchronized worms were grown on NGM plates seeded with E. coli
HB101 at 25°C until they reached the L4 stage. The infection plates were prepared as described (37
). Briefly, TSA plates with Nal (10 μg/ml) were prepared 1 week before the experiment and stored at 4°C in the dark. S. aureus
was grown in TSB + Nal overnight at 37°C. Five hundred microliters of the overnight S. aureus
culture was uniformly spread onto the entire surface of 100-mm TSA + Nal plates and incubated at 37°C for 6 hours. L4 worms were washed three times with sterile 50 mM NaCl and seeded onto infection TSA plates that were previously warmed to room temperature. After 8 hours of infection at 25°C, animals were washed off the plates and resuspended in TRI Reagent as described above.
Total RNA was isolated with TRI Reagent according to the manufacturer’s instructions (Sigma-Aldrich, T9424). To ensure the highest purity of the RNA samples isolated from worms, the subsequent phenol/chloroform extraction has been performed according to standard protocols. Before RT-qPCR, RNA-seq and DRS library preparation, RNase H treatment, and PAT total RNA was treated with TURBO DNase (Thermo Fisher Scientific). RNA was then purified by phenol/chloroform extraction and ethanol precipitation.
RNA-seq and data analysis
cultures and RNA extraction are described above. Three independent replicate sample sets were prepared for each strain [wild type and tent-5(tm3504)
] and condition (worms that were grown on E. coli
HB101 or infected by S. aureus
for 8 hours). Two micrograms (worms) or 1 μg (wild-type BMDMs stimulated with LPS) of DNase-treated total RNA was used for the library preparation. Ribosomal RNA (rRNA) was removed using a Ribo-Zero rRNA removal kit (human/mouse/rat; Epicentre, RZG1224). Sequencing libraries were prepared using a KAPA stranded RNA-seq library preparation kit (KAPA Biosystems, KR0934), and their quality was assessed with an Agilent 2100 Bioanalyzer (Agilent Technologies Inc.). The libraries were sequenced in the 75-nt single-end (C. elegans
samples) or 75-nt pair-end (BMDMs) mode on the NextSeq500 Illumina platform. RNA-seq reads were adapter-clipped and quality-filtered with cutadapt (version 1.18) to remove adapters, low-quality fragments (minimum quality score was set to 20), and too short sequences (threshold set to 30 nt) (105
). Quality-filtered reads were mapped to the respective reference genomes of C. elegans
(WBCel235; ENSEMBL, release 94) or mouse (GRCm38; ENSEMBL, release 94) using the STAR (Spliced Transcripts Alignment to a Reference) aligner (version 2.6.1b or version 2.7.6a for worm and mouse, respectively) (106
). Read counts were assigned to genes using featureCounts from the Subread package (version 1.6.3) with options -Q 10 -p -B -C -s 2 -g gene_id -t exon and respective annotation files for C. elegans
(WBCel235; ENSEMBL, release 94) or mouse (Gencode vM25) (107
). Multimappers and reads overlapping multiple features were not counted. Differential expression analysis was performed with DESeq2 (version 1.22) Bioconductor package (108
) with default settings. For C. elegans
, most of the analyses were performed for the genes, which expression was down-regulated at least 1.5-fold [log2
fold change < −log2
(1.5), FDR < 0.05] in mutant worms. Venn diagrams were drawn with VENNY (www.stefanjol.nl/venny
). Gene sets were submitted for GO enrichment analysis to the WormBase Enrichment Suite (WS278) (109
) and WormCat tool (110
DRS and data analysis
For tethering assay, technical replicate sample sets were prepared from 293T cells transfected with the wild-type or catalytically inactive NHA–TENT-5. For C. elegans
DRS, two independent replicate sample sets were prepared for tent-5(tm3504)
and wild-type worms [the same input RNA samples as for the RNA-seq experiment (replicates 1 and 2) were used]. For mice samples, BMDMs from Tent5aFlox/Flox Tent5c−/−
and wild-type mice were isolated, cultured, and transduced as described above. On the 14th day after isolation, cells were stimulated with LPS (100 ng/ml) for 8 hours. Two replicate sample sets were prepared from BMDMs for DRS analysis. Total RNA from 293T, C. elegans
, and BMDM samples was isolated with TRI Reagent according to the manufacturer’s instructions (Sigma-Aldrich, T9424). The cap-enriched mRNA was prepared from 100 μg of total RNA with GST-eIF4EK119A
protein (homemade) and glutathione sepharose 4B (GE Healthcare, 17-0756-01) as described previously (23
). Nanopore direct RNA libraries were prepared with a DRS Kit [Oxford Nanopore Technologies (ONT), SQK-RNA002] from 3 μg (worms and 293T) and 3.5 μg (BMDMs) of cap-enriched mRNA mixed with 150 ng of Saccharomyces cerevisiae
oligothymidilate [oligo(dT)]–enriched mRNA to optimize sequencing efficiency. Sequencing was performed with a MinION device (ONT; Flow cell type FLO-MIN106, RevD). Raw reads were basecalled with the standalone version of Guppy 4.0.11 (ONT). Sequencing reads were mapped to Gencode v36 supplemented with sequences of reporter transcripts (293T), WBCel235 (worms), or Gencode vM26 (BMDMs) reference transcriptomes using MiniMap 2.17 (111
) with options -k 14 -ax map-ont –secondary = no and processed with samtools 1.9 to filter out supplementary alignments and reads mapping to reverse strand (samtools view -b -F 2320). The poly(A) tail lengths were estimated with Nanopolish (version 0.13.2) polya function (112
). In subsequent analyses, only length estimates with quality control tag reported by Nanopolish as PASS were considered. Statistical analysis was performed using functions provided in the NanoTail R package (23
). Poly(A) length distributions in analyzed conditions were compared using the Wilcoxon test, filtering out transcripts that had a low number of supporting reads under each condition (<10). Collected P
values were adjusted for multiple comparisons using the Benjamini-Hochberg method. Transcripts were considered as having a significant change in poly(A) tail length if the adjusted P
value was <0.05. Transcripts were considered as TENT-5 or TENT5A/C substrates if, in addition to being significantly changed, their median poly(A) tail length was at least 5 nt shorter in the mutant worms or double Tent5aFlox/Flox Tent5c−/−
mutant compared to wild-type worms or mouse BMDMs, respectively. For differential expression estimates, reads were mapped to C. elegans
(WBCel235; ENSEMBL, release 94) or mouse (GRCm38; ENSEMBL, release 94) reference genomes using MiniMap 2.17 (111
), with options -ax splice –secondary = no -uf. Read counts were assigned to genes using featureCounts from the Subread package (version 2.0.1) with options -L –fracOverlap 0.5 –fracOverlapFeature 0.2 -s 1 and respective annotation files for C. elegans
(WBCel235; ENSEMBL, release 94) or mouse (Gencode vM25) (107
). Multimappers and reads overlapping multiple features were not counted. Differential expression analysis was performed with DESeq2 (version 1.28) Bioconductor package (108
) with default settings. Gene sets were submitted for GO enrichment analysis to the WormBase Enrichment Suite (WS278) (109
) or g:Profiler (113
Transcript and UTR length analysis
Data regarding coordinates of the 5′ and 3′UTRs in the WBCel235 genome and percent guanine-cytosine content for each gene were downloaded from ParaSite BioMart (WS276) (https://parasite.wormbase.org/biomart/martview/
). Data regarding transcript and coding sequences lengths were obtained from BioMart using biomaRt R package (biomart = “ensembl” and dataset = “celegans_gene_ensembl”). For each gene, only the longest possible 5′ and 3′UTRs or coding sequence was considered for the analysis. The lengths of the 5′ and 3′UTRs were calculated on the basis of obtained coordinates. Statistics were calculated using the Wilcoxon test.
Motif enrichment analysis
Data regarding coordinates of the 3′UTRs in the WBCel235 genome were downloaded from ParaSite BioMart (WS276) (https://parasite.wormbase.org/biomart/martview/
). Only the longest possible 3′UTR sequence for each gene was considered for the analysis. Coordinates of 3′UTRs of TENT-5 substrates (DRS) and genes that expression levels were down-regulated at least 1.5-fold in tent-5(tm3505)
mutant worms compared to wild type (RNA-seq), as well as coordinates of 3′UTRs of all remaining genes identified by DRS or RNA-seq (background), were saved as bed file and were used for respective FASTA sequences collection using the bedtools getfasta tool (version 2.29.2) (114
) and WBCel235 (ENSEMBL, release 94) genome sequence. FASTA sequences of 3′UTRs of TENT5A/C substrates and all Gencode-annotated transcripts in mm10 genome (background) were obtained with bedtools getfasta tool (version 2.29.2) (114
), using bed files with 3′UTR coordinates downloaded from the UCSC Browser Table tool (Gencode vM23 track and known_gene table) and GRCm38 genome sequence. Sequence motifs were searched using the DREME tool (version 5.3.0) (59
) with options -rna -norc -k 8 -l with the respective background (described above) specified.
Reverse transcription quantitative polymerase chain reaction
One microgram of the DNase-treated total RNA was reverse-transcribed with 1 μl of oligo(dT)25 and random primers mix (50 mM and 50 ng/μl, respectively) using the SuperScript III Reverse Transcriptase (Thermo Fisher Scientific, 18080085). cDNA samples were diluted 10× and used for RT-qPCR analysis using the LightCycler 480 SYBR Green I Master Mix (Roche, 04887352001) and 0.25 μM primers on the LightCycler 480 Instrument (Roche). Primers were designed with Primer-BLAST (National Center for Biotechnology Information) to be exon-junction spanning where possible and tested for amplification efficiencies with a series of template dilutions. Each experimental replicate was measured in technical triplicate. Expression levels for each sample were normalized to act-1. Gene expression changes were calculated using the 2−ΔΔC(t) method. Unpaired two-sample two-sided t test using ΔCt values were performed for most comparisons except for induction of gene expression during infection, where one-sided t tests were performed. A P < 0.05 was considered statistically significant: ns, not significant; *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001. Primer sequences are listed in table S2.
RNase H treatment
Twenty micrograms of DNase-treated total RNA was mixed with 2 μl of 50 mM oligo(dT)25, 1 μl 10× hybridization buffer [25 mM tris-HCl (pH 7.5), 1 mM EDTA, and 50 mM NaCl], and water in 10 μl. RNA was denatured at 70°C for 10 min and slowly cooled down to 42°C. Next, 10 μl of prewarmed to 37°C 2× reaction buffer [40 mM tris-HCl (pH 7.5), 20 mM MgCl2, 200 mM KCl, 2 mM DTT, and 10% sucrose] and 1 μl of RNase H (2 U; Thermo Fisher Scientific, EN0201) were added, and the reactions were carried out at 37°C for 1 hour. RNA was recovered with phenol/chloroform extraction, precipitated with 96% ethanol, and analyzed using Northern blots or PAT.
Poly(A) tail analysis
One microgram of DNA-free total RNA was ligated with 125 pmol of RA3_15N 3′ adaptor at 18°C for 16 hours in 20-μl mixtures containing 1× T4 RNA ligase buffer, 10% PEG 8000 (polyethylene glycol, molecular weight 8000), 50 U of RiboLock RNase inhibitor, and 300 U of T4 RNA ligase 2 truncated KQ [New England Biolabs (NEB), M0373L]. The samples were purified with AMPure XP magnetic beads [1:0.75 (v/v) RNA-to-beads ratio] to discard nonligated adapters and short RNA fragments. RNA was eluted in 15 μl of RNase-free water and reverse-transcribed using 200 U of the SuperScript III Reverse Transcriptase (Thermo Fisher Scientific) and 100 pmol of the RPI PCR Index Primer (TruSeq Illumina) according to the manufacturer’s protocol. The cDNAs were purified using AMPure XP beads [1:1 (v/v) ratio], eluted with 20 μl of RNase-free water, and used for the nested PCR. Briefly, 1 μl of the cDNA was used for PCR-1 (Phusion Hot Start II; 25 cycles), with a gene-specific forward primer and a universal reverse primer RPuni. The PCR-1 samples were diluted 100× and used as a template for PCR-2, with a second forward gene-specific primer and RPuni. The PCR-1 and PCR-2 amplicons were analyzed in the 2% agarose gels in 1× TBE buffer (90 mM tris-borate and 2 mM EDTA). All primers used for PAT are described in table S2.
High-molecular weight RNA samples were separated on 1.2% agarose gels containing 1.7% formaldehyde in 1× NBC buffer (50 mM boric acid, 1 mM sodium acetate, and 5 mM NaOH) and transferred to Hybond-N+ membranes (GE Healthcare) by overnight capillary elution using 8× SSC buffer (1.2 M NaCl and 120 mM sodium citrate). Low–molecular weight RNA samples were separated on 4 to 6% acrylamide gels containing 7 M urea in 0.5× TBE buffer (45 mM tris-borate and 1 mM EDTA) and electrotransferred to membranes in 0.5× TBE buffer at 300 to 350 mA at 4°C for 3 hours. RNA was immobilized on membranes by 254-nm UV light using a CL-1000 cross-linker (UVP) with the auto cross-link function (120 mJ/cm2). Next, membranes were stained with 0.03% methylene blue in 0.3 M NaAc (pH 5.3), and staining was digitized. Random primed RL probes were PCR-amplified with primers RL_Fw and RL_Rev using pRL-5Box plasmid as a template and radioactively labeled with 20 μCi of [α-32P]-dATP and the DECAprime II DNA Labeling Kit (Thermo Fisher Scientific, AM1456). Membranes were prehybridized in the PerfectHyb Plus Hybridization Buffer (Sigma-Aldrich) at 65°C for 30 min and incubated with probes in PerfectHyb buffer at 65°C overnight with rotation. Membranes were washed three times in prewarmed 0.5× SSC with 0.1% SDS at 65°C for 20 min and then exposed overnight to PhosphorImager screens (Fujifilm). The screens were scanned with a Typhoon FLA 7000 scanner (GE Healthcare) and analyzed with Multi Gauge software version 2.0 (Fujifilm).
Subcellular protein fractionation
Mixed-stage worm populations were grown on NGM plates seeded with E. coli HB101 at 20°C. Two independent replicate sample sets were prepared for each strain. Subcellular protein fractionation was performed using the Subcellular Protein Fractionation Kit (Thermo Fisher Scientific, 78840). Briefly, all buffers were supplemented with protease inhibitors (Invitrogen), and the entire procedure was performed at 4°C. For each strain, worms were washed from three 100-mm plates and washed three times with 50 mM NaCl (900g for 2 min at room temperature). Worms were resuspended in ice-cold 1.5 ml of cytoplasmic extraction buffer and lysed using Omni tissue homogenizer for 1 min following 10 min of incubation with gentle mixing. Lysates were centrifuged at 500g, and the supernatant was collected as a cytoplasmic fraction, while the pellet was resuspended in membrane extraction buffer. After 10 min of incubation, the sample was centrifuged at 3000g, and the supernatant was collected as a membrane fraction. Last, the pellet was resuspended in nuclear extraction buffer, incubated for 30 min, and centrifuged at 5000g. The supernatant was collected as a nuclear fraction. The concentration of protein was measured by Bradford assay, and samples for SDS-PAGE and Western blot were prepared using the same amount of protein from each fraction. Control total protein input samples were prepared by boiling worms in 1× SDS sample buffer.
Total worm extract preparation for Superdex 200 chromatography
Protein extracts were prepared from mixed-stage worm populations grown at 20°C. Animals were washed three times with 50 mM NaCl and then briefly with lysis buffer [50 mM Hepes (pH 7.4), 100 mM NaCl, 3 mM MgCl2, 0.5 mM DTT, 0.05% NP-40 substitute, and 10% (v/v) glycerol]. The supernatant was discarded, and the pellet was resuspended in 5 V of lysis buffer supplemented with 1 mM phenylmethylsulfonyl fluoride (PMSF), 1× chymostatin, and 1× protease inhibitors and drop-frozen in liquid nitrogen. Lysis was performed by grinding frozen worms in liquid nitrogen. The extract was allowed to melt on ice and supplemented with 1 mM PMSF, 1× chymostatin, and 1×protease inhibitors. The lysate was cleared by two centrifugation steps at 20,000g at 4°C for 20 min. Five hundred microliters of the lysate (7 μg/μl) was subjected to the Superdex 200 chromatography column equilibrated with lysis buffer. Two hundred microliters of each collected fraction was precipitated using methanol/chloroform; pellets were resuspended in 20 μl of 1× SDS sample buffer, boiled, and used for the SDS-PAGE and Western blot.
BMDMs from Tent5aFlox/Flox Tent5c−/−
-3xFLAG, and wild-type mice were isolated, cultured, and transduced as described above. On the 13th day after isolation, cells were counted and seeded on six-well plates, with 0.5 million cells per well. The next day, cells were stimulated with LPS (100 ng/ml) for 8 hours (Tent5aFlox/Flox Tent5c−/−
and wild type) or for 0 to 16 hours of time points (Tent5a
-3xFLAG, and wild type). BMDMs were scratched and pelleted by centrifugation for 3 min at 350g
. Cells were lysed with 0.1% NP-40 in phosphate-buffered saline (PBS) supplemented with protease inhibitors and viscolase (final concentration of 0.1 U/ml; A&A Biotechnology). The samples were incubated at 37°C for 30 min with shaking before 3× SDS sample buffer [187.5 mM tris-HCl (pH 6.8), 6% SDS, 150 mM DTT, 0.02% bromophenol blue, 30% glycerol, and 3% 2-mercaptoethanol] was added, and the lysates were boiled for 10 min. Lysates from the 293T cells following tethering assay were prepared using the same protocol. Protein samples from a mixed population of worms were prepared by boiling ~100 worms in 3× SDS sample buffer for 5 min. Boiled protein mixtures from cells or worms were cleared by centrifugation for 5 min at maximum speed at room temperature and resolved on 10 to 12% SDS-PAGE gels. Proteins were transferred to Protran nitrocellulose membranes (GE Healthcare) by wet transfer at 300 mA at 4°C for 2 hours in 1× transfer buffer [25 mM tris base, 192 mM glycine, and 20% methanol (v/v)]. After transfer, the membranes were stained with 0.3% (w/v) Ponceau S in 3% (v/v) acetic acid, and the staining was digitized. Next, membranes were soaked in 5% (w/v) nonfat milk in 1× TBS-T (20 mM tris base, 150 mM NaCl, and 0.01% Tween 20) for 1 hour with gentle agitation at room temperature, followed by the overnight incubation at 4°C with specific primary antibodies diluted 1:3000 (anti-RL
antibody, clone 5B11.2; Millipore, MAB4400), 1:3000 (GFP B-2; Santa Cruz Biotechnology, sc-9996), 1:1000 (FLAG; Proteintech, 20543-1-AP), 1:10,000 RFP (red fluorescent protein); Erdogan, AB233), 1:5000 [glyceraldehyde phosphate dehydrogenase (GAPDH); Proteintech, 10494-1-AP], 1:10,000 (α-tubulin, clone DM1A; Millipore, MABT205), 1:1000 (iNOS; Cell Signaling Technology, 13120), 1:1000 (CD80; Cell Signaling Technology, 54521), 1:30,000 (lysozyme; Abcam, ab108508), 1:30,000 (cathepsin S; Invitrogen, MA5-29695), 1:10,000 (cathepsin B; Abcam, ab214428), and 1:20,000 (cathepsin D; Abcam, ab75852). Membranes were washed three times for 10 min each in 1× TBS-T and then incubated for 2 hours with gentle agitation at room temperature with horseradish peroxidase–conjugated secondary anti-mouse (Millipore, 401215) or anti-rabbit (Millipore, 401393) antibodies diluted 1:5000. Following three washes in 1× TBS-T, blots were incubated with the Clarity Western ECL Substrate (Bio-Rad) for 1 to 3 min, and signals were detected either through exposure to a CL-Exposure film (Thermo Fisher Scientific) and developed in an AGFA Curix CP-1000 device or visualized using the ChemiDoc Imaging System (Bio-Rad). Protein bands from Western blots were quantified with ImageJ as described in www.yorku.ca/yisheng/Internal/Protocols/ImageJ.pdf
. Final relative quantification values represent the ratio of net band intensity from the protein of interest to net GAPDH (loading control).
Protein extracts were prepared in eight replicate sample sets. The tent-5(tm3504)
mutant and wild-type worms were grown at 20°C on E. coli
HB101 until they reached the L4 stage. Worms were washed three times with 50 mM NaCl and once in 1× lysis buffer [50 mM Hepes (pH 7.4), 100 mM NaCl, 3 mM MgCl2
, 0.5 mM DTT, 0.05% NP-40 substitute, and 10% (v/v) glycerol]. Pellets were resuspended in 900 μl of 1× lysis buffer supplemented with 1 mM PMSF, 1× chymostatin, and 1× protease inhibitors; transferred to 2-ml tubes containing 200 μl of Zirconia beads (BioSpec Products); and drop-frozen in liquid nitrogen. Next, tubes were inserted into the Fast Prep-24 machine (MP Biomedicals), and worms were crushed for 1 min at maximum speed. Following centrifugation at 14,000g
for 10 min at 4°C, lysates were transferred to the new tubes and subjected to sonication at high amplitude for 20 min (30-s on/30-s off cycle) (Diagenode Bioruptor XL) and then cleared by centrifugation at 14,000g
for 30 min at 4°C. A Millipore Direct Detect infrared spectrometer was used to determine the total protein concentration of the lysate. Sample preparation was done on the basis of modified FASP (Filter-aided sample preparation) protocol (115
). Briefly, a supernatant was placed at Vivacon 30-kDa filter (Sartorius), centrifuged, and washed three times with 200 μl of 8 M urea in 100 mM NH4
. Next, samples were reduced (DTT at room temperature for 30 min) and alkylated (indole-3-acetic acid at room temperature for 15 min) following overnight digestion with trypsin (Promega) and acidified with trifluoroacetic acid to a final concertation of 0.1%. Mass spectrometry (MS) analysis was performed by liquid chromatography–MS in the Laboratory of Mass Spectrometry (Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warsaw) using a nanoACQUITY UPLC system (Waters, 176016000) coupled to an LTQ-Orbitrap Velos mass spectrometer (Thermo Fisher Scientific). Peptides were separated by a 180-min linear gradient of 95% solution A (0.1% formic acid in water) to 35% solution B (acetonitrile and 0.1% formic acid). The measurement of each sample was preceded by three washing runs to avoid cross-contamination; the final MS washing run was searched for the presence of cross-contamination between samples. If the protein of interest was identified in the washing run and the next measured sample at the same or smaller intensity, then the sample was regarded as contaminated and excluded from the final graphs. The mass spectrometer was operated in the data-dependent MS-MS2 mode, and data were acquired in the mass/charge ratio range of 300 to 2000. MS raw data files were used to calculate protein abundance in the samples using the MaxQuant (version 22.214.171.124) platform (116
). The reference proteome of C. elegans
database from UniProt (27,805 protein entries) and common contaminates list included in MaxQuant were used, and analysis was performed with the following settings: match between runs, variable modification: oxidation (M), and 4.5 parts per million of error tolerance. Label-free quantification (LFQ) intensity values were calculated using the MaxLFQ algorithm to estimate quantities of identified proteins. Protein abundance was defined as the LFQ value calculated by MaxQuant software for a protein (sum of intensities of identified peptides of a given protein) divided by its molecular weight. The Scafold4 Q + S platform was used for statistical analysis. Protein abundance in analyzed samples was compared using the Mann-Whitney test with Benjamini-Hochberg correction.
Life-span and killing assays
The 35-mm plates with TSA + Nal (final concentration of 10 μg/ml) for killing assays on S. aureus
NCTC8325 were prepared 1 week before the experiments and stored at 4°C protected from light (37
). Survival analysis on S. marcescens
Db10, P. luminescens
, and respective control E. coli
OP50 were performed on the 35-mm plates with NGM without the addition of drugs or antibiotics. Life-span assays on E. coli
HB101, P. aeruginosa
PAO1, and some assays on E. coli
OP50 were performed with the addition of FUdR (5-Fluoro-2′-deoxyuridine) (final concentration of 0.1 mg/ml; Sigma-Aldrich, F0503) to prevent progeny production. All bacteria strains were freshly seeded from the −80°C stock 3 days before an experiment to the appropriate solid medium. E. coli
OP50 was cultured overnight in LB, and 50 μl of overnight culture was spread onto the center of 35-mm NGM plates and incubated at 37°C for 16 hours. E. coli
HB101 was cultured overnight in LB + streptomycin (0.1 mg/ml). Fifty microliters of overnight culture was spread onto the center of 35-mm NGM or NGM + FUdR plates (depending on the experiment setup) and incubated at 37°C for 16 hours. For survival analysis on UV-killed E. coli
HB101, NGM + FUdR plates seeded with bacteria were exposed to UV light in a UV Stratalinker 2400 for 30 min at maximum. For survival analysis on heat-killed bacteria, E. coli
HB101 and OP50 were cultured overnight in 50 ml of LB and centrifuged at 7000g
for 10 min at room temperature, and pellets were resuspended in 5 ml of fresh LB. Next, to kill bacteria, mixtures were incubated in the water bath at 65°C for 30 min. Fifty microliters of culture containing dead bacteria was spread onto the center of 35-mm plates and incubated at 25°C for 24 hours. Bacterial killing was evaluated by inoculating LB medium with UV- or heat-treated bacteria, and lack of growth at 37°C confirmed effective killing. S. aureus
was grown overnight in TSB + Nal (10 μg/ml). Ten microliters of overnight culture was spread onto the center of 35-mm TSA + Nal plates and incubated at 37°C for 6 hours and then cooled down to 25°C and used for the killing assays. P. aeruginosa
and S. marcescens
were grown overnight at 37°C in LB. Ten microliters of overnight cultures was spread onto the center of 35-mm NGM or NGM + FUdR plates (for S. marcescens
and P. aeruginosa
, respectively) and incubated at 37°C for 24 hours and then at 25°C for another 24 hours. P. luminescens
was cultured overnight at 30°C in LB, and 10 μl of the overnight culture was spread onto the center of 35-mm NGM plates and incubated at 30°C for 24 hours and after that at 25°C for 24 hours. All worms were grown on NGM + E. coli
HB101 at 20° or 25°C (depending on the temperature in which the assays were performed) for three to four generations before the experiments. Animal populations were synchronized by bleaching of the gravid adults and the starvation of L1 larvae for 16 hours at appropriate temperatures. Forty to 60 L4-staged worms were transferred to each of the three replicate assay plates per strain. Beginning on the next day, the number of dead and live worms on each plate was recorded daily (S. marcescens
, P. aeruginosa
, P. luminescens
, and E. coli
) or twice a day (S. aureus
). Live worms were transferred daily to new plates to avoid contamination with the progeny (S. marcescens
, P. luminescens
, and related E. coli
OP50 control). For P. aeruginosa
tests, worms were not transferred to the new plates. For E. coli
HB101 life spans performed on the NGM + FUdR plates, worms were transferred to the new plates every 3 days until day 12 and then left on the same plates. Worms that left the plates in the first several days of the assay were removed from the counts of subsequent days. Animals were considered dead if they failed to respond to a gentle touch. For each survival experiment, at least two biological replicates were carried out. Kaplan-Meier survival analyses were performed using GraphPad Prism 7 software. Life-span survival data were compared using the log-rank significance test and presented as median survival. A P
< 0.05 was considered significantly different from control: ns, not significant; *P
< 0.05, **P
< 0.01, ***P
< 0.001, and ****P
Exposure of wild-type and mutant worms to E. coli
HB101 and P. aeruginosa
PAO1 was carried out exactly as for life-span and killing assays described above. CFU assays were performed essentially as described in (49
) with minor modifications. Briefly, on the 5th (HB101) and 4th (PAO1) day of adulthood, 30 worms of each genotype from each of the three technical replicate plates were collected into 50 μl of 1× M9 supplemented with 25 mM levamisole (Sigma-Aldrich, L9756) to inhibit pharyngeal pumping and expulsion. Worms were washed in 1× M9 + 25 mM levamisole three times and then surface-sterilized in 1× M9 + 25 mM levamisole + kanamycin (100 μg/ml) for 45 min at room temperature. Following three washes with 1× M9 + 25 mM levamisole, worms were resuspended in 150 μl of PBS containing 0.1% Triton X-100. A 100-μl aliquot of the supernatant was removed from each replicate to test for external bacterial contamination. Animals were homogenized in the remaining 50 μl of PBS + 0.1% Triton X-100 with pellet pestle (Bel-Art, BAF199230001) and motor for 30 s, and then 450 μl of PBS was added to each sample. Dilution series of homogenates was spread to the LB plates without antibiotics and grown at 37°C for 24 hours. CFU value per worm was counted as follows: number of CFU/worm = (number of colonies × dilution factor)/number of worms in lysate − external CFU. For each CFU experiment, three biological replicates were carried out, each comprising at least two technical replicates. A P
< 0.05 was considered significantly different from control: ns, not significant; *P
< 0.05 and **P
Statistical analysis was performed in Microsoft Excel (RT-qPCR), GraphPad Prism 7 (life span assays), or with R 4.0 (117
). Details of the particular statistical analyses, significance, number of replicates and sample sizes, and the features of all plots are described in the figure legends. Data plotted as box plots have the following features: whiskers (25th and 75th percentiles), minima and maxima (5th and 95th percentiles), and thick lines (median). Data presented as heatmaps were normalized to a sequencing depth using DESeq2 and transformed with regularized log transformation for visualization purposes.
We thank M. Krzyszton for helpful comments and the Dziembowski laboratory members for discussions. We are grateful to B. Goldstein, W. Filipowicz, D. Trono, and J. Jaworski for sharing plasmids; G. Jagura-Burdzy for sharing P. aeruginosa PAO1; D. Adamska for assistance with RNA-seq; O. Gewartowska, M. Szpila, E. Borsuk, and J. Gruchota for mice lines generation; and M. Hyjek-Składanowska for help with S200 chromatography. Some C. elegans strains were provided by the CGC, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440).
Funding: This work was supported by National Science Center (OPUS 17 UMO-2019/33/B/NZ2/01773 to A.D., OPUS 14 UMO-2017/27/B/NZ2/01234 to S.M., and PRELUDIUM 19 UMO-2020/37/N/NZ2/02893 to A.B.). This research was supported by the funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no 810425. A.B. was also supported by the Foundation for Polish Science (FNP).
Author contributions: A.D. acquired funding and directed the studies. A.D., V.L., S.M., and A.B. designed the experiments. V.L. performed all C. elegans experiments and analyzed initial RNA-seq data. A.B. established primary BMDM cell cultures and performed all subsequent BMDM experiments. P.S.K. analyzed RNA-seq and C. elegans DRS data. S.M. performed tethering assays and preliminary BMDM experiments. N.G. analyzed BMDM DRS data. T.W. performed statistical colocalization analysis. D.C. analyzed MS data. Z.M. participated in C. elegans life span, CFU analysis, and transgenic strains generation. K.D. and J.J.E. contributed resources and helped to design experiments. V.L. and A.D. wrote the manuscript with input from P.S.K., A.B., and J.J.E.
Competing interests: The authors declare that they have no competing interests.
Data and materials availability:
All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. RNA-seq data have been deposited to GEO database under the accession number GSE163549. Nanopore DRS data have been deposited to ENA with the following accession numbers: C. elegans
and BMDM DRS, PRJEB40892; tethering experiment DRS, ERS12230818, ERS12230819, ERS12230820, and ERS6477295. Proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (118
) partner repository with identifier PXD023238. All plasmids and C. elegans
strains generated in this study are available upon request from A.D. The mouse lines can be provided by A.D.’s pending scientific review and a completed material transfer agreement. Requests for the mice lines should be submitted to A.D.