Intelectin 3 is dispensable for resistance against a mycobacterial infection in zebrafish (Danio rerio)

1 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 www.nature.com/scientificreports Intelectin 3 is dispensable for resistance against a mycobacterial infection in zebrafish (Danio rerio) Markus J. T. Ojanen1,2, Meri I. E. Uusi-Mäkelä1, Sanna-Kaisa E. Harjula1, Anni K. saralahti1, Kaisa E. Oksanen1, Niklas Kähkönen3, Juha A. E. Määttä3, Vesa P. Hytönen 3, Marko Pesu 2,4 & Mika Rämet1,5,6,7 Tuberculosis is a multifactorial bacterial disease, which can be modeled in the zebrafish (Danio rerio). Abdominal cavity infection with Mycobacterium marinum, a close relative of Mycobacterium tuberculosis, leads to a granulomatous disease in adult zebrafish, which replicates the different phases of human tuberculosis, including primary infection, latency and spontaneous reactivation. Here, we have carried out a transcriptional analysis of zebrafish challenged with low-dose of M. marinum, and identified intelectin 3 (itln3) among the highly up-regulated genes. In order to clarify the in vivo significance of Itln3 in immunity, we created nonsense itln3 mutant zebrafish by CRISPR/Cas9 mutagenesis and analyzed the outcome of M. marinum infection in both zebrafish embryos and adult fish. The lack of functional itln3 did not affect survival or the mycobacterial burden in the zebrafish. Furthermore, embryonic survival was not affected when another mycobacterial challenge responsive intelectin, itln1, was silenced using morpholinos either in the WT or itln3 mutant fish. In addition, M. marinum infection in dexamethasone-treated adult zebrafish, which have lowered lymphocyte counts, resulted in similar bacterial burden in both WT fish and homozygous itln3 mutants. Collectively, although itln3 expression is induced upon M. marinum infection in zebrafish, it is dispensable for protective mycobacterial immune response. Tuberculosis is an epidemic multifactorial disease caused by Mycobacterium tuberculosis1. The susceptibility to tuberculosis depends on the M. tuberculosis strain and on a number of host-related factors such as environmental conditions, other underlying diseases as well as genetic variation2–4. Critical genes of the adaptive immunity required for the mycobacterial immune response such as interferon gamma (IFNG)5,6 and interleukin 12 (IL12)7,8 were identified already in the 1980’s and 1990’s, respectively. The importance of these genes has later been verified in human tuberculosis patients9 and by using experimental gene knockout mouse models of tuberculosis10–13. More recently, pattern recognition receptor (PRR) gene polymorphisms of Toll-like receptors (TLRs)14–16 and C-type lectins17,18, have been associated with M. tuberculosis susceptibility, delineating also the central role of the innate immunity in controlling the mycobacterial infection. Lectins are carbohydrate-binding proteins important for numerous biological processes such as intracellular glycoprotein secretion, leukocyte trafficking and microbial recognition19,20. Consequently, lectins act as recognition molecules inside cells, on the cell surface and in extracellular fluids20. Intelectins (ITLNs) are a distinct family of lectins, which were first identified in Xenopus laevis21 and were later found in a number of chordates including human, mouse and zebrafish (Danio rerio)22–25. Although ITLN function has been linked to a number of processes such as iron absorption26, metabolic disorders27 as well as cancer development28,29, their exact biological functions are elusive. Suggesting a role for ITLNs in the immune response, itln gene expression is highly up-regulated upon a bacterial infection in fish25,30–32. Moreover, human ITLN1 (also known as Omentin) has been 1Laboratory of experimental immunology, BioMeditech institute and faculty of Medicine and Life Sciences, University of tampere, tampere, finland. 2Laboratory of immunoregulation, BioMeditech institute and faculty of Medicine and Life Sciences, University of tampere, tampere, finland. 3Laboratory of Protein Dynamics, BioMeditech institute and faculty of Medicine and Life Sciences, University of tampere, tampere, finland. 4Department of Dermatology, tampere University Hospital, tampere, finland. 5Department of Pediatrics, tampere University Hospital, tampere, finland. 6Department of children and Adolescents, Oulu University Hospital, Oulu, finland. 7PeDeGO Research Unit and Medical Research center Oulu, University of Oulu, Oulu, finland. correspondence and requests for materials should be addressed to M.R. (email: [email protected]) Received: 9 October 2018 Accepted: 7 December 2018 Published: xx xx xxxx OPEN www.nature.com/scientificreports/ 2 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 shown to bind to the Mycobacterium bovis Bacillus Calmette-Guérin (BCG)33, and more specifically to exocyclic 1,2-diol glycan epitopes that are expressed selectively on microbial surfaces34. The importance of ITLNs for immunity in vivo, however, is less clear. Previously, Voehringer et al., used transgenic mice with lung-specific ITLN1 and ITLN2 over-expression to study the effects of these proteins in the mouse infection models of the parasite Nippostrongylus brasiliensis and the M. tuberculosis bacterium35. In these settings, the authors could not detect enhanced pathogen clearance in the Itln transgenic mice. In contrast, a so called “natural deletion” of the Itln2 gene has been previously associated with a higher susceptibility against the parasite Trichinella spiralis in a C57BL/10 mouse strain36. Recently, an Itln1 knockout mouse strain was created to study inflammatory bone diseases37. In the aforementioned study, the lack of Itln1 was associated with a proinflammatory phenotype characterized by elevated TNF and IL6 levels in bone tissue and in serum, and was shown to result in osteoporosis37. The genome of the zebrafish was assembled for the first time in 2002 and the prevalent 11th assembly (GRCz11) is an invaluable tool for research using zebrafish as a disease model38. Over 70% of human genes have at least one zebrafish orthologue and for this reason, the zebrafish immune system is highly similar compared to humans38. In fact, most of the human immune cell populations such as Tand B-cells39–41, neutrophils and macrophages42, dendritic cells43 as well as the complement system44 and immunoglobulins45,46, are found in the zebrafish. Importantly, zebrafish can be modified genetically with the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated 9 (Cas9) mutagenesis47,48, which allows disease modeling using reverse genetics, although some genes appear difficult to target successfully49. A Mycobacterium marinum infection of zebrafish is nowadays a commonly used model for studying tuberculosis in both larvae and adult fish50,51. Compared to several other tuberculosis models, the mycobacterial model of zebrafish is considered safe, cost-effective and ethical52,53. More importantly, M. marinum is closely related to M. tuberculosis, and the two bacterial species have comparable pathogenic characteristics in the natural hosts; macrophage mediated intracellular multiplication as well as the formation of granuloma structures54–56. The larval model enables studying specifically the innate immunity57,58, whereas the adult zebrafish model allows studying also components of the adaptive immune system in both an acute mycobacterial infection59 as well as during mycobacterial latency56,60. In order to identify candidate genes associated with the host response against mycobacteria, we conducted a gene expression microarray in M. marinum infected adult zebrafish. Here, we identified a zebrafish ITLN orthologue itln3 among the genes that were most induced upon infection. In order to gain more insights into the function of ITLNs, we used CRISPR/Cas9 mutagenesis to create knockout itln3 mutant zebrafish lines, and used the zebrafish M. marinum infection model to determine the in vivo significance of Itln3 in a mycobacterial infection. Results Genome-wide gene expression microarray analysis of M. marinum infected adult zebrafish. In order to identify genes involved in the host immune response against mycobacteria, we used the zebrafish M. marinum infection model and conducted a genome-wide gene expression analysis using the microarray platform. To this end, we infected wild-type (WT) AB zebrafish with M. marinum (20 CFU; SD 6 CFU) and isolated their organ blocks (includes all the organs of the abdominal cavity) for a transcriptomic analysis at 14 days post infection (dpi). From a total of 43603 probes used in the analysis, we found 93 probes, corresponding to 70 genes, that were up-regulated and 26 probes, corresponding to 21 genes, that were down-regulated (log2 fold change > 3 ) compared to the mock-treated (PBS) controls (Supplementary Table1). Further evaluation of the up-regulated probes with a GOrilla gene ontology (GO) enrichment analysis61,62 revealed 22 enriched (p < 0.001) processes including response to carbohydrates (GO:0009743), cholesterol homeostasis (GO:0042632) and antigen processing and presentation (GO:0019882) (Supplementary Table2). Among the up-regulated genes we found five genes; si:busm1-194e12.11 (mhc2 family gene), arachidonate 5-lipoxygenase b, tandem duplicate 3 (alox5b.3), zgc:113912 (mhc2 family gene), CD59 molecule (cd59) and si:busm1-194e12.12 (mhc2 family gene) with well-known immunological functions in antigen processing, inflammation and in the regulation of the complement system (Fig.1A, Supplementary Table1). Of the 21 down-regulated genes, five were associated with the immune response; CD58 molecule (cd58), myeloid-specific peroxidase (mpx), complement factor b-like (cfbl), immunoresponsive gene 1, like (irg1l) and si:busm1-266f07.1 (mhc2 family gene) (Fig.1A, Supplementary Table1). Interestingly, approximately 38% of the up-regulated probes i.e. parvalbumin 1 (pvalb1), alpha-tropomyosin (tpma), troponin I, skeletal, fast 2b, tandem duplicate 2 (tnni2b.2) and myosin, heavy polypeptide 1.1 (myhz1.1) were related to muscle associated biological processes including muscle contraction (GO:0006936), muscle system process (GO:0003012) and myofibril assembly (GO:0030239) (Supplementary Tables1 and 2). The GO-analysis of the down-regulated probes also showed a significant enrichment of another 22 processes including response to external biotic stimulus (GO: 0043207) and cholesterol biosynthetic process (GO:0006695) and immunological processes, such as response to other organism (GO:0051707), response to bacterium (GO:0009617) and the induction of bacterial agglutination (GO:0043152) (Supplementary Table2). Mycobacterial infection up-regulates itln3 expression in both zebrafish embryos and adult fish. Previous studies in several animal models have shown the expression of the Intelectin (ITLN) gene to be induced upon a bacterial infection25,30,32. Accordingly, the expression of the zebrafish itln3 (ENSDARG00000003523) was increased on average 3.3-fold (log2 change) upon a M. marinum infection in our microarray analysis (Fig.1A, Supplementary Table1). In contrast, two other itln genes; itln2 (ENSDARG00000036084) and itln2-like (ENSDARG00000093796) were down-regulated compared to the PBS controls (−3.5 and −3.2 log2 fold change, respectively) (Fig.1A, Supplementary Table1), suggesting a diverse regulation of itln genes in the M. marinum infected zebrafish. Since both ENSDARG00000036084 and www.nature.com/scientificreports/ 3 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 Figure 1. Zebrafish intelectin genes are differentially expressed upon M. marinum infection. (A) A genome-wide gene expression microarray was conducted in adult WT AB zebrafish injected with M. marinum (20 CFU; SD 6 CFU) (n = 2) or PBS (n = 3). Average numerical results (log2) for each probe in both infected fish (y-axis) and PBS controls (x-axis) are shown. Upand down-regulated transcripts (log2 fold change 3 ) in the organ blocks are shown in grey, and the common immunological genes are annotated. Two itln3 probes as well as itln2 and itln2-like probes are highlighted. (B–E) The expression of zebrafish itln genes (itln1, itln2, itln2-like and itln3) was measured with qPCR in the organ blocks of the M. marinum infected (6 CFU; SD 3 CFU) and PBS injected adult WT e46 zebrafish at 1 (n = 12 and n = 4, respectively) and 6 dpi (n = 12 and n = 8, respectively) as well as 4 (n = 12 and n = 11, respectively) and 9 wpi (n = 12 and n = 10, respectively). (F–I) The expression of itln1, itln2, itln2-like and itln3 was determined with qPCR in the M. marinum (39 CFU; SD 47 CFU) infected WT AB embryos (n = 5 at all timepoints) and in PBS injected controls (n = 5 at all timepoints) at 1–7 dpi. Note the different scales of the y axes in B-I. Gene expressions were normalized to eef1a1l1 expression and target genes were run once in the qPCR analyses. A two-tailed Mann-Whitney test was used in the statistical comparison of differences in B–I. www.nature.com/scientificreports/ 4 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 ENSDARG00000093796 share the same gene name, itln2, in Ensembl genome browser, ENSDARG00000093796 is referred to as itln2-like throughout the text. To confirm the differential expression pattern of the itln family members in the zebrafish mycobacterial infection and to study the kinetics of the host response more carefully, we analyzed itln1 (ENSDARG00000007534), itln2, itln2-like and itln3 gene expression from the abdominal cavity organ blocks of M. marinum infected (6 CFU; SD 3 CFU) WT e46 background adult zebrafish with qPCR at 1 and 6 dpi, as well as at 4 and 9 weeks post infection (wpi) (Fig.1B–E). In line with our microarray data, itln3 was significantly induced at 4 wpi (3.8-fold, P = 0.002) and 9 wpi (5.9-fold, P = 0.003) (Fig.1E), whereas itln2 was down-regulated compared to the PBS controls both at 6 dpi (0.3-fold, P = 0.019) and 4 wpi (0.2-fold, P < 0.001) (Fig.1C). No significant differences in the relative mRNA expression levels of the itln1 (Fig.1B) or itln2-like (Fig.1D) genes were observed between infected and the PBS injected adult fish at any of the measured time points. Next, we infected WT AB zebrafish embryos with mycobacteria and performed an expression analysis of the itln genes by qPCR. Here, M. marinum (39 CFU; SD 47 CFU) was microinjected into the yolk sac of the embryos and the gene expression was quantified daily between 1 and 7 dpi (Fig.1F–I). In the mycobacteria infected embryos we detected the up-regulation of both itln1 (1.8 to 6.4-fold, P = 0.008–0.016) (Fig.1F) and itln3 (1.8 to 111.4-fold, P = 0.008–0.032) (Fig.1I) starting at 2 dpi and continuing until 7 dpi, as well as the induction of itln2 at 7 dpi (21.6-fold, P = 0.008) (Fig.1G) and itln2-like (Fig.1H) between 4 and 7 dpi (20.7 to 76.1-fold, P = 0.008–0.032) compared to the PBS controls. Also, in line with previous reports suggesting that other infectious diseases up-regulate ITLN expression, a significant induction of itln3 expression (8.4-fold at 7hpi; 11.8-fold at 18hpi; 5.5-fold at 24hpi; 4.4-fold at 48hpi, P = 0.002 in all comparisons) was observed in Streptococcus pneumoniae (T4 serotype) infected (296 CFU; SD 32 CFU) embryos (Supplementary Figure1). In order to understand the infection-inducible nature of the zebrafish itln genes at steady state, we quantified the relative mRNA levels of itln1, itln2, itln2-like and itln3 in the liver, spleen, kidney and intestine of unchallenged WT e46 zebrafish by qPCR (Fig.2A–D). Here, we found that itln2 expression was restricted to the intestine (Fig.2B), whereas itln3 showed the highest relative expression in the liver and the highest overall expression compared to the housekeeping gene (eukaryotic translation elongation factor 1 alpha 1, like 1; eef1a1l1) (Fig.2D). Conversely, itln1 was expressed in all of the studied tissues with the second highest overall expression levels (Fig.2A), while itln2-like was primarily expressed in the zebrafish kidney and the intestine (Fig.2C). These results are in line with a previous qPCR analysis of the itln gene family members in unchallenged adult zebrafish25. Creating itln3 mutant zebrafish using CRISPR/Cas9 mutagenesis. The type II CRISPR/Cas system is an invaluable technology for targeted genome editing63,64, and to date it has been utilized in a number of model organisms. We and others have used the CRISPR/Cas9 mutagenesis method successfully in the zebrafish47,49,65,66. Here, we used the CRISPR/Cas9 method to create zebrafish carrying nonsense itln3 mutations for our in vivo studies (Fig.3). To this end, we identified a functional gRNA targeting the second exon of the itln3 gene with an average mutagenesis efficiency of 39.5% (Fig.3A,B). After an outcross of parental mutation carriers (F0-generation) with WT TL zebrafish, we observed two germ-line transmitted frameshift mutations in the F1-progeny corresponding to a total loss of five base pairs (−5 bp; loss of GCATC) and to a total gain of eight base pairs (+8 bp; loss of GGAGCATC and gain of TGCTAGGTAAGTATCA) at the target loci (Fig.3C). Analyses with the Translate tool (Expasy; SIB, Swiss Institute of Bioinformatics)67 of both the −5 bp and +8 bp mutations confirmed the disrupted reading frames from amino acids 47 and 45 onwards resulting in premature stop-codons after 79 and 71 amino acids, respectively (Fig.3C). These two different itln3 null mutant zebrafish lines were named itln3uta145 (−5 bp mutation) and itln3uta148 (+8 bp mutation). qPCR analysis of uninjected and M. marinum infected (422 CFU; SD 221 CFU, 2 wpi) adult zebrafish revealed diminished itln3 transcript levels in the homozygous itln3uta145/uta145 (residual expression less than 1%, P < 0.001) and itln3uta148/uta148 mutants (residual expression less than 0.1%, P < 0.001) compared to the WT controls (Supplementary Figure2), suggesting that the indel-mutations lead to the nonsense-mediated RNA decay of the mutant mRNAs68. Furthermore, the inheritance of the mutations followed Mendelian ratios for both of the mutant lines, and the homozygous itln3uta145/uta145 and itln3uta148/uta148 mutants did not show any developmental defects nor phenotypical differences compared to their WT siblings (Supplementary Figure3). Figure 2. Expression of zebrafish itln genes in adult zebrafish tissues. Relative expression of (A) itln1, (B) itln2, (C) itln2-like and (D) itln3 was measured with qPCR in the uninfected adult WT e46 zebrafish liver (n = 10), spleen (n = 10), kidney (n = 10) and the intestine (n = 10). Note the different scales of the y axes. Gene expressions were normalized to eef1a1l1 expression and target genes were run once in the qPCR analyses. www.nature.com/scientificreports/ 5 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 Nonsense mutation in itln3 does not affect host resistance against M. marinum in zebrafish embryos. The up-regulation of the expression of the itln3 gene in a M. marinum infection suggests a possible role for Itln3 in the host immunity against mycobacterial infections. To test if the resistance towards a mycobacterial infection is altered in homozygous itln3 mutant embryos, we first infected M. marinum (40 CFU; SD 30 CFU) into the yolk sac of the ungenotyped F2-progeny of heterozygous itln3uta145/+ and itln3uta148/+ zebrafish and followed their survival until 7 days post fertilization (dpf) (Fig.4A,B). Post-experiment genotyping revealed an average survival of 47% in the itln3uta145 background embryos and 48% in the embryos with the itln3uta148 background. However, any significant differences in the survival between the homozygous and heterozygous itln3 mutants or WT fish could not be observed in either itln3uta145 (Fig.4A) or itln3uta148 fish lines (Fig.4B) before 7 dpi (7 dpf). Next, we quantified the mycobacterial burden in the embryos that had survived by qPCR using primers for M. marinum internal transcribed spacer (MMITS)56 (Fig.4C). The M. marinum quantification revealed bacterial copy number medians (log10) of 4.18 and 4.15 in 100 ng of zebrafish DNA in the itln3uta145 and itln3uta145 WT groups, respectively. Comparably, heterozygous itln3uta145/+ and itln3uta148/+ fish had copy number medians of 4.02 and 4.22 (in 100 ng of zebrafish DNA, log10), respectively, and the homozygous itln3uta145/uta145 and itln3uta148/uta148 mutants 3.53 and 4.25 (in 100 ng of zebrafish DNA, log10). Thus, there were no statistically significant differences in the mycobacterial burdens between the different genotypes in neither itln3uta145 nor itln3uta148 zebrafish. The site of the bacterial injection can affect the immune response in the embryos50. Therefore, we next treated the ungenotyped F2-progeny of itln3uta145/+ and itln3uta148/+ zebrafish by injecting M. marinum into the blood circulation valley of 2-day-old embryos. In these fish, the M. marinum infection (46 CFU; SD 31 CFU) was not able to cause any mortality prior to the experimental end-point of 5 dpi (7 dpf). However, this allowed us to quantify the M. marinum burden in all of the infected embryos at the end-point (Fig.4D). Here, the bacterial copy number medians (log10) in 100 ng of zebrafish DNA were 3.61 (WT itln3uta145), 3.83 (WT itln3uta148), 3.63 (itln3uta145/+), 3.77 (itln3uta148/+), 3.75 (itln3uta145/uta145) and 3.79 (itln3uta148/uta148). Similarly to the yolk sac infection, mycobacterial quantification did not reveal any differences between the individuals of the different genotypes in either the itln3uta145 or the itln3uta148 zebrafish background. Noteworthy, we also infected the ungenotyped F2-progeny of itln3uta145/+ and itln3uta148/+ zebrafish with S. pneumoniae (serotypes 1 and T4, blood circulation valley infection at 2 dpf) and followed the survival of the fish to 5 dpi69. There was no difference between WT embryos and the itln3 mutants (Supplementary Figure1). Deleterious mutations may lead to genetic compensation, which in turn can affect the observed phenotype in gene knockout models70. To address this, we used a morpholino-oligonucleotide to silence itln1 in our itln3 mutant zebrafish together with the yolk sac mycobacterial infection of zebrafish embryos. In order to ensure efficient termination of translation in all of the four zebrafish itln1 transcripts, we targeted the second exon (E2) Figure 3. Generation of itln3uta145 and itln3uta148 mutant zebrafish lines using CRISPR-Cas9 mutagenesis. (A) An appropriate guide RNA (gRNA) target site was identified in the second exon of itln3. (B) 2.5% agarose TAE gel electrophoresis was performed to evaluate occurrence of target site mutations in zebrafish. The in vivo CRISPR/ Cas9 mutagenesis efficiency was estimated with the T7EI assay in the gRNA and Cas9 mRNA injected embryos. The size of the uninjected WT control PCR product is 210 bp, whereas the PCR products of the mutated embryos are partially cleaved at the target site. The cleavage efficiency was calculated from the band intensities and the mutagenesis efficiency calculated according to thefollowing formula: % mutagenesis = 100 × (1 – (1-fraction of cleavage)1/2)64. GeneRuler 50 bp DNA Ladder (#SM0373, Thermo Fischer Scientific) was used as a molecular weight marker (MW). Gel image is cropped to exclude portions that do not contain experimental samples. (C) gRNA target sites were sequenced from F1-generation mutant zebrafish and two frameshift mutations (-5 bp deletion, itln3uta145 and +8 bp insertion, itln3uta148) detected, leading to truncated protein products of 79 and 71 amino acids, respectively. www.nature.com/scientificreports/ 6 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 Figure 4. The lack of itln3 does not affect the survival or the mycobacterial burden of M. marinum infected zebrafish embryos. (A,B) M. marinum (40 CFU; SD 30 CFU) was injected into the yolk sac of the WT (itln3uta145) (n = 31), itln3uta145/+ (n = 74), itln3uta145/uta145 (n = 22), WT (itln3uta148) (n = 19), itln3uta148/+ (n = 27) and itln3uta148/ uta148 (n = 16) zebrafish embryos at 0 dpf and the survival recorded until 7 dpi. A log-rank (Mantel-Cox) test was used for the statistical comparison of differences. The data was collected from a single experiment. (C) Mycobacterial burden was measured by qPCR from the yolk sac infected WT (itln3uta145) (n = 17), itln3uta145/+ (n = 32), itln3uta145/uta145 (n = 10), WT (itln3uta148) (n = 9), itln3uta148/+ (n = 15) and itln3uta148/uta148 (n = 7) embryos that were alive at 7 dpi. (D) M. marinum (46 CFU; SD 31 CFU) was injected into the blood circulation valley of the WT (itln3uta145) (n = 29), itln3uta145/+ (n = 77), itln3uta145/uta145 (n = 36), WT (itln3uta148) (n = 31), itln3uta148/+ (n = 57) and itln3uta148/uta148 (n = 19) zebrafish embryos at 2 dpf and the M. marinum burden quantified at 5 dpi. Bacterial load is represented in panels C and D as bacterial copies (log10) in 100 ng of zebrafish DNA. A two-tailed MannWhitney test was used in the statistical comparison of differences in C and D. www.nature.com/scientificreports/ 7 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 of the gene with a splice-blocking (SB) morpholino (Fig.5A). In our initial SB morpholino titration experiments, 2.8 ng of itln1-blocking morpholino did not reveal any adverse effects on the survival or on the phenotype of unchallenged zebrafish embryos within the first 7 dpf. However, lower WT itln1 mRNA levels were observed in the SB morphants with residual expression of 17.1% at 4 dpi, 21.8% at 5 dpi and 33.9% at 6 dpi compared to the random control (RC) injected embryos (Fig.5B), demonstrating that this amount of the SB morpholino silences the expression of itln1 efficiently during embryonic development. In addition, detectable itln1 expression levels were observed in the RC morphants already at 1 dpi, whereas in the SB morpholino injected embryos itln1 expression was evident later starting at 2 dpi based on qPCR (Fig.5B, Supplementary Figure4). Next, we performed morpholino-M. marinum co-injections (20 CFU; SD 19 CFU) into the yolk sac of the un-genotyped F2-progeny Figure 5. Morpholino mediated silencing of itln1 expression does not alter the survival of the WT or itln3 knockout zebrafish ina M. marinum infection. (A) A schematic representation of the effects of the morpholino mediated silencing of itln1. A splice site blocking morpholino (SB) was used to prevent the normal splicing event between exon 2 and exon 3 in itln1. Morpholino binding to its target site leads to an alternative splicing event that deletes the start codon containing exon 2 from the transcript. Consequently, this prevents translation of the Itln1 protein. In order to quantify the relative amount of the WT itln1 transcript, qPCR primers were designed to specifically amplify only the WT itln1 mRNA. (B) WT itln1 expression was quantified with qPCR from the itln1 SB morpholino (n = 3) and random control morpholino (RC) injected zebrafish (n = 3) at 1–7 dpf. Gene expression was normalized to eef1a1l1 expression. All samples were run once as technical duplicates. (C–E) Survival of the morpholino and M. marinum (20 CFU; SD 19 or 13 CFU; SD 10 CFU) co-injected embryos were followed until 7 dpi. In panel C, WT (itln3uta145 and itln3uta148) embryos injected with either SB (n = 47 and n = 53, respectively) or RC morpholino (n = 29 and n = 54) are shown, whereas in panels D and E the itln3uta145 background (n = 45–73) and itln3uta148 background embryos (n = 18–35) injected with SB morpholino are depicted, respectively. Note that the SB morpholino injected WT (itln3uta145) embryo group is shown in both C and D panels in order to simplify data representation. The data in panel C was collected from two individual experiments, whereas other data is from a single experiment. A log-rank (Mantel-Cox) test was used for the statistical comparison of differences. MO = morpholino. www.nature.com/scientificreports/ 8 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 of itln3uta145/+ and itln3uta148/+ zebrafish (Fig.5C–E) and the F3-progeny of WT itln3uta148 (13 CFU; SD 10 CFU) (Fig.5C) and followed their survival up to 7 dpi. There were few dying embryos among uninfected embryos upon RC or itln1 morpholino injection (Supplementary Figure4), whereas the mortality reached 77.8–100% in the morpholino-M. marinum co-injected embryos. Noteworthy, the comparison between the infected RC and SB morpholino injected WT itln3uta145 and itln3uta148 embryos did not show any differences in survival (Fig.5C). Moreover, inhibiting itln1 expression in homozygous itln3uta145/uta145 and itln3uta148/uta148 mutants lead to a similar mortality compared to the corresponding heterozygous and WT siblings of the same genetic background (Fig.5D,E), indicating that the simultaneous lack of itln1 and itln3 functionality does not affect mycobacterial resistance in the zebrafish embryo. Consistently, we did not detect any differences in the mRNA levels of itln1, itln2 and itln2-like between the homozygous itln3 mutants and the WT controls either in uninjected (4 dpf) or M. marinum (25 CFU; SD 23 CFU, 4 dpf/4 dpi) infected embryos (Supplementary Figure5), suggesting that there is no transcriptional compensation by the other studied intelectin gene members in the itln3uta145/uta145 and itln3uta148/uta148 mutant fish. Similarly, no transcriptional compensation by itln1, itln2 or itln2-like was observed in the adult itln3 mutant zebrafish either in steady state or upon M. marinum infection (Supplementary Figure2). Adult itln3 mutant zebrafish have a normal immune response towards a M. marinum infection. In order to test the mycobacterial susceptibility of the itln3 mutants in adult zebrafish, we performed a low-dose (48 CFU; SD 5 CFU) mycobacterial inoculation into the abdominal cavity of the fish and followed their survival for up to 24 wpi (Fig.6A,B). After the follow-up, an average of 67% of the itln3uta145 background zebrafish had survived, corresponding to 74% of the WT, 73% of the itln3uta145/+ and 59% of the itln3uta145/uta145 fish. In the itln3uta148 background fish, a combined survival percentage of 81% was observed (78% in the WT, 82% in the itln3uta148/+ and 84% in the itln3uta148/uta148 fish). Similarly to the embryonic survival experiments, no statistically significant differences in the survival between the genotypes were observed. We and others have previously shown that the outcome of a mycobacterial infection in adult zebrafish depends not only on the host genotype but also on the infection dose. While, a so called low-dose inoculate can result in latency and a chronic disease56, a higher dose leads to a fast progressing acute infection59,71. We hypothesized Figure 6. Adult itln3 mutant zebrafish have comparable survival and mycobacterial burden compared to WT fish upon M. marinum infection. (A) The WT (itln3uta145) (n = 38), iltn3uta145/+ (n = 40), iltn3uta145/uta145 (n = 38) and (B) WT (itln3uta148) (n = 38), iltn3uta148/+ (n = 38), iltn3uta148/uta148 (n = 38) zebrafish were infected with M. marinum (48 CFU; SD 5), and their survival followed for 24 weeks. A log-rank (Mantel-Cox) test was used for the statistical comparison of differences. The data was collected from a single experiment. (C,D) The itln3uta145 and itln3uta148 background zebrafish were infected with M. marinum (422 CFU; SD 221 CFU) and bacterial burden (log10) in 100 ng of zebrafish DNA determined at 2 and 4 wpi from the organ blocks (without the kidney). Group sizes at 2 and 4 wpi, respectively, were as follows: WT (itln3uta145) n = 10, n = 8; iltn3uta145/+ n = 12, n = 12; iltn3uta145/uta145 n = 12, n = 12; WT (itln3uta148) n = 9, n = N/A; iltn3uta148/+ n = 9, n = 14 and iltn3uta148/uta148 n = 8, n = 12. All samples were run once. A two-tailed Mann-Whitney test was used in the statistical comparison of differences. N/A = no fish available for analysis. www.nature.com/scientificreports/ 9 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 that the effects caused by the lack of Itln3 could be more prominent in an infection with a higher mycobacterial dose. Consequently, we infected WT fish as well as heterozygous and homozygous itln3 mutants from both the itln3uta145 and itln3uta148 backgrounds with a higher M. marinum dose (422 CFU; SD 221 CFU) and quantified the bacterial burden at 2 and 4 wpi (Fig.6C,D). In these fish, we detected M. marinum copy number medians (log10) of 4.19 (WT itln3uta145), 3.76 (itln3uta145/+), 4.16 (itln3uta145/uta145), 3.69 (WT itln3uta148), 3.73 (itln3uta148/+) and 3.92 (itln3uta148/uta148) in 100 ng of zebrafish DNA at 2 wpi and 4.72 (WT itln3uta145), 4.36 (itln3uta145/+), 4.60 (itln3uta145/ uta145), 4.10 (itln3uta148/+) and 3.72 (itln3uta148/uta148) at 4 wpi. Noteworthy, no WT itln3uta148 fish were available at 4wpi for a bacterial quantification. Altogether, these data indicate that the loss of Itln3 function is dispensable for the host resistance against abdominal cavity M. marinum infection in adult zebrafish. Dexamethasone mediated lymphocyte depletion in itln3 knockout zebrafish does not affect the survival or mycobacterial burden in a M. marinum infection. We have recently published a zebrafish immune-suppression model for mycobacterial reactivation using orally administered dexamethasone60. The dexamethasone treatment decreases the total amount of lymphocytes by an average of 36% (from a relative proportion of 19.3% to 12.4%), and consequently leads to reactivation of the M. marinum infection. In turn, a number of studies have suggested that Itln3 functions in microbial surveillance and therefore in the innate immunity23,24,34. In order to highlight the importance of innate immune mechanisms in the mycobacterial defense, we used the dexamethasone treatment to specifically deplete the lymphocyte population in the adult itln3 mutation carrying zebrafish lines itln3uta145 and itln3uta148, and subsequently infected both WT and homozygous itln3 mutants with M. marinum (47 CFU; SD 4 CFU) (Fig.7A). Expectedly, our flow cytometric analysis demonstrated a significant decrease in the lymphocyte counts of both WT itln3uta145 and itln3uta148 fish (31.5%, P = 0.002 and 23.7%, P = 0.010, respectively) as well as the itln3uta145/uta145 and itln3uta148/uta148 mutants (31.5% and 40.5%, P < 0.001 in both comparisons) three weeks after initiating the dexamethasone administration at 2 wpi (Fig.7B–D). In addition, neither the total cell count nor the amount of myeloid cells and blood cell precursors were affected by dexamethasone (Supplementary Figure6). We did not detect any substantial mortality of either the itln3uta145 or the itln3uta148 mutants or WT fish during the five-week follow-up period. As is shown in the Fig.7D,E, the bacterial amounts did not differ between the groups; in 100 ng of zebrafish DNA, mycobacterial copy number medians (log10) of 2.60 and 2.65 in WT itln3uta145, 2.55 and 3.10 in itln3uta145/uta145, 2.87 and 2.43 in WT itln3uta148 and 2.25 and 2.91 in itln3uta148/uta148 zebrafish were observed at 2 and 4 wpi, respectively. In conclusion, our data are in accordance with previous literature on the possible role for itlns in immunity, as the zebrafish itln3 is highly induced in a mycobacterial infection. However, M. marinum infection experiments using both zebrafish embryos and adult fish suggest that itln3 is dispensable for a protective mycobacterial host response. Moreover, itln1 does not seem to compensate for the lack of functional itln3 in the embryonic infection model. Of note, unlike has been reported for human ITLN1, we were unable to demonstrate direct binding of recombinant Itln3 to mycobacteria (or S. pneumoniae or Escherichia coli) in vitro (Supplementary Figure7), which may explain the nonessential role of Itln3 for zebrafish immunity in our models. Discussion The genetics of the host affect the outcome of a M. tuberculosis infection, i.e. the development of active tuberculosis4. Genome-wide expression analyses using microarray and RNA sequencing platforms are important for understanding complicated biological processes such as the host immune defense against pathogens. To date, a handful of transcriptome studies have been done in the zebrafish M. marinum infection model using microarray technology59,72,73, the digital gene expression (DGE) method74 and RNA sequencing75–77. Collectively, by using both zebrafish embryos and adult fish, these studies have provided important insights into the innate and adaptive host response against mycobacterial infections. We used the adult zebrafish M. marinum (ATCC 927) infection model together with a zebrafish gene expression microarray to identify novel candidate genes in a mycobacterial infection. From this data, we identified a total of 91 differentially expressed genes (log2 fold change > 3 ) that were linked to 44 enriched processes, including genes associated with the immune response. Previous studies have shown several genes of the complement system (e.g. complement component c3b, c3b; complement component 6, c6) to be up-regulated in an infection59,72–74,76, whereas the expression of some complement associated genes (e.g. complement factor b, cfb; mannose binding lectin, mbl) has been shown to be reduced72,74. In line with previous results, we also saw an induction of cd59 (regulation of membrane attack complex formation) as well as reduced expression of cfbl (component of the C3 convertase). Conversely, although previous transcriptomic studies have shown the induction of genes that are involved in neutrophil and macrophage related functions (e.g. mpx and irg1l)73,76, our data indicated down-regulation of these transcripts in an infection. In summary, the aforementioned similarities and differences between these transcriptomic studies can result from a number of factors including the developmental stage of the host (embryos vs. adult fish), the time points chosen for sample collection, the different outcomes of an infection (chronic vs. acute), the use of different bacterial strains (E11, Mma20 or ATCC 927) and doses, and they can be due to differences in the technical execution of sample preparation and analyses. Interestingly, circa 38% of the up-regulated probes were related to muscle associated biological processes including muscle contraction (GO:0006936), muscle system process (GO:0003012) and myofibril assembly (GO:0030239). Supporting the relevance of this finding, a genome-wide expression analysis in the fruit fly Drosophila melanogaster identified several muscle specific genes such as actin88F (Act88F) and tropomycin 2 (Tm2) to be induced after a Pseudomonas aeruginosa infection78. Consistently, the down-regulation of muscle expressed genes (troponin C41C, TpnC41C; glutathione S-Transferase 2, Gst2) was later connected to an increased susceptibility to infection, suggesting an immunological role for muscle tissue79,80. Although the differential expression of muscle specific genes can be indirectly linked to the immune response through the regulation of other physiological functions, as has been also suggested by Chatterjee et al.,81, both mouse and zebrafish muscle www.nature.com/scientificreports/ 16 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 81. Chatterjee, A., Roy, D., Patnaik, E. & Nongthomba, U. Muscles provide protection during microbial infection by activating innate immune response pathways in Drosophila and zebrafish. Dis Model Mech 9, 697–705 (2016). 82. Lin, S., Fan, T., Wu, J., Hui, C. & Chen, J. Immune response and inhibition of bacterial growth by electrotransfer of plasmid DNA containing the antimicrobial peptide, epinecidin-1, into zebrafish muscle. Fish Shellfish Immunol. 26, 451–458 (2009). 83. Frost, R. A., Nystrom, G. J. & Lang, C. H. Lipopolysaccharide regulates proinflammatory cytokine expression in mouse myoblasts and skeletal muscle. Am. J. Physiol. Regul. Integr. Comp. Physiol. 283, 698 (2002). 84. French, A. T. et al. Up-regulation of intelectin in sheep after infection with Teladorsagia circumcincta. Int. J. Parasitol. 38, 467–475 (2008). 85. Pemberton, A. D., Knight, P. A., Wright, S. H. & Miller, H. R. P. Proteomic analysis of mouse jejunal epithelium and its response to infection with the intestinal nematode, Trichinella spiralis. Proteomics 4, 1101–1108 (2004). 86. Datta, R. et al. Identification of novel genes in intestinal tissue that are regulated after infection with an intestinal nematode parasite. Infect. Immun. 73, 4025–4033 (2005). 87. Podok, P., Xu, L., Xu, D. & Lu, L. Different expression profiles of Interleukin 11 (IL-11), Intelectin (ITLN) and Purine nucleoside phosphorylase 5a (PNP 5a) in crucian carp (Carassius auratus gibelio) in response to Cyprinid herpesvirus 2 and Aeromonas hydrophila. Fish Shellfish Immunol. 38, 65–73 (2014). 88. Winkelhake, J. L., Vodicnik, M. J. & Taylor, J. L. Induction in rainbow trout of an acute phase (C-reactive) protein by chemicals of environmental concern. Comp. Biochem. Physiol. C, Comp. Pharmacol. Toxicol. 74, 55–58 (1983). 89. Kovacevic, N., Hagen, M. O., Xie, J. & Belosevic, M. The analysis of the acute phase response during the course of Trypanosoma carassii infection in the goldfish (Carassius auratus L.). Dev. Comp. Immunol. 53, 112–122 (2015). 90. Talbot, A. T., Pottinger, T. G., Smith, T. J. & Cairns, M. T. Acute phase gene expression in rainbow trout (Oncorhynchus mykiss) after exposure to a confinement stressor: A comparison of pooled and individual data. Fish Shellfish Immunol. 27, 309–317 (2009). 91. Gerwick, L., Corley-Smith, G. & Bayne, C. J. Gene transcript changes in individual rainbow trout livers following an inflammatory stimulus. Fish Shellfish Immunol. 22, 157–171 (2007). 92. Kerr, S. C. et al. Intelectin-1 is a prominent protein constituent of pathologic mucus associated with eosinophilic airway inflammation in asthma. Am. J. Respir. Crit. Care Med. 189, 1005–1007 (2014). 93. Renshaw, S. A. & Trede, N. S. A model 450 million years in the making: zebrafish and vertebrate immunity. Dis. Model. Mech. 5, 38–47 (2012). 94. Tang, R., Dodd, A., Lai, D., McNabb, W. C. & Love, D. R. Validation of zebrafish (Danio rerio) reference genes for quantitative realtime RT-PCR normalization. Acta Biochim. Biophys. Sin. (Shanghai) 39, 384–390 (2007). 95. Hruscha, A. & Schmid, B. Generation of zebrafish models by CRISPR /Cas9 genome editing. Methods Mol. Biol. 1254, 341–350 (2015). 96. Krebs, J. CRISPR design tool and protocol (2015). Data retrieved: 07:16, May 19, GMT. https://figshare.com/articles/CRISPR_ Design_Tool/1117899 (2015). 97. Altschul, S. F. et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic acids research 25, 3389–3402 (1997). 98. Meeker, N. D., Hutchinson, S. A., Ho, L. & Trede, N. S. Method for isolation of PCR-ready genomic DNA from zebrafish tissues. BioTechniques 43, 614 (2007). 99. Turpeinen, H. et al. Proprotein Convertase Subtilisin/Kexin Type 7 (PCSK7) Is Essential for the Zebrafish Development and Bioavailability of Transforming Growth Factor beta1a (TGFbeta1a). J. Biol. Chem. 288, 36610–36623 (2013). 100. Harjula, S. E., Ojanen, M. J. T., Taavitsainen, S., Nykter, M. & Rämet, M. Interleukin 10 mutant zebrafish have an enhanced interferon gamma response and improved survival against a Mycobacterium marinum infection. Sci Rep 8, 10360 (2018). 101. Traver, D. et al. Transplantation and in vivo imaging of multilineage engraftment in zebrafish bloodless mutants. Nat. Immunol. 4, 1238–1246 (2003). Acknowledgements This study was financially supported by the Academy of Finland (M.R., 277495, M.P. 295814 and 286477), the Sigrid Juselius Foundation (M.R.), the Jane and Aatos Erkko Foundation (M.R.), the Competitive State Research Financing of the Expert Responsibility Area of Tampere University Hospital (M.R., M.P. 9U047 and 9V049), the Competitive State Research Financing of the Expert Responsibility area of Oulu University Hospital (M.R.), the Tampere Tuberculosis Foundation (M.O., M.R., S.-K.H., M.P.), the City of Tampere Science Foundation (S.- K.H), the Väinö and Laina Kivi Foundation (M.O., S.-K.H.), the Finnish Cultural Foundation, the Central Fund (S.-K.H.), the Finnish Concordia Fund (S.-K.H.), the Orion Research Foundation sr (S.-K.H), the Maud Kuistila Memorial Foundation (M.O.), the University of Tampere Doctoral Programme in Biomedicine and Biotechnology (M.O.), the Cancer Society of Finland (M.P.) and Tays tukisäätiö (Tays Support Foundation) (M.P.). We thank the Tampere Zebrafish Core Facility, partly funded by Biocenter Finland, for maintaining and providing the zebrafish. The use of the facilities and expertise of the Protein Technologies core facility of the University of Tampere, a member of Biocenter Finland, is also gratefully acknowledged. We also greatly acknowledge Hannaleena Piippo, Jenna Ilomäki, Leena Mäkinen, Carina Bäuerlein, Mirja Niskanen, Juha Saarikettu, Janey Barron, Christopher Gault, Janne Kärnä, Marianne Karlsberg, Ine Herman, Anna Grönholm and Latifeh Azizi for technical assistance and Jukka Lehtiniemi for taking the adult zebrafish images. In addition we thank Henna Myllymäki, Hannu Turpeinen, Mataleena Parikka and Tero Järvinen for scientific advice and support, as well as Hannah Pratt and Helen Cooper for proof-reading this manuscript. Author Contributions M.P., V.H. and M.R. provided materials and facilities for the research. M.O., M.U., J.M., V.H., M.P. and M.R. designed and/or directed the experiments. M.O., A.S., N.K., S.-K.H., K.O. and M.U. performed the experiments. M.O. and M.U. analyzed the data. M.O., M.U. and M.R. wrote the paper. All authors reviewed and approved the manuscript. Additional Information Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-018-37678-1. Competing Interests: The authors declare no competing interests. www.nature.com/scientificreports/ 17 Scientific RepoRts | (2019) 9:995 | https://doi.org/10.1038/s41598-018-37678-1 Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2019