Technical advance: The use of tree shrews as a model of pulmonary fibrosis

Background Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive disease with a high morbidity and mortality. Some of the mechanisms of fibrosis development have been described using rodent models; however, the relevance of findings in these animal models is difficult to assess. New innovative models are needed that closely mimic IPF disease pathology. Methods To overcome this unmet need of investigating IPF with a relevant model, we utilized tree shrews, which are genetically, anatomically, and metabolically similar to primates and humans. Using human antibodies and primers, we investigated the role of macrophage phenotypic switching in normal and IPF subjects and bleomycin-injured tree shrews. Results Bronchoalveolar lavage (BAL) cells from tree shrews expressed human markers, and there was recruitment of monocyte-derived macrophages (MDMs) to the lung in IPF subjects and bleomycin-injured tree shrews. MDMs were polarized to a profibrotic phenotype in IPF and in bleomycin-injured tree shrews. Resident alveolar macrophages (RAMs) expressed proinflammatory markers regardless of bleomycin exposure. Tree shrews developed bleomycin-induced pulmonary fibrosis with architectural distortion in parenchyma and widespread collagen deposition. Conclusion The profibrotic polarization of macrophages has been demonstrated to be present in IPF subjects and in fibrotic mice. Although the lung macrophages have long been considered to be homogeneous, recent evidence indicates that these cells are heterogeneous during multiple chronic lung diseases. Here, we show new data that indicate a critical and essential role for macrophage-fibroblast crosstalk promoting fibroblast differentiation and collagen production. in the development and progression of fibrosis. The current data strongly suggest development of therapeutics that attenuate of the profibrotic activation of MDMs may mitigate macrophage-fibroblast interaction. These observations demonstrate that tree shrews are an ideal animal model to investigate the pathogenesis of IPF as they are genetically, anatomically, and metabolically closer to humans than the more commonly used rodent models.


Introduction
Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive disease that is rapidly increasing in prevalence with over 80 per 100,000 persons developing the disease [1,2]. The median survival remains to be dismal at 3-5 years [3]. Some of the mechanisms, including alveolar epithelial injury, fibroblast differentiation and extracellular matrix deposition/stiffening, and profibrotic polarization of myeloid cells, have been described using rodent models of fibrosis [4][5][6]. The relevance of findings in these animal models is difficult to assess due to the dissimilarity of rodents to humans.
We recognize that multiple cell types contribute to IPF development. While existing paradigms have focused on abnormal epithelial-fibroblast interactions in fibrogenesis, a new paradigm based on emerging information from our group and others indicate a critical and essential role for macrophage-fibroblast crosstalk in the development and progression of fibrosis [5,7,8]. While macrophages contribute importantly in the immune response to both infectious and non-infectious agents, they are critical in the repair responses to tissue injury [9]. Fibroblasts respond to cytokines and growth factors produced by macrophages to differentiate into myofibroblasts, which generate extracellular matrix (ECM) proteins, possess contractile properties, and invade tissues to mediate repair and remodeling [10,11].
Macrophage activation is critical in disease development as they can be classically or alternatively activated. These diverse functions are dependent on their ability to polarize between phenotypically distinct sub-populations. Macrophages develop mixed phenotypes in complex pathological conditions, such as lung injury and fibrosis, and polarize to a predominant phenotype depending on the duration and stage of injury repair [8,[12][13][14]. Alternatively activated, or profibrotic, macrophages express characteristic marker proteins involved in profibrotic changes, such as 1) anti-inflammatory and profibrotic immune effectors, 2) up-regulated Larginine metabolism by arginase 1 to generate polyamines and proline, and 3) alteration in extracellular matrix dynamics. An imbalance to a predominant profibrotic phenotype can result in a fibrotic response to injury [12,15].
We and others have shown that lung macrophages from fibrotic mice are profibrotic [5,7,8,16,17]. Although bronchoalveolar lavage (BAL) cells from IPF subjects are polarized to a profibrotic phenotype, the timing and regulation of this phenotypic switching in fibrosis development/progression is poorly understood. To overcome the unmet need of determining the role of profibrotic polarization of macrophages in humans, we utilized a new animal model, bleomycin-injured tree shrews, to investigate the role of macrophage polarization and macrophage-fibroblast interaction in fibrosis. The beneficial features of this animal model is that tree shrews are genetically, anatomically, and metabolically closer to humans than mice or rats [18][19][20][21].

Human subjects
We obtained BAL cells from normal subjects and IPF subjects. Normal subjects had to meet the following criteria: 1) age between 18 and 80 years; 2) no history of cardiopulmonary disease or other chronic disease; 3) no prescription or nonprescription medication except oral contraceptives; 4) no recent or current evidence of infection; and 5) lifetime nonsmoker. IPF subjects had to meet the following criteria: 1) FVC (forced vital capacity) at least 50% predicted; 2) current nonsmoker; 3) no recent or current evidence of infection; and 4) evidence of restrictive physiology on pulmonary function tests and usual interstitial pneumonia on high-resolution chest computed tomography. Fiberoptic bronchoscopy with bronchoalveolar lavage was performed after subjects received local anesthesia. Three sub-segments of the lung were lavaged with five 20-ml aliquots of normal saline, and the first aliquot in each was discarded. The recruitment and bronchoscopy were approved by the University of Alabama at Birmingham and the Birmingham VAMC Institutional Review Boards and conducted between April-December 2019. All subjects were recruited in the UAB or VAMC ILD clinic and provided prior written consent to participate in the study. Subject demographics ( Table 1) are representative of a larger population. Table 2 shows forced vital capacity (FVC) measurements for normal and IPF subjects.

Tree shrew (Tupaia belangeri)
This study was carried out using three-to five-month-old male and female northern tree shrews (Tupaia Belangeri), which average cost ranges between $800-1000. Tree shrews were bred in the University of Alabama at Birmingham (UAB) Tree Shrew Core and raised by their mothers until weaning. After weaning, tree shrews were housed in individual cages (1.52m 3 cage volume) under 14/10 hours light/dark cycles with continuous access to water and dry food. Each cage had a resting board installed. The mating of tree shews can difficult due to their aggressive and territorial nature, but well-matched tree shrews can lead to repeated pregnancies and a successful breeding colony with each litter producing 2-3 tree shrews. Tree shews were intratracheally administered 1.75 U/kg of bleomycin (n = 8) or saline (n = 9) after being anesthetized with a mixture of ketamine/xylazine (90 mg/kg Ketamine, 10 mg/kg Xylazine, i.m.) and isoflurane supplement as needed (0.5-2.0%) using a precision Fortec vaporizer. The health of the animals was monitored daily throughout the experiment. Twenty-one days after the intratracheal administration, euthanasia was performed using a lethal injection of Xylazine and bilateral thoracotomy. Bronchoalveolar lavage (BAL) was harvested by cannulating the trachea and performing total lavage of the lungs three times with 7 ml sterile saline using the flow of gravity to collect cells and fluid. All protocols were approved by the University of Alabama at Birmingham Institutional Animal Care and Use Committee, and in accordance with the National Institutes of Health guidelines for animal care.

Quantitative real time PCR
Total RNA was isolated, reverse transcribed, and quantitative real-time PCR was performed as described previously [7]. The following primer sets were used: human arginase-1:

IPF fibroblast incubation with BAL fluid
IPF fibroblasts were cultured in DMEM supplemented with 10% fetal bovine serum, streptomycin, and amphotericin B. All experiments were conducted in BAL fluid that was centrifuged to remove debris and cells. IPF fibroblast were incubated in equal concentrations of BAL fluid from tree shrews exposed to saline or bleomycin. IPF fibroblasts were harvested after 24 hr of incubation with BAL fluid.

Immunoblot analysis
Cell lysates were harvested in lysis buffer containing a protease inhibitor mix (Roche Applied Science, Complete Mini tablets) and a phosphatase inhibitor mix (Calbiochem). Cell lysates were assayed for protein content using a DCTM protein assay kit (Bio-Rad) and separated by SDS-PAGE and transferred to PVDF membranes. Immunoblot analyses on the membranes were performed with the designated antibodies, followed by the appropriate secondary antibody cross-linked to horseradish peroxidase. Primary antibodies used were human β-actin (A5441, Sigma) and human α-SMA (03-61001, American Research Products).

Hydroxyproline assay
Lung tissue was dried to a stable weight and acid hydrolyzed with 6N HCl for 24 h at 110˚C. Samples were resuspended in 1.5 ml phosphate-buffered saline followed by incubation at 60˚C for 1 h. Samples were centrifuged at 13,000 rpm, and the supernatant was taken for hydroxyproline analysis by using chloramine-T. Hydroxyproline concentration was normalized to the dry weight of the tissue.

Statistics
Statistical comparisons were performed using a student's t test when only two groups of data are presented, or one-way ANOVA with a Tukey's post hoc test when multiple data groups are present. All statistical analyses were expressed as ± S.E.M. unless otherwise noted and p < 0.05 was considered to be significant. GraphPad Prism 5.0 (GraphPad Software) statistical software was used for all analyses. For comparing 3 normal and 5 IPF subjects, the comparison tests were performed with 80% power to detect 2.06 effect size (smaller mean is 2.06 standard deviation apart from the larger mean) at one-sided α = 0.05. Experiments were performed with 3 to 6 tree shrews per group for two group comparisons with 80% power to detect 1.73 to 2.49 effect sizes at one-sided α = 0.05. For four group comparisons with 3 tree shrews per group at onesided α = 0.05, the study would be able to detect that difference between the smallest and largest mean is 2.95 standard deviation. As the experiments were performed with expectation of detecting 5 to 15-fold differences of means, utilized number of tree shrews provided sufficient power to observe the group differences.

Study approval
We obtained BAL cells from normal and IPF subjects under an approved protocols (300001124) by the Human Subjects Institutional Review Board of the University of Alabama at Birmingham and (1670) from the Birmingham Department of Veteran Affairs. Human BAL specimens were used for research only. All subjects provided prior written consent to participate in the study. Animal experiments were approved by the University of Alabama at Birmingham Institutional Animal Care and Use Committee under protocol 21869 and were performed in accordance with NIH guidelines.
https://doi.org/10.1371/journal.pone.0241323.g001 tree shrews had normal lung parenchyma (Fig 1G), whereas there was a dramatic increase in cellular infiltrate and distortion of parenchymal architecture after bleomycin (Fig 1H). Lung histology from bleomycin-injured tree shrews was similar to previous findings from bleomycinexposed mice. Collagen deposition was widespread in bleomycin-injured animals compared to the saline controls (Fig 1I and 1J). The histological findings were verified biochemically by hydroxyproline analysis (Fig 1K). In aggregate, these observations suggest that macrophages from tree shrews express human markers, there is recruitment of MDMs to the lung after bleomycin injury, and tree shrews develop bleomycin-induced pulmonary fibrosis.

Recruited monocyte-derived macrophages are polarized to a profibrotic phenotype
We and others have shown that BAL cells from subjects with pulmonary fibrosis and fibrotic mice have a profibrotic phenotype [5,8,12,16,24]. We questioned if BAL cells from IPF subjects expressed profibrotic genes compared to normal subjects. BAL macrophages from IPF subjects expressed greater arginase 1 (Fig 2A), VEGF (Fig 2B), and CCL18 ( Fig 2C). In contrast, proinflammatory genes, iNOS and TNF-α, were significantly greater in the normal subjects (Fig 2D and 2E).
To determine the macrophage subset that polarized to the profibrotic phenotype, BAL cells were subjected to FACS sorting to separate RAMs and MDMs. All of the profibrotic markers, arginase 1 (Fig 3A), VEGF (Fig 3B), TGF-β1 (Fig 3C), and PDGF-B (Fig 3D) were significantly greater in MDMs than RAMs from bleomycin-injured tree shrews. The gene expression of iNOS ( Fig 3E) and TNF-α ( Fig 3F) were reduced in MDMs and no different from the MDMs from the saline controls. The reverse was seen with the RAMs as they had a proinflammatory phenotype regardless of bleomycin. Taken together, these data demonstrate that the profibrotic polarization of lung macrophages occurs exclusively in MDMs and supports the fact that their recruitment is necessary for dysregulated repair in pulmonary fibrosis. Furthermore, tree shrews represent a novel model to investigate IPF since these animals are genetically and immunologically similar to humans.

Macrophage-derived profibrotic gene expression regulates fibroblast differentiation
We have shown that active TGF-β1 in BAL fluid is primarily derived from lung macrophages and is essential for fibrosis as Tgfb1 -/-Lyz2-cre mice are protected from fibrosis development [7]. To determine if BAL fluid from tree shrews could regulate fibroblast differentiation and ECM gene expression, we incubated IPF fibroblasts with the BAL fluid from tree shrews. An immunoblot analysis revealed that BAL fluid from bleomycin-injured tree shrews greatly increased α-SMA protein in IPF fibroblasts compared to the saline controls (Fig 4A and 4B). The expression of collagen 1a ( Fig 4C) and fibronectin (Fig 4D) was also significantly increased by culturing IPF fibroblasts in BAL fluid from bleomycin-injured tree shrews. These observations suggest that the macrophage-derived profibrotic mediators from bleomycininjured tree shrews promote fibroblast activation.

Discussion
Fibrotic remodeling affects all organ systems, including the heart, kidney, liver, vasculature, and lung. It is estimated that 45% of deaths in the United States are accredited to fibrotic disease annually [25]. IPF, in particular, carries a high mortality with a median survival rate of less than three years, and only 20% of patients survive 5 years after diagnosis [26]. The incidence and mortality from IPF is increasing worldwide with an estimated 40,000 lives lost in the U.S. each year [2,27]. Due to the extreme morbidity and mortality of IPF, there is a critical and unmet need for a more innovative model that will facilitate the study of IPF and further the development and testing of novel therapeutics to treat this devastating disease.
Although the lung macrophages have long been considered to be homogeneous, recent evidence indicates that these cells are heterogeneous during multiple chronic lung diseases [5,17,[28][29][30]. Importantly, the critical role of recruited monocyte-derived macrophages is known to be necessary for the development of fibrosis [5,17,22]; however, little is known about the subset of macrophages that are profibrotic, which is essential for the dysregulated fibrotic repair. The profibrotic polarization of macrophages has been demonstrated to be present in IPF subjects and in bleomycin-injured and asbestos-exposed fibrotic mice [5,7,8,16,17]. The presence of proinflammatory macrophages in IPF is much different. For example, conflicting data exists on the expression of iNOS in humans [31,32]; however, little is known about the expression in IPF. Our data shows that iNOS is significantly decreased in macrophages from IPF subjects, and a similar trend was seen in bleomycin-injured tree shrew macrophages. We further determined that expression is subset specific. MDMs had significantly reduced iNOS expression compared to RAMs from bleomycin-injured tree shrews. In aggregate, our data demonstrate that MDMs are the macrophage subset that polarized to a profibrotic phenotype in bleomycin-injured tree shrews. While there are clear advantages to using mice, including inbred strains, large litter size, availability of antibodies, and ease of developing transgenic/knockout mice, tree shrews are an innovative model to study human diseases because they are anatomically and genetically similar to humans [18,20,21]. Although non-human primates may be the most human-like model, several impracticalities exist including long reproductive cycle, large animal size, high cost, and slow aging process. Tree shrews present an unique opportunity for translational pulmonary research as they are small animals (150-220 grams), have a low cost of maintenance, short reproductive cycle (41-55 days), and attain sexual maturity at 3-6 months [18]. Ophthalmologic studies are one area in which tree shrews are a valuable model because myopia and collagen remodeling of the sclera are similar to the human disease [33,34]. In regard to lung disease, tree shrews have been used for lung cancer [35], Zika viral infection [36], and H1N1 influenza infection [37]. The influenza infection in tree shrews is physiologically analogous to the infection in humans. While the immune system in the mouse greatly differs from that of humans, tree shrews possess the same ligand families as humans, including the major histocompatibility complex (MHC) class I-related chain (MIC) gene and the UL16-binding protein (ULBP) family [18,38]. Anatomically, the tree shrew lung is more similar to humans than mice. Despite the advantages of the tree shrew as an animal model, limited access to this animal and lack of specific antibodies/reagents restrict the usefulness of this model. A limitation of our study is that we did not evaluate the self-resolving nature of bleomycin-induced lung injury in our tree shrews, as our experiments analyzed fibrosis 21 days after exposure. Additionally, the current study did not assess the ability of the tree shrew model to recapitulate the pathological features of human IPF. Nonetheless, the utilization of human antibodies and primers for the first time in an animal model of pulmonary fibrosis strongly suggest that tree shrews are an ideal model to investigate dysregulated fibrotic repair.
The current dogma of the pathogenesis of pulmonary fibrosis has focused on abnormal epithelial-fibroblast interactions [39][40][41]. Here, we show observations that indicate a critical and essential role for monocyte-derived macrophage-fibroblast crosstalk in the development and progression of fibrosis. Fibroblasts are the final effector cells in fibrosis. We have demonstrated that the growth factors that induce fibrosis by regulating fibroblast activation are macrophagederived [7]. More specifically, mice with a conditional deletion of Tgfb1 in macrophages were protected from pulmonary fibrosis. The current data strongly suggest that attenuation of the profibrotic activation of MDMs mitigates macrophage-fibroblast interaction. Thus, we propose that monocyte-derived macrophage-fibroblast crosstalk can induce fibrosis and is the primary driver in mediating disease progression. Taken together, these observations demonstrate that tree shrews are an ideal animal model to investigate the pathogenesis of IPF as they are genetically, anatomically, and metabolically closer to humans than the more commonly used rodent models.