Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Comparison of the broncoalveolar lavage fluid proteomics between foals and adult horses

  • Alejandra A. Rivolta,

    Roles Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Washington State University, Pullman, Washington, United States of America

    ⨯
  • Adina R. Bujold,

    Roles Conceptualization, Investigation, Methodology

    Affiliations Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Washington State University, Pullman, Washington, United States of America, Department of Pathobiology, University of Guelph, Guelph, Ontario, Canada

    ⨯
  • Phillip A. Wilmarth,

    Roles Data curation, Formal analysis, Software

    Affiliation Proteomic Shared Resource, Oregon Health & Science University, Portland, Oregon, United States of America

    ⨯
  • Brett S. Phinney,

    Roles Conceptualization, Data curation, Methodology

    Affiliation Genome Center Proteomics Core Facility, UC Davis, Davis, California, United States of America

    ⨯
  • Joseph P. Navelski,

    Roles Software, Visualization

    Affiliation School of Economic Sciences, Washington State University, Pullman, Washington, United States of America

    ⨯
  • David W. Horohov,

    Roles Funding acquisition

    Affiliation Gluck Equine Research Center, University of Kentucky, Lexington, Kentucky, United States of America

    ⨯
  • Macarena G. Sanz

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Project administration, Resources, Supervision, Writing – review & editing

    macarena@wsu.edu

    Affiliation Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Washington State University, Pullman, Washington, United States of America

    ⨯

Abstract

Neonates have different cellular composition in their bronchoalveolar lavage fluid (BALF) when compared to foals and adult horses; however, little is known about the non-cellular components of BALF. The objective of this study was to determine the proteomic composition of BALF in neonatal horses and to compare it to that of foals and adult horses. Bronchoalveolar lavage fluid samples of seven neonates (< 1 week age), four 5 to 7-week-old foals, and six adult horses were collected. Quantitative proteomics of the fluid was performed using tandem mass tag labeling followed by high resolution liquid chromatography tandem mass spectrometry and protein relative abundances were compared between groups using exact text. A total of 704 proteins were identified with gene ontology terms and were classified. Of these, 332 proteins were related to the immune system in neonates, foals, and adult horses. The most frequent molecular functions identified were binding and catalytic activity and the most common biological processes were cellular process, metabolic process, and biological regulation. There was a significant difference in the proteome of neonates when compared to foals and to adult horses. Neonates had less relative expression (FDR < 0.01) of many immune-related proteins, including immunoglobulins, proteins involved in the complement cascade, ferritin, BPI fold-containing family B member 1, and macrophage receptor MARCO. This is the first report of equine neonate BALF proteomics and reveals differential abundance of proteins when compared to BALF from adult horses. The lower relative abundance of immune-related proteins in neonates could contribute to their susceptibility to pulmonary infections.

Citation: Rivolta AA, Bujold AR, Wilmarth PA, Phinney BS, Navelski JP, Horohov DW, et al. (2023) Comparison of the broncoalveolar lavage fluid proteomics between foals and adult horses. PLoS ONE 18(9): e0290778. https://doi.org/10.1371/journal.pone.0290778

Editor: Toryn Poolman, University College London, UNITED KINGDOM

Received: October 31, 2022; Accepted: August 15, 2023; Published: September 5, 2023

Copyright: © 2023 Rivolta et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Raw data and PAW workflow output is available from the MASSIVE repository # MSV000090531 and Proteome Exchange #PXD037366.

Funding: This study was funded by intramural grants from the University of Kentucky and Washington State University. PW was supported in part by NIH core grants P30EY010572 and P30CA069533 The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The pulmonary immune system of equine neonates is not mature at birth [1]. This may explain why they are susceptible to pathogens such as Rhodococcus equi (R. equi) infection while older foals are less susceptible and adult horses are resistant to infection with this bacterium unless they are immunocompromised [2–4]. The underlying mechanism for age resistance to this and other pathogens remains poorly understood and evaluation of the neonatal respiratory immune system has been mostly limited to the cellular component of bronchoalveolar lavage fluid (BALF) [2, 5, 6]. Equine neonate BALF is comprised of ~ 80% macrophages, 5–20% lymphocytes and ~ 9% of neutrophils. This differs from equine adult BALF that has 40–65% of macrophages, 30–70% of lymphocytes and 5% of neutrophils [7, 8]. The bactericidal activity of macrophages [2] and the phagocytic capacity of neutrophils [9] is lower in foals than in adult horses. As foals age, the percentage of macrophages decreases and the percentage of lymphocytes increases until they reach adult levels by 3 months of age [8]. Neutrophil activity and numbers reach adult levels at 3–4 weeks of age [9, 10].

Little is known about the non-cellular immune components of foal BALF. This fraction includes proteins, immunoglobulins, and other macromolecules such as surfactant complex. Interferon gamma (IFN-γ) production is deficient in BALF-originated lymphocytes of neonates and only reaches adult levels in yearlings [11]. Neonates are also born hypogammaglobulinemic [12] and require colostrum ingestion for passive transfer of immunoglobulins until endogenous production begins at around 1–3 months of life [13]. After the ingestion of colostrum, maternal IgG is detected in high concentrations in nasal secretions during the first 2 weeks of life but endogenous production is not seen for the first 42 days [14]. Pulmonary surfactant proteins play a role in lung protection as they modulate the inflammatory response against pathogens [15]. Surfactant analysis showed a significant difference in the composition of the phospholipids present in surfactant between neonates and adult horses [16].

While other directed protein identification techniques are limited to a small number of proteins, quantitative proteomics methods allow characterization of the whole proteome of a sample and enables for comparison of relative abundances between samples. The proteome of BALF from healthy adult horses [17] and horses with asthma [18] has been reported. To the authors’ knowledge, the proteomic analysis of the equine neonate BALF has not been reported. Age-related differences in protein composition of BALF may be important in the future understanding of the pulmonary immune system and neonatal susceptibility to pulmonary infection. The objective of this study was to determine the proteomic composition of BALF in neonate horses and to compare it to that of older foals and adult horses.

Material and methods

Animals and sample collection

All samples used in the present study were collected as part of a previously reported study [3] that was approved by the Institutional Animal Care and Use Committee at the University of Kentucky (ASAF 2010–0624). Seven neonates (<1 week age, 3 males and 4 females), four 5 to 7-week-old foals (sex not recorded), and six adult Thoroughbred horses (> 10 years old, all females) that belonged to the University of Kentucky research herd were used. The age of the foals was selected based on the fact that this age group shows little to no susceptibility to R. equi infection [3, 4]. All included animals were deemed healthy based on physical examination. All neonates and foals had adequate passive transfer (IgG > 800 mg/dL) and unremarkable thoracic ultrasonography [19]. Bronchoalveolar lavage fluid samples were collected as previously described [20]. Briefly, horses were sedated using xylazine (0.15 mg/kg IV; Butler Co., Dublin, OH) and butorphanol tartrate (0.01 mg/kg IV; Fort Dodge Animal Health, IA, USA). Thereafter, a sterile BAL tube lubricated at the distal end (8mm OD, 160cm for neonates and foals, and 10 mm OD, 300 cm for adult horses) was passed through the nasal passage, through the trachea, and into the distal airway until gently seated in a bronchiole. Luer-lock 60 mL syringes (Tyco Health Care, MA, USA) filled with sterile phosphate-buffered saline (PBS, Corning, NY, USA) were attached to the catheter on the proximal end of the inner BAL tube. The total amount of PBS infused was 100 mL for all neonates and foals and 240 mL for adult horses. The PBS was instilled into the lung and immediately withdrawn using syringe suction. Sampling of the lower airway was visually confirmed by the presence of surfactant (foam) in the fluid recovered.

Sample preparation

Samples were centrifuged at 500 x g for 15 minutes, filtered using 0.2 μm pore filter to remove additional cellular components, and stored at -80 °C until preparation. Thereafter, they were lyophilized to further concentrate them to 75 μg/mL (largest concentration obtained from the more diluted sample). Total protein concentration was determined by BCA assay. Protein preparation was accomplished with S-trap columns (Protifi, NY, USA) according to manufacturer’s instruction and digestion into peptides with trypsin (Promega, WI, USA). Samples were stored at -80 °C until processing.

TMT labeling

All sample peptides were labeled as previously described [21] using TMT Mass Tagging kits (Thermo Scientific, MA, USA). Briefly, peptides were resuspended in 100 μL of triethylammonium bicarbonate (TEAB). A pooled sample for normalization between runs was prepared by combining 6.25 μg of peptide from each sample. TMT labeling was accomplished on 25 μg of peptides by adding 40 μL of TMT label previously resuspended in acetonitrile. Samples were randomly distributed between 10-plex label sets with two pooled samples per set and submitted to UC Davis Proteomics Core Facility for LC-MS/MS analysis.

Nano LC-MS/MS

The proteomics analysis was conducted at UC Davis Proteomics Core Facility (CA, USA). High resolution liquid chromatography (LC) separation was done on a Dionex nano Ultimate 3000 UPCL (Thermo Scientific, MA, USA) followed by electrospray ionization with an Easyspray source and mass spectrometry on Thermo Scientific Fusion Lumus (Thermo Scientific, MA, USA), as previously described [22]. Briefly, multiplexed TMT-labeled samples were resuspended in 5% formic acid and separated by LC. Eluted peptides were dissociated using CID fragmentation energy of 35%, turbo scan speed, 50 ms max inject time, 1x104 automatic gain control (AGC) and maximum parallelizable time turned on, and fragment ions were measured in the linear ion trap. Reporter ion detection was performed in the Orbitrap mass analyzer using MS3 scans with a resolution of 50K and a scan range of 100–500 Da, 105 ms max inject time and 1x105 AGC, following synchronous precursor isolation of the top 10 ions in the linear ion trap, and higher-energy collisional dissociation in the ion-routing multipole.

Database searching

Raw data were converted to text files and compressed using MSConvert (Proteowizard 2.1, http://proteowizard.sourceforge.net). MS2 format files were created for each RAW file and reporter ion peak heights extracted using Python scripts. A canonical FASTA file (taxon = 9796, horse, version 2019.05, 21455 sequences) was downloaded from Uniprot (http://uniprot.org) in May 2019. Common contaminant sequences and sequence-reversed decoys were added and database searching was conducted using Comet (version 2016 release 3) [23]. The searches used semi-tryptic cleavage, 1.25 Da precursor mass tolerance, 1.0005 Da fragment ion tolerances, variable oxidation of methionine, and static modifications for alkylated cysteines and TMT tags (at peptide N-term and lysine). The proteomic analysis workflow (PAW) pipeline was used for peptide spectrum match (PSM) validation using the target/decoy method as previously described [24]. Basic parsimony and extended parsimony analyses were performed to obtain the final protein groups. Identified proteins were mapped to human orthologs using scripts available at https://github.com/pwilmart/PAW_BLAST. Accession numbers were transformed to Gene Ontology (GO) terms using gene ontology enrichment analysis and visualization tool (GOrilla) (http://cbl-gorilla.cs.technion.ac.il/) and classified using PANTHER (http://pantherdb.org/) software and GOrilla to facilitate data visualization.

Statistical analysis

Internal reference scaling (IRS) used the pooled reference samples to put each TMT plex on a common reporter ion intensity scale so that all biological samples could be directly compared [21]. The pool reference sample was prepared using a pool of all the samples submitted for the present study. All statistical analyses were performed using the Bioconductor R statistical software (https://www.r-project.org/) package edgeR [25]. To compare differences in protein expression among the different groups exact test was performed [26, 27]. Benjamini-Hochberg multiple testing correction (a false discovery rate, FDR) of edgeR p-values was set at 0.01. Maximum peptide intensities were log 2 transformed to provide the protein expression levels (PELs) [28].

Results

IRS normalization process

The graphs before and after IRS normalization were plotted (S1 Fig) for the eight identical pooled reference samples, which demonstrated that the IRS method successfully removed the large variation from data acquisition random MS2 scan selection. The trended variance for moderated testing was calculated and a trimmed mean of M-values (TMM) normalization method was performed on the 17 samples used for this study (S2 Fig) [29].

Proteins in BALF

Bronchoalveolar lavage fluid from all samples were analyzed using nano LC-MS/MS. Data were combined and a total of 1,341 proteins were confidently identified. The IRS method requires protein intensity measurements for the pooled reference channels from each of the plexes and resulted in 925 proteins that could be quantified. Eight hundred and eighty-four of the quantifiable proteins had reporter ion signals present in the 17 samples and successfully converted from human ortholog Uniprot accession numbers to NCBI gi accession numbers using Uniprot retrieve tool (S1A Table). A total of 704 proteins had a GO term and were classified through their molecular function and biological process. The most predominant molecular functions present were binding (42.9%, n = 302) and catalytic (38.8%, n = 273) activities. The most common biological processes were cellular process (29.4%, n = 443) followed by metabolic process (18.5%, n = 278); 249 proteins were represented in both processes. The percentage distribution of each category was the same for all groups (Fig 1).

thumbnail
Download:
Fig 1. Percentage of frequency of (A) Molecular functions and (B) Biological processes of the proteins with GO terms present in all the BALF samples from neonates, foals, and adult horses.

Molecular functions with < 1% of frequency and biological processes with < 3.5% of frequency were grouped under “Others” category in each histogram.

https://doi.org/10.1371/journal.pone.0290778.g001

The number of proteins related to immune processes, according to Uniprot, were reported as a percentage of the total proteins identified. Three hundred and thirty-two proteins (37.5%) were related to immune responses in neonates, foals, and adult horses. The complete list of immune-related proteins is detailed in S2 Table.

Comparison between neonates, foals, and adult horses

Following identification of the peptides and normalization of the results, the three groups were compared using the exact test to evaluate differences in the relative abundance level of the BALF proteins. The detailed differences in protein expression levels between groups can be seen in the heatmaps of S3 Table and are listed in S1B and S1C Table.

Immune-related proteins.

For these comparisons, the GO terms of all the immune-related proteins were inputted into GOrilla and grouped in 16 biological processes based on their similarity (Fig 2A). Volcano plots showing the relative protein abundance changes between groups are shown in Fig 3 and the protein candidates for each comparison are mentioned in discussion section.

thumbnail
Download:
Fig 2. Comprehensive scatterplots of relative abundances of the biological processes identified of (A) Immune-related and (B) Non-immune related processes between neonates, foals, and adult horses.

The sum of the average protein reporter ion totals for each biological process is represented in the x axes for neonates (solid triangle), foals (gray circle) and adult horses (open square). Biological processes identified inputting GO terms into GOrilla are shown in the y axes.

https://doi.org/10.1371/journal.pone.0290778.g002

thumbnail
Download:
Fig 3. Volcano plots illustrating the protein abundance of immune-related (left panel) and non-immune related (right panel) changes between (A) Neonates vs adult horses, (B) Neonates vs foals, and (C) Foals vs adult horses.

White circles represent proteins significantly different (FDR < 0.01). The log2 fold change is shown in the x axis. The -log of the FDR is shown in the y axis.

https://doi.org/10.1371/journal.pone.0290778.g003

Neonates vs adult horses.

After annotation at FDR < 0.01 and a fold change > 1.80, which is the relative ratio between the two groups [30], 34 immune-related proteins were down regulated in neonates. A total of 29 immune-related proteins were up regulated in neonates (FDR < 0.01, fold change > 1.79) (Fig 2A). However, these differences in proteins did not impact the relative abundance of the immune-related processes which were all lower in neonates.

Neonates vs foals.

Forty-seven immune-related proteins were down regulated (FDR < 0.01, fold change > 1.68) and 15 immune-related proteins were up regulated (FDR < 0.01, fold change > 2.14) in neonates (Fig 2A). The relative abundance of all the immune-related processes was lower in neonates.

Foals vs adult horses.

Only one immune-related protein had a reduced relative abundance (FDR < 0.01, fold change = 4.45) and 5 immune-related proteins had higher relative abundance (FDR < 0.01, fold change > 3.67) in foals. This group had higher relative abundance of regulation of coagulation, inflammatory response, and platelet-related processes. Adults had higher relative abundances of the remaining immune-related processes (Fig 2A).

Other proteins.

GO terms of the non-immune related proteins were grouped in 33 biological processes based on their process similarities using GOrilla (Fig 2B). Volcano plots showing protein relative abundance differences are presented in Fig 3.

Neonates vs adult horses.

Twenty-four proteins were down regulated (FDR < 0.01, fold change > 1.77), and 81 proteins were up regulated (FDR < 0.01, fold change > 1.62) in neonates. Glycolysis, anatomical structure development, and monosaccharide biosynthetic processes had higher relative abundance in neonates. The rest of the non-immune related processes had higher relative abundance in adult horses (Fig 2B).

Neonates vs foals.

Thirty-four proteins were down regulated (FDR < 0.01, fold change > 1.70) and 26 proteins were up regulated (FDR < 0.01, fold change > 1.78) in neonates. Neonates had higher relative abundances of glycolysis, interaction with symbiont, monosaccharide biosynthetic process and regulation of cell activation; foals had higher relative abundance of the other non-immune processes (Fig 2B).

Foals vs adult horses.

Two proteins had a higher relative abundance in foals (FDR < 0.01, fold change > 2.90). Foals had higher relative abundance in anatomical structure development and iron cholesterol-related processes (Fig 2B). Both groups had similar relative abundance of multicellular organismal process, monosaccharide biosynthetic process, homeostasis, endocytosis, exocytosis, detoxification, cellular oxidant detoxification, secretion, response to stress, regulation of biological quality, and protein-lipid process. Adult horses had higher relative abundance of the remaining non-immune processes (Fig 2B).

Immunoglobulins.

The classification of immunoglobulins into subclasses is not possible with shotgun proteomic methods in part because Uniprot does not have a deep sequence coverage of immunoglobulins [31]. Immunoglobulin isotypes can be classified based on their heavy chain [32]. Therefore, all the peptides identified as part of the same immunoglobulin isotype based on the heavy chain were grouped for comparison. Adult horses and foals presented more relative abundance of immunoglobulins. Immunoglobulin heavy constant γ was the most common peptide in all samples, followed by secretory component, immunoglobulin heavy constant α, immunoglobulin heavy constant μ and in last place immunoglobulin heavy constant ε. The relative abundances were significantly different between groups. Immunoglobulin heavy constant γ, immunoglobulin heavy constant α, immunoglobulin heavy constant μ and secretory component had lower relative abundance in neonates (FDR < 0.01) and there was no difference between foals and adult horses (FDR > 0.05). Immunoglobulin heavy constant ε had lower relative abundance in neonates and foals than in adult horses (FDR < 0.01), Fig 4.

thumbnail
Download:
Fig 4. Heatmap representing the differences in total immunoglobulin and immunoglobulin heavy chain relative abundance between neonates, foals, and adult horses.

Yellow tones indicate a significant lower relative abundance. Groups are shown in columns, the rows represent the immunoglobulin isotypes.

https://doi.org/10.1371/journal.pone.0290778.g004

Discussion

The present study describes a marked variation between the BALF proteome of neonate horses and that of foals and adult horses. To the best of the authors’ knowledge, this is the first time the proteomics of equine neonates and foals BALF is evaluated. Differences in proteome by age was previously reported using not only the TMT method [33], but also others such as two-dimensional difference gel electrophoresis [34, 35]. Profile differences were assessed after ortholog mapping of the identified peptides in all the individuals with BLAST alignment sequence similarity between horse and human proteins. This methodology is commonly used for species that have less developed database annotations [36, 37]. The authors acknowledge that while protein names have several regularities, some can differ among different species [38].

A total of 704 proteins were successfully classified based on their molecular function and biological process. The most frequent molecular functions identified were binding and catalytic activity and the most common biological processes were cellular process, metabolic process, and biological regulation. This is consistent with that described for adult healthy horses, non-smoking humans, mice, and newborn pigs [17, 39–41].

For the group comparisons, the biological processes were identified using GO terms and GOrilla. This resulted in 101 biological processes that were combined into 49 processes (16 immune-related and 33 non-immune related) based on their similarity. Neonates had a reduced relative abundance of most processes when compared to foals and to adult horses.

Neonates had less relative abundance of immune-related proteins than the other two groups. Circulating cytokines and complement proteins have been shown to be lower in neonates than in adult horses [42, 43]. In the present study, the relative abundance of complement proteins in the BALF, which are key for microbial opsonization [44], was lower in neonate than in foals and adult horses. Specifically, C2a, C4b-binding protein, properdin, C1q subunits B and C, C4, and C8 were lower in neonates. C1q is deemed essential for opsonization and killing of R. equi [45]. Thus, this difference in complement composition and its role in age-associated susceptibility to R. equi warrants further investigation. Other immune-related proteins that had a lower relative abundance in BALF of neonates were macrophage receptor with collagenous structure (MARCO), bactericidal/permeability increasing (BPI) fold-containing family B member 1 protein (BPIFB1), and ferritin. Macrophage receptor MARCO mediates binding and ingestion of unopsonized environmental particles and pathogens including Gram-negative and Gram-positive bacteria [46, 47]. BPFB1 belongs to a group of innate immune proteins that exerts its antimicrobial effect binding to lipopolysaccharides of Gram-negative bacteria present in the respiratory tract [48]. This protein is abundant in nasopharyngeal and respiratory secretions of fetal and adult humans [49]. Ferritin is essential for iron storage and iron availability needed for cellular activities [50]. In horses and humans, ferritin is part of the acute phase response increasing as a result of bacterial and viral infection and inflammation [51–53]. This lower abundance of ferritin may result in higher availability of iron for pathogens.

When compared to adult horses, neonates had relatively lower abundance of lipopolysaccharide-binding protein (LBP), a modulator that enables rapid response against LPS of Gram-negative bacteria in mice [54] among other immune-related proteins (S3 Table). Factor H, C6 and C9, all part of the complement cascade [44] and macrophage colony stimulator factor, a protein that establishes and maintains tissue-resident macrophages [55] had lower relative abundance in neonates than in foals (Fig 2A, S3 Table).

Neonates had the highest relative abundance of mucin-1 and lysozyme C. Mucin-1 is a transmembrane glycoprotein that along with other mucins, provides a protective barrier against pathogens [56, 57]. However, this glycoprotein also plays an anti-inflammatory role following the activation of host inflammation in response to a variety of infectious insults [58]. Whether this is also the case in equine neonates with respiratory infection remains to be determined. Lysozyme C is an antimicrobial protein that can hydrolyze the β-1,4-glycosidic bond present in the polysaccharide layer of Gram-positive bacteria cell walls [59]. Similar to what is reported here, this protein decreases with age in humans [60].

Surfactant proteins have a role in lung protection. This group of proteins reduces the surface tension at the air-liquid interface of the lung and modulates the inflammatory response against pathogens [15]. In humans, surfactant protein A is higher in children than in adults [61]. Similar differences were not observed in our study, but this may be due to the low number of individuals included.

Foals and adult horses had similar immune-related protein abundance (Fig 2A). Five proteins were relatively more abundant in foals including apolipoprotein AI which is part of the apolipoproteins family and has been identified in human BALF [62]. This high-density lipoprotein prevents viral penetration, facilitates complement-mediating bacterial killing, and provides protection against parasites [63]. Apolipoprotein I expression was found higher in adult humans that in children [64]. Ferritin was the only protein with a higher relative expression in adult horses, and has also been shown to increase with age in BALF of rats [34].

Non-immune related proteins such as peroxiredoxin-6, aquaporin-1 and ezrin, all of them related to lung development and metabolism, had a higher relative abundance in neonates than in the other groups. Similar age differences have been reported in humans [65–67]. Foals and adult horses had similar protein expression levels of non-immune related proteins. Only superoxide dismutase and beta-galactosidase were higher in foals than in adult horses.

Immunoglobulin G is the most abundant Ig in BALF of adult horses, followed by IgA, IgE and IgM [68]. If all the peptides grouped based on the heavy chain are part of each specific immunoglobulin isotype [32], the Ig abundances in the BALF of neonates and foals in this study followed a similar pattern of abundances. As expected, neonates had the lowest relative abundance of all the immunoglobulins, as production of this immunoglobulin does not reach adult levels until 5–8 weeks of life [69] and mare’s placentation prevents in utero immunoglobulin transfer [1, 12].

This study is not without limitations. Only a small number of horses was included in each group and all the adult horses were females. All horses were Thoroughbreds and were housed in the same environment. Additional work is needed before this information can be extrapolated to other breeds. The TMT labeling is an isobaric method that allows the analysis of peptides and infers protein ID based on unique peptides. This method is followed by IRS normalization, that removes compositional bias, ensuring fully confidentiality when comparing proteins across groups with equal variances [21, 29]. However, as in most of quantitative proteomic methods, the results are expressed as relative abundance changes instead of absolute quantification. Other techniques such as enzyme-linked immunosorbent assay and immunoblotting analysis are needed to confirm the single protein differences observed in our study. An additional limitation is the use of algorithms from open-access tools to predict the protein function which may not be completely accurate.

Overall, equine neonates had lower relative abundance of immune-related proteins than foals and adult horses. In particular, those associated with antibacterial function such as complement proteins, macrophage receptor MARCO, BPI fold-containing family B member 1, ferritin, and immunoglobulins. The lower relative abundance of immune-related proteins in BALF of neonates may contribute to their susceptibility to pulmonary infections. Our study provides the bases for further evaluation of the role some of these single proteins may have in the protection of the neonatal lung.

Supporting information

S1 Fig. Scatterplots of protein intensities from the two pool channels from each of the four TMT experiments.

(A) Pooled standards before IRS. (B) Pooled standards after IRS normalization process.

https://doi.org/10.1371/journal.pone.0290778.s001

(DOCX)

S2 Fig. Intensity distributions for all 17 samples after TMM normalization.

The X axis represents the samples and Y axis represent the log of the intensity. Foals are represented in red, adult horses in green, and neonates in blue.

https://doi.org/10.1371/journal.pone.0290778.s002

(TIF)

S1 Table. Proteins present in all neonate samples and their protein expression levels (PEI).

(A) Uniprot identification, molecular function, and biological processes are included. (B) Fold change of immune related proteins of neonates compared to adults. (C) Fold change of non-immune related proteins of neonates compared to adults.

https://doi.org/10.1371/journal.pone.0290778.s003

(XLSX)

S2 Table. Subclassification and percentages of the proteins involved in the immune system process.

Proteins not recognized by GOrilla but involved in immunity were added to the “Immune system process” section. Uniprot accession numbers are included.

https://doi.org/10.1371/journal.pone.0290778.s004

(DOCX)

S3 Table. Heatmap representing the differences of protein expression levels of (A) proteins involved in the immune system process and (B) proteins involved in other biological processes (FDR < 0.01).

The maroon and orange colors indicate higher expression levels. Yellow color indicates a significant lower expression. Each column represents the group of horses (neonate, foal, and adult) and rows represent the proteins.

https://doi.org/10.1371/journal.pone.0290778.s005

(DOCX)

References

  1. 1. Giguère S, Polkes AC. Immunologic disorders in neonatal foals. Vet Clin North Am Equine Pract. 2005;21(2):241–72, v. pmid:16051049.
  2. 2. Berghaus LJ, Giguère S, Sturgill TL. Effects of age and macrophage lineage on intracellular survival and cytokine induction after infection with Rhodococcus equi. Vet Immunol Immunopathol. 2014;160(1–2):41–50. Epub 2014/03/29. pmid:24736188.
  3. 3. Sanz M, Loynachan A, Sun L, Oliveira A, Breheny P, Horohov DW. The effect of bacterial dose and foal age at challenge on Rhodococcus equi infection. Vet Microbiol. 2013;167(3–4):623–31. Epub 2013/09/24. pmid:24139178.
  4. 4. Horowitz ML, Cohen ND, Takai S, Becu T, Chaffin MK, Chu KK, et al. Application of Sartwell’s model (lognormal distribution of incubation periods) to age at onset and age at death of foals with Rhodococcus equi pneumonia as evidence of perinatal infection. J Vet Intern Med. 2001;15(3):171–5. pmid:11380023.
  5. 5. Hines SA, Stone DM, Hines MT, Alperin DC, Knowles DP, Norton LK, et al. Clearance of virulent but not avirulent Rhodococcus equi from the lungs of adult horses is associated with intracytoplasmic gamma interferon production by CD4+ and CD8+ T lymphocytes. Clin Diagn Lab Immunol. 2003;10(2):208–15. pmid:12626444
  6. 6. Bordin AI, Cohen ND, Giguère S, Bray JM, Berghaus LJ, Scott B, et al. Host-directed therapy in foals can enhance functional innate immunity and reduce severity of Rhodococcus equi pneumonia. Sci Rep. 2021;11(1):2483. Epub 20210128. pmid:33510265
  7. 7. Hostetter SJ, Clark SK, Gilbertie JM, Wiechert SA, Jones DE, Sponseller BA. Age-related variation in the cellular composition of equine bronchoalveolar lavage fluid. Vet Clin Pathol. 2017;46(2):344–53. Epub 20170327. pmid:28346682.
  8. 8. Balson GA, Smith GD, Yager JA. Immunophenotypic analysis of foal bronchoalveolar lavage lymphocytes. Vet Microbiol. 1997;56(3–4):237–46. pmid:9226838.
  9. 9. Bernoco M, Liu IK, Wuest-Ehlert CJ, Miller ME, Bowers J. Chemotactic and phagocytic function of peripheral blood polymorphonuclear leucocytes in newborn foals. J Reprod Fertil Suppl. 1987;35:599–605. pmid:3479614.
  10. 10. Leclere M, Lavoie-Lamoureux A, Lavoie JP. Heaves, an asthma-like disease of horses. Respirology. 2011;16(7):1027–46. pmid:21824219.
  11. 11. Breathnach CC, Sturgill-Wright T, Stiltner JL, Adams AA, Lunn DP, Horohov DW. Foals are interferon gamma-deficient at birth. Vet Immunol Immunopathol. 2006;112(3–4):199–209. Epub 2006/04/18. pmid:16621024.
  12. 12. Sellon DC. Secondary immunodeficiencies of horses. Vet Clin North Am Equine Pract. 2000;16(1):117–30. pmid:10752142.
  13. 13. Rouse BT. The immunoglobulins of adult equine and foal sera: a quantitative study. Br Vet J. 1971;127(1):45–52. pmid:5541943.
  14. 14. Sheoran AS, Timoney JF, Holmes MA, Karzenski SS, Crisman MV. Immunoglobulin isotypes in sera and nasal mucosal secretions and their neonatal transfer and distribution in horses. Am J Vet Res. 2000;61(9):1099–105. pmid:10976743.
  15. 15. Wright JR. Immunoregulatory functions of surfactant proteins. Nat Rev Immunol. 2005;5(1):58–68. pmid:15630429.
  16. 16. Christmann U, Livesey LC, Taintor JS, Waldridge BM, Schumacher J, Grier BL, et al. Lung surfactant function and composition in neonatal foals and adult horses. J Vet Intern Med. 2006;20(6):1402–7. pmid:17186857.
  17. 17. Bright LA, Mujahid N, Nanduri B, McCarthy FM, Costa LR, Burgess SC, et al. Functional modelling of an equine bronchoalveolar lavage fluid proteome provides experimental confirmation and functional annotation of equine genome sequences. Anim Genet. 2011;42(4):395–405. Epub 20110206. pmid:21749422.
  18. 18. Feutz MM. Proteomic analysis of bronchoalveolar lavage fluid in an equine model of asthma during a natural antigen exposure trial. In: Riley CP, Zhang X., Adamec J., Thompson C., & Couetil L. L., editor. Journal of Integrated OMICS, 2 (2)2012. p. 123–31.
  19. 19. Sanz MG, Bradway DS, Horohov DW, Baszler TV. Rhodococcus equi-specific hyperimmune plasma administration decreases faecal shedding of pathogenic R. equi in foals. Vet Rec. 2019;185(1):19. Epub 20190417. pmid:30995996.
  20. 20. Hoffman AM. Bronchoalveolar lavage: sampling technique and guidelines for cytologic preparation and interpretation. Vet Clin North Am Equine Pract. 2008;24(2):423–35, vii–viii. pmid:18652963.
  21. 21. Plubell DL, Wilmarth PA, Zhao Y, Fenton AM, Minnier J, Reddy AP, et al. Extended Multiplexing of Tandem Mass Tags (TMT) Labeling Reveals Age and High Fat Diet Specific Proteome Changes in Mouse Epididymal Adipose Tissue. Mol Cell Proteomics. 2017;16(5):873–90. Epub 2017/03/21. pmid:28325852
  22. 22. Tiwari S, Mishra M, Salemi MR, Phinney BS, Newens JL, Gomes AV. Gender-specific changes in energy metabolism and protein degradation as major pathways affected in livers of mice treated with ibuprofen. Sci Rep. 2020;10(1):3386. Epub 20200225. pmid:32099006
  23. 23. Eng JK, Jahan TA, Hoopmann MR. Comet: an open-source MS/MS sequence database search tool. Proteomics. 2013;13(1):22–4. Epub 20121204. pmid:23148064.
  24. 24. H D Sagawa C, de A B Assis R, Zaini PA, Wilmarth PA, Phinney BS, Moreira LM, et al. Proteome Analysis of Walnut Bacterial Blight Disease. Int J Mol Sci. 2020;21(20). Epub 20201009. pmid:33050347
  25. 25. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26(1):139–40. Epub 20091111. pmid:19910308
  26. 26. Robinson MD, Smyth GK. Moderated statistical tests for assessing differences in tag abundance. Bioinformatics. 2007;23(21):2881–7. Epub 20070919. pmid:17881408.
  27. 27. Robinson MD, Smyth GK. Small-sample estimation of negative binomial dispersion, with applications to SAGE data. Biostatistics. 2008;9(2):321–32. Epub 20070829. pmid:17728317.
  28. 28. Tesena P, Kingkaw A, Vongsangnak W, Pitikarn S, Phaonakrop N, Roytrakul S, et al. Preliminary Study: Proteomic Profiling Uncovers Potential Proteins for Biomonitoring Equine Melanocytic Neoplasm. Animals (Basel). 2021;11(7). Epub 20210627. pmid:34199079
  29. 29. Robinson MD, Oshlack A. A scaling normalization method for differential expression analysis of RNA-seq data. Genome Biol. 2010;11(3):R25. Epub 20100302. pmid:20196867
  30. 30. Aguilan JT, Kulej K, Sidoli S. Guide for protein fold change and p-value calculation for non-experts in proteomics. Mol Omics. 2020;16(6):573–82. Epub 20200924. pmid:32968743.
  31. 31. Omenn GS, Lane L, Overall CM, Paik YK, Cristea IM, Corrales FJ, et al. Progress Identifying and Analyzing the Human Proteome: 2021 Metrics from the HUPO Human Proteome Project. J Proteome Res. 2021;20(12):5227–40. Epub 20211020. pmid:34670092
  32. 32. Hood L, Campbell JH, Elgin SC. The organization, expression, and evolution of antibody genes and other multigene families. Annu Rev Genet. 1975;9:305–53. pmid:813561.
  33. 33. Anoshchenko O, Prasad B, Neradugomma NK, Wang J, Mao Q, Unadkat JD. Gestational Age-Dependent Abundance of Human Placental Transporters as Determined by Quantitative Targeted Proteomics. Drug Metab Dispos. 2020;48(9):735–41. Epub 20200626. pmid:32591415
  34. 34. Umstead TM, Freeman WM, Chinchilli VM, Phelps DS. Age-related changes in the expression and oxidation of bronchoalveolar lavage proteins in the rat. Am J Physiol Lung Cell Mol Physiol. 2009;296(1):L14–29. Epub 20081017. pmid:18931054.
  35. 35. Hollung K, Grove H, Færgestad EM, Sidhu MS, Berg P. Comparison of muscle proteome profiles in pure breeds of Norwegian Landrace and Duroc at three different ages. Meat Sci. 2009;81(3):487–92. Epub 20081010. pmid:20416601.
  36. 36. McGorum BC, Pirie RS, Eaton SL, Keen JA, Cumyn EM, Arnott DM, et al. Proteomic Profiling of Cranial (Superior) Cervical Ganglia Reveals Beta-Amyloid and Ubiquitin Proteasome System Perturbations in an Equine Multiple System Neuropathy. Mol Cell Proteomics. 2015;14(11):3072–86. Epub 20150913. pmid:26364976
  37. 37. Dong Z, Haines S, Coates D. Proteomic Profiling of Stem Cell Tissues during Regeneration of Deer Antler: A Model of Mammalian Organ Regeneration. J Proteome Res. 2020;19(4):1760–75. Epub 20200316. pmid:32155067.
  38. 38. Franzén K, Eriksson G, Olsson F, Asker L, Lidén P, Cöster J. Protein names and how to find them. Int J Med Inform. 2002;67(1–3):49–61. pmid:12460631.
  39. 39. Chen J, Ryu S, Gharib SA, Goodlett DR, Schnapp LM. Exploration of the normal human bronchoalveolar lavage fluid proteome. Proteomics Clin Appl. 2008;2(4):585–95. pmid:21136857
  40. 40. Guo Y, Ma SF, Grigoryev D, Van Eyk J, Garcia JG. 1-DE MS and 2-D LC-MS analysis of the mouse bronchoalveolar lavage proteome. Proteomics. 2005;5(17):4608–24. pmid:16240291.
  41. 41. Bartlett JA, Albertolle ME, Wohlford-Lenane C, Pezzulo AA, Zabner J, Niles RK, et al. Protein composition of bronchoalveolar lavage fluid and airway surface liquid from newborn pigs. Am J Physiol Lung Cell Mol Physiol. 2013;305(3):L256–66. Epub 20130524. pmid:23709621
  42. 42. Bernoco MM, Liu IK, Willits NH. Hemolytic complement activity and concentrations of its third component during maturation of the immune response in colostrum-deprived foals. Am J Vet Res. 1994;55(7):928–33. pmid:7978631.
  43. 43. Boyd NK, Cohen ND, Lim WS, Martens RJ, Chaffin MK, Ball JM. Temporal changes in cytokine expression of foals during the first month of life. Vet Immunol Immunopathol. 2003;92(1–2):75–85. pmid:12628765.
  44. 44. Noris M, Remuzzi G. Overview of complement activation and regulation. Semin Nephrol. 2013;33(6):479–92. pmid:24161035
  45. 45. Folmar CN, Cywes-Bentley C, Bordin AI, Rocha JN, Bray JM, Kahn SK, et al. In vitro evaluation of complement deposition and opsonophagocytic killing of Rhodococcus equi mediated by poly-N-acetyl glucosamine hyperimmune plasma compared to commercial plasma products. J Vet Intern Med. 2019;33(3):1493–9. Epub 2019/04/29. pmid:31034109
  46. 46. Kraal G, van der Laan LJ, Elomaa O, Tryggvason K. The macrophage receptor MARCO. Microbes Infect. 2000;2(3):313–6. pmid:10758408.
  47. 47. Kanno S, Hirano S, Sakamoto T, Furuyama A, Takase H, Kato H, et al. Scavenger receptor MARCO contributes to cellular internalization of exosomes by dynamin-dependent endocytosis and macropinocytosis. Sci Rep. 2020;10(1):21795. Epub 20201211. pmid:33311558
  48. 48. Marra MN, Wilde CG, Griffith JE, Snable JL, Scott RW. Bactericidal/permeability-increasing protein has endotoxin-neutralizing activity. J Immunol. 1990;144(2):662–6. pmid:2295804.
  49. 49. Xu Y, Tao Z, Jiang Y, Liu T, Xiang Y. Overexpression of BPIFB1 promotes apoptosis and inhibits proliferation via the MEK/ERK signal pathway in nasopharyngeal carcinoma. Int J Clin Exp Pathol. 2019;12(1):356–64. Epub 20190101. pmid:31933752
  50. 50. Theil EC. Ferritin: structure, gene regulation, and cellular function in animals, plants, and microorganisms. Annu Rev Biochem. 1987;56:289–315. pmid:3304136.
  51. 51. Kawamata R, Yokoyama K, Sato M, Goto M, Nozaki Y, Takagi T, et al. Utility of serum ferritin and lactate dehydrogenase as surrogate markers for steroid therapy for Mycoplasma pneumoniae pneumonia. J Infect Chemother. 2015;21(11):783–9. Epub 20150819. pmid:26298038.
  52. 52. Guan WJ, Ni ZY, Hu Y, Liang WH, Ou CQ, He JX, et al. Clinical Characteristics of Coronavirus Disease 2019 in China. N Engl J Med. 2020;382(18):1708–20. Epub 20200228. pmid:32109013
  53. 53. Dondi F, Lukacs RM, Gentilini F, Rinnovati R, Spadari A, Romagnoli N. Serum amyloid A, haptoglobin, and ferritin in horses with colic: Association with common clinicopathological variables and short-term outcome. Vet J. 2015;205(1):50–5. Epub 20150319. pmid:25981935.
  54. 54. Knapp S, Florquin S, Golenbock DT, van der Poll T. Pulmonary lipopolysaccharide (LPS)-binding protein inhibits the LPS-induced lung inflammation in vivo. J Immunol. 2006;176(5):3189–95. pmid:16493079.
  55. 55. Laoui D, Van Overmeire E, De Baetselier P, Van Ginderachter JA, Raes G. Functional Relationship between Tumor-Associated Macrophages and Macrophage Colony-Stimulating Factor as Contributors to Cancer Progression. Front Immunol. 2014;5:489. Epub 20141007. pmid:25339957
  56. 56. Kashyap B, Kullaa AM. Regulation of mucin 1 expression and its relationship with oral diseases. Arch Oral Biol. 2020;117:104791. Epub 20200701. pmid:32652493.
  57. 57. Dhar P, Ng GZ, Dunne EM, Sutton P. Mucin 1 protects against severe Streptococcus pneumoniae infection. Virulence. 2017;8(8):1631–42. Epub 20170713. pmid:28605238
  58. 58. Kato K, Lillehoj EP, Lu W, Kim KC. MUC1: The First Respiratory Mucin with an Anti-Inflammatory Function. J Clin Med. 2017;6(12). Epub 20171129. pmid:29186029
  59. 59. Ferraboschi P, Ciceri S, Grisenti P. Applications of Lysozyme, an Innate Immune Defense Factor, as an Alternative Antibiotic. Antibiotics (Basel). 2021;10(12). Epub 20211214. pmid:34943746
  60. 60. Mackie IA, Seal DV. Quantitative tear lysozyme assay in units of activity per microlitre. Br J Ophthalmol. 1976;60(1):70–4. pmid:1268164
  61. 61. Ratjen F, Rehn B, Costabel U, Bruch J. Age-dependency of surfactant phospholipids and surfactant protein A in bronchoalveolar lavage fluid of children without bronchopulmonary disease. Eur Respir J. 1996;9(2):328–33. pmid:8777972.
  62. 62. Wattiez R, Hermans C, Bernard A, Lesur O, Falmagne P. Human bronchoalveolar lavage fluid: two-dimensional gel electrophoresis, amino acid microsequencing and identification of major proteins. Electrophoresis. 1999;20(7):1634–45. pmid:10424490.
  63. 63. Georgila K, Vyrla D, Drakos E. Apolipoprotein A-I (ApoA-I), Immunity, Inflammation and Cancer. Cancers (Basel). 2019;11(8). Epub 20190801. pmid:31374929
  64. 64. Thompson A, Danesh J. Associations between apolipoprotein B, apolipoprotein AI, the apolipoprotein B/AI ratio and coronary heart disease: a literature-based meta-analysis of prospective studies. J Intern Med. 2006;259(5):481–92. pmid:16629854.
  65. 65. Arevalo JA, Vázquez-Medina JP. The Role of Peroxiredoxin 6 in Cell Signaling. Antioxidants (Basel). 2018;7(12). Epub 20181124. pmid:30477202
  66. 66. Song Y, Fukuda N, Bai C, Ma T, Matthay MA, Verkman AS. Role of aquaporins in alveolar fluid clearance in neonatal and adult lung, and in oedema formation following acute lung injury: studies in transgenic aquaporin null mice. J Physiol. 2000;525 Pt 3:771–9. pmid:10856128
  67. 67. Xie JJ, Zhang FR, Tao LH, Lü Z, Xu XE, Jian-Shen , et al. Expression of ezrin in human embryonic, fetal, and normal adult tissues. J Histochem Cytochem. 2011;59(11):1001–8. Epub 20110810. pmid:21832146
  68. 68. Halliwell RE, McGorum BC, Irving P, Dixon PM. Local and systemic antibody production in horses affected with chronic obstructive pulmonary disease. Vet Immunol Immunopathol. 1993;38(3–4):201–15. pmid:8291200.
  69. 69. Perkins GA, Wagner B. The development of equine immunity: Current knowledge on immunology in the young horse. Equine Vet J. 2015;47(3):267–74. Epub 20150106. pmid:25405920.