The sialidase NEU3 promotes pulmonary fibrosis in mice

Chinese hamster ovary (CHO-K1) cells were cultured in CDM4CHO medium with l-glutamine (Cat# SH30557.02, Hyclone/Cytiva, Marlborough, MA, USA). Cells (1 × 10⁵) were mixed with 2 μg of 100 μg/ml of murine NEU3 expression plasmid (MR223297; Origene, Rockville, MD, USA) in 100 μl PBS (GE Lifesciences, Marlborough, MA, USA) and were transfected by electroporation using a 4D-Nucleofector System (Lonza, Walkersville, MD, USA) following the manufacturer’s protocol. Before use, the plasmid was sequenced for verification as previously described [25, 27‐29]. The transfected cells were kept at room temperature for 15 min for recovery, after which the CHO-K1 cells were cultured in 25 ml CDM4CHO medium in a humidified incubator at 37 °C with 5% CO₂. After 24 h, 400 μg/ml of G418 (345812; Calbiochem EMD, San Diego, CA) was added to select for transfected cells. After 10 days, the cells were collected and lysed, and c-Myc–tagged recombinant murine NEU3 was purified using a Myc-Trap agarose kit (ytak-20; Chromotek, Hauppauge, NY, USA) following the manufacturer’s protocol. Recombinant protein was checked for protein concentration by OD 260/280/320 using a Take3 micro-volume plate with a SynergyMX plate reader (BioTek, Winooski, VT, USA). NEU3 is 51 kDa, and was further purified by centrifugation at 10,000×g for 5 min at 4 °C through an Amicon Ultra 0.5 ml 100 kDa cutoff spin filter (Millipore, Billerica, MA, USA). The NEU3 was analyzed for size and purity by PAGE on, and Coomassie staining of, 4–20% Tris/glycine gels (Bio-Rad, Hercules, CA, USA), as described previously [25, 27‐29]. The NEU3 was stored in 50 μl of 10% glycerol, 100 mM glycine, and 25 mM Tris–HCl, pH 7.3, at 4 °C.

Starting with the murine NEU3 expression plasmid MR223297, a variant mutated to change the tyrosines at positions 179 and 181 to phenylalanines was generated using a QuikChange II Site-Directed Mutagenesis Kit (#200523; Agilent, Santa Clara, CA, USA) and the primer 5′ CATCCCCGCCTTCGCCTTCTATGTCTCACGTTGG 3′, with the underscored sequences representing the point mutation sites. Other workers found that these two mutations eliminate NEU3 sialidase activity [18]. The resulting plasmid was sequenced to confirm the point mutations and absence of other mutations.

Human peripheral blood was collected from healthy volunteers who gave written consent with specific approval from the Texas A&M University human subjects review board (IRB2009-0671D). All methods were performed in accordance with the relevant guidelines and regulations. Blood collection, isolation of peripheral blood mononuclear cells (PBMC), and cell culture were done as described previously [22, 23, 25, 30]. Murine spleen cells were isolated by forcing diced spleen fragments through a 100 μm cell strainer (BD Biosciences, San Jose, CA, USA) using the plunger of a 1 ml syringe (BD Medical, Franklin Lakes, NJ, USA), as described previously [31]. To determine the activity of native and mutated NEU3 we assayed the ability of NEU3 to induce extracellular accumulation of IL-6 from human PBMC and murine spleen cells [23, 25], using human (BioLegend) and murine (PeproTech, Cranbury, NJ, USA) IL-6 ELISA kits following the manufacturers’ protocols.

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by Texas A&M University Animal Use and Care Committee (IACUC 2020-0272). All procedures were performed under 4% isoflurane in oxygen anesthesia, and all efforts were made to minimize suffering. Animals were housed with a 12-h/12-h light–dark cycle with free access to food and water, and all procedures were performed between 09:00 and noon. To induce inflammation and fibrosis, 7–8 week old 20–25 g male and female C57BL/6 mice (Jackson Laboratories, Bar Harbor, ME, USA) were given an oropharyngeal aspiration of 3 U/kg (equivalent to 0.06 U/20 g mouse) bleomycin (2246–10; BioVision Incorporated, Milpitas, CA, USA) in 50 µl of 0.9% saline or oropharyngeal saline alone, as a control, as previously described [22, 23, 25, 32]. Starting 10 days after saline or bleomycin had been administered, some of the mice received an oropharyngeal aspiration of 15 ng of recombinant (rec) murine NEU3 or mutated NEU3 in 50 µl of 0.9% saline, or saline, every 48 h. An additional cohort of mice received only 15 ng of recombinant murine NEU3, mutated NEU3, or saline, every 48 h over the course of 21 days. Independent sets of animal experiments were performed three times over the course of 6 months. Three male and three female mice that did not receive saline, bleomycin, or NEU3 were defined as naïve mice. All the mice were monitored twice daily to observe any sign of distress. At the indicated time points, mice were euthanized by CO₂ inhalation, and bronchoalveolar lavage fluid (BALF) and lung tissue was obtained and analyzed as described previously [22, 23, 25, 32, 33].

BALF cells were counted and processed to prepare cell spots as described previously [22, 23, 25, 32]. After air drying for 48 h at room temperature, some of the cell spots were fixed and immunochemistry was performed as described previously [22, 23, 25, 32]. Briefly, slides with cell spots were incubated with antibodies diluted to 5 µg/ml in PBS containing 2% IgG-free BSA (BSA-50, Rockland Immunochemicals, Pottstown, PA, USA). Antibodies included anti-CD3 (NB600-1441, rabbit clone SP7, Novus Biologicals, Centennial, CO, USA) to detect T-cells, anti-CD11b (101202, clone M1/70, BioLegend, San Diego, CA, USA) to detect blood and inflammatory macrophages, anti-CD11c (M100-3, clone 223H7, MBL International, Woburn, MA, USA) to detect alveolar macrophages and dendritic cells, anti-CD45 (147702, clone I3/2.3, BioLegend) for total leukocytes, anti-Ly6g (127602, clone 1A8, BioLegend) to detect neutrophils, with isotype-matched irrelevant rat (BioLegend) and rabbit (Novus Biologicals) antibodies as controls. After washing six times with PBS for 5 min each, the slides were incubated at room temperature for 30 min with 1 µg/ml biotinylated donkey anti-rat (NBP1-75379, Novus) or biotinylated donkey F(ab′)₂ anti-rabbit (711-066-152, Jackson Laboratories, West Grove, PA, USA) secondary antibodies. After washing, the biotinylated antibodies were detected with a 1/500 dilution of ExtrAvidin alkaline phosphatase (SA-5100-1, Vector Laboratories, Newark, CA, USA). Staining was developed with the Vector Red Alkaline Phosphatase Kit (SK-5100, Vector), following the manufacturers’ protocols. Slides were then counterstained with Gill’s hematoxylin (Sigma-Aldrich). Using a 40 × objective, at least 150 cells from each stained BALF spot were examined and the percent positive cells was recorded.

After collecting BALF, the lungs from the mice were harvested and inflated with Surgipath frozen section compound (#3801480, Leica, Buffalo Grove, IL, USA), frozen on dry ice, and stored at − 80 °C. 10 μm cryosections of lungs were placed on Superfrost Plus glass slides (VWR) and were air dried for 48 h. Immunohistochemistry was done as previously described [22, 23, 25, 32] and as detailed above using anti-CD3, anti-CD11b, anti-CD11c, anti-CD45, anti-Ly6g, or anti-Mac2 (clone M3/38; BioLegend) antibodies with isotype-matched irrelevant rat and rabbit as controls. Positively stained cells were counted from randomly selected fields, and presented as the number of positive cells per mm², as described previously [22, 23, 25, 32]. Lung sections were also stained with Sirius red to detect collagen and analyzed as previously described [23, 25, 34, 35]. Briefly, images were converted to RGB stacks using Image J, and then the green channel (which shows the red staining) was analyzed. Intensity of staining was determined using the threshold tool, which was kept the same for analyzing each set of images. The area stained as a percentage of the total area of the lung tissue was then determined using Image J. Other investigators and ourselves have shown that histological analysis of fibrosis and collagen content by sirius red staining, correlates well with collagen content as measured by antibody staining and hyproxyproline and sirius red (Sircol) assays [23, 25, 34‐38]. Lung sections were also stained using rabbit polyclonal anti-collagen VI antibodies (NB120-6588, Novus Biologicals), as described previously [34, 39]. Lung sections stained with picrosirius red were assessed histologically for fibrosis, using the Ashcroft system, as described previously [38, 40].

Statistical analysis was performed using Prism Version 7.05 (GraphPad Software, La Jolla, CA, USA). Statistical significance between two groups was determined by t test, or between multiple groups using analysis of variance (ANOVA) with both Bonferroni’s and Sidak’s post-tests, and significance was defined as p < 0.05.

We previously observed that the BALF from fibrotic lungs, and fibrotic lesions from human and mouse lungs, have elevated levels of NEU3, that NEU3 inhibitors attenuate bleomycin-induced pulmonary fibrosis, and that Neu3^−/− knockout mice are resistant to bleomycin-induced inflammation and fibrosis [22, 23, 25]. To further elucidate the contribution of NEU3 to fibrosis, we generated recombinant murine NEU3 and a mutated recombinant murine NEU3 protein with two point mutations in the active site. NEU3 and mutated NEU3 were 51 kDa as assessed by PAGE and Coomassie staining (Fig. 1A). Recombinant human NEU3 induces the accumulation of IL-6 from human peripheral blood mononuclear cells (PBMC), and this is blocked by NEU3 inhibitors [22, 23, 25], indicating that the effect on IL-6 is dependent on NEU3 activity. The recombinant murine NEU3, but not the mutated NEU3, upregulated extracellular accumulation of IL-6 in both human and murine cells (Fig. 1B, C), indicating that the recombinant murine NEU3 has NEU3 activity and that the mutated NEU3 does not and is thus inactive.

To determine if elevated levels of NEU3 protein in the lung, without any other exogenous insult, could induce inflammation and/or fibrosis, mice received 15 ng of either recombinant (rec) murine NEU3 or rec murine inactive NEU3 every 48 h by oropharyngeal aspiration. This dose of NEU3 corresponds to the total amount of native mouse NEU3 in the BALF of mice at day 21 after bleomycin [23]. There was no discernable effect of rec NEU3 on the appearance or behavior of the mice, aspiration did not significantly affect body weight, or white fat, liver, kidney, spleen, heart, or brown fat weights as a percent of total body weight (Additional file 1: Fig. S1A–C). Compared to naive controls, rec NEU3 aspiration increased BALF cell counts and B BALF AL CD11b, CD11c and CD45 cell counts in male and female mice (Fig. 2). Compared to rec NEU3-treated mice, rec inactive NEU3-treated mice had lower BALF cell counts (Fig. 2A), and this trend was observed in both male and female mice (Additional file 2: Fig. S2A). Compared to NEU3 aspiration, aspiration of inactive NEU3 caused a smaller increase in CD45 cells in the BALF (Fig. 2B), which was also due to the effect on female but not male mice (Additional file 2: Fig. S2B, C). There were no significant differences in BALF protein levels between naïve mice (1.2 ± 0.2 mg/ml: mean ± SEM, n = 6), NEU3 (1.8 ± 0.4 mg/ml; mean ± SEM, n = 6) and inactive NEU3 (1.4 ± 0.9 mg/ml; mean ± SEM, n = 6) treated mice.

After removing BALF, lung sections were stained to detect immune cells that were not removed by lavage. Compared to naive mice, NEU3 increased the number of CD11b, CD11c, CD45, Mac2, and Ly6g positive cells in the lung tissue (Fig. 3; Additional file 3: Fig. S3). In male mice, compared to naive mice, NEU3 increased the number of CD3, CD11b, CD45, and Mac2 positive cells in the lungs (Additional file 4: Fig. S4A–F). In female mice, NEU3 decreased CD3 positive cells and increased Mac2 and Ly6g positive cells (Additional file 4: Fig. S4E, F). Compared to NEU3 aspiration, aspiration of inactive NEU3 did not significantly affect counts of cells expressing any of the 6 markers (Fig. 3; Additional file 4: Fig S4). Together, the data indicate that aspiration of either form of NEU3 induces inflammation in the lungs of mice.

Lung sections were also stained with picrosirius red to detect total collagen (Fig. 4A–F). In male and female mice, aspiration of NEU3 but not inactive NEU3 caused an increase in picrosirius red staining (Fig. 4G–I). Lung sections were also stained with anti-collagen-VI antibodies (Fig. 4J–O). In male but not female mice, aspiration of NEU3 but not inactive NEU3 caused an increase in collagen VI staining (Fig. 4J, K, L). These data suggest that NEU3 induces fibrosis, that the effect is dependent on NEU3 activity, and that for unknown reasons, the fibrosis in female mice is not accompanied by an increase in collagen VI staining.

Bleomycin aspiration is a standard way to increase inflammation and fibrosis in the lungs of mice [41, 42], and we found that this also increases levels of NEU3 in the lungs [22, 23, 25]. To determine if further increasing levels of NEU3 in the lungs would alter inflammation and fibrosis after aspiration of bleomycin, male and female mice were treated with saline or bleomycin at day 0, and then from days 10 to 20 received 15 ng of either NEU3, inactive NEU3, or saline control every 2 days by oropharyngeal aspiration. As observed previously [43, 44], male mice that received bleomycin have a drop in body weight between days 3 and 10, but there was no significant weight loss in the female mice (Fig. 5A, B). Aspiration of NEU3 or inactive NEU3 did not significantly alter the weights of the mice (Fig. 5A, B). As previously observed, compared to saline controls, bleomycin instillation led to an increase in the number of cells recovered from the BALF in both male and female mice at day 21 (Fig. 5C, D). In the lungs of male mice that received bleomycin, NEU3 aspiration had significantly higher BALF cell counts compared to saline or inactive NEU3 treated mice (Fig. 5C). In female mice that received either saline or bleomycin, NEU3 aspiration did not significantly alter BALF cell counts (Fig. 5D). In bleomycin-treated male but not female mice, aspiration of NEU3 but not inactive NEU3 increased the number of CD11c and CD45 positive cells in the BALF at day 21 (Fig. 6A).

In the lung tissue of bleomycin-treated mice after BAL, NEU3 aspiration, but not aspiration of inactive NEU3, increased the numbers of CD11c and CD45 positive cells (Fig. 7), which was due to the effect on male but not female mice (Additional file 5: Fig. S5A, B). Together, the results suggest that in male but not female mice, increasing levels of NEU3 starting at day 10 after bleomycin causes a further increase in inflammation.

Lung sections were also stained with picrosirius red to detect total collagen (Fig. 8A–L). In saline treated male and female mice, aspiration of NEU3, but not inactive NEU3, from day 10–20 increased picrosirius red staining (Fig. 8M and Additional file 6: Fig S6A, B). In bleomycin treated mice, NEU3 aspiration increased picrosirius red staining (Fig. 8M), which was due to the effect on male but not female mice (Additional file 6: Fig. S6A, B). In mice treated with bleomycin, compared to aspiration of NEU3, aspiration of inactive NEU3 did not significantly increase picrosirius red staining (Fig. 8M; Additional file 6: Fig. S6A, B). Lung sections were also stained with anti-collagen-VI antibodies (Fig. 8N–T). In bleomycin treated mice, NEU3 aspiration increased collagen VI staining (Fig. 8N–T), which was statistically significant in male but not female mice (Additional file 6: Fig. S6C, D). Lung fibrosis was also assessed by histology (Fig. 8U). In saline treated male and female mice, aspiration of NEU3 but not inactive NEU3 increased fibrosis (Fig. 8U; Additional file 6: Fig. S6E, F). In bleomycin treated mice, compared to inactive NEU3, NEU3 aspiration increased fibrosis (Fig. 8U) in both male and female mice (Additional file 6: Fig. S6A, B). These data suggest that increasing NEU3 can increase fibrosis, and that this effect requires NEU3 activity.