López-Serrano et al. BMC Veterinary Research (2020) 16:28 https://doi.org/10.1186/s12917-020-2250-9 RESEARCH ARTICLE Open Access Heterogeneity of Moraxella isolates found in the nasal cavities of piglets Sergi López-Serrano1, Nuria Galofré-Milà1, Mar Costa-Hurtado1, Ana M. Pérez-de-Rozas1 and Virginia Aragon1,2* Abstract Background: Previous studies have shown that the genus Moraxella is commonly present in the nasal microbiota of swine. Results: In this study, 51 isolates of Moraxella were obtained from nasal swabs from 3 to 4 week old piglets, which represented 26 different fingerprintings by enterobacterial repetitive intergenic consensus (ERIC)-PCR. Whole 16S rRNA gene sequencing allowed the identification at species level of the Moraxella spp. isolates. The majority of the field strains were identified as Moraxella pluranimalium, but Moraxella porci was also detected. In addition, a cluster of 7 strains did not group with any described Moraxella species, probably representing a new species. Subsequent phenotypic characterization indicated that strains of Moraxella pluranimalium were mainly sensitive to serum complement, while the cluster representing the putative new species was highly resistant. Biofilm formation capacity was very variable among the Moraxella spp. isolates, while adherence to epithelial cell lines was similar among selected strains. Additionally, variability was also observed in the association of selected strains to porcine alveolar macrophages. Antimicrobial tests evidenced the existence of multidrug-resistance in the strains. Conclusions: In summary, phenotypic characterization revealed heterogeneity among Moraxella strains from the nasal cavity of piglets. Strains with pathogenic potential were detected as well as those that may be commensal members of the nasal microbiota. However, the role of Moraxella in porcine diseases and health should be further evaluated. Keywords: Swine, Nasal microbiota, Virulence, Bacterial diversity, Moraxella Background relevance of this genus in swine health is not well estab- Moraxella is a microbiota member of the upper respira- lished. Nonetheless, Moraxella may play a role in swine as a tory tract in vertebrates, but some species may cause op- member of the nasal microbiota, since it has been detected portunistic infections. Best-known Moraxella infections as one of the most abundant genera in the nasal cavity of are produced by Moraxella catarrhalis, which causes weaned piglets [5], is abundant in the nasal microbiota of respiratory infections, including pneumonia and otitis in slaughter age pigs [6] and can be found in environmental human [1]. In livestock, Moraxella bovis is well-known as samples in farrowing buildings [7]. the etiological agent of infectious keratoconjunctivitis in The microbiota plays an important role in host health cattle [2]. In swine, the genus Moraxella has representative and disease through different mechanisms, such as mat- species, such as Moraxella porci, which was discovered and uration of the immune system, improvement of the mu- isolated from meninges [3] and Moraxella pluranimalium, cosal barrier and resistance against pathogens [8]. In the described from an abdominal cavity isolate [4], but the case of the nasal microbiota, early colonizers can deter- mine the stability and composition of bacterial commu- nity leading to a healthier status, as reported for children * Correspondence: virginia.aragon@irta.es 1 IRTA, Centre de Recerca en Sanitat Animal (CReSA, IRTA-UAB), Campus de la early colonized by high abundance of Moraxella, which Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain was associated with fewer respiratory infections [9]. 2 OIE Collaborating Centre for the Research and Control of Emerging and However, the specific role of Moraxella, among other Re-emerging Swine Diseases in Europe (IRTA-CReSA), Bellaterra, Barcelona, Spain residents of the respiratory microbiota, in health is still © The Author(s). 2020 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided 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 Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 2 of 12 unknown in pigs. In this study, we performed a geno- repetitive intergenic consensus (ERIC)-PCR determined typic and phenotypic characterization of 51 Moraxella 26 different ERIC fingerprintings, which were not shared field isolates from the nasal turbinates of weaned piglets. among farms (Table 1). Up to 7 different fingerprinting We assessed their capacity of evasion of the innate profiles were isolated from the same farm, and up to 3 immunity by examining their complement susceptibility from a single piglet. One isolate of each fingerprinting and macrophage association capacity, as well as other (from now on referred as different strains for clarity characteristics closely related to its ecological niche, throughout the text) were selected for further analysis. such as biofilm formation, mucin adhesion and cell ad- Sequencing of approximately 1360 bp of the 16S rRNA herence. Our results showed genotypic and phenotypic gene allowed a more precise identification of the Moraxella heterogeneity among the Moraxella isolates. isolates. Sequences from the nasal isolates and sequences from Moraxella sp. type strains from the Ribosomal data- Results base were used to build an UPGMA (unweighted pair Bacterial identification and genotyping group method with arithmetic mean) tree (Fig. 1). Sixteen Initial identification of bacterial isolates by partial 16S nasal strains clustered with more than 99.5% identity with rRNA gene sequencing showed a total collection of 51 Moraxella pluranimalium, while one strain, LL-3, was a Moraxella spp. isolates. Genotyping by enterobacterial bit more divergent, with 99.4% identity. Strain EJ45–1 Table 1 Sampled farms and nasal Moraxella spp. isolated from piglets at weaning (3–4 weeks of age). Ten piglets representing 5 litters from each farm were sampled. Name of strain indicates farm of origin, followed by piglet number and finally the number of isolate. Exceptions to this are LL and CR isolates, which indicate only the number of isolate Farm Selected strains Isolates with same ERIC PCR profile Health status Antimicrobial treatmenta GM GM3–2 GM4–1 Healthy AMOX GM8–1 – GM5–1 GM7–2, GM9–1, GM10–1 GM5–5 GM7–6 GM5–7 GM7–8, GM7–9, GM8–4 AR AR-5A AR-5C Healthy AMOX EJ EJ43-3A – Glässer’s disease ENR EJ44-2A – EJ45–1 EJ45–5 UK/LG UK1–12 UK1–32, LG1–4, LG1–6, LG2–2, LG3–7 Healthy – UK1–20 LG5–5, LG8–3 LG6–2 – LG6–4 LG6–6 LG6-7 g – CR CR-18 – Healthy AMOX CR-10 – CR-7A CR-12 KD KD4–7 – Healthy – VL VL6–4 – Healthy TUL, CEFT VL9–7 VL9–8, VL10–1, VL10–2 VL6–6 – VL1–4 VL1–7, VL5–3 VL3–9 – VL2–5 – VL1–5 VL1–6 LL LL-3 – Nervous signs Unknown AMOX Amoxicillin, ENR enrofloxacin, CEFT ceftiofur, TUL tulathromycin a Antibiotics administered before weaning (before sampling) López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 3 of 12 Fig. 1 UPGMA tree built with the 16S rRNA gene sequences of the Moraxella spp. isolates from the nasal cavities of piglets at weaning. Type strains (T) of different Moraxella species were included in the analysis clustered with Moraxella porci, with 99.2% identity. Strain resistance rate. Other antimicrobials with low resist- CR-7A showed similarity to M. porci and Moraxella cuni- ance rates were ceftiofur, gentamicin and colistin. Six culi, but did not clearly cluster with any of them. Finally, a strains showed resistance to amoxicillin, which was group of 7 nasal strains showed homology among them of abolished with the β-lactamase inhibitor clavulanate. more than 98%, but did not clustered with any of the M. pluranimalium UK1–12 and UK1–20 showed only already described Moraxella species, indicating that they resistance to tetracycline; both strains were isolated may represent a new species. Biochemical characterization from the same farm, which did not use antimicrobials of representative isolates can be found in Additional file 1. before the sampling time. Six strains were resistant to 2 different antimicrobials. Multidrug-resistant (MDR), Antimicrobial susceptibility defined as resistance to 3 or more classes of antimi- Moraxella strains from nasal microbiota showed high crobials, was found in 18 out of the 26 strains. A diversity in antibiotic susceptibility (Table 2). In gen- higher percentage of MDR strains was observed in eral, all the strains were sensitive to amoxicillin+cla- the farms using antimicrobials than in those not using vulanate and florfenicol, in contrast to trimetoprim + these drugs during lactation (75% versus 50%, respect- sulfonamide and tetracycline, where we found a high ively), although no statistical significance was found. Table 2 Antimicrobial susceptibility of nasal Moraxella strains isolated from 3 to 4 week old piglets Strain COL ENR TET DOX S+T AMOX AMC CEFT GEN LI + SP ERY TUL MARB FLOR M. pluranimalium GM3–2 S (25.6) I (18.7) R (9) S (27.3) R (9) S (20.4) S (38.5) S (28.2) S (21.3) S (21.2) S (28.7) S (0.5) R (8) S (1) GM8–1 S (24.4) I (17.6) R (9) S (24.8) I (15.8) I (17.6) S (36.4) S (28.3) S (23.9) S (20) S (24.5) S (0.5) R (8) S (1) AR-5A S (26.5) I (20.9) R (9) I (20.7) R (9) R (15.9) S (38.2) S (23.9) S (24) R (14.5) S (28.2) S (16) R (4) S (1) EJ43-3A S (28) I (19.6) I (20.3) S (24.2) S (27) R (13.4) S (33.1) I (18.7) S (22.9) R (9) R (9) R (> 64) R (8) S (2) EJ44-2A R (9) R (16.1) R (9) R (19.7) R (9) R (9) S (29.9) R (14.5) R (9) R (9) S (24.5) S (16) R (4) S (1) LG6–2 S (21) S (35.9) R (13.6) I (20.8) R (9) I (19.6) S (33) S (30.7) S (17.5) S (20.6) S (25.4) S (4) S (0.125) I (4) UK1–12 S (25.8) S (42.6) R (13.2) S (26) S (17.5) S (21) S (39.5) S (27.2) S (23.9) S (20.3) S (28.7) S (1) S (0.0625) S (0.5) López-Serrano et al. BMC Veterinary Research UK1–20 S (33.4) S (41.5) R (18.5) S (32.8) S (41.1) S (23.5) S (38.4) S (30.3) S (27.5) S (29.6) S (31.9) S (2) S (0.0625) S (0.5) CR-18 S (27.7) S (26.5) R (9) S (29.2) R (9) R (13.2) S (36.2) S (26.1) S (21.5) S (28.9) S (26.1) S (0.5) S (0.5) S (0.5) GM5–1 S (23.6) I (22.6) R (9) S (25.8) I (15.7) S (22.7) S (37.7) S (27.6) S (25.1) S (30.7) S (28.8) S (16) R (4) S (0.5) KD4–7 S (25.2) S (28.2) R (15.7) S (24.9) R (9) S (33.9) S (45.7) S (33.4) S (25.3) S (31.6) S (30) S (2) R (16) S (0.5) (2020) 16:28 LG6–4 S (23.1) S (30.4) R (11.9) I (23.6) R (9) I (16.5) S (36.4) S (29.9) I (15.4) R (15.1) S (25.9) S (8) S (1) S (1) LG6-7 g S (23) S (28) R (12.5) I (22.4) R (9) I (19.7) S (37.5) S (31.6) I (15.4) R (13.6) S (28.3) S (8) R (8) I (4) VL6–4 R (9) I (18.1) R (9) I (20.7) R (9) R (9) S (29.9) S (23.3) R (9) R (9) R (9) R (> 64) R (4) S (2) CR-10 S (31.6) I (18.4) R (9) I (23.6) R (9) I (17.9) S (38.9) S (23.4) S (21.8) I (19.2) S (30) S (0.5) R (8) S (1) VL9–7 I (18.3) I (21.4) R (9) S (29.3) R (9) S (20.8) S (39.4) S (26.7) S (26.8) R (9) R (9) R (> 64) R (4) S (1) LL-3 S (20.5) I (17.1) R (9) I (23.3) R (9) I (17.4) S (34.9) I (21.1) S (18.6) I (16.6) I (20.6) S (8) S (1) S (0.5) M. porci EJ45–1 R (14.5) I (19.7) R (9) R (9) S (22.1) S (23.3) S (40.6) S (33.8) R (9) S (37) S (29.5) S (4) I (2) S (0.125) Other Moraxella GM5–5 S (27.5) S (27) R (9) S (34.2) R (9) I (19.2) S (41.4) S (33.7) R (9) S (24) R (9) R (64) S (1) S (0.5) VL6–6 S (29.2) S (31.7) R (9) S (33.7) R (9) S (41.3) S (57.3) S (51.2) S (29.8) S (65.6) R (9) R (> 64) S (1) S (0.5) VL1–4 R (9) I (20.5) R (9) I (21.8) R (9) S (29.1) S (52.9) S (33.4) S (40.1) S (23.8) R (9) R (> 64) R (4) S (2) VL3–9 S (18.7) I (20.5) R (9) I (22.6) R (9) R (12.5) S (55.5) S (41.1) S (32.7) S (35.6) R (9) R (> 64) I (2) S (2) VL2–5 S (34.4) S (42.3) R (19.6) S (38.5) R (9) S (29.8) S (52.5) S (40.3) S (27.8) I (18.3) R (9) S (0.125) S (0.0625) S (0.125) GM5–7 S (27.6) S (> 30) R (15.5) S (36.9) S (50.2) S (61.4) S (60) S (62.6) S (31.9) S (22.6) R (9) R (64) S (0.625) S (0.125) VL1–5 S (26.3) S (42.7) R (9) S (33.2) I (15.2) S (65.4) S (67.6) S (56.4) S (31.3) S (59) R (9) S (< 0.125) S (0.25) S (0.125) CR-7A S (26.7) I (17.8) R (9) S (24) R (9) S (20.8) S (39.6) S (28.3) S (23.6) I (19.2) S (29.7) S (0.5) R (8) S (1) Diffusion disk breakpoints in mm: COL colistin S ≥ 20 R ≤ 16, ENR enrofloxacin S ≥ 23 R ≤ 16, TET tetracycline S ≥ 24 R ≤ 20, DOX doxycycline S ≥ 24 R ≤ 20; S + T, trimethroprim-sulfonamide S ≥ 16 R ≤ 10, AMOX amoxicillin S ≥ 20 R ≤ 16, AMC amoxicillin + clavulanate S ≥ 30 R ≤ 27, CEFT ceftiofur S ≥ 21 R ≤ 17, GEN, gentamicin S ≥ 16 R ≤ 12, LI + SP lincomycin-spectinomycin S ≥ 20 R ≤ 16, ERY erythromycin S ≥ 23 R ≤ 13 MIC breakpoints in μg/ml, TUL tulathromycin S ≤ 16 R ≥ 64, MARB marbofloxacin S ≤ 1 R ≥ 4, FLOR florfenicol S ≤ 2 R ≥ 8 Page 4 of 12 López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 5 of 12 Serum resistance Biofilm assays Initially, bacterial suspensions were prepared in PBS In an attempt to select non-agglutinating variants, we per- obtaining irregular results, due to auto-agglutination by formed several passes of the agglutinating strains UK1–20 some strains. To overcome this problem, all the strains and VL6–6 in broth culture. Each passage was performed were resuspended in 20% glycerol in PBS and passed with bacteria taken from the surface of the culture that did through a needle if necessary; this process did not affect not agglutinate. Curiously, a thick biofilm was observed in the viability of the bacteria (not shown). the culture tubes of UK1–20 after 27 passes (UK1–20p27); When the strains were assayed for survival after incu- the bacteria in these cultures did not agglutinate but stuck bation with fresh rabbit serum, 17 of the 26 (65%) to the glass. In the case of VL6–6, a biofilm in the air- strains showed a reduction in viability of more than 3 liquid interphase was observed after 27 passes (VL6–6p27) logarithms after incubation with serum (Fig. 2). Within (Fig. 3). This observation prompted us to examine the nat- M. pluranimalium, 12 out of the 16 strains (75%) ural level of biofilm formation capacity of the rest of the showed a reduction of viability of more than 3 logs, nasal Moraxella sp. strains. We found different rates of similar to the type strain of this species, while the rest 4 biofilm formation that were not dependent on the Morax- (25%) showed a reduction of 2–3 logs. On the other ella species. Observing the results, different levels of bio- hand, only 2 of the 7 (29%) strains from the cluster film formation could be arbitrarily established: production representing a putative new Moraxella species showed a of biofilm when Abs590 ≥ 0.6 and of strong biofilm when reduction of more than 3 logs in viability, while the rest Abs590 > 1.5. With those definitions, strains showing bio- of strains in this cluster (5/7; 71%) showed a reduction film production capacity were M. pluranimalium GM3–2, of viability of less than 2 logarithms (Fig. 2), indicating GM8–1, EJ44-2A, LG6–2,, UK1–20, GM5–1, VL6–4 and resistance to the serum complement. Moreover, strains VL9–7; Moraxella sp. LL-3; Moraxella sp. (putative new showing the highest resistance to serum were observed species cluster) VL6–6; and also the variants UK1–20p27 within the isolates from the cluster of the putative new and VL6–6p27 used as controls (Fig. 4). Few isolates (M Moraxella species, and belonged to the same farm pluranimalium GM3–2 and LG6–2, and Moraxella sp. (VL1–4, VL3–9 and VL1–5; Fig. 2). M. porci type strain LL-3) were able to form strong biofilms (Fig. 4). In general, SN9-4 M, as well as strain M. porci EJ45–1, showed high strains forming biofilm under static conditions formed bio- susceptibility to the treatment with rabbit sera (Fig. 2). film also under shaking conditions, although at different 6 5 4 3 2 1 Fig. 2 Serum resistance assay showed diverse CFU reduction among the nasal Moraxella spp. strains. Survival of Moraxella pluranimalium strains (gray bars), uncertainly classified Moraxella strains (white bars), strains from the putative new Moraxella species cluster (black bars) and M. porci (streaked bars) was tested with 90% rabbit serum. Initial bacterial inoculum was approximately 106 CFU/ml. Results are expressed as average of the log CFU reduction after 1 h of incubation at 37 °C with the serum, error bars indicate standard deviation from three replicate assays López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 6 of 12 differences in cell adhesion to both PK-15 (not shown) or A549 cells (Fig. 5). Association to porcine alveolar macrophages Selected strains representing the diversity in phylogen- etic, serum resistance and biofilm formation were tested with porcine alveolar macrophages (PAMs). Different levels of association with PAMs were observed among the strains (Fig. 6). Within the 4 Moraxella pluranima- lium strains tested, only LG6–2 presented a clear associ- ation with PAMS, similar to the level observed in the susceptible H. parasuis strain used as control. Moraxella porci EJ45–1, unlike the M. porci type strain SN9-4 M, showed high levels of association with PAMs. The three isolates from the putative new species cluster presented low association with macrophages, even lower than the H. parasuis phagocytosis resistant strain used as control (Fig. 6). Levels of association to macrophages were not directly linked to biofilm formation capacity or serum resistance (Table 3). In vitro culture passages affected association to macrophages In vitro passages affected the autoagglutination and bio- film formation capacity of M. plurianimalium UK1–20 and Moraxella sp. VL6–6 as indicated above. Although enhanced adhesion to epithelial cells was not observed Fig. 3 Effect of culture passages on biofilm formation. Biofilm in UK1–20p27 (not shown), this strain showed higher produced in glass tubes by strains UK1–20 and VL6–6 after 27 association with PAMs than the original strain (T test, culture passages in the laboratory (UK1–20p27 and VL6–6p27). The original strains are included in the pictures for comparison. Cultures P = 0.04; Fig. 6). In contrast, VL6–6p27 showed the same were incubated for different times (indicated in the figure) and level of interaction with PAMs as the original strain pictures were taken after elimination of the bacterial culture and VL6–6 (Fig. 6); this may be due to a different adaptation staining with crystal violet. Although it is not clear in the picture, to laboratory conditions, as observed by the different VL6–6p27 formed biofilm as a floating pellicle at 24 h type of biofilm formed by this strain. level. Under static conditions, we observed a common Discussion trend of decreasing biofilm production when comparing 24 Moraxella spp. from the nasal microbiota of piglets and 48 h. In contrast, we observed the opposite effect showed heterogeneous characteristics, from antimicro- under shaking conditions, with most of the strains showing bial resistance profiles to virulent mechanisms and ad- a tendency of increasing biofilm production with time. herence properties. The majority of the isolates belonged Strains with slow growth (unclassified cluster) showed the to M. plurianimalium species, but we also detected M. same trend in decreasing biofilm production under both porci, a couple of isolates of uncertain classification and conditions. Biofilm formation was not affected by pre- a cluster of strains that may constitute a new Moraxella coating the wells with mucin or bovine serum albumin species. This cluster comprised strains with a slower (BSA) (data not shown). growth rate in broth than the rest of the nasal strains and contained most of the strains with high resistance to the serum complement. Cell adhesion Higher rates of antimicrobial resistance were found in To explore whether the biofilm formation was associated strains from farms undergoing antimicrobial treatments, with cell adhesion capacity, selected strains with differ- but further studies are needed to demonstrate that the ent rates of biofilm formation were assayed in PK-15 administration of these drugs promotes resistance in this and A549 cell lines. Although the assayed strains pre- bacterial genus. MDR strains were broadly detected, and sented adhesion ability to cells, the observed differences only one strain, UK1–20, from a farm where no antibi- in biofilm formation capacity did not correlated with otics were used, did not show resistance to any of the López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 7 of 12 Fig. 4 Biofilm formation by Moraxella spp. isolates. Biofilm formation under static and shaking conditions in microtiter plates for Moraxella spp. nasal strains. Biofilm was measured at 24 h (open circles) and 48 h (black squares) by staining with crystal violet. In slow growers (cluster of putative new Moraxella species GM5–5, GM5–7, VL1–4, VL1–5, VL2–5, VL3–9, VL6–6) incubations were 48 h and 72 h (open circles and black squares, respectively). Quantification of biofilm was determined the absorbance at 590 nm of the dissolved dye. Strains UK1–20 and VL6–6 after 27 in vitro culture passages (UK1–20p27 and VL6–6p27) were included in the assay. Blank wells with uninoculated medium were also processed and included in the figure. Vertical dashed lines indicate the levels used in this study to consider that a strain had ability to form biofilm (Abs590 ≥ 0.6) or strong biofilm (Abs590 > 1.5) Fig. 5 Bacteria adhesion to A549 cells. Two biofilm forming strains, LG6–2 and UK1–20p27, and two non-forming strains, LG6-7 g and SN9-4 M, were incubated with pulmonary A549 cells to examine their adhesion capacity. After 1 h of incubation, non-attached bacteria were washed and attached bacteria were quantify by dilutions and plating. Results represent the average adherent bacteria from 3 different assays (error bars: standard deviation) López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 8 of 12 Fig. 6 Phagocytosis susceptibility of selected Moraxella spp. strains. Selected M. pluranimalium strains (gray bars), strains from the putative new species cluster (black bars) and M. porci strains (streaked bars). FITC-labelled Moraxella spp. strains were incubated with porcine alveolar macrophages (PAMs) during 1 h at 37 °C. After washing to eliminate unbound bacteria, association of bacteria with PAMs was measured by flow cytometry and expressed as % of fluorescent PAMs. H. parasuis susceptible strain SW114 and resistant strain Nagasaki were included as control for the assay (white bars) and SW114 levels of phagocytosis were used to normalize the results of different tests. In addition, strains UK1–20 and VL6–6 after 27 passages on agar plates were also tested (dotted bars). Results represent the average of duplicate wells from 2 different assays (error bars: standard deviation) antimicrobials tested. It is important to highlight the presence of four MDR strains of Moraxella pluranima- lium, with a high number of drug resistances, including colistin in two of them. This latter observation is consistent with the recent discovery of colistin resistance Table 3 Summary of results from serum susceptibility, genes in Moraxella [10–12]. Coding genes of antibiotic association to alveolar macrophages and biofilm formation resistances can be horizontally transferred between bac- capacity for selected Moraxella spp. Isolates teria of the same or even different species, and therefore, Strains Biofilm formationb Serum susceptibilityc Association to the presence of resistance genes in the resident micro- macrophages biota is of concern since they could be transferred to M. pluranimalium potential pathogens [13]. Diversity of the nasal Morax- LG6–2 + S S ella isolates was also evident in the in vitro assays, where UK1–20 + S R strains showed a wide heterogeneity. LG6-7 g – S R Biofilms are traditionally considered forms of resistance VL6–4 + S R to environmental conditions and a virulence mechanism a in bacteria of clinical importance [14]. As for other char- Moraxella sp. acteristics, we observed heterogeneity in biofilm formation VL6–6 + S R capacity of the porcine nasal strains of Moraxella. Vari- VL1–4 – R R ability in biofilm formation has been already reported for VL1–5 – R R other species of Moraxella from ruminants [15]. Biofilm M. porci formation capacity can be associated also to the col- EJ45–1 – S S onizing ability of the bacteria, as was observed for H. parasuis [16]. The interaction of the nasal Moraxella SN9-4 M – S R strains with the respiratory mucosa needs more study, a Strains in the putative new Moraxella species cluster b Abs590 > 0.8 in the crystal violet assay but attachment to mucin as wells as to epithelial cells c Viability reduction of more than 3 logarithms could play a role, as suggested by our results. López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 9 of 12 On the other hand, bacteria from the upper respiratory to explain the function of this genus in swine health and tract can occasionally reach the lower tract, where they disease. will be confronted with other components of the im- mune system, such as the alveolar macrophages. In vitro Methods assays with PAMs provided insight into the potential of Isolation and bacteria cultures the nasal isolates to survive this first barrier of cell im- Nasal cavities from 10 piglets of around 3 weeks of age munity in the lung. Heterogeneity in this feature has from 5 different litters were sampled from 8 commercial been shown for other bacterial species, such as H. para- farms (Table 1) with standard management practices for suis [17] or Klebsiella pneumoniae [18]. In our case, husbandry and welfare of the animals. Sampling of strains with high association to PAMs, such as LG6–2, piglets was done under institutional authorization and were observed, while other strains did not associate with followed good veterinary practices. According to European the phagocytic cells, as the case of the virulent SN9-4 M. (Directive 2010/63/EU of the European Parliament and of According to the results, macrophage association is not the Council of 22 September 2010 on the protection of directly associated with the phylogeny of the isolates, animals used for scientific purposes) and Spanish (Real since heterogeneity in this trait was observed within the Decreto 53/2013) normative, this procedure did not require distinct species or clusters. One paradigmatic case could specific approval by an Ethical Committee. Nasal sampling be EJ45–1 and SN9-4 M from M. porci, having high differ- was performed only once to each piglet and is not likely to ences between them. The importance that phagocytosis cause pain, suffering, distress or lasting harm equivalent to, resistance could have for in vivo infection/colonization by or higher than, that caused by the introduction of a needle M. pluranimalium is supported by the fact that originally in accordance with good veterinary practice (Chapter I, the UK1–20 strain showed no association with PAMs Article 1, 5 (f) of 2010/63/EU). Nasal swabs were trans- (indicative of resistance to phagocytosis), but lost this trait ported in Amies medium to the laboratory, where they after in vitro culture passages, in agreement with the loss were plated on chocolate agar (Biomérieux, Marcy l’Étoile, of virulence after laboratory adaptation in other bacteria France) to isolate colonies. After 24 and 48 h of incubation [19]. Loss of phagocytosis resistance in UK1–20 was at 37 °C and 5% CO2, different colonies were selected and concurrent with increase biofilm formation capacity. stored at − 80 °C in 20% glycerol-Brain Heart Infusion Thus, biofilm formation may be relevant in environmental broth (BHI) for further characterization. Bacterial suspen- persistence, whereas phagocytosis resistance will be im- sions were also performed in phosphate-buffered saline portant during host infection. However, this phenomenon (PBS) for DNA extraction. seems to be strain-dependent, since it was not observed In addition, reference strains of Moraxella porci SN9- with strain VL6–6, which did not show an adaptation to 4 M and Moraxella pluranimalium CD12-CA4, both laboratory conditions that affected its original resistance isolated from lesions of diseased pigs, were used in this to PAMs. study. For assays with PAMs, 2 reference strains of Hae- Most of the assayed Moraxella spp. strains showed sen- mophilus parasuis were used as controls. sitivity to the serum complement, with some exceptions, including most of the strains in the cluster representing a Bacterial identification and genotyping putative new species. Thus, most nasal Moraxella strains DNA extraction was performed using Chelex based Insta- are probably poorly invasive and will be kept on the nasal gene™ Matrix (Bio-Rad Laboratories, Hercules, CA, USA) mucosa, but the new putative species cluster deserves following manufacturer’s instructions. Preliminary identifica- more attention to define their role in disease in pigs. tion of isolates was performed by partial sequencing of the 16S rRNA gene using primers 358F (CTACGGGAGGCAG- CAGT) and 907R (CCGTCWATTCMTTTGAGTTT) [20]. Conclusions Sequences were analyzed by blasting against the Ribosomal The characterization of Moraxella spp. isolates from the database (http://rdp.cme.msu.edu). nasal cavities of young piglets demonstrated a high het- All isolates identified as Moraxella were then genotyped erogeneity within this genus. Most of the isolates were by ERIC-PCR [21] with primers ERIC-1F (ATGTAAGCTC identified as M. pluranimalium, and only one as M. CTGGGGATTCAC) and ERIC-2R (AAGTAAGTGACTG porci. A group of 7 isolates clustered together by 16S GGGTGAGCG). PCR reaction mixture consisted of 3 mM rRNA gene sequence and did not show homology to any of MgCl2, 1.2 μM of each primer, 0.23 mM of dNTPs, 0.75 known Moraxella species. Many of the Moraxella spp. U of GoTaq® polymerase (Promega, Madison Wisconsin, strains were resistant to multiple antimicrobial classes. USA) and 100 ng of DNA sample. Amplification was carried Heterogeneity of the Moraxella strains was also evident out with an initial denaturation of 94 °C for 2 min followed in adhesion and virulence-associated assays. Further ana- by 30 cycles of 30 s at 94 °C, 1 min at 50 °C and 2.5 min at lyses, including experimental infections, need to be done 72 °C, and a final extension of 20 min at 72 °C. One isolate López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 10 of 12 from each fingerprinting profile was selected for further Association to porcine alveolar macrophages analysis. Association to macrophages was tested as previously Final identification of the different Moraxella strains was described with PAMs [17]. PAMs were obtained from performed by 16S rRNA gene amplification with universal healthy piglets euthanized by intravenous pentobarbital primers 8F (AGAGTTTGATCCTGGCTCAG) and 1492R overdose under institutional authorization. All proce- (CGGTTACCTTGTTACGACTT) [22] and sequencing dures involving animals followed EU and Spanish nor- with 8F, 1492R, 358F and 907R primers. Sequence analysis mative (Directive 2010/63/EU and Real Decreto 53/ was performed with Fingerprinting II v3.0 software (Bio- 2013). Lungs were removed and PAMs were isolated by Rad) and a UPGMA dendrogram was built for species lung lavage with PBS containing 70 μg/mL of gentami- assignment, including sequences from different Moraxella cin. PAMs were collected by centrifugation at 241 x g spp. type strains. Biochemical characterization of represen- for 15 min, washed twice with sterile PBS and stored at tative isolates was performed using the Vitek 2 system − 150 °C in 10% Dimethylsulfoxide (DMSO) in fetal bo- (Biomerieux). vine serum (FBS) until use. Selected nasal strains of Moraxella were tested to de- termine their level of association to PAMs. Moraxella Antimicrobial susceptibility porci reference strain SN9-4 M was also included in the Susceptibility to several antimicrobials was tested as pre- assay. In addition of two reference strains of Haemophi- viously described [23]. Neo-Sensitabs™ diffusion tablets lus parasuis, Nagasaki (virulent) and SW-114 (non-viru- (Rosco Diagnostica, Taastrup, Denmark) were used for lent), were included as control for the technique. Gentamicin, Ceftiofur, Colistin, Erythromycin, Lincoe- For the assay, PAMs were plated in Dulbecco’s Modi- spectin, Tetracycline, Doxycycline, Trimetoprim + Sulfa- fied Eagle Medium (DMEM) supplemented with 10% of mide (T + S), Enrofloxacin, Amoxicillin and Amoxicillin FBS and 1% glutamine at a concentration of 5 × 105 cells + Clavulanic acid testing. Marbofloxacin, florfenicol and per well in 6 well-plates. Plates were incubated 1–2 h to tulathromycin were tested in chocolate agar plates at dif- allow the attachment of the macrophages to the bottom ferent dilutions, from 16 to 0.125 μg/mL in the case of of the wells. For each strain duplicate wells were inocu- florfenicol and marbofloxacin; and from 64 to 0.125 μg/ lated with 107 CFU of fluorescein isothiocyanate (FITC)- mL in the case of tulathromycin. McFarland suspensions labelled bacteria (bacterial inocula were confirmed by of 0.5 in 0.9% NaCl were prepared for each strain and dilutions and plating). After 1 h of incubation at 37 °C streaked on the plates with the different antibiotic con- and 5% CO2, plates were transferred to an ice bath to centration. Minimal inhibitory concentration (MIC) was stop interaction and were washed twice with PBS to determined observing the bacterial growth after 24 h of eliminate unbound bacteria. PAMs were then scrapped incubation. As no clinical breakpoints are available for in PBS and were analyzed by flow cytometry using an these species, available CLSI breakpoints for Moraxella EPICS XL-MCLTM Flow Cytometer (Beckman Coulter, catarrhalis were taken for Tetracycline, Doxycycline, Madrid, Spain) or MACSQuant Analyzer 10 (Myltenyi Trimetoprim+Sulfonamide, and veterinary practice ac- Biotec, Bergisch Gladbach, Germany). Assays were cording to CLSI breakpoints was followed for the rest of repeated with PAMs from different animals that were drugs (EUCAST-and CLSI potency NEO-SENSITABS, already available at the laboratory. Since each batch of Rosco Diagnostica, 2013). macrophages showed different level of activity, values were harmonized considering the reference H. parasuis Serum resistance assay strain SW114 (phagocytosis sensitive) as 100% phagocyt- Serum resistance assay was carried out with rabbit osis in each assay; i.e., using the following calculation: (% serum (EU Directive 2010/63/EU and Spanish normative PAMs with test bacteria - % fluorescence background in Real Decreto 53/2013 were followed). A bacterial sus- PAMs without bacteria / % PAMs with positive control pension of each strain (representative of each ERIC fin- SW114 - % fluorescence background in PAMs without gerprinting) was prepared in PBS with 20% glycerol to bacteria) × 100. reach an OD600 of 0.3 in a VIS 7200 spectrophotometer (Dinko Instruments, Barcelona, Spain). In duplicate Biofilm assays wells, 10 μl of the bacterial suspension (approx. 106 col- Biofilm assays were performed in 96 well cell culture ony forming units [CFU]/mL) were mixed with 90 ul of plates under static and shaking conditions following fresh filtered rabbit serum and mixtures were incubated previous published protocol with some modifications for 1 h at 37 °C and 100 rpm. Bacterial survival was cal- (Bello-Ortí et al. 2014). Bacterial suspensions were made culated by comparing bacterial counts (obtained by serial to reach an OD600 0.3 in BHI. Wells in 96 well plates dilutions and plating) at time 0 and after 1 h incubation. were then inoculated in quadruplicate with a 1:100 dilu- The assay was carried out 3 times for each strain. tion of the bacterial suspension in BHI. Plates were López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 11 of 12 incubated at 37 °C, under static conditions with 5% CO2 Authors’ contributions or under agitation at 100 rpm for 24 and 48 h, except for SLS performed and analyzed the experiments and drafted the manuscript, NGM, MCH and AMPR performed experiments, and VA conceived the study, slow growing isolates, which were incubated for 48 and analyzed results and helped to draft the manuscript. All authors have read 72 h. After incubation, wells were emptied and rinsed and approved the manuscript. with tap water to remove unattached bacteria. Wells were then stained with 0.1% (w/v) crystal violet (Merck, Funding This work was supported by the Ministry of Economy and Competitiveness Darmstadt, Germany) for 2 min at room temperature. (MINECO) of the Spanish Government (grant numbers AGL2013–45662 and Wells were washed thrice with tap water to remove the AGL2016–77361). MCH is a recipient of a Juan de la Cierva fellowship, from excess of crystal violet and let dried at 37 °C. The dye in MINECO (Spain). The authors are also grateful to the Centres de Recerca de Catalunya (CERCA) Programme. Funding institutions did not influence this the stained biofilms was solubilized with 100 μL of 70% study. ethanol and quantified at 590 nm in a Powerwave XS Microplate Spectrophotometer (Biotek Instruments Inc., Availability of data and materials The datasets analysed during the current study are available from the Winooski, VT, USA). Biofilm formation was also corresponding author on reasonable request. assessed on wells pre-coated with 5 μg of mucin or BSA as control. Ethics approval and consent to participate Verbal consent for sampling from the owners of the animals was obtained, following common practice for this type of sampling. Piglet sampling was Cell adhesion assay part of routine veterinary practice in the farms. Sampling of piglets was done under institutional authorization and followed good veterinary practices. Porcine epithelial cells PK-15 (ATCC® CCL-33) and hu- According to European (Directive 2010/63/EU of the European Parliament man lung A549 (ATCC® CCL-185) cell lines were cultured and of the Council of 22 September 2010 on the protection of animals used in DMEM supplemented with 1% glutamine and 5% FBS for scientific purposes) and Spanish (Real Decreto 53/2013) normative, this procedure did not require specific approval by an Ethical Committee. Nasal for PK-15 and 10% for A549 cells. sampling was performed only once to each piglet and is not likely to cause Concentration of cells per ml was calculated to obtain pain, suffering, distress or lasting harm equivalent to, or higher than, that a monolayer after overnight incubation, approximately caused by the introduction of a needle in accordance with good veterinary practice (Chapter I, Article 1, 5 (f) of 2010/63/EU). Blood sampling of rabbit to 50,000 cells/well in cell culture 96 well plates. After obtain serum was performed following good veterinary practices and the overnight incubation, wells with PK-15 or A549 cells above mentioned normative. For PAMs, following Article 18 of 2010/63/EU were washed once with sterile PBS. Duplicate wells were (Sharing organs and tissues), lungs were obtained from animals killed for other reasons. inoculated with 107 CFU/mL bacteria from each isolate tested. Bacterial inoculum quantification was confirmed Consent for publication by colony counts. Microplates were then centrifuged 10 Not applicable. min at 100 x g to facilitate the contact between bacteria and cells. After incubation for 1 h at 37 °C and 5% CO2, Competing interests The authors declare that they have no competing interests. None of the wells were washed twice with sterile PBS to remove un- authors has any financial or personal relationships that could inappropriately attached bacteria, and attached bacteria were released with influence or bias the content of the paper. 0.1% of saponin and pipetting. The resulting suspension Received: 6 November 2019 Accepted: 20 January 2020 was quantified by plating dilutions on agar plates. References Supplementary information 1. Verduin CM, Hol C, Fleer A, van Dijk H, van Belkum A. Moraxella catarrhalis: Supplementary information accompanies this paper at https://doi.org/10. from emerging to established pathogen. Clin Microbiol Rev. 2002;15:125–44. 1186/s12917-020-2250-9. 2. Postma GC, Carfagnini JC, Minatel L. Moraxella bovis pathogenicity: an update. Comp Immunol Microbiol Infect Dis. 2008;31:449–58. Additional file 1. Biochemical characterization by Vitek 2 (Biomerieux) 3. Vela AI, Sanchez-Porro C, Aragon V, Olvera A, Dominguez L, Ventosa A, of representative strains. Fernandez-Garayzabal JF. Moraxella porci sp. nov., isolated from pigs. Int J Syst Evol Microbiol. 2010;60:2446–50. 4. Vela AI, Arroyo E, Aragon V, Sanchez-Porro C, Latre MV, Cerda-Cuellar Abbreviations M, Ventosa A, Dominguez L, Fernandez-Garayzabal JF. Moraxella BHI: Brain Heart Infusion; BSA: Bovine serum albumin; CFU: Colony forming pluranimalium sp. nov., isolated from animal specimens. Int J Syst Evol units; DMEM: Dulbecco’s Modified Eagle Medium; DMSO: Dimethylsulfoxide; Microbiol. 2009;59:671–4. ERIC: Enterobacterial repetitive intergenic consensus; FBS: Fetal bovine serum; 5. Correa-Fiz F, Fraile L, Aragon V. Piglet nasal microbiota at weaning may FITC: Fluorescein isothiocyanate; MDR: Multidrug-resistant; MIC: Minimal influence the development of Glasser's disease during the rearing period. inhibitory concentration; PAMs: Porcine alveolar macrophages; BMC Genomics. 2016;17:404. UPGMA: Unweighted pair group method with arithmetic mean 6. Weese JS, Slifierz M, Jalali M, Friendship R. Evaluation of the nasal microbiota in slaughter-age pigs and the impact on nasal methicillin- resistant Staphylococcus aureus (MRSA) carriage. BMC Vet Res. 2014;10:69. Acknowledgements 7. Cormier Y, Tremblay G, Meriaux A, Brochu G, Lavoie J. Airborne microbial A preliminary version of this study was presented at the Prato Conference contents in two types of swine confinement buildings in Quebec. Am Ind on the Pathogenesis of Bacterial Infections of Animals 2016, as a poster Hyg Assoc J. 1990;51:304–9. presentation (http://vetpath-2016.p.asnevents.com.au/days/2016-10-12/ 8. Fung TC, Olson CA, Hsiao EY. Interactions between the microbiota, immune abstract/38673). and nervous systems in health and disease. Nat Neurosci. 2017;20:145–55. López-Serrano et al. BMC Veterinary Research (2020) 16:28 Page 12 of 12 9. Biesbroek G, Tsivtsivadze E, Sanders EA, Montijn R, Veenhoven RH, Keijser BJ, Bogaert D. Early respiratory microbiota composition determines bacterial succession patterns and respiratory health in children. Am J Respir Crit Care Med. 2014;190:1283–92. 10. AbuOun M, Stubberfield EJ, Duggett NA, Kirchner M, Dormer L, Nunez- Garcia J, Randall LP, Lemma F, Crook DW, Teale C, Smith RP, Anjum MF. mcr-1 and mcr-2 (mcr-6.1) variant genes identified in Moraxella species isolated from pigs in Great Britain from 2014 to 2015. J Antimicrob Chemother. 2017;72:2745–9. 11. Poirel L, Kieffer N, Fernandez-Garayzabal JF, Vela AI, Larpin Y, Nordmann P. MCR-2-mediated plasmid-borne polymyxin resistance most likely originates from Moraxella pluranimalium. J Antimicrob Chemother. 2017;72:2947–9. 12. Kieffer N, Nordmann P, Poirel L. Moraxella Species as Potential Sources of MCR-Like Polymyxin Resistance Determinants. Antimicrob Agents Chemother. 2017;61:e00129–17. 13. Penders J, Stobberingh EE, Savelkoul PH, Wolffs PF. The human microbiome as a reservoir of antimicrobial resistance. Front Microbiol. 2013;4:87. 14. Hall-Stoodley L, Costerton JW, Stoodley P. Bacterial biofilms: from the natural environment to infectious diseases. Nat Rev Microbiol. 2004;2:95–108. 15. Ely VL, Vargas AC, Costa MM, Oliveira HP, Potter L, Reghelin MA, Fernandes AW, Pereira DIB, Sangioni LA, Botton SA. Moraxella bovis, Moraxella ovis and Moraxella bovoculi: biofilm formation and lysozyme activity. J Appl Microbiol. 2019;126:369–76. 16. Bello-Orti B, Deslandes V, Tremblay YD, Labrie J, Howell KJ, Tucker AW, Maskell DJ, Aragon V, Jacques M. Biofilm formation by virulent and non- virulent strains of Haemophilus parasuis. Vet Res. 2014;45:104. 17. Olvera A, Ballester M, Nofrarias M, Sibila M, Aragon V. Differences in phagocytosis susceptibility in Haemophilus parasuis strains. Vet Res. 2009;40:24. 18. Cavalcanti FCN, Rodrigues JF, Cabral AB, Azarias LCB, Morais Junior MA, Castro C, Lopes ACS. Relationships between phagocytosis, mucoid phenotype, and genetic characteristics of Klebsiella pneumoniae clinical isolates. Rev Soc Bras Med Trop. 2019;52:e20190089B. 19. Gaillard ME, Bottero D, Castuma CE, Basile LA, Hozbor D. Laboratory adaptation of Bordetella pertussis is associated with the loss of type three secretion system functionality. Infect Immun. 2011;79:3677–82. 20. Lane DJ. 16S/23S rRNA sequencing. In: Stackebrandt E, Goodfellow M, editors. Nucleic acid techniques in bacterial systematics. England: Wiley; 1991. p. 115–65. 21. Versalovic J, Koeuth T, Lupski JR. Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Res. 1991;19:6823–31. 22. Weisburg WG, Barns SM, Pelletier DA, Lane DJ. 16S ribosomal DNA amplification for phylogenetic study. J Bacteriol. 1991;173:697–703. 23. Lorenzo de Arriba M, Lopez-Serrano S, Galofre-Mila N, Aragon V. Characterisation of Bergeyella spp. isolated from the nasal cavities of piglets. Vet J. 2018;234:1–6. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.