Taenia solium and Taenia crassiceps: miRNomes of the larvae and effects of miR-10-5p and let-7-5p on murine peritoneal macrophages

Abstract Neurocysticercosis (NCC), a major cause of neurological morbidity worldwide, is caused by the larvae of Taenia solium. Cestodes secrete molecules that block the Th1 response of their hosts and induce a Th2 response permissive to their establishment. Mature microRNAs (miRs) are small noncoding RNAs that regulate gene expression and participate in immunological processes. To determine the participation of Taenia miRs in the immune response against cysticercosis, we constructed small RNA (sRNA) libraries from larvae of Taenia solium and Taenia crassiceps. A total of 12074504 and 11779456 sequencing reads for T. solium and T. crassiceps, respectively, were mapped to the genomes of T. solium and other helminths. Both larvae shared similar miRNome, and miR-10-5p was the most abundant in both species, followed by let-7-5p in T. solium and miR-4989-3p in T. crassiceps, whereas among the genus-specific miRs, miR-001-3p was the most abundant in both, followed by miR-002-3p in T. solium and miR-003a-3p in T. crassiceps. The sequences of these miRs were identical in both. Structure and target prediction analyses revealed that these pre-miRs formed a hairpin and had more than one target involved in immunoregulation. Culture of macrophages, RT-PCR and ELISA assays showed that cells internalized miR-10-5p and let-7-5p into the cytoplasm and the miRs strongly decreased interleukin 16 (Il6) expression, tumor necrosis factor (TNF) and IL-12 secretion, and moderately decreased nitric oxide synthase inducible (Nos2) and Il1b expression (pro-inflammatory cytokines) in M(IFN-γ) macrophages and expression of Tgf1b, and the secretion of IL-10 (anti-inflammatory cytokines) in M(IL-4) macrophages. These findings could help us understand the role of miRs in the host–Taenia relationship.


Introduction
Neurocysticercosis (NCC) in humans is caused by larvae of Taenia solium. It is the leading cause of seizures and epilepsy and is considered a serious health problem worldwide [1]. In human NCC, the intensity of symptoms depends primarily on the inflammatory response, which is associated with the Th1 response with high levels of TNF, IFN-γ, IL-17, and IL-23 and coincides with degeneration of the larvae [2,3], whereas the Th2 response (anti-inflammatory response) is associated with asymptomatic NCC with high level of IL-10, IL-4, IL5, and IL-13 and viable larvae [3,4]. In animal NCC, swine and murine granulomas and their inflammatory responses are similar than in humans [5][6][7][8]. However, the exact role of classically and alternatively activated macrophages/microglia and their secreted molecules in NCC patients remains to be elucidated. In contrast, in the model of murine peritoneal cysticercosis by Taenia crassiceps, the macrophages play a key role in promoting the Th1 (protective) and Th2 (permissive) responses against the larvae [9,10]. The depletion of alternatively activated macrophages during early infection decreases the parasitic burden and restores the antigen-specific proliferative response in T lymphocytes [11,12]. It is also known that the products of T. crassiceps larvae block the toll-like receptor (TLR) response and the production of the inflammatory cytokines in both macrophages and dendritic cells from mice, biasing to Th2 environment that includes T cells (Th2 and Treg), eosinophils, and alternatively activated macrophages with the secretion of type 2 cytokines that facilitate larval growth [13][14][15].
Mature microRNAs (miRs) are small noncoding RNAs (∼22 nt) that bind to complementary sequences in the 3 untranslated region (3 -UTR) of target mRNAs and mediate either translational repression or mRNA degradation [16,17]. They are involved in processes, such as developmental transitions, responses to the environment, cell signaling, and in the regulation of the immune response [16][17][18]. In addition, miRs are also associated with M1 and M2 activation in human and murine macrophages [19][20][21][22]. In Taenia genus, adult-stage miRs have been reported in T. multiceps, T. saginata, T. asiatica, and T. solium, showing similar profiles; in contrast, only one report of the larval stage miR of T. ovis was found [23][24][25][26]. However, functional tests for these miRs have not been performed.
The present study aimed to sequence the miRs from the larvae of T. solium and T. crassiceps to compare their expression profiles, as well as bioinformatics analysis of their structure and abundance and the prediction of their immunological targets. In addition, we show the effects of the two most abundant miRs on naive and activated peritoneal macrophages.

Biological material
Taenia crassiceps (ORF strain) larvae were obtained from experimentally infected mice, as previously described [10], and then, killed in a CO 2 chamber 90 days later. T. solium larvae were dissected from naturally infected rural pigs acquired from local farms. Larvae were washed four times with sterile ice-cold phosphate-buffered saline (PBS) pH 7.2 and stored at −70 • C. Naive macrophages were obtained from non-infected mice.

Uptake of miRs by macrophages
Naive murine macrophages were isolated from the peritoneum by aspiration with RPMI and 2 × 10 6 cells/ml were seeded on sterilized coverslips coated with poly-d-lysine (Neuvitro, Vancouver, WA, U.S.A.) in a culture plate, as previously described [28]. Cultures were incubated in RPMI-1640 with 10% FBS, 100 U/ml penicillin, and 100 μg streptomycin in an atmosphere of 5% CO 2 at 37 • C for 15, 30, or 60 min in the absence or presence of synthetic miR-10-5p or let-7-5p (100 ng/ml) coupled to Cyanine 5 (Cy5) (Sigma, St Louis, MO, U.S.A.). Cells were washed once with PBS and fixed with 4% paraformaldehyde for 1 h at 4 • C. The cells were washed three times with PBS, and then, stained with 4 ,6-diamidino-2-phenylindole (DAPI) for 5 min. After one more wash with PBS, the macrophages adhered to the coverslips were mounted on to slides using 50% glycerin in PBS. Fluorescence was detected using a laser scanning confocal microscope at the excitation wavelengths of 405 and 603 nm for DAPI and Cy5, respectively. Images were captured under a vertical LEICA TCS-SP5 II microscope and processed with the manufacturer's software. Three-channel images were acquired in the light field microscopy and fluorescence microscopy to detect DAPI and Cy5, and then, merged for analysis. Orthogonal images were obtained by laser scanning in an XYZ configuration (3D reconstruction), taken in projections of 17-18 sequential sections, scanned five to seven times each, with a thickness of 10.5 μm.

Statistical analysis
One-way Analysis of Variance (ANOVA) with Tukey's multiple comparisons test was used for the statistical analysis of collected data and P-values <0.05 were considered significant. We performed all possible comparisons, however only those comparisons relevant to the discussion are described in the result: to macrophages M(N) vs. M(IFNγ), M(N) vs. M(IL-4), No incubation vs. miR-10-5p, and No incubation vs. let-7-5p.

Isolation of sRNAs from T. solium and T. crassiceps larvae
Total RNA extracted from T. solium and T. crassiceps larvae presented the characteristic pattern of the RNAs from cestodes ( Figure 1A, lane 1), showing the majority of mRNAs between 0.5 and ∼7000 nt, a single ribosomal RNA (rRNA) band of ∼2438 nts (composed of 18S and 28S rRNAs), and an sRNA fraction of ∼200 nts. The RNA fraction of >200 nts (lane 2) and the sRNA fraction (lane 3) were isolated. Figure 1B shows the electropherogram plots of the fluorescence intensities of the bands that form the T. solium (Up) and T. crassiceps (Down) total RNA, with a peak of ∼25 nts (peak 1) corresponding to putative miRs, a peak of ∼100 nts (peak 2) corresponding to other sRNAs (transfer RNA (tRNA), small interfering RNA (siRNA), and small nucleolar RNA (snoRNA)), and peaks between 2000 and 2500 nts (peaks 3 and 4) corresponding to rRNAs and larger RNAs.

Characterization of the sRNA libraries
The sequences of sRNA libraries showed a total of 12074504 raw reads obtained for T. solium, of which 8300154 reads were mappable. For T. crassiceps, a total of 11779456 reads were obtained, of which 8256769 were mappable. The mappable reads from T. solium (58.2%) and T. crassiceps (50.7%), correspond to know and predicted miRs, 12.8 and 10.2%, respectively were mapped to mRNAs, and 14.2 and 11.4%, respectively, were mapped to snoRNAs, siRNA, tRNAs, and rRNA fragments. We found 19.4% reads with no hits to T. solium sequence data, and 32% reads with no hits to T. crassiceps sequence data. Table 1 shows the known and predicted miRs grouped as A, B, C, D, and E, as well as the number of unique miRs and corresponding reads. Notably, a higher number of unique miRs belonged to the group of predicted miRs mapped to the genome of T. solium, but not to that of other helminths (Group D). Moreover, we found 41 unique known miRs in T. solium and 40 in T. crassiceps that were mappable to the genomes of T. solium and other helminths. The total numbers of unique miRs were 335 in T. solium and 192 in T. crassiceps. For the known T. solium miRs, 48.8% were detected at the 5p arm and 51.22% at the 3p arm in 93 clusters of which 67.74% showed reads with zero errors and a score of >500. For the known T. crassiceps miRs, 47.73% were detected at the 5p arm and 52.27% at the 3p arm in 99 clusters, of which 64.65% showed reads with zero errors. The GC contents in these miR sequences were between 50 and 70%, and an sRNA population of between 17 and 28 nts (with a predominant size at 22 nts) was observed in both species ( Figure 2).   Figure 3 shows the sequences and number of reads of the most abundant miRs in T. solium larvae; namely Tso-miR-10-5p (670475 reads), Tso-let-7-5p (184435 reads), Tso-miR-71-5p (107153 reads), and Tso-miR-4989-3p (94010 reads), followed by Tso-miR-61-3p, Tso-miR-125-5p, Tso-miR-2c-3p, and Tso-bantam-3p with more than 20000 reads. It also shows the sequences and number of reads of the most abundant miRs in T. crassiceps larvae; they were Tcr-miR-10-5p (907685 reads), Tcr-miR-4989-3p (199774 reads), Tcr-let-7-5p (118312 reads), and

solium (gray) and T. crassiceps (black)
Reads were obtained after removing the 3 -adapter.  The cutoff for the number of copy sequences was 300 reads. The symbol ✪ shows miRs found only in the T. solium (Tso-) library.
The seed sequences of each miR are in black and underlined.
The sequence, length, and identification of total known miRs found in both parasites are shown in Supplementary  Table S1. We observed a similar abundance of miRs for both Taenia larvae, likewise, we also detected several miRs that were unknown in T. solium and T. crassiceps and that did not map to the genomes of other helminths different to genus Taenia; these miR sequences and copy numbers are shown in Supplementary Table S2, and the most abundant miRs in this group are shown in Figure 4. The most abundant genus-specific unknown miR (miR-001-3p) was identical in both Taenia larvae (11414 reads to T. solium and 7720 reads to T. crassiceps). Others abundant miRs were Tso-miR-002-3p (7252 reads) and Tcr-003-3p (2816 reads), where Tso-miR-001-3p represented more than 30% of the total of reads in this group. We also found some novel miRs in T. solium that were absent from T. crassiceps, but with a low frequency of ∼1000 reads (listed in Supplementary Table S3). Table 2 shows the location of the most abundant miRs and two genus-specific unknown miRs (miR-001-3p and miR-002-3p) on the T. solium genome. Notably, miR-71-5p and the miR-2 family were localized in the same cluster, as were miR-4989-3p and miR-277-3p in both species. Figure 5A shows the predicted hairpin structure for miR-10-5p with a minimum free energy of −26.50 kcal/mol, and its predicted immunologic targets, the transcriptional repressor B-cell lymphoma 6 (Bcl6), coding for a molecule involved in the stability or regulatory T cells [33], the nuclear receptor corepressor 2 (Ncor2), and the trans-acting T-cell-specific transcription factor (Gata3), which inhibit proliferation and promote apoptosis in diffuse large B-cell lymphomas [34]. Figure 5B shows the let-7-5p hairpin structure with a minimum free energy of −34.10 kcal/mol and the two predicted targets, the interleukin 13 (Il13) and de CC-chemokine receptor 7 (Ccr7) cytokines involved in anti-inflammatory response [35]. Similarly, Supplementary Figure S1 shows the predicted structure of novel pre-miR Tso-mir-001-3p, with the minimum free energy of −44.90 kcal/mol, and its predicted immunologic targets; namely, early T-cell activation antigen P60 (Cd69), an important regulator of the immune response highly expressed by memory and regulatory T cells in the gut, which is associated with negative regulation of Th1 and Th17-mediated immune response [36]. Notably, this gene is the principal target for this miR, matching at three positions in the 3 -UTR of its Members of the same family are shown in the same row.  mRNA. Other targets of this miR were the TNF receptor-associated factor 3 (Traf3), which is involved in the signal transduction of CD40 and the regulation of NF-kB1 activation [37] and the suppressor of cytokine signaling 5 (Socs5), a cytokine-inducible negative regulator of cytokine signaling that plays a role as a specific negative regulator for Th2 differentiation [38]. Supplementary Figure S2 shows the secondary structure of Tso-mir-002-p3, with a free energy of −69.40 kca/mol, and the putative immunologic target, signaling lymphocytic activation molecule family member 1 (Slamf1) and the V-Set immunoregulatory receptor (Vsir). The Slamf1 has a costimulatory effect on T cells and participates in the activation of B lymphocytes and some authors have suggested that may be involved in the generation of the adaptive and inflammatory immune responses [39], Vsir (VISTA), is a negative immune-checkpoint protein associated with suppression of T-cell response and suppression of cytokines like IL-2 [40]. Table 3 lists the other abundant T. solium and T. crassiceps miRs and their putative immunologic target genes, such as Il10, Il12, Tnf , Mmd2, Tgfbr1, Irf4, Mapk4, Nfκb1, and Il23p40, which are involved in immunologic regulatory functions in the innate and adaptive immune responses of the cells.

Uptake of miRs by peritoneal macrophages
For further assays, we chose miR-10-5p and let-7-p5 due to their abundance and predictive targets involved in the classical and alternative activation of macrophages, respectively. The M(N) macrophages were incubated with miR-10-5p-Cya5 (miR10-Cya5) or let-7-5p-Cya5 (let7-Cya5) in vitro and the internalization was monitored at different times by confocal microscopy. Figure 6 shows the presence of miR-10-Cya5 in the cytoplasm, where few faint red specks were observed at 15 min, while the images at 30 and 60 min, showed a clear pattern of red specks in the cytoplasm. A blue signal (DAPI) was observed in the nucleus of all macrophages. In addition, macrophages incubated without miRs showed no red signals. Orthogonal sections showed red specks inside the cytoplasm of cells. Similar results were observed with let7-Cya5 (Supplementary Figure S3). The quantification shows a significant increase in the fluorescence intensity inside the cells, >3 units at 30 and 60 min with respect to 15 min for miR10-Cya5, while the increase in let7-Cya5 had significance only between 15 and 60 min (Supplementary Figure S4).

Effect of T. solium miRs on peritoneal macrophages
The pattern of cytokines in M(N), M(IFN-γ), and M(IL-4) macrophages and the effect of miRs over them is shown in Figure 7. In M(N) macrophages only Il1b and Tgfb1 were expressed, the treatment with IFN-γ induced an M1-like activation with a significant increase in Nos2, Il6, and Il1b expression, while the treatment with IL-4 increases the Tgfb1 expression. Likewise, the treatment with the miRs altered the expression of this cytokines in all groups ( Figure  7A,B). For instance, in the M(IFN-γ) macrophages the expression level of Nos2 and Il1b, decreased significantly upon incubation with miR-10-5p or let-7-5p and the Il6 expression, was blocked with the incubation of any of the miRs. In the M(N) macrophages, the Il1b expression decreased significantly upon incubation with miR-10-5p and was totally blocked with let-7-5p. In M(IL-4) macrophages. the Il1b expression increased moderately with miR-10 and decreased with let-7. Concerning Tgfb1, the expression was reduced by both miRs in M(N) and M(IL-4) macrophages but not was affected in the M (IFN-γ). The miR-10-5p showed a bigger effect over Nos2 and Tgfb1 and let-7-5p over Il1b. A constant expression of the Gapdh (housekeeping gene) was observed in all the three macrophage groups ( Figure 7A). Likewise, in the culture supernatant (Figure 8), we observed a basal secretion of TNF, IL-12, and IL-10 in the M(N) macrophages, that is not significantly modified upon incubation with any of the miRs. As expected, the M(IFN-γ) macrophages increased the secretion of TNF and Il-12 significantly, but it is reduced at basal level upon incubation with both miRs. The M(IL-4) macrophages produced basal levels of this cytokines, and we did not find any statistical differences between treatments. In contrast, M(IL-4) macrophages increased strongly the secretion of IL-10, which is decreased by one-third upon incubation with any of the miRs.

Discussion
In the present study, we described, as the first step, the miRNomes of the larvae of the closely related organisms T. solium and T. crassiceps. A total of 18.26 million reads of high quality for T. solium and 15.24 million for T. crassiceps were obtained, with a total of 335 miRs and 192 miRs, respectively. We found at least 40 miRs that were classified as known miRs, resembling the miRs of cestodes like Echinococcus genus with high identity (100% identity in more than 50% of miRs) and more than 100 unknown miRs (not found in other cestodes); similar sequences were shared by both the species, and we also found great similarities in the kinds, sizes, and distributions by copy number among the larvae of these two Taenia species. In addition, for both larvae, the most abundant miR was miR-10-5p, while the other abundant miRs were let-7-5p, miR-71-5p, and miR-4989-3p. It was in concordance with the results of other cestodes, such as Mesocestoides corti, Echinococcus multilocularis, and Echinococcus canadensis [41,42]. In contrast, in Taenia adult stage, miR-7-5p, miR-71, miR-277, miR-219-5p, and miR-2b-3p were predominant in T. solium and T. asiatica, and miR-71 and miR-219-5p were the most abundant in T. saginata [25,26]. Notably, miR-71 was localized in the same cluster with miR-2b and miR-2c, while miR-4849 was in the same cluster with miR-277; these clusters are conserved across Cestodes [24,43]. In addition, miR-71 is highly expressed in larvae and adults, and it has been involved in processes, such as longevity and neuron development, suggesting that it could be involved in important biological functions in the life cycle of Taenia genus [44,45]. Moreover, we found some putative miRs, present only in T. solium and others in T. crassiceps, but they all matched with the T. solium genome. This result could not be confirmed to T. crassiceps since the number of reads was low, and the T. crassiceps genome is not currently available. In our study, the predicted hairpin structures, the presence of homologs to Drosha, Dicer, and Pasha (as identified in the T. solium Genome Project) [46], and the differences in miR profiles between the Taenia larvae and adults suggest that: (1) the process for the synthesis of miRs is similar to that described for mammals, and (2) the expression of miRs is in accordance with the environment confronted by the larvae or adult in the hosts.
In our target prediction analysis, we found that Taenia larvae miR-10-5p, miR-125-5p, and Tso-miR-001 have target transcription factors and adaptor proteins that are involved in macrophage IRF/STAT pathways, such as Cd69 and Tnf , involved in the IFN-signaling pathway, which regulate the expression of cytokine receptors, cell activation markers (CD38, CD69, CD97), and cell adhesion molecules that activate macrophages to secrete TNF to stimulate more macrophages through the TNFR1, leaving to classical activation [19,47,48]. Moreover, the targets of miR-125 and miR-9, such as Irf4 and Nfkb, also are important in the classical activation of cells because TNF is mediated by canonical NF-κB and MAPK signaling that activates Il1 and Il6, while IRF4-deficiency allows the generation of Foxp3 + Tregs cells [49]. In cysticercosis, Tregs induction seems to participate in the control of the inflammatory responses since a negative correlation between the percentage of peripheral Tregs and activated CD8 + and CD4 + T cells, along with a depressed T-cell proliferative response has been observed [50]. Likewise, the miR-10-5p targets, such as Il12 and Il23, could interfere with the IL-12 family signaling pathway. In addition, let-7-5p showed a predicted target that encodes cytokines involved in M2 polarization, such as Il10 that triggers M2c subtype phenotype and Il13 that elicits the M2a phenotype [51]. In contrast, Tso-mir-002-3p has targets that participate in the activation of T and B lymphocytes, and therefore, is involved in the adaptive immune responses [39]. These analyses suggest an important role for Taenia miRs in the polarization of macrophages. Macrophages in cysticercosis promote a transient Th1 protective response with classical activated macrophages that is changed by parasite products to a Th2 permissive response with alternatively activated macrophages [14]. Moreover, studies in human and murine macrophages showed that miR-9, miR-127, miR-155, and miR-125b are associated with classical activation directly affecting IFNγ or Bcl6 [19,20,52]. In addition, other miRs, such as miR-223, miR-34a, miR-132, miR-146a, miR-125a-5p, let-7c, and miR-124 promote alternative activation by regulating TLR4 and IL-4 signaling [21,22,52].
Already knowing that the more abundant Taenia miRs (miR-10-5p, let-7-5p) putatively have target genes of immune response and that macrophages in murine cysticercosis promote Th1 or Th2 responses [9,10]. We demonstrated by confocal microscopy that synthetic miR-10-5p and let-7-5p were internalized into the cytoplasm of M(N), M(IFN-γ), and M(IL-4) murine peritoneal macrophages in vitro. Notably, the incubation of M(IFN-γ) macrophages with miR-10-5p or let-7-5p significantly down-regulated the expression of Il6, Il1b, and TNF, IL-12 secretion (pro-inflammatory cytokines). Moreover, in the M(IL-4) macrophages, these miRs reduced the expression of cytokines involved in M2/Th2 differentiation. Even in our predicted analysis, let-7-5p targets IL-10 and moderately down-regulates this cytokine. However, our results for let-7-5p are in agreement with those observed for let7, let-7a, and let-7d that down-regulate IL-6 and IL-10 directly or TNF, IL-6, and IL-1β indirectly in MCF10A, fibroblasts, and breast cancer cells [53,54,55]. These results were important, because murine resistant to cysticercosis display high levels of TNF, IL-12, IL1-β, and NO during early infection (Th1 response), which is associated with the elimination of larvae [12]. On the other hand, high levels of pro-inflammatory cytokines (IL-6 and TNF) cause damage to the microglia promoting autoimmune and neurodegenerative diseases [56,57]. This tissue damage is also observed in human NCC at the beginning of larvae degeneration [2,3] and in pig NCC when they are treated with praziquantel [6]. In contrast, viable larvae are associated with a long initial asymptomatic phase that correlate with a undetectable inflammation in the Central Nervous System, presumably due to T. solium larvae factors prevent inflammation [58]. On the other hand, evidence exists that an sRNA-peptide of T. solium significantly reduces the inflammation around the larvae and decreases the antibody responses in the murine cysticercosis [59]. Therefore, the property of Taenia miRs, i.e. the down-regulation of pro-inflammatory cytokines in M(IFN-γ) macrophages could be used in the future to control tissue damage and the process of chronic neurodegeneration in Alzheimer, Parkinson, amyotrophic lateral sclerosis, multiple sclerosis, and NCC [60,61]. However, more studies are needed to be carried out to explore it further. The understanding the role of miRs in host-parasite relationship could help find other targets to develop new molecules for the regulation of inflammation [59]. Additionally, these miRs could be targets of miRNA inhibitors with the aim of favoring a protective response against the parasite. On the other hand, the abundant and specific Taenia miRs could be used as markers in the diagnosis of this parasitic disease [62].
Future studies are currently under way to evaluate the precise mechanism of the uptake of nude Taenia larvae miRs by macrophages and the time required for macrophage differentiation and also to determine the effect of Taenia unknown and specific miRs on immune cells.

Conclusion
Our results show that larval miRNomes of T. crassiceps and T. solium are similar in composition. Both miR-10-p5 and let-7-5p, the most abundant miRs, are conserved in the larvae belonging to the Taenia genus, and strongly down-regulate the production of pro-inflammatory cytokines in macrophages M(IFN-γ) and moderately anti-inflammatory cytokines in macrophages M(IL-4). It suggests a new immunosuppressive mechanism that helps the larvae in their establishment and permanence inside the host. Furthermore, we propose that therapies based in the Taenia miRs could be employed in the control of larvae of cestode and inflammatory diseases.

Ethics Approval
All protocols were carried out in the animal facilities of the Facultad de Medicina Universidad Nacional Autonoma de Mexico, under controlled conditions of temperature (22 • C), a relative humidity of 50-60% and 12-h dark/light cycles, in strict accordance with the Official Mexican Norm for the Production, Care, and Use of Laboratory Animals (NOM-062-ZOO-1999) and the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, U.S.A. The project was also approved by the Internal Committee for the Care and Use of Laboratory Animals (CICUAL) from the Medical Faculty of Universidad Nacional Autónoma de Mexico.